In 2023, The Joint United Nations Programme on human immunodeficiency virus (HIV) and acquired immunodeficiency syndrome (AIDS) reported the estimated prevalence of 380,000 of people living with HIV (PLWH) across all ages in Mexico (1). Of those, 64% of PLWH are on antiretroviral therapy (ART) and 59% have suppressed viremia (defined as plasma viral load below 40 copies of HIV-1 RNA/mL) (1). The human immunodeficiency virus type 1 (HIV-1) requires both the CD4 receptor and a co-receptor, either C-C chemokine receptor 5 (CCR5) or C-X-C chemokine receptor 4 (CXCR4) to infect hosts cells. The preference for either co-receptor is referred to as viral tropism. HIV viral tropism is underreported in Mexico. This is of concern as it is crucial for understanding HIV pathogenesis and disease progression, particularly in a country where individuals are diagnosed with advanced HIV disease (late presenters) (1). The entry is mediated by the glycoproteins (gp) gp120-gp41 (2). HIV-1 transmitted/founder (T/F) virus typically uses CCR5 receptor (R5 virus) for entry (3), which dominates during the first weeks or years of HIV-1 infection. The virus can evolve and switch to use CXCR4 receptor (X4 virus) if the X4 virus has a higher fitness. The X4 phenotype is mostly seen in late stage of HIV infection (4, 5). Elevated immune activation and coreceptor switch have been associated with HIV-1 disease progression (6–10). There are three hypotheses that attempt to explain the processes underlying coreceptor switch and its association with disease progression, based on empirical evidence or model support. First, molecular determinants located in the V3 loop of gp120 confer the tropism coreceptor phenotype and few genetic changes in the sequence are required to determine which coreceptor is used to trigger the process of entry (2, 11). These changes are assumed to be the result of evolution of the viral populations during infection, where the quick turnover and high mutation rate of the virus increase the probability of X4 virus emergence as a random event during any stage of the disease (12, 13). Second, in immunocompetent people living with HIV (PLWH), X4 viruses do not have the fitter phenotype and are better recognized by the immune system than R5 viruses (7, 14). The CD8+ T cell response against X4 virus replication appears to be more effective than that against R5 virus. However, the progressive erosion of the immune system leads to reduced susceptibility to X4 virus (7, 15). If the X4 virus exhibits higher replicative fitness than the R5 counterparts, it could lead to its emergence and, in some cases, replace the predominance of the R5 virus variants (13, 16, 17). Third, HIV-1 exhibits different preferences for target cells, whereby CD4+ T-cells represent the primary target. Naive CD4+ T-cells express high levels of CXCR4+, while memory CD4+ T cells express both CXCR4+ and CCR5+ (18). Over time, the activation of naive CD4+ T cells into memory CD4+ T cells increases, as does the frequency of their proliferation. This leads to selective pressure in favor of the R5 virus, as the number of target cells increases. In the later stages of the infection, the proliferation rate of naive CD4+ T cells further increases, shifting the selection pressure in favor of X4 variants (19–22). This switch from CCR5 to CXCR4 coreceptor is associated with rapid CD4+ T cell depletion and disease progression. Chronic immune activation characterizes HIV-1 disease progression and is often measured by the expression of the surface markers HLA-DR and CD38 on both CD4+ and CD8+ T cells (22). In addition to the aforementioned surface markers, other surface markers have been identified as indicators of T cell exhaustion (programmed death-1, PD-1) and senescence (CD57+). The expression of these markers has been implicated in the impairment of effector functions and T cell terminal differentiation and extensive proliferation, thereby contributing to the compromise of the immune system’s ability to control viral replication (22–26). Furthermore, a broad spectrum of biomarkers have been used to quantify chronic immune activation, and these biomarkers remain elevated even in PLWH with antiretroviral therapy (ART) with undetectable viraemia (27). Among these biomarkers, interleukin-6 (IL-6) is of significant interest. IL-6 is a cytokine that plays a critical role in activating and regulating the immune response to viral infections. Numerous studies have demonstrated a correlation between IL-6 and mortality and/or morbidity in PLWH (28, 29). In addition, microbial translocation (the passage of bacteria and/or microbial products from the gut into the systemic circulation) triggers the activation of monocytes and/or macrophages during the early phase of the disease, resulting in the shedding of sCD14 and sCD163 (22). sCD14 is a marker of monocyte activation, is shed in response to lipopolysaccharide (LPS) stimulation which reflect CD4+ T cell depletion in the gut-associated mucosal lymphoid tissue (GALT) and correlates with the expression of HLA-DR and CD38 on CD8+ T cells (27, 30). Soluble CD163 (sCD163) is a marker of HIV activity reflecting the virus replication and T cell activation (31). Several studies have reported discrepancies regarding whether X4 variants are the cause or consequence of a dysfunctional immune response. Recent studies have examined the relationship between the presence of X4 variants and disease outcome. In 2017, Hayashida, et al. (32) reported that the rate of decrease of CD4+ T cell count or the rise of HIV-1 load accelerated significantly after the emergence of X4 variants in two individuals experiencing slow progression. Also, in 2020, Connell et al. (7) demonstrated that X4-tropism is not the favorable viral phenotype in the immunocompetent host. They observed that higher levels of immune activation can predict the subsequent selection of X4 variants. Therefore, it is essential to have a better understanding of the mechanisms that drive viral infection via X4 variants, from minority to dominant viral populations. Evidence suggest that X4 variants exhibit a low replication rate in early HIV-1 infection. This is due to the fact that X4 variants target preferentially naive CD4+ T cells, which results in insignificant viral production (33). For this reason, next generation sequencing (NGS) platforms are well-suited for the characterization and genotyping of viral diversity. Furthermore, bioinformatic algorithms such as geno2pheno (34) are used to predict the tropism in the viral population. Consequently, X4 variants can be detected at very low levels shortly after HIV transmission. Also, X4 variants have been observed to exhibit increased pathogenicity which contributes to CD4+ T cell depletion. This, in turn, leads to the proliferation of target cells that are adequate for replication in terms of R5 and X4 variants. Such changes may create an environment in the host that maintains chronic immune activation, which creates the optimal conditions for switching coreceptor preference. Immune activation is critical for HIV-1 to maintain replication and cellular infection. To determine whether early immune activation is associated with the presence of X4 variants, it is crucial to examine the most important cellular markers, in particular HLA-DR and CD38, in CD4+ and CD8+ T cells. Additionally, the significance of soluble markers such as sCD163, sCD14 and IL-6, which are associated with inflammation and the activation of antigen-presenting cells in PLWH with the presence of X4 variants, has yet to be fully elucidated. The aim of this study is to determine the levels of immune activation and their relationship with viral tropism in PLWH who have not been previously treated in México City. We used NGS to determine the viral tropism of HIV; enzyme-linked immunosorbent assays to quantify soluble markers of inflammation and immune activation; and flow cytometry to quantify T cell immune activation, senescence, exhaustion, and expression of coreceptors (CCR5+ and CXCR4+).

