ISSN 0003-6838, Applied Biochemistry and Microbiology, 2006, Vol. 42, No. 6, pp. 573–583. © MAIK “Nauka /Interperiodica” (Russia), 2006. Original Russian Text © A.V. Markov, A.V. Gusakov, E.I. Dzedzyulya, B.B. Ustinov, A.A. Antonov, O.N. Okunev, A.O. Bekkarevich, A.P. Sinitsyn, 2006, published in Prikladnaya Biokhimiya i Mikrobiologiya, 2006, Vol. 42, No. 6, pp. 654–664.
Properties of Hemicellulases of the Enzyme Complex from Trichoderma longibrachiatum A. V. Markova, A. V. Gusakova, E. I. Dzedzyulyaa, B. B. Ustinova, A. A. Antonova, O. N. Okunevb, A. O. Bekkarevichb, and A. P. Sinitsyna a Faculty
of Chemistry, Moscow State University, Moscow, 119899 Russia e-mail:
[email protected] b Institute of Biochemistry and Physiology of Microorganisms, Russian Academy of Sciences, Moscow Oblast, Pushchino, 142292 Russia e-mail:
[email protected] Received January 10, 2006
Abstract—Six xylan-hydrolyzing enzymes have been isolated from the preparations Celloviridin G20x and Xybeten-Xyl, obtained earlier based on the strain 1 Trichoderma longibrachiatum (Trichoderma reesei) TW-1. The enzymes isolated were represented by three xylanases (XYLs), XYL I (20 kDa, pI 5.5), XYL II (21 kDa, pI 9.5), XYL III (30 kDa, pI 9.1); endoglucanase I (EG I), an enzyme exhibiting xylanase activity (57 kDa, pI 4.6); and two exodepolymerases, β-xylosidase (β-XYL; 80 kDa, pI 4.5) and α-L-arabinofuranosidase I (α-LAF I; 55 kDa, pI 7.4). The substrate specificity of the enzymes isolated was determined. XYL II exhibited maximum specific xylanase activity (190 U/mg). The content of the enzymes in the preparation was assessed. Maximum contributions to the total xylanase activities of preparations Celloviridin G20x and Xybeten-Xyl were made by EG I and XYL II, respectively. Effects of temperature and pH on the enzyme activities, their stabilities under various conditions, and the kinetics of exhaustive hydrolysis of glucuronoxylan and arabinoxylan were studied. Combinations of endodepolymerases (XYL I, XYL II, XYL III, or EG I) and exodepolymerases (α-LAF I or β-XYL) produced synergistic effects on arabinoxylan cleavage. The reverse was the case when endodepolymerases, such as XYL I or EG I, were combined with α-L-AF I. DOI: 10.1134/S000368380606007X
Xylan accounts for up to 30% of the total polysaccharides in the cell walls of plants [1]. Depending on the source and means of isolation, natural xylans may have linear or branched structures [1, 2]. Among natural xylans, the most widespread species are glucuronoxylan—the major xylan of hardwood (deciduous) tree species—and arabinoxylan, which is a component of softwood (coniferous) tree species and cereals [2]. Enzymatic hydrolysis of natural xylans is achieved with the participation of hemicellulases, endo-1,4-βxylanases (EC 3.2.1.8), β-xylosidases (EC 3.2.1.37), αL-arabinofuranosidases (EC 3.2.1.55), and α-glucuronidases (EC 3.2.1.139) [3–10]. The fungi of the genus Trichoderma rank first among hemicellulase-forming microorganisms, because of their high secretory capacity and the diversity of the enzymes produced [11, 12]. Consequently, enzyme preparations involving Trichoderma strains enjoy universal use in a variety of biological processes [13–17]. The Russian preparations Celloviridin G20x and Xybeten-Xyl, obtained using the strain T. longibrachiatum (T. reesei) TW-1, compete successfully with analogs manufactured by premier world suppliers of enzymes. In addition to serving as supplements to poultry and animal forage, they are used in the industrial production of ethanol (for increasing the yield of starch
obtained from grain raw material), beer brewing, and breadmaking [18–23]. In all the processes indicated, the positive effect of the preparations is related to hydrolysis of nonstarch polysaccharides, such as xylans and β-glucans [24, 25]. However, the properties of individual enzymes contained in these preparations have not been described at any appreciable length: the majority of published data on these subjects have been derived from studies of enzyme preparations based on the strain T. reesei QM9414. In this work, we sought to isolate and study hemicellulases from the preparations Celloviridin G20x and Xybeten-Xyl, which hydrolyze glucuronoxylan and arabinoxylan: XYL I, XYL II, XYL III, β-XYL, and αL-AF I, as well as EG I, which exhibits xylanase activity in addition to acting as a hemicellulase. MATERIALS AND METHODS Enzyme preparations. Commercial preparations Celloviridin G20x and Xybeten-Xyl (Joint Stock Company Biovet, Bulgaria) were obtained based on the strain T. longibrachiatum (T. reesei) TW-1, using the technologies of Promferment LLC (Russia).
573
574
MARKOV et al.
