Degenerate Oligonucleotide Sequence

Downloaded from genome.cshlp.org on October 1, 2021 - Published by Cold Spring Harbor Laboratory Press

Degenerate Oligonucleotide Sequence- directed Cross-species PCR Cloning of the BCP 54/ALDH 3 cDNA: Priming from Inverted Repeats and Formation of Tandem Primer Arrays David L. Cooper and Edward W. Baptist

The Department of Pathology, Duke University Medical Center, Durham, North Carolina 27710

Bovine corneal protein 54 (BCP 54) these was a 552-bp product occur- PCR (13) to generate the first reported is the major soluble protein of the ring at elevated primer concentra- cDNA probe to BCP 54. (6) The BCP 54 bovine cornea, and immunoreactive tions that formed through bidirec- cDNA probe generated by this mixed forms of this protein have been de- tional amplification from a single oligonucleotide primed amplification scribed in a wide range of mam- DOS annealing to an inverted of cDNA (MOPAC) technique C8,9) was mals. Dideoxy sequence determina- repeat located In the BCP 54 coding cloned and dideoxy sequenced. A Gen- tion of a previously synthesized sequence. The second artifactual Bank library search (version 63.0) 420-bp cDNA to BCP 54 generated product was a 212-bp sequence revealed a strong similarity to the pre- by the novel mixed oligonucleotide that contained several unreported viously cloned cDNA of rat liver (class primer amplification of cDNA amplification anomalies, including 3) tumor-associated aldehyde dehydro- (MOPAC) procedure revealed ex- the formation of a tandem primer genase (RATALD). (12) Nucleotide and tensive similarity to the cDNA en- array. amino acid sequence alignment of the coding tumor-associated rat liver BCP 54 translation product revealed it (class 3) aldehyde dehydrogenase as 78% and 84% identical with (RATALD). PCR amplification with RATALD at the nucleotide and amino additional pairs of degenerate Except for the most abundant soluble acid levels, respectively. Conservation oligonucleotide sequence (DOS) protein of the bovine cornea (BCP 54, of amino acid elements common to primers derived from both BCP 54- molecular weight 54,000 daltons), (1,2) the ALDH supergene family thought to amino-acid sequence and amino little attention has been paid to the be of structural/functional signficance acid and nucleotide sequence data water-soluble structural proteins of the was further substantiated by this analy- from RATALD produced three PCR cornea. Although the initial descrip- sis. products that were cloned and sub- tion of the major soluble corneal Degenerate oligonucleotide se- sequently sequenced. The major protein can be traced back almost 30 quence (DOS) and PCR have been pre- product was 716-bp BCP 54 cDNA years, (3,4) BCP 54 had undergone, until viously utilized successfully to clone clone encompassing the BCP 54 recently (s,6) only minimal molecular cDNAs for specific proteins. This ap- carboxy-terminal amino acid se- characterization with no known func- proach relies on the synthesis of quence for which the DOS pair was tion ascribed. We have previously util- oligonucleotides of limited degeneracy designed. Sequence alignment of ized mixed oligonucleotide primers from reverse translation of protein sub- the BCP 54 cDNA and its transla- complementary to the reverse transla- sequences. The cDNAs so generated tion product with RATALD demon- tion products of amino acid sequence have been used as probes for Northern strated 81% and 85% Identity at obtained from Staphylococcus aureus and Southern blotting and for screen- the nucleotide and amino acid V8-digested BCP 54 fragments (s) and ing cDNA and genomic recombinant levels, respectively. Analysis of the DNA libraries. (8,9) DOS primers have additional two clones established Present address: Department of Pathology, University also been used to amplify unknown that they were the results of PCR of Pittsburgh School of Medicine, Pittsburgh, members of gene families (1~ and artifactual processes. The first of Pennsylvania 15261. homologous genes in different

