Ethanol Adaptation Strategies in Salmonella enterica Serovar Enteritidis Revealed by Global Proteomic and Mutagenic Analyses

Stress adaptation in foodborne pathogens has been recognized as a food safety concern since it may compromise currently employed microbial intervention strategies. While adaptation to sublethal levels of ethanol is able to induce ethanol tolerance in foodborne pathogens, the molecular mechanism underlying this phenomenon is poorly characterized. Hence, global proteomic analysis and mutagenic analysis were conducted in the current work to understand the strategies employed by Salmonella enterica subsp. enterica serovar Enteritidis to respond to ethanol adaptation. It was revealed that coordinated regulation of multiple pathways involving metabolism, ABC transporters, regulators, enterobactin biosynthesis and uptake, the ribosome, flagellar assembly, and virulence was responsible for the development of ethanol adaptation response in this pathogen. Such knowledge will undoubtedly contribute to the development and implementation of more-effective food safety interventions.

ABSTRACT

Salmonella enterica subsp. enterica serovar Enteritidis is able to adapt to sublethal concentrations of ethanol, which subsequently induce tolerance of this pathogen to normally lethal ethanol challenges. This work aims to elucidate the underlying ethanol adaptation mechanisms of S. Enteritidis by proteomic and mutagenic analyses. The global proteomic response of S. Enteritidis to ethanol adaptation (5% ethanol for 1 h) was determined by isobaric tags for relative and absolute quantification (iTRAQ), and it was found that a total of 138 proteins were differentially expressed in ethanol-adapted cells compared to nonadapted cells. A total of 56 upregulated proteins were principally associated with purine metabolism and as transporters for glycine betaine, phosphate, d-alanine, thiamine, and heme, whereas 82 downregulated proteins were mainly involved in enterobactin biosynthesis and uptake, the ribosome, flagellar assembly, and virulence. Moreover, mutagenic analysis further revealed the functions of two highly upregulated proteins belonging to purine metabolism (HiuH, 5-hydroxyisourate hydrolase) and glycine betaine transport (ProX, glycine betaine-binding periplasmic protein) pathways. Deletion of either hiuH or proX resulted in the development of a stronger ethanol tolerance response, suggesting negative regulatory roles in ethanol adaptation. Collectively, this work suggests that S. Enteritidis employs multiple strategies to coordinate ethanol adaptation.

IMPORTANCE Stress adaptation in foodborne pathogens has been recognized as a food safety concern since it may compromise currently employed microbial intervention strategies. While adaptation to sublethal levels of ethanol is able to induce ethanol tolerance in foodborne pathogens, the molecular mechanism underlying this phenomenon is poorly characterized. Hence, global proteomic analysis and mutagenic analysis were conducted in the current work to understand the strategies employed by Salmonella enterica subsp. enterica serovar Enteritidis to respond to ethanol adaptation. It was revealed that coordinated regulation of multiple pathways involving metabolism, ABC transporters, regulators, enterobactin biosynthesis and uptake, the ribosome, flagellar assembly, and virulence was responsible for the development of ethanol adaptation response in this pathogen. Such knowledge will undoubtedly contribute to the development and implementation of more-effective food safety interventions.

INTRODUCTION

Ethanol adaptation in foodborne pathogens has been a subject of great interest in food safety in the last 2 decades. A number of pathogenic bacteria (e.g., Listeria monocytogenes, Bacillus cereus, Vibrio parahaemolyticus, and Cronobacter sakazakii) are able to adapt to sublethal concentrations of ethanol (1–4). Ethanol adaptation can enhance the tolerance of these pathogens to homologous and heterologous stressing agents commonly applied during food processing and storage, thus increasing microbial food safety risks (5–7). In fact, there is an increasing number of stress-adapted pathogenic bacteria involved in foodborne outbreaks (8). In this context, it is of paramount importance to uncover ethanol adaptation mechanisms in understudied foodborne pathogens, such as Salmonella enterica.

To date, the strategies employed by pathogenic bacteria to respond to ethanol adaptation are poorly characterized, especially at the molecular level. Chiang et al. found that ethanol adaptation increased the ratio of unsaturated to saturated fatty acids, indicating an enhancement in cell membrane fluidity (6). Moreover, two-dimensional gel electrophoresis (2-DE) analysis revealed that the expression of eight proteins was enhanced 1.11- to 1.94-fold, while the expression of seven proteins was reduced 0.22- to 0.64-fold by ethanol adaptation (9). Unfortunately, further identification and functional analysis of these differentially expressed proteins have not yet been reported. Furthermore, traditional gel-based methods such as 2-DE suffer from their lack of proteome coverage, sensitivity, and reproducibility (10). A novel approach, isobaric tags for relative and absolute quantification (iTRAQ), can overcome these shortcomings. The iTRAQ approach has been extensively utilized to characterize bacterial stress response mechanisms at the proteomics level in recent years (11–14). Maserati et al. reported that global proteomic analysis by iTRAQ contributed to a better understanding of the regulatory systems involved in the response of Salmonella enterica subsp. enterica serovar Typhimurium to low water activity (a_w), desiccation, and heat (15). Additionally, quantitative proteomics revealed the important role of YbgC in the survival of Salmonella enterica subsp. enterica serovar Enteritidis in egg white (16). It is therefore expected that this technology will be helpful to provide an insight into the molecular and cellular bases of ethanol adaptation in foodborne pathogens.

