CD26+CD39−CD73+CD4+ T lymphocytes decrease after HIV-1 infection but show an intrinsic association with low viral replication and high CD4+ T-cell count

To the Editor: Cluster of differentiation 39 (CD39) is an extracellular ATPase that cleaves adenosine triphosphate (ATP) to adenosine diphosphate and adenosine monophosphate. By doing so, it depletes extracellular ATP that has been generated by the activated T cells or released by other cells in the stressed environment, such as human immunodeficiency virus (HIV) invasion. CD39-generated adenosine monophosphate is further cleaved to adenosine by CD73, the other main ectoenzyme in purinergic signaling. CD39 and CD73 function as metabolic enzymes to produce adenosine, thereby exerting inhibitory effects in tumors1,2; by contrast, CD26 facilitates adenosine conversion to inosine and plays complex biological roles in diseases.3,4 However, the nonenzymatic immunological roles of CD4 lymphocytes coexpressing these three molecules after the onset of HIV-1 infection remain unclear. Here, 42 individuals with acute HIV-1 infection (AHI), 44 chronic HIV-1 infection (CHI) individuals, and 25 HIV-1-negative healthy controls (HCs) were enrolled to investigate the roles of CD4 lymphocytes expressing CD26, CD39, and CD73 at different HIV-1 infection stages. Flow cytometry analyzed the immune profiles and functions of these CD4 lymphocytes, CD26 and CD73 expression on CD4 lymphocytes was significantly decreased in both AHI and CHI groups, whereas CD39 expression showed no statistically significant variation in these infected groups Supplementary Figure 1A, B, https://links.lww.com/CM9/C901. Among the eight CD4 lymphocyte subsets the proportion of CD26+CD39−CD73+CD4+ cells, the third most abundant subset, was significantly decreased in AHI and CHI individuals, and CD26–CD39–CD73+CD4+ cells also showed a reduction among three groups Supplementary Figure 1C, D, https://links.lww.com/CM9/C901. During HIV-1 infection, programmed cell death protein 1 (PD-1) and T cell immunoreceptor with Ig and ITIM domains (TIGIT) expression were lower on CD26+CD4+ T cells and CD73+CD4+ T cells than on their CD26– and CD73– counterparts, respectively Supplementary Figure 1E and F, https://links.lww.com/CM9/C901. Consistently, CD26+CD39–CD73+CD4+ lymphocytes exhibited the lowest surface PD-1 and TIGIT expression among all T-cell subpopulations, followed by CD26+CD39−CD73−CD4+ lymphocytes. Notably, CD26−CD39−CD73+CD4+ cells displayed higher expression levels of PD-1 and TIGIT compared with the latter Supplementary Figure 1G, https://links.lww.com/CM9/C901. Strikingly, in AHI and CHI, PD-1 expression was downregulated, while TIGIT expression remained unchanged, on both the CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T-cell subsets. These results indicate that CD26+CD39−CD73+CD4+ T cells display a reduced degree of immune exhaustion in individuals infected with HIV-1. During HIV-1 infection, activated CD26+CD4+ T cells were present at higher levels than their CD26− counterparts. By contrast, in both HIV-1-infected individuals and healthy controls, CD73+CD4+ T cells showed lower activation levels than CD73− counterparts. However, both CD26+CD4+ and CD73+CD4+ T cells exhibited lower proliferative capacity than their respective CD26−CD4+ or CD73–CD4+ counterparts Supplementary Figure 2A, https://links.lww.com/CM9/C901. Across the three study groups, the frequencies of activated CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T cells were comparable to those of the other four CD4+ T-cell subsets, while their activation levels were indeed elevated in HIV-1 infection. On the contrary, among the multiple CD4+ T-cell subpopulations, the CD26+CD39−CD73+CD4+ subset displayed the lowest proliferative activity, followed by CD26−CD39−CD73+CD4+ subset across all three study groups, with their proliferation capacities failing to be elevated following HIV-1 infection. In contrast, the CD26−CD39+CD73−CD4+ and CD26+CD39+CD73−CD4+ T cells displayed the strongest proliferative potential. Senescent cells (characterized by CD28−CD57+) were few in the CD26+CD39−CD73+CD4+ subset, whereas those in the CD26−CD39−CD73–CD4+ T-cell subset were more abundant Supplementary Figure 2B, https://links.lww.com/CM9/C901. Given that CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T cells express lower levels of the exhaustion markers PD-1 and TIGIT, we further investigated whether these lymphocytes possess greater capacity to produce degranulation substance and cytokines. Across three study groups, the capacity of CD26+CD39−CD73+CD4+ T cells to release CD107a was