AIDS
DOI: 10.1097/QAD.0000000000001272
Increasing Prevalence of Synonymous Mutations K65K and K66K in HIV-1 Subtype B Reverse Transcriptase
Sushama TELWATTE1,2, Chanson J. BRUMME3, Anna C. HEARPS1,4, Catherine F. LATHAM1, Joshua A. HAYWARD 1,2, Secondo SONZA5, Nicolas
SLUIS-CREMER6, P. Richard HARRIGAN3,7, Gilda TACHEDJIAN1,2,4,5,8*
1
Centre for Biomedical Research, Burnet Institute, Melbourne, Victoria 3004, Australia
2
Department of Microbiology, Monash University, Clayton, Victoria 3800, Australia 3
British Columbia Centre for Excellence in HIV/AIDS, Vancouver, BC V6Z1Y6, Canada
4
Department of Infectious Diseases, Monash University, Melbourne, Victoria 3004, Australia
5
Department of Microbiology and Immunology at the Doherty Institute for Infection and Immunity, The University of Melbourne, Melbourne, Victoria 3000, Australia 6
Department of Medicine, Division of Infectious Diseases, University of Pittsburgh School of Medicine, Pittsburgh, PA 15261, USA
7
University of British Columbia, BC, Canada 8
School of Science, College of Science, Engineering and Health, RMIT University, Melbourne, Victoria 3000, Australia
* To whom correspondence should be addressed. Tel: 61 3 9282 2256; Fax: 61 3 9282 2100;
Email: [email protected]
Present Address: Retroviral Biology and Antivirals Laboratory, Centre for Biomedical Research, Burnet Institute, GPO Box 2284, Melbourne, Victoria 3001, Australia.
Authors declare no conflict of interest
Financial Support:
GT was supported by a National Health and Medical Research Council of Australia (NHMRC) grant [603704] and NHMRC Senior Research Fellowship [543105]. ST was supported by a Monash Postgraduate Publication Award and an Australian Postgraduate Award through Monash University. N.SC was supported by: United States National Institutes of Health [R01AI081571]. We acknowledge the contribution to this work of the Victorian Operational Infrastructure Support Program received by the Burnet Institute.
Abstract
Objective: Synonymous substitutions K65K or K66K in HIV-1 reverse transcriptase (RT) alleviate fitness and fidelity defects in HIV-1 molecular clones harboring thymidine analog mutations (TAMs); however their potential for transmission and persistence is unknown. Here we investigated the temporal appearance of K65K/K66K relative to TAMs in a HIV-1 cohort, their prevalence over time, and their impact on viral fitness in the context of patient-derived RT sequences.
Methods: Retrospective analyses of the temporal appearance and longitudinal prevalence of synonymous substitutions and drug resistance mutations were performed using the British Columbia Centre for Excellence in HIV/AIDS Drug Treatment Program (DTP) database. Plasma-derived HIV-1 from the DTP was used to generate infectious molecular clones. Growth competition assays were performed to determine viral fitness.
Results: The prevalence of K65K/K66K in drug-naïve individuals tripled from 11% in 1997 to 37% in 2014 (P<0.0001, n=5221), with K66K mainly accounting for the increase. These mutations emerged in drug-treated individuals without TAMs in 14% of the cohort and conferred a fitness advantage in the context of patient-derived multidrug-resistant (MDR) virus in the absence of drug.
Conclusions: The appearance of K65K/K66K in drug-treated individuals was largely independent of TAMs, suggesting alternative factors are likely associated with their emergence. The increasing K65K/K66K prevalence to over a third of treatment-naïve
individuals in the mostly subtype B DTP cohort and their ability to confer a fitness advantage to MDR virus might explain the transmission and persistence of virus harbouring K65K/K66K in untreated individuals, and highlights their role in adaptive HIV-1 evolution.
