Depolymerization of β-chitin to mono- and disaccharides by the serum fraction from the para rubber tree, Hevea brasiliensis

Article abstract of DOI:10.1016/j.carres.2004.09.011

The serum fraction of latex from Hevea brasiliensis, the para rubber tree, is known to contain an endo-chitinolytic enzyme, hevamine. Herein the activity of the rubber serum towards β-chitin is investigated. The serum contained 6 mg/mL of protein and a chitinolytic activity of 18 mU per mg of protein. The optimum ratio of enzyme to chitin was 0.22 mU/mg, and the optimum substrate concentration was 60 mg/mL. The optimum pH range was pH 2-4, and the optimum temperature was 45°C. At these conditions both (GlcNAc)₂ and GlcNAc were produced in a molar ratio of approximately 2:1. The hydrolysis of 300 mg of chitin with 64 mU of the rubber serum for 8 days under the optimum conditions gave 39 mg of GlcNAc and 108 mg of (GlcNAc)₂ as determined by HPLC. Mixing the rubber serum preparation with an Aspergillus niger pectinase preparation containing β-N-acetylhexosaminidase can be used to produce almost exclusively the GlcNAc monomer in about 50% yield.

Full text of DOI:10.1016/j.carres.2004.09.011

Carbohydrate
RESEARCH
Carbohydrate Research 339 (2004) 2799–2804
Depolymerization of b-chitin to mono- and disaccharides by
the serum fraction from the para rubber tree, Hevea brasiliensis
Akamol Klaikherd,a M. L. Siripastr Jayanta,a Jariya Boonjawat,b
Sei-ichi Aibac and Mongkol Sukwattanasinitta,*
aDepartment of Chemistry, Faculty of Science, Chulalongkorn University, Phyathai Road, Pathumwan, Bangkok 10330, Thailand
bDepartment of Biochemistry, Faculty of Science, Chulalongkorn University, Phyathai Road, Pathumwan, Bangkok 10330, Thailand
cGreen Biotechnology Research Group, The Special Division for Human Life Technology, National Institute of Advanced
Industrial Science and Technology, Osaka 563-8577, Japan
Received 13 July 2004; accepted 21 September 2004
Available online 18 October 2004
Abstract—The serum fraction of latex from Hevea brasiliensis, the para rubber tree, is known to contain an endo-chitinolytic
enzyme, hevamine. Herein the activity of the rubber serum towards b-chitin is investigated. The serum contained 6mg/mL of protein
and a chitinolytic activity of 18mU permg of protein. The optimum ratio of enzyme to chitin was 0.22mU/mg, and the optimum
substrate concentration was 60mg/mL. The optimum pH range was pH2–4, and the optimum temperature was 45ꢀC. At these con-
ditions both (GlcNAc)2 and GlcNAc were produced in a molar ratio of approximately 2:1. The hydrolysis of 300mg of chitin with
64mU of the rubber serum for 8days under the optimum conditions gave 39mg of GlcNAc and 108mg of (GlcNAc)2 as determined
by HPLC. Mixing the rubber serum preparation with an Aspergillus niger pectinase preparation containing b-N-acetylhexosamini-
dase can be used to produce almost exclusively the GlcNAc monomer in about 50% yield.
ꢁ 2004 Elsevier Ltd. All rights reserved.
Keywords: Hevea brasiliensis; Chitin; Hevamine; D-Glucosamine
1. Introduction
building blocks for biologically important glycolipids
and glycoproteins.10
Chitin [a polymer of (1!4)-2-acetamido-2-deoxy-b-D-
glucose] is a polysaccharide abundant in invertebrate
exoskeletons including crustacean shells and fungal cell
walls.1,2 A gigantic amount of chitin is processed as solid
waste from seafood processing industry.2,3 The natural
degradation of chitin presents an important feature,
not only in the global recycling of carbon and nitrogen
sources, but also in the production of useful chemical
reagents.4 N-Acetyl-D-glucosamine (GlcNAc) and
glucosamine (GlcN), the monomer of chitin and chito-
san, are therapeutic agents for osteoarthritis,5 inflamma-
tory bowel diseases6 and gastritis.7 Chitinous oligomers
also have been reported as having antitumor, antifungal
and antimicrobial activities,8,9 as well as being chemical
In nature, chitin-degrading enzymes are widely dis-
tributed among bacteria, arthropods, protozoans and
plants.11,12 Nevertheless, commercial preparations of
chitinases are inevitably expensive, especially, the puri-
fied chitinases. Hence, the use of crude enzymes has
become an interesting alternative in the preparation of
GlcNAc and chito-oligomers.13–15
Hevea brasiliensis, the para rubber tree, is known to
produce an endochitinase called hevamine, which be-
longs to glycoside hydrolase family no. 18. The enzyme
was isolated from the lutoid body of the rubber latex.16
A few reports have described chitinolytic activities and
structures of hevamine.17–19 In the present work, we
illustrate here that the rubber serum fraction from H.
