Transcriptome analysis of Escherichia coli O157:H7 EDL933 during heat shock

Abstract

The response to elevated temperature [heat shock (HS)] is highly conserved. The transcriptome of Escherichia coli K-12 has been studied under a variety of conditions while such studies involving E. coli O157:H7 are only now being conducted. To better understand the impact of HS on E. coli O157:H7, global transcript levels of strain EDL933 cells shifted from 37 to 50 °C for 15 min were compared with cells held at 37 °C by microarray. Using a mixed model analysis, 193 genes were found to be differentially transcribed at P<0.0042 with a q value <0.1. The 111 downregulated genes include the curli pili-associated genes csgABCDEFG, maltose transport-associated proteins malEFK, and NADH dehydrogenase subunit encoding nuoCEHIJN. The 82 genes upregulated include the HS-induced genes rpoH, dnaK, dnaJ, groEL, groES, and grpE along with two LEE-encoded genes: hypothetical gene Z5121 and sepZ. Twenty-three additional genes located in O-islands were found to be differentially expressed. Quantitative real-time PCR (qRT-PCR) was performed to validate the microarray results. Also, samples subjected to a 30–42 °C shift were examined by qRT-PCR to confirm differential transcription of selected genes. These results indicate that this pathogen may regulate its virulence factors in response to temperature changes.

Escherichia coli O157:H7, heat shock, microarray, transcriptome

Introduction

Since its isolation and identification over two decades ago (Riley et al., 1983), Escherichia coli O157:H7 has remained a significant food-borne pathogen. It causes acute gastrointestinal disease in humans, which manifests from mild diarrhea to hemorrhagic colitis, with the possibility of developing hemolytic uremic syndrome, a potentially fatal sequelae that is the leading cause of acute renal failure in children (Nataro & Kaper, 1998). Upon entering a host, E. coli O157:H7 is subjected to a variety of environments that must be endured to ultimately colonize and/or cause disease.

One common factor between all of these host environments is an elevation of temperature. This may be a shift from environmental ambient temperature to internal host temperature during the initial infection process or an increase in internal host temperature (pyrexia) due to bacteremia and inflammation. This pathogen is also subjected to extreme temperature conditions in the food-processing industry. Heat shock (HS) has the potential to be detrimental due to an increase in unfavorable protein-folding interactions. The HS response in E. coli K-12 is well studied (Gross, 1996), but the E. coli O157:H7 EDL933 genome contains additional genes in O-islands and plasmids (Perna et al., 2001), which may respond to HS in unknown ways.

The HS response is often considered a general stress condition. It has been widely studied in many organisms including E. coli K-12. Most, if not all, organisms elicit a response to HS by a rapid, selective, and transient synthesis of a number of universally conserved proteins known as HS proteins (Bardwell & Craig, 1984, 1987). Effects on cellular responses such as HS by the unique information encoded within O-islands and on the plasmid pO157 are poorly understood.

The focus of this study was to characterize the transcriptional response of E. coli O157:H7 to HS. To accomplish this, the transcriptional profile of pathogenic E. coli O157:H7 EDL933 during HS was compared with cells grown under normal conditions using two-color microarrays and quantitative real-time PCR (qRT-PCR). The results of this experiment demonstrate that while classical HS proteins were upregulated as expected, genes not previously identified as being differentially expressed during HS and genes not found within E. coli K-12 were differentially expressed in E. coli O157:H7 during HS as well.

Materials and methods

Strains and culture conditions

Escherichia coli O157:H7 EDL933 was used in this study. To obtain cultures for study, 100 mL of Luria–Bertani broth in 250-mL Erlenmeyer flasks was inoculated with 1 mL of an overnight culture. After a 3-h incubation at 37 °C and shaking at 200 r.p.m., half of the flasks were shifted from 37 to 50 °C for 15 min. One milliliter of cells were quickly pelleted by centrifugation at 12 000 g for 45 s, 500 μL of the culture supernate was aspirated, 1 mL of RNAprotect Bacteria Reagent (Qiagen, Valencia, CA) was added, pellets were homogenized by vortexing, samples were incubated at room temperature for 10 min, and then RNA was isolated. An additional study was performed with a temperature upshift from 30 to 42 °C, which was analyzed by semi-quantitative RT-PCR for specific genes.