Key words: synonymous mutations, HIV-1, transmitted drug resistance, thymidine analogue mutations, drug resistance, fitness, replicative capacity
Introduction
Transmission of drug-resistant virus is a threat to treatment success in HIV-infected individuals. The estimated prevalence of transmitted drug resistance (TDR) has in general remained stable over the past 15 years at 10-15% in Europe, USA and Australia [1-4]. Drug resistance mutations (DRM) with low fitness costs are slower to revert to wild-type in the absence of drug pressure [5, 6] and their presence can potentially influence onward transmission [7], while the genetic background of DRM can influence their persistence in vivo [5].
Synonymous or „silent‟ mutations at codons K65K (AAAAAG) and K66K (AAAAAG) in subtype B HIV-1 reverse transcriptase (RT) can emerge in individuals on antiretroviral therapy and are strongly associated with the presence of thymidine analogue mutations (TAMs)[8]. K65K/K66K alleviate a 100-fold increase in insertion/deletion (indel) frequency and a decrease in viral fitness conferred by the TAMs D67N and K70R in the context of the NL4.3 genetic backbone [9]. Rescue of indel formation by either K65K or K66K has been ascribed to disruption of a homopolymeric stretch of eight adenines due to the presence of D67N that causes pausing of recombinant HIV-1 RT during DNA synthesis [8, 9]. Unlike subtype C virus that naturally carries the K65K (AAG) and K66K (AAG) polymorphisms, the HIV-1 subtypes A, D, G, CRF_AE and CRF_AG, which together with subtype B represent a significant global HIV-1 burden, all harbour an A-triplet (AAA) at codons 65 and 66 [10]. However, the temporal appearance of K65K and K66K synonymous mutations relative to TAMs, their prevalence in a population and their impact on viral fitness in the genetic background of multidrug resistant (MDR) HIV-1 are unknown.
Materials and Methods Cohort Participants
Subtype B HIV-1 infected individuals included in this study (n=8443) were enrolled in the Drug Treatment Program (DTP) at the British Columbia (BC) Centre for Excellence, Vancouver, Canada. The demographics of the cohort are well-established [11, 12] and are described in Methods, Supplemental Digital Content 1, http://links.lww.com/QAD/A988. Ethics approval for use of blinded patient samples was granted by the University of British Columbia- Providence Healthcare Research Ethics Board (certificate H06-03350).
Temporal Appearance of Mutations
The temporal appearance of K65K/K66K relative to the emergence of TAMs was assessed by an analysis of available sequential genotypes from individuals enrolled in the DTP. The analysis was performed with 2131 combination antiretroviral therapy (cART)-naïve individuals who did not carry any of the TAMs D67, K70 and K219Q or the synonymous mutations K65K and K66K in their first sample.
Longitudinal Analysis of DRMs, Prevalence of Synonymous Mutations in RT and Growth Competition Assays
The prevalence of K65K/K66K, M184V/I and K103N was determined by a population-based analysis of cART-naïve HIV-1-infected (n=5221 persons) and drug-treated individuals (n=5713 persons) aged 18 or older enrolled in the DTP between 1997 and June 2014 for whom HIV-1 RT sequences were available. In this analysis, individuals appeared in a median of two years (IQR: 1-4). Additionally, the
prevalence of both K65K/K66K was assessed using available subtype B sequences from the US (n= 2771 sequences), downloaded from the Stanford HIV Drug Resistance Database [13] on October 26th 2015. The analysis focused on sequence data collected between 1995 and 2009 from the US as subtype B predominates the epidemic in this region. Determination of the prevalence of silent substitutions (AAAAAG) at 22 consensus lysine residues across HIV-1 RT and growth competition assays were performed as described in Methods, Supplemental Digital Content 1, http://links.lww.com/QAD/A988.
