913

Nucleic Acids Research, Vol 18, No. 4

Construction of recombinant DNA molecules by the use of a single stranded DNA generated by the polymerase chain reaction: its application to chimeric hepatitis A virus/ poliovirus subgenomic cDNA Czeslaw Wychowski 14 *, Suzanne U.Emerson1, Jonathan Silver2 and Stephen M.Feinstone3 laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, NIH, Bethesda, MD 20892, laboratory of Molecular Biology, National Institute of Allergy and Infectious Diseases, NIH, Bethesda MD 20892, laboratory of Hepatitis Research, Division of Virology, CBER/FDA, Bethesda, MD 20892, USA and 4Unrte de Virologie Moleculaire, UA CNRS 545, Instrtut Pasteur, Pans, France Received October 16, 1989, Revised and Accepted January 12, 1990

ABSTRACT In order to study the importance of VP4 in picornavirus replication and translation, we replaced the hepatitis A virus (HAV) VP4 with the poliovirus (PV1) VP4. Using a modification of oligonucleotide site directed mutagenesls and the polymerase chain reaction (PCR), we created a subgenomic cDNA chimera of hepatitis A virus In which the precise sequences coding for HAV VP4 capsld protein were replaced by the sequences coding for the poliovirus VP4 capsid protein. The method involved the use of PCR primers corresponding to the 3' and 5' ends of the poliovirus VP4 sequence and that had HAV VP4 3' and 5' flanking sequences on their 5'ends. Single stranded DNA of 240 and 242 nt containing the 204 nt coding for the complete poliovirus VP4 were produced by using a limiting amount of one of the primers in a PCR reaction. These single stranded PCR products were used like mutagenic oligonucleotides on a single stranded phagemid containing the first 2070 bases of the HAV genome. Using this technique, we precisely replaced the HAV VP4 gene by the poliovirus VP4 gene as determined by DNA sequencing. The cDNA was transcribed into RNA and translated In vitro. The resulting protein could be precipitated by antibody to poliovirus VP4 but not to HAV VP4. INTRODUCTION Hepatitis A virus (HAV) is assigned to the picornavirus family (1,2,3), although it has several features that are unique among these viruses (for review 4). HAV differs significantly from the other picornaviruses in its growth characteristics and the stability of its capsid (4,5,6). One striking difference is that the length predicted by sequence analysis of the VP4 capsid protein of HAV * To whom correspondence should be addressed

is only 23 amino acids (7) while the other members of this viral family have VP4 peptides about three times that long (for review 8). Since the VP4 protein of the prototype picornavirus, poliovirus, appears to be important for viral morphogenesis and uncoating (9), the question arises whether the truncated VP4 protein of HAV is related to the unique growth and stability properties of the virus. In order to answer this question, we have begun to construct chimeric viral mutants. In the course of this work we developed a new and versatile method for exchanging DNA sequences in a precise and efficient manner. First, we wished to replace the short (69 bases) VP4 gene of HAV with the much longer one (204 bases) of poliovirus. In vitro oligonucleotide site-directed mutagenesis is generally used to introduce base substitutions, deletions, or insertions into the DNA of interest (10). This method usually involves hybridization of a chemically synthesized oligonucleotide containing the desired mutation to the DNA followed by extension of the oligonucleotide on the original template DNA (11,12,13,14). However large substitutions such as the one we wished to make are restricted by the cost and the technical limitations involved in the synthesis of long oligonucleotides. Therefore, we produced a single stranded DNA using the polymerase chain reaction (PCR) (15) and devised a method to use it much like a mutagenic oligonucleotide to introduce a large substitution in a cDNA clone. This technique can be used to introduce relatively large mutations at precise sites by substitution or insertion of nucleotide sequences from another source without regard to restriction enzyme sites or any other sequence considerations. The technique would also be useful as a general method for producing recombinant molecules without utilizing specific cloning sites. In general the technique combines site directed mutagenesis of single stranded DNA with PCR. The 3' ends of the PCR primers are constructed to amplify exactly the region of the donor

914 Nucleic Acids Research molecule to be substituted or inserted. The 5'ends of both primers consist of sequences that direct the single stranded PCR product precisely to the location on the single stranded target molecule where the substitution or insertion is to be made (16). A single stranded PCR product is synthezised by holding one primer at a limiting concentration as described (15).

