Phytoecdysteroids and Anabolic Effect of Atriplex dimorphostegia: UPLC-PDA-MS/MS Profiling, In Silico and In Vivo Models

Atriplex dimorphostegia (Saltbush) is an annual halophytic shrub that is widely distributed across various parts of Asia. The current study is the first to report the metabolites profile of the total ethanol extract of the aerial parts of A. dimorphostegia (TEAD), and its anabolic activity together with the isolated 20-hydroxyecdysone (20-HE) in orchidectomized male rats. TEAD was analyzed and standardized utilizing UPLC-PDA-ESI–MS/MS and UPLC-PDA-UV techniques, resulting in tentative identification of fifty compounds including polyphenols, steroids and triterpenoids. In addition, 20-HE was quantified, representing 26.79 μg/mg of the extract. Phytochemical investigation of TEAD resulted in the isolation of 20-HE from the ethyl acetate fraction (EFAD) and was identified by conventional spectroscopic methods of analysis. Furthermore, the anabolic effect of the isolated 20-HE and TEAD was then evaluated using in silico and in vivo models. Molecular docking experiments revealed in vitro selectivity of 20-HE towards estrogen receptors (ERs), specifically ERβ over ERα and androgenic receptor (AR). The anabolic efficacy of TEAD and 20-HE was studied in orchidectomized immature male Wistar rats using the weight of gastrocnemius and soleus muscles. The weights of ventral prostate and seminal vesicles were used as indicators for androgenic activity. Rats administered 20-HE and TEAD showed a significant increase (p = 0.0006 and p < 0.0001) in the net muscle mass compared to the negative control, while the group receiving TEAD showed the highest percentage among all groups at p < 0.0001. Histopathological investigation of skeletal muscle fibers showed normal morphological structures, and the group administered 20-HE showed an increase in cross sectional area of muscle fibers comparable to methandienone and testosterone groups at p > 0.99. A. dimorphostegia exhibited promising anabolic activity with minimal androgenic side effects.


Introduction
Since the discovery of testosterone in 1935, many derivatives have been synthesized to prolong its biological activity, rendering them orally active and developing products referred to as anabolic androgenic steroids (AAS) with maximum anabolic effects and minimum androgenic side effects [1]. Although AAS therapy has been associated with treating muscle contusion injury, hypogonadism and other chronic diseases such as cachexia, they are very common among recreational athletes [2,3]. Their prolonged administration, however, have been associated with a negative impact on fertility in both sexes related to the strong androgenic 5-α-dihydrotestosterone effect [3,4].

Characterization of Total Extract of A. dimorphostegia (TEAD) by UPLC-PDA-ESI-MS/MS
This is the first time to report the metabolic profile of TEAD utilizing Ultra-performance liquid chromatography (UPLC) coupled to photodiode array detection (PDA) and electrospray ionization (ESI) tandem mass spectrometry (MS) methods (UPLC-PDA-ESI-MS/MS) ( Table 2). Fifty compounds were tentatively identified, including phenolic acids and their derivatives, flavonoids, and their derivatives [41]. In addition, phytosterols, triterpenoids and steroids were also identified.

