Almost 40 years after its initial discovery, HIV-1 remains a major health crisis worldwide with over 39.9 million individuals infected in 2023 and 1.3 million new infections in 2018. While HIV-1 infections occur across the globe, the heaviest disease burden is present in sub-Saharan Africa, harboring over 2/3 of all HIV-1 cases (UNAIDS, 2023).

HIV-1 demonstrates a great potential for generating genetic diversity due to lack of proof-reading ability in the reverse transcriptase and high viral recombination rates (Hemelaar, 2012; Hemelaar et al., 2019). In order to organize the amount of genetic diversity, HIV-1 group M is divided into nine different subtypes, A–D, F–H, J, K, and L (Hemelaar, 2012; Hemelaar et al., 2019; Yamaguchi et al., 2020). Globally, the two most common subtypes observed are subtype C (47% of global infections) and A (10% of global infections); these two subtypes are also the most common subtypes observed in sub-Saharan Africa (Hemelaar et al., 2019).

Clinical differences can be observed in infections by the nine subtypes globally. Subtype D, a common subtype in sub-Saharan Africa, was found to be associated with increased mortality compared to subtype A and has a 4-fold higher rate of CD4⁺ T cell decline in the absence of anti-retroviral drugs compared to other subtypes (Easterbrook et al., 2010; Hemelaar, 2012). Individuals infected with subtype D have also been reported to have weaker humoral immune responses to HIV infections compared to individuals infected with other subtypes (Longosz et al., 2014).

While there are numerous studies documenting characteristics of subtype C infections, studies exploring the features of subtype A infections are less common. However, one of the early studies looking at subtype A infections found that female sex workers infected with non-subtype A viruses were 8 times more likely to develop AIDS compared to those infected with subtype A (Kanki et al., 1999).

A multi-site study of acute-early HIV infections, termed IAVI Protocol C, enrolled over 600 newly infected individuals, who were infected by subtype A, C, D and recombinant viruses, and who were followed for as long as 10 years with regular viral load and CD4⁺ T cell measurements (Landais et al., 2016; Price et al., 2020). Studies within Protocol C found that infections with subtype A and C viruses resulted in different clinical presentations in the infected patients. Individuals infected with subtype C were found to have a 60% faster disease progression compared to subtype A, based on viral load, CD4⁺ T cell decline, and time to AIDS (Amornkul et al., 2013). Moreover, in the IAVI Protocol C cohort, a higher proportion of viral controllers and long-term non-progressors were observed following subtype A infections, compared to individuals infected with subtype C in the absence of antiretroviral therapy (Price et al., 2019). In addition, in multivariate analyses viral control was shown to be independent of favorable HLA alleles (Prentice et al., 2014; Price et al., 2019). Finally, in HIV discordant couples, where one partner was infected with HIV, subtype C in Zambia was associated with higher post-counseling HIV-1 incidence compared to subtype A in Rwanda despite the same counseling and interventional approaches in each country (Kamali et al., 2015). Collectively, these data indicate that virus-specific characteristics may determine a less pathogenic outcome in subtype A infected individuals.

In this study, we sought to investigate possible causes for the observed clinical differences seen between HIV-1 subtype A and C infections and if these differences could be observed during acute HIV-1 infection. To address this question, we compared the replicative capacity of transmitted-founder viruses from each subtype, since we have shown previously that the trajectory of HIV pathogenesis is defined in part by this trait of the virus. In addition, because inflammation and inflammatory cytokines are linked to more rapid disease progression in HIV-1 infected individuals (Kuller et al., 2008; Prince et al., 2012; Claiborne et al., 2015), we examined the cytokine/chemokine profiles of individuals pre- and post-infection with either subtype A or C HIV-1 to see if there were any observable differences in the biomarker response in the acute phase of infection between subtypes.

