Short Communication: Selection of Thymidine Analogue Resistance Mutational Patterns in Children Infected from a Common HIV Type 1 Subtype G Source

Abstract

In HIV-1, thymidine analogue mutations (TAMs) cluster in one of two groups (215Y, 41L, 210W, or 215F, 219E/Q), representing two independent mutational patterns (T215Y and T215F cluster, respectively). The mechanisms by which these pathways are selected are not fully understood. To investigate possible factors driving the selection of the TAMs, we analyzed the TAM patterns with regard to the respective treatment, viral load, and HLA in 18 children all infected from a common source of HIV-1 clade G virus and initially treated with zidovudine. The HIV reverse transcriptase sequences of 14/18 children carried at least one TAM. At first sampling date, the T215Y-linked pattern was observed in five cases and the T215F cluster was seen in nine. During the follow-up period, three patients changed their patterns. Children treated with identical NRTI combinations at the first sampling date developed different pathways. Under AZT/d4T therapies, an association was found between the HLA B*13 (in combination with HLA DRB1*0701) and the mutation T215Y. The mutation T215Y reverted in three out of four patients who discontinued AZT/d4T treatment. We speculate that in the context of these subtype G viruses, the development of the T215Y mutation may be strongly disfavored whereas the presence of HLA B*13 may counteract this effect and permit its development.

T hymidine analogues such as zidovudine (AZT) and stavudine (d4T) select for thymidine analogue mutations (TAMs), which accumulate at reverse transcriptase (RT) codons 41, 67, 70, 210, 215, and 219.^1,2 In HIV-1 clade B, T215Y clusters with M41L and L210W, representing the TAM1 pathway; T215F is frequently found together with K219E/Q, constituting the TAM-2 pathway. The direction of thymidine analogue resistance development is of absolute clinical relevance for future therapies as viruses exhibiting the TAM-1 pattern are often fitter and more cross-resistant to tenofovir than those showing the TAM-2 cluster.³

The factors determining the selection of the resistance pattern under AZT/d4T drug pressure are not fully clarified. These may include the viral genetic background, the immune pressure, or the order in which the drugs are administrated. Human leukocyte antigen system (HLA)-dependent immune responses can influence the evolution of drug resistance as the development of resistance mutations may alter HLA-restricted CTL epitopes.⁴ For instance, HLA B*7 has been linked to development of the TAM-1 pathway in clade B viruses under AZT-containing therapies.⁵

In the present study we analyzed the TAM patterns with regard to therapy history, viral load, and HLA A, B, and DR type in 18 children infected with HIV-1 clade G from the same infection source. The common source of infection presents an important advantage in that the viral genetic background can be excluded as variable for differential TAM development. Unfortunately, these types of outbreaks also limit the number of patients to be analyzed.

The 18 children analyzed here belong to a group of 106 patients, not familiarly related, from Rostov on Don (Russian Federation) and nearby areas who acquired nosocomial HIV-1 infection as newborns after exposure to nonsterile needles while receiving treatment in the intensive care unit in 1988/1989.⁶ All children in this study, except patient 3, were treated with AZT monotherapy in the early 1990s.

Plasma samples for genotypic resistance testing, clinical data, and therapy histories were collected between 2002 and 2006. Viral RNA isolation, reverse transcriptase polymerase chain reaction (RT-PCR) and nested PCR, and sequencing of the RT region were performed as previously described.⁷

TAMs (M41L, D67N, K70R, L210W, T215F/Y, and K219E/Q) and mutation clusters were defined in accordance with previous work.^8
–10 Sequences were classified as T215Y linked when the mutation T215Y itself was present or, in its absence, when the M41L or L210W but not the T215F was detected. The T215F-linked sequences showed the T215F mutation or, in its absence, the K219Q/E but not the T215Y. Sequences were classified as “mixed” when they presented no T215Y/F mutation but both M41L and K219Q. The TAMs K70R and D67N were not specifically associated with any specific pattern as they are in both clusters in clade B samples.

HLA A and B genotypes were determined using sequence-specific primer polymerase chain reaction (SSP-PCR). HLA DRB1 typing was done using standard sequencing of exon-2 and exon-3.

Viruses from each patient's plasma were genotyped at least once. The sequences can be found under GenBank accession numbers GQ167727–GQ167766. All patients harbored HIV-1 subtype G viruses.

