Evidence that HIV-1 CRF01_AE is associated with low CD4+T cell count and CXCR4 co-receptor usage in recently infected young men who have sex with men (MSM) in Shanghai, China

Abstract

Men who have sex with men (MSM) have recently accounted for an alarmingly increasing proportion of HIV-1 transmission in China. In order to investigate the immune status as a result of CRF01_AE infection and CXCR4 co-receptor usage in a young Shanghai-based HIV-1-infected MSM population in Shanghai, 364 HIV-1-infected MSM with average age of 22.7 years old, newly diagnosed between Jan 2009 and Jul 2013 were analyzed for CD4+T cell count, subtyping using phylogenetic analysis, and viral co-receptor tropism using Geno2pheno and webPSSM in combination. A total of 276 individuals were identified as recently infected. Subtype assignment were as follows: 176 (63.8%) CRF01_AE, 77 (27.9%) CRF07_BC, and 23 (8.3%) subtype B. Besides, 24 second-generation recombinant strains were identified. A lower CD4+T cell count at baseline survey was observed among CRF01_AE strain-infected individuals, compared to those who were infected with CRF07_BC (P<0.01). The frequency of baseline CD4+T cell count <200 was higher and the frequency of CD4 T counts >500 lower in CRF01_AE infection than CRF07_BC infection. It is worth noting that 32.4%-40.9% of CRF01_AE strain-infected individuals were predicted to carry CXCR4-tropic viruses whereas none of CRF07_BC and subtype B were found to be as CXCR4-tropic viruses (P<0.001). As could be expected CXCR4 tropism was associated with lower CD4 T counts. This study revealed that CRF01_AE strains with high frequency of CXCR4 tropism are prevailing in the young MSM population in China and could potentially cause a severe loss of CD4+T cell count and rapid disease progression. A regular surveillance of HIV-1 subtypes, CD4+T cell count and viral co-receptor usage would be greatly beneficial for effectively monitoring disease progression, improvement of antiretroviral therapy strategy and prompt intervention of transmission.

Figure 1. Phylogenetic tree analysis of HIV-1 env and pol gene sequences among MSM with recent infections in Shanghai.

The phylogenetic trees were constructed using neighbor-joining methods (Mega 5.0) based on pol (A) and env (B) sequence regions. The bootstrap values of 1000 replicates above 75% are labeled on the major clusters nodes. The CXCR4-tropic strains determined with algorithm I were indicated by both solid and open circles, whereas the CXCR4-tropic strains determined with algorithm II were only indicated by solid circles. CRF01_AE sequences are marked in red, CFR07_BC sequences are marked in green, and subtype B/B’ sequences are marked in blue. U stands for unidentified subtypes/recombinants. The subtype reference sequences from the Los Alamos HIV sequence database (http://hiv-web.lanl.gov/content/index) were indicated by solid triangles. Trees were rooted using group O as a out group.

Figure 2. The levels of CD4+T cell count in Shanghai young MSM infected with HIV-1 CRF01_AE, CRF07_BC, and subtype B.

The statistical significance in levels of CD4+T cell count (Median and Interquartile Range [IQR]) among three different subtypes and recombinants, CRF01_AE, CRF07_BC, and subtype B was calculated using the Two Independent Samples Nonparametric Tests (Man-Whitney U).

Figure 3. A strong association observed between HIV-1 CRF01_AE- and CRF07_BC-infected MSM, based on the analysis of stratified baseline CD4+T cell count.

The statistical significance of the association between HIV-1 CRF01_AE and CRF07_BC infections based on the stratified baseline CD4+T cell count was evaluated using the Fisher’s exact test. The data are shown as the constituent ratio (95%CI).

Figure 4. Association between HIV-1 tropisms and CD4+T cell count using two genotypic algorithms.

The statistical significance of association between CXCR4- and CCR5-tropic strain infections and CD4+T cell count (Median and Interquartile Range [IQR]) was evaluated using the Two Independent Samples Nonparametric Tests (Man-Whitney U). Algorithm I: Geno2pheno (FPR = 10%) and webPSSM in combination, Algorithm II: Geno2pheno (FPR = 5%) and webPSSM in combination. X4 in light-grey square and R5 in dark-grey square were indicated as CXCR4- and CCR5-tropic strains, respectively.

Figure 5. Association between CXCR4-tropic strain infection and CD4+T cell count stratified groups in HIV-1 CRF01_AE infections (n = 176) when using two genotypic algorithm predictions.

The statistical significance of correlation between CXCR4-tropic strain infection and stratified CD4+T cell counts (percentage [95%CI]), when using two algorithms, was evaluated using the Fisher’s exact test.