Remember me
RESEARCH ARTICLE
Year : 2020 | Volume
: 52
| Issue : 6 | Page : 495-504
Evaluation of the possibility of selective modulation of retinal glucose transporters in diabetic complications: An experimental study
Kanuj Mishra1, Madhu Nath2, Nabanita Halder2, Thirumurthy Velpandian2
1 Department of Biotechnology, All India Institute of Medical Sciences, New Delhi, India
2 Dr. Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi, India
Correspondence Address:
Prof. Thirumurthy Velpandian
Dr. Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences, New Delhi - 110 029
India
Source of Support: None, Conflict of Interest: None
CheckDOI: 10.4103/ijp.IJP_403_17
PURPOSE: To identify the possibility of modulating retinal glucose transporters in diabetic conditions to prevent retinal complications of diabetic retinopathy.
MATERIALS AND METHODS: In silico and in vitro binding assays were performed to assess the effect of genistein and positive controls (pioglitazone and estradiol) on nuclear receptor estrogen receptor beta and peroxisome proliferator-activated receptor gamma (PPARγ). In vivo effects of compounds were tested on diabetic rats. Structural and functional analysis of retina was performed at 28th day followed by gene expression analysis of glucose transporters and nuclear receptors. Pioglitazone and genistein levels were analyzed by liquid chromatography with tandem mass spectrometry.
RESULTS: Genistein showed equi-affinity toward PPARγ in in silico experiments contrary to in vitro findings. In multidose study, their therapeutic effects were observed by analyzing the retinal function. Retinal gene expression studies revealed that both test agents significantly up regulated PPARγ, GLUT4, and down regulated GLUT1. Genistein showed significant up regulation of GLUT4 and down regulation of GLUT1 as compared to PGZ which has been well correlated with the Electroretinography (ERG) outcome.
CONCLUSION: This study showed the possibility of selective upregulation of GLUT4 (independent of PPARγ activation) in the retina of diabetic rats using genistein. Selective modulation of retinal glucose transporters as therapeutic target in ocular diabetic complications can be possibly explored.
Keywords: Estrogen receptor beta, genistein, peroxisome proliferator-activated receptor gamma, Slc2a1 (GLUT-1), Slc2a4 (GLUT-4)
Diabetes presents a multi-faceted challenge to health system, which eventually results in the macro- and micro-health complications. Hyperglycemia, which is one of the hallmarks of the diabetes, results in various micro-complications such as nephropathy, neuropathy, and retinopathy. Glucose transport from blood to tissues is accomplished by GLUT family of transporters. GLUT1 (Slc2a1) is ubiquitously present transporter, which in conjunction with tissue specific other GLUT transporters, enables the entry of glucose. However, in terms of blood retinal barrier, GLUT1 is exclusively responsible for the transportation of glucose inside the retinal cells.[1],[2] Majority of the cells in body, regulate the entry of glucose using insulin sensitive GLUT receptors, although, cells in the tissues having high metabolic demand like retina, are more susceptible for hyperglycemia insult, as it mainly contains insulin independent GLUT receptors. Another equally important member of GLUT transporter is the GLUT4, which is an insulin sensitive transporter and mostly accountable for the rate-limiting glucose intake into the heart, fat, and skeletal muscle.[3]
Peroxisome proliferator-activated receptor-γ (PPARγ) is a ligand-dependent transcription factor which is one the subtype of nuclear hormone receptor super family of PPARs. It is a diabetic retinopathy (DR) contender receptor, as it plays an important role in regulating many biological process dysregulated in DR such as adipogenesis, glucose metabolism, angiogenesis, and inflammation.[4] Interestingly, significant down-regulation of PPARγ receptors has been observed in several experimental studies and this alteration also reported to be involved in the pathophysiological features of DR such as neurodegeneration and microangiopathy.[5] Compounds, like thiazolidinediones such as pioglitazone and troglitazone, show even higher affinity to activate PPARγ than its own natural ligands.
Estrogen receptor (ER) is also a member of nuclear receptor. The signaling of estrogens is mediated through two ERs, ERα, and ER-β. Like PPARγ, ER-β has also been documented for playing important role in various pathological diseases such as age-related macular degeneration,[6] multiple sclerosis,[7] diabetes, and influencing GLUT transporters.[8],[9],[10]
Genistein a bioactive flavonoid compound has been widely recognized for its promiscuity when comes to its action mechanism. It is a known positive activator of PPARγ[9] and also an inhibitor of GLUT1 glucose transporter.[10]
As the activation of PPARγ and ER-β is expected to change the expression of glucose transporters, this study was conducted to evaluate the genistein-induced modulation of glucose transport mechanisms in the retina in the streptozotocin (STZ)-induced early diabetes in rats.
