Molecular and Cellular Biochemistry

, Volume 385, Issue 1, pp 7–15

Regulation of carbohydrate metabolism by indole-3-carbinol and its metabolite 3,3′-diindolylmethane in high-fat diet-induced C57BL/6J mice


  • Jayakumar Poornima
    • Department of Biochemistry and Biotechnology, Faculty of ScienceAnnamalai University
    • Department of Biochemistry and Biotechnology, Faculty of ScienceAnnamalai University

DOI: 10.1007/s11010-013-1808-2

Cite this article as:
Poornima, J. & Mirunalini, S. Mol Cell Biochem (2014) 385: 7. doi:10.1007/s11010-013-1808-2


Indole glucosinolates, present in cruciferous vegetables have been investigated for their putative pharmacological properties. The current study was designed to analyse whether the treatment of the indole glucosinolates—indole-3-carbinol (I3C) and its metabolite 3,3′-diindolylmethane (DIM) could alter the carbohydrate metabolism in high-fat diet (HFD)-induced C57BL/6J mice. The plasma glucose, insulin, haemoglobin (Hb), glycosylated haemoglobin (HbA1c), glycogen and the activities of glycolytic enzyme (hexokinase), hepatic shunt enzyme (glucose-6-phosphate dehydrogenase), gluconeogenic enzymes (glucose-6-phosphatase and fructose-1,6-bisphosphatase) were analysed in liver and kidney of the treated and HFD mice. Histopathological examination of liver and pancreases were also carried out. The HFD mice show increased glucose, insulin and HbA1c and decreased Hb and glycogen levels. The elevated activity of glucose-6-phosphatase and fructose-1,6-bisphosphatase and subsequent decline in the activity of glucokinase and glucose-6-phosphate dehydrogenase were seen in HFD mice. Among treatment groups, the mice administered with I3C and DIM, DIM shows decreased glucose, insulin and HbA1c and increased Hb and glycogen content in liver when compared to I3C, which was comparable with the standard drug metformin. The similar result was also obtained in case of carbohydrate metabolism enzymes; treatment with DIM positively regulates carbohydrate metabolic enzymes by inducing the activity of glucokinase and glucose-6-phosphate dehydrogenase and suppressing the activity of glucose-6-phosphatase and fructose-1,6-bisphosphatase when compared to I3C, which were also supported by our histopathological observations.


GluconeogenesisInsulin resistantAntihyperglycemic effectTreatmentHexokinaseGlycogen


The incidence of diabetes in India will increase by 95 % in 2025 and the sufferers will be young age individuals [1]. The high-calorie diets that predispose to the development of the type 2 diabetes (T2DM) was the major concern in this era because the epidemiological observations provide an evidence that the development of insulin resistance and T2DM was positively related to fat consumption and negatively related to carbohydrate consumption [2]. The C57BL/6J mouse has become an important model for understanding the interplay between lipid and carbohydrate metabolism that underlie T2DM as it mimics the human T2DM with major pathological evidence of insulin resistance. Treatment of T2DM necessitates some effective oral antihyperglycemic drugs of naturally derived nutritive phytochemicals due to the side effects of existing drugs.

The carbohydrate metabolic enzymes include hexokinase, insulin-dependant predominant enzyme in glycolytic pathway which is partially responsible for the glucose homoeostasis where it increases the glycolysis and glucose utilization for energy production [3]. Glucose-6-phosphate dehydrogenase (G6PD) is the key in maintaining redox potential and cell survival via production of NADPH and pentose phosphate [4]. The two rate-limiting gluconeogenic enzymes, glucose-6-phosphatase and fructose-1,6-bisphosphatase, contribute to hepatic glucose production contributing of about 50–60 % of the glucose release and is partially responsible for elevated glucose levels in T2DM condition and under normal condition the action of insulin suppresses the superfluous gluconeogenic enzymes activation [5]. The diabetes linked with the abnormal lipid metabolism contributes to insulin resistance, in which the anomalous action of insulin correlates with the negative regulation of the above enzymes in the diabetic condition. Metformin, the standard drug is used in this study which could positively regulate the above enzymes [6].

