Unveiling the Expression Characteristics of IspC, a Cell Wall-Associated Peptidoglycan Hydrolase in Listeria monocytogenes, duringGrowth under Stress Conditions

Jennifer Ronholm,a,b Xudong Cao,c and Min Lina,b

Canadian Food Inspection Agency, Ottawa Laboratory Fallowfield, Ottawa, Ontario, Canadaa; Department of Biochemistry, Microbiology and Immunology, University ofOttawa, Ottawa, Ontario, Canadab; and Department of Chemical and Biological Engineering, University of Ottawa, Ottawa, Ontario, Canadac

Listeria monocytogenes serotype 4b is a food-borne pathogen of public health concern, since it accounts for approximately 40%of human listeriosis cases. We have recently identified IspC, a surface-localized peptidoglycan hydrolase, as the antigen recog-nized by a number of monoclonal antibodies (MAbs) produced against a serotype 4b strain for diagnostic applications. To deter-mine whether IspC, which is well conserved among various serotype 4b strains, is a useful diagnostic marker in antibody-basedmethods, we assessed the expression of IspC in L. monocytogenes cultured under normal and stress conditions. A functional pro-moter directing the transcription of the ispC gene was identified upstream of the ispC open reading frame by constructing a pro-moterless lacZ gene fusion with the putative ispC promoter region and by 5= rapid amplification of cDNA ends analysis. Usingboth the lacZ reporter gene system and immunofluorescent staining with an IspC-specific MAb, we provide evidence that IspC isexpressed on the cell surface in all growth conditions tested (temperature, osmotic stress, pH, ethanol, oxidative stress, anaero-bic conditions, carbon source, and type of growth media) that allow for cellular division, although the level of ispC gene expres-sion varies. These results demonstrated the usefulness of IspC as an excellent diagnostic marker for the serotype 4b strains andimply that IspC, in conjunction with specific MAbs, can be targeted for detection and isolation of L. monocytogenes serotype 4bstrains directly from food, environmental, and clinical samples with minimal or no need for culture enrichment.

Listeria monocytogenes is a Gram-positive pathogenic bacteriumthat can lead to listeriosis, a life-threatening opportunistic in-

fection caused by the ingestion of contaminated foods. Clinicaloutcomes of listeriosis range from asymptomatic infection tononspecific flu-like symptoms, gastroenteritis, septicemia, men-ingitis, and fetal infection followed by abortion in pregnantwomen (24). In the environment, L. monocytogenes is extremelyhardy and actively divides between 3 and 45°C (26), in up to 10%salt (16), at a pH of between 4.4 and 9.2 (5), and under anaerobicconditions (15). The ability of L. monocytogenes to grow in ex-treme environments makes it a concern for the food industry,particularly in food-processing plants where ready-to-eat foodsare prepared.

L. monocytogenes is divided into 13 serotypes; however, 98% ofhuman illness is caused by serotype 1/2a, 1/2b, and 4b strains (8).Serotype 4b strains account for more cases of human listeriosisthan serotype 1/2a and 1/2b isolates combined, although 1/2a and1/2b strains are much more commonly found in foods and theenvironment (24, 25). This suggests that serotype 4b strains arespecifically adapted to infecting human hosts (24, 25). Serotype 4bstrains are also more often isolated from patients with meningo-encephalitis than from patients where the infection has been lim-ited to the bloodstream (24). Listeriosis patients also suffer a 26%mortality rate when infected with a serotype 4b strain compared toa 16% mortality rate in patients infected with a serotype 1/2a or1/2b strain (6). These observations suggest that serotype 4b ismore virulent in humans than other serotypes. Therefore, the de-velopment of a diagnostic test specific for L. monocytogenes sero-type 4b strains is important.

Current gold standard methods for isolating and detecting L.monocytogenes are culture based and labor intensive and take 5 to7 days. Molecular methods, such as PCR, have been developed to

expedite the detection of L. monocytogenes in food samples buthave some disadvantages. They still require the culture enrich-ment steps prior to detection. Inhibitory substances and back-ground bacteria present in food samples can confound the PCRresults tremendously. In addition, determining the viability of anorganism, which is important in a food recall, is impossible withmolecular methods. Antibody-based methods are promising toovercome these drawbacks for rapid isolation and detection of L.monocytogenes from food and environmental samples. Severalmonoclonal antibodies (MAbs) which react specifically with L.monocytogenes serotype 4b were developed by our laboratory (13)and have been shown to recognize a surface-localized autolysin,IspC (homologous to LMOf2365_1093) (J. Ronholm, H. vanFaassen, R. MacKenzie, Z. Zhang, X. Cao, and M. Lin, unpub-lished data). These MAbs, together with IspC as a surface maker,have the potential for use in diagnostic tests for L. monocytogenesserotype 4b strains. The surface expression of IspC at a level allow-ing these specific antibodies to bind L. monocytogenes cells origi-nating from various growth conditions is critical to success inculture-independent, antibody-based detection methods and re-mains to be assessed. Surface protein expression is unstable andgenerally dependent upon growth conditions (3, 4, 19). The vari-ability of surface epitope expression has also been previouslyshown to limit the usefulness of antibodies in L. monocytogenes

Received 28 June 2012 Accepted 5 August 2012

Published ahead of print 24 August 2012

Address correspondence to Min Lin, Min.Lin@inspection.gc.ca.

