Journal Club
Toyoda T, Mae SI, Tanaka H, Kondo Y, Funato M, Hosokawa Y, Sudo T,
Kawaguchi Y, Osafune K.
Cell aggregation optimizes the differentiation of human ESCs and iPSCs into
pancreatic bud-like progenitor cells.
Stem Cell Res. 2015 Jan 28;14(2):185-197. doi: 10.1016/j.scr.2015.01.007.
Choi JA, Ko SH, Park YR, Jee DH, Ko SH, Park CK.
Retinal Nerve Fiber Layer Loss Is Associated with Urinary Albuimin Excretion
in Patients with Type 2 Diabetes.
Ophthalmology. 2015 Feb 6. pii: S0161-6420(15)00003-2.
doi: 10.1016/j.ophtha.2015.01.001.
2015年3月12日 8:30-8:55
８階 医局
埼玉医科大学 総合医療センター 内分泌・糖尿病内科
Department of Endocrinology and Diabetes,
Saitama Medical Center, Saitama Medical University
松田 昌文
Matsuda, Masafumi
Organogenesis of the murine pancreas. The pancreatic bud begins around day 9
of gestation (E9.5) in the mouse embryo. At E11.5 the pancreatic ductal epithelium
grows and branches into the surrounding mesenchyme. The epithelial cells
differentiate into exocrine and endocrine cells at E15.5. Ductal, acinar, and islet
cells are clearly found by E18.5.
Front. Physiol., 18 July 2013 | doi: 10.3389/fphys.2013.00170
Pancreatic cell lineage and the role of transcription factors during islet cell type
development. All mature pancreatic cell types are derived from progenitors that express Pdx-1
and/or Ptf1a. Mnx-1 and Mib-1 act to regulate notch ligand activity. Ngn3 expression in pancreatic
progenitor cells gives rise to ductal, acini, and endocrine progenitor cells. Nkx2.2 and Nkx6.1 work
alongside other transcription factors including PDX-1, MafA Mnx1, Isl1 and NeuroD1 to regulate the
formation of the majority of islet β-cells.
Front. Physiol., 18 July 2013 | doi: 10.3389/fphys.2013.00170
Transcription factors
Pdx-1. Pancreatic duodenal homeobox 1 (Pdx-1) is one of the first transcription factors expressed, with gene expression starting as early as E8.5 in the
mouse in the foregut endoderm (Offield et al., 1996). All the cells derived from the endoderm have been shown to express Pdx-1 (Gu et al., 2002). Both
the ventral and dorsal pancreatic buds express Pdx-1 at E9.5 (Offield et al., 1996). At approximately E10 expression of Pdx-1 mRNA is then
downregulated with expression becoming restricted only to endocrine cells in the pancreas and this is maintained in adult β-cells (Ohlsson et al., 1993;
Ahlgren et al., 1998). Pdx-1 itself is an essential mediator of mesenchymal signaling, necessary for the branching morphogenesis involved in ductal
network formation of the pancreas at E10.5 (Ahlgren et al., 1996). Germline knockout studies have shown that while knockdown of Pdx-1 prior to E10.5
has no effect on pancreatic developmental processes (Wescott et al., 2009), the targeted pancreatic deletion at E10.5 or later results in pancreatic
agenesis (Ahlgren et al., 1996; Offield et al., 1996). Pdx-1 contains three principal transcription initiation sites (Sharma et al., 1996) and in the β-cell
each of these sites may be activated by the binding of a specific set of transcription factors (Melloul et al., 2002); FOXA2, HNF6, PTF1a, MNX-1, MAFA,
HNF, SP1/3, USF1/2 and PDX-1 itself (Harrison et al., 1999; Melloul et al., 2002; Jacquemin et al., 2003; Gao et al., 2008; Vanhoose et al., 2008).
Interestingly, reduced expression of the PDX-1 gene within the human pancreas has been linked with type 2 diabetes, a rare autosomal dominant form
of type 2 diabetes called maturity onset diabetes of the young (MODY4) and pancreatic agenesis (Lin and Vuguin, 2012).
Ptf1a/p48. Ptf1a (pancreas specific transcription factor 1a/p48) is a basic helix-loop-helix (bHLH) transcription factor (Beres et al., 2006). It was first
identified as a subunit of the trimeric PTF1 transcription factor complex, and has been found to be crucial for the regulation of exocrine gene
transcription (Ohlsson et al., 1993; Krapp et al., 1996). Endodermal expression of Ptf1a/p48 is pancreas specific throughout development; being
expressed in endocrine, exocrine and ductal cell types (Kawaguchi et al., 2002). PTF1a protein has been detected as early as E8–8.75 in the ventral
and dorsal pancreatic ducts (Hald et al., 2008) but by E13.5 expression becomes restricted to acinar precursor cells (Kawaguchi et al., 2002). In adult
rodents, PTF1a/p48 transcription factor protein is expressed in acinar tissue and induces amylase and elastase gene expression.
