Transcriptional and Phenotypic Responses of Listeria monocytogenes to Chlorine Dioxide

Abstract

Significant food-borne disease outbreaks have occurred from consumption of ready-to-eat foods, including produce, contaminated with Listeria monocytogenes. Challenging food matrices (e.g., cantaloupe, sprouts) with limited processing steps postharvest to reduce pathogen loads have underscored a need for new mitigation strategies. Chlorine dioxide (ClO₂) is increasingly being used in produce and other food systems to reduce food-borne pathogen levels. The goal of this study was to characterize the transcriptional response and survival of L. monocytogenes 10403S exposed to ClO₂. The transcriptional profile of log-phase cells exposed to 300 mg/liter ClO₂ for 15 min was defined by whole-genome microarray. A total of 340 genes were significantly differentially expressed. Among the differentially expressed genes, 223 were upregulated (fold change ≥ 1.5; adjusted P value < 0.05) in role categories responsible for protein fate, cellular processes, and energy metabolism. There were 113 and 16 genes differentially expressed belonging to regulatory networks of σ^B and CtsR, respectively. We assessed L. monocytogenes 10403S survival after exposure to 100, 300, and 500 mg/liter aqueous ClO₂ in brain heart infusion (BHI) broth; there was a significant difference between cells exposed to 500 mg/liter ClO₂ and those exposed to all other conditions over time (P value < 0.05). Isogenic ΔsigB and ΔctsR mutants exposed to 300 mg/liter ClO₂ were more sensitive to ClO₂ than the wild type under the same conditions. These results provide an initial insight into the mechanisms that L. monocytogenes employs to survive sublethal ClO₂ and further our understanding of the inactivation mechanisms of this increasingly used sanitizer.

INTRODUCTION

Listeria monocytogenes is a Gram-positive, opportunistic food-borne pathogen which has the capacity to cause severe disease in humans and animals. Listeriosis can manifest as febrile gastroenteritis, sepsis, meningoencephalitis, and spontaneous abortion in pregnant women. There are approximately 1,600 cases of listeriosis in the United States annually, 99% of which are attributed to contaminated food (1). L. monocytogenes has been isolated from diverse environments (as reviewed in reference 2) and has the capacity to persist in agricultural (3), food manufacturing (4), and retail food (5) environments. Due to the high mortality risk of listeriosis (∼16% [1]), there is a definitive need for practical, safe, and effective mitigation strategies to control this organism from farm to fork. It is becoming more evident that fresh produce products transmit pathogens to humans (6). Next-generation sanitizers, such as chlorine dioxide (ClO₂), are increasingly being used in the food industry, notably in produce (7). ClO₂ is a strong, rapid oxidizing agent which is used in both gas and aqueous forms in food systems to reduce microbial loads. It effectively decreases microbial populations on the surfaces of foods (8) and food processing equipment (9). For example, ClO₂ reduced L. monocytogenes levels by up to 4.3 CFU/cm² on cantaloupe (10) and >3 log CFU/cm² on stainless steel coupons (11). Evidence of degraded proteins extracted from Saccharomyces cerevisiae (12) and reduced commercially purified NADH to NAD⁺ in vitro (13) from ClO₂ exposure has been proposed among the mechanisms of microbial inactivation. Subsequent studies in Escherichia coli have reported an observed drop in intracellular pH of surviving bacterial cells exposed to ClO₂ (14) which results in cell death.

L. monocytogenes has many well-characterized stress response regulators that allow rapid, adaptive responses to intrinsic factors (e.g., salt and low pH) and to the intervention methods (e.g., high-pressure processing, low and high temperatures) used to control this pathogen in food. The roles of alternative sigma factors in regulation of genes related to stress response and virulence have been documented across many bacterial species as reviewed in Kazmierczak et al. (15). In L. monocytogenes, alternative sigma factors σ^B, σ^C, σ^H, and σ^L and other transcriptional regulators (e.g., CtsR and HrcA) play significant roles in stress response and resistance (16). Specifically, the general stress response sigma factor σ^B not only regulates genes that promote resistance to osmotic and acid stress but also increases the transcription of major virulence genes in L. monocytogenes (17). Exposure and adaptation to environmental stress result in cross-protection upon exposure to subsequent stresses. Evidence of cross-protection has been observed among stress conditions, including alkali, ethanol, and osmotic (18); heat and acid (19); and osmotic, oxidative, and low temperature (20) stresses. Exposure of L. monocytogenes to stress conditions (e.g., high acid, high salt) prior to ingestion by the host can lead to a more robust and efficient host cell invasion (21).

This study aimed to characterize transcriptional changes in L. monocytogenes exposed to ClO₂ to better understand potential methods of inactivation and to elucidate stress response mechanisms that may facilitate adaptation, survival, and, potentially, tolerance. We hypothesized that L. monocytogenes employs multiple oxidative stress response systems as well as general stress response regulatory networks (e.g., σ^B) to overcome ClO₂ oxidative stress. To test this hypothesis, we (i) identified changes in the global transcriptome after exposure to ClO₂ and (ii) tracked the ability of wild-type L. monocytogenes 10403S and isogenic ΔsigB and ΔctsR mutants to survive ClO₂ exposure over time. Our data suggest that several transcriptional regulatory networks, including σ^B and CtsR, and oxidative stress response systems help L. monocytogenes survive sublethal ClO₂ exposure.

MATERIALS AND METHODS

Bacterial strains and storage.

L. monocytogenes 10403S (22) and its isogenic mutant strains L. monocytogenes 10403S ΔsigB (23) and L. monocytogenes 10403S ΔctsR (24) were used in this study. Stock cultures were stored at −80°C in brain heart infusion (BHI; Difco, Detroit, MI) broth–15% glycerol. Prior to each experiment, stock cultures were streaked onto BHI agar in order to obtain an isolated colony. A single colony was used to inoculate 5 ml of sterile BHI broth into a culture tube (16 by 125 mm). Following overnight incubation (15 to 18 h) at 37°C with 230 rpm shaking, 5 ml of preheated sterile BHI broth was inoculated with 50 μl of overnight culture. Following growth to the early log phase (optical density at 600 nm [OD₆₀₀] = 0.4) at 37°C with 230 rpm shaking, 500 μl of early log phase was used to inoculate 50 ml of preheated sterile BHI broth into a 300-ml Nephelo flask (Bellco, Vineland, NJ). Cultures were grown to the early log phase (OD₆₀₀ = 0.4) at 37°C and 230 rpm; these conditions were used for all phenotypic and transcriptional work.

Chlorine dioxide effect on log-phase cell survival.

