Characterization of genomic and complementary DNA sequence of ...

THEJOURNAL

OF

BIOLOGICAL CHEMISTRY

Vol. 260, No. 24, Issue of October 25, pp. 13384-13388 1985 Printed in C.S.A.

0 1985 by The American Society of Biological Chemists, Inc.

Characterization of Genomic and Complementary DNA Sequence of Human C-reactive Protein, and Comparisonwith the Cornpiementary DNA Sequence of Serum Amyloid P Component* (Received for publication, March 26, 1985)

Patricia Woo, Julie R. KorenbergS, and Alexander S. Whiteheads From the Division of Cell Biology, Children’s Hospital, Department of Pediatrics, Harvard Medical School, Boston, Massachusetts 02115

ComplementaryandgenomicDNA clonescorre(reviewed in Ref. 5 ) . Human CRP and SAP show considerable sponding to the human C-reactiveprotein(CRP) amino acid homology (6). The structuralgenes for both have mRNAand structural gene have been analyzed and been mapped to chromosome 1 (7, S), and are probable prodcompared. Nucleotide sequencing of the coding regions ucts of gene duplication. The entire amino acid sequence of of both cDNA and genomic DNA revealed an additional human CRP has been reported (9). In thispaper, we present 19 amino acid peptide not described in the published nucleotide sequence of the CRP mRNA obtained from analCRP amino acid sequence. The CRP gene contains a ysis of CRP complementary DNA clones. The derived amino single 278 base pair intron within the codon specifying acid sequence of mature CRP revealed a region of 19 amino the third residue of matureCRP. The intron contains acids within the mature product not previously reported. a repetitive sequence (GT)1SG(GT)3 which is similar to structurescapableofadoptingthe Z-DNA form. A Analysis of a CRP genomic clone confirmed the presence of comparison of CRP coding and amino acid sequences this additional peptide. The genomic clone also permitted determination of the complete nucleotide sequence of the with those of serum amyloid P component revealed gene and definition of a single intron of 27.8 base pairs which striking overall homology which was not uniform: a region of limited conservation is bounded bytwo highly exhibits several unusual features. The amended amino acid sequence of CRP iscompared with that of SAP and thedegree conserved regions, of homology at thenucleotide and protein level is presented. MATERIALS ANDMETHODS

C-reactive protein (CRPl) is one of the major plasma proteins that increase in concentration during the acute phase response to tissue injury or inflammation. It was originally defined by its ability to bind to the C-polysaccharide of Streptococcus pneumoniae (1).CRP displays several functions associated with host defense; for example, it promotes agglutination, bacterial capsular swelling, phagocytosis, and complement fixation through its calcium-dependent binding to phosphorylcholine (reviewed in Ref. 2). Recently, Robey et al. (3) described the binding of CRP tochromatin and postulated that itsmajor role is as a scavenger for chromatin during cell necrosis. Serum CRP belongs to the family of pentraxins. These proteins have a discoid arrangement of 5 noncovalently bound subunits and are found in the sera of many species ranging from horseshoe crab to higher mammals (4). In human serum, there are two pentraxins, CRP and serum amyloid P component (SAP). The latterdiffers from CRP in that itis not an acute phase protein, it is glycosylated, and in serum it is present as a complex of two pentameric discs instead of one. SAP is a precursor of amyloid P component which is found in basement membranes and associated with amyloid deposits

* This work was supported by United States Public Health Service Grant A120959 and the Charles P. Hood Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact. $ Present address: Department of Pediatrics, M650, University of California, San Francisco, CA 94143. 3 Supported by a Helen Hay Whitney FoundationFellowship (488). The abbreviations used are: CRP,C-reactive protein; SAP,serum amyloid P component.

