The consumption of functional foods with established antidiabetic components is perhaps one of the best documented strategies to treat and/or inhibiting the prevalence of diabetes, worldwide (Riezzo et al. Reference Riezzo, Chiloiro and Russo2005). Dairy foods constitute a family of natural functional foods with their established health benefits. Dahi is an Indian home made variant of yogurt, which differs from yoghurt in the use of mixed starters of mesophilic lactococci. The principle flavour inducing metabolite in dahi is diacetyl, which is relished more by people of South-Asian origin than the acetaldehyde flavour in yoghurt. Dahi has been considered as a functional food due to its several health benefits i.e. antidiarreal, anticarcinogenic, cholesterol lowering and antiatherogenic properties (Chawla & Kansal, Reference Chawla and Kansal1984; Abbas & Jafri, Reference Abbas and Jafri1992). Probiotics are the live microbial food supplements which give health benefits beyond basic nutrition upon consumption in sufficient amount (FAO/WHO, 2001). Different strains of lactic acid bacteria (LAB) such as Lactobacillus and Bifidobacterium are considered as important probiotics regimens. Fermented milk products such as yogurt, containing probiotic LAB, are well established for several health benefits (Reid et al. Reference Reid, Anand, Bingham, Mugua, Wadstrom, Fuller, Anukam and Katsivo2005). Few reports have dealt with the antidiabetic effect of LAB. Matsuzaki et al. (Reference Matsuzaki, Nagata, Kado, Uchida, Kato, Hashimoto and Yokokura1997a, Reference Matsuzaki, Yamazaki, Hashimoto and Yokokurab), reported that oral administration of Lb. casei has a preventive effect on elevation of plasma glucose and reduction of plasma insulin levels by preventing immune mediated destruction of pancreatic β-cells in NOD and KK-Ay mice. Moreover, they also reported that feeding of Lb. casei supplemented diet strongly inhibited disappearance of β-cells and nitric oxide production in alloxan injected BALB/c mice (Matsuzaki et al. Reference Matsuzaki, Nagata, Kado, Uchida, Hashimoto and Yokokura1997c). Tabuchi et al. (Reference Tabuchi, Ozaki, Tamura, Yamada, Ishida, Hosoda and Hosono2003) reported that oral administration of Lb. rhamnosus GG extensively delayed the induction of type 1 diabetes in STZ-injected rats. Lb. acidophilus NCDC14 and Lb. casei NCDC19 used in present study were selected for probiotic attributes i.e. acid and bile tolerance, colonization in intestinal tract, protoplast regeneration, surface hydrophobicity, antagonistic effect against pathogenic bacteria. From 50 lactobacilli strains (un-published data) these were chosen for preparation of probiotic dahi (PD), which has shown anti-diabetic effect in type 2 diabetic rats (Yadav et al. Reference Yadav, Jain and Sinha2007a). The purpose of present study was to investigate the suppressive effect of dahi supplemented with probiotic Lb. acidophilus and Lb. casei during the progression of STZ-induced diabetes in rats in comparison with skim milk (SM) and control dahi (CD) fed groups.
Material and methods
Bacterial cultures and preparation of dahi
Lactococcus lactis ssp diacetylactis NCDC60, Lb. acidophilus NCDC14 and Lb. casei NCDC19 were obtained from National Collection of Dairy Cultures, of the National Dairy Research Institute, Karnal, India. Raw buffalo milk was procured from the experimental dairy, of the institute. Milk was adjusted to 25 g fat/l by adding fresh SM and heated at 90°C for 15 min, and cooled to 37°C, aseptically. PD was prepared by inoculating this milk with ∼107 cfu/ml Lb. acidophilus, ∼104 cfu/ml Lb. casei and ∼106 cfu/ml Lc. lactis (Yadav et al. Reference Yadav, Jain and Sinha2005). CD was prepared by inoculating with 1% Lc. Lactis (∼107 cfu/ml). The final products contained bacterial counts of ∼73×108 cfu/g. The fat, protein, lactose, total solids and ash contents were ∼0·45 to 2·5, 4·5, 3·27 to 4·18, 16 and 0·9%, respectively in milk and dahi samples, which were analysed by methods described in Association of Official Analytical Chemists (1995).
