297 1. Appl. Glyeosei., 51,297-301 (2004)
© 2004 The Japanese Society of Applied Glycoscience
Hydrolysis of Water-Soluble and Water-Insoluble Cellulosic Substrates by
Endo-,B -1,4-Glucanase from Acetobacter xylinum
(Received April 3D, 2004 ; Accepted June 9, 2004)
Fuyu Ito,' Yoshihiko Amano,1 Kouichi Nozaki,' Inder M. Saxena,2
2 Malcolm R. Brown Jr. and Takahisa Kanda'
'Department of Chemistry and Material Engineering, Faeulry of Engineering, Shinshu University
(4--17-1, Wakasato, Nagano 380-8553, Japan)
'School of Biological Science, The University of Texas at Austin (Austin, Texas 78712, USA)
Abstract: Morphology changes in bacterial cellulose produced by Acetobacter xylinum ATCC23769 were ob served in the presence of /3 -glucodisaccharides such as gentiobiose and cellobiose. Endo-/3 -1,4-glucanase ac tivity in culture broth was higher than that in the absence of those sugars. So we have investigated the prop erties of endo-/3 -1,4-glucanase (AEG) produced by this bacterium. This enzyme could hydrolyze water-soluble cellulose such as CMC, hydroxyethyl cellulose and cellodextrin, and decreased the viscosity of the substrate solution. On the other hand, AEG could not produce any soluble sugars from water-insoluble cellulose such as Avicel and bacterial cellulose. These properties were completely different from endo-glucanase from fungi. AEG hydrolyzed cellohexaose and produced cellobiose, cellotriose and cellotetraose, but in the presence of bacterial cellulose, the soluble sugars produced from cellohexaose disappeared in the reaction mixture. It is suggested that AEG might have transglycosyl activity, though it belongs to glycosidase family 8. It is proposed that this activity is closely related to cellulose synthesis. Key words: Acetobacter xylinum, endo-/3 -1,4-glucanase, Irpex lactells
A. xy/inum is a very simple and convenient system for bated at 25°C for 15 days under the static condition. studying the biochemical genetic aspects of cellulose bio Enzymes. An extracellular enzyme produced by A. synthesis. Why does A. xylinum synthesize cellulose? xyiinum (designated AEG) was obtained by the following. One possibility is that this bacterium would use cellulose Be-free cells and the culture supernatant were recovered as a form of glucose storage and its polymerizations cause from the broth by filtration through a nylon membrane a reduction in osmotic strength. If this is the case, A. (l3-XXlOO, SEFAR, Swiss). The filtrate was centrifuged xyhnum must also encode a cellulase gene. Various re at 30,000 X 9 for 15 min at 4"C. The supernatant was ports have described the production of cellulolytic en treated with ammonium sulfate at 80% saturation, the pre zymes by Aeetobacler strains. -), One of their genes en cipitates formed were dissolved in 50 !TIL sodium acetate codes the endo-/5 -1 ,4-glucanase (CMCax) that can decom buffer (pH 5.0)_ The solution was dialyzed for 24 h at pose cellulose. Because cellulose biosynthesis in Aceto 4 C against distilled water and was used as the crude en zyme solution. The fungal en do-type cellulase (designated baeler )"ylinum was inhibited by the addition of the anti En- I) used in the present work was obtained from Drise CMCax serum to its growth medium," this enzyme is thought to be essential for cellulose biosynthesis. How lase, a commercial product of Irpex iaetetts, manufactured ever, the relation between cellulose biosynthesis and cel by Kyowa Hakko (Tokyo, Japan) according to procedures lulolytic enzyme has not yet been established. previously reported_'" From this point of view, it has been expected to make Substrates. Avicel, a microcrystalline cellulose pow
the properties of endo-glucanase clear, but there are few der (Alt. 2331), was purchased from Merck (Darmstadt,
Germany). Carboxymethyl cellulose (CMC) and ostazin
reports about the mode of action of Aeetobaeler-cellulase brilliant red-hydroxyethyl cellulose (OBR-HEC) were pur
on various cellulosic materials. We have investigated the action pattern of this enzyme on water-soluble and chased from Sigma Chemical (USA). Cotton was prepared
-insoluble substrates. as described by Hoshino et ai.;) Phosphoric acid-swollen
cotton (HC) and Avicel (HA) were prepared as described
MATERIALS AND METHODS by Wood.' Tamarind xyloglucan was donated by T.