1:57-629 by Cold Spring Harbor Laboratory Press ISSN 1054-9803/91 $3.00 PCR Methods and Applications 57 Downloaded from genome.cshlp.org on October 1, 2021 - Published by Cold Spring Harbor Laboratory Press

species. (11) These applications of DOS incubation at 72~ followed to insure BCP 54 cDNA that encompasses the 3' have become common, but little has completion of all polymerization prod- most terminal protein encoding region been reported concerning the composi- ucts. Aliquots (10 ~l) of reaction mix- as well as two minor products that tion and nature of inappropriate prod- tures were analyzed by electrophoresis were generated during amplification. ucts generated by PCR with such through 1.2% agarose gels. Following Our strategy utilizes single-stranded primers. In this report we communi- with ethidium bromide and photo- (ss) bovine corneal epithelial cDNA as cate not only the successful genera- graphed with UV illumination. template for PCR amplification primed tion, again by the MOPAC technique, A second set of experiments by two DOS (Fig. 1) based on pre- of a second extended 716-bp BCP 54 employed identical reaction condi- viously determined BCP 54 and RAT- cDNA that encompasses the 3' most tions, except that 400 ng each of ALD amino acid and nucleotide se- terminal protein-encoding region, but primers DOS#1 and DOS#2 or DOS#1 quence. (s,12) The 5' sense primer additionally discuss the nature of two alone were used. This reproducibly (DOS#l) and an internal reverse primer PCR-generated artifacts associated with generated the 552-bp band that we had (DOS#2) were previously designed (6) by the utilization of DOS as PCR primers. previously observed. (6) reverse translation of amino acid sequence determined from a BCP 54 V8 METHODS AND MATERIALS Cloning of PCR Products protease fragment (Fig. 2A). The 3' antisense primer (DOS#3) was Materials Reaction mixtures with both the 716- bp and 552-bp bands were extracted designed (16) from the 3' nucleotide PCR primers DOS#1 (AACCCYCAYTA- with CHC13 after most of the mineral and amino acid sequence of the rat YGTGGACAARGA) and DOS#2 (GAAC- oil was removed by pipetting. The homolog RATALD. Based on dinucleo- TGRATRGCYTCYTC) were designed as DNA was precipitated with ethanol, tide frequency utilization in human previously described C6) by reverse trans- and the pellet was dissolved in 10 ~l of proteins, (17) the 3' sequence syn- lation of amino-terminal amino acid 10 mM Tris-HCl (pH 8.5), 1 mM EDTA. thesized consisted of the singular sequences of BCP 54 peptides produced After restriction enzyme digestion methionine codon, the most frequent- by partial digestion with Staphylococcus with EcoRI, the PCR product was ly utilized proline codon, the two most aureus V8 protease. Primer DOS#3 (GT- ligated into plasmid pBS- which had frequent codons of asparagine and GYCKGGGCATRTTRGC) was similarly been previously digested with EcoRI alanine, and four of six possible prepared from reverse translation of and treated with bacterial alkaline codons of arginine. We also designed a the carboxy-terminal 6 amino acids of phosphatase. Transformation of com- fourth DOS primer to the 5 ' terminus the RATALD sequence ~12/ (Fig. 2A). petent Escherichia coli strain NM522 (is) of RATALD and paired it with our suc- Each degenerate oligonucleotide also with the ligation reaction mixture cessful DOS#2 and DOS#3 as well as a contained an 8-nucleotide linker on its resulted in a number of white colonies number of other internal BCP 54 5' terminus that included a 6-nucleo- on ampicillin (100 mg/ml) IPTG (0.5 cDNA-based nondegenerate synthetic tide EcoRI restriction site (GGGAA- mM) XGAL (40 mg/ml) LB solid me- oligonucleotide primers. These at- TTC). Restriction enzymes and Taq dium. Twelve of these were picked, tempts failed to amplify the 5' DNA polymerase were purchased from restreaked, and analyzed by agarose gel terminus. We believe this may indicate Promega (Madison, WI), and T4 DNA electrophoresis of alkaline lysis DNA the incomplete nature of 5' ends con- ligase and bacterial alkaline phospha- minipreparations digested with EcoRI. tained in our ss cDNA preparation used tase from GIBCO BRL/Life Tech- in this work, and not necessarily an nologies (Gaithersburg, MD). Plasmid Sequence Analysis alternative 5 ' bovine sequence. pBS- was obtained from Stratagene (La Plasmid DNA from colonies containing Plasmid DNA from clone #6 which Jolla, CA). recombinant plasmids was used as contained a PCR-generated insert of template in double-stranded dideoxy approximately 720 bp (Fig. 2B) was PCR Amplification sequencing reactions with T7 DNA used as template in double-stranded The first strand cDNA synthesized from polymerase kits from either Pharmacia dideoxy sequencing reactions bovine corneal epithelium mRNA was or US Biochemical. By sequencing from (Pharmacia or US Biochemical). By se- used directly in 50-~1 PCR reac- primer sites on both sides of the pBS- quencing from T3/T7 promoter sites tions ~13,14J which contained 0.2 mM of polylinker, and the generation of inter- on both sides of the pBS- polylinker, each dNTP, 200 ng each of primers nal deoxyoligonucleotide primers and utilizing two internal synthetic DOS#1 and either DOS#2 or DOS#3, homologous to the cDNA sequence, oligonucleotides made to previous BCP and 2 units of Taq DNA polymerase in the full sequence of the PCR cDNA was 54 cDNA sequence, the full sequence a buffer of 10 mM Tris-HC1 (pH 9.0 at obtained. To control for Taq DNA of the PCR cDNA was obtained (Fig. 3). 25~ mM KC1/1.5 mM MgC12/0.1% polymerase misincorporations, DNA The cDNA contains 716 nucleotides, gelatin/0.01% Triton X-IO0. Reactions from 10 independent recombinant iso- or, when translated, 239 amino acids. were overlaid with 50 ~l of mineral oil lates identical to clone #6 as described Sequence alignment of the BCP 54 to prevent evaporation and heated at in the text were combined and se- cDNA is colinear with the RATALD se- 94~ for 4 min to denature initially quenced as above. quence without any insertions or dele- the template cDNA. This was followed tions. At the nucleotide level, these se- by 35 cycles of 1 min at 94~ 1 min at RESULTS quences are 81% identical compared 55~ and 2 min at 72~ A final 5-min In this report we describe an extended with an 85% identity when translated.

