7/27/2019 ejb_94

1/6

Eur. J. Biochem. 224, 519-524 (1994)0 EBS 1994

Sequence similarities and evolutionary relationshipsof microbial, plant and animal a-amylasesStefan JANECEKInstitute of Ecobiology, Slovak Academy of Sciences, Bratislava, Slovakia(Received March 22lJune 17, 1994) - EJB 94 039213

Amino acid sequence comparison of 37 a-amylases from microbial, plant and animal sourceswas performed to identify their mutual sequence similarities in addition to the five already describedconserved regions. These sequence regions were examined from structure/function and evolutionaryperspectives. An unrooted evolutionary tree of a-amylases was constructed on a subset of 55 resi-dues from the alignment of sequence similarities along with conserved regions. The most importantnew information extracted from the tree was as follows: (a) the close evolutionary relationship ofAlteromonas haloplanctis a-amylase (thermolabile enzyme from an antarctic psychrotroph) with thealready known group of homologous a-amylases from streptomycetes, Thermomonospora cuwata,insects and mammals, and (b) the remarkable 40.1% identity between starch-saccharifying Bacillussubtilis a-amylase and the enzyme from the ruminal bacterium Butyrivibrio fibrisolvens, ana-amylase with an unusually large polypeptide chain (943 esidues in the mature enzyme). Due toa very high degree of similarity, the whole amino acid sequences of three groups of a-amylases,namely (a) fungi and yeasts, (b) plants, and (c) A. haloplanctis, streptomycetes, T cuwata , insectsand mammals, were aligned independently and their unrooted distance trees were calculated usingthese alignments. Possible rooting of the trees was also discussed. Based on the knowledge of thelocation of the five disulfide bonds in the structure of pig pancreatic a-amylase, the possible disul-fide bridges were established for each of these groups of homologous a-amylases.

Although all a-amylases catalyze the hydrolysis of a-l,4-D-glucosidic bonds, their amino acid sequences have highvariability [11.Generally, five highly conserved sequence re-gions [2, 1 are described. Despite the very low sequenceidentity (10% or less) [2],he a-amylases have been groupedin one family of similar sequences along with other enzymesinvolved in starch metabolism [4]. s indicated from predic-tion studies [l, 51 and confirmed by the crystal structuresof four a-amylases (pig pancreas [6], Aspergillus niger [7],Aspergillus olyzae [S] and barley [9]), the a-amylases con-tain a (j3/a),-bmel domain [lo].

More than 50 complete amino acid sequences of a-amy-lases from different microbial, plant and animal sources arecurrently known [ll, 121, he vast majority being deducedfrom the DNA sequences. The total number of availablea-amylase sequences is much higher due to the presence ofmultiple a-amylase genes or isozymes in one organism. Al-though there is extensive information on a-amylases se-quences from different sources, an overview of their diversityis lacking. Several pairs or groups of a-amylases are knownto be homologous, such as liquefying Bacillus a-amylases[13], he a-amylases from A. oryzae and Saccharomycopsisfibuligera 1141, Aeromonas hydrophila and Xanthomonascampestris a-amylases [1.51, the Streptomyces a-amylasesalong with Thermomonospora curvata, Drosophila melano-gaster and mammalian a-amylases [16, 71, nd the a-amy-

Correspondence to,5. JaneEek, Institute of Ecobiology, SlovakAcademy of Sciences, Stefhnikova 3, SK-81434 Bratislava, SlovakiaF a x : +42 7 334967.Enzymes. a-Amylase (EC 3.2.1 I); cyclodextrin glycosyltrans-ferase (EC 2.4.1.19).

lases from plants [ lS] . For all these examples, sequenceanalysis remains to be performed.Therefore, the main goal of the present study was to sum-marize the findings resulting from the analysis of amino acidsequences of available microbial, plant and animal a-amy-lases. To further elucidate the evolutionary aspects of a-amy-lases, unrooted distance trees were constructed and discus-sed.

