Mechanism of inhibition of human immunodeficiency virus type 1 ...

15 downloads 176 Views 3MB Size Report
(0 1991 by The American Society for Biochemistry and Molecular Biology, Inc. Vol. 266, No. 3, Issue of January 25, pp. 1754-1762, 1991. Printed in U. S. A.
THEJOURNAL OF BIOLOGICAL CHEMISTRY (0 1991 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 266, No. 3, Issue of January 25, pp. 1754-1762, 1991 Printed in U.S.A.

Mechanism of Inhibition of Human Immunodeficiency Virus Type 1 Reverse Transcriptase and Human DNA Polymerases CY,@, and y by the 5’-Triphosphates of Carbovir, 3’-Azido-3’-deoxythymidine, 2’,3’-Dideoxyguanosine, and 3’-Deoxythymidine ANOVEL

RNATEMPLATEFORTHEEVALUATION

OF ANTIRETROVIRALDRUGS* (Received for publication, July 16, 1990)

William B. Parker$§, E. Lucile White$, Sue C. ShaddixS, Larry J. Ross$, Robert W. Buckheit, Jr.$, Julie M. Germany$, John A, Secrist IIIZ, Robert Vincell, and William M. Shannon$ From the SKettering-MeyerLaboratory, Southern Research Institute, Birmingham, Alabama 35205 and the IDepartment of Medicinal Chemistry, College of Pharmacy, University of Minnesota, Minneapolis, Minnesota 55455

Carbovir (the carbocyclic analog of 2’,3’-didehydro- only drug generally available forAIDS patients, inhibits HIV2’,3’-dideoxyguanosine) is a potent inhibitorof human 1 replication due to its specific inhibition of HIV-1 reverse immunodeficiency virus type 1 (HIV-1) replication. transcriptase (1-3). Other deoxynucleoside analogs effective Assays were developed to assess the mechanism of against HIV-1, such as ddC and ddI, are also inhibitors of inhibition bythe 5”triphosphate of carbovir of HIV-1 HIV-1 reverse transcriptase once they aremetabolized to the reverse transcriptase using eitherRNA or DNA tem- appropriate triphosphate (4-9). These compounds may interplates that contain all four natural nucleotides. Car- fere with viral replication by blocking the enzymes’ ability to bovir-TP wasa potent inhibitorof HIV-l reverse transcriptase using either template with K i values similar copy either the viral RNA genome or the complementary to thatobserved by AZT-TP, ddGTP, and ddTTP. The DNA strand. In either case the drug after anabolism to the kinetic constants for incorporation of these nucleotide triphosphate may interfere with chain elongation by reversibly competing with the naturalnucleoside triphosphate or by analogsinto DNA byHIV-1 reversetranscriptase chain. using either template were similar to the values seen terminating chaingrowth after insertion into the DNA for their respective natural nucleotides. In addition, Unfortunately, all of these compounds have toxicities that the incorporation of either carbovir-TP or AZT-TP inlimit their usefulness (10, 11).The challenge today is to find the presence of dGTP or dTTP, respectively, indicatedcompounds that are potent inhibitorsof HIV-1 reverse tranthat themechanism of inhibition by these two nucleo- scriptase with few unwanted cytotoxicities. One way to evaltide analogs was due to their incorporation into the uate the potential toxicity of a deoxynucleoside analog is to DNA resulting in chain termination. Carbovir-TP was study its interaction with cellular polymerases. Such studies not a potent inhibitorof DNA polymerase a,8, or y, or should reveal differences among the various HIV-1 reverse DNA primase. Given the potent activity of carbovir- transcriptase inhibitors that may suggest rational combinaT P against HIV- 1reverse transcriptase and its lack of tions which would maximize effect against HIV-1 and miniactivity against human DNA polymerases, we believe mize host toxicities. that further evaluation of this compound as a potential Carbovir (Fig. 1, thecarbocyclic analog of 2’,3’-didehydrodrug for the treatment of HIV-1 infectionis warranted. 2‘,3‘-dideoxyguanosine; NSC 614846) has recently been identified as a potent inhibitor of HIV-1 replication (12). Little, if any, toxicity has been detectedinprimates withhigh Human immunodeficiency virustype 1 (HIV-1)’ reverse concentrations of carbovir, suggesting that carbovir may have transcriptase (EC 2.7.7.49) is a proven target for the devel- little toxicity in humans. In a preliminarystudy(13), we opment of anti-AIDS drugs. The 5”triphosphate of AZT, the found that the triphosphate of carbovir inhibited HIV-1 reobserved verse transcriptase at concentrations similar to those * This work was funded in part by a grant from the University of with AZT-TP, ddTTP, and ddGTP. Furthermore, carbovirMinnesota and by National Institutes of Health (NIH) Grant R01 TP was not a potent inhibitor of the host enzymes DNA AI29157. The synthesis of carbovir-TP was funded by NIH Grant polymerases a , p, and y (EC 2.7.7.7) and DNA primase (EC R01 CA-23263. The costs of publication of this article were defrayed 2.7.7.6). The most striking difference between carbovir-TP in part by the payment of page charges. This article must therefore deoxynucleotide analogs was the be hereby marked “aduertisernent” in accordance with 18 U.S.C. andtheotheranti-HIV Section 1734 solely to indicate this fact. relative insensitivity of DNA polymerase y to carbovir-TP. 5 To whom correspondence should be addressed: 2000 Ninth Ave. Unfortunately, direct comparisonsof activity against HIVSouth, Birmingham, AL 35205. Tel: 205-581-2797. 1 reverse transcriptase are clouded by the necessity of using The abbreviations and trivial names used are:HIV-1, human enzyme which employs immunodeficiency virus type 1; AZT, 3’-deoxy-3’-azidothymidine; a highly artificial assay system for the AZT-MP and AZT-TP, the 5”mOnO- or triphosphate of A Z T car- different synthetichomopolymers as templates for each base. of action of bovir, the carbocyclic analog of 2‘,3‘-didehydro-2‘,3’-dideoxyguano- In order to better understand the mechanism sine; carbovir-MPand carbovir-TP,the 5“mOno- or triphosphate of carbovir, we have developed a n assay which uses rRNAas the carbovir; ddC, 2’,3’-dideoxycytidine; ddI, 2’,3’-dideoxyinosine; template for HIV-1 reverse transcriptase. In this work, the ddGMP and ddGTP, the 5’-mono- or triphosphate of 2‘,3’-dideoxyguanosine; ddTMP and ddTTP, the 5’-mono or tri-phosphate of inhibition by carbovir-TP of HIV-1 reverse transcriptase has 3’-deoxythymidine; Tricine, N-[2-hydroxy-l,l-bis(hydroxymethyl)- been characterizedwith respect to itsactivity on botha RNA template (negative strand synthesis) and a DNA template ethyllglycine; dNTPs, deoxynucleotide triphosphates.

