Disclosure: The authors have no ownership or intellectual property rights or any financial ties with the sponsors. The authors, not the sponsors, are responsible for data collection, analysis and interpretation as well as manuscript preparation. The sponsors were aware of the study design which determined the budget and underwriting of the project.
Efficacy and tolerability of a novel herbal formulation for weight management
Article first published online: 20 JUN 2013
Copyright © 2012 The Obesity Society
Volume 21, Issue 5, pages 921–927, May 2013
How to Cite
Stern, J.S., Peerson, J., Mishra, A.T., Sadasiva Rao, M.V. and Rajeswari, K.P. (2013), Efficacy and tolerability of a novel herbal formulation for weight management. Obesity, 21: 921–927. doi: 10.1002/oby.20211
Funding agencies: This research was supported by the InterHealth Nutraceuticals Inc.
- Issue published online: 20 JUN 2013
- Article first published online: 20 JUN 2013
- Accepted manuscript online: 29 NOV 2012 04:08PM EST
- Manuscript Accepted: 16 NOV 2012
- Manuscript Received: 12 DEC 2011
- InterHealth Nutraceuticals Inc
To evaluate the efficacy of an herbal blend.
Design and Methods:
A randomized, double-blind, clinical trial in 60 subjects with body mass index (BMI) between 30 and 40 kg/m2. Participants were randomized into two groups receiving either 400 mg herbal capsules or 400 mg placebo capsules twice daily. The herbal blend comprises of extracts from Sphaeranthus indicus and Garcinia mangostana. Participants received a standard diet (2,000 kcal per day) and walked 30 min 5 days per week.
After 8 weeks, significant net reductions in body weight (3.74 kg; P < 0.0001), BMI (1.61 kg/m2; P < 0.0001), and waist circumference (5.44 cm; P < 0.05) were observed in the herbal group compared with placebo. Additionally, a significant increase in serum adiponectin concentration was found in the herbal group versus placebo (P = 0.001). Adverse events were mild and were equally distributed between the two groups. In vitro studies in the 3T3-L1 adipocyte cell line showed that the herbal extract markedly downregulated the expression of peroxisome proliferator-activated receptor gamma, adipocyte-differentiation related protein, and cluster of differentiation 36 but increased adiponectin expression. The herbal extract also reduced the expression and the recruitment of perilipin onto the membrane of lipid droplets.
Supplementation with the herbal blend resulted in a greater degree of weight loss than placebo over 8 weeks.
The prevalence of excess weight gain is a global health problem (1–3). In 2008, the World Health Organization estimated that about 1.5 billion adults (aged 20 and above) were overweight with nearly 500 million of these individuals classified as obese (4). In the United States, over 60% of the adults are either overweight or obese (5). The impact of these demographic changes on overall healthcare costs is dramatic and continues to rise (6).
Existing treatments for obesity include diet, exercise, behavior therapy, pharmacotherapy, and surgical intervention (7). Several pharmacotherapies for treating overweight and obesity exert their primary action by either suppressing appetite, limiting food absorption, reducing food intake, altering metabolism, or increasing energy expenditure (8). Each intervention has its distinct advantages. They may also have significant adverse effects and limitations (9). For example, sibutramine increases cardiovascular risk in a dose-dependent manner (10). Some weight loss drugs may be habit-forming (11). Therefore, many people seek remedies having a better safety profile to achieve and to maintain reductions in body weight.
Ayurveda is an ancient system of health care that originated in India. Over 2,000 plants are described for alleviating diabetes, obesity, allergy, inflammation, pain, and promoting general well-being (12). With the goal of developing an herbal weight loss ingredient, we screened extracts from multiple medicinal plants for their ability to prevent fat accumulation (adipogenesis) while promoting fat breakdown (lipolysis). Extracts exhibiting potent efficacy in the initial screening were selected for further studies both individually and in various combinations. A formulation consisting of the extracts of the flower heads from Sphaeranthus indicus and the fruit rinds of Garcinia mangostana was selected for further development.
Here, we report the results of a randomized, double-blind, placebo-controlled clinical study designed to evaluate the efficacy and tolerability of the herbal blend for managing body weight in obese human subjects with weight loss as the primary outcome.
