- Email: [email protected]
Proteolytic modification of raw-starch-digesting amylase from Bacillus circulans F-2 with subtilisin: separation of the substrate-hydrolytic domain and the raw substrate-adsorbable domain
Biochimica et Biophysica Acta, 1122(1992) 243-250
.L, '
© 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/$05.00
BBAPRO 34279
Proteolytic modification of raw-starch-digesting amylase from Bacillus circulans F-2 with subtilisin: separation of the substrate-hydrolytic domain and the raw substrate-adsorbable domain Cheorl-Ho Kim a, Suk-Tae Kwon b, Hajime Taniguchi c and Dae-Sil Lee b Genome Research Program, Genetic Engineering Research Institute, K.L S. T., Yusung-ku, Taejon (South Korea), h Laboratory of Molecular Biology, Genetic Engineering Research Institute, K.I.S. Z, Yasung-ka, Taejon (South Korea) and c Laboratory of Biocom'ersion, National Food Research Institute, Kannodai, Tsukuba-city, lbaraki (Japan)
(Received 14 February 1992)
Key words: Proteolysis: Raw-starch-digestingamylase: Amylase:Subtilisin digestion: Hydrolyticdomain: Peptide: Activesite: (B. circulans F-2) Raw starch-digesting amylase (BF-2A, 93000 Da) from Bacillus circulans F-2 was converted into two components during digestion with subtilisin. The two components were separated and designated BF-2A' (63 kDa) and BF-2B (30 kDa), respectively. BF-2A' exhibited the same hydrolysis curve for soluble starch as the original amylase (BF-2A). Moreover, the catalytic activities of original and modified enzymes were indistiguishable in K m, Vmu and in their specific activity for soluble starch hydrdysis. However, its adsorbability and digestibility on raw starch was greatly decreased. Furthermore, the enzymatic action pattern on soluble starch was differed greatly from that of BF-2A. The stability of the enzymes decreased below pH 5.5 and at 50°C, while it was quite stable even at pH 12. On the other hand, the smaller peptide (BF-2B) could be adsorbed onto raw starch. From these results, it is suggested that the larger peptide (BF-2A') has a region responsible for the expression of the enzyme activity to hydrolyze soluble substrate, and the smaller peptide (BF-2B) plays a role on raw starch adsorption and also contributes to the original enzyme-to-enzyme stabilization. A proposed model of the raw-starch-digesting enzyme from this strain is extensively discussed.
Introduction Even though many papers [1-4] have been published about amylases which can digest raw starch granules without heat gelatinization, most o f these amylases cannot effectively digest raw potato starch, one o f the most resistant starches to enzymatic digestion. In previous papers [5,6], we reported that an amylase p r o d u c e d by Bacillus circulans F-2 could effectively digest the raw starch at an appropriate rate and that it p r o d u c e d a maltohexose, preferentially from the soluble starch in the early stage of the hydrolysis. Bacillus circulans F-2 p r o d u c e d little amylase on media containing soluble carbohydrates such as glucose, maltose and soluble starch as a carbon source, though these carbon sources were rapidly utilized and good bacterial growth was obtained. Unlike o t h e r amy-
Correspondence to: C.-H. Kim, Genome Research Program, Genetic Engineering Research Institute, K.I.S.T., Taeduk Science Town, P.O. Box 17, Yusung-ku, Taeion 305-606. South Korea.
lase-producing microorganisms, this bacterium produced amylase only when raw starches were used as a carbon source, among which the raw potato starch was found to give the highest amylase production. Raw potato starch induced about 5-times as much amylase as that induced by the raw corn starch [7]. T o study the molecular structure of this amylase, we cloned and analyzed its gene [8] and found a special region in the deduced polypeptide chain that was rich in serine and threonine (data not shown). This result suggests that this amylase has a special region on its polypeptide chain which is responsible for its raw starch adsorption, as is found in some raw starch-digesting glucoamylases [9]. With these specific characteristics of this amylase, to measure raw starch digestibility, we treated this amylase with some proteinases such as Asp-endopeptidase, Lys-endopeptidase, Arg-endopeptidase and subtilisin under appropriate conditions. W h e n this amylase was digested with subtilisin, it was converted into two components (BF-2A' and BF-2B). These preliminary re-
244 suits have been described in a previous paper [10]. These assays showed that BF-2A' was a gelatinized starch-digestible amylase and BF-2B was a adsorption peptide. BF-2A' had a decreased ability to adsorb onto raw starch and had little ability to digest raw starch. These results showed that the modified amylase (BF2A') had a essential region for soluble starch digestion and a hydrolytic domain of the original amylase. This paper deals with the characteristics and properties of the subtilisin-modified amylase (BF-2A') and a smaller peptide (BF-2B) from the raw starch-digesting amylase (BF-2A) of Bacillus circulans F-2. Materials and Methods
Microorganism and cultit'ation. The microorganism was isolated from potato starch granules which were prepared in this laboratory. Isolation and characterization of Bacillus circulans F-2 was reported earlier [5]. The strain cultivation was carried out essentially according to our method described previously [5] in medium containing cross-linked potato starch. Crosslinked starch was manufactured by the cross-linking reaction between starches and epichlorohydrin according to the method of Caldwell et al. [11]. Materials. The soluble starch from E. Merck was reduced with sodium borohydride and used as a substrate on the original amylase and the subtilisin-modified amylase assays. Raw potato starch granules were obtained from Central Research Laboratories of Oji Corn-Starch. Molecular weight determination reagents on SDS-PAGE and gel filtration were purchased from Sigma (USA) and Boehringer-Mannheim, respectively. Lys-endopeptidase, Arg-endopeptidase, Asp-endopeptidase and subtilisin were obtained from Boehringer-Mannheim. All other chemicals were obtained from commercial sources. Polyacrylamide gel electrophoresis and amylase activity-staining. Polyacrylamide slab gel electrophoresis was performed according to the method of Davis [11]. SDS-PAGE was carried out essentially according to the method of Laemmli [12]. Gels were stained for protein with Coomassie brilliant blue R-250 and for amylase activity. For amylase activity staining the electrophoresised gel was incubated for 30 min at 37°C in 50 mM sodium phosphate buffer containing 2% soluble starch, and the enzymatic activity on the gel was detected by iodine staining [5]. Western blotting analysis. Preparation of rabbit antiserum against the purified B. circulans F-2 amylase (93 kDa) was conventionally performed by boosting the antigen into rabbit. SDS-immunoblotting of protein was carried out according to Gooderham et al. [14] using biotinyl-anti-IgG (Vector Laboratories) of goat and streptavidin-horse radish peroxidase system (Amersham).
