The Hypoglycemic and Hypocholesterolemic Activity of Dioscorea deltoidea, Tribulus terrestris and Panax japonicus Cell Culture Biomass in Rats with High-Fat Diet-Induced Obesity

Obesity, and its consequences for human health, is a huge and complicated problem that has no simple solution. The constant search for natural and safe compounds with systemic action that can be used for obesity prophylactics and treatment is hampered by the limited availability and variable quality of biomass of wild medicinal plants. Plant cell biotechnology is an alternative approach for the sustainable production of vegetative biomass or individual phytochemicals with high therapeutic potential. In this study, the suspension cell biomass of the medicinal plants, Dioscorea deltoidea Wall., Tribulus terrestris L., and Panax japonicus (T. Nees) C.A. Mey, produced in 20 L and 630 L bioreactors, were tested for therapeutic effects in rat models with alimentary-induced obesity. Three-month intake of water infusions of dry cell biomass (100 mg/g body weight) against the background of a hypercaloric diet reduced weight gain and the proportion of fat mass in the obese animals. In addition, cell biomass preparation reduced the intracellular dehydration and balanced the amounts of intra- and extracellular fluids in the body as determined by bioimpedance spectroscopy. A significant decrease in the glucose and cholesterol levels in the blood was also observed as a result of cell biomass administration for all species. Hypocholesterolemic activity reduced in the line P. japonicus > D. deltoidea > T. terrestris/liraglutide > intact group > control group. By the sum of parameters tested, the cell culture of D. deltoidea was considered the most effective in mitigating diet-induced obesity, with positive effects sometimes exceeding those of the reference drug liraglutide. A safety assessment of D. deltoidea cell phytopreparation showed no toxic effect on the reproductive function of the animals and their offspring. These results support the potential application of the biotechnologically produced cell biomass of medicinal plant species as safe and effective natural remedies for the treatment of obesity and related complications, particularly for the long-term treatment and during pregnancy and lactation periods when conventional treatment is often contraindicated.


Introduction
The prevalence of obesity and the burden of obesity-related diseases are increasing worldwide [1,2]. According to WHO Global Health Observatory data, over 1.9 billion 1.
Suspension cell cultures of medicinal plants with enhanced production of active metabolites were received from All-Russian collection of plant cell cultures, Institute of Plant Physiology of Russian Academy of Sciences (Appendix A, Figures A1 and A2).
The following cell cultures were used in the study:
The analysis of content and composition of bioactive secondary metabolites in the cell cultures was described previously [50].
Cell suspensions were cultured in 20 L (T. terrestris) and 630 L (D. deltoidea and P. japonicus) bubble-type bioreactors under semi-continuous regime at 25 ± 1 • C in darkness (Appendix A, Figure A3). Cultivation conditions for each culture including medium composition, agitation, oxygen concentration, inoculum density etc. were described earlier [49,54,55]. Cell biomass was harvested from bioreactors at the point of maximum dry weight accumulation and separated from liquid medium on Nutsche filters under vacuum, then rinsed three times with distilled water. The resulting cell mass was dried on stainlesssteel shelves at 45 ± 1 • C for 18 h. This dried cell powder (Appendix A, Figure A3) was used to prepare aqueous suspensions by infusing dried biomass into purified water at a ratio of 1:20 at room temperature for 7 h [50]. These suspensions were orally administrated to laboratory animals following the experimental scheme as described below.

Laboratory Animal Husbandry
Inbred male Brown Norway rats were used in this study. The animals were kept under standard conditions in accordance with «OECD. Principles of good laboratory practice». The animals were provided with standard feed ("Complete feed for laboratory animals", LLC "Laboratorkorm", Moscow, Russia) and water (GOST 2874-82 "Drinking water") ad libitum. Access to food was restricted prior to blood collection. Before the study, animals were quarantined in a separated room for 14 days. Each animal was assigned an individual number (a mark on the tail area with a permanent indelible marker, periodically updated).

Experimental Model of Alimentary Obesity and Antiobesity Effects of Cell Biomass Preparations
The animals were randomly divided into two groups: group 1 (n = 10) was an intact group receiving standard feed (SF); group 2 (n = 50) served to model alimentary obesity and received hypercaloric feed (HF) with an excess of easily digestible carbohydrates (Table 1). Hypercaloric feed was composed of standard feed (63%) with addition of rendered beef fat (19%), sucrose (10%), and isolated soy protein (8%) [56].
After three months on HF diet, changes in the body parameters of the animals including body weight (BW), fat mass (FM), fat-free mass (FFM), total body water (TBW), the amount of intra-and extracellular fluid (ICF and ECF), and BMI were analyzed by bioimpedance method (see below) [57,58]. Formation of obesity in group 2 rats receiving HF was confirmed by a significant increase in fat mass and BMI compared to group 1 receiving SF (Table 2).
During the next 3 months, subgroup 1 (intact animals) continued to receive standard feed while animals of group 2 with developed obesity were divided into subgroups 2 to 6 ( Table 1). Subgroup 2 continued to receive HF only and served as an untreated control. Subgroup 3 received HF and daily subcutaneous injections of a reference drug liraglutide (Saxenda ©, Novo Nordisk, Denmark) (0.3 mg/kg in 1 mL 0.9% NaCl). Subgroups 4-6 received HF and phytopreparations (Table 1). Phytopreparations were administered intragastrically in concentrations equivalent to 100 mg dry cell biomass per kg animal weight once a day. The doses of liraglutide and phytopreparations used in this study were selected based on previously published recommendations [50,59,60].  The effects of the phytopreparations and liraglutide on the obese rats were evaluated after 6 months (including 3 months of obesity formation followed by 3 months of treatment with liraglutide or phytopreparations) using several parameters measured by bioimpedance spectroscopy (BIS) including fat mass, fat-free mass, total body water, the amount of intraand extracellular fluid, as well as body weight, blood glucose, total cholesterol, and daily urine output (see below).

