Global characterization of the Pho regulon in
Caulobacter crescentus
by
Emma A. Lubin
A.B. Biochemistry & Molecular Biology
Dartmouth College, Hanover, 2006
SUBMITTED TO THE DEPARTMENT OF BIOLOGY IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY IN BIOLOGY
AT THE
OF TECHNOLOGY
INSTITUTE
MASSACHUSETTS
TC4OLOGY~~-
FEBRUARY 2014
BR
@2014 Emma A. Lubin. All rights reserved.
The author hereby grants MIT permission to reproduce and 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:
Emma A. Lubin
Department of Biology
January 10, 2014
Certified by:
Michael T. Laub
Associate Professor of Biology
Thesis supervisor
Accepted by:
Stephen P. Bell
Professor of Biology
I
i
Global characterization of the Pho regulon in
Caulobacter crescentus
by
Emma A. Lubin
Submitted to the Department of Biology
January 10, 2014 in partial fulfillment of the requirements of the degree of
Doctor of Philosophy in Biology at the Massachusetts Institute of Technology
Abstract
Bacteria must sense and respond to their environment in order to survive and proliferate.
Adapting to phosphate-limited conditions is particularly critical, as phosphate is a central
component of many important biomolecules. Most bacteria respond to phosphate
limitation through a widely conserved pathway, composed of the phosphate transport Pst
system, and downstream signal transduction pathway, PhoR-PhoB, termed the Pho
system.
In this thesis, I use the model organism Caulobactercrescentus to characterize the
response to phosphate limitation. I use ChIP-Seq on the transcriptional regulator PhoB to
globally map the Pho regulon in Caulobacterin both phosphate-starved and -replete
conditions. I find that the regulatory regions of over 50 genes are bound by PhoB
following phosphate limitation, and I identify a consensus PhoB binding motif in
Caulobacter.I then examine the function of PhoU, which is a putative negative regulator
of the Pho regulon in Caulobacterand many other bacteria. I use morphological and
microarray data to demonstrate that PhoU is not a negative regulator of the Pho regulon,
and that it instead acts outside the PhoR-PhoB pathway. I find that the function of PhoU
is tightly linked to cellular phosphate metabolism. This work offers insight into how
Caulobacterresponds to nutrient stress, as well as a better understanding of the
connectivity and output of the phosphate limitation response pathway.
Thesis advisor: Michael T. Laub
Title: Associate Professor of Biology
3
Acknowledgements
This work could not have been completed without the help of many people. First
and foremost, I would like to thank my advisor, Mike Laub. If I had to hazard a guess, I
would say that raising a graduate student must be a pretty thankless task. I would like to
take this opportunity to thank Mike for not only teaching me a tremendous amount of
science but for the faith he has always shown in me. I've always been grateful for his
guidance and insight, but it's been only since I started preparing to leave the lab that I
really understood how much effort and thought he'd invested into making sure I learned
the professional skills I'd need to go on in science, not just the scientific ones. I feel very
lucky to have stumbled into this lab, and I couldn't have asked for a better mentor.
I would like to thank my committee members, Graham Walker and Dennis Kim,
for their scientific advice, and invaluable and constant support and encouragement over
the past few years. I am especially grateful to Sue Lovett for serving as my outside
committee member.
Barrett Perchuk, the Laub lab manager and keeper of lab knowledge, is an
essential component of all work done in the Laub lab; without him most of these
experiments would never have gotten done. I would like to thank Sally MacGillivray, the
lab administrative assistant, who is the sort of person who makes any place she is in a
better place to be.
I would also like to thank my labmates; I couldn't imagine having gone through
this without them. I would particularly like to thank those who have invested
considerable time and effort into scavenging free food with me: Erin Chen, Josh Modell,
Orr Ashenberg, Chris Aakre, Diane Baer, Mike Salazar, and Sri Kosuri. I would also like
to thank my past and current baymates, Celeste Peterson and Leonor Garcia Bayona, for
putting up with the aesthetic horror that is my lab bench; Erin Chen, Christos Tsokos, and
Kasia Gora, for starting off in the lab with me; Christos Tsokos, Diane Baer, Chris Aakre,
and Andy Yuan for their technical help; and the specificity group Jeff Skerker, Emily
Capra, Anna Podgomia, and Orr Ashenberg.
My neighbors on the fifth and sixth floors have made Building 68 a fantastic place
to work, and I particularly would not have liked to have gone through grad school
without Shankar Sundar and Steve Glynn nearby in the Sauer lab.
I would like to thank Steve Bell for his advice early on, and for being so generous
in letting me use his lab equipment, and Frank Solomon for being Frank Solomon.
I would like to thank Officer Sean Collier and the MIT police department for
protecting this campus, its students, and the work that is done here.
Nothing I've done at MIT would have been remotely possible without the Biology
department administrative staff, particularly Betsey Walsh, without whose reminders I
would have lost my entire stipend to late fees. I would also like to thank Janice Chang,
and the Biology Education staff, Maggie Cabral, Luke and Bio Headquarters, the Biology
financial office, especially Mary Mango, and John Fucillo and the safety and facilities
staff for all of their patience and help over the years.
My work would have taken a lot longer if not for the incredible Building 68
kitchen staff; getting to benefit from their work is a luxury I'm grateful for daily.
I would also like to thank the Building 68 custodial staff, particularly Richie, and
Francisco from the late-late-night custodial crew.
4
I got into science in college because I was fortunate enough to have fantastic
undergraduate mentors. I would like to thank Jim DiRenzo at Dartmouth Medical School
for giving me my first lab job and showing me the ropes, and the graduate students and
postdocs who supervised me over four years: Allison Abbott, Chris Hammell, Maria Ow,
and Chris Harmes. I would especially like to thank Candy Lee and Victor Ambros, whose
lab I worked in during my last 3 years of college. Aside from being two of the most
inspiring scientists around, they've mentored and supported me through everything I've
done since, and I'm grateful that they let me continue to bother them for advice years
later.
In my family, I am surrounded by scientists on all sides. I would especially like to
acknowledge my grandfather Martin Lubin, a Professor Emeritus of Microbiology, and
my grandmother Dorothy Lubin, who edited his papers, for inspiring a love of science
early on. I would also like to acknowledge my maternal grandmother Barbara Gimbel,
who received her PhD in philosophy of science, and taught history of science.
I would like to thank my siblings, Amos, Isabella, and Rebecca, who are the
reason I do anything, and my parents Adam and Victoria, and my many aunts, uncles,
and cousins for their love and support.
I could never have imagined being so welcomed by a family other than my own,
but I had never met the Tsokoses: thank you George and Maria, Sophia and Ben -- and
especially Theo, for your considerable contributions to science at the age of 18 months. I
don't deserve any of you.
I would like to thank my friends, and in particular Rashelle Lee, Julie Valastyan,
and Liza Bouton, as well as Liza's parents, Judi and Jon.
During orientation my first week at MIT, Steve Bell introduced a graduate student
speaker who had married her classmate. He joked that we should all look around the
room because we might be sitting near our future husband or wife. I was skeptical, but
only because Christos had decided to skip orientation that day. I would like to thank
Christos Tsokos for many helpful discussions -- some about science, most about cats --
but mostly, for everything.
5
Table of Contents
Abstract
3
Acknowledgements
4
Chapter 1: Introduction
9
Bacterial responses to phosphate limitation
10
Transport
11
Phosphatescavenging
11
Membrane composition
The oxidative stress response
13
14
Polyphosphate
Virulence
15
17
Motile-to-sessile transitions
Stalk elongation
19
19
The phosphate limitation response pathway
21
The Pstsystem: an ABC-type transporter
PhoR/PhoB: a two-component signaling system
Histidine kinases
Response regulators
Two-component specificity
22
26
27
29
29
Hybrid histidine kinases
30
Regulation of two-component systems
30
The general stress response in bacteria
E. coli: RpoS and the generalstress response
a-proteobacteria:ECF ufactors
PhoU and PhoU-like domains
PhoU: a putative regulatorof the PhoR/PhoBpathway
33
34
35
36
36
YjbB: a transporter-fusedPhoU domain protein in Escherichiacoli
39
Archaeal PhoUhomologs
Eukaryotic chaperonecofactors
39
40
Global characterization of the Pho regulon in Caulobactercrescentus
41
References
42
Chapter 2: Global characterization of the Pho regulon in Caulobactercrescentus
48
Abstract
49
Introduction
50
Results
Epitope-taggedPhoB retainswild-type function
ChIP-Seq reveals genome-wide binding patternsof PhoB
53
53
55
Identification of the PhoB regulon
PhoU is not a negative regulatorof the Pho regulon in Caulobacter
Mutations in both the Pst and Pho systems suppress a phoU mutant
6
60
62
65
Discussion
68
69
PhoB regulates a different set ofgenes in Caulobacterthan in E. coli
PhoU does not regulate PhoR activity in Caulobacterand instead likely regulatesphosphate
70
metabolism
Characterizationof the response to phosphate limitation in Caulobactercrescentus
72
Experimental Procedures
Strains and growth conditions
Microscopy
/-galactosidase assays
Immunoblots
ChIP-Seq and analysis
DNA microarrays
Colony forming units
Transposon mutagenesis and rescue cloning
73
73
74
74
74
75
76
76
77
References
78
Chapter 3: Conclusions and Future Work
82
The Pho regulon in Caulobacter crescentus
83
83
Pho regulon specialization in Caulobacter
84
What genes regulatestalk elongation in Caulobacter?
How does Caulobacterintegrate the response to phosphate limitation with two different cell
86
types?
What other mechanisms does Caulobacter use to respond to phosphate limitation? 89
Indirect PhoB targets
The PhoB-independentphosphate response
phoH: an uncharacterizedstress responsegene
What is the function of PhoU?
89
90
92
93
Phosphate-dependencyof phoU depletion lethality
93
PhoUand polyphosphate
96
Regulation of PhoR
98
References
100
Appendix 1: The Pho regulon in Caulobacter crescentus
103
References
106
Appendix 2: Complete set of peaks identified by PhoB ChIP-Seq
107
References
117
Appendix 3: Genes regulated in response to phoU depletion in Caulobacter
118
7
List of Figures and Tables
Chapter 1: Introduction
Figure 1.1 - Polyphosphate
Figure
Figure
Figure
Figure
Figure
Figure
1.2
1.3
1.4
1.5
1.6
1.7
- The Pst/Pho phosphate limitation response pathway
- Mechanism of ABC-type transporters
- Two-component signal transduction systems
- Regulation of NtrB by glutamine and a-ketoglutarate through P11
- The general stress response
- PhoU is a putative regulator of the PhoR/PhoB pathway
Chapter 2: Global characterization of the Pho regulon in Caulobactercrescentus
Figure 2.1 - A strain harboring C-terminally 3XFLAG tagged phoB behaves like wildtype in phosphate-replete and phosphate-limited conditions
Figure 2.2 - ChIP-Seq reveals genome-wide binding patterns of PhoB
Figure 2.3 - PhoB binds Pho regulon genes upon phosphate limitation
Figure 2.4 - ChIP-Seq differentiates between direct and indirect PhoB targets, and
identifies PhoB-repressed genes
Figure 2.5 - pho U depletion does not phenocopy pstS mutation
Figure 2.6 - pho U functions independently of the Pho regulon
Figure 2.7 - Mutations in the pst and pho genes suppress pho U depletion lethality
Chapter 3: Conclusions and future work
Figure 3.1 - The Caulobactercrescentus cell cycle
Figure 3.2 - Deletion of CC3094 suppresses the pstS mutant large swarm
Figure 3.3 - Growth on minimal medium, but not at low temperature, suppresses pho U
depletion lethality
Table 3.1 - Candidate regulators of stalk elongation in Caulobacter
Table 3.2 - Indirect PhoB-regulated genes
Table 3.3 - PhoB-independent genes
8
Chapter 1: Introduction
9
Cells must rapidly sense and respond to their external environment to survive. For
bacteria, some of which often face widely fluctuating conditions, this adaptability is
especially important. Although various nutrients are essential for bacterial life, effectively
sensing and responding to the phosphate state of the environment is particularly crucial.
Phosphorus is a component of multiple central biomolecules, but is often a limiting
resource in bacterial environments. Compounds that rely on phosphate range from DNA
to phospholipids, and the levels and usage of these molecules must be tightly regulated in
phosphate-starved conditions. In addition to containing free phosphate at a concentration
of roughly 10 mM (Rao et al., 2009; Wanner, 1996), the dry biomass of E. coli has been
calculated to be composed of roughly 20% RNA, 9% lipids, 4% free metabolites, and 3%
DNA -- all of which contain phosphate (Blank, 2012).
Bacterial responses to phosphate limitation
In phosphate-limited conditions, bacteria must execute specific responses to adapt and
survive. Several types of responses to low environmental phosphate are common to
multiple bacterial species. In particular, regulation of transport and phosphate scavenging
genes, alteration of membrane composition, and activation of oxidative stress genes have
all been observed as responses to low phosphate in diverse bacteria. When sufficient
levels of phosphate are available after starvation, an increase in levels of polyphosphate is
also observed (Ohtake et al., 1998). Additionally, in several species, low phosphate has a
more dramatic effect, stimulating a change in the lifestyle of the bacterium. Each of these
various responses to phosphate limitation are discussed below in detail.
10
Transport
A common response to low phosphate is to increase import of both inorganic phosphate
and phosphorylated compounds. This is accomplished by upregulating the expression of
transporters for phosphate and alternative phosphate sources. Among these transporters is
the high-affinity Pst system, which imports inorganic phosphate (discussed below). In
addition, E. coli increases its uptake of two classes of phosphorylated compounds in
response to phosphate limitation. It upregulates expression of the ugpBAEC locus which
encodes an ABC transporter that imports glycerol-3-P (Blank, 2012), as well as the 14gene phn locus, which is responsible for importing and metabolizing phosphonates.
Phosphonates are phosphate-containing compounds characterized by the presence of a
stable C-P bond. In E. coli the phnC-phnE genes encode a phosphonate transporter,
which is also capable of transporting inorganic phosphate, while the phnG-phnM genes
encode a C-P lyase complex capable of hydrolyzing C-P bonds (Dyrman et al., 2006;
Kamat and Raushel, 2013; Metcalf et al., 1990; Metcalf and Wanner, 1993).
Phosphate scavenging
In addition to importing phosphate and phosphate-containing compounds, some bacteria
export enzymes that can non-specifically remove phosphoryl groups from compounds in
the extracellular environment; the resulting inorganic phosphate is then imported into the
cell.
The majority of these exported enzymes are alkaline phosphatases, and the best
characterized of them is phoA. Expression of phoA is induced upon phosphate limitation
in E. coli (Brickman and Beckwith, 1975; Torriani, 1960), and induction ofphoA
expression is often used as a marker of activation of the phosphate limitation response. In
11
Gram negative bacteria, PhoA is localized to the periplasm, while in Gram positive
bacteria it is membrane-bound (Zaheer et al., 2009).
A phoA homolog has not been identified in Sinorhizobium meliloti and other closely
related a-proteobacteria. These bacteria, along with many marine bacteria, instead
encode phoX (Kathuria and Martiny, 2011; Sebastian and Ammerman, 2009), which is
the major alkaline phosphatase in S. meliloti (Zaheer et al., 2009). Expression ofphoX is
induced by phosphate limitation, and the PhoX protein contains a Tat translocation
signal; accordingly, the mature protein has been found localized to the periplasm (Zaheer
et al., 2009). In vitro characterization of PhoX protein purified from S. meliloti has found
that it has optimal phosphatase activity in the presence of calcium and at pH 9-11, and
exhibits low substrate specificity for C-O-P bonds, and is able to remove phosphoryl
groups from a range of nucleotides, phosphorylated carbohydrates, and amino acids
(Zaheer et al., 2009).
A third class of secreted alkaline phosphatases, phoD, is similarly produced in response
to phosphate limitation, and has been identified in Bacillus subtilis as well as in marine
bacteria. It does not appear to act as the primary phosphate-induced alkaline phosphatase
in these species, and is found in addition to the presence of phoA or phoX(Eder et al.,
1996; Kageyama et al., 2011). In the marine bacterium Aphanothece halophytica,
expression of the phoD phosphatase is induced by salt stress as well as phosphate
limitation (Kageyama et al., 2011).
12
Membrane composition
In response to phosphate limitation, E. coli and other bacteria alter their membrane
composition. This process has been proposed to have two different functions. First, it
alters membrane fluidity to increase resistance to environmental stress that may be
concomitant with phosphate limitation, and second, it replaces phosphorylated membrane
components with unphosphorylated substitutes to save the phosphate for other, more
critical, cellular processes.
Both models are supported by evidence from strains of E. coli, which show altered
resistance to stress, as well as lower levels of some phosphorylated lipids, in low
phosphate conditions. Mutations in the regulatory pathway that controls the response to
phosphate limitation have been found to alter membrane composition and permeability. A
mutation in the pst phosphate transport operon in an extraintestinal pathogenic
Escherichiacoli (ExPEC) strain, which results in hyperactivation of the Pho regulon,
results in increased outer membrane permeability, and copy number of the pst genes has
been found to influence fatty acid regulation (Lamarche and Harel, 2009). Further,
mutation of the pst system in this strain results in increased sensitivity to antimicrobial
peptides, as well as to the antibiotic vancomycin, further supporting the notion that the
external barriers of the cell are more permeable in this mutant (Lamarche et al., 2008).
This study also found that a phosphorylated form of lipid A was less abundant in this
mutant, indicating that the pathway regulating the response to phosphate starvation may
control lipid A modifications in response to starvation in pathogenic E. coli (Lamarche et
al., 2008). Other bacterial species have also been observed to alter membrane
composition in response to phosphate limitation. For example, the ca-proteobacterium
13
Sinorhizobium meliloti replaces its membrane phospholipids with a set of three
phosphate-free lipids during phosphate limitation (Zavaleta-Pastor et al., 2010).
Membrane lipid rearrangements in response to phosphate starvation are not unique to
bacteria. Some species of plants remodel their membranes when phosphate-starved, using
phosphorus-containing lipids as a source of internal phosphate, and replacing them with
the non-phosphorus lipid digalactosyldiacylglycerol (Nakamura, 2013).
The oxidative stress response
Growth under phosphate-limited conditions results in an up-regulation of oxidative stress
genes. In particular, in Sinorhizobium meliloti, Agrobacterium tumefaciens, and
Pseudomonasaeruginosa,increased expression of catalase-encoding genes has been
observed upon phosphate limitation (Yuan et al., 2005). Further in E. coli, ahpCF,which
encodes an alkylhydroperoxide reductase, has been found to be required for protecting
cells in aerobic, phosphate-starved conditions from oxidative stress (Moreau et al., 2001).
Although these findings suggest that phosphate limitation may induce oxidative stress
within a cell, the precise reason for this is unknown. One model is based on the finding
that while growth slows following phosphate limitation in E. coli, bacteria continue to
undergo aerobic respiration. It has been hypothesized that the continued production of
hydrogen peroxide in this process, without its dilution by cell division, leads to hydrogen
peroxide accumulation, and thus increased oxidative stress on the cell (Gerard et al.,
1999; Yang et al., 2012).
14
Polyphosphate
Production of polyphosphate is increased upon phosphate limitation. Polyphosphate is
composed of chains of tens to hundreds of negatively charged orthophosphate groups,
linked by high-energy phosphoanhydride bonds (Figure 1. IA). It is found in both
prokaryotes and higher organisms, and has been reported to have a number of different
functions. Perhaps its most well-defined role is that of a phosphate storage polymer in
starved conditions in bacteria (Rao et al., 2009). While polyphosphate is thought to be
present at low levels in nutrient-replete conditions, in response to stringency,
polyphosphate levels have been found to increase more than 100-fold (Rao et al., 2009).
The phosphate backbone of polyphosphate is complexed with divalent cations such as
Ca, Mg, Mn, Fe, and Co, and may therefore serve as a storage facility for metals as well
as phosphate.
Although polyphosphate levels are known to increase in phosphate-limited conditions, it
is debated whether this increase is under control of the main phosphate response pathway
in bacteria (discussed below). An increase in polyphosphate stores may instead represent
a parallel mechanism to respond to phosphate limitation.
In addition to its role as a storage polymer, polyphosphate has been assigned a number of
other proposed functions. It is thought to contribute to bacterial resistance to harsh
environmental conditions by forming a capsule, which is speculated to benefit the cell by
protecting it or by chelating environmental metals (Tinsley et al., 1993). In this role,
polyphosphate has been found localized to the periplasm, as well as the cytoplasm, in
some bacteria. In Neisseriaand others, 50% of cellular polyphosphate has been found to
be a component of the cell capsule (Kulaev and Kulakovskaya, 2000). Polyphosphate has
been implicated in bacterial competence as well; in this capacity, polyphosphate forms a
15
helical polymer in complex with polyhydroxybutyrate and calcium, and this complex has
been discovered in the membranes of bacteria that have been made competent, leading to
the hypothesis that these complexes make the membrane permeable to DNA (Kulaev and
Kulakovskaya, 2000; Rao et al., 2009). Polyphosphate has been proposed to have
regulatory roles as well; in vitro data suggest it may control the activity of DNA
polymerase and the Lon protease, among other enzymes (Achbergovi, and Nahika,
2011), and in vivo evidence suggests that polyphosphate may inhibit cell cycle
progression in Caulobactercrescentus (Boutte et al., 2012).
Bacterial, cytoplasmic polyphosphate stores are regulated largely by the opposing
activities of polyphosphate kinases (Ppk), which transfer the gamma phosphoryl group of
ATP to the growing polyphosphate chain, and exopolyphosphatases (Ppx), which
hydrolyze a phosphoryl group from the end of this chain (Figure 1.1 B). In some bacteria,
such as E. coli, Ppkl is the primary known polyphosphate kinase, while Pseudomonas
aeruginosaPAO 1, Caulobactercrescentus, and other bacteria encode two polyphosphate
kinases, Ppkl and Ppk2 (Boutte et al., 2012; Rao et al., 2009). Ppkl is known to use ATP
to produce polyphosphate, while Ppk2 has been found to use ATP and GTP equally well
as a substrate (Rao et al., 2009).
In eukaryotes polyphosphate has also been found to be widely abundant, and present in
numerous subcellular compartments. As in prokaryotes, it is thought to function as a
storage compound in eukaryotes, and it has been identified in membranes and in complex
with polyhydroxybutyrate. Regulatory roles for polyphosphate have also been identified
in eukaryotes, most notably in the stimulation of the Ser/Thr protein kinase TOR, which
functions in cell growth and proliferation (Rao et al., 2009).
16
A
0
--
0
0
P-C -P-C -P-0-
I
B
I
O- 0-P-. L
n
I
PPK1I
PPK2
ATP
H0
PO4
PPX11
PPX2
Figure 1.1 - Polyphosphate
(A) Polyphosphate is composed of tens to hundreds of orthophosphate groups linked by
phosphoanhydride bonds. (B) Polyphosphate is synthesized by polyphosphate kinases
(PPKI/PPK2), and degraded by exopolyphosphatases (PPXI/PPX2)
Virulence
In addition to these physiological responses that allow bacteria to adapt to lower
phosphate conditions, some bacteria exploit low phosphate levels as a signal to trigger
lifestyle changes. In particular, for pathogenic bacteria that colonize the gut, phosphate
levels can dictate the decision to become more virulent.
Outside of a host, the phosphate limitation response in a number of pathogenic bacteria
includes the up-regulation of virulence genes, including exopolysaccharide and secretion
genes (Chakraborty et al., 2011; Faure et al., 2013; von KrUger et al., 2006), and a
relationship between host phosphate levels and bacterial virulence has subsequently been
identified in pathogenic bacteria in both infection models and humans. Pseudomonas
aeruginosa,which colonizes the intestine of the microscopic worm C. elegans, has been
found to kill its host when grown on low phosphate medium, but not on high phosphate
17
medium (Zaborin et al., 2009). This trend is not universal; for example, one study in the
avian pathogenic strain Escherichiacoli 078 found that an increase in expression of
phosphate limitation response genes resulted in decreased virulence of the strain
(Bertrand et al., 2010).
Phosphate levels influence bacterial infection of vertebrates as well. In vertebrates,
surgical injury results in the release of products of physiological stress into the gut, which
alters the gut environment; in particular, intestinal phosphate levels are lowered. This
decrease in environmental phosphate is then used by opportunistic pathogens in the gut as
a means to sense host vulnerability, and accordingly to up-regulate expression of
virulence genes (Alverdy et al., 2000; Long et al., 2008). The precise cause of phosphate
depletion after surgical injury is unknown, but it is thought to occur due to the
involvement of phosphate in myocardial performance and arterial pressure (Shor et al.,
2006).
In humans, severe phosphate depletion in septic individuals is a predictor of patient
mortality (Shor et al., 2006), supporting a model in which, counter-intuitively, reduction
of an environmental nutrient necessary for bacterial life actually supports bacterial
success in pathogenesis. The effect of phosphate on virulence has been better studied in
mouse models. In one study, after surgical injury to the livers of mice, P. aeruginosawas
able to colonize their intestines, which resulted in host death. Strikingly, this lethality was
blocked when oral phosphate was administered to the mouse. Pseudomonasisolates
recovered from infected mice that were not administered oral phosphate were found to
have upregulated expression of pstS, which encodes a periplasmic phosphate binding
protein produced in response to low phosphate levels, compared to those that were
18
administered the phosphate, indicating that the virulent isolates were indeed phosphatestarved (Long et al., 2008).
Motile-to-sessile transitions
In addition to making the decision to turn on virulence genes, many bacteria use
phosphate as a signal to trigger other lifestyle changes. A well-studied example is that of
Pseudomonasfluorescens,For this bacterium, high phosphate stimulates free swimming
Pseudomonas to form biofilms, and low phosphate stimulates release from biofilms. This
process is mediated through main phosphate response pathway, PhoR-PhoB (discussed
below), which regulates production of c-di-GMP, a small signaling molecule that has
been implicated in regulation of motile-to-sessile transitions in bacteria (Monds et al.,
2006; Newell et al., 2011). Similar effects of phosphate levels on motile-to-sessile
transitions have been observed in diverse bacteria. Another example is that of
Agrobacterium tumefaciens, in which phosphate limitation has the opposite effect of that
found in Pseudomonas, increasing biofilm formation, again through the PhoR-PhoB
pathway (Danhorn et al., 2004).
Stalk elongation
While some tactics to respond to phosphate limitation are widespread among different
bacteria, others are more specialized. One example is that of the freshwater aproteobacterium Caulobactercrescentus. Phosphate is the most limiting nutrient in
Caulobacter'senvironment, and this bacterium undergoes a distinct, morphological
change in response to phosphate limitation. Caulobacterbears a polar stalk that it
elongates as much as twenty-fold its normal length in response to phosphate limitation
(Gonin et al., 2000). Stalk elongation has been found to occur only in response to a few
19
nutrient perturbations: phosphate limitation or calcium excess (Poindexter, 1984). The
specificity of stalk elongation underscores the importance of the phosphate limitation
response in this bacterium and indicates a possible relationship between these two
nutrients in bacteria.
The precise function of stalk elongation in Caulobacteris unknown. It has been
hypothesized to play a role in increased nutrient uptake during starvation by increasing
cell surface area without proportionally increasing cell body size, which would require
costly protein production. In support of this hypothesis, proteomic studies of the stalk
have found that it is enriched for some, but not all, of the proteins known to be involved
in inorganic phosphate import in Caulobacter,suggesting that phosphate is absorbed in
the stalk and then transported to the cell body to be metabolized (Ireland et al., 2002).
However, more recent evidence has suggested that a protein-mediated diffusion barrier
exists between the stalk and cell body, preventing the exchange of both membrane and
soluble proteins between the two compartments, calling this model into question
(Schlimpert et al., 2012).
Several alternate roles for stalk elongation have been suggested including a function for
the stalk as a nutrient antenna that allows cells to take up phosphate, which, as a small
molecule, may then be able to diffuse through the barrier. Alternative putative roles for
the stalk include a function as a storage compartment for damaged proteins, and as a
mechanism to distance the Caulobactercell body from attachment surfaces, allowing
better nutrient flux around the cell (Baldi and Barral, 2012; Klein et al., 2013).
Although the Caulobacterstalk has been studied for several decades, little is known
about how stalk elongation is regulated. Several studies have aimed to determine whether
20
there are genes involved specifically in stalk elongation, and not in general cell wall
synthesis. Recently, it was found that deletions of PbpX and PbpC, two of six
glycosyltransferases encoded in the Caulobactergenome, resulted in defects in stalk
elongation (Yakhnina and Gitai, 2013), and that one of these, PbpC, localizes primarily to
the stalk (Kuhn et al., 2010; Yakhnina and Gitai, 2013). Thus, although some headway
has been made in identifying the composition of the stalk, the mechanism by which stalk
elongation is regulated remains unknown.
