Deciphering why Salmonella Gallinarum is less invasive in vitro than Salmonella Enteritidis
© Rossignol et al.; licensee BioMed Central Ltd. 2014
Received: 28 February 2014
Accepted: 23 July 2014
Published: 30 August 2014
Salmonella Gallinarum and Salmonella Enteritidis are genetically closely related however associated with different pathologies. Several studies have suggested that S. Gallinarum is less invasive in vitro than S. Enteritidis. In this study we confirm that the S. Gallinarum strains tested were much less invasive than the S. Enteritidis strains tested in cells of avian or human origin. In addition, the S. Gallinarum T3SS-1-dependent ability to invade host cells was delayed by two to three hours compared to S. Enteritidis, indicating that T3SS-1-dependent entry is less efficient in S. Gallinarum than S. Enteritidis. This was neither due to a decreased transcription of T3SS-1 related genes when bacteria come into contact with cells, as transcription of hilA, invF and sipA was similar to that observed for S. Enteritidis, nor to a lack of functionality of the S. Gallinarum T3SS-1 apparatus as this apparatus was able to secrete and translocate effector proteins into host cells. In contrast, genome comparison of four S. Gallinarum and two S. Enteritidis strains revealed that all S. Gallinarum genomes displayed the same point mutations in each of the main T3SS-1 effector genes sipA, sopE, sopE2, sopD and sopA.
Salmonella Enteritidis (SE) and Salmonella Gallinarum (SG) are responsible for very different infections in poultry. SE causes transient systemic infection and asymptomatic intestinal carriage, whereas SG shows poor intestinal invasion but is responsible for typhoid fever associated with a high mortality rate . Furthermore, SE is able to infect plants and different species of warm and cold blooded animals, whereas SG is highly restricted to avian species. Nevertheless, these two serotypes are closely related genetically and present 99.7% homology between orthologous genes . To date, despite their genetic similarities, reasons for their different pathological behavior are still poorly understood even though some differences at the genomic and proteomic levels have been described ,.
SG displays several distinct features suspected in part to be responsible for its distinct infection phenotype. The absence of type 1 fimbriae  and flagella could explain a reduced pro-inflammatory response compared to SE, facilitating systemic organ colonization from the gut ,. Moreover, SG has lost many metabolic pathways, such as 1,2-propanediol degradation and ornithine decarboxylation, leading to restriction of usable carbon and energy sources . These limited metabolic capabilities could explain SG's reduced ability to colonize the gut, but not why it exhibits a systemic lifestyle compared to SE.
Two studies have suggested that SG displays a deficiency for invasion of epithelial cells compared to SE. Setta et al. demonstrated that SG strain 287/91 was about 60 times less invasive than the SE strain P125109 in chicken kidney cells . Similarly Barrow and Lovell showed that the SG strain 9 was ten times less invasive than an SE strain in Vero cells . However, reasons for this invasion deficiency compared to SE remain unknown.
Salmonella harbors several virulence factors allowing invasion of non-phagocytic cells and is the sole bacterium described as being able to enter cells by either a Zipper or a Trigger mechanism . Rck, which is absent on the SG genome , is an outer membrane protein mediating the Zipper entry mechanism . PagN is another invasin present in both SG and SE strains . However, both invasins are not expressed in cell culture media . The Trigger entry mechanism involves the well-characterized Type Three Secretion System 1 (T3SS-1) encoded by the Salmonella Pathogenicity Island-1. This secretion apparatus delivers effector proteins directly into the host cytosol after contact with the cell. Some effectors, such as SipA, SipC, SopB, SopE and SopE2, trigger extensive rearrangements of the actin cytoskeleton leading to marked membrane ruffling and bacterial internalization . Some others, for example SopD or SopA, contribute to a variety of post-invasion processes, such as vacuolar development (reviewed in ) and modulation of the inflammatory response ,.
The aim of this study was to investigate why SG is less invasive in vitro than SE and whether this difference is related to the functionality of the T3SS-1. We demonstrated that SG's low invasion ability was related to a delayed T3SS-1 entry mechanism despite the presence of a functional T3SS-1 apparatus and a similar expression of T3SS-1-related genes compared to SE.
Materials and methods
Bacterial strains and plasmids
Salmonella Enteritidis strains 02-4232, En9, ST180 and SG vaccine strain 9S were provided by the WHO Collaborating Center for Salmonella (Pasteur Institute, Paris, France). SE strains 1630-02, 01-7587 and SG strains 2210 and 7285 were obtained from the French Laboratory for Food safety (ANSES, Maisons-Alfort, France). SE strain LA5 was provided by the Veterinary Laboratories Agency (Addlestone, UK) , while SG strains 12B, 287/91 and 9 were provided by INRA Centre Val de Loire, France, the National Collection of Type Cultures (NCTC 13346) and by P. Barrow's lab (Nottingham, UK) respectively. The genome sequence of strains LA5 (CAGR00000000), P125109 (AM933172), 287/91 (AM933173) and 9 (CM001153 to CM001154) are available on line ,,. When unspecified, bacteria were grown in Tryptic Soy Broth (TSB, Difco, Fisher-003, Illkirch, France) medium at 37 °C with shaking. Plasmids used in this study and their characteristics are listed in Additional file 1.
Cell lines and culture conditions
Human intestinal cell line HT-29 (85061109 ECACC, Salisbury, UK), human cervical cell line HeLa (ATCC CRM-CCL-2), chicken hepatoma cell line LMH (ATCC CRL-2117), chicken lung epithelial cells CLEC213  and chicken fibroblast cell line DF-1  were grown in the different recommended cell culture media. Cells were routinely grown in 75°Cm2 plastic tissue culture flasks at 37 °C under 5% CO2. Throughout this study, cells were grown without antimicrobial compounds.
