Brief treatment with iNKT cell ligand α-galactosylceramide confers a long-term protection against lupus

Abstract

CD1d presents glycolipid antigens such as α-galactosylceramide (αGalCer) to invariant natural killer T cells (iNKT). We have reported that activated iNKTs inhibit IL-10-producing autoreactive B cells, while promoting or leaving intact the normal B cell responses, making iNKT modulation an attractive therapeutic modality. Here, we report that a brief treatment of young lupus-prone (NZB/NZW)F1 (BWF1) mice with two injections of αGalCer conferred a long-term protection against lupus. Long-term repeated administrations of αGalCer, however, afforded no clinical benefit. These disparate clinical effects correlated with iNKT responsiveness. While a brief treatment with αGalCer enhanced iNKT responses upon in vitro recall, the long-term αGalCer treatment resulted in reduced iNKT responses in BWF1 mice. The improvement in disease with αGalCer treatment was associated with the reduced IL-10 production. Furthermore, iNKTs directly inhibited IL-10-secreting cells in vivo in reconstituted SCID mice and inhibited IL-10-secreting B cells in vitro in co-cultures. Thus, a brief treatment with a CD1d-binding glycolipid enhances iNKT responses, reduces IL-10 production, and delays the onset of lupus, whereas long-term repeated treatments induce marked iNKT hyporesponsiveness and do not affect disease outcome in BWF1 mice. Identifying glycolipid regimens that can modulate iNKT responsiveness will have important implications for developing iNKT-based therapies for autoimmune diseases.

Fig. 1

A brief treatment with αGalCer confers long-term protection from lupus in BWF1 mice. Seven-week-old BWF1 mice were treated with αGalCer (4 µg i.p.) or vehicle twice at a 3-day interval. a Results are shown as the percentage of mice with severe proteinuria (≥300 mg/dl) on two consecutive assessments. **P≤0.01 (log-rank test; n=21 mice per group pooled from two separate experiments). B The animals were bled and their sera tested for IgG anti-dsDNA Ab (*P≤0.05; mean±SE; n=12 mice per group)

Fig. 2

Long-term repeated administration of αGalCer does not protect from lupus in BWF1 mice. Five-week-old mice at a preclinical stage, i.e., without circulating anti-dsDNA Abs or proteinuria (n=22 mice per group; a, d); 18-week-old mice with early clinical disease, i.e., with IgG anti-dsDNA Abs but with mild or no proteinuria (n=21 mice per group; b, e); and 26-week-old mice with advanced disease, i.e., with IgG anti-DNA Abs and moderate proteinuria (n=12 mice per group; c, f) were treated with αGalCer or vehicle twice a week for 11–18 weeks. Results are shown as the percentages of mice with severe proteinuria (≥300 mg/dl) on two consecutive assessments (a–c) and as the mean±SE units per milliliter serum IgG anti-dsDNA Ab (d–f)

Fig. 3

Effect of short-term administration of αGalCer on iNKT cell responses in BWF1 mice. a Cytokine responses. Seven-week-old BWF1 mice were injected i.v. once with 4 µg αGalCer or vehicle alone (five mice per group). Spleen cells from these mice were cultured without or with 10–100 ng/ml αGalCer for 2 days, and culture supernatants assayed for cytokines by ELISA. Results are shown as the mean±SE nanogram per milliliter. *P≤0.01; #P≤0.05 compared to vehicle control mice. Similar data were obtained in another five each of 14-week-old BWF1 mice (data not shown). b Proliferative responses. Ten-week-old BWF1 mice were injected i.v. with 4 µg αGalCer or vehicle. Spleens were harvested 2 h later, and spleen cells cultured with medium alone or 50 ng/ml αGalCer for 48 h, with ³H thymidine in the last 16 h of culture. Results are shown as the mean±SE (n=3 per group). *P<0.01 compared to vehicle-injected mice

Fig. 4

Effect of long-term administration of αGalCer on iNKT cell responses in BWF1 mice. Nine-week-old female BWF1 mice were treated with αGalCer (4 µg i.p. twice a week) or control vehicle for 3 months. Three months later, mice were sacrificed and their spleen cells cultured with medium alone, αGalCer (50 ng/ml), Con A (5 µg/ml), or plate-bound anti-CD3 (5 µg/ml) for 2 days. Supernatants were assayed for cytokines, which are shown as the mean+SE. *P<0.05; n=4 mice per group

Fig. 5

Effect of αGalCer treatment on iNKT cell numbers in BWF1 mice. Seven- to 9-week-old BWF1 mice were treated with αGalCer (4 µg i.p. twice a week) or equal volume of vehicle for a short-term (1 week, as described in Fig. 1a) (a) or for a long-term (12 weeks, as described in Fig. 2) (b). At 8–9 months of age, animals were euthanized and their spleen cells analyzed for CD1d/αGalCer tetramer⁺ TCRβ⁺ cells (n=6–9 mice per group). Splenic iNKT cell numbers were not statistically significantly different between αGalCer- and vehicle-treated animals

Fig. 6

Effect of αGalCer treatment on IL-10 production. a Spleen cells were harvested from BWF1 mice ≥6 months after treatment with αGalCer or vehicle (as in Fig. 1) and cultured with LPS without or with αGalCer. Supernatants were assayed for IL-10 by ELISA (*P<0.05; n=10 mice per group). b BALB/c SCID mice were reconstituted with B cells and T cells (CD1d–αGalCer dimer⁺ cells from Vα14^Tg mice or Jα18^−/− B cell-depleted spleen cells), as described in “Materials and Methods”. Spleen cells from the reconstituted mice were analyzed for IL-10-secreting cells using a cytokine secretion assay. Co-staining with CD11b was performed to distinguish between adoptively transferred cells (CD11b⁻) and SCID mouse spleen cells (CD11b⁺). Numbers of IL-10-secreting cells per million spleen cells are shown. Results represent two independent experiments. c T cells from Vα14^Tg mice and B cells from Jα18^−/− mice were cultured overnight with LPS or CpG and vehicle or αGalCer. IL-10-producing B cells, as detected by cytokine secretion assay, are shown as percentages (%) and total cells per million in the culture. Results represent two independent experiments