Ethanol adaptation in S. Enteritidis was evaluated in our previous study, and it was demonstrated that this bacterium acquired tolerance to normally lethal ethanol challenges upon adaptation to sublethal concentrations (2.5 to 10%) of ethanol, which was defined as the ethanol tolerance response (17). The current work was carried out to unravel ethanol adaptation mechanisms in S. Enteritidis by iTRAQ and mutagenic analyses.

RESULTS AND DISCUSSION

Global changes in the proteome of S. Enteritidis during ethanol adaptation.

Exposure to 5% ethanol for 1 h has been identified as an optimal adaptation condition that induces the highest magnitude of ethanol tolerance response in S. Enteritidis (17). This adaptive response was reconfirmed in the current work, as ethanol-adapted cells exhibited a significantly (P < 0.05) higher survival rate than nonadapted cells under normally lethal ethanol challenge conditions (15% ethanol for 1 h) (Fig. 1). The proteomic response of S. Enteritidis to the aforementioned sublethal treatment (5% ethanol for 1 h) was thus determined by iTRAQ analysis to provide an insight into ethanol adaptation mechanisms. A total of 2,174 proteins were detected and quantified in two independent trials. A significant correlation (P < 0.0001; correlation coefficient, >0.67) between protein expression levels in the two biological replicates for nonadapted and ethanol-adapted groups was observed (Fig. S1), confirming the repeatability of the iTRAQ experiment.

FIG 1.

Ethanol adaptation (5% ethanol for 1 h) induces tolerance to a normally lethal ethanol treatment (15% ethanol for 1 h) in S. Enteritidis. The survival rate was calculated by dividing the surviving population by the initial population (corresponding to 100%). Data are presented as mean ± standard deviation. Different lowercase letters indicate significant differences (P < 0.05).

Considerable cutoff criteria (P < 0.05 and iTRAQ ratios of >1.3 or <0.77) were then employed for protein quantification in the current work. The same iTRAQ ratio cutoff was used by Allan et al. to determine whether a protein was differentially expressed in Streptococcus pneumoniae in response to nitric oxide (18). It should be noted that differential proteins with moderate iTRAQ ratios may also be important to bacterial stress responses. For example, YbgC was upregulated by 1.20- and 1.46-fold, as identified by the iTRAQ experiments, after the exposure of S. Enteritidis to 50% and 80% egg white, respectively; mutagenic analysis further revealed that YbgC was indeed a key protein contributing to S. Enteritidis survival in egg white (16). Therefore, all significantly differentially expressed proteins, including those with moderate iTRAQ ratios, were utilized to uncover ethanol adaptation mechanisms in S. Enteritidis in the current study.

Based on the aforementioned criteria, a total of 138 differentially expressed proteins (56 upregulated and 82 downregulated) were screened (Data Set S1). These proteins were assigned to functional groups by Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis for a better understanding of their cellular functions. The major clustered functional categories were metabolism, ABC transporters, regulators, enterobactin biosynthesis and uptake, the ribosome, flagellar assembly, and virulence (Table 1). Moreover, approximately 20% of proteins were poorly characterized, with an unknown function or with a general function based on predictions only (Table S1). A proposed model highlighting the major proteomic changes, presented in Fig. 2, shows the complex regulatory networks governing ethanol adaptation of S. Enteritidis.

TABLE 1.

Representative functional groups for differentially expressed proteins during ethanol adaptation of S. Enteritidis