MATERIALS AND METHODSAmino acid sequences of a-amylases (Table 1) were ex-tracted from the Swiss-Prot protein and GenBank DNA se-quence data banks [ll, 121.Conserved positions and semi-conservative substitutions

from the authors previous alignment of eight a-amylases [3]and other alignments [ l ,15,16, 191 were used in searchingfor sequence similarities in the present set of a-amylases.The identified similarities, along with the isolated fiveconserved sequence regions of each a-amylase [2, 1 werealigned to give sub-sets of 55 amino acid residues. An un-rooted distance tree was calculated for 30 out of 37 a-amy-lases based on this alignment by the neighbour-joiningmethod [20]mplemented in the program CLUSTAL V [21].The seven a-amylase sequences, three of which (A. niger andAspergillus shirousami, and Streptomyces griseus) were inthe sequence sub-sets identical to A. oryzae and S. limosusenzymes, respectively, and four of which (mouse, rat andhuman pancreas, and human saliva) were different from theirpancreatic (pig pancreas) and salivary (mouse saliva) coun-

7/27/2019 ejb_94

2/6

520Table 1. a-Amylases ncluded in the present study. All Swiss-Protdatabase numbers start with P.Abb revia- Organism (specification) SwissProttion or GenBankaccessionnumberAerhyAlthaBacamBacliBacmeBacstBac suButfiDicthEsccoMicspSaltyStrgrStrhyStrliStrthStrviThecuXancaAspniAsporAspshSacfiSchocBarleMaizeRiceaVigmuWheatAnogaDromeMuspaMussaPigpaRatpaHumpaHumsa

Aeromonas hydrophilaAlteromonas haloplanctisBacillus amyloliquefaciensBacillus licheniformisBacillus rnegateriumBacillus stearothermophilusBacillus subtilisButyrivibrio fibrosolvensDictyoglomus therrnophilurn (amy B)Escherichia coliMicrococcus sp.Salmonella typhimuriumStreptomyces griseusStreptomyces hygroscopicusStreptomyces limosusStreptomyces thermoviolaceusStreptomyces violaceusThermomonospora curvataXanthomonas campestrisAspergillus nigerAspergillus oryzaeAspergillus shirousamiSaccharomycopsis fibuligeraSchwanniomyces occidentalisHordeum vulgare (barley isozyme A )Zea mays (maize)Oryza sativa (rice isozyme 1B)Vigna mungo (rice bean, black gram)Triticum aestivum (wheat amy 3)Anopheles gambiaeDrosophila melanogasterM us musculus (mouse pancreas)M us musculus (mouse saliva)Sus scrofa (pig pancreas)Rattus norvegicus (rat pancreas)Homo sapiens (human pancreas)Homo sapiens (human saliva)

P22630P29957PO0692PO6278P20845PO6279PO0691P30269P14898P26612x55799P26613P30270PO8486PO9794P27350P22998P29750M85252X.52755P10529P30292P21567P19269PO0693L2580.5P17654P178.59PO8117LO4753PO8 144PO0688PO0687PO0690PO0689PO4746PO4745

terparts in only one residue, were not used in the constructionof this tree.Whole amino acid sequences of three groups of a-amy-lases with high degree of mutual sequence similarities thatformed compact clusters in the evolutionary tree werealigned independently. These alignments were made usingthe program CLUSTAL V [21]. The groups were (a ) fungiand yeasts, (b) plants, and (c) streptomycetes, insects andmammals along with Alteromonas haloplanctis and T cur-vatu. Based on these alignments, the unrooted distance treeswere constructed by the neighbour-joining method [20]. Thecysteine residues involved in probable disulfide bridges ofthe third group of a-amylases represented by streptomycetesand animals were extracted from the amino acid sequencealignment of this group by using the known cystines of pigpancreatic a-amylase [22] as the template.RESULTS AND DISCUSSIONSequence similarities

a-Amylases contain five conserved sequence regions,four of which are well established [23] and usually noted in

AerhyAlthaBacamBacliBacmeBacStBaCSUButfiDicthESCCOMicspSaltyStrgrStrhyStrliStrthStrviThecuXancaAspniAsporASpShSacfiSchocBarleMaizeRiceaVigmuWheatAnogaDromeMuspaMussaPigpaRatpaHumpaHumsa

