View
0
Download
0
Category
Preview:
Citation preview
ArgP’s role in osmoregulation Chapter 3
62
Introduction
Studies on the ArgP protein were initiated in this laboratory because of the
finding that in a Glu synthase (GOGAT encoded by gltBD)- deficient strain (defective in
one of the two pathways of ammonium assimilation and Glu synthesis, the other being
that through Glu dehydrogenase [GDH], see Figure 1.6 in Chapter 1), the argP mutant is
crippled for ammonium assimilation and is osmosensitive. This gltBD argP strain
showed reduced intracellular levels of Glu and GDH compared to a gltBD single mutant
in both low and high osmolarity conditions (Nandineni et al., 2004). Accordingly, it was
hypothesized that (i) the osmosensitivity of the gltBD argP strain is because of limited
Glu as a compatible solute and counter-ion to the potassium that accumulates on
hyperosmotic stress, (ii) ArgP has a role in control of ammonium assimilation and Glu
synthesis pathway in a strain wherein only the GDH pathway was functional. This
control is probably through its direct or indirect regulation of expression of gdhA, the
gene encoding GDH (Nandineni et al., 2004).
Another finding that supports this hypothesis is that by Bender and coworkers
who have shown that gdhA transcription in Klebsiella aerogenes is repressed about 3-
fold upon Lys supplementation; to explain this, these authors postulated that an
unidentified regulator protein which is Lys sensitive activates gdhA in the strain (Goss et
al., 2002). Since E. coli ArgP at its well-studied target gene argO is known to repress
transcription in the presence of Lys, it is likely that this regulator protein is the
orthologous ArgP.
This chapter describes in vivo promoter-lac fusion and in vitro EMSA
experiments performed to investigate the role of ArgP in the transcriptional regulation of
gdhA. Subsequently, the combined effect of ArgP, GltBD and hyperosmotic stress on
gdhA expression was determined. Evidence of increased expression of gdhA relieving the
osmosensitive phenotype of a gltBD argP is provided. The ArgP regulated promoter was
mapped by primer extension and verified by site directed mutagenesis.
Results
3.1 Construction of transcriptional lac fusion to cis regulatory region of gdhA
To enable studies on transcriptional regulation of gdhA, 435-bp of its upstream
regulatory region from 322 to +113 (w.r.t. transcription start site taken as +1) was PCR-
amplified from E. coli genomic DNA using primers 5’-
ArgP’s role in osmoregulation Chapter 3
63
ATTTTGATCCTGCAGAACGCAGCACTG-3’ and 5’-
GTTTGATTCGGATCCCGCTTTTGGACATG-3’(restriction sites in italics) and cloned
at the PstI and BamHI sites of plasmid vector pCU22 (Ueguchi and Mizuno, 1993) which
is a derivative of pUC19 with strong, tandem, phage fd transcription terminators flanking
the MCS useful for preparing supercoiled DNA template for in vitro transcription
reactions. This plasmid was designated as pHYD2601. Subsequently sub-cloning at the
PstI and BamHI sites in pMU575 (Andrews et al., 1991), which is a single copy number
plasmid vector for generating lacZ transcriptional fusions that encodes a trimethoprim
resistance marker, was achieved. The plasmid construct was designated as pHYD2602.
3.2 In vivo expression of gdhA-lac fusion in argP+, argP
d and argP strains
To test whether gdhA expression is under ArgP control, pHYD2602 was
introduced into argP strain derivatives carrying argP+ or the argP
d alleles on pCL1920
(pSC101-based, spectinomycin resistance) vector backbone (Lerner and Inouye, 1990).
-Galactosidase assays were performed after growing strains in minimal A (MA)-glucose
media without or with supplementation of previously reported co-effectors, Arg or Lys at
a concentration of 1 mM. The results from these experiments are presented in Table 3.1.
The data in the following table demonstrated that, (i) ArgP activates gdhA
expression by about 4-fold (compare values of argP+
[Nil] and argP [Nil], see
Variant/ ratio for argP+) (ii) Unlike at argO, Arg addition did not cause further
activation of expression (compare values of Nil and Arg for argP+) but there was an
ArgP-dependent Lys repression of about 3-fold (compare values of Nil and Lys for
argP+, see second row of Nil/Lys ratio). (iii) ArgP
d variants -A68V, -S94L, -V144M and
-P217L behaved more or less like wild-type ArgP (compare ratios Nil/Lys and Variant/
of each of these variants with those of argP+). On the other hand, ArgP
d variants -P108S
and -R295C showed much higher activation than argP+
(about 7-fold, see Variant/ ratio
for these variants) but ArgPd-P274S acted like argP itself in rendering gdhA expression
very low and non-repressible by Lys (see ratios Nil/Lys and Variant/ for ArgPd-P274S
and argP).
ArgP’s role in osmoregulation Chapter 3
64
Table 3.1 Regulation of gdhA-lac expression by ArgP and its variants a
-gal sp. act. Ratio
ArgP
variant Nil Arg Lys Nil/ Lys Variant/Δ
argP 211 198 192 1.0 1.0
argP+ 801 711 307 2.6 3.7
-A68V 743 749 313 2.3 3.5
-S94L 700 687 314 2.2 3.3
-P108S 1502 1401 383 3.9 7.1
-V144M 614 555 278 2.2 2.9
-P217L 609 612 314 1.9 2.8
-P274S 253 219 177 1.4 1.1
-R295C 1512 1360 331 4.5 7.1
a Values shown are of -Galactosidase specific activity (Miller units) after growth in glucose-MA without
(Nil) or with supplementation of Arg (Arg) or Lys (Lys) at 1 mM concentration. The ratios Nil/Lys and
Variant/ indicate, respectively, the degree of Lys repression for the concerned strain, and the degree of
regulation imposed by the concerned ArgP variant (or the wild type ArgP) relative to that in the argP.
3.3 In vitro binding of ArgP to upstream regulatory region of gdhA
The above results indicate that ArgP is a transcriptional regulator of gdhA.
Consequently, the binding ability of ArgP to the upstream regulatory region of gdhA was
checked by EMSA. Varying concentrations of purified, His6-tagged ArgP protein
(Laishram and Gowrishankar, 2007) were incubated with radiolabelled DNA fragment
encompassing the gdhA cis regulatory region (same as that used for generating the lac
fusion construct), before examining the DNA-protein complexes by gel electophoresis
and phosporimaging. The dissociation constants (Kd) for this binding were estimated as
equivalent to the protein concentration (nM) at which half maximal binding of the DNA
was observed by considering the following equation:
ArgP’s role in osmoregulation Chapter 3
65
Kd= [ArgP][DNA] / [ArgP-DNA]
When [ArgP-DNA] equals [DNA] (that is when 50% of DNA fraction is protein bound),
and when [ArgP] >> [ArgP-DNA]
Kd= the concentration of ArgP present in the EMSA reaction…………...Equation 3.1
Where [ArgP] is the concentration of unbound ArgP protein, [DNA] is the concentration
of unbound DNA and [ArgP-DNA] is the concentration of the ArgP protein-DNA
complex.
Figure 3.1 shows that ArgP binds to cis elements at gdhA with an apparent Kd of
80 nM; the binding is inhibited on addition of Lys to the binding reaction and the Kd
increased to 150 nM. This is in contrast to the previous binding profile of ArgP at argO,
where binding is not inhibited by Lys.
Figure 3.1 EMSA with ArgP and cis regulatory regions of gdhA in absence or presence of the co-
effector Lys. ArgP monomer concentrations are indicated for each lane. Bands corresponding to free DNA
and to DNA in binary complex with ArgP are marked by open and filled arrowheads, respectively.
3.4 The combined effect of ArgP, GltBD and high osmolarity on in vivo expression
of gdhA-lac fusion
The next objective in this study was the determination of gdhA expression in the
gltBD argP strain under high osmolarity conditions and evaluation of a probable
correlation of these expression levels with the osmosensitive phenotype of the strain. For
this pHYD2602 (gdhA-lac fusion) was introduced into wild-type (MC4100), argP
(GJ9602), gltBD (GJ4652) and argP gltBD (GJ9611) strains and lac expression was
measured after growing the strains in glucose-MA without ( NaCl) and with (+ NaCl)
supplementation of 0.4 M NaCl. (The plates were also supplemented with 1 mM glycine
ArgP’s role in osmoregulation Chapter 3
66
Table 3.2 Regulation of gdhA expression by ArgP, GltBD and high osmolarity a
Strain genotype NaCl + NaCl
gltBD+
argP+ 1016 331
gltBD+ argP 340 102
gltBD argP+ 1307 998
gltBD argP 540 424
a Values reported are the specific activities of -Galactosidase in derivatives of the indicated strains
carrying gdhA-lac on plasmid pHYD2602, after growth in glucose-MA without ( NaCl) and with (+
NaCl) supplementation of 0.4 M NaCl and 1 mM glycine betaine.
betaine as an osmoprotectant to partially alleviate the growth-inhibitory effects of NaCl).
The results are presented in Table 3.2.
The results obtained suggest that:
(i) ArgP increases gdhA transcription by about 3-fold in both low and high-osmolar
growth media, and in both gltBD+
and gltBD backgrounds (compare column-wise
values in gltBD+
argP+
with gltBD+ argP and values in gltBD argP
+ with gltBD
argP).
(ii) GltBD loss increases gdhA transcription by about 4-fold only at high osmolarity,
in both argP+ and argP backgrounds (consider values in + NaCl column and compare
values of gltBD argP+
with gltBD+
argP+ and values of gltBD argP with gltBD
+
argP).
(iii) Independent of the argP status, high osmolarity decreases gdhA transcription by
about 3-fold in gltBD+ but not in the gltBD background (see rows gltBD
+ argP
+ and
gltBD+ argP and compare across each row, values obtained without and with NaCl).
(iv) At high osmolarity, the activating effects of presence of ArgP and absence of
GltBD are independent and additive, resulting in as high as 10-fold increase in gdhA-lac
transcription (compare values of gltBD argP+
and gltBD
+ argP on NaCl addition).
Taking into account the above mentioned gdhA expression dynamics, as also the
facts that GltBD serves as the primary enzyme for Glu synthesis and that gltBD argP+
is not osmosensitive, it may be concluded that the gltBD argP mutant is osmosensitive
because of loss of the all too important activation of gdhA by ArgP. In a gltBD argP
ArgP’s role in osmoregulation Chapter 3
67
strain, gdhA expression (540 units) is apparently sufficient to meet cellular Glu
requirements under ordinary growth conditions; under hyperosmotic (+ NaCl) conditions,
gdhA expression (424 units) is not sufficient to meet the combined requirements of Glu
both for cellular processes and for serving as a counter-ion for the accumulated
potassium.
3.5 Direct correlation between the ArgPd-effected gdhA expression and
osmosensitivity of gltBD argP
Based on the expression levels effected at the gdhA promoter (Table 3.1), the
ArgPd variants can be classified into three classes namely (i) similar to wild-type (an
example being -A68V), (ii) higher expression levels than wild-type (examples being
-P108S and -R295C), and (iii) constitutively low and argP-like (an example being
-P274S). Following this observation, an attempt was made to determine if there is a
correlation between different extents of gdhA expression and the osmosensitivity of the
gltBD argP mutant strain. Accordingly, the wild-type argP+
or the argPd
alleles, -A68V,
-P108S, -R295C and -P274S on the pSC101-based plasmid vector pCL1920 were
introduced into the gltBD argP (GJ9611) strain. These derivatives were then streaked
on MA-glucose plates containing 0.5 M NaCl without or with 1 mM Lys. (The plates
were also supplemented with 1 mM glycine betaine as an osmoprotectant to partially
alleviate the growth-inhibitory effects of NaCl). The pictures of these plates are shown in
Figure 3.2 and results tabulated in Table 3.3.
The results indicate that strains with high gdhA-lac expression (argP+, -P108S,
-R295C, -A68V) are osmotolerant (see plate [A] and rows Nil and + NaCl corresponding
to these alleles); whereas strains with a low gdhA-lac expression (argP and -P274S) are
osmosensitive (see plate [A] and rows Nil and + NaCl corresponding to this genotype).
Lys addition decreases gdhA-lac expression and this is reflected in the decreased growth
of strains bearing argP+, -P108S, -R295C, -A68V alleles on a plate containing NaCl and
Lys (compare values Nil and Lys, + NaCl and + NaCl + Lys and growth on plate [A] to
growth on plate [B] of these alleles); growth of argP and -P274S strains is more or less
indistinguishable from the previous condition.
Thus there exists a direct correlation between the level of gdhA expression and
the osmosensitivity of the different argPd bearing strains. The most noteworthy result
was for strain derivatives of argPd-P274S, which like argP, showed both low gdhA
ArgP’s role in osmoregulation Chapter 3
68
expression and inability to grow on medium with NaCl (compare values for -P274S and
argP).
Figure 3.2 and Table 3.3 Correlation between the ArgPd-effected gdhA expression and
osmosensitivity of gltBD argP mutant strain b
-gal sp. act.a Growth score
b
argP genotype Nil Lys +NaCl +NaCl +Lys
argP 216 201 + +
argP+ 801 291 ++++ ++
-A68V 765 311 ++++ ++
-P108S 1502 383 ++++ ++
-R295C 1571 314 ++++ ++
-P274S 216 165 + +
a Values indicated are -Galactosidase specific activity as previously described for gdhA-lac in table 3.1.
b The gltBD argP strain carrying argP
+ or the specified argP
d alleles were streaked on MA-glucose plates
with 0.5 M NaCl + 1 mM glycine betaine without (+ NaCl) or with 1 mM Lys (+ NaCl + Lys) and
incubated for 42-48 hrs at 37oC. Growth was scored on the following qualitative 5-point scale (in
increasing order): (no growth), +, ++, +++ and ++++ (full growth).
ArgP’s role in osmoregulation Chapter 3
69
3.6 Multicopy gdhA is able to relieve osmosensitivity of gltBD argP strain
As a corollary of the above result, it was next examined if multicopy gdhA could
relieve the osmosensitivity of gltBD argP strain.
Figure 3.3 and Table 3.4 Effect of multicopy gdhA on osmosensitivity of gltBD argP
strain a
Strain Plasmid
– NaCl
+ NaCl + NaCl
+ IPTG
wild-type
pTrc99A
(vector)
++++ ++++ ++++
pHYD2685
(m.c. gdhA)
++++ ++++ ++++
gltBD argP
pTrc99A
(vector)
++++ + +
pHYD2685
(m.c. gdhA)
++++ ++++ ++++
a The indicated strain derivatives were streaked on MA-glucose plates without (– NaCl) or with 0.5 M
NaCl + 1 mM glycine betaine without (+ NaCl) or with 0.5 mM IPTG (+ NaCl + IPTG) and incubated for
48 hrs at 37oC. Growth was scored on the following qualitative 5-point scale (in increasing order): (no
growth), +, ++, +++ and ++++ (full growth).
The gdhA coding region was PCR-amplified using forward primer JGJgdhAfullF
(5’-ACAAGAATTCCTGCAAAAGCACATG-3’) and reverse primer JGJgdhAfullR (5’-
AACAAAGCTTTGTAGGCCTGATAAG- 3’) to give a 1468-bp DNA fragment.
Utilising the restriction sites that were incorporated in the primers, the fragment was
cloned downstream of the trc promoter at the EcoRI and HindIII site of pTrc99A vector
ArgP’s role in osmoregulation Chapter 3
70
(Amersham Pharmacia) which is a ColE1 based plasmid conferring ampicillin resistance
and with an isopropyl--D-1-thiogalactopyranoside (IPTG)-inducible trc promoter. The
clone was designated as pHYD2685. This plasmid or the pTrc99A vector (serving as
negative control) was transformed into wild-type and gltBD argP strains. The resultant
strain derivatives were streaked on MA-glucose media without or with (i) 0.5 M NaCl, or
(ii) 0.5 M NaCl with 0.5 mM IPTG, and scored for growth after 48 hrs incubation at
37oC (Figure 3.3 and Table 3.4).
It was observed that the gltBD argP strain with pTrc99A was unable to grow
on medium with NaCl (without or with addition of IPTG) whereas pTrc99A bearing
gdhA (pHYD2685) restores growth of the strain on the same medium (without or with
addition of IPTG) (compare rows corresponding to gltBD argP [pTrc99A] and
gltBD argP [pHYD2685] and growth of same strains in panel [B] and [C]). The wild-
type strain showed growth under all tested conditions.
These results show that expression of gdhA from the multicopy pHYD2685,
irrespective of presence or absence of IPTG in the medium alleviates the osmosensitive
phenotype of a gltBD argP strain.
3.7 Mapping of gdhA transcription start site by primer extension analysis
One gdhA transcription start site has previously been mapped to a position 63-bp
upstream of the translation start site. DNA sequence analysis has suggested the existence
of three possible promoters (Valle et al., 1983). Results in the previous section showed
that there are multiple factors (ArgP, GltBD and osmolarity) affecting gdhA
transcription. Hence, it was of interest to find out if gdhA transcriptional regulation
involved a system of multiple promoters, where each promoter is responsive to a specific
factor or growth condition.
To examine this possibility, primer extension assays for mapping probable
multiple transcription start sites of gdhA were performed. Three primers namely
JGJgdhA1 (5’-AGAGAATATGTCTGATCCAT-3’), JGJgdhA2 (5’-
CATGTGCTTTTGCAGTTTTC-3’) and JGJgdhA3 (5’-
ATAACGAGAGTAATCTCATA-3’) end labelled with 32
P-γ-ATP were used; the 3’
ends of these primers were complementary, respectively to +1, 70 and 150 bp w.r.t.
the translation start site of gdhA. These primers were annealed to total cellular RNA
extracted from strains gltBD+
argP+
grown in MA-glucose and ΔgltBD argP+
grown in
ArgP’s role in osmoregulation Chapter 3
71
MA-glucose supplemented with 0.4 M NaCl and 1 mM glycine betaine (high level gdhA
expression was observed for these strains under these conditions, see Table 3.2) followed
by a primer extension reaction with reverse transcriptase. The reaction products were
resolved by electrophoresis on a 6 % urea denaturing polyacrylamide gel and visualized
by phosphorimaging.
Figure 3.4 Primer extension analyses to map transcription start site of gdhA. Strains gltBD
+ argP
+ grown
in MA-glucose and gltBD argP+
grown in MA-glucose with 0.4 M NaCl were used. Of the three primers,
namely JGJgdhA1(1), JGJgdhA2 (2) and JGJgdhA3 (3) used, a transcription start site in both strains and
conditions was mapped only by primer JGJgdhA1 to +63-bp w.r.t. translation start site (shown by arrows).
Lanes A, C, G, T represent sequence ladder with dideoxy ATP or CTP or GTP or TTP in the sequencing
ladder reaction mixture. The sequencing ladder on the left hand side of the gel was generated by using
pUC19 as template and universal M13 forward primer, while the one at the right hand was generated using
pHYD2601 as template and JGJgdhA1.
ArgP’s role in osmoregulation Chapter 3
72
As depicted in the Figure 3.4, a prominent extension product was visible with
primer JGJgdhA1 in RNA preparations from gltBD+
argP+
grown without NaCl and
ΔgltBD argP+ strains grown with NaCl. This was equivalent to the previously reported
transcription start site at 63-bp w.r.t. translation start site. No other extension products
were seen in the case of this primer or of the other two primers in either strains i.e. no
alternative transcription start sites were detected.
3.8 Determination of ArgP-regulated gdhA promoter by site directed mutagenesis
Figure 3.5 Upstream regulatory sequences of gdhA. The mapped transcription start site at –63-bp and the
putative 10 and 35 sequences are shown (A). The mutated 10 sequences in the constructs pHYD2653
(B) and pHYD2654 (C) are also shown.
Figure 3.5 shows that in the sequence upstream of the mapped gdhA transcription
start site, a putative 10 hexamer motif TATAGT was observed (the consensus hexamer
sequence being TATAAT, with more than 90% conservation for second A and last T
McClure, 1985; Browning and Busby, 2004). Situated 19 nucleotides upstream is a
putative 35 hexamer motif TTGCTT (the consensus for which is TTGACA, the first
three residues TTG being more conserved). To determine if this is the ArgP-regulated
promoter, the consensus residues of the putative 10 hexamer motifs were changed to Cs
by a site directed mutagenesis approach. The corresponding nucleotides were modified
on plasmid pHYD2601 template with the aid of primers in which the alternative
nucleotides have been incorporated (shown in primer sequence below) using
QuikChange Site Directed Mutagenesis kit (Stratagene) as described in Materials and
Methods. Thus, the putative 10 sequence was changed at the first and second positions
from TATAGT to CCTAGT (plasmid pHYD2649) with primers JGJgdhAqc1 (5’-
TTCTTGATGGCCTAGTCGAAAAC-3’) and JGJgdhAqc2 (5’-
TTTTCGACTAGGCCATCAAGAATG-3’), and at the last position to TATAGC with
ArgP’s role in osmoregulation Chapter 3
73
primers JGJgdhAqc3 (5’-TGATGGTATAGCCGAAAACTGC-3’) and JGJgdhAqc4 (5’-
GCAGTTTTCGGCTATACCATC-3’). The fragments with mutations incorporated were
then sub-cloned in the PstI and BamHI sites of pMU575 to measure the promoter
activity. The resultant plasmids were designated as pHYD2653 (bearing the TATAGT to
CCTAGT change) and pHYD2654 (bearing the TATAGT to TATAGC change).
The plasmids pHYD2653 and pHYD2654 were transformed into wild-type
(MC4100) and argP (GJ9602) and promoter activity was judged by -Galactosidase
enzyme activity after growth of the derivatives in MA-glucose without and with
supplementation of either 1 mM Lys or 0.4 M NaCl (Table 3.4).
Table 3.4 Determination of ArgP regulated promoter of gdhA a
Strain Plasmid Nil Lys + NaCl
wild-type
pHYD2602 (wt promoter)
1095 402 423
pHYD2653 (mutant 1
promoter)
34 38 35
pHYD2654 (mutant 2
promoter)
43 48 48
argP
pHYD2602 (wt promoter)
399
380 110
pHYD2653 (mutant 1
promoter)
36 33 29
pHYD2654 (mutant 2
promoter)
46 40 42
a Values reported are the specific activities of -Galactosidase of the indicated strains carrying the indicated
plasmids, after growth in MA-glucose without (Nil) and with supplementation of 1 mM Lys (Lys) or 0.4
M NaCl (+ NaCl) and 1 mM glycine betaine.
The above experiment revealed that mutations in the conserved residues of the
putative 10 motif upstream of the mapped transcription site at 63, results in significant
loss of promoter activity both in the wild-type and argP strains in all growth conditions
(compare activity of wild-type promoter on pHYD2602 with that of mutant promoters on
ArgP’s role in osmoregulation Chapter 3
74
pHYD2653 and pHYD2654 in both strains). These observations lead to the conclusion
that this hexamer motif indeed constitutes the authentic 10 element of the functional
gdhA promoter and that ArgP exerts its regulation through this promoter itself.
Discussion
In this study it was shown that ArgP activates gdhA, the gene encoding Glu
dehydrogenase by about 4-fold in a Lys sensitive manner; this clearly correlated with the
manner of ArgP binding to the cis regulatory region. At the time that this study was
being carried out, similar findings were also reported for gdhA in Klebsiella aerogenes
by Goss (2008).
Ensuing work found that gdhA-lac expression is activated 3-fold upon loss of
gltBD. The activating effects of presence of ArgP and loss of gltBD were additive and
resulted in about 10-fold activation. Two lines of evidence clearly implicate a role for
ArgP in osmoregulation through regulation of GDH and hence Glu levels in the cell: (i)
osmosensitive phenotype of the two strains namely, of gltBD argP+ on supplementation
with Lys and of gltBD argP-P274S (harboring the argPd allele that is argP-like w.r.t.
gdhA-lac expression); and (ii) the ability of multicopy gdhA to relieve the osmosensitive
phenotype of the gltBD argP double mutant strain. Primer extension analysis and site
directed mutagenesis also enabled the precise location of the gdhA functional and ArgP
regulated promoter at hexamer motifs located immediately upstream of the sole
transcription start site mapped at 63-bp w.r.t. start site of gdhA translation.
An interesting observation made in this study is that a gene involved in Glu
biosynthesis is regulated by a Lys-sensitive regulator. This suggests a close inter-
dependence between regulation of cellular pools of these two amino acids. In enteric
bacteria, Glu serves the important role of being the primary donor of amino groups in the
biosynthesis of other nitrogenous compounds (Magasanik, 1993) and as the potassium
counter-ion in osmotic homeostasis (Csonka and Epstein, 1996). Of the 20 common
amino acids, Lys is noteworthy in that, despite its relatively high nitrogen content, enteric
bacteria do not readily use Lys as a source of nitrogen or even carbon (Brenchley et al.,
1973; Gutnick et al., 1969; Tyler 1978). Further, Lys can be decarboxylated to
cadaverine (Charlier and Glansdorff, 2004), which is either used as a polyamine or
excreted as a means of neutralizing an acidic environment (Slonczewski and Foster,
1996). Additionally, the size of the intracellular pool of Lys is likely to be related to that
ArgP’s role in osmoregulation Chapter 3
75
of its immediate precursor, diaminopimelate (Park, 1996), which is required for cell wall
biosynthesis.
A related observation is that E. coli permits an exception to the rule of one
synthetase per amino acid for Lys such that it has two distinct genes encoding lysyl
tRNA synthetase activity, namely the constitutively expressed lysS and the inducible
(under high temperature, low pH, anaerobiosis or presence of leucine) lysU (Hirshfield et
al., 1977; Matthews and Neidhardt, 1988; Clark and Neidhardt, 1990; Leveque et al.,
1991; Brevet et al., 1995).These unique roles and features of Lys could account for its
possible use as a sensor for various physiological states of the cell. In fact, Goss (2008)
has proposed that Lys may serve as a stand-in for Glu in signaling the extent of Glu
overflow from osmotic pressure homeostasis into the biosynthetic pathways for other
nitrogenous compounds.
It is also to be noted that, under low pH conditions, E. coli sequentially employs
three amino acid decarboxylases namely the Lys, Arg and Glu decarboxylases
(Slonczewski and Foster, 1996). It may therefore not be coincidental that it is these three
amino acids that are in some way implicated in the ArgP regulon. This is an additional
clue that implies a close relationship between Glu and Lys intracellular pools.
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
76
Introduction
As previously mentioned, the functional role of the ArgP or IciA protein in E. coli
is an enigma. It has been variously described as an inhibitor of chromosome replication
initiation, a transcriptional activator, and a nucleoid-associated protein.
First identified as a protein that binds specific sequences in oriC to inhibit
chromosomal initiation of replication (Hwang and Kornberg, 1990, 1992; Hwang et al.,
1992), ArgP was subsequently described as a transcriptional regulator belonging to the
LTTR family (Thony et al., 1991). Studies from this laboratory has indicated it to be
involved in osmoregulation (Nandineni et al., 2004) and to be essential both in vivo and
in vitro for transcription of the gene encoding the Arg exporter, ArgO (Nandineni and
Gowrishankar, 2004; Laishram and Gowrishankar, 2007). Furthermore, ArgP-regulated
transcription in vivo has been demonstrated for the genes dapB of E. coli (Bouvier et al.,
2008) and gdhA of Klebsiella aerogenes (Goss, 2008). As described in Chapter 3, ArgP
is a transcriptional regulator of gdhA in E. coli also, which likely explains its role in
osmoregulation. In addition, in vitro studies have suggested that ArgP activates the
transcription of the dnaA and nrdA genes involved in DNA replication and metabolism
(Lee et al., 1996, 1997; Han et al., 1998).
As mentioned above and in Chapter 1 (Section 1.2.7 c), the argO gene, coding for
an Arg exporter, is one of the well-characterized targets for transcriptional regulation by
ArgP (Nandineni and Gowrishankar, 2004; Laishram and Gowrishankar, 2007; Peeters et
al., 2009). Transcription of argO in vivo is activated by both Arg as well as its toxic
analog CAN, both effects being mediated by ArgP. On the other hand, argO transcription
is drastically reduced in medium supplemented with Lys, to a level equivalent to that
observed in ΔargP mutants. Dominant, gain-of-function mutations in argP have been
identified (designated argPd) that confer a CAN
r phenotype and act in trans to
considerably increase argO transcription in vivo over that obtained in the argP+ strain
(Nandineni and Gowrishankar, 2004).
In an attempt to get a better insight into the role of ArgP as a transcriptional
regulator in E. coli, a whole genome differential gene expression microarray was
performed. This chapter describes the methodology used for short-listing and identifying
from the microarray data, genes activated by ArgP and their subsequent validation. It
also describes in vivo and in vitro experiments to demonstrate and describe the features
of ArgP transcriptional regulation of the Lys-specific permease, lysP.
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
77
Results
Section A: Differential gene expression microarray to identify ArgP-
regulated genes
4.1.1 Details of strains used for microarray profiling
The strains provided for the microarray-based comparison of whole-genome
mRNA abundances were an isogenic pair of strains that was either ΔargP (GJ9602) or
expressing the dominant gain of function ArgPd-S94L variant in the argP
+ strain
(MC4100/pHYD926). It was expected that the ArgPd variant would activate expression
of all ArgP-regulated genes irrespective of their need for other co-effectors in the
medium, whereas these genes would remain un-activated in the ∆argP mutant. The
strains were grown to exponential phase in Arg-supplemented MA-glucose medium. The
RNA extraction, microarray hybridization and initial data processing were performed by
Genotypic Technology Pvt. Ltd., Bengaluru.