Copyright © 2012, American Society for Microbiology. All Rights Reserved.

doi:10.1128/AEM.02065-12

November 2012 Volume 78 Number 22 Applied and Environmental Microbiology p. 7833–7840 aem.asm.org 7833

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from

detection (19). Therefore, the objective of this work was to char-acterize the surface expression of IspC during growth under vari-ous physical and chemical stresses and to determine if IspC is auseful marker for the development of antibody-based detectionmethods. By analyzing the activity of the ispC promoter in re-sponse to various environmental stresses with the lacZ reporterand the surface expression of IspC with immunofluorescence mi-croscopy, we provide evidence that in a wide range of environ-mental conditions that allow L. monocytogenes cells to activelydivide, IspC is expressed on the surface at high enough levels forimmunological detection.

MATERIALS AND METHODSBacterial strains and monoclonal antibodies. L. monocytogenes serotype4b LI0521, a human clinical isolate, was used in this study and was grownin either brain heart infusion (BHI) broth (BD Diagnostics) or on BHIagar plates at 37°C and supplemented with 10 �g/ml kanamycin or 50�g/ml erythromycin as required for bacteria transformed with pTCV-PispC. L. monocytogenes cell concentrations were estimated as previouslydescribed (14). The Escherichia coli strain DH5� was used for plasmidpropagation. E. coli was cultured in Luria-Bertani (LB) media supple-mented with 50 �g/ml kanamycin or 150 �g/ml erythromycin as re-quired.

The M2773 monoclonal antibody in tissue culture fluid (TCF), whichwas previously shown to specifically interact with the IspC protein (Ron-holm et al., unpublished), was used for immunofluorescent staining.

5= RACE analysis. The transcription start site for the ispC gene wasdetermined by 5= rapid amplification of cDNA ends (RACE) analysis us-ing total RNA derived from L. monocytogenes. All solutions, water, glass-ware, and utensils used for RNA extraction were treated with 0.1% diethyl-pyrocarbonate (DEPC) (Sigma) overnight at 37°C and autoclaved. TotalRNA was extracted from 1.5 ml mid-log-phase L. monocytogenes culture.Bacterial cells were lysed after being treated with RNAprotect bacterialreagent (Qiagen) per the supplier’s instructions by digestion with 13,000U of lysozyme (Sigma) in 0.1 ml of Tris-EDTA (TE) buffer for 30 min at37°C, followed by mechanical disruption with the FastPrep system andLysing Matrix B (MP Biomedicals) according to the manufacturer’s in-structions. RNA was purified from the cell lysate with the RNeasy minikit(Qiagen) per the supplier’s instructions. The integrity of the RNA samplewas confirmed by agarose gel electrophoresis. 5= RACE was carried out, toidentify the transcriptional start site, according to the manufacturer’s in-structions (Roche) using IspC gene-specific primers (Table 1).

Construction of PispC-lacZ transcriptional fusions. The pTCV-lacplasmid vector (21), a low-copy-number and broad-host-range plasmidthat contains a promoterless lacZ gene, was used to evaluate the ispCpromoter activity in various growth conditions. A 454-bp DNA fragmentcontaining the ispC promoter was amplified by PCR using primer pairP920/P921 (Table 1) from L. monocytogenes genomic DNA. After diges-

tion with BamHI and EcoRI (New England BioLabs), the amplified pro-moter region was cloned into BamHI and EcoRI sites of the pTCV-lacplasmid, resulting in the recombinant plasmid pTCV-PispC. The recom-binant plasmid was sequenced with the Vlac primers (P918 and P919) toverify the inserted sequence. pTCV-PispC was subsequently introducedinto competent L. monocytogenes by electroporation as described previ-ously (28).

Growth conditions. Various growth conditions were evaluated fortheir effects on ispC gene expression, including low temperature (4°C),high temperature (42°C), salt concentration (1 to 10%), acidity (pH 4),alkaline growth (pH 10), sublethal concentrations of ethanol (2 to 5%),oxidative media, anaerobic growth, alternative carbon sources, and vari-ous Listeria enrichment media. L. monocytogenes harboring pTCV-PispCwas grown from frozen stock bacteria on a BHI agar plate and subse-quently stored at 4°C. A single colony was inoculated into BHI broth andincubated for 16 to 18 h at 37°C. A 1:100 dilution of the overnight culturewas subcultured into the media associated with each test condition (seebelow). As a control, L. monocytogenes containing the promoterless plas-mid pTCV-lac was also tested under the same conditions. Each conditionwas examined using triplicate cultures. In addition, each condition wasindependently examined a minimum of 3 times, therefore a minimum of9 measurements was recorded for each condition. Samples were taken andassayed for the �-galactosidase activity at 150 min, 350 min, and 24 h toprovide a detailed look at when IspC expression occurs, except for the 4°Cculture, where samples were collected after 168, 336, and 504 h of growth.

For temperature evaluation, BHI media were each equilibrated to thepredetermined temperatures (4, 37, and 42°C) prior to bacterial inocula-tion.

The effects of various salt concentrations were examined over a 24-hperiod by supplementing BHI broth with 1 to 10% NaCl (wt/vol) at 1%intervals. The effects of pH were monitored by buffering BHI broth withHCl or NaOH to the desired pH (pH 2, 4, 6, 8, 10, or 12). The effect ofsublethal concentrations of ethanol was assessed by adding the solventdirectly to BHI broth to a final concentration of 2, 5, or 10% (vol/vol). Theeffects of highly oxidative conditions on IspC expression were tested byadding 7 mM cumene hydroperoxide (CHP) (Sigma) to BHI broth. An-aerobic conditions were created by autoclaving BHI in serum bottles. Thebottles were sealed and the oxygen was removed by 5 cycles of vacuumingand then bubbling with a gaseous mix of 80% N2, 10% CO2, and 10% H2.Resazurin was added at 0.005% (vol/vol) to verify that the medium wasanaerobic. To determine if the carbon source affected ispC expression, L.monocytogenes was grown in the minimal media developed by Premaratneet al. (22) and supplemented individually with 20% (wt/vol) glucose,mannose, or fructose. Selective enrichment media have been shown toaffect the antigen expression in L. monocytogenes (3 and Ronholm et al.,unpublished). Therefore, the effects of various growth media on IspCexpression were examined with three common enrichment culturebroths: University of Vermont modified enrichment broth (UVM) (BDDiagnostics), Fraser broth (Oxiod), and Palcam broth (Oxoid).