While a deficiency in PTF1a/p48 protein does not inhibit the initial formation of the pancreas it does cause a complete lack of acinar cell development
(Krapp et al., 1998; Kawaguchi et al., 2002). Cell lineage studies have shown that this is through cells adopting an intestinal fate rather than becoming
cells within the ventral pancreas (Kawaguchi et al., 2002). Defects in the human PTF1a protein are associated with permanent neonatal diabetes
mellitus (Masui et al., 2007). It has recently been suggested that Ptf1a-mediated control of Delta-like ligand 1 (Dll1) expression is crucial for Notchmediated control of early pancreas development (Ahnfelt-Ronne et al., 2012). The Ptf1a activation of Dll1 within multipotent progenitor cells (MPC)
stimulates proliferation and pancreas growth by maintenance of HES1 (hairy and enhancer of split 1) expression and PTF1a protein levels (AhnfeltRonne et al., 2012).
Endocrine lineage specification
Differentiation of the cells into each endocrine cell type found within the islet of Langerhans begins at specific time points during embryogenesis. For αcells this is E9.5, for β-cells E10.5, for δ-cells at E14.5 and lastly PP cells at E18.5. The critical window of differentiation of endocrine cells in humans is
from weeks 7 to 23 of gestation (Lin and Vuguin, 2012). Glucagon-producing α-cells are found at week 7, followed by insulin-producing β-cells,
somatostatin-producing δ-cells and pancreatic polypeptide PP cells at weeks 8–10 gestation (Lin and Vuguin, 2012). Importantly, the processes that
control the development of each cell type are governed by a complex signaling cascade.
Front. Physiol., 18 July 2013 | doi: 10.3389/fphys.2013.00170
Notch cell signaling. The notch cell signaling pathway is a well-conserved signaling pathway that regulates cell fate determination during embryonic
development as well as the maintenance of tissue homeostasis during adult life (Bigas and Espinosa, 2012). During development Notch signaling must
be regulated to balance the expansion of the progenitor pool with the differentiation of mature cell types (Greenwood et al., 2007), with genetic
perturbations of this pathway resulting in a change in pancreatic cell fate (Apelqvist et al., 1999). Notch-mediated signaling ensures the appropriate
maintenance of the progenitor cell population through the activation of the bHLH factor Hes1 (Jensen et al., 2000). Loss-of-function experiments have
shown that Notch signaling promotes pancreatic progenitor cell self-renewal and/or exocrine lineage commitment (Murtaugh et al., 2003). A reduction in
Notch signaling increases the expression of the pro-endocrine gene Ngn3, promoting an endocrine fate; while those cells exposed to Notch signaling
express Hes-1 and p48 mRNA and adopt an exocrine fate (Apelqvist et al., 1999).
Ngn3. Neurogenin 3 (NGN3) is one of the most important transcription factors for endocrine development. All endocrine cell types are derived from
progenitor cells expressing the Ngn3 gene (Gu et al., 2002). The bHLH transcription factor is expressed in two distinct waves (Villasenor et al., 2008).
The first phase of Ngn3 expression begins in murine pancreatic progenitor cells at E8.5–9 resulting in glucagon producing cells, while the second
begins at E12 with expression peaking at E15.5 corresponding with endocrine cell differentiation (Gradwohl et al., 2000; Schwitzgebel et al., 2000).
Ngn3 null animals are born without islets, and lack all endocrine cell types, develop type 1 diabetes and die 1–3 days after birth (Gradwohl et al., 2000).
Ngn3 has been associated with several downstream factors involved in endocrine differentiation and cell subtype specification and maintenance,
including Nkx2.2, Nkx6.1, Pax4 Isl1, b2/NeuroD1, Pax6 and Pdx-1 (Huang et al., 2000; Smith et al., 2003; Collombat et al., 2005).
Pax4—Arx. The paired box containing gene 4 (Pax4) mRNA is first detected at E9.5 and is transiently expressed in all endocrine progenitors during
pancreatic development, being downregulated shortly after birth (Lin and Vuguin, 2012). Pax4 is expressed downstream of Ngn3, with expression being
lost in Ngn3 null mice but not vice versa (Gradwohl et al., 2000; Wang et al., 2004). Pax4 activity appears essential for the appropriate initiation of β-cell
differentiation (Collombat et al., 2005). In the absence of Pax4, β- and δ-cells fail to develop and more α- cells are observed (Collombat et al., 2003,
2005; Sosa-Pineda, 2004). The loss of Pax4 prevents the expression of Pdx-1, HB9 and insulin mRNA in β-cell precursors (Wang et al., 2004). PAX4
has an opposing action to the transcription factor aristaless-related homeobox gene (ARX) (Collombat et al., 2003). Arx gene expression begins at E9.5
during mouse pancreatic development and persists into mature α-cells (Collombat et al., 2003). Arx null mice lack α-cells despite having no changes in
total islet cell number (Collombat et al., 2005). Therefore, it is the balance of Pax4 and Arx that is crucial for cell fate determination of both α- and βcells within the islet (Collombat et al., 2005).
