Diabetes mellitus is a serious chronic metabolic disorder that has a significant impact on the health, quality of life, and life expectancy of patients, as well as the healthcare system. Nojirimycine was first discovered from a streptomycete, and its naturally occurring hydrogenated product, 1-deoxynojirimycin (1-DNJ) was first isolated from the mulberry tree (Yoshikaki and Hivonu, 1976). Today, more than 20 polyhydroxy alkaloids have been identified from mulberry and silkworm (Asano et al., 1994a; 1994b; 2001). 1-DNJ is a piperidine alkaloid that is a highly effective alpha-glycosidase inhibitor (Yoshikuni, 1988; Yoshikuni et al., 1988; Hughes and Rudge, 1994) and an effective chemical for the treatment of hyperglycemia. Presently, 1-DNJ and its analogs have been isolated from a wide range of plants and microbes (Asano et al., 1998; 2000; Kim et al., 1999), but the 1-DNJ content in the mulberry tree is the highest among plants (Kimura et al., 2007; Yatsunami et al., 2008). Mulberry has been utilized as a Chinese medicine against diabetes mellitus for years. Ryu et al. (1997) first reported that the silkworm larval powder of the 5th instar (prepared by lyophilization) had beneficial effects on diabetic patients (Ryu et al., 1997; 1999), and the lowering of blood sugar levels was further elucidated by subsequent research (Han et al., 2007). Silkworm powder has blood glucose-lowering effects (Ryu et al., 2002), and mulberry leaves the diet of silkworms effectively inhibit alpha-glucosidase in the small intestine of humans (Oku et al., 2006). In Korea, Japan, and China, products made from mulberry and silkworm larvae are becoming popular for auxiliary therapy for diabetes mellitus. In this report, we compared the 1-DNJ content among silkworm varieties and investigated the antidiabetic effects of silkworm extracts on the db/db mouse.

Silkworm larvae were reared by feeding mulberry leaves in the spring season of 2012 at the National Academy Agricultural Science. The silkworm varieties used for the experiment were Yeonnokjam, Yangwonjam, and Hansaengjam. The 5^th instar 3^rd d larvae were quickly frozen with liquid nitrogen and lyophilized, and the resulting dried silkworm powder was extracted with ethanol.

1-DNJ content was measured according to the method reported by Kim et al. (2003). 1-DNJ in the silkworm larva or mulberry leaf powder was extracted with 0.05 mol/L HCl, treated with 9-fluorenylmethyl (FMOC) to produce a 1-DNJ-FMOC complex compound, and finally detected by high pressure liquid chromatography (HPLC).

Fasting blood glucose levels were measured after the animals fasted for 3 h for 4 wk. The db/db mice were obtained from Japan SLC, Inc (Japan), and animals at 6 wk of age were used. Blood glucose levels were determined in mouse blood samples from the tail vein using a glucose analyzer (Accu-Chek Active; Roche Diagnostics GmbH, Germany). Blood glucose levels were determined in blood samples from the tail vein at 0 min (prior to maltose administration), 30 min, 60 min, 90 min, and 120 min after maltose administration for glucose tolerance tests.

Male C57BL/KSJ-(db/db) mice (6 wk old) were purchased from Japan SLC, Inc (Japan). They were housed in a conventional state with the appropriate temperature (23℃ ± 3℃) and humidity (55% ± 15%) under a 12-h light/12-h dark cycle with free access to food and water. All groups were fed a standard diet (certified irradiated global 18% protein rodent diet). After a 1-wk adaptation period, the 7-wk-old mice were divided into the following 6 groups (n = 10 in each group): diabetic control group, low silkworm extract group (db/db low dose, 22.5 mg/kg), high silkworm extract group (db/db high dose, 180 mg/kg), and acarbose groups.

The mice fasted for 3 h, and then blood samples were taken after their autopsy. Biochemical analysis of blood TG, TCHO, LDL, GLU, AST, and ALT was performed using a blood biochemical analyzer (AU680, Beckman Coulter, Japan). Fat was taken from the circumference of the perirenal and epididymis during the autopsy, and the weight was measured.

