Antiretroviral resistance, genotypic characterization and origin of Human Immunodeficiency Virus among the infected wives of Intravenous drug users in Manipur

Increasing incidence of drug resistance is ascertained to be the main obstacles in limiting the virus among the human immunodeficiency virus (HIV) infected individuals. This study investigates the drug resistance mutations (DRMs), genetic variants and origin of transmitted drug resistance of HIV-1 among the HIV-1 infected wives of intravenous drug users (IDUs) in Manipur. 44 HIV pol gene sequences were generated from 56 blood samples by viral gene amplification and sequencing. Sequences were then analysed for drug resistance, genetic variants and origin. The result revealed that among the treatment naive cases, 35.7% had Transmitted Drug Resistance Mutations (TDRMs) while among treatment experienced cases, 50% had Acquired Drug Resistant Mutations (ADRMs). These TDRMs and ADRMs conferred resistance to nucleoside reverse transcriptase inhibitors (NRTIs), non-nucleoside reverse transcriptase inhibitors (NNRTIs) and/or protease inhibitors (PIs). Majority of the isolated HIV-1 sequences (77.3%) were subtype C while 9.1% was discordant subtype, 6.8% was subtype B, 4.5% was CRF_01AE and 2.3% was URF_BC. TDRM strains were found to be introduced from Myanmar, Vietnam and mainland India. This study also reveals the appearance of CRF_01AE for the first time in Manipur. The finding of this study indicates high prevalence of drug resistant mutations and complex molecular epidemiology in Manipur.

to guide the patient for timely switching to second-line ART, to avoid severe clinical outcomes, and to prevent onward HIV transmission including resistant virus 10 . Previously we have conducted a study to investigate the HIV-1 drug resistance among the ART experienced HIV individual in the region to understand the genetic basis of drug resistance. The data revealed that 53% of HIV-1 infected individuals residing in Manipur harbored drug resistance mutations at different drug resistance positions 11 . Nevertheless, there is no data on the wives of IDUs even though there is clear evidence of HIV/AIDS expansion from IDUs to general population through them. Wives of IDUs are gradually accounting for a larger proportion of HIV infections and ultimately becoming the new face of the HIV epidemic. Therefore, this current study is conducted to investigate the TDR and ADR (to the NRTIs, NNRTIs and PIs drug classes), genetic variant and origin of the TDR among the HIV-1 infected wives of IDUs residing in Manipur state during the year 2016-2017.

