Modulation of human monocytes by Escherichia coli heat-labile enterotoxin B-subunit; altered cytokine production and its functional consequences

Abstract

In murine systems, the B subunit of Escherichia coli heat-labile enterotoxin (EtxB) is a potent immunomodulator capable of suppressing Th1-mediated autoimmune diseases. This results from its ability to bind cell surface receptors, principally GM1-ganglioside, and as a consequence down-regulate the pathological T helper type 1 (Th1) response. The capacity of EtxB to alter human T-cell responses has not been investigated. Here we show that EtxB, but not the receptor non-binding mutant EtxB (G33D), triggers the release of interleukin (IL)-10, IL-6 and tumour necrosis factor-alpha (TNF-alpha) by human monocytes. The production of IL-8, transforming growth factor-beta (TGF-beta) or IL-12 was not enhanced by EtxB. Indeed, EtxB was shown to inhibit IL-12 secretion in monocytes stimulated with interferon-gamma (IFN-gamma) and lipopolysaccharide (LPS) by an IL-10-independent mechanism. When EtxB-treated monocytes were used as antigen presenting cells in an allogeneic mixed lymphocyte reaction (MLR), IL-10 and IFN-gamma production were increased in comparison to levels seen in cultures stimulated with untreated monocytes; proliferation was unaltered. This modulation of the T-cell response was not only evident in the primary MLR triggered by EtxB-treated monocytes, but also upon restimulation of the responding T cells with fresh untreated monocytes; indicating that presentation by EtxB-treated monocytes leads to altered T-cell differentiation. Sorting experiments showed that IL-10 secreting T cells from the MLR cultures were strong suppressors of T-cell proliferation following their addition into a fresh primary MLR.

Figure 1

Cytokine production by EtxB-treated human monocytes. EtxB was added to freshly magnetic-activated cell sorting (MACS)-purified human CD14⁺ cells, then the cells were transferred to celELISA plates either immediately or after 24 or 48 hr and incubated for further 24 hr, allowing the measurement of IL-10 production in the intervals 0–24 hr, 24–48 hr, and 48–72 hr. LPS (from Salmonella abortus equi) was added as a positive control for cytokine production. Values shown are the means of triplicate samples+SEM (although error bars are often too small to be visible over the symbols), and analysis of each cytokine was carried out on at least three separate occasions.

Figure 2

The effect of EtxB upon IL-12 secretion by activated human monocytes. (a) MACS-purified human CD14⁺ cells were cultured in the presence of 40 µg/ml EtxB overnight and were then activated with IFN-γ (100 ng/ml)+LPS (10 µg/ml). The cells were then transferred to IL-12 celELISA plates and IL-12p40 or IL-12p70 production was measured after 24 hr (b) Monocytes were cultured with increasing quantities of EtxB in the presence or absence of a neutralizing anti-IL-10 antibody (50 µg/ml). IL-12p40 and p70 levels were measured after 24 hr by celELISA. Values shown are from a representative experiment (n=3) and are the mean of 3 separate samples+SEM. Dashed lines represents the levels of cytokine detected in unstimulated cell cultures.

Figure 3

EtxB effects upon IL-10 and IL-12 production require ganglioside-GM1 binding and are abrogated by boiling. MACS-purified human CD14⁺ cells were cultured in the presence of 40 µg/ml EtxB for 24 hr, then IL-10 production was measured by celELISA (a) or were further activated with IFN-γ (100 ng/ml)+LPS (10 µg/ml) to assess IL-12p40 secretion in the following 24 hr (b). EtxB(G33D), a mutant protein that does not bind to GM1 lacks the effects of the wild-type protein. Also, heat denaturation of EtxB (100°, 20 min) eliminates the IL-10-inducing/ IL-12-suppressing activity of the EtxB preparation. The values are the mean+SEM for three separate samples, and the data are representative from three similar experiments.

Figure 4

Cell proliferation and cytokine secretion in MLR induced using EtxB-treated monocytes as APC. (a) MACS-isolated CD14⁺ APC were either cultured overnight in the presence of 40 µg/ml EtxB or left untreated, then were washed and cocultured with allogeneic CD4⁺ responder T cells. After 7 days, the remaining cells were harvested and washed before being used to establish secondary MLR using untreated monocytes that had been stored frozen from the original donor. Cell proliferation was assessed by measuring [³H]thymidine incorporation. (b,c). To measure cytokine secretion in the primary (b) and secondary (c) MLR, cells were transferred to celELISA plates and cytokine production was measured after 24 hr (Open bars, EtxB-treated APC; full bars, untreated APC). The cytokine measurements were carried out on day 5 of the primary MLR and on the first day following establishment of the secondary MLR. The values for proliferation and cytokine levels are the mean+SEM from three samples, and the data are representative of experiments performed with three donor combinations.

Figure 5

IL-10-secreting CD3⁺ lymphocytes in MLR cultures can suppress T cell proliferation. (a) Allogeneic MLR was set up using CD14⁺ monocytes either treated overnight with 40 µg/ml EtxB or left untreated as APC while the responders were allogeneic CD4⁺ T cells. After restimulation on day 7 with fresh untreated APC, cells were encapsulated into agarose microdroplets containing bound IL-10 capture antibodies. Cells were then cultured overnight to allow IL-10 secretion, washed, stained for CD3 and IL-10 and analysed by flow cytometry. (b) Unsorted cells (open bars) or cells sorted as CD3⁺ IL-10⁺ (hatched bars) or CD3⁺ IL-10⁻ (closed bars) were obtained, the agarose capsules were digested and the released regulatory cells were added (at a ratio of 1:10, regulators:responders) to a new primary MLR. Proliferation in the recipient MLR in the absence of added cells (shaded bar) is also shown. Cell proliferation was assessed by measuring [³H]thymidine incorporation and the data shown represent the means and SEM of triplicate cultures and are representative of two identical experiments.