Cloning and nucleotide sequence of the gene (amyP) for maltotetraose-forming amylase from Pseudomonas stutzeri MO-19

The gene (amyP) coding for maltotetraose-forming amylase (exo-maltotetraohydrolase) of Pseudomonas stutzeri MO-19 was cloned. Its nucleotide sequence contained an open reading frame coding for a precursor (547 amino acid residues) of secreted amylase. The precursor had a signal peptide of 21 amino acid residues at its amino terminus. An extract of Escherichia coli carrying the cloned amyP had amylolytic activity with the same mode of action as the extracellular exo-maltotetraohydrolase obtained from P. stutzeri MO-19. A region in the primary structure of this amylase showed homology with those of other amylases of both procaryotic and eucaryotic origins. The minimum 5' noncoding region necessary for the expression of amyP in E. coli was determined, and the sequence of this region was compared with those of Pseudomonas promoters.

Under appropriate culture conditions, Pseudomonas stutzeri produces an extracellular amylase that forms maltotetraose (EC 3.2.1.60, exo-maltotetraohydrolase; abbreviated as G4-amylase) (21). This enzyme is a unique amylase because it catalyzes the release of a-anomeric oligosaccharide (ot-maltotetraose) exoglycolytically from the nonreducing ends of starch (23), whereas other exo-type amylases (glucoamylase and ,-amylase) release P-anomeric products by exoglycolytic cleavage, and ox-amylases hydrolyze starch endoglycolytically to produce ot-malto-oligosaccharides. Thus, the G4-amylase has unique activity for hydrolysis of starch intermediate between those of (x-amylases and Por gluco-amylase; G4-amylase produces a similar product to a-amylases but hydrolyzes starch exoglycolytically like Por gluco-amylase. Robyt and Ackerman (21) reported that the G4-amylase of P. stutzeri consists of isozymes. Multiple forms of this enzyme (seven isozymes) varying in molecular weight and isoelectric point were also reported by Schmidt and John (26). Sakano et al. (23,24) purified two forms of enzyme (F-1 and F-2), each giving a single band on polyacrylamide gel electrophoresis with or without sodium dodecyl sulfate. The enzymatic properties of F-1 and F-2 were the same, but their isoelectric points were different. The reason for the multiple forms of the enzyme is still unknown.
G4-amylase is commercially important for producing maltotetraose, which is used as a substrate of amylases in studies on their mode of action and as a highly sensitive substrate for detection of ot-amylase when coupled with a chromogenic compound. Maltotetraose is also being tested for use as a food additive to improve the texture and moisture retention of foods.
As an initial step in understanding the molecular basis of the unique action mechanism of G4-amylase and the genetic basis of its multiple forms and also for elucidating the regulation of the synthesis of this industrially important enzyme, we cloned and sequenced the G4-amylase gene (amyP) from P. stutzeri. Then, we compared its amino acid sequence with those of other amylases of both procaryotic and eucaryotic origins.
Purification of G4-amylase produced by P. stutzeri. P. stutzeri MO-19 was grown for 20 h at 30°C with vigorous aeration in 2.5 liters of medium. Cells were separated from the culture fluid by centrifugation. Solid (NH4)2SO4 was added to the fluid to 20% saturation, and the mixture was left to stand for 16 h at 4°C. The precipitate was removed by centrifugation (10,000 x g, 20 min), and the supernatant was then brought to 40% saturation of (NH4)2SO4 and left to stand for 16 h at 4°C. The resulting precipitate was dissolved in 10 mM phosphate buffer (pH 7.0) and dialyzed against the same buffer at 4°C. Insoluble material that appeared was removed by centrifugation, and the supernatant was dialyzed against 10 mM Tris hydrochloride (pH 8.0) at 4°C. The dialyzed enzyme was applied to a column of DEAE-Toyopeal 650S (Toyo Soda, Tokyo, Japan), and material was eluted with a linear gradient of 0 to 0.5 M NaCl. G4-amylase was eluted with a concentration of between 0.1 and 0.2 M NaCl. The purity of each fraction containing enzyme activity was checked by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (14), and the fractions giving a single protein band were pooled and used for further studies.
Amino acid sequence analysis of the purified G4-amylase.  The NH2-terminal amino acid sequence of the purified G4-amylase was determined by step-wise Edman degradation of 100 ,ug of the enzyme in an Applied Biosystems protein sequencer 470A fitted with a Senshu Pak AQUASIL SEQ-4(K) column (Senshu Scientific, Tokyo, Japan).