Results
Clinical characteristics of study participants. Table 1 show the clinical characteristics of participants. 56 HIV-1 infected wives of IDUs were enrolled for the study during August 2016 and September 2017 (Supplementary Table S1). All of them were infected with HIV from their spouses probably through sexual mode. Among the 56 recruited individuals, 18 had never been experienced to ART drugs while remaining 38 individuals had experienced ART for more than 6 months to the following ART drug combinations at the time of sample collection: ZLN; Zidovudine + lamivudine + nevirapine, (AZT + 3TC + NVP) and TLE; Tenofovir + Lamivudine + efavirenz (TDF + 3TC + EFV) as a fixed-dose combination but none started a PI-based regimen. The CD4 T cell counts of the participants were ≤200 cells/μl for 14.3%, between 201 and 400 cells/μl for 48.2%, and ≥401cells/μl for 37.5%. The average CD4 count among the ART naïve and ART experienced individuals were 220 cells/μl and 350 cells/μl respectively. The percentage of each category of particular factors (age, CD4 counts, year of HIV diagnose, ART treatment status, participants' level of education, etc.) is summarized in Table 1. From the 56 blood samples, a total of 44 samples were amplified and sequenced with a success rate of 78% (44/56). The failure of PCR amplification or sequencing (22%) might be due to the degradation of samples, poor transportation and/or sequencing primer specificity.
HIV-1 genetic characterization. 44 pol nucleotide sequences (33 PR and 44 RT sequences) were obtained from 56 blood samples by viral gene amplification, sequencing. The newly generated sequences were submitted to the GenBank under GenBank accession numbers: MG251453-MG251529. Out of the 44 pol nucleotides sequences, 14 were obtained from treatment naïve individuals. Using either PR and/or RT regions of HIV pol gene, 44 samples were genotyped successfully. Based on the phylogenetic bootstrap consensus tree constructed with 33 PR sequences and reference sequences, four samples Manipur021_PR, Manipur037_PR, Manipur049_PR and Manipur055_PR formed cluster with subtype B group, one sample Manipur005_PR with CRF_01AE and another sample Manipur051_PR was identified as recombinant BC while the rest of samples were cluster with subtype C (Fig. 1a). The analysis of independent phylogenetic trees of 44 RT sequences revealed that two samples (Manipur005_RT and Manipur036_RT) were CRF_01AE, 3 samples (Manipur037_RT, Manipur049_RT and Manipur055_RT) were formed cluster with subtype B, and five samples were intersubtype recombination BC while remaining sequences were identified as subtype C (Fig. 1b). This result is consistent when the samples were genotyped with REGA/COMET. BC intersubtype recombination and CRF_01AE at RT gene were also confirmed by RIP, jPHMM (Supplementary Fig S1) and bootscanning in Simplot software ( Supplementary Fig. S2). The analysis of PR region revealed that 81.8% was subtype C, 3% was URF_BC, 12.2% was subtype B and 3% was CRF_01AE while analysis of RT region showed that 77.3% was subtype C, 11.4% was URF_BC, 6.8% was subtype B and 4.5% was CRF_01AE (Fig. 2a,b).
Finally, based on PR and RT regions, 34 out of 44 samples (77.3%) were subtype C group while 4 (9.1%) were discordant subtype, 3 (6.8%) were subtype B, 2 (4.5%) were CRF_01AE and 1 (2.3%) was URF_BC (  Phylogenetic trees were generated from the sequenced HIV-1 PR (a) and RT (b) genes together with the corresponding viral gene of reference. The tree was inferred using the Neighbor-Joining method. The optimal tree with the sum of branch length = 2.5 is shown. The tree was computed using the Kimura 2-parameter method and are in the units of the number of base substitutions per site. The rate variation among sites was modeled with a gamma distribution (shape parameter = 1). The phylogenetic trees analyses were conducted in MEGA7. Manipur sequences were indicated with solid red triangle, Drug resistance of the ART naïve and ART treatment experienced individuals were represented with solid circle and solid stars respectively.
Among the remaining 30 ART treatment experienced individuals, 15 (50%) individuals harbored HIV-1 drug resistance strains that conferred resistance to NRTIs, NNRTIs and/or PIs (Fig. 3a). The results further showed that 6 (20%), 12 (40%) and 4 (13.3%) individuals had DRMs at the target sites for NRTIs, NNRTIs and PIs respectively (Fig. 3b). Twenty-five different types of ADRMs in fifteen individuals have been defined based on the mutation

