|Home | About | Journals | Submit | Contact Us | Français|
The invasion-associated type III secretion system (T3SS-1) of S. Typhimurium is required to initiate and sustain an acute inflammatory response in the intestine. We investigated the relationship of S. Typhimurium T3SS-1-induced IL-8 expression and invasion with intracellular Ca2+ mobilization in HeLa cells. Compared to the sipAsopABDE2 mutant, strains carrying a mutation in sipA, or mutations in sopABDE2 induced higher levels of IL-8 and greater bacterial internalization despite the fact that these mutants elicited similarly low intracellular concentrations of Ca2+. Likewise, complemented sipAsopABDE2 mutant with sopE2 did not affect intracellular Ca2+ concentrations or IL-8 expression, but significantly increased bacterial internalization. Treating HeLa cells with the calcium chelator BAPTA-AM or with D-BAPTA-AM, a derivative with greatly reduced Ca2+ chelating activity, yielded strong evidence that BAPTA-AM does not affect invasion and inhibits IL-8 secretion by a calcium-dependent mechanism. These findings suggest that, although wild-type S. Typhimurium-induced IL-8 expression and bacterial internalization in HeLa cells coincides with increased cytosolic Ca2+, the differing levels of IL-8 and invasion induced by strains carrying different effector proteins are unrelated to levels of intracellular Ca2+.
The Salmonella enterica serovar Typhimurium (S. Typhimurium) effector proteins SipA, SopA, SopB, SopD and SopE2, translocated through the Salmonella Pathogenicity Island-1 (SPI-1) encoded type III secretion system (T3SS-1), play important roles in generating an inflammatory response in the bovine ligated ileal loop model. A S. Typhimurium strain carrying mutations in sipA, sopA, sopB, sopD and sopE2 causes 60% less inflammatory responsiveness and decreased fluid accumulation relative to wild type infected loops. On the other hand, mutations in the slrP, avrA, sspH1 or sptP genes that encode other T3SS-1 effector proteins did not reduce fluid accumulation or the degree of inflammation in bovine ligated ileal loops. The degree of attenuation observed in a sipAsopABDE2 mutant was similar to that of a S. Typhimurium sipB mutant , a strain deficient for translocating T3SS-1 effector proteins into host cells. These findings suggest that SipA, SopA, SopB, SopD and SopE2 are the major effector proteins required for S. Typhimurium-induced enteritis in calves.
S. Typhimurium T3SS-1-translocated effector proteins, SipA, SopA, SopB, SopD and SopE2, induce cytoskeletal rearrangements leading to bacterial internalization . SopA associates with a host ubiquitin E3 ligase inducing bacterial escape into the cytosol of epithelial cells . SopE2, is a guanine nucleotide exchange factor , and SopB, is an inositol phosphate phosphatase . Phosphatidylinositol phosphates (generated by SopB) and activated Rho-family GTPases (produced by SopE2) act in concert to activate WASP/Scar proteins, which in turn recruit the Arp2/3 complex to initiate the formation of new branches on actin filaments . SopD acts cooperatively with SopB to promote membrane fission and macropinosome formation during the invasion process . SipA, which is delivered into host cells as quickly as 10 sec after exposure , localizes in the host cell plasma membrane  where it acts as an actin-binding protein .
Once in contact with host cells, S. Typhimurium was found to alter cytosolic Ca2+ concentrations within 2 to 4 minutes after infection [11-14]. Ca2+ is an important intracellular messenger involved in many cellular functions, including vesicular trafficking , cytoskeletal rearrangements  and gene expression . Therefore, changes in intracellular Ca2+ level, depending on the amplitude of transients, may influence different cell signaling pathways .
Other bacterial pathogens have been found to alter Ca2+ homeostasis in host cells. Although intracellular Ca2+ has been implicated in mediating cytoskeletal rearrangements, the dependence of bacterial invasion on changes in Ca2+ fluxes is controversial. Invasion of Shigella flexneri in epithelial cells induces an increase in intracellular Ca2+ that is dependent on a functional T3SS, but cytoskeletal rearrangements are also observed in cells with no detectable Ca2+ response . Listeria monocytogenes and Campylobacter jejuni induce intracellular Ca2+ changes in host cells [20, 21]. While chelation of intracellular Ca2+ interferes with C. jejuni internalization , it has no effect on L. monocytogenes invasion . Previous studies suggest that S. Typhimurium invasion is dependent on the induction of cytosolic Ca2+ increases , and that chelation of Ca2+ inhibits bacterial internalization . However, a non-invasive S. Typhimurium ΔhilA mutant induces cytosolic Ca2+ changes , indicating that host cell internalization is not a prerequisite for inducing Ca2+ mobilization.
