Class I–unrestricted noncytotoxic anti–HTLV-I activity of CD8⁺ T cells

Human T-lymphotropic virus type I (HTLV-I) causes 2 distinct diseases, a T-cell malignancy designated adult T-cell leukemia/lymphoma (ATLL) and a chronic inflammatory nervous system disease designated HTLV-I-associated myelopathy/tropical spastic paraparesis (HAM/TSP).1 However, the latency between viral transmission and disease progression is often very long, typically several years to decades; furthermore, the majority (95% or more) of infected individuals are healthy carriers.1 What determines the outcome of HTLV-I infection has not been fully understood; however, CD8⁺ T-cell response to HTLV-I infection probably plays a role in disease progression. Although it has been widely considered that CD8⁺ T cells mediate antiviral activity principally by cytolytic mechanisms, noncytolytic antiviral response of CD8⁺ T cells has been demonstrated in controlling infection with human immunodeficiency virus (HIV),2 simian immunodeficiency virus (SIV),2feline immunodeficiency virus (FIV),3 or hepatitis B virus.4 In this study, we investigated whether MHC class I-unrestricted noncytolytic mechanism is also used by CD8⁺T cells to control HTLV-I infection.

Peripheral blood mononuclear cells (PBMCs) were isolated from HTLV-I seropositive or seronegative healthy individuals (Nagasaki Red Cross Blood Center, Nagasaki, Japan). CD8⁺ T cells were positively selected, as described previously.5CD8⁺ T-cell-depleted PBMCs (CD8⁻ PBMCs) contained less than 1.5% CD8⁺ cells, as determined by flow cytometric analysis (data not shown). In some experiments CD8⁻ PBMCs and autologous or allogeneic CD8⁺ T cells were cultured together or separated by a semipermeable membrane filter with 0.4-μmol/L pores (Costar, Acton, MA).

Replication of HTLV-I was determined by HTLV-I p19 Ag levels in cell-free supernatants from HTLV-I carriers' PBMCs, using a commercially available enzyme-linked immunosorbent assays (ELISA; Cellular Product, Inc, Buffalo, NY).

It has been shown that HIV-1 is more readily isolated from PBMCs of infected individuals when CD8⁺ T cells are depleted.6 CD8⁺ T-cell–mediated antiviral activity has also been demonstrated in controlling other lentiviruses such as SIV and FIV.2,3 Furthermore, CD8⁺depletion in vivo dramatically increased plasma viral loads in SIV-infected macaques.7 These studies suggest that CD8⁺ T cells play a critical role in the pathogenesis of retroviral infections.

To explore the possibility that CD8⁺ T cells play a role in preventing replication of HTLV-I, another human retrovirus, CD8⁺ T cells were depleted from PBMCs of healthy HTLV-I carriers. Depletion of CD8⁺ T cells from PBMCs resulted in a significant increase in HTLV-I p19 Ag expression, and reconstitution of autologous CD8⁺ T cells suppressed viral expression in a dose-dependent manner (Table 1).

Next, we investigated whether CD8⁺ T-cell–mediated anti–HTLV-I activity resulted principally from HTLV-I-specific class I-restricted cytotoxic T-lymphocyte response (CTL). To do so, we first performed coculture experiments in which CD8⁻ PBMCs of HTLV-I carriers were reconstituted with autologous or allogeneic CD8⁺ T cells, and p19 Ag levels in cell-free supernatants were compared. As shown in Table 1, autologous CD8⁺ T cells markedly, and allogeneic CD8⁺ T cells modestly suppressed HTLV-I replication. Allogeneic CD8⁺ T cells derived from HTLV-I seronegative healthy individuals had little activity against HTLV-I (data not shown); however, it was not possible in this study to determine whether Ag-specific stimulation is essential for CD8⁺ T-cell anti–HTLV-I activity. These results suggest that, although CTL response plays a critical role in controlling HTLV-I replication, there exists another antiviral activity that does not require MHC class I compatibility.

Because it is possible that anti–HTLV-I activity of allogeneic CD8⁺ T cells was attributed to CTL response mediated by partially matched class I molecules, CD8⁺ T cells were separated from CD8⁻ PBMCs by a semipermeable membrane filter (trans-well cultures). Although less efficient than CD8⁺ T cells in cocultures, these cells could suppress HTLV-I replication even when direct cell-to-cell contact, which is essential for class I-restricted CTL response, was prohibited (Table2). Furthermore, the percentage viabilities of the control cells and the cells in trans-well cultures were similar (88% ± 8% and 90% ± 9%, respectively), and HTLV-I proviral DNA was detected in trans-well cultures at levels comparable to CD8⁻ PBMC cultures that yielded substantial amounts of p19 Ag (data not shown). These results suggest that class I-unrestricted noncytolytic response of CD8⁺ T cells may play a role in controlling HTLV-I infection.

A previous study demonstrated the inverse relationship between numbers of peripheral CD8⁺ T cells and levels of gamma interferon production and HTLV-I expression.8 Furthermore, it has also been reported that persons with asymptomatic HTLV-I infection have heightened immune reactivity, whereas those with ATLL do not,9 suggesting that there may be immune regulation of HTLV-I infection. Recently, HTLV-I–specific CTLs have been found in healthy carriers or patients with HAM/TSP,10 and it has been described that HTLV-I is not latent and is actively replicating in infected individuals, and that the chief determinants of the equilibrium viral load of HTLV-I are virus-specific CTL response.11 In this study, we identified anti–HTLV-I activity other than CTL response. The significance of the non-CTL response in vivo warrants further investigation.