Human immunodeficiency viruses (HIVs) are one of the most crucial members of the retroviral family Retroviridae. Among the two main types of HIVs i.e., HIV-1 and HIV-2, HIV-1 has been characterized as the major causative agent of acquired immune deficiency syndrome (AIDS) (Blattner et al., 1988; Weiss, 1993).

Natural variability of HIV-1 is the key underlying the complex biology of the menacing virus. Evolutionary and epidemiological history of HIV can be investigated through phylogenetic analysis of erratic forms of the virus. Interestingly, due to rapid availability of nucleotide sequence data of viral genes, it is nowadays possible to illustrate a detailed phylogenetic relationship of viruses like HIV. Phylogenetic arrangements of HIV-1 are commonly carried out using nucleotide sequences of different sub-genomic regions of the same HIV-1 genome i.e., gag, pol, and envelope (env). This approach has already revealed unique inter-subtype recombinant forms of virus isolates (McCutchan, 2006). Envelope (env) gene encodes known targets for cytotoxic T lymphocytes and neutralizing antibodies (Goudsmit et al., 1988; Chesebro et al., 1991). Despite extensive study of sequence variation of env gene of HIV-1, the exact impact of selection in governing the patterns of variation in the env gene still remains somewhat obscure. In some earlier studies, based on lower frequency of synonymous nucleotide substitutions within the V3 loop in the env genes, it has been circuitously verified that the variation in env gene is maintained by selection for antigenic diversity (Simmonds et al., 1991; Bonhoeffer et al., 1995). Theoretically, it has also been implicated that selection pressure and mutability dynamics are the vital forces that affect viral fitness and robustness (Brown, 1997). It has been hypothesized that negative (purifying) selection constraints, operational in HIV-1, execute a crucial role in devising molecular evolutionary patterns of HIV-1 in contrast to positive (diversifying) selection which has been reported to play a minor role (Seo et al., 2002; Drummond et al., 2003).

Investigations pertaining to viral divergence in HIV-1 patients, associated with varying rates of progression of disease, have often produced interesting and conflicting results (Ganeshan et al., 1997; Arts and Quiñones-Mateu, 2003; Rangel et al., 2003). Some researchers have noted that non-synonymous substitutions in the env gene have been higher in long-term non-progressor (LTNP) whereas, there has been no difference between slow and normal progressors (NPs) regarding the influence of synonymous substitution (Strunnikova et al., 1995; Bagnarelli et al., 1999). However, Markham and co-workers observed opposite results with greater accumulation of non-synonymous substitutions in NPs (Markham et al., 1998). The event of T-cell activation can be effectively employed to estimate the progression to AIDS in HIV-1 infected individuals and might be utility in apt assessment of viral replication rates and evolutionary complexities. Consequently, activation of immune responses might enforce critical constraints on the evolutionary tricks and viral generation rates of HIV. However, such effects yet remain unexplored from the evolutionary perspective. Extensive investigations pertaining to evolutionary signatures might prove handy in elucidating the patterns of molecular adaptation of HIV in human host. Variations in synonymous substitution rates reflect changes in generation time or mutation rate, while rate of non-synonymous mutations tend to be affected by changes in selective pressure and effective population size (Lemey et al., 2007).

Envelope (env) gene is the fastest evolving one in the HIV-1 genome. Conflicting selective pressures shape the evolutionary dynamics of the virus (Korber et al., 2000; Ross and Rodrigo, 2002; Yoshida et al., 2011). In the present study, investigations pertaining to the evolutionary traits of the env genes representing rapid progressor (RP), slow progressor (SP), and LTNP human patients (available in public domain), have been performed extensively through detailed phylogenetic analysis, followed by subsequent estimation of synonymous and non-synonymous substitution rates. Rapid-progressors are associated with speedy development of AIDS within 3 years of infection whereas; slow-progressors are characterized by a comparatively slower gradual onset of AIDS after seroconversion that might take a span 3–10 years (Kumar, 2013) to develop. Comparative analysis of relative synonymous codon usage (RSCU) patterns of the three categories of env genes marked by varying degrees of disease progression i.e., RP, SP, and LTNP types and their similarity index with human host has also been executed in the present endeavor with a motif to address the adaptive strategies of the viruses for successful residence and infection. Results pertaining to RSCU patterns and evolutionary dynamics might prove utility to unravel the molecular underpinnings of viral adaptation and infection in human host and excavate subtle discrepancies among the viral types associated with varying rates of disease progression.

Intragenomic and intergenomic variations pertaining to codon and amino acid usage patterns can be well explained in the light of multivariate statistical analysis. Correspondence analysis (CoA) is one such type of comprehensive statistical tool which highlights the major variations among codon and amino acid usage data and places them in accordance with such observed variations (Greenacre, 1984). CoA on the basis of RAAU of the three concerned types of env gene sequences revealed their unique amino acid compositional features and separated them according to their diverse compositional traits along the two principle axes of separation of genes i.e., Axis 1 and Axis 2 of RAAU data (Figure 2). SP and LTNP sequences, to certain extent, had identical amino acid usage features as was reflected from COA on RAAU data (Figure 2). However, RP sequences clustered separately along Axis 1 signifying completely diverse amino acid usage patterns from SP and LTNP sequences. Such contrasting features of amino acid composition of RP sequences might have an implication in their rapid rate of disease progression in human host, characterized by a short stay in host cellular environment and speedy proliferation and infection (Kumar, 2013; Jarrin et al., 2015).

