Identification of Pluripotent Stem Cells and Characterization of Glia
in the Planarian Schmidtea mediterranea
by
Irving E. Wang
B.A. Biology (2003), M.S. Molecular, Cell and Developmental Biology (2004)
The Johns Hopkins University, Baltimore, MD
SUBMITTED TO THE DEPARTMENT OF BIOLOGY IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY IN BIOLOGY
AT THE
MASSACHUSETTS INSTITUTE OF TECHNOLOGY
SEPTEMBER 2014
MASSACHUSETT ISTo
1ITUTE
OF TECI-NWLCGy
SEP02
201fl
C 2014 Irving E. Wang. All rights reserved.
LBRARiES
This author hereby grants to MIT permission to reproduce
and to distribute publicly paper and electronic
copies of this thesis document in whole or in part
in any medium now known or hereafter created.
Signature of Author....
Signature redacted
Irving E. Wang
Department of Biology
August 29, 2014
Certified by.Signature
redacted
Signature redacted
Peter W. Reddien
Associate Professor of Biology
Thesis Supervisor
Accepted by.....
Michael T. Hemann
Associate Professor of Biology
Chair, Committee for Graduate Students
1
Identification of Pluripotent Stem Cells and Characterization of Glia
in the Planarian Schmidtea mediterranea
by
Irving E. Wang
Submitted to the Department of Biology
on September 2, 2014 in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy in Biology
ABSTRACT
Given their regenerative capacity, the planarian Schmidtea mediterraneahas emerged as a model
system for the study of stem cell biology, tissue specification, and axis formation. Many aspects
of the regenerative machinery have yet to be characterized. Although it is known that neoblasts,
the population of all proliferative cells in the adult planarian, are the source of new tissue during
regeneration, it is unknown whether neoblasts consist of multiple subpopulations of lineagerestricted multipotent stem cells or if there exists a pluripotent stem cell type. We developed two
methods for performing clonal analysis to determine the potential of neoblasts: sublethal
irradiation and single-cell transplantation. Colonies consisting of both self-renewed neoblasts
expressing stem cell markers and differentiated cells expressing specialized tissue markers from
single cells that we have termed clonogenic neoblasts. These cells are capable of generating all
differentiated cell types in the adult animal and restoring regeneration in hosts where endogenous
neoblasts have been ablated. These findings provide insight into the overall process of
regeneration and the regulation of pluripotent adult stem cells during regeneration.
Regeneration and homeostasis, the gradual turnover and replacement of cells in the adult
planarian, both require the formation of new cells and signaling pathways to control their
specification and function. The Hedgehog signaling pathway has been implicated in anteriorposterior polarity specification, but no role in planarian nervous system regeneration has been
described despite that hedgehog is expressed in neurons in the brain. Although Hedgehog
signaling is critical for central nervous system development in vertebrates and Drosophila, our
data indicate that it is not involved in patterning the planarian brain. Instead, Hedgehog
signaling regulates gene expression in a cell type identified as planarian glia from its localization
to the axon-rich neuropil, expression of planarian orthologs of astrocyte genes, and branching
morphology in close association with neurons. Evidence of both the existence of planarian glia
and their regulation by Hedgehog signaling offers the opportunity to dissect glial cell biology in
a highly regenerative model organism and understand the evolution of the cell type.
Thesis Advisor: Peter W. Reddien
Title: Associate Professor of Biology
3
4
Table of Contents
.. 7
Chapter 1: Introduction .........................................................................................
8
--------------.............
.
...
Foreword......................................................................................--9
I. G lial Cells of V ertebrates and Invertebrates................................................................
20
II. Adult Gliogenesis in CNS Repair and Regeneration ..............................................
34
III. Planaria as a M odel for Glial Cell Biology.............................................................
43
IV . Content Overview ................................................................................................
46
......................
Figures.................................................................................................
56
-..........
References..............................................................................................................
Chapter 2: Clonogenic neoblasts are pluripotent stem cells
that underlie planarian regeneration...................................................................................
. ---------------.......
A bstract.....................................................................................................
.. .....
Introduction................................................................................................
Results and D iscussion .................................................................................................
Figures...........................................................................................................................
M aterials and M ethods.....................................................................................................137
A cknow ledgem ents..........................................................................................................145
References........................................................................................................................146
73
74
75
76
86
Chapter 3: Hedgehog signaling regulates gene expression
.--.... ... 151
in planarian glia ..................................................................................................
152
Abstract ..........................................................................................................................
154
..
..
Introduction....................................................................................................
. --................ 157
Results.....................................................................................................
Discussion........................................................................................................................175
....182
... .
Figures...........................................
M aterials and M ethods.....................................................................................................216
.... 220
References....................................................................................................
227
C hapter 4: D iscussion ................................................................................................................
Cells........................................................................................228
Stem
dult
A
I. Pluripotent
II. Functions of Planarian Glia........................................................................................232
III. Roles of G lia Cells in Regeneration...........................................................................236
...................................... 240
IV . G lial Cell Evolution...........................................................o
243
V . Conclusion ..................................................................................................................
References........................................................................................................................244
5
6
Chapter 1
Introduction
7
Foreword
Until recently, the most complex nervous system known, the mammalian brain, was thought to
produce no new cells during adulthood.
Recently, however, numerous studies have
demonstrated the formation of new neurons in certain regions of the brain and the potential for
neural stem cells to regenerate damaged portions following major injury (Lois and AlvarezBuylla, 1993; Nakatomi et al., 2002).
Glia are an indispensible cell type that comprise the
majority of cells in the mammalian brain.
Their roles in the development, function, and
refinement of neural networks are only beginning to be touched upon. Understanding how these
cells, characterized by their highly specialized and elaborate morphologies, are regenerated and
aid in the regeneration of neurons will benefit the treatment of brains compromised by traumatic
injury or disease.
Planarians, free-living flatworms, are noted for their remarkable regenerative potential
and provide a tractable system in which to study glia regeneration (Morgan, 1898). Whereas
mammalian nervous system regeneration is limited to replenishment of depleted cell populations
and reformation of certain structures such as myelin, planarians can regenerate any missing piece
of the nervous system and adults can form new brains entirely de novo.
By comparing
postembryonic gliogenesis between mammalian systems where regeneration is limited and
invertebrate systems where regeneration is pervasive and accessible, much can be learned about
the conserved mechanisms shared between both and the reasons behind limited regenerative
potential in vertebrate glia.
8
I. Glial Cells of Vertebrates and Invertebrates
Specification and Function of Vertebrate CNS Macroglia
Introduction to glia subtypes in the CNS
Vertebrate glia in the central nervous system can be classified into two general classes based on
their lineage. Microglia are mesodermal in origin and act as resident immune cells in the brain
(Hess et al., 2004).
Macroglia are ectodermal, deriving from the same progenitor cells as
neurons during development, and can be subdivided into at least three types based on
morphology and function: ependymal cells, astrocytes, and oligodendrocytes.
Although
microglia are critical for CNS response to various types of damage (Hanisch and Kettenmann,
2007), I will focus on the specification of macroglia from radial glia and the biology of postnatal
astrocytes and oligodendrocytes.
Developmentalorigins of neurons and glia
CNS macroglia and neurons derive from neural stem cells known as radial glia. These cells were
originally identified in the developing brain as glial cells because of their bipolar morphology
and were thought to provide a track for migrating neurons and axons. They are now recognized
as multipotent progenitors and as the primary source of ectodermal cells in the central nervous
system (Malatesta et al., 2000; Noctor et al., 2001). These cells form from neuroepithelial stem
cells in the ventricular zone, receiving extrinsic signals such as Notchl and FGF2 to suppress
neural and glial fates and allow expansion of the population (Gaiano et al., 2000; Smith et al.,
9
2006; Yoon et al., 2004). During brain formation, radial glia divide asymmetrically to form
neurons that migrate along the radial glial processes and form the cortical layers (Noctor et al.,
2001; 2004).
Oligodendrocytespecificationandfunction
In later embryonic stages, the radial glia switch from neural fates to glial fates, producing
oligodendrocyte precursors that migrate to the outer surface of the brain and spinal cord.
Oligodendrocytes were originally thought to only be formed in the ventral domains of
developing CNS structures specified by the morphogen Sonic Hedgehog (SHH) and the
transcription factors Nkx2.2, Oligl, and Olig2 (Wegner, 2008).
Deletion of oligi and olig2
genes results in loss of an entire domain of the neural tube and formation of astrocytes in place of
oligodendrocytes (Zhou and Anderson, 2002). Recent studies, however, have demonstrated that
oligodendrocytes can also derive from more dorsal domains of the neural tube. Intriguingly, the
progeny from the dorsal domains disappear at late stages of embryogenesis, suggesting only a
temporary role during development (Kessaris et al., 2006).
Oligodendrocyte progenitor cells (OPCs) derive from neural progenitors by repression of
neural fates through Notch signaling (Grandbarbe et al., 2003). Many more OPCs are formed
than are necessary, and will either form mature oligodendrocytes in the presence of survival
factors such as PDGF, or undergo programmed cell death (Barres et al., 1992).
Mature
oligodendrocytes are primarily involved in providing insulation for axons, allowing dramatic
increases in action potential propagation speed through saltatory conduction (Kaplan et al.,
1997).
High metabolic rates are necessary for creating myelin sheaths, leaving the cells
10
susceptible to oxidative damage (Juurlink et al., 1998).
As a result, oligodendrocytes are
hypothesized to be replaced regularly (Young et al., 2013).
Astrocyte specificationandfunction
During the late stages of embryonic development, the radial glia detach from the ventricular
surface and either migrate into the brain parenchyma to form astrocytes or persist in the
subventricular zone as multipotent neural stem cells (Merkle et al., 2004).
Astrocyte
differentiation is a multi-stage process that begins upon activation of Notch signaling in neural
stem cells by adjacent, newly-formed neurons expressing the Notch ligands JAGGED1 (JAGI)
and DELTA LIKE 1 (DLL1) (Namihira et al., 2009).
Activated Notch signaling triggers
chromatin remodeling and repression of genes involved in repressing astrocyte differentiation,
including neurogenin 1 (ngnl) and neurogenin 2 (ngn2), as well as expression of astrocyte genes
(Hirabayashi et al., 2009; Sun et al., 2001).
The progenitor cell, now primed to adopt an astrocyte fate, can respond to the neuronallyreleased cytokine Cardiotrophin 1 (CT-1). This factor activates the JAK/STAT pathway, leading
to transcription of astroglial genes derepressed through the actions of Notch signaling (Fan et al.,
2005; He et al., 2005). A protein involved in the unfolded protein response, Oasis, has also been
implicated in astrocyte differentiation.
Interestingly, Oasis acts through ER stress-mediated
upregulation of gcml, a gene encoding the vertebrate ortholog of the fly glia master regulator
Glial Cells Missing. Production of Gcm1 results in demethylation of promoters for astrocytic
genes (Saito et al., 2012).
11
Differentiated astrocytes are generally subdivided into two classes based on morphology
and regionalization: protoplasmic astrocytes of the gray matter and fibrous astrocytes of the
white matter (Miller and Raff, 1984). However, recent studies point to an even greater diversity
of types (Bachoo et al., 2004; Cregg et al., 2014; Freeman, 2010). How these different fates are
specified is of considerable interest; it is unknown whether these different fates are determined in
the neural tube domain from which the cells emerge or from local signals where the terminally
differentiated cells reside.
As the most abundant cell type in the mammalian brain, astrocytes perform a variety of
functions, including clearance of neurotransmitters from the extracellular environment,
formation the blood-brain barrier, and modulation of synaptic activity (Sofroniew and Vinters,
2010). These cells are noted for their long processes that either form end feet on vasculature and
neuron cell bodies to provide a blood-brain barrier, or branch into many peripheral astrocyte
processes (PAPs) that form tripartite synapses (a synapse consisting of a pre-synaptic neuron, a
post-synaptic neuron, and an astrocyte encasing the synaptic cleft) (Derouiche and Frotscher,
2001).
Near synapses, neurotransmitter transporter proteins, such as glutamate transporters
GLAST/EAAT1 and GLT-1/EAAT2, accumulate to uptake excitotoxic glutamate from the
synaptic cleft into the cytoplasm (Anderson and Swanson, 2000).
Intracellular glutamate is
converted into glutamine by the enzyme Glutamine Synthetase, a necessary process for
preventing neuronal damage (Zou et al., 2010).
Another prominent astrocyte marker is the
cytoplasmic intermediate filament Glial Fibrillary Acidic Protein (GFAP), which is particularly
useful in labeling astrocytes in a reactive state, as will be discussed below (Brenner et al., 1994;
12
Condorelli et al., 1990). Many of these astrocyte markers are not astrocyte-specific; GFAP is
expressed in radial glia, and glutamate transporters are expressed in both radial glia and
oligodendrocytes (DeSilva et al., 2009; Hartfuss et al., 2001; Shibata et al., 1997).
Ependymal Cells
Radial glia also give rise to ependymal cells, ciliated glia that comprise the barrier between
neurons and the cerebrospinal fluid of the ventricle (Bruni, 1998; Spassky et al., 2005). More
recently, however, it was suggested that ependymal cells retain stem cell-like properties (selfrenewal and differentiation) and are capable of producing neurons and astrocytes upon ischemic
injury (from oxygen deprivation) to the brain (Carldn et al., 2009; Johansson et al., 1999). These
findings are controversial as other labs claimed that these studies incorrectly identified astrocytes
as ependymal cells, and have been unable to identify proliferative ependymal cells with
ultrastructural resolution (Spassky et al., 2005). Little is currently known about ependymal cells
compared to astrocytes and oligodendrocytes, but it remains a very interesting glial cell type with
regards to participation in nervous system regeneration.
Overview of invertebrateglia
Glia have been identified in many invertebrate species and well-characterized in the two major
invertebrate
model
organisms, Drosophila melanogaster and Caenorhabditis elegans
(Hartenstein, 2011; Oikonomou and Shaham, 2011).
In fact, glia have been found in almost
every Bilaterian lineage through electron microscopy studies or via identification of glia-specific
proteins (Figure 1).
Some of the major exceptions include hemichordates, bryozoans, and
13
rotifers. In most cases, the evidence relies on the cellular morphology and spatial relationship
with neurons. Many invertebrate glia have what are described as primitive features, such as nonensheathing morphologies, which some speculate may be more similar to ancestral glial cell
types than vertebrate glia (Hartline, 2011).
C. elegans has several distinct glial cell types: socket and sheath cells that support
amphid sensory neurons, CEPsh cells that send processes to interact with synapses in the nerve
ring, and GLR cells that form connections between head muscles and motor neurons
(Oikonomou and Shaham, 2011). In addition to providing support to neurons, glia in C. elegans
are also active in the development of the nervous system by secreting axon guidance cues
(Wadsworth et al., 1996).
Drosophilaglia specificationandfunction
Drosophilaglia are the most well studied of the invertebrate glia and arguably the most relevant
in the context of understanding vertebrate glia. Four major types of fly glia have been described,
each consisting of a number of subtypes.
The surface glia are flattened cells that undergo
endocycles to increase their size and connect to one another through pleated septate junctions
(Baumgartner et al., 1996; Unhavaithaya and Orr-Weaver, 2012). These cells form a structure
analogous to the blood-brain barrier (Stork et al., 2008). Cortex glia extend lamellar processes
that wrap cortical neuron cell bodies and provide nutrient support and gas exchange (Freeman
and Doherty, 2006).
The neuropil glia are the most diverse type of Drosophila glia, assuming roles such as
axon isolation, trophic support, debris clearance, and nerve bundle fasciculation (Doherty et al.,
14
2009; Freeman and Doherty, 2006). Members of this class express glutamate transporters and
Glutamine Synthetase (Rival et al., 2004).
The last class of glia, the peripheral glia, are
associated with the peripheral nervous system and perform similar functions to the previously
mentioned types (Banerjee et al., 2006; Freeman and Doherty, 2006).
During embryonic development, Drosophila glia can be divided into two classes based
on their germ layer origin: the midline glia that derive from the mesectoderm and the lateral glia
that derive from the neurectoderm (Bossing and Technau, 1994; Coutinho-Budd and Freeman,
2013).
The transcription factor Glial Cells Missing is the master regulator of lateral glia
specification. gcm mutations result in a switch from glial fate to neural fate and overexpression
results in the opposite phenotype (Hosoya et al., 1995). Studies show that expression of Gcm
results in chromatin modification in neural stem cells that prime them for a glial fate (Flici et al.,
2011).
Thus, the involvement of Gcm in chromatin remodeling is similar to the role of
mammalian Gcm discussed above, but in fly Gcm plays a much more central role to the
specification of glia. In lateral glia, gcm expression activates two genes, reversedpolarity(repo)
and pointed (pnt), that encode transcription factors that promote glial cell fate, as well as a third
gene, tramtrack (ttk), that encodes a transcription factor that represses neural fate (Giesen et al.,
1997).
Another target of Gcm, locomotion defects (loco), encodes a regulator of G protein
signaling and is necessary for glia to form proper connections (Granderath et al., 1999). The
lateral glia cells differentiate into cortex, surface, or neuropil glia depending on which
neuroglioblast or glioblast they originated from (Hartenstein, 2011).
The midline glia follow a different developmental pathway than lateral glia, using
signaling pathways such as Notch and transcription factors such as Pointed to control
15
differentiation and function (Menne and KlAmbt, 1994; Scholz et al., 1997).
During
development, these cells control axonal midline crossing through the expression of cues such as
Slit and Netrin, a function analogous to that of the vertebrate floor plate (Mitchell et al., 1996;
Rothberg et al., 1990). At later stages, the midline glia population is split into the anterior
midline glia, specified by Notch signaling, and the posterior midline glia, specified by Hedgehog
signaling. Whereas the function of the posterior midline glia is unknown, the anterior midline
glia persist into adulthood and wrap commissural axons (Watson et al., 2011).
Evolutionary Origins of Glia
Evidencefor a common origin
A major long-standing question about the evolution of glia is whether vertebrate glia and
invertebrate glia evolved from a common ancestral glial cell type. The implications of this are
whether the study of invertebrate glia can offer insight into the function of vertebrate glia.
Although there are noted exceptions, complex features are more likely to have arisen once in
evolution and lost in extant organisms of the same lineage that lack them than to have arisen
multiple times in different branches.
Comparison between vertebrate and fly glia reveals a number of similarities in both
morphology and function that support the hypothesis for a common origin.
The surface glia
form the blood-brain barrier and cortex glia provide trophic support to neurons, which are both
functions of vertebrate astrocytes (Hartenstein, 2011). Additionally, Drosophila surface glia and
vertebrate Schwann cells (peripheral nervous system glia) adopt a flattened morphology and
16
form pleated septate junctions (Baumgartner et al., 1996; Boyle et al., 2001).
Drosophila
neuropil glia wrap around axons, a morphological characteristic of oligodendrocytes, but
function to maintain the extracellular environment, a functional characteristic of astrocytes
(Hartenstein, 2011).
The function of common genes expressed in the glia of invertebrates and vertebrates can
be used to infer the function of ancestral glia, but few genes are currently known to be shared.
Three mammalian glia proteins have been used to attempt to identify glia in other species:
Glutamine Synthetase (GS), S100 Calcium Binding Protein P (S1000), and GFAP (Hartline,
2011). The presence of Glutamine Synthetase in multiple branches of the bilateria lends credit to
the theory that glia evolved originally to have environmental regulation functions (Anderson and
Swanson, 2000).
S100p is an EF-hand calcium-binding protein that can bind to and inhibit
polymerization of GFAP, but has not been identified by immunoreactivity in Drosophila or C.
elegans (Donato, 1999; Hartline, 2011).
The use of GFAP as a glial marker in invertebrates may be unreliable because it is a
member of a cytoplasmic intermediate filament family that expanded in the vertebrate lineage
(Erber et al., 1998). Cytoplasmic intermediate filaments expressed in other cell types may crossreact with the GFAP antibody, leading to false positive results. Additionally, certain species
where glia have been conclusively identified, such as Drosophila,lack cytoplasmic intermediate
filaments altogether (Goldstein and Gunawardena, 2000). Therefore, although it is tempting to
identify glia in invertebrates based on immunoreactivity with anti-GFAP antibodies, it cannot
conclusively demonstrate the presence of a glial cell type.
17
Evidence againsta common origin
A common requirement could be the impetus for cell types with analogous functions to arise
multiple times throughout evolution.
Astrocytes and astrocyte-like cells are likely to have
evolved to fulfill extracellular environment maintenance that neurons are unable to perform. If
glutamate, an excitatory neurotransmitter, is not taken up from the synaptic cleft after release
from the pre-synaptic terminal, it will continue to evoke post-synaptic responses, cause surges in
intracellular calcium ion levels, and result in damage to the post-synaptic neuron (Manev et al.,
1989). Neurons are intrinsically poor at reuptake of glutamate because of shifting membrane
potentials that affect glutamate transporter activity and high levels of intracellular glutamate that
drive diffusion of the molecule into vesicles or out of the cell (Anderson and Swanson, 2000).
This innate inability to transport glutamate presents a need for another cell to carry out this
specialized function.
The functional and morphological characteristics shared between invertebrate and
vertebrate glia may have arisen through convergent evolution. A line of evidence that supports
this is the lack of developmental similarities, specifically the known transcription factors that
specify these cell types during development. Orthologs of DrosophilaGcm have been identified
in mammals, but are found at low levels in the nervous system and mainly participate in placenta
development (Altshuller et al., 1996; Anson-Cartwright et al., 2000). One group reports that
upregulation of gcml expression by the unfolded protein response results in demethylation of the
GFAP promoter in astrocytes, but the loss of this mechanism in mutants only delays the
differentiation of astrocytes (Saito et al., 2012). Gcm in flies is also involved in plasmocyte
differentiation during hematopoesis, which would suggest that Gcm is not glia specific (Jacques
18
et al., 2009).
These results raise the possibility that Gem is a general factor for chromatin
remodeling and was co-opted into regulating glia gene expression independently in vertebrates
and flies.
Mammalian glial specification factors are also present in Drosophila, yet are not
involved in glia formation.
The Drosophila Olig2 ortholog OR controls motor neuron axon
projection in flies rather than glia identity (Oyallon et al., 2012).
The lack of conserved
specification factors between fly and vertebrate glia suggest different origins of the two cell
types.
In order to gain a better understanding of glia evolution, not only will glia need to be
identified in more species, but they will also need to be better characterized. Comparisons of
development, gene expression profile, and function can provide better indications of homologous
cell types than morphology alone.
19
I. Adult Gliogenesis in CNS Repair and Regeneration
Reactive Astrogliosis in Vertebrates
Function of reactive astrogliosis
Terminally differentiated astrocytes outside the neurogenic zones of the brain are capable of
responding to brain injuries during a process known as reactive astrogliosis (Sofroniew, 2009).
In the brain parenchyma, resident astrocytes are normally non-proliferative and can be identified
by their expression of GLAST/EAATI and Glutamine Synthetase. Upon traumatic injury to the
brain or spinal cord, a series of events unfold that lead to the formation of an astrocyte scar
(Burda and Sofroniew, 2014) (Figure 2A).
In addition to upregulated expression of GFAP,
astrocytes near the lesion experience a short burst of proliferation within two days following
injury, as evidenced by cells labeled by BrdU and the transient expression of cell cycle markers
(Barreto et al., 2011; Buffo et al., 2008; Zamanian et al., 2012). These reactive astrocytes also
undergo hypertrophy, an increase in cell body and process size (Buffo et al., 2008; Li and
Raisman, 1995). At the lesion, reactive astrocytes deposit proteins such as Collagen Type IV,
Laminin, and Fibronectin, forming a barrier of extracellular matrix near newly formed
vasculature (McKeon et al., 1991; Silver and Miller, 2004).
Dedifferentiationof reactive astrocytes
During astrogliosis, reactive astrocytes have been shown to undergo dedifferentiation to a state
similar to neural stem cells found during development (Robel et al., 2011; Vaccarino et al.,
20
2007).
Reactive astrocytes collected following traumatic spinal cord or brain injury form
neurospheres in culture that can both self-renew and differentiate
into neurons and
oligodendrocytes (Buffo et al., 2008; Lang et al., 2004; Sirko et al., 2009). The dedifferentiated
cells activate pathways, such as Akt/PI3K and ErbB2, previously silent in astrocytes and express
radial glia markers such as Nestin, a molecular profile highly reminiscent of neural stem cells
(Feng et al., 2014; Laywell et al., 2000; Yang et al., 2011). The basic helix-loop-helix (bHLH)
transcription factor Olig2, which controls glial cell differentiation during development, is
expressed during the reactive astrocyte proliferative phase and conditional knockout of the gene
in astrocytes abrogates their proliferation (Chen et al., 2008).
Although reactive astrocytes are capable of differentiating into oligodendrocytes and
neurons in vitro, their potential is restricted to astrocytes in vivo (Buffo et al., 2008).
The
mechanisms that limit reactive astrocyte to unipotency have not yet been found, and why these
controls are in place is unclear. Regardless, the multipotency of reactive astrocytes makes them
an attractive target for in vivo stem cell-mediated regeneration of the nervous system.
Activation of reactive astrogliosis
Healthy astrocytes are activated by signals released from damaged neurons and from immune
cells that invade the lesion. The type of damage, such as from hypoxia or inflammation, results
in differential responses as indicated by different reactive astrocyte expression profiles,
suggesting that either astrocytes are capable of sensing certain types of damage or that there
exists multiple subtypes that each launch specific programs (Sirko et al., 2009; 2013; Sofroniew,
2009; Zamanian et al., 2012). Additionally, the proliferative response of astrocytes is dependent
21
on their distance from the lesion site, indicating both a mechanism for confinement of astrocyte
scar formation and for regulation of astrocytes by local signals (Barreto et al., 2011).
A number of factors have been identified as activators of reactive astrocytes. Epidermal
Growth Factor (EGF), basic Fibroblast Growth Factor (bFGF) (Laywell et al., 2000), Endothelin1 (ET-1) (Gadea et al., 2008), Tumor Necrosis Factor (TNF) (Selmaj et al., 1990), and Sonic
Hedgehog (SHH) (Sirko et al. 1990) have been demonstrated to induce neurosphere formation in
cultured astrocytes from unwounded brains. SHH is found at greater concentration in cerebral
spinal fluid (CSF) than in the brain parenchyma. Injuries that cause seepage of CSF into the
lesion result in reactive astrocyte proliferation whereas confined focal damage such as from
amyloid plaque formation does not (Sirko et al., 2013).
Another factor, Fibroblast Growth
Factor 4 (FGF4), has been shown to be released by reactive astrocytes themselves to stimulate
proliferation of nearby astrocytes and may be a mechanism for expanding astrocyte scars (Feng
et al., 2014).
The variety of extrinsic signals that are necessary and sufficient for
dedifferentiation of reactive astrocytes may be further evidence of a heterogeneous astrocyte
population.
Effect of reactive astrogliosison CNS repair
Though limited in their regenerative potential, reactive astrocytes themselves block regeneration
by preventing axons from entering the lesion (Silver and Miller, 2004). Normally the brain
parenchyma does not secrete signals that inhibit axon outgrowth (Davies et al., 1997). Damaged
neurons will attempt to regenerate lost axons by sprouting new neurites, but proximity to
astrocyte scars results in collapse and formation of a dystrophic growth cone (Houle and Jin,
22
2001; Li and Raisman, 1995). Studies found that the overall neurite outgrowth length of was
decreased for injured neurons either layered onto astrocyte scar explants or proximal to lesion
sites in vivo (McKeon et al., 1991; Rudge and Silver, 1990). These studies also found that the
level of neurite outgrowth inhibition was correlated with the recruitment of astrocytes to the
lesion, and the presence of chondroitin sulfate proteoglycans (CSPGs) and Cytotactin/Tenascin
on astrocyte processes (McKeon et al., 1991). Degradation of CSPG by Chondroitinase ABC
(ChABC) allowed extending neurites to enter the lesion, suggesting that CSPG is the main
inhibitory factor to axon outgrowth (Bradbury et al., 2002; Moon et al., 2001).
Similarly,
neurons lacking PTP, the CSPG receptor, lose sensitivity to CSPG gradients and are able to
extend axons into lesions (Fry et al., 2010; Shen et al., 2009). PTPo is functionally redundant
with the myelin-associated inhibitor receptors NgR1 and NgR3 that mediate neurite avoidance of
myelin, raising the possibility that this mechanism is employed in neurons when avoiding
oligodendrocytes as well as reactive astrocytes (Dickendesher et al., 2012).
The inhibitory nature of reactive astrogliosis towards regeneration of the nervous system
may at first appear detrimental for neural tissue, but astrocyte scar formation is essential for
preventing the primary immune response from inflicting further damage.
Ablation of all
dividing astrocytes during injury response results in increased invasion of the brain parenchyma
by leukocytes, leading to inflammation and damage to previously unaffected regions of the brain
(Bush et al., 1999; Faulkner et al., 2004; Voskuhl et al., 2009). Astrocyte scar formation thus has
beneficial effects at early stages following traumatic brain injury, but at later stages appears more
disadvantageous by preventing axon regeneration. Some hypothesize that axon growth arrest
during early phases of CNS repair is essential for preserving damaged neurons (Rolls et al.,
23
2009). Interestingly, an astrocyte's capacity to form neurospheres and to permit axon growth all
diminish with increasing age of the animal (Laywell et al., 2000; McKeon et al., 1991; Sirko et
al., 2013).
Oligodendrocyte Progenitors and Remyelination
Identification of adult oligodendrocyteprogenitors
Demyelination can occur from age or disease and results in impairment of motor and cognitive
ability (McTigue and Tripathi, 2008; Nishiyama, 2007). Restoring myelination requires new cell
production (Figure 2B).
Previous studies have shown that remyelination can occur in focal
lesions where oligodendrocytes
have been specifically ablated by antibodies
against
sphingolipids or by infection of glia with coronavirus. However, mature oligodendrocytes were
unproliferative based on lack of BrdU incorporation and were insufficient to remyelinate when
all other dividing cells were ablated by ionizing radiation (Keirstead and Blakemore, 1997;
Redwine and Armstrong, 1998).
Cells expressing the proteoglycan NG2 were found to proliferate near lesion sites and
differentiate into mature oligodendrocytes (Greenwood and Butt, 2003; Levine and Card, 1987;
Redwine and Armstrong, 1998; Watanabe et al., 2002). These cells derive from the OPCs found
in the embryo. Most OPCs specified during development in mice will differentiate into mature
oligodendrocytes within the first ten postnatal days, but some will remain in an undifferentiated,
perpetually cycling state (Reynolds and Hardy, 1997; Young et al., 2013). These cells can also
migrate into and repopulate neighboring regions of the brain where proliferative cells have been
24
ablated by ionizing radiation, suggesting a capacity to regenerate their own pool (Chari and
Blakemore, 2002).
NG2+ progenitors in adult animals are distinct from OPCs found during development
based on their slower cell cycle and proliferation rate, and are referred to as either NG2 glia or
polydendrocytes (Greenwood and Butt, 2003; Nishiyama, 2007)).
Under uninjured conditions,
NG2 glia are thought to continually proliferate to replace dying mature oligodendrocytes (Young
et al., 2013). The rate of proliferation and time required to fully differentiate both increase with
the age of the animal (Sim et al., 2002; Young et al., 2013). Myelin itself has been shown to
exert inhibitory effects on the differentiation of NG2 glia into myelinating oligodendrocytes,
raising the possibility that the age-related decrease in proliferation rates may be a result of
increased amounts of myelin debris throughout the CNS (Kotter et al., 2006).
Differentiationof NG2 glia
Upon demyelination, astrocytes secrete the mitogens PDGF and FGF2 to activate remyelination
programs in nearby NG2 glia (Frost et al., 2003; Talbott et al., 2005). The differentiation of
NG2 glia into mature myelinating oligodendrocytes is dependent on early expression of
transcription factors, including Nkx2.2 and Oligl, which leads to constitutive activation of
myelin genes such as Myelin Gene Regulatory Factor (MRF) (Koenning et al., 2012; Qi et al.,
2001; Xin et al., 2005). Olig2 is required for differentiation of NG2 glia progeny, as deletion of
the gene results in the formation of astrocytes rather than oligodendrocytes (Zhu et al., 2012).
To identify more regulators of differentiation, a study examining gene expression levels
in differentiating oligodendrocyte progenitors found the neural stem cell marker Sox2 to be
25
upregulated during initial response to demyelination and downregulated during differentiation
along with transcriptional repressors such as HesI, Hes5, Id2, and Id4. This study, interestingly,
identified, along with Oligl, two histone deacetylases (HDACs) that are upregulated during the
differentiation phase (Shen et al., 2008).
Repression of pro-neural genes and activation of
oligodendrocyte genes has since been shown with Chromatin Immunoprecipitation Sequencing
(ChIP-Seq) to be controlled epigenetically (Shen et al., 2008; Swiss et al., 2011). The SWI/SNF
complex component Brgl is recruited by Olig2 to modify chromatin at the promoters of
oligodendrocyte genes (Yu et al., 2013). Chromatin modification factors are in turn controlled
by extrinsic factors such as SHH and BMP4, which have been found to drive oligodendrocyte
and astrocyte fates, respectively, in vitro (Wu et al., 2012).
Axonal regulationof oligodendrocytes
Notably, NG2 glia have the capacity to receive instructions directly from the neurons they
myelinate. In the hippocampus, stimulation of CA3 pyramidal neurons results in depolarization
of NG2 glia in the CAI region, providing evidence that oligodendrocyte precursors can directly
communicate with neurons (Bergles et al., 2000). In vitro, activation of glutamatergic AMPA
receptors in cultured NG2 glia results in inhibition of their proliferation and differentiation
(Gallo et al., 1996). Additionally, neuronal stimulation of NG2 glia through glutamate receptors
causes translation of myelin component proteins and modification of the membrane in cultured
NG2 glia, signs of active myelination (Wake et al., 2011).
One source of glutamate that evokes excitation in NG2 glia is glutamate-filled vesicle
fusion along unmyelinated axons, which supports the hypothesis of activity-dependent
26
myelination (Ziskin et al., 2007). An in vivo experiment using optogenetics to stimulate cortical
neurons resulted in proliferation and differentiation of NG2 glia into mature oligodendrocytes as
well as increased myelin sheath thickness (Gibson et al., 2014).
The glutamatergic NMDA
receptor, on the other hand, is active in NG2 glia but its deletion does not result in any
morphological, functional or survival changes. Instead, NMDA receptors are thought to regulate
the composition of AMPA receptor populations at synapses (De Biase et al., 2011; K~rad6ttir et
al., 2005).
NG2 glia response to neuronal activity has implications for the reactive repair of
demyelination as well as possible mechanisms for modulating neural function during learning.
Stem cell propertiesof NG2 glia
The multipotency of NG2 glia is disputed. NG2 glia isolated in culture have been demonstrated
to differentiate into both oligodendrocytes and astrocytes, but not neurons, even when cultured in
neural stem cell media (Zhu et al., 2008).
A number of reports using Cre recombinase
techniques for lineage tracing have reported NG2 glia differentiating into astrocytes, neurons, or
both (Guo et al., 2009; Rivers et al., 2008; Zhu et al., 2008). Another report claimed that the
previous studies were flawed in their use of non-specific promoters to drive expression of Cre
recombinase, and that in vivo NG2 glia only generate mature oligodendrocytes (Kang et al.,
2010). Yet another report states that differentiation of NG2 cells into astrocytes occurs during
embryonic stages but not in adults (Zhu et al., 2011).
NG2 glia are at least capable of self-renewal, suggesting a stem cell capacity. Studies
using cuprizone treatment for demyelination of the corpus callosum showed depletion of the
NG2 glia pool after repeat cycles of oligodendrocyte loss and restoration, which was interpreted
27
as an inability of the cells to restore their own population (Mason et al., 2004). Other studies
using focal demyelination with lysolecithin injections found that myelination ability is not
attenuated after repeated insults (Penderis et al., 2003).
This second result is supported by the
finding that NG2 glia can repopulate regions where those cells have been lost (Chari and
Blakemore, 2002). More recently, using clonal analysis, NG2 glia were shown to undergo both
asymmetric division, forming an NG2 glial cell and a myelinating oligodendrocyte, and
symmetric division, forming either two NG2 glial cells or two oligodendrocytes (Zhu et al.,
2011). It may be that the original studies generated a non-permissive environment, such as one
filled with inhibitory myelin debris, for NG2 glia proliferation and differentiation.
Gliogenesis in Neurogenic Zones
Adult neural stem cells
Within the mature central nervous system lie two neurogenic zones, the Subependymal Zone
(SEZ) and the Subgranular Zone (SGZ), populated by adult neural stem cells (NSCs). These two
sources of neurons and glia are essential for maintaining the sensitive olfactory bulb and for
memory formation in the hippocampus. The neurogenic zones are comprised of three cell types.
The B 1 cells are slowly dividing neural stem cells that give rise to the transit amplifying C cells.
The fast dividing C cells in turn form the migrating neuroblasts known as A cells (Doetsch et al.,
1997). The SEZ neural progenitors then form a continuous chain known as the Rostral Migratory
Stream (RMS) that supplies the olfactory bulb with fresh neurons (Lois and Alvarez-Buylla,
1993; Lois et al., 1996). Although this is the primary role of the SEZ neurogenic zone, recent
28
studies have shown that the NSCs are capable of contributing to gliogenesis, particularly in
response to brain injuries (Marshall et al., 2005; Li et al., 2010).
NSCs have much in common with radial glia from developmental stages as well as with
astrocytes from postnatal stages. These cells express astrocyte genes such as GFAP, vimentin,
and S100p, and are capable of forming neurospheres in culture. The neurospheres are capable of
self-renewal and differentiation in astrocytes, oligodendrocytes, and neurons, both in vitro and in
vivo (Doetsch et al., 1999; Laywell et al., 2000). Tracing studies indicate that NSCs are capable
of differentiating in vivo into astrocytes and oligodendrocytes (Levison and Goldman, 1993;
Levison et al., 1993; Menn et al., 2006). This result was elegantly shown in clonal analysis
experiments where clusters of labeled cells stemming from one progenitor were comprised of
astrocyte and oligodendrocyte cell types (Levison and Goldman, 1993).
Specification of glia in neuralstem cells
Olig2 is essential for specification of glial cell fates in NSCs.
Cells where Olig2 is
overexpressed are prevented from adopting neural fates and entering the rostral migratory
stream, whereas cells where Olig2 is inhibited fail to differentiate into oligodendrocytes or
astrocytes (Marshall et al., 2005). The fate of the progenitors appears linked to their destination,
as NSCs contribute to oligodendrocyte populations in the white matter and both oligodendrocyte
and astrocyte populations in the gray matter (Levison and Goldman, 1993). This multipotency in
the cortical region appears reduced with increasing age (Levison et al., 1993).
NSC differentiation is also dependent on its immediate environment. Culturing NSCs
atop a feeder layer composed of neurons results in increased oligodendrocyte formation. When a
29
feeder layer of astrocytes is used, proliferation and neural differentiation are both increased
(Song et al., 2002). This effect may be because of contact-mediated inhibition of glial fates, as
reduction of Notch signaling activity by deleting both GFAP and Vimentin from astrocytes
resulted in a larger ratio of neurons formed from progenitors (Wilhelmsson et al., 2012).
Activation of glialcell production
Injury to the brain increases proliferation, migration, and differentiation of NSCs in the SEZ
(Nait-Oumesmar et al., 2007).
Following ischemia, recently proliferated cells were observed
translocating from the SEZ into the brain parenchyma while simultaneously upregulating mature
glia markers and downregulating NSC markers (Zhang et al., 2001). Progeny tracing of the
gliogenic Nestin+ cells in the SEZ shows that these cells, which normally favor astrocytic fates,
can also become oligodendrocytes and neurons in response to injury (Li et al., 2010). Normally
the B 1 cells produce neurons, but damage from demyelination is sufficient to direct more NSCs
to proceed down the oligodendrocyte lineage (Menn et al., 2006).
Perfusion of exogenous EGF into the ventricles elicits the same upregulation of cell
proliferation and differentiation as damage, suggesting that this molecule may be involved in
injury response (Craig et al., 1996; Gonzalez-Perez et al., 2009; Kuhn et al., 1997).