Early log-phase cultures of L. monocytogenes 10403S were grown as described above and exposed to 100, 300, or 500 mg/liter commercially available aqueous chlorine dioxide (ClO₂; CDG Environmental, Bethlehem, PA). A stock solution of ClO₂ (3,000 mg/liter) was added directly to the broth culture at specific volumes to attain the target concentration. Prior to cell exposure, the ClO₂ concentration of the stock solution was verified using the N,N-diethyl-p-phenylenediamine (DPD) method and the solution was scanned on a Hach DR/2500 spectrophotometer (Hach Instruments, Loveland, CO) (25). Phosphate-buffered saline (PBS) was added to the control samples at a volume equal to the volume of ClO₂ added with the aim of minimizing the effects of volume and headspace differences on cell activity. Both PBS and ClO₂ were preheated to 37°C prior to addition to the culture in order to limit the effects of temperature shifts on transcription. Cells exposed to 100, 300, or 500 mg/liter ClO₂ were incubated at 37°C and 230 rpm, and a 500-μl aliquot was removed at 2.5, 5, 7.5, 10, 12.5, 15, and 20 min postexposure to enumerate cell populations. L. monocytogenes 10403S ΔsigB and ΔctsR were also exposed to 300 mg/liter ClO₂ to assess survival as previously described. All exposed samples were combined with 500 μl of neutralizing buffer (BD Difco, Franklin Lakes, NJ) (1:1 ratio) to stop ClO₂ activity and were stored on ice until further processing (≤30 min postexposure). All samples were serially diluted in PBS and spread plated onto BHI. Plates were incubated at 37°C; colonies were enumerated after 48 h of incubation. Statistical analyses were performed using SAS v 9.3 (SAS Institute, Inc., Cary, NC). A general linear model (GLM) and multivariate repeated measure analysis of variance (ANOVA) with a Tukey pairwise comparison were used to test the effects of “time” and “time*concentration” or “time*strain” interactions. Ratios (calculated as log CFU/ml treated − log CFU/ml untreated) were calculated to normalize each strain or condition to the initial cell population before ClO₂ exposure. The variable “time” represented a repeated measure of the ratio of log CFU/ml cells treated with ClO₂ to log CFU/ml unexposed cells after 2.5, 5, 7.5, 10, 12.5, 15, and 20 min. The variables “concentration” and “strain” represented different [ClO₂] and isogenic deletion mutants. Significance was defined as P < 0.05. Three independent experiments were performed.

RNA extraction.

L. monocytogenes 10403S was grown to early log phase as previously described. Cells were exposed to 300 mg/liter aqueous ClO₂ at 37°C and 230 rpm for 15 min. An equal volume of PBS was added to the control sample. Following exposure to ClO₂, 50-ml cultures were combined with ice-cold neutralizing buffer (1:1 ratio) and immediately added to an ice-cold stop solution of 10% acid–phenol chloroform (Life Technologies, Grand Island, NY) in ethanol (10:1 ratio). Bacterial cells were collected by centrifugation at 15,300 × g for 20 min at 4°C. Cells were resuspended in 1 ml TRI reagent (Life Technologies, Grand Island, NY). Cell lysis was performed by bead beating (0.1 zircon beads) for 5 cycles of 30 s per cycle at speed setting 4 in a FastPrep FP120 device (BIO101/Savant, Santa Ana, CA). Samples were chilled on ice between cycles. Following centrifugation at 1,700 × g for 10 min at 4°C, 0.1 ml of 1-bromo-3-chloropropane was added to the supernatant and the mixture was incubated at room temperature for 10 min. RNA was precipitated in isopropanol at −20°C overnight. DNase treatment and further RNA purification were performed as outlined previously (26). RNA quantity and quality were assessed by spectroscopy using a NanoDrop 2000c spectrophotometer (Thermo Scientific, Wilmington, DE) and a 2100 Bioanalyzer system (Agilent, Foster City, CA), respectively. RNA integrity number (RIN) scores of greater than 7.0 were considered acceptable for further work.

cDNA synthesis and microarray hybridization.

Synthesis of cDNA from RNA was performed as previously outlined (27) with the exception of using SuperScript III reverse transcriptase (RT; Life Technologies, Grand Island, NY) for reverse transcription to cDNA. Quadruplicate samples were prepared with dye swapping to minimize dye bias.

The Cornell University Food Safety Laboratory (CUFSL) Listeria 6K oligonucleotide array (NCBI GEO GPL5029) was utilized as previously outlined (26, 28). This 70-mer oligonucleotide DNA microarray includes 2,857 open reading frames (ORFs) from L. monocytogenes EGD-e, 5 ORFs from Saccharomyces cerevisiae (negative control), L. monocytogenes chromosomal DNA (positive control), and salmon sperm DNA (printing quality control). It should be noted that the microarray used here was designed using the L. monocytogenes EGD-e genome sequence. The sequences of the targeted ORFs on the microarray were matched with the L. monocytogenes 10403S genome. Cross-hybridization index (CHI) values of 100%, >95%, and >90% were calculated for 2,107, 2,578, and 2,695 of the probes on the microarray, respectively (28). A total of 45 probes had a CHI value < 90, and 117 of the probes designed for the EGD-e genome were missing from the 10403S genome (28) and were omitted during analysis.

Preparation of the microarray slides, hybridization, and posthybridization washing and drying were performed as previously outlined (29). The microarrays were scanned using a GenePix 4000B microarray scanner (Molecular Devices, Sunnyvale, CA) to provide working TIFF files.

Microarray statistical analysis.

Spots were aligned and identified by probe name within the software-guided grid and analyzed using GenePix Pro 6.0 software (Molecular Devices, Sunnyvale, CA). Spots that were empty or saturated in both channels or had poor morphology were flagged and removed from further analysis. Microarray data were analyzed using LIMMA (linear models for microarray data) package version 3.18.2 (30) through the use of BioConductor software supported by R version 2.15.1 (31). Background correction (normexp method), within-array normalization (print-tip loess method), and between-array normalization (scale method) were all used for data preprocessing (32, 33). Differential expression analysis was performed by first fitting a linear model to the normalized data followed by a smoothing correction (eBays function) (30). Differentially expressed genes were defined as having a fold change (FC) ≥ 1.50 and an adjusted P value < 0.05. The activity of defined operons was based on those previously annotated by our group through transcriptome sequencing (RNA-Seq) analyses (34). The continuity of RNA-Seq coverage across neighboring genes and the presence or absence of Rho-independent terminators were used for operon characterization (34). Differentially expressed genes were categorized into the published J. Craig Venter Institute (JCVI) role categories and functional groups (http://cmr.jcvi.org/tigr-scripts/CMR/CmrHomePage.cgi) as utilized previously to investigate regulatory networks (17). An overall χ² test was performed to determine if role categories were over- and underrepresented; odds ratios (OD) were calculated to identify which categories were over- and underrepresented. The Department of Energy Joint Genome Institute Integrated Microbial Genomes (DOE JGI IMG) (http://img.jgi.doe.gov/cgi-bin/w/main.cgi) listings were consulted further for additional functional group annotation.

TaqMan qRT-PCR.

