Molecular cloning and nucleotide sequence of full-length cdna for sweet potato catalase mrna

Transcription

1 Eur. J. Biochem. 165, (1987) 0 FEBS 1987 Molecular cloning and nucleotide sequence of full-length cdna for sweet potato catalase mrna Shigeru SAKAJO, Kenzo NAKAMURA and Tadashi ASAHI Laboratory of Biochemistry, Faculty of Agriculture, Nagoya University (Received December 15, 1986/February 9, 1987) - EJB A nearly full-length cdna clone for catalase (pcaso1) was obtained through immunological screening of cdna expression library constructed from size-fractionated poly(a)-rich RNA of wounded sweet potato tuberous roots by Escherichia coli expression vector-primed cdna synthesis. Two additional catalase cdna clones (pcas10 and pcas13), which contained cdna inserts slightly longer than that of pcasol at their 5 -termini, were identified by colony hybridization of another cdna library. Those three catalase cdnas contained primary structures not identical, but closely related, to one another based on their restriction enzyme and RNase cleavage mapping analyses, suggesting that microheterogeneity exists in catalase mrnas. The cdna insert of pcasl3 carried the entire catalase coding capacity, since the RNA transcribed in vitro from the cdna under the SP6 phage promoter directed the synthesis of a catalase polypeptide in the wheat germ in vitro translation assay. The nucleotide sequencing of these catalase cdnas indicated that 1900-base catalase mrna contained a coding region of 1476 bases. The amino acid sequence of sweet potato catalase deduced from the nucleotide sequence was 35 amino acids shorter than rat liver catalase [Furuta, S., Hayashi, H., Hijikata, M., Miyazawa, S., Osumi, T. & Hashimoto, T. (1986) Proc. Nut1 Acad. Sci. USA 83, Although these two sequences showed only 38% homology, the sequences around the amino acid residues implicated in catalytic function, heme ligand or heme contact had been well conserved during evolution. Catalase in sweet potato tuberous roots comprises four identical subunits with a relative molecular mass of about and is localized in microbodies which belong to neither glyoxysomes nor leaf peroxisomes, so-called unspecified microbodies [l, 21. When slices of the root tissue are incubated at moderate temperatures, the enzyme protein increases after a lag period of about 1 day [3]. In order to analyse the mechanism of the increase. we attemdted to clone a cdna for sweet potato catalase mrna, a probe for determining the mrna. Sweet potato catalase protein, strictly its subunit, is synthesized in vitro in the same size as the mature form [4]. Most microbody proteins have been shown to be synthesized in vitro in the mature size and are thought to contain, in the mature proteins themselves, a signal(s) for the post-translational delivery into the organelle [5,6]. Comparisons among the structures of these proteins, as well as analysis of the extra peptides of larger precursors for a few microbody proteins, may be necessary for deducing the structure(s) of the signal(s). In other words, it may be a help in the deduction of the signal(s) to determine the structures of these proteins. In the present work, we have cloned full-length cdnas for sweet potato catalase mrna and deduced the amino acid sequence from the nucleotide sequence. Correspondence to S. Sakajo, Laboratory of Biochemistry, Faculty of Agriculture, Nagoya University, Chikusa, Nagoya, Japan 464 Enzymes. CataIase (EC 1.I 1.1.6); nuclease Bd31 (EC ); restriction endonuclease (EC ); RNApolymerase (EC ); RNase A (EC ). MATERIALS AND METHODS Sweet potato (Zpomoea batatus, Kokei no. 14) roots harvested in autumn were stored at 14 C until use. Slices (4 mm thick), prepared from the parenchymatous tissue of the roots, were incubated at 29 C for 1 day. Preparation and size-fractionation of poly(a)-rich RNA Poly(A)-rich RNA was prepared from the incubated tissue slices as described previously [7]. The poly(a)-rich RNA (0.1 mg) in 0.1 ml buffer A (10 mm Tris/HCl, ph 7.5, 10 mm EDTA, 0.2% SDS) was layered on 10.5 ml of a 5-25% linear sucrose density gradient containing buffer A and centrifuged at rpm for 18 h in a Hitachi RPS40T rotor, after which the gradient was divided into 20 equal volume fractions. Poly(A)-rich RNA in each fraction was precipitated with ethanol, dissolved in sterilized water, and assayed for the activity of catalase protein synthesis with the wheat germ in vitro translation system and immunoprecipitation with anti- (sweet potato catalase) antibody as described previously [4]. Construction and screening of cdna library A cdna library was constructed for a size-fractionated poly(a)-rich RNA fraction with the highest catalase-synthesizing activity by Vector-Primer and linker method [81 using Escherichiu coli expression vector puc8 [9] as described previously [lo].