Origin of transmitted drug resistance sequences of Manipur HIV-1. The maximum likelihood (ML)
tree of the five TDRM reverse transcriptase sequences is shown in Fig. 4. The strains from Manipur (Manipur005_ RT, Manipur017_RT, Manipur021_RT, Manipur034_RT and Manipur056_RT) are highlighted in red colour and those from other countries are indicated with different colour in the taxa. The sequence (Manipur005_RT) is phylogenetically related and branched within high support with CRF_01AE strains isolate from Vietnam and Czech Republic (Europe) (Fig. 4a). The sequence (Manipur017_RT) is branched with the other BC strains from China, but significant positioned within a highly supported monophyletic clade that was composed by BC isolates from Myanmar (Fig. 4b).
Manipur021_RT is formed well-supported monophyletic clusters with recombinant BC sequences of Myanmar and China (Fig. 4c). The sequence Manipur034_RT is closely related with all subtype  (Fig. 4d).
The phylogenetic analysis of Manipur056_RT (subtype C) sequence showed a close relationship with subtype C strain from India (Fig. 4e)  and/or protease inhibitors (PIs) drugs which hinder before and during treatment of HIV. When we compared the prevalence of NRTIs, NNRTIs and PIs related mutations in ART naive and ART experienced individuals, the findings of this study suggest that ART naïve individuals had harbored HIV with 35.7% DRMs while 23.3% DRMs among the ART experienced individuals in NRTIs drug target site (Fig. 3b). Nevertheless, In NNRTIs drug target site, 21.4% DRMs among the ART naïve as compared to 40% DRMs in ART experienced individuals (Fig. 3b). While in PIs drug target site, ART naïve individuals has harbored HIV with 7.1% mutation compared to 13.3% among the ART experienced individuals (Fig. 3b). Therefore this indicates that the levels of drug resistance in RT region were high in the Manipur state. Further, M184V was identified to be the most frequent mutation among the ART naïve (21.4%) and ART experienced (10%) individuals conferring resistance to NRTIs. Interestingly, M184V could reduce pathogenicity of HIV, and increase the susceptibility to other drugs targeted at NRTI sites 13,14 . Among the ART naïve patients, most of the HIV-1 sequences have resistance against the anti-viral drugs; stavudine (d4T) while among the treatment experienced samples, resistance against the efavirenz (EFV), nevirapine (NVP) and rilpivirine (RPV) were frequency detected (Fig. 3c). The overall finding of this study has raised the necessity for genotyping the drug resistance mutation to assist the clinicians in selecting the potent antiretroviral drug regimens that will enhance the appropriate treatment responses. Further, the findings reveal that the majority of the isolated HIV-1 sequences (77.3%) were subtype C while 9.1% was discordant subtype, 6.8% was subtype B, 4.5% was CRF_01AE and 2.3% was URF_BC (Fig. 2e). The prevalence of subtype C in Manipur is almost consistent with previous finding. However, this study also shows the appearance of circulating recombinant form CRF_01AE for the first time after three decade of HIV prevalence in Manipur. The viral CRF_01AE strain was highly prevalent in the neighbouring countries viz China, Myanmar and other south-east Asian countries. Previously Chen R. et al. 15 has reported that CRF01_AE viruses had highly drug resistant RT which significantly influences on HIV-1 mutation frequencies. Therefore, appearance of CRF_01AE for the first time shows the changing molecular epidemiology and evolution of drug resistance strain.

Discussion
Further, we investigated the origin of five HIV-1 transmitted drug resistance strains circulating among the wives of IDUs of Manipur. The BLAST search of Manipur005_RT mainly retrieved CRF_01AE sequences from Thailand (51%), Japan (15%), China (9%) and Indonesia (7%) while the BLAST similarity search of the samples; Manipu017_RT and Manipur021_RT retrieved BC sequences mainly isolated in China (94%). The phylogenetic analysis Manipur034_RT and Manipur056_RT sequence showed a close relationship between Manipur and mainland Indian samples (Fig. 4d,e). The finding is consistent when we analyze four HIV-1 sequences without    Fig. S3). The URF_BC variant of Manipur was found to be close-associated with BC strain from Myanmar, the subtype C variant was found to be introduced from mainland Indian while the CRF_01AE recombinant has traced its origin from Vietnam. The results suggested that samples which are identified as URF_BC, CRF_01AE and subtype C among the transmitted drug resistant strains were found to be introduced from multiple directions; Myanmar, Vietnam, and mainland India respectively (Fig. 5).
The reason for prevalence of complex HIV-1 genetic variant including CRF_01AE in Manipur may be explained due to the geographic location near the south East Asia. It is also noteworthy to mention that certain unique and unclassified mutations which could impact the ART treatment regime were also detected among the study population. Future research on the identification and characterization of the same could be proved beneficial in constantly monitoring the response of individuals to achieve optimal therapeutic outcome. Nevertheless, the study has some limitation. First, as the study focuses on the recently diagnosed among the wives of IDUs in Manipur, the sample size of the study individuals are limited. Moreover, the numbers of ART naïve individuals are also small as compared to the ART experienced individuals. Second, the study focused on data from the PBMC of blood donors from HIV infected wives of IDUs, the rates of TDRMs reported cannot necessarily be concluded to other populations of HIV infected blood donors. Despite these limitations, the current study of HIV-1 drug resistance mutation, genetic diversity and origin of TDR among the HIV infected wives of IDUs in Manipur shows the appearance of CRF_01AE even though subtype C is most frequently found virus. This is the first report of detecting CRF_01AE in Manipur. The study also reveals the high prevalence of drug resistance mutations to NRTIs and NNRTIs among the HIV-1 infected wives of IDUs in Manipur. HIV drug resistant mutations in these individuals represent a great challenge for the future of the ART program. Therefore, our finding suggested that ART programs and the monitoring system for the effectiveness of ART need to be improved. Patients on ART should be constantly monitored for treatment failure and other classes of ART drugs such as protease inhibitors should be included in the regimen for ART-failed patients after the first-line regimen.