Preparation of chromosomal DNA from P. stutzeri and cloning of the amyP gene for G4-amylase. Chromosomal DNA of P. stutzeri MO-19 was prepared (22), partially digested with restriction enzyme Sau3AI, and ligated to pBR322 DNA which had been treated with BamHI and bacterial alkaline phosphatase. The ligated DNA was used to transform E. coli HB101 by the conventional CaCl2 procedure. Transformants were selected on LB plates supplemented with 50 jig of ampicillin per ml and 0.5% soluble starch. G4-amylase-positive clones were detected by staining the plates with 10 mM KI-42 solution; the positive colonies had a semitransparent halo. DNA manipulations were done essentially as described by Maniatis et al. (15). Plasmids were isolated by the alkaline lysis procedure (1). Enzymes were obtained from commercial sources and used according to the recommendations of the suppliers.
Assay of G4-amylase activity. A reaction mixture containing 0.2 ml of enzyme solution and 5 ml of 1% soluble starch buffered with 20 mM phosphate buffer (pH 7.0) was incubated for 20 min at 40°C. The reaction was stopped by mixing 1 ml of the reaction mixture with 2 ml of Somogyi reagent. Increase of reducing power was determined by the Somogyi-Nelson method (29). One unit of enzyme was defined as the activity forming 1 ,umol of maltotetraose from soluble starch in 1 min.
Analysis of hydrolysis products from starch. A mixture of S ml of 5% soluble starch buffered with 20 mM phosphate buffer (pH 7.0) and 1 ml of enzyme solution (1 U) was incubated for 1 h at 40°C. Then an aliquot of the mixture was spotted onto Toyo filter paper (no. 50; Toyo Roshi, Tokyo, Japan). Paper chromatography was carried out by the ascending technique with a solvent system of n-butanol-pyridine-water (6:4:3) at 60°C. Then the chromatogram was dried and dipped in silver nitrate solution (32).
Southern blot hybridization. Chromosomal DNA was digested with Sall and subjected to electrophoresis on a 1% agarose gel. The fragments separated were located by staining with ethidium bromide (0.5 ,ug/ml) and were transferred to a Zeta-probe membrane (Bio-Rad Japan, Tokyo, Japan). As a probe, the 3.0-kilobase-pair (kb) SalI DNA fragment prepared from pTPS618 was labeled with [ct-32P]dCTP (3,000 Ci/mmol) using a nick translation kit from Takara Shuzo (Kyoto, Japan). Hybridization was carried out as previously described (30).

RESULTS
Cloning of amyP in E. coli. A library of P. stutzeri MO-19 DNA was constructed in E. coli HB101 by using pBR322, and about 18,000 transformants were screened for G4-amylase activity as described in Materials and Methods. One G4-amylase-positive clone was obtained from this library. The recombinant plasmid from the positive clone was designated pTPS6. The restriction map of the 7.2-kb insert in pTPS6 is shown in Fig. 1. pTPS6 DNA was partially digested with SalI to reduce the size of the insert and then ligated and used to transform E. coli HB101. Several deletion plasmids conferring the G4-amylase-positive character were obtained; one of these plasmids, designated pTPS618, lacked the C and D regions (Fig. 1). Thus, the amyP was found to be included in the A and B regions.
somal DNA of P. stutzeri MO-19 digested with Sall, whereas no hybridization band was detected with Sall-digested chromosomal DNA from E. coli HB101 (data not shown). The probe also hybridized with the same-size fragment from Sall-digested chromosomal DNA of P. stutzeri NRRL B-3389, which has been used in studies of G4-amylase by Robyt and Ackerman (21), Schmidt and John (26), and Sakano et al. (23,24). Thus, these two P. stutzeri strains seem to have homologous enzyme.
Starch hydrolysis with G4-amylase produced by E. coli. Paper chromatography was used to determine whether the starch-hydrolyzing activity conferred by pTPS618 was due to the synthesis of G4-amylase. A cell extract prepared by sonication of E. coli HB101 harboring pTPS618 was used as the crude enzyme because no amylolytic activity was detectable in the culture medium. The crude enzyme solution was incubated with soluble starch, and the products were analyzed by paper chromatography. The crude enzyme formed maltotetraose from soluble starch (Fig. 2), indicating that the cell extract from E. coli HB101 harboring pTPS618 contained G4-amylase.
DNA sequence of amyP. The nucleotide sequence of both strands of the 2.1-kb Sau3AI-Eco47III segment (Fig. 1) was determined. The sequence contains a single open reading frame (1,641 base pairs) that encodes a protein of 547 amino acids corresponding to the amyP product (Fig. 3). A protein with a calculated molecular weight of 59,736 could be translated from this open reading frame. No in-frame initiation codon could be found upstream of the open reading frame. The sequence GGAG, found 8-bp upstream of the first letter of the initiation codon is presumably the ribosome-binding site (27).