Results
No temporal association between K65K/K66K and TAMs
As K65K/K66K are co-localised on the same genome as TAMs D67/K70/K219 (Table, Supplemental Digital Content 2, http://links.lww.com/QAD/A988), we hypothesised that K65K/K66K were more likely to emerge after the appearance of TAMs to compensate for fitness defects, particularly those mediated by D67N [9]. Accordingly, we assessed the temporal appearance of TAMs and synonymous mutations in individuals on cART using a retrospective population analysis. Surprisingly, our analysis did not support our initial hypothesis. Of the 29 individuals for whom K65K/K66K and TAMs showed any temporal association, no significant pattern was observed [P=0.52; Table 1 comparing (a), (b) and (c)]. Instead K65K/K66K emerged before or independently of TAMs in 299 individuals [Table 1, (a)+(d); 14%] compared to 17 individuals [Table 1, (b)+(c); 0.8%] in whom K65K/K66K were observed concurrently with or following TAMs [comparison of (a)+(d) and (b)+(c); P<0.0001, n=316]. Although the majority of individuals included
in this analysis (83%) had neither synonymous mutations nor TAMs [Table 1(f), n=1769], the appearance of either K65K or K66K alone (n=287) was more likely than TAMs (D67, K70 or K219; n=46) in the absence of K65K/K66K [Table 1 comparing (d) and (e), P<0.0001].
K65K/K66K prevalence is decreasing in drug-treated individuals from 1997-2014 while their prevalence is increasing in drug-naïve individuals with K66K largely contributing to this increase
A retrospective longitudinal analysis of the DTP cohort revealed that the prevalence of K65K/K66K declined (Fig 1a, 31% in 1997 to 23% in 2014, P<0.0001) in individuals with drug resistance mutations (DRM) undergoing therapy (Individuals with DRM) and in „All Individuals‟ (Fig 1a, 27% in 1997 to 22% in 2014, P<0.0001), encompassing every person included in the analysis. This decline coincided with the decreasing number of individuals failing therapy in the DTP cohort and thereby requiring a drug resistance test (see Figure, Supplemental Digital Content 3, http://links.lww.com/QAD/A988, P<0.0001). Strikingly, in drug-naïve individuals the prevalence of K65K/K66K increased from 11% in 1997 to 37% in 2014 in the DTP cohort (Fig. 1a, P<0.001; n=5221). This trend was confirmed in an independent population analysis using sequences from the US, which showed the prevalence of K65K/K66K increased from 9% in 1997 to 23% in 2009 (Fig. 1b, P=0.003; n=2771). Comparatively, within the same DTP population, the prevalence of K103N increased (0-4.5% from 1997-2013; P<0.01) and M184V/I remained relatively low over time (1.94-0% from 1997-2013; P<0.01)(Fig. 1c). These data suggest that K65K/K66K may confer an advantage to HIV-1 by potentially increasing their transmission efficiency and persistence in therapy-naïve individuals.
We next analysed the relative contribution of K66K compared to K65K in addition to the evolution of silent substitutions (i.e. AAAAAG) at 20 other consensus lysine residues across HIV-1 RT in antiretroviral drug naïve patients initiating cART in the DTP cohort between 1996-2014 (see Table, Supplemental Digital Content 4, http://links.lww.com/QAD/A988). Our analysis reveals that while 11 of the 22 positions showed a significant (p<0.002) increase in prevalence of AAA to AAG only two sites showed dramatic increases (slope >0.5) over time. These were K66K (slope = 0.79, p<<0.0001) that had the greatest increase in prevalence and K220K (slope =0.67, p<<0.0001) with the second highest increase in prevalence. While K65K also increased in prevalence, the magnitude of the increase was lower (slope = 0.22, p=0.0013) than K66K. These data show that K66K is the most important AAA-to-AAG silent mutation with regard to increasing prevalence in HIV-1 RT compared to K65K and silent mutations at other lysine (AAA) codons.