Materials and Methods
In vitro studies
Chemicals and reagents
Insulin, dexamethasone, 3-isobutyl-1-methylxanthine, Dulbecco's modified Eagle's medium (DMEM), penicillin, streptomycin, sodium pyruvate, dimethyl sulfoxide (DMSO), D-glucose, Triton X-100, Tris buffer, sodium chloride, ethylenediamine tetraacetic acid (EDTA), sodium deoxycholate, sodium vanadate, sodium fluoride, aprotinin, leupeptin, pepstatin, phenylmethylsulfonyl fluoride (PMSF), and anti-actin antibody were obtained from Sigma Chemical Co. (St. Louis, MO). Fetal bovine serum was purchased from Hyclone (Logan, UT). Anti-PPARγ2/1 antibody was procured from Millipore (Billerica, MA). Antibodies specific to cluster of differentiation 36 (CD36) and adipocyte-differentiation related protein (ADRP) were purchased from R&D Systems (Minneapolis, MN) and Santa Cruz Biotechnology (Santa Cruz, CA), respectively. Antibody against perilipin was obtained from Abcam (Cambridge, UK).
Cell culture and treatment
3T3-L1 mouse embryo fibroblasts (American Type Culture Collection, Manassas, VA) were cultivated in maintenance medium comprised of DMEM supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 μg/ml streptomycin, 1 mM sodium pyruvate, and 4.5 g/l D-glucose. The herbal blend was dissolved in DMSO at 10 mg/ml and stored in aliquots at −20°C and diluted immediately before use in a culture medium to a final concentration of 0.1% (v/v) DMSO.
Equal number of 3T3-L1 cells (6 × 104 cells/well) was seeded in each well of 24-well tissue culture plates and grown to confluence. Cells were pretreated with different concentrations of the herbal blend (5-15 μg/ml) for 1 h and incubated further with the differentiation medium (maintenance medium supplemented with 500 nM insulin, 1 μM dexamethasone, and 0.5 mM 3-isobutyl-1-methylxanthine) for 2 days. The cells were further maintained in the post-differentiation medium comprised of the maintenance medium and 100 nM insulin in the presence or absence of the herbal blend for 8 days. The cell culture medium was changed every 48 h. The control cultures received 0.1% (v/v) DMSO vehicle.
Treated cells were washed twice with chilled phosphate buffered saline (PBS) and cell lysates were prepared in a lysis buffer (50 mM Tris pH 7.5, 150 mM sodium chloride, 2 mM EDTA, 1% Triton X-100, 0.5% sodium deoxycholate, 1 mM sodium vanadate, 1 mM sodium fluoride, 1 μg/ml aprotinin, 1 μg/ml leupeptin, 1 μg/ml pepstatin, and 1 mM PMSF). Cell lysates were clarified by centrifugation at 14,000g for 20 min at 4°C. Protein concentrations were estimated by Bradford reagent (13).
Equal amount of proteins (20 μg/lane) was resolved in sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Following SDS-PAGE, the electro-blotted nitrocellulose membranes were reacted with primary antibody specific to PPARγ, ADRP, CD36, and adiponectin. The same membrane was reprobed with anti-actin antibody. Specific signals were detected with enhanced chemiluminescence (Thermo Scientific, Rockford, IL) and signal intensities were analyzed (Molecular Imaging Software, version 4.0, Eastman Kodak Company, Rochester, NY).
Perilipin immunofluorescence microscopy
Subcellular localization of perilipin in adipocytes was performed by immunofluorescence assay as previously described with modifications (14). Briefly, 3T3-L1 cells were differentiated ± herbal blend (10 μg/ml). Control cultures received 0.2% (v/v) DMSO as the vehicle. Following treatment, cells were washed with PBS and fixed in 4% paraformaldehyde for 15 min at room temperature. Fixed cells were permeabilized with 0.1% Triton X-100 for 5 min then treated with perilipin antibody in 1% bovine serum albumin and 0.1% Triton X-100 overnight at 4°C. After washing, cells were treated with FITC-conjugated, anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) for 15 min and prepared for visualization of intracellular localization of perilipin using a 40× objective under a Nikon Eclipse TS 100F microscope equipped with a Nikon Coolpix camera (Nikon Corporation, Japan).