Indole glucosinolates are a group of sulphur-containing secondary metabolites derived from tryptophan which are biologically active based on 2,3-benzopyrrole, formed during the catabolism of tryptophan, were found commonly in plants of the family Cruciferae (cabbage, broccoli, cauliflower, brussel sprouts, etc.). Indole-3-carbinol (I3C) and 3,3′-diindolylmethane (DIM) are naturally occurring plant alkaloids formed by the hydrolysis of indole glucosinolate (glucobrassicin), found in significant concentrations in cruciferous vegetables such as broccoli and brussels sprouts [7]. The amount of I3C found in the diet can vary greatly, ranging from 20 to 120 mg daily, and is dependent on dietary intake of cruciferous vegetables and their changeable concentrations [8]. Glucosinolate contents in the vegetables are tabulated in Table 1 [9].
Table 1

Glucosinolate (I3C and DIM) contents in cruciferous vegetables (Naughton et al. [9])


Glucosinolate contents (mg/100 g)













Horse radish




Brussels sprouts


Following oral administration, I3C undergoes dimerization in stomach acid. Among the several metabolites the major metabolite absorbed in the small intestine was its dimer, DIM. DIM accounts for about 10–20 % of the breakdown products of I3C; therefore, the typical daily ingestion of I3C from the diet provides between 2 and 24 mg of DIM [7]. Earlier studies showed that I3C and its metabolite DIM were suggested to be an effective drugs in anticancer therapy and it also had antimutagenic effect and antioxidant effect [10, 11]. This study was hypothesised to investigate whether either I3C or its metabolite DIM could positively regulate the carbohydrate metabolism enzymes. The study was also extended to prove whether I3C or its metabolite could exert accelerated glycaemic control over HFD-induced C57BL/6J mice by modulating those enzymes.

Materials and methods


I3C and DIM were obtained from Sigma-Aldrich (U6753; St. Louis, MO, USA). All other chemicals used in this study were of analytical grade obtained from E. Merck or HIMEDIA, Mumbai, India.

Animals and diet

Male C57BL/6J mice were purchased from the National Institute of Nutrition (NIN), Hyderabad and maintained in Central Animal House, Department of Experimental Medicine, Rajah Muthiah Medical College and Hospital, Annamalai University, and were kept in an air-conditioned room (25 ± 1 °C) with a 12-h light/dark cycle. Studies were carried out in accordance with Indian National Law on Animal Care and Use. Institutional Animal Ethics Committee of Rajah Muthiah Medical College and Hospital (Reg No. 160/1999/CPCSEA, Proposal number: 912), Annamalai University, Annamalainagar, provided ethical clearance. Two dietary regimens were feeding either normal diet or HFD ad libitum along with water, for the initial period of 10 weeks. The composition and preparation of HFD and standard diet were described by Arjunan and Sundaresan [12].

Experimental design

C57BL/6J mice test animals were fed initially, before the study, with standard diets. In the preliminary doze fixation study we fixed the doses for I3C and DIM. The optimum doze for I3C was fixed as 40 mg/kg. b.wt from 20, 40 and 80 mg/kg. b.wt. The optimum doze for DIM was fixed as 10 mg/kg. b.wt from 5, 10 and 20 mg/kg. b.wt in the doze fixation study. After induction of diabetes by high-fat diet the animals were assigned to one of seven groups with ten mice in each group:
Group I:

Control (0.5 % DMSO)

Group II:

Control + optimum dose of I3C (40 mg/kg. b.wt )

Group III:

Control + optimum dose of DIM (10 mg/kg. b.wt )

Group IV:


Group V:

HFD + optimum dose of I3C (40 mg/kg. b.wt )

Group VI:

HFD + optimum dose of DIM (10 mg/kg. b.wt)

Group VII:

HFD + Metformin (200 mg/kg. b.wt )

I3C, DIM and metformin were administered as suspensions directly into the stomach using a gastric tube in the morning for 10 weeks, by mixing with vehicle 0.5 % dimethyl sulphoxide (DMSO). Body weight and food intake of mice were measured weekly once. After 10 weeks, the animals were fasted for 12 h, anaesthetized between 8:00 and 9:00 am using ketamine chloride (24 mg/kg. b.wt, intramuscular injection), and sacrificed by cervical dislocation in the laboratory. Blood was collected by cutting the jugular vein into heparinised glass tubes. Plasma was obtained from blood samples after centrifugation (1,500×g for 10 min) and stored at 4 °C until analysis. After collecting the blood, the liver and pancreas were removed, rinsed with physiological saline and used for the various biochemical parameters and histopathological study.

