Correspondence: Mónica N. Giacomodonato, CEFYBO-CONICET, Departamento de Microbiología, Parasitología e Inmunología, Facultad de Medicina, Universidad de Buenos Aires, Buenos Aires, Argentina. Tel.: +54 11 5950 9500; fax: +54 11 4964 2554; e-mail: firstname.lastname@example.org
DNA adenine methylation regulates virulence gene expression in certain bacteria, including Salmonella Typhimurium. The aim of this study was to investigate the involvement of DNA adenine methylase (Dam) methylation in the expression and secretion of the SPI-1 effector protein SopA. For this purpose, SopA–FLAG-tagged wild-type and dam strains of Salmonella Typhimurium were constructed. The expression and secretion of SopA were determined in bacterial culture and in intracellular bacteria recovered from infected HEp-2 epithelial cells. Bacterial culture supernatants and pellets were used to investigate secreted proteins and cell-associated proteins, respectively. Western blot and quantitative reverse transcriptase PCR analysis showed that the dam mutant expresses lower levels of SopA than the wild-type strain. Interestingly, the strain lacking Dam synthesizes SopA under nonpermissive conditions (28 °C). In addition, SopA secretion was drastically impaired in the dam mutant. In vivo experiments showed that the intracellular Salmonella dam mutant synthesizes SopA although in lower amounts than the wild-type strain. Taken together, our results suggest that Dam methylation modulates the expression and secretion of SopA in Salmonella Typhimurium.
Alteration of DNA adenine methylase (Dam) activity has been shown to attenuate the virulence of several pathogens and to confer protective immune responses in vaccinated animals (Wion & Casadesús, 2006). The molecular basis of virulence attenuation and protection conferred in dam mutant strains appears to involve ectopic gene expression and the resultant elaboration of an expanded repertoire of antigens. Additionally, the low-grade persistence of dam mutant vaccines in appropriate lymphoid tissues (e.g. Peyer's patches) in Salmonella spp. (García del Portillo et al., 1999; Heithoff et al., 2001) and in Yersinia (Julio et al., 2001) may provide a stable source of antigens in sufficient quantity and duration for the transition toward the development of potent adaptive immune responses (Dueger et al., 2001; Heusipp et al., 2007). This suggestion is supported by work with Salmonella wherein loss of the Dam function results in a number of changes in the bacterial physiology. Dam mutants appear to express in vitro a number of genes that are normally only produced in vivo during the initiation and progression of bacterial infection (Heithoff et al., 1999, 2001; Mahan et al., 2000); additionally, both bacteria-associated and secreted proteins are affected by the loss of Dam regulation (García del Portillo et al., 1999; Heithoff et al., 2001; Pucciarelli et al., 2002). A recent report by Balbontín et al. (2006) provided evidence that Dam methylation regulates the invasion genes of the pathogenicity island 1 (SPI-1); they proposed a correlation between specific alterations of gene expression and certain virulence defects of Salmonella dam mutants. The need for Dam methylation to activate the expression of SPI-1 genes seems to provide a straightforward explanation for the reduced secretion of SPI-1 effectors such as SipA, SipB and SipC reported earlier (García del Portillo et al., 1999). Certain effector proteins, such as SopA, whose secretion is mediated by the SPI-1 type III secretion system (TTSS-1), are encoded by genes that are located outside SPI-1. SopA participates in triggering inflammation through its E3 ligase activity (Zhang et al., 2006) and facilitates the bacterial escape out of the Salmonella-containing vacuoles into the cytosol (Zhang et al., 2005). In this work, we found that Dam methylation regulates the expression and secretion of SopA effector protein. These findings would contribute towards the understanding of attenuation of bacteria lacking the Dam protein, proposed as live vaccines.
