Biomedical Research (Tokyo) 36 (1) 63-70, 2015

Bifidobacterium species lower serum glucose, increase expressions of insulin signaling proteins, and improve adipokine profile in diabetic mice

Thi Kim Chung LE1, Toshio HOSAKA

2, Thanh Trung NGUYEN1, Afework KASSU

4, Thi Oanh DANG1, Hong Ba

TRAN1, Tran Phuong PHAM

3, Quang Binh TRAN3, Thi Hong Hao LE

1, and Xuan Da PHAM1

1 Department of Food Microbiology, National Institute for Food Control, 13 Phan Huy Chu, Hoan Kiem, Hanoi, Vietnam; 2 Third De-partment of Internal Medicine, Division of Diabetes, Endocrinology and Metabolism, Kyorin University School of Medicine, Mitaka, Tokyo, Japan; 3 National Institute of Hygiene and Epidemiology, 1 Yersin, Hanoi, Vietnam; and 4 Department of Medical Microbiology, University of Gondar, PO Box 196, Gondar, Ethiopia

(Received 20 November 2014; and accepted 7 December 2014)

ABSTRACTThis study, using C57BL/6J mice with streptozotocin (STZ)-induced diabetes, aimed to determine whether Bifidobacterium species (spp.) both induces the expressions of proteins in the insulin sig-naling pathway and enhances the expressions of certain adipocytokines. The protein expressions of IκB kinase alpha (IKKα), IκB kinase beta (IKKβ), nuclear factor-kappaB inhibitor alpha (IκBα), and the mitogen-activated protein kinase (MAPK) pathway were also investigated. Oral adminis-tration of Bifidobacterium spp. reduced blood glucose levels significantly and increased the protein expressions of insulin receptor beta, insulin receptor substrate 1, protein kinase B (Akt/PKB), IKKα, and IκBα. Extracellular-signal-regulated kinase 2 (ERK2) showed increased expression. Bifidobacterium spp. also induced the adiponectin expression and decreased both macrophage che-moattractant protein-1 (MCP-1) and interleukin-6 (IL-6) expression. In addition, IKKβ, c-Jun NH2-terminal kinase (JNK) and p38 MAP kinase expressions showed no significant changes in both groups. In conclusion, Bifidobacterium spp. may be the promising bacteria for treating diabetes.

Strains including B. bifidum, B. longum, B. infantis, and B. animalis are Bifidobacterium spp., which comprise genus of gram-positive, non-motile, often branched anaerobic bacteria. Bifidobacteria are used as probiotics for supporting digestion in many coun-tries. To date, several studies have demonstrated the benefits of probiotics in managing metabolic disor-ders including diabetes. At present, there are research groups focusing on this novel concept. Dietary supplementation with multiple probiotic strains, in-cluding L. acidophilus, L. casei, L. rhamnosus, L. bulgaricus, B. breve, B. longum, and S. thermophi-

lus, has been shown to prevent elevations of fasting plasma glucose in diabetic patients (3). The insulin signaling pathway controls glucose transport in mus-cle and fat cells. Insulin binds to insulin receptors on the surfaces of target cells. This binding activates insulin receptor beta (IR-β), and then activates insu-lin receptor substrate 1 (IRS-1), thereby recruiting phosphatidylinositol 3-kinase (PI3K) to this loca-tion. One important target of PI3K in muscle and fat cells is protein kinase B (Akt/PKB) which has a key function in glucose uptake (8). However, the influ-ence of Bifidobacterium spp. on insulin signaling pathway has not as yet been studied in detail. Probi-otic effects on the innate immune response path-ways, including the Toll-like receptor (TLR), nuclear factor kappa B (NF-κB), mitogen-activated protein kinase (MAPK), and c-Jun NH2-terminal kinase (JNK) (37) pathways, have been extensively investi-gated. Nevertheless, whether Bifidobacterium spp.

Address correspondence to: Dr Thi Kim Chung LeDepartment of Food Microbiology, National Institute for Food Control, 13 Phan Huy Chu, Hoan Kiem, Hanoi, VietnamFax: +84-4-39335738E-mail: lethikimchung27@yahoo.com

T. K. C. Le et al.64

IKKβ polyclonal antibody, rabbit anti-mouse IκBα polyclonal antibody, rabbit anti-mouse anti-ERK2 polyclonal antibody, rabbit anti-mouse JNK poly-clonal antibody, rabbit anti-mouse p38 MAP kinase polyclonal antibody, horseradish peroxidase (HRP)-conjugated anti-mouse and anti-rabbit antibodies were from Sigma-Aldrich (St. Louis, MO, USA). All other reagents used were of analytical grade.