2 438 The cdna expression library was screened with anti- (sweet potato catalase) antibody by the in situ colony radioimmunoassay method described by Helfman et al. [ll] except that '251-labelled protein A (30 mci/mg, Amersham) was used instead of the second antibody. RNase mapping analysis pcas13, a plasmid containing a cdna for sweet potato catalase mrna, was digested with SulI and treated with BaZ31 to remove the dg tail at the 5'-terminus of the cdna insert. After Sol1 linkers were attached to both termini, the DNA was digested with SalI and PvuII. A 2.0-kb SaZI-PvuII fragment, which contained the cdna and a 0.1-kb sequence derived from puc8, was sub-cloned into SulI - SmuI sites of psp64 [12]. RNAs were synthesized in vitro from this plasmid (pcas113) using SP6 RNA polymerase (Amersham) in the absence of m7gpppg [13]. RNase mapping was performed with 32P-labelled RNA synthesized from StuI-digested pcasll3 as a probe by the method of Myers et al. [14]. The RNA probe containing pcas13 cdna (from the 5'-end to the StuI site) was annealed with cdnas for sweet potato catalase mrna and digested with RNase A. In practice, 20ng of pcaso1, pcaslo or pcasl3 previously digested with StuI was incubated with the RNA probe (6 x lo5 cpm) in 30 pl of 80% formamide, 0.4 M NaCl, 1 mm EDTA, 40 mm Pipes (ph 6.4) at 95 "C for 10 min and then at 45 "C for 40 min, after which the RNA probe was digested by incubation at 25 C for 60 min after addition of 350 pl of an RNase solution (40 pg/ml RNase A, 0.2 M NaCl, 0.1 M LiC1, 1 mm EDTA, 10 mm Tris/HCl, ph 7.5). Then, RNase-resistant RNA fragments were recovered by phenol/ chloroform extraction and ethanol precipitation. The fragments were separated by electrophoresis on a 4.5% polyacrylamide gel under denaturing conditions and detected by autoradiography. Under these conditions, RNA probe alone or the RNA probe hybridized with vector (puc8) DNA was degraded completely. Fig. 1. Detection of a cdna clone with the in situ colony radioimmunoassay method using anti- (sweet potato catalase) antibody. The cdna expression library constructed with the size-factionated poly(a)-rich RNA from incubated slices of sweet potato tuberous root tissue was screened as described in the text. The arrow indicates the clone reacted strongly with the antibody Synthesis of cdna-encodedprotein in vitro EcoRI-digested pcasl13 was transcribed in vitro as described above, and the RNA produced (about 50 ng) was translated with the wheat germ in vitro translation system as in [4]. Other methods Antibody against sweet potato catalase was prepared as described by Esaka and Asahi [2]. Translation in vitro, immunoprecipitation, SDSipolyacrylamide gel electrophoresis and fluorography were performed as in [4]. Methods for nick translation of cdna inserts with [a-32p]dctp (3000 Ci/ mmol, Amersham), colony hybridization, and hybridization selection of mrna were according to [15]. Nothern blot hybridization with a 32P-labelled cdna was performed as in [16] and DNA sequencing was by the M13-derived dideoxynucleotide chain-termination method [17]. RESULTS We constructed a cdna expression library (about 9 x lo3 colonies) from the size-fractionated poly(a)-rich RNA of the incubated tissue slices of sweet potato tuberous roots by Fig. 2. Translation in vitro of hybrid-selectedpoly(a)-rich RNA with pcasoi. RNA selected by hybridization with pcasol out of the total poly(a)-rich RNA from incubated slices of sweet potato tuberous root tissue was translated with the wheat germ in vitro translation system. Then the translation products were electrophoresed on the SDS/polyacrylamide gel and visualized by