Although stalk elongation is not a widespread response to phosphate limitation, the
Caulobacterstalk is not the only bacterial appendage that has been found to have a
specialized role in responding to phosphate limitation. For example, studies of clinical
isolates of P. aeruginosahave found that low phosphate induces formation of appendagelike structures on their cell surfaces that contain higher concentrations of PstS, the same
periplasmic phosphate-binding protein found in the Caulobacterstalk (Zaborina et al.,
2008).
The phosphate limitation response pathway
Most of the responses discussed above, and many others, are controlled either
directly or indirectly by a signaling pathway that senses and responds to the phosphate
state of the environment. E. coli and many other bacterial species respond to phosphate
limitation through the same conserved pathway. This is composed of the Pst system, an
ABC-type transport system that imports phosphate, and a downstream two-component
signaling pathway, PhoR-PhoB. The Pst system regulates the activity of the PhoR-PhoB
pathway in response to the external phosphate environment, and this pathway in turn
21
executes a transcriptional response to that environment (Figure 1.2), as will be described
below in detail.
In many bacteria, a second, low-affinity transporter, the PitA system, takes up phosphate
in phosphate-abundant environments. However, the primary function of this transporter is
unclear; it has more recently been implicated as a zinc transport system, and may import
phosphate in conjunction with metal ions (Beard et al., 2000; Graham et al., 2009;
Jackson et al., 2008).
A
high phosphate
*
B
low phosphate
0
0
P
Figure 1.2 - The Pst/Pho phosphate limitation response pathway
The Pst system imports phosphate and is thought to regulate activity of the PhoR/B two-component signal transduction
system. (A) In high phosphate conditions, PhoR and PhoB are repressed. (B) In low phosphate conditions, PhoR is
active to autophosphorylate and phosphotransfer to PhoB. Phosphorylated PhoB dimerizes and binds DNA, activating
transcription of a set of genes termed the Pho regulon.
The Pst system: an ABC-type transporter
ATP-binding cassette (ABC) transporters employ ATP hydrolysis to power the transport
of substrates across membranes. This class of transporter is widely conserved, from
bacteria to humans, but appears much more common in lower life forms. In humans,
roughly 50 ABC transporters have been identified, while in E. coli, 80 such reporters
22
have been annotated, comprising 5% of the genome (Rees et al., 2009). Further, although
both ABC importers and exporters exist in prokaryotes, only ABC-type exporters have
been identified in eukaryotes (Rees et al., 2009).
The minimum architecture of an ABC transport system consists of four domains: two
more variable transmembrane domains, which form a channel, and two conserved
cytoplasmic ATPase domains, which power substrate transport. ABC-type importers
contain a fifth domain, a high-affinity binding protein for the system's ligand, which
resides in the periplasm of Gram-negative bacteria, and can be membrane associated or
fused to a transmembrane component in Gram-positive bacteria (van der Heide and
Poolman, 2002). Although this basic architecture remains conserved between different
ABC transporters, their polypeptide arrangement varies. In some cases each domain is
encoded as a separate protein; in others, each transmembrane domain is encoded
separately along with a single ATPase domain; in still others, all four core domains are
found encoded in a single polypeptide (Rees et al., 2009).
The ABC-type Pst system is composed of PstA and PstC, which form the transmembrane
channel, the periplasmic phosphate binding protein, PstS, and two subunits of the ATPase
PstB, which associate with the cytoplasmic portion of PstA and PstC. In E. coli, all four
components are encoded within a single operon. Although the mechanism of action of the
Pst system has not been well studied, it is expected to behave similarly to other, bettercharacterized ABC-type transport systems, such as the bacterial maltose (Mal)
transporter. The Mal system, MalFGK 2, is composed of the transmembrane components
MalF and MaIG, the periplasmic maltose-binding protein, MBP, and two subunits of the
cytoplasmic ATPase, MalK (Chen, 2013).
23
Structural studies of the Mal system have captured it in multiple conformations, allowing
inferences about its molecular mechanism. A crystal structure of MalFGK 2, in the
absence of MBP and nucleotides, shows the resting state conformation (Figure 1.3A) of
the transporter, in which the maltose-binding site is exposed to the cytoplasm, and the
dimerization interface of the two MalK subunits is reduced (Chen, 2013). A structure in
the presence of MBP but in the absence of nucleotides shows the pre-translocation state
(Figure 1.3B), with MBP binding to the periplasmic surface of the transmembrane MalF
and MaIG in a closed conformation. In this state, the two MalK subunits are rotated
towards each other, forming a semi-open dimer. Finally, a third structure, in the presence
of both MBP and nucleotides, shows closure of the MalK dimer, rotation of the
transmembrane subunits, and opening of MBP, forming the outward-facing state (Figure
1.3C) of the transporter. In this state, the substrate is transferred from MBP to the
transmembrane domains, and two ATP molecules are bound by the MalK dimer (Chen,
2013).
A
B
inward-facing
C
pre-translocation
outward-facing
Figure 1.3 - Mechanism of ABC-type transporters
Structural studies of the bacterial maltose transporter have captured it in three different conformations and provided
insights into its molecular mechanism. The transmembrane components are shown in light and dark blue; the ATPase
subunits are shown in orange and red, and the periplasmic binding protein is shown in green. In (A), the transporter is
in the inward-facing conformation. In (B) it is in the pre-translocation state. In (C) it is in the outward facing conformation. (Based on structural studies reviewed in Chen et al., 2013.)
24
In addition to ensuring proper substrate translocation, it is necessary for ABC transporters
to maintain low ATPase activity in the absence of substrate. Structural studies of the
maltose transporter suggest that substrate binding properly positions catalytic residues in
the MalK domains to permit ATPase activity. Binding of MBP, resulting in partial
closure of the MalK dimer in the pre-translocation state, brings catalytic residues from
one MalK subunit in proximity with the nucleotide binding region of the other MalK
subunit. When the outward-facing state of the transporter is achieved, a conserved
LSGGQ motif in one MalK subunit is then better oriented with the nucleotide binding
site of the other subunit. This motif functions to orient the y-phosphate of the ATP of the
ATP molecule, positioning it for hydrolysis (Chen, 2013).
Although the process of substrate translocation has been well-characterized in this and
other ABC transporters, several regulatory aspects of transporter function are less well
understood. In particular, further studies are necessary to understand more generally how
ABC transporters can be regulated by additional factors, and can in turn impose
regulation on downstream proteins.
Additional domains fused either to the transmembrane or ATPase domains have been
found to be involved in the regulation of ABC transporter activity in some systems.
Trans-inhibition has been observed for some systems, in which the intracellular
concentration of the ligand can inhibit the uptake of additional ligand. In the E. coli
methionine ABC importer (MetNI), binding of methionine to the C-terminal domain of
the ATPase subunits has been found to reduce transporter activity (Rees et al., 2009).
Although a similar mechanism could allow the Pst system to regulate the downstream
25
PhoR/PhoB pathway, no such domain has been identified in the Pst systems of E. coli
and other examined bacteria.
PhoR/PhoB: a two-component signaling system
The PhoR/PhoB signal transduction system belongs to the class of two-component
signaling systems, the predominant signaling modality in bacteria (Capra and Laub,
2012; Stock et al., 2000). The Pst system somehow represses activity of the downstream
PhoR/PhoB pathway in phosphate-replete conditions, and permits its activation in
phosphate-limited conditions (Figure 1.2) (Hsieh and Wanner, 2010).
A two-component signal transduction system is composed of a sensor histidine kinase
(e.g. PhoR) and its cognate response regulator (e.g. PhoB) (Figure 1.4A). Between 20 and
200 of these pathways have been identified in almost all sequenced bacteria (Alm et al.,
2006), and they have been found to be responsible for a wide range of processes, from
sensing changes in the extracellular environment to regulating cell cycle progression.
Despite this, little is known about how the majority of them are controlled, and what their
precise output regulons are (Galperin, 2010; Krell et al., 2010).
26
A
B
(1) HK-His + ATP
HK-His-P + ADP
P
P00H
HK
(2) HK-His-P + RR-Asp
(3) RR-Asp-P + H20
HK-His + RR-Asp-P
RR-Asp + Pi
RR
Figure 1.4 - Two-component signal transduction systems
(A) Two-component systems are composed of a sensor histidine kinase (HK), which is often membrane-bound, and
cytoplasmic response regulator (RR). The kinase autophosphorylates on a conserved histidine residue, and phosphotransfers to a conserved aspartate residue on the response regulator. (B) These systems participate in three reactions:
(1) Autophosphorylation of the kinase on a conserved histidine residue; (2) Phosphoryl transfer from the histidine
kinase to a conserved aspartate residue on the response regulator; (3) Dephosphorylation of the response regulator,
which is stimulated by the histidine kinase.
Histidine kinases
The core cytoplasmic portion of a histidine kinase is composed of an N-terminal DHp
(dimerization/histidine-containing phosphotransfer) domain and C-terminal CA (ATPbinding/catalytic) domain. Histidine kinases form dimers, mediated by the DHp domain,
and this interaction is highly specific (Ashenberg et al., 2011). The DHp domain also
contains a conserved histidine residue, which it phosphorylates when activated, and is
also responsible for mediating the interaction with the kinase's cognate response
regulator. The CA domain contains residues necessary to bind ATP/ADP and
magnesium, as well as the catalytic residues required to phosphorylate the DHp domain
(Stock et al., 2000).
Many histidine kinases are membrane bound. For these kinases, the cytoplasmic portion
is linked via, at minimum, a transmembrane domain to a periplasmic or extracellular
domain, which typically binds to an external ligand that regulates the activity of the
27
kinase. Most kinases, both cytoplasmic and transmembrane, contain additional domains
upstream of the cytoplasmic core domains. A HAMP domain is typically found directly
upstream of the DHp domain, which is thought to transmit the signal received by input
domains to the core cytoplasmic portion of the kinase through conformational changes. In
place of, or in addition to, extracytoplasmic signal receptor domains, kinases often
contain additional cytoplasmic domains N-terminal to the HAMP domain, such as PAS
domains, which can act as points of regulation on kinase activity (Krell et al., 2010).
Histidine kinases participate in three biochemical reactions: autophosphorylation,
phosphotransfer, and phosphatase (Figure 1.4B). Once the kinase autophosphorylates on
a conserved histidine residue, it can then transfer that phosphoryl group to the conserved
aspartate residue of its cognate response regulator. When acting as a phosphatase, the
histidine kinase again interacts with its cognate response regulator, likely through an
overlapping binding interface, and stimulates hydrolysis of the regulator's phosphoryl
group. Although the kinase contains the catalytic residues necessary for the
autophosphorylation reaction, the regulator contains the residues sufficient for hydrolysis
of its Asp~P bond. In the absence of kinase, phosphorylated response regulators are
capable of autodephosphorylation, while interaction with the kinase stimulates this
reaction (Stock et al., 2000).
Ligand binding to an input domain is thought to stimulate either autophosphorylation or
phosphatase activity by altering the conformation of the DHp and CA domains. A variety
of histidine kinase input domains have been identified. Of the 14 types of sensor domains
currently characterized, PAS domains are the most abundant, and can be found in both
the cytoplasmic and extracytoplasmic regions of the kinase (Krell et al., 2010). These
28
domains have poor conservation at the sequence level, but form a conserved a/n fold.
PAS domains have been found to receive signals through several mechanisms, both by
directly binding ligand, and by binding ligand through a cofactor (Krell et al., 2010).
Response regulators
Response regulators typically have a less complex domain architecture than their
histidine kinase counterparts. The core domains of a response regulator are the conserved
N-terminal receiver domain, which receives a phosphoryl group from the kinase on a
conserved aspartate residue, and a more variable C-terminal effector domain, which
enacts an output in response to phosphorylation. Typically, the response regulator
effector domain is a DNA-binding domain, and enacts a transcriptional output. However,
a variety of alternate effector domains have been identified in bacteria, including some
that function as enzymes such as phosphodiesterases or methylesterases (Galperin, 2010).
There is diversity within the set of DNA-binding effector domains as well; annotated
DNA binding domains on response regulators include winged helix domains (the
OmpR/PhoB family), helix-turn-helix domains (the NarL/FixJ family), and several others
(Galperin, 2010).
Most response regulators are thought to dimerize or form multimers upon
phosphorylation, which is the active form of the regulator. As with histidine kinases, in
vitro work has found that in almost all studied cases, this dimerization is highly specific
(Gao et al., 2008).
Two-component specificity
Although tens or hundreds of these histidine kinase/response regulator pairs can exist in a
single bacterial cell, a histidine kinase recognizes its cognate response regulator with
29
exquisite specificity (Grimshaw et al., 1998; Skerker et al., 2005). In contrast to
eukaryotic signaling pathways, many of which require the aid of subcellular localization
and scaffolding proteins to ensure specificity between systems, two-component
specificity is mediated by molecular recognition alone.
The interface responsible for ensuring kinase-regulator specificity has been identified,
and found to include the c-helical DHp domain of the kinase, cc-helix 1 of the response
regulator, and a loop region in each protein (Skerker et al., 2008).
Hybrid histidine kinases
In addition to canonical histidine kinase/response regulator pairs, more complex bacterial
phosphorelays can have three or more components. Most frequently, these pathways are
composed of a hybrid histidine kinase, which contains both the histidine kinase DHp and
CA domains, as well as a response regulator receiver domain. The hybrid kinase is
capable of autophosphorylating and transferring this phosphoryl group to its own receiver
domain. A second protein, called an HPt (histidine-containing phosphotransfer), then
receives the phosphoryl group from this receiver domain, and transfers it to a third
protein, a response regulator, which is then activated to enact the downstream output of
the pathway (Capra and Laub, 2012).
Regulation of two-component systems
In addition to the regulation of histidine kinase autophosphorylation through signal
binding to a receptor domain, several two-component signal transduction systems have
been found to be subject to further regulation. Only a few examples of this type of
regulation have been well characterized; in most, it is the histidine kinase, rather than the
response regulator, that is the site of signal integration. Within the class of proteins that
30
have been found to regulate histidine kinase activity, regulators that interact with the
DHp, CA, and transmembrane domains have all been identified. Examples of each of
these types of regulation are discussed below.
The first characterized example of regulation of a two-component system by an
additional factor was that of the histidine kinase NtrB and its regulation by the small
protein P11. NtrB (also called NRII), controls the transcriptional response to nitrogen in .
coli and other bacteria, by phosphorylating and activating its cognate response regulator
NtrC (NRI). PII activity is regulated both by ca-ketoglutarate and glutamine levels, which
transmit information about the carbon and nitrogen states of the cell, respectively (Figure
1.5), allowing NtrB to sense and respond to both signals (Ninfa and Jiang, 2005). When
cellular glutamine concentrations are high, a uridylyltransferase/uridylyl-removing
enzyme (UTase/UR) catalyzes the removal of a UMP group from P11; in low glutamine
concentrations, it catalyzes the uridylylation of P11. Unmodified PIH is able to bind to
NtrB and promote its phosphatase activity (Ninfa and Jiang, 2005). The function of
unmodified PH1 is repressed by high a-ketoglutarate levels; thus, PII acts as an AND
NOT gate, requiring glutamine and not a-ketoglutarate (Figure 1.5). PII activation of
NtrB phosphatase activity is thought to occur through binding to the NtrB CA domain
(Pioszak and Ninfa, 2003a, b). Several residues in the CA domain have been identified
that influence PII binding to NtrB; however, the precise binding site has not yet been
mapped.
31
glutamine
UTase
Fill
PIl-UMP
UR
a-ketoglutarate -f
glutamine
HK
NtrB -P
RR
P
---
NtrB
P
nitrogen response genes
Figure 1.5 - Regulation of NtrB by glutamine and a-ketoglutarate through PH1
UTase/UR: uridylyltransferase/urdyly-removing enzyme. Glutamine regulates the activity of the UTase/UR enzyme,
which can add or remove a UMP group from the regulatory protein P11. When PH1 is not modified by UMP, it binds to the
histidine kinase NtrB and promotes its phosphatase activity for the response regulator NtrC. This process inhibited by
a-ketoglutarate. When phosphorylated, NtrC promotes transcription of nitrogen response genes.
SipA, a small (81 amino acid) protein found in cyanobacteria has been shown to regulate
the activity of the histidine kinase NblS in several cyanobacterial species (Sakayori et al.,
2009). The interaction between SipA and NblS was discovered in a yeast two-hybrid
screen of Synechococcus libraries using a portion of NblS as bait (Espinosa et al., 2006).
Additional yeast two-hybrid studies and in vitro binding assays have indicated that,
similar to PII, SipA acts by binding to the CA domain of NbS (Salinas et al., 2007).
Other regulators of histidine kinases act by binding to the DHp domain. Two examples of
this are the inhibitors Sda and KipI, both of which regulate the sporulation cascade kinase
32
KinA in Bacillus subtilis through an interaction with its DHp domain (Cunningham and
Burkholder, 2008).
Another example of regulation of a histidine kinase through its DHp domain is that of
regulation of the oxygen-responsive two-component system FixL/J by FixT. FixT is
thought to inhibit the kinase FixL by acting as a competitive inhibitor. FixT has been
shown to contain the residues necessary to receive a phosphoryl group from a histidine
kinase, and is proposed to inhibit this pathway by competing with FixJ for phosphoryl
groups from FixL (Krell et al., 2010).
In addition to regulators that act by binding to the cytoplasmic domains of the kinase,
those that control kinase activity through interaction with its transmembrane regions have
also been identified. The magnesium-responsive two-component signaling pathway,
PhoQ/PhoP, is regulated by the 47 amino acid peptide MgrB in Salmonella typhimuium
and other bacterial species (Lippa and Goulian, 2009). MgrB regulates the activity of the
histidine kinase PhoQ by interacting with its transmembrane domain (Lippa and Goulian,
2009).
The general stress response in bacteria
Although the conserved Pst/Pho pathway responds specifically to phosphate limitation,
many bacteria are also thought to encode a general stress response that can be activated
by a wide range of stressors.
33
A
RssB active
RssB inactive
phosphate
starvation
magnesium
starvation
RssB
RssB
RssB
RssB
ClipX/P
6general
B
stress response
Starvation/stress
RsiB1
RsiB2
anti-anti-sigma factors
RsiAl
RsiA2
anti-sigma factors
RpoE2
general stress response
Figure 1.6 - The general stress response
Diverse groups of bacteria have different mechanisms of activating a general stress response. (A) In gammaproteobacteria, the general stress response is activated by cp (RpoS). Adapted from Bougdour et al., 2006.
(B) In the alpha-proteobacterium Sinorhizobium meliloti the general stress response is under control of
RpoE2. Adapted from Bastiat et al., 2010.
E. coli: RpoS and the general stress response
In E. coli and other gamma-proteobacteria, a single a factor, rpoS, serves as a point of
integration of multiple stress signals to enact the general stress response (Figure 1.6A).
RpoS is regulated in response to phosphate starvation, magnesium starvation, DNA
damage, and other stresses via control of its proteolysis (Bougdour et al., 2006; Merrikh
et al., 2009). A response regulator, RssB, is known to promote degradation of RpoS by
the protease ClpXP. RssB in turn is regulated by a suite of anti-adaptor proteins. In
response to a particular type of stress, a specific anti-adaptor is activated to bind RssB
34
and inhibit it from mediating degradation of RpoS. IraP is the anti-adaptor responsible for
inhibiting RssB in response to phosphate starvation (Bougdour et al., 2006).
a-proteobacteria: ECF a factors
No homolog of rpoS has been identified in C. crescentus, S. meliloti, and other aproteobacteria, leaving open the question of how these bacteria respond to stress, and
whether, as in E. coli, there is a central protein that coordinates the response to multiple
stressors. Work in S. meliloti has indicated that this is indeed the case. An extracytoplasmic (ECF) o factor, RpoE2, has been found to regulate a general stress response
in this bacterium (Figure 1.6B). RpoE2 is regulated by two anti-o factors, which are in
turn be regulated by anti-anti-a factors. These additional levels of regulation are
proposed to integrate multiple signals into a single ECF sigma factor, although the
p
Ireis 'Ieanism1
uy VVIich vatIous
stressrs iaU
Le
J LUiCLLJi1 VI L111 iLp
VVtay 1s
unknown (Bastiat et al., 2010). A similar pattern has been observed in C. crescentus.
Here, however, although Caulobacterencodes an rpoE2 homolog, the general stress
response in this bacterium is regulated by a different ECF c factor, o , which, similarly
to RpoE2, is regulated by the anti-o factor NepR, and the anti-anti-o factor PhyR
(Foreman et al., 2012; Lourenco et al., 2011).
35
PhoU and PhoU-like domains
PhoU: a putative regulator of the PhoR/PhoB pathway
In most bacteria, PhoR does not have a large extracytoplasmic domain, leaving open the
question of how it senses external phosphate conditions. The gene phoU is widely
conserved in bacteria and often found co-operonic with the pst system and phoR/phoB
genes; thus, it has been proposed to act as a regulator of the phoR/phoB system. In this
model, in phosphate-replete conditions, the Pst system activates PhoU to repress PhoR.
When transport through the Pst system slows, this repression of PhoR is relieved (Figure
1.7C).
phoU was first designated phoT and cloned in (Amemura et al., 1982) and subsequently
renamed. Structural studies of the PhoU protein in Aquifex aeolicus (Figure 1.7A) and
Thermatoga maritima (Figure 1.7B) have revealed two repeats of a three-helix bundle
that together form six-helix bundle structure. One three-helix bundle is considered a
"PhoU domain" (Liu et al., 2005; Oganesyan et al., 2005). These crystal structures
formed a new class of domains not previously found in the Protein Data Bank; the
structure of Bag domains, which belong to a class of eukaryotic chaperone proteins, has
been noted as the structure most similar to that of PhoU (Oganesyan et al., 2005). I
discuss these domains in greater detail below.
These studies also revealed two conserved patches of aspartate and glutamate residues on
the surface of PhoU, implicating PhoU in metal binding. Two histidine residues proximal
36
to these patches suggest that PhoU may favor binding of Zn and Fe, rather than Mn or
Mg (Oganesyan et al., 2005). In the Thematoga maritima structure, PhoU was
crystallized with multinuclear iron clusters, supporting this notion. Iron was not present
in the purification or crystallization buffers, indicating that PhoU was purified in an ironbound form (Liu et al., 2005).
Although PhoU has been proposed to act as a signaling intermediate between the Pst and
Pho systems (Hsieh and Wanner, 2010; Steed and Wanner, 1993), evidence for this
model has been inconclusive. If PhoU does act to repress the PhoR/PhoB pathway,
mutation ofpho U should result in de-repression of Pho regulon genes. In accordance with
this model, expression of alkaline phosphatase, a marker of the Pho regulon, is induced in
phoUmutants of E. coli (Muda et al., 1992; Surin et al., 1985). Further, genome-wide
expression profiling of aphoU mutant in the E. coli W3 110 strain showed increased
expression of a number of other genes proposed to function in phosphate limitation (Li
and Zhang, 2007). Notably, however, this change in gene expression was not compared
to that in a pst mutant, in which the Pho regulon is known to be upregulated.
Characterization of a pst mutant would allow identification of Pho regulon genes in the
W3 110 strain, which would allow determination of the effect of the pho U mutation on
Pho regulon genes. Although phosphate-related genes are upregulated in the pho U
mutant, these genes may not encompass the entire Pho regulon.
Despite evidence in support of this model, a direct interaction between PhoU and the
Pst/Pho systems has not been conclusively identified. A yeast two-hybrid study identified
an interaction between PhoU and PhoB, as well as an interaction between PhoU and the
ferric uptake regulatory protein Fur (Chakraborty et al., 2011). The same study did not
37
identify an interaction between PhoR and PhoB, possibly due to the transient nature of
this interaction. Further, it did not find a strong interaction between PhoR and itself, or
between PhoB and itself (Chakraborty et al., 2011), although these proteins are known to
form dimers. A second study, which used FRET to examine interactions between PhoU
and the PhoR/PhoB system identified an interaction between PhoR and PhoB, but not
between PhoU and either of these proteins (Baek et al., 2007). These conflicting results,
and the lack of evidence of a direct interaction between the Pst system and any of the
PhoR/B/U components, have left open the question of what the relationship of PhoU is to
the Pst/Pho pathway.
A
B
Aquifex aeolicus
Thermatoga maritima
.00,
Figure 1.7 - PhoU is a putative regulator of the PhoR/PhoB pathway
(A),(B) Crystal structures of PhoU purified from Aquifex aeolicus and Thermatoga maritima. (C) Proposed
model for PhoU function. The Pst system imports phosphate and is thought to act through PhoU to repress
PhoR.
38
In addition to pho U homologs encoded with pst/pho genes, several species, including
Thermatoga maritima and Mycobacterium tuberculosis, contain multiple pho U homologs
(Liu et al., 2005; Shi and Zhang, 2010), suggesting that the function of PhoU-like
domains can be diversified. Additional PhoU domains have been identified in bacteria, as
well as in archaea, as part of multi-domain proteins. Finally, PhoU has been proposed to
share structural similarity with Bag domains, which act as chaperone cofactors in
eukaryotes. Below I discuss each of these classes of PhoU-related proteins.
YjbB: a transporter-fused PhoU domain protein in Escherichia coli
yjbB is an E. coli gene encoding an annotated N-terminal Na+/Pi cotransporter, and Cterminal PhoU domain (Motomura et al., 2011). Although this gene has not been well
studied, it appears to influence intracellular phosphate levels in E. coli. Mutation ofphoU
in E. coli has been found to result in a 1000-fold increase in intracellular polyphosphate
levels (Morohoshi et al., 2002), further discussed in Chapters 2 and 3. Overproduction of
YjbB in this mutant background results in reduction of polyphosphate to near wild-type
levels. This reduction is thought to occur due to increased phosphate export through the
YjbB transporter domain. In support of this hypothesis, overproduction of YjbB has been
found to increase the rate of phosphate export (Motomura et al., 2011).
Archaeal PhoU homologs
PhoU homologs have been identified in archaea, although few have been characterized. A
PhoU homolog in the methanogenic archaeon Methanococcus maripaludisJJ has been
identified in an operon encoding a predicted nitrate/sulfonate/bicarbonate ABC
transporter. Mutation of this gene results in an increased lag phase during growth on
39
formate, indicating that the substrate of this transporter may be a component needed for
growth on formate (Sattler et al., 2013).
In addition, a class of archaeal proteins has been identified that contains a C-terminal
PhoU-like domain and an N-terminal AbrB-like domain, a putative DNA-binding
domain, separated by a central domain of unknown function. Intriguingly, these proteins
are often found in operons encoding pst system homologs. In some of these archaeal
species, no phoR/phoB homologs have been identified, suggesting that in these PhoU-like
proteins, the PhoU domain might act to regulate the phosphate limitation response
through control of the transcription factor activity encoded in the AbrB domain (Coles et
al., 2005).
Eukaryotic chaperone cofactors
Although PhoU-like domains belong to no previously known structural family, a
crystallographic study of a PhoU homolog from Aquifex aeolicus noted a structural
similarity between PhoU and the Bcl2-associated athanogene (Bag) domain, which is a
cofactor for the eukaryotic chaperone Hsp70 family (Oganesyan et al., 2005). The Bag
domain identified as similar to PhoU was found to share approximately 50% sequence
similarity with each of the two domains of this PhoU homolog, and an RMSD of
approximately 2.9 A was found when the Bag domain was overlaid with either of the two
PhoU domains. These findings led the authors to propose a possible role for PhoU as a
scaffold in the Pst/Pho signaling system, as the PhoR CA domain bears a resemblance to
the ATPase domain of Hsp70 (Oganesyan et al., 2005).
40
Global characterization of the Pho regulon in Caulobacter
crescentus
Although the response to phosphate limitation has been studied in many bacteria using
microarray analysis and reporter genes, little work has been done to differentiate between
direct and indirect targets of PhoB, and to determine its activity in phosphate-replete as
well as starved conditions. Further, although PhoU has long been hypothesized to act as a
negative regulator of the PhoR/PhoB system, evidence in support of this model remains
inconclusive.
In this work, I have aimed to understand how the freshwater a-proteobacterium
Caulobactercrescentus responds to phosphate limitation. Caulobacteris an excellent
model in which to understand these questions, because in addition to responding to
phosphate through the widely conserved Pst/Pho pathway, it has a specific,
morphological response to phosphate limitation, elongating its stalk as much as twentyfold its normal length. I have used ChIP-Seq on the response regulator PhoB to identify
the set of genes regulated in response to phosphate limitation in Caulobacter.Further,I
have tested the proposed function of PhoU in the phosphate limitation response. I have
found that PhoU does not regulate the Pho regulon through the PhoR/PhoB pathway as
has been proposed, but that it functions elsewhere in cellular phosphate metabolism.
41
References
Achbergovi, L., and J. Nahalka, 2011, Polyphosphate - an ancient energy source and
active metabolic regulator: Microbial Cell Factories, v. 10, p. 14.
Alm, E., K. Huang, and A. Arkin, 2006, The evolution of two-component systems in
bacteria reveals different strategies for niche adaptation: PLoS Computational
biology, v. 2, p. 1.
Alverdy, J., C. Holbrook, F. Rocha, L. Seiden, R. Licheng, Wu, M. Musch, E. Chang, D.