Because Ca2+ is central to many cell signaling events, including the expression of IL-8 and the induction of cytoskeletal changes, and the S. Typhimurium effector proteins SipA, SopA, SopB, SopD and SopE2 induce inflammatory responses and bacterial invasion, we focused our studies on the role(s) of intracellular Ca2+ in S. Typhimurium-induced invasion and the induction of IL-8 expression.
To correlate cytosolic calcium changes with IL-8 expression, bacterial internalization, and effector proteins, we used HeLa S3 cells, in which, as opposed to T84 cells, different profiles of responses to infection of S. Typhimurium wild type and mutant can be detected , therefore, questions related to specific effector proteins can be addressed. HeLa S3 cells (human cervical epithelial cells – ATCC) were grown in Kaighn's modification of Ham's F12 medium with 2 mM L-glutamine and 1.5 g/L sodium bicarbonate and supplemented with 10% fetal bovine serum. Before performing assays, HeLa cells were grown to a concentration of 0.2 × 105 cells/cm2 in 24 well polystyrene plates (Corning, Acton, MA), or in 0.1 mg of collagen type I-pretreated Coverglass Chambers (Nalge Nunc International, Rochester, NY) for 48 hr. Strains of bacteria used in this study are listed in Table 1. S. Typhimurium strains were grown in Luria-Bertani (LB) medium overnight with nalidixic acid 50 mg/liter, then sub-cultured in LB for 4 hr .
Expression levels of IL-8 in S. Typhimurium infected cells were measured by quantitative real time-PCR (qRT-PCR). HeLa cells were infected with Salmonella strains (Table 1) at multiplicity of infection (MOI) 200 [24, 25] for 1 hr. After infection, total RNA was extracted and reverse transcription was performed using the Cells to cDNA kit (Applied Biosystem, Foster City, CA). The qRT-PCR was performed in a Smart Cycler II (Cepheid, Sunnyvale, CA) by mixing the cDNA with OmniMix beads (Cepheid, Sunnyvale, CA), SYBG Green I 1× (Invitrogen Corporation, Carlsbad, CA) and 500 nM of each primer set (IL-8 forward primer: AAGGAACCATCTCACTGTGTGTAAAC, IL-8 reverse primer: ATCAGGAAGG CTGCCAAGAG). The cDNA concentration of each sample was normalized to an internal control, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (GAPDH forward primer: GAAGGTGAAGGTCGGAGTC, GAPDH reverse primer: GAAGATGGTGATGGGATTTC). The normalized levels of gene expression in infected cells were calculated relative to uninfected cells as previously described . The effect of an intracellular Ca2+ chelator on gene expression was performed by treating HeLa cells with 50 μM BAPTA-AM (Calbiochem, San Diego, CA), or D-BAPTA–AM (provided by Dr. Rousseau, CEA/Saclay, France) in HBSS media for 45 min  prior to Salmonella infection. Gene expression analysis of BAPTA-AM treated cells was performed as described above.
Expression levels of sipA in bacteria culture were measured by qRT-PCR as described above, using sipA specific primers (SipA forward primer: AAGATTGCTGCGGGTTAACG, SipA reverse primer: TGGCTGCCAGAAACAA AGAA). S. Typhimurium strains were grown overnight and sub-cultured for 4 hr. S. Typhimurium RNA was extracted using the SV total RNA isolation kit (Promega Corporation, Madison, WI). The cDNA concentration of each sample was normalized to an internal control, rrsB 16S ribosomal RNA (RrsB forward primer: GGCAGGCCTAACACATGCA, RrsB reverse primer: CTTGCGACGTTATGCGG TATT). The normalized levels of gene expression in sipAsopABDE2 mutant complemented with sipA and sopABDE2 mutant were calculated relative to S. Typhimurium wild type .