HB101 bacteria (21) were transformed with the ligated mixtures and resulting clones were screened by colony hybridization (22) using [32P]-labelled DNA coding for poliovirus VP4 (23). Hybridization was at 42°C and washing at 68°C in the presence of 50% formamide (17).

MATERIALS AND METHODS Enzymatic Amplification of DNA and Mutagenesis System Oligodeoxynucleotide primers specific for Hepatitis A virus and for poliovirus type 1 sequences were synthesized on an ABI 380 A DNA Synthesizer (Applied Biosystems, Foster City, CA) by die methoxyphosphoramidite method (Table 1). Production and amplification of single stranded DNAs were performed according to Gyllensten and Erlich (15). Briefly, donor sequences representing the PV1 VP4 gene were amplified in a 100 ul reaction volume containing 0.1 /tg of poliovirus cDNA, 2.5 units of Taq polymerase (Perkin Elmer Cetus), 200 uM each dNTP, 50 pmol primers CT1 or CT2, 1 pmol primer CT3, 50 mM KC1, 10 mM Tns-HCl pH 8.3, 1.5 mM MgCl2 and 0.01% (wt/vol) gelatin. The reaction was performed for 35 cycles in a programmable DNA Thermal Cycler (Perkin-Elmer Cetus) as follows: 94°C for 1.0 min, 50°C for 1.5 min and 1 min at 72°C. The DNA from the reaction mixture was then electrophoresed through 3% Nu Sieve GTG agarose (FMC Byproducts, Rockland, ME) gel in TAE IX (17) and the single stranded DNA was recovered from die gel according to Wieslander (18). The mutagenesis was performed by using phagemid vector pTZ HAV Hindm ss DNA (see the text) and the technique of Taylor

DNA Sequencing Sequencing was performed directly on mini-preparations of plasmid DNA according to the method of Zagursky et al. (24). The DNA (1 /tg) was alkali denatured and hybridized to an oligonucleotide primer (60 ng), the sequence of which is complementary to that of HAV nucleotides 688-704 (5' CTGAGGTACTCAGGGGC 3') or HAV nucleotides 841-856 (5' CCAGTCACTGCAGTCC 3')(25). Sequencing reactions were carried out at 42 °C in the presence of 15 uCi of 35 S dATP and 14 units of reverse transcriptase using the dideoxynucleotide chain termination method (26).

In Vitro Transcription pTZ HAV Hindm and mutant recombinant DNAs (pTZ/PCRl and pTZ/PCR3) were linearized by digestion with the restriction enzyme BglE or BstEII (Fig. 1) and the resulting linear DNA was extracted sequentially with phenol/chloroform and chloroform and then precipitated with ethanol. RNA transcription was performed on 3 ng of linearized DNA using T7 RNA polymerase (Promega, Biotec) in reaction conditions defined by

BttEII Table 1 : Sequence of Synthetic Oligonucleotide Primers and Positions In the Genomes of Pollovlrus and Hepatitis A Virus

Primers

Sequence

CT1

S1 CATTCTTAAATAATAA'iTGGaTGCTCAGGTTTCATC: HAV 51 NOH-COOtNQ

HAV VP4

724

CT2

OrlE

PV1 VP4 5-END

743 748

781

51 CTTAAATAATAATQAACATGGGTQCTCAGQTTTCATC 3'

I

I

HAV51 NON-CODING

CT3

17 Promoter

"1 1

I

HAV VP4

v

I

PV1 VP4 5-END

ft 10

i

311 GGGGTTACGATTTGAGCTAACTCCTTCTCGTTTAC G 51 PV1 VP4TEND HAV VP2 TEND (complcfTwnt) (comptinMnt)

T A B L E 1: For the amplification the primers are within the sequence coding for the poliovirus type 1 VP4 gene. The locations of the primers were determined from the available sequences of poliovirus type 1. The primers CT1 and CT2 were complementary to the minus strand and trie primer CT3 was complementary to the plus strand. The fragment corresponding to pohovirus and to Hepatitis A is indicated below the primers and the positions in the respective genomes are indicated.