Materials and MethodsDrugs and chemicals
Pioglitazone and STZ were procured from Sigma Aldrich (USA), genistein was gifted by Xi'an BoSheng Biological Technology Co. Ltd., China. Insulin (Humulin70/30), ketamine and xylazine were procured (C. B Pharma, India). Sodium citrate dihydrate, citric acid, D-Glucose, and gum acacia were procured from Merck, Germany. Accu Check Active (USA) blood glucometer was used for the quantification of blood glucose and was recalibrated for rat blood. Other chemicals and solvents were of the highest analytical grade, obtained from their respective commercial sources. Primers were purchased from Oligo IDT Technologies, USA. RNAlater was obtained from Ambion, USA.
Modeling and docking studies
The three-dimensional structure modeling and docking were performed for PPARγ and ER-β with genistein and their respective positive controls, i.e., pioglitazone and estradiol respectively using Cache software version 6.1.12 (Fujitsu, Japan). All calculations were performed on Window XP professional version 2002, equipped with 2 processor Intel (R) Core™ 2 Duo with 208GhHz and 2GB RAM. The standard Cache ver. 6.1.12 of Fujitsu, Japan includes the docking procedure in the embedded in the “Fast Dock compute engine.” The purpose of molecular docking of genistein in the active site of PPARγ and ER-β is to prove its promiscuous nature in activating different molecules for the betterment of the various dysregulated biological processes in DR. The docking procedure was optimized for the PPARγ and ER-β. The method was used to individually dock and score genistein as a flexible ligand with rigid receptor mode. The chemical sample files (with CSF extensions) were stored individually with the final docked ligand geometry and other information.
In vitro nuclear receptor ligand binding assays
A sensitive fluorescent polarization-based method was used for analyzing the binding affinities of PPARγ and ER-β ligands with respective positive controls. The assay was conducted according to the manufacturer's instructions given in the PPARγ and ER-β ligand screening kits, respectively (Cayman Chemical, Sydney, Australia, item no. 10007685 and Invitrogen catalog no. P2615). In brief, PPARγ and ER-β both are bound with a small fluorescent probe which in combination represents a large fluorescence molecule having high degree of fluorescent polarization. The addition of a small amount of PPARγ and ER-β high affinity ligands will displace the PPARγ and ER-β from the small fluorescent probes and result in drop in the fluorescent polarization. This polarization is quantified as milli-polarisation (mP) and plotting this mP versus ligand concentration allows the construction of IC50 curve with a broad dynamic range.
Experimental diabetes model
Female Wistar rats weighing 150 g-200 g were used in the study and procured from Central Animal Facility. The study was approved by the Institutional Standing Animal Ethics Committee (File No. 751/IAEC/13). Diabetes was elicited at zero day by intraperitoneal injection of STZ in overnight fasted rats (n = 5 in each group) at dosage of 45 mg/kg in cold 0.1M citrate buffer (pH-4.5) intraperitoneally.[11] The diabetic state was established by the measurement of fasting blood glucose levels after 2nd, 15th, and 21st day. The animals exhibiting blood sugar level more than 250 mg/dl were considered diabetic and included in the study.
Effect of single-dose genistein on blood glucose level in diabetic rats
The 15th day diabetic rats were divided into four groups (n = 5) to evaluate the effect of single-dose genistein on blood sugar levels at various time points in comparison to the positive control (pioglitazone). The overnight fasted rats were orally administered with genistein at the dose of 30 and 60 mg/kg in gum acacia by oral gavage. Control animals received only vehicle (gum acacia of the same quantity) and pioglitazone administered rats served as a positive control at the dose of 3.5 mg/kg in gum acacia. The quantification of blood glucose levels was done using glucometer at 0, 2, 4, 6, and 8 h post dosing.
