Reverse transcriptase pausesat N2-methylguanine ... - Europe PMC

1 downloads 0 Views 1MB Size Report
been used to prime reverse transcriptase (avian myeloblastosis virus RNA-directed DNA nucleotidyltransferase; deoxynu- cleosidetriphosphate:DNA ...
Proc. Natil. Acad. Sci. USA Vol. 76, No. 8, pp. 3751-3754, August 1979

Biochemistry

Reverse transcriptase pauses at N2-methylguanine during in vitro transcription of Escherichia coli 16S ribosomal RNA (avian myeloblastosis virus/RNA-directed DNA nucleotidyltransferase/primer/base methylation/kinetics)

DOUGLAS C. YOUVAN* AND JOHN E. HEARST*t *Group in Biophysics and Medical Physics and tDepartment of Chemistry and Laboratory of Chemical Biodynamics, University of California, Berkeley, California 94720

Communicated by Melvin Calvin, May 14,1979

ABSTRACT A restriction fragment strand complementary to a sequence near the 3' end of Escherichia coli 16S rRNA has been used to prime reverse transcriptase (avian myeloblastosis virus RNA-directed DNA nucleotidyltransferase; deoxynucleosidetriphosphate:DNA deoxynucleotidyltransferase, EC 2.7.7.7). In addition to transcripts that were extended to the 5' end of the RNA, two major transcription intermediates were observed. These discrete-sized cDNA intermediates are the result of a kinetic barrier imposed by monomethylation of the amino group on guanine that participates in base-pairing. Both major transcription intermediates correspond to attenuation at the known positions of N2-methylguanine (m2G) in the rRNA sequence. The relaxation time for elongation of the cDNA through m2G is approximately 3 min. No other major kinetic pauses were observed in the 1340 bases transcribed. Synchronous and uniquely primed synthesis of cDNA on an RNA or DNA template by various polymerases can be a useful technique in the elucidation of the modulators of such enzymes.

Recently described rapid sequence-determination procedures utilize polymerases and chain-terminating nucleoside triphosphate analogs- (1, 2). We are interested in coupling these techniques in order to study the kinetics and fidelity of transcription on templates that contain various natural or chemical modifications and secondary structures. Our approach has been to investigate the time course of primed transcription on a large template at low resolution and then to reexamine the major attenuators of transcription at sequence resolution. We have observed two strong kinetic pauses during the reverse transcription of Escherichia coli 16S rRNA primed at sequence position 1341.0 At the resolution of.agarose gel electrophoresis of the cDNA products, both kinetic pauses correlate with the known positions (3) of N2-methylguanine (m2G) in the RNA template. Methylation of a group involved in base-pair hydrogen bonding is a probable attenuator of the rate of transcription. However, the possible coincidence of the positions of m2G with strong secondary structures necessitated investigating the kinetic pauses at higher resolution. Both kinetic barriers are highly localized and at sequence resolution the first pause occurs at the known position of m2G in the RNA template. Most of the cDNA elongates through the m2G at sequence position 1206 with a relaxation time of approximately 3 min.