Determination of the NH2-terminal amino acid sequence. The sequence of the 20 NH2-terminal amino acids of the extracellular form of G4-amylase from P. stutzeri MO-19 was determined with a peptide sequencer to be Asp-Gln-Ala-Gly-Lys-Ser-Pro-Asn-Ala-Val-Arg-Tyr-His-Gly-Gly-Asp-Glu-Ile-Ile-Leu. This sequence was identical to that deduced from the DNA sequence starting at position 64 (Fig. 3). These results suggest that the first 21 amino acid residues (Met-1 to Ala-21) constitute a signal peptide involved in secretion of the protein. The molecular weight of the mature G4-amylase calculated from the DNA sequence was 57,547, which is consistent with the value of 58,000 obtained by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the purified enzyme. The amino acid composition calculated from the deduced amino acid sequence was essentially consistent with that of the purified enzyme (data not shown).
Homology of G4-amylase with other a-amylases. The G4amylase sequence was compared with the primary structures of a-amylases from eucaryotes and procaryotes. Four regions (I to IV) which are conserved in cx-amylases (19) were also found in G4-amylase (Fig. 4). The homology of region III is limited, although the spacings of the four regions in the sequences of G4-amylase and a-amylases are similar. However, no homology was found between G4-amylase and ,- (13) or gluco-amylase (12; data not shown).
Construction and characterization of deletion plasmids of the amyP promoter region. Plasmid pTPS618 was digested with Eco47III, which has one site in the vector pBR322, ligated with pUC18 cleaved by SmaI, and then used to transform E. coli HB101. One of the G4-amylase-positive clones was selected. Restriction analysis showed that this plasmid, designated pSE86R, carried the 2.2-kb Eco47III fragment containing amyP. The amyP gene is in the opposite orientation to the lac promoter on pUC18 (Fig. 5). The 2.2-kb Eco47III fragment contains a 119-bp Eco47III-BamHI portion derived from pBR322. This 119-bp portion does not contain the promoter, since this portion is the middle part of the Tetr gene of pBR322. Thus, these results suggest that the amyP is expressed in E. coli by its own promoter.
To determine the location of the promoter, we constructed plasmids having a deletion in the 5' noncoding region of amyP. pSE86R was linearized with BamHI and PstI and digested with exonuclease III to produce various extents of deletion of the amyP promoter region. After treatment with mung bean nuclease and Klenow enzyme, the resultant blunt-ended fragments were ligated, and the ligated products were used to transform E. coli HB101. The extents of deletion were determined by estimating the sizes of fragments generated by EcoRI and HindlIl by 1% agarose gel electrophoresis. Five deletion plasmids were selected, and their G4-amylase activities in E. coli were assayed as a function of the amyP promoter (Fig. 5). The deletion endpoint of each plasmid was determined by DNA sequencing. The results revealed that the minimum 5' noncoding region required for G4-amylase activity in E. coli is 77 nucleotides upstream from the initiation codon; this 77-nucleotide region is essential for the function of amyP promoter in E. coli. The endpoint of the region is located within an inverted repeat sequence (Fig. 3).

DISCUSSION
We have cloned and analyzed the DNA fragment encoding the G4-amylase structural gene (amyP) from P. stutzeri MO-19. Its nucleotide sequence contained a single open reading frame of 1,641 bp, beginning from an ATG initiation codon at position 1 and ending with a TAG termination codon at position 1642 (Fig. 3). The deduced amino acid sequence corresponding to nucleotide positions 64 to 123 was identical to that of the NH2-terminal 20  Nucleotide sequence of amyP and its amino acid translation product. Only the sequence of the antisense strand is shown. The numbering of nucleotides begins at the first letter of the potential initiation codon. The cleavage site between the signal peptide and extracellular mature G4-amylase is indicated by the upward arrow. The stem-and-loop structures upstream and downstream from the structural gene are indicated by horizontal arrows below the sequence. The sequence matching that obtained by Edman degradation of the NH2-terminal region of the purified G4-amylase is underlined. A putative promoter is indicated by an overline. The line marked SD indicates the proposed Shine-Dalgarno (ribosome-binding) sequence. The upstream endpoint of the deletion, which is the minimum potential region for the promoter, is indicated by the downward arrow. The four sequences (I to IV) homologous to those of a-amylases are boxed. peptide, which is removed during secretion of the enzyme. The deduced amino acid sequence of this region, which has several positively charged amino acids near the NH2 terminus followed by a hydrophobic amino acid core and Ala at ACGCCAAGCTTGCATGCCTGCAGGTCGACTCTAGAGGATCCCCGGGTACCGAGCTCGMATCACTG the COOH-terminal end, is similar to that typical of signal peptides of other secreted proteins (28,33,34).