We have previously shown that the phenotype of the virus that initiates infection in an individual, the transmitted founder virus, can significantly impact both viral load and disease trajectory (Prince et al., 2012; Claiborne et al., 2015; Yue et al., 2015). In a previous study of more than 120 individuals newly-infected with subtype C HIV-1, we demonstrated that the replicative capacity (RC) of the infecting virus was associated both with set-point viral load and the speed of CD4⁺ T cell decline (Prince et al., 2012; Claiborne et al., 2015). A number of studies comparing the replicative capacity of viruses, chimeric in either gag or env genes, have suggested that different subtypes have different replicative capacities (Archary et al., 2010; Venner et al., 2016). In a limited study of five subtype A and five subtype D authentic molecularly cloned TFVs, we showed that subtype D TFVs replicated better than their subtype A counterparts (Baalwa et al., 2013). Because subtype A HIV-1 infected individuals exhibit similar disease characteristics as individuals infected with low vRC viruses of other subtypes, we sought to determine whether authentic transmitted founder viruses derived from acutely infected subtype A and subtype C individuals similarly differed in their replicative capacity.

In this study, where we compared a total of 30 viruses derived from molecularly cloned TFVs; we did not find a significant difference in the replicative capacity of HIV-1 subtype A and C TFVs. Based on this data, it does not appear that replicative capacity in activated CD4⁺ T cells in vitro can explain the differences in disease progression observed in the two subtypes. However, it should be noted that the small number of viruses analyzed from each subtype limits the strength of this conclusion and may make subtle but potentially biologically important differences harder to detect. Nevertheless, the results are consistent with the findings of Prentice et al. (2014), who showed that viral loads at 3 months after infection were similar for subtype A and C infected individuals.

When we examined some of the characteristics associated with disease progression, we saw a lack of association between RC and set point viral load (data not shown), consistent with a previous study where the association between the two factors was weak (Prince et al., 2012). In contrast, we were able to observe a clear separation of the Kaplan–Meier curves for low RC and high RC viruses and time to CD4 counts under 300, which while not significant, likely due to the small number of viruses, does support an association between RC vs. CD4 decline in this study of authentic TFV. This is consistent with findings from a previous study that examined 127 gag-MJ4 chimeric viruses and showed that low viral RC was significantly associated with slower CD4⁺ T cell decline, while high viral RC was associated with rapid CD4⁺ T cell decline (p = 0.002; Claiborne et al., 2015). In addition, both this previous study and our current study found that high RC viruses are linked to higher levels of inflammatory cytokines and chemokines (Claiborne et al., 2015).

Since inflammation is known to be a driver of disease progression and CD4⁺ T cell loss in HIV-1 infection (Fiebig et al., 2003; Roberts et al., 2010; Jiao et al., 2012; Shive et al., 2014; Claiborne et al., 2015), we wanted to determine whether there were differences in inflammatory biomarkers that could explain the differences in CD4⁺ T cell decline in subtype A and C infected individuals in the Protocol C cohorts (Amornkul et al., 2013). We found a greater number of biomarkers were increased as a result of HIV-1 subtype C infection compared to subtype A infections, but, while differences in biomarkers levels were observed within subtypes before and after infection, the overall levels of the biomarkers following HIV-1 acute infection were the same in the two subtypes. In addition, while subtype C infections result in the increase in multiple biomarkers post infection, subtype A infection is mostly defined by an increase of TNFα. The reduced number of cytokines that increased following subtype A HIV-1 infection was due in part to the higher levels of biomarkers observed prior to infection in these individuals. We cannot rule out the possibility that it is the magnitude of increase from uninfected levels of inflammatory cytokines/chemokines to those after infection, rather than the absolute levels themselves, that mediates the immunological dysfunction and elevated CD4⁺ T cell decline observed in subtype C.