The study material is an herbal blend consisting of aqueous alcohol extracts of the flower heads from S. indicus and the fruit rinds of G. mangostana. These extracts were concentrated and mixed together in a 3:1 ratio of S. indicus to G. mangostana to achieve a final product containing 3% 7-hydroxyfrullanolide and 2% α-mangostin. These two components serve as endogenous reference standards for controlling batch-to-batch consistency. Quality assurance was conducted by standardizing for the above two components as well as bioassay-guided testing to ensure both the chemical and the biological properties of the herbal blend. Reverse phase high-performance liquid chromatography analysis (C18, acetonitrile) demonstrated that the herbal extract is consistent across multiple lots (data not shown).
The herbal blend (400 mg) was encapsulated in size zero hard gelatin capsules with excipients. Each placebo capsule contained only excipients and was identical to that of the active capsules. The herbal blend was provided by InterHealth Nutraceuticals (Benicia, CA) and Laila Nutraceuticals, (Vijayawada, India).
Recruitment of subjects
This randomized, double-blind, clinical trial (RCT) was performed at ASRAM, Eluru, Andhra Pradesh, India from March 2010 to July 2010 (clinical trial registration number: ISRCTN52261953). The study protocol was evaluated and approved by the ASRAM Institutional Review Board (IRB).
One hundred eleven subjects were selected out of a pool of 182 candidates using a questionnaire based screening procedure. Inclusion criteria were: adults (21-50 years), body mass index (BMI) (30-40 kg/m2), not pregnant, willingness to follow method of birth control if premenopausal. Exclusion criteria were: history of thyroid disease, cardiovascular disease, diabetes, hepatitis, pancreatitis, lactic acidosis or hepatomegaly with steatosis, motor weakness, peripheral sensory neuropathy, allergies to spices and herbal products, using weight loss medications, laxatives or diuretics taken solely for the purpose of weight loss, recent unexplained weight loss or gain, HIV positive, undergone surgery before 30 days of screening or planning to undergo within the study period, any evidence of organ dysfunction or any clinically significant deviation from normal that could impact subjects well-being or clinical outcome. Out of the 111 candidates, 60 subjects were enrolled in the study. Each participant voluntarily signed the IRB approved informed consent form. After recruitment, the subjects were randomly assigned to either placebo or herbal blend group.
At baseline evaluation, subjects were provided active or placebo capsules, compliance cards, the list of instructions for moderate exercise, schedules for delivery of daily meals, and dates for follow-up evaluations. The subjects were provided with free prepared meals totaling 2,000 kcal/day, with 61% of calories from carbohydrate, 14% from protein, and 25% from fat. The subjects were instructed to take 2 capsules a day, one each 30 min before breakfast and dinner.
All subjects filled out a questionnaire providing details regarding medical history plus compliance with the exercise regimen and the diet at baseline as well as all follow-up evaluations at 14, 28, and 56 days. At baseline and subsequent follow-up visits, subjects were assessed for body weight, height, waist plus hip circumference, vital signs, and blood and urine were collected.
Serum adipokine assay
Serum adiponectin levels were determined using human adiponectin enzyme-linked immunosorbent assay kit (Millipore).
Outcome variables were assessed for conformance to the normal distribution and transformed if necessary. Mean values of outcomes were compared with a mixed model analysis, including treatment group and time as main effects, the interaction between treatment group and time, and baseline value of the variable and gender as covariates; follow-up analyses of covariance were used to compare the groups at each time point. For Day 0 comparison, a t-test was used. P-values less than 0.05 were considered to be statistically significant for between-group comparisons. All comparisons reported are between the placebo and the herbal groups at each specified time point. Analysis was performed with SAS for Windows Release 9.2 (Cary, NC). All statistical results were independently generated by the Department of Nutrition, University of California, Davis.
