Journal of Bioscience and BioengineeringVOL. 107 No. 2, 124–129, 2009

www.elsevier.com/locate/jbiosc

Characterization of the C-terminal truncated form of amylopullulanasefrom Lactobacillus plantarum L137

Jong-Hyun Kim,1 Michihiro Sunako,2 Hisayo Ono,1 Yoshikatsu Murooka,1,§

Eiichiro Fukusaki,1 and Mitsuo Yamashita1,⁎,§§

⁎ CorrespondE-mail add

§ HiroshimaJapan.§§ DepartmenTechnology, 3-

1389-1723/$doi:10.1016/j.

Department of Biotechnology, Graduate School of Engineering, Osaka University, Yamadaoka 2-1, Suita, Osaka 565-0871, Japan,1 and Sanwa Cornstarch Co.,Ltd., 594 Unate, Kashihara, Nara 634-8585, Japan2

Received 22 July 2008; accepted 17 October 2008

A gene (apuA) encoding amylopullulanase from a starch-hydrolyzing lactic acid bacterium, Lactobacillus plantarum L137,which had been isolated from traditional fermented food made from fish and rice in the Philippines, was found to containtwo unique amino acid repeating units in the N- and C-terminal region. The former is a six amino acid sequence (Asp-Ala/Thr-Ala-Asn-Ser-Thr) repeated 39 times, and the latter is a three amino acid sequence (Gln-Pro-Thr) repeated 50 times. Toclarify the role of these repeating units, a truncated apuA in the C-terminal region was constructed and expressed in L.plantarum NCL21, which is the ApuA− derivative of strain L137. The recombinant truncated amylopullulanase (ApuAΔ),which lacks the 24 kDa of the C-terminal repeat region, was purified and characterized, and compared with wild-typeamylopullulanase (ApuA). The enzyme production and specific activity of ApuAΔ were higher than those of ApuA. The twoenzymes, ApuA and ApuAΔ, showed similar pH (4.0–4.5) and temperature (40–45 °C) optima. However, the activity ofApuAΔ was more stable in the pH and temperature than that of ApuA. The catalytic efficiencies of ApuAΔ toward solublestarch, pullulan and amylose were higher than those of ApuA, although their substrate specificities towards saccharideswere similar. From these results, we conclude that the C-terminal repeating region of ApuA is negatively involved in thestability of amylopullulanase and binding of substrates. Thus, the truncated amylopullulanase is more useful in processingof amylose and pullulan.

© 2008, The Society for Biotechnology, Japan. All rights reserved.

[Key words: Amylopullulanase; Truncated forms; Repeating amino acid sequence; Stability; Catalytic efficiency]

Amylolytic enzymes degrade starch by hydrolyzing α-1,4- and/or α-1,6-glucosidic linkages. They can differ greatly in theirspecificity and properties depending on their source (1). In thelast decade there have been reports of novel enzymes possessingboth starch- and pullulan-hydrolyzing activities (2, 3). The enzymeshydrolyzing α-1,4- and α-1,6-glucosidic linkages in starch andpullulan are generally referred to as amylopullulanases (E.C.3.2.1.41) (2, 3). Amylopullulanase can be further divided into twosubgroups based on the number of active sites within the protein(3). Most amylopullulanases from the thermophilic anaerobespossess one active site (4–7). On the other hand, amylopullulanasesfrom aerobic microorganisms contain either one (8, 9) or twoactive sites (10, 11).

In recent years, a considerable number of amylopullulanases havebeen isolated from awide variety of microorganisms (5, 8, 12–14). The

ing author. Tel.: +81 (0) 3 5859 8156; fax: +81 (0) 3 5859 8101.ress: yamashi@sic.shibaura-it.ac.jp (M. Yamashita).Institute of Technology, 2-1-1 Miyake, Saeki-ku, Hiroshima 731-5193,

t of Applied Chemistry, Faculty of Engineering, Shibaura Institute of7-5 Toyosu, Koto-ku, Tokyo, 135-8548, Japan.

- see front matter © 2008, The Society for Biotechnology, Japan. Alljbiosc.2008.10.019

amylopullulanases from the lactic acid bacteria (LAB), such as Lacto-bacillus amylophilus (15) and Bifidobacterium breve (16), have beenisolated, purified and characterized.

Quite recently, we cloned, sequenced, and expressed a gene (apuA)encoding amylopullulanase from a plant-originated lactic acid bacte-rium, Lactobacillus plantarum L137 (17). The apuA gene has twounique amino acid repeating units in the N- and C-terminal region.The repeat sequence of the N-terminal region of apuA did not showsimilarity to any sequence in the GenBank database.

However, the repeated amino acid sequences of the C-terminalregion exhibited 54% and 50% similarities to a bacterial peptido-glycan-bound protein from Lactobacillus johnsonii NCC 533 (proteinID: AAS09580) and a cell surface protein from Aspergillus fumigatusAf293 (protein ID: EAL92679), respectively. Repeating units in theN- and C-terminal region have been found in amylopullulanasesfrom Bacillus sp. (8, 11), but the functions have been poorlycharacterized.

In the present work, the functional role of the C-terminalrepeat sequences of amylopullulanase (ApuA) from L. plantarumL137 is investigated using a recombinant clone carrying a gene thatencodes a truncated amylopullulanase (ApuAΔ), in which the C-terminal repeat sequences of wild-type ApuA are deleted.

rights reserved.