K65K and K66K confer a fitness advantage to MDR virus
The increasing prevalence of K66K and to a lesser extent K65K in drug-naïve individuals suggests that these mutations may confer a fitness advantage to MDR virus in a manner similar to that seen previously in a laboratory strain [9]. We subjected HIV-1, derived from the plasma of individual B, with distinct RT coding regions carrying either K65K (MDR1K65K) or K66K (MDR2K66K)(see Tables, Supplemental Digital Content 2 and 5, http://links.lww.com/QAD/A988) to growth competition assays with their corresponding isogenic clones MDR1 and MDR2 lacking silent mutations in the absence of drug (Figures 1d-e). The fitness vector slopes for MDR1K65K and MDR2K66K were positive with linear coefficients of 0.007±0.001 and 0.005±0.001, respectively (Figure 1f, p<0.0001, n=3 for each)
indicating that the presence of either K65K or K66K in the context of MDR RT increased viral fitness compared to virus without either synonymous mutation. These data demonstrate that synonymous mutations potentiate the fitness of virus with RT coding regions from a MDR clinical isolate.
Discussion
We report that synonymous mutations K66K and to a lesser extent K65K in the HIV-1 RT are increasing in prevalence in the DTP cohort and that both K65K and K66K confer a fitness advantage to MDR virus. Our observations suggest that virus harbouring these synonymous mutations have the potential to be transmitted more efficiently and persist in treatment-naïve individuals. Despite K65K and K66K appearing on the same genome as TAMs when both are present, our data indicate that K65K/K66K are more likely to arise in drug-treated individuals independent of TAMs.
Our analysis of the evolution of silent mutations at consensus lysine residues in HIV-1 RT demonstrates that K66K and K220K occurred most frequently (Table, Supplemental Digital Content 4, http://links.lww.com/QAD/A988). Similar to K66K, the silent A to G change at position K220 occurs at the ultimate nucleotide in a homopolymeric stretch of 6A‟s. This observation is consistent with the hypothesis that extensive homopolymeric stretch of A‟s can negatively impact HIV-1 fitness with 4 runs of A‟s associated with RT pausing on the DNA template, decrease in RT processivity, and an increase in indels (frameshift) and base substitutions as reported by us and others [8, 9, 14]. Notably, we have previously reported that K66K is
strongly selected by antiviral drug pressure, while K220 does not appear to be similarly selected [8].
Other lysine-encoding positions in RT were also present in the context of a homopolymeric run of 4 A‟s including positions comprising a run of 6A‟s where a small (i.e. slope = 0.22, K30K and K65K) or no (K101K, K103K, K104K and K219K) significant increase in the prevalence of AAA to AAG silent mutations were observed. Positions K101, K103, K104 and K219 are associated with known polymorphisms and/or drug resistance mutations observed in drug naïve and/or drug treated individuals [15]. The lack of evolution of silent mutations at positions associated with drug resistance is consistent with a previous study showing that synonymous mutations are less likely to occur at positions where nonsynonymous drug resistance mutations are selected in individuals on cART [16]. For the remaining residues with 4 runs of A‟s, constraints on the evolution of silent mutations imposed by RNA secondary structure or cis-acting regulatory elements that are critical for HIV-1 replication [17] might explain the lack of silent mutation evolution at these positions.
Our observation of significant yet mostly modest increases in AAA to AAG synonymous mutations at positions comprising a single triplet of A‟s (K22K, K154K, K223K and K238K) suggest that HIV-1 in this cohort is evolving towards the use of AAG codons preferred by the host [18], as previously reported for early expressed HIV-1 genes [19]. The exceptions in our analysis were K70K, a known drug resistance site, and K73K, which is a polymorphic site in drug-treated individuals [15]. Synonymous mutations might therefore also emerge in HIV-1 due to
evolutionary pressure to use codons and the corresponding tRNA favoured by its host for improving translation efficiency [19, 20].
An intriguing question is whether silent AAA to AAG mutations can mutate to drug resistant forms under antiviral drug pressure. In this regard we analysed the predominately subtype B infected DTP cohort to determine if the K65R tenofovir resistance mutation is preferentially selected in ARV-naïve individuals who start therapy with AAG vs. AAA at position 65 (described in Methods, Supplemental Digital Content 1, http://links.lww.com/QAD/A988). We found no preference for selection of the K65R tenofovir resistance mutation (data not shown). Unfortunately, the rarity of K65R and the relatively low number of individuals with AAG at baseline are a limitation of this analysis.