TaqMan primers (IDT DNA, Coralville, IA) and probes (Applied Biosystems, Foster City, CA) were designed using Primer Express 3.0 (Applied Biosystems, Foster City, CA) for L. monocytogenes 10403S genes rpoB, lexA, clpC, dnaK, sigB, lmo0669, and lmo1433 (Table 1). Quantitative reverse transcriptase PCR (qRT-PCR) was performed as outlined (35) with TaqMan One-Step RT-PCR Master Mix reagent and MultiScribe RT for single-tube reverse transcription and amplification using a StepOnePlus real-time PCR system (Applied Biosystems, Foster City, CA). Negative controls to detect genomic DNA contamination were performed by omitting reverse transcriptase for each replicate and primer and probe set. All reactions were performed in duplicate for each of three independent RNA isolations. A standard curve based on genomic DNA copy number was generated for each gene. Log-transformed (log₁₀) values of mRNA expression values were normalized against the expression value of the housekeeping gene rpoB, which is routinely used for normalization of qRT-PCR data in L. monocytogenes (35). Statistical analyses were performed using SAS v 9.3 (SAS Institute, Inc., Cary, NC) with a general linear model (GLM) and Dunnett correction for multiple comparisons. Significantly different transcript levels between the experimental and control samples were defined by a P value < 0.05.

TABLE 1.

TaqMan primers and probes^a

Target	Forward primer	Probeb	Reverse primer
rpoB	CGATCTTGGAGAGCCGAAATA	CGGTAGAAGAATCTAAGAAC	GAGCCGCATAGTTTGCATCAC
sigB	GTAGAGTCCATCGCCCGAAA	TTCCCAAGGTAAATCT	TCCCAACTTGAACTAAATCTTCATGA
lmo0669	TGAAAAGCAACGCCAGGATT	CCTGGTTTGCAAAGTA	TCGCCCGCATCTGGAA
lmo1433	AGCACAGGCAGCAGGGTTAA	AGCAATTGTAGAAGAGCG	GTACACACGTTCCACCCCAAT
clpC	AAAGCTCGGCAGCAAGTTCTA	TTAGGCGGCGGTGAT	GTCTCCCCGCCCCAGTAG
dnaK	CGAACGTAATACAACTATCCCAACA	AAATCACAAACTTTCTCTACAGC	ATCTACGGCTGGTTGGTTATCAG
lexA	ACTGTTCATGGGCATCTTGCT	CTTGAAGGTAAAGGGC	GAGGCTTTGTAGGGTCACGTCTA

^aAll sequences are written 5′ → 3′.

^bAll probes have a 5′ 6-carboxyfluorescein (FAM) reporter dye with a 3′ MGB dark quencher dye.

Microarray data accession number.

The raw and normalized microarray data are available through the NCBI Gene Expression Omnibus (GEO) genomics data repository under accession number GSE48187.

RESULTS

Effect of [ClO₂] on L. monocytogenes 10403S survival.

Early log-phase cultures of L. monocytogenes 10403S were exposed to three concentrations of aqueous ClO₂ (Table 2) to (i) assess survival and (ii) determine exposure concentrations that would elicit a transcriptional effect for subsequent microarray analyses. The variables of “time,” “concentration,” and “time*concentration” had a significant effect on L. monocytogenes survival (P values = 0.0003, 0.0084, and 0.0001, respectively). There was a significant difference in survival between cells exposed to 500 mg/liter ClO₂ and those exposed to 300 and 100 mg/liter after 12.5 min (Table 2). Specifically, fewer cells survived the 500 mg/liter ClO₂ treatment. Exposure of cells to 300 mg/liter ClO₂ for 15 min resulted in a decrease in cell population that was not statistically significant; 300 mg/liter ClO₂ was selected as a suitable condition to investigate transcriptional effects of sublethal ClO₂ concentrations due to the observable subtle effect on cell populations.

TABLE 2.

L. monocytogenes ClO₂ stress survival over time

[ClO2] or strain	Ratio ± SD of log CFU/ml cells treated with ClO2 to log CFU/ml unexposed cells at indicated time (min)b
	2.5	5	7.5	10	12.5	15	20
[ClO2] (mg/liter)a
100	−0.14 ± 0.30A	−0.31 ± 0.19A	−0.34 ± 0.19A	−0.34 ± 0.16A	−0.34 ± 0.13A	−0.25 ± 0.11A	0.20 ± 0.18A
300	0.02 ± 0.26A	−0.13 ± 0.38A	−0.06 ± 0.17A	−0.06 ± 0.10A	−0.17 ± 0.05A	−0.33 ± 0.37A	−0.21 ± 0.15A
500	−0.90 ± 0.62A	−1.45 ± 1.51A	−1.65 ± 1.45A	−1.97 ± 1.42A	−2.59 ± 1.07B	−3.65 ± 0.55B	−3.65 ± 0.55B
Strain
Wild typec	0.02 ± 0.26A	−0.13 ± 0.38A	−0.06 ± 0.17A	−0.06 ± 0.10A	−0.17 ± 0.05A	−0.33 ± 0.37A	−0.21 ± 0.15A
ΔsigBd	−0.59 ± 0.17B	−0.61 ± 0.10A	−0.59 ± 0.18B	−0.59 ± 0.14C	−0.69 ± 0.14A	−0.65 ± 0.21B	−0.67 ± 0.21B
ΔctsRe	−0.37 ± 0.16B	−0.38 ± 0.34A	−0.42 ± 0.26B	−0.42 ± 0.12B	−0.44 ± 0.38A	−0.40 ± 0.32AB	−0.42 ± 0.18A

^aConcentration of ClO₂ added to log-phase wild-type L. monocytogenes 10403S.

^bDifferent superscript letters denote statistically significant ratios among concentrations or strains at that time point.

^cRatio of log CFU/ml wild-type cells treated with 300 mg/liter ClO₂ to log CFU/ml unexposed wild-type cells after 2.5, 5, 7.5, 10, 12.5, 15, and 20 min.

^dRatio of log CFU/ml ΔsigB cells treated with ClO₂ to log CFU/ml unexposed ΔsigB cells after 2.5, 5, 7.5, 10, 12.5, 15, and 20 min.

^eRatio of log CFU/ml ΔctsR cells treated with ClO₂ to log CFU/ml unexposed ΔctsR cells after 2.5, 5, 7.5, 10, 12.5, 15, and 20 min.

Microarray analyses identified 340 differentially expressed genes upon exposure to ClO₂.

Whole-genome microarray analyses identified 340 genes as significantly differentially expressed following exposure of log-phase L. monocytogenes 10403S to ClO₂. These genes constituted 114 putative operons among which at least one gene showed differential expression. Of the 340 genes differentially expressed, 223 genes were upregulated (≥1.5-fold change [FC]; adjusted P value < 0.05) upon exposure to ClO₂ compared to unexposed wild-type cells (Table 3; see also Table S1 in the supplemental material). In contrast, 117 genes were significantly downregulated in cells exposed to ClO₂ compared to unexposed cells (FC ≥ 1.50; adjusted P value < 0.05) (Table 4; see also Table S2).

TABLE 3.