Microarray construction

The E. coli O157:H7 microarray consists of 85.7% of the ORFs encoded within the chromosome of E. coli O157:H7 EDL933 (5453 ORFs) and pO157 (101 ORFs) represented as 100–500 bp PCR products (probes) spotted onto glass substrates. Determination of primer sequences and generation and purification of PCR products were carried out as described previously (Madsen et al., 2006). Primers were purchased from Integrated DNA Technologies (Coralville, IA). A custom printer at the Microarray Facility at Washington University was used to print the microarrays onto Corning UltraGAPS substrates (Corning, Big Flats, NY). The array contains 48 subarrays in a 15-column by 15-row configuration with probes spotted in duplicate in a noncontiguous fashion. Slides were UV cross-linked at 450 mJ in a Stratalinker (Stratagene, La Jolla, CA) to immobilize the DNA. A description of the array can be found at the National Center for Biotechnology Information Gene Expression Omnibus (Edgar et al., 2002) under platform GPL6272.

RNA isolation

RNA was isolated from samples using the RNeasy Mini Kit (Qiagen) following the lysis and digestion kit protocol. Samples were treated using Turbo DNAse (Ambion, Austin, TX) as per the manufacturer's instructions. The lysis and digestion protocol was followed with two 50 μL RNAse-free water elutions. Each sample was treated with 2 μL of DNAse (Ambion) at 37 °C for 30 min. Samples were purified and concentrated using Microcon YM-30 columns (Millipore, Billerica, MA), and the quantity and purity were determined using an ND-1000 spectrophotometer (Nanodrop, Wilmington, DE). Samples were determined to be free of contaminating genomic DNA by the absence of a band on a DNA electrophoresis gel after 39 rounds of PCR as follows: samples were diluted to 100 ng μL⁻¹ and used as a template in PCR with native Taq DNA polymerase (Invitrogen, Carlsbad, CA) as per the manufacturer's instructions. The primers used were forward primer 5′-CAGGGTTCTGAGCAGGTGTG-3′ and reverse primer 5′-GACAACCCGGACTGGGAG-3′. Thermocycler conditions were 94 °C for 5 min, and then 39 cycles of 94 °C for 30 s, 62 °C for 30 s, and 72 °C for 30 s, with a final 72 °C for 5 min. RNA sample integrity was assessed using the 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA) with all samples having an RNA integrity number of 9.0 and higher.

Labeled cDNA generation and hybridization

Targets were prepared as described previously, excluding the specific hexamer primer set (Oneal et al., 2008). To prime the cDNA synthesis reactions, 10 μg of random hexamer (Integrated DNA Technologies) was used. The frequency of dye incorporation and yield was determined using the ND-1000 spectrophotometer. From each paired sample, 1.5 μg of labeled cDNA was combined, dried, resuspended in 60 μL Pronto! cDNA/long oligonucleotide hybridization solution (Corning), incubated at 95 °C for 5 min, and centrifuged at 13 000 g for 2 min at room temperature. This solution was then pipetted on a freshly prehybridized microarray slide, covered with a 22 × 60 mm HybriSlip (Grace Bio Labs, Bend, OR), placed in a Corning hybridization chamber, and incubated in a 42 °C water bath for 12–16 h. Array prehybridization was conducted using sodium borohydride as described previously (Raghavachari et al., 2003). Posthybridization washes were conducted according to Corning's UltraGAPS protocol, dried by centrifugation, and then scanned.

Microarray experimental design

Twelve heat-shocked RNA samples were paired with 12 control RNA samples for hybridization on a two-color microarray slide. Cy3 and Cy5 dye assignments to control and treated samples were reversed for six of the arrays to account for variation in labeling efficiencies (dye bias).

Image acquisition and data analysis

Scanning, image segmentation, and normalization were conducted as described previously (Madsen et al., 2006). The normalized values for duplicate spots were averaged within each array to produce one normalized measure of expression of fluorescence for each probe sequence and each of the 24 experimental units. Mixed model analysis was conducted as described previously (Madsen et al., 2006), excluding slide region effects and slide × region interactions.

To discover potential regulatory motifs in upstream regions of significant differentially transcribed genes, a one-block motif analysis was carried out using bioprospector and meme in conjunction with biooptimizer (Bailey & Elkan, 1994; Liu et al., 2001; Jensen & Liu, 2004). We limited the analysis to genes upregulated that are not found in E. coli K-12 and have >50 bp of upstream intergenic sequence (Z0968, Z2323, Z2400, Z2565, Z3341, hopD, sepZ, Z0402, Z1453, Z1456, Z1648, Z4321, Z4650, Z4883, and Z5121). Motifs were deemed significant if identified by biooptimizer using both meme and bioprospector output as input. bprom (http://www.softberry.com) was used for promoter prediction.