These data are in stark contrast to the role of a natural silent polymorphism (AAG) at position K65 in subtype C that is reported to accelerate selection of K65R during ARV therapy [14]. Notably, in subtype C, the ultimate A in a homopolymeric stretch of six A‟s is located at the middle base of the AAG triplet. Accordingly, a single base substitution can occur as a result of dislocation mutagenesis (involving Streisinger strand slippage during DNA synthesis) resulting in a nonsynonymous change to AGG (i.e. K65R) with the mutagenesis mechanism dependent on the presence of a G at the end of the homopolymeric A stretch [14]. The importance of this sequence context is highlighted by the accelerated in vitro selection of K65R in the presence of tenofovir in a subtype B virus engineered with the 64-65 (AAA AAG) motif of subtype C, compared to subtype B HIV [21]. In contrast to subtype C, the probability of strand slippage and mutagenesis to AGG encoding an arginine at codon 65 with a
synonymous mutation (AAG) for subtype B HIV-1 strains is unlikely, as the middle A of this codon is not at the end of 4 A homopolymeric stretch (codon 64 is AAG). However, dislocation mutagenesis could explain why the prevalence of synonymous mutations at position K66 in subtype B is greater than K65 as the A to G change occurs at the ultimate A in the homopolymeric run of As with this A immediately upstream of a “G” at codon 67.
The observed tripling of K65K/K66K prevalence in treatment-naïve individuals in the DTP cohort over the past 20 years (Fig 1a) is suggestive of the transmission of virus harbouring these mutations and their persistence in cART-naïve individuals. This observation alludes to the potential of these mutations to confer a transmission and/or persistence advantage to HIV-1, a hypothesis supported by our in vitro data showing that both K65K and K66K confer a fitness advantage to MDR virus (Figure 1d-e). By comparison, our data showing that the M184V/I mutations exist at low levels in drug-naïve individuals, while the prevalence of K103N increased from 1997 to 2014 (Figure 1c), is consistent with previous studies [22, 23]. Mutations with relatively low fitness costs such as K103E/N, Y181C and G190A [24] account for >80% of nonnucleoside reverse transcriptase inhibitor-associated TDR across all regions and subtypes while transmission and/or persistence of high-fitness-cost mutations such as M184V/I are declining [4]. Consistent with our hypothesis, transmission efficacy has been previously linked to fitness. For instance, M184V/I exhibits diminished HIV-1 transmission efficacy in an in vitro model due to lower replicative capacity whereas K103N transmissibility is comparable to HIV-1 WT [7]. Although the contribution of viral fitness to transmission efficiency is debated [25], our data support the notion that the fitness cost of mutations likely plays a role in their transmission efficiency.
Treatment-naïve individuals are a major source of TDR [26, 27]. K103N, which can persist in the absence of drug pressure [6], is detected significantly more often in recently-infected individuals compared to those with unknown infection duration as reported in the European SPREAD program [28] and can exhibit higher transmissibility [29]. Given that the prevalence of K65K/K66K is increasing in drug-naïve individuals in the DTP cohort, coupled with the increased replicative capacity conferred by these mutations it is likely that these synonymous mutations persist in treatment-naïve individuals who consequently serve as a source for onward transmission of HIV-1 harbouring K65K and K66K.
Our finding of an increasing prevalence of K65K/K66K in the BC patient population is uniquely representative of all patients receiving drug resistance tests in BC as the DTP is the sole source of clinical HIV resistance testing in this province. Moreover, our findings are generalizable outside of the BC cohort, as this trend was confirmed in an independent population analysis using sequences from the US deposited in the Stanford HIV database, which showed the prevalence of K65K/K66K increased from 9% in 1997 to 23% in 2009 (Fig. 1b). However, it is likely that chance variation in the initial “seed” virus and many other factors could play a role, and we have no evidence that this is necessarily a universal phenomenon in all regions with predominantly subtype B virus.
Our study highlights the unappreciated roles of silent mutations in the adaptive evolution of HIV-1 and in the transmission and persistence of this virus in untreated populations.