Substrates. Carboxymethylcellulose sodium salt (CMC) of medium viscosity, birch glucuronoxylan, laminarin (from the alga Laminaria digitata), polygalacturonic acid (PGA) potassium salt, p-nitrophenyl-β-Dlactopyranoside (PNP-β-D-D-Lac), p-nitrophenyl-β-Dglucopyranoside (PNP-β-D-Gluc), p-nitrophenyl-β-Dcellobioside (PNP-β-D-Cell), p-nitrophenyl-α-D-arabinofuranoside (PNP-α-D-AF), and p-nitrophenyl-βD-xylopyranoside (PNP-β-D-Xyl) were obtained from Sigma (United States). Barley β-glucan, tamarind xyloglucan, and wheat arabinoxylan were supplied by Megazyme (Australia). Microcrystalline cellulose (avicel) was from Serva (Germany). Isolation of individual enzymes. Desalting and low-pressure anion exchange chromatography were performed using a Bio-Rad Econo System (United States). Chromatofocusing and preparative high-resolution gel permeation chromatography were performed on a Pharmacia Biotech FPLC system (Sweden). A 1.2 × 29 (or 0.7 × 16) cm Biogel P-4 column (Bio-Rad) was used for desalting. Depending on the conditions of the separation, the column was equilibrated with a 0.25 M imidazole–HCl buffer (pH 7.0 or 7.5) or a 0.1 M sodium-acetate buffer (pH 5.0). The proteins were eluted with the starting buffers. Low-pressure anion exchange chromatography was performed on a 1.2 × 29 cm DEAE-Spheron column (Bio-Rad) equilibrated with a 0.25 M imidazole–HCl buffer (pH 7.0). Proteins bound under starting conditions were eluted using a gradient of NaCl concentration. Chromatofocusing was performed on a Mono P HR 5/20 column (gel volume, 4 ml) equilibrated with a 0.25 M imidazole–HCl buffer (pH 7.5). Proteins bound under starting conditions were eluted using a pH gradient that was created by sequential passing through the column of Pharmacia Biotech polybuffers PB 74 with pH 4.0 and 3.0. The flow rate was 0.7 ml/min. High-resolution gel permeation chromatography was performed on a Superose 12 HR 10/30 column (gel volume, 25 ml) equilibrated with 0.1 M sodium-acetate buffer (pH 5.0); the proteins were eluted using the same buffer. A Phenyl-Superose HR 5/5 column (gel volume, 1 ml) was used, equilibrated with 0.05 M sodium-acetate buffer (pH 5.5) containing 1 M (NH4)2SO4 (buffer A). Proteins bound under starting conditions were further eluted in a gradient of 0.05 M sodium-acetate buffer (buffer B). A solution of the enzyme in buffer A (2 ml; protein concentration, 4–5 g/l)) was loaded on the column. Following binding of the sample to the carrier, the column was washed in sequence with 3 ml
starting buffer and 5 ml linear gradient (0–100%) of buffer B; the flow rate was 0.5 ml/min. Thereafter, elution was carried out using 100% buffer B. Measurement of enzymatic activities. The activities of CMCase, β-glucanase, laminarinase, avicelase, xylanase, polygalacturonase, and xyloglucanase were measured by the initial rate of formation of reducing sugars (RSs). The RSs were determined using the Somogyi–Nelson technique for CMC, β-glucan, laminarin, avicel, xylan, PGA, and xyloglucan hydrolyses, respectively [27, 28]. The concentration of the polysaccharide substrate in the reaction mixture was 5 g/l. With low-molecular-weight synthetic substrates (the p-nitrophenyl-glycosides PNP-β-D-Lac, PNP-βD-Gluc, PNP-β-D-Cell, PNP-α-D-AL, and PNP-β-DXyl) the activities of the enzymes were measured by the initial rate of p-nitrophenol (PNP) formation, as described in [27, 28]. A 0.05 M solution of a substrate in 0.1 M sodium-acetate buffer (pH 5.0) was incubated with an enzyme at 40°ë for 10 min. The reaction was stopped by adding 1 M Na2CO3. The amount of PNP formed in the solution was measured spectrophotometrically at 400 nm, using the value of the molar extinction coefficient (ε400 = 18300 M–1 cm–1). Enzyme activities were expressed in international units (IUs). One such unit corresponds to the formation of 1 µmol product per minute that an enzyme acts on the appropriate substrate. In comparing the enzymes with each other, specific activities were used. The content of protein in the samples was measured using the Lowry method. Characterization of individual enzymes. Analytical isoelectric focusing (IEF) and SDS-PAGE of individual proteins were performed on Bio-Rad instruments, Model III Cell and Mini Protean II, respectively, according to the manufacturers’ manuals. The proteins in the gels were stained with Coomassie Brilliant Blue R-250 (Ferak, Germany). Prior to electrophoresis, the enzyme solutions under study were pretreated with 1% SDS and 5% β-mercaptoethanol at 100°ë for 5– 10 min. Protein mixtures from Sigma were used as standards during IEF and SDS-PAGE. MALDI-TOF mass spectrometry of tryptic hydrolysates of the proteins. The assignment of the majority of the enzymes isolated was validated by analysis of their tryptic hydrolysates using MALDI-TOF mass spectrometry. The mass spectra were recorded on a Bruker Daltonics Reflex III system (Germany). The data obtained were processed using Mascot software (www.matrixscience.com) using NCBI and Swiss-Prot databases. Experiments involving mass spectrometry were performed at the Department of proteomic studies
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 42
No. 6
2006