4.1.2 Methodology for short-listing candidate genes and validation of microarray
data to identify genes activated by ArgP
In order to consolidate the data from the microarray experiment, genes with high-
level mRNA expression in MC4100/pHYD926 compared to GJ9602 and showing log2-
fold difference ≥ 1.0 were further considered. This derived list comprising a total of 344
genes (Appendix I) represented the probable genes activated, at least 2-fold, by ArgP.
Similarly, a second list of genes with high-level mRNA expression in GJ9602 compared
to MC4100/pHYD926 and showing log2-fold difference ≥ 1.0 was obtained. This
comprised a total of 738 entries (Appendix II) and represented probable genes repressed
at least 2-fold by ArgP. However, since ArgP has been previously described as a
transcriptional activator, it was decided to initially explore its regulation of the genes in
the first (activated genes) list.
Of the genes previously reported to be ArgP regulated, dnaA and nrdA did not
feature in the list of activated genes, but argO, dapB and gdhA showed log2-fold
differences of 2.3, 2.7, and 1.3 respectively. A common feature of regulation among
these three genes is that their expression is repressed in the presence of Lys. Hence, other
genes were then short-listed in this list that were previously reported to be Lys-
repressed, as candidates for activation by ArgP and for further validation experiments.
These genes comprised of (gene function and log2-fold microarray expression difference
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
78
between argPd
and ΔargP in parentheses) lysP (Lys-specific permease, 2.3), lysC (Lys-
sensitive aspartokinase, 1.9) and asd (aspartate semialdehyde dehydrogenase, 1.1). The
proposed strategy for validation was (i) to construct promoter-lac fusions of these genes
followed by β-Galactosidase assays in argP+, argP
d and argP strain backgrounds; and
(ii) for promoters showing in vivo ArgP activation in the lac fusion experiments, to
undertake EMSA of regulatory regions in vitro. This strategy is schematically depicted in
Figure 4.1.
Figure 4.1 Methodology for short-listing candidate genes and validation of microarray data
Section B: ArgP-mediated regulation of the Lys-specific permease lysP
Using the above-described methodology and as described in detail below, it was
found that the Lys-specific permease lysP had the highest fold activation in argPd
compared to argP strain backgound. Accordingly, the following systematic in vivo and
in vitro studies were carried out to examine the features of lysP regulation by ArgP.
4.2.1 Thialysine resistance test of argP strain
As described in Chapter 1, two systems for the transport of Lys exist in E. coli
(Patte, 1996): (i) the Lys-Arg-ornithine (LAO) system, transports Lys and ornithine and
is inhibited by Arg, although Arg is not a substrate (Rosen, 1971, 1973) (ii) the Lys-
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
79
specific permease (LysP) system, is inhibited by the Lys analog S-(-aminoethyl)-L-
cysteine (thialysine or thiosine) but not by ornithine or Arg (Rosen, 1971). lysP mutation
confers resistance to thialysine such that a lysP strain grows on a plate containing up to
100 g/ml thialysine, whereas the wild-type strain fails to grow (Steffes et al., 1992).
It was reasoned that if lysP is regulated by ArgP, then a argP mutant would also
be thialysine-resistant since lysP will not be expressed in the strain. Hence, preliminary
phenotypic tests to check for thialysine resistance (Tlr) of argP strain were performed.
On a MA-glucose plate containing 100 g/ml of thialysine, it was found that incubation
for 24 hours at 37C allowed comparable growth of lysP, argP and arg lysP but
not of the wild-type strain (Figure 4.2). Thus, the argP strain is Tlr i.e. it phenocopied
the lysP strain suggesting that ArgP may regulate lysP. The observation that argP
lysP is not more Tlr
than lysP suggests that lysP and argP perhaps act in a single
pathway to mediate thialysine resistance.
Figure 4.2 Thialysine resistance test of argP strain. Wild-type (wt), lysP, argP and argP lysP
strains were streaked on MA-glucose plates without and with 100 g/ml thialysine and growth was
observed after incubation for 24 hours at 37C.
4.2.2 Construction of lysP-lac fusion and in vivo expression in argP+, argP
d and
argP strains
Next, the promoter-lac fusion of the lysP gene was constructed. A 263-bp
amplicon comprising the lysP regulatory region from 219 to +43 relative to
transcription start site taken as +1, and expected to encompass all the cis regulatory
sequences, was PCR-amplified using forward primer JGJlysPF (5’ -
GCGCTTTCTGCAGTATTGCGATCC-3’) and reverse primer JGJlysPR (5’-
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
80
TAGTTTCGGATCCCATACAAAAATGC-3’). The primers were designed to
incorporate PstI and BamHI sites at the upstream and downstream ends respectively, of
the amplicon (shown in italics in the sequences in parentheses). From the 263-bp
amplicon, a 238-bp fragment extending from –206 to +32 of lysP promoter region was
obtained after digestion with PstI and BamHI and was cloned at the same sites of the
plasmid vector pMU575 (Andrews et al. 1991). The resultant plasmid was designated as
pHYD2636. This plasmid was used in all the subsequent in vivo β-Galactosidase assays
to report for lysP promoter activity in different strain backgrounds.
Plasmid pHYD2636 (lysP-lac) was transformed into argP (GJ9602) strain
carrying either argP+, or any of the argP
d alleles on the pSC101-based plasmid vector
pCL1920; similar derivatives of argP with vector pCL1920 alone served as negative
controls in these experiments. The β-Galactosidase specific activities of these derived
strains are shown in Table 4.1.
The results in Table 4.1 indicated that:
(i) lysP-lac expression was very low in the argP strain in all media tested but it
was substantially elevated in the argP+
or argPd
bearing strains. In fact the fold activation
of lysP transcription by wild-type ArgP was as high as 34-fold (compare Variant/Δ value
of argP+
and argP).
(ii) Transcription from the lysP promoter in the strains bearing argP+
or argPd
variants showed no further activation on Arg supplementation (compare columns Nil and
Arg for each of these strains). This indicated that the co-effector Arg has no effect on lysP-
lac expression.
(iii) lysP-lac expression exhibited repression upon Lys supplementation in the argP+
or argPd variants but not the argP strain background (compare Nil/Lys ratios of argP
+ or
argPd variants with that of argP), suggesting that Lys repression of this promoter is
ArgP-mediated.
(iv) Among the argPd alleles, -A68V, -S94L, -V144M, -P217L showed similar behavior to
wild-type i.e. about 20- to 30-fold activation and 3- to 5-fold Lys repression; -P108S and
-R295C were somewhat different by showing about 30-fold activation but only 2-fold Lys
repression (compare Variant/ and Nil/Lys ratios of argP+
with that of these argPd
variants). However, the most striking difference was presented by -P274S that showed
much lower activation (of about 10-fold) compared to argP+ and 3-fold Lys repression
(compare Variant/ and Nil/Lys ratios of argP+
with that of -P274S).
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
81
Table 4.1 Regulation of lysP-lac expression by ArgP and its variants a
-gal sp. act. Ratio
ArgP
variant Nil Arg Lys Nil/ Lys Variant/Δ
argP 51 50 48 1.0 1.0
argP+ 1731 1592 325 5.3 34
-A68V 1558 1455 587 2.6 31
-S94L 1561 1653 473 3.3 30
-P108S 1378 1434 922 1.4 27
-V144M 1174 1136 251 4.6 23
-P217L 881 858 166 5.3 17
-P274S 499 465 169 2.9 10
-R295C 1938 1883 987 1.9 38
a Values shown are of -Galactosidase specific activity (Miller units) after growth in glucose-MA without
(Nil) or with supplementation of Arg (Arg) or Lys (Lys) at 1 mM concentration. The ratios Nil/Lys and
Variant/ indicate, respectively, the degree of Lys repression for the concerned strain, and the degree of
regulation imposed by the concerned ArgP variant (or the wild-type ArgP) relative to that in the argP.
4.2.3 In vitro binding of ArgP to the lysP upstream regulatory region
The simplest explanation to account for the above in vivo observation on lysP
regulation is that of direct regulation of the lysP promoter by ArgP. To examine this
possibility, the purified His6-tagged ArgP protein (Laishram and Gowrishankar, 2007)
was tested for its ability to bind DNA of the lysP regulatory region. An EMSA was
undertaken using the 263-bp amplicon (–219 to +43) described above.
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
82
Figure 4.3 Binding of His6-ArgP protein to 263-bp (219 to +43) lysP fragment in absence (A) or
presence of co-effectors, Arg (B) or Lys (C) shown by EMSA. ArgP monomer concentrations are indicated
for each lane. In Figure 4.3 (A), lanes CS and NS depict results of addition of 100-fold excess of unlabelled
specific (263-bp lysP fragment) or non-specific (253-bp fragment from the ilvG locus) competitor DNA,
respectively, to the EMSA reactions undertaken with 160 nM ArgP. The previously described 427-bp argO
DNA fragment (Laishram and Gowrishankar, 2007) was used as positive control for ArgP binding. Bands
corresponding to free DNA and to DNA in binary complex with ArgP, are marked by open and filled
arrowheads respectively.
The fragment was radiolabelled as described in Chapter 2, Section 2.2.4.5, and it
was incubated with varying concentrations of purified ArgP (from 0 to 160 nM) in
absence (Figure 4.3 A) or presence of Arg (Figure 4.3 B) or Lys (Figure 4.3 C), before
being subjected to electrophoresis on a 5% non-denaturing polyacrylamide gel and then
being visualized by phosphorimaging.
It was observed that ArgP binds the lysP regulatory region in absence of any
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
83
amino acids and that the addition of Arg (at a concentration of 0.1 mM) did not hinder
the binding. However, the presence of Lys (at a concentration of 0.1 mM) in the reaction
considerably reduced ArgP binding. The results in Figure 4.3 suggest that ArgP binds the
lysP regulatory DNA specifically, such that, it is competed by excess (100-fold) of the
cognate unlabelled fragment (CS) but not so by non-specific unlabelled DNA fragment
(NS: NS was a 253-bp PCR amplicon from the ilvG coding region obtained using
primers JGBPMAILVGFP1, 5’- CAGCAACACTGCGCGCAGCTGCGTGATC-3’ and
JGBPMAILVGRP1 5’-CTTGTGCGCCAACCGCCGCCGGTAAAC-3’).
4.2.4 Binding curves and dissociation constants for ArgP and lysP interaction in
absence and presence of Arg and Lys
To obtain a quantitative estimate of the strength of the above-described binding,
the distribution of radioactivity of unbound DNA in Figure 4.3 A, B and C was
quantitated by densitometry of the autoradiograms. Taking the value of radioactivity in
the lane without ArgP protein as total or hundred percent radioactivity, ‘% unbound
fraction’ for each lane was calculated. From this ‘% bound fraction’ (which is the
fraction of the total radioactivity of lysP DNA that is shifted upon ArgP binding) was
derived as ‘100 % unbound fraction’. Binding curves were obtained by plotting the
values of ‘% bound fraction’ as a function of the added ArgP concentration (Figure 4.4).
A hyperbolic curve was obtained in the absence of Arg or Lys, suggesting that the
binding of ArgP to lysP DNA is non-cooperative. In presence of Arg, there was no
significant change in the nature of protein binding, whereas when Lys was added, a sharp
decrease in the slope of the curve was seen. By considering Equation 3.1, previously
described in Chapter 3, the dissociation constants (Kd) were calculated from the various
binding curves obtained in Figure 4.4 as equivalent to the protein concentration at which
half maximal binding was observed.
The calculations revealed that the Kd of ArgP binding to lysP DNA was about 40
nM in absence of Arg or Lys, 50 nM in presence of Arg and >160 nM in the presence of
Lys. It can be concluded that whereas Arg addition does not cause any significant change
in the binding affinity, Lys causes at least a 5-fold decrease in affinity of binding of ArgP
to lysP.
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
84
Figure 4.4 Binding curves for ArgP binding to lysP DNA in the absence (Nil) or presence of co-effectors
Arg or Lys. These curves were established from the EMSA experiment shown in Figure 4.3.
4.2.5 Multiple sequence alignment of upstream region of lysP with those of
previously identified ArgP targets (argO and dapB)
To further investigate the extent of ArgP binding on the lysP upstream region, a
multiple sequence alignment of lysP upstream regulatory region (extending from its
intervening intergenic region with yeiE to the 10 sequence), with DNase I footprinted
region of ArgP at argO (Laishram and Gowrishankar, 2007) and the ArgP binding
sequence at dapB (Bouvier et al., 2008), was executed using Clustal 2.1 available on
www.genome.jp/tools/clustalw (Figure 4.5).
Considerable sequence identity of the three compared sequences (denoted by
asterisks), extending till –98 (w.r.t. transcription start site taken as +1) of lysP, was seen.
This observation indicates of a probable common ArgP binding site within the upstream
region of these genes. Further, it suggests that sequences up to –98 of lysP may be
sufficient for ArgP regulation. In fact, an examination of dapB (93 to 48) and lysP
(91 to 47) sequences reveals a pair of closely resembling imperfect inverted repeats in
each. As with other LTTRs (Maddocks and Oyston, 2008) the promoter distal and
proximal sequences were designated as the Recognition Binding Site (RBS) and
Activation Binding Site (ABS), respectively (Figure 4.6).
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
85
Figure 4.5 Clustal 2.1 alignment of lysP regulatory region (encompassing its intervening intergenic
region with yeiE to 10 sequence) with argO (80 to 25) and dapB (104 to 10 sequence). Considerable
sequence identity was seen till 98 of lysP. Underlined are the 35 sequence of argO and the 35 and 10
sequence of dapB and lysP. Asterisks indicate sequence identity.
Figure 4.6 Clustal 2.1 alignment of regulatory regions of lysP
(91 to 47) and dapB (93 to 48) revealing ~ 57% identity (A). Probable ArgP Recognition Binding Site
(RBS) and Activation Binding Site (ABS) boxed in the dapB and lysP sequences based on the closely
resembling imperfect inverted repeats in each (B).
4.2.6 In vivo lac-fusion assays of nested deletion constructs of lysP regulatory region
Based on the sequence alignment described in the previous section, it was
decided to construct two nested deletion products of the 238-bp lysP DNA fragment.
Both these fragments extended to the same base (+32) as the 238-bp fragment at the
downstream end but were different and truncated to various extents at the upstream end.
The first truncated fragment extended from 114 to +32 and was generated by PCR
amplification, using forward primer JGJlysPFP1 (5’-
ACAACTGCAGTTCGCCAGAAAA-3’) and reverse primer JGJlysPR (5’-
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
86
TAGTTTCGGATCCCATACAAAAATGC-3’). The second truncated fragment extended
from 76 to +32 and was generated by PCR amplification, using forward primer
JGJlysPFP2 (5’-ACAACTGCAGCTGGGCGATCAT-3’) and reverse primer JGJlysPR
(5’-TAGTTTCGGATCCCATACAAAAATGC-3’). The fragments were cloned into the
PstI and BamHI site of plasmid vector pMU575 (Andrews et al., 1991). The respective
plasmids were designated as pHYD2647 and pHYD2648. Either of these plasmids or
pHYD2636 (bearing the 238-bp lysP fragment and serving as a control plasmid) were
introduced by transformation into argP+
(MC4100) and argP (GJ9602) strains. -
Galactosidase assay of the resultant strains was done in MA-glucose media without and
with supplementation of Lys. The results are as shown in Table 4.2.
Table 4.2 ArgP regulation of lac fusions to nested deletions of lysP a
____________argP+__________ ___________argP__________
Plasmid
(lysP extent) Nil Lys Nil Lys
pHYD2636
(219 to +32) 802 150 31 30
pHYD2647
(114 to +32) 852 122 52 50
pHYD2648
(76 to +32) 43 47 41 42
a Values reported are the specific activities of -Galactosidase in derivatives of MC4100 (argP
+) or
GJ9602 (ΔargP) carrying the indicated lysP-lac fusion plasmids, after growth in glucose-MA without (Nil)
or with (Lys) Lys supplementation at 1 mM. Extents of lysP shown are in bp relative to start-site of
transcription.
It was observed from values in Table 4.2 that, the lysP(114 to +32)-lac
expression (like the previously described lysP[219 to +32]-lac) was activated
tremendously (17-fold) in the presence of ArgP (compare Nil values of argP+
and
argP strains for the two lac fusions); an 8-fold, ArgP-dependent Lys repression was
also evident (compare Nil, Lys values in argP+
and argP strains for the two lac
fusions). On the other hand, the lysP(76 to +32)-lac fusion showed low levels of lac
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
87
expression in both argP+
and argP strain backgrounds (see values of argP+
and argP
strains under both Nil, Lys conditions for this lac fusion) and did not behave like
lysP(219 to +32)-lac. This suggests that the lysP(76 to +32)-lac fusion no longer bears
the ArgP responsive cis element. Thus, the lysP upstream region between 114 and 76
carries critical sequence determinants for regulation of lysP by ArgP.
4.2.7 In vitro EMSA experiments with nested deletion fragments of lysP regulatory
region
To determine if the above in vivo observations could be supported by in vitro
ArgP binding data, EMSA experiments using each of the DNA fragments, i.e. lysP (114
to +32) and lysP (76 to +32) and varying concentrations of the ArgP protein, were
performed. As can be seen from Figure 4.7, lysP (114 to +32) showed similar binding
affinity and Lys sensitive ArgP binding as that described in Section 4.2.3 for the 263-bp
(219 to +43) fragment. In contrast, the lysP (76 to +32) fragment showed insignificant
ArgP binding even at high ArgP concentrations used, irrespective of the absence or
presence of Lys (Figure 4.8). These findings corroborate the previous in vivo
observations and suggest that sequences up to 114 are sufficient for ArgP binding and
regulation, and that the sequences between 114 to 76 house an important determinant
for ArgP binding and ArgP mediated regulation of lysP.
Figure 4.7 ArgP binding to lysP (114 to +32) DNA in the absence (A) or presence of 0.1 mM Lys (B).
ArgP monomer concentrations are indicated for each lane. Bands corresponding to free DNA and DNA in
binary complex with ArgP are marked by open and closed arrowheads respectively.
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
88
Figure 4.8 ArgP binding to lysP (76 to +32) DNA in the absence (A) or presence of 0.1 mM Lys (B).
ArgP monomer concentrations are indicated for each lane. Bands corresponding to free DNA are marked
by open arrowheads.
Section C: Role of ArgP in cadBA regulation
As discussed in Chapter 1 (Section 1.1.5), the lysP gene, which encodes the Lys-
specific permease, is also described in literature as cadR, because it has been implicated
in transcriptional regulation of the Lys decarboxylation system encoded by the two gene
operon, cadBA (cadB codes for Lys decarboxylase and cadA codes for the Lys-
cadaverine antiporter). cadBA expression is induced only in conditions of low pH,
anaerobiosis and in the presence of Lys and is dependent on a membrane bound
activator, CadC (Neely et al., 1994; Tetsch et al., 2008). In lysP defective mutants,
cadBA is induced at low pH even in absence of Lys, and the model is that LysP
negatively regulates cadBA in absence of Lys by sequestering CadC (Neely et al., 1994;
Tetsch et al., 2008). Since, it was found that ArgP is needed for activation of lysP
transcription, it was pertinent to determine if cadBA expression would be rendered Lys
independent even in the argP strain. Accordingly, the following experiments were
performed.
4.3.1 Construction of cadBA-lac promoter fusion
A cadBA-lac fusion was first constructed by PCR-amplifying cadBA upstream
regulatory region using forward primer JGJcadBF (5’-
TTTAATTTACCTGCAGGGGCAAAC-3’) and reverse primer JGJcadBR (5’-
TGCAATACCGGATCCCATCATATTA-3’); PstI and BamHI restriction sites were
incorporated into the forward and reverse primer respectively and are shown in italics in
the primer sequences. Following restriction digestion of the PCR amplicon with the same
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
89
enzymes, a 527-bp fragment extending from 387 to +140 with respect to transcription
start site of cadB, taken as +1, was cloned at the PstI and BamHI sites of vector pMU575
(Andrews et al., 1991). The resultant plasmid was designated as pHYD2674.
4.3.2 Effects of CadC, LysP, and ArgP on cadBA-lac expression
Next, the plasmid pHYD2674 bearing cadBA-lac fusion was introduced by
transformation into the following strains: wild-type (MC4100), lysP (GJ9623), cadC
(GJ9647), argP (GJ9602), lysP cadC (GJ9649), argP cadC (GJ9648), argP
lysP (GJ9624). -Galactosidase assays were performed after growth in MA-glucose
based media of pH 7.4 or pH 5.8, supplemented either with 19 amino acids other than
Lys (Nil) or with 19 amino acids and 10 mM Lys (Lys). The results of this assay are
shown in Table 4.3.
Table 4.3 Effects of CadC, LysP, and ArgP on cadBA-lac expression a
________ pH 7.4___________ _________pH 5.8____________
Strain Nil Lys Nil Lys
wild-type 25 28 24 98
lysP 27 27 94 96
cadC 25 26 26 24
argP 23 22 26 96
lysP cadC 21 28 24 26
argP cadC 25 26 24 28
argP lysP 27 24 87 99
a Values reported are the specific activities of -Galactosidase in the indicated strain derivatives carrying
the cadBA-lac fusion plasmid pHYD2674, after growth in MA-glucose based medium of pH 7.4 or pH 5.8
supplemented with: Nil, 19 amino acids (other than Lys); or Lys, 19 amino acids and 10 mM Lys.
The results show that, as previously reported, cadBA expression was induced only
at low pH, in the presence of Lys (compare Nil, Lys values derived at pH 7.4 for all
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
90
strains with Lys value of wild-type at pH 5.8) and the CadC transcriptional activator
(compare Lys values of wild-type at pH 5.8 with Lys values of cadC at pH 5.8); lysP
mutation rendered its expression constitutive only in the presence of CadC (compare Nil
values at pH 5.8 of wild-type, lysP, cadC and lysP cadC). However the argP
mutant failed to phenocopy the lysP mutant (compare Nil, Lys values of lysP, argP
and argP lysP strains at pH 5.8) in that cadBA expression continued to be Lys-
dependent in the former. Hence, ArgP does not seem to have a role in the regulation of
the cadBA operon.
Section D: In vitro binding studies of the ArgPd proteins -S94L, -P108S
and -P274S with the lysP regulatory region
4.4.1 Cloning of argPd alleles -S94L, -P108S and -P274S in protein expression vector
pET21b
As can be seen in Table 4.1 and as described in section 4.2.2, the argPd alleles
can be classified into three categories depending on the effects on lysP transcription (i)
alleles (-A68V, -S94L, -V144M, -P217L) that were more or less like wild-type with 20-
to 30-fold activation and 3- to 5-fold Lys repression; (ii) alleles (-P108S and -R295C)
that showed about 30-fold activation but only 2-fold Lys repression; and (iii) allele
-P274S that showed much less activation of about 10-fold and 3-fold Lys repression. To
find out if these in vivo regulatory differences exhibited by the argPd alleles compared to
the wild-type, were an outcome of differences in native ArgP and ArgPd binding patterns
at this locus, the ArgPd
proteins -S94L, -P108S and -P274S, each being representative of
one of the above three mentioned categories, respectively, were overexpressed and
purified.
As templates for PCR amplification of the argPd
alleles -S94L, -P108S and
-P274S, plasmids pHYD926, pHYD927 and pHYD2606 were respectively used. These
plasmids are derived from the low-copy-number cloning vector pCL1920 and contain a
1.86 kb SalI fragment from the E. coli chromosome which encompasses the argP locus
bearing the mutant alleles -S94L, -P108S and -P274S along with the native promoter
(Nandineni and Gowrishankar, 2004). The primers used for PCR amplification of the
920-bp protein coding region of each of the three argPd variants were JGARGP1r (5΄-
AGCAGACAACACATATGAAACGCCCGGA-3΄) as forward primer and JGARGP3r
(5΄-ATTATTTGATCTCGAGATCCTGACGAAG-3΄) as reverse primer; restriction
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
91
enzyme sites pertaining to NdeI and XhoI were respectively incorporated in the forward
and the reverse primers which are shown in the primer sequences in italics. The ATG
sequence of the NdeI site corresponds to the initiation codon of ArgPd proteins, and the
CGA sequence of the XhoI site on the reverse primer, eliminates the termination codon
of ArgPd proteins. The three resultant amplicons encoding the full length ArgP
d proteins
-S94L, -P108S and -P274S were then digested with NdeI and XhoI and cloned at the
same sites in plasmid pET21b (a medium-copy-number ColE1-based vector, with a
strong T7 promoter and an efficient ribosome binding site lying upstream of the NdeI site
in the multiple cloning site region) (Novagen, EMD Biosciences). The plasmids thereby
derived for argPd alleles -S94L, -P108S and -P274S were designated as pHYD2678,
pHYD2679, pHYD2680 respectively. The sequence of each of the three plasmids was
verified by DNA sequencing. Since the stop codon was abolished by incorporating XhoI
restriction enzyme site in the reverse primer, the reading frame of the alleles has been
extended in the plasmids to include, and to terminate just beyond, the His6-codons
situated after the multiple cloning site region of the vector. The proteins thus have a His6-
tag at the C-terminal end to enable protein purification with the aid of Ni-NTA affinity
chromatography.
4.4.2 Functional complementation studies with pHYD2678, pHYD2679 and
pHYD2680 encoding His6-ArgPdS94L, His6-ArgP
dP108S and His6-ArgP
dP274S
respectively
For functional complementation studies of the His6-tagged ArgPd
constructs, test
for CAN tolerance was done. It has been previously reported (Nandineni and
Gowrishankar, 2004) that a strain with wild-type argP allele cannot grow on medium
with CAN above a concentration of 40 g/ml i.e. it is CANs. However, a strain with the
argPd alleles can grow on medium supplemented with 65-100 g/ml of CAN i.e. it is
CANr. This is correlated with the ability of the argP
d alleles to activate the Arg exporter,
argO, to a much higher degree than wild-type argP. CAN resistance is also exhibited by
an argR
strain, in which there is derepression of the Arg biosynthetic pathway and CAN
resistance is achieved because of inability of CAN to compete with the excess cytosolic
Arg. For the CAN tolerance test, plasmids pHYD2678, pHYD2679 and pHYD2680 were
transformed into argP strain (GJ9602) and the ampicillin resistant colonies were
streaked on 0.2% glucose-MA agar plates containing 65 μg/ml CAN. Strains argR
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
92
(GJ4748) harboring plasmid vector pET21b and argP (GJ9602) carrying pHYD1705
(wild-type ArgP cloned in pET21b) served as positive and negative controls respectively.
After 36 hours incubation at 37C, as expected, it was found that the positive control and
the strains harboring the argPd plasmids exhibited CAN resistance whereas the strain
carrying wild-type ArgP failed to grow. This indicates that pHYD2678, pHYD2679 and
pHYD2680 can recapitulate the previously reported argPd phenotype and that the C-
terminal His6-tag does not affect function of the ArgPd proteins.
4.4.3 Overexpression and purification of ArgPd proteins -S94L, -P108S and
-P274S
Strain BL21 (DE3) was used for overexpression of ArgPd proteins -S94L,
-P108S and -P274S. In this strain, the T7 RNA polymerase gene is under the control of
the lac repressor-operator system that is induced by IPTG (Studier and Moffatt, 1986).
The plasmids, pHYD2678, pHYD2679 and pHYD2680 were transformed into BL21
(DE3) and protein overexpression and purification was achieved by the procedure
Figure 4.9 Overexpression and purification of His6-ArgP
dS94L, His6-ArgP
dP108S, His6-ArgP
dP274S,
demonstrated by Coomassie blue staining following sodium dodecyl sulfate-polyacrylamide gel (12%)
electrophoresis. Lanes 1, 2 indicate cell lysates of strain BL21 (DE3) carrying plasmid vector, pET21B,
without (U) and with induction (I) with 1mM IPTG respectively. Similarly, lanes 3 to 11 indicate
uninduced (U), induced (I) cultures and purified (P) proteins pertaining to ArgPdS94L, ArgP
dP108S,
ArgPdP274S respectively. Lane 12 shows the profile of protein molecular weight marker (M) of sizes in
kilodaltons as indicated alongside.
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
93
described in the Material and Methods Section 2.2.3.1. Briefly, the strains were grown in
LB to an A600 of around 0.6, were then induced with 1 mM IPTG and harvested after 4-
hrs of induction. Bacterial cells were lysed by sonication and protein purification was
achieved by passing the lysate through Ni-NTA (Qiagen) chromatographic column.
Figure 4.9 shows overexpression and purification of ArgPd proteins -S94L, -P108S and
-P274S. It can be estimated that the extent of protein purification achieved in each case
was about 95 %.
4.4.4 EMSA studies of the ArgPd proteins -S94L, -P108S and -P274S with the lysP
regulatory region
To determine the binding pattern of ArgPd proteins -S94L, -P108S and -P274S at
lysP, EMSA was performed in which varying concentrations of ArgPd proteins (0 to 300
nM) were incubated with the radiolabelled, 263-bp long, lysP DNA. -S94L, -P108S and
-P274S binding at lysP was more or less comparable to native ArgP: firstly, in terms of
binding affinities, approximate Kds being 60 nM, 100 nM and 60 nM respectively, the
previously determined Kd for native ArgP being about 40 nM (Figure 4.10 A) and
secondly with respect to reduced affinity of binding to the lysP regulatory region in the
presence of Lys (Figure 4.10 B).
