`
`
`
`AMERICAN
`AssOCIATION FOR THE
`ADVANCEMENT OF
`SciENCE
`
`Editorial
`Letters
`
`• CIENCE
`
`ISSN 0036-8075
`5 SEPTEMBER 1986
`VoLUME 233
`NUMBER 4768
`
`1015 This Week in Science
`
`1017 Beauty, Balance, and Mathematics
`
`1027 Probability Estimates: A. M. GRIMWADE • NASA's Bureaucracy:
`A. M. SQUIREs; P. BEARSE • Protein Structure: D . B. WETLAUFER
`
`News & Comment
`
`1029 Chernobyl : Errors and Design Flaws
`
`1031 Briefing: Earthquake Research Center Siting Triggers California Tremors
`1032 Medicine as Business: Are Doctors Entrepreneurs?
`
`1034 B1iejing: Lindow Microbe Test Delayed by Legal Action Until Spring • PHS
`Invites Industry Collaboration on AIDS Vaccine • Will an AIDS Vaccine
`Bankrupt the Company That Makes It? • NASA Council Sees Continued Erosion
`of Space Program • Soviet Union Suspends Plans to Divert Four Rivers
`
`Mat1Mik411~14i¥-
`
`1037 Trying to Crack the Second Half of the Genetic Code
`
`1039 Do California Quakes Portend a Large One?
`
`1040 Washington Embraces Global Earth Sciences
`
`Articles
`
`Research Articles
`
`Reports
`
`1043 Strong Ground Motion from the Michoacan, Mexico, Earthquake:
`]. G. ANDERSON, P. BODIN,] . N. BRUNE,] . PRINCE, s. K. SINGH, R. QUAAS,
`M. ONATE
`
`1050 Multiple DNA-Protein Interactions Governing High-Precision DNA
`Transactions: H. EcHOLS
`
`1056 Natural Philosophy in the Constitution: G. PIEL
`
`1061
`
`Induction of Membrane Ruffling and Fluid-Phase Pinocytosis in Quiescent
`Fibroblasts by ms Proteins: D. BAR-SAGI AND]. R. FERAMISCO
`
`1069 Catalytic H ydration of Terminal Alkenes to Primary Alcohols: C. M. ]ENSEN
`AND W. C. TROGLER
`
`1071 Coastal Uplift and Mortality of Intertidal Organisms Caused by the September
`1985 Mexico Earthquakes: P. BoDIN AND T. KLINGER
`
`1073 Transition State Analogs as Ligands for Affinity Purification of]uvenile Hormone
`Esterase: Y. A. I. ABDEL-AAL AND B. D. HAMMocK
`
`•
`
`•
`
`SCIENCE Is published weekly on Friday, except the last week In December, and with a plus Issue In May by the
`American Association lor the Advancement of Science, 1333 H Street, NW, Washington, DC 20005. Second-class
`postage (publication No. 484460) paid at Washington, DC, and at an additional entry. Now ~ombined with The Scientific
`Monthly® Copyright © 1986 by the American Association lor the Advancement of Science. The title SCIENCE is a reg·
`istered trademark ol the AAAS. Domestic individual membership and subscription (51 issues): $65. Domestic institutional
`subscription (51 issues): $98. Foreign postage ex1ra: Canada $24, other (surface mail) $27, air-surface via-Amsterdam
`$65. First class, airmail, school-year, and student rates on request. Single copies $2.50 ($3 by mail); back issues $4
`($4.50 by mail); Biotechnology issue, $5.50 ($6 by mail); classroom rates on request; Guide to Biotechnology Products
`and Instruments $16 ($17 by mail). Change of address : allow 6 weeks, giving old and new addresses and seven-digit
`account number. Authorization to photocopy material lor internal or personal use under circumstances not falling within
`the fair use provisions of the Copyright Act is granted by AAAS to libraries and other users registered with the Copyright
`Clearance Center (CCC) Transactional Reporting Service, provided that the base fee of $1 per copy plus $0.10 per page
`is paid directly to CCC, 21 Congress Street, Salem, Massachusetts 01970. The identification code for Science is 0036·
`8075/83$1 + .10. Postmaster : Send Form 3579 to Science, 1333 H Street, NW, Washington, DC 20005. Science is in·
`dexed in the Reader's Guide to Periodical Literature and in several specialized indexes.