1754

Inhibition of HIV-1 Reverse Transcriptase (positive strand synthesis). We report the inhibitionconstants, thetype of inhibition, and the Michaelis constant for carbovir-TP as a substrate. These results have been compared with those obtained for AZT-TP, ddGTP, and ddTTP. In addition, these compounds have also been characterized in a similar manner against various host enzymes (DNA polymerase a , p, and y, and DNA primase) isolated from a human cell line. EXPERIMENTAL PROCEDURES

Materials-Carbovir-TPwas prepared from (+)-carbovir as described (13). AZT-TPwas a generous gift of Burroughs Wellcome Co. (Research Triangle Park, NC). Poly(rA).p(dT)12-18r dATP, dGTP, dCTP, dTTP, ddGTP, ddTTP, SephadexG-25, and T4polynucleotide kinase were purchased from Pharmacia LKB Biotechnology Inc. [methyl,l',2'-3H]dTTP (100 Ci/mmol) in Tricinebuffer was obtained (11 Ci/mmol), [afrom Du Pont-New England Nuclear. [8-'H]dGTP '"PIGTP (3000 Ci/mmol),and[y-"P]ATP (7000 Ci/mmol) were obtained from ICN Ftadiochemicals. DEAE-cellulose, cellulose-phosphate, and single strand DNA-cellulose were obtained from Sigma. 16 S and 23 S Escherichia coli rRNAand two DNA oligomers complementary to positions519-536 (primer A) and 907-926 (primer B) of 16 S rRNAusingthe E. coli 16 S rRNA deoxynucleotide numbering system (14) were obtained from Boehringer Mannheim. Three DNA primers,15or 16 bases long, and the 47-base DNA template were purchased from Genetic Designs, Inc. (Houston, TX). The sequence of the 15-base primer used for the enzyme inhibition studies was 5'-TAACCTTGCGGCCGT-3'. The sequence of the 47base template DNAwas 5"TTCATTTGGGAAACCCTTGGAACCT GACTGACTGGCCGTCGTTTTAC-3'. The sequence of the 15-base primer used for the primer extension assays was 5"GTAAAACGACGGCCA-3'; the 16-base primer had an added guanosine monophosphate at the3' end. Purification of Polymerases-HIV-1 reverse transcriptase was purified from HIV-1 virions (SK1 strain) obtained fromt.he medium of cultures of an infected human T-lymphoblastoid (CEM) cell line (15). This virusisolate was chosen because of itscapacity to provide extremely high levels of infectious virus. The cells were removed by low speed centrifugation and the virus was collected fromthe clarified supernatant by centrifugation at 16,000 X g for 2 h a t 4 "C. The virus pellet was resuspended in l/lOthof the initial supernatantvolume in 10 mM Tris (pH 7.8), 1 mM EDTA, 200 mM KCI, 10 mM P-mercaptoethanol, 3 p~ leupeptin,and 0.5% Triton X-100, dispersed by vortexing, and incubated a t room temperature for 1 h to ensure that there was complete disruption of the virus coat, This extract was diluted with 1 volume of column equilibrationbuffer and loaded onto a 1 X 7-cm single-strand DNA-cellulose column equilibrated with buffer containing 25mM potassium phosphate (pH7.5), 10% glycerol, 1 mM phenylmethylsulfonyl fluoride, and 10 mM p-mercaptoethanol. The column was washed with the equilibrationbuffer and the HIV-1 reverse transcriptase was eluted by a 140-ml linear gradientof 0-1 M KC1. The column was monitored by measuring reverse transcriptase activity (with poly(rA) .p(dT)IZ-IR as the template-primer) singleand strand DNA nuclease activity.Fractions were pooled so that the HIV1 reverse transcriptase activity was essentially free of the nuclease activity. Bovine serum albumin was added to a final concentrationof 0

HOCHV

I

Carbovir

FIG. 1. Structure of carbovir. (+)-cis-2-Amino-1,9-dihydro-9[4-oxymethyl-2-cyclopenten-l-yl]purine-6-one; NSC 614846.

by Carbovir-TP

1755

200 pg/ml. The enzyme was dialyzed against 25 mM Tris (pH 7.5) buffer containing 10 mM (3-mercaptoethanol, 30% glycerol, and 0.01% Triton X-100, aliquoted, and storedat -70 "C until needed. Accurate determinations of the specific activity were not possible because the concentration of protein in our preparations was below the level of detection. However, we calculate that the specific activity of HIV-1 reverse transcriptase was greater than4,000 units/mg of protein. One unit of enzyme activity is defined as the amount of enzyme needed to incorporate 1 nmol of ['HIdTTP into acid-precipitable material per h at 37 "C using poly(rA) .p(dT),,.l, as template. Human DNA polymerases a, p, and y and DNA primase were purified from K562 cells (5 ml of packed cells) grown in cell culture as previously described (9) with some modifications. After 3 cycles of freezing and thawing, the cell pelletwasresuspended in 20 ml of 7.5), 20 mM buffer containing 10 mM potassiumphosphate(pH dithiothreitol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 0.5 pg/ml pepstatin A, 0.5 pg/ml leupeptin, and 10% glycerol, and homogenized with 25 strokes in a Teflon/glass homogenizer. All steps were done a t 4 "C. KC1 was added to a final concentration of 1 M and the extract was incubated on ice for 1 h. Deionized water was added to bring thevolume to 45 ml, and particulate matterwas removed by centrifugation at 19,000 X g for 30 min. Nucleic acids were removed from this extract by chromatography on a DEAE-cellulose column (1.5 X 25 cm) equilibrated with 400 mM potassium phosphate (pH 7.5). The unabsorbed fractions were collected and dialyzed against 1.5 liters of 50 mM potassium phosphate buffer (pH 7.5) containing 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 1 mM EDTA, and 5% glycerol. After 2.5 h, the extract was transferred to fresh dialysis buffer containing the above ingredients plus 0.5 pg/ml pepstatin A and leupeptin, and dialysis was continued for 16 h. All buffers beyond this step, except thefinal storage buffer, contained 1 mM dithiothreitol, 1mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 5% glycerol, 0.5 pg/ml pepstatin A, and 0.5 pg/ml leupeptin, and all gradients ranged from 0 to 1 M KC1 in the appropriate dialysis buffer. This extractwas loaded onto a DEAE-cellulose column (1.5 X 25 cm) connected in tandem to a cellulose-phosphate column (1.5 X 25 cm) equilibrated with the dialysis buffer. The DNA polymerase a-DNA primase complex and DNA polymerase y bound to the DEAE-cellucellulose-phosphate lose column, and DNA polymerase (3 bound to the column. The enzymes were eluted from the DEAE-cellulose column with a 150-ml linear gradient. The fractions containing the DNA polymerase a-DNA primase complex and DNA polymerase y were pooled, dialyzed against 1 liter of 100 mM potassium phosphate (pH 7.5) buffer for 16 h, and loaded onto a cellulose-phosphate column dialysis buffer. The enzymes were (1.5 X 25 cm) equilibrated with the eluted with a 150-ml linear gradient. DNA polymerase y was usually resolved from the DNA polymerase a-DNA primase complex at this step. The fractions containing DNA polymerase y and DNA polymerase a were pooled separately anddialyzed against 25 mM potassium phosphate buffer (pH 7.5). Each pool was then loaded separately onto a single strand DNA-cellulose column (1 X 7 cm) equilibrated with the dialysis buffer, and theenzymes were eluted from their respective columns with a 50-ml linear gradient. (When DNA polymerase y was not resolved from the DNA polymerase (Y peak during chromatography on cellulose-phosphate, the activities were pooled together and purified further on single strand DNA-cellulose. These enzymes are always resolved by this step.) Under these mild purification procedures DNApolymerase a and DNA primase were not separated from one another. DNA polymerase (3 was eluted from the phosphocellulose column with a 100-ml linear gradient. The fractions containing DNA polymerase P werepooled and dialyzed against 25 mM potassium phosphate buffer (pH 7.5) for 16 h. This extractwas then loaded onto a single strand DNA-cellulose column (1 X 7 cm) equilibrated with the dialysisbuffer andeluted witha50-ml lineargradient. The fractions containing the desired enzymes from the final purification step were pooled and dialyzed against 25 mM potassium phosphate buffer (pH 7.5) containing 1 mM dithiothreitol, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, and 20% glycerol. The three DNA polymerases were distinguished by their chromatographic behavior, their differences in K,,, for deoxynucleoside triphosphates, and their sensitivities to dideoxynucleotides, aranucleotides, and N-ethylmaleimide. The specific activities of DNA polymerase a, P, and y used in these studieswere 500, 1666, and 8 units/mg protein, respectively. One unit of enzyme activity is defined as the amount ofenzyme needed toincorporate 1 nmol of [3H]dTTP into acid-precipitable material/h at 37 "C using activated DNA as template. Measurement of Enzyme Actiuity-For the studyof RNA transcription,HIV-1 reverse transcriptase activity was measured in 5 0 - ~ l

