Logo of iaiPermissionsJournals.ASM.orgJournalIAI ArticleJournal InfoAuthorsReviewers
Infect Immun. 2007 Feb; 75(2): 792–800.
Published online 2006 Dec 4. doi:  10.1128/IAI.00679-06
PMCID: PMC1828496

Involvement of a Salmonella Genomic Island 1 Gene in the Rumen Protozoan-Mediated Enhancement of Invasion for Multiple-Antibiotic-Resistant Salmonella enterica Serovar Typhimurium


Multiple-antibiotic-resistant Salmonella enterica serotype Typhimurium is a food-borne pathogen that may be more virulent than related strains lacking the multiresistance phenotype. Salmonella enterica serotype Typhimurium phage type DT104 is the most prevalent of these multiresistant/hypervirulent strains. Multiresistance in DT104 is conferred by an integron structure, designated Salmonella genomic island 1 (SGI1), while we recently demonstrated DT104 hyperinvasion mediated by rumen protozoa (RPz) that are normal flora of cattle. Hyperinvasion was also observed in other Salmonella strains, i.e., other S. enterica serovar Typhimurium phage types and other S. enterica serovars, like S. enterica serovar Infantis, possessing SGI1, while DT104 strains lacking SGI1 were not hyperinvasive. Herein we attempted to identify SGI1 genes involved in the RPz-mediated hyperinvasion of Salmonella strains bearing SGI1. Transposon mutagenesis, coupled with a novel reporter system, revealed the involvement of an SGI1 gene previously designated SO13. Disruption of SO13 expression led to an abrogation of hyperinvasion as assessed by tissue culture invasion assays and by bovine challenge experiments. However, hyperinvasion was not observed in non-SGI1-bearing strains of Salmonella engineered to express SO13. That is, SO13 and another SGI1 gene(s) may coordinately upregulate invasion in DT104 exposed to RPz.

Salmonella enterica is a major cause of food-borne illnesses throughout the world (25). Unfortunately, numerous Salmonella strains have become resistant to multiple antibiotics. This is especially true for S. enterica serotype Typhimurium phage type DT104, which is often resistant to five or more antibiotics as the result of the acquisition of an integron structure, designated Salmonella genomic island 1 (SGI1), that contains genes encoding resistance to five different antibiotics (3).

DT104 may be more virulent than other Salmonella strains, and this putative phenomenon may be related to the presence of SGI1 (23). Humans are two to three times more likely to be hospitalized upon DT104 infection than upon infection with other strains (27), and calves infected with DT104 are 13 times more likely to die than calves infected with antibiotic-sensitive S. enterica serovar Typhimurium (13). This finding also extended to other Salmonella serotypes, e.g., Agona and Infantis, that possess SGI1 (23).

Rumen protozoa (RPz), microbiota that natively inhabit the major forestomach of ruminants such as cattle, were identified as mediators of DT104 hyperinvasion. In this model, RPz engulf DT104 and then hyperactivate the enteroinvasive phenotype. The RPz/DT104 bacteria are then moved to the abomasum, the “true” stomach of ruminants, where the RPz are digested and DT104 is released. DT104 then moves to the small intestine where invasion, or in this case hyperinvasion, ensues. Hyperinvasion leads to a faster onset of clinical signs, a greater recovery rate of the pathogen, and a poorer prognosis. This phenomenon was not observed in the absence of RPz or SGI1 (23). Thus, RPz and SGI1 appear to cocontribute to the enhancement of invasion.

The study presented herein is an attempt to identify the molecular basis for the RPz-mediated activation of invasion-related genes in DT104. The first phase involved the preliminary identification of DT104 virulence genes involved in hyperinvasion. The second phase entailed the use of a novel reporter system and transposon mutagenesis in order to identify SGI1 genes involved in the phenomenon. The final phase was devoted to confirming the role of the SGI1 gene using in vitro invasion assays and in vivo bovine experiments.


Bacterial strains and preparation.

Bacterial strains are summarized in Table Table1,1, with strain 795 (5) serving as the model strain for DT104. Bacteria were stored in cryopreservation tubes containing 50% glycerol-50% culture medium at −70°C and grown in Lennox L broth (GIBCO-BRL) with antibiotics such as ampicillin (Sigma Chemicals, 32 μg/ml), chloramphenicol (Sigma Chemicals, 32 μg/ml), kanamycin (Sigma Chemicals, 64 μg/ml), or zeocin (Invitrogen, 25 μg/ml).

Summary of strains and plasmids used in this studyd

Our previous studies revealed that zeocin proved to be the antibiotic that markedly reduced background bacteria from the RPz milieu (23). For Salmonella strains that were zeocin sensitive, i.e., strains lacking TnZeo, zeocin resistance was conferred via transformation with pECFP (14). This plasmid has a pCRII-Blunt (Invitrogen) backbone with genes encoding enhanced cyan fluorescent protein, kanamycin resistance, and zeocin resistance.

Isolation of RPz and incubations with Salmonella.

