Microbiology Papers in Press. Published February 8, 2015 as doi:10.1099/mic.0.000052
Microbiology A single amino acid substitution in the Ω-like loop of E. coli PBP5 disrupts its ability to maintain cell-shape and intrinsic beta-lactam resistance --Manuscript Draft-Manuscript Number:
MIC-D-14-00066R2
Full Title:
A single amino acid substitution in the Ω-like loop of E. coli PBP5 disrupts its ability to maintain cell-shape and intrinsic beta-lactam resistance
Short Title:
Involvement of Leu182 in the physiology of E. coli PBP5
Article Type:
Standard
Section/Category:
Physiology and metabolism
Corresponding Author:
Anindya S Ghosh Indian Institute of Technology Kharagpur Kharagpur, WB INDIA
First Author:
Mouparna Dutta
Order of Authors:
Mouparna Dutta Debasish Kar, M. Tech. Ankita Bansal Sandeep Chakraborty Anindya S Ghosh
Abstract:
PBP5, a DD-carboxypeptidase, maintains cell shape and intrinsic beta-lactam resistance in E. coli. A strain lacking PBP5 loses intrinsic beta-lactam resistance and simultaneous lack of two other PBPs results in aberrantly shaped cells. PBP5 expression in trans complements the shape and restores the lost beta-lactam resistance. PBP5 has a 'Ω loop' like region similar to that in Class A beta-lactamases. It was previously predicted that Leu182 present in the 'Ω -like' loop of PBP5 corresponds to Glu166 in PER-1 beta-lactamase. Here, we studied the physiological and biochemical effects of the Leu182Glu mutation in PBP5. Upon the over-expression in septuple PBP mutants, ~75% cells were abnormally shaped and intrinsic beta-lactam resistance maintenance was partially lost. Biochemically, the purified soluble mutated PBP5 (smPBP5) retained low acylation ability for penicillin. The turnover number of smPBP5 for artificial and peptidoglycan mimetic substrates were ~10 times lesser than that of wild type. Superimposition of the active-site residues of smPBP5 on PBP5 revealed that perturbation in the orientating key residues is a possible reason for the low turnover numbers. Therefore, we establish the involvement of Leu182 in maintaining the physiological and biochemical behavior of E. coli PBP5.
Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation
Manuscript_MIC-D-14-00066R2 Click here to download Manuscript Including References (Word document): Revised_Manuscript_#MIC-D-14-00066-R2.docx
1
Title:
2
A single amino acid substitution in the Ω-like loop of E. coli PBP5 disrupts its ability to
3
maintain cell-shape and intrinsic beta-lactam resistance
4 5
Short Title: Involvement of Leu182 in the physiology of E. coli PBP5
6 7
Authors: Mouparna Dutta, Debasish Kar, Ankita Bansal, Sandeep Chakraborty1,2, Anindya S.
8
Ghosh*
9 10
Affiliation: Department of Biotechnology, Indian Institute of Technology Kharagpur,
11
Kharagpur-721302, India
12
1
Plant Sciences Department, University of California, Davis, 95616, USA
13
2
Department of Biological Sciences, Tata Institute of Fundamental Research, Mumbai, India
14 15 16
*Corresponding Author:
17
Anindya S. Ghosh, E-mail:
[email protected]; Tel: +91-3222-283798
18 19
(i) Summary: 191 words
20
(ii) Main text: 4165 words
21
Tables: 04
22
Figures: 04
23
Supplementary Material: 01 (contains 02 figures and 01 table)
24 25 26
Summary:
27
PBP5, a DD-carboxypeptidase, maintains cell shape and intrinsic beta-lactam resistance
28
in E. coli. A strain lacking PBP5 loses intrinsic beta-lactam resistance and simultaneous lack of
29
two other PBPs results in aberrantly shaped cells. PBP5 expression in trans complements the
30
shape and restores the lost beta-lactam resistance. PBP5 has a ‘Ω loop’ like region similar to that
31
in Class A beta-lactamases. It was previously predicted that Leu182 present in the ‘Ω –like’ loop
32
of PBP5 corresponds to Glu166 in PER-1 beta-lactamase. Here, we studied the physiological and
33
biochemical effects of the Leu182Glu mutation in PBP5. Upon the over-expression in septuple
34
PBP mutants, ~75% cells were abnormally shaped and intrinsic beta-lactam resistance
35
maintenance was partially lost. Biochemically, the purified soluble mutated PBP5 (smPBP5)
36
retained low acylation ability for penicillin. The turnover number of smPBP5 for artificial and
37
peptidoglycan mimetic substrates were ~10 times lesser than that of wild type. Superimposition
38
of the active-site residues of smPBP5 on PBP5 revealed that perturbation in the orientating key
39
residues is a possible reason for the low turnover numbers.
40
involvement of Leu182 in maintaining the physiological and biochemical behavior of E. coli
41
PBP5.
Therefore, we establish the
42 43
Keywords: E. coli; penicillin-binding protein 5; omega loop; cell-shape; site-directed
44
mutagenesis; DD-CPase
45 46
47 48 49
1.0. Introduction
50
One of the main bacterial cell wall components is the peptidoglycan layer, which is
51
indispensible for cellular viability and maintenance of cellular morphology. Penicillin-binding
52
proteins (PBPs) are the enzymes required for the cross-linking of the glycan chain
53
(transglycosylase) and peptide chain (transpeptidase). Majority of these proteins have three
54
signature motifs, namely, SXXK, SXN and KTG, and these motifs are responsible for their
55
respective enzymatic activities (Ghuysen, 1991). In E. coli, five PBPs (PBP4, 5, 6, DacD and
56
AmpH) are documented as DD-Carboxypeptidase (DD-CPase) (Korat et al., 1991; Holtje, 1998;
57
Nelson and Young 200 and 2001; Baquero et al., 1996 and Gonzalez-Leiza et al., 2011). These
58
PBPs are thought to prevent the unwanted crosslink formation by removing the terminal D-
59
alanine residue from the pentapeptide side-chain of N-acetyl muramic acid (Holtje, 1998 and
60
Ghosh et al., 2008). PBP5 is the key DD-CPase that helps maintain a uniform cell-shape in E.
