Genetic Characteristics of HIV-1 CRF12_BF First Identified in Guangdong Province,China

HIV-1 CRF12_BF is a circulating recombinant form of HIV-1 composed of subtype B and subtype F. CRF12_BF was first found in Argentina and Uruguay in 1999 and was assessed to be associated with injection drug users, mother-to-child, and sexual transmission. ¹ CRF12_BF is mainly prevalent in Latin America ² and has also been reported in western and central Europe ³ and in east Asia, ^4,5 but has not been previously reported in mainland China.

Guangdong Province is one of the commercially developed and populous areas located on the southern coast of China, with frequent foreign exchanges and a large transient population. HIV-1 spread quickly in Guangdong Province after the first HIV/AIDS case was identified in 1986. ^6,7 There are several HIV-1 genotypes circulating in Guangdong Province. The predominant genotype in Guangdong Province was CRF01_AE, and subtypes A1, B, C, CRF07_BC, CRF55_01B, CRF08_BC, and CRF59_01B have also been reported. ^8,9

Previous genomic analysis of the partial pol gene indicated that an HIV strain in Guangzhou city, Guangdong Province, might be classified as CRF12_BF (GenBank accession number: MT589616). In this study, we obtained the near full-length genome (NFLG) sequence for this CRF12_BF strain and present the genetic characteristics of this strain.

Serum was collected from a 48-year-old male patient (Patient ID: 10-097) who is a resident of Guangzhou and was diagnosed with HIV-1 on June 22, 2018. This study was approved by the institutional review board of the Guangzhou Eighth People's Hospital, and written informed consent of the patient was obtained before collecting serum samples. After serum collection, the patient was started on first-line antiretroviral drug therapy, comprising tenofovir (TDF), lamivudine (3TC), and efavirenz (EFV).

The RNA template was extracted using the QIAamp Viral RNA Mini Kit (Qiagen, Hilden, Germany) and reverse transcribed to cDNA using SuperScript IV VILO Master Mix (Invitrogen, CA) according to the manufacturer's instructions. The NFLG of HIV-1 was amplified with Platinum SuperFi II PCR Master Mix (Invitrogen) using the same conditions and primers as previously described. ^10,11 PCR products were purified and then sequenced by Tianyi Huiyuan Company (Guangzhou, China) with 28 specific primers and the Sanger sequencing method. Sequencing fragments were assembled by ContigExpress (Vector NTI software) with a minimum overlap of 100 bp. The assembled NFLG sequence was deposited in GenBank (accession number: MT712270).

The NFLG sequence of the HIV-1 isolate 10-097 was 8,835 bp (from 780 to 9,588 bp according to the HXB2 calibrator) in length. To check for potential contamination and to confirm the sequence quality, the sequence was submitted to the HIV-1 Sequence Quality Control Tool (https://www.hiv.lanl.gov/content/sequence/QC/index.html). The percentage of non-ACGT content was zero, no hypermutations were detected, and all of the gene structures were normal with the correct open reading frames.

The sequence was then submitted to the HIV Basic Local Alignment Search Tool (BLAST) available at the Los Alamos HIV Database website (https://www.hiv.lanl.gov/content/sequence/BASIC_BLAST/basic_blast.html) to determine the HIV-1 subtypes. ¹² The results of BLAST analysis identified the virus to be CRF12_BF, having 91% sequence identity with a CRF12_BF sequence submitted from Argentina in 1999 (GenBank accession number: AF332867).

This was further confirmed by a neighbor-joining (NJ) phylogenetic tree in Mega 6.06 software. ¹³ The NFLG sequence was aligned with reference sequences downloaded from the Los Alamos HIV database (https://www.hiv.lanl.gov/content/index), including nine subtypes (A–D, F–H, J, and K) and CRF12_BF, by BioEdit Software. The NJ tree was constructed using the Kimura two-parameter model with 1,000 bootstrap replications. The phylogenetic tree showed that the sequence from the 10-097 strain formed a well-supported (bootstrap value 100%) cluster with the CRF12_BF reference sequences (Fig. 1).

OPEN IN VIEWER

FIG. 1.

The NJ tree was constructed using the NFLG sequences of HIV-1 isolates by the program MEGA 6.0.6. The reference sequences of subtypes A–D, F–H, J, K, and CRF12_BF were downloaded from the Los Alamos HIV database (https://www.hiv.lanl.gov/). Each reference was labeled with the HIV-1 subtype, followed by the sequence name and accession number. The 10-097 strain is labeled with a black dot. The scale bar represents 2% genetic distance. Bootstrap values >70% are presented at the corresponding nodes of the tree. NFLG, near full-length genome; NJ, neighbor-joining.

The patient was divorced and claimed to have a history of obtaining dental fillings abroad. He denied that he had been infected through sexual behavior, and his ex-wife was HIV antibody negative. The patient was negative for syphilis, hepatitis B virus, hepatitis C virus, and tuberculosis. The baseline CD4⁺ and CD8⁺ T cell counts of the patient were 369 and 1,624 cells/μL, respectively, whereas the HIV-1 viral load was 3,500 copies/mL.