Adhesion-invasion assays were performed as previously described . Briefly, 24-well tissue culture plates were seeded five days before infection in order to obtain confluent monolayers for the day of infection. Bacteria were grown overnight in TSB medium at 37 °C without shaking. Each experiment used two plates: one to determine the total number of adhered and invaded bacteria (adhesion/invasion assay) and the second the invaded bacteria only (invasion assay). Cell monolayers were incubated in medium without antibiotics for 24 h and then infected for different times (0.5 h, 1 h, 1.5 h, 3 h or 4.5 h) at 37 °C with 107°CFU in 300 μL (multiplicity of infection 10). For adhesion/invasion assays, the cell monolayers were gently washed six times with phosphate buffered saline (pH 7.3) and then disrupted with 1 mL cold distilled water (4 °C). Viable bacteria (intra- and extracellular) were counted after plating serial dilutions on TSA (Difco). For the invasion assay, the second plate was washed with Dulbecco's modified Eagle's medium (DMEM, Life Technologies, Fisher-003, Illkirch, France) and incubated in culture medium containing 100 μg of gentamicin per mL. After 1.5 h at 37 °C, cells were washed with phosphate-buffered saline and lysed with 1 mL cold distilled water (4 °C). Viable intracellular bacteria were assessed by serial dilutions plated on TSA. Cell viability was checked in parallel to infection by the trypan blue exclusion test. The results are expressed as the percentage of CFU recovered relative to the number of bacteria deposited per well. Experiments were carried out in duplicate and repeated at least twice for each strain.
Construction of LA5invA and 287/91invA mutants
The Datsenko and Wanner method  was used to construct SE LA5invA and SG 287/91invA mutants with two modifications. Firstly, 35 mL of LB medium (Difco) containing 100 μg.mL-1 carbenicillin were seeded with 350 μL of an overnight culture of strains SE LA5 or SG 287/91 carrying pKD46 and incubated at 30 °C without shaking. One hour before OD600nm reached 0.3, the medium was supplemented with 5 mM arabinose (final concentration). Secondly, bacteria were washed three times with 10% glycerol water supplemented with 1 mM 3-morpholinopropane-1-sulfonic acid buffer and were finally resuspended in 100 μL of GYT medium (10% glycerol, 0.125% yeast extract (Difco) and 0.25% tryptone (Difco)). Kanamycin (50 μg/mL) or chloramphenicol (30 μg/mL) resistant mutants for SE LA5 and SG 287/91 respectively were selected and checked using PCR.
In order to analyze the T3SS-1 secretion profile of our strains, in vitro secretion assays were performed as previously described, after culture of bacteria in 30 mL of LB broth containing 0.3 M NaCl, a medium inducing SPI-1 gene expression . Bacteria were grown until the OD reached 1.8-2.0 and then cultures were harvested by centrifugation for 15 Min at 7000 g at 4 °C. Culture supernatants were filtered through 0.2 μm filters. Secreted proteins were precipitated by adding trichloroacetic acid (10% final) and an incubation at 4 °C for 30 Min. Proteins were recovered by centrifugation at 12 000 g for 20 Min and resuspended in 100 μL of Laemmli buffer. Samples were boiled for 5 Min at 99 °C. β-lactoglobuline was added to each culture supernatant to control for protein precipitation and sample loading. Proteins were separated using SDS-PAGE (10% acrylamide), and then stained with Coomassie brilliant blue G-250 . Gels were visualized with a Fusion FX7 imaging system (Vilber Lourmat).
A β-lactamase (TEM-1) translocation assay of effector proteins was performed as previously described by Fookes et al. . SG 287/91 wild-type and its invA mutant strain carrying TEM-1 fusions were incubated for 3 h with HeLa and CLEC213 cells at MOI 100. Translocation was detected by fluorescence microscopy (Leica) using fluorescent CCF4/AM β-lactamase substrate (LiveBLAzer FRET-B/g Loading Kit, Invitrogen, Fisher-003, Illkirch, France).
Actin foci quantification
HeLa cells (1× μ105) were seeded two days before infection in 24-well plates on 12 mM microscope cover glasses. The day before infection, 1 μg of pEGFP-actin was used to transfect cells using Nanofectin as recommended by the manufacturer (PAA-The Cell Culture Company, PAA Labora, Les Mureaux, France). After 3 h, fresh serum containing medium was added for 16 h and the cells were then antibiotic-starved. After three washes, cells were incubated at MOI 50 with mCherry labeled bacteria, i.e. carrying the pFPVmcherry plasmid  (Additional file 1), which were grown overnight in LB broth medium at 37 °C without shaking. After centrifugation for 5 Min at 50 g at room temperature and 0.5 h at 37 °C, the cells were washed three times and fixed in PBS 4% paraformaldehyde at room temperature. Coverslips were mounted in fluorescence mounting medium and analyzed using an Apotome system (Zeiss). Actin foci associated with each strain were scored visually for 100 cells labelled with actin-EGFP.
Quantification of intracellular bacteria replication
Quantification of bacterial replication in the cellular culture medium
Bacteria were grown overnight in TSB at 37 °C without shaking. Prior to the infection, bacteria were diluted in DMEM-F12 medium (Invitrogen). Bacteria (1.2× μ107 per well) were distributed in 24-well non-seeded plates and were incubated for 1.5 h, 3 h and 4.5 h at 37 °C under 5% CO2. For each time point, bacterial multiplication in the cellular culture medium was determined by plating serial dilutions on TSA.
