Thymic stromal lymphopoietin (TSLP) acts as a potent mucosal adjuvant for HIV-1 gp140 vaccination in mice

Abstract

The development of a successful vaccine against HIV is likely to require the induction of strong and long-lasting humoral immune responses at the mucosal portal of virus entry. Hence, the design of a vaccine strategy able to induce mucosal antibodies and in particular specific IgA, may be crucial to providing immune protection. Nasal immunisation is known to induce specific IgG and IgA responses in the cervicovaginal mucosa; however, there is an urgent need for the development of safe, effective and accessible mucosal adjuvants for nasal application in humans. To reduce the potential for adverse events associated with some nasal adjuvants, we have assessed whether the B-cell-activating cytokines APRIL, BAFF and TSLP enhance humoral immune responses to HIV-1 gp140. Following intranasal immunisation, TSLP but not APRIL or BAFF induced strong humoral responses both in serum and mucosa. The adjuvant effect of TSLP on humoral responses was similar to that of cholera toxin (CT). The use of TSLP as an adjuvant skewed both the cellular and humoral immune responses towards Th2 cells. This is the first time that TSLP has been demonstrated to have a positive effect as a mucosal adjuvant, and specifically to promote mucosal and systemic responses to HIV gp140.

Figure 1

TSLP induces gp140-specific immune responses after intranasal immunisation. Mice were immunised with 10 μg gp140 in the presence or absence of 5 μg TSLP, APRIL or BAFF i.n. three times at 3 week intervals. Anti-gp140 IgA (left) and IgG (right) were measured 3 weeks after the final immunisation by ELISA in (A) serum and (B) vaginal lavage. Data are shown as endpoint titres, with lines representing geometric means of n=5 mice; data are representative of two experiments. ^*p<0.05, ^**p<0.01, ^***p<0.001, one-way analysis of variance (ANOVA) with Tukey's multiple comparison post test.

Figure 2

TSLP induces similar humoral but not cellular responses to CT. Mice were immunised three times i.n. with 10 μg gp140 alone (+),with 5 μg TSLP (△) or CT (◊). Anti-gp140 IgA and IgG were measured by ELISA in (A, B) serum, (C, D) vaginal lavage and (E, F) faecal extracts obtained from mice 3 weeks after the final immunisation. (G) Anti-gp140 IgA and IgG ASC were measured by B-cell ELISPOT in spleen from mice 3 wks after the final immunisation. (H) 10 days after the final immunisation, splenocytes were isolated and cultured for 5 days with CN54 gp140 protein and proliferation measured by ³H thymidine incorporation. Data points represent single mice, bars represent mean of n=5 mice+SD (G, H). Data are representative of two experiments. ^*p<0.05, ^**p<0.01, ^***p<0.001, one-way analysis of variance (ANOVA) with Tukey's multiple comparison post test.

Figure 3

TSLP shifts the immune response to gp140 to Th2 Mice were immunised three times i.n. with 10 μg gp140 alone or with 5 μg TSLP or CT. (A) Anti-gp140 IgG₁ and IgG_2a and (B) IgE were measured in serum by ELISA. (C) Splenocytes were stimulated with a CN54 gp140 peptide pool and specific cytokine production was measured in CD4⁺ T cells by flow cytometry. (D) Mice were immunised twice at 2 week intervals, followed by three daily i.n. challenges with 20 μg gp140 in 100 μL and were culled 3 days later. (E) gp140-specific IgE was measured in sera. (F) Cells were collected from bronchoalveolar lavage (BAL) and counted and (G) differential cell counts performed. Representative PAS-stained sections of lungs from (H) PBS, (I) gp140, (J) gp140:TSLP, (K) gp140:CT, (L) IP gp140:alum. Data are shown as mean of n=5 mice+SD, experiment performed once. ^*p<0.05, ^**p<0.01, ^***p<0.001, one-way analysis of variance (ANOVA) with Tukey's multiple comparison post test.

Figure 4

I.n. immunisation induces higher gp140-specific IgA responses than i.d immunisation. Mice were immunised i.n. (open symbols) or i.d. (black symbols) three times with 10 μg gp140 alone or with 5 μg TSLP or CT. (A, C) Anti-gp140 IgG and (B, D) anti-gp140 IgA were measured by ELISA (A, B) in serum 3 weeks after priming (first immunisation), first boost (second immunisation) and final immunisation (third immunisation) and (C, D) in vaginal lavage, faecal extracts and nasal lavage 3 weeks after the final immunisation. Data are shown as endpoint titres, with lines representing geometric means of n=5 mice; data are representative of two experiments. ^***p<0.001, one-way analysis of variance (ANOVA) with Tukey's multiple comparison post test.

Figure 5

TSLP induces sustained humoral and cellular immune responses to gp140. Mice were immunised four times i.n. with 10 μg gp140 alone (▪) or with 5 μg TSLP (○) or CT (◊). The fourth immunisation was given 6 months after the third immunisation. Anti-gp140 IgA and IgG were measured by ELISA in (A, B) serum and (C, D) vaginal lavage before every immunisation and 3 weeks after the final immunisation. (E) CN54 gp140 IgA and IgG ASC were measured by B-cell ELISPOT in spleens from mice 3 weeks after the final immunisation. Antigen-specific T-cell proliferation of splenocytes was assessed after in vitro stimulation with gp140 after 120 h using (F) ³H-thymidine and (G) CFSE dilution. Data are shown as mean of n=5 mice+SD, experiment was performed once. ^*p<0.05, ^**p<0.01, ^***p<0.001, one-way analysis of variance (ANOVA) with Tukey's multiple comparison post test.