Analytical method. Protein content was determined according to the method of Lowry [15] with bovine serum albumin as standard. Protein content in the eluates was routinely followed by measurement of the absorbance at 280 nm. Oligosaccharide patterns produced by the enzymatic action were examined by thinlayer chromatography (TLC), as described by Taniguchi [5]. Glycoprotein was stained on the polyacylamide gel by PAS-staining method [16]. Purification and limited proteolysis of raw starch-digesting amylase from Bacillus circulans F-2 and isolation of the components by HPLC. Bacillus circulans F-2 amylase was purified to a state showing a single band on SDS-PAGE by raw starch adsorption, ion-exchange and hydrophobic-interaction chromatography, as reported previously [5]. 50 mg of this purified amylase (93 kDa) was incubated with 0.1 mg (5 units/mg) of subtilisin in 1.6 ml of 50 mM sodium phosphate buffer (pH 7.2) at 37°C for 11 h. These digestion methods for each proteinase were described in our previous paper [10]. Gel filtration on a TSK-Gel column (Tosoh) was carried out conventionally as follows. The digested sample was isolated by high-performance liquid chromatography (HPLC), Shimadzu LC-4A on a column of TSK Gel G3000 SW-XL (1.6 x 300 mm), which had been previously equilibrated with 50 mM sodium phosphate buffer (pH 6.0) containing 300 mM sodium chloride. After application of the enzyme on the column, it was eluted with the same buffer at a flow rate of 0.5 ml/min. Adsorption of amylase onto raw starch. Adsorption onto raw starch was measured as the decrease in A280 of the supernatant with the given amounts of the enzyme as follows. Each amylase (0.3 units/ml) in 1.0 rnl of 0.25 mM sodium phosphate buffer (pH 6.0) was applied to 0.5 g of raw starch and left to stand at 4°C for 2 h. To determine the adsorbability of the enzyme, raw starch was removed by centrifugation (3000 rpm, 10 min, 4°C) and the adsorbance at 280 nm was determined in 500 tzl of the supernatant fluid. Control samples without any raw starch or without enzyme solmion were treated in the same manner. Enzyme activity assay, digestion of raw starch granules and soluble starch. The amylase activity was determined and the digestion of raw starch achieved essentially according to the method described by Taniguchi [5]. 0.3 units per ml of the original (BF-2A) and modified amylase (BF-2A') were incubated with 20 mg of raw starch granules at 37°C. The concentration of the reducing sugar (as o-glucose equivalents) in the supernatant was determined after digestion using a colorimetric method (dinitrosalicylic acid reagent) [6]. Digestion of soluble starch was carried out as follows: The digestion mixture contained 2.5 ml of 2% soluble starch solution, 0.5 ml 50 mM sodium phosphate buffer (pH 6.0), 0.5 ml deionized water and 0.3
245 u n i t s / m l o f the enzyme solution. Digestion was carried out at 37°C. Portions o f the digestion mixture were withdrawn at intervals. T h e amount of total sugars was d e t e r m i n e d with phenol-sulfuric acid [17].
0.5,
0.45:
I
o.
]i,,
0.3S0.3-
Results
.0.25.
Proteolysis of raw-starch-digesting amylase with subtilisin and isolation of the ~,o components BF-2A' and BF-2B
0.15' I
T h e purified raw-starch-digesting amylase (93 kDa) o f Bacillus circulans F-2 has b e e n incubated separately with proteolytic enzymes such as Asp-endopeptidase, Lys-endopeptidase, Arg-endopeptidase, and subtilisin (not shown, see Ref. 10). O f these enzymes, only subtilisin gave an enzymatically active amylase band o n the n a t i v e - P A G E followed by iodine-staining after extensive proteolysis. T h e chromatograms obtained during subtilisin digestion showed two increasing peaks and one decreasing peak, strongly suggesting that two predominant fragments are formed. After 11 h digestion with subtilisin, the chromatographic pattern remained constant. A preparative gel filtration on the T S K Gel G3000 S W - X L column after subtilisin digestion is presented in Fig. 1. The results o f S D S - P A G E shown in Fig. 2A clearly indicated that the original peptide (93 kDa) disappeared after subtilisin treatment and that a new peptide of 63 kDa had appeared together with a n o t h e r smaller peptide (30 kDa). We concluded that subtilisin hydrolyzed the amylase into two peptide chains and the two peptides were tentatively desig-
A 1
0.10.05-
0----~ "-" "- "--" -- ~. . 0
20
start
3
.
.
60
"1,
70
80
hated as BF-2A' (63 kDa) and BF-2B (30 kDa), respectively. Activity stainirrg (Fig. 2B) showed that a larger peptide still had the amylase activity, in contrast to the smaller one (A broad white zone observed in the lower part of lane 4 is a nonspecific band that appeared during iodine staining). Next, to characterize the modified components further, two peptides were isolated
B
2
.
30 40 50 RemnlIon time (rain)
Fig. 1. Gel chromatography of subtilisin digests of the original amylase (BF-2A) on a TSK Gel G3OO0SW-XI column. Gel filtration was carried out as described in Materials and Methods. (n), the sample before digestion with subtilisin; (•), the sample after 4 h digestion with subtilisin; (11), the sample after 8 h digestion with subtilisin. Peaks of 1, II, and III indicate BF-2A, BF-2A' and BF-2B, respectively.
C
4
MrXlO 3 97.4 66
10
MrXl03~ 1O0
.~
~ =qE
~
-
2
5O
45 20
31 10 0
I
I
I
!
0.2
0.4
0.6
O.B
1.0
R.M
Fig. 2. Polyacrylamidegel electrophoresis of the original (BF-2A) and the subtilisin-modifiedamylase(BF-2A'). Arrowheads indicate the protein (A) or the enzyme activity(B) bands. A: Protein staining on SDS-PAGE. Molecular weight markers from Sigma were used. Lane 1, the original amylase (BF-2A); lane 2, the modified amylase (BF-2A'). B: Enzyme activitystaining of the native PAGE. Lane 3, the original amylase (BF-2A): lane 4, the modified amylase (BF-2A'). C: Determination of the molecular weight of the purified BF-2A and the two components (BF-2A' and BF-2B) by SDS-PAGE. 1, the original amylase (BF-2A); 2, the modified amylase (BF-2A'); 3, peptide (BF-2B).
246 06 I~'-2A
0~ ...02
[
8_. °
__L
C Mr xl03 .g
500'
A
300:
oB
200"
.~ 06
1oo. D-
I~-2A'
04
50"
02.
~,
20 I
I
10
20
i
0
~
50
60
Retention time (rain)
0
~E
30
RT (min)
Fig. 3. High-performance liquid chromatogram of the original amylase (BF-2A) and two components (BF-2A' and BF-2B). Column, TSK-Gel G3000 SW-XI (1.6x300 ram); mobile phase, 50 mM sodium phosphate buffer (pH 6.0), containing 0.3 M sodium chloride; flow rate, 0.5 ml/min; detection, UV at 280 nm. A: Before the digestion with subtilisin. B; After the extensive digestion with subtilisin (11 h). C; Determination of the molecular weight of the original amylase (BF-2A) and two components (BF-2A' and BF-2B) by gel filtration. Molecular weight markers (Boehringer-Mannheim) used were as follows (M r x 100):A, ferritin (450); B, catalase (240); C, aldolase (160); D, bovine serum albumin (68); E, tysozyme (14). l, 2, and 3 indicate BF-2A, BF-2A' and BF-2B, respectively.
from the digestion mixture using HPLC with a TSK-Gel G3000 SW-XL column (1.6 × 300 mm). As shown in Fig. 3, the modified components eluted slower than the original amylase (BF-2A). And then, to see the molecular identity of the two isolates, each of the peptides was separated on SDS-PAGE, transferred to a nitrocellulose membrane and Western blotting analysis with rabbit antiserum against the original amylase, BF-2A, was performed. The results are shown in Fig. 4, which
1
BF-2A
2
3
4
5
6
--- 3F - 2A'
--, B F - 2 B
Fig. 4. SDS-Western blotting analysis. Protein samples were separated by 12.0% SDS-PAGE, transferred and treated with anti-BF-2A.
Reaction was carried out as described in Materials and Methods. Lanes 1-3, the purified BF-2A; lane 4, after extensive digestion with subtilisin; lane 5, the purified BF-2A'; lane 6, the purified BF-2B.
clearly indicated that two peptides are identical to the original amylase.
Enzymatic properties of modified amylase, BF-2A' (I) A;'olecular mass of the two components. The molecular masses of the two components were estimated to be about 63 and 30 kDa by SDS-PAGE (Fig. 2C). The molecular masses determined by gel-filtration chromatography on a TSK Gel G3000 SW-XL (1.6 x 300 mm) were about 61 and 30 kDa, respectively (Fig. 3C). The carbohydrate content of the peptide was checked for glycosylation status. However, no detectable value was obtained in the peptides at all (not shown), as confirmed in the previous papers [5,10]. Thus, the values obtained by SDS-PAGE might be reliable.
