Genetic Diversity of HIV-1 in Krasnoyarsk Krai: Area with High Levels of HIV-1 Recombination in Russia

More than a quarter of HIV-infected individuals registered in Russia live in Siberia. Unlike Central Russia where HIV-1 subtype A6 is predominant, in most Siberian regions since 2012, a new HIV-1 CRF63_02A1 genetic variant has spread, with the share of this variant attaining 75–85% among newly identified HIV cases. Krasnoyarsk Krai is considered to be a high-risk territory according to morbidity rate and HIV infection incidence among the population. The current paper aims to study the molecular epidemiologic characteristics of HIV-1 spreading in Krasnoyarsk Krai. Phylogenetic and recombination analyses of pol (PR-RT, IN) and env regions of the virus were used for genotyping 159 HIV-1 isolated in Krasnoyarsk Krai. 57.2% of the isolates belonged to subtype A (A6) specific to Russia, 12.6% to CRF63_02A1, and 0.6% to CRF02_AGСА, and in 29.6% HIV-1 URFs were detected, including URF63/А (23.9%), URFА/В (4.4%), and URF02/А (1.3%). In 6 of 7, HIV-1 URFА/В identical recombination model was detected; the origin of 38 URF63/А was proven to be the result of individual recombination events. Since 2015, a share of the population with newly diagnosed HIV who were infected with HIV-1 URF reached an exceptionally high rate of 38.6%. As distinct from adjacent Siberian regions, the HIV-1 CRF63_02A1 prevalence rate in Krasnoyarsk Krai is within 16%; however, the increased contribution of new HIV-1 into the regional epidemic development was observed due to the recombination of viruses of subtypes А, В, and CRF63_02A1. The difference between the described molecular epidemiologic picture in Krasnoyarsk Krai and in adjacent areas is likely caused by differences in predominant routes of HIV transmission and by more recent HIV-1 CRF63_02A1 transmission in the PWID group, which had a high prevalence of HIV-1 subtype A by the time of the new virus transmission, resulting in increased possibility of coinfection with various HIV-1 genetic variants.


Introduction
HIV infection epidemic in Russia continues to evolve [1], and HIV infection cases have been registered in all territorial entities of the Russian Federation. e number of Russian regions with a high prevalence of HIV infection (more than 0.5% of population size) reached 34 in 2018. Siberia (Siberian Federal District, SFD) significantly contributes to epidemic development, where more than a quarter of HIV-infected population of Russia is registered. Five SFD regions make the list of top 10 Russian territories with the highest HIV infection rates. In 2018, Krasnoyarsk Krai ranks 8 th in Russia and 5 th in SFD according to this index [2,3], with adjacent Kemerovo Oblast, Irkutsk Oblast, Novosibirsk Oblast, and Tomsk Oblast being on the top of the list (Figure 1).
Krasnoyarsk Krai is the second largest federal subject with an area of 2 339 700 km 2 and 2 876 497 residents, with the urban population accounting for 77.4%. Approximately 80% of the population lives south of Angara on one-tenth of the region area [4]. As reported by Krasnoyarsk Regional Center for Prevention and Control of AIDS, the first HIV infection in the region was officially registered in 1989. Before 1999, sporadic HIV infection cases were identified, and in 1999 HIV infection was detected in 156 people in 14 territorial entities of the region; this number was 3-fold higher as compared to the total number of identified HIVinfected people during ten preceding years. Over the following two years, the manifold increase in the number of newly detected HIV-infected patients was registered as well ( Figure 2). e period from 2002 to 2006 can be considered as a time of epidemic process stabilization, and beginning from 2007 up until 2017 annual increase in registration of new cases had been observed again.
As in the case of other Russian regions, from the late 1990s onwards Krasnoyarsk Krai was demonstrating a sharp increase in the number of people who inject drugs (PWID) resulting from the beginning of large volume of opioids delivery to Russian Federation from Afghanistan through Central Asia countries (so-called North Route). At the same time, the first HIV infection cases were registered among PWID. Common for that time practice of risk behavior (shared use of syringes/needles) led to the fast spreading of HIV among PWID [5,6].