Measurement of invF, sipA and hilA mRNA levels
Expression of invF, sipA and hilA was measured using semi-quantitative Reverse Transcription PCR. CLEC213 cells were seeded in 100 mM diameter dishes to obtain confluent monolayers on the day of infection. Cells were infected for 0.5 h with strain SE LA5 at MOI 100 or strain 287/91 at MOI 1000 in order to obtain the same amount of adhered bacteria per dish for the two strains. CLEC213 were lysed using a 0.1% SDS, 1% acidic phenol, 19% ethanol mixture  and a bacterial pellet was obtained through centrifugation. Bacterial RNA was purified using the RNeasy Mini Kit (Qiagen, Courtaboeuf, France) and genomic DNA contamination was removed using the TURBO DNA-free kit, in accordance with the manufacturer's instructions (Ambion, Spitfire Close-Ermine Business, UK). Bacterial RNA was reverse transcribed to cDNA in the presence of random hexamers using the SuperScript III First-Strand Synthesis SuperMix (Invitrogen) and PCR amplifications were performed using the appropriate primers (Additional file 1). Reactions without reverse transcription were performed to check the lack of DNA contamination in the RNA preparation. Expression of the housekeeping gene tufA was also measured as a control for bacterial RNA quantity. PCR samples were loaded on an agarose gel and band intensity was assessed using Quantity One software (BioRad).
Unpaired t test with Welch's correction (because variances are significantly different) was used with the number of CFU as a variable to analyze the differences in the invasion assays. The Gaussian distribution of our values was tested by the d'Agostino and Pearson omnibus normality test. Data were analyzed using Graph Pad Prism software 5.0.
S. Gallinarum is less invasive in avian epithelial cells in vitro than S. Enteritidis
In order to confirm the low invasive phenotype of SG compared to SE, the invasive ability of six strains of each serotype were compared using adhesion/invasion assays performed in avian hepatoma cells (LMH) (Figure 1A). On average SG strains were 1500-fold less invasive than SE strains after 0.5 h of bacteria-host cell contact (1.97 - 0.41 versus 5.17× μ0.38 log10 CFU respectively). An unpaired t test showed that this difference was highly significant (p <-0.0001). These data suggest that for all the strains tested SG was less invasive than SE. Figure 1A shows that on average the SG strains were six times less adhesive than the SE strains (5.07× μ0.19 versus 5.88× μ0.22 log10 CFU respectively (p-<-0.0001)), suggesting that the marked invasion deficiency of SG could not be explained by a deficiency in adhesion alone. The lack of a link between poor adhesion and poor invasion observed for SG was confirmed by centrifuging the plate immediately after infection using the SG strain 2210 as an example (Figure 1A, white bar). Centrifugation enabled the main part of the bacterial population to come rapidly into contact with cells independently of an adhesion defect or lack of motility. Under this condition, the number of SG that adhered was four times higher (white bar) than for non-centrifuged bacteria, however, the number of intracellular bacteria (hatched bar) remained unchanged, confirming that the adhesion defect did not explain the invasion defect.
S. Gallinarum is less invasive in vitro than S. Enteritidis regardless of the cell origin
We previously observed that the SG strains tested were less invasive in avian cells than the SE strains. To determine whether their low invasive phenotype depends on the cell origin, adhesion/invasion assays were performed using other avian (CLEC213 and DF-1) and human (HeLa and HT-29) cell lines (Figure 1B). For practical reasons, further experiments were performed using only one representative strain of each serotype: the SE LA5 strain, responsible for transient systemic infection and intestinal carriage in the chicken and the SG 287/91 strain, responsible for lethal systemic infection in poultry. The results show that the SG 287/91 strain was always significantly less invasive than the SE LA5 strain (30- to 4000-fold, depending on the cell line tested), demonstrating that SG is less invasive in vitro than SE regardless of the cell origin. As previously observed, Figure 1B shows that SG displayed an adhesion defect compared to SE in all the cell lines tested, but the difference was insufficient to explain the invasion defect. Moreover, no correlation was observed between adhesion and invasion rates.
It is interesting to note that the SG 287/91 strain was significantly less invasive for avian than for human cell lines (p < 0.0001) (less than 10 intracellular bacteria compared to 70 for 1× μ107 bacteria deposited respectively). This was not observed with the SE LA5 strain, which displayed similar invasive abilities regardless of the origin of the cell line. Further experiments are required to explain this phenotype.
S. Gallinarum exhibits a functional T3SS-1 apparatus but is unable to induce membrane ruffles
As the T3SS-1 is the main invasion factor known in Salmonella, we investigated further whether a dysfunctional T3SS-1 apparatus could explain the low invasive ability of SG. The analysis of the genomes of the four SG strains 287/91, 2210, 9 and 9S revealed that all T3SS-1 genes as well as all effectors encoding genes known to be secreted by the T3SS-1 are present in SG genomes (data not shown).
To determine whether SG was able to translocate T3SS-1 effectors into the host cell cytosol, a fluorescence-based translocation assay was performed using a fusion protein of the effector SopD with mature TEM-1 β-lactamase and fluorescent β-lactamase substrate CCF4. The emission of blue fluorescence by eukaryotic cells previously loaded with the substrate reveals the cleavage of the CCF4 substrate by β-lactamase and therefore the translocation of SopD into these cells. Uncleaved CCF4 emits green fluorescence. Figure 2B shows that blue cells were observed when avian (CLEC213) and human (HeLa) cells were infected with the SG 287/91 strain carrying pCX340sopD, the plasmid encoding SopD-TEM-1, demonstrating that SG was able to translocate the fusion protein into the cytosol of both avian and human cells. In contrast, cells infected with SG 287/91 carrying the empty vector pCX340 emitted a green fluorescence, indicating that conversion of the fluorescence of CCF4 was only due to the presence of the fusion protein in the host cell. This translocation was T3SS-1-dependent since the 287/91invA strain carrying pCX340sopD was not able to translocate SopD-TEM-1 into the host cell cytosol. These data demonstrate that the SG 287/91 strain was able to translocate at least the SopD effector through its T3SS-1 apparatus into the cytosol of host cells and thus that the T3SS-1 apparatus of SG was functional.