61
coli (Nelson et al, 2000, 2001 and Ghosh et al., 2008). E. coli mutants lacking at least three DD-
62
CPases including PBP5 show aberrant morphology. However, the abnormalities are abolished
63
upon ectopic expression of PBP5 (Nelson and Young, 2000; Ghosh and Young, 2003). The
64
ability of PBP5 to repair the cellular oddities depends on a stretch of 20 amino acids (200-219)
65
around the KTG motif of PBP5, known as the ‘morphology maintenance domain’ (MMD)
66
(Ghosh and Young, 2003). Apart from maintaining the cell shape, PBP5 also plays an important
67
role in maintaining the intrinsic beta-lactam resistance in E. coli (Sarkar et al., 2010). The
68
absence of PBP5 sensitizes the E. coli cell to beta-lactam antibiotics, while in trans expression of
69
the same helps to reverse the lost resistance (Sarkar et al., 2010). It is speculated that the high
70
copy number of PBP5 at the logarithmic phase of growth may be responsible for trapping the
71
beta-lactam antibiotics, which prevents them from binding to the essential PBPs (Sarkar et al.,
72
2010; Sarkar et al., 2011).
73
Several mutation analyses have been studied for assessing the activity of PBP5 in vitro.
74
The PBP5 ‘74-90 loop’ deletion mutant, Gly105Asp mutant and the modified Cys115 resulted in
75
the loss of DD-CPase activity individually due to inefficient deacylation (Nicholas et al., 2003;
76
Nicola et al., 2005). The particular defects in deacylation reveal the intimate role of SXN motif
77
especially Ser110 in the catalytic mechanism (Nicola et al., 2005). In addition to the active-site
78
residues, Asp175 also serves as a determinant of in vitro DD-CPase activity, but not for the
79
penicillin-binding activity (van der Linden et al., 1994).
80
It is speculated that PBP5 may have beta-lactamase like activity as it shares a common
81
ancestry with beta-lactamases (Zhang et al., 2007). Nevertheless, PBP5 in its soluble or in
82
membrane-bound form does not show beta-lactamase activity under physiological pH condition
83
(Sarkar et al., 2010), thoughPBP5 has a ‘Ω-loop’ like region (a characteristic for the beta-
84
lactamases) similar to that of TEM-1 beta-lactamases (Zhang et al., 2007; Ghosh et al., 2008).
85
The orientation of His151 in this loop is similar to that of Glu166 in TEM-1, which might help in
86
accessing water molecule during deacylation of beta-lactams by PBP5 (Davies et al., 2001;
87
Ghosh et al., 2008). It has been reported that the soluble form of PBP-A of
88
Thermosynechococcus elongatus possesses 50-fold increase in penicillin hydrolysis upon L158E
89
mutation at the ‘Ω-loop’ like region, but its physiological effect is yet unknown (Urbach et al.,
90
2008). Recently, based on in silico predictions, it is hypothesized that the Leu153 of ‘Ω-loop’
91
like region in E. coli PBP5 could be the best candidate for mimicking Glu166 residue of Class A
92
beta-lactamases (Chakraborty, 2012).
93
In the present study, we investigated whether the alteration in the 'Ω-loop' could affect the
94
physiological function of PBP5. Accordingly, the leucine present in the 'Ω-loop’ of both
95
membrane-bound (at 182nd position) and the soluble (at 153rd position) constructs of E. coli
96
PBP5 are replaced with glutamic acid residue. Physiological functions of the mutant in terms of
97
cell shape and intrinsic beta-lactam resistance are assessed. Subsequently, the alterations in
98
functions of the mutant are verified through in vitro biochemical analyses and finally the possible
99
reason for the altered functions are explained by using in silico molecular modeling and
100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115
superimposition studies.
116
2.0. Materials and Methods
117
2.1. Bacterial strains, reagents and growth conditions
118
E. coli K12 (CS109) strain and its PBP deletion mutants, CS703-1 (∆PBP1, 4, 5, 6, 7,
119
AmpC and AmpH) and AM15-1(∆PBP5) were used for this study (Nelson et al., 2002 and
120
Sarkar et al., 2010) (Table. 1). The strains were grown in Luria-Bertani (LB) broth and Muller-
121
Hinton (MH) broth (HiMedia, Mumbai, India). Chloramphenicol (20 µg ml-1) or kanamycin (50
122
µg ml-1) was added wherever necessary. Unless otherwise specified all the chemical and reagents
123
were purchased from Sigma-Aldrich (St. Louis, MO, USA).
124 125
2.2. Construction of the clones for full-length PBP5L182E and soluble form of the protein
126
(smPBP5)
127
The clone for PBP5L182E was created on the membrane-bound form and smPBP5 were
128
created on soluble form of PBP5 by using commercially available site-directed mutagenesis kit
129
(Stratagene Co., La Jolla, Calif.) to generate the plasmids pPJ5L182E and pTS5LE, respectively
130
(Table. 1). The plasmid, pPJ5 was used as template to make pPJ5L182E, while pT7cPBP5 was
131
used as template to make pTS5LE. Same set of oligonucleotide primer pairs were used for each
132
reaction.
133
ctgaccatcagcatcctcaccatgtaccgtctgg-3'. The gene for both sPBP5 and smPBP5 were sub-cloned
134
from pT7-7K to pET28a (+) at the sites of NdeI and HindIII to generate pET5 and pET5LE,
135
respectively (Table. 1). Finally, the mutations in the respective clones were confirmed by
136
sequencing through commercial sequencing service (MWG Biotech Inc, Bangalore, India).
Those
were
137 138
2.3. Assessment of Cell shape
5'-ccagacggtacatggtgaggatgctgatggtcag-3'
and
5'-
139
The E. coli strain CS703-1 containing pPJ5L182E was expressed by inducing with
140
0.005% arabinose at the early log phase as previously described (Nelson and Young, 2001).
141
Cells were then incubated for 2-4 h and treated with aztreonam at the final concentration of 5 µg
142
ml-1 (Nilsen et al, 2004). Further, cells were washed with 1X PBS (pH 7.4) and cell morphology
143
was checked by fixing the cells onto poly-lysine coated glass slides and subsequently visualized
144
the live cells under the OLYMPUS IX 71 phase-contrast microscope (Olympus Inc., Japan).
145 146
2.4. Determination of Minimum Inhibitory Concentration (MIC)
147
MIC values were determined by the microtiter plate dilution method as described earlier
148
(Sarkar et al., 2011). Total volume of MH broth in each well was made up to 300 µl by adding
149
overnight grown culture of the respective E. coli strains at a concentration ~105 cells per well.