Nkx. Members of the NK homeodomain transcription factor family are critical regulators of organ development (Stanfel et al., 2005). Of these, Nkx2.2
and Nkx6.1 are involved in endocrine cell lineage and are likely to be the most important during pancreatic development.
Nkx2.2 NKX2.2 is able to bind to both the insulin and Pax4 promoters (Cissell et al., 2003). The Nkx2.2 gene is initially dorsally expressed with Pdx-1 at
E8.75 and ventrally at E9.5 (Jorgensen et al., 2007). However, by E15.5 its expression becomes restricted to only endocrine cells (Sussel et al., 1998).
Nkx2.2 deficiency results in a reduction of α- and PP-cells, and the complete loss of β-cells (Sussel et al., 1998) and in the absence of Nkx2.2, β-cells
fail to activate insulin gene transcription and lack Nkx6.1, suggesting that Nkx2.2 is essential for the specification of the mature β-cell phenotype
(Sussel et al., 1998).
Front. Physiol., 18 July 2013 | doi: 10.3389/fphys.2013.00170
Nkx6.1 Nkx6.1 pancreatic gene expression is similar to that of Nkx2.2. In mouse models Nkx6.1 is expressed in the prospective ventral pancreas
domain at E8.75. By E9.0 Nkx6.1 expression switches from the ventral domain to the dorsal domain until E10.5, when ventral expression of Nkx6.1
reappears (Jorgensen et al., 2007) and then by E11.5, its expression becomes restricted to the central epithelium (Jorgensen et al., 2007). In adult
murine islets, Nkx6.1 expression is restricted to β-cells (Jensen et al., 1996) where it suppresses glucagon expression and modulates glucose
stimulated insulin secretion (Schisler et al., 2005). Nkx6.1 acts downstream of Nkx2.2 (Sussel et al., 1998) Not only do Nkx2.2 single mutant animals
lack expression of Nkx6.1 within the β-cell (Sussel et al., 1998) but the Nkx6.1 promoter contains a conserved binding site for Nkx2.2 (Sander et al.,
2000). In fact, Nkx6.1 deficiency does not affect early pancreas development, but it does result in compromised β-cell development (Sander et al.,
2000) and mutant animals lacking both Nkx2.2/Nkx6.1 have a phenotype identical to that of Nkx2.2 knockout animals (Sander et al., 2000).
Maintenance of β-cell identity
NeuroD1/Beta2. NeuroD1/Beta2, a bHLH, is first expressed in mouse pancreatic cells from E9.5 (Naya et al., 1997). By E12.5, NeuroD1/Beta2 mRNA
is found in both dorsal and ventral pancreatic buds and from E17.5 gene and protein expression is restricted to small clusters of endocrine cells and is
rarely detected in ductal epithelial cells (Chu et al., 2001). In adult islets, NeuroD1/Beta2 is expressed in all endocrine cells of the islet, although its
function still remains unknown. While it is believed that NeuroD1/Beta2 plays a role in regulating insulin (Sharma et al., 1999) and glucagon mRNA
expression, mice lacking NeuroD1/Beta2 are still able to produce functional insulin and glucagon (Chu et al., 2001).
Large Maf transcription factors. The Maf family of transcription factors belong to the basic leucine-zipper (bZIP) family that have been associated with
the regulation of a number of differentiation processes (Hang and Stein, 2011). The Maf family is divided into two groups (small and large) based on
their molecular size (Hang and Stein, 2011). The large Maf proteins MAFA, MAFB, c-MAF and NRL contain a transactivation domain in the N-terminal
region and are key regulators of cellular differentiation (Yang and Cvekl, 2007; Hang and Stein, 2011). In adult islets, MAFA, MAFB, and c-MAF are
expressed unlike NRL (Matsuoka et al., 2003), supporting a role for NRL in islet cell development.
MafA. MafA gene expression starts at E13.5 and continues throughout adulthood in the mouse (Matsuoka et al., 2003). MafA knockout mice display no
changes in pancreas development, but the mice do go on to develop glucose intolerance and type 2 diabetes as a result of reduced β-cell mass and
increased β-cell apoptosis (Zhang et al., 2005). Furthermore, MAFA binds to the insulin promoter and through its interaction with PDX-1 and NEUROD1,
activates insulin gene transcription (Olbrot et al., 2002; Aramata et al., 2005).
MafB. In the adult mouse pancreas MafB is expressed within the islet α-cells where it regulates glucagon gene expression (Artner et al., 2006, 2007).
Expression of MafB mRNA within the mouse pancreas begins as early as E10.5 in glucagon-positive cells, before expression in islet positive cells at
E12.5 (Artner et al., 2006). MafB becomes progressively restricted to α-cells postnatally through the downregulation of its expression in β-cells in the
first 2 weeks after birth and its expression is undetectable in these cells 3 weeks after birth (Artner et al., 2007). MAFB plays a pivotal role in the
development and differentiation of α- and β-cells during pancreatogenesis by directly affecting the transcription of specific genes present in α- or βcells as opposed to genes which are more ubiquitously expressed in other islet cell types (e.g., Pax6, Isl1 and NeuroD1) (Artner et al., 2007).