`The American Association for the Advancement of Science was founded in 1848 and incorporated in 1874. lis objects
`are to further the work of scientists, to facilitate cooperation among them, to foster scientific freedom and responsibility,
`to improve the effectiveness of science in the promotion of human welfare, and to increase public understanding and ap·
`preciation of the importance and promise of the methods of science in human progress.
`
`1012
`
`SCIENCE, VOL. 233
`
`GeneDX 1010, pg. 2
`
`
`
`13. H. G. Bull, N. A. Thornberry, E. H . Cordes,]. Bioi.
`Cbcm. 260, 2963 (1985).
`14. M. A. Ashour ct nl. , personal communication; M.
`H. Gelb, J. P. Svaren, R. H. Abeles, Bioc!Jwust>)' 24,
`1813 (1985).
`15. R. Wolfenden, A111111. RCI'. Biopbys. Biocng. 5, 271
`(1976); Mct/Jods Enzymo/. 46, 15 (1977); U. Brod(cid:173)
`beck, inEnzyme ln/JibitOI~, U. Brodbeck, Ed. (Verlag
`Chemic, Deerfield, FL, 1980), p. 3.
`
`16. N. 0 . Kaplan ct nl., Proc. Nntl. A end. Sci. U.S.A . 71 ,
`3450 (1974); T. ]. Bazzone ct nl., Bioc!Jcmistl)' 18,
`43~2 (1979); N. ]. Hoogenraad, T. M. Sutherland,
`G. ]. Howlett, Ann/. Bioc/Jcm. 101 , 97 (1980) ; L.
`Andersson and R. Wolfenden, J. Bioi. CIJCm. 225,
`11106 (1980); R. Wolfenden and L. Andersson, in
`C/Jcmicnl Appronc/Jes to Undc>!tnnding Enz_yme Cntnl(cid:173)
`)'sis: Biomimetic C/Jcmist>y nnd Tmnsition State Ann(cid:173)
`logs, B. S. Green, Y . Ashani, D. Chipman, Eds.
`
`(Elsevier, Amsterdam, 1982), vol. 10, pp. 287-297.
`17. Supported by grants from the Herman Frasch Foun(cid:173)
`dation and NIEHS , NSF, and USDA (ES02710-
`06, DCB-8518697, and 85-CRCR-1-1715, respcc(cid:173)
`tivcl\•) . vVe thank P. Bartlett, R. Wolfenden, S.
`Duffey, T. Sparks, and other colleagues for valuable
`comments.
`
`27 January 1986; accepted 12 May 1986
`
`Direct Cloning and Sequence Analysis of
`Enzymatically Amplified Genomic Sequences
`
`STEPHEN J. ScHARF, GLENN T. HoRN, HENRY A. ERLICH
`
`A method is described for directly cloning enzymatically amplified segments of
`genomic DNA into an Ml3 ve<;tor for sequence analysis. A llO-base pair fragment of
`the human ~-globin gene an~ a 242-base pair fragment of the human leukocyte
`antigen DQa locus were amplified by the polymerase chain reaction method, a
`procedure based on repeated cycles of denaturation, primer annealing, and extension
`by DNA polymerase I. Oligonucleotide primers with restriction endonuclease sites
`added to their 5' ends were used to facilitate the cloning of the amplified DNA. The
`analysis of cloned products allowed the quantitative evaluation of the amplification
`method's specificity and fidelity. Given the low frequency of sequence errors observed,
`111-is approach promis~,:s to be a rapid method for obtaining reliable genomic sequences
`from nanogram amounts of DNA.
`
`U NDERSTANDING THE MOLECULAR
`
`basis of genetic disease or of com(cid:173)
`plex genetic polymorphisms, such
`as those ·in the human leukocyte antigen
`(HLA) region , requires detailed nucleotide
`sequence information from a variety of indi(cid:173)
`viduals to localize relevant variations. Cur(cid:173)
`rently, the an<! lysis of each allelic variant
`requires a substantial effort in library con(cid:173)
`stmction, screening, mapping, subcloning,
`and sequencing. We report here a method
`for the enzymatic an1plification of specific
`segmepts of genomic DNA and their direct
`cloning into Ml3 vectors for sequence anal(cid:173)
`ysis, using modifications of tl1e polymerase
`chain reaction (PCR) (1, 2). This in vitro
`amplification procedure is based on repeated
`cycles of denaturation, oligonucleotide
`primer annealing, and primer extension by
`the Klenow fragment of DNA polymerase,
`and results in an exponential increase in
`copies of the region flanked by the primers.