In 1999-2001, a share of PWID among newly identified incidents of HIV infection in Krasnoyarsk Krai exceeded 95%. From 2002, Krasnoyarsk Krai has shown a decrease in the rate of HIV spreading among PWID; at the same time, PWID sexual partners became involved in the development of the epidemic process and this was reflected by the increase of heterosexual (HS) transmission route contribution in HIV spreading.
In 2007-2008, in adjacent Kemerovo Oblast and Novosibirsk Oblast, the second period of 3-to 4-fold increase in the number of new HIV infection cases among PWID was observed affecting the situation in Krasnoyarsk Krai. In 2007-2008, an increase in morbidity rate was observed, with the following relative stabilization but with higher quantitative indices of HIV distribution [7]. In 2013-2014, an epidemic outbreak in adjacent Tomsk Oblast was observed with 6-to 10-fold increase in newly registered HIV infection cases in the PWID group; Kemerovo Oblast, Novosibirsk Oblast, Omsk Oblast, Altai Krai, and Krasnoyarsk Krai demonstrated an increase in morbidity rates [8]. Starting in 2010, territory epidemics expansion in SFD has been caused by the beginning of synthetic drugs spreading [9], switching from opioid drugs to the new psychoactive substances (the so-called "synthetics" and "salts") among PWID, and the involvement of large amount of population into the use of such drugs.
In total, 35 779 HIV-infected patients were registered in Krasnoyarsk Krai, and as of 30.06.2018, the HIV infection rate was 953.5 per 100 000 population. e cities of Krasnoyarsk and Norilsk demonstrate the highest HIV incidence rate. Men were predominant among the HIV-infected population in Krasnoyarsk Krai for the duration of the epidemic. A share of men among PWID has been stably registered at the level of 75-78%. In 2004-2007, two-thirds of detected HIV cases among people infected via heterosexual contacts (HS) account for women. Beginning from 2008, the gradual growth of a share of men infected via HS is registered: in 2016, the number of men in that group was equal to the number of women infected via HS. In recent years, 30-to 50-year-old people have mainly contributed to the HIV infection distribution in Krasnoyarsk Krai; in 2018, their number was 71.7% among newly detected HIV-positive patients, of which 60.5% were infected via HS and 36.5% were PWID. Studies by Bobkov et al. demonstrated that epidemic development in Russia, including SFD, in the late 1990s and early 2000s, was caused by HIV-1 subtype A (A6) distribution among PWID [10][11][12][13].
We proposed that the activation of HIV infection distribution in Krasnoyarsk Krai in 2013-2016 may have contributed to changes in the genetic variability of circulating HIV strains in that territory. is study aims to analyze the characteristics of current HIV-1 distribution in Krasnoyarsk Krai.  28,2017). Peripheral blood of the HIV-infected individuals was sampled by the medical staff of the "Krasnoyarsk Regional Center for Prevention and Control of AIDS," with pretest and posttest psychological consultations. Epidemiologic data were collected by a professional epidemiologist from individuals under the study after they were acquainted with the project goals.

Study Participants and Sample
Collection. From September to December 2017, 162 peripheral blood plasma samples were collected from HIV-infected patients in Krasnoyarsk Krai living in Krasnoyarsk (132), Norilsk (12), and other communities of the South of Krasnoyarsk Krai (18) who sought assistance in Krasnoyarsk Regional Center for Prevention and Control of AIDS. Plasma was separated within 3 hours after collection and frozen at −80°C for further use. 125 Krasnoyarsk patients were ART-naïve; 37 patients took APBT medication. e blood samples were linked with demographic and clinical data via coded ID numbers according to the requirements of medical ethics in Russia. e recorded characteristics for patients included their gender, age, most probable route of transmission, dates of the last negative and first positive tests for HIV, drug use, viral load, and CD4 cell count.