RPz were isolated under CO2 as described recently (23). Briefly, rumen fluid was removed from an adult bovine via a rumen fistula. RPz were isolated using Coleman's buffer D (8), and pelleted protozoa were resuspended in 30 ml of the same buffer. One milliliter of this mixture was used for RPz enumeration, and 3 ml (approximately 105 RPz) was used in each experiment. RPz profiles varied from sample to sample.

Approximately 109 bacteria were added to approximately 105 RPz. The Salmonella-RPz mixture was then gently rolled for 16 h at 37°C in a sealed 5-ml glass tube. Extracellular Salmonella bacteria were then killed using 300 μg/ml florfenicol (Schering-Plough). For in vivo studies, the Salmonella-RPz mixture was used as the inoculum. For in vitro studies, RPz were lysed for 60 s at 4,800 rpm using 2.5 mM glass beads and a Mini-Beadbeater (Biospec Products). The lysate was centrifuged at 15,000 rpm for 2 min and then resuspended in 350 μl Lennox L broth. Lysates were then used in reverse transcription-PCR (RT-PCR) or invasion assays.

Salmonella incubations with Acanthamoeba castellanii.

Acanthamoeba castellanii strain AC30 (ATCC 50374) was grown axenically at 25°C in ATCC medium 712. Protists were maintained in 25-cm tissue culture flasks, and medium was replenished every 10 to 14 days following 1:2,500 dilution of confluent cells.

For incubations with Salmonella, 20% of confluent cells were incubated with 109 Salmonella bacteria in a sealed glass tube. The coincubation mixture was gently rolled for 8 to 16 h, after which amoebae were lysed and prepared as described herein for RPz.

RT-PCR assays.

RT-PCR was conducted in order to assess the expression of hilA and SO13. RNA was isolated from bacteria using the RNeasy Mini Kit with enzymatic lysis for the initial cell wall disruption (QIAGEN). RT-PCR was carried out using the SuperScript One-Step RT-PCR with Platinum Taq kit (Invitrogen) using the hilA or SO13 primers described in Table Table2.2. The hilA amplicon is 599 bp while the SO13 amplicon is 858 bp (oligonucleotide sequences are provided in Table Table2).2). PCR reagents have been described recently (4) with visualization entailing 1.5% agarose gel electrophoresis.

DNA sequences of oligonucleotides used in this study

Salmonella invasion assays using HEp-2 cells.

Invasion assays, using Salmonella grown aerobically or anaerobically or recovered from protozoa (RPz or Acanthamoeba castellanii), were performed using HEp-2 cells as described previously (23) with a multiplicity of infection equal to approximately 40 and nine replicates per strain. For protozoan studies, 300 μl of protozoal lysates was used for invasion assays (i.e., 100 μl/well) while 25 μl was used for Salmonella enumeration. Percent invasion was calculated by dividing CFU recovered by CFU added.

Creation of the reporter plasmid pSC104.

The reporter plasmid pSC104 was created using pCR2.1 (Invitrogen) as the backbone. The first phase was to PCR amplify floR, an SGI1 gene encoding a chloramphenicol/florfenicol efflux protein (1), from DT104 and then to ligate the gene into pCR2.1 in the antisense orientation. This intermediate plasmid, designated pSC103, yields numerous floR antisense transcripts that anneal to and thus prevent the translation of the floR sense transcript. The second phase was to PCR amplify the hilA promoter with overhanging HindIII and BamHI sites, digest the amplicon and pSC103 with HindIII and BamHI (HindIII and BamHI sites are upstream from the TA cloning site in pCR2.1), and then ligate the hilA promoter into pSC103 using T4 DNA ligase. The third phase entailed deletion of the lac promoter from pCR2.1. This involved mutagenesis, using the QuikChange mutagenesis kit (Stratagene), whereby a new HindIII site was introduced into the lac promoter. The resulting plasmid was then digested with HindIII and then self ligated, using T4 DNA ligase, under dilute conditions, thus generating pSC104.

Chloramphenicol resistance assays and transposon mutagenesis.

The MIC for chloramphenicol, as determined by CLSI standards (21), was determined for DT104 and DT104/pSC104 recovered from RPz or exposed to either aerobic or anaerobic growth conditions. Chloramphenicol MICs were also assessed for DT104/pSC103 grown in the presence of IPTG (isopropyl-β-d-thiogalactopyranoside; 100 μM; Sigma), for activation of lac expression, and for DT104/pSC103 transformed with the lac-repressing plasmid pLacIq.

The EZ::TnZeo transposome (4) was electrotransformed into DT104/pSC104 using approximately 1010 bacteria and 100 ng of transposome. Transformants were selected for zeocin resistance (25 μg/ml) and for chloramphenicol resistance (32 μg/ml) following exposure to RPz. For mutants that grew in zeocin and chloramphenicol, genomic DNA was isolated (G NOME DNA kit; Bio 101), then digested with EcoRV, and then self ligated with T4 DNA ligase under dilute conditions. To amplify the DNA fragments containing the TnZeo insertion, inverse PCR (primers provided in Table Table2)2) was performed on the self-ligated fragments (“minicircles”). Amplicons were then agarose gel purified (QIAEX II; QIAGEN), cloned into pCR2.1, and transformed into Top10′ cells (Invitrogen). Purified DNA was prepared from the individual transformants and submitted for DNA sequencing.