Preparation of Bifidobacterium spp. strains for ad-ministration. Bifidobacterium spp. strains including B. bifidum ATCC 15700, B. longum, B. infantis, and B. animalis were obtained from DVS Biolife Limit-ed (Hyderabad, India). Bifidobacterium spp. strains were cultured overnight in MRS broth (Merck, Ger-many) added to 0.25% L-cysteine (Sigma) in anaero-bic condition at 37°C. Concentrations of the cultured strains were counted by culturing on MRS agar (Merck) containing 0.25% L-cystein and incubated at 37°C for 72 h in an anaerobic chamber containing an atmosphere of 85% N2, 10% H2 and 5% CO2. The cultured strains in MRS broth were collected and then washed three times with phosphate buffered saline (PBS) by centrifugation at 9,000× g for 30 min at 4°C. Mixing of freshly prepared live bacteria was performed daily for administration to each mouse just prior to use. The mixture of Bifidobacterium spp. included 109 colony-forming units (CFU) of B. longum, 109 CFU of B. bifidum, 109 CFU of B. infan-tis, and 109 CFU of B. animalis. This mixture was prepared in 0.4 mL of PBS and fed to the mice once a day.

Experimental design. Ten male C57BL/6J mice of seven-week-old were obtained from the Jackson Laboratory (Bar Harbor, ME, USA) for this study. All mice were maintained and treated in accordance with the Guidelines and Regulations for Experiments on Vertebrate Animals of the National Institute for Food Control in Testing, Research, and Training. The mice were housed individually in plastic cages in a room with controlled light (08 : 30–20 : 30), temperature (25 ± 1°C) and humidity (60 ± 5%) un-der conventional conditions.　Diabetes was induced in normal mice by five se-quential daily intraperitoneal injections of a freshly prepared solution of Streptozotocin (STZ) in citrate buffer (pH 4.5) at 45 mg/kg of body weight (31, 33). The mice were supplied with 10% sucrose wa-ter to avoid sudden post-injection hypoglycemia. The mice were tested to assure sufficient levels of hyperglycemia at 4 weeks post-injection. After all of the mice had been confirmed to be in a diabetic

exerts effects on those pathways remains as yet un-known.　The mixture of Bifidobacterium infantis and Lac-tobacillus acidophilus reportedly decreased the in-terleukin-6 (IL-6) mRNA level in primary intestinal epithelial cells isolated from resected ileum with necrotizing enterocolitis (11). Bifidobacteria longum significantly down-regulated levels of macrophage chemoattractant protein-1 (MCP-1) and IL-6 in por-cine intestinal epithelial cells challenged with heat-killed enterotoxigenic Escherichia coli (34). The mixture that contained B. longum and B. animalis inhibited MCP-1 production in colitis of BALB/c mice (25). B. bifidum was shown to suppress MCP-1 expression in an in vitro co-culture of intestinal epithelial cells and T cells (18). However, only one recent study focused on the anti-diabetic effect of Bifidobacterium spp. and its molecular mechanism (13). Infection with Kilham rat virus leads to in-creased Bifidobacterium spp. in the intestine, which raises the possibility that virus-induced type 1 dia-betes may involve alterations in the balance among bacterial groups, resulting in the down-regulation of protective mechanisms or, conversely, the up-regula-tion of pro-inflammatory innate pathways culminat-ing in anti-islet T cell responses (13).　Therefore, we examined whether oral administra-tion of Bifidobacterium spp. ameliorates diabetes, possibly via effects on the insulin signaling pathway and inflammatory adipocytokines, in a lean insulin-depleted murine diabetic model; mice with strepto-zotocin (STZ)-induced diabetes. This murine model of diabetes is not influenced by obesity which is re-lated to insulin resistance. We evaluated the effects of Bifidobacterium spp. on the mRNA expressions of adiponectin, MCP-1 and IL-6, as well as the pro-tein levels of IRS-1, IR-β and Akt in adipose tissues from mice with diabetes induced by STZ. Further-more, the protein expressions of IκB kinase alpha (IKKα), IκB kinase beta (IKKβ), nuclear factor-kap-pa B inhibitor alpha (IκBα) and MAP kinase path-way proteins, such as extracellular-signal-regulated kinase 2 (ERK2), JNK, and p38 MAP kinase, were also investigated.

MATERIALS AND METHODS

Antibodies. Mouse anti-mouse IR-β monoclonal anti-body was purchased from Abcam (Cambridge, UK). Rabbit anti-mouse IRS-1 polyclonal antibody, rabbit anti-mouse Akt polyclonal antibody, mouse anti-mouse β-Actin monoclonal antibody, rabbit anti-mouse IKKα polyclonal antibody, rabbit anti-mouse

Bifidobacterium species enhance insulin signaling 65

amplification by melting curve analysis. Data were normalized by the 18S ribosomal RNA expression levels in each sample and shown as the means ± SE of 3 independent quantitative-RT-PCR analyses.