PROPERTIES OF HEMICELLULASES OF THE ENZYME COMPLEX
of the Research Institute of Biomedical Chemistry, Russian Academy of Medical Sciences (Moscow). Analyses of products of xylan hydrolysis. Studies of products of xylan hydrolysis were performed using a Bio-Rad Chromatography Workstation 700 (United States), equipped with a Knauer refractometer (Germany). Molecular weight distributions (MWDs) of arabinoxylan and glucuronoxylan hydrolysis products were determined by gel permeation chromatography on a Bio-Gel TSK 30 XL column; 0.1 M sodium-acetate buffer (pH 5.0) served as an eluant. The composition of xylan hydrolysis products was determined by HPLC on a column packed with grafted amino phase. A mixture of acetonitrile with water (7 : 3) was used as an eluant (to improve the resolution of xylose, arabinose, and xylobiose, the ratio was changed to 8 : 2). Measurement of pH and temperature optima of the enzymes. Profiles of pH dependence of the enzyme activity (with glucuronoxylan as a substrate) were obtained at 50°ë using 0.1 M citrate–phosphate buffer (pH range, 3.0–7.5). In the case of XYL I, which required lower pH values, a mixture of HCl and KCl was used. The values of pH indicated on the graphs was measured in the reaction mixture. Profiles of the temperature dependence of the enzyme activity were studied within the range of 20°– 90°ë using 0.1 M citrate–phosphate buffer (pH 5.0). The results of measurements were expressed as percentages of the maximum activity value (corresponding to a pH or a temperature optimum). Studies of pH- and thermostability of the enzymes. The solutions of the enzymes in 0.1 M sodium-acetate buffer (pH 5.0) or 0.1 M citrate–phosphate buffer (pH 7.0) were incubated at a given temperature. Aliquots of the solutions were taken at intervals from the incubation mixtures and used as samples for measuring the activities of the enzymes (in each case, an appropriate substrate was used, as described above). The results are presented as dependences of the residual activity (expressed as a percentage of the maximum) on the incubation time (at a given temperature and a given pH). Studies of the kinetics of enzyme-catalyzed xylan hydrolysis. The enzymes were incubated with the substrates (arabinoxylan or glucuronoxylan) at 50°C and pH 5.0 (0.1 M sodium-acetate buffer) until the kinetic curve reached its plateau. The concentrations of the enzymes were adjusted in such a way so as to ensure the attainment of 1–2% hydrolysis depth in 10 min. In studying hydrolysis catalyzed by combinations of the enzymes, the components were added at a ration of 1 : 1 APPLIED BIOCHEMISTRY AND MICROBIOLOGY
575
(protein content). Aliquots were taken from the reaction mixture in the course of the hydrolytic reaction, and the concentrations of RSs were measured by the Somogyi– Nelson technique [27, 28]. RESULTS AND DISCUSSION Isolation of individual enzymes. EG I and XYL I were isolated in two steps. The preparations Celloviridin G20x and Xybeten-Xyl were desalted on a Biogel P-4 column equilibrated with 0.025 M imidazole–HCl buffer (pH 7.5). Thereafter, chromatofocusing was performed (Fig. 1) on a Mono Q column equilibrated with 0.25 M imidazole–HCl buffer (pH 7.5) at pH 7.5 to 3.0). The fractions obtained were analyzed by the method described in [26], which allowed us to identify (among other enzymes constituting Celloviridin G20x) fractions containing EG I and XYL I, which were pure by SDS-PAGE and IEF (Figs. 2 and 3). The contents of the enzymes in the preparation were assessed (Table 1). Purified β-XYL was isolated from the preparation Xybeten-Xyl; the contents in the preparations of this an other enzymes were also assessed (Figs. 2 and 3). The procedure for obtaining XYL II, XYL III, and α-L-AF comprised 5 steps. During the first step, the preparations were desalted on a Biogel P-4 column equilibrated with 0.25 M imidazole–HCl buffer (pH 7.0). At the second step, the desalted preparation was subjected to low-pressure anion-exchange chromatography on a DEAE-Spheron column equilibrated with 0.25 M imidazole–HCl buffer (pH 7.0). The fraction that was not bound to the carrier was collected. To prevent the enzyme inactivation, pH of the fraction was adjusted to 5.0 (by adding 1 M sodium-acetate buffer) immediately after its collection. As a result of the third step (hydrophobic chromatography), fractions were obtained, which contained XYL II and α-L-AF, as well as an insignificant amount of XYL III. Prior to further separation, these fractions were desalted on a Biogel P-4 column equilibrated with 0.1 M sodium-acetate buffer (pH 5.0) and subjected to gel filtration on a Superose 12 (the column was equilibrated with the same buffer). Gel filtration produced electrophoretically homogeneous fractions of XYL II, XYL III, and α-L-AF (Figs. 2 and 3). Substrate specificity and properties of the enzymes isolated. To classify the enzymes isolated, their molecular characteristics were studied (molecular weight, pI, substrate specificity, and mechanism of depolymerization of polymeric substrates). The substrate specificity was judged by comparing specific activities of the enzymes obtained with various substrates (Table 1).