Bioimpedance Spectroscopy (BIS), Collection, and Evaluation of Biological Material
Bioimpedance spectroscopy, a noninvasive method for the estimation of body composition [57], was performed using ImpediVet BIS1 spectroscopic system (ImpediMed, USA). The device allows accurate measurement of the total body weight, extracellular and intracellular fluid volumes, fat-free mass, and fat mass, as well as body mass index. Impe-diVet is a single-channel tetrapolar device that scans 256 frequencies from 4 to 1000 kHz. Low-frequency current passes through extracellular fluids but not through cell membranes; hence, the impedance at low frequencies relates to ECF volume. High-frequency current can travel through cell membranes and passes through both extracellular fluids and cells, thus relating to TBW. Calculation of ICF was performed by subtracting ECF from TBW [57]. Evaluation of fat mass and lean body mass is based on the difference in the water composition of muscles and adipose tissues [58]. The animals were anesthetized by a mixture of zoletil (25 mg/kg) and xylazine (10 mg/kg) and placed on a heated platform to maintain body temperature. Electrodes equipped with needles were inserted in the area of the nose, between the ears, at the base of the tail, and the coccygeal zone (Appendix A, Figure A4). Three consecutive measurements were performed on each animal with an interval of 3 s. The following constant parameters associated with the type of laboratory animals were used by default: body proportion (length to girth ratio) 1.0; body density 1.05 g/cm 3 ; hydration constant 0.732; resistance coefficients ]ρ i = 325; ρ e = 752 [59].
Blood sampling was performed in alive animals by incising the gum between the lower incisors. After fixing the rat by hand with a sharp blade, a small incision was made in the gums (mixed blood). The blood was applied to glucometer test strips and/or collected in vacuum tubes with a blood coagulation activator. Blood glucose concentration was measured using Accu-Chek Active glucometer (Roche Diagnostics, Switzerland) with an accuracy of 0.1 mmol/L.
To measure total cholesterol (TC) level, blood sampling was performed in alive animals after an overnight (10-12 h) fast. After taking blood from the gums into Vacuette vacuum tubes with a blood coagulation activator, the tubes were left for 30 min to settle, then the blood was centrifuged for 10 min at 1000 rpm, the resulting serum was separated, and then centrifuged again for 15 min at 4000 rpm. The analysis of the serum was performed using a biochemical blood analyzer (Erba xl100 (Erba Lachema, Czech Republic).
Urine was collected within 24 h in metabolic cages ensuring the separation of feces and urine. The volume of daily urine was determined using laboratory volumetric glassware.