I1-6 FNW 151-8 FEW 161-8 FEW 22~1-10 FEW 221-18 YVN 371-11 FEW 221-14 WNW 151-120 FCW 151-139 FI D 361-9 FHW 221-353 LTD 431-9 FHW 221-12 FEW 161-12 FER 161-12 FEW 161-11 FEW 161-12 FEW 161-12 FQW 161-6 FNW 151-14 LT D 381-14 LTD 381-14 LTD 381-15 VT D 381-25 VTD 381-7 FNW 231-7 FNW 231-6 FNW 231-6 FNW 221-7 FNW 231-16 FEW 161-16 FEW 161-16 FEW 161-16 FEW 161-16 FEW 161-16 FEW 161-16 FEW 161-16 FEW 16

11GYXQVLISPGYAAVQVSPGITAVWIPPGITAVWIPPGIWMMPVNPGITALWLPPGYTAIQTSPGYTAVQTSPGINTIWISPGINMVWLPPGVNTIWISPGINMWLPPGYGYVQVSPG Y G W N S PGYGWQVSPGYGWQVSPGYGYVQVSPGFGAVQVSPGYRKVLVAPGFTAIWITPGFTAIWITPGFTAIWITPGFTAIWISPGFTAIWISPGVTHVWLPPGATHVWLPPGITHVWLPPGITHVWLPPGATHVWLPPGYGGVQLSPGYAGVQVSPGFGGVQVSPGFAGVQVSPGFGGVQVSPGFGGVQVSPGFGGVQVSPGFGGVQVSP*

10889

11911983

119879578

14190

141898789898998

10889898989898787868786

10194

102105105102105105

I11DWSDDYGNDWDEDWDESWGQDWDENWSDGYTDDYETGWNDRFSDGWNDDYGNDYTNDYGNDYQDNYQDDNYNDNYEDNYEDNYEDNYDDNYNDKYSDQYSNPYGDAYSDXYSNDWGNNYNDNYNDNYQDSYNDNYNDNYNDNYND

177167191189185190165198160172180172163159163163163166177165165165165165184181184181166176179180180180180180180

IVGSPLVYSDHGYPKVMSSYGYPQVFYGDGYPQVFYGDGNPYIYYGEGYPCVFYGDGSTPLFFSRGTPLFFSRPAIPIIYNGQGVPSVFYPDGQPVIYYGEGVPSVFYPDGSPDVHSGYGSPNVYSGYGSPDVHSGYGAPDINSGYGSPDVHSGYGTPKVMSSYGVPMVYTDNGIPIIYAGQGIPIIYAGQGIPIIYAGQGIPVIYYGQGIPIIYYGQGIPCIFYDHGTPCIFYDHGNPCIFYDHGTPSIFYDHGIPCIFYDHGQLRIMSSFG T P R W S FGFTRVMSSYGFTRVMSSYGFTRVMSSYGFTRVMSSYGFTRVMSSYGFTRVMSSY

3.12-443306-453366-483366-483349-493368-51531J7-619454-9434.19-562370- 5692-11 430-494306-538300-448306-5383115-431306-541318-5723J2-4403.12-4773.12-4783.?2-4783.13-468343-487327-414324-412325-403322-398358-3893 15-489311-476340-493343-496343-496340-493343-496343-496

