The M184V mutation reduces the selective excision of zidovudine 5'-monophosphate (AZTMP) by the reverse transcriptase of human immunodeficiency virus type 1

Abstract

The M184V mutation in human immunodeficiency virus type 1 (HIV-1) reverse transcriptase (RT) causes resistance to lamivudine, but it also increases the sensitivity of the virus to zidovudine (3'-azido-3'-deoxythymidine; AZT). This sensitization to AZT is seen both in the presence and the absence of the mutations that confer resistance to AZT. AZT resistance is due to enhanced excision of AZT 5'-monophosphate (AZTMP) from the end of the primer by the RT of the resistant virus. Published data suggest that the excision reaction involves pyrophosphorolysis but that the likely in vivo pyrophosphate donor is not pyrophosphate but ATP. The mutations that lead to AZT resistance enhance ATP binding and, in so doing, enhance pyrophosphorolysis. The excision reaction is specific for AZT because HIV-1 RT, which can form a closed complex with a dideoxy-terminated primer and an incoming deoxynucleoside triphosphate (dNTP), does not form the closed complex with an AZTMP-terminated primer and an incoming dNTP. This means that an AZTMP-terminated primer has better access to the site where it can be excised. The M184V mutation alters the polymerase active site in a fashion that specifically interferes with ATP-mediated excision of AZTMP from the end of the primer strand. The M184V mutation does not affect the incorporation of AZT 5'-triphosphate (AZTTP), either in the presence or the absence of mutations that enhance AZTMP excision. However, in the presence of ATP, the M184V mutation does decrease the ability of HIV-1 RT to carry out AZTMP excision. Based on these results, and on the results of other excision experiments, we present a model to explain how the M184V mutation affects AZTMP excision.

FIG. 1.

Polymerization and excision reactions carried out by HIV-1 RT. (A) Diagram of a simplified version of the polymerization reaction and the relationship of the nucleotide binding (N) and priming (P) sites. At the top, the incoming dNTP is bound at the N site; the end of the primer is at the P site. In the second step (second from the top), the α phosphate is joined to the end of the primer, releasing pyrophosphate. This leaves the end of the primer at the N site. Translocation (third from the top) moves the end of the primer to the P site. The next dNTP binds (bottom), and the cycle continues. (B) The drawings on the left show the steps of AZT excision; the drawing on the right shows the stable closed complex that forms when HIV-1 RT has incorporated a dideoxynucleotide, translocated the ddNMP block to the P site, and bound the incoming dNTP. This stable complex does not form with an AZTMP-terminated primer; the AZTMP-terminated primer resides preferentially at the N site (see text). Because the AZTMP-terminated primer preferentially resides in the N site, the binding of ATP, which acts as a pyrophosphate donor (middle of left drawing), can lead to excision, producing a dinucleotide tetraphosphate (bottom of left drawing).

FIG. 2.

Inhibition of polymerase activity by AZTTP and 3TCTP. The four HIV-1 RTs used in the subsequent experiments (wild-type [WT], M184V, AZT-21, and M184V/AZT-21) were tested for inhibition by AZTTP and 3TCTP. To simplify the comparisons, the activities of each of the enzymes were normalized to 100%. Various concentrations of AZTTP and 3TCTP were added to polymerization reactions containing a −47 sequencing primer annealed to an M13mp18 DNA template (see Materials and Methods). After 30 min, the reactions were stopped by the addition of trichloroacetic acid and the newly synthesized DNA was collected on Whatman GF/C filters. Panel A shows the effects of adding AZTTP to the polymerization reactions; panel B shows the effects of adding 3TCTP.

FIG. 3.

Processivity and primer extension of wild-type HIV-1 RT and the various RT mutants. As described in Materials and Methods, the presence of a poly(rC)·oligo(dG) trap binds the HIV-1 RT after it disassociates from the labeled template-primer. This limits the HIV-1 RT to one round of extension of the labeled primer and is a measure of processivity (on left side of figure). Without the trap, the HIV-1 RT can undergo multiple rounds of primer extension (on right side). The reactions were run for 10 min, and the reaction products were fractionated on a 6% polyacrylamide gel. The products were visualized by autoradiography. The scale on the left indicates the sizes of the products. WT, wild type.

FIG. 4.

Low dNTP extension assay. The ability of the various enzymes to extend the −47 primer on an M13mp18 template was measured at a final concentration of either 0.1 or 0.5 μM each of the four dNTPs. The reactions were run as a time course with samples taken at 15, 30, and 60 min (see Materials and Methods). The reaction products were fractionated on a 6% polyacrylamide gel, and the DNA products were visualized by autoradiography. The scale at the left shows the sizes of the products. WT, wild type.

FIG. 5.

Comparison of the abilities of the various HIV-1 RTs (mutant and wild-type [WT]) to excise AZTMP from the end of the primer and extend the primer in the presence of AZTTP and various concentrations of ATP. The primer is fully blocked with AZTMP at the beginning of the reaction (see Materials and Methods). A short (25 nucleotides long) template extension provides an additional opportunity for the enzymes to incorporate AZTTP and excise AZTMP. The assay measures the percentage of the primer that is completely extended by each enzyme at various ATP concentrations. (A) Results obtained at low concentrations (10 μM) of each of the four dNTPs. (B) Results obtained with high concentrations (100 μM) of each of the four dNTPs.

FIG. 6.

Comparison of the abilities of the various HIV-1 RTs (mutant and wild-type [WT]) to excise ddTMP and extend the primer in the presence of ddTTP and various concentrations of ATP. The assay conditions are similar to those described in the legend to Fig. 5, except that instead of AZTMP, the primer was terminated with ddTMP and, in the reaction mixtures, ddTTP was substituted for AZTTP. (A) Reactions run in the presence of 10 μM concentrations of each of the dNTPs. (B) Reactions run in the presence of 100 μM concentrations of each of the dNTPs.

FIG. 7.

Comparison of the abilities of the various HIV-1 RTs (mutant and wild-type [WT]) to excise AZTMP from the end of the primer and extend it in the presence of AZTTP and various concentrations of sodium pyrophosphate (NaPP_i). These reactions are quite similar to the reactions whose results are depicted in Fig. 5; however, in this case, the pyrophosphate donor is pyrophosphate. (A) Reactions run in the presence of 10 μM concentrations of each of the dNTPs. (B) Reactions run in the presence of 100 μM concentrations of each of the dNTPs.

FIG. 8.

Relative efficiency of AZTMP incorporation or excision at various positions. The abilities of the various HIV-1 RTs to be blocked by AZTMP incorporation at various positions on an M13mp18 template were compared at various concentrations of ATP (see Materials and Methods). The primer was labeled with ³²P, and the reaction products were fractionated by electrophoresis on a 6% polyacrylamide gel. On the left is a scale showing the sizes of the DNA products. Arrows on the right indicate positions at which the excision efficiency differs for the RT mutants.