Figure 4.10 EMSA of 263-bp lysP fragment with ArgP (WT) or its variants -S94L, -P274S and
-P108S at indicated protein concentrations in the absence (A) or presence (B) of Lys (0.1 mM). Bands
corresponding to free DNA and to DNA in binary complex with each of the ArgP proteins are marked by
open and filled arrows respectively.
These observations suggest that the differences in in vivo regulation seen
between strains harboring argP or the argPd alleles cannot be simply correlated with
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
94
binding affinities. This is particularly true for -P274S, which is least effective for lysP
activation but shows proficient binding to the regulatory element.
A striking difference that was seen between native ArgP and the ArgPd variants in
these EMSA experiments was that the DNA-protein complexes formed by all the three
ArgPd variants migrated faster than those formed by native ArgP. It has been previously
reported that such migration differences of DNA-protein complexes are a result of the
differences in DNA bending such that less bent DNA produces faster migrating
complexes than highly bent DNA (Wu and Crothers, 1984). For a number of LTTRs
(Colyer and Kredich, 1996; Parsek et al., 1995; Akakura and Winans, 2002; Dangel et
al., 2005) less bent DNA in the DNA-protein complex is associated with higher
activation. Figure 4.10 reveals that the fastest mobility was shown by the DNA-protein
complex formed by -P274S, followed by -P108S and -S94L; this suggests that -P274S
induces the least DNA bending. However this observation was inconsistent with the
finding that -P274S exhibited the least extent of lysP activation.
SECTION E: Interplay of other transcription factors in regulation of
lysP
4.5.1 Role of LysR in lysP regulation
ArgP belongs to the LTTR family of which LysR is the prototypic member
(Maddocks and Oyston, 2008). Since LysR has also been reported to mediate the
repression by Lys at lysA (Stragier et al. 1983a, 1983b), we tested for the possibility of
an interplay between the two transcription factors at lysP in vivo.
4.5.1.1 Construction of a Para-lysA derivative and its functional complementation
in a lysR strain
A lysR strain fails to express lysA and hence is a Lys auxotroph (Stragier et.
al., 1983a, 1983b). Therefore to test for LysR regulation of lysP in the absence of Lys,
it was decided to construct a plasmid derivative in which lysA was ectopically
expressed from the Para promoter on pBAD18 plasmid (Guzman et al. 1995). For this,
the lysA coding region was PCR amplified from E. coli genomic DNA with primer
pairs JGJlysAF (5’-ACAAGGTACCTTTTATGATGTGGCGT-3’) and JGJlysAR (5’-
ACAATCTAGAAGTCATCATGCAACC-3’); KpnI and XbaI sites were incorporated
in the primers and are shown in italics in the sequences. The 1341-bp sized, KpnI-
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
95
XbaI digested product was cloned into the corresponding sites of pBAD18. The
plasmid was designated as pHYD2673. For verification of this construct, its ability to
alleviate auxotrophy of a lysR strain on arabinose induction was examined. Plasmid
pHYD2673 was hence transformed into lysR (GJ9652); while to serve as positive
and negative control, respectively, pBAD18 was transformed into wild-type (GJ9650)
and lysR strain. These three strains were streaked on MA ampicillin plates with
glycerol as carbon source and arabinose added at 0.2% concentration, for induction.
After overnight incubation at 37 C, it was found that the positive control and lysR
bearing pHYD2673 strain showed healthy growth while the negative control failed to
grow (since the plate was not supplemented with Lys). Thus it was concluded that,
pHYD2673 encodes a functional LysA.
4.5.1.2 Effect of LysR in lysP-lac expression
To determine if LysR has any role in lysP regulation, we transformed plasmids
pHYD2673 (Para-lysA) and pHYD2636 (lysP-lac) into the following strains: wild-type
(GJ9650), argP (GJ9651), lysR (GJ9652) and argP lysR (GJ9653). The resultant
strains were grown in MA supplemented with 0.2% each of glycerol and L-arabinose
before performing β-Galactosidase assay as described in Section 2.2.1.5 of Chapter 2.
The results are as listed in Table 4.4.
Table 4.4 Effect of LysR on lysP-lac expression a
Strain lysP-lac
wild-type 1004
lysR 1052
argP 49
argP lysR 49
a Values shown are of -Galactosidase specific activity (Miller units) after growth in glycerol-arabinose-
MA without co-effector(s).
It was noted from the results above that, as expected, lysP-lac expression was
activated in the presence of ArgP (compare values of wild-type and argP strains).
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
96
However, there was no difference in expression levels in the wild-type and lysR strains,
suggesting that LysR has no role in lysP regulation. Also, the argP or argP lysR
strains showed identical regulatory outcomes indicating that there is no cross-regulation
between LysR and ArgP at this locus.
4.5.2 Role of Lrp in lysP regulation
The transcriptional regulator Lrp has previously been implicated in regulation of
one of the ArgP targets, argO (Peeters et al., 2009). Hence, it was checked if Lrp had any
effect on lysP transcription. Accordingly, pHYD2636 (lysP-lac) was transformed into
wild-type (MC4100), argP (GJ9602), lrp (GJ9625), and argP lrp (GJ9626). -
Galactosidase assays were performed after growth of these strains in MA-glucose
medium, without or with supplementation of Arg, Lys or leucine (known to be a co-
effector of Lrp at some loci) at a concentration of 1 mM. The results are tabulated in
Table 4.5.
Table 4.5 Effect of Lrp on lysP-lac expression a
Strain Nil Leu Arg Lys
wild-type 1029 1099 985 158
lrp 468 323 456 141
argP 49 45 47 47
argP lrp 57 38 54 54
a Values shown are of -Galactosidase specific activity (Miller units) after growth in glucose-MA without
(Nil) or with supplementation of leucine (Leu), Arg (Arg) or Lys (Lys) at a concentration of 1 mM.
Results as shown in Table 4.5 demonstrated that Lrp activates lysP by about 2-
fold (compare Nil values of wild-type and lrp) and leucine has no effect on this
activation (compare Nil, Leu values of wild-type and lrp). Unlike at argO, Arg
supplementation did not enhance lysP-lac expression in a lrp strain (Peeters et al.,
[2009] reported that argO-lac expression in a lrp strain is induced much further than in
the wild-type strain on Arg supplementation. This is because Lrp competes for ArgP
binding and hence activation at argO). Furthermore, Lys repression of expression was
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
97
independent of Lrp (compare Lys values of wild-type and lrp). Also, the previously
described activation of lysP by ArgP is much more pronounced than Lrp, such that
argP and argP lrp strains show more or less the same values for β-Galactosidase
activity.
Discussion
To examine ArgP’s role as a transcriptional regulator in E. coli, a microarray-
based comparison of whole genome mRNA abundances between an argPd and argP
strain was done. Data obtained from a microarray experiment do not represent the
regulation targets under the direct control of the test transcription factor alone but instead
also include large numbers of genes, which are affected indirectly due to the changes in
the expression levels of direct target genes. Generally, therefore the direct targets
represent only a minor fraction of the differentially expressed genes detected by
microarray analysis. A bottom up approach was therefore taken in this study and genes
were short-listed from the microarray, that showed (i) at least 2-fold activation in argPd
compared to argP, strain and (ii) whose expression (like previously identified ArgP
targets) was repressed in the presence of Lys. Three genes, namely, lysP (Lys-specific
permease), lysC (Lys-sensitive aspartokinase) and asd (aspartate semialdehyde
dehydrogenase) were arrived at and considered for validation experiments. The primary
aim of the approach taken was to be able to identify as many direct targets as possible
from the vast list identified from the microarray experiment. However, it is to be noted
that this approach may have created a bias and may have allowed identification of only a
subset of genes under direct regulation of ArgP.
Further detailed experiments with lysP, showed that ArgP activates lysP by about
35-fold and that there is about 5-fold ArgP dependent repression, in the presence of Lys.
A argP mutant phenocopies a lysP mutant for resistance to the Lys analog, thialysine.
In vitro, ArgP binds the lysP regulatory region with a Kd of around 40nM and the binding
affinity is diminished upon Lys addition (to a Kd>160nM) suggesting that Lys represses
lysP in vivo by engendering the loss of ArgP binding to the lysP operator region. This is
unlike the argO locus, where ArgP binding to the cis regulatory region is not Lys
sensitive. The sequences upto 114 of the lysP regulatory region are sufficient for in vivo
regulation and in vitro binding; also sequences between 114 and 76 carry important
determinants for ArgP regulation and binding of lysP.
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
98
In addition to its role in active uptake of Lys, LysP also participates in Lys
dependent transcriptional regulation of the cadBA operon (encoding genes for Lys
decarboxylation), by sequestering the CadC activator in the absence of Lys but not in its
presence (Neely et al., 1994; Tetsch et al., 2008). We found in this study that the argP
mutant does not phenocopy a lysP mutant for cadBA expression, suggesting that the
basal level of LysP in a argP strain is sufficient for the negative regulation of cadBA in
absence of Lys. Also, the regulation of cadBA by LysP is not through regulation of its
(LysP) abundance but predominantly through the presence or absence of its interaction
with CadC. ArgP thus does not seem to have a role in the regulation of the cadBA
operon.
The argPd
alleles, based on their in vivo effect on lysP-lac, could be classified
into three categories, namely, (i) alleles -A68V, -S94L, -V144M, -P217L that were more
or less like wild-type with 20- to 30-fold activation and 3- to 5-fold Lys repression, (ii)
alleles -P108S and -R295C that showed about 30-fold activation but only 2-fold Lys
repression, and (iii) allele -P274S that showed much less activation of about 10-fold and
3-fold Lys repression. To further analyse the basis of these differences, the ArgPd
proteins -S94L, -P108S, -P274S, each protein being representative of one of the three
categories described above were cloned, overexpressed and purified. Subsequent EMSA
of the lysP regulatory region and native ArgP or ArgPd proteins did not reveal any
significant differences in binding affinities. On the other hand, differences in mobility of
the protein-DNA complexes formed by native ArgP and the ArgPd variants were seen,
such that the ArgPd-DNA complexes migrated faster than native ArgP-DNA complex, in
the order (fastest to slowest) -P274S, -P108S and -S94L. Such differences in mobility of
protein-DNA complexes have been reported to be due to differences in DNA bending
such that less bent DNA migrates faster.
Numerous transcriptional regulators have been reported to bend the DNA of their
target sequences but the effect caused by DNA bending is different depending on the
type of regulator. Extensive DNA bending studies have been done for catabolite activator
protein (CAP) also known as cAMP receptor protein (CRP) (Gartenberg and Crothers,
1988; Schultz et al., 1991), p4 of Bacillus subtilis phage 29 (Rojo and Salas, 1991;
Salas, 1998), factor for inversion stimulation (FIS) (Finkel and Johnson, 1992; Betermier
et al., 1993) and integration host factor (IHF) (Friedman, 1988; Sugimura and Crothers,
2006). These proteins generally increase DNA bending and activation is as a result of the
Microarray profiling and ArgP-mediated regulation of lysine permease Chapter 4
99
simultaneous achievement of favorable promoter architecture and a stabilized regulator-
RNAP contact to form a productive nucleoprotein transcription initiation complex
(Perez-Martin et al., 1994). However, certain LTTRs (OccR, CysB, ClcR, CbbR) that
activate only in the presence of an inducing co-effector, apparently do so through a
decrease (rather than increase) in DNA bending (Akakura and Winans, 2002; Colyer and
Kredich, 1996; Parsek et al., 1995; Dangel et al., 2005).
Contrary to these LTTR studies, ArgPd-P274S, even though showed least DNA
bending, it was also the most deficient in lysP activation. A probable explanation can
perhaps be drawn from in vivo experiments that have shown that ArgP does not require
an inducing co-effector at lysP, suggesting that lysP promoter architecture or ArgP
mechanism of activation at lysP may be different from the prototypic LTTRs described.
There are also reports which indicate that any event that bends promoter sequences in a
direction unfavorable for RNA polymerase binding can effect transcription repression
(Rojo and Salas, 1991). It may be speculated that ArgPd-P274S bending at lysP may be
unfavorable for proper RNA polymerase contacts and hence least activation is seen with
this ArgPd variant.
We also examined the effect of the regulators LysR and Lrp on lysP transcription.
LysR was found not to have any role in lysP expression; Lrp on the other hand activated
lysP by about 2-fold. However, Lrp was not a competitive activator of ArgP as at argO
(Peeters et al., 2009). In comparison with Lrp, ArgP was found to be the predominant
and determining transcriptional regulator at the lysP locus. Lrp is known to be a global
transcriptional regulator that is involved in regulation of many loci associated with amino
acid metabolism (Calvo and Matthews, 1994; Seshasayee et al., 2008). It is thus, not
completely surprising to find its involvement in lysP regulation, although this may be
only to the extent of fine-tuning of its expression.
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
100
Introduction
As described in Chapter 4, three genes namely lysP, lysC and asd were shortlisted
for validation from the microarray data, as candidate targets of the LTTR ArgP. Ensuing
work showed that lysP is indeed a transcriptional target of ArgP. The present chapter
begins with a description of experiments to test whether ArgP regulates transcription of
the other two candidate genes lysC and asd, both of which are part of the Lys
biosynthetic pathway. Since the results from these experiments indicated that ArgP does
mediate Lys repression of lysC and asd, additional experiments were then undertaken to
test whether ArgP also regulates two other genes, dapD and lysA that did not feature in
the microarray, but from literature are known to be Lys-repressed and which also
function as part of the Lys biosynthetic pathway.
As described in Chapter 1 (Section 1.1.4 and Figure 1.4), in E. coli Lys is
synthesized from aspartate and involves nine successive enzymatic reactions, eight being
common with DAP and the first two being common with methionine and threonine
(Patte, 1996). The specific biosynthetic branch leading to DAP and eventually Lys
includes enzymes (coding gene names in parentheses), aspartokinase III (lysC), aspartate
semialdehyde dehydrogenase (asd) dihydrodipicolinate synthase (dapA),
dihydrodipicolinate reductase (dapB), tetrahydrodipicolinate succinylase (dapD), N-
succinyl diaminopimelate aminotransferase (dapC or argD), N-succinyl diaminopimelate
desuccinylase (dapE), diaminopimelate epimerase (dapF) and diaminopimelate
decarboxylase (lysA). The penultimate step catalysed by diaminopimelate epimerase
(dapF) results in the formation of DAP, which is then decarboxylated by
diaminopimelate decarboxylase (lysA) to Lys.
Of all the genes of this pathway, lysC, asd, dapA, dapB, dapD and lysA are
reported to be Lys repressed. For LysC and DapA this is achieved through feedback
inhibition (Patte, 1996); additional regulation through a Lys-responsive riboswitch
mechanism has also been described for lysC (Rodionov et al., 2003; Sudarshan et al.,
2003). During the course of this study, Bouvier et al. (2008) reported that transcription of
dapB is Lys repressed through ArgP. Also, lysA expression is strictly dependent on the
regulatory protein LysR, a transcriptional activator responding to the internal
concentrations of DAP and Lys (Stragier et al., 1983a, 1983b). However, the mechanism
of Lys repression of asd and dapD have not been elucidated. Also, no common trans-
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
101
acting transcriptional regulatory factor that could effect repression of these genes in
response to increasing cellular Lys concentration has been identified.
This Chapter describes in vivo studies using argP+, argP
d and argP strains to
conclusively show ArgP regulation of lysC, asd, dapD and lysA. In vitro EMSAs show
that direct association with the upstream regulatory region of these genes, mediates ArgP
regulation. Studies investigating inter-relationships between ArgP regulation and the
previously described regulatory elements of this pathway are also discussed.
Results
Section A: ArgP regulation of lysC, asd, dapD and lysA
5.1.1 Construction of transcriptional lac fusions to cis regulatory regions of lysC,
asd, dapD and lysA
For in vivo studies of ArgP-mediated regulation of lysC, asd, dapD and lysA,
promoter-lac fusions were constructed by cloning the corresponding upstream regulatory
regions in pMU575, which is a single-copy number plasmid vector for generating
promoter-lacZ transcriptional fusions that encodes a trimethoprim resistance marker.
Four plasmids, namely, pHYD2664 (lysC-lac), pHYD2668 (asd-lac), pHYD2610 (dapD-
lac) and pHYD2670 (lysA-lac) were consequently obtained. The sequences of resultant
clones of each lac fusion were verified by sequencing. A detailed description of the
construction of each plasmid is provided below:
1) pHYD2664 (lysC-lac) : 615-bp of the lysC upstream regulatory region extending
from 244 to +371 was PCR-amplified from E. coli genomic DNA using primers
(restriction sites shown in italics), JGJlysCF (5’-ACAACTGCAGGTCTGCGTTGGATT-
3’) and JGJlysCR (5’-ACAATCTAGACGGTTCATGGCGT-3’). The PCR product was
digested with restriction enzymes, PstI and XbaI and cloned at the same sites in
pMU575.
2) pHYD2668 (asd-lac): Primers (restriction sites shown in italics) JGJasdF (5’-
GTATGTTTCAGTGTCGACATGAAAATAG-3’) and JGJasdR (5’-
GGCGTCGAAGCTTCGCTCTTCAACCA-3’) were used to amplify 366-bp of the asd
regulatory region. Digestion with SalI and HindIII resulted in a 343-bp DNA fragment,
encompassing the region from 207 to +136 that was cloned at the corresponding sites of
pMU575.
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
102
3) pHYD2610 (dapD-lac): 401-bp of the dapD regulatory region was PCR-
amplified with primers (restriction sites shown in italics) JGJdapDF (5’-
CGCCATTTTACTGCAGAAACCGAAG-3’) and JGJdapDR (5’-
CGGGTAACGGATCCTGCATTGGCT-3’). The resulting amplicon was digested with
PstI and BamHI, to obtain a 373-bp (273 to +100) DNA fragment, which was cloned at
the same sites in pMU575.
4) pHYD2670 (lysA-lac): Primers JGJlysAFnew (5’-
ACAACTGCAGCCTCAGTCAGGCTTC-3’), and JGJlysARnew (5’-
ACAAGGATCCCAGCCAAATTCAGC-3’) were used for PCR amplification of lysA
promoter region extending from 139 to +129. Restriction sites, PstI and BamHI were
incorporated in the primer sequence allowing digestion and subsequent cloning of the
same fragment at the corresponding sites of pMU575.
5.1.2 In vivo expression of lysC-lac, asd-lac, dapD-lac and lysA-lac fusions in argP+,
argPd
and argP strains
To ascertain the effect of ArgP and the ArgPd variants on the lysC-lac, asd-lac,
dapD-lac and lysA-lac, derivatives of the argP strain (GJ9602) each carrying two
plasmids namely, (i) plasmid vector pCL1920 or its derivatives with argP+ or the
different argPd variants and (ii) one of the following lac fusion plasmids, pHYD2664
(lysC-lac), pHYD2668 (asd-lac), pHYD2610 (dapD-lac) or pHYD2670 (lysA-lac) were
contructed. As control, strains with pHYD1723 (argO-lac) were also employed. The
strains were grown in glucose-MA supplemented with 18 amino acids other than Arg or
Lys, 18 amino acids with 1 mM Arg, or 18 amino acids with 1 mM Lys. β-Galactosidase
assays were then performed for each lac fusion set of strains.
The data from these experiments as documented in Table 5.1, permitted the
following interpretations:
(i) As has already been reported earlier (Nandineni and Gowrishankar, 2004; Laishram
and Gowrishankar, 2007; Peeters et al., 2009), argO-lac expression was stimulated 4-
fold in the argP+
strain relative to argP, but only in Arg-supplemented medium.
(ii) Transcription from promoters of the genes lysC, asd, dapD and lysA, in cultures of
the argP+ strains grown in Arg- or Lys-unsupplemented medium was higher than that in
the corresponding argP derivatives, by factors of approximately 4, 2.5, 4 and 2
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
103
TABLE 5.1 Expression of lac fusions in different argP variant derivatives a
_____________________________________________________________________________________________________________
argO lysC asd
------------------------------------------------ ---------------------------------------------- ------------------------------------------------
-gal sp. act. Ratio -gal sp. act. Ratio -gal sp. act. Ratio
argP ----------------------- -------------------- ------------------------ -------------------- ------------------------ --------------------
genotype Nil Arg Lys Nil/ Variant/Δ Nil Arg Lys Nil/ Variant/Δ Nil Arg Lys Nil/ Variant/Δ
Lys Lys Lys
______________________________________________________________________________________________________________
ΔargP 38 31 32 1.0 1.0 553 52 107 5.1 1.0 265 284 302 0.8 1.0
argP+ 41 106 25 4.2 3.4 2095 197 214 10 3.7 608 632 281 2.1 2.2
-A68V 2248 2613 1036 2.5 84 1477 1436 197 7.4 2.6 447 443 209 2.1 1.6
-S94L 3615 3227 3776 0.9 104 1806 1719 214 8.4 3.2 478 450 265 1.8 1.8
-P108S 1621 1633 755 2.2 53 2928 2736 250 12 5.2 871 760 283 3.0 3.2
-V144M 737 930 176 5.3 30 1264 1268 158 8.0 2.2 331 344 229 1.4 1.2
-P217L 2126 2436 623 3.9 79 1677 1610 186 9.0 3.0 352 283 157 2.2 1.3
-P274S 7884 8246 8392 1.0 266 678 646 165 4.1 1.2 195 207 186 1.0 0.7
-R295C 115 256 688 27 8.3 2399 2282 236 10 4.3 662 675 295 2.2 2.4
____________________________________________________________________________________________________________________________________________________
a Values shown are of -Galactosidase specific activity (Miller units) in glucose-MA supplemented with the 18 amino acids other than Arg or Lys (Nil), 18 amino acids
and 1 mM Arg (Arg) or 18 amino acids and 1 mM Lys (Lys). The ratios Nil/Lys and Variant/ indicate, respectively, the degree of Lys repression for the concerned
strain, and the degree of regulation imposed by the concerned ArgP variant (or the wild-type ArgP) relative to that in the argP.
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
104
TABLE 5.1 (-continued) Expression of lac fusions in different argP variant derivatives
a
_____________________________________________________________________________
dapD lysA
------------------------------------------------ ----------------------------------------------
-gal sp. act. Ratio -gal sp. act. Ratio
argP ----------------------- -------------------- ------------------------ --------------------
genotype Nil Arg Lys Nil/ Variant/Δ Nil Arg Lys Nil/ Variant/Δ
Lys Lys
_____________________________________________________________________________
ΔargP 161 152 162 0.9 1.0 493 535 80 6.1 1.0
argP+ 642 572 213 3.0 3.9 1004 980 77 13 2.0
-A68V 437 322 130 3.3 2.7 984 980 71 14 1.9
-S94L 672 567 249 2.6 4.1 910 1300 71 13 1.8
-P108S 659 757 238 2.7 4.0 1075 1038 68 16 2.1
-V144M 229 278 134 1.7 1.4 782 836 77 10 1.5
-P217L 262 216 120 2.1 1.6 841 820 64 13 1.7
-P274S 160 139 144 1.1 0.9 720 511 92 7.8 1.4
-R295C 301 230 152 1.9 1.8 893 892 69 13 1.8
_______________________________________________________________________________________________________
a Values shown are of -Galactosidase specific activity (Miller units) in glucose-MA supplemented with the 18 amino acids other than Arg or Lys (Nil), 18 amino acids
and 1 mM Arg (Arg) or 18 amino acids and 1 mM Lys (Lys). The ratios Nil/Lys and Variant/ indicate, respectively, the degree of Lys repression for the concerned
strain, and the degree of regulation imposed by the concerned ArgP variant (or the wild-type ArgP) relative to that in the argP.
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
105
respectively (see Variant/Δ ratio of argP+ for each lac fusion). For each of these lac-
fusions, similar β-Galactosidase values were obtained for strains grown in cultures
without and with Arg supplementation (compare Nil and Arg values of argP and its
variants for all these lac fusions), suggesting that Arg supplementation was without effect
for any of these genes. This implies that ArgP activates these genes and, unlike at argO,
activation is not dependent on Arg as a co-effector.
(iii) All the genes above also exhibited much higher repression upon Lys
supplementation in the argP+ than the argP strain, suggesting that Lys repression of
their promoters is ArgP-mediated (see Nil/Lys ratio of argP+ for each lac fusion).
(iv) The argPd variants used in this study had been selected to confer L-canavanine-
resistance (that is, for enhanced expression of ArgO which is the exporter of Arg and L-
canavanine) (Celis, 1999; Nandineni and Gowrishankar, 2004). As expected, all of them
activated argO-lac to a much greater extent than argP+
(see Variant/∆ ratio for argO-
lac) and this was observed both without and with Arg or Lys supplementation. The
-P274S variant in particular was the most effective for constitutive argO expression
(with 270-fold degree of activation).
(v) In the case of lysC, asd, dapD and lysA, many of the argPd variants behaved much
like the argP+ allele itself for both activation as well as Lys-repression, although a few
combinations exhibited differences. For example, the levels of activation obtained with
the -P108S variant at lysC (5-fold) and asd (4-fold) were higher than that with argP+ at
these promoters; on the other hand, both -V144M and -P217L were ineffective for
activation of asd and dapD (compare Variant/ ratios of argP+ and these argP
d alleles
for asd-lac and dapD-lac fusions). However, the most conspicuous discrepancy was that
seen with the -P274S variant; although it was the most proficient of all the ArgPd mutants
for constitutive argO expression, this variant was amongst the least effective for
activation of expression from all the other ArgP-regulated promoters, and indeed
behaved like ΔargP at three of them, namely lysC, asd and dapD (compare Variant/
ratios of argP+ and argP
d-P274S for these lac fusions).
5.1.3 In vitro binding studies of ArgP to upstream regulatory region of lysC, asd,
dapD and lysA
To determine whether the in vivo expression differences between argP+ and argP
mutant strains for the various promoter-lac fusion constructs was associated with ArgP
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
106
Figure 5.2 EMSAs with ArgP and cis regulatory regions of different genes in absence or presence of the
co-effector Lys. ArgP monomer concentrations are indicated for each lane. Bands corresponding to free
DNA and to DNA in binary complex with ArgP are marked by open and filled arrowheads, respectively.
binding in vitro to the corresponding cis regulatory regions, EMSAs were performed.
Earlier studies have shown that ArgP-mediated Lys repression can occur by two
alternative mechanisms at different promoters. At argO, Lys-liganded ArgP binds the
regulatory region (Laishram and Gowrishankar, 2007; Bouvier et al., 2008; Peeters et al.,
2009) but it then inhibits productive transcription at a step further downstream by
trapping RNAP at the promoter (Laishram and Gowrishankar, 2007). On the other hand,
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
107
at dapB (Bouvier et al., 2008), gdhA of K. aerogenes (Goss, 2008), or at the gdhA and
lysP loci previously described in this thesis, ArgP binding to the regulatory region is
diminished upon Lys addition. Accordingly in this study, EMSA with His6-ArgP were
done in the absence or presence of 0.1 mM Lys, and the cis regulatory regions of lysC,
asd, dapD, lysA, argO, dapB and lysP were employed. The data from these experiments
are shown in Figure 5.2.
The apparent Kds for binding of ArgP to the different templates in the presence or
absence of Lys were estimated as equivalent to the protein concentration at which half
maximal binding was seen (from equation 3.1 in Chapter 3) and the values are given in
Table 5.2.
TABLE 5.2 Apparent Kds (nM) of ArgP binding to cis regulatory regions of different
genes
cis regulatory region Lys + Lys
argO 15 15
dapB 120 >150
lysP 55 >140
lysC 70 110
asd 170 >200
dapD 70 120
lysA 150 210
Kd values listed in Table 5.2, show that ArgP exhibits high-affinity binding to
argO (Kd = 15 nM) which is not diminished in presence of Lys, which is consistent with
the data from earlier reports (Bouvier et al. 2008, Laishram and Gowrishankar, 2007;
Peeters et al., 2009). Under identical experimental conditions as for argO, the regulatory
regions of the other ArgP-regulated genes (dapB, lysP, lysC, asd, dapD and lysA) were
also bound by ArgP, with apparent Kds ranging from 55 nM to 170 nM (monomeric
concentrations). In all these cases (unlike the situation with argO), Lys addition was
associated with increase in the apparent Kds, indicating that ArgP binding to the DNA
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
108
regions in these instances is Lys-sensitive.
5.1.4 In vivo expression of dapB-lac in argPd
and argP strains
It has previously been shown that ArgP activates expression of dapB (coding for
dihydropicolinate reductase), by 4-fold in a Lys-sensitive manner (Bouvier et al., 2008).
To document the effect of the various argPd alleles on dapB expression, a dapB-lac
fusion was constructed. For this, the dapB regulatory region was PCR-amplified using
primers (restriction sites in italics) JGJdapBF (5’- CCCTGTTTTGCTGCAGTGGAAAC-
3’) and JGJdapBR (5’-AATGCCAGCGGATCCTGAATCAAC-3’). The resulting 411-bp
amplicon was digested with restriction enzymes PstI and BamHI to generate a 386-bp
fragment extending from 285 to +101, which was subsequently cloned in pMU575 at
the same sites. This plasmid clone was designated pHYD2669. Similar strain
backgrounds and conditions as described in Section 5.1.2 were used to set up β-
Galactosidase assays using pHYD2669. The result of these β-Galactosidase assays
demonstrating the effect of the ArgPds on dapB-lac is shown in Table 5.3.