Mib-1. Mind bomb 1 (Mib-1) encodes an E3 ubiquitin ligase essential for Notch ligand activity (Itoh et al., 2003) that is required for correct proximodistal
patterning in the developing pancreas as well as in β-cell formation Horn et al., 2012). Endodermal-specific inactivation of Mib-1 causes the proximal
cells of the developing pancreas to adopt a distal fate resulting in a reduction in β-cells (Horn et al., 2012).
Front. Physiol., 18 July 2013 | doi: 10.3389/fphys.2013.00170
Mnx-1. The homeodomain transcription factor MNX1 (HB9) plays a key role in pancreatic development and function (Harrison et al., 1999; Li et al.,
1999; Dalgin and Prince, 2012). Murine Mnx-1 gene expression begins at E8 (Lin and Vuguin, 2012). Work in zebrafish suggests that Mnx-1 may
suppress α- cell fate in β-cell precursors (Dalgin and Prince, 2012). Furthermore, Mnx-1 knockout mice fail to develop the dorsal lobe of the pancreas
with the remnant pancreas having smaller islets of Langerhans with reduced β-cell numbers expressing low levels of Glut2 and Nkx6.1 (Harrison et al.,
1999).
Lbl-11/Isl-1. The transcriptional regulator Islet-1 (Isl-1) gene is expressed in the developing mouse pancreas at E8–8.5 (Lin and Vuguin, 2012) but is
later restricted only to the endocrine cells of adult islets, where it is involved in the maturation, proliferation and survival of the second wave of islet cells
(Du et al., 2009). Isl-1 directly regulates MafA, a potent regulator of the insulin gene and β-cell function (Du et al., 2009). Isl-1 deficient animals fail to
develop functional β-cells, and have a reduced ability to expand their endocrine cell mass postnatally and consequently become diabetic (Du et al.,
2009). A nonsense mutation of the ISL-1 gene has been found in a case of human type 1 diabetic patient. Activation of Isl-1 is regulated in part by the
transcriptional co-regulator, LIM domain-binding protein 1 (LBL-1) (Hunter et al., 2013).
Front. Physiol., 18 July 2013 | doi: 10.3389/fphys.2013.00170
Aims/Hypothesis:
Embryonic pancreatic bud cells, the earliest
pancreas-committed cells, generated from human
embryonic stem cells (hESCs) and induced
pluripotent stem cells (hiPSCs) have been shown
to differentiate into mature pancreatic β-cells in
vivo, indicating the feasibility of hESC/iPSCbased cell therapy for diabetes. However, the key
factors required for the differentiation of these
cells into pancreatic bud cells are incompletely
understood. The purpose of this study was to
establish culture conditions that efficiently induce
PDX1+NKX6.1+ pancreatic bud cells from
hESCs/iPSCs.
Methods:
In vitro differentiation of hESCs/iPSCs
Immunostaining
Flow cytometry
Quantitative real-time reverse transcriptionpolymerase chain reaction (qRT-PCR)
Animal studies and transplantation
experiments
Statistical analysis
In vitro differentiation of hESCs/iPSCs
The maintenance culture of a hESC line, KhES-3 (Suemori et al., 2006), and five hiPSC lines; 585A1, 604B1,
692D2, 648B1 and 409B2 (Okita et al., 2011 and Kajiwara et al., 2012) was performed as described previously (Mae
et al., 2013). For feeder-free cultures, cells were maintained with Essential 8 medium (Thermo Fisher Scientific,
Waltham, MA) according to the manufacturer's instructions. Experiments with hESCs/iPSCs were approved by
the ethics committee of the Department of Medicine and Graduate School of Medicine, Kyoto University. Cells
were directed into key stages of pancreatic development, including definitive endoderm (Stage 1), primitive gut
tube (Stage 2), posterior foregut (Stage 3) and pancreatic bud (Stage 4). The final protocol was as follows (Fig.
1A):
Stage 1 hESC/iPSC colonies grown on a feeder layer were first deprived of feeder cells and dissociated into
single cells as described previously (Suemori et al., 2006). The cells were resuspended with Stage 1 medium
containing RPMI 1640 medium (NACALAI TESQUE, Kyoto, Japan) supplemented with 2% (vol/vol) growth factor
reduced B27 (GFR-B27, Thermo Fisher Scientific), 50 U/ml penicillin/streptomycin (P/S, Thermo Fisher Scientific),
100 ng/ml activin A (R&D Systems, Minneapolis, MN), 3 μM CHIR99021 (Axon Medchem, Groningen, Netherlands)
and 10 μM Y-27632 (Wako, Osaka, Japan), seeded on Matrigel (Becton Dickinson, Franklin Lakes, NJ)-coated
plates at a density of 1 × 105 cells/cm2 and cultured for one day. For the next three days, the cells were cultured
in RPMI 1640 medium with 2% GFR-B27, 50 U/ml P/S, 100 ng/ml activin A and 1 μM CHIR99021.