Author Contributions
S.T., P.R.H., A.C.H., G.T. designed research; S.T., C.J.B., A.C.H., J.A.H., performed research; S.T., C.J.B., A.C.H., C.F.L., J.A.H., S.S., N.S-C., P.R.H., G.T., analysed data; and S.T., C.F.L., G.T. wrote the manuscript.
References:
1. Vercauteren J, Wensing AM, van de Vijver DA, Albert J, Balotta C,
Hamouda O, et al. Transmission of drug-resistant HIV-1 is stabilizing in Europe. J Infect Dis 2009;200:1503-1508.
2. Frentz D, Van de Vijver DA, Abecasis AB, Albert J, Hamouda O, Jorgensen LB, et al. Increase in transmitted resistance to non-nucleoside reverse transcriptase inhibitors among newly diagnosed HIV-1 infections in Europe. BMC Infect Dis 2014;14:407.
3. Pham QD, Wilson DP, Law MG, Kelleher AD, Zhang L. Global burden of transmitted HIV drug resistance and HIV-exposure categories: a systematic review and meta-analysis. AIDS 2014;28:2751-2762. 4. Rhee SY, Blanco JL, Jordan MR, Taylor J, Lemey P, Varghese V, et al.
Geographic and Temporal Trends in the Molecular Epidemiology and Genetic Mechanisms of Transmitted HIV-1 Drug Resistance: An Individual-Patient- and Sequence-Level Meta-Analysis. PLoS Med 2015;12:e1001810.
5. Yang WL, Kouyos RD, Boni J, Yerly S, Klimkait T, Aubert V, et al. Persistence of Transmitted HIV-1 Drug Resistance Mutations Associated with Fitness Costs and Viral Genetic Backgrounds. PLoS Pathog 2015;11:e1004722.
6. Gianotti N, Galli L, Boeri E, Maillard M, Serra G, Ratti D, et al. In vivo dynamics of the K103N mutation following the withdrawal of non-nucleoside reverse transcriptase inhibitors in Human
Immunodeficiency Virus-infected patients. New Microbiol 2005;28:319-326.
7. Pingen M, Sarrami-Forooshani R, Wensing AM, van Ham P, Drewniak A, Boucher CA, et al. Diminished transmission of drug resistant HIV-1 variants with reduced replication capacity in a human transmission model. Retrovirology 2014;11:113.
8. Harrigan RP, Sheen C-W, Gill VS, Wynhoven B, Hudson E, Lima VD, et al. Silent mutations are selected in HIV-1 reverse transcriptase and affect enzymatic efficiency. AIDS 2008;22:2501-2508.
9. Telwatte S, Hearps AC, Johnson A, Latham CF, Moore K, Agius P, et al. Silent mutations at codons 65 and 66 in reverse transcriptase alleviate indel formation and restore fitness in subtype B HIV-1 containing D67N and K70R drug resistance mutations. Nucleic Acids Res 2015;43:3256-3271.
10. Buonaguro L, Tornesello ML, Buonaguro FM. Human immunodeficiency virus type 1 subtype distribution in the worldwide epidemic:
pathogenetic and therapeutic implications. J Virol 2007;81:10209-10219.
11. Hogg RS, Yip B, Chan KJ, Wood E, Craib KJ, O'Shaughnessy MV, et al. Rates of disease progression by baseline CD4 cell count and viral load after initiating triple-drug therapy. JAMA 2001;286:2568-2577.
12. Heath K, Samji H, Nosyk B, Colley G, Gilbert M, Hogg RS, et al. Cohort profile: Seek and treat for the optimal prevention of HIV/AIDS in
British Columbia (STOP HIV/AIDS BC). Int J Epidemiol 2014;43:1073-1081.
13. Rhee SY, Gonzales MJ, Kantor R, Betts BJ, Ravela J, Shafer RW. Human immunodeficiency virus reverse transcriptase and protease sequence database. Nucleic Acids Res 2003;31:298-303.