Genes upregulated upon exposure to ClO₂ among over- and underrepresented JCVI role categories^a

Locus tagb	Gene name	Annotationc	Fold change	Adjusted P value
Cell envelope
lmo0584		Conserved hypothetical membrane protein	2.58	0.006
lmo0880		Similar to wall-associated protein precursor (LPXTG motif)	8.68	0.005
Cellular processes
lmo0220	ftsH	Highly similar to cell division protein FtsH	1.55	0.005
lmo0321		Similar to unknown proteins	4.86	0.003
lmo0515		Conserved hypothetical protein	8.05	0.018
lmo0669		Similar to oxidoreductase	5.78	0.049
lmo0906		Similar to glutathione reductase	1.84	0.038
lmo0983		Similar to glutathione peroxidase	2.76	4E-04
lmo1288		Conserved hypothetical protein	1.9	0.01
lmo1433		Similar to glutathione reductase	2.53	0.021
lmo1439	sod	Superoxide dismutase	2.44	2E-04
lmo1577		Similar to unknown proteins	2.07	0.012
lmo1580		Similar to unknown protein	4.76	0.002
lmo1601		Similar to general stress protein	3.37	0.003
lmo1694		Similar to CDP-abequose synthase	2.64	0.014
lmo1708		Similar to aminoglycoside N3-acetyltransferases	1.57	0.03
lmo1879	cspD	Similar to cold shock protein	3.87	0.004
lmo2016	cspB	Similar to major cold shock protein	1.76	0.002
lmo2190	mecA	Competence negative regulator MecA	1.87	0.002
lmo2230		Similar to arsenate reductase	8.07	1E-03
lmo2398	ltrC	Low-temperature-requirement C protein, also similar to B. subtilis YutG protein	2.39	0.005
lmo2426		Conserved hypothetical proteins	1.81	0.003
lmo2673		Conserved hypothetical protein	8.85	0.003
Energy metabolism
lmo0232	clpC	Endopeptidase Clp ATP-binding chain C	4.84	3E-04
lmo0261		Similar to phospho-beta-glucosidase	1.98	0.03
lmo0268		Similar to phosphoglycerate mutase	6.77	1E-04
lmo0342		Similar to transketolase	2.96	0.014
lmo0347		Similar to dihydroxyacetone kinase	4.54	0.024
lmo0521		Similar to 6-phospho-beta-glucosidase	1.82	0.049
lmo0613		Similar to oxidoreductase	5.22	2E-04
lmo0646		Similar to unknown proteins	4.18	2E-04
lmo0722		Similar to pyruvate oxidase	2.96	0.007
lmo0758		Unknown	118	2E-07
lmo0759		Unknown	141	2E-07
lmo0781		Similar to mannose-specific phosphotransferase system (PTS) component IID	6.35	6E-04
lmo0811		Similar to carbonic anhydrase	2.2	0.015
lmo0907		Similar to phosphoglycerate mutase	2.27	0.008
lmo0913		Similar to succinate semialdehyde dehydrogenase	8.15	0.002
lmo0915		Similar to phosphotransferase system enzyme IIC	12.5	0.009
lmo0917			11.5	0.042
lmo0943	fri	Nonheme iron-binding ferritin	3.23	0.007
lmo1054			1.48	0.046
lmo1055	pdhD	Highly similar to dihydrolipoamide dehydrogenase, E3 subunit of pyruvate dehydrogenase complex	1.98	0.006
lmo1233	trxA	Thioredoxin	2.37	0.023
lmo1254		Similar to alpha,alpha-phosphotrehalase	7.47	2E-04
lmo1255		Similar to PTS system trehalose-specific enzyme IIBC	6.61	2E-04
lmo1293	glpD	Similar to glycerol-3-phosphate dehydrogenase	3.11	0.009
lmo1348		Similar to aminomethyltransferase	5.98	0.022
lmo1350		Similar to glycine dehydrogenase (decarboxylating) subunit 2	4.84	0.034
lmo1514		Similar to unknown protein	1.74	0.007
lmo1538		Similar to glycerol kinase	11.9	5E-05
lmo1579		Similar to alanine dehydrogenase	2.34	0.006
lmo1789		Weakly similar to NAD(P)H oxidoreductase chain B	1.62	0.041
lmo1883		Similar to chitinases	3.17	0.031
lmo1995	dra	Similar to deoxyribose-phosphate aldolase	1.91	0.001
lmo2000		Similar to PTS mannose-specific enzyme IID component	2.47	0.039
lmo2057	ctaB	Highly similar to heme A farnesyltransferase	1.85	0.015
lmo2122		Similar to maltodextrose utilization protein MalA	4.29	0.017
lmo2205		Similar to phosphoglyceromutase 1	2.2	0.038
lmo2363		Similar to glutamate decarboxylase	1.78	0.031
lmo2373		Similar to phosphotransferase system (PTS) beta-glucoside-specific enzyme IIB component	1.98	0.036
lmo2425			1.45	0.01
lmo2437		Unknown	1.6	0.027
lmo2494		Similar to negative regulator of phosphate regulon	2.37	0.012
lmo2528	atpC	Highly similar to H+-transporting ATP synthase chain epsilon	1.77	0.006
lmo2573		Similar to zinc-binding dehydrogenase	1.98	0.038
lmo2650		Similar to hypothetical PTS enzyme IIB component	5.69	0.002
lmo2674		Similar to ribose-5-phosphate epimerase	5.26	0.002
lmo2724		Similar to unknown proteins	2.44	0.004
lmo2743		Similar to transaldolase	2.31	0.01
lmo2830		Similar to thioredoxin	1.89	0.012
Hypothetical protein
lmo0170		Unknown	2.72	0.003
lmo0229	ctsR	Highly similar to transcription repressor of class III stress genes (CtsR)	4.82	0.042
lmo0231		Similar to arginine kinase	2.2	0.043
lmo0439		Weakly similar to a module of peptide synthetase	2.44	0.02
lmo0496		Similar to B. subtilis YnzC protein	4.93	7E-04
lmo0596		Similar to unknown proteins	4.09	0.002
lmo0670		Unknown	4.49	1E-03
lmo0720		Unknown	2.3	0.041
lmo0761		Similar to unknown proteins	36.8	4E-05
lmo0796		Conserved hypothetical protein	15.9	0.001
lmo0869		Unknown	2.9	0.035
lmo0911		Unknown	2.58	0.023
lmo0964		Similar to B. subtilis YjbH protein	2.28	0.022
lmo1059		Unknown	1.79	0.003
lmo1069		Similar to B. subtilis YlaI protein	1.68	0.008
lmo1137		Unknown	2.65	2E-04
lmo1332		Similar to conserved hypothetical proteins	1.53	0.038
lmo1501		Similar to unknown proteins	1.55	0.023
lmo1502		Similar to unknown proteins	1.72	0.013
lmo1503		Unknown	1.52	0.019
lmo1526		Similar to unknown proteins	4.24	0.002
lmo1535		Similar to unknown proteins	1.88	0.023
lmo1602		Similar to unknown proteins	3.15	0.007
lmo1608		Similar to unknown proteins	2.54	0.001
lmo1704		Similar to conserved hypothetical proteins	3.67	0.001
lmo1728		Some similarities to cellobiose-phosphorylase	4.33	0.021
lmo1921			1.47	0.027
lmo2054		Similar to unknown proteins	1.55	0.014
lmo2191	spxA	Similar to unknown proteins	2.12	0.006
lmo2213		Similar to unknown protein	23.6	5E-05
lmo2387		Conserved hypothetical protein	2.39	0.025
lmo2391		Conserved hypothetical protein similar to B. subtilis YhfK protein	10.8	2E-04
lmo2392			1.46	0.023
lmo2393			1.47	0.03
lmo2570		Unknown	3.18	0.003
lmo2572		Similar to chain A, dihydrofolate reductase	3.6	5E-04
lmo2585		Similar to B. subtilis YrhD protein	16.1	0.003
lmo2603		Unknown	2.3	0.014
lmo2646		Unknown	6.42	1E-03
lmo2692		Unknown	1.52	0.03
lmo2705		Unknown	1.52	0.031
Protein fate
lmo0764		Similar to lipoate-protein ligase	4.72	0.001
lmo1051		Similar to formylmethionine deformylase and to B. subtilis YkrB protein	1.55	0.024
lmo1268	clpX	ATP-dependent Clp protease ATP-binding subunit ClpX	1.79	0.018
lmo1354		Similar to aminopeptidase P	1.68	0.026
lmo1473	dnaK	class I heat shock protein (molecular chaperone) DnaK	2.02	0.024
lmo1474	grpE	Heat shock protein GrpE	2.22	0.008
lmo1578		Similar to X-Pro dipeptidase	2.44	0.004
lmo1611		Similar to aminopeptidase	1.97	0.004
lmo1620			1.5	0.031
lmo2068	groEL	Class I heat shock protein (chaperonin) GroEL	2.22	7E-04
lmo2069	groES	Class I heat shock protein (chaperonin) GroES	2.51	0.002
lmo2199	ohrA	Similar to unknown protein	9.67	3E-06
lmo2206	clpB	Similar to endopeptidase Clp ATP-binding chain B (ClpB)	3.07	0.012
lmo2398	ltrC	Low-temperature-requirement C protein, also similar to B. subtilis YutG protein	2.39	0.005
lmo2468	clpP	ATP-dependent Clp protease proteolytic subunit	2.54	0.009
lmo2510			1.5	0.031
Protein synthesis
lmo0211	ctc	Similar to B. subtilis general stress protein	2.58	0.02
lmo0240		Highly similar to B. subtilis YazC protein	2.03	0.004
lmo1218		Similar to rRNA methylase	1.68	0.013
lmo1703		Similar to RNA methyltransferases	1.52	0.039
lmo2511		Similar to conserved hypothetical proteins such as B. subtilis YvyD protein	6.26	0.003