�-Galactosidase assay. �-Galactosidase assays were carried out essen-tially as described by Miller (18), with the exception that cells were col-lected by centrifugation for 2 min at 16,100 � g and resuspended in theassay buffer (0.06 M Na2HPO4, 0.04 M NaH2PO4, 0.01 M KCl, 0.001 MMgSO4, and 0.05 M �-mercaptoethanol at pH 7) to remove backgroundfrom the growth media prior to the assay. Enzyme activity was expressedin Miller units, defined as 1,000 � [OD420 � (1.75 � OD550)]/(incuba-tion time � volume of culture � OD600), where OD420 is the opticaldensity at 420 nm. As a negative control, L. monocytogenes was trans-formed with an empty pTCV-lac plasmid, and this was grown side by sidewith test samples. The number of Miller units used for the negative con-trol was always less than 5. �-Galactosidase activity analyses were carriedout in three independent experiments, each using triplicate samples, toensure reproducibility. Results were analyzed for statistical significancewith a Mann-Whitney U test, and statistical significance was defined asP � 0.001.

TABLE 1 Oligonucleotide primers used in this study

Primer Nucleotide sequence

P998 (cDNAsynthesis)

5=-CTTTGTTGTAGCAACAGATACTA-3=

P995 (GSP1) 5=-GAGGCTCCATTGCAGTTACTTTA-3=P996 (GSP2) 5=-GCGGAATTCAGCATCAATTTTAA-3=P997 (GSP3) 5=-CCCCAACCAGATTCTAGAATTCG-3=P920 5=-AGAGAATTCAAAATATCAAAAAGAGCATAA-3=a

P921 5=-CGCGGATCCAATTTGTTTATTGTCCTAATT-3=b

P918 (Vlac1) 5=-GTTGAATAACACTTATTCCTATC-3=P919 (Vlac2) 5=-CTTCCACAGTAGTTCACCACC-3=a The EcoRI restriction site is underlined.b The BamHI restriction site is underlined.

Ronholm et al.

7834 aem.asm.org Applied and Environmental Microbiology

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from

Immunofluorescent staining. Immunofluorescent staining was car-ried out essentially as described previously (14). Briefly, L. monocytogeneswas grown in each of the tested growth conditions for 24 h (504 h for the4°C culture) and the cells collected by centrifugation for 2 min at 16,100 �g to remove the culture supernatant. Cell pellets were resuspended inphosphate-buffered saline (PBS) containing 5% bovine serum albumin(BSA) for 1 h at room temperature to block nonspecific protein bindingsites. Cells were then allowed to react for 1 h with the MAb M2773 in tissueculture fluid at a dilution of 1:50 in PBS containing 5% BSA. After washingtwice with PBS, cells were incubated with a 1:2,000 dilution of Dylight488-conjugated goat anti-mouse IgG(H�L) (Jackson ImmunoResearch)in PBS containing 5% BSA for 1 h. Cells were washed 3 times with PBS andresuspended in PBS. Cells were examined with a fluorescence microscopefirst under phase-contrast and then fluorescence settings. Fluorescent im-ages were captured with a charge-coupled device (CCD) camera using theQCapture Pro software (Q Imaging) with a 3-s exposure time.

RESULTSIdentification of the ispC gene promoter and the TSS. The se-quence upstream of the ispC open reading frame (ORF) was ex-amined for the presence of regulatory elements. It should be notedthan an ORF (LMOF2365_1092) for a putative teichoic acid ABCtransporter, transcribed in the same direction as ispC, ends 224 bpupstream of the ispC start codon. Sequence analysis with a pro-karyotic promoter prediction algorithm (www.fruitfly.org) re-vealed elements of two putative promoters and correspondingpredicted transcriptional start sites (TSSs) with a perfect score of1.0 within a region of 400 bp upstream of the ispC ORF. To deter-mine if either of these putative promoters, or an alternative pro-moter within this region, was active, a 454-bp DNA fragment im-mediately upstream of the ispC translation start codon was clonedinto the pTCV-lac plasmid (21) containing a promoterless lacZreporter gene, making pTCV-PispC. Approximately 40 Millerunits of �-galactosidase activity were detected at mid-log phase(350 min) of L. monocytogenes(pTCV-PispC), which is serotype 4bstrain LI0521 that had been transformed with pTCV-PispC. Thisindicated that a functional ispC promoter exists within the 454-bpregion immediately upstream of the ispC ORF.

To define the location of the ispC promoter more precisely, wemapped the TSS using 5= RACE to an adenine nucleotide 31 bpaway from the ispC translation start codon (Fig. 1). Surprisingly,this was not one of the two TSSs predicted for two putative pro-moters having a perfect score of 1.0. However, within this func-tional promoter region another promoter was predicted with ascore of 0.73, containing a predicted TSS one nucleotide awayfrom the experimentally determined TSS. Corresponding to thisexperimentally determined TSS, there are �10 (TGGTAAAAT)and �35 (TTGTTA) elements spaced 19 bp apart, as predicted bya bacterial promoter program (Softberry) (Fig. 1). Examination ofthe sequence in this functional promoter region did not identify

consensus sequence elements typical for a sigma B transcriptionfactor (�35 GTTT and �10 GGGnAn [where n represents anynucleotide]) (10) in L. monocytogenes or for a PrfA box (TTAA-CAnnTGTTAA) (25), which is the binding site of the virulencefactor transcriptional activator PrfA.

