Enzyme and Microbial Technology 27 (2000) 406 413

www.elsevier.com/locate/enzmictec

Purication and characterization of an extracellular -amylase produced by Lactobacillus manihotivorans LMG 18010T, an amylolytic lactic acid bacterium
G. Aguilar,a J. Morlon-Guyot,b B. Trejo-Aguilar,a J.P. Guyotb,*
a

UNAM, Departamento de Alimentos y Biotecnologia, Facultad de Quimica, Cd. Universitaria, Qumica Conj. E., A.P. 04510 Mexico, D.F., Mexico b IRD (ex-ORSTOM), Laboratoire de Biotechnologie Microbienne Tropicale, 911 avenue Agropolis, B.P. 5045, Montpellier cedex 1, France Received 24 June 1999; received in revised form 10 April 2000; accepted 18 April 2000

Abstract This work presents the purication and characterization of an extracellular -amylase (1,4- -D-glucan glucanohydrolase, EC 3.2.1.1) produced by a new lactic acid bacterium: Lactobacillus manihotivorans able to produce L( ) lactic acid from starch. The molecular weight was found to be 135 kDa. The temperature and pH optimum were 55C and 5.5, respectively, and pI was 3.8. The -amylase had good stability at pH range from 5 to 6 and the enzyme was sensitive to temperature, losing activity within 1 h of incubation at 55C. Higher thermal stability was observed when the enzyme was incubated in presence of soluble starch. Km value and activation energy were 3.44 mg/ml and 32.55 kJ/mol, respectively. Amylose was found to be a better substrate than soluble starch and amylopectin. Al3 , Fe3 , and Hg2 (10 mM) almost completely inhibited the -amylase. 2000 Elsevier Science Inc. All rights reserved.
Keywords: -Amylase; Lactic acid bacteria; Lactobacillus manihotivorans, Starch; Cassava sour starch

1. Introduction Amylolytic lactic acid bacteria (ALAB) have been isolated from different environments and geographical areas, such as swine and cattle waste-corn fermentations in the United States (Lactobacillus amylophilus and Lactobacillus amylovorus) [1,2], fermented cassava roots in Congo and Niger (Lactobacillus plantarum strains) [3,4], chicken crop in France (Lactobacillus strains) [5], sh silage in Sweden (Leuconostoc strains) [6], fermented sh and rice food in Japan (Lactobacillus plantarum) [7], and maize sourdoughs in Benin (Lactobacillus fermentum) [8]. These ALAB may play an important role in converting directly amylaceous substrates for food or lactic acid production, avoiding prehydrolysis steps for starch utilization. Reports on the purication and characterisation of -amylase (1,4- -D-glucan glucanohydrolase, EC 3.2.1.1) produced by ALAB are scant, in spite of some studies on the amyA gene of ALAB [9,10] which demonstrated that these
* Corresponding author. Tel.: 33-4-67416285; 67416283. E-mail address: jpguyot@mpl.ird.fr (J.P. Guyot). fax: 33-4-

amylases share some uncommon molecular characteristics. Amylases from L. amylophilus, L. amylovorus, and L. plantarum A6 share similar properties [9 13]. The studies on puried amylases from different ALAB indicate that they generally are of high molecular weight, in contrast to that from Bacillus species [14]. For instance amylases from L. plantarum L137, L. amylovorus and L. amylophilus were reported to be of 230 kDa [7], 140 kDa [15], and 100 kDa [12], respectively. L. manihotivorans, a new ALAB species, was isolated from the process to produce cassava sour starch in Colombia [16] and the type strain LMG 18010T produces an extracellular -amylase. The lactic acid bacteria involved in this process imparts bread making properties to sour starch from cassava, products obtained from this starch are very popular in Colombia and Brazil. L. manihotivorans is able to convert starch to L( )lactic acid directly like L. amylophilus (unlike L. plantarum and L. amylovorus) and appears to display higher amylase activity and growth rate than this ALAB when cultivated on starch. The study of this new species will help to understand better its role in the cassava sour starch process, and also to exploit as best as possible its potentialities based on both amylase and lactic acid produc-

0141-0229/00/$ see front matter 2000 Elsevier Science Inc. All rights reserved. PII: S 0 1 4 1 - 0 2 2 9 ( 0 0 ) 0 0 2 3 0 - 1

G. Aguilar et al. / Enzyme and Microbial Technology 27 (2000) 406 413

407

tion. The purication and characterization of the amylase produced by L. manihotivorans LMG 18010T is attempted in this paper.

