J Antimicrob Chemother 2010; 65: 1926 – 1935

doi:10.1093/jac/dkq246 Advance Access publication 17 July 2010

Molecular analysis of porin gene transcription in heterogenotypic

multidrug-resistant Escherichia coli isolates from scouring calves
Heather M. Vinson 1, Ablesh Gautam 2, Susan Olet 1,3, Penelope S. Gibbs 1,4 and Robert Barigye 1,4*
1
Department of Veterinary and Microbiological Sciences, North Dakota State University, 1523 Centennial Blvd, Fargo, ND 58108, USA;
2
Maxwell H. Gluck Equine Research Center, Department of Veterinary Science, University of Kentucky, Lexington, KY 40546-0099, USA;
3
Department of Statistics, North Dakota State University, PO Box 6060, Fargo, ND 58108, USA; 4Department of Veterinary Diagnostic

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

Services, North Dakota State University, 1523 Centennial Blvd, Fargo, ND 58108, USA

*Corresponding author. Tel: +1-701-231-5271; Fax: +1-701-231-7514; E-mail: [email protected]

Received 1 April 2010; returned 5 May 2010; revised 19 May 2010; accepted 6 June 2010

Background: Despite evidence that altered membrane porins may impair microbial drug uptake thereby poten-
tially compounding efflux pump-mediated multidrug resistance, few studies have evaluated gene transcription
to identify multidrug-resistance-associated porins and other potential drug targets.
Methods: Genes that encode six membrane porins (fadL, lamB, ompC, ompF, ompW and yiaT) and two mem-
brane proteins (tolC and ompT) were assessed by PCR and by quantitative real-time PCR (qRT –PCR) analysis of
10 multidrug-resistant (MDR) and 10 antibiotic-susceptible (AS) Escherichia coli isolates. The mean DDCt values
for the study E. coli genes were analysed by the Wilcoxon test (P ¼ 0.05).
Results: All 20 E. coli isolates tested positive for tolC, lamB, ompC, ompF genes, while 10 MDR and 9/10 (90%) AS
isolates were positive for the fadL gene. Seven out of 10 (70%) MDR and 7/10 (70%) AS isolates were positive for
the yiaT gene, while 7/10 (70%) MDR and only 4/10 (40%) AS isolates were positive for the ompT gene. The
mean DDCt values for the tolC and yiaT genes were significantly higher in MDR than in AS isolates (Wilcoxon
test; P,0.05). No significant difference was seen with respect to fadL, lamB, ompC, ompF, ompT and ompW
gene transcription (Wilcoxon test; P .0.05).
Conclusions: Findings suggest up-regulated transcription of tolC and yiaT genes in the MDR E. coli isolates.
These results indirectly suggest that TolC and YiaT proteins may play some role(s) in multidrug resistance,
but proteomic studies are needed before the two proteins are considered potential drug targets.

Keywords: qRT –PCR, TolC, YiaT

Introduction not limited to, reduced transmembrane diffusion of drug mol-

ecules, mutation of the drug target and enzymatic modification
Multidrug resistance is a serious public and animal health or degradation of the antibacterial agent.9 It is widely known
problem that presents a major challenge to the treatment of that increased expression of membrane-based efflux pumps
bacterial infections. Recently, various species of extensively (EPs) is associated with multidrug resistance in most clinically
drug-resistant (XDR) bacteria including important pathogens relevant bacterial pathogens.10 – 18 This is because EPs expel a
like Mycobacterium spp., various members of the family wide range of structurally unrelated compounds including
Enterobacteriaceae1 – 6 and other XDR bacterial strains such as various antibacterial agents.10 – 13,16 The outer membrane of
Pseudomonas aeruginosa 7 and Acinetobacter baumannii 8 have Gram-negative bacteria also contains protein channels called
been reported. These reports underscore the challenge posed porins that serve as conduits through which various compounds
by these pathogens to antimicrobial therapy of human and including several groups of antibacterial agents gain access to
animal disease. Since drug molecules must cross the bacterial the interior of the organism.19 – 21 Interaction of antibiotic mol-
cell membrane and wall before reaching the target site, the ecules and porin channels determines drug translocation effi-
outer membrane of Gram-negative bacteria presents a consider- ciency; and likewise, the ability of bacteria to reduce the influx
able physical barrier to the influx of drug molecules. Other of antibiotic molecules through altered membrane porins
general mechanisms of bacterial resistance include, but are contributes to the emergence of antibiotic resistance.20 – 22

# The Author 2010. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved.
For Permissions, please e-mail: [email protected]

1926
Porin gene transcription in multidrug-resistant E. coli JAC
A number of researchers have reported that impaired uptake of Materials and methods
the drug molecules due to altered porins may compound
EP-mediated drug resistance.20,21,23,24 In fact, various research- Study E. coli isolates
ers have reported altered porin expression in drug-resistant All 10 MDR and 9 antimicrobial-susceptible (AS) E. coli strains used in this
Gram-negative bacteria.4,18,25,26 Clinically significant multidrug research were isolated from faecal samples and/or intestinal contents
resistance related to altered membrane permeability has been from scouring calves at the North Dakota State University (NDSU) Veter-
reported in Gram-negative enteric bacteria.20,21 These research- inary Diagnostic Laboratory (VDL). The scouring neonatal calves from
ers specifically document porin loss as a major bacterial resist- which the study isolates were cultured were raised at various beef and
ance mechanism that restricts the influx of b-lactam and dairy farms located within the State of North Dakota as well as
fluoroquinolone antibiotics.20,21 Porin modification, which may western parts of Minnesota. The AS1 isolate (ATCC E. coli strain
involve either up- or down-regulation of porin genes, has also #25922) used as the experimental control was an ATCC strain (Manassas,
VA, USA) with known broad-spectrum antimicrobial agent susceptibility.
been cited as a mechanistic event in the development of drug

