Metabolism of kalopanaxsaponin K by human intestinal bacteria and antirheumatoid arthritis activity of their metabolites.

Article abstract of DOI:10.1248/bpb.25.68

When kalopanaxsaponin K (KPK) from Kalopanax pictus was incubated for 24 h at 37 degrees C with human intestinal microflora, KPK was mainly metabolized to kalopanaxsaponin I (KPI) via kalopanaxsaponin H (KPH) rather than via kalopanaxsaponin J (KPJ), and then transformed to kalopanaxsaponin A (KPA) and hederagenin. Bacteroides sp., and Bifidobacterium sp. and Fusobacterium sp. transformed KPK to KPI and KPA and hederagenin via KPH or KPJ. However, Lactobacillus sp. and Streptococcus sp. transformed KPK to KPI, KPA, and hederagenin only via KPJ. The metabolite KPA of KPK showed potent antirheumatoid arthritis activity.

Full text of DOI:10.1248/bpb.25.68

68
Biol. Pharm. Bull. 25(1) 68—71 (2002)
Vol. 25, No. 1
Metabolism of Kalopanaxsaponin K by Human Intestinal Bacteria and
Antirheumatoid Arthritis Activity of Their Metabolites
d
Dong-Hyun KIM, ,a Eun-Ah BAE,a Myung Joo HAN,b Hee-Juhn PARK,c and Jong-Won CHOI
*
College of Pharmacy,a and Department of Food and Nutrition,b Kyung-Hee University, 1, Hoegi, Dongdaemun-ku, Seoul
130–701, Korea, Department of Botanical Resources, Sangji University,c Wonju 220–702, Korea, and College of
Pharmacy, Kyung-Sung University,d Pusan 608–736, Korea. Received July 9, 2001; accepted September 17, 2001
When kalopanaxsaponin K (KPK) from Kalopanax pictus was incubated for 24 h at 37 °C with human in-
testinal microflora, KPK was mainly metabolized to kalopanaxsaponin I (KPI) via kalopanaxsaponin H (KPH)
rather than via kalopanaxsaponin J (KPJ), and then transformed to kalopanaxsaponin A (KPA) and hedera-
genin. Bacteroides sp., and Bifidobacterium sp. and Fusobacterium sp. transformed KPK to KPI and KPA and
hederagenin via KPH or KPJ. However, Lactobacillus sp. and Streptococcus sp. transformed KPK to KPI, KPA,
and hederagenin only via KPJ. The metabolite KPA of KPK showed potent antirheumatoid arthritis activity.
Key words kalopanaxsaponin K; antirheumatoid arthritis activity; intestinal bacteria; Kalopanax pictus
The stem bark of Kalopanax pictus belonging to the fam- won province, Korea, and the plant was identified by profes-
ily Araliaceae has been used in Korea as a tonic, analgesic, sor S. Y. Yun (Division of Applied Plant Sciences, Sangji
antiinflammatory and antidiabetic medicine. From this plant, University, Wonju, Korea). A voucher specimen was de-
hederagenin glycosides, syringin, liriodendrin, and conifer- posited in the herbarium of Life Science and Natural Re-
ylaldehyde glucosides have been isolated.1—3) Related to sources, Sangji University, Wonju, Korea.
biological activities of K. pictus, it was reported that the
Dried stem bark (4.8 kg) of K. pictus was cut and extracted
active components on antidiabetic,4) cytotoxic,5) and antifun- three times with MeOH under reflux. The extract was filtered
gal assays6) may be hederagenin monodesmosides. We also and evaporated under reduced pressure to give a viscous
reported that kalopanaxsaponin B (KPB) and H (KPH) mass (630 g) of MeOH extract. This extract was suspended in
were easily metabolized by human intestinal bacteria to water and fractionated with ethylacetate and n-butanol.5,7)
kalopanaxsaponin A (KPA) and kalopanaxsaponin I (KPI), Physicochemical data of isolated KPK and kalopanaxsaponin
resepectively.7) However, these metabolites were slowly trans- J (KPJ) were recorded and the values were in good agree-
formed to hederagenin. KPI was slowly transformed to hed- ment with those of authentic samples.