lowest among CD4+ T-cell subpopulations, and this capacity did not increase in AHI or CHI relative to HC. The interferon-γ (IFN-γ) secretion ability of this subset was comparable to that of other subsets except CD26+CD39+CD73–CD4+ T cells, with no upregulation observed in AHI or CHI relative to HC Supplementary Figure 3A, https://links.lww.com/CM9/C901. Additionally, CD26−CD39−CD73+CD4+ T cells exhibited similarly reduced capacities for CD107a and IFN-γ production. However, both CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T cells showed interleukin-2 (IL-2) and IL-17a secretion abilities comparable with those of other subsets except CD26+CD39+CD73−CD4+ T cells. Notably, compared to HC, the IL-17a production level of CD26−CD39−CD73+CD4+ T cells was significantly reduced in both AHI and CHI, whereas the IL-2 and IL-17a secretion capacities of other subsets remained unaffected following HIV-1 infection. These results demonstrate that (1) CD26+CD39−CD73+CD4+ T cells have similar IFN-γ, IL-2, and IL-17a secretion abilities to other CD4+ T-cell subsets except CD26+CD39+CD73−CD4+ T cells; and (2) the production capacities of CD107a, IFN-γ, and IL-2 in CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T cells are not affected by HIV-1 infection, with only IL-17a secretion significantly regulated (specifically reduced in CD26−CD39−CD73+CD4+ T cells). The number of functional CD26+CD39−CD73+CD4+ subset cells, mainly those releasing CD107a and IL-17a, were inversely associated with HIV-1 viral load in AHI. The number of IFN-γ+CD26+CD39−CD73+CD4+ T cells showed a trend of inverse association with viral load in the AHI group, while this cell population was significantly inversely correlated with viral load in the CHI group Supplementary Figure 3B, C, https://links.lww.com/CM9/C901. Additionally, the number of functional CD26–CD39−CD73+CD4+ subset cells (including those releasing IL-2 and IFN-γ) also displayed an inverse association with HIV-1 viral load in AHI Supplementary Figure 3D, https://links.lww.com/CM9/C901. These outcomes indicate that higher number of functional CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T cells are associated with stronger inhibition of HIV-1 replication. Although CD39 expression on total CD4+ T cells was not upregulated in AHI and CHI groups, a higher proportion of CD39+CD4+ T cells, particularly the CD26−CD39+CD73−CD4+ subset, was correlated with higher HIV-1 viral load and lower CD4+ T-cell count in the CHI group Supplementary Figure 4A–D, https://links.lww.com/CM9/C901. Additionally, in the AHI group, a higher proportion of CD26+CD39+CD73−CD4+ T cells correlated with higher viral loads, while in the CHI group, this subset was associated with lower CD4+ T-cell count Supplementary Figure 4D, 4E, https://links.lww.com/CM9/C901. Conversely, in the CHI group, downregulated CD73 expression on CD4+ T cells showed a trend toward association with higher viral load and was significantly correlated with lower CD4+ T-cell count. Although the frequencies of CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T cells were reduced in AHI and CHI groups, a higher frequency of these cells was only associated with a higher CD4+ T-cell count in the CHI group Supplementary Figure 4D, 4F, https://links.lww.com/CM9/C901. These findings indicate that despite the reduction in CD73+CD4+ T cells, particularly the CD26+CD39−CD73+CD4+ subset, in both AHI and CHI individuals, these cells are associated with higher CD4+ T-cell count, especially in those with chronic infection. The three ecoenzymes CD26, CD39, and CD73 have been widely reported to play complex immunoregulatory roles in tumors and viral infections, including HIV-1 infection. In the context of HIV-1 infection, CD26 expression has been shown to promote apoptosis and enhance the cytopathic effect of T cells, a finding consistent with our observation that the proportion of CD26+CD4+ T cells, particularly CD26+CD39−CD73+CD4+ T cells, was reduced in HIV-1-infected individuals. Additionally, another study also observed decreased soluble dipeptidyl peptidase-4 (sDDP-4) activity and CD26 expression on CD4+ T cells in HIV-1 infection.5 This suggests that CD26-mediated T-cell apoptosis may partially explain the depletion of these CD26+CD4+ T-cell subsets observed in our study. In our study, CD39 expression on total CD4+ T cells remained unchanged during HIV-1 infection, consistent with some previous reports, but in contrast to studies showing expanded CD39 expression on regulatory T (Treg) cells in HIV-1-infected individuals (regardless of antiretroviral therapy ART status) and even unaltered