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

E. coli isolates that were resistant to ≥80% of the antimicrobial agents
resistance.4,8,25,27
tested were designated MDR and then selected for the study. The Sensi-
TolC is a ubiquitous outer membrane protein (OMP) that par- titre panel included the following antibacterial agents: ampicillin; ceftio-
ticipates in the assembly of tri-component EP systems in Gram- fur; chlortetracycline; clindamycin; danofloxacin; florfenicol; gentamicin;
negative bacteria10,14 – 16,28,29 that allow various substrates to be neomycin; oxytetracycline; penicillin; spectinomycin; sulfadimethoxime;
transported from the inside to the outside of the microbial cell. tiamulin; tilmicosin; trimethoprim/sulfamethoxazole; tulathromycin; and
Along with TolC, Escherichia coli expresses a number of trimeric tylosin–tartrate base. All the E. coli isolates studied are currently kept
porins otherwise called ‘the classical porins’. These include in cryopreservation at 2808C in 20% glycerol and Luria–Bertani (LB)
outer membrane protein F (OmpF) and outer membrane broth at Research Laboratory 154, Department of Veterinary and Micro-
protein C (OmpC).23,30,31 Down-regulation of porin expression biological Sciences, NDSU. The antimicrobial susceptibility profiles of the
has been associated with drug resistance in Pseudomonas study E. coli isolates were determined by the Bovine/Porcine Tulathromy-
cepacia,32 while an ompF-null mutant Salmonella enterica cin MIC Sensititre microplate using the Sensititre susceptibility system
(Trek Diagnostic Systems, OH, USA) according to the manufacturer’s
serovar Typhi isolate was also found to be resistant to chloram-
instructions.
phenicol.33 Likewise, several strains of drug-resistant Haemophilus
influenzae with decreased porin gene expression have been
reported.34 Whereas drug-resistance-associated down-regulated Extraction of bacterial DNA
porins have been widely documented, a Chinese research team Study E. coli isolates were inoculated on MacConkey agar plates and incu-
recently reported up-regulation of TolC and the membrane bated overnight at 378C. A single colony of each isolate was added to
porins OmpC and OmpW, along with simultaneous down- 40 mL of 10 mM Tris/1 mM EDTA (TE) buffer with 1% 20 mg/mL protein-
regulation of the porins OmpF and FadL in a nalidixic acid- ase K. Samples were incubated at 558C for 10 min followed by 10 min
resistant E. coli isolate.27 The same workers also reported incubation at 808C. The extracted DNA was then diluted with 80 mL of
up-regulation of TolC and the porin LamB, and down-regulation sterile water, centrifuged for 5 min and stored at 2208C for future PCR
of FadL and OmpW in streptomycin-resistant E. coli isolates.25 analysis.
These observations have justified mechanistic studies into the
dynamics of porin gene regulation and how altered expression
PFGE
of porin and other OMP genes relates to drug resistance, with
the hope that the new knowledge can be used in the design The genetic/clonal relationship between each of the study isolates in the
of novel efficacious treatments for infections caused by two E. coli groups was determined by PFGE analysis. The molecular size
standard Salmonella Braenderup BAA-664 (ATCC) along with E. coli test
multidrug-resistant (MDR) pathogens. Our research group specu-
cultures were streaked out on LB agar (Sigma, St Louis, MO, USA) and
lates that in the long term, certain membrane porins may serve
incubated overnight at 378C. A sterile swab was used to transfer bacteria
as potential targets for novel drugs with efficacy against to a tube containing 2 mL of cell suspension buffer (100 mM Tris/100 mM
MDR/XDR bacterial pathogens. However, there is need for a EDTA, pH 8.0) and the concentration was adjusted to an absorbance
thorough understanding of porin gene transcription and reading of 1.35 at a wavelength of 610 nm. SeaKem Gold agarose
expression dynamics in MDR/XDR bacteria so that the porins (Lonza, Rockland, ME, USA) was prepared as a 1% solution in TE buffer
and other membrane proteins with the greatest relevance to (pH 8.0) and stored in a 588C water bath. A 400 mL aliquot of the bacterial
drug resistance are targeted for drug development. Recently, a suspension was gently mixed with 20 mL of proteinase K (20 mg/mL stock
porin-targeting vaccine designed by scientists at Epitopixw in solution) in a 1.5 mL microcentrifuge tube. An equal volume of the pre-
MN, USA was licensed to prevent colonization of cattle with heated agarose was then added to the bacterial suspension, gently
E. coli O157:H7 (http://www.epitopix.com). Despite numerous mixed and immediately dispensed into plug moulds (Bio-Rad Labora-
studies documenting altered porin gene expression in tories, Hercules, CA, USA). After the plugs solidified, they were placed
drug-resistant bacteria,9,26,27 comprehensive studies have not into a 50 mL tube containing 5 mL of cell lysis buffer (50 mM Tris/
50 mM EDTA/1% Sarkosyl, pH 8.0) along with 25 mL of proteinase K
been performed to evaluate porin gene expression in MDR
(20 mg/mL stock). Samples were allowed to incubate at 548C for 2 h at
bacteria. Therefore, the present research was designed to
180 rpm in a shaking incubator. Following incubation, the cell lysis
assess the relative transcription rates of genes that encode six buffer was poured off and 10 mL of sterile water preheated to 508C
membrane porins (FadL, LamB, OmpC, OmpF, OmpW and YiaT), was added. The samples continued to incubate at 508C in a shaking
a component of trimeric drug efflux pumps (TolC) and a drug incubator for 15 min at 180 rpm. The water was then poured off and
resistance-related membrane protease (OmpT) in MDR E. coli the water rinse was repeated followed by four washes with TE buffer
isolates. Plans are underway to undertake translational proteo- (pH 8.0). An 2.0 mm wide slice of plug was cut with a single-edge
mic studies of the corresponding proteins. razor blade and transferred to a 1.5 mL microcentrifuge tube containing