fluorography. Lanes 1, 2 and 3, total products synthesized without RNA, with RNA hybridization-selected with puc8, and with RNA hybridization-selected with pcaso1, respectively; lanes 4, 5 and 6, immunoprecipitates from the products for lanes 1,2 and 3, respcctively, with anti-(sweet potato catalase) antibody E. coli expression vector-primed cdna synthesis [lo]. When this library was screened by in situ colony radioimmunoassay [l 11 with anti-(sweet potato catalase) antibody, one colony gave a highly positive reaction with antibody (Fig. 1). This

3 439 colony harbored a with a cdna insert of 1.9 kbp. In order to confirm that pcasol is a catalase cdna clone, we selected RNA hybridizing with the plasmid DNA out of the total poly(a)-rich RNA from the tissue slices. When the hybrid-selected RNA was translated in vitro with the wheat germ system, a polypeptide with the same mobility on the SDS/polyacrylamide gel as the mature catalase subunit was synthesized (Fig. 2, lane 3). This polypeptide was immunoprecipitated with anti-(sweet potato catalase) antibody Fig. 3. Northern blot analysis of sweetpotato catalase mrna. Poly(A)- rich RNA (1.2 pg) from sweet potato tuberous roots was glyoxylated and separated on a 1.5% agarose gel. After transfer of RNA onto a Bio-dyne membrane, catalase mrna was detected by hybridization with P-labelled pcasol and autoradiography. Size markers were rrnas from E. coli and potato tubers (Fig. 2, lane 6). Consequently, we concluded that pcasol carried a cdna for sweet potato catalase mrna. Northern blot analysis of tuberous root poly(a)-rich RNA with j2p-labe1led pcasol showed that sweet potato catalase mrna was about 1.9 kb long (Fig. 3). Since the cdna insert of pcasol was also about 1.9 kbp long, pcasol was thought to be a full- or nearly full-length catalase cdna clone. The restriction enzyme cleavage map of pcasol cdna insert is shown in Fig. 4, and its entire nucleotide sequence was determined with the sequencing strategy shown in Fig. 4. The sequence contained a large open reading frame which was continuous from the 5 -terminal nucleotide of the insert (cf. Fig. 7). The first ATG codon appeared between nucleotides from the 5 -terminus. We deduced the amino acid sequence from the nucleotide sequence of the open reading frame (cf. Fig. 7) and compared it with the amino acid sequence of purified beef liver catalase [18]. As a result, we observed that the methionine residue coded by the first ATG codon of the open reading frame corresponded to the 10th amino acid residue (Gln) of the beef liver catalase sequence. This raised a possibility that the cdna insert of pcasol might lack some part of the N-terminal coding sequence as well as the 5 -untranslated region of the catalase mrna. We screened another cdna library by the colony hybridization method using 32P-labelled MluI - SstI fragment of pcasol as a probe to search colonies with longer cdnas. As a result, two colonies were found to contain plasmids (pcas10 and pcas13) with cdna inserts slightly longer than the cdna insert of pcasol at their 5 -termini. Restriction enzyme maps of these three catalase cdna clones for BarnHI, BglII, HindIII, MluI, PstI, SphI, SstI, StuI and XhoI (Fig. 4) showed that the cdna insert of pcas13 was the same as that of pcaso1, whereas the cdna for pcas10 was slightly different from that for pcasol and pcas13, namely it was lacking only in the site of PstI as compared with the others. We also analysed the three cdnas by the RNase mapping procedure [14]; that is, the RNA probe prepared from the cdna for pcas13 was hybridized with each cdna and then digested with RNase A to be cleaved at regions that did not match with the cdna, if any. Only one I I L I kb Fig. 4. Restriction enzyme maps of the cdna inserts in clonespcaso1, pcasl0 andpcasi3. The oligo(dg) tails and poly(da) are exluded. The cleavage sites of restriction endonucleases are shown above (six-base-recognizing enzymes) and under (other enzymes) the bars. A, AluI; H, HaeIII; R, RsaI; S, Sau3A. Arrows show the sequencing strategy