Western blot analysis. At the end of the experimen-tal period, adipose tissues were taken from mice and then rinsed three times with ice-cold PBS and lysed in RIPA buffer (50 mM Tris, 150 mM NaCl, 1 mM EDTA, 0.5% sodium deoxycholate, 1% NP-40, 0.1% SDS, 1 μM aprotinin, 10 μM leupeptin, 0.1 μM phenylmethylsulfonyl fluoride, 20 mM sodium fluo-ride, 20 mM beta glycerophosphate and 1 mM sodi-um orthovanadate, pH 7.4). The supernatants were obtained by centrifugation at 13,000 rotations per minute at 4°C for 20 min, and the protein content was determined for each sample employing the Nanodrop 1000 system (Thermo Scientific, USA). Equal amounts of protein (100 μg) were subjected to sodium dodecyl sulfate-polyacrylamide gel electro-phoresis. Gels were transferred to an Immobilon-P membrane (Millipore, Bedford, MA, USA). Immu-noblotting was performed with an enhanced chemi-luminescence kit (Pierce, Thermo Scientific, USA) according to the manufacturer’s instructions. Densi-tometric analyses were performed using Image J1.38 x software (NIH, Bethesda, MD, USA).

Statistical analysis. The values obtained are present-ed as means ± SE. Significant differences (*P < 0.05 or **P < 0.01) between means were evaluated by applying the unpaired Student’s t-test.

RESULTS

Bifidobacterium spp. improved adipose insulin sig-naling sensitivity in diabetic miceBlood glucose testing is the gold standard for sub-clinical diagnosis of diabetes. Mice with blood glu-cose levels higher than 16.7 mmol/L one week after STZ injection were considered to have diabetes (33). The insulin concentrations in these mice could not be measured due to the insufficient sensitivity of the enzyme-linked immune sorbent assay (ELISA) used. All of the diabetic mice were randomly divided into 2 groups. After 5 weeks with (trial group) or with-out (control group) probiotic administration, blood glucose levels of the trial group were significantly lower than those of the control group (P < 0.01) (Table 1). Neither body nor adipose tissue weights differed significantly between the trial and control groups of mice (Table 1). To elucidate the role of Bifidobacterium spp. in the expressions of insulin

state (blood glucose greater than 16.7 mmol/L) with-in 1 week (26), they were randomly assigned to trial and control groups. The trial group (n = 5) received a normal diet (Oriental Yeast Tokyo, Japan) with the mixture of Bifidobacterium spp. for 5 weeks by oral gavage while the control group (n = 5) was given only the normal diet with saline at an equivalent volume by oral gavage for 5 weeks. The normal diet (AIN-93M) contained (per 100 g of the diet) 14.00 g of milk casein, 0.18 g of L-cysteine, 46.57 g of corn starch, 15.50 g of α-corn starch, 10.00 g of sucrose, 4.00 g of soybean oil, 5.00 g of powdered cellulose, 3.50 g of AIN-93M mineral mixture, 1.00 g of AIN-93M vitamin mixture, 0.25 g of choline bitartrate, and 0.0008 g of tert-butylhydroquinone. At the end of the experimental period, all mice were fasted for 6 h, anesthetized with diethyl ether and then exsan-guinated from the heart. The blood of each mouse was collected to measure glucose levels. The mes-enteric fat of each mouse also was collected and weighed. Total RNA and protein of adipose tissue were isolated from adipose tissue of both the trial and the control group.

RNA extraction and reverse transcription PCR. TRIzol reagent (Life Technologies, Carlsbad, CA, USA) was used to extract total RNA from adipose tissue of C57BL/6J mice. The RNA was treated with DNase I (Life Technologies), and 0.5 μg RNA was reverse transcribed using TaKaRa PrimeScript™ RT reagent kits according to the manufacturer’s in-structions (Takara Bio, Kyoto, Japan).

