The stress protein gp96 is not an activator of resting rat bone marrow-derived dendritic cells, but is a costimulator and activator of CD3+ T cells

Abstract

Although low doses of tumor-derived stress protein gp96 elicit protective immunity to the tumor from which it is isolated, protection is lost at high doses because of the induction of immunoregulatory CD4+ T cells. This study evaluated the influence of gp96 on resting rat bone marrow-derived dendritic cells (BMDCs) and purified CD3+ T cells. In contrast to previous reports, gp96 had no effect on adhesion and costimulatory molecule expression by BMDCs, nor did it influence interleukin (IL)-10 and IL-12 secretion or their allostimulatory capacity. Gp96 did not bind to BMDCs but dose-dependently bound to CD4+ and CD8+ T cells. At low concentrations (1 and 25 microg/mL), gp96 acted as a costimulator of CD3+ T cells, inducing proliferation and the secretion of interferon (IFN)-gamma- and IL-10. Gp96 also increased the proliferation of CD28-costimulated CD3+ T cells and their secretion of IFN-gamma, IL-4, and IL-10. Gp96 had no effect at higher concentrations (50 and 100 microg/mL), despite the occurrence of cell surface binding at these concentrations. These findings indicate that gp96 can act as a costimulatory molecule for CD3+ T cells, and an observed increase in the IL-10: IFN-gamma secretion ratio induced by gp96 suggests that it might, at appropriate concentrations, promote a regulatory T-helper 2 (Th2)-like phenotype.

Fig 1.

Influence of gp96 on cell surface antigen expression by resting rat BMDCs. Adherent Wistar BMDCs were generated and cultured for an additional 24 hours with gp96 or LPS (1 μg/mL). The proportion (proportion, %; left panel) of OX62⁺ BMDCs expressing CD11b (n = 12 experiments), CD40 (n = 4), CD54 (n = 13), CD80 (n = 4), CD86 (n = 12), and MHC class II (n = 12) and its intensity of expression (median channel of fluorescent intensity, MFI; right panel) were determined by flow cytometry. Data are presented as mean ± SEM

Fig 2.

Influence of gp96 on IL-10 and IL-12 subunit (IL-12_p40) and heterodimer (IL-12_p70) secretion by resting rat BMDCs. Adherent Wistar BMDCs were generated and cultured with gp96 or LPS (1 μg/mL) without harvesting for an additional 24 hours. Levels of IL-10 (n = 10 independent experiments), IL-12_p40 (n = 10), and IL-12_p70 (n = 5) in cell-free culture supernatants were determined by enzyme immunoassay. Data are presented as mean ± SEM

Fig 3.

Forward (FSC) and side (SSC) light scatter characteristics of resting gp96-treated and LPS-activated rat BMDCs. Adherent Wistar BMDCs were generated and cultured for an additional 24 hours with gp96 or LPS (1 μg/mL). The region in which the majority of cells were apparent in control cultures has been superimposed on FSC vs SSC scatterplots for cells cultured with gp96 and LPS for illustrative purposes. Plots are representative of at least 5 independent experiments

Fig 4.

Effect of gp96 on the proliferation of CD3⁺ T cells. CFSE-labeled Lewis CD3⁺ T cells were cultured for 5 days in anti-CD3 mAb– coated plates with different concentrations of gp96 in the absence (left panel) and presence (right panel) of costimulation with an anti-CD28 mAb. Proliferation (as indicated by a decrease in fluorescence intensity—relative proliferation) was determined by flow cytometry. Fluorescent histograms are representative of data that were obtained from a minimum of 7 independent experiments. The M1 (negative-staining) and M2 (positive-staining) regions were set with the use of the CFSE-stained unstimulated control. The proportion of cells in the M1 region (ie, exhibited a reduced fluorescence intensity compared with the unstimulated control) is indicated

Fig 5.

Effect of gp96 on the proliferation of CD3⁺ T cells. CFSE-labeled Lewis CD3⁺ T cells were cultured for 5 days in anti-CD3 mAb– coated plates with different concentrations of gp96 in the absence (left panel) and presence (right panel) of costimulation with an anti-CD28 mAb. Proliferation (as indicated by a decrease in fluorescence intensity) was determined by flow cytometry. Data (mean ± SEM) are presented as relative proliferation and were obtained from a minimum of 7 independent experiments

Fig 6.

Effect of gp96 on cytokine secretion by unstimulated (left panel) and costimulated (right panel) CD3⁺ T cells. CD3⁺ Lewis T cells were cultured for 5 days in anti-CD3 mAb–coated plates with different concentrations of gp96 in the absence of costimulation with an anti-CD28 mAb. Cell-free supernatants were harvested, and IFN-γ, IL-4, and IL-10 levels were determined by enzyme immunoassay. Data were obtained from 4 (for unstimulated cells) and 3 (for costimulated cells) independent experiments and are presented as mean ± SEM

Fig 7.

Uptake of FITC-gp96 and FITC-BSA into OX62⁺ BMDCs at 4°C (upper) and 37°C (lower). Propidium iodide (PI)–stained (nonviable) cells were excluded from the analysis. The M1 (negative-staining) and M2 (positive-staining) regions were set against the control (incubated in the absence of FITC-labeled protein). Although there is an apparent binding of FITC-gp96 to BMDCs at 4°C, this binding is nonspecific because FITC-BSA bound to a similar degree. Fluorescent histograms are representative of 2 independent experiments

Fig 8.

(a) Binding of 100 μg of FITC-gp96 and 100 μg of FITC-BSA to freshly isolated CD3⁺ (light shading) and CD4⁺ T cells (dark shading; data from 2 independent experiments). (b) Binding of FITC-gp96 (light shading) and FITC-BSA (dark shading) to cultured CD3⁺ and CD4⁺ T cells (data from 3 independent experiments). T cells were cultured for 5 days in anti-CD3 mAb–coated plates in the absence and presence of costimulation with an anti-CD28 mAb, at which time they were harvested, washed, and fixed. Differential binding of FITC-gp96 to CD4⁺ and CD8⁺ T cells was determined by costaining purified CD3⁺ T cells with PE-conjugated anti-CD4 mAb and analyzing FITC-gp96 and FITC-BSA binding to the CD4⁻ population (data from 2 independent experiments). In all instances, FITC-gp96 and FITC-BSA bindings were determined by flow cytometry, and data are presented as mean ± SEM

Fig 9.

Binding of FITC-gp96 and FITC-BSA to costimulated CD3⁺ T cells. Upper panel: concentration-dependent binding of FITC-gp96 to CD3⁺ T cells. Fixed CD3⁺ T cells were incubated with FITC-gp96 (left) or FITC-BSA (right) for 45 minutes at 4°C and washed twice, and binding was measured by flow cytometry. Concentrations of FITC-gp96 and FITC-BSA are indicated and fluorescence histograms are representative of 3 independent experiments. Lower panel: representative fluorescent micrographs of FITC-gp96 and FITC-BSA binding (both 50 μg) to activated CD3⁺ T cells (magnification 40×)