stipulated by the National Institutes of Health guidelines. The plasmid was digested with Hinfi restriction endonuclease (Bethesda Research Laboratories, Rockville, MD) and end labeling was accomplished by elongation of the 3' ends of the total restriction digest (10 Aug of DNA) with 50 units of avian myeloblastosis virus reverse transcriptase (RNA-directed DNA nucleotidyltransferase; deoxynucleosidetriphosphate:DNA deoxynucleotidyltransferase, EC 2.7.7.7) in the presence of 100 ,uCi (1 Ci 3.7 X 1010 becquerels) of deoxyriboadenosine 5'[a-32P]triphosphate (Amersham, 400 Ci/mmol) for 1 hr at 370C in RT buffer (50 mM Tris, pH 8.3/6 mM MgCl2/40 mM KCI). AMV reverse transcriptase was provided by Joseph Beard (Life Sciences Inc., St. Petersburg, FL) through the auspices of the National Cancer Institute. The 137-base HinfI fragment can be simultaneously purified from the restriction digest and strand-separated on an 8% polyacrylamide gel by a described strand-separation procedure (6). The slow-migrating strand is the sense (primer) strand homologous to the rRNA 1341-1477 bases from the 5' end. Restriction endonuclease HindIII-digested simian virus 40 markers were labeled with reverse transcriptase similarly to the above procedure. Typically, 105-106 cpm (Cerenkov radiation) of 32p was incorporated in 1 Ag of restriction fragments. Reverse Transcription of 16S rRNA. E. coli MRE 600 rRNA, a gift from Pallaiah Thammana, was prepared as described (7). Reverse transcriptase used for the synthesis of large cDNA was rechromatographed by gel filtration on Bio-Gel P-100 in the buffer in which the enzyme was supplied [E buffer, 50% (vol/vol) glycerol/0.2 M KPO4, pH 7.2/2 mM dithiothreitol/ 0.2% Triton X-100] except that the column was run in 10% glycerol. The protein that chromatographed near the exclusion volume was concentrated by dialysis against E buffer. The end-labeled primer (2 X 103 cpm) was annealed to excess 16S rRNA template (0.5 Ag per time point) for 5 min at 65"C in RT buffer. The reaction volume for each time point was 20 =

MAl with each deoxyribonucleoside triphosphate (dNTP) present at 200MM. Each reaction mixture was equilibrated at 370C for 2 min prior to addition of 25 units of AMV reverse transcriptase. Transcription was stopped at each point by addition of 7 Ml of 0.5% sodium dodecyl sulfate/10 mM EDTA, pH 9.0/50%

MATERIALS AND METHODS 32P-Labeled Primer. The bacterial clone containing pER18 DNA, recombinant for a portion of E. coli rDNA from the rmB cistron, was grown (4), lysed (5), and banded in an ethidium bromide/cesium chloride density gradient. Recombinant DNA experiments were carried out under P1,EK1 conditions as

glycerol/0.25% xylene cyanol/0.15% bromophenol blue and quenched on ice. Samples were electrophoresed on a 1.4% alkaline agarose slab gel containing 30 mM NaOH (with 2 mM EDTA in the gel and running buffer) for 5 hr at 4 V/cm. Data from the gel were quantitated by densitometry of the autoradiogram.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Abbreviations: m2G, N2-methylguanine; dNTP, deoxyribonucleoside triphosphate; ddNTP, 2',3'-dideoxyribonucleoside triphosphate. t Sequence positions will be given in reference to the RNA template; the 5'-end base corresponds to sequence position 1. 3751

3752

Dideoxy Sequencing. Ten nanograms of unlabeled 137-base HinfI primer was annealed to excess 16S rRNA and reverse transcribed in the presence of 2',3'-dideoxyribonucleoside triphosphates (ddNTPs) (P-L Biochemicals) which are chainterminators (1, 2). The dideoxy sequencing method modified for reverse transcriptase will be published elsewhere (R. Swanstrom, personal communication). Briefly, labeling is accomplished by a 20-min pulse with a 32P-labeled dNTP (10 ,4Ci) other than the particular ddNTP generating the sequence and then the dNTP is chased for 20 min. The reaction buffer is the same as above but the dNTP concentrations are decreased to 100 tiM for each dNTP other than the labeled dNTP or ddNTP. The ddNTP-to-dNTP (25 AM) ratios are as follows: 1:2.5 A, 1:5 C, 1:2.5 G, and 1:2.5 T. Natural pauses were studied in relation to the sequence by synthesizing cDNA in the absence of ddNTPs. cDNA from the reactions was electrophoresed on a 40 cm 8% polyacrylamide/7 M urea slab gel run in 50 mM Tris-borate, pH 8.6/1 mM EDTA for 8 hr at 1250 V.

RESULTS Transcription Intermediates. Primed synthesis of cDNA on 16S rRNA as a function of time is shown in Fig. 1. Clearly, the transcription process was not uniformly processive-i.e., discrete partial cDNA intermediates exist. After 5 min of transcription, the 137-base primer had elongated up to the first major kinetic barrier and some cDNA had progressed to the second kinetic barrier. After 10 min of reaction, cDNA that corresponded in length to transcription to the end of the RNA began to accumulate. The amount of cDNA that had paused at the first major kinetic barrier decreased after 5 min of reaction. Minor transcription intermediates also were visible in this gel.