Hayashi, Kyoto University. Cellohexaose (G6) and p Culture. A. xylinum ATCC 23769 was used in this nitrophenyl jJ-D-glucopyranoside (PNPG) were purchased study by picking up "rough colonies" in 5-days-old agar from Seikagaku Kogyo (Tokyo, Japan). plate. For seed culture, Hestrin and Schramm (SH) me Protein measurements. Enzyme protein was deter dium was used." The liquid medium (1000 mL) was ster mined by the Lowry method 9 ) using bovine serum albu ilized at 121°C for 15 min, and inoculated and then incumin as the standard protein.
Saccharification activity on water-soluble cellulose. >
Corresponding author (Tel. & Fax.
[email protected]).
--81--26--269-5394, E-mail:
The reaction mixture contained O. I !TIL of I _Owt% CMC solution, O. I mL of enzyme solution, and 0.2 !TIL of 0.05
298
1. Appl. Glycosci., Vol. 51, NO.4 (2004)
M sodium acetate buffer (pH 5.0) in a total volume of 0.4
mL. After incubation at 30T for 24 h, reducing sugars produced per mL of the reaction mixture were determined by the method of Somogyi'OI-Nelson.") One unit was de fined as the amount of enzyme producing 1 p. mol of glu cose per min.
Saccharification activity on water-insoluble cellulose. The reaction mixtures contained 0.1 mL of a substrate (l wt% Avicel, 0.1 wt% HA, HC or 0.25wt% BC), 0.1 mL of enzyme solution, and 0.2 mL of 0.05 \1 sodium acetate v buffer pH 5.0. Incubation was conducted at 30 C with mechanical shaking (l00 strokes/min) for 24 h, and the mixtures were filtered through a glass filter. Reducing sugars produced were measured in the same way as the above. One unit was defined as the amount of enzyme producing 1 ,umol of glucose per min. p -Nitrophenyl f3 -glucoside hydrolyzing activity. A reaction mixture contained 0.1 mL of 5 ffiJV1 of pNPG, 0.1 mL of enzyme solution, and 0.2 mL of 0.05 M sodium acetate buffer, pH 5.0. After incubation at 30°C for 24 h, 1.0 mL of 1.0wt% Na,C0 3 and 2.0 mL of distilled water were added to the mixture. The amount of p-nitrophenol liberated was measured at 420 nm. One unit was defined as the amount of enzyme producing I ,umol of p nitrophenol per min. Hydrolyzing activity on aBR -HEe. The reaction mixture contained 0.35 mL of 1.0wt% OBR-HEC solu tion, 0.35 mL of enzyme solution, and 0.7 mL of 0.05 M sodium acetate buffer, pH 5.0. After incubation at 30°C for 24 h, 0.4 mL of the reaction mixture was mixed with 1.2 mL of acetone. The precipitated substrate was re moved by centrifugation and the absorbance of the super natant was measured at 550 nm. One unit was defined as the amount of enzyme changing the absorbance (0.05/h). Viscometry of degradation of CMC and XG. The ac tivity was assayed viscometrically in a viscometer at 30°C with 1 !TIL of enzyme solution and 4 mL of 0.05 M so dium acetate buffer (pH 5.0) and 1.0 mL of 1.0wt% CMC or 0.5wt% XG. One unit of activity was defined as the amount of enzyme required to cause a 10% decrease in relative viscosity for 24 h under these conditions.
Molecular weight (Mw) distribution of CMC and XG. The molecular weight distributions of CMC and XG were determined using HPLC (801; JASCO, Tokyo, Ja pan). They were detected using a RI monitor (410; Wa ters, USA) equipped with a G 3000 PWXL column (7.8 X 300 mm, Tosoh, Tokyo, Japan). The mobile phase is dis tilled water at the flow rate of 0.8 mL/min. MOlecular weight was also determined using Shodex STANDARD P-82 (Hayashibara Biochemical Laboratories, Japan) as the standard carbohydrate. Thin -layer chromatography (TLC). The reaction mixture consisted of 0.1 mL of enzyme solution, 0.2 mL of 5 mg/mL cellohexaose or 5 mg/mL cellohexaose, to which was added 1 mg/mL bacterial cellulose and 0.1 mL of 0.05 M sodium acetate buffer at pH 5.0. After in cubation for appropriate periods, 30,uL aJiljuots of reac tion mixture and authentic sugars (ccllo-oligosaccharides) solution were spotted individually. Analytical TLC was performed with silica-gel 60 (0.5 mm thickness, Merck) in the solvent system of chloroform-methanol-water (90:
65 : 15). The resolved sugars were detected by heating the plate at 120°C for 10 min after spraying with 30% sulfu ric acid.