Vol. 42
No. 6
2006
576
MARKOV et al. A280 1.0
(‡)
0.8 0.6 EG I
0.4
XYL I α-L-AF I
0.2
XYL II, XYL III
β-XYL
0 (b)
1.0 0.8
XYL II XYL III α-L-AF I
0.6 EG I β-XYL
0.4 XYL I
0.2 0
20
40
60
80
100
120 min
Fig. 1. Profiles of chromatofocusing of the preparations (a) Celloviridin G20x and (b) Xybeten-Xyl.
M
1
M
2
M
66
66
36
45 36
45 36
29
29
29
66
3
M 66
45
4
M 66
14
14
20
6
66
45
36 45 36
20
M
45
29 20
5
20
14
36 29
29 20
14
20
Fig. 2. SDS-Page of the enzymes (1) XYL I, (2) XYL II, (3) XYL III, (4) β-XYL, (5) α-L-AF I, and (6) EG I; M, marker proteins varying in molecular weight (kDa).
The enzyme with a molecular weight of 20 kDa and a pI of 5.5 exhibited a high specific activity with glucuronoxylan and arabinoxylan, but failed to hydrolyze CMC, β-glucan, or synthetic low-molecular-weight substrates (PNP-α-L-AF and PNP-β-D-Xyl) (Table 1).
When compared to the published data [3–5], the results obtained made it possible to identify the enzyme with XYL of T. reesei. Enzymes with molecular weights of 21 kDa (pI 9.5) and 30 kDa (pI 9.1) had similar substrate specificity. Such properties are characteristic of
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 42
No. 6
2006
PROPERTIES OF HEMICELLULASES OF THE ENZYME COMPLEX M
1
3
2
9.3
8.8
M
4
5
M
8.3
9.3
7.7
6.6
8.2
6.5
7.2
6.0
577
6
5.1 4.6
6.6 5.9 5.1
5.1
4.6
4.6 3.5
Fig. 3. Isoelectric focusing of the enzymes (1) XYL I, (2) XYL II, (3) XYL III, (4) β-XYL, (5) α-L-AF I, and (6) EG I; M, marker proteins varying in the value of pI.
the two known xylanases of T. reesei, XYL II and XYL III [3–6]. The enzyme with a molecular weight of 57 kDa and a pI of 4.6 exhibited a high specific activity with CMC, β-glucan, and xyloglucan (Table 1); it also hydrolyzed glucuronoxylan and arabinoxylan, as well as the low-molecular-weight synthetic substrates (PNPβ-D-Lac and PNP-β-D-Cell). Based on these properties, the enzyme was identified with EG I of T. reesei [11, 12]. Of note, EG I is the only known cellulase of T. reesei, which exhibits a high xylanase activity. The enzyme with a molecular weight of 80 kDa and a pI of 4.5 exhibited insignificant activity with glucuronoxylan and arabinoxylan (Table 1); it hydrolyzed synthetic substrates (PNP-α-L-AF and PNP-β-D-Xyl), but not CMC and β-glucan. When compared to the published data [7, 8], the results obtained made it possible to identify the enzyme with β-XYL of T. reesei. The enzyme with a molecular weight of 55 kDa and a pI of 7.4 exhibited high activity with PNP-α-L-AF; hydrolyses of glucuronoxylan and arabinoxylan were weak; no activity was detected with β-glucan, CMC, and PNP-βD-Xyl (Table 1). Taken together, the properties of the enzyme corresponded to α-L-AF of T. reesei [7, 8]. The assignment of the enzymes as XYL III, β-XYL, α-L-AF I, and EG I was further validated by analyzing tryptic hydrolysates of the enzyme proteins by MALDI-TOF mass spectrometry. For each of the above enzymes, 11–17 peptides were found, which had the same molecular weights as the theoretical peptides of exact T. reesei enzymes. For example, experimental values of molecular weights of 17 peptides of α-L-AF from the preparation Celloviridin G20x matched the weights of theoretical tryptic peptides of α-L-AF I (Abf1) of T. reesei (family 54 of glycosyl hydrolases; APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Swiss-Prot database number Q92455). In other words, the enzyme isolated was identical to one of the two known arabinofuranosidases of T. reesei, viz., α-L-AF I. It should be noted that a protein is identified if no less than five specific tryptic peptides corresponding to its amino acid sequence are detected by mass spectrometry [29]. To determine the mechanism of action of the enzymes, we used three types of data: (1) MWD of substrates in the course of hydrolysis; (2) the composition of low-molecular-weight end products of xylan hydrolysis, and (3) the yield of RSs resulting from deep xylan hydrolysis. Processes of glucuronoxylan and arabinoxylan hydrolysis catalyzed by XYL I, XYL II, XYL III, and EG I had pronounced features of endodepolymerization. The mean molecular weight of the polymers decreased considerably at early stages of the hydrolysis (depth, 1–2%). The effects of β-XYL and α-L-AF I on the xylans were not associated with the changes of average molecular weights of the polymers, up to a hydrolysis depth on the order of 5–10%, which was indicative of the exodepolymerase type of process. These features of the mechanism of action of the enzymes are in accord with the published data on T. reesei hemicellulases [3–12]. The chromatograms shown in Fig. 4 exemplify changes in the MWD of glucuronoxylan observed during its hydrolysis by XYL II or β-XYL (which follow endo- and exodepolymerase mechanisms, respectively). In the case of XYL I and EG I, low-molecularweight products of glucuronoxylan hydrolysis were represented largely by xylose and xylobiose; in the case of XYL II and XYL III, xylose was predominant.