Amplification of HIV-1 Gene Fragments and Sequence
Analysis. Viral RNAs were extracted from 250 μl of plasma with RealBest DeltaMag kit (Vector-Best, Russia) according to the manufacturer's recommendations. e RNA was utilized for reverse transcription PCR and nested PCR. Amplification was performed using a lyophilized ready-touse Reverse Transcription Master Mix containing all the components for a single-tube reverse transcription and PCR (Vector-Best, Russia) and an in-house set of primers as described previously [15].
is procedure generates an amplicon of the pol gene encoding the protease-reverse transcriptase (PR-RT, 1400 nt), integrase (IN, 960 nt), and a fragment encoding a part of the major envelope protein, env (732 nt). For several HIV-1 samples, gag (1457 nt) fragment was additionally obtained. After purification, PCR fragments were sequenced by ABI PRISM 3130xl DNA Analyzer (Applied Biosystems, United States).

Sequence Analysis.
All original sequence fragments of the pol and env gene regions were assembled in whole sequences in Sequencher 4.1 software (Gene Codes Corporation, Ann Arbor, MI, United States). e assembled sequences of pol fragments (PR-RT and IN) and env gene fragments were compared to the corresponding reference sequences of various HIV-1 subtypes and recombinant forms extracted from the Los Alamos HIV Sequence Database (http://www.hiv.lanl.gov) using ClustalW Multiple Alignment and BioEdit software 7.2.5 [25].
MEGA v6.0 was used to construct a phylogenetic tree employing the neighbor-joining method based on the Kimura two-parameter model with 1000 bootstrap replicates based on Kimura's two-parameter model [26]. Statistical significance of phylogenetic tree topologies was estimated using bootstrap analysis. Discordant gene regions or outlier positions in the trees were further analyzed using the jumping profile Hidden Markov Model program (jpHMM; http://jphmm.gobics.de).
e similarity between HIV sequences is plotted using SimPlot 3.5.1 software [27] using a 200 nt window with tree construction by the neighborjoining method applying Kimura's two-parameter substitution model. e possible intertype mosaicisms of URFs were screened using the Recombinant Identification Program (RIP, http://www.hiv.lanl.gov/content/sequence/RIP/ RIP.html) and then verified via bootscanning and informative site analysis using the program SimPlot v3.5.1. When determining the most probable geographic origin of novel HIV variants circulating in Krasnoyarsk Krai, each novel URF sequence was aligned with HIV-1 sequences isolated worldwide with the highest identities using the online tool HIV BLAST, available at the LANL HIV database. e V3 nucleotide sequences were analyzed to estimate the genotypic virus tropism and to confirm the phenotypic coreceptor specificity using the online tool Geno2pheno [coreceptor] (G2P, http://coreceptor.geno2pheno.org) [28] and interpreted using the FPR (false positive rate) cutoff values of 10%. V3 sequences were also submitted to Position-Specific Scoring Matrix (PSSM) (http://indra. mullins.microbiol.washington.edu/webpssm) and the sequence codes were additionally checked for positively charged amino acid residues at 11 and/or 25 codons of the V3 loop [29]. e nucleotide sequences of pol gene containing the fulllength protease, integrase, and the first 300 codons of reverse transcriptase gene were submitted to Stanford HIV Drug Resistance Database (http://hivdb.stanford.edu, assayed for the presence of mutations determining resistance to protease, reverse transcriptase, and integrase inhibitors (DR mutations)) [30]. e DR mutations were identified based on the WHO-recommended list of mutations for DR surveillance [31].
2.6. Statistical Analysis. Analysis of qualitative data was carried out using a modification of Fisher's test for contingency tables of mxn dimension. Evaluation of p value was conducted using the Monte Carlo method; the number of simulations during each calculation was one million. R programming language (version 3.4.0) by Rstudio 1.1.442 was used for statistical analyses of the obtained data.

High Diversity of HIV-1 Circulating in Krasnoyarsk Krai.
For 159 (100%) HIV-1 samples, fragments encoding protease-reverse transcriptase (PR-RT), for 112 (70.4%) integrase (IN), and for 108 (57.9%) fragments encoding V3 region of the major envelope protein (env) were obtained. e obtained nucleotide HIV-1 sequences were used for genotyping by constructing phylogenetic trees using the neighbor-joining (NJ) method. Detection of potential recombinant events between virus subtypes was performed using specialized program resources (jpHMM, SimPlot software, RIP). e complex of data obtained during recombination and phylogenetic analyses was used for genotyping of 159 HIV-1 samples ( Figure 3).