Functional group or GenBank/UniProt accession no.	Descriptiona	Gene name	Ratio
Metabolism
Q4VYA5	5-Hydroxyisourate hydrolase	hiuH	1.98
B5QTE0	UPF0345 protein YaiE	yaiE	1.36
A0A1R2HTP8	Histidine biosynthesis bifunctional protein HisIE	hisI	1.34
B5QWI9	Arginine N-succinyltransferase	astA	1.81
P29848	Cysteine synthase	cysM	0.59
M7RD94	Aminotransferase, class I/II	STM1557	0.69
M7RWR3	Ketose-bisphosphate aldolase	gatY	1.52
A0A0F6AY02	Glyoxylate carboligase	gcl	1.43
A0A1R2VYV9	Transketolase protein	STM2340	1.40
EPI76970	PTS system fructose-specific EIIBBC component	fruA	1.30
B5R1U0	Ribulokinase	araB	0.72
A0A0F5BCE2	Succinate dehydrogenase hydrophobic membrane anchor subunit	sdhD	0.73
A0A1R2PZL4	PTS system glucitol/sorbitol-specific IIBC component	srlE	0.49
B5R5F8	Phosphopantetheine adenylyltransferase	coaD	1.36
P0A1A1	Thioesterase family protein	yciA	0.61
AMN27763	Penicillin binding protein	mrdA	0.73
A0A2R4HP12	NDH-1 subunit M	nuoM	1.37
B5R2Z7	NADH-quinone oxidoreductase subunit N	nuoN	1.38
AIN67540	Polyprenyl synthetase	ispA	0.69
ABC transporters
Q8ZML1	Glycine betaine-binding periplasmic protein	proX	1.65
P17328	Glycine betaine/l-proline transport ATP-binding protein ProV	proV	1.42
E8XHU4	Phosphate transport system permease protein	pstC	1.56
Q8ZKX5	Phosphate-binding protein PstS	pstS	1.76
A0A0F7J509	Thiamine/thiamine pyrophosphate ABC transporter	tbpA	1.35
A0A2R4HK85	Heme exporter protein	ccmC	1.36
EPI87418	ABC transporter, substrate-binding protein, family 3	dalS	1.31
P36638	Peptide transport system ATP-binding protein SapF	sapF	0.68
AHO69181	Manganese transport system ATP-binding protein MntA	sitB	0.71
Regulators
B5R4S3	Sigma factor-binding protein Crl	crl	1.40
B5R1C8	DNA-binding protein Fis	fis	0.52
A0A0F6B247	Regulator of RpoS	rssB	1.33
B5QV29	HTH-type transcriptional repressor PurR	purR	0.75
A0A1R2QC05	HTH-type transcriptional regulator MetR	metR	0.68
P0A9Y5	Cold shock protein CspA	cspA	0.47
Ribosome
B5QUQ1	50S ribosomal protein L34	rpmH	0.70
B5R1F9	50S ribosomal protein L30	rpmD	0.72
B4TKK2	30S ribosomal protein S14	rpsN	0.74
B5R0L7	30S ribosomal protein S9	rpsI	0.75
AIN66722	ATP-dependent RNA helicase RhlE	rhlE	0.66
A0A2R4HLT0	ATP-dependent RNA helicase DeaD	deaD	0.70
EPJ03745	ATP-dependent RNA helicase DbpA	dbpA	0.74
B5R578	GTPase Der	der	0.73
B5R0H4	GTPase Obg	obg	0.74
B5QZV7	Ribosome-binding factor A	rbfA	0.71
B5QUG2	tRNA (guanine-N1-)-methyltransferase	trmD	0.72
Enterobactin biosynthesis and uptake
AIN08681	2,3-Dihydroxybenzoate-2,3-dehydrogenase	entA	0.59
Q8ZR31	Isochorismatase	entB	0.71
AIN08678	Isochorismate synthase	entC	0.67
AIN08671	Enterobactin synthetase component F	entF	0.73
B5QVK0	Proofreading thioesterase EntH	entH	0.72
EPI62949	Colicin I receptor	cirA	0.56
V7IPX8	Ferrienterobactin receptor	fepA	0.60
B5R716	Ferrienterobactin-binding periplasmic protein	fepB	0.76
A0A0F6AY78	Achromobactin ABC transporter, ATP-binding protein	fepC	0.70
Flagella
P16323	Flagellar basal body protein	flgF	0.63
AHQ29209	Flagellar assembly protein FliH	fliH	0.75
Salmonella pathogenicity island
AIN69063	Outer membrane autotransporter barrel domain protein	misL	0.54
AIN69734	Type III secretion apparatus lipoprotein, YscJ/HrcJ family	prgK	0.68
P0A1B9	Invasion protein InvC	invC	0.69
AIN02111	Type III secretion apparatus needle protein	prgI	0.72

^aPTS, phosphotransferase; HTH, helix-turn-helix.

FIG 2.

A proposed model for major cellular changes occurring during ethanol adaptation of S. Enteritidis. The star symbol (★) indicates that the function of a protein belonging to this pathway has been validated by mutagenic analysis in the current study.

Metabolism.

S. Enteritidis altered the expression of a considerable proportion of metabolism-related proteins in the current study (Table 1), reflecting a coordinated regulation of metabolic processes in response to ethanol adaptation. These differentially expressed proteins belong to carbohydrate, terpenoid and polyketide, energy, amino acid, glycan, cofactor and vitamin, lipid, and nucleotide metabolic pathways. Some proteins involved in coenzyme A (CoA) biosynthesis (CoaD), the pentose phosphate pathway (STM2340), or oxidative phosphorylation (NuoM and NuoN) or in purine (HiuH and YaiE), galactose (GatY), glyoxylate (Gcl), histidine (HisIE), or arginine (AstA) metabolism were enhanced 1.34- to 1.98-fold. These diverse metabolic upregulations suggested the nutritional requirements of S. Enteritidis for ethanol adaptation.

The relationship between amino acid biosynthesis and microbial ethanol tolerance has been reported before. Transcriptome analysis revealed that biosynthetic pathways for amino acids (e.g., histidine, tryptophan, and branched-chain amino acids) were commonly upregulated in ethanol-tolerant strains of Escherichia coli obtained by parallel evolution (19). Furthermore, Hirasawa et al. found that overexpression of tryptophan biosynthesis genes or supplementation of tryptophan to the culture medium conferred ethanol tolerance to Saccharomyces cerevisiae (20). In the current study, it was noted that several proteins (e.g., HisIE, CysM, and STM1557) related to amino acid metabolism were differentially expressed. In particular, histidine biosynthesis bifunctional protein HisIE was elevated by 1.34-fold in ethanol-adapted S. Enteritidis (Table 1). Therefore, amino acid biosynthesis seems to be involved in ethanol adaptation of S. Enteritidis.

ABC transporters.

A large number of differentially expressed proteins were related to ABC transporters in the current study (Table 1). Briefly, proteins responsible for transport of glycine betaine (ProX and ProV), phosphate (PstC and PstS), d-alanine (DalS), thiamine (TbpA), and heme (CcmC) were upregulated; while those for manganese (SitB), cationic peptide (SapF), and siderophore (FepB and FepC) were downregulated. The differential expression of such a sizable fraction of ABC transporters certainly highlights their importance in ethanol adaptation of S. Enteritidis. There were only two transporters, ProVWX and PstSCAB, with more than one upregulated protein.

ProVWX, one of the three osmoprotectant systems in S. enterica, utilizes ATP hydrolysis to drive transport of compatible solutes (21). The pore, ATPase, and substrate binding proteins in this system are called ProW, ProV, and ProX, respectively. The contribution of compatible solutes (e.g., glycine betaine) transported by ProVWX to bacterial survival under NaCl stress has been outlined (22, 23). In the current study, the expression of ProX and ProV was enhanced 1.65- and 1.42-fold, respectively (Table 1). Nevertheless, ethanol-adapted S. Enteritidis did not mount cross-protection against NaCl (17), reflecting the possibility that regulatory pathways mediating osmotic stress tolerance may be different from those involved in the ethanol stress response.