`This method greatly reduces tl1e nt.Jmber of
`cloned DNA fragments to be screened, cir(cid:173)
`cumvents the need for full genomic libraries,
`and may allow cloning from nanogram
`quantities of genomic DNA. In addition,
`the cloning and sequencing of PCR-ampli(cid:173)
`fied DNA i~ a powerful analytical tool for
`tl1e study of the specificity and fidelity of
`tl1is newly developed teclmique.
`To develop this teclmique for genomic
`sequence determination and to analyze the
`individual products of PCR amplification,
`we chose the oligonucleotide primers and
`
`probes previously described for tl1e diagno(cid:173)
`sis of sickle cell anemia (2) . These primers
`amplifY a 110-bp segment of the hwnan f3-
`hemoglobin gene containing the Hb-S plU(cid:173)
`tation (Fig. 1). They were modified near
`their 5' ends (Fig. l ) to produce convenient
`restriction sites (linkers) for cloning directly
`into the Ml3mpl0 sequencing vector (3).
`These modifications did not affect the effi(cid:173)
`ciency of PCR an1plification of tl1e specific
`f3-globin segment (Fig. 2B), even tl10ugh
`tl1e overall pattern of PCR-amplified prod(cid:173)
`ucts is different (Fig. 2A). After amplifica(cid:173)
`tion, the PCR products were cleaved with
`tl1e appropriate restriction enzymes and dia(cid:173)
`lyzed to remove inhibitors of ligation. These
`fragments were ligated into the Ml3 vector,
`transformed into the JM103 host, and plat(cid:173)
`ed out, and the resulting plaques were
`screened by hybridization witl1 a labeled
`oligonucleotide probe to detect tl1e f3-globin
`clones. The plaques were also screened with
`the labeled PCR oligonucleotide prin1ers to
`identify all of tl1e clones containing ampli(cid:173)
`fied DNA. Individual clones were tl1en se(cid:173)
`quenced directly by using the dideoxy prim(cid:173)
`er-extension method (4).
`The incorporation of different restriction
`sites at the termini of tl1e amplified product
`allows digestion of the vector with two
`restriction enzymes, thereby reducing the
`background of blue plaques from vectors
`without inserts to less tl1an 11% (Table l).
`More tl1an 80% of the clones contained
`tl1e PCR primer se-
`DNA inserts witl1
`
`quences but only about l% of the clones
`hybridized to tl1e internal f3-globin probe.
`These nonglobin fragments presumably rep(cid:173)
`resent amplifications of other segments of
`the genome. This observation is consistent
`witl1 the gel and Soutl1ern blot analysis of
`tl1e PCR-amplified DNA from a f3-globin
`deletion mutant and a normal ceU line (Fig.
`2). The sil11ilarity of the observed gel pro(cid:173)
`files reveals tl1at most of the amplified geno(cid:173)
`mic DNA fragments arise from nonglobin
`templates. Sequence analysis of two of these
`nontarget clones showed tl1at tl1e segments
`between the PCR primer sequences were
`unrelated to the f3-globin gene and con(cid:173)
`tained an abw1dance of dinucleotide repeats,
`sin1ilar to some genomic intergenic spacer
`sequences (5).
`When ten of the clones that hybridized to
`the f3-globin probe were sequenced, nine
`proved to be identical to the f3-globin gene
`and one contained five nucleotide differ(cid:173)
`ences but was identical to the &-globin gene.
`The f3-globin PCR primers each have two
`mismatches witl1
`tl1e &-globin sequence
`(Fig. l) . The preferential PCR an1plification
`of f3-globin relative to &-globin observed in
`our clonal analysis agrees with tl1e results
`obtained with the oligomer restriction assay
`on the PCR reaction (2, 6) . Each of tl1ese
`ten sequenced clones contains a segment of
`70 bp originally syntl1esized from tl1e geno(cid:173)
`mic DNA template during the PCR amplifi (cid:173)
`cation process. Since no sequence alterations
`were seen in tl1ese clones, the frequency of
`nucleotide misincorporation during 20 cy(cid:173)
`cles of PCR amplification is less tl1a11 l in
`700.
