CRF19_cpx is an Evolutionary fit HIV-1 Variant Strongly Associated With Rapid Progression to AIDS in Cuba

Abstract

Background: Clinicians reported an increasing trend of rapid progression (RP) (AIDS within 3 years of infection) in Cuba.

Methods: Recently infected patients were prospectively sampled, 52 RP at AIDS diagnosis (AIDS-RP) and 21 without AIDS in the same time frame (non-AIDS). 22 patients were sampled at AIDS diagnosis (chronic-AIDS) retrospectively assessed as > 3 years infected. Clinical, demographic, virological, epidemiological and immunological data were collected. Pol and env sequences were used for subtyping, transmission cluster analysis, and prediction of resistance, co-receptor use and evolutionary fitness. Host, immunological and viral predictors of RP were explored through data mining.

Findings: Subtyping revealed 26 subtype B strains, 6 C, 6 CRF18_cpx, 9 CRF19_cpx, 29 BG-recombinants and other subtypes/URFs. All patients infected with CRF19 belonged to the AIDS-RP group. Data mining identified CRF19, oral candidiasis and RANTES levels as the strongest predictors of AIDS-RP. CRF19 was more frequently predicted to use the CXCR4 co-receptor, had higher fitness scores in the protease region, and patients had higher viral load at diagnosis.

Interpretation: CRF19 is a recombinant of subtype D (C-part of Gag, PR, RT and nef), subtype A (N-part of Gag, Integrase, Env) and subtype G (Vif, Vpr, Vpu and C-part of Env). Since subtypes D and A have been associated with respectively faster and slower disease progression, our findings might indicate a fit PR driving high viral load, which in combination with co-infections may boost RANTES levels and thus CXCR4 use, potentially explaining the fast progression. We propose that CRF19 is evolutionary very fit and causing rapid progression to AIDS in many newly infected patients in Cuba.

Keywords: CRF19; Cuba; HIV-1; Progression to AIDS; Variant.

Supplementary Fig. 1

Years of inclusion and follow-up for AIDS patients. A) Year of HIV diagnosis and AIDS definition of chronic-AIDS patients. B) Years of follow-up for AIDS-RP and chronic-AIDS patients.

Supplementary Fig. 2

Distribution of geno2pheno[co-receptor] FPRs among V3 loop sequences with phenotypic co-receptor information. A) Boxplots of geno2pheno[co-receptor] FPRs for 170 HIV-1 V3 loop sequences that were published at the LANL HIV sequence database between 2009 and July 2013 and that were classified as R5, dual/mixed or X4 by phenotypic co-receptor testing. The central mark is the median and the edges of the box are the 25th and 75th percentiles. Outliers are plotted individually as crosses. Note that none of these sequences were used in the training data set for geno2pheno[co-receptor]. B) Boxplots of geno2pheno[co-receptor] FPRs for 7446 HIV-1 V3 loop sequences from the LANL HIV sequence database that were classified as R5, dual/mixed or X4 by phenotypic co-receptor testing. Note that this set of sequences also contains sequences that had been used for training geno2pheno[co-receptor].

Supplementary Fig. 3

Phylogenetic analysis of all subtypes. Phylogenetic analyses were performed on 2121 sequences for the analysis of all subtypes (average length of 918 nucleotides). The ML phylogenetic tree was constructed using FastTree and transmission clusters were identified with the program Cluster Picker using as cut-off genetic distances of 0.03, 0.045 and 0.06 with a bootstrap support of 98%. AIDS-RP (red), Chronic-AIDS (blue), non-AIDS (green) and reference sequences (magenta). Two transmission clusters included four patients from the AIDS-RP and chronic-AIDS cohorts. One subtype B cluster involved two patients from the non-AIDS, one patient from the chronic-AIDS and one patient from the AIDS-RP group, while one CRF20_BG cluster included two patients from the non-AIDS group together with two individuals of the AIDS-RP group. None of the patients from our study carrying a CRF19_cpx virus clustered together (Fig. 1C).

Supplementary Fig. 4

Comparison of CD4, viral load, RANTES, MCP-1, IP-10, neopterin and β2 microglobulin according to the disease progression rates (non-AIDS, chronic-AIDS and AIDS-RP). A) CD4⁺ levels at HIV diagnosis (Kruskal–Wallis test, p < 0.0001, with Dunn's multiple comparison test, ****p < 0.0001). B) CD4⁺ levels at sampling (Kruskal–Wallis test, p < 0.0001, with Dunn's multiple comparison test, ***p < 0.001 and ****p < 0.0001). C) Log of viral load at HIV diagnosis (Kruskal–Wallis test, p < 0.0001, with Dunn's multiple comparison test, ****p < 0.0001 and **p < 0.01). D) Log of viral load at sampling (Kruskal–Wallis test, p = 0.0012, with Dunn's multiple comparison test, **p < 0.01). E) RANTES levels at sampling (Kruskal–Wallis test, p = 0.0006, with Dunn's multiple comparison test, **p < 0.01). F) MCP-1 levels at sampling (Kruskal–Wallis test, p = 0.046, with Dunn's multiple comparison test, *p < 0.05). G) IP-10 levels at sampling (Kruskal–Wallis test, p = 0.047). H) Neopterin levels at sampling (Kruskal–Wallis test, p = 0.020, with Dunn's multiple comparison test, *p < 0.05). I) β2 microglobulin levels at sampling (Kruskal–Wallis test, p = 0.0012, with Dunn's multiple comparison test, **p < 0.01).