Interestingly, NSC cells in the hippocampal SGZ do not respond to EGF application, providing
evidence for the heterogeneity of the NSC population (Kuhn et al., 1997). FGF2 has also been
identified to regulate NSCs in culture, and can function interchangeably with EGF but stimulates
lower proliferation rates than EGF. It has been suggested that varying concentrations of EGF
and FGF2 may be used to modulate NSC kinetics in vivo (Gritti et al., 1999).
30
Glia Regeneration in Invertebrates
Adult gliogenesis in Drosophila
There is increasing evidence of glial regeneration in response to nervous system injury in
invertebrates. The two most well-characterized invertebrate systems, Drosophilaand C. elegans,
belong to the ecdysozoa, a bilaterian superphylum whose members have lost most of their
capacity to regenerate (Bely and Nyberg, 2010) (Figure 1). C. elegans lacks all somatic cell
proliferation in adulthood, but Drosophila CNS cells have been suggested to divide following
wounding (Fernindez-Hernandez et al., 2013).
Programmed neuronal cell death or damage to the antenna lobe caused proliferation of
repo+ glial cells near the wound site. This response was dependent on Eiger (Egr), a TNF
homolog, and was restricted to the first week of adulthood (Kato et al., 2009). The interface glia,
a class of neuropil glia, are normally capable of dividing during adulthood but remain in GI from
Prospero-mediated inhibition of CycE. Upon injury, Egr promotes expression of Dorsal in the
interface glia and, together with Notch signaling, results in reactivation of the cell cycle. Dorsal
also promotes expression of Prospero, which accumulates and causes the cell to undergo
differentiation. The newly formed glia close the wound site and repair the neuropil (Kato et al.,
2011).
A number of interesting parallels are drawn between the Drosophila and vertebrate
responses to injury. Egr is a homolog of TNF, which is involved in reactive astrogliosis, and
Prospero is a homolog of Proxl, a neural stem cell factor that promotes cell cycle exit and
differentiation (Karalay et al., 2011; Kato et al., 2011; Selmaj et al., 1990).
31
Examples of glia regenerationin other invertebrates
Damage to the nervous system results in the recruitment of blood cells to the lesion site in insects
and molluscs (Bale et al., 2001; Howes et al., 1989). In cockroaches, loss of the surface glia that
form the blood brain barrier causes an accumulation of haemocytes at the lesion site (Treherne et
al., 1984). These haemocytes provide a temporary barricade while nearby glia proliferate to
replace the lost surface cells (Smith et al., 1987; 1990). This process parallels the formation of
the astrocyte scar, where leukocytes invade the lesion site early while astrocytes proliferate to
form a barrier.
In the leech Hirudo medicinalis, damage to the nervous system results in almost
immediate expression of nitric oxide synthesis genes in ensheathing glia (Shafer et al., 1998).
Because nitric oxide can act as a neurotransmitter, this response may be a means of signaling to
nearby neurons. However, ablation of ensheathing glia cells does not affect the regeneration of
the axon bundles after transection, suggesting that the glia do not participate in axon guidance
(Elliot and Muller, 1982). When crayfish axon bundles are severed, the axons sprout additional
processes that break down the bordering membranes of adjacent glia and engulf their nuclei. The
function of this syncytium is proposed to be for large scale protein transfer necessary to support
the regenerating axon (Pearce et al., 2003).
Whereas some aspects of invertebrate glia regeneration are reminiscent of vertebrate
processes, others appear different.
Surveying more invertebrate species is necessary to
determine which regenerative features are ancestral to all Bilateria and how studying them in a
32
non-mammalian organism can lead to insights into nervous system repair and regeneration in
vertebrate species.
33
III. Planaria as a Model for Glial Cell Biology
Planarian Regenerative Potential
Overview of the model system
The planarian Schmidtea mediterranea is a free-living, freshwater flatworm that has recently
emerged as a model system for the study of many aspects of regeneration biology, such as axis
polarity determination, tissue patterning, and stem cells.
Planarians are a member of the
Lophotrochozoan superphylum, an understudied branch of the Bilateria and sister group to the
Ecdysozoans, which includes Drosophila and C. elegans (Adoutte et al., 2000).
Lophotrochozoans
and Ecdysozoans
constitute the Protostomia,
Together,
the sister group
to
Deuterostomia, which includes all vertebrate model organisms. Two strains of S. mediterranea
are widely used: a hermaphroditic strain that is capable of sexual reproduction by crossfertilization, and an asexual strain that derived from the sexual strain following a chromosomal
translocation that prevents germ line formation (Newmark and SAnchez Alvarado, 2002).
Gene expression can be easily and specifically inhibited in planarians by introducing
double stranded RNA, either through injection directly into the intestinal tract or by feeding the
animals bacteria expressing double stranded RNA (dsRNA) (Newmark et al., 2003; Reddien et
al., 2005a; Sanchez Alvarado and Newmark, 1999).
Whole-mount fluorescent RNA in situ
hybridization allows the spatial localization of gene expression in the context of the entire
animal, providing a powerful technique to examine changes resulting from inhibition of target
genes (Pearson et al., 2009; Umesono et al., 1997).
34
Planariananatomy
Although planarian anatomy is simple compared to vertebrate model systems, a number of tissue
types have been identified with homologous functions and specification mechanisms to these in
other model systems. Externally, planarians have an epidermis that is ciliated on the ventral
surface to allow for motility. Beneath the epidermis lies a pigmented epithelium, and then a
muscular layer formed from a regular network of longitudinal, circular, and diagonal muscle
fibers.
The animal ingests food through muscular pharynx located in the midbody that is
extruded through a mouth on the ventral surface. The pharynx is connected to an intestinal tract
that splits into three primary branches and many secondary branches extending into almost every
region of the animal. Between the gut and the muscle layer is the parenchyma, mesenchymal
tissue consisting of many cell types (Hyman, 1940).
At the dorsal surface in the anterior are two light-sensing organs known as photoreceptors
that inform the animal's light-avoidance behavior (Carpenter et al., 1974). The photoreceptor
neurons extend axon bundles to the central nervous system (CNS). The planarian CNS consists
of a pair of ventrally-localized cephalic ganglia, each comprised of a cortex of neuron cell bodies
fully encompassing a neuropil of neurites (Morita and Best, 1965) (Figure 3A). Each cephalic
ganglion lobe lies atop a ventral nerve cord that extends from the anterior to the poster. The
majority of synapses lie within the neuropil (Morita and Best, 1966) (Figure 3B). Distributed
throughout the ventral surface is a grid-like network of axon bundles known as the orthogon.
Commissural axon bundles emerge perpendicularly from the nerve cords and either extend
medially to connect the two ventral nerve cords or laterally towards the periphery (Reuter et al.,
35
1998). Neurons are most concentrated in the cortical regions of the cephalic ganglia and nerve
cords but are also found distributed throughout the body and in the pharynx.
Planarianregeneration
Planarians can heal almost any injury without scar formation or permanent changes to normal
morphology (Morgan, 1898). Almost any piece of the animal, when separated from the body,
can regenerate all missing portions and form a new individual. The only regions incapable of
regenerating when isolated are the pharynx and the tip of the head, for reasons discussed below.
When significant portions of the body are amputated, the remaining body must both generate
new tissue to replace what was lost (epimorphosis) and reorganize old tissue to accommodate the
now smaller size of the animal (morphallaxis) (Morgan, 1898). The blastema, an unpigmented
outgrowth of newly formed cells emerging from the amputation site, is a common feature in
other regenerative species and is useful for studying formation and differentiation of cells.
Pieces of an original animal are capable of fully regenerating all missing structures in a week and
retain the capacity to regenerate following subsequent wounding.
Planarian Stem Cells
Characterizationofplanarianstem cells
Proliferative cells can be identified through RNA in situ hybridization by their expression of
smedwi-1, a gene encoding an ortholog of Piwi ortholog (Reddien et al., 2005b). This marker
has been confirmed as a proliferation mark by overlapping expression of S phase marker histone
36
H2B (Newmark and Sanchez Alvarado, 2000) and M phase marker anti-phosphorylated Histone
H3 (Wenemoser and Reddien, 2010), as well as by expression of smedwi-J in cells with >2c
DNA content isolated by FACS (Hayashi et al., 2006). When ionizing radiation is used to ablate
all dividing cells, planarians can survive for up to four weeks but are unable to regenerate
following amputation (Reddien and Sinchez Alvarado, 2004).
These data suggest that the
source of new tissue during regeneration is the population of all proliferative cells in the adult
animal; this population is known as the neoblasts.
Neoblasts are distributed throughout the
parenchyma of the body except in the pharynx and at the tip of the head, which explains why
those two regions are unable to regenerate on their own.
Neoblasts are dynamic cells that are capable of differential responses to wounding.
During homeostasis in unwounded animals, neoblasts undergo mitosis at a low rate to replace
cells lost as a consequence of normal tissue turnover (Pellettieri et al., 2010; Sal6 and BaguftA,
1984). If the animal is injured without tissue loss, such as from a stab, a wave of mitoses is
observed throughout the animal six hours after wounding.
If the animal suffers an injury that
results in lost tissue, the six-hour mitotic peak is followed by a second wave of mitoses localized
to the injury site at 48 hours after wounding (Wenemoser and Reddien, 2010). In addition to
proliferating in response to wounding, neoblasts also migrate towards sites of injury
(Guedelhoefer and Sanchez Alvarado, 2012). Descendants of proliferating neoblasts can be
identified for a short period of time after dividing by immunofluorescence of SMEDWI-1
protein, which perdures after smedwi-J is downregulated during differentiation (Wenemoser and
Reddien, 2010). When combined with fluorescent in situ hybridization for differentiated cell
markers, whether differentiation occurs can be determined (Figure 4A).
37
Neoblastpotential
The term neoblast refers to all proliferative cells in the animal, which is likely a heterogeneous
population.
The irradiation sensitive cell population consists of stem cells as well as their
immediate descendants, identified by markers prog-1 and agat-1 (Eisenhoffer et al., 2008). In
the first experiment that examined neoblasts on a single cell level, qPCR of individual FACSisolated neoblasts demonstrated that subsets of the population expressed differentiated cell
markers (Hayashi et al., 2010). Finding heterogeneity in the neoblast compartment has raised the
question of whether there exists a pluripotent stem cell that is capable of differentiating into any
cell type, or whether multiple lineage-restricted stem cell types must work in concert to power
regeneration (Figure 4B).
In one experiment, a large number of cells collected from a donor animal were
transplanted into an irradiated host animal that lacked all endogenous neoblasts.
The
transplantation was able to restore regenerative abilities in the host animal, suggesting that the
transplanted cells were capable of generating all differentiated tissue types in the adult animal
(BagufiA et al., 1989).
These studies, however, were insufficient in examining the stem cell
potential of individual cells. Therefore, the question of whether a pluripotent stem cell type
exists in planarians was left unanswered.
38
Evidence of Planarian Glia
Ultrastructuralidentification ofglia
Electron microscopy studies have provided evidence for planarian glia in the neuropil of the
central nervous system.
Originally described as accessory cells, these structures are
distinguishable from neurons and secretory cells based on their relatively clear cytoplasm
(Morita and Best, 1966).
The low abundance of vesicles and granules has led some to
hypothesize that the primary role of these cells is mechanical support, but the presence of
glycogen granules suggests additional roles in providing nutrient support to neighboring neurons
(Golubev, 1988). These studies provide no functional or molecular evidence that validate these
cells as glia or allow comparisons with established glia in other organisms, but provide a
foundation on which future studies of planarian glia may be conducted.
Candidateglia cells by marker localization
Although the expression patterns for many genes have been determined, currently only
one has been shown in cells localized to the neuropil, the region where glia are hypothesized to
reside.
This gene, netrin 2 (net2), encodes a secreted ligand that binds to Netrin receptors
expressed in neurons, and when its expression is inhibited by RNAi, causes disrupted axon
guidance and defective brain patterning (CebriA and Newmark, 2005). The axon guidance cue
Netrin is expressed in glia in C. elegans (Wadsworth et al., 1996), the floor plate glia of
vertebrates (Serafini et al., 1994), and the midline glia of Drosophila (Mitchell et al., 1996),
suggesting that the planarian net2 may be a marker for candidate glial cells.
39
Slit, another secreted axon guidance factor, is expressed in Drosophila midline glia
(Rothberg et al., 1990) and the floor plate and roof plate of the vertebrate neural tube (Holmes et
al., 1998), and the gene encoding the planarian ortholog is expressed in a ventral midline stripe
lying between the cephalic ganglia lobes (Cebrii et al., 2007). Inhibition of slit in planarians
results in midline defects including cyclopic photoreceptors and failed separation between the
brain lobes (CebriA et al., 2007). The localization and function of the slit+ cells is reminiscent of
the midline of both Drosophilaand vertebrates, raising the possibility that the cells within this
domain are glia.
Potentialglia-associatedgenes
A number of genes encoding orthologs of proteins involved in glia specification in other systems
has been described in planarians, but none of these genes have been examined in the context of
planarian glia specification.
The basic helix-loop-helix (bHLH) family includes many
transcription factors involved in determining neural versus glial cell fates, including Olig2 and
Hairy/Enhancer of Split (Nieto et al., 2001; Zhou and Anderson, 2002). Many members of the
bHLH family in planarians are expressed in the nervous system and result in cephalic ganglia
patterning defects upon inhibition (Cowles et al., 2013).
Inhibition of planarian BMP4-J, which encodes an ortholog of a factor involved in OPC
fate specification (Wu et al., 2012), results in ectopic photoreceptors in addition to defects in
regeneration of lateral domains of the body (Reddien et al., 2007). EGFs have been implicated
in reactive astrogliosis and NSC proliferation, and three genes encoding orthologs have been
identified in planarians.
In addition to its expression in pharynx and neoblasts, egfr-3 is
40
expressed at the outer edge of the cephalic ganglia lobes, and its inhibition results in failed
differentiation of neuronal cell types in the anterior blastema (Fraguas et al., 2011). FGF is an
important mitogen for maintaining stem cell characteristics and promoting proliferation in
astrocytes and neural stem cells (Feng et al., 2014; Gritti et al., 1999).
nou-darake (ndk), a planarian gene encoding a protein similar to FGF receptors, is
expressed in the nervous system.
Posterior expansion of the cephalic ganglia with ectopic
photoreceptors is observed upon inhibition of ndk, suggesting that its role is to restrict the
cephalic ganglia to the anterior domain (CebriA et al., 2002). The genes mentioned here encode
proteins with roles in nervous system regeneration in planarians and may be involved in
regulating glial cell biology as well, but without a reliable marker for glial cells, either by in situ
hybridization or by immunofluorescence, those roles cannot be assessed.
There remain many other signaling pathways with demonstrated roles in Drosophilaand
vertebrate glia development and biology that either have not been described as yet or appear to
have no role in nervous system regeneration. Notch signaling controls neuron versus glia fate
choices through contact inhibition in both developing glia and oligodendrocyte progenitors
(Gaiano et al., 2000; Grandbarbe et al., 2003; Wilhelmsson et al., 2012). Aside from studies of
the role of the downstream bHLH transcription factor-encoding gene hesl-3 in nervous system
regeneration, Notch signaling in planarians has not been described in the literature.
The Hedgehog signaling pathway has known roles in planarians, but none described yet
for nervous system regeneration (Rink et al., 2009) (Figure 5A).
Inhibition of Hedgehog
pathway components results in anterior-posterior polarity defects during regeneration (Rink et
al., 2009; Yazawa et al., 2009) (Figure 5B). Interestingly, hedgehog is expressed in cells in the
41
nervous system adjacent to the neuropil, suggesting that it may have a previously unexplored
role in neural or glial cell specification (Rink et al., 2009) (Figure 5C). Given that Hedgehog
signaling is involved in oligodendrocyte specification (Wegner, 2008) and reactive astrocyte
activation (Sirko et al., 2013) in vertebrates and posterior midline glia specification in
Drosophila(Watson et al., 2011), it is an interesting pathway to investigate for potential roles in
regulation of planarian glia biology.
Evidence for a glial cell type in planarians would open the field to the study of glia in a
highly regenerative organism. A comparison of mechanisms, both cell autonomous and system
wide, between mammalian glia and planarian glia may offer insight into how and why
mammalian glia evolved to limit regenerative potential. The stem cell potential of the organism
offers a unique perspective of observing the complete developmental pathway of the cell in an
adult animal, from naive to terminally differentiated. Additionally, continued characterization of
glia in other species will contribute to our understanding of the homology between vertebrate and
invertebrate glia types and also the origins of this complex and vital cell type.
42
IV. Content Overview
Determining the source of specialized cell types, such as glia, during planarian regeneration is
integral to understanding how they are specified and whether their differentiation has similarities
to developmental mechanisms in other species. Planarian neoblasts are the source of new tissue
during regeneration, but the pathways by which specified cells are determined and whether pools
of progenitors exist to repopulate depleted differentiated cell types are largely unknown. For
example, do planarian glia arise from a unipotent progenitor population that can be depleted,
similarly to newly formed astrocytes during reactive astrogliosis, or do they stem from more
potent cells that contribute to other cell types, such as all specialized cells during embryonic
development? In order to gain a better understanding of how planarian glia are regenerated and
contribute to regeneration, the origin of these cells must be identified.
Once uncovered,
questions pertaining to how new planarian glia are made throughout adulthood and whether the
system for replenishing these cells is similar to mammalian mechanisms can be asked.
Chapter 2 of this thesis demonstrates the existence of a pluripotent stem cell type within
the neoblast population. These cells, called clonogenic neoblasts, were shown to be capable of
self-renewal and differentiation using sublethal irradiation or single cell transplantation colony
assays. Additionally, a single transplanted clonogenic neoblast was demonstrated to be sufficient
to restore regeneration in animals lacking all endogenous neoblasts. These results suggest that,
within the adult animal, all specialized cells can arise from a single cell type, which expands our
understanding of how terminally differentiated cells such as glia are regenerated after injury.
With this knowledge, the formation of specialized cell types can begin to be elucidated.
43
In
addition to answering a long-standing question in the planarian field and providing a basis for
continued study of basic regenerative mechanisms and stem cell biology, this discovery presents
an opportunity to dissect the differentiation mechanisms of planarian glial cells during
regeneration and compare the process to macroglia development or injury response in
mammalian systems.
To further investigate the function of planarian glia and their similarity to glial cells of
other species, genes with known function in glia cells and pathways with known roles in glia
specification in other model systems were investigated in planarians. Molecular characteristics
are important for establishing cell types, deducing function, and finding homology with similar
cell types in other species. With these data, arguments can be made concerning the evolutionary
origins of glial cells. Furthermore, it would aid analysis to look for similar functions shared
between invertebrate and vertebrate glia, especially in the context of wound response and
regeneration.
The Hedgehog signaling pathway is involved in both the specification of glia
during development and the modulation of astrocyte activity in mammals and, therefore, is an
ideal target for perturbation in order to characterize putative glial cell types.
Chapter 3 of this thesis provides molecular evidence for the existence of planarian glia
and demonstrates that gene expression in these cells is regulated by Hedgehog signaling from
adjacent neurons.
The identification of planarian glia by both marker expression and
morphology builds upon ultrastructural studies done in the past as well as provide a foundation
on which comparisons of planarian glia can be made with glia of other model systems. These
data contribute to our understanding whether planarian glia share a common evolutionary origin
with mammalian glia, and thereby whether roles during regeneration and nervous system repair
44
are likely to be similar. Additionally, the involvement of Hedgehog signaling in gene expression
in these terminally differentiated cells shows surprising similarity to a mechanism by which
neurons communicate with nearby astrocytes using Sonic hedgehog. This conserved role of
Hedgehog points to the possibility of an ancestral function of the pathway shared between
deuterostomes and protostomes.
This new evidence for planarian glia creates a number of
opportunities to study specification and development of this cell type, its role in regeneration of
the nervous system, and the evolution of glia.
45
Figure 1
M0
IF
UI
Ctenophora
SMC3
Placozoa
ND
Cnidaria
NO
Acoelomorpha
ND
Rotifera
ND
Platyhelminthes MDN
Gastrotricha
-U
Bryozoa
ND
Entoprocta
-UC
Cycliophora
-U
Annelida
DD
Mollusca
-U
Nemertia
-N0
Brachiopodia
-U
Phoronida
DD
Nematoida
-U
Tardigrada
ND
Onychophora
OU
Arthropoda
Hemichordata
Echinodermata
Cephalochordata NED
Urochordata
-U
Craniata
46
Regeneration Key
No Data
No Regeneration
D3 Limb
Segment
Whole Body
r0
0
OR
0
Nk
m
0
N
0
(D
0
(A
0)
Glia Key
D:
N
No Data
No Glia
Primitive Glia
Developed Glia
Figure 1. Presence or absence of regeneration and glia in the Metazoa.
Phylogenetic tree of the Metazoa with Bilaterian superphyla (Deuterostomia, Ecdysozoa, and
Lophotrochozoa) noted on right side. Regenerative potential of each taxa (based on highest
regenerative ability of a single species within the taxon) is classified based on no regeneration of
body structures (red), regeneration of limbs (yellow), regeneration of segments of the primary
body axes (green), and regeneration of whole bodies from small pieces (blue) (Bely and Nyberg,
2010).
Presence of glia-like cells in each taxa (based on closest morphological similarity to
mammalian glia of a single species within the taxon) is classified by absence of glia (red),
presence of primitive non-ensheathing glia (yellow), and well-developed glia (green) (Hartline,
2011).
47
Figure 2
A
Reactive Astrogliosis
B
Remyelination
Neuron
0
0
Astrocyte
0
OL
NG2 Glia
*
V
z
~JNG2-
GLAST+
GS+
EGF/FGFI
0
t
TNF/SHH
PDGFaR+
Myelin
PDGFFGF2
;IWo
0)
E
gA2+
Oligl+
Sox2+
G4G
2utamate
0
LU
0 0
EL I..
0
a
*
Oli401g2+
I
I
-
0
0
0L
0.
0
w
MRF+
_______
Collagen / Fibronectin / CSPG
48
MRF+
Figure 2. Overview of reactive astrogliosis and remyelination.
(A) Astrocytes upregulate expression of GFAP and Olig2 in response release of EGF, FGF,
TNF, of SHH from damaged portions of the nervous system. Activated astrocytes proliferate
and secrete FGF4 to promote proliferation in nearby astrocytes. Reactive astrocytes accumulate
and seal the wound site by secretion of extracellular matrix proteins collagen, fibronectin and
CSPG.
(B) Myelin represses proliferation and differentiation in NG2 glia.
Demyelination
induces response in astrocytes, which in turn secrete PDGF and FGF2 to activate Nkx2.2, Oligl
and Sox2 expression in NG2 glia. After proliferation, immature oligodendrocytes express Olig2
to begin differentiation and MRF to synthesize myelin proteins.
Depolarization of immature
oligodendrocytes by unmyelinated axons triggers the formation of the myelin sheath.
49
Figure 3
A
B
Dc2 mRNA
Synapsin Protein
DAPI
syn mRNA Synapsin Protein
50
Figure 3. Localization of cell bodies and synapses in the planarian nervous system.
(A) Fluorescent in situ hybridization (FISH) of pan-neuronal marker prohormone convertase 2
(pc2) (green) (Collins et al., 2010) with DAPI for nuclei (blue). Overview image (left) shows
cephalic ganglia lobes (CG) and ventral nerve cords (VNC). Detail image (right) shows cephalic
ganglia with DIC.
Cell bodies labeled by FISH are localized to the cortex (ctx), which
encompasses the cell body-sparse neuropil (np).
(B) Immunofluorescence (IF) for synapse-
associated protein Synapsin (magenta) and FISH for synapsin (syn) transcript (green, right
images only). Overview image (left) shows high concentration of synapses in the neuropil of the
CG and VNC as well as smaller clusters throughout the ventral parenchyma. Inset (i) (top right)
shows detail of cephalic ganglia, with no overlap between Synapsin transcript and protein. Inset
(ii) (bottom right) shows detail of ventral nerve cords. Anterior up, ventral surface shown for all.
Scale bars: I00um for A and left panel of B; 50um for right panels of B.
51
Figure 4
A
0)
4L
0)
C
smedwi-1+
SMEDWI 4
smedw-1+
SMEDWI-1+
C.
0
Neoblast
smedwi-1+
SMEDWI 1+
SMEDWI- i
SMEDWI-1+
marker+
Lineage Restricted
marker+
MARKER+
Pluripotent
Ii,
*
0
*
B
Differentiated Cell
I
I
52
4K
Figure 4. Planarian neoblast stem cell characteristics and potential.
(A) Self-renewal of neoblasts can be traced by detection of either smedwi-J mRNA by in situ
hybridization or SMEDWI-1 protein by immunofluorescence.
Differentiation of neoblasts can
be traced by detection of overlapping expression of SMEDWI- 1 protein, which perdures after
smedwi-J transcription ceases, and differentiated cell markers. (B) Lineage restricted model and
pluripotent model of neoblast population potential. In the lineage restricted model, the neoblast
population contains multiple subpopulations, each of which can only generate specific cell types.
In the pluripotent model, the neoblast population contains a pluripotent stem cell type capable of
giving rise to all differentiated cell types in the adult animal.
53
Figure 5
B
C
control RNAI
hh RNAi
ptc RNAI
hh chat
54
ptc RNAI
Figure 5. The Hedgehog signaling pathway in planaria.
(A) Simplified schematic representing major Hedgehog signaling pathway components studied
in planaria. Hedgehog (Hh), a secreted ligand, binds and inhibits the transmembrane receptor
Patched (Ptc). Inhibition of Ptc relieves inhibition of the transmembrane protein Smoothened
(Smo). Activated Smo inhibits cytoplasmic protein Suppressor of Fused (Sufu). Inhibition of
Sufu releases sequestration of Gli-1 transcription factor. (B) Images of anterior and posterior
blastemas of live animals following Hedgehog signaling perturbation and six days of
regeneration. Right-most image shows severe ptc RNAi phenotype. Red dotted lines delineate
approximate amputation plane.
Red arrowhead indicates fused photoreceptors.
(C) Double
FISH of hh (red) and neuronal marker choline acetyltransferase(chat) (green). Co-expression is
observed in the medial cortex of the cephalic ganglia.
55
References
Adoutte, A., Balavoine, G., Lartillot, N., Lespinet, 0., Prud'homme, B., and de Rosa, R. (2000).
The new animal phylogeny: reliability and implications. Proc. Natl. Acad. Sci. U.S.a. 97,44534456.
Altshuller, Y., Copeland, N.G., Gilbert, D.J., Jenkins, N.A., and Frohman, M.A. (1996). GcmI, a
mammalian homolog of Drosophilaglial cells missing. FEBS Lett. 393, 201-204.
Anderson, C.M., and Swanson, R.A. (2000). Astrocyte glutamate transport: review of properties,
regulation, and physiological functions. Glia 32, 1-14.
Anson-Cartwright, L., Dawson, K., Holmyard, D., Fisher, S.J., Lazzarini, R.A., and Cross, J.C.
(2000). The glial cells missing-i protein is essential for branching morphogenesis in the
chorioallantoic placenta. Nat. Genet. 25, 311-314.
Bachoo, R.M., Kim, R.S., Ligon, K.L., Maher, E.A., Brennan, C., Billings, N., Chan, S., Li, C.,
Rowitch, D.H., Wong, W.H., et al. (2004). Molecular diversity of astrocytes with implications
for neurological disorders. Proc. Natl. Acad. Sci. U.S.a. 101, 8384-8389.
BagufiA, J., Sal6, E., and Auladell, C. (1989). Regeneration and pattern formation in planarians.
III. that neoblasts are totipotent stem cells and the cells. Development (Cambridge, England) 7786.
Bale, S.D., Howard, T.A., and Moffett, S.B. (2001). Neuronal and non-neuronal responses to
nerve crush in a pulmonate snail, Melampus bidentatus. Invert. Neurosci. 4, 105-117.
Banerjee, S., Pillai, A.M., Paik, R., Li, J., and Bhat, M.A. (2006). Axonal ensheathment and
septate junction formation in the peripheral nervous system of Drosophila.Journal of
Neuroscience 26, 3319-3329.
Barres, B.A., Hart, I.K., Coles, H.S., Burne, J.F., Voyvodic, J.T., Richardson, W.D., and Raff,
M.C. (1992). Cell death and control of cell survival in the oligodendrocyte lineage. Cell 70, 3146.
Barreto, G.E., Sun, X., Xu, L., and Giffard, R.G. (2011). Astrocyte proliferation following stroke
in the mouse depends on distance from the infarct. PLoS ONE 6, e2788 1.
Baumgartner, S., Littleton, J.T., Broadie, K., Bhat, M.A., Harbecke, R., Lengyel, J.A., ChiquetEhrismann, R., Prokop, A., and Bellen, H.J. (1996). A Drosophila neurexin is required for
septate junction and blood-nerve barrier formation and function. Cell 87, 1059-1068.
Bely, A.E., and Nyberg, K.G. (2010). Evolution of animal regeneration: re-emergence of a field.
Trends Ecol. Evol. (Amst.) 25, 161-170.
Bergles, D.E., Roberts, J.D., Somogyi, P., and Jahr, C.E. (2000). Glutamatergic synapses on
56
oligodendrocyte precursor cells in the hippocampus. Nature 405, 187-191.
Bossing, T., and Technau, G.M. (1994). The fate of the CNS midline progenitors in Drosophila
as revealed by a new method for single cell labelling. Development (Cambridge, England) 120,
1895-1906.
Boyle, M.E., Berglund, E.O., Murai, K.K., Weber, L., Peles, E., and Ranscht, B. (2001).
Contactin orchestrates assembly of the septate-like junctions at the paranode in myelinated
peripheral nerve. Neuron 30, 385-397.
Bradbury, E.J., Moon, L.D.F., Popat, R.J., King, V.R., Bennett, G.S., Patel, P.N., Fawcett, J.W.,
and McMahon, S.B. (2002). Chondroitinase ABC promotes functional recovery after spinal cord
injury. Nature 416, 636-640.
Brenner, M., Kisseberth, W.C., Su, Y., Besnard, F., and Messing, A. (1994). GFAP promoter
directs astrocyte-specific expression in transgenic mice. J. Neurosci. 14, 1030-1037.
Bruni, J.E. (1998). Ependymal development, proliferation, and functions: a review. Microsc.
Res. Tech. 41, 2-13.
Buffo, A., Rite, I., Tripathi, P., Lepier, A., Colak, D., Horn, A.-P., Mori, T., and G~tz, M. (2008).
Origin and progeny of reactive gliosis: A source of multipotent cells in the injured brain.
Proceedings of the National Academy of Sciences 105, 3581-3586.
Burda, J.E., and Sofroniew, M.V. (2014). Reactive gliosis and the multicellular response to CNS
damage and disease. Neuron 81, 229-248.
Bush, T.G., Puvanachandra, N., Homer, C.H., Polito, A., Ostenfeld, T., Svendsen, C.N., Mucke,
L., Johnson, M.H., and Sofroniew, M.V. (1999). Leukocyte infiltration, neuronal degeneration,
and neurite outgrowth after ablation of scar-forming, reactive astrocytes in adult transgenic mice.
Neuron 23, 297-308.
Carldn, M., Meletis, K., G6ritz, C., Darsalia, V., Evergren, E., Tanigaki, K., Amendola, M.,
Barnab&Heider, F., Yeung, M.S.Y., Naldini, L., et al. (2009). Forebrain ependymal cells are
Notch-dependent and generate neuroblasts and astrocytes after stroke. Nat. Neurosci. 12, 259267.
Carpenter, K.S., Morita, M., and Best, J.B. (1974). Ultrastructure of the photoreceptor of the
planarian Dugesiadorotocephala.Cell Tissue Res 148.
Cebri&, F., and Newmark, P.A. (2005). Planarian homologs of netrin and netrin receptor are
required for proper regeneration of the central nervous system and the maintenance of nervous
system architecture. Development (Cambridge, England) 132, 3691-3703.
Cebria, F., Guo, T., Jopek, J., and Newmark, P.A. (2007). Regeneration and maintenance of the
planarian midline is regulated by a slit orthologue. Developmental Biology 307, 394-406.
57
CebriA, F., Kobayashi, C., Umesono, Y., Nakazawa, M., Mineta, K., Ikeo, K., Gojobori, T., Itoh,
M., Taira, M., Sinchez Alvarado, A., et al. (2002). FGFR-related gene nou-darake restricts brain
tissues to the head region of planarians. Nature 419, 620-624.
Chari, D.M., and Blakemore, W.F. (2002). Efficient recolonisation of progenitor-depleted areas
of the CNS by adult oligodendrocyte progenitor cells. Glia 37, 307-313.
Chen, Y., Miles, D.K., Hoang, T., Shi, J., Hurlock, E., Kernie, S.G., and Lu, Q.R. (2008). The
basic helix-loop-helix transcription factor olig2 is critical for reactive astrocyte proliferation after
cortical injury. Journal of Neuroscience 28, 10983-10989.
Collins, J.J., III, Hou, X., Romanova, E.V., Lambrus, B.G., Miller, C.M., Saberi, A., Sweedler,
J.V., and Newmark, P.A. (2010). Genome-Wide Analyses Reveal a Role for Peptide Hormones
in Planarian Germline Development. PLoS Biol 8, e1000509.
Condorelli, D.F., Dell'Albani, P., Kaczmarek, L., Messina, L., Spampinato, G., Avola, R.,
Messina, A., and Giuffrida Stella, A.M. (1990). Glial fibrillary acidic protein messenger RNA
and glutamine synthetase activity after nervous system injury. J. Neurosci. Res. 26, 251-257.
Coutinho-Budd, J., and Freeman, M.R. (2013). Probing the enigma: unraveling glial cell biology
in invertebrates. Current Opinion in Neurobiology.
Cowles, M.W., Brown, D.D.R., Nisperos, S.V., Stanley, B.N., Pearson, B.J., and Zayas, R.M.
(2013). Genome-wide analysis of the bHLH gene family in planarians identifies factors required
for adult neurogenesis and neuronal regeneration. Development (Cambridge, England) 140,
4691-4702.
Craig, C.G., Tropepe, V., Morshead, C.M., Reynolds, B.A., Weiss, S., and van der Kooy, D.
(1996). In vivo growth factor expansion of endogenous subependymal neural precursor cell
populations in the adult mouse brain. J. Neurosci. 16, 2649-2658.
Cregg, J.M., DePaul, M.A., Filous, A.R., Lang, B.T., Tran, A., and Silver, J. (2014). Functional
regeneration beyond the glial scar. Exp. Neurol. 253, 197-207.
Davies, S.J., Fitch, M.T., Memberg, S.P., Hall, A.K., Raisman, G., and Silver, J. (1997).
Regeneration of adult axons in white matter tracts of the central nervous system. Nature 390,
680-683.
De Biase, L.M., Kang, S.H., Baxi, E.G., Fukaya, M., Pucak, M.L., Mishina, M., Calabresi, P.A.,
and Bergles, D.E. (2011). NMDA receptor signaling in oligodendrocyte progenitors is not
required for oligodendrogenesis and myelination. Journal of Neuroscience 31, 12650-12662.
Derouiche, A., and Frotscher, M. (2001). Peripheral astrocyte processes: monitoring by selective
immunostaining for the actin-binding ERM proteins. Glia 36, 330-341.
DeSilva, T.M., Kabakov, A.Y., Goldhoff, P.E., Volpe, J.J., and Rosenberg, P.A. (2009).
58
Regulation of glutamate transport in developing rat oligodendrocytes. Journal of Neuroscience
29, 7898-7908.
Dickendesher, T.L., Baldwin, K.T., Mironova, Y.A., Koriyama, Y., Raiker, S.J., Askew, K.L.,
Wood, A., Geoffroy, C.G., Zheng, B., Liepmann, C.D., et al. (2012). NgR1 and NgR3 are
receptors for chondroitin sulfate proteoglycans. Nat. Neurosci. 15, 703-712.
Doetsch, F., Caill6, I., Lim, D.A., Garcia-Verdugo, J.M., and Alvarez-Buylla, A. (1999).
Subventricular zone astrocytes are neural stem cells in the adult mammalian brain. Cell 97, 703716.
Doetsch, F., Garcia-Verdugo, J.M., and Alvarez-Buylla, A. (1997). Cellular composition and
three-dimensional organization of the subventricular germinal zone in the adult mammalian
brain. J. Neurosci. 17, 5046-5061.
Doherty, J., Logan, M.A., Tagdemir, O.E., and Freeman, M.R. (2009). Ensheathing glia function
as phagocytes in the adult Drosophilabrain. Journal of Neuroscience 29, 4768-4781.
Donato, R. (1999). Functional roles of S100 proteins, calcium-binding proteins of the EF-hand
type. Biochim. Biophys. Acta 1450, 191-231.
Eisenhoffer, G.T., Kang, H., and Sanchez Alvarado, A. (2008). Molecular analysis of stem cells
and their descendants during cell turnover and regeneration in the planarian Schmidtea
mediterranea.Cell Stem Cell 3, 327-339.
Elliot, E.J., and Muller, K.J. (1982). Synapses between neurons regenerate accurately after
destruction of ensheathing glial cells in the leech. Science 215, 1260-1262.
Erber, A., Riemer, D., Bovenschulte, M., and Weber, K. (1998). Molecular Phylogeny of
Metazoan Intermediate Filament Proteins. J Mol Evol 47, 751-762.
Fan, G., Martinowich, K., Chin, M.H., He, F., Fouse, S.D., Hutnick, L., Hattori, D., Ge, W.,
Shen, Y., Wu, H., et al. (2005). DNA methylation controls the timing of astrogliogenesis through
regulation of JAK-STAT signaling. Development (Cambridge, England) 132, 3345-3356.
Faulkner, J.R., Herrmann, J.E., Woo, M.J., Tansey, K.E., Doan, N.B., and Sofroniew, M.V.
(2004). Reactive astrocytes protect tissue and preserve function after spinal cord injury. Journal
of Neuroscience 24, 2143-2155.
Feng, G.-D., He, B.-R., Lu, F., Liu, L.-H., Zhang, L., Chen, B., He, Z.-P., Hao, D.-J., and Yang,
H. (2014). Fibroblast Growth Factor 4 Is Required but not Sufficient for the Astrocyte
Dedifferentiation. Mol. Neurobiol.
FernAndez-HernAndez, I., Rhiner, C., and Moreno, E. (2013). Adult neurogenesis in Drosophila.
Cell Rep 3, 1857-1865.
59
Flici, H., Erkosar, B., Komonyi, 0., Karatas, O.F., Laneve, P., and Giangrande, A. (2011).
Gcm/Glide-dependent conversion into glia depends on neural stem cell age, but not on division,
triggering a chromatin signature that is conserved in vertebrate glia. Development (Cambridge,
England) 138,4167-4178.
Fraguas, S., Barberin, S., and Cebrik, F. (2011). EGFR signaling regulates cell proliferation,
differentiation and morphogenesis during planarian regeneration and homeostasis.
Developmental Biology 354, 87-101.
Freeman, M.R., and Doherty, J. (2006). Glial cell biology in Drosophilaand vertebrates. Trends
Neurosci.
Freeman, M.R. (2010). Specification and morphogenesis of astrocytes. Science 330, 774-778.
Frost, E.E., Nielsen, J.A., Le, T.Q., and Armstrong, R.C. (2003). PDGF and FGF2 regulate
oligodendrocyte progenitor responses to demyelination. J. Neurobiol. 54, 457-472.
Fry, E.J., Chagnon, M.J., L6pez-Vales, R., Tremblay, M.L., and David, S. (2010). Corticospinal
tract regeneration after spinal cord injury in receptor protein tyrosine phosphatase sigma
deficient mice. Glia 58, 423-433.
Gaiano, N., Nye, J.S., and Fishell, G. (2000). Radial glial identity is promoted by Notchl
signaling in the murine forebrain. Neuron 26, 395-404.
Gallo, V., Zhou, J.M., McBain, C.J., Wright, P., Knutson, P.L., and Armstrong, R.C. (1996).
Oligodendrocyte progenitor cell proliferation and lineage progression are regulated by glutamate
receptor-mediated K+ channel block. J. Neurosci. 16, 2659-2670.
Gibson, E.M., Purger, D., Mount, C.W., Goldstein, A.K., Lin, G.L., Wood, L.S., Inema, I.,
Miller, S.E., Bieri, G., Zuchero, J.B., et al. (2014). Neuronal Activity Promotes
Oligodendrogenesis and Adaptive Myelination in the Mammalian Brain. Science.