It was observed that in media without supplementation, dapB-lac expression in
argP+ was 25-fold higher than in ∆argP indicating that ArgP activates dapB expression
by this extent. Arg supplementation did not result in further activation of dapB-lac, but
Lys supplementation caused about a 6-fold decrease in expression only in argP+
(see
Nil/Lys ratio pertaining to argP+) confirming previously reported findings of ArgP-
mediated Lys repression. It was also noted that the fold activation of dapB-lac by ArgP,
seen in the above experiments, is much higher than the 4-fold activation reported by
Bouvier et al. and may be due to differences in the medium used; while MA-glucose
supplemented with 18 amino acids other than Arg and Lys were used in the above
experiments, MA-glucose was used by Bouvier et al. (2008). The MA-glucose medium
supplemented with 18 amino acids is in principle similar to the Arg free (AF) rich
medium that has been used in studies of Arg biosynthesis and transport to obtain
derepression by the repressor protein ArgR which uses Arg as a corepressor (Novick and
Maas, 1961; Celis 1977; Celis et al., 1998). It has been reported that growth in AF
medium is twice as fast as in minimal medium and causes depletion of the limiting
amounts of intracellular Arg, thereby resulting in derepression of ArgR targets. Similarly
in the above experiments, it may be reasoned that growth in MA-glucose medium
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
109
supplemented with 18 amino acids (unlike growth in MA-glucose medium) causes
depletion of intracellular Lys and hence higher expression levels of dapB-lac.
Table 5.3 Effect of the ArgPds on dapB-lac expression
a
-gal sp. act. Ratio
ArgP
variant Nil Arg Lys Nil/ Lys Variant/Δ
argP 100 108 140 0.7 1.0
argP+ 2344 2380 401 5.8 23
-A68V 1074 1199 289 3.7 11
-S94L 1561 1468 299 5.2 16
-P108S 2891 2882 421 6.8 28
-V144M 1692 1592 305 5.5 17
-P217L 1143 1024 275 4.1 11
-P274S 190 173 127 1.4 1.9
-R295C 3717 3071 572 6.4 37
a Values shown are of -Galactosidase specific activity (Miller units) in glucose-MA supplemented with
the 18 amino acids other than Arg or Lys (Nil), 18 amino acids and 1 mM Arg (Arg) or 18 amino acids and
1 mM Lys (Lys). The ratios Nil/Lys and Variant/ indicate, respectively, the degree of Lys repression for
the concerned strain, and the degree of regulation imposed by the concerned ArgP variant (or the wild-type
ArgP) relative to that in the argP.
It was also found that ArgPd variants -P108S, -R295C were as potent in activation
of dapB-lac as the wild-type allele (see Variant/∆ ratio of the respective alleles showing
28-, 37- and 23-fold activation respectively), variants -A68V, -S94L, -V144M, -P217L
were not as effective in activation (see Variant/∆ ratio showing 11-, 16-, 17- and 11-fold
activation respectively). The most drastic difference, as previously described for lysC,
asd and dapD, was exhibited by -P274S, which failed to activate and phenocopied
argP. In case of all the argPd
variants, β-Galactosidase values obtained on Arg
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
110
supplementation were comparable to those without such supplementation (compare Nil
and Arg values for each argPd), suggesting that Arg does not have an effect on dapB-lac
expression. On the other hand, Lys addition (other than in variant -P274S), resulted in 4-
to 7-fold decrease in expression (see Nil/Lys ratios for each argPd), allowing the
conclusion that argPd variants like wild-type argP cause Lys-repression of dap-lac
expression.
Section B: Inter-relationships between ArgP regulation and the
previously described regulatory elements for different genes
5.2.1 Lysine repression mediated by riboswitch and ArgP at lysC
As previously mentioned, lysC expression is regulated at a post-transcriptional
level through a Lys-responsive riboswitch that is proposed to exist in the leader region of
the lysC mRNA (Rodionov et al., 2003; Sudarshan et al., 2003). In vivo lysC-lac fusion
studies (Table 5.1) showed that there is a 5-fold Lys-repression in the argP strain that is
ArgP-independent (see Nil/Lys ratio of argP for lysC-lac,); this presumably represents
the regulation imposed by the riboswitch, but there is additional repression to 10-fold in
an argP+ strain suggestive of an additional component of control that was ArgP
dependent.
5.2.2 LysR, ArgP regulation at lysA
lysA coding for diaminopimelate decarboxylase, the enzyme that decarboxylates
DAP into Lys, is strictly dependent on the regulatory protein LysR, a transcriptional
activator responding to the internal concentrations of DAP and Lys (Stragier et al.,
1983b). lysR is divergently transcribed from and shares a common regulatory region with
lysA. Also, LysR expression is negatively autoregulated. As both LysR and ArgP are
reported to be involved in regulation at lysA, it was of interest to find out if ArgP directly
regulated lysA or mediated its effect indirectly through regulation of lysR. Further an
examination of the existence of any crosstalk between the two regulators at this locus
was planned. For achieving these objectives, the following experiments were performed:
5.2.2.1 Construction of lysR-lac
The lysR regulatory region (268-bp in length and extending from –127 to +141)
was amplified using primers JGJlysRF (5’- ACAACTGCAGCAGCCAAATTCAGC-3’)
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
111
and JGJlysRR (5’- ACAAGGATCCCCTCAGTCAGGCTTC-3’), digested with PstI and
BamHI and cloned at the same sites in the single copy, promoter expression vector,
pMU575, to obtain pHYD2675.
5.2.2.2 Combined effect of LysR and ArgP on lysA-lac and lysR-lac expression
To check the effect of ArgP and LysR (along with co-effectors) at lysA-lac or
lysR-lac, a ΔlysR strain was purposed to be constructed. However, a ΔlysR strain fails to
express LysA and hence is a Lys auxotroph. Plasmid pHYD2673, in which lysA was
ectopically expressed from the Para promoter (details of construction having been
described in Chapter 4, Section 4.5.1.1) was thus transformed into strains: wild-type
(GJ9650), ΔlysR (GJ9652), ΔargP (GJ9651), ΔargP ΔlysR (GJ9653), already carrying
lysA-lac (pHYD2670) or lysR-lac (pHYD2675). β-Galactosidase assays were performed
after growing strains in MA-glucose supplemented with 0.2% each of glycerol and L-
arabinose along with 18 (other than Arg and Lys) amino acids; or 18 amino acids and
1 mM Arg or 18 amino acids and 1 mM Lys.
The results tabulated for lysA-lac expression as tabulated in Table 5.4 allowed the
following conclusions:
(i) In the wild-type strain, lysA expression was about 4-fold lower on Lys
supplementation than without or with Arg supplementation. This shows that lysA
expression is Lys-sensitive
(ii) Comparison of β-Galactosidase values of wild-type and ΔlysR show that lysA-lac
expression is about 10-fold lower in ΔlysR. Also, in ΔlysR, Lys addition did not result in
further decrease in lysA-lac expression. These observations substantiate previous reports
of involvement of LysR in the activation and Lys-repression at lysA.
(iii) lysA-lac expression was about 3-fold lower in ∆argP compared to wild-type in all
media tested, suggesting that ArgP activates lysA. However, this ability to activate
occurred only in a lysR+ background such that a ΔlysR and ΔargP ΔlysR showed
identical values.
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
112
Table 5.4 Cross-regulation by LysR and ArgP at lysA-laca
Strain Nil Arg Lys
wild-type 215 300 68
ΔlysR 23 26 22
ΔargP 72 75 29
ΔargP ΔlysR 24 23 23
a Values shown are of -Galactosidase specific activity (Miller units) obtained on growing strains in
glycerol arabinose-MA supplemented with the 18 amino acids other than Arg or Lys (Nil), 18 amino acids
and 1 mM Arg (Arg) or 18 amino acids and 1 mM Lys (Lys).
Table 5.5 Cross-regulation by LysR and ArgP at lysR-lac a
Strain Nil Arg Lys
wild-type 30 33 33
ΔlysR 59 65 63
ΔargP 29 33 34
ΔargP ΔlysR 66 66 65
a Values shown are of -Galactosidase specific activity (Miller units) obtained on growing strains in
glycerol arabinose-MA supplemented with the 18 amino acids other than Arg or Lys (Nil), 18 amino acids
and 1 mM Arg (Arg) or 18 amino acids and 1 mM Lys (Lys).
Likewise, data for lysR-lac expression (Table 5.5) showed that:
(i) Expression of lysR-lac is increased two-fold in ΔlysR in comparison to wild-type,
irrespective of amino acid supplementation, indicative of negative autoregulation that has
been described earlier (Stragier et al., 1983a, 1983b).
(ii) Expression levels in wild-type and ∆argP or ΔlysR and ΔargP ΔlysR were similar
indicating that ArgP does not have an effect on lysR-lac expression and that ArgP
regulation of lysA is not through regulation of lysR.
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
113
Discussion
In the results described in this chapter, the genes lysC, asd, dapD and lysA were
found to be activated by ArgP to varying extents (2- to 4-fold), this activation being
independent of the co-effector Arg but being Lys-sensitive. Several of the various argd
variants used in this study behaved much like the argP+ allele both for activation and Lys
repression at these four loci and at dapB. However, allele -P274S was odd in showing
ΔargP behaviour; this being strikingly different from its effect at argO where it activates
by 270-fold. EMSAs showed Lys-sensitive binding of ArgP to the regulatory regions of
lysC, asd, dapD and lysA suggesting that Lys repression is due to inability of the
activator to bind to the operator.
Three lines of evidence thereby suggest that the mechanism of ArgP regulation at
argO is fundamentally different from that at the loci lysC, asd, dapD, lysA and dapB.
First, argO is the only gene that requires Arg as co-effector for its activation. The second
is the differential response to the various argPd mutations (although it must be noted that
these mutations were selected on the basis of their ability to confer greatly increased
argO expression). Of these the most prominent distinction is that obtained with the
-P274S variant of ArgP; this mutant is by far the most effective for constitutive argO
expression, whereas it exhibits diametrically opposite effects and behaves much like
ΔargP for a majority of other target genes. Finally, argO is the only example in which
ArgP’s binding to the cis regulatory region is not Lys-sensitive, so that repression by Lys
is achieved by an RNA polymerase-trapping mechanism at the argO promoter. Also, it is
to be mentioned here that while ArgP is essential for argO transcription (Nandineni and
Gowrishankar, 2004; Laishram and Gowrishankar, 2007), just as LysR itself is essential
for lysA transcription (Stragier et al., 1983a, 1983b; Maddocks and Oyston, 2008), ArgP
only plays an auxiliary role in transcription of genes of the Lys biosynthetic pathway.
Of the ArgP-regulated genes, lysC and lysA exhibit additional mechanisms for
Lys-repression involving, respectively, a postulated riboswitch (Rodionov et al., 2003,
Sudarshan et al., 2003) and the LysR regulator protein (Stragier et al., 1983; Maddocks
and Oyston, 2008). Our results indicate that ArgP-mediated regulation of lysC is
independent of and additive to the putative riboswitch regulatory mechanism. On the
other hand ArgP activates lysA transcription 2-fold, but only in cells that are also
proficient for LysR. These results lend support to the notion that multi-target regulators
and multi-factor promoters represent the norm in bacterial gene regulation (Ishihama,
ArgP regulation of enzymes of the Lys biosynthetic pathway Chapter 5
114
2010).
Of the ArgP-regulated and Lys-repressed genes identified earlier and in this
study, five encode enzymes for biosynthesis of DAP and Lys: two in the pathway that is
common for Lys, threonine and methionine biosynthesis (lysC, asd), and three in the
Lys-specific pathway (dapB, dapD, lysA). One may speculate that Lys-dependent
modulation by ArgP of multiple genes of the Lys pathway provides a fine-tuning effect
in controlling the flux of intermediates serving the biosynthesis of three different amino
acids for protein synthesis as also of DAP for peptidoglycan assembly.
Our results also allow us to suggest that all Lys-liganded repression in E. coli is
mediated by ArgP. LysR also mediates Lys-repression (of lysA), but in this case it has
been suggested that Lys repression is indirect and that the co-effector ligand is actually
DAP (which is the substrate for lysA), and that the latter’s binding to LysR converts it
into an activator conformation (Maddocks and Oyston, 2008).
ArgP as a non-canonical transcriptional regulator Chapter 6
115
Introduction
As described in Chapter 1 (Sections 1.2.7 a), there are previous reports of ArgP
(IciA) binding the three AT-rich 13-mers at oriC, preventing opening of these regions
and thereby blocking the initiation of replication (Hwang and Kornberg, 1990; 1992;
Thony et al., 1991). ArgP (IciA) has also been described as a transcriptional activator of
dnaA and nrdA, which are genes involved in DNA replication and metabolism (see
Section 1.2.7 a and 1.2.7 b in Chapter 1). Lee et al. (1997) showed that IciA binds the
dnaA 1P promoter and that in vivo overproduction of the protein enhances transcription
from the same promoter. EMSA and DNase I footprinting assays also indicated that IciA
binds the AT-rich sites in the upstream region of nrdA. RNase protection assays
demonstrated that in vivo overexpression of IciA enhances transcription from the nrd
promoter, but purified IciA protein did not affect in vitro transcription from the same
promoter (Han et al., 1998).
Despite these reports of IciA regulating crucial genes in DNA replication and
metabolism, an iciA mutant or an overproducing strain is not compromised for DNA
replication or growth rate (Thony et al., 1991).
In other studies, the argP genetic locus was described independently by Celis et
al., (1973) and Rosen (1973) as mutations conferring canavanine resistance (CANr) and
it was postulated to be a regulator of Arg uptake. Although subsequent work (Nandineni
and Gowrishankar, 2004; Laishram and Gowrishankar, 2007) conclusively showed that
ArgP is a regulator of the Arg exporter, argO, it has never been determined if ArgP has
any role in transcriptional regulation of the genes involved in Arg uptake or import.
As described in Chapter 1 (Section 1.1.3), three uptake systems for Arg have been
described in E. coli: (i) the LAO (Lysine, Arg, Ornithine) system for basic amino acids
(ii) the AO (Arg, Ornithine) system and (iii) the Arg specific system. The LAO system is
encoded by argT-hisJQMP with two transcription units of the same polarity. The AO and
Arg specific system are encoded by the artPIQM- artJ locus that are organized into two
transcriptional units of the same polarity (see Figure 1.3 in Chapter 1).
At the same time, ArgP (IciA) has also been reported as a nucleoid-associated
protein that shows apparently sequence-nonspecific DNA binding activity (Azam and
Ishihama, 1999). The protein exhibits affinity for AT-rich and curved DNA sequences as
determined in vitro by DNase I footprinting or EMSAs (Wei and Bernander, 1996; Azam
and Ishihama, 1999).
ArgP as a non-canonical transcriptional regulator Chapter 6
116
This chapter describes in vivo lac fusion studies performed in argP, argP+,
argPd strain backgrounds to verify and analyse the previously reported involvement of
ArgP in transcriptional regulation of DNA replication genes (dnaA, nrdA) and Arg
uptake genes (argT, hisJ, artP, artJ). In vitro EMSAs with ArgP and regulatory regions
of dnaA, nrdA, argT and a DNA fragment from the coding region of E. coli lacZ are also
reported.
Results
Section A: In vivo experiments to examine the role of ArgP in the
regulation of dnaA, nrdA, argT, hisJ, artJ and artP
6.1.1 Construction of transcriptional lac fusions to cis regulatory regions of dnaA,
nrdA, argT, hisJ, artJ, artP
With the objective to examine the role of ArgP in the regulation of dnaA, nrdA,
argT, hisJ, artJ and artP, promoter-lac fusions were constructed. The respective
upstream regulatory regions were cloned in promoter expression vector pMU575. The
plasmids pHYD2671 (dnaA-lac), pHYD2672 (nrdA-lac), pHYD2660 (argT-lac),
pHYD2661 (hisJ-lac), pHYD2658 (artJ-lac) and pHYD2659 (artP-lac) were thereby
obtained. The sequences of resultant clones of each lac fusion were verified by
sequencing. A detailed description of the construction of each plasmid is provided below
(restriction sites incorporated in primers being indicated in italics):
1) pHYD2671 (dnaA-lac) : 668-bp of the dnaA upstream regulatory region extending
from 460 to +208 and inclusive of the previously described ArgP-regulated dnaA 1P
and 2P promoters was PCR-amplified from E. coli genomic DNA using primers
JGJdnaAF (5’-ACAACTGCAGTTTCATGGCGATT-3’) and JGJdnaAR (5’-
ACAAGGATCCGCTGGTAACTCAT-3’). The PCR product was digested with
restriction enzymes, PstI and BamHI and cloned at the same sites in pMU575.
2) pHYD2672 (nrdA-lac): Primers JGJnrdAF (5’-
ACAACTGCAGCGCCATCAACAAT-3’) and JGJnrdAR (5’-
ACAAGGATCCGTCGAGATTGATGC-3’) were used to amplify 481-bp of the ArgP-
regulated nrdA regulatory region which have beforehand been defined (Han et al., 1998).
Digestion with PstI and BamHI was followed by cloning at the same sites in pMU575.
ArgP as a non-canonical transcriptional regulator Chapter 6
117
3) pHYD2660 (argT-lac): The argT regulatory region was PCR-amplified with
primers JGJargTF (5’-ACAACTGCAGATCTCTTTGCCCGC-3’) and JGJargTR (5’-
ACAAGGATCCGTAGCGCCGCATA-3’). The resulting 367-bp amplicon was digested
with PstI and BamHI and cloned at the same sites in pMU575.
4) pHYD2661 (hisJ-lac): Primers JGJhisJF (5’-
ACAACTGCAGTGTCTACGGTGACTG-3’) and JGJhisJR (5’-
ACAAGGATCCTCGCAGCAAACG-3’) were used for PCR-amplification of hisJ
promoter region extending from 187 to +119. Restriction sites, PstI and BamHI were
incorporated in the primer sequence allowing digestion and subsequent cloning of the
same fragment at the corresponding sites of pMU575.
5) pHYD2658 (artJ-lac): The artJ-lac fusion was constructed by cloning the 416-bp
region from 270 to +146 at the PstI and BamHI sites of pMU575. The primers used for
amplification of this region were JGJartJF (5’-ACAACTGCAGCAAAGCGCTGGCA-
3’) and JGJartJR (5’-ACAAGGATCCGGATAGGTGGCTGAAA-3’).
6) pHYD2659 (artP-lac): PCR using genomic DNA as template and primers
JGJartPF (5’-ACAACTGCAGGAATCGCTAACGCC-3’) and JGJartPR (5’-
ACAAGGATCCTATCGAACAGCGCC-3’) resulted in an amplicon of 288-bp
containing the artP regulatory region from 192 to 196. Following digestion with PstI
and BamHI, this DNA fragment was cloned at the corresponding sites in pMU575.
6.1.2 In vivo expression of dnaA-lac, nrdA-lac, argT-lac, hisJ-lac, artJ-lac and artP-
lac fusions in argP+, argP
d and argP strains
The effect of ArgP and the ArgPd variants on dnaA-lac, nrdA-lac, argT-lac, hisJ-
lac, artJ-lac and artP-lac expression was evaluated by transforming plasmids
pHYD2671 (dnaA-lac) or pHYD2672 (nrdA-lac) or pHYD2660 (argT-lac) or
pHYD2661 (hisJ-lac) or pHYD2658 (artJ-lac) or pHYD2659 (artP-lac) into a argP
(GJ9602) strain bearing the argP or argPd alleles on the plasmid pCL1920 vector
backbone, or the vector alone. Plasmid pHYD1723 (previously described argO-lac) was
also transformed into this strain to serve as a positive control for argP and argPd
activation. β-Galactosidase assays were performed after growing the strains in glucose-
MA supplemented with the 18 amino acids other than Arg or Lys, along with (i) 1 mM
Arg, (ii) 1 mM Lys, or (iii) neither of the above. The outcomes of these experiments are
recorded in Table 6.1.
ArgP as a non-canonical transcriptional regulator Chapter 6
118
TABLE 6.1 Expression of lac fusions in different argP variant derivatives a
________________________________________________________________________________________________________________
dnaA nrdA argT
-------------------------------------------------- -------------------------------------------------- ----------------------------------------------
-gal sp. act. Ratio -gal sp. act. Ratio -gal sp. act. Ratio
argP ----------------------- -------------------- ---------------------- -------------------- ---------------------- --------------------
genotype Nil Arg Lys Nil/ Variant/Δ Nil Arg Lys Nil/ Variant/Δ Nil Arg Lys Nil/ Variant/
Lys Lys Lys
________________________________________________________________________________________________________________
ΔargP 88 89 95 0.9 1.0 338 348 377 0.8 1.0 208 218 204 1.0 1.0
argP+ 85 94 90 0.8 0.8 377 357 400 0.9 1.1 214 194 205 1.0 1.0
-A68V 87 87 82 1.0 0.9 419 399 391 1.0 1.2 211 197 211 1.0 1.0
-S94L 72 71 85 0.8 0.8 361 331 409 0.8 1.0 227 221 224 1.0 1.0
-P108S 78 78 91 0.8 0.8 386 398 426 0.9 1.1 196 200 194 1.0 0.9
-V144M 98 85 85 1.1 1.1 421 413 429 0.9 1.2 184 187 189 0.9 0.8
-P217L 92 90 95 0.9 1.0 340 418 377 0.9 1.0 216 215 211 1.0 1.0
-P274S 87 85 83 1.0 0.9 374 369 399 0.9 1.1 187 197 206 0.9 0.8
-R295C 93 87 90 1.0 1.0 389 417 387 1.0 1.1 191 197 196 0.9 0.9
________________________________________________________________________________________________________________________________________________________
a Values shown are of -Galactosidase specific activity (Miller units) in glucose-MA supplemented with the 18 amino acids other than Arg or Lys (Nil), 18 amino acids
and 1 mM Arg (Arg) or 18 amino acids and 1 mM Lys (Lys). The ratios Nil/Lys and Variant/ indicate, respectively, the degree of Lys repression for the concerned
strain, and the degree of regulation imposed by the concerned ArgP variant (or the wild-type ArgP) relative to that in the argP.
ArgP as a non-canonical transcriptional regulator Chapter 6
119
TABLE 6.1 (-continued) Expression of lac fusions in different argP variant derivatives a
___________________________________________________________________________________________________________________________
hisJ artJ artP
-------------------------------------------------- -------------------------------------------------- -----------------------------------------------
-gal sp. act. Ratio -gal sp. act. Ratio -gal sp. act. Ratio
argP ----------------------- -------------------- ---------------------- -------------------- ---------------------- -------------------
genotype Nil Arg Lys Nil/ Variant/Δ Nil Arg Lys Nil/ Variant/Δ Nil Arg Lys Nil/ Variant/Δ
Lys Lys Lys
________________________________________________________________________________________________________________
____
ΔargP 244 114 296 0.8 1.0 1256 231 1252 1.0 1.0 184 122 210 0.9 1.0
argP+ 227 115 262 0.9 0.9 1353 286 1480 0.9 1.0 170 105 197 0.9 0.9
-A68V 211 98 208 1.0 0.9 1419 299 1345 1.0 1.1 190 107 211 1.0 1.0
-S94L 200 91 221 0.9 0.8 1361 331 1409 1.0 1.0 207 121 211 1.0 1.1
-P108S 237 98 219 1.0 1.0 1387 328 1406 1.0 1.1 196 100 194 1.0 1.0
-V144M 228 105 209 1.0 0.9 1421 313 1429 0.9 1.1 184 107 189 1.0 1.0
-P217L 211 109 213 0.9 0.9 1340 318 1371 0.9 1.0 186 115 198 1.0 1.0
-P274S 227 95 222 1.0 0.9 1364 269 1394 0.9 1.0 197 117 206 1.0 1.0
-R295C 241 118 234 1.0 1.0 1412 317 1387 1.0 1.1 191 107 196 1.0 1.0
________________________________________________________________________________________________________________________________________________________ a Values shown are of β-Galactosidase specific activity (Miller units) in glucose-MA supplemented with the 18 amino acids other than Arg or Lys (Nil), 18 amino acids
and 1 mM Arg (Arg) or 18 amino acids and 1 mM Lys (Lys). The ratios Nil/Lys and Variant/ indicate, respectively, the degree of Lys repression for the concerned
strain, and the degree of regulation imposed by the concerned ArgP variant (or the wild-type ArgP) relative to that in the argP.
ArgP as a non-canonical transcriptional regulator Chapter 6
120
The above results indicate that each of the promoter-lac fusions were unaltered
for expression in the argP+or argP
d variants compared to that in the argP strain (see
ratio Variant/ which is 1.0 in each case). Also, for all promoter-lac fusions and in each
strain, expression under both conditions that is without or with Lys supplementation, was
nearly identical (see ratio Nil/ Lys which is 1.0 in each case). The expression of argO-
lac, under the same conditions as that employed for these promoter-lac fusions showed
ArgP-dependent activation and Lys repression as previously described in Table 5.1.
Thus, the data in Table 6.1 indicates that, ArgP does not activate transcription of DNA
replication associated genes, dnaA and nrdA in vivo nor does it regulate the genes of Arg
uptake system, contrary to previous reports. In the case of hisJ-lac, artJ-lac and artP-lac,
2-fold, 6-fold and 2-fold repression of expression in cultures grown with Arg
supplementation was observed, respectively (compare columns Nil and Arg for
respective lac fusions). Arg-dependent repression of these Arg uptake genes has been
previously reported to be ArgR-mediated (Caldara et al., 2007).
Section B: ArgP as a non-canonical transcriptional regulator
6.2.1 In vitro binding of ArgP to upstream regulatory region of dnaA and nrdA
Contrary to previous reports (Lee et al., 1996, 1997; Han et al., 1998), results in
the previous section show that neither ArgP nor the ArgPd variants activate dnaA or nrdA
in vivo. To investigate the in vitro finding from the same reports, of ArgP binding to
upstream regulatory region of dnaA and nrdA, EMSAs were performed. This involved
using radiolabelled DNA fragments encompassing the dnaA and nrdA cis regulatory
regions (same as that used for generating the lac fusion constructs) and varying
concentrations of purified His6-ArgP. The binding reactions were set up in the absence
and presence of 0.1 mM Lys.
Figure 6.1 shows that like previously reported, ArgP does bind to the upstream
regulatory region of dnaA and nrdA, with an apparent Kd of 150 nM in each case. The
binding was unchanged in presence or absence of Lys.
ArgP as a non-canonical transcriptional regulator Chapter 6
121
Figure 6.1 EMSAs with ArgP and cis regulatory regions of dnaA and nrdA in absence or presence of the
co-effector Lys. ArgP monomer concentrations are indicated for each lane. Bands corresponding to free
DNA and to DNA in binary complex with ArgP are marked by open and filled arrowheads, respectively.
Lanes 10-12 in each panel represent control EMSA reactions with lysP template.
6.2.2 In vitro binding studies of ArgP to upstream regulatory region of argT and to
the coding region of lacZ
Since ArgP showed inability to activate dnaA and nrdA in vivo but showed
proficient binding of their regulatory elements in vitro, it was pertinent to find out if in
vitro binding in the absence of in vivo regulation was also observed for the genes
involved in Arg uptake. EMSA reactions were set up using argT regulatory region (with
the same genomic co-ordinates as that used for generating the lac fusion construct) and
varying concentrations of purified His6-ArgP, in the absence and presence of 0.1 mM
Lys. It was found that the upstream regulatory region of argT revealed a very low affinity
of binding by ArgP (Kd>400nM) that was Lys-independent (Figure 6.2 A).
Based on these results that showed ArgP binding to several DNA templates in
vitro without a corresponding regulatory effect in vivo, it was next determined whether at
high concentrations it would exhibit binding to a completely non-specific control DNA
fragment. When tested by EMSA (Figure 6.2 B), very negligible binding of ArgP was
observed even at 400 nM to an internal 295-bp DNA sequence from lacZ (obtained by
PCR-amplification from genomic DNA using primer pair 5’-
GTGGTGCAACGGGCGCTGGGTCGGTTAC-3’ and 5’-
CAACTCGCCGCACATCTGAACTTCAG-3’). These findings are consistent with
ArgP as a non-canonical transcriptional regulator Chapter 6
122
Figure 6.2 EMSAs with ArgP and cis regulatory region of argT and internal lacZ DNA fragment in
absence and presence of co-effector Lys. ArgP monomer concentrations are indicated for each lane. Bands
corresponding to free DNA and to DNA in binary complex with ArgP (latter not observed for lacZ) are
marked by open and filled arrowheads, respectively. Lanes 10-12 in panel B represent control EMSA
reactions with dapD template.
other reports that ArgP does not bind DNA non-specifically (Hwang et al., 1992; Ruiz et
al., 2011).
Table 6.2 is a comprehensive tabulation of the apparent Kds of ArgP binding to
cis regulatory regions of the in vivo regulated genes namely, argO, gdhA, lysP, dapB,
lysC, asd, dapD, lysA (as previously described in Chapters 3, 4 and 5) and the in vivo
non-regulated genes namely, dnaA, nrdA, argT and coding region of lacZ, in the absence
and presence of 0.1 mM Lys.