The cause for the differences observed between HIV-1 subtype A and C infections remains unclear. One confounding aspect is that subtypes are generally linked to geographic locations, with subtype A being predominantly found in East Africa (Rwanda, Kenya and Uganda) while subtype C is strongly associated with southern African countries (South Africa and Zambia). The ethnic and genetic differences that distinguish individuals in Rwanda and Zambia could therefore influence the immune responses to these viruses. Indeed, one possible explanation is that subtype C viruses are better adapted to the immune system of their new hosts than subtype A viruses. We have previously shown that infections by subtype C TFVs that have a higher fraction of epitopes preadapted to the HLA-directed immune response of their new host result in higher viral loads and faster CD4⁺ T cell decline (Carlson et al., 2016; Monaco et al., 2016). In the study by Monaco et al. (2016), viruses from 169 subtype C heterosexual transmission pairs were sequenced and on average one-quarter of the possible HLA-linked target sites in the transmitted virus Gag proteins were already adapted to the new host’s HLA types, with transmitted preadaptation significantly reducing early immune recognition of epitopes. While the extent of immune adaptation in subtype A viruses to their hosts is currently unknown, it is possible that the ongoing viral control observed in subtype A infections reflects less adaptation to their Rwandan hosts compared to subtype C infections in Zambian individuals; ongoing studies are aimed at addressing this question.

In an unexpected finding, we were able to find an association between CD4⁺ T cell count decline and higher levels of the inflammatory chemokine ITAC when we combined the cohorts in this study. ITAC, or CXCL11, is produced by multiple immune cells (neutrophils, monocytes, and macrophages) and endothelial cells in response to IFNγ and binds to CXCR3, which is preferentially expressed on Th1 cells (Cox et al., 2008). Of the defined CXCR3 ligands, ITAC is the most potent and efficacious for chemotaxis of activated T cells and has the greatest affinity for CXCR3 as well (Petkovic et al., 2004). One study found that the presence of ITAC in an in vitro transwell migration model resulted in a 4–6 fold increase in the migration of T cells (Mohan et al., 2002). Another study from 2005 found that ITAC mRNA levels were upregulated in HIV-1 infected human monocyte-derived macrophages and dendritic cells. Medium from these infected cells were then found to be chemotactic for freshly isolated human CD4⁺ T cells; however, when the CD4⁺ T cells were pretreated with an anti-CXCR3 antibody (the receptor that ITAC binds to), the previously observed chemotaxis was abolished (Foley et al., 2005). This study also found that ITAC mRNA levels were upregulated within a lymph node isolated from an HIV-1 infected individual, which implicates ITAC in the recruitment of susceptible CD4⁺ T cells to HIV-1 infected lymph nodes (Foley et al., 2005). The authors of this study hypothesize that since CCR5⁺ T cells usually also express CXCR3, ITAC plays an important role to recruitment of these susceptible cells to HIV-1 infected regions or cells; this recruitment might then be able to “enhance the sequestration of T cells in infected lymphoid organs and spread of HIV-1 infection between cells” (Foley et al., 2005). Overall, this previous study is consistent with the results presented here that ITAC can be a contributor to AIDS immunopathology.

Previous studies have identified CD4⁺ T cell decline and viral replicative capacity are both markers of more rapid disease progression (Giorgi et al., 1999; Prince et al., 2012; Claiborne et al., 2015). Our results raise the possibility that high ITAC levels may be a marker for rapid disease progression. This result falls in line with our previous observations that high ITAC levels in uninfected partners were associated with HIV-1 acquisition and that the ITAC levels could be used as a predictive variable to identify individuals that would eventually seroconvert within serodiscordant couples (McInally et al., 2021). The findings of the current study show that ITAC may not only be associated with susceptibility to HIV-1 acquisition, but also disease progression in HIV-1 infected individuals. They highlight the need to further examine the role of ITAC in HIV-1 infection.

Overall, the data presented here demonstrate similar in vitro replicative capacities for subtype C and A transmitted founder viruses, arguing against this phenotypic trait as the basis for different pathogenic outcomes in the two subtypes. Moreover, levels of inflammatory cytokines that might drive disease progression were similar during acute infection indicating that additional studies are required to understand the mechanism underlying differences in disease progression between the two subtypes.