Enteric-coated capsaicin (2%) pellets were provided by Phytosol India Pvt Ltd, Bangalore, India. Sodium alginate (Vivapharm®; JRS PHARMA GmbH & Co. KG, Rosenberg, Germany), microcrystalline cellulose powder (Pharmacel® 101; DFE Pharma, Goch, Germany) and ethyl cellulose (ASHACEL® M-20; Asha Cellulose Pvt Ltd, Abrama, Valsad, Gujarat, India) were used. Besides, n-hexane, ethyl acetate, isopropyl alcohol, polyvinyl pyrrolidine K-30, sucrose and sodium dodecyl sulfate were procured from Sigma-Aldrich, Merck, Bangalore, India. Orlistat 97% purity was purchased from TCI Chemicals Pvt. Ltd, Chennai, India. Normal fat diet and high-fat diet from Research Diet Inc, New Brunswick, NJ, USA were used in this study. The reference standard capsanthin was purchased from Sigma-Aldrich, Merck, St. Louis, MO, USA.

Red bell pepper (Capsicum annuum) fruits were procured from the local market in Bangalore, India, dried, deseeded and flaked. The flaked fruits were extracted with n-hexane (16), then filtered and concentrated to obtain the oleoresin. The oleoresin was further purified by supercritical extraction and subjected to alkaline saponification. The saponified extract was washed with water and purified with ethyl acetate to obtain capsanthin-enriched extract (17).

In a glass bowl, capsanthin-enriched extract, sucrose and microcrystalline cellulose were mixed geometrically, and 5% (m/V) polyvinyl pyrrolidine K-30 solution in isopropyl alcohol was added to form a coherent mass. Wet pellets from the coherent mass were prepared by using a twin-screw axial extruder (Anish Pharma Equip Pvt. Ltd., Nashik, India). The wet pellets were transferred to a spheronizer (18) and dried. Pellets were coated with sodium alginate and ethyl cellulose barrier using a fluidized bed coater with bottom spray (Anish Pharma Equip Pvt. Ltd.). The pellets were removed and cured at room temperature.

The physical, chemical and microbial analyses were performed using the methods described in the United States Pharmacopeia (USP). The micrographs of capsanthin-enriched extract and pellets, and capsaicin pellets were captured by Canon digital camera (SX160 IS; Cannon USA Inc., Melville, NY, USA). The solubility of the capsanthin-enriched pellets (1 g) in alcohol and water was analyzed using USP method 561 (19). The particle size was determined by dry sieve analysis as per the method described in USP 786 (20). In this method, 5 g of pellets passed through US sieve number 30. The loss on drying was determined by drying 1 g of pellets at 105 °C to constant mass as per USP 731 (21) procedure.

The Joint Ethical Committee on Food Additives (JECFA) method was employed to quantify capsanthin and carotenoids (22). As per the method, capsanthin was extracted using acetone, then saponified and subjected to high-performance liquid chromatography (HPLC). The HPLC chromatograph equipped with LC solution software (model LC2030C, i-Series plus; Shimadzu, Kyoto, Japan) was used. The C18 reversed-phase column with the dimensions 250 mm×4.6 mm, 5 microns was used. The binary gradient mobile phase consisted of acetone (A) and water (B). The gradient program was set at 0 to 5 min 75% B, 5 to 10 min 75–95% B, 10 to 17 min 95% B, 17 to 22 min, 95-100% B, 22 to 75 min 100-75% B. The total flow rate was maintained at 1.2 mL/min, and peaks were detected at 474 nm. The standard and the samples were prepared in acetone and sonicated for 10 min. The retention time of capsanthin in the standard was used to identify the capsanthin in capsanthin-enriched extract and pellets. Carotenoids were estimated by spectrophotometry. The UV spectrophotometer equipped with Lab solution software (model 1900i; Shimadzu) was used. A suitable quantity of pellets was dissolved in acetone and the absorbance was measured at 462 nm (22).

The elemental impurities like lead, arsenic, cadmium and mercury were estimated by using USP 233 (23) method. In this method, inductively coupled plasma-optical emission spectroscopy (ICP-OES) equipment (model 5510; Agilent Technologies Inc., Santa Clara, CA, USA) was used. The ICP-OES instrument conditions were argon plasma 15 L/min, auxiliary gas 0.9–1.0 L/min, analysis mode 500-700 kPa and nebulizer gas maintained at 0.7–1.2 L/min.

The microbiological parameters of the pellets were also assessed by USP procedures. The total plate, yeast and mould counts were determined by USP 2021 method (24) using 1 g pellets. Pathogens such as Escherichia coli, Salmonella sp. and Staphylococcus aureus were analyzed as per method USP 2022 (25) using 10 g pellets. The Pseudomonas aeruginosa analysis was performed as per USP method 62 (26) with 10 g sample.