Quantitative RT-PCR. Samples were prepared using TaKaRa SYBR Premix Ex Taq according to the manufacturer’s instructions (Takara Bio). Quantita-tive real-time reverse transcription PCR with 100 ng cDNA was performed using the CFX96 system (Bio-Rad Laboratories, Hercules, CA, USA). The fol-lowing oligonucleotide primers were used: adiponec-tin 5’-TGTTGGAATGACAGGAGCTG-3’ (forward) and 5’-CGAATGGGTACATTGGGAAC-3’ (reverse); IL-6 5’-ACAACCACGGCCTTCCCTACTT-3’ (for-ward) and 5’-CACGATTTCCCAGAGAACATGTG-3’ (reverse); MCP-1 5’-CCCAATGAGTAGGCTGGAG A-3’ (forward) and 5’-TCTGGACCCATTCCTTC TTG-3’ (reverse); 18S ribosomal RNA 5’-AAACGG CTACCACATCCAAG-3’ (forward) and 5’-GGC CTCGAAAGAGTCCTGTA-3’ (reverse). The an-nealing temperature for the PCR reactions was 60°C, product sizes ranged from 172 to 225, and the total reaction mixture volume was 20 μL. After the reac-tion, each PCR product was verified for its single

T. K. C. Le et al.66

Bifidobacterium spp. treatment increased the adipo-nectin mRNA level and reduced MCP-1 and IL-6 mRNA levels in adipose tissue of mice with STZ-in-duced diabetesTo assess the role of Bifidobacterium spp. in im-proving the blood glucose levels of STZ-insulin-de-pleted mice, the mRNA levels of adiponectin, one of the adipokines influencing insulin sensitivity, in adipocytes were measured in both the trial and the control group. As shown in Fig. 3, the adiponectin mRNA level was markedly elevated in the trial group as compared to the control group (P < 0.01). In order to explore the role of Bifidobacterium spp. in the production of inflammatory adipokines, the mRNA expressions of IL-6 and MCP-1 were evalu-ated in both groups of mice with the STZ-induced diabetes. Both MCP-1 and IL-6 mRNA levels were markedly reduced in the trial group as compared to the control group (Fig. 4).

signaling pathway proteins, diabetic mice were as-sessed for the presence of IRS-1, IR-β and Akt. As shown in Fig. 1, the expressions of IR-β, IRS-1, and Akt were significantly increased in adipose tissue from the trial group as compared to that from the control after normalization by β-Actin expression.

Bifidobacterium spp. increased IKKα, IκBα and ERK2 signals but did not affect JNK and p38 MAP kinase protein expressionsWe examined the expressions of IKKα, IKKβ, and IκBα to investigate the impact of the mixture of Bi-fidobacterium spp. The levels of ERK2, JNK and p38 MAP kinase proteins involved in the MAPK pathway were also assessed. As shown in Fig. 2, ex-pressions of IKKα, IκBα, and ERK2 were signifi-cantly increased in the trial group as compared to the control group. However, the expressions of IKKβ, JNK, and p38 MAP kinase were unchanged in both groups.

Fig. 1　Expressions of insulin signaling pathway proteins in diabetic mice. Twelve-week-old mice with STZ-induced diabetes (n = 10) were treated with (trial group) or without (control group) a mixture of Bifidobacterium spp. for 5 weeks. Adipose tis-sue was harvested from the diabetic mice of both groups after a 6-hour fast. Total protein was extracted from adipose tis-sue and subjected to Western blot using IR-β, IRS-1, Akt, and β-Actin antibodies as described in Materials and Methods (Fig. 1A). Results were also quantified and the fold-increase from the trial group as compared to the control group was nor-malized with β-Actin (Fig. 1B). Data shown are the means ± SE (n = 5) (**P < 0.01 compared to the control group).

Table 1　Body weight, dietary intake and mesenteric fat weight in diabetic mice

Parameter Control (n = 5) Trial (n = 5)Body weight (initial) (g) 26.05 ± 1.02 27.44 ± 1.33NS

Body weight (final) (g) 32.21 ± 0.74 31.00 ± 1.66NS

Dietary intake (g/2 days) 3.96 ± 0.24 3.93 ± 0.41NS

Mesenteric fat weight (g) 0.46 ± 0.05 0.45 ± 0.06NS

Blood glucose (mmol/L) 17.9 ± 1.8 11.6 ± 1.0**

Twelve-week-old male C57BL/6J mice with diabetes (n = 10) were fed a normal diet with (trial group) or without (control group) a mixture of Bifidobacterium spp. for 5 weeks by oral gavage. The control group was given the same volume of saline for 5 weeks. Values are expressed as means ± SE (n = 5). N.S., no significant differences between control and Bifidobacterium spp. groups. (**P < 0.01 compared to the control)