To analyze whether mutations associated with drug resistance existed, the pol gene of this strain was submitted to the Stanford University HIV Drug Resistance Database (https://hivdb.stanford.edu/hivdb/by-sequences). The results showed that the E138A mutation conferred resistance against a non-nucleoside reverse transcriptase inhibitor (NNRTI). The drug resistance profile of the patient indicates potential low-level resistance against EFV and low-level resistance against rilpivirine. We then performed drug resistance analysis of other CRF12_BF sequences (partial pol genes and nucleotides 2253-3367 according to the HXB2 calibrator, n = 362) previously submitted to the Los Alamos HIV Sequence database to compare the drug resistance mutations (DRMs) observed in the 10-097 strain with DRMs in other CRF12_BF strains.

Out of 362 sequences, 9 sequences (2.49%) also had the NNRTI mutation E138A. As the patient had no prior exposure to antiretroviral drugs, we speculate that the resistance site of this sequence is derived from transmitted drug resistance. On June 2019, 1 year after highly active antiretroviral therapy (HAART), the CD4⁺ count of the patient had increased to 595 cells/μL, and the HIV-1 RNA was below the low limit of detection (COBAS AmpliPrep/COBAS TaqMan HIV-1 Test v2.0, the quantitative range is 20 to 10,000,000 copies/mL; Roche, Switzerland). Currently, 2 years after HAART, the CD4⁺ count of the patient is 639 cells/μL, and the HIV-1 RNA is still below the low limit of detection. This suggested that the pre-existing mutation E138A may have a rare effect on the TDF/3TC/EFV combination treatment.

To explore the possible origins of the 10-097 strain, further analysis was performed to investigate the phylogenetic relationship of the 10-097 strain with strains from other countries. We downloaded all CRF12_BF sequences from the Los Alamos HIV database. There were 481 CRF12_BF sequences in total, 14 of which were complete genomes, and the other sequences were from the gag, pol, and env genes. The complete genome, pol gene, gag gene, and env gene were selected to maximize the length and the number of segments for further analysis. Segments spanning 8,690 bp of the NFLG (nucleotides 796-9485 according to the HXB2 calibrator) from 14 CRF12_BF patients, 567 bp of the pol gene (nucleotides 2661-3227 according to the HXB2 calibrator) from 378 patients, 422 bp of the gag gene (nucleotides 1616-2038 according to the HXB2 calibrator) from 26 patients, and 329 bp of the env gene (nucleotides 7064-7392 according to the HXB2 calibrator) from 24 patients were included in the final data set.

A total of 378 sequences of the pol gene were aligned with the 10-097 strain and reference F2 sequences by BioEdit Software. The alignment was submitted to the ATGC website to build a maximum likelihood (ML) phylogenetic tree by using the online program PhyML 3.0. ¹⁴ The ML tree was constructed using the general time reversible substitution model with branch support estimated using the approximate likelihood ratio test (aLRT). ¹⁵ The final trees were visualized using the program Figtree 1.4.2 (Fig. 2). The sequences of the complete genome, gag gene and env gene, were analyzed in the same way. As shown in the phylogenetic tree of the pol gene, the sequence from the 10-097 strain clustered with three CRF12_BF sequences from Argentina and formed a better-supported (aLRT branch support value 0.918) cluster with an Argentina sequence (accession number AY365719, Fig. 2). There were no clusters formed with the sequences of the complete genome, gag gene and env gene (data not shown). Therefore, we speculated that the 10-097 strain in this study may have originated from Argentina.

OPEN IN VIEWER

FIG. 2.

Phylogenetic tree analysis. (A) The polar layout of a ML tree was constructed with sequences from the 10-097 strain (labeled with blue line) and CRF12_BF strains from other countries by the online program PhyML 3.0 in the ATGC website. The fragments were partial pol genes (nucleotides 2661-3227 according to the HXB2 calibrator). Subtype F2 was used as the outgroup (labeled with green line). The scale bar represents 4% genetic distance. The sequence from the 10-097 strain and three CRF12_BF sequences from Argentina (labeled with red line) formed a cluster. (B) Partially enlarged rectangular layout of the cluster in A. The aLRT branch support values are indicated at the corresponding nodes of the tree. aLRT, approximate likelihood ratio test; ML, maximum likelihood. Color images are available online.

In this research, we obtained the sequence of one NFLG of CRF12_BF isolated in Guangdong Province. This is the first report on the CRF12_BF of the full-length genome in mainland China, suggesting that a more complex HIV genotype is circulating in Guangdong Province. Although this pre-existing NNRTI-mutated virus was fortunately suppressed after TDF/3TC/EFV combination treatment, surveillance of newly diagnosed patients for evidence of transmitted resistance is still recommended so that the appropriate treatment can be tailored to the patient to prevent the spread of drug-resistant viruses.