Animals and feeding schedule
Male Wistar rats of 6–8 weeks old [156±8·91 g body weight (b.w.)] were housed 3 animals per cage at 23–25°C and 12/12 h light dark cycle in small animal house of the institute. Animals were divided into five groups (n=6 in each group) viz. (1) normal group (NCG); fed with standard chow (SC) of the small animal house, (2) diabetic control group (DCG); also fed with SC, (3) SM fed group (SMG); fed with SM with drinking water (15 ml/rat/day) and SC, (4) CD fed group (CDFG), fed with CD (15 g/rat/day) along with SC and (5) PD fed group (PDFG); fed with PD (15 g/rat/day) and SC. The SC contained 61% wheat, 28% chick pea, 1% casein, 5% refined groundnut oil, 4% salt mixture and 1% vitamin mixture. The salt and vitamin mixtures were prepared according to the American Official of Analytical Chemists (1995). Diabetes was induced by a single injection (i.p.) of 45 mg/kg b.w. of STZ (freshly prepared in 0·1 m-citrate buffer, pH 6·3) to overnight food deprived animals of DCG, SMG, CDFG and PDFG on d 0 of the experiment. However, NCG animals were injected with an equal volume of citrate buffer only. Food and water intake, urine volume and b.w. were analysed at weekly intervals by keeping animals in metabolic cages. Urine sugar, albumin and ketone bodies were determined by using Benedict's reagent, sulphosalicylic acid coagulation test and Rothera's reagent, respectively. The study was approved by the National Dairy Research Institute Animal Ethics Committee, and rats were maintained in accordance with the National Institute of Nutrition, India guidelines for the care and use of laboratory animals (Saionton, Reference Saionton2002).
Oral glucose tolerance test and insulin response test
Oral glucose tolerance tests (OGTT) were performed at weekly interval throughout the experimental period. The rats were food deprived for 12 h before the administration of an oral glucose load (2 g/kg b.w.). Tail blood was drawn at 0, 30, 60, 90 and 120 min after glucose feeding. Glucose levels were determined with the help of Accu-Check Advantage Blood Glucose Monitor (Roche Group, Indiana, USA) by using one drop of whole blood and remaining blood sample was collected in heparinised (2 U/μl) vials and plasma was used for estimation of insulin levels by ELISA kit (Mercodia Diagnostics, Uppsala, Sweden). Area under the curve for glucose (AUCglucose) was determined using the trapezoidal rule.
Blood and tissues collection
At the last day of experiment, blood was collected in heparinized vials (2 U/μl) from overnight fasted animals by puncturing tail vein. Plasma was separated by centrifuging the blood samples at 5000 g for 10 min at 4°C and used for analysis of blood lipids as described below. After blood collection, animals were killed by cervical dislocation and liver, pancreas, epididymal, retroperitoneal and abdominal fat, soleus and extensor digitorum longus (EDL) muscles were excised and weighed. The liver and pancreas were chopped in small pieces and stored at −70°C till further use.
Biochemical Estimations
Plasma total-cholesterol (Tc), triglyceride (TG) and HDL-cholesterol (HDLc) were estimated using enzymatic kits procured from Bayer Diagnostics India Ltd, Gujarat, India and LDL-cholesterol (LDLc), and VLDL-cholesterol (VLDLc) were calculated by using the Friedewald's equation (Friedewald et al. Reference Friedewald, Levi and Fredrickson1972). The hepatic tissue lipids were extracted according to Folch et al. (Reference Folch, Lees and Sloane-Stanley1957) and estimated as above. Liver glycogen content was determined by the method of Vander Varies (Reference Vander Vries1954). Tissue lipid peroxidation as TBARS (Ohkawa et al. Reference Ohkawa, Ohishi and Yagi1979), total nitrite (Moshage et al. Reference Moshage, Kok, Huizenga and Jansen1995), GSH (Ellman, Reference Ellman1959), and the activities of catalase (Sinha, Reference Sinha1972), superoxide dismutase (SOD) (Marklund & Marklund, Reference Marklund and Marklund1974), glutathione peroxidase (GPx) (Rotruck et al. Reference Rotruck, Papeat, Ganther and Swanson1973) and, total protein (Lowry et al. Reference Lowry, Rosebrough, Farr and Randall1951) were determined according to the methods described previously.
Statistical analysis
Results were subjected to analysis of variance by using SPSS (SPSS Inc. Chicago) and significant differences among mean values of variables were determined by Benferroni's test. The values differed at the levels of P<0·05 were considered significant. Data were expressed as mean±standard deviation (sd).
Results
Dietary intake and physiological variables of animals
After STZ-injection, the b.w. of rats was significantly decreased during the 28 d experimental period. The rate of reduction was highest in CDG and lowest in PDFG animals among all the STZ-injected groups (Table 1). The type of diet did not significantly affect the intake of food and water, as well as excretion of urine volume. The liver and pancreatic organ weights were not statistically different in all the groups. Weights of epididymal fat were significantly lower in DCG, SMG and CDFG (P<0·05) while these measures were closer to the normal in CDFG and PDFG. Similarly, abdominal fat pad was also higher in PDFG among all the STZ-injected rats (P<0·05). No significant changes were observed in retroperitoneal fat content among all the groups. Moreover, no significant differences were observed in the weight of soleus and EDL muscles in all groups. During last week of experiment, urine sugar was significantly increased and appeared as 4+ (>2 gm/dl) in all the STZ-injected rats (Table 1), while urine protein and ketone bodies were not detected in any group of the animals.