Finally, since the SG 287/91 strain harbors a fully functional T3SS-1 apparatus, and since cell invasion by T3SS-1 leads to strong actin reorganization, we investigated the impact of SG T3SS-1-dependent entry on the host cell actin cytoskeleton. The quantification of actin foci only takes into account the Trigger mechanism, which mediates extensive actin rearrangements in contrast to the Zipper mechanism . Actin-GFP transfected cells were used in order to visualize the actin reorganization occurring during cell invasion mediated by the T3SS-1 using fluorescence microscopy (Figure 2C). Unlike the other cell lines tested, HeLa cells displayed a good transfection rate and were therefore used to perform these experiments. Quantification of actin foci shows that after 0.5 h of infection the SG 287/91 strain was associated with 50-fold less actin foci than the SE LA5 strain (0.7× μ0.6 for 100 cells vs 34.3× μ13.3, respectively), strongly suggesting that the SG invasion defect is mainly related to a T3SS-1 invasion defect.
S. Gallinarum's T3SS-1 invasive ability increases with time
In order to investigate the time course of the invasion process, invasion kinetics was studied using the wild-type strains and their invA mutants (used already in Figure 3A). To determine the contribution of the T3SS-1 over time, the number of intracellular bacteria obtained for each invA mutant was subtracted from the number obtained for the corresponding wild-type strain (Figure 3B, raw data Additional file 2). Experiments were carried out in avian (LMH, CLEC213 and DF-1) and human cell lines (HeLa and HT-29). The results show that after 0.5 h, the SG 287/91 T3SS-1-dependent invasion ability was dramatically impaired compared to that of the SE LA5 strain (30- to 5000-fold lower according to the cell line tested). For between 1 h and 4.5 h of interaction, intracellular number of bacteria increased over time for both strains in all tested cell lines , due to longer bacteria-host cell contact time, but the results for SG remained lower. However, the number of intracellular bacteria increased much more strongly with SG than with SE and particularly in avian cells. Indeed, in avian cells, SG's T3SS-1 invasion abilities increased by 15 to 150 times more than that observed for SE from 0.5 to 3 h. This increase in the number of intracellular SG over time was also observed for human cells but to a lower extent.
Increase in the number of intracellular bacteria between 0.5 h and 3 h is not related to replication of intracellular bacteria
Time of bacteria-cells contact*
Number of bacteria per vacuole
Bacterial replication in the cell culture medium during the infection phase could also explain the increase in the number of intracellular bacteria between 0.5 h and 3 h by artificially increasing the MOI. To test this hypothesis, bacterial replication rates were assessed in the cell culture medium over time. The results show that SG replicated more slowly than SE in the cell culture medium, as previously described in bacterial culture medium. Indeed, the SG 287/91 bacteria number increased only 25-fold between t0 and 3 h, while the SE LA5 bacteria number increased 150 times in cell culture medium. This result shows that the increase in intracellular SG 287/91 was not related to this serotype having a higher replication rate than SE in the cell culture medium during the infection period.
Overall, these data demonstrate that the greater increase in the number of intracellular SG than of SE was not related to a higher replication rate but to an increase in T3SS-1-dependent invasive ability over time.
S. Gallinarum SPI-1 gene expression is similar to S. Enteritidis SPI-1 gene expression and S. Gallinarum effectors contain point mutations
S . Gallinarum T3SS-1 effectors display potential causal mutations
Position of mutation
length: 783 aa
length: 446 aa
GEF catalytic domain
GEF catalytic domain
In order to determine whether one of these effectors was inactive in SG, an attempt to complement the SG 287/91 strain with one of the corresponding genes of the invasive strain SE LA5 was carried out. However, single complementation did not increase the invasive phenotype of the SG strains, suggesting that either the mutations have no deleterious impact on the invasive phenotype or, more probably, that more than one mutated effector is responsible for the low invasive ability of SG (Additional file 3). Further studies should be performed to make the point with different polarized and non-polarized cell lines because it has been described that the T3SS-1 dependent entry process is dependent on used cell lines and polarization ,.
Salmonella is a facultative intracellular bacterium and several reports have shown that cell invasion is critical for disease to develop. Previous work performed with one epithelial cell line and one strain for each, SG and SE reported that the SG strain was less invasive than the SE strain after 0.5 h of bacteria-host cell contact ,. Our work has extended this observation to six strains of each serotype and to five cell lines, strengthening the notion that SG is less invasive than SE and demonstrating that its low invasive ability is at the serotype scale. These findings were observed with cells originating from avian and human sources, suggesting that host specificity and lethal systemic spread of SG in the chicken are not related to specific invasiveness for avian non-phagocytic cells. Large metabolic networks which are intact in genomes of gastrointestinal Salmonella like SE but degraded in genomes of extra-intestinal serotypes and which enable growth of Salmonella in the inflamed gastrointestinal tract, could be involved in host specificity and disease pattern . Furthermore, our in vitro results showing that SG poorly invades non-phagocytic cells suggest that enterocytes are probably not the main entry route for SG in vivo from the gut to the systemic apparatus, although it is not possible to exclude the hypothesis that a low invasion rate of these cells could be sufficient to initiate systemic infection.