From this data, it may be concluded that the genes exhibiting ArgP regulation in
vivo can be divided into two categories based on the Kd of binding of ArgP to their cis
regulatory region in absence and presence of Lys: (i) gene (argO) exhibiting high
affinity, Lys-insensitive binding and (ii) genes (gdhA, lysP, dapB, lysC, asd, dapD, lysA)
exhibiting moderate affinity, Lys-sensitive binding. At the cis regulatory region of genes
that did not display regulation in vivo (dnaA, nrdA and argT), ArgP effected low affinity
binding but unlike for the regulated genes of the second category, binding was Lys-
insensitive.
ArgP as a non-canonical transcriptional regulator Chapter 6
123
Table 6.2 Apparent Kds (in nM) of ArgP binding to cis regulatory regions of different
genes
cis regulatory region Lys + Lys
argO 15 15
gdhA 80 130
lysP 55 >140
dapB 120 >150
lysC 70 110
asd 170 >200
dapD 70 120
lysA 150 210
dnaA 150 150
nrdA 150 150
argT >400 >400
lacZ >400 >400
Discussion
In its description as IciA, ArgP has been reported to bind oriC in E. coli (Hwang
and Kornberg, 1990; 1992; Hwang et al., 1992) and M. tuberculosis (Kumar et al.,
2009), and to regulate transcription from promoters of the dnaA and nrdA genes (Lee et
al., 1996, 1997; Lee and Hwang 1997; Han et al., 1998).Contrary to these reports, in this
study, dnaA-lac or nrdA-lac expression was unaffected between argP+, argP
d and argP
strains, suggesting that ArgP does not regulate their transcription in vivo. These results
also allow the conclusion that this regulator protein does not play a role in transactions
involving DNA metabolism or replication in E. coli, a conclusion that is supported by the
facts that neither a ΔargP mutant nor an ArgP-overproducing strain is compromised for
DNA replication or growth rate (Thony et al., 1991). Additional confidence in the above
results showing lack of regulation was offered by use of the argPd
alleles, which showed
variable activation of the ArgP-regulated genes argO, gdhA, lysP, dapB, lysC, asd, dapD
and lysA, but consistently showed absence of activation of dnaA and nrdA.
Since ArgP is required for transcription of the Arg exporter ArgO, ArgP
ArgP as a non-canonical transcriptional regulator Chapter 6
124
dependence of expression of other genes known to be involved in Arg transport (artP,
artJ, argT, and the hisJ) was determined. None of them was affected by either the ΔargP
mutation or Lys supplementation, indicating that ArgP does not regulate Arg import in E.
coli.
On revisiting the findings of in vitro binding of ArgP at dnaA and nrdA (Lee et
al., 1996, 1997; Lee and Hwang 1997; Han et al., 1998), it was found that ArgP does
bind to the regulatory regions of dnaA and nrdA with moderate affinity and in a Lys-
insensitive manner. On checking for binding at argT regulatory region, a similar weak
affinity, Lys-insensitive binding was observed. Nonetheless, on using an internal lacZ
DNA fragment in EMSA reactions, negligible binding was noticed. With the exception
of argO (which showed Lys-insensitive, high affinity binding), a distinguishing feature
between in vivo ArgP regulated and non-regulated genes was Lys-sensitive binding in
vitro.
As described in Chapter 4, probable ArgP binding sequences were detected in the
regulatory regions of lysP and dapB (genes with highest fold activation by ArgP).
However, an attempt to determine similar sites in the regulatory regions of lysC, asd,
dapD and lysA was unsuccessful. This observation, combined with the fact that ArgP
shows binding affinity for AT-rich regions (oriC [as described in Chapter 1], upstream
regions of dnaA, nrdA and argT) which are known to have higher bending and twisting
flexibilities (Zhang et al., 2004) suggests that DNA structure may also determine binding
site recognition. High specificity of binding despite a relaxed sequence dependency has
also been reported to be a common feature for IHF and CRP in E. coli (Steffen, 2002;
Chen et al., 2001).
Although a genome wide analysis was not undertaken, the above findings suggest
that the ArgP or IciA protein is a non-canonical transcriptional regulator, in that it does
exhibit specific DNA binding to mediate the transcriptional regulation of particular genes
in vivo but in addition it also binds at other sites in the genome that are not associated
with regulatory outcomes. Genome-wide chromatin immunoprecipitation studies reveal
that DNA binding by the activator protein CRP is also apparently non-canonical since
there are about 70 sites where it binds strongly and exerts its transcriptional regulatory
function, and a further approximately 104 sites of low-affinity binding (Grainger et al.,
2005). Another example of non-canonical binding is RutR, many of whose binding sites
fall within gene coding regions (Shimada et al., 2008). LexA is described as binding to
ArgP as a non-canonical transcriptional regulator Chapter 6
125
targets containing degenerate consensus sequences (Wade et al., 2005). On the other
hand, ChIp-chip studies of transcriptional regulators such as MelR (Grainger et al.,
2004), MntR (Yamamoto etal., 2011), NsrR (Partridge et al., 2009), and PurR (Cho et
al., 2011) reveal that they behave canonically in that each exhibits binding only to those
sites where it serves to control transcription of the adjacent genes or operons. Lrp also
appears to be a canonical transcriptional regulator, although it binds nearly 140 sites in
the genome (Cho et al., 2008). CRP’s closely related paralog FNR binds canonically at
around 65 locations without any noise of low-affinity binding (Grainger et al. 2007). In
this context, transcriptional regulators in E. coli can be imagined as a spectrum with non-
canonical regulators such as CRP, RutR and ArgP at its one end and canonical regulators
such as MelR, MntR, NsrR, PurR, FNR and Lrp at the other.
The purpose of non-canonical binding by ArgP remains to be determined. It has
been speculated in the case of CRP that by bending the DNA at its non-canonical binding
sites, the protein may contribute to chromosome shaping and chromatin compaction
(Grainger et al., 2005; Wade et al., 2007). It is also possible that lack of in vivo
regulation of genes to which in vitro binding is demonstrated may be because the CRP
transcription factor functions at these loci only in specific conditions (Wade et al., 2007).
Like CRP, ArgP too may similarly participate in shaping the nucleoid architecture, given
its reported propensity to bind curved DNA (Wei and Bernander, 1996; Azam and
Ishihama, 1999). In the case of RutR it is proposed that evolution has been slow to
eliminate non-functional DNA sites because they do not have an adverse effect on cell
fitness (Shimada et al., 2008).
It is also possible that the putative dual functions of ArgP, in transcriptional
regulation and in chromosome organization, are in some way related to its ability to
dimerize in two different modes (as deduced from the crystal structure of the M.
tuberculosis ortholog [Zhou et al., 2010]). It is possible that the two dimer interfaces on
ArgP engage in sequential alternating interactions to assemble as a polymeric scaffold,
analogous to that which has been described for another nucleoid protein H-NS (Arold et
al., 2010). Ruangprasert et al., (2010) had deduced from structural studies a similar
infinite polymer arrangement for the LTTR BenM (Figure 1.15 in Chapter 1). In this
context, a ChIp-chip analysis would greatly facilitate the exact measurement of the
chromosome-wide DNA-binding profile of ArgP in vivo.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
126
Introduction
The arginine exporter argO is presently the most extensively studied, bona fide
transcriptional target of ArgP. As detailed in Chapter 1, it was previously designated as
yggA and was proposed to be an amino acid exporter primarily based on its sequence
homology to LysE, the Lys and Arg exporter in the Gram-positive bacterium C.
glutamicum (Nandineni and Gowrishankar, 2004). lysE is under the transcriptional
control of LysG, the ArgP ortholog in C. glutamicum (Bellmann et al. 2001). Thus, the
pairs ArgP-argO of E. coli and LysG-lysE of C. glutamicum are orthologous, with the
first member of each pair being a LTTR and the second its target encoding a basic amino
acid exporter.
The ArgP and LysG proteins show 35% identity and 53% similarity. Further, a
sequence alignment of the DNase I footprinted region of argO (85 to 20) with the
previously described upstream regulatory region of lysE (Bellmann et al., 2001) shows
about 44% identity. However, whereas LysE is an exporter of Arg and Lys whose
expression is induced by LysG in presence of co-effectors Arg, Lys or His, ArgO exports
Arg alone and ArgP activates expression in presence of co-effectors Arg but not Lys or
His. Based on the similarities between these orthologous transcriptional pairs and in an
attempt to address the differences in their features of regulation, transcriptional cross-
regulation studies between ArgP-argO of E. coli and LysG-lysE of C. glutamicum were
carried out.
This chapter describes experiments to examine the cross-regulation ability of
ArgP on lysE and LysG on argO. The effect of the argPd alleles on lysE-lac expression
was also determined; in vitro EMSA experiments to analyse the probable causes for the
differences in regulation exerted by wild-type argP and the argPd
alleles at argO and
lysE are described. Subsequently, the combined effects of ArgP, ArgPds and LysG on
lysE-lac expression are discussed.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
127
Results
Section A: In vivo transriptional cross-regulation studies in E. coli of
ArgP on lysE and LysG on argO
7.1.1 Construction of Para-lysG derivative and lysE-lac fusion
The lysG coding region was derived from C. glutamicum ATCC 13032 genomic
DNA as a 939-bp PCR-amplicon using primer pair (restriction sites indicated in italics)
5´-ACAAGAATTCGGTTCTTAACATGGT-3´ and 5´-
ACAAAAGCTTGCGAAGAAGTGAAA-3’. This DNA fragment was cloned at the
EcoRI and HindIII sites downstream of Para in plasmid pBAD18 (pMB9 replicon for
Ara-induced expression of target genes, ampicillin) to enable arabinose-inducible
expression of LysG in E. coli. This plasmid was designated as pHYD2676.
Similarly, a 334-bp PstI-BamHI fragment with the C. glutamicum lysE regulatory
region (from 289 bp to +45 bp with respect to start site of transcription taken as +1) was
obtained with the primer pair 5´-TAGTTTCTGCAGGCAGCAACAC-3´ and 5´-
GTCCGATGGATCCTAAAAGACTGG-3´ from C. glutamicum ATCC 13032 genomic
DNA. Plasmid pHYD2677 was derived after cloning this fragment at the PstI-BamHI
sites of plasmid pMU575 (IncW single-copy-number replicon with promoter-less lacZYA
operon, trimethoprim) and represented a lysE-lacZ transcriptional fusion.
7.1.2 In vivo β-Galactosidase assays to determine effects of C. glutamicum LysG and
E. coli ArgP on respective non-cognate targets argO and lysE
As mentioned above, ArgP has previously been shown to activate argO in
presence of Arg or citrulline and to repress it in presence of Lys (Nandineni and
Gowrishankar, 2004; Laishram and Gowrishankar, 2007; Peeters et al., 2009). On the
other hand, LysG activates lysE expression in the presence of any of the co-effectors Arg,
Lys, citrulline, or His (Bellmann et al., 2001).
To determine the effect of LysG on argO-lac expression, plasmid pHYD2676
(Para-lysG) or pBAD18 (as negative control) was transformed into a argP (GJ9651)
strain carrying pHYD1723 (argO-lac fusion). The resultant strains were grown in MA
with 0.2% each of glycerol and arabinose in absence and presence of co-effectors Arg (or
its intermediate citrulline), Lys or His or their corresponding dipeptides (with Ala); β-
Galactosidase assays were performed to determine the extent of argO-lac expression in
each case. Table 7.1 shows the results of these assays.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
128
Comparison of -Galactosidase specific activity units in strains without or with
expression of LysG (and in the presence of any co-effector) indicates that LysG is unable
to activate expression of argO-lac under any growth condition. Also, ectopic LysG
expression in an argP+
(GJ9650) strain did not interfere with the ability of ArgP to
regulate argO-lac.
Table 7.1 Absence of LysG effect on argO-lac expression a
∆argP argP+
Supplement LysG + LysG LysG + LysG
Nil 25 29 29 33
Lys 30 27 12 14
His 26 30 16 17
Arg 24 30 150 (563) 174 (655)
Citrulline 28 27 135 198
a Values shown are -Galactosidase specific activity (Miller units) in MA-glycerol arabinose without (Nil)
or with supplementation of Arg or Arg-Ala dipeptide (Arg) or Lys or Lys-Ala dipeptide (Lys), His or His-
Ala dipeptide (His), and citrulline (as free amino acid only) at 1 mM. Values obtained using free amino
acids or dipeptides were identical (within error) except in the case of argP+ strain grown in presence of
Arg-Ala, where there was a 4-fold higher induction than that obtained on use of Arg as a free amino acid;
the latter values are indicated in parentheses.
Similarly, the effect of ArgP on lysE-lac expression was ascertained by using
derivatives of argP+
(MC4100) and argP (GJ9602) strains carrying pHYD2677 (lysE-
lac fusion) in β-Galactosidase assays. The strains were grown in MA-glucose
supplemented either with Arg, Lys, His or with their corresponding dipeptides (with
Ala). Table 7.2 shows the results of these assays.
It was observed that the lysE-lac fusion directed a low level of -Galactosidase
expression which was not different between ΔargP and argP +strains in all media tested,
suggesting that native ArgP also cannot cross-activate lysE.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
129
Table 7.2 Absence of ArgP effect on lysE-lac expression a
Supplement argP argP+
Nil 25 25
Lys 24 27
His 28 25
Arg 28 27
a Values shown are -Galactosidase specific activity (Miller units) in MA-glucose without (Nil) or with
supplementation of Arg or Arg-Ala dipeptide (Arg) or Lys or Lys-Ala dipeptide (Lys), His or His-Ala
dipeptide (His) at 1 mM. Values obtained using free amino acids or dipeptides were identical (within
error).
7.1.3 Reconstitution of C. glutamicum LysG-lysE regulation in E. coli
A possible reason for the lack of activation of argO by LysG could be the
inability of the Gram-positive C. glutamicum transcriptional regulator to recruit Gram-
negative E. coli RNAP. To examine this possibility, reconstitution of C. glutamicum
LysG-lysE regulation in E. coli was attempted. As previously described, in C.
glutamicum, LysG activates lysE expression by about twenty-fold in the presence of co-
effectors Arg, Lys or His (Bellmann et al., 2001).
Table 7.3 shows that in a ΔargP strain with Ara-induced LysG, lysE-lac
expression was significantly up-regulated (4- to 5- fold) in presence of Arg, Lys or His,
in comparison with the level of expression observed in absence of any co-effector. For
Lys and His, activation was observed upon their supplementation either as free amino
acids or as dipeptides (with Ala). For Arg however, it was only the dipeptide and not the
free amino acid that activated lysE expression.
Although the possibility exists that it is the Arg-containing dipeptide which is the
co-effector for LysG-mediated activation of lysE, it is more likely for the following
reasons that Arg is indeed the co-effector and that uptake of exogenously provided Arg is
not as efficient as that of the dipeptide in generating the intracellular concentrations of
amino acid needed for lysE induction: (i) it is known that the E. coli Arg uptake systems
are feedback repressed in presence of Arg (Caldara et al., 2007); (ii) since the data
indicate that the Lys-Ala and His-Ala effects of lysE are mediated by the free amino
acids Lys and His, respectively, it is reasonable to assume that the Arg-Ala effect is also
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
130
mediated by Arg; (iii) LysG’s ortholog ArgP possesses a single binding pocket for
competitive binding of the co-effectors Arg and Lys ( Laishram and Gowrishankar, 2007;
Zhou et al., 2010); and (iv) ArgP-mediated argO-lac induction was also determined to be
4-fold higher with Arg-Ala than with Arg (data shown in Table 7.1).
Table 7.3 lysE-lac activation by LysG in E. coli a
Supplement Free amino acid Dipeptide (with Ala)
Nil 22 18
Lys 110 95
His 95 89
Arg 22 111
Citrulline 21 n.d.
a Values shown are -galactosidase specific activity (Miller units). A derivative of strain ΔargP (GJ9651)
with plasmids pHYD2676 and pHYD2677 bearing Para-lysG and lysE-lac, respectively, were grown for -
galactosidase assays in MA medium with 0.2% each of glycerol and Ara without (Nil) or with addition at 1
mM of Lys, His, Arg or citrulline either as free amino acid or as dipeptide with Ala (for all but citrulline;
that is, Lys-Ala, His-Ala or Arg-Ala). n.d. implies not determined.
Thus, the results recapitulate in E. coli, at least qualitatively, the LysG regulation
of lysE as described in C. glutamicum.
7.2 Section B: Transcriptional cross-regulation studies of ArgPd
variants on lysE
7.2.1 In vivo -Galactosidase assays to show the effect of ArgPd
variants on lysE-lac
expression
As mentioned in the previous chapters, several gain-of-function single amino acid
substitutions in ArgP are known (Nandineni and Gowrishankar, 2004) that mediate high
and constitutive expression of argO in E. coli (values for argO-lac expression effected
by these argP variants are shown in chapter 5, Table 5.1). These are designated as the
argPds and seven such ArgP
d variants were tested for their ability to cross-regulate lysE.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
131
The argP strain derivatives harboring two plasmids namely pHYD2677 (lysE-
lac) and any one of the seven argPd variants cloned in vector pCL1920 (pSC101
replicon, streptomycin and spectinomycin), were employed in these experiments. -
Galactosidase assays were performed after growth in MA-glucose with or without
supplementation of amino acids Arg and Lys.
Table 7.4 demonstrates that while four of the ArgP variants namely -A68V,
-V144M, -P217L, and -R295C were unable to activate lysE-lac in E. coli, the remaining
three namely -S94L, -P108S, and -P274S were able to do so to different extents. ArgPds
-P274S and -S94L were the most effective with a 5- to 6-fold activation of lysE
expression, which in both cases was independent of co-effector addition and comparable
to the maximal activation obtained with LysG itself upon co-effector supplementation
(see Table 7.3). ArgPd-P108S activated mildly by 2-fold and this was Lys-sensitive.
Table 7.4 lysE-lac activation by certain ArgPd
variants a
argP genotype Nil Arg Lys
∆argP 15 15 14
argP+ 15 19 18
-A68V 25 24 19
-S94L 84 85 67
-P108S 35 36 18
-V144M 15 16 14
-P217L 16 19 14
-P274S 126 138 118
-R295C 11 12 14
a Values shown are -galactosidase specific activity (Miller units) obtained after growth of strains in MA-
glucose without (Nil) or with supplementation of amino acids Arg and Lys at 1 mM.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
132
7.2.2 In vitro binding studies of the ArgPd
proteins -S94L, -P274S and -P108S with
the argO and lysE regulatory regions
To determine if the increased activation at argO and lysE by the ArgPd variants
-S94L, -P274S and -P108S (relative to that by native ArgP) was due to alterations in
either affinity or pattern of binding of the proteins to the corresponding regulatory
regions, EMSAs were first performed with the purified (His6-tagged) proteins and
upstream regulatory regions of argO (427-bp DNA fragment from 293 to +109 bp) or
lysE (334-bp DNA fragment from 289 to +45 bp). Attempts to use purified LysG as
well in these experiments were unsuccessful, since the protein could not be recovered in
active form from insoluble inclusion bodies following its overexpression.
Figure 7.1 EMSAs with native ArgP (WT) or its variants -S94L, -P274S and -P108S at the indicated
protein concentrations and cis regulatory region of argO (427-bp fragment) in the absence of co-effector
(Nil) or in the presence of Arg or Lys at 0.1 mM concentration. Bands corresponding to free DNA and to
DNA in binary complex with each of the ArgP proteins are marked by open and filled arrowheads,
respectively.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
133
Figure 7.1 A shows that at argO, the Kds of binding of ArgP and its variants
-S94L, -P274S and -P108S were not significantly different from one another (around 10
nM in all cases); however, significant differences were observed in migration rates of the
DNA-protein complexes, in decreasing order of -P274S, -P108S, -S94L and native ArgP.
Neither the Kd of binding nor migration rate of the DNA-protein complexes at argO was
affected by Arg or Lys addition (Figure 7.1 B and 7.1 C).
Figure 7.2 EMSAs with native ArgP (WT) or its variants -S94L, -P274S and -P108S at the indicated
protein concentrations and cis regulatory region of lysE (334-bp fragment) in the absence of co-effector
(Nil) and in the presence of Arg or Lys at 0.1 mM concentration. Bands corresponding to free DNA and to
DNA in binary complex with each of the ArgP proteins are marked by open and filled arrowheads,
respectively.
At lysE by contrast (Figure 7.2 A), native ArgP and the ArgPd
variants showed
significant differences in their binding affinities; whereas native ArgP displayed
negligible binding even at 300 nM, the -S94L, -P274S and -P108S variants exhibited Kds
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
134
of ~300 nM, 60 nM, and >300 nM, respectively. As with argO, the complexes of lysE
DNA with ArgPd variants also migrated faster than the complex of lysE with wild-type
ArgP. The EMSA features above were unaltered by Arg or Lys supplementation (Figure
7.2 B, Figure 7.2 C respectively).
In the case of other LTTRs, differences in degree of DNA bending upon protein
binding (Akakura and Winans, 2002; Colyer and Kriedrich, 1996; Dangel et al., 2005;
Kreusch et al., 1995; Kullik et al, 1995, McFall et al., 1997; Parsek et al., 1995; Wang et
al., 1992) or in protein pIs (Leveau et al., 1996) have earlier been invoked to explain
differences in migration rates of the protein-DNA complexes in EMSA experiments.
Since native ArgP is a homodimer (Azam and Ishihama, 1999; Laishram and
Gowrishankar, 2007; Zhou et al., 2010), a third possibility that the ArgPd proteins may
function as monomers to therefore exhibit faster migration after binding to DNA could
also be considered. These different possibilities were tested in the experiments described
in the following sections.
7.2.3 pI measurements of ArgP and the ArgPd proteins
It was reasoned that the difference in mobilty of the DNA-protein complexes of
native ArgP and ArgPd variants could be due to altered (lower) pI compared to the wild-
type protein (predicted pI being 6.8). Hence, in a native EMSA gel of pH 8.3, the ArgPd
variants would have more negative charge and show enhanced mobility.
To test this possibility, native isoelectric focusing was done (Figure 7.3) using
Pharmacia Phast Gel Apparatus and precast IEF gel (pH 3-9) from GE healthcare. The
results indicated that the pI of the ArgPd proteins (~6.6) was not lower than that of native
ArgP itself (~6.3), thereby excluding pI changes as the explanation for the observed
EMSA mobility differences.
7.2.4 Gel filtration chromatography to determine the oligomeric state of the ArgPd
proteins -S94L, -P274S and -P108S
To determine the oligomeric state of the ArgPd proteins -S94L, -P274S and -
P108S, the purified proteins were passed through Sephadex G100 gel filtration column,
along with native ArgP as control. Each of the four proteins -S94L, -P274S and -P108S
eluted as a single peak at similar elution volumes (Figure 7.4 A). Molecular weights
corresponding to the peaks, as estimated from the calibration curve derived from running
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
135
Figure 7.3 Native isoelectric focusing of ArgP and its variants -S94L, -P274S -P108S. Lane at right
depicts protein markers of pIs as indicated alongside. Bands corresponding to native ArgP and the variants
are indicated by arrows.
Figure 7.4 Elution profiles of wild-type ArgP, -S94L,-P274S and -P108S on gel filtration
chromatography through sephadex G-100 (A). Subunit composition was determined from a plot of log10
molecular masses in kilodaltons (MW) of protein standards (solid squares denoting in descending order,
bovine serum albumin, ovalbumin, chymotrypsinogen A and ribonuclease A) against the fraction number
representing their peaks of elution; shown by the interrupted lines are the fraction number representing the
elution peak for each of the four proteins (WT, -S94L,-P274S and -P108S) and the corresponding log10MW
value as intercepts of the X- and Y- axes, respectively (B).
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
136
molecular mass markers approximated for each protein, to a dimeric state (Figure 7.4 B).
Thus, the ArgPd proteins, like native ArgP, exist as homodimers in solution and an
altered oligomeric state of the ArgPd proteins is not an explanation for their altered
electrophoretic mobility.
7.2.5 DNA bending studies with native ArgP and ArgPd proteins -S94L, -P274S,
-P108S at argO
The degree of electrophoretic mobility retardation of a bent DNA fragment
depends on position of the bend with respect to fragment ends, so that retardation is most
pronounced when the bend is located near the centre of the fragment (Wu and Crothers,
1984; Wang et al., 1992). Accordingly, migration rates of argO DNA complexes were
determined using (i) 100 nM each of purified ArgP or ArgPd proteins and (ii) three PCR-
generated, radiolabelled DNA fragments of nearly identical size but in which the 85 to
20 argO segment (which is the region footprinted by ArgP) was located near the
downstream end (D), the middle (M), or the upstream end (U) of the fragment. The
fragments (length and extent with respect to argO transcription start site, indicated in
parentheses): D (335-bp, 305 to +30), M (333-bp, from 212 to +121) and U (332-bp,
from 105 to +227) were generated using primer pairs 5’-
GTGCGCCTGAACGAACTTGGTG-3’ and 5’-CACGTTGGATATTCCGAATT-3’, 5’-
CTGGAGCGTATTAAACGTGA-3’ and 5’-GTATGCCCTGATTCATCACAAAAG-3’,
5’-CGCTGAGGCCAGATAATACT-3’ and 5’-CACGGCGACTGCATCAATAA-3’,
respectively. Although the fragment sequences were not permuted with respect to one
another, it was reasoned that the different neighbouring sequences in the three fragments
would not affect their mobility under the experimental conditions.
Figure 7.5 shows the results of these EMSAs which indicated the following: (i) In
the absence of any co-effector (Figure 7.5 A), the migration for the standard argO
fragment (with ArgP binding site at middle) in complexes with the different ArgP
proteins were reproduced, so that the complexes with native ArgP and ArgPd-P274S
migrated the slowest and fastest, respectively (compare encircled bands in the different
lanes labelled M). (ii) These migration differences between the different ArgP proteins
were largely abolished for argO fragments in which the protein binding sites were
located near the downstream or upstream ends, so that all complexes now migrated at the
faster rate (see lanes labelled D and U, respectively). These data therefore strongly
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
137
suggest that argO DNA is bent upon native ArgP binding, and that bending is
considerably less in case of binding by the three ArgPd variant proteins.
Figure 7.5 B and Figure 7.5 C shows similar binding assays performed in the
presence of co-effector Arg and Lys at 0.1 mM concentration. These figures reveal that
(i) unlike for other reported LTTRs, addition of the positive co-effector Arg did not cause
decreased DNA bending by native ArgP (compare lanes M in Figure 7.5 A and Figure
7.5 B for WT); (ii) neither did addition of the negative co-effector Lys result in increased
DNA bending by native ArgP (compare lanes M in Figure 7.5 A and Figure 7.5 C for
WT); and (iii) among the ArgPd-DNA complexes, the migration profiles were more or
less similar to that in the condition without co-effector. However, one significant finding
was that there was a speeding up of the argO complex with the -S94L variant (compare
lanes M in Figure 7.5 A and Figure 7.5 B for -S94L) but not that with native
Figure 7.5 EMSAs with native ArgP (WT) or its variants -S94L, -P274S and -P108S at 100 nM and the
argO templates D, M and U (with LTTR-binding site at the downstream end, middle and upstream end of
the fragments, respectively and as illustrated adjoining panel A). The experiments were done in absence
(Nil) or presence of co-effectors Arg or Lys at 0.1 mM concentration. Open arrowhead in each panel
denotes the unbound DNA probe, and bands corresponding to binary complexes of protein with template M
are circled.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
138
ArgP in presence of Arg, suggesting that the bend induced only by the former is reversed
upon Arg addition (at least at the concentration used in the experiments).
7.2.6 DNA bending studies with native ArgP and ArgPd proteins -S94L, -P274S,
-P108S at lysE
Figure 7.6 EMSAs with native ArgP (WT) or its variants -S94L, -P274S and -P108S at 400 nM and the
lysE templates D, M and U (with LTTR-binding site at the downstream end, middle and upstream end of
the fragments, respectively). The experiments were done in absence (Nil) or presence of co-effectors Arg
or Lys at 0.1 mM concentration. Open arrowhead in each panel denotes the unbound DNA probe, and
bands corresponding to binary complexes of protein with template M are circled.
Similar binding experiments were undertaken with respect to lysE using 400 nM
of purified ArgP and ArgPd
proteins; the three fragments (extent with respect to lysE
transcription start site, indicated in parentheses) D (268 to +21), M (188 to +101) and
U (88 to +201) were each 289-bp in length. They were generated by PCR amplication
from C. glutamicum ATCC 13032 genomic DNA using the respective primer pairs: 5’-
CTGCTTGCACAAGGACTTCACC-3’ and 5’-ACCTGTAATGAAGATTTCCAT-3’,
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
139
5’-TCGAGAGCTTTAACGCGCTGAC-3’ and 5’-CCTTCGCGCTTAATTCCTTGTT-
3’, 5’-CCAGTTGAATGGGGTTCATGA-3’ and 5’-
CACGATCGGCGCGGCATTGGAC-3’.
Figure 7.6 demonstrates that except under the condition of Arg addition, native
ArgP binding to lysE is associated with more pronounced DNA bending than is binding
of either of the ArgPd variants (compare lanes M of WT and of the ArgP
d variants). Arg
addition led to faster mobility of the complex of wild-type ArgP with the M DNA
fragment (compared to that in the absence of Arg), so that the M fragment complexes
with ArgP and the different ArgPd
proteins now exhibited similar mobilities.
