Radical-scavenging effects of Aloe arborescens Miller on prevention of pancreatic islet B-cell destruction in rats

Journal of Ethnopharmacology 89 (2003) 37–45

Radical-scavenging effects of Aloe arborescens Miller on prevention of pancreatic islet B-cell destruction in rats Hidehiko Beppu∗ , Takaaki Koike, Kan Shimpo, Takeshi Chihara, Motoyuki Hoshino, Chikako Ida, Hiroshi Kuzuya Fujita Memorial Institute of Pharmacognosy, Fujita Health University, 1865 Isshiki-cho, Hisai, Mie 514-1296, Japan Received in revised form 5 June 2003; accepted 7 June 2003

Abstract We evaluated the possible scavenging effects of Aloe arborescens Miller var. natalensis Berger (Kidachi aloe in Japanese) on free radicals generated by streptozotocin (Sz) or alloxan (Ax). The components of Kidachi aloe were added to a reaction system in which • OH radicals derived from Sz or Ax as pancreatic islet B-cell toxins and hypoxanthine-xanthine oxidase (HX-XO)-derived O2 radicals destroy isolated islet B-cells, and we observed its preventive effects. The Kidachi aloe components inhibited the destruction of rat pancreatic islet B-cells by Sz, Ax or HX-XO. These components were prepared in the form of a freeze-dried powder of the boiled leaf skin of Kidachi aloe, and measurement of 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical-scavenging activity showed higher radical-scavenging activity in this boiled leaf skin powder than the non-boiled leaf skin powder. Furthermore, HPLC chromatograms of the “Boiled leaf skin powder” were similar to those of commercially available aloin (barbaloin content: approximately 20%). Therefore, the main component may be a phenol compound. In addition, the phenolic fraction of the Boiled leaf skin contained large amounts of 2 -O-p-coumaroylaloesin and 2 -O-feruloylaloesin, which have higher DPPH radical-scavenging activity than barbaloin. These results suggest that the action mechanism of Kidachi aloe Boiled leaf skin components, which prevent destruction of the pancreatic islets by specific pancreatic islet toxins such as Sz, Ax, and HX-XO, involves inhibition of free radical-scavenging effects, and may be associated with a thermostable low molecular component. The co-existence of Kidachi aloe-derived 2 -O-p-coumaroylaloesin, 2 -O-feruloylaloesin, and aloin may result in the potentiation of radical-scavenging activity. © 2003 Elsevier Ireland Ltd. All rights reserved. Keywords: Aloe; Anti-diabetes; Streptozotocin; Alloxan; Langerhans’ islets; Hypoxanthine-xanthine oxidase

1. Introduction Previously, several studies have reported the effects of Aloe arborescens Miller var. natalensis Berger (Kidachi aloe) and Aloe barbadensis Miller (Aloe vera Linne) on streptozotocin (Sz) or alloxan (Ax)-induced diabetes in animals (Ghannam et al., 1986; Hikino et al., 1986; Ajabnoor, 1990; Beppu et al., 1993; Chithra et al., 1998; Okyar et al., 2001). However, the action mechanism was not examined in those studies. Takasu et al. (1988) found the production of hydrogen peroxide (H2 O2 ) by the pancreatic islets. They also reported that Sz or Ax administration to cultured cells of the iso∗ Corresponding

author. Tel.: +81-59-252-1010; fax: +81-59-252-0710. E-mail address: [email protected] (H. Beppu).

lated rat pancreatic islets and the rat jugular vein resulted in increased H2 O2 production in the islets, twice that in the controls (Takasu et al., 1991; Komiya and Takasu, 1993). Okamoto (1985) reported that destruction of pancreatic islet B-cells was associated with pancreatic B-cell DNA injury related to free radicals. The pancreatic islets are relatively susceptible to active oxygen, and are more easily damaged than other tissues. This is because the activities of antioxidant enzymes such as superoxide dismutase (SOD) are low in pancreatic islet B-cells, and because these cells are easily affected by active oxygen and radicals (Okamoto, 1990; Burkart et al., 1992; Hotta et al., 1998). We performed this experiment to examine the inhibitory effects of Kidachi aloe leaf skin or leaf pulp on pancreatic B-cell injury to • OH free radicals produced by Ax and Sz as well as to clarify the features of the active constituents.