1756

Inhibition of HIV-1 Reverse Transcriptase Carbovir-TP by

reactions containing 50 mM Tris (pH8.0), 50 mM KCI, 10 mM MgC12, reactions were stopped and the products of primer extension sepa4 mM B-mercaptoethanol, 3% glycerol, 1 mg/ml bovine serum albu- rated asdescribed above. min, 3.33 pg/ml of primed 16 S rRNA from E. coli, 10 p~ dATP, 10 Quantitation of the Primer Externion Assays-The bands on the p M dCTP, 10 p M dGTP, and various concentrations of 13H]dTTP. x-ray film representing various labeled DNA chains were quantitated The primer was annealed to the template at a ratio of 3 to 1 as with the aid of a densitometer (Shimadzu; CS-9OOOU). The linear described below. HIV-1 reverse transcriptase was also studied using range of the x-ray film relating the band intensity to the actual gapped duplex DNA (16) as template. This activitywas measured in disintegrations was determined by electrophoresis of 2-fold serial 50-p1 reactions containing 50 mM Tris (pH 8.0), 125 mM KCl, 2 mM dilutions of "P-labeled primer followed by autoradiography.The MgCl,, 4 mM 8-mercaptoethanol, 1 mg/mlbovine serum albumin, bands were scanned and the resulting peaks were integrated with the 100 pg/ml of gapped duplex DNA, 10 p~ dATP, 10 p~ dCTP, 10 p~ densitometer.Mostexperimentalmeasurements were taken from dGTP, and various concentrations of I3H]dTTP. Activity with the results falling within the linear range of the countsuersus integration homopolymer template was measured in50-p1 volumes containing 38 curve, but some of the more intense bands were corrected with the mM Tris (pH 7.9), 180 mM KCl, 8 mM MgCI,, 4 mM P-mercaptoeth- use of an experimentally determined factor to account for the loss of anol, 1 mg/ml bovine serum albumin, 1 pg/ml of poly(rA) .p(dT)12-18, linearity. The extension of the primerby HIV-1 reverse transcriptase and various concentrations of [3H]dTTP. DNA polymerase a activity is expressed as the percentage of the average concentration of subwas measured in 5O-pl volumes containing 50 mM Tris (pH 8.0), 1 strate present throughout the experiment and was calculated accordmg/ml bovine serum albumin, 10 mM MgCl,, 1 mM dithiothreitol, ing to the formula: 100 pg/ml gapped duplex DNA, 50 p~ dGTP, 50 p~ dCTP, 50 p~ I1 dATP, various concentrations of [3H]dTTP, and 20 mM potassium %/min = 100 X phosphate (pH8.0). DNA polymerase p and y activity were measured ( I o + OSI1)t as described for DNA polymerase 01 except that 20 mM potassium phosphate was replaced with 100 mM KCl. DNA primase activitywas as described (20-25). I , is the intensity of the band (product) representing the incorporationof deoxynucleotide analog into thegrowing measured in 5O-pl volumes containing 50 mM Tris (pH 8.0), 1 mg/ml bovine serumalbumin,10 mM MgCI,, 1 mM dithiothreitol, 0.083 DNA chain. I, is the intensity of t.he primer band at the conclusion of the experiment. For experiments where I , was further than one units/ml poly(dC), and [a-"PIGTP. For each enzyme reaction the base from the original primer, the relative velocity of the incorporalabeled dNTP used will depend on the analog being studied. When the guanosine analogs were evaluated against HIV-1 reverse tran- tion was determined by dividing I , by the intensity of the band one scriptase on the RNA template, the ethanol was removed from the nucleotide shorter (20). The incubation time, t, of all experiments labeled dGTP stock prior to use. After incubation the DNA in each was chosen so that products were still accumulating linearly as a sample was precipitated onto glass fiber filters using a 5% trichloro- function of time during the experiments. Furthermore, the enzyme activity was kept low so that only a small portion of primers were acetic acid solution containing 10 mM pyrophosphate. These filters were batched washed and counted for radioactivity (17). All enzyme extended by the polymerase (typically less than 25%). The K , and V,,, values were determined from Lineweaver-Burk analyses (26,27) reactions, except DNA polymerase y with AZT-TP, were linear a t of the data. All lines were obtained from unweighted least squares fit 37 "C for 60 min. Extension of DNA Primer Annealedto 16 S rRNA-The extension to the data points. of a '"P-labeled DNA primer annealed toE . coli 16 S rRNA was done RESULTSANDDISCUSSION using RNA sequencing technology described by Lane et al. (18).The 5' end of the desired primer (A or B) was labeled with P ' in 50-pl In this work we have evaluated a number of active antireactions containing 0.3 pg of oligomer, 50 mM Tris (pH 8.0), 50 mM HIV deoxynucleoside analogs for their ability to inhibit HIVNaCI, 10 mM MgCl?, 2.8 mM dithiothreitol, 1 mM EDTA, 0.15 mCi 1 reverse transcriptase. The effect of these deoxynucleotide of [y-:"P]ATP,and 10 units of T4 polynucleotidekinase.After incubation for 2 h at 37 "C, the reaction was stopped by boiling for 5 analogs was determined using botha natural RNA and DNA deoxynucleotides. We believe min. The labeled product was separated from [y-32P]ATPby centrif- template containing all four ugation of the sample through a 1.0-ml Sephadex G-25 column and that assaysusing a natural rRNA templateoffer a number of mixed with 100 pg of a mixture of E. coli 16 S and 23 S rRNA in 100 advantages for this typeof study over the homopolymers that mM NaCl, 200 mM Tris (pH 8.0). This mixture was placed in boiling are normally used in the assay of HIV-1 reverse transcriptase. water and allowed to cool slowly to room temperature. Assuming First, this template, which is commercially available, can be equal amounts of 16 S and 23 S rRNA in these preparations, the ratio of primers to template was approximately 1 to 1. The ratio of used to study anydeoxynucleoside analog regardless of base. primer to template in experiments involving guanine deoxynucleo- HIV-1 reverse transcriptasereads only thepoly(rA)and tides was 3 to 1, and theIo (see below) was divided by 3 to obtain the poly(rC) templates efficiently (2, 28) making it difficult to true measure of extendable primers. The extension of these primers study any analog other than thoseof thymidine and guanoby HIV-1 reverse transcriptase was done in 10-pl mixtures containing sine. Second, a natural RNA template assaymay more accu50 mM Tris (pH8.0), 10 mM MgCl,, 1 mM dithiothreitol, 50 mM KC1, rately represent the true in vivo effect of a deoxynucleoside natural deoxynucleotides and nucleotide analogs as defined in each experiment, and 0.05 mg/ml of 32P-labeled primer annealed to 16 S analog on HIV-1 reverse transcriptase than assays using a rRNA for 40 min at 37 "C. The reactions were terminated by adding homopolymer as template and only one deoxynucleotide as 10 p l of 90% formamide, 10 mM EDTA, and 0.03% bromphenol blue, substrate. Third, comparisonsbetween deoxynucleoside anafollowed by boiling for 5 min. The extension products of the primers log triphosphates of different bases can be made under idenwere separated by electrophoresis (at 2000 volts for 4 h) on a 20% tical assay conditions. Finally, this template can easily be polyacrylamide gel containing 7 M urea (19), and they were visualized of any deoxynucleotide adapted to study the incorporation by autoradiography. The sequence of addition of deoxynucleotides after primer A was 5'-GCACGGAGT-3', and after primer B it was analog into the DNAby HIV-1 reverse transcriptase using a 5'-TAACCTTGC-3' (14). The sequences were confirmed by sequenc- RNA template. ing with dideoxynucleotides. HIV-1 reverse transcriptase required MgClz with both hetExtension of DNA Primer Annealed to DNA Template-Theeffect eropolymer templates; 8 mM gives optimal activity using the of deoxynucleoside triphosphate analogs on the ability of these polym- RNA template and2 mM with the DNA template.MnCL did erases to extend a "P-labeled primer that has been annealed to a M e whenemploying eithertemplate. single-strand DNA template was examined as described (19). Briefly, notsubstitutefor the desired polymerase was incubated in 10-pl reactions that con- Furthermore, thetwo templates hadslightly different requiretained the ingredients listed above for each polymerase, except that ments for monovalent salt; 50 mM KC1 maximally stimulated the gapped duplex DNAwas replaced with 0.15 pg/ml of single-strand the enzyme with the RNA template, whereas 125 mM was 47-base DNA oligomer that was annealed to an equimolar concentra- required for maximal stimulation with the DNA template. tion of either a 15- or 16-base primer labeled on the 5' end with [y- Kinetic constants for the two templates and the homopolymer '"PIATP by T4 polynucleotide kinase. AMP (2 mM) was included in I. One disadvanpoly(rA) .p(dT)12-18 are presented in Table 3'-exonuclease the DNApolymerase y experiments to inhibit the is apparent from this table, i.e. the that copurified with DNA polymerase y. Addition of AMP did not tage of the rRNA template affect DNA polymerase y activity. Afterincubation a t 37 "C the relatively low activity of HIV-1 reverse transcriptase using