`To analyze the molecular basis of genetic
`polymorphism and disease susceptibility in
`tl1e HLA class II loci, tl1is approach has been
`extended to the an1plification and cloning of
`a 242-bp fragment from the second exon of
`tl1e HLA DQo: locus, which exhibits local(cid:173)
`ized aUelic variability. In tlus case the prim(cid:173)
`ers, whose sequence is based on conserved
`regions of this exon, contain 5' -terminal
`restriction sites witl1 no homology to the
`DQo: sequence (Fig. 1). The specificity of
`amplification achieved with tl1ese primers is
`greater than that achieved with tl1e f3-globin
`
`Cetus Corporation , Department of Human Genetics,
`1400 Fifty-Third Street, Emeryville, CA 94608.
`
`SC IEN CE, VOL. 233
`
`J
`
`GeneDX 1010, pg. 3
`
`
`
`A
`
`Human {3 -hemoglobin
`
`PC03: ACACAACTGTGTTCACTAGC ~
`Met .•.
`GH18: CTTCTGcagCAACTGTGTTCACTAGC ~
`... TTACATTTGCTTCTGACACAACTGTGTTCACTAGCAACCTCAAACAGACACCATGGTGCAC ... TGCCCTGTGGGGCAAGGTGAACGTGGATGAAGTTGGTGGTGAGGCCC .. .
`.. . AATGTAAACGAAGACTGTGTTGACACAAGTGATCGTTGGAGTTTGTCTGTGGTACCACGTG ... ACGGGACACCCCGTTCCACTTGCACCTACTTCAACCACCACTCCGGG .. .
`~ CCACTTGCACCTACTTCgAaCAC
`:GH19
`~ CCACTTGCACCTACTTCAAC
`:PC04
`
`B
`
`119 bp
`
`Human HLA DQ a
`
`GH26: gtgctgcaGGTGTAAACTTGTACCAG ~
`... ACCACGTTGCCTCTTGTGGTGTAAACTTGTACCAGTTTTACGGTCCCTCTGGCCAGTACAC ... TTAAACGCTACAACTCTACCGCTGCTACCAATGAGGTTCCTGAGGTC .. .
`... TGGTGCAACGGAGAACACCACATTTGAACATGGTCAAAATGCCAGGGAGACCGGTCATGTG ... AATTTGCGATGTTGAGATGGCGACGATGGTTACTCCAAGGACTCCAG .. .
`~ GTTGAGATGGCGACGATGGcctaggCAc
`:GH27
`
`242 bp
`
`Fig. l. Oligonucleotide primers for PCR amplification. (A) The two central
`sequence lines show both strands of the human l3-hcmoglobi..t1 gene (5) from
`52 bp upstream of the i..t1itiation codon (Met) to 84 bp downstream. Above
`and below this sequence are the oligonucleotide primers used to amplif)r this
`segment of the genome. The 3' end of each primer is marked by an arrow to
`i..t1dicate the direction of extension by DNA polymerase. Dots indicate
`nucleotide differences between 13- and o-globin . The primers PC03 and
`PC04 were used in our previous study (1, 2 ) and are completely homologous
`to the target gene. These sequences were extended and modified to produce
`the linker-primers GH18 and GH19. The modifications (lowercase letters)
`create a Pst I site in GH18 and a Hind III site in GH19. The amplified
`
`segment is 110 bp with PC03 and PC04, and 119 bp with GH18 and
`GH19. (B) The central sequences show a segment of the human HLA DQa
`gene (10 ) from codons 9 to 92 (codons are numbered for the mature
`protei..t1). Tlus region codes for the highly polymorphic outer domain of the
`protein. As above, the oligonucleotide lmker-primcr sequences used to
`an1plif)r this segment arc shown, with lowercase letters dcsignati..t1g nonho(cid:173)
`mologous bases. In this case the desired restriction sites (Pst I for primer
`GH26 and Ban1 HI for GH27) were added to the 5' end of the primers
`without regard to the target sequence. The length of the amplified product
`from HLA DQa is 242 bp.
`
`Table l. Distribution of[3-globin PCR clones. An
`entire PCR reaction was digested at 37"C with Pst
`I (20 U) and Hind III (20 U ) for 90 mi..tmtes (for
`13-globm) or Bam HI (24 U ) and Pst I (20 U ) for
`60 minutes (for HLA DQa ). After phenol extrac(cid:173)
`tion (11 ), the DNA was dialyzed to remove low
`molecular weight inhibitors of ligation (presum(cid:173)
`ably the deoxynucleoside triphosphates used in
`PCR), and concentrated by ethanol precipitation.