5.5 to assay stability. Samples were cooled in an ice-water bath after incubation and the residual activities were determined. Half life of -amylase was assayed both at 40 and 60C. The residual activity was determined as described above. 2.6. Effect of metal ions

2. Material and methods 2.1. Microorganism L. manihotivorans LMG 18010T is available at the BCCM/LMG culture collection. The strain is conserved in 40% glycerol at 80C. 2.2. Culture conditions MRS-medium [17] (with glucose or soluble starch as substrates at 15 g/l) was routinely used for cultivation of L. manihotivorans. For amylase production, cells were grown with soluble starch (Prolabo, Paris, France) as substrate in 2-1 bioreactors (Biolatte, Poissy, France) with pH controlled at 6 by NaOH (5 N). Growth medium was gently stirred to maintain homogeneity. Bioreactors were inoculated (10% v/v) with 12-h cultures pregrown on MRSglucose. 2.3. Extracellular amylase activity Extracellular -amylase activity was assayed by measurement of the iodine-complexing ability of starch at pH 5.5 and 55C as previously described [13]. One enzyme unit (U) is dened as the amount of enzyme that permits the hydrolysis of 10 mg of starch in 30 min in the conditions of the assay. Activity was also determined by quantication of the amount of reducing sugars by the Dinitrosalicylic (DNS) method [18]. 2.4. Effect of pH The relative -amylase activity by using 1.2% (w/v) soluble starch was determined at several pH values, from 2.6 to 8 with citrate-phosphate (0.1 M) and Tris-HCl (0.1 M) buffers. The same buffers were used to determine pH stability. One volume of -amylase was dialysed against distilled water and diluted in an equal volume of the above buffers, and the mixture was incubated during 24 h at 4 and 40C. At this time, residual activities of samples were determined under standard conditions. The stability and optimal pH of the puried enzyme were tested by diluting 10 to 20-fold and assaying as described above. 2.5. Effect of temperature Activity of samples was determined at temperatures from 30 to 90C with 1.2% (w/v) soluble starch. Enzyme samples were incubated for 10 and 60 min in the presence and absence of starch in citrate-phosphate buffer (0.1 M) at pH Enzyme assays were performed in presence of different metal ions, all at 10 mM nal concentration. The relative activity of the enzyme was compared in 0.1 M of citratephosphate buffer. The chloride salts of Na , K , Mn2 , Co2 , Ba2 , Hg2 , Ca2 , Ni2 , Fe2 , Al3 , and sulphate salts of Mg2 , Fe3 , Zn2 , and Cu2 were used. 2.7. Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) and in situ detection of -amylase activity (zymography) SDS-PAGE was performed according to Laemmli [19], with a resolving gel containing 7.5% acrylamide and 2.7% bis-acrylamide. The stacking gel contained 4% acrylamide and 2.7% bis-acrylamide. The SDS (Bio-Rad, Richmond, CA, USA) concentration was 0.1% in the gel and running buffer. Samples were boiled for 90 s in sample buffer (4% SDS, 20% glycerol, 10% 2-mercaptoethanol, 125 mM TrisHCl, pH 6.8, and 0.005% bromophenol blue). Samples were submitted to electrophoresis at 30 mA for 1 to 2 h through either 0.75 or 1 mm gel in a vertical slab gel unit SE-600 (Hoefer Scientica Instruments, USA). Acrylamide gels were protein stained with a solution containing 0.125% (v/v) Coomassie Blue R-250 (Bio-Rad) 50% (v/v) methanol, and 10% acetic acid. The gels were destained in 10% acetic acid with heating. For zymography, acrylamide gels were incubated for 0.5 to 1 h with shaking, in Tris-HCl 0.01 M, pH 7.6, and with a change of citrate-phosphate buffer 0.1 M, pH 5.5. After this time, gels were immersed in a solution of 1.0% soluble starch (Prolabo) and maintained at 40C with shaking for a period of 1 h. After incubation, gels were stained with iodine-solution until the appearance of clear zones under a dark brown background. 2.8. Isoelectric point determination Analytical isoelectric focusing was performed with the Pharmacia Phast System (Piscatawy, NJ, USA) by using a Phast Gel IEF 39, as per the instruction. Protein bands were stained with a solution containing 0.125% (v/v) Coomassie Blue R-250 (Bio-Rad), 50% (v/v) methanol, and 10% acetic acid. The gel was destained in 10% acetic acid with heating. For the localisation of the -amylase activity in the gel, it was incubated as described for zymograms (see above). The isoelectric point of -amylase was determined by using broad range protein markers ranging from pI 3 to 10 (Pharmacia Biotech, Uppsala, Sweden).