0378-8741/$ – see front matter © 2003 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/S0378-8741(03)00268-X

38

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45

2. Materials and methods 2.1. Preparation of Kidachi aloe material Fresh aloe leaves (5–6 years old) were harvested from the herb garden of Yurika Co., Ltd. (Hisai, Japan). Leaves weighing over 80 g were selected and immediately. 2.1.1. Preparation of freeze-dried whole Kidachi aloe leaf, freeze-dried leaf skin and freeze-dried leaf pulp Three kilograms of fresh whole Kidachi aloe leaves were homogenized in a polytron homogenizer and then freeze-dried (abbreviated as Whole leaf FD). Three kilograms of the fresh whole Kidachi aloe leaves were separated into the superficial layer of the leaf (leaf skin, weight approximately 1 kg) and the succulent layer of the leaf (leaf pulp weight approximately 2 kg) with a knife, and prepared by the same method used for the Whole leaf FD. The leaf skin gave a freeze-dried Kidachi aloe leaf skin powder (abbreviated as Leaf skin FD), while the leaf pulp gave a freeze-dried leaf pulp powder (abbreviated as Leaf pulp FD). 2.1.2. Preparation of acetone-precipitated Kidachi aloe leaf skin juice Three kilograms of fresh whole Kidachi aloe leaves were separated into the leaf skin and pulp with a knife, and the skin was homogenized in a polytron homogenizer and then homogenates were filtered through a Whatman GF/A filter. Approximately 1 l of aloe leaf skin juice was treated with a twofold volume of cold acetone, precipitated and lyophilized. The resulting powder was referred to as the acetone-precipitated Kidachi aloe leaf skin powder (abbreviated as Leaf skin AP). 2.1.3. Preparation of boiled Kidachi aloe leaf extract powder The leaf skin was cut into small pieces, and put in a 3 l glass beaker. To a wet weight of 2 kg, 800 ml of distilled water was added, then boiled for 40 min. The boiled extract was centrifuged at 18,000×g for 15 min, and the supernatant was freeze-dried, yielding 2.4 g of powder (abbreviated as Boiled leaf skin). The leaf pulp was similarly processed by heat extraction to yield 1.4 g of powder (abbreviated as Boiled leaf pulp). 2.2. Chemicals In this experiment, the main low molecular components of aloe, commercially available aloin (A-0451, barbaloin content: approximately 20%, Sigma Chemical Co., USA) and aloin A (B-6906, barbaloin, content: approximately 97%, Sigma Chemical Co.), were used as control reagents. Furthermore, 400 mg of aloin or 40 mg of aloin A was placed in a beaker, mixed with 8 ml of distilled water, and heated in boiled water for 40 min. Subsequently, these aloin and aloin

A solutions were freeze-dried to obtain powders (abbreviated Boiled aloin and Boiled aloin A). The aloe components name of aloe and barbaloin quantity (mg/g dry powder) in aloe are shown in Fig. 1. On the other hand, the method of Shimpo et al. (2003) was used to isolate aloenin, 2 -O-p-coumaroylaloesin, 2 -O-feruloylaloesin, isobarbaloin, and barbaloin as phenolic compounds characteristic of Kidachi aloe for the measurement of DPPH radical-scavenging activity. In addition, commercially available aloe-emodin (A-7687, content: approximately 95%, Sigma Chemical Co.) was used. 2.3. Reversed-phase HPLC To compare the components of commercially available aloin and aloin A (barbaloin) to those of leaf skin and Boiled leaf skin, high performance liquid chromatography (HPLC) was performed. The presence of phenolic compounds characteristic of Kidachi aloe was also confirmed under similar conditions. These aloe components were mixed with an equivalent volume of methanol, deproteinized, and centrifuged at 900 × g for 5 min to isolate the supernatant with Inersil ODS-2. The HPLC system for the determination of the compounds of interest consisted of two Shimadzu (Kyoto, Japan) LC-10AT pumps equipped with a Shimadzu SCL-10A system controller connected to a Shimadzu column oven, CTO-10A, and a Shimadzu diode array detector, SPD-M10AV. Samples were injected with a Model 7125 (Rheodyne, Cotati, CA) sample valve equipped with a 100-␮l loop. Peak areas were calculated with a Shimadzu workstation (CLASS-LC10/M10A). A degasser (Shimadzu DGU-4A) for mobile-phase buffers was attached to the pumps. Reversed-phase HPLC was performed according to the methods described by Kuzuya et al. (2001), as shown in Fig. 5. 2.4. Preparation of doses of aloe components for concentrations for cell culture These Kidachi aloe-derived powders and chemicals were dissolved in phosphate buffered saline (PBS, pH 7.4) at concentrations of 1–20 mg/ml, centrifuged at 7700 × g for 15 min, and filtrated with a 0.22-␮m filter. In cell culture experiments, these Kidachi aloe-derived powders and chemicals were dissolved in RPMI1640 at concentrations of 1–16 mg/ml. 2.5. Preparation of islet cells As described by Appels et al. (1989), Wistar rats weighing 250 g were sacrificed, and the duodenal side of the ductus choledochus was ligated. HBSS containing 1 mg/ml of collagenase was infused into the pancreatic duct and the pancreas was removed and heated at 37 ◦ C for 45 min to soften