In line with previous studies, we also observed that all tested SG strains were less adhesive than the SE strains ,. However, this phenotype was not sufficient on its own to explain the difference in invasion between these serotypes. Indeed, SG's adhesion rate was, on average, only six-fold lower than SE, whereas its ability to enter was up to 4000-fold lower depending on the cell line tested. Furthermore, no correlation was observed between the differences in adhesion rates and in entry levels for the five cell lines and the twelve Salmonella strains tested. The low invasive phenotype of SG was also observed when bacteria were in close contact with cells (after centrifugation of plates), which strongly suggests that poor invasion rates were related to a deficient entry ability.
Salmonella is able to enter cells through at least three invasion routes mediated by the T3SS-1, which induces extensive actin cytoskeletal rearrangements, and two invasins (PagN and Rck) which induce limited actin cytoskeletal rearrangement . Rck is absent in SG , hence it could be thought to be responsible for its poor invasion behavior. However, Rck and PagN are not expressed in Salmonella under our culture conditions ,, suggesting that the invasion deficiency observed for SG cannot be explained by differences in these two invasins. In contrast, fluorescent microscopy of actin transfected cells shows that SG induced a very low number of actin foci compared to SE suggesting that the invasion defect could be related to a defect in the Trigger mechanism.
In Salmonella, the Trigger entry mechanism is mediated by the T3SS-1. Our study demonstrates that the low invasive phenotype cannot be related to an inability of SG to use its T3SS-1. Indeed, SG was able to secrete T3SS-1 effectors in vitro and to translocate them within host cells in a T3SS-1 dependent manner. However, a SG invA mutant unable to use its T3SS-1 was defective in entering cells compared to the wild-type strain, showing that the T3SS-1 is, at least in part, functional in SG. Finally, the kinetics of cell invasion demonstrate that SG was able to invade cells efficiently using its T3SS-1, but only after a very long contact time with host cells, whereas the invasion of SE was more efficient.
As SG harbors a functional T3SS-1 apparatus, the T3SS-1 defect could be explained by a lower SPI-1 gene expression compared to that of SE or by dysfunctional T3SS-1 effectors. The present study shows that T3SS-1 regulatory genes hilA and invF and the effector gene sipA were expressed at similar levels to those in SE after bacteria-host cell contact, suggesting that all T3SS-1 effectors are well expressed by SG. Moreover, whole genome comparison revealed that the four SG genomes analyzed (287/91, 9, 9S, and 2210) displayed all the known genes encoding T3SS-1 effectors. However, these genomes displayed the same point mutations in the main T3SS-1 effector genes sopE, sopE2, sipA, sopD and sopA unlike SE strains P125109 and LA5. For sopE, sopE2 and sipA, these mutations change the amino-acid sequence. Furthermore, sopA has an early STOP codon in the mutated version leading to a shorter SopA protein. This mutation has been previously described by Thomson et al. . All these changes could affect the proper folding of the proteins or could modify their function as they are in functional domains (Table 2). SopE and SopE2 are paralogues and act as eukaryotic guanine exchange nucleotide factors (GEF). They trigger the exchange of GDP for GTP leading to activation of Rho GTPases Rac1 and Cdc42, responsible for actin polymerization and formation of membrane ruffles ,. A mutation in the GEF catalytic domain of these proteins (Table 2) could thus modify the invasive ability of SG. SipA binds actin directly and induces its polymerization by decreasing the critical concentration for actin assembly ,. Two of the four mutations in SipA were present in the chaperone binding domain suggesting that SipA of SG could be weakly or slowly translocated within the host cells. Thus the mutations in the genes encoding these three proteins could explain why SG induced fewer actin foci than SE and also the delayed entry during early stages of infection, as a S. Typhimurium sipA mutant was shown to enter cells more slowly than its wild-type parent especially in polarized cells . Finally, SopD plays a role in membrane fission after internalization of the bacteria , and SopA exhibits an E3 ubiquitin ligase activity in addition to its potential to modulate the intestinal inflammation response . SopD and SopA are also both involved in invasion of polarized epithelial cells . Overall, the present study shows that SG harbors mutations in genes encoding the main effectors which could contribute to its defective invasion.
In order to test the impact of these mutations on cell invasion, an attempt was made to complement the SG 287/91 strain with either sipA, sopE or sopA from the SE LA5 strain. SopE2 was not introduced due to its redundancy with SopE . However, the invasive ability of our single complemented strains was not restored after 1 h of bacteria/LMH cell contact (Additional file 3). Complementation was also not observed after infection of polarized Caco-2 cells with sipA complemented SG strain (data not shown). This result suggests that SG's T3SS-1 defective invasion could be related to a combination of mutations in several T3SS-1 effector genes. Another hypothesis, which is compatible with the first one, is that the lower adhesion observed with SG compared to SE hampered an intimate contact between SG and its host cells, which is necessary for the injection of T3SS-1 effectors. This hypothesis is supported by the demonstration that establishment of close contact is a prerequisite for subsequent T3SS-1-dependent invasion of some cell lines . Nevertheless, we have demonstrated by β-lactamase (TEM-1) translocation assay that SG, like SE is able to inject T3SS-1 effectors. The fact that this test is not quantitative does not allow us to exclude this second hypothesis. Further experiments are required to investigate the impact of adhesion on the intimate contact of SG and the role of these mutations on the biological function of these effectors and on the pathogenicity of SG.