Baseline demographic characteristics
Overall, the patient profiles with respect to age, sex, height, weight, BMI, waist circumference, and hip circumference were not significantly different between the herbal blend (800 mg/day, n = 30) and the placebo (n = 30) groups (Table 1).
|Characteristics||Placebo (n = 30)||Herbal (n = 30)||P value|
|Number of females (%)||22 (73.3)||23 (76.7)||NA|
|Age (years)||38.77 ± 1.39||38.93 ± 1.45||0.649|
|Height (m)||1.58 ± 0.02||1.55 ± 0.02||0.226|
|Body weight (kg)||84.12 ± 2.42||81.92 ± 1.69||0.406|
|Body mass index (kg/m2)||33.77 ± 0.48||34.14 ± 0.54||0.614|
|Waist circumference (cm)||98.26 ± 1.85||99.10 ± 1.64||0.749|
|Hip circumference (cm)||115.57 ± 1.42||116.76 ± 1.37||0.555|
Reduction in body weight, BMI, and waist circumference
Table 2 summarizes the body weight changes over time for subjects supplemented with either herbal supplement or placebo. The herbal group experienced a statistically significant greater reduction in body weight at 2, 4, and 8 weeks compared with placebo group (P < 0.0001) with significant net weight losses of 1.28, 2.17, and 3.74 kg at 2, 4, and 8 weeks, respectively. These net changes represent a 1.6%, 2.7%, and 4.6% decrease in overall body weight. A similarly statistically significant decrease in BMI was observed for subjects consuming the herbal blend (Table 2). Supplementation with the herbal blend resulted in a statistically significant net reduction in BMI of 0.53, 0.92, and 1.61 kg/m2 (P < 0.0001) at 2, 4, and 8 weeks.
|Parameters||Week||Placebo (n = 27)||Herbal (n = 29)||Net reductionab||P value|
|Reduction in body weight (kg)||2||0.56 ± 0.16||1.84 ± 0.12||1.28 (1.6%)||0.0001*|
|4||0.82 ± 0.15||2.99 ± 0.26||2.17 (2.7%)||0.0001*|
|8||1.34 ± 0.26||5.08 ± 0.43||3.74 (4.6%)||0.0001*|
|Reduction in BMI (kg/m2)||2||0.24 ± 0.06||0.77 ± 0.05||0.53 (1.6%)||0.0001*|
|4||0.33 ± 0.06||1.25 ± 0.11||0.92 (2.7%)||0.0001*|
|8||0.53 ± 0.10||2.14 ± 0.18||1.61 (4.7%)||0.0001*|
|Reduction in waist circumference (cm)||2||0.84 ± 0.89||4.22 ± 0.79||3.38 (3.4%)||0.016*|
|4||3.63 ± 0.99||7.37 ± 1.04||3.73 (3.8%)||0.013*|
|8||6.40 ± 1.09||11.84 ± 1.07||5.44 (5.5%)||0.0004*|
Supplementation with the herbal blend also resulted in statistically significant net decreases (P < 0.05) in waist circumference of 5.44 cm (Table 2). This corresponds to a 5.5% reduction in waist circumference versus baseline. The net impact of the herbal blend on waist circumference was statistically significant as early as 2-weeks of supplementation (reduction = 3.38 cm, P < 0.05) with the reduction in waist size continuing to increase through 4 weeks of supplementation (reduction = 3.73 cm, P < 0.05).
Modulation of serum adiponectin, cholesterol, and triglyceride levels
In this study, we found that herbal supplementation for 8 weeks resulted in a statistically significant increase in adiponectin levels compared with placebo (0.93 μg/ml vs. 0.04 μg/ml; P = 0.0011). It has been shown that the circulating levels of this hormone are inversely proportional to the amount of body visceral fat (15, 16). These results suggest that the herbal blend may influence lipid metabolism, which is consistent with the reduction in waist circumference noted above.
The herbal supplementation also significantly decreased total cholesterol compared with the placebo group (26.17 mg/dl vs. 10.7 mg/dl; P = 0.016). Similarly, the triglyceride levels were significantly reduced in herbal supplement group more than in placebo group (56 mg/dl vs. 31.82 mg/dl; P = 0.012) (Table 3).