`All (13-globin) or one tenth (DQa) of the material
`was ligated to 0.5 fLg of the cut M13 vector under
`standard conditions (11) and transformed into
`approxilllately 6 x 109 cells of freshly prepared,
`competent JM103 (3) in a total volwne of200 fLl.
`Fresh JM103 culture (150 fLl ) was then mixed
`with 10 to 30 fLl of the transformed cells, plated
`on IPTG/X-gal agar plates (3 ), and i..tKubatcd
`overnight. The plates were scored for blue (paren(cid:173)
`tal) plaques and lifted onto BioDyne A filters (3) .
`These filters were hybridized (2) either with the
`labeled PC04 oligonucleotide to visualize aU of
`the clones containi..t1g PCR-an1plified DNA
`(prm1er plaques) or with the RS24 oligonucleo(cid:173)
`tide probe (the exact complement of the RS06 (6)
`probe] to specifically visualize the clones contain(cid:173)
`i..tlg hemoglobin sequences (globin plaques). Ten
`clones from the latter category were sequenced by
`the dideo:-)' extension method (4 ). Nine were
`identical to the expected [3-hemoglobi..t1 target
`sequence and one was identical to the homolo(cid:173)
`gous region of the human o-hemoglobm gene (5).
`
`Category
`
`Number
`
`Total plaques
`White plaques
`Primer plaques
`Globill plaques
`
`1496
`1338
`1206
`15
`
`Frequency
`(%)
`
`100
`89
`81
`1
`
`A
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`B
`
`2
`
`3 4
`
`5
`
`6
`
`7
`
`8
`
`- 242bp -
`
`-119 bp -
`
`Globin
`
`DQ a.
`
`Fig. 2. PCR an1plification with oligonucleotide linker-prm1ers. DNA was amplified by mi..xing 1 fLg of
`genomic DNA in polymerase bufrer (2) with 100 pmol of each primer. San1plcs were subjected to 20
`cycles of PCR, each consisting of 2 mmutes of denaturation at 95°C, 2 minutes of coolillg to 37°C, and
`2 minutes of polymerization with 1 U of Klenow DNA polymerase. After an1plification, the DNA was
`concentrated by ethanol precipitation and half of the total reaction subjected to electrophoresis on a gel
`of 4% NuSievc agarosc in tris-boratc-EDTA butter. The gel was photographed, and the DNA was
`transferred to Genatran nylon membrane and hybridized (2) to a labeled probe specific for the target
`sequence. The blot was then washed and autoradiographed. For the amplification of 13-globm, the
`starti..tlg DNA was either from the Molt-4 cell li..t1e or from the globin deletion mutant GM2064 (2 ). For
`lane 5, the starting material was 11 pg of the 13-globin recombinant plasmid pBR328 :: 13A (6) , the molar
`equivalent ofS fLg of genomic DNA. For lane 6, the reaction was performed as in lane 1 except that no
`enzyme was added. To mcrease the efficiency of an1plification of the longer HLA DQa segment,
`dill1ethyl sulfoxide was added to 10% and the polymerization was carried out at 37°C for 27 cycles. The
`starting DNA for the amplification of DQa was either from the consanguineous HLA-typing ceU line
`LG-2 or from the HLA class II deletion mutant LCL72l.180 (7) . (A) Ethidiwn bronude-stamed gel
`showing total an1plified products. (Lane 1) Pri..tners PC03 and PC04 on Molt-4 DNA; (lane 2) PC03
`and PC04 on GM2064 DNA; (lane 3) GH18 and GH19 on Molt-4; (lane 4) GH18 and GH19 on
`GM2064; (lane 5) GH18 and GH19 on pBR328::13A DNA; (lane 6) PC03 and PC04 on Molt-4, no
`enzyme; (lane 7) primers GH26 and GH27 on LG-2 DNA; (lane 8) GH26 and GH27 on
`LCL72l.l80 DNA. (B) Southern blots showing specific amplified products. Lanes arc nW11bered as m
`(A). Lanes 1 through 6 were hybridized to the labeled RS06 oligonucleotide probe (2). Lanes 7 and 8
`were hybridized to a cloned DQa eDNA probe labeled by nick-translation (11).