(2) Effect of temperature on activity and stability of modified amylase, BF-2A'. The effect of temperature on the activity o,f the modified amylase, BF-2A', is shown in Fig. 5A, which shows that the modified amylase exhibits maximum activity at 55°C, while the original amylase shows maximum activity at 59°C, indicating that thermal shift in the optimum temperature is possible due to limited proteolysis. The thermostability of the modified amylase is shown in Fig. 5B. The enzyme was unstable at higher temperature, retaining about 40% of its original activity after being incubated for 30 rain at 50°C. The enzyme showed maximum
247 A
B
A
B
lOO
100 80
8o
>,
60
'~ "G 6C
•g 60
0
40
a, 40 .>
"6
Q, 40
"6 2C tY
2(
_A
2 --Z
6
8 --f0---
6
pH
20
40 '
60 80 0 2b ' ' Temperature (*C)
4b
B-if- 80
Fig. 5. Effect of temperature on activity and stability of the purified amylases. (A) Effect on activity. Reaction mixtures containing 50 pl of the enzyme solution, 50/zl of Mcllvaine buffer (pH 6.5) and 100 /zl 2% soluble starch were incubated at various temperatures for l h and the amount of the reducing sugars produced was determined by the DNS method [5]. (B) Effect on stability. 50 # l of the enzyme solution was preincuhated for 30 min and the remaining activity was assayed as described above with addition of I00 ttl 2% soluble starch. ( o L the original amylase (BF-2A); (e), the modified amylase (BF-2A').
activity at 55°C, as shown in Fig. 5A, but it was almost inactivated when incubated at the same temperature for 30 min in the absence of the added substrate. This result indicated that the BF-2A' is more unstable tban BF-2A.
(3) Optimum pH and pH stability of the modified amylase, BF-2A'. The modified amylase, BF-2A' showed maximum activity at pH 6.5, and about 50% of maximum activity at pH 5.0 and 8.0, respectively, as shown in Fig. 6A. Stability of the BF-2A' between pH 3 and pH 12 was examined at 37°C, as shown in Fig. 6B. Under these conditions, BF-2A' was found to be quite stable at pH 12, but it was rapidly inactivated at pH values below 5.5. On the other hand, the BF-2A showed broader curve for optimum pH than the BF-2A' and exhibited pH stability even between pH 5 and pH 6.5. With above results, it was suggested that BF-2A' is more unstable than BF-2A; this probably results from the removal of the smaller peptide, BF-2B, implying that BF-2B has a important role in stabilizing the enzyme under environmental conditions.
(4) Action of the modified amylase, BF-2A' on soluble starch. Original amylase, BF-2A is a G6-forming amylase as described by Taniguchi et al. [5]. As shown in Fig 7A, in the early stage of hydrolysis, maltohexose(G 6) was the main product detected and its amount increased rapidly. Maltotetrose(G 4) and maltose(Gz) began to appear concomitantly after G 6 had accumulated to a certain degree. On the other hand, the action pattern of the BF-2A' on soluble starch was studied by thin-layer chromatography. As shown in Fig. 7B, a
8
10
12
pH
Fig. 6. Effect of pH on activity and stability of the purified amylases. (A) Effect on activity. Reaction mixtures containing 50 ~1 of the enzyme solution, 100 #1 2% soluble starch and 50 pl buffer indicated below were preincuhated at 37°C for 5 rain and the amount of reducing su~;, s produced was determined by the DNS method [5]. pH 3.0-8.0, Mcllvaine buffer; pH 8.5-11.0, Atkins-Pantin buffer [23]. (B) Effect on stability. 50 p l of the enzyme solution was preincubated with 50 p] of buffer at 37°C for 1 h. After the incubation, 100/ti of I M sodium phosphate buffer (pH 6.5) and 100 pl 2% soluble starch were added and the remaining activity was assayed as described above, the buffers were used as described above, except for pH 1.0-12.0, where 50 mM sodium phosphate buffer was used. (o), the original amylase (BF-2A); (e), the modified amylase (BF-2A').
drastic ch~:nge in the action pattern of the amylase after subtiliisin treatment was observed. Thus, the BF2A' produced a series of maltooligosaccharides from soluble sta,rch. On further incubation, maltopentose (Gs), malt, )tetrose(G4 ), maitotriose(G3) and maltose were formed with a concomitant decrease in G 9 and A
B
J
-"
G2
."--.
"
"~
IG 1 G2
G3
G4 G6
!: i.
-
0.5 5 10 15 300
0 5
G8
G9
1020/4060905
T,rne (min) Fig. 7. Thin-layer chromatography of the hydrolysis products formed from soluble starch by action of the original and modified amylases. 10-pl portions were withdrawn from standard reaction mixtures at time intervals and spotted on the thin layer plates. For experimental details of the thin-layer chromatography, see Materials and Methods. (A), the original amylase (BF-2A); (B), the; modified amylase (BF-2A').
248
I,t
1 O0
80
,o
60
cn
'~
4G
20 B
0
2
-----'-r
i
L ;|l
4
B
10
6
Discussion
24
h r of i n c u b Q t i o n Fig. 8. Effect of the original and modified amylaseson raw-starch or soluble-starch digestion. (A) The curve on soluble-starch hydrolysis. (o), the original amylase (BF-2A); (e), the modified amylase (BF2A'). (B) The curve on raw-starch hydrolysis. ([3), the original amylase (BF-2A); (11), the modified amylase (BF-2A'). Detailed experimental conditions were described in Materials and Methods.
G 8. In final stage of hydrolysis, (36 and G 7 accumulated with the disappearance of G9 and G s having mobilities close to those of G 6 and G 7 o n thin-layer chromatography.
(5) Effect of the original and modzfied amylases on raw and soluble starch-adsorbability and digestibility. Binding activities of BF-2A and BF-2A' on raw starch were compared. BF-2A was able to effectively adsorb onto insoluble polysaccharide of raw starch, but the adsorbability of BF-2A' onto raw starch decreased to about 17% of that of BF-2A. No detectable difference was seen compared with BF-2A, even though the adTABLE 1
Properties of the original and modified enzymes Enzyme
BF-2A BF-2A'
Specific activity (units/rag of protein) a
Vmax (/zg/min per ml) d
Km (%) d
62.4 64.7
8.4 7.8
0.49 0.48
sorbability of the smaller peptide, BF-2B, onto raw starch was examined (data not shown). The digestibilities of BF-2A and BF-2A' into raw starch were observed as shown in Fig. 8. The digestibility of BF-2A' decreased to about 18% of BF-2A. However, the digestibilities of the BF-2A' and BF-2A onto soluble starch were observed to be almost the same as those of BF-2A, as shown in Fig. 8. Even though K m, Vmax and specific activity for hydrolysis of soluble starch were determined for the original and modified enzyme, no differences in ar, y of these parameters were detected for the two e n ~ m e s (Table I). These results indicate that the loss of taw starch-digesting activity of BF-2A' is apparently due to its changed binding properties and that the peptide BF-2B does not have an important role in the catalytic function itself.
Relative activity soluble raw starch b
starch c
1.0 0.97
1.0 0.16
a Specific activity against soluble starch for enzyme purified to homogeneity. h Activities expressed relative to the activity of the original enzyme (BF-2A) against soluble starch, which was designated as 1. c Activities expressed relative to the activity of the original enzyme against raw starch, which was designated at 1. d Measured for the hydrolysis of soluble starch.