Effect of D. deltoidea Cell Biomass Preparation on Reproductive Functions of Laboratory Rats
From three cell lines tested for obesity treatments described in Section 2.2, phytopreparation from D. deltoidea cell biomass demonstrated the most profound positive effect. This phytopreparation was further tested for its potential effects on reproductive functions of laboratory rats.
Experiments were performed on clinically healthy SPF Wistar rats. Rats of the initial colony (generation F0), males and females, weighing 220 ± 15 g were randomly selected from the stock maintained at Animal Breeding Facility of BIBCh, RAS (the Unique Research Unit Bio-Model of the IBCh, RAS; the Bioresource Collection-Collection of SPF-Laboratory Rodents for Fundamental, Biomedical and Pharmacological Studies), which has an AAALACi accreditation. Animal husbandry conditions during the whole experiment were described earlier [49].
After 14 days of adaptation, F0 rats were divided into 2 groups (n = 14), experimental and control. Rats of the experimental group received a phytopreparation of D. deltoidea cell biomass which was supplied daily at the dose of 100 mg biomass/kg through polysulfone drinkers (Techniplast, Italy) for 80 days.
Reproductive function was evaluated by assessing the fertility of F0 animals and the postnatal development of F1 generation. For fertilization, female rats (n = 7 in each group) were housed together with males in a ratio of 2:1 for 14 days. The pregnancy of one or both females confirmed the fertility of the male. Fertility was determined as a percentage of the females that became pregnant and the males with fertilization ability out of total number of females/males. The average litter size, male-to-female ratio in the litter, and the number of live and dead newborns were also determined. Pup survival rate was determined twice: (i) from day 0 to day 5 as the percentage of pups survived by day 5 out of live newborns, and (ii) from day 6 to day 25 as the percentage of pups survived by day 25 out of number of pups alive at day 6.
The postnatal development of F1 generation was evaluated during the first 30 days of life by the number of live and dead newborns, dynamics of body weight, and general physical development [61][62][63]. Rat body weight was measured using Adventurer Pro electronic technical scales (Ohaus, Parsippany, NJ, USA) with an accuracy of ±0.1 g on days 4,7,11,14,18,22, and 25 after birth.
At the end of the experiment (80 days), females of parental generation (F0) and generation F1 (30-day-old) were euthanized in the carbon dioxide chamber (VetTech, Congleton, UK), blood was collected, and general autopsy was performed.
Quantification of erythrocytes and platelets in the blood was performed on a fully automatic hematology analyzer Abacus junior vet 2.7 (Diatron Messtechnik GmbH, Aus-tria) using Diatron reagent kits. Lymphocyte (LYM), granulocyte (GRA), and monocyte (MON) contents were determined on a Guava Easy Cyte flow cytometer (Merck Millipore, Germany) by detecting cell size and granularity. Total leukocyte count (WBC) was determined using the formula: WBC = LYM + GRA + MON. The relative content of lymphocytes, granulocytes, and monocytes was determined according to the formulas: LYM/WBC*100%, GRA/WBC*100%, MON/WBC*100%.
Biochemical parameters of blood serum were analyzed on a BioChem FC-360 automatic biochemical analyzer (HTI, Farmington, MI, USA) using HighTechnology reagent kits. The antioxidant status of the blood serum was evaluated as described earlier [64]. The level of malondialdehyde (MDA), the major lipid peroxidation product, in plasma was evaluated using the thiobarbituric acid reactive substances (TBARS) assay. Catalase (CAT) activity in blood plasma was determined on a spectrophotometer SF-2000 (OCB «Spectr», Russia) according to [65]: 720 µL of 50 mM phosphate buffer (pH = 7.0) was mixed with 800 µL of 0.1% hydrogen peroxide and absorbance (D0) was measured (wavelength 240 nm, 1 cm cuvettes) relative to the control sample. Next, 20 µL of plasma were added to the tubes and the absorbance (D1) was measured after 1.5 min of incubation. In the control, 800 µL of phosphate buffer was added instead of hydrogen peroxide. Concentration of reduced glutathione and glutathione reductase activity in serum was determined on a spectrophotometer SF-2000 (OCB «Spectr», Russia) using the Ellman's reagent: 5,5 -dithiobis-(2-nitrobenzoic acid) (DTNB) dissolved in methanol. Total antioxidant activity was determined by the rate of the oxidation of the reduced form of 2,6-dichlorophenolindophenol (DCPIP) in the reaction medium using a spectrophotometer (HTI, HTI, Farmington, MI, USA) [65]. The de Ritis coefficient was calculated as the ratio of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) activity according to the following formula [66]: Autopsy of animals of all groups included a thorough examination of the external surface of the body, all orifices, intracranial, thoracic, and abdominal cavities, and their contents.
The research was approved by the bioethical commission of the V.M. Gorbatov Federal Research Centre for Food Systems of the Russian Academy of Sciences.

Statistical Data Analysis
Statistical analysis of the data obtained in the experiments on antiobesity effect of phytopreparations was performed using the GraphPad Prism 8 software package (GraphPad Software, San Diego, CA, USA). When assessing the significance of differences between the study groups, the hypothesis of the normal distribution of the trait was tested using the Kolmagorov-Smirnov test. In the case of a normal distribution, the parametric Student's t-test (p < 0.1) was used, otherwise the nonparametric Mann-Whitney tests [67] were applied at p < 0.1. Data in figures and tables are presented as means followed by standard deviations.
Statistical analysis of data on phytopreparation effect on fertility was performed using STATISTICA 10 software. Results were presented as mean ± standard deviation, minimum and maximum values, or percentiles (P25-P75) depending on the dataset. Statistical validity was calculated using one-way ANOVA with Tukey's test at p < 0.05.

Alimentary Induced Obesity Model
As mentioned above, a hypercaloric, high-fat diet with an excess of digestible carbohydrates was used to model alimentary obesity. The rats receiving HF weighed 23.64% more than the animals on the standard diet ( Table 2). After three months of receiving a hypercaloric diet, the fat mass of the animals and BMI were significantly higher, and the fat-free mass was significantly lower than in the group receiving the standard feed ( Table 2). The total amount of water in the body decreased in the animals fed with hypercaloric food (Table 2). Simultaneously, we observed a decrease in ICF and an increase in ECF in rats on HF, which reflected a tendency to the intracellular dehydration.
After six months of the experiment, changes in animal body weight and blood glucose level were analyzed, and the results for subgroups 1-3 are given in Tables 3 and 4. The final weight change was 12.57% in group №2 and −22.41% in group №3 compared to the intact animals of subgroup №1 ( Table 3). The data presented in Tables 2-4 confirm that alimentary-induced obesity was successfully reproduced in the rat model according to the selected criteria [67]. Liraglutide effectively reduced both weight gain (Table 3) and the blood glucose level (Table 4) which justified its choice as a reference drug.  In the induced obesity model, treatment with phytopreparations effectively reduced the accumulation of body weight ( Figure 1) compared to animals on HF without any treatment. However, their action was milder compared to the reference drug liraglutide.
The three-month treatment of obese animals with the cell biomass of D. deltoidea resulted in a significant decrease in the proportion of body fat mass (28.37 ± 5.49%) compared to both the control (41.5 ± 7.72%, p = 0.0007) and intact (36.23 ± 4.1%, p = 0.0034) groups ( Figure 2). Consequently, there was a significant increase in FFM in the group of obese animals that received D. deltoidea cell biomass against the background of a hypercaloric diet (73.05 ± 6.70%) relative to the control (59.20 ± 4.61%, p = 0.0001) and intact (64.95 ± 4.07%, p = 0.0001) animals ( Figure 2). Interestingly, these parameters were even better in the cell biomass-treated animals than in the animals of the intact group.