FIG. 1. Construction of the plasmids pRN-PulSa and pRN-PulΔ3. Plasmid pRN-PulΔ3 was derived from pRN-PulSa after deletion of the 450-bp C-terminal repeat region. ss and repeatunits indicate signal amino acid sequences and amino acid repeating units of the N- and C-terminal regions, respectively.

CHARACTERIZATION OF TRUNCATED AMYLOPULLULANASE 125VOL. 107, 2009

Comparison of the enzymatic properties of the two amylopullula-nases is described.

MATERIALS AND METHODS

Bacterial strains, plasmids, and culture conditions L. plantarum L137 wasisolated from a traditional fermented food prepared in the Philippines (18), and itsderivative strain NCL21 was constructed by curing plasmids with novobiocin (Sigma-Aldrich, St. Louis, MO, USA) as described previously (19). Escherichia coli DH5αwas usedas a host for DNA manipulation and transformation to construct vector DNAs. DeMan-Rogosa-Sharpe (MRS) medium (Difco, Detroit, MI, USA) was routinely used for culti-vation of Lactobacillus strains at 30 °C. For amylopullulanase production, cells weregrown with 1% (w/v) soluble starch (Wako, Osaka) as a carbon source. For detection ofthe starch hydrolyzing activity of bacterial colonies, a BS plate (blue starch agar plate)was prepared by adding 1% (w/v) starch and 0.2% (w/v) starch azure (Sigma-Aldrich) tothe MRS agar plate. E. coli was routinely grown at 37 °C in Luria–Bertani (LB) medium(20). When necessary, the media were added with ampicillin (Ap) (50 μg/ml) for E. coli,or erythromycin (Em) (2.5 μg/ml) for L. plantarum.

DNA manipulation and transformation Plasmid DNA preparation from E. coli,and restriction enzyme digestion, ligation, and transformation of E. coliwere carried outas described in Sambrook et al. (20).

Plasmid DNAs, pBluescriptII KS+ (Stratagene, La Jolla, CA, USA) and pGEM-T Easy(Promega Corp., Tokyo), were used to delete the target region from the apuA gene ofL. plantarum L137.

One ml of an overnight culture of the Lactobacillus strain in MRS broth wasinoculated into 3 ml fresh MRS broth and incubated at 30 °C for 2 h. Plasmid DNAwas extracted from the cells to check the plasmid profiles as described by Andersonand McKay (21), with the exception that 10 μg/ml mutanolysin (N-acetylmuramidaseSG; Seikagaku Kyogyo, Tokyo) was used in cell lysis. Plasmid DNA from E. coli wasisolated by the alkaline lysis method (20) or using a Quantum Prep plasmid miniprepKit (Bio-Rad Laboratories, Tokyo). The method of Hanahan was used to preparecompetent cells of E. coli for transformation (22). Other recombinant DNA techniqueswere carried out as described by Sambrook et al. (20). Transformation of L. plantarumwas performed as described previously (23). The transformants were selected byadding Em to the solid culture medium. The production of amylopullulanase with therecombinant clones was detected by visualization of halos due to starch degradationon the BS plates.

Construction of the truncated apuA gene The pRN-PulSa plasmid carrying thefull-length amylopullulanase gene (apuA) (accession no. AB369265) from L. plantarumL137 and the pBS-PulNcore plasmid carrying the 3.5-kb region containing theC-terminal region of apuA in the pBluescriptII KS+ vector were constructed previously(17) and used as the template for the gene truncation experiment. A pair of specificprimers, Pul_Ncofo and Pul_3del, which are 5′-CCATGGCAATTATTGGCCAGGTAG-TAATCT-3′ and 5′-GATGGGCACTTCAGGGCATTACTAATGAA-3′, respectively, weredesigned to amplify double-stranded DNAs of the truncated apuA gene, whichcorresponded to the amino acid sequence from the N-terminal residues Thr943 toThr1839 at the C-terminus. The amplified truncated apuA gene was subcloned intopGEM-T Easy vector and confirmed that no point mutations occurred during the PCR

TABLE 1. Purification of amylopullulanases from L. plan

Plasmid (enzyme) Purification step Total protein (mg) Total

pRN-PulSa (ApuA) Culture supernatant –a

Ultrafiltration (100 kDa cut-off) 7.5DEAE Sepharose CL-4B 3.9Superose 6 1.6

pRN-PulΔ3 (ApuAΔ) Culture supernatant –a

Ultrafiltration (100 kDa cut-off) 9.7DEAE Sepharose CL-4B 5.2Superose 6 2.4

a Not shown because of the high background protein content resulting from the MRS+1%

cloning by DNA sequencing. The amplified DNA fragment of approximately 2.9-kb ofthe N-terminal region of the apuA gene was purified by gel electrophoresis and thensubcloned into pGEM-T Easy vector. The resulting plasmid containing the 2.9-kb DNAfragment was restricted with NcoI/NotI, and ligated with NcoI/NotI-digested pRN-PulSa plasmid (17), resulting in pRN-PulΔ3. Finally, the recombinant plasmid con-taining the truncated apuA gene, pRN-PulΔ3, was then transformed into L. plantarumNCL21 by electroporation (23).

Enzyme assay α-Amylase and pullulanase activities were measured at 40 °C ina 0.2 ml reaction mixture that contained 0.1 ml of a 1% (w/v) solution of soluble starch(Wako) or pullulan (Mr 99,800; Hayashibara Biochemical Laboratories Inc., Okayama) in0.1M sodium acetate buffer (pH 4.5) and 0.1 ml of a suitably diluted solution of enzyme.The reducing sugar formed was quantified by the DNS (dinitrosalicylic acid) method(24). One unit (U) of enzymatic activity was defined as the amount of protein thatproduced 1 pmol/min of reducing sugar as glucose under the conditions of the assay.