In this cross-sectional study, we assessed coreceptor usage by using a highly sensitive assay to detect X4 viral variants and examined levels of chronic immune activation in PLWH. Although no statistically significant values were observed, our data appear to be consistent with the increased levels of CD38+ alone and HLA-DR+CD38+ in the X4 group in relation to the R5 group (see Figures 3 , 4 ), providing suggestive data of T cell activation. However, our data did not provide sufficient evidence to conclude that PLWH harboring X4 viral variants exhibit elevated levels of markers of immune activation. Conversely, evidence was provided that PLWH exclusively harboring the R5 virus exhibited increased levels of IL-6 in comparison to PLWH with a mixed R5 and X4 virus population.

The initial report, which sought to establish a relationship between the cellular markers of activation and the emergence of X4 viral variants, was conducted by Connell and colleagues (7). This study observed that PLWH with X4 viral variants exhibited elevated levels of HLA-DR+ CD38+ in CD4+ T cells. Moreover, at five years of follow-up, a positive correlation has been identified between X4-tropism and CD4+ HLA-DR+ T cells and CD4+ HLA-DR+ CD38+ T cells (7). It has been demonstrated that the expression of CD38+ alone in CD8+ T cells is strongly associated with disease progression, irrespective of the CD4+ T cell count. This provides evidence that CD38+ expression in CD8+ T cells is a useful marker of lymphocyte activation in studies such as those conducted by Giorgi JV and colleagues (22, 43, 44). Furthermore, research has been conducted on the expression of CD38+ alone, or in combination with HLA-DR+ in CD4+ T cells. This research revealed that CD38+ expression in CD4+ T cells is an effective prognostic marker of impaired immune function (19, 45–48). One hypothesis that may explain the relationship between cellular activation, particular that of CD4+ T cells, and the presence of X4 virus with the deterioration of the immune system, is the high cytotoxicity of the X4 virus in relation to the R5 virus (8, 11, 49, 50). Holm and colleagues (46) have demonstrated that the infection of primary T cell cultures with the CXCR4-tropic HIV-1 variant ELI6, induce activation of CD4+ and CD8+ T cells, which subsequently undergo apoptosis. Additionally, this activation is attributed to the presence of replicating and non-replicating viruses (46, 51). In contrast, Meditz et al. (45) report that the primary source of HIV-1 propagation in lymph nodes was the activated CD4+ T cells expressing HLA-DR+ CD38+ and high levels of CCR5 which are the main target of R5 virus. This is likely the main cause for CD4+ T cell depletion in lymphoid tissue. Furthermore, in the same study, in vitro models have demonstrated that most R5 and X4 viruses replicate in CD4+ T cells expressing CD38+, irrespective of which coreceptor is expressed on these cells. This suggests that the depletion of CD4+ T cells is driven by either direct cell lysis or bystander apoptosis. The precise mechanism by which coreceptor tropism contributes to this process will be discussed in more detail later. However, it is evident that activated lymphocytes are central to the virus’s propagation.

It is noteworthy that when CD4+ T cells expressing both CCR5+CXCR4+ coreceptors and CD38+ were selected, a significant difference was found between R5-tropism PLWH and X4-tropism (7.8 vs 10.7, p = 0.021, see Figure 4 ). The underlying reasons for the observed discrepancy compared to CD4+ T cells expressing only CD38+ require further discussion. One plausible hypothesis is that the small sample size of the X4 group (n=5) may have influenced the observed outcomes. The data suggest that PLWH within this group exhibited elevated markers of immune activation, although not to a significant extent, a limitation previously acknowledged by Cassadella et al. (6, 52). These authors propose that implementing NGS technology in conjunction with a more stringent FPR threshold, could refine the identification of PLWH harboring the X4 virus. Such an approach would likely reduce the number of PLWH who are stratified within the X4-group, thereby addressing the current limitations in statistical power due to sample size constraints. Additionally, it is also interesting to note that circulating central memory CD4+ T cells express both coreceptors (CCR5 and CXCR4) and exhibit an activated phenotype characterized by CD38 expression (53). These cells reside outside the GALT. Given that the cells analyzed in this study were derived from PBMC, this may explain why significant levels of cellular activation were specifically observed in these cells. However, no differences were detected between groups in terms of lymphocyte subtypes (data not shown), nor was there any evidence to suggest that the expression levels of coreceptors predispose PLWH to the presence of X4 variants. Importantly, the relationship between cellular activation and the emergence or presence of X4 variants does not necessarily imply changes in coreceptor expression levels or the co-expression of both markers associated with cellular activation (HLA-DR+ and CD38+) (10, 54).