58 PCR Methods and Applications Downloaded from genome.cshlp.org on October 1, 2021 - Published by Cold Spring Harbor Laboratory Press

SS BCP54 cDNA maintenance of tertiary structure and , 5' 3' D._gOS#1 , hints that the tertiary structures DOS#3 DOS#1 formed by the corresponding regions of family members of the ALDH super- 1 Polymerase 1 gene are similar. chain reaction amplification Sequencing of clone #3, which contained a 552-bp insert, from the T7 promoter primer revealed complete 716-bp Amplified cDNA 552-bp identity with the 5' end of our pre- product viously reported BCP 54 cDNA sequence. (6) However, sequencing from RI EcoRI ~ the T3 promoter primer yielded one of the 16 possible sequences of DOS#1 followed by 129 nucleotides of BCP 54 Ligate sequence, of which 31 overlap with our previously reported sequence and the remainder of the nucleotides identical to the appropriate region of Transform with bacteria and the clone #6 (716 bp) cDNA sequence screen by XGal color selection reported above. Comparison of the DNA sequence of clone #6 with DOS#1 degenerate sequence of the cloned iso- Dideoxy DNA sequencing of clone late, since this clone represents only FIGURE 1 Strategy for the PCR-based amplification cloning of BCP 54 cDNA utilizing one recombinant isolate, demonstrates degenerate oligonucleotide sequences as primers. First-strand cDNA synthesis was ac- that the apparent deoxyoligonucleo- complished by AMV reverse transcription of oligo(dT)-primed poly(A) + rnRNA isolated from tide used to prime the amplification of bovine corneal epithelial cells (Fast Track, Invitrogen). this product, AACCCTCACTATGTG- GACAAAGA, anneals to the BCP 54 Apparent once more is the striking I DEHUE1 and RATALD in correspond- nucleotides corresponding to RATALD conservation of glycine residues, as the ing positions. (6,7) The overrepresenta- nucleotides 1345 to 1357. The 3' 5 additional glycines (Gly-403, -404, tion of glycines among residues con- terminus of this oligonucleotide ex- -409, -411, -415) present in this ex- served between distantly related hibits 12 matches and only 1 mis- tended BCP 54 sequence are present in I proteins has been correlated with the matched pair of bases (Fig. 3), allowing

Sense Primer: DOS#1

BCP54 amino acid sequence from 33kDa S5aphylococcus aureus V8 fragment

I ggknphyvdkdrd Idi ac rr i awgk fm

n p h y v d k d 5'-GGGAATTC AAC CC(C/T) CA(C/T) TA(C/T) GTG GAC AA(A/G) GA -3' EcoRI

Antisense primer: DOS#3

Tumor-inducible (class 3) rat aldehyde dehydrogenase amino acid and nucleotide sequence