`
`5 SEPTEMBER 1986
`
`REPORTS 1077
`
`GeneDX 1010, pg. 4
`
`
`
`primers, since gel electrophoresis of the
`PCR products reveals a discrete band at 240
`bp absent from the HLA deletion mutant
`(7) (Fig. 2). In addition, hybridization
`screening of the Ml3 clones from this am(cid:173)
`plification indicates that about 20% are ho(cid:173)
`mologous to the DQa probe, an increase of
`20 times over the [3-globin amplification.
`The basis for the difference in the specificity
`of amplification, defined as the ratio of
`target to nontarget clones, is not dear. It is
`Likely to reflect the primer sequences and
`their genomic distribution rather than the
`difFerent reaction conditions used (Fig. 2).
`The differences between the sequence of the
`[3-globin primers GH18, GH19 and PC03,
`PC04 (Fig. l ) may account for the observed
`change in the gel profile of the PCR-ampli(cid:173)
`fied products (Fig. 2, lanes l and 2 versus
`lanes 3 and 4), reflecting some difference in
`the amplification of nontarget segments.
`Three HLA DQa PCR clones derived
`from the homozygous typing cell LG2 were
`subjected to sequence analysis (8). Two
`clones were identical to a DQa complemen(cid:173)
`tal)' (eDNA) done from the same cell line.
`One differed by a single nucleotide, indicat(cid:173)
`ing an error rate of approximately l/600,
`assuming the substitution occurred during
`the 27 cycles of amplification. This proce(cid:173)
`dure has also been used to analyze sequences
`from polymorphic regions of the HLA
`DQf3 and DRf3 loci.
`Our rapid method for the cloning of
`specifically an1plified genomic fragments re(cid:173)
`lies on the ability of oligonucleotides to
`function as PCR primers with w1paired
`bases near their 5' ends. In the later cycles of
`amplification these primers anneal primarily
`to the amplified products rather than to the
`original genomic sequences, and are there(cid:173)
`fore fully complementary. The [3-globin
`linker-primers GH18 and GH19 appear to
`be approximately as efficient and specific as
`the fully matched primers. The HLA DQa
`linker-primers GH26 and GH27, with even
`more 5' mismatches, show an amplification
`specificity 20 times higher than the [3-globin
`primers. A large number of DQa clones
`were obtained from just 100 ng of this
`amplified genomic DNA. Our data suggest
`that the error rate over many cycles of
`amplification is sufficiently low so that reli(cid:173)
`able genomic sequences can be determined
`directly from PCR an1plification and don(cid:173)
`mg.
`This procedure offers significant advan(cid:173)
`tages over standard cloning protocols for the
`analysis of sequence polymorph isms in that
`it circumvents the construction and screen(cid:173)
`ing of full genomic libraries and could po(cid:173)
`tentially be
`initiated
`from nanogram
`amounts of DNA. It is capable of isolating
`only a limited region of tl1e genome, howev-
`
`10 78
`
`er, and requires sequence information to
`identify conserved primer and probe seg(cid:173)
`ments.' Unlike direct genomic sequencing
`(9), tl1e cloning of amplified DNA allows
`the separation of related genes and alleles
`prior to sequencing and does not require a
`detailed knowledge of adjacent restriction
`sites. The application of oligonucleotide
`linker-primers to introduce specific restric(cid:173)
`tion sites into PCR-amplified DNA may al(cid:173)
`so prove useful as a general cloning strategy.
`
`REFERENCES AND NOTES
`
`l. F. Faloona and K. B. Mullis, Methods E11zymol., in
`press.
`2. R. K. Saiki et rrl., Scie11ce 230, 1350 (1985 ).
`3. J. Messing, Methods E11zymo/. 101, 20 (1983).
`
`IITG, isopropyl-J3-o·thiogalacto·pyranoside; X-gal,
`5-bromo-4-chloro-3-indolyl-J3-o-galactoside.
`4. L Guo and R. Wu, ibid. 100, 60 (1983).
`5. M. Poncz, E. Schwartz, M. Ballantine, S. Surrey, f .
`Bioi. C!Jem. 258, 11 599 (1983).
`6. R. Saiki, N. Arnheim, H . Erlich, Biotech11oiO!fY 3,
`1008 (1985).
`7. H. Erlich, ]. S. Lee, J. W. Petersen, T. Bugawan, R.
`DeMars, Hum. lmmtmol. 16, 205 (1986).