Stage 2 The cells were exposed to Improved MEM Zinc Option (iMEM) medium (Thermo Fisher Scientific)
supplemented with 1% GFR-B27, 100 U/ml P/S (iMEM-B27) and 50 ng/ml keratinocyte growth factor (KGF; R&D
Systems) for three days.
Stage 3 The cultures were continued for three days in iMEM-B27 with 0.5 μM 3-Keto-N-aminoethyl-N′aminocaproyldihydrocinnamoyl cyclopamine (KAAD-CYC; Toronto Research Chemicals, Ontario, Canada), 0.5 nM
4-[(E)-2-(5,6,7,8-Tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-1-propenyl]benzoic acid (TTNPB, Santa Cruz
Biotechnology, Dallas, TX) and 100 ng/ml NOGGIN (Pepro-tech, Rocky Hill, NJ).
Stage 4 The cells were cultured for four to 20 days in iMEM-B27 with 100 ng/ml KGF, 100 ng/ml NOGGIN and 50
ng/ml epidermal growth factor (EGF, R&D Systems). For monolayer and aggregation cultures, the cells after Stage
3 treatments were dissociated into single cells by gentle pipetting after treatment with 0.25% trypsin–EDTA. The
same inducing factors were used as described above, except for the addition of 10 μM Y-27632 to the Stage 4
treatments. For monolayer cultures, the cells were seeded on Matrigel-coated plates at a density of 6–48 × 104
cells/cm2. To evaluate various extracellular matrices, the cells were seeded on laminin-111 (10 μg/ml; BioLamina,
Sundbyberg, Sweden), fibronectin (5 μg/ml, Merck, Whitehouse Station, NJ), Synthemax (Corning, Corning, NY)
and collagen I plates (Becton Dickinson). For aggregation cultures, the cells were seeded on a low-binding plate
at a density of 3–30 × 103 cells/well.
Figure 1. PDX1+NKX6.1+ cells
are localized at aggregated areas.
(A) A schematic diagram of the
procedures used for the
differentiation of pancreatic bud
cells from human ESCs. (B–D) A
hESC line, KhES-3, was
differentiated into definitive
endoderm, primitive gut tube
endoderm, posterior foregut and
pancreatic bud, and was
analyzed by immunostaining for
representative markers. (D) A
magnified image of (C). Note that
only the aggregated areas of
PDX1high cells co-expressed
NKX6.1. SOX17, SRY (sex
determining region Y)-box17;
FOXA2, forkhead box protein A2;
HNF, hepatocyte nuclear factor;
PDX1, pancreatic and duodenal
homeobox 1; GATA4, GATA
binding protein 4; NKX6.1, NK6
transcription factor related, locus
1. Scale bars, 300 μm in B and
100 μm in C and D.
Immunostaining
The cells were fixed with 4% paraformaldehyde (PFA) for 20 min at 4 °C. Then,
immunostaining was performed as described previously (Mae et al., 2013). The primary
antibodies used are detailed in Supplementary Table 1. The cell aggregates or
implanted grafts were fixed with 4% PFA for one to two days at 4 °C. After washing
with PBS, the samples were equilibrated in a 10–30% sucrose solution at room
temperature for 1 h, and then mounted and frozen. The frozen blocks were sectioned at
10–30 μm, and immunostaining was performed after removing the mounting medium.
For quantification of the Ki-67+ and cleaved-Caspase3+ cell ratios, immunostained cells
were analyzed using an image analyzer CellInsight NXT (Thermo Fisher Scientific,
Waltham, MA) or manual counting.
Flow cytometry
The cells were dissociated into single cells with 0.25% trypsin–EDTA treatment, fixed,
permeabilized and blocked with a BD Cytofix/Cytoperm Kit (Becton Dickinson). Then,
cells were stained with the antibodies detailed in Supplementary Table 1.
Quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR)
Total RNA was isolated from the cells with an RNeasy kit (Qiagen, Venlo, Netherlands),
and cDNA was prepared with a ReverTra Ace qPCR RT Master Mix (TOYOBO, Osaka,
Japan) and oligo (dT)20 primer, according to the manufacturer's instructions. The qRTPCR analysis was carried out with SYBR Premix Ex Taq II (Takara, Otsu, Japan). The
expression of each gene was normalized to the level of glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) expression. The primer sequences used are shown in
Supplementary Table 2.
Animal studies and transplantation experiments
All animal experiments were performed in accordance with the Guidelines for Animal
Experiments of Kyoto University. Male seven- to 14-week-old NOD.CB17-Prkdcscid/J
mice (NOD–SCID, Charles River Laboratories Japan, Yokohama, Japan) were
maintained on a 12-h light/dark cycle with ad libitum access to a standard irradiated
diet. Mice were anesthetized with inhalable isoflurane and received implants of hESCderived cell aggregates after Stage 4. Stage 4 day 4 cell aggregates
(30 × 103 cells/aggregate) were cultured in Stage 4 medium with or without an ALK5
inhibitor (Santa Cruz) for one to two additional days before implantation. Then, two
hundred cell aggregates per mouse were implanted under a kidney subcapsule. At 30–
210 days after implantation, the mice were sacrificed and the serial sections of grafts
were examined by immunostaining, as described above.