CRP Complementary DNA Clones-The isolation of two complementary DNA clones, pCRPl andpCRP5, from an adult humanliver library (10) has been described previously (7, 11). Isolation of CRP GenomicClones-A 5.4-kilobase pair genomic CRP clone was isolated from a partial EcoRI human genomic library (12) (gift of T. Maniatis,Harvard University) by the method of Benton and Davis (13). The library was plated out at high density and transferred to nitrocellulose filters. Thesewere hybridized under standard conditions (14) with the cDNA probe pCRP5 which had been radiolabeled by nick translation (15). Following hybridization the filters were washed in 30 mM NaCl, 3 mM Na citrate, 0.1% sodium dodecyl sulfate a t 65 “C for 1 h, dried and visualized by autoradiography. Clones giving positive hybridization signals were plaque-purifled. One CRP specific clone CRPg2 was selected for further analysis. This clone was digested with the restriction endonuclease EcoRI, and the three fragments obtained were subcloned into the plasmid vector obtained were subcloned into theplasmid vector pKT218. Sequence Analysis of CRP Complementary and Genomic DNACRP specific complementary and genomicDNA clones were sequenced by both thechemical cleavage method of Maxam and Gilbert (16) and by the dideoxy chain termination method of Sanger (17) following subcloning into thephage vectors M13mpll andM13mp18. Oligonucleotide Synthesis-Eleven unique sequence oligonucleotides for use as sequencing primers were synthesized by the modified phosphotriester method using a Biosearch Sam One automated nucleotide synthesizer. Determination of Amino Acid and Nucleotide Homologies-Two databases, the protein identification resource from National Institutes of Health and Genbank from Bolt, Beranek, and Newman were searched to identify entries showing homology with CRP at the amino acid and nucleotide levels, respectively. RESULTS

Nucleotide and Derived Amino Acid Sequence of cDNATwo cDNA clones (pCRP1, pCRP5) were sequenced by both the chemical cleavage and the dideoxy chaintermination methods according to thestrategy shown in Fig. 1.The largest

13384

-

Determination of Human C-reactive Protein mRNA and Gene Structure ”

t

L

-

”

/ / //

RI

i c

”

I I

c” ” ”

1

KK

RI

KK

RI I

pCRPS

I

1’ I ,.... .....

.........

+

” 4

pCRPl

KK

Y

I

13385

-

“

a, f

CRPgZ

c ”

” t c-

FIG. 1. Sequence strategy of cDNA and genomic CRP. El, leader sequence; intron; H,coding sequence; 0, untranslated regions; K, KpnI cut site; V, stop codon; R1,EcoRI cut site. The cDNA clones (pCRP1, pCRP5) are bounded by PstI restriction enzyme cut sites. Both cDNAs and genomic CRP (CRPg2) were sequenced by the chemical cleavage method of Maxam and Gilbert (16) from the restriction enzyme cut sites: PstI and EcoR1. The entire pCRPl and pCRP5inserts were subcloned into the PstI site of M13mpll. pCRP5 was also cut with KpnI and the resultant PstI-KpnI, KpnI-PstI fragments subcloned into M13mp18. The genomic clone CRPg2 was digested with EcoRI and the three resultant fragments subcloned into M13mpll. Thecentral EcoRI fragment was also cut with KpnI and theresulting Kpn-EcoRI and EcoRI-KpnI fragments were subcloned into M13mp18. These M13 clones were sequenced by the dideoxy chain termination method of Sanger (17) using unique sequence oligonucleotide primers. The genomic sequence was derived from sequencing 89% of the coding strand and63% of the anticoding strand. ThecDNA sequence was derived from sequencing 85% of the coding strand and63% of the anticoding strand of pCRP5, and 29% of the coding strand and17% of the anticoding strand of pCRP1.