Tissue homogenate preparation

Liver and kidney tissues were sliced into pieces and homogenized in Tris–HCL buffer (pH 7.0) in cold condition to give 10 % homogenate (w/v). The homogenate was centrifuged at 1,000 rpm for 10 min at 4 °C in a cold centrifuge. The supernatant was separated and used to measure carbohydrate metabolic enzyme activities.

Biochemical estimations

Glucose was estimated by the method of Trinder using a reagent kit [13]. The plasma insulin (Medgenix-INS-ELISA, Biosource, Europe S.A., Belgium) level was determined using a radioimmunoassay kit. The assay was based on the oligoclonal system in which several monoclonal antibodies directed against distinct epitopes of insulin [14]. Hb was estimated by the method of Drabkin and Austin [15]. HbA1c was estimated by the method of Sudhakar and Pattabiraman [16]. Glycogen content was determined as described by Morales et al. [17]. Glucokinase was assayed according to the method of Brandstrup et al. [18]. Glucose-6-phosphatase was assayed according to the method of Koide and Oda [19]. Fructose-1,6-bisphosphatase was assayed by the method of Gancedo and Gancedo [20]. G6PD activity was measured according to the method of Bergmeyer [21].

Histopathological examination

Liver and pancreas were excised, fixed in 10 % neutral formalin then dehydrated in graded alcohol (80–100 %), cleared in xylene and embedded in paraffin. Then the tissues were sliced into 5 μm pieces using microtome, deparaffinated in xylene, passed through 80–100 % alcohol and stained with haematoxylin and eosin (HE) for Olympus BX40 photomicroscope assessments at the resolution of 40X.

Statistical analysis

Values were given as mean ± S.D. for six readings from ten mice in each group. Data were analysed by one-way analysis of variance followed by Duncan’s Multiple Range Test (DMRT) using SPSS version 11.5 (SPSS, Chicago, IL). The limit of statistical significance was set at p ≤ 0.05.


Table 2 shows the effect of I3C and DIM on plasma glucose, insulin, Hb, HbA1c and glycogen of control and HFD mice. The levels of plasma glucose, insulin and HbA1c were found to be elevated and Hb and glycogen levels were reduced in HFD mice. Administration of I3C and DIM significantly decreased plasma glucose, insulin and HbA1c and increased the Hb and glycogen levels to near normalcy which were comparable with the standard drug metformin.
Table 2

Effect of I3C and DIM on plasma glucose, insulin, Hb, HbA1c and glycogen in HFD-fed C57BL/6J mice


Glucose (mg/dL)

Insulin (μU/mL)

Haemoglobin (g/dL)

Glycosylated haemoglobin (%)

Glycogen (mg/100 g tissue)


92.14 ± 5.86a,d

19.87 ± 1.68a

15.11 ± 0.94a,d

5.23 ± 0.40a

57.66 ± 3.81a,d

Control + I3C (40 mg/kg b.wt)

88.62 ± 7.67a,d

19.73 ± 1.45a

15.44 ± 1.02a,d

5.11 ± 0.45a

58.32 ± 3.90a

Control + DIM (10 mg/kg b.wt)

83.95 ± 7.12a

19.71 ± 1.99a

15.96 ± 1.12a

4.97 ± 0.37a

59.13 ± 3.45a


282.41 ± 24.90b

68.32 ± 6.30b

9.75 ± 0.86b

9.11 ± 0.72b

30.05 ± 2.95b

HFD + I3C (40 mg/kg b.wt)

146.71 ± 10.51c

53.99 ± 5.09c

12.67 ± 1.09c

7.10 ± 0.41c

49.31 ± 3.81c

HFD + DIM (10 mg/kg b.wt)