brasiliensis contained interestingly different chitinolytic
activities from hevamine in certain aspects. The rubber
serum can potentially be an excellent source of low-cost
*
Corresponding author. Tel.: +66 2 218 7620; fax: +66 2 218 7598;
e-mail: smongkol@chula.ac.th
0008-6215/$ - see front matter ꢁ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.09.011

2800
A. Klaikherd et al. / Carbohydrate Research 339 (2004) 2799–2804
chitinolytic enzymes, as it is regarded as a massive waste
product that must be disposed of from rubber-process-
ing plants.
35
30
25
20
15
10
2. Results and discussion
The hydrolysis of b-chitin with enzyme in the rubber
serum gave two major chitinolysis products, (GlcNAc)2
and GlcNAc. The production yield of (GlcNAc)2 and
GlcNAc increased proportionally with the increment
of enzyme:chitin ratio from 0.11 to 1.08mU/mg
(Fig. 1). The enzyme:chitin ratio of 0.22mU/mg was
thus selected for further condition optimization as
this enzyme:chitin ratio also produced appropriate
concentration of the products to be effectively analyzed
by HPLC throughout the course of the hydrolysis
studied.
5
0
0
10
20
30
40
50
60
70
80
Concentration of chitin (mg/mL)
Figure 2. The effect of concentration of chitin on the chitinolysis of
b-chitin with rubber serum. (r) GlcNAc, (j) (GlcNAc)2, enzyme:
chitin = 0.22mU/mg, pH4.0 (0.05M NaOAc), 37ꢀC, 8days.
The most convenient means to increase the product
yields and concentration is to raise the concentration
of both chitin substrate and the enzyme while maintain-
ing the enzyme:chitin ratio. In the next experiment, the
concentration of both chitin substrate and the enzyme
were increased while the enzyme:chitin ratio was kept
constant at 0.22mU/mg. As expected, the yield of both
saccharides went up with the increasing concentration
of chitin (Fig. 2) up to 60mg/mL where the stirring be-
came ineffective due to the high content of solid chitin.
A pH-dependence study of chitinolytic activity of the
enzyme in the rubber serum revealed that the chitino-
lytic enzyme in the serum was acidophilic with an opti-
mum pH range of pH2–4 (Fig. 3). The results found
here indicated that the enzyme complexes in the serum
respond to pH rather differently from hevamine, the re-
ported major chitinase found in the rubber serum. Hev-
amine was reported to show the optimum pH in the
range of pH4.0–6.0.17 This difference may be attributed
40
35
30
25
20
15
10
5
0
1.5
2.0
2.5
3.0
3.5
pH
4.0
4.5
5.0
5.5
6.0
Figure 3. The effect of pH on the chitinolysis of b-chitin with rubber
serum. pH2–4.5 (0.05M citrate–phosphate buffer): (r) GlcNAc, (j)
(GlcNAc)2, pH4.0–5.5 (0.05M NaOAc): (m) GlcNAc, (·) (GlcNAc)2,
[chitin] = 60mg/mL, enzyme:chitin = 0.22mU/mg, 37ꢀC, 8days.
to some chitin-binding proteins and other cooperative
enzymes present in the serum.20,21
30
25
20
15
10
5
The optimum temperature for chitinases has been
reported to vary with the sources and generally lies
between 30 and 50ꢀC. The optimum temperature for
hevamine was reported to be 45ꢀC.17 In this particular
study, the enzyme in the rubber serum displayed the
highest chitinolytic activity at 45ꢀC (Fig. 4), which is
the same as that for hevamine. At 37ꢀC and at 55ꢀC
the activity dropped from that at the optimum tempera-
ture by approximately 25–30%.