408

G. Aguilar et al. / Enzyme and Microbial Technology 27 (2000) 406 413

Fig. 1. Elution prole of strain LMG 18010T -amylase on DEAE-cellulose column chromatography. The ultraltered enzyme solution was applied to the column which was equilibrated with 0.1 M Tris-HCl buffer, pH 7.2. The amylase activity was eluted at 0.25, 0.30, and 0.34 M NaCl. Detailed experimental conditions are given in section 2. Symbols () A280, (F) -amylase activity and (- - - - -) NaCl gradient.

2.9. Purication procedure After 14 h of batch cultivation, fermentation broth was collected and centrifuged at 10 000 rev./min for 20 min at 4C. One liter of supernatant uid was concentrated to 50 ml by an ultraltration device (Amicon, Beverly, MA, USA) with a PM-10 membrane. The concentrated enzyme solution was dialysed against 10 l of citrate-phosphate buffer 10 mM, pH 5.5, and ltered through 0.2 m pore size membrane (Millipore, Saint Quentin les Yvelines, France) and applied to DEAE-Cellulose column (DE-52, Whatman Laboratory Sales, Hillsboro, OR, USA) 25 250 mm at a ow rate of 2.5 ml/min at 25C, previously equilibrated with 0.1 M Tris-HCl, pH 7.2. Elution was done with a linear NaCl gradient (0 to 0.8 M) in the same buffer. Fractions of 2.5 ml were collected and analysed. The most active fractions were pooled, dialysed against buffer and applied to a Sephadex G-200 (10 250 mm) column and the active fractions from this column were used for further characterisation. 2.10. Kinetic determinations Initial rates of soluble starch hydrolysis were determined at various substrate concentrations (112 mg/ml). The ki-

netic constants Km and Vmax were estimated by the method of LineweaverBurk.

3. Results 3.1. Enzyme production and purication Strain LMG 18010T was grown on medium MRS containing 1.5% (w/v) soluble starch. The -amylase activity was evident since the early stages of fermentation, culminating at the beginning of stationary phase of growth. The crude cell-free culture supernatant contained about 2225 U/ml. After concentration by ultraltration and dialysis against citrate-phosphate buffer the activity increased to around 843 U/ml. The -amylase was rst puried by anion-exchange chromatography on DEAE-cellulose column. The elution prole showed three peaks of -amylase activity (Fig. 1). The -amylase activity eluted from column at 0.25, 0.30, and 0.34 M NaCl for peaks I, II, and III, respectively. A summary of the purication procedure is given in Table 1. For peak III, fractions 67 68 were pooled dialysed and applied to a Sephadex G-200 column. From this column a single peak of -amylase activity was obtained (data not

Table 1 Summary of purication steps of extracellular -amylase from Lactobacillus manihotivorans LMG 18010T Purication step Culture supernatant Concentration, ultraltration and dialysis DEAE-Cellulose chromatography Peak I (fractions 5253) Peak II (fraction 60) Peak III (fractions 6768) Sephadex G-200 gel ltration Total protein (mg) 1,032 556 2.21 1.9 3.6 2.27 Total activity (U/l) 23,000 16,864 509 208 1,617 1,432 Specic activity (U/mg) 22.30 30.30 231.00 109.50 449.00 631.00 Yield (%) 100 73.3 2.2 0.9 7.0 6.2 Purication (fold) 1.0 1.36 10.4 4.91 20.0 28.3

G. Aguilar et al. / Enzyme and Microbial Technology 27 (2000) 406 413

409

Fig. 2. SDS-PAGE (Lanes 1, 2, 3, 4, 5) and zymogram (Lanes a, b, c, d, e) of -amylase containing DEAE fractions. Lanes 1 and a, ultraltered enzyme solution; Lanes 2 and b, peak I; Lanes 3 and c, peak II, Lanes 4 and d, peak III after DEAE-cellulose chromatography; and Lanes 5 and e, active fraction after molecular sieving chromatography of peak III from DEAE-cellulose column.