Bacteria have developed intricate strategies to sense and respond to changes in environmental phosphate, thus maintaining intracellular phosphate pools, which are essential for their survival (24). The Pho regulon mediates the response of S. enterica and E. coli to phosphate starvation conditions. In this regulon, the PstSCAB transporter senses phosphate concentrations and communicates with the two-component system PhoRB via PhoU (25). The genes belonging to this signal transduction pathway in E. coli are only expressed when external phosphate is limited (24). In the current study, PstS (a phosphate binding protein), PstC (a phosphate transport system permease protein), and PhoU (a phosphate-specific transport system accessory protein) were upregulated following ethanol adaptation (Table 1; Data Set S1). It was therefore indicative that a phosphate limitation response was triggered by the sublethal ethanol treatment. Similarly, the expression of genes encoding the PstSCAB transporter was enhanced in response to acid and oxidative stresses in S. enterica (25, 26). Overall, maintaining cellular phosphate homeostasis may be essential for S. Enteritidis to mount an adaptive response.

Regulators.

Regulators are indispensable for S. enterica to mount appropriate responses to food processing and storage-related stresses (26–28). Altogether, six regulators (Crl, Fis, RssB, PurR, MetR, and CspA) in S. Enteritidis showed differential expression in the current study. Crl and RssB were upregulated by 1.40- and 1.33-fold, respectively, while Fis was downregulated by 1.92-fold (Table 1). Interestingly, all three of these proteins are involved in the regulation of the sigma factor RpoS in S. enterica. Crl is an unconventional transcription factor known to enhance RpoS activity by a direct interaction, thus controlling the expression of RpoS-regulated genes (29, 30). The RssB response regulator plays a central role in RpoS degradation by delivering it to the ClpXP protease (31, 32); mutation of rssB (mviA) led to a higher level of RpoS and stronger acid tolerance in S. Typhimurium (33). Fis acts as a regulator that mediates the transcriptional induction of RpoS (34). In fact, RpoS abundance is regulated at many levels, including those of protein activity, protein turnover, transcription, and translation (35). Although RpoS was not detected by iTRAQ in the current study, a significant increase in rpoS mRNA level was observed upon ethanol adaptation (36). The importance of RpoS to the survival of S. enterica under food processing and storage-related stresses has been well documented (27, 28, 37). Therefore, it is reasonable to speculate that Fis, RssB, and Crl play a role in ethanol adaptation of S. Enteritidis, considering their regulatory effect on RpoS.

PurR is a transcriptional repressor of purine nucleotide biosynthesis in S. enterica (38). In the current study, ethanol adaptation led to a repression of PurR, along with the induction of HiuH and YaiE, which are involved in purine metabolism (Table 1). Cho et al. provided evidence for the involvement of PurR in bacterial stress responses; deletion of purR decreased the expression of acid tolerance genes (e.g., hdeA, hdeB, and hdeD) in E. coli K-12 MG1655 (39). In this sense, downregulation of PurR in the current work correlated well with our previous finding that ethanol adaptation failed to induce cross-protection in S. Enteritidis against hydrochloric, citric, lactic, ascorbic, and acetic acids (36).

MetR is a transcription factor of the LysR family that regulates the expression of methionine biosynthesis genes (e.g., metE, metF, and metH) in S. enterica (40). MetR and two other proteins (STM1557 and CysM) involved in methionine and cysteine metabolism showed downregulation in the current study (Table 1). Furthermore, the above-mentioned six proteins displayed a close interaction by STRING (Search Tool for the Retrieval of Interacting Genes/Proteins) analysis (Fig. S2). These results indicate the repression of the methionine biosynthesis pathway by ethanol adaptation. In fact, methionine biosynthesis is generally sensitive to environmental signals such as heat shock and oxidative stress (41).

CspA, the major and best-characterized cold shock protein, acts as a global regulator by binding to mRNA or single-stranded DNA. It can influence the transcriptional and translational properties of bacterial cells (42). For instance, Rennella et al. highlighted the RNA binding and chaperone activity of CspA in E. coli (43). In S. Typhimurium, CspA targeted about 25% of the RNA encoded by the genome. These targets were responsible for stress responses, motility, virulence, metabolic processes, cellular transport, transcription regulation, and metal binding (44). Further mutational analysis provided evidence that cold shock proteins were involved in ethanol, pH, and NaCl stress responses of Clostridium botulinum (45). In the current study, ethanol adaptation reduced the expression of CspA by 2.13-fold (Table 1). Meanwhile, our previous study showed that cross protection against –20°C occurred in S. Enteritidis following ethanol adaptation (17). Hence, the role of CspA in ethanol adaptation and its induced cross-protection effect against freezing temperature in S. Enteritidis can be anticipated.

Ribosome.

The expression of ribosome-related proteins was decreased in ethanol-adapted cells of S. Enteritidis in the current study. As shown in Table 1, GTPases (Der and Obg) and ATP-dependent RNA helicases (DeaD, RhlE, and DbpA), which are involved in ribosome maturation in at least E. coli (46), were repressed. The ribosome binding factor (RbfA) and tRNA methyltransferase (TrmD) were also downregulated. In a similar vein, lower expression levels were observed for 30S (RpsN and RpsI) and 50S (RpmH and RpmD) ribosomal proteins. The number of ribosomal units determines the rate of protein synthesis, which is strictly related to the rate of cellular growth (47). This means that the amount of ribosome-related proteins plays an important role in the bacterial growth rate. In our previous study, sublethal concentrations of ethanol (2.5 to 10%) used for ethanol adaptation inhibited the growth of S. Enteritidis (17). This inhibitory effect can be explained by the repression of ribosome-related proteins under ethanol adaptation conditions.