1927
Vinson et al.

Table 1. Annealing temperatures of the primers used in routine PCR analysis of porin and OMP genes

Porin gene Primer sequence (3′ 5′ ) Annealing temperature (8C) Amplicon size (bp)

tolC forward: ACAACGCCAAGCAAGAGC 60 197

reverse: CGCATAACCATCAGCAATAGC

ompC forward: CAGGATGTGGGTTCTTTCG 53 162

reverse: GAAGTCAGTGTTACGGTAGG

ompF forward: ACCTGGCAGCGAACTACG 54 191

reverse: AACATCACCGATACCTTCTACG

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

ompTa forward: TCCTCAACGAACCCAATTACC 54 198
reverse: TTCCAGTCAAGCCAATGTAGG

ompW forward: GGCGACCGACAACATTGG 60 107

reverse: TTGGTGGCAGATGATGAACG

lamB forward: GGTCACATGCTGCGTATCC 60 111

reverse: GGTGCCGTTGTCGTTATCC

fadL forward: CTTCTATTACCTCTAACTATGG 53 294

reverse: TACTGTCAATACCGTTGG

yiaT forward: CGGTCTGGCAAGTTATTCC 54 136

reverse: GCTGTCGGTAATCTCTTCC

a
Note that the ompT gene encodes an outer membrane-based protease, OmpT.

100 mL of 30 U of XbaI (Promega Corporation, Madison, WI, USA) in 1× Reverse Transcription Kit (Qiagen) according to the manufacturer’s
restriction buffer and 1.0 mL of BSA (Promega Corporation), followed by instructions. Four reactions containing reverse transcriptase (RT positive)
incubation at 378C for 2 h. After incubation, the restriction digestion and four reactions without reverse transcriptase (RT negative) were pre-
mixture was replaced with 200 mL of 0.5× Tris/borate/EDTA (TBE) and pared for each of the RNA samples. The cDNA synthesis was completed
allowed to incubate at room temperature for 5 min. Electrophoresis on a DNA Engine Thermocycler (Bio-Rad Laboratories) after which the
was then carried out on the CHEF-mapper system (Bio-Rad Laboratories) resulting cDNA product was diluted 10-fold, and a 1:100 dilution was
using a 1% SeaKem Gold agarose gel and 0.5× TBE running buffer using used in the qRT–PCR.
the following electrophoresis conditions: initial switch time, 2.16 s; final The Quantitect SYBR Green PCR Kit (Qiagen) was used for all real-time
switch time, 63.8 s; run time, 18 h; 148C; and ramping factor linear. PCRs in accordance with the manufacturer’s instructions. Primers used in
Upon completion of electrophoresis, the gel was stained for 30 min in this study (Table 2) were designed based on specific porin and OMP gene
an ethidium bromide solution (1 mg/mL) and destained with double dis- sequences obtained from the EcoCyc Database for E. coli K-12, strain
tilled water. Samples were further characterized using the BioNumerics MG1655 (SRI International) and manufactured by Trilink Biotechnologies.
software (Applied Maths, Austin, TX, USA). The primers include a housekeeping gene, gapA, used for a quantitative
control. Working concentrations were determined by testing each
primer at 300, 600 and 900 nM for each of the target genes. Each PCR
PCR assay reaction contained 5 mL of cDNA, 12.5 mL 2× QuantiTect SYBR Green
The PCR assay for the porin genes was carried out according to the con- PCR master mix, the appropriate concentration of primer (Table 2) and
ditions shown in Table 1. Primers used in this study (Table 2) were enough nuclease-free water to bring the volume to 25 mL. qRT– PCR
designed based on specific porin and OMP gene sequences obtained was performed using an iCycler IQ (Bio-Rad Laboratories) with the follow-
from the EcoCyc Database for E. coli K-12, strain MG1655 (SRI Inter- ing cycling conditions: an initial incubation of 958C for 15 min; followed
national, Menlo Park, CA, USA) and manufactured by Trilink Biotechnolo- by 40 cycles of 15 s at 948C, 30 s at the appropriate annealing tempera-
gies (San Diego, CA, USA). PCR products were run in a 1.5% agarose gel, ture (Table 2) and 728C for 30 s. All test samples were run in duplicate
and stained with ethidium bromide for visualization. including an RT-negative control for each sample set along with a
blank control consisting of nuclease-free water in place of cDNA. Quanti-
fication for each target gene was determined by the DDCt method using
Quantitative real-time PCR the housekeeping gene, gapA.35
To isolate RNA, 10 MDR and 10 AS E. coli isolates were inoculated into
4.0 mL of LB broth (EMD Chemicals, Gibbstown, NJ, USA) and incubated
at 378C to mid-logarithmic phase with OD600 values ranging from 0.15
Statistical analysis
to 0.30. Purification of total RNA from each of the study E. coli strains The statistical analysis on the DDCt values obtained by qRT–PCR was per-
was performed using the RNeasy Protect Bacteria Mini Kit (Qiagen, formed using SAS Version 9.13. Descriptive statistics entailed numerical
Valencia, CA, USA), according to the manufacturer’s instructions. An summaries and graphs. A non-parametric approach Wilcoxon rank sum
additional DNase treatment was carried out using Ambion’s DNA-free test was used to compare the location parameters of the DDCt values
Kit (Ambion, Austin, TX, USA) according to the manufacturer’s instruc- for the two groups of E. coli (MDR and AS E. coli isolates). All tests were
tions. The synthesis of cDNA was performed using the Quantitect done at a 5% level of significance.