45O aa h r p m n a nt gtg ccg ggg cat gtt ggc 1530 1520 5'-GGGAATTC GTG (C/T)C(T/G) GGG CAT (G/A)TT (G/A)GC -3' EcoR I I FIGURE 2 Selection of DOS primers, and product analysis in the PCR generation of BCP 54 cDNA. Amino acid sequence is represented by one- letter lowercase code and nucleotide sequence is represented by one-letter uppercase code. (A) The amino sequence (underlined) from the 33- kD S. aureus V8 generated fragment of BCP 54 was used to derive a 5' sense DOS#1 (bold uppercase letter, region of limited nucleotide degeneracy [N/N] indicated) by reverse translation. The antisense DOS#3 was derived from BCP 54 amino acid and RATALD amino acid and nucleotide 3 ' sequence. The RATALD amino acid and nucleotide numbering are above and below the sequences utilized, respectively. (B) Gel electrophoresis analysis of BCP 54 cDNA PCR products utilizing the DOS#l/DOS#3 and DOS# l/DOS#2 primer pairs. (i) (Lane 1) 1-kb DNA ladder (GIBCO/BRL); (lane 2) production of a single BCP 54 cDNA PCR fragment of 716 bp primed by DOS#I and DOS#3. (ii) (Lane 1) 1-kb DNA ladder (GIBCO/BRL); (lane 2) optimized production of 420-bp BCP 54 cDNA primed by DOS#I/I]~DS#2.C6) (Lane 3) Nonoptimized production of 420-bp BCP cDNA (primed by DOS#I/DOS#2) and 552-bp BCP cDNA (primed by DOS#1 ).

PCR Methods and Applications 59 Downloaded from genome.cshlp.org on October 1, 2021 - Published by Cold Spring Harbor Laboratory Press

BCP54 nucleotide 42, there is another primer N P H Y V D K D R D L D I A C R R I A aoa#1AAC CCY CAY TAY GTG GAC AAR GA-4~ dimer of DOS#1 and DOS#2 (nucleo- AAC CCT CAC TAC GTG GAT AAA GAT CGC GAC CTG GAC ATT GCC TGC AGA CGC ATC GCC tides 43-80) with a 2-nucleotide over- AgC CCT tgt TAC GTG GAc AAg GAc tGt GAt CTa GAt gTg GCt TGC AGg CGt ATC GCC 872 215 s P c Y V D K D c D L D v A C R R I A lap. However, at positions 79-81, there RATALD is a 3-nucleotide overlap between the W G K F M N S G Q T C V A P D Y I L C TGG GGA AAG TTC ATG AAC AGT GGC CAG ACC TGC GTG GCC CCC GAC TAC ATC TTG TGT 5' end of primer DOS#2 and the TFC TGG GGg AAa TTt ATG AAC AGT GGC CAG ACC TGt GTG GCC CCa GAC TAC ATC cTc TGT 929 234 W G K F M N S G Q T C V A P D Y I L C of the EcoRI linker on DOS#1. The

D P S I Q S Q V V E K L K K S L K E F nonambiguous G at position 21 of GAC CCC TCT ATC CAG AGC CAA GTC GTG GAG AAG CTC AAA AAG TCC CTG AAA GAG TTC DOS#1 has undergone a transition to A GAC CCC TCg ATt CAG AaC CAA aTC GTG GAG AAG CTC AAg AAG TCa CTc AAA GAt TTC 986 253 D P S I Q n Q i V E K L K K S L K d F at nucleotide 94 of our sequence.