^aAll genes that showed a higher transcript level in the ClO₂-exposed samples over the control sample without treatment are listed with an adjusted P value < 0.05 and fold change of ≥1.50. Only genes categorized into JCVI role categories with over- and underrepresentation on the basis of the odds ratio are listed.

^bLocus tag identification is on the basis of the lmo number from the L. monocytogenes EGD-e strain.

^cAnnotation was defined on the basis of the JCVI role category listings.

TABLE 4.

Genes downregulated upon exposure to ClO₂ among over- and underrepresented JCVI role categories^a

Locus tagb	Gene name	Annotationc	Fold change	Adjusted P value
Cell envelope
lmo0198	gcaD	Highly similar to UDP-N-acetylglucosamine pyrophosphorylase	1.604	0
lmo0695		Unknown	1.88	0.01
lmo0971	dltD	DltD protein for d-alanine esterification of lipoteichoic acid and wall teichoic acid	1.959	0.05
lmo0972	dltC	d-Alanyl carrier protein	2.081	0
lmo0973	dltB	DltB protein for d-alanine esterification of lipoteichoic acid and wall teichoic acid	2.265	0
lmo1082		Similar to dTDP-sugar epimerase	1.749	0.01
lmo1548	mreB	Similar to cell-shape determining protein MreB	1.977	0
lmo2038	murE	Similar to UDP-N-acetylmuramoylalanyl-d-glutamate-2,6-diaminopimelate ligase	1.793	0.04
lmo2549	gtcA	Wall teichoic acid glycosylation protein GtcA	1.774	0.02
Cellular processes
lmo0217		Similar to B. subtilis DivIC protein	1.955	0.01
lmo0677		Similar to flagellar biosynthesis protein FliQ	4.934	0
lmo0679		Similar to flagellar biosynthetic protein FlhB	1.934	0.02
lmo0693		Similar to flagellar motor switch protein FliY C-terminal part	1.506	0.04
lmo0696		Similar to flagellar hook assembly protein	1.731	0.02
lmo1071		Similar to cell division protein RodA and FtsW	1.885	0.01
lmo1297		Similar to aluminum resistance protein and to B. subtilis YnbB protein (hypothetical)	1.604	0.05
lmo1364	cspL	Similar to cold shock protein	1.808	0
lmo1544	minD	Highly similar to cell division inhibitor (septum placement) protein MinD	1.594	0.02
lmo1699		Some similarities to methyl-accepting chemotaxis proteins	2.075	0
lmo1700		Unknown	1.782	0.03
lmo2040	ftsL	Similar to cell division protein FtsL	1.738	0.02
lmo2427		Similar to cell division proteins RodA, FtsW	1.73	0.05
lmo2428		Similar to cell division proteins RodA, FtsW	1.996	0.01
lmo2506	ftsX	Highly similar to cell division protein FtsX	1.81	0.03
lmo2507	ftsE	Highly similar to the cell division ATP-binding protein FtsE	1.828	0.01
lmo2569		Similar to dipeptide ABC transporter (dipeptide-binding protein)	1.635	0.04
Energy metabolism
lmo1072	pycA	Highly similar to pyruvate carboxylase	1.521	0.04
lmo2367	pgi	Glucose-6-phosphate isomerase	1.597	0.01
Fatty acid and phospholipid metabolism
lmo0970		Similar to enoyl-acyl-carrier protein reductase	1.634	0.02
lmo1806	acpA	Highly similar to acyl carrier proteins	2.068	0.01
lmo1807	fabG	Similar to 3-ketoacyl-acyl-carrier protein reductase	1.828	0.01
lmo2202		Similar to 3-oxoacyl-acyl-carrier protein synthase	2.412	0
lmo2450		Similar to carboxylesterase	1.803	0.02
Protein synthesis
lmo0177	metS	Methionyl-tRNA synthetase	1.811	0
lmo0244		Similar to ribosomal protein L33 type II	1.944	0
lmo0695		Unknown	1.88	0.01
lmo1294	miaA	Similar to tRNA isopentenylpyrophosphate transferase	2.196	0.01
lmo1658	rpsB	30S ribosomal protein S2	1.55	0.03
lmo1755	gatA	Glutamyl-tRNA (Gln) amidotransferase (subunit A)	1.506	0.02
lmo2047	rpmF	Ribosomal protein L32	2.195	0
lmo2597	rplM	Ribosomal protein L13	2.077	0
lmo2633	rpsJ	Ribosomal protein S10	1.523	0.02
lmo2811		Similar to GTPase	1.727	0.02
lmo2856	rpmH	Ribosomal protein L34	1.979	0.02
Purines, pyrimidines, nucleosides, and nucleotides
lmo1827		Similar to guanylate kinases	1.771	0
lmo1929	ndk	Similar to nucleoside diphosphate kinase	2.104	0
lmo0509	prs	Similar to phosphoribosyl pyrophosphate synthetase	1.768	0.01
lmo1096	guaA	Highly similar to GMP synthetase	2.904	0
lmo1929	ndk	Similar to nucleoside diphosphate kinase	2.104	0
lmo1463		Similar to cytidine deaminase	1.75	0.01
lmo1885		Similar to xanthine phosphoribosyltransferase	2.201	0.01
Unknown function
lmo1240		Conserved hypothetical protein, similar to B. subtilis YsnB protein	1.763	0
lmo0998		Similar to hypothetical protein	1.914	0.02
lmo1067		Similar to GTP-binding elongation factor	2.936	0
lmo1479	lepA	Highly similar to GTP-binding protein LepA	1.544	0.01
lmo1500		Similar to unknown proteins	2.351	0.02
lmo2779		Similar to probable GTP-binding protein	2.916	0
lmo2802	gidB	GidB protein	1.511	0.03