†# Values (mean±sd of six animals in each group) were recorded as g/100 g and mg/100 g of body weight, respectively after 28 days and *were estimated during 28 day of the experimental period
a–e Values with common superscript in a row are not significantly different at the level of P value <0·05
Effect on OGTT and insulin response
The OGTT showed a significant decrease of insulin response in DCG, SMG and CDFG after 7 d (data not shown). However, feeding of PD maintained OGTT as normal on d 7 but tolerance was reduced 14 d after the STZ-injection. At the end of study (28 d), the AUCglucose values were lowest in PDFG among treated groups, followed by CDG<SMG<DCFG. In contrast, the AUCinsulin values were highest in PDFG among all the STZ-injected animals (Fig. 1).
Effects on liver glycogen content
As shown in Table 2, the liver glycogen was lowest in DCG among all the groups. The feeding of PD maintained the liver glycogen near to the normal values in STZ-injected rats. The oral administration of SM and CD also inhibited the reduction of glycogen accumulation in liver but the effect of SM was lower than CD (P<0·05).
a–d The values (mean±sd) with common superscript in a row are not significantly different (P<0·05)
Effects on blood and liver lipid content
Table 2 also depicts that the plasma Tc, TG, LDLc, VLDLc, LDLc/HDLc ratio were highest and HDLc was lowest in the DCG among all the groups after 28 d (P<0·05). However these variables were maintained near to the normal in PDFG. Similarly, liver Tc and TG also highest in DCG while maintained as normal by feeding of PD. However feeding SM or CD was moderately (or partially) effective in mitigating the effect of STZ-induced diabetes on plasma and liver lipid profiles.
Effects on oxidative stress measures
Twenty eight d after STZ-injection, the TBARS and total nitrite were drastically increased in the pancreatic tissues of DCG compared with NCG (Table 3). The feeding of PD arrested the elevation of these factors which remained significantly lower after 28 d than in other diabetic animals (P<0·05). Moreover, GSH and enzymatic activities of catalase, SOD and GPx were significantly depleted in pancreatic tissues of DCG compared with NCG (P<0·05). The oral administration of PD also arrested the depletion of GSH and enzymatic activities of catalase, SOD and GPx in STZ-injected rats, while, SM and CD have moderate activities for these effects.
a–d The values (mean±sd) with common superscript in a row are not significantly different (P<0·05)
Discussion
In present study, the effect of oral administration of dahi containing probiotic Lb. acidophilus and Lb. casei along with dahi culture on progression of STZ-induced diabetes in rats was evaluated. It was found that PD was more effective in suppression of the progression of STZ-induced diabetes and its complications such as dyslipidemia and oxidative stress compared with the SM and CD. It is well known that STZ selectively kills insulin secreting pancreatic β-cells resulting in a decrease in endogenous insulin release, which paves the ways to diminished utilization of glucose by tissues (Wilson et al. Reference Wilson, Patton, McCord, Millins and Mossman1984). In type 1 diabetes β-cells are destroyed by autoimmune reactions, also long term type 2 diabetes leads to β-cell damage by various unknown mechanisms (Robertson & Harmon, Reference Robertson and Harmon2006). In present study, the feeding of PD significantly delayed the progression of STZ-induced diabetes by suppressing the elevation of glucose intolerance and blood glucose as well as reduction of insulin levels during 28 d of the experiment. It indicates that PD may delay the STZ-induced alteration in glucose homeostasis by maintaining insulin levels which might be due to protection of pancreatic β-cell from damage. In addition, Fig. 1 indicates that the levels of insulin were not altered 14 d after STZ-injection among all groups of animals unlike the increased blood glucose levels, the reason behind these results may be the allied production of non-function insulin by STZ-exposed β-cells. PD was unable to fully suppress the induction of STZ-induced diabetes in rats after 28 d, this might be due to the chronic effect of STZ on β-cells and suppressive effect of PD was overcome in a time dependent manner in STZ-injected rats.
Insulin stimulates the glycogen synthesis in liver tissues in the presence of higher levels of glucose in blood (Klover & Mooney, Reference Klover and Mooney2004). Depleted insulin levels in STZ-induced diabetes may results in reduction of liver glycogen content that might be due to slow down of the insulin activated glycogenesis pathway. A similar trend was also observed in the present study. The oral administration of PD maintained liver glycogen content near normal; demonstrating that PD has the ability to attenuate insulin depletion sufficiently to stimulate glycogen synthesis in liver tissues till 28 d.