Y G E D A K K S R D Y G R I I N S R H DOS#1 also overlaps with primer TAC GGG GAG GAC GCC AAG AAG TCC AGG GAC TAC GGG AGA ATC ATC AAC TCC CGG CAC TAt GGG GAa GAt GCt AAG cAG TCC cGt GAt TAt GGG AGg ATC ATC AAt gaC CGt CAC 1043 DOS#2 in a 2-bp common sequence 272 Y G E D A K q S R D Y G R I I N d R H (nucleotides 103 and 104). At positions F Q R V M G L L E G Q K V A Y G G T G 121-122 and 141-142 are insertions of TTC CAG AGG GTG ATG GGC CTG CTG GAG GGC CAG AAG GTC GCC TAC GGA GGC ACC GGG TTC CAG cGG GTc Aaa GGC CTG aTt GAc aaC CAG AAa GTa GCC cAt GGA GGC ACt tGG ii00 AC dinucleotides between tandem 291 F Q R V k G L i d n Q K V A H G G T w repeats of the DOS#2 primer. The last D A T T R Y I A P T I L T D V D P E S GAC GCG ACC ACC CGC TAC ATC GCC CCC ACC ATC CTT ACG GAC GTG GAC CCC GAG TCC of these DOS#2 repeats overlaps in its GAC caG tCC tCa CGa TAC ATa GCt CCa ACC ATC CTg gtG GAt GTG GAC CCC cAG TCC 1157 EcoRI linker with the EcoRI linker of a 310 D q s s R Y I A P T I L v D V D P q S DOS#1 primer. This linker sequence is P V M Q E E V F G P V L P I M C V R S CCG GTG ATG CAG GAG GAG GTC TTC GGG CCC GTC CTG CCC ATC ATG TGT GTG CGC AGC interrupted by the insertion of a T-A CCa GTG ATG CAG GAG GAG aTC TTt GGG CCg GTg aTG CCC ATt gTG TGT GTt CGa AGC 1214 329 P V M Q E E i F O P V m P I v C V R S base pair (nucleotide 165) in the TT dinucleotide of the EcoRI site. The final L E E A I Q F I T Q R E K P L A L Y V CTG GAG GAA GCC ATC CAG TTC ATC ACC CAG CGC GAG AAG CCC CTG GCG CTC TAC GTC 46 nucleotides are again a classical tTG GAG GAg GCC ATt CAa TTt ATC AaC CAG CGt GAG AAG CCC CTG GCa CTC TAt GTg 1271 348 L E E A I Q F I n Q R E K P L A L Y V primer dimer with 2 bp overlap at

F S P N D K V I K K M I A E T S S G G nucleotides 189-190. TTC TCA CCG AAC GAC AAG GTG ATT AAG AAG ATG ATC GCG GAG ACC TCC AGC GGC GGG The type of primer dimers rep- TTC TCc aac AAt GAg AAG GTG ATc AAG AAa ATG ATC GCa GAG ACa TCC AGC GGt GGa 1328 367 F S n N e K V I K K M I A E T S S G G resented by nucleotides 43-81 and

V T A N D V V V H I S V H S L P Y G G nucleotides 160-212 has been pre- GTG ACA GCC AAC GAC GTC GTT GTC CAC ATC AGC GTG CAC TCG CTG CCC TAC GGG GGT GTG ACA GCC AAt GAC GTC aTT GTt CAC ATC Act GTG CcC act tTG CCC Ttt GGt GGT 1385 viously described.OS) We report here -~-AG rAA CAG GTG TAt Tac CcC caa dos#1 two other types of artifactual sequence 386 V T A N D V i V H I t V -p t L P f G G resulting from primer overlaps. The V G D S G M G S Y H G R K S F E T P S GTG GGG GAC AGC GGC ATG GGG TCC TAC CAT GGC AGG AAG AGC TTC GAG ACC TTC TCG first of these occurs at nucleotides GTG GGG aAC AGC GGC ATG GGG gCC TAC CAT GGC AaG AAa AGC TTC GAG ACC TTC TCc 1442 405 V G n S G M G a Y H G k K S F E T F S 26-29 where the 3' pentanucleotide AAAGA of DOS#1 overlaps the 3' H R R S C L V R P L L N E E T L K A R CAC CGC CGC TCC TGC CTG GTG AGG CCT CTG CTG AAC GAA GAG ACC CTC AAA GCC AGA pentanucleotide TCTTC of DOS#2. CAC CGC CGC TCt TGC CTG GTG AaG tCT CTG CTG AAt GAA GAa gCt CaC AAg GCC AGg 1499 424 H R R S C L V k s L L N E E a h K A R However the 3' C from DOS#2 is