Materials and methods
Salmonella Typhimurium American Type Culture Collection (ATCC) 14028 and derived strains tagged with the 8-aa FLAG epitope tag peptide were used in this work. Strain SSM2795 (sopA::3 × FLAG) was obtained using the method described by Uzzau et al. (2001). 3 × FLAG epitope tails were added to the ends of the sopA gene. The 3 × FLAG epitope is a sequence of three tandem FLAG epitopes (22 aa). A pair of primers was designed to amplify a 3 × FLAG and kanR coding sequence using plasmid pSUB11 (Uzzau et al., 2001). The 3′ ends of these oligonucleotides were complementary to the first 20 nt of the pSUB11 3 × FLAG coding region (GACTACAAAGACCATGACGG, forward primers) and to the 20 nt of the pSUB11 priming site 2 (CATATGAATATCCTCCTTAG, reverse primers). The 5′-ends of the oligonucleotides were designed to be homologous to the last 40 nt of each tagged gene, not including the stop codon (forward primers), and to the 40 nt immediately downstream of the gene stop codon (reverse primers). The high-frequency generalized transducing bacteriophage P22HT was used for the transduction. The Δdam-230 zge-6313::Tn10dCmR allele was transduced from the Salmonella Typhimurium SV4712 strain (kindly provided by Dr Casadesús) into the tagged strains SSM2795, resulting in a dam strain called STD2795 (sopA::3 × FLAG Δdam-230 zge-6313::Tn10dCmR). Complementation assays were performed using plasmid pIZ833 – a pMM40 derivative carrying the dam gene of Salmonella Typhimurium strain SL1344 under the control of a tac promoter (A.I. Prieto, unpublished data).
Preparation of secreted proteins
Bacterial strains were grown under conditions that either induce or not the expression of SPI-1 genes, as described by Miki et al. (2004). Bacterial culture supernatants and pellets were obtained to investigate secreted proteins and cell-associated proteins, respectively (Pucciarelli et al., 2002). Bacteria were grown overnight, in Luria–Bertani (LB) broth containing 0.3 M NaCl, without aeration, at 37 °C (SPI-1-inducing conditions) or at 28 °C (noninducing conditions). For the isolation of cell-associated proteins, bacterial strains carrying the epitope-tagged gene were grown in 1.5-mL cultures to the stationary phase and centrifuged. Bacterial pellets were resuspended in 100 μL of H2O and immediately mixed with 100 μL of Laemmli lysis buffer (Laemmli, 1970). Suspensions were incubated at 100 °C for 5–10 min. For the isolation of proteins released into the culture supernatants (secreted proteins), cells were pelleted by centrifugation and 2-mL supernatant was collected from each sample. The supernatants were then filtered (0.45-μm pore size), and the proteins were precipitated with 25% TCA and sedimented by high-speed centrifugation (14 000 g for 30 min). The pellet was washed in cold acetone and resuspended in phosphate-buffered saline (PBS) and Laemmli buffer. Four independent extractions for each sample were added together to minimize differences in protein recovery from sample to sample. The proteins were then boiled for 5–10 min, and an aliquot of each sample was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (10% gel) (Raffatellu et al., 2005).
3 × FLAG fusion SopA was immunodetected using mouse anti-FLAG M2-peroxidase mAbs (Sigma). SopA expression and secretion were normalized to 106 CFU and shown as arbitrary units. Detection was performed by chemiluminescence (Luminol, Santa Cruz Biotechnology). Blots were scanned, and the intensity of the signals was determined using the public domain nih image program (http://rsb.info.nih.gov/nih-image/).