Enterobactin biosynthesis and uptake.

Enterobactin is a major siderophore produced by S. enterica under iron restriction conditions to solubilize exogenous iron, thereby making this metal available for bacterial cells (48). In the current study, ethanol adaptation led to a repression of proteins responsible for enterobactin biosynthesis (EntA, EntB, EntC, EntF, and EntH; 1.38- to 1.69-fold) and uptake (CirA, FepA, FepB, and FepC; 1.32- to 1.78-fold) (Table 1). In fact, there is increasing evidence implicating the role of enterobactin in bacterial stress responses. For example, enterobactin biosynthesis and uptake were induced following exposure of S. Enteritidis to egg white to facilitate iron acquisition, thus providing a survival advantage to this bacterium (16, 49). Peralta et al. found that enterobactin protected E. coli against oxidative stress and that this effect was independent of its capacity to scavenge iron (50). Moreover, the growth of the E. coli ΔtolC mutant was impaired by the accumulation of periplasmic enterobactin (51). Therefore, in the current study, blocking enterobactin synthesis and uptake pathways could be an adaptation strategy of S. Enteritidis under sublethal ethanol stress.

Flagella.

As shown in Table 1, two proteins (FliH, a flagellar assembly protein, and FlgF, a flagellar basal body rod protein) related to flagellar assembly were repressed during the course of ethanol adaptation. Similarly, flgF in S. Typhimurium was downregulated in response to oxidative stress, which might serve as an energy conservation strategy (25). Moreover, the expression of both fliH and flgF was reduced when an acid tolerance response was stimulated in S. Typhimurium; further functional analysis revealed that mutation of flgD, which encodes a scaffolding protein required for flagellar hook formation, led to the absence of an acid adaptation phenotype (26). Hence, the results above enforce the knowledge of flagellum-related proteins playing a crucial role in the stress response of S. enterica.

Salmonella pathogenicity island.

During food processing and storage, foodborne pathogens encounter many of the same stresses that they experience during host infection. Therefore, many stress tolerance-related genes are also likely involved in bacterial survival within the host, and stress adaptation can thus alter the virulence potential of foodborne pathogens (52). In the current study, four Salmonella pathogenicity island (SPI)-related proteins (PrgI, PrgK, MisL, and InvC) were downregulated by 1.39- to 1.85-fold after ethanol adaptation (Table 1). SPIs are highly conserved across the genus and are essential for virulence (26). The prgI and prgK genes encode secretion apparatus proteins of SPI-1 (53). The misL gene encodes an outer membrane autotransporter in SPI-3 (54). The invasion gene invC is a key component of the type III secretion system (55). Ryan et al. found that several virulence factors in S. Typhimurium were differentially regulated during acid adaptation, including invACE in SPI-1 and ssaCGJNQRV in SPI-2 (26). Furthermore, heat shock led to a repression of SPI-1 genes (e.g., prgK and prgH) and an induction of SPI-2 and SPI-5 genes in S. Typhimurium, accompanied by a greater adhesion to Caco-2 cells (56). There is still a lack of knowledge on the virulence of ethanol-adapted S. Enteritidis, which can be further addressed in future studies.

Validation of iTRAQ results at the mRNA level.

The iTRAQ data were verified by reverse transcription-quantitative real-time PCR (RT-qPCR) to determine the transcriptional profile of ten differentially regulated proteins. Six proteins (ProV, SecE, STM2506, YlaC, HiuH, and ProX) were upregulated and four (EntA, FepA, SitB, and CspA) were downregulated. As shown in Fig. 3, nine of the ten proteins and their corresponding mRNAs displayed a similar expression pattern, and the only exception was ProV. This finding provided evidence for the reliability of data derived from the aforementioned proteomic analysis.

FIG 3.

Comparison between protein and mRNA levels of ten differentially regulated proteins revealed by iTRAQ and RT-qPCR. An asterisk (*) signifies that a gene in S. Enteritidis was significantly (P < 0.05) differentially expressed in response to sublethal ethanol adaptation (5% ethanol for 1 h).

Functional analysis of HiuH and ProX in ethanol adaptation of S. Enteritidis.

In the current study, we hypothesized that highly upregulated proteins revealed by iTRAQ are important to ethanol adaptation of S. Enteritidis. Therefore, two proteins (HiuH, involved in nucleotide metabolism, and ProX, associated with the ABC transporter) that showed enhanced expression in both iTRAQ and RT-qPCR tests were selected for mutagenic analysis. We constructed S. Enteritidis mutants in which hiuH or proX was deleted and compared their abilities to develop an ethanol tolerance response with that of the wild-type (WT) strain. No significant difference (P > 0.05) was found in the growth curve of wild-type, ΔhiuH, and ΔproX strains in Luria-Bertani (LB) broth (Fig. S3). Nevertheless, ΔhiuH and ΔproX mutants mounted a significantly (P < 0.05) higher ethanol tolerance response than that of the wild-type strain (Fig. 4). Furthermore, the ethanol tolerance response was restored in complemented strains (Fig. 4), thereby confirming the negative regulatory role of these two proteins in ethanol adaptation of S. Enteritidis.

FIG 4.