150
The antibiotics were tested by two-fold step dilution (Sarkar et al., 2010) and finally, the
151
microtiter plates were incubated at 37º C for 18 h. Bacterial growth was monitored by measuring
152
the optical density at 600 nm using a Multiskan Spectrum spectrophotometer (model 1500;
153
Thermo Scientific, Nyon, Switzerland). MIC values were tested in triplicate and repeated at least
154
six times for accuracy.
155 156
2.5. Expression and purification of the soluble form of wild type PBP5 (sPBP5) and its mutant
157
(smPBP5)
158
The expression of sPBP5 and smPBP5 were carried out in E. coli BL21 star under the
159
control of T7 promoter. The bulk expression was obtained in cytoplasmic fraction by growing
160
the cells at 30º C for 8 h incubation in the presence of 0.05 mM isopropyl-β-
161
thiogalactopyranoside (IPTG). The N-terminal His-tag containing sPBP5 and smPBP5 were
162
purified using Ni-NTA affinity chromatography. The proteins were eluted from the affinity
163
column in the elution buffers containing 50 mM and 100 mM of imidazole in a buffer (10 mM
164
Tris-HCl, 300 mM NaCl buffer, pH 8.0) at 4° C. The eluted fractions were subjected to 12%
165
SDS-PAGE to determine their molecular weights and their activity was assessed prior to
166
collection. The fractions collected were dialyzed for 12 h with three changes of same buffer
167
without imidazole and was further concentrated by using Amicon Ultra-4 Devices (Millipore,
168
Billerica, MA).
169 170
2.6. Fluorescent penicillin binding assay
171
The stability and activity of the PBPs were checked by labeling with Bocillin-FL
172
(Invitrogen, Carlsbad, CA), a fluorescent penicillin. The membrane bound PBPs (~300 µg) and
173
the soluble PBPs (~25 µg) were incubated at 35º C for 30 min with 50 µM and 100 µM of
174
Bocillin-FL, respectively (Zhao et al., 1999; Chowdhury et al., 2010). Further, the proteins were
175
analyzed through 12% SDS-PAGE and visualized in Typhoon FLA 7000 (GE Healthcare Bio-
176
Sciences AB, Uppsala, Sweden).
177 178
2.7. Kinetics analysis of smPBP5 with penta-peptide substrate
179
The DD-CPase activity exerted by both sPBP5 and smPBP5 were checked for the
180
artificial substrate, Nα,Nє-diacetyl-Lys-D-Ala-D-Ala (AcLAA) and peptidoglycan mimetic
181
pentapeptide substrate, L-Ala-γ-D-Glu-L-Lys-D-Ala-D-Ala (AGLAA) (USV custom peptide
182
synthesis, Mumbai, India) in 10 mM Tris-HCl buffer (pH 8.0) (Chowdhury et al.,2010). The
183
reactions were carried out in 60 µl for each soluble PBPs where 2 mg of the each proteins was
184
mixed with the respective peptides at a concentration range of 0.25–12 mM. The remaining
185
volume was adjusted with 50 mM Tris-HCl, pH 8.5. The reaction mixture was incubated at 37
186
°C for 30 min and after which freshly prepared enzyme–coenzyme mixture (140 µl) was added.
187
The enzyme-co-enzyme mixture was prepared by mixing 50 mM Tris-HCl, pH 8.5; 0.3 mg ml-1
188
FAD; 10 mg ml-1 horseradish peroxidase; and 5 mg ml-1 D-amino acid oxidase at a ration 20: 10:
189
5: 1). This resultant mixture was incubated at 37 °C for 5 min. Free D-alanine generated in this
190
reaction was detected using the method described by Frere et al. (1976), and compared with a
191
standard D-alanine solution using a Multiskan Spectrum-1500 spectrophotometer (Thermo
192
Scientific, Nyon-1, Switzerland) set at 460 nm. (Frere et al., 1976). The kinetic parameters for
193
this assay were analyzed from the linear regression of the double reciprocal plot (Lineweaver and
194
Burk, 1934).
195
.
196
2.8. Kinetic study of the interaction of soluble enzymes with penicillin
197
The acylation rate k2/K (i.e. the rate of formation of acyl-enzyme complex with sub-
198
saturating concentration of the substrate) was obtained from time course study of enzyme-
199
substrate complex formation with Bocillin-FL (Chambers et al., 1994; Chowdhury et al., 2010).
200
The sPBP5 and smPBP5 were incubated with different concentration of Bocillin-FL (substrate)
201
for 30, 60, 90 and 120s. To stop the reaction, denaturing buffer was added and samples were
202
analyzed in 12% SDS-PAGE. The labeled proteins were quantified by Typhoon FLA 7000 (GE
203
Healthcare Bio-Sciences AB, Uppsala, Sweden).
204 205
The deacylation rate constant (k3) of soluble enzymes were determined by plotting the
206
value of remaining Bocillin-FL versus time (Di Guilmi et al., 2000; Chowdhury et al., 2010). In
207
brief, the enzymes were incubated with Bocillin-FL (50 µM) at 35º C for 30 min. Penicillin G (3
208
mM) was added at t=0 and fluorescent intensity was measured by removing the aliquots time to
209
time and subsequently denaturing the protein.
210 211
2.9. Assay of beta-lactamase activity of soluble enzymes
212
The beta-lactam hydrolysis of sPBP5 and smPBP5 were determined with different groups
213
of beta-lactam antibiotics in the presence of 10 mM phosphate buffer (pH 7.4). The purified
214
proteins were incubated with the beta-lactam antibiotics at different concentrations and the
215
hydrolysis was monitored at respective wavelengths by using BioSpectrometer kinetic
216
(Eppendorf, Hamburg, Germany). The molar extinction coefficients for turnover of penicillin
217
and nitrocefin were ∆ε232 = -800 M-1cm-1 and ∆ε482 = 15,000 M-1cm-1 respectively (Urbach et al.,
218
2008).