Raw starch-digesting amylase (BF-2A, 93 kDa) was digested with :~ubtilisin and it was degraded into two components: a soluble substrate-digestibable component (BF-2A', 63 kDa) and a peptide (BF-2B, 30 kDa). Each of them was separated to the homogeneous states by high-performance size-exclusion chromatography (Fig. 3). Several properties of the original amylase (BF-2A), the modified amylase (BF-2A') and the smaller peptide (BF-2B) were investigated. The elution pattern on a TSK Gel G3000 SW-XL column, molecular mass, the effect of temperature and pH on enzyme activities and enzyme stabilities, raw starch-digestibility, adsorbability onto -aw starch, action pattern on soluble starch, and hydrolysis for soluble starch (Figs. 3-8). Differences in enzymatic properties between the original amylase (BF - 2A) and the modified amylase (BF-2A') were observed. The modified enzyme showed differences in the elution pattern on a TSK Gel G3000 SW-XL column, in action pattern on soluble starch and in adsorbability and digestibility onto raw starch. No difference was observed in the hydrolysis curve for soluble starch. Additionally, to investigate the enzymatic properties the specific activity, K m and Vm~ for hydrolysis of soluble starch were determined for the original enzyme and the modified enzyme. Results (Table I) revealed no differences between the two forms of the enzyme, strongly suggesting that the BF-2B region does not have an important role in the catalytic or enzyme function per se. A drastic change in action pattern of the amylase after subtilisin treatment was observed. There has been no previous report on changes in raw starch digestibility of a-amylase after proteinase treatment. Therefore, this is to our knowledge the first report on change in raw starch-digestibility of a-amylase and in action pattern of amylase after limited protcolysis. The most
249 interesting results from this work are that the hydrolytic activity of the original amylase from B. circulans F-2 against soluble starch is not affected by digestion with subtilisin, whereas raw-starch digestibility rapidly decreases (Fig. 8). Furthermore, the loss in activity toward raw starch is paralleled by a decrease of enzyme adsorption onto this insoluble substrate, while the adsorbability of the smaller peptide on raw substrate was the same as that of the original amylase. From the evidence presented it appears that subtilisin splits off the terminal region of the original enzyme corresponding to the carboxyl-termini, which has a lot of serine and threonine residues. This conclusion is dependent on the amino-acid sequence data deduced from the cloned BF-2A gene (Kim et al., data not shown). The stability of the modified amylase (BF-2A') activity under environmental conditions, such as pH or thermal changes was compared with that of the original amylase, BF-2A (Fig. 6). BF-2A' largely lost its stability under the above conditions. This result can support the hypothesis that the smaller peptide stabilizes the BF-2A in the environment (Figs. 1-8). In similar work, S. Hayashida reported that the raw-starch-digestive glucoamylase I (90 kDa) of Aspergillus awamori was digested wilh subtilisin or acid proteinase of the same mold strain [18,J9] and the enzyme was converted to raw starch-indigestive glucoamylase I' (83 kDa) and a glycopeptide (7 kDa). The glycopeptide (7 kDa) showed a significant adsorbability onto raw starch. On the other hand, H.V. Tilbeurgh et al. reported that limited proteolysis of cellobiohydroiase ! (CBH-I, 65 kDa) from Trichoderma reesei by papain yields a core protein and a carboxyl-termini (10 kDa) [20]. Of them, the core protein was fully active against the small and soluble substrates. Activity against an insoluble substrate such as avicel, was completely lost. Recently, crystalline cellulose-adsorbable cellobiose oxidase from Phanerochaete chrososporium was also cleaved by papain into two domains, one of which could be responsible for adsorption onto avicel, the other being unable to be adsorbed onto cellulose [21]. Recently, some of the enzyme genes have been cloned and have shown a great variety in the structures. However, despite this, many of the different enzymes showed some interesting common features in their protein architecture [22]. As described by Tilbeurgh [20], it was considered that raw starch-digestive amylase could he proteolysed by some proteolytic enzymes, and that the carboxyl-terminal domains, consisting of the serine and threonine rich sequence, could be removed if it had a domain structure. Our BF-2A' had the same digesitibility of soluble starch as the original amylase, BF-2A, and it can weakly digest raw starch. So, it seems that the hydrolytic and the enzyme-active site must be localized in this larger
peptide, while the smaller peptide of carboxyl-terminal domain seems to play a role in substrate-binding or solubilization [22]. Therefore, a functional organisation of the original amylase (BF-2A) is proposed; one domain, corresponding to the carboxyl-terminal of the enzyme, is implicated in the adsorption process, whereas the other domain contains the hydrolytic, functional domain, corresponding to the amino-terminal region of the enzyme, even though we have no data on its amine-.terminal amino-acid sequence data yet, which can be strongly argued. The relative ease with which the terminal peptide is split off suggests that two parts of the enzyme molecule are structural domains, themselves, probably linked by some hinge region susceptible to proteolytic attack. Compared with other cases such as raw-starch-digesting glucoamylases and crystalline-cellulose-hydrolyzing cellulases, this proteolytic system absolutely coincides with them in a similar manner. In this respect, we can mention Trichoderma reesei [20] and Aspergillus awamori [18,19]. Compared with these above cases, the most outstanding features of the BF-2A' would be that BF-2A' has little adsorption (about 17%) and digestion (about 18%) activity on raw starch, whereas no other case have shown these abilities at all to date and BF-2A has no any carbohydrate moiety in the polypeptide, as determined in this study and a previous paper [10]. However, all the other cases indicated that the glycosylated carboxyl-terminal region is responsible for adsorption onto raw substrates. The site is proposed as raw substrate-affinity site. Amino-acid residues of the serine and threonine are closely related to the function of the raw substrate-affinity site. Thus, a raw starch adsorption site, as opposed to the hydrolytic site, should be considered and this would be a first report indicating that the non-glycosylated terminal region having the serine and threonine residues can be adsorbed onto raw substrate. The present results demonstrated that raw starch-digestibility of a-amylase from the bacterial strain was from its two distinct regions of a hydrolytic domain and a raw starch-adsorption domain.
Acknowledgement The authors acknowledge the late Professor Yoshiharu Maruyama, Laboratory of Biological Chemistry, Department of Agricultural Chemistry, The University of Tokyo, Tokyo, Japan for his kind direction in this work.
References 1 Mizokami, K., Kozai, M. and Kitahara, K. (1985) J. Jpn. Soc. Starch Sci. 32, 132-135. 2 Ueda, S. (1974) J. Jpn Soc. Starch Sci. 21,210-217.
250 3 Kainuma, K., lshigami, H. and Kobayashi. S. (1985) J. Jpn. Soc. Starch Sci. 32, 136-141. 4 Abe, J., Bergmann, F.W., Ohnata, K. and Hizukuri, S. (1985) J. Jpn. Soc. Starch Sci. 32, 128-133. 5 Taniguchi, H.. Odashima, F., lgarashi, M., Maruyama, Y. and Nakamura, M. (1982) Agric. Biol. Chem. 46, 2107-2115. 6 Taniguchi, H., Chung, M.-J., Yoshigi, N. and Maruyama. Y. (1983) Agric. Biol. Chem. 47, 511-519. 7 Sata, H., Taniguchi, H. and Maruyama, Y. (1986) Agric. Biol. Chem. 50, 2803-2809. 8 Kim, C-H., Sara, H., Taniguchi, H. and Maruyama, Y. (1990) Biochim. Biophys. Acta 1048, 2233-2238. 9 Hayashida, S., Nomura, T., Yoshino, E. and Hongo, M. (1976) Agric. Biol. Chem. 40, 141-146. 10 Kim, C-H., Zhang, G.W., Maruyama, Y. and Taniguchi, H. (1990) Agric. Biol. Chem. 54, 2767-2768. I1 Bernfeld, P. (1955) Methods Enzymol. 149. 12 Davis, B.J. (1970) Ann. NY Acad. Sci. 121,404-427. 13 Laemmli, U.K. (1970) Nature 227, 680-685. 14 Gooderham. K. (1983) in Techniques in Molecular Biology
(Walkerand, J.M. and Gaastra, W., eds.), pp. 49-87, Croom Helm Publishers, London. 15 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 16 Leach, B.S., Collawn, J.F. and Fish, W.W. (1980) Biochemistry 19, 5738-5742. 17 Dubois, M., Gilles, K.A., Hamillon, J.K., Rebers, P.A. and Smith, F. (1956) Anal. Chem. 28, 350-360. 18 Hayashida, S., Kunisaki, S.I., Nakao, M. and Fior, P.Q. (1982) Agric. Biol. Chem. 46, 83-89. 19 Hayashida, S., Nomura, T., Yoshino, E. and Hongo, M. (1978) Agric. Biol. Chem. 40, 141-146. 20 Tilbeurgh, H.V., Tomme, P., Claeyssens, M., Bhikhabhai, R. and Pettersson, G. (1986) FEBS Left. 204, 223-227. 21 Eriksson. G., Pettersson, G., Johansson, G., Ruiz, A. and Uzcategui, E. (1990) Ear. J. Biochem. 196, 101-106. 22 Knowles, J., Lehtovaara, P. and Teeri, T. (1987) Trends Biotechnol. 5, 255-261. 23 Atkins, W.R.G. and Pantin, C.F.A. (1926) Biochem. J. 20, 102104.