Creation of a recombinant plasmid encoding SO13.

The plasmid pSO13 was created by PCR amplifying SO13 from DT104 and then ligating the gene into pCR2.1. Proper insert orientation was confirmed using PCR with a vector-specific primer and a gene-specific primer (Table (Table22).

In vivo infection experiments (48-h duration).

Salmonella strains DT104-795, DT104ΔSO13, and DT104ΔSO13/pSO13 were incubated with RPz as described above. Following the overnight incubation, calves were orally infected with RPz still containing Salmonella (n = 5 calves per strain). Specifically, RPz were resuspended in 350 μl of Lennox L broth of which 25 μl was lysed for bacterial enumeration while 325 μl (approximately 4 × 108 CFU of Salmonella) was used for infection. The 325 μl was placed in a gelatin capsule, and the capsule was orally introduced into 1- to 2-week-old Jersey calves (approximately 50 to 100 lb each) immediately followed by 500 ml of commercial milk replacer. Calves of this age were used since they do not have a functional rumen and thus have no native RPz.

Control calves (n = 3 per strain) were challenged with Salmonella strain DT104-795, DT104ΔSO13, or DT104ΔSO13/pSO13 present in HEp-2 cells. Specifically, approximately 109 bacteria were added to approximately 105 HEp-2 cells in a tissue culture dish. Following an overnight incubation, extracellular Salmonella bacteria were then killed using 300 μg/ml florfenicol (Schering-Plough), and HEp-2 cells were then dislodged via trypsinization. HEp-2 cells were then resuspended in 350 μl of Lennox L broth of which 25 μl was used for bacterial enumeration, following HEp-2 cell lysis, while 325 μl (containing approximately 4 × 108 CFU of Salmonella) was used for infection as described above.

Various clinical parameters were assessed every 8 to 12 h postinfection. After 48 h, calves were euthanized using xylazine (1 mg/lb of body weight; intramuscular; Phoenix Laboratories) and pentobarbital (2.6 mg/lb; intravenous; Fort Dodge Laboratories). Splenic samples, which are a reliable indicator of relative systemic burden (23), were collected following euthanasia. Ileal contents were also collected for enumerations in order to determine the lysis of RPz relative to that of HEp-2 cells. Blood was also collected into EDTA tubes for hematological examination performed at the Clinical Pathology Laboratory at Iowa Sate University's College of Veterinary Medicine.

In vivo infection experiments (time course).

Salmonella strains DT104ΔSO13 and DT104ΔSO13/pSO13 were orally inoculated into calves (n = 5 calves per strain) that were recently transfaunated (oral, 1 day prior to infection) with 50 ml of rumen fluid from the fistulated cow. Blood was drawn, and various clinical parameters, including rectal temperatures and hydration status, were assessed every 12 h postinfection. Individual calves were euthanized if rectal temperatures exceeded 105°F or if percent dehydration was 8% or greater. Blood samples were subjected to quantitative PCR for estimation of systemic infection. RPz were also enumerated, as described previously (23), from euthanized calves in order to ensure the efficacy of transfaunation. Calves contained 104 to 105 RPz/ml of rumen fluid.

Salmonella MPN experiments.

Splenic samples (1.5 to 3 g) taken from calves were subjected to the most-probable-number (MPN) enumeration procedure using a series of selective medium preparations (28). Briefly, cultures are grown to extinction in serial dilutions in order to calculate the concentration of the recovered pathogen. The identity of Salmonella strains was confirmed using antiserum/agglutination-based serogrouping.

Quantitative PCR.

Quantitative PCR was used to quantify copies of the target gene sipB/sipC present in samples collected from infected animals. Primers and the TaqMan probe used have been described previously (24). Five-milliliter aliquots of samples collected from infected animals were centrifuged at 5,000 × g for 10 min to pellet bacterial cells. Genomic DNA from cell pellets was isolated using the DNeasy kit and by following the protocol recommended for isolation of total DNA from gram-negative bacteria (QIAGEN Inc., Valencia, CA). The concentration of genomic DNA was determined by using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Inc., Wilmington, DE). Genomic DNA was diluted 1:10 in Tris-HCl (pH 8.5), and 5 μl of diluted DNA was used. PCR was performed in a total volume of 25 μl containing 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 2.5 mM MgCl2, 0.2 mM deoxynucleoside triphosphates, 30 nM of passive reference dye ROX, 600 nM of each primer, 100 nM of sipB/sipC-specific TaqMan probe, and 1.25 units of AmpliTaq Gold DNA polymerase. A standard curve was generated using six duplicate sets of samples containing known quantities of genomic DNA purified from DT104. Two independent sets of PCR assays were performed on MX3005P (Stratagene, La Jolla, CA) by first heating samples to 95°C for 10 min, followed by 40 cycles of heating at 94°C for 30 s, 55°C for 60 s, and 72°C for 30 s. The data were analyzed by MX3005P real-time detection system and software. A standard curve was generated by the instrument by plotting the cycle threshold for each standard against the copy number of sipB/sipC. The linearity of each standard was validated by the amplification efficiency and the R2 values. The copy number of sipB/sipC present in unknown samples was automatically estimated by the instrument from the standard curve.

