Background
Diabetes mellitus (DM) is a global menace, which ranks fourth among the diet-related non-communicable chronic diseases, after cardiovascular diseases, cancers and chronic respiratory diseases [
1]. In 2014, an estimated 387 million people were identified to be diabetic worldwide, and the global health expenditure accounted for 612 billion US dollars [
2]. Globally, around 80% of the DM population is represented by low and middle-income countries [
2,
3], which could probably due to the unaffordable medical expenditures. One strategic approach to curb diabetes is, therefore, to promote the intake of regionally available herbs and functional food derivatives in the daily diet, either as food or beverages, which could afford a considerable amount of protection against this lifestyle disease. The herbs which could be used as such either in whole or in parts deserve special attention due to their feasibility for daily use, easy availability and relatively low cost.
Various cross-sectional surveys across different countries and regions of the world revealed that 17–72.8% patients rely on complementary and alternative medicine (CAM) methods in managing DM, either as a supportive measure along with the modern medicine or occasionally as a standalone therapy [
4]. A cross-sectional study conducted in Malaysian population showed that the prevalence of CAM use in the management of DM was 62.5%, wherein among the use of herbs,
Orthosiphon stamineus (OS) ranked second at the preference of 38.7% of the patients, to bitter gourd (
Momordica charantia), which stood first with 48.7% [
5]. OS is renowned for its consumption in the form of a herbal infusion known as Java tea. For ages, OS leaves have been widely used in traditional medicine practices across South East Asian (SEA) region due to its curative effects in diabetes, hypertension, edema, hepatitis, jaundice, renal calculi, gout, and rheumatism. Numerous reports confirming their effectiveness as a cure for nephrolithiasis, hydronephrosis, diabetes, vesical calculi, and arteriosclerosis could be retrieved [
6‐
8].
Few pharmacological reports on the antidiabetic activity of OS in in vivo experimental models are available, whereby hypoglycemic and antihyperglycemic potentials of OS extracts have been reported. Repeated daily oral administration of OS aqueous extract for 14 days at a dose of 500 mg/kg bw elucidated comparable plasma glucose lowering effects with that of 5 mg/kg bw of glibenclamide [
9]. Furthermore, it was identified that OS does not potentiate the glucose-induced insulin secretion [
10]. However, no studies explored the underlying mechanism of glucose or any other DM biomarker regulation by OS treatment. Although potent inhibitory activity of OS on the pancreatic
α-amylase and intestinal
α-glucosidase, two enzymes which cause a sudden postprandial hyperglycemia in type 2 DM by enhancing the hydrolysis of starch and uptake of glucose was suggested [
11], until date, the exact protective mechanism of OS in DM is not reported, to the best of our knowledge. Therefore, it is of great significance to investigate the underlying mechanisms of antidiabetic activity to derive deeper insight into their therapeutic effects and thus to ensure its safe and effective usage.
Metabolomics identifies possible metabolic pathways by unraveling the complex inter-relationships of cellular metabolites by their identification and quantification using sophisticated analytical and statistical techniques. Metabolomics has already demonstrated the suitability in the evaluation of the pharmacological effects and mechanism of action of various herbs [
12,
13]. As metabolomics involves the quantification of small endogenous molecules such as amino acids, sugars, lipids and organic acids, which are either end products or intermediates in various metabolic pathways, it offers an efficient and simple tool to get a comprehensive snapshot of the internal cellular environment [
14,
15]. Furthermore, the compatibility of peripheral fluids such as urine and serum in a metabolomic analysis has promoted its application in pharmacological-toxicological studies by the holistic and global determination of metabolites, or patterns of biomarkers that altered as the result of a stimulus [
16]. Although many kinds of advanced instrumental approaches could be utilized in metabolomics studies, Nuclear Magnetic Resonance (NMR) spectroscopy is one of the most preferred methods as it offers simple sample preparation, handling and data analysis techniques [
17]. Recently, reports on the use of metabolomics in the study of diabetes with an objective of identifying new metabolic biomarkers and therefore to further the understanding on the underlying biochemical mechanisms and metabolic pathways are in plenty [
18‐
20].
Accumulating all necessary points in an attempt to solve the stated problem, this study was designed to investigate the effect of various standardized OS extracts in the endogenous metabolites of the diabetic rat after 14 days of oral administration, through 1H NMR metabolomic analysis of urine samples. Herein, Streptozotocin (STZ) induced rat model was used. The modulatory effects of OS on potential diabetic biomarkers in the rats were investigated and a pathway analysis identified the most relevant therapeutic targets, and thus, the metabolic pathways involved in the treatment, thereby unraveling the mechanism. To the best of our knowledge, this study is the first report on the protective effect of OS on pathological changes of STZ-induced diabetes mellitus based on metabolomic approach, which may serve to fill the knowledge gap about the possible mechanisms and metabolic targets of OS on DM.