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45

39

leaf pul p

Fig. 1. Diagram of the separation of the aloes material and quantity of barbaloin in aloe. (∗ ): Barbaloin content: mg/g dry powder; isolation method by Kuzuya et al. (2001).

the tissue. Pancreatic islets were collected under a stereoscopic microscope. The pancreatic islets were thoroughly washed in HBSS, and cultured at 37 ◦ C under the 5% CO2 gaseous phase for 24 h. These cultured cells were treated with EGTA-trypsin to isolate pancreatic islet cells. 2.6. Measurement of the oxygen radical-generating system: experiments of the superoxide-generating system and free radical-scavenging system With respect to the oxygen radical-generating system, 0.5 mM hypoxanthine (HX, 6-hydroxypurine, Sigma Chemical Co.) was mixed with 2×10−5 M Luminol (Sigma Chemical Co.) as an emission-enhancing agent, and then mixed with 0.025 U/ml of xanthine oxidase (XO, Sigma Chemical Co.) to generate oxygen radicals by the HX-XO system (Burkart et al., 1992). Experimentally, aloe components and SOD were added to the HX-XO system using a chemiluminescence device (BLR 201, Aloka Inc.), and lumines-

cence intensity (count/min) was measured. Reagents and aloe component concentrations were confirmed. HX was mixed with 10% FCS, penicillin, and streptomycin (Gibco BRL Inc.), and dissolved in RPMI1640 medium (Nissuisha) to prepare a final concentration of 0.5 mmol/l immediately before use. In an experiment on cell injury related to HX-XO using isolated pancreatic islet cells, 100 ml of pancreatic islet cells (3 × 105 cells/ml) were added to a 96 half-well plate with lid (CORNING, #25870, Growth area 16 mm2 ). HX was prepared at a final concentration of 0.5 mmol/l immediately before use, while XO was prepared at a final concentration of 25 mU/ml. To obtain aloe extract for each area, freeze-dried powder was dissolved in RPMI1640 medium to prepare a concentration of 0.16 g/ml. The survival rate for cells exposed to oxygen radicals was blue 4 h after exposure. In the presence of 0.25% trypan blue, viable or dead cells in 10 visual fields were counted under an inverted microscope (200×), and the survival rate was calculated.

40

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45

2.7. Measurement of DPPH radical-scavenging activity

× 1000

2.7.2. Measurement of DPPH radical-scavenging activity DPPH radical-scavenging activity was measured by the method of Yamaguchi et al. (1998). Extract solution (100 ␮l) was mixed with 400 ␮l of 100 mM Tris–HCl buffer (pH 7.4) and 500 ␮l of 500 ␮M DPPH (Wako Pure Chemical Industries, Ltd.) dissolved in ethanol, thoroughly stirred, and left at room temperature in a dark room for 20 min. After reactions, the DPPH decrease rate was measured as radical-scavenging activity (%) by HPLC using the method of Abe et al. (2000). As a positive control, Trolox (Aldrich), a water-soluble vitamin E derivative, was used. To produce a standard curve, ethanol solutions of Trolox were prepared to obtain the final concentrations of 20, 40, 60, 80, 100, 120, and 140 ␮M, and the scavenging activity of each solution was measured. The HPLC apparatus and analytic conditions were as follows: column, Handy-ODS (4.6 mm × 250 mm, Wako Pure Chemical Industries, Ltd.); detector, UV-970 Intelligent UV-Vis (JASCO Corp.); detection, 517 nm; flow-rate, 1.0 ml/min; mobile phase, 0.15% potassium phosphate solution (pH 3.5):acetonitrile, 1:3. 2.8. Statistical analysis Values are expressed as means ± S.D. or means ± S.E. Significance was tested using Student’s t-test for unpaired samples. P < 0.05 was regarded as significant.