Statistical analyses.

Statistical analysis was performed using an analysis of variance with Scheffe's F test for multiple comparisons. Comparisons were made between strains or between RPz-dependent and RPz-independent data. Analyses were performed using StatView (SAS Institute).


Increased hilA expression in DT104 exposed to RPz.

Our previous studies indicated that DT104 is hyperinvasive following exposure to RPz (23). We therefore postulated that hilA, a transcriptional regulator of Salmonella invasion whose upregulation can lead to hyperinvasion (19), may be overexpressed in response to RPz exposure. RT-PCR was thus used to evaluate hilA expression in DT104 recovered from RPz. This expression was also evaluated in relationship to HEp-2 cell invasion.

As depicted in Fig. Fig.1,1, RPz-mediated hyperinvasion was associated with enhanced hilA expression in DT104. That is, RPz-dependent hilA expression and invasion were markedly increased relative to these parameters in DT104 grown anaerobically (i.e., in the absence of RPz). As expected, invasion and hilA expression were repressed (invasion of <0.1%, no hilA amplicon detected) in aerobic growth which negatively impacts Salmonella invasion via oxygen-regulated genes (16). Increased hilA expression was also detected in Salmonella enterica serotype Infantis (5) bearing SGI1 (data not shown).

FIG. 1.
Relationship between RPz exposure, hilA expression, and hyperinvasion for DT104 and EE419 grown under various conditions. (Bottom) Invasion. (Top) RT-PCR for hilA expression. The hilA amplicon is 599 bp. Specific molecular size markers are indicated on ...

As a control, hilA expression and invasion were assessed in Salmonella strain EE419, a hyperinvasive strain that constitutively expresses hilA from a transposon (Tn5B50) in the hilA promoter (19). Hyperinvasion and hilA overexpression were evident under standard invasion-inducing conditions, and these activities were comparable to that observed for DT104 isolated from RPz. RPz had no additional effect on invasion and hilA expression in EE419. Invasion and hilA expression of EE419 were detectable even in aerobic conditions (Fig. (Fig.11).

HilA overexpression-based reporter system and identification of SO13.

As shown in Fig. Fig.1,1, hilA overexpression is associated with RPz-mediated hyperinvasion of DT104. In order to identify genes involved in this phenotype, it thus was feasible to exploit a reporter system that detects hilA “nonoverexpression.” To do so, we created a plasmid (pSC104) containing a fusion between the hilA promoter and an antisense copy of the chloramphenicol resistance gene floR found in SGI1 of DT104. That is, activation of the hilA promoter would lead to the synthesis of antisense floR transcripts that anneal to floR mRNA, thus preventing the translation of sense floR transcripts for DT104/pSC104 transformants. Activation of the hilA promoter would thus lead to a decrease in the chloramphenicol MIC while hyperactivation of the hilA promoter could lead to overt chloramphenicol sensitivity.

As shown in Fig. Fig.2,2, the basal chloramphenicol MIC is 128 μg/ml for native DT104. The chloramphenicol MIC went from 128 μg/ml under aerobic growth (noninvasive, hilA repressed) to 32 μg/ml during anaerobic growth (invasive, hilA expressed) for DT104/pSC104. The MIC was further reduced to 8 μg/ml (chloramphenicol sensitive) when hilA expression was hyperactivated as part of exposure to RPz for DT104/pSC104.

FIG. 2.
Invasion and chloramphenicol MIC measurements for DT104, DT104 transformed with a plasmid expressing an antisense copy of the chloramphenicol resistance-conferring gene floR (DT104/pSC104), and the latter also transformed with EZ::TnZeo (SC423). Antisense ...

Other controls included DT104/pSC103, i.e., DT104 expressing antisense floR under the control of the lac promoter, either grown in the presence of IPTG or transformed with a plasmid encoding the lac repressor LacIq. The former control illustrates maximal synthesis of the antisense floR transcript (MIC = 8 μg/ml) while the latter control establishes minimal antisense activity and thus maximal expression of genomic floR (MIC = 128 μg/ml). Figure Figure22 also provides the invasion characteristics of each strain under the various conditions.

In order to identify genes involved in the RPz-mediated phenotype, DT104/pSC104 was transformed with the EZ::TnZeo transposon (4), and pools (5 × 108 transformants/pool) of transformants were examined for chloramphenicol resistance after en masse exposure to RPz. Subpooling yielded one mutant, designated SC423, that exhibited a fourfold increase in the MIC for chloramphenicol without displaying RPz-mediated hyperinvasion. Inverse PCR-based sequence analysis revealed that the TnZeo transposon sequences had inserted in SO13 (accession number AF261825), an 885-bp SGI1 open reading frame with an unknown function (2), in SC423. Analysis of the 295 deduced amino acids revealed the following: an isoelectric point of 8.36, a putative helix-turn-helix-turn-helix from amino acids 9 to 44, and an alpha helix with a weak amphipathic character (average hydrophobic moment per residue [17] equal to −0.3 on the polar face) at amino acids 69 to 82.