Cons f w G v SP y d g P vFig. 1. Sequence similarities in a-amylases. The abbreviations ofenzyme sources are given in Table 1. a-Amylases are numberedfrom the N-terminus of the mature (when known) enzymes. Thenumbers represent the number of amino acid residues between tworegions and the sizes of the peptide chain preceding the first andending the last region, respectively. Invariable amino acid residuesare indicated (*). A residue is written in the consensus ( Cons ) se-quence if it is present in more than half of the enzymes.new a-amylase sequences. The fifth conserved regicln hasbeen demonstrated only recently [3]. It comprises the se-quence region around Asp1 67 (pig pancreatic a-amylasenumbering) involved in the binding of Ca2+ 6]. For example,the regions in the amino acid sequence of pig pancreatica-amylase are as follows: 1-95 DAVINH 63 LLDLA 23GFRLDASKH 31 EVID 58 FVDNHD 301-496; the num-bers indicate the size of the sequence between the differentregions and the sizes of the start and end of the sequence.The present study adds four sequence similarities (Fig. 1) tothese conserved regions. These regions were identified alongthe sites where certain single amino acid residues arestrongly conserved or where so-called semi-conservativesubstitutions (hydrophobic or hydrophilic residues) are pre-sent. Taking into account the refined three-dimensional struc-ture of the (Pla),-barrel of pig pancreatic a-amylase [6]. thelocations of short regions found in this study to be similar inall a-amylases are as follows: region I (FEW, pig pancreatica-amylase in Fig. l ) , loop 1 ; region I1 (GFGGVQVSP),strand P2 ; region I11 (SYND), the longest loop 3 (domainB); region IV (GFTRVMSSY), strand /38. Both the loops(loop 1, loop 3) are the loops joining the C-terminus of a p-strand to the N-terminus of the adjacent helix, i e . the loopsin P-loop-a units. For comparison, the locations of previouslydescribed conserved regions are also in P-strands and loops,i.e. none of the a-helices of the @/a),-barrel is involved inthe conserved regions. This reflects that the sequencerequirements of the inner /3-barrel are more stringent than therequirement of the outside helical cylinder. This is a teaturecharacteristic of all known @/a),-barrel starch hydrolases andrelated enzymes [24].

There are two main differences between the previouslydescribed conserved regions [l -31 and the sequence simi-

7/27/2019 ejb_94

3/6

7/27/2019 ejb_94

4/6

522PigpaAltha

PigpaAltha

PigpaAltha

PigpaAltha

PigpaAltha

PigpaAltha

PigpaAltha

PigpaAltha

PigpaAltha

PigpaAltha

QYAPQTQSGRTDIVHLFEWRWVDIALEcERYLGPKGFGGVQVSPPNENVWTNPS- - --.- - PTTPVHLFEWNWQDVAQECEQYLGPKGYAAVQVSPPNEHIT----G* ****** * * * *** ****** ********RPWWERYQPVSYKLcTRSGNENEFRDMVTFlCNNVGVRIYVDAVINHMCGSGAAAGS Q W W T R Y Q P V S Y E L Q S R G G N R A Q F I D ~ c S A A G V D 1 Y S G** ******* * t rt * *** ** ** **** **** * *'IGTTccSYCNPGNRBFPAVPYSAWDFNDGKCKTASGGIESYNDP-YQVRDcQLVGTGTAGNSF---GNKSFPI--YSPQDFHES-cTI"----SDYGN~RVRVQNCEIVC,*** * ** ** ** ** t * * * ***LLDLALEKDYVRSHIADYLNKLIDIGVAGFKLDASKHUWPGDIKAVLDKLHNLNTLADLDTASNYVQNTIAAYINDLQAIGVKGFRFDASKHVAASDIQSLNAKVN----t tt ** tt * * * *t t *** ***** **NWFPAGSRPFIFQEVIDLGGEAIKSGEYFSNGRVTEFKYGAKLGTVVRKWSGEKM------ SPWFQEVIDQGGEAVGASEYLSTGLVTEFKYSTELGNTFRN---GSL* ****** **** ** * * ****** ** *S Y L K N W G E G W G f H P S D R A L V F V D N H D N Q R G I l G A G G S S I LAWLSNFGEGWGFHPSCSAWFVDIltiDNQRGtiGGAGN-VITFEDGRLYDLANVFML* t ***"***** * "*******..*** .* * * ***AIiPYGFTRVHSSYRWARNFVNGEDVNDWIGPPNNNGVIKEVTINADTTC-GNDWVAYPYGYPKVHSSY----DFIlGDTDAGGPNVPVHNNG~E---------cAsN~I( ** * *r*** * " * ** r t rCEHRWSY I GGVDFRNNTADNWAVTNWDNTNN~~SFGRGSSGHMAINKEDSGLT***** * * *** ***** ** **** * * * *S T L Q T G L P A G T Y C D V I S G D K V G N - - S c T G I K V Y V S S D G K A A IATVQTDMASGQYc NVLKGELSADAKSc SGEVI T~SDGTI NLNI GAWDA-- -MAI* * * * * * * * ** * *** * **HAESKL------ 496HKNAKLNTSSAS 453* **