3. Results As shown in Fig. 2, the free radical-scavenging effects of Leaf skin FD and Boiled leaf skin were examined using an HX-XO system. Boiled leaf skin had no curves suggesting radical generation, demonstrating the oxygen radical-scavenging effects of Boiled leaf skin components. Subsequently, to investigate free radical-scavenging effects, we designed an experimental system that mixed free radicals generated by HX-XO with isolated pancreatic islet cells, then examined cellular disorder, and calculated the survival rate with trypan blue. The survival rate for pancreatic islet cells was rapidly decreased 4–6 h after addition of

6 5 4 3 2 1

12 0

10 5

90

75

60

45

30

15

5

-1

10

0 0

2.7.1. Preparation of aloe components and chemicals Each powder (5 mg) of aloenin, 2 -O-p-coumaroylaloesin, 2 -O-feruloylaloesin, isobarbaloin, barbaloin, and aloeemodin was mixed with 1 ml of 50% methanol aqueous solution using a multi-tube mixer MT-30 (JASCO Corp.) for 1 h and centrifuged at 15,000 × g for 10 min, and the supernatant was obtained as extract solution. However, aloeemodin, which is slightly soluble in methanol, was dissolved in dimethyl sulfoxide. Each powder (50 mg) of Whole leaf FD, Leaf skin FD, and Boiled leaf skin was mixed with 1 ml of 50% methanol solution, and extract solution was obtained as samples by a method similar to the above.

Chemiluminescence activity (cpm)

7

Time(min) Fig. 2. Radical scavenge effects of aloe components on release of oxygen radicals from the xanthine oxidase-hypoxanthine system as detected by luminol-enhanced chemiluminescence. (䊉—䊉) HX-XO system alone, (䊊—䊊) HX-XO system + Boiled leaf skin, (䉱—䉱) HX-XO system + SOD.

HX-XO (survival rate 6 h after addition: 14.1%). However, in the control group without HX-XO addition, the survival rate was 86.2%, 6 h after HX-XO was added in the other groups. When Boiled leaf skin was added to this reaction system, the survival rate was 74.5%, 6 h after addition, suggesting that pancreatic islet cells were not damaged by free radicals. Based on the above experiment, the survival rate for pancreatic islet cells was compared 4 h after each aloe component was added to the HX-XO experimental system (Fig. 3). As a result, survival rates were 87.6 ± 3.05% (n = 5) in the untreated system and 37.4 ± 5.8% (n = 5) in the HX-XO system alone. Next, damage in isolated pancreatic islet cells was measured using Sz and Ax in the HX-HO system. The results are shown in Table 1. The cell survival rates 4 h after cell damage by Sz and Ax in the absence of Boiled leaf skin were 57.2 and 31.6%, respectively, but those in the presence Table 1 Effects of Boiled leaf skin on free radical-scavenging action in islet cells of rats exposed to Sz and Ax, comparison by concentration Treatment

Concentration (mg/ml)

No.

Viability (%)

Sz(+) Sz(+) Sz(+) Sz(−)

+ Boiled + Boiled + Boiled + Boiled

leaf leaf leaf leaf

skin skin skin skin

(−) (+) (+) (+)

– ×2 (8) ×4 (4) ×2 (8)

5 5 5 5

57.2 81.3 68.4 88.4

± ± ± ±

8.2 12.2∗∗ 23.5∗ 5.3

Ax(+) Ax(+) Ax(+) Ax(−)

+ Boiled + Boiled + Boiled + Boiled

leaf leaf leaf leaf

skin skin skin skin

(−) (+) (+) (+)

– ×2 (8) ×4 (4) ×2 (8)

5 5 5 5

31.6 82.8 59.1 88.1

± ± ± ±

6.5 8.1∗∗ 33.8∗ 8.4

–

5

93.7 ± 2.8

Vehicle

Values give absorbance units as mean ± S.D. Significant compared with Sz or Ax(+) + Boiled leaf skin (−) ∗ P < 0.01, ∗∗ P < 0.001. No.: cell culture number in each group.