1757

Inhibition of HIV-1 Reverse Transcriptase by Carbovir-TP TABLE I Kinetic constants for templates and dTTP with HIV-1 reverse transcriptase The rates of incorporation of deoxynucleotides were linear with respect to time. The kinetic constants were determined from linear plots of 1 / V versus l/[substrate] (26, 27). Each value is the average of a t least two separate experiments. Template

Km pglrnl

rRNA gd DNA 2.44 PoMrA)

0.17 0.12 0.05

Vmax K, V,.,/K, unitsf ml

0.01 0.16 1.12

5.0

LI

I

1

/

D N A TEMPLATE

for dTTP PM

0.57

0.06 1.33 48.80

3.5

TABLEI1 Inhibition of HIV-1 reverse transcriptase by carbouir-TP, ddGTP, AZT-TP, and ddTTP The rates of incorporation of deoxynucleotides in the presence of triphosphate analogs were linear with respect to time. The kinetic constants were determined from replots of the slopes from linear plots of 1/V versus l/[dNTP] (53). Inhibition of HIV-1 reverse transcriptase was competitive with respect to the natural triphosphate for each compound. Each value is the average of a t least two separate experiments. 16 S rRNA template

ComDound

(+)-CBV-TP 0.96 0.16 0.82 ddGTP AZT-TP 0.46 ddTTP 0.68

W 0.09

0.023 0.024

0.04

CARBOVIR .TP IpMI

Gapped template DNA

PM

0.09 0.20 0.05

RNA TEMPLATE

0.74 0.07 0.05 0.54 0.02 1.02 0.18 1.22 0.10

.l.O

-0.5

0

0.5

10

1 /dGTP [ flu”11

0.04 0.18 0.08

FIG. 2. Inhibition of HIV-1 reverse transcriptase by carbovir-TP. A replot of the slopes from a Lineweaver-Burk (53) analysis of the inhibition by carbovir-TP of HIV-1 reverse transcriptase using both templateswith respect to dGTPwas done as described under “Experimental Procedures.”

this template compared to its activity with poly(rA). However, the K,,, values for both the template and dTTP and the optimal monovalent cation concentration were closer to the values phate analogs were competitiveinhibitors of the enzymes studied when compared with their respective normal deoxyreported using native viral RNA as template(2, 29). The effect of carbovir-TP on isolated HIV-1reverse tran- nucleoside triphosphates. Noneof these analog triphosphates scriptase was determined and compared with the activity of were potent inhibitors of DNA polymerase a , the enzyme primarily involved in cell replication. The Kifor each deoxddGTP,AZT-TP,andddTTP(Table 11). Withboththe rRNA and DNA template, the rateof incorporation of deox- ynucleotide analog was at least 20-fold greater than the K , ynucleotides by HIV-1 reverse transcriptase in the presence for the normaldeoxynucleotide. Therefore, inhibitionof DNA polymerase a would not be expectedin cells treated with any of inhibitory concentrations of carbovir-TP, AZT-TP, of these deoxynucleosides. DNA polymerase p was much less ddGTP, or ddTTP was linear with respect to time. In addition, was DNA polymerase (Y but was under conditions in which the template was in large excess, sensitive to carbovir-TP than all four analogs appeared to bereversible inhibitors andcould more sensitive to ddGTP and AZT-TP. DNA polymerase y be classified as fully competitive with the corresponding nat- was extremely sensitive to ddGTP but less sensitive to carural deoxynucleoside triphosphate (data shown only for car- bovir-TP. Because the inhibition of DNA polymerase y by bovir-TP, Fig. 2). Carbovir-TPwas a very potent inhibitor of AZT-TP was not linear with respect to time, we were not able HIV-1 reverse transcriptase using the RNA template. It is t o determine a Kifor this reaction. However, in reactions likely that carbovir-TP is twice as potent as the results in containing 2.5 FM dTTP the ICsa for AZT-TP with DNA this paper indicate, because the carbovir-TP used in these polymerase y (11 @)I was similar to that seenwith DNA experiments was produced from racemic carbovir; only the polymerase /3 (13). T h e K,,, for d T T P with DNA polymerase (-)-isomershows antiviralactivityin cell culture(30). It y was approximately 10-fold lower than i t was withDNA KJK, ratios for all four com- polymerase p, suggestilig that AZT-TP was a more potent should be stressed that the pounds, on both templates, were less than 1,indicating strong inhibitor of DNA polymerase y than DNA polymerase p. binding of theinhibitorinthepresence of substrate.In Recently, it has been suggested that the inhibition of mitoaddition, theKi values for carbovir-TP and AZT-TP on both chondrial replication by dideoxynucleosides, such as ddC and templates were less than the intracellular triphosphatelevel ddI, is responsible for the painful peripheral neuropathy asachieved in cell cultures incubated withcarbovir (31) or AZT sociated with these drugs (35). Consistentwith this observa(1, 32-34). tion is the result that only ddGTP and otherdideoxynucleoAn important consideration in evaluating any inhibitor of tides (Table I11 and Refs. 9 and 36) have a K J K m ratio (0.19 HIV-1 reverse transcriptase is examining itseffect on human for ddGTP) which is favorable for inhibition of this enzyme DNA polymerases (Table 111). Inhibition of these enzymes activity. If this is thecase, then our resultswould predict that could lead to cytotoxicity, limiting theusefulness of the com- treatment of patients with carbovir will not result in this pound.Except for DNA polymerase y andAZT-TP,the particular toxicity, and therefore it could be safely used in reaction rate of all enzymes with thedeoxynucleotide analogs combination with ddI orddC. Inhibition of DNA polymerase was linear with respect t o time. All deoxynucleoside triphos- y could also be responsible for the toxicity of AZT (37, 38).