4 Probe M, 9L Fig. 5. RNase mapping analysis of the cdna inserts in clonespcaso1, pcas10 andpcasi3. The RNA probe prepared from pcas113 was hybridized with pcas03 (lane i), pcaslo (lane 2) and pcas13 (lane 3) and digested with RNase A, after which RNA fragments were separated and detected as described in the text. The position of the RNA probe and the lengths in base numbers of size makers are indicated at the left side RNA fragment with the same size as the probe was detected when the probe was hybridized with pcas13 cdna (Fig. 5, lane 3), whereas several fragments were observed with pcaslo cdna (Fig. 5, lane 2), which indicates that the nucleotide sequences of these two cdnas differ from each other at several locations. In the case of pcasol cdna, a few RNA fragments were produced (Fig. 5, lane l), suggesting that the cdnas for pcasol and pcas13 are very similar, but not identical, to each other. To examine whether the cdna insert of pcas13 had all the coding region for the catalase protein, the cdna insert sequence was transcribed in vitro under the SP6 phage promoter, and the transcripts were analysed by in vitro translation with the wheat germ system. When immunoprecipitates from the translation products with anti-(sweet potato catalase) antibody were analysed by SDS/polyacrylamide gel electrophoresis, a single polypeptide with the same mobility as the catalase protein synthesized with the poly(a)-rich RNA from sweet potato tuberous root tissue was detected (Fig. 6, lanes 2 and 3). The result indicates that pcasl3 carries the entire proteincoding region for sweet potato catalase. The nucleotide sequence of 5 -terminal part of pcas13 cdna were determined. It contained an additional 54-bp sequence at the 5 side of a 224-bp sequence identical to the 5 -terminal sequence of pcasol cdna (Fig. 7). There was not ATG codon within the 5 -terminal 54-bp sequence of Fig. 6. Synthesis in vitro ofproteins encoded b,v pcas13 cdna insert. RNA transcribed in vitro from pcasl3 cdna insert cloned in psp64 with SP6 RNA polymerase (lanes 1 and 2) or the total poly(a)-rich RNA from incubated slices of sweet potato tuberous root tissue (lane 3) was translated with the wheat germ in vitro translation system, and immunoprecipitates with control serum (lane 1) or anti-(sweet potato catalse) antibody (lanes 2 and 3) from the translation products were electrophoresed on the SDS/polyacrylamide gel and visualized by fluorography pcasl3 cdna which was absent in pcasol. Since pcasl3 cdna carried the full coding capacity for catalase and the next ATG codon in the open reading frame appeared 373 nucleotides downstream from the first ATG codon, we concluded that the first ATG codon in the open reading frame in the pcasol cdna sequence was an initiation codon for the catalase translation (Fig. 7). The N-terminal amino acid sequence of purified sweet potato catalase could not be determined, possibly due to a minor modification in the N-terminal part of the protein [4]. The open reading frame was 1476 bp in length, and it can code for a polypeptide with 492 amino acid residues. The relative molecular mass (56984) and the amino acid composition of catalase deduced from the DNA sequence were similar to those of