The graft function was assessed by measuring human C-peptide levels in mouse
plasma in response to glucose administration. The mice were fasted for > 5 h, and a
30% glucose solution was administered by intraperitoneal injection at a dose of
3.0 g/kg body weight. Blood samples were collected prior to and at 30 min after the
glucose administration via a tail vein into heparinized capillaries. The plasma human
C-peptide level was analyzed by an ELISA (Mercodia, Uppsala, Sweden), according to
the manufacturer's instructions. All metabolic analyses were performed in conscious
and restrained animals.
Statistical analysis
The statistical analyses were performed using paired t-tests (SigmaStat 3.5, Systat,
San Jose, CA). The differences between groups were considered significant for values
of p < 0.05.
Figure 2. Cell aggregation cultures promote
PDX1+NKX6.1+ cell induction from PDX1+ posterior
foregut cells. (A) A schematic diagram of the procedures
used to assess the effects of the cell density on the
cultures with regard to the induction of pancreatic bud
cells. Posterior foregut cells differentiated in adhesion
cultures were treated with the soluble factors required for
pancreatic bud induction (KGF, NOGGIN and EGF), either
with only a medium change (S4F) or with re-seeding at
different cell densities for monolayer cultures (6–
48x104 cells/cm2, S4F-2D) or to form cellular aggregates
(3-30x103cells/aggregate) (S4F-AG). (B) Representative
bright field images of the three culture formats. (C)
Quantification of the PDX1+NKX6.1+ cell ratio. After the
induction of pancreatic bud cells in the three culture
conditions described above, cells were examined by flow
cytometry. (D) The mRNA expression of pancreatic bud
markers, PDX1, PTF1A and NKX6.1, was assessed by
quantitative real-time polymerase chain reaction (qRTPCR) before and during pancreatic bud induction with cell
aggregation (black circle, solid line) and in monolayer
cultures (open circle, dotted line). The data are presented
as the fold-change in gene expression relative to the peak
value. (E) A hESC line (KhES-3) and five hiPSC lines
(585A1, 604B1, 692D2, 648B1 and 409B2) were
differentiated into posterior foregut cells in adhesion
cultures. Then, the foregut cells were dissociated, reseeded either for monolayer cultures (12 × 104 cells/cm2,
2D) or to form cellular aggregates
(3 × 103 cells/aggregate, AG) and were treated with KGF,
NOGGIN and EGF to induce pancreatic bud cells. After
the differentiation, the PDX1+NKX6.1+ cell ratio was
measured by flow cytometry. The data are presented as
the mean ± S.E.M. from three to seven independent
experiments in C and from three independent experiments
in D and E. Scale bar, 300 μm.
Figure 3. PDX1+NKX6.1+ cells emerge from
PDX1high cells in cellular aggregates. Cell
populations were analyzed by flow cytometry
after 0–20 days of aggregation culture at Stage 4.
(A) Representative dot plots indicating that
NKX6.1+ cells emerged from PDX1high cells. (B)
Quantification of the induction rate of
PDX1+NKX6.1+ (black circle, solid line) and the
total PDX1+ cells (open circle, dotted line). The
data are from three independent experiments
presented as the mean ± S.E.M.
Figure 4. PDX1+NKX6.1+ cells
in cellular aggregates coexpress multiple pancreatic
bud markers. The localization
of PDX1+NKX6.1+ cells in
aggregates was assessed.
Cryosections of cellular
aggregates at Stage 4, days 4
(A) and 12 (B), were stained
for the indicated markers of
pancreatic buds (PDX1,
NKX6.1, SOX9 and GATA4)
and endocrine cells (INS, GCG,
SST and GHR). Pancreatic
bud cells were distributed
throughout the entire
aggregates, and were
segregated from endocrine
compartments. SOX9, SRY
(sex determining region Y)box9; INS, insulin; GCG,
glucagon; SST, somatostatin;
GHR, ghrelin. Scale bar,
100 μm.
Figure 5.
Both cell aggregation cultures and soluble factor
treatment are required for the efficient induction of
PDX1+NKX6.1+ cells. (A) NKX6.1 expression was
assessed by qRT-PCR in the cells treated with
various combinations of soluble factors (100 ng/ml
KGF, 100 ng/ml NOGGIN and 50 ng/ml EGF) for four
days of Stage 4 aggregation culture. Cell aggregation
alone did not induce the expression of NKX6.1.