Y P A/P* S P/V* A N M P R H deleted and a T is present to pair with TA- CCC GCG AGC CCG GCC AAT ATG CCC CGA CAC 81~ nt identity TAT CCC cCa AGC CCa GCC AAg ATG CCC CGg CAC 1532 the A of DOS#1. It is unlikely that the -~-CGR TTR TAC GGG KCY GTG dos#3 lack of a C is due to an error in 443 Y P P S P A K M P R H 85~ aa identity oligonucleotide synthesis because this FIGURE 3 Sequence alignment of BCP 54 (top sequence) and RATALD (boSom sequence). ]he is the nucleotide that is directly at- conventional one-letter nucleotide and amino acid codes are employed. Uppercase type indi- tached to the column matrix. The cates amino acid or nucleotide identity between BCP 54 and RATALD. DOS#1 and DOS#3 are other type of overlap is at nucleotides italicized where they bound to the BCP 54 cDNA to generate clones #3 (552 bp) and #6 (716 79-81 and at nucleotides 161-166. In bp), respectively. Arrowheads indicate the direction of priming. Translation of the eDNA se- both of these, the overlap is between quence differs from amino acid sequence determination at two positions (*). 5' primer sequences which implies that ligation is necessary for unbroken the BCP 54 sequence in this region to (tandem array) of primer DOS#2 sepa- strands. function as an inverted repeat and rated by AC dinucleotides with primer serve as a second site of DOS#1 bind- dimer (tail-to-tail) derivatives at each DISCUSSION ing so that PCR amplification could end (Fig. 4). This study represents our continued ef- and did occur with DOS#1 as the sole An earlier report (17) described addi- fort to characterize the cDNA encoding primer. tion of a single dAMP nucleotide to the the BCP 54 gene product. By use of the We also found that clone #12 con- 3' end of a blunt-ended DNA duplex. PCR and a pair of degenerate oligonu- tained an unexpected small insert of This can be seen at nucleotide 42 of cleotide sequence primers derived from 212 nucleotides. When this was se- our sequence. Nucleotides 1-41 both BCP 54 amino acid sequence and quenced, it was obvious that this was a originate in a dimer of primers DOS#1 sequence data, both amino acid and previously unreported type of PCR- and DOS#2 with a 4-nucleotide overlap nucleotide, from the homologous generated artifact, consisting of a and loss of the 3'-most nucleotide of cDNA encoding the tumor-associated threefold head-to-tail tandem repeat DOS#2. After the A insertion at rat liver (class 3) aldehyde dehydro-

60 PCR Methods and Applications Downloaded from genome.cshlp.org on October 1, 2021 - Published by Cold Spring Harbor Laboratory Press

I 50 that contained sequences generated dos#l dos#1 gaattcaacccycaytaygtggacaarga aacccyca partially or completely as PCR artifacts. GAATTCAACCCTCACTACGTGGACAAAGAGGCCATCCAGTq]'I~CCCTCA The larger of these, clone #3 (552 bp), CTTAAGTTGGGAGTGATGCACCTG~TCTCCGGTAGGTC~TGGGAGT is interesting because it represents an ctyctycgrtargtcaag dos#2 incorrect priming on the desired

