Generation of Th1 T cell responses directed to a HLA Class II restricted epitope from the Aspergillus f16 allergen

Abstract

The Aspergillus allergen Asp f16 has been shown to confer protective Th1 T cell-mediated immunity against infection with Aspergillus conidia in murine models. Here, we use overlapping (11-aa overlap with preceding peptide) pentadecapeptides spanning the entire 427-aa coding region of Asp f16 presented on autologous dendritic cells (DC) to evaluate the ability of this antigen to induce Th1 responses in humans. Proliferative responses were induced in five out of five donors, and one line with a high frequency of interferon (IFN)-gamma-producing CD4(+) T cells in response to the complete peptide pool was characterized. This line was cytotoxic to autologous pool-pulsed and Aspergillus culture extract-pulsed targets. Limitation of cytotoxicity to the CD4(+) T cell subset was demonstrated by co-expression of the degranulation marker CD107a in response to peptide pool-pulsed targets. Cytotoxic T lymphocytes (CTL) killed Aspergillus hyphae and CTL culture supernatant killed Aspergillus conidia. By screening 21 smaller pools and individual peptides shared by positive pools we identified a single candidate sequence of TWSIDGAVVRT that elicited responses equal to the complete pool. The defined epitope was presented by human leucocyte antigen (HLA)-DRB1-0301. These data identify the first known Aspergillus-specific T cell epitope and support the use of Asp f16 in clinical immunotherapy protocols to prime protective immune responses to prevent or treat Aspergillus infection in immunocompromised patients.

Fig. 1

Proliferative response to Asp f16 protein-derived overlapping pentadecapeptides. (a) PBMC from five different donors were stimulated with soluble Asp f16-PPC (not presented on autologous DC) in the presence or absence of IL-2 added at day 3 of a 6-day culture. Data are presented as the counts per minute (cpm) [³H]-thymidine uptake and as the stimulation index (SI) of triplicate cultures from all five doors compared to PBMC alone. (b) Proliferation in response to autologous PPC-pulsed fast-DC (PPC-DC). Decreasing numbers of 25 Gy-irradiated, PPC-pulsed or non-pulsed (NP-DC), mature fast-DC from the same five donors were cultured with autologous lymphocytes for 6 days. Data are presented as the mean cpm and as the SI index of triplicate cultures from all five donors at the indicated DC : T cell ratios. Asp f16-PPC: complete pool of Asp f16 protein-derived overlapping pentadecapeptides.

Fig. 2

Lytic activity of the Asp f16-specific T cell line. Asp f16-effectors were generated by three stimulations with autologous, mature, Asp f16-PPC-pulsed fast-DC. Cytolytic activity of the culture was assessed 5 days thereafter by chromium release using autologous non-, Asp f16-PPC- and CF-pulsed DC (a) or non- and Asp f16-PPC-pulsed BLCL (b) as targets. Data are presented as mean value of triplicate wells ± s.d. There was no significant difference in lytic activity towards CF- and PPC-pulsed targets (P > 0·05, one-way analysis of variance). NP: non-pulsed; PPC: complete overlapping pentadecapeptides pool; CF: commercial Aspergillus culture-filtrate antigen.

Fig. 3

Conidiacidal activity of Asp f16-specific T cell line bulk-culture supernatant. Live and heat-inactivated conidia were labelled with a fluorescent molecular stain (FUN-1). Asp f16-effectors were restimulated 48 h prior to the assay. FUN-1-stained live conidia (10⁶) were incubated overnight in fresh RPMI-1640 culture medium or in CTL culture medium. Wells containing live and heat-inactivated FUN-1-labelled conidia alone were used as positive and negative controls, respectively. Metabolically active conidia accumulate orange fluorescence in vacuoles, while dormant and dead conidia stain green. FUN-1-stained live conidia (a), FUN-1-stained dead conidia (b), FUN-1-stained live conidia in fresh RPMI-1640 culture medium (c) and FUN-1-stained live conidia in Asp f16-CTL culture medium (d).

Fig. 4

Screening with small peptide pools. The Asp f16-specific T cell line was screened using 21 smaller pools containing 4–11 peptides in ⁵¹Cr release assays (a, b, c) and in intracellular IFN-γ production assays (d). Reactivity was seen to only three smaller pools. Data are presented as mean values ± s.d. of triplicate assays. Asp f16-effectors were sensitized only and to the same extent to PP3, PP4, and PP15 (P > 0·05, one-way analysis of variance). Thus, the recognized Asp f16 epitope should be in SP43 and SP44. PP: peptide pool; SP: single peptide.

Fig. 5

Asp f16 epitope. Asp f16 bulk-culture showed lytic activity (a) and IFN-γ production (b) in response to single peptides sharing the epitope. Data are presented as mean values ± s.d. The effectors were not sensitive to SP42 and SP45 indicating that the epitope is contained in SP43 and SP44. There was no significant difference (P > 0·05, one-way analysis of variance) in lytic activity and IFN-γ production in response to either PPC, SP43 or SP44. The shared amino acid sequence between SP43 and SP44 is TWSIDGAVVRT (c). PP: peptide pool; SP: single peptide.

Fig. 6

Frequency of T cells in PBMC from normal DRB1-0301⁺ donors reactive to peptide epitope WSIDGAVVR. Cryopreserved PBMC from six HLA-DRB1-0301⁺ normal blood donors in addition to donor RD0308 were stimulated with peptide SP43 containing WSIDGAVVR and with the complete Asp f19 peptide pool and were screened for IFN-γ production by ELISPOT. Response to SEB was used as a positive control and PBMC in the absence of peptide were used as the background control. Data are expressed as the number of IFN-γ-producing cells per 10⁵ cells plated.