Ethanol tolerance response in the wild type (WT), deletion mutants (ΔhiuH and ΔproX) and complemented strains (ΔhiuH-C and ΔproX-C) of S. Enteritidis. Ethanol-adapted cells (5% ethanol for 1 h) were further exposed to 15% ethanol for 4 h. The survival rate, calculated by dividing the initial population size (corresponding to 100%) by the surviving population size, was then employed to assess the development of an ethanol tolerance response. A survival rate of 50% indicates that the population of S. Enteritidis cells was reduced by half. Data are presented as mean ± standard deviation. Different lowercase letters indicate significant differences (P < 0.05).

The 5-hydroxyisourate (5-HIU) hydrolase HiuH is involved in bacterial purine metabolism (57). This pathway requires four enzymatic steps that convert xanthine to uric acid, uric acid to 5-HIU, 5-HIU to OHCU (2-oxo-4-hydroxy-4-carboxy-5-ureidoimidazoline), and OHCU to allantoin (58). HiuH catalyzes the third step of this reaction to metabolize 5-HIU to OHCU (59). Hennebry et al. found that mutation of hiuH (yedX) did not affect the responses of S. Typhimurium to oxidative stress, reduced nutrient provision, and temperature alteration (58). In the current study, it was demonstrated that the deficiency of hiuH contributed to the development of an ethanol tolerance response in S. Enteritidis. Although the mechanism for this negative regulation is unclear, these findings suggest that the purine metabolism pathway is involved in ethanol adaptation of S. Enteritidis.

Deletion of proX also led to a significantly (P < 0.05) higher magnitude of ethanol tolerance response in the current study. Similarly, the tolerance of E. coli to n-hexane and cyclohexane was improved by the disruption of proX (60). These observations provided evidence for the involvement of the ProVWX uptake system in bacterial organic solvent tolerance. In the ProVWX system, ProX recognizes a compatible solute and delivers it to a protein complex consisting of ProV and ProW. The ProVWX transporter permits the uptake of various compatible solutes (e.g., glycine betaine, ectoine, taurine, proline, and structural analogues of glycine betaine) involved in bacterial stress responses (61). It was therefore speculated that disruption of the proX gene improved the ethanol tolerance response of S. Enteritidis by acting on the intracellular concentration of these solutes. Taken together, mutational analysis supports our hypothesis that highly upregulated proteins, such as HiuH and ProX, play a role in ethanol adaptation of S. Enteritidis.

Conclusions.

Proteomic characterization revealed that complex regulatory pathways associated with metabolism, ABC transporters, regulators, enterobactin biosynthesis and uptake, the ribosome, flagellar assembly, and virulence were at play during ethanol adaptation of S. Enteritidis. Moreover, mutagenic and complementation analyses demonstrated a negative regulatory role of ProX and HiuH in ethanol adaptation. Collectively, our work provides important insights into ethanol adaptation mechanisms of S. Enteritidis, as well as a framework for further investigation on this subject. For example, functional analysis of more proteins belonging to different pathways will deepen our understanding of ethanol adaptation in S. Enteritidis. It would also be interesting to address the effect of ethanol adaptation on the virulence properties of S. Enteritidis in future studies.

MATERIALS AND METHODS

Bacterial strains and storage conditions.

S. Enteritidis strain ATCC 13076, obtained from the American Type Culture Collection, was used in the current study. The bacterial strain was maintained in LB broth supplemented with 25% glycerol at −80°C and streaked onto LB agar, followed by incubation at 37°C for 24 h prior to use. For each experiment, a single colony was inoculated in 5 ml LB broth and incubated overnight at 37°C. A 500-μl aliquot of the active culture was inoculated into 50 ml LB broth and incubated at 37°C and 200 rpm for 5 h to reach the late exponential phase (17).

Ethanol adaptation assays.

The ethanol adaptation assay was carried out as previously described (17). Briefly, late-exponential-phase cultures (5 ml) of S. Enteritidis were centrifuged, washed with phosphate-buffered saline (PBS, pH 7.4), and resuspended in 50 ml fresh LB broth (control) or in LB broth containing 5% (vol/vol) ethanol. These samples were incubated at 25°C with shaking (170 rpm) for 1 h to prepare nonadapted and ethanol-adapted cultures for iTRAQ testing. Moreover, these two cultures were subjected to ethanol tolerance assessment according to our previously described method (17).

Protein extraction, quantification, and digestion.

Nonadapted and ethanol-adapted cells of S. Enteritidis were washed twice with PBS, resuspended in SDT buffer (1 mM dithiothreitol [DTT], 4% SDS, and 150 mM Tris-HCl [pH 8.0]), boiled for 5 min, and ultrasonicated for another 5 min. The lysates were centrifuged at 14,000 × g for 10 min to remove cellular debris (16). The resulting supernatants were transferred to new tubes and stored at −80°C for subsequent use. Bicinchoninic acid (BCA) protein assay reagent (Promega, Madison, WI) was then utilized to determine the protein concentration.

Protein digestion was performed according to a filter-aided sample preparation (FASP) protocol (62). Briefly, 200 μg protein from each sample was mixed with 30 μl SDT buffer and was washed three times by ultrafiltration (Pall units, 10 kDa) with 200 μl UA buffer (8 M urea and 150 mM Tris-HCl [pH 8.0]). Proteins were then alkylated with 50 mM iodoacetamide in the dark for 30 min, then washed three times with 100 μl UA buffer and then twice with 100 μl dissolution buffer (50 mM triethylammonium bicarbonate [pH 8.5]). Finally, proteins were digested overnight with 2 μg trypsin (Promega, Madison, WI) in 40 μl dissolution buffer at 37°C. The resulting peptides were collected as a filtrate in clean tubes by centrifugation at 14,000 × g for 10 min. The peptide content was estimated by UV light spectral density at 280 nm using an extinction coefficient of 1.1 for a 0.1% solution that was calculated on the basis of the frequency of tryptophan and tyrosine in vertebrate proteins.

iTRAQ labeling and strong cation-exchange chromatography fractionation.