i00 template. Theoretically, such a PCR dos#1 ytaygtggacaargg t t caacccycayt aygt ggaca product should occur whenever a set of CTATGTGGACAAGGAGGAGGCTATTCAGTTCAACC CTCACTATATGGACA degenerate primers can anneal to a GATACACCTGTTCCTCCTCCGATAAGTCAAGTTGGGAGTGATATACC TGT template DNA molecule containing in- ctyctvcgrtargtcaag dos#2 verted repeat sequences with sufficient sequence similarity to effectively prime Taq DNA polymerase activity. In prac- 150 tice, observation of this artifact was de- AGGAGGAGGCCATTCAGTT~--~GAGGAGGCCATTCAGTT~--~GAGGAGGC pendent on the relative concentrations TCC~TCCTCCGGTAAGTCAA~CTCCTCCGGTAAGTCAA~TCCTCCG of primers and template and not docu- ctyctvcgrtargtcaag ctyctvcgrtargtcaag ctyctvcg dos#2 dos#~ " dos#2 mented until doubling of the DOS primer concentration. Priming from the inverted repeat present within the dos#l gaatt caacccycaytaygtggacaar~a 200 BCP 54/RATALD protein coding se- TATTCAGTTCGAAT%IT~AACCCTCATTACGTGGACAAAGAAGAAGCTATT quence is analogous to single primer ATAAGTCAAGCTTA/~TTGGGAGTAATGCACCTGTTTC~TTCTTCGATAA PCR, e.g., inter-Alu PCR, (23) a tech- rtargtcaagcttaa g ctyctvcgrtar dos#2 nique used to amplify human DNA from complex mixtures of human and other species DNA. 212 Our third clone, #12 (212 bp), an CAGTTCGAATTC obvious product of in vitro processes, GTCAAGCTTAAG is interesting for two reasons. The first gtcaagct taag is the wide variety of tail-to-tail primer FIGURE 4 Double-stranded DNA sequence of clone #12 (212). Underlined sequences on the dimers resulting from overlap and sub- upper strand are from primer DOS#1 and those on the lower strand are from DOS#2; the sequent extension of these primers. degenerate sequences of these primers are in lowercase letters. Inserted nucleotides at The second is the formation of head- nucleotides 42, 121, 122, 141, 142, and 165 are enclosed in boxes. The A at nucleotide 94 to-tail tandem direct repeats of one that is the result of a transition from the nonambiguous G found at this position in DOS#1 is starred (*). primer separated by dinucleotide in- serts that generates a tandem primer array. genase, a second more complete 716- BCP 54/RATALD is a taxon-specific bp cDNA encompassing the BCP 54 protein limited to mammals. However, ACKNOWLEDGMENTS carboxy-terminal amino acid sequence we did detect enzymatic activity con- This work was supported in part by re- was amplified, cloned, and dideoxy se- sistent with either class 1 or class 2 search grants RO1-EY-00146 and P30- quenced. Nucleotide and amino acid ALDH family members (distinct from EY-05722 from the National Eye In- sequence alignment of the BCP 54 the class 3 forms) in two species of stitute. We thank Mr. Allan Summers translation product revealed it as 81% bony fish. It has also been reported for expert technical assistance and Drs. and 85% identical with RATALD at the that the taxon-specific crystallin (~l- Gordon Klintworth and ]oram nucleotide and amino acid levels, crystallin) of the elephant shrew (19) is a Piatigorsky for their continuing en- respectively. Conservation of amino member of the cytoplasmic (class 1) couragement of our work in the field acid sequence elements common to ALDH family, which leads us to of visual science. the ALDH supergene family thought to hypothesize that the phenomena of be of structural/functional signficance gene sharing may be more widespread REFERENCES were further substantiated by this anal- than originally thought, including the 1. Alexander, R.J., B. Silverman, and ysis. utilization of gene products from W.L. Henley. 1981. Isolation and It is apparent that in normal devel- alternative members of the same gene characterization of BCP 54, the major opmental biology, MCP/ALDH 3 is family, in different tissues across soluble protein of bovine cornea. Exp. transcribed and expressed in a highly species boundaries.(22) Consistent with Eye Res. 32: 205-216. tissue-specific manner, not typical of the hypoth-esis that gene sharing oc- 2. Silverman, B., R.J. Alexander, and other ALDH isozymes. Preliminary curs through neutral selection of one W.L. Henley. 1981. Tissue and species analysis of the major soluble corneal of a family of possible genes, corneal specificity of BCP 54, the major proteins from other taxa (vertebrate expression of BCP 54/RATALD is there- soluble protein of bovine cornea. Exp. classes: bird, bony fish, reptile, fore not based on its enzymatic ac- Eye Res. 33: 19-29. amphibia) found the class 3 ALDH ab- tivity. 3. Francois, J. and M. Rabaey. 1963. sent in these cornea,(22) indicating that We have also analyzed two clones Immunoelectrophoresis of the

I~CR Methods and Applications 61 Downloaded from genome.cshlp.org on October 1, 2021 - Published by Cold Spring Harbor Laboratory Press