Ohman, and S. Suh, 2000, Gut-derived sepsis occurs when the right pathogen
with the right virulence genes meets the right host: evidence for in vivo virulence
expression in Pseudomonas aeruginosa:Annals of Surgery, v. 232, p. 10.
Amemura, M., H. Shinagawa, K. Makino, N. Otsuji, and A. Nakata, 1982, Cloning of and
complementation tests with alkaline phosphatase regulatory genes (phoS and
phoT) of Escherichiacoli: Journal of Bacteriology, v. 152, p. 10.
Ashenberg, 0., K. Rozen-Gagnon, M. T. Laub, and A. E. Keating, 2011, Determinants of
homodimerization specificity in histidine kinases: Journal of Molecular Biology,
v. 413, p. 14.
Baek, J. H., Y. J. Kang, and S. Y. Lee, 2007, Transcript and protein level analyses of the
interactions among PhoB, PhoR, PhoU and CreC in response to phosphate
starvation in Escherichiacoli: FEMS Microbiology Letters, v. 277, p. 254-259.
Baldi, S., and Y. Barral, 2012, Bacterial border fence: Cell, v. 151, p. 2.
Bastiat, B., L. Sauviac, and C. Bruand, 2010, Dual control of Sinorhizobium meliloti
RpoE2 sigma factor activity by two PhyR-type two-component response
regulators: Journal of Bacteriology, v. 192, p. 11.
Beard, S. J., R. Hashim, G. Wu, M. R. B. Binet, M. N. Hughes, and R. K. Poole, 2000,
Evidence for the transport of zinc(II) ions via the Pit inorganic phosphate
transport system in Escherichiacoli: FEMS Microbiology Letters, v. 184, p. 5.
Bertrand, N., S. Houle, G. LeBihan, E. Poirier, C. M. Dozois, and J. Harel, 2010,
Increased Pho regulon activation correlates with decreased virulence of an avian
pathogenic Escherichiacoli 078 strain: Infect Immun, v. 78, p. 5324-5331.
Blank, L. M., 2012, The cell and P: from cellular function to biotechnological
application: Current Opinion in Biotechnology, v. 23, p. 6.
Bougdour, A., S. Wickner, and S. Gottesman, 2006, Modulating RssB activity: IraP, a
novel regulator of Ts stability in Escherichiacoli: Genes & Development, v. 20, p.
14.
Boutte, C. C., J. T. Henry, and S. Crosson, 2012, ppGpp and polyphosphate modulate cell
cycle progression in Caulobactercrescentus: Journal of Bacteriology, v. 194, p. 8.
Brickman, E., and J. Beckwith, 1975, Analysis of the regulation of Escherichiacoli
alkaline phosphatase synthesis using deletions and phi-80 transducing phages:
Journal of Molecular Biology, v. 96, p. 10.
Capra, E. J., and M. T. Laub, 2012, Evolution of two-component signal transduction
systems: Annual Review of Microbiology, v. 66, p. 23.
Chakraborty, S., J. Sivaraman, K. Y. Leung, and Y.-K. Mok, 2011, Two-component
PhoB-PhoR regulatory system and ferric uptake regulator sense phosphate and
iron to control virulence genes in type III and VI secretion systems of
Edwardsiellatarda: Journal of Biological Chemistry, v. 286, p. 13.
42
Chen, J., 2013, Molecular mechanism of the Escherichiacoli maltose transporter:
Current Opinion in Structural Biology, v. 23, p. 7.
Coles, M., S. Djuranovic, J. Soding, T. Frickey, K. Koretke, V. Truffault, J. Martin, and
A. N. Lupas, 2005, AbrB-like transcription factors assume a swapped hairpin fold
that is evolutionarily related to double-psi P barrels: Structure, v. 13, p. 10.
Cunningham, K. A., and W. F. Burkholder, 2008, The histidine kinase inhibitor Sda
binds near the site of autophosphorylation and may sterically hinder
autophosphorylation and phosphotransfer to SpoOF: Molecular Microbiology, v.
71, p. 19.
Danhorn, T., M. Hentzer, M. Givskov, M. R. Parsek, and C. Fuqua, 2004, Phosphorus
limitation enhances biofilm formation of the plant pathogen Agrobacterium
tumefaciens through the PhoR-PhoB regulatory system: Journal of Bacteriology,
v. 186, p. 10.
Dyrman, S. T., P. D. Chappell, S. T. Haley, J. W. Moffett, E. D. Orchard, J. B.
Waterbury, and E. A. Webb, 2006, Phosphonate utilization by the globally
important marine diazotroph Trichodesmium: Nature, v. 439, p. 4.
Eder, S., L. Shi, K. Jensen, K. Yamane, and F. M. Hulett, 1996, A Bacillus subtilis
secreted phosphodiesterase/alkaline phosphatase is the product of a Pho regulon
gene, phoD: Microbiology, v. 142, p. 7.
Espinosa, J., I. Fuentes, S. Burillo, F. Rodriguez-Mateos, and A. Contreras, 2006, SipA, a
novel type of protein from Synechococcus sp. PCC 7942, binds to the kinase
domain of NblS: FEMS Microbiology Letters, v. 254, p. 7.
Faure, L., M. Llamas, K. Bastiaansen, S. de Bentzmann, and S. Bigot, 2013, Phosphate
starvation relayed by PhoB activates the expression of Pseudomonas aeruginosa
sigma-vrel ECF factor and its target genes: Microbiology.
Foreman, R., A. Fiebig, and S. Crosson, 2012, The LovK-LovR two-component system is
a regulator of the general stress pathway in Caulobactercrescentus: Journal of
Bacteriology, v. 194, p. 12.
Galperin, M. Y., 2010, Diversity of structure and function of response regulator output
domains: Current Opinion in Microbiology, v. 13, p. 10.
Gao, R., Y. Tao, and A. M. Stock, 2008, System-level mapping of Escherichiacoli
response regulator dimerization with FRET hybrids: Molecular Microbiology, v.
69, p. 15.
Graham, A. I., S. Hunt, S. L. Stokes, N. Bramall, J. Bunch, A. G. Cox, C. W. McLeod,
and R. K. Poole, 2009, Severe zinc depletion of Escherichiacoli: roles for high
affinity zinc binding by ZinT, zinc transport and zinc-independent proteins: The
Journal of Biological Chemistry, v. 284, p. 13.
Grimshaw, C. E., S. Huang, C. G. Hanstein, M. A. Strauch, D. Burbulys, L. Wang, J. A.
Hoch, and J. M. Whiteley, 1998, Synergistic kinetic interactions between
components of the phosphorelay controlling sporulation in Bacillus subtilis:
Biochemistry, v. 37, p. 11.
Gerard, F., A.-M. Dri, and P. L. Moreau, 1999, Role of Escherichiacoli RpoS, LexA and
H-NS global regulators in metabolism and survival under aerobic, phosphatestarved conditions: Microbiology, v. 145, p. 16.
Hsieh, Y. J., and B. L. Wanner, 2010, Global regulation by the seven-component Pi
signaling system: Curr Opin Microbiol, v. 13, p. 198-203.
43
Ireland, M. M. E., J. A. Karty, E. M. Quardokus, J. P. Reilly, and Y. V. Brun, 2002,
Proteomic analysis of the Caulobactercrescentus stalk indicates competence for
nutrient uptake: Molecular Microbiology, v. 45, p. 13.
Jackson, R. J., M. R. B. Binet, L. J. Lee, R. Ma, A. 1. Graham, C. W. McLeod, and R. K.
Poole, 2008, Expression of the PitA phosphate/metal transporter of Escherichia
coli is responsive to zinc and inorganic phosphate levels: FEMS Microbiology
Letters, v. 289, p. 6.
Kageyama, H., K. Tripathi, A. K. Rai, S. Cha-um, R. Waditee-Sirisattha, and T. Takabe,
2011, An alkaline phosphatase/phosphodiesterase, PhoD, induced by salt stress
and secreted out of the cells of Aphanothece halophytica,a halotolerant
cyanobacterium: Applied Environmental Microbiology, v. 77, p. 6.
Kamat, S. S., and F. M. Raushel, 2013, The enzymatic conversion of phosphonates to
phosphate by bacteria: Current Opinion in Chemical Biology, v. 17, p. 8.
Kathuria, S., and A. C. Martiny, 2011, Prevalence of a calcium-based alkaline
phosphatase associated with the marine cyanobacterium Prochlorococcusand
other ocean bacteria: Environmental Microbiology, v. 13, p. 10.
Klein, E. A., S. Schlimpert, V. Hughes, Y. V. Brun, M. Thanbichler, and Z. Gitai, 2013,
Physiological role of stalk lengthening in Caulobactercrescentus:
Communicative & Integrative Biology, v. 6, p. 1.
Krell, T., J. Lacal, A. Busch, H. Siva-Jimenez, M.-E. Guazzaroni, and J. L. Ramos, 2010,
Bacterial sensor kinases: diversity in the recognition of environmental signals:
Annual Review of Microbiology, v. 64, p. 21.
Kuhn, J., A. Briegel, E. Morschel, J. Kahnt, K. Leser, S. Wick, G. J. Jensen, and M.
Thanbichler, 2010, Bactofilins, a ubiquitous class of cytoskeletal proteins
mediating polar localization of a cell wall synthase in Caulobactercrescents: The
EMBO Journal, v. 29, p. 13.
Kulaev, I., and T. Kulakovskaya, 2000, Polyphosphate and phosphate pump: Annual
Reviews of Microbiology, v. 54, p. 26.
Lamarche, M. G., and J. Harel, 2009, Membrane homeostasis requires intact pst in
extraintestinal pathogenic Escherichiacoli: Curr Microbiol v. 60, p. 356-359.
Lamarche, M. G., S.-H. Kim, S. Crdpin, M. Mourez, N. Bertrand, R. E. Bishop, J. D.
Dubreuil, and J. Harel, 2008, Modulation of hexa-acyl pyrophosphate lipid A
population under Escherichiacoli phosphate (Pho) regulon activation: Journal of
Bacteriology, v. 190, p. 9.
Li, Y., and Y. Zhang, 2007, PhoU Is a Persistence Switch Involved in Persister Formation
and Tolerance to Multiple Antibiotics and Stresses in Escherichiacoli:
Antimicrobial Agents and Chemotherapy, v. 51, p. 8.
Lippa, A. M., and M. Goulian, 2009, Feedback inhibition in the PhoQ/PhoP signaling
system by a membrane peptide: PLoS Genetics, v. 5.
Liu, J., Y. Lou, H. Yokota, P. D. Adams, R. Kim, and S. H. Kim, 2005, Crystal structure
of a PhoU protein homologue: a new class of metalloprotein containing
multinuclear iron clusters: Journal of Biological Chemistry, v. 280, p. 1596015966.
Long, J., 0. Zaborina, C. Holbrook, A. Zaborin, and J. Alverdy, 2008, Depletion of
intestinal phosphate after operative injury activates the virulence of P. aeruginosa
causing lethal gut-derived sepsis: Surgery, v. 144, p. 9.
44
Lourenco, R. F., C. Kohler, and S. L. Gomes, 2011, A two-component system, an antisigma factor and two paralogous ECF sigma factors are involved in the control of
general stress response in Caulobactercrescentus: Molecular Microbiology, v.
80, p. 15.
Merrikh, H., A. E. Ferrazzoli, and S. T. Lovett, 2009, Growth phase and (p)ppGpp
control of IraD, a regulator of RpoS stability, in Escherichia coli: Journal of
Bacteriology, v. 191, p. 11.
Metcalf, W. W., P. M. Steed, and B. L. Wanner, 1990, Identification of phosphate
starvation-inducible genes in Escherichiacoli K-12 by DNA sequence analysis of
psi:.:lacZ(Mu d]) transcriptional fusions: Journal of Bacteriology, v. 172, p. 31913200.
Metcalf, W. W., and B. L. Wanner, 1993, Evidence for a fourteen-gene, phnC to
phnP locus for phosphonate metabolism in Escherichiacoli: Gene, v. 129, p. 6.
Monds, R. D., P. D. Newell, R. H. Gross, and G. A. O'Toole, 2006, Phosphate-dependent
modulation of c-do-GMP levels regulates PseudomonasfluorescensPfo-1 biofilm
formation by controlling secretion of the adhesion LapA: Molecular
Microbiology, v. 63, p. 24.
Moreau, P. L., F. Gerard, N. W. Lutz, and P. Cozzone, 2001, Non-growing Escherichia
coli cells starved for glucose or phosphate use different mechanisms to survive
oxidative stress: Molecular Microbiology, v. 39, p. 13.
Morohoshi, T., T. Maruo, Y. Shirai, J. Kato, T. Ikeda, N. Takiguchi, H. Ohtake, and A.
Kuroda, 2002, Accumulation of inorganic polyphosphate inphoU mutants of
Escherichiacoli and Synechocystis sp. strain PCC6803: Applied and
Environmental Microbiology, v. 68, p. 4.
Motomura, K., R. Hirota, N. Ohnaka, M. Okada, T. Ikeda, T. Morohoshi, H. Ohtake, and
A. Kuroda, 2011, Overproduction of YjbB reduces the level of polyphosphate in
Escherichiacoli: a hypothetical role of YjbB in phosphate export and
polyphosphate accumulation: FEMS Microbiology Letters, v. 320, p. 8.
Muda, M., N. N. Rao, and A. Torriani, 1992, Role of PhoU in phosphate import and
alkaline phosphatase regulation: Journal of Bacteriology, v. 174, p. 8.
Nakamura, Y., 2013, Phosphate starvation and membrane lipid remodeling in seed plants:
Progress in Lipid Research, v. 52, p. 8.
Newell, P. D., C. D. Boyd, H. Sondermann, and G. A. O'Toole, 2011, A c-di-GMP
effector system controls cell adhesion by inside-out signaling and surface protein
cleavage: PLoS Biology, v. 9.
Ninfa, A. J., and P. Jiang, 2005, P1H signal transduction proteins: sensors of alphaketoglutarate that regulate nitrogen metabolism: Current Opinion in
Microbiology, v. 8, p. 6.
Oganesyan, V., N. Oganesyan, P. D. Adams, J. Jancarik, H. A. Yokota, R. Kim, and S. H.
Kim, 2005, Crystal structure of the "PhoU-like" phosphate uptake regulator from
Aquifex aeolicus: Journal of Bacteriology, v. 187, p. 4238-4244.
Ohtake, H., J. Kato, A. Kuroda, H. Wu, and T. Ikeda, 1998, Regulation of bacterial
phosphate taxis and polyphosphate accumulation in response to phosphate
starvation stress: Journal of Biosciences, v. 23, p. 9.
45
Pioszak, A. A., and A. J. Ninfa, 2003a, Genetic and biochemical analysis of phosphatase
activity of Escherichiacoli NRII (NtrB) and its regulation by the P1I signal
transduction protein: Journal of Bacteriology, v. 185, p. 17.
Pioszak, A. A., and A. J. Ninfa, 2003b, Mechanism of the P11-activated phosphatase
activity of Escherichiacoli NRII (NtrB): how the different domains of NRII
collaborate to act as a phosphatase: Biochemistry, v. 42, p. 15.
Poindexter, J. S., 1984, The role of calcium in stalk development and in phosphate
acquisition in Caulobactercrescentus: Archives of Microbiology, v. 138, p. 13.
Rao, N. N., M. R. G6mez-Garcia, and A. Kornberg, 2009, Inorganic polyphosphate:
essential for growth and survival: Annual Review of Biochemistry, v. 78, p. 43.
Rees, D. C., E. Johnson, and 0. Lewinson, 2009, ABC transporters: the power to change:
Nature Reviews Molecular Cell Biology, v. 10, p. 10.
Sakayori, T., Y. Shiraiwa, and 1. Suzuki, 2009, A Synechocystis homolog of SipA
protein, Ss1345 1, enhances the activity of the histidine kinase Hik33: Plant & Cell
Physiology, v. 50, p. 10.
Salinas, P., D. Ruiz, R. Cantos, M. L. Lopez-Redondo, A. Marina, and A. Contreras,
2007, The regulatory factor SipA provides a link between NblS and NbIR signal
transduction pathways in the cyanobacterium Synechococcus sp. PCC 7942:
Molecular Microbiology, v. 66, p. 13.
Sattler, C., S. Wolf, J. Fersch, S. Goetz, and M. Rother, 2013, Random mutagenesis
identifies factors involved in formate-dependent growth of the methanogenic
archaeonMethanococcus maripaludis:Molecular Genetics and Genomics, v. 288,
p. 12.
Schlimpert, S., E. A. Klein, A. Briegel, V. Hughes, J. Kahnt, K. Bolte, U. G. Maier, Y. V.
Brun, G. J. Jensen, Z. Gitai, and M. Thanbichler, 2012, General protein diffusion
barriers create compartments within bacterial cells: Cell, v. 151, p. 13.
Sebastian, M., and J. W. Ammerman, 2009, The alkaline phosphatase PhoX is more
widely distributed in marine bacteria than the classical PhoA: The ISME Journal,
v.3,p. 10.
Shi, W., and Y. Zhang, 2010, PhoY2 but not PhoYl is the PhoU homologue involved in
persisters in Mycobacterium turbuculosis:J Antimicrob Chemother, v. 65, p.
1237-1242.
Shor, R., A. Halabe, S. Rishver, Y. Tilis, Z. Matas, A. Fux, M. Boaz, and J. Weinstein,
2006, Severe hypophosphatemia in sepsis as a mortality predictor: Annals of
Clinical and Laboratory Science, v. 36, p. 6.
Skerker, J. M., B. S. Perchuk, A. Siryaporn, E. A. Lubin, 0. Ashenberg, M. Goulian, and
M. T. Laub, 2008, Rewiring the specificity of two-component signal transduction
systems: Cell, v. 133, p. 12.
Skerker, J. M., M. Prasol, B. Perchuk, E. Biondi, and M. T. Laub, 2005, Two-component
signal transduction pathways regulating growth and cell cycle progression in a
bacterium: a systems-level analysis: PLoS Biology, v. 3, p. 334-353.
Steed, P. M., and B. L. Wanner, 1993, Use of the rep technique for allele replacement to
construct mutants with deletions of the pstSCAB-phoU operon: evidence of a new
role for the PhoU protein in the phosphate regulon: Journal of Bacteriology, v.
175, p. 6797-6809.
46
Stock, A. M., V. L. Robinson, and P. N. Goudreau, 2000, Two-component signal
transduction: Annual Review of Biochemistry, v. 69, p. 33.
Surin, B., H. Ronsenberg, and G. Cox, 1985, Phosphate-specific transport system of
Escherichiacoli: nucleotide sequence and gene-polypeptide relationships: Journal
of Bacteriology, v. 161, p. 10.
Tinsley, C. R., B. N. Manjula, and E. C. Gotschlich, 1993, Purification and
characterization of polyphosphate kinase from Neisseriameningitidis: Infection
and Immunity, v. 61, p. 8.
Torriani, A., 1960, Influence of inorganic phosphate in the formation of phosphatases by
Escherichiacoli: Biochim et Biophys Acta, v. 38, p. 10.
van der Heide, T., and B. Poolman, 2002, ABC transporters: one, two or four
extracytoplasmic substrate-binding sites?: EMBO reports, v. 3, p. 6.
von Kriger, W. M. A., L. M. S. Lery, M. R. Saures, F. S. de Neves-Manta, C. M. Batista
e Silva, A. G. da Costa Neves-Ferriera, J. Perales, and P. Mascarello Bisch, 2006,
The phosphate-starvation response in Vibrio cholerae 01 and phoB mutant under
proteomic analysis: disclosing functions involved in adaptation, survival, and
virulence: Proteomics, v. 6, p. 17.
Wanner, B. L., 1996, Phosphorus assimilation and control of the phosphate regulon, in
NeidhardtFC, ed., Escherichiacoli and Salmonella: Cellular and Molecular
Biology, v. 1: Washington, DC, ASM Press, p. 1357-1381.
Yakhnina, A. A., and Z. Gitai, 2013, Diverse functions for six glycosyltransferases in
Caulobactercrescentus cell wall assembly: Journal of Bacteriology, v. 195, p. 9.
Yang, C., T.-W. Huang, S.-Y. Wen, C.-Y. Chang, S.-F. Tsai, W.-F. Wu, and O.-H.
Chang, 2012, Genome-Wide PhoB Binding and Gene Expression Profiles Reveal
the Hierarchical Gene Regulatory Network of Phosphate Starvation inEscherichia
coli PLoS ONE, v. 7, p. 1.
Yuan, Z.-C., R. Zaheer, and T. M. Finan, 2005, Phosphate limitation induces catalase
expression in Sinorhizobium meliloti, Pseudomonas aeruginosaand
Agrobacterium tumefaciens: Molecular Microbiology, v. 58, p. 8.
Zaborin, A., K. Romanowski, S. Gerdes, C. Holbrook, F. Lepine, J. Long, V. Poroyko, S.
P. Diggle, A. Wilke, K. Righetti, 1. Morozova, T. Babrowski, D. C. Liu, 0.
Zaborina, and J. C. Alverdy, 2009, Red death in Caenorhabditiselegans caused
by Pseudomonas aeruginosaPAO 1: Proceedings of the National Academy of
Sciences, v. 106, p. 6.
Zaborina, 0., C. Holbrook, Y. Chen, J. Long, A. Zaborin, I. Morozova, H. Fernandez, Y.
Wang, J. R. Turner, and J. C. Alverdy, 2008, Structure-function aspects of PstS in
multi-drug-resistant Pseudomonas aeruginosa:PLoS Pathogens, v. 4, p. e43.
Zaheer, R., R. Morton, M. Proudfoot, A. Yakunin, and T. M. Finan, 2009, Genetic and
biochemical properties of an alkaline phosphatase PhoX family protein found in
many bacteria: Environmental Microbiology, v. 11, p. 16.
Zavaleta-Pastor, M., C. Sohlenkamp, J.-L. Gao, Z. Guan, R. Zaheer, T. M. Finan, C. R.
H. Raetz, I. M. L6pez-Lara, and 0. Geiger, 2010, Sinorhizobium meliloti
phospholipase C required for lipid remodeling during phosphorus limitation:
Proceedings of the National Academy of Sciences, v. 107, p. 6.
47
Chapter 2: Global characterization of the Pho regulon in
Caulobactercrescentus
48
Abstract
Sensing and responding to phosphate is critical to bacterial life. For Caulobacter
crescentus, which lives in a phosphate-limited environment, this process is especially
crucial. Caulobacterand many other bacteria respond to phosphate limitation through a
widely conserved signaling pathway, but precisely how bacteria sense the phosphate state
of the environment, and what genes they regulate in response to phosphate limitation, is
poorly characterized. Here, we map the global binding patterns of the phosphateresponsive transcriptional regulator PhoB in both phosphate-limited and -replete
conditions. We demonstrate that PhoB binds to a handful of genes in phosphate-replete
conditions, but then is induced to bind over 50 genes, termed the Pho regulon, in
phosphate-starved conditions. Further, we find that PhoU, a protein thought to coordinate
phosphate import with repression of the Phn regaulcIn dcs nnt influence the Pho regulon,
and instead likely links phosphate import to other metabolic processes.
49
Introduction
Bacteria typically must sense and rapidly respond to the nutrient state of their
environment to survive. Although this ability to adapt to extracellular changes is critical,
the mechanisms by which it occurs remain incompletely understood. Sensing the
availability of extracellular phosphate is particularly important as phosphate is required
for the synthesis of many biomolecules, from ATP to phospholipids. An ability to sense
and respond to phosphate is thus important for maximal growth of bacteria (Steed and
Wanner, 1993), and is implicated in biofilm formation (O'May et al., 2009) and the
virulence of some pathogens (Bertrand et al., 2010; Jacobsen et al., 2008; Lamarche et
al., 2008; Pratt et al., 2010).
Most bacteria respond to phosphate limitation through a widely conserved signal
transduction pathway whose connectivity and functions remains only partly characterized
(Hsieh and Wanner, 2010). In this pathway, phosphate conditions are thought to be
sensed through changes in phosphate uptake by the low-affinity Pst transporter in
conjunction with a, two-component signaling pathway, PhoR-PhoB, known as the Pho
system. For E. coli cells, the Pst transporter is active during growth in phosphate-replete
conditions, which somehow inhibits autophosphorylation of the histidine kinase PhoR.
When phosphate becomes limiting and flux through the Pst transporter is reduced, PhoR
is stimulated to autophosphorylate and then transfer its phosphoryl group to PhoB.
Phosphorylated PhoB undergoes a conformational change and dimerizes along its a4-P5ca5 interface (Mack et al., 2009), allowing it to then bind conserved DNA sequences
called pho boxes in certain promoters usually leading to increased transcription of target
genes (Blanco et al., 2002). X-ray crystallography and mutational studies indicate that
50
PhoB binds to region 4 of &", stabilizing its association with the -35 region of target
promoters (Blanco et al., 2011; Makino et al., 1993).
In E. coli the expression of more than 40 genes changes following phosphate starvation,
including the pst and pho genes themselves (Yang et al., 2012). These genes were
identified through a combination of reporter studies and microarray analysis, but could
not distinguish between direct and indirect targets (Baek and Lee, 2006; Metcalf et al.,
1990). More recently, a ChIP-chip study of PhoB identified some putative direct targets,
but did not examine PhoB binding in high phosphate conditions (Baek and Lee, 2006;
Metcalf et al., 1990; Yang et al., 2012).
Two-component systems are the predominant means by which bacteria sense and respond
to external stimuli (Capra and Laub, 2012; Stock et al., 2000). Although many histidine
kinases bind extracellular ligands, others lack a large extracellular domain, and are
thought instead to respond to an intracellular signal (Hsieh and Wanner, 2010; Krell et
al., 2010). PhoR is a membrane-embedded kinase, but does not contain a significant
periplasmic domain. PhoR has been suggested to sense extracellular phosphate status
through an interaction with the Pst transporter, which also resides in the inner membrane,
but the precise mechanism by which the Pst system regulates PhoR is unclear. A protein
of unknown function, PhoU, has been proposed as an intermediate between the Pst and
Pho systems, inhibiting PhoR when the Pst system is actively transporting phosphate
(Baek et al., 2007). phoU is widely conserved in bacteria and frequently co-regulated
with the pst and pho genes (Baek et al., 2007; Steed and Wanner, 1993). Work in E. coli
has shown that the expression of alkaline phosphatase and some other members of the
Pho regulon are upregulated in pho U mutants (Li and Zhang, 2007; Wanner and Latterell,
51
1980), indicating that PhoU may function as a negative regulator of the Pho regulon.
However, evidence of a direct interaction between PhoU and any of the membranelocalized Pst or Pho proteins has not been documented.
Although many bacteria use the Pst-Pho signaling pathway to respond to phosphate
limitation, they must ultimately adapt to changes in phosphate levels in widely differing
ways. For example, in the freshwater a-proteobacterium Caulobactercrescentus low
extracellular phosphate stimulates elongation of a polar appendage called the stalk, which
is a tubular extension of the cell envelope. Phosphate starvation can lead cells to extend
their stalks up to 20-fold their lengths in phosphate-replete conditions (Gonin et al., 2000;
Schmidt and Stanier, 1966). It was initially suggested that stalk elongation may increase
the nutrient scavenging ability of phosphate-starved Caulobactercells, but subsequent
studies found that a diffusion barrier exists between the stalk and cell body, preventing
free exchange of membrane and periplasmic proteins (Klein et al., 2013).
Here, we used a combination of microarray analysis, bioinformatics, and ChIP-Seq to
identify the direct targets of CaulobacterPhoB. The CaulobacterPhoB regulon includes
nearly 50 genes, with relatively few genes in common with the E. coli PhoB regulon,
beyond the pst and pho genes. These results demonstrate how a highly conserved
signaling pathways can be used to drive vastly different programs of gene expression in
different bacteria, highlighting the plasticity of bacterial regulatory networks. Further, we
show that the conserved protein PhoU does not act as a negative regulator of the PhoRPhoB signaling pathway, but instead functions as a critical player in cellular phosphate
metabolism, perhaps by modulating the levels of polyphosphate.
52
Results
Epitope-tagged PhoB retains wild-type function
To perform ChIP-Seq on PhoB, we constructed a strain in which the chromosomal copy
of phoB encodes a 20 amino acid linker and C-terminal 3XFLAG tag (phoB-3XFLA G).
To test whether this version ofphoB supports wild-type-like growth in both phosphatereplete and starved conditions, we grew the phoB-3XFLAG strain in minimal medium
with 10 mM phosphate (M2G), and then washed and resuspended it in minimal medium
with either 10 mM phosphate (M2G) or 50 [tM phosphate (M5G). The phoB-3XFLAG
strain exhibited growth nearly indistinguishable from the wild type in both conditions, in
contrast to a phoB deletion strain, which grew poorly in both conditions (Figure 2. lA).