Hypoglycemic and Hypocholesterolemic Activity in Liraglutide and Phytopreparations of D. deltoidea, P. japonicus, and T. terrestris Cell Cultures
In the induced obesity model, treatment with phytopreparations effectively reduced the accumulation of body weight ( Figure 1) compared to animals on HF without any treatment. However, their action was milder compared to the reference drug liraglutide. obese animals that received D. deltoidea cell biomass against the background of a hypercaloric diet (73.05 ± 6.70%) relative to the control (59.20 ± 4.61%, p = 0.0001) and intact (64.95 ± 4.07%, p = 0.0001) animals ( Figure 2). Interestingly, these parameters were even better in the cell biomass-treated animals than in the animals of the intact group. Rats that were injected with liraglutide against the background of a hypercaloric diet also demonstrated a significant decrease in FM and an increase in FFM when compared with the control HF group: 36.07 ± 7.60% vs. 41.50 ± 7.72% (p = 0.0209) for FM and 71.62% ± 10.13% vs. 63.93% ± 7.60% (p = 0.0462) for FFM ( Figure 3). These values were close to the animals of the intact group.
The FM and FFM of the animals treated with the cell biomass of P. japonicus and T. terrestris laid in between the control (HF) and intact (SF) groups, i.e., these preparations caused a slight decrease in the fat mass accumulation of the obese animals, but the differences from the control group without treatment were insignificant ( Figure 2). The hypercaloric feed led to a notable decrease in the percentage of the total body water and ECF in rats compared to the group with the standard feed ( Figure 3). Animals that received liraglutide treatment against the background of a hypercaloric diet also showed a significantly lower TBW and ECF compared to the intact group: 46.80 ± 5.57% vs. 52.43 ± 7.42% (p = 0.0372) for TBW and 41.89 ± 4.07% vs. 49.54 ± 6.30% (p = 0.0047) for ECF, but the differences from the control (HF) group were insignificant. The amount of Rats that were injected with liraglutide against the background of a hypercaloric diet also demonstrated a significant decrease in FM and an increase in FFM when compared with the control HF group: 36.07 ± 7.60% vs. 41.50 ± 7.72% (p = 0.0209) for FM and 71.62% ± 10.13% vs. 63.93% ± 7.60% (p = 0.0462) for FFM ( Figure 3). These values were close to the animals of the intact group. the balance was shifted toward intracellular fluid accumulation. Indeed, the amount of intracellular fluid in the Dioscorea animal group exceeded the amount of extracellular fluid: 58.87% ± 4.58% vs. 41.13% ± 4.32% (Figure 3). Administration of the T. terrestris and P. japonicus cell biomass did not cause any significant changes in the TBW and fluid distribution compared to the control HF group (Figure 3). In this study, phytopreparations and liraglutide showed hypocholesterolemic and hypoglycemic effects ( Figure 4). The most pronounced effect on the glucose level in the blood serum was noted in the animal groups that received the cell biomass of D. deltoidea (2.96 ± 0.29 mmol/L) and T. terrestris (2.92 ± 0.65 mmol/L) ( Figure 4). The highest cholesterol-lowering activity was recorded for the cell biomass of P. japonicus (0.60 ± 0.015 mmol/L) in comparison with the control (p = 0.0004) and intact (p = 0.0240) groups ( Figure  4). The FM and FFM of the animals treated with the cell biomass of P. japonicus and T. terrestris laid in between the control (HF) and intact (SF) groups, i.e., these preparations caused a slight decrease in the fat mass accumulation of the obese animals, but the differences from the control group without treatment were insignificant ( Figure 2).
The hypercaloric feed led to a notable decrease in the percentage of the total body water and ECF in rats compared to the group with the standard feed ( Figure 3). Animals that received liraglutide treatment against the background of a hypercaloric diet also showed a significantly lower TBW and ECF compared to the intact group: 46.80 ± 5.57% vs. 52.43 ± 7.42% (p = 0.0372) for TBW and 41.89 ± 4.07% vs. 49.54 ± 6.30% (p = 0.0047) for ECF, but the differences from the control (HF) group were insignificant. The amount of the intracellular fluid of the animals receiving liraglutide was slightly higher than in the intact group: 58.11 ± 4.07% vs. 53.49 ± 5.13% (p = 0.0387).