.L, '
© 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4838/92/$05.00
BBAPRO 34279
Proteolytic modification of raw-starch-digesting amylase from Bacillus circulans F-2 with subtilisin: separation of the substrate-hydrolytic domain and the raw substrate-adsorbable domain Cheorl-Ho Kim a, Suk-Tae Kwon b, Hajime Taniguchi c and Dae-Sil Lee b Genome Research Program, Genetic Engineering Research Institute, K.L S. T., Yusung-ku, Taejon (South Korea), h Laboratory of Molecular Biology, Genetic Engineering Research Institute, K.I.S. Z, Yasung-ka, Taejon (South Korea) and c Laboratory of Biocom'ersion, National Food Research Institute, Kannodai, Tsukuba-city, lbaraki (Japan)
(Received 14 February 1992)
Key words: Proteolysis: Raw-starch-digestingamylase: Amylase:Subtilisin digestion: Hydrolyticdomain: Peptide: Activesite: (B. circulans F-2) Raw starch-digesting amylase (BF-2A, 93000 Da) from Bacillus circulans F-2 was converted into two components during digestion with subtilisin. The two components were separated and designated BF-2A' (63 kDa) and BF-2B (30 kDa), respectively. BF-2A' exhibited the same hydrolysis curve for soluble starch as the original amylase (BF-2A). Moreover, the catalytic activities of original and modified enzymes were indistiguishable in K m, Vmu and in their specific activity for soluble starch hydrdysis. However, its adsorbability and digestibility on raw starch was greatly decreased. Furthermore, the enzymatic action pattern on soluble starch was differed greatly from that of BF-2A. The stability of the enzymes decreased below pH 5.5 and at 50°C, while it was quite stable even at pH 12. On the other hand, the smaller peptide (BF-2B) could be adsorbed onto raw starch. From these results, it is suggested that the larger peptide (BF-2A') has a region responsible for the expression of the enzyme activity to hydrolyze soluble substrate, and the smaller peptide (BF-2B) plays a role on raw starch adsorption and also contributes to the original enzyme-to-enzyme stabilization. A proposed model of the raw-starch-digesting enzyme from this strain is extensively discussed.
Introduction Even though many papers [1-4] have been published about amylases which can digest raw starch granules without heat gelatinization, most o f these amylases cannot effectively digest raw potato starch, one o f the most resistant starches to enzymatic digestion. In previous papers [5,6], we reported that an amylase p r o d u c e d by Bacillus circulans F-2 could effectively digest the raw starch at an appropriate rate and that it p r o d u c e d a maltohexose, preferentially from the soluble starch in the early stage of the hydrolysis. Bacillus circulans F-2 p r o d u c e d little amylase on media containing soluble carbohydrates such as glucose, maltose and soluble starch as a carbon source, though these carbon sources were rapidly utilized and good bacterial growth was obtained. Unlike o t h e r amy-
Correspondence to: C.-H. Kim, Genome Research Program, Genetic Engineering Research Institute, K.I.S.T., Taeduk Science Town, P.O. Box 17, Yusung-ku, Taeion 305-606. South Korea.
lase-producing microorganisms, this bacterium produced amylase only when raw starches were used as a carbon source, among which the raw potato starch was found to give the highest amylase production. Raw potato starch induced about 5-times as much amylase as that induced by the raw corn starch [7]. T o study the molecular structure of this amylase, we cloned and analyzed its gene [8] and found a special region in the deduced polypeptide chain that was rich in serine and threonine (data not shown). This result suggests that this amylase has a special region on its polypeptide chain which is responsible for its raw starch adsorption, as is found in some raw starch-digesting glucoamylases [9]. With these specific characteristics of this amylase, to measure raw starch digestibility, we treated this amylase with some proteinases such as Asp-endopeptidase, Lys-endopeptidase, Arg-endopeptidase and subtilisin under appropriate conditions. W h e n this amylase was digested with subtilisin, it was converted into two components (BF-2A' and BF-2B). These preliminary re-
244 suits have been described in a previous paper [10]. These assays showed that BF-2A' was a gelatinized starch-digestible amylase and BF-2B was a adsorption peptide. BF-2A' had a decreased ability to adsorb onto raw starch and had little ability to digest raw starch. These results showed that the modified amylase (BF2A') had a essential region for soluble starch digestion and a hydrolytic domain of the original amylase. This paper deals with the characteristics and properties of the subtilisin-modified amylase (BF-2A') and a smaller peptide (BF-2B) from the raw starch-digesting amylase (BF-2A) of Bacillus circulans F-2. Materials and Methods
Microorganism and cultit'ation. The microorganism was isolated from potato starch granules which were prepared in this laboratory. Isolation and characterization of Bacillus circulans F-2 was reported earlier [5]. The strain cultivation was carried out essentially according to our method described previously [5] in medium containing cross-linked potato starch. Crosslinked starch was manufactured by the cross-linking reaction between starches and epichlorohydrin according to the method of Caldwell et al. [11]. Materials. The soluble starch from E. Merck was reduced with sodium borohydride and used as a substrate on the original amylase and the subtilisin-modified amylase assays. Raw potato starch granules were obtained from Central Research Laboratories of Oji Corn-Starch. Molecular weight determination reagents on SDS-PAGE and gel filtration were purchased from Sigma (USA) and Boehringer-Mannheim, respectively. Lys-endopeptidase, Arg-endopeptidase, Asp-endopeptidase and subtilisin were obtained from Boehringer-Mannheim. All other chemicals were obtained from commercial sources. Polyacrylamide gel electrophoresis and amylase activity-staining. Polyacrylamide slab gel electrophoresis was performed according to the method of Davis [11]. SDS-PAGE was carried out essentially according to the method of Laemmli [12]. Gels were stained for protein with Coomassie brilliant blue R-250 and for amylase activity. For amylase activity staining the electrophoresised gel was incubated for 30 min at 37°C in 50 mM sodium phosphate buffer containing 2% soluble starch, and the enzymatic activity on the gel was detected by iodine staining [5]. Western blotting analysis. Preparation of rabbit antiserum against the purified B. circulans F-2 amylase (93 kDa) was conventionally performed by boosting the antigen into rabbit. SDS-immunoblotting of protein was carried out according to Gooderham et al. [14] using biotinyl-anti-IgG (Vector Laboratories) of goat and streptavidin-horse radish peroxidase system (Amersham).
Analytical method. Protein content was determined according to the method of Lowry [15] with bovine serum albumin as standard. Protein content in the eluates was routinely followed by measurement of the absorbance at 280 nm. Oligosaccharide patterns produced by the enzymatic action were examined by thinlayer chromatography (TLC), as described by Taniguchi [5]. Glycoprotein was stained on the polyacylamide gel by PAS-staining method [16]. Purification and limited proteolysis of raw starch-digesting amylase from Bacillus circulans F-2 and isolation of the components by HPLC. Bacillus circulans F-2 amylase was purified to a state showing a single band on SDS-PAGE by raw starch adsorption, ion-exchange and hydrophobic-interaction chromatography, as reported previously [5]. 50 mg of this purified amylase (93 kDa) was incubated with 0.1 mg (5 units/mg) of subtilisin in 1.6 ml of 50 mM sodium phosphate buffer (pH 7.2) at 37°C for 11 h. These digestion methods for each proteinase were described in our previous paper [10]. Gel filtration on a TSK-Gel column (Tosoh) was carried out conventionally as follows. The digested sample was isolated by high-performance liquid chromatography (HPLC), Shimadzu LC-4A on a column of TSK Gel G3000 SW-XL (1.6 x 300 mm), which had been previously equilibrated with 50 mM sodium phosphate buffer (pH 6.0) containing 300 mM sodium chloride. After application of the enzyme on the column, it was eluted with the same buffer at a flow rate of 0.5 ml/min. Adsorption of amylase onto raw starch. Adsorption onto raw starch was measured as the decrease in A280 of the supernatant with the given amounts of the enzyme as follows. Each amylase (0.3 units/ml) in 1.0 rnl of 0.25 mM sodium phosphate buffer (pH 6.0) was applied to 0.5 g of raw starch and left to stand at 4°C for 2 h. To determine the adsorbability of the enzyme, raw starch was removed by centrifugation (3000 rpm, 10 min, 4°C) and the adsorbance at 280 nm was determined in 500 tzl of the supernatant fluid. Control samples without any raw starch or without enzyme solmion were treated in the same manner. Enzyme activity assay, digestion of raw starch granules and soluble starch. The amylase activity was determined and the digestion of raw starch achieved essentially according to the method described by Taniguchi [5]. 0.3 units per ml of the original (BF-2A) and modified amylase (BF-2A') were incubated with 20 mg of raw starch granules at 37°C. The concentration of the reducing sugar (as o-glucose equivalents) in the supernatant was determined after digestion using a colorimetric method (dinitrosalicylic acid reagent) [6]. Digestion of soluble starch was carried out as follows: The digestion mixture contained 2.5 ml of 2% soluble starch solution, 0.5 ml 50 mM sodium phosphate buffer (pH 6.0), 0.5 ml deionized water and 0.3
245 u n i t s / m l o f the enzyme solution. Digestion was carried out at 37°C. Portions o f the digestion mixture were withdrawn at intervals. T h e amount of total sugars was d e t e r m i n e d with phenol-sulfuric acid [17].