Earlier reports suggested that the env proteins of LTNP were subject to purifying selection for their overall mutability and tendencies of variation in LTNP. The impact of stronger purifying selection on env genes of LTNP in contrast to RP might be an outcome of the survival value conferred by adaptive stability of the envelop protein. To further validate our observation, we then fitted linear regression lines between Ka and Ks through the origin assuming that Ka and Ks should both initially be zero at the moment of lineage divergence. The slopes for the lines of whole env genes for LTNP and RP were found to be 0.48 and 0.58, respectively. These results indicated that the increase in Ka with respect to Ks in RP was around 1.2 times faster than in LTNP. Such observations were suggestive of the fact that env proteins of LTNP genes have been subject to more severe evolutionary selection with respect to RP, which eventually has been reflected at a lower level of non-synonymous nucleotide substitution for a given level of synonymous substitution. Similarly, the slope for SP was found to be 0.49, suggesting that the increase in Ka with respect to Ks in SP was around 1.01 times faster than in LTNP. This, in turn, reinforced the hypothesis that unusual functional constraints have been instrumental on the env protein sequences of LTNP. Higher average Ka/Ks value of RP over SP and LTNP sequences (statistically significant) indicated that the evolutionary constraint on the RP sequences has been lesser compared to the impact of evolutionary forces on the SP and LTNP sequences. Such an instance of relaxed selection pressure might confer an added advantage to the RP sequences, associated with rapid infection in a short span of time, in accumulation of non-synonymous mutations and subsequent evasion of host (human) immune response (Khanlou et al., 1996; Jarrin et al., 2015).

HIV-1 shows variation in susceptibility of infection to human (Fellay et al., 2007). Many attempts have been made to understand the viral genetics associated with disease progression in HIV-1 affected patients. However, molecular mechanism and host genetic influence on effective viral pathogenesis and subsequent progression of disease still remain unclear (Michael, 1999; Carrington et al., 2001).

Viruses mimic the codon usage profile of their respective hosts and rely largely on the chaperon apparatus of their hosts for proficient replication and enhanced robustness (Kunec and Osterrieder, 2016). In this pretext, similarity index of a concerned virus with its respective host has been suggested to be an effective tool to properly assess the impact of host genomic influence on the patterns of associated viral codonic signatures. Several earlier studies have used similarity index to demonstrate the influence of host genome over the viral genome (Nasrullah et al., 2015; Butt et al., 2016). Estimation of similarity index with human host among the concerned types of env sequences associated with varying degrees of disease progression i.e., RP, SP, and LTNP signified the fact that the impact of selection by human host was most intense in LTNP compared to SP and RP. Such observations seemed logical in light of the fact that LTNP sequences reside in the human system for long and are characterized by high CD4+ and CD8+ T-cell counts (Zeller et al., 1996; Kumar, 2013). Thus, higher similarity index with human host in LTNP and usage of a higher proportion of identically selected codons with human host tend to be a consequence of prolonged residence and adaptive efficacy of LTNP in human body. In contrast, the instance of lower similarity index in RP with human host might be well explained from the viewpoint that RP stays in human body for a shorter span and causes rapid infection (Khanlou et al., 1996). However, SP sequences displayed higher similarity index with human host in comparison to RP which might be attributed to a comparatively longer residence and slower disease progression than RP (Hogervorst et al., 1995). Thus, observations pertaining to similarity index among the three types of env sequences with human host seemed well correlated with the rates of disease progression and duration of abode in human host.

Decreased HIV-1 strain evolution correlates with weaker viral fitness and the inability to evade the host immune system. Some studies have shown that viral strains from LTNPs are less evolved and thus, less capable of evading the host immunological response when compared with progressor strains (Wang et al., 2003; Sandonís et al., 2009). In our study, Ka/Ks decreased significantly for LTNP env gene sequences compared to RP gene sequences. Similar trend was observed for glycosylation sites of the respective sequences. Findings of the present study confirm that HIV-1 env gene of LTNP have been under stronger purifying selection compared to the respective RP counterpart. The env genes, coding for the env glycoproteins, experience severe selection constraints from the host due to their constant exposure to the host immune system. The env protein mediates HIV-1 entry into host CD4+ T cells. Reduction in Ka/Ks for env genes of LTNP might result in significantly slower infection of CD4+ T cells compared with RP. Our hypothesis seems well justified in light of the fact that the env protein plays essential role in receptor binding and cell entry (Wyatt and Sodroski, 1998) and efficiently maintains extensive glycosylation on its surface to avoid antibody neutralization (Reitter et al., 1998). In this perspective, it might be suggested that env gene evolution occurs mainly by negative selection with the occurrence of mutation that might not reach fixation in the viral population.

Present study addresses an evolutionary perspective that recognizes the ongoing dynamic interplay between human host and genetic diversity of HIV. Our analysis also provides an elucidation of the driving force accountable for the selection of favorable host genetic profiles as well as for viral mutations that are best adapted to the concerned host effects. Outcomes of the current research work confer a vivid portrait about the role of host factors in determining the overall progression of HIV infection.

SB planned and designed the work. AR and RB generated the data. SB, AR, and RB analyzed data and prepared the manuscript. All authors read and approved the final manuscript.