et al. (19). Single stranded HAV Hind m phagemid DNA was produced by standard methods using M13KO7 as a helper phage (20) (Bio Rad Laboratories, Richmond, CA).

FIG. 1: Construction of pTZ HAV Hindm The fragment Hindm-Hindin containing Hepatitis A nucleotides 1 - 2 0 7 0 isolated from plasmid pHAV/7 (34) was inserted into the plasmid pTZ (12) Bacteria JM109 transformed by the resulting plasmid pTZ HAV Hindm which contains the 5'non coding region and the sequences coding for the 23 amino-acids of VP4 HAV, the 222 amino-acids of VP2 HAV and a part of VP3 HAV (200 < amino-acids from 246) was used after infection by M13 KO7 (20) for producing'single stranded DNA pTZ HAV Hindm used as a template for the site directed mutagenesis On, E coti origin of replication Lac Z', complementation fragment of the /3-galactosidase gene fllG, origin of replication and morphogenetK signal from phage fl intergenic region. Abbreviations, AMP, ampKillin resistance gene PL, polylinker 5'NC region, 5' non coding region.

Nucleic Acids Research the manufacturer (40 mM Tris HC1 pH 7.5, 6 mM MgC12, 2 mM spermidine hydrochloride, 10 mM Nacl, 10 mM dithiothreitol, RNasin and 0.5 mM each ATP, CTP, UTP and GTP. The 100 /tl reaction mixture was incubated at 37°C for 2h, DNase I (30 U/ ml) (RQI, Promega, Biotec) was added and incubation was continued at 37°C for 15 mn. The reaction mixture was extracted sequentially with phenol/chloroform and chloroform and then precipitated with ethanol and finally redisolved in diethylpyrocarbonate treated water at a concentration of 1 /tg//tl. In Vitro Translation and Analysis of Chimeric Translated Products Translation in vitro was performed in rabbit reticulocyte lysate (Promega Biotec, Madison, WI) essentially as described by the supplier. The translation mixture contained 1 mCi/ml ([35]S) methionine (TRAN 35 S-LABEL TM, 1179 Ci/mmole, ICN BIOMEDICALS.INC), 20 /tM each amino acid except methionine and RNA at 100 /tg/ml. The proteins were immunoprecipitated from the translational mixture essentially as descnbed (27) and analyzed on 1 % sodium dodecyl sulfate, 12% polyacrylamide gels (SDS- PAGE) that were electrophoresed at 200 constant volts for 45 min in a Biorad Mini Protean II slab cell apparatus at room temperature, followed by drying and autoradiography. RESULTS Subcloning cDNA of HAV in Phagemid pTZ The Hindm-Hindffl restnction fragment from HAV HM-175/7 MK-5 cDNA sequences representing nucleotides 1 to 2070 of the viral RNA which contains the 5'non coding region, the sequence coding for VP4, VP2 and a part of VP3, was subcloned into the Hind m site of the phagemid vector pTZ (12). The pTZ recombinant DNA containing the Hindin fragment is designated pTZ HAV Hindffl (Fig. 1). Single stranded DNA (minus sense) was produced as descnbed above and was used as a template in the site-directed in vitro mutagenesis system.

