MPs play pivotal roles in antigen capture and presentation, pathogen and tissue debris clearance, and cellular secretory functions. However, activated MPs can infiltrate through the blood brain barrier and contribute to the CNS inflammation by secreting various inflammatory cytokines and growth-inhibiting proteins. In HAM/TSP, MPs are reservoirs of HTLV-I, induce proinflammatory cytokines and excessive antigen-specific T cell responses, and can also infiltrate the CNS. In our study, we analyzed CD14+ cell subpopulation in PBMCs of patients with HAM/TSP and demonstrated that CD14lowCD16+ subset of patients with HAM/TSP showed significantly higher CX3CR1 and HLA-DR expression, compared to NDs and ACs. Since it has been reported that CX3CR1 expression is regulated by IL-2 and IL-15 , activated T cells expressing these cytokines might affect CX3CR1 expression on monocytes in patients with HAM/TSP [19, 38, 39]. In mice, GR1-CX3CR1high monocytes (homolog of human CD16+ monocytes) patrol vascular endothelium by mechanisms involving LFA-1 and CX3CR1 and are rapidly recruited into inflamed tissues, such as spleen, gut, lung and brain, where they differentiate into macrophage [23, 40]. In humans, CD16+ monocytes that have the potential to migrate preferentially in response to fractalkine, a ligand of CX3CR1, have more Fc receptor mediated phagocytosis function and are at a more advanced stage of differentiation to macrophage and dendritic cell [41–43]. These findings suggest that CD14lowCD16+ and CD14+CD16- cells are recruited into different anatomic sites under constitutive or inflammatory conditions and play distinct functional roles in immunity and disease pathogenesis. Fractalkine is expressed on activated endothelial cells , neuron , apoptotic cells , and brain with inflammation . Therefore, HTLV-I-activated or infected cells might induce fractalkine expression at the site of inflammation such as the spinal cord to recruit and adhere CX3CR1+ cells. The hypothesis was supported by the accumulation of CX3CR1+ cells immunohistochemically detected in the meninges and parenchyma of HAM/TSP spinal cords as well as around blood vessels (Figure 1D). The CX3CR1+ cells were CD68+ and also morphologically consistent with MPs. Therefore, these results suggested that CX3CR1+ MPs could accumulate in spinal cords of patients with HAM/TSP. Moreover, the increase of degranulation and IFN-γ expression in CD8+ T cells were significantly correlated with the increase of CX3CR1 and HLA-DR expression in CD14lowCD16+ subset of HTLV-I-infected patients. These results support the hypothesis that strong correlation between CD8+ T cell activation and MP activation contribute to the pathogenesis of HAM/TSP. These differential changes in peripheral MP subpopulations in vivo may also be associated with the infiltration of MPs into the CNS and CD8+ T cell activation in patients with neurologic inflammatory disease.
MP activation in patients with HAM/TSP was also suggested by TNF-α and IL-1β expression in CD14+ cells. Expression of IL-1β and TNF-α was detected in perivascular infiltrating macrophages and microglia in the spinal cords of patients with HAM/TSP and in infiltrating macrophage in the muscle of patients with HTLV-I-related myositis [27, 48]. Thus, the proinflammatory cytokine expression in peripheral MPs might be related to the infiltration of MPs into the inflammatory site of patients with HTLV-I-related diseases. Moreover, CD14+ cells accelerated HTLV-I Tax expression of autologous CD4+CD25+ T cells in patients with HAM/TSP, which was dependent on cell-cell contact. In patients with HAM/TSP, high HTLV-I Tax expression is mainly detected in CD4+ T cells after ex vivo culture, but dendritic cells and CD14+ cells can also express HTLV-I Tax, consistent with the observation that HTLV-I infects dendritic cells to effectively transfer cell-free virus to CD4+ T cells [18, 19]. In HIV, human CD16+ monocytes have been shown to be more susceptible to infection than CD16- monocytes, to preferentially harbor the virus over the long-term, and to promote high levels of HIV replication upon differentiation into macrophages and interaction with activated T cells [30, 49]. Therefore, HTLV-I infected and activated MP might likewise contribute to T cell activation and virus dissemination in HTLV-I associated disease.
Minocycline is a well known as inhibitor of MP activation and has been reported to have beneficial effects on inflammation, microglial activation, matrix metalloproteinases, nitric oxide production, and apoptotic cell death . Furthermore, minocycline has been suggested to have neuroprotective effects in human as well as in animal models of a number of neurologic diseases including stroke, multiple sclerosis, and Parkinson's disease . In our study, minocycline treatment significantly inhibited proinflammatory cytokine expression (TNF-α and IL-1β) in CD14+ cells of patients with HAM/TSP, while TNF-α expressions in CD4+ T cells of patients with HAM/TSP did not change. These results suggest that the effects of minocycline may act through inhibition of MP activation rather than HTLV-associated T cell activation. Unexpectedly, minocycline treatment also effectively inhibited spontaneous lymphoproliferation and IFN-γ expression of CD8+ T cells, which are well-described observations of T cell activation in patients with HAM/TSP. While these T cell responses have been reported to be due to IL-2/IL-2 receptor and IL-15/IL-15 receptor autocrine loop following expression of HTLV-I Tax in T cells [32, 38], a number of studies have demonstrated that non-T cells and CD14+ cells can also play a stimulatory role in HTLV-I-associated T cell activation [5, 19, 38]. Therefore, our results support the view that T cell responses in patients with HAM/TSP are due, in part, to the activation of MPs.
Inhibition of MPs resulted in the suppression of CD8+ T cell dysregulation (degranulation and IFN-γ expression). Elevated IFN-γ expression is an important immunological marker in the pathogenesis of HAM/TSP , and CD8+ T cell dysregulation was mediated by various factors, including virus infection, enhanced IL-2/IL-15, and expression of cellular molecules [19, 51–54]. Unexpectedly, minocycline inhibited spontaneous degranulation/IFN-γ expression in CD8+ T cells of HAM/TSP patients as well as HTLV-I Tax11-19-specific CD8+ T cell responses. Antiviral CD8+ T cells can elaborate at least two effector functions, cytotoxicity and inflammatory cytokine production, which are determined primarily by antigen concentration . Interestingly, minocycline treatment suppressed inflammatory IFN-γ production, but not total cytotoxicity (CD107a expression) in Tax-specific CD8+ T cells of patients with HAM/TSP. Moreover, after the treatment with minocycline, MHC class I expression on CD14+ cells of patients with HAM/TSP was gradually suppressed in cultured cells, compared to untreated MPs. These results suggested that the activation of CD8+ T cells was inhibited through MHC class I downregulation on CD14+ cells after minocycline treatment. This may be one mechanism involved in the reduction of CD8+ T cell inflammatory IFN-γ production in the presence of minocycline. Moreover, minocycline significantly inhibited spontaneous degranulation/IFN-γ expression in CD8+ T cells of HAM/TSP patients. As previously reported, the spontaneous degranulation/IFN-γ expression in CD8+ T cells of HAM/TSP patients was mediated by various factor(s) [19, 52]. To evaluate regulatory effects of CD8+ T cell by minocycline, further analysis would be needed. In addition, even though minocycline down-modulates the capacity of antigen-presenting cells to trigger CD8+ T cell effector responses, the cytotoxic function of Tax-specific CD8+ T cells might be still maintained and continue to provide control of virus-infected cells. This may have a positive clinical consequence for use of minocycline in treatment of HTLV-I-associated disease.