purified sweet potato catalase [2] (data not shown). The nucleotide sequence of the catalase cdna also contained a 247-bp-long 3 -untranslated region (Fig. 7) and a poly(da) tail. A consensus polyadenylation signal (AATAAA) was present 20 bp upstream of the poly(da) tail. DISCUSSION The catalase cdna clone pcaso1 identified by immunological screening of the cdna expression library contained 27 bp of the 5 -untranslated region in addition to the entire coding region, the 3 -untranslated region and poly(a) tail of the catalase mrna. Since the cdna insert of pcasol was linked to the PstI terminus of the puc8 vector by the (dg). (dc) homopolymer tail of 18 bp (date not shown), the coding frame for catalase protein does not match to that for the N-

5 441-8lAGTGTGATCACTGCTACCTCTCTTTATTCTACAAGCTrTCCACTTTCTTCTCCCCATTATTCTCTCTGTCCCCTCATCTCC A 1 ATG GAT CCT TCA AAG TAC CGT CCA TCA AGT AGC TTC AAC ACA CCC TTC TGC ACT ACC AAC 1 Met Asp Pro Ser Lys Tyr Arg Pro Ser Ser Ser Phe Asn Thr Pro Phe Cys Thr Thr Asn 61 TCC GGA GCT CCG GTA TGG AAC AAC ACC TGC GCA CTC ACC GTC GGC AGC AGA GGA CCA ATT 21 Ser & Ala Pro Val Trp Asn Asn Thr Cys Ala Lue Thr Val% Ser Arg Gly Pro Ile 121 CTG CTA GAA GAT TAT CAC TTG GTG GAG AAA ATT CAA AAC TTC ACT CGT GAA AGG ATC CCA 41 &Leu Glu & Tyr His Leu Val Glu Lys Ile Gln Asn & Thr Arg Glu Arg Ile Pro v 181 GAA CGA GTG GTG CAT GCC AGG GGT GCA AGT GCC AAG GGC TTC TTT GAG GTC ACT CAT GAC 61 Glu Arg Val Val His Ala Arg Gly Ala Ser & Lys Phe Phe Glu Val Thr His Asp ATT ACA CAC CTC ACC TGC GCC GAC TTC CTC CGT GCC CCC GGC GTT CAG ACA CCT CTC ATC Ile Thr His Leu Thr Cys & t Asp Phe Leu Arg Ala Pro Q Val Gln Thr Pro Leu lle GTC CGC TTC TCC ACT GTC ATC CAT GAG CGT GGT AGC CCC GAA ACT ATC AGA GAT CCC CGT Val & Phe Ser Thr Val Ile His Glu Arg Gly Ser Pro Glu Thr Ile Arg Asp Pro Arg - - GGT TTT GCC GTC AAG ATG TAC ACC CGT GGA GGA AAC TGG GAT TTG GTG GGC AAC AAT TTC Gly Phe Ala Val Lys Met Tyr Thr Arg Gly Gly Asn Trp Asp Leu Val Gly Asn Asn Phe - CCG GTG TTC TTT ATC CGG GAC GGA ACG CAG TTC CCG GAC GTG ATC CAC GCG TTC AAG CCA Pro Val Phe Phe Ile Arg Asp Gly Thr Gln Phe Pro Asp Val Ile His Ala Phe & Pro - AAC CCG AAA TCC CAC ATC CAG GAG AAC TGG AGA ATC CTG GAT TAC TTA TCC CAT CTC CCG Asn Pro Lys Ser His Ile Gln Glu Asn Trp Arg Ile Leu & lyr Leu His Lru Pro -_ 541 GAG AGT CTC AAC ACC TTC GCC TGG TTC TAC GAC GAT GTC GGT ATC CCC ACC GAT TAC CGC 181 Glu Ser Leu Asn Thr Phe Ala Trp Phe Tyr Asp & Val Gly Ile Pro Thr Asp Tyr & 601 CAC ATG GAA GGC TTC GGC GTC CAC ACT TTC ACC ATG ATC AAC AAG GAA GGC AAG GCC AAT 201 His Met Glu & Phe 9 Val H is Thr Phe Thr Met Ile & Lys Glu Q Lys & Asn 661 TAT GTA AAA TTC CAC TGG AAA CCC ACC TGC GGC GTC AAA TGT CTG CTG GAA GAG GAG GCG 221 Val Lys Phe His Trp & Pr Thr Cys Q Val Cys Leu Glu Glu Glu Ala P 721 ATT AGG ATC GGC GGC GAG AAT CAC AGC CAC GCC ACC CAG GAT TTA TAC GAG TCC ATC GCC 241 Ile & Ile Gly Gly Glu Asn H is Ser H is Ala Thr Gln Asp Leu Tyr Glu Ser Ile Ala GCC GGG AAT TAC CCG GAG TGG AAG CTT TAT ATC CAA GTG ATG GAC CCG GAT CAC GAG GAC Ala Gly Asn Tyr Pro Glu Lys Leu Tyr Ile Gln Val Met Asp Pro Asp His Glu Asp CGG TTC GAT TTT GAC CCG CTG GAC ACG ACC AAG ATC TGG CCG GAA GAG TTG ATT CCT CTG Arg & Asp Phe Asp Pro Leu & Thr Thr Lys Ile Trp Pro Glu Glu Leu Ile Pro Leu CAG CCG GTG GGG AGA ATG GTG TTG AAC AAG AAT ATT GAT AAT TTC TTT GCG GAG AAT GAG Gln Pro Val Gly Arg Met Val Leu Asn Lys & Ile Asp & Phe Phe Ala Glu Asn Glu ATG TTG GCG ATG GAC CCG GCG CAT ATT crc ccc GGA ATA TAC Trc TCC GAT GAT AAG ATG Met Leu & Met Asp Pro Ala H is Ile Val Pro Gly Ile Tyr Phe Ser Asp Asp Lys Met CTC CAG GCT CGA GTC TTT GCC TAC GCC GAC ACT CAC CGC CAC CGC CTT GGC CCC AAC TAT Leu Gln