Effect of temperature on ispC expression. The ispC expressionin response to growth temperature (at 37, 42, and 4°C) was inves-tigated by assessing the �-galactosidase activity under the controlof the ispC gene promoter with L. monocytogenes(pCTV-PispC)(Fig. 2). The enzyme activity showed a significant decrease at 350min and 24 h of growth at 42°C compared to that of the culture at37°C (P � 0.001) (Fig. 2A). Because of a long generation time at4°C, the bacteria were allowed to grow for an extended period inorder to obtain sufficient numbers of cells for the accurate mea-surement of the �-galactosidase activity. The enzyme activity wassubstantially reduced when cultured at 4°C compared to that at37°C. These results indicate that the ispC gene expression is sub-jected to regulation by temperature. In spite of a low level of ispCgene expression at 24 h of growth at 42°C or at 504 h at 4°C(Fig. 2A), the IspC protein was clearly detectable on the surface byimmunofluorescence microscopy with the anti-IspC MAb,M2773 (Fig. 2B to E).

Osmotic stress and ispC expression. The expression of ispC inresponse to growth in high salt conditions (1 to 10% NaCl) wassimilarly investigated using the L. monocytogenes(pCTV-PispC)

FIG 1 ispC transcription start site. Using 5= RACE, the transcription start siteof IspC was identified to be an adenine residue located 31 bp upstream fromthe translation start codon. The approximate �10 and �35 regions, as pre-dicted by BPROM, are also indicated in boldface italics.

FIG 2 Temperature-dependent regulation of ispC expression. The ispC pro-moter activity was measured in L. monocytogenes(pTCV-PispC) using a �-ga-lactosidase assay. (A) After 150 min of growth at 42°C, there was no change inpromoter activity compared to growth at 37°C at the same time point. How-ever, after 350 min and 24 h of growth at 42°C, IspC expression was signifi-cantly less (P � 0.001) than that in cells grown at 37°C. IspC expression in cellsgrown at 4°C was shown to be very low. Statistically significant changes, in thisand other figures, of promoter activity are indicated by two asterisks. In Fig. 2to 8, the bars indicate the median value of all observations (n � 9), and theerror bars are used to indicate the interquartile deviation. Cells grown for 24 hat 42°C were visualized using phase contrast microscopy (B) and were detect-able by immunofluorescence microscopy with an anti-IspC MAb (C). Cellsgrown for 504 h at 4°C were visualized using phase contrast microscopy (D)and were also detectable by immunofluorescence microscopy (E).

Regulated Expression of IspC

November 2012 Volume 78 Number 22 aem.asm.org 7835

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from

cells. After 150 min of growth, cells grown in 4 and 5% (wt/vol)sodium chloride had statistically significant reductions in enzymeactivity (P � 0.001) compared to cells grown in BHI broth withoutadditional salt. Cultures grown in the presence of 6 to 10% (wt/vol) NaCl did not show detectable levels of �-galactosidase activityat the 150-min time point (Fig. 3A). Cells grown for 350 min in 5to 10% (wt/vol) NaCl had significant ispC expression (P � 0.001).After 24 h of growth, cultures supplemented with 3 to 10% (wt/vol) NaCl had significantly reduced enzyme activity (P � 0.001)(Fig. 3A). This indicated that the ispC gene expression is subjectedto regulation by osmotic stress. Despite significantly decreasedispC promoter activity in osmotically stressed cells, the IspC pro-tein was detectable on the surface of cells after 24 h of growth inBHI containing various salt concentrations by immunofluores-cence microscopy with M2773, although the fluorescence signaldecreased in a NaCl concentration-dependent manner (Fig. 3B toI). Cell morphological changes, such as chaining, started to occurin bacteria grown at 5% (wt/vol) NaCl and became more promi-nent as the salt concentration increased, although chained cellswere still able to express the surface IspC, which was detectable byimmunofluorescence microscopy with M2773.

Minimal effect of culture pH on ispC expression. L. monocy-togenes has been reported to grow in the pH range of 4.4 to 9.2 (5);

however, our preliminary experiments indicated that cellular di-vision does not occur below pH 6 or above pH 10, and the activityof �-galactosidase was investigated in L. monocytogenes(pTCV-PispC) only within the pH range of 6 to 10. At the 150-min timepoint, the activity of the enzyme expressed under the control ofthe ispC promoter was not affected by the culture medium pH.However, cultures at pH 6 and 10 had significantly lower enzymeactivity than the positive control (pH 7.6) (P � 0.001) at the 350-min time point (Fig. 4A). At 24 h there was no difference in theenzyme activity between cultures at different pHs (Fig. 4A). Con-sistent with these results is that after 24 h of growth, surface ex-pression of IspC was detected by immunofluorescence micros-copy in cells cultured under each condition, although qualitativeobservations indicated that growth in alkaline conditions makescells easier to detect with the anti-IspC MAb (Fig. 4B to G).