We also tested whether PhoB-3XFLAG regulated PhoB-dependent genes in a manner
comparable to untagged PhoB. We constructed lacZ reporters for the pstC and pstS
promoters in Caulobacter,and then assessed the ability of phoB-3XFLAG to induce
expression of each reporter following phosphate limitation. Cells were shifted from M2G
to M5G, or mock-shifted and retained in M2G, and grown for 7 hours to mid-exponential
phase before measuring P-galactosidase activity. In M2G, the activity of both reporters
was ~2,000 Miller units in both the phoB-3XFLA G and wild-type strain. In M5G, Pgalactosidase activity was induced to -10,000 Miller units for the Pps,, reporter in both
the wild type and the phoB-3XFLA G strain, and around 8,000 Miller units for the P/ss
reporter in both strains. These results indicate that FLAG-tagged PhoB functions as well
as untagged PhoB to induce the expression of PhoB-dependent genes. As a control, we
53
confirmed that in a phoB deletion strain, both reporters exhibited less than 2,000 Miller
units of activity, consistent with these promoters being PhoB-dependent (Figure 2.1 B).
Finally, we tested PhoB-3XFLAG in a strain in which PhoB is constitutively activated.
Null mutations in pstS, which encodes the periplasmic phosphate binding protein, block
phosphate import and result in a hyper-activation of the Pho regulon in Caulobacter,
including stalk elongation, even in phosphate-replete conditions (Gonin et al., 2000). We
found that in a pstS::Tn5 background, a strain producing PhoB-3XFLAG also exhibited
extensive stalk elongation (Figure 2. 1C), further supporting our conclusion that the
epitope-tagged version of PhoB binds and regulates the same set of target genes as wildtype PhoB.
54
1.6
1,4
A
B
10 mM PI
/
0.8
0.6/
0.4
/
0.2
1.6
1.
.w
i. phoB-3XF
APWo
*,
4
12 16
time (h)
8
20
0.6
0 WT
4
8
6000
4000
4000
2000
2000
phoB-3F
12 16 20
time (h)
24
28
So pM Pi
WT
12.000
10,000
10,000
WOO
g 8000
6000
6000
4000
4000
2000
2000
0
0
PpatS48CZ
Ppstc-lacZ
PYE pstS::Tn5
PYE
C
Ppsts-IacZ
Ppac4acZ
0.8
.v
8000
6000
12,000
1.2
0.2
0
0
w phoB-3XF
28
24
so PM PI
0.4
12.000@
10,000
800 0
q
0
10 mM Pi
10.000
1.2
0
12.000
phoB-3XF
WT
phoB-3XF
Figure 2.1 - A strain harboring C-terminally 3XFLAG tagged phoB behaves like wild-type in phosphate-replete
and phosphate-limited conditions.
(A) Growth curves of the noted strains in minimal medium with 10 mM or 50 pM phosphate. (B) Beta-galactosidase
assay of expression of a Ppstc-IacZ (left) and Ppsts-lacZ reporter (right) in 10 mM phosphate minimal medium (top) and
50 pM phosphate minimal medium (bottom). Strains were grown in 10 mM phosphate medium, washed, and resuspended in 10 mM or 50 pM phosphate medium, and outgrown for 7 hours to 0.3 < OD, < 0.4. (C) Light microscopy of
the indicated strains grown in PYE (rich medium).
ChIP-Seq reveals genome-wide binding patterns of PhoB
We performed ChIP-Seq analysis on cells expressing phoB-3XFLAG and grown to midexponential phase in (1) PYE, a complex rich medium in which PhoB should be
predominantly unphosphorylated and inactive, and (2) minimal medium containing 50
[tM phosphate (M5G), in which PhoB is phosphorylated and active, as judged by the
PhoB-dependent reporters for Ppstc and Ppss (Figure 2.1 B). Additionally, we performed
ChIP-Seq on a strain grown in phosphate-replete PYE that, in addition to expressing
phoB-3XFLAG, harbors a disruption of pstS, leading to constitutive activation of PhoB.
In each case, PhoB-bound DNA was immunoprecipitated using an anti-FLAG antibody,
55
and DNA subsequently purified (see Experimental Procedures). As controls, we also
performed ChIP using the anti-FLAG antibody on a strain producing untagged PhoB and
grown in phosphate-replete and phosphate-limited conditions, as well as pstS.: Tn5 cells
producing untagged PhoB and grown in phosphate-replete conditions.
We first used qPCR to verify that the ChIP samples for strains producing tagged PhoB
were enriched for a chromosomal region (the pstC promoter) predicted to be PhoBbound, based on E. coli studies and the expression analyses presented below. As a control
locus we used PCC12 94 whose expression is not PhoB-regulated. To determine the amount
by which a ChIP sample was enriched for these loci, we also performed qPCR on an
input DNA control for each sample: a portion of each sample was reserved before
subjecting the sample to ChIP, and DNA was isolated from this input sample, and
analyzed by qPCR at the pstC and CC 1294 loci. "Fold enrichment" was then calculated
as the amount of the pstC or CC1294 locus found in a ChIP output sample, relative to the
amount found in its input sample (see Experimental Procedures).
56
A
B
BpstC
65
600
M Pt
1OmM Pi+ 3XF
5
5050
46
+3XF
E3~~~50 um Pits:n
uM P + 3XF
pstS::Tn5 PYE
pstS:: Tn5 PYE + 3XF
50
9
pstS::TnS
PYE
35
0
L.
0
9
wt
0
50 PM Pi
WI
20
+3XF
16
10
000
PYE
305
400
80
__
L00
J
im-:-
11400
8
--
0
-
1800
F_
1050
0
1052
1400
11400
00
1698
0
1400
0
1698
genomic position {kb)
PCC1294
PpstC
C
307
10
1400
F ~
0
9000
4 00
0
9
5
0
LA
-
90L
25
SF~oo
0
800
1
+3XF
30
patS
CC0925
woo
+3XF
1
pstS::Tn5
PYE
-1
'M __ - a -
M
Wt
+3XF
. I
50 PM PI
a a
Wt
+3XF
PYE
W
4016
genomic position (kb)
Figure 2.2 - ChIP-seq reveals genome-wide binding patterns of PhoB.
(A) qPCR at predicted phoB-activated (PpstC) and unactivated (PCC1294) loci of DNA samples obtained through ChIP
of the phoB-3XFLAG strain, using anti-FLAG antibody. Y-axis shows fold enrichment of a locus in ChIP output sample
compared to input DNA (see Experimental Procedures for methods and a description of the fold enrichment calculation). (B) Read plots for example loci from pstS::Tn5, 50 pM Pi medium, and PYE (rich medium) samples. Corresponding no-FLAG control is shown for each phoB-3XFLAG experimental sample. Wild-type and 3XFLAG designations refer to the genotype at the phoB locus. (C) Genomic read plots from pstS::Tn5, 50 pM Pi medium, and PYE
phoB-3XFLAG and no-FLAG control samples. Plots are scaled to maximum read height in the pstS::Tn5 sample
(9000 reads).
All three experimental samples were enriched over 20-fold for the pstC promoter in the
ChIP output DNA sample compared to the input DNA sample (Figure 2.2A), in contrast
to the wild-type control samples, which were less than 2.5-fold enriched for this locus. As
expected, the CC1294 promoter was not significantly enriched in the experimental
samples above the level of enrichment observed in control samples (Figure 2.2A).
We then constructed libraries from, and deep-sequenced, the rich medium, phosphate
limited, and pstS mutant ChIP samples taken from cells producing epitope-tagged PhoB,
57
along with control ChIP samples taken from strains treated identically but harboring the
wild-type copy of phoB. Equal numbers of reads (860,000) were analyzed from each
sample using the peak-calling software MACS (Zhang et al., 2008), and peaks with a pvalue < 10- identified.
Read profiles at individual loci across all three conditions indicate that PhoB is
constitutively bound to some promoters, even in rich medium, while at other loci,
phosphate starvation stimulates PhoB binding (Figure 2.2B), consistent with our qPCR
data (Figure 2.2A). However, genome-wide PhoB binding patterns (Figure 2.2C)
indicated that the majority of PhoB binding is induced by phosphate-limitation. Only 5
significant peaks were identified in the PYE sample, while 102 significant peaks were
found in the 50 [M phosphate sample, and 204 were identified in the pstS::Tn5 sample,
consistent with PhoB being hyper-activated in this condition (Figure 2.3A).
Peak fold-enrichment for the ChIP-Seq data was determined by comparing numbers of
reads at a peak in the epitope-tagged experiment to reads at that locus in the non-epitopetagged control. At most loci, the observed peak fold-enrichment was greater in the pstS
mutant sample than in the low phosphate sample (Figure 2.3B). In total, 92 peaks
identified in the pstS sample were also identified in the 50 xM phosphate sample (Figure
2.3A). The overlap between these two independent samples, in which PhoB is activated
in two different nutrient conditions, supports the notion that these peaks represent bona
fide, direct PhoB targets, and suggests that PhoB regulates the same core set of genes at
different levels of phosphate limitation.
A complete set of peaks identified in each sample is given in Appendix 2. To delineate
the high-confidence members of the PhoB regulon and a consensus PhoB binding site, we
58
focused on the set of 50 peaks that were > 7.5-fold enriched in the pstS mutant. Of these
peaks, 43 were also found to be significant in the 50 [M phosphate condition (Figure
2.3B). This set includes genes responsible for phosphate transport and metabolism, as
well as motility and additional transport genes (Figure 2.3B).
The vast majority of the 50 peaks > 7.5-fold enriched were found in intergenic regions.
Only 4 out of 50 peaks were 90% or more contained within an annotated intragenic
region. This pattern is consistent with the proposed model in which PhoB activates
transcription by binding near the -35 region of promoters (Blanco et al., 2002).
B
A
L
0.
2
=L
M
C pstS::Tn5 (204 peaks)
C) 50 pM Pi (102 peaks)
fold-enrichment
I60
40
20
10
no peak
C0290 Phosphate transport system permease PstC
C0996 TonB-dependent receptor
C0722 TonB-dependent receptor
C0170 Hypothetical protein
C3094 Sensory box/GGDEF family protein
CC0172 General secretion pathway protein C
C1791 TonB-dependent receptor
C2819 TonB-dependent receptor
CC0361 Phosphonates transport ATP-blnding protein PhnC
CC1515 Phosphate binding protein PatS
CC2149 TonB-dependent receptor
CC3301 PeplIdoglycan-speclc endopeptidase, M23 family
CC2225 Conserved hypothetical protein
CC3014 Hypothetical protein
CC1970 TonB-dependent receptor
CC2810 Methyl-accepting chernotaxis protein
CC2923 TonS-dependent receptor
CC0210 TonB-dependent receptor
CC0487 Ribose-phosphate pyrophosphokinase
CC0316 Hypothetical protein
CCNA03638 Putative adenylate cyclase family
CC2967 Cell wall hydrolase family protein
CC0925 OAR outer membrane protein precursor
CC2657 Transposase
CC1015 Type I secretion outer membrane protein rsaFa
CC2237 Deoxycytidine triphosphate deaminase
!CC3171 Hypothetical protein
CC3344 UDP-2,3-dlacylglucosamine hydrolase
CC0453 Ribosomal large subunit pseudouridine synthase D
CC0794 Flagellin
CC0756 Phosphatidylglycerol glycosyliransferase
CC3703 Glutamate racemase
CC3706 Hypothetical protein
CC1182 Two-component response regulator
CCNAR0037 tRNA-Glu
CC1263 LSU ribosomal protein L6P
CC1163 Acyl carrier protein
CC1203 Periplasmic multidrug efflux lipoprotein precursor
CC1099 TonB-dependent outer membrane receptor
CC3504 Zinc metalloprotease
CC0162 Hypothetical protein
CC3718 Tetratricopeptide repeat family protein
CC3395 Transcriptional regulator, AlgH
CC1666 TonS-dependent outer membrane receptor
CCNARROO79 Minimal medium expressed sRNA
CC0551 Hypothetical protein
CC2017 Conserved hypothetical protein
CC0264 AMP nudeosidase
CC2336 TonS accessory protein ExbB
CC1103 Transporter, MFS superfamily
Figure 2.3 - PhoB binds Pho regulon genes upon phosphate limitation.
(A) Overlap between the sets of genes containing significant peaks in the 50 pM phosphate and pstS mutant ChIP-seq
samples. (B) Fold-enrichment of genes 7.5-fold or more enriched the pstS::Tn5 sample is shown for each sample. Blue
indicates that no significant peak was called.
59
Identification of the PhoB regulon
To validate ChIP-Seq peaks as representing PhoB binding sites, the motif-finding
program MEME (Bailey et al., 2009) was used to identify a consensus site based on the
sequences of the 24 peaks that were > 13-fold enriched in the pstS mutant. The resulting
motif is composed of two 6 base pair sites that appear to be direct repeats flanking an
AT-rich region (Figure 2.4B). Although similar to the PhoB consensus site predicted in
E. coli (Yang et al., 2012), this Caulobactersite is one base pair shorter and has a higher
GC content, the latter likely reflecting the higher GC content of the Caulobactergenome.
In E. coli, the central AT-rich region has been proposed to be a modified -35 binding site
(Yang et al., 2012). The putative CaulobacterPhoB binding motif, or Pho box, was then
used to predict PhoB binding sites across the Caulobactergenome using MAST (Bailey
et al., 2009), which identified Pho boxes within 37 of the 50 most enriched peaks
(Appendix 1).
60
A
B
pstS::Tn5;
patS:: Tn5 ephoB
ChIP
peak
ptio box
phiobo
putative -35 site
C
o
8
7
6
S4
25
3.
log,
fold change
0
00
100 -220
3.00
*
*
-10
0F
40
60
80
100
120
140
180
-2 .
-3
ChIP-seq fold ennchment In pstS::Tn5 in PYE
Figure 4 - ChIP-seq differentiates between direct and indirect PhoB targets, and identifies PhoB-repressed
genes
(A) PhoB-dependent genes determined by microarray analysis. Genes upregulated two-fold or more in the pstS::Tn5
strain are shown. Presence of a significant ChIP-seq peak in the gene or gene's operon in the pstS::Tn5 sample is
indicated in green. (B) MEME was used to identify a motif from sequences of peaks >13 fold upregulated in the
pstS::Tn5 ChIP-seq sample. E-value is 1.3E-20. PhoB binding sites (pho boxes) and putative -35 site are labeled. (C)
Comparison of ChIP-seq fold enrichment to microarray expression change in the pstS::Tn5 mutant background. 10 out
of 50 genes most enriched the ChIP-seq sample were downregulated in the pstS::Tn5 microarray.
To further pinpoint direct PhoB targets, we used whole genome DNA microarrays to
identify genes whose expression depends on PhoB. We harvested RNA from strains
harboring either the pstS::Tn5 mutation alone or the pstS.: Tn5 mutation in a AphoB
background, with each strain grown to mid-exponential phase in rich medium. RNA from
these strains was compared on microarrays to RNA obtained from wild-type Caulobacter
grown under the same conditions. To identify PhoB-activated genes, we selected those
that were upregulated in the pstS mutant, but not in the pstS.: Tn5;AphoB double mutant.
Specifically, we found 48 genes that had peaks in the pstS.: Tn5 ChIP-Seq data and were
at least 2-fold upregulated in the pstS::Tn5 mutant (Figure 2.4A), indicating that these
genes are likely directly activated by PhoB. We also found 10 genes that had PhoB ChIP-
61
Seq peaks and that exhibited at least 2-fold downregulation in the pstS::Tn5 mutant
(Figure 2.4C), suggesting that PhoB likely represses these genes.
We identified an additional 27 genes that are at least 2-fold upregulated in the pstS:: Tn5
mutant, but that did not have peaks in the pstS:. Tn5 ChIP-Seq data. These 27 genes are
likely indirectly activated by PhoB. The putative regulatory regions of these indirect
targets were subjected to analysis by MEME (Bailey et al., 2009); the best predicted
motif had an E-value of only 2 x 10-3 and, as expected, showed little resemblance to the
inferred PhoB binding site (data not shown).
PhoU is not a negative regulator of the Pho regulon in Caulobacter
The activation of PhoB as a transcription factor depends on the histidine kinase PhoR,
which somehow senses changes in flux through the phosphate transporter PstABC.
Whether this sensing is direct is unknown. A highly conserved protein called PhoU is
often encoded in the same operon as the pst genes and has been suggested to couple PhoR
with the Pst transporter. Specifically, PhoU was suggested to repress PhoR activity in
phosphate-replete conditions when the transporter is active, implying that PhoU is a
negative regulator of the Pho regulon (Hsieh and Wanner, 2010).
To test this hypothesis in Caulobacterwe constructed a strain in which pho U is deleted
from its native locus and is instead inserted at the van locus under the control of the Pvan
promoter, permitting inducible expression ofpho U by the addition of vanillate to the
medium. The phoU depletion strain was cultured overnight in rich medium supplemented
with vanillate, and cells were then shifted to medium without vanillate, and diluted as
needed to maintain exponential growth. We found that the depletion of PhoU, upon
removal of vanillate, did not result in stalk hyper-elongation, as would be expected upon
62
loss of a negative regulator of the Pho regulon, and as occurs in pstS mutants. Instead, the
loss of PhoU led to enlarged and slightly filamentous cells after 8 hours of depletion, and,
after 16 hours of depletion, modest chromosome accumulation (Figure 2.5A). Further,
and also in contrast to a pstS mutant, we found that the depletion ofphoU was lethal. To
measure viability, the phoU depletion strain was shifted from medium with vanillate to
medium without vanillate, and diluted once approximately every two generations to
maintain growth in mid-exponential phase. Samples were taken at 3-hour intervals to
measure colony forming units (CFUs), normalized at each time point to CFUs per 1 mL
of culture at OD60 0 =1. We observed a three-log decrease in colony forming units over 30
hours of depletion (Figure 2.5B). In contrast, when wild-type orpstS::Tn5 cells were
treated in the same manner, we observed no loss in viability. Collectively, these results
indicate that PhoU likely does not negatively regulate the Pho regulon in Caulobacter.
A
-
+VANILLATE
:Skan
Pvwn-pho4J;AphoU
Pvar-phoU;.phoU
Pva.-phoU AphoU
VANILLATE
WT
IS::k.
WT
PnphoU;U
IN 2N
1N 2N
B
DNA contencell
DNA contentcell
+ VANILLATE
1010
a108
0108
0
10
106
105
-VANILLATE
101
'
0
WT
107 -
Pvan-phoUAphoU
108
A pstS::Tn5
5
10
16
20
25
30
10~
WT
" Pvn-phoU:AphoU
A pstS::Tn5
0
5
10
15
20
25
30
Figure 2.5 - phoU depletion does not phenocopy pstS mutation.
(A) Light microscopy of the noted strains in the presence and absence of vanillate, 8 hours after
removal of vanillate, and flow cytometry of the phoU depletion strain in the presence and absence of
vanillate,16 hours after removal of vanillate. (B) Colony forming units for the noted strains in PYE.
CFUs given are the average of two replicates and indicate the CFUs per 1 ml of normalized OD 1
culture.
63
To directly assess whether PhoU influences the Pho regulon, we examined global
patterns of gene expression in the pho U depletion strain. A culture was grown in the
presence of vanillate and then shifted to medium with or without vanillate. Samples were
removed at 2, 5, and 7 hours post-shift and RNA from the two conditions directly
compared on DNA microarrays. The vast majority of genes upregulated 2-fold or more in
a pstS mutant were not upregulated after phoU was depleted (Figure 2.6A). Expression of
the pstCAB genes, but not pstS, did increase up to 2.6-fold after 7 hours, although not as
much as in the pstS mutant.
We also assayed the effect of depleting PhoU using lacZ reporters for the pstC and pstS
promoters, which are PhoB-regulated (Figure 2.1 B, 2.3B). We again shifted the phoU
depletion strain to non-inducing conditions, and observed a modest increase in activity of
the Pps'c reporter after 7 or more hours, but not the Ppst, reporter. In both cases,
substantially higher activity was seen in the pstS mutant. Taken together, these data
indicate that a loss ofphoU does not induce the same expression patterns seen in a pstS
mutant in which the Pho regulon is upregulated (Figure 2.6B). These data support our
assertion that PhoU does not function as a negative regulator of the Pho regulon in
Caulobacter.
64
A
phoU depleted
2 hrs
5 hrs
7 hrs
B
ChIP
pstS::Tn5 peak
-
30.000
U
25.000
PYE
PYE
+
vanillate
20,000
CC0996
phos
15,000
S10,000
wt AphoB patS::
tn5
pstA
7 hrs
13 hrs
phoU depletion
30,000
N PYE
PYE
25.000
+
vanillate
20.000
log,
fold change
15.000
10,000
2.00
-200
0005000
JII
UEEOIE
-1.00
3.00
wt AphoB psts::
tn5
13 hrs
7 hrs
phoU depletion
Figure 6 - phoU functions independently of the Pho regulon
(A) Expression changes upon phoU depletion were assayed by microarray at 2, 5, and 7 hours and are the average
of two replicates. Genes which were two-fold or more upregulated in the pstS disruption strain are shown. Green
indicates direct phoB targets, as determined by ChIP-Seq. Genes labeled are those 2-fold or more upregulated in
the phoU depletion strain at the 7 hour timepoint and found to be direct phoB targets. (B) Beta-galactosidase assay
of Ppstc-lacZ and Ppsts-lacZ reporter expression after phoU depletion in PYE. Hours after removal of vanillate are
indicated.
Mutations in both the Pst and Pho systems suppress a phoU mutant
To further probe the function of PhoU, we isolated mutations that restore viability to a
strain depleted of PhoU. We performed transposon mutagenesis of the Pvan-phoU
depletion strain using kanamycin-marked Tn5. We plated the transposon-mutagenized
cells on rich medium with kanamycin and without vanillate, to select for mutants that
were viable in the absence ofphoU. We selected colonies that grew up after 2-4 days, and
identified the site of transposon insertion for 18 candidate suppressors (Figure 2.7A).
Two insertions mapped near vanR, the vanillate repressor, which regulates phoU
transcription in this strain, and one mapped to cobT, a cobaltochelatase, which we did not
65
independently verify. Seven Tn5 insertions mapped to phoR or phoB while the remaining
8 insertions were distributed among the four components of the pst system, pstSCAB.
A
44
I
pstC
phoR
+ AohoR
marR
C
Pvan-phoU; AphoU
phoU
pstB
pstA
vanR
pstS
B
4 W1
101
phoB
cobT
+ VANILLATE
+ DstS::Tn5
10
PYE
+van
108
0
L.
U
107
Pvan-phoU;AphoU
PYE
-van
A Pvan-phoU;AphoU;pstS::Tn5
106
*Pvan-phoU;AphoU;AphoR
0
5
10
D
15
20
25
30
time (h)
10 1
- VANILLATE
A.
A
A~
10,
108
LL
a
transcription
10 7
x Pvan-phoU;AphoU
?
A Pven-phoU;AphoU;pstS::Tn5
106
cel death
* Pvan-phoU;AphoU;AphoR
105
0
5
10
15
20
25
30
time (h)
E
Pvan-phoU AphoU
S6k,-.
I A h-
PYE
-van
Figure 2.7 - Mutations in the pst and pho genes suppress phoU depletion lethality.
(A) Location of Tn5 insertions which suppressed phoU depletion lethality, indicated by arrows. Bars denote directly
adjacent, but not co-operonic, loci. phoU, which was deleted in this background, is shown in dark gray. (B) Light
microscopy of the noted strains in the presence and absence of vanillate. (C) Colony forming units (CFUs) are the
average of two replicates and are given as the CFUs in 1 ml of normalized OD 1 culture. Cultures were grown in the
presence (top panel) and absence (bottom panel) of vanillate. (D) Transcription of the Pst system is regulated by the
PhoR/PhoB pathway. Disruption of either the Pst or Pho systems decreases phosphate import, suggesting that PhoU
regulates a phosphate-dependent process. (E) Light microscopy of the phoU depletion strain grown in the absence of
vanillate for 8 hours (left) or 14 hours (right). Granules are marked with white arrowheads.
66
To verify that mutations in both the pst and pho systems can suppress the lethality of a
pho U depletion, we independently transduced the pstS:.: Tn5 allele and a AphoR mutation
into the pho U depletion strain. We found that in each case the mutation introduced was
epistatic to the depletion of pho U. Cells depleted of PhoU and harboring a phoR deletion
had a morphology similar to the AphoR strain and no longer lost viability when shifted to
medium lacking vanillate (Figures 2.713-C). Similarly, pstS::Tn5 cells depleted of PhoU
exhibited the long-stalk phenotype associated with pstS:: Tn5 and retained viability upon
shift to medium without vanillate. These results confirm that mutations in both the PhoRPhoB signaling pathway and the Pst transporter can suppress the lethality of depleting
PhoU.
These results corroborate our conclusion that PhoU is not a negative regulator of the Pho
regulon. If it were, and if the lethality of depleting phoU were due to overexpression of
the Pho regulon, our screen would have identified suppressor mutations in phoR and
phoB, but not in the pst genes as pst mutations upregulate the Pho regulon (Figure 2.4A
above). Instead, the results of our screen strongly suggest that the lethality of a PhoU
mutant is suppressed by a reduction in the levels of the Pst transporter, which is achieved
directly by disrupting pst genes or indirectly by disrupting phoR or phoB, which are
required for expressing the pst genes (Figures 2.1 B, 2.3C, 2.7D). Further, these genetic
data suggest that PhoU may participate in regulating or metabolizing cellular pools of
inorganic phosphate that are taken up by the Pst system. Without PhoU, the inorganic
phosphate that gets imported may be converted to a toxic form, leading to cell death.
Alternatively, activity of the Pst transporter may be increased in the absence of PhoU,
and cells may not tolerate the resulting excessive concentrations of inorganic phosphate.
67
Discussion
Two-component systems are the predominant signaling modality in bacteria. Although
these systems employ a variety of output mechanisms, the majority uses a response
regulator to enact a transcriptional response. However, the regulons of most response
regulators remain uncharacterized (Galperin, 2010; Krell et al., 2010). Here, we used
global expression studies and ChIP-Seq to create a high-resolution map of PhoB binding
sites in Caulobactercrescentus in both phosphate-limited and -replete conditions. We
found that PhoB binds the predicted regulatory regions of over 50 genes (Figure 2.3C).
Microarray analysis indicates that the transcription of at least 27 other genes may also be
activated by PhoB (Figure 2.4A). The activation of these 27 genes likely represents an
indirect response to phosphate limitation that could be mediated by PhoB-activated
transcriptional regulators, or by other PhoB-regulated signaling mechanisms such as
TonB-dependent transport systems, which can effect downstream transcriptional changes
through control of ECF sigma factors (Koebnik, 2005).
We also found that PhoB negatively regulates gene expression with microarray analysis
indicating that 10 PhoB targets are likely directly repressed by PhoB (Figure 2.4C). By
comparing PhoB ChIP-Seq peak positions to transcription start sites inferred from RNASeq data (Fang et al., 2013), we propose that PhoB activates gene expression when bound
near the -35 region of target promoters, but represses expression when bound further
from +1 site (Appendix 1).
68
PhoB regulates a different set of genes in Caulobacter than in E. coli
A recent ChIP-chip-based study of the Pho regulon in E. coli (Yang et al., 2012) enables
a comparison of the PhoB regulons in Caulobacterand E. coli. We find that the set of
genes regulated by PhoB in Caulobacterdiffers substantially from that regulated in E.
co/i, indicating that although the upstream signaling pathway remains intact, the output
regulon has likely been tailored to each organism's needs and ecological niche. Of the 50
genes comprising the CaulobacterPho regulon, we could identify a reciprocal best
BLAST hit in E. coli for only 18. Of these 18 genes, only 4 are members of the E. coli
Pho regulon (Appendix 1); these include phoB itself, the pst operon, and the phosphonate
(phn) transport system operon. This minimal overlap in the PhoB regulon could reflect
differences in the ways in which genes were called as PhoB targets, although this is
unlikely to explain the lack of overlap entirely. In some cases, there could be different
genes in the two organisms that fulfill similar functions, particularly as both regulons
encode a number of transport-related proteins. However, it generally appears that the two
organisms have evolved fundamentally different Pho regulons.
Some of the differences in the CaulobacterPhoB regulon relative to that of E. coli
presumably account for their different physiological responses, most notably the stalk
elongation that is a hallmark of phosphate-starved Caulobacter.It is possible that the
process of stalk elongation requires the activity of the cell wall and membrane-modifying
proteins encoded in the CaulobacterPho regulon (Figure 2.3B). However, the E. coli Pho
regulon regulates similar, though not orthologous, genes (Yang et al., 2012), suggesting
that the PhoB-regulated genes identified here participate in maintenance of the cell
envelop, rather than stalk elongation. Additionally, stalk elongation may not be affected
69
by genes directly regulated by PhoB and instead could depend on a PhoB activated
transcriptional regulator or signaling protein, which in turn carries out a program of stalk
elongation. Future studies of the individual members of the CaulobacterPho regulon are
needed.