0.5,
0.45:
I
o.
]i,,
0.3S0.3-
Results
.0.25.
Proteolysis of raw-starch-digesting amylase with subtilisin and isolation of the ~,o components BF-2A' and BF-2B
0.15' I
T h e purified raw-starch-digesting amylase (93 kDa) o f Bacillus circulans F-2 has b e e n incubated separately with proteolytic enzymes such as Asp-endopeptidase, Lys-endopeptidase, Arg-endopeptidase, and subtilisin (not shown, see Ref. 10). O f these enzymes, only subtilisin gave an enzymatically active amylase band o n the n a t i v e - P A G E followed by iodine-staining after extensive proteolysis. T h e chromatograms obtained during subtilisin digestion showed two increasing peaks and one decreasing peak, strongly suggesting that two predominant fragments are formed. After 11 h digestion with subtilisin, the chromatographic pattern remained constant. A preparative gel filtration on the T S K Gel G3000 S W - X L column after subtilisin digestion is presented in Fig. 1. The results o f S D S - P A G E shown in Fig. 2A clearly indicated that the original peptide (93 kDa) disappeared after subtilisin treatment and that a new peptide of 63 kDa had appeared together with a n o t h e r smaller peptide (30 kDa). We concluded that subtilisin hydrolyzed the amylase into two peptide chains and the two peptides were tentatively desig-
A 1
0.10.05-
0----~ "-" "- "--" -- ~. . 0
20
start
3
.
.
60
"1,
70
80
hated as BF-2A' (63 kDa) and BF-2B (30 kDa), respectively. Activity stainirrg (Fig. 2B) showed that a larger peptide still had the amylase activity, in contrast to the smaller one (A broad white zone observed in the lower part of lane 4 is a nonspecific band that appeared during iodine staining). Next, to characterize the modified components further, two peptides were isolated
B
2
.
30 40 50 RemnlIon time (rain)
Fig. 1. Gel chromatography of subtilisin digests of the original amylase (BF-2A) on a TSK Gel G3OO0SW-XI column. Gel filtration was carried out as described in Materials and Methods. (n), the sample before digestion with subtilisin; (•), the sample after 4 h digestion with subtilisin; (11), the sample after 8 h digestion with subtilisin. Peaks of 1, II, and III indicate BF-2A, BF-2A' and BF-2B, respectively.
C
4
MrXlO 3 97.4 66
10
MrXl03~ 1O0
.~
~ =qE
~
-
2
5O
45 20
31 10 0
I
I
I
!
0.2
0.4
0.6
O.B
1.0
R.M
Fig. 2. Polyacrylamidegel electrophoresis of the original (BF-2A) and the subtilisin-modifiedamylase(BF-2A'). Arrowheads indicate the protein (A) or the enzyme activity(B) bands. A: Protein staining on SDS-PAGE. Molecular weight markers from Sigma were used. Lane 1, the original amylase (BF-2A); lane 2, the modified amylase (BF-2A'). B: Enzyme activitystaining of the native PAGE. Lane 3, the original amylase (BF-2A): lane 4, the modified amylase (BF-2A'). C: Determination of the molecular weight of the purified BF-2A and the two components (BF-2A' and BF-2B) by SDS-PAGE. 1, the original amylase (BF-2A); 2, the modified amylase (BF-2A'); 3, peptide (BF-2B).
246 06 I~'-2A
0~ ...02
[
8_. °
__L
C Mr xl03 .g
500'
A
300:
oB
200"
.~ 06
1oo. D-
I~-2A'
04
50"
02.
~,
20 I
I
10
20
i
0
~
50
60
Retention time (rain)
0
~E
30
RT (min)
Fig. 3. High-performance liquid chromatogram of the original amylase (BF-2A) and two components (BF-2A' and BF-2B). Column, TSK-Gel G3000 SW-XI (1.6x300 ram); mobile phase, 50 mM sodium phosphate buffer (pH 6.0), containing 0.3 M sodium chloride; flow rate, 0.5 ml/min; detection, UV at 280 nm. A: Before the digestion with subtilisin. B; After the extensive digestion with subtilisin (11 h). C; Determination of the molecular weight of the original amylase (BF-2A) and two components (BF-2A' and BF-2B) by gel filtration. Molecular weight markers (Boehringer-Mannheim) used were as follows (M r x 100):A, ferritin (450); B, catalase (240); C, aldolase (160); D, bovine serum albumin (68); E, tysozyme (14). l, 2, and 3 indicate BF-2A, BF-2A' and BF-2B, respectively.
from the digestion mixture using HPLC with a TSK-Gel G3000 SW-XL column (1.6 × 300 mm). As shown in Fig. 3, the modified components eluted slower than the original amylase (BF-2A). And then, to see the molecular identity of the two isolates, each of the peptides was separated on SDS-PAGE, transferred to a nitrocellulose membrane and Western blotting analysis with rabbit antiserum against the original amylase, BF-2A, was performed. The results are shown in Fig. 4, which
1
BF-2A
2
3
4
5
6
--- 3F - 2A'
--, B F - 2 B
Fig. 4. SDS-Western blotting analysis. Protein samples were separated by 12.0% SDS-PAGE, transferred and treated with anti-BF-2A.
Reaction was carried out as described in Materials and Methods. Lanes 1-3, the purified BF-2A; lane 4, after extensive digestion with subtilisin; lane 5, the purified BF-2A'; lane 6, the purified BF-2B.
clearly indicated that two peptides are identical to the original amylase.
Enzymatic properties of modified amylase, BF-2A' (I) A;'olecular mass of the two components. The molecular masses of the two components were estimated to be about 63 and 30 kDa by SDS-PAGE (Fig. 2C). The molecular masses determined by gel-filtration chromatography on a TSK Gel G3000 SW-XL (1.6 x 300 mm) were about 61 and 30 kDa, respectively (Fig. 3C). The carbohydrate content of the peptide was checked for glycosylation status. However, no detectable value was obtained in the peptides at all (not shown), as confirmed in the previous papers [5,10]. Thus, the values obtained by SDS-PAGE might be reliable.