The products from soluble starch, amylose (Nakano Vinegar Co., Ltd., Handa) andpullulan (Shodex standard P-82; Showa Denko K.K., Tokyo) hydrolyzed by the wild-type and truncated amylopullulanase, were examined with HPLC using a SugarSZ5532 column (Showa Denko K.K.) connected to a Shodex RI SE-6I detector (ShowaDenko K.K.). The products were eluted with a mixture of acetonitrile:water (70:30:v/v)at a flow rate of 1 ml per min and a column temperature of 20 °C.

The apparent Michaelis constant (Km) of each enzyme was determined at 10different concentrations of soluble starch, amylose and pullulan (from 0 to 40 g/l) attheir temperature and pH optima. Kinetic parameters were calculated by fitting initialvelocities and substrate concentrations to the Michaelis–Menten equation using theHanes–Woolf plot method (25).

Purification of wild-type and truncated amylopullulanases Cells of a 1-lculture of L. plantarum NCL21 carrying pRN-PulSa and pRN-PulΔ3 left standing at 30 °Cfor 48 h in MRS-starch (1%) mediumwere centrifuged at 8000 ×g for 15 min at 4 °C. Thesupernatant obtained was used as a crude enzyme solution. Wild-type and mutantamylopullulanases were purified as described previously (17).

Electrophoresis and active staining The concentration of protein was deter-mined by measuring the absorbance at 280 nm or by the Lowry method (26) usingbovine serum albumin as the standard. SDS-PAGE was carried out with an 8%polyacrylamide gel by the method of Laemmli (27) using a Compact PAGE twin (ATTO,Tokyo). A prestained protein molecular marker was purchased from Nippon Bio-RadLaboratories, and protein bandswere detected by staining with Coomassie Brilliant BlueR-250 (CBB). For active staining of the amylopullulanase band, soluble starch was addedin an 8% SDS-polyacrylamide gel at a final concentration of 1% (w/v). After electro-phoresis, the gel was immersed and equilibrated in 0.1 M acetate buffer (pH 4.5) for 2 h.To detect the active band, the gel was stained by soaking in 0.2% (w/v) I2 /2% (w/v) KIsolution.

Characterization of recombinant amylopullulanases from L. plantarumNCL21 The amylase activity was determined at various pH values (2.5–8.0) and atvarious temperatures (20–60 °C) in 0.1 M sodium acetate buffer (pH 4.5). The effect oftemperature was examined under standard conditions (0.1 M sodium acetate, pH 4.5)by varying the temperature (20–60 °C). The optimal pH was determined under severaldifferent pH values using 0.1 M glycine-HCl (pH 2.5–4.0), sodium acetate (pH 3.5–6.0)and sodium phosphate (pH 6.0–8.0). To determine the thermostabilities of recombinantenzymes, samples were preincubated in 0.1 M sodium acetate buffer at optimum pH

tarum NCL21 harboring pRN-PulSa and pRN-PulΔ3

activity (U) Specific activity (U/mg protein) Purification (fold) Yield (%)

2800 –a –a 1002100 280 1 751390 356 1.3 50690 431 1.5 25

4960 –a –a 1003840 396 1 772460 473 1.2 501430 596 1.5 29

soluble starch medium.

FIG. 2. Effects of pH on the activity and stability of the ApuA (closed circles) and ApuAΔ(open circles). (A) The optimum pH for ApuA and ApuAΔ. The optimum pH curve wasdetermined at 40 °C at various pH values in 0.1 M buffers (glycine-HCl, pH 2.5–4;acetate, pH 3.5–6; phosphate, pH 6–8), with 0.2 μg enzyme, and soluble starch as asubstrate. The values are shown as percentages of the maximum activities, which weretaken as 100%. (B) The pH stability for the ApuA and ApuAΔ activities. The purifiedenzyme (0.2 μg) was incubated for 1 h at 40 °C in 0.1 M of each buffer, and then sampleswere used for the measurements of the residual activity under the standard conditionsof the enzymatic assay. The values are shown as percentages of the original activity,which was taken as 100%.

FIG. 3. Effect of temperature on the activity and stability of the ApuA (closed circles) andApuAΔ (open circles). (A) Optimum temperatures of the ApuA and ApuAΔ activities.The enzymes were assayed at the indicated temperatures and in 0.1 M acetate buffer(pH 4.5) with 0.2 μg enzyme and starch as a substrate. The values shown arepercentages of the maximum activities, which were taken as 100%. (B) Thermostabilityof the enzyme. The enzyme (0.2 μg) was preincubated at the indicated temperatures in1.0 ml of 0.1 M acetate buffer (pH 4.5) for 1 h. Aliquots (0.1 ml) were withdrawn and theresidual ApuA and ApuAΔ activities were measured at 40 °C in 0.1 M acetate buffer (pH4.5). The activity of the untreated enzyme (0.2 μg) was taken as 100%.

126 KIM ET AL. J. BIOSCI. BIOENG.,

under several temperatures (20–60 °C) for 1 h. These samples were then put on ice forat least 30 min, and the residual activity was determined at pH 4.5 at 40 °C, using theDNS method as described above.