159.06 ± 11.73d

27.54 ± 2.55d

14.22 ± 1.21d

5.44 ± 0.43a

53.13 ± 4.02d

HFD + Metformin

109.39 ± 6.07d

24.72 ± 2.34d

14.98 ± 1.09a, d

5.37 ± 0.40a

55.26 ± 3.68d

Values were means ± SD for six samples from 10 mice in each group. Values not sharing a common superscript differ significantly at p ≤ 0.05. Duncan’s Multiple Range Test (DMRT)

Table 3 shows the effect of I3C and DIM on hexokinase and glucose-6-phosphate dehydrogenase in the liver and kidney tissues of HFD mice. Administration of I3C, DIM and metformin showed a significant increase in the levels of hexokinase and G6PD in HFD mice.
Table 3

Effect of I3C and DIM on hexokinase and glucose-6-phosphate dehydrogenase in the tissues of HFD-fed C57BL/6J mice


Hexokinase (U*/h/mg protein)

Glucose-6-phosphate dehydrogenase (U^/mg protein)






0.33 ± 0.02a

0.28 ± 0.01a

0.77 ± 0.06b

1.37 ± 0.10a

Control + I3C (40 mg/kg b.wt)

0.32 ± 0.02a

0.30 ± 0.02a

0.78 ± 0.05a

1.39 ± 0.07a

Control + DIM (10 mg/kg b.wt)

0.29 ± 0.02a

0.31 ± 0.02a

0.80 ± 0.06a

1.40 ± 0.09a


0.09 ± 0.008b

0.11 ± 0.009b

0.39 ± 0.01b

0.56 ± 0.03b

HFD + I3C (40 mg/kg b.wt)

0.18 ± 0.01c

0.19 ± 0.01c

0.49 ± 0.02c

0.75 ± 0.05c

HFD + DIM (10 mg/kg b.wt)

0.25 ± 0.02d

0.24 ± 0.02d

0.68 ± 0.03d

1.19 ± 0.08d

HFD + Metformin

0.29 ± 0.02d

0.27 ± 0.02a, d

0.70 ± 0.04e

1.23 ± 0.09d

Values were means ± SD for six samples from ten mice in each group. Values not sharing a common superscript differ significantly at p ≤ 0.05. Duncan’s multiple range test (DMRT)

U* µmol of glucose phosphorlated/hr

U^ nmol of NADPH formed/min

Table 4 shows the effect of I3C and DIM on the activities of gluconeogenic enzyme, glucose-6-phosphatase and fructose-1,6-bisphosphatase in the tissues (liver and kidney) of control and HFD mice. The activities of glucose-6-phosphatase and fructose-1,6-bisphosphatase were increased significantly in HFD mice; in contrast the treated group shows decreased activities of the above enzymes.
Table 4

Effect of I3C and DIM on glucose-6-phosphatase and fructose-1,6-bisphosphatase in the tissues of HFD-fed C57BL/6J mice


Glucose-6-phosphatase (U**/min/mg protein)

Fructose-1,6-bisphosphatase (U^^/h/mg protein)






0.19 ± 0.01a

0.20 ± 0.01a

0.39 ± 0.03a,d

0.75 ± 0.07a,d

Control + I3C (40 mg/kg b.wt)

0.17 ± 0.01a

0.18 ± 0.01a

0.37 ± 0.02a

0.73 ± 0.07a,d

Control +DIM (10 mg/kg b.wt)

0.15 ± 0.01a

0.16 ± 0.01a

0.34 ± 0.02a

0.69 ± 0.07a


0.42 ± 0.04b

0.47 ± 0.04b

0.78 ± 0.06b

1.47 ± 0.10b

HFD + I3C (40 mg/kg b.wt)

0.30 ± 0.03c

0.39 ± 0.03c

0.60 ± 0.06c

1.15 ± 0.09c

HFD + DIM (10 mg/kg b.wt)