(2) Effect of temperature on activity and stability of modified amylase, BF-2A'. The effect of temperature on the activity o,f the modified amylase, BF-2A', is shown in Fig. 5A, which shows that the modified amylase exhibits maximum activity at 55°C, while the original amylase shows maximum activity at 59°C, indicating that thermal shift in the optimum temperature is possible due to limited proteolysis. The thermostability of the modified amylase is shown in Fig. 5B. The enzyme was unstable at higher temperature, retaining about 40% of its original activity after being incubated for 30 rain at 50°C. The enzyme showed maximum
247 A
B
A
B
lOO
100 80
8o
>,
60
'~ "G 6C
•g 60
0
40
a, 40 .>
"6
Q, 40
"6 2C tY
2(
_A
2 --Z
6
8 --f0---
6
pH
20
40 '
60 80 0 2b ' ' Temperature (*C)
4b
B-if- 80
Fig. 5. Effect of temperature on activity and stability of the purified amylases. (A) Effect on activity. Reaction mixtures containing 50 pl of the enzyme solution, 50/zl of Mcllvaine buffer (pH 6.5) and 100 /zl 2% soluble starch were incubated at various temperatures for l h and the amount of the reducing sugars produced was determined by the DNS method [5]. (B) Effect on stability. 50 # l of the enzyme solution was preincuhated for 30 min and the remaining activity was assayed as described above with addition of I00 ttl 2% soluble starch. ( o L the original amylase (BF-2A); (e), the modified amylase (BF-2A').
activity at 55°C, as shown in Fig. 5A, but it was almost inactivated when incubated at the same temperature for 30 min in the absence of the added substrate. This result indicated that the BF-2A' is more unstable tban BF-2A.
(3) Optimum pH and pH stability of the modified amylase, BF-2A'. The modified amylase, BF-2A' showed maximum activity at pH 6.5, and about 50% of maximum activity at pH 5.0 and 8.0, respectively, as shown in Fig. 6A. Stability of the BF-2A' between pH 3 and pH 12 was examined at 37°C, as shown in Fig. 6B. Under these conditions, BF-2A' was found to be quite stable at pH 12, but it was rapidly inactivated at pH values below 5.5. On the other hand, the BF-2A showed broader curve for optimum pH than the BF-2A' and exhibited pH stability even between pH 5 and pH 6.5. With above results, it was suggested that BF-2A' is more unstable than BF-2A; this probably results from the removal of the smaller peptide, BF-2B, implying that BF-2B has a important role in stabilizing the enzyme under environmental conditions.
(4) Action of the modified amylase, BF-2A' on soluble starch. Original amylase, BF-2A is a G6-forming amylase as described by Taniguchi et al. [5]. As shown in Fig 7A, in the early stage of hydrolysis, maltohexose(G 6) was the main product detected and its amount increased rapidly. Maltotetrose(G 4) and maltose(Gz) began to appear concomitantly after G 6 had accumulated to a certain degree. On the other hand, the action pattern of the BF-2A' on soluble starch was studied by thin-layer chromatography. As shown in Fig. 7B, a
8
10
12
pH
Fig. 6. Effect of pH on activity and stability of the purified amylases. (A) Effect on activity. Reaction mixtures containing 50 ~1 of the enzyme solution, 100 #1 2% soluble starch and 50 pl buffer indicated below were preincuhated at 37°C for 5 rain and the amount of reducing su~;, s produced was determined by the DNS method [5]. pH 3.0-8.0, Mcllvaine buffer; pH 8.5-11.0, Atkins-Pantin buffer [23]. (B) Effect on stability. 50 p l of the enzyme solution was preincubated with 50 p] of buffer at 37°C for 1 h. After the incubation, 100/ti of I M sodium phosphate buffer (pH 6.5) and 100 pl 2% soluble starch were added and the remaining activity was assayed as described above, the buffers were used as described above, except for pH 1.0-12.0, where 50 mM sodium phosphate buffer was used. (o), the original amylase (BF-2A); (e), the modified amylase (BF-2A').
drastic ch~:nge in the action pattern of the amylase after subtiliisin treatment was observed. Thus, the BF2A' produced a series of maltooligosaccharides from soluble sta,rch. On further incubation, maltopentose (Gs), malt, )tetrose(G4 ), maitotriose(G3) and maltose were formed with a concomitant decrease in G 9 and A
B
J
-"
G2
."--.
"
"~
IG 1 G2
G3
G4 G6
!: i.
-
0.5 5 10 15 300
0 5
G8
G9
1020/4060905
T,rne (min) Fig. 7. Thin-layer chromatography of the hydrolysis products formed from soluble starch by action of the original and modified amylases. 10-pl portions were withdrawn from standard reaction mixtures at time intervals and spotted on the thin layer plates. For experimental details of the thin-layer chromatography, see Materials and Methods. (A), the original amylase (BF-2A); (B), the; modified amylase (BF-2A').
248
I,t
1 O0
80
,o
60
cn
'~
4G
20 B
0
2
-----'-r
i
L ;|l
4
B
10
6
Discussion
24
h r of i n c u b Q t i o n Fig. 8. Effect of the original and modified amylaseson raw-starch or soluble-starch digestion. (A) The curve on soluble-starch hydrolysis. (o), the original amylase (BF-2A); (e), the modified amylase (BF2A'). (B) The curve on raw-starch hydrolysis. ([3), the original amylase (BF-2A); (11), the modified amylase (BF-2A'). Detailed experimental conditions were described in Materials and Methods.
G 8. In final stage of hydrolysis, (36 and G 7 accumulated with the disappearance of G9 and G s having mobilities close to those of G 6 and G 7 o n thin-layer chromatography.
(5) Effect of the original and modzfied amylases on raw and soluble starch-adsorbability and digestibility. Binding activities of BF-2A and BF-2A' on raw starch were compared. BF-2A was able to effectively adsorb onto insoluble polysaccharide of raw starch, but the adsorbability of BF-2A' onto raw starch decreased to about 17% of that of BF-2A. No detectable difference was seen compared with BF-2A, even though the adTABLE 1
Properties of the original and modified enzymes Enzyme
BF-2A BF-2A'
Specific activity (units/rag of protein) a
Vmax (/zg/min per ml) d
Km (%) d
62.4 64.7
8.4 7.8
0.49 0.48
sorbability of the smaller peptide, BF-2B, onto raw starch was examined (data not shown). The digestibilities of BF-2A and BF-2A' into raw starch were observed as shown in Fig. 8. The digestibility of BF-2A' decreased to about 18% of BF-2A. However, the digestibilities of the BF-2A' and BF-2A onto soluble starch were observed to be almost the same as those of BF-2A, as shown in Fig. 8. Even though K m, Vmax and specific activity for hydrolysis of soluble starch were determined for the original and modified enzyme, no differences in ar, y of these parameters were detected for the two e n ~ m e s (Table I). These results indicate that the loss of taw starch-digesting activity of BF-2A' is apparently due to its changed binding properties and that the peptide BF-2B does not have an important role in the catalytic function itself.
Relative activity soluble raw starch b
starch c
1.0 0.97
1.0 0.16
a Specific activity against soluble starch for enzyme purified to homogeneity. h Activities expressed relative to the activity of the original enzyme (BF-2A) against soluble starch, which was designated as 1. c Activities expressed relative to the activity of the original enzyme against raw starch, which was designated at 1. d Measured for the hydrolysis of soluble starch.