The 8 multiplex iTRAQ labelings were carried out according to the manufacturer’s instructions (Applied Biosystems, Foster City, CA). iTRAQ reagents 113 and 114 were employed to label the peptides from nonadapted S. Enteritidis, whereas reagents 117 and 118 were utilized to label the peptides from ethanol-adapted S. Enteritidis. Four other labels (115, 116, 119, and 121) were used in other experiments. Samples were combined and vacuum dried after labeling. The iTRAQ-labeled peptides were dissolved in 2 ml buffer A (10 mM KH₂PO₄ in 25% acetonitrile, pH 3.0) and fractionated using an AKTA purification system (GE Healthcare, Sweden) and a PolySULFOETHYL A column (4.6 × 100 mm, 5 μm; PolyLC, Columbia, MD). The gradient elution was conducted with 0% to 10% buffer B (500 mM KCl and 10 mM KH₂PO₄ in 25% acetonitrile, pH 3.0) for 32 min, 10% to 20% buffer B for 10 min, 20% to 45% buffer B for 5 min, and 45% to 100% buffer B for 13 min. The tryptic peptides were separated at a flow rate of 1,000 μl/min and monitored by absorbance at 214 nm. The fractions were collected every minute, combined into 15 pools, desalted using C₁₈ cartridges (Empore SPE; Sigma, St. Louis, MO), and concentrated by vacuum centrifugation.

Liquid chromatography-electrospray ionization-tandem mass spectrometry analysis.

The labeled peptides were analyzed on a Q Exactive mass spectrometer coupled to an Easy-nLC liquid chromatography system (Proxeon Biosystems, Thermo Fisher, Fairlawn, NJ) and equipped with a C₁₈ trap column (5 μm, 100 μm × 20 mm) and a C₁₈ analytical column (3 μm, 75 μm × 100 mm). A 10-μl aliquot of sample was loaded along with reversed-phase (RP)-C₁₈ 5-μm resin in buffer A (0.1% formic acid). Separation was achieved using a linear gradient of buffer B (84% acetonitrile in 0.1% formic acid) controlled by IntelliFlow technology at a flow rate of 250 nl/min. A data-dependent top 10 method was utilized to acquire mass spectrometry (MS) data, dynamically selecting the most abundant precursor ions from the survey scan (300 to 1,800 m/z) for higher-energy collisional dissociation (HCD) fragmentation. The predictive automatic gain control (pAGC) system was employed with the instrument, using a dynamic exclusion duration of 60 s to determine the target value. Survey scans were obtained at m/z 200 at a resolution of 70,000, and resolution for HCD spectra was set to 17,500 at m/z 200. Normalized collision energy was 30 eV, and the underfill ratio was defined as 0.1%. The instrument was run with the peptide recognition mode enabled.

Data analysis.

Tandem mass spectrometry (MS/MS) spectra were searched against the S. Enteritidis UniProt database (37,314 sequences downloaded on 22 June 2016) and the decoy database using Mascot 2.2 (Matrix Science, London, UK) embedded in Proteome Discoverer 1.4 (Thermo Electron, San Jose, CA). The following parameters were used for protein identification: enzyme, trypsin; MS/MS tolerance, 0.1 Da; missed cleavage, 2; variable modification, Oxidation (M); fixed modification, Carbamidomethyl (C); iTRAQ8plex(K), iTRAQ8plex(N-term); false discovery rate (FDR), ≤0.01.

Proteins with iTRAQ ratios of >1.3 (increased) or <0.77 (decreased) and a P value of <0.05 were considered to be differentially expressed. The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis (http://www.genome.jp/kegg/pathway.html) was employed to determine the metabolic pathway for all differentially expressed proteins.

Gene expression analysis.

Differentially expressed proteins in iTRAQ tests were validated at the mRNA level by RT-qPCR. A total of ten differentially regulated proteins were selected to determine their corresponding transcription levels. RNA extraction and RT-qPCR analysis were carried out on nonadapted and ethanol-adapted S. Enteritidis cells as previously described (36), using primers listed in Table 2. Alterations of gene expression in ethanol-adapted cells compared to that in nonadapted counterparts were calculated by the cycle threshold (2^−ΔΔCT) method. The 16S rRNA gene was employed as a nonregulated control for data normalization.

TABLE 2.

Primers used for RT-qPCR analysis

Gene	Forward primer sequence (5′ to 3′)	Reverse primer sequence (5′ to 3′)
proV	CTCGGGTAAATCCACAA	TTATCCAGCACGGTCAT
secE	CAATCGTCGGCAACTAC	TACCAGGCGAACCAGA
STM2506	CGCATGACCCGTATCGT	CGGCGTGGTGACAGAAA
ylaC	AGCGAAACTATTGATGAC	CCGTTGTAACAGACCC
hiuH	CAGCAAACAGGCAAAC	TAATAACAGCGGCACA
proX	GGCATTACCGTCCAAC	CGACTTCACTCGGCTTA
entA	TTTGCGGTCAATGTGGG	GCTGTTCGGCATCTTCG
fepA	CGTATCCACCATCACCG	ACTCGCTACCGCCTTTT
sitB	TGGTAGGCGTAAATGGT	CCCTGGCAAGAAACAC
cspA	TTCGGCTTTATTACTCCTG	CTTTCTGACCTTCGTCCA
16S rRNA	CAGAAGAAGCACCGGCTAAC	GACTCAAGCCTGCCAGTTTC

Generation of hiuH and proX deletion mutants and complementing strains.