Subroto et al. also reported that the chitinolytic activ-
ity of hevamine depended on the concentration of salts
in the reaction media.17 In this study, we found that
the chitinolytic activity of the enzyme in the rubber ser-
um depended on the concentration of buffer up to 0.1M
NaOAc (Fig. 5). At higher concentration of the buffer,
the activity was relatively constant. The buffer concen-
0
0
0.5
1
1.5
2
2.5
Enzyme/chitin ratio (mU/mg)
Figure 1. The effect of the enzyme:chitin ratio on the chitinolysis of b-
chitin with rubber serum. (r) GlcNAc, (j) (GlcNAc)2, [chitin] =
20mg/mL, pH4.0 (0.05M NaOAc), 37ꢀC; 8days.

A. Klaikherd et al. / Carbohydrate Research 339 (2004) 2799–2804
2801
60
50
40
30
20
10
60
50
40
30
20
10
0
0
0
2
4
6
8
10
12
14
16
18
25
35
45
55
Temperature (oC)
Time (day)
Figure 6. The time course of the hydrolytic products from chitinolysis
of b-chitin with rubber serum. (r) GlcNAc, (j) (GlcNAc)2,
[chitin] = 60mg/mL enzyme:chitin ratio = 0.22mU/mg, pH4.0 (0.1M
NaOAc), 45ꢀC.
Figure 4. The effect of temperature on the chitinolysis of b-chitin with
rubber serum. (r) GlcNAc, (j) (GlcNAc)2, [chitin] = 60mg/mL,
enzyme:chitin ratio = 0.22mU/mg, pH4.0 (0.1M NaOAc), 8days.
tration of 0.1M was thus the minimum concentration
required for maintaining the effective activity level of
the enzyme throughout the course of the hydrolysis.
After 8days of incubation, the HPLC yields of Glc-
NAc and (GlcNAc)2 from the hydrolysis of b-chitin
(300mg) with the serum at the optimum conditions were
38 and 109mg, respectively, representing 12% and 36%
of the starting chitin. The total conversion of chitin to
GlcNAc and (GlcNAc)2 was thus around 48%, and
the reaction was virtually terminated after 8days (Fig.
6). We first speculated that the enzymatic hydrolysis
reactions stopped at only 50% conversion rather than
going to completion due to either enzyme denaturation
or product inhibition. Various enzyme:chitin ratios
0.22, 0.44, 0.66 and 0.88mU/mg were used in the hydroly-
sis. The production of both GlcNAc and (GlcNAc)2
during the first four days tended to increase with the
enzyme:chitin ratio, but the final production yields
after 8days were virtually the same at the total yield
around 50% (Fig. 7). Furthermore, addition of more
rubber serum after 8days gave no additional yields.
These results indicated that enzyme denaturation was
unlikely to be the major cause for limiting the final
yields; otherwise the final yields should increase with
the increase of the serum.
GlcNAc product inhibition was investigated by carry-
ing out the reaction in the presence of GlcNAc inten-
tionally added in various concentrations at the
beginning of the hydrolysis. The results showed that
high initial concentration of GlcNAc did not inhibit
the reaction (Table 1). We also found evidence, which
will be described later, that (GlcNac)2 was not the cause
for the hydrolysis to be halted.
Since neither denaturation of enzyme nor the product
inhibition could explain the self-termination of the
hydrolysis, we speculated that the hydrolysis of chitin
stopped halfway to completion because the chitin start-
ing material remained in the reaction in an unsuitable
form for further degradation by the hevamine in the
serum. This endo-type chitinase, known to be the major
chitinolytic enzyme in para rubber trees, has a (ꢀ4, ꢀ3,
ꢀ2, ꢀ1, +1, +2) binding cleft that hydrolyzes chitooligo-
saccharides containing a GlcNAc unit longer than
(GlcNAc)4.18,19 The remaining unhydrolyzable chitin
particles may contain GlcNAc units densely packed in
a crystalline nucleus surrounded by a loose polymeric
shell enriched with GlcN units that prevent the attack
by hevamine (Fig. 8).
60
50
40
30
20
10
0
0
0.1
0.2
0.3
0.4
0.5
From the data in all previous experiments, the major
products from the enzymatic hydrolysis of chitin were
(GlcNAc)2 and GlcNAc with (GlcNAc)2:GlcNAc molar
ratio of approximately 2:1. It is known that GlcNAc can
be produced from the hydrolysis of (GlcNAc)2 by b-N-
acetylhexosaminidase. The b-N-acetylhexosaminidase
Concentration of buffer (M)
Figure 5. The effect of concentration of buffer on the chitinolysis
of b-chitin with rubber serum. (r) GlcNAc, (j) (GlcNAc)2, [chitin] =
60mg/mL, enzyme:chitin ratio = 0.22mU/mg, pH4.0 (NaOAc), 45ꢀC,
8days.