In 14 patients the RT gene revealed one or more TAMs in at least one sample, while four remained wild type despite receiving nucleoside reverse transcriptase inhibitor (NRTI)-containing therapies (Table 1). In the first sampling time, five viruses followed the T215Y pattern (all five displaying the mutation T215Y) and nine presented the T215F cluster (five showing the T215F itself ).

Table 1.

Therapies and TAM Patterns of Children Infected from a Common Source

		Sampling time 1			Sampling time 2			Sampling time 3
Patient ID	Therapy before sampling time ^a	Therapy ^a	TAMs ^b	TAM profile ^c	Therapy ^a	TAMs ^b	TAM profile ^c	Therapy ^a	TAMs ^b	TAM profile ^c
1	AZT, d4T, ddI, NFV	ddI, d4T, NFV	wt^d	—
2	AZT, 3TC, d4T, NFV	d4T, 3TC, NFV	wt	—	d4T, 3TC, NFV	wt	—
3	—	—	wt	—	—	wt	—
4	AZT, 3TC, NFV	d4T, 3TC, NFV	wt	—
5	AZT, others unknown	ddI, EFV, NFV	T215F, K219Q, D67N, K70R	F
6	AZT, ddI, NVP, NFV	ddI, NVP, NFV	T215F	F	ddI, NVP, NFV	T215F	F
7	AZT, others unknown	d4T, EFV, LPV	T215F, M41L	F	Unknown	T215F, M41L, D67G	F
8	AZT, 3TC, NFV, ddI, d4T, IDV, EFV	3TC, EFV, LPV	T215F, K219EQ, M41L, D67N	F	d4T, ddI, LPV	T215F, 219EQ, 41L, D67N, K70R	F
9	AZT, 3TC, d4T	d4T, 3TC	T215F	F	d4T, 3TC	T215F, M41L	F	d4T, ddI, EFV	T215F, M41L	F
10	AZT, 3TC, NVP	3TC, NVP	K219E	F	d4T, ddI, NFV	K219E	F
11	AZT, 3TC, d4T, NVP, EFV, IDV, NFV	IDV, EFV	K219Q, D67N	F	IDV, 3TC	K219Q	F
12	AZT, 3TC	AZT, 3TC	K219Q, D67N/N, K70R	F	AZT, 3TC	K219Q, D67N, K70R	F
13	AZT, 3TC, NFV	AZT, 3TC, NFV	T215L, K219Q, D67N	F	ddI, 3TC, NFV	T215L, K219Q, D67N	F	AZT, 3TC, NFV	T215L, M41L, K219E, D67N	M
14	AZT, 3TC, ddI, NFV, SQV	d4T, EFV, LPV	T215Y, M41L, L210W, K219E, D67N	Y	ddI, NVP, LPV	T215Y, M41L, L210W, K219E, D67N	Y	3TC, ddI, LPV	T215Y, M41L, L210W, K219E, D67N	Y
15	AZT, 3TC, d4T	AZT, 3TC	T215Y, M41L, L210RW	Y	NVP, AZT	T215Y, M41L, L210RW, D67N	Y	unknown	T215Y, M41L, L210W, D67N	Y
16	AZT, 3TC	AZT, 3TC	T215Y, M41L	Y	EFV	T215AD/N/T, M41L	Y	No therapy	wt	wt
17	AZT, others unknown	d4T, 3TC, IDV	T215Y, M41L, K219Q, D67N, K70R	Y	SQV, EFV	T215C, M41L, K219Q, D67N	M	ddI, 3TC, LPV	T215C, M41L, K219Q, D67N	M
18	AZT, others unknown	AZT, others unknown	T215Y, D67N, K219Q, K70R	Y	IDV, 3TC	K219Q	F	AZT, 3TC, EFV	K219Q	F

AZT, zidovudine; 3TC, lamivudine; d4T, stavudine; ddI, didanosine; NVP, nevirapine; EFV, efavirenz; IDV, indinavir; NFV, nelfinavir; LPV, lopinavir; SQV, saquinavir. AZT and d4T are highlighted in bold as described to select for TAM development.

Main definitors for the TAM pathway (T215F and T215Y) are marked in bold. Mutations in italics are TAMs, although not TAM-pattern defining ones.

Y, T215Y-linked cluster; F, T215F-linked cluster; M, mixed cluster.

wt, wild type at TAM positions.