RESULTS
Enzyme activity in crude enzyme preparation. The protein content of the crude enzyme preparation (AEG) was determined to be 1.43 mg/mL. Table 1 sum marizes the hydrolyzing activity of AEG against water soluble and insoluble substrates. AEG hydrolyzed water soluble cellulose, but did not hydrolyze water-insoluble cellulose. AEG also containing j.1-glucosidase as p NPG was degraded. On the other hand, En-1 hydrolyzed water soluble and insoluble cellulose as reported previously.'2 "I
Hydrolysis of CMC by AEG and En-I. The time courses of hydrolysis of CMC were compared with typical endo-glucanase En-I from I. laeteus (Fig. I). En-1 hydrolyzed CMC rapidly at the initial stage, and de creased the viscosity of CMC following the typical curve of endo cellulase. On the other hand, AEG decreased the viscosity very slowly. The molecular weight distribution of CMC was determined using HPLC (Fig. 2, Table 2). It was estimated at about 180,000 initially, and decreased to 300 after treatment with En-l for 24 h. This suggests that the main products produced by En-l were celJo oligosaccharides. On the other hand, the elution profile of CMC treated with AEG revealed that the original peak of CMC shifted to three clear peaks showing low molecular weights. Their molecular weights were estimated as 180,000, 50,000 and 5000 respectively. It is suggested this enzyme cleaved specific glycosidic bonds of CMC and then stopped. From these results, the mode of action of AEG was completely different from En-I.
Table 1. Specific activities (SP) of AEG against various sub strates. SP
Substrate (water insoluble)
(X 10-' U/mg)
0.66
AviceJ
0
0.4
H,PO,-Avicel
0
OBR-HEC
0.68
H,PO,-cotton
0
pNPG
1.06
BC
0
Substrate (water soluble)
(X 10-' U/mg)
CMC XG
SP
0.8 ~
~
~
E C
'v; 0
:ii
;;:
0.6
AEG
0.4 0.2 0
f\ 0
En-l
2
[
Time (h) Fig. 1. Viscosity changes of CMC during incubation with AEG and En-I. v
The activity was assayed viscometrically at 30 C with an enzyme solution and 1.0wt% CMC. One unit of activity was defined as the amount of enzyme required to cause a 100/< decrease in relative vis cosity for 24 h under these conditions.
299
Hydrolysis of Cellulosic Substrates by Endo-;S' -I ,4-Glucanase from Acetobacter xylinum 0.4
AEG
En-I
.~0.2 o
() U)
:>
AEG
o
o Fig. 3.
Oh
~I
~)!\~
®
24
48 Time (h)
96
72
Viscosity changes of XG during incubation with AEG and En-I.
Experimental dc;tails are the same as described in Fig. 1.
~~
;JC ~.=:J[
o
7.5
15.0
o
7.5
15.0
Retention time (min)
Fig. 2. HPLC pattern of hydrolysis products from CMC by AEG and En-I. The molecular weight distributions of CMC were deterrnined us ing HPLC. Numbers and arrows indicate molecular weight markers.
CD,
180,000; 0), 50,000; Q), 5000; @, 300.
Table 2. Molecular weight (Mw) changes of CMC during in cubation with AEG and En-I. AEG Mw
En-l
oh
48 h
Oh
(1)180000
60%
55%
60%
0)50000
40%
25%
40%
0)5000
48 h
o
15.0
o
7.5
15.0
Retention time (min)
18%
20%
@300
7.5
82%
Fig. 4. HPLC pattern of hydrolysis products from XG by AEG and En-I. Experimental details are the same as described in Fig. I. Num bers and an'ows indicate molecular weight markers. (1), 9,000,000;
0), 380,000; Q), 120,000; @, 60,000.