219 220
2.10. In silico analysis of smPBP5
221
A constrained based homology program MODELLER 9v12 was adopted to build a 3D
222
model of the smPBP5 (Sali and Blundell, 1993). For model building, a PDB coordinate, 1NZO
223
chain A [The crystal structure of wild type PBP5 from E. coli, (Nicholas et al., 2003)] was used
224
as a template. The final model was validated through PROCHECK (Laskowski et al., 1993) and
225
Verify-3D (Luthy et al., 1992). To distinguish structural organization of the active-site
226
orientation, the active site of smPBP5 was superimposed to that of sPBP5 through Swiss PDB
227
deep view program and the result was analyzed in PyMol visualization tool. The active-site
228
groove volume of the enzyme was measured through CASTp calculation (Dundas et al., 2006).
229 230
2.11. Secondary structure determination of soluble proteins
231
To compare the overall folding of sPBP5 and smPBP5, Circular Dichroism (CD)
232
spectroscopy was done in the "far-UV" spectral region (200-240 nm) for determination of the
233
structural distribution of α -helix, β-sheet and random coil in each protein. Spectral data of the
234
samples were collected at different temperatures (i.e. 28, 35 and 40° C) with a 0.2 nm step
235
resolution, 1 s time constant and 10 milli degrees sensitivity at a 2.0 mm spectral bandwidth,
236
with a scanning speed of 50 nm min_1. Before, spectral data collection, all the soluble proteins
237
were diluted to same concentration (0.1 mg ml-1). Corrected spectra were obtained by subtracting
238
the solvent spectrum. To reduce noise and random instrumental error, an average spectrum was
239
compiled from four successive accumulations over a range of 200–240 nm. The percentage of α-
240
helix, β-sheet and random coil were calculated using program k2d3 software (Louis-Jeune et al.,
241
2012). This method compares the secondary structure of unknown protein with known data set
242
and analyzes the percentage of secondary structure of the sample protein.
243 244
Simultaneously, the secondary structure of sPBP5 (1NZO) and smPBP5 were also
245
verified by using two independent algorithms, PREDICT PROTEIN (Rost et al., 2004) and
246
PSIPRED (Jones, 1999).
247 248 249 250 251 252 253
254 255
3.0. Results
256
3.1. Site-directed mutagenesis at the ‘Ω type loop’ affects the cell shape maintaining ability of
257
PBP5
258
The strain CS703-1, derived from E. coli (CS109), lacks seven non-essential PBPs, and is
259
morphologically deformed (Nelson and Young, 2001; Ghosh et al., 2006). In trans expression of
260
PBP5 in this deletion mutant restores cell shape to near normal, i.e. rod shaped (Ghosh and
261
Young 2003; Nelson and Young, 2000). To check whether the alteration in the ‘Ω type loop’ of
262
PBP5, i.e. the Leu182Glu mutation, could affect its ability to revert the abnormal cell shape of
263
CS703-1, the deformities of CS703-1 cells were checked upon ectopic expression of both PBP5
264
and PBP5L182E. Surprisingly, after complementing with PBP5 and its L182E mutant
265
individually in CS703-1, the levels of abnormalities in the cell shape were reduced to about 16%
266
and 75% respectively, as compared to the strain without any complementation (Fig. 1; Table.
267
S1). As the genes encoding PBP5 and PBP5L182E were cloned in pBAD18, the strain CS703-1
268
with pBAD18 was taken as a control.
269 270
To compare the cellular expression level and stability of PBP5L182E with that of wild
271
type PBP5, both PBP5 and PBP5L182E were expressed under the control of arabinose promoter
272
and isolated from the membrane of CS703-1 after 12 hrs of incubation at 37°C. It is noteworthy
273
to mention that the expression level and the stability of the proteins could have been checked by
274
immune-blotting technique. However, the similar procedure was not employed here because the
275
antibodies developed against PBP5 may also cross-react with DacD, which was present in the
276
CS703-1 cells, and it shares considerable identity with PBP5 (Potluri et al., 2010). Moreover, for
277
assessing the stability of the PBP5 and its mutant immune-blotting might not be the ideal choice,
278
because the PBP5 antibodies, in addition to binding to the active protein, can also bind to its
279
inactive variant. Therefore, it would be more meaningful to assess the in vivo expression for a
280
considerable interval of time and the stability of these proteins by labeling them subsequently
281
with a substrate like, Bocillin-FL. Upon labeling the equal amount of each protein with Bocillin-
282
FL (Fig. 2a), the similar intensities of the bound Bocillin-FL were found associated to PBP5 and
283
PBP5L182E indicating that both the PBPs were active to the same extent during their expression
284
in CS703-1. The overall results indicate that the mutation L182E in PBP5 almost aborted its
285
ability to restore the normal rod shape in E. coli CS703-1 strain.
286 287
3.2. The PBP5L182E can partially maintain the intrinsic beta-lactam resistance in E. coli
288
It is known that the deletion of PBP5 sensitizes E. coli cells to beta-lactam antibiotics.
289
Ectopic complementation of PBP5 restores the lost beta-lactam resistance of AM15-1 (PBP5
290
deletion mutant of CS109, an E. coli K-12 strain), since PBP5 possibly traps beta-lactams
291
(Sarkar et al., 2010). To check whether the Leu182Glu mutation can alter the intrinsic beta-
292
lactam resistance property of PBP5, the beta-lactam sensitivity of AM15-1 was tested by
293
expressing the PBP5L182E ectopically, using 0.0005% arabinose. It was observed that, unlike
294
the wild type, PBP5L182E was not able to restore the lost beta-lactam resistance of AM15-1
295
completely (Table. 2). The result signifies that PBP5L182E partially lost its ability to maintain
296
intrinsic beta-lactam resistance in E. coli. It can be speculated that PBP5 might have lost its
297
ability to exert efficient DD-CPase activity in vivo, which is required for maintaining cell shape.
298
It also retains its ability to bind beta-lactams, although the binding ability to beta-lactams may
299
have been reduced.