Bacteria were grown in LB medium overnight with appropriate antibiotics  then sub-cultured in LB for 4 hr. The bacteria cultures were centrifuged, and the bacteria extracts were obtained by lysing cells with 20 mM MOPS, pH 7.0, 0.5% Triton X-100, and protease inhibitor mixture. The homogenates were sonicated twice for 15 sec, centrifuged for 30 min at 16,000 × g, and the supernatant removed for further assays. The protein concentration of bacteria cell extracts was determined using a commercial Bradford assay (Sigma, St. Louis, MO). An equal amount of protein obtained from bacteria cell extract was separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), transferred to immobilon transfer membrane (Millipore Coorporation, Bedford, MA), and subsequently reacted with mouse anti-S. Typhimurium SipA antibody (Center for Innovations in Medicine, Biodesign Institute, Arizona State University, Tempe, Arizona), and anti-mouse peroxidase secondary antibody (Sigma, St. Louis, MO).
HeLa cells were infected with Salmonella strains (Table 1) at MOI 200 [24, 25] for 1 hr. After infection, plates were centrifuged for 5 min at 1000 rpm and incubated at 37°C for 1 hr. Cells were washed three times with 500 μl/well of phosphate buffer saline (PBS) and treated with 100 μg/ml of gentamicin (Invitrogen Corporation, Carlsbad, CA) for 1 hr at 37°C. After antibiotic treatment, cells were washed again with PBS and incubated with 1% Triton X-100 for 5 min at 37°C. The number of internalized bacteria was determined by plating ten-fold serial dilutions of the cell lysates in nalidixic acid containing LB plates. To analyze the effect of BAPTA-AM on bacteria internalization, cells were treated with 50 uM BAPTA -AM (Calbiochem, San Diego, CA), or D-BAPTA – AM in HBSS media for 45 min  followed by Salmonella infection, and an invasion assay was performed as described above.
The intracellular Ca2+ levels were assessed by confocal microscopy as described previously [18, 27, 28]. HeLa cells grown on Coverglass Chamber slides (Nalge Nunc International, Rochester, NY) were loaded with 3 uM of Fluo-4 (Invitrogen Corporation, Carlsbad, CA) for 1 hr at 37°C and washed with serum- and phenol red-free medium. Fluo-4 loaded epithelial cells were then examined with Meridian Ultima confocal microscope using 488 nm wavelength for excitation and 530 nm for emission. Basal intracellular Ca2+ level was obtained during the first four scans, then the cells were infected with S. Typhimurium strains and Ca2+ was monitored for 5 min. We examined cells using differential interference contrast and confocal microscopy immediately after infection, and we were able to readily detect bacterial contact with the cell surface (data not shown). The changes in intracellular Ca2+ after infection at any time (t) were normalized to basal Ca2+ level at time (0). Cells treated with10 μM thapsigargin (Sigma, St. Louis, MO) were used as a positive Ca2+ control and PBS as a negative control. To verify that the magnitude of intracellular Ca2+ responses is biologically meaningful, absolute concentrations of cytosolic Ca2+ were assessed using the Ca2+ calibration kit #2 (Invitrogen Corporation, Carlsbad, CA). To chelate intracellular Ca2+, HeLa cells were treated with 50 uM BAPTA-AM (Calbiochem, San Diego, CA), or D-BAPTA–AM in HBSS media for 45 min  followed by Salmonella infection.
To analyze the effects of S. Typhimurium effector proteins on intracellular Ca2+, induction of gene expression, and invasion in HeLa cells, geometric means and standard deviations were determined and statistical significance was calculated using Student's t-test or one-way ANOVA followed by Tukey's multiple comparison test at P < 0.05.
Infecting HeLa cells with a series of defined S. Typhimurium mutants defective in secreting individual or combinations of T3SS-1 effector proteins demonstrated that the S. Typhimurium effector proteins SipA, SopA, SopB, SopD and SopE2 act in concert to trigger IL-8 expression in HeLa cells, as previously demonstrated . As shown in Fig. 1A, strains carrying sipA (sopABDE2 mutant and sipAsopABDE2 mutant complemented with plasmid encoding sipA), or strains lacking sipA, but carrying the genes sopA, sopB, sopD and sopE2 induced higher levels of IL-8 expression than strains lacking the genes encoding all five effector proteins (sipAsopABDE2 mutant). Previously, we have shown, that induction of IL-8 expression detected by qRT-PCR leads to secretion of IL-8 . Most observations on S. Typhimurium invasion of HeLa cells paralleled those on triggering IL-8 expression. The sipAsopABDE2 mutant complemented with a plasmid encoding sipA, the sopABDE2 mutant, and the sipA mutant invaded HeLa cells at significantly higher levels than the sipAsopABDE2 mutant (Fig. 1B). However, the sipAsopABDE2 mutant complemented with a plasmid encoding sopE2 did not induce IL-8 expression, but it partially complemented the invasion defect of the sipAsopABDE2 mutant (Fig. 1B).