1928
Porin gene transcription in multidrug-resistant E. coli JAC
Table 2. Working concentrations of the primers used in the qRT– PCR assay for porin and membrane protease genes

Name Reference Primer sequence (5′ 3′ ) Concentration (nM) Annealing temperature (8C)

gapA-F 18 ACTTCGACAAATATGCTGGC 300 53

gapA-R CGGGATGATGTTCTGGGAA

TolC-F this study ACAACGCCAAGCAAGAGC 600 60

TolC-R CGCATAACCATCAGCAATAGC

LamB-F this study GGTCACATGCTGCGTATCC 300 60

LamB-R GGTGCCGTTGTCGTTATCC

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

FadL-F this study CTTCTATTACCTCTAACTATGG 900 53
FadL-R TACTGTCAATACCGTTGG

YiaT-F this study CGGTCTGGCAAGTTATTCC 600 54

YiaT-R GCTGTCGGTAATCTCTTCC

OmpC-F this study CAGGATGTGGGTTCTTTCG 600 53

OmpC-R GAAGTCAGTGTTACGGTAGG

OmpF-F this study ACCTGGCAGCGAACTACG 600 54

OmpF-R AACATCACCGATACCTTCTACG

OmpT-F this study TCCTCAACGAACCCAATTACC 600 54

OmpT-R TTCCAGTCAAGCCAATGTAGG

OmpW-F this study GGCGACCGACAACATTGG 300 60

OmpW-R TTGGTGGCAGATGATGAACG

Note that ompT is a gene that encodes the membrane-based protease OmpT.

Results MDR and 10 AS E. coli isolates) were computed, and the Wilcoxon
test (P ¼ 0.05) was used to determine the level of statistical sig-
PFGE nificance between the gene transcription in the two E. coli
PFGE (Figure 1) did not indicate a clonal relationship between any groups. Based on the qRT–PCR data, the mean DDCt values for
of the isolates tested. The data indicate that all the study E. coli the tolC (P ¼ 0.0002436) and yiaT (P¼ 0.0041) genes were signifi-
isolates were distinct from each other, thus none of the data cantly higher in the MDR E. coli isolates than in the AS E. coli iso-
is a duplicate of any other isolates. lates, but no significant difference was seen for the fadL
(P ¼ 0.3154), lamB (P¼ 0.1903), ompC (P ¼0.1716), ompF
(P ¼ 0.5787), ompT (P¼ 0.2303) and ompW (P ¼ 0.1903) genes
Routine PCR (Table 3). Of the two membrane protein genes that were appar-
Analysis of PCR products showed the relevant bands for the OMP ently significantly hypertranscribed in the MDR isolates, tolC
and porin genes tolC (197 bp), fadL (294 bp), yiaT (136 bp), lamB showed the highest level of significance (P ¼ 0.0002436) followed
(111 bp), ompC (162 bp) and ompF (191 bp) and for the mem- by yiaT (P ¼0.0041). The mean DDCt values for the other
brane protease gene ompT (198 bp) in all the MDR and AS E. coli OMP-encoding genes appeared higher, but statistical analysis
isolates (Figure 2). With reference to results of routine PCR, all did not reveal any significant difference between the transcrip-
MDR and AS E. coli isolates had detectable tolC, lamB, ompC and tion of MDR and AS E. coli isolates.
ompF genes (Figure 2). PCR products for all the studied
outer-membrane-encoding genes were also detected in all of
the 10 MDR strains, while the fadL gene was detected in 9/10 AS Discussion
E. coli isolates (Figure 2). Seven out of 10 (70%) MDR and 7/10
(70%) AS E. coli isolates tested positive for the yiaT gene PFGE data did not indicate a significant clonal relationship
(Figure 2), while 7/10 MDR isolates and 4/10 AS E. coli isolates between the E. coli isolates tested during the study. Of the six
were positive for the ompT gene (Figure 2). All the PCR results porin and two OMP genes studied, the mean DDCt values for
were mirrored by the individual DDCt values [see the Supplemen- the tolC and yiaT genes were significantly higher (P,0.05) in
tary data, available at JAC Online (http://jac.oxfordjournals.org/)] MDR than for the AS E. coli isolates, while that of the ompT
for each of the porin genes as determined by qRT–PCR. gene was borderline (Table 3). Interestingly, no statistically sig-
nificant difference (P .0.05) was seen in the transcription
levels of fadL, lamB, ompC, ompF and ompW genes even
qRT– PCR though the sum of the scores was higher than what was
The mean DDCt values for the study porin and the other two expected under the null hypothesis of no difference. Overall,
membrane protein genes in the 20 study E. coli isolates (10 these findings are consistent with up-regulated transcription of

1929
Vinson et al.

Dice (Opt:1.00%) (Tol 1.0%-1.0%) (H>0.0% S>0.0%) [0.0%-100.0%]

PFGE XBA
PFGE XBA

100
55
60
65
70
75
80
85
90
95
76.9 AS 2
69.8 AS 8
76.9 MDR 9
69.2
AS 5
65.6 MDR 1

65.1 AS 1
84.2 MDR 2

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

61.8 72.5 MDR 5
60 AS 10
AS 4
MDR 7
58.2
83.3 MDR 6
71.8 AS 6
57.1 61.5 MDR 4
MDR 10
55.3 MDR 3
71.4 AS 3
53.1 70 AS 9
AS 7
MDR 8

Figure 1. PFGE of 10 MDR and 10 AS E. coli isolates.