5543

11095

164141

219191

274238

329292

383334

437389

490441

Fig.3. Amino acid sequence alignment of the a-amylases frompig pancreas (Pigpa) and A. haloplanctis (Altha).The sequencesare numbered from the N-terminus of the mature enzymes. Identicalamino acid residues (*)and gaps (-) are indicated. The ten cysteineresidues involved in the five disulfide bridges of pig pancreatica-amylase 1221 an d the eight equivalent cysteine residues of theenzyme from A. haloplunctis are shown in lower-case letters an d areitalicized.

served regions [ l , 21, the fifth conserved region [3] and thesequence similarities (this study, Fig. 1).The evolutionary relationships of fungal and yeasta-amylases are shown in Fig. 2B. Taxonomy is respected de-spite the very similar amino acid sequences, and fungi andyeasts form their own evolutionary groups. The fungala-amylases are mutually more closely related than yeast en-zymes. The position of these a-amylases in the whole tree(Fig. 2A) next to the enzymes with the longest branches(Dictyoglomus thermophilum, Micrococcus sp., Bacillusmegaterium) is remarkable.

The plant a-amylases appear between the liquefying andsaccharifying a-amylases of the genus Bacillus (Fig. 2A).These a-amylases form another group of closely related en-zymes (Fig. 2C). They differ from the rest of the a-amylasesespecially with respect to the sequence in the third conservedregion, e.g. AWRLDFARG in barley (low PI isozyme) com-pared with GFRLDASKH in pig pancreatic a-amylase.

As mentioned above, the a-amylases from D. thermophi-lum, Micrococcus sp. and B. megaterium are on the longestbranches of the tree in Fig. 2A. This indicates that these en-zymes least resemble the other enzymes. Indeed, several dis-tinct differences from the rest of the a-amylases can be foundalso in the regions of sequence similarity shown in Fig. 1,such as threonine residues at the fourth position of region ITfor D. thermophilum and Micrococcus sp. a-amylases and aproline residue at the sixth position of this region for theB. rnegateriurn enzyme. Moreover, D. thermophilum and B.rnegateriurn enzymes possess an intermediary sequence inregion I, i.e. Phe (Tyr) and Asp (Asn) at the first and the thirdpositions, respectively, in comparison with FEW or LTD formost a-amylases (compare Fig. 1). Furthermore, the a-amy-lase from Micrococcus sp. has the largest polypeptide chain

(1104 amino acid residues) of all a-amylases known to date[11, 121. Despite its size, this a-amylase, in addition to thesupposed @/a),-barrel domain, apparently does not containparts recognized elsewhere, e.g. the C-terminal starch-hind-ing domain [26]. D. thermophilum produces, in addition tothe enzyme investigated in this study (a-amylase B). twoother a-amylases designated A and C [27, 281. The sequenceof a-amylase A [27] is exceptional, since the conserved se-quence regions [2, 31 are not apparent in this sequence Theposition of B. megaterium a-amylase in the tree in Fig. 2Acan be explained only by the lower degree of sequence simi-larity throughout the conserved regions (data not shown) andthe sequence similarities (Fig. 1).As for the rooting of the trees presented in Fig. 2 , themidpoint method [29] could be used. The root in this methodcan be placed at the half-way mark along the longest recon-structed lineage between two taxa. Thus, the tree presentingthe evolutionary relationships of all a-amylases in this study(Fig. 2A) could be rooted along the branch of D. therniophi-lum a-amylase. Correct rooting in this case should be war-ranted by the fact that this organism produces one a-amylase(a-amylase A) that is homologous with the Pyrococcus-furio-sus a-amylase [ 3 0 ] ,which is probably one of the mo\t an-cient a-amylases. The rest of the trees in Fig. 2 can be rootedas follows by analogy: (a) fungal a-amylases would be sepa-rated from the yeast enzymes in the tree in Fig. 2B; (b) plac-ing the root along the branch of Wgna rnungo a-amylase inthe plant enzyme tree (Fig. 2C), two different taxonomicgroups [Leguminosae (V. mungo) and Gramineae (the rest ofplant a-amylases)] would result in this case, and (c) the rootplaced along the branch of 7: cuwata a-amylase in the treein Fig. 2D would divide the enzymes produced by microor-ganisms and animals with the exception of A . haloplanctisa-amylase. It should be pointed out, however, that the A.haloplanctis enzyme is on a branch with length comparableto the branch length of T cuwata a-amylase.