RESULTS

Amylopullulanase production Constructs of the apuA gene(Fig. 1) were introduced by electroporation into L. plantarum NCL21,and the resulting transformants, NCL21 harboring pRN-PulSa andpRN-PulΔ3, were selected for their resistance to Em. The colonies ofEmr were replicated on BS medium showing starch hydrolyzingactivity to examine the production of amylopullulanase. The largesthalo-forming cell was selected (data not shown). The total plasmidDNA from selected transformants, NCL21 harboring pRN-PulSa andpRN-PulΔ3, was extracted and used to retransform E. coli cells.Plasmid DNAs, pRN-PulSa and pRN-PulΔ3, from recombinant cells of

E. coli were confirmed to be the same constructs from L. plantarumNCL21 transformants by restriction enzyme profile and DNA sequen-cing (data not shown). L. plantarum NCL21 carrying pRN-PulΔ3 orpRN-PulSa was grown for 24 h on various carbon sources, solublestarch, lactose, and glucose. The truncated amylopullulanase from therecombinant NCL21 harboring pRN-PulΔ3 showed a higher amylo-pullulanase activity on soluble starch than that produced from NCL21harboring pRN-PulSa (Table 1), while glucose decreased the produc-tion of both amylopullulanases as described previously (17). To purifyamylopullulanase cultures from both recombinant NCL21 strains, cellswere harvested in the late logarithmic growth phase.

Purification of the recombinant ApuA and ApuAΔ Wild-typeand truncated amylopullulanases were purified as described in theMaterials and methods section. The purification step and totalrecovery of the two enzymes are summarized in Table 1. Approxi-mately 2800 U per liter of amylolytic activity was obtained in the

TABLE 2. Substrate specificity of amylopullulanases

Substrate (1% (w/v)) Relative activity (%)

ApuA ApuAΔ

Soluble starch 100 100Amylopectin 106 118Amylose 76 87Pullulan 70 83Glycogen 46 51Cyclodextrins 0 0Dextran 0 0

Assays were performed at 40 °C in 0.1 M sodium acetate buffer (pH 4.5) with 0.2 μg ofenzyme under the standard reaction conditions.

TABLE 4. Effects of various metal ions on amylase and pullulanase activities ofamylopullulanases

Metal Finalconcentration

(mM)

Relative residual activity (%)

ApuA ApuAΔ

Amylasea Pullulanaseb Amylasea Pullulanaseb

None 100 100 100 100NaCl 50 mM 107 95 103 94KCl 50 mM 112 99 106 93CaCl2 1 mM 111 101 105 94CdCl2 1 mM 83 94 71 84CoCl2 1 mM 150 140 128 130CuCl2 1 mM 60 53 62 60FeCl2 1 mM 97 65 83 62HgCl2 1 mM 0 0 0 0MnCl2 1 mM 133 133 109 121NiCl2 1 mM 116 112 87 100ZnCl2 1 mM 99 99 84 98

Each enzymatic activity was assayed after treatment of enzymes with eachmetal ion for1 h at 30 °C in 0.1 M sodium acetate buffer (pH 4.5). a and b indicate the activity ofhydrolysis for soluble starch and pullulan, respectively.

CHARACTERIZATION OF TRUNCATED AMYLOPULLULANASE 127VOL. 107, 2009

culture supernatant from the recombinant strain NCL21 harboringpRN-PulSa. At the final purification step amylopullulanase wasobtained with a 25% overall yield and specific activity of 431 U permg protein. In the culture supernatant from the recombinant strainNCL21 harboring pRN-PulΔ3, approximately 4960 U per liter ofamylolytic activity was obtained. At the final purification step thetruncated amylopullulanasewas obtainedwith a 29% overall yield andspecific activity of 596 U per mg protein. This result means that thetruncated enzyme has 1.4-fold higher activity than that of the wild-type ApuA. The purified enzyme of ApuAΔ showed smear proteinbands of 150 to 250 kDa on SDS-PAGE by staining with CBB (data notshown), similar to those seen for the wild-type ApuA in our previousstudy (17). To confirm that these smear protein bands possessed theamylolytic activity, the purified enzyme was analyzed by amylaseactive staining (data not shown). The starch degradation activity wasalso detected at the same location of smear protein bands from 150 to250 kDa on SDS-PAGE. These results suggest that the C-terminalrepeating units of ApuA do not contribute to folding of the structuralplatform. The truncated enzyme probably appears as a single proteinas in the case of the wild-type enzyme.

Effects of temperature and pH The optimal activity and stabi-lity of pH and temperature were examined for the purified ApuA andApuAΔ with four buffers covering the range between pH 2.5 and 8.0(Fig. 2A). Maximum activities were observed around pH 4.0–4.5 forthe amylopullulanase activity in 0.1 M acetate buffer. The twoamylopullulanases showed similar optimum pH profiles. To measurethe pH stability of the enzymatic activities, each of the enzymes wasincubated in various buffers and then assayed at 40 °C in 0.1 Msodium acetate buffer at pH 4.0. As shown in Fig. 2B, the activity ofApuA was less stable, especially above pH 7.0, than that of truncatedApuAΔ.

The dependence on temperature of the amylopullulanase activitywas determined by measuring the activities at various temperaturesin 0.1 M sodium acetate buffer at pH 4.0. Maximum activities weredetected at around 40–45 °C in both cases, as shown in Fig. 3A. Thethermal stabilities of both enzymes were assessed in 0.1 M sodiumacetate buffer after incubating for 1 h at various temperatures. Boththe activities were stable up to 30 °C, as shown in Fig. 3B. Above thistemperature, ApuA activity decreased more rapidly than that ofApuAΔ. The temperature at which 50% of ApuA activity was lost was41 °C, whereas the corresponding temperature for ApuAΔ was 45 °C.Thus, ApuAΔ exhibited a little more thermostability than ApuA.