Characterization of Total Extract of A. dimorphostegia (TEAD) by UPLC-PDA-ESI-MS/MS
This is the first time to report the metabolic profile of TEAD utilizing Ultra-performance liquid chromatography (UPLC) coupled to photodiode array detection (PDA) and electrospray ionization (ESI) tandem mass spectrometry (MS) methods (UPLC-PDA-ESI-MS/MS) ( Table 2). Fifty compounds were tentatively identified, including phenolic acids and their derivatives, flavonoids, and their derivatives [41]. In addition, phytosterols, triterpenoids and steroids were also identified.   Compound 1 was identified as caffeoyl-hexose-deoxyhexoside with a parent ion at m/z 487.01, showing MS 2 ions at m/z 308 due to the loss of caffeic acid moiety, and at m/z 179 due to the expulsion of deoxyhexose and hexose moieties [42]. Compound 2 was identified with a [M−H] − ion at m/z 451. 12 showing fragmentation peaks at m/z 353, indicating the presence of caffeoyl-quinic acid or chlorogenic acid. The peak at m/z 191 is due to quinic acid and was identified as an unknown chlorogenic acid derivative [43,44]. Compound 3, with a molecular ion peak at m/z 448.2 and exhibiting typical fragmentation patterns of C-glycosides, was assigned isoorientin [45,46]. to the loss of a rhamnose moiety, and was assigned kaempferol rhamnoside [49]. Compound 8 has a [M−H] − peak at m/z 153 characteristic of dihydroxybenzoic acid, while the ion at m/z 108 is due to the expulsion of the carboxyl moiety [50]. Compound 9 was identified as caffeic acid, showing a molecular ion peak [M−H] − at m/z 179.01 and a fragment ion at m/z 135 corresponding to [M−H−CO 2 ] − [26,51]. Compound 10, having precursor ion peak [M−H] − at m/z 487.08, was identified as caffeoyl-hexose-deoxyhexoside. The fragment ion peak at m/z 308 is due to the loss of caffeic acid moiety and at m/z 179 is due to the expulsion of hexose and deoxyhexose [42].  [54]. Compounds 16, 42 and 48, with molecular ion peaks [M−H] − at m/z 273, were assigned tetrahydroxyflavan (afzelechin), with the diagnostic product ion at m/z 179 due to the loss of the dihydroxybenzene moiety [48]. The data on compounds 17 and 43 were in agreement with the reported data on the fragmentation pattern of kaempferol aglycone [49,55]. Compounds 18 and 45 were tentatively identified as isorhamnetin, with [M−H] − at m/z 315.2 and fragment ions at m/z 301 and 272 characteristic of isorhamnetin [26,56]. Compounds 19 and 46 with a molecular ion [M−H] − at m/z 317 were tentatively identified as myricetin. The MS 2 fragment ions at m/z 179 and 151 corresponds to the retrocyclization on the A-C ring and the loss of CO, respectively [26,56]. Compounds 20 and 47, with base peak at m/z 301, matched the data reported for quercetin [26,57].  [63]. Compound 33 showed a molecular ion peak [M+H] + at m/z 577 representing daucosterol (sitosterol-glucoside), with a fragment ion [M+H−162] + at m/z 413 due to the loss of the hexose moiety corresponding to sitosterol, and the fragment ion at m/z 369 is due to the loss of glucosyl moiety followed by dehydration [64]. a base peak at m/z 315 (tamerexetin), and was tentatively identified as tamerexetin deoxyhexosyl hexoside [20]. Compound 39 was identified as caffeoyl-coumaroyl spermidine with a molecular ion peak [M−H] − at m/z 452.4 [51]. Compound 40, with a fragment peak ion [M−H 2 O+H] + at m/z 397, was identified as β-sitosterol based on the fragments in the MS 2 chromatogram [64,67]. Compound 41 had a [M−H] − ion at m/z 519. 5 showing MS 2 peaks a at m/z 357 (pinoresinol), and the prominent ion at m/z 151 was due to the cleavage of the tetrahydrofuran ring [68]. Compound 44, with a molecular ion peak [M−H] − ion at m/z 237 and based on the spectral data, was identified as portulasoid (butoxyseptanoside) [36]. Compound 49 was identified as ursolic acid with a parent ion peak [M+H] + at m/z 457 and a fragment ion peak [M−H 2 O-COOH+H] + at m/z 393 [67,69] Compounds 50 and 52 both produced molecular ion peaks at m/z 427.2 and a base peak at m/z 218. They were identified as α-amyrin or β-amyrin, respectively [67,70]. Compound 51 with a molecular ion peak [M+H] + at m/z 481 was tentatively identified as 20-hydroxyecdysone. The MS 2 chromatogram showed sequential loss of water with a base ion peak M+H−2H 2