One interesting finding can be extracted from the nlign-ment of the amino acid sequences of a-amylases from \trep-tomycetes, 7: cuwata, A. haloplanctis, insects and manimals(data not shown), i.e. the alignment of cysteine residueswhen compared with these residues of pig pancreatic a-amy-lase, which is known to contain five disulfide bridges [22].All of these bridges are present probably only in the a-amy-lases of mammals. The results for all a-amylases of thisgroup are summarized in Table 2. There is no evidence thatthe disulfide bonds are really formed, but the unambiguousalignment of the cysteine residues of pig pancreatic a-amy-lase (involved in the five disulfide bridges [22]) with thecysteine residues of the rest of the a-amylases (Fig. 3) sup-port the possibility of the presence of structurally equivalentbonds. Analogical data for fungal and yeast a-amylases arealready implemented in the Swiss-Prot protein sequence databank 1111. As for the plant enzymes, only two invariant cys-teine residues (CyslO6 and Cys126; barley a-amylase num-bering) can be found in their alignment (data not shown).The first cysteine residues could be equivalent to C jsl15,the second residue could be equivalent to the Cysl41 of pigpancreatic a-amylase, both cysteines of the animal enzymebeing involved in two different disulfide bridges [22] (com-pare Table 2). The determination of possible locations of di-sulfide bonds in plant a-amylases should be based, however,on the knowledge of locations of these bonds in one of theseenzymes since plant a-amylases form, like the enzymes fromstreptomycetes and animals, their own homologous group

7/27/2019 ejb_94

5/6

523Table 2. Possible disulfide bridges of a-amylases. Full names of the sources of a-amylases are given in Table 1. The locations of disulfidebridges in the sequence of pig pancreatic a-amylase were taken from [22].a-Amylase Pig pancreatic a-amylase disulfide bond positions

28-86 70-115 141- 60 378-384 450-462possible positions of disulfide bonds in other a-amylase

120- 137 328-335 402 -41 6ltha 20-74 -124- 140 323-330trgr 24- 8 382- 91trhy 24-78 - 122-138 -124- 40 323-330trli 24-78 -Strth 24-78 - -124-140 323-330trvi 24-78 - 133- 49 340-347 411 -423hecu 24-83 -Anoga 28- 8 142-156 362-368 434-446Drome 28- 4 135- 49 358- 64 430-442Muspa 28-86 70-115 141-157 375-381 447-459Mussa 28- 6 70-115 141-160 378-384 450-462Ratpa 28-86 70-115 141-157 375-381 447-459Humpa 28-86 70-115 141- 60 378-384 450-462Humsa 28- 6 70-115 141- 60 378-384 450-462

----

--

(Fig. 2A and C ) .Such information will probably be availablesoon [9].The last remark concerns the three a-amylases that wereexcluded from the studied set of a-amylases (Table 1). Thesea-amylases are the enzymes from Bacillus circulans, Bacillussp. BlOl8 and Clostridium thermosulfurogenes, with the ac-cession numbers in Swiss-Prot protein sequence databaseP08137, P17692 and P26827, respectively [ll]. Based onthe presence of some sequence features characteristic of acyclodextrin glycosyltransferase, including the presence ofthe sequence Phe-Ala-Pro in cyclodextrin glycosyltransfer-ases in the first conserved region [31] and the insertion ofthe glutamine residue in front of the invariant proline in cy-clodextrin glycosyltransferases in the second sequence simi-larity [32], the possibility of an erroneous identification ofthese enzymes seems probable. There is no biochemical evi-dence as yet to support the re-classification of the three a-amylases. When talung into account that the functions ofthese two enzymes are related (the bond cleavage in cyclo-dextrin glycosyltransferases is followed by a transglycosyla-tion step [33]), their biochemical re-evaluation is relevant.The author would like to express his gratitude to D r B. Svensson(Carlsberg Laboratory, Copenhagen) for her valuable comments onthe construction of the evolutionary tree of starch hydrolases andrelated enzymes, and for kindly providing her manuscript prior topublication. The author also thanks two anonym ous referees for theirremarks on the manuscript.