It is important to note that Krasnoyarsk subtype A HIV-1 variants are distributed in the phylogenetic tree within subtype A genetic cluster mixing with samples isolated in other Siberian territories. Several HIV-1 subtype A from Krasnoyarsk Krai formed separate subbranches that frequently aggregated viruses isolated from Krasnoyarsk residents PWID and HS patients who probably had sexual contacts with PWID but failed to know about it or to report that their partners might belong to PWID.
Such an example may include separate subbranch of a phylogenetic cluster of subtype A including Krasnoyarsk 53,62,135,39,66,34,43,114,12,26, and 69 ( Figure 3). All those HIV variants were isolated from PWID except for Krasnoyarsk 43 isolated from a man with HIV infection detected in 2017 who reported a probable infection route via HS.
In contrast to HIV-1 subtype A, more than half of HIV-1 CRF63_02A1 from Krasnoyarsk Krai formed a separate subbranch ( is CRF63_02A1 branch specific to Krasnoyarsk Krai also comprises a substantial number of HIV-1 isolates that were obtained from 2014 to 2016 in Tomsk Oblast from PWID who were using synthetic drugs. ese data point to probable HIV-1 CRF63_02A1 transmission from Tomsk Oblast. Among Krasnoyarsk Krai HIV-1, which group together with viruses CRF63_02A1 isolated from Novosibirsk Oblast residents, in one case the virus was detected in PWID who arrived from Kemerovo Oblast adjacent to Novosibirsk Oblast (Krasnoyarsk 10), and in the second case in a man with HIV detected in August 2017, who reported HS (Krasnoyarsk 117).
Out of the remaining eight HIV-1 CRF63_02A1, which did not noticeably group with other HIV-1, seven were isolated from people infected via HS, and one from a patient infected via contact with PWID.
HIV-1 CRF02_AG SA genetic variant was detected in one case (Krasnoyarsk 72) in 37-year-old PWID from the city of Norilsk with HIV detected in 2002.
For URF02/A, the second ancestor form was HIV-1 CRF02_AG CA [32,33]. In one case, URF02/A virus was isolated from PWID from the city of Norilsk with HIV In most cases, PR-RT regions of HIV-1 URF63/A demonstrated the mosaic structure of genome with significantly differed length and location of interchanged regions identical with HIV-1 subtype A6 and CRF63_02A1 sequences.
ese data indicate that the origin of HIV-1 URF63/A genetic variants is the result of separate recombination events ( Figure 4).
In some cases, we observed a combination of HIV-1 URF63/A in separate subbranches with a high bootstrap, for instance, HIV-1 samples Krasnoyarsk 54 and 56. However, the recombination analysis of HIV variants in the PR-RT region showed differences in the recombination model, and in the phylogenetic analysis carried out for the virus IN region HIV-1, Krasnoyarsk 54 grouped with HIV-1 CRF63_02A1, and Krasnoyarsk 56 grouped with viruses of subtype A6 (Figures 5 and 6).   .3) † HIV-1 genotype: the result of HIV-1 genotyping obtained after analysis of the complex of data: phylogenetic analysis from RP-RT, IN, and env regions and additional recombination analysis in case of inconsistency of genotyping results; † † gradation "homosexual contacts" was excluded from the analysis due to a low number of observations; * p < 0.01. * * PWID: people who inject drugs. 6 BioMed Research International Additional sequencing and analysis of gag gene for HIV-1 URFA/B made it possible to conclude that those virus variants differed from previously described HIV-1 CRF03_AB [34] since, unlike CRF03_AB viruses, they have an additional insertion of subtype B sequence similar to HIV-1 subtype A gag region ( Figure 8). All HIV-1 URFA/B demonstrate a similar structure of gag-pol region (PR-RT) but differ in IN and env sequences: phylogenetic analysis showed that 6 of 7 samples grouped with HIV-1 subtype A (A6) by corresponding genome regions, while Krasnoyarsk 100 sample belonged to URF63/A according to the IN region structure (nucleotide sequence is partly identical to SRF63_02A1 sequence and partly to A6 subtype), and phylogenetic analysis of env region of Krasnoyarsk 100 showed that it belongs to HIV-1 CRF63_02A1 phylogenetic cluster ( Figures 5-7).