shown). The purication procedure yielded an -amylase with a specic activity of 631 U mg/protein, exhibiting a purication factor of 28.3 and yield of 6.2%. The apparent molecular mass of strain LMG 18010T -amylase was found to be 135 kDa as estimated by SDSPAGE (Fig. 2). The isoelectric point of the enzyme was found to be 3.8 by PAGE-isoelectric focusing. For peak I and II, SDS-PAGE and zymograms revealed active protein bands in the range of 43 64 and 4390 kDa for peak I and II, respectively (Fig. 2). It could be stated that the main activity was found in the protein band of 135 kDa. It is believed that these bands could be produced by partial digestion of the 135-kDa -amylase during fermentation. 3.2. Effect of pH and temperature The inuence of pH and temperature on activity is shown in Fig. 3. The amylase showed higher activity from pH 4 to 6 with a maximum observed at 5.5 (Fig. 3a). However, it could be considered that the enzyme has a broad pH range of activity. The optimum temperature for activity was found to be 55C. The activity reduced to 98% at 65C (Fig. 3b). The Arrhenius law was followed between 30 and 50C and an activation energy of 32.55 kJ/mol was calculated (insert Fig. 3b). A sharp decrease of activity was observed above 65C. 3.3. Stability at pH and temperature The enzyme showed good stability at a pH range from 5 to 6 at 4 and 40C. The maximum was observed at pH 5.5 at both temperatures (Fig. 4a). At 40C, a higher decrease in activity was seen at pH 7 and 8. The stability of -amylase was poor at acidic pH whatever the incubation temperature (Fig. 4a). Thermal stability of the enzyme was monitored by measuring the residual enzymatic activity, on incubation of 10

and 60 min with or without substrate at the prescribed temperatures (Fig. 4b). The results showed that the -amylase was sensitive to temperature inactivation above 40C, losing about 30% of their activity at 50C without substrate and practically all the activity at 55C during 1 h of incubation (Fig. 4b). Higher stability was observed when the incubation time was reduced to 10 min. The stability was also higher by incubation in presence of 1% starch, and particularly at higher temperatures. In fact, around 10% of

Fig. 3. Effect of pH at 55C (a) and temperature at pH 5.5 (b) on enzyme activity. Insert in (b) Arrhenius plot.

410

G. Aguilar et al. / Enzyme and Microbial Technology 27 (2000) 406 413 Table 2 Effect of various metal ions on puried -amylase from strain LMG 18010T Ion (10 mM) Control (No addition) K Na Ca2 Co2 Mg2 Ba2 Mn2 Fe2 Zn2 Ni2 Cu2 Hg2 Fe3 Al3
a

Relative -amylase activitya 1.00 1.04 1.03 1.01 0.95 0.94 1.03 0.99 0.87 1.03 0.41 0.51 0.02 0.08 0.10

The activity of control was 19.76 U/ml and it was taken as one.

ity and Fe2 , Cu2 , and Ni2 inhibition, respectively. 3.5. Catalytic properties

showed 13, 49, and 59%

Starch hydrolysis was simultaneously measured assessing decrease in iodine-staining power and production of reducing sugars (Fig. 5). A rapid decrease in the iodinestaining property was observed together with a low production of reducing sugars which ranged from 5 to 8% of total starch, indicating a typical -amylase starch breackdown pattern. The results indicate the action of an endoenzyme. The substrate specicity of the -amylase was evaluated
Fig. 4. Effect of pH (a) and temperature (b) on stability of -amylase and half life of the enzyme (c). (a) Incubation was carried out during 24 h at 4C (E) and 40C (F) in 0.1 M citrate-phosphate buffer (pH between 2.6 7) and Tris-HCl (pH 8). (b) Incubation was carried out during 10 min ( , s) or 60 min (, ) in the presence (open symbols) and in the absence (closed symbols) of 1% starch. (c) half-life of the enzyme was estimated at 40C (}) and 60C ({) during different length of time in 0.1 M citratephosphate buffer, pH 5.5.

