the journal of vol. 268, no. 14, issue of may 15, pp. 10312 … · 1999-01-13 · the journal of...

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 268, No. 14, Issue of May 15, pp. 10312-10323,1993 Printed in U.S.A.

Mechanism of HIV- 1 Reverse Transcriptase TERMINATION OF PROCESSIVE SYNTHESIS ON A NATURAL DNA TEMPLATE IS INFLUENCED BY THE SEQUENCE OF THE TEMPLATE-PRIMER STEM*

(Received for publication, October 15, 1992)

John AbbottsS, Katarzyna Bebeneks, Thomas A. Kunkeljq, and Samuel H. Wilson From the Laboratory of Biochemistry, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892 and the $Laboratory of Molecular Genetics, National Institute of Environmental Health Sciences, Research Triangle Park, North Carolina 27709

During processive DNA synthesis in vitro, the human immunoefficiency virus, type 1 (HIV-1) reverse transcriptase encounters template nucleotide positions at which continued synthesis is difficult. At these positions, the enzyme has a relatively high probability of dissociating from the template, and product molecules of corresponding length accumulate as the incubation proceeds. These positions, which are known as termination sites, could be associated with template secondary structures in some cases, but many termination sites appear to be template sequence-related rather than secondary structure-related. Mechanisms produc- ing these blocks in processive DNA synthesis are not well understood. In this study, to examine further the effects of template sequence on termination, we engineered selected single-base changes in the M13mp2 template, and we found that such changes can influence termination. Several general trends emerged from the study. First, strong termination sites rarely correspond to dATP as the “incoming” substrate opposite template T. Second, the sequence of the template-primer stem is more important for termination than the sequence of the single-stranded template ahead of the primer. Thus, we note the phenomenon of action at a distance: changing sequence at one nucleotide position in the template-primer stem alters termination at other positions, a few nucleotides distant at the primer 3‘ end. A and C as template bases in the template-primer stem have opposite effects. A is the strongest terminator residue, and C is the weakest terminator residue, fol- lowed by G. Since termination sites are produced by reverse transcriptase dissociation from the template- primer, the results suggest that the HIV-1 reverse transcriptase has properties reminiscent of a sequence- specific double-stranded DNA-binding protein in that its binding mechanism can distinguish both base residues and positions in the double-stranded DNA template-primer stem.

* This work was supported in part by the National Institutes of Health Intramural AIDS Targeted Antiviral Program. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduer- tisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ Holder of a National Research Council-National Institutes of Health Research Associateship while this work was done.

B To whom correspondence should be addressed Laboratory of Molecular Genetics E3-01, National Institute of Environmental Sci- ences, P. 0. Box 12233, Research Triangle Park, NC 27709. Tel.: 919- 541-2644: Fax: 919-541-7613.

When DNA-polymerizing enzymes conduct on natural DNA templates one cycle of synthesis (binding, processive DNA synthesis, and dissociation) the template position for termination of DNA synthesis is nonrandom. Termination occurs preferentially at certain nucleotide positions (Weaver and DePamphilis, 1982; LaDuca et al., 1983; Fairfield et al., 1983; Abbotts et al., 1988), and termination probabilities at different template positions can vary by orders of magnitude (Bebenek et al., 1989). Although the mechanism of termination is not well understood, the process can be studied exper- imentally because product molecules accumulate in a reaction mixture at chain lengths corresponding to termination sites, and these accumulated molecules can be easily quantified after sequencing gel electrophoresis. The specific nucleotide position at which termination preferentially occurs ( i x . a termination site) is presumably governed by a combination of factors, such as enzyme/template-primer contacts, effects of the incoming dNTP on stabilizing the pretransition state complex, and stabilization of the pyrophosphate leaving group. All of these factors could influence polymerase transit time across a given nucleotide position. It is possible that the longer the polymerase remains a t a specific template position the greater is the likelihood that the enzyme will dissociate from the template at that position (Abbotts et al., 1988).

Eucaryotic DNA polymerases in vitro exhibit an inverse correlation between processivity (i.e. lack of termination) and frameshift mutation frequency in an M13mp2 mutagenesis system. The more processive enzyme, DNA polymerase 7 , shows the lowest frameshift error rate; the least processive enzyme, DNA polymerase p, shows the highest propensity for frameshifts; and DNA polymerase a, a moderately processive enzyme, shows an intermediate frameshift error rate (Kunkel, 1985). The HIV-1’ reverse transcriptase, a processive DNA polymerase, shows a strong tendency to terminate DNA synthesis at certain nucleotide positions in the M13mp2 template system. Further, the mutation spectrum of the HIV-1 reverse transcriptase in the M13mp2 fidelity assay shows frameshifts within “runs” of three or more of the same nucleotide, and each of these runs has at least one position with relatively high probability of terminating DNA synthesis (Bebenek et al., 1989). These results indicate that template runs containing a strong termination site are hot spots for frameshift mutations by the HIV-1 reverse transcriptase.

In the present work, we have changed the wild type M13mp2 DNA template to produce two altered template

~~

The abbreviations used are: HIV-1, human immunodeficiency virus, type 1; wt, wild type; pwt, pseudowild type; AMV, avian myeloblastosis virus; Pol I, DNA polymerase I; aS-dNTP, 2”deoxy- nucleoside 5’-O-(l-thio)triphosphate derivatives.

~~~

10312

Processive Synthesis by HIV-1 Reverse Transcriptase 10313

TABLE I HIV-1 reverse transcriptase termination probabilities on M13mp2 templates

Conditions of DNA synthesis, analysis of products of synthesis, and determination of termination probabilities are described under “Experimental Procedures.” Termination Probabilities are normalized on each template to residue -84 by assigning a termination probability of 1.0 by definition. Termination probabilities were calculated through residue 171, the last residue at which a mutation could be detected (Bebenek et al.. 1989). Seauence residues on the M13 temalates are indicated in the accompanying manuscript (Bebenek et al., 1993).