Raw starch-digesting amylase (BF-2A, 93 kDa) was digested with :~ubtilisin and it was degraded into two components: a soluble substrate-digestibable component (BF-2A', 63 kDa) and a peptide (BF-2B, 30 kDa). Each of them was separated to the homogeneous states by high-performance size-exclusion chromatography (Fig. 3). Several properties of the original amylase (BF-2A), the modified amylase (BF-2A') and the smaller peptide (BF-2B) were investigated. The elution pattern on a TSK Gel G3000 SW-XL column, molecular mass, the effect of temperature and pH on enzyme activities and enzyme stabilities, raw starch-digestibility, adsorbability onto -aw starch, action pattern on soluble starch, and hydrolysis for soluble starch (Figs. 3-8). Differences in enzymatic properties between the original amylase (BF - 2A) and the modified amylase (BF-2A') were observed. The modified enzyme showed differences in the elution pattern on a TSK Gel G3000 SW-XL column, in action pattern on soluble starch and in adsorbability and digestibility onto raw starch. No difference was observed in the hydrolysis curve for soluble starch. Additionally, to investigate the enzymatic properties the specific activity, K m and Vm~ for hydrolysis of soluble starch were determined for the original enzyme and the modified enzyme. Results (Table I) revealed no differences between the two forms of the enzyme, strongly suggesting that the BF-2B region does not have an important role in the catalytic or enzyme function per se. A drastic change in action pattern of the amylase after subtilisin treatment was observed. There has been no previous report on changes in raw starch digestibility of a-amylase after proteinase treatment. Therefore, this is to our knowledge the first report on change in raw starch-digestibility of a-amylase and in action pattern of amylase after limited protcolysis. The most
249 interesting results from this work are that the hydrolytic activity of the original amylase from B. circulans F-2 against soluble starch is not affected by digestion with subtilisin, whereas raw-starch digestibility rapidly decreases (Fig. 8). Furthermore, the loss in activity toward raw starch is paralleled by a decrease of enzyme adsorption onto this insoluble substrate, while the adsorbability of the smaller peptide on raw substrate was the same as that of the original amylase. From the evidence presented it appears that subtilisin splits off the terminal region of the original enzyme corresponding to the carboxyl-termini, which has a lot of serine and threonine residues. This conclusion is dependent on the amino-acid sequence data deduced from the cloned BF-2A gene (Kim et al., data not shown). The stability of the modified amylase (BF-2A') activity under environmental conditions, such as pH or thermal changes was compared with that of the original amylase, BF-2A (Fig. 6). BF-2A' largely lost its stability under the above conditions. This result can support the hypothesis that the smaller peptide stabilizes the BF-2A in the environment (Figs. 1-8). In similar work, S. Hayashida reported that the raw-starch-digestive glucoamylase I (90 kDa) of Aspergillus awamori was digested wilh subtilisin or acid proteinase of the same mold strain [18,J9] and the enzyme was converted to raw starch-indigestive glucoamylase I' (83 kDa) and a glycopeptide (7 kDa). The glycopeptide (7 kDa) showed a significant adsorbability onto raw starch. On the other hand, H.V. Tilbeurgh et al. reported that limited proteolysis of cellobiohydroiase ! (CBH-I, 65 kDa) from Trichoderma reesei by papain yields a core protein and a carboxyl-termini (10 kDa) [20]. Of them, the core protein was fully active against the small and soluble substrates. Activity against an insoluble substrate such as avicel, was completely lost. Recently, crystalline cellulose-adsorbable cellobiose oxidase from Phanerochaete chrososporium was also cleaved by papain into two domains, one of which could be responsible for adsorption onto avicel, the other being unable to be adsorbed onto cellulose [21]. Recently, some of the enzyme genes have been cloned and have shown a great variety in the structures. However, despite this, many of the different enzymes showed some interesting common features in their protein architecture [22]. As described by Tilbeurgh [20], it was considered that raw starch-digestive amylase could he proteolysed by some proteolytic enzymes, and that the carboxyl-terminal domains, consisting of the serine and threonine rich sequence, could be removed if it had a domain structure. Our BF-2A' had the same digesitibility of soluble starch as the original amylase, BF-2A, and it can weakly digest raw starch. So, it seems that the hydrolytic and the enzyme-active site must be localized in this larger
peptide, while the smaller peptide of carboxyl-terminal domain seems to play a role in substrate-binding or solubilization [22]. Therefore, a functional organisation of the original amylase (BF-2A) is proposed; one domain, corresponding to the carboxyl-terminal of the enzyme, is implicated in the adsorption process, whereas the other domain contains the hydrolytic, functional domain, corresponding to the amino-terminal region of the enzyme, even though we have no data on its amine-.terminal amino-acid sequence data yet, which can be strongly argued. The relative ease with which the terminal peptide is split off suggests that two parts of the enzyme molecule are structural domains, themselves, probably linked by some hinge region susceptible to proteolytic attack. Compared with other cases such as raw-starch-digesting glucoamylases and crystalline-cellulose-hydrolyzing cellulases, this proteolytic system absolutely coincides with them in a similar manner. In this respect, we can mention Trichoderma reesei [20] and Aspergillus awamori [18,19]. Compared with these above cases, the most outstanding features of the BF-2A' would be that BF-2A' has little adsorption (about 17%) and digestion (about 18%) activity on raw starch, whereas no other case have shown these abilities at all to date and BF-2A has no any carbohydrate moiety in the polypeptide, as determined in this study and a previous paper [10]. However, all the other cases indicated that the glycosylated carboxyl-terminal region is responsible for adsorption onto raw substrates. The site is proposed as raw substrate-affinity site. Amino-acid residues of the serine and threonine are closely related to the function of the raw substrate-affinity site. Thus, a raw starch adsorption site, as opposed to the hydrolytic site, should be considered and this would be a first report indicating that the non-glycosylated terminal region having the serine and threonine residues can be adsorbed onto raw substrate. The present results demonstrated that raw starch-digestibility of a-amylase from the bacterial strain was from its two distinct regions of a hydrolytic domain and a raw starch-adsorption domain.
Acknowledgement The authors acknowledge the late Professor Yoshiharu Maruyama, Laboratory of Biological Chemistry, Department of Agricultural Chemistry, The University of Tokyo, Tokyo, Japan for his kind direction in this work.
References 1 Mizokami, K., Kozai, M. and Kitahara, K. (1985) J. Jpn. Soc. Starch Sci. 32, 132-135. 2 Ueda, S. (1974) J. Jpn Soc. Starch Sci. 21,210-217.
250 3 Kainuma, K., lshigami, H. and Kobayashi. S. (1985) J. Jpn. Soc. Starch Sci. 32, 136-141. 4 Abe, J., Bergmann, F.W., Ohnata, K. and Hizukuri, S. (1985) J. Jpn. Soc. Starch Sci. 32, 128-133. 5 Taniguchi, H.. Odashima, F., lgarashi, M., Maruyama, Y. and Nakamura, M. (1982) Agric. Biol. Chem. 46, 2107-2115. 6 Taniguchi, H., Chung, M.-J., Yoshigi, N. and Maruyama. Y. (1983) Agric. Biol. Chem. 47, 511-519. 7 Sata, H., Taniguchi, H. and Maruyama, Y. (1986) Agric. Biol. Chem. 50, 2803-2809. 8 Kim, C-H., Sara, H., Taniguchi, H. and Maruyama, Y. (1990) Biochim. Biophys. Acta 1048, 2233-2238. 9 Hayashida, S., Nomura, T., Yoshino, E. and Hongo, M. (1976) Agric. Biol. Chem. 40, 141-146. 10 Kim, C-H., Zhang, G.W., Maruyama, Y. and Taniguchi, H. (1990) Agric. Biol. Chem. 54, 2767-2768. I1 Bernfeld, P. (1955) Methods Enzymol. 149. 12 Davis, B.J. (1970) Ann. NY Acad. Sci. 121,404-427. 13 Laemmli, U.K. (1970) Nature 227, 680-685. 14 Gooderham. K. (1983) in Techniques in Molecular Biology
(Walkerand, J.M. and Gaastra, W., eds.), pp. 49-87, Croom Helm Publishers, London. 15 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 16 Leach, B.S., Collawn, J.F. and Fish, W.W. (1980) Biochemistry 19, 5738-5742. 17 Dubois, M., Gilles, K.A., Hamillon, J.K., Rebers, P.A. and Smith, F. (1956) Anal. Chem. 28, 350-360. 18 Hayashida, S., Kunisaki, S.I., Nakao, M. and Fior, P.Q. (1982) Agric. Biol. Chem. 46, 83-89. 19 Hayashida, S., Nomura, T., Yoshino, E. and Hongo, M. (1978) Agric. Biol. Chem. 40, 141-146. 20 Tilbeurgh, H.V., Tomme, P., Claeyssens, M., Bhikhabhai, R. and Pettersson, G. (1986) FEBS Left. 204, 223-227. 21 Eriksson. G., Pettersson, G., Johansson, G., Ruiz, A. and Uzcategui, E. (1990) Ear. J. Biochem. 196, 101-106. 22 Knowles, J., Lehtovaara, P. and Teeri, T. (1987) Trends Biotechnol. 5, 255-261. 23 Atkins, W.R.G. and Pantin, C.F.A. (1926) Biochem. J. 20, 102104.