14. Coutsinos D, Invernizzi CF, Moisi D, Oliveira M, Martinez-Cajas JL, Brenner BG, et al. A Template-Dependent Dislocation Mechanism Potentiates K65R Reverse Transcriptase Mutation Development in Subtype C Variants of HIV-1. PloS One 2011;6.
15. Ceccherini-Silberstein F, Gago F, Santoro M, Gori C, Svicher V, Rodriguez-Barrios F, et al. High sequence conservation of human
immunodeficiency virus type 1 reverse transcriptase under drug pressure despite the continuous appearance of mutations. J Virol 2005;79:10718-10729.
16. Ross L, Johnson M, DeMasi R, Liao Q, Graham N, Shaefer M, et al. Viral genetic heterogeneity in HIV-1-infected individuals is associated with increasing use of HAART and higher viremia. AIDS 2000;14:813-819.
17. Mayrose I, Stern A, Burdelova EO, Sabo Y, Laham-Karam N, Zamostiano R, et al. Synonymous site conservation in the HIV-1 genome. BMC Evol Biol 2013;13:164.
18. Benson DA, Cavanaugh M, Clark K, Karsch-Mizrachi I, Lipman DJ, Ostell J, et al. GenBank. Nucleic Acids Res 2013;41:D36-42.
19. Pandit A, Sinha S. Differential trends in the codon usage patterns in HIV-1 genes. PLoS One 2011;6:e28889.
20. Meintjes PL, Rodrigo AG. Evolution of relative synonymous codon usage in Human Immunodeficiency Virus type-1. J Bioinform Comput Biol 2005;3:157-168.
21. Cox S, Margot N, Miller M, Callebaut C. Antiviral activity of tenofovir alafenaimide (TAF) against HIV-1 subtypes and emergence of K65R. In: European Meeting on HIV & Hepatitis, Treatment Strategies & Antiviral Drug Resisance. Rome, Italy; 2016.
22. Jain V, Sucupira MC, Bacchetti P, Hartogensis W, Diaz RS, Kallas EG, et al. Differential persistence of transmitted HIV-1 drug resistance mutation classes. J Infect Dis 2011;203:1174-1181.
23. Chin BS, Shin HS, Kim G, Wagner GA, Gianella S, Smith DM. Short Communication: Increase of HIV-1 K103N Transmitted Drug Resistance and Its Association with Efavirenz Use in South Korea. AIDS Res Hum Retroviruses 2015;31:603-607.
24. Collins JA, Thompson MG, Paintsil E, Ricketts M, Gedzior J, Alexander L. Competitive fitness of nevirapine-resistant human
immunodeficiency virus type 1 mutants. J Virol 2004;78:603-611. 25. Wainberg MA, Moisi D, Oliveira M, Toni TD, Brenner BG. Transmission
dynamics of the M184V drug resistance mutation in primary HIV infection. J Antimicrob Chemother 2011;66:2346-2349.
26. Drescher SM, von Wyl V, Yang WL, Boni J, Yerly S, Shah C, et al.
Treatment-naive individuals are the major source of transmitted HIV-1 drug resistance in men who have sex with men in the Swiss HIV Cohort Study. Clin Infect Dis 2014;58:285-294.
27. Mbisa JL, Fearnhill E, Dunn DT, Pillay D, Asboe D, Cane PA. Evidence of Self-Sustaining Drug Resistant HIV-1 Lineages Among Untreated Patients in the United Kingdom. Clin Infect Dis 2015;61:829-836. 28. Hofstra LM, Sauvageot N, Albert J, Alexiev I, Garcia F, Struck D, et al.
Transmission of HIV Drug Resistance and the Predicted Effect on Current First-line Regimens in Europe. Clin Infect Dis 2015;pii: civ963. 29. Winand R, Theys K, Eusebio M, Aerts J, Camacho RJ, Gomes P, et al.
Assessing transmissibility of HIV-1 drug resistance mutations from treated and from drug-naive individuals. AIDS 2015;29:2045-2052.