Giesen, K., Hummel, T., Stollewerk, A., Harrison, S., Travers, A., and KlAmbt, C. (1997). Glial
development in the DrosophilaCNS requires concomitant activation of glial and repression of
neuronal differentiation genes. Development (Cambridge, England) 124, 2307-2316.
Goldstein, L.S., and Gunawardena, S. (2000). Flying through the Drosophilacytoskeletal
genome. J. Cell Biol. 150, F63-F68.
Golubev, A.I. (1988). Glia and neuroglia relationships in the central nervous system of the
Turbellaria (Electron microscopic data). Fortschr Zool 36, 31-37.
Gonzalez-Perez, 0., Romero-Rodriguez, R., Soriano-Navarro, M., Garcia-Verdugo, J.M., and
Alvarez-Buylla, A. (2009). Epidermal growth factor induces the progeny of subventricular zone
type B cells to migrate and differentiate into oligodendrocytes. Stem Cells 27, 2032-2043.
60
Grandbarbe, L., Bouissac, J., Rand, M., Hrab6 de Angelis, M., Artavanis-Tsakonas, S., and
Mohier, E. (2003). Delta-Notch signaling controls the generation of neurons/glia from neural
stem cells in a stepwise process. Development (Cambridge, England) 130, 1391-1402.
Granderath, S., Stollewerk, A., Greig, S., Goodman, C.S., O'Kane, C.J., and KlAmbt, C. (1999).
loco encodes an RGS protein required for Drosophila glial differentiation. Development
(Cambridge, England) 126, 1781-1791.
Greenwood, K., and Butt, A.M. (2003). Evidence that perinatal and adult NG2-glia are not
conventional oligodendrocyte progenitors and do not depend on axons for their survival. Mol.
Cell. Neurosci. 23, 544-558.
Gritti, A., Fr6lichsthal-Schoeller, P., Galli, R., Parati, E.A., Cova, L., Pagano, S.F., Bjornson,
C.R., and Vescovi, A.L. (1999). Epidermal and fibroblast growth factors behave as mitogenic
regulators for a single multipotent stem cell-like population from the subventricular region of the
adult mouse forebrain. J. Neurosci. 19, 3287-3297.
Guedelhoefer, O.C., and SAnchez Alvarado, A. (2012). Amputation induces stem cell
mobilization to sites of injury during planarian regeneration. Development (Cambridge, England)
139, 3510-3520.
Guo, F., Ma, J., McCauley, E., Bannerman, P., and Pleasure, D. (2009). Early postnatal
proteolipid promoter-expressing progenitors produce multilineage cells in vivo. Journal of
Neuroscience 29, 7256-7270.
Hanisch, U.-K., and Kettenmann, H. (2007). Microglia: active sensor and versatile effector cells
in the normal and pathologic brain. Nat. Neurosci. 10, 1387-1394.
Hartenstein, V. (2011). Morphological diversity and development of glia in Drosophila. Glia 59,
1237-1252.
Hartfuss, E., Galli, R., Heins, N., and G6tz, M. (2001). Characterization of CNS precursor
subtypes and radial glia. Developmental Biology 229, 15-30.
Hartline, D.K. (2011). The evolutionary origins of glia. Glia 59, 1215-1236.
Hayashi, T., Asami, M., Higuchi, S., Shibata, N., and Agata, K. (2006). Isolation of planarian Xray-sensitive stem cells by fluorescence-activated cell sorting. Dev. Growth Differ. 48, 371-380.
Hayashi, T., Shibata, N., Okumura, R., Kudome, T., Nishimura, 0., Tarui, H., and Agata, K.
(2010). Single-cell gene profiling of planarian stem cells using fluorescent activated cell sorting
and its "index sorting" function for stem cell research. Dev. Growth Differ. 52, 131-144.
He, F., Ge, W., Martinowich, K., Becker-Catania, S., Coskun, V., Zhu, W., Wu, H., Castro, D.,
Guillemot, F., Fan, G., et al. (2005). A positive autoregulatory loop of Jak-STAT signaling
controls the onset of astrogliogenesis. Nat. Neurosci. 8, 616-625.
61
Hess, D.C., Abe, T., Hill, W.D., Studdard, A.M., Carothers, J., Masuya, M., Fleming, P.A.,
Drake, C.J., and Ogawa, M. (2004). Hematopoietic origin of microglial and perivascular cells in
brain. Exp. Neurol. 186, 134-144.
Hirabayashi, Y., Suzki, N., Tsuboi, M., Endo, T.A., Toyoda, T., Shinga, J., Koseki, H., Vidal,
M., and Gotoh, Y. (2009). Polycomb limits the neurogenic competence of neural precursor cells
to promote astrogenic fate transition. Neuron 63, 600-613.
Holmes, G.P., Negus, K., Burridge, L., Raman, S., Algar, E., Yamada, T., and Little, M.H.
(1998). Distinct but overlapping expression patterns of two vertebrate slit homologs implies
functional roles in CNS development and organogenesis. Mech. Dev. 79, 57-72.
Hosoya, T., Takizawa, K., Nitta, K., and Hotta,'Y. (1995). glial cells missing: a binary switch
between neuronal and glial determination in Drosophila. Cell 82, 1025-1036.
Houle, J.D., and Jin, Y. (2001). Chronically injured supraspinal neurons exhibit only modest
axonal dieback in response to a cervical hemisection lesion. Exp. Neurol. 169, 208-217.
Howes, E.A., Smith, P.J., and Treherne, J.E. (1989). Adult insect glial culture: Activation,
substrate effects and proliferation. Tissue Cell 21, 759-772.
Hyman, L.H. (1940). The invertebrates: Platyhelminthes and Rhynchocoela, the acoelomate
Bilateria. (McGraw-Hill).
Jacques, C., Soustelle, L., Nagy, I., Diebold, C., and Giangrande, A. (2009). A novel role of the
glial fate determinant glial cells missing in hematopoiesis. Int. J. Dev. Biol. 53, 1013-1022.
Johansson, C.B., Momma, S., Clarke, D.L., Risling, M., Lendahl, U., and Frisdn, J. (1999).
Identification of a neural stem cell in the adult mammalian central nervous system. Cell 96, 2534.
Juurlink, B.H., Thorburne, S.K., and Hertz, L. (1998). Peroxide-scavenging deficit underlies
oligodendrocyte susceptibility to oxidative stress. Glia 22, 371-378.
Kang, S.H., Fukaya, M., Yang, J.K., Rothstein, J.D., and Bergles, D.E. (2010). NG2+ CNS glial
progenitors remain committed to the oligodendrocyte lineage in postnatal life and following
neurodegeneration. Neuron 68, 668-681.
Kaplan, M.R., Meyer-Franke, A., Lambert, S., Bennett, V., Duncan, I.D., Levinson, S.R., and
Barres, B.A. (1997). Induction of sodium channel clustering by oligodendrocytes. Nature 386,
724-728.
Karalay, 0., Doberauer, K., Vadodaria, K.C., Knobloch, M., Berti, L., Miquelajauregui, A.,
Schwark, M., Jagasia, R., Taketo, M.M., Tarabykin, V., et al. (2011). Prospero-related
homeobox 1 gene (Proxi) is regulated by canonical Wnt signaling and has a stage-specific role
in adult hippocampal neurogenesis. Proceedings of the National Academy of Sciences 108,
62
5807-5812.
Kato, K., Awasaki, T., and Ito, K. (2009). Neuronal programmed cell death induces glial cell
division in the adult Drosophilabrain. Development (Cambridge, England) 136, 51-59.
Kato, K., Forero, M.G., Fenton, J.C., and Hidalgo, A. (2011). The glial regenerative response to
central nervous system injury is enabled by pros-notch and pros-NFicB feedback. PLoS Biol 9,
e1001 133.
Kdrad6ttir, R., Cavelier, P., Bergersen, L.H., and Attwell, D. (2005). NMDA receptors are
expressed in oligodendrocytes and activated in ischaemia. Nature 438, 1162-1166.
Keirstead, H.S., and Blakemore, W.F. (1997). Identification of post-mitotic oligodendrocytes
incapable of remyelination within the demyelinated adult spinal cord. J. Neuropathol. Exp.
Neurol. 56, 1191-1201.
Kessaris, N., Fogarty, M., Iannarelli, P., Grist, M., Wegner, M., and Richardson, W.D. (2006).
Competing waves of oligodendrocytes in the forebrain and postnatal elimination of an embryonic
lineage. Nat. Neurosci. 9, 173-179.
Koenning, M., Jackson, S., Hay, C.M., Faux, C., Kilpatrick, T.J., Willingham, M., and Emery, B.
(2012). Myelin gene regulatory factor is required for maintenance of myelin and mature
oligodendrocyte identity in the adult CNS. Journal of Neuroscience 32, 12528-12542.
Kotter, M.R., Li, W.-W., Zhao, C., and Franklin, R.J.M. (2006). Myelin impairs CNS
remyelination by inhibiting oligodendrocyte precursor cell differentiation. Journal of
Neuroscience 26, 328-332.
Kuhn, H.G., Winkler, J., Kempermann, G., Thal, L.J., and Gage, F.H. (1997). Epidermal growth
factor and fibroblast growth factor-2 have different effects on neural progenitors in the adult rat
brain. J. Neurosci. 17, 5820-5829.
Lang, B., Liu, H.L., Liu, R., Feng, G.D., Jiao, X.Y., and Ju, G. (2004). Astrocytes in injured
adult rat spinal cord may acquire the potential of neural stem cells. Neuroscience 128, 775-783.
Laywell, E.D., Rakic, P., Kukekov, V.G., Holland, E.C., and Steindler, D.A. (2000).
Identification of a multipotent astrocytic stem cell in the immature and adult mouse brain. Proc.
Natl. Acad. Sci. U.S.a. 97, 13883-13888.
Levine, J.M., and Card, J.P. (1987). Light and electron microscopic localization of a cell surface
antigen (NG2) in the rat cerebellum: association with smooth protoplasmic astrocytes. J.
Neurosci. 7, 2711-2720.
Levison, S.W., and Goldman, J.E. (1993). Both oligodendrocytes and astrocytes develop from
progenitors in the subventricular zone of postnatal rat forebrain. Neuron 10, 201-212.
63
Levison, S.W., Chuang, C., Abramson, B.J., and Goldman, J.E. (1993). The migrational patterns
and developmental fates of glial precursors in the rat subventricular zone are temporally
regulated. Development (Cambridge, England) 119,611-622.
Li, L., Harms, K.M., Ventura, P.B., Lagace, D.C., Eisch, A.J., and Cunningham, L.A. (2010).
Focal cerebral ischemia induces a multilineage cytogenic response from adult subventricular
zone that is predominantly gliogenic. Glia 58, 1610-1619.
Li, Y., and Raisman, G. (1995). Sprouts from cut corticospinal axons persist in the presence of
astrocytic scarring in long-term lesions of the adult rat spinal cord. Exp. Neurol. 134, 102-111.
Lois, C., and Alvarez-Buylla, A. (1993). Proliferating subventricular zone cells in the adult
mammalian forebrain can differentiate into neurons and glia. Proc. Natl. Acad. Sci. U.S.a. 90,
2074-2077.
Malatesta, P., Hartfuss, E., and G6tz, M. (2000). Isolation of radial glial cells by fluorescentactivated cell sorting reveals a neuronal lineage. Development (Cambridge, England) 127, 52535263.
Manev, H., Favaron, M., Guidotti, A., and Costa, E. (1989). Delayed increase of Ca2+ influx
elicited by glutamate: role in neuronal death. Mol. Pharmacol. 36, 106-112.
Marshall, C.A.G., Novitch, B.G., and Goldman, J.E. (2005). Olig2 directs astrocyte and
oligodendrocyte formation in postnatal subventricular zone cells. Journal of Neuroscience 25,
7289-7298.
Mason, J.L., Toews, A., Hostettler, J.D., Morell, P., Suzuki, K., Goldman, J.E., and Matsushima,
G.K. (2004). Oligodendrocytes and progenitors become progressively depleted within
chronically demyelinated lesions. Am. J. Pathol. 164, 1673-1682.
McKeon, R.J., Schreiber, R.C., Rudge, J.S., and Silver, J. (1991). Reduction of neurite outgrowth
in a model of glial scarring following CNS injury is correlated with the expression of inhibitory
molecules on reactive astrocytes. J. Neurosci. 11, 3398-3411.
McTigue, D.M., and Tripathi, R.B. (2008). The life, death, and replacement of oligodendrocytes
in the adult CNS. J. Neurochem. 107, 1-19.
Menn, B., Garcia-Verdugo, J.M., Yaschine, C., Gonzalez-Perez, 0., Rowitch, D., and AlvarezBuylla, A. (2006). Origin of oligodendrocytes in the subventricular zone of the adult brain.
Journal of Neuroscience 26, 7907-7918.
Menne, T.V., and Klimbt, C. (1994). The formation of commissures in the Drosophila CNS
depends on the midline cells and on the Notch gene. Development (Cambridge, England) 120,
123-133.
Merkle, F.T., Tramontin, A.D., Garcia-Verdugo, J.M., and Alvarez-Buylla, A. (2004). Radial
64
glia give rise to adult neural stem cells in the subventricular zone. Proc. Natl. Acad. Sci. U.S.a.
101, 17528-17532.
Miller, R.H., and Raff, M.C. (1984). Fibrous and protoplasmic astrocytes are biochemically and
developmentally distinct. J. Neurosci. 4, 585-592.
Mitchell, K.J., Doyle, J.L., Serafini, T., Kennedy, T.E., Tessier-Lavigne, M., Goodman, C.S.,
and Dickson, B.J. (1996). Genetic analysis of Netrin genes in Drosophila: Netrins guide CNS
commissural axons and peripheral motor axons. Neuron 17, 203-215.
Moon, L.D., Asher, R.A., Rhodes, K.E., and Fawcett, J.W. (2001). Regeneration of CNS axons
back to their target following treatment of adult rat brain with chondroitinase ABC. Nat.
Neurosci. 4,465-466.
Morgan, T.H. (1898). Experimental studies of the regeneration of Planariamaculata. Dev.
Genes Evol. 7, 364-397.
Morita, M., and Best, J.B. (1965). Electron microscopic studies on Planaria. II. Fine structure of
the neurosecretory system in the planarian Dugesia dorotocephala.J. Ultrastruct. Res. 13, 396408.
Morita, M., and Best, J.B. (1966). Electron microscopic studies of planaria. III. Some
observations on the fine structure of planarian nervous tissue. J. Exp. Zool. 161, 391-411.
Nait-Oumesmar, B., Picard-Riera, N., Kerninon, C., Decker, L., Seilhean, D., H6glinger, G.U.,
Hirsch, E.C., Reynolds, R., and Baron-Van Evercooren, A. (2007). Activation of the
subventricular zone in multiple sclerosis: evidence for early glial progenitors. Proc. Natl. Acad.
Sci. U.S.a. 104,4694-4699.
Nakatomi, H., Kuriu, T., Okabe, S., Yamamoto, S.-I., Hatano, 0., Kawahara, N., Tamura, A.,
Kirino, T., and Nakafuku, M. (2002). Regeneration of hippocampal pyramidal neurons after
ischemic brain injury by recruitment of endogenous neural progenitors. Cell 110, 429-441.
Namihira, M., Kohyama, J., Semi, K., Sanosaka, T., Deneen, B., Taga, T., and Nakashima, K.
(2009). Committed neuronal precursors confer astrocytic potential on residual neural precursor
cells. Dev. Cell 16, 245-255.
Newmark, P.A., and Sanchez Alvarado, A. (2000). Bromodeoxyuridine specifically labels the
regenerative stem cells of planarians. Developmental Biology 220, 142-153.
Newmark, P.A., and SAnchez Alvarado, A. (2002). Not your father's planarian: a classic model
enters the era of functional genomics. Nat Rev Genet 3, 210-219.
Newmark, P.A., Reddien, P.W., Cebrii, F., and S~nchez Alvarado, A. (2003). Ingestion of
bacterially expressed double-stranded RNA inhibits gene expression in planarians. Proc. Natl.
Acad. Sci. U.S.a. 100 Suppl 1, 11861-11865.
65
Nieto, M., Schuurmans, C., Britz, 0., and Guillemot, F. (2001). Neural bHLH genes control the
neuronal versus glial fate decision in cortical progenitors. Neuron 29, 401-413.
Nishiyama, A. (2007). Polydendrocytes: NG2 cells with many roles in development and repair of
the CNS. Neuroscientist 13, 62-76.
Noctor, S.C., Flint, A.C., Weissman, T.A., Dammerman, R.S., and Kriegstein, A.R. (2001).
Neurons derived from radial glial cells establish radial units in neocortex. Nature 409, 714-720.
Noctor, S.C., Martinez-Cerdeflo, V., Ivic, L., and Kriegstein, A.R. (2004). Cortical neurons arise
in symmetric and asymmetric division zones and migrate through specific phases. Nat. Neurosci.
7, 136-144.
Oikonomou, G., and Shaham, S. (2011). The glia of Caenorhabditiselegans. Glia 59, 12531263.
Oyallon, J., Apitz, H., Miguel-Aliaga, I., Timofeev, K., Ferreira, L., and Salecker, I. (2012).
Regulation of locomotion and motoneuron trajectory selection and targeting by the Drosophila
homolog of Olig family transcription factors. Developmental Biology 369, 261-276.
Pearce, J., Lnenicka, G.A., and Govind, C.K. (2003). Regenerating crayfish motor axons
assimilate glial cells and sprout in cultured explants. J. Comp. Neurol. 464, 449-462.
Pearson, B.J., Eisenhoffer, G.T., Gurley, K.A., Rink, J.C., Miller, D.E., and Sinchez Alvarado,
A. (2009). Formaldehyde-based whole-mount in situ hybridization method for planarians. Dev.
Dyn. 238,443-450.
Pellettieri, J., Fitzgerald, P., Watanabe, S., Mancuso, J., Green, D.R., and SAnchez Alvarado, A.
(2010). Cell death and tissue remodeling in planarian regeneration. Developmental Biology 338,
76-85.
Penderis, J., Shields, S.A., and Franklin, R.J.M. (2003). Impaired remyelination and depletion of
oligodendrocyte progenitors does not occur following repeated episodes of focal demyelination
in the rat central nervous system. Brain 126, 1382-1391.
Qi, Y., Cai, J., Wu, Y., Wu, R., Lee, J., Fu, H., Rao, M., Sussel, L., Rubenstein, J., and Qiu, M.
(2001). Control of oligodendrocyte differentiation by the Nkx2.2 homeodomain transcription
factor. Development (Cambridge, England) 128, 2723-2733.
Reddien, P.W., and SAnchez Alvarado, A. (2004). Fundamentals of planarian regeneration.
Annu. Rev. Cell Dev. Biol. 20, 725-757.
Reddien, P.W., Bermange, A.L., Kicza, A.M., and SAnchez Alvarado, A. (2007). BMP signaling
regulates the dorsal planarian midline and is needed for asymmetric regeneration. Development
(Cambridge, England) 134, 4043-4051.
66
Reddien, P.W., Bermange, A.L., Murfitt, K.J., Jennings, J.R., and Sinchez Alvarado, A. (2005a).
Identification of genes needed for regeneration, stem cell function, and tissue homeostasis by
systematic gene perturbation in planaria. Dev. Cell 8, 635-649.
Reddien, P.W., Oviedo, N.J., Jennings, J.R., Jenkin, J.C., and Sanchez Alvarado, A. (2005b).
SMEDWI-2 is a PIWI-like protein that regulates planarian stem cells. Science 310, 1327-1330.
Redwine, J.M., and Armstrong, R.C. (1998). In vivo proliferation of oligodendrocyte progenitors
expressing PDGFalphaR during early remyelination. J. Neurobiol. 37, 413-428.
Reuter, M., Mntyl1, K., and Gustafsson, M.K.S. (1998). Organization of the orthogon - main
and minor nerve cords - Springer. Hydrobiologia 383, 175-182.
Reynolds, R., and Hardy, R. (1997). Oligodendroglial progenitors labeled with the 04 antibody
persist in the adult rat cerebral cortex in vivo. J. Neurosci. Res. 47, 455-470.
Rink, J.C., Gurley, K.A., Elliott, S.A., and Sanchez Alvarado, A. (2009). Planarian Hh Signaling
Regulates Regeneration Polarity and Links Hh Pathway Evolution to Cilia. Science 326, 14061410.
Rival, T., Soustelle, L., Strambi, C., Besson, M.-T., Ichd, M., and Birman, S. (2004). Decreasing
glutamate buffering capacity triggers oxidative stress and neuropil degeneration in the
Drosophilabrain. Curr. Biol. 14, 599-605.
Rivers, L.E., Young, K.M., Rizzi, M., Jamen, F., Psachoulia, K., Wade, A., Kessaris, N., and
Richardson, W.D. (2008). PDGFRA/NG2 glia generate myelinating oligodendrocytes and
piriform projection neurons in adult mice. Nat. Neurosci. 11, 1392-1401.
Robel, S., Berninger, B., and G6tz, M. (2011). The stem cell potential of glia: lessons from
reactive gliosis. Nat. Rev. Neurosci. 12, 88-104.
Rolls, A., Shechter, R., and Schwartz, M. (2009). The bright side of the glial scar in CNS repair.
Nat. Rev. Neurosci. 10, 235-241.
Rothberg, J.M., Jacobs, J.R., Goodman, C.S., and Artavanis-Tsakonas, S. (1990). slit: an
extracellular protein necessary for development of midline glia and commissural axon pathways
contains both EGF and LRR domains. Genes Dev. 4, 2169-2187.
Rudge, J.S., and Silver, J. (1990). Inhibition of neurite outgrowth on astroglial scars in vitro. J.
Neurosci. 10, 3594-3603.
Saito, A., Kanemoto, S., Kawasaki, N., Asada, R., Iwamoto, H., Oki, M., Miyagi, H., Izumi, S.,
Sanosaka, T., Nakashima, K., et al. (2012). Unfolded protein response, activated by OASIS
family transcription factors, promotes astrocyte differentiation. Nat Commun 3, 967.
Sal6, E., and Bagufia, J. (1984). Regeneration and pattern formation in planarians. I. The pattern
67
of mitosis in anterior and posterior regeneration in Dugesia (G) tigrina, and a new proposal for
blastema formation. J Embryol Exp Morphol 83, 63-80.
Sanchez Alvarado, A., and Newmark, P.A. (1999). Double-stranded RNA specifically disrupts
gene expression during planarian regeneration. Proc. Natl. Acad. Sci. U.S.a. 96, 5049-5054.
Scholz, H., Sadlowski, E., Klaes, A., and KlAmbt, C. (1997). Control of midline glia
development in the embryonic DrosophilaCNS. Mech. Dev. 62, 79-91.
Selmaj, K.W., Farooq, M., Norton, W.T., Raine, C.S., and Brosnan, C.F. (1990). Proliferation of
astrocytes in vitro in response to cytokines. A primary role for tumor necrosis factor. J. Immunol.
144, 129-135.
Serafini, T., Kennedy, T.E., Galko, M.J., Mirzayan, C., Jessell, T.M., and Tessier-Lavigne, M.
(1994). The netrins define a family of axon outgrowth-promoting proteins homologous to C.
elegans UNC-6. Cell 78,409-424.
Shafer, O.T., Chen, A., Kumar, S.M., Muller, K.J., and Sahley, C.L. (1998). Injury-induced
expression of endothelial nitric oxide synthase by glial and microglial cells in the leech central
nervous system within minutes after injury. Proc. Biol. Sci. 265, 2171-2175.
Shen, S., Sandoval, J., Swiss, V.A., Li, J., Dupree, J., Franklin, R.J.M., and Casaccia-Bonnefil,
P. (2008). Age-dependent epigenetic control of differentiation inhibitors is critical for
remyelination efficiency. Nat. Neurosci. 11, 1024-1034.
Shen, Y., Tenney, A.P., Busch, S.A., Horn, K.P., Cuascut, F.X., Liu, K., He, Z., Silver, J., and
Flanagan, J.G. (2009). PTPsigma is a receptor for chondroitin sulfate proteoglycan, an inhibitor
of neural regeneration. Science 326, 592-596.
Shibata, T., Yamada, K., Watanabe, M., Ikenaka, K., Wada, K., Tanaka, K., and Inoue, Y.
(1997). Glutamate transporter GLAST is expressed in the radial glia-astrocyte lineage of
developing mouse spinal cord. J. Neurosci. 17, 9212-9219.
Silver, J., and Miller, J.H. (2004). Regeneration beyond the glial scar. Nat. Rev. Neurosci. 5,
146-156.
Sim, F.J., Zhao, C., Penderis, J., and Franklin, R.J.M. (2002). The age-related decrease in CNS
remyelination efficiency is attributable to an impairment of both oligodendrocyte progenitor
recruitment and differentiation. Journal of Neuroscience 22, 2451-2459.
Sirko, S., Behrendt, G., Johansson, P.A., Tripathi, P., Costa, M., Bek, S., Heinrich, C., Tiedt, S.,
Colak, D., Dichgans, M., et al. (2013). Reactive glia in the injured brain acquire stem cell
properties in response to sonic hedgehog. [corrected]. Cell Stem Cell 12, 426-439.
Sirko, S., Neitz, A., Mittmann, T., Horvat-Br6cker, A., Holst, von, A., Eysel, U.T., and Faissner,
A. (2009). Focal laser-lesions activate an endogenous population of neural stem/progenitor cells
68
in the adult visual cortex. Brain 132, 2252-2264.
Smith, K.M., Ohkubo, Y., Maragnoli, M.E., Rasin, M.-R., Schwartz, M.L., Sestan, N., and
Vaccarino, F.M. (2006). Midline radial glia translocation and corpus callosum formation require
FGF signaling. Nat. Neurosci. 9, 787-797.
Smith, P.J., Howes, E.A., and Treherne, J.E. (1987). Mechanisms of glial regeneration in an
insect central nervous system. J. Exp. Biol. 132, 59-78.
Smith, P.J., Howes, E.A., and Treherne, J.E. (1990). Cell proliferation in the repairing adult
insect central nervous system: incorporation of the thymidine analogue 5-bromo-2-deoxyuridine
in vivo. J. Cell. Sci. 95 (Pt 4), 599-604.
Sofroniew, M.V. (2009). Molecular dissection of reactive astrogliosis and glial scar formation.
Trends Neurosci. 32, 638-647.
Sofroniew, M.V., and Vinters, H.V. (2010). Astrocytes: biology and pathology. Acta
Neuropathol. 119, 7-35.
Song, H., Stevens, C.F., and Gage, F.H. (2002). Astroglia induce neurogenesis from adult neural
stem cells. Nature 417, 39-44.
Spassky, N., Merkle, F.T., Flames, N., Tramontin, A.D., Garcia-Verdugo, J.M., and AlvarezBuylla, A. (2005). Adult ependymal cells are postmitotic and are derived from radial glial cells
during embryogenesis. Journal of Neuroscience 25, 10-18.
Stork, T., Engelen, D., Krudewig, A., Silies, M., Bainton, R.J., and Klambt, C. (2008).
Organization and function of the blood-brain barrier in Drosophila.Journal of Neuroscience 28,
587-597.
Sun, Y., Nadal-Vicens, M., Misono, S., Lin, M.Z., Zubiaga, A., Hua, X., Fan, G., and Greenberg,
M.E. (2001). Neurogenin promotes neurogenesis and inhibits glial differentiation by independent
mechanisms. Cell 104, 365-376.
Swiss, V.A., Nguyen, T., Dugas, J., Ibrahim, A., Barres, B., Androulakis, I.P., and Casaccia, P.
(2011). Identification of a gene regulatory network necessary for the initiation of
oligodendrocyte differentiation. PLoS ONE 6, e18088.
Talbott, J.F., Loy, D.N., Liu, Y., Qiu, M.S., Bunge, M.B., Rao, M.S., and Whittemore, S.R.
(2005). Endogenous Nkx2.2+/Olig2+ oligodendrocyte precursor cells fail to remyelinate the
demyelinated adult rat spinal cord in the absence of astrocytes. Exp. Neurol. 192, 11-24.
Treherne, J.E., Harrison, J.B., Treherne, J.M., and Lane, N.J. (1984). Glial repair in an insect
central nervous system: effects of surgical lesioning. J. Neurosci. 4, 2689-2697.
Umesono, Y., Watanabe, K., and Agata, K. (1997). A planarian orthopedia homolog is
69
specifically expressed in the branch region of both the mature and regenerating brain. Dev.
Growth Differ. 39, 723-727.
Unhavaithaya, Y., and Orr-Weaver, T.L. (2012). Polyploidization of glia in neural development
links tissue growth to blood-brain barrier integrity. Genes Dev. 26, 31-36.
Vaccarino, F.M., Fagel, D.M., Ganat, Y., Maragnoli, M.E., Ment, L.R., Ohkubo, Y., Schwartz,
M.L., Silbereis, J., and Smith, K.M. (2007). Astroglial cells in development, regeneration, and
repair. Neuroscientist 13, 173-185.
Voskuhl, R.R., Peterson, R.S., Song, B., Ao, Y., Morales, L.B.J., Tiwari-Woodruff, S., and
Sofroniew, M.V. (2009). Reactive astrocytes form scar-like perivascular barriers to leukocytes
during adaptive immune inflammation of the CNS. Journal of Neuroscience 29, 11511-11522.
Wadsworth, W.G., Bhatt, H., and Hedgecock, E.M. (1996). Neuroglia and pioneer neurons
express UNC-6 to provide global and local netrin cues for guiding migrations in C. elegans.
Neuron 16, 35-46.
Wake, H., Lee, P.R., and Fields, R.D. (2011). Control of local protein synthesis and initial events
in myelination by action potentials. Science 333, 1647-1651.
Watanabe, M., Toyama, Y., and Nishiyama, A. (2002). Differentiation of proliferated NG2positive glial progenitor cells in a remyelinating lesion. J. Neurosci. Res. 69, 826-836.
Watson, J.D., Wheeler, S.R., Stagg, S.B., and Crews, S.T. (2011). Drosophilahedgehog
signaling and engrailed-runt mutual repression direct midline glia to alternative ensheathing and
non-ensheathing fates. Development (Cambridge, England) 138, 1285-1295.
Wegner, M. (2008). A matter of identity: transcriptional control in oligodendrocytes. J. Mol.
Neurosci. 35, 3-12.
Wenemoser, D., and Reddien, P.W. (2010). Planarian regeneration involves distinct stem cell
responses to wounds and tissue absence. Developmental Biology 344, 979-991.
Wilhelmsson, U., Faiz, M., de Pablo, Y., Sjdqvist, M., Andersson, D., Widestrand, A., Potokar,
M., Stenovec, M., Smith, P.L.P., Shinjyo, N., et al. (2012). Astrocytes negatively regulate
neurogenesis through the JaggedI-mediated Notch pathway. Stem Cells 30, 2320-2329.
Wu, M., Hernandez, M., Shen, S., Sabo, J.K., Kelkar, D., Wang, J., O'Leary, R., Phillips, G.R.,
Cate, H.S., and Casaccia, P. (2012). Differential modulation of the oligodendrocyte
transcriptome by sonic hedgehog and bone morphogenetic protein 4 via opposing effects on
histone acetylation. Journal of Neuroscience 32, 6651-6664.
Xin, M., Yue, T., Ma, Z., Wu, F.-F., Gow, A., and Lu, Q.R. (2005). Myelinogenesis and axonal
recognition by oligodendrocytes in brain are uncoupled in OligI-null mice. Journal of
Neuroscience 25, 1354-1365.
70
Yang, H., Ling, W., Vitale, A., Olivera, C., Min, Y., and You, S. (2011). ErbB2 activation
contributes to de-differentiation of astrocytes into radial glial cells following induction of
scratch-insulted astrocyte conditioned medium. Neurochemistry International 59, 1010-1018.
Yazawa, S., Umesono, Y., Hayashi, T., Tarui, H., and Agata, K. (2009). Planarian
Hedgehog/Patched establishes anterior-posterior polarity by regulating Wnt signaling.
Proceedings of the National Academy of Sciences 106, 22329-22334.
Yoon, K., Nery, S., Rutlin, M.L., Radtke, F., Fishell, G., and Gaiano, N. (2004). Fibroblast
growth factor receptor signaling promotes radial glial identity and interacts with Notchi
signaling in telencephalic progenitors. Journal of Neuroscience 24, 9497-9506.
Young, K.M., Psachoulia, K., Tripathi, R.B., Dunn, S.-J., Cossell, L., Attwell, D., Tohyama, K.,
and Richardson, W.D. (2013). Oligodendrocyte dynamics in the healthy adult CNS: evidence for
myelin remodeling. Neuron 77, 873-885.
Yu, Y., Chen, Y., Kim, B., Wang, H., Zhao, C., He, X., Liu, L., Liu, W., Wu, L.M.N., Mao, M.,
et al. (2013). Olig2 targets chromatin remodelers to enhancers to initiate oligodendrocyte
differentiation. Cell 152, 248-261.
Zamanian, J.L., Xu, L., Foo, L.C., Nouri, N., Zhou, L., Giffard, R.G., and Barres, B.A. (2012).
Genomic analysis of reactive astrogliosis. Journal of Neuroscience 32, 6391-6410.
Zhang, R.L., Zhang, Z.G., Zhang, L., and Chopp, M. (2001). Proliferation and differentiation of
progenitor cells in the cortex and the subventricular zone in the adult rat after focal cerebral
ischemia. Neuroscience 105, 33-41.
Zhou, Q., and Anderson, D.J. (2002). The bHLH transcription factors OLIG2 and OLIG1 couple
neuronal and glial subtype specification. Cell 109, 61-73.
Zhu, X., Bergles, D.E., and Nishiyama, A. (2008). NG2 cells generate both oligodendrocytes and
gray matter astrocytes. Development (Cambridge, England) 135, 145-157.
Zhu, X., Hill, R.A., Dietrich, D., Komitova, M., Suzuki, R., and Nishiyama, A. (2011). Agedependent fate and lineage restriction of single NG2 cells. Development (Cambridge, England)
138, 745-753.
Zhu, X., Zuo, H., Maher, B.J., Serwanski, D.R., LoTurco, J.J., Lu, Q.R., and Nishiyama, A.
(2012). Olig2-dependent developmental fate switch of NG2 cells. Development (Cambridge,
England) 139, 2299-2307.
Ziskin, J.L., Nishiyama, A., Rubio, M., Fukaya, M., and Bergles, D.E. (2007). Vesicular release
of glutamate from unmyelinated axons in white matter. Nat. Neurosci. 10, 321-330.
Zou, J., Wang, Y.-X., Dou, F.-F., L, H.-Z., Ma, Z.-W., Lu, P.-H., and Xu, X.-M. (2010).
Glutamine synthetase down-regulation reduces astrocyte protection against glutamate
71
excitotoxicity to neurons. Neurochemistry International 56, 577-5 84.
72
Chapter 2
Clonogenic neoblasts are pluripotent adult stem cells
that underlie planarian regeneration
Daniel E. Wagner', Irving E. Wang', and Peter W. Reddien
'These authors contributed equally
Experiments shown in Figures 1-3, Supplemental Figures S1-S16, and Supplemental Table 1
were performed by DEW. Experiments shown in Figures 4-6 and Supplemental Figures S17S19 were performed by IEW in collaboration with DEW. All authors contributed to the design
of experiments and editing of the manuscript.
Published as:
Wagner, D.E., Wang, I.E., and Reddien, P.W. (2011). Clonogenic neoblasts are pluripotent adult
stem cells that underlie planarian regeneration. Science 332, 811-816
73
Abstract
Pluripotent cells in the embryo can generate all cell types, but lineage-restricted cells are
generally thought to replenish adult tissues. Planarians are flatworms and regenerate from tiny
body fragments, a process requiring a population of proliferating cells (neoblasts).
Whether
regeneration is accomplished by pluripotent cells or by the collective activity of multiple lineagerestricted cell types is unknown. Using ionizing radiation and single-cell transplantation, we
identified neoblasts that can form large descendant-cell colonies in vivo. These clonogenic
neoblasts (cNeoblasts) produce cells that differentiate into neuronal, intestinal, and other known
post-mitotic cell types and are distributed throughout the body. Single transplanted cNeoblasts
restored regeneration in lethally irradiated hosts. We conclude that broadly distributed, adult
pluripotent stem cells underlie the remarkable regenerative abilities of planarians.
74
Introduction
Pluripotent cells, such as embryonic blastomeres, differentiate into mature cell types spanning
three germ layers (Evans, 1972; Evans and Kaufman, 1981; Martin, 1981). Although essential
for development, pluripotent cells are generally not known to be present in adult animals
(Wagers and Weissman, 2004; Wagers et al., 2002). Adult tissues, by contrast, are typically
maintained by specialized, tissue-specific adult stem cells (Barker et al., 2007; Blanpain et al.,
2004; Ohlstein and Spradling, 2006; Spangrude et al., 1988; Uchida et al., 2000; Wagers and
Weissman, 2004; Weissman, 2000).
Planarians are flatworms well known for the ability to
regenerate whole animals from small pieces of tissue (Morgan, 1898). Planarian regeneration
requires a population of proliferative cells, historically known as neoblasts, that exist throughout
the body and collectively produce all known differentiated cell types (Keller, 1894; Reddien and
Sanchez Alvarado, 2004).
Neoblasts have great potential for molecular genetic studies in
Schmidtea mediterranea where a sequenced genome and molecular tools (including RNAi
technology) enable identification and study of genes controlling regeneration(Reddien and
Sanchez Alvarado, 2004; Reddien et al., 2005a). To date, however, neoblast properties have
only been studied at the level of a population (BagufiA et al., 1989; Eisenhoffer et al., 2008;
Lange and Gilbert, 1968; Newmark and Sanchez Alvarado, 2000; Reddien and Sanchez
Alvarado, 2004; Salvetti et al., 2009; Wolff and Dubois, 1948). The cell population known as
neoblasts, therefore, could either contain only lineage-restricted cells that together allow
regeneration, or could contain, within the population, stem cells that are pluripotent at the single-
75
cell level. A fundamental issue to address for understanding planarian regeneration, therefore, is
the in vivo potential of individual proliferating planarian cells.
Results and Discussion
Colonies are generatedfromsingle smedwi-J+ cellsfollowing irradiation
We developed an in vivo method, utilizing ionizing radiation, that permits study of rare,
individual proliferating cells and their descendants. Irradiation eliminates dividing cells and is a
classic strategy for studying stem cells (Becker et al., 1963; Till and McCulloch, 1961). All
dividing cells in adult planarians express the smedwi-J gene (Guo et al., 2006) (Fig. IA); these
cells are specifically, rapidly, and completely depleted following exposure to high irradiation
doses (e.g., 6,000 rads) (Eisenhoffer et al., 2008; Guo et al., 2006; Reddien et al., 2005b). Low
irradiation doses (i.e., 500 rads) eliminate some proliferating cells, leaving a large number spread
ventrally throughout the animal (Salvetti et al., 2009). We identified an irradiation dose (1,750
rads) that eliminated all smedwi-J+ cells from most (78%) animals (Fig. 1B). However, seven
days after 1,750 rad exposure, rare smedwi-J+ cells were present in the minority of animals
(22%) as sparse "clusters" (Fig. lB-C).
Clusters consistently displayed compact, isolated,
colony-like morphology and originated ventrally throughout the body (Fig. ID, S lA-C) but were
not associated with specific known tissues (Fig. S1D-E). These clusters, if resulting from clonal
growth of single smedwi-J+cells, provide the opportunity to study the developmental potential of
individual planarian cells.
76
Numerous smedwi-J+ cluster attributes indicate they result from clonal growth. smedwi1J clusters were preceded by isolated smedwi-J+ cells present 3-4 days following irradiation, and
typically displayed 3-10 cells after one week (Fig. 1E-F).
Based on the low proportion of
animals with smedwi-J+ cells in close proximity 3 days post-irradiation with 1,750 rads (Fig. S2),
it is improbable that clusters arose from multiple adjacent smedwi-J' cells (P = 0.0138, 2-tailed
Fisher's Exact Test).
Cluster size increased dramatically over time, suggesting exponential
growth, ultimately yielding hundreds of smedwi-J+ cells 14-18 days post-irradiation (Fig. 1E-F).