Ala & Val Phe Ala Tyr Ala Asp Thr His Arg His - Arg Leu Gly Pro Asn Tyr ATG CTG CTT CCG GTT AAT GCC CCC AAG TGC GCT CAT CAC AAC AAT AGC TAT GAT GGT TAC Met Leu Leu Pro Val Asn Ala Pro Lys Cys & H is His Asn & Ser Tyr Asp Gly Tyr 1141 ATG AAC TTT GTC CAC AGG GAT GAA GAG GTT GAT TAC TTT CCC TCG AAG TTT GAT AAC ACA 381 Met Asn Phe Val His Arg Asp Glu Glu Val Asp Tyr Phe Pro Ser Lys Phe Asp Asn Thr 1201 CGT AAC GCT GAG AGG TTC CCA ACT CCG TTG CGT ATC GTG ACG GGG CAA CGT GAT AAG TGT 401 Arg Asn Ala Glu Arg Phe Pro Thr Pro Leu Arg Ile Val Thr Gly Gln Arg & Lys Cys 1261 GTT ATT GAG AAG GAG AAC AAC TTC AAG CAA CCT GGA GAT AGA TAC AGA TCC TGG GCT CCA 421 Val Ile Glu Lys Glu & Asn Phe Lys Gln Pro Gly Asp & Tyr Arg Ser Trp Ala Pro 1321 GAC AGG CAA GAT AGA TTC ATC AAC CGA TGG GTC AAG GCC TTG TCT GAG CCC CGA GTC ACC 441 Asp Arg Gln Asp Arg Phe Ile Asn Arg Trp Val Lys Ala Leu Ser Glu Pro Arg Val Thr 1381 CAT GAA ATT CGC AGC ACT TGG ATT TCT TAC CTC ACT CAG GCT GAT AGG TCT CTT GGA CAG 461 His Glu Ile Arg Ser Thr Trp Ile Ser Tyr Leu Thr Gln Ala Asp Arg Ser Leu Gly Gln 1441 AAG GTG GCT TCC CGT CTG AAC ATA AGA CCG ACG ATG TAA GAGAGGTGGAATTAAAGGCAGTATAAA 481 Lys Val Ala Ser Arg Asn Ile Arg Pro Thr Met term 1507 TTGCTTTTGAGTTGAAATACATGCATGTTGTACTAAATGGGACGGCACCGTTTATTACTGTAAATTTGATGTTTATCAA 1586 TGTCGTATTTTGGGTTTTACAAAATACTATGTTGTTATTTGGTCCTATATGTGACGATGCAAGTGAACTGGTACGAT(;T 1665 AAGTAGACTGGGACAATGTTCTAG1 TTGAATAAT-TTATCGTTATTGTGTGGTC 1723 Fig. I. Nucleotide sequence of cdna for sweet potato catalase mrna. The sequences of the cdna insert of pcasol (from the nucleotide indicated by an upward arrowhead to the 3 -terminal) and the 5 -end part of the insert of pcas13 (from the 5 -terminal to the nucleotide indicated by a downward arrowheat) are shown in a continuity (see Fig. 4 for the sequencing strategy). Nucleotides are numbered from the first base of the start codon (ATG). The deduced amino acid sequence of sweet potato catalase is indicated under the nucleotide sequence. Underlined amino acid residues are homologous to rat liver catalase [20] (a single line) and correspond to the residues proposed to be catalytically important for beef liver catalase [22] (double lines). An underlined nucleotide sequence is the consensus poly(a) addition signal sequence

6 442 terminal coding sequence of j-galactosidase on the vector [9]. Therefore, the translation of catalase-related antigen in E. coli cells harboring pcasol is most likely to be re-initiated from an initiation codon inside the cdna. We have previously reported the direct immunological identification of full-length cdna clones for several other proteins by the same approach as used in the present study [7, 101. The three catalase cdna clones obtained in the present study are not identical, but closely related, to one another in their primary structures (Figs 4 and 5). The microheterogeneity of cdnas strongly suggests the pesence of microheterogeneous catalase mrnas, in other words the presence of multiple catalase genes closely related to each other in the hexaploid genome of sweet potato. It is not known whether this heterogeneity is solely due to polyploidy of the genome and polymorphism among the same allele or due to duplicated genes on a chromosome. The maize genome has been reported to contain three loci coding for three distinct catalase polypeptides [ 191. We compared the amino acid sequence deduced in the present work with those of mammalian catalases [18, 20, 211. Sweet potato catalase protein is smaller than the mammalian protein. We found the best homology between the amino acid sequences deduced for sweet potato and mammalian liver [20, 211 