proteins of the corneal epithelium. K.B. Mullis, and H.A. Erlich. 1988. Exp. Eye Res. 2: 196-202. Primer-directed enzymatic amplifica- 4. Berger, B. 1971. Immunoelectropho- tion of DNA with a thermostable resis of extracts from bovine corneal DNA polymerase. Science 239: epithelium using antisera specific to 4487-4491. individual protein fractions. Acta 14. Cooper, D.L. and N.R. Isola. 1990. Ophthalmol. 49: 685-701. Full-length cDNA utilizing the 5. Cooper, D.L., E.W. Baptist, J. Enghild, polymerase chain reaction, a H. Lee, N.R. Isola, and G.K. degenerate oligonucleotide sequence, Klintworth. 1990. Partial amino acid and a universal primer. BioTechniques sequence determination of the major 9: 60-65. soluble bovine corneal protein. Curr. 15. Dagert, M. and S.D. Ehrlich. 1979. Eye Res. 9: 781-786. Prolonged incubation in calcium 6. Cooper, D.U, E.W. Baptist, J. Enghild, chloride improves the competence of N.R. Isola, and G.K. Klintworth. 1991. Escherichia coil Gene 6: 23-28. Bovine corneal protein 54K (BCP 54) 16. Lathe, R. 1985. Synthetic oligonucle- is a homologue of the tumor- otide probes deduced from amino associated (class 3) rat aldehyde acid sequence data: Theoretical and dehydrogenase-encoding gene (RAT- practical applications. J. Mol. Biol. ALD). Gene 98: 201-207. 183: 1-12. 7. Hempel, J. and R. Lindahl. 1990. 17. Clark, J.M. 1988. Novel non- Class III aldehyde dehydrogenase templated nucleotide addition reac- from rat liver: Superfamily rela- tions catalyzed by procaryotic and tionship to classes I and II and func- eucaryotic DNA polymerases. Nucleic tional interpretations. In Enzymology Acids Res. 16: 9677-9686. of carbonyl metabolism (ed. H. Weiner 18. Watson, R. 1989. The formation of and T.G. Flynn), vol. 2, pp. 3-17. primer artifacts in polymerase chain Alan R. Liss, Inc., New York. reactions. Amplifications 1(2): 5-6. 8. Lee, C.C., X. Wu, R.A. Gibbs, R.G. 19. Wistow, G. and H. Kim. 1991. Lens Cook, D.M. Muzny, and C.T. Caskey. protein expression in mammals: 1988. Generation of cDNA probes Taxon-specificity and the recruitment directed by amino acid sequence: of crystallins. ]. Mol. Evol. 32: Cloning of urate oxidase. Science 239: 262-269. 1288-1291. 20. Piatigorsky, J., W.E. O'Brien, B.L. 9. Lee, C.C. and C.T. Caskey. 1989. Norman, K. Kalumuck, G.J. Wistow, Generation of cDNA probes by T. Borras, J.M. Nickerson, and E.F. reverse translation of amino acid Wawrousek. 1988. Gene sharing by 8- sequence. In Genetic engineering: crystallin and arginio-succinate lyase. Principles and methods (ed. J.K. Proc. Natl. Acad. Sci. 85: 3479-3483. Setlow), vol. II, pp. 159-170. Plenum, 21. Piatigorsky, J. and G.J. Wistow. 1989. New York. Enzyme/crystallins: Gene sharing as 10. Wilks, A.F., R.R. Kurban, C.M. an evolutionary strategy. Celt 57: Hovens, and S.J. Ralph. 1989. The 197-199. application of the polymerase chain 22. Cooper, D.L., E.W. Baptist, N.R. Isola, reaction to cloning members of the G.K. Klintworth, and M. Gottsman. protein tyrosine kinase family. Gene 1991. Bovine corneal protein 54K is a 85: 67-74. homologue of rat tumor-associated 11. Sommer, R. and D. Tautz. 1989. aldehyde dehydrogenase: Another Minimal homology requirements for example of gene sharing? Invest. PCR primers. Nucleic Acids Res. 17: Ophthalmol. Visual Sci. (suppl.) 32: 6749. 781. 12. Jones, D.E., M.D. Brennan, J. Hempel, 23. Nelson, D.L. and C.T. Caskey. 1989. and R. Lindahl. 1988. Cloning and Alu PCR: The use of repeat sequence complete nucleotide sequence of a primers for amplification of human full-length cDNA encoding a DNA from complex sources. In PCR catalytically functional tumor- technology (ed. H.A. Erlich), pp. associated aldehyde dehydrogenase. 113-116. Stockton Press, New York. Proc. Natl. Acad. Sci. 85: 1782-1786. 13. Saiki, R.K., D.H. Gelfand, S. Stoffel, Received May 31, 1991; accepted in S.J. Scharf, R. Higuchi, G.T. Horn, revised form June 18, 1991.

62 PCR Methods and Applications Downloaded from genome.cshlp.org on October 1, 2021 - Published by Cold Spring Harbor Laboratory Press

Degenerate oligonucleotide sequence-directed cross-species PCR cloning of the BCP 54/ALDH 3 cDNA: priming from inverted repeats and formation of tandem primer arrays.

D L Cooper and E W Baptist

Genome Res. 1991 1: 57-62 Access the most recent version at doi:10.1101/gr.1.1.57

References This article cites 20 articles, 3 of which can be accessed free at: http://genome.cshlp.org/content/1/1/57.full.html#ref-list-1

License

Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the Service top right corner of the article or click here.

To subscribe to Genome Research go to: https://genome.cshlp.org/subscriptions