Hydrolysis of XG by AEG and En-I. The time courses for the hydrolysis of the XG by endo glucanases were examined viscometrically (Fig. 3). Both AEG and En-l decreased the viscosity of the XG solution very slowly, but AEG decreased to a larger extent than En-I, and this result was the reverse of CMC degradation. The molecular weight of XG was estimated to be about 9,000,000 initially using HPLC (Fig. 4, Table 3). The mo lecular weight of XG after treatment with AEG or En-l shifted from 9,000,000 to 380,000, 120,000 and 60,000 after 48 h. From these results, the modes of action of both enzymes were almost similar to each other.
Degradation of cellohexaose (G 6) with or without bac terial cellulose. The hydrolysis products from Go by AEG were ana lyzed by TLC (Fig. 5 (A)). The enzyme produced cel lotriose in the early stage, but cellobiose and cellotetraose were also detected after 6 h incubation. This enzyme could not attack cellodextrin from cellobiose to cellopen taose and accumulated in the reaction mixture, though cel lopentaose was degraded after prolonged incubation. This suggests that AEG required more than a hexaose unit of glucose for catalysis. The degradation pattern of Go in the presence of BC is shown in Fig. 5 (B). Although the hy drolysis products were similar to those from Go only, the
Table 3. Molecular weight (Mw) changes of XG during incu bation with ALG and En-I. En-I
AEG Mw
oh
(1)9000000
70%
0)380000
30%
48 h
oh
67%
30%
48 h
70% 15vo
Q) I20000
30%
70%
@60000
3%
15%
amount of the hydrolysis products in the presence of bac terial cellulose obviously decreased compared with prod ucts from Go only. Interestingly, cello-oligosaccharides produced at the initial stage disappeared after prolonged incubation with AEG. It is suggested that AEG has a transglycosyl activity especially in the presence of insol uble cellulose.
DISCUSSION Cellulase activity was detected in the culture broth of A. xylinum, as some researchers have reported. v ; We in vestigated the properties of cellulase, especially the mode of action on various cellulosic substrates, as the role of
300
J. Appl. G/ycosci., Vol. 5 l, No.4 (2004) (8) ..........."..._-.-_---.,..---
(A)
•
Gl G2 G3 G4 G5 G6
S
0
0.5
3
6
12
24
S
Time (h) Fig. 5.
S
0
3
6
12
24
48
S
Time (h)
Thin-layer chromatograms of hydrolysis products from cellohexaose (A) and cellohexaose in the pres ence of bacterial cellulose (B) by ALG.
The reaction mixture consisted of 0.1 mL of enzyme solution, 0.2 mL of 5 mg/mL cellohexaose or 5 mg/ml. cellohexaose with the addition of J mg/mL bacterial cellulose and 0.1 mL of buffer. After incubation for appro priate periods, 30 pL aliquots of the reaction mixture and authentic suga.r (cello-oligosaccharides) solution were spotted individually. Symbols: S, standard sugars; G , glucose; G, cellobiose; G" cellotriose; G., cellotetraose; G" cellopentaose; G6, cellohexaose.