Consistent with clusters originating from pre-existing smedwi-J+ cells that survived irradiation, a
cluster location scatterplot resembled the normal smedwi-J]-expression pattern (Fig. ID), and
cluster frequency decreased with increasing irradiation doses (see below, Fig. 3A). BrdU
delivery labels smedwi-J-expressing cells (Guo et al., 2006) (Fig. S3A), followed by a rapid
decline in incorporation within 24-48 hours (Eisenhoffer et al., 2008) (Fig. S3B-D)
demonstrating unincorporated BrdU does not perdure long term. A BrdU pulse followed by
irradiation resulted in clusters consisting entirely of BrdU+; smedwi-J1
cells (Fig. S3E),
indicating smedwi-I+-clusterexpansion results from division of existing smedwi-J+ cells (i.e., by
clonal growth). If some other process, such as dedifferentiation, produced smedwi-J+ cluster
cells, these cells should have been BrdU~.
Following irradiation, every proliferative cell detected by an 8-hour BrdU pulse (Fig.
1G), or by using probes for the conserved proliferation marker genes histone h2b (Smed-h2b)
(Hewitson et al., 2006), pcna (Smed-pcna) (Bravo et al., 1987), or ribonucleotide reductase
(Smed-RRM2-1) (Eisenhoffer et al., 2008; Eriksson et al., 1984) (Fig. S3F-H), existed in clusters
and expressed smedwi-J (n=815). Therefore, no other source (non-smedwi-J*) for proliferating
77
cells exists outside of smedwi-J clusters in irradiated animals.
Furthermore, if additional
sources for smedwi-1] cells (other than clonal growth) existed, cluster number would be expected
to increase with time and small, newly formed clusters might be present at late timepoints
following irradiation.
Neither of these possibilities was observed (Fig. IF). New cluster
production was also not observed following amputation or feeding (Fig. S4), which elicit
proliferative responses (BagufiA, 1976; Wenemoser and Reddien, 2010). These data indicate that
clonal expansion (producing colonies) represents the source of new smedwi-J+ dividing cells
during cluster formation and growth.
Not all proliferating cells (neoblasts) necessarily have the capacity to form colonies. We
term cells displaying this capacity clonogenic neoblasts (cNeoblasts); these cells express
smedwi-1, have a body-wide (head-to-tail) distribution (Fig. ID), and generate large, expanding
colonies of smedwi-J+ cells. The ability of small numbers of colonies ultimately to restore both
smedwi-J+ cells and mitotic activity to normal levels (Fig. S5) suggests a stem cell-like capacity
for self-renewal.
To investigate the potential of individual cNeoblasts, we analyzed smedwi-J+ colonies
using three well-established differentiation assays involving a SMEDWI-1 antibody (Guo et al.,
2006; Scimone et al., 2010; Wenemoser and Reddien, 2010), BrdU pulse-chase (Eisenhoffer et
al., 2008; Newmark and Sanchez Alvarado, 2000; Scimone et al., 2010), and post-mitotic cell
type markers (Eisenhoffer et al., 2008; Pearson and Sanchez Alvarado, 2010; Scimone et al.,
2010; Wenemoser and Reddien, 2010).
SMEDWI-1 protein is present in smedwi-I mRNA+
dividing cells and temporarily remains detectable in post-mitotic descendant cells (Guo et al.,
2006). Differentiating cells therefore transit through a SMEDWI-1(protein)+; smedwi-J(mRNA)-
78
state (Scimone et al., 2010; Wenemoser and Reddien, 2010). All colonies examined (12/12)
contained SMEDWI-l+; smedwi-1- cells (Fig. 1H). Independently, BrdU can label cells that
divide and exit the smedwi-J' state. All colonies analyzed by 4-day BrdU pulse-chase (31/31)
contained BrdU+; smedwi-1- cells (Fig. 11) and no BrdU+ cells existed in worms lacking smedwi1J colonies (25/25 animals), indicating that colonies produce, and are the only source for, cells
exiting the smedwi-J+ undifferentiated state. SMEDWI-l+ or BrdU+ colony cells can thus be
assessed for lineage-specific marker expression to determine the developmental potential of
individual cNeoblasts.
cNeoblasts display broaddifferentiation capacity
Described adult stem cells typically only produce differentiated cells corresponding to their germ
layer and tissue of origin (Wagers and Weissman, 2004). To address whether cNeoblasts, by
contrast, could produce cell types derived from multiple germ layers, we identified and
characterized markers for neuronal (ectoderm-derived) and intestinal (endoderm-derived)
lineages. In untreated animals, some SMEDWI-1+ descendant cells expressed a choline acetyltransferase ortholog, Smed-chat (Fig. S6); chat expression is widely conserved in cholinergic
neurons (Nishimura et al., 2010). SMEDWI-1*; chati cells were enriched in brain regions, had
neuronal morphology, and chate cells co-expressed additional neuronal markers (Fig. S6),
indicating that SMEDWI- 1; chat7 cells are differentiating neurons. Smed-gata4/5/6 and Smedhnf4, orthologs
of endoderm-promoting
GATA4/5/6
and
HNF4 transcription factors,
respectively (Morrisey et al., 1998), were expressed in intestinal cells and also in interspersed
cells surrounding the intestine (Fig. S7-10). Many of these interspersed cells were irradiation-
79
sensitive and SMEDWI-1*, indicating they represent differentiating endodermal cells (Fig. S8,
S 10). A third endoderm marker gene used, Smed-mat, was expressed in intestinal branches (Fig.
S8, S10).
Finally, additional differentiation marker genes used (Smed-AGA T-1, NB.21.11E,
Smed-MCP-1, Smed-ODC-1, Smed-CYPJAI-1, and NB.52.12F) are expressed in partially
overlapping mesenchymal populations of post-mitotic cells (Fig. SIlA-F) (Eisenhoffer et al.,
2008). These populations have unknown lineage relationships but turnover rapidly, and are
consequently depleted following irradiation (Eisenhoffer et al., 2008) (Fig. S IIG-L).
Using SMEDWI-1 to label colony cell descendants, individual colonies were examined
for the presence of both gata4/5/6+ and chat*, both gata4/5/6' and AGAT-1], or both AGAT-1
and chati differentiating cells.
In nearly all cases (n=20/22), individual colonies contained
newly produced cells of both lineages tested (Fig. 2A-C, S12). The 1,750 rad dose yields rare,
well-separated colonies (See Fig. 1C); animals fixed 7 days after irradiation contained single
colonies (12/28 animals), no colonies (12/28), and only rarely, more than one colony (4/28).
Given the high frequency of colonies producing multiple lineages (20/22), it is improbable that
all such cases were the result of multiple colonies merging (P<0.0001, Fisher's Exact Test). In
addition, using a 4-day BrdU pulse chase as an independent method, we identified several
colonies containing both BrdU+; chat+ (neuronal) and BrdU+; mat+ (intestinal) cells (Fig. 2D).
Nearly all colonies examined, using the SMEDWI-1 antibody or BrdU, produced differentiated
cells for any single lineage analyzed (n=61/64) (Fig. S13). These colonies were distributed
throughout the body and not restricted to specific anatomical regions (Fig. 2E). Finally, nearly
every
smedwi-J+ colony
examined had associated
cells expressing every
additional
differentiation marker tested (NB.21.1 lE, MCP-1, ODC-1, CYPJAJ-1, NB.52.12F) (n1 10/115,
80
Fig. S14A-E); descendant cell clusters, furthermore, were never observed in regions lacking
smedwi-J+ colonies (Fig. S 11G-L). In addition, even early colonies (7 days post-irradiation) had
associated differentiated cells (Fig. S14F). Together, these data indicate that broad multipotency
and a body-wide distribution are fundamental attributes of individual cNeoblasts.
Small numbers of cNeoblasts restore regenerativeability
Irradiated planarians cannot regenerate (Bardeen and Baetjer, 1904) and suffer massive tissue
loss because of failed replacement of aged differentiated cells
(Bardeen and Baetjer, 1904; Dubois, 1949). However, transplantation of large numbers of cells
(Bagufti et al., 1989) or tissue fragments can restore regenerative ability to irradiated hosts and
change sexual behavior to that of a donor (BaguA et al., 1989; Lange and Gilbert, 1968; Lender
and Garcia, 1965; Wolff and Dubois, 1948). We sought to determine whether small numbers of
cNeoblasts would restore regenerative ability to irradiated animals. Following irradiation, some
animals were fixed and colony numbers determined; remaining animals were followed to assess
survival and regeneration frequencies. Increasing irradiation doses resulted in decreasing colony
numbers (Fig. 3A-B) and survival rates (Fig. S15A). Regeneration, which involves production
of diverse cell types (Fig. S16), was initially impeded in animals cut 4 days post-irradiation (Fig.
S15B-C); however, many animals both survived and ultimately regenerated at doses that
produced sparse, measurable colony numbers (Fig. 3A-D, Table S1). These animals regenerated
heads containing neurons (ectoderm), muscle (mesoderm), and intestine (endoderm) (Fig. 3D-E).
The minimum number of cNeoblasts initially present in irradiation survivors can be estimated by
comparing the number of colonies present (in fixed animals) to observed regeneration
81
frequencies (see Table Sl). Our data indicate as few as three (P=0.0478), four (P=0.0017), or
five colonies (P<0.0001, Fisher's Exact Test) can be sufficient to restore regenerative ability to
entire animals.
Transplantationof individualcNeoblasts
To determine whether a single cNeoblast can generate all essential adult cells, we developed a
method for isolating and transplanting individual cNeoblasts into lethally irradiated hosts.
Previous flow cytometry studies identified an irradiation-sensitive cell population with a high
percentage of smedwi-J+ cells (Hayashi et al., 2006; Reddien et al., 2005b).
However, the
Hoechst 33342 DNA dye used in this method is cytotoxic. Therefore, size and complexity
properties of cells within the Xl fraction were used to define a gate for sorting unlabeled cells,
which we refer to as the X1(FS) fraction (Fig. 4A-B).
X1(FS) cells are heterogeneous; however, cells with a similar morphology to Xl cells can
be identified microscopically (Fig. 4C). Single selected cells were loaded into needles and
transplanted post-pharyngeally into lethally irradiated hosts (Fig. 4D). To confirm that only
single cells were transplanted, needles were loaded and the contents expelled into media. In all
cases, only a single cell was observed exiting the needle (n=136/136).
Furthermore, some
animals were fixed immediately following transplant and labeled with a smedwi-J RNA probe.
All injected animals had either one (n=20/60) or zero (n=40/60) smedwi-J+cells (Fig. 4E-F).
If a transplanted cell was a cNeoblast capable of engraftment, then clonal growth of
progeny cells would be expected.
Indeed, animals examined 6 days after single cell
transplantation displayed clusters with 1 to 13 smedwi-J+ cells (n=23/100) (Fig. 4G).
82
Furthermore, selecting for XI(FS) cells that were approximately 10pm in diameter and that had
blebs and/or cytoplasmic processes increased engraftment rates, ranging from 12% (n=17) to
75% (n=20) (Fig. S17). Cells with properties of cNeoblasts, therefore, are present in the Xl(FS)
fraction and can be successfully transplanted.
If cNeoblasts are pluripotent stem cells capable of self-renewal, then a single cNeoblast
should, in principle, be capable of restoring tissue turnover and regenerative capacity to lethally
irradiated hosts. However, for this to occur the irradiated host must survive long enough for the
cNeoblast to repopulate the smedwi-1] population and replenish dying tissue. We therefore used
a sexual S. mediterranea strain (S2F1L3F2) that can survive longer than the asexual strain
(CIW4) after a 6,000 rad irradiation dose (Fig. S18).
Sexual hosts transplanted with single
asexual cells had colonies consisting of large numbers of SMEDWI-1+ cells 30 days after
transplantation (n=4/17) (Fig. 4H).
Every colony examined contained SMEDWI-1*; Smed-
gata4/5/6 double-positive cells (n=4/4) and most of these colonies also contained SMEDWI- 1;
Smed-chat' double-positive cells (n=3/4) (Fig. 4H). Transplant data thus independently confirm
attributes of colonies described post 1,750 rads, further indicating that clonal growth and
multipotency are important features of individual cNeoblasts.
Entire animals and strains regeneratedfrom a single transplantedcNeoblast
Two weeks after irradiation, lesions appeared at sexual animal head tips, followed by progressive
anterior to posterior tissue regression with 100% penetrant animal death after approximately six
weeks (n=78) (Fig. 5A). Remarkably, several transplant recipients lived past seven weeks and
eventually developed blastemas at the site of tissue regression (n=7/130) (Fig. 5A). Animals that
83
developed blastemas regenerated anterior and midbody structures, such as photoreceptors and
pharynges (Fig. 5A), and regained feeding behavior by eight weeks after irradiation. Of the
seven rescued animals, three were expanded into strains (RI, R2, and R3) by serial amputation
and regeneration (Fig. S19).
These animals exhibited normal blastema formation and the
capacity to regenerate photoreceptors and intestine following amputation (Fig. 5B). The ability
to produce multiple regenerating animals from a single transplanted cell indicates a selfrenewing capability of cNeoblasts. Rescued strains did not display sexual features, such as large
size and a gonopore; by contrast, animals in all three strains reproduced by binary fission, an
asexual behavior seen very rarely in sexual animals (Fig. S 19).
To confirm that all new tissue in rescued strains resulted from clonal division from the
donor cNeoblast, we genotyped the animals using SNPs identified between the asexual strain and
the sexual strain (see Materials and Methods). Genomic DNA was isolated from strain Ri, R2,
and R3 animals following two rounds of regeneration; growth and regeneration should replace
host tissues with donor-derived cells (Fig. 6A).
If, on the other hand, host cells continue to
replenish tissues after irradiation, host SNPs in the collected genomic DNA would be expected.
PCR-RFLP analysis of two loci (RFLP 00310 and RFLP 00463) revealed that the rescued strains
have the asexual strain RFLPs, indicating that the majority of cells in these animals were donorderived (Fig. 6B). Sequencing of three independent homozygous haplotypes (00163, 00463, and
02716), each containing six SNPs that distinguish asexual CIW4 and sexual S2F1L3F2 strains,
confirmed that the rescue strains possessed the donor, rather than host, genotype (Fig. 6C).
These data indicate that descendants of a single cNeoblast ultimately transformed the recipient
into a genetic clone of the donor by replacing all cells present in the original host. We conclude
84
that cNeoblasts are pluripotent stem cells with a broad, body-wide distribution and that
persistence into adulthood of a pluripotent stem cell enables the remarkable regenerative feats of
planarians.
85
Figure 1
1I
EV
J1 aw
JAI
Rads
B
A
D
1 00
93.7%
(468/499)
80 "'
60
40
A 20
4
Ime chumbm 111Cm
12cjm
0 1 2 3 4 >5
# Clusters I Worm
F
-
4)
1000
size
6
4
100
2 Cr
4) 4
7 10
1I
Frequency 0
0
5 10
15 20
Days after Irradiation
86
D
Figure 1. Expanding colonies are generated from isolated smedwi-1+ cells following
irradiation
(A-B) Proliferating cells were detected by smedwi-J expression using whole-mount in situ
hybridization (ISH). Anterior, up. Ventral surface shown. (B) Representative images 7 days
after 1,750-rad treatment show clusters (arrowheads) of smedwi-J+ cells (individual purple dots).
(C) Histogram of cluster frequencies following I 750 rads. (D) Clusters observed by smedwi-J
ISH 7 days post-1,750 rads displayed in a scatterplot. phx, pharynx. (E-F) Animals fixed in a
timecourse after 1,750-rad treatment analyzed by smedwi-) fluorescence in situ hybridization
(FISH). (F) Mean cluster frequency (#clusters/worm) and size (#smedwi-1J cells/cluster) are
plotted. Error bars, standard deviation (n=17-22 animals/timepoint). (G) IF (BrdU) and FISH
(smedwi-1).
234/234 BrdU+ cells (8-hour BrdU-pulse in seven-day-irradiated worms) were
smedwi-J+. (H) IF (SMEDWI-1) and FISH (smedwi-1); 12/12 colonies contained SMEDWI-1*;
smedwi-1~ cells (arrowheads) 7 days post-1,750 rads. (I) IF (BrdU) and FISH (smedwi-1). 31/31
colonies (with BrdU pulse days 7-11 post-1,750 rads) contained BrdU+; smedwi-~ cells. Scale
bars, 200pm (A-B), 20pm (E, G-I).
87
Figure 2
E Colony
D
ALLocations
t-
-
88
Figure 2. Clonogenic neoblasts display broad differentiation capacity
(A-C) Triple-labeling of individual colonies 22 days after irradiation. Shown are projections
through optical sections from irradiated animals. Left, tiled images (images from overlapping
regions assembled) of representative animals with individual colonies are shown (anterior, up).
Circles indicate approximate location of region imaged at high magnification for middle panels;
middle images are optical sections with anterior to the right. Example differentiating cells from
individual colonies labeled by IF (SMEDWI-1) and double FISH for gata4/5/6 and chat (A),
gata4/S/6 and AGAT-1 (B), or AGAT-1 and chat (C) are shown.
Proportions of colonies
displaying multiple differentiating cell types are indicated. Roman numerals indicate doublepositive cells, with individual channels shown in columns to the right. Additional double-positive
cells are indicated with arrowheads. See also Figure S12. (D) IF (BrdU) and double FISH (Smedchat; Smed-mat) worms with BrdU-pulse days 14-18 after irradiation.
Single colonies (n=7)
contained both BrdU+; chat- (neuronal) and BrdU+; mat* (intestinal) descendants.
Boxes
indicate zoomed regions. (E) Scatterplots showing locations of individual colonies producing
differentiated cell types (see also Figure S 13).
Colony cell differentiation was assessed by
labeling with SMEDWI-1 (circles) or BrdU (diamonds). Scale bars, (A-C) left, 100im; middle
20ptm; right 5p[m; (D) top image, 20tm; others 5Rm.
89
Figure 3
smedwi*1
C
B
-
.4
100
.
>121
.
A
----
so
C
with 26 colonies
moo animis
10,
0w
E
with a3 colonies
LM 60-
+4.-4---.--84 ------ *4
*
40
-
-1,260 Rads
--------... % animals
1,500 Reds
0
C,
Ana
40II
44
-4--i---00 **
.4
+
-
0
-......--
C
6,
4,
++
4W
4
2,
6,000 Rads
01
---
--- - - -
0
t-- ----
-
1,750 Rads
2-- -
20
Irradiation Dose (Rads)
Irradiation Dose (Rads)
90
Figure 3. Small numbers of cNeoblasts can restore tissue turnover and regenerative ability
(A-B) Animals were irradiated at different doses. Some of these animals were fixed 7 days postirradiation (1,000 rads, 25 animals; between 1,125 and 1,875 rads, >38 animals/dose; for 6,000
rads, 26 animals) and labeled by smedwi-J+ ISH. (A) Representative smedwi-1] ISH images.
Anterior, left. (B) Colony numbers/worm are plotted (each dot represents one animal). (C) The
percentage of animals with restored regeneration following irradiation (>98 worms/dose were
examined; animals were from the same irradiated cohort as in A-B). Data indicate 3 or more
cNeoblasts were sufficient to restore regeneration (see also Table S1).
(D) Normal head
regeneration in 97/99 worms amputated 39 or 40 days after 1,250 rads. (E) Heads regenerated
after irradiation contained differentiated neuronal (chat), intestinal (mat-]+), and muscle (mhc1) cells (41/41 worms, 1,250 rads; 15/15 worms, 1,500 rads). SMEDWI-1* cells were also
restored (n=9/9 worms, 1,250 rads). Dotted lines, approximate amputation plane. Arrowheads,
photoreceptors. Scale bars, 200pm (A), 20Rm (D-E).
91
Figure 4
BIE
i
Hoechst Red
G
|
Day 9
0%
-F
FS
a
er 6,000
23%
WSo 231100
1 UnInjected
Il
Injecled
92
p
Figure 4.
Single transplanted cNeoblasts display properties of clonal growth and
multipotency
Irradiation-sensitive cells (polygonal gate) were identified by Hoechst 333342 labeling (A) and
back-gated to set the X1(FS) gate (oval) based on size (FS) and complexity (SS) parameters (B).
The X I(FS) fraction is heterogeneous and contains some cells approximately 1 Otm in diameter
with processes (arrowhead) (C). (D) Individual cells were loaded into needles (one needle used
per injection) and transplanted into the medial, post-pharyngeal, parenchymal space of hosts. (EF) FISH (smedwi-1) of a host immediately after transplantation. Anterior, up. Ventral surface
shown.
Zero (n=40/60) or one (n=20/60) smedwi-1] cells were observed in all cases, with
expected size and morphology. (F) is a zoomed image of(E). (G) Colony formation 9 days after
irradiation, 6 days after transplant. Anterior, up. Ventral surface shown. Colonies of smedwi-J+
cells (arrowhead) appeared in transplant recipients (n=23/100) but not in untreated animals
(n=5). (H) IF (SMEDWI-1) and double FISH (Smed-gata4/5/6; Smed-chat) 33 days after
irradiation, 30 days after transplant.
Single colonies were observed (n=4/17); example
differentiating cells from displayed colony are shown. Scale bars, 10pim (C), 50tm (E), 5pm (F)
and zoomed images in (H), 20ptm (H), 200tim (G).
93
FIgure 5
A
DaV 15
<
Day 54
Dy3
Dy3
ppx
B
ay
0
Day
8
Dy1+
px
I..
94
Dy 6+
Figure 5. Restoration of regeneration in lethally irradiated hosts by single transplanted
cNeoblasts
(A) Representative images of transplant hosts. Tissue regression (asterisks) began anterior to
photoreceptors (arrowheads) and progressed from anterior to posterior (px, pharynges). Rescued
animals developed blastemas (arrow) at regression site (dotted line) after seven weeks and fully
regenerated after eight weeks. Anterior, up. Dorsal surface shown. (B) Representative images of
rescue strains undergoing regeneration following amputation. Blastemas formed at approximate
amputation plane (dotted line). Intestine (labeled with red food coloring) and photoreceptors
(arrowheads) were observed in blastemas after 12 days of regeneration. Anterior, up. Dorsal
surface shown. Scale bars, 1mm (A), 50O0m (B).
95
Figure 6
A
Number of Reads
C
Asexual HaDlotvye I Sexual Haplotvye
Asexual (Donor)
Sexual (Host)
Rescue Strain 1
94
Rescue Strain 2
94
Co
Rescue Strain 3
Locus
0E00163
0 00463
96
Figure 6. Genotype conversion by single transplanted cNeoblasts
(A) Schematic showing replacement of host tissue by transplanted donor cells (blue); animals for
genotyping were amputated (dotted line) and allowed to regenerate twice. (B) PCR-RFLP
analysis of rescued strains. Locus 00310 was cut by Hpal in sexual animals (S) but not asexual
animals (A) or the rescued strains (1, 2, 3). Locus 00463 was cut by Sca in asexual animals and
the rescued strains, but not in sexual animals. (C) Haplotype sequencing. Stacked histogram
representing number of sequencing reads from each locus for each strain. Bars extend left for
number of reads corresponding to the asexual haplotype and right for number of reads
corresponding to the sexual haplotype. Bar absence indicates no reads.
97
Figure SI
C
B
0
|
Sing le Clusters, 7 da s after 1,750 Rads
0
Untreated
98
Figure S1. smedwi-1+ clusters originate in ventral positions throughout the body
(A-B) Transverse tissue sections shown, dorsal up. Control worms display smedwi-)+ cells
distributed in parenchyma throughout the dorsal-ventral axis (A). Representative image of a
smedwi-J+ cluster (arrowhead) seven days post-1,750 rad treatment (B). (C) Representative
images of animals displaying single smedwi-1+ clusters seven days post-1,750 rad treatment.
Whole animals, anterior up. Individual clusters were scattered throughout body, but consistently
displayed compact, isolated, colony-like morphology. (D-E) Triple FISH for synapsin (central
nervous system), mat (intestine), and smedwi-) seven days after 1,750 rads. Nuclei are labeled
with Hoechst. Shown are projections through Apotome optical sections. Individual smedwi-J'
clusters were distributed throughout the ventral regions and were not invariantly associated with
specific organs or anatomical features. Some clusters had cells adjacent (within -1 cell diameter)
to the ventral nerve cords (28/44); some clusters were located large distances from the nerve
cords (16/44). Scale bars 20mm (A-B, D-E), 200mm (C).
99
Figure S2
100
Figure S2. Following irradiation, clusters are preceded by isolated smedwi-1] cells
(A-E) Representative images of smedwi-J (FISH) after 1,750 rad exposure. Whole animal,
anterior up. Magnified regions are indicated by boxes. Arrowheads denote individual smedwi-J'
cells. Three days after irradiation, the majority of animals (16/20) displayed isolated smedwi-J+
cells (A). Animals with >1 smedwi-J+ cell in close proximity (within 50mm) were rare (2/20)
(B). Remaining animals at this condition (2/20) displayed no smedwi-JI cells (C). Animals fixed
8 days post-irradiation displayed either small smedwi-J+ clusters (9/19) (D) or were devoid of
smedwi-J+ cells (E). Based on these proportions, the possibility that all 8-day clusters arose from
multiple cells can be excluded (P = 0.0138, 2-tailed Fisher's Exact Test). Scale bars 200mm (ae), 20mm (zoomed boxes).
101
Figure S3
cesny, S pmSt.
11 C.Ifly ,w
poa
W
D
* sne*4-1
Ssmdwf-1* ; BrdU+
~75
* BrdU*
25
60
0
Hour pos4kdU
C
12hr befor Irradiation
; BrdU
C
-
7
atr1,70 Radb
102
Figure S3. BrdU incorporation and proliferation of smedwi-1] colony cells
(A) Immunofluorescence (BrdU) and FISH (smedwi-J). All dividing cells (571/571) detected in
unirradiated, intact worms by a 12-hour BrdU pulse expressed smedwi-J+ (see also, Guo et al,
2006). Thus, smedwi-J+ cells are the sole source for BrdU-labeled cells in these animals. (B-D)
BrdU incorporation and dilution kinetics in expanding clusters of smedwi-J+ cells. Animals were
injected with BrdU 7 days after 1,750 rads and fixed at various timepoints. Representative
immunofluorescence (BrdU) and FISH (smedwi-1) images are shown (B-C). Percentage of
smedwi-1, BrdU+, and smedwi-1J; BrdU+ (double-positive) cells are indicated for each
timepoint (n> 154 cells analyzed/timepoint) (D). BrdU is rapidly incorporated into a majority of
smedwi-J+ cells within the first 8 hours following labeling. Following a brief 48-hour chase
period, the first signs of BrdU dilution from the continuously dividing smedwi-1 population are
evident. At 24 hours post-BrdU, 123/133 (92.5%) smedwi-1 cells were BrdU+, compared with
91/119 (76.5%) at 48 hours post injection. Fisher's Exact Test indicates that this difference is
significant (P=0.0006, 2-tailed). Unincorporated BrdU, therefore, is unlikely to remain within
injected animals for longer than 48 hours post-injection (see also, Eisenhoffer et al, 2008). (E)
Animals were irradiated (1,750 rads) 12 hours after BrdU injection and fixed 5-6 days later.
Shown is a representative cluster following IF detection of BrdU+ cells and FISH (smedwi-1).
Several colonies of 3-12 smedwi-J+ cells were identified (n=9), and nearly all of these cells
(47/48) contained BrdU+ nuclei (BrdU signal is anticipated to be absent from some cells after
many cell divisions). (F-H) Double FISH seven days after 1,750-rad irradiation. All proliferating
cells expressing Smed-h2b (F, 304/304), Smed-RRM2 (G, 111/111), or Smed-pcna (H, 166/166)
were smedwi-] +-colony cells. Scale bars 200mm (A-E), 20mm (zoomed images, F-H).
103
Figure S4
Irradiation Only
Amputation after Irradiation
C
,
A
25/54
Colonies
I
29/54
D
31/55
24/55
Colonies
No Colonies
I
I
| No Colonies
Irradiation Only
I
104
I
Figure S4. Amputation and feeding fail to stimulate new colony formation following
irradiation
(A-D) Animals were exposed to 1,750 rads and fixed 12 days later for smedwi-1 ISH. Shown are
whole animals (anterior up) amputated into fragments five days after irradiation (A-B), or left
intact (C-D). Dotted lines indicate amputation plane. Amputated fragments from individual
worms were cultured, fixed, and stained independently. The proportion of amputated worms
displaying colonies (25/54) was not significantly altered from that of intact worms (24/55),
indicating that amputation did not stimulate formation of additional colonies (P=0.8483, Fisher's
Exact Test, 2-tailed). (E-H) Animals were exposed to 1,750 rads and fixed on day 12 for
smedwi-] FISH. Animals were either fed four days after irradiation (E-F), or left untreated (GH). The proportion of fed worms displaying colonies (18/49) was not significantly different from
that of control worms (14/54), indicating that feeding failed to stimulate formation of additional
colonies (P=0.2885, Fisher's Exact Test, 2-tailed). Scale bars, 200 mm.
105
Figure S5
Irradiated 1500 Rads
250-
........
ft
.....
mm :1:0-0-0-oooel m
-m
..........
..........
200
..... .... ft
E
-
G
...ft..mmm ......
150
cjn
0
100
50
C
5
15
iO
Days After 1,500 Rads
El
7
10
|[
106
Figure S5. smedwi-1+ cells and mitotic activity are restored to normal levels following
irradiation with 1,500 rads
(A-F) Whole-mount smedwi-J ISH (A-C) and IF detection of Histone-H3 (phospho-serine-10)+
mitotic cells (D-F). Representative images demonstrate that the number and body-wide
distribution of both smedwi-1
cells and mitotic activity are gradually restored following
irradiation at 1,500 rads. (G) A timecourse illustrates that mitotic numbers (normalized by worm
area) are almost completely depleted 7 days after irradiation. Successive timepoints display
increased levels of mitotic activity. Furthermore, the increase in mitotic activity slowed between
days 17 and 21, suggesting that colony expansion is a regulated (rather than neoplastic) process.
Shown are means and standard deviations for each data point. Mean and standard deviation
mitotic levels for untreated worms (184.6
22.4 mitoses / mm2 tissue) are represented by solid
and dotted lines, respectively. Scale bars, 200mm.
107
Figure S6
108
Figure S6. A subset of SMEDWI-1 cells undergo neuronal differentiation
Double-labeling analysis of cephalic ganglia in intact, unirradiated planarians. Anterior, up. (A)
IF (SMEDWI-1) and FISH (Smed-chat) identifies SMEDWI-1I cells co-expressing Smed-chat, a
marker of differentiated neurons (double-positive cells are indicated by arrowheads). Zoomed
regions are indicated by boxes. These cells were enriched in brain regions and often adopted a
non-neoblast cell morphology that included long axon-like cytoplasmic processes, suggesting
differentiation into neurons. (B-C) Double FISH in intact, unirradiated planarians indicates that
Smed-chat+ cells exist in the planarian CNS and co-express Smed-synapsin (b) and Smedproprotein- convertase-2 (Smed-PC2) (C), two independent markers of the planarian nervous
system. (D) A representative FISH image shows that Smed-chat mRNA is localized within or in
close proximity to the nucleus. Scale bars, 20mm (A-C), 5mm (zoomed images, D).
109
Figure S7
MmGata3
HsGata3
DrGata3
821802
HsGata2
MmGata2
DrGata2
TcGatu123
GATA 112/3
AgGata 23
AmGatal23
DpGata123
73 1 735
DmGraln
871 939
NvGata
HsGatal
92/941 MmGatal
DrGatal
601761
MmGata4
HsGata4
921 930
DrGata6
MmGata6
57/330
HsGataG
MmGata6
HsGata5
DrGata5
AmGata46Sbba
GA l A 415/6
AgGata456bba
TcGata466bba
87/845
DmPannler
521/ 504
- 5DpGata46bb
DpGatata456ba
AgG456ba
TcGata456bn
~ ~AmGata466ba
$71939
mrpn
DpGaui456a
TcGata456a
41/556
80
L-=64'
694
Ag
Smed-Gata4/5/6
i 801
DmGatae
AmGata456a
110
Figure S7. Phylogenetic analysis of the Smed-gata4/5/6 gene
Maximum likelihood and neighbor-joining analysis provide strong support for the Schmidtea
mediterraneagene Smed-gata4/S/6 falling within the GATA4/5/6 clade with known protostome
members of this family. Genes used in this analysis are well-established representatives of the
GATAl/2/3 or GATA4/5/6 gene families (Gillis et al., 2008). Accession numbers for genes
listed in this tree can be found in ref (Gillis et al., 2008). Mm, Mus musculus; Hs, Homo sapiens;
Dr, Danio rerio; Tc, Tribolium castaneum; Ag, Anopheles gambiae; Am, Apis mellhfera; Dp,
Daphnia pulex; Dm, Drosophila melanogaster; Nv, Nematostella vectensis; Smed, Schmidtea
mediterranea.
111
Figure S8
112
Figure S8. A subset of SMEDWI-1* cells express Smed-gata4/5/6
Double-labeling analysis of posterior intestinal branches in untreated or lethally- irradiated
planarians. (A-B) Double FISH on untreated (A) or 5 day, lethally irradiated (B) planarians
indicate that Smed-gata4/S/6 is expressed in fully differentiated Smed- mat+ cells of the planarian
intestine. Smed-gata4/S/6 is also expressed in isolated, irradiation-sensitive cells associated with
the intestine (arrowheads). (C) Many isolated Smed-gata-4/5/6+ cells were co-labeled by
SMEDWI-1 (IF). (D) A representative FISH image shows subcellular localization of Smedgata4/S/6 mRNA. Scale bars, 50mm (A- C), 5mm (zoomed images, D).
113
Figure S9
C.NHR2 5
I NR5
NR3
100 100 DERR
ER
HSERRa
02DmDHR4
99/
NR6
7CNHR91
7
NvNR1
811433
DmDHR38
681 984
HsNGFIB
981976
CNHR6
78) 435
I
NR4
HsRORa
100
11001
CoNHR23
S100 /
1000
*
DmDHR3
DmUSP
-
=99/ 1000
I
HsRXRa
NR2-B
DmDHR78
CNHR41
0 1971814
HsTR2
811433
831659
66
NR2-E
DmDSF
10
691I963
--
59
NvNR10
100 11000
HsCOUPTFa
941I998
DmSVP
NR2-F
CoNHR49
Smed-HNF4
641N923
81
DmHNF4NRNvNR4
sNF4I
114
NR2-C
I
NRI
Figure S9. Phylogenetic analysis of the Smed-hnf4 gene
Maximum likelihood and neighbor-joining analysis provide strong support for the Schmidtea
mediterraneagene Smed-hnf4 falling within the NR2A/HNF4 family of nuclear receptors. Genes
used in this analysis are well-established representatives of the six families of nuclear receptor
genes NRI-6. Sequences for Nematostella genes used in this trte can be found in ref (Reitzel and
Tarrant, 2009). Hs, Homo sapiens; Dim, Drosophilamelanogaster;Nv, Nematostella vectensis;
Ce, Caenorhabditiselegans; Smed, Schmidtea mediterranea.
115
Figure SIO
116
Figure S10. A subset of SMEDWI-1+ cells express Smed-hnf4
Double-labeling analysis of posterior intestinal branches in untreated or lethally irradiated
planarians. (A-B) Double FISH on untreated (A) or 5 day, lethally irradiated (B) planarians
indicate that similarly to Smed-gata4/5/6, Smed-hnf4 is expressed in filly differentiated Smedmat' cells of the planarian intestine. Smed-hnf4 is also expressed in isolated, irradiation-sensitive
cells associated with the intestine (arrowheads). (C) Many isolated Smed-hnf4+ cells were colabeled by SMEDWI-l (IF). (D) A representative FISH image shows subcellular localization of
Smed-hnf4 mRNA. Scale bars, 50mm (A-C), 5mm (zoomed images, D).
117
Figure S11
118
Figure S11. Characterization of post-mitotic cell populations
(A-F) Double FISH analysis of post-mitotic cell types in unirradiated, intact planarians (see also,
Eisenhoffer et. al., 2008). Shown are ventral, prepharyngeal regions of the animal. A previously
reported panel of marker genes labels multiple populations of cells. "Category 3" gene SmedAGAT-1 only partially overlapped in expression with other category 3 genes Smed-MCP-1 (A),
Smed-ODC-1 (B), Smed-CYPJAJ-J (C), and NB.52.12F (D). However, cells expressing SmedRas-related, another category 3 gene, showed extensive overlap with the Smed-AGAT-J+
population (E). Similarly, "category 2" markers NB.21.11 E and NB.32.1G were expressed by
the same population of cells (F). A revised panel of six markers NB.21.11E, Smed-AGAT-1,
Smed-MCP-1, Smed-ODC-1, Smed-CYPJA1-1, and NB.52.12F therefore encompasses a
heterogeneous set of cell populations with minimal redundancy. (G-L) FISH showing presence
of NB.21.1 lE, Smed-AGAT-1, Smed-MCP-1, Smed-ODC-1, Smed-CYPJAJ-1, and NB.52.12Fexpressing cells. Shown are ventral, anterior regions from untreated or day 19-irradiated worms
lacking smedwi-]+ colonies. All six cell types were depleted after 1,750 rads.
119
Figure S12
120
Figure S12. Clonogenic neoblasts display broad differentiation capacity
Triple labeling of individual colonies 22 days after irradiation. Shown are additional cells from
the same colonies depicted in Fig. 2. Each row of panels shows individual channel images for a
single cell. (A-B) Additional examples of colony cells positive for chat (A) and gata4/S/6 (B)
expression from the same colony shown in Fig. 2A. (C-D) Additional examples of colony cells
positive for AGAT-J (C) and gata4/S/6 (D) expression from the same colony shown in Fig. 2B.
(E-F) Additional examples of colony cells positive for chat (E) and AGA T-J (F) expression from
the same colony shown in Fig. 2C. Scale bars, 5pm.
121
Figure S13
122
Figure S13. A high percentage of cNeoblast colonies produce differentiated cell types
Double-labeling
analysis of individual colonies 15-20 days after irradiation. Shown are
representative optical sections from irradiated animals. Boxes indicate zoomed regions within the
same colony that contain double-positive cells. Locations of all colonies in these analyses are
shown in a scatterplot in Figure 2E. (A-B) Two representative colonies labeled by IF (SMEDWI1) and FISH (Smed-chat) from worms 15-19 days after 1,750-1,800 rads. 29/30 such colonies
contained
differentiating
neurons
(SMEDWI+;
Smed-chat+ double-positive
cells).
(C)
Representative colony labeled by IF (BrdU) and double FISH (Smed-chat; smedwi-1) from 1,750
rad-treated animals. Worms with growing colonies were injected with a pulse of BrdU 14 days
after irradiation and examined 4 or 5 days later to detect differentiating cells. Every such colony
examined (n=8/8) contained Smed-chat*; BrdU+ double-positive cells. (D-E) Two representative
colonies labeled by IF (SMEDWI-1) and FISH (Smed-gata4/5/6) from worms 19 days after
1,750 rads. 14/15 such colonies contained differentiating intestinal cells SMEDWI-1*; Smedgata4/5/6+ cells. (F-G) Two representative colonies labeled by IF (SMEDWI-1) and FISH
(Smed-hnf4) from worms 19 days after 1,750 rad exposure. 10/11 such colonies contained
SMEDWI-1; Smed-hnf4+ cells. Scale bars, 20mm (5mm, zoomed images).
123
Figure S14
D
1750
E
+7
124
Figure S14. Nearly all colonies locally produce post-mitotic cell types
(A-E) smedwi-J' colonies 20 days after 1,750 rads contained post-mitotic cells expressing
NB.21.l lE (A, 16/16 colonies), NB.52.12F (B, 13/15 colonies), CYPA1-1 (C, 15/17), MCP-1
(D, 20/20 colonies), and ODC-1 (E, 20/20 colonies). (F) Representative colony labeled by
double FISH (smedwi-1 and NB.21.1 lE) from worms 7 days after exposure to 1,750 rads. Nearly
all individual
colonies analyzed (14/15) examined
differentiating cells. Scale bars, 20mm.
125
at this early timepoint displayed
Figure S15
A
10
----------- -- ---
0
--.....
40
------------------ -
- --
20
------------------ -
----
- - ----.......
---......
......... -.
-......-----..
0
---....................