PhoU does not regulate PhoR activity in Caulobacter and instead likely
regulates phosphate metabolism
Although the Pho pathway is highly conserved throughout the bacterial kingdom and has
been studied for decades, the mechanism by which the histidine kinase PhoR senses
changes in extracellular phosphate has remained unclear. PhoU has been suggested to
couple the Pst transporter with PhoR, inhibiting PhoR only when flux through the
transporter is high. However, our data strongly suggest that PhoU does not function in
this manner, at least in Caulobacter,leaving open the questions of what role PhoU plays
and why it is essential for viability in phosphate-replete conditions.
Our genetic data indicate that an activity of the pst transporter is ultimately responsible
for the lethality observed upon pho U depletion. Thus, two general models for PhoU
function can be drawn: (1) PhoU negatively regulates the phosphate import activity of the
Pst transporter or (2) PhoU regulates a phosphate-dependent cellular process. In both
models, an excessive accumulation of intracellular phosphate or another metabolite that
accumulates in a phosphate-dependent manner ultimately kills cells lacking PhoU.
Consequently, mutations in either the pst or pho genes, which both prevent expression of
the Pst transporter and, hence, slow the import of phosphate, restore viability.
Further studies are necessary to differentiate between the two models. The first model
makes two predictions. One is that the two known activities of the Pst transporter,
70
phosphate import and negative regulation of the Pho regulon, are separable, as activation
of the Pho regulon is not observed when pho U is depleted. Although this has not been
studied in Caulobacter,work in E. coli suggests that the activities can be separated.
Mutation of residues at the interface of PstA and PstC, the two proteins that form the
transmembrane channel through which phosphate is imported, has been shown to ablate
phosphate import, but to maintain repression of the Pho regulon in phosphate-replete
conditions (Cox et al., 1988; Cox et al., 1989). These residues are conserved in
Caulobacterand other bacteria. The first model also predicts that increased phosphate
import will be observed in a phoU loss-of-function background. Several labs have
measured phosphate import in pho U mutants in E. coli, with differing results: pho U
mutants have been found to have little effect on phosphate import (Steed and Wanner,
1993), to increase phosphate import (Rice et al., 2009), and to reduce phosphate import
(Muda et al., 1992), although this last strain may have been harboring a suppressor
mutation in the pst or pho genes (Steed and Wanner, 1993).
PhoU may not affect the Pst transporter and instead could regulate a phosphate-dependent
process. There are several processes that could be targets of PhoU regulation, perhaps the
most intriguing of which is polyphosphate synthesis. Studies in several bacterial species
have found that phoU mutants accumulate large polyphosphate granules (Kashihara et al.,
2010; Morohoshi et al., 2002). It is possible that this accumulation is simply a
consequence of the increased import of inorganic phosphate in these mutants; expression
of a phosphate exporter or mutation of the pst system has been found to reduce
polyphosphate accumulation (Hirota et al., 2013; Motomura et al., 2011). If PhoU does
affect polyphosphate, it could do so by inhibiting a polyphosphate kinase or activating an
71
exopolyphosphatase; alternatively, PhoU could act as a structural component of
polyphosphate granules, perhaps limiting the length of the growing polyphosphate chain.
A PhoU homolog has been crystallized in the presence of divalent cations, clustered into
two negatively charged pockets (Liu et al., 2005; Oganesyan et al., 2005). These regions
could allow PhoU to act in a structural role by binding to the divalent cations known to
associate with polyphosphate chains (Rao et al., 2009).
PhoU may instead regulate another critical, phosphate-dependent process. For example,
misregulation of phospholipid production, which affects membrane stability, could also
explain the lethality caused by phoU depletion (Contreras et al., 1979).
Characterization of the response to phosphate limitation in Caulobacter
crescentus
We have characterized both the transcriptional output enacted in response to phosphate
limitation and the signaling pathway that regulates it in Caulobactercrescentus. We have
shown that PhoU does not act as a negative regulator of the Pho regulon, but that it is a
critical player in cellular phosphate metabolism, a role that is likely to be conserved given
the wide conservation ofphoU. We have also identified the genes that PhoB regulates in
response to phosphate limitation in Caulobactercrescentus; these genes show little
similarity to the set of PhoB regulated genes in E. coli highlighting the flexibility and
dynamics of transcriptional networks in bacteria. The delineation of the CaulobacterPho
regulon will enable a better understanding of how bacteria respond to phosphatelimitation, including how the synthesis of a polar organelle such as the stalk is regulated.
72
Experimental Procedures
Strains and growth conditions
C. crescentus strains were grown in PYE (rich medium), M2G (minimal medium), or
M5G (low phosphate medium), supplemented when necessary with oxytetracycline (1
[tg/ml), kanamycin (25 tg/ml), or gentamycin (0.6 [tg/ml). Cultures were grown at 300 C
unless otherwise noted, and diluted when necessary to maintain exponential growth. E.
coli cultures used for cloning were grown at 370 C in Super broth and supplemented
when necessary with oxytetracycline (12 mg/ml), kanamycin (50 mg/ml), or gentamycin
(15 mg/ml).
The pho U depletion strain was constructed by first integrating a copy of phoU at the
vanillate locus using plasmid pVGFPN-4 (Thanbichler et al., 2007), with the phoU open
reading frame cloned into the Ndel and XbaI sites, which removes the GFP coding region
from the plasmid. We subsequently deleted phoU from the pstC-A-B-phoU-phoB operon
using allelic replacement, as described previously (Skerker et al., 2005). The resulting
strain contains pho U at the van locus, and a markerless deletion of pho U in the pstC
operon, in which the entire coding region ofphoU has been removed except for the first
and final 9 bases.
lacZ reporter plasmids were derived from pRKlac290 (Gober and Shapiro, 1992).
pRKlac290 was digested with KpnI and Xbal, and a DNA fragment containing the 200
bp directly upstream of either the pstC orpstS annotated translation start sites with
flanking KpnI and XbaI cut sites was cloned into the multiple cloning site upstream of
the lacZ open reading frame.
73
The pstS: Tn5 strain was obtained from Yves Brun (Gonin et al., 2000), and transduced
into a clean CB 1 5N background. Lysate from this strain was also used to construct the
pstS::Tn5;phoB-3XFLAG and pstS:: Tn5;Pian-phoU;AphoU strains. The AphoR::tet strain
was constructed previously (Skerker et al., 2005). The AphoR;Pan-phoU;AphoUdouble
mutant was constructed by transduction of the AphoR allele into the phoU depletion
strain.
Microscopy
Cells in mid-exponential phase were fixed with 0.5% paraformaldehyde and mounted on
M2G 1.5% agarose pads and imaged as described in (Tsokos et al., 2011).
P-galactosidase assays
Strains were grown in mid-exponential phase in medium supplemented with I tg/mL
oxytetracycline. Assays were performed essentially as in (Miller, 1972).
Immunoblots
Immunoblotting was performed as in (Modell et al., 2011): samples were normalized in
20 [tL 1:4 sample buffer:dH 20 to OD600=0.2, resolved on 12% sodium dodecyl sulfatepolyacrylamide gels, and transferred to polyvinylidene difluoride transfer membrane
(Pierce). Membranes were probed with monoclonal mouse cc-Flag (Sigma) at a 1:1000
dilution. Secondary HRP-conjugated a-mouse Pierce) was used at a 1:3000 dilution.
74
ChIP-Seq and analysis
Chromatin immunoprecipitation for ChIP-Seq was performed as in (Fioravanti et al.,
2013), with modifications: mid-exponential cultures were cross-linked in 10 mM sodium
phosphate (pH 7.6) and 1% formaldehyde at room temperature for 10 minutes. Reactions
were quenched with 0.1 M glycine at room temperature for 5 minutes, and on ice for 15
minutes. Cells were washed 3X in phosphate buffered saline (PBS) and lysed with
Ready-Lyse lysozyme solution (Epicentre, Madison, WI) according to manufacturer
instructions. Lysates were diluted 1:1 in ChIP buffer (1.1% Triton X- 100, 1.2 mM
EDTA, 16.7 mM Tris-HCl (pH 8.1), 167 mM NaCl) plus Roche Protease Inhibitor
Tablets (Roche) and incubated at 370 C for 10 minutes. Lysates were sonicated (Branson
Sonicator) on ice for 6 bursts of 10 seconds each at 15% amplitude, and then cleared by
centrifugation at 14,000 rpm for 5 minutes at
0
C CilevA spernta-tsif
nrnorm1i'7 eA
by protein content in I mL of ChIP buffer + 0.0 1% SDS, and pre-cleared with 50 [tL of
Protein-A DynaBeads (Invitrogen) (pre-blocked with 100 ig Ultra Pure BSA in ChIP
buffer + 0.01% SDS) by 1 hour rotation at 40 C.
10% of each supernatant was removed and used as total chromatin input sample. The
remaining supernatant was incubated with a 1:1000 dilution of anti-M2 antibody
overnight at 40 C. Each sample was then incubated with 50 [L of pre-blocked Protein-A
DynaBeads for 6 hours at 40 C with rotation. DynaBeads were washed consecutively at
40 C for 15 minutes with 1 mL of the following buffers: low salt wash buffer (0.1% SDS,
1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), 150 mM NaCl), high salt
wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl (pH 8.1), 500
mM NaCl), LiCl wash buffer (0.25 M LiCl, 1% NP-40, 1% deoxycholate, 1 mM EDTA,
75
10 mM Tris-HCl (pH 8.1)), and twice with TE buffer (10 mM Tris-HCl (pH 8.1), 1 mM
EDTA). Complexes were eluted twice by incubation with 250 [tL freshly prepared
elution buffer (1% SDS, 0.1 M NaHCO 3) at 300 C for 15 minutes.
Cross-links were reversed by addition of 300 mM NaCl and 2 [LL of 0.5 mg/mL RNase A
and overnight incubation at 65' C. Samples were treated with 5 [tL of Proteinase K (20
mg/mL, NEB) in 40 mM EDTA and 40 mM Tris-HFCl (pH 6.8) for 2 hours at 450 C.
DNA was extracted using a PCR purification kit (Qiagen) and resuspended in 80 [tL of
water.
Library was prepared using the SPRI-works system and sequenced on an Illumina MiSeq
(MIT BioMicroCenter). ChIP-Seq results were analyzed using MACS software analysis
package (Zhang et al., 2008). 860,000 reads were analyzed for each sample, and peaks
were called with a p-value < 10-5.
DNA microarrays
RNA was collected from cultures grown to mid-exponential phase in rich medium at 300
C. For the pstS mutant, wild-type was used as a reference. For the phoU depletion, the
phoU depletion strain grown in the absence of vanillate was compared to the same strain
grown in the presence of vanillate. Gene expression profiles were obtained as described
previously (Tsokos et al., 2011) using custom 8x1 5K Agilent expression arrays.
Complete data sets are provided in Appendices I and 2.
Colony forming units
Strains were grown overnight in PYE with vanillate, and then washed and released into
medium with or without vanillate. Cultures were subsequently grown for 30 hours, and
76
diluted once every two doubling times to maintain mid-exponential growth. Samples
were removed every three hours and plated on PYE + vanillate. Colonies were counted
after two days of growth for all strains except those with a pstS mutation, which have a
growth defect and were counted after three days growth.
Transposon mutagenesis and rescue cloning
50 mL of electrocompetent Pan-phoU;AphoUcells were transformed with 0.5 [L of EZTn5 transposon mix (EZ-Tn5 <R6Kyori/KAN-2> Insertion Kit, Epicentre,
www.epibio.com), and outgrown in I mL of PYE for 1.5 hours at 30' C. Cells were then
plated on PYE supplemented with kanamycin. Colonies were picked after 2, 3, and 4
days growth at 30' C.
Colonies were restruck onto fresh PYE+kanamycin plates and chromosomal DNA
subsequently prepared from single colonies cultured in PYE+kanamycin. DNA was
digested with BfuCI for 2 minutes at room temperature and 20 minutes at 800 C to yield
approximately 5 kb fragments. Sheared DNA was ligated with T4 DNA ligase and the
reaction was dialyzed for 1 hour using .45 tM nitrocellulose filters (Millipore) From
each dialyzed ligation reaction 1.5 [IL was electroporated into 25 [tL pir- 116 cells and
plated on LB medium supplemented with kanamycin. DNA was extracted from the
resulting colonies and sequenced using KAN-2 FP- 1 and R6KAN-2 RP-1 primers
(Epicentre, www.epibio.com).
77
References
Baek, J. H., Y. J. Kang, and S. Y. Lee, 2007, Transcript and protein level analyses of the
interactions among PhoB, PhoR, PhoU and CreC in response to phosphate
starvation in Escherichiacoli: FEMS Microbiology Letters, v. 277, p. 254-259.
Baek, J. H., and S. Y. Lee, 2006, Novel gene members in the Pho regulon of Escherichia
coli: FEMS Microbiology Letters, v. 264, p. 104-109.
Bailey, T. J., M. Boden, F. A. Buske, M. Frith, C. E. Grant, L. Clementi, J. Ren, W. W.
Li, and W. S. Noble, 2009, MEME SUITE: tools for motif discovery and
searching: Nucleic Acids Research, v. 37, p. 7.
Bertrand, N., S. Houle, G. LeBihan, E. Poirier, C. M. Dozois, and J. Harel, 2010,
Increased Pho regulon activation correlates with decreased virulence of an avian
pathogenic Escherichiacoli 078 strain: Infect Immun, v. 78, p. 5324-5331.
Blanco, A. G., A. Canals, J. Bernues, M. Sola, and M. Coll, 2011, The structure of a
transcription activation subcomplex reveals how sigma-70 is recruited to PhoB
promoters: EMBO Journal.
Blanco, A. G., M. Sola, F. X. Gomis-Ruth, and M. Coll, 2002, Tandem DNA recognition
by PhoB, a two-component signal transduction transcriptional activator: Structure,
v. 10, p. 701-713.
Capra, E. J., and M. T. Laub, 2012, Evolution of two-component signal transduction
systems: Annual Review of Microbiology, v. 66, p. 23.
Contreras, I., R. A. Bender, J. Mansour, S. Henry, and L. Shapiro, 1979, Caulobacter
crescentus mutant defective in membrane phospholipid synthesis: Journal of
Bacteriology, v. 140, p. 8.
Cox, G. B., D. Webb, J. Godovac-Zimmermann, and H. Rosenberg, 1988, Arg-220 of the
PstA protein is required for phosphate transport through the phosphate-specific
transport system in Escherichiacoli but not for alkaline phosphatase repression:
Journal of Bacteriology, v. 170, p. 4.
Cox, G. B., D. Webb, and H. Rosenberg, 1989, Specific amino acid residues in both the
PstB and PstC proteins are required for phosphate transport by the Escherichia
coli Pst system: Journal of Bacteriology, v. 171, p. 4.
Fang, G., K. Passalacqua, J. Hocking, P. M. Llopis, M. Gerstein, N. H. Bergman, and C.
Jacobs-Wagner, 2013, Transcriptomic and phylogenetic analysis of a bacterial
cell cycle reveals strong associations between gene co-expression and evolution:
BMC Genomics, v. 14.
Fioravanti, A., C. Fumeaux, S. S. Mohapatra, C. Bompard, M. Brilli, A. Frandi, V.
Castric, P. H. Viollier, and E. G. Biondi, 2013, DNA Binding of the Cell Cycle
Transcriptional Regulator GcrA Depends on N6-Adenosine Methylation
in Caulobactercrescentus and other Alphaproteobacteria:PLoS Genetics, v. 9, p.
1.
Galperin, M. Y., 2010, Diversity of structure and function of response regulator output
domains: Current Opinion in Microbiology, v. 13, p. 10.
Gober, J. W., and L. Shapiro, 1992, A Developmentally Regulated CaulobacterFlagellar
Promoter is Activated by 3' Enhancer and IHF Binding Elements: Molecular
Biology of the Cell, v. 3, p. 14.
78
Gonin, M., E. M. Quardokus, D. O'Donnol, J. Maddock, and Y. V. Brun, 2000,
Regulation of stalk elongation by phosphate in Caulobactercrescentus: Journal of
Bacteriology, v. 182, p. 337-347.
Hirota, R., K. Motomura, S. Nakai, T. Handa, T. Ikeda, and A. Kuroda, 2013, Stable
polyphosphate accumulation by a pseudo-revertant of an Escherichiacoli pho U
mutant: Biotechnology Letters, v. 35, p. 7.
Hsieh, Y. J., and B. L. Wanner, 2010, Global regulation by the seven-component Pi
signaling system: Curr Opin Microbiol, v. 13, p. 198-203.
Jacobsen, S. M., M. C. Lane, J. M. Harro, M. E. Shirtliff, and H. L. T. Mobley, 2008, The
high-affinity phosphate transporter Pst is a virulence factor forProteus
mirabilis during complicated urinary tract infection: FEMS Immunology and
Medical Microbiology, v. 52, p. 14.
Kashihara, H., B. M. Kang, T. Omasa, K. Honda, Y. Sameshima, A. Kuroda, and H.
Ohtake, 2010, Electron microscopic analysis of heat-induced leakage of
polyphosphate from a phoU mutant of Escherichiacoli, Biosci Biotechnol
Biochem, p. 865-868.
Klein, E. A., S. Schlimpert, V. Hughes, Y. V. Brun, M. Thanbichler, and Z. Gitai, 2013,
Physiological role of stalk lengthening in Caulobactercrescentus:
Communicative & Integrative Biology, v. 6, p. 1.
Koebnik, R., 2005, TonB-dependent trans-envelope signalling: the exception or the rule?:
Trends in Microbiology, v. 13, p. 5.
Krell, T., J. Lacal, A. Busch, H. Siva-Jimenez, M.-E. Guazzaroni, and J. L. Ramos, 2010,
Bacterial sensor kinases: diversity in the recognition of environmental signals:
Annual Review of Microbiology, v. 64, p. 21.
Lamarche, M. G., B. L. Wanner, S. Crepin, and J. Harel, 2008, The phosphate regulon
and bacterial virulence: a regulatory network connecting phosphate homeostasis
and pathogenesis: FEMS Microbiology Reviews, v. 32, p. 13.
Li, Y., and Y. Zhang, 2007, PhoU Is a Persistence Switch Involved in Persister Formation
and Tolerance to Multiple Antibiotics and Stresses in Escherichiacoli:
Antimicrobial Agents and Chemotherapy, v. 51, p. 8.
Liu, J., Y. Lou, H. Yokota, P. D. Adams, R. Kim, and S. H. Kim, 2005, Crystal structure
of a PhoU protein homologue: a new class of metalloprotein containing
multinuclear iron clusters: Journal of Biological Chemistry, v. 280, p. 1596015966.
Mack, T. R., R. Gao, and A. M. Stock, 2009, Probing the Roles of the Two Different
Dimers Mediated by the Receiver Domain of the Response Regulator PhoB:
Journal of Molecular Biology, v. 389, p. 16.
Makino, K., M. Amemura, S. K. Kim, A. Nakata, and H. Shinagawa, 1993, Role of the
sigma 70 subunit of RNA polymerase in transcriptional activation by activator
protein PhoB in Escherichiacoli: Genes & Development, v. 7, p. 149-160.
Metcalf, W. W., P. M. Steed, and B. L. Wanner, 1990, Identification of phosphate
starvation-inducible genes in Escherichiacoli K-12 by DNA sequence analysis of
psi::/acZ(Mu d]) transcriptional fusions: Journal of Bacteriology, v. 172, p. 31913200.
Miller, J. H., 1972, Experiments in molecular genetics: Cold Spring Harbor, N.Y., Cold
Spring Harbor Laboratory, 466 p.
79
Modell, J. W., A. C. Hopkins, and M. T. Laub, 2011, A DNA damage checkpoint in
Caulobactercrescentus inhibits cell division through a direct interaction with
FtsW: Genes & Development, v. 25, p. 16.
Morohoshi, T., T. Maruo, Y. Shirai, J. Kato, T. Ikeda, N. Takiguchi, H. Ohtake, and A.
Kuroda, 2002, Accumulation of inorganic polyphosphate in pho U mutants of
Escherichiacoli and Synechocystis sp. strain PCC6803: Applied and
Environmental Microbiology, v. 68, p. 4.
Motomura, K., R. Hirota, N. Ohnaka, M. Okada, T. Ikeda, T. Morohoshi, H. Ohtake, and
A. Kuroda, 2011, Overproduction of YjbB reduces the level of polyphosphate in
Escherichiacoli: a hypothetical role of YjbB in phosphate export and
polyphosphate accumulation: FEMS Microbiology Letters, v. 320, p. 8.
Muda, M., N. N. Rao, and A. Torriani, 1992, Role of PhoU in phosphate import and
alkaline phosphatase regulation: Journal of Bacteriology, v. 174, p. 8.
Oganesyan, V., N. Oganesyan, P. D. Adams, J. Jancarik, H. A. Yokota, R. Kim, and S. H.
Kim, 2005, Crystal structure of the "PhoU-like" phosphate uptake regulator from
Aquifex aeolicus: Journal of Bacteriology, v. 187, p. 4238-4244.
O'May, G. A., S. M. Jacobsen, M. Longwell, P. Stoodley, H. L. Mobley, and M. E.
Shirtliff, 2009, The high-affinity phosphate transporter Pst in Proteusmirabilis
H14320 and its importance in biofilm formation: Microbiology, v. 155, p. 15231535.
Pratt, J. T., A. M. Ismail, and A. Camilli, 2010, PhoB regulates both environmental and
virulence gene expression in Vibrio cholerae: Molecular Microbiology, v. 77, p.
1595-1605.
Rao, N. N., M. R. G6mez-Garcia, and A. Kornberg, 2009, Inorganic polyphosphate:
essential for growth and survival: Annual Review of Biochemistry, v. 78, p. 43.
Rice, C. D., J. E. Pollard, Z. T. Lewis, and W. R. McCleary, 2009, Employment of a
promoter-swapping technique shows that PhoU modulates the activity of the
PstSCAB 2 ABC transporter in Escherichiacoli: Applied and Environmental
Microbiology, v. 75, p. 10.
Schmidt, J. M., and R. Y. Stanier, 1966, The development of cellular stalks in bacteria:
The Journal of Cell Biology, v. 28, p. 14.
Skerker, J. M., M. Prasol, B. Perchuk, E. Biondi, and M. T. Laub, 2005, Two-component
signal transduction pathways regulating growth and cell cycle progression in a
bacterium: a systems-level analysis: PLoS Biology, v. 3, p. 334-353.
Steed, P. M., and B. L. Wanner, 1993, Use of the rep technique for allele replacement to
construct mutants with deletions of the pstSCAB-pho U operon: evidence of a new
role for the PhoU protein in the phosphate regulon: Journal of Bacteriology, v.
175, p. 6797-6809.
Stock, A. M., V. L. Robinson, and P. N. Goudreau, 2000, Two-component signal
transduction: Annual Review of Biochemistry, v. 69, p. 33.
Thanbichler, M., A. A. Iniesta, and L. Shapiro, 2007, A comprehensive set of plasmids
for vanillate- and xylose-inducible gene expression in Caulobactercrescentus:
Nucleic Acids Research, v. 35, p. 1.
Tsokos, C. G., B. S. Perchuk, and M. T. Laub, 2011, A dynamic complex of signaling
proteins uses polar localization to regulate cell-fate asymmetry in Caulobacter
crescentus: Developmental Cell, v. 20, p. 13.
80
Wanner, B. L., and P. Latterell, 1980, Mutants affected in alkaline phosphatase
expression: evidence for multiple positive regulators of the phosphate regulon in
Escherichiacoli: Genetics, v. 96, p. 353-366.
Yang, C., T.-W. Huang, S.-Y. Wen, C.-Y. Chang, S.-F. Tsai, W.-F. Wu, and O.-H.
Chang, 2012, Genome-Wide PhoB Binding and Gene Expression Profiles Reveal
the Hierarchical Gene Regulatory Network of Phosphate Starvation inEscherichia
coli PLoS ONE, v. 7, p. 1.
Zhang, Y., T. Liu, C. A. Meyer, J. Eeckhoute, D. S. Johnson, B. E. Bernstein, C.
Nusbaum, R. M. Myers, M. Brown, W. Li, and X. S. Liu, 2008, Model-based
analysis of ChIP-Seq (MACS): Genome Biology, v. 9, p. 1.
81
Chapter 3: Conclusions and Future Work
82
In this work, I used Caulobactercrescentus as a model to understand the bacterial
response to phosphate limitation. I mapped the global binding patterns of PhoB, the
transcriptional activator of the phosphate limitation response, in phosphate-starved and replete conditions, and defined the Pho regulon in Caulobactercrescentus. I examined
the role of PhoU, an uncharacterized member of the Pst/Pho operon. I tested its proposed
function as a negative regulator of the PhoR/PhoB system, and demonstrated that it
instead functions outside this pathway, likely in a process central to phosphate
metabolism.
The Pho regulon in Caulobacter crescentus
Pho regulon specialization in Caulobacter
Although the CaulobacterPho regulon shares some functional overlap with that in E. coli
(Chapter 2, Discussion), there are notable points of specialization within each regulon.
One of the largest subsets of the CaulobacterPho regulon is TonB-dependent receptors.
The Caulobactergenome codes for 60 such receptors, whereas E. coli, whose genome
has only 9 annotated TonB-dependent receptors (Koebnik, 2005), instead regulates
transcription of a smaller and more diverse set of transporters in low phosphate
conditions (Yang et al., 2012). While the functions of many of these low phosphateregulated transporters in Caulobacterare unknown, E. coli regulates genes controlling
import of phosphate- and carbon-containing compounds, including components of the
PTS system, in response to phosphate limitation. Although there are two duplicated PTS
systems annotated in the Caulobactergenome (Nierman et al., 2001), we did not find
83
these to be PhoB regulated genes, which may indicate that carbon import is a more
common mechanism to respond to phosphate starvation in E. coli than in Caulobacter.
Other important nutrient response pathways, such as carbon, nitrogen, and magnesium,
have not been as well characterized in Caulobacter;analysis of the global binding
patterns of regulators that control these responses will help determine whether regulation
of a range of transporters is a general strategy Caulobacteremploys to adapt to different
types of starved conditions. However, the annotated TonB-dependent system that is most
strongly regulated by PhoB in Caulobacterhas been identified as a system for secreting
an extracellular lipoprotein (Le Blastier et al., 2010), suggesting that these genes may
have more diverse functions than previously supposed.
What genes regulate stalk elongation in Caulobacter?
The most striking example of Pho regulon specialization in Caulobacteris that of stalk
elongation. As discussed in Chapter 2, a handful of cell wall and membrane-modifying
genes are regulated by PhoB. These and several other candidate regulators of stalk
elongation are shown in Table 3.1. PhoB may regulate a regulator that in turn controls
stalk elongation. The CaulobacterPho regulon includes several transcription factors that
could fill this role; additionally, since at least one TonB dependent system has been
identified as a secretion system in Caulobacter,it is possible that a TonB system could
export components needed to elongate the polar stalk. Finally, a number of hypothetical
proteins that are activated by PhoB and conserved only in Caulobacterand closely
related species are candidates for a regulator of stalk elongation.
84
Table 3.1 - Candidate regulators of stalk elongation in Caulobacter
ChIP peak fold-enrichment
pstS.::Th
PYE
MSG
Gene #
Annotation
CC3301
Peptidoglycan-specific endopeptidase, M23 family
0
19.5
10.2
CC2967
Cell wall hydrolase family protein
0
13.81
6.12
CC3094
Sensory box/GGDEF family protein
0
34.22
19.63
CC1791
TonB-dependent receptor
0
28.88
20.93
CC2819
TonB-dependent receptor
2.89
26.61
22.85
CC1970
TonB-dependent receptor
0
16.4
7.95
CC2923
TonB-dependent receptor
0
15.29
13.17
CC0210
TonB-dependent receptor
0
15.14
7.92
CC1099
TonB-dependent outer membrane receptor
0
9.23
7.08
CC1666
TonB-dependent outer membrane receptor
0
7.99
4.93
CC2336
TonB accessory protein exbB
0
7.63
6.78
CC0996
TonB-dependent receptor
0
52.11
25.55
CC0722
TonB-dependent receptor
0
41.21
33.76
CC1182
Two-component response regulator
0
10.67
7.32
CC3395
Transcriptional regulator, algH
0
8.05
4.53
CC2225
Conserved hypothetical protein
0
19.11
10.48
CC3014
Hypothetical protein
0
17.93
10.84
CC0316
Hypothetical protein
0
13.98
7.37
CC3171
Hypothetical protein
0
12.15
10.92
CC3706
Hypothetical protein
0
11.04
6.37
CC0162
Hypothetical protein
0
9.21
0
CC0551
Hypothetical protein
0
7.84
8.33
CC2017
Conserved hypothetical protein
0
7.8
5.56
CC1820
Hypothetical protein
0
7.28
7.24
CC3330
Hypothetical protein
0
7.25
4.52
PYE
Another likely candidate for a regulator of stalk elongation is the gene CC3094, which is
annotated as encoding a GGDEF/EAL family protein. This family of proteins is
responsible for the synthesis and degradation of the signaling molecule cyclic-di-GMP:
GGDEF domains synthesize cyclic-di-GMP from two GTP molecules, while EAL
domains hydrolyze cyclic-di-GMP to its linear form, pGpG (Jenal and Malone, 2006).