915

In Vitro Mutagenesis The principle of the mutagenesis method is diagrammed in Fig.2. The primers used to perform the PCR are listed in Table 1. Chimeric clones in which the PV1 VP4 was substituted for the HAV VP4 were produced in the pTZ HAV Hindm as described. Two different constructs were produced because HAV has methionine codons at positions 738 and 744 (25) and it is not yet known which one is used for translation initiation. Primer CT1 was designed to insert the PV1 VP4 at position 738 and the primer CT2 to insert the PV1 VP4 at position 744. Primer CT3 was complementary to the 3'end of PV1 VP4 plus the first codon of PV1 VP2 and codons 2 through 6 of HAV VP2. Primer CT3 was used in the PCR reaction at a limiting concentration in order to produce a positive sense single stranded molecule to be used in the mutagenesis reaction on the negative sense single strand template pTZ HAV Hindm (Fig. 2). The site directed mutagenesis was conducted by annealing template to single stranded DNA. The annealing mix was heated to 80°C for 5 min and was gradually cooled by incubating the reaction tube at 56°C for 20 min and at 37°C for 20 min. The tube was then held in an ice water bath 10 min. Then the second strand synthesis in presence of thiolated dCTP was performed with the large fragment of DNA polymerase 1 (klenow fragment) according to the manufacturer. When the final mutated plasmid was constructed successfully as judged by agarose gel electrophoreseis, 20 /tl of competent E. coli HB101 (BRL, Bethesda, MD) were transformed with 3/tl of the total reaction. One hundred colonies were picked, regrown and fixed on nitrocellulose filters (22), hybridized to the poliovirus VP4 probe labelled according to Feinberg and Vogelstein (23) and washed as described (17). Five percent of the colonies transformed by the PCR1 construct and ten percent of those transformed by the PCR3 construct hybridize with the PV1 VP4 probe. DNA Sequence Analysis Colonies that reacted positively with the PV1 VP4 probe were isolated and the DNA purified. The nucleotide sequences of

FIG. 2: Principle of an In Vitro Mutagenesis System Using Single Stranded DNA Produced by PCR Like a Mutageruc Oligonucleotide. Single stranded D N A was produced by PCR according to the method of Gyllensten and Erlich (15) as detailed in the Materials and Methods. The single stranded DNA was used as an mutagenic oligonucleotide based on the methodology of Taylor and co-workers (19). All details of the different steps of the mutagenesis were as descnbed in the manual of the Amersham Kit.

916 Nucleic Acids Research several of the resulting plasmids were determined. All plasmids (five of each construct) that we sequenced for the junctions at 5' and 3' ends of VP4 contained poliovirus VP4 sequences substituted in the predicted position for the VP4 sequences of HAV. Two plasmids pTZ.PCRl and pTZ.PCR3 were selected for further analysis. The sequences at the 5'end and the 3'end of the insertion are shown in Fig. 3. In the plasmid pTZ.PCRl the sequences coding for the poliovirus VP4 gene were substituted for HAV nucleotides 737 through 807, and in the plasmid pTZ.PCR3 between the nucleotides 743 and 807 (Fig. 3). In all cases the poliovirus VP4 gene was intact and replaced exactly the hepatitis A VP4 gene as intended. Characterization of the Chimeric Gene in an In Vitro Translation Assay The DNA of pTZ HAV Hindlll, pTZ.PCRl and pTZ.PCR3 were cleaved by Bgin or BstEII restriction enzymes and transcribed by the T7 RNA polymerase. The transcripts of these

PTZ.

pTZ. JPCR3 A C G T

linearized plasmids were used in rabbit reticulocyte lysate in vitro translation mixtures. The truncated protein PI (corresponding to the first 357 aminoacids of the capsid proteins of hepatitis A) expressed by the RNA pTZ HAV Hindm linearized with Bglll was smaller than the translation products of either pTZ.PCRl or .PCR3 linearized Bglll as determined by SDS-PAGE (Fig. 4). The apparent molecular weight of the truncated PI chimeric protein expressed by pTZ.PCRl or/PCR3 linearized with Bglll was approximatively 45 Kdaltons compared to the 40K daltons for the protein expressed from the clone pTZ HAV Hindin linearized with Bglll. Lower bands present on the gel may represent internal initiations or degradation products. In order to verify the expression of the poliovirus gene in our constructs, the DNAs of pTZ HAV Hindlll and pTZ.PCRl or PCR3 were linearized by BstEII, transcribed by T7 RNA polymerase and the RNAs translated in vitro. The reaction mixtures were immunoprecipitated as described (27) by the anti-

-ESS!