the activity was retained at 80C during 10 min of incubation and a good stability was observed from 30 to 55C. Half life of the enzyme was estimated to be 10 min at 60C and around 100 min at 40C (Fig. 4c). 3.4. Effect of metal ions The -amylase activity was measured at pH 5.5 and 55C in presence of different metal ions (Table 2). None of the metal ions was required for catalytic activity. The following metal ions had no effect on -amylase activity: K , Na , Ca2 , Co2 , Mg2 , Ba2 , Mn2 , and Zn2 . Al3 , Fe3 , and Hg2 almost completely inhibited enzyme activ-

Fig. 5. Reduction of iodine-staining power (F, E) and release of reduced groups (s, ) from starch (open symbols) and amylopectin (closed symbols) of puried -amylase from strain LMG 18010T at pH 5.5 and 55C.

G. Aguilar et al. / Enzyme and Microbial Technology 27 (2000) 406 413

411

Fig. 6. MichaelisMenten type plot of -amylase hydrolysis rate at different starch concentrations. Insert: LineweaverBurk representation of MichaelisMenten plot.

on soluble starch, amylose, amylopectin, glycogen, maltodextrins, and - and -cyclodextrins. Amylose was found to be the best substrate for strain LMG 18010T -amylase. The relative activity observed on amylose was 1.6 times greater than that obtained on starch. The hydrolysis of amylopectin was 90% in relation to starch. Glycogen and dextrins with 10 and 20 DP were hydrolysed at low rate reaching 6.7, 7.3, and 8.9%, respectively, and no activity was found on - and -cyclodextrins. The -amylase showed a MichaelisMenten type kinetics when hydrolysing soluble starch. As calculated from LineweaverBurk plots (Fig. 6) the Km and Vmax values at 55C were 3.44 mg/ml and 0.45 mg hydrolysed starch/ml/ min, respectively.

4. Discussion The -amylase was isolated from the supernatant of strain LMG 18010T grown on starch and puried 28 times with a yield of 6.2%. The yield of amylase activity recovered from the all three peaks after the DEAE-Cellulose column was 10%. Increasing protein bands of low molecular weight with amylase activity were observed on zymograms, after prolonged incubation without protease inhibitor of crude or concentrated enzyme solutions at room temperature. The addition of protease inhibitors reduced this problem (data not shown). Giraud et al. [13] also observed on zymograms protein bands (between 50 and 150 kDa) with amylase activity during the purication of the -amylase produced by L. plantarum A6. They concluded that their extract consisted of a population of aggregates of 50 kDa amylase. But further molecular works on the L. plantarum A6 -amylase gene invalidated this hypothesis [10]. From the deduced amino acid sequence of the amyA gene of L.

plantarum A6, it was concluded that the molecular weight of the enzyme was about 100 kDa [10]. Then, partial proteolysis would probably generate the different active bands observed on zymograms, and also inactive products during preliminary steps of purication procedure, that would explain the low recovered amylase activity yield. The apparent purity of the enzyme was demonstrated by SDS-PAGE (Fig. 2). The puried -amylase exhibited an isoelectric point of 3.8. This value is lower than those reported for other microorganisms, e.g., B. subtilis [20] and Clostridium acetobutylicum [21]. The apparent molecular mass was found to be 135 kDa, which is consistent with the predicted molecular mass from sequenced gene. This value is similar to the -amylase from L. amylovorus, L. plantarum, A6, and L. amylophilus (which range from 100 to 140 kDa [15,10,12]). The optimal pH (5.5) and temperature (55C) were similar to other bacterial -amylases. However, they differ from the amylases of Clostridium acetobutylicum [21] and Alicyclobacillus acidocaldarius [22] which showed resistance to acidic pH. The stability of strain LMG 18010T amylase was less than 10% at pH values below 4. Activation energy of 32.6 kJ/mol was closed of the value found for L. plantarum A6 (30.6 kJ/mol) [13], but higher or lower than that reported for other amylases produced by different bacteria (B. licheniformis: 14 kJ/mol [23], B. subtilis: 8 kJ/mol [24], B. stearothermophilus: 22 kJ/mol [24], Pyrococcus woesei: 74 kJ/mol [25], and C. acetobutylicum: 26.3 kJ/mol [21]). However, the thermal stability of the enzyme was limited. One h of incubation at 55C suppressed the activity completely. The stability could be improved by addition of 1% starch, suggesting that the high starch concentration of some processes may improve the performance of strain LMG 18010T amylase. Although it is difcult to compare the kinetic values of amylases obtained by other workers in view of the different starch substrates used and the assay conditions, the Km value for starch of strain LMG 18010T -amylase (3.44 mg/ml) was within the range of many amylases (0.35 to 4.7 mg/ml). It has to be noted that, for the same type of soluble starch, the afnity of strain LMG 18010T -amylase was lower than that of L. plantarum A6 (Km : 2.38 mg/ml [13]). Furthermore, similarly to L. amylophilus [12], Fe3 and some divalent cations (Fe2 , Cu2 , Ni2 ) totally or partially inhibited strain LMG 18010T -amylase unlike the L. plantarum A6 and L. amylovorus amylases [12,13]. It appears that Ca2 did not enhance amylase activity of these Lactobacillus species except for L. amylovorus [12]. Comparative studies on the molecular structure of these enzymes will probably explain these differences. Unlike L. amylophilus and L. amylovorus, strain LMG 18010T had a higher activity with amylose than with soluble starch and amylopectin (no data are available for L. plantarum A6). Furthermore, activity (relative to starch) on amylopectin remains signicantly higher for strain LMG