In-frame deletions of hiuH and proX were performed according to previously described homologous recombination knockout procedures using primers in Table 3 (63). Plasmids and strains used are listed in Table 4. The fragments of homologous arms were obtained from S. Enteritidis genomic DNA by overlap extension PCR. This product was cloned into the pMD₁₉-T vector (TaKaRa, Dalian, China) to generate pMD₁₉ΔhiuH and pMD₁₉ΔproX, respectively. The correct construction was confirmed by DNA sequencing. Both pMD₁₉ΔhiuH and pMD₁₉ΔproX were digested with SacI and XbaI and then ligated into pRE112 (a suicide vector carrying a sucrose-sensitive gene and a chloramphenicol resistance gene). The resulting pREΔhiuH and pREΔproX plasmids were introduced into E. coli SM10λpir by CaCl₂ transformation. These two plasmids were then extracted from E. coli cells and transformed into the wild-type S. Enteritidis ATCC 13076 by electroporation (2,400 V, 4.2 ms) to accomplish a single crossover. The single-crossover strains were grown in LB broth supplemented with 8% sucrose to accomplish a second crossover. Colonies that were resistant to sucrose and sensitive to chloramphenicol were selected. The resulting mutants, S. Enteritidis ΔhiuH and S. Enteritidis ΔproX, were confirmed by PCR analysis and DNA sequencing.

TABLE 3.

Primers used for the construction and complementation of S. Enteritidis deletion mutants

Primer	Sequence (5′ to 3′) (restriction enzyme)a
hiuH-F1	CTCTAGAGCCGTCAGGCAAATAA (XbaI)
hiuH-R1	AGGCTCTAAAGCTTCACTCCTTTACGGTAT
hiuH-F2	GGAGTGAAGCTTTAGAGCCTATCCCATTAG
hiuH-R2	GCGAGCTCAAGCGGGATAACCACC (SacI)
proX-F1	CTCTAGAAGGTGCCTGCCGACTT (XbaI)
proX-R1	AAAAACGATCCGTTGTTCCTTTAATTATGG
proX-F2	AGGAACAACGGATCGTTTTTTATGCCGGAT
proX-R2	GCGAGCTCTGCTAAGCGACTGACTGC (SacI)

^aThe 20-bp overlap sequences that were engineered into primers for amplification of the fragments of homologous arms are in bold, and restriction sites are underlined.

TABLE 4.

Strains and plasmids used for the construction and complementation of S. Enteritidis deletion mutants

Strain or plasmid	Description	Source or reference
S. Enteritidis ATCC 13076	Wild-type strain	American Type Culture Collection
ΔhiuH	hiuH deletion mutant of S. Enteritidis ATCC 13076	This study
ΔhiuH-C	Complementary strain for hiuH deletion mutant	This study
ΔproX	proX deletion mutant of S. Enteritidis ATCC 13076	This study
ΔproX-C	Complementary strain for proX deletion mutant	This study
E. coli DH5α	Host for cloning	Laboratory stock
E. coli SM10(λpir)	thi thr-1 leu6 proA2 his-4 arg E2 lacY1 galK2 ara14 xyl5 supE44 λpir	Laboratory stock
pMD19-T	Cloning vector, Ampr	TaKaRa, Japan
pRE112	pGP704 suicide plasmid, pir dependent, oriT oriV sacB, Cmr	Laboratory stock
pRE112-ΔhiuH	pRE112 containing a 686-bp hiuH deletion PCR product	This study
pRE112-ΔhiuH-C	pRE112 containing a wild-type copy of hiuH and its two flanking sequences; used to complement strain ΔhiuH	This study
pRE112-ΔproX	pRE112 containing a 905-bp proX deletion PCR product	This study
pRE112-ΔproX-C	pRE112 containing a wild-type copy of proX and its two flanking sequences; used to complement strain ΔproX	This study

To generate complemented strains, the constructed plasmids pREΔhiuH-C and pREΔproX-C were transferred into the corresponding mutant strains by electroporation at 2,400 V for 4.2 ms. A double selection was then carried out as described above. The complementation of these two genes was confirmed by PCR and DNA sequencing.

Determination of the role of hiuH and proX in ethanol adaptation.

The role of hiuH and proX in ethanol adaptation of S. Enteritidis was assessed by comparing the capacity of their deletion mutants (ΔhiuH and ΔproX) and complemented strains (ΔhiuH-C and ΔproX-C) to develop an ethanol tolerance response with that of the wild type (WT). The ethanol tolerance response, defined as the induced tolerance to normally lethal ethanol challenge conditions following adaptation to mild ethanol stress, was determined as previously detailed (17). The wild type, deletion mutants, and complemented strains of S. Enteritidis were adapted in 5% ethanol for 1 h as described above. Ethanol-adapted cells (100 μl) were then inoculated into 10 ml LB broth containing 15% ethanol. The viable bacterial population was determined after incubation at 25°C/170 rpm for 4 h by plating the appropriate dilutions onto LB agar. The survival rate was then calculated by dividing the initial population size (corresponding to 100%) by the surviving population size.

Statistical analysis.

Gene expression levels and survival rates were subjected to a one-way analysis of variance (ANOVA) with SAS version 8.0 (SAS Institute Inc., Cary, NC). Duncan’s test was then employed to detect the statistical significance at the level of P < 0.05.