1758

TABLEI11 Inhibition of human polymerasesby carbouir-TP, ddGTP,and AZT-TP The rates of incorporation of deoxynucleotides in the presence of triphosphate analogs were linear with respect to time. The kinetic constants were determined from replots of the slopes from linear plots of l / V uers’sus 1/ [dNTP] (53). Inhibition of each enzyme was competitive with respect to the natural triphosphate for each compound. The DNA polymerases were assayed using gapped duplexDNA as template. DNA primase was assayed using poly(dC) as template. K,

K,/Km

K,

1.5 0.03

DNA 26 pol 01 22 229DNA pol p320 0.14 DNA89poly y 23 DNA primase

AZT-TP

Km

CM

1.3

ddCTP

(&)-CBV-TP

Enzyme

65

21

Inhibition of HIV-1 Reverse Transcriptase by Carbouir-TP

CM

0.85 1.4 0.26 5000.11

CM

5.0 57

500 0.09

The Ki of carbovir-TP for all three DNA polymerases was greater than the level of carbovir-TP which is produced in cells treated with high concentrations of carbovir (31), suggesting that any cytotoxicity observed with carbovir would not be a result of the inhibition of any of these enzymes. DNA primase was also inhibited by both carbovir-TP and ddGTP (39). Although the KJK,,, ratio was favorable for inhibition ) much higher than concentrations (-O.l), the Ki (50 p ~ was achieved in cell culture with either carbovir (31) or dideoxynucleosides (4,7,9,40), indicating that significant inhibition of this enzyme would not occur with pharmacological doses of any of these compounds. Studies on the effect of these analogs on DNA polymerase 6 are in progress. The ability of the various enzymes to incorporate carbovirTP into the growing DNA strand was determined using a DNA template, and this was compared to the results seen with AZT-TP, ddTTP, and ddGTP. These experiments were similar to the work of others who have evaluated the incorporation into DNA of deoxynucleoside analogs by reverse transcriptases (2, 5, 41) and DNA polymerases (19, 42-46). We have expanded our ability to study antiviral deoxynucleoside analogs by developing an assay which uses a RNA template to study primer extension by reverse transcriptases. It is not known whether the type of template will affect the ability of these enzymes to substitute antiviral deoxynucleotides as substrates. Because there is no 3”hydroxyl on carbovir-TP, ddGTP, ddTTP, or AZT-TP, these compounds are chain terminators, if they are incorporated into the DNA chain. However, it is also possible that a particular inhibitor of HIV-1 reverse transcriptase is notasubstrate for the polymerase and, therefore, would not be incorporated into the DNA. For those analogs that are incorporated into the growing DNA chain, it is possible to determine a V,,, and K,,, for the incorporation (20-25). Mendelmann etal. (23) have shown that for natural deoxynucleotidesthese constants are dependent on whether or not the base is added immediately after the primer (standing start) or at a position further along the template (running start). We have therefore evaluated the incorporation of these analogs at both standing and running sites. The incorporation of carbovir-MP into the growing DNA strand by HIV-1 reverse transcriptase was studied using 16 S E. coli rRNA that was annealed to primer B. In the absence of dGTP theprimer was extended seven deoxynucleotides (27 bases total length) where it was stopped due to a lack of substrate (Fig. 3). As the concentration of carbovir-TP in the reaction was increased from 0 to 30 p M , more and more product was observed that was 28 deoxynucleotides in length, indicating that carbovir-MP was incorporated into the growing strand in place of dGMP. Unfortunately, thereis a natural

21

4.2

45

stop siteat position 27 in reactions containing all four natural deoxynucleotides (Fig. 3, lane C) which interfered with the interpretation of the results. However, a comparison of the bands in lane C with the lane containing 30 p~ carbovir-TP in place of dGTP, shows that the intensity of the DNA oligomer 27 deoxynucleotides in length is the same, suggesting that in the presence of saturating concentrations of carbovirTP, HIV-1 reverse transcriptase behaves at position 27 exactly as it would with dGTP. That is, it either incorporates the deoxynucleotide or it dissociates from the template independent of the deoxynucleotide. The K,,, for the addition of carbovir-MP to theDNA chain at a running start was determined to be approximately 0.23 p~ (Fig. 3). Using primer A with which dGTP is the first base to be inserted, we found that carbovir-MP was also incorporated intothe growing DNA strand with a K,,, similar to thatfound with the running start (Table IV). The ability of HIV-1 reverse transcriptase to incorporate carbovir was also studied using a DNA template (Fig. 4), and similar results were found. The kinetic constants for the incorporation of the various nucleotides into DNA from standing starts by HIV-1 reverse transcriptase are shown in Table IV. In the evaluation of this table it should be emphasized that the kinetic values for the incorporation of normal nucleotides is dependent on the sequence under evaluation (47). Therefore, no attempt should be made to compare directly the kinetic values of nucleotides at different insertion sites. However, the ratio for any nucleotide analog to that of the natural nucleotide at the same position should give relative measures of the ability of that particular analog to be a substrate. The V,,,/K, is a measure of the efficiency of nucleotide insertion by the enzyme (21). The general conclusion that can be drawn from the results shown in Table IV is that all four nucleotide analogs are efficiently used as substrates by HIV-1 reverse transcriptase with both templates. The most dramatic difference among the nucleotides occurs with the DNA template. There theenzyme inserts the two deoxyguanosine analogs with one-half the efficiency of dGTP and thetwo thymidine analogs with onetenth theefficiency of dTTP. With the RNA template, HIV1 reverse transcriptase appeared 5 times more efficient’ at inserting carbovir-TP and AZT-TP than the natural nucleotide and 2.5 times more efficient with ddGTP and ddTTP. Because of the high variability of some of these determinations, the statistical significance of the data on the RNA template is questionable. Results for the running start arenot reported because we were unable to obtain all four sets of kinetic constants for the naturalnucleotides with our primers. (To force a stop at either the deoxyguanosine or thymidine insertion site, the next base in sequence has to be left out of the reaction. If that base is also needed for chain extension

Inhibition of HZV-1Reverse Transcriptase FIG.3. Incorporation of carbovirMP into the DNA by HIV-1 reverse transcriptase usinga rRNA template. HIV-1 reverse transcriptase was incubated with carbovir-TP (0-30 p ~ ) , 10 p~ each of dATP, dCTP, and dTTP, and ‘”P-labeled primer B annealed to 16 S rRNA. Lane C contained no carbovirTP and 10 JIM each of dATP, dCTP, dCTP, and dTTP. The extension products were separated by gel electrophoresisand visualized by autoradiography. The bandswere quantitated with the aid of a densitometer and the relativevelocity was calculated as described (20). A Lineweaver-Burkplot (26, 27) of the data is shown.

by Carbouir-TP

1759

5-

0. GC, GT TCCAA

” -

-

”.