Effect of sublethal ethanol concentrations on ispC expres-sion. �-Galactosidase activity was assessed in L. monocytogenes(pTCV-PispC) cultivated in BHI broth containing sublethalethanol concentrations (2, 5, and 10%, vol/vol). Ethanol at a con-centration of 10% (vol/vol) was inhibitory to cell division, there-fore this condition was eliminated from the study. Although gen-eration time was increased for bacterial cells in the presence of

FIG 3 Sodium chloride downregulation of ispC expression. The ispC pro-moter activity was measured in Miller units at 1 to 10% NaCl. Concentration-dependent activity suppression was observed at all time points. (A) Statisticallysignificant decreases in ispC gene expression (P � 0.001) were found for sev-eral conditions compared to cells at the same time point grown in BHI withoutadditional salt (positive control). Although cellular growth and division werealso suppressed by osmotic stress, the IspC protein was detectable on the cellsurface by immunofluorescence in 1% (C), 6% (E), 7% (G), and 10% (I) NaCl.Panels B, D, F, and H show phase contrast images to demonstrate the presenceand abundance of L. monocytogenes cells. Significant cellular chaining wasapparent in the presence of 5% NaCl.

FIG 4 pH has a minimal effect on ispC expression. Cultures with a pH outsidethe range 6 to 10 could not actively divide within the time periods for thisexperiment. (A) Within the pH 6 to 10 range, ispC expression was relativelyconstant, with statistically different (P � 0.001) expression only occurring at350 min postinoculation at pH 6 and 10. At 24 h of growth there was nosignificant difference in ispC expression between any of the cultures. IspC wasalso detectable on the cell surface by immunofluorescence at pH 6 (C), pH 8(E), and pH 10 (G). Corresponding phase contrast images are shown in panelsB, D, and F for comparison.

Ronholm et al.

7836 aem.asm.org Applied and Environmental Microbiology

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ethanol, the level of ispC expression was not significantly affected(P 0.001) by the ethanol concentration used at any of the timepoints (Fig. 5A). Therefore, expression of the ispC gene is notregulated by exposure to sublethal ethanol, and the presence of theIspC protein in cells grown in 2 and 5% ethanol was easily detectedby immunofluorescence microscopy using the anti-IspC MAbM2773 (Fig. 5B and E).

Effect of oxidative and anaerobic growth conditions on ispCexpression. L. monocytogenes(pTCV-PispC) was grown in CHP toprovide highly oxidative conditions and also under anaerobicconditions before being analyzed for �-galactosidase activity (Fig.6). Previous studies have shown that L. monocytogenes survives in13.8 mM CHP (2); however, our preliminary experiments indi-cated that L. monocytogenes cannot actively divide in CHP concen-trations as low as 7 mM; therefore, this condition was eliminatedfrom the study. Similar levels of enzyme activity were observed at150- and 350-min time points for cell cultures under both aerobicand anaerobic conditions, revealing no effect of these conditionson the expression of ispC. However, after 24 h of growth, the ispCpromoter showed more activity in aerobically cultured cells thanin anaerobic conditions (P � 0.001) (Fig. 6A). Bacterial cellsgrown in both aerobic and anaerobic conditions express the IspCprotein and are easily detectable by using immunofluorescencemicroscopy with the M2773 antibody (Fig. 6B to E).

Carbon source does not regulate ispC expression. The ex-pression of the ispC gene was examined in response to three car-bon sources (glucose, fructose, and mannose). Carbon source has

FIG 5 Sublethal dose of ethanol does not affect ispC expression. (A) Growth inthe presence of 2 or 5% ethanol did not significantly (P � 0.001) affect ispCexpression compared to the culture in the presence of 0% ethanol. IspC isreadily detectable on the cell surface of L. monocytogenes cells grown in 2% (C)and 5% (E) ethanol. Phase contrast images are shown in panels B and D forcomparison.

FIG 6 ispC is expressed in anaerobic conditions. (A) ispC expression was measured from L. monocytogenes(pTCV-PispC) grown in an anaerobic culture andcompared to cells grown in an aerobic culture. At 350 min after inoculation, there were no differences in ispC expression. Although ispC was also expressed at 24h after inoculation, levels of expression were significantly (P � 0.001) lower in the anaerobic culture than in the aerobic culture. However, IspC was detectableon the cell surface 24 h after inoculation of both aerobic (C) and anaerobic (E) growth media. The same field of view is shown in panels B and D, respectively, inphase contrast to allow for comparison.

Regulated Expression of IspC

November 2012 Volume 78 Number 22 aem.asm.org 7837

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from

been cited as having a major effect on the expression of variousproteins in L. monocytogenes (17, 22). L. monocytogenes(pTCV-PispC) showed similar growth (data not shown) and �-galactosi-dase activity in the minimal media supplemented with each car-bon source (Fig. 7A). Therefore, the ispC gene expression is notregulated by carbon source. Cells grown with each carbon sourcewere also easily detected by the presence of the IspC protein byimmunofluorescence microscopy (Fig. 7B to G).

Effect of various L. monocytogenes enrichment media onispC expression. Several types of media are routinely used for theselective enrichment of L. monocytogenes in most Listeria methodsprior to bacterial isolation and detection. To expand the potentialof MAbs which recognize IspC beyond direct culture-indepen-dent detection into postenrichment detection, the expression ofthe ispC gene, as assessed by the activity of �-galactosidase tran-scribed by the ispC promoter in L. monocytogenes(pTCV-PispC),was examined over the span of the growth curve in four media,including BHI broth (Fig. 8A), Fraser broth (Fig. 8B), Palcambroth (Fig. 8C), and UVM (Fig. 8D). Although the levels of �-ga-lactosidase activity varied between media, it was detectable in all ofthe media used. Expression of the ispC gene was significantlyhigher (P � 0.001) at all time points when cells were grown in BHIbroth than when they were grown in UVM. ispC gene expression isthe same in cells grown in Fraser and Palcam broths (P 0.001)except for at the 750-min time point, when the ispC gene is ex-pressed more in Fraser broth (P � 0.001). The activity of �-galac-tosidase under the control of the ispC promoter is generally higherin cells grown in BHI broth than in cells grown in Fraser andPalcam broths, although significance varies between time points.IspC was present on the cell surface at high enough levels to allowcells to be easily detected by immunofluorescence microscopy af-ter 24 h of growth regardless of growth media (Fig. 8E to J).