Treatment with all soluble factors with aggregation
cultures most potently induced the expression of
NKX6.1. (B) NOGGIN or all of the soluble factors
(KGF, NOGGIN and EGF) was/were removed at the
indicated periods during 12 days of aggregation
culture at Stage 4. (Upper) A schematic diagram of
the procedures used to assess the specific timing
requirements for the soluble factors. (Lower) The
quantification of the induction rate of PDX1+NKX6.1+
(black bars) and PDX1+NKX6.1− (white bars) cells.
The data are from three independent experiments
presented as the mean ± S.E.M. n/a, not applicable.
Figure 6. The PDX1+NKX6.1+ cells
generated with aggregation cultures
differentiate into branched pancreatic
epithelia in vivo. Cellular aggregates a
Stage 4, days 4–5, were implanted into
the kidney subcapsules of NOD–SCID
mice. (A) An image of a host kidney
harvested 30 days after implantation.
The arrow indicates the location of the
graft. (B–D) Cryosections of grafts
were stained for the indicated markers
Representative images of human
pancreatic cells are shown. Asterisks
indicate the host mouse kidney in B
and C. HuN, human nuclei. Scale bars
100 μm in B and D and 300 μm in C.
Figure 7.
PDX1+NKX6.1+ cells generated with aggregation cultures mature into functional β-cells in
vivo. A total of six million cells at Stage 4, day 5, of aggregation cultures that had been pretreated with an ALK5 inhibitor were implanted under the kidney subcapsule of NOD–SCID
mice. (A) The average plasma human C-peptide levels in host mice were measured 30 min
after glucose injection (3.0 g/ kg body weight, i.p.) at various time points after implantation, as
indicated. (B) The average plasma human C-peptide levels after > 5 h of fasting (before) and
30 min after glucose injection (after) in the host mice were examined on day 150 after
implantation. (C) A representative image of hESC-derived islet-like endocrine clusters on day
210 after implantation. The data are presented as the mean ± S.E.M. of 13 mice from four
independent cohorts of implantation experiments in A and four mice from two independent
cohorts of implantation experiments in B. *P < 0.05. Scale bar, 100 μm.
Results:
We differentiated a hESC line, KhES-3, into pancreatic
lineages with a stepwise protocol recapitulating
developmental process. The induction rate of
PDX1+NKX6.1+ cells was correlated with cell density in
adherent cultures, and markedly improved with cell
aggregation cultures. The positive effects of cell aggregation
cultures on the differentiation of pancreatic bud cells were
reproduced in multiple hESC/iPSC lines. The human
PDX1+NKX6.1+ cells developed into pancreatic epithelia
after implantation into immunocompromised mice. Moreover,
human C-peptide secretion into mouse bloodstream was
stimulated by glucose challenges after in vivo maturation.
Conclusions:
Taken together, these results suggest that
cultures with high cell density are crucial for the
differentiation of pancreas-committed progenitor
cells from hESCs/iPSCs. Our findings may be
applicable for the development of hESC/iPSCbased cell therapy for diabetes.
Message
糖尿病治療に向けてヒトES/iPS細胞から移植用の膵細胞を効率よく作製する方法を開
発 豊田太郎助教（京都大学CiRA増殖分化機構研究部門）、長船健二教授（京都大学
CiRA増殖分化機構研究部門）らの研究グループ
ヒト多能性幹細胞（ES細胞およびiPS細胞）を膵臓の元となる膵芽（すいが）細胞へ
と高効率に作製する培養条件を確立し、さらに作製した細胞が移植後に血糖値に応じ
たインスリン分泌をする細胞へと成熟可能であることを示しました。この研究成果は
2015年1月28日(英国時間)に「Stem Cell Research」にオンライン公開されました。
膵臓は、膵前駆細胞と呼ばれる一層の細胞シートから膵芽と呼ばれる細胞の塊をつ
くることで初めて形として認識することができます。つまり、膵芽は膵臓の最初の組
織であると考えられます。このため、膵芽細胞は糖尿病に対する細胞移植療法をはじ
めとした膵臓再生医療の基盤となる細胞源として期待されています。これまでにヒト
多能性幹細胞から膵芽細胞を作製する方法はいくつか報告されていますが、分化の仕
組みが完全には分かっておらず、安定性、効率などの点について改良の余地がありま
した。
本研究では、ヒトの膵発生過程において膵芽細胞が出現する際に細胞の塊をつくる
ことに着目し、それを培養皿上で再現しました。その結果、細胞密度と相関して膵芽
細胞（PDX1+ NKX6.1+ cell）への分化が促進し、細胞塊を作製して培養することでさ
らに効率よく分化することを発見しました。また、細胞塊の形成で分化誘導に効果が
見られたことから、細胞間相互作用を介した新たな分化の仕組みが存在することも示
唆されました。さらに、作製した膵芽細胞をマウスに移植すると、生着して胎児の膵
臓に似た組織構造を形成し、最終的には血糖値に応じてインスリンを分泌する成熟し
た膵β細胞へと分化しました。
網膜神経線維層欠損
視神経乳頭部から扇状に広がる、周りの網膜の色と比
べて少しくすんで見える部分が網膜神経線維層欠損部
です。網膜神経線維層欠損は最も早期に生じる緑内障
性眼底変化といわれ、その部位に一致する視野障害を
認めることが多い所見です。
網膜神経線維層欠損（白黒写真）
網膜神経線維層欠損を無赤色フィルター光にて撮影す
ると、その部分が周りの網膜に比べ暗く写り、通常の検
査よりはっきりと観察することが可能となります。
緑内障を疑う重要な所見です。文字通り神経が減っている所見です。神経が減っている
ところは眼から脳への情報が伝えられませんので、視野が欠損します。
http://www.ningyocho-ganka.com/?page_id=42
1 Department of Ophthalmology, St. Vincent’s Hospital College of Medicine, The
Catholic University of Korea, Seoul, Republic of Korea.