7.3 Section C: Combined effect of LysG and ArgPd variants on lysE
expression
Since the typical LTTR is dimeric in solution and assembles on DNA as a dimer
of dimers or as higher-order oligomers to activate transcription (Maddocks and Oyston
2008, Momany and Neidle, 2012), we tested the effects of the combined presence of
LysG and ArgP or its variants on lysE regulation in E. coli.
For this, -Galactosidase assays were performed in argP (GJ9651) strain
derivatives carrying three plasmids namely, (i) pHYD2677 (lysE-lac fusion), (ii)
pHYD2676 (Para-lysG) or pBAD18 vector and (iii) argP+ or argP
d derivatives on
pCL1920 vector. The strains were grown in MA-glycerol arabinose and co-effectors Arg,
Lys and His were added at 1 mM concentration as free amino acids or as dipeptides.
Table 7.5 shows the results of these experiments.
It was found that activation of lysE transcription in presence of LysG and its co-
effectors Arg, Lys and His (added as their respective dipeptides with Ala) was unaffected
by native ArgP (that is, in the argP+
strain in comparison with that in argP).
Interestingly, however, strains in which LysG was co-expressed with any of the lysE
activation-proficient ArgPd
variants (-S94L, -P274S, or -P108S) displayed significantly
lower lysE expression than those with only one of the LTTRs present in certain growth
media (compare values of -S94L, -P274S, or -P108S in LysG and + LysG sections),
indicative of a mutual antagonism or reciprocal dominant negative effect. For example,
such mutual interference was observed for LysG with either ArgPd-S94L or ArgP
d-
P274S in cultures supplemented with either Arg-Ala or His-Ala, and LysG was also
dominant-negative over all three ArgPd variants in cultures without co-effector
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
140
Table 7.5 Antagonism between ArgPd variants and LysG for lysE-lac activation
a
LysG + LysG
argP
genotype
Nil Arg Arg-
Ala
Lys Lys-
Ala
His His-
Ala
Nil Arg Arg-
Ala
Lys Lys-
Ala
His His-
Ala
∆argP 25 27 27 24 24 29 24 21 24 155 100 115 96 116
argP+ 25 28 22 27 23 28 21 19 25 118 106 131 108 111
-S94L 95 111 97 106 109 120 93 21 21 19 91 110 55 56
-P274S 181 162 185 188 175 174 179 23 23 20 96 103 45 43
-P108S 45 51 40 31 20 46 39 19 23 106 114 104 111 105
a Values shown are -galactosidase specific activity (Miller units) obtained after growth of strains in MA-glycerol arabinose without (Nil) or with supplementation of
amino acids Arg, Lys, His or the corresponding dipeptides (with Ala) at 1 mM. Values indicative of dominant-negative effect of LysG are in bold, and those indicative
of mutual dominant negativity of LysG and ArgPd in bold italics.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
141
supplementation. On the other hand, in cultures supplemented with Lys-Ala, lysE
expression in the combined presence of LysG and ArgPd
was similar to that with either
LTTR alone, and no antagonistic regulation was observed.
When free amino acids were used, His behaved like His-Ala in provoking the
mutual antagonistic effect, and likewise Lys behaved like Lys-Ala in activating lysE-lac
in presence of both LTTRs. With Arg supplementation, the results were indistinguishable
from those obtained in the unsupplemented cultures, which may be consistent with the
data described in Section 7.1.2 of this chapter and as shown in Table 7.3 that it is only the
Arg-containing dipeptide (and not Arg as free amino acid) which is proficient for
activating LysG in E. coli.
Discussion
The main results described in this chapter are the following: (i) C. glutamicum
LysG can activate its cognate target lysE in the heterologous milieu of E. coli; (ii) some
ArgPd variants of E. coli can activate C. glutamicum lysE whereas native LysG and ArgP
do not cross-regulate their non-cognate targets (argO and lysE, respectively) in vivo; and
(iii) binding of the ArgPd variant proteins to the regulatory regions of argO or lysE is
associated with a reduced angle of DNA bending compared to that obtained with native
ArgP. Each of these is further discussed below.
a) Reconstitution of C. glutamicum LysG-lysE regulation in E. coli
In the classical model for gene activation (Ptashne and Gann, 1997),
transcription factor binding to a target gene regulatory region is followed by RNAP
recruitment to enable productive transcription. The result showing the ability of C.
glutamicum LysG to activate its cognate target lysE in E. coli in presence of Lys, His or
Arg is remarkable for two reasons. Firstly, it demonstrates that a Gram-positive
transcription factor is successful in recruiting the Gram-negative RNA polymerase
enzyme to make appropriate promoter contacts for transcriptional activation. There have
been no earlier reports of in vivo transcriptional activation across the divide between
Gram-positive and Gram-negative bacteria which some estimates indicate were
established about two billion years ago (Feng et al., 1997; Pace 1997). Since
transcriptional activation requires high-fidelity interactions across macromolecular
interfaces, our results imply that such contacts (between LysG and RNAP) have been
conserved across a vast evolutionary distance. The LTTRs CatR and OccR from,
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
142
respectively, Pseudomonas putida and Agrobacterium tumefaciens have earlier been
shown to activate transcription with E. coli RNAP in vitro, that is, within the Gram-
negative kingdom itself (McFall et al., 1997; Wang et al., 1992). In this context, the
alternative explanation that ArgP and LysG are not orthologs that have diverged from a
single gene in the ancient common ancestor, but instead represent an example of
horizontal gene transfer (HGT) that occurred more recently in evolution was explored.
One way to distinguish between these possibilities is to compare an organism’s overall
GC content with that of the gene in question, since it is the discordance between the two
that is the often taken as a signature of HGT (Gogarten et al., 2002) Data presented in
Table 7.6 shows that the GC content of the gene encoding the closest relative of E. coli
ArgP in each of seven different bacteria more or less matches that for the entire genome
of the corresponding bacterium, suggesting that the genes are indeed orthologs rather
than outcomes of HGT.
Table 7.6 Percent GC content of genomes and argP orthologs of different bacteriaa
Organism Ortholog IDa %GC for gene %GC for genome
E. coli NP_417391.1 54.4 50.8
C. glutamicum EGV41297.1 55.2 53.8
M. tuberculosis NP_216501.1 65.0 65.6
Streptomyces
coelicolor
NP_631362.1 75.8 72.0
Klebsiella
pneumoniae
YP_002236622.1 59.7 56.9
Haemophilus
influenzae
ZP_01783604.1 33.4 38.0
Pseudomonas
aeruginosa
YP_001350270.1 71.5 64.4
a The ortholog ID is as listed in the NCBI protein database.
A few other instances have been reported of conserved interactions between
Gram-positive and Gram-negative bacteria, that is manifested as successful cross-
complementation of genes whose products (proteins or RNA) function within
macromolecular complexes; the examples include F0-F1 ATPase between Bacillus
megaterium and E. coli (Scarpetta et al., 1991), protein complexes involving YidC
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
143
between Streptococcus mutans and E. coli (Dong et al., 2008) or Era/Bex between
Bacillus subtilis and E. coli (Minkovsky et al., 2002), and the protein-RNA components
of RNase P between B. subtilis and E. coli (Wegscheid et al., 2006). In vitro
reconstitution of the 30S ribosome complex with protein subunits and rRNA inter-mixed
between E. coli and Bacillus stearothermophilus has also been described (Higo et al.,
1973; Nomura et al., 1968). However, several proteins of the E. coli and B. subtilis
divisome complexes do not productively interact with one another although they are
orthologous (Robichon et al., 2008), nor is the SecA homolog from either B. subtilis
(McNicholas et al., 1995) or Streptomyces lividans (Blanco et al., 1996) able to
participate in protein translocation in E. coli.
Secondly, the result showing the ability of C. glutamicum LysG to activate its
cognate target lysE in E. coli indicates that E. coli RNAP recognizes the C. glutamicum
lysE promoter. A previous study had found that Corynebacterial promoters of varying
similarity to the consensus E. coli promoter sequences and all recognized by the sigA
gene encoded principal sigma factor, are correctly recognized in E. coli (Patek et al.,
2003). The canonical 35 and 10 hexamers found in E. coli and Bacillus sp. are
TTGACA and TATAAT with a 14 to 17 bp intervening region. In the consensus C.
glutamicum promoters, the prominent feature is a conserved extended 10 region
tgngnTA(c/t)aaTgg, while the 35 region is much less conserved (Patek et al., 2003).
Analysis of the upstream regulatory region of lysE revealed CACGAT as a motif
analogous by position and sequence to 10, also 17 bp upstream to this sequence is a
putative 35 bearing the sequence TTTACT. While this 10 is 50% identical, the 35 is
67% identical to the E. coli consensus. Thus considering that the lysE promoter
sequences bear significant identity with those in E. coli and is also recognized by the
principal sigma factor (Patek et al., 2003), we had assumed this promoter may be
functional in E. coli.
b) Cross-regulation between ArgP-argO and LysG-lysE
Several examples of successful cross-regulation between different transcription
factors have been described earlier, including those between BenR-XylS of the AraC
family (Cowles et al., 2000; Jeffrey et al., 1992) XylR-DmpR of the NtrC family
(Fernandez et al., 1994); and CatR-ClcR and TfdR-TfdT-TcbR-ClcR of the LTTR family
of proteins (Parsek et al., 1994; Parsek et al., 1995; Leveau and van der Meer, 1996). In
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
144
all these cases, the target genes are involved in catabolism of aromatic compounds in
Pseudomonas species or related Gram-negative bacteria and are most often plasmid-
borne; hence it is unclear whether the different regulator proteins are paralogs or merely
orthologs that have been re-united through horizontal gene transfer (Gogarten et al.,
2002). Amongst other LTTRs, limited or no cross-regulation is observed between NocR
and OccR involved in catabolism of nopaline and octopine respectively in A. tumefaciens
(Kreusch et al., 1995); however, overexpressed GcvA (that regulates glycine cleavage) of
E. coli is able apparently to partially substitute for Citrobacter freundii AmpR in
regulation of the ampC gene encoding β-lactamase (Everett et al., 1995).
In this study, no cross-regulation between the native ArgP and LysG proteins for
their respective non-cognate targets lysE and argO was observed. For the LysG-argO
combination, the failure to activate is not because of an inability of LysG to recruit E.
coli RNAP, given that LysG activates lysE in E. coli. Hence it can be suggested that this
failure reflects a decreased binding affinity of LysG to the argO regulatory region.
Likewise, ArgP’s inability to cross-regulate lysE is at least in part explained by the high
Kd of protein binding to lysE, since the two ArgPd variants that were most effective for
activation (-P274S and -S94L) also exhibited improved binding affinities. The fact that
neither native protein exerts a dominant negative effect on the other’s activating ability in
E. coli (that is, LysG on lysE and ArgP on argO) would also suggest that inability to
cross-regulate is likely due to a defect in an early rather than late step in the process.
The ArgPd single amino acid substitution variants used here had originally been
selected to confer elevated argO expression (Celis 1999; Nandineni and Gowrishakar,
2004); the finding that some of them have also become proficient for cross-regulation of
lysE in vivo serves to strengthen the notion that ArgP-argO and LysG-lysE are indeed
orthologous transcription factor-target pairs which are both functionally and
evolutionarily related. The two ArgPd derivatives that were most effective for lysE
activation (-P274S, -S94L) are also those that conferred the highest constitutive levels of
argO expression, but whether this is just a coincidence remains to be determined.
Furthermore, an important distinction between native ArgP and LysG is that the former
fails to activate its target in presence of Lys, suggesting that the -P274S and -S94L ArgPd
variants are more akin to LysG in this regard.
The crystal structure of M. tuberculosis ArgP has been determined (Zhou et al.,
2010) and may be described with the structure of the LTTR CbnR of Ralstonia eutropha
(Muraoka et al., 2003) as being comprised of an N-terminal DNA binding domain (DBD)
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
145
Figure 7.7 Amino acid sequence alignment of CbnR of R. eutropha, ArgP of E. coli and LTTRs: AmpR
of C. freundi, NodD of R. leguminosarum, OxyR of E. coli and AphB of V. cholerae. The colored lines
below the sequence alignment table represent the domains in CbnR; the coloring scheme is as follows: red,
yellow, blue and green represent DNA-binding domain, the α4 linker helix, and RD-I and RD-II,
respectively. The hinge regions (hinges 1,2 and 3) are indicated below the colored line. Hinge 3 is
composed of two segments, hinge 3(a) and hinge 3(b). Numbering of residues is as those of CbnR. The
residues corresponding to the position of mutation in ArgP variants described in this study are shown in
red. Mutations described in the text with respect to AmpR, NodD, OxyR and AphB are indicated in blue in
the respective sequences.
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
146
followed sequentially by a linker helix and the C-terminal regulatory domain (RD)-I and
-II (with hinge regions 1, 2 and 3 located between (i) DBD and linker helix, (ii) linker
helix and RD-I, and (iii) RD-I and RD-II, respectively). Upon mapping of the seven
ArgPd substitutions (that render E. coli constitutive for argO expression) it was found
that none is in the DBD, A68V maps in the linker helix which is proposed to be involved
in establishing intra-dimer contacts, S94L, P108S, V144M, P274S, R295C map to RD-I
and P217L was in RD-II. Specifically, the three mutants shown to activate lysE localized
next to hinge 2 (ArgPd-S94L and ArgP
d-P108S) and hinge 3 (ArgP
d-P274S). Hinge 2 is
responsible for the conformational change between the compact and the extended forms
of the monomer, while hinge 3 is the crossover region between RD-I and RD-II that is
postulated to accommodate the regulatory ligand and make associated conformational
changes. Mutation in residues around hinge 2 with LTTRs NodD (residue 95), AmpR
(residue 102), AphB (residue 100) and OxyR (residue 100) also result in co-effector
independent phenotype in each case (Burn et al., 1989; McIver et al., 1989; Kullik et al.,
1995; Bartowsky and Normark, 1991; Taylor et al., 2012) (Figure 7.7). Thus these hinge
regions appear to be crucial for conferring the flexibility needed for a single LTTR to
recognize both the argO and lysE transcriptional targets.
c) Mutual dominant negativity between LysG and ArgPd variants for lysE
regulation
The data from experiments examining the effects of combined presence of LysG
and ArgPd variants on lysE expression in vivo indicate that both LysG on the one hand,
and ArgPd-S94L or -P274S on the other, mutually antagonize each other’s ability to
activate lysE in the Arg- or His-supplemented cultures. Furthermore, LysG prevents lysE
activation by the ArgPd proteins in media not supplemented with any co-effector. Our
findings suggest that mixed oligomers between ArgPd and LysG are being generated
under these conditions, at the level either of the dimer itself or of dimer of dimers upon
DNA binding, which are inactive for appropriate RNAP recruitment and gene activation.
The implication is that the homo-oligomers of LysG and ArgPd themselves activate lysE
by subtly different mechanisms that cannot be successfully integrated in the mixed
oligomers. In cultures with Lys supplementation, however, lysE activation was more or
less the same when LysG and the ArgPd variants were present either alone or together,
indicating that in this case gene regulation is mediated by a common or shared
mechanism; an alternative explanation is that the LysG and ArgPd protomers in their Lys-
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
147
bound states exclude one another and assemble only as homo-oligomers both in solution
and on DNA. Thus, although LysG mediates activation by all three amino acids of lysE
expression, it appears to do so by at least two different mechanisms that can be
distinguished by their interference or otherwise by the ArgPd proteins.
d) Target DNA bending upon binding by ArgP and its variants
As described in Chapter 1, for a number of LTTRs reduced DNA bending is
correlated with gene activation (Colyer and Kredich, 1996; Kullik et al., 1995; Akakura
and Winans, 2002; Kreusch et al., 1995; Wang et al., 1992; Kreusch et al., 1995; Parsek
et al., 1995; McFall et al., 1997; Dangel et al., 2005). Binding of the activating co-
effector causes a pronounced decrease in DNA bending and in the length of DNAse I
footprinted region. In this context, a sliding dimer model has been proposed (see Chapter
1 Section 1.2.6 and Figure 1.14). In this model, a non-activated LTTR tetramer binds to
the high affinity RBS site and to the proximal ABS’ subsite of the promoter with a high
angle bend. Activating co-effector induced conformational changes in the regulator
causes the regulator to slide to a more distal ABS’’ subsite, while still maintaining
binding to the RBS site, at the same time relaxing the angle of bending of the bound
DNA (Porrua et al., 2007; Monferrer et al., 2010).
However, in a few cases, such as MetR (Lorenz and Stauffer, 1995), DntR
(Smirnova et al., 2004) and TrpI (Pineiro et al., 1997), co-effector addition does not
affect protein induced DNA bending, and for NodD (Chen et al., 2005) the bending is in
fact increased in presence of the co-efffector. Studies with LTTR mutants such as those
for OxyR (Kullik et al., 1995), CysB (Colyer and Kredich, 1996), OccR (Akakura and
Winans, 2002) and CbbR (Dangel et al., 2005) indicate that constitutive activation can be
achieved by different mechanisms even for a single protein-target pair. Some variants
confer less DNA bending than their native counterparts, but others confer unchanged or
even increased bending. For OxyR, improved RNAP recruitment and a role for protein
conformational changes following its binding to DNA have also been suggested to
distinguish the constitutive mutants from the corresponding wild-type proteins (Kullik et
al., 1995). Binding to DNA can also dictate changes in LTTR oligomerization leading to
gene activation, as has been demonstrated for some mutants of the NAC protein (Rosario
et al., 2010).
Transcriptional cross-regulation between ArgP-argO and LysG-lysE Chapter 7
148
In this context, the present studies indicate that native ArgP binding induces DNA
bending at both argO and lysE, and that the three active ArgPd variants tested confer
relaxation of bending (relative to native ArgP) at these target loci. Thus, the results are
broadly consistent with those for other LTTRs on DNA bending induced by native
protein binding and its alteration with substitutions conferring constitutivity. These
results may be corroborated by more direct read-outs such as AFM-imaging or single
molecule FRET. At lysE, the ArgPd variants also show increased affinity of binding that
is proportionate to the magnitude by which they activate expression in vivo, suggestive of
an induced fit mechanism.
One possible discrepancy that remains to be explained is the finding that bending
by native ArgP at argO is unaffected by the activating co-effector Arg. An examination
of the ArgP bound argO regulatory region does reveal probable ABS’’ and ABS’
sequences at sites previously described for other LTTRs (Figure 7.8).
Figure 7.8 Probable RBS, ABS’’ and ABS’ sites on the ArgP DNase I footprinted region at argO. These
sites are proposed on the basis of sequence comparison with ArgP binding sequences at lysP and dapB
(Chapter 4, figure 4.6) and on the co-ordinates of the ABS’ sites in other LTTR targets. The –35 sequence
is underlined.
However, previous studies have shown that the DNase I footprint of ArgP at
argO is also unaffected by co-effector (Arg or Lys) addition (Laishram and
Gowrishankar, 2007), thus contradicting a probable explanation of activation through the
‘sliding dimer’ hypothesis. This suggests that either (i) the activating co-effector Arg acts
by mechanisms other than (or in addition to) that related to ArgP-induced DNA bending;
or (ii) although the concentration of Arg used (0.1 mM) is in accordance with the highest
concentration of co-effectors used in the case of other LTTRs for similar experiments,
this concentration may not be sufficient to elicit sustained, lower angle DNA bends by
the native ArgP at argO.
Genes activated at least 2-fold by ArgP Appendix I
151
Appendix I: Genes activated at least 2-fold in argP+/argP
d compared to ΔargP
Gene
name
Fold
difference Description
ygiL 7.6 Predicted fimbrial-like adhesin protein
glpB 6.5 sn-glycerol-3-phosphate dehydrogenase (anaerobic), membrane
anchor subunit
glpQ 6.1 Periplasmic glycerophosphodiester phosphodiesterase
glpC 5.9 sn-glycerol-3-phosphate dehydrogenase (anaerobic), small
subunit
glpA 5.9 sn-glycerol-3-phosphate dehydrogenase (anaerobic), large
subunit, FAD/NAD(P)-binding
argP 5.8 DNA-binding transcriptional activator, replication initiation
inhibitor
ibpB 5.2 heat shock chaperone
glpD 4.9 sn-glycerol-3-phosphate dehydrogenase, aerobic, FAD/NAD(P)-
binding
tnaA 4.9 tryptophanase/L-cysteine desulfhydrase, PLP-dependent
atoS 4.7 Sensory histidine kinase in two-component regulatory system
with AtoC
glpK 4.6 Glycerol kinase
ytfQ 4.5 predicted sugar transporter subunit: periplasmic-binding
component of ABC superfamily
ygeV 4.3 predicted DNA-binding transcriptional regulator
dctA 3.9 C4-dicarboxylic acid, orotate and citrate transporter
ytfR 3.8 predicted sugar transporter subunit: ATP-binding component of
ABC superfamily
glpT 3.8 sn-glycerol-3-phosphate transporter
yqeB 3.8 conserved protein with NAD(P)-binding Rossman fold
sgcR 3.8 KpLE2 phage-like element; predicted DNA-binding
transcriptional regulator
cstA 3.7 carbon starvation protein
garR 3.7 Tartronate semialdehyde reductase
yghW 3.6 predicted protein
Genes activated at least 2-fold by ArgP Appendix I
152
Gene
name
Fold
difference Description
mglA 3.6 fused methyl-galactoside transporter subunits of ABC
superfamily:ATP-binding components
tdcB 3.5 Catabolic threonine dehydratase, PLP-dependent
ibpA 3.4 heat shock chaperone
garL 3.3 alpha-dehydro-beta-deoxy-D-glucarate aldolase
garK 3.2 Glyceratekinase I
fucO 3.2 L-1,2-propanediol oxidoreductase
ygfJ 3.2 conserved protein
ycgB 3.2 conserved protein
yniA 3.2 Predicted phosphotransferase/kinase
garP 3.2 predicted (D)-galactarate transporter
ydcS 3.1 Predicted spermidine/putrescine transporter subunit
clpB 3.1 Protein disaggregation chaperone
bssR 3.0 conserved protein
lsrK 2.9 autoinducer-2 (AI-2) kinase
atoC 2.9 fused response regulator of ato operon, in two-component system
with AtoS: response regulator/sigma54 interaction protein
ptsA 2.9 fused predicted PTS enzymes: Hpr component/enzyme I
component/enzyme IIA component
yeiT 2.9 Predicted oxidoreductase
hyaC 2.9 hydrogenase 1, b-type cytochrome subunit
fumC 2.8 Fumarate hydratase (fumarase C),aerobic Class II
fumA 2.8 Fumarate hydratase (fumarase A), aerobic Class I
glgC 2.8 glucose-1-phosphate adenylyl transferase
yjfF 2.8 predicted sugar transporter subunit
araG 2.8 fused L-arabinose transporter subunits of ABC superfamily:
ATP-binding components
ytfT 2.7 predicted sugar transporter: membrane component of ABC family
Genes activated at least 2-fold by ArgP Appendix I
153
Gene
name
Fold
difference Description
dapB 2.7 Dihydrodipicolinate reductase
hyaD 2.7 protein involved in processing of HyaA and HyaB proteins
glgS 2.7 predicted glycogen synthesis protein
lsrB 2.7 AI2 transporter
ydiR 2.6 predicted electron transfer flavoprotein, FAD-binding
ybaE 2.6 predicted transporter subunit: periplasmic-binding component of
ABC superfamily
pck 2.6 Phosphoenol pyruvate carboxykinase
yccJ 2.6 predicted protein
wrbA 2.6 Predicted flavoprotein in Trp regulation
yraK 2.6 Predicted fimbrial-like adhesin protein
araF 2.5 L-arabinose transporter subunit
ymgF 2.5 predicted protein
yeiA 2.5 Predicted oxidoreductase
ygcB 2.5 Conserved protein, member of DEAD box family
araB 2.4 L-ribulokinase
rhsA 2.4 rhsA element core protein RshA
lsrG 2.4 autoinducer-2 (AI-2) modifying protein LsrG
hha 2.4 Modulator of gene expression, with H-NS
ybaJ 2.4 predicted protein
cpxP 2.4 periplasmic protein combats stress
melR 2.4 DNA-binding transcriptional dual regulator
aldA 2.4 Aldehydedehydrogenase A, NAD-linked
hyaE 2.4 protein involved in processing of HyaA and HyaB proteins
yejG 2.4 predicted protein
yfaW 2.4 Predicted enolase
argO 2.3 arginine transporter
