Increased predominance of HIV-1 CRF01_AE and its recombinants in the Philippines

Abstract

The growth rate of new HIV infections in the Philippines was the fastest of any countries in the Asia-Pacific region between 2010 and 2016. To date, HIV-1 subtyping results in the Philippines have been determined by characterizing only partial viral genome sequences. It is not known whether recombination occurs in the majority of unsequenced genome regions. Near-full-length genome (NFLG) sequences were obtained by amplifying two overlapping half genomes from plasma samples collected between 2015 and 2017 from 23 newly diagnosed infected individuals in the Philippines. Phylogenetic analysis showed that the newly characterized sequences were CRF01_AE (14), subtype B (3), CRF01/B recombinants (5) and a CRF01/CRF07/B recombinant (1). All 14 CRF01_AE formed a tight cluster, suggesting that they were derived from a single introduction. The time to the most recent common ancestor (tMRCA) for CRF01_AE in the Philippines was 1995 (1992-1998), about 10-15 years later than that of CRF01_AE in China and Thailand. All five CRF01/B recombinants showed distinct recombination patterns, suggesting ongoing recombination between the two predominant circulating viruses. The identification of partial CRF07_BC sequences in one CRF01/CRF07/B recombinant, not reported previously in the Philippines, indicated that CRF07_BC may have been recently introduced into that country from China, where CRF07_BC is prevalent. Our results show that the major epidemic strains may have shifted to an increased predominance of CRF01_AE and its recombinants, and that other genotypes such as CRF07_BC may have been introduced into the Philippines.

Keywords: DRMs; HIV-1; NFLG; Philippines; subtype; tMRCA.

Fig. 1.

Phylogenetic analysis of near-full-length genome sequences. Newly obtained NFLG sequences from 23 HIV-1-infected individuals in the Philippines were aligned together with reference sequences from the HIV-1 sequence database (www.hiv.lanl.gov). The phylogenetic tree was constructed using the neighbour-joining method and Kimura two-parameter model. The scale bar represents 0.02 nucleotide substitutions per site. Asterisks indicate bootstrap values in which the cluster to the right is supported in 80 % or more replicates (out of 1000). The newly characterized viral sequences are shown in red, and other subtype reference sequences in black.

Fig. 2.

Recombination breakpoint analysis of NFLG sequences. Recombination breakpoints of the newly obtained NFLG sequences were determined using similarity plot and jpHMM. The recombination pattern for each NFLG sequence is shown using RecDraw. Sequences CRF01_AE, subtype B and CRF07_BC are indicated by orange, blue and brown boxes, respectively.

Fig. 3.

Phylogenetic analysis of new NFLG sequences with additional reference sequences. The 23 new NFLG sequences from the Philippines were aligned with additional CRF01_AE, CRF07_BC, CRF08_BC, subtype B and other reference sequences. The phylogenetic tree was constructed using the neighbour-joining method and the Kimura two-parameter model. The scale bar represents 0.02 nucleotide substitutions per site. Asterisks indicate bootstrap values in which the cluster to the right is supported in >80 % of replicates (per 1000). Newly characterized sequences in the Philippines are shown in red. The recombinant sequences are indicated by sample ID, and the sequence derived from Japan is indicated by JP. Sequences within a smaller cluster were collapsed and are shown as a triangle.

Fig. 4.

Phylogenetic analysis of partial pol and env gene sequences. Partial pol (left) and env (right) sequences available from the Philippines in the HIV sequence database were analysed, together with the newly characterized NFLG sequences and reference sequences. Since the sequences in each region were from different studies and did not fully overlap, neighbour-joining trees were constructed to include as many sequences as possible and to maximally utilize the sequence length for all available sequences. The CRF01_AE sequences from this study and from others are indicated in red and blue, respectively. Reference sequences are shown in black.

Fig. 5.

Identification of CRF07_BC-like sequences in the 1023 genome. Three subtype C-origin fragment sequences (a, b and c) in the 1023 genome were aligned together with seven additional CRF07_BC reference sequences. Phylogenetic trees were constructed using the neighbuor-joining method and the Kimura two-parameter model. The position of each recombinant region based on itse location in the HXB2 genome is indicated at the bottom of the tree. The scale bar represents 0.01 nucleotide substitutions per site. The sequences from 1023 are shown in red.

Fig. 6.

Estimated time of introduction of CRF01_AE in the Philippines. Maximum-clade credibility trees were generated for the subtype CRF01_AE NFLG sequences (red) using the Bayesian MCMC approach implemented in beast1.8.2. Each Markov chain Monte Carlo (MCMC) analysis was run for 50 million steps and sampled every 10 000 states. Posterior probabilities were calculated with a 10 % burn-in and checked for convergence using Tracer v1.6. FigTree 1.4.2 was used for visualization of the annotated trees. The mean time and 95 %HPD of the most common ancestor (tMRCA: year) are shown for the key notes based on relaxed (non-correlated log-normal) molecular clocks under HKY nucleotide substitution models in a gamma-distribution of among-site rate heterogeneity with four rate categories (HKY+γ4). All posterior probability values for key nodes are 1.0.