A second mutant was identified but not pursued further because RPz-mediated hyperinvasiveness was intact. Inverse PCR identified murE as the insertion site.

In order to study further the relationship between SO13 and DT104 hyperinvasion, it seemed prudent to eliminate pSC104 from SC423. DT104ΔSO13 was thus created by curing, via growth at 42°C, pSC104 from SC423.

In vitro characterization of SO13 and protozoan-mediated hyperinvasion.

To assess the role that SO13 has in regards to RPz-mediated hyperinvasion by DT104, SO13 expression was evaluated under various conditions including exposure to RPz. Additionally, DT104ΔSO13 was complemented with pSO13 and invasion was evaluated. TH11, a wild-type SGI1-free strain of DT104 (5), served as a control strain. Additionally, SO13 expression and tissue culture cell invasion were evaluated for DT104 recovered from free-living Acanthamoeba castellanii. This latter phase was performed in order to assess the extension of the RPz-mediated effect to other protozoa.

Figure Figure33 demonstrates the relationship between RPz exposure, SO13 expression, and hyperinvasion. Deletion of a secreted invasion protein gene (sipD [18]), a downstream effector of hilA (15), resulted in a noninvasive phenotype in the presence of RPz. Deletion of SO13 eliminated RPz-mediated hyperinvasion. RPz-mediated hyperinvasion was noted for DT104ΔsipD and for DT104ΔSO13 complemented with pSipD and pSO13, respectively. Thus, for DT104, there appears to be a relationship between RPz exposure, SO13 expression, and hyperinvasion. However, the phenotype conferring this relationship was not transferable, via SO13 expression alone, to a DT104 strain lacking SGI1 as seen with TH11/pSO13. Also, the RPz-independent expression of SO13 (via pSO13) did not lead to hyperinvasion for DT104 (Table (Table33).

FIG. 3.
Association between protozoan exposure, SO13 expression, and hyperinvasion for various DT104 strains subjected to various conditions. The designation ΔSGI1 is synonymous with strain TH11. Culture conditions include anaerobic growth (white bars) ...
Summary of the relationship between hyperinvasion and SO13 mRNA for Salmonella cultured aerobically, anaerobically, or within RPz or Acanthamoeba castellanii (amoebae)

As shown on the right side of Fig. Fig.3,3, SO13-independent hyperinvasion was observed for DT104 exposed to Acanthamoeba castellanii. However, this effect was 57% ± 5.2% of that observed for RPz exposure.

As summarized in Table Table3,3, hyperinvasion occurred under three conditions in this study. First, growth in anaerobic conditions can lead to hyperinvasion for strains overexpressing hilA as seen in EE419. The second condition is exposure to RPz for Salmonella containing SGI1 (e.g., S. enterica serovar Typhimurium phage type U302 and S. enterica serotype Infantis), while the third condition involves exposure to Acanthamoeba castellanii for Salmonella containing SGI1. SO13 expression accompanied hyperinvasion for RPz only.

In vivo recovery of wild-type DT104, DT104ΔSO13, and complemented DT104ΔSO13.

In vivo studies were undertaken to potentially extrapolate the role that SO13 has in conferring RPz-mediated hyperinvasion. To do so, DT104, DT104ΔSO13, or DT104ΔSO13/pSO13 was incubated with RPz and then separately inoculated into neonatal calves as described previously (23). As a control, DT104-loaded HEp-2 cells were separately inoculated into these subjects. Neonatal calves were chosen since they are a good model for assessing Salmonella burden and since they are natively free of RPz until about 6 to 8 weeks of age.

By inoculation of the DT104-loaded cells into milk, the inoculum will bypass the rumen (albeit it is nonfunctional in neonatal calves) and be deposited in the abomasum which is the “true” stomach of ruminants. Digestive processes then release the DT104, which can then move to the ileum, the preferred invasion site for Salmonella (7). Since it is possible that RPz and HEp-2 cells may have different sensitivities to abomasal lysis and thus different kinetics of Salmonella release, ileal contents were subjected to the MPN procedure and found to contain similar numbers of Salmonella. Specific ileal enumerations were as indicated: DT104, 3.2 × 106/ml for RPz loaded and 4.3 × 106/ml for HEp-2 loaded; DT104ΔSO13, 1.9 × 106/ml for RPz loaded and 1.1 × 106/ml for HEp-2 loaded; DT104ΔSO13/pSO13, 9.2 × 105/ml for RPz loaded and 9.1 × 105/ml for HEp-2 loaded.

As shown in Fig. Fig.4,4, significantly more (i.e., 20 to 40 times more) DT104 bacteria were recovered from calves that were inoculated with DT104-laden RPz than were recovered from calves infected with DT104-loaded HEp-2 cells. The same can be said for calves challenged with the complemented mutant. Statistical analyses revealed P values of <0.0001 despite the relatively small numbers of animals, i.e., 15 principals and nine controls, that were used in the oral infection studies lasting 48 h. No difference (P > 0.05) was detected between calves infected with DT104ΔSO13-loaded RPz and calves infected with DT104ΔSO13-loaded HEp-2 cells. Also, no difference (P > 0.05) was detected in pathogen loads for calves infected with HEp-2 cells bearing any of the three different strains.