We compared the 1-DNJ content among silkworm extracts of 5^th instar 3^rd d larvae and 5^th instar 5^th d larvae of the Yeonnokjam, Yangwonjam, and Hansaengjam. The Yeonnokjam and Hansaengjam larvae contained higher 1-DNJ than those of Yangwonjam. Furthermore, the 1-DNJ content of 5^th instar 5^th d larva powder was lower than that of the 5^th instar 3^rd d larvae. The best new candidate among these extracts was YR53. Yeonnokjam is one of the special silkworm varieties in Korea (Table 1).

[Table 1.] 1-Deoxynojirimycin content in silkworm varieties and its ratio in lyophilized material and extracts

We established a profile pattern of highly polyhydroxylated alkaloids, such as 1-deoxynojirimycin (1-DNJ), fagomine, and 1,4-dideoxy-1,4-imino-_D-arabinitol (DAB), of Yeonnokjam, a silkworm variety. We standardized the manufacturing process with this profile pattern, and now its development, including the establishment of standards as a natural medicine, is in progress (Fig. 1).

[Fig. 1.] HPLC chromatogram of 1-deoxynojirimycin (1-DNJ), fagomine, and DAB of silkworm (Yeonnokjam) extract.

In order to determine the optimal silkworm powder extraction, we selected conditions based on how they lowered glucose levels with varying ethanol concentrations. All groups, with the exception of control and YR30 groups, exhibited glucose-lowering effects. Among these, the YR70 group showed the highest glucose-lowering effects for 4 wk. In conclusion, the optimal extraction condition was 70% ethanol (Fig. 2).

[Fig. 2.] Effects of silkworm (Yeonnokjam) extracts on blood glucose levels in db/db mice

Based on blood glucose measurements, there was no statistically significant reduction in blood glucose levels in the G5 (180 mg/kg/d) group compared to that of the control group (P < 0.05 or P < 0.01). The blood glucose levels from wk 2 to 4 of the G3 (45 mg/kg/d) group did not decrease significantly. In contrast, a statistically significant reduction was observed for the G2 (22.5 mg/kg/d) group for 4 wk after administration (P < 0.05). The positive control (G6) group also exhibited a statistically significant reduction for 4 wk after administration (P < 0.01). The dosage test of silkworm extract powder needed an additional dosage lower than 22.5 mg/kg/d. Nevertheless, it is unclear why the effects of silkworm extract powder on blood glucose levels are dose independent (Fig. 3).

[Fig. 3.] Effects of Silkworm (Yeonnokjam) extracts on change of blood glucose in db/db mice.

The silkworm extract group did not show statistically significant changes in water intake compared to that of the control group. However, the 22.5 mg/kg/d and 45 mg/kg/d groups reduced their water intake at 3 and 4 wk after administration by approximately 47% and 33% and 41% and 20%, respectively. The water intake of the positive control group was reduced by approximately 70–75% compared to that of the control group over the course of 4 wk (Fig. 4).

[Fig. 4.] Effects of silkworm (Yeonnokjam) extracts on db/db mouse water consumption *

The two groups of G2 (22.5 mg/kg/d) and G5 (180 mg/kg/d) showed statistically significant decreases in epididymal fat weight (P < 0.05) compared to those of the control groups. For perirenal fat weights, the silkworm extract group did not show statistically significant decreases compared to those of control groups (Fig. 5).

[Fig. 5.] Effects of silkworm (Yeonnokjam) extracts on the weight of epididymal and perirenal fat weights

The positive control group did not show statistically significant decreases in epididymal fat weight, whereas its perirenal fat weight increased (P < 0.05 or P < 0.01).

The blood biochemical analysis indicated that the silkworm extract (180 mg) showed similar decreases in AST and ALT levels. Similar decreases were found in the positive control group as well. G2 (22.5 mg/kg/d) showed statistically significant decreases in GLU levels compared to that of the control groups. The positive control group (G6) also exhibited statistically significant decreases in GLU levels compared to the control groups (P < 0.05 or P < 0.01). The silkworm extract groups did not show statistically significant changes in TCHO, LDL, or HDL levels compared to those of the control groups, whereas the positive control group (G6) showed a statistically significant decrease compared to those of control groups (P < 0.05 or P < 0.01). G3 (45 mg/kg/d) showed statistically significant decreases in TCHO, LDL, and HDL levels compared to those of control groups (P < 0.05). The positive control group (G6) did not show statistically significant changes compared to those of control groups (P < 0.05 or P < 0.01) (Table 2).

[Table 2.] Blood biochemical analysis in db/db mice.