Genes activated at least 2-fold by ArgP Appendix I
154
Gene
name
Fold
difference Description
lysP 2.3 lysine transporter
glgB 2.3 1,4-alpha-glucan branching enzyme
yfiQ 2.3 fused predicted acyl-CoA synthetase: NAD(P)-binding
subunit/ATP-binding subunit
yeaH 2.3 conserved protein
uspF 2.3 stress-induced protein, ATP-binding protein
sdhC 2.3 Succinate dehydrogenase, membrane subunit, binds cytochrome
b556
ydcJ 2.3 Conserved protein
tnaC 2.2 tryptophanase leader peptide
ftnB 2.2 Predictedferritin-like protein
lsrF 2.2 putative autoinducer-2 (AI-2) aldolase
fadI 2.2 beta-ketoacyl-CoA thiolase, anaerobic, subunit
yqeC 2.2 conserved protein
yjfN 2.2 predicted protein
ybdD 2.2 conserved protein
glgA 2.1 Glycogen synthase
ypfM 2.1 hypothetical protein b4606
ydcH 2.1 predicted protein
frdD 2.1 Fumarate reductase (anaerobic), membrane anchor subunit
modA 2.1 Molybdate transporter subunit
yggE 2.1 Conserved protein
yqeF 2.0 Predicted acyl transferase
yjfO 2.0 Conserved protein
srlE 2.0 glucitol/sorbitol-specific enzyme IIB component of PTS
ygdH 2.0 conserved protein
ycbJ 1.9 conserved protein
Genes activated at least 2-fold by ArgP Appendix I
155
Gene
name
Fold
difference Description
tdcC 1.9 L-threonine/L-serine transporter
hyaB 1.9 hydrogenase 1, large subunit
glpF 1.9 glycerol facilitator
sgcQ 1.9 KpLE2 phage-like element; predicted nucleoside triphosphatase
yfcH 1.9 conserved protein with NAD(P)-binding Rossmann-fold domain
mglB 1.9 methyl-galactoside transporter subunit
sgcC 1.9 KpLE2 phage-like element; predicted phospho-transferase
enzyme IIC component
ydhV 1.9 Predicted oxidoreductase
hsdR 1.9 endonuclease R
bssS 1.9 predicted protein
ydcA 1.9 predicted protein
rihB 1.9 Ribonucleoside hydrolase 2
gudD 1.9 (D)-glucarate dehydratase 1
lysC 1.9 aspartokinase III
ybeL 1.9 conserved protein
lsrR 1.9 lsr operon transcriptional repressor
srlB 1.9 glucitol/sorbitol-specific enzyme IIA component of PTS
lsrA 1.9 fused AI2 transporter subunits of ABC superfamily: ATP-
bindingcomponents
yaiY 1.9 predicted inner membrane protein
ugpB 1.9 Glycerol-3-phosphate transporter subunit
sdhD 1.9 Succinatedehydrogenase, membrane subunit, binds cytochrome
b556
glcC 1.9 DNA-binding transcriptional dual regulator, glycolate-binding
yiaY 1.8 predicted Fe-containing alcohol dehydrogenase
uspB 1.8 predicted universal stress (ethanol tolerance) protein B
sgcE 1.8 KpLE2 phage-like element; predicted epimerase
Genes activated at least 2-fold by ArgP Appendix I
156
Gene
name
Fold
difference Description
acs 1.8 acetyl-CoA synthetase
ylaC 1.8 predicted inner membrane protein
glgX 1.8 Glycogen debranching enzyme
hyaA 1.8 hydrogenase 1, small subunit
rbsB 1.8 D-ribose transporter subunit
ompW 1.8 outer membrane protein W
ucpA 1.8 Predicted oxidoredutase, sulfate metabolism protein
glgP 1.8 Glycogen phosphorylase
garD 1.8 (D)-galactarate dehydrogenase
sgcA 1.8 KpLE2 phage-like element; predicted phospho transferase
enzyme IIA component
ydeN 1.8 conserved protein
yhhA 1.8 conserved protein
bdm 1.8 Biofilm-dependent modulation protein
ydhX 1.8 predicted 4Fe-4S ferridoxin-type protein
hycF 1.8 Formate hydrogen lyase complex iron-sulfur protein
galS 1.8 DNA-binding transcriptional repressor
alpA 1.7 CP4-57 prophage; DNA-binding transcriptional activator
frc 1.7 Formyl-CoA transferase, NAD(P)-binding
yrhD 1.7 hypothetical protein b4612
fxsA 1.7 inner membrane protein
ydcT 1.7 Predicted spermidine/putrescine transporter subunit
dnaK 1.7 Chaperone Hsp70, co-chaperone with DnaJ
xylF 1.7 D-xylose transporter subunit
ydjF 1.7 predicted DNA-binding transcriptional regulator
frdC 1.7 Fumarate reductase (anaerobic), membrane anchor subunit
ydhW 1.7 predicted protein
Genes activated at least 2-fold by ArgP Appendix I
157
Gene
name
Fold
difference Description
ygcO 1.7 predicted 4Fe-4S cluster-containing protein
macB 1.7 Fused macrolide transporter subunits of ABC superfamily: ATP-
binding component/membrane component
rbsA 1.7 fused D-ribose transporter subunits of ABC superfamily: ATP-
binding components
yidE 1.7 predicted transporter
sgcB 1.7 predicted enzyme IIB component of PTS
allS 1.7 DNA-binding transcriptional activator of the all Doperon
caiF 1.7 DNA-binding transcriptional activator
zraP 1.7 Zn-binding periplasmic protein
arnT 1.7 4-amino-4-deoxy-L-arabinose transferase
bolA 1.6 regulator of penicillin binding proteins and beta-lactamase
transcription (morphogene)
fumB 1.6 anaerobic class I fumarate hydratase (fumarase B)
malF 1.6 maltose transporter subunit
frdB 1.6 Fumarate reductase (anaerobic), Fe-S subunit
ldrD 1.6 toxic polypeptide, small
yagX 1.6 predicted aromatic compound dioxygenase
malM 1.6 Maltose regulon periplasmic protein
fadD 1.6 acyl-CoA synthetase (long-chain-fatty-acid—CoAligase)
idnK 1.6 D-gluconate kinase, thermosensitive
hspQ 1.6 DNA-binding protein, hemimethylated
tdcD 1.6 Propionatekinase/acetate kinase C, anaerobic
ybjD 1.6 conserved protein with nucleoside triphosphatehydrolase domain
yqeG 1.5 predicted transporter
kdpF 1.5 Potassium ion accessory
transporter subunit
rbsC 1.5 D-ribose transporter subunit
Genes activated at least 2-fold by ArgP Appendix I
158
Gene
name
Fold
difference Description
adiY 1.5 DNA-binding transcriptional activator
cysQ 1.5 PAPS (adenosine 3'-phosphate 5'-phosphosulfate) 3'(2'),5'-
bisphosphate nucleotidase
gutM 1.5 DNA-binding transcriptional activator of glucitol operon
yjjM 1.5 predicted DNA-binding transcriptional regulator
feaR 1.5 DNA-binding transcriptional dual regulator
yebE 1.5 conserved protein
hyaF 1.5 Protein involved in nickel incorporation into hydrogenase-1
proteins
ydaS 1.5 Rac prophage; predicted DNA-binding transcriptional regulator
aldB 1.5 Aldehydedehydrogenase B
mhpR 1.5 DNA-binding transcriptional activator, 3HPP-binding
mtlA 1.5 Fused mannitol-specific PTS enzymes: IIA components/IIB
components/IIC components
sgcX 1.5 KpLE2 phage-like element; predicted endoglucanase with Zn-
dependent exopeptidase domain
uspE 1.5 stress-induced protein
hybA 1.5 hydrogenase 2 4Fe-4S ferredoxin-type component
yibA 1.5 lyase containing HEAT-repeat
ydhM 1.4 predicted DNA-binding transcriptional regulator
deoC 1.4 2-deoxyribose-5-phosphate aldolase, NAD(P)-linked
yaaW 1.4 conserved protein
tnaB 1.4 tryptophan transporter of low affinity
yeaC 1.4 conserved protein
yneM 1.4 hypothetical protein b4599
ydiL 1.4 conserved protein
ybiI 1.4 conserved protein
hycG 1.4 hydrogenase 3 and formate hydrogenase complex, HycG subunit
Genes activated at least 2-fold by ArgP Appendix I
159
Gene
name
Fold
difference Description
araH 1.4 fused L-arabinose transporter subunits of ABC superfamily:
membrane components
aspA 1.4 aspartate ammonia-lyase
yjhI 1.4 KpLE2 phage-like element; predicted DNA-binding
transcriptional regulator
yhcO 1.4 Predicted barnase inhibitor
hypB 1.4 GTP hydrolase involved in nickel liganding into hydrogenases
ybdJ 1.4 predicted inner membrane protein
ivbL 1.4 ilvB operon leader peptide
sufB 1.4 component of SufBCD complex
yecT 1.4 predicted protein
mtlD 1.4 mannitol-1-phosphate dehydrogenase, NAD(P)-binding
dadA 1.4 D-amino acid dehydrogenase
ybdK 1.4 gamma-glutamyl:cysteineligase
yeaG 1.4 conserved protein with nucleoside triphosphatehydrolase domain
mtlR 1.4 DNA-binding repressor
citC 1.4 Citrate lyase synthetase
ybiA 1.3 conserved protein
mokC 1.3 regulatory protein for HokC, overlaps CDS of hokC
lsrC 1.3 AI2 transporter
yrhB 1.3 predicted protein
ygcL 1.3 predicted protein
hcaR 1.3 DNA-binding transcriptional activator of 3-phenylpropionic acid
catabolism
modB 1.3 molybdate transporter subunit
maeB 1.3 Fused malic enzyme predicted oxidoreductase/predicted
phosphotransacetylase
Genes activated at least 2-fold by ArgP Appendix I
160
Gene
name
Fold
difference
Description
msrB 1.3 Methionine sulfoxide reductase B
eptB 1.3 predicted metal dependent hydrolase
yibT 1.3 predicted protein
ygcN 1.3 Predicted oxidoreductase with FAD/NAD(P)-binding domain
ybaT 1.3 predicted transporter
hycE 1.3 Hydrogenase 3, large subunit
dcuB 1.3 C4-dicarboxylate antiporter
smf 1.3 conserved protein
uspC 1.3 universal stress protein
yeeN 1.3 conserved protein
gdhA 1.3 Glutamatedehydrogenase, NADP-specific
hypC 1.3 protein required for maturation of hydrogenases 1 and 3
glcB 1.3 Malatesynthase G
frdA 1.3 Fumarate reductase (anaerobic) catalytic and NAD/flavoprotein
subunit
tisB 1.3 lexA-regulated toxic peptide
lamB 1.3 maltose outer membrane porin (maltoporin)
gspG 1.3 Pseudopilin, cryptic, general secretion pathway
gudX 1.3 Predicted glucarate dehydratase
aidB 1.2 Isovaleryl CoA dehydrogenase
yfjY 1.2 CP4-57 prophage; predicted DNA repair protein
yjgL 1.2 predicted protein
ycgY 1.2 predicted protein
yacL 1.2 conserved protein
lsrD 1.2 AI2 transporter
raiA 1.2 cold shock protein associated with 30S ribosomal subunit
fre 1.2 Flavin reductase
Genes activated at least 2-fold by ArgP Appendix I
161
Gene
name
Fold
difference Description
ydcU 1.2 Predicted spermidine/putrescine transporter subunit
yojO 1.2 hypothetical protein b4604
yqiA 1.2 predicted esterase
yehA 1.2 Predicted fimbrial-like adhesin protein
tdcA 1.2 DNA-binding transcriptional activator
gltS 1.2 Glutamate transporter
ybhF 1.2 fused predicted transporter subunits of ABC superfamily: ATP-
binding components
pdhR 1.2 DNA-binding transcriptional dual regulator
yraJ 1.2 predicted outer membrane protein
yfgF 1.2 predicted inner membrane protein
ymfL 1.2 e14 prophage; predicted DNA-binding transcriptional regulator
gspD 1.2 General secretory pathway component, cryptic
ygcP 1.2 predicted anti-terminator regulatory protein
ade 1.2 cryptic adenine deaminase
wcaF 1.2 Predicted acyltransferase
rhaB 1.2 rhamnulokinase
dedD 1.2 conserved protein
yibI 1.2 predicted inner membrane protein
rcsA 1.2 DNA-binding transcriptional activator, co-regulator with RcsB
ydgD 1.2 predicted peptidase
wcaD 1.2 Predicted colanic acid polymerase
ydiB 1.2 quinate/shikimate 5-dehydrogenase, NAD(P)-binding
ydaU 1.2 Rac prophage; conserved protein
ubiD 1.2 3-octaprenyl-4-hydroxybenzoate decarboxylase
ydcI 1.2 predicted DNA-binding transcriptional regulator
aslA 1.2 Acryl sulfatase-like enzyme
Genes activated at least 2-fold by ArgP Appendix I
162
Gene
name
Fold
difference Description
trxC 1.2 thioredoxin 2
hokC 1.2 toxic membrane protein, small
malG 1.2 maltose transporter subunit
clpA 1.2 ATPase and specificity subunit of ClpA-ClpP ATP-dependent
serine protease, chaperone activity
yieE 1.1 Predicted phosphopantetheinyl transferase
yihT 1.1 Predicted aldolase
yebW 1.1 predicted protein
hybO 1.1 hydrogenase 2, small subunit
rhsB 1.1 rhsB element core protein RshB
arcA 1.1 DNA-binding response regulator in two-component regulatory
system with ArcB or CpxA
ygjR 1.1 predicted NAD(P)-binding dehydrogenase
ykfM 1.1 hypothetical protein, no homologs
ychH 1.1 predicted inner membrane protein
hybD 1.1 predicted maturation element for hydrogenase 2
uspG 1.1 universal stress protein UP12
asd 1.1 aspartate-semialdehyde dehydrogenase, NAD(P)-binding
zraS 1.1 Sensory histidine kinase in two-component regulatory system
with ZraR
ybhH 1.1 conserved protein
yfdP 1.1 CPS-53 (KpLE1) prophage; predicted protein
ybbC 1.1 Predicted protein
yhcF 1.1 predicted transcriptional regulator
ycaP 1.1 Conserved inner membrane protein
citD 1.1 Citrate lyase, acyl carrier (gamma) subunit
ydaM 1.1 Predicted diguanylate cyclase, GGDEF domain signalling protein
hycH 1.1 protein required for maturation of hydrogenase 3
Genes activated at least 2-fold by ArgP Appendix I
163
Gene
name
Fold
difference Description
malQ 1.1 4-alpha-glucanotransferase (amylomaltase)
gltI 1.1 Glutamate and aspartate transporter subunit
lyx 1.1 L-xylulose kinase
sdhA 1.1 Succinate dehydrogenase, flavoprotein subunit
yicJ 1.1 predicted transporter
ycbB 1.1 Predicted carboxypeptidase
yadI 1.1 predicted PTS Enzyme IIA
yadM 1.1 Predicted fimbrial-like adhesin protein
abgT 1.1 predicted cryptic aminobenzoyl-glutamate transporter
yedP 1.1 Conserved protein
tdcR 1.1 DNA-binding transcriptional activator
sfmA 1.1 Predicted fimbrial-like adhesin protein
yggG 1.1 predicted peptidase
dgsA 1.1 DNA-binding transcriptional repressor
ymgE 1.1 predicted inner membrane protein
yodD 1.1 predicted protein
ygcG 1.1 predicted protein
rhaD 1.1 rhamnulose-1-phosphate aldolase
yqgD 1.1 predicted inner membrane protein
ybhM 1.1 Conserved inner membrane protein
ldrC 1.1 toxic polypeptide, small
citX 1.0 apo-citrate lyase phosphoribosyl-dephospho-CoA transferase
ydhT 1.0 conserved protein
yhfL 1.0 conserved secreted peptide
srlA 1.0 glucitol/sorbitol-specific enzyme IIC component of PTS
fliC 1.0 flagellar filament structural protein (flagellin)
Genes activated at least 2-fold by ArgP Appendix I
164
Gene
name
Fold
difference Description
uspD 1.0 stress-induced protein
cspD 1.0 cold shock protein homolog
folC 1.0 Bifunctional folyl polyglutamate synthase/ dihydrofolate synthase
argT 1.0 lysine/arginine/ornithine transporter subunit
ydhI 1.0 Predicted inner membrane protein
yfgG 1.0 Predicted protein
fdoG 1.0 Formate dehydrogenase-O, large subunit
yohC 1.0 Predicted inner membrane protein
ychN 1.0 conserved protein
yejA 1.0 Predicted oligopeptide transporter subunit
sfsA 1.0 Predicted DNA-binding transcriptional regulator
ulaG 1.0 predicted L-ascorbate 6-phosphate lactonase
yliI 1.0 Predicteddehydrogenase
fsaB 1.0 fructose-6-phosphate aldolase 2
yjiS 1.0 Conserved protein
nagE 1.0 fused N-acetyl glucosamine specific PTS enzyme: IIC, IIB , and
IIA components
fruR 1.0 DNA-binding transcriptional dual regulator
yabI 1.0 Conserved inner membrane protein
malP 1.0 Maltodextrin phosphorylase
ydaE 1.0 Rac prophage; conserved protein
ansP 1.0 L-asparagine transporter
Genes activated at least 2-fold in ΔargP Appendix II
165
Appendix II: Genes activated at least 2-fold in ΔargP compared to argP+/argP
d
Gene
name
Fold
difference Description
glnK 4.5 nitrogen assimilation regulatory protein for GlnL, GlnE, and
AmtB
yhjX 3.4 predicted transporter
metF 3.3 5,10-methylenetetrahydrofolate reductase
tppB 3.3 predicted transporter
pyrC 3.3 dihydro-orotase
narK 3.2 nitrate/nitrite transporter
metN 3 DL-methionine transporter subunit
metI 2.9 DL-methionine transporter subunit
ydjN 2.8 predicted transporter
yeeD 2.8 conserved protein
yciW 2.8 predicted oxidoreductase
cysP 2.7 thiosulfate transporter subunit
metA 2.6 homoserine O-transsuccinylase
mhpF 2.6 acetaldehyde-CoA dehydrogenase II, NAD-binding
deaD 2.6 ATP-dependent RNA helicase
upp 2.6 uracil phosphoribosyltransferase
metK 2.6 methionine adenosyltransferase 1
moeB 2.5 molybdopterin synthase sulfurylase
glnQ 2.5 glutamine transporter subunit
asnA 2.5 asparagine synthetase A
napB 2.4 nitrate reductase, small, cytochrome C550 subunit,
periplasmic
cysD 2.4 sulfate adenylyltransferase, subunit 2
prs 2.4 phosphoribosylpyrophosphate synthase
focA 2.4 formate transporter
purU 2.4 Formyltetrahydrofolate hydrolase
Genes activated at least 2-fold in ΔargP Appendix II
166
Gene
name
Fold
difference Description
guaA 2.3 GMP synthetase (glutamine aminotransferase)
cysJ 2.2 sulfite reductase, alpha subunit, flavoprotein
ilvA 2.2 threonine deaminase
cysB 2.2 DNA-binding transcriptional dual regulator, O-acetyl-L-
serine-binding
napA 2.2 nitrate reductase, periplasmic, large subunit
yfgB 2.2 predicted enzyme
narL 2.2 DNA-binding response regulator in two-component regulatory
system with NarX (or NarQ)
aroH 2.2 3-deoxy-D-arabino-heptulosonate-7-phosphate synthase,
tryptophan repressible
pyrD 2.2 Dihydro-orotate oxidase, FMN-linked
metR 2.2 DNA-binding transcriptional activator, homocysteine-binding
lpxA 2.2 UDP-N-acetylglucosamine acetyltransferase
ycaD 2.2 predicted transporter
lnt 2.2 apolipoprotein N-acyltransferase
glnA 2.2 glutamine synthetase
ybjX 2.2 conserved protein
glnP 2.1 glutamine transporter subunit
rimM 2.1 16S rRNA processing protein
purK 2.1 N5-carboxyaminoimidazole ribonucleotide synthase
cysS 2.1 cysteinyl-tRNA synthetase
cysH 2.1 3'-phosphoadenosine 5'-phosphosulfate reductase
purM 2.1 Phosphoribosyl amino imidazole synthetase
ybiS 2.1 conserved protein
cysU 2.1 sulfate/thiosulfate transporter subunit
metB 2.1 cystathionine gamma-synthase, PLP-dependent
prmC 2.1 N5-glutamine methyltransferase, modifies release factors RF-
1 and RF-2
Genes activated at least 2-fold in ΔargP Appendix II
167
Gene
name
Fold
difference Description
rhlE 2.1 RNA helicase
yohK 2 predicted inner membrane protein
lipB 2 lipoyl-protein ligase
cysI 2 sulfite reductase, beta subunit, NAD(P)-binding, heme-
binding
yohJ 2 conserved inner membrane protein
thiI 2 Sulfur transferase required for thiamine and 4-thiouridine
biosynthesis
yebZ 2 predicted inner membrane protein
yciT 2 predicted DNA-binding transcriptional regulator
gltP 2 glutamate/aspartate:proton symporter
dinI 2 DNA damage-inducible protein I
prmB 2 N5-glutamine methyltransferase
narJ 2 molybdenum-cofactor-assembly chaperone subunit (delta
subunit) of nitrate reductase 1
ycdO 2 conserved protein
ychF 1.9 predicted GTP-binding protein
gltX 1.9 glutamyl-tRNA synthetase
yggT 1.9 predicted inner membrane protein
carB 1.9 carbamoyl-phosphate synthase large subunit
msbA 1.9 fused lipid transporter subunits of ABC superfamily:
membrane component/ATP-binding component
yfaY 1.9 conserved protein
ybhC 1.9 predicted pectinesterase
ybgF 1.9 predicted protein
purE 1.9 N5-carboxyaminoimidazole ribonucleotide mutase
folD 1.9 bifunctional 5,10-methylene-tetrahydrofolate dehydrogenase/
5,10-methylene-tetrahydrofolate cyclohydrolase
ycdY 1.9 conserved protein
Genes activated at least 2-fold in ΔargP Appendix II
168
Gene
name
Fold
difference Description
amiC 1.9 N-acetylmuramoyl-L-alanine amidase
yagJ 1.9 CP4-6 prophage; predicted protein
cysC 1.9 adenosine 5'-phosphosulfate kinase
shiA 1.9 shikimate transporter
mtr 1.9 tryptophan transporter of high affinity
ubiG 1.9 bifunctional 3-demethylubiquinone-9 3-methyltransferase/ 2-
octaprenyl-6-hydroxy phenol methylase
sbp 1.9 sulfate transporter subunit
exoX 1.9 DNA exonuclease X
purR 1.9 DNA-binding transcriptional repressor, hypoxanthine-binding
ddlA 1.9 D-alanine-D-alanine ligase A
mprA 1.9 DNA-binding transcriptional repressor of microcin B17
synthesis and multidrug efflux
napC 1.9 nitrate reductase, cytochrome c-type, periplasmic
sohB 1.9 predicted inner membrane peptidase
yggW 1.9 predicted oxidoreductase
rpsJ 1.9 30S ribosomal subunit protein S10
rfbD 1.9 dTDP-4-dehydrorhamnose reductase subunit, NAD(P)-
binding, of dTDP-L-rhamnose synthase
fabF 1.8 3-oxoacyl-ACP-synthii-monomer
cspA 1.8 major cold shock protein
yejM 1.8 predicted hydrolase, inner membrane
narI 1.8 nitrate reductase 1, gamma (cytochrome b(NR)) subunit
mnmA 1.8 tRNA (5-methylaminomethyl-2-thiouridylate)-
methyltransferase
ilvC 1.8 ketol-acid reductoisomerase, NAD(P)-binding
lysS 1.8 lysine tRNA synthetase, constitutive
ycgM 1.8 predicted isomerase/hydrolase
glyA 1.8 serine hydroxymethyltransferase
dapA 1.8 dihydrodipicolinate synthase
Genes activated at least 2-fold in ΔargP Appendix II
169
Gene
name
Fold
difference Description
dfp 1.8 fused 4'-phosphopantothenoylcysteine decarboxylase/phospho
pantothenoyl cysteine synthetase, FMN-binding
aroC 1.8 chorismate synthase
apt 1.8 Adenine phosphoribosyl transferase
rplS 1.8 50S ribosomal subunit protein L19
lolB 1.8 chaperone for lipoproteins
trmD 1.8 tRNA (guanine-1-)-methyltransferase
cysA 1.8 sulfate/thiosulfate transporter subunit
mhpD 1.8 2-keto-4-pentenoate hydratase
tyrB 1.8 Tyrosine aminotransferase, tyrosine-repressible, PLP-
dependent
ispF 1.8 2C-methyl-D-erythritol 2,4-cyclodiphosphate synthase
chaA 1.8 Calcium/sodium:proton antiporter
guaC 1.8 GMP reductase
cysN 1.8 sulfate adenylyltransferase, subunit 1
purB 1.8 adenylosuccinate lyase
glnH 1.8 glutamine transporter subunit
hemA 1.8 glutamyl tRNA reductase
proS 1.8 prolyl-tRNA synthetase
ilvD 1.8 dihydroxyacid dehydratase
pncB 1.8 nicotinate phosphoribosyl transferase
rho 1.8 transcription termination factor
modF 1.8 fused molybdate transporter subunits of ABC superfamily:
ATP-binding components
asnB 1.8 asparagine synthetase B
ycdU 1.8 predicted inner membrane protein
ispB 1.8 octaprenyl diphosphate synthase
ycbX 1.8 predicted 2Fe-2S cluster-containing protein
narX 1.8 Sensory histidine kinase in two-component regulatory system
with NarL
Genes activated at least 2-fold in ΔargP Appendix II
170
Gene
name
Fold
difference Description
ynfM 1.8 predicted transporter
prfB 1.8 peptide chain release factor RF-2
rnb 1.8 ribonuclease II
rstB 1.7 Sensory histidine kinase in two-component regulatory system
with RstA
potA 1.7 polyamine transporter subunit
purN 1.7 Phosphoribosyl glycinamide formyl transferase 1
rplC 1.7 50S ribosomal subunit protein L3
adk 1.7 adenylate kinase
miaB 1.7 isopentenyl-adenosine A37 tRNA methylthiolase
rfbB 1.7 dTDP-glucose 4,6 dehydratase, NAD(P)-binding
rhtC 1.7 threonine efflux system
mdoB 1.7 phosphoglycerol transferase I
yedD 1.7 predicted protein
kdsB 1.7 3-deoxy-manno-octulosonate cytidylyl transferase
prfA 1.7 peptide chain release factor RF-1
rhlB 1.7 ATP-dependent RNA helicase
nupC 1.7 nucleoside (except guanosine) transporter
pyrB 1.7 aspartate carbamoyl transferase, catalytic subunit
metQ 1.7 DL-methionine transporter subunit
ribC 1.7 riboflavin synthase, alpha subunit
glnL 1.7 Sensory histidine kinase in two-component regulatory system
with GlnG
glnB 1.7 regulatory protein P-II for glutamine synthetase
rluF 1.7 23S rRNA pseudouridine synthase
dnaX 1.7 DNA polymerase III/DNA elongation factor III, tau and
gamma subunits
gnsB 1.7 Qin prophage; predicted protein
moeA 1.7 molybdopterin biosynthesis protein
ilvM 1.7 acetolactate synthase II, small subunit
Genes activated at least 2-fold in ΔargP Appendix II
171
Gene
name
Fold
difference Description
purF 1.7 Amido phosphoribosyl transferase
dppB 1.7 dipeptide transporter
ompX 1.7 outer membrane protein
cysK 1.7 Cysteine synthase A, O-acetylserine sulfhydrolase A subunit
yciH 1.7 conserved protein
glnG 1.7
fused DNA-binding response regulator in two-component
regulatory system with GlnL: response regulator/sigma54
interaction protein
dcm 1.7 DNA cytosine methylase
yjcB 1.7 predicted inner membrane protein
nikE 1.7 nickel transporter subunit
ygaD 1.7 conserved protein
hydN 1.7 Formate dehydrogenase-H,
ansA 1.7 cytoplasmic L-asparaginase I
ybjT 1.7 conserved protein with NAD(P)-binding Rossmann-fold
domain
ypaA 1.7 predicted protein
ccmA 1.7 heme exporter subunit
aroA 1.7 5-enolpyruvylshikimate-3-phosphate synthetase
yicC 1.7 conserved protein
thrA 1.7 fused aspartokinase I and homoserine dehydrogenase I
yedJ 1.7 predicted phosphohydrolase
nhaB 1.7 sodium:proton antiporter
tyrS 1.7 tyrosyl-tRNA synthetase
yicE 1.6 predicted transporter
ppiD 1.6 Peptidyl-prolyl cis-trans isomerase (rotamase D)
dpiA 1.6 DNA-binding response regulator in two-component regulatory
system with citA
yoaE 1.6 fused predicted membrane protein/conserved protein
malY 1.6 bifunctional beta-cystathionase, PLP-dependent/ regulator of
Genes activated at least 2-fold in ΔargP Appendix II
172
maltose regulon
Gene
name
Fold
difference Description
rplD 1.6 50S ribosomal subunit protein L4
apaG 1.6 protein associated with Co2+
and Mg2+
efflux
ychM 1.6 predicted transporter
yccA 1.6 inner membrane protein
nusA 1.6 transcription termination/antitermination L factor
ymgD 1.6 predicted protein
cyoB 1.6 cytochrome o ubiquinol oxidase subunit I
manA 1.6 mannose-6-phosphate isomerase
mukF 1.6 Involved in chromosome partioning, Ca2+
binding protein
yigB 1.6 predicted hydrolase
sdaC 1.6 predicted serine transporter
ybaL 1.6 predicted transporter with NAD(P)-binding Rossmann-fold
domain
mltB 1.6 membrane-bound lytic murein transglycosylase B
ybfK 1.6 hypothetical protein b4590
ybjM 1.6 predicted inner membrane protein
fis 1.6 global DNA-binding transcriptional dual regulator
nrdH 1.6 glutaredoxin-like protein
dppD 1.6 dipeptide transporter
ispE 1.6 4-diphosphocytidyl-2-C-methylerythritol kinase
nrfG 1.6 heme lyase (NrfEFG) for insertion of heme into c552, subunit
NrfG
rdgC 1.6 DNA-binding protein
greA 1.6 transcription elongation factor
nadE 1.6 NAD synthetase, NH3/glutamine-dependent
leuB 1.6 3-isopropylmalate dehydrogenase
hemE 1.6 uroporphyrinogen decarboxylase
potB 1.6 polyamine transporter subunit
dusB 1.6 tRNA-dihydrouridine synthase B
Genes activated at least 2-fold in ΔargP Appendix II
173
Gene
name
Fold
difference Description
tolQ 1.6 membrane spanning protein in TolA-TolQ-TolR complex
aroG 1.6 3-deoxy-D-arabino-heptulosonate-7-phosphate synthase,
phenylalanine repressible
cysW 1.6 sulfate/thiosulfate transporter subunit
oppD 1.6 oligopeptide transporter subunit
ybcJ 1.6 predicted RNA-binding protein
thyA 1.6 thymidylate synthetase
hflD 1.6 predicted lysogenization regulator
rpsA 1.6 30S ribosomal subunit protein S1
torS 1.6 hybrid sensory histidine kinase in two-component regulatory
system with TorR
rsgA 1.6 ribosome small subunit-dependent GTPase A
ygfB 1.6 predicted protein
mhpE 1.6 4-hyroxy-2-oxovalerate/4-hydroxy-2-oxopentanoic acid
aldolase, class I
ybiT 1.6 fused predicted transporter subunits of ABC superfamily:
ATP-binding components
amtB 1.6 ammonium transporter
yadS 1.6 conserved inner membrane protein