tolC and yiaT genes, but not the other six OMP genes in the yiaT transcription in MDR E. coli isolates in contrast to extremely
MDR E. coli isolates. Except for the yiaT gene, altered low transcription levels in the AS E. coli isolates. YiaT is a 27.4 kDa
porin-gene-associated drug resistance has been investigated protein that was first reported as a putative outer membrane
for the other seven proteins in drug-resistant E. coli isolates,27,36 porin.37 Despite the current knowledge of its structural identity
but not in MDR bacteria. The present findings strongly suggest, and spatial location within the bacterial cell,37 the physiological
albeit indirectly, that expression of TolC and YiaT proteins may and other biochemical functions of the protein have not been
be crucial in mediating MDR mechanisms. Plans are underway described to date. For the first time, data from our research
to study both proteins by translational proteomics as a prelude demonstrate that up-regulated transcription of the yiaT gene
to evaluating their suitability as potential targets for developing may be associated with yet to be described drug resistance
efficacious drugs against potentially deadly MDR pathogens. mechanisms. Subject to corroboration by translational proteo-
Based on routine PCR data, the tolC gene was present in all mics, the present findings further suggest, albeit indirectly, that
study MDR and AS E. coli isolates (Figure 2). However, when the this porin may play some role(s) in MDR mechanisms. Since
qRT –PCR data were analysed, the mean DDCt value for the YiaT protein is predicted to be related to transmembrane sub-
gene was significantly higher in the MDR isolates (P,0.05) strate transport, these mechanisms deserve to be unravelled
than in the AS E. coli isolates (Table 3 and Figure 3). TolC has and characterized before the protein is fully considered a poten-
been widely reported for its drug efflux properties and plays a tial target for developing new drugs against XDR/MDR pathogens.
critical role in mediating clinically significant MDR.15,17,29 However, the fact that a minority of both MDR and AS E. coli iso-
Up-regulation of TolC and other membrane proteins was also lates did not transcribe the yiaT gene suggests that if it were to
recently reported in E. coli isolates with resistance to chloram- play a role in drug resistance mechanisms in MDR E. coli, the
phenicol,36 streptomycin25 and nalidixic acid.27 As would have mechanism in question would perhaps be an adjunct to some
been expected, therefore, the present data strongly suggest, other system. This finding is not surprising since most membrane
albeit indirectly, that possible TolC overexpression may be an porins are physiologically redundant and work together in multi-
important mechanism of MDR in Gram-negative bacteria. component networks. For example, the expression of OmpC and
These findings need to be corroborated by translational proteo- OmpF is modulated by the transcription regulators OmpR and
mics to explore the potential for developing TolC-targeting thera- EnvZ, and both are encoded by the ompB operon.38 What
peutic agents with efficacy against MDR/XDR Gram-negative needs to be defined for YiaT are the possible regulatory pathways
bacteria. involved in its expression in addition to determining the precise
Routine PCR analysis of the genes that encode the study manner in which those expression patterns may affect outer
membrane proteins revealed that 2/10 MDR and 3/10 AS E. coli membrane permeability and drug resistance.
isolates were negative for the yiaT gene. On the other hand, FadL is an outer membrane b-barrel protein involved in the
the qRT–PCR data were remarkably consistent with up-regulated uptake of long-chain fatty acids.25,39 The amino terminus of the

1930
Porin gene transcription in multidrug-resistant E. coli JAC
er

er
dd

dd
bp 1 2 3 4 5 6 7 8 9 10 11 bp 1 2 3 4 5 6 7 8 9 10 11
La

La
200 300
100 200

12 13 14 15 16 17 18 19 20 12 13 14 15 16 17 18 19 20
200 300
100 200
A = tolC B = fadL
er

er
dd

dd

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

1 2 3 4 5 6 7 8 9 10 11 1 2 3 4 5 6 7 8 9 10 11
La

La
bp bp
200
100 100

12 13 14 15 16 17 18 19 20
12 13 14 15 16 17 18 19 20
200
100 100

C = yiaT D = lamB

er
er

dd
dd

bp 1 2 3 4 5 6 7 8 9 10 11 bp 1 2 3 4 5 6 7 8 9 10 11 12

La
La

300 300
200 200

12 13 14 15 16 17 18 19 20 12 13 14 15 16 17 18 19 20
300 300
200 200
E = ompC F = ompF
er
dd

bp 1 2 3 4 5 6 7 8 9 10 11
La

300
200

12 13 14 15 16 17 18 19 20

300
200

G = ompT

Figure 2. Analysis of porin and membrane protease gene PCR products in 1% agarose gels stained with ethidium bromide. Lanes 1 –10 represent MDR
E. coli isolates and lanes 11 –20 represent AS E. coli isolates. Note that duplicate lanes were loaded for each of the samples.

protein also contains an attachment site for bacteriophage T2, statistically significant difference between the mean DDCt
while the carboxyl end has a crucial amino acid sequence needed values for the 10 MDR and 10 AS E. coli isolates. OmpC is a
for fatty acid transport.40 Based on digital analysis of the relevant small trimeric porin through which transmembrane transport
two-dimensional gel protein spot in a single E. coli isolate, a of small ions and other hydrophilic solutes ≤500 Da takes
Chinese research team recently reported down-regulation of FadL place.41 The up-regulation of OmpC has been reported in
expression in nalidixic acid- and streptomycin-resistant E. coli iso- drug-resistant bacteria,27 while decreased production of the
lates.25,27 The qRT–PCR data from the present study did not protein has also been observed in response to exogenous polya-
demonstrate a statistically significant difference in fadL gene tran- mines.42 The latter compounds are known to induce resistance
scription rates between the MDR and AS E. coli groups. to ColE7, a product of an SOS regulon that encodes a bacteriocin
Routine PCR analysis showed ompC expression by all 20 study with potency against susceptible E. coli and related enterobac-
E. coli isolates, while qRT–PCR analysis did not reveal a teria under conditions of stress.42 The rather conflicting findings