Residue

-84 -83 -82 -81 -80 -79 -78 -77 -76 -75 -74 -73 -72 -71 -70 -69 -68 -67 -66 -65 -64 -63 -62 -61 -60 -59 -58 -57 -56 -55 -54 -53 -52 -51 -50 -49 -48 -47 -46 -45 -44 -43 -42 -41 -40 -39 -38 -37 -36 -35 -34 -33 -32 -31 -30 -29 -28 -27 -26 -25 -24 -23 -22 -21 -20 -19 -18 -17 -16 -15

. wt

1.000 0.180 0.740 0.048 0.021 0.021 0.020 0.025 0.620 0.030 0.016 0.067 0.009 0.024 0.006 0.003 0.011 0.072 0.006 0.019 0.006 0.069 0.006 0.003

<0.001 <0.001

0.006 0.009 0.004 0.004 0.038 0.004 0.003 0.003 0.001 0.001 0.001 0.006 0.004

(0.001 0.001 0.001 0.002 0.007 0.010 0.040 0.038 0.045 0.020 0.008 0.006 0.011 0.012 0.005 0.006 0.002 0.002 0.003 0.004 0.035 0.006 0.008 0.002 0.002 0.003 0.004 0.012 0.009 0.007

Pwtl 1.000 0.200 0.720 0.040 0.024 0.016 0.010 0.014 0.560 0.017 0.014 0.074 0.005 0.016 0.006 0.004 0.004 0.072 0.005 0.016 0.009 0.095 0.006 0.002 0.003 0.002 0.006 0.014 0.004 0.006 0.024 0.008 0.007 0.014 0.006 0.002 0.007 0.026 0.023 0.005 0.008 0.006 0.002 0.009 0.010 0.016 0.100 0.047 0.040 0.010 0.010 0.010 0.013 0.016 0.018 0.008 0.006 0.004 0.003 0.043 0.008 0.014 0.006 0.002 0.005 0.006 0.022 0.018 0.013

t0.001 a o o 1

Pwt2 1.000 0.160 0.740 0.072 0.027 0.027 0.014 0.026 0.580 0.034 0.018 0.072 0.013 0.019 0.006 0.006 0.007 0.050 0.007 0.013 0.010 0.100 0.005 0.004 0.003 0.005 0.006 0.014 0.002 0.003 0.016 0.008 0.007 0.008 0.004 0.006 0.003 0.004 0.007 0.007 0.001 0.006 0.013 0.005 0.025 0.070 0.220 0.160 0.080 0.064 0.024 0.110 0.074 0.037 0.008 0.006 0.003 0.003 0.005 0.030 0.004 0.008 0.003 0.002 0.002 0.003 0.028 0.016 0.013

Residue

-14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56

wt

<0.001 <0.001

0.050 0.002 0.002 0.005 0.018 0.007 0.004 0.011 0.002 0.004 0.012 0.120 0.046 0.010 0.019 0.012 0.031 0.130 0.017 0.094 0.011 0.064 0.014 0.065 0.230 0.062 0.011 0.050 0.038 0.049 0.029 0.028 0.007 0.006 0.004 0.012 0.053 0.100 0.017 0.020 0.009 0.019 0.009 0.002 0.005

<0.001 <0.001

0.003 0.008 0.001 0.002 0.005 0.004 0.037 0.019 0.008 0.013 0.017 0.018 0.085 0.053 0.014 0.012 0.100 0.022 0.110 0.005 0.003

Pwtl

0.004 0.008 0.063 0.007 0.011 0.004 0.014 0.011 0.004 0.017 0.003 0.007 0.013 0.130 0.045 0.014 0.022 0.008 0.026 0.110 0.011 0.054 0.009 0.051 0.016 0.071 0.260 0.066 0.009 0.051 0.054 0.077 0.040 0.051 0.006 0.004

<0.001 0.010 0.058 0.140 0.028 0.030 0.019 0.031 0.011 0.003 0.007 0.002 0.002 0.005 0.012 0.003 0.005 0.010 0.010 0.044 0.026 0.016 0.017 0.019 0.022 0.063 0.037 0.012 0.014 0.120 0.042 0.094 0.010 0.004

Pwt2

0.002 0.005 0.042 0.009 0.007 0.011 0.015 0.011 0.006 0.023 0.008 0.011 0.018 0.120 0.036 0.018 0.028 0.013 0.020 0.095 0.010 0.048 0.008 0.047 0.009 0.067 0.230 0.051 0.012 0.046 0.037 0.068 0.034 0.042 0.004 0.003 0.001 0.007 0.046 0.120 0.024 0.026 0.014 0.019 0.008 0.002 0.009 0.001 0.001 0.003 0.010 0.002 0.002 0.005 0.004 0.029 0.022 0.010 0.014 0.016 0.019 0.065 0.036 0.018 0.015 0.074 0.028 0.110 0.009 0.004

10314 Processive Synthesis by HIV-1 Reverse Transcriptase

TABLE I-continued "

57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99

100 101 102 103 104 105 106 107 108 109 110 111 112 113

0.005 0.002

<0.001 0.001

<0.001 0.003 0.007 0.023 0.011 0.008 0.056 0.023 0.022 0.180 0.013 0.018 0.008 0.012 0.004 0.005 0.030 0.007 0.013 0.100 0.005 0.004 0.037 0.004 0.019 0.022 0.015 0.008 0.046 0.083 0.043 0.014 0.002 0.002 0.002 0.001 0.004 0.004 0.009 0.007 0.007 0.002 0.058 0.007 0.002 0.003 0.008 0.024 0.024 0.026 0.004 0.030 0.014

0.005 0.005 0.004 0.006 0.004 0.004 0.010 0.008 0.007 0.007 0.014 0.007 0.016 0.046 0.017 0.010 0.004 0.010 0.008 0.005 0.036 0.009 0.013 0.040 0.004 0.004 0.028 0.005 0.024 0.024 0.012 0.014 0.069 0.084 0.052 0.020 0.006 0.004 0.004 0.002 0.006 0.009 0.010 0.011 0.012 0.010 0.180 0.024 0.006 0.008 0.018 0.016 0.017 0.036 0.008 0.024 0.012

0.008 0.003 0.002 0.001 0.001 0.004 0.008 0.027 0.011 0.021 0.140 0.030 0.039 0.110 0.018 0.008 0.005 0.008 0.002 0.002 0.026 0.005 0.006 0.046 0.005 0.003 0.010 0.004 0.004 0.001 0.004 0.004 0.013 0.007 0.011 0.005 0.004 0.002 0.002 0.002 0.010 0.005 0.008 0.012 0.018 0.006 0.076 0.017 0.004 0.005 0.018 0.014 0.016 0.032 0.006 0.056 0.013

Residue w t Pwt l D d 2

114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138

142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 168 170 171

139-141

0.004 0.004 0.010 0.004 0.003 0.040 0.007 0.043 0.008 0.005 0.006 0.005 0.002 0.006 0.002 0.002 0.001 0.001 0.002 0.002 0.004 0.008 0.026 0.020 0.012 0.018 0.001 0.003 0.004 0.002 0.022 0.004 0.032 0.035 0.010 0.028 0.080 0.003 0.015 0.034 0.024 0.052 0.078 0.006 0.029 0.012 0.008 0.028 0.022 0.016 0.026 0.022 0.035 0.020 0.034 0.006

0.002 0.003 0.006 0.006 0.002 0.037 0.004 0.004 0.008 0.007 0.011 0.009 0.008 0.012 0.004 0.009 0.006 0.010 0.002 0.002 0.003 0.006 0.016 0.012 0.010 0.030 0.002 0.004 0.005 0.002 0.011 0.004 0.054 0.044 0.013 0.026 0.082 0.004 0.018 0.028 0.053 0.033 0.074 0.004 0.021 0.010 0.004 0.020 0.024 0.014 0.019 0.014 0.022 0.024 0.026 0.005

___

~

0.004 0.004 0.006 0.005 0.002 0.024 0.004 0.010 0.007 0.006 0.008 0.006 0.002 0.006 0.002 0.002 0.002 0.002 0.002 0.002 0.003 0.006 0.016 0.005 0.008 0.020 (compression) 0.001 0.004 0.006 0.002 0.016 0.004 0.033 0.034 0.020 0.023 0.098 0.004 0.013 0.024 0.041 0.024 0.054 0.006 0.009 0.008 0.003 0.016 0.015 0.012 0.016 0.014 0.016 0.015 0.020 0.004

molecules, each with a different set of seven single-base changes. The base substitutions are described in further detail in the accompanying manuscript (Bebenek et al., 1993). These changes were designed to investigate the effects of altering sequence in or around a run on mutations and the effects of DNA sequence on termination. We report here that several of the single-base changes in the M13mp2 template are associated with significant changes in the termination pattern. Overall, the results suggest trends for HIV-1 reverse transcriptase termination on a natural DNA template and indicate the importance of sequence composition in the template- primer stem.

EXPERIMENTAL PROCEDURES

Materials Unlabeled dNTPs were obtained from Pharmacia LKB Biotech-

nology Inc. [y3*P]ATP was obtained from ICN Radiochemicals. Formamide and urea were obtained from Bethesda Research Labo- ratories. Electrophoresis grade acrylamide and bisacrylamide were obtained from Bio-Rad.