A C G T

A C G T

^S5

ACGT

•A PV1 VP4 PV1 VP4 PV1 VP4

C

—

HAV 5'NCrsgion HAV

HAV VP2

5'NC region

FIG. 3: DNA Sequence Determination of Plasmids pTZ PCR1 and PCR3 at the 5' and 3' Junction of the Inserted Poliovirus VP4 Sequences. The DNAs were prepared and sequenced by the method of Zagursky et al. (24). Shown is an autoradiograph of a 7% polyacrylamide gel containing the sequence of poliovirus VP4 capsid protein inserted between nucleotide 737 (pTZ.PCRl)/or 743 (pTZ.PCR3) at the 5'end of a subgenomic Hepatitis A cDNA (panel A) and nucleotide 807 (pTZ.PCRl and pTZ.PCR3 ) at the 3'end of a subgenomic Hepatitis A cDNA (panel B). Panel A is the sequence of the plus strand. Panel B is the sequence of the minus strand.

Nucleic Acids Research HAV VP4 serum or anti-poliovirus VP4 serum (Fig. 5). Only the chimeric proteins from the translated RNA pTZ.PCRl and PCR3 linearized by BstEII (which encoded the 69 amino acids of poliovirus and the amino-acids 2 to 129 of VP2 of HAV) were

HAV PCR1 PCR3 9Z5»» 69

•

—

46

»•

—

30

»•

—

FIG. 4: SDS-PAGE Analysis of Translated Products of T7 Transcribed RNA from pTZ HAV Hindm Cleaved with BglH and pTZ.PCRl or pTZ.PCR3 Cleaved with Bglll. Translations were performed as described in Materials and Methods. The products were electrophoresed on a 12% NaDodSO4/polyacrylamide gel and visualized by fluorography. Arrows mark the positions of 14C labeled marker polypeptides: phosphorylase (92.5 Kdal), bovine serum albumin (69 Kdal), ovalbumin (46 Kdal), carbonic anhydrase (30 Kdal), lysozyme (14.3 Kdal).

pTZ.HAV pTZ.PCRl pTZ.PCR3 6

f

FIG. 5: SDS-PAGE Analysis of Immunoprecipitates with Anti-HAV Capsid VP4 Serum or Anti-Poliovirus Capsid VP4 Rabbit Serum. Translation reactions from T7 transcribed RNA from pTZHAV HindUI cleaved by BstEII and pTZ.PCRl and pTZ.PCR3 cleaved by BstEII were immunoprecipitated by anti-HAV VP4 rabbit serum (lanes 1,3 and 5) or by anti-poliovirus VP4 rabbit serum (lanes 2, 4 and 6). The immune precipitates were then electrophoresed on a NaDodSO4 20% polyacrylamide gel and visualized by fluorography.