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45

41

HX-XO + Leaf skin AP HX-XO + Leaf pulp FD HX-XO + Leaf skin FD

*

HX-XO + Boiled leaf skin

**

HX-XO + Boiled leaf pulp HX-XO alone Untreated

*

0

10

20

30

40

50

60

70

80

90 100

Viability (%) Fig. 3. Damage to isolated islet cells and effects of aloe components on the viability of cells from rats 4 h after exposure to oxygen radicals generated by HX-XO system. Values are the mean for the number of wells used (n = 5). Significantly different from HX-XO system alone ∗ P < 0.01, ∗∗ P < 0.001.

of 8 mg/ml Boiled leaf skin were 81.3 and 82.8%, respectively, showing significant inhibition of cell damage induced by Sz or Ax. Thus, Boiled leaf skin protected cells from free radicals (H2 O2 ) generated in the cell culture solution. Therefore, the same HX-XO system was used to separately test boiled components and non-boiled components, and the results obtained with these components were compared. These results are shown in Table 2. Boiled leaf skin and Boiled aloin increased survival rates compared to non-boiled components. However, aloin A (barbaloin) did not increase the survival rate even when it was boiled. Fig. 4 shows micrographs of cultured cells in the HX-XO reaction system after 4 h. The external appearance of isolated normal islet cells as the control (Fig. 4a) was similar between addition of (Fig. 4c) Boiled leaf skin, (Fig. 4d) addition of HX-XO plus Boiled leaf skin. But addition of (Fig. 4b) HX-XO alone, many destroyed cell fragments and swollen cells were observed.

In addition, we isolated and identified aloin, Boiled leaf skin, and phenolic compounds in Kidachi aloe by HPLC (Fig. 5), and found that chromatograms of components dissolved in methanol resembled chromatograms containing isobarbaloin, barbaloin, and aloenin, 2 -O-p-coumaroylaloesin, 2 -O-feruloylaloesin, which are specific to phenolic compounds. The number of peaks for Boiled leaf skin (Fig. 5D) was similar to that for aloin (Fig. 5C) rather than that for aloin A (Fig. 5B), but none of the peaks for aloin (Fig. 5C) was consistent with the peaks for aloenin (a). In addition, all the five types of peaks for Boiled leaf skin were higher than those for Leaf skin FD (Fig. 5E). The peaks of isobarbaloin and barbaloin were higher than those of 2 -O-p-coumaroylaloesin and 2 -O-feruloylaloesin in aloin, whereas the latter was higher than the former in Boiled leaf skin. To confirm the radical-scavenging activity of Boiled leaf skin, DPPH radical-scavenging activity of each phenolic

Table 2 Effects of aloe components on oxygen radical-scavenging action in islet cells of rats exposed in HX-XO, comparisons among leaf skin fraction, commercial aloin and aloin A (barbaloin) Treatment

Concentration (mg/ml)

No.

Viability (%)

HX-XO + Leaf skin FD HX-XO + Boiled leaf skin

×1 (16) ×2 (8) ×4 (4) ×16 (1) ×2 (8) ×2 (8) ×4 (4) ×2 (8) ×2 (8) ×4 (4)

5 5 5 5 5 5 5 5 5 5 5

56.3 85.0 77.8 42.8 75.3 65.2 50.2 39.3 38.7 34.2 30.4

HX-XO + Boiled aloin HX-XO + Non-boiled aloin HX-XO + Boiled aloin A HX-XO + Non-boiled aloin A HX-XO + vehicle

± ± ± ± ± ± ± ± ± ± ±

3.7∗ 18. 2∗∗,† 5.2∗∗,† 8.7 21.6∗,† 15.2∗ 8.3 13.3 11.5 8.7 12.2

Values give absorbance units as mean ±S.D. Significant compared with control (HX-XO + vehicle) ∗ P < 0.01, ∗∗ P < 0.001, and Leaf skin FD † P < 0.01.