FIG 7 Carbon source regulation of ispC expression. (A) ispC was expressed atthe same level in minimal media when supplemented individually with a glu-cose, fructose, or mannose carbon source. L. monocytogenes cells are also de-tectable by an anti-IspC antibody when grown in glucose (C), fructose (E), ormannose (G). In addition, phase contrast images of cells grown in glucose (B),fructose (D), or mannose (F) are shown for comparison.

FIG 8 ispC expression in various enrichment media. The �-galactosidase activity under the control of the IspC promoter was measured in Miller units in L. monocytogenes cellsgrown from inoculation to lag phase in BHI broth (A), Fraser broth (B), Palcam broth (C), and UVM (D). A clear peak in promoter activity during mid-log phase could be seenonly in BHI broth. The presence of the protein IspC on the surface of cells grown in Fraser broth (F), Palcam broth (H), and UVM broth (J) was detectable after 24 h of growthby immunofluorescence microscopy. The phase-contrast images of the same field of view are shown in panels E, G, and I, respectively, for comparison.

Ronholm et al.

7838 aem.asm.org Applied and Environmental Microbiology

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from

DISCUSSION

In this study, we characterized the expression characteristics ofIspC, a surface-associated peptidoglycan hydrolase, through theuse of the lacZ reporter gene system and immunofluorescencemicroscopy in order to assess the value of IspC as a diagnosticmarker for L. monocytogenes serotype 4b. A functional IspC pro-moter was active in L. monocytogenes and was identified immedi-ately upstream of the ispC translation start site. This promoterremained active in ispC gene transcription in all environmentalconditions tested which are conducive to bacterial cell division;however, the levels of its activity varied. We have further demon-strated that L. monocytogenes serotype 4b cells are capable of dis-playing IspC on the cell surface in each of the tested conditions,making cells detectable by immunofluorescence microscopy us-ing an anti-IspC MAb. Even in an extreme stress condition, wherethere was only weak expression of IspC, there was still a sufficientamount of IspC localized on the cell surface to allow detectionusing the anti-IspC MAb. We recently have demonstrated thatIspC is highly conserved among L. monocytogenes serotype 4b iso-lates but not in other serotypes (Ronholm et al., unpublished).Since IspC is unique to serotype 4b isolates and its expression isdetectable under normal or stress growth conditions, it can serveas an excellent diagnostic marker, with the aid of anti-IspC anti-bodies, for detection or isolation of this important serotype of L.monocytogenes.

Transcription of IspC started from a TSS 31 bp upstream of thetranslation start site under the direction of its own promoter (Fig.1). Expression of ispC appears to peak, when bacterial cells aregrown in BHI broth, during mid-log phase (350 min) (Fig. 8A).This finding correlated well with the result of a previous studywhich showed by reverse transcription-PCR (RT-PCR) that theIspC gene was transcribed at the highest levels during early logphase (27). The IspC protein was detectable in all growth condi-tions examined in this study; however, particularly low levels ofexpression were observed during growth in BHI broth supple-mented with 10% (vol/wt) NaCl (Fig. 3) or during growth at 4°C(Fig. 2).

An anti-IspC MAb was able to detect the target protein on thecell surface by immunofluorescence microscopy after 24 h ofgrowth (504 h of culture at 4°C) regardless of growth conditions.These findings were an important addition to the ispC promoteractivity analysis, which suggested that IspC can serve as a gooddiagnostic marker because gene expression is always observed inL. monocytogenes. In contrast, other studies conducted to deter-mine which L. monocytogenes growth conditions allow for expres-sion of the antigens targeted by various antibodies did not use cellsgrown entirely under a stress condition (3, 7). In these studies,cells were grown either in a nonstressed condition followed bysubculturing for a limited time in the stress condition (7) or in astress environment and then rescued in an enrichment broth (3).The problem with growing cells for a limited time in the stresscondition followed by transfer to enrichment broth is that residualprotein from the original culture does not get degraded quicklyenough to disappear after the short incubation time and thereforeis still detected (Ronholm et al., unpublished). Our findings thatIspC was detectable with this specific MAb in all growth condi-tions tested makes this protein antigen a novel potential diagnos-tic marker for diagnostics for L. monocytogenes serotype 4b, sincestudies which have used a methodology similar to the one pre-

sented here have found that their antigens were not expressed atdetectable levels under all tested conditions (4, 11, 19, 20). Sincethe primary objective of this study was to validate IspC as a markerfor diagnostic use in pre-enrichment bacterial detection, possiblyin a biosensor in a microfludics-based flow cytometer, our meth-odology, where cells were grown entirely in the stress conditionand detected by immunofluorescence signal, allowed us to exam-ine IspC expression in cells from selected environments underdetection conditions simulating those that would be present in abiosensor applied to the detection of L. monocytogenes directlyfrom food or environmental samples.

L. monocytogenes was shown to have chained morphologywhen subjected to extreme osmotic stress (Fig. 3). This findingwas in agreement with other studies (4, 9). Cellular chaining indi-cates that the autolysins involved in cell division either are notexpressed or are not active under these conditions.