Our first aim was to correlate the two major kinetic pauses with some known feature of the RNA: sequence, secondary structure, or methylation. We found that both major cDNA intermediates were coincident in length to transcripts terminating at the m2Gs that have been mapped 965 and 1206 bases from the 5' end of the molecule (3). The resolution of the alkaline agarose gel and calibration by markers were not sufficiently accurate to prove this point conclusively, so we studied the first kinetic barrier at sequence resolution. The two major kinetic intermediates were identified in reference to the RNA sequence (Fig. 2). The sequencing ladder was generated by the incorporation of ddNTPs during primed reverse transcription from the 137-base Hinfl primer. The 3' end of the primer is homologous to sequence position 1341 and the sequence can be read up to the first major kinetic pause. The known sequence (3) near the m2 G at 1206 is: 5'-C-A-U-G-m2G-C-C-C-U-U-3'. Our interpretation of the alignment of the first major kinetic

_

.,...

_S

I.

S 2..

.f

M

5

10

20

_*.|_2;wk1

Proc. Natl. Acad. Sci. USA 76 (1979)

Biochemistry: Youvan and Hearst

40

A-.w^

I

1200

:t

1220 1230

.:.

U

zl.-,.

2

s. z.

"

.1.

-U..

I,

:.

-iNt4

.s::,.

1240

a

1250 1260 1270

*

1 280 1290

_

V .

4

.mt

_

w * f a

1210

s i_

_

U R l._

B F

J

w w

...

_ s. E # ;

W

_K

_

Sg

S,

.eR5

.;A i1

....

..

;.

_t

_

1300

.-

w

w

..ij"W.B

1310 1320

1330

~

~ ~ ~

FIG. 1. Autoradiogram of an alkaline agarose gel showing the elongation of the 137-base restriction strand primer (labeled P) by reverse transcriptase. The transcription time in minutes is shown above each lane. Lane M contains end-labeled HindIII-digested simian virus 40 markers (216, 448, 527, 1102, 1170, and 1769 bases, bottom to top). Bands labeled 1 and 2 are the first and second major kinetic barriers. Band E is coincident in length to transcripts extending to the 5' end of the RNA.

X A C G U Y FIG. 2. Autoradiogram of a dideoxy sequencing gel in which reverse transcriptase has been used to elongate the 137-base restriction strand primer. Lanes labeled A, C, G, and U refer to the rRNA sequence and result from reactions containing the complementary ddNTP. Lanes X and Y are cDNA products transcribed in the absence of ddNTPs; bands 1 and 2 represent natural pauses. Lanes A, C, G, U, and X were pulse labeled for 20 min and chased for 20 min whereas lane Y was labeled for 5 min and chased for only 5 min. Dots to the left of bands mark every 10 bases; numbering refers to the rRNA sequence position starting from the 5' end base. The entire sequence of 16S rRNA has been published by Brosius et al. (3).

barrier with this sequence is that the pause corresponds to the attenuation of transcription at the position of the m2 G at 1206. The resolution in the sequencing ladder is not sufficiently good to show whether or not the m2G directs the incorporation of cytosine into the cDNA. A second strong natural pause is observed in this gel beyond sequence resolution but consistent with the position of the m2G at 965. Kinetic Analysis. The overall kinetics of reverse transcription on 16S rRNA can be simplified to the following scheme: k1

k2

P + T -*PT -*1206

k3 -*

k4

k5

k6

1205 -*965 -*964 -- 1.