eragenin via KPA. Among these compounds, the most antidi-
KPK (1.2 g): Amorphous powder, mp 217—219 °C (dec.),
abetic compound was KPA. However, studies related to the 1H-NMR and 13C-NMR: literature (ref. 3).
metabolic pathway of KPI to hederagenin in the intestine, the
KPJ (0.8 g): Amorphous powder, mp 217—219 °C (dec.),
biotransfromation of kalopanaxsaponin K (KPK) to KPA, 1H-NMR and 13C-NMR: literature (ref. 5).
which is an antitumor compound,5) and their biological activ-
ities, such as antirheumatoid arthritis, are incomplete.
Extraction and Isolation of Metabolites To obtain the
metabolites of KPK by human intestinal bacteria, the reac-
Therefore we isolated the main components KPB, KPH, tion mixture contained 0.4 g KPK (or KPJ) and 2 g fresh
and KPK from the stem bark of K. pictus, investigated the stool specimen of human (male, twenty) in a final volume of
metabolic pathway of these kalopanaxsaponins by human in- 500 ml of an anaerobic dilution medium.8) The mixture was
testinal bacteria, and measured the antirheumatoid arthritis incubated at 37 °C for 24 h. The reaction mixture was ex-
activity of the main metabolites. We also screened the bio- tracted three times with ethylacetate. EtOAc soluble portion
transforming bacteria of KPK to the bioactive compound of KPK-reacted mixture was evaporated under reduced pres-
KPA from human intestinal bacteria.
sure and the metabolites were isolated by silica gel column
chromatography with the solvent of CHCl3/MeOH/H2O
(7 : 3 : 1, lower phase) according to the previous method.7)
Physicochemical data of KPJ (3 mg), KPI (35 mg), KPA
MATERIALS AND METHODS
Instrument Melting points were determined on an elec- (3 mg), HA (hederagenin 3-O-a-L-arabinopyranoside, 0.2
trothermal digital melting point apparatus. 1H- and 13C-NMR mg) and hederagenin (1.2 mg) were taken and their values
spectra were recorded on a Brucker-AM 500 with tetra- were in good agreement with authentic specimens, respec-
methylsilane (TMS) as an internal standard. The TLC chro- tively.
1
matogram of metabolites was quantitatively analyzed with a
Shimadzu CS-920 TLC-scanner.
Hederagenin: Amorphous powder, mp 318—320 °C. H-
and 13C-NMR: literature (ref. 7).
Materials Pontamine and histamine were purchased
HA: Colorless needles from MeOH, mp 228—230 °C
from Sigma Chemical Co. (U.S.A.), Freund’s complete adju- (dec.). 1H-NMR: literature (ref. 7).
vant (FCA) reagent was from Difco Co. (U.S.A.) and general
KPA: Colorless needles from MeOH, mp 265—268 °C
anaerobic medium (GAM) was from Nissui Pharmaceutical (dec.). 1H- and 13C-NMR: literature (ref. 7).
Co., Ltd. (Japan).
KPI: Colorless needles from MeOH, mp 218—220 °C
Extraction and Isolation of KPK and KPJ from K. pic- (dec.). 1H- and 13C-NMR: literature (ref. 7).
tus K. pictus NAKAI was collected in August 1998 in Kang-
KPJ: Amorphous powder, mp 217—219 °C (dec.), 1H- and
To whom correspondence should be addressed. e-mail: dhkim@khu.ac.kr
© 2002 Pharmaceutical Society of Japan

January 2002
69
13C-NMR: literature (ref. 5).
was measured at 570 nm.