The total body water content in the animals receiving the cell biomass of D. deltoidea was significantly higher compared with the control group: 51.92% ± 8.08% vs. 43.76% ± 4.18% (p = 0.0161) and similar to the intact group ( Figure 3). In addition, both the ICF and ECF values in the Dioscorea-treated rats were slightly different from the intact group, and the balance was shifted toward intracellular fluid accumulation. Indeed, the amount of intracellular fluid in the Dioscorea animal group exceeded the amount of extracellular fluid: 58.87% ± 4.58% vs. 41.13% ± 4.32% (Figure 3). Administration of the T. terrestris and P. japonicus cell biomass did not cause any significant changes in the TBW and fluid distribution compared to the control HF group (Figure 3).
In this study, phytopreparations and liraglutide showed hypocholesterolemic and hypoglycemic effects (Figure 4). The most pronounced effect on the glucose level in the blood serum was noted in the animal groups that received the cell biomass of D. deltoidea (2.96 ± 0.29 mmol/L) and T. terrestris (2.92 ± 0.65 mmol/L) (Figure 4). The highest cholesterollowering activity was recorded for the cell biomass of P. japonicus (0.60 ± 0.015 mmol/L) in comparison with the control (p = 0.0004) and intact (p = 0.0240) groups (Figure 4).  The daily urine output in the intact and HF groups recorded before drug administration (3 months after the beginning of different feeding) ranged from 1.6 mL to 5.2 mL (indicated in Figure 5 by two horizontal lines). At the end of the experiment (after 3 months of treatment), an increase in the daily urine output was observed in the animals that received the cell biomass of D. deltoidea, T. terrestris, and P. japonicus compared to the control, intact, and liraglutide-receiving groups ( Figure 5). The daily urine output of the rats treated with the P. japonicus cell biomass was more than two times higher than in the intact group and added up to 10.3 ± 2.2 mL. In the groups treated with the D. deltoidea and T. terrestris cell biomass, the daily urine volume was nearly similar (6.8 ± 3.1 mL and 6.1 ± 2.7 mL, respectively). In the rats that were injected with liraglutide against the background of a hypercaloric diet, no statistically significant change in the daily urine output (4.1 ± 2.0 mL) in comparison with the control and intact groups was observed ( Figure 5). The daily urine output in the intact and HF groups recorded before drug administration (3 months after the beginning of different feeding) ranged from 1.6 mL to 5.2 mL (indicated in Figure 5 by two horizontal lines). At the end of the experiment (after 3 months of treatment), an increase in the daily urine output was observed in the animals that received the cell biomass of D. deltoidea, T. terrestris, and P. japonicus compared to the control, intact, and liraglutide-receiving groups ( Figure 5). The daily urine output of the rats treated with the P. japonicus cell biomass was more than two times higher than in the intact group and added up to 10.3 ± 2.2 mL. In the groups treated with the D. deltoidea and T. terrestris cell biomass, the daily urine volume was nearly similar (6.8 ± 3.1 mL and 6.1 ± 2.7 mL, respectively). In the rats that were injected with liraglutide against the background of a hypercaloric diet, no statistically significant change in the daily urine output (4.1 ± 2.0 mL) in comparison with the control and intact groups was observed ( Figure 5).
In conclusion, phytopreparation from the Dioscorea deltoidea cells was the most efficient in normalizing the fat mass and liquid balance in animals with induced obesity. All three types of phytopreparations tested were comparable with the reference drug liraglutide in reducing the glucose and cholesterol level in the blood. Administration of cell biomass also increased urine output. Based on these results, the phytopreparation of D. deltoidea was selected for further analysis of its potential effect on the reproductive functions and antioxidant activity in rats. In conclusion, phytopreparation from the Dioscorea deltoidea cells was the most efficient in normalizing the fat mass and liquid balance in animals with induced obesity.
All three types of phytopreparations tested were comparable with the reference drug liraglutide in reducing the glucose and cholesterol level in the blood. Administration of cell biomass also increased urine output. Based on these results, the phytopreparation of D. deltoidea was selected for further analysis of its potential effect on the reproductive functions and antioxidant activity in rats.