0
0
10
20
30
40
50
60
Days After Irradiation
-1,000
-1,125
-1,250
-1,375
rads
rads
rads
rads
1,500
-1,750
-1,875
-6,000
rads
rads
rads
rads
126
Figure S15. Effects of irradiation on planarian survival and regeneration
(A) Survival curves of irradiated worms irradiated at various doses of irradiation. Animals are
from the same experiment shown in Figure 3. Viability decreased sharply above 1,500 rads
(n>98 worms/sample). (B-C) Head regeneration was initially impeded by even low doses of
irradiation. Shown are head regions from worms 8 days post- amputation. (B) Control worm. (C)
Worm amputated 4 days after exposure to 1,000 rads (49/49 animals). Similar results were
obtained after 1,250 rads (50/50 worms), or 1,500 rads (50/50 worms). Scale bars, 20mm.
127
Figure S16
128
Figure S16. Diverse types of differentiating cells are produced during head regeneration
Planarians were injected with BrdU 18 hours prior to decapitation. IF (BrdU) together with FISH
shows 5-day-regenerating heads containing newly formed (BrdU+) differentiated cells. These
cells include (A) excretory, (B) muscle, (C) intestinal, and (D) neuronal lineages as determined
by expression of Smed-carbonic-anhydrase (Smed-CA), Smed-myosin-heavy-chain-J (Smedmhc-1), Smed-methionine-adenosyltransferase(Smed-mat), and Smed-choline-acetyl-transferase
(Smed-chat), respectively. Scale bars, 100mm. Anterior, up. Dotted line, approximate amputation
plane.
129
Figure S17
i
I
10%
2120
130
Figure S17. Morphological characteristics and transplant frequencies of different X1(FS)
cells
(A-E) Representative images of X1(FS) cells. Nuclei were labeled with Hoechst 33342. (F-J)
Representative images of animals 9 days after irradiation, 6 days after transplant. Anterior, up.
Ventral surface shown. Transplantation of cells 10-14mm in diameter with low cytoplasmic
granularity (A) resulted in few smedwi-]+ clusters (n=2/20) (F). Transplantation of cells 8-10mm
in diameter with low cytoplasmic granularity (B) resulted in zero smedwi-J+ clusters (n=0/20)
(G) Transplantation of cells 10-12mm in diameter with low cytoplasmic granularity and
processes (C) resulted in many formed smedwi-1* clusters (n=15/20) (H). Transplantation of
cells 10-14mm in diameter with high cytoplasmic granularity (D) resulted in few formed
smedwi-J+ clusters (n=5/20) (I). Transplantation of cells 8-10mm in diameter with high
cytoplasmic granularity (E) resulted in few formed smedwi-J+ clusters (n=1/20) (J). Scale bars,
10mm (A-E).
131
Figure S18
120-
.
100-
E
0
M Asexual Strain
Sexual Strain
6040-
20
0
10
20
30
40
Days Post Irradiation
132
50
60
70
Figure S18. Survival of asexual and sexual strain animals following lethal irradiation
Exact survival times vary between experiments. In this particular experiment, asexual CIW4
animals exposed to a 6,000 rad dose of radiation had a median survival period of 17 days and a
longest survival period of 21 days (n=105). Sexual S2F1L3F2 strain animals exposed to identical
conditions within the same experiment had a median survival period of 39 days and a longest
survival period of 63 days (n=97).
133
Figure S19
Irradiation
Transplantation
Day 50
Day 100
Day 200
Day 150
I
_
_
_
_
Rescue Strain 1
Rescue Strain 2
a
0
Rescue Strain 3
T--O
-4
Survivor 4
m
-
Survivor 7
-
Survivor 5
Survivor 6
134
1111
Figure S19. Timeline of rescued transplant hosts
Trees detailing survival of rescued animals, with branches indicating expansion of a single
animal into multiple individuals by amputation or fissioning. Red end points indicate natural
death. White end points indicate sacrifice of the individual for experimental purposes. Green
intersections indicate amputations to expand the population. Blue intersections indicate
fissioning events. Black arrowheads indicate which individuals are still alive at the time of
writing.
135
100
100
1 (1)
0
1 (1)
0
40
26
2(5)
0
1(3)
0
0
0
0
0
0
0
1875
6000
99
3(3)
4(4)
38
9(24)
5(13)
3(8)
2(5)
0.0017
0.0032
colonies
5 or more
0.0478
4 or more
colonies
and restore regenerative ability.
as few as three (P=0.0478), four (P=0.001 7), or five colonies (P<0.0001) colonies are sufficient to rescue entire animals
frequencies of smedwi-1 colonies observed from in situ data, a set of predictions can be generated based on a
hypothetical minimum number of colonies required for restoring regeneration. Fisher's Exact test (2-tailed, a = 0.05)
facilitates a direct comparison between the number of colonies present and number of worms regenerated. P-values for
tests in which significantly more animals regenerated than predicted are shown in bold. Together these data indicate that
and regenerated are shown. Nearly all worms that survived also displayed normal head regeneration. Given the
Planarians were exposed to a range of irradiation doses. A portion of these animals were fixed 7 days later and colonies
visualized by smedwi-1 ISH. The number and percentage (in parentheses) of animals displaying various numbers of
colonies are shown. N indicates the total number of animals analyzed. Remaining animals were followed for several
weeks and decapitated 39-40 days after irradiation. Number and percentage of total worms that survived or both survived
1750
1(3)
28 (28)
30 (30)
57
24 (42)
16 (28)
2 (4)
1500
11(19)
100
4(7)
0.0302
<0.0001
100
80(80)
82(82)
51
47 (92)
43 (84)
34 (67)
32 (63)
1375
39 (76)
0.0002
99
97(98)
98(99)
52
52 (100)
47 (90)
44(85)
41(79)
1250
52 (100)
100
98
25 100(100)
47 98(100)
25(100)
47(100)
25(100)
47(100)
25(100)
47(100)
25(100)
46(98)
25(100)
46(98)
1000
1125
Survived
N
3 or more
colonies
4 or more
colonies
5 or more
colonies
6 or more
colonies
Dose
(Rads)
2 or more
colonies
100(100)
98(100)
6 or more
colonies
Fisher's Exact Test (2-tailed)
N
# (%) worms recovered (d54-55)
Survived &
Regenerated
# (%) worms with smedwi-1* colonies (d7)
Table S1. A small number of smedwl-r* colonies can rescue entire animals from irradiation and
restore regenerative ability
Materials and Methods
PlanariaCulture and Irradiation
Schmidtea mediterraneaasexual strain C1W4 was maintained as described (SAnchez Alvarado et
al., 2002). Animals were starved in the presence of Gentamicin (Gibco) for ten days prior to
irradiation experiments. Individual irradiation experiments used size-matched animals with
identical feeding and culturing histories. Irradiation was delivered to animals at 79-82 rads/min
using dual Gammacell-40 '"Cesium sources (i.e., sources positioned both above and below the
specimen). For survival experiments, irradiated animals were maintained in 6 cm Petri dishes (10
worms/dish) in the dark; water and dishes were changed every 3-4 days.
In situ Hybridizationand Tissue Sectioning
Whole-mount in situ hybridizations (ISH) and fluorescent in situ hybridizations (FISH) were
performed and RNA probes prepared as described (Pearson et al., 2009). Tyramide-conjugated
fluorophores were generated from AMCA, Fluorescein, Rhodamine (Pierce), and Cy5 (GE
Healthcare) NHS esters as previously reported (Hopman et al., 1998). For double/triple labeling,
HRP- inactivation was performed between labelings in 4% formaldehyde, 45 min. Tissue
sectioning was performed as previously described (Reddien et al., 2005b).
BrdUlabeling and Immunofluorescence
Animals were fed or injected with 5 mg/mL BrdU (Fluka) as previously described (Newmark
and Sanchez Alvarado, 2000). Specimens were fixed in 4% formaldehyde as described (Pearson
137
et al., 2009) and antibody labelings performed as reported (Reddien et al., 2005b) using rat antiBrdU (1:100, Oxford Biotech), rabbit anti-H3P (1:100, Millipore), or rabbit anti-SMEDWI-1
(Guo et al., 2006) (1:2000).
Microscopy
Microscopy images were captured with an AxioCam HRm on a Zeiss Stereo Lumar V12 or an
Axio Imager ZI using Zeiss Axiovision software. Double-positive cells were scored in optical
sections obtained with an Apotome (Zeiss). Additional images were collected on a Zeiss LSM
700 confocal microscope using Zen software.
PhylogeneticAnalysis
Putative members of gata4/S/6 and hnf4 gene families were identified in the Schmidtea
mediterranea genome. Peptide sequences were aligned with well-known members of these
families using ClustalW with default settings (Thompson et al., 1994). Alignments were trimmed
using GBlocks (Castresana, 2000). Neighbor-joining trees were generated using ClustalW using
default settings and 1,000 bootstrap replicates. Maximum likelihood analyses using 100
bootstrap replicates were run on each alignment using PhyML (Guindon and Gascuel, 2003) with
WAG model of amino acid substitution, four substitution rate categories, and the proportion of
invariable sites estimated from the dataset. Maximum likelihood trees are shown in
Supplemental Figures 7 and 9. Maximum likelihood bootstrap values greater or equal to 50
(50%) and neighbor-joining bootstrap values greater than 500 (50%) are indicated in bold and
italics, respectively.
138
XJ(FS) Cell Collection
Animals were starved for at least seven days prior to harvesting. For control cells, animals were
macerated in 1.0 mg/ml collagenase (Sigma) and 0.3 mM N-acetyl-L- cysteine (Sigma) for 1
hour and labeled in 0.4mg/ml Hoechst 33342 (Invitrogen) for 45 minutes. For transplant cells,
animals were macerated in 1.0 mg/ml collagenase and 0.3 mM N-acetyl-L-cysteine for 20
minutes. The X1 population from Hoechst-labeled control cells was used to define the forward
scatter/side scatter gate. Cells were sorted with a Dako Cytomation MoFlo sorter.
Single Cell Transplantation
Animals to receive transplants were starved in the presence of Gentamicin for at least seven days
prior to onset of experiments. Three days prior to transplantation, irradiation was delivered to
animals at 79-82 rads/min for 76 minutes. Cells collected by flow cytometry were loaded at low
density onto glass cover slips treated with 2% dimethyldichlorosilane (Sigma) in chloroform.
Individual cells were selected based on morphology with lOx magnification and loaded by
mouth pipetting into the tip of pulled borosilicate glass microcapillaries (Sutter) treated with
0.1% polyvinylpyrrolidone (Sigma). Loaded cells were injected into the post-pharyngeal midline
of cold- immobilized animals at 1.5-2.5 psi (Eppendorf FemtoJet). For survival experiments,
transplant recipients were maintained in 6 cm Petri dishes (3 worms/dish) in the dark; water and
dishes were changed every 3 days.
139
SNP Discovery
Short (36 bp) sequencing reads from both asexual Clone 4 and sexual S2FlL3F2 strain expressed
sequences were obtained by mRNA-Seq (Illumina). Reads were mapped to an assembly of
planarian expressed sequences and Single Nucleotide Polymorphisms (SNPs) were identified
with MAQ using default settings (Li et al., 2008). Only SNPs based on at least 10X read depth
for both strains were considered. Candidate loci were selected based on presence of multiple
homozygous SNPs. SNP-containing loci were validated by PCR and Sanger sequencing (see
SNP sequencing below).
Genomic DNA Isolation
Animals used for genotyping were subjected to two rounds of regeneration and starved for at
least five days. DNA was isolated from intact regenerated animals using Easy- DNA kit
(Invitrogen).
PCR-RFLPAnalysis
Restriction Fragment Length Polymorphism (RFLP) loci were amplified from genomic DNA
samples (Finnzymes Phusion Polymerase). PCR product was purified and digested with
restriction enzymes (HpaI or ScaI-HF, New England Biolabs) for two hours. Digested DNA was
purified by phenol-chloroform extraction and run on a 1.4% agarose gel.
SNP Sequencing
SNP loci were amplified from genomic DNA. PCR products were gel-purified, A-tailed (Roche
140
Taq Polymerase), and ligated into pGEM-T Easy vector (Promega). 94 bacterial colonies from
each locus for each strain were Sanger sequenced with M13F primer. For genotyping, reads
were counted if at least four SNPs corresponding to a single haplotype were present.
RFLP loci sequences
> RFLP 00310 (S2FlL3F2)
TCGGATACAGTAAATCACCTGATACTATTGCTACGGGCTATTCTGGTGAT
GCTCCCCCGTCTATAACTGCTGCACAATTACAACTGAGTCCTGGTCAAGC
GGACACGGGATACGTATCATTGACGTGGAATATACTGACCCAGTCCGACA
TTGCCACGAATGTGAACGGATTTTTCCGTGGATATCGAATTGAATGGTGC
TTGGCAAACCTGATTGATGCGGAATGTGATGCATCAACTCAATATCAGGT
AAATAGGTAATTAGACGTTTTATGTTTAATATTATAAGGATGTGATTCTC
GCAACACAAAACCTCCCGGTTCTTTATGGAAATAAGCGCCGTAAAAGATC
AGTACAAGATGATGATGAAGACACACAGACGGATGAAGGAAGATTTAAAT
ATGATACTAAATATCGCCAGGTCATACCAGACAACCCAGCAACGTCAGTT
AATTTTCAAGTCTTAAGTCGTAGAAAACGAGCTGCATTAAAAAATCCTGA
TGATTGGAATTATGGAAAAAATATCACTGTAAAATTGACGATGATTCCAG
GCAATACTTGGATCAAGGTTTGGCTGAGAGTTTTGAAT
> RFLP 00310 (CIW4)
TCGGATACAGTAAATCACCTGATACTATTGCTACGGGCTATTCTGGTGAT
GCTCCCCCGTCTATAACTGCTGCACAATTACAACTAGGTCCTGGTCAAGC
GGACACGGGCTACGTATCATTGACGTGGAATATACTGACCCAGTCCGACA
TTGCCACGAATGTTAACGGATTTTTCCGTGGATATCGAATTGAATGGTGC
TTGGCAAACCTGATTGATGCAGAATGTGATGCATCAACTCAATATCAGGT
AAATAGGTAATTAGACGTTTTATGTTTAATATTATAAGGATGTGATTCTC
GCAACACAAAACCTCCCGGTTCTTTATGGAAATAAGCGCCGTAAAAGATC
AGTACAAGATGATGATGAAGACACACAGACGGATGAAGGAAGATTTAAAT
ATGATACTAAATATCGCCAGGTCATACCAGACAACCCAGCAACGTCAGTT
AATTTTCAAGTCTTAAGTCGTAGAAAACGAGCTGCATTAAAAAATCCTGA
TGATTGGAATTATGGAAAAAATATCACTGTAAAATTGACGATGATTCCAG
GCAATACTTGGATCAAGGTTTGGCTGAGAGTTTTGAAT
> RFLP 00463 (S2FlL3F2)
ATCGGATCACCTATCAATATTTGCCTCCGGCTGCATTCAACATTGAACTC
GTTCCGCAATCTTCATCAGCCAATAACAGCAGCAAAACATCTTCGGATTG
CCACAGGAATTCAGATGGCAGCCGGAAATTGAGATCGCATACTCTCCCAG
GCGACAAAATCGCTCCTGTTGTCATTGGCAATGCGCCCGCTCAACAGTCG
GCCTCCACAGCAGATTCGCCTATCATGGCAACGAGAAACCTTCGCGGATG
GATTGTAATACTCAAGGAGTACTTTGGATTCGTGGAAACGGCCGATCACA
141
ACGCGCTATACAAGTTCAGCCCGTTCACAATCAAGAAGAGCAAATTGGGA
GTGGAATTGAAGGTTGGCTCGGCGATTGAATTTCTGGCGGTCCCGAGCTC
TGGCAGTCGGCCTCGTCGCATCATTGAGCAGTTCCTGAAGGTCCTCACCG
AGCCGTTATCCAATGAG
> RFLP 00463 (CIW4)
ATCGGATCACCTATCAATATTTGCCTCCGGCTGCATTCAACATTGAACTC
GTTCCGCAATCTTCATCAGCCAATAACAGCAGCAAAACATCTTCGGATTG
CCACAGGAATTCAGATGGCAGCCGGAAATTGAGATCGCATACTCTCCCAG
GCGACAAAATCGCTCCTGTTGTCATTGGCAATGCGCCCGCTCAACAGTCG
GCCTCCACAGCAGATTCGCCTATCATGGCAACGAGAAACCTTCGCGGATG
GATTGTAATACTCAAGGAGTATTTTGGATTCGTGGAAACGGCCGATCACA
ACGCGCTATACAAGTTCAGCCCGTTCACAATCAAGAAGAGCAAATTGGGA
GTGGAATTGAAGGTTGGCTCGGCGATTGAATTTCTGGCGGTCCCGAGCTC
TGGCAGTCGGCCTCGTCGCATCATTGAGCAGTTCCTGAAGGTCCTCACCG
AGCCGTTATCCAATGAG
Sequences of SNP loci
> SNP 00163 (S2F1L3F2)
CCCAGTGAAAAACCCAAACATGGAAATGAAGATTGTTTCAATACATTTTT
CAGTGAGACTGGAAATGGAAAATATGTTCCTCGGGCTCTTTTTGTCGATT
TGGAACCAAGTGTAATTGGTAAGTTTTAGAATTTTGTGTTTTATATTTTT
AATGCTCTCCTGCTTTAGGTGAAGTGAGAAATGGGGCTTATAGACAACTG
TTCCATCCGGAACAACTTATTAGTGGTAAAGAAGATGCAGCTAATAATTA
CGCAAGAGGACATTATACAGTGGGTAAAGAATTGATCGATCAAGTTTTAG
ATAGAATTAGAAAGGTTGCTGATAATTGTACCGGTTTGCAAGGGTTTCTA
ATGTTTCATTCATTTGGTGGTGGAACTGGTTCCGGGTTTACTTCTCTGTT
AATGGAACGGTTAAGTGTTGATTATGGTAAAAAATCCAAGTTAGAGTTTG
CTGTTTATCCTGCTCCACAAATCG
> SNP 00163
(CIW4)
CCCAGTGAAAAACCCAAACATGGAAATGAAGATTGTTTCAATACATTTTT
CAGTGAGACTGGAAATGGAAAATATGTTCCTCGGGCTCTTTTTGTCGATT
TGGAACCAAGTGTAATTGGTAAGTTTTAGATTTTTGTGTTTTCTATTTTT
AATGCTCTCCTGCTTTAGGTGAAGTGAGAAATGGGGCTTATAGACAACTG
TTCCATCCGGAACAACTTATTAGTGGTAAAGAAGATGCAGCTAATAATTA
CGCAAGAGGACATTATACAGTTGGGAAAGAATTGATCGATCAAGTTTTAG
ATAGGATAAGAAAGGTTGCTGATAATTGTACCGGTTTGCAAGGGTTTCTA
ATGTTTCATTCATTTGGTGGTGGAACTGGTTCCGGGTTTACTTCTCTGTT
AATGGAACGGTTAAGTGTTGATTATGGTAAAAAATCCAAGTTAGAGTTTG
CTGTTTATCCTGCTCCACAAATCG
> SNP 00463
(S2F1L3F2)
GACGATATTGCCGGATTGATTGAGCCGGATTTATCGGTAAAGGGTGATGA
142
GGACATGCTGGTAGCGTTCAAGGCCAAGGAATGGCTACCAAGCGATGAGA
AACTGACAAAGTTTGCGGTAGTTTCCTACACGCCGTTGCAAGTGTCCAGG
AACGTCGGAGGCCACCAGAAGCTGACGACAACCGCGCTGCGCATCATGCC
ATCAAAGGAGAGGGACAAAATGACCCACCTGAAGCTGTATCCGGCCGGCC
AGTTACGGGGAGTTGTCAATACTGCAGTTCACAATCCCGATGATCTCGGC
ATCATTTTCTGTCAGTTTCCCAATTCATCTACCAAACGAACGGTCGCATT
TACCAGTGAGGATCTTGTCAACTGCAAGCCCAAGGT
> SNP 00463
(CIW4)
GACGATATTGCCGGATTGATTGAGCCGGATTTATCGGTAAAGGGCGATGA
GGACATGCTGGTAGCGTTCAAGGCCAAGGAATGGTTGCCAAGCGATGAGA
AACTGACAAAGTTTGCGGTAGTTTCCTACACGCCGTTGCAAGTGTCCAGG
AACGTCGGAGGCCATCAGAAGCTGACGACAACCGCGCTGCGCATCATGCC
ATCAAAGGAGAGGGACAAAATGACCCACCTGAAGCTGTATCCGGCCGGCC
AGTTGCGGGGAGTTGTCAATACTGCAGTTCACAATCCCGATGATCTCGGC
ATCATTTTCTGTCAGTTTCCCAATTCATCTACCAAACGGACGGTCGCATT
TACCAGTGAGGATCTTGTCAACTGCAAGCCCAAGGT
> SNP 02716
(S2F1L3F2)
GCTTTCGTCTATATGTTAGAGCGTTTCATCAATACTTCACTGACTTCGGT
AAATAAATGGTGTGCTATAACATTTTTTGTTTACCTATTTACATAGGATC
AAGTACAAAATCATTTGTTAGATCTTTGCCAAATAATTTCTTCACCGAAA
AAAGAAAAAATTATCGATTTCTTGAAAAAAGAAGGACCTAAATTGATTCC
CCGAATACTTAAAAAAGACTGTCCAATAAAGATTTGTCAAATGGAAAACT
TTTGTCACAAAATCGAAGTTATATTCGAAAATGATGATAAGGTTCCAAGT
AATTTAATTTACTCTAAGGGATTTTCGAATATATCTTCTAATTAGATCCA
TCAAAATATTGTTCGATCTGCCAATCGG
> SNP 02716
(CIW4)
GCTTTCGTCTATATGTTAGAGCGTTTCATCAATACTTCACTGACTTCAGT
AAATAAATGGTGTGCTATAACATTTTTTGTTTACCTATTTACATAGGATC
AAGTACAAAATCATTTGTTAGATCTTTGCCAAATAATTTCTTCACCGAAA
AAAGAAAAAATTATCGATTTCTTGAAAAAAGAAGGACCTAAATTGATTCC
ACGAATACTTAAAAAAGACTGTCCAATGAAGATTTGTCAAATGGAAAACT
GTTGTCACAAAATCGAAGTTATATTCAAAAATGATGATAAGGTTCCAAGT
AATTTAATTTACTCTAAGGGATTTTCGAATATATCTTCTACTTAGATCCA
TCAAAATATTGTTCGATCTGCCAATCGG
Primers for PCR-RFLP analysis.
RFLP_00310_F1
RFLP_00310_R2
RFLP_00463_F1
5'-TCGGATACAGTAAATCACCTGATAC-3'
5'-ATTCAAAACTCTCAGCCAAACC-3'
5'-ATCGGATCACCTATCAATATTTGC-3'
RFLP_00463_Ri
5'-GATAACGGCTCGGTGAGGAC-3'
Primers for SNP loci sequencing.
143
SNP_00163_Fl
SNP 00163 F2
SNP 00163_Ri
SNP 00163 R2
SNP 00463_Fl
SNP 00463_F2
SNP_00463_Ri
SNP 00463 R2
SNP 02716 Fl
SNP 02716 F2
SNP_02716_Ri
SNP_02716_R2
5'-GTCTTGAACATGGTATTCAACAAGA-3'
5' -CCCAGTGAAAAACCCAAACA-3'
5'-ATGGTTCAACAACCGCTGTA-3'
5'-CGATTTGTGGAGCAGGATAAA-3'
5'-CTCGACATATCGGAGTTGTGAA-3'
5'-GACGATATTGCCGGATTGA-3'
5'-CAACTAACTGACAGGCAGCAAC-3'
5'-ACCTTGGGCTTGCAGTTG-3'
5'-TCACGATGGAAACCAAAAAG-3'
5'-GCTTTCGTCTATATGTTAGAGCGTTTC-3'
5'-TTTTCTAAGGCTACCCAGCTGAT-3'
5'-ACCGATTGGCAGATCGAA-3'
144
Acknowledgements
We thank D. Kim for manuscript comments, S. Lapan for intestinal markers, M. Srivastava for
phylogenetics advice, D. Wenemoser for SMEDWI-l antibody purification, J. Owen for Illumina
data, M. Griffin for flow cytometry assistance, as well as P. Hsu, G. Bell, R. Young, and all
members of the Reddien Lab for extensive comments and discussion. P.W.R. is an early career
scientist of the Howard Hughes Medical Institute.
R01GM080639 and Keck Foundation support.
145
We acknowledge support by NIH
References
BagufiA, J., Sal6, E., and Auladell, C. (1989). Regeneration and pattern formation in planarians.
III. that neoblasts are totipotent stem cells and the cells. Development (Cambridge, England) 7786.
BagufliA, J. (1976). Mitosis in the intact and regenerating planarian Dugesiamediterranean.sp. I.
Mitotic studies during growth, feeding and starvation. J. Exp. Zool. 195, 53-64.
Bardeen, C.R., and Baetjer, F.H. (1904). The inhibitive action of the Roentgen rays on
regeneration in planarians. J. Exp. Zool. 1, 191-195.
Barker, N., van Es, J.H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen, M., Haegebarth,
A., Korving, J., Begthel, H., Peters, P.J., et al. (2007). Identification of stem cells in small
intestine and colon by marker gene Lgr5. Nature 449, 1003-1007.
Becker, A.J., McCulloch, E.A., and Till, J.E. (1963). Cytological demonstration of the clonal
nature of spleen colonies derived from transplanted mouse marrow cells. Nature 197, 452-454.
Blanpain, C., Lowry, W.E., Geoghegan, A., Polak, L., and Fuchs, E. (2004). Self-renewal,
multipotency, and the existence of two cell populations within an epithelial stem cell niche. Cell
118, 635-648.
Bravo, R., Frank, R., Blundell, P.A., and Macdonald-Bravo, H. (1987). Cyclin/PCNA is the
auxiliary protein of DNA polymerase-delta. Nature 326, 515-517.
Castresana, J. (2000). Selection of conserved blocks from multiple alignments for their use in
phylogenetic analysis. Mol. Biol. Evol. 17, 540-552.
Dubois, F. (1949). Contribution A 1 'etude de la migration des cellules de regeneration chez les
Planaires dulcicoles. Bull. Biol. Fr. Belg. 83, 213-283.
Eisenhoffer, G.T., Kang, H., and SAnchez Alvarado, A. (2008). Molecular analysis of stem cells
and their descendants during cell turnover and regeneration in the planarian Schmidtea
mediterranea. Cell Stem Cell 3, 327-339.
Eriksson, S., GrAslund, A., Skog, S., Thelander, L., and Tribukait, B. (1984). Cell cycledependent regulation of mammalian ribonucleotide reductase. The S phase-correlated increase in
subunit M2 is regulated by de novo protein synthesis. J. Biol. Chem. 259, 11695-11700.
Evans, M.J. (1972). The isolation and properties of a clonal tissue culture strain of pluripotent
mouse teratoma cells. J Embryol Exp Morphol 28, 163-176.
Evans, M.J., and Kaufman, M.H. (1981). Establishment in culture of pluripotential cells from
mouse embryos. Nature 292, 154-156.
146
Gillis, W.Q., Bowerman, B.A., and Schneider, S.Q. (2008). The evolution of protostome GATA
factors: molecular phylogenetics, synteny, and intron/exon structure reveal orthologous
relationships. BMC Evol. Biol. 8, 112.
Guindon, S., and Gascuel, 0. (2003). A simple, fast, and accurate algorithm to estimate large
phylogenies by maximum likelihood. Syst. Biol. 52, 696-704.
Guo, T., Peters, A.H.F.M., and Newmark, P.A. (2006). A Bruno-like gene is required for stem
cell maintenance in planarians. Dev. Cell 11, 159-169.
Hayashi, T., Asami, M., Higuchi, S., Shibata, N., and Agata, K. (2006). Isolation of planarian Xray-sensitive stem cells by fluorescence-activated cell sorting. Dev. Growth Differ. 48, 371-380.
Hewitson, T.D., Kelynack, K.J., and Darby, I.A. (2006). Histochemical localization of cell
proliferation using in situ hybridization for histone mRNA. Methods Mol. Biol. 326, 219-226.
Hopman, A.H., Ramaekers, F.C., and Speel, E.J. (1998). Rapid synthesis of biotin-, digoxigenin, trinitrophenyl-, and fluorochrome-labeled tyramides and their application for In situ
hybridization using CARD amplification. J. Histochem. Cytochem. 46, 771-777.
Keller, J. (1894). Die ungeschlechtliche Fortpflanzungder Stlsswasser-Turbellarien. Jen Zeit
Naturw 28, 370-407.
Lange, C.S., and Gilbert, C.W. (1968). Studies on the cellular basis of radiation lethality. 3. The
measurement of stem-cell repopulation probability. Int. J. Radiat. Biol. Relat. Stud. Phys. Chem.
Med. 14, 373-388.
Lender, T., and Garcia, A. (1965). Les ndoblasts marqu6s par l"uridine tritide migrent et 6difient
le blasteme des planaires d"eau douce. C. R. Acad. Sci. Paris 160, 4095-4097.
Li, H., Ruan, J., and Durbin, R. (2008). Mapping short DNA sequencing reads and calling
variants using mapping quality scores. Genome Res. 18, 1851-1858.
Martin, G.R. (1981). Isolation of a pluripotent cell line from early mouse embryos cultured in
medium conditioned by teratocarcinoma stem cells. Proc. Natl. Acad. Sci. U.S.a. 78, 7634-7638.
Morgan, T.H. (1898). Experimental studies of the regeneration of Planariamaculata. Dev.
Genes Evol. 7, 364-397.
Morrisey, E.E., Tang, Z., Sigrist, K., Lu, M.M., Jiang, F., Ip, H.S., and Parmacek, M.S. (1998).
GATA6 regulates HNF4 and is required for differentiation of visceral endoderm in the mouse
embryo. Genes Dev. 12, 3579-3590.
Newmark, P.A., and Sanchez Alvarado, A. (2000). Bromodeoxyuridine specifically labels the
regenerative stem cells of planarians. Developmental Biology 220, 142-153.
147
Nishimura, K., Kitamura, Y., Taniguchi, T., and Agata, K. (2010). Analysis of motor function
modulated by cholinergic neurons in planarian Dugesiajaponica. Neuroscience 168, 18-30.
Ohlstein, B., and Spradling, A. (2006). The adult Drosophilaposterior midgut is maintained by
pluripotent stem cells. Nature 439, 470-474.
Pearson, B.J., and Stnchez Alvarado, A. (2010). A planarian p53 homolog regulates proliferation
and self-renewal in adult stem cell lineages. Development (Cambridge, England) 137, 213-221.
Pearson, B.J., Eisenhoffer, G.T., Gurley, K.A., Rink, J.C., Miller, D.E., and Sanchez Alvarado,
A. (2009). Formaldehyde-based whole-mount in situ hybridization method for planarians. Dev.
Dyn. 238,443-450.
Reddien, P.W., and SAnchez Alvarado, A. (2004). Fundamentals of planarian regeneration.
Annu. Rev. Cell Dev. Biol. 20, 725-757.
Reddien, P.W., Bermange, A.L., Murfitt, K.J., Jennings, J.R., and SAnchez Alvarado, A. (2005a).
Identification of genes needed for regeneration, stem cell function, and tissue homeostasis by
systematic gene perturbation in planaria. Dev. Cell 8, 635-649.
Reddien, P.W., Oviedo, N.J., Jennings, J.R., Jenkin, J.C., and SAnchez Alvarado, A. (2005b).
SMEDWI-2 is a PIWI-like protein that regulates planarian stem cells. Science 310, 1327-1330.
Reitzel, A.M., and Tarrant, A.M. (2009). Nuclear receptor complement of the cnidarian
Nematostella vectensis: phylogenetic relationships and developmental expression patterns. BMC
Evol. Biol. 9,230.
Salvetti, A., Rossi, L., Bonuccelli, L., Lena, A., Pugliesi, C., Rainaldi, G., Evangelista, M., and
Gremigni, V. (2009). Adult stem cell plasticity: neoblast repopulation in non-lethally irradiated
planarians. Developmental Biology 328, 305-314.
SAnchez Alvarado, A., Newmark, P.A., Robb, S.M., and Juste, R. (2002). The Schmidtea
mediterranea database as a molecular resource for studying platyhelminthes, stem cells and
regeneration. Development (Cambridge, England) 129, 5659-5665.
Scimone, M.L., Meisel, J., and Reddien, P.W. (2010). The Mi-2-like Smed-CHD4 gene is
required for stem cell differentiation in the planarian Schmidtea mediterranea.Development
(Cambridge, England) 137, 1231-1241.
Spangrude, G.J., Heimfeld, S., and Weissman, I.L. (1988). Purification and characterization of
mouse hematopoietic stem cells. Science 241, 58-62.
Thompson, J.D., Higgins, D.G., and Gibson, T.J. (1994). CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence weighting, positionspecific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673-4680.
148
Till, J.E., and McCulloch, E.A. (1961). A direct measurement of the radiation sensitivity of
normal mouse bone marrow cells. Radiat. Res. 14, 213-222.
Uchida, N., Buck, D.W., He, D., Reitsma, M.J., Masek, M., Phan, T.V., Tsukamoto, A.S., Gage,
F.H., and Weissman, I.L. (2000). Direct isolation of human central nervous system stem cells.
Proc. Natl. Acad. Sci. U.S.a. 97, 14720-14725.
Wagers, A.J., and Weissman, I.L. (2004). Plasticity of adult stem cells. Cell 116, 639-648.
Wagers, A.J., Sherwood, R.I., Christensen, J.L., and Weissman, I.L. (2002). Little evidence for
developmental plasticity of adult hematopoietic stem cells. Science 297, 2256-2259.
Weissman, I.L. (2000). Stem cells: units of development, units of regeneration, and units in
evolution. Cell 100, 157-168.
Wenemoser, D., and Reddien, P.W. (2010). Planarian regeneration involves distinct stem cell
responses to wounds and tissue absence. Developmental Biology 344, 979-991.
Wolff, E., and Dubois, F. (1948). Sur la migration des cellules de r6g6ndration chez les planaires.
Rev Suisse Zool 55, 218-227.
149
150
Chapter 3
Hedgehog signaling regulates gene expression in planarian glia
Irving E. Wang', Sylvain W. Lapani, and Peter W. Reddien
'These authors contributed equally
Experiments shown in Figure 1-2 and Supplemental Figure Si, S3, S5 were performed by SWL
and IEW. Experiments shown in Figures 3-6 and Supplemental Figures S2, S4, S6-S8 were
performed by IEW. All authors contributed to the design of experiments and editing of the
manuscript.
151
Abstract
Hedgehog signaling has important roles in central nervous system (CNS) development in
vertebrates and Drosophila, yet it specifies different CNS cell types along distinct axes in these
disparate animals. Exploring the role of Hedgehog signaling in the nervous systems of a broader
range of animal phyla is critical for understanding which functions of Hedgehog in the CNS are
ancestral to the Bilateria, in contrast to those that represent clade-specific adaptations. Hedgehog
ligand is expressed in a population of medially-located neurons in the planarian brain. To
investigate the function of Hedgehog signaling in the brain, we performed RNA-Sequencing on
brain tissue samples from the planarian Schmidtea mediterranea following inhibition of
hedgehog (hh) and the gene encoding its receptor, patched (ptc), by RNAi.
No significant
differences were observed in the expression of transcription factors that specify neurogenic
domains, but transcripts for two genes expressed in the central nervous system, intermediate
filament-i (f-1) and protocadherin-19 (pcdh-19), were significantly misregulated.
if-1 and
pcdh-19 were expressed in cells within the neuropil of the planarian CNS and adjacent to hh+
neurons. Expression was not detected following hh RNAi and was present in ectopic locations
outside the CNS following ptc RNAi. if-J/pcdh-J9- cells did not express neural genes but did
express planarian orthologs of astrocyte markers, such as glutamine synthetase (gs) and
excitatory amino acid transporter-2/glt-J (eaat2). IF-1 processes associate closely with axons
and encapsulate regions of high synapse density. We conclude that if-1J/pcdh-19* cells represent
previously unidentified planarian glia.
Inhibition of hh or ptc did not block expression of
astrocyte markers, suggesting that certain aspects of gene expression, but not cell specification or
152
survival, was regulated by constitutive signaling from hh+ neurons. Hedgehog signaling has
been implicated in astrocyte response to brain injury in vertebrates and posterior midline glia
specification in Drosophila. We conclude that Hedgehog signaling confers the if-1*/pcdh-J9*
identity to planarian glia within the CNS and hypothesize the control of glial identity by
neighboring neuronal populations is an ancestral role for the pathway in the nervous system.
Furthermore, the identification of planarian glia presents a novel tractable system for dissection
of glia biology.
153
Introduction
The Hedgehog signaling pathway has been implicated in numerous developmental processes
across the Metazoa, including limb and midline development in vertebrates and segmentation in
Drosophila (Ingham et al., 2011).
Little is known, however, about the role of Hedgehog
signaling in the Lophotrochozoa, one of the three superphyla that comprise the Bilateria. Further
study and comparison with the other two Bilaterian superphyla, the Deuterostomes and the
Ecdysozoa, is required to understand the evolution of this signaling pathway and its roles in
Metazoan biology. One member of the lophotrochozoa, the planarian Schmidtea mediterranea,
has recently become a model system for the study of stem cell biology, wound responses, and
tissue patterning as a consequence of the development of powerful techniques such as RNA
interference (Reddien et al., 2005a) and fluorescent RNA in situ hybridization (Pearson et al.,
2009). Planarians are free-living platyheminthes capable of regenerating almost any lost tissue
through the continued expression of developmental factors and the maintenance of a pluripotent
stem cell population throughout adulthood (Morgan, 1898; Reddien and SAnchez Alvarado,
2004; Wagner et al., 2011).
Previous studies showed that perturbation of the Hedgehog
signaling pathway causes an anterior-posterior polarity defect during regeneration. Inhibition of
the planarian hedgehog ortholog (hh) resulted in bifurcated or absent tail formation and
inhibition of the patched ortholog (ptc) resulted in anterior tails in place of heads. These defects
are attributed to hh-mediated modulation of Wnt signaling during regeneration (Rink et al., 2009;
Yazawa et al., 2009).
154
The planarian central nervous system consists of a pair of cephalic ganglia and ventral
nerve cords, each structure comprised of a cortex of cell bodies encompassing a neuropil (Morita
and Best, 1965). hh is expressed in cells present in the medial cortex of the cephalic ganglia
(Rink et al., 2009; Yazawa et al., 2009), a similar location for a protostome to the vertebrate
neural tube floor plate (Dessaud et al., 2008). However, direct effects of Hedgehog signaling on
planarian nervous system regeneration have not been described, despite a wealth of information
of its involvement in the CNS of other systems. The vertebrate ortholog Sonic hedgehog (SHH)
is secreted from the floor plate and forms a ventral-to-dorsal morphogenetic gradient that
establishes distinct domains of transcription factor expression in the ventral neural tube (Dessaud
et al., 2008).
The Drosophila neurectoderm has a similar ventral-to-dorsal distribution of
orthologous transcription factors, but Hedgehog signaling is not required to establish these
domains (Cornell and Ohlen, 2000).
One neural tube domain specified by SHH generates
oligodendrocytes, the glial cells responsible for myelination (Rowitch, 2004). Additional roles
for Hedgehog signaling in glial cell biology include inducing reactive astrogliosis in response to
brain injury in adult mammals (Sirko et al., 2013) and specifying subtypes of midline glia during
Drosophila development (Watson et al., 2011). Examining the role of Hedgehog signaling in
planarian brain regeneration presents an opportunity to determine whether the pathway has
ancestral roles in the differentiation and regulation of CNS cell types.
Here we show that although the cephalic ganglion displays similarity in transcription
factor domain distribution to the vertebrate neural tube, perturbation of Hedgehog signaling does
not disrupt these expression domains. Through a tissue-specific mRNA-Sequencing approach,
we identified two CNS-associated genes, intermediatefilament-i (If-1) and protocadherin-19
155
(pcdh-19), with expression levels impacted by inhibition of either hh orptc. Expression of these
genes is found in cells with molecular and morphological characteristics of glia. Hedgehog
signaling is not required for the formation or maintenance of glia, despite having a strong ability
to modulate gene expression in these cells. This discovery presents an opportunity to not only
study the regulation and function of glia in a highly regenerative organism, but also gain insight
into ancestral roles of Hedgehog signaling in CNS development.