42

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45

Fig. 4. Light microscopic views of effects of the oxygen radical-scavenging action of aloe components on the viability of islet cells (magnification 360×). (a) Islet cells, (b) islet cells plus HX-XO, (c) islet cells plus Boiled leaf skin, (d) islet cells plus HX-XO plus Boiled leaf skin.

compound of Kidachi aloe was measured. Table 3 shows 50% scavenging concentration SC50 (␮M) of each compound. Among these phenolic compounds, 2 -O-pcoumaroylaloesin and 2 -O-feruloylaloesin as chromone derivatives had more marked DPPH radical-scavenging effects than isobarbaloin and barbaloin as anthrone derivatives (Table 3). On the other hand, the scavenging effects of aloenin, the content of which is relatively high, were not so marked. The SC50 of Trolox and l-ascorbic acid as the controls were 49.9 and 43.6 ␮M, respectively, being similar to Table 3 50% radical-scavenging activity by concentration of phenol compounds by the DPPH-HPLC method Phenol compound

SC50 (␮M)a

Aloenin 2 -O-p-Coumaroylaloesin 2 -O-Feruloylaloesin Isobarbaloin Barbaloin Aloe-emodin (Sigma Chemical Co.)

1480.5 ± 69.3 ± 38.7 ± 373.7 ± 377.7 ± >2000.0

Trolox l-Ascorbic acid a

Each value is the mean ± S.E. (n = 3).

52.3 2.8 0.6 7.8 8.4

49.9 ± 1.6 43.6 ± 0.1

Table 4 Radical-scavenging activity of Leaf skin FD and Boiled leaf skin by the DPPH-HPLC method Aloe compound

Radical-scavenging activity (mM Trolox equivalent/g)a

Whole leaf FD Leaf skin FD Boiled leaf skin

48.2 ± 0.6 35.0 ± 0.8 61.6 ± 0.3

a

Each value is the mean ± S.E. (n = 3).

those of 2 -O-p-coumaroylaloesin and 2 -O-feruloylaloesin. The DPPH radical-scavenging activities of Whole leaf FD, Leaf skin FD, and Boiled leaf skin were 48.2, 35.0, and 61.6 mM Trolox equivalent/g. DPPH radical-scavenging activity was increased by boiling of Leaf skin FD (Table 4). These results show that a boiled Kidachi aloe leaf skin powder has scavenging effects on HX-XO-derived O2 radicals, free radicals produced by Sz and Ax, and DPPH radicals. 4. Discussion Previous studies have investigated the free radical-scavenging effects of aloe. It has been reported that mangano-SOD

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45 Fig. 5. Separation of standard compound and aloe components by HPLC. (A) standard: (a) aloenin, (b) 2 -O-p-coumaroylaloesin, (c) 2 -O-feruloylaloesin, (d) isobarbaloin, (e) barbaloin, (f) aloe-emodin; (B) aloin A (barbaloin 97%); (C) aloin (barbaloin 20%); (D) Boiled leaf skin; (E) Leaf skin (non-boiled). Samples were injected with aloin A (97% barbaloin); 2 ␮l of 1 mg/ml, commercial aloin; 2 ␮l of 1 mg/ml, Leaf skin (non-boiled); 20 ␮l of 1 mg/ml, Boiled leaf skin; 20 ␮l of 1 mg/ml. Column: Inertsil ODS-2 column (4.6 mm × 150 mm, GL Sciences Inc., Tokyo, Japan), mobile phase: water–methanol, elution profile: 0 min, 75:25; 10 min, 75:25; 25 min, 60:40; 35 min, 60:40; 45 min, 50:50; 65 min, 0:100; 75 min, 0:100. Chromatography was performed at 40 ◦ C with a flow-rate of 1.0 ml/min, and detection was at 254 nm.