TABLE 3. Kinetic parameters of ApuA and ApuAΔ

Substrate ApuA ApuAΔ

Km (g/l) kcat (s−1) kcat/Km

(g/l−1 s−1)Km (g/l) kcat (s−1) kcat/Km

(g/l−1 s−1)

Starch 7.7 25.5 3.3 5.0 23.1 4.6Amylose 2.6 15.3 5.9 3.3 21.1 6.4Pullulan 6.9 18.0 2.6 2.6 14.0 5.4

The purified wild-type and truncated enzymes at concentrations of 0.2 μg were used.

Substrate specificity The relative activities of ApuA and ApuAΔwere tested toward various polysaccharide substrates (Table 2). Bothenzymes exhibited significant activity toward soluble starch andamylopectin, while amylose, pullulan and glycogen were hydrolyzedat a moderate rate. Both enzymes were unable to hydrolyze towardcyclodextrins and dextran. The reaction catalyzed by the purifiedamylopullulanases for soluble starch, amylose and pullulan exhibitedMichaelis–Menten kinetics. The kinetic parameters of the purifiedApuA and ApuAΔ were determined at their pH and temperatureoptima (Table 3). The results indicated that the catalytic efficiency(kcat/Km) of the ApuAΔ with soluble starch, amylose and pullulan was1.4-, 1.1- and 2.1-fold higher, respectively, than that of the wild-typeenzyme ApuA, independent on their kcat and Km values. In particular,the apparent Km values of pullulan hydrolysis for ApuA and ApuAΔwere 6.9 g/l and 2.6 g/l, respectively, indicating that the affinity ofApuAΔ to pullulan increased approximately 2.7-fold and the substratebinding to α-1, 6-glucosidic linkage was altered.

Effects of metal ions and various reagents ApuA and ApuAΔwere incubated with various metal ions at 30 °C for 1 h, and theresults are summarized in Table 4. The α-amylase activities of ApuAand ApuAΔ were both inhibited strongly by Hg2+ (100% inhibition)and by Cu2+ ions (40%), while Co2+ ions moderately stimulated theactivity (by 30–50%). Similar results were obtained for pullulanaseactivities.

The effects on the amylopullulanase of various chemical reagentswere examined at 30 °C for 1 h, as shown in Table 5. The amylo-pullulanase activities of ApuA and ApuAΔ were inhibited strongly by0.1 mM N-bromosuccinimide, 1 M guanidine-HCl, and 5 M urea,moderately by 10mMα-cyclodextrin andγ-cyclodextrin, andnot at allby 10 mM EDTA and β-cyclodextrin. The same inhibition pattern of N-bromosuccinimide, α-cyclodextrin and β-cyclodextrin suggests thatthere might be the new conserved active site in the commonamylopullulanase (3, 28). In any case, these results strongly suggestthat the active site in the ApuA is a single site and C-terminal repeatingunits do not contribute on catalytic domain.

Hydrolyzed products of recombinant ApuA and ApuAΔ Fin-ally, we examined the time course of hydrolyzed products for solublestarch, amylose and pullulan with the two recombinant enzymes. Thetwo purified enzymes hydrolyzed soluble starch tomainly maltotriose(G3) and maltotetraose (G4) without any difference in the time course(17). However, ApuAΔ hydrolyzed amylosemore quickly to G3–G5 andpullulan more quickly to G3 than ApuA (Tables 6 and 7). Neitherenzyme produced glucose (G1) or maltose (G2). HPLC analysis of thehydrolysis products for amylose and pullulan showed that thetruncated ApuAΔ hydrolyzed both α-1,4- and α-1,6-glucosidic

TABLE 5. Effects of various chemicals on amylase and pullulanase activities of amylopullulanases

Chemicals Final concentration (mM) Relative residual activity (%)

ApuA ApuAΔ

Amylasea Pullulanaseb Amylasea Pullulanaseb

None 100 100 100 100EDTA 10 mM 98 96 101 90α-Cyclodextrin 10 mM 68 52 68 50β-Cyclodextrin 10 mM 101 105 99 110γ-Cyclodextrin 10 mM 49 35 53 40N-Bromosuccinimide 0.1 mM 0 0 0 0Guanidine-HCl 1 M 27 34 23 27Urea 1 M 78 80 75 70Urea 5 M 7 3 0 0

Each enzymatic activity was assayed after treatment of enzyme with chemicals for 30 min at 30 °C in 0.1 M acetate buffer (pH 4.5). a and b indicate the activity of hydrolysis for solublestarch and pullulan, respectively.

128 KIM ET AL. J. BIOSCI. BIOENG.,

linkages more liquefying as end-type amylolytic enzyme than saccha-rifying, compared with ApuA.

DISCUSSION

In the present study, to clarify the role of repeating amino acidsequences in the C-terminal region of amylopullulanase (ApuA) fromL. plantarum L137, we constructed the deletion of C-terminal of a geneencoding ApuA and purified the truncated enzyme ApuAΔ fromrecombinant NCL21 carrying pRN-Pul3Δ. Although we also attemptedto delete the N-terminal repeat region of the ApuA gene, cloning of theN-terminal deletion vector was unsuccessful.