Time Courses of Metabolism of KPK and KPJ by
Human Intestinal Bacteria The assay mixture contained
4 ml of KPK (or KPJ) 0.6 mM and 0.1 g fresh fecal suspen-
sion (human intestinal bacteria fraction) in a final volume of
20 ml of an anaerobic dilution medium.8) The mixture was
incubated at 37 °C for 24 h and an aliquot (1 ml) of the reac-
tion mixture was periodically extracted twice with 5 ml of
ethylacetate. The ethylacetate extract was analyzed by TLC.
TLC TLC for KPK, KPH, KPB, KPJ, KPI, KPA, HA,
and hederagenin were performed on silica gel plates (Merck,
Silica gel 60F254) with CHCl3/MeOH/water (65 : 35 : 10, lower
layer). The chromatograms of these compounds were quanti-
tatively analyzed with a TLC scanner after spraying with 5%
H2SO4 solution, followed by heating in a dry oven at 110 °C
for 15 min (detection ranges of KPK, KPJ, and their metabo-
lites, 0.5—10 mg).
Screening of Intestinal Bacterial Strains Metabolizing
KPK According to our previous method,7) a suspension of
the fresh feces of a healthy male volunteer was diluted 105 to
108-fold with GAM broth. An aliquot (200 ml) of the 107-di-
luted human feces was inoculated on a GAM agar plate,
which was anaerobically incubated at 37 °C for 4 d. Colonies
were isolated and each colony was incubated in 5 ml of GAM
broth containing 0.4% KPK for 1 d. The cultured medium
was extracted with 5 ml of ethylacetate. After evaporating the
ethylacetate fraction, it was analyzed for KPJ, KPI, KPA, and
hederagenin by TLC. Twenty bacterial strains containing pre-
viously isolated bacteria were isolated.9) Identification of the
isolated bacteria was performed according to Bergey’s man-
ual.10)
Animals Male Sprague–Dawley (SD) rats (200—220 g)
and male ICR mice (20—25 g) were supplied from Daehan
Experimental Animal Breeding Center. All rats and mice
were housed in wire cages, fed with standard laboratory
chow (Samyang Feed Production Co.) and allowed water ad
libitum.
Anti-inflammatory activity was measured according to the
method of Saito et al.11) For the induction of rheumatoid
arthritis, 0.05 ml of FCA reagent was injected to right foot-
pads of the rats. The induction of rheumatoid arthritis in rats
was confirmed 2 weeks after FCA treatment. KPA and KPI
(5, 10, 20 mg/kg/d) were dissolved in saline and intraperi-
toneally administered daily for 3, 5, 7, and 10 d. The effect
was taken by plethysmometer (Ugo Basile, Italy). The in-
hibitory effect was calculated as follows:
RESULTS AND DISCUSSION
Metabolites of KPK by Human Intestinal Microflora
To investigate the metabolites of KPK broken down by
human intestinal bacteria, KPK was anaerobically incubated
for 24 h with a bacterial mixture from human feces. Then the
metabolites were extracted with ethylacetate and analyzed by
TLC. Five metabolites, three main metabolites and two
1
minor ones, were observed by TLC. By analysis of H- and
13C-NMR spectra, the three main metabolites were identified
as KPI, KPA and hederagenin, and the two minor ones were
identified as KPJ and HA. When KPJ was anaerobically in-
cubated for 24 h with a bacterial mixture from human feces,
its main metabolites were KPI, KPA, and hederagenin.
When KPK was anaerobically incubated for 24 h with 20
bacterial strains isolated from human feces, Bacteroides sp.,
and Bifidobacterium sp. and Fusobacterium sp. transformed
KPK to KPI and/or KPA and hederagenin via KPH or KPJ
(Table 1). However, Lactobacillus sp. and Streptococcus sp.
transformed KPK to KPI, KPA, and hederagenin only via
KPJ.
