Synthetic double-stranded RNAs are adjuvants for the induction of T helper 1 and humoral immune responses to human papillomavirus in rhesus macaques

Abstract

Toll-like receptor (TLR) ligands are being considered as adjuvants for the induction of antigen-specific immune responses, as in the design of vaccines. Polyriboinosinic-polyribocytoidylic acid (poly I:C), a synthetic double-stranded RNA (dsRNA), is recognized by TLR3 and other intracellular receptors. Poly ICLC is a poly I:C analogue, which has been stabilized against the serum nucleases that are present in the plasma of primates. Poly I:C(12)U, another analogue, is less toxic but also less stable in vivo than poly I:C, and TLR3 is essential for its recognition. To study the effects of these compounds on the induction of protein-specific immune responses in an animal model relevant to humans, rhesus macaques were immunized subcutaneously (s.c.) with keyhole limpet hemocyanin (KLH) or human papillomavirus (HPV)16 capsomeres with or without dsRNA or a control adjuvant, the TLR9 ligand CpG-C. All dsRNA compounds served as adjuvants for KLH-specific cellular immune responses, with the highest proliferative responses being observed with 2 mg/animal poly ICLC (p = 0.002) or 6 mg/animal poly I:C(12)U (p = 0.001) when compared with immunization with KLH alone. Notably, poly ICLC -- but not CpG-C given at the same dose -- also helped to induce HPV16-specific Th1 immune responses while both adjuvants supported the induction of strong anti-HPV16 L1 antibody responses as determined by ELISA and neutralization assay. In contrast, control animals injected with HPV16 capsomeres alone did not develop substantial HPV16-specific immune responses. Injection of dsRNA led to increased numbers of cells producing the T cell-activating chemokines CXCL9 and CXCL10 as detected by in situ hybridization in draining lymph nodes 18 hours after injections, and to increased serum levels of CXCL10 (p = 0.01). This was paralleled by the reduced production of the homeostatic T cell-attracting chemokine CCL21. Thus, synthetic dsRNAs induce an innate chemokine response and act as adjuvants for virus-specific Th1 and humoral immune responses in nonhuman primates.

Conflict of interest statement

AMS is the CEO of Oncovir, Inc., which provided the poly ICLC for the current study. He was not involved in the acquisition and analysis of the data. All other authors declare they have no competing interests.

Figures

Figure 1. Induction of KLH-specific proliferation by immunization with KLH plus poly ICLC or poly I:C12U.

(A) Live cells in a T cell gate (SSClow cells) were analyzed using anti-CD3 PerCP- and anti-CD4 or anti-CD8 APC-conjugated mAbs. CD3+CD4+ and CD3+CD4−, or CD3+CD8+ and CD3+CD8− cells, respectively, were further analyzed regarding their CFSE expression. The indicated percentages are calculated relative to the number of gated cells. (B) KLH (200 µg) was administered either alone or in combination with 2 mg poly ICLC, 2 mg poly I:C12U, or 6 mg poly I:C12U per animal (five animals per group). CFSE-dilution was used to assess KLH-specific proliferation of CD3+CD4+ T cells before as well as 2 and 6 weeks after immunization. Background proliferation (percentage of CFSElow cells in medium alone) was subtracted from the percentage of CFSElow cells after stimulation with 100 µg/ml KLH for 6 days, and means±SD for each group are presented. * Statistical differences at 2 weeks: 6 mg poly I:C12U versus KLH alone (p = 0.001), 2 mg poly ICLC versus KLH alone (p = 0.002), 2 mg poly I:C12U versus KLH alone (p = 0.16); at 6 weeks: 2 mg poly I:C12U versus KLH alone (p = 0.013).

Figure 2. Induction of HPV-specific Th1 immune responses by injection of HPV16 L1 protein and poly ICLC.