The HIV-1 reverse transcriptase, expressed in Escherichia coli from a plasmid, was obtained from Genetics Institute, Cambridge, MA. The expression plasmid contains the coding sequence of HIV-1 reverse transcriptase with the carboxyl-terminal Leu omitted (Huber et al., 1989). The final fraction of the expressed enzyme is devoid of 3' + 5' exonuclease activity (Roberts et al., 1988). When analyzed


by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, the final fraction is >90% pure and displays two major bands of approximately equal intensity at about 66 and 51 kDa, similar to results seen with the immunoaffinity-purified HIV enzyme derived from virions (di Marzo Veronese, et al., 1986). The specific activity of the expressed enzyme was 1000 nmol of dTMP incorporated/min/mg at 37 “C on a poly(rA) . oligo(dT) template-primer with dTTP-Mg. We have found that the recombinant enzyme and the reverse transcriptase purified from virions have approximately the same specific activity (Kedar et al., 1990; Abbotts et al., 1991). On an M13mp2 DNA template, the enzymes show similar termination patterns and similar mutation spectra (Bebenek et al., 1989).

The templates were M13mp2 DNA (designated “wild type”) and two M13mp2 DNA templates engineered with different combinations of seven single-nucleotide changes, designated pseudowild type 1 (pwtl) and pseudowild type 2 (pwt2). The templates and the rationale for each set of changes are described in the accompanying paper (Bebenek et al., 1993). Primers were synthetic oligodeoxynucleotides complementary to the M13mp2 lacZ sequence; primer 59 is complementary to positions 75-59, with the 3’-hydroxyl opposite position 59; primer 177 is complementary to residues 177-191, with the 3’- hydroxyl opposite position 177. DNA synthesis reactions described below were carried out with either primer and one of the three templates.

Methods Preparation of Labeled DNA Primers-Synthetic DNA primers

were 5’ end labeled with 32P in accordance with the procedure of Maxam and Gilbert (1980). DNA was extracted with phenol; residual phenol was removed with ether, and the DNA was precipitated with ethanol.

In Vitro DNA Synthesis with Labeled Primers-A 32P-labeled primer, either primer 59 or primer 177, was hybridized to one of the M13 DNA templates a t a 1:l molar ratio of primer to template by heating to 100 “C and slowly cooling to room temperature. DNA synthesis reactions (10 pl) contained 20 mM Tris-HCI, pH 8.0; 10 mM MgC12; 2 mM dithiothreitol; a 1 mM concentration each of dATP, dCTP, dGTP, and dTTP; 12 fmol of HIV reverse transcriptase; and 0.15 pg (63 fmol) of hybridized template-primer. Reaction mixtures were incubated at 37 ”C, and reactions were stopped by adding EDTA to a final concentration of 18 mM. A dye mixture in deionized formamide was then added to a total volume of 15 pl. 5 - ~ l portions

were loaded for gel electrophoresis. Chain Length Analysis of Products of DNA Synthesis-Products

were analyzed by gel electrophoresis as described previously (Detera and Wilson, 1982). Electrophoresis was conducted on gels containing 8-12% polyacrylamide and 7 M urea. The gel was prerun for 1 h at 40-45 watts without cooling, and electrophoresis was performed at 40-45 watts. After electrophoresis, the gel was transferred to What- man 3MM paper, covered with Saran Wrap, and dried in a Bio-Rad gel dryer. Products were visualized by autoradiography with Kodak XAR-5 film at room temperature. DNA sequencing reactions, to determine the positions of identified termination sites, were carried out by the dideoxynucleotide method (Sanger et al., 1977) using the appropriate M13 DNA template and either synthetic oligomer as 32P- labeled primer.

Quantification of Termination Probability-Exposed autoradiography films were evaluated for band intensity with a Zeineh soft laser scanning densitometer (Biomed Instruments). Quantification of band intensities and calculation of termination probabilities were performed as described previously; the termination probability at a given position is the number of product molecules at that position, divided by this same number plus the number of all longer product molecules (Abbotts et al., 1988). Termination probabilities were compared for DNA synthesis reactions on templates wt, pwtl, and pwt2, where the reactions were carried out under the same conditions and where synthesis reached residue -84, the 5’ position on the M13mp2 lacZ sequence (see, for example, Fig. 1, a and b); probabilities reported represent the average of at least two independent determinations. To establish the reproducibility of the termination probability analysis, triplicate measurements were made at 27 sites spanning a range of termination probabilities from 0.29 to 15% (data not shown). Stand- ard deviations were typically 10-40% of mean values, and in only one instance was a single value more than 2-fold higher than the average of the other two determinations. Thus, the analysis is generally reproducible, but an occasional outlying value is obtained.

The pattern of strong termination sites is reproducible, but termination probabilities reported here may have values that are different (2-fold in some cases) from those reported earlier on the wt template (Bebenek et al., 1989). In the earlier paper, we reported termination probabilities for residues in the 3’ region of the M13mp2 template based on short incubations of 5-15 min. In the present study termination probabilities were quantified for reactions in which synthesis proceeded to residue -84. The extent of synthesis was

TABLE I1 Nucleotide residues around thQ 45 strongest termination sites and 78 weakest termination sites on wild type M13mp2 DNA template with

HIV-1 reverse transcriptase A. The strong termination sites for DNA synthesis by HIV-1 reverse transcriptase on a 255-residue segment of M13mp2 template have

been identified previously (Bebenek et al., 1989). The sequences around these 40 sites, with changes (six additional sites, one removed, and one reassigned) determined in this work, were aligned from position +7 to -7, with the termination site itself representing position 0 and the + positions representing the direction of synthesis 3’ of the termination site. B. 78 sites from this work with the weakest termination were aligned in sequence in a similar manner. The table indicates the number of times each template base was scored at a given position. On a random basis, each base would be expected to occur one-fourth of the time. Boxed numbers are significantly different ( p 5 0.05) from random determined usine the normal aDDroximation to a binomial distribution (Mendenhall et al.. 1981) and are summarized as trends.

~~ ~

Incidence of residues at termination site and flanking sequence

Termination probability of site !::f N~~~~~~ . . ss template Site ds primer stem

+7 +6 +5 +4 +3 +2 +1 -1 -2 -3 -4 -5 -6 -7

A

B.

Strongest (25%) 45 A

T G C

Trend High termination

Trend: Low termination

Weakest (50.4%)

likelihood

likelihood

78 A T G C

Trend High termination

Trend Low termination likelihood

likelihood

11 10 9 14 14 10 1 2 11 16 [23] 14 9 16 [20] 16 10 16 16 13 12 10 [ l ] 15 15 10 15 17 9 8 10 15 11 11 9 7 6 1221 8 7 7 10 9 10 [4] 9 9 8 9 9 12 [19] 10 11 7 [5] 6 10 10 13 10 n n n n n C G n n A n n n A n

n n n n n n T n n C n n n G n

18 21 24 23 21 17 23 20 17 [9] 18 16 [8] [7] 13 23 23 17 18 19 25 23 18 20 19 14 15 26 24 24 13 17 20 15 18 14 [5] 13 15 19 17 22 19 20 21 24 17 17 22 20 22 1271 [27] 26 [31] 1291 25 25 1271 20 n n n n n n G n n A n n A A n

n n n n n n C C n C C n n C n


similar in these simultaneous reactions, and we conclude that these conditions provide a valid comparison of termination probabilities across the three templates.