Time Course of Metabolism of KPK and KPJ by
Human Intestinal Microflora The time course of metabo-
lism of KPK by human intestinal microflora is shown in
Fig. 1. When KPK was incubated with human intestinal mi-
croflora, KPH, KPJ, and KPI were produced 4 h after incuba-
tion. The main metabolites were KPH and KPI (KPH was
dramatically decreased and KPI was increased 9 h after the
start of incubation). KPA, KPJ, and hederagenin were also
observed from 9 to 24 h after incubation, but these com-
pounds were minor metabolites. After 15 h of incubation,
KPI was found to be a main metabolite while KPA, HA, and
hederagenin were detected as the minor metabolites. KPI and
hederagenin were the main metabolites 24 h after incubation.
When KPJ was incubated with human intestinal mi-
croflora, KPI was observed after 4 h of incubation. However,
the transforming activity of KPJ to KPI was weak. Therefore
the transformation of KPJ to hederagenin was also slow dur-
ing 24 h of incubation. The main metabolic pathway of KPK
by human intestinal microflora proceeds to KPI via KPH
rather than KPJ.
inhibition of edema (%)ϭ
volume of control groupϪvolume of treated group
)ϭ
ϫ100
volume of control group
The vascular permeability test was performed according to
Whittle’s method.12) Pontamine sky blue saline solution 4%
(0.1 ml/10 g mouse body weight) was injected into the tail
vein 30 min after the intraperitoneal administration of KPA
or KPI. Histamine 0.5% (0.1 ml/10 g mouse body weight)
was injected into the abdominal cavity 15 min after injection
of pontamine sky blue saline solution. The mice were killed
20 min after the final injection, and then the pigment exu-
dated into abdominal cavity was washed with 10 ml of dis-
tilled water. This washed solution was centrifuged (3000
rpm, 10 min) and the absorbance of the resulting supernatant
Fig. 1. Time Course of Metabolism of KPK by Human Intestinal Mi-
croflora
Human fecal suspensions were prepared and their metabolites were assayed accord-
ing to Materials and Methods. Symbols indicate the following: ᭛, KPK; ᭜, KPJ; ᭺,
KPH; ᭹, KPI; ᮀ, KPB; ᭿, KPA; ᭡, hederagenin 3-O-a-L-arabinopyranside (HA); ϫ,
hederagenin (H).

70
Vol. 25, No. 1
When KPK was incubated with Bacteroides JY-6, its meta- were the main metabolites of KPK produced by human in-
bolic time course was similar to that of human intestinal bac- testinal bacteria, were measured in a FCA reagent-induced
teria (data not shown). However, when KPK was incubated rat rheumatoid arthritis model (Table 2). After the induction
with Eubacterium A-44, KPJ was a main metabolite 4 h after of rheumatoid arthritis, KPA or KPI (5, 10, 20 mg/kg) were
incubation. KPH, KPB and KPJ were observed as metabo- intraperitoneally administered for 10 d, respectively. The
lites 9 h after incubation. Another metabolite, KPA, was ob- edema was significantly reduced 5 d after the administration
served 20 h after incubation as a main metabolite. This KPA of KPA at the two higher doses. The administration of KPI
was slowly metabolized to hederagenin.
5 mg/kg did not inhibit the edema, and the 10 mg/kg dose
Based on these findings, the time course of KPK metabo- showed weak inhibitory effects. Both KPA and KPI quickly
lism by human intestinal bacteria was as follows: in the early showed significant effects when their doses were increased.
stage, KPK began to be quickly converted to KPH and slowly The prolongation of KPA and KPI administration exhibited
to KPJ. Then these compounds were metabolized to KPI, and more potent antiedema effects.
then slowly metabolized to KPA, HA, and hederagenin
(Chart 1).
To confirm the antiedema effects, the inhibitory activity of
KPA and KPI on vascular permeability was also measured
Antiinflammatory Effect of the Main Metabolites KPI (Table 3). KPA and KPI reduced the vascular permeability,
and KPA The inhibitory effects of KPI and KPA, which and KPA was more effective than KPI.