We compared the primary structure of G4-amylase with those of a-amylases from eucaryotes and procaryotes and found that G4-amylase has regions homologous to the four regions (I to IV) conserved in the primary structures of a-amylases (19; Fig. 4). A molecular model of Taka-amylase (ot-amylase from Aspergillus oryzae) proposed by Matsuura et al. (16) suggests that Glu-230 and Asp-297 in regions III and IV of this enzyme, respectively (Fig. 4), function as the catalytic residues. Corresponding residues, Glu-240 and Asp-314 were found in regions III and IV, respectively, of G4-amylase. These common regions may contain the amino acid residues responsible for cleaving 1,4-a-D-glucosidic linkages. However, G4-amylase is unique in that it has Arg-217 and Gly-218 in region II; all other a-amylases studied except barley a-amylase, have Lys and His in this position (19). These amino acid residues found only in region II of G4-amylase may be related to the unique action of the enzyme in producing a-maltotetraose from starch by an exomechanism. We mentioned previously that the mode of hydrolysis of starch by G4-amylase is intermediate between those of a-amylases and Por gluco-amylase. However since no homologous regions were found in the primary structures of G4-amylase and ,Bor gluco-amylase, the evolutionary origin and catalytic mechanism of G4-amylase are closer to those of a-amylases than to those of Por gluco-amylase. FIG. 5. G4-amylase activity in E. coli as a function of the amyP promoter of pSE86R and its deletion derivatives. A 2.2-kb Eco47III DNA fragment containing amyP inserted into the SmaI site of pUC18 was designated pSE86R. The upper part of the figure shows the multiple cloning sites of pUC18. lacP indicates the lacZ promoter. The construction of deletions in the 5' noncoding region of amyP is described in the text. 0 and *, Noncoding and coding regions for G4-amylase, respectively; 0, 119-bp Eco47III-BamHI portion derived from pBR322; N and C, NH2 and COOH termini of G4-amylase, respectively. Numbers on the left indicate the nucleotide positions of deletion endpoints numbered from the first letter of the initiation Met codon. Crude cell extracts of E. coli harboring the plasmids were prepared, and their G4-amylase activities were assayed as described in Materials and Methods. Activities are shown on the right in milliunits per milliliter of culture. stutzeri; P. aer., Pseudomonas aeruginosa; P. put., Pseudomonas putida; P. sp., Pseudomonas sp. The sequences homologous with the consensus sequence of ntrlnif promoters are underlined. All sequences except that for amyP are cited from Deretic et al. (2). Multiplicities in molecular weight or isoelectric point of G4-amylase have been reported by others (21,23,24,26). We also observed two forms of the enzyme purified from our strain by gel electrophoresis (unpublished data). These multiplicities may be ascribed to limited proteolysis during culture of the organisms or purification of the enzyme, because Southern blot hybridization revealed that the G4amylase gene (amyP) exists in a single locus of chromosomal DNA as a single copy and G4-amylase is primarily synthesized as a single polypeptide. However, the possibility that other nonhomologous genes for G4-amylase exist cannot be ruled out.
The minimum 5' noncoding region necessary for the expression of amyP in E. coli has been mapped by deletion analysis. This region was within 77-bp upstream from the initiation codon. Several characteristics of this region can be pointed out. (i) It contains an inverted repeat which has a high potential free energy -17.2 kcal [ca. -72 kJ]/mol as calculated by the method of Tinoco et al. (31; Fig. 3). (ii) It does not show extensive homology to the consensus sequence recognized by the RNA polymerase holoenzyme Eu70 in E. coli (17,20). (iii) It has weak homology to the consensus sequence for ntrlnif promoters recognized by Eu54 in E. coli (6,7) and the sequences of several Pseudomonas promoters found to have homology with the consensus sequence (2) (Fig. 6). These observations suggest that this region may act as a native promoter in P. stutzeri MO-19 and may be involved in regulation of gene expression.
Determination of the transcription start site in P. stutzeri is necessary in order to clarify these suggestions. The total G4-amylase activity in E. coli is 0.05% of that in P. stutzeri (data not shown). Since the nucleotide sequence preceding the initiation codon for amyP is complementary to the 3' end of the 16S rRNA not only in Pseudomonas aeruginosa but also in E. coli (27), the mRNA of amyP is expected to be translated in E. coli. Therefore, the weak expression of amyP in E. coli may be due to inefficient recognition of the amyP promoter by E. coli RNA polymerase. A similar reason for the weak expression of Pseudomonas genes in E. coli has been suggested by others (3,4,(8)(9)(10)(11). There have been few studies on Pseudomonas RNA polymerase, and further work is needed to elucidate its interaction with the promoter in Pseudomonas spp.