Since the repeating amino acid sequences of the C-terminal regionof ApuAwere homologous to those of peptidoglycan-binding proteinsand the cell surface protein, it was expected that the repeating unit ofthe C-terminal would participate in the adhesion to saccharide and thestability of the enzyme structure. The biochemical study of ApuAΔrevealed that the enzyme hydrolyzed α-1,4- and α-1,6-glucosidiclinkages at almost the same efficacy as ApuA. However, the stability ofthe enzyme and the rate of production of the hydrolyzed products byApuAΔwere different from thosebyApuA. Thus,we concluded that therepeating unit of the C-terminal region of ApuA has a strong effect onthe catalytic activity through substrate binding and enzyme stabilities.

The recombinant wild-type (ApuA) and truncated (ApuAΔ) formsof amylopullulanase were purified from L. plantarum NCL21 harboringpRN-PulSa and pRN-PulΔ3, respectively. Despite identical promoterand signal sequences, the total activity of ApuAΔ in the culture super-natant was higher than that of ApuA. This difference in the enzymeproduction might have been due to the secretion efficiency andspecific activity of the recombinant protein by the deletion of extra C-

TABLE 6. Comparison of hydrolyzed products from amylose with ApuA and ApuAΔ

Enzyme Reactiontime (min)

End products formed (%)

G1 G2 G3 G4 G5 Gn

ApuA 5 0 0 0 0 0 10010 0 0 0 0 0 10020 0 0 0 0 0 10030 0 0 2.4 3.7 4.3 89.660 0 0 2.4 3.7 1.2 92.6

ApuAΔ 5 0 0 0.9 1.5 1.6 96.010 0 0 3.2 1.7 0.7 94.420 0 0 5.8 8.2 8.6 77.430 0 0 6.0 8.4 11.2 74.460 0 0 10.7 14.2 14.1 61.0

Assays were performed at 40 °C in 0.1 M sodium acetate buffer (pH 4.5) with 0.2 μg ofenzyme under the standard reaction conditions. The reaction mixture contained 1%pullulan in 0.1 M acetate buffer (pH 4.5) and the mixture was incubated at 30 °C. Theproducts measured were glucose (G1), maltose (G2), maltotriose (G3), maltotetraose(G4), maltopentaose (G5), and oligosaccharide larger than maltohexaose (Gn). Theamount of oligosaccharide formed was indicated by the mole percentage of the totalamount of products of hydrolysis at a given time.

terminal repeating units. The specific activity of purified ApuAΔ wasalso 1.4-fold higher than that of the purified ApuA, evenwhen enzymeactivity was evaluated on amolar basis, suggesting that the C-terminalrepeating unit of the amylopullulanase was involved in the enzymeactivity and the correct protein folding of truncated amylopullulanase.

The substrate specificities of both amylopullulanases were almostthe same to all substrates tested. However, ApuAΔ showed slightlyhigher affinity for soluble starch and pullulan than did ApuA, with adecrease in the Km value. The Km values of ApuAΔ revealed that thisenzyme had a slightly higher affinity for amylose than did ApuA, indi-cating the substrate binding was altered in ApuAΔ. The catalyticefficiencies (kcat/Km) of ApuAΔ for soluble starch, amylose and pullulanwere 1.4-, 1.1- and 2.1-higher, respectively than those of ApuA. Thepositive effects were not observed under the hydrolyzed products ofsoluble starch, but were strongly observed under the hydrolyzedproducts of amylose and pullulan. The truncated ApuAΔ hydrolyzedamylose and pullulan more quickly and produced greater amounts ofshort-chain end products in amylose and pullulan digestion than ApuA(Tables 6, 7). The apparent Km values of ApuAΔ with soluble starch,amylose and pullulanwere 0.65, 1.27, and 0.38-fold lower than those ofApuA (Table 3), respectively, indicating that the affinity of ApuAΔincreased to pullulan,, but decreased to amylose. It seems that the C-terminus region interfere with the binding of regular structural oligo-saccharides (homogenous α-1,4- or α-1,6-glucosidic linkages) hydro-lyzed from amylose and pullulan, but does not interfere significantlywith that of irregular structure from starch. Further work is needed todefine more clearly the nature of the function of the C-terminal repeatsequences using a different molecular size of polysaccharides. Sincesome ofα-amylases and pullulanases have a substrate-binding functionin their C-terminus (29–31), the deletion of the C-terminus would be

TABLE 7. Comparison of hydrolyzed products from pullulan with ApuA and ApuAΔ

Enzyme Reactiontime (min)

End products formed (%)

G1 G2 G3 G4 G5–7 Gn

ApuA 5 0 0 0.9 3.7 0 95.510 0 0 0.9 8.7 0 90.320 0 0 2.7 6.1 0 91.230 0 0 24.7 0 0 75.360 0 0 22.6 0 77.4 0

ApuAΔ 5 0 0 35.6 0 64.4 010 0 0 33.2 5.1 61.7 020 0 0 52.8 9.6 37.6 030 0 0 79.1 0 20.9 060 0 0 92.1 0 7.9 0

Assays were performed at 40 °C in 0.1 M sodium acetate buffer (pH 4.5) with 0.2 μg ofenzyme under the standard reaction conditions. The reaction mixture contained 1%pullulan in 0.1 M acetate buffer (pH 4.5) and the mixture was incubated at 30 °C.Products indicated were glucose (G1), maltose (G2), maltotriose (G3), maltotetraose(G4), maltapentaose to maltoheptaose (G5–7), and oligosaccharide larger thanmaltooctaose (Gn). The amount of oligosaccharide formed was indicated by the molepercentage of the total amount of products of hydrolysis at a given time.