1931
Vinson et al.

by Li et al.25 and our data are not surprising since current evi- lead to decreased expression or total OmpC loss.18 Li et al.25
dence suggests that progression of drug resistance mechanisms have suggested that this porin along with TolC may play impor-
like those caused by continued exposure to subinhibitory concen- tant roles in drug resistance mechanisms. However, their con-
trations of certain antimicrobial agents may in a step-by-step clusions on OmpC were based on digital analysis of relevant
manner select for porin expression modifications that eventually gel spots from a single drug-resistant E. coli strain.25 Whereas

Table 3. Wilcoxon analysis (P¼0.05) of mean DDCt values for the 10 MDR and 10 AS E. coli isolates

Primer No. of MDR E. coli isolates No. of AS E. coli isolates Wilcoxon statistic P value

tolC 10 10 148 0.0002436**

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

lamB 10 10 123 0.1903
ompC 10 10 123 0.1716
fadL 10 9 77 0.3154
yiaT 7 7 74 0.0041*
ompF 10 10 113 0.5787
ompT 7 4 17 0.2303
ompW 10 10 123 0.1903

A statistically significant difference was verified for the transcription levels of the tolC and yiaT genes (P,0.05), but
not for fadL, lamB, ompC, ompF, ompT and ompW genes (P.0.05). A single asterisk indicates a statistically
significant P value, while double asterisks denote a highly significant P value.

900.0 12 000.0
478.83 yiaT ompT
800.0 6897.33
10 000.0
700.0

600.0 8000.0
Ct

Ct

500.0
6000.0
Mean

Mean

400.0

300.0 4000.0

200.0 1158.02
2000.0
100.0
7.16
0.0 0.0
MDR AS MDR AS

1200.0 tolC 800.0 lamB

711.47 472.53
700.0
1000.0
600.0
800.0
500.0
Ct

Ct

600.0 400.0
Mean

Mean

300.0
400.0
200.0
91.63
200.0 72.39
100.0

0.0 0.0
MDR AS MDR AS

Figure 3. Bar graphs showing the mean DDCt values for the yiaT, ompT, tolC, lamB, ompC, fadL, ompW and ompF genes in the 10 MDR and 10 AS E. coli
isolates.

1932
Porin gene transcription in multidrug-resistant E. coli JAC
10 000.0 ompC 2000.0 fadL
1052.83
6668.41
9000.0 1800.0
8000.0 1600.0
7000.0 1400.0
3874.25
Ct

Ct
6000.0 1200.0
5000.0 1000.0
Mean

Mean
4000.0 800.0
3000.0 600.0

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

2000.0 400.0 90.02
1000.0 200.0
0.0 0.0
MDR AS MDR AS

120 000.0 ompW 450.0 ompF

262.11
54557.89 400.0
100 000.0
350.0

80 000.0 300.0
Ct

Ct

250.0
60 000.0
Mean

Mean

200.0

40 000.0 150.0

100.0 59.11
20 000.0
1053.94 50.0

0.0 0.0
MDR AS MDR AS

Figure 3. Continued.

current evidence on the unequivocal role of TolC in multidrug OmpF production may decrease in response to exogenous polya-
resistance justifies targeting it for the development of novel mines.22 The latter have been shown to inhibit the influx of anti-
drugs, the conflicting results on OmpC transcription dynamics biotic molecules into bacteria.31 Most recently, down-regulation
suggest that despite the potential role in drug resistance mech- of OmpFexpression was reported in a drug-resistant E. coli isolate.25
anisms, targeting this protein may be problematic and unre- In the present study, routine PCR analysis showed the pres-
warding. This is due to the consideration that both ence of the lamB gene in all 20 study E. coli isolates. The mean
up-regulation and down-regulation may have mechanistic impli- DDCt values obtained by qRT –PCR analysis of the 10 MDR
cations for drug resistance. E. coli were higher than for the 10 AS E. coli isolates, but this
Routine PCR analysis showed the presence of the ompF gene in difference was statistically insignificant. LamB is a sugar-specific
all 20 study E. coli isolates, while qRT–PCR analysis revealed no stat- porin involved in transmembrane transport of maltose and mal-
istically significant difference between the mean DDCt values for todextrins that are crucial for bacterial metabolism.46 This porin
the 10 MDR and 10 AS E. coli isolates. OmpF is a large porin that was reported to be down-regulated in a tetracycline-resistant
allows transmembrane diffusion of solutes such as sugars, ions E. coli isolate,47 but was also later shown to be up-regulated in
and amino acids ,600 Da in size.30,43 Loss of large-channel a streptomycin-resistant E. coli isolate.25 Our finding with
porins is often associated with increased resistance to hydrophilic regard to the presence of the gene in all 20 E. coli isolates
antibacterial agents.33 For instance, ompF E. coli mutant strains studied is neither surprising nor of much significance.
have been reported to develop resistance to b-lactam antibiotics,44 Transcription dynamics of the ompT gene in the two E. coli
while clinical isolates of porin-deficient Serratia marcescens were groups were also evaluated and found to be borderline. This
resistant to aminoglycosides and b-lactam antibiotics.45 An ompF- gene encodes an outer membrane protease that is known to
null mutant Salmonella enterica serovar Typhi isolate was also hydrolyse the antimicrobial peptide protamine.48 Recently,
resistant to chloramphenicol.33 Under physiological conditions, up-regulation of OmpT was reported in a drug-resistant E. coli
OmpF imports group A colicins, the bactericidal molecules pro- isolate25,36 suggesting it may play a role in drug resistance
duced by certain E. coli isolates.20 Other studies have shown that most probably through degradation of antibiotic molecules.