clone pCRP5 (11)contains 84 base pairs of 5’ untranslated region, 54 base pairs encoding the CRP leader sequence and 618 base pairs which define an amino acid sequence identical to that previously published (9) except for an additional 19 amino acid peptide inserted after residue 64 (Fig. 3). This region is not flanked by the consensus sequences for intronexon splice junctions, indicatingthat thepeptide is an integral part of the CRPmolecule and not theconsequence of cloning an incompletely processed RNA species. A portion of the 3‘ untranslated region from the stop codon to nucleotide 1754 is defined by pCRP5. An unusual feature of this region is a TATA box beginning at nucleotide position 1571. Nucleotide and Derived Amino Acid Sequence of Genomic DNA-A genomic CRP clone (CRPg2) was isolated from a partial EcoRI library using pCRP5 insert as a hybridization probe. The size of the clone and the sequence strategy employed is shown in Fig. 1 and its structure isshown in Fig. 2. Thereis a polyadenylation signal sequence (AATAAA) starting at -322 which may define the 3‘ limit of a flanking gene. The 316 base pairsbetween this signal and thesequence for mature CRP containsa CAAT box a t -239, a TATA box a t -187 and 54 nucleotides encoding the CRPleader sequence. The nucleotides defining the 5‘ untranslated region from -138, the preCRP coding sequences, and the following 855 base pairs of 3’ untranslated region are identical to those specified by the cDNA pCRP5, except for T instead of C a t position 1122. The remaining 357 base pairs of 3’ untranslated region bounded by the polyadenylation signal AATAAA beginning a t nucleotide 2111 are contained in this clone. A single intron of 278 base pairs begins at nucleotide 8 (i.e. in the codon specifying the thirdresidue of mature CRP) and is flanked at the 5’ and 3’ ends by the exon-intron splice junction consensus sequences AGGTAAG, and CAG, respectively. It contains two unusual repetitive sequences. One is a run of 16 adenines from nucleotide 84-99, and the otheris a block of (GT)15G(GT)3from nucleotide 136-172. Determination of Amino Acid and Nucleotide HomologiesThe derived amino acid sequence of CRP was used to search

for homologies with entries other than SAP in the National Institutes of Health protein sequence database. There is28% overall homology of amino acid residues 1-25 of CRP with residues 7-31 of human apolipoprotein CII (18), residues 5175of the Australian tiger snake phospholipase A (19) and residues 22-46 of bovine and human histone H2B (20,21). Of the 8 residues from position 13 to20 of CRP and34 to 41 of human and bovine histone HZB, there are 5 strict identities and 1 neutral substitution. There are a further 3 identities and 4 neutral substitutions when the following 35 residues are considered. There is a 32% homology of residues 101-126 of CRP with residues 2-26 of chicken and residues 89-113 of Drosophila histone H2B (22) and H1(23), respectively. There is also a 32% homology of residues 180-206of CRP and residues 184-208 ofthe Drosophila heat shock protein 26 (24). There were no furthersignificant homologies, other thanwith SAP, at the nucleotide level when the CRP nucleotide sequence was compared with entries in the Genbank database. Comparison of the Nucleotide and Derived Amino Acid Sequencesof Human CRP and SAP-The nucleotide sequencing of both cDNA and genomic clones identified an additional peptide of 19 amino acids in the primary structure of human CRP notpreviously recognized. Recently, Mantzouranis et al. (8) have described the complete amino acid sequence of human SAP derived from the nucleotide sequence of cDNA clones. The previously reported protein sequence of human SAP (6) had been determined by alignment of SAP peptides based on homology with the published CRP sequence. Nucleotide sequencing of SAP cDNA clones showed that some peptides were incorrectly aligned, and others were omitted. We have compared the amended sequences of human CRP and SAP at the nucleotide and amino acid level. Fig. 3 shows an alignment of these two pentraxins. A minimal number of gaps have been introduced into both sequences in order to align regions showing marked homology overseveral residues. Overall amino acid identity is 51%, a figure which rises to 66% when neutral substitutions are considered. The level of nucleotide homology is 59%. Of the 105 amino acids showing

Determination of Human C-reactive Protein mRNA

13386

and Gene

Structure

-3z A A T A A A T A A C T C A C A T T G A T T T C T C T G G T C T G A A A T A A T T G T T G G A A A A T T A T T

-TACATAGTGGCGCAAACTCCCTTACTGCTTTGGATATAAATCCAGGCAGGAGGAGGTAGCTCTAAGGCAAGAGATCTAGGACTTCTAGCCCCTGAACTTTC 221

-1P

AGCCGAATACATCTTTTCCAAAGGAGTGAATTCAGGCCCTTGTATCACTGGCAGCAGGACGTGACC-----------------------------------

- 10

_"""""_"