Vol. 42
No. 6
2006
578
MARKOV et al.
Table 1. Properties of homogeneous enzymes and enzyme preparations of T. longibrachiatum (T. reesei) Parameter MW, pI Depolymerization type
XYL I
XYL II
XYL III
EG I
β-XYL
α-L-AF I
21 kDa, pI 5.5
21 kDa, pI 9.5
30 kDa, pI 9.1
57 kDa, pI 4.6
80 kDa, pI 4.5
55 kDa, pI 7.4
endo
endo
endo
endo
exo
exo
Celloviridin G20x
XybetenXyl
endo
endo
Content in the preparation, % Celloviridin G20x
0.3–0.5
<0.05
<0.05
7–8
<0.05
0.5–1
Xybeten-Xyl
0.3–0.5
15–20
2–5
7–8
1–2
<0.05
pH optimum, (pH50%)
2.5 (1.6–4.5)
5.0 (4.0–7.0)
5.0–5.5 (4.0–7.5)
5.0–5.5 (4.0–6.7)
6.5 (2.5–7.0)
3.5–4.0 (<2.5–5.3)
6.0 (3.0–6.7)
5.0 (4.0–7.0)
Temperature optimum, (T50%)
50 (<25–67)
65 (47–73)
55–60 (40–72)
65 (40–68)
65 (52–73)
60 (45–73)
60 (45–70)
60 (45–70)
Stability at 50°C, pH 5.0 3 h, 96% 3 h, 100% 3 h, 100% 3 h, 100% 3 h, 100% Stability at 50°C, pH 7.0 20 min, 0% 3 h, 70%
3 h, 35%
3 h, 97%
3 h, 100% 3 h, 100%
3 h, 26% 10 min, 0% 10 min, 0% 3 h, 30%
3 h, 84%
Specific activity, U/mg* CMC
0
0
0
23.8
0
0
7.4
7.3
β-glucan
0
0
0
43.7
0
0
8
7.8
Avicel
0
0
0
0.22
0
0
0.47
0.42
Glucuronoxylan
7.8
191
75
7.9
4.6
1.2
0.9
32.8
Arabinoxylan
8.3
170
65
8.3
0.55
2.6
1.8
31
Xyloglucan
0
0
0
45.1
0
0
4.9
4
Laminarin
0
0
0
0
0
0
0.43
0.8
PGA
0
0
0
0
0
0
0.63
0.15
PNP-β-D-Cell
0
0
0
0.34
0
0
ND
ND
PNP-β-D-Lac
0
0
0
0.33
0
0
ND
ND
PNP-β-D-AF
0
0
0
0
0.64
14
0.14
0.03
PNP-β-D-Xyl
0
0
0
0
1.6
0
0.006
0.03
PNP-β-D-Gluc
0
0
0
0
0
0
0.14
0.39
* Values of specific activity are expressed in U/mg protein (individual enzymes) or U/mg preparation; ND, activity not determined.
Hydrolysis of arabinoxylan by β-XYL resulted in the formation of xylose as the major low-molecular-weight product, whereas α-L-AF I produced arabinose. Thus, the preparations under study contained six enzymes hydrolyzing xylans, of which four catalyzed endodepolymerase hydrolysis (XYL I, XYL II, XYL III, and EG I) and two (β-XYL and α-L-AF I), exodepolymerase hydrolysis. Specific activities of the enzymes isolated varied considerably: XYL I and EG I exhibited near-identical activity with both xylan types
(~ 10 U/mg protein), whereas XYL II was almost 20 times more active (again, with both substrates). The contribution of individual enzymes to the total xylanase activity was determined from the quantitative content and specific activity of each species in the preparations. Figure 5 shows the values of absolute xylanase activity for each enzyme (i.e., of the product of multiplying together the enzyme’s specific activity and the content of its protein in the preparation). The comparison of the values led us to conclude that xylanase activity of Celloviridin G20x (Fig. 5a) is largely due to
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 42
No. 6
2006
PROPERTIES OF HEMICELLULASES OF THE ENZYME COMPLEX mV 4
mV 3
(‡)
579
(b)
4 1 3
3
2
2
1
3
2
4
2 1
1
0 10
12
14
16
18
0 10
20 min
12
14
16
18
20 min
Fig. 4. Changes in the MWD of glucuronoxylan, caused by isolated (a) XYL II and (b) β-XYL: (1) original glucuronoxylan sample; (2) hydrolysis depth, 1%; (3) hydrolysis depth, 5%; (4) hydrolysis depth, 11%.