CD39+ Treg cell expansion during acute HIV infection despite ART initiation.6 Notably, elevated frequencies of CD39+CD4+ T cells and CD39+ Treg cells have also been observed in severe acute respiratory syndrome coronavirus 2 infection, suggesting a conserved role of CD39+ T cells in viral infections.7 Besides, the higher proportions of CD39+CD4+ T cells and CD26−CD39+CD73–CD4+ T cells were associated with higher viral load and lower CD4+ T-cell count exclusively in CHI. These indicate that CD39 likely plays a detrimental role in CHI, owing to its predominant expression on Treg cells. As observed in HIV, hepatitis B virus (HBV), and SARS-CoV-2 infections, Treg cells correlate with disease progression by suppressing effector T-cell function, an effect potentially mediated by CD39-dependent adenosine production. Previous studies consistently show reduced CD73 expression in HIV-1-infected individuals: surface, intracellular, and transcriptional CD73 levels are decreased in CD8+ T cells (regardless of differentiation status), a downregulation potentially driven by elevated plasma ATP and mediated by increased miRNA-30b, miRNA-30c, and miRNA-30e.8 CD73 expression is also markedly diminished on both regulatory and effector T cells in HIV-1 infection, consistent with our findings of reduced CD73 on CD4+ T cells in infected individuals. Moreover, CD73+CD4+ T lymphocytes exhibit lower activation, proliferation, and exhaustion than CD73−CD4+ T lymphocytes in both HIV-1-infected individuals and healthy controls. Notably, despite the overall depletion of CD73+CD4+ T cells post HIV-1 infection, higher proportions of these cells correlate with lower viral load and higher CD4+ T-cell count in CHI individuals. These results confirm that CD73 expression confers T lymphocytes intrinsic properties that support HIV-1 infection control. In our study, both CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T lymphocytes harbored few senescent cells, indicating their longer lifespan. Indeed, human peripheral CD4+ T cells expressing CD73 are long-lived, highly functional, and capable of differentiating into tissue-residing cells. A decline in this subset in older adults may impair T-cell memory.9 Therefore, it was unsurprising that the long-lived CD73-expressing cells, particularly CD26+CD39−CD73+CD4+ and CD26−CD39−CD73+CD4+ T cells, decreased during HIV-1 infection. Notably, CD26+CD39−CD73+CD4+ T lymphocytes exhibited minimal exhaustion, with PD-1 and TIGIT expression either downregulated or unchanged during HIV-1 infection. Additionally, in both AHI and CHI groups, these cells showed moderate activation capacity and very limited proliferation. Nevertheless, their ability to produce IL-2 and IL-17a was comparable to that of other T-cell subsets. Furthermore, HIV-1 infection did not impair their production of the degranulation substance CD107a or the cytokines IFN-γ and IL-2. These results suggest that despite reduced numbers of CD26+CD39−CD73+CD4+ T cells, their immunological profile enables effective suppression of HIV-1 infection. Furthermore, several immune functions may explain the association between CD26+CD39−CD73+CD4+ T cells with effective HIV suppression. In both AHI and CHI groups, the frequency of multifunctional CD26+CD39−CD73+CD4+ T cells did not decrease significantly postinfection (data not shown). Moreover, the number of specific functional CD4+ T-cell subsets, particularly IFN-γ+CD26+CD39–CD73+CD4+ T cells, correlated with low viral load, indicating these cells may critically suppress HIV-1 replication via IFN-γ secretion. Additionally, CD26+CD39−CD73+CD4+ T cells may inhibit HIV-1 replication in AHI through IL-17a and CD107a secretion, while CD26−CD39−CD73+CD4+ T cells exert this effect via secreting IL-2 and IFN-γ. These subsets share common antiviral mechanisms such as IFN-γ secretion, but also rely on distinct immune factors to mediate HIV-1 control. In summary, CD26+CD39−CD73+CD4+ T cells are the third most prevalent T-cell subset in healthy individuals, but markedly diminished during HIV-1 infection. This unique subset presents low proliferation and weak cytotoxicity with comparable activation and IL-2/IL-17a secretion to other CD4+ T cell subsets. Notably, they remain resistant to immune exhaustion and senescence irrespective of HIV status, maintaining stable functional potential under chronic viral infection. Funding This work was supported by the National Natural Science Foundation of China (Grant No. 82472266 to B.S.), the National Key R&D Program of China (Grant Nos. 2023YFE0116000, 2023YFC2308300, and 2023YFC2308302). Conflicts of interest None.