Termination probabilities are normalized for each template to residue -84, which was assigned a termination probability of 1.0 by definition. This convention is appropriate because termination patterns and intensities do not change significantly on templates wt, pwtl, and pwt2 in the region 5' of residue -40. When one examines the absolute termination probabilities, compared with downstream product molecules, at the three termination sites at residues -76, -82, and -84, the following values are obtained. At residue -76, termination probabilities are 0.345,0.343, and 0.282 on templates wt, pwtl, and pwt2, respectively; at residue -82, values are 0.225, 0.265, and 0.230; and at residue -84, values are 0.099, 0.122, and 0.074. These differences are small enough such that normalizing termination on each template to residue -84 is a reasonable approach for evalu- ating termination changes across templates. With these conventions, Table I gives termination probabilities for each residue on the three templates under identical reaction conditions and synthesis on each template to residue -84.

RESULTS

Tabulation of Strong Termination Sites for Reverse Tran- scriptases of HIV and AMV-In our previous work, we identified on the M13mp2 DNA template 40 strong termination sites for the reverse transcriptase of HIV-1 and 15 strong termination sites for the reverse transcriptase of avian myeloblastosis virus (AMV) (Bebenek et al., 1989). Some level of termination occurs at each residue along the template, but sites with termination probabilities of 5% or greater were considered strong termination sites. As a result of the present work, we have made some additions and corrections to the earlier identification. Because of better resolution of positions in or near the lac2 regulatory region of M13mp2, we have identified new strong termination sites for HIV-1 reverse transcriptase at residues -67, -63, -12, 8, 10, and 12 and have lowered the termination probability assigned for residue -39 (Table I). In addition, the strong site for HIV-1 reverse transcriptase at residue 118 has been reassigned to residue 119; better resolution of a compression in the sequencing lanes allowed this new assignment (data not shown).

To examine if strong termination sites might be associated with nucleotide sequences of the template, we tabulated the sequences at which such sites appeared in the earlier work (Bebenek et al., 1989), with the additions and corrections from present work, for a total of 45 sites (Table IIA). Template sequences corresponding to each strong termination site were examined from positions +7 to -7, where the termination site itself was designated as position 0, and plus (+) positions were in the direction of synthesis relative to the termination site.

When the tabulated sequences around the 45 strongest termination sites are considered for the HIV-1 reverse transcriptase (Table IIA), no specific consensus sequence appears, and the base at the termination site itself shows little deviation from random (11-12 occurrences): 11 times the base is

an A, 11 times a C, 8 times a G, and 15 times a T. At other positions, however, the occurrence of given bases is significantly different from random. The most striking deviation occurs at position +1, where only once in 45 termination sites is the base a T; thus, termination is much less likely to occur with the combination of template T and incoming nucleotide dATP. A second deviation from random also is noted at the +1 position. 22 of the 45 sites contained a template G residue, corresponding to incoming dCTP. Hence, the difference in occurrence of T versus G at the template position +1 is 1:22, indicating an effect on termination by the incoming dNTP substrate/template residue combination.

Other than position +1, the most striking deviations occur in the template-primer stem at positions -2 and -6; at each position, one nucleotide is scored much more frequently and another much less frequently, than random. At position -2, A occurs with more than one-half (23) of the termination sites, and the bases C and G occur only five and seven times, respectively. At position -6, A occurs 20 times, whereas G occurs only four times. This search for average or consensus nucleotide assignment for an HIV-1 reverse transcriptase terminator sequence reveals (Table IIA) that positions +2, +1, -2 and -6 of the 15 template positions tabulated show a strong deviation ( p value 5 0.05) from random: C at the +2 position, G at the +1 position, and A at positions -2 and -6. Nucleotides and positions that are inversely correlated with the strongest termination sites are tabulated in Table IIA also. As noted above, the most striking lack of correlation with termination occurs at template position +1 where T occurs once in the 45 termination sites. C at position -2 and G at position -6 occur severalfold less frequently than random.

These trends are corroborated by a similar analysis of the 78 sites in the M13mp2 wild type template which show the lowest termination probabilities, 5 0.4% (Table IIB). Thus, G a t +1 occurs only five times, or about 4-fold less than random (19-20 occurrences), and A a t positions -2, -5 and -6 occurs 2-3-fold less than random. By contrast, C a t positions +1,0, -2, -3, and -6 were more frequent than random, and the template moiety of the entire 7-residue template- primer stem was C-rich. That higher deviations occur more often at the minus positions suggests that enzyme contacts in the template-primer stem are more important in determining termination than contacts in the single-stranded template region. Finally, extending this analysis in each direction to positions +15 and -15 revealed random occurrence for nucleotide residues a t all of these more distal positions (not shown).

For the AMV reverse transcriptase, deviations from random for sequence around strong termination sites are seen also (Table 111). Some deviations from random occur a t positions different from those seen with HIV-1 reverse transcriptase. For example, a deviation from random (equal to 4) is seen at

TABLE I11 Nucleotide residues around the 15 strongest termination sites on wild type M13mp2 DNA template with A M V reverse transcriptase

Tabulation of sequences for the 15 strongest termination sites for AMV RT on the M13mp2 template (Bebenek et al., 1989). The table indicates the number of times each template base was scored at a given position, as described in Table 11. Underlined numbers are significantly different (p < 0.05) from random, based on a binomial distribution. "

Incidence of residues at termination site and flanking sequence Termination probability No. of Nucleotide

of site sites residue ss template Site ds primer stem

+7 +6 +5 +4 +3 +2 +I " -1 -2 -3 -4 -5 -6 -7

Strongest (25%) 15 A 4 4 2 3 2 3 7 6 1 0 5 3 4 2 5 2

T 2 2 3 3 5 4 2 3 3 4 7 7 5 7 6 G 7 8 4 3 3 2 3 1 ! 1 _ 0 3 1 2 5 C 2 1 6 6 5 6 3 5 2 5 5 1 2 3 2 ~~ ~~


a) b) 1 2 3 4 5 6 1 2 3 4 5 6

3 -32--38 “84 .-e4

-67 3 89-91

‘121

”””

“38 +-33

-25

C ) 1 2 3 4 5 6

~26

+67 ‘70

J ‘89-91

103

FIG. 1. HIV-1 reverse transcriptase termination on M13 templates. DNA synthesis reactions were carried out with the HIV- 1 reverse transcriptase on M13mp2 templates using labeled primers, and products of synthesis were displayed by gel electrophoresis, as described under “Experimental Procedures.” Panel a, synthesis was carried out with labeled primer 177. Electrophoresis was conducted on a 12% polyacrylamide, 7 M urea gel, until unextended primer 177 ran nearly to the bottom of the gel. DNA synthesis conditions were as described under “Experimental Procedures” in duplicate 30-min incubations and these templates: w t (lanes Z and 2 ) , pwtl (lanes 3 and 4 ) , pwt2 (lanes 5 and 6 ) . Positions -84 and 177 are indicated for reference by dots. Other positions are indicated where changes in termination occur; these changes are described in the text. Panel b, synthesis was carried out with labeled primer 59. Electrophoresis was conducted on an 8% polyacrylamide, 7 M urea gel, until the position at the bottom of the gel was approximately -15. DNA synthesis reactions were as described under “Experimental Procedures,” with these differences in template and reaction time: wt, 15 min (lane Z 1; wt, 30 min (lane 2) ; pwtl, 15 min (lane 3); pwtl, 30 min (lane 4 ) ; pwt2.15 min (lane 5); pwt2,30 min (lane 6 ) . Positions -84 and -25 are indicated for reference; positions -33 through -38 are indicated as the region in which changes in the termination pattern occur. Panel c, synthesis was carried out with labeled primer 177. Electro- phoresis was conducted on an 8% polyacrylamide, 7 M urea gel, until the position at the bottom of the gel was approximately 115. The samples displayed were the same as those in lanes 1-6 of panel a, except lanes 5 and 6 are reversed. Positions 26 and 103 are indicated for reference; other positions are indicated where changes in the termination pattern occur.