E 70
E 4000
(‡)
(b)
60 3000
50 40
2000
30 20
1000
10 0
1
2
3
4
5
0
6
1
2
3
4
5
6
Fig. 5. Contribution of individual enzymes to the total xylanase activity of the preparations (a) Celloviridin G20x and (b) XybetenXyl: (1) XYL I, (2) XYL II, (3) XYL III, (4) EG I, (5) β-XYL, and (6) α-L-AF I.
EG I, whereas that of Xybeten-Xyl (Fig. 5b) is accounted for by XYL II. Note that the preparations Celloviridin G20x and Xybeten-Xyl were obtained from the same strain TW-1 under different conditions of fermentation. Temperature and pH optima of the enzymes and their stability. The stability of an enzyme is a major characteristic determining its use in various biotechnological processes. Figure 6 shows pH dependences of the activity of the enzymes obtained in the course of hydrolysis of specific substrates (glucuronoxylan, in the case of XYL I, XYL II, XYL III, β-XYL, and EG I; APPLIED BIOCHEMISTRY AND MICROBIOLOGY
PNP-α-L-AF, in the case of α-L-AF I). The values of optimum pH (pHopt) determined for individual enzymes varied considerably. The most acidic enzymes were XYL I (pHopt 2.5) and α-L-AF I (pHopt 3.5–4.0); β-XYL exhibited optimum activity at the highest pH value (pHopt 6.5), being the most alkaline species out of the enzymes studied. Another important characteristic, in addition to pHopt, was the range of pH within which the enzyme retained 50% of its maximum activity (pH50%). The latter parameter is frequently used in practice, because it makes possible to compare enzymes with distinct types of pH dependences (e.g., flat-slope and
Vol. 42
No. 6
2006
580
MARKOV et al.
% 100 80
5
3
1 2
60
4
7
6
8
40 20 0
1
2
3
4
5
6
7
8
91 pH
2
3
4
5
6
7
8
91 pH
2
3
4
5
6
7
8
9 pH
Fig. 6. Dependences on pH of the xylanase activity of individual enzymes and the preparations Celloviridin G20x and XybetenXyl: (1) XYL I, (2) XYL II, (3) XYL III, (4) EG I, (5) β-XYL, (6) α-L-AF I, (7) Celloviridin G20x and (8) Xybeten-Xyl.
peak-shaped). XYL II, XYL III, and β-XYL exhibited maximum activity at neutral pH (the range of ç50%is shifted to neutral values; see Table 1); XYL I and α-LAF I were most active at weakly acidic pH. Table 2. Depth of exhaustive glucuronoxylan or arabinoxylan hydrolysis by purified enzymes and the preparations Celloviridin G20x and Xybeten-Xyl (50°C, pH 5.0) Depth of hydrolysis, % Preparation, enzyme glucuronoxylan
arabinoxylan
Celloviridin G20x
40
43
Xybeten-Xyl
42
30
XYL I
37
28
XYL I + α-L-AF I
26
40
XYL I + β-XYL
46
37
XYL II
38
28
XYL II + α-L-AF I
38
48
XYL II + β-XYL
72
43
XYL III
42
30
XYL III + α-L-AF I
42
56
XYL III + β-XYL
72
42
EG I
42
26
EG I + α-L-AF I
37
51
EG I + β-XYL
49
36
β-XYL
52
19
α-L-AF I
12
25
α-L-AF I + β-XYL
49
73
Figure 7 shows the effects of temperature on the activity of the hemicellulases. Temperature optima of the enzymes did not differ considerably, ranging from 60 to 65°ë; XYL I, which exhibited optimum activity at 50°ë, was an exceptional case. The range of í50% values (i.e., temperatures, at which the enzyme retains more than 50% of its maximum activity) was shifted to the region below the temperature optima (<25°C; Table 1). The dependences of xylanase activity of the preparation Celloviridin G20x on pH and temperature (Figs. 6 and 7) were very similar to those of EG I, which is natural given the fact that this enzyme is the major contributor to the xylanase activity of the preparation. The presence in the pH dependence curve of a shoulder in the region of low values is accounted for by the effect of XYL I, which has the most acidic optimum among the enzymes studied. Temperature and pH dependences of xylanase activity of Xybeten-Xyl are similar to those of XYL II (Figs. 6 and 7), which is the major contributor to the enzymatic activity of the preparation. All individual enzymes isolated were highly stable at pH 5.0 and 50°ë, retaining more than 90% of the original activity under these conditions after 3 h of incubation. If pH was increased to 7.0, appreciable inactivation of the enzymes was observed (Fig. 8). For example, inactivation of XYL I or β-XYL was complete in 20 min. Maximum stability at pH 7.0 and 50°ë was observed with XYL II, which retained 68% of the original activity throughout 3 h of incubation. The activity of xylanase of the preparations Celloviridin G20x and Xybeten-Xyl exhibited little, if any, changes during 3 h of incubation. At pH 7.0 and 50°ë, the prep-
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 42
No. 6
2006
PROPERTIES OF HEMICELLULASES OF THE ENZYME COMPLEX
aration Celloviridin G20x retained 30% of the original xylanase activity after 3 h of incubation (Fig. 8). This observation is accounted for by the properties of EG I, the major contributor to the total xylanase activity of the preparation. The preparation Xybeten-Xyl was more stable at pH 7.0 and 50°ë (Fig. 8), retaining more than 80% of the original xylanase activity after 3 h of incubation due to the high stability of XYL II under these conditions.