|Parameter (U)||Placebo (n = 27)||Herbal (n = 29)||P value (between groups)|
|SGOT (U/l)||−1.52 ± 1.13||−3.83 ± 1.12||0.6904|
|SGPT (U/l)||−1.26 ± 1.48||−3.38 ± 1.57||0.5582|
|Systolic BP (mm Hg)||−0.37 ± 1.89||−1.03 ± 2.12||0.4955|
|Diastolic BP (mm Hg)||0.44 ± 1.70||−0.76 ± 1.73||0.4369|
|Pulse rate (beats/min)||−1.00 ± 1.41||0.83 ± 1.63||0.1552|
|Cholesterol (mg/dl)||−10.70 ± 4.21||−26.17 ± 4.82||0.0159*|
|Triglycerides (mg/dl)||−31.82 ± 6.88||−56.00 ± 6.80||0.0120*|
|LDL (mg/dl)||−1.98 ± 1.42||−8.30 ± 4.22||0.2830|
|HDL (mg/dl)||1.63 ± 1.29||2.85 ± 1.17||0.3557|
|Fasting glucose (mg/dl)||−3.52 ± 2.52||−10.97 ± 4.23||0.1168|
Modulation of adipogenesis marker proteins in 3T3-L1 cells
Figure 1A and 1B, respectively, depicts a representative immunoblot and densitometric analysis showing that the herbal blend downregulated the expression of peroxisome proliferator-activated receptor gamma (PPARγ), ADRP, and CD36 in a dose-dependent manner in 3T3-L1 cells. By contrast, the herbal blend upregulated the expression of adiponectin, consistent with the clinical results described above. PPARγ, CD36 protein, and ADRP play a role in regulating adipogenesis, cholesteryl ester uptake, and fat storage, respectively (17–19). These observations imply that the herbal blend modulates several key aspects of adipogenesis that include adipocyte differentiation, lipid droplet formation, and fatty acid uptake by adipocytes.
Downregulation of perilipin expression on lipid droplets in 3T3-L1 cells
Perilipin is an integral membrane protein primarily localized on the surface of lipid droplets that prevents the access of lipases to the triglycerides stored in the fat droplets (18). Figure 2 illustrates the immunofluorescence images of adipocytes stained with antibody against perilipin. The results show that treatment with the herbal blend markedly inhibited the expression of perilipin (Figure 2B) as compared with untreated control (Figure 2A).
No major adverse events were reported during this study. Minor adverse events (headache, nausea, gastrointestinal irritation, plus back, leg, ankle, and joint pain) were evenly distributed between the two groups.
In addition, we measured the effect of herbal supplementation on various biochemical parameters plus other markers associated with liver, heart, kidney, and metabolic function. No changes between groups were observed for these biochemical markers after 8 weeks of supplementation (Table 3).
Four subjects (one from the herbal group and three from the placebo group) were dropped from the study and were lost to follow-up. No subject discontinued participating in the study as a result of an adverse event. The results attributed to the dropouts were excluded from all statistical analyses as per the predefined criteria established in the study protocol.
We evaluated an herbal blend of S. indicus and G. mangostana extracts using RCT and in vitro studies in 3T3-L1 adipocytes model system. The herbal extracts were screened in vitro for their ability to inhibit adipogenesis and promote lipolysis. The main findings of this RCT were that the herbal blend, when provided together with a standard diet and moderate daily exercise, promoted a significantly higher reduction of body weight, BMI, and waist circumference than diet management and exercise alone.
The present research report describes the results of a clinical study conducted in 60 human subjects with BMI between 30 and 40 kg/m2. Subjects received an 800 mg daily dose of either herbal supplement or placebo for 8 weeks. We observed that treatment with the herbal blend for 8 weeks yielded a 3.7 kg weight loss relative to placebo. The net weight reducing effects of the herbal blend occur as early as 2 weeks after initiating supplementation. These effects include statistically significant reductions in weight, BMI, and waist circumference. The relatively low drop-out rate (one in herbal and three in placebo group) may be due to weight loss experienced in both groups plus the availability of free meals (20). When taken together, these observations support the potential of the herbal blend for managing weight loss.
Preclinical studies have provided insights into the possible mode of action by which the herbal blend exerts its effect on weight loss. In vitro studies with 3T3-L1 adipocytes demonstrate that the herbal blend reduces the expression of PPARγ, CD36, and ADRP. Additionally, the herbal blend markedly decreased the expression of perilipin on the surface of lipid droplets. These results suggest that the herbal blend may attenuate fat accumulation by partially blocking adipogenesis as well as fat uptake by mature adipocytes and by rendering lipid droplets more susceptible to lipases.