the termination site itself, where only once does a G occur. An effect of an incoming nucleotide/template residue combination also is seen, where seven times an A residue occurs a t position +1. As noted with HIV-1 reverse transcriptase, significant deviations from random with AMV reverse transcriptase are greater at minus positions. The most frequent occurrences of a base are 10 times an A a t position -1 and 9 times an A a t position -5; the least frequent occurrences are none a t positions -1 ( G ) , -3 ( G ) , -5 (C), and -6 ( G ) . There are,

in addition, some deviations from random a t plus positions (e.g. 8 G a t +6), but overall the data with AMV reverse transcriptase suggest that the template-primer stem is important in determining termination and that there is an opposite effect for A versus C or G, A being terminatory and C or G not terminatory. An additional suggestion from Tables I1 and I11 is that it might be possible to modulate termination a t one position by making nucleotide changes at the corresponding template-primer stem positions.

Termination Changes on the Altered Templates-The preparation of altered M13mp2 templates provided the opportu- nity for us to examine the effect of a limited number of single- base changes on termination probability of the HIV-1 reverse transcriptase. Fig. l a illustrates how the termination pattern changes across three templates, wt and two altered templates, pwtl and pwt2. Many of the same strong termination sites are maintained across all three templates. However, there are several sites a t which changes can be observed by inspection of the autoradiogram, and these are indicated on Fig. la. Termination a t residue 121, a moderate site on wt template, is considerably reduced on templates pwtl and pwt2. At residues 89-91, termination is significantly diminished on pwt2. The strength of termination at residues 67 and 70, relative to each other, changes across the three templates. As will be described below, changes in the termination patterns also occur in the region -32 to -38. Changes also can be seen in the region 130-144, and these are described below as well. An additional observation is that changes in termination tend to be limited to those regions in which nucleotide sequence changes were made. For example, no changes were made to the template sequence between positions 69 and -30, and changes in the termination pattern between positions -29 and 63 are not significant.

Fig. l b shows the -30 to -40 region in more detail, display- ing the products of reaction with the HIV-1 reverse transcriptase and labeled primer 59. We have noted previously that changing the primer position does not significantly change the termination pattern at residues downstream of the primer (Abbotts et al., 1988; Bebenek et al., 1989), and we find the termination pattern with a given M13mp2 template downstream of residue 59 is very similar with primer 59 and with primer 177 (data not shown). When the three templates are compared, one sees that the wild type template shows three termination sites a t residues -37, -38, and -39 and a weaker site a t residue -36; with pwtl, the strongest termination site is seen at residue -38, and termination at residues -37 and -36 are decidedly weaker; with pwt2, one observes several strong termination sites in the -30 to -40 region, with the strongest appearing a t -32, -33, -37, and -38.

Fig. IC displays another region in which the termination pattern changes as a function of template sequence. On the wt template, termination at residue 70 is much stronger than a t residue 67. On pwtl, termination at both positions is reduced, compared with termination at downstream sites. On pwt2, termination a t residue 67 is stronger than at residue 70. Another change is seen in the region of residues 89-91, a t which both wt and pwtl show a triplet of strong termination sites, but termination at these positions is reduced on template pwt2.

Quantitative Assessment of Termination Probabilities-Ter- mination probability can be quantified a t each position on the M13mp2 template, allowing a more detailed examination of changes across the three templates. Fig. 2, panels a-f, illustrates the regions on the M13mp2 templates at which termination probabilities change significantly and represents all the regions at which nucleotide sequence changes were made with templates pwtl and pwt2. We found that changes


FIG. 2. HIV-1 reverse transcriptase termination probabilities on M13mp2 templates. Termination probabilities were determined from the autoradiography films of displayed products of DNA synthesis as described under “Experimental Procedures.” Posi- tions of products of synthesis were determined with parallel DNA sequencing reactions. Probabilities are indicated for termination under the same conditions on templates wt (solid bars), pwtl (striped bars), and pwt2 (stippled bars). Termination probabilities and template sequences are indicated for six different template regions in panels a-f. Panels a- d and f indicate all of the seven single- base sequence changes made in templates pwtl and pwt2. For the region in panel e, the sequence was the same on all three templates.

E 0 - L

m E .- E L Q I-

- 4 4 ~ a a ~ x t z a ~ ~ ~ ~ ~ ~ ? a a 1 r ) m - - n w t : C C C C A G C C T T T A C A C T T T pWt1: A G put2 : A A

M13, 62 to 76 0.2 $1

62 63 64 65 66 67 68 6 9 76 71 72 73 7 4 75 7 6 n t : G G C C G T C G T T T T A C A put1 : C pwt2: A

c) M13, 81 t o 99

.- “ - ,x 0.08

n CI 0.06 2 m

0.

E 0.04 .- 0

m c.

.s 0.02 E L Q

I- 0.00 81828384858687888990919293949596919899

u t : C G T G A C T G G G A A A A C C C T G pwtl: A put2 : G


FIG. 2-continued

M13, 100 to 110 0.2 I I

0.0 100 101 102 103 104 105 106 107 108.109 110

w t : G C G T T A C C C A A p w t l : C T put2 : A T

M13, 115 to 125

115 116 117 118 119 120 121 122 .I23 124 125 A T C G C C T T G C A

0*04 * M13, 130 to 144

C 0 .- c m C .- L E

I- a)

t

130131132133134135136137138 142143144 * : A T C C C C C T T T C G C . C A pnt 1 : A put2 : A


TABLE IV List of residues in which increase in termination probability was 20-

fold or greater

Nucleotide Templates Change in position residue compared sequence of change in

template increase

86 86 90 90

-33 121 -33

67 131

pwt2 us. pwtl pwt2 us. wt pwt2 us. wt pwt2 us. pwtl pwtl us. pwt2 pwtl us. wt wt us. pwt2 pwtl us. pwt2 wt us. pwtl

G - A G - A G - + A G - A C - A A - T C - A C - A T + A

-6 -6 -2 -2 -3

-16 -3 -2 -6

24 22 12 12 11 11 10 10 10

in the termination pattern which can be detected by visual inspection are reflected by changes in termination probability values; more subtle changes in termination pattern usually do not produce significant changes in probability values (Table

Fig. 2a indicates the termination probabilities in the region -27 to -44. Sequence changes in this region were made a t residues -37 and -41 on template pwtl and at residues -30 and -37 on pwt2. The most striking termination changes in this region occur with template pwt2; although only 2 bases are changed, changes in termination probability are seen at residues -31 to -40. At 4 different residues (-31, -32, -33, and -35), differences exceed 6-fold, with the greatest difference being 11-fold (residue -33). The data suggest that template-primer stem differences influence termination in this region. As indicated in Table IIA, the incoming dNTP/+1 template residue combination may also influence termination. At residue -36, incoming dCTP or dTTP appears to be associated with a higher termination probability than incoming dGTP.