HPV16 capsomeres (10 µg/animal) were injected either alone or in combination with 2 mg poly ICLC or 2 mg CpG-C per animal at weeks 0 and 8 (six animals per group), and immune responses were monitored in triplicates in IFN-γ ELISPOT assays (A), CFSE dilution assays gated on CD4+ T cells (B), and IFN-γ concentrations in supernatants of re-stimulated PBMCs (C), at baseline (pre) and the indicated points in time after the first immunizations. At week 19, only five of the six poly ICLC-injected animals could be analyzed for proliferation and IFN-γ secretion. (A) To determine the numbers of HPV-specific IFN-γ secreting cells, PBMCs were re-stimulated with HPV or control antigen (A447) at 1.25 µg/ml for 20 hours, the ELISPOT plates developed, and the spot numbers counted and extrapolated to 106 PBMCs (HPV16, black bars; A447, light grey bars). Average spot numbers of background responses (to A447) plus twice the standard deviation were considered positive (*). (B) HPV-specific proliferative responses were determined using CFSE-labeled PBMCs expanded for 6 days in the presence of HPV16 or control antigen (A447) at 1.25 µg/ml and re-stimulated with peptide pools 1–4 (B) or 5–8 (data not shown) for the final 6 hours of the assay. Data shown as mean %CFSElow (HPV16) minus background proliferation (%CFSElow with A447)±SEM. * p = 0.008. (C) PBMCs were re-stimulated with HPV16 or control antigen (A447) at 1.25 µg/ml for 2 days, supernatants were collected, and IFN-γ concentrations determined using a monkey IFN-γ specific ELISA assay. Data shown as mean IFN-γ secretion (pg/ml) by cells re-stimulated with HPV16 minus background secretion (by cells incubated in the presence of A447)±SEM. * p = 0.033 at week 10, p = 0.031 at week 12, and p = 0.014 at week 19.

Figure 3. Both poly ICLC and CpG-C act as adjuvants for the induction of HPV16-specific binding and neutralizing antibodies.

HPV16 capsomeres (10 µg/animal) were injected either alone or in combination with 2 mg poly ICLC or 2 mg CpG-C per animal at weeks 0 and 8 (six animals per group), and L1-specific antibodies in plasma samples were measured (A) by ELISA at baseline (pre) and the indicated points in time after the first immunization, and (B) by neutralization assays at week 12. Data given as mean titers +/− SD (A), and as the percentage of neutralization of pseudovirions by serum samples at the dilutions indicated [(B), line, median; *, individual animals)], measured by reduction of SEAP activity (see Materials and Methods) and compared to controls of untreated pseudovirions.

Figure 4. Upregulation of CXCL10 and CXCL9 in draining lymph nodes following injection of poly ICLC.

Expression of CXCL10 (A,B) and CXCL9 (C,D) mRNA (in situ hybridization) and protein (insets) in control lymph nodes before (left panels) and in draining lymph nodes (right panels) 18 hours after KLH-immunization with 0.5 (A,C) or 0.1 mg/kg (B,D) poly ICLC, respectively. The presence of chemokine mRNA is shown by the black dots in the T cell area (100× magnification). Cells positive for chemokine protein expression are visible in red (insets: 400× magnification). Immunohistochemistry was performed on sections consecutive to the one in which in situ hybridization was performed to confirm protein expression. Data are representative for two animals for each dose of poly ICLC.

Figure 5. Downregulation of CCL21 in draining lymph nodes 18 hours after immunization with KLH plus poly ICLC.

(A) Expression of CCL21 mRNA (determined by in situ hybridization) in control lymph nodes before (black dots, left panel) and in draining lymph nodes (right panel) 18 hours after KLH-immunization with 0.5 mg/kg poly ICLC; in situ hybridization, 100× magnification. (B,C) Cells positive for CCL21 protein expression (determined by immunohistochemistry) are visible in brown in the T cell areas of control lymph nodes (left panels) and of draining lymph nodes (right panels) after immunization with KLH plus 0.5 mg/kg (B) or 0.1 mg/kg (C) poly ICLC; 400× magnification.

Figure 6. dsRNA-induced CXCL10 and CXCL9 in serum.

(A) Mean serum levels (±SD) of CXCL10 after immunization of animals with either KLH alone (n = 4), KLH+0.5 mg/kg poly I:C (n = 4 animals), or KLH+0.5 mg/kg poly ICLC (n = 6 animals). (B) The effect of lower doses of poly ICLC on CXCL10 serum levels was assessed in two animals immunized with 0.1 mg/kg. (C) Mean serum levels (±SD) of CXCL10 after immunization of five animals per group with either KLH alone, KLH+6 mg poly IC12U, KLH+2 mg poly IC12U, or KLH+2 mg poly ICLC. (D) Serum levels of CXCL9 after immunization with either KLH alone (n = 2, dotted lines), KLH+0.5 mg/kg poly I:C (full lines, open symbols), or KLH+0.5 mg/kg poly ICLC (full lines, full symbols). Chemokine levels were determined at the indicated points in time; *, p = 0.027; **, p≤0.001.

Figure 7. Monkey DCs secrete CXCL10 upon stimulation with poly ICLC.

Monocyte-derived DCs were generated from rhesus macaque CD14+ monocytes and either stimulated with poly ICLC at the indicated concentrations or left in medium alone. CXCL10 concentrations in the supernatants collected 48 hours later were determined by ELISA. Results of seven independent experiments are given, the horizontal lines indicate the mean concentrations.