CHARACTERIZATION OF TRUNCATED AMYLOPULLULANASE 129VOL. 107, 2009

expected to result in a loss of this activity. However, the 100 C-terminusamino acid residues of amylopullulanase from Thermoanaerobacterethanolicus 39E have no effect on structure and activity (32). In theamylopullulanase from L137 strain, the C-terminal extension had anegative effect on the activity of the protein containing substratebinding and hydrolytic production rate for amylose and pullulan.

Since ApuAΔ showed greater pH stability and thermostability thanApuA, the C-terminus of ApuA appears to have a negative effect on thethermostability of the enzyme. These results are in agreement withthe previous observations for the amylase from Bacillus stearothermo-philus (33) and Bacillus subtilis (34) for which the C-terminally trun-cated enzymes were more thermostable than the entire enzyme, butare different from those obtained with the amylase from Lactobacillusamylovorus (35). Our results suggest that the compact structure(24 kDa deletion) of ApuAΔ might increase its stability, and that theextra C-terminal region did not greatly change the protein structure ofamylopullulanase, which folded correctly.

Although the 3-D structure of the amylopullulanase containing theextra C-terminal region is still unknown, we speculate that this regionforms a domain without affecting the folding of the whole enzyme.We concluded that the extra C-terminal repeating amino acids, QPT, ofApuA were not necessary for the structure and function of theamylopullulanase, suggesting that the C-terminal repeating region ofApuA serves no apparent purpose. In other words, we succeeded inthe construction of a highly active and stable amylopullulanase.

ACKNOWLEDGMENTS

This work was supported by a Grant-in-Aid for Scientific Researchfrom the Ministry of Education, Culture, Sports, Science andTechnology of Japan (Kaken no. 20404023).

References

1. Fogarty, W. M. and Kelly, C. T.: Recent advances in microbial amylases, in: W. M.Fogart, C. T. Kelly (Eds.), Microbial enzymes and biotechnology, 2nd edn., AppliedScience, London, 1990, pp. 71–132 (1990).

2. Takasaki, Y.: Pullulanase-amylase complex enzyme from Bacillus subtilis, Agric.Biol. Chem., 51, 9–16 (1987).

3. Ara, K., Saeki, K., Igarashi, K., Takaiwa, M., Uemura, T., Hagihara, H., Kawai, S.,and Ito, S.: Purification and characterization of an alkaline amylopullulanase withboth α-1,-4 and α-1,-6 hydrolytic activity from alkalophilic Bacillus sp. KSM-1378,Biochim. Biophys. Acta, 1243, 315–324 (1995).

4. Coleman, R. D., Yang, S. S., and McAlister, M. P.: Cloning of the debranching-enzyme gene from Thermoanaerobium brockii into Escherichia coli and Bacillussubtilis, J. Bacteriol., 169, 4302–4307 (1987).

5. Mathupala, S. P., Lowe, S. E., Podkovyrov, S. M., and Zeikus, J. G.: Sequencing ofthe amylopullulanase (apu) gene of Thermoanaerobacter ethanolicus 39E, andidentification of the active site by site-directed mutagenesis, J. Biol. Chem., 268,16332–16344 (1993).

6. Plant, A. R., Clemens, R. M., Morgan, H. W., and Daniel, R. M.: Active-site- andsubstrate-specificity of Thermoanaerobium Tok6-B1 pullulanase, Biochem. J., 246,537–541 (1987).

7. Ramesh, M. V., Podkovyrov, S. M., Lowe, S. E., and Zeikus, J. G.: Cloning andsequencing of the Thermoanaerobacterium saccharolyticum B6A-RI apu gene andpurification and characterization of the amylopullulase from Escherichia coli, Appl.Environ. Microbiol., 60, 94–101 (1994).

8. Lee, S. P., Morikawa, M., Takagi, M., and Imanaka, T.: Cloning of the aapT gene andcharacterization of its product, an amylase-pullulanase (AapT), from thermophilicand alkaliphilic Bacillus sp. strain XAL601, Appl. Environ. Microbiol., 60, 3764–3773(1994).

9. Brunswick, J. M., Kelly, C. T., and Fogarty, W. M.: The amylopullulanase of Bacillussp. DSM 405, Appl. Microbiol. Biotechnol., 51, 170–175 (1999).

10. Kim, C. H. and Kim, Y. S.: Substrate specificity and detailed characterization ofa bifunctional amylase pullulanase enzyme from Bacillus circulans F-2 havingtwo different active sites on one polypeptide, Eur. J. Biochem., 227, 687–693(1995).

11. Hatada, Y., Igarashi, K., Ozaki, K., Ara, K., Hitomi, J., Kobayashi, T., Kawai, S.,

Watabe, T., and Ito, S.: Amino acid sequence and molecular structure of an alkalineamylopullulanase from Bacillus that hydrolyzes α-1,4 and α-1,6 linkages inpolysaccharides at different active sites, J. Biol. Chem., 271, 24075–24083 (1996).

12. Ganghofner, D., Kellermann, J., Staudenbauer, W. L., and Bronnenmeier, K.:Purification and properties of an amylopullulanase, a glucoamylase and an α-glucosidase in the amylolytic enzyme system of Thermoanaerobacterium thermo-saccharolyticum, Biosci. Biotechnol. Biochem., 62, 302–308 (1998).

13. Lin, L. -L., Tsau, M. -R., and Chu, W. -S.: Purification and properties of a 140 kDaamylopullulanase from the thermophilic and alkaliphilic Bacillus sp. strain TS-23,Biotechnol. Appl. Biochem., 24, 101–107 (1996).