Pharmacokinetic study of single and multidose treatment of genistein and ocular penetration
Twenty days following the diabetes induction, rats were subjected for single and multidose pharmacokinetic analysis. On the 21st day, the first dose study was conducted in overnight fasted rats (n = 4) by administering genistein suspension (60 mg/kg body weight) in gum acacia (as oral gavage). Subsequently, multidose kinetic study was conducted in the same rats by administering genistein twice daily for 7 days. At the end of first dose and last dose, whole blood samples amounting to 20 μl were collected in 0.2 ml EDTA vials from the tail vein at 0, 2, 4, 6, 8, 10, and 12 h for genistein quantification. Briefly, for plasma extraction, the whole blood was centrifuges at 1800 g for 15 min using a refrigerated centrifuge and plasma (10–12 ul in volume) was carefully aspirated and transferred to the cold 0.2 ml vials. Moreover, after collecting the last blood sample at 12 h after the multiple dosing, the animals were sacrificed using carbon dioxide euthanasia for the isolation of vitreous and retina choroid. All samples were stored at − 20°C until analysis of genistein levels by liquid chromatography coupled tandem mass spectroscopy. The genistein levels were subjected to pharmacokinetic curve fitting as shown in the using two compartmental model in WinNonlin version 5.1 software. The two compartment model was used to analyze the pharmacokinetic of vitreous and retina choroid, as being the closed cavity, eye is guarded by the ocular barriers. These unique exceptions make the distribution of drug slower in the eye and make it a peripheral second compartment.
Ocular penetration studies of pioglitazone
Overnight fasted rats were administered with pioglitazone at the dose of 3.5 mg/kg and they were sacrificed at 2 h using carbon-di-oxide euthanasia (n = 4). Blood obtained by cardiac puncture, vitreous and retina were harvested immediately after the death. Samples were processed for the quantification using liquid chromatography with tandem mass spectrometry (LC-MS/MS) .
Genistein and pioglitazone levels quantification using liquid chromatography with tandem mass spectrometry
Levels of genistein in vitreous, retina, and blood were quantified using liquid chromatography (Surveyor, Thermo, USA) coupled with tandem mass spectroscopy (4000QTrap, Absciex, USA). Briefly for genistein quantification, isocratic method for separation was achieved using C18 Purospher star column (55 mm × 4 mm, 3.5 μm, Merck, Germany) using a 50% organic gradient elution with water (0.1% formic acid) and acetonitrile (0.3% formic acid) at flow rate of 500 μl/min, where probenecid was used as the internal standard. Analytes were quantified in multiple reaction monitoring mode and transitions 269/133 and 283.9/240 were used for genistein and probenecid, respectively. The lower limit of quantification (LLOQ) for genistein was 62.5 ng/ml with accuracy of 96.57% and %CV of 2.32.
For pioglitazone quantification also isocratic method of separation was used. The separation of pioglitazone and internal standard sulfadimethoxine (SDM) was achieved using C18 Purospher star column (55 × 4 mm, 3.5 μm, Merck, Germany) following a 50% organic gradient elution with water (0.1% formic acid) and methanol (0.1% formic acid) at flow rate of 500 μl/min. The total runtime was 4 min. Analytes were quantified in multiple reactions monitoring mode and transitions 357.1/134.1 and 357.1/119.1 were used for pioglitazone. For internal standard 311.0/156 transition was used. The LLOQ for developed method of pioglitazone was 23.43 ng/ml with accuracy of 139.12% and %CV of 139.72.
Biological samples (blood, plasma, and tissue samples) of measured quantity were added with extraction solvent for direct deproteinization (90% acetonitrile with 0.1% formic acid and internal standard). It was subjected for vortexing and centrifugation, 10 μl of supernatant was subjected for quantification using the aforesaid method by LC-MS/MS. Representative chromatograms are shown in [Figure 1].
Figure 1: Representative LC-MS/MS chromatograms of Pioglitazone (upper panel) and Genistein (lower panel) in the blank, standard and vitreous humorEffect of multi dose genistein and pioglitazone on retinal functions of diabetic rats
For this study, three groups of 21 days' diabetic rats (n = 5 each) were compared with normal rats to evaluate the protective effect of genistein on retinal functions and the expressions of Esr2, GLUT1, GLUT4, and Pparg in retina and to compare it with positive control (pioglitazone). Group 1 having normal (No diabetes induction) rats, Group 2 diabetic rats (untreated) received only vehicle (gum acacia) twice a day orally and Group 3 received pioglitazone at the dose of 3.5 mg/kg in gum acacia twice a day orally served as positive control. Experimental Group 4 received genistein at the dose of 60 mg/kg in gum acacia twice a day orally. This treatment continued for 7 days and at the end of 28th day after the induction of diabetes, retina functional assessment was done through fundus imaging and ERG, later retina was harvested and used for messenger RNA (mRNA) expression study.