The Analysis of Possible Toxic Effects of D. deltoidea Phytopreparation on Reproductive Functions
The fertility of both females and males in the control and experimental groups was 100%. No birth-related mortality was observed in the F0 females of both groups.
A comparative analysis of the F0 rats mating results is presented in Table 5. The total number of newborn pups in the experimental group was 12% lower compared to the control group. The experimental group showed a lower variation in the average litter size (7-9 pups) compared to the control group (4-11 pups). The ratio of males and females in the experimental group was nearly equal while in the control group the number of males was 1.7 times higher than females. In both groups, the variations in reflex development were typical for the rats of this line ( Table 5).
Analysis of the weight dynamics in the F1 generation ( Figure 6) revealed that the pups of the experimental group whose parents received the phytopreparations of D. deltoidea gained weight more intensively than the animals of the control group during the whole period of observation. This tendency was the most pronounced from the 4th to 7th day and by the 25th day. The increase in body weight during this period in the animals of the experimental group reached 49% in comparison with 36% in the control group.

The Analysis of Possible Toxic Effects of D. deltoidea Phytopreparation on Reproductive Functions
The fertility of both females and males in the control and experimental groups was 100%. No birth-related mortality was observed in the F0 females of both groups.
A comparative analysis of the F0 rats mating results is presented in Table 5. The total number of newborn pups in the experimental group was 12% lower compared to the control group. The experimental group showed a lower variation in the average litter size (7-9 pups) compared to the control group (4-11 pups). The ratio of males and females in the experimental group was nearly equal while in the control group the number of males was 1.7 times higher than females. In both groups, the variations in reflex development were typical for the rats of this line (Table 5). Analysis of the weight dynamics in the F1 generation ( Figure 6) revealed that the pups of the experimental group whose parents received the phytopreparations of D. deltoidea gained weight more intensively than the animals of the control group during the whole period of observation. This tendency was the most pronounced from the 4th to 7th day and by the 25th day. The increase in body weight during this period in the animals of the experimental group reached 49% in comparison with 36% in the control group.  Analysis of the hematological test panel revealed no significant changes between the control and experimental groups of the F0 (parent) and F1 rats. The only significant Analysis of the hematological test panel revealed no significant changes between the control and experimental groups of the F0 (parent) and F1 rats. The only significant increase was observed in the leukocyte content in the F0 rats, mostly due to lymphocytes (up to 30%), while the content of immature leukocyte forms, granulocytes and monocytes, remained at the control level (Table A1). Among the indices of the functional activity of erythrocytes (Table A2), a slight decrease was recorded in hematocrit while the concentration of hemoglobin in erythrocytes increased (up to 4%). No statistically significant changes between the groups were found with regard to the platelet component (Table A2).
The comparative analysis of the blood serum biochemistry (Table A3) revealed that the administration of D. deltoidea caused a significant increase in alanine aminotransferase activity (by over 15%), gamma-glutamyltranspeptidase (up to 40%), and urea (by 10%) in the F0 parent rats compared to the control. Minor increases in the activity of aspartate aminotransferase, alkaline phosphatase, and lactate dehydrogenase (up to 10%) were statistically insignificant. At the same time, the de Ritis ratio also did not significantly differ between the groups. In the experimental F1 rats, a nonsignificant increase in the total protein content (by 7%, mostly due to the albumin fraction) and a significant increase in creatinine (by 11%) were recorded compared to the F1 control group (Table A3). Minor insignificant decreases in aspartate aminotransferase and alkaline phosphatase activity (to 15% and 7%, respectively) were also detected.
The analysis of the antioxidant potential of the blood (Table 6) demonstrated a significant decrease in the total antioxidant activity of the blood (by 7%) in the rats of the F1 experimental group alongside an elevated malondialdehyde content (over 10%), reduced glutathione (up to 30%), and catalase activity (up to 40%). In the F1 generation, a significant increase in the total blood antioxidant activity (by 15%) and reduced glutathione (up to 40%) were detected in the experimental group compared to the control. It should be noted that the observed fluctuations in the hematological and biochemical blood parameters were within the physiological norm for this animal type.
Based on the results of the autopsy, macroscopic examination, and analysis of the relative weight of the internal organs, no differences were found between the animals of the experimental and control groups of the F0 and F1 generations (Table A4).

Discussion
Bioreactor-grown plant cell cultures are frequently acknowledged as a sustainable and renewable source of vegetative biomass rich in valuable phytochemicals. However, the phytochemical composition of such biomass may be altered compared to their respective donor plants. This, for example, was the case in the Digitalis lanata cell culture which contained a complex mixture of phenylethanoid compounds, including digiciliside A, digiciliside B, maxoside, purpureaside E, and their methyl derivatives and isomers, as well as seven furostanol glycosides with aglycones tigogenin and gitogenin but not cardiac glycosides that are major bioactive metabolites of the plants [68]. Phytoecdysteroid composition in the cell cultures of Ajuga turkestanica was narrowed compared to hairy roots or whole plants while a number of phenylpropanoids were detected in callus cultures [69,70]. Significant differences in the composition of phenolic compounds were revealed between the in vitro cell cultures and plant roots of Phlojodicarpus sibiricus, an endangered endemic species of Eastern Siberia [71]. The ratio of furostanol and spirostanol-type glycosides differed between the plant rhizomes and several cell culture strains of Dioscorea deltoidea [72]. The cell culture of Panax japonicus used in the present study accumulated high levels of "acidic" ginsenosides (malonyl derivatives of dammarane ginsenosides and glycosides of oleanolic acid) in contrast to the plant roots where neutral ginsenosides are usually prevailing [73]. Phenylethanoids are major compounds in the cell culture of Tribulus terrestris, but only trace amounts of these chemicals are usually found in plants [74]. Therefore, there is a nonzero chance that cell biomass will show different biological actions compared to what was reported for plants of the same species. In view of these differences, the biological activity cell-culture-based phytopreparations or their crude extracts should be thoroughly investigated before they could be recommended for use.