43

44

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45

(Sabeh et al., 1996) with an estimated molecular weight of 32–42 kDa is present in Aloe vera leaf skin and leaf pulp. Cuand Zn-SOD activities (30 kDa or more) were detected in Kidachi aloe leaf skin and leaf pulp. SOD activities suggesting low molecular vitamins C, E, and carotinoids have been reported in 5 kDa fractions (Kato and Arai, 1997). It is known that the main low molecular substances found in aloe vegetables, anthraquinone and anthrone, exhibit antioxidant effects and radical-scavenging effects (Malterud et al., 1993). Takasu et al. reported H2 O2 production at concentrations of 1.45 and 0.92 nmol/islet/min after addition of 1 mM Sz and Ax, respectively, to the culture solution of isolated rat pancreatic islet cells. They also measured H2 O2 production in the islets 20 min after administration of Sz or Ax to the rat jugular vein and observed H2 O2 production 2.7 times and about twice, respectively, as much as the control value (Takasu et al., 1991; Komiya and Takasu, 1993). In this experiment, pancreatic islet cells isolated from rats were exposed to HX-XO system-generated free radicals and • OH free radicals generated by Sz and Ax. The oxygen radical-scavenging effects of aloe components were demonstrated (Fig. 3 and Table 1). Kidachi aloe Boiled leaf skin and commercially available Boiled aloin showed more potent free radical-scavenging effects than non-boiled samples, and protected pancreatic islet cells (Table 2). Furthermore, a HPLC chromatogram of Boiled leaf skin was similar to that of commercially available aloin (barbaloin content: approximately 20%) (Fig. 5). Though Boiled leaf skin contains aloenin that is characteristic of Kidachi aloe and not contained in Boiled aloin, the DPPH radical-scavenging action of this component was very weak (Table 3). Therefore, 2 -O-p-coumaroylaloesin and 2 -O-feruloylaloesin, which were increased by boiling (Fig. 5C–E), may have enhanced the radical-scavenging effects of Boiled leaf skin, resulting in additive effects (Table 4). In this experiment, Boiled leaf skin more potently protected cells than fresh Leaf skin FD. This suggests that not Kidachi aloe-derived SOD enzyme or SOD-like substances such as vitamins C and E but vegetable phenols are involved in antioxidant effects. Yagi et al. (2002) recently evaluated the anti-oxidation effects, free radical-scavenging effects, and anti-inflammatory effects of seven types of aloesin derivatives (chromone derivatives) that are present in Aloe vera. Our results concerning DPPH radical-scavenging activities of 2 -O-pcoumaroylaloesin and 2 -O-feruloylaloesin among these derivatives were consistent with those in their study. Furthermore, a solitary sample, aloin A (barbaloin content: approximately 97%), did not show any HX-XO systemgenerated radical-scavenging effects. However, rough aloin showed twofold free radical-scavenging effects. Therefore, components involved in the free radical-scavenging effects of Kidachi aloe and Aloe vera may disappear in the process

of purification from commercially available aloin to aloin A (barbaloin). In addition, since 2 -O-p-coumaroylaloesin and 2 -O-feruloylaloesin had free radical-scavenging effects, we intend to isolate and purify them and evaluate their in vivo effects.

Acknowledgements We thank Ms. K. Matsumori for assisting with this experiment. Furthermore, we thank to Professor Dr. Akira Yagi for supplying 2 -O-feruloylaloesin (include 2 -O-pcoumaroylaloesin) and Yurika Co., Ltd. for supplying fresh Kidachi aloe leaves. This study was conducted under a Fujita Gakuen Teaching Staff Subsidy.