14. Matuschek, M., Burchhardt, G., Sahm, K., and Balh, H.: Pullulanase of Thermoa-naerobacterium thermosulfurigenes EM1 (Clostridium thermosulfurogenes): molecu-lar analysis of the gene, composite structure of the enzyme, and a common modelfor its attachment to the cell surface, J. Bacteriol., 176, 3295–3302 (1994).

15. Vishnu, C., Naveena, B. J., Altaf, M. d., Venkateshwar, M., and Reddy, G.:Amylopullulanase-A novel enzyme of Lactobacillus amylophilus GV6 in directfermentation of starch to L(+) lactic acid, Enzyme Microb. Technol., 38, 545–550(2006).

16. Ryan, S. M., Fitzgerald, G. F., and van Sinderen, D.: Screening for and identificationof starch-, amylopectin-, and pullulan-degrading activities in bifidobacterialstrains, Appl. Environ. Microbiol., 72, 5289–5296 (2006).

17. Kim, J. H., Sunako, M., Ono, H., Murooka, Y., Fukusaki, E., and Yamashita, M.:Characterization of gene encoding amylopullulanase from plant-originated lacticacid bacterium, Lactobacillus plantarum L137, J. Biosci. Bioeng., 106, 449–459(2008).

18. Olympia, M., Ono, H., Shinmyo, A., and Takano, M.: Lactic acid bacteria in a fer-mented fishery product, “burong bangus”, J. Ferment. Bioeng., 73, 193–197 (1992).

19. Olympia, M., Fukuda, H., Ono, H., Kaneko, Y., and Takano, M.: Characterization ofstarch-hydrolyzing lactic acid bacteria isolated from a fermented fish and rice food,“burong bangus”, and its amylolytic enzyme, J. Ferment. Bioeng.,80,124–130 (1995).

20. Sambrook, J. and Russel, D. W.: Molecular cloning: a laboratory manual, 3rd ed.,Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y., 2001

21. Anderson, D. G. and McKay, L. L.: Simple and Rapid method for isolating largeplasmid DNA from lactic streptococci, Appl. Environ. Microbiol., 46, 549–552(1983).

22. Hanahan, D.: Studies on transformation of Escherichia coli with plasmid, J. Mol.Biol., 166, 577–580 (1983).

23. Kaneko, Y., Kobayashi, H., Kiatpapan, P., Nishimoto, T., Napitupulu, R., Ono, H.,andMurooka, Y.: Development of a host–vector system for Lactobacillus plantarumL137 isolated from a traditional fermented food produced in Philippines, J. Biosci.Bioeng., 89, 62–67 (2000).

24. Miller, G. L.: Use of dinitrosalicylic acid reagent for determination of reducingsugar, Ann. Biochem., 31, 426–428 (1959).

25. Hanes, C. S.: Studies on plant amylases: the effect of starch concentration upon thevelocity of hydrolysis by the amylase of germinated barley, Biochem. J., 26,1406–1421 (1932).

26. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.: Protein measurementwith the folin phenol reagent, J. Biol. Chem., 193, 265–275 (1951).

27. Laemmli, U. K.: Cleavage of structural proteins during the assembly of the head ofbacteriophage T4, Nature, 227, 680–685 (1970).

28. Lin, F. -P. and Leu, K. -L.: Cloning, expression, and characterization of thermostableregion of amylopullulanase gene from Thermoanaerobacter ethanolicus 39E, Appl.Biochem. Biotechnol., 97, 33–44 (2002).

29. Svensson, B., Jespersen, H., Sierks, M. R., and MacGregor, E. A.: Sequencehomology between putative raw-starch binding domains from different starch-degrading enzymes, Biochem. J., 264, 309–311 (1989).

30. Lo, H. F., Lin, L. L., Chiang, W. Y., Chie, M. C., Hsu, W. H., and Chang, C. T.: Deletionanalysis of the C-terminal region of the α-amylase of Bacillus sp. strain TS-23, Arch.Microbiol., 178, 115–123 (2002).

31. Erra-Pujada,M., Chang-Pi-Hin, F., Debeire, P., Duchiron, F., andO’Donohue,M. J.:Purification and properties of the catalytic domain of the thermostable pullulanasetype II from Thermococcus hydrothermalis, Biotechnol. Lett., 23, 1273–1277 (2001).

32. Lin, H. Y., Chuang, H. H., and Lin, F. P.: Biochemical characterization of engineeredamylopullulanase from Thermoanaerobacter ethanolicus 39E-implicating the non-necessity of its 100 C-terminal amino acid residues, Extremophiles, 12, 641–650(2008).

33. Vihinen, M., Peltonen, T., Litia, A., Suominen, I., and Mantsala, P.: C-terminaltruncations of thermostable Bacillus stearothermophilus α-amylase, Protein Eng., 7,1255–1259 (1994).

34. Ohdan, K., Kuriki, T., Kaneko, H., Shimada, J., Takada, T., Fujimoto, Z., Mizuno,H., and Okada, S.: Characteristics of two forms of α-amylases and structuralimplication, Appl. Environ. Microbiol., 65, 4652–4658 (1999).

35. Rodriguez Sanoja, R., Morlon-Guyot, J., Jore, J., Pintado, J., Juge, N., and Guyot,J. P.: Comparative characterization of complete and truncated forms of Lactoba-cillus amylovorus-amylase and role of the C-terminal direct repeats in raw-starchbinding, Appl. Environ. Microbiol., 66, 3350–3356 (2000).