Different TAM patterns were developed in the first sampling date although all children were treated with similar NRTIs regimens, as reported in another study.¹⁰ For instance, patient #12 displayed the T215F cluster while patient #16 displayed the T215Y-linked one, in spite of an identical therapy history (AZT, 3TC). Patients #11 and #15 received the same NRTI backbone (AZT, d4T, 3TC) with or without NNRTIs/PIs and also developed a different TAM profile. Our results could not corroborate other studies that reported the duration of AZT treatment and the number of previous therapies as factors modulating the selection of TAM clusters.^11,12

During the follow-up period, three patients (#13, #17, #18) showed a change in their TAM cluster: patients #13 and #17 developed a “mixed” pattern, patient #18 switched from a 215Y- to a T215F-linked cluster, and patient #9 reverted to wild type. No reversion of the T215F mutation after cessation of AZT/d4T administration was observed (patients #5, #6, and #8). On the other hand, the T215Y reverted to a T215C/A/D/N/T in three out of four patients who discontinued their AZT/d4T-containg treatment (patients #16, #17, and #18).

No association was detected between viral load values (data not shown) and TAM pathway selectivity.^10
–12

We also investigated whether HLA has an impact on the evolution of TAM profiles (Table 2). Four of the five patients carrying a virus with the T215Y mutation at sampling time 1 exhibited HLA B*13 (patients #14, #15, #16, and #18). The HLA subtype DRB1*0701 allele was detected in all patients displaying HLA B*13 and also in another child who did not present the B*13 allele and displayed the T215F cluster (patient #7). Additionally, one virus showed the T215Y mutation in the absence of DRB1*0701 (patient #17). From these results, we presume a linkage between the HLA B*13 with the mutation T215Y in this specific subtype G viral context. This association is not yet statistically significant due to the small population size. Extended studies are required to confirm this hypothesis.

Table 2.

TAM Patterns and HLA Types of Children Infected from a Common Source

	Sampling time 1	HLA ^a
Patient ^a	TAMs ^b	A	B	DRB1
1	wt^c	10/11	22/27	nd
2	wt	02/31	39/51	0301/1301
3^d	wt	11/24	13/15	0701/0901
4	wt	02/09	35/35	nd
5	T215F, K219Q, D67N, K70R	02/24	07/07	1401/1501
6	T215F	02/25	18/27	0401/0801
7	T215F, M41L	02/25	15/44	0701/1101
8	T215F, K219EQ, M41L, D67N	02/68	38/53	1001/1302
9	T215F	11/33	14/39	0102/0801
10	K219E	01/02	49/57	0403/1303
11	K219Q, D67N	03/24	08/27	0301/1301
12	K219Q, D67D/N, K70R	09/09	15/39	nd
13	T215L, K219Q, D67N	02/11	07/18	nd
14	T215Y, M41L, L210W, K219E, D67N	24/24	13/18	0701/1501
15	T215Y, M41L, L210LRW	24/30	13/52	0701/1502
16	T215Y, M41L	01/01	13/57	0701/1601
17	T215Y, M41L, K219Q, D67N, K70R	02/32	44/58	1104/1303
18	T215Y, D67N, K219Q, K70R	01/66	13/41	0401/0701

Mutations and HLA haplotypes marked in bold are associated with the detection of T215F.

Main designators for the TAM pathway (T215F and T215Y) are marked in bold. Mutations in italics are TAMs, although not TAM-pattern defining ones.

wt, wild type at TAM positions.

Patient #3 did not receive antiretroviral therapy.

The reduced prevalence of the T215Y pattern in the first samples and its reversion in three out of four patients discontinuing their AZT/d4T-containing treatment contrast with previous studies with cohorts of patients carrying viruses of subtypes B and C, where the T215Y pattern represented 60–70% of the sequences and reverted in only 23 out of 52 (44%) subtype B viruses after suppression of AZT/d4T.^10

–14 Our reduced proportion of T215Y-linked mutations cannot be justified by a different prevalence of HLA B*13 in comparison to these previous studies,^10
–12,14 as the prevalence of HLA B*13 in our study (0.025) matches the frequency reported for Eurasia (from 0.016 for western Europe to 0.037 for Asia, available from http://www.allelefrequencies.net). Instead, we presume an influence of the viral background. Different HIV-1 subtypes may respond differently to HAART regimens. M41L is less frequent in C than in B¹⁴ whereas 69N and V75M (NTRI resistance-associated mutations, no TAMs) are more frequent in CRF01_AE than in B.¹⁵

Therefore, we conclude that in the context of these subtype G viruses, the development of the T215Y mutation may be strongly disfavored while the presence of HLA B*13 may counteract this effect and permit its development.