Functional assessment of retina
At the 28th day post diabetic induction, ERG measurement was taken according to ISCEV guideline and by method, as described by Vetrivel et al.[12] Briefly, dark adapted animals were anesthetized using ketamine (75 mg/kg body weight) and xylazine (5 mg/kg body weight). The pupil of eye was dilated with tropicamide ophthalmic solution. Upon complete anaesthetization, animals were positioned on a platform containing heating pad to maintain body temperature. The reference electrode was placed on the forehead; ground electrode was inserted at the base of tail, and corneal electrode was firmly in contact with cornea of animal. White light intensity of 30 cds/m2 with 6 ms flash duration was used to obtain 25 responses, and average of all responses was considered as a single ERG response from the retina of each animal. MICRON III, inbuilt software (Labscribe software) was used to calculate the “a” and “b” wave amplitude and latency. Retinal images were taken using MICRON III rodent imaging system (phoenix lab, USA). The retinal images were then evaluated for tortuosity index of arteries and veins using Image J software (available at http://rsb.info.nih.gov/nih-image developed by Wayne Rasband, NIH, Bethesda, MD) as per the method described by Vetrivel et al.[12]
Expression analysis by quantitative real-time polymerase chain reaction
For this study, three groups of 28 days diabetic rats (n = 5 each) were compared with normal healthy rats (Group 1, n = 5) after 7 day treatment of vehicle in Group 1 diabetic rats, pioglitazone (3.5 mg/kg) in Group 3 diabetic rats and genistein (60 mg/kg) in group 4 diabetic rats. Rats were sacrificed humanely using excess of pentobarbitone after the functional and structural assessment of retina and kidney (positive control) and pooled retinal tissues were immediately processed for the gene expression analysis through quantitative real-time polymerase chain reaction (qRT-PCR). Housekeeping genes (18s and Ywhaz) were taken as reference genes while normal tissue served as control. The fold expression of PPARg, GLUT4, and GLUT1 was done using inbuilt BioRad software of CFX96. 2–delta delta CT (Livak) Method was used to analyze relative gene expression. The primers sequences for: 1) PPARg: 5'-AGATGACAGACCTCAGGCAGA-3'and Reverse 5'-TCCTGGAGCAGAGGGTGAAG-3',2) GLUT4: Forward, 5'-CTCCAACTGGACGAGCAAC-3' and Reverse, 5'-CAGCAGGAGGACCGCAAATA-3',3) GLUT1: Forward, 5'-CATTGGTCTGGCTGGCATGG-3' and Reverse, 5'-GGCCACGATACTCAGATAGGAC-3' were used for the analysis.
Statistical analysis
The experimental data are expressed as mean ± standard error of the mean appropriate statistical methods were employed using graph pad prism and P < 0.05 has been considered as statistically significant. Statistical analysis was done using one way ANOVA, two-way ANOVA, and Student's t-test. Gene expression data were statistically analyzed using REST 2009 software.[13] For determination of EC50, the value of genistein and pioglitazone in in vitro nuclear receptor ligand binding assay, nonlinear regression of concentration curve was prepared using GraphPad Prism version 7.0 (GraphPad Software Inc, San Diego, California USA).
ResultsDocking simulation
The docking simulation and scoring were performed on genistein using a genetic algorithm which creates random populations based on random positions, orientations, and confirmations of the ligand (genistein) against the fixed binding pocket of the PPARγ and ER-β to a maximum of 40,000 generations. The affinity of genistein toward the PPARγ and ER-β receptors was classified on the basis of the potential mean force values which evaluates pairwise atomic potentials from the structural information of protein-ligand bound complex. Compared to the positive ligands of pioglitazone and estradiol on PPARγ and ER-β, respectively, genistein affinity was found to have 114.7% and 105.7% higher binding affinity, respectively.
In vitro nuclear receptor ligand binding assay
In in vitro nuclear receptor ligand binding assays, EC50 values of genistein and pioglitazone on PPARγ was found to be 23.9 μM and 0.22 μM, respectively, whereas the EC50 values of genistein and estradiol on ER-β were found to be 18.4 pM and 3.3 pM, respectively [Figure 2]a and [Figure 2]b.