Bifidobacterium species enhance insulin signaling 67

DISCUSSION

Previous studies indicated that an IRS-1 knockout murine model showed mild insulin intolerance and ablation of the insulin receptor in adipose tissues, which resulted in diet-induced obesity and marked resistance to the development of glucose intolerance (35). In mice, homozygous knockout of the insulin receptor caused death at 3–7 days post parturition (1, 16). Other study indicated that null alleles of the in-sulin receptor and IRS-1 (IRS-1 null animals are mildly insulin intolerant) developed frank diabetes at 4–6 months of age (7). Although skeletal muscle accounts for more than 80% of postprandial glucose disposal, the ablation of the insulin receptor specifi-cally in muscle resulted in normal glucose tolerance (6). Otherwise, probiotics may compensate for defi-cits in the insulin-like growth factor I receptor/PI3K/Akt survival pathways and attenuate cardiac apopto-sis in spontaneously hypertensive rats (21). As for the insulin signaling pathway, we demonstrated that

oral administration of Bifidobacterium spp. increased the levels of IR-β, IRS-1, and Akt proteins in dia-betic mice. In this study, Akt up-regulation may have improved glucose uptake as shown by reduced plasma glucose levels. Therefore, we can reasonably speculate that Bifidobacterium spp. promote the re-covery of β-cells of pancreas or increase insulin sensitivity in mice with STZ-induced diabetes by enhancing the function of the insulin signaling path-way. Recovery of β-cell function in the trial group is even possible. However, serum insulin levels were not detectable by the ELISA method within the 5-week treatment period of this study.　Toll-like receptors are related to host immune re-sponses (12). To date, three major MAPK pathways have been identified in mammals: ERK, stress-acti-vated protein kinase/JNK, and p38 (12). Our obser-vation that there were no changes in the levels of JNK and p38 MAP kinase under any of the study conditions, while expression of ERK2 was in-creased, raises the possibility of probiotics playing a

Fig. 2　Expressions of Toll-like receptor and MAPK pathway proteins in diabetic mice. Twelve-week-old mice with STZ-in-duced diabetes (n = 10) were treated with (trial group) or without (control group) a mixture of Bifidobacterium spp. for 5 weeks. Adipose tissue was harvested from the diabetic mice of both groups after a 6-hour fast. Total protein was extracted from adipose tissue and subjected to Western blot using IKKα, IKKβ, IκBα, ERK2, JNK, p38MAP kinase, and β-Actin anti-bodies as described in Materials and Methods (Fig. 2A). Results were also quantified and the fold-increase from the trial group as compared to the control group was normalized with β-Actin (Fig. 2B). Data shown are the means ± SE (n = 5) (**P < 0.01 compared to the control). N.S., not significant.

T. K. C. Le et al.68

with type 1 diabetes reportedly have higher adipo-nectin levels than non-diabetic children (29) and this might be an autoimmune mechanism underlying in-sulin depletion. We showed adiponectin mRNA to be significantly higher in the trial group than in the control group (P < 0.01). This suggests that adipo-nectin may ameliorate the disease state in mice with STZ-induced diabetes. This mechanism may under-lie one of the benefits of Bifidobacterium spp. inges-tion. IL-6 was reported to reduce insulin-dependent hepatic glycogen synthesis (19, 28) and glucose up-take in adipocytes (27). Subjects with a high after-therapy IL-6 level showed poorer periodontal healing than those with lower levels (24). Significant ex-pression of MCP-1 was documented in the myocar-dial tissues of cases with type 1 diabetes suffering from diabetic keto-acidosis, while smaller amounts of MCP-1 were expressed in the myocardial tissues of overweight/obese individuals (23). Another study involving 70 patients with type 1 diabetes showed that all indices of pro-inflammatory cytokines were higher in patients than in healthy control subjects. These observations suggest a positive correlation be-tween blood pro-inflammatory cytokine levels and disease severity (20). In this study, we showed oral administration of a mixture of Bifidobacterium spp. to decrease the mRNA expressions of IL-6 and