156
Results
Transcriptionfactor domain specification in the planarian cephalic ganglia occurs normally
following Hedgehog signalingpathway perturbation
Cells in the medial cortex of cephalic ganglia express hh continually throughout the life of adult
animals. This expression pattern is reminiscent of the vertebrate neural tube floor plate. SHH is
secreted from cells in the floor plate, establishing a ventral-to-dorsal gradient that results in the
formation of distinct domains of transcription factor expression.
Near the floor plate, two
homeodomain transcription factors, Nkx6.1 and Nkx2.2, are expressed and give rise to the pV3
progenitors. The pMN domain immediately dorsal to pV3 does not express Nkx2.2 but does
express low levels of the transcription factor Pax6. The pV2 domain has increased Pax6 levels
and continued expression of Nkx6.1 when compared to pMN, and the dorsal two domains of the
ventral neural tube, pVI and pVO, express Pax6 only (Dessaud et al., 2008).
Planarian orthologs of Nkx2.2, Nkx6.1, and Pax6 were identified as Smed-nkx2 (nkx2),
Smed-nkx6 (nkx6), and Smed-pax6b (pax6b), respectively.
nkx2 was expressed in the medial
cortex and part of ventral cortical region lateral to the medial cortex (Figure IA). nkx6 was
expressed more broadly, with expression concentrated most highly in the medial cortex but
extending more laterally than nkx2, reaching and extending slightly beyond the outer lobe of the
cephalic ganglia (Figure IB). pax6b had a complementary expression domain to nkx2. The
domain extends from the outer lobe to the chemosensory branches (Figure 1 C). These mediallateral expression domains were similar to the ventral-dorsal domains observed in the vertebrate
157
neural tube, with nhx2 being closest to the source of hh ligand, pax6b being furthest, no overlap
between nkx2 and pax6b, and overlap between nkx6 and both nkx2 and pax6b (Figure ID).
Loss of SHH expression in the vertebrate floor plate results in failed domain formation
and loss of the entire ventral neural tube (Ruiz i Altaba et al., 2003). To determine whether the
planarian CNS requires Hedgehog signaling to establish its transcription factor domains, hh and
ptc were targeted for inhibition by RNAi. Animals were fed a mixture of liver and bacteria
expressing dsRNA for a control gene not present in the planarian genome, hh, orptc a total of six
times over twenty days. The effects were examined under unamputated homeostatic conditions
as ptc(RNAi) animals are not expected to regenerate heads (Rink et al., 2009; Yazawa et al.,
2009). In both hh(RNAi) and ptc(RNAi) animals, no changes in the expression pattern of nkx2,
nkx6, and pax6b compared to control were observed in any animals (Figure 1E). Therefore,
under pathway perturbation conditions sufficient to obtain polarity defects, no role for Hedgehog
signaling in maintaining transcription factor expression domains in the planarian CNS was
detected.
mRNA-Seq identifies a set ofCNS-enrichedgenes affected by inhibitionof hedgehog
To identify nervous system defects resulting from perturbation of Hedgehog signaling, we
examined changes in gene expression within the CNS following inhibition of hh or ptc. Because
Hedgehog signaling is likely to have roles in homeostasis of other tissue types, we developed a
method to isolate CNS tissue for RNA collection and sequencing. Amputated head fragments
collected from CIW4 asexual strain animals after six RNAi feedings were used as reference
samples (Figure IF). For CNS-specific sample collection, large (>2cm) sexual animals were
158
dissected following a brief acid-based fixation that permitted the cephalic ganglia to be separated
from surrounding tissue and collected for RNA isolation (Figure 1 G). mRNA purified from both
head and cephalic ganglia samples from control, hh(RNAi), and ptc(RNAi) animals was used in
expression profiling by quantitative RNA-Seq.
The transcriptomes of the head and cephalic ganglia samples from control animals were
compared to determine whether the dissection technique enriched for neural transcripts.
Whereas the expression level of the housekeeping gene gapdh was similar between the two
samples, known neuronal transcripts synapsin (syn) (Nakazawa et al., 2003), prohormone
convertase 2 (pc2) (Collins et al., 2010), and choline acetyltransferase (chat) (Nishimura et al.,
2010) were between 2.6 to 7 fold increased in CNS-specific samples. The prohormone-encoding
gene 1020HH (Collins et al., 2010), which is expressed in cells in the cephalic ganglia and
ventral nerve cords, was enriched 81 fold. The relatively lower enrichment levels for a subset of
these genes can be explained by their expression in peripheral nervous system (PNS), cells of
which were present in the head fragment samples but not the cephalic ganglia sample.
All
examined genes with expression known to be restricted from the nervous system were, indeed,
greatly diminished in the CNS-specific sample (Figure 1H). hh, which is expressed in medial
cortical neurons, was enriched by 57-fold in CNS samples whereas ptc, which is expressed in
many tissues in planarians, showed little difference in expression (Figure 1 H, Supplemental
Table 1). The enrichment for neural transcripts indicates that analysis of these data can reveal
effects of hh and ptc inhibition on the central nervous system.
A comparison of hh(RNAi) and ptc(RNAi) animals to control animals showed
insignificant differences in transcript levels of the housekeeping gene gapdh and neural genes
159
syn and pc2. Expression of hh in hh(RNAi) animals was, as expected, significantly reduced (pdaj
< 0.05). ptc expression was decreased in ptc(RNAi) animals as well as in hh(RNAi) animals.
Patched protein is a negative regulator of Hedgehog signaling and acts as a repressor of ptc
transcription (Varjosalo and Taipale, 2008); increased Patched activity in animals lacking
Hedgehog ligand is likely responsible for the reduction of ptc transcript levels in hh(RNAi)
animals. Additionally, confirming our previous results, the expression levels of nkx2, nkx6, and
pax6b were not significantly changed in the hh(RNAi) and ptc(RNAi) animals versus controls
(Figure 11).
We selected a set of genes that fit the criteria of at least 2 fold enrichment or depletion in
hh(RNAi) or ptc(RNAi) samples and at least 1,000 reads per kilobase per million reads (RPKM)
to account for minor discrepancies when harvesting tissue (Supplemental Table 2). A wholemount in situ hybridization (WISH) screen for these genes identified two that were expressed in
the central nervous system (Supplemental Figure SOl).
The first gene, Smed-intermediate
filament-i (f-1), is an ortholog of cytoplasmic intermediate filament genes (Supplemental Figure
S02) (Kuo and Weisblat, 2011). Intermediate filaments are cytoskeletal proteins that provide
structural support and mechanical stress resistance in a variety of cell types (Herrmann et al.,
2007).
The second gene, Smed-protocadherin-19 (pcdh-19), although lacking identifiable
cadherin domains, has BLAST homology to protocadherin-19, a calcium-dependent cell
adhesion molecule that has been implicated in nervous system development and synapse
formation (Frank and Kemler, 2002).
160
Expression and localization of f
andpcdh-19 is alteredby Hedgehogpathwayperturbation
Fluorescent in situ hybridization (FISH) revealed that if-] and pcdh-19 are co-expressed
primarily in cells in the neuropil, the cell body sparse region within the neuron-dense cortex
(6.5% if-+, 4.0% pcdh-19+, 89.5% ifJ+/pcdh-19+)(Figure 2A, 2B). mRNA signal was detected
in processes extending from the cell body, a characteristic not commonly observed by FISH,
indicating that the morphology of the cells is likely to be highly branched (Supplemental Figure
SO3B).
Small numbers of cells were also observed outside the neuropil, in locations such as
within the cortex of the cephalic ganglia, in between the two lobes of the cephalic ganglia, and
sparsely throughout the ventral parenchyma (Supplemental Figure S03A, SO3B). if-J/pcdh-19+
cells both inside and outside the neuropil express ptc, indicating that these cells may be
responsive to Hedgehog signaling (Figure 2C).
While the majority of ptc+ cells within the
neuropil were if-J*/pcdh-19+, expression of if-] and pcdh-19 occurred in only a small subset of
ptc+ cells in the CNS cortex and parenchymal space. Additionally, if-]J/pcdh-19+neuropil cells
were adjacent to the hh+ neurons in the medial cortex, suggesting that these cells were not only
competent to respond to hh but also in close proximity to a source of the ligand (Figure 2D).
To characterize the effect of Hedgehog signaling on the neuropil cells, we performed
FISH for if-] and pcdh-19 following inhibition of hh or ptc and quantified the density of cells in
two regions of the head, inside the neuropil and outside the neuropil. Cells expressing one or
both of the two markers were found inside the neuropil of the cephalic ganglia at a density of
about 2200 cells/mm 2 of cephalic ganglion and outside the neuropil at a density of about 65
cells/mm 2 of head (Figure 2E, 2F).
Upon RNAi of hh, the density of if-1/pcdh-19+ cells
decreased 92% in the neuropil and 98% outside the neuropil (Figure 2E, 2F).
161
As ptc is a
negative regulator of Hedgehog signaling (Varjosalo and Taipale, 2008), RNAi of ptc is
expected to result in an increase in if-1 and pcdh-19 expression. Accordingly, the density of cells
expressing either or both if-1 and pcdh-19 in ptc(RNAi) animals increased 57% in the neuropil
and 615% outside the neuropil (Figure 2E, 2F). A portion of the increased cell count consisted
of cells expressing only one of the two markers, with if-1-/pcdh-19* comprising 38.2% of the
population (Figure 2H). Expression at ectopic locations outside the neuropil was localized to the
parenchyma near the ventral surface of the animal, with concentration of expression at the rim of
the head where presumptive chemosensory neurons reside (Supplemental Figure SO3A, SO3C).
To ensure that ablation of if-1 and pcdh-19 signal resulted from loss of Hedgehog
signaling, the Gli transcription factors, which are the downstream effectors of the pathway, were
also targeted for inhibition. gli-i and gli-2, encoding activating transcription factors, and gli-3,
encoding a repressing transcription factor, have been found in the S. mediterranea genome.
Inhibition of gli-i results in a similar defective tail regeneration phenotype as inhibition of hh
(Rink et al., 2009). We found that RNAi of gli-i also resulted in loss of if-1 and pcdh-19 signal
whereas RNAi of gli-2 and gli-3 did not have any discernable effect on expression of if-1 and
pcdh-19 (Supplemental Figure SO4A). These data indicate that the Hedgehog signaling pathway
regulates the biology of specific cell type in the axon rich neuropil of the planarian nervous
system.
During regeneration, if-1*/pcdh-19+ cells accumulate in the blastema, a protruding mass
of newly formed tissue that replaces missing portions of the body. As expected, no if-1+/pcdh19* cells were observed in hh(RNAi) animals, both in the blastema and throughout the preexisting tissue. Inhibition of plc results in defective head regeneration; the cephalic ganglia in
162
the anterior blastema appear as masses of cells without any discernable neuropil region. An
increase in the total number of ifI/pcdh-19* cells, however, was observed in ptc(RNAi) anterior
blastemas despite impaired head formation (Supplemental Figure SO4B).
if1/pcdh-19' cells are not neurons
To understand the role of Hedgehog signaling in the planarian nervous system, we sought to
identify the if-J+/pcdh-19+ cell type. Given the localization of these cells within the central
nervous system, we first tested for co-expression of neural markers. Both pc2 and syn were
expressed in a few cells in the neuropil region and but were not in any ifJ]/pcdh-19+cells inside
or outside the neuropil (Figure 2G, 2H, Supplemental Figure S05A, S05B). The planarian
genome contains a number of voltage-gated ion channels, some of which are expressed in a
CNS-like pattern.
Three genes, one encoding a potassium channel, one encoding a sodium
channel, and one encoding a calcium channel, were examined but did not show co-expression
with if-1 or pcdh-19 (Figure 21).
A gene encoding a sodium and potassium co-transporter
involved in maintaining ion concentration gradients in neurons also was not expressed in ifJ+/pcdh-19+ cells (Figure 21). Furthermore, choline acetyltransferase (chat), which encodes an
acetylcholine synthesis enzyme, glutamate decarboxylase (gad), which encodes a GABA
synthesis enzyme, tyrosine hydroxylase (th), which encodes a dopamine synthesis pathway
protein (Fraguas et al., 2011), tryptophan hydroxylase (tph), which encodes a serotonin synthesis
enzyme (Fraguas et al., 2011), were all not found in if-J+/pcdh-19+ cells even though their
expression was readily detected in other CNS cells (Figure 21). Finally, we examined genes
encoding orthologs of proteins involved in synapse function. Genes encoding orthologs of three
163
Synaptotagmin synaptic vesicle docking and fusion proteins, the synaptic vesicle associated
protein Synaptogyrin 2, synaptic vesicle fusion proteins SNAP25 and Unc-13, and the vesicular
neurotransmitter transporters VAchT and VGluT (Mairz et al., 2013) were not expressed in f
1*/pcdh-19+ cells (Figure 21).
Additional genes encoding synaptic proteins, neurotransmitter
biosynthesis enzymes, and transcription factors were also tested and showed no overlapping
expression with if1 and pcdh-19 (Supplemental Figure S05C). Netrin 2, a marker previously
described to be expressed in cells in the neuropil (CebriA and Newmark, 2005), also was not
expressed in if1/pcdh-19+ cells (Supplemental Figure SO5C). We conclude that, despite the
localization within the central nervous system and the presence of cytoplasmic extensions, the f
I/pcdh-J9' cells are not neurons.
if1J/pcdh-19* cells express neurotransmitterreuptake and metabolism genes
In addition to neurons, the other predominant cell type in the central nervous systems of various
animals is glia. Glia act as neuronal support cells by providing trophic support, axon insulation,
environmental maintenance, blood-brain barrier, and synapse pruning (Pfeiffer et al., 1993;
Sofroniew and Vinters, 2010).
Invertebrate glia have been extensively characterized in
Drosophila(Hartenstein, 2011) and C. elegans (Oikonomou and Shaham, 2011), and have been
identified in annelids (Deitmer et al., 1999) and molluscs (Reinecke,
1976).
Electron
microscopy performed on transverse sections of the planarian Dugesia tigrina revealed cells
distributed throughout the ventral nerve cords that lack neuronal characteristics; these have been
described as candidate planarian glial cells (Golubev, 1988; Morita and Best, 1966).
We
hypothesized that the if 1/pcdh-19+ cells might be glia, and therefore we performed a candidate
164
screen for expression of known glial markers from vertebrates and Drosophila in the planarian
neuropil.
Four astrocyte-associated genes were found to be co-expressed in if-J*/pcdh-19+ cells.
Smed-gs (gs) encodes an ortholog of Glutamine Synthetase, which converts the neurotransmitter
glutamate to glutamine (Anderson and Swanson, 2000), and was expressed in cells in the
neuropil as well as cells in the ventral parenchyma and the intestine (Supplemental Figure
S06A).
The majority of if-1/pcdh-19+ cells expressed gs both inside the neuropil (97%) and
outside the neuropil (72%), although gs*/if-1~/pcdh-19" cells were abundant outside the neuropil
(Figure 3A). Similar to if-1 and pcdh-19, gs transcripts were observed in cytoplasmic branches
extending from the cell body (Supplemental Figure S06B).
Smed-gat-1 (gat-]) is predicted to encode a protein with high similarity to GABA
transporters (Featherstone, 2011) and was expressed almost exclusively in cells within the
neuropil (Supplemental Figure S06C).
if-]J/pcdh-19+ cells showed almost complete overlap
with gat-] both inside the neuropil (95%) and outside the neuropil (99%) (Figure 3B). Unlike gs
expression, few gat-I+ cells outside the neuropil did not express if-1 or pcdh-19 (10%), and the
transcript appeared to be restricted to the cell body close to the nucleus (Supplemental Figure
S06D).
Smed-eaat2 (eaat2), which encodes a protein similar to the vertebrate glutamate
transporter and glial cell marker GLT-1/EAAT2 (Featherstone, 2011), was expressed most
abundantly in the neuropil but was also expressed in cells throughout the ventral parenchyma
with some concentration along the rim of the entire animal and in the pharynx (Supplemental
Figure S06E). FISH for eaat2 indicated that the gene is expressed in the majority of if-1+/pcdh-
165
19+ cells in the neuropil and to a lesser degree outside the neuropil (Figure 3C). Branches were
also observed in eaat2' cells (Supplemental Figure S06F). Smed-eaat3 (eaat3) encodes a protein
similar to a neuronally-expressed vertebrate glutamate transporter, EAAC1/EAAT3.
eaat3
expression was punctate and dispersed throughout the entire neuropil (Supplemental Figure
S06G, S06H). Whereas overlapping expression with if-1 and pcdh-19 was confirmed, it was
difficult to determine whether other cell types within the cephalic ganglia also expressed this
gene (Figure 3D).
Because if-*/pcdh-19+ cells express genes encoding proteins that reuptake extracellular
glutamate (eaat2 and eaat3) and metabolize glutamate (gs), we propose that these cells
participate in maintaining the extracellular environment, a role typically performed in vertebrates
by astrocytes. Together with the lack of neuronal marker expression, these results are consistent
with the possibility that if-1]/pcdh-19+ cells are planarian glia.
The similar expression of gs+ cells and ectopic if-l/pcdh-19+ cells in the ventral
parenchyma of ptc(RNAi) animals led us to further characterize marker expression outside the
neuropil. gs+ cells outside the neuropil were observed to express ptc at low rates, suggesting that
at least a subset of the cells might respond to Hedgehog signaling (Figure 3E). ptc was also coexpressed with gat-], eaat2, and eaat3 in a few cells outside the neuropil (Supplemental Figure
S061, S06J, S06K).
Cells expressing gs, gat-1, eaat2, or eaat3 did not express the pc2,
suggesting that they were not neurons (Figure 3I, Supplemental Figure S06L, S06M, S06N). In
the gs+ cell population outside the neuropil, some cells expressed eaat2 (Figure 3F) and very few
expressed gat-1 and eaat3 (Figure 3E, 3G), presumably because of the scarcity of those markers
in cells outside the neuropil. To ensure RNA probe specificity for its target molecule, WISH was
166
performed on animals following gs, gat-1, eaat2, or eaat3 RNAi.
Detection by in situ
hybridization of the target molecules was reduced in animals following inhibition of the same
molecule by RNAi (Supplemental Figure S060). Co-expression of the astrocyte markers and
lack of neuronal markers in these cells outside the neuropil raise the possibility that other glial
cell types with differing expression profiles may exist in the planarian peripheral nervous system.
IF-] protein localizes to cytoplasmic extensions that closely associatewith neurons
To examine the morphology of cells that express if-1, we raised a polyclonal antibody against a
peptide corresponding to a segment of the IF-1 protein primary structure.
Whole-mount
immunofluorescence revealed an extensive network of IF-i+ branches concentrated in the
neuropil and appearing in the parenchyma near the ventral surface (Figure 4A). RNAi of if-1
resulted in complete loss of IF-i antibody immunolabeling, confirming that labeling did not
result from cross-reactivity with other planarian proteins (Supplemental Figure S07A). In the
cephalic ganglia, the IF-I+ processes formed a mesh that congregated into tracts leading out of
the neuropil between the chemosensory branches to the periphery of the head (Figure 4B). The
IF-1 cytoplasmic extensions were also observed forming hollow columns oriented along the
dorsal-ventral axis (Figure 4C).
In the periphery, IF-i+ processes ran along tracts that were
mostly devoid of cell bodies (Figure 4D). These peripheral branches varied between animals in
extent, number, and location along the AP axis. In the ventral nerve cords, the processes ran
parallel to one another in a tight bundle in the dorsal domain of the VNC and formed lacunae in
the ventral portion of the VNC (Figure 4E, Supplemental Figure S07B). IF-i appeared to be
167
localized mainly to the processes of if-1 cells given the minimal overlap between the protein
and if-] transcripts (Supplemental Figure S07C).
To determine the association of the IF-i+ processes with neurons, if any, we performed
immunofluorescence in conjunction with antibodies against a-tubulin and synapsin. Antibodies
against a-tubulin label axons of both the central and peripheral nervous system in planarians
(Sanchez Alvarado and Newmark, 1999). Axons run parallel to the AP axis through the neuropil
and regularly exit the ventral nerve cords to form orthogonal commissures that extend from the
VNC to the edge of the body. The IF-1 processes emerging from the cephalic ganglion neuropil
appeared to follow the same tracts as the a-tubulin+ axon bundles (Figure 4F). A similar effect
was observed in the orthogonal branches throughout the parenchyma (Figure 4G). The IF-1I
processes were embedded within the nerve bundles and did not appear to fully enclose the
commissural axon fascicle.
An anti-Synapsin antibody labels large clusters of synapses within the neuropil and in
nerve plexuses in the grid-like network of commissural axon bundles called the Orthogon (Adell
et al., 2009; Reuter et al., 1998). Accumulations of synapses formed discrete, regularly spaced
structures in the ventral nerve cord that strongly resembled synaptic glomeruli described in insect
species (Boeckh and Tolbert, 1993; Leise, 1990). Immunofluoresence with both the anti-IF-i
antibody and the anti-Synapsin antibody showed IF-1 branches weaving through and filling
gaps within the synapse-dense cephalic ganglion neuropil (Figure 4H). Interestingly, the IF-1I
cytoplasmic extensions also appeared to fully encapsulate the synaptic glomeruli of ventral nerve
cords (Figure 41). The highly branched morphology of the if-1/pcdh-19* cells and their close
contact with both axons and areas of high synaptic density support our prediction that these cells
168
are acting in a similar fashion to astrocytes. We conclude that if-1/pcdh-19Y cells represent
planarian glia.
Perturbationof Hedgehog signalingaffects expression and localizationof IF-] protein
To determine the effect of Hedgehog signaling on intermediate filament protein expression and
localization, we performed IF-I immunofluorescence following RNAi of hh and ptc. Inhibition
of hh resulted in complete ablation of IF-i immunofluorescence signal and no change in
expression or localization of Synapsin protein, whereas inhibition of ptc caused increase of IF-I
protein throughout the animal (Figure 4J).
The increase observed in ptc(RNAi) animals
manifested primarily as an increase in the number of orthogonal commissures that came in
contact with an IF-I + process. Normally 15.1% of orthogonal axon bundles are associated with
IF-I + processes, whereas the percentage decreased to 2.1% following hh inhibition and increased
to 61.4% following ptc inhibition (Figure 4K). Because no peripheral branches in ptc(RNAi)
animals were observed deviating from the orthogonal axon network, and because Synapsin
localization was unchanged under ptc inhibition, IF- I+ processes appear strictly limited to axon
bundles even when ectopically formed (Supplemental Figure S07D).
Accumulation of IF-I
protein in ptc(RNAi) animals was also observed at the rim of the head where small Synapsin*
clusters are normally seen with minimal IF protein in control animals (Supplemental Figure
S07E).
The observation that IF-1 processes necessarily associate with axon bundles led us to
hypothesize that newly formed planarian glia use axon tracks to guide growing processes. To
determine whether IF-1 branches extend along a pre-existing neuronal network, we examined
169
the regeneration of these structures in the anterior blastema. Heads of wild type animals were
amputated and trunks were fixed at two-day intervals. Synapsin protein was detected in the
blastema after four days of regeneration, indicating the terminally differentiated neurons have
reformed an active network at that time point. IF-I protein was first detected in the blastema at
low levels at six days post-amputation and more substantially at eight days post-amputation. IF1+ branches in the blastema were always found associating with the Synapsin*
clusters that
formed earlier, suggesting that the glia cells extend their branches through pre-existing neuronal
networks (Figure 4L).
Inhibitionof hedgehog does not ablate glial cells
To determine whether inhibition of hh results in ablation of planarian glia, expression of gs, gat1, eaat2, and eaat3 was examined in hh(RNAi) and ptc(RNAi) animals.
We found that in
hh(RNAi) the expression of if-1 and pcdh-19 was eliminated, but expression of the other glia
markers was still observed throughout the neuropil and indistinguishable from control animals.
Similarly, inhibition of ptc had no effect on the expression or localization of gs, gat-1, eaat2, or
eaat3 (Figure 5A).
To confirm these results, we examined the differential expression of each
gene in the RNA-Seq data described above. The expression level differences between RNAi
samples were statistically significant for none of the four glia markers, and most had only log2
fold changes of less than 0.5 between RNAi conditions (Figure 5B). gat-] expression levels
displayed a similar trend to that seen for if-] and pcdh-19 transcript levels. Inhibition of hh
resulted in depletion of gat-1 transcripts in both cephalic ganglia and whole head samples, and
170
inhibition of ptc resulted in slight increase. Expression of gat-] may show a delayed or slower
response to inhibition of hh or ptc if it is an indirect target of Hedgehog signaling.
We quantified numbers of gs+ and gat-JX cells in animals following inhibition of hh or
ptc. eaat2 and eaat3 were omitted from this analysis due to their wider expression pattern in the
neuropil. We observed no significant change in the total number of gs+ cells within the neuropil
in hh(RNAi) and ptc(RNAi) (Figure 5C).
Similar results were obtained when examining
expression of gat-] in RNAi conditions (Figure 5D). However, we do observe a shift in the ratio
of gs+ or gat-JX cells that express if-1 and pcdh-19 to those that do not (Figure 5C, 5D). The
shift in ratio suggests that upon loss of Hedgehog signaling, gs+/gat-1+/if-1+/pcdh-19'cells stop
expression of if I and pcdh-19, or that the gs+/gat-1+/if-1+/pcdh-19+cells die and are replaced by
gs+/gat-JI/'f-JI/pcdh-19~cells.
In an effort to ablate the glial cells, we performed RNAi on candidate transcription
factors with known roles in glia specification in other systems.
We have identified two
homologs of the Drosophilagene glial cells missing, gcm1-1 and gcm1-2, which regulates the
formation of glia during development (Hosoya et al., 1995). We also extended our analysis to
two other genes, olig2 and nkx2, that encode orthologs of transcription factors necessary for
specification of oligodendrocytes in the vertebrate neural tube (Qi et al., 2001; Zhou and
Anderson, 2002). Following RNAi, animals were assayed for expression of if-1 and pcdh-19 by
WISH and were found to have no observable difference with control animals, both in
homeostasis and regeneration conditions (Supplemental Figure S08). Therefore, these homologs
of glial transcription factors do not appear to have an effect on planarian glia. We note that
171
vertebrates and flies do not share any known common glia specification factors, indicating that
gene regulatory networks for these cells may be highly divergent.
Hedgehogsignalingpromotes expression of if-1 andpcdh-19 in existing glia-like cells
To determine whether neuropil cells that lack if-1 and pcdh-19 expression in hh(RNAi) animals
are viable, we examined whether they are still specified during regeneration when the Hedgehog
pathway is perturbed. Following RNAi, heads and tails of animals were amputated and the trunk
blastemas were examined after six days of regeneration. Blastemas of control animals contained
cells expressing planarian glia markers, both inside the forming neuropil and outside (Figure
6A).
Similar results were observed in ptc(RNAi) animals (Figure 6A).
Because ptc(RNAi)
animals have defective anterior blastema formation, comparison of number of cells found within
the blastema could be unreliable. In hh(RNAi) animals, expression of if-1 and pcdh-19 was
eliminated, but cells expressing gat-1, eaat2, and eaat3 were observed throughout the blastema
(Figure 6A). The presence of these markers in newly formed cells of the blastema suggests that
the animal is capable of regenerating this cell type in the absence of positive Hedgehog
signaling.
There are two possible mechanisms for loss of if-1 and pcdh-19 in the neuropil glia
population.
The first possibility is that Hedgehog signaling promotes expression of these
specific genes and that loss of the ligand results in inactivation of those genes. The second
possibility is that Hedgehog signaling is responsible for differentiation of an if-1*/pcdh-19+
subtype of glia from a pool of existing gs*/gat-J+/eaat2+/eaat3+glia. If this were the case, then
we would expect continued expression of if-1 and pcdh-19 in neuropil glia following hh RNAi
172
until they were replaced during normal tissue turnover. We distinguished between these two
possibilities by examining whether IF-1 protein persists longer than if-1 transcript following hh
RNAi. If if-1 expression is no longer detected due to cell death, then the mRNA should be
ablated along with the protein, but if, by contrast, the cell inactivates transcription of if-1, then
the protein is expected to perdure longer than the mRNA transcripts. We performed a time
course where subsets of animals were examined after each RNAi treatment interval for presence
of if-1 and pcdh-19 transcripts as well as IF-I protein. In animals treated with hh RNAi for 12
days, the number of cells expressing if-1 and pcdh-19 transcripts was greatly reduced compared
to control, but the level of protein detection was similar, suggesting that there existed
cells (Figure 6B).
-1~/IF-1+
These data support the model that Hedgehog signaling is required for
continued expression of certain genes in neuropil glia, but not for the survival of these cells.
Similarly, the ectopic expression of if-1 and pcdh-19 observed in ptc(RNAi) animals may
be due to formation of new cells or to promoted expression of the two genes in pre-existing cells.
To determine whether perturbation of Hedgehog signaling affects gene expression in existing
cells or formation of new ifJ1/pcdh-19* cells, we examined the effects of hh and ptc RNAi in
animals where all neoblasts have been ablated by irradiation (Wolff and Dubois, 1948). After
irradiation, animals are capable of surviving for a short duration before their tissues begin to
regress. We exposed animals to 6,000 rads of ionizing radiation and immediately afterwards
began RNAi. If ectopic expression resulting from inhibition ofptc requires new cell production,
then we would not expect to see an increase in numbers of if-1]/pcdh-19+ cells outside the
neuropil in irradiated animals. Otherwise, it would suggest that differentiated cells that existed
prior to irradiation are induced to express if-] and pcdh-19.
173
Indeed, we observed greater
numbers of if-i+/pcdh-i9f cells in animals where ptc was inhibited after irradiation compared to
control (Figure 6C). This result is consistent with a role of Hedgehog signaling in controlling
gene expression rather than specification of a cell type.
Based on these and previously discussed data, we believe that Hedgehog ligand
constitutively secreted from hh+ neurons in the medial cortex activates transcription of if-1 and
pcdh-19 in neuropil glia. Loss of hedgehog ligand results in inactivation of if-1 andpcdh-19 but
does not affect the viability of the cell, judging from continued expression of gs, gat-i, eaa2,
and eaat3. Cells competent to respond to Hedgehog signaling exist outside the neuropil and will
transcribe if-1 and pcdh-19 ectopically when repression of gli-i is relieved by RNAi ofptc.
174
Discussion
Planarianglial cell biology is controlled by Hedgehog signaling
We have described two principle findings in this work: molecular evidence for a planarian glial
cell type and a specific role for Hedgehog signaling in the planarian nervous system. Previous
studies using electron microscopy have identified neuronally-associated cells with branching
morphology and sparse cytoplasm in the planarian ventral nerve cord. Given their localization,
association with neurons, and morphological similarity to other invertebrate glia, these cells were
classified as candidate planarian glia (Golubev, 1988; Morita and Best, 1966). These studies,
however, did not provide any molecular evidence, which is important not only for providing
markers for continued study but also for discerning function, specification, and evolution of the
cell type. We have described here the first molecular markers for a planarian glial cell type, as
well as a morphological characterization of these cells and a mechanism to control gene
expression within the cell.
One of the functions of planarian glia can be inferred from our data. First, the greatest
accumulation of planarian glia is in the neuropil, a region filled with axons (based on a-tubulin
immunofluorescence) and synapses (based on Synapsin immunofluorescence). Second, the cells
have long processes that are closely associated with neurons. These processes extend through
the synapse-rich regions of the neuropil, travel along orthogonal commissures of the peripheral
nervous system, and encapsulate synaptic glomeruli. Third, three neurotransmitter transporters
are expressed in planarian glia.
Orthologs of the proteins encoded by eaat2 and eaat3 have
known roles in the transport of glutamate from the extracellular environment into the cytoplasm
175
where it is metabolized by orthologs of the enzyme encoded by glutamine synthetase (Anderson
and Swanson, 2000), another gene expressed in planarian glia. Glutamate released from the presynaptic neuron can continue to activate glutamate receptors on the post-synaptic neuron,
resulting in high intracellular levels of calcium and activation of pathways that lead to cellular
damage (Manev et al., 1989). We hypothesize, given these data and by analogy to the function
of astrocytes in other animals, that planarian glia uptake the excitotoxic neurotransmitter
glutamate from areas near synapses to prevent damage to the nervous system.
Constitutive expression of hh is required for expression of if1 and pcdh-19 in planarian
glial cells during homeostasis. Upon inhibition of hh expression, cells cease transcription of if-1
and pcdh-19, with loss of detectable levels of ifI transcript within two weeks of RNAi followed
by loss of detectable levels of IF-I protein by four weeks of RNAi. Inhibition of ptc results in
transcription of if-1 and pcdh-19 in cells distributed throughout the ventral parenchyma, likely
from derepression of the Gli-1 transcription factor.
This indicates that cells competent to
respond to hedgehog ligand normally exist outside the central nervous system. Additionally, the
accumulation of if1 and pcdh-19 in cells outside the neuropil in ptc(RNAi) animals following
ablation of new cell formation indicates that Hedgehog signaling induces expression of the two
genes in existing competent cells.
Our data support the possibility of at least one other glial cell type. Both gs and eaat2,
genes abundantly expressed in fly glia (Freeman et al., 2003) and vertebrate glia (Cahoy et al.,
2008), are co-expressed in cells outside the neuropil. These cells show a branched morphology
by in situ hybridization similar to that found in cells expressing if
and pcdh-19.
Their
concentration at the rim of the head where chemosensory neuron processes accumulate suggest
176
that if these cells are indeed glia, they may be a type specifically associated with the peripheral
nervous system. Interestingly, in ptc(RNAi) animals we observed an increase in the number of
gs+/eaat2+/if-1+/pedh-19F cells throughout the ventral parenchyma of the animal without a
concomitant increase in the number of all cells expressing both eaat2 and gs. While these data
lead us to hypothesize that if-1 and pcdh-19 are induced in cells already expressing eaat2 and gs
upon inhibition of ptc, lineage tracing techniques that would allow us to definitively determine
the origin of ectopic if-J*/pcdh-19F cells outside the neuropil are currently unavailable.
The
ability of Hedgehog signaling to modulate function in already existing glia has been described in
vertebrates; Sonic hedgehog secreted from neurons has been proposed to regulate distinct
subpopulations of astrocytes in adult brains (Garcia et al., 2010).
The functions of if-1 and pcdh-19 are currently unknown. Inhibition of these genes has
resulted in no observable behavioral or morphological defect.
If if-1 and pcdh-19 confer
specialized functions to planarian glia cells, then it is possible that the phenotype cannot be
detected with our current assays.
In reactive astrogliosis, the mammalian CNS response to
injury, Sonic hedgehog is one of the inductive signals that induces expression of the gene
encoding the intermediate filament GFAP (Sirko et al., 2013). Increased levels of GFAP protein
result in an increase in cell size, which is necessary for the formation of an astrocytic scar at the
wound site (Wilhelmsson et al., 2004). While there is currently no evidence that planarian glia
are involved in wound response or that a glial scar is formed, this role of Hedgehog signaling in
reactive astrogliosis raises a number of possible functions for if-1 that can be explored.
177
Ancestral roles of Hedgehog signalingin central nervous system development
In central nervous system development, Hedgehog signaling plays a critical role in vertebrates
but a seemingly less direct role in Drosophila. Sonic hedgehog expression in the vertebrate floor
plate establishes distinct domains of transcription factor expression in the ventral neural tube.
These domains first give rise to neurons and then, at later stages of development, glia (Dessaud
et al., 2008; Yu et al., 2013). The dorsal-ventral distribution of homologous transcription factors
in the developing Drosophilacentral nervous system bears a resemblance to the distribution seen
in vertebrates (Cornell and Ohlen, 2000). Hedgehog, however, appears to play a role in the
anterior-posterior patterning of neuroblasts rather than dorsal-ventral patterning (Bhat, 1999).
Hedgehog signaling, while crucial to vertebrate neural tube patterning, has not been shown to be
involved in specifying similar progenitor domains in Drosophila or planarians. The most
parsimonious explanation is that while hedgehog was expressed in or near the midline of
developing nervous systems of the common bilaterian ancestor, only in vertebrates was the
pathway co-opted into dorsal-ventral patterning.
The floor plate, which is induced by Sonic hedgehog secreted from the notochord, serves
as a moderator of axonal midline crossing through the secretion of axon guidance cues
(Colamarino and Tessier-Lavigne, 1995). Sonic hedgehog continues its involvement in neural
patterning by acting as a chemoattractant and by mediating cellular responses to other guidance
cues (Parra and Zou, 2010). The Drosophila midline glia are considered to be an analogous
structure to the vertebrate floor plate due similar gene expression and roles in controlling midline
crossing (Evans and Bashaw, 2010). Hedgehog is required for the decision to form posterior
midline glia, whose function is still not fully understood, instead of anterior midline glia, which
178
develop into ensheathing glia in the Drosophila neuropil (Watson et al., 2011).
A shared
function of Hedgehog signaling among Deuterostomes, Ecdysozoans, and Lophotrochozoans
appears to be in the control of glia near the midline. Thus, it will be interesting to see if
planarian neuropil glia participate in axon guidance during regeneration, suggesting a possible
ancestral role of Hedgehog signaling to all Bilaterians.
Implications ofmolecular evidencefor planarianglial cells
Planarians are an ideal model for the study of regeneration due to their unrivaled regenerative
ability and the molecular tools developed for the system. The role of glia in regeneration in other
systems has been described in vertebrates, where glia proliferate in response to brain injury, and
in insects, where surface glia can reform the blood-brain barrier (Sofroniew, 2009; Treherne et
al., 1984). Interestingly, astrocytic scars appear to counteract neural regeneration by blocking
the extension of axons into the damaged region (Silver and Miller, 2004).
Glial scars are
unlikely to form in planarians, given that regeneration results in re-establishment of IF-i protein
patterns seen in unwounded animals.
Whether glia actively participate in repatterning the
nervous system after injury is an interesting topic to explore, possibly leading to studies on both
mechanisms of glia-neuron interaction and glial roles in neural network connectivity. If, on the
other hand, glia passively extend their processes into existing neural architecture, then the
mechanisms that guide glial cell development and migration may be studied instead. Several
lines of evidence support the second hypothesis. IF-i+ processes are not seen deviating from
axonal tracts, inhibition of hh affects glial processes but overall disruption to the neural network
179
has not been observed, and synapses are formed prior to IF-1 processes in the regeneration
blastema.
The work we present here opens the field to a number of opportunities for continued
research. Glia are now gaining recognition as an active player in nervous system development,
function, and regeneration, and not just as a passive cell type that glues neurons together
(Freeman and Rowitch, 2013; Perea and Araque, 2010; Robel et al., 2011). Planarians are a
tractable model organism that will be amenable to the study of glia in a highly regenerative
system and member of the understudied Lophotrochozoan superphylum. Additionally, further
characterization of planarian glia, especially in its developmental pathway, will surely provide
more insight into the long-standing question of whether invertebrate and vertebrate glia share a
common origin (Hartline, 2011).
180
181
00
U
z
z
I
*
U
"3
gapdh
pc2
synapsin
chat
1020HH
Smedwi-I
agat-1 * *mag
ca *mat * collagen
*1
mhc6
hh
ptc
smo
sufu
*m
gli-I
gli-2
gli-3
*
*
Log 2 Fold Change
-
Log 2 Fold Change
x
U
0
I
U
synapsin
*
nkx6
r*
pcdh-19
if-I
pax6b
nkx2
-
ptc
hh
pc2
*
gapdh
U
-Z)
Clb
I
pax6b
jesjop
nkx6
0A
.eje
nkx2
e.
r
14
m
z
z
0
0
Eu
Eu
11z
I
1
W-
.9
'n
Figure 1. Hedgehog signaling in the planarian cephalic ganglia
(A-C) Expression of neural transcription factors by fluorescent in situ hybridization (FISH). (A)
Expression of nkx2 in medial domain and outer edge of cephalic ganglia lobes. (B) Expression
of nlx6 in medial and lateral domains of lobes. (C) Expression of pax6b in lateral and outer
edge of lobes. (D) Comparison of transcription factor domains and neuron subtype distribution
between planarian cephalic ganglia (top) and vertebrate neural tube (bottom). Transverse view
for both. Planarian cephalic ganglia (one lobe shown) oriented dorsal up, medial left, lateral
right; neural tube oriented dorsal right.