In this study, the hypervariable V3 region of HIV-1 was analyzed using the online platform geno2pheno, which provides false-positive rate (FPR) cutoffs to estimate the probability of incorrectly classifying R5-tropic virus sequences as X4-tropic. A lower FPR indicates a reduced likelihood of misclassifying X4-tropic sequences. To characterize highly diverse viral populations, we employed the Illumina MiSeq platform, which offers significant advantages over other next generation sequencing platforms by minimizing errors associated with homopolymers, insertions, deletions, and other sequencing artifacts. Additionally, this technology allows for increased sequencing depth per sample, reduces error rates, and avoids artifacts derived from PCR-mediated recombination. These advantages enabled accurately identifying and quantifying of low-frequency X4-tropic viral subpopulations within a broader set of viral quasispecies. However, inherent issues with the technology, such as PCR resampling and PCR/sequencing errors, limit the study of these low-frequency variants. In order to address these issues, a number of studies have developed unique molecular identifiers (not applied in this study), such as Primer ID (55–57), with the aim of identifying true low-frequency variants and reducing errors in genetically diverse populations. In this study, we explore different cut-off points for the viral population, with thresholds of >2% and >5% (a more conservative approach) being examined. Our findings indicate that there are no discrepancies in the stratification of the PLWH by tropism when both cut-offs are employed (data not shown).

In this cohort, the median number of viral variants was 1836 (IQR = 937.5 - 2942.5), reflecting the high viral diversity and genetic variability of HIV-1. The prevalence of X4-tropic viruses in late HIV presenters was 18.4%, consistent with findings reported in other studies (16, 17, 39, 52, 58–60). In contrast, in PLWH with advanced disease stages, up to 50% of PLWH have been documented to harbor X4-tropic viruses. The remaining viral variants likely represent populations with reduced viral fitness, which appear and disappear continuously. However, these variants may play a relevant role in CD4+ T-cell depletion, either through direct lysis or bystander apoptosis. Furthermore, it has been observed that the gp120 can spontaneously shed from the surface of virions and infected cells, suggesting a potential role in modulating immune responses. Variable levels of soluble gp120 have been reported in PLWH (61). This soluble form of gp120 has been shown to interact with the CD4 receptor on T lymphocytes, monocytes, dendritic cells, and macrophages, inducing the secretion of proinflammatory cytokines such as IL-6, IL-10, IL-1β, interferon-alpha and gamma, and tumor necrosis factor-alpha (TNF-α) (62). These findings highlight the importance of gp120 in not only viral entry but also in the pathogenesis of HIV through the activation of inflammatory responses.

With regard to the clinical characteristics exhibited by the cohorts under investigation, significant differences were identified in CD4+ T cell count and pVL. The INER cohort was comprised of PLWH who had probably delayed even more time their diagnosis and treatment for various reasons. As tertiary-level of care hospital, most PLWH arrived with advance disease. By contrast the IMSS cohort exhibited favorable CD4+ and pVL values, which could be indicative of reduced levels of soluble markers (sCD14, sCD163 and IL-6; data not shown). Conversely, no significant differences were observed between the R5 and X4 groups with respect to CD4+, CD8+ T cell count and pVL. Connell and colleagues (7) observed a negative association of X4-tropism and CD4+ T cell count; however, our findings related to CD4+ T cell count did not show significant differences between both groups. It is acknowledged that a potential confounder in this study is the fact that over 60% of the PLWH in both groups had CD4+ T cells <200 cells/µL, a likely consequence of late diagnosis of HIV-1 in these individuals. The low CD4+ T cell count at the time of diagnosis indicates that the majority of PLWH were at a stage of advanced disease, which is also reflected by elevated levels of pVL. Despite the sustained efforts to implement strategies to initiate early treatment with antiretroviral therapy in Mexico, late diagnosis remains frequent among individuals attending health services due to fear of discrimination, stigma, and social inequalities, in particular with respect to access to health care. Consequently, the stratification of these individuals with early-stage disease was complicated. Clinical data presented in this study is consistent with the findings of previous studies (63, 64), which reported that more than 50% of individuals were diagnosed and treated in a late stage of the disease in Mexico. These individuals typically presented with a CD4+ T cell count of less than 200 cells/mm³ and/or a history of AIDS-defining illness.

