Distinct isoforms of phospholipase A2 mediate the ability of Salmonella enterica serotype typhimurium and Shigella flexneri to induce the transepithelial migration of neutrophils

Abstract

Salmonella spp. and Shigella spp. are responsible for millions of cases of enteric disease each year worldwide. While these pathogens have evolved distinct strategies for interacting with the human intestinal epithelium, they both induce significant proinflammatory responses that result in massive transepithelial migration of neutrophils across the intestinal mucosa. It has previously been shown with Salmonella enterica serotype Typhimurium that the process of neutrophil transmigration is mediated in part by the secretion of hepoxilin A(3) (HXA(3); 8-hydroxy-11,12-epoxy-eicosatetraenoic acid), a potent neutrophil chemoattractant, from the apical surface of infected model intestinal epithelium. This study confirms that HXA(3) is also secreted in response to infection by Shigella flexneri, that it is produced by a pathway involving 12/15-lipoxygenase (12/15-LOX), and that S. enterica serovar Typhimurium and S. flexneri share certain elements in the mechanism(s) that underlies the otherwise separate signal transduction pathways that are engaged to induce polymorphonuclear leukocyte (PMN) transepithelial migration (protein kinase C and extracellular signal-regulated kinases 1 and 2, respectively). PMN transepithelial migration in response to infection with S. flexneri was dependent on 12/15-LOX activity, the enzyme responsible for the initial metabolism of arachidonic acid to HXA(3). Probing further into this pathway, we also found that S. enterica serovar Typhimurium and S. flexneri activate different subtypes of phospholipase A(2), a critical enzyme involved in the liberation of arachidonic acid from cellular membranes. Thus, although S. enterica serovar Typhimurium and S. flexneri utilize different mechanisms for triggering the induction of PMN transepithelial migration, we found that their reliance on 12/15-LOX is conserved, suggesting that enteric pathogens may ultimately stimulate similar pathways for the synthesis and release of HXA(3).

FIG. 1.

Identification of HXA₃ in the apical supernatants of T-84 monolayers. HXA₃ was identified by employing a triple quadrupole LC/MS-MS system as described in Materials and Methods. (A) Diagnostic fragment ions were identified by MS-MS enhanced-product analyses of synthetic HXA₃. (B) Released endogenous HXA₃ was identified by multiple reaction monitoring for the specific transition 335→127 m/z and matching retention time (14.5 min) with synthetic HXA₃; exposure was to HBSS⁺ (dashed line) or S. flexneri (solid line).

FIG. 2.

The effect of 12- and 5-LOX inhibitors on PMN transepithelial migration across T-84 monolayers. Prior to infection, T-84 monolayers were exposed to buffer only (black bars) or to the 12-LOX inhibitor baicalein (gray bars) or the 5-LOX inhibitor caffeic acid (white bars) for 48 or 24 h, respectively. (A) PMN transepithelial migration induced by wild-type (WT) S. flexneri or the virulence plasmid-cured mutant BS103. (B) PMN migration across uninfected T-84 monolayers in the presence of 1 μM PMN chemoattractant (fMLP) or buffer only (−). Values are shown as percentages of PMNs that crossed from the basolateral to the apical chamber in untreated monolayers within a given condition (i.e., 100% of wild-type or fMLP response). The data are expressed as the mean ± standard deviation for triplicate samples and represent one of at least three independent experiments performed with similar results. *, P < 0.01.

FIG. 3.

HCT-8 cells transfected with a vector control (black bars) or a vector modified to generate siRNAs aimed at decreasing the expression of ALOX15, the gene encoding human 12/15-LOX (gray bars). (A) PMN transepithelial migration induced by wild-type (WT) S. enterica serovar Typhimurium. (B) PMN transepithelial migration induced by wild-type S. flexneri. (C) PMN migration across uninfected T-84 monolayers in the presence of 1 μM PMN chemoattractant (fMLP). (D) Western blot analysis of HCT-8 monolayers infected with S. flexneri demonstrating reduced levels of 15-LOX (∼66 kDa) in monolayers transfected with the siRNA-generating plasmid compared to those seen for of the vector control. Detection of GAPDH levels served as the internal control for protein loading conditions. The data represent a Western blot from an individual experiment performed at least three times. PMN values are shown as percentages of PMNs that crossed from the basolateral to the apical chamber in untreated monolayers within a given condition (i.e., 100% of wild-type or fMLP response). The data are expressed as the mean ± standard deviation for triplicate samples and represent one of at least three independent experiments performed with similar results. (−), HBSS⁺ only; *, P < 0.01.

FIG. 4.

The effect of the general PLA₂ inhibitor ONO-RS-082 on PMN transepithelial migration across T-84 monolayers. Prior to infection, T-84 monolayers were exposed to buffer only (black bars) or to ONO-RS-082 at 5 μM (gray bars), 10 μM (white bars), or 20 μM (striped bars) for 2 h. (A) PMN transepithelial migration induced by wild-type (WT) S. enterica serovar Typhimurium or S. flexneri. (B) PMN migration across uninfected T-84 monolayers in the presence of 1 μM PMN chemoattractant (fMLP) or buffer only (−). Values are shown as percentages of PMNs that crossed from the basolateral to the apical chamber in the untreated monolayers within a given condition (i.e., 100% of wild-type or fMLP response). The data are expressed as the mean ± standard deviation for triplicate samples and represent one of at least three independent experiments performed with similar results. *, P < 0.01; **, P < 0.05.