FIG. 4.
MPN-based quantitation of Salmonella recovered from the spleens of calves experimentally challenged with DT104, DT104ΔSO13, or DT104ΔSO13/pSO13 loaded in either RPz (gray bars) or HEp-2 cells (white bars). Each bar represents the mean ...

Figure Figure44 also illustrates a difference in the immunologic response between calves infected with the hyperinvasive strains, i.e., RPz containing DT104 or the complemented mutant. As shown within the bars, banded neutrophils were common in calves infected with the hyperinvasive strains but were rare in the calves infected with DT104ΔSO13-loaded cells. Since 120 banded neutrophils per ml is considered to be significantly elevated (11), a pronounced left shift (1,110 to 1,800 banded neutrophils per ml) was clearly evident in 10 of 10 calves infected with the putative hypervirulent strains while only one of five “parallel” calves (i.e., infected with DT104ΔSO13-laden RPz) responded similarly. None of the nine control calves, i.e., those inoculated with HEp-2 cells, had any banded neutrophils at the time of euthanasia. Hemograms from all 24 calves revealed neutrophilia (data not shown).

As shown in Fig. Fig.5,5, the systemic pathogen burden for DT104ΔSO13/pSO13 exceeded that for DT104ΔSO13 even if the latter infection was allowed to progress. Systemic infection began to wane for all DT104ΔSO13-infected calves when DT104ΔSO13/pSO13-infected calves began to be euthanized. Specifically, two DT104ΔSO13/pSO13-infected calves were euthanized at 48 h, one calf was euthanized at 60 h, and the remaining two calves were euthanized at 72 h postinfection.

FIG. 5.
PCR-based quantitation of Salmonella recovered from the blood of transfaunated calves experimentally challenged with DT104ΔSO13 (circles) or DT104ΔSO13/pSO13 (squares). Each point represents the mean ± standard error of the mean ...


The objective of this study was to identify the SGI1 gene(s) involved in the augmentation of invasion for multiresistant Salmonella exposed to RPz. SGI1 genes were the focus of this study since our recent research indicated that RPz elicit hyperinvasion only in Salmonella strains possessing SGI1 (23). Although DT104 was the first strain found to harbor SGI1 (3), other S. enterica serovar Typhimurium phage types (e.g., DT120, DT193, and U302) have acquired SGI1 (5). Furthermore, SGI1 has been found in other serotypes such as Agona (2), Infantis (5), Meleagridis (12), Paratyphi (20), Albany (9), Newport (10), and Indiana and Senftenberg (26).

In this study we also provide evidence that other protozoa, i.e., Acanthamoeba castellanii, can mediate hyperinvasion in DT104 like that observed for Acanthamoeba castellanii and the invasion of Legionella pneumophila (6). Our recent studies have shown that hyperinvasion does not occur in DT104 recovered from bovine macrophages or from mammalian epithelial cells (23). Therefore, hyperinvasion is not a ubiquitous response after passive or active eukaryotic cell entry. Intraprotozoal processes, perhaps those involved in bacterial predation and/or digestion, appear to selectively hyperactivate DT104 invasion. Identification of these protozoal processes is currently a line of research that we are pursuing. There does appear to be a divergence between the mechanisms underlying the effects mediated by RPz and those observed following exposure to Acanthamoeba castellanii.

The loss-of-function mutant and the complemented mutant provide evidence that SO13, an 885-bp SGI1 gene without an ascribed function (2), mediates the RPz-associated effect in DT104. SO13 is one of 15 SGI1 open reading frames that have not yet been characterized, i.e., protein database searches revealed no orthologs. Secondary structure examinations revealed some novel characteristics but nothing that facilitated ascribing a specific function to SO13.

Studies with TH11/pSO13 suggest that SO13 is not the sole determinant of the phenotype. Since TH11 and DT104 differ only by the presence of SGI1 in DT104, it appears that another SGI1 component is a cofactor. Current studies are under way whereby TH11/pSO13 is transformed with an SGI1 library in search of a subclone that is hyperinvasive in the presence of protozoa. Our studies with the complemented mutant demonstrate that protozoa are required for the phenotype since expression of SO13 did not lead to RPz-independent hyperinvasion even for DT104.

The in vivo studies confirm the in vitro studies in regards to the role of SO13 in facilitating the hyperinvasion of DT104. Based on the inflammatory leukograms observed in the in vivo studies, it appears that a massive inflammatory response is occurring in calves infected with the hyperinvasive strains. As reported in our previous study (23), calves within this group often display pyrexia whereas the control calves remained afebrile at the time of euthanasia. Time course experiments also show that DT104 infections progress rapidly and infected calves exhibit dire clinical signs that warrant intervention. While treatment failures can still be to blame for some reports of apparent DT104 hyperpathogenicity, the peracute nature of bovine DT104 salmonellosis seems to be responsible for the severe prognoses of the cases.