cellulase for cellulose synthesis of A. xylinum was not es tablished. Cellulase is defined as the enzyme that degrades cellulose and produces cello-oligosaccharides. However, Acetobacter cellulase could not degrade ceJlo oligosacharides less than DP 5. which is the smallest sub strate as a completely soluble sugar. Furthermore. it could not produce soluble sugars from insoluble celluloses. such as A vice]. bacterial cellulose and phosphoric acid-swollen cellulose. We have the questions of which substrates this cellulase reacts with and how it degrades cellulose. One possibility was derived from our results of cellohexaose degradation. Three products 0,>02=0. were formed from 0 6 that was partially solubilized in water. but it had a weak crystalline structure. From these results, Acelobac fer cellulase is an endo-glucanase, but is diffcrent from fungal endo-glucanases such as En-l from lrpex lacleus." It is interesting that A. xylinum produced enzymes that have transglycosyl activity. because cello-oligosaccharides were reused and disappeared in the reaction mixture of cellohexaose and bacterial cellulose. This phenomenon was observed only in the presence of bacterial cellulose that had the same conditions as the A. xylinul11 culture. IL is reported that the endo-glucanase from A. xylinum be longs to glycosidase family 8, which is an inverting en zyme (http://afmb.cnrs-mrs.fr/CAZYIOH_8.html). As inverting enzymes merely catalyze the transglycosyl reac tion, this reaction rrlight be catalyzed by other enzymes because we did not purify this enzyme completely. An other possibility of the cellulase action in the culture is acetan degradation. as it degrades xyloglucan, which is a sirrlilar structure to acetan. Both substrates have the 1,4-g1ucan backbone attached side chain composed of rrlixed oligosaccharides. Indeed, we detected some sugars composed of side chains such as rhamnose. mannose and gentiobiose (data not shown). The cellulase from A. xylinum did not attack purified
/3
bacterial cellulose produced by itself when we detected reducing sugars produced. However, it is repoJ1ed that cellulose produced under the condition that it produced more cellulose than usual has a low degree of polymeriza tion.": We also measured the DP of bacterial cellulose when it was cultivated in the presence of glucodisaceharides such as gentiobiose and cellobiose. The DP of bacterial cellulose produced in the presence of ,3-glucodisaccharides was lower than that in the absence of those sugars. Furthermore, cellulase activity in the cul ture broth was also higher than usual (data not shown). From these results. we propose that the cellulase is very important for cellulose production, and it might attack bacterial cellulose though it does not release soluble prod ucts. Matthysse el al. 16 . 17 ) reported that an endoglucanase gene (designated ceIC), homologous to A. xylinum ATeC 23769, is contained in an operon of cellulose synthase in Agrobacferium lumefaciens. and that there is a transposon insertion in celC blocks cellulose synthesis, indicating that the production of CMC-hydrolyzing activity may be specifically associated with cellulose production. In higher plants, the activities of endo-/J-l,4-glucanase are closely related to various physiological aspects of plant growth. Moreover, Tonouchi et al. ". reported that cellulose pro duction by strain BPR2001 was enhanced by the addition of a small amount of an endo- /3-1 ,4-glucanase from Ba cillus sublilis. Cellulase, especially endo-/),-I,4-glucanase, is somehow related to the biosynthesis of cellulose, The decrease in the DP of a polymer material is generally thought to influence its quality. We expect that AEG can contribute to the unique physical properties of Be.
/3
This work was supposed by Grants-in-Aid for 21st Century COE Program by the Ministry of Education. Culture, Sports, Science, and Technology. We wish to thank Dr. T. Hayashi of the Wood
Hydrolysis of Cellulosic Substrates by Endo-,9-1,4-Glucanase from AcelObacter xylinum Research Institute, Kyoto University, for supplying the xyloglucan of tamarind.