During the course of the Caulobactercell cycle, stalk synthesis is stimulated by the
increased production of cyclic-di-GMP at the stalked pole (Abel et al., 2011). An
85
attractive model for phosphate-dependent stalk elongation is, then, that c-di-GMP levels
are again increased at the stalked pole, activating the same mechanism that elongates
stalks in response to cell cycle cues. This proposed increase in c-di-GMP levels could be
effected by CC3094. To test this hypothesis, we obtained a null allele of CC3094 (Urs
Jenal, unpublished). We found that deleting the CC3094 gene has no effect on stalk
elongation, either in low phosphate medium or in the pstS-: Tn5 mutant background (data
not shown).
As ChIP-Seq has identified a set of promising candidates for regulators of stalk
elongation, we will make deletion strains of each of these to test their function in rich and
low phosphate medium. If deletion of these candidates has no effect on stalk elongation, a
screen could be used to identify stalk elongation regulators. Other studies have used the
fact that the buoyant density of Caulobacteris lowered by increased stalk length to
isolate mutants that have hyperelongated stalks. This strategy allowed for the isolation of
the CaulobacterpstS mutant, among others (Gonin et al., 2000). To identify factors that
promote stalk elongation, we could mutagenize the pstS mutant strain, employ density
gradient centrifugation to isolate cells in which the long stalk phenotype is now
suppressed, and map the resulting mutations.
How does Caulobacter integrate the response to phosphate limitation with
two different cell types?
Although elongated stalks are a clear marker of phosphate limitation, they are not found
on all, or even a clear majority, of phosphate-limited Caulobactercells. Caulobacter
crescentus has two distinct cell types: the sessile stalked cell, and motile swarmer cell
(Figure 3.1). The swarmer cell bears a polar flagellum, and can chemotax towards
nutrients. The stalked cell has a polar stalk with a holdfast at the distal tip, allowing it to
86
adhere to surfaces (Ong et al., 1990; Tsokos and Laub, 2012). Although the Caulobacter
growth rate slows during phosphate limited conditions, a mixed population of cell types
is observed (data not shown); it is unknown whether the ratio of swarmer to stalked cells
changes upon phosphate limitation, but it seems possible, as numerous other species bias
their lifestyles in starved conditions; in Pseudomonasfluorescens,for example, biofilm
formation is increased in phosphate-limited conditions (Monds et al., 2006; Newell et al.,
2011). To determine whether the ratio of stalked to swarmer cells is altered upon
phosphate limitation in Caulobacter,we will introduce cell-type specific reporters.
Although we may find that the ratio of cell types is biased toward stalked cells, a study of
carbon starvation in Caulobactersuggests that expression changes in the two cell types
may be an interesting area for further study. This study found that some starvation
response genes are differentially regulated between the stalked and swarmer cell types
(Britos et al., 2011).
swarmer/G1
stalked/S
predivisional/S
Figure 3.1 - The Caulobacter crescentus cell cycle
Caulobacterhas two cell types, swarmer and stalked. Swarmer cells have a polar flagellum, are free-swimming and
arrested in GI phase. These cells can undergo a transition to the stalked cell type, forming a stalk and undergoing
DNA replication and division into two distinct daughter cells, one swarmer, and one stalk.
The experiments presented in Chapter 2 were performed on mixed populations of cells.
Repeating the microarray and ChIP-Seq examination of gene expression and PhoB
binding patterns in isolated swarmer and stalked cell populations would allow us to better
87
understand the extent to which the phosphate limitation response is tailored to a particular
cell type.
One indication that phosphate limitation may have a swarmer cell-specific function and
not solely a stalked cell one comes from assays of swarming motility. When placed on
low agar plates, Caulobacterswarms outward. Changes in swarm size can indicate a
perturbation in the cell cycle, or an alteration of chemotaxis ability (Skerker et al., 2005).
While mutation ofpstS in Caulobacterleads to hyperactivation of the Pho regulon and
thus stalk hyperelongation (Gonin et al., 2000), we have found that pstS mutation also
leads to a defect in swarming motility. The pstS mutant displays a larger, more diffuse
swarm phenotype, compared to wild-type cells (Figure 3.2).
Proteins that synthesize or degrade c-di-GMP have been found to regulate the motile-tosessile transition in other bacteria (Jenal and Malone, 2006; Simm et al., 2004); as we
found that CC3094 does not have a function in regulating stalk elongation in
Caulobacter,we hypothesized that this protein might play a role in regulating the motile
state in this bacterium.
We found that although a deletion of CC3094 alone had no effect on either stalk
elongation or swarming motility, deletion of CC3094 in the pstS mutant background
resulted in a small swarm phenotype (Figure 3.2). This result suggests a model in which
CC3094 is responsible for promoting swarming motility in low phosphate conditions.
This model predicts that the deletion of CC3094 should suppress swarming motility on
low phosphate swarm plates; however, this is technically difficult to test. It remains to be
tested whether CC3094 is expressed differentially in the two cell types in phosphate
limited medium.
88
WT
pstS::Tn5
ACC3094
ACC3094;pstS::Tn5
Figure 3.2 - Deletion of CC3094 suppresses the pstS mutant large swarm
The noted strains were innoculated into 0.3% agar plates and grown at 300 C for 4 days to assay swarming
motility. Mutation of pstS results in a diffuse swarm phenotype, while deletion of CC3094 alone has no effect
on swarm phenotype on rich medium. Tthe double pstS::Tn5;AC3094 mutant has a small swarm, indicating
that CC3094 is responsible for promoting swarming motility in response to activation by PhoB.
What other mechanisms does Caulobacter use to respond to
phosphate limitation?
Although tens, and possibly hundreds, of genes are directly regulated by PhoB in
response to phosphate limitation, a number of other gene expression changes are
observed upon phosphate limitation. Some appear to be secondary effects that are
indirectly dependent on phoB, rather than direct targets. Others appear to be entirely
independent ofphoB.
Indirect PhoB targets
Comparison of ChIP-Seq data with several microarray profiles obtained in different
phosphate-limited conditions can allow us to better classify indirect PhoB targets as well
as PhoB-independent genes. To identify indirect targets of PhoB, we examined genes that
were upregulated in a pstS mutant and that did not contain peaks in the PhoB ChIP data
(Chapter 2). We also compared the ChiP results to another published microarray data set
from the lab (Capra et al., 2012), in which gene expression in a phoR mutant is compared
to wild-type in low phosphate and high phosphate medium; in this dataset, genes which
are downregulated in low phosphate medium represent targets of the phoR/B system. The
89
resulting set of genes includes those involved in responding to oxidative stress, as well as
genes encoding ribosomal proteins and additional phosphate scavenging genes. A subset
of these genes is given in Table 3.2.
Table 3.2 - Indirect PhoB-regulated genes
Microarray data are given as Iog2. ratios
PstS:':Tfts/WT
in rich
pstS:*:TnS;AphoB1WT
medium
in rich medium
Gene #
Annotation
CC0995
Peptide methionine sulfoxide reductase msrA
6.051999333
-0.07575657
CC2924
Conserved hypothetical protein
3.857865826
0.580230107
CC1277
phosphodiesterase I / nucleotide pyrophosphatase
2.297777663
0.1788563
CC0844
Hypothetical protein
2.20692293
-0.11193652
CC1777
Superoxide dismutase
2.000797292
-1.686875087
CC3031
Non-hemolytic phospholipase C
1.698086594
-0.104225494
CC1781
TonB-dependent receptor
1.628852075
-2.477420153
CC1776
Transcriptional regulator, GntR family
1.386810069
-1.161440974
AphoR/WT
AphoR/WT in10 mM
Gene #
Annotation
in 50 pM Pi
Pi
CC3172
Glycerophosphoryl diester phosphodiesterase
-4.836476
0.396620838
CC3031
Non-hemolytic phospholipase C
-3.007872947
0.241721691
CC2027
TonB-dependent outer membrane receptor
-2.980794262
0.503564389
CC1254
SSU
-2.832312358
-0.167875838
CC0496
LSU ribosomal protein L10P
-2.745336787
0.11754357
CC1375
dTDP-glucose 4,6-dehydratase
-2.736032386
0.15625205
CC0289
Phosphate regulon sensor protein phoR
-2.658887733
-2.12611176
CC1250
LSU ribosomal protein L23P
-2.600328236
0.172595505
ribosomal protein S3P
The PhoB-independent phosphate response
Of particular interest are those genes that appear entirely phoB-independent. One way to
obtain this set is to examine genes up- (or down-) regulated in both a pstS mutant and a
phoB mutant. Phosphate import is blocked in both mutants (Gonin et al., 2000). Thus,
both cells are phosphate-starved, even in rich medium, but have opposite effects on Pho
regulon expression, eliminating genes that are primarily phoB-regulated from
consideration.
90
Examination of this set ofphoB-independent genes (Table 3.3) may reveal interesting
factors for further study. One example is phoH (discussed below), which may function in
multiple starvation responses. A number of genes encoding TonB-dependent receptors
and hypothetical proteins may also function in a phoB-independent response to phosphate
limitation.
Table 3.3 - PhoB-independent genes
Microarray data are logz ratios
Gene #
Annotation
AphoBlwt
PStS::Tn51Wt
AphoBapsts::
Tnslwt
CC1401
cytochrome c oxidase, CcoN subunit
2.856071651
2.480663053
2.255284442
CC0559
2.758075058
2.342942654
1.749223053
CC0584
hypothetical protein
succinylornithine transaminase,
putative
2.513915048
1.382388698
1.347142344
CC0683
HIyD family secretion protein
2.409116113
0.878749236
1.65752615
CC0027
conserved hypothetical protein
2.328817247
1.891461456
1.422493383
CC1404
cytochrome c oxidase, CcoP subunit
2.258629601
1.958824105
2.082846649
CC2644
PhoH-related protein
2.194834042
2.48808386
1.730049808
CC0210
TonB-dependent receptor
2.187539396
1.746816766
0.06849114
CC3263
conserved hypothetical protein
2.07446333
2.365164219
1.955231376
CC2194
TonB-dependent receptor
2.038394298
1.913780129
1.722405222
CC0581
arginine N-succinyltransferase
2.032575441
0.981988002
1.07291693
CC3060
conserved hypothetical protein
1.873367308
1.502859496
1.391471415
CC2928
TonB-dependent receptor
1.869275335
1.252542659
1.542473009
CC3291
hypothetical protein
1.794132457
1.375430825
1.301927043
CC2534
histidinol-phosphate aminotransferase
1.765496764
1.288274349
2.046547794
CC2645
hypothetical protein
1.653637462
1.165703258
0.26539645
CC0682
hypothetical protein
1.528742267
1.471871248
1.139793149
CC0681
hypothetical protein
1.416392097
1.436464216
1.14321183
CCO557
hypothetical protein
1.298510506
0.694044949
1.968088333
CC2159
hypothetical protein
1.29366102
0.734183316
1.153038966
0.242369903
CC0669
hypothetical protein
1.22155875
1.317663724
CC0277
hypothetical protein
1.193379429
0.743864872
1.161451318
CC2532
homogentisate 1,2-dioxygenase
1.18724203
0.771879716
1.101365809
CC3574
alanine dehydrogenase
1.111340426
0.804004795
1.150187132
CC0712
ferrous iron transport protein B
1.105256446
1.097889831
1.241959005
CC0167
hypothetical protein
1.094100742
1.25510933
0.199891198
CC1357
serine hydroxymethyltransferase
1.050175983
1.187378893
0.640228858
91
IfphoB does not account for part of the response to phosphate limitation in Caulobacter,
what does? One possibility is that some of these phoB-independent genes comprise the
general stress response in Caulobacter,which is currently poorly defined. Although it is
known that iraP is responsible for integrating phosphate starvation with the general stress
in E. coli, a similar gene has not been identified in Caulobacter.Further investigation
into the set ofphoB-independent genes may identify the factor(s) responsible for inducing
the general stress response in phosphate-limited conditions. Some of the phoBindependent genes, such as phoH and CC329 1, encoding a hypothetical protein, are
upregulated by carbon starvation as well (Britos et al., 2011), making them likely
candidates for members of this response.
phoH: an uncharacterized stress response gene
phoH is a widely conserved gene that was first identified as psiH, a phosphate-starvation
induced gene in E. coli (Metcalf et al., 1990), and subsequently was cloned and renamed
(Kim et al., 1993). The phoH gene encodes an approximately 39 kDa protein with
homology to AAA+ ATPases; its ATPase activity has been confirmed in E. coli (Kim et
al., 1993).
Despite its ubiquity and the importance of other well-characterized AAA+ ATPases such
as DnaA and ClpX, little is known about the role of PhoH. It appears to function in
phosphate limitation in diverse bacteria; phoH is encoded in genomes of a number of
viruses that infect marine bacteria. These viruses often also encode pst genes (Monier et
al., 2012). Viral encoding of the pst genes ensures that sufficient phosphorus is present to
support viral replication (Zeng and Chisholm, 2012); thus the presence ofphoH in viral
genomes is indicative of an important role in the host starvation response.
92
phoH may respond to multiple types of stress; in a study of gene expression changes
upon carbon starvation in Caulobactercrescentus, phoH (CC2644) was ~44-fold
upregulated. phoH is also regulated by CtrA, the master regulator of the cell cycle in
Caulobacter,and is cell-cycle regulated (Britos et al., 2011). To further investigate the
function ofphoH, we will delete and overexpress this gene, and characterize the
phenotype of the resulting strains in rich medium, as well in medium limited for various
nutrients.
What is the function of PhoU?
pho U is widely conserved and found encoded with the pst and pho systems in many
diverse bacteria, suggesting it plays an important role in bacterial phosphate metabolism.
Further, although Caulobacteris the first bacterium for which phoU has been found to be
essential, deletion ofphoU in E. coli results in frequent acquisition of suppressor
mutations (Steed and Wanner, 1993), underscoring the importance of this gene to
bacterial life.
The data shown in Chapter 2 support a function for PhoU in regulating or metabolizing
cellular phosphate, and suggest that phoU depletion is lethal because inorganic phosphate
is imported to toxic levels or is converted to a toxic form. In future work, we will test the
role of phosphate in phoU depletion lethality, and investigate specific hypotheses into
phoU function.
Phosphate-dependency of phoU depletion lethality
One possibility is that PhoU regulates activity of the Pst system, and that in the absence
of PhoU, excess phosphate is imported. The effect of PhoU on phosphate uptake has been
93
tested in E. coli, with conflicting results (Chapter 2, Discussion). We measured phosphate
import in Caulobacterupon pho U depletion and did not find that it was significantly
increased at depletion time points up to 8 hours (data not shown). We did observe a twofold increase in phosphate import at later (14 hour) depletion time points (data not
shown); however, lacZ reporter data indicate that transcription of the pst genes has
increased comparably by this time (Chapter 2, Figure 2.7B). Thus, we cannot rule out the
possibility that increased production of the Pst transporter is responsible for the observed
increase in phosphate import, and further study is needed to determine the effect of PhoU
on phosphate import.
We also intend to test the effect of PhoU on phosphate uptake in a system in which pst
system transcription is made independent of PhoB regulation. One study (Rice et al.,
2009) tested this scenario in E. coli and found an increase in phosphate uptake in a phoU
mutant. This result was promising; however, the observed increase in uptake was modest,
and was measured only in low phosphate medium, not in high phosphate medium in
which increased phosphate uptake is expected to have an adverse effect. Further, the
increase in phosphate uptake was not shown to depend on the pst system. We will attempt
to perform a similar line of experiments in Caulobacter,and address the concerns noted.
Regardless of whether PhoU functions in phosphate import or elsewhere, a model in
which phoU depletion is lethal because of excess or misdirected intracellular phosphate
predicts that the viability of aphoU depletion should be affected by the levels of
available extracellular inorganic phosphate. Accordingly, we examined the ability of the
phoU depletion strain to grow on minimal medium agar containing 50 [tM phosphate, a
relatively limited concentration of extracellular phosphate. Our preliminary results
94
suggest that this strain is viable in the absence of vanillate (i.e. after depletion of PhoU)
under these conditions, while it is not on PYE agar in the absence of vanillate (Fig-show
plates) where the extracellular phosphate concentration is likely much higher than 50 M.
Surprisingly, the phoU depletion strain was also viable when grown on plates containing
10 mM phosphate; as the concentration of phosphate in PYE, a complex rich medium, is
unknown, it is possible that the concentration of available phosphate in PYE is
significantly higher than 10 mM. One possibility is that the slow growth that results from
growth on minimal medium is responsible for restoring phoU viability rather than the
lowered phosphate levels. Such an effect has been observed for another putatively
essential Caulobactergene, gcrA, which is viable when grown on low phosphate medium
at 30' C or on rich medium at 170 C (Diane Baer, personal communication). However,
we found that the phoU mutant remains non-viable when grown on PYE agar at 170 C,
suggesting that it is minimal medium and low phosphate, rather than a slower growth
rate, which suppresses phoU depletion lethality (Figure 3.3). Another prediction of this
model is that increased phosphate levels will result in phoU depletion lethality in a dosedependent manner, which we will test in future experiments.
PYE
M5G
M20
PYE 18* C
+vanlllato
-vanillat
-
Figure 3.3 - Growth on minimal medium, but not at low temperature, suppresses phoU depletion lethality
The phoU depletion strain was grown at 300 C on agar plates made with rich medium (PYE), or minimal medium
supplemented with 10 mM or 50 pM phosphate, or at 18* C on agar plates made with PYE. All strains grew when
plates were supplemented with vanillate (top row), compared to plates without vanillate (bottom row).
95
PhoU and polyphosphate
In addition to testing the phosphate-dependence of pho U mutant lethality, we will
investigate the relationship ofphoU with polyphosphate. Studies in several other bacteria
have noted a remarkable accumulation of polyphosphate in phoU mutants (Morohoshi et
al., 2002). However, it remains unclear whether phoU mutants accumulate polyphosphate
due to increased phosphate uptake, or whether polyphosphate is directly regulated by
phoU. While the majority of our current data are consistent with either model, if
overproduction of polyphosphate is the reason for pho U depletion lethality, one would
expect that mutations in ppk genes would have been identified as suppressors ofpho U
(Chapter 2, Figure 2.7A). The absence of mutations in these genes may be due to a
potential functional redundancy between ppkl and ppk2 in Caulobacter,in which case
mutation of one would not rescue the phoU mutant.
Another prediction of this model is that overexpression of a ppk gene should phenocopy
the pho U mutant. ppkl has been overexpressed in Caulobacter,leading to the formation
of intracellular polyphosphate granules (Henry and Crosson, 2013). However, this strain
is viable, and the granules appear small relative to the granule accumulation we have
observed in the pho U mutant. Higher overexpression of ppkI may be required to observe
a comparable effect.
We will also assay directly whether the CaulobacterphoU mutant, like those in other
bacteria, accumulates polyphosphate. If polyphosphate is produced, and if it is
responsible for the lethality observed in the pho U mutant, what precisely is the
mechanism of cell death? One possibility is that excess polyphosphate sequesters metals,
drawing them away from essential cellular processes; this could explain the general
96
changes in expression of metabolic genes observed in Appendix 3 and in a microarray
analysis of a phoU mutant in E. coli.
Another possibility is that regulatory functions of polyphosphate are responsible for
phoU lethality. Polyphosphate has important regulatory roles in many bacteria. Most
notably, it stimulates activity of Lon protease in E. coli (Kuroda et al., 2001; Rao et al.,
2009). Work in E. coli has shown that overproduction of Lon is lethal (Christensen et al.,
2004; Goff and Goldberg, 1987); as Caulobacterencodes a Lon ortholog,
overstimulation of Lon activity through polyphosphate could be the cause ofphoU
depletion lethality. Polyphosphate has additional regulatory functions in Caulobacteras
well. One study found that polyphosphate can regulate the swarmer-to-stalked cell
transition in this bacterium; when polyphosphate levels are lowered by deletion ofppk1,
swarmer cells develop polar stalks and improperly initiate DNA replication (Boutte et al.,
2012). This is not entirely consistent with the phoU mutant phenotype. The phoU mutant
displays modest chromosome accumulation, the opposite of what would be predicted by
this model, although this phenotype appears after cell death has already begun (Chapter
2, Figure 2.5A).
In future work, we aim to test whether modulation of polyphosphate levels through
mutation of ppk or ppx genes affects the viability of the pho U mutant. Deletion of both
ppk genes, which is predicted to ablate polyphosphate production, should suppress pho U
depletion lethality. Further, deletion of the ppx genes in Caulobactershould result in
higher levels of polyphosphate and phenocopy the phoU mutant. Intriguingly, while
mutation of either ppk gene individually orppxl has little effect on Caulobacterviability,
deletion of ppx2 has been found to result in a severe growth defect (Christen et al., 2011).
97
A second notable characteristic ofphoU mutants is their apparent membrane
permeability. This could be consistent with a role in polyphosphate regulation, as
polyphosphate-polyhydroxybutyrate complexes are thought to increase membrane
permeability (Rao et al., 2009), but could also indicate a role for phoU in another
important phosphate-dependent process, production of phospholipids. One study in E.
coli found increased susceptibility of a pho U mutant to antibiotics and other compounds
and proposed a role for pho U in cell persistence (Li and Zhang, 2007), We have not
tested the permeability of the pho U mutant in Caulobacterdirectly, but have found that
cells become enlarged and display apparent membrane blebbing (Chapter 2, Figure 2.5).
Electron microscopy studies of this mutant would be informative, to better characterize
its membrane and granule accumulation phenotypes.
Regulation of PhoR
Our data have ruled out a function for PhoU in regulation of PhoR, leaving open the
question of how PhoR is regulated. One possibility is that the Pst system interacts directly
with PhoR to regulate its activity. PhoR contains a cytoplasmic PAS domain; as PAS
domains can act as points of regulation on kinase activity (Krell et al., 2010), this domain
could mediate a direct interaction between PhoR and the Pst system. In some ABC type
transport systems, the cytoplasmic ATPase subunits interact with and regulate other
factors (Rees et al., 2009). Thus, we hypothesized that an interaction between the Pst and
Pho systems might be mediated by PstB, the ATPase subunit of the Pst system, and
PhoR. We used a bacterial two-hybrid assay to test this hypothesis, but did not identify an
interaction between PhoR and PstB (data not shown). It is possible, however, that
98
introducing a tag onto PstB for use in the bacterial two-hybrid assay prevented it from
interacting with PhoR. In vitro studies using smaller tags may identify an interaction
between PhoR and PstB. Alternatively, the transmembrane components of the Pst system,
PstA and PstC, could interact with the transmembrane portion of PhoR to regulate
activity of PhoR.
Recent work in E. coli has suggested that PhoR activity is regulated by other nutrient
signals, in addition to phosphate limitation. The PTSNtr system, paralagous to the sugar
transport phosphotransferase system (PTS) has been implicated in potassium homeostasis
in E. coli and found to stimulate the histidine kinase KdpD, which regulates expression of
potassium-responsive genes. A component of the PTSNtr system, EIIANt, regulates
expression of tested Pho regulon genes in E . coli, and promotes phosphorylation of PhoB
by PhoR in vitro (Lfittmann et al., 2012), suggesting multiple nutrient responses may
integrated through regulation of PhoR activity.
99
References
Abel, S., P. Chien, P. Wassmann, T. Schirmer, V. Kaever, M. T. Laub, T. A. Baker, and U.
Jenal, 2011, Regulatory cohesion of cell cycle and cell differentiation through
interlinked phosphorylation and second messenger networks: Molecular Cell,
v. 43, p. 11.
Boutte, C. C., J. T. Henry, and S. Crosson, 2012, ppGpp and polyphosphate modulate
cell cycle progression in Caulobactercrescentus: Journal of Bacteriology, v.
194, p. 8.
Britos, L., E. Abeliuk, T. Taverner, M. Lipton, H. McAdams, and L. Shapiro, 2011,
Regulatory response to carbon starvation in Caulobactercrescentus: PLoS
One.
Capra, E. J., B. S. Perchik, J.M. Skerker, and M. T. Laub, 2012, Adaptive mutations that
prevent crosstalk enable the expansion of paralogous signaling protein
families: Cell, v. 150, p. 11.
Christen, B., E. Abeliuk, J. M. Collier, V. S. Kalogeraki, B. Passarelli, J. A. Coller, M. J.
Fero, H. H. McAdams, and L. Shapiro, 2011, The essential genome of a
bacterium: Molecular Systems Biology, v. 7, p. 7.
Christensen, S. K., G. Maenhaut-Michel, N. Mine, S. Gottesman, K. Gerdes, and L. Van
Melderen, 2004, Overproduction of the Lon protease triggers inhibition of
translation in Escherichiacoli: involvement of theyefM-yoeB toxin-antitoxin
system: Molecular Microbiology, v. 51, p. 13.
Goff, S. A., and A. L. Goldberg, 1987, An increased content of protease La, the Ion gene
product, increases protein degradation and blocks growth in Escherichiacoli:
The Journal of Biological Chemistry, v. 262, p. 8.
Gonin, M., E. M. Quardokus, D. O'Donnol, J.Maddock, and Y. V. Brun, 2000,
Regulation of stalk elongation by phosphate in Caulobactercrescentus:
Journal of Bacteriology, v. 182, p. 337-347.
Henry, J.T., and S. Crosson, 2013, Chromosome replication and segregation govern
the biogenesis and inheritance of inorganic polyphosphate granules:
Molecular Biology of the Cell, v. 24, p. 10.
Jenal, U., and J.Malone, 2006, Mechanisms of cyclic-di-GMP signaling in bacteria:
Annual Review of Genetics, v. 40, p. 23.
Kim, S.-k., K. Makino, M. Amemura, H. Shinagawa, and A. Nakata, 1993, Molecular
analysis of the phoH gene, belonging to the phosphate regulon in Escherichia
coli: Journal of Bacteriology, v. 175, p. 9.
Koebnik, R., 2005, TonB-dependent trans-envelope signalling: the exception or the
rule?: Trends in Microbiology, v. 13, p. 5.
Krell, T., J. Lacal, A. Busch, H. Siva-Jimenez, M.-E. Guazzaroni, and J. L. Ramos, 2010,
Bacterial sensor kinases: diversity in the recognition of environmental
signals: Annual Review of Microbiology, v. 64, p. 21.
Kuroda, A., K. Nomura, R. Ohtomo, J. Kato, T. Ikeda, N. Takiguchi, H. Ohtake, and A.
Kornberg, 2001, Role of inorganic polyphosphate in promoting ribosomal
protein degradation by the Lon protease in E. coli: Science, v. 293, p. 4.
100
Le Blastier, S., A. Hamels, M. Cabeen, L. Schille, F. Tilquin, M. Dieu, M. Raes, and J.-Y.
Matroule, 2010, Phosphate starvation triggers production and secretion of an
extracellular lipoprotein in Caulobactercrescentus: PLoS One.
Li, Y., and Y. Zhang, 2007, PhoU Is a Persistence Switch Involved in Persister
Formation and Tolerance to Multiple Antibiotics and Stresses in Escherichia
coli: Antimicrobial Agents and Chemotherapy, v. 51, p. 8.
Luttmann, D., Y. G6pel, and B. G6rke, 2012, The phosphotransferase protein EIIANtr
modulates the phosphate starvation response through interaction with
histidine kinase PhoR in Escherichiacoli: Molecular Microbiology, v. 86, p. 15.
Metcalf, W. W., P. M. Steed, and B. L. Wanner, 1990, Identification of phosphate
starvation-inducible genes in Escherichiacoli K-12 by DNA sequence analysis
of psi::iacZ(Mu dl) transcriptional fusions: Journal of Bacteriology, v. 172, p.
3191-3200.
Monds, R. D., P. D. Newell, R. H. Gross, and G. A. O'Toole, 2006, Phosphate-dependent
modulation of c-do-GMP levels regulates PseudomonasfluorescensPf0-1
biofilm formation by controlling secretion of the adhesion LapA: Molecular
Microbiology, v. 63, p. 24.
Monier, A., R. M. Welsh, C. Gentemann, G. Weinstock, E. Sodergren, E. V. Armbrust, J.
A. Eisen, and A. Z. Worden, 2012, Phosphate transporters in marine
phytoplankton and their viruses: cross-domain commonalities in viral-host
gene exchanges: Environmental Microbiology, v. 14, p. 15.
Morohoshi, T., T. Maruo, Y. Shirai, J.Kato, T. Ikeda, N. Takiguchi, H. Ohtake, and A.
Kuroda, 2002, Accumulation of inorganic polyphosphate in phoU mutants of
Escherichiacoli and Synechocystis sp. strain PCC6803: Applied and
Environmental Microbiology, v. 68, p. 4.
Newell, P. D., C. D. Boyd, H. Sondermann, and G. A. O'Toole, 2011, A c-di-GMP effector
system controls cell adhesion by inside-out signaling and surface protein
cleavage: PLoS Biology, v. 9.
Nierman, W. C., T. V. Feldblyum, M. T. Laub, I. T. Paulsen, K. E. Nelson, J. Eisen, J.F.