Quantitative reverse transcriptase PCR (qRT-PCR)
Total bacterial RNA was extracted using Trizol reagent (Life Technologies Inc., Grand Island, NY). Total RNA (1 μg per sample) was reverse-transcribed using Transcriptor Reverse Transcriptase (Roche, Basel, Switzerland) according to the manufacturer's protocol. qRT-PCR was performed using the SYBR Green PCR kit (PE Applied Biosystems, Foster City, CA) using an Applied Biosystems 7700 sequence detector. Measured mRNA levels were normalized to the mRNA levels of the 16S rRNA gene. The primer sequences were SopA forward, TCCACCGTGAAGTTGATTGA, and reverse, GCACTGAGGATGTGCTGGTA, and 16S forward, TGTAGCGGTGAAATGCGTAG, and reverse, CAAGGGCACAACCTCCAAG. Cycling conditions were 95 °C for 10 min, followed by 45 cycles of 95 °C for 10 s, 55 °C for 10 s and 72 °C for 15 s and one cycle of 40 °C for 30 s.
Six- to eight-week-old BALB/c mice were purchased from the Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires, and kept in our animal house throughout the experiments. All experiments were performed in accordance with the guidelines of the School of Medicine Animal Care and Use Committee. The model described by Jones et al. (1994) was used to study the invasion capacity of ATCC 14028, SSM2795, STD2795 and STD2795/pIZ833 strains of Salmonella Typhimurium. Mice were infected intraloop with 107 CFU of each bacterial strain and sacrificed 60 min later. Ileal loops were aseptically removed and incubated for 60 min in gentamicin before homogenizing. Intracellular bacteria were recovered by plating homogenate dilutions.
Bacterial infection of eukaryotic cells
Human laryngeal epithelial (HEp-2) cells (ATCC, CCL-23) were maintained in DMEM containing 10% fetal bovine serum. Confluent monolayers were inoculated with bacteria grown standing overnight in LB broth, at a multiplicity of infection of 10 : 1. Infected monolayers were then incubated for 20 min at 37 °C in 5% CO2, washed twice with PBS and then incubated in fresh tissue culture medium containing 100 mg mL−1 gentamicin for 30 min to remove extracellular bacteria. Finally, monolayers were washed three times with PBS and lysed with 1% Triton X-100 in PBS to release intracellular bacteria. An aliquot of this suspension was used to determine the number of intracellular bacteria by plating serial dilutions onto LB agar plates. Released bacteria were then prepared for immunoblotting analysis as described above. In selected experiments, intracellular CFU were counted both 20 min and 24 h after infection to determine the replication rate of intracellular Salmonella. Invasion rates of nonphagocytic cells were determined as the ratio of viable intracellular bacteria recovered shortly after infection (20 min) vs. viable bacteria added to infect the eukaryotic cells.
Results and discussion
Epitope tagging does not modify wild-type or dam phenotypes
In order to investigate whether tagging of SopA with the FLAG epitope affects virulence properties, we performed experiments using tagged and untagged wild-type Salmonella Typhimurium. No significant differences in invasiveness or intracellular proliferation were observed in any of the wild-type strains studied (Table 1). These findings are in agreement with an earlier work performed in a murine model of infection (Giacomodonato et al., 2007). It has been well documented that lack of the Dam protein affects DNA methylation status, bacterial invasion and intracellular proliferation capacities (García del Portillo et al., 1999; Heithoff et al., 2001). Our results show that tagging of the SPI-1 gene sopA does not modify the attenuated phenotype of the Salmonella Typhimurium dam mutant. DNA from all the strains tested was cleaved with Sau3AI, which recognizes both methylated and unmethylated DNA, but only DNA from the dam mutant was cut with MboI, which requires unmethylated adenine residues; wild-type DNA was cleaved with DpnI, which digests methylated adenine residues. The digestion pattern was restored to that of the wild type in the dam mutant complemented with pIZ833 plasmid, indicating that the differences observed between dam and wild-type methylation patterns are due to lack of the Dam protein (data not shown). In line with previous studies on Salmonella dam mutants (García del Portillo et al., 1999), we found that tagged strain STD2795 (sopA::3 × FLAG Δdam-230 zge-6313::Tn10dCmR) shows defects in invasion within nonphagocytic cells (Table 1). Invasiveness was investigated using both HEp-2 cultured cells and the murine ileal loop; the results are presented in Table 1. Salmonella dam strain was partially impaired for invading nonphagocytic cells (45% of wild-type values in cultured cell and 40% of wild-type values in murine ileal mucosa). A significant defect in the proliferation of the dam strain within HEp-2 cells was also detected (60% of wild-type values). The DNA methylation pattern and proliferation capacity of the mutant were restored by introducing the dam gene cloned in plasmid pIZ833. The invasion defect of the dam mutant, however, could not be fully restored by complementation (Table 1). Failure in restoring certain virulence traits after complementation of dam mutants has been reported earlier (García del Portillo et al., 1999; Heithoff et al., 2001; Balbontín et al., 2006) and could be explained by the fact that overproduction of Dam methylase reproduces certain phenotypes of mutant strains lacking the Dam protein (Torreblanca & Casadesus, 1996; Løbner-Olesen et al., 2005). Altogether, our results show that epitope tagging does not affect the dam phenotype of Salmonella mutants; consequently, tagged strains were used in this work.