(E) Inhibition of hh (center column) or ptc (right
column) shows no change in expression pattern of nkx2 (top row), nkx6 (middle row), or pax6b
(bottom row) from controls (left column). (F) Head amputation for control Illumina libraries.
Circle indicates portion of animal taken for RNA isolation. (G) Cephalic ganglia dissection was
performed by creating an incision (inc) to peel away the dorsal epidermis, followed by removal
of intestinal tissue (gut) and the pharynx (phx), revealing the cephalic ganglia (cg) and ventral
nerve cords (vnc). (H) Bar graph depicting log2 fold change of selected markers between head
fragment reads and cephalic ganglia reads. Statistically significant log2 fold change (pdj < 0.05)
For a list of all analyzed genes, please see Table 1. (I) Bar graph
indicated by asterisks.
depicting log2 fold change of selected markers between cephalic ganglia reads from hh(RNAi)
and ptc(RNAi) animals.
observed for hh (pj
=
Statistically significant log2 fold change, indicated by asterisks,
6.9e-5), if-] (p4
=
0.04), and pcdh-19 (paj = 1.6e-24) in hh(RNAi)
animals. For a list of all affected genes, please see Table 2. Anterior up, ventral surface shown
for A-C and E. Anterior up, dorsal surface shown for F and G.
183
Figure 2
E
ptc RNAi
hh RNA
Control RNAi
F
M if-1+
[3 if-1+/pcdh-19+
M
pcdh-19+
-
Control RNAi
4'
K
-
hh RNAi
I
-
ptc RNAi
0
I
I
I
I
U-
1000 2000 3000 4000
Cells / mm 2 inside neuropil
0
200
2
400
600
Cells mm outside neuropil
if-I /ipcdh-19
184
Figure 2.
Expression of if-1 and pcdh-19 in neuropil cells is dependent on Hedgehog
signaling
(A) Double FISH for if1 and pcdh-19 in wild type animals. DAPI-labeled nuclei (blue) enriched
in cortical region surrounding cell-body sparse neuropil.
(B) Detail of if-1 and pcdh-19
expression in the cephalic ganglia neuropil. (C) Double FISH for if1/pcdh-19 and ptc indicates
co-expression. (D) Double FISH for if1/pcdh-J9 and hh indicates lack of co-expression. (E)
Double FISH for if-1 and pcdh-19 in animals following inhibition of control gene, hh, or ptc.
(F) Quantification of results from (E), with
White dotted line delineates edge of animal.
distribution of if1I only cells (green), pcdh-19+ only cells (magenta), and if1+/pcdh-19* cells
(white). Within the neuropil, cells that express one or both markers are present at 2135.6 265.8
cells/mm2 in control RNAi conditions (n
=
5 animals), 169.3 118.6 cells/mm 2 in hh RNAi
conditions (n = 4 animals), and 3354.0+249.5 cells/mm2 inptc RNAi conditions (n = 5 animals).
Differences were significant in both hh RNAi (p = 2.7e-6, two-tailed t test) and ptc RNAi (p
=
7.1 e-5, two-tailed t test). In the head not including the neuropil region, cells that express one or
both markers are present at 64.4+16.6 cells/mm 2 in control RNAi conditions (n = 5 animals),
1.5 2.9 cells/mm2 in hh RNAi conditions (n
RNAi conditions (n
=
=
4 animals), and 465.4 68.7 cells/mm2 in ptc
5 animals). Differences were significant in both hh RNAi (p
=
1.5e-5,
two-tailed t test) and ptc RNAi (p = 1.4e-6, two-tailed t test). (G) Double FISH for if-1/pcdh-19
and pc2 indicates no co-expression. (H) Double FISH for if-1/pcdh-19 and syn indicates no coexpression.
(I) Double FISH for if-1/pcdh-19 and panel of neural markers indicates no co-
expression. For a list of all neuronal markers tested, please see Table 3. Anterior up, ventral
surface shown for all. Scale bars: I00um for all.
185
Figure 3
U
186
Figure 3. if-1]/pcdh-19+ cells express neurotransmitter reuptake and metabolism genes
(A-D) Double FISH showing overlapping expression of if-J'/pcdh-19+ neuropil cells with (A)
glutamine synthetase, (B) GABA transporter gat-], (C) glutamate transporter eaat2, and (D)
glutamate transporter eaat3. Detail of one representative cell included with each cephalic ganglia
overview image, split into single channels. (E-I) Schematic indicates portion of animal shown.
Images show one hemisphere of the cephalic ganglia and the lateral parenchymal space. White
dotted line delineates edge of animal. Arrowheads denote double positive cells. (E) Double
FISH for ptc and gs showing significant overlap within the neuropil and rare overlap outside the
neuropil (F) Double FISH for ptc and gs showing no overlap in expression. (G-I) Double FISH
indicating that gs is co-expressed with (G) eaat2, (H) eaat3, and (I) gat-1 in the majority of cells
in the neuropil and in few cells outside the CNS. Anterior up, ventral surface shown for all.
Scale bars: I00um for overviews; lOum for insets for A-D; 50um for E-I.
187
Figure 4
ptc RNAi
Control RNAi
80
-
J
60
-
K0
-
c08e 40
0>>
20
-
(U
(0
-
0
4
**
UN
WC
L
Day 0
Day 2
Day 6
Day 4
C
rC
188
Day 8
Figure 4. if-1]Ipcdh-19+ cells have long processes that closely associate with neurons
(A) Whole-mount immunofluorescence for IF-1 protein in wild type untreated animals. Top
dotted box refers to panel (B). Middle dotted box refers to panel (D). Bottom dotted box refers
to panel (E). (B) Collapsed stack of IF-I immunofluorescence in the cephalic ganglia. Dotted
box refers to panel (C). (C-E) Detail of IF-i immunofluorescence in (C) the cephalic ganglia
neuropil, (D) the lateral ventral parenchyma of the trunk, and (E) the ventral nerve cord. (F-G)
Immunofluorescence of IF-I (magenta) and a-tubulin (green) in (F) the head and (G) the lateral
ventral parenchyma of wild type untreated animals.
(H-1) Immunofluorescence of IF-I
(magenta) and Synapsin (green) in (H) the head and (I) the ventral nerve cord of wild type
untreated animals. (J) Immunofluorescence of IF-i (magenta) and Synapsin (green) following
inhibition of hh, ptc, or a control gene. (K) Quantification of hh and ptc RNAi phenotype based
on percentage of orthogonal axon bundles in contact with IF-I+ processes. In control RNAi
animals, 15.1 5.1% of orthogonal axon bundles contained IF-lI processes (n = 5 animals). In hh
RNAi animals, 2.1 2.8% of orthogonal axon bundles contained IF-1+ processes (n = 5 animals).
In ptc RNAi animals, 61.4 7.8% of orthogonal axon bundles contained IF-lI processes (n = 4
animals). Difference between control and hh(RNAi) (p = 1.Oe-3, two-tailed t test) and control
and ptc(RNAi)
(p =
1.3e-5, two-tailed
t test) were
statistically
significant.
(L)
Immunofluorescence for IF-i (magenta) and Synapsin (green) in anterior regeneration time
course. White dotted line delineates edge of animal. Yellow dotted line delineates approximate
amputation plane. Anterior right, ventral surface shown. Scale bars: 1 00um for A, B, F, H, and
J; 5Oum for L; lOum for C, D, E, G, and I.
189
0
Q"0
0
0+0
0/
00
0
0I
0
01
0)
U 0U
1111
(11
CA
N)
0
0
0
Cells /mm
+0
0
0
0
-.L
elsim
0
0U
10
2
2
0
L"1
0
0
wi
0
0
0
0
--
'1
-I
(~u
W
z
z
;0
0
0) C"
0
ptc Head
hh Head
ptc CG
hh CG
Log 2 Fold Change
Log 2 Fold Change
Log 2 Fold Change
Log 2 Fold Change
-L
gs
gat-I
if-I Ipcdh-19 gs
eaat2
if-I /pcdh-19 gat-I
eaat3
if-1I/pcdh-19 eaat3 if-1I/pcdh-19 eaat2
Figure 5. hh inhibition does not ablate glial cells
(A) Double FISH for if-1/pcdh-19 with gs (first row), gat-] (second row), eaat2 (third row), and
eaat3 (fourth row) following inhibition of a control gene (first column), hh (second column), or
ptc (third column). (B) Comparison of gene expression levels from cephalic ganglia (CG) or
head fragment (Head) tissue samples following inhibition of hh or ptc. Expression level changes
for each gene are not statistically significant. (C-D) Stacked bar graph of depicting number of
cells expressing gs (C) or gat-] (D) per square millimeter following inhibition of a control gene,
hh, or ptc. Bar sections denote ratio of if-1]/pcdh-19+ subpopulation (white) to if-1-/pcdh-19~
subpopulation (green). Differences between RNAi conditions not statistically significant (twotailed t test). Anterior up, ventral surface shown for all. Scale bars: I00um for A.
191
t0J
if-I pcdh-19
Lethal Irradiation
Reduced RNAi
if-1 pcdh-19 IF-1
W
Anterior Blastema (d6)
if-1 pcdh-19 eaat2 eaat3
if-1 pcdh-19 gat-1
Anterior Blastema (d6)
z
z
z
0
:3
l>
;>
0)
C
-n
Figure 6. Hedgehog signaling is required for if-1 and pcdh-19 expression in planarian glia
(A) Double FISH Qf-1/pcdh-19; gat-]) or triple FISH (f-1/pcdh-19; eaat2; eaat3) in d6 anterior
blastemas of trunks following inhibition of control gene, hh, or ptc. White dotted line delineates
edge of animal. Yellow dotted line delineates approximate amputation plane. (B) FISH of ifJ/pcdh-19 and immunofluorescence of IF-1 in animals following reduced RNAi treatment (fed
dO, d4, d8, fixed d12) of control gene, hh, orptc. (C) FISH of if-/pcdh-19 in animals following
lethal irradiation and subsequent RNAi treatment (irradiated dO, fed dO, d4, d8, fixed dl1).
Anterior up, ventral surface shown for all. Scale bars: I00um for all.
193
Supplemental Figure S1
7131
3461
9961
6605
433
12264
648
69
821
1206
16532
122
13510
284
2627
331
2649
57
1530
348
1242
I
750
1636
2169
I
628
9223
13594
447
194
Supplemental Figure S1.
Whole mount in situ hybridization (WISH) screen for genes with significant differential
expression levels following inhibition of hh or ptc. Labels correspond to contig ID numbers
listed in Supplemental Table S2.
Anterior up, dorsal surface shown on left, ventral surface
shown on right. Scale bars: 500um for all.
195
I 543
543
294
126
--'=
234
go
35
878~C
9998
|9
sapiens
H. sapienslammn AJC
1411'
.sain
Oegns mu"-
963
C. slogans dfa-3
i2
82
H. robustaIamin2
H. robustalamiri1
famin
ertn1
Ii"fi chan
7340
poei 861
C. intestinells protein 2
943
9396
23084
C. elegans mn-i
L.iate ro>1 042
C. angns4gan ane protein
H.po
1
8atspoen24
protn384
cyloplnmmucin
00L.Me
.al
Laplcantarmproteltila23084
itate filament 8
10. L.
D. molanopaster lamin C
C.
voctonsis protein 170348
sapiens lammn 81
motanogastor tannin
N.
H moic
CCslogans ibb-1
Ob-an
C.ean
iai
dilfmentt1
ntMd
938 H. r.busto ChaustC etlme nntrmd
0lgn
1000
robusta1000
gigantoe protein 284411
H,
543 76
H. robust&a o
otuuayt cytoplmamiedtlem
C.
I. pfganlsm prate n 179001
D.
u383
H. modicirialsmacrfli
L..
34900
1O0O0
921~~H
H, tobusaactoplasmc mntermediate Ifiaent 5
H8 m.mdcnu
H
92
medctinalis gierin
H. robWW cytplamtc intermediate faMwnW 3
H. robusta cytopasmtc Intrmediate iamont 9
68
H.
tnstinanllsprotein 297340
C.ninsstina s protein 29
448C C.
C.
H. sapiensdesmin
-p--s 9meti
173 H.saien vme~inH. sapiens nwuomt
Cn2281 00
F414.C
4
4.9tsinlspoei
r91812ane
H.
s. logans #ta-1
H
10383
73--.
295 $a
~~484
601 C. t";et
372
12H.
160
518
504
~ ~
9 390
57
351
584
1F --
M92H.
so
to
925
2E85
295
1000
H. Saplons keratin
sapiens keratin 5
intestinallsprotein 206773
H.
1000
sOwd-keain1
-
H. sapiens keratin 9
C.
C. slegans
ifc-2
ifc-1
tlet protein 220856
oftgens
C.
C elaainslfd-2
elgans ifd-1
C.
C netnlspoen264
oteans tpIII
0
C)
In
0
V
0
0
0
070
Supplemental Figure S2. Phylogenetic tree for cytoplasmic intermediate filaments
Maximum likelihood tree with 1,000 bootstrap replicates of cytoplasmic intermediate filament
genes from all three superphyla of the bilateria, including S. mediterranea if-1.
intermediate filament genes were used as an outgroup to root the tree.
197
Nuclear
Supplemental Figure S3
B
A
Head rim
... . . . .
F eripheral
par enchyma
if-I DCdh-19
if-1 Dcdh-19
ii
Between
cephalic
ganglia
iv
Nerve cord
Between
ne rve cords
C
Control RNAi
Control RNAi
ptc RNAi
I'
198
ptc RNAi
Supplemental Figure S3. if-1]/pcdh-19+ cells are found in multiple regions
(A) Schematic of the body and central nervous system of the planarian.
Red dotted boxes
indicate regions of interest focused on in (B) and (C). (B) Double FISH for if-1 and pcdh-19 in
wild type untreated animls. Images show detail of single cells near the periphery of the head
(i), in the parenchyma at the ventral midline between the lobes of the cephalic ganglia (ii), within
the ventral lateral parenchyma of the trunk (iii), and embedded in the ventral nerve cord (iv). (C)
Double FISH for if-1 and pcdh-19 in animals following inhibition of control gene orptc. Images
show detail of the head rim region (i) and the tail region between the ventral nerve cords (v).
Anterior up, ventral surface shown for all. Scale bars: 10um for A; 50um for B.
199
Supplemental Figure S4
A
gli-2 RNAi
gli-I RNAi
Control RNAi
gli-3 RNAi
<L
b1
20
D
0
200)
Supplemental Figure S4. Further characterization of if-1+/pcdh-19r cells
(A) FISH for if-1/pcdh-J9 in animals following inhibition of control gene, gli-1, gli-2, or gli-3.
DAPI, nuclei (blue). Anterior up, ventral surface shown for all. (B) FISH for if-1/pcdh-19 in d6
anterior blastemas and trunk region ventral to the pharynges of trunk fragments following
inhibition of control gene, hh, or ptc. Images of anterior blastemas show accumulation of if1*/pcdh-J9r cells during regeneration. Images of pharyngeal region show presentation of hh or
ptc phenotype. Scale bars: I00um for all.
201
Supplemental Figure S5
A
if-I Dcdh-19 Dc2 DAPI
B
if-I Dcdh-19 svn DAP
C
if-1 pcdh-19
202
Supplemental Figure S5.
if-1 and pcdh-19 expression does not overlap with neuronal
marker expression
(A-B) Double FISH for if-1/pcdh-19 and (A) pc2 or (B) syn. Detail of four cell clusters shown
for each. Images show merged image in top left quadrant and single channels for remaining
quadrants. DAPI, nuclei (blue). (C) Double FISH for if-J/pcdh-19 and other described neuronal
markers in wild type untreated animals. Anterior up, ventral surface shown for all. Scale bars:
I0um for A, B; I00um for C.
203
Supplemental Figure S6
gs
C
t-1
E
. tat 2
G
eamt3
0
c\<
z
J
K
2
z
OZ
NC
CQ~
CO Z
qCOl
204
Supplemental Figure S6. Further characterization of markers for if-1]Ipcdh-19+ cells
(A-H) WISH for (A) gs, (C) gat-], (E) eaat2, and (G) eaat3 with FISH for single cell detail for
(B) gs, (D) gat-i, (F) eaat2, and (H) eaat3. DAPI, nuclei (blue). (I-K) Double FISH for ptc and
(I) gat-] (J) eaat2, and (K) eaa3 in wild type untreated animals. (L-N) Double FISH for pc2
and (L) gat-], (M) eaat2, and (N) eaat3 in wild type untreated animals.
(0) RNA probe
specificity controls for gs, gat-], eaat2, and eaat3. 4/4 gs(RNAi) animals display reduced gs
expression in the CNS compared to unc-22(RNAi) controls animals. 4/5 gat-J(RNAi) animals
display slightly reduced gat-1 expression in the CNS compared to unc-22(RNAi) controls
animals. 9/9 eaat2(RNAi) animals display reduced eaat2 expression in the CNS compared to
unc-22(RNAi) controls animals. 6/6 eaat3(RNAi) animals display reduced eaat3 expression in
the CNS compared to unc-22(RNAi) controls animals. Anterior up, ventral surface shown for A,
C, E, G. Anterior up, ventral surface of right cephalic ganglia lobe and lateral head parenchyma
shown for I-N. Anterior left, ventral surface shown for 0. Scale bars: 500um for A, C, E, G, 0;
10um for B, D, F, H; 50um for I-N.
205
Supplemental Figure S7
A Control RNAi
if-I RNAi
B
Pcdh-19 RNAi
0
C
C
Li-
D
C
Control RNAi
ptc RNAi
IE
0
(U
C
Co
U-
(U
Co
U-
206
Control RNAi
ptc RNAi
Supplemental Figure S7. IF-i protein accumulates in specific tissues
(A) Immunofluorescence of IF-I and Synapsin in animals following inhibition of control gene,
if-1, and pcdh-19. (B) Immunofluorescence of IF-I in dorsal (top) domain and the ventral
(bottom) domain of the ventral nerve cord. Arrowheads depict lacunae in the IF-1 process
bundle. (C) FISH of if-1/pcdh-19 and immunofluorescence of IF-i in the cephalic ganglia of
wild type untreated animals. (D) Detail of immunofluorescence of IF-I and Synapsin in lateral
ventral parenchyma of the trunk following inhibition of control gene or ptc.
(E) Detail of
immunofluorescence of IF-I and Synapsin in the head rim of animals following inhibition of
control gene or ptc. Dotted box in top row refers to corresponding image in bottom row.
Anterior up, ventral surface shown for A, C, D, E. Anterior right, ventral surface shown for B.
Scale bars: I00um for A, C, top row of E; IOum for B, D, bottom row of E.
207
Supplemental Figure S8
unc-22 RNAi
olig2 RNAi
jog
nkx2 RNAi
gcml-1 RNAi
S
S
gcml-2 RNAi
b
9,
ee
208
Supplemental Figure S8. Inhibition of if-1 or pcdh-19 does not produce any observable
phenotype
WISH of if-1/pcdh-19 in whole animals or d6 regenerating head, trunk, and tail fragments
following inhibition of control gene or putative glial transcription factor.
For each RNAi
condition, left image shows intact animal, top right image shows head fragment, middle right
image shows trunk fragment, bottom right image shows tail fragment.
surface shown.
209
Anterior up, ventral
Table S1. Enrichment of neuronal markers and depletion of non-neuronal markers
in cephalic ganglia tissue libraries
CNS control (RHAI) vs.
Head control (RNAI)
Conto ID
ddSmed_v4_78_0_1
ddSmed_v4_41965_0_1
dd._Smed_v4_7063_0_2
ddSmed_.v4_9474.0.1
ddSmed_v4_10354_0_1
dd_Smed-V4j7470_0_1
dd Smed v4 46372_0_1
ddSmed_v4_17566_.0.1
ddSmed_v4_3135_0_1
ddSmedyv4_6208_0_1
ddSmnedv4_1566_0_1
ddSmed-v4_5133_0_2
dd_Smedv4_9795_0_1
dd_Smed-V4_14852_0_1
dd_Smedv4_8392_0_2
dd_Smedv4_16581_0_1
dd_Smedv4_2688_0_1O
ddSmed_v4_28398_0_1
ddSmedv4_7326_0_1
ddSmedv4_14865_0_1
ddSmedyv4_659_0_1
ddSmed-v4_756_0_1
dd_.Smedv4_899_0_1
ddSmedv4_4575_0_1
ddSmedv4_9774_0_1
ddSmedv4_8234_0_1
dd_Srned_v4_4841_0_1
ddSmed v4_34399_0_1
ddSmedv4_19040_0_1
dd_Smedyv4_17385_0_1
ddSmedv4_48430_0_1
ddSmedv4769_0_1
dd_Smed_v4_702_0_1
dd_Smed_v4_3131_0_1
ddSmedv4_7223_0_1
dd_.Smedv4_907_0_1
ddSmed v4 249 ,,0 1
, ,
Annotation
gapdh
hedgehog
patched
smoothened
sufu
gil-1
gli-2
gli-3
synapsin
choline acetyltransferase
prohonnone convertase 2
1020HH
netrin I
nefrin 2
tryptophan hydroxylase
tyrosine hydroxylase
beta-catenin
wnbP-1
wntP-2
wntP-3
smedwi-I
smedwi-2
agat-1
cubulin
sine oculis
pou2/3
carbonic anhydrase
tyrosinase,
distal-less
sp6-9
ovo
marginaladhesive gland
collagen
mboat2
mboat7
met
mhc6
Expression
References
Global
CNS
Rink 2009
CNS, PHX, PCYM
CNS, PHX, PCYM
PHX, PCYM
GUT, MRG
Rink 2009
Rink 2009
Rink 2009
Rink 2009
Rink 2009
Rink 2009
CNS, PNS
CNS, PNS
CNS, PNS
CNS
CNS
NP
CNS, PR
CNS, PHX
Global
Tail
Tail
PHX
NB
NB
PCYM
NPH
NPH
NPH
NPH
PR
PR, PCYM
PR
PR
RIM
MUS
GUT
GUT
GUT
PHX
210
Nishimura 2010
Collins 2010
Collins 2010
Cebuda 2005
Cebria 2005
Curie 2013
Fraguas 2011
Petersen 2008
Petersen 2008
Petersen 2008
Petersen 2008
Reddien 2005
Reddien 2005
Eisenhoffer 2008
Scmone 2011
Scimone 2011
Scimone 2011
Scimone 2011
Lapan 2011
Lapan 2011
Lapan 2011
Lapan 2012
Zayas 2010
Witchley 2013
Wenemoser 2010
Lop Fold
Change
pul
-0.11
5.4E-01
5.85
-0.42
0.87
0.41
-2.10
nla
5.7E-57
3.9E-02
3.8E-07
1.OE-01
3.3E-29
2.3E-01
0.40
1.41
2.82
2.65
6.35
1.50
2.51
0.33
2.92
-1.67
1.16
0.36
-3.06
-1.58
-0.75
-2.09
-3.49
0.03
-1.71
-2.48
-2.43
1.77
1.36
-2.77
-4.48
-4.40
0.53
1.24
-2.05
-2.73
3.8E-01
1.7E-25
6.1E-95
5.1E-88
3.OE-304
3.6E-16
1.2E-21
2.7E-01
3.OE-49
9.8E-33
3.2E-01
3.5E-01
1.3E-02
2.OE-31
1.5E-05
2.5E-51
4.8E-104
1.OE+00
7.5E-12
2.1E-34
2.OE-07
1.6E-10
1.4E-04
4.2E-01
4.3E-196
7.4E-197
8.7E-04
3.4E-15
3.1E-44
3.2E-93
Supplemental Table S1. Enrichment of neuronal markers and depletion of non-neuronal
markers in cephalic ganglia tissue libraries
For each gene, general expression pattern and log2 fold enrichment of CNS tissue expression
over head fragment expression is listed. CNS, central nervous system; GUT, intestinal tract;
MUS, muscle layer; NB, neoblasts; NP, neuropil; NPH, nephridia; PCYM, parenchyma; PHX,
pharynx; PR, photoreceptors; RIM, body peripheral edge. References for previously published
genes are listed (Cebria and Newmark, 2005; Collins et al., 2010; Currie and Pearson, 2013;
Eisenhoffer et al., 2008; Fraguas et al., 2011; Lapan and Reddien, 2011; 2012; Nishimura et al.,
2010; Petersen and Reddien, 2009; Reddien et al., 2005b; Rink et al., 2009; Scimone et al., 2011;
Witchley et al., 2013; Zayas et al., 2010).
211
........
........
k)
Contig ID
ddSmed v4_7131 0 1
ddSmedv4_3451_0_1
ddSmed v4_99610_1
ddSmedv4_6605_0_1
ddSmed_v4_433_0_1
ddSmed_v4_12254_0_1
ddSmedv4_648_0_1
ddSmedv4_69_0_2
dd Smed v4_821 0 1
ddSmedv4_332_0_1
ddSmedv4_1206_0_1
ddSmedv4_16532_0_1
ddSmedv4 122 1 1
dd_Smed_v4_13510_0 1
ddSmedv4 284 0 1
ddSmedv4_493_0_1
ddSmedv4_2627 0 1
dd_Smed_v4_331_0_1
dd Smed v4 2649 0 1
dd_Smed_v4_57_0_1
ddSmedv4_750_0_1
ddSmedv4_16360_1
ddSmedv4_2169_0_1
ddSmed v4_1530 0 1
ddSmed_v4_348_0_1
ddSmedv4_1242 0 1
ddSmed v4_628_0_1
ddSmedv4_9223 0 1
ddSmedv4_13594_0_ 1
dd Smed v4 447 0 1-
Annotation
GLIPRi-like protein 1-like (Schmidtea mediterranea)
FAM115A (Homo sapiens)
protocadherin-19 (Mus musculus)
No Similarity
GLIPR1-like protein 1-like (Schmidtea mediterranea)
Protein IFA-1 (Caenorhabdiis elegans)
tubulin alpha 1A chain (Homo sapiens)
No Similarity
protein disulfide-isomerase AS (Mus musculus)
prog-1 (Schmidtea mediterranea)
hematopoetic prostaglandin D synthase (Mus musculus)
No Similarity
lysosomal acid lipase/cholesteryl ester hydrolase precursor
No Similarity
Saposin-related (Drosophila melanogaster)
reticulocalbin-1 (Schmidtea mediterranea)
E3 ubiquitin-protein ligase MYLIP (Homo sapiens)
No Similarity
fibrillin-2 (Homo sapien)
WAP four-disulfide core domain protein 6B (Mus musculus)
No Similarity
zonadhesin (Homo sapiens)
No Similarity
GLIPR1-like protein 1-like (Schmidtea mediterranea)
peptidyl-prolyl cis-trans isomerase C (Mus musculus)
peptidyl-prolyl cis-trans isomerase B (Mus musculus)
SCO-spondin-like (Mus musculus)
No Similarity
No Similarity
No Similaritv
2.00E-44
5.00E-46
7.00E-46
3.00E-25
4.00E-33
8.00E-142
7.00E-12
2.00E-64
0.00E+00
3.00E-05
9.00E-104
-1.09
0.51
-0.38
0.21
-0.65
0.33
-0.82
0.05
-1.15
-0.43
-1.84
-0.50
-1.12
-0.50
-0.35
-0.19
-3.24
-3.56
0.26
-1.15
1.OOE-24
2.00E-32
0.OOE+00
2.00E-44
1.00E-82
0.00E+00
-3.00
Change
Log2 Fold
-2.90
-2.80
-2.40
-2.16
-2.10
-1.19
-1.19
-1.18
-1.16
E Value
2.00E-44
6.00E-38
1.OOE-03
padj
4.58E-21
4.87E-24
1.61 E-24
8.60E-12
1.29E-1 3
0.044
8.02E-03
8.02E-03
0.043
0.023
0.044
0.015
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
0.851
1.000
1.000
CNS hh (RNA) vs.
CNS control (RNAI)
Change
1.47
1.15
0.26
1.63
1.40
0.18
1.19
-0.93
1.28
-0.83
-0.13
-2.53
-4.26
-0.91
-0.73
0.90
1.00
1.26
1.29
1.38
1.49
1.55
1.65
1.62
2.00
2.29
2.65
2.75
2.85
2.90
Log2 Fold
pad
3.13E-07
0.105
1.000
2.69E-10
2.13E-05
1.000
0.012
0.175
0.850
1.000
1.000
6.96E-14
4.85E-06
8.61 E-05
0.010
0.033
2.97E-05
8.04E-04
2.73E-09
3.88E-04
3.81 E-09
0.005
6.10E-15
1.50E-03
3.99E-06
3.81E-08
4.65E-40
1.IOE-04
2.42E-06
1.71E-07
CNS ptc (RNAI) vs.
CNS control (RNAI)
Change
0.25
-0.35
3.75
2.25
1.23
4.68
-1.61
-1.15
-0.51
0.46
2.86
7.22
-0.08
6.99
0.72
-1.89
-0.31
-0.45
-1.94
-0.20
0.83
-0.55
0.78
0.84
-0.97
-0.85
-3.34
2.21
2.76
-1.01
Pad
0.737
0.528
9.34E-06
0.034
0.132
7.84E-09
4.86E-12
1.12E-07
0.250
0.242
7.34E-1 2
1.01 E-09
0.941
3.82E-33
0.015
1.24E-40
0.147
0.321
4.75E-46
0.733
0.145
0.434
0.180
0.253
0.016
0.078
6.77E-95
0.143
0.094
0.119
CNS control (RNAI) vs.
Head control (RNAI)
Log2 Fold
Table S2. Genes with significant differential expression levels following inhibition of hh or ptc.
Supplemental Table S2.
Genes with significant differential expression levels following
inhibition of hh or plc
Criteria for selecting genes was (1) adjusted p value (padj) of less than 0.05, (2) greater than
1000 RPKM, and (3) greater than 2 fold change in expression level either between control and
hh(RNAi) or between control and ptc(RNAi). Annotations by best BLAST hit listed for each
gene; No Similarity listed if no significant BLAST hit was found. Two genes, prog-J and
reticulocalbin-1, were described in planaria previously (Eisenhoffer et al., 2008; Zayas et al.,
2010). Blue text indicates greater than 2 fold change in expression level. Green text indicates
enrichment in CNS tissue versus whole head fragment.
213
v7iz
I&I- &I
&I
&I
&I I&& I&f
~
1I I II If X0
-&~-
II & l I& I &
If If01
00
I
I
II I
[II00
I
f41
40j-h8hn
~
4
eel
~A
0
I"
to..
lbi*I
.4
NJ
-n -n
8 tg
g
-W
-S
"vM
tog
-A
"M
w
;
-&
.,0 -4
N
g
"-qw88
"
h.
!
0boo
5*
VP9PP
I.
M!
I
PP~b~bb~bPPb.bbbPbbb
-h
74
1"
I
w i4ioIs
C13
Mg;!
N
Zbe
i
06
Supplemental Table S3. Neuronal markers used in co-expression studies
For each gene, log2 fold change between control and hh(RNAi) and between control and
ptc(RNAi) cephalic ganglia samples are listed. References for previously published genes are
listed (Cebrih and Newmark, 2005; Collins et al., 2010; Cowles et al., 2013; Currie and Pearson,
2013; Fraguas et al., 2011; MArz et al., 2013; Nishimura et al., 2010).
215
Materials and Methods
PlanarianCulture
Animals were maintained in Ix Montjuic planarian water at 200 C. S2F1L3F2 sexual animals
were used in dissection experiments and CIW4 asexual animals were used in all other
experiments.
RNA Interference
300ml of bacterial culture expressing dsRNA was pelleted and mixed with lml of 70% liver in
planarian water as previously described (Reddien et al., 2005a). Asexual animals were fed 6
times at four-day intervals unless otherwise noted. Sexual animals were fed 12 times at four-day
intervals. A gene not present in the planarian genome, unc-22, was used as a control in each
RNAi experiment.
Dissection
After four days of starvation, the animals were immersed in a 0.33N HCl solution for 30
seconds, washed once in PBS, washed once in PBS + 1% BSA, and immobilized dorsal-side up
on a silicon elastomer pad with insect pins. One longitudinal incision and one lateral incision
were made through the dorsal epidermis near the base of the pharynx. The epidermis was peeled
away to expose the pharynx and a layer of gut tissue overlying the central nervous system.
Collected tissue was placed immediately in Trizol Reagent (Invitrogen) and stored at -80C until
all samples were processed.
216
mRNA-Seq Analysis
Illumina libraries were generated with 1.Oug total RNA from head fragments and 0.2ug total
RNA from dissected CNS samples using Illumina TruSeq kits.
Libraries were prepared in
duplicate and sequenced with Illumina HiSeq. Reads were mapped to the Dresden Assembly
with Bowtie and read counts were analyzed with DESeq R package (10% false discovery rate).
MolecularBiology
cDNA libraries from planarian multi-stage total RNA were synthesized using SuperScriptIll
(Invitrogen). DNA fragments were amplified from cDNA with primers designed from Dresden
Assembly sequences and cloned into pGEM (Promega).
For RNAi constructs, inserts were
amplified from pGEM constructs and introduced using BP clonase (Invitrogen) into a Gateway
vector containing flanking LacZ inducible promoters. Full-length sequences for if-1 and pcdh-19
were obtained with 5' and 3' RACE (Ambion).
Phylogenetics
Amino acid sequences for full-length if-1 were aligned to previously published intermediate
filament protein sequences using MUSCLE (EMBL-EBI) with default parameters (Kuo and
Weisblat, 2011).
Poorly aligned segments were eliminated using GBlocks (Castresana Lab).
Phylogenetic trees were constructed using maximum likelihood with 1000 bootstraps and
adjusted with FigTree.
217
RNA in situ Hybridizationand Immunofluorescence
RNA probes were synthesized with digoxygenin, fluorescein, or dinitrolphenyl nucleotides
(Roche). For whole-mount in situ hybridization (WISH) and fluorescent in situ hybridization
(FISH), animals were fixed in 4% formaldehyde according to published protocols (Pearson et al.,
2009).
FISH protocols were followed as previously described using RNA probe dilutions at
1:1000, anti-digoxygenin peroxidase at 1:500, anti-fluorescein peroxidase at 1:300, and antidinitrophenyl at 1:100 (Pearson et al., 2009). Probes for if- 1 and pcdh-19 typically combined
into single channel to improve coverage and signal intensity. Rabbit polyclonal antibodies for
IF-1 protein were raised against peptides with amino acid sequence "TENNQIENSKEKTVC"
(GenScript). For immunofluorescence, animals were fixed in Carnoy's fixative and stained as
previously described (Newmark and Sanchez Alvarado, 2000; Wenemoser and Reddien, 2010).
Anti-IF-1
antibody
was
used
at
0.4ug/ml,
anti-Synapsin
antibody
(Anti-SYNORF,
Developmental Studies Hybridoma Bank) at 1:1000, and anti-a-tubulin (DM1 A, NeoMarkers) at
1:1000, and were developed with tyramide signal amplification (Invitrogen). To detect nuclei,
animals were stained in DAPI overnight prior to mounting in VectaShield (Vector Labs).
Samples were imaged by confocal microscopy (Zeiss LSM 700) and processed with Fiji/ImageJ.
Cell counts for neuropil regions were normalized to cross-sectional area of the cephalic ganglia
lobes.
Cell counts for heads excluding the neuropil region were normalized to the cross-
sectional area of the head.
218
RNA Probe Specificity
RNA probe specificity for a target gene was determined by performing whole-mount in situ
hybridization on animals following inhibition of the gene. Animals were fed one to four times
with bacteria expressing dsRNA for a control gene or the target gene. After the last feeding, the
animals were given five days to clear the intestine of lingering RNAi food.
Irradiation
Animals were exposed to 6,000 rads of ionizing radiation (GammaCell) to ablate all dividing
cells. Treated animals were subsequently fed dsRNA-expressing bacteria three times at dO, d4,
and d8. Animals displayed signs of anterior regression at dl I and were fixed immediately.
219
References
Adell, T., Sal6, E., Boutros, M., and Bartscherer, K. (2009). Smed-Evi/Wntless is required for
beta-catenin-dependent and -independent processes during planarian regeneration. Development
(Cambridge, England) 136, 905-910.
Anderson, C.M., and Swanson, R.A. (2000). Astrocyte glutamate transport: review of properties,
regulation, and physiological functions. Glia 32, 1-14.
Bhat, K.M. (1999). Segment polarity genes in neuroblast formation and identity specification
during Drosophilaneurogenesis. Bioessays 21, 472-485.
Boeckh, J., and Tolbert, L.P. (1993). Synaptic organization and development of the antennal lobe
in insects. Microsc. Res. Tech. 24, 260-280.
Cahoy, J.D., Emery, B., Kaushal, A., Foo, L.C., Zamanian, J.L., Christopherson, K.S., Xing, Y.,
Lubischer, J.L., Krieg, P.A., Krupenko, S.A., et al. (2008). A Transcriptome Database for
Astrocytes, Neurons, and Oligodendrocytes: A New Resource for Understanding Brain
Development and Function. Journal of Neuroscience 28,264-278.
Cebrii, F., and Newmark, P.A. (2005). Planarian homologs of netrin and netrin receptor are
required for proper regeneration of the central nervous system and the maintenance of nervous
system architecture. Development (Cambridge, England) 132, 3691-3703.
Colamarino, S.A., and Tessier-Lavigne, M. (1995). The role of the floor plate in axon guidance.
Annu. Rev. Neurosci. 18, 497-529.
Collins, J.J., III, Hou, X., Romanova, E.V., Lambrus, B.G., Miller, C.M., Saberi, A., Sweedler,
J.V., and Newmark, P.A. (2010). Genome-Wide Analyses Reveal a Role for Peptide Hormones
in Planarian Germline Development. PLoS Biol 8, e1000509.
Cornell, R.A., and Ohlen, T.V. (2000). Vnd/nkx, ind/gsh, and msh/msx: conserved regulators of
dorsoventral neural patterning? Current Opinion in Neurobiology 10, 63-71.
Cowles, M.W., Brown, D.D.R., Nisperos, S.V., Stanley, B.N., Pearson, B.J., and Zayas, R.M.
(2013). Genome-wide analysis of the bHLH gene family in planarians identifies factors required
for adult neurogenesis and neuronal regeneration. Development (Cambridge, England) 140,
4691-4702.
Currie, K.W., and Pearson, B.J. (2013). Transcription factors lhxl/5-1 and pitx are required for
the maintenance and regeneration of serotonergic neurons in planarians. Development
(Cambridge, England) 140, 3577-3588.
Deitmer, J.W., Rose, C.R., Munsch, T., Schmidt, J., Nett, W., Schneider, H.P., and Lohr, C.
(1999). Leech giant glial cell: functional role in a simple nervous system. Glia 28, 175-182.
220
Dessaud, E., McMahon, A.P., and Briscoe, J. (2008). Pattern formation in the vertebrate neural
tube: a sonic hedgehog morphogen-regulated transcriptional network. Development (Cambridge,
England) 135, 2489-2503.
Eisenhoffer, G.T., Kang, H., and SAnchez Alvarado, A. (2008). Molecular analysis of stem cells
and their descendants during cell turnover and regeneration in the planarian Schmidtea
mediterranea.Cell Stem Cell 3, 327-339.
Evans, T.A., and Bashaw, G.J. (2010). Axon guidance at the midline: of mice and flies. Current
Opinion in Neurobiology 20, 79-85.
Featherstone, D.E. (2011). Glial solute carrier transporters in Drosophilaand mice. Glia 59,
1351-1363.
Fraguas, S., Barberin, S., and CebriA, F. (2011). EGFR signaling regulates cell proliferation,
differentiation and morphogenesis during planarian regeneration and homeostasis.
Developmental Biology 354, 87-101.
Frank, M., and Kemler, R. (2002). Protocadherins. Current Opinion in Cell Biology 14, 557-562.
Freeman, M.R., Delrow, J., Kim, J., Johnson, E., and Doe, C.Q. (2003). Unwrapping Glial
Biology: Gem Target Genes Regulating Glial Development, Diversification, and Function.
Neuron.
Freeman, M.R., and Rowitch, D.H. (2013). Evolving concepts of gliogenesis: a look way back
and ahead to the next 25 years. Neuron 80, 613-623.
Garcia, A.D.R., Petrova, R., Eng, L., and Joyner, A.L. (2010). Sonic hedgehog regulates discrete
populations of astrocytes in the adult mouse forebrain. Journal of Neuroscience 30, 1359713608.