Antiobesity Effects of Cell Biomass Preparations of D. deltoidea, T. terrestris, and P. japonicus
A hypercaloric, high-fat diet with an excess of easily digestible carbohydrates was used to model alimentary obesity. This diet affects fat and carbohydrate metabolism in animals, in particular contributing to the development of liver steatosis.
There are many experimental models of obesity: genetic, neuroendocrine, surgical, diet-induced, and others. However, the diet-induced obesity is believed to most closely resemble the obesity in humans based on both etiology and developmental mechanisms. Among diet-induced obesity models, high-fat, high-carbohydrate and combined highfat and high-carbohydrate models are the most commonly used [75,76]. High-fat and high-carbohydrate models are not very effective, and obesity develops slowly on highcarbohydrate/low-fat models [77]. Combined diets ("cafeteria diet", "western diet") do not always allow clear differentiation between the obesity-induced disorders and disorders caused by the deficiency of vitamins and minerals [78,79]. Therefore, in this study, a hypercaloric, high-fat diet was used to induce obesity [78,80]. Animal fats used in such diets are rich in saturated fatty acids that usually cause a more rapid and profound weight gain than the same amounts of polyunsaturated fats [77,81]. Male rats of the wild type are often used in the obesity studies [82,83] since they are omnivores like humans, have similar taste buds, food identification, and digestion systems. Neuroanatomically, they are also close to humans in terms of the brain areas that control food intake [84]. According to several researchers, diet-induced models on the wild rat type are closer to the pathogenesis of human obesity than models on genetically modified animals which can develop obesity on conventional diets [84,85].
In our study, the bioimpedance analysis clearly demonstrated the developed differences in the body composition between rats receiving hypercaloric or standard feed. The BIS method is relatively noninvasive and shows excellent reproducibility when assessing the body condition in dynamics, which is particularly important while modeling alimentary obesity in vivo [86]. There was a direct significant relationship between the excess body weight and fat mass. According to the results of our study, there was an increase in the fat mass and, consequently, in the body weight of the animals receiving a hypercaloric diet. The weight gain was mostly due to an increase in the fat mass and less due to an increase in the muscle mass. As obesity progressed, the animals receiving the hypercaloric feed had a decreased total body water (by ca. 20%) compared to the animals on standard feed, mostly due to intracellular dehydration.
Liraglutide was chosen as a reference drug due to the absence of the similarly effective herbal remedies for obesity treatment. Liraglutide also has an advantage over orlistat and sibutramine due to its pleiotropic effects, including hypoglycemic action, proven safety profile, and absence of excitatory activity. Liraglutide administered at 200 mg/kg twice a day significantly reduced the blood glucose, serum total cholesterol, TG, and lowdensity lipoprotein (LDL) cholesterol. It also reduced the intensity of lipid deposition in the liver in the db/db mice receiving a hypercaloric diet [87]. Bugáňová et al. [88] reported that liraglutide treatment reduced glucose levels by 1.8 times and triglycerides by 28% against a background of fasting compared to the control group of mice. Similarly, in the present study, liraglutide reduced the cholesterol content in the blood serum of the obese animals by 36% and glucose levels by 42% compared to the animals without treatment. Hansen et al. [89] found that liraglutide significantly reduced the body weight of the male Sprague Dawley rats with developed alimentary-induced obesity. Similar to our study, the drug promoted a decrease in the fat mass in rats on a hypercaloric diet [89]. In the study of Bugáňová et al. [88], liraglutide demonstrated the ability to reduce the body weight of animals with HCD-induced obesity. In our study, liraglutide contributed to a 19.5% reduction in the fat mass in Brown Norway rats with induced obesity.
As discussed earlier, the effectiveness of plant cell biomass extracts against induced obesity was assessed using bioreactor-produced cell cultures of the medicinal plants Panax japonicus, Dioscorea deltoidea, and Tribulus terrestris with bioactive secondary metabolites: triterpene and steroid glycosides. Plant extracts of these species show a wide range of biological actions, including reduction in body weight and appetite suppression, effects on leptin, ghrelin, and adiponectin levels, inhibition of pancreatic lipase activity which prevents digestion and absorption of fats and carbohydrates, reduction in the glucose, total cholesterol, and triglyceride content in the blood, etc. [35][36][37][38][39][40][41][42][43][44][45][46][47][48]. The weight-reducing ability of T. terrestris plant extracts has been reported [43]. Although the effect of D. deltoidea plant extracts on weight during obesity has never been explored, Jeong et al. [90] reported the antiobesity effect of the extract from the closely related species, D. oppositifolia, on dietinduced obese mice. The authors observed a decrease in total body weight and parametrial adipose tissue weight, as well as a decrease in total cholesterol, triglyceride level, and LDLcholesterol in the blood serum. An earlier study of the methanolic extract of D. nipponica powder demonstrated its effectiveness against body and adipose tissue weight gains in rodents induced by a high-fat diet [47]. Lee et al. [91] showed that the intake of fresh leaf extracts (3.3 mg/kg/day) or dried leaf extracts (3.3 mg/kg/day) of P. japonicus caused a slight decrease in body weight in obese rats, but the effect was statistically insignificant. In the present study, no significant changes in body weight were observed in animals with induced alimentary obesity who received aqueous extracts of biomass of the cell cultures compared to the other groups. However, the intake of D. deltoidea cell biomass infusion led to a significant decrease in the body fat mass (up to 27%) of the obese animals and a significant increase in the fat-free mass (up to 11%).
All three preparations of cell biomass showed hypocholesterolemic and hypoglycemic effects that were most pronounced in the animal groups receiving D. deltoidea and T. terrestris extracts. These data are in agreement with the literature data. For example, T. terrestris plant extracts were shown to significantly reduce the level of the total cholesterol and glucose in the blood [41,44]. Similar effects were observed for the extracts of various species of Dioscorea also containing steroid glycosides. Several animal studies have shown the antilipemic effects of sapogenin and diosgenin-rich extracts of Dioscorea species on hypercholesterolemic animals such as mice and rats, thus resulting in the reduction in the concentrations of blood cholesterol [92]. Another study demonstrated constant improvement in the cholesterol profile of the liver and plasma of mice fed with 50% raw lyophilized yam for 21 days [93]. A dose-dependent glucose-lowering effect on glucose during type 2 diabetes was observed for P. japonicus [94].