^aAll genes that showed a lower transcript level in the ClO₂-exposed samples over the control sample without treatment are listed with an adjusted P value < 0.05 and a fold-change of ≥1.50. Only genes categorized into JCVI role categories with over- and underrepresentation on the basis of the odds ratio are listed.

^bLocus tag identification is on the basis of the lmo number from the L. monocytogenes EGD-e strain.

^cAnnotation was defined on the basis of the JCVI role category listings.

Six JCVI role categories were either over- or underrepresented among significantly upregulated genes.

Genes that were upregulated in cells exposed to ClO₂ were classified by their JCVI role categories (Fig. 1) to identify categories that were over- and underrepresented after stress exposure (Table 5). The energy metabolism role category was overrepresented (OR = 2.1) and included six genes encoding components of phosphotransferase systems (PTS), lmo0613 (oxidoreductase), lmo2830 (thioredoxin), lmo1789 [NAD(P)H oxidoreductase chain B], and lmo2528 (proton-transporting ATP synthase chain). The cellular processes role category group was overrepresented (OR = 1.7) and included genes lmo0669 (oxidoreductase), lmo0906 and lmo1433 (glutathione reductases), and lmo1439 (superoxide dismutase). The activity of genes encoding enzymes such as oxidoreductase or glutathione reductase would allow increased resistance or adaptability to oxidative stress. The protein fate role category was overrepresented as well (OR = 2.2); seven of the genes in that category encoded heat shock proteins. A large representation of transcriptional activity of genes encoding chaperone proteins suggests a demand for protein folding and degradation. As mentioned, ctsR, the regulator for class III heat shock genes, was upregulated. In addition, hrcA, the regulator for class I heat shock genes, was upregulated but not found to be statistically significantly differentially expressed (FC = 1.7; adjusted P value = 0.05). The protein synthesis (OR = 0.4) and cell envelope (OR = 0.2) role categories were underrepresented.

FIG 1.

Differentially expressed gene distribution among JCVI role categories. The JCVI role categories are listed on the y axis. The total number of genes in each role category is in parentheses. Black bars represent the numbers of genes upregulated following ClO₂ exposure; hashed bars represent the numbers of genes downregulated following exposure to ClO₂.

TABLE 5.

JCVI role categories over- or underrepresented for genes upregulated upon exposure of strains to ClO₂

JCVI role categorya	No. of annotated genes	No. of positively expressed genes in ClO2-exposed strainsb	Odds ratioc	Chi-square test P valued
Cellular processes	165	21	1.8	0.018
Energy metabolism	380	48	1.8	0.003
Protein fate	100	16	2.3	0.002
Protein synthesis	142	5	0.4	0.048
Cell envelope	119	2	0.2	0.010
Hypothetical	705	41	0.7	0.021

^aJCVI role categories for L. monocytogenes EDG-e.

^bNumber of genes upregulated upon ClO₂ exposure within the selected role category.

^cOdd ratios were employed to characterize the representation within each category.

^dThe chi-square test was used to identify the significant association between the number of genes overly active and the number of total genes within the selected role category. A P value of <0.05 was used to determine significance.

Significantly downregulated genes were either over- or underrepresented in 7 JCVI role categories.

Genes that were downregulated in cells exposed to ClO₂ were categorized by their JCVI role categories (Fig. 1) to identify categories that were over- or underrepresented after stress exposure (Table 6). The cell envelope role category was overrepresented (OR = 2.3); this group includes genes responsible for cell wall manufacturing, specifically, teichoic acid formation and accumulation (gtcA, dltD, and dltC) and cell shape determination (mreB). Similarly to the upregulated genes, the cell processes category was also overrepresented among downregulated genes (OR = 3.5). The genes downregulated in this role category include lmo0217 (similar to divIC septum formation), minD (septum placement), lmo1071, lmo2427, and lmo2428 (similar to genes encoding cell division proteins RodA and FtsW), and ftsL, ftsX, and ftsE (cell-division proteins). It should be noted that there was increased expression of minC (FC = 1.47; adjusted P value = 0.028), but constraints were not met for minC to be considered significantly differentially expressed. MinC is responsible for septum site determination. The fatty acid and phospholipid metabolism category was overrepresented (OR = 2.6) and included genes encoding acyl-carrier proteins (lmo0970, lmo1806, lmo1807, lmo2202, and lmo2450) responsible for membrane lipid production (36), which suggests reduced availability of membrane structural components. This correlates well with decreased transcription of genes encoding teichoic acid production. The purines, pyrimidines, nucleosides, and nucleotides role category, which includes ndk, prs, guaA, and lmo1827, was also overrepresented (OR = 3.1). Although not included in this role category, lepA, encoding a GTP-binding protein that inhibits tRNA through ribosomal back-translocase activity, was also downregulated. The energy metabolism role category was underrepresented among downregulated genes, but it was overrepresented by upregulated genes.

TABLE 6.

JCVI role categories over- or underrepresented for genes downregulated upon exposure of strains to ClO₂

JCVI role categorya	No. of annotated genes	No. of positively expressed genes in ClO2-exposed strainsb	Odds ratioc	Chi-square test P valued
Cell envelope	119	9	2.3	0.015
Cellular processes	165	17	3.5	<0.001
Fatty acid and phospholipid metabolism	60	5	2.6	0.042
Protein synthesis	142	11	2.4	0.005
Purines, pyrimidines, nucleosides, nucleotides	72	7	3.1	0.004
Unknown	109	7	2.2	0.030
Energy metabolism	380	2	0.13	0.001

^aJCVI role categories for L. monocytogenes EDG-e.

^bNumber of genes downregulated upon ClO₂ exposure within the selected role category.

^cOdd ratios were employed to characterize the representation within each category.

^dThe chi-square test was used to identify the significant association between the number of genes overly inactive and the total number of genes within the selected role category. A P value of <0.05 was used to determine significance.

Genes in two major regulatory networks are differentially expressed under conditions of ClO₂ stress.