917

immunoprecipitated by the anti-poliovirus VP4 serum (Fig. 5, lanes 4 and 6) and no reactivity with anti-HAV VP4 serum was observed (Fig. 5, lanes 3 and 5). However the nonchimeric protein representing PI of HAV was precipitated by anti-HAV VP4 serum (Fig. 5, lane 1) but not by anti-VP4 poliovirus serum (Fig. 5, lane 2). These results confirmed that the sequences coding for VP4 poliovirus have been correctly inserted and translated. DISCUSSION The construction of chimeric DNA molecules can be a useful method to study the function of both nucleotide sequences and their translation products. HAV is a picornavirus that has several unique features including slow, non-cytolytic replication in cell culture (4), high thermal stability (5,6) and a relatively small VP4 capsid protein that distinguish it from the polioviruses and most other enteroviruses (7). In order to begin to determine if these features are related we have constructed a chimeric cDNA in which the HAV VP4 is replaced by the PV1 VP4. In this report we describe a new method for site specific mutagenesis of DNA sequences cloned in a single stranded pTZ phagemid (13). This mutagenesis method uses a modification of oligonucleotide site directed mutagenesis methods and involves a simple PCR reaction for producing specific single stranded DNAs (15) to be used like large mutagenic oligonucleotides. This technique can be used to modify precisely any DNA sequences cloned into a phagemid vector and in a sense is an alternate method of producing recombinant DNA molecules. This approach also offers the following advantages over the currently available procedures for mutagenesis: i) it is rapid; ii) the method permits replacement of exactly the sequences desired; iii) no restriction sites are required; iv) the sequence of only the termini and not the entire internal region need to be known; v) the ssDNA used as a large mutagenic oligonucleotide can be easily generated by PCR. While we generated single stranded PCR products by use of unequal primer concentrations, several other methods were available (29,30). We have successfully used the method of lambda exonuclease digestion (29) for the production of recombinant DNA by our technique. Other techniques which utilize PCR for mutagenesis have been described that are useful for making small substitutions (31), inversions (31,32,33) and deletions (31,32). The technique we have described in this report is useful for making relatively large substitutions and insertions without the necessity for restriction enzyme sites. As determined by the sequencing, we have successfully substituted the totality of 204 nucleotides coding for the 68 aminoacids of the poliovirus VP4 capsid protein in place of nucleotides 737-807 or 743-807 of HAV that code for the HAV VP4. It should be possible to insert much longer segments by this method. The method also could be adapted for making insertions or deletions as well as the type of substitutions we described in this paper. By using this novel technique it should be possible to produce cDNA molecules of many types that will enable the analysis of many aspects of the biology of HAV. For instance, specific genes can now be subcloned into expression vectors and studied in the absence of extraneous sequences derived from more commonly used methods involving fusion protein strategies. Genes or regions from an unsequenced mutant virus can be substituted for those of a well characterized genome and the product genome expressed

918 Nucleic Acids Research and examined for phenotype to locate important mutations or define gene functions. Hybrid viruses of virtually any type can be constructed as long as some form of the resultant nucleic acid is infectious. The ability to exchange any sequence without requiring restriction sites provides a powerful new tool to analyze gene function. In this context, we are presently inserting the subgenomic fragments described in this manuscript into a full length infectious clone of HAV cDNA (34) under approved containment levels in an attempt to produce viable chimenc viruses. At this time viable viruses have not been detected. ACKNOWLEDGEMENTS We thank Dr R. H. Purcell and Dr R. M. Chanock for encouragement and continuous interest in this work. We are grateful to Dr. J. Sninsky for assistance with the PCR, Dr E. Wimmer for the gift of anti-VP4 poliovirus serum and W. L. Maloy for the anti-VP4 HAV serum. We thank Terry Popkin and David Stec for photography and Linda Jordan for editorial assistance. This work was supported in part by Cetus Inc. Dr. C. Wychowski is a Fogarty Visiting Scientist. This study was supported in part by a grant from the World Health Organization Programme for Vaccine Development.