412

G. Aguilar et al. / Enzyme and Microbial Technology 27 (2000) 406 413

18010T (90%) than that observed for L. amylophilus (52%) and L. amylovorus (2%) [12]. This suggests that strain L. manihotivorans would have a greater ability to degrade starches with high amylose ratio and also, whatever starch composition, starch degradation by strain LMG 18010T would be better than the former species. Regarding the use of glycogen, in spite of a very low relative activity when compared to starch, it has been shown that L. manihotivorans was able to degrade partially mussel glycogen at similar rates than L. plantarum A6 [26]. Such an activity could be of potential interest for some applications, such as lactic acid production from glycogen containing products [26] (e.g., for seafood product preservation).

References
[1] Nakamura LK, Crowell CD. Lactobacillus amylophilus, a new starchhydrolyzing species from swine waste-corn fermentation. Dev Ind Microbiol 1979;20:531 40. [2] Nakamura LK. Lactobacillus amylovorus, a new starch-hydrolyzing species from cattle waste-corn fermentations. Int J Syst Bacteriol 1981;31:56 63. [3] Nwankwo D, Anadu E, Usoro R. Cassava-fermenting organisms. MIRCEN J 1989;5:169 79. [4] Giraud E, Brauman A, Keleke S, Lelong B, Raimbault M. Isolation and physiological study of an amylolytic strain of Lactobacillus plantarum. Appl Microbiol Biotechnol 1991;36:379 83. [5] Champ M, Szylit O, Raibaud P, Abdelkader A. Amylase production by three Lactobacillus strains isolated from chicken crop. J Appl Bacteriol 1983;55:48793. [6] Lindgren S, Refai O. Amylolytic lactic acid bacteria in sh silage. J Appl Bacteriol 1984;57:221 8. [7] Olympia M, Fukuda H, Ono H, Kaneko Y, Takano M. Characterisation of starch-hydrolyzing lactic acid bacteria from a fermented sh and rice food, burong isda, and its amylolytic enzyme. J Ferm Bioeng 1995;80:124 30. [8] Agati A, Guyot JP, MorlonGuyot J, Hounhouigan J. Isolation and characterization of new amylolytic strains of Lactobacillus fermentum from fermented maize doughs (mawe and ogi) from Benin. ` J Appl Microbiol 1998;85:51220. [9] Fitzsimons A, Hols P, Jore J, Leer RJ, OConnell M, Delcour J. Development of an amylolytic Lactobacillus plantarum silage strain expressing the Lactobacillus amylovorus -amylase gene. Appl Environ Microbiol 1994;60:3529 35. [10] Giraud E, Cuny G. Molecular characterization of the -amylase genes of Lactobacillus plantarum A6 and Lactobacillus amylovorus reveals an unusual 3 end structure with direct tandem repeats and suggest a common evolutionary origin. Gene 1997;198:149 57. [11] BurgessCassler A, Imam S. Partial purication and comparative characterization of -amylase secreted by Lactobacillus amylovorus. Curr Microbiol 1991;23:20713. [12] Pompeyo CC, Gomez MS, Gasparian S, MorlonGuyot J. Compari son of amylolytic properties of Lactobacillus amylovorus and of Lactobacillus amylophilus. Appl Microbiol Biotechnol 1993;40: 266 9. [13] Giraud E, Gosselin L, Marin B, Parada JL, Raimbault M. Purication and characterization of an extra-cellular amylase from Lactobacillus plantarum strain A6. J Appl Bacteriol 1993;75:276 82. [14] Fogarty WM. Microbial amylase. In: Fogarty WM, editor. Microbial enzymes and biotechnology. London: Applied Science Publishers, 1983. p. 292. [15] Imam SH, BurgessCassler A, Cote GL, Gordon SH, Baker FL. A study of corn starch granule digestion by an unusually high molecular weight -amylase secreted by Lactobacillus amylovorus. Curr Microbiol 1991;22:36570. [16] MorlonGuyot J, Guyot JP, Pot B, Jacobe de Haut I, Raimbault M. Lactobacillus manihotivorans sp. nov., a new starch-hydrolyzing lactic acid bacterium isolated from cassava sour starch fermentation. Int J Syst Bacteriol 1998;48:11019. [17] deMan JC, Rogosa M, Sharpe ME. A medium for the cultivation of lactobacilli. J Appl Bacteriol 1960;123:130 5. [18] Miller GL. Use of dinitrosalicylic acid reagent for determination of reducing sugars. Anal Chem 1959;31:426 8. [19] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680 5. [20] Moseley MH, Keay L. Purication and characterisation of the amylase of B. subtilis NRRL B3411. Biotechnol Bioeng 1970;12:25171. [21] Paquet V, Croux C, Goma G, Soucaille P. Purication and characterization of -amylase from Clostridium acetobutylicum ATCC 824. Appl Environ Microbiol 1991;57:212 8.