The in vitro release profile of capsanthin-enriched extract and pellets was quantified using dissolution apparatus (model DS 8000; Lab India Analytical Instruments, Bangalore, India) and following the dissolution method described in USP 711 method (27). The experiment adopted the paddle method, with the addition of 900 mL of phosphate buffer at pH=6.8 and 1% sodium lauryl sulfate as the medium, at 100 rpm rotating speed and 37 °C, to determine the dissolution of capsanthin. About 5.0 mL of each sample were withdrawn at regular time intervals and replaced by the fresh medium. Capsanthin was quantified by the HPLC method described above.

The stability study was conducted as per ICH guidelines Q1A (R2) (28). Optimized formulations of capsanthin-enriched pellets were placed in a low-density polyethylene bag, packed in high-density polyethylene (HDPE) containers. The containers were kept in a stability chamber, and temperature ((25±2) °C) and relative humidity ((60±5) %) were maintained. The samples were analysed initially and after the end of a period of 3, 6, 9 and 12 months (28). The following parameters were examined: change in the appearance, i.e. colour, identification and content of capsanthin and carotenoids by HPLC.

A spectroscopically pure capsanthin was isolated from partially purified capsanthin-enriched extract. The separation was performed with a preparative HPLC (model 1200; Agilent Technologies Inc.). The pure capsanthin fraction was characterized by LC-MS, ¹H-NMR and ¹³C-NMR.

The preparative HPLC fraction 3 (F3) was analyzed to determine the mass by in-house developed LC-MS with a binary gradient elution using Waters LCMS Micromass ZQ quadrupole mass analyzer coupled with tandem HPLC (model e2695, Xevo G2-Xs QTof; Waters Corporation, Milford, MA, USA).

¹H and ¹³C nuclear magnetic resonance spectra (29) were recorded to determine the structure of the isolated capsanthin on Agilent-NMR (model VNMRS-400; Agilent Technologies Inc.). The samples were prepared in deuterated chloroform. Various proton environments were monitored by analysing an H-H correlation within the chemical structure.

The Institutional Animal Ethics Committee (IAEC) approved this in vivo study protocol with approval number VIP/IAEC/156/2019. In this study we implemented the measures recommended by the CPCSEA (Committee for Control and Supervision of Experiments on Animals) (30). The four-week-old male C57BL/6 mice (Hylasco, Hyderabad, India) were acclimatized for one week to a specific temperature ((22±3) °C), humidity ((30-70)±5) %) and lighting 08:00–20:00 conditions after their procurement. Polycarbonate cases were used to house the animals, and free access to drinking water and food was provided. The C57BL/6 mice were randomly divided into seven groups of ten mice each after adaptation. Group G1 (normal diet, ND) was fed with a 42% kJ fat diet, and the remaining groups (G2 to G7) were fed with a 251% kJ fat diet to induce obesity for seven weeks. The induction of obesity was confirmed by body mass increase. The capsanthin-enriched pellets were suspended in 0.5% carboxymethyl cellulose (CMC) and administered orally to groups on body mass basis: G3 (20.5 mg/kg, i.e. low dose), G4 (41 mg/kg, i.e. medium dose) and G5 (82 mg/kg, i.e. high dose) for five weeks. Groups G6 and G7 received capsaicin pellets (16 mg/kg) and orlistat (16 mg/kg) in 0.5% CMC, respectively, for five weeks. The normal diet control (G1) and high-fat diet (HFD) control (G2) mice were provided with only 0.5% CMC. The body mass and food intake were measured twice weekly during the obesity induction period and thrice weekly thereafter up to the end of the treatment.

At the end of the study, blood samples were collected for haematology, clinical pathology investigations and biomarker analysis. The haematology analyzer (model ABX Micros ESV 60; HORIBA Instruments Inc., Irvine, CA, USA) was used for haematological evaluation. Clinical chemistry parameters were estimated using the RX Daytona+ (Randox Laboratories, Kearneysville, WV, USA) instrument. Commercially available enzyme-linked immunosorbent assay kits (KinesisDx kits; Krishgen Biosystems, Mumbai, India) were used to quantify insulin, leptin, adiponectin and free fatty acids. At the end of the study, all animals were subjected to a detailed gross necropsy examination. External body surfaces, orifices and cavities such as abdominal, thoracic and cranial were included in the examination.

When the body mass was measured, the liver, abdominal adipose tissues (inguinal and white epididymal) and kidney tissues were fixed in 10% natural buffered formalin, stained with hematoxylin and eosin (Everon Life Sciences, New Delhi, India), and subjected to histopathological evaluation. Adipose tissues were observed microscopically (at 40×, BX53; Olympus Corporation, Shinjuku-ku, Japan) for any change in adipocytes. Both internal and external gross necropsy examinations were done on all the tested animals.

A one-way analysis of variance (ANOVA) was used for multiple comparisons. The effects of treatment compared to control were analyzed using the GraphPad software, v. 8.0 (31). The values were expressed as mean±S.D. (standard deviation), compared and evaluated at a 5% (p≤0.05) level.