Table 1. Invasion and proliferation of tagged strains
Dam mutants of Salmonella Typhimurium synthesize low amounts of SopA
We examined the effect of the absence of Dam on SopA synthesis under SPI-1-inducing (37 °C) and noninducing (28 °C) conditions. The relative amount of the effector protein present in the whole bacterial extract was quantified. As expected, SopA was synthesized by wild-type strain grown under SPI-1 conditions (Fig. 1a). Densitometry analysis showed a significant reduction of SopA synthesis under SPI-1 conditions in the dam mutant compared with the wild-type strain (P<0.01) and the complemented strain (P<0.05) (Fig. 1b). Compared with the wild-type strain, the Salmonella dam mutant expressed (under inducing conditions) only 20% of the amount of SopA. Interestingly, the dam mutant expressed SopA under noninducing conditions (28 °C), whereas no synthesis of the effector was observed in the wild-type or the complemented strain at this temperature. Moreover, no significant differences were found in the amount of SopA expressed by the dam mutant grown at 37 or 28 °C (Fig. 1b). This downregulation of SopA expression in dam mutants was confirmed by qRT-PCR in relation to 16S rRNA gene expression. The transcript level of sopA was decreased in the dam mutant approximately twofold (P<0.05) compared with the wild-type or the complemented strains under SPI-1-inducing conditions. Again, no differences in the expression of sopA were detected in the dam mutant grown under inducing or noninducing conditions (Table 2). These findings suggest that lack of Dam relaxes the temperature regulation of SopA synthesis. Our results are in agreement with previous data obtained from Yersinia spp., showing that Dam overproduction leads to the expression and secretion of Yop virulence proteins under nonpermissive conditions (Julio et al., 2001, 2002). These authors demonstrated that Dam overproduction disrupts both thermal and calcium regulation of YopE synthesis and relaxes the thermal (but not calcium) dependence of YopE secretion.
Table 2. qRT-PCR for sopA expression
Expression levels of sopA gene
SPI-1-inducing conditions (37°C)
SPI-1-non-inducing conditions (28°C)
Relative mRNA amounts were determined by qRT-PCR and related to mRNA levels in wild-type strain grown under inducing conditions, set as 1. Values are means ± SDs from three independent experiments.
Significance of difference (P<0.05) from level in wild-type strain grown under inducing conditions was calculated by Student's t-test.