REFERENCES I) W.E. Husemann and R. Werner: Cellulosesynthese durch Ace tobacter xylinum. I. Ober Molekulargewicht und Molekularge wichtsvel1eillung von Bakteriencellulose in Abhangigkeit von der Synthesedauer. Makromo!. Chem., 59, 43-60 ([ 963). 2) R. Standal, TG. Iversen, D.H. Coucheron, E. Fjaervik, J.M. Blantny and S. Valla: A new gene required for ceJlulose pro duction and a gene encoding cellulolytic activity in Acerobac ter xylinum are colocalized with the bcs operon. 1. Bacteriol., 176, 665-672 (1994). 3) T Okamoto, S. Yamamoto, H. Ikeaga and K. Nakamura: Cloning of the Acetobacter xylinum cellulase gene and its ex pression in Escherichia coli and Zymomonas mobilis. App!. Microbio!. Biotechnol., 42,563-'568 (1994). 4) H.M. Koo, S.H. Song, Y.R. Pyun and Y.S. Kim: Evidence that a ,9-1,4-endoglucanase secreted by Acetobacter xylinum plays an essential role for the fonnation of cellulose fiber. Biosei. Biotechnol. Biochem., 62, 2275-2259 (1998). 5) S. Hestrin and M. Schramm: Synthesis of cellulose by Aceto bacter xylinum 1. Micromethod for the determination of cellu lose. Biochem. J., 56, 163-[66 (1954). 6) T Kanda, S. Nakakubo, K. Wakabayashi and K. Nisizawa: Purification and properties of a lower-molecular-weight endo cellulase from Irpex lacteus (Polyporus tulipijerae). 1. Bio chem.,84, 1217-,1226 (1978). 7) E. Hoshino, T. Kanda, Y. Sasaki and K. Nisizawa: Adsorption mode of exo- and endo-cellulase from Irpex lacteus (Poly porus tulipiferae) on cellulose with different crystallinities. 1. Biochem., 111, 600-605 (1992). 8) TM. Wood: Preparation of crystalline, and dyed cellulose sub strates. Methods in Enzymology, W.A. Wood and S.T Kel logg, eds., Academic Press, New York, pp. 19-25 (1988). 9) O.H. Lowry, N.J. Rosebrough, A.L. Fan and R.J. Randall: Protein measurement with the Folin phenol reagent. 1. Bioi. Chem., 193, 265-275 (1951). 10) M. Somogyi: Notes on sugar detelmination. 1. Bioi. ClleIn., 195, 19-23 (1952). II) N. Nelson: A photometric adaptation of Somogyi method for the detennination of glucose. 1. Bio!. Chem., 153, 375-380 (1944). 12) T Kanda, K. Wakabayashi and K. :--Jisizawa: Purification and properties of a lower-molecular-weight endo-cellulase from 11' pex lacteus (Polyporus tulipijerae). 1. Biochem., 87, 1625- 1634 (1980) 13) T Kanda, Y. Amano, M. Shiroishi and E. Hoshino: Mode of action of exo- and endo-type cellulase from fungi in the hy drolysis of various substrates. 1. Appl. Glycosei., 41, 273--282 (1994). 14) M. Shiroishi, Y. Amano, E. Hoshino, K. Nisizawa and T. Kanda: Hydrolysis of various cellulose, (I ~3), (1->4)-/9-0 glucans, and xl' log lucan by three endo-type cellulases. Moku zai Gakkaishi, 43,178-187 (1997). IS) N. Tahara, M. Tabuchi, K. Watanabe, H. Yano, Y. MOIinaga and F. Yoshinaga: Degree of polymerization of cellulose from Acetobacter xyLinum BPR2001 decreased by cellulase pro duced by the strain. Biosei. Bioteehnol. Biochem., 61, 1862 1865 (1997). 16) A.G. Matthysse, S. White and R. Lightfoot: Genes required for
301
cellulose synthesis in Agrobacterium tumefaciens. 1. Bacte riol., 177, 1069-1075 (1995). 17) A.G. Matthysse, D.L. Thomas and A.R. White: Mechanism of cellulose synthesis in Agrobacterium tumefaeiens. 1. Bacte riol., 177,1076-1081 (1995). 18) N. Tonouchi, N. Tahara, T. Tsuchida, F Yoshinaga, T Beppu and S. Horinouchi: Addition of small amount of an endogluca nase enhances cellulose production by Acetobacter xylinum. Biosci. Biotechno!. Biochem., 59, 805--808 (1995).
-tz )[; D
- 7.. ~£ 00 Acetobacter xylinum (J) ~£ 9" Q
I:--
~'-fJ -1,4-1')[;1.77- iZ·(J)*~'t1s.tV'
=1'~'t1¥~':M9"Qfl=ffl 1f1'!ipJ\::$T', :.Riff Bt~', f)'I!Jjij'Jj]-', Inder M. Saxena',
Malcolm R. Brown Jr 2
,
t$83r.~/\'
, 181+1:**I*gMm14tI*H
(380-8553 :&f)'m::S~ 4-17-])
'School of Biological Science, The University of
Texas at Austin
(Austin, Texas 78712, USA)
1:: }v 0 . - :A. IL);if I:ii -C' 06 Ut ~m (;I:, l:il1:f5'H=-.L /' /--" (3 -1,4- 7' )v j; T - --c' ~ 5j'-i£, -9' 6, i t.:. If /' T ;t ~' ;;j- - :A. -\"1:: 0 ~';;j- -:A. 7J: C"O)tJ1l'~jJ±iJJ.9=J 1::.1JO.z 6 C 7f~!'L~1J'~1t L, 2' G 1>(-0) C !§ O)m1:f5'i-~*.L /' V-P'-1,4-7"}v j; T - --c'ititJ: 'b ~