Methods
Chemicals and reagents
3-Trimethylsilylpropionic acid (TSP) and Streptozotocin (STZ) were obtained from Sigma-Aldrich (St. Louis, USA). Deuterium oxide (D2O, 99.9%), deuterated methanol (CD3OD, 99.9%), potassium dihydrogren phosphate, deuterated sodium hydroxide and sodium azide were purchased from Merck (Darmstadt, Germany). The normal rat chow feed was purchased from Specialty feeds (Glen Forrest, Australia).
OS plants of eight to 10 weeks old were collected from Kampung Repuh, Batu Kurau (GPS coordinates: 4.52° N, 100.48° E), Perak, Malaysia in January 2012. After authenticated by a botanist, a voucher specimen (SK1997/12) was deposited at the herbarium of Institute of Bioscience, Universiti Putra Malaysia, Malaysia. The separated leaves were cleaned and then dried using industrial scale continuous drying microwave equipment with an output frequency of 2.45 ± 0.05 GHz for a period of 40 s. The selection of the microwave frequency was based on a trial and error basis in ensuring an efficient drying process that caused little stress to the composition of the leaves as assessed by their color and texture. The dried leaf material was then ground in a blender to powder; size uniformity was ensured by sieving through a stainless steel mesh of 200 mm diameter and stored in airtight containers at 3 ± 2 °C for further processing. The powdered leaf material was then extracted by ultrasonic assisted extraction method. Briefly, weighed quantity of the leaf material was extracted in a measured volume of solvent (ratio of 1 g: 20 mL) by subjecting to sonication for 30 min, while maintaining the sonication bath at a temperature window of 30–40 °C. The extract was filtered before repeating twice with fresh solvent and sonication step, for each time. The filtrate was combined and the solvent was removed using rotary evaporator at 40 °C. The resulted crude extracts of OS (OSE) were lyophilized (extraction yield; water: 14% w/w; methanol: 9% w/w; ethanol 13%, w/w and 50% ethanol water: 8% w/w) and kept frozen until use.
Phytochemical analysis of the OS extracts
The HPLC quantification of two major marker compounds of OS, namely rosmarinic acid and sinensetin, in OS aqueous extract was carried out, in accordance with the methods described in our published report on OS [
21].
Animal experimental design and dose preparation
All the animal experiments were conducted in Animal Biosafety Level – 2 (ABSL - 2) housing complex located at Laboratory of Animal Resource, Universiti Kebangsaan Malaysia (Bangi, Malaysia). A total of 60 male Sprague Dawley (SD) rats, 11 weeks old (225 ± 50 g) were used. The animals were maintained in an air conditioned room at 24 ± 2 °C and acclimatized for 7 days before the experiment. Three rats were housed per polycarbonate cages. The light cycle was maintained at 12 h of light and 12 h of darkness and the rats were allowed free access to food and water.
Rats were randomly divided into 12 groups, with five rats in each group. Six groups were normal rats injected with 0.9% saline for control (N), normal rats treated with OSE water (N-A), normal rats treated with OSE methanol (N-M), normal rats treated with OSE ethanol (N-E), normal rats treated with OSE 50% ethanol (N-50E) and normal rats treated with glibenclamide (N-G). The other six groups were diabetic rats divided into diabetic-control (D), diabetic rats treated with OSE aqueous (D-A), diabetic rats treated with OSE methanol (D-M), diabetic rats treated with OSE ethanol (D-E), diabetic rats treated with OSE 50% ethanol (D-50E) and diabetic rats treated with glibenclamide (DG). All animal handling and experimental protocols were performed in strict accordance with the ethics guidelines approved by Universiti Putra Malaysia Animal Ethics Committee (Approval number: UPM/FPSK/PADS/BR-UUH/00485).
The stock solutions of OSE were prepared separately using 1% CMC as the vehicle. The OSE dose used in the study was 500 mg/kg of the rat body weight (bw) prepared from the respective stock solution and administrated by oral force feed. For glibenclamide, 10 mg/kg bw dose was used and dissolved in DMSO (25 mg/ml). All the doses of extracts and glibenclamide were preserved at 4 °C and used within 3 days.
Induction of diabetes
The fasted rats were injected intraperitoneally with 60 mg/kg freshly dissolved STZ in 0.9% saline. One week after the STZ administration, the rats with fasting blood glucose concentrations of over 300 mg/dl were considered to be diabetic and used in further experiments. The diabetic rats were then treated for 14 days with 500 mg/kg bw of each extract. The experiment is a modified method described by Sriplang et al., (2007), wherein 14 days has proven to have anti-hyperglycemic effect on rat plasma, without any obvious adverse effects [
9]. The positive control, glibenclamide was given in similar manner to that of treatment. For urine collection, rats from all groups were kept in metabolic cages for 14 h of fasting. Each urine sample was collected into 0.1 ml of 1% sodium azide solution and then centrifuged for 10 min at 4 °C, from which the collected supernatant was stored at −80 °C until analysis.