In this study, levels of sCD163 are consistent with Connell and colleagues findings (7), who identified elevated levels of sCD163 in the R5 group, though not significant. In some studies, this soluble marker has been identified as a useful predictor of mortality and morbidity and is associated with elevated plasma viral load (31, 65, 66). Regarding sCD14, similar values were observed across groups. Given that these PLWH were diagnosed at a late stage, it is plausible to infer a process of monocyte/macrophage stimulation via LPS, which is characteristic of the chronic phase of infection. This is associated with elevated levels of IL-6, immune activation, gut dysbiosis, and microbial translocation (30, 67).

In contrast, our study found that IL-6 was significantly elevated in the R5 group. This finding appears counterintuitive, as it would be hypothetically expected that the X4 group would cause greater immune activation. To date, there are no reports linking elevated IL-6 levels to the presence of X4 viral variants in PLWH. Nonetheless, there are a few reports where IL-6 is strongly induced by gp120. Del Cornò, et al. (68) showed that the engagement of coreceptor CCR5 by a HIV-1 R5-tropic gp120 promotes STAT3 activation concentration-dependently and time-dependently leading to IL-6 secretion by immature monocyte-derived dendritic cells. STAT3 belongs to family of proteins known as signal transducers and activators of transcription (STAT), which are involved in diverse biological processes, such as cell differentiation, apoptosis and inflammatory response (69, 70). In order to corroborate the involvement of the coreceptor, the authors used a CCR5 antagonist, Tak779, which blocks the function of the coreceptor CCR5 and the signaling mediated by CCL4 (the natural ligand of CCR5). CCL4 did not induce the secretion of IL-6 by itself, which confirms that this ligand does not promote STAT3 activation in a natural manner. Moreover, Shah, et al. (71) showed that gp120 increased the expression of IL-6 at both mRNA and protein levels in astrocytes. These cells express both coreceptor CXCR4 and CCR5. Their findings indicated varying levels of IL-6 expression in response to several strains of gp120, including X4-tropic gp120IIIB which exhibited lower expression levels compared with R5-tropic gp120 CN54, gp120 CM and gp120 Bal strains. The involvement of the coreceptor in the viral entry mechanism may also initiate signaling events that can modify the cellular function or impact the post-entry stage of infection. Therefore, it can be demonstrated that CCL4 does not intervene in the secretion of IL-6, nor can it be excluded that CXCR4 does not implicate in the activation of STAT3. The hypothesis that lower levels of IL-6 in PLWH with X4 viruses could promote IL-6 secretion via STAT3 to a lesser degree is therefore proposed, as suggested by the aforementioned results.

Furthermore, principal component analysis was performed to assess interdependencies among the studied variables. This analysis revealed a significant negative correlation between CD4+ T cell counts and IL-6 levels, consistent with our findings using Spearman test described earlier. In the PCA, IL-6 emerged as a positive vector, clustering subjects harboring R5-tropic viruses on the right side of the biplot. In contrast, CD4+ T-cell counts were represented as a negative vector, grouping PLWH with X4-tropic viruses on the left side. Notably, the soluble markers sCD14 and sCD163 did not contribute substantially to the variance explained by the first principal component (PC1). However, these markers aided in distinguishing PLWH with elevated levels of soluble markers, which were positioned on the right side of the graph. It is worth noting that the majority of PLWH exhibited markedly low CD4+ T cell counts, which may account for the observed dispersion in the PCA plot and the absence of well-defined clusters separating the X4 and R5 groups. Moreover, a multiple logistic regression model was used to evaluate the predictive value of the variables. The model underscored the significance of IL-6, indicating that elevated levels of this cytokine may serve as a predictive biomarker for the presence of exclusively R5-tropic viral populations. These findings are consistent with the hypothesis that R5-tropic viruses are associated with an enhanced inflammatory response, as reflected by increased IL-6 levels, whereas X4-tropic viruses may associate with a less pronounced inflammatory state but exhibit greater cellular activation in terms of CD38+ and/or HLA-DR+ expression, leading to greater CD4+ T-cell depletion.