REFERENCES1. MacGregor, E. A . (1988) a-Amylase structure and activity, J.Protein Chem. 7, 399-415.2. Nakajima, R. , Imanaka, T. & Aiba, S . (1986) Comparison ofamino acid sequences of eleven different a-amylases, Appl.Microbcol. Biotechnol. 23, 355- 60.3. JaneEek, S . (1992)New conserved amino acid region of a-amy-lases in the third loop of their (Plu),-barrel dom ains, Biochem.4. Henrissat, B. (1991) A classification of glycosyl hydrolasesbased on amino acid sequence similarities, Biochem. J. 280,309-316.

J . 288. 1069-1070.

5. Raimbaud, E., Buleon, A., Perez, S . & Henrissat, B. (1989)Hydrophobic cluster analysis of the primary sequences of a-amylases, Int. J. Biol. Macromol. l l , 217-225.6. Qian, M. X. , Haser, R. & Payan, F. (1993) Structure and molec.ular model refinement of pig pancreatic a-amylase at 2.1 Aresolution, J. Mol. Biol. 231, 785-799.7. Brady, R. L., Brzozowski, A. M., Derewenda, Z. S., Dodson, E.J. & Dodson, G . G. (1991) Solution of the structure of Asper-gillus niger acid a-amylase by combined molecular replace-ment and multiple isomorphous methods, Acta Crystallogr:Sect. B 47, 527-535.8. Swift, H. J., Brady, L., Derewenda, Z. S . , Dodson, E. J.,Dodson, G. G., Turkenburg, J. P. & Wilkinson, A. J. (1991)Structure and molecular m odel refinement of Aspergillus oryzae (TAKA) a-amylase. An application of the simulated-an-nealing method, Acta Crystullogr. Sect. B 47, 535-544.9. Kadziola, A., Abe, J., Svensson, B. & Haser, R. (1992) Crystalstructure and function of barley malt a-amylase, Abstractfrom the Workshop on Cereal Polysaccharides, Le Croisic,France."10. Janezek, S . & BaliZ, 3. (1993) Evolution of parallel Pla-barrelenzyme family lightened by structural data on starch-proces-sing enzymes, J. Protein Chem. 12, 509-514.11. SwissProt protein sequence database , release 27, October 1993.12. GenBank DNA sequence database, release 80, October 1993.13. Yuuki, T., Nomura, T., Tezuka, H., Tsuboi, A., Yamagata, H.,Tsukagoshi, N. & Udaka, S. (1985) Complete nucleotide se-quence of a gene coding for heat- and pH-stable a-amylase ofBacillus licheniformis. Comparison of the amino acid se-quences of three bacterial liquefying a-amylases deducedfrom the DNA sequences, J. Biochem. (Tokyo) 98 , 1147-1156.14. Itoh, T., Yamashita, I. & Fukui, S. (1987) Nucleotide sequenceof the a-amylase gene ( ALP ] ) n the yeast Saccharomycopsisfibuligera, FEBS Lett. 219, 339-342.15. Hu, N.-T., Hung, M.-N., Huang, A.-M., Tsai, H.-F., Yang, B.-Y., Chow, T.-Y. & Tseng, Y.-H. (1992) Molecular cloning,characterization and nucleotide sequence of the gene forsecreted a-amylase from Xunthomonas campestris pv . cum-pestris, J. Gen. Microbiol. 138, 1647-1655.16. Long, C. M., Virolle, M.-J., Chang, S.-Y., Chang, S. & Bibb,M. J. (1987) a-Amylase gene of Streptomyces limosus. Nucle-otide sequence, expression motifs, and amino acid sequencehomology to mammalian and invertebrate a-amylases, J. Bac-teriol. 169. 5745 -5754.