Golubev, A.I. (1988). Glia and neuroglia relationships in the central nervous system of the
Turbellaria (Electron microscopic data). Fortschr Zool 36, 31-37.
Hartenstein, V. (2011). Morphological diversity and development of glia in Drosophila.Glia 59,
1237-1252.
Hartline, D.K. (2011). The evolutionary origins of glia. Glia 59, 1215-1236.
Herrmann, H., Bar, H., Kreplak, L., Strelkov, S.V., and Aebi, U. (2007). Intermediate filaments:
from cell architecture to nanomechanics. Nat. Rev. Mol. Cell Biol. 8, 562-573.
Hosoya, T., Takizawa, K., Nitta, K., and Hotta, Y. (1995). glial cells missing: a binary switch
between neuronal and glial determination in Drosophila. Cell 82, 1025-1036.
Ingham, P.W., Nakano, Y., and Seger, C. (2011). Mechanisms and functions of Hedgehog
221
signalling across the metazoa. Nat Rev Genet 12, 393-406.
Kuo, D.-H., and Weisblat, D.A. (2011). Intermediate filament genes as differentiation markers in
the leech Helobdella. Dev. Genes Evol. 221, 225-240.
Lapan, S.W., and Reddien, P.W. (2011). dlx and sp6-9 Control optic cup regeneration in a
prototypic eye. PLoS Genet. 7, e1002226.
Lapan, S.W., and Reddien, P.W. (2012). Transcriptome analysis of the planarian eye identifies
ovo as a specific regulator of eye regeneration. Cell Rep 2, 294-307.
Leise, E.M. (1990). Modular construction of nervous systems: a basic principle of design for
invertebrates and vertebrates. Brain Res. Brain Res. Rev. 15, 1-23.
Manev, H., Favaron, M., Guidotti, A., and Costa, E. (1989). Delayed increase of Ca2+ influx
elicited by glutamate: role in neuronal death. Mol. Pharmacol. 36, 106-112.
MArz, M., Seebeck, F., and Bartscherer, K. (2013). A Pitx transcription factor controls the
establishment and maintenance of the serotonergic lineage in planarians. Development
(Cambridge, England) 140, 4499-4509.
Morgan, T.H. (1898). Experimental studies of the regeneration of Planariamaculata. Dev.
Genes Evol. 7,364-397.
Morita, M., and Best, J.B. (1965). Electron microscopic studies on Planaria. II. Fine structure of
the neurosecretory system in the planarian Dugesiadorotocephala.J. Ultrastruct. Res. 13, 396408.
Morita, M., and Best, J.B. (1966). Electron microscopic studies of planaria. III. Some
observations on the fine structure of planarian nervous tissue. J. Exp. Zool. 161, 391-411.
Nakazawa, M., CebriA, F., Mineta, K., Ikeo, K., Agata, K., and Gojobori, T. (2003). Search for
the evolutionary origin of a brain: planarian brain characterized by microarray. Mol. Biol. Evol.
20, 784-791.
Newmark, P.A., and Sanchez Alvarado, A. (2000). Bromodeoxyuridine specifically labels the
regenerative stem cells of planarians. Developmental Biology 220, 142-153.
Nishimura, K., Kitamura, Y., Taniguchi, T., and Agata, K. (2010). Analysis of motor function
modulated by cholinergic neurons in planarian Dugesiajaponica.Neuroscience 168, 18-30.
Oikonomou, G., and Shaham, S. (2011). The glia of Caenorhabditiselegans. Glia 59, 12531263.
Parra, L.M., and Zou, Y. (2010). Sonic hedgehog induces response of commissural axons to
Semaphorin repulsion during midline crossing. Nat. Neurosci. 13, 29-35.
222
Pearson, B.J., Eisenhoffer, G.T., Gurley, K.A., Rink, J.C., Miller, D.E., and SAnchez Alvarado,
A. (2009). Formaldehyde-based whole-mount in situ hybridization method for planarians. Dev.
Dyn. 238,443-450.
Perea, G., and Araque, A. (2010). GLIA modulates synaptic transmission. Brain Res Rev 63, 93102.
Petersen, C.P., and Reddien, P.W. (2009). A wound-induced Wnt expression program controls
planarian regeneration polarity. Proceedings of the National Academy of Sciences 106, 1706 117066.
Pfeiffer, S.E., Warrington, A.E., and Bansal, R. (1993). The oligodendrocyte and its many
cellular processes. Trends Cell Biol. 3, 191-197.
Qi, Y., Cai, J., Wu, Y., Wu, R., Lee, J., Fu, H., Rao, M., Sussel, L., Rubenstein, J., and Qiu, M.
(2001). Control of oligodendrocyte differentiation by the Nkx2.2 homeodomain transcription
factor. Development (Cambridge, England) 128, 2723-2733.
Reddien, P.W., and SAnchez Alvarado, A. (2004). Fundamentals of planarian regeneration.
Annu. Rev. Cell Dev. Biol. 20, 725-757.
Reddien, P.W., Bermange, A.L., Murfitt, K.J., Jennings, J.R., and SAnchez Alvarado, A. (2005a).
Identification of genes needed for regeneration, stem cell function, and tissue homeostasis by
systematic gene perturbation in planaria. Dev. Cell 8, 635-649.
Reddien, P.W., Oviedo, N.J., Jennings, J.R., Jenkin, J.C., and SAnchez Alvarado, A. (2005b).
SMEDWI-2 is a PIWI-like protein that regulates planarian stem cells. Science 310, 1327-1330.
Reinecke, M. (1976). The glial cells of the cerebral ganglia of Helix pomatia L. (Gastropoda,
Pulmonata). Cell Tissue Res 169.
Reuter, M., Mhntyli, K., and Gustafsson, M.K.S. (1998). Organization of the orthogon - main
and minor nerve cords - Springer. Hydrobiologia 383, 175-182.
Rink, J.C., Gurley, K.A., Elliott, S.A., and Sanchez Alvarado, A. (2009). Planarian Hh Signaling
Regulates Regeneration Polarity and Links Hh Pathway Evolution to Cilia. Science 326, 14061410.
Robel, S., Berninger, B., and G6tz, M. (2011). The stem cell potential of glia: lessons from
reactive gliosis. Nat. Rev. Neurosci. 12, 88-104.
Rowitch, D.H. (2004). Glial specification in the vertebrate neural tube. Nat. Rev. Neurosci. 5,
409-419.
Ruiz i Altaba, A., Nguyen, V., and Palma, V. (2003).The emergent design of the neural tube:
prepattern, SHH morphogen and GLI code. Curr. Opin. Genet. Dev. 13, 513-521.
223
Sanchez Alvarado, A., and Newmark, P.A. (1999). Double-stranded RNA specifically disrupts
gene expression during planarian regeneration. Proc. Natl. Acad. Sci. U.S.a. 96, 5049-5054.
Scimone, M.L., Srivastava, M., Bell, G.W., and Reddien, P.W. (2011). A regulatory program for
excretory system regeneration in planarians. Development (Cambridge, England) 138, 43874398.
Silver, J., and Miller, J.H. (2004). Regeneration beyond the glial scar. Nat. Rev. Neurosci. 5,
146-156.
Sirko, S., Behrendt, G., Johansson, P.A., Tripathi, P., Costa, M., Bek, S., Heinrich, C., Tiedt, S.,
Colak, D., Dichgans, M., et al. (2013). Reactive glia in the injured brain acquire stem cell
properties in response to sonic hedgehog. [corrected]. Cell Stem Cell 12, 426-439.
Sofroniew, M.V. (2009). Molecular dissection of reactive astrogliosis and glial scar formation.
Trends Neurosci. 32, 638-647.
Sofroniew, M.V., and Vinters, H.V. (2010). Astrocytes: biology and pathology. Acta
Neuropathol. 119, 7-35.
Treherne, J.E., Harrison, J.B., Treherne, J.M., and Lane, N.J. (1984). Glial repair in an insect
central nervous system: effects of surgical lesioning. J. Neurosci. 4, 2689-2697.
Varjosalo, M., and Taipale, J. (2008). Hedgehog: functions and mechanisms. Genes Dev. 22,
2454-2472.
Wagner, D.E., Wang, I.E., and Reddien, P.W. (2011). Clonogenic neoblasts are pluripotent adult
stem cells that underlie planarian regeneration. Science 332, 811-816.
Watson, J.D., Wheeler, S.R., Stagg, S.B., and Crews, S.T. (2011). Drosophilahedgehog
signaling and engrailed-runt mutual repression direct midline glia to alternative ensheathing and
non-ensheathing fates. Development (Cambridge, England) 138, 1285-1295.
Wenemoser, D., and Reddien, P.W. (2010). Planarian regeneration involves distinct stem cell
responses to wounds and tissue absence. Developmental Biology 344, 979-991.
Wilhelmsson, U., Li, L., Pekna, M., Berthold, C.-H., Blom, S., Eliasson, C., Renner, 0.,
Bushong, E., Ellisman, M., Morgan, T.E., et al. (2004). Absence of glial fibrillary acidic protein
and vimentin prevents hypertrophy of astrocytic processes and improves post-traumatic
regeneration. Journal of Neuroscience 24, 5016-5021.
Witchley, J.N., Mayer, M., Wagner, D.E., Owen, J.H., and Reddien, P.W. (2013). Muscle cells
provide instructions for planarian regeneration. Cell Rep 4, 633-641.
Wolff, E., and Dubois, F. (1948). Sur la migration des cellules de regdndration chez les planaires.
Rev Suisse Zool 55, 218-227.
224
Yazawa, S., Umesono, Y., Hayashi, T., Tarui, H., and Agata, K. (2009). Planarian
Hedgehog/Patched establishes anterior-posterior polarity by regulating Wnt signaling.
Proceedings of the National Academy of Sciences 106,22329-22334.
Yu, K., McGlynn, S., and Matise, M.P. (2013). Floor plate-derived sonic hedgehog regulates
glial and ependymal cell fates in the developing spinal cord. Development (Cambridge, England)
140, 1594-1604.
Zayas, R.M., Cebria, F., Guo, T., Feng, J., and Newmark, P.A. (2010). The use of lectins as
markers for differentiated secretory cells in planarians. Dev. Dyn. 239, 2888-2897.
Zhou, Q., and Anderson, D.J. (2002). The bHLH transcription factors OLIG2 and OLIGI couple
neuronal and glial subtype specification. Cell 109, 61-73.
225
226
Chapter 4
Discussion
227
I. Pluripotent adult stem cells
The existence of a pluripotent stem cell type in planarians had been speculated upon but not
conclusively demonstrated until the use of sub-lethal irradiation and single cell transplantation.
Unfortunately the types of experiments that can be performed using clonogenic neoblasts are
currently limited, as there is no way to determine a priori whether any particular neoblast is
pluripotent. A neoblast is clonogenic if it is capable of forming a colony, which currently cannot
be assayed in vivo. Also, there is increasing evidence that the neoblast population is highly
heterogeneous, containing clonogenic neoblasts as well as lineage restricted stem cells that
express both neoblast markers and differentiated cell markers. These properties limit certain
types of experiments that require pure populations of cells, such as transcriptomic, epigenetic,
and proteomic analysis. Even so, the development of transplantation has opened the field to a
number of experiments that can further both our understanding of stem cell biology and a
number of technologies that can improve planarians as a model system.
Many questions of the nature of clonogenic neoblasts remain unanswered. Are neoblasts
able to sense their environment and decide to divide symmetrically or asymmetrically?
We
estimate based on our data that the cells undergo symmetric division on average once every two
days. However, the number of cells per colony can vary widely at specific time points after
single cell transplant, with some colonies examined consisting of only one cell. This may result
from varying rates of cell division, where colonies with fewer cells stem from a clonogenic
neoblast that was cycling more slowly.
Another possibility that is not necessarily mutually
exclusive is that the cells of the early colony underwent one or more rounds of asymmetric
228
division, producing a clonogenic neoblast and a differentiated cell. One line of evidence that
argues against the ability of a clonogenic neoblast to sense population depletion is from
experiments done with shielded irradiation. When a portion of the animal is protected against
ionizing radiation, the neoblasts within that region survive while the remainder of the animal is
left devoid of proliferative cells. In this situation, few smedwi-J+ cells are observed migrating
into the areas lacking neoblasts unless the animal is wounded (Guedelhoefer and SAnchez
Alvarado, 2012). Therefore, depletion of the neoblast population does not appear to stimulate a
response in these cells.
It would also be interesting to see what types of cells are produced by the clonogenic
neoblasts early after transplantation into irradiated hosts and whether the transplantation location
has an effect on the identity of the cells. Two possibilities are evident. The first is that the
progeny of clonogenic neoblasts are completely naYve of their environment and differentiate
randomly. If this were the case, then we would expect either the progeny cell to migrate long
distances to its destination tissue or undergo apoptosis if unable to find a niche. The second
possibility is that the progeny cell is capable of sensing its surroundings and engages a
differentiation pathway befitting its location. Recent improvements in the reliability of BrdU
labeling will enhance clonal analysis by single cell injection, as it will allow the tracing of the
differentiated progeny of the transplanted cell (van Wolfswinkel et al., 2014).
What proportion of the smedwi-1] cell population consists of clonogenic neoblasts? We
can determine the success rate of neoblast transplantation by performing an in situ hybridization
for smedwi-J on transplant hosts immediately following the procedure. We can also determine
the Pate of successful clonogenic neoblast grafting by examining colony formation several days
229
later.
However, attempting to infer the proportion of clonogenic neoblasts in the neoblast
population by calculating the ratio between these two rates will not likely to be accurate. One
possibility for an underestimate is that the transplanted cell is damaged during the procedure and,
although it may survive for a short time it may fail to engraft and divide. Evidence for this
hypothesis is the presence of dynamic processes on X1(FS) cells which have been observed to
break off upon loading or ejection from the injection needle. Another source of error in this
estimate is the use of visual identification of cells to be transplanted, which biases the
transplanted cells towards colony forming cells. Still another possibility is whether the state of
the cell, such as cell cycle or region of the animal it was collected from, has any bearing on its
success in colony formation upon transplantation. This illustrates the need for specific markers
for clonogenic neoblasts, as the transplant protocol is limited to certain types of experiments.
One major benefit offered by the development of single cell transplantation is its utility
for the generation of transgenic animals. If the cell can be manipulated prior to transplant into an
irradiated host, one could imagine creating entire transgenic lines using this technique.
Unfortunately, the current success rate of creating lines from single cells is prohibitively low
(about 1 in 45 transplant hosts recovered from lethal irradiation in the most successful
experiments). A reason for this low success rate is the inability of any given expanding colony
to restore regeneration before the entire animal succumbs to lethal irradiation.
One possible
solution to this issue is to alter the state of the host to be more amenable to colony expansion.
Upon wounding, planarians undergo two waves of synchronized mitoses (Wenemoser and
Reddien, 2010). Transplant hosts may be fed or lightly wounded to promote cell division during
early colony expansion phases.
However, wounding an animal also increases the rate of
230
programmed cell death, so attempting to stimulate proliferation this way may prove more
detrimental than beneficial (Pellettieri et al., 2010). More protocol development, especially for
methods to isolate a purer population of clonogenic neoblasts and to prevent hosts from
succumbing to lethal irradiation before the graft can restore regeneration, will be required to
allow transplantation to be a viable approach to transgenics.
Asexual planarians are incapable of sexual reproduction. Instead, asexual planarians of
sufficient size will undergo fission, where the tail will detach from the body and both pieces will
regenerate into two complete individuals. In this fissiparous form of reproduction, pre-existing
neoblasts are partitioned between the two individuals. It is conceivable that, over time, neoblasts
accrue mutations, thus generating a mixed population of cells with varying genomes. A unique
situation arises where each variant neoblast genotype can be thought of as an individual
subpopulation with its own fitness as well as a contribution to the overall fitness of the animal.
Heterogeneity in the clonogenic neoblast population may confer selective advantage to the host
animal by allowing adapted subpopulations to thrive under stress conditions. Selection against
deleterious mutations may be carried out by creating an environment where clonogenic neoblasts
must compete for limited resources.
Currently these theories do not have any supporting
evidence, but can be explored by examining the level of DNA sequence polymorphism within an
animal and seeing if strains generated from single cells have any differences from wild type
animals.
231
I. Functions of Planarian Glia
The function of planarian glia is currently unknown.
This is unsurprising because, despite
finding a signaling pathway that regulates the identity of the cell, we have no way of completely
ablating the cell. Previous studies have inhibited transcription factor expression by RNAi to
prevent the formation of a specific cell type during regeneration, but we currently do not know of
any transcription factors to be expressed in glial cells (Lapan and Reddien, 2012; Scimone et al.,
2011). Toxic small molecules that specifically act on Glutamine Synthetase have been used to
ablate glial cells in vertebrate systems, but two issues may confound analysis. First, a second
gene encoding a glutamine synthetase enzyme has been identified in the planarian genome and
initial in situ hybridization studies have indicated that it is expressed in cells outside the nervous
system. Second, the gs gene that has been identified in planarian glia is also expressed in other
tissues, most notably the intestine. Thus, a pharmacological approach will likely result in offtarget effects that cannot be divorced from the glia ablation phenotype.
Inhibition of the
expression of transcription factors necessary for specification is commonly used to ablate a cell
type in the study of planarian regeneration, but none have yet been identified as necessary for
glia formation. A candidate approach using homologs of transcription factors known to play a
role in glial cell specification in mouse and fly, such as Gcm and Olig2, has been attempted to no
avail (Hosoya et al., 1995; Zhou and Anderson, 2002). Compiling a list of transcription factors
expressed in planarian glia would require RNA-Seq on either a pure population of cells or a
single cell, both of which we currently have no method of obtaining. Until a means of ablating
232
the cell type is discovered, any functional data will have to be inferred based on morphology and
gene expression.
We have identified four planarian glial genes encoding orthologs of vertebrate and
invertebrate proteins with known functions. Interestingly, each of the four genes are predicted to
be involved in either neurotransmitter reuptake or neurotransmitter metabolism, and the
orthologs of two of the four genes, eaat2 and gs, are commonly used markers for astrocytes in
vertebrates and astrocyte-like glia in Drosophila. Three of these genes, eaat2, eaat3, and gat-],
encode members of the solute carrier family (Hediger et al., 2004). The excitatory amino acid
transporters are often co-expressed with glutamine synthetase enzymes, which catalyze the
processing of glutamate into glutamine (Anderson and Swanson, 2000). lonotropic glutamate
receptors and the glutamate biosynthesis gene glutaminase have been identified in the planarian
genome, thus providing evidence towards both the presence of glutamatergic neurons and a need
for a system to maintain extracellular levels of excitotoxic glutamate. The GABA transporter
gat-1 encodes a member of the SLC6 solute carrier family, of which gat-3 is expressed in
vertebrate glia and gat-] is expressed in insect glia (Kinjo et al., 2013; Oland et al., 2010).
Although we can predict that neurotransmitter uptake is performed by these cells, we have no
direct evidence. RNAi of these genes may result in a behavioral phenotype, but that alone is
insufficient to determine the solute specificity of these transporters.
Ideally, radiolabelled
neurotransmitters would be applied to live cells to track transport into the cell.
The planarian glial cell morphology is an indication of the putative extracellular
environment maintenance function. The processes of planarian glia are embedded in the axon
bundles of the neuropil and the peripheral nervous system, and also form a continuous layer
233
around synaptic glomeruli in the ventral nerve cords.
We posit that planarian glia extend
processes along axons to reuptake neurotransmitter molecules released at synapses. Although
we have described the highly branched nature of these cells as well as their close association
with neurons, axons and synapses, the actual shape of the cell is still unknown.
Immunofluorescence of GFAP labels cytoskeletal elements within vertebrate astrocytes and does
not provide any information of the topology of the cell membrane. Actin and Ezrin antibodies
label thousands of smaller peripheral astrocyte processes that emerge from the major GFAP+
processes to interact with synapses (Derouiche and Frotscher, 2001). We currently have no
means of observing the true structure of the cell, as we have no antibodies for other structures in
these cells, but we know that the neurotransmitter transporters are transmembrane domain
proteins and are likely to accumulate near synapses. Producing antibodies against these targets
will aid in the determining the morphology of the cell.
RNAi of if-1 and pcdh-19 in unamputated and regenerating fragments has not produced
any detectable phenotype. It is possible that these two genes are involved in a specific neuropil
glia function that would not have been tested in our assays. Intermediate filaments play two
well-known functions in cells of other organisms: providing mechanical support and increasing
cell size. These functions are well conserved among cytoplasmic intermediate filaments such as
GFAP (Herrmann et al., 2007). Given the density of synapses within the neuropil compared to
the peripheral nervous system, more regulation of the extracellular environment may be required.
Increasing the size of the cell may allow wider coverage of synapses by neuropil glia. Another
possibility is that the central nervous system is more sensitive to damage, given the greater
number of axons in a confined region that are more important to protect. The formation of IF- 1'
234
glial processes primarily in the neuropil may be a means of directing resources towards the
central nervous system at the expense of the more expendable peripheral nervous system.
Assays that specifically probe the sensitivity of animals to neuropil damage or neurotransmitter
reuptake may reveal a previously undescribed aspect of the phenotype.
235
III. Roles of Glia Cells in Regeneration
Although it is known that clonogenic neoblasts are the source of new tissue in planarian
regeneration, the mechanism by which the progeny of clonogenic neoblasts differentiate into
planarian glia has yet to be established. Given that glia are capable of proliferating in adulthood
in both vertebrates and invertebrates, it would be interesting to determine whether any of the
mechanisms regulating adult gliogenesis are present in planarians, or if they specifically follow
developmental paths of determination.
Heterogeneity observed in the neoblast compartment
suggests that there are intermediate stages between naYve stem cell and terminally differentiated
cell, and the characterization of other cell types within the neoblast population is currently the
focus of many projects in the planarian community.
One question is whether planarian glia arise from the same progenitor lineage as neurons.
This would entail an intermediate cell committed to an ectodermal lineage but retains the
capacity to form either neurons or glia, as well as another system for determining the fate of the
cell. The Notch signaling pathway is a principle mediator of neuron versus glia decisions both
during development and adult gliogenesis (Gaiano and Fishell, 2002; Wheeler et al., 2008), and
is an attractive target for continued study of planarian glia.
It is also possible that glia
progenitors, are specified independently of neural progenitors as well as all other progenitors,
which is an interesting possibility as it would suggest the presence of a master glial regulatory
factor.
The existence of such a factor would have implications in both comparisons of
differentiation mechanisms between species and the understanding of the evolution of the glial
cell type.
236
Glial involvement in axon guidance has been well established. Glia in C. elegans, the
midline glia in Drosophila, and the floor plate glia in vertebrates all provide cues that control
midline crossing of axons (Mitchell et al., 1996; Serafini et al., 1994; Wadsworth et al., 1996). In
adulthood, however, the glia are involved repression of outgrowth, as seen in neurite collapse by
myelinating oligodendrocytes and astroglial scars (McKeon et al., 1991; Wang et al., 2002).
Currently, there is no evidence that planarian glia provide instructive cues to neurons during
regeneration. Neural networks form prior to IF-W processes in the regeneration blastema. Also,
IF-1 processes also always contact with Synapsin+ or a-Tubulin+ axon bundles, but not all axon
bundles are associated with IF-1* processes. However, these experiments use the IF-I antibody
to trace the location of glial processes.
It is possible that glial processes exist prior to the
expression of if-1, as evidenced by the persistence of glial cells following inhibition of hh, the
possible existence of if-J glia in the peripheral nervous system, and the branched morphology of
gs+1'fF cells. One could imagine that planarian glia extend processes first and then stabilize
them with intermediate filaments upon induction by Hedgehog signaling.
The study of
homeostatic effects of glia on neural architecture may also reveal any roles in synapse pruning,
maintenance of connectivity, and overall stability of the neural network, but without a means of
determining the minutiae of glial morphology, this will be difficult to do.
Whether glia play a role in the AP polarity defect observed in hh(RNAi) and ptc(RNAi)
animals has not been fully explored. The defective tail regeneration resulting from inhibition of
hh or the regeneration of anterior tails from inhibition of ptc is a striking but rare phenotype,
requiring long-term inhibition by RNAi as well as multiple rounds of regeneration to clear
existing, pre-patterned tissues (Rink et al., 2009; Yazawa et al., 2009). Even then, a range of
237
expressivity of the respective phenotypes is observed.
hh(RNAi) animals can regenerate
bifurcated tails, half tails, or no tails. ptc(RNAi) animals can regenerate heads with cyclopic
photoreceptors or anterior tails (Rink et al., 2009; Yazawa et al., 2009).
Additionally, it is
possible that hh is expressed at low levels in other cells throughout the animal, but the major site
of hh production is the medial cortex of the cephalic ganglia and the anterior-most portions of the
ventral nerve cords.
The distance between the primary source of ligand, the head, and the
apparent target of signaling, the tail, raises questions of how Hedgehog signaling is controlling
polarity decisions so remotely. These data raise the possibility that Hedgehog signaling may be
indirectly involved in the establishment of AP polarity.
In contrast to the late-appearing AP polarity defect, changes in if-1 and pcdh-19
expression occur as early as 12 days after the onset of RNAi treatment.
Also, planarian glia
express ptc, providing evidence that these cells are direct targets of Hedgehog signaling. One
possible role of glia in polarity regeneration is to provide a route for hh ligand to travel to its site
of action. The Hedgehog protein has a cholesterol modification that allows it to associate with
cell membranes and limits its diffusion (Varjosalo and Taipale, 2008). One group has shown,
using fluorescently-tagged Sonic hedgehog, that the ligand binds to and is transported along
filopodia in order to cover longer distances (Sanders et al., 2013). A similar mechanism may
exist in planarians, where hh not only is transported by a glia track, but also promotes the
formation of that track. If that is the case, then it would suggest that if-1 or pcdh-19 are directly
involved in this mechanism, since only expression of these two genes and not the entire cell is
ablated by hh inhibition. Long-term inhibition of if-1 or pcdh-19 does not recapitulate the hh
238
RNAi morphological defect.
It is possible that inhibition of multiple genes regulated by
Hedgehog signaling is required to produce the AP defect.
239
IV. Glial Cell Evolution
Continued study of planarian glia will offer insight into whether the glia of protostomes and
deuterostomes originated from a single primordial glial cell type, or whether glia arose
independently in those two branches of the bilateria. Although we have found three planarian
genes, gs, gat-], and eaat2, that encode orthologs expressed in both vertebrate and Drosophila
glia, they are not glia-specific and cannot be used as evidence of common origin. Glutamine
Synthetase is found in the vertebrate liver (Gebhardt and Mecke, 1983), and related solute carrier
family members of gat-1 and eaat2 are also expressed in neurons (Jin et al., 2011; Zhou and
Danbolt, 2013). Comparisons of morphology are limited because of the lack of ultrastructural
resolution, but there are similarities, such as encapsulation of synaptic glomeruli (Boeckh and
Tolbert, 1993; Leise, 1990). Interestingly, the glial processes observed in the peripheral nervous
system do not appear to wrap orthogonal commissures. Complete encasement of neurons or
axons is thought to be a trait of "well-developed" glia, suggesting that the glial form we see in
the peripheral nervous system may have a morphology similar to the hypothetical ancestral glial
cell type and offer an opportunity to study the origins of glia (Hartline, 2011).
Whereas
invertebrate and vertebrate glia share a number of commonalities, such as morphology and
function, the developmental pathways of the two are dissimilar.
We do not yet know the key
factors responsible for specifying planarian glia, but we have evidence that orthologs for
Drosophila Gcm and mammalian Olig2 are not involved.
If planarian glia are specified in a
way unrelated to vertebrates and other invertebrates, it would be consistent with the theory that
the cell type arose multiple times during the course of animal evolution.
240
Identification of a role for Hedgehog signaling in glial cell identity in planarians draws
interesting parallels. In vertebrates, Sonic hedgehog released from neurons has been implicated
in generating heterogeneity in the astrocyte population (Garcia et al., 2010). In Drosophila,
Hedgehog signaling confers posterior midline glia identity during differentiation of midline glia
(Watson et al., 2011). Hedgehog signaling therefore may be involved in the specification of a
neuropil-localized subtype of planarian glia.
The similarity between if-1/pcdh-19 glia and
reactive astrocytes should also be considered. Sonic hedgehog is one of the extrinsic factors that
activate astrocytes to undergo hypertrophy during astrocyte scar formation (Sirko et al., 2013).
Expression of GFAP, an intermediate filament, is upregulated during hypertrophy to increase the
size of the cell and allow more efficient sealing of the wound site (Wilhelmsson et al., 2004). In
planaria, glia in the vicinity of hh ligand upregulate expression of if-1. The functions may be
different, but there is striking resemblance between the effect of Hedgehog signaling and cellular
response.
Hedgehog signaling does not appear to control central nervous system patterning in
planarians, based on both expression patterns and expression levels of neurogenic transcription
factors following inhibition of hh or ptc. In vertebrates, Sonic hedgehog is critical for specifying
the floor plate and patterning the ventral domain of the neural tube, and in Drosophila,
Hedgehog controls cell identity specification in developing neuroblasts (Bhat, 1999; Dessaud et
al., 2008). A question arises about whether the role of Hedgehog signaling in nervous system
patterning arose in the last common Bilaterian ancestor and was lost in the planarian lineage, or
whether Hedgehog signaling has been co-opted into nervous system patterning multiple times
throughout evolution, or a combination of the two. Clearly many more organisms from many
241
other phyla will need to be analyzed before any conclusions can be made, but our results begin to
support a certain viewpoint. The ventral to dorsal arrangement of neurogenic transcription factor
expression domains is similar to that of Drosophila. However, unlike vertebrates, expression
and localization of ventral neuroblastsdefective (vnd), intermediate neuroblastsdefective (ind)
and eyeless (ey) (which encode Drosophilaorthologs of Nkx2.1, Gsh2, and Pax6, respectively) is
independent of Hedgehog signaling (Cornell and Ohlen, 2000). Planaria have a medial to lateral
distribution of Nkx2 and Pax6 domains similar to the ventral to dorsal distribution in fly and
vertebrate, and those domains too are independent of Hedgehog signaling. Given these data, it is
likely that the involvement of Hedgehog signaling in patterning these neurogenic domains is a
vertebrate innovation.
A secondary topic raised by the discovery of if-1 in planarian glia is the evolution of
intermediate filaments in the context of glia. Intermediate filament proteins such as Vimentin,
Nestin, and GFAP are used as markers for glia, neural stem cells, and reactive astrocytes,
respectively. These subtypes of cytoplasmic intermediate filaments, however, arose when the
intermediate filaments radiated into six different families in the vertebrate lineage (Erber et al.,
1998). Even so, intermediate filaments have been found in invertebrate glia, the major exception
being Drosophila (Goldstein and Gunawardena, 2000; Hartline, 2011).
Given the highly
specialized morphology of glial cells, the need for additional cytoskeletal structure elements is
not surprising.
Whether intermediate filaments impart a structural property especially
advantageous to glia such that it would be incorporated multiple times throughout evolution, or
whether the use of intermediate filaments was inherited from a common glial cell ancestor is
unknown.
242
V. Conclusion
This work presents a number of opportunities for continued study of planarian stem cells and
glia. Of particular interest are the means by which glia are specified at the clonogenic neoblast
level and how their differentiation is modulated by the neural network into which they must
integrate.
Not only would elucidating this mechanism increase out understanding of the
specification of cell types in a highly regenerative model organism, but will also permit
comparisons between extant species with glia and provide insight into long standing questions of
the evolution of the cell type. In addition to understanding how glia are regenerated, continued
investigation of whether glia participate in the regeneration of other cell types and structures is
also pertinent, as it would aid in our understanding of adult gliogenesis and nervous system
repair in mammalian systems.
243
References
Anderson, C.M., and Swanson, R.A. (2000). Astrocyte glutamate transport: review of properties,
regulation, and physiological functions. Glia 32, 1-14.
Bhat, K.M. (1999). Segment polarity genes in neuroblast formation and identity specification
during Drosophilaneurogenesis. Bioessays 21, 472-485.
Boeckh, J., and Tolbert, L.P. (1993). Synaptic organization and development of the antennal lobe
in insects. Microsc. Res. Tech. 24,260-280.
Cornell, R.A., and Ohlen, T.V. (2000). Vnd/nkx, ind/gsh, and msh/msx: conserved regulators of
dorsoventral neural patterning? Current Opinion in Neurobiology 10, 63-71.
Derouiche, A., and Frotscher, M. (2001). Peripheral astrocyte processes: monitoring by selective
immunostaining for the actin-binding ERM proteins. Glia 36, 330-341.
Dessaud, E., McMahon, A.P., and Briscoe, J. (2008). Pattern formation in the vertebrate neural
tube: a sonic hedgehog morphogen-regulated transcriptional network. Development (Cambridge,
England) 135, 2489-2503.
Erber, A., Riemer, D., Bovenschulte, M., and Weber, K. (1998). Molecular Phylogeny of
Metazoan Intermediate Filament Proteins. J Mol Evol 47, 751-762.
Gaiano, N., and Fishell, G. (2002). The role of Notch in promoting glial and neural stem cell
fates. Annu. Rev. Neurosci. 25, 471-490.
Garcia, A.D.R., Petrova, R., Eng, L., and Joyner, A.L. (2010). Sonic hedgehog regulates discrete
populations of astrocytes in the adult mouse forebrain. Journal of Neuroscience 30, 1359713608.
Gebhardt, R., and Mecke, D. (1983). Heterogeneous distribution of glutamine synthetase among
rat liver parenchymal cells in situ and in primary culture. Embo J. 2, 567-570.
Goldstein, L.S., and Gunawardena, S. (2000). Flying through the Drosophilacytoskeletal
genome. J. Cell Biol. 150, F63-F68.
Guedelhoefer, O.C., and Sdnchez Alvarado, A. (2012). Amputation induces stem cell
mobilization to sites of injury during planarian regeneration. Development (Cambridge, England)
139, 3510-3520.
Hartline, D.K. (2011). The evolutionary origins of glia. Glia 59, 1215-1236.
Hediger, M.A., Romero, M.F., Peng, J.-B., Rolfs, A., Takanaga, H., and Bruford, E.A. (2004).
The ABCs of solute carriers: physiological, pathological and therapeutic implications of human
membrane transport proteinsIntroduction. Pflugers Arch. 447, 465-468.
244
Herrmann, H., Bar, H., Kreplak, L., Strelkov, S.V., and Aebi, U. (2007). Intermediate filaments:
from cell architecture to nanomechanics. Nat. Rev. Mol. Cell Biol. 8, 562-573.
Hosoya, T., Takizawa, K., Nitta, K., and Hotta, Y. (1995). glial cells missing: a binary switch
between neuronal and glial determination in Drosophila.Cell 82, 1025-1036.
Jin, X.-T., Galvan, A., Wichmann, T., and Smith, Y. (2011). Localization and Function of
GABA Transporters GAT-1 and GAT-3 in the Basal Ganglia. Front Syst Neurosci 5, 63.
Kinjo, A., Koito, T., Kawaguchi, S., and Inoue, K. (2013). Evolutionary history of the GABA
transporter (GAT) group revealed by marine invertebrate GAT- 1. PLoS ONE 8, e8241 0.
Lapan, S.W., and Reddien, P.W. (2012). Transcriptome analysis of the planarian eye identifies
ovo as a specific regulator of eye regeneration. Cell Rep 2, 294-307.
Leise, E.M. (1990). Modular construction of nervous systems: a basic principle of design for
invertebrates and vertebrates. Brain Res. Brain Res. Rev. 15, 1-23.
McKeon, R.J., Schreiber, R.C., Rudge, J.S., and Silver, J. (1991). Reduction of neurite outgrowth
in a model of glial scarring following CNS injury is correlated with the expression of inhibitory
molecules on reactive astrocytes. J. Neurosci. 11, 3398-3411.
Mitchell, K.J., Doyle, J.L., Serafini, T., Kennedy, T.E., Tessier-Lavigne, M., Goodman, C.S.,
and Dickson, B.J. (1996). Genetic analysis of Netrin genes in Drosophila:Netrins guide CNS
commissural axons and peripheral motor axons. Neuron 17, 203-215.
Oland, L.A., Gibson, N.J., and Tolbert, L.P. (2010). Localization of a GABA transporter to glial
cells in the developing and adult olfactory pathway of the moth Manduca sexta. J. Comp.
Neurol. 518, 815-838.
Pellettieri, J., Fitzgerald, P., Watanabe, S., Mancuso, J., Green, D.R., and SAnchez Alvarado, A.
(2010). Cell death and tissue remodeling in planarian regeneration. Developmental Biology 338,
76-85.
Rink, J.C., Gurley, K.A., Elliott, S.A., and Sanchez Alvarado, A. (2009). Planarian Hh Signaling
Regulates Regeneration Polarity and Links Hh Pathway Evolution to Cilia. Science 326, 14061410.
Sanders, T.A., Llagostera, E., and Barna, M. (2013). Specialized filopodia direct long-range
transport of SHH during vertebrate tissue patterning. Nature 497, 628-632.
Scimone, M.L., Srivastava, M., Bell, G.W., and Reddien, P.W. (2011). A regulatory program for
excretory system regeneration in planarians. Development (Cambridge, England) 138, 43874398.
Serafini, T., Kennedy, T.E., Galko, M.J., Mirzayan, C., Jessell, T.M., and Tessier-Lavigne, M.
245
(1994). The netrins define a family of axon outgrowth-promoting proteins homologous to C.
elegans UNC-6. Cell 78,409-424.
Sirko, S., Behrendt, G., Johansson, P.A., Tripathi, P., Costa, M., Bek, S., Heinrich, C., Tiedt, S.,
Colak, D., Dichgans, M., et al. (2013). Reactive glia in the injured brain acquire stem cell
properties in response to sonic hedgehog. [corrected]. Cell Stem Cell 12, 426-439.
van Wolfswinkel, J.C., Wagner, D.E., and Reddien, P.W. (2014). Single-Cell Analysis Reveals
Functionally Distinct Classes within the Planarian Stem Cell Compartment. Cell Stem Cell.
Varjosalo, M., and Taipale, J. (2008). Hedgehog: functions and mechanisms. Genes Dev. 22,
2454-2472.
Wadsworth, W.G., Bhatt, H., and Hedgecock, E.M. (1996). Neuroglia and pioneer neurons
express UNC-6 to provide global and local netrin cues for guiding migrations in C. elegans.
Neuron 16, 35-46.
Wang, K.C., Koprivica, V., Kim, J.A., Sivasankaran, R., Guo, Y., Neve, R.L., and He, Z. (2002).
Oligodendrocyte-myelin glycoprotein is a Nogo receptor ligand that inhibits neurite outgrowth.
Nature 417, 941-944.
Watson, J.D., Wheeler, S.R., Stagg, S.B., and Crews, S.T. (2011). Drosophilahedgehog
signaling and engrailed-runt mutual repression direct midline glia to alternative ensheathing and
non-ensheathing fates. Development (Cambridge, England) 138, 1285-1295.
Wenemoser, D., and Reddien, P.W. (2010). Planarian regeneration involves distinct stem cell
responses to wounds and tissue absence. Developmental Biology 344, 979-991.
Wheeler, S.R., Stagg, S.B., and Crews, S.T. (2008). Multiple Notch signaling events control
DrosophilaCNS midline neurogenesis, gliogenesis and neuronal identity. Development
(Cambridge, England) 135, 3071-3079.
Wilhelmsson, U., Li, L., Pekna, M., Berthold, C.-H., Blom, S., Eliasson, C., Renner, 0.,
Bushong, E., Ellisman, M., Morgan, T.E., et al. (2004). Absence of glial fibrillary acidic protein
and vimentin prevents hypertrophy of astrocytic processes and improves post-traumatic
regeneration. Journal of Neuroscience 24, 5016-5021.
Yazawa, S., Umesono, Y., Hayashi, T., Tarui, H., and Agata, K. (2009). Planarian
Hedgehog/Patched establishes anterior-posterior polarity by regulating Wnt signaling.
Proceedings of the National Academy of Sciences 106, 22329-22334.
Zhou, Q., and Anderson, D.J. (2002). The bHLH transcription factors OLIG2 and OLIGI couple
neuronal and glial subtype specification. Cell 109, 61-73.
Zhou, Y., and Danbolt, N.C. (2013). GABA and Glutamate Transporters in Brain. Front
Endocrinol (Lausanne) 4, 165.
246