300 301
3.3. The smPBP5 exhibits poor DD-CPase activity
302
To know whether PBP5L182E had retained the ability to exert DD-carboxypeptidase
303
activity it was mandatory to analyze the biochemical nature of PBP5L182E in vitro and to
304
compare the same with the wild type PBP5. In order to verify the biochemical function of
305
PBP5L182E, the soluble form of it (smPBP5) was generated. Though PBP5 is a membrane
306
bound protein, for ease of purification, both the genes encoding sPBP5 and smPBP5 were cloned
307
in pET28a(+)and finally the proteins were purified by Ni-NTA affinity chromatography. The
308
concentrations of the proteins produced were ~1 mg ml-1. The molecular mass of the purified
309
proteins were ~44.0 kDa and both the enzymes were active as confirmed by labeling them with
310
Bocillin-FL and subsequent visualization in 12% SDS-PAGE (Fig. 2). The sPBP5 and smPBP5
311
were checked for its DD-CPase activity. Both the artificial peptidoglycan substrate (Nα,Nє-
312
diacetyl-Lys-D-Ala-D-Ala or, AcLAA) and a peptidoglycan mimetic pentapeptide substrate (L-
313
Ala-γ-D-Glu-L-Lys-D-Ala-D-Ala or, AGLAA) were used (Chowdhury et al., 2012) as substrates
314
for the assay. The turnover numbers for AcLAA and AGLAA, i.e., kcat value of smPBP5
315
indicated ~10 fold lower product formation by smPBP5 as compared to sPBP5 (Table. 3a).
316
Therefore, the result indicates that indeed smPBP5 has lost the ability to exert in vitro DD-CPase
317
activity for both AcLAA and AGLAA substrates as compared to sPBP5, which is consistent with
318
the result obtained from in vivo study. Hence, we can infer that mutation of L182E not only
319
affects in vivo physiology of PBP5, but also the same is reflected in vitro.
320 321
3.4. Penicillin acylates smPBP5 at a rate that is little lesser as compared to sPBP5
322
To check the penicillin-binding efficacies of sPBP5 and the smPBP5, kinetic parameters
323
for acylation and deacylation were determined with fluorescent penicillin, Bocillin-FL by
324
calculating acylation and deacylation rate constants. The resultant rate of acylation (k2/K) of
325
smPBP5 for Bocillin was ~40% lower than that of sPBP5 (Table. 3b) indicating that the L153E
326
mutation affected the penicillin-binding efficiency of smPBP5. Further, the rate of dissociation
327
of penicillin from these enzymes, as determined by calculating the deacylation rate constant (k3),
328
showed ~10% lower rate in smPBP5 than sPBP5 (Table. 3b). Therefore, it can be stated that
329
L153E mutation has reduced the rate of penicillin-binding to sPBP5 by 40% without
330
significantly affecting its dissociation.
331 332
3.5. The L153E mutation doesn’t alter the beta-lactamase activity of sPBP5 in vitro
333
The glutamic acid at the Ω-loop of class A beta-lactamase is believed to provide water
334
molecule for the hydrolysis of beta-lactam antibiotics (Banerjee et al., 1998 and Sun et al.,
335
2003). To check whether the replacement of Glu at Leu153 in the purified soluble construct of
336
PBP5 could change in beta-lactamase activity, the hydrolysis of both penicillin and
337
cephalosporin group antibiotics were studied. However, there was no substantial difference in the
338
hydrolysis pattern of penicillin and nitrocefin by smPBP5 as compared sPBP5 (data not shown).
339 340
3.6. Replacement of Leu153 to Glu disrupt the association of the key residues in the sPBP5
341
active-site
342
To understand the reason behind the lower DD-CPase activity of smPBP5, both
343
biophysical and in silico studies were conducted. The structural model of the same was built to
344
compare with the crystal structure of wild type PBP5. The similarities in the secondary structure
345
analyses of the soluble proteins by both CD and in silico prediction (Fig. S1) revealed that there
346
was no significant change in overall folding of smPBP5 (Table. 4). In addition, both the proteins
347
(sPBP5 and smPBP5), were apparently stable within a physiological temperature range of 28 to
348
40 °C as revealed by the thermal stability assessment through CD (Fig. S2). However, the
349
superimposition study of the residues present in catalytic-sites exhibited a drastic deviation of the
350
γ-OH of both Ser44 and Ser110 and ε-NH2 of Lys47 residue (Fig. 3). In smPBP5, both γ-OH of
351
Ser44 and ε-NH2 of Lys47 are moved away from each other by 1.90 Ǻ and 2.18 Ǻ, respectively.
352
Thereby, the distance between Ser44 and Lys47 has been increased (Fig. 3) that prevents the H-
353
bond formation. In addition, the γ-OH of Ser110 has also deviated by 2.14 Ǻ creating an inter-
354
atomic distance of 4.2 Ǻ between Ser110 and Lys213 in smPBP5 whereas it is 2.7 Ǻ in case of
355
sPBP5 (Fig. 3). As a result, the higher distance between Ser110 and Lys213 might affect the H-
356
bond formation in smPBP5. This may interfere in the deacylation of peptidoglycan mimetic
357
substrates while exerting DD-CPase activity (Sauvage et al., 2008), though the deacylation of
358
Bocillin is not much affected (Table. 3b). Due to the adverse changes in the organization of the
359
key residues in smPBP5, the groove volume of the active-site has been decreased by almost 200
360
Ǻ3, which is the most likely reason for its lower ability to catalyse peptidoglycan mimetic
361
substrates (Fig. 4).
362 363 364 365 366 367
368 369 370 371
4.0. Discussion:
372
4.1. The possible reason for the functional limitation of L182E mutant of PBP5
373
The expression of PBP5 offers sufficient DD-CPase activity to preserve the consistency
374
in the cell morphology (Nelson and Young, 2001) and to maintain the intrinsic beta-lactam
375
resistance property in E. coli (Sarkar et al., 2011). The partial inability of PBP5L182E to reverse
376
the cell shape oddities of CS703-1 to normal suggests impairment in its DD-CPase activity (Fig.
377
1). To verify the physiological reason of the particular mutation, biochemical activity of smPBP5
378
has been examined, where it has shown a ~10 fold lower DD-CPase activity with the
379
peptidoglycan mimetic pentapeptide substrate than sPBP5 (Table. 3a). Therefore, it seems
380
obvious that L182E mutation of PBP5 drastically affect the DD-CPase activity, both in vivo and
381
in vitro.
382 383
Additionally, we find that PBP5 with L182E mutation lacks the ability to reverse the lost
384
intrinsic beta-lactam resistance in ∆PBP5 mutant of E. coli (Table. 2). It has been speculated
385
earlier that the high copy number PBP5 traps beta-lactam antibiotics to shield over other
386
essential PBPs (Sarkar et al., 2010). So, if the rate of acylation of PBP5 were lowered towards
387
beta-lactam substrates, the ability of PBP5 to act as traps would supposedly diminish. Indeed,
388
the rate of acylation for smPBP is ~40% lower for the penicillin substrate than sPBP5 (Table.