The reporter plasmid pSC104 is a novel aspect of this study that may have broader implications for the assessment of other hyperactivating phenotypes in other bacteria. The main goal of this study was to identify a gene insertion mutant with normal invasive characteristics yet existing within a hyperinvasive background. Since this goal did not seem to be achievable using standard invasion assays, it became necessary to develop a reporter system in which a lack of hyperinvasion leads to activation of a selection marker such as antibiotic resistance. By linking hilA promoter activity to an inhibition of an innate antibiotic resistance gene encoding chloramphenicol resistance, hyperactivation of the hilA promoter would lead to chloramphenicol sensitivity in all transformants exposed to RPz. Disruption of overexpression, via transposon mutagenesis, would therefore lead to a restoration of chloramphenicol resistance. The chloramphenicol-resistant mutant was detected in merely 2 days using 2,000 small pools of transposon mutants. This system could be applied to other strains if the promoter is known for the hyperactivated gene and if the strain has a known antibiotic resistance gene.

In summary, our study suggests that DT104 hyperinvasion is related to RPz exposure and SO13 expression that in turn activates hilA overexpression. Since SO13 is part of SGI1, this phenomenon is restricted to certain multiresistant strains of Salmonella. However, this phenotype may become more prevalent considering the expanding list of SGI1-bearing strains. RPz dependence would thus limit the hyperinvasion to ruminants, although this study now extends the phenotype to other protozoa, such as free-living Acanthamoeba castellanii and including protozoa that may inhabit the intestinal tract. That is, amoeba-contaminated water could be a source of hyperinvasive/multiresistant Salmonella for nonruminants. It is currently unclear if Salmonella would be released from amoebae in monogastric animals. Future studies will assess this possibility as well as identifying other SGI1 genes that mediate hyperinvasion, RPz processes involved in the activation of SO13 expression, and the consistency of this phenomenon in other ruminants. Hopefully, exhaustive characterization will provide the basis for perturbing this aspect of salmonellosis.


We thank Ruth Willson and Deb Lebo for technical assistance, Sandy Johnson and Linda Miller for secretarial assistance, Nathan Horman and Brian Conrad for animal husbandry, and Brad Jones and Jeffrey Beetham for reading the manuscript.

Z. McCuddin and this research were partially supported by beef and veal producers and importers through their $1-per-head checkoff via the Cattlemen's Beef Board and state beef councils through the National Cattlemen's Beef Association.

Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.


Editor: F. C. Fang


Published ahead of print on 4 December 2006.