P and T represent the primer and template; the major pauses and the following bases are represented by their sequence position from the 5' end of the RNA (base 1). Because the primer is end labeled, the radioactivity in each gel band (Fig. 1) is directly proportional to each species. We set the total area of the densitometer tracing of each gel lane to 1.0, so the concentration of each component can be represented by a dimensionless fraction. The rate constant ki is the second-order rate constant for the hybridization of primer to template. Because the template is in vast excess over primer under the conditions of the experiment, a semilogarithmic plot of nonelongated primer concentration ([P]) versus time (Fig. 3) reveals a pseudo-first-order rate constant kl[T] = 0.024 min-1. The plot also reveals that, due to the preannealing, 85% of the primers are hybridized to templates at zero time and elongate essentially instantaneously upon addition of reverse transcriptase. Intrinsic to this analysis is the assumption that the elongation rate designated k2 is very fast relative to kl[T]. The rather slow hybridization or elongation of the remaining 15% of the primer does not significantly modify the following kinetic analysis because the relaxation times of subsequent processes are considerably shorter than the time associated with priming

-1 = 42 min. T1= kki[T] Fig. 4 is a plot of the fractional concentration of cDNA pausing at sequence position 1206, designated [1206], as a function of time. Experimentally, some trailing of [1206] is observed with time in excess of concentrations predicted for

, 0.15V

aU

0

°0.100 c

a,

C

_\

0

0.050

3753

Proc. Natl. Acad. Sc. USA 76 (1979)

Youvan and Hearst

Biochemistry:

40 20 Transcription time, min

5 10

FIG. 3. Semilogarithmic plot of nonelongated primer concentration (dimensionless fraction) plotted against transcription time. Data were obtained from densitometry of the autoradiogram shown in Fig. 1. See text for the kinetic analysis.

1.0a -W

0.9-

~0 O

0.8-

C 0

0.7-

*r, a

Co

c ° 0.6.

_

C-

..- c 4)

0.5o.2 .C

.2 X

0

0.4-

+-,X U

C

0.3-

'r

0.2

0 u a,

-

0.1 0.1-

1 -

.

l0

20

40

Transcription time, min

FIG. 4. Plot of the concentration of cDNA pausing at the m2G 1206 kinetic barrier as a function of transcription time. The curve drawn is generated by Eq. 1 (see text) with k3 = 0.35 min-1 and f = 0.06.

first-order kinetics. Some fraction, designated f, of cDNA pausing at m2G 1206 does not proceed further with an appreciable rate. We have modified a first-order kinetic analysis to include the variable f in addition to the first-order rate constant k3: [1206] = (1 -f) [1206]o e -k3t + f[1206]o. [1] The best fit of this equation to the data yields [1206]o = 0.85, f = 0.06, and k3 = 0.35 min-'. This means that 94% of the cDNA that initiated from hybridized primer proceeded through. the m2G at 1206 with a relaxation time of approximately 3 min. The caus, of 6% of the total cDNA stopping at m2G 1206 has not been investigated. This could involve RNase H nicking of the template, which may occur when the enzyme is in a nonprocessive mode (8). DISCUSSION Partial cDNA transcripts have been observed during the in vitro reverse transcription of messenger and viral RNAs (9-11). Discrete transcription intermediates are synthesized by the endogenous transcription of disrupted Rous sarcoma virus virions (12, 13). Presumably, some partial transcripts are kinetic intermediates that accumulate transiently during the elongation of cDNA due to kinetic modulators in the RNA template. Possible template attenuators of the rate of transcription include signals in the primary sequence, base or ribose methylations, and RNA secondary structure. Recently, the sequence of an E. coli 16S rRNA gene has been determined and the positions of various methylations were established by correlation with rRNA oligonucleotide data (3). In addition, a secondary structure has been proposed for 16S rRNA consisting of long-range loop interactions (14). These data and the availability of cloned DNA as a source of primers makes 16S rRNA an ideal template for the study of kinetic modulators of reverse transcriptase. In this communication we have shown that m2G is the principal kinetic barrier for reverse transcription in the 1340