f

8

AGGTAAGGGCCACCCCAGGCTATGGGAGAGTTTTGATCTGAGGTATGGGGGTGGGGTCTAAGACTGCATGAACAGTC

83

TCAAAAAAAAAAAAAAAAGACTGTATGAACAGAACAGTGGAGCATCCTTCATGGTGTGTGTGTGTGTGTGTGTGTGTGTGTGTGGTGTGTAACTGGAGAAG 184

GGGTCAGTCTGTTTCTCAATCTTAAATTCTATACGTAAGTGAGGGGATAGATCTGTGTGATCTGAGAAACCTCTCACATTTGCTTGTTTTTCTGGCTCA~

an TGAGGCCCAGCTGTGGGTCCTGAAGGTACCTCCTCCCGGTTTTTTACACCGCATGGGCCCCACGTCTCTGTCTCTGGTACCTCCCGCTTTTTTACACTGCATGG 998

TTCCCACGTCTCTGTCTCTGGGCCTTTGTTCCCCTATATGCATTGAGGCCTGCTCCACCCTCCTCAGCGCCTGAGAATGGAGGTAAAGTGTCTGGTCTGGG o lo p

AGCTCGTTAACTATGCTCGGAAA~GGTCCAAAAGAATCAGAATTTGAGGTGTTTTGTTTTCATTTT-ATTTCAAGTTGGACAGATCTTGGAGATAATTTCT 1100 TACCTCACATAGATGAGAAAACTAACACCCAGAAAGGAGAAATGATGTTATAAAAAACTCATAAGGCAAGAGCTGAGAAGGAAGCGCTGATCTTCTATTTA 1301

ATTCCCCACCCATGACCCCCAGAAAGCAGGAGCATTGCCCACATTCACAGGGCTCTTCAGTATCAGAATCAGGACACTGGCCAGGTGTCTGGTTTGGGTCC 1402

AGAGTGCTCATCATCATGTCATAGAACTGCTGGGCCCAGGTCTCCTGAAATGGGAAGCCCAGCAATACCACGCAGTCCCTCCACTTTCTCAAAGCACACTG 1505

GAAAGGCCATTAGAATTGCCCCAGCA~AGCAGATCTGCTTTTTTTCCAGAGCAAAATGAAGCACTAGGTATAAATATGTTGTTACTGCCAAGAACTTAAAT 1604

GACTGGTTTTTGTTTGCTTGCAGTGCTTTCTTAATTTTATGGCTCTTCTGGGAAACTCCTCCCCTTTTCCACACGAACCTTGTGGGGCTGTGAATTCTTTC 1705

TTCATCCCCGCATTCCCAATATACCCAGGCCACAAGAGTGGACGTGAACCACAGGGTGTCCTGTCAGAGGAGCCCATCTCCCATCTCCCCAGCTCCCTATC i806

TGGAGGATAGTTGGATAGGTACGTGTTCCTAGCAGGACCAACTACAGTCTTCCCAAGGATTGAGTTATGGACTTTGGGAGTGAGACATCTTCTTGCTGCTG 1907

GATTTCCAAGCTGAGAGGACGTGAACCTGGGACCACCAGTAGCCATCTTGTTTGCCACATGGAGAGAGACTGTGAGGACAGAAGCCAAACTGGAAGTGGAG 2006

GAGCCAAGGGATTGACAAACAACAGAGCCTTGACCACGTGGAGTCTCTGAATCAGCCTTGTCTGGAACCAGATCTACACCTGGACTGCCCAGGTCTATAAG 2io9

CCAATAAA

FIG. 2. Nucleotide sequence of the CRP gene. The nucleotides common to both cDNA (pCRP5) and specified by pCRP5instead of T genomic (CRPg2) clones are in italics. *, a t position 1122, aCis -, leader sequence as sxown in Fig. 3;+, coding sequence as shown in Fig. 3. The CAAT box, TATA (TATAAT) boxes, polyadenylation signals (AATAAA), and the 5' (at positions 8 and 920) and 3' (at positions 283 and 1055) consensus splice junctions are underlined. ""

strict conservation, 62 have the same codon usage, and single base changes account for 11of the 31 neutral substitutions. DISCUSSION