Material and Methods
Study design, subjects and inclusion criteria. This community-based cross-sectional study was conducted during August 2016 to September 2017 in Manipur, India. The participants were enrolled depending on their suitability with the inclusion and exclusion criteria. Only the HIV-1 infected wives of IDUs residing in Manipur state were enrolled for this study. Regardless of patient ART status, participants were included in this study. Based on the above mentioned inclusion and exclusion criteria, 56 HIV infected wives of IDUs were finally recruited to obtain the objective of this study. A personal inquiry to the official staffs of Manipur AIDS control Society (MACS) revealed that our sampling size represents more than 10% of the HIV infected wives of Manipur state annually.     [17][18][19][20] and phylogenetic tree with HIV-1 reference sequences of subtypes A-D, F-H, J, K and CRF_01AE retrieved from the Los Alamos HIV database (www.hiv.lanl.gov). Reference CPZ sequences were also retrieved from HIV database and used as outgroup in the phylogenetic tree analysis. The phylogenetic analyses were performed with the bulk of sequences using the neighbor-joining model based on the Kimura 2-parameter distance matrix and substitution includes transition + transversion using MEGA software (Version7.0) 21 . Tree topology was tested by bootstrap analysis with 500 replicates. The recombination structures were further detected using recombination identification program (RIP) 22 and confirmed by bootstrapping in Simplot 3.5.1 software 23 .

Transmitted and acquired drug resistance analysis of HIV-1. The Transmitted Drug Resistance
Mutations (TDRMs) and Acquired Drug Resistance Mutations (ADRMs) were identified using the Calibrated Population Resistance (CPR) tool (http://cpr.stanford.edu/cpr.cgi) 24 and the Genotypic Resistance Interpretation Algorithm of the HIVdb program (http://sierra2.stanford.edu/sierra/servlet/JSierra) respectively 25,26 , both available through the Stanford University HIV Drug Resistance Database. In accordance to the World Health Organization (WHO) guidelines, the presence of one or more major resistance mutations to any drug class in treatment-naïve patients was considered as TDRM. The HIVdb program was also used to infer the resistance profile of the HIV-1 sequences. The level of resistivity against antiretroviral drugs; protease inhibitors (PI), nucleoside and non-nucleoside reverse transcriptase inhibitors (NRTI and NNRTI respectively) was estimated according to the HIVdb Interpretation Algorithm version 6.0.11 (Stanford University, Palo Alto, CA, USA) 27 .
Determination of the origin of TDRMs HIV-1 variants of Manipur. The determination of most probable geographical origin of the transmitted drug resistance genetic variant circulating in Manipur was performed as described by Mendoza Y et.al. 28 . Each TDRM Manipur sequence was aligned with the 52 HIV-1 sequences isolated world-wide with the highest Basic local alignment search tool (BLAST) search (http://www.hiv.lanl. gov/content/sequence/BASIC_BLAST/basic_blast.html) similarity score (>94) using clustal W and subject to Maximum Likelihood (ML) phylogenetic analysis using the GTR + I + Γ nucleotide substitution model. The ML tree was reconstructed with the PhyML program 29 using an online web server (https://www.hiv.lanl.gov/content /sequence/PHYML/interface.html). Heuristic tree search was performed using the SPR branch-swapping algorithm and the reliability of the obtained topology was estimated with the approximate likelihood-ratio test (aLRT) based on the Shimodaira-Hasegawa-like procedure 30

Data Availability
HIV-1 pol (Protease and reverse transcriptase) nucleotide sequences analyzed in this study are available in the NCBI database under the GeneBank accession numbers MG251453-MG251529.