1. Bolton, L. F., L. C. Kelley, M. D. Lee, P. J. Fedorka-Cray, and J. J. Maurer. 1999. Detection of multidrug-resistant Salmonella enterica serotype Typhimurium DT104 based on a gene which confers cross-resistance to florfenicol and chloramphenicol. J. Clin. Microbiol. 37:1348-1351. [PMC free article] [PubMed]
2. Boyd, D., G. Peters, A. Cloeckaert, K. Boumedine, E. Chaslus-Dancla, H. Imberechts, and M. Mulvey. 2001. Complete nucleotide sequence of a 43-kilobase genomic island associated with the multidrug resistance region of Salmonella enterica serovar Typhimurium DT104 and its identification in phage type DT120 and serovar Agona. J. Bacteriol. 183:5725-5732. [PMC free article] [PubMed]
3. Briggs, C. E., and P. M. Fratamico. 1999. Molecular characterization of an antibiotic resistance gene cluster of Salmonella typhimurium DT104. Antimicrob. Agents Chemother. 43:846-849. [PMC free article] [PubMed]
4. Carlson, S., M. Wu, Z. McCuddin, and V. Sharma. 2005. SlyA partially regulates the collagenase-mediated cytopathic phenotype specific for multiresistant Salmonella enterica serotype Typhimurium phagetype DT104. Microb. Pathog. 38:181-187. [PubMed]
5. Carlson, S. A., L. F. Bolton, C. E. Briggs, H. S. Hurd, V. K. Sharma, P. Fedorka-Cray, and B. D. Jones. 1999. Detection of Salmonella typhimurium DT104 using multiplex and fluorogenic PCR. Mol. Cell. Probes 13:213-222. [PubMed]
6. Cirillo, J. D., S. Falkow, and L. S. Tompkins. 1994. Growth of Legionella in Acanthamoeba castellanii enhances invasion. Infect. Immun. 62:3254-3261. [PMC free article] [PubMed]
7. Clark, M. A., M. A. Jepson, N. L. Simmons, and B. H. Hirst. 1994. Preferential interaction of Salmonella typhimurium with mouse Peyer's patch M cells. Res. Microbiol. 145:543-552. [PubMed]
8. Coleman, G., and D. Reynolds. 1982. The uptake of bacteria and amino acids by Ophryoscolex caudatus, Diploplastron affine and some other rumen Entodiniomorphid protozoa. J. Appl. Bacteriol. 52:135-144.
9. Doublet, B., R. Lailler, D. Meunier, A. Brisabois, D. Boyd, M. R. Mulvey, E. Chaslus-Dancla, and A. Cloeckaert. 2003. Variant Salmonella genomic island 1 antibiotic resistance gene cluster in Salmonella enterica serovar Albany. Emerg. Infect. Dis. 9:585-591. [PMC free article] [PubMed]
10. Doublet, B., F. X. Weill, L. Fabre, E. Chaslus-Dancla, and A. Cloeckaert. 2004. Variant Salmonella genomic island 1 antibiotic resistance gene cluster containing a novel 3′-N-aminoglycoside acetyltransferase gene cassette, aac(3)-Id, in Salmonella enterica serovar Newport. Antimicrob. Agents Chemother. 48:585-591. [PMC free article] [PubMed]
11. Duncan, J., and K. Prasse. 2003. Veterinary laboratory medicine, 4th ed. Iowa State University Press, Ames.
12. Ebner, P., K. Garner, and A. Mathew. 2004. Class 1 integrons in various Salmonella enterica serovars isolated from animals and identification of genomic island SGI1 in Salmonella enterica var. Meleagridis. J. Antimicrob. Chemother. 53:1004-1009. [PubMed]
13. Evans, S., and R. Davies. 1996. Case control study of multiple-resistant Salmonella typhimurium DT104 infection of cattle in Great Britain. Vet. Rec. 139:557-558. [PubMed]
14. Frana, T., and S. Carlson. 2001. Development and use of a plasmid encoding green fluorescence protein in multiple antibiotic resistant Salmonella. BioTechniques 30:28-32. [PubMed]
15. Johnston, C., D. Pegues, C. Hueck, A. Lee, and S. Miller. 1996. Transcriptional activation of Salmonella typhimurium invasion genes by a member of the phosphorylated response-regulator superfamily. Mol. Microbiol. 22:715-727. [PubMed]
16. Jones, B. D., and S. Falkow. 1994. Identification and characterization of a Salmonella typhimurium oxygen-regulated gene required for bacterial internalization. Infect. Immun. 62:3745-3752. [PMC free article] [PubMed]
17. Jones, M., G. Anantharamaiah, and J. Segrest. 1992. Computer programs to identify and classify amphipathic alpha helical domains. J. Lipid Res. 33:287-296. [PubMed]
18. Kaniga, K., D. Trollinger, and J. E. Galan. 1995. Identification of two targets of the type III protein secretion system encoded by the inv and spa loci of Salmonella typhimurium that have homology to the Shigella IpaD and IpaA proteins. J. Bacteriol. 177:7078-7085. [PMC free article] [PubMed]
19. Lee, C. A., B. D. Jones, and S. Falkow. 1992. Identification of a Salmonella typhimurium invasion locus by selection for hyperinvasive mutants. Proc. Natl. Acad. Sci. USA 89:1847-1851. [PMC free article] [PubMed]
20. Meunier, D., D. Boyd, M. R. Mulvey, S. Baucheron, C. Mammina, A. Nastasi, E. Chaslus-Dancla, and A. Cloeckaert. 2002. Salmonella enterica serotype Typhimurium DT 104 antibiotic resistance genomic island 1 in serotype Paratyphi B. Emerg. Infect. Dis. 8:430-433. [PMC free article] [PubMed]
21. NCCLS. 2001. Performance standards for antimicrobial disk and dilution susceptibility tests, 2nd ed., vol. 20. NCCLS, Wayne, PA.
22. Penheiter, K. L., N. Mathur, D. Giles, T. Fahlen, and B. D. Jones. 1997. Non-invasive Salmonella typhimurium mutants are avirulent because of an inability to enter and destroy M cells of ileal Peyer's patches. Mol. Microbiol. 24:697-709. [PubMed]
23. Rasmussen, M., S. A. Carlson, S. K. Franklin, Z. P. McCuddin, M. T. Wu, and V. K. Sharma. 2005. Exposure to rumen protozoa leads to enhancement of pathogenicity of and invasion by multiple-antibiotic-resistant Salmonella enterica bearing SGI1. Infect. Immun. 73:4668-4675. [PMC free article] [PubMed]
24. Sharma, V., and S. Carlson. 2000. Simultaneous detection of Salmonella strains and Escherichia coli O157:H7 with fluorogenic PCR and single-enrichment-broth culture. Appl. Environ. Microbiol. 66:5472-5476. [PMC free article] [PubMed]
25. Slutsker, L., S. F. Altekruse, and D. L. Swerdlow. 1998. Foodborne diseases. Emerging pathogens and trends. Infect. Dis. Clin. N. Am. 12:199-216. [PubMed]
26. Vo, A., E. van Duijkeren, A. Fluit, W. Wannet, A. Verbruggen, H. Maas, and W. Gaastra. 2006. Antibiotic resistance, integrons and Salmonella genomic island 1 among non-typhoidal Salmonella serovars in The Netherlands. Int. J. Antimicrob. Agents 28:172-179. [PubMed]
27. Wall, P., D. Morgan, K. Lamden, M. Ryan, M. Griffin, E. Threlfall, L. Ward, and B. Rowe. 1994. A case control study of infection with an epidemic strain of multiresistant Salmonella typhimurium DT104 in England and Wales. Commun. Dis. Rep. Rev. 4:R130-R135. [PubMed]
28. Wood, R., and R. Rose. 1992. Populations of Salmonella typhimurium in internal organs of experimentally infected carrier swine. Am. J. Vet. Res. 53:69-76. [PubMed]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...