catalase proteins, when the plant protein was proposed to be lacking in the 10 and 25 amino acid residues in the N- and C-terminal parts, respectively, of the mammalian proteins. Even in such a comparison, only 38% of the total amino acid residues of the plant protein are homologous to the residues of the rat protein (Fig. 7). However, close similarites between the sequences are seen in parts, especially in regions around the amino acid residues which have been shown with beef liver catalase to play important roles in the catalytic function (His- 65 and Asn-138 in Fig. 7) and the heme ligand (Tyr-348 in Fig. 7) [22]. Out of the 39 amino acid residues proposed to contact with the heme for the beef liver enzyme (amino acid residues numbered as 51, , 62-65, 102, 104, 106, 107, , 121, 123, 134, , 143, 144, 148, 151, 154, 155, 158, 159, , 201, 340, 343, 344, 347, 348, 351, 352, and 355 in Fig. 7) [22], only eight residues are different between the plant and mammalian proteins. In contrast there are dissimilarities in the sequences of other parts, especially the N-terminal (1-17) and C-terminal ( ) parts, between the plant and mammalian proteins. These observations show that amino acid residues important for the catalytic activity had been conserved very well through evolution, whereas other residues had been changed, deleted and/or added. It has been proposed from studies with trypanosome glycosomes that a region with positive charges on proteins may function as a signal for delivery of the proteins into microbodies 161. Actually, clusters of positively charged amino acid residues exist in mammalian catalases [18, 20, 211. In the case of sweet potato catalase, such clusters are not observed but some regions (for instance, amino acid residues ) in the sequence tend to be rich in positively charged residues. One of these regions may possibly play an important role in delivery of the protein into the microbodies. REFERENCES 1. Esaka, M. & Asahi, T. (1979) Plant Cell Physiol. 20, Esaka, M. & Asahi, T. (1982) Plant Cell Physiol. 23, Esaka, M., Maeshima, M. & Asahi, T. (1983) Plant Cell Physiol. 24, Sakajo, S. & Asahi, T. (1986) FEBS Lett. 205, Lazarow, P. B. & Fujiki, Y. (1985) Annu. Rev. Cell Biol. I, Borst, P. (1986) Biochirn. Biophys. Acta 866, Hattori, T., Nakagawa, T., Maeshirna, M., Nakarnura, K. & Asahi, T. (1985) Plant Mol. Biol. 5, Okayama, H. & Berg, P. (1982) Mol. Cell Biol. 2, Vieira, J. & Messing, J. (1982) Gene 19, Nakamura, K., Hattori, T. & Asahi, T. (1986) FEBS Lett. 198, Helfman, D. M., Feramisco, J. R., Fiddes, J. C., Thomas, G. P. & Hughes, S. H. (1983) Proc. Natl Acad. Sci. USA 80, Melton, D. A,, Krieg, P. A,, Rebagliati, M. R., Maniatis, T., Zinn, K. & Green, M. R. (1984) Nucleic Acids Res. 12, Krieg, P. A. & Melton, D. A. (1984) Nucleic Acids Res. 12, Myers, R. M., Larin, Z. & Maniatis, T. (1985) Science (Wash. DC) 230, Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular cloning: A laboratory rnannual, Cold Spring Harbor Laboratory, New York. 16. Thomas, P. S. (1983) Methods Enzyrnol. 100, Messing, J. (1983) Methods Enzyrnol. 101, Schroeder, W. A., Shelton, J. R., Shelton, J. B., Robberson, B. R., Apell, G., Fang, R. S. & Bonaventura, J. (1982) Arch. Biochern. Biophys. 214, Roupakias, D. G., McMillin, D. E. & Scandalios, J. G. (1980) Theor. Appl. Genet. 58, Furuta, S., Hayashi, H., Hijikata, M., Miyazawa, S., Osurni, T. & Hashimoto, T. (1986) Proc. NatlAcad. Sci. USA83, Bell, G. I., Najarian, R. C., Mullenbach, G. T. & Hallewell, R. A. (1986) Nucleic Acids Res. 14, Murthy, M. R. N., ReidIII, T. J., Sicignano, A., Tanaka, N. & Rossman, M. G. (1981) J. Mol. Biol. 152,