Feeding of PD arrested the elevation of Tc and TG and depletion of HDLc in blood as well as liver which suggests that PD feeding may delay the progression of diabetes induced atherosclerosis [a major causative agent for induction of the foremost diabetic complication (Henry, Reference Henry2001)]. Several workers also reported, the cholesterol lowering activity of milk and its fermented products containing LAB (Taylor & Williams, Reference Taylor and Williams1998). It has been proposed that various milk components such as orotic acid, pyrimidine like nucleotide, calcium or hydroxyl methyl glutaric acid attenuate the de-novo synthesis of cholesterol through inhibition of NADPH formation (a reducing power required for biosynthesis of cholesterol) HMG-CoA synthase (a rate limiting step in cholesterol biosynthesis) in liver and increase cholesterol clearance from blood stream (by enhancing excretion of bile acids) (Agrawal & Kansal, Reference Agrawal and Kansal1991). In addition, fermentation of milk with LAB increases bioavailability of these components and increased absorption in gastrointestinal tract allows increased function (Roberfroid, Reference Roberfroid2000).
Oxidative stress is a hallmark of STZ-induced β-cell damage. STZ destroys β-cells increasing production of superoxide radical, H2O2 and hydroxyl radicals (Szkudelski, Reference Szkudelski2001). The most commonly used indicator of lipid peroxidation is TBARS (Lyons, Reference Lyons1991). In the present study, TBARS levels in pancreatic tissues were drastically increased in DCG after STZ administration. However, it was significantly lower in PDFG, indicated that the formation of TBARS was suppressed by PD feeding which may be due to inhibition of free radical formation and insulin reduction. Furthermore, nitric oxide (NO) has been demonstrated to be an effector molecule responsible for pancreatic β-cells destruction. STZ liberates toxic amounts of NO by increasing NO synthase activity that inhibits the activity of aconitase and participates in DNA damage. As a result of this, β-cells undergo destruction by necrosis (Szkudelski, Reference Szkudelski2001). The oral administration of PD also diminished the elevation of total nitrite in pancreatic tissues and thus might be preventing β-cell destruction. A biodefensive antioxidant system protects tissue from free radical damage. Diabetes reduces the antioxidant capacity of tissues which increases deleterious effects of free radicals. A sulphydryl tripeptide; glutathione is known to protect the cellular system against toxic effects of lipid peroxidation (Ahmed, Reference Ahmed2005). During diabetic condition and increased oxidative stress, the GSH content decreases due to either higher utilization rate such as through Gpx activity and/or lower production (Anuradha & Selvam, Reference Anuradha and Selvam1993). Gpx is an enzyme that catalyzes the decomposition of H2O2 and other organic hydroperoxide to non-toxic products at the expense of the GSH (Bruce et al. Reference Bruce, Freeman and James1982). Its activity may be reduced by radical induced inactivation and/or glycation of enzyme (Rajasekaran et al. Reference Rajasekaran, Sivagnanam and Subramanian2005). The decreased GPx activity may allow accumulation of toxic products. In the present study, feeding of PD to STZ-injected rats significantly inhibited the decline of GSH content suggesting that it either enhanced biosynthesis of GSH or reducing the oxidative stress leading to less degradation of GSH or both. Moreover, fermented milk products containing several antioxidant sulphydryl group components may also increase the substrate pool for GSH biosynthesis (Aneja et al. Reference Aneja, Mathur, Chandan and Benerjee2002). SOD and catalase are two antioxidant enzymes which sequentially detoxify oxygen free radicals viz. SOD scavenges superoxide radical to H2O2 and molecular oxygen (McCord et al. Reference McCord, Keele and Fridovich1976) and H2O2 is further detoxified by catalase and converted into water and molecular oxygen (Chance et al. Reference Chance, Greenstein and Roughton1952). In the present study, the activities of SOD and catalase were maintained near to normal by feeding of PD, which indicates that PD diminished the formation of superoxide anions and H2O2 (substrates of these enzymes).
As observed above, the PD was additionally effective in suppression of STZ-induced consequences in carbohydrates and lipid metabolism as well as oxidative stress compared with SM and CD. This may be due to higher availability of biologically active substances in PD than CD, i.e. bioactive peptides, organic acids (un-published data), free fatty acids, conjugated linoleinic acid and complex oligosaccharides (Yadav et al. Reference Yadav, Jain and Sinha2007b, Reference Yadav, Jain and Sinhac).
Authors are grateful to the Dr GBKS Prasad, Jiwaji University, Gwalior, India for his invaluable support and guidance.