0.21 ± 0.09d

0.28 ± 0.01d

0.45 ± 0.03d

0.88 ± 0.05d

HFD + Metformin

0.22 ± 0.02d

0.24 ± 0.02d

0.42 ± 0.03e

0.81 ± 0.08d

Values were means ± SD for six samples from ten mice in each group. Values not sharing a common superscript differ significantly at p ≤ 0.05. Duncan’s multiple range test (DMRT)

U** µmol of Pi liberated/min

U^^ µmol of Pi liberated/hr

Figure 1 shows the histology of liver at 40X (H&E staining). Control shows normal hepatocytes with central portal vein; in contrast HFD mice show hepatic necrosis with micro- and macro-vesicular stenosis whereas treatment group shows the normal hepatic cell with reduced hepatic stenosis.
Fig. 1

Histopathological changes in liver (H&E, 40X). a Control: Normal morphological features of hepatocytes with central portal vein. b Control+ I3C and c Control+ DIM: Normal hepatic cells. d HFD: Diabetic control mice shows diffused cytoplasmic vacuolization, hepatic necrosis with micro- and macro-vesicular stenosis, shown in arrows (S). e HFD+ I3C: Shows moderate steatosis and microparticulate lipid droplets, shown in arrows (M). f HFD + DIM: Improved hepatocyte architecture with reduction in hepatic stenosis. g HFD+ Metformin: Shows reduced hepatic stenosis and regaining of normal hepatic structure

Figure 2 shows the histology of pancreas at 40X (H&E staining). HFD mice show hypertrophied islets cells with fatty infiltration; and treatment with I3C, DIM and metformin shows reduced lipid accumulation in the acinar cells and regaining normal islet cell structure.
Fig. 2

Histopathological changes in pancreas (H&E, 40X). a Control: Typical histological structure of mice pancreas with normal islet. b. Control+ I3C & c. Control+ DIM: Normal islet cells and acinar pancreas architecture. d. HFD: Hypertrophied islets of Langerhans (marked as H) with intracellular lipid accumulation (marked as L). e. HFD+ I3C: Reduction in size of islet cells with mild fatty infiltrations. f. HFD+ DIM. g. HFD+ Metformin: Near-normal pancreatic architecture with islet cells


The carbohydrate metabolic enzymes were potentially excellent targets in the treatment of T2DM; hence these enzymes were monitored to control the levels of blood glucose. In the present study the blood glucose, insulin and HbA1c levels were found to be increased and the haemoglobin and glycogen levels were decreased in HFD mice. Insulin was the major hormone that regulates the glucose transport by stimulating the sugar uptake from circulation to muscles and adipose tissue [22]. During defective insulin receptor (insulin-resistant) condition the translocation of GLUT4 (member of glucose transporter family) was blocked which in turn leads to excessive accumulation of glucose in the blood of HFD mice, which was monitored in our study and was in agreement with the earlier reports [23]. In the treatment condition the efficacy of DIM in reducing the blood glucose level was much higher than the I3C which was comparable with the standard drug metformin, suggest that our drugs could modulate an effective GLUT4 translocation thereby reduce the blood glucose and further utilize it for oxidation to provide energy to the cells.

HbA1c was a standard biochemical marker for the diabetes as elevated HbA1c levels reflects the high glucose content in the blood which in turn leads to glycosylation of amino groups of lysine residue in protein due to the direct reaction between sugar and aminogroups in protein [24]. In our study the mice administered with I3C and DIM decreased HbA1c levels which further confirm the reduced blood glucose levels in HFD mice.

In our study the suppressed glycogen content in the liver of HFD mice was due to the defective action of insulin which in turn deregulates glycogenesis in liver or due to oxidative stress that developed during diabetic condition which inactivates glycogen synthase may also partially be responsible for the reduced glycogen levels, in accord of the another study [25]. During treatment condition the DIM restores the level of glycogen accumulation in the liver similar to that of the standard drug to near normalcy when compared to I3C, which indicates that our drug could regulate anabolism of glycogen.