URFA/B was isolated in four Krasnoyarsk patient cases with two of the patients belonging to PWID with HIV
Decrease in sensitivity to protease inhibitors (PIs) was not detected. We analyzed primary or transmitted HIV-1 drug resistance for 122 ARVT-naïve patients. In this sample of patients, we detected mutations resulting in the development of different-level HIV-1 resistance in 8.2% (10/122) (Table 3). Transmitted drug resistance was detected in 4 patients infected via HS and 6 PWID patients. As for the virus protease inhibitors, in one case, mutation L76V (0.8%) was found to belong to main PI-resistance mutations. L76V is a          nonpolymorphic mutation, which reduces sensitivity to PIs and FPV and NFV; it increases sensitivity to ATV, SQV, and TPV. In the other case, we detected relatively nonpolymorphic mutation M46I. M46I in combination with other PI resistance mutations is associated with reduced sensitivity to each of the PIs except for DRV. Among mutations associated with the resistance to NNRTIs, there were mutations causing significant HIV sensitivity reduction to the following drugs: K103N causing high-level resistance to NVP and EFV (3/122, 2.5%); K101E (1/122, 0.8%) associated with intermediate resistance to NVP and RPV and low-level resistance to EFV and ETR; M230L (0.8%) causing intermediate to high-level resistance to each of the NNRTIs; and V108I (1.6%) resulting in low-level resistance to NVP and potentially low-level resistance to EFV. K103N was found in three women infected via HS with HIV detected from July to November 2017.
Among mutations impacting HIV sensitivity to NTRIs, in individual cases, we found TAMs M41L, D67E, L210W, and T215S. We disregarded A62V since it is widespread among subtype A viruses in the countries of the former Soviet Union [30].
Among other mutations impacting the development of anti-HIV drugs resistance, we detected E138A (5/122, 4.1%), which is a common polymorphic accessory mutation to NNRTIs; it confers a borderline low-level reduction in RPV sensitivity.
In the case of one PWID man with HIV detected in 2016, Major Resistance Mutation associated with the resistance to INSTI (integrase strand transfer inhibitors) E138K (1/122, 0.8%) was found. Alone it does not reduce INSTI sensitivity. When E138K occurs in combination with Q148 mutations, they are associated with high-level resistance to RAL and EVG and moderate reductions in DTG and BIC sensitivity.

HIV-1 pol Gene Subtype-Specific Polymorphism and Genotypic Prediction of HIV-1 Coreceptor Usage.
A comparative analysis was carried out for pol gene sequences encoding protease, reverse transcriptase, and integrase of HIV-1 CRF63_02A1 and subtype A.
For protease region of viruses CRF63_02A1, the presence of amino acid substitutions K20I and I64M was registered in all cases, whereas only one of the mutations, K20I, was found among HIV-1 subtype A, and it was detected in 2.3% of all cases. Substitution I64M is usual for the CRF02_AG genetic variant. Combination of mutations I64M and G17E in HIV protease is believed to be connected with the development of hypersensitivity to NFV, ATV, and     IDV. K20I is the consensus amino acid in HIV-1 subtypes G and CRF02_AG. Mutation V77I in protease common for HIV-1 subtype A in a number of Russian regions was detected only among HIV-1 subtype A in 60.8%. Substitution A62V is specific for reverse transcriptase of HIV subtype A, in the countries of the former Soviet Union. In the studied sample, A62V was registered in 69.1% of HIV subtype A [35][36][37][38]. In 100% of studied sequences of reverse transcriptase CRF63_02A1, V60I and D121Y polymorphisms were found that were not associated with the resistance and did not occur among viruses of subtype A.