SopA secretion is impaired in dam mutants of Salmonella Typhimurium
The relative amounts of SopA were assessed in wild-type and dam strain culture supernatants. Under conditions that mimic the intestinal environment (37 °C), only the wild-type strain secreted SopA (Fig. 2a). Despite the fact that SopA was expressed in dam cells grown at 37 and 28 °C (although in low amount), the protein could not be detected in the supernatants at any temperature. Moreover, SopA secretion in the dam mutant was not restored after complementation (Fig. 2a and b). These results are in line with our previous observation that SopA is the effector protein secreted in the lowest amount following intragastrical or intraperitoneal infection of mice with Salmonella Typhimurium (Giacomodonato et al., 2007). Lack of SopA secretion in dam mutants could be explained considering that Dam methylation is required to activate the expression of certain SPI-1 syringe-encoding genes (Balbontín et al., 2006). Nevertheless, dam mutants showed not only impairment in SopA secretion but also a reduced synthesis of this effector protein. These results are in agreement with those reported earlier showing that dam mutants of Salmonella Typhimurium are significantly reduced in their ability to synthesize and secrete SipC (Balbontín et al. 2006). It would be premature, however, to conclude that dam strains are totally unable to secret SopA. On the one hand, it is possible that the amount of SopA secreted by dam mutants is below the detection limit of our method. On the other, Zhang et al. (2005) demonstrated that SopA has a short life span.
A reduced secretion of the SPI-1 effector proteins such as SipA, SipB and SipC in Salmonella dam mutants has been reported earlier (García-Del Portillo et al., 1999). More recently, Balbontín et al. (2006) identified genes that are up- or downregulated in dam mutants, indicating that Dam methylation represses or activates their expression. Therefore, our results suggest that the sopA gene is activated by Dam methylation. In contrast to other Salmonella effector proteins, such as SopB, SopD and SopE2, relatively little is known about SopA. An earlier work by Wood et al. (2000) demonstrated a role for SopA in the Salmonella-induced movement of polymorphonuclear leukocytes across the intestinal epithelium. In addition, Zhang et al. (2002) showed that SopA acts in concert with other TTSS-1 secreted effector proteins. In addition, Layton et al. (2005) have reported that SopA localizes in the mitochondria; the correlation of this fact with the role of SopA in virulence remains unknown.
Dam methylation triggers SopA synthesis in Salmonella Typhimurium intracellular bacteria
We investigated the synthesis of SopA in dam mutants during early stages of HEp-2 cells infection. Confluent HEp-2 cells were infected with SopA–FLAG-tagged Salmonella mutants. Intracellular bacteria were collected 20 min after infection. As shown in Fig. 3a, the epitope-tagged SopA could be detected specifically. Soon after infection, SopA appears to be expressed at a lower level in dam mutant STD2795 compared with wild-type strain SSM2795 (P<0.05). Of note, the amount of SopA detected intracellularly in wild-type and dam strains at 20 min postinfection (Fig. 3) was fivefold higher than that synthesized by bacteria cultured under SPI-1-inducing condition (Fig. 1).
Previously, we demonstrated that the expression of SPI-1 genes persists for several hours after infection of mice (Giacomodonato et al., 2007). Then, we investigated whether SopA is expressed in dam mutants during late stages of HEp-2 cells infection. For this purpose, HEp-2 cells were infected and the intracellular bacteria were collected 24 h after infection. The results are shown in Fig. 4a and b. We found that SopA is expressed in intracellular bacteria 24 h after infection. Once again, SopA was synthesized at a lower level in the dam mutant compared with the wild-type strain (P<0.05) (Fig. 4a and b). This finding suggests a sustained expression of these effectors after invasion. Therefore, it would be important to carefully consider the dichotomous roles of SPI-1 and SPI-2 in the intestinal and/or the systemic paradigm of Salmonella infection (Coburn et al., 2005; Schlumberger & Hardt, 2006).
In summary, we found that in vitro and in vivo expression of SopA is impaired in dam mutants of Salmonella Typhimurium; moreover, no secreted SopA could be detected. Altered expression and/or secretion of proteins in dam strains may contribute to the decreased virulence and heightened immunity observed in vaccinated hosts.
We are very grateful to Ms María Isabel Bernal for her excellent technical assistance and to Dr J. Casadesús for providing the SV4712 strain. This work was supported in part by grants from Agencia Nacional de Promoción a la Ciencia y Tecnología (PICT – 2006-00407) and from Universidad de Buenos Aires (UBACyT M608 and M009), Argentina.