The hydrophilic matrix formulation of capsanthin-enriched pellets was developed using microcrystalline cellulose, sucrose and polyvinyl pyrrolidine K 30 and further coated with sodium alginate and ethyl cellulose as barrier. The polymers on the exterior surface hydrate and swell, forming a protective gel layer and slowly and steadily release capsanthin. The release of capsanthin is due to either diffusion or erosion of the gel layer or the combination of both processes (32). The in vitro cumulative capsanthin release profiles of capsanthin-enriched extract and capsanthin-enriched pellets using the dissolution apparatus in the buffer at pH=6.8 are shown in Fig. S2. As can be seen, the capsanthin-enriched extract and capsanthin-enriched pellets exhibited different release patterns. The release profile of capsanthin-enriched extract was observed to be rapid, while that of capsanthin-enriched pellets occurred in a sustained release manner.

The physical and chemical stability of capsanthin-enriched pellets was evaluated as per the ICH guidelines (28). At the end of one year, no change in the physical characteristics such as degradation or change in the colour was observed. Chemical analysis revealed that no degradation in the content of total carotenoids and capsanthin was observed. In conclusion, capsanthin-enriched pellets were stable at 25 ºC and 60 % relative humidity. The stability data are presented inTable S3.

LC-MS/MS with HPLC technique was employed to determine the chemical profile of capsanthin-enriched extract. Liquid chromatogram patterns of capsanthin-enriched extract showed major peaks at their retention times of 5.88, 6.24, 6.84 and 7.11 min (Fig. S3). Furthermore, capsanthin-enriched extract showed a molecular ion peak at a retention time of 7.11 min with the fragmentation of m/z=585.30 [M⁺]⁺ (Fig. S4) and dehydrated productions ([MH-H₂O]⁺ at m/z=567, which was correlated with capsanthin from literature studies (33). In order to unequivocally prove the capsanthin identification, after isolation from capsanthin-enriched extract, the structure was confirmed using ¹H-NMR and ¹³C-NMR spectroscopy. The NMR data were consistent with those in the literature (33). The ¹H-NMR, ¹³C-NMR data and the structure of capsanthin are shown in Figs. S5a, S5b and S5c respectively.

All surviving animals were subjected to necropsy and both external and internal gross pathological examinations at the end of the study. No gross abnormality was found in the treated and control groups. One animal was found dead in the group fed HFD with capsaicin pellet during the dosing period, which was due to autolysis according to the necropsy analysis. Mice in the group receiving ND showed normal histology without any microscopic alterations. In all treated groups, mildly to moderately diffuse fatty infiltration was observed in the liver. No hepatitis, haemorrhage, hepatocyte necrosis and fat cysts were observed in the treated groups.

The histopathological scoring of the liver is presented in Table 1. Animals treated with capsanthin-enriched pellets and capsaicin pellets did not reveal any microscopic alteration in the liver, kidneys and white adipose tissues (inguinal and epididymal). Histological examination revealed that the sizes of the white adipose tissues (inguinal and epididymal) were significantly smaller in the treated group, and a significant mass reduction in inguinal white adipose tissue (37.0% lower) and epididymal white adipose tissue (43.64% lower) was noted in the group receiving a high dose of capsanthin-enriched pellets (Fig. 3). White inguinal and epididymal adipose tissues of various treated and control groups of mice are shown in Fig. 4 and Fig. 5 respectively. The larger adipocytes were observed in the HFD group than in the ND group. In the ND group, the adipocytes were normal with regular sizes. The obese mice from the HFD control group showed fat stored in adipocytes as accumulated lipid droplets that occupy most of the cytoplasm. In the HFD group, larger adipocytes were observed in the adipose tissue than in the ND group. The normal adipocyte distribution was observed in the ND group with regular sizes of cells (Fig. 4 and Fig. 5), whereas in the HFD control, the enlarged adipocytes that occupied most of the cytoplasm were noted. However, the adipocytes of the groups receiving HFD with capsanthin-enriched pellets and HFD with capsaicin pellets were smaller and comparable to the groups fed HFD with orlistat and ND.

Blood acts as a pathological reflector of the status of the animals exposed to phytochemicals and plays a vital role in their biological, pathological and nutritional status (46). The examination of haematological parameters provides the opportunity to clinically investigate the impact of highly standardized capsanthin, which has not been assessed and reported previously. We examined the clinical chemistry and haematological parameters and reported the obesity biomarkers of mice fed capsanthin-enriched pellets for the first time. There was no significant change in the haematological (Table S5) and clinical chemistry parameters (Table 2) in all control and treated groups. However, a nonsignificant reduction of the total cholesterol, LDL, HDL, triglycerides and blood glucose levels was observed in the groups treated with capsanthin-enriched pellets compared to the control group receiving HFD.