1H NMR spectroscopic analysis of urine
Urine samples were thawed, centrifuged at 5000 rpm for 10 min and then, 400 μl supernatant obtained from each sample was mixed with 200 μl of phosphate buffer solution (0.1232 g of KH2PO4), containing 10 mg trimethylsilylpropionic acid sodium salt (TSP) prepared in 10 ml D2O with 1.0 M NaOD solution (used to adjust the pH to 7.4), in 5 mm standard NMR tube (Norell, Sigma-Aldrich, Canada). The NMR spectra were recorded using 500 MHz NMR spectrometer (Varian Inova 500, Illinois, USA) at 25 °C with the parameter of pulse width (PW) 21.0 μs (90°) and relaxation delay (RD) 2.0 s. Deuterium oxide was used as the internal lock and TSP was used as the calibration standard and referenced the chemical shift at δ 0.0 ppm.
The 2D NMR spectra such as J resolved (JRES), COSY and HMBC were acquired using Bruker Ascend 700 MHz instrument at room temperature (25 °C). The JRES and COSY spectra were acquired using 4 scans, 1 K data points at 128 increments and a spectral width of 16 ppm in dimensions and a relaxation delay of 2 s. The heteronuclear multiple bond coherence (HMBC) spectra were obtained using 8 scans, 1 K data points, 256 t1 increments at a spectral width of 13 ppm and 220 ppm in the proton and carbon dimensions respectively. The relaxation delay was 1.5 s.
Statistical analysis of 1H NMR spectra
All of the NMR spectra were manually phased, baseline corrected and calibrated to TSP at 0.00 ppm. The chemical shift (δ) region 0 to 10 was reduced to integrated bins of 0.04 ppm width to use in Chenomx NMR software package (Chenomx NMR Suite 5.1 Professional, Edmonton, Canada) for multivariate pattern recognition analysis. The region of the spectra associated with residual water and urea (4.66–5.05 and 5.54–6.0 ppm) were removed. The remaining spectral segments for each NMR spectrum were normalized to the total sum of the spectral intensity to partially compensate for differences in concentrations of the samples. NMR data was then imported to SIMCA-P 13.0 software package (Umetrics, Umea°, Sweden) for analysis and visualization by multivariate statistical methods. Data was mean-centered and Pareto scaled prior to analysis by Principal Component Analysis (PCA) and Orthogonal Partial Least Squares-Discriminant Analysis (OPLS-DA). Data were visualized with the scores plot of the two principal components (PC1 and PC2) in which each point represented an individual spectrum of a sample. The metabolites associated with the group separation were indicated by the corresponding loading plots, in which each point stood for a single NMR spectral bin. The validation and significance of the model were done by using permutation test, CV-ANOVA and R2Y/Q2Y values as and when applicable [
22].
The pathway analysis and heat map was generated using Metaboanalyst 3.0, (
http://www.metaboanalyst.ca), which is a freely available web-based platform for comprehensive analysis of metabolomic data. The univariate analysis of the integration areas of the metabolites was performed. Kolmogorov-Smirnov test was used to check the normality of the distribution. One-way analysis of variance (ANOVA) was done using GraphPad Prism V 7.0 (GraphPad Software Inc., San Diego, USA), Tukey’s test was chosen as the post hoc analysis method.
P ≤ 0.05 was considered to be statistically significant and the values were expressed as mean ± SEM.
Conclusions
A 1H NMR based urine metabolomics tool has been used for the first time to evaluate the protective effects of OS in DM using STZ induced experimental model in rats. Pattern recognition combined with multivariate statistical analysis identified that 14 days oral administration of OSAE at the dose of 500 mg/kg bw caused the reversal of DM. A total of 15 metabolites, which levels changed significantly upon treatment were identified as the biomarkers of OSAE in diabetes. These biomarkers suggested the involvement of several metabolic pathways, whence, a systematic metabolic pathways analysis identified that OSAE exerted the antidiabetic activity through regulating the TCA cycle, glycolysis/gluconeogenesis, lipid and amino acid metabolism. Thus, metabolomics approach aided in exploring the effects of OS in DM biomarkers and provided a better understanding of the mechanistic pathways involved, and proved as a promising tool in the ethnopharmacological validation and mechanism of action studies in traditional medicine research.
Acknowledgements
The authors are grateful to the Ministry of Agriculture and Agro-based Industry, Malaysia and Research Management Centre of Universiti Putra Malaysia for facilitating this work. The authors would like to acknowledge the staffs of Laboratory of Natural Products, Institute of Bioscience, Universiti Putra Malaysia for all the helps provided during the project.