Validation of microarray data

To confirm significant transcriptional differences between genes in both temperature upshift samples, qRT-PCR was performed on nine genes (csgB, dnaK, groEL, malF, rpoH, sepZ, yceP, Z1443, and Z5121) shown to have significant transcriptional differences during HS as determined by microarrays. In addition to these genes, clpB was also analyzed by qRT-PCR with both temperature upshift samples. The Stratagene Brilliant II One Step qRT-PCR kit, along with the Stratagene Mx3005P QPCR System, was used for RT-PCR. The ISU method was used to calculate fold change between treatment and control samples (Gallup & Ackermann, 2006). Table 1 describes the qRT-PCR primers used in this study.

1

RT-PCR primers used to validate the microarray results

Genes	Sequences	Tm (°C)
clpB-F	GGT GCG CGT TCT TAA TCT TTG CGA	60.2
clpB-R	CCT CCA CGC ATT TGT TCA ATC GCT	60.2
csgB-F	AGG TAG TAG CAA CCG GGC AAA GAT	60.9
csgB-R	GCA CCT TGC GAA ATA CTG GCA TCA	60.1
dnaK-F	TGG TGG TCA GAC TCG TAT GCC AAT	60.1
dnaK-R	ATT GCT ACA GCT TCG TCC GGG TTA	60.3
groEL-F	GTG GGT ATC AAA GTT GCA CTG CGT	60.0
groEL-R	TTT GGT TGG GTC CAG GAT ACC CAT	60.2
malF-F	ATT CAA CTG TTA ACC AAC GGC GGC	60.4
malF-R	TGG CTT TCA GGT TCA CTA TCG CCA	60.5
phoP-F	AGC TGG CAG GAC AAA GTC GAA GTA	60.2
phoP-R	CAT TCG CGC CAT CAC CTC TTC AAT	59.9
rpoH-F	TCG TCA AAG TTG CGA CCA CCA AAG	60.0
rpoH-R	ATC GTC GTC GGA AGA CAG GTC AAA	60.0
sepZ-F	TGG CGA CCT CAC TCA GTG GAA ATA	59.9
sepZ-R	CGG CTA TAA CTC TAA CGG TGC GAT	58.5
Z1443-F	AAA GCG CGA GGA AGT AAG CAA G	58.0
Z1443-R	TGT CAT CAG AAG GGC TTA TGA ACT	56.0
Z5121-F	CGT ACG CAG GGA GTG ATT GAA CAT	59.0
Z5121-R	CAT CCT GCG AAC GCG CTC AAT AAT	60.0
yceP-F	AGT CAT TCA GAC TCA TCC GCT CGT	59.8
yceP-R	TGG TAG TGC AAA CGC AAC ATC AGC	60.3

Results

Array development and construction

Of the 5555 ORFs encoded within the E. coli O157:H7 EDL933 genome and pO157, unique PCR primers could be designed and successfully amplify 4756. Agarose gel electrophoresis was conducted postamplification and -purification to determine the specificity of primer pairs and product sizes. Arrays were quality controlled postprinting using spot qc (Integrated DNA Technologies). The hybridized arrays were scanned and spot morphology, spacing, and overall array quality were examined.

HS

To determine the effect of HS on the transcriptome of E. coli O157:H7 EDL933, microarrays were used to compare transcript levels within planktonic cultures subjected to elevated temperature to cultures held at a constant temperature. Cultures were at the same phase of growth. Extraction of total RNA from c. 6.0 × 10⁸ cells yielded a minimum of 25 μg of total RNA post-DNAse treatment. Equal amounts of paired control and heat-shocked total RNA from all samples were subjected to cDNA synthesis and dye coupling to allow for an equal number of dye swaps. Data from each of the twelve replicates were used in the statistical analysis.

Statistical analysis indicated that 193 genes demonstrated transcriptional differences with a P value <0.0042 and an estimated false discovery rate <10%. The significance and differences in transcript levels for all genes are depicted as a volcano plot (Fig. 1). Of the 193 genes differentially expressed, 82 genes were upregulated while 111 genes were downregulated. The 50 most significantly differentially expressed genes, 25 up- and 25 downregulated, are shown (Tables 2 and 3, respectively). The remaining genes are shown in Supporting Information, Table S1. Table 4 lists genes that are differentially expressed that are not present in E. coli K-12. Nine genes, seven upregulated and two downregulated, were chosen for validation using qRT-PCR along with phoP, a control gene that showed no differential expression. We also chose to examine clpB by qRT-PCR since it had previously been shown to be upregulated during HS (Gross, 1996). In all cases the qRT-PCR results confirmed the direction of the transcript difference and indicated that the microarrays show comparatively lower fold change values (Tables 2 and 3). In the case of clpB, the microarray failed to identify any changes, but its upregulation was confirmed by qRT-PCR (Table 2). The dataset from the microarrays can be accessed from the Gene Expression Omnibus (NCBI) using series accession number GSE11463 (http://www.ncbi.nlm.nih.gov/geo/).