4

.~ ” -

-~

1

I

Primer .03

0

0.1 0.3

1

3

10

30

1

C

CARBOVIR - TP ( p h 4 )

3.33

10

- TP ( M - 1 )

l/CARBOVIR

TABLE IV Incorporation of deoxynucleotide analogs into DNA by HIV-1 reverse transcriptase The kinetic constants for the incorporation of the deoxynucleotides a t standing starts were determined as described under “Experimental Procedures.” Primer A annealed to 16 S rRNA was used as the RNA template for experiments with nucleotides containing guanine. Primer B annealed to 16 S rRNA was used as the RNA template for experiments with nucleotides containing thymine. The values for the kinetic constants of d C T P a n d d 7 T P were done in assays inwhich the only nucleotide present was either dGTPor d T T P so that once incorporated the DNA chain would not be further extended. Each value represents the mean f S.D. of a t least three separate determinations. DNA template

RNA template Compound

Vm..lKm K,

V-.

V-./K, K m

PM

%/rnin

dCTP (f)-CBV-TP ddCTP

0.72 f 0.71 0.22 f 0.08 0.31 0.23 f 0.15

0.20 f 0.06 0.27 0.51 1.23 f 0.04 0.65 0.74 f 0.03

dTTP AZT-TP ddTTP

0.230.07 f 0.10 0.13 f 0.08 0.14 f 0.07

f0.30 0.01 0.1835.2 f1.38 0.06b 0.10 f 0.05

0.28

V-.

PM

%/min

0.66 f 0.20 0.49 f 0.23 f 0.44 0.27

1.0 f 0.17 f 0.96 0.15“ f 0.19”

0.68

6.25 f 2.4 f 1gh 34.2 f 2Sb

0.98 f 0.51 0.92 f 0.47 0.29 f 0.06‘

0.16 0.03 0.01

1.52

0.71 Significantly less than the corresponding value observed with d C T P determined by one-tailed Student’s t test ( p < 0.05). Significantly greater than the corresponding valueobserved with d T T P determined by one-tailed Student’s t test ( p < 0.05). Significantly less than the corresponding value observed with AZT-TP or d?TP determined by one-tailed Student’s t test ( p < 0.05). a

4-

G-

3G-

A-

T-

C-

2-

GAC-

1-

I C

0

0.01 0.03 0.1 0.3

1

CARBOVIR - TP I N )

3

10

30

13.3 10 l/CARBOVIR

- TP

33.33

FIG.4. Incorporation of carbovir MP into DNA by HIV-1 reverse transcriptaseusing a DNA template. HIV-1 reverse transcriptase was incubated with carbovir-TP (0-30 JIM), 10 JIM each of dATP, dCTP. and dTTP. and a “P-labeled DNA primer 16 bases long annealed to a 47-base DNA template. Lane C contained TTP. The extension productswere separated hy gel no carhovir-TP and 10 p~ each of dATP, dCTP, dCTP, and electrophoresis andvisualized by autoradiography. The bandswere quantitated with the aid of a densitometer, and the relative velocity was calculated as described (20). A Lineweaver-Rurk plot (26, 25) of the data is shown.

Inhibition of HIV-1Reverse Transcriptase Carbovir-TP by

1760

RNA TEMPLATE

FIG.5. Incorporation of carbovirMP into DNA by HIV-1 reverse transcriptase in thepresence of dGTP. The experiment was done as described in the leaends to Fias. 3 and 4 except that all assays contilined 2 crl'.

T-

..

.

CCAAI

pM

.

?\.,

..

,. '$A&

T-d PA"[ 0

0.03 0.1 0.3

1

"c -1 "0 -A "c

'

3

10

I

-T .- -

- -

"

0 0.01 003 0 1 0.3

30

1

3

10

30

-

CARBOWR TP [MI

length and there is only accumulationof product 28 bases in length. The build-up of only this product indicates that car-A bovir-TP acts asa chain terminator. Because of the lack of a "T -1 natural stop site at the position just prior to the incorporation -G A is clearerusingthe DNA of thefirstdGMP,thisresult "G template (Fig. 5). As with the RNA template, only one product -G "c (20 bases in length) accumulated after incubation of HIV-1 -A reverse transcriptase with carbovir-TP. This is the length of -C "0 product expected for the incorporationof a chain-terminating "PnIllW deoxynucleotide analog with guanine as the base. No accumulation of product 19 bases long wasobserved in these 0 0.03 0.1 0.3 1 3 10 30 experiments, indicating that with the DNA template inhibiAZT-TP ( p M ) tion of HIV-1 reverse transcriptase by carbovir-TP was enFIG. 6. Incorporation of AZT-MP into the DNA by HIV-1 tirely due to the incorporation of carbovir-MP into thegrowreverse transcriptase in the presence of dTTP. HIV-1 reverse ing chain resultingin chain termination. A similar result was transcriptase was incubated with AZT-TP (0-30 p ~ ) 2, phi dTTP, 10 p~ each of dATP, dCTP, and dCTP, and '"P-labeled primer A obtained with AZT-TP (Fig. 6), indicating that the observed annealed to 16 S rRNA. The extension products were separated by inhibition of HIV-1 reverse transcriptase was a result of chain termination. This conclusion differs from that of Kedar et 01. gel electrophoresis and visualized by autoradiography. (41) who suggestedthat low concentrations of AZT-TP inhibbetween the primer and the nucleotide of interest, then the ited HIV-1 reverse transcriptase primarilyby reversible comreaction will not proceed to the base of interest.) However, petitionwith d T T P in a n in vitro assaywithpoly(rA). p(dT)12-IRasthe enzyme's template.Ono et al. (48) also from the datawe were able to obtain (all analog triphosphates on both templates and d T T P using the DNA template), we concluded from similar kinetic analysis that AZT-TP inhibwould conclude that there is no difference between running ited the reverse transcriptase from another retrovirus, avian and standing startsin the way the enzyme inserts analogsin myeloblastosis virus, by reversible competition with dTTP. These primer extension assaysallow us to directly assess the the growing cDNA chain. After completion of these experiments it was clear that all relative role of the incorporation of AZT-MP compared to of these deoxynucleotide analogs were chain terminators using that of reversible competitive inhibition of the enzyme withboth templates. However, it was still not clearhow important out having to make the assumptions used by others. Therefore, we feel that our data reflect more accurately the actual this was to the inhibition of HIV-1 reverse transcriptase. It was possible that the inhibition of HIV-1 reverse transcriptase mechanism of inhibition of HIV-1 reverse transcriptase by by these deoxynucleotide analogs was due to a dual mecha- AZT-TP. The ability of the host enzymes to utilize these deoxynunism involving both competition a t low concentrations and incorporation a t higher concentrations asproposed for AZT- cleotide analogs as substrates for DNA synthesis was also TP by Kedar et al. (41). Therefore, the ability of HIV-1 reverse evaluated. DNA polymerase (Y did not incorporate AZT-MP transcriptase to incorporate both carbovir-MP and AZT-MP into the DNA chain a t concentrations of AZT-TP as high as into the growing DNA strand in the presenceof either dGTP 1 mM. However, incorporation of AZT-MP into the DNA by or dTTP, depending on the analog being studied, was deter- DNApolymerase B and y wasobserved at concentrations mined. Under these circumstances, itwas reasoned thatif the which could be achieved in cell culture. Both enzymes were growing DNA deoxynucleotides were acting solely as chain terminators, thencapable of incorporatingAZT-MPintothe we would expect to see the accumulation of DNA products of strand with similar K , values (100 PM each). Recently, AZT a size consistent with the additionof the nucleotide. Since we has been shown to be incorporated into the DNA of human bone marrow cells (34) and K562 cells (491, and it has been have shown that these compounds are competitive with the natural deoxynucleotide, reversible inhibition should resultin proposed that this incorporation may be responsible for the the accumulation of a DNA product that is one base shorter toxicity observed with treatment with AZT. However, these responsible for the thanthat expected if theanalog were incorporated. The studiesdidnotdeterminetheenzyme results obtained with carbovir-TP can be seen in Fig. 5. As incorporation.Fromour work with isolated enzymes(also DNA synthesis on the RNA template is inhibited by carbovir- Ref. 2) it is probable that DNA polymerase /j.and y. but not TP, there is no change in the amountof product 27 bases in DNA polymerase CY, are responsible for the incorporation of "