role in the MAPK pathway. ERK, a widely ex-pressed protein kinase, is an intracellular signaling molecule involved in functions including the regula-tion of proliferation, differentiation, and survival of cells (36). The ERK2 increase in mice fed Bifido-bacterium spp. may indicate differentiation into a cell type capable of inducing insulin sensitivity in diabetic mice. One prior study indicated dysregula-tion of nuclear IKKα to be linked to diabetes (30). In that study, increased cytosolic IKKα was suggest-ed to increase the phosphorylation of IKKα by Akt, thereby leading to nuclear accumulation of IKKα and supporting the diverse functions of IKKα such as apoptosis, tumor suppression, actions on the im-mune system, cell proliferation and chromatin re-modeling in an NF-κB independent manner (14). In addition, the increased cytosolic IκBα observed in our Bifidobacterium spp. treated diabetic mice may have inhibited the effects of NF-κB, thereby leading to decreased transcription of certain pro-inflammato-ry genes such as those of IL-6 and MCP-1. JNK is known to participate in the inhibition of insulin ac-tion and its phosphorylation is presumed to disrupt the interaction between insulin receptor and IRS-1 (5). A study by Tanaka et al. indicated that IKKβ induces both insulin action and diabetes, by a mech-anism involving defective secretion in response to inflammatory cytokines (32). Another study showed that p38 MAP kinase is related to insulin resistance (22). Therefore, the lack of effects of JNK, IKKβ and p38 MAP kinase in the current study may sug-gest that deleterious actions associated with diabetes do not appear in Bifidobacterium spp. treated diabet-ic mice. Furthermore, Akt may affect IKKα and even result in the activation of IκBα which may in turn inhibit the effects of NF-κB (15) and lead to reduced transcription of target genes such as those of IL-6 and MCP-1. Herein, we did not clarify the individual roles of B. longum, B. bifidum, B. infantis, and B. animalis in the Bifidobacterium spp. strain mixture. Further studies are needed to clarify the ef-fects of each bacterial species.　Adipose tissues produce leptin and adiponectin to regulate feeding behavior and generate pro- and an-ti-inflammatory adipokines which modulate inflam-matory responses (9). Adiponectin enhances insulin sensitivity by increasing hepatic IRS-2 expression via a macrophage-derived IL-6-dependent pathway (4), while IL-6 and MCP-1 inhibit insulin sensitivity (17, 19). Adiponectin gene expression and secretion are inhibited by IL-6 in 3T3-L1 adipocytes (10). In-sulin signaling may also be impaired by altered se-cretion of cytokines and chemokines (2). Children

Fig. 3　Adiponectin expression increased in Bifidobacterium spp. treated diabetic mice. Twelve-week-old mice with STZ-induced diabetes (n = 10) were treated with (trial group) or without (control group) a mixture of Bifidobacterium spp. for five weeks. Total RNA isolated from adipose tissue was subjected to quantitative real-time RT-PCR with primers specific for adiponectin as described in Materials and Meth-ods. Data were normalized by 18S ribosomal RNA. Results were quantified, and the fold-increase from the control groups is shown. Data shown are the means ± SE (n = 5) (**P < 0.01 compared to the control).

Bifidobacterium species enhance insulin signaling 69

MCP-1 in diabetic mice. This may represent true benefit of Bifidobacterium spp. ingestion for elimi-nating the unwanted features of diabetes in mice.　In summary, oral administration of Bifidobacteri-um spp. enhances the expressions of proteins in-volved in the insulin-signaling pathway and increases levels of proteins related to innate immune respons-es. Moreover, the current study demonstrated that a mixture of Bifidobacterium spp. also decreased the blood glucose level and increased the expression of adiponectin mRNA while decreasing those of in-flammatory adipokines (MCP-1 and IL-6) in insulin-depleted diabetic mice. Our findings may facilitate understanding the novel effects of Bifidobacterium spp. and suggest a newly-recognized benefit of Bifi-dobacterium spp. treatment for diabetes. It is neces-sary to elucidate the underlying mechanisms of action and to evaluate the efficacy of Bifidobacteri-um spp. treatment in diabetic patients.

Acknowledgments

This research was funded by the Vietnam National Foundation for Science and Technology Development (NAFOSTED) under grant number 106.05-2011.58.

REFERENCES

Fig. 4　Bifidobacterium spp. decreased MCP-1 and IL-6 mRNA levels in adipose tissue of mice with STZ-induced diabetes. Twelve-week-old mice with STZ-induced diabetes (n = 10) were treated with (trial group) or without (control group) a mixture of Bifidobacterium spp. for 5 weeks. Adipose tissue was harvested after a 6-hour fast. Total mRNA was extracted from adi-pose tissue and subjected to quantitative real time RT-PCR with specific primers for MCP-1 (A) and IL-6 (B) as described in Materials and Methods. Data were normalized by 18S ribosomal RNA. Results were quantified, and the fold-increase from the control groups is shown. Data shown are the means ± SE (n = 5) (*P < 0.05 compared to the control).

1. Accili D, Drago J, Lee EJ, Johnson MD, Cool MH, Salvatore P, Asico LD, Jose PA, Taylor SI and Westphal H (1996) Ear-

ly neonatal death in mice homozygous for a null allele of the insulin receptor gene. Nat Genet 12, 106–109.

2. Antuna-Puente B, Feve B, Fellahi S and Bastard JP (2007) [Obesity, inflammation and insulin resistance: which role for adipokines]. Therapie 62, 285–292.

3. Asemi Z, Zare Z, Shakeri H, Sabihi SS and Esmaillzadeh A (2013) Effect of multispecies probiotic supplements on meta-bolic profiles, hs-CRP, and oxidative stress in patients with type 2 diabetes. Ann Nutr Metab 63, 1–9.