2802
A. Klaikherd et al. / Carbohydrate Research 339 (2004) 2799–2804
100
80
60
40
20
0
2
4
6
8
Time (day)
2
4
6
8
2 4 6 8
2
4 6 8
0.22
0.66
0.44
0.88
Enzyme/chitin ratio (mU/mg)
Figure 7. The yields of GlcNAc and (GlcNAc)2 in the chitinolysis of b-chitin with rubber serum at various enzyme:chitin ratios. (j) GlcNAc, (
(GlcNAc)2, [chitin] = 60mg/mL, pH4.0 (0.1M NaOAc), 45ꢀC, enzyme:chitin ratio = 0.22, 0.44, 0.66, 0.88mU/mg.
)
Table 1. The effect of GlcNAc concentration in the product inhibition
study
count for the significant amount of GlcNAc observed
and the absence of higher oligosaccharides in the
hydrolysis of chitin with the serum, the serum must con-
tain at least one more chitinolytic enzyme besides hev-
amine. We would like to propose here that the major
oligosaccharide produced by the hydrolysis of chitin
with hevamine was likely to be (GlcNAc)5, and another
enzyme presented in the serum was chitobiosidase, an
exochitinase that usually hydrolyzes chitin and oligosac-
charides from the non-reducing end to release (Glc-
NAc)2. The products obtained from the hydrolysis of
(GlcNAc)5 with chitobiosidase would be (GlcNAc)2
and (GlcNAc) with a (GlcNAc)2:(GlcNAc) molar
ratio of 2:1, which corresponds to the experimental
observations.
Initial concentration of GlcNAc (mM)
Yield of products
(mM)
GlcNAc
(GlcNAc)2
0.00
9.31
32.4
31.5
29.9
32.5
33.2
31.0
45.0
44.4
43.5
46.4
47.3
45.2
18.17
28.21
36.31
47.19
[chitin] = 60mg/mL, enzyme:chitin ratio = 0.22mU/mg, pH4.0 (0.1M
NaOAc), 45ꢀC, 6days.
Although the rubber serum exhibited reasonably high
chitinolytic activity, it produced two major hydrolytic
products, GlcNAc and (GlcNAc)2. We were interested
in producing GlcNAc from a one-step hydrolysis of chit-
in with the rubber serum. To achieve this goal, the
serum must contain both chitinase (EC 3.2.1.14) and
b-N-acetylhexosaminidase (EC 3.2.1.52) activities. The
chitinase is responsible for hydrolyzing polymeric chitin
chains into smaller chitooligosaccharides, which are in
turn further hydrolyzed by b-N-acetylhexosaminidase
to GlcNAc. Unfortunately, the serum contained no b-
N-acetylhexosaminidase activity for this purpose. Com-
mercial pectinase from Aspergillus niger (An) was
reported to contain high b-N-acetylhexosaminidase
activity.13 It should thus be interesting to mix the rubber
serum with pectinase An to produce a low-cost enzyme
complex for production of GlcNAc from the direct
hydrolysis of chitin.
Figure 8. The proposed structure of the unhydrolyzable chitin particles
)
in the chitinolysis of b-chitin with rubber serum. (d) GlcNAc and (
GlcN.
activity in the rubber serum was thus evaluated by incu-
bating (GlcNAc)2 with the serum. No GlcNAc from the
mixture of (GlcNAc)2 and the serum was detected even
after 24h of incubation. These results clearly indicated
the absence of b-N-acetylhexosaminidase activity in
the serum. These results also agreed well with a rela-
tively constant (GlcNAc)2:(GlcNAc) molar ratio of ꢁ2
throughout the hydrolysis time course. However, to ac-
In order to find a suitable mixing ratio between pec-
tinase An and the rubber serum, the pectinase An (b-
N-acetylhexosaminidase activity = 21.35mU/mg) and
the rubber serum (chitinase activity = 108mU/mL) were
mixed at various activity ratios in the chitinolysis of b-
chitin. When the activity ratio of pectinase An to the

A. Klaikherd et al. / Carbohydrate Research 339 (2004) 2799–2804
2803
120
100
80
of chitin around 50%. This limited conversion should
be a result of hevamine present as a main enzyme in
the serum capable of hydrolyzing only the amorphous
part of chitin with at least four connecting GlcNAc
units. Mixing the serum with an enzyme preparation
containing b-N-acetylhexosaminidase is used for con-
verting (GlcNAc)2 to GlcNAc in situ. The final yield
of GlcNAc was nevertheless 50%. Further investigation
in mixing the serum with another low-cost enzyme con-
taining different types of endochitinase to improve the
conversion is currently under investigation.