Fig. 2b indicates sequences and termination patterns near residue 69. Residue 70 is a very strong termination site on the wt template, and the termination probability at this residue changes with sequence changes in pwtl and pwt2. The changes in termination a t residue 70 may be related to changes in the incoming substrate/template residue combination and are consistent with a terminator influence of incoming dCTP (Table IIA). Significant termination changes are also seen at residue 67. Thus, termination is 10-fold more frequent with the pwt2 template than for pwtl, and the only difference is the nucleotide change a t residue 69. The strongest termination occurs when residue 69 is an A, which is consistent with an influence on termination by A at the -2 position in the template-primer stem (Table IIA).

Fig. 2c indicates changes in the region of residues 81-99. When the base at residue 98 is changed to A in pwtl, there are only small changes in the termination pattern. However, when residue 92 is changed to G, termination decreases at the downstream residues 83 and 85-92. Again, the effects of a single-nucleotide change are substantial, reducing termination by 12-fold a t residue 90 and 22-fold at residue 86. These observations provide further support for the idea that sequences in the template-primer stem region are important in influencing termination.

For the template region shown in Fig. 2d, the major change in termination occurs at residue 103. The change in sequence at residue 107 does not seem responsible, since the base at this residue is the same in pwtl and pwt2. Yet the base at residue 102 is different on each template, as is termination at residue 103. The suggestion is that the incoming nucleotide/ +1 template position combination has the dominant effect.

In Fig. 2e, changes are seen in termination at residues 119

I).

. . . I . . . , . . . , . . . , . . .

X

0 0.2 0.4 0.6 0.8 1

Termination Probability

1 3 c . . . . . . . . . . . . . . . . . . . ,

X X

X

X

X

@ X x x X

X

x x c 3 x X X

X X

x x x X X

e

X

3

0 0.02 0.04 0.06 0.08 0.1

Termination Probability

ability (abscissa) and free energy of hybrid stability (-kcal/ FIG. 3. Computer-derived comparison of termination prob-

mol) in a putative 7-base pair template-primer stem. All residues in the M13mp2 template evaluated (-84 to 171) are shown in the upper panel, and residues corresponding to termination probability of 10% or less are shown in the lower panel. Free energy values were computed under the conditions of the DNA polymerase incubation using the method described by Rychlik and Rhoads (1989). Nucleotide residues noted by circles (0) correspond to sites preceding stem-loop secondary structures predicted in the M13mp2 template, also as described by Rychlik and Rhoads (1989).

and 121, yet there are no nearby changes in sequence. The closest positions at which the sequence is altered are 137 and 107. Fig. 2f indicates changes in the region of residues 130- 144. Analysis in this region is made difficult by the existence of a compression in sequencing gel lanes at residues 139-141, such that these positions cannot be distinguished. Nonethe- less, a sequence change at position 137 has some influence on termination at several positions. We note that the termination pattern changes slightly between pwtl and pwt2, although the sequence in this region is identical for the two templates.

Comparison between the Global Analysis (Table II) and Results with Mutated Templates-Several general trends on nucleotide sequence influence over termination sites were surmised from the information in Table 11. In summary, when T is in the template position +I, termination is rare; this trend is independent of the sequence context, since in only one of the 45 strong termination sites was the +1 nucleotide a T residue. Second, in the template strand when A is in the -2 or -6 position, there is an increased chance for termina-


TABLE V Analysis of template-primer stem position effects for changes of C to A and G to A

Altered residue Change from Changed residue at Change in termination probability

+ to relative position + ,, -1 -2 -3 -4 -5 -6 -7

-37 (wt us. pwt2) -30 (wt us. pwt2)

69 (pwtl us. pwt2) No. of changes per position

-37

69

92

(pwtl us. pwt2)

(wt us. pwtl)

(pwt2 us. wt) No. of changes per position

Overall no. of changes per position

C - A Residue

C - A Residue

C - A Residue

Result

Result

Result

G - A Residue

G - A Residue

G - A Residue

Result

Result

Result

-37"

-30 N 69 N

0

-37"

69 N 92 N 0

0

-38 tt t

-31

68 N

2

-38 N 68

N 91

1

3

t

-39 N

-32 tt 67

2

-39

67

90

3

5

tT

t t tt

-40

-33

66

3

-40 N 66

89

2

5

t tt t

t t

-41 N

-34

65 N

1

-41 N 65

N 88

N 0

1

f

-42

-35

64

3

-42

64 N 87

N 1

4

tt

tt t

tt

-43

-36 tt t 63

N 2

-43 N 63

N 86

1

3

tt

-44 N

-37

62 N

1

-44

62 N 85

2

3

t

44

t

a Sites at which a second nearby change complicates the interpretation. N = 52.49-fold change; t = >2.49-5-fold change; Tt = >5-fold change. Notice that arrows go up in all cases but one. Thus, changes to template-strand A in the primer stem usually increase termination probability. G + A and C + A changes at residues 102 and -41 were eliminated because of another nearby change confounding the interpretation.

tion. Conversely, G in the template strand at -6 is associated with reduced termination, and C in the template strand, especially at the -2 position, is associated with reduced termination. With these general trends in mind, it is interesting to consider the termination values for the two mutated templates. Several of the single-base changes are associated with increases in termination probability of 10-24-fold (Table IV). These changes occur in template regions that are neither predicted to have nor known to have long range secondary structures (Fig. 3 and see below). All but one of these increases correspond to a change to A in the template moiety of the template-primer stem and are in positions -2, -3, or -6.

In Fig. 2, termination across all regions with single-nucleotide changes for the two mutated templates are compared with the wild type template. Consider first the trend for the incoming nucleotide/+l template position. At residue 70, termination is high when 69 is a G, corresponding to incoming dCTP, and is 4-fold lower when 69 is a C. This trend is the same as that noted in Table 11. Similarly, termination at -36 is higher when residue -37 is changed from C to G or A, corresponding to the incoming dCTP and TTP. The antiter- mination effect of C in the +1 position is not absolute, however, as seen for residue 102, where termination at 103 is higher for C a t +1 than for G or A at +1 (Fig. 2d).

Next, the effect of nucleotide residue and position in the template-primer stem was evaluated for six of the changes to A (Table V). Consider, for example, termination differences corresponding to the G to A substitution a t residue 92 pwt2 versus wt, (Fig. 2c). With wt template, strong termination sites are observed for positions 89, 90, and 91. With pwt2, termination was reduced modestly when the G substitution represented the 0, -4, and -5 positions but had a stronger effect when the substitution represented positions -1, -2, -3, -6, and -7; when the base change represented positions -2 and -6, termination was reduced 11- and 22-fold, respectively (Tables IV and V). This picture of A in the -2 and -6 positions being associated with more termination than G is similar to the trend noted in Table IIA.

From Table IIA, one might predict a strong difference in termination when a C to A change occurs at position -2, but not at -4. This prediction was only partially borne out with respect to position effect (Table V). Although the C to A

change produced increases in termination when the mutation was in the template-primer stem, termination increased when the substitution represented the -4 position in at least one case, and strong increases in termination were seen with substitutions at positions other than -2. Consider, for example, the C to A change at residue -30. Increased termination was seen with all positions, including very strong increases when the substitution represented the -3 and -5 positions. These latter two increases could not be predicted from the tabulation in Table IIA. Nevertheless, there does appear to be a position effect in the template-primer stem for the mutated residue. Summation of the number of termination changes per position represented by base alteration reveals a bimodal distribution on either side of the -4 position (Table V).