Among HIV subtype A, 10 variants comprised V90I in the region of reverse transcriptase; furthermore, in 8 of 10 cases, HIV with V90I was detected in patients, which have been treated for more than 5 years with confirmed ARVT inefficiency (virological failure), and only in two instances in naïve patients who were infected 1 and 3 years ago.
Subtype-specific mutations were also detected in the HIV integrase region. In the case of subtype A virus, L74I was detected in 100% of the cases, while among CRF63_02A1 this substitution was detected only in 3%. Alone, L74M/I have minimal, if any, effect on INSTI sensitivity, but in combination with mutation Y143H/R/C L74I, causes reduction of HIV sensitivity to raltegravir [39].
Mutation M50I specific to integrase of HIV-1 CRF63_02A1 spreading in Tomsk Oblast and Novosibirsk Oblast (where it is detected in 85-98%) was detected only in 31.3% among HIV-1 circulating in Krasnoyarsk Krai. In this position in circulating HIV-1 in Krasnoyarsk Krai, the substitution M50T was found most frequently-in 56.3%. Data on the connection between the development of resistance to virus integrase inhibitors and mutation M50T are not available.
M50I is a polymorphic mutation selected in vitro by DTG and BIC in combination with R263K. It appears to contribute to reduced DTG sensitivity in combination with R263K [40].
To predict HIV-1 tropism, after two or three re-received nucleotide sequences of V-3 loop for each HIV-1 variant these nucleotide sequences were used for analysis with Geno2pheno [coreceptor] 2.5 software with FPR (false positive rate) equal to 10%. e tropism was successfully predicted for 108 (67.9%) patients. In 91 cases (84.3%), virus tropism was predicted to coreceptor SSR5, and in 17 to SXCR4 (15.74%). Out of 17 SXCR4-tropic HIV-1, 14 belonged to subtype A, two to URF63/A, and one to CRF63_02A1. In 6 cases, HIV

Discussion
Krasnoyarsk Krai is a high-risk territory according to HIV epidemic indices. Despite some stabilization of the epidemic, annual detection of new HIV infection, cases in Krasnoyarsk Krai remain at a high level and increase in HIV incidence is observed among the population. erefore, molecular epidemiologic analysis of current HIV infection spreading in the region conducted in our study is of current importance for characterization of the epidemic situation.
e previous study of HIV-1 isolated in Krasnoyarsk Krai in 2008 revealed that the epidemic development in the region was accompanied by the distribution of the genetically homogeneous HIV-1 subtype A (A6) population [24]. Phylogenetic analysis carried out for HIV-1 isolated in different Siberian regions led to the conclusion that there were at least two independent HIV-1 CRF63_02A1 transmissions to Krasnoyarsk Krai from Tomsk Oblast and Kemerovo Oblast. e complex of data on the detection of phylogenetic clustering and analysis of subtype-specific HIV-1 polymorphism points out that, in Krasnoyarsk Krai, its region specific HIV-1 CRF63_02A1 population of variants with specific differences in genome nucleotide sequences has begun to spread, as well.
In 2015-2017, a remarkably high share of people with newly diagnosed HIV infection and with the detection of newly emerged recombinant HIV (38.6%) was registered in the region.
Furthermore, we detected the distribution of two HIV-1 URF of different origins that differed by one of the ancestral forms: URFA/B and URF63/A. New HIV-1 URFA/B was isolated in 7 cases. In 5 URFA/ B cases, we detected a high level of identity for all studied virus regions. One more HIV from that group was identical to 5 HIV-1 URFA/B in gag-pol region but was genotyped as URF63/A virus by IN region and was clustered with HIV-1 CRF63_02A1 (Krasnoyarsk 100) in env region analysis, which seem to point to the likelihood of the patient reinfection and, consequently, the second recombination between HIV-1 URFA/B and CRF63_02A1 viruses.