A significant number of genes under the regulation of the alternative sigma factor σ^B were differentially expressed. Specifically, a total of 113 genes previously identified as σ^B dependent (26, 34, 37) were upregulated in cells exposed to ClO₂. In total, 51 of the 113 upregulated genes (i) were directly preceded by a σ^B-dependent promoter or (ii) were in one of 43 operons preceded by a σ^B-dependent promoter identified as indicating direct σ^B regulation by Oliver et al. (34). lmo0895, which encodes σ^B, was upregulated in cells exposed to ClO₂ (FC = 1.62; adjusted P value < 0.05). There was an increase in expression of σ^B regulators, including rsbV (anti-anti-sigma factor encoded by lmo0893) and rsbW (anti-sigma factor encoded by lmo0894); however, they were not significantly differentially expressed as defined by our inclusion criteria. There was an increase in activity for genes regulated by the negative regulator CtsR. The entire putative operon for CtsR (ctsR, mcsA, mcsB, and clpC) was significantly upregulated (adjusted P value < 0.05). A total of 10 genes (mcsA, mcsB, clpC, ynzC, lmo0811, lmo0822, lmo2054, clpB, lmo2205, and clpP) of 42 previously reported as being negatively controlled by CtsR were upregulated (24). Upregulation of 6 of 22 genes identified as indirectly and positively regulated by CtsR (24) was also observed.

TaqMan qRT-PCR confirms selected differentially expressed genes upon exposure of strains to ClO₂.

TaqMan qRT-PCR was used to confirm differences in transcript levels of select differentially expressed genes identified by microarray analysis. TaqMan qRT-PCR confirmed statistically significant increases (P value < 0.05) in transcript levels in sigB, lmo0669, lmo1433, clpC, and dnaK in cells exposed to ClO₂ on the basis of log copy number (Fig. 2). The increased transcript levels of sigB and clpC support microarray results in that oxidative stress by ClO₂ leads to the increased transcriptional activity of both sigB and ctsR and subsequently the σ^B and CtsR regulons. Transcript levels of lexA were not significantly increased in cells exposed to ClO₂ on the basis of qRT-PCR analyses, although a small increase in log copy number was observed (Fig. 2). Limited increases in the lexA log copy number are consistent with fold change values identified by microarray analysis of lexA (FC = 1.61; adjusted P value = 0.044) (see Table S1 in the supplemental material).

FIG 2.

Confirmation of differentially expressed genes by TaqMan qRT-PCR. Transcript levels in lexA, clpC, lmo0669, sigB, dnaK, and lmo1433 were quantified in wild-type strain 10403S by TaqMan qRT-PCR. Transcript levels are expressed as log copy number of cDNA normalized to rpoB. Control samples were not exposed to ClO₂; L. monocytogenes was exposed to 300 mg/liter ClO₂ for 15 min. Samples were collected and analyzed in triplicate, with standard deviations represented by error bars. Significant differences between control and ClO₂-exposed samples were determined by PROC GLM with Dunnett correction; significantly different transcript levels are denoted by an asterisk (P value < 0.05).

σ^B and CtsR contribute to L. monocytogenes ClO₂ survival.

We exposed L. monocytogenes 10403S ΔsigB and ΔctsR strains to 300 mg/ml ClO₂ to determine if these regulators affect the survival of L. monocytogenes. The mutant strains were compared to wild-type control cells exposed to 300 mg/liter ClO₂ over time (Table 2). There was a significant effect of “strain” (P < 0.001), but there was a not a significant effect of “time” (P = 0.393) or “time*strain” (P = 0.917). At 2.5, 7.5, 10, 15, and 20 min postexposure to 300 mg/liter ClO₂, there were significant differences in survival among L. monocytogenes 10403S, ΔsigB, and ΔctsR. Specifically, the ΔsigB and ΔctsR strains were more sensitive to ClO₂ than the wild-type cells under the same conditions. This suggests that these regulators play a direct role in ClO₂ survival as shown by a significant transcriptional change in the regulons and differences in survival.

DISCUSSION

In this study, we used transcriptomic and phenotypic approaches to investigate the effects of ClO₂ on L. monocytogenes. Our data indicate that (i) ClO₂ exposure at sublethal levels resulted in differential gene expression of 340 genes, including genes involved in heat shock response, redox reactions, cell replication, and universal stress response, and (ii) stress response regulon networks of σ^B and CtsR contribute to ClO₂ exposure survival. Consistent with other studies investigating L. monocytogenes transcriptional responses to stress (e.g., references 20, 28, and 38), a significant percentage (>11%) of currently identified genes in strain 10403S were differentially expressed. In Bacillus subtilis, oxidative stress induced similar degrees of differential gene expression, with >200 genes upregulated upon exposure to either hydrogen peroxide or paraquat (39). Among the significantly differentially expressed genes, 223 were upregulated and 117 were downregulated. Our data support the hypothesis that L. monocytogenes employs a wide array of stress response systems and that stress response regulators σ^B and CtsR play direct roles in responding to ClO₂ oxidative stress.

Chaperone proteins play a role in oxidative stress management.

Oxidizing agents commonly damage the stability and functionality of proteins. We found 16 genes involved with protein fate activity to be upregulated upon ClO₂ exposure. It is likely that increased transcription of peptidase- and protease-based chaperone protein genes (e.g., clpC, clpB, and clpP) is needed for protein degradation and protein recycling and otherwise for repair of proteins damaged by oxidation. Increased activity of genes encoding chaperone proteins (e.g., dnaK, groEL, and groES) indicated a need for maintenance or repair of protein structure damaged by oxidation. These genes are involved in the heat shock response of L. monocytogenes (38), indicating that they have cross-functional stress response utility. Similar activity of heat shock proteins has been observed in E. coli to manage oxidative stress (40). In E. coli, the roles of chaperone repair proteins DnaK, Hsp33, GroEL, and GroES are cross-functional for oxidative and heat shock response activities (41, 42). Increased transcript levels of dnaK, groES, groEL, and clpP have been observed following exposure to H₂O₂ and chlorine (40). lmo0222, an orthologue to hsp33, was not differentially expressed; however, this specific chaperone has been observed to be controlled posttranslationally (43). A partnership between Hsp33 and DnaK occurs where the oxidized dimers of Hsp33 bind to damaged proteins and, once a redox reaction occurs, transfer the protected substrate protein to DnaK for refolding (42).

Altered activity of cell division genes, a universal stress response, and an SOS response are evident following ClO₂ exposure.