Chart 1. Proposed Metabolic Pathway of Kalopanaxsaponin K by Human Intestinal Microflora
➞, main pathway; ➝, minor pathway.
Table 1. KPK and KPJ Transforming Activity of Intestinal Bacteria Isolated from Human Feces
Transformed producta) (mM)
Microbe
KPK
KPI
KPJ
KPH
KPB
KPJ
KPA
HA
H
KPI
KPA
HA
H
Bacteroides HJ15
Bacteroides JY6
Bifidobacterium K103
Bifidobacterium K110
Eubacterium A44
Eubacterium L8
Fusobacterium K60
Lactobacillus L2
Streptococcus S2
Streptococcus S10
0
2.13
7.70
1.31
2.02
4.15
0
0.82
0
0
1.43
Ͻ0.05
0.96
0.90
4.28
Ͻ0.05
0.83
Ͻ0.05
0.83
0.78
25.31
1.37
0.78
0
2.29
0
2.10
3.07
1.61
0.68
7.39
1.62
1.37
0.99
1.24
1.55
7.02
6.77
2.80
0
0
0
0
0
0
0.81
0.75
1.18
0
5.03
6.16
3.04
0
0.21
3.33
4.46
0
0
1.99
8.70
2.42
0.81
5.78
0
0.99
2.42
0
0.93
0
0
0
0.93
0
0.75
1.12
0.81
0
7.0
6.16
4.46
0
4.46
11.54
4.74
0
32.30
2.79
2.79
2.79
0
2.79
0
0
78.58
6.86
Ͻ0.05
Ͻ0.05
Ͻ0.05
Ͻ0.05
Ͻ0.05
11.69
<0.05
0
0
0
0
0
0
Ͻ0.05
0.99
a) To assay transformed products, the reaction mixture was incubated at 37 °C for 20 h with a 25 mg pellet of each intestinal bacterium.

January 2002
71
Table 2. Inhibitory Effect of KPA and KPI on Freund’s Complete Adjuvant Reagent-Induced Rat Edema Model
Swelling (%)
Dose
Group
(mg/kg)
0
3
5
7
10 (d)
Control
KPA
74.4Ϯ5.58a,b
78.6Ϯ4.39a
80.6Ϯ11.3b
73.7Ϯ3.44a,b
73.6Ϯ4.78a,b
75.3Ϯ5.26a,b
76.2Ϯ4.40a
74.7Ϯ3.36a,b
75.9Ϯ3.27a,b
66.5Ϯ3.57c,d,e
67.9Ϯ2.58c,f
63.0Ϯ3.07f
71.3Ϯ4.92a,b
63.3Ϯ2.82e,f
53.4Ϯ3.73d,e
53.9Ϯ3.86d,e
73.9Ϯ4.59a
62.1Ϯ2.50e,f
62.1Ϯ2.19e,f
74.2Ϯ5.49a,b
61.8Ϯ1.44e,f
50.6Ϯ2.45d,e
49.5Ϯ1.94d,e
72.4Ϯ2.80e,f
62.8Ϯ2.17e,f
60.2Ϯ2.50f
5
10
20
5
10
20
KPI
75.3Ϯ2.41a,b
71.5Ϯ2.00a,c,d
65.8Ϯ3.42d,e,f
KPA and KPI were intraperitoneally administered to rats once a day for 10 d. The assay procedure was described in the experimental methods. Values represent meansϮS.D.
(nϭ10). Values followed by the same letter are not significantly different from control (pϽ0.05).
Table 3. Inhibitory Effect of KPA and KPI on Vascular Permeability in by intestinal bacteria and then the metabolites KPI and KPA
Rats
could play an important role in the antirheumatoid activity of
these compounds. Finally, we believe that KPK, KPB and
KPH in K. pictus are natural prodrugs for the antirheumatic
compound KPA.