In Silico Modelling of Steroid Receptor Binding
The docked pose of 20-HE in the binding site of ERβ reveals a binding mode similar to that of the co-crystallized ligand, estradiol. The inhibitor occupies the hydrophobic central cavity of the binding pocket. The inhibitor is then anchored to the binding pocket via an array of hydrogen bonds. The cyclic end of the inhibitor is hydrogen bonded to the side chain of Arg346 and the backbone carbonyl of Leu339. Although the hydroxyl group of 20-HE occupies a virtually nearly identical position in space as the hydroxyl group of estradiol, it forms hydrogen bonds to different residues. The estradiol hydroxyl group is hydrogen bonded to the side chain of Glu305. Extending to the opposite side of the binding cavity, the aliphatic chain of 20-HE also contains a hydroxyl group that occupies a very similar position to that of the other hydroxyl group of estradiol. This hydroxyl group hydrogen bonds to the carbonyl of Gly472. Overall, the positioning of 20-HE is very similar to that of estradiol in the binding cavity of ERβ (Figures 3 and 4).
Looking at the binding modes of 20-HE in the cavity of ERα and AR immediately points out a critical determinant for selectivity towards ERβ ( Figure 5). The cyclic end of 20-HE clashes with the side chain of Glu52. The aliphatic chain of the ligand clashes extensively with the side chains of His524 and Leu525. This clashing is governed by the secondary hydroxyl group and the terminal tertiary hydroxyl group ( Figure 6). This suggests that despite the absence contribution of the extensive hydroxylation of the aliphatic chain of 20-HE to its binding to ERβ, the presence of these groups govern its selective binding towards ERβ over ERα indirectly. Similarly, extensive clashing with the binding site of AR inhibits 20-HE binding to AR. The ketone group of the ligand clashes with the side chain of Met780. The aliphatic chain of the ligand clashes extensively with the side chains of Gln711 and Arg752. Similar to the case of ERβ, the secondary hydroxyl group and the terminal methyl group of the ligand aliphatic chain are responsible for the ligand clashes, in addition to the methyl group of the C1 position of the aliphatic chain.

In Silico Modelling of Steroid Receptor Binding
The docked pose of 20-HE in the binding site of ERβ reveals a binding mode similar to that of the co-crystallized ligand, estradiol. The inhibitor occupies the hydrophobic central cavity of the binding pocket. The inhibitor is then anchored to the binding pocket via an array of hydrogen bonds. The cyclic end of the inhibitor is hydrogen bonded to the side chain of Arg346 and the backbone carbonyl of Leu339. Although the hydroxyl group of 20-HE occupies a virtually nearly identical position in space as the hydroxyl group of estradiol, it forms hydrogen bonds to different residues. The estradiol hydroxyl group is hydrogen bonded to the side chain of Glu305. Extending to the opposite side of the binding cavity, the aliphatic chain of 20-HE also contains a hydroxyl group that occupies a very similar position to that of the other hydroxyl group of estradiol. This hydroxyl group hydrogen bonds to the carbonyl of Gly472. Overall, the positioning of 20-HE is very similar to that of estradiol in the binding cavity of ERβ (Figures 3 and 4).
Looking at the binding modes of 20-HE in the cavity of ERα and AR immediately points out a critical determinant for selectivity towards ERβ ( Figure 5). The cyclic end of 20-HE clashes with the side chain of Glu52. The aliphatic chain of the ligand clashes extensively with the side chains of His524 and Leu525. This clashing is governed by the secondary hydroxyl group and the terminal tertiary hydroxyl group ( Figure 6). This suggests that despite the absence contribution of the extensive hydroxylation of the aliphatic chain of 20-HE to its binding to ERβ, the presence of these groups govern its selective binding towards ERβ over ERα indirectly. Similarly, extensive clashing with the binding site of AR inhibits 20-HE binding to AR. The ketone group of the ligand clashes with the side chain of Met780. The aliphatic chain of the ligand clashes extensively with the side chains of Gln711 and Arg752. Similar to the case of ERβ, the secondary hydroxyl group and the terminal methyl group of the ligand aliphatic chain are responsible for the ligand clashes, in addition to the methyl group of the C1 position of the aliphatic chain.
The C1 methyl group of the aliphatic chain of the ligand is of specific interest. It is responsible for additional clashing with the side chain of Gln711 of AR. However, it is also responsible for the only steric clash in the binding pose of 20-HE to ERβ with the carbonyl of Gly472 at 2.75 Å. Therefore, although this group potentially contributes to In conclusion, 20-HE exhibits a binding pattern to ERβ similar to the co-crystallized ligand estradiol. The flexibility and poly-substitution of its aliphatic side chain mediates its selectivity towards ERβ via extensive clashing with the residue side chains within the binding site of ERα and AR. These steric clashes negatively influence the binding stability of 20-HE in the binding site of ERα and AR, and indirectly improve selectivity towards ERβ.