`8. T. L Bugawan et rr/.,unpublished results.
`9. G. M. Church and W. Gilbert, Pl'oc. Nrrtl. Acrrd. Sci.
`U.S.A. 81 , 1991 (1984).
`10. C. Auffray et rrl., Nrrtul'e (Lo11do11) 308, 327 (1984).
`11. T. Maniatis, E. F. Fritsch, J. Sambrook, in .Moleculm·
`Clo11i11g (Cold Spring Harbor Laboratory, Cold
`Spring Harbor, NY, 1982).
`12. We thank C. Long for the sequencing of PCR
`amplified DNA's; C. Levenson, L Gada, and D.
`Spasic for
`the synthesis of oligonucleotides; ].
`Sninsky, R. Saiki, and K. Mullis for critical review;
`and K. Levenson for the preparation of this manu(cid:173)
`script.
`21 April1986; accepted 17 July 1986
`
`Inhibition of Endothelial Regeneration by Type-Beta
`Transforming Growth Factor from Platelets
`
`RONALD L. HEIMARK, DANIEL R. TWARDZIK, STEPHEN M . SCHWARTZ
`
`Damage to the vessel wall is a signal for endothelial migration and replication and for
`platelet release at the site of injury. Addition of transforming growth factor-beta
`(TGF-f3) purified from platelets to growing aortic endothelial cells inhibited [3H]thy(cid:173)
`midine incorporation in a concentration-dependent marmer. A transient inhibition of
`DNA synthesis was also observed in response to wounding; cell migration and
`replication are inhibited during the first 24 hours after wounding. By 48 hours after
`wounding both TGF-[3-treated and -untreated cultures showed similar responses.
`Flow microfiuorimetric analysis of cell cycle distribution indicated that after 24 hours
`of exposure to TGF-f3 the cells were blocked from entering S phase, and the fraction of
`cells in G 1 was increased. The inhibition of the initiation of regeneration by TGF-f3
`could allow time for recruitment of smooth muscle cells into the site of injury by other
`platelet components.
`
`T HE CONTINUITY OF THE VASCUlAR
`
`endothelium is lost as atherosclerotic
`lesions progress (1) . In contrast,
`even in large wounds the endotheliwn rap(cid:173)
`idly regenerates both in vitro and in vivo
`(2). This raises the possibility that some
`property of the atl1eromatous wall or ele(cid:173)
`ment of the blood interacting with the wall
`at tl1e developing plaque can act to prevent
`the normal processes of endothelial regener(cid:173)
`ation. An obvious source of such an effect at
`sites of denudation is tl1e platelet. Among
`tl1e components released from platelets at a
`site of vascular injury are a group of growth
`regulatory proteins, including platelet-de(cid:173)
`rived growm factor (PDGF), a growth fac(cid:173)
`tor similar
`to epidermal growtl1 factor
`(EGF), and transforming growth factor(cid:173)
`beta (TGF-[3) (3, 4).
`The endotheliwn grows as a strictly densi(cid:173)
`ty-inhibited monolayer. Although endothe(cid:173)
`lial cell growth requires serum or plasma,
`PDGF and EGF are not required (5) . In the
`presence of acid-treated serum, aortic endo-
`
`thelia! cells have been shown to form colo(cid:173)
`nies in soft agar ( 6) . Similar results were
`obtained from mouse embryo cells (AKR-
`2B) or normal rat kidney (NRK) cells and
`tl1e activity that promotes growth in soft
`agar was identified as TGF-[3 (4) . We report
`here that TGF-f3 purified from human plate(cid:173)
`lets can inhibit the endothelial regeneration
`process by inl1ibiting both replication and
`migration.
`Transforming growth factors were orig(cid:173)
`inally found in viral-transformed rodent
`cells, and it was postulated that they were
`involved in neoplastic cell growth (7, 8).
`However, their presence in normal tissues,
`including kidney, placenta, and platelets,
`suggests a more general function ( 4, 9).
`Platelets contain 40 to l 00 times as much
`TGF-f3 as other nonneoplastic tissues. TGF-
`
`R. L. Heimark and S. M. Schwartz, Department of
`Pathology, SJ-60, University of Washington, Seartle,
`WA 98195.
`D. R. Twardzik, Oncogcn, Seattle, WA 98121.
`
`SC I ENCE, VOL. 233
`
`GeneDX 1010, pg. 5