References Abe, N., Nemoto, A., Tsuchiya, Y., Hojo, H., Hirota, A., 2000. Studies on the 1,1-diphenyl-2-picrylhydrazyl radical scavenging mechanism for a 2-pyrone compound. Bioscience, Biotechnology and Biochemistry 64, 306–313. Ajabnoor, M.A., 1990. Effect of aloes on blood glucose levels in normal and alloxan diabetic mice. Journal of Ethnopharmacology 28, 215–220. Appels, B., Burkart, V., Kantwerkfunke, G., Funda, J., Kolb-bachofen, V., Kolb, H., 1989. Spontaneous cytotoxicity of macrophages against pancreatic islet cells. Journal of Immunology 142, 3803–3808. Beppu, H., Nagamura, Y., Fujita, K., 1993. Hypoglycaemic and antidiabetic effects in mice of Aloe arborescens Miller var. natalensis Berger. Phytotherapy Research 7, 37–42. Burkart, V., Koike, T., Brenner, H.H., Kolb, H., 1992. Oxygen radicals generated by the enzyme xanthine oxidase lyse rat pancreatic islet cells in vitro. Diabetologia 35, 1028–1034. Chithra, P., Sajithial, G.B., Chandrakasan, G., 1998. Influences of Aloe vera on the healing of dermal wounds in diabetic rats. Journal of Ethnopharmacology 59, 195–201. Ghannam, N., Kingston, M., Al-Meshaal, I.A., Tariq, M., Parman, N.S., Woodhouse, N., 1986. The anti-diabetic activity of aloes: preliminary clinical and experimental observations. Hormone Research 24, 288– 294. Hikino, H., Takahashi, M., Murakami, M., Konno, C., Mirin, Y., Karikura, M., Hayashi, T., 1986. Isolation and hypoglycemic activity of arborans A and B, glycans of Aloe arborescens var. natalensis leaves. International Journal of Crude Drug Research 24, 183–186. Hotta, M., Tashiro, F., lkegami, H., Niwa, H., Ogihara, T., Yodoi, J., Miyazaki, J., 1998. Pancreatic B-cell-specific expression of thioredoxin, an anti-oxidative and anti-apoptotic protein revents, autoimmune and streptozotocin-induced diabetes. Journal of Experimental Medicine 188, 1445–1451. Kato, S., Arai, Y., 1997. SOD of Kidachi aloe. New Food Industry 39, 1–8 (in Japanese). Komiya, I., Takasu, N., 1993. Diabetogenic drug and free radical. Journal of Active Oxygens and Free Radical 4, 149–156 (in Japanese). Kuzuya, H., Tamai, I., Beppu, H., Shimpo, K., Chihara, T., 2001. Determination of aloenin, barbaloin and isobarbaloin in Aloe species by micellar electrokinetic chromatography. Journal of Chromatography B 752, 91–97. Malterud, K.E., Farbrot, T.L., Huse, A.E., Sund, R.B., 1993. Antioxidant and radical scavenging effects of anthraquinones and anthrones. Pharmacology 47, 77–85.

H. Beppu et al. / Journal of Ethnopharmacology 89 (2003) 37–45 Okamoto, H., 1985. Molecular basis of experimental diabetes: degeneration, oncogenesis and regeneration of pancreatic ␤-cells of islets of Langerhans. Bioessays 2, 15–21. Okamoto, H., 1990. The molecular basis of experimental diabetes. In: Okamoto, H. (Ed.), Molecular Biology of the Islets of Langerhans. Cambridge University Press, pp. 209–231. Okyar, A., Can, A., Akev, N., Baktir, G., Sütlüpinar, N., 2001. Effect of Aloe vera leaves on blood glucose level in type I and type II diabetic rat models. Phytotherapy Research 15, 157–161. Sabeh, F., Wright, T., Norton, S.J., 1996. Isozymes of superoxide dismutase from Aloe vera. Enzyme Protein 49, 212–221. Shimpo, K., Beppu, H., Chihara, T., Hoshino, M., Ida, C., Kuzuya, H., 2003. DPPH radical scavenging activity of Kidachi aloe its phenolic compounds. Fujita-Gakuen Igakkaishi (Bulletin of the Fujita Medical Society) 27, 1, in press.

45

Takasu, N., Komatsu, M., Aizawa, T., 1988. Hydrogen peroxide generation in islets synergistic regulation by cytoplasmic free calcium and protein kinase-C. Biochemical and Biophysical Research Communication 155, 569–575. Takasu, N., Komatsu, M., Asawa, T., 1991. Streptozotosin and alloxaninduced H2 O2 generation and DNA fragmentation in pancreatic islets. Diabetes 40, 1141–1145. Yagi, A., Kabash, A., Okamura, N., Haraguchi, H., Moustafa, S.M., Khalifa, T.I., 2002. Antioxidant, free radical scavenging and antiinflammatory effects of aloesin derivatives in Aloe vera. Planta Medica 68, 957–960. Yamaguchi, T., Takamura, H., Matoba, T., Terao, J., 1998. HPLC method for evaluation of the free radical-scavenging activity of foods by using 1,1-diphenyl-2-picrylhydrazyl. Bioscience, Biotechnology and Biochemistry 62, 1201–1204.