4. Awazawa M, Ueki K, Inabe K, Yamauchi T, Kubota N, Kaneko K, Kobayashi M, Iwane A, Sasako T, Okazaki Y, Ohsugi M, Takamoto I, Yamashita S, Asahara H, Akira S, Kasuga M and Kadowaki T (2011) Adiponectin enhances in-sulin sensitivity by increasing hepatic IRS-2 expression via a macrophage-derived IL-6-dependent pathway. Cell Metab 13, 401–412.

5. Boura-Halfon S and Zick Y (2009) Phosphorylation of IRS proteins, insulin action, and insulin resistance. Am J Physiol Endocrinol Metab 296, E581–591.

6. Bruning JC, Michael MD, Winnay JN, Hayashi T, Horsch D, Accili D, Goodyear LJ and Kahn CR (1998) A muscle-specific insulin receptor knockout exhibits features of the metabolic syndrome of NIDDM without altering glucose tolerance. Mol Cell 2, 559–569.

7. Bruning JC, Winnay J, Bonner-Weir S, Taylor SI, Accili D and Kahn CR (1997) Development of a novel polygenic model of NIDDM in mice heterozygous for IR and IRS-1 null alleles. Cell 88, 561–572.

8. Bryant NJ, Govers R and James DE (2002) Regulated trans-port of the glucose transporter GLUT4. Nat Rev Mol Cell Biol 3, 267–277.

9. Carvalho-Filho MA, Ueno M, Hirabara SM, Seabra AB, Carvalheira JB, de Oliveira MG, Velloso LA, Curi R and Saad MJ (2005) S-nitrosation of the insulin receptor, insulin receptor substrate 1, and protein kinase B/Akt: a novel mech-

T. K. C. Le et al.70

anism of insulin resistance. Diabetes 54 959–967.10. Fasshauer M, Kralisch S, Klier M, Lossner U, Bluher M,

Klein J and Paschke R (2003) Adiponectin gene expression and secretion is inhibited by interleukin-6 in 3T3-L1 adipo-cytes. Biochem Biophys Res Commun 301, 1045–1050.

11. Ganguli K, Meng D, Rautava S, Lu L, Walker WA and Nanthakumar N (2013) Probiotics prevent necrotizing entero-colitis by modulating enterocyte genes that regulate innate immune-mediated inflammation. Am J Physiol Gastrointest Liver Physiol 304, G132–141.

12. Ha U, Lim JH, Jono H, Koga T, Srivastava A, Malley R, Pages G, Pouyssegur J and Li JD (2007) A novel role for IkappaB kinase (IKK) alpha and IKKbeta in ERK-dependent up-regulation of MUC5AC mucin transcription by Strepto-coccus pneumoniae. J Immunol 178, 1736–1747.

13. Hara N, Alkanani AK, Ir D, Robertson CE, Wagner BD, Frank DN and Zipris D (2012) Prevention of virus-induced type 1 diabetes with antibiotic therapy. J Immunol 189, 3805–3814.

14. Huang WC and Hung MC (2013) Beyond NF-kappaB acti-vation: nuclear functions of IkappaB kinase alpha. J Biomed Sci 20, 3.

15. Jacobs MD and Harrison SC (1998) Structure of an IkappaB alpha/NF-kappaB complex. Cell 95, 749–758.

16. Joshi RL, Lamothe B, Cordonnier N, Mesbah K, Monthioux E, Jami J and Bucchini D (1996) Targeted disruption of the in-sulin receptor gene in the mouse results in neonatal lethality. EMBO J 15, 1542–1547.

17. Kanda H, Tateya S, Tamori Y, Kotani K, Hiasa K, Kitazawa R, Kitazawa S, Miyachi H, Maeda S, Egashira K and Kasuga M (2006) MCP-1 contributes to macrophage infiltration into ad-ipose tissue, insulin resistance, and hepatic steatosis in obesi-ty. J Clin Invest 116, 1494–1505.

18. Kim N, Kunisawa J, Kweon MN, Eog JG and Kiyono H (2007) Oral feeding of Bifidobacterium bifidum (BGN4) pre-vents CD4(+) CD45RB(high) T cell-mediated inflammatory bowel disease by inhibition of disordered T cell activation. Clin Immunol 123, 30–39.

19. Klover PJ, Zimmers TA, Koniaris LG and Mooney RA (2003) Chronic exposure to interleukin-6 causes hepatic insulin re-sistance in mice. Diabetes 52, 2784–2789.

20. Kyiak I, Fartushok NV, Onyshchuk I, Fedevych I and Bashta HV (2012) [Profile of proinflammatory cytokines in type 1 diabetes mellitus]. Fiziol Zh 58, 65–69.