Because bacteria-host cell interactions induce Ca2+ responses in the early stages of infection [11-14, 19-21], and intracellular Ca2+ is involved in the activation of many cell signaling pathways [15-17], we investigated whether the level of Ca2+ is proportional to the level of invasion and gene expression in S. Typhimurium infected HeLa cells. We found that the S. Typhimurium wild type significantly increased the intracellular Ca2+ concentration (from resting [Ca2+] ~70nM to ~519 nM) at 30 sec after infection (Fig. 2A and and2B),2B), which has been previously shown [11-14]. Changes in the Ca2+ concentration were also observed when cells were infected with other S. Typhimurium strains (Fig. 2C). However, S. Typhimurium wild type increased intracellular Ca2+ concentrations in HeLa cells at significantly higher levels than the sipAsopABDE2 mutant. Complementation of sipAsopABDE2 mutant with the cloned sipA gene increased intracellular Ca2+ concentrations at levels similar to that induced by wild type (Fig. 2C). The sipAsopABDE2 mutant complemented with sopA induced intermediate levels of Ca2+. On the other hand, the sipAsopABDE2 mutant complemented with sopD, the sipA mutant, and the sopABDE2 mutant induced lower levels of Ca2+ as compared to sipAsopABDE2 mutant (Fig. 2C). The reduced Ca2+ response induced by sipA mutant, sopABDE2 mutant and sopD complemented strains were unexpected and to further understand it, experiments in parallel with SPI-1 mutant using multiple mutants will need to be performed. The current study, however, was focused on finding whether levels of Ca2+ are responsible for the different levels of IL-8 and invasion induced by several S. Typhimurium strains.
Compared to the sipAsopABDE2 mutant, strains carrying a mutation in sipA, or mutations in sopABDE2 induced higher levels of IL-8 and greater bacterial internalization despite the fact that these mutants elicited similarly low intracellular concentrations of Ca2. Likewise, the sipAsopABDE2 mutant complemented with sopE2 induced higher levels of bacterial internalization while the intracellular concentrations of Ca2+ remained low (Table 2). These findings suggest that the level of Ca2+ induced was not proportional to the levels of IL-8 expression and invasion induced by each S. Typhimurium strain.
In summary, infection of HeLa cells with S. Typhimurium was generally accompanied by a significant increase in the intracellular Ca2+ concentration when compared to uninfected cells, which averaged approximately 70nM; however, the cellular events (invasion and IL-8 expression) induced by different S. Typhimurium strains are more likely due to the presence of effector proteins (SipA or SopE2 or combination of SopA, SopB, SopD and SopE2) rather than the levels of Ca2+ (Table 2).
We performed additional experiments to investigate the discrepancies regarding alterations in Ca2+ generated by strains carrying the sipA gene cloned on a low copy number plasmid (sipAsopABDE2 mutant complemented with sipA) and a strain carrying a chromosomal copy of sipA (sopABDE2 mutant). qRT-PCR indicated that sipA mRNA levels were similar in the sopABDE2 mutant and the S. Typhimurium wild type. In contrast, the sipAsopABDE2 mutant complemented with the cloned sipA gene expressed sipA at 177.5-fold higher levels than the wild type (Fig. 3A). Western blot analysis of the bacteria whole cell lysate confirmed a higher level of SipA in sipAsopABDE2 mutant complemented with sipA than in sopABDE2 mutant and in wild type (Fig. 3B). The higher level of Ca2+ found in cells infected with sipAsopABDE2 mutant complemented with sipA may be due to the artificially increased expression of sipA by cloning the gene on a plasmid. Extracellular addition of SipA (up to 40 μg), however, failed to mobilize Ca2+ (data not shown). Interestingly, it has been found that SipA is delivered into host cells through T3SS-1 beginning at 10-90 sec after bacteria-cell contact , which may explain the SipA-induced increases in cytosolic Ca2+ levels at 30 sec after infection. Immediately after cell contact, the sipAsopABDE2 mutant complemented with sipA may secrete higher amounts of SipA into the host cell, thereby triggering higher levels of cytosolic Ca2+ responses.