3754

Biochemistry: Youvan and Hearst

bases proximal to the 5' end of E. coli 16S rRNA. Previously, m6A, which is dimethylated on the base-pairing amino group, was shown to be an efficient terminator of reverse transcription near the 3' end of rRNA (15). It has also been shown that methylation of the amino group in adenine and guanosine results in destabilization of base pairing (16-18). The effect is observed at the monomer level for m6A by high-resolution proton magnetic resonance of base analogs due to restricted rotation of the amino methyl group about the nitrogen-ring carbon bond and a syn orientation. This is not observed for m2G, for which destabilization results only at the polymer level as demonstrated by thermal denaturation and circular dichroism analysis of methylated polyguanylic acid complexes with poly(C). It is notable that Rous sarcoma virus RNA contains several monomethylated adenines, m6A, the function of which are unknown (19). These may prove to be natural kinetic modulators of reverse transcriptase. Independent of possible kinetic barriers due to base methylation, the principal attenuator of the rate of T4 DNA polymerase synthesis on the bacteriophage fd single-stranded DNA template is secondary structure (C. Huang, personal communication). A similar attenuation of reverse transcriptase by secondary structure is not observed for 16S rRNA. This may be due to a basic difference in abilities of the two polymerases to polymerize through helical regions in the template or to a less-stable and possible fluctuating secondary structure in 16S rRNA. We thank Craig Squires for restriction fragments and preliminary sequence data and Harry Noller for providing us with the pER18 clone. Ron Swanstrom's advice and dideoxy sequencing recipes are especially appreciated. This work was supported in part by funds from the National Institutes of Health (Grant T32-ES07075), from the American Cancer Society (Grant NP185), and from the Division of Environmental Research and Development, U.S. Department of Energy.

Proc. Natl. Acad. Sci. USA 76 (1979) 1. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl. Acad. Sci. USA 74,5463-5467. 2. Hamlyn, P. H., Brownlee, G. G., Cheng, C. C., Gait, M. J. & Milstein, C. (1978) Cell 15, 1067-1075. 3. Brosius, J., Palmer, M. L., Kennedy, P. J. & Noller, H. F. (1978) Proc. Natl. Acad. Sci. USA 75,4801-4805. 4. Palmer, M. L., Raker, M. A., Kennedy, P. J., Young, J. W., Barnes, W. M., Rodriquez, R. F. & Noller, H. F. (1979) Mol. Gen. Genet., 172, 171-178. 5. Zasloff, M., Ginder, G. D. & Felsenfeld, G. (1978) Nucleic Acids Res. 5, 1139-1152. 6. Maxam, A. M. & Gilbert, W. (1977) Proc. Nati. Acad. Sci. USA 74,560-564. 7. Thammana, P. & Cantor, C. R. (1978) Nucleic Acids Res. 5, 805-823. 8. Myers, J. C. & Spiegelman, S. (1978) Proc. Natl. Acad. Sci. USA 75,5329-5333. 9. Efstratiadis, A., Maniatis, T., Kafatos, F. C., Jeffrey, A. & Vournakis, J. N. (1975) Cell 4,367-378. 10. Frolova, L. Y., Metelyev, V. G., Ratmanova, K. I., Smirnov, V. D., Shabarova, Z. A., Prokofyev, M. A., Berzin, V. M., Jansone, I. V., Gren, E. J. & Kisselev, L. L. (1977) Nucleic Acids Res. 4, 2145-2159. 11. Devos, R., Van Emmelo, J., Celen, P., Gillis, E. & Fiers, W. (1977) Eur. J. Biochem. 79,419-432. 12. Darlix, J.-L., Bromley, P. A. & Spahr, P.-F. (1977) J. Virol. 23, 659-668. 13. Lai, M. M. C. & Hu, S. S. F. (1978) Nature (London) 271, 481-483. 14. Wollenzien, P., Hearst, J., Thammana, P. & Cantor, C. R. (1979) J. Mol. Biol., in press. 15. Hagenbuchle, O., Santer, M., Steitz, J. A. & Mans, R. J. (1978) Cell 13,551-563. 16. Ikehara, M. & Hattori, M. (1971) Biochemistry 10, 35853590. 17. von Hippel, P. H. & Engel, J. D. (1974) Biochemistry 13, 4143-4158. 18. Ginell, S. L. & Parthasarathy, R. (1978) Biochem. Biophys. Res. Commun. 84, 886-894. 19. Dimock, K. & Stoltzfus, C. M. (1977) Biochemistry 16, 471478.