HIV-1 URFA/B viruses were isolated in PWID (3), PWID sexual partners (1), and individuals infected via HS (3) living in the city of Krasnoyarsk and other cities of Krasnoyarsk Krai. Considerably restricted (as compared to CRF63_02A1) distribution of this genetic variant suggests that the detected HIV-1 URFA/B, which was first discovered in this study, originated in PWID/or was transmitted by PWID and has been distributed among closed PWID group and their sexual partners. As HIV-1 heterogeneity in Krasnoyarsk Krai has begun to significantly increase since 2012, we carried out additional analysis of HIV-1 genetic variants registered in 2012-2017 among people practicing risk behavior (HS and PWID). Table 4 demonstrates that among people infected via HS, an infection with SRF63_02A1 is registered more frequently, while a share of HIV-1 URF63/A in PWID is higher almost by 10% as compared to the HS group.
Distribution of HIV-1 SRF63_02A1 in Krasnoyarsk Krai is registered in different population strata. Due to the high HIV-1 subtype A incidence among Krasnoyarsk Krai population, HIV reinfection is observed including infection with SRF63_02A1 viruses. It is likely that, in PWID, which is the most affected group by HIV infection compared to other population, HIV reinfection occurs more frequently.
It is notable that HIV-1 SRF63_02A1 distribution in Krasnoyarsk Krai significantly differs from other cases of HIV-1 SRF63_02A1 rapid spreading in Novosibirsk Oblast, Tomsk Oblast, Kemerovo Oblast, and Altai Krai that were described earlier [7,8,22]. In those Siberian regions, HIV-1 subtype A was predominant up to 2008-2012. HIV-1 SRF63_02A1 was spreading in those regions during local epidemic outbreaks that were accompanied by a multifold increase in morbidity among PWID. After that, further spreading of that HIV genetic variant was registered among HS partners of PWID, and currently, HIV-1 SRF63_02A1 is being registered among the whole population. Among newly diagnosed patients, HIV-1 SRF63_02A1 infection is registered in more than 80-85% of cases in these territories. Kemerovo Oblast is the only exception where the distribution of SRF63_02A1 was detected at the level of 74% already in 2014-2016, and in 21% of the cases, HIV-1 URF63/A infection was registered [7].
In the first years of HIV infection distribution and in 2013-2015, the contribution of PWID transmission to HIV spreading was predominant in Krasnoyarsk Krai; in the course of other periods, HIV was predominantly distributed via HS. is dynamic may be a key factor in the model of epidemic process development [41].
To genotype and detect recombinant HIV-1, we studied not the entire genome, but only three HIV-1 genome regions that probably prevented us from the detection of all recombinant viruses in the analyzed sample.
In support of this hypothesis, we point out that in HIV-1 PR-RT region both "hot" and "cold spots" of recombination were described; a number of studies confirmed that analysis of HIV-1 PR-RT region reveals only about 30% of all recombinant events occurring in HIV-1 genome [42,43].
When carrying out recombinant analysis, we also encountered some difficulties. In some cases, recombinant analysis of several deduced HIV-1 sequences was not possible due to insufficient availability of publicly accessible close homologous sequences which are necessary for the analysis. To carry out HIV-1 URF recombinant analysis using SimPlot software, in each case, we preliminary performed search for viral sequences with the most close genetic identity to the analyzed HIV sequences from GenBank (using HIV BLAST software (https://www.hiv.lanl.gov/ content/sequence/BASIC_BLAST/basic_blast.html)) and among HIV-1 sequences that were obtained from the study of HIV-1 heterogeneity in SFD. Additionally, we carried out the comparable procedure of searching for the most genetically close sequences for each HIV-1 URF recombinant segment without recombination joint; however, in some cases no genetically close sequences were found.
Recombination analysis was complicated by the fact that, in Krasnoyarsk Krai, multiple HIV-1 transmissions from other regions are believed to take place including transmission of CRF63_02A1 and possibly URF63/A. As described previously [7,8], the considerably heterogeneous HIV-1 CRF63_02A1 and URF63/A population is currently distributed in SFD. An absence of genetically close HIV-1 to be used for comparison (references) does not allow for performing reliable determination of recombinant HIV-1 origin in some cases.