A current hypothesis is that ubiquitous serotypes, such as SE, elicit a strong pro-inflammatory response in the gut limiting the development of infection ,. On the contrary to ubiquitous serotypes, host specific serotypes such as SG are not thought to cause this strong pro-inflammatory response, thus allowing systemic dissemination and severe disease ,. Several bacterial antigens including T3SS-1 effectors have been shown to elicit the pro-inflammatory response. Wood et al. have shown that S. Dublin SopA induces fluid secretion and polymorphonuclear cell influx in a bovine ligated loop model . An SE sipA mutant has also been shown to impair its capacity to promote the expression of two pro-inflammatory chemokines CXCLi1 and CXCLi2, . Further work is necessary to determine whether the T3SS-1 effector mutations identified in our study can help SG to escape the pro-inflammatory response.
It is interesting to note that, like SG, both S. Typhi and S. Abortusovis have a low entry rate into host cells and are host specific and induce systemic infection. To determine the relationship between the respective invasion rates and induction of systemic infection, it would be interesting to develop a SG chimera, namely a SG strain having the SE effectors, to analyze the impact on cell entry capability, on the immune response induced and thus on pathogenicity.
We are grateful to Gad Frankel (Imperial College London, London, UK) for providing the plasmid pCX340sopD (pICC612) and Olivia Steele-Mortimer (NIH, Hamilton, USA) who gave us the pFPVmcherry plasmid (Addgen plasmid 20956). We also thank Louise Baugé who adapted Datsenko and Wanner's method for mutant construction in S. Gallinarum. We would also like to thank Sue Edrich for correcting the English. This work and AR's doctoral fellowship were supported by the Région Centre and INRA. J.F. has been funded by a fellowship from the Pasteur Foundation, and J.E. is supported by a grant from the European Union (Grant "Rupteffects", ERC2010-StG-261166).
- Chappell L, Kaiser P, Barrow P, Jones MA, Johnston C, Wigley P: The immunobiology of avian systemic salmonellosis. Vet Immunol Immunopathol. 2009, 128: 53-59. 10.1016/j.vetimm.2008.10.295.View ArticlePubMedGoogle Scholar
- Thomson NR, Clayton DJ, Windhorst D, Vernikos G, Davidson S, Churcher C, Quail MA, Stevens M, Jones MA, Watson M, Barron A, Layton A, Pickard D, Kingsley RA, Bignell A, Clark L, Harris B, Ormond D, Abdellah Z, Brooks K, Cherevach I, Chillingworth T, Woodward J, Norberczak H, Lord A, Arrowsmith C, Jagels K, Moule S, Mungall K, Sanders M, Whitehead S, Chabalgoity JA, Maskell D, Humphrey T, Roberts M, Barrow PA, Dougan G, Parkhill J: Comparative genome analysis ofSalmonellaEnteritidis PT4 andSalmonellaGallinarum 287/91 provides insights into evolutionary and host adaptation pathways.Genome Res 2008, 18:1624-1637.,
- Eswarappa SM, Janice J, Balasundaram SV, Dixit NM, Chakravortty D: Host-specificity ofSalmonella entericaserovar Gallinarum: insights from comparative genomics.Infect Genet Evol 2009, 9:468-473.,
- Osman KM, Ali MM, Radwan MI, Kim HK, Han J: Comparative proteomic analysis onSalmonellaGallinarum andSalmonellaEnteritidis exploring proteins that may incorporate host adaptation in poultry.J Proteomics 2009, 72:815-821.,
- Wilson RL, Elthon J, Clegg S, Jones BD: Salmonella entericaserovarsgallinarumandpullorumexpressingSalmonella entericaserovartyphimuriumtype 1 fimbriae exhibit increased invasiveness for mammalian cells.Infect Immun 2000, 68:4782-4785.,
- Kaiser P, Rothwell L, Galyov EE, Barrow PA, Burnside J, Wigley P: Differential cytokine expression in avian cells in response to invasion bySalmonella typhimurium,Salmonella enteritidisandSalmonella gallinarum.Microbiology 2000, 146:3217-3226.,
- Iqbal M, Philbin VJ, Withanage GS, Wigley P, Beal RK, Goodchild MJ, Barrow PA, McConnell I, Maskell DJ, Young J, Bumstead N, Boyd Y, Smith AL: Identification and functional characterization of chicken toll-like receptor 5 reveals a fundamental role in the biology of infection withSalmonella entericaserovar Typhimurium.Infect Immun 2005, 73:2344-2350.,
- Setta A, Barrow PA, Kaiser P, Jones MA: Immune dynamics following infection of avian macrophages and epithelial cells with typhoidal and non-typhoidalSalmonella entericaserovars; bacterial invasion and persistence, nitric oxide and oxygen production, differential host gene expression, NF-kappaB signalling and cell cytotoxicity.Vet Immunol Immunopathol 2012, 146:212-224.,
- Barrow PA, Lovell MA: Invasion of Vero cells bySalmonellaspecies.J Med Microbiol 1989, 28:59-67.,
- Velge P, Wiedemann A, Rosselin M, Abed N, Boumart Z, Chausse AM, Grepinet O, Namdari F, Roche SM, Rossignol A, Virlogeux-Payant I: Multiplicity ofSalmonellaentry mechanisms, a new paradigm forSalmonellapathogenesis.Microbiologyopen 2012, 1:243-258.,