In this report we present the complete cDNA sequence of human CRP, including a 57 nucleotide long region corresponding to a previously unreported 19 amino acid peptide following position 64 of the published protein sequence. In several recent studies, human CRP has been observed to migrate "anomalously" on polyacrylamide gels (4, l l ) , with an apparent M , of 24,000, about 1500 larger than that calculated from the published amino acid sequence (9). The size of the extra peptide described above accounts for this discrepancy in apparent molecular weight. A CRP genomic clone was completely sequenced. The region 5' to the coding sequence contains a polyadenylation signal (AATAAA) beginning at nucleotide -322, which probably defines the 3' limit of a flanking gene. A CAAT box at -239 and a TATA box at -187 define the probable initiation and attachment sites for RNA polymerase 11. Of particular interest in the 5' portion of the CRP gene are three regions

similar to the Drosophila heat shock consensus sequence CTnGAAnnTTCnAG (25). These regions lie between the TATA box and thecodon specifying the pre-CRP N-terminal methionine, i.e. between nucleotides -146 and -134 (CTagGACTTCtAG), between nucleotides -130 and -119 (CTGAActTTCAG), and between nucleotides -95 and -86 (GTGAATTCAG). In Drosophila, the presence of one, or more, heat shock consensus sequences in the5' regions of the heat shock genes is necessary for heat-induced chromosome puffing and the coincident increase i q transcription of the heat shock gene products (25). The "consensus elements" in the 5' region of the CRP gene may have analogous functions leading to enhancement of CRP synthesis inthe acute phase response. A single intron of 278 base pairs begins at nucleotide 8 following the codons specifying the firsttwo amino acids of the mature CRP. Its position is similar to that observed for introns at the junctions of the leader sequence and mature product coding regions of mouse X1 immunoglobulin genes (26). It contains two unusual repetitive elements: a run of 16 adenines from nucleotide 84 to 99 and a block of (GT)16G(GT)Sfrom nucleotide 136 to 172. Sections of re-

Determination of Human C-reactive Protein

mRNA and Gene

Structure

13387

FIG. 3. Comparison of nucleotide and derived amino acid sequences of CRP andSAP. Only nucleotides in the SAPcDNA that aredifferent from CRP are shown. 0,strict amino acid homology; i:i, neutral substitutions as defined by Dayhoff et al. (36), i.e. (A,P,G), (N,Q), (D,E), (S,T), (C), (V,I,M,L,), (K,R,H), and (F,Y,W).

peated As have previously been observed in some 3' untranslated regions (for example, human atrial natriuretic factor precursor (27)), however, their significance is unknown. Blocks of purine-pyrimidine repeats ((GT)n or (CA)n) are found in several regions of the genome: the introns of the human y-globin gene (28), mouse immunoglobulin CHand VH genes, and human actin gene; the 3' noncoding region of mouse immunoglobulin K U-V gene and in thehuman intergenic a-globin gene cluster (reviewed in Ref. 29). Hybridization with (CA)n and (GT)n probes has revealed that such sequences are present in the genomes of mammals, frogs, slime mold, and yeast (29). (GT)n sequences can adopt the left-handed Z-DNA form which may participate in the activation of chromatin domains (29,30). Both(GT)n and(CA)n are associated with sites of recombination: in man, (GT)n sequences mark the boundaries ofgene conversion events between globin genes (28), and theboundaries of an a-globin gene duplication (31). Furthermore, Stringer (32) has demonstrateda recombinational event a t a(CA)n/(GT)n sequence associated with the insertion of simian virus 40 into

the rat genome. The function of the (GT),,G(GT), sequence in theCRP intron is unknown. In the 3' untranslated region, there are two unusual features. There are 5' and 3' consensus splice junctions beginning at nucleotides 920 (AGGT) 1055 (Py,CAG), respectively, in both the genomic and cDNA sequences. There isno experimental evidence for two different CRP mRNA species in human acute phaseliver (11).It is nevertheless possible that there is a poly mRNA for CRP and that different mRNA processing occurs in tissues other than liver, and/or in nonacutephase liver. Alternatively, thesejunctional sequences may be nonfunctional. The second unusual finding is a TATA box beginning at nucleotide 1571, suggesting that sequences downstream may be transcribed. However, the derived amino acid sequences in all 3 reading frames downstream from this TATA box are not open. No homology with other known proteins was found when this downstream region was compared with entries in both the Genbank and the National Institutes of Health databases. Therefore we consider it likely thatthis TATA box is nonfunctional. At position 1122,