Effect of D. deltoidea Cell Biomass Preparation on Reproductive Functions of Laboratory Rats
Our study demonstrated that the aqueous extract of D. deltoidea cell culture was the most promising for the complex treatment of obesity. Therefore, an additional series of experiments was performed to assess the possible toxic effect of this biotechnologically produced substance on the reproductive functions and offspring development in laboratory rats. Administration of D. deltoidea cell biomass at the dose of 100 mg/kg to rats during the period of mating and nursing had no toxic effect on their reproductive functions: the fertility of rats remained 100% and no mortality was observed. At the same time, longterm intake of D. deltoidea cell biomass can affect liver function, while de Ritis coefficient values indicate the absence of hepatocellular damage to the liver [95]. Such effects require further research.
After the completion of the lactation period, the health status of both the F0 females and pups was investigated by means of clinical, hematological, biochemical, and pathological anatomical studies. A minor positive effect of D. deltoidea cell preparation on blood immune activity was observed as reflected by an increase in the leukocyte and lymphocyte content and hemoglobin content in erythrocytes. A slight shift in the antioxidant system activity in the blood of the F0 animals was also observed, which to some extent was compensated by the proportional increase in glutathione (up to 30%) and catalase (up to 40%) [96,97].
Comparative analysis of the postnatal development data of the F1 offspring from the animals of the experimental parental group consuming D. deltoidea phytopreparation revealed a high survival rate, even number of cubs in a litter, and more even distribution of females and males than in the control group, as well as a more intensive weight gain (body weight gain by day 30 was 49% against 36% in the control). The experimental animals showed an increase in total protein and albumin contents (by 7% and 5%) and a statistically significant creatinine increase up to 11%. As for the blood antioxidant potential, there was an increase in total antioxidant activity of 15% and reduced glutathione up to 40%. Taken together, these data may indirectly indicate a more active protein assimilation in the animals of the experimental groups.

Conclusions
This study presents the results of the preliminary screening of the antiobesity activity of the phytopreparations based on the biotechnologically produced cell cultures of D. deltoidea, T. terrestris, and P. japonicus. Cell culture in vitro is not a full equivalent of the whole plant and, as such, requires thorough investigation and clinical trials before it could be recommended for human treatment.
A comparative evaluation of cell culture extracts in relation to the reference drug liraglutide in the model of alimentary obesity indicated that phytopreparations were efficient in reducing the body fat mass and restored the intracellular-to-extracellular fluid balance in the obese animals. The positive effects were decreasing in the line D. deltoidea > liraglutide > intact group > T. terrestris > P. japonicus > control group (no treatment).
In addition, the aqueous extracts of the cell cultures showed hypocholesterolemic activity which could be ranged by efficiency in the line P. japonicus > D. deltoidea > T. terrestris/liraglutide > intact group > control group. The hypoglycemic effect was most prominent for the cell biomass of D. deltoidea and T. terrestris and decreased in the order T. terrestris > D. deltoidea > P. japonicus > liraglutide > intact group > control group. Cell preparations influenced the daily diuresis of the animals which increased with the treatment with P. japonicus > T. terrestris > D. deltoidea > liraglutide > intact group > control group. Phytopreparation of P. japonicus significantly increased the daily diuresis compared to the control level.
Based on the results presented, the water extract of the D. deltoidea cell biomass was the most efficient for the obesity treatment and its positive effects sometimes exceeded those of the reference drug liraglutide.
An additional safety assessment of D. deltoidea cell phytopreparation showed no toxic effect on the reproductive function of the animals and their offspring. Therefore, this substance has great potential for the treatment of obesity in women of fertile age.
The experiments presented here should be expanded to test the functional role of biotechnologically derived phytoproducts in other models of obesity to further investigate their effects on induced metabolic disorders. We hope that the results of this study will stimulate a wider application of plant cell cultures as safe and effective supportive remedies for the treatment of obesity and obesity-related complications, particularly during the longterm treatment.     * significantly different from the control group at p < 0.05 1 according to [99]. * significantly different from the control group at p < 0.05 1 according to [99].