An anti-IspC MAb was able to detect L. monocytogenes cellsgrown in each of the selective enrichment broths, although therewas variation in the activity of the ispC promoter between broths(Fig. 8A to D). These experiments were carried out to assess if IspCcan be detected by anti-IspC antibodies on the surface of L. mono-cytogenes after enrichment. Selective enrichment broths have beenshown to affect the expression of several surface antigens (3, 11,19, 20). Similar studies, using different MAbs (EM-7G1 andC11E9) which also recognize an Listeria autolysin (lmo2691) in L.monocytogenes serotype 1/2a, found that growth in UVM entirelyinhibits the expression of the antigen, rendering these antibodiesuseless for detection when enrichment takes place in this medium(3, 19, 20). UVM was also found to reduce the expression of ActA,making detection impossible using an anti-ActA antibody afterUVM enrichment (11). In agreement with these findings, we alsoshowed that expression of IspC was more reduced in UVM than inany other media. However, despite weak activity of the ispC pro-moter, cells grown in UVM were still detectable by immunofluo-rescence microscopy with an anti-IspC MAb. This is likely becauseIspC is a very stable protein (Ronholm et al., unpublished) (27),and even minor expression during any growth phase leads to sur-face protein accumulation. The anti-IspC MAb was also able todetect cells grown in Fraser broth almost as well as cells grown inBHI broth. The EM-7G1 and C11E9 antibodies were unable todetect cells grown in Fraser broth (3, 19).

IspC is a peptidoglycan hydrolase with N-acetylglucosamini-dase activity (23). It is involved in virulence, as evidenced by theattenuation seen in the ispC deletion mutant (28). Elucidatingthe mechanisms behind regulation of virulence-associated auto-lysins is an important step in determining their role in pathogen-esis, which appears to be separate from their roles in cellular divi-sion or growth (12). We are very interested in the biologicalsignificance of the IspC autolysin, which appears to be serotype 4bspecific and is expressed over a broad range of environmentalconditions. The ispC promoter appears to be most active duringearly log phase. In addition, we found that ispC is always expressedin cells which are able to divide. These two observations suggestthat IspC has a role in cell division. However, previous work indi-cates that IspC is not directly involved in cellular division, sinceispC knockout mutants do not have the characteristic chainingphenotype that is seen in cells where essential division moleculeshave been knocked out (28). One possibility is that IspC has a rolein cell division; however, a phenotype is not observed in the dele-tion mutant because another autolysin with an overlapping func-

Regulated Expression of IspC

November 2012 Volume 78 Number 22 aem.asm.org 7839

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from

tion is able to functionally compensate for the ispC knockout.IspC expression is also highly upregulated during rabbit infection(30), and based on the rate of cell division proportionate to itsupregulation, it is likely that there are requirements for IspC dur-ing infection besides those related to growth and division. The roleof IspC in infection is not currently understood but is of interest.

Food-borne bacteria encounter a variety of environmentalstresses which alter surface protein expression, and this affects ourability to detect bacteria using antibody-based methods (19). Sev-eral studies have demonstrated that the ability of polyclonal andmonoclonal antibodies to detect Listeria is highly dependent onculture conditions (3, 4, 7, 11, 19, 20, 29). Ideally, next-generationrapid detection techniques will allow for specific pathogen detec-tion without the need for enrichment. Developing technologies,such as flow cytometry, immunomagnetic separation, and biosen-sors, have the potential to eliminate the need for cultural enrich-ment (1). However, a major hurdle to overcome for such technol-ogies is the need to identify surface antigen targets that areconsistently expressed regardless of sample matrix or growthphase.

Our laboratory has produced MAbs which are specific to L.monocytogenes serotype 4b and bind to the cell wall-associatedIspC with high affinity (13 and Ronholm et al., unpublished).However, before these antibodies are used in diagnostics, it is cru-cial to evaluate the influence of stress on expression of the targetprotein and determine how this affects antibody reactions (7). Wehave shown that IspC is expressed in each of our tested conditionsand that a MAb to IspC is able to detect bacterial cells in everytested growth environment, including extreme stress conditionsand enrichment media, supporting the claim that IspC, togetherwith its specific MAbs, has value in antibody-based diagnostics forL. monocytogenes serotype 4b.

ACKNOWLEDGMENTS

J.R. was supported by the Canadian Institutes of Health Research (CIHR)through a doctoral award–Frederick Banting and Charles Best CanadaGraduate Scholarship. This study was also supported in part by a StrategicGrant from the Natural Sciences and Engineering Research Council ofCanada (NSERC) to X.C. and M.L. and in part by the Ontario Centres ofExcellence.

We also acknowledge Brian Brooks and Susan Logan for reviewing themanuscript and providing helpful suggestions.

REFERENCES1. Dwivedi HP, Jaykus LA. 2011. Detection of pathogens in foods: the current

state-of-the-art and future directions. Crit. Rev. Microbiol. 37:40–63.2. Ferreira A, O’Byrne CP, Boor KJ. 2001. Role of �B in heat, ethanol, acid,

and oxidative stress resistance and during carbon starvation in Listeriamonocytogenes. Appl. Environ. Microbiol. 67:4454 – 4457.

3. Geng T, Hahm BK, Bhunia AK. 2006. Selective enrichment media affectthe antibody-based detection of stress-exposed Listeria monocytogenes dueto differential expression of antibody-reactive antigens identified by pro-tein sequencing. J. Food Prot. 69:1879 –1886.

4. Geng T, et al. 2003. Expression of cellular antigens of Listeria monocyto-genes that react with monoclonal antibodies C11E9 and EM-7G1 underacid-, salt- or temperature-induced stress environments. J. Appl. Micro-biol. 95:762–772.

5. George S, Lund B, Brocklehurst T. 1988. The effect of pH and temper-ature on initiation of growth of Listeria monocytogenes. Lett. Appl. Micro-biol. 6:153–156.

6. Gerner-Smidt P, et al. 2005. Invasive listeriosis in Denmark 1994 –2003:a review of 299 cases with special emphasis on risk factors for mortality.Clin. Microbiol. Infect. 11:618 – 624.