7/27/2019 ejb_94

6/6

52417. Pet'tizek, M., T ichq, P. & Kuncovi , M. (1992) haracterizationof the a-amylase-encoding gene from Thermomonospora cur-vatu, Gene (Amst.) 112,77-83.18. Huang, N., tebbins, G. L. & Rodriquez, R. L. (1992) lassifi-cation and evolution of a-amylase genes in plants, Proc. Nut1Acad. Sci. USA 89, 7526-7530.19. Rumbak, E., awlings, D. E., Lindsey, G. G. & Woods, D. R.(1991)Cloning, nucleotide sequence, and enzymatic charac-terization of an a-am ylase from the ruminal bacterium Butyri-vibrio jibrisolvens H17c, J. Bacteriol. 173,4203 -4211.20. Saitou, N.& Nei, M. (1987) he neighbour-joining method: Anew method for reconstructing phylogenetic trees, Mol. Biol.Evol. 4, 406-425.21. Higgins, D. ., Bleasby, A. J. & Fuchs, R. (1992) LUSTA L V.Improved software for multiple sequence alignment, Comput.Appl. Biosci. 8 , 189- 91.22. Pasero, L., M azzei-Pierron, Y., Abadie, B., Chicheportiche, Y. &Marchis-Mouren, G. (1986) Complete amino acid sequenceand location of the five disulfide bridges in porcine pancreatica-amylase, Biochim. Biophys. Acta 869, 147- 57.23. Vihinen, M. & Mantsala, P. (1989)Microbial amylolytic en-zymes, Crit. Rev. Biochem. Mol. Biol. 24, 329-418.24. Jespersen, H. M., MacGregor, E. A., Henrissat, B., Sierks, M.R. & Svensson, B. (1993) tarch- and glycogen-debranchingand branching enzymes. Prediction of structural features ofthe catalytic (@&-barrel doma in and evolutionary relation-

ship to other amylolytic enzy mes, J. Protein Chem. 12, 791-805.25. Raha, M. A., Kawagishi, I ., Mueller, V., Kihara, M. & Macnab,R. M. (1992) scherichia coli produces a cytoplasmic a-amy-lase, Amy A, J. Bacteriol. 174, 6644-6652.

26. Svensson, B., espersen, H., Sierks, M. & MacGregor, E. A.(1989) equence homology between putative raw-starch bind-ing domains from different starch-degrading enzymes. Bio-chem. J. 264, 309-311.27. Fukusumi, S., amizono, A., H orinouchi, S. &Beppu, T. (1988)Cloning and nucleotide sequence of a heat-stable amylasegene from an anaerobic thermophile, Dictyoglomus themo-philum, Eul: J. Biochem. 174, 15-21.28. Horinouchi, S.,Fukusumi, S. , Ohshima, T. & Beppu, T. (1988)Cloning and expression in Escherichia coli of two additionalamylase genes of a strictly anaerobic thermophile, Dictyoglo-mus therrnophilum, and their nucleotide sequences wi:h ex-tremely low guanine-plus-cytosine contents, Eur J. Biochern.176, 243-253.29. Stewart, C.-B. (1993)The powers and pitfalls of parsimony,Nature 361, 603-607.

30.Laderman, K. A., Asada, K., Uemori, T., Mukai, H. , T a p c h i ,Y., Kato, I. & Anfinsen, C. B. (1993) -Amylase from thehyperthermophilic archaebacterium Pyrococcus furiosus.Cloning and sequencing of the gene and expression in Ercher-ichiu coli, J. Biol. Chem. 268, 24402-24407.31. MacGregor, E. A. & Svensson, B. (1989) super-secondarystructure predicted to be common to several a - l ,4 -~-g luca n-cleaving enzymes, Biochem. J. 259, 145- 152.32. JaneEek, S. & Tbth, D. (1994)Amino acid sequence alignmentof microbial, plant and animal a-amylases : Some evolutionaryimplications, Biologia (Bratislava) 49, in the press.33. Svensson, B. & Segaard, M. (1993)Mutational analysis of gly-cosylase function, J . Biotechnol. 29, 1-37.