While high levels of SipA (expressed from sipAsopABDE2 mutant complemented with sipA) induced Ca2+ responses at wild type levels, IL-8 expression and IL-8 secretion  were observed when sipA was present either on the chromosome (low levels of SipA) or on a plasmid (high levels of SipA). As mentioned previously, the Ca2+ response occurs immediately after bacteria-cell contact [11-14, 19-21], whereas IL-8 expression was detected at 1 hr post infection. Thus, even low concentrations of internalized SipA or other S. Typhimurium proteins (SopA, SopB, SopD and SopE2) are able to interact with host cell signaling molecules and induce IL-8 expression at relatively low levels of cytosolic Ca2+.
Previous studies indicated that chelating intracellular Ca2+ with BAPTA-AM inhibits S. Typhimurium invasion and IL-8 expression, suggesting that those cellular responses are mediated through changes in Ca2+ homeostasis [11, 22]. We tested the effect of BAPTA-AM on S. Typhimurium invasion and IL-8 gene expression. Treating HeLa cells with 50 μM of BAPTA-AM had no significant effect on S. Typhimurium internalization (Fig. 4A), which was consistent with results from a previous study demonstrating that 10 μM of BAPTA-AM had no effect on S. Typhimurium or L. monocytogenes invasion in Hep2 cells . It should be mentioned that the use of BAPTA-AM to study bacteria internalization has yielded conflicting results, as others have found that 200 μM of BAPTA-AM reduces S. Typhimurium invasion by 87% in HeLa cells , and also affects C. jejuni internalization in human intestinal cells .
It has been previously demonstrated that S. Typhimurium stimulates Ca2+ mobilization in epithelial cells [11-14], which was required to activate IL-8 production, yet increases in Ca2+ levels and IL-8 secretion were absent in S. Typhimurium infected T84 cells previously treated with BAPTA-AM . In agreement with these studies [11-14], we found that S. Typhimurium alters Ca2+ levels as early as 30 sec after infection, and that the use of BAPTA-AM inhibited cytosolic Ca2+ transients and S. Typhimurium and thapsigargin-induced IL-8 expression in HeLa cells (Fig. 4B). Several other studies have demonstrated, however, that BAPTA-AM can interfere with cell signaling events independently of changes in the Ca2+ concentration [29, 30]. In different cell types, 10-50 μM of BAPTA-AM acts on actin and microtubule disassembly and affects mitochondrial function independent of its Ca2+ chelating activity . Additionally, BAPTA-AM has a dose-dependent-neurotoxic effect on primary neuronal cells, reduces global protein synthesis and induces an endoplasmic reticulum (ER)-specific stress response by influencing the ER Ca2+ homeostasis . Therefore, the use of BAPTA-AM to elucidate the role of Ca2+ on cell signaling pathways may lead to controversial results not only due to dose dependent effects, but also due to non-specific effects on metabolic pathways.
To test whether down-regulation of IL-8 in BAPTA-AM-treated cells is due to reduced levels of Ca2+ or BAPTA-AM-side effects, we treated HeLa cells with BAPTA-AM or D-BAPTA-AM, a BAPTA-AM derivative lacking one acetic acid group with reduced Ca2+ chelating activity . As shown previously , BAPTA-AM but not D-BAPTA-AM inhibited thapsigargin-induced Ca2+ transients (data not shown). Treating HeLa cells with either D-BAPTA or BAPTA-AM reduced S. Typhimurium-induced IL-8 expression (Fig. 4C), indicating that down-regulation of IL-8 in BAPTA-AM treated cells is due to effects of this reagent unrelated to its ability to chelate Ca2+.
The authors thank Drs. Shuping Zhang, Paul Wilson and Quynhtien Tran for providing selected mutants strains and plasmids used in this study and for assisting on protein precipitation procedures. This work was supported by grants from the NIH grant A144170 and the USDA NRICGP 2002-35204-11624. Dr. Sara Lawhon was supported by NIH grant AI060933.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.