Analysis of the Zidovudine Resistance Mutations T215Y, M41L, and L210W in HIV-1 Reverse Transcriptase

Abstract

Although anti-human immunodeficiency virus type 1 (HIV-1) therapies have become more sophisticated and more effective, drug resistance continues to be a major problem. Zidovudine (azidothymidine; AZT) was the first nucleoside reverse transcriptase (RT) inhibitor (NRTI) approved for the treatment of HIV-1 infections and is still being used, particularly in the developing world. This drug targets the conversion of single-stranded RNA to double-stranded DNA by HIV-1 RT. However, resistance to the drug quickly appeared both in viruses replicating in cells in culture and in patients undergoing AZT monotherapy. The primary resistance pathway selects for mutations of T215 that change the threonine to either a tyrosine or a phenylalanine (T215Y/F); this resistance pathway involves an ATP-dependent excision mechanism. The pseudo-sugar ring of AZT lacks a 3' OH; RT incorporates AZT monophosphate (AZTMP), which blocks the end of the viral DNA primer. AZT-resistant forms of HIV-1 RT use ATP in an excision reaction to unblock the 3' end of the primer strand, allowing its extension by RT. The T215Y AZT resistance mutation is often accompanied by two other mutations, M41L and L210W. In this study, the roles of these mutations, in combination with T215Y, were examined to determine whether they affect polymerization and excision by HIV-1 RT. The M41L mutation appears to help restore the DNA polymerization activity of RT containing the T215Y mutation and also enhances AZTMP excision. The L210W mutation plays a similar role, but it enhances excision by RTs that carry the T215Y mutation when ATP is present at a low concentration.

FIG 1

Structure of the region near the HIV-1 RT polymerase active site that is involved in AZT resistance. (A) Structure of the region near the polymerase active site of an AZT-resistant HIV-1 RT (AZT-R) containing the dinucleoside tetraphosphate AZTppppA bound to the RT (2). π-π stacking between Y215 and the adenine ring stabilizes the binding of the ATP moiety to the mutant RT. The AZTMP has been removed from the end of the primer in an ATP-mediated excision reaction, but the reaction product has not yet been released. The L210W and M41L mutations do not appear to make contact with the ATP moiety, suggesting that they may facilitate the excision reaction indirectly. L210W has been modeled into the structure and is located behind the T215Y mutation on the opposite side of the ATP moiety. It is possible that L210W may interact with T215Y to enhance π-π stacking with the adenine ring. The L residue of the M41L mutation is deeper in the RT structure than the WT M residue and does not appear to interact with the substrate directly. (B) Location and orientation of the Y215 side chain in different RT structures. In the apo form of RT, the side chain appears to interact with P217 and is tucked away from the cleft between the finger and thumb subdomains. When the dinucleoside tetraphosphate is bound, the tyrosine side chain (Y215) rotates away from P217 into an orientation that can interact with the ATP moiety. The two structures are compared in the far right diagram. (C) Interactions of M41L with F116. The leucine side chain is far enough away from the dinucleoside tetraphosphate that it is unlikely to interact directly with this substrate. However, the leucine (L41) side chain could interact with F116, which is near the azido group of the AZT moiety. Nonetheless, this interaction does not appear to alter the position of F116.

FIG 2

Comparison of the abilities of the various mutant RTs to excise AZTMP from the end of the blocked primer and extend the primer in the presence of AZTTP. (A) General outline of the experimental protocol. The template can be varied at the designated “N” site so that any NRTITP can be tested for its ability to be excised; however, in the experiments described in the text, the “N” is an “A” so that AZTMP will be incorporated. The asterisk indicates the presence of the ³²P end label on the primer. The end of the primer is blocked with the desired NRTIMP (in the experiments described in the text, the analog is AZTMP, which is represented by the stop sign). Either a DNA or an RNA template can be used in the reaction. The blocked template/primer is purified and used as a substrate for the mutant RTs in the presence of the indicated concentrations of ATP, 1.0 μM AZTTP, and 10.0 μM each dNTP at 37°C for the indicated amounts of time. The reactions are stopped, the template/primer is purified, and the products are fractionated by PAGE on a 15% gel. The amount of full-length product (extension of the primer to the end of the template) is measured by using the PhosphorImager system and is expressed as a percentage of the total amount of primer. (B and C) Ability of WT RT and the T215Y, M41L/T215Y, and M41L/L210W/T215Y mutants to excise AZTMP from the end of the primer using 3.0 mM ATP and extend the primer on either a DNA template (B) or an RNA template (C). (D and E) Excision/extension by these mutants with 0.6 mM ATP for a DNA template (D) and an RNA template (E).