Figure 2: Fluorescence polarization-based ligand screen assay was done to analyze the affinity of genistein for peroxisome proliferator-activated receptor gamma and estrogen receptor beta and EC50 values for the genistein and the respective positive controls were measure by plotting a graph between concentration (pM) and milli-polarization unit (mP), (a) demonstrating the genistein and pioglitazone affinity against peroxisome proliferator-activated receptor gamma whereas, (b) represents affinity of genistein and estradiol against estrogen receptor beta, (c) shows the blood glucose levels of diabetic rats at different time interval in the whole study, (d) represents the percentage elevation in blood glucose level in diabetic untreated, pioglitazone 3.5 mg/kg treated, Genistein 30 mg/kg and 60 mg/kg treated groups (treatment vs. diabetic control P < 0.001, ANOVA) (e) was obtained from pharmacokinetic analysis of genistein, plasma concentration versus time curve observed (•) and software predicted (−), plasma concentration over the 7 days of experimental time points was plottedEffect of single-dose genistein on blood glucose level in diabetic rats
[Figure 2]c demonstrates the sustained hyperglycemic state alluding to the diabetic condition of experimental animals found during the experimental period at the time points 15th and 21st day. Initial studies in diabetic rats were found to be suboptimal effect of genistein at 30 mg/kg BW.[14],[15],[16] This formed the basis for the two dosages of 30 and 60 mg/kg BW given in our study to the experimental groups of genistein treatment against the positive control pioglitazone (3.5 mg/kg BW). Graph was plotted between the percentage elevation of blood glucose level at different interval of time for the respective drug treatment and diabetic control groups as shown in [Figure 2]d. The drug concentrations of both the positive control and test drug exhibited significant anti-hyperglycemia at the different time points. The anti-hyperglycemic activity of genistein at the dosage of 60 mg/kg exhibited significant lowering of the blood glucose elevation, extrapolating its anti-diabetic property comparable to pioglitazone and on this basis the 60 mg/kg dose was selected for further analysis.
Pharmacokinetic analysis and ocular penetration of genistein
The pharmacokinetic simulation of the genistein blood levels generated using the first dose and after multiple dosing (60 mg/kg twice daily) in diabetic rats is shown in [Figure 2]e. The curve fitting was achieved with the correlation coefficient of 0.949 between observed and predicted values using the two compartment model. The Cmax of the first dose 1.7 μg/ml increased to 1.97 μg/ml on the 7th day after multiple dosing. However, this increase was found not statistically significant and thus indicates the lack of any drastic increase in steady state levels after multiple dosing during the study. The AUC was found to be 107.47 h*μg/ml after multiple dosing. Genistein levels in vitreous and retina choroid were found to be 0.19 μM (50.7 ng/ml) and 0.37 μM (100.8 ng/ml) respectively after its multidose administration for 7 days.
Functional assessment of retina
The representative fundus images of the experimental groups are shown in [Figure 3]. Retinal vascular changes (early symptoms of DR) in terms of tortuosity index were higher in diabetic animals in comparison to normal control, in which it was statistically significant only in arterioles. However, this retinal vascular changes in the pioglitazone and genistein treated groups were found to be insignificant either with normal or with diabetic controls [Figure 4]a and [Figure 4]b. A significant increase in “b” and “a” wave amplitude was observed in early diabetic stage which was found to be statistically reduced in drug treatment groups [Figure 4]c and [Figure 4]d. While a significant increase in latency time was observed in drug-treated groups in comparison to normal as well as untreated diabetic group [Figure 4]e and [Figure 4]f.
Figure 3: Representative fundus images of experimental groups. (a) Normal group (untreated), (b) Diabetic rats, (c) Pioglitazone (3.5 mg/kg)-treated diabetic group, (d) Genistein (60 mg/kg)-treated diabetic group
Figure 4: Tortuosity Index. (a) Tortuosity index of arterioles, (b) Tortuosity index of venules (no significance difference among groups was found). ERG parameters, (c) a wave amplitude in experimental groups, (d) b wave amplitude in experimental groups, (e) b wave implicit time in experimental groups (no significance among groups was found), (f) a wave latency in experimental groups. Unpaired t-test was used for statistical comparisonOcular penetration of pioglitazone
The concentration of pioglitazone in plasma, vitreous, and retina was quantified at 2 h after administration of 3.5 mg/kg orally using LC-MS/MS. The concentration of pioglitazone in plasma was found to be 6.15 μg/ml or 17.2 μM, in retina, it was 0.371 μg/ml or 1.041 μM and it was found to be least in vitreous, i.e., 0.180 μg/ml or 0.51 μM.