Dose
(mg/kg)
Permeability
(dye amount, mg)
Group
Normal
106.7Ϯ10.05a
Control
KPA
213.6Ϯ14.53b (100.0)
189.9Ϯ11.98b,c (88.9)
156.0Ϯ9.95d,e (73.0)
150.9Ϯ10.04e (70.6)
200.1Ϯ13.91b (93.7)
186.9Ϯ9.85b,c (87.5)
173.2Ϯ9.85b,c (81.1)
109.9Ϯ12.36f (51.5)
Acknowledgements This work was supported by the BK
21 grant from the Ministry of Education and Kyung Hee
University.
5
10
20
5
10
20
100
KPI
REFERENCES
Aminopyrine
1) Sano K., Sanada S., Ida Y., Shoji J., Chem. Pharm. Bull., 39, 865—
870 (1991).
2) Shao C. J., Kasai R., Xu J. D., Tanaka O., Chem. Pharm. Bull., 37,
311—314 (1989).
3) Lee M., Hahn D. R., Arch. Pharm. Res., 14, 124—128 (1991).
4) Park H. J., Kim D.-H., Choi J. W., Park J. H., Han Y. N., Arch. Pharm.
Res., 21, 24—29 (1998).
The assay procedure was described in the experimental methods. Values represent
meansϮS.D. (nϭ10). Values followed by the same letter are not significantly different
from normal (pϽ0.05). The values in the parentheses represent the percentage in-
hibitory ratio.
5) Park H. J., Kwon S. K., Lee J. H., Lee K. K., Miyamoto K., Lee K. T.,
Planta Medica, 67, 118—121 (2001).
6) Kim D. W., Bang K. H., Rhee Y. H., Lee K. T., Park H. J., Arch.
Pharm. Res., 21, 688—691 (1998).
7) Kim D.-H., Yu K. W., Bae E. A., Park H. J., Choi J. W., Biol. Pharm.
Bull., 21, 360—365 (1998).
8) Hattori M., Shu Y.-Z., Shimizu M., Hayashi T., Morita N., Kobashi K.,
Xu G.-J., Namba T., Chem. Pharm. Bull., 33, 3838—3846 (1985).
9) Bae E.-A., Park S.-Y., Kim D.-H., Biol. Pharm. Bull., 23, 1481—1485
(2000).
When the biological activities of kalopanaxsaponins were
investigated in hyperglycemia,4,7) cytotoxicity,5) mutagenic-
ity,13) and hepatotoxicity model,14) KPA was found to be the
most active compound. In this FCA-induced rat rheumatoid
arthritis model, KPA was also effective.
In Korea, the stem bark of K. pictus belonging to the fam-
ily Araliaceae has been used as a tonic, analgesic, an-
tirheumatic and antidiabetic agent. Its main components were
kalopanaxsaponins, such as KPH, KPB, and KPK. However,
it is thought that these compounds are not absorbed from the
intestine. Therefore these compounds come into contact with
intestinal bacteria after oral administration and are trans-
10) Kreig N. R., “Bergey’s Manual of Systemic Bacteriology,” Williams
and Wilkins, Baltimore, 1984, pp. 603—661 and 1418—1434.
11) Saito B., Ohashi T., Togashi M., Koyanagi T., Nippon Hinyokika
Gakkai Zasshi, 81, 993—996 (1990).
12) Whittle B. A., Brit. J. Pharmacol., 22, 246—260 (1949).
formed into KPI, KPA, and hederagenin. Particularly if Eu- 13) Lee K. T., Sohn I. C., Park H. J., Kim D. W., Jung K. O., Park K. Y.,
Planta Medica, 66, 329—332 (2000).
14) Park H. J., Lee K. T., Jung W. T., Choi J. W., Kadota S., Nat. Med., 53,
bacterium A-44 is involved in the metabolism of these
kalopanaxsaponins, the main metabolite should be KPA. This
biotransformation should be catalyzed in the human intestine
113—117 (1999).