REFERENCES 1 Siegl, G., Frosner, G G , Gauss-Muller, V , Tratschw, J D and Deinhardt, F (1981) J Gen Virol 57,331-341 2 Coulepsis, A G , Locarniru, S A , Westaway, E G., Tannock, A and Gust, I D (1982) Intervirology 18, 107-127. 3. Gust.I D , Coulepsis, A G , Feinstone, S M , Locarniru, S. A , Montsugu, Y , Najera, R and Siegl, G (1983) Intervirology 20, 1-7 4 Gust, I D. and Feinstone, S M (1988) Hepatitis A CRC Press, Boca Raton, Fla) 5 Anderson, D A (1987) J Med Virol. 22, 35-44 6 Siegl, G , Weitz, M. and Kronauer, G (1984) Intervirology 22, 218-226 7. Baroudy, B M , Ticehurst, J R., Miele, T. A., Maizel, V , Jr , Purcell, R.H. and Feinstone, S M (1985) Proc Natl. Acad. Sci USA 82, 2143-2147 8. Koch, F and Koch, G (1985) The molecular biology of poliovirus SpnngerVerlag, Vienna 9. Rueckert, R R. (1985) In Fields, B N , Kmpe, D M., Chanock, R M , Melmck, J L., Roizman, B and Shope, R E. (eds ), Virology Raven Press, New York. 10 Smith, M. and Gillam, S (1981) In Setlow, J and Hollaender, A (eds ), Genetic Engineering Plenum, New York, Vol 3, pp 1 —32 11 Zoller, M.D and Smith, M. (1983) Meth. Enzymol. 101, 469-500. 12. Mead, D A , Skorupa, E. S. and Kemper, B. (1985) Nucleic Acids Res 13, 1103-1117 13 Mead, D. A , Skorupa, E S and Kemper, B (1986) Protein Engineering 1, 6 7 - 7 4 . 14 Schold, M , Colombero, A , Reyes, A A and Wallace, R B (1984) DNA 3, 469-477 15. Gyllensten, V. B and Erlich, H. A (1988) Proc. Natl Acad Sci USA 85, 7652-7656 16 Scharf, S J , Horn, G T andEruch, H A. (1986) Science 233, 1076-1078. 17. Maruaus, T., Fritsch, E F. and Sambrook, J (1982) Molecular Cloning: A Laboratory Manual Cold Spring Harbor Laboratory Press, Cold Spring Harbor. 18 Wieslander, L (1979) Anal Biochem 98, 305-309. 19. Taylor, J. W , Ott, J. and Eckstein, F (1985) Nucleic Acids Res 13, 8765-8785. 20 Viera, J. and Messing, J (1987) Methods Enzymol. 153, 3 - 1 1 21 Boyer, H. W. and Roulland-Dussoix, D (1969) J Mol. Biol 41,459-472 22 Grunstein, M. and Hogness, D S (1975) Proc. Natl. Acad Sci. USA 72, 3 % 1-3965 23 Feinberg, A P and Vogclstein, B (1983) Anal. Biochem 1 3 2 , 6 - 1 3 . 24. Zagursky, R. J., Baumeister, K., Lomax, N. and Bcrman, M. L (1985) Gene Anal. Tech. 2, 89-94.

25. Cohen, J I , Trcehurst, J R , PurceU, R H., Buckler-White, A. and Baroudy, B M (1987) J Virol 61, 50-59 26 Sanger, F , Nicklen, S and Coulson, A R (1977) Proc Natl Acad Sci USA 74, 5463-5467 27 Wilbams, P M , Williamson, K J , Emerson, S U and Schubert, M (1988) Virology 164, 176-181. 28 Erruni, E A, Doroer, A J , Dorner, L. F , Jameson, B A and Wimmer, E. (1983) Virology 124, 144-151 29 Higuchi, R G. and Ochman, H. (1989) Nucleic Acids. Res. 17, 5865. 30 Mitchell, L G and Meml, C R (1989) Anal Brochem 178, 239-242 31 Vallette, F , Mege, E , Reiss, A. and Adesrik, M (1989) Nucleic Acids Res 17, 723-733 32. Ho, S N , Hunt, H D , Horton, R M , Pullen, J K and Pease, L R (1989) Gene 77, 51-59 33 Kammann, M , Laufs, J , Schell, i and Gronenborn, B (1989) Nucleic Acids Res 17, 5404 34. Cohen, J I , Ticchurst, J R , Feinstone, S M , Rosenblum, B and Purcell, R H (1987) J Virol 61, 3035-3039