21. Lin PP, Hsieh YM, Kuo WW, Lin YM, Yeh YL, Lin CC, Tsai FJ, Tsai CH, Huang CY and Tsai C (2013) Probiotic-fermented purple sweet potato yogurt activates compensatory IGFIR/PI3K/Akt survival pathways and attenuates cardiac apoptosis in the hearts of spontaneously hypertensive rats. Int J Mol Med 32, 1319–1328.

22. Liu Z and Cao W (2009) p38 mitogen-activated protein ki-nase: a critical node linking insulin resistance and cardiovas-cular diseases in type 2 diabetes mellitus. Endocr Metab Immune Disord Drug Targets 9, 38–46.

23. Niu J, Gilliland MG, Jin Z, Kolattukudy PE and Hoffman WH (2014) MCP-1and IL-1beta expression in the myocardia of two young patients with Type 1 diabetes mellitus and fatal diabetic ketoacidosis. Exp Mol Pathol 96, 71–79.

24. Passoja A, Knuuttila M, Hiltunen L, Karttunen R, Niemela O, Raunio T, Vainio O, Hedberg P and Tervonen T (2011) Se-rum interleukin-6 may modulate periodontal inflammation in type 1 diabetic subjects. J Clin Periodontol 38, 687–693.

25. Roselli M, Finamore A, Nuccitelli S, Carnevali P, Brigidi P, Vitali B, Nobili F, Rami R, Garaguso I and Mengheri E (2009) Prevention of TNBS-induced colitis by different Lac-tobacillus and Bifidobacterium strains is associated with an expansion of gammadeltaT and regulatory T cells of intestinal intraepithelial lymphocytes. Inflamm Bowel Dis 15, 1526–1536.

26. Rosenberg L, Lipsett M, Yoon JW, Prentki M, Wang R, Jun HS, Pittenger GL, Taylor-Fishwick D and Vinik AI (2004) A pentadecapeptide fragment of islet neogenesis-associated pro-tein increases beta-cell mass and reverses diabetes in C57BL/ 6J mice. Ann Surg 240, 875–884.

27. Rotter V, Nagaev I and Smith U (2003) Interleukin-6 (IL-6) induces insulin resistance in 3T3-L1 adipocytes and is, like IL-8 and tumor necrosis factor-alpha, overexpressed in hu-man fat cells from insulin-resistant subjects. J Biol Chem 278, 45777–45784.

28. Senn JJ, Klover PJ, Nowak IA and Mooney RA (2002) Inter-leukin-6 induces cellular insulin resistance in hepatocytes. Diabetes 51, 3391–3399.

29. Solomon JR and Varadarajan P (2013) Adiponectin levels in south Indian children with type 1 diabetes mellitus and non-diabetic children and its correlation with anthropometry and glycemic control. Pediatr Endocrinol Rev 11, 34–43.

30. Starkey JM, Haidacher SJ, LeJeune WS, Zhang X, Tieu BC, Choudhary S, Brasier AR, Denner LA and Tilton RG (2006) Diabetes-induced activation of canonical and noncanonical nuclear factor-kappaB pathways in renal cortex. Diabetes 55, 1252–1259.

31. Szkudelski T (2001) The mechanism of alloxan and strepto-zotocin action in B cells of the rat pancreas. Physiol Res 50, 537–546.

32. Tanaka M, Fuentes ME, Yamaguchi K, Durnin MH, Dalrymple SA, Hardy KL and Goeddel DV (1999) Embryonic lethality, liver degeneration, and impaired NF-kappa B activation in IKK-beta-deficient mice. Immunity 10, 421–429.

33. Tang J, Du Y, Petrash JM, Sheibani N and Kern TS (2013) Deletion of aldose reductase from mice inhibits diabetes- induced retinal capillary degeneration and superoxide genera-tion. PLoS One 8, e62081.

34. Tomosada Y, Villena J, Murata K, Chiba E, Shimazu T, Aso H, Iwabuchi N, Xiao JZ, Saito T and Kitazawa H (2013) Immu-noregulatory effect of bifidobacteria strains in porcine intesti-nal epithelial cells through modulation of ubiquitin-editing enzyme A20 expression. PLoS One 8, e59259.

35. Watson RT, Kanzaki M and Pessin JE (2004) Regulated membrane trafficking of the insulin-responsive glucose trans-porter 4 in adipocytes. Endocr Rev 25, 177–204.

36. Wortzel I and Seger R (2011) The ERK cascade: distinct functions within various subcellular organelles. Genes Can-cer 2, 195–209.

37. Yoon SS and Sun J (2011) Probiotics, nuclear receptor sig-naling, and anti-inflammatory pathways. Gastroenterol Res Pract 971–938.