Evaluation of the Cytotoxic Activity of 20-HE and TEAD (SRB-U Assay)
Before assessing the anabolic activity of 20-HE and TEAD, their cytotoxicity was evaluated against human skeletal muscle cells (SkMC) using SRB-U assay for 24 h, human SkMC was subjected to 20-HE and TEAD at concentrations of 0.01, 0.1, 1, 10 and 100 μg/mL (Figure 7). 20-HE and TEAD exhibited no effect on cell viability at concentrations equal to or greater than 0.01 μg/mL with an IC50 of > 100 μg/mL.  The C1 methyl group of the aliphatic chain of the ligand is of specific interest. It is responsible for additional clashing with the side chain of Gln711 of AR. However, it is also responsible for the only steric clash in the binding pose of 20-HE to ERβ with the carbonyl of Gly472 at 2.75 Å. Therefore, although this group potentially contributes to selectivity towards ERβ, its elimination might be advantageous as it would increase the binding stability of the ligand to ERβ. Its contribution in steric clashes with the side chain of both ERβ and AR suggests that the elimination of this would still conserve selectivity towards ERβ. Future medicinal chemistry studies can be directed towards the total synthesis of 20-HE analogues and the structure-activity relationship investigation of the effect of elimination of this methyl group on both the potency and selectivity of 20-HE.
In conclusion, 20-HE exhibits a binding pattern to ERβ similar to the co-crystallized ligand estradiol. The flexibility and poly-substitution of its aliphatic side chain mediates its selectivity towards ERβ via extensive clashing with the residue side chains within the binding site of ERα and AR. These steric clashes negatively influence the binding stability of 20-HE in the binding site of ERα and AR, and indirectly improve selectivity towards ERβ.

Evaluation of the Cytotoxic Activity of 20-HE and TEAD (SRB-U Assay)
Before assessing the anabolic activity of 20-HE and TEAD, their cytotoxicity was evaluated against human skeletal muscle cells (SkMC) using SRB-U assay for 24 h, human SkMC was subjected to 20-HE and TEAD at concentrations of 0.01, 0.1, 1, 10 and 100 µg/mL (Figure 7). 20-HE and TEAD exhibited no effect on cell viability at concentrations equal to or greater than 0.01 µg/mL with an IC 50 of > 100 µg/mL.

Evaluation of the Cytotoxic Activity of 20-HE and TEAD (SRB-U Assay)
Before assessing the anabolic activity of 20-HE and TEAD, their cytotoxicity was evaluated against human skeletal muscle cells (SkMC) using SRB-U assay for 24 h, human SkMC was subjected to 20-HE and TEAD at concentrations of 0.01, 0.1, 1, 10 and 100 μg/mL (Figure 7). 20-HE and TEAD exhibited no effect on cell viability at concentrations equal to or greater than 0.01 μg/mL with an IC50 of > 100 μg/mL.   After 10 days, there was no significant (p = 0.54) increase in total body weight among all groups. However, groups receiving methandienone and testosterone showed a significant (p < 0.0001) increase in the mass of muscle bundle compared to the negative control, both by 80%. Similarly, the administration of 20-HE and TEAD increased the net mass of muscles significantly (p = 0.0006 and p < 0.0001) compared to the negative control by 76% and 112%, respectively. There was no significant (p > 0.5) difference between 20-HE and TEAD ( Figure 8). After 10 days, there was no significant (p = 0.54) increase in total body weight among all groups. However, groups receiving methandienone and testosterone showed a significant (p < 0.0001) increase in the mass of muscle bundle compared to the negative control, both by 80%. Similarly, the administration of 20-HE and TEAD increased the net mass of muscles significantly (p = 0.0006 and p < 0.0001) compared to the negative control by 76% and 112%, respectively. There was no significant (p > 0.5) difference between 20-HE and TEAD (Figure 8).

Weight of Secondary Sex Organs
Group C receiving testosterone showed a significant (p < 0.0001) increase in the mass of prostate and seminal vesicles, a 14-fold increase in comparison with the negative control. Moreover, when comparing group C to groups receiving 20-HE and TEAD, the mass of prostate and seminal vesicles also increased significantly (p < 0.0001), by 17-folds (Figure 9).

Weight of Secondary Sex Organs
Group C receiving testosterone showed a significant (p < 0.0001) increase in the mass of prostate and seminal vesicles, a 14-fold increase in comparison with the negative control. Moreover, when comparing group C to groups receiving 20-HE and TEAD, the mass of prostate and seminal vesicles also increased significantly (p < 0.0001), by 17-folds (Figure 9).

Anabolic to Androgenic Ratios
The anabolic to androgenic ratio of each group was calculated first using the weight of the muscle bundle ( Figure 10) and then using the weight of the levator ani muscle (Figure 11 The muscle bundle to prostate and seminal vesicles ratio showed a significant difference in the group receiving TEAD when compared to the negative control and testosterone groups at p values of 0.002 and <0.0001 respectively. There is no significant (p > 0.99) difference between the group receiving TEAD when compared to methandienone, and neither is there a significant difference when compared to 20-HE (p > 0.99). When comparing the group receiving 20-HE compared to the negative control, the difference is not significant, but the difference compared to testosterone is significant at p = 0.0036. Although the group receiving total extract showed a higher ratio when compared to 20-HE, the difference was not significant (p > 0.1).