389
3b). Therefore, such reduction in the acylation rate might be the contributor for the partial
390
inability of PBP5L182E to reverse intrinsic beta-lactam resistance.
391 392
Definitely, the speculative insight into the mechanism behind the inefficiency of smPBP5
393
to exert low DD-CPase activity, especially towards peptidoglycan mimetic substrate, lies into the
394
packing of the active site. The distance between γ-OH of Ser44 and ε-NH2 of Lys47 is increased
395
in smPBP5, which may result in an unsuccessful interaction and possibly weakens the
396
nucleophilicity of the active-site Ser44 (Davies et al., 2001 and Stefanova et al., 2002). Thus,
397
greater distance may also hinder the acylation rate of the peptidoglycan mimetic substrates
398
(Table. 3a). Moreover, Ser110, which helps in hydrolyzing peptide substrate (s) by polarizing
399
water molecules with the help of Lys213 (Nicola et al., 2005), also undergoes deviation in
400
smPBP5 and thereby raising the distance between γ-OH of Ser110 and ε-NH2 of Lys213. Such
401
alteration in distance might hamper the stability of the acyl-enzyme complex during deacylation
402
(Malhotra and Nicholas, 1992). Furthermore, as there is a reduction in the active-site grove
403
volume by ~25% in the smPBP5 than in sPBP5, the space at the binding pocket is possibly
404
inadequate for proper fitting of the pentapeptide substrate (Fig. 4). Hence, the release of terminal
405
D-alanine may be less during DD-CPase reaction, lowering the enzymatic efficiency of smPBP5.
406
However, the interaction of the smPBP5 with the relatively smaller substrate Bocillin-FL is
407
barely affected as it can fit comfortably into the groove (Table. 3b). A similar effect has been
408
reported for E. coli DacD (Chowdhury et al., 2012). Therefore, the present observation is in line
409
with our earlier speculation that though there is no remarkable difference in overall folding of the
410
enzyme, the activity of the PBP5 towards the substrate depends not only on the orientation of the
411
active-site residues but also on the active-site groove-volume. In a nutshell, it can be stated that
412
the discrete organization of the active-site residues in smPBP5 may be attributed to two big
413
deflections, one between Ser44 and Lys47 and another between Ser110 and Lys213, which
414
adversely affect their orientation and the active-site groove volume. This in turn impedes the
415
enzymatic activity both in vivo and in vitro. Therefore, we conclude that Leu182 plays a pivotal
416
role in maintaining the physiology of E. coli PBP5.
417 418 419 420 421 422
Acknowledgement
423
This work was supported in part by two different grants. One from the Department of
424
Biotechnology (DBT-NE program), Government of India and another from Sponsored Research
425
and Industrial Consultancy (SRIC) SGIRG seed Grant to ASG. MD and DK were supported by
426
Institute Fellowship of IIT Kharagpur and AB was Council of Scientific and Industrial Research,
427
Govt. of India.
428 429 430 431 432 433 434 435 436
437 438 439 440
References
441
Baquero, M.R., Bouzon, M., Quintela, J.C., Ayala, J.A. & Moreno, F. (1996). dacD, an
442
Escherichia coli gene encoding a novel penicillin-binding protein (PBP6b) with DD-
443
carboxypeptidase activity. J Bacteriol 178, 7106-7111
444
Banerjee, S., Pieper, U., Kapadia, G., Pannell, L.K. & Herzberg, O. (1998). Role of the Ω-
445
loop in the activity, substrate specificity, and structure of class A β-lactamase. Biochemistry
446
37, 3286–3296.
447
Chakraborty, S., (2012). An Automated Flow for Directed Evolution Based on Detection of
448
Promiscuous Scaffolds Using Spatial and Electrostatic Properties of Catalytic Residues.
449
PLoS One 7, e40408.
450 451
Chambers, H.F., Sachdeva, M.J. & Hackbarth, C.J. (1994). Kinetics of penicillin binding to penicillin-binding proteins of Staphylococcus aureus. Biochem J 301, 139–144.
452
Chowdhury, C., Nayak, T.R., Young, K.D. & Ghosh, A.S. (2010). A weak DD-
453
carboxypeptidase activity explains the inability of PBP 6 to substitute for PBP 5 in
454
maintaining normal cell shape in Escherichia coli. FEMS Microbiol Lett 303, 76–83.
455
Chowdhury, C., Kar, D., Dutta, M., Kmar, A. & Ghosh, A.S. (2012). Moderate deacylation
456
efficiency of DacD explains its ability to partially restore beta-lactam resistance in
457
Escherichia coli PBP5 mutant. FEMS Microbiol Lett 337, 73–80.
458 459
Davies, C., Whites, S.W. & Nicholas, R.A. (2001). Crystal Structure of a Deacylation-defective Mutant of Penicillin-binding Protein 5 at 2.3-Å Resolution. J BiolChem 276, 616–623.
460
Di Guilmi, A.M., Mouz, N., Petillot, Y., Forest, E., Dideberg, O. & Vernet, T. (2000).
461
Deacylation kinetics analysis of Streptococcus pneumoniae penicillin-binding protein 2x
462
mutants resistant to beta-lactam antibiotics using electrospray ionization- mass spectrometry.
463
Anal Biochem 284, 240–246.
464
Dundas, J., Ouyang, Z., Tseng, J., Binkowski, A., Turpaz, Y. & Liang, J. (2006). CASTp:
465
computed atlas of surface topography of proteins with structural and topographical mapping
466
of functionally annotated residues. Nucleic Acids Res 34, W116–8.
467
Frere, J.M., Leyh-Bouille, M., Ghuysen, J.M., Nieto, M. & Perkins, H.R. (1976). Exocellular
468
DD-carboxypeptidases-transpeptidases from Streptomyces. Methods Enzymol 45, 610–636.
469
Ghosh, A.S., Chowdhury, C. & Nelson, D.E. (2008). Physiological functions of D-alanine
470
carboxypeptidases in Escherichia coli. Trends Microbiol 16, 309–317.
471
Ghosh, A.S., Melquist, A.L. & Young, K.D. (2006). Loss of O-antigen increases cell shape
472
abnormalities in penicillin-binding protein mutants of Escherichia coli. FEMS Microbiol Lett
473
263, 252-257.