2 Division of Endocrinology and Metabolism, Cheonan Chungmu Hospital,
Cheonan, Republic of Korea.
3 Division of Endocrinology and Metabolism, St. Vincent’s Hospital, College of
Medicine, The Catholic University of Korea, Seoul, Republic of Korea.
4 Department of Ophthalmology, Seoul St. Mary’s Hospital, College of Medicine,
The Catholic University of Korea, Seoul, Republic of Korea.
DOI: http://dx.doi.org/10.1016/j.ophtha.2015.01.001
Objectives
To identify the factors associated with retinal
nerve fiber layer (RNFL) loss in patients
with type 2 diabetes.
Design
Cross-sectional study.
Participants
Ninety-six nonglaucomatous patients with type 2 diabetes without
renal impairment (estimated glomerular filtration rate, ≥60 ml/minute
per 1.73 m2).
Methods
Eyes were divided into 2 groups based on the presence or absence
of RNFL defects detected by red-free retinal fundus photography. All
participants underwent an eye fundus examination, and the urinary
albumin-to-creatinine ratio (ACR) was determined. A cardiovascular
autonomic function test was performed using the following heart rate
variability parameters: expiration-to-inspiration ratio, response to the
Valsalva maneuver, and standing. Multiple logistic regression
analyses were performed to determine potential risk factors related to
the presence of RNFL defects in these patients.
Main Outcomes and Measures
Main Outcomes and Measures
The association between RNFL defects and diabetic complications.
Figure 1. Red-free
fundus photograph
showing a
representative case of
retinal nerve fiber layer
(RNFL) defect in a 43year-old man with type
2 diabetes. A narrow
RNFL defect is shown
at the 1-o’clock
position in the red-free
photograph (white
arrows). A cotton-wool
spot is seen on the
inferior-temporal side
of the retina.
In this study, we excluded eyes with a glaucomatous optic disc because an RNFL defect
is also a characteristic feature of glaucoma. Consistent with previous studies, we found
that RNFL defects occurred predominantly in the optic disc on the superior side. The
locations of the RNFL defects in patients with diabetes seemed to be clearly different
from those resulting from glaucomatous RNFL damage, which occurs predominantly in
the inferior temporal retina.
Results
Among the patients, 43 (44.8%) had localized RNFL defects (group
1), whereas the others (55.2%) did not (group 2). The RNFL defects
occurred more frequently on the superior side (75.6% and 71.0% in
right and left eyes, respectively) compared with the inferior side
(13.8% and 0.0% in right and left eyes, respectively). Patients with
RNFL defects (group 1) had significantly higher rates of diabetic
retinopathy (60.5%) compared with those without RNFL defects
(group 2; 32.1%; P = 0.007). The urinary ACR was significantly
higher in patients with RNFL defects than in those without defects
(45.3±72.1 μg/mg vs. 15.4±17.3 μg/mg creatinine, respectively; P =
0.015), whereas autonomic function test grading was similar between
the groups. The urinary ACR was the only factor related to visual field
defect location in both univariate (P = 0.021) and multivariate (P =
0.036) logistic regression analyses after adjusting for age; gender;
presence of diabetic retinopathy; diabetes duration; smoking; statin
use; and antiplatelet, angiotensin-converting enzyme inhibitor or
angiotensin receptor blocker treatment.
Conclusions
Urinary albumin excretion was associated
with nerve fiber layer loss in patients with
type 2 diabetes. Careful examination of the
optic nerve head may be necessary,
particularly in patients with type 2 diabetes
exhibiting albuminuria.
Message
腎障害のない非緑内障の2型糖尿病（DM）患者96
人を対象に、網膜神経線維層（RNFL）欠損の関
連因子を横断研究で特定。RNFL欠損発生率は
44.8％だった。糖尿病性網膜症および尿中アル
ブミン/クレアチニン比（ACR）はRNFL欠損なし
群に比べ、欠損あり群で有意に増加した。尿中
アルブミン排泄量がRNFL欠損と関連すると示唆
された。
もともと糖尿病で網膜神経線維層（RNFL）欠損
が言われていた。眼圧正常者を集めているが、
少なくとも眼圧は表示すべきだろう。
http://www.m3.com/clinical/journal/15179