Heidelberg, M. R. K. Alley, N. Ohta, J. R. Maddock, I. Potocka, W. C. Nelson, A.
Newton, C. Stephens, N. D. Phadke, B. Ely, R. T. DeBoy, R. J. Dodson, A. S.
Durkin, M. L. Gwinn, D. H. Haft, J.F. Kolonay, J.Smit, M. B. Craven, H. Khouri, J.
Shetti, K. Berry, T. Utterback, K. Tran, A. Wolf, J.Vamathevan, M. Ermolaeva,
0. White, S. L. Salzberg, J. C. Venter, L. Shapiro, and C. M. Fraser, 2001,
Complete genome sequence of Caulobactercrescentus: PNAS, v. 98, p. 6.
Ong, C. J., M. L. Wong, and J. Smit, 1990, Attachment of the adhesive holdfast
organelle to the cellular stalk of Caulobactercrescentus: Journal of
Bacteriology, v. 172, p. 9.
Rao, N. N., M. R. G6mez-Garcia, and A. Kornberg, 2009, Inorganic polyphosphate:
essential for growth and survival: Annual Review of Biochemistry, v. 78, p.
43.
Rees, D. C., E. Johnson, and 0. Lewinson, 2009, ABC transporters: the power to
change: Nature Reviews Molecular Cell Biology, v. 10, p. 10.
Simm, R., M. Morr, A. Kader, M. Nimtz, and U. R6mling, 2004, GGDEF and EAL
domains inversely regulate cyclic di-GMP levels and transition from sessility
to motility: Molecular Microbiology, v. 53, p. 12.
101
Skerker, J. M., M. Prasol, B. Perchuk, E. Biondi, and M. T. Laub, 2005, Two-component
signal transduction pathways regulating growth and cell cycle progression in
a bacterium: a systems-level analysis: PLoS Biology, v. 3, p. 334-353.
Tsokos, C. G., and M. T. Laub, 2012, Polarity and cell fate asymmetry in Caulobacter
crescentus: Current Opinion in Microbiology, v. 15, p. 7.
Yang, C., T.-W. Huang, S.-Y. Wen, C.-Y. Chang, S.-F. Tsai, W.-F. Wu, and 0.-H. Chang,
2012, Genome-Wide PhoB Binding and Gene Expression Profiles Reveal the
Hierarchical Gene Regulatory Network of Phosphate Starvation inEscherichia
coli PLoS ONE, v. 7, p. 1.
Zeng, Q., and S. W. Chisholm, 2012, Marine viruses exploit their host's twocomponent regulatory system in response to resource limitation: Current
Biology, v. 22, p. 5.
102
Appendix 1: The Pho regulon in Caulobacter crescentus
The 50 genes associated with ChIP-Seq peaks greater than 7.5 fold enriched in the pstS: :Tn5 mutant sample are given, along with the
PhoB binding sites predicted using the MEME/MAST suite. We attempted to identify a transcription start site for each gene using
RNA-Seq data published in (Fang et al., 2013). Where both a PhoB binding motif and a transcription start site were identified, we
have noted in parentheses after the motif sequence the position of the 5' end of the PhoB binding motif relative to the predicted
transcription start site.
For each of the 50 genes, we also attempted to identify a reciprocal best BLAST hit in the E coli K- 12 genome. E-values are given for
identified reciprocal best BLAST hits, and it is noted where the reciprocal best BLAST hit was also found to be a PhoB-regulated
gene in E. coli in (Yang et al., 2012).
Genes shaded in gray are predicted to be PhoB-repressed, based on microarray expression data given in Appendix 2.
103
Gene #
CC0290
CC0996
CC0722
CCO170
CC3094
CC0172
CC1791
CC2819
CC0361
CC1515
CC2149
CC3301
CC2225
CC3014
Annotation
Phosphate transport
system permease protein
pstC
TonB-dependent
receptor
TonB-dependent
receptor
Hypothetical protein
Sensory box/GGDEF
family protein
General secretion
pathway protein C
TonB-dependent
receptor
TonB-dependent
receptor
Phosphonates transport
ATP-binding protein
phnC
Phosphate-binding
protein
TonB-dependent
receptor
Peptidoglycan-specific
endopeptidase, M23
family
Conserved hypothetical
protein
pstS::TnS
foldenrichment
Reciprocal
best
BLAST E-
Motif
E-
value
Motif 1 (position)
148.55
0.064
TC TCAAACT TCAT
52.11
5.8
TCTT AAACT ACAT
41.21
5.9
TCACACAACCATTAC
35.12
4.2
ACTCAAAAATATCAA
34.22
6
29.94
3.1
28.88
0.47
26.61
5.6
23.15
0.47
Motif 2 (positin
TCAC (kACT
ATTATCAAA
TC TC
C TC C
T iT AA, TTTTC C AA
TA TAAAATC TT CA
(-48)
TC CCAAACT TCTT
TCACAA ACT TA CC
(-175)
1TCACAAATCC
TCAT AATATTAAATA
T T- TC CATTTCA AC
(-90)
yes
TT C
L.OOE-91
(-61)
22.97
0.79
19.54
1.7
19.5
0.072
19.11
0.25
17.93
11
16.4
16
; TTACCAAAAA
TTC !CAACATT
(-85)
1TC TCAAACT T ACA
TCATCA TCT
yes (second
best BLAST
hit)
CAC
ACACC AA CTTC C
TCA AA
CCTTC C
TCACCAATCkTTC CA
C
TTT AA TTTTT T AC
( ACAC AAACT TCAT
(-94)
20
T CTTC
CTTT T AC
TC T AATCTTTCCAA
CC2923
15.29
0.83
CAT AC
TTTCAC TC
TCATCAATCCTT ATC
CC0210
TonB-dependent
15.14
0.19
104
_
TCAAAAAACTTCTCA
16.2
CC2810
vanue
3.OOE-28
Hypothetical protein
TonB-dependent
receptor
Methyl-accepting
chemotaxis protein
TonB-dependent
receptor
CC1970
Motif 3 (plOsitiofl
BLAST hit
found in
PhoB ChIPchip in .
"nfi?
yes
TCAC__
T' AC
TCACCAAACCTTCTAA_
_
receptor
CC0487
CC0316
Ribose-phosphate
pyrophosphokinase
(-37)
14.5
13.98
CC2967
Hypothetical protein
Putative adenylate
cyclase family
Cell wall hydrolase
family protein
CC0925
CC2657
0
CC1015
CC2237
13.96
2.3
13.81
2.4
OAR protein precursor
13.79
13
transposase
Type I secretion outer
membrane protein rsaFa
Deoxycytidine
triphosphate deaminase
13.6
0.27
12.52
7.9
12.42
19
Hypothetical protein
UDP-2,3diacylglucosamine
hydrolase
Ribosomal large subunit
pseudouridine synthase
D
12.15
CC0756
Flagellin
Phosphatidylglycerol
glycosyltransferase
CC3703
Glutamate racemase
CC3706
CC3171
CC3344
CC0453
CC0794
CC1
182
0
CC1263
CC 1163
CC1203
CC1099
0.28
TCACC ATCC TCAT
(-83)
2.00E-1 16
TCAAAA A C TC TT
TCAT e ATAC. TCATA
(-24)
TAAAACA ATTT T AA
(-319)
.OOE+00
1.80E-01
T T AC ATTTT T TC
C C
iC- CTTT CC AC
TCAAk AATT TCATA
3.90E-01
11.94
1.3
TTCATCAAA C ACAT
7.00E-05
11.53
5.6
4.00E-62
11.35
7.7
C C AC
CT AT AC
TC TCAAAAC' ACACC
(-27)
11.26
11.19
7.3
TTCACAAAAAC
Hypothetical protein
Two-component
response regulator
11.04
1.6
tRNA-Glu
10.53
1.8
9.8
4.6
LSU ribosomal protein
L6P
Acyl carrier protein
Periplasmic multidrug
efflux lipoprotein
precursor
TonB-dependent outer
membrane receptor
yes
AC
iC
9.00E-23
TC
Co t ATTT T AC
6.00E-31
8.OOE-64
10.67
TATTTCA ATTT TTTC
TAA(AAA
CC TC. CA
8.00E-58
9.62
3.00E-03
9.4
6.OOE-28
9.23
105
CC3504
Zinc metalloprotease
9.22
CC0162
Hypothetical protein
Tetratricopeptide repeat
family protein
Transcriptional
regulator, algH
TonB-dependent outer
membrane receptor
minimal medium
expressed sRNA
9.21
7.88
2
Hypothetical protein
conserved hypothetical
protein
7.84
2.8
T C TCA ATT
7.8
1.7
C A AA ATTTC T AC
AP nucleosidase
TonB accessory protein
exbB
Transporter, MFS
superfamily
7.68
CC3718
CC3395
CC1666
0
CC0551
CC2017
CC0264
CC2336
CC 1103
9.3
TC TC ATTTC-A AC
8.66
8.05
1.7
T T AC
TTTCAA TC
1.00E-41
7.99
AAC TCAAATTC T AC
C AT
4.30E-01
7.63
3.OOE-44
7.6
3.00E-46
References
Fang, G., K. Passalacqua, J. Hocking, P. M. Llopis, M. Gerstein, N. H. Bergman, and C. Jacobs-Wagner, 2013, Transcriptomic and
phylogenetic analysis of a bacterial cell cycle reveals strong associations between gene co-expression and evolution:
BMC Genomics, v. 14.
Yang, C., T.-W. Huang, S.-Y. Wen, C.-Y. Chang, S.-F. Tsai, W.-F. Wu, and 0.-H. Chang, 2012, Genome-Wide PhoB Binding and Gene
Expression Profiles Reveal the Hierarchical Gene Regulatory Network of Phosphate Starvation inEscherichiacoli PLoS
ONE, v. 7, p. 1.
106
Appendix 2: Complete set of peaks identified by PhoB ChIP-Seq
The complete set of statistically significant peaks identified by MACS from the ChIP-Seq data in Chapter 2 is shown below. For a
given peak, Gene 1 is the closest gene with an annotated translation start site downstream of the peak. Both the CCNA number, and
equivalent CC number are given for Gene 1; these are the gene numbers for the laboratory strain of Caulobacter,CB 1 5N (CCNA
number), and for the closely related CB 15 strain (CC number). In cases where the peak was found between two divergently
transcribed genes, Gene 2 is also listed. For the bulk of our analyses, Gene 1 is considered the PhoB-regulated gene. Any gene for
which a peak was called in at least one of the three ChIP samples is shown, with a "0" given for any sample in which a significant
peak was not identified.
These data are annotated with microarray expression data from two separate experiments. (1) A pstS:: Tn5 mutant and
pstS::Tn5;AphoB strains grown in PYE (rich medium) and compared to a wild-type reference. (2) A AphoR strain grown in minimal
medium with 50 [tM phosphate (M5G) or 10 mM phosphate (M2G) and compared to a wild-type reference grown under the same
conditions. This second experiment was published in (Capra et al., 2012). All microarray data are given as raw ratios of experiment to
reference.
107
Fold enrichments are given as
experiment
(3XFLAG tag) over control (no
epitope tag)
Gene I
CC#
CCNA 00292
CC0290
Annotation
Phosphate transport system
permease protein pstC
CCNA 01047
CC0996
CCNA 00759
CCNA 00169
CCNA 03191
CC3094
CCNA 00171
pstS::Tn5
PYB
PYE
These data are given as raw ratios of experiment to reference (wt)
M5G
Gene 2
pstS::Tn5/wt
pstS::Tn5;
AphoB,*'t
AphoRlwt M5G
AphoRlwt
M2G
27.96
148.55
46.11
6.208690342
0.303878518
0.229498636
1.645822776
TonB-dependent receptor
0
52.11
25.55
8.039888022
1.279130928
0.020062866
1.680932631
CC0722
TonB-dependent receptor
0
41.21
33.76
87.096359
1.028647718
0.010441759
1.224653106
CCO170
5.09
35.12
15.66
180.925591
1.041927498
0.048972373
3.024894913
0
34.22
19.63
5.13767546
1.192010278
0.348455196
1.576897094
CC0172
Hypothetical protein
Sensory box/GGDEF family
protein
General secretion pathway
protein C
0
29.94
0
1.110495152
0.691416915
0.250546849
1.238547281
CCNA 01869
CC1791
TonB-dependent receptor
0
28.88
20.93
3.408962486
0.94691525
0.241891062
1.055389671
CCNA 02910
CC2819
2.89
26.61
22.85
0.325087297
2.855398105
10.54816487
0.680917218
CCNA 00366
CC0361
TonB-dependent receptor
Phosphonates transport ATPbinding protein phnC
0
23.15
9.5
2.143987367
0.820996132
0.471326672
0.904183661
CCNA 01583
CC1515
Phosphate-binding protein
0
22.97
12.65
2.620193307
1.000537413
0.363961258
1.195091291
CCNA 02232
CC2149
0
19.54
17.09
14.83798747
1.430128298
0.099625088
1.11200689
CCNA 03410
CC3301
TonB-dependent receptor
Peptidoglycan-specific
endopepidase, M23 family
0
19.5
10.2
0.988524642
0.494908599
0.754055612
0.942981724
CCNA 02308
CC2225
Conserved hypothetical protein
0
19.11
10.48
CCNA 02309
3.595699031
0.996368525
0.210314313
1.132086378
CCNA 03109
CC3014
Hypothetical protein
0
17.93
10.84
CCNA 03108
1.046887466
1.142944126
1.439800639
1.006795871
CCNA 02048
CC1970
0
16.4
7.95
0.353210278
0.886883688
4.521902316
0.969819898
CCNA 02901
CC2810
TonB-dependent receptor
Methyl-accepting chemotaxis
protein
4.71
16.2
9.32
0.897842171
1.199430256
2.528878449
1.209571633
CCNA 03019
CC2923
TonB-dependent receptor
0
15.29
13.17
CCNA 03020
0
0
0.077603052
1.092758853
CCNA 00210
CC0210
0
15.14
7.92
CCNA 00209
1.770108958
1.188258996
0.854394618
0.966270828
CCNA 00520
CC0487
TonB-dependent receptor
Ribose-phosphate
pyrophosphokinase
0
14.5
10.17
0.823095069
1.349221748
0.679141058
0.892474773
00318
CC0316
Hypothetical protein
0
13.98
7.37
0.67842212
1.399587323
4.128879581
0.829898284
0
13.96
7.04
0
0
0.327708413
1.342544792
CCNA
CCNA 03638
108
0 Putative adenylate cyclase family
CCNA 00365
CCNA 02231
CCNA 00317
Gene 1
IAnnotation
CC#
pstS::TnS5
PYBE
PYEB
M5G
Gene 2
pstS::Tn5;
nstS::Tn5Aw'i
AphoR/wt
AnIhoB/wt
AnhnIt MSG
AG
CCNA 03062
CC2967
Cell wall hydrolase family
protein
0.398819759
0.749134848
5.089533982
0.926379498
CCNA 00974
CC0925
OAR protein precursor
0
13.79
8.59
CCNA 00973
0.317443667
1.326783311
7.507897295
0.783700065
CCNA 02740
CC2657
0
13.6
11.05
CCNA 02741
0
0
0
0
CCNA 01067
CC1015
0
12.52
6.21
1.421510264
0.868160456
0.85621018
1.072537859
CCNA 02320
CC2237
transposase
Type I secretion outer membrane
protein rsaFa
Deoxycytidine triphosphate
deaminase
0.901459563
0.972943229
0.995974802
0.906641639
CCNA 03274
CC3171
1.333521432
0.653130553
0.95795842
0.892704861
CCNA 03454
CC3344
CCNA 00485
CC0453
CCNA 00836
CC0794
CCNA 00793
0
13.81
6.12
0
12.42
7.72
Hypothetical protein
UDP-2,3-diacylglucosamine
hydrolase
Ribosomal large subunit
pseudouridine synthase D
0
12.15
10.92
0
11.94
6.31
CCNA 03453
4.355118737
1.099118313
0.20664848
1.050833237
0
11.53
9.58
CCNA 00486
0.836317971
0.995798411
1.071704386
1.115766027
0
11.35
6.58
1.097779206
0.900533763
1.40362049
0.865956053
CC0756
Flagellin
Phosphatidylglycerol
glycosyltransferase
0
11.26
10.62
CCNA 00792
6.302935415
0.832122266
0.065702514
1.606337135
CCNA 03817
CC3703
Glutamate racemase
0
11.19
5.29
CCNA 03816
0.823505809
0.71285303
0.977618829
0.921611023
CCNA 03821
CC3706
0
11.04
6.37
0.752257744
0.776664282
1.046350103
1.078326837
CCNA 01240
CC 1182
Hypothetical protein
Two-component response
regulator
0
10.67
7.32
0.873600977
0.88270042
0.930557047
0.906017858
CCNA R0037
0
tRNA-Glu
0
10.53
0
0
0
0
0
CCNA 01321
CC1263
LSU ribosomal protein L6P
0
9.8
7.52
0.486183258
0.628058359
0.215143
0.938165655
CCNA 01221
CC1163
0
9.62
9.08
0.444777549
0.783262155
2.274896105
1.156808883
CCNA 01261
CC1203
0
9.4
5.72
1.379113469
0.659680024
0.556404374
1.211503332
CCNA 01155
CC1099
Acyl carrier protein
Periplasmic multidrug efflux
lipoprotein precursor
TonB-dependent outer
membrane receptor
0
9.23
7.08
0.235768464
0.925668988
4.402611339
0.996932165
CCNA 03619
CC3504
Zinc metalloprotease
0
9.22
5.31
0.419436931
1.227627665
1.689710446
1.014098621
CCNA 00161
CC0162
0
9.21
0
0.982778294
1.225650557
1.771720274
1.045345459
CCNA 03834
CC3718
Hypothetical protein
Tetratricopeptide repeat family
protein
0
8.66
6.44
0.949020295
1.037110425
1.019975221
1.164336042
CCNA 03506
CC3395
0
8.05
4.53
0.865871276
0.707674152
0.838499522
0.987703518
CCNA 01738
CC1666
0
7.99
4.93
0.352533181
3.689775986
7.677752363
0.995274754
CCNA R0079
0
Transcriptional regulator, algH
TonB-dependent outer
membrane receptor
minimal medium expressed
sRNA
0
7.88
3.89
0
0
0
0
CCNA 02321
CCNA 03505
109
pstS::Tn5
PYE
M5G
Hypothetical protein
0
7.84
8.33
1.075969597
conserved hypothetical protein
0
7.8
5.56
CCNA 02097
0.659173895
CC0264
AMP nucleosidase
0
7.68
0
CCNA 00264
1.278595975
CCNA 02421
CC2336
TonB accessory protein exbB
0
7.63
6.78
CCNA 01159
CC 1103
Transporter, MFS superfamily
0
7.6
CCNA 02933
CC2840
Aminopeptidase
0
CCNA 01896
CC1820
Hypothetical protein
CCNA 03439
CC3330
CCNA 02283
CC2200
CCNA 03839
CC3723
CCNA 02738
CC2655
CCNA 02035
CC1958
CCNA 01327
CC1269
CCNA 00451
CC0442
CCNA 01773
CC#
Annotation
CCNA 00585
CC0551
CCNA 02096
CC2017
CCNA 00265
AphoRlwt
pstS::Tn5;
PYE
Gene I
AphoB/wt
AphoRkwt M5G
M2G
1.06194005
1.281926817
0.940127984
1.611882325
2.224763892
0.934616371
0.834065197
0.431735634
0.788564968
1.777175564
1.077484697
0.590470621
1.441597758
4.98
1.065123582
0.698574132
0.859212223
1.197566462
7.37
4.35
0.620052654
0.953078277
1.466012768
0.914673259
0
7.28
7.24
4.083193863
1.200788906
0.447402681
1.517667155
Hypothetical protein
HNH endonuclease family
protein
Acylamino-acid-releasing
enzyme
Granaticin polyketide synthase
putative ketoacyl reductase 2
0
7.25
4.52
3.559043802
0.99296361
0.21164951
0.873070946
0
7.23
5.94
2.313840045
4.368509988
0.313538337
1.133849441
0
7.15
0
0.405041945
0.730718327
1.575963117
0.946284182
0
6.77
4.51
0.815126477
0.975962954
0.725681935
0.820315203
Glycosyltransferase
Adenylate kinase / Nucleosidediphosphate kinase
0
6.46
5.35
3.157427106
0.83010558
0.714359564
1.055661079
0
6.42
6.18
0.572532309
0.973329755
0.933232862
0.969780183
0
6.39
5.85
0.977912506
0.756078114
1.393075701
1.114799249
CC1701
TonB-dependent receptor
SNARE-associated family
membrane protein
0
6.27
4.65
1.512864388
1.022821857
0.418246544
0.961607519
CCNA 02895
CC2804
TonB-dependent receptor
3.41
6.12
3.59
CCNA 02896
0.910056951
0.821531624
1.041580486
1.088159812
CCNA 01185
CC 1127
Hypothetical protein
0
6.05
0
CCNA 01186
0.812942821
0.869725092
0.629740541
1.010294536
CCNA 01533
CC1466
Hypothetical protein
0
5.91
4.38
1.204786243
0.913974334
0.621716754
0.905065992
5.89
4.05
2.117385818
1.18722582
0.03688341
1.458970057
Gene 2
pstS::Tn5/wt
CCNA 00776
CC0739
OAR protein precursor
0
CCNA 01189
CC1131
TonB-dependent receptor
0
5.78
5
1.69482554
0.965047633
0.494384187
1.100602495
CCNA 00598
CC0563
TonB-dependent receptor
0
5.71
3.36
16.18080038
1.27987238
0.207962216
1.320362139
CCNA 01439
CC1375
dTDP-glucose 4,6-dehydratase
0
5.68
0
11.93072037
0.809903607
0.150097059
1.114388326
CCNA 00639
CC0603
heme:hemopexin-binding protein
0
5.67
3.14
0.863733969
1.323152799
1.484893352
1.268192985
CCNA 02001
CC1924
Phosphoserine phosphatase
0
5.64
5.07
1.005657276
0.496878265
0.042325273
1.049920669
CCNA 01636
CC1565
Alkaline phosphatase
0
5.56
3.8
4.892778507
1.109305359
0.322064539
0.965877387
110
CCNA 01440
CCNA 02000
pstS::Tn5
Gene 1
CC#
CCNA 03701
CC3586
CCNA 03064
CC2969
Annotation
Integration host factor betasubunit
Transcriptional regulator, PadR
family
CCNA 01296
CC1238
CCNA 01568
pstS::Tn5;
PYB
M5G
0
5.55
4.46
0
5.53
0
hypothetical protein
0
5.49
CC1500
Na(+)/H(+) antiporter
0
CCNA 00425
CC0420
Hypothetical protein
CCNA 01400
CC1339
CCNA 00225
PYEB
Gene
AphoRlwt
pstS::Tn5/wt
AnhoR/wt
AnhnR/vst M5G
0.611980381
1.600340883
0.903114279
0.855247485
0.874063243
0.784435325
1.212014649
1.058429095
6.41
1.425607594
1.058644145
0
0
5.41
3.82
0.913326902
0.874127936
1.068605587
0.919773867
0
5.27
2.92
1.660861175
0.436180923
1.793618583
1.468984061
Nitrogen regulatory protein GlnK
0
5.27
5.69
0.345302721
0.480977765
1.290491735
0.972952726
CC0225
Hypothetical protein
0
5.21
0
1.005321568
1.341297144
1.512298416
1.249259889
CCNA 01041
CC0990
Hypothetical protein
0
5.15
5.7
1.020939484
1.28528666
1.357360775
1.142504274
CCNA 00157
CC0158
Glyoxalase family protein
0
5.1
0
0.961224872
0.483893736
0.906986942
0.74749652
CCNA 02562
CC2477
Sell repeat domain protein
0
5.09
3.22
0.874008298
1.370460956
1.093510048
0.912345246
CCNA 03246
CC3144
0
5
3.55
1.071354833
0.760910072
1.669704821
0.722833925
CCNA 00893
CC0850
0
4.95
0
0.804637002
1.215719361
0.593817388
0.930605436
CCNA 03268
CC3165
0
4.91
3.55
1.043305909
0.850354472
1.609020332
1.152114783
CCNA 03719
CC3605
Hypothetical protein
Hypothetical transcriptional
regulatory protein
diguanylate receptor protein
dgrB
Undecaprenyl-diphosphatase
(Bacitracin resistance protein)
0
4.91
0
1.091607892
0.933827517
0.645902625
0.888117411
CCNA 01944
CC1868
Hypothetical cytosolic protein
0
4.91
0
1.021723384
1.961853922
1.589832118
1.113240483
CCNA 02830
CC2744
0
4.89
0
1
1
1.748304779
1.297311052
CCNA 02132
CC2053
0
4.88
0
1.093830427
0.965867382
1.206575756
0.99185752
CCNA 01606
CC1537
0
4.79
0
0.979374969
1.146481365
0.745288988
1.255420076
CCNA 03791
CC3677
DNA repair protein radC
Transcriptional regulator, Lacl
family
Chromosome partitioning protein
parA
Transcriptional regulator, MarR
family
0
4.74
0
5.181633204
0.874648052
0.095692715
1.063435236
CCNA 01170
CCl113
TonB-dependent receptor
0
4.74
7.81
10.77291748
0.798648271
0.076804736
1.465825258
CCNA 02706
0
0
4.73
0
0
0
0.998343066
1.485026797
CCNA 03252
CC3150
hypothetical protein
Hypothetical transcriptional
regulatory protein
0
4.72
0
0.8259904
0.891815467
1.324401021
1.239742404
CCNA 00128
CCO129
Argininosuccinate synthase
0
4.71
0
0.796098246
0.801678063
0.61188292
0.969489396
CCNA 01150
CC1094
Hypothetical protein
0
4.71
0
1.29807564
1.0351374
1.649247656
1.571645365
2
CCNA 03065
CCNA 00224
CCNA 03247
CCNA 01607
CCNA 03251
CCNA 01149
M2GC
111
PYE
pstS::Tn5
PYB
M5G
Cytochrome p450
0
4.7
CC2385
hypothetical protein
0
CCNA 00982
CC0933
CCNA 03278
CC3174
Transcriptional regulator
Aminobutyraldehyde
dehydrogenase
CCNA 01862
CC1784
Hypothetical protein
CCNA 00863
CC0820
CCNA 01271
CC1213
CCNA 01865
CC1787
xylonolactonase xylC
Transcriptional regulator, MarR
family
Periplasmic multidrug efflux
lipoprotein precursor
CCNA 00718
CC0679
CCNA 03054
Gene 2
pstS::Tn5/wt
pstS::Tn5;
AphoB/wt
AphoRlwt MSG
2.72
CCNA 00060
0.900188236
0.859739077
1.171273108
0.985256489
4.67
4.08
CCNA 02469
1.302266894
1.120598405
2.230013914
1.569739623
0
4.63
0
2.184238209
0.807854843
1.097403486
0.859106694
0
4.63
0
0.799045351
0.67021522
1.078262996
0.806720479
0
4.55
4.47
0.657657837
0.905593576
0.825631208
0.97241199
0
4.49
0
1.026795602
1.207841646
0.729049059
0.933480973
0
4.48
0
1.276291862
1.22179966
0.748077309
1.013349002
0
4.47
0
0.783489775
0.531740819
0.916351652
0.942706824
abortive infection protein
0
4.45
3.33
0.340800325
0.615885529
1.485213925
1.074892659
CC2960
Dioxygenase
0
4.42
0
1.752265862
1.16375082
0.648202935
1.211097222
CCNA 02179
CC2094
hypothetical protein
0
4.4
2.26
0.659680024
0.311649667
1.258757455
1.250880952
CCNA 00834
CC0792
0
4.35
0
1.230835461
1.191516333
1.560714693
0.804612227
CCNA 02821
CC2735
Flagellin
hipA-related phosphatidylinositol
3/4-kinase
Polyribonucleotide
nucleotidyltransferase/
Polynucleotide
adenylyltransferase
0
4.35
0
1.100102959
0.824707606
1.145087905
1.191319963
0
4.34
0
0.789826005
0.47705827
0.548342348
0.977079621
Gene 1
CC#
Annotation
CCNA 00061
CC0063
CCNA 02468
CCNA 00719
AphoRlwt
M2G
CCNA 00033
CC0034
CCNA 03549
CC3436
TonB-dependent receptor
0
4.32
0
1.064633187
1.010531836
1.187976658
1.065004423
CCNA 00835
CC0793
Flagellin
0
4.3
0
1.16251884
1.147809989
1.470408224
0.803147534
CCNA 00919
0
0
4.3
3.84
0
0
0.830914071
1.219831347
CCNA 02059
CC1980
Hypothetical protein
Ribosomal large subunit
pseudouridine synthase C
0
4.28
0
0.981552047
0.945003317
1.145841836
0.958171976
CCNA 02761
CC2678
0
4.27
0
0.529798972
1.278805348
1.928520142
0.969285403
CCNA 00359
CC0354
hypothetical protein
SH3 domain-containing cell
surface protein
0
4.24
0
1.000253316
1.098246935
0.921046346