Given the importance of the direction in which thymidine analogue resistance develops for successive therapies, further studies of TAM development/evolution in non-B viruses in association with HLA types are required.

Footnotes

Acknowledgments

The authors thank Dörte Hammerschmidt for excellent technical assistance and Nadine Sichtig for helpful discussions. This work was supported by the Bundesminister fuer Gesundheit within the RESINA-Project (BMG-IIA5-2007-2507AUK361) and EU project URESIST (IST-4-027173). All authors contributed to the critical revision and final approval of the manuscript.

Author Disclosure Statement

No competing financial interests exist.

References

Boucher

, O'Sullivan

, Mulder

et al. Ordered appearance of zidovudine resistance mutations during treatment of 18 human immunodeficiency virus-positive subjects. J Infect Dis, 1992; 165:105–110.

Larder

, Kemp

. Multiple mutations in HIV-1 reverse transcriptase confer high-level resistance to zidovudine (AZT) Science, 1989; 246:1155–1158.

Berg

, Dupke

, Hintsche

, Mayr

, Schlote

, Schuler

. Impact of different TAMs on outcome of a HAART regimen with ABC + 3TC + AZT (Trizivir), TDF. Proceedings of the 4th European HIV Drug Resistance Workshop, Monte Carlo, Monaco, 2006Abstract 69.

Svicher

, Sing

, Santoro

et al. Involvement of novel human immunodeficiency virus type 1 reverse transcriptase mutations in the regulation of resistance to nucleoside inhibitors. J Virol, 2006; 80:7186–7198.

John

, Moore

, James

, Mallal

. Interactive selective pressures of HLA-restricted immune responses and antiretroviral drugs on HIV-1. Antiviral Ther, 2005; 10:551–555.

Bobkov

, Garaev

, Rzhaninova

et al. Molecular epidemiology of HIV-1 in the former Soviet Union: Analysis of env V3 sequences and their correlation with epidemiologic data. AIDS, 1994; 8:619–624.

Balduin

, Sierra

, Däumer

et al. Evolution of HIV resistance during treatment interruption in experienced patients and after restarting a new therapy. J Clin Virol, 2005; 34:277–287.

, Giguel

, Hatano

, Reid

, Lu

, Kuritzkes

. Fitness comparison of thymidine analog resistance pathways in human immunodeficiency virus type 1. J Virol, 2006; 80:7020–7027.

Hanna

, Johnson

, Kuritzkes

et al. Patterns of resistance mutations selected by treatment of human immunodeficiency virus type 1 infection with zidovudine, didanosine, and nevirapine. J Infect Dis, 2000; 181:904–911.

10.

Marcelin

, Delaugerre

, Wirden

et al. Thymidine analogue reverse transcriptase inhibitors resistance mutations profiles and association to other nucleoside reverse transcriptase inhibitors resistance mutations observed in the context of virological failure. J Med Virol, 2004; 72:162–165.

11.

De Luca

, Di Giambenedetto

, Romano

et al. Frequency and treatment-related predictors of thymidine-analogue mutation patterns in HIV-1 isolates after unsuccessful antiretroviral therapy. J Infect Dis, 2006; 193:1219–1222.

12.

Cozzi-Lepri

, Ruiz

, Loveday

et al. Thymidine analogue mutation profiles: Factors associated with acquiring specific profiles and their impact on the virological response to therapy. Antiviral Ther, 2005; 10:791–802.

13.

Balduin

. Dynamic evolution of the pol gene in HIV-1 infected patients under different selective pressures. Mathematics and Natural Sciences University of Cologne (Germany): Cologne, 2008; 120.

14.

Novitsky

, Wester

, DeGruttola

et al. The reverse transcriptase 67N 70R 215Y genotype is the predominant TAM pathway associated with virologic failure among HIV type 1C-infected adults treated with ZDV/ddI-containing HAART in southern Africa. AIDS Res Hum Retroviruses, 2007; 23:868–878.

15.

Ariyoshi

, Matsuda

, Miura

, Tateishi

, Yamada

, Sugiura

. Patterns of point mutations associated with antiretroviral drug treatment failure in CRF01_AE (subtype E) infection differ from subtype B infection. J Acquir Immune Defic Syndr, 2003; 33:336–342.