Messenger RNA expression analysis
qRT-PCR analysis of the isolated retina samples in the diabetic group showed upregulation of GLUT1 and down-regulation of PPARg and GLUT4. Significant up-regulation of PPARg and GLUT4 and downregulation of GLUT1 was seen in the genistein treatment group and also in pioglitazone-treated test tissues [Figure 5]. No significant change in Esr2 expression was observed (data not shown).
Figure 5: mRNA expression levels of peroxisome proliferator-activated receptor gamma, GLUT-1 (slc2a1) and GLUT-4 (slc2a4) in retina of all the four experimental groups (normal, diabetic vehicle administered, pioglitazone 3.5 mg/kg treated group, and genistein 60 mg/kg treated group) were analyzed after the respective treatments where (a) represents peroxisome proliferator-activated receptor gamma expression in experimental groups, (b) represents GLUT-1 (slc2a1) expression in experimental groups and (c) demonstrates GLUT-4 (slc2a4) expression in experimental groups (Statistical analysis by REST software (Biorad) Discussion The ever-increasing global prevalence of diabetes alone accounts for 7% all-cause mortality and health-care cost burden of 12%, worldwide. The present drug regimens and therapeutic modalities rely on sensitizing the insulin receptors in tissues for glucose uptake in type II diabetes using various agents such as biguanides, glucagon-like peptide-1 receptor agonists, thiazolidinediones and the dipeptidyl peptidase IV inhibitors. Despite of good blood sugar level control, vulnerability of retina to microvascular complications at an early stage of diabetes is a matter of concern.[17] Therefore, this study was undertaken to identify the possibility of developing retina-specific agents to selectively regulate glucose transport in diabetic conditions to prevent micro vascular complications.
GLUT1 isoform of glucose transporter is reported to be insulin insensitive and present at high density in the membrane of human erythrocytes, blood brain, blood placental, and blood ocular barriers.[1] In contrast GLUT4 is reported to be insulin sensitive and present at fat, skeletal muscle, and myocardial tissues and reported to be involved in insulin resistance in type-II diabetes.[18]
Ontogenetically conserved Glut 1 has been recognized as a main glucose transporter of the human eye reported to meet the glucose demand of the high-energy consuming photoreceptors by the constant influx of glucose. Glut 1 is reported to be expressed by the epithelial cells which form the blood-ocular barrier, in outer segments of the photoreceptors, and in the retinal cell of the adult eye.[2] Retina of 28-week-old diabetic and galactosemic rats showed significantly elevated levels of glucose transporter protein mRNA.[19] However, studies have also showed the significant decrease in levels of GLUT1 protein during chronic hyperglycemia without much change in GLUT1 mRNA, where ubiquitinylation of GLUT1 has been speculated to be a possible mechanism of GLUT1 degradation in diabetes.[20] Reduction of GLUT1 transporter in diabetes has been well documented[1],[21] which leads to reduced glucose intake in retina causing decreased reactivity of microglia.[22]
Anti-diabetic drug like pioglitazone belongs to the class of TZD exhibits its action selectively through PPARγ that regulates the transcription of number of insulin responsive genes.[23] Interestingly, multi-targeted nature of the flavonoids have provided a reasonable opportunity to fight the diabetic complications in various in vivo studies.[24] In diabetic conditions, therapeutic up-regulation of GLUT4 represents one of the promising targets for controlling hyperglycemia. GLUT4 transporter has been reported to be activated through two independent pathways involving ER-β and PPARγ.[8],[21] Individual studies have documented genistein to be a ligand for ER-β and PPARγ both.[25],[26] Hence, in the present study a preliminary in silico docking experiment was employed to evaluate the interaction of genistein against PPARγ and ER-β. The high docking score exhibited by genistein against both of the targets justified its multi target binding nature.
In our in vitro receptor binding assays, genistein showed higher binding affinity towards ER-β as reported in the literature.[27] This has explained the potential benefits of genistein through its agonism on ER-β in metabolic disorders, chronic inflammatory diseases, and also in cancer.[28] However, the present study genistein affinity on PPARγ was found to be 100 times inferior to pioglitazone while comparing their EC50 values. In order to understand the possibility of genistein to activate PPARγ by reaching plasma levels more than EC50, a pharmacokinetic study was also performed and simulated to evaluate the cumulative levels after multiple dosing in diabetic rats.