60
40
20
0
0
1
2
3
4
Pectinase An / serum rubber ratio (mU/mU)
3. Experimental
3.1. Materials
Figure 9. The yields of GlcNAc and (GlcNAc)2 obtained from the
chitinolysis of b-chitin with the mixed pectinase An/rubber serum
at various activity mixing ratios. (r) GlcNAc; (j) (GlcNAc)2,
[chitin] = 60mg/mL, 0.1M NaOAc pH4.0, 45ꢀC, 2days.
b-Chitin from squid pen was purchased from Ta-Ming
Enterprises (Samutsakorn, Thailand) and was ground
in an ultracentrifugal mill (Recter 970) to give puffy
fibrous chitin about 50lm in diameter and 100lm in
length. Rubber serum was obtained from Pan-Asia Bio-
technology Co., Ltd. (Thailand) and analyzed for its
protein content by settling protein with the ISO/FDIS
procedure22 following LowryÕs method.23 The analysis
showed that the serum contained 6.0mg protein per
1mL of the serum solution. The rubber serum possessed
chitinolytic activity of 18mU/mg protein determined
according to the modified SchaleÕs method.24
rubber serum was higher than 1.5, (GlcNAc)2 was quan-
titatively converted to GlcNAc (Fig. 9).
It is also noteworthy to point out here that the hydroly-
sis of b-chitin by the mixed enzyme preparation also
stopped at 50% yield of GlcNAc (Fig. 10), which is in
line with the results observed in the hydrolysis solely
with the serum described previously. These results sug-
gested that (GlcNAc)2 product inhibition was not a
cause for ending the hydrolysis at this low level of
conversion.
In conclusion, we have demonstrated herein that the
serum fraction from the latex of para rubber tree (H.
brasiliensis) possesses interesting chitinolytic activities
that included both endochitinase, hevamine, and exochi-
tinase, chitobiosidase activities. The enzyme complex in
the serum cooperatively hydrolyzes chitin to produce
(GlcNAc)2 and GlcNAc with 2:1 molar ratio. The
hydrolysis, however, gave a relatively low conversion
3.2. General procedures
A typical enzymatic hydrolysis was carried out by incu-
bating a mixture of the rubber serum and the chitin sub-
strate in a buffer solution in the presence of NaN3
(500ppm) as a preservative. At each time point, a por-
tion of the reaction mixture was sampled out, diluted
with water, then mixed with CH3CN at the ratio of
30:70 (v/v), filtered, and analyzed by HPLC (controller:
Water controller 600, column: Shodex Asahipak NH2P-
50, flow rate: 1mL/min, mobile phase: 30:70 CH3CN–
water, detector: Water 991 photodiode array at
210nm). The amounts of GlcNAc and (GlcNAc)2 in
the reaction mixture were determined from the calibra-
tion lines of the corresponding compounds. The percent-
age yields were calculated based on the number of moles
of GlcNAc or (GlcNAc)2 produced against the number
of moles of the corresponding repeating units in the
starting chitin.
The unit (U) of endo-type chitinolytic activity
was determined based on the number of reducing end
groups of the sugars produced from the hydrolysis of
colloidal chitin at pH4.0 and 37ꢀC using a modified
SchaleÕs method (1U = the amount of enzyme capable
for producing 1lmol of reducing end of sugar per
minute).22
50
40
30
% yield
20
10
0
8
0.03
6
0.16
0.33
4
Time (day)
1.65
2
3.29
ratio pectinase An/serum
rubber (mU/mU)
Figure 10. The conversion of chitin to GlcNAc and (GlcNAc)2
obtained from the chitinolysis of b-chitin with mixed pectinase
An and rubber serum at various activity mixing ratios. [chitin] =
60mg/mL; 0.1M NaOAc pH4.0; 45ꢀC; 2–8days.

2804
A. Klaikherd et al. / Carbohydrate Research 339 (2004) 2799–2804
7. Gindzienski, A.; Zwierz, K.; Rozanski, A. Biochem. Med.
1971, 5, 188–194.