% 100 60
20
2
0 100
5
60 40 30
6
20 4 0 100
Addition of α-L-AF I to XYL II or XYL III did not affect the depth of exhaustive hydrolysis of glucuronoxylan, whereas its addition to XYL I, EG I, or β-XYL decreased this parameter (Table 2). Thus, concerted action of XYL I, EG I, or β-XYL, on the one hand, and α-L-AF I, on the other hand, produced antagonistic effects. This observation is accounted for by the decrease in the reactivity of glucuronoxylan caused by cleavage of arabinose side groups of the substrate under the effect of α-L-AF I.
7
60
8
40 30 20
The depth of arabinoxylan hydrolysis by individual endodepolymerases increased considerably on addition of α-L-AF I or β-XYL (Table 2), the effect being more pronounced in the case of α-L-AF I. Thus, concerted action of an endoxylanase (XYL I, XYL II, XYL III, or EG I), on the one hand, and α-L-AF I or β-XYL, on the other hand, produced synergistic effects.
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
3
30
The depth of exhaustive hydrolysis of glucuronoxylan and arabinoxylan by endodepolymerases XYL I, XYL II, XYL III, and EG I amounted to 37–41% and 26–30%, respectively. In the case of glucuronoxylan, however, the difference between the enzymes was more pronounced: the values obtained with β-XYL and α-LAF I equaled 52 and 12%, respectively.
The difference between the preparations turned out to be more significant with arabinoxylan as the substrate. The depth of hydrolysis of arabinoxylan by the preparations Celloviridin G20x and Xybeten-Xyl was
1
40
Exhaustive hydrolysis of xylans by the enzymes. Xylanases are frequently used in situations where a deep conversion of the substrate is called for. Therefore, we studied the kinetics of hydrolysis of glucurono- and arabinoxylans. In these experiments, we used both individual enzymes and combinations thereof (at 1 : 1 ratios; protein content). The data obtained were compared to those available for the original enzyme preparations Celloviridin G20x and Xybeten-Xyl (Table 2).
The depth of glucuronoxylan hydrolysis by the preparations Celloviridin G20x and Xybeten-Xyl equaled 40 and 42%, respectively. Thus, the preparations largely mimicked the effects of EG I and XYL II, the enzymes determining their major xylanase activity.
581
0 20
30
40
50
60
70
80
90 T, °C
Fig. 7. Temperature dependences of the xylanase activity of individual enzymes and the preparations Celloviridin G20x and Xybeten-Xyl: (1) XYL I, (2) XYL II, (3) XYL III, (4) EG I, (5) β-XYL, (6) α-L-AF I, (7) Celloviridin G20x and (8) Xybeten-Xyl.
equal to 40 and 30%, respectively. The higher conversion degree of arabinoxylan by Celloviridin G20x than individual endoxylanases may be accounted for by their synergy with α-L-AF I. The content of α-L-AF I in the preparation Xybeten-Xyl is considerably lower than in Celloviridin G20x; correspondingly, the depth of arabi-
Vol. 42
No. 6
2006
582
MARKOV et al.
% 100 8 80
2
60 3
40 20
6
1 4
0
7
30
60
90
120 150 180 0 min
5 30
60
90
120 150 180 0 min
30
60
90
120 150 180 min
Fig. 8. Stability of individual enzymes and the preparations Celloviridin G20x and Xybeten-Xyl at pH 7.0 and 50°C: (1) XYL I, (2) XYL II, (3) XYL III, (4) EG I, (5) β-XYL, (6) α-L-AF I, (7) Celloviridin G20x and (8) Xybeten-Xyl.
noxylan hydrolysis is lower. Synergistic effects may also underlie the higher activity of the preparation Celloviridin G20x, observed with arabinoxylan (as compared to glucuronoxylan; Table 1). In the case of the preparation Xybeten-Xyl, the activities measured with either substrate are nearly identical to each other. In conclusion, the preparations Celloviridin G20x and Xybeten-Xyl are composed of six enzymes involved in biodegradation of natural xylans. Of these, four enzymes act as endodepolymerases (XYL I, XYL II, XYL III, and EG I), and two (β-XYL and α-L-AF I), as exodepolymerases. Distinct fermentation conditions used in obtaining the preparations produce differences in (1) the total xylanase activity (which is considerably higher in the case of Xybeten-Xyl) and (2) the content of individual hemicellulases (EG I and XYL II are major contributors to the activity of the preparations Celloviridin G20x and Xybeten-Xyl, respectively). Depending on the type of the polymeric substrates (xylans) subjected to exhaustive hydrolysis, concerted action of endoxylanases of the preparations and β-XYL or α-L-AF I may result in synergistic (arabinoxylan) or antagonistic (glucuronoxylan) effects. These properties, as well as specific features of temperature and pH dependences of the enzyme activities pave the way for their use in a variety of biotechnological processes. ACKNOWLEDGMENTS This work was supported in part by the Federal Target Science and Technology Program Studies and developments in Priority Fields of Science and Technology (project no. ZhS-KP.4/001).