From Table IIA, a G to A change at the -6 position can be predicted to cause a strong increase in termination. This prediction was borne out for the G to A change at residue 92, but not for the G to A changes at residues -37 and 69. Results in Table IIA also predict an increase in termination with the G to A change at the -2 position. This prediction was borne out with each of the G to A changes; however, increases also were seen when the change represented position -3, and this could not have been predicted from Table IIA. In the case of the -37 change, the G to A mutation at the -7 position produced a decrease in termination. For both types of mutation, C to A and G to A, the position effect for the template- primer stem shows a bimodal distribution with peaks when the base substitution represents residues -1 to -3, and -5 to -7.

Note that in three cases a change to A (residues 102, 98, and 137) did not result in strong changes in termination probability, implying that factors other than the mere pres- ence of A in the template-primer stem are important also. The changes at 102,98, and 137 correspond to C-rich regions for the template-primer stem, and it is conceivable that the terminator effect of A is squelched by a reduced termination effect of a C-rich region.

Ten C substitutions were created. The global trend from Table I1 predicts little change in termination probability for these alterations. The residue 69 G + C change, however, produced a moderate reduction in termination when the sub-


stitution represented the -1, -2, and -5 positions. Thus, C appears to be a weaker terminator than G in the residue 69 sequence context and in the residue -37 sequence context. The G -+ C change at 102 has very little effect, however, in the G- and C-rich context around residue 102.

Analysis of Template-Primer Stability-The possibility of correlation between template-primer stem hybrid stability and termination probability across the entire template was evaluated (Fig. 3). The computer-derived free energy of an- nealing for each consecutive 7-residue hybrid across the template corresponding to positions -1 through -7 was obtained by the computer program described by Rychlik and Rhoads (1989). Termination probability values for the wt template were taken from Table I. This analysis reveals no correlation between increased or decreased template-primer stem hybrid free energy and termination probability. Also, additional analysis failed to reveal a correlation between template secondary structure potential and termination probability (Fig. 3). The analysis of potential stem-loop structures was also conducted by the computer program described by Rychlik and Rhoads (1989).

DISCUSSION

The tendency of DNA-polymerizing enzymes to terminate synthesis preferentially at specific positions on natural DNA templates has been recognized for some time (Weaver and DePamphilis, 1982; LaDuca et al., 1983; Fairfield et al., 1983). Other than putative secondary structures, factors that induce termination are not well understood. Our results, here and elsewhere (Abbotts et al., 1988; Bebenek et al., 1989), suggest that sequence contexts independent of secondary structure can also influence termination. Table 11, tabulating sequences around the strongest and weakest termination sites for the HIV-1 reverse transcriptase, and Table 111, tabulating sequences around the strong termination sites for AMV reverse transcriptase, indicate that sequence composition can influence termination.

We have shown previously with the E. coli Pol I large fragment that the incoming nucleotide has a strong influence on termination probability during synthesis on a DNA template. When a single dNTP was reduced in concentration by an order of magnitude or replaced by the corresponding dNTPaS, new strong termination sites appeared at the position where the altered dNTP was the incoming nucleotide (Abbotts et al., 1988). This phenomenon is extended with the HIV-1 reverse transcriptase, by Table 11. The distributions at the +1 position indicate an influence on termination of the incoming dNTP/template residue combination; however, one cannot rule out the sole influence of either component, the +1 template position or dNTP, on termination independent of the other component.

Our data indicate that the sequence of the template-primer stem region influences dissociation at the primer terminus. Sequence distributions of terminator sites are decidedly nonrandom at positions -2 and -6 for the HIV-1 reverse transcriptase (Table 11) and at several template-primer stem positions for the AMV reverse transcriptase (Table 111). It should be noted that the trends are not absolute in that for the HIV-1 reverse transcriptase all nucleotides were scored at each position a t least once; for the AMV reverse transcriptase, at least three nucleotides were scored a t least once at each position. The quantitative relationship between sequence and termination, therefore, appears to be complex. Nonetheless, the data indicate that sequence can influence termination, and they imply that the sequence in the template-primer stem region can influence termination for a greater distance than sequence in the single-stranded template region.

The ability of sequence to influence termination is further evaluated by experiments with the mutated templates (Fig. 2). The trends deduced from the tabulation of the strongest and weakest (Table 11) sites were only generally corroborated by the results with the mutated templates. The precise position effect for template at -2 and -6 was not corroborated. Several of the changes in termination illustrated in Fig. 2 (positions -36,69, and 103) can be attributed to the effect of the incoming nucleotide/+l template residue combination. The results in Fig. 2 also support the idea that changes in sequence a t one position can influence termination at other positions; in some cases, the changes in Fig. 2 are more dramatic than could be predicted from the data in Table 11. For example, in Fig. 2a, only two positions, -30 and -37, are changed in sequence, but termination changes are seen across positions -31 to -40; the absence of other local changes in termination suggests that enzyme-template-primer binding is the most straightforward explanation for the observed changes.

Another dramatic change is seen in Fig. 2c. The sequence at positions 88-97 (GGGAAAACCC) represents an obvious potential stem-loop structure, and we have observed that several polymerases show strong termination or pause sites in this vicinity (Kumar et al., 1990; Abbotts and Wilson, 1991). With HIV-1 reverse transcriptase, however, the strongest termination sites occur a t positions 89, 90, and 91; con- cerning the possible influence of template secondary structure, the enzyme will not reach these sites until after it has opened the putative stem starting at residue 97. The sequence change in template pwtl, an A at residue 98, would increase the stem by two nucleotides; AG at 98-99 can base pair with C T a t 86-87. Nonetheless, this change has little effect on termination in the region. On the other hand, the change from A to G at residue 92, in the putative loop, is accompanied by dramatic reductions in termination up to residue 83. These observations imply that in this region, the template-primer stem sequence is more important for HIV-1 reverse transcriptase termination than the putative template secondary structure. Finally, it should be noted that we did not see obvious secondary structure effects on termination in other regions shown in Fig. 2 (see Fig. 3).

Other groups have investigated the binding site size of DNA polymerases. Fisher and Korn (1981) found with DNA polymerase (Y from KB cells a minimal length of 8 nucleotides for optimal priming on synthetic homopolymer templates. Joyce et al. (1986) found by DNA footprinting that Pol I large fragment, when bound to a template-primer structure of natural DNA, protected about 19 nucleotides on the template strand and 12 nucleotides on the primer. On the basis of experiments with fluorescent DNA probes, or with a DNA photoprobe, Benkovic and colleagues determined that the DNA binding region of Pol I large fragment covers between 5 and 7 base pairs of duplex DNA (Allen et al., 1989; Catalan0 et al., 1990). Nevinsky et al. (1990) investigated four polymerases: Pol I large fragment; replicative and repair-like polymerases of the lower eucaryote Physarum polycephalum; and DNA polymerase (Y from human placenta. For all of these enzymes, the efficiency of a primer on synthetic homopoly- mers was optimal at 9-12 nucleotides. Our results with the reverse transcriptases of HIV and AMV suggest that enzyme- primer contacts occur at least 6 nucleotides upstream from the primer terminus (Tables I1 and 111). Individual experiments suggest contacts further removed from the primer terminus; for example, in Fig. 2c a change in the template- primer stem sequence at residue 92 is accompanied by termination changes up to residue 83.