1

Volcano plot of transcriptional differences in Escherichia coli O157:H7 EDL933 during HS. Individual differences are plotted as log₂ fold change vs. −log₁₀P value. Points above the line at P<0.0042 indicate differential expression at a false discovery rate of 10%.

2

Top 25 upregulated genes based on P value of microarray data

ID	Locus	Product/function	P value	q value	FC	FC	FC
clpB	Z3886	Protein disaggregation chaperone				5.74	35.4
dinl	Z1698	Putative damage-induced protein I	<0.0004	0.002	1.4
dnaJ	Z0015	Chaperone with DnaK HS protein	<0.0004	0.001	4.7
dnaK	Z0014	Molecular chaperone DnaK	<0.0004	0.001	5.4	20.3	18.3
groEL	Z5748	Chaperonin GroEL	<0.0004	0.001	6.6	12.8	13.7
groES	Z5747	Co-chaperonin GroES	<0.0004	0.000	7.9
grpE	Z3907	HS protein GrpE	<0.0004	0.002	2.7
hflB	Z4540	Integral membrane peptidase	<0.0004	0.022	2.2
htpG	Z0590	HS protein 90	<0.0004	0.005	1.9
htpX	Z2876	HS protein HtpX	<0.0004	0.000	4.9
hydN	Z4021	Electron transport from formate to hydrogen Fe–S centers	<0.0004	0.018	1.2
ibpA	Z5183	HS protein	<0.0004	0.002	5.8
ibpB	Z5182	HS protein	<0.0004	0.006	2.6
prlC	Z4898	Oligopeptidase A	<0.0004	0.025	1.3
rpoH	Z4835	RNA polymerase σ factor	<0.0004	0.027	2.5	5.2	3.1
rrmJ	Z4541	Cell division protein	<0.0004	0.014	2.9
secG	Z4537	Protein-export membrane protein	<0.0004	0.025	2.0
slpA	Z0033	Probable peptidyl-prolyl cis–trans isomerase	<0.0004	0.022	1.2
topA	Z2536	DNA topoisomerase I	<0.0004	0.010	1.4
ybaP	Z0601	Putative ligase	<0.0004	0.016	1.2
yceP	Z1697	Hypothetical protein	<0.0004	0.001	3.5	11.7	4.7
ygiM	Z4408	Hypothetical protein	<0.0004	0.016	1.1
Z1443	Z1443	Unknown protein encoded by prophage BP-933W	<0.0004	0.025	1.4		6.9
Z4321	Z4321	Putative PagC-like membrane protein	<0.0004	0.020	1.2
Z5745	Z5745	Hypothetical protein	<0.0004	0.010	1.6

*

FC, fold change microarray (37–50°C).

†

FC, fold change qRT-PCR (37–50°C).

‡

FC, fold change qRT-PCR (30–42°C).

§

Indicates not found in Escherichia coli K-12.