Figure 1 Legend
Fig. 1. Prevalence of K65K, K66K, K103N and M184V/I in the DTP cohort in British Columbia and increased fitness of drug-resistant HIV-1 variants carrying silent mutations K65K or K66K. (a) Prevalence of silent mutations K65K and/or K66K in patients in the DTP cohort from 1997 to January 1 2014. „All individuals‟ includes total individuals analysed, „Individuals with DRM‟ harboured ≥1 IAS-USA-defined resistance mutation, „Drug-Naïve Individuals‟ were treatment-naïve at baseline irrespective of DRM, „Drug-Naïve Individuals (-D67/K70/K219)‟ were treatment-naïve and did not harbour any D67, K70 and K219 mutations. (b) Prevalence of K65K and K66K in subtype B HIV-1 infected treatment-naïve
individuals from the US (n=2771). Available subtype B sequences from the US were downloaded from the Stanford HIV Drug Resistance Database on October 26th 2015. The presence of K65K and/or K66K was assessed between the available collection dates (1995-2009) (c) Prevalence of K103N and M184V/I in same subset of treatment-naïve patients as in (a). The Cochran-Armitage test assessed the trends observed for the prevalence of mutations in the DTP during 1997-2014 and Stanford Database cohort during 1995 and 2009. All statistical analyses were performed using XLSTAT 2015 version 16.01 (Addinsoft, New York, NY) or SAS version 9.2 (SAS Institute, Cary NC). (d-f) Growth competition dynamics of MDR HIV-1 carrying K65K or K66K. Representative plots from n=3 assays performed in MT-2 cells with (d) MDR1 and MDR1K65K and (e) MDR2 and MDR2K66K in the absence of drug. (f) Average fitness vectors from n=3 independent assays representing the rate of change in the proportion of MDR1K65K and MDR2K66K relative to their isogenic variant lacking either K65K or K66K (MDR1 and MDR2 respectively).
List of Supplemental Digital Content
Supplemental Digital Content 1. Methods document detailing procedure for cloning and sequencing of the RT region of HIV-1 clinical isolates, construction of MDR infectious molecular clones of HIV-1, growth competition assays in MT-2 cells, silent mutation evolution at 22 lysine-encoding consensus sites in HIV-1 RT, and K65K and K65R association analysis .pdf
Supplemental Digital Content 2, Table detailing the number of clones carrying TAMs and a silent mutation .pdf
Supplemental Digital Content 3, Figure showing the declining number of individuals per year included in these analyses .pdf
Supplemental Digital Content 4, Table detailing the sequence context and the change in prevalence of AAG mutation over time at codons with baseline AAA. pdf
Supplemental Digital Content 5, Table showing predicted genotypic resistance for MDR1, MDR1K65K, MDR2 and MDR2K66K .pdf
Table 1. Temporal appearance of K65K and K66K relative to TAMs in a retrospective analysis of samples from subtype B HIV-1 infected individuals
Temporal Appearance Count # % † P ‡
a) Silent mutation before TAM¶ 12 0.56 0.52&
b) TAM before silent mutation 10 0.47
c) Concurrent emergence 7 0.33
d) Silent mutation only 287 13.47 <0.0001§
e) TAM only 46 2.16 <0.0001+
f) None 1769 83.01 <0.0001⌃
#
Number of drug-naïve individuals †
Percentage of individuals; total n= 2131 ‡
Statistical analysis performed using the chi square (χ2) test to evaluate the temporal association between silent mutations and TAMs by comparing the frequencies of each category. P<0.05 indicates statistical significance. The χ2
tests indicate the difference between observed and expected frequencies of each mutation. ¶
TAM defined as mutations at D67, K70 or K219 &
Comparison of differences between (a), (b) and (c), n=29. No differences in frequency between the outcomes were detected.
§
Comparison of (a)+(d) and (b)+(c), assessing the difference in frequency of independent emergence of K65K and K66K versus their emergence in concert with TAMs.
+
Comparison of (d) and (e), assessing the difference between the frequency of TAMs and the silent mutations, K65K or K66K.
Represents subtype B HIV-1-infected individuals in whom neither silent mutations nor TAM were detected
⌃