- Haneda T, Okada N, Nakazawa N, Kawakami T, Danbara H: Complete DNA sequence and comparative analysis of the 50-kilobase virulence plasmid ofSalmonella entericaserovar Choleraesuis.Infect Immun 2001, 69:2612-2620.,
- Rosselin M, Virlogeux-Payant I, Roy C, Bottreau E, Sizaret PY, Mijouin L, Germon P, Caron E, Velge P, Wiedemann A: Rck ofSalmonella enterica, subspeciesentericaserovar Enteritidis, mediates Zipper-like internalization.Cell Res 2010, 20:647-664.,
- Lambert MA, Smith SG: The PagN protein ofSalmonella entericaserovar Typhimurium is an adhesin and invasin.BMC Microbiol 2008, 8:142-145.,
- Patel JC, Galan JE: Manipulation of the host actin cytoskeleton by Salmonella-all in the name of entry. Curr Opin Microbiol. 2005, 8: 10-15. 10.1016/j.mib.2004.09.001.View ArticlePubMedGoogle Scholar
- Steele-Mortimer O: TheSalmonella-containing vacuole: moving with the times.Curr Opin Microbiol 2008, 11:38-45.,
- Raffatellu M, Wilson RP, Chessa D, Andrews-Polymenis H, Tran QT, Lawhon S, Khare S, Adams LG, Baumler AJ: SipA, SopA, SopB, SopD, and SopE2 contribute toSalmonella entericaserotype Typhimurium invasion of epithelial cells.Infect Immun 2005, 73:146-154.,
- Galyov EE, Wood MW, Rosqvist R, Mullan PB, Watson PR, Hedges S, Wallis TS: A secreted effector protein ofSalmonella dublinis translocated into eukaryotic cells and mediates inflammation and fluid secretion in infected ileal mucosa.Mol Microbiol 1997, 25:903-912.,
- Dibb-Fuller MP, Allen-Vercoe E, Thorns CJ, Woodward MJ: Fimbriae- and flagella-mediated association with and invasion of cultured epithelial cells bySalmonella enteritidis.Microbiology 1999, 145:1023-1031.,
- Grepinet O, Rossignol A, Loux V, Chiapello H, Gendrault A, Gibrat JF, Velge P, Virlogeux-Payant I: Genome sequence of the invasiveSalmonella entericasubsp.entericaserotype enteritidis strain LA5.J Bacteriol 2012, 194:2387-2388.,
- Richardson EJ, Limaye B, Inamdar H, Datta A, Manjari KS, Pullinger GD, Thomson NR, Joshi RR, Watson M, Stevens MP: Genome sequences ofSalmonella entericaserovar Typhimurium, Choleraesuis, Dublin, and Gallinarum strains of well- defined virulence in food-producing animals.J Bacteriol 2011, 193:3162-3163.,
- Esnault E, Bonsergent C, Larcher T, Bed hom B, Vautherot JF, Delaleu B, Guigand L, Soubieux D, Marc D, Quéré P: A novel chicken lung epithelial cell line: characterization and response to low pathogenicity avian influenza virus. Virus Res. 2011, 159: 32-42. 10.1016/j.virusres.2011.04.022.View ArticlePubMedGoogle Scholar
- Himly M, Foster DN, Bottoli I, Iacovoni JS, Vogt PK: The DF-1 chicken fibroblast cell line: Transformation induced by diverse oncogenes and cell death resulting from infection by avian leukosis viruses. Virology. 1998, 248: 295-304. 10.1006/viro.1998.9290.View ArticlePubMedGoogle Scholar
- Roche SM, Gracieux P, Milohanic E, Albert I, Virlogeux-Payant I, Temoin S, Grepinet O, Kerouanton A, Jacquet C, Cossart P, Velge P: Investigation of specific substitutions in virulence genes characterizing phenotypic groups of low-virulence field strains ofListeria monocytogenes.Appl Environ Microbiol 2005, 71:6039-6048.,
- Datsenko KA, Wanner BL: One-step inactivation of chromosomal genes inEscherichia coliK-12 using PCR products.Proc Natl Acad Sci U S A 2000, 97:6640-6645.,
- Arricau N, Hermant D, Waxin H, Ecobichon C, Duffey PS, Popoff MY: The RcsB-RcsC regulatory system ofSalmonellaTyphi differentially modulates the expression of invasion proteins, flagellin and Vi antigen in response to osmolarity.Mol Microbiol 1998, 29:835-850.,
- Neuhoff V, Arold N, Taube D, Ehrhardt W: Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis. 1988, 9: 255-262. 10.1002/elps.1150090603.View ArticlePubMedGoogle Scholar
- Fookes M, Schroeder GN, Langridge GC, Blondel CJ, Mammina C, Connor TR, Seth-Smith H, Vernikos GS, Robinson KS, Sanders M, Petty NK, Kingsley RA, Baumler AJ, Nuccio SP, Contreras I, Santiviago CA, Maskell D, Barrow PA, Humphrey T, Nastasi A, Roberts M, Frankel G, Parkhill J, Dougan G, Thomson NR: Salmonella bongoriprovides insights into the evolution of theSalmonellae.PLoS Pathog 2011, 7:e1002191.,
- Drecktrah D, Levine-Wilkinson S, Dam T, Winfree S, Knodler LA, Schroer TA, Steele-Mortimer O: Dynamic behavior ofSalmonella-induced membrane tubules in epithelial cells.Traffic 2008, 9:2117-2129.,
- Eriksson S, Lucchini S, Thompson A, Rhen M, Hinton JC: Unravelling the biology of macrophage infection by gene expression profiling of intracellularSalmonella enterica.Mol Microbiol 2003, 47:103-118.,