Numerous studies have shown that the nuclear transcription factor PPARγ plays a pivotal role in activating adipocyte specific-gene expression and differentiation as well as the control of energy accumulation in the form of adipose tissue mass (21). A reduced level of this protein significantly impacts an organism's ability to store fat. For example, inhibition of PPARγ with specific antagonists ameliorates diet-induced obesity and insulin resistance in mouse models (22). Likewise, adipose-specific PPARγ knockout mice display impaired adipose tissue development even when fed a high-fat diet (23). Therefore, in vitro findings suggest that the reduction in weight caused by this herbal blend may result from reduced fat storage.
CD36 is a transmembrane receptor that facilitates the uptake of long chain fatty acids by adipocytes and is upregulated in subcutaneous and visceral adipose tissue in obese subject (24). The downregulation of this receptor by this herbal blend provides another mechanism for preventing fat accumulation in adipocytes.
The observation that the herbal blend blocks both perilipin and ADRP expression in vitro is a key finding. Perilipin and ADRP are proteins that coat the lipid droplets of adipocytes and play an important role in their formation (18, 25). Both proteins are members of the PAT (Perilipin, ADRP, and TIP47) domain family that regulates the access of lipases to lipid esters within lipid droplets (18). Perilipin knock-out animals are reported to exhibit an increased level of basal lipolysis and are resistant to diet-induced obesity (26). Even though these mice consume equal amounts of food compared with wild-type mice, they demonstrate a 30% reduction in adipose tissue mass accompanied by a greater lean body mass (26). Likewise, in vitro research in the MIN6 pancreatic beta cell line has shown that downregulation of ADRP suppresses the accumulation of triglycerides along with a decrease in the expression of lipogenic genes such as diacylglycerol O-acyltransferase-2 and fatty acid synthase (25). Given these observations, it seems reasonable to propose that the herbal blend may enhance weight loss by two distinct mechanisms: decreased fat accumulation and increased fat breakdown.
Adiponectin levels were significantly higher in in vitro 3T3-L1 cells and clinical studies suggesting that the herbal blend acts directly at the level of adipocytes. Adiponectin is predominantly secreted by white adipose tissue and plays an important role in pathogenesis of obesity and type II diabetes (15).
Circulating adiponectin levels are lower in people who have obesity, diabetes, and cardiovascular disease than those of healthy control subjects (27–29). Plasma levels of adiponectin have been shown to negatively associate with degree of adiposity and positively associate with insulin sensitivity (30). In addition, adiponectin is capable of promoting fat oxidation in tissues such as skeletal muscles and liver (31, 32). Therefore, the beneficial effects observed on serum triglycerides and total cholesterol may at least in part be via an increase in adiponectin levels by this herbal blend.
In conclusion, supplementation of the herbal blend at a daily dose of 800 mg resulted in statistically significant reductions in body weight, BMI, and waist circumference that exceeded those achieved via diet management and moderate exercise alone. Its significant effect on body weight reduction was seen as early as 2 weeks. Supplementation with the herbal blend also resulted in statistically significant reductions in serum cholesterol and triglyceride concentrations. Analysis of comprehensive blood panel and perusal of adverse events observed during the study, indicate that the herbal blend is well-tolerated with minimal side effects. Thus, these observations indicate that this herbal blend is promising as a weight loss ingredient; however, longer term studies are needed as the current study was relatively short-term with a small sample size. Based on the in vitro modulation of proteins associated with lipid metabolism, the weight reduction effect of this herbal blend appears to be mediated via (1) reduced adipocyte differentiation; (2) decreased uptake of fatty acids by adipose tissue; and (3) increased lipolysis of stored fat. In conclusion, the blend of S. indicus and G. mangostana we tested demonstrated weight loss efficacy over a short duration of 8 weeks and was well tolerated.
This work was supported by an unrestricted grant from InterHealth Nutraceuticals Inc. Benicia, CA to Judith S. Stern, University of California, Davis.
- 4World Health Organization (WHO). Obesity and overweight. Fact sheet N°311. March, 2011.
- 5Prevalence and trends in obesity among US adults, 1999–2008. JAMA 2010; 303235-303241., , , .
- 7Current and future treatment of metabolic syndrome and type 2 diabetes in children and adolescents. Diabetes Spectrum 2005; 18: 220-228., , , , .