Besides, in a number of cases, it was extremely complicated to identify possible intra-and intersubtypical recombinant events due to a high level of identity of HIV-1 subtype A and CRF63_02A1 sequences (e.g., in the region encoding virus integrase), since it is subtype A that is one of the ancestral viruses for the newly emerged recombinant virus CRF63_02A1 [22].
Significant differences in the distribution of two HIV-1 CRF: CRF63_02A1 and CRF02_AG SA should be noted. Individual cases of HIV-1 CRF02_AG SA registration have been observed in all Russian regions beginning from the 2000s. Usually, HIV-1 CRF02_AG SA infected women who reported HS contacts with people from Central Asia, while men belong to the PWID group.
Even though HIV-1 CRF02_AG SA is continued to be detected in different Russian regions including cases among PWID, this HIV-1 genetic variant and new URF based on this variant have limited distribution in Russia, except for single recombinant HIV-1 CRF63_02A1 variant that emerged from this virus [22,44]. e analysis of the resistant HIV-1 transmission in Krasnoyarsk Krai revealed infection with HIV-1 carrying mutations associated with the development of the virus resistance to PI, NRTI, and NNRTI in 8.2% of studied cases. Described HIV-1 mutations and the degree of their incidence and analysis of HIV-1 tropism correspond well with data obtained by other researchers for other Russian regions [15,23,[45][46][47][48][49]. Most frequently detected was substitution K103N (3/10), which is DR-mutation to NNRTI. is mutation is the most widespread among HIV-infected Russian patients undergoing ARVT since the first generation NNRTIs were the most commonly prescribed drugs for a long time and they have the lowest genetic barrier for resistance development [50]. Resistant HIV-1 transmission was detected both among patients infected via HS and among PWID.

Conclusions
e study of the molecular epidemiological picture of HIV-1 distribution in Krasnoyarsk Krai enabled us for the first time since 2008 to register considerable changes in the features of the developing epidemic. Although subtype A (A6) remains predominant HIV-1 genetic variant, we detected not only transmission and distribution of previously rare HIV-1 genetic variants in Krasnoyarsk Krai but also the emergence of a large variety of new unique recombinant HIV-1 forms, which originated from viruses of subtypes A, C, SRF63_02A1, and CRF02_AG. Our study demonstrated a high frequency of HIV-1 reinfection for newly diagnosed HIV infection cases among Krasnoyarsk Krai population, suggesting that the region population, including PLHIV, is probably insufficiently informed on the negative impact of HIV-1 reinfection on the progress of the disease. Although general coverage of Krasnoyarsk Krai population with HIV testing is high, targeted HIV testing among people practicing risk behavior seems to be insufficient.
High HIV-1 prevalence among Krasnoyarsk Krai population and parallel circulation of the different genetic variants of the virus in certain territories and among risk groups facilitate HIV-1 reinfection including reinfection with genetically different viruses. ese conditions are necessary and sufficient for the increase in the frequency of sporadic recombinant events occurrence in the virus genome allowing for HIV-1 rapid evolution. It was shown that random shuffling of HIV genetic material not only increases viral diversity but also enables the virus to evade the response of the human immune system [51]. e increase in circulating HIV-1 genetic diversity is a general characteristic for many current territorial epidemics [43,49,[52][53][54][55][56]. e uniqueness of the situation (detection of URF HIV-1 in 38.6% of HIV diagnosed people from Krasnoyarsk Krai in 2015-2017) also lies in the fact that monitoring for HIV, in this case, allows for observing the selection process of new viable viruses from a variety of recombinant forms of HIV that occur in the human body. Our studies of HIV-1 isolated from recently infected individuals in Siberian regions carried out in 2015-2018, including this study, for the first time demonstrated the uniquely high contribution of the virus recombinant variability to the increase in heterogeneity of HIV-1 circulating in Russia. is situation is an adverse factor of epidemic development and requires special attention in order to achieve the stabilization of molecular epidemiologic processes in the region. Data Availability e text of the article contains references to the nucleotide sequences of HIV-1 deposited in the Genebank, which are the main primary results of this study. e nucleotide sequences of HIV in the Genebank are publicly available. Disclosure e funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this paper.