FIG. 5.

The effect of the specific sPLA₂ inhibitor 4-bromophenacyl bromide on PMN transepithelial migration across T-84 monolayers. Prior to infection, T-84 monolayers were exposed to buffer only (black bars) or to 4-bromophenacyl bromide at 0.07 μM (gray bars), 0.7 μM (white bars), or 7.0 μM (striped bars) for 2 h. (A) PMN transepithelial migration induced by wild-type (WT) S. enterica serovar Typhimurium or the noninvasive hilA mutant vv341. (B) PMN transepithelial migration induced by wild-type S. flexneri or the virulence plasmid-cured mutant BS103. (C) PMN migration across uninfected T-84 monolayers in the presence of 1 μM PMN chemoattractant (fMLP) or buffer only (−). Values are shown as percentages of PMNs that crossed from the basolateral to the apical chamber in the untreated monolayers within a given condition (i.e., 100% of wild-type or fMLP response). The data are expressed as the mean ± standard deviation for triplicate samples and represent one of at least three independent experiments performed with similar results. No significant differences were observed with increased doses of the inhibitor.

FIG. 6.

The effect of the specific cPLA₂α inhibitor on PMN transepithelial migration across T-84 monolayers. Prior to infection, T-84 monolayers were exposed to buffer only (black bars) or cPLA₂α inhibitor at 0.06 μM (gray bars), 0.6 μM (white bars), or 6.0 μM (striped bars) for 2 h. (A) PMN transepithelial migration induced by wild-type (WT) S. enterica serovar Typhimurium or the noninvasive hilA mutant vv341. (B) PMN transepithelial migration induced by wild-type S. flexneri or the virulence plasmid-cured mutant BS103. (C) PMN migration across uninfected monolayers in the presence of 1 μM PMN chemoattractant (fMLP) or buffer only (−). Values are shown as percentages of PMNs that crossed from the basolateral to the apical chamber in the untreated monolayers within a given condition (i.e., 100% of wild-type or fMLP response). The data are expressed as the mean ± standard deviation for triplicate samples and represent one of at least three independent experiments performed with similar results. *, P < 0.01; **, P < 0.05.

FIG. 7.

The effect of the specific iPLA₂ inhibitor BEL on PMN transepithelial migration across T-84 monolayers. Prior to infection, T-84 monolayers were exposed to buffer only (black bars) or to BEL at 1 μM (gray bars), 5 μM (white bars), or 25 μM (striped bars) for 2 h. (A) PMN transepithelial migration induced by wild-type (WT) S. enterica serovar Typhimurium or the noninvasive hilA mutant vv341. (B) PMN transepithelial migration induced by wild-type S. flexneri or the virulence plasmid-cured mutant BS103. (C) PMN migration across uninfected monolayers in the presence of 1 μM PMN chemoattractant (fMLP) or buffer only (−). Values are shown as percentages of PMNs that crossed from the basolateral to the apical chamber in the untreated monolayers within a given condition (i.e., 100% of wild-type or fMLP response). The data are expressed as the mean ± standard deviation for triplicate samples and represent one of at least three independent experiments performed with similar results. *, P < 0.01; **, P < 0.05.

FIG. 8.

HCT-8 cells transfected with a vector control (black bars) or a vector modified to generate siRNAs aimed at decreasing the expression of the gene encoding iPLA₂ (gray bars). (A) PMN transepithelial migration induced by wild-type (WT) S. enterica serovar Typhimurium. (B) PMN migration across uninfected T-84 monolayers in the presence of 1 μM PMN chemoattractant (fMLP). (C) Western blot analysis of HCT-8 monolayers infected with S. enterica serovar Typhimurium demonstrating reduced levels of iPLA₂ (∼80 kDa) in monolayers transfected with the siRNA-generating plasmid compared to those seen for the vector control. Detection of GAPDH levels served as the internal control for protein loading conditions. The data represent a Western blot from an individual experiment performed at least three times. PMN values are shown as percentages of PMNs that crossed from the basolateral to the apical chamber in untreated monolayers within a given condition (i.e., 100% of wild-type or fMLP response). The data are expressed as the mean ± standard deviation for triplicate samples and represent one of at least three independent experiments performed with similar results. (−), HBSS⁺ only; **, P < 0.05.

FIG. 9.

S. enterica serovar Typhimurium and S. flexneri induce PMN transepithelial migration through the utilization of distinct PLA₂ enzymes. S. enterica serovar Typhimurium interacts with intestinal epithelial cells from the apical surface, and the effector protein, SipA, activates a novel signal transduction cascade that leads to the activation of PKC (*, activated PKC). We postulate (dashed arrow) that this PKC-dependent signal transduction cascade leads to the activation of iPLA₂, which liberates arachidonic acid (AA) from membranes for subsequent metabolism by the 12/15-LOX pathway, culminating in the synthesis of HXA₃. S. flexneri gains access to and interacts with the intestinal epithelium from the basolateral surface. Once inside the host cell, S. flexneri secretes the effector proteins OspF and OspC1, which leads to the phosphorylation (P) of ERK1/2. ERK1/2 activation is a prominent trigger (dashed arrow) which leads to the activation of cPLA₂. This enzyme releases arachidonic acid from membranes, where is becomes an available substrate for the 12/15-LOX pathway and leads to the eventual production of HXA₃. IL-8, interleukin-8. Large black ovals represent cell nuclei.