Kohlstaedt et al. (1992) have recently reported the structure


of HIV-1 reverse transcriptase at 3.5 A resolution. They represented the structural model of the reverse transcriptase heterodimer as a right hand, containing a “palm” or long cleft that can accommodate double-stranded DNA. Other enzyme regions that might contact DNA include “fingers,” “thumb,” and a region with the RNase H catalytic site. In the context of this model and a hypothetical template-primer-enzyme complex, the different enzyme regions extend over the following approximate template positions where the primer 3‘ terminus represents position 0: fingers, from +6 on the template strand to -5 on the template-primer stem; thumb, from 0 to -8; and RNase H region, from -12 to -21. Our results are consistent with an interpretation that enzyme-DNA contacts in the fingers and thumb regions are more important for influencing termination than are contacts in the RNase H region.

It can be noted in Fig. 2a that increasing the A T composition of template pwt2 was accompanied by an increase in termination, and in Fig. 2c, decreasing the AT composition of pwt2 was accompanied by a decrease in termination. In addition, the relatively high frequency at which A or T is found in minus positions, particularly positions -1 through -3 (Table II), suggests that the melting of the primer stem may be important in influencing termination. Huber et al. (1989) reported for the HIV-1 reverse transcriptase a tendency of the enzyme to terminate DNA synthesis at positions within one or two nucleotides after sequences containing at least 3 A and/or T residues in a row. However, in systematic computer-based analysis across the template, we observed no correlation between hybrid stability of the 7-base pair template-primer stem and termination (Fig. 3). Moreover, termination of synthesis cannot be entirely explained by the melting of the template-primer because termination patterns are enzyme-specific, and the positions at which sequences of termination sites show the greatest deviation from random distribution are different for the two reverse transcriptases examined here (Tables I1 and 111). For synthesis on the wild type M13mp2 template, the termination pattern of HIV-1 reverse transcriptase is distinct from that of AMV reverse transcriptase (Bebenek et al., 1989) and from that of Pol I large fragment (Abbotts and Wilson, 1991). We also have noted a difference in the pause site patterns on the M13mp2 template for DNA polymerase /3, a generally distributive enzyme, and its 31-kDa catalytic fragment (Kumar et al., 1990). It thus appears that the termination pattern is characteristic of an individual DNA-synthesizing enzyme rather than a constant feature of the template sequence. The present data suggest that contacts between the enzyme and the template- primer stem influence termination and have a bearing on variations among enzymes in their observed termination patterns.

We cannot rule out the possibility that the sequence changes in Fig. 2 may have long range template structural effects that influence termination. However, significant termination changes occur much more frequently in regions relatively local to sequence changes. For example, the termination pattern is remarkably similar across the three templates between positions -29 to +63 (Fig. 1). We therefore conclude that the major effects of sequence alteration on termination are relatively local and that base changes in the first seven template-primer stem positions can influence ter-

mination at the incoming nucleotide position. With the benefit of hindsight, we can now recognize that

having different sequences around the changes in the templates has complicated some of the conclusions on termination. Thus, future work might be devoted to examining the effects of single-nucleotide changes on termination as a function of sequence composition in the template-primer stem. The present work, nonetheless, is sufficient to evaluate a number of single-base changes in different sequence context (Table V) and to indicate that alterations in incoming nucleotide/+l template residue or the primer stem region have the greatest influence on termination. Clearly a nucleotide sequence change at one position in the template-primer stem can influence termination several nucleotides distant at the primer terminus. Based upon the importance of the template- primer stem, we suggest that the HIV-1 reverse transcriptase can thus be viewed as having characteristics of a sequence- specific double-stranded DNA-binding protein, in that its contacts with the template-primer stem can be influenced by nucleotide sequence. In addition, some of the observations here can be applied to the question of whether termination influences mutation, which is addressed in the accompanying manuscript (Bebenek et al., 1993).

Acknowledgments-We thank Dr. William Beard for assistance with computations of template-primer hybrid stability and for assistance in the analysis in Fig. 3. We thank Dr. Kristin Eckert for assistance with the reproducibility analysis of termination probability values.

REFERENCES Abbotts, J., and Wilson, S. H. (1991) in Advances in Molecular Biology and

Co., New York Targeted Treatment for AIDS (Kumar, A,, ed) pp. 1-19, Plenum Publishing

Abbotts, J., SenGupta, D. N., Zon, G., and Wilson, S. H. (1988) J. Biol. Chem. 2 6 3 , 15094-15103

Abbotts, J., Jaju, M., and Wilson, S. H. (1991) J. Biol. Chem. 266,3937-3943 Allen, D. J., Darke, P. L., and Benkovic, S. J. (1989) Biochemistry 2 8 , 4601-

4607 Bebenek, K., Abbotts, J., Roberts, J. D., Wilson, S. H., and Kunkel, T. A.

Bebenek, K., Abbotts, J., Wilson, S. H., and Kunkel, T. A. (1993) J. Biol. Chem.

Catalano, C. E., Allen, D. J., and Benkovic, S. J. (1990) Biochemistry 29,3612-

1989) J. Bml. Chem. 2 6 4 , 16948-16956

268,10324-10334

2Wl

De::& S. D., and Wilson, S. H. (1982) J. Bid. Chem. 257,9770-9780 di Marzo Veronese, F., Copeland, T., DeVico, A. L., Rahman, R., Oroszlan, S.,

Fairfield, F. R., Newport, J. W., Dolejsi, M. K., and von Hippel, P. H. (1983)

Fisher, P. A., and Korn, D. (1981) Biochemistry 20,4570-4578 Huber, H. E., McCoy, J. M., Seehra, J. S., and Richardson, C. C. (1989) J. Bid.

Chem. 264,4669-4678 Joyce, C. M., OKs, D. L., Rush, J., Steitz, T. A,, Konigsberg, W. H., and

Grindley, N. D. F. (1986) UCLA Symp. Mol. Cell. B i d , New Series 3 9 , 197- 9nr,

Gallo, R. C., and Sarngadharan, M. G. (1986) Science 231,1289-1291

J. Biomol. Struct. & Dyn. 1,715-727

Kedar, P. S., Abbotts, J., Kovacs, T., Lesiak, K., Torrence, P., and Wilson, S.

Kohlstaedt, L. A., Wang, J., Friedman, J. M., Rice, P. A,, and Steitz, T. A.

Kumar, A., Abbotts, J., Karawya, E. M., and Wilson, S. H. (1990) Biochemistry

”-

H. (1990) Biochemistry 29,3603-3611

(1992) Science 2 5 6 , 1783-1790

29. 71%-71.59 - - . - - - . - - - Kungel, T. A. (1985) J. Biol. Chem. 260, 12866-12874 LaDuca, R. J., Fay, P. F., Chuang, C., McHenry, C. S., and Bambara, R. A.

Maxam, A. M., and Gilbert, W. (1980) Methods Enzymol. 66,499-560 Nevinsky, G. A., Veniaminova, A. G., Levina, A. S., Podust, V. N., and Lavrik,

Mendenhall, W., Scheaffer, R. L., and Wackerly, D. D. (1981) Mathematical 0. I. (1990) Biochemistry 29,1200-1207

Roberts, J. D., Bebenek, K., and Kunkel, T. A. (1988) Science 242,1171-1173 Statistics with Applicatio~~~, 3rd ed., pp. 287-288, Duxbury Press, Boston

Rychlik, W., and Rhoads, R. E. (1989) Nucleic Acids Res. 17,8543-8551 Sanger, F., Nicklen, S., and Coulson, A. R. (1977) Proc. Natl. Acod. Scr. U. S. A.

Weaver, D. T., and DePamphilis, M. L. (1982) J. Biol. Chem. 267,2075-2086

(1983) Biochemistry 22,5177-5188

74,5463-5467

the journal of vol. 268, no. 14, issue of may 15, pp. 10312 … · 1999-01-13 · the journal of...

Documents