Inhibition of HIV-1 Reverse Transcriptase by Carbouir-TP

1761

REFERENCES AZT into the DNA. We have not evaluated DNA polymerase 1. Furman, P. A., Fyfe, J. A., St. Clair, M. H., Weinhold, K., Rideout, 6, and therefore, can not exclude this enzyme as a potential J. L., Freeman, G.A., Lehrman, S. N., Bolognesi, D. P., Broder, source for the incorporation of AZT-MP into the DNA of S., Mitsuya, H., and Barry, D. W. (1986) Proc. Natl. Acad. Sci. host cells. No incorporation into the DNAof carbovir-MP (1 U. S. A . 83,8333-8337 mM) by DNA polymerase B was observed. Carbovir-TP was a 2. St. Clair, M. H., Richards, C. A., Spector, T., Weinhold, K. J., substrate for DNA polymerase a and y although the K, for Miller, W. H., Langlois, A. J.,andFurman,P. A. (1987) Antimicrob. Agents Chemother. 31, 1972-1977 incorporation was high (40 pM and greater than 1 mM, re3. Matthes, E., Lehmann, Ch., Scholz, D., von Janta-Lipinski, M., spectively). DNA polymerase cy and p incorporated ddGMP Gaertner, K., Rosenthal, H. A., and Langen,P. (1987) Biochem. and ddTMP into the DNA at high concentrations. Of note Biophys. Res. Commun. 148, 78-85 was the small amountof ddGMP (and ddTMP) incorporated 4. Hao, Z., Cooney, D. A., Hartman, N. R., Perno, C. F., Fridland, into the DNA by DNA polymerasey even though this enzyme A., DeVico, A. L., Sarngadharan, M. G., Broder, S., and Johns, D. G. (1988) Mol. Pharmacol. 34,431-435 of 0.03 p ~ )suggesting , was potently inhibitedby ddGTP (Ki 5. Hao, Z., Cooney, D. A,, Farquhar, D., Perno, C. F., Zhang, K., that inhibition of DNA polymerase y by ddGTP may be due Masood, R., Wilson, Y., Hartman, N. R., Balzarini, J., and t o reversible competition with dGTP and not by chain terJohns, D. G. (1990) Mol. Pharmacol. 37, 157-163 mination. Furtherwork is necessary to more adequately char- 6. Chen, M.S., Oshana, S. C. (1987) Biochem. Pharmacol.36,4361acterize the mechanism of inhibition of DNA polymerase y 4362 7. Johnson, M. A,, Ahluwalia, G., Connelly, M. C., Cooney, D. A., by dideoxynucleotide analogs. Broder, S., Johns, D. G., and Fridland, A. (1988) J. Biol. Chem. Insummary, carbovir isaninteresting deoxynucleoside 263,15354-15357 analog with anti-HIV activity that we feel should be further 8. Mitsuya, H., Jarrett, R. F., Matsukura, M., Veronese, F. D. M., evaluated for the treatment of AIDS. In this work we have DeVico, A. L., Sarngadharan, M. G., Johns, D. G., and Reitz, developed HIV-1 reverse transcriptase assays using natural M. S. (1987) Proc. Natl. Acad. Sci. U. S. A . 84, 2033-2037 9. Starnes, M. C., and Cheng,Y.-C. (1987) J . Biol. Chem.262,988RNAandDNAtemplatestoevaluatethemechanism of 991 inhibition of this enzyme by carbovir-TP and to compare the 10. Richman, D. D., Fischl, M. A., Grieco, M. H., Gottlieb, M. S., effect of carbovir-TP with those of other relevant anti-HIV Volberding, P. A., Laskin, 0. L., Leedom, J. M., Groopman, J. deoxynucleoside analogs. Although it appears as though carE., Mildvan, D., Hirsch, M. S., Jackson, G. G., Durack, D. T., bovir inhibits HIV-1 growth due to the inhibition of HIV-1 Nusinoff-Lehrman, S., and the AZTCollaborative Working Group (1987) New Engl. J . Med. 317, 192-197 reverse transcriptase by its triphosphate, similar to that observed with the other anti-HIV deoxynucleoside analogs, we 11. Sommadossi, J.-P., Carlisle, R., Schinazi,R. F., and Zhou, Z. (1988) Antimicrob. Agents Chemother. 32,997-1001 believe that it is sufficiently different from these analogs to 12. Vince, R., Hua, M., Brownell, J., Daluge, S., Lee, F., Shannon, warrant evaluation in the clinic as a possible anti-AIDS drug. W. M., Lavelle,G. C., Qualls, J., Weislow, 0. S., Kiser, R., One of its most important characteristics is its lack of activity Canonico, P. G., Schultz, R. H., Narayanan, V. L., Mayo, J. G., Shoemaker, R. H., and Boyd, M. R. (1988) Biochem. Biophys. against human polymerases, indicating that these enzymes Res. Commun. 156,1046-1053 will not be inhibited by pharmacological doses of carbovir. 13. White, E.L., Parker, W. B., Macy, L. J., Shaddix,S. C., McCaleb, Indeed, in vivo toxicity experiments in primates have found G., Secrist, J. A., 111, Vince, R., and Shannon, W. M. (1989) carbovir to have very low toxicity. A specific difference beBiochern. Biophys. Res. Commun. 161,393-398 tweencarbovir and the dideoxynucleosides is its relatively 14. Brosius, J., Dull, T. J., Sleeter, D. D., and Noller, H. F. (1981) J. Mol. Biol. 148, 107-127 low activity against DNA polymerase y. If the peripheral neuropathy of the dideoxynucleosides is dueto their inhibition 15. Cloyd, M. W., and Moore, B. E. (1990) Virology 174, 103-116 a d , E., Mitchener, J., Lee, L., and Baril, B. (1977) Nucleic of DNA polymerase y as has been suggested (35), then car- 16. B Acids Res. 4 , 2641-2653 bovir could be combined with thedideoxynucleosides without 17. White, E. L., Shaddix, S. C., Brockman, R. W., and Bennett, L. exacerbation of this particular toxicity. Because AZT-resistL., Jr. (1982) Cancer Res. 42, 2260-2264 18. Lane, D. J., Pace,B., Olsen, G. J., Stahl, D. A., Sogin, M. L., and a n t isolates of HIV-1 have been obtained from patients treated Pace, N. R. (1985) Proc. Natl. Acud. Sci. U. S. A . 82, 6955with AZT (50, 51), there is great need to develop new drugs that are active against viral isolates that are resistant AZT.to 19. 6959 Parker, W. B., Bapat, A. R., Shen, J.-X., Townsend, A. J., and Fortunately, most other anti-HIV deoxynucleoside analogs, Cheng, Y.-C. (1988) Mol. Pharmacol. 34, 485-491 including carbovir, are active against HIV-1 strains resistant 20. Boosalis, M. S., Petruska, J., and Goodman,M. F. (1987) J . Biol. Chem. 262,14689-14696 t o AZT (52), indicating that these drugs may have a role either as second line therapy of HIV infection after AZT 21. Randall, S. K., Eritja, R., Kaplan, B. E., Petruska, J., and Goodman, M. F. (1987) J. Biol. Chem. 262, 6864-6870 failure or in combination with AZT to prevent the occurrence 22. Petruska, J., Goodman, M. F., Boosalis, M. S., Sowers, L. C., of drug resistant strainsof HIV-1. Two other chemical charCheong, C., and Tinoco, I., Jr. (1988) Proc. Natl. Acad. Sci. U. acteristics of carbovir that distinguish this compound from S. A. 85,6252-6256 dideoxynucleosides, such as ddI, is its acid stability and the 23. Mendelman, L. V., Boosalis, M. S., Petruska, J., and Goodman, M. F. (1989) J . Biol. Chem. 264, 14415-14423 resistance to cleavage of the bond between the base and the 24. Boosalis, M. S., Mosbaugh, D. W., Hamatake, R.,Sugino, A,, cyclopentane ringby purine nucleoside phosphorylase. Acknowledgment-Discussions with L. Lee Bennett, Jr. regarding the enzyme inhibition studies are gratefullyacknowledged. Addendum-Since submission of this work, Huang et al. (54) have published ontheincorporation of AZT-MPinto DNA by HIV-1 reverse transcriptase and human DNA polymerase a and 6. In agreement with our conclusions, they showed that the inhibition of HIV1 reverse transcriptase was due to the incorporation of AZT-MP into the growing DNA chain with subsequent inhibitionof chain elonga6 did not tion.Inaddition,they showed that DNApolymerase incorporate AZT-MP into the DNA.