The activity of b-N-acetylhexosaminidase (chitobiase)
was determined by HPLC using (GlcNAc)2 as a sub-
strate. One unit of b-N-acetylhexosaminidase activity
was defined as the amount of an enzyme able to hydro-
lyze 1lmol of (GlcNAc)2 to 2lmol of GlcNAc per
minute.
8. Shahidi, F.; Kamil, J.; Arachchi, V.; Jeon, Y. J. Trends
Food Sci. Technol. 1999, 10, 37–51.
9. Suzuki, S. Fragrance J. 1996, 15, 61–68.
10. Takahashi, S.; Terayama, H.; Koshino, H.; Kuzuhara, H.
Tetrahedron 1999, 55, 14871–14884.
11. Stintzi, A.; Heitz, T.; Prasad, V.; Wiedermamn, S. M.;
Kauffmann, S.; Geoffroy, P.; Legrand, M.; Fritig, B.
Biochimie 1993, 75, 687–706.
12. Kramer, K. J.; Koga, D. Insect Biochem. 1986, 16, 851–
877.
Acknowledgements
13. Sashiwa, H.; Fujishima, S.; Yamano, N.; Kawasaki, N.;
Nakayama, A.; Muraki, E.; Hiraga, K.; Oda, K.; Aiba, S.
Carbohydr. Res. 2002, 337, 761–763.
14. Sukwattanasinitt, M.; Zhu, H.; Shashiwa, H.; Aiba, S.
Carbohydr. Res. 2002, 337, 133–137.
15. Pichyangkura, R.; Kudan, S.; Kuttiyawong, K.; Sukwat-
tanasinitt, M.; Aiba, S. Carbohydr. Res. 2002, 337, 557–
559.
16. Terwisscha van Scheltinga, A. C.; Kalk, K. H.; Beintema,
J. J.; Dijkstra, B. W. Structure 1994, 2, 1181–1189.
17. Subroto, T.; Beintema, J. J.; Schreuder, H. A.; Soed-
janaatmadja, U. M. S.; Van Koningsveld, G. A. Phyto-
chemistry 1996, 43, 29–37.
This work was financially supported by Thailand–Japan
Technology Transfer Project (TJTTP) under the Over-
sea Economy Cooperation Fund (OECF), Thailand
Toray Science Foundation (TTF) and National Metal
and Materials Technology Center (Code MT-B45-
POL-09-206-G). We would also like to gratefully thank
to Metallurgy and Material Science Research, Chul-
alongkorn University for permission to use an ultracen-
trifugal mill, and Pan-Asia biotechnology Co., Ltd, for
providing the rubber serum used in this work.
18. Bokma, E.; Barends, T.; Terwisscha van Scheltinga, A. C.;
Dijkstra, B. W.; Beintema, J. J. FEBS Lett. 2000, 478,
119–122.
References
19. Terwisscha van Scheltinga, A. C.; Armand, S.; Kalk, K.
H.; Isogai, A.; Henrissat, B.; Dijkstra, B. W. Biochemistry
1995, 34, 15619–15623.
1. Knorr, D. Food Technol. 1984, 38, 85–97.
`
2. Jolles, K.; Muzzarelli, R. A. A. Chitin and Chitinases;
Birkhauser: Basel, 1999.
20. Subroto, T.; Vries, H.; Jan Schuringa, J.; Soedjanaat-
madja, U. M. S.; Hofsteenged, J.; Jekelc, P. A.; Beintemac,
J. J. Plant Physiol. Biochem. 2001, 39, 1047–1055.
21. Martin, M. N. Plant Physiol. 1990, 95, 469.
22. Beezhold, D.; Swanson, M.; Zehr, B. D.; Kostyal, D.
Annals Allergy Asthma Immunol. 1996, 76, 520–526.
23. Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randal,
R. J. J. Biol. Chem. 1951, 193, 265–275.
3. Tracey, M. V. Rev. Pure Appl. Chem. 1957, 7, 1–14.
4. Gooday, G. W. In Advances in Microbial Ecology.
Marshall, K. C., Ed.; Plenum: New York, 1990; pp 387–
430.
5. Talent, J. M.; Gracy, R. W. Clin. Ther. 1996, 18, 1184–
1190.
6. Salvatore, S.; Heuschkel, R.; Tomlin, S.; Davies, S. E.;
Edwards, S.; Walker-Smith, J. A.; French, I.; Murch,
S. H. Aliment. Pharmacol. Ther. 2000, 14, 1567–1579.
24. Imoto, T.; Yagishita, K. Agric. Biol. Chem. 1971, 35,
1154–1156.