13388

Determination of Human C-reactiue Protein mRNA and Gene Structure

CRPg2 specifies a T whereas pCRP5 specifies a C. This discrepancy could arise either as a result of a mutation in the cloned DNA or due to allelic variation in the CRP gene. The search for homologies with proteins other than SAP in the amino acid and nucleotide databases has revealed 28% homology between the first25 amino acids of CRP andsnake phospholipase A and human apolipoprotein C11. Therefore this region of CRP may contain its phosphorylcholine binding site. The homologies seen between residues 1-25 and 101-126 of CRP and human histone H2B and chicken and crocodile H2B, respectively, may indicate regions involved in thebinding of CRP t o chromatin. Both the human histone H3, H4 gene cluster and human CRP have been assigned to chromosome 1 (7, 33): the histone gene cluster is mapped to band lqZl(33) andregional localization of the CRPgene willreveal whether it is part of the same gene cluster. The 32% homology Drosophila heat between residues 180 and 206 ofCRP and the shock protein 26 may indicate a common functional domain. The concentrations of both heatshock proteins and CRPcan increase 1000-fold on induction. Furthermore, the stimulus for CRP production is interleukin I’ which is a major component ineliciting fever. The comparison of the amended CRPcoding sequence with that of SAP at both the amino acid and nucleotide level shows a high degree of homology. Both leader sequences are hydrophobic and share extensive amino acid sequence homology with a wide range of signal peptides reported for other secreted products (34). The overall amino acid identity of the mature product is 51% (66% when neutral substitutions are considered). CRP and SAP are probably products of gene duplication: homologues of both proteins are found in a wide variety of species (e.g. horseshoe crab, mouse, and man) suggesting that the duplication event occurred early in evolution. This possibility is supported by the findings that only 59% of the amino acid residues showing strict conservation have the same codon usage and only 35% of neutral substitutions can be accounted for by single base changes. The overall nucleotide homology is 59%. However, the level of homology between CRP and SAP is not constant throughout their sequences. From residue 1 to 66 of CRP (1-64 of SAP) amino acid conservation is 53% (70% if neutral substitutions areconsidered), from residue 67 to 91 of CRP (65-89 of SAP) conservation is only 20% (36% if neutral substitutions are considered), and from residue 92 to 206 of CRP (90-205 of SAP) conservation is 57% (71% if neutral substitutions are considered). The degree of nucleotide conservation in these three regions is 62%, 33%,and 69%, respectively. These strikingly different levels of homology between the different regions of these proteinsmay reflect the portions of the molecules giving rise to similar and dissimilar physiological functions: that is, the peptides from residues 67 to 91 of CRP and 65 to 89 of SAP may have specialized functions, and themore conserved flanking regions may represent protein structures with functions sharedbetween CRP andSAP. These threeregions may represent distinct domains. However, the CRP gene does not contain corresponding separate exons as are often found in T,-Y. Liu, G. Goldberger; personal communications.

the gene structure of other proteins with distinct domains. CRP also shares 50% amino acid and 60% nucleotide homology with the Syrian hamster “female protein” (35), another acute phase pentraxin whose reponse is further modulated by sex steroids. Female protein shares the property of calcium-dependent binding to phosphorylcholine with CRP. However, female protein has a greater overall homology with human SAP a t both the amino acid (72%) and thenucleotide level (77%). The comparative analysis of the genomic structures of human CRP, SAP, and hamster female protein will provide insights into the differences in theregulation of gene expression during the acute phaseresponse. Acknowledgments-We thank Professor H. R. Colten and Dr. S. B. Dowton for help and advice on this work, Dr. Gary Gray and Dr. Tanya Falbel for assistance with computing, Ari D. Baron for technical assistance and Helen Hourihan for secretarial assistance. REFERENCES 1. Tillet, W. S., and Francis, T., Jr. (1930) J. Exp. Med. 5 2 , 561-571