mdtK 1.6 multidrug efflux system transporter
hscB 1.6 DnaJ-like molecular chaperone specific for IscU
rpmA 1.6 50S ribosomal subunit protein L27
dnaC 1.6 DNA biosynthesis protein
mqo 1.6 malate dehydrogenase, FAD/NAD(P)-binding domain
suhB 1.6 inositol monophosphatase
tig 1.6 Peptidyl-prolyl cis/trans isomerase (trigger factor)
kdsD 1.6 D-arabinose 5-phosphate isomerase
queA 1.6 S-adenosylmethionine:tRNA ribosyltransferase-isomerase
purD 1.6 Phosphoribosyl glycinamide synthetase phosphoribosylamine-
glycine ligase
Genes activated at least 2-fold in ΔargP Appendix II
174
Gene
name
Fold
difference Description
yrbD 1.6 predicted ABC-type organic solvent transporter
mukE 1.6 protein involved in chromosome partitioning
rnhA 1.6 ribonuclease HI, degrades RNA of DNA-RNA hybrids
metE 1.5 5-methyltetrahydropteroyltriglutamate-homocysteine S-
methyltransferase
rnt 1.5 ribonuclease T (RNase T)
yfhA 1.5 predicted DNA-binding response regulator in two-component
system
secD 1.5 SecYEG protein translocase auxillary subunit
yraL 1.5 predicted methyltransferase
speC 1.5 ornithine decarboxylase, constitutive
ydcP 1.5 predicted peptidase
tsf 1.5 protein chain elongation factor EF-Ts
cld 1.5 regulator of length of O-antigen component of
lipopolysaccharide chains
yfcD 1.5 predicted NUDIX hydrolase
rpsB 1.5 30S ribosomal subunit protein S2
holA 1.5 DNA polymerase III, delta subunit
yjbJ 1.5 predicted stress response protein
lepB 1.5 leader peptidase (signal peptidase I)
ybeA 1.5 conserved protein
yieG 1.5 predicted inner membrane protein
nusG 1.5 transcription termination factor
yqiC 1.5 conserved protein
mgtA 1.5 magnesium transporter
rlmL 1.5 predicted methyltransferase
fecB 1.5 KpLE2 phage-like element; iron-dicitrate transporter
slmA 1.5 Division inhibitor
speD 1.5 S-adenosylmethionine decarboxylase
bioD 1.5 dethiobiotin synthetase
Genes activated at least 2-fold in ΔargP Appendix II
175
Gene
name
Fold
difference Description
guaB 1.5 IMP dehydrogenase
leuA 1.5 2-isopropylmalate synthase
tyrP 1.5 Tyrosine transporter
metJ 1.5 DNA-binding transcriptional repressor, S-
adenosylmethionine-binding
murA 1.5 UDP-N-acetylglucosamine 1-carboxyvinyltransferase
dsbG 1.5 periplasmic disulfide isomerase/thiol-disulphide oxidase
cysZ 1.5 predicted inner membrane protein
kdgK 1.5 Keto deoxy gluconokinase
gcvH 1.5 glycine cleavage complex lipoyl protein
rplW 1.5 50S ribosomal subunit protein L23
panD 1.5 aspartate 1-decarboxylase
mltD 1.5 predicted membrane-bound lytic murein transglycosylase D
yajC 1.5 SecYEG protein translocase auxillary subunit
alaS 1.5 alanyl-tRNA synthetase
ydiK 1.5 predicted inner membrane protein
folE 1.5 GTP cyclohydrolase I
rplM 1.5 50S ribosomal subunit protein L13
typA 1.5 GTP-binding protein
ydeA 1.5 predicted arabinose transporter
yeiR 1.5 predicted enzyme
fkpB 1.5 FKBP-type peptidyl-prolyl cis-trans isomerase (rotamase)
iaaA 1.5 L-asparaginase
kdsC 1.5 3-deoxy-D-manno-octulosonate 8-phosphate phosphatase
gyrB 1.5 DNA gyrase, subunit B
aroK 1.5 shikimate kinase I
speA 1.5 biosynthetic arginine decarboxylase, PLP-binding
putA 1.5 fused DNA-binding transcriptional regulator/proline
dehydrogenase/pyrroline-5-carboxylate dehydrogenase
rpsP 1.5 30S ribosomal subunit protein S16
Genes activated at least 2-fold in ΔargP Appendix II
176
Gene
name
Fold
difference Description
acrA 1.5 multidrug efflux system
ybbB 1.5 tRNA 2-selenouridine synthase, selenophosphate-dependent
pepQ 1.5 proline dipeptidase
appY 1.5 DLP12 prophage; DNA-binding transcriptional activator
fdx 1.5 [2Fe-2S] ferredoxin [Escherichia coli K12]
bcp 1.5 thiol peroxidase, thioredoxin-dependent
btuF
1.5
vitamin B12 transporter subunit: periplasmic-binding
component of ABC superfamily
era 1.5 membrane-associated, 16S rRNA-binding GTPase
gyrA 1.5 DNA gyrase (type II topoisomerase), subunit A
rrmA 1.5 23S rRNA m1G745 methyltransferase
truB 1.4 tRNA pseudouridine synthase
rpsH 1.4 30S ribosomal subunit protein S8
pyrF 1.4 orotidine-5'-phosphate decarboxylase
bcr 1.4 bicyclomycin/multidrug efflux system
yebC 1.4 conserved protein
glpR 1.4 DNA-binding transcriptional repressor
marB 1.4 predicted protein
ruvC 1.4 component of RuvABC resolvasome, endonuclease
yidR 1.4 conserved protein
rplB 1.4 50S ribosomal subunit protein L2
psd 1.4 Phosphatidyl serine decarboxylase
yjcE 1.4 predicted cation/proton antiporter
hemN 1.4 coproporphyrinogen III oxidase, SAM and NAD(P)H
dependent, oxygen-independent
dut 1.4 Deoxyuridine triphosphatase
metC 1.4 cystathionine beta-lyase, PLP-dependent
csrA 1.4 pleiotropic regulatory protein for carbon source metabolism
menC 1.4 o-succinylbenzoyl-CoA synthase
yfbQ 1.4 predicted aminotransferase
Genes activated at least 2-fold in ΔargP Appendix II
177
Gene
name
Fold
difference Description
rdgB 1.4 dITP/XTP pyrophosphatase
serU 1.4 Ser tRNA
cyoD 1.4 cytochrome ubiquinol oxidase subunit IV
rplQ 1.4 50S ribosomal subunit protein L17
map 1.4 methionine aminopeptidase
rsmC 1.4 16S rRNA m2G1207 methylase
dsbA 1.4 periplasmic protein disulfide isomerase I
glyQ 1.4 glycine tRNA synthetase, alpha subunit
ygdE 1.4 predicted methyltransferase
ribF 1.4 bifunctional riboflavin kinase/FAD synthetase
pth 1.4 peptidyl-tRNA hydrolase
yjaG 1.4 conserved protein
trpD 1.4 fused glutamine amidotransferase (component II) of
anthranilate synthase/anthranilate phosphoribosyl transferase
fmt 1.4 10-formyltetrahydrofolate:L-methionyl-tRNA(fMet) N-
formyltransferase
recJ 1.4 ssDNA exonuclease, 5' --> 3'-specific
lolD 1.4 outer membrane-specific lipoprotein transporter subunit
yqfB 1.4 conserved protein
spoT 1.4 bifunctional (p)ppGpp synthetase II/ guanosine-3',5'-bis
pyrophosphate 3'-pyrophosphohydrolase
plsX 1.4 fatty acid/phospholipid synthesis protein
allR 1.4 DNA-binding transcriptional repressor
yheS 1.4 fused predicted transporter subunits of ABC superfamily:
ATP-binding components
motA 1.4 proton conductor component of flagella motor
rsxG 1.4 predicted oxidoreductase
pyrI 1.4 aspartate carbamoyltransferase, regulatory subunit
ycdX 1.4 predicted zinc-binding hydrolase
pnp 1.4 polynucleotide phosphorylase/polyadenylase
Genes activated at least 2-fold in ΔargP Appendix II
178
Gene
name
Fold
difference Description
thrC 1.4 threonine synthase
rpsI 1.4 30S ribosomal subunit protein S9
glmU 1.4 fused N-acetyl glucosamine-1-phosphate
uridyltransferase/glucosamine-1-phosphate acetyl transferase
pdxB 1.4 erythronate-4-phosphate dehydrogenase
rsmB 1.4 16S rRNA m5C967 methyltransferase, S-adenosyl-L-
methionine-dependent
ydiJ 1.4 predicted FAD-linked oxidoreductase
yheT 1.4 predicted hydrolase
hslU 1.4 molecular chaperone and ATPase component of HslUV
protease
nemA 1.4 N-ethylmaleimide reductase, FMN-linked
nlpB 1.4 lipoprotein
rpsS 1.4 30S ribosomal subunit protein S19
hrpB 1.4 predicted ATP-dependent helicase
tgt 1.4 tRNA-guanine transglycosylase
ackA 1.4 acetate kinase A and propionate kinase 2
yceB 1.4 predicted lipoprotein
ycaR 1.4 conserved protein
bcsA 1.4 cellulose synthase, catalytic subunit
rplX 1.4 50S ribosomal subunit protein L24
pldA 1.4 outer membrane phospholipase A
mdoG 1.4 glucan biosynthesis protein, periplasmic
mdtE 1.4 multidrug resistance efflux transporter
argS 1.4 arginyl-tRNA synthetase
yadG 1.4 predicted transporter subunit: ATP-binding component of
ABC superfamily
glmM 1.4 phosphoglucosamine mutase
hpt 1.4 hypoxanthine phosphoribosyltransferase
rplU 1.4 50S ribosomal subunit protein L21
Genes activated at least 2-fold in ΔargP Appendix II
179
Gene
name
Fold
difference Description
grxC 1.4 glutaredoxin 3
ybbO 1.4 predicted oxidoreductase with NAD(P)-binding Rossmann-
fold domain
yobA 1.4 conserved protein
gpp 1.4 guanosine pentaphosphatase/exopolyphosphatase
ftsY 1.4 fused Signal Recognition Particle (SRP) receptor: membrane
binding protein/conserved protein
lepA 1.3 GTP-binding membrane protein
rplN 1.3 50S ribosomal subunit protein L14
yjeP 1.3 predicted mechanosensitive channel
trpE 1.3 component I of anthranilate synthase
thrS 1.3 threonyl-tRNA synthetase
oxyR 1.3 DNA-binding transcriptional dual regulator
dhaK 1.3 dihydroxyacetone kinase, N-terminal domain
metG 1.3 methionyl-tRNA synthetase
mreB 1.3 cell wall structural complex MreBCD, actin-like component
MreB
ccmD 1.3 cytochrome c biogenesis protein
ygfZ 1.3 predicted folate-dependent regulatory protein
ndh 1.3 respiratory NADH dehydrogenase 2/cupric reductase
ydfO 1.3 Qin prophage; predicted protein
tdk 1.3 thymidine kinase/deoxyuridine kinase
napG 1.3 ferredoxin-type protein essential for electron transfer from
ubiquinol to periplasmic nitrate reductase (NapAB)
atpH 1.3 F1 sector of membrane-bound ATP synthase, delta subunit
dnaB 1.3 replicative DNA helicase
dsbC 1.3 protein disulfide isomerase II
yjjK 1.3 fused predicted transporter subunits of ABC superfamily:
ATP-binding components
csgE 1.3 predicted transport protein
Genes activated at least 2-fold in ΔargP Appendix II
180
Gene
name
Fold
difference Description
yggS 1.3 predicted enzyme
ydgH 1.3 predicted protein
yadH 1.3 predicted transporter subunit: membrane component of ABC
superfamily
gcd 1.3 glucose dehydrogenase
atpE 1.3 F0 sector of membrane-bound ATP synthase, subunit c
dacB 1.3 D-alanyl-D-alanine carboxypeptidase
yacG 1.3 conserved protein
rpe 1.3 D-ribulose-5-phosphate 3-epimerase
ycdZ 1.3 predicted inner membrane protein
iscX 1.3 conserved protein
panB 1.3 3-methyl-2-oxobutanoate hydroxymethyl transferase
hda 1.3 ATPase regulatory factor involved in DnaA inactivation
narH 1.3 nitrate reductase 1, beta (Fe-S) subunit
cca 1.3 fused tRNA nucleotidyl transferase/2'3'-cyclic
phosphodiesterase/2'nucleotidase and phosphatase
ycfS 1.3 conserved protein
rpsF 1.3 30S ribosomal subunit protein S6
rarA 1.3 recombination protein
pyrH 1.3 uridylate kinase
apaH 1.3 diadenosine tetraphosphatase
rluB 1.3 23S rRNA pseudouridylate synthase
ydiA 1.3 conserved protein
yneH 1.3 predicted glutaminase
yieH 1.3 predicted hydrolase
tehB 1.3 predicted S-adenosyl-L-methionine-dependent
methyltransferase
solA 1.3 N-methyltryptophan oxidase, FAD-binding
udk 1.3 uridine/cytidine kinase
gnsA 1.3 predicted regulator of phosphatidylethanolamine synthesis
Genes activated at least 2-fold in ΔargP Appendix II
181
Gene
name
Fold
difference Description
racR 1.3 Rac prophage; predicted DNA-binding transcriptional
regulator
mrdA 1.3 transpeptidase involved in peptidoglycan synthesis (penicillin-
binding protein 2)
yciB 1.3 predicted inner membrane protein
priB 1.3 primosomal protein N
selA 1.3 selenocysteine synthase
mreC 1.3 cell wall structural complex MreBCD transmembrane
component MreC
yrbK 1.3 conserved protein
yddW 1.3 predicted liprotein
purL 1.3 phosphoribosylformyl-glycineamide synthetase
yggX 1.3 protein that protects iron-sulfur proteins against oxidative
damage
hemF 1.3 coproporphyrinogen III oxidase
ruvA 1.3 component of RuvABC resolvasome, regulatory subunit
cyaY 1.3 frataxin, iron-binding and oxidizing protein
rffD 1.3 UDP-N-acetyl-D-mannosaminuronic acid dehydrogenase
gfcA 1.3 predicted protein
nadC 1.3 quinolinate phosphoribosyltransferase
napD 1.3 assembly protein for periplasmic nitrate reductase
hcaB 1.3 2,3-dihydroxy-2,3-dihydrophenylpropionate dehydrogenase
ppx 1.3 exopolyphosphatase
yacF 1.3 conserved protein
glnS 1.3 glutamyl-tRNA synthetase
emrA 1.3 multidrug efflux system
yfiH 1.3 conserved protein
proQ 1.3 predicted structural transport element
cydC 1.3 fused cysteine transporter subunits of ABC superfamily:
membrane component/ATP-binding component
Genes activated at least 2-fold in ΔargP Appendix II
182
Gene
name
Fold
difference Description
creC 1.3 sensory histidine kinase in two-component regulatory system
with CreB or PhoB, regulator of the CreBC regulon
hscA 1.3 DnaK-like molecular chaperone specific for IscU
cydD 1.3 fused cysteine transporter subunits of ABC superfamily:
membrane component/ATP-binding component
ybjS 1.3 predicted NAD(P)H-binding oxidoreductase with NAD(P)-
binding Rossmann-fold domain
bioC 1.3 predicted methltransferase, enzyme of biotin synthesis
pdxA 1.3 4-hydroxy-L-threonine phosphate dehydrogenase, NAD-
dependent
ribB 1.3 3,4-dihydroxy-2-butanone-4-phosphate synthase
nudJ 1.3 bifunctional thiamin pyrimidine pyrophosphate hydrolase/
thiamin pyrophosphate hydrolase
rpoA 1.3 RNA polymerase, alpha subunit
rfaG 1.3 glucosyltransferase I
ptsG 1.3 fused glucose-specific PTS enzymes: IIB component/IIC
component
lgt 1.3 phosphatidylglycerol-prolipoprotein diacylglyceryl transferase
gph 1.3 phosphoglycolate phosphatase
recA 1.3 DNA strand exchange and recombination protein with
protease and nuclease activity
dacA 1.3 D-alanyl-D-alanine carboxypeptidase (penicillin-binding
protein 5)
trxB 1.3 thioredoxin reductase, FAD/NAD(P)-binding
diaA 1.3 DnaA initiator-associating factor for replication initiation
proA 1.3 gamma-glutamylphosphate reductase
yeaE 1.3 predicted oxidoreductase
ydhJ 1.3 undecaprenyl pyrophosphate phosphatase
ygiF 1.3 predicted adenylate cyclase
proB 1.3 gamma-glutamate kinase
Genes activated at least 2-fold in ΔargP Appendix II
183
Gene
name
Fold
difference Description
lpxK 1.3 lipid A 4'kinase
hisS 1.2 histidyl tRNA synthetase
ybeB 1.2 predicted protein
yhbY 1.2 predicted RNA-binding protein
yaiL 1.2 nucleoprotein/polynucleotide-associated enzyme
purH 1.2 fused IMP cyclohydrolase/phosphoribosyl amino imidazole
carboxamide formyl transferase
sppA 1.2 protease IV (signal peptide peptidase)
kgtP 1.2 alpha-ketoglutarate transporter
ybhA 1.2 predicted hydrolase
minD 1.2 membrane ATPase of the MinC-MinD-MinE system
rstA 1.2 DNA-binding response regulator in two-component regulatory
system with RstB
yaeB 1.2 conserved protein
yiaF 1.2 conserved protein
ytjB 1.2 conserved protein
rluD 1.2 23S rRNA pseudouridine synthase
yigL 1.2 predicted hydrolase
yciN 1.2 predicted protein
hsdM 1.2 DNA methylase M
yjgP 1.2 conserved inner membrane protein
folM 1.2 dihydrofolate reductase isozyme
uup 1.2 fused predicted transporter subunits of ABC superfamily:
ATP-binding components
ybdG 1.2 predicted mechanosensitive channel
rluE 1.2 23S rRNA pseudouridine synthase
pcnB 1.2 poly(A) polymerase I
yegQ 1.2 predicted peptidase
rplE 1.2 50S ribosomal subunit protein L5
ycgL 1.2 conserved protein
Genes activated at least 2-fold in ΔargP Appendix II
184
Gene
name
Fold
difference Description
ybiV 1.2 predicted hydrolase
yfgM 1.2 conserved protein
essD 1.2 DLP12 prophage; predicted phage lysis protein
mrcA 1.2 fused penicillin-binding protein 1a: murein
transglycosylase/murein transpeptidase
yceD 1.2 conserved protein
lspA 1.2 prolipoprotein signal peptidase (signal peptidase II)
rfaQ 1.2 lipopolysaccharide core biosynthesis protein
serC 1.2 3-phosphoserine/phosphohydroxythreonine aminotransferase
cyoE 1.2 protoheme IX farnesyltransferase
cueO 1.2 multicopper oxidase (laccase)
mnmC 1.2 fused 5-methylaminomethyl-2-thiouridine-forming enzyme
methyltransferase/FAD-dependent demodification enzyme
ydiI 1.2 conserved protein
serA 1.2 D-3-phosphoglycerate dehydrogenase
yebY 1.2 predicted protein
yiaD 1.2 predicted outer membrane lipoprotein
menB 1.2 dihydroxynaphthoic acid synthetase
yfiO 1.2 predicted lipoprotein
ribA 1.2 GTP cyclohydrolase II
yhcC 1.2 predicted Fe-S oxidoreductase
napH 1.2 ferredoxin-type protein essential for electron transfer from
ubiquinol to periplasmic nitrate reductase (NapAB)
hflC 1.2 modulator for HflB protease specific for phage lambda cII
repressor
surA 1.2 peptidyl-prolyl cis-trans isomerase (PPIase)
dcyD 1.2 D-cysteine desulfhydrase, PLP-dependent
metL 1.2 fused aspartokinase II/homoserine dehydrogenase II
ksgA 1.2 S-adenosylmethionine-6-N',N'-adenosyl (rRNA)
dimethyltransferase
Genes activated at least 2-fold in ΔargP Appendix II
185
Gene
name
Fold
difference Description
avtA 1.2 valine-pyruvate aminotransferase
holC 1.2 DNA polymerase III, chi subunit
focB 1.2 predicted formate transporter
yajG 1.2 predicted lipoprotein
fnr 1.2 DNA-binding transcriptional dual regulator, global regulator
of anaerobic growth
mpl 1.2 UDP-N-acetylmuramate:L-alanyl-gamma-D-glutamyl-meso-
diaminopimelate ligase
plsC 1.2 1-acyl-sn-glycerol-3-phosphate acyltransferase
yfdH 1.2 CPS-53 (KpLE1) prophage; bactoprenol glucosyl transferase
yiiX 1.2 predicted peptidoglycan peptidase
cysM 1.2 cysteine synthase B (O-acetylserine sulfhydrolase B)
thrB 1.2 homoserine kinase
pykF 1.2 pyruvate kinase I
fabA 1.2 beta-hydroxydecanoyl thioester dehydrase
rlpB 1.2 minor lipoprotein
asnS 1.2 asparaginyl tRNA synthetase
sieB 1.2 Rac prophage; phage superinfection exclusion protein
cpxA 1.2 sensory histidine kinase in two-component regulatory system
with CpxR
proC 1.2 pyrroline-5-carboxylate reductase, NAD(P)-binding
uraA 1.2 uracil transporter
rffC 1.2 TDP-fucosamine acetyltransferase
rnlA 1.2 CP4-57 prophage; RNase LS
gshA 1.2 gamma-glutamate-cysteine ligase
ttcA 1.2 predicted C32 tRNA thiolase
xseA 1.2 exonuclease VII, large subunit
fabD 1.2 malonyl-CoA
tolA 1.2 membrane anchored protein in TolA-TolQ-TolR complex
btuE 1.2 predicted glutathione peroxidase
Genes activated at least 2-fold in ΔargP Appendix II
186
Gene
name
Fold
difference Description
tatD 1.2 DNase, magnesium-dependent
galR 1.2 DNA-binding transcriptional repressor
yhbE 1.2 conserved inner membrane protein
narP 1.2 DNA-binding response regulator in two-component regulatory
system with NarQ or NarX
recO 1.2 gap repair protein
amn 1.2 AMP nucleosidase
ygaH 1.2 predicted inner membrane protein
dnaA 1.2 chromosomal replication initiator protein DnaA, DNA-binding
transcriptional dual regulator
bioA 1.2 7,8-diaminopelargonic acid synthase, PLP-dependent
yraN 1.2 conserved protein
brnQ 1.2 predicted branched chain amino acid transporter (LIV-II)
ribE 1.2 riboflavin synthase beta chain
fklB 1.2 FKBP-type peptidyl-prolyl cis-trans isomerase (rotamase)
bisC 1.2 biotin sulfoxide reductase
ydiH 1.2 predicted protein
ygiQ 1.2 conserved protein
yecE 1.2 conserved protein
rffT 1.2 TDP-Fuc4NAc:lipidIIFuc4NAc transferase
yjjU 1.1 predicted esterase
racC 1.1 Rac prophage; predicted protein
oppF 1.1 oligopeptide transporter subunit
nudK 1.1 predicted NUDIX hydrolase
fabB 1.1 3-oxoacyl-ACP-synthase I
crcB 1.1 predicted inner membrane protein associated with
chromosome condensation
gloA 1.1 glyoxalase I, Ni-dependent
fldA 1.1 flavodoxin 1
lptB 1.1 predicted transporter subunit: ATP-binding component of
Genes activated at least 2-fold in ΔargP Appendix II
187
ABC superfamily
Gene
name
Fold
difference Description
tolC 1.1 transport channel
nusB 1.1 transcription antitermination protein
minE 1.1 cell division topological specificity factor
rdoA 1.1 Thr/Ser kinase implicated in Cpx stress response
yfiP 1.1 conserved protein
yghG 1.1 predicted protein
aroL 1.1 shikimate kinase II
rpsG 1.1 30S ribosomal subunit protein S7
fecE 1.1 KpLE2 phage-like element; iron-dicitrate transporter subunit
fusA 1.1 protein chain elongation factor EF-G, GTP-binding
fabH 1.1 3-oxoacyl-ACP synthase III
moaE 1.1 molybdopterin synthase, large subunit
ynaI 1.1 conserved inner membrane protein
dam 1.1 DNA adenine methylase
speE 1.1 spermidine synthase (putrescine aminopropyltransferase)
rplV 1.1 50S ribosomal subunit protein L22
malZ 1.1 maltodextrin glucosidase
yqgE 1.1 predicted protein
rpsK 1.1 30S ribosomal subunit protein S11
marA 1.1 DNA-binding transcriptional dual activator of multiple
antibiotic resistance
glyS 1.1 glycine tRNA synthetase, beta subunit
panC 1.1 pantothenate synthetase
menE 1.1 o-succinylbenzoate-CoA ligase
yegD 1.1 predicted chaperone
yfeX 1.1 conserved protein
lon 1.1 DNA-binding ATP-dependent protease La
ptsP 1.1
fused PTS enzyme: PEP-protein phosphotransferase (enzyme
I)/GAF domain containing protein
Genes activated at least 2-fold in ΔargP Appendix II
188
Gene
name
Fold
difference Description
dnaE 1.1 DNA polymerase III alpha subunit
speB 1.1 agmatinase
fabI 1.1 enoyl-ACP reductase
yfgD 1.1 predicted oxidoreductase
apbE 1.1 predicted thiamine biosynthesis lipoprotein
trpB 1.1 tryptophan synthase, beta subunit
gadW 1.1 DNA-binding transcriptional activator
psiE 1 predicted phosphate starvation inducible protein
yqaB 1 predicted hydrolase
yceA 1 conserved protein
secE 1 preprotein translocase membrane subunit
yqgF 1 predicted Holliday junction resolvase
rplL 1 50S ribosomal subunit protein L7/L12
gcvT 1 Amino methyl transferase, tetrahydrofolate-dependent, subunit
(T protein) of glycine cleavage complex
alsE 1 allulose-6-phosphate 3-epimerase
cdsA 1 CDP-diglyceride synthase
hcaF 1 3-phenylpropionate dioxygenase, small (beta) subunit
cho 1 endonuclease of nucleotide excision repair
yjiH 1 conserved inner membrane protein
nac 1 DNA-binding transcriptional dual regulator of nitrogen
assimilation
yhhQ 1 conserved inner membrane protein
hslJ 1 heat-inducible protein
ybaB 1 conserved protein
mrp 1 antiporter inner membrane protein
yfhG 1 conserved protein
yfcZ 1 conserved protein
yheU 1 conserved protein
Genes activated at least 2-fold in ΔargP Appendix II
189
Gene
name
Fold
difference Description
bglF 1 fused beta-glucoside-specific PTS enzymes: IIA
component/IIB component/IIC component
ilvE 1 branched-chain amino-acid aminotransferase
ycbL 1 predicted metal-binding enzyme
pepE 1 (alpha)-aspartyl dipeptidase
osmF 1 predicted transporter subunit: periplasmic-binding component
of ABC superfamily
rpsT 1 30S ribosomal subunit protein S20
ygbE 1 conserved inner membrane protein
yfgC 1 predicted peptidase
rbn 1 binuclear zinc phosphodiesterase
rpsC 1 30S ribosomal subunit protein S3
tyrR 1 DNA-binding transcriptional dual regulator, tyrosine-binding
xthA 1 exonuclease III
yraM 1 conserved protein
ynjE 1 predicted thiosulfate sulfur transferase
yccS 1 predicted inner membrane protein
gpmM 1 phosphoglycero mutase III, cofactor-independent
yeaZ 1 predicted peptidase
emtA 1 lytic murein endotransglycosylase E
uof 1 ryhB-regulated fur leader peptide
grxD 1 conserved protein
rpsM 1 30S ribosomal subunit protein S13
hycA 1 regulator of the transcriptional regulator FhlA
cvpA 1 membrane protein required for colicin V production
rlpA 1 minor lipoprotein
cobS 1 cobalamin 5'-phosphate synthase
poxB 1 pyruvate dehydrogenase (pyruvate oxidase), thiamin-
dependent, FAD-binding
fabZ 1 (3R)-hydroxymyristol acyl carrier protein dehydratase
Genes activated at least 2-fold in ΔargP Appendix II
190
Gene
name
Fold
difference Description
atpB 1 F0 sector of membrane-bound ATP synthase
adhP 1 ethanol-active dehydrogenase/acetaldehyde-active reductase
trpT 1 Trp tRNA
dusA 1 tRNA-dihydrouridine synthase A
ogt 1 O-6-alkylguanine-DNA:cysteine-protein methyltransferase
dppF 1 dipeptide transporter
creB 1 DNA-binding response regulator in two-component regulatory
system with CreC
paaY 1 predicted hexapeptide repeat acetyltransferase
ybjL 1 predicted transporter
zwf 1 glucose-6-phosphate dehydrogenase
rlmB 1 23S rRNA (Gm2251)-methyltransferase
sgrR 1 DNA-binding transcriptional regulator
yibN 1 predicted rhodanese-related sulfurtransferase
pmbA 1 predicted peptidase required for the maturation and secretion
of the antibiotic peptide MccB17
mnmG 1 glucose-inhibited cell-division protein
ruvB 1 ATP-dependent DNA helicase, component of RuvABC
resolvasome
rplJ 1 50S ribosomal subunit protein L10
aroB 1 3-dehydroquinate synthase
yliG 1 predicted SAM-dependent methyltransferase
yfcC 1 predicted inner membrane protein
yigZ 1 predicted elongation factor
bioB 1 biotin synthase
yliJ 1 predicted glutathione S-transferase
cmoB 1 predicted S-adenosyl-L-methionine-dependent
methyltransferase
yecM 1 predicted metal-binding enzyme
tatA 1 TatABCE protein translocation system subunit
Genes activated at least 2-fold in ΔargP Appendix II
191
Gene
name
Fold
difference Description
menD 1 bifunctional 2-oxoglutarate decarboxylase/ SHCHC synthase
yfiF 1 predicted methyltransferase
gsk 1 inosine/guanosine kinase
hisF 1 imidazole glycerol phosphate synthase, catalytic subunit with
HisH
otsB 1 trehalose-6-phosphate phosphatase, biosynthetic
aroD 1 3-dehydroquinate dehydratase
ybjO 1 predicted inner membrane protein
lpcA 1 D-sedoheptulose 7-phosphate isomerase
nagA 1 N-acetylglucosamine-6-phosphate deacetylase
dppC 1 dipeptide transporter
yrbB 1 predicted protein
serB 1 3-phosphoserine phosphatase
dnaQ 1 DNA polymerase III epsilon subunit
ygiC 1 predicted enzyme
yfgL 1 protein assembly complex, lipoprotein component
ccmE 1 periplasmic heme chaperone
ycbK 1 conserved protein
yhgF 1 predicted transcriptional accessory protein
uvrY 1
DNA-binding response regulator in two-component regulatory
system with
BarA
yafJ 1 predicted amidotransfease
sapD 1 predicted antimicrobial peptide transporter subunit
wbbI 1 conserved protein
lptA 1 predicted transporter subunit: periplasmic-binding component
of ABC superfamily
topB 1 DNA topoisomerase III
rsxE 1 predicted inner membrane NADH-quinone reductase
cusR 1 DNA-binding response regulator in two-component regulatory
Genes activated at least 2-fold in ΔargP Appendix II
192
system with CusS
Gene
name
Fold
difference Description
kefA 1 fused conserved protein
hemL 1 glutamate-1-semialdehyde aminotransferase (aminomutase)
emrY 1 predicted multidrug efflux system
yffS 1 CPZ-55 prophage; predicted protein
ycgV 1 predicted adhesin
tsx 1 nucleoside channel, receptor of phage T6 and colicin K
Recommended