0.801680211
CCNA 02460
CC2375
conserved hypothetical protein
0
4.21
0
0.913988539
0.953783928
1.525908048
0.969427225
CCNA 01021
CC0970
TonB-dependent receptor
0
4.21
3.42
1.719953347
0.901225213
0.964641376
0.703189709
CCNA 00957
CC0911
Porin O precursor
0
4.19
0
0.808288997
1.167705519
1.464140086
1.000640364
112
CCNA 01020
AphoRlwt
pstS::Tn5
PYE
M5G
0
4.16
0
0
4.15
3.1
Cold shock protein
0
4.14
0
CC2396
Aryl-alcohol dehydrogenase
0
4.12
CC0548
Hypothetical protein
0
4.12
RNA polymerase sigma factor
0
4.12
0
0
0
1.363513671
1.176675389
Hypothetical protein
0
4.1
0
1.684224575
4.024080565
0.520541016
1.201729039
small non-coding RNA
Transcriptional regulator, ArsR
family
0
4.09
3.05
0
0
0
0
0
4.07
0
1.492745305
0.853043993
1.16671015
1.583160389
General stress protein 170
Xanthine dehydrogenase large
subunit
Tetratricopeptide repeat family
protein
1-acyl-sn-glycerol-3-phosphate
acyltransferase
0
4.05
0
1.262796394
1.072012872
1.03047746
1.199125103
0
4.02
0
0.623322064
0.955707513
1.535399177
1.062019054
0
3.98
0
1.008223154
1.041528387
1.208485124
0.998220645
0
3.95
0
1.104926356
0.768245456
0.750784937
1.479230012
0
3.91
3.4
0.98529588
0.881319406
1.040535033
1.061853015
0
3.91
2.7
0.874983775
1.039441375
0.812899184
0.920750647
Gene I
CC#
Annotation
CCNA 03368
CC3259
CCNA 02790
CC2707
Hypothetical CheE protein
RNA polymerase ECF-type
sigma factor
CCNA 00701
CC0665
CCNA 02479
CCNA 00582
PYB I
pstS::Tn5;
pstS::Tn5/wt
AphoBlwt
AphoR/wt M5G
M2G
0.704287532
1.231685986
1.760662074
1.069855487
CCNA 02789
1.099808318
0.542360559
2.156944978
0.988054467
CCNA 00700
0.771554604
0.956312869
1.061121867
0.866623657
0
0.817946691
0.901008214
1.182566815
0.928545416
7.38
0.855460576
1.319268925
5.597091796
0.758545048
Gene 2
CCNA 00685
0
CCNA 00705
CC0669
CCNA 00686
CCNA R0025
0
CCNA 01573
CC1505
CCNA 00990
CC0941
CCNA 02701
CC2618
CCNA 02120
CC2039
CCNA 01932
CC1856
CCNA 00841
CC0799
CCNA 00326
CC0323
Para-nitrobenzyl esterase
Vegetatible incompatibility
protein HET-E-1
CCNA 03119
CC3024
Hypothetical protein
0
3.89
0
1.26813844
0.861919422
1.47949988
1.322866679
CCNA 00447
CC0438
Chemotaxis protein cheD
0
3.89
0
0.961250695
0.934782463
0.766787849
0.907399369
CCNA 02034
CC1957
0
3.86
0
0.93225202
0.758810502
1.074384147
1.15452259
CCNA 02601
CC2516
Luciferase-like monooxygenase
Glutathione-dependent
formaldehyde dehydrogenase
0
3.86
0
1.144266393
0.569227291
1.205870014
0.903763749
CCNA 02770
CC2688
Conjugal transfer protein trbL
0
3.86
0
0
0
1.33403011
1.013807542
CCNA 03527
CC3414
Glutamate--cysteine ligase
0
3.84
0
1.125598052
0.642145337
0.684337957
1.018034361
CCNA 01365
CC1307
Aspartyl protease perP
0
3.81
0
0.866163742
1.020375448
1.237077854
0.849720639
CCNA 03743
CC3628
0
3.81
0
0.876442306
0.86782735
1.349031614
0.922325316
CCNA 03621
CC3506
Beta-lactamase family protein
Transcriptional regulator, AraC
family
0
3.79
0
0.980442711
0.987382796
1.538662893
1.092794714
CCNA 01574
CCNA 00325
CCNA 03622
113
Gene
1
CC#
PYE
pstS::Tn5
PYE
M5G
0
3.76
pstS::Tn5;
Gene 2
vstS::TnS/wt
AphoB/wt
AphoR/wt M5G
AphoRiwt
M2G
0
CCNA 00455
0.988429175
0.724287702
0.829189587
1.136236272
CCNA 01115
0.656649069
0.880616193
1.316617856
1.09581856
CCNA 00454
CC0445
Annotation
Transcriptional regulator, GntR
family
CCNA 01116
CC1063
Histidine protein kinase divJ
0
3.72
3.58
CCNA 03629
CC3514
Hypothetical protein
0
3.71
0
0.987643023
1.472312502
3.078087838
1.391662857
CCNA 02347
CC2264
Phosphomannomutase
0
3.7
0
2.036156294
0.615035239
0.322231407
0.857085032
CCNA 03169
CC3073
Conserved hypothetical protein
0
3.68
0
0.951371199
0.759844982
0.789895165
1.404092919
CCNA 01513
CC1446
C4-dicarboxylate-binding protein
0
3.68
3.98
0.942947471
1.077333092
1.726127623
1.449266893
CCNA 01837
CC1761
Transposase
0
3.66
0
1.128868444
0.939362747
1.547105805
1.162957595
CCNA 02999
CC2904
0
3.62
0
1.170307532
1.434112869
1.149236438
0.819071684
CCNA 00752
CC0715
0
3.6
0
0.81245628
0.691233854
1.504357907
0.942058072
CCNA 03355
CC3246
Conserved hypothetical protein
3-hydroxybutyryl-CoA
dehydrogenase
5-formyltetrahydrofolate cycloligase
0
3.58
0
0.775830175
0.564503536
1.514589299
0.874356234
CCNA 00628
CC0591
Chemotaxis protein cheY
0
3.58
0
CCNA 00629
1.448423479
1.230983472
2.133144756
1.270937249
CCNA 03325
CC3218
0
3.55
0
CCNA 03326
0.236455816
1.769599553
2.017283867
1.025313572
CCNA 02499
CC2416
Hypothetical protein
Type IV secretory pathway,
VirB2 protein
0
3.53
2.84
1.075375159
1.230597419
1.060104358
0.976833767
CCNA 01557
CC1490
Uronate isomerase
0
3.53
3.71
1.167536977
0.896877921
0.845230772
0.93681124
CCNA 01143
CC1089
Translation initiation inhibitor
0
3.51
0
1.006468066
1.059131781
0.76316638
0.945731469
CCNA 00544
CC0510
0
3.51
0
CCNA 00543
1.350750074
0.768909099
1.290374736
0.838635572
CCNA 02658
CC2575
Acetyl-CoA acetyltransferase
3-hydroxybutyryl-CoA
dehydratase
0
3.49
0
CCNA 02659
1.091130428
0.632411851
1.016113338
1.011566018
CCNA 02597
CC2511
Conserved hypothetical protein
0
3.49
0
0.868010543
0.785235635
0.731933493
1.317506265
GDP-mannose 4,6 dehydratase
0
3.47
0
0
0.920233599
CCNA 03168
CCNA 00472
0
0
1.722296707
CCNA 01455
CC1392
Hypothetical protein
0
3.47
0
0.955955112
0.991824884
1.067421981
1.148943707
CCNA 00529
CC0495
conserved hypothetical protein
0
3.46
0
1.217586993
0.812830516
0.598017581
0.983167236
CCNA 00319
CC0317
Acetyltransferase, GNAT family
0
3.44
0
0.627335696
0.431519077
0.666668535
1.167254899
CCNA 01708
CC1636
0
3.44
0
1.175844792
1.132139648
0.917432035
0.969392843
CCNA 01584
CC1516
Conserved hypothetical protein
Multimodular transpeptidasetransglycosylase PBP IA
0
3.44
0
0.888960574
0.998714258
0.676652816
0.915984434
CCNA 03313
CC3207
Hypothetical protein
0
3.43
0
0.848268486
0.507379987
1.948516149
1.422685598
114
CCNA 03314
pstS::Tn5wt
pstS::Tn5;
AphoB/wt
AphoRlwt M5G
AphoRlwt
M2G
7.418794358
0.602559586
0.205185362
0.922946376
1.095447748
1.116906111
0.871986291
0.990420232
0
1.225389064
2.229769651
1.576683887
0.969885637
3.36
0
0.710044739
0.571478637
0.365232395
0.993023049
0
3.31
0
1.77827941
0.881319406
1.024137512
1.119031205
Nucleoside diphosphate kinase
0
3.31
0
1.465922838
0.457907522
0.441323799
1.036050422
Ribosome-associated factor Y
0
3.31
0
1.022421939
0.9100809
1.083082286
1.289308352
0.900096119
0.553182084
0.953144113
PYE
pstS::Tn5
PYE
M5G
0
3.42
0
CCO018
3-phytase /6-phytase
Molybdenum cofactor
biosynthesis protein A
0
3.38
0
CCNA 00635
CC0599
Hypothetical protein
0
3.37
CCNA 03305
CC3201
0
CCNA 02747
CC2664
SSU ribosomal protein S7P
ABC transporter substratebinding protein
CCNA 01770
CC1699
CCNA 03711
CC3597
Gene 1
CC#
Annotation
CCNA 01353
CC1295
CCNA 00018
CCNA 03640
CC3525
Gene 2
CCNA 00019
CCNA 01769
Ferredoxin--NAD(+) reductase
0
3.29
0
0.817146641
3.29
0
1.023198748
0.685646084
0.94448488
0.943540358
0.875991718
1.372355626
0.876935352
0.988188823
1.036453924
0.878364955
1.11709213
0.958303376
CCNA 01610
CC1541
2-isopropylmalate synthase
0
CCNA 00273
CC0272
0
3.27
0
CCNA 02726
CC2643
Peptide deformylase
Putative
acetyltransferase/acyltransferase
0
3.19
0
CCNA 00048
CCO050
S-adenosylmethionine synthetase
0
3.14
0
1.087140202
0.757452766
0.469514272
0.802505879
CCNA 00636
CC0600
Hypothetical protein
0
3.12
0
0.986506611
1.648162392
0.930220537
0.852344121
CCNA R0021
0
tRNA-Pro
0
3.12
0
0
0
0
0
CCNA 02663
CC2580
DnaK suppressor protein dksA
0
3.1
CCNA 02727
0
CCNA 03575
0.890170777
3.010694346
0.603235732
0.951020675
0.866438018
1.374880651
1.400787842
0.66286204
CCNA 03574
CC3461
TonB-dependent receptor
0
3.09
2.69
CCNA 01810
CC1736
Hypothetical cytosolic protein
0
3.09
0
0.761465094
0.938885215
0.879135119
0.958900116
CCNA 00082
CC0084
0
3.09
0
0.879326173
0.908290977
0.692808813
0.913474008
CCNA 03372
CC3263
Class I lysyl-tRNA synthetase
Bacterioferritin-associated
ferredoxin
0
3.08
0
0.72443596
1.770108958
0.585660526
1.398781989
CCNA 03043
CC2948
Type IV pilin protein pilA
0
3.08
0
1.885818969
1.02789795
1.407845019
1.182227082
CCNA 01244
CC1186
Hypothetical protein
0
3.08
3.26
1.197843278
0.773927003
1.264032834
1.058313925
CCNA 01059
CC1007
S-layer protein rsaA
0
3.06
0
0.929651127
0.95856154
1.848253623
1.215605479
CCNA 03263
CC3161
TonB-dependent receptor
0
3.03
3.02
2.051162179
1.021644967
1.128379737
1.038626614
CCNA 02369
CC2286
Alpha-amylase
0
3.02
0
1.013277901
0.844197936
1.160426778
0.994604817
CCNA 01051
CC0999
TonB-dependent receptor
0
3.01
0
1.114814221
1.203271831
1.25701101
1.042329744
CCNA 03044
CCNA 02370
-
115
Gene I
CC#
CCNA 03871
CC3755
CCNA 01475
CC1409
CCNA 01556
CC1489
CCNA 03728
CC3613
CCNA 03129
Annotation
Glucose inhibited division
protein A
OmpW family outer membrane
protein
Transcriptional regulator, LacI
family
PYE
pstS::Tn5
PYE
M5G
Gene 2
AphoRlwt
pstS::Tn5/wt
pstS::TnO;
AphoB/vt
AphoR/wt M5G
M2G
0
3.01
0
0.942052269
0.810587682
1.009498171
1.011352395
0
2.98
0
0.966540374
1.15425031
1.277586418
1.615745753
0
2.98
0
0.854205769
0.917104343
0.981548085
0.90626556
0
2.9
0
0.4852885
1.261972816
2.849009935
1.028876425
CC3034
Hypothetical protein
Peptidoglycan-specific
endopeptidase, M23 family
0
2.87
0
0.90153308
1.303366838
1.041533636
0.978619532
CCNA 02766
CC2684
Conserved hypothetical protein
0
2.82
0
1.244981382
1.600610693
1.127419076
1.068375654
CCNA 00438
CC0429
Hypothetical protein
0
2.82
0
1.21604599
0.572136955
1.070515486
1.014933414
CCNA 01085
CC1033
DNA-cytosine methyltransferase
0
2.63
0
0.682391067
0.820855414
1.290508106
1.300583525
CCNA 03715
CC3601
Protein yhbN precursor
0
2.63
0
1.034786871
0.782681376
0.875131234
1.124173205
CCNA 03870
CC3754
Methyltransferase gidB
0
2.56
0
0.91862868
1.065874323
0.763706489
0.971846222
CCNA 00886
CC0843
0
2.47
0
0.687727938
0.998508272
0.451254645
0.918857679
CCNA 02039
CC1961
0
2.46
0
1.041757575
0.427891178
0.73989377
0.873812609
CCNA 03259
CC3157
Aspartokinase
ATP-dependent endopeptidase
clp ATP-binding subunit clpX
Hypothetical cupin 2 domain
containing protein
0
2.14
0
0.754628726
0.867627548
1.074009794
1.121156996
CCNA R0014
0
0
0
3.66
0
0
0
0
CCNA 01451
CC1387
Cold shock protein cspD
0
0
4.5
0.742164198
0.97525159
3.63885628
0.979593559
CCNA 01782
CC1710
0
0
4.53
1.301034728
0.670930875
0.265111666
0.930119178
CCNA 00202
CC0202
Polyphosphate kinase
DesA-family fatty acid
desaturase
CCNA 00081
CC0083
Hypothetical protein
CCNA 01186
CC 1128
CCNA 01307
CC1249
CCNA 00732
CC0696
CCNA 00041
CCNA 03224
116
0
CCNA 01555
CCNA 00439
CCNA 00885
CCNA 00778
0
0
4.35
0.991194766
1.120278497
1.177717203
0.980492295
1.88
0
0
0.7724434
1.075721875
2.867635863
1.197967895
Conserved hypothetical protein
LSU ribosomal protein LIE (=
L4P)
0
0
4.81
1.423420883
0.705504844
1.089080685
0.895665076
0
0
3.34
0.460468578
0.987174663
0.179532546
1.001394482
0
0
4.72
0.864503323
1.028700353
0.540999852
0.992739185
CC0042
GumN superfamily protein
Bacterial Protein Translation
Initiation Factor 2 (IF-2)
2.01
0
0
0.660541335
0.79754566
0.388655623
0.943554283
CC3124
Hypothetical protein
0
0
3.53
1.255307143
0.785687783
0.777818812
0.964242693
CCNA 01185
CCNA 00731
Gene 1
CC#
annotation
CCNA 00497
CC0465
CCNA 03381
CC3272
Putative rhamnosyl transferase
Glycerophosphoryl diester
phosphodiesterase
pstS::Tn
PYB
PYE
Gene 2
M5G
0
0
3.64
0
0
4.13
CCNA 03380
pstS::Tn5/wt
pstS::Tn5;
AphoB/wt
AphoRlwt M5G
AphoRiwt
M2G
0.587038611
1.753880502
0.757405012
0.935733961
9.562150085
0.860311159
0.04330754
1.332464688
References
Capra, E. J., B. S. Perchik, J. M. Skerker, and M. T. Laub, 2012, Adaptive mutations that prevent crosstalk enable the expansion of
paralogous signaling protein families: Cell, v. 150, p. 11.
117
Appendix 3: Genes regulated in response to pho U depletion in Caulobacter
Expression changes in the phoU depletion strain were assayed by microarray at 2, 5, and 7 hours after the removal of vanillate
when the strain was grown in rich medium (PYE). The same strain treated identically and then grown in the presence of
vanillate for 2, 5, or 7 hours was used as a reference (see Chapter 2, Experimental Procedures). Genes 1.5-fold or more
upregulated at the 7 hour timepoint are given in Appendix 3A, and genes 1.5 fold or more downregulated at the 7 hour
timepoint are given in Appendix 3B. Raw ratios are given. Gene numbers for the laboratory CB15N Caulobacterstrain (CCNA
number), as well as the corresponding gene numbers for the CB15 Caulobacterstrain (CC number), are given.
118
Appendix 3A - Genes 1.5 fold or more upregulated after 7 hours phoU depletion
ratio Shrs
ratio 2hrs
CCNA number
CC number
Annotation
CCNA 01354
CC 1296
Myo-inositol 2-dehydrogenase idhA
1.870255121
2.98745332
3.533483504
CCNA 01359
CC 1301
1.598470802
2.643635328
3.074655194
CCNA 00902
CC 0859
myo-inositol catabolism protein ioIB
inositol ABC transporter, periplasmic inositol-binding protein
ibpA
1.526559721
2.453578135
3.015514314
CCNA 01358
CC 1300
myo-inositol catabolism protein iolE
1.43222845
2.272977437
2.821133181
CCNA 01357
CC 1299
myo-inositol catabolism protein
iolD
1.46043968
2.304278807
2.758301108
CCNA 00292
CC 0290
Phosphate transport system permease protein pstC
0.956475524
1.015005697
2.5616353
CCNA 00294
CC 0292
Phosphate transport ATP-binding protein pstB
0.970539747
1.058595427
2.561563427
CCNA 01356
CC 1298
5-dehydro-2-deoxygluconokinase
1.451470512
2.445808441
2.540327217
CCNA 01360
CC 1302
Malonate-semialdehyde dehydrogenase iolA
1.277339714
2.087538514
2.419288197
CCNA 00904
CC 0861
inositol ABC transport system, permease protein iatP
1.347529987
2.084258228
2.347669308
CCNA 00293
CC 0291
Phosphate transport system permease protein pstA
0.971752842
1.040282299
2.232511728
CCNA 01855
CC 1777
Superoxide dismutase
1.056707624
0.421232561
2.215668026
CCNA 01854
CC 1776
Transcriptional regulator, GntR family
1.079674291
0.406957824
2.137495898
CCNA 00296
CC 0294
Phosphate regulon response regulator PhoB
0.920083128
0.990745798
2.10456132
CCNA 00619
CC 0583
Succinylarginine dihydrolase
1.624844179
2.174647392
2.08067288
CCNA 01859
CC 1781
TonB-dependent receptor
0.952325524
0.396201107
2.042021612
Hypothetical protein
1.561669498
1.057583431
1.998112954
0.895805402
0.918060892
1.984834238
1.91679215
0.476868236
1.944665541
0
CCNA 03452
ioiC
ratio 7hrs
CCNA 01047
CC 0996
TonB-dependent receptor
CCNA 03615
CC 3500
Sensory transduction protein kinase
CCNA 03711
CC 3597
Ribosome-associated factor Y
1.349967233
1.532145927
1.926277998
CCNA 02878
CC 2789
genetic exchange protein, putative
1.402637664
0.93794196
1.902927308
CCNA 01856
CC 1778
Ferrichrome-iron receptor
0.962564761
0.401363795
1.87920714
CCNA 01860
CC 1782
Rare lipoprotein B precursor
1.03618342
0.416000239
1.862631097
CCNA 02109
CC 2029
Thiamine biosynthesis protein thiC
1.224885033
1.79112399
1.854975752
119
CCNA 03116
Hypothetical protein
1.332060558
1.013076014
1.78314912
CCNA 00615
CC 0579
0
TonB-dependent outer membrane receptor
1.232594643
1.49512401
1.782327282
CCNA 00438
CC 0429
Hypothetical protein
1.579786071
1.911775279
1.7819354
CCNA 00617
CC 0581
Arginine N-succinyltransferase, beta chain
1.472473239
1.730380068
1.725777583
CCNA 02980
CC 2886
Tryptophan 2,3-dioxygenase
1.580825199
1.728150884
1.718261746
CCNA 01857
CC_1779
Transcriptional regulator, AraC family
1.083908724
0.452064112
1.715360258
CCNA 00679
CC 0642
Transposase
1.380382489
1.033341007
1.705955621
CCNA 00338
CC 0335
TonB-dependent receptor
1.104557309
1.399578475
1.701202004
CCNA 01463
CC 1397
3-deoxy-7-phosphoheptulonate synthase
1.040784444
1.123733524
1.692161541
CCNA 00169
CC 0170
Hypothetical protein
0.985867064
1.014085039
1.677168726
CCNA 00206
CC 0206
Succinyl-CoA:3-ketoacid-coenzyme A transferase subunit A
0.948835354
1.279094997
1.675330935
CCNA 02176
0
hypothetical protein
1.406924906
1.131981833
1.667906133
CCNA 01565
0
Hypothetical protein
1.400247606
1.142285876
1.662587283
CCNA 00701
CC 0665
Cold shock protein
1.017091814
1.190449496
1.652068546
CCNA 01853
CC 1775
Hypothetical protein
1.060458236
0.562591035
1.648604677
CCNA 03632
CC 3517
Hypothetical protein
1268023581
1.007536412
1.646172231
CCNA 03214
CC 3115
Conserved hypothetical protein, putative zinc finger
1.059306359
1.319391443
1.64005588
CCNA 01186
CC 1128
Conserved hypothetical protein
1.440512254
1.691422365
1.624831154
Hypothetical protein
1.225924927
0.945390332
1.615574689
Hypothetical cytosolic protein
1.409863859
1.660742965
1.60498804
CCNA 03182
CCNA 01405
0
CC 1343
CCNA 00431
0
Hypothetical protein
1.19443765
1.07568073
1.602931114
CCNA 00364
CC 0359
Deoxyhypusine synthase
0.838919122
0.862513472
1.578716424
CCNA 00771
CC 0734
Hypothetical protein
1.308440444
1.103011659
1.562440006
CCNA 02421
CC 2336
TonB accessory protein exbB
1.374849306
1.494878946
1.555455009
hypothetical protein
1.170952766
0.964539984
1.547902925
CCNA 02892
0
CCNA 03400
CC 3291
Hypothetical protein
1.200632969
1.537628005
1.54774272
CCNA 01298
CC 1240
Protein Translation Elongation Factor Tu (EF-TU)
1.273456624
1.517020044
1.536130561
120
CCNA 02597
CC 2511
Conserved hypothetical protein
1.273044592
1.360933446
1.534474086
CCNA 01476
CC 1410
CRP-family transciiption regulator ftrB
1.391288707
1.008694482
1.528979208
CCNA 00903
CC 0860
inositol transport ATP-binding protein iatA
1.191019225
1.44777301
1.528626277
Hypothetical protein
1.135710163
0.843870738
1.527914166
DNA-binding protein HU
1.637221191
1.676532428
1.523211545
Hypothetical protein
1.131884468
0.85067077
1.511760184
CCNA 03196
CCNA 02416
0
CC 2331
CCNA 01000
0
CCNA 03150
CC 3055
Hypothetical protein
1.486079878
1.053533053
1.510900534
CCNA 01406
CC 1344
Predicted metal-dependent hydrolase
1.291765025
1.583571625
1.508393746
CCNA 03631
CC 3516
1.28304663
1.039062454
1.507391233
CCNA 00620
CC 0584
Hypothetical protein
Acetylornithine aminotransferase / Succinyldiaminopimelate
aminotransferase
CCNA 02177
0
1.459425203
1.759401111
1.507236549
hypothetical protein
1.331446943
1.065392682
1.505977702
CCNA 03444
CC 3336
TonB-dependent receptor
1.324077825
1.426025097
1.502750829
CCNA 03327
CC 3220
Hypothetical protein
1.191454747
1.1596742
1.502593371
CCNA 00678
CC 0641
LSU ribosomal protein L II P
1.308392458
1.395050873
1.500630356
121
Appendix 3B - Genes 1.5 fold or more downregulated after 7 hours phoU depletion
number
CCNA number
CC
CCNA 02488
CC 2405
acetate/3-ketoacid CoA transferase, subunit A
0.030222324
0.033480539
0.03943157
CCNA 02491
CC 2408
Protocatechuate 3,4-dioxygenase beta chain
0.046023887
0.049574526
0.055810985
CCNA 02487
CC 2404
4-hydroxybenzoate 3-monooxygenase
0.042364935
0.037771218
0.059549709
CCNA 02477
CC 2394
Vanillate demethylase oxidoreductase subunit vanB
0.048535881
0.0442215
0.064089624
CCNA 02489
CC 2406
acetate/3-ketoacid CoA transferase, subunit B
0.04990814
0.047423691
0.066502051
CCNA 02476
CC 2393
Vanillate demethylase oxygenase subunit A vanA
0.061629193
0.05744188
0.073434992
CCNA 01168
CC
0.068871847
0.057163749
0.073443332
CCNA 02494
CC 2411
3-ketoacyl-CoA thiolase
3-oxoadipate enol-lactonase / 4-carboxymuconolactone
decarboxylase
0.080267341
0.061650223
0.079182371
CCNA 02490
CC 2407
3-ketoacyl-CoA thiolase
0.075276031
0.066866473
0.081830518
CCNA 02493
CC 2410
3-carboxy-cis,cis-muconate cycloisomerase
0.109565471
0.091872817
0.097571221
CCNA 02600
CC 2515
S-formylglutathione hydrolase
0.103912309
0.104131023
0.112535844
CCNA 00295
CC 0293
Phosphate transport system protein phoU
0.061354361
0.082121265
0.113625724
CCNA 02601
CC 2516
Glutathione-dependent formaldehyde dehydrogenase
0.114541556
0.128604616
0.114437847
CCNA 02495
CC 2412
4-hydroxybenzoate transporter
0.194074631
0.150600619
0.209292221
CCNA 02599
CC 2513
antitoxin protein relB-2
0.265994603
0.273344727
0.288682793
CCNA 02475
CC 2392
Transcriptional regulator vanR
0.28811076
0.322655571
0.349515091
CCNA 02598
Annntttrn
1111
ratin 2hre
ratinn
r ti 7h
Shrs
CC 2512
toxin protein relE-2
0.366564013
0.308208311
0.380911459
CCNA 02492
CC 2409
Protocatechuate 3,4-dioxygenase alpha chain
0.437355664
0.455469714
0.459553926
CCNA 02474
CC 2391
Transcriptional regulator, MarR family
0.386234012
0.426985735
0.460864142
Hypothetical protein
,
CCNA 01176
0
0.655408323
0.614684554
0.530015515
CCNA 01178
CC 1120
Oxidoreductase
0.630290434
0.567372884
0.553820509
CCNA 00214
CC 0214
TonB-dependent receptor
0.507903549
0.505640724
0.557655625
CCNA 01179
CC 1121
Phosphoadenosine phosphosulfate reductase
0.644211023
0.555571133
0.561176306
CCNA 01177
CC 1119
Sulfite reductase (Ferredoxin)
0.689248365
0.608278689
0.591026909
122
CCNA 01664
CC 1592
Potassium-transporting ATPase B chain
0.595458375
0.580201743
0.605718104
CCNA 02004
CC 1927
Hypothetical protein
0.699409624
0.625870358
0.610858406
CCNA 01847
CC 1769
Cytochrome c oxidase assembly protein Surf1
0.615082594
0.436248511
0.617001584
CCNA 01550
CC 1483
Sulfate adenylyltransferase subunit 2
0.701188351
0.672924117
0.622850396
CCNA 01021
CC 0970
TonB-dependent receptor
0.512747385
0.506023785
0.625625238
CCNA 01851
CC 1773
Quinol cytochrome oxidase polypeptide II
0.663725374
0.453892797
0.638501768
CCNA 01549
CC 1482
Sulfate adenylyltransferase subunit
0.660642997
0.639991622
0.644477461
CCNA 01516
CC 1449
NAD(P)H-dependent quinone reductase
0.802099905
0.749208613
0.646054803
hypothetical protein
0.891677089
1.055489259
0.653562932
Quinol cytochrome oxidase polypeptide I
0.690651324
0.431776035
0.659475551
Hypothetical protein
0.900202779
0.942216906
0.668656537
0
CCNA 02706
CCNA 01850
CCNA 01953
CC 1772
0
1/ Adenylylsulfate
kinase
123