Ontogeny and specificities of mucosal and blood human immunodeficiency virus type 1-specific CD8(+) cytotoxic T lymphocytes

Abstract

Induction of adaptive immunity to human immunodeficiency virus type 1 (HIV-1) at the mucosal site of transmission is poorly understood but crucial in devising strategies to control and prevent infection. To gain further understanding of HIV-1-specific T-cell mucosal immunity, we established HIV-1-specific CD8(+) cytotoxic T-lymphocyte (CTL) cell lines and clones from the blood, cervix, rectum, and semen of 12 HIV-1-infected individuals and compared their specificities, cytolytic function, and T-cell receptor (TCR) clonotypes. Blood and mucosal CD8(+) CTL had common HIV-1 epitope specificities and major histocompatibility complex restriction patterns. Moreover, both systemic and mucosal CTL lysed targets with similar efficiency, primarily through the perforin-dependent pathway in in vitro studies. Sequence analysis of the TCRbeta VDJ region revealed in some cases identical HIV-1-specific CTL clones in different compartments in the same HIV-1-infected individual. These results clearly establish that a subset of blood and mucosal HIV-1-specific CTL can have a common origin and can traffic between anatomically distinct compartments. Thus, these effectors can provide immune surveillance at the mucosa, where rapid responses are needed to contain HIV-1 infection.

FIG. 1.

HIV-1 Gag-specific cytolysis and functional avidity of blood and mucosal clones. (A) The optimal epitope of NP014 semen CTL clone is the Gag 9-mer (aa 147 to 155). Autologous B-LCL were pulsed with the 20-mer (aa 141 to 160), 10-mer (aa 147 to 156), and 9-mer (aa 147 to 155) at the peptide concentrations indicated, and specific lysis depicted was performed at an E:T ratio of 5:1. (B) Comparison of cytolysis of semen and blood clones from NP014 at different peptide concentrations of the optimal 9-mer (aa 147 to 155) expressed by autologous B-LCL. (C) Similar comparison of cytolysis by cervical and blood clones from NS0909 at an E:T ratio of 5:1.

FIG. 2.

Selective inhibition of HIV-1-specific cytolytic activity by CMA but not anti-FasL or anti-TNF-α antagonistic antibodies. Blood and mucosal CD8⁺ clones were pretreated with CMA (1 to 100 nM), anti-FasL MAb (12 to 50 μg/ml), anti-TNF-α MAb (5 to 20 μg/ml), or control mouse immunoglobulin G for 2 h. In some experiments, a combination of two or three blocking agents at the highest (experiment 1) or lowest (experiment 2) concentrations was used. The figure depicts cytolysis of CD8⁺ clones isolated from patients CC0710 (blood and cervix), NP002 (rectum), and NP014 (semen). Blood and cervical clones from patient CC0710 recognized HIV-1 Gag within aa 264 to 272. Blood (data not shown) and semen HIV-specific CTL clones from patient NP014 recognized HIV-1 Gag within aa 164 to 172. Blood (data not shown) and rectal T-cell clones from patient NP002 recognized HIV-1 Gag within aa 269 to 277. Ab, antibody.

FIG. 3.

Blood and mucosal Gag-specific CTL clones recognize the same epitope and display the same TCR β VDJ rearrangement. (A to C) Blood, rectal, and seminal CTL clones, established from volunteer NP002 from the same visit. (A) CTL clones derived from blood (B22), semen (S18), and rectum (R9) recognized the same HIV-1 Gag epitope within aa 269 to 277. (B) The four CTL clones (blood-derived B18 and B22, semen-derived S18, and rectum-derived R9) utilize TCR Vβ22 and present identical rearrangements of their TCR β VDJ segments. The four left lanes contain PCR products from the clonal cDNA amplified with pooled Vβ21 to Vβ25 primers. The four right lanes contain the PCR products from the clonal cDNA amplified with the Vβ22-specific primers. The coding and amino acid sequence of the five clones (including B21) TCR BVDJ region is presented to the right. (C) TCR Vβ22 usage was confirmed by flow cytometry. Expression of Vβ22 fluorescein isothiocyanate-conjugated MAb by the CTL clone B22 (solid peak) is shown in comparison to the isotype control (open peak). (D) PBMC from volunteer NS0909 were probed for the presence of TCR Vβ6 and C20 clone-specific VDJ segment. A positive response was detected with Vβ6 specific primers (lane 2) and C20 clone-specific primers (lane 3). Negative and positive control PCR assays were performed in the absence of Vβ6-specific primers (lane 1) or with the cDNA product from cervical clone C20 (lane 5).