It is hereby proposed that the HIV-1 T/F R5-tropic viruses are the primary causative force of T cell activation and significant CD4+ T cell depletion, particularly within the gut-associated lymph tissue during the initial phase of the disease. The mucosal site represents the primary route of viral entry into the host, wherein a significant proportion of immune cells express the CCR5 coreceptor (e.g. macrophages, dendritic cells, and memory CD4+ T cells), thereby conferring a selective advantage to the R5-tropic viruses over X4-tropic viruses (3, 35). The R5 viruses compete with X4 viruses for the cellular targets that express both coreceptors, especially those activated CD4+ T cells, as these cells are metabolically more active, thus providing a highly favorable environment for replication (72). This dynamic limit the replication of the X4 viruses, consequently leading to a reduced production of the virions. In addition, the R5-tropic viruses have been distinguished by their increased viral burst, defined as the number of virions released per infected cell (10). Therefore, it is rarely observed that the X4 viruses predominate during the early stages of the disease. We propose that the relationship between coreceptor switching and immune activation is bidirectional; in one direction, immune activation facilitates the change of coreceptor use by creating an environment more suitable for the selection of X4 viruses. In the absence of antiretroviral therapy, the infection process gives rise to the emergence of competent viruses that exhibit diverse phenotypes. This phenomenon is a direct consequence of the host environment undergoing constant alterations in response to the dynamic interplay between the virus and its host. Consequently, this process generates a series of selective pressures that favor proliferation of variants better adapted to the prevailing environmental conditions (3, 51, 72). In the opposite direction, at the late stage of infection, the X4 virus has been observed to diminish immune system function by means of the activation of additional CD4+ T cells and the more efficient destruction of immune cells, thus perpetuating the inflammatory cycle and immune damage (8, 46, 49–51). It is imperative to better understand the interconnected processes of chronic immune activation and coreceptor switching in HIV infection, as these may associate to the progression of the disease. Elucidation of these mechanisms is essential for the formulation of therapeutic strategies that modulate immune activation and prevent viral tropism switching. The objective of such strategies is to preserve immune function and enhance clinical outcomes in individuals living with HIV.

During the course of this study, we encountered several limitations. Firstly, we were limited in reporting causality due to the cross-sectional nature of this study. Secondly, as previously mentioned, the statistical power of our sample size in the X4 group was limited. Thirdly, there was an absence of available PBMCs in the IMSS cohort. These limitations might account for the lack of statistical differences when comparing markers of immune activation between the R5 and X4 groups. Lastly, the majority of the PLWH recruited in both cohorts showed reduced CD4+ T cell counts, suggesting prolonged viral exposure, which could result in a sustained immune activation with increments in the levels of several biomarkers in individuals with greater time exposure. The study could not determine whether participants had acute or early HIV infection, thus Fiebig staging could not be performed. Lastly, unmeasured confounders may have influenced levels of cytokine and immune activation, in particular coinfections with cytomegalovirus, hepatitis C and B, and opportunistic infections, and smoking. These confounders were not recorded for most of PLWH in our study, limiting our ability to include them in our analyses. Contrastingly, the study’s strengths are supported by the use of advanced technological instruments, such as the employment of flow cytometry for the quantification of immune activation markers and the integration of paired-end sequencing data with the geno2pheno algorithm for the identification of individuals harboring X4 variants. Furthermore, the participants in this study represent a diverse sample of the PLWH residing in the metropolitan area of Mexico City.

This study explored coreceptor usage and chronic immune activation in late HIV-1 presenters, revealing trends of elevated CD38+ and HLA-DR+CD38+ in CD4+ or CD8+ or both expression in the X4 group and significantly higher IL-6 levels in the R5 group. While no statistically significant differences were observed in the cellular markers of inflammation, the data is consistent with prior research (7). The findings suggest that R5-tropic viruses drive inflammatory responses in early stages, while X4-tropic viruses may enhance CD4+ T cell depletion through immune activation. These insights highlight the complex interplay between viral tropism and immune activation. Future studies with larger sample sizes and longitudinal designs are needed to confirm these findings and explore the causal relationships between viral tropism, immune activation and disease progression.