5. Conclusion In the present study, -amylase of L. manihotivorans LMG 18010T was studied because it is produced by a new species of lactic acid bacterium that displays some interesting characteristics. Among them, L. manihotivorans LMG 18010T produces almost exclusively L( ) lactic acid from starch [16], and it grows faster than L. amylophilus, making both amylase and lactic acid production by strain LMG 18010T very attractive. Nevertheless, starch degradation by ALAB amylases could be a limiting step for lactic acid production. Therefore, it is necessary to improve our knowledges by comparing different ALAB amylases. This would allow to implement appropriate improvement strategies by genetic or protein engineering. Based on their properties and on molecular evidences [10,12], amylases produced by L. plantarum A6, L. manihotivorans, L. amylovorus, and L. amylophilus seem to belong to the same group of Lactobacillus -amylases, which differ in only minor characteristics. Notwithstanding, these minor characteristics would probably explain the different efciencies of ALAB to degrade various types of starches. For instance, it is expected that L. manihotivorans will hydrolyse more efciently starches with high amylose content. It is also evident that despite similarities between Bacillus species in terms of optimal pH and temperature, the high molecular weight of Lactobacillus amylases (approximately the double) is a distinctive characteristic. It remains anyway to understand the reasons of such differences between Lactobacillus and Bacillus amylases and the respective advantages that such differences may offer.

Acknowledgments We are indebted to Dr. Matthew George (ICAR scientist visiting India) for kindly revising the manuscript.

G. Aguilar et al. / Enzyme and Microbial Technology 27 (2000) 406 413 [22] Buonocore V, Caporale C, De Rosa M, Gambacorta A. Stable, inducible thermoacidophilic -amylase from Bacillus acidocaldarius. J Bacteriol 1976;128:51521. [23] Chung YC, Kobayashi T, Kanai H, Akiba T, Kudo T. Purication and properties of extracellular amylase from the hyperthermophilic archaeon Thermococcus profundus DT5432. Appl Environ Microbiol 1995;61:1502 6. [24] Ogasawara K, Imanishi A, Isemura T. Studies on thermophilic

413

-amylase from Bacillus stearothermophilus. J Biochem 1970;67: 6575. [25] Koch R, Spreinat A, Lemke K, Antranikian G. Purication and properties of a hyperthermoactive -amylase from the archaebacterium Pyrococcus woesei. Arch Microbiol 1991;155:572 8. [26] Pintado J, Guyot JP, Raimbault M. Lactic acid production from mussel processing wastes with an amylolytic bacterial strain. Enzyme Microb Technol 1999;24:590 8.