7. Hahm BK, Bhunia AK. 2006. Effect of environmental stresses on anti-body-based detection of Escherichia coli O157:H7, Salmonella enterica se-rotype Enteritidis and Listeria monocytogenes. J. Appl. Microbiol. 100:1017–1027.

8. Jacquet C, Gouin E, Jeannel D, Cossart P, Rocourt J. 2002. Expressionof ActA, Ami, InlB, and listeriolysin O in Listeria monocytogenes of humanand food origin. Appl. Environ. Microbiol. 68:616 – 622.

9. Jorgensen F, Stephens PJ, Knochel S. 1995. The effect of osmotic shockand subsequent adaptation on the thermotolerance and cell morphologyof Listeria monocytogenes. J. Appl. Microbiol. 79:274 –281.

10. Kazmierczak MJ, Mithoe SC, Boor KJ, Wiedmann M. 2003. Listeriamonocytogenes sigma B regulates stress response and virulence functions.J. Bacteriol. 185:5722–5734.

11. Lathrop AA, Banada PP, Bhunia AK. 2008. Differential expression ofInlB and ActA in Listeria monocytogenes in selective and nonselective en-richment broths. J. Appl. Microbiol. 104:627– 639.

12. Lenz LL, Mohammadi S, Geissler A, Portnoy DA. 2003. SecA2-dependent secretion of autolytic enzymes promotes Listeria monocyto-genes pathogenesis. Proc. Natl. Acad. Sci. U. S. A. 100:12432–12437.

13. Lin M, et al. 2009. Screening and characterization of monoclonal anti-bodies to the surface antigens of Listeria monocytogenes serotype 4b. J.Appl. Microbiol. 106:1705–1714.

14. Lin M, et al. 2006. Monoclonal antibodies binding to the cell surface ofListeria monocytogenes serotype 4b. J. Med. Microbiol. 55:291–299.

15. Lungu B, Ricke SC, Johnson MG. Growth, survival, proliferation andpathogenesis of Listeria monocytogenes under low oxygen or anaerobicconditions: a review. Anaerobe 15:7–17.

16. McClure PJ, Roberts TA, Otto Oguru P. 1989. Comparison of the effectsof sodium chloride, pH and temperature on the growth of Listeria mono-cytogenes on gradient plates and in liquid medium. Lett. Appl. Microbiol.9:95–99.

17. Milenbachs AA, Brown DP, Moors M, Youngman P. 1997. Carbon-source regulation of virulence gene expression in Listeria monocytogenes.Mol. Microbiol. 23:1075–1085.

18. Miller JH. 1972. Experiments in molecular genetics, p 352–355. ColdSpring Harbor Laboratories, Cold Spring Harbor, NY.

19. Nannapaneni R, Story R, Bhunia AK, Johnson MG. 1998. Unstableexpression and thermal instability of a species-specific cell surface epitopeassociated with a 66-kilodalton antigen recognized by monoclonal anti-body EM-7G1 within serotypes of Listeria monocytogenes grown in non-selective and selective broths. Appl. Environ. Microbiol. 64:3070 –3074.

20. Nannapaneni R, Story R, Bhunia AK, Johnson MG. 1998. Reactivities ofgenus-specific monoclonal antibody EM-6E11 against Listeria species andserotypes of Listeria monocytogenes grown in nonselective and selectiveenrichment broth media. J. Food Prot. 61:1195–1198.

21. Poyart C, Trieu-Cuot P. 1997. A broad-host-range mobilizable shuttlevector for the construction of transcriptional fusions to beta-galactosidasein gram-positive bacteria. FEMS Microbiol. Lett. 156:193–198.

22. Premaratne RJ, Lin WJ, Johnson EA. 1991. Development of an improvedchemically defined minimal medium for Listeria monocytogenes. Appl.Environ. Microbiol. 57:3046 –3048.

23. Ronholm J, et al. 2012. The Listeria monocytogenes serotype 4b autolysinIspC has N-acetylglucosaminidase activity. Glycobiology 22:1311–1320.

24. Swaminathan B, Gerner-Smidt P. 2007. The epidemiology of humanlisteriosis. Microbes Infect. 9:1236 –1243.

25. Vazquez-Boland JA, et al. 2001. Listeria pathogenesis and molecularvirulence determinants. Clin. Microbiol. Rev. 14:584 – 640.

26. Walker SJ, Archer P, Banks JG. 1990. Growth of Listeria monocytogenesat refrigeration temperatures. J. Appl. Bacteriol. 68:157–162.

27. Wang L, Lin M. 2007. Identification of IspC, an 86-kilodalton protein targetof humoral immune response to infection with Listeria monocytogenes sero-type 4b, as a novel surface autolysin. J. Bacteriol. 189:2046–2054.

28. Wang L, Lin M. 2008. A novel cell wall-anchored peptidoglycan hydro-lase (autolysin), IspC, essential for Listeria monocytogenes virulence: ge-netic and proteomic analysis. Microbiology 154:1900 –1913.

29. Yang L, Banada PP, Bhunia AK, Bashir R. 2008. Effects of dielectropho-resis on growth, viability and immuno-reactivity of Listeria monocyto-genes. J. Biol. Eng. 2:6.

30. Yu WL, Dan H, Lin M. 2007. Novel protein targets of the humoralimmune response to Listeria monocytogenes infection in rabbits. J. Med.Microbiol. 56:888 – 895.

Ronholm et al.

7840 aem.asm.org Applied and Environmental Microbiology

on March 27, 2020 by guest

http://aem.asm

.org/D

ownloaded from