REFERENCES 1. Goodwin, T.W. and Merser, E.I., Introduction to Plant Biochemistry, Oxford: Pergamon, 1983. 2. Coughlan, M.P. and Hazlewood, G.P., Hemicellulose and Hemicellulases. Portland Press Research Monograph, London and Chapel Hill, 1993. 3. Torronen, A., Mach, R.L., Messner, R., Gonzalez, R., Kalkkinen, N., Harkki, A., and Kubicek, C.P., Biotechnol., 1992, vol. 10, pp. 1461–1465. 4. Biely, P., Vranska, M., Kremnicky, L., Tenkanen, M., Poutanen, K., and Hayn, M., Proc. II TRICEL Symposium on Trichoderma reesei Cellulases and other Hydrolases. Espoo. (Finland), 1993, pp. 125–135. 5. Mach, R.L., Butterweck, A., Schindler, M., Messner, R., Herzog, P., and Kubicek, C., Proc. II TRICEL Symposium on Trichoderma reesei Cellulases and other Hydrolases. Espoo. (Finland), 1993, pp. 211–216. 6. Xu, J., Takakuwa, N., Nogawa, M., Okada, H., and Morikawa, Y., Appl. Biochem. Biotechnol., 1998, vol. 49, pp. 718–724. 7. Herrmann, M.C., Vrsanska, M., Jurikova, M., Hirrsch, J., Biely, P., and Kubicek, C.P., Biochem. J., 1997, vol. 321, pp. 375–381. 8. Poutanen, K. and Puls, J., Appl. Microbiol. Biotechnol., 1988, vol. 28, pp. 425–432. 9. Poutanen, K., J. Biotechnol., 1998, vol. 7, pp. 271–282. 10. Nogawa, M., Yatsui, K., Tomioka, A., Okada, H., and Morikawa, Y., Appl. Environ. Microbiol., 1999, vol. 65, pp. 3964–3968. 11. Bhikhabhai, R., Johansson, G., and Petterson, G., J. Appl. Biochem., 1984, vol. 6, pp. 336–345. 12. Biely, P., Vrsansksa, M., and Claeyssens, M., Eur. J. Biochem., 1991, vol. 200, pp. 157–163. 13. Uhlig, H., Industrial Enzymes and Their Applications, New York: Willey, 1998. 14. Buchert, J., Tenkanen, M., Kantelinen, A., and Viikari, L., Bioresourse Technol., 1994, vol. 50, pp. 65– 72.
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
Vol. 42
No. 6
2006
PROPERTIES OF HEMICELLULASES OF THE ENZYME COMPLEX 15. Bhat, M.K., Biotechnol. Adv., 2000, vol. 18, pp. 355– 383. 16. Markov, A.V., Grishutin, S.G., Salanowich, T.N., Kondratyeva, E.G., and Sinitsyn, A.P., Abstracts of Int. Conf. “Biocatalysis-2002: Fundamentals & Applications”, Moscow, 2002, pp. 112–113. 17. Chesson, A., Animal Feed Sci. Technol., 1993, vol. 45, pp. 65–79. 18. Okolelova, T., Rumyantsev, S., Morozov, A., and Kuznetsova, T., Zhivotnovodstvo Rossii, 2001, no. 8, pp. 38– 41. 19. Komarova, Z. and Khrishchataya, E., Kombikorma, 2000, no. 1, p. 4. 20. Vasil’eva, N.Ya., Tsurikova, N.V., Shirokova, T.Yu., Keleinikova, E.V., and Sinitsyn, A.P., Khran. Pererab. Sel’khozsyr’ya, 2001, no. 4, pp. 46–47. 21. Burachevskii, I.I., Proizv. Spirta Likerovod. Izd., 2001, no. 1, p. 11. 22. Kislukhina, O.V., Fermenty v proizvodstve pishchi i kormov (Enzymes in Production of Food and Fodders), Moscow: DeLi print, 2002.
APPLIED BIOCHEMISTRY AND MICROBIOLOGY
583
23. Koryachkina, S.Ya., Kuznetsova, E.A., and Gulyaeva, E.V., Khran. Pererab. Sel’khozsyr’ya, 2003, no. 1, pp. 43–45. 24. Annison, G. and Choct, M., World’s Poultry Sci. J., 1991, vol. 47, pp. 164–172. 25. Girhammar, U. and Nair, B.M., Food Hydrocolloids, 1992, vol. 6, pp. 329–343. 26. Markov, A.V., Gusakov, A.V., Kondrat’eva, E.G., Okunev, O.N., Bekkarevich, A.O., and Sinitsyn, A.P., Biokhimiya, 2005, no. 6, pp. 796–804. 27. Sinitsyn, A.P., Gusakov, A.V., and Chernoglazov, V.M., Biokonversiya lignotsellyuloznykh materialov (Bioconversion of Lignocellulose Materials), Moscow: Mosk. Gos. Univ., 1995. 28. Schejter, A. and Marcus, L., Methods Enzymol., 1988, vol. 161, pp. 366–373. 29. Rappsilber, J., Moniatte, M., Nielsen, M.L., Podtelejnikov, A.V., and Mann, M., Int. J. Mass Spectrom., 2003, vol. 226, pp. 223–237.
Vol. 42
No. 6
2006