2. Kushner, I., Volanakis, J. E., and Genvurz, H. (1982) Ann. N. Y.Acad. Sei. 389,235-274 3. Robey, F. A., Jones, K. D., Tanaka, T., andLlu, T.-Y. (1984) J. Biol. Chem. 259,7311-7316 4. Baltz, M.L., DeBeer, F.C., Feinstein, A., Munn, E. A., Milstein, C. P., Fletcher, T. C., March, J. F., Bruton, C., Clamp, J. R Davies, A. J. S., and Pepys, M. B. (1982) Ann. N. Y. Acad. Sci. 389,4$-75 5. Pepys, M. B., and Baltz, M. L. (1983) Adu. Immunol. 34,141-211 6. Anderson, J. K., and Mole, J. E. (1982) Ann. N. Y. Acud. Sei. 3 8 9 , 216234 7. Whitehead, A. S., Bruns,, G. A. P., Markham, A. F., Colten, H. R., and Woods, D. E. (1983) Sccence 221,69-71 8, Mantzouranis, E. C. Dowton, S. B., Whitehead, A. S., Edge, M. D., Bruns, G. A. P., and Colt&, H. R. (1985) J. Biol. Chem. 260,7752-7756 9. Oliveira, E. B., Gotschlich, E. C., and Liu, T. (1979) J. Biol. Chem. 254, 489-502 10. Woods, D. E., Markham, A. F., Rlcker, A. T., Goldberger, G., and Colten, H. R. (1982) Proc. Natl. Acad. Sci. U. S. A. 79,5661-5665 11. Tucci, A., Goldberger, G., Whitehead, A. S., Kay, R. M., Woods, D. E., and Colten, H. R. (19833 J. Immunol. 131,2416-2419 12. Maniatis, T., Hardison, R. C., Lacy, E., Lauer, J., O’Connell, C., Quon, D., Sim, G. K., and Efstratiadis, A. (1978) Cell 1 5 , 687-701 13. Benton, M. D., and Davis, R. W. (1977) Science 196,180 14. Jeffreys, A. J., and Flavell, R. A. (1977) Cell 12, 429-439 15. Rigby, P. W. J., Dieckmann, M., Rhodes, C., and Berg, P. (1977) J. Mol. Biol. 113, 237-251 16. Maxam, A. M., and Gilbert, W. (1980) Methods Enzymol. 65,499-560. 17. Sanger, F., Nicklen, S., and Coulsen, A. R. (1977) Proc. Natl. Acad. Scr. U. S. A. 74,5463-5467 18. Jackson, R. L., Baker H. N., Gllliam E. B., and Gotto, A. M. (1977) Proc. Natl. Acud. Sci. U. A. 74,1942-1945 19. Halpert, J., and Eaker, D. (1975) J. Biol. Chem. 250,6990-6997 20. Iwai, K., Hayashi, H., and Ishikawa, K. (1972) J. Biochem. (Tokyo) 72, 357-367 21. Ohe, Y., Hayashi, H., and Iwai, K. (1979) J. Biochem. (Tokyo)85,615-624 22. Van Helden, P., Strickland, W. N., Brandt, W. F., and Von Holt, C. (1978) Biochem. Biophys. Acta 533,278-281 23. Goldberg, M.L. (1979) Ph.D. Thesis, Stanford University 24. Ingolia, T. D., and Craig, E. A. (1982) Proc. Natl. Acud.Sci. U. S. A. 79,2360-2364 25. Simon, J. A., Sutton, C. A., Lobell, R. B., Glaser, R. L., and Lis, J. T. (1985) Cell 40,805-817 26. Bernard, O., Hozumi, N., and Tonegawa, S. (1978) Cell 15,1133-1144 27. Greenbera. B. D.. Bencen. G. H., Seilhamer.. J. J... Lewicki,. J. A.,. and Fiddes, J. C. (1384) Niture 312,656-658 28. Slightom, J. L., Blechl, A. E., and Smithies, 0. (1980) Cell 21,627-638 29. Rogers J. (1983) Nature 305,101-102 30. Nordhiim, A,, and Rich, A. (1983) Proc. Natl. Acud. Sci. U. S. A. 80,1821-

h.

1 R9K

31. Proi%oot, N., and Maniatis, T. (1980) Cell 21,537-544 32. Stringer, J. R. (1982) Nature 296,363-366 33. Green, L., van Antwerpen, R., Stein, J.,.Stein, G., Tripputi, P., Emanuel, B Selden, J., and Croce, C . (1984) Sezence 226,838-840 34. Von’Heijne, G. (1983) Eur. J. Biochem. 133,17-21 35. Dowton, S. B., Woods, D.E., Mantzouranis, E. C., and Colten, H. R. (1985) Science 228,1206-1208 36. Da hoff, M. O., Eck, R. V., and Park, C. M. (1972) in Atlas of Protein sequeme and Structure(Dayhoff, M. O.! ed) Vol. 3, pp. 89-99, National Biomedical Research Foundation, Washmgton, D. C.