The effect of anti-diabetic compounds namely pioglitazone and genistein was evaluated in the in vivo STZ-induced diabetic rat model. In the present work, a preliminary single-dose study was conducted to derive pharmacodynamic equivalent dose of genistein with pioglitazone (3.5 mg/kg). This experiment concluded that 60 mg/kg genistein showed blood sugar lowering effect similar to that of pioglitazone (3.5 mg/kg). Therefore, genistein oral dose of 60 mg/kg was selected and administered twice daily in diabetic rats for our multi dose study.
In our study, abnormal increase in ERG parameters within the experimental period of 4 weeks with the mild but significant change in vascular architecture was found, the findings were in the concordance with the earlier studies documented the occurrence of earlier changes in diabetic complications at 4 weeks in diabetic rats.[29] This implies that in early diabetes, neuronal changes related to retinal function can precede or coincide with vascular changes. The increase in ERG parameter was alleviated in both of the treated groups but more in the genistein-treated group which could be due to its higher level of control over glucose transporters (as seen with the expression of both GLUT4 and GLUT1), in contrast to pioglitazone which showed a prominent role on the activation of GLUT1 as compared to GLUT4 in retina.
In this current study, diabetic rats exhibited a significant increase in the expression of Glut1 and decrease in Glut4 in the retina as compared to their age-matched normal rats. Whereas, after 7 days multidose treatment with PPARγ agonist pioglitazone showed a significant reduction of increased levels of the expression of GLUT1 and normalization of GLUT4 in diabetic rats. Interestingly, genistein showed a complete normalization of GLUT1 along with the 6-fold increase in the expression of GLUT4. The dual mechanism followed by genistein by down regulating GLUT1 along with the up regulation of GLUT4 shows the possibility of region specific glucose regulation in the eye which would ultimately beneficial in preventing the development of DR. In this study, no change was found in the expression of estrogen (Esr2) receptor; however, we found a significant up-regulation of PPAR in the retina of genistein treated animals.
Increase in GLUT1 expression has been reported in diabetic retina by Lu et al. in 2012 and decrease in the expression of GLUT4 has been reported in the diabetic rats.[30] Our study also showed a similar increase in expression of GLUT1 in diabetic group supporting the aforesaid observations. In addition, decreased expression of GLUT4 expression can lead to lesser intake of glucose inside photoreceptors can increase persistent extracellular glucose. Such an increased level of extracellular glucose might be instrumental in initiating the cascade of events leading to neuronal complications particularly glutamate excitotoxicity (ERG) as seen in our study. This finding also validates the studies by Vera et al.[10] who have characterized the anti-diabetic activity of genistein by GLUT1 inhibition.
While considering the vitreous and retinal levels of pioglitazone and genistein after multidose administration, pioglitazone reached levels above EC50 for the PPAR-γ whereas genistein reached levels much lower than that of its EC50 observed from in vitro studies. The vitreous and retinal levels of genistein reached after multiple dosing for 7 days was adequate to activate ER-β, and would have favored toward the upregulation of GLUT4 transporters in the euglycemic state that of pioglitazone. This observation shows that PPAR-γ independent pathway followed by genistein for the retinal expression of GLUT4 more than that of pioglitazone.
To conclude, this study showed the possibility of selective upregulation of GLUT4 in the diabetic retina of experimental animals using genistein and the mechanism of this upregulation may be independent of PPAR-γ activation considering the drug levels reached in the retina and vitreous humor. Up-regulation of GLUT-4 protected photoreceptors as predicted in the “a” wave of ERG. This selective up regulation can reduce the metabolic deprivation of glucose in neural retina which in turn prevent the incidence of subsequent neuro and vascular complications of DR. Implication of this observation provides the possibility of exploring selective modulation of retinal glucose transporters as therapeutic modality in preventing ocular diabetic complications.
Acknowledgment
We would like to acknowledge the help of Dr. Sharmilee Vetrivel and Dr. Rajesh Kumar while conducting this study.
Financial support and sponsorship
This work was supported by the All India Institute of Medical Sciences, New Delhi, India.
Conflicts of interest
There are no conflicts of interest.
References
Comments (0)