- Fardini Y, Chettab K, Grepinet O, Rochereau S, Trotereau J, Harvey P, Amy M, Bottreau E, Bumstead N, Barrow PA, Virlogeux-Payant I: The YfgL lipoprotein is essential for type III secretion system expression and virulence ofSalmonella entericaSerovar Enteritidis.Infect Immun 2007, 75:358-370.,
- Rosselin M, Abed N, Virlogeux-Payant I, Bottreau E, Sizaret PY, Velge P, Wiedemann A: Heterogeneity of type III secretion system (T3SS)-1-independent entry mechanisms used bySalmonellaEnteritidis to invade different cell types.Microbiology 2011, 157:839-847.,
- Boyen F, Pasmans F, Van Immerseel F, Morgan E, Adriaensen C, Hernalsteens JP, Decostere A, Ducatelle R, Haesebrouck F: SalmonellaTyphimurium SPI-1 genes promote intestinal but not tonsillar colonization in pigs.Microbes Infect 2006, 8:2899-2907.,
- Nuccio SP, Baumler AJ: Comparative analysis ofSalmonellagenomes identifies a metabolic network for escalating growth in the inflamed gut.MBio 2014, 5:e00929-00914.,
- Guo A, Cao S, Tu L, Chen P, Zhang C, Jia A, Yang W, Liu Z, Chen H, Schifferli DM: FimH alleles direct preferential binding ofSalmonellato distinct mammalian cells or to avian cells.Microbiology 2009, 155:1623-1633.,
- Bulgin R, Raymond B, Garnett JA, Frankel G, Crepin VF, Berger CN, Arbeloa A: Bacterial guanine nucleotide exchange factors SopE-like and WxxxE effectors. Infect Immun. 2010, 78: 1417-1425. 10.1128/IAI.01250-09.PubMed CentralView ArticlePubMedGoogle Scholar
- Hardt WD, Chen LM, Schuebel KE, Bustelo XR, Galan JE: S. typhimurium encodes an activator of Rho GTPases that induces membrane ruffling and nuclear responses in host cells. Cell. 1998, 93: 815-826. 10.1016/S0092-8674(00)81442-7.View ArticlePubMedGoogle Scholar
- Agbor TA, McCormick BA: SalmonellaEffectors: Important players modulating host cell function during infection.Cell Microbiol 2011, 13:1858-1869.,
- McGhie EJ, Hayward RD, Koronakis V: Control of actin turnover by aSalmonellainvasion protein.Mol Cell 2004, 13:497-510.,
- Perrett CA, Jepson MA: Regulation ofSalmonella-induced membrane ruffling by SipA differs in strains lacking other effectors.Cell Microbiol 2009, 11:475-487.,
- Bakowski MA, Cirulis JT, Brown NF, Finlay BB, Brumell JH: SopD acts cooperatively with SopB duringSalmonella entericaserovar Typhimurium invasion.Cell Microbiol 2007, 9:2839-2855.,
- Zhang Y, Higashide WM, McCormick BA, Chen J, Zhou D: The inflammation-associatedSalmonellaSopA is a HECT-like E3 ubiquitin ligase.Mol Microbiol 2006, 62:786-793.,
- Orchard RC, Alto NM: Mimicking GEFs: a common theme for bacterial pathogens. Cell Microbiol. 2012, 14: 10-18. 10.1111/j.1462-5822.2011.01703.x.PubMed CentralView ArticlePubMedGoogle Scholar
- Gerlach RG, Claudio N, Rohde M, Jackel D, Wagner C, Hensel M: Cooperation ofSalmonellapathogenicity islands 1 and 4 is required to breach epithelial barriers.Cell Microbiol 2008, 10:2364-2376.,
- Wigley P: Immunity to bacterial infection in the chicken. Dev Comp Immunol. 2013, 41: 413-417. 10.1016/j.dci.2013.04.008.View ArticlePubMedGoogle Scholar
- Henderson SC, Bounous DI, Lee MD: Early events in the pathogenesis of avian salmonellosis. Infect Immun. 1999, 67: 3580-3586.PubMed CentralPubMedGoogle Scholar
- Setta AM, Barrow PA, Kaiser P, Jones MA: Early immune dynamics following infection withSalmonella entericaserovars Enteritidis, Infantis, Pullorum and Gallinarum: cytokine and chemokine gene expression profile and cellular changes of chicken cecal tonsils.Comp Immunol Microbiol Infect Dis 2012, 35:397-410.,
- Wood MW, Jones MA, Watson PR, Siber AM, McCormick BA, Hedges S, Rosqvist R, Wallis TS, Galyov EE: The secreted effector protein ofSalmonella dublin, SopA, is translocated into eukaryotic cells and influences the induction of enteritis.Cell Microbiol 2000, 2:293-303.,
- Li S, Zhang MZ, Yan L, Lillehoj H, Pace LW, Zhang S: Induction of CXC chemokine messenger-RNA expression in chicken oviduct epithelial cells bySalmonella entericaserovar Enteritidis via the type three secretion system-1.Avian Dis 2009, 53:396-404.,
- Ehsani S, Santos JC, Rodrigues CD, Henriques R, Audry L, Zimmer C, Sansonetti P, Tran Van Nhieu G, Enninga J: Hierarchies of host factor dynamics at the entry site ofShigella flexneriduring host cell invasion.Infect Immun 2012, 80:2548-2557.,
- Chang AC, Cohen SN: Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J Bacteriol. 1978, 134: 1141-1156.PubMed CentralPubMedGoogle Scholar
- Knodler LA, Bestor A, Ma C, Hansen-Wester I, Hensel M, Vallance BA, Steele-Mortimer O: Cloning vectors and fluorescent proteins can significantly inhibit Salmonella enterica virulence in both epithelial cells and macrophages: implications for bacterial pathogenesis studies. Infect Immun. 2005, 73: 7027-7031. 10.1128/IAI.73.10.7027-7031.2005.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.