474
Ghosh, A.S. & Young, K.D. (2003). Sequences near the Active Site in Chimeric Penicillin
475
Binding Proteins 5 and 6 Affect Uniform Morphology of Escherichia coli. J Bacteriol 185,
476
2178–2186.
477 478
Ghuysen, J.M. (1991). Serine β-lactamases and penicillin-binding proteins. Annu Rev Microbiol 45, 37–67.
479
González-Leiza, S.M., de Pedro, M.A. & Ayala, J.A. (2011). AmpH, a bifunctional DD-
480
endopeptidase and DD-carboxypeptidase of Escherichia coli. J Bacteriol 193, 6887-6894.
481
Holtje, J.V. (1998). Growth of the stress-bearing and shape-maintaining murein sacculus of
482
Escherichia coli. Microbiol Mol Biol Rev 62, 181–203.
483 484
Jones, D.T. (1999). Protein secondary structure prediction based on position-specific scoring matrices. J Mol Biol 292, 195–202.
485 486
Korat, B., Mottl, H. & Keck, W. (1991). Penicillin-binding protein 4 of Escherichia coli:
487
molecular cloning of the dacB gene, controlled overexpression, and alterations in murein
488
composition. Mol Microbiol 5, 675-684.
489
Laskowski, R.A., MacArthur, M.W., Moss, D.S. & Thornton, J.M. (1993). PROCHECK: a
490
programme to check the stereochemical quality of protein structures. J Appl Crystallogr 26,
491
283–291.
492 493
Lineweaver, H. & Burk, D. (1934). The determination of enzyme dissociation constants. J Am Chem Soc 56, 658–666.
494
Louis-Jeune, C., Andrade-Navarro, M.A. & Perez-Iratxeta, C. (2012). Prediction of protein
495
secondary structure from circular dichroism using theoretically derived spectra. Proteins
496
80, 374-381.
497 498
Luthy, R., Bowie, J.U. & Eisenberg, D. (1992). Assessment of protein models with threedimensional profiles. Nature. 356, 83–85.
499
Malhotra, K.T. & Nicholas, R.A. (1992). Substitution of lysine 213 with arginine in penicillin-
500
binding protein 5 of Escherichia coli abolishes D-alanine carboxypeptidase activity without
501
affecting penicillin binding. J Biol Chem 267, 11386–91.
502 503
Nelson, D.E. & Young, K.D. (2000). Penicillin Binding Protein 5 Affects Cell Diameter, Contour and Morphology of Escherichia coli. J Bacteriol 183, 1714–1721.
504
Nelson, D.E. & Young, K.D. (2001). Contributions of PBP 5 and DD-Carboxypeptidase
505
Penicillin Binding Proteins to Maintenance of Cell Shape in Escherichia coli. J Bacteriol
506
183, 3055–3064.
507
Nelson, D.E., Ghosh. A.S., Paulson, A.L. & Young, K.D. (2002). Contribution of membrane-
508
binding and enzymatic domains of penicillin binding protein 5 to maintenance of uniform
509
cellular morphology of Escherichia coli. J Bacteriol 184, 3630–3639.
510
Nicola, G., Fedarovich, A., Nicholas, R.A. & Davies, C. (2005). A large displacement of the
511
SXN motif of Cys115-modified penicillin-binding protein 5 from Escherichia coli. Biochem
512
J 392, 55–63.
513
Nicola, G., Peddi, S., Stefanova, M., Nicholas, R.A., Gutheil, W.G. & Davies, C. (2005).
514
Crystal Structure of Escherichia coli Penicillin-Binding Protein 5 Bound to a Tripeptide
515
Boronic Acid Inhibitor: A Role for Ser-110 in Deacylation. Biochemistry 44, 8207–8217.
516
Nicholas, R.A., Krings, S., Tomberg, J., Nicola, G. & Davies, C. (2003). Crystal structure of
517
wild-type penicillin-binding protein 5 from Escherichia coli: implications for deacylation of
518
the acyl-enzyme complex. J Biol Chem 278, 52826–33.
519
Nilsen T, Ghosh A.S., Goldberg M. B.
& Young K.D. (2004). Branching sites and
520
morphological abnormalities behave as ectopic poles in shape-defective Escherichia coli.
521
Mol Microbiol 52 (4), 1045–1054
522
Potluri, L., Karczmarek , A., Verheul, J., Piette, A., Wilkin, J.M., Werth, N., Banzhaf, M.,
523
Vollmer, W., Young, K.D., Nguyen-Distèche, M. & den Blaauwen, T. (2010). Septal and
524
lateral wall localization of PBP5, the major D,D-carboxypeptidase of Escherichia coli,
525
requires substrate recognition and membrane attachment. Mol Microbiol 77, 300-323.
526 527 528 529
Rost, B., Yachdav, G. & Liu, J. (2004). The PredictProtein server. Nucleic Acids Res 32, W321–W326. Sali, A. & Blundell, T.L. (1993). Comparative protein modelling by satisfaction of spatial restraints. J Mol Biol 234, 779–815.
530
Sarkar, S.K., Chowdhury, C. & Ghosh, A.S. (2010). Deletion of penicillin-binding protein 5
531
(PBP5) sensitises Escherichia coli cells to beta-lactam agents. Int J Antimicrob Agents 35,
532
244–249.
533
Sarkar, S.K., Dutta, M., Chowdhury, C., Kumar, A. & Ghosh, A.S. (2011). PBP5, PBP6 and
534
DacD play different roles in intrinsic β-lactam resistance of Escherichia coli. Microbiology
535
157, 2702–2707.
536
Sauvage, E., Powell, A.J., Heilemann, J., Josephine, H.R., Charlier, P., Davies, C. & Pratt,
537
R.F. (2008). Crystal Structures of Complexes of Bacterial DD-Peptidases with
538
Peptidoglycan-Mimetic Ligands: The Substrate Specificity Puzzle. J Mol Biol 381, 383–393.
539
Stefanova, M.E., Davies, C., Nicholas. R.A. & Gutheil, W.G. (2002). pH, inhibitor, and
540
substrate specificity studies on Escherichia coli penicillin-binding protein 5. Biochim.
541
Biophys.Acta. 1597, 292–300.
542
Sun, T., Nukaga, M., Mayama, K., Braswell, E.H. & Knox, J.R. (2003). Comparison of β-
543
lactamases of classes A and D: 1.5-Å crystallographic structure of the class D OXA-1
544
oxacillinase. Protein Sci 12, 82–91.