Supplementary Fig. 5

Correlation between viral load at diagnosis and sampling and estimated evolutionary fitness at sampling (based on the protease sequence) pooling all three study groups (non-AIDS, chronic-AIDS and AIDS-RP). A) Estimated evolutionary fitness vs viral load at HIV diagnosis. B) Estimated evolutionary fitness vs viral load at sampling. Pearson's correlation coefficient.

Supplementary Fig. 6

Comparison of MIP-1β and IL-8 levels according to the disease progression rates (non-AIDS, chronic-AIDS and AIDS-RP). A) MIP-1β levels at sampling (Kruskal–Wallis test, p = 0.51). B) MIP-1β levels at sampling (Kruskal–Wallis test, p = 0.52).

Fig. 1

CRF19_cpx is associated with rapid progression to AIDS. A) HIV subtype distribution among the three studied groups: non-AIDS, chronic-AIDS and AIDS-RP, classified according to the disease progression rates (see Patients and methods). B) Variables found to be significantly associated with AIDS-RP (AIDS-RP versus non-AIDS + chronic-AIDS) or non-AIDS (non-AIDS versus chronic-AIDS + AIDS-RP) patients in univariate analysis were included in the BN model. A central role for CRF19_cpx in rapid progression to AIDS was demonstrated, showing direct influences (arcs) between AIDS-RP, CRF19_cpx, oral candidiasis, higher RANTES, not always using condoms and not subtype B (red contours and arcs). The stability of the dependency was assessed with a non-parametric bootstrap (100 × replicates). All arcs with bootstrap over 35% are depicted in the network. All arcs with dashed lines represent antagonistic dependencies. C) The ML phylogenetic tree for subtype CRF19_cpx (pol region) was constructed using FastTree. CRF19-cpx-infected AIDS-RP (red), CRF19_cpx reference sequences and all other CRF19_cpx sequences retrieved from Los Alamos database (black), all other available CRF19_cpx sequences from IPK (black), and subtype B Los Alamos reference sequences (blue). One of the nine CRF19_cpx sequences was incomplete for pol region and not included in the ML phylogenetic tree analysis. D) Mosaic structure of the CRF19_cpx genome as taken from the Los Alamos HIV Sequence Database (http://www.hiv.lanl.gov/content/sequence/HIV/CRFs/breakpoints.html#CRF19). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

Comparison of viral load, RANTES, estimated evolutionary fitness, oral candidiasis and co-receptor use prediction among the most frequent HIV subtypes in our study (non-AIDS, chronic-AIDS and AIDS-RP groups). A) Viral load (log copies/ml) at HIV diagnosis (One-way ANOVA, p = 0.0002, with Holm–Sidak's multiple comparison test, *p < 0.05 and ***p < 0.001). B) Viral load (log copies/ml) at sampling (One-way ANOVA, p = 0.0043, with Holm–Sidak's multiple comparison test, *p < 0.05). C) Evolutionary fitness estimated from protease sequence (log of fitness value, logF) at sampling (Kruskal–Wallis test, p < 0.0001, with Dunn's multiple comparison test, ****p < 0.0001). D) Proportion with oral candidiasis at sampling (Chi-square test, p = 0.0005). E) RANTES levels (pg/ml) at sampling (Kruskal–Wallis test, p = 0.038, with Dunn's multiple comparison test, *p < 0.05). F) Proportion of co-receptor use at sampling (Chi-square test, p = 0.0051).

Fig. 3

Comparison of viral load, RANTES, estimated evolutionary fitness, oral candidiasis and co-receptor use between CRF19_cpx and all other HIV subtypes among AIDS-RP. A) Viral load (log copies/ml) at HIV diagnosis (Mann Whitney test, *p < 0.05). B) Evolutionary fitness estimated from protease sequence (log of fitness landscape, logF) at sampling (Mann Whitney test, ***p < 0.001). C) Viral load (log copies/ml) at sampling (Mann Whitney test, p = 0.33). D) Proportion with oral candidiasis at sampling (Chi-square test, p = 0.013). E) RANTES levels (pg/ml) at sampling (Mann Whitney test, *p < 0.05). F) Proportion of co-receptor use at sampling (Chi-square test, p = 0.014).