In the present study, it was observed that the hexokinase, which was the predominant enzyme in the glycolytic pathway, was suppressed in HFD condition in the liver and kidney tissues, this inhibition or inactivation of this enzyme may be due to the altered action of insulin or due to the defective insulin receptor action which could resist the further energy production, where glucose acts as a substrate [26]. Administration of I3C and DIM in HFD mice increased the level of hexokinase in the liver and kidney tissues thereby it could bypass its action towards increased glycolysis and glucose utilization. The principal intracellular reductant was NADPH whose production was mainly dependent on G6PD activity. The possible reason for reduced G6PD activity in HFD condition in the tissues (liver and kidney) will be due to an increased oxidative stress which was a hallmark for the diabetic complication caused by the activation of protein kinase A and inhibition of G6PD further driven by the decreased availability of NADPH. It was evidenced that there was a reduced G6PD in diabetic nephropathic conditions [27].

The levels of hepatic and renal gluconeogenic enzymes were found to be suppressed in treatment condition as the inhibition of gluconeogenesis (suppression of glucose-6-phosphatase and fructose-1,6-bisphosphatase expression) contributes to glycemic control in the HFD mice. The earlier studies suggest that increased activity of glucose-6-phosphatase in diabetic condition provides hydrogen which binds with NADP+ in the form of NADPH and enhances synthesis of fat (lipogenesis) from carbohydrate which finally contributes to augmented blood glucose [28]. The inactivation of glucose-6-phosphatase was followed by the treatment of I3C and DIM; hence in the treatment condition these drugs could defend against the lipogensis and thereby they could impart some antihyperglycemic effect. Another gluconeogenic enzyme was fructose-1,6-bisphosphatase, which catalyzes the rate-limiting step in glycolysis which was responsible for the increased glycolytic flux in the diabetic condition. Under normal condition, insulin functions as a suppressor for the above enzyme hence the defective insulin action will lead to amplified production of fructose-1,6-bisphosphatase which was monitored in our study and was in agreement with the earlier studies in HFD mice [29]. Our result demonstrates that the administration of I3C and DIM could alter hepatic and renal fructose-1,6-bisphosphatase to some extent as it could overcome gluconeogenesis. Also the ability of DIM to suppress the gluconeogenic enzymes in liver and kidney was found to be more effective when compared to I3C which was comparable with the standard drug; hence DIM could act as better insulin sensitizers rather than I3C. The possible carbohydrate mechanism by the action of I3C and DIM is shown in Fig. 3. Metformin was selected for evaluating antihyperglycemic activity in this study, because it was documented as a well-proven antidiabetes drug [6].
Fig. 3

Possible carbohydrate mechanism regulated by I3C and DIM in C57BL/6J mice. I3C and DIM administrated in HFD mice significantly increased the activities of liver hexokinase, which is the first rate-limiting enzyme in the glyolytic pathway and it regulates hepatic shunt enzyme (glucose-6-phosphate dehydrogenase) which is responsible to overcome the oxidative stress. Another point is it decreased gluconeogenic enzymes (glucose-6-phosphatase and fructose-1,6-bisphosphatase) and thereby reduces the hepatic glucose production. Finally the cellular mechanism leads to increased transport of glucose into the cells; thereby it maintains normal blood glucose homoeostasis

In a nutshell this study show that the antihyperglycemic effect of indole glucosinolate—DIM was found to be more potent when compared to I3C; the possible explanation may be that in the acid environment of the stomach, I3C undergoes hydrolysis to a number of products, including a dimeric product—DIM which was its major active metabolite. DIM was acid stable and was detected in the bloodstream after oral intake of I3C which was explored from the earlier studies [30]. Even though the I3C could possess antihyperglycemic activity it cannot act as an effective antihyperglycemic agent due to its instability. Another reason could be that the I3C contains single nitrogen atom in its structure and DIM contains two nitrogens, thus it could effectively interact with the proteins (enzymes) to form stable hydrogen bonds for the interactions.


From this study we can warrant the conclusion that the ability of DIM to modulate the carbohydrate metabolism was more potent when compared to I3C. Therefore, in this emerging era DIM can be investigated more to reveal its antihyperglycemic action by tracing out the signalling pathways in which it could be used for further clinical trials to overcome diabetic epidemic.

Copyright information

© Springer Science+Business Media New York 2013