545
Urbach, C., Fastrez, J. & Soumillion, P. (2008). A new family of cyanobacterial penicillin-
546
binding proteins: a missing link in the evolution of class A betalactamases. J Biol Chem 283,
547
32516–32526.
548
van der Linden, M.P., de Haan, L., Dideberg, O. & Keck, W. (1994). Site-directed
549
mutagenesis of proposed active-site residues of penicillin-binding protein 5 from Escherichia
550
coli. Biochem J 303, 357–362.
551
Zhang, W., Shi, Q., Meroueh, S.O., Vakulenko, S.B. & Mobashery, S. (2007). Catalytic
552
mechanism of penicillin-binding protein 5 of Escherichia coli. Biochemistry 46, 10113–21.
553
Zhao, G., Meier, T.I., Kahl, S.D., Gee, K.R. & Blaszczak, L.C. (1999). BOCILLIN FL, a
554
sensitive and commercially available reagent for detection of penicillin-binding proteins.
555
Antimicrob Agents Chemother 43, 1124–1128.
556 557 558 559 560 561 562 563 564 565 566 567 568 569 570
571 572
Figure Legends
573
Fig. 1. Cell shape of CS703-1 with the expression of PBP5L182E. Row 1: without aztreonam
574
and Row 2: with aztreonam (5 µg ml-1). From Left to right Column 1: CS703-1 with pBAD18
575
cam; Column 2: CS703-1 with pPJ5 and Column 3: CS703-1 with pPJ5L182E
576 577
Fig. 2a. SDS-PAGE analysis of membrane proteins: Lane 1, 3 and 4 show the proteins that are
578
stained with comassie blue stain; Lane 5, 7 and 8 are the Bocillin-binding of the membranes of
579
CS703-1/pBAD, CS703-1/pPJ5 and CS703-1/pPJ5L182E, respectively. Lanes 2 and 6 are for
580
Protein molecular mass marker.
581
2b. SDS-PAGE (12%) analysis of the purified soluble enzymes: Lane 1: Protein molecular mass
582
marker, Lane 2: purified sPBP5; Lane 3: purified smPBP5; Lane 4 and Lane 5 are purified
583
sPBP5 and smPBP5 labeled with Bocillin-FL, respectively.
584 585
Fig. 3. Superimposition of the active-site residues of sPBP5 (1NZO) over smPBP5 (RMSD value
586
is 0.91 Å). The residues of sPBP5 and smPBP5 are colored in ‘Green’ and ‘Magenta’,
587
respectively. In smPBP5, the γ-OH of Ser110, γ-OH of Ser44 and ε-NH2 of Lys47 are twisted
588
from their original position by 100.28°, 85.34° and 125.80°, respectively.
589 590
Fig. 4. The surface-view of (a) 1NZO and (b) modeled smPBP5. The active-site groove volumes
591
of 1NZO and smPBP5 are shown in yellow color. (a) 1NZO = 960 Å3 and smPBP5= 782 Å3
592
593 594 595 596 597 598
Table 1. List of E. coli strains and plasmids
599
600 601 602 603 604 605 606 607
E. coli strains
Genotypes
Source or References
CS109
W1485 rpoS rph
C. Schnaitman
AM15-1
CS109∆PBP5
Sarkar et al., 2010
CS703-1
CS109∆PBP1, 4, 5, 6, 7, AmpC and AmpH
Nelson et al., 2002
Plasmids
Features
Source or References
pPJ5
E. coli PBP5 cloned in pBAD18 Cam
Kevin D. Young
pT7-cPBP5
E. coli sPBP5 cloned in pT7-7k
Robert A. Nicholas
pET5
E. coli sPBP5 cloned in pET28a(+)
This study
pPJ5L182E
E. coli PBP5L182E cloned in pBAD 18 Cam
This study
pTS5LE
E. coli sPBP5L153E (smPBP5) cloned in pT7-7k
This study
pET5LE
E. coli sPBP5L153E (smPBP5) cloned in pET28a(+)
This study
608 609 610 611 612 613
Table 2. Beta-lactam sensitivity of E. coli strains carrying pPJ5L182E
614
E. coli strains CS1O9 AM15-1 AM15-1/pPJ5 AM15-1/pPJ5L182E 615 616 617 618 619 620 621 622 623 624 625 626
PenicillinG 250 64 250 125
MIC (µg ml-1) Piperacillin Cefadroxil 2 32 0.5 8 2 32 1 16
Cefaclor 4 0.5 4 2
627 628 629 630 631
Table 3. Kinetic parameters of the enzymes with pentapeptides (a) and penicillin (b)
632
(a) sPBP5
smPBP5
Km (mM)
kcat (s-1)
Km (mM)
kcat (s-1)
L-Ala-γ-D-Glu-L-Lys-D-Ala-D-Ala
0.434±0.06
1.97±0.2
0.042±0.002
0.194±0.06
Nα,Nє-diacetyl-Lys-D-Ala-D-Ala
0.538±0.03
2.45±0.4
0.035±0.007
0.2±0.04
Substrate
633 634
(b) sPBP5
smPBP5
Substrate
Penicillin (Bocillin FL) 635 636 637 638 639 640 641 642
k2/K (M-1s-1)
k3 (s-1)
k2/K (M-1s-1)
k3 (s-1)
730±17.9
14.9±1.7
463±13.3
16.4±6.5
643 644 645 646 647 648 649 650
Table 4. Percentage distribution of secondary structures obtained by using different
651
techniques Different
Circular Dichroism
Techniques Secondary
α-Helix
β-Sheet
Random coil
α-Helix
β-Sheet
Random coil
sPBP5
22.60%
28.71%
48.69%
24.73%
29.67%
45.60%
smPBP5
21.41%
27.39%
51.2%
24.45%
29.95%
45.60%
structures
652 653 654 655
PredictProtein server
Figure-1-#MIC-D-14-00066-R1 Click here to download high resolution image
Figure-2_#MIC-D-14-00066-R1 Click here to download high resolution image
Figure-3_#MIC-D-14-00066-R1 Click here to download high resolution image
Figure-4_#MIC-D-14-00066-R1 Click here to download high resolution image
Supplementary Material Files_#MIC-D-14-00066-R1 Click here to download Supplementary Material Files: Supplementary-materials_#MIC-D-14-00066R1.pdf