1933
Vinson et al.

The membrane protease has also been reported to be a virulence 6 Jain A, Mondal R. Extensively drug-resistant tuberculosis: current
factor in urinary tract infections.49 Based on this background, we challenges and threats. FEMS Immunol Med Microbiol 2008; 53: 145–50.
wanted to assess the transcription dynamics of the ompT gene in 7 Bonomo RA, Szabo D. Mechanisms of multidrug resistance in
MDR E. coli isolates to determine whether membrane-based Acinetobacter species and Pseudomonas aeruginosa. Clin Infect Dis
enzyme might indeed participate in drug resistance mechan- 2006; 43: S49–56.
isms. In the present study, routine ompT PCR results were the 8 Vila J, Martı́ S, Sánchez-Céspedes J. Porins, efflux pumps and multidrug
most dramatic as 7/10 of the MDR and only 4/10 AS E. coli resistance in Acinetobacter baumannii. J Antimicrob Chemother 2007; 59:
strains had a detectable PCR signal for the ompT gene. 1210– 5.
The ultimate goal of this research was to predict which mem- 9 McDermont PF, Walker RD, White DG. Antimicrobials: modes of action
brane porins or protease/protein might be involved in MDR mech- and mechanisms of resistance. Int J Toxicol 2003; 22: 135– 43.
anisms so that new targets for developing efficacious anti-MDR 10 Tikhonova EB, Zgurskaya HI. AcrA, AcrB and TolC of Escherichia coli
pathogen drugs are identified. Experimental data indicated for form a stable intermembrane multidrug efflux complex. J Biol Chem

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

the first time that the novel porin YiaT is likely to be associated 2004; 279: 32116–24.
with the MDR phenomenon. The findings of our research
11 Chang G. Structure of MsbA from Vibrio cholerae: a multidrug
suggest, albeit indirectly, that in addition to TolC, the porin YiaT resistance ABC transporter homolog in a closed conformation. J Mol
may play a role in mediating MDR mechanisms. As such, the Biol 2003; 330: 419–30.
two OMPs should be studied further by translational proteomics
12 Neyfakh AA, Bidnenko VE, Chen LB. Efflux-mediated multidrug
and explore their potential as targets for developing novel drugs
resistance in Bacillus subtilis: similarities and dissimilarities with the
against infections caused by troublesome MDR pathogens. mammalian system. Proc Natl Acad Sci USA 1991; 88: 4781– 5.
Subject to corroboration by translational proteomics of the
13 Nikaido H, Vaara M. Molecular basis of bacterial outer membrane
study OMPs, the current qRT– PCR data appear to suggest that
permeability. Microbiol Rev 1985; 49: 1 –32.
transcriptional differences within the lamB, ompC, ompF and
ompW genes may not be associated with MDR mechanisms. 14 Andersen C, Hughes C, Koronakis V. Chunnel vision. Export and efflux
through bacterial channel-tunnels. EMBO Rep 2000; 1: 313–8.
15 Koronakis V, Sharff A, Koronakis E et al. Crystal structure of the
bacterial membrane protein TolC central to multidrug efflux and
Funding protein export. Nature 2000; 405: 914–9.
This work was supported by financial resources provided through the 16 Nikaido H. Multidrug efflux pumps of Gram-negative bacteria.
United States Department of Agriculture (USDA)’s Animal and Plant J Bacteriol 1996; 178: 5853– 9.
Health Inspection Service (APHIS) Biosurveillance (grant number
17 Fralick A. Evidence that TolC is required for functioning of the Mar/
FARG014465) and the North Dakota State University Development
AcrAB efflux pump of Escherichia coli. J Bacteriol 1996; 178: 5803– 5.
Foundation (grant number FARG0013382).
18 Nikaido H. Multidrug resistance in bacteria. Annu Rev Biochem 2009;
78: 119–46.
19 Chang CF, Mizushima S, Glaeser RM. Projected structure of the
Transparency declarations pore-forming OmpC protein from Escherichia coli outer membrane.
None to declare. Biophys J 1985; 47: 629– 39.
20 Zakharov SD, Cramer WA. On the mechanism and pathway of colicin
import across the E. coli outer membrane. Front Biosci 2004; 9: 1311– 7.
Supplementary data 21 Pagès JM, Chloë EJ, Winterhalter M. The porin and the permeating
antibiotic: a selective diffusion barrier in Gram-negative bacteria. Nature
Supplementary data are available at JAC Online (http://jac.oxfordjournals.
org/). Rev Microbiol 2008; 6: 893– 903.
22 Pan YH, Liao CC, Kuo CC et al. The critical roles of polyamines in
regulating ColE7 production and restricting ColE7 uptake of the
colicin-producing Escherichia coli. J Biol Chem 2006; 281: 13083–91.
References
23 Cohen SP, McMurry LM, Levy SB. marA locus causes decreased
1 Alcaide F, Santin M. Multidrug-resistant tuberculosis. Enferm Infec
expression of OmpF porin in multiple-antibiotic-resistant (Mar) mutants
Microbiol Clin 2008; 26: 54 –60.
of Escherichia coli. J Bacteriol 1988; 170: 5416 –22.
2 Banerjee R, Schecter GF, Flood J et al. Extensively drug-resistant
tuberculosis: new strains, new challenges. Expert Rev Anti Infect Ther 24 Gambino L, Gracheck SJ, Miller PF. Overexpression of the MarA positive
2008; 6: 713–24. regulator is sufficient to confer multiple antibiotic resistance in
Escherichia coli. J Bacteriol 1993; 175: 2888– 94.
3 Bratu S, Brooks S, Burney S et al. Detection and spread of Escherichia
coli possessing the plasmid-borne carbapenemases KPC-2 in Brooklyn, 25 Li H, Wang BC, Xu WJ et al. Identification and network of outer
New York. Clin Infect Dis 2007; 44: 972– 5. membrane proteins regulating streptomycin resistance in Escherichia
coli. J Proteome Res 2008; 7: 4040– 9.
4 Brink A, Feldman C, Richards G et al. Emergence of extensive drug
resistance (XDR) among Gram-negative bacilli in South Africa looms 26 Mallea M, Chevalier J, Bornet C et al. Porin alteration and active efflux:
nearer. S Afr Med J 2008; 98: 586– 92. two in vivo drug resistance strategies used by Enterobacter aerogenes.
Microbiology 1998; 144: 3003 –9.
5 Elliott E, Brink A, Greune JV et al. In-vivo development of ertapenem
resistance in a patient with pneumonia caused by Klebsiella 27 Lin M, Li H, Wan BC et al. Proteomic analysis of nalidixic acid
pneumoniae with extended-spectrum b-lactamase. Clin Infect Dis resistance in Escherichia coli: identification and functional
2006; 42: e95–8. characterization of OM protein. J Proteome Res 2008; 7: 2399– 405.