3

Top 25 downregulated genes based on P value of microarray data

ID	Locus	Product/function	P value	q value	FC	FC	FC
aspA	Z5744	Aspartate ammonia-lyase	<0.0002	0.002	−2.7
crp	Z4718	Cyclic AMP receptor protein	<0.0002	0.008	−1.4
csgA	Z1676	Curlin major subunit	<0.0002	0.006	−2.7
csgB	Z1675	Minor curlin subunit precursor similar to CsgA	<0.0002	0.010	−1.7	−13.4	−4.1
csgC	Z1677	Putative curli production protein	<0.0002	0.010	−1.4
cydB	Z0901	Cytochrome d terminal oxidase polypeptide subunit II	<0.0002	0.005	−1.7
dctA	Z4942	C4-dicarboxylate transport protein	<0.0002	0.016	−1.4
fba	Z4263	Fructose-bisphosphate aldolase	<0.0002	0.004	−2.0
hdeA	Z4922	Hypothetical protein	<0.0002	0.016	−4.7
malF	Z5631	Maltose permease protein	<0.0002	0.009	−10.9	−4.6	−20.3
ompR	Z4760	Osmolarity response regulator	<0.0002	0.008	−1.3
pepD	Z0298	Aminoacyl-histidine dipeptidase	<0.0002	0.012	−1.3
tdcB	Z4469	Threonine dehydratase	<0.0002	0.005	−3.1
tdH	Z5043	l-Threonine 3-dehydrogenase	<0.0002	0.010	−1.8
yciD	Z2034	Putative outer membrane protein	<0.0002	0.009	−1.7
ydaA	Z2435	Hypothetical protein	<0.0002	0.006	−2.3
ydiA	Z2732	Hypothetical protein	<0.0002	0.006	−1.2
yeiA	Z3402	Dihydropyrimidine dehydrogenase	<0.0002	0.010	−1.7
Z0968	Z0968	Unknown protein encoded by prophage CP-933K	<0.0002	0.003	−2.9
Z1479	Z1479	Unknown protein encoded by prophage BP-933W	<0.0002	0.016	−1.1
Z2220	Z2220	Putative sensor kinase	<0.0002	0.010	−1.7
Z2565	Z2565	Putative chaperone protein	<0.0002	0.016	−2.0
Z2869	Z2869	Hypothetical protein	<0.0002	0.003	−1.4
Z3332	Z3332	Unknown protein encoded by prophage CP-933V	<0.0002	0.006	−3.3

*

FC, fold change microarray (37–50°C).

†

FC, fold change qRT-PCR (37–50°C).

‡

FC, fold change qRT-PCR (30–42°C).

§

Indicates not found in Escherichia coli K-12.

4

Differentially transcribed genes not found in Escherichia coli K-12

ID	Locus	Product/function	P value	q value	FC	FC
Downregulated
Z0968	Z0968	Unknown protein encoded by prophage CP-933K	0.000	0.003	−2.9
Z1479	Z1479	Unknown protein encoded by prophage BP-933W	0.000	0.016	−1.1
Z1495	Z1495	Unknown protein encoded by prophage BP-933W	0.001	0.052	−1.1
Z2323	Z2323	Hypothetical protein	0.001	0.057	−1.3
Z2400	Z2400	Partial putative repressor protein encoded by prophage CP-933R	0.003	0.079	−1.3
Z2565	Z2565	Putative chaperone protein	0.000	0.016	−2.0
Z3332	Z3332	Unknown protein encoded by prophage CP-933V	0.000	0.006	−3.3
Z3341	Z3341	Unknown protein encoded by prophage CP-933V	0.001	0.039	−1.6
Z4333	Z4333	Putative cytotoxin	0.002	0.067	−1.1
Upregulated
hopD	Z4693	Putative leader peptidase	0.004	0.091	1.1
sepZ	Z5122	Type three secretion system associated protein	0.004	0.097	1.1	24.0
Z0402	Z0402	Putative beta-barrel outer membrane protein	0.001	0.066	1.2
Z1059	Z1059	Hypothetical protein	0.001	0.053	1.6
Z1338	Z1338	Putative DNA replication factor encoded by cryptic prophage CP-933M	0.004	0.091	1.1
Z1340	Z1340	Unknown protein encoded by cryptic prophage CP-933M	0.003	0.079	1.1
Z1443	Z1443	Unknown protein encoded by prophage BP-933W	0.000	0.025	1.4	6.9
Z1453	Z1453	Unknown protein encoded by prophage BP-933W	0.002	0.068	1.3
Z1456	Z1456	Unknown protein encoded by prophage BP-933W	0.004	0.092	1.4
Z1648	Z1648	Unknown in putative ISEc8	0.001	0.048	1.2
Z2394	Z2394	Unknown protein encoded by prophage CP-933R	0.003	0.079	1.2
Z4321	Z4321	Putative PagC-like membrane protein	0.000	0.020	1.2
Z4650	Z4650	Putative transferase	0.001	0.035	1.2
Z4883	Z4883	Hypothetical protein	0.002	0.067	1.2
Z5121	Z5121	Hypothetical protein	0.002	0.067	1.1	21.3

*

FC, fold change microarray (37–50°C).

†

FC, fold change qRT-PCR (30–42°C).

In a second series of experiments, an HS study using a temperature shift from 30 to 42 °C was performed, and select genes were examined by qRT-PCR. In accordance to the 37 to 50 °C shift, all of the genes examined (csgB, dnaK, groEL, malF, rpoH, sepZ, yceP, Z1443, and Z5121) showed the same transcriptional shift (Tables 2 and 3). Additionally, clpB was not differentially transcribed on our microarrays, but it was by qRT-PCR (Table 2).