FIG 3

Comparison of the abilities of additional mutant RTs to excise AZTMP from the end of the blocked primer and extend the primer in the presence of AZTTP. Excision/extension reactions were carried out by WT RT and the L210W/T215Y, M41L/L210W/T215Y, and AZT-R (M41L/D67N/K70R/T215Y/K219Q) mutants on either a DNA (A) or an RNA (B) template, as described in the legend of Fig. 2 and in Materials and Methods.

FIG 4

Ability of WT RT and the T215Y, M41L/T215Y, and M41L/L210W/T215Y mutants to incorporate AZTTP in the presence of normal dNTPs. The RTs were challenged with a range of AZTTP concentrations while the concentration of dNTPs was held constant. If the mutant RTs differ in their ability to bind and incorporate AZTTP, the amount of full-length product will decrease at a rate that differs from that of WT RT. Temp(UA) refers to the fact that the first 2 nucleotides in the template after the double-stranded region are uracil-adenine.

FIG 5

Comparison of the polymerase activities of the WT and mutant RTs. A 5′-end-labeled primer was annealed to single-stranded M13mp18 DNA (A, B, E, and F) or PPT-LTR-PBS RNA (C, D, G, and H). (A, C, E, and G) In the low-dNTP assay, RT was allowed to extend the labeled primer in the presence of suboptimal levels of dNTPs (0.1, 0.2, 0.5, and 1.0 μM each dNTP) for 15 min at 37°C. (B, D, F, and H) For the processivity assay, the 5′-end-labeled primer was annealed to single-stranded M13mp18 DNA or PPT-LTR-PBS RNA. RT was allowed to bind in the absence of dNTPs. dNTPs (20 μM each) and a “cold trap” [poly(rC)·oligo(dG)] were added to initiate the reaction, and the reaction mixtures were incubated at 37°C. The cold trap limits the polymerase reaction to one cycle of extension. The DDDP or RDDP reaction had no trap, and multiple rounds of binding and polymerization will occur. nt, nucleotide.

FIG 5

Comparison of the polymerase activities of the WT and mutant RTs. A 5′-end-labeled primer was annealed to single-stranded M13mp18 DNA (A, B, E, and F) or PPT-LTR-PBS RNA (C, D, G, and H). (A, C, E, and G) In the low-dNTP assay, RT was allowed to extend the labeled primer in the presence of suboptimal levels of dNTPs (0.1, 0.2, 0.5, and 1.0 μM each dNTP) for 15 min at 37°C. (B, D, F, and H) For the processivity assay, the 5′-end-labeled primer was annealed to single-stranded M13mp18 DNA or PPT-LTR-PBS RNA. RT was allowed to bind in the absence of dNTPs. dNTPs (20 μM each) and a “cold trap” [poly(rC)·oligo(dG)] were added to initiate the reaction, and the reaction mixtures were incubated at 37°C. The cold trap limits the polymerase reaction to one cycle of extension. The DDDP or RDDP reaction had no trap, and multiple rounds of binding and polymerization will occur. nt, nucleotide.

FIG 6

(Right) Diagram of the kinetic reaction. The primer is labeled at the 5′ end and then annealed to the appropriate template, depending on which dNTP is being tested (in this diagram, dTTP is shown as being added where the template contains an “A”). (Left) Diagram indicating the various reactants and rate constants during the polymerization process. The reaction mixture contained 25.0 nM labeled template/primer (T/P) and 0.1 nM the appropriate RT. The reaction was initiated by the addition of the dNTP. Reactions were performed at 37°C for 10 min, the reactions were stopped, the T/P was purified, and the samples were fractionated on a 15% PAGE gel. The products were analyzed on a PhosphorImager instrument. The variable substrate (S) is the concentration of the dNTP being tested. P is technically the concentration of PP_i released from the reaction, but this is equal to the amount of primer extended by one nucleotide (primer + 1). The ratio of the primer band to the primer + 1 band was obtained from the PhosphorImager data and is presented as nanomolar P. E is the amount of enzyme (0.1 nM RT).