1934
Porin gene transcription in multidrug-resistant E. coli JAC
28 Benz B, Maier E, Gentsdiev I. TolC of Escherichia coli functions as an 40 Cristalli G, DiRusso CC, Black PN. The amino-terminal region of the
outer membrane channel. Zentralbl Bakteriol 1993; 278: 187– 96. long-chain fatty acid transport protein FadL contains an externally
29 Sulavik C, Houseweart C, Cramer C et al. Antibiotic susceptibility exposed domain required for bacteriophage T2 binding. Arch Biochem
profiles of Escherichia coli strains lacking multidrug efflux pump genes. Biophys 2000; 377: 324– 33.
Antimicrob Agents Chemother 2001; 45: 1126–36. 41 Macnab RM. Escherichia coli and Salmonella. In: Neidhardt FC, Curtiss
30 Inokuchi K, Mutoh N, Matsuyama S et al. Primary structure of the R, Ingraham JL et al., eds. Cellular and Molecular Biology. 2nd Edn.
ompF gene that codes for a major outer membrane protein of Washington, DC: ASM Press, 1996; 123–45.
Escherichia coli K-12. Nucleic Acids Res 1982; 10: 6957 –68. 42 Paulsen IT, Park JH, Choi PS et al. A family of Gram-negative bacterial
31 Iyer R, Delcour AH. Complex inhibition of OmpF and OmpC bacterial outer membrane factors that function in the export of proteins,
porins by polyamines. J Biol Chem 1997; 272: 18595– 601. carbohydrates, drugs and heavy metals. FEMS Microbiol Lett 1997; 156:
32 Aronoff SC. Outer membrane permeability in Pseudomonas cepacia: 1 –8.
diminished porin content in a b-lactam-resistant mutant and in 43 Cowan SW, Garavito RM, Jansonius JN et al. The structure of OmpF

Downloaded from https://academic.oup.com/jac/article/65/9/1926/722338 by guest on 19 February 2023

resistant cystic fibrosis isolates. Antimicrob Agents Chemother 1998; 32: porin in a tetragonal crystal form. Structure 1995; 3: 1041– 50.
1636– 9. 44 Sigumura K, Nishihara T. Purification, characterization, and primary
33 Toro CS, Lobos SR, Calderon I et al. Clinical isolate of a porin-less structure of Escherichia coli protease VII with specificity for paired basic
Salmonella typhi resistant to high levels of chloramphenicol. Antimicrob residues: identity of protease VII and OmpT. J Bacteriol 1988; 170:
Agents Chemother 1990; 34: 1716 –9. 5625– 32.
34 Burns JL, Smith AL. A major outer-membrane protein functions as a 45 Goldstein FW, Gutmann L, Williamson R et al. In vivo and in vitro
porin in Haemophilus influenzae. J Gen Microbiol 1987; 133: 1273 –8. emergence of simultaneous resistance to both b-lactam and
35 Fitzmaurice J, Glennon M, Duffy G et al. Application of real-time PCR aminoglycoside antibiotics in a strain of Serratia marcescens. Ann
and RT–PCR assays for the detection and quantification of VT1 and VT Microbiol 1983; 134A: 329–37.
2 toxin genes in E. coli O157:H7. Mol Cell Probes 2004; 18: 123–32. 46 Ranquin A, Van Gelder P. Maltoporin: sugar for physics and biology.
36 Li H, Lin XM, Wang SY et al. Identification and antibody-therapeutic Res Microbiol 2004; 155: 611–6.
targeting of chloramphenicol-resistant outer membrane proteins in 47 Zhang DF, Jiang B, Xiang ZM et al. Functional characterization of
Escherichia coli. J Proteome Res 2007; 6: 3628– 36. altered outer membrane proteins for tetracycline resistance in
37 Zhai Y, Saier MH Jr. The b-barrel finder (BBF) program, allowing Escherichia coli. Int J Antimicrob Agents 2008; 32: 315–9.
identification of outer membrane b-barrel proteins encoded within 48 Stumpe S, Schmid R, Stephens DL et al. Identification of OmpT as
prokaryotic genomes. Protein Sci 2002; 11: 2196– 207. the protease that hydrolyzes the antimicrobial peptide protamine
38 Mizuno T, Wurtzel ET, Inouye M. Cloning of the regulatory genes before it enters growing cells of Escherichia coli. J Bacteriol 1998; 180:
(ompR and envZ) for the matrix proteins of the Escherichia coli outer 4002– 6.
membrane. J Bacteriol 1982; 150: 1462– 6. 49 Kanamaru S, Kurazono H, Ishitoya S et al. Distribution and genetic
39 Black PN. The fadL gene product of Escherichia coli is an outer association of putative uropathogenic virulence factors iroN, iha, kpsMT,
membrane protein required for uptake of long-chain fatty acids and ompT and usp in Escherichia coli isolated from urinary tract infections
involved in sensitivity to bacteriophage T2. J Bacteriol 1988; 170: 2850– 4. in Japan. J Urol 2003; 170: 2490 –3.

1935