No significant one-block motifs were identified in the E. coli O157:H7-specific upregulated genes that have >50 bp of upstream intergenic sequence. All of these genes have the predicted σ⁷⁰ promoter two-block sequences upstream identified using bprom.

Discussion

Overview

The data in this study clearly demonstrate that E. coli O157:H7 exhibits transcriptional differences in response to HS. Not surprisingly, previously identified E. coli K-12 HS genes were found to be upregulated in O157:H7 as in other systems. Interestingly, genes common to E. coli O157:H7 and E. coli K-12 that had not previously been reported as differentially regulated in HS were identified. This may be attributed to different experimental conditions or array platforms, or to a real change in gene expression due to the difference in the genomic composition. Further studies will be needed to differentiate these two possibilities. Some genes found within O-islands were found to be differentially expressed as well.

When comparing our HS results with other microarray studies conducted on similar species, the fold change across all genes was slightly lower. A comparison of fold change across microarray platforms is difficult as there are many factors that can affect these differences. Our qRT-PCR results indicate that microarrays underestimate the fold change of transcripts, which is common (Morey et al., 2006).

Classic HS response

Many previously identified HS response genes were upregulated during HS treatment. These include genes encoding major chaperone subunits groEL, groES, dnaK, dnaJ, and grpE, σ factors rpoH and rpoD, HS proteins htpGX, and ibpAB, and proteases clpX, ftsH, hslV, and hflX, among others. These results indicate that the core HS response in E. coli O157:H7 is similar to that of E. coli K-12. This was expected as HS results in an increase of unfavorable protein interactions such as misfolding and aggregation. This requires a quick and sustained response to avoid detrimental effects by producing proteins involved in aiding protein folding, protein disaggregation, and proteolysis. Several previously identified HS genes were not identified in these studies; however, variation in growth conditions, treatment conditions, and microarray platforms may be the potential reasons for this.

Escherichia coli O157:H7-specific genes

The sequencing of the E. coli K-12 and E. coli O157:H7 genomes revealed a 4.1-Mbp shared backbone with E. coli O157:H7 EDL933 containing an additional 1.34 Mbp in O-islands with E. coli K-12 containing an additional 0.53 Mbp in K-islands (Blattner et al., 1997; Perna et al., 2001). Of the 1379 genes encoded within O-islands 24 genes were differentially expressed, 15 up-, and nine downregulated. Twelve of these genes are encoded on bacteriophage including five genes found within the Stx2 encoding bacteriophage BP-933W and two genes are within the Stx1 encoding bacteriophage CP-933V. No increase in transcripts related to the Shiga-like toxins was observed, however. This indicates that the bacteriophage within E. coli O157:H7 are responding to HS, but this increase in temperature does not result in increased levels of Shiga-toxin transcripts.

Z5121 and sepZ, genes adjacent to one another within the locus of enterocyte effacement, were found to be upregulated during HS. Although it has no assigned function, the sepZ (espZ) gene product has been shown to be an effector protein transported via a type III secretion system into host cells by enteropathogenic E. coli (Kanack et al., 2005). This is preliminary evidence that genes located in an important pathogenesis island are upregulated as a result of an increase in temperature. This could offer an advantage to a pathogen by only producing virulence factors when in the host and limit energy expenditures.

It is interesting to note that all genes assayed for regulatory sequences have predicted σ⁷⁰ consensus sequences. The gene encoding σ⁷⁰, rpoD, was significantly upregulated as a result of HS.

In conclusion, we have demonstrated that E. coli O157:H7 shows a gene expression pattern similar to E. coli K-12 in response to HS. Additionally, E. coli O157:H7 not only upregulates genes found within O-islands but also upregulates genes in the enterocyte effacement locus in response to HS. It is unlikely that the latter genes are involved in the classic view of the HS response (i.e. limiting unfavorable protein interactions), but rather are upregulated during HS for another purpose such as pathogenesis. This will require further study.

Acknowledgements

C.M. was funded for the microarray construction through United States Department of Agriculture Specific Cooperative Agreement 58-3625-2-127. We thank Melissa L. Madsen for microarray construction and advice during this project, and Jack Gallup for assistance with qRT-PCR.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Table S1. Differentially expressed genes of Escherichia coli O157:H7 during HS.

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

Editor: Wolfgang Schumann