An abundance of genes related to cell division and cell macromolecule synthesis were downregulated upon exposure to ClO₂ in this study. Downregulation of genes that encode cell division proteins (rodA, ftsW, ftsL, ftsX, and ftsE), septum formation (divIC), and movement (minD) suggests that there is altered activity of the cell division machinery. Genes encoding cell wall structural component assembly, including those encoding teichoic acid, acyl-carrier proteins (lmo0970 to lmo0974), and components for DNA synthesis (genes encoding steps in nucleotide production), also showed decreased transcription. It is tempting to speculate that the absence of significantly differentially expressed genes responsible for cell division indicates stalled cell replication. The decrease in expression levels for genes encoding cell wall manufacturing components bolsters the claim for arrested cell division. While genes involved in cell division and replication were downregulated, universal stress protein (Usp) genes, including lmo0515, lmo1580, and lmo2673, were upregulated. It has been suggested that these three Usp genes are involved with DNA damage repair and the arrest of cell division (37). Their transcription is σ^B dependent (37), and they aid in acid and oxidative stress survival (44). Usp has been shown to play a key role in survival of E. coli with respect to oxidative species (45). Increased activity of several key SOS response-related genes, including lexA, uvrA, recN, recO, lmo1881, and ynxZ, was also observed here. The role of the SOS regulon and protection of the cell from the reactive species generated through oxidative stress were also observed in B. subtilis (39).

Genes involved with redox reactions help manage ClO₂ oxidative stress.

We found increased transcription of thioredoxin, oxidoreductase, superoxide dismutase, catalase, and glutathione reductase genes in cells exposed to ClO₂, which outlines the approach of L. monocytogenes to reduce or manage oxidative species. Multiple redox-sensitive transcriptional regulators, including spxA (lmo2191), an ortholog of oxidative stress regulator Spx, were upregulated in this study. In B. subtilis, Spx regulates the thioredoxin-related genes trxA and trxB; thioredoxin aids in oxidative stress management in B. subtilis, working as a substrate for preferential reduction at the expense of NADPH (46). Similarly, we observed increased transcription of thioredoxin genes trxA and lmo2830 as well as lmo1789, which encodes an NADPH oxidoreductase. Further, clpX and clpP, which encode the protease complex ClpXP, which regulates Spx levels (47), were upregulated here, supporting the hypothesis that NADPH-dependent oxidoreductases play roles in managing chlorine dioxide stress. lmo2199 and lmo2200, which have similar levels of homology to the ohrAR operon in B. subtilis, were upregulated as well. OhrA, encoded by ohrA, is involved in organic hydroperoxide resistance in B. subtilis (48). Although OhrR has been identified as a repressor of ohrA, it was observed here that both were upregulated, suggesting that the cell may be regulating an overabundance of OhrA. This same phenomenon was observed in L. monocytogenes under conditions of oxidative stress in the cytosol during intracellular host adaptation (49). Transcription of lmo0613 was significantly increased upon ClO₂ exposure. Its orthologue, marA (an oxidoreductase), and its entire operon have been shown to increase resistance to oxidative and antibiotic stress in E. coli (50). In the case of lmo0669, this oxidoreductase has been identified in previous studies as a general stress response protein (26, 35, 37). Fri, a nonheme iron-binding ferritin, encoded by lmo0943, and kat, which encodes catalase, were upregulated as well. Fri has been associated with resistance to hydrogen peroxide, and kat, a catalase gene with a heme group cofactor, has been shown to reduce reactive oxygen species (51, 52). Glutathione has been shown to protect Lactococcus lactis against oxidative stress, specifically, that caused by H₂O₂ (53). Glutathione reductase reduces glutathione disulfide to glutathione. Increased transcription of an abundance of oxidoreductase genes suggests that the cell may be directing the reactive species toward oxidizing renewable energy targets [e.g., glutathione, NAD(P)H, ATP, etc.] as these are essential for proliferation (53).

σ^B and CtsR contribute to L. monocytogenes ClO₂ survival.

We observed significant transcriptional changes among genes in the σ^B and CtsR regulatory networks and, to some extent, the HrcA regulatory network upon exposure to ClO₂. A total of 113, 16, and 9 genes previously identified as regulated by σ^B (54), CtsR (55), and HrcA (56), respectively, were upregulated in this study. These three regulons constitute an abundance of stress response genes with overlapping and, in some cases, competing responsibilities for transcriptional control (16). We found that the ΔsigB and ΔctsR mutants had a significantly larger cell population decrease than the wild-type strain. Similar rapid responses were observed in B. subtilis within 10 min of exposure to superoxide and peroxide stressors inducing a notable transcriptional response (39). Stress response genes under direct and indirect control of σ^B have been found to be differentially expressed during osmotic (35, 57), acid (23, 26, 35), cold (28), and ethanol (26) stress. The CtsR (class three stress gene repressor) regulon negatively regulates heat shock-type proteins; these proteins are important for resistance to high pressure, high temperature, and osmotic stress (58). The HrcA (heat regulation at CIRCE) regulon negatively regulates a separate class of heat shock-type proteins (class I stress response genes). This group includes the genes constituting the dnaK and groESL operons (56), classic, well-characterized chaperone protein systems. A discrepancy occurs between the upregulation of genes directly negatively regulated by CtsR and increased expression of ctsR itself. It is possible that CtsR was being degraded or was otherwise inactive to enlist the stress response genes controlled by this repressor. As inactivation by ClO₂ is rapid, as shown by the phenotypic assays described above, and as it is likely that the long-term reactivity of ClO₂ is limited by the organic load in the growth medium, the cells may already be trending toward basal expression levels.

Previous studies have demonstrated the ability of ClO₂ to inactivate food-borne pathogens on food and food contact surfaces (e.g., reference 8), but this is the first study, to our knowledge, to investigate the effects of ClO₂ at the transcriptional level. In this study, we found a substantial number of genes differentially expressed upon sublethal exposure to ClO₂, including genes involved in heat shock response and redox reactions. We also found that stress response regulators σ^B and CtsR directly contribute to ClO₂ exposure survival. We observed a clear concentration-dependent effect of ClO₂ on survival of stationary-phase L. monocytogenes cells. While oxidation by ClO₂ is an immediate reaction, we observed a secondary population decrease, which may result from secondary derivative reactive oxygen species stemming from ClO₂ (e.g., ClO₂⁻, Cl⁻, and ClO₃⁻). We recognize that this study was limited to the transcriptional response of L. monocytogenes to ClO₂ in a BHI broth system (as opposed to on produce or food contact surfaces). The volatility of ClO₂ results in experimental design challenges. Bacterial cultures were exposed to ClO₂ in a BHI broth system, as centrifugation and resuspension of cells in buffer have been observed to alter transcriptional profiles (28). Additionally, increased volatility at higher temperatures was a concern at 37°C; however, lowering the temperature of exposure would have led to differential gene expression from temperature shifts alone. Further analysis should be considered to assess the effects of ClO₂ on L. monocytogenes to include (i) multiple times, as only 15 min of exposure to 300 mg/liter was assessed, and (ii) multiple strains from various serotypes and lineages, as only strain 10403S, a 1/2a serotype, was chosen. This initial report provides significant insight into the mechanisms used by L. monocytogenes to respond to and survive oxidative stress from a sanitizer that is increasingly being used in the agricultural commodities and food industries. A thorough understanding of bactericidal mechanisms is pivotal in understanding the utility of sanitizers. This understanding will allow the synergistic application of multiple sanitizer agents with different modes of actions to boost the efficacy and optimization of future sanitizers, both of which will lead to more effective approaches to inactivation of L. monocytogenes in food processing environments.