25. 26. 27. 28. 29.

Kunkel, T. A., and Goodman, M. F. (1989) J. Biol. Chem. 264, 11360-11366 Mendelman, L. V., Petruska, J., and Goodman, M. F. (1990) J . Biol. Chem. 265, 2338-2346 Haldane, J . B. S., and Stern,K. G. (1932) Allgemeine Chemie der Enzyme, pp. 119-120, Steinkopff Verlag, Dresden and Leipzig Lineweaver, H., and Burk, D. (1934) J . Am. Chem. Soc. 56,658666 Cheng, Y.-C., Dutschman, G. E., Bastow, K. F., Sarngadharan, M. G., and Ting, R. Y. C. (1987) J . Biol. Chem. 262, 21872189 Yong, W. H., Wyman, S., and Levy, J. A. (1990) AIDS (Lond.) 4, 199-206

1762

Inhibition of HIV-1 Reverse Transcriptase Carbovir-TP by

30. Vince, R., and Brownell, J. (1990) Biochem. Biophys. Res. Commun. 168,912-916 31. Rose, L. M., Schmid, S. M., Tomaszewski, J. E., Greishaber, C. K., and Page, J. G. (1990) Proc. Am. Assoc. Cancer Res. 3 1 , 2270 32. Balzarini, J., Pauwels, R., Baba, M., Herdewijn, P., de Clercq, E., Broder, S., and Johns, D. G. (1988) Biochem. Phurmacol. 3 7 , 897-903 33. Ho, H.-T., and Hitchcock, M. J. M. (1989) Antimicrob. Agents Chemother. 33,844-849 34. Sommadossi, J.-P., Carlisle, R., and Zhou, Z. (1989) Mol. Pharmacol. 3 6 , 9-14 35. Chen, C.-H., and Cheng, Y.-C. (1989) J. Bwl. Chem. 264,1193411937 36. Ono, K., Nakane, H., Heredewijn, P., Balzarini, J., and De Clercq, E. (1989) Mol. Phurmacol. 35,578-583 37. Simpson, M.V., Chin, C.D., Keilbaugh, S. A., Lin, T.-S., and Prusoff, W. H. (1989) Biochem. Pharmucol. 3 8 , 1033-1036 38. Konig, H., Behr, E., Lower, J., and Kurth,R. (1989) Antimicrob. Agents Chemother. 33, 2109-2114 39. Yagura, T., Kozu, T., and Seno, T. (1983) J. Biochem. (Tokyo) 9 1,607-618 40. Agarwal, R. P., Busso, M. E., Mian, A. M., and Resnick, L. (1989) AIDS Res. Hum. Retroviruses 5 , 541-550 41. Kedar, P. S., Abbotts, J., Kovacs, T., Lesiak, K., Torrence, P., and Wilson, S. H. (1990) Biochemistry 29,3603-3611 42. Townsend, A. J., and Cheng, Y.-C. (1987) Mol. Pharmucol. 3 2 , 330-339

43. Reid, R., Mar, E.-C., Huang, E.-S.,and Topal, M.D. (1988) J. Biol. Chem. 263,3898-3904 44. Ohno, Y., Spriggs, D., Matsukage, A,, Ohno, T., and Kufe, D. (1988) Cancer Res. 48,1494-1498 45. Ohno, Y., Spriggs, D., Matsukage, A., Ohno, T., and Kufe, D. (1989) Cancer Res. 49,2077-2081 46. Hentosh, P., Koob, R., and Blakley, R. L. (1990) J. Bwl. Chem. 265,4033-4040 47. Ricchetti, M., and Buc, H. (1990) EMBO J. 9 , 1583-1593 48. Ono, K., Ogasawara, M., Iwata, Y., Nakane, H., Fujii, T., Sawai, K., and Saneyoshi, M. (1986) Biochem. Biophys. Res. Commun. 140,498-507 49. Vazques-Padua, M. A., Starnes, M. C., and Cheng, Y.-C. (1989) Proc. Am. Assoc. Cancer Res. 3 0 , 2386 50. Larder, B. A,, Darby, G., and Richman, D. D. (1989) Science 243,1731-1734 51. Rooke, R., Tremblay, M., Soudeyns, H., DeStephano, L.,Yao, X.-J., Fanning, M., Montaner, J. S. G., O'Shaughnessy, M., Gelmon, K., Tsoukas, C., Gill, J., Ruedy, J., Wainberg, M. A., and theCanadian Zidovudine Multi-Centre StudyGroup (1989) AIDS (Lond.)3,411-415 52. Larder, B. A., Chesebro, B., and Richman, D. D. (1990) Antimicrob. Agents Chemother. 3 4 , 436-441 53. Dixon, M., and Webb, E. C. (1979) Enzymes, 3rd Ed., pp. 345353, Academic Press, New York 54. Huang, P., Farquhar, D., and Plunkett, W. (1990) J. Biol. Chem. 265,11914-11918