Anabolic to Androgenic Ratios
The anabolic to androgenic ratio of each group was calculated first using the weight of the muscle bundle ( Figure 10) and then using the weight of the levator ani muscle ( Figure 11

Muscle Mass Percentage
The muscle mass percentage was calculated using the mean mass of the muscle bundle of both right and left hindlimbs against the mean final body weight of rats in each group at the end of the study (Figure 12). The group receiving TEAD showed the highest percentage (0.64%) among all groups, two-fold compared to the negative control. However, the difference between the groups receiving TEAD and 20-HE was not significant. The muscle bundle to prostate and seminal vesicles ratio showed a significant difference in the group receiving TEAD when compared to the negative control and testosterone groups at p values of 0.002 and <0.0001 respectively. There is no significant (p > 0.99) difference between the group receiving TEAD when compared to methandienone, and neither is there a significant difference when compared to 20-HE (p > 0.99). When comparing the group receiving 20-HE compared to the negative control, the difference is not significant, but the difference compared to testosterone is significant at p = 0.0036. Although the group receiving total extract showed a higher ratio when compared to 20-HE, the difference was not significant (p > 0.1).

Muscle Mass Percentage
The muscle mass percentage was calculated using the mean mass of the muscle bundle of both right and left hindlimbs against the mean final body weight of rats in each group at the end of the study (Figure 12). The group receiving TEAD showed the highest percentage (0.64%) among all groups, two-fold compared to the negative control. However, the difference between the groups receiving TEAD and 20-HE was not significant. All samples stained with H&E demonstrated almost normal morphological structures of skeletal muscle fibers, without abnormal subcellular alterations with normal intact intermuscular fine connective tissue, without abnormal inflammatory cell infiltrates and with normal vasculatures (Figure 13).

Histopathological Examination with Hematoxylin and Eosin (H&E) Stain
All samples stained with H&E demonstrated almost normal morphological structures of skeletal muscle fibers, without abnormal subcellular alterations with normal intact intermuscular fine connective tissue, without abnormal inflammatory cell infiltrates and with normal vasculatures (Figure 13). The size of muscles increased (p < 0.0001) in all groups compared to the negative control ( Figure 14). While testosterone treatment showed a maximum increase in muscle fiber size (1017 ± 100 μm²), there was no significant (p = 0.67) difference when compared

Number of Muscle Fibers
The number of muscle fibers increased in all groups when compared to the negative control at p < 0.0001 ( Figure 15). The group receiving methandienone showed the highest number of muscle fibers (49.5 ± 4.5), significantly higher than the groups receiving 20-HE and TEAD at p < 0.0001. There was no significant difference (p = 0.6) between the groups receiving 20-HE and TEAD.

Number of Muscle Fibers
The number of muscle fibers increased in all groups when compared to the negative control at p < 0.0001 ( Figure 15). The group receiving methandienone showed the highest number of muscle fibers (49.5 ± 4.5), significantly higher than the groups receiving 20-HE and TEAD at p < 0.0001. There was no significant difference (p = 0.6) between the groups receiving 20-HE and TEAD.

Number of Muscle Fibers
The number of muscle fibers increased in all groups when compared to the negative control at p < 0.0001 ( Figure 15). The group receiving methandienone showed the highest number of muscle fibers (49.5 ± 4.5), significantly higher than the groups receiving 20-HE and TEAD at p < 0.0001. There was no significant difference (p = 0.6) between the groups receiving 20-HE and TEAD.

Discussion
The purpose of the current study was to evaluate the anabolic effect of A. dimorphostegia. TEAD was found to contain 20-HE (26.79 µg/mg), a phytoecdysteroid that was proven to have low acute toxicity [8] and an anabolic potency that have been well documented [6,7,19,76]. Several steroids have been employed in treating various muscle injuries and chronic diseases, as well as among athletes [2,3,16]. However, the negative impact of AAS on fertility in both sexes presents ecdysteroids as an alternative for AAS due to their anabolic potency with minimal androgenic side effects [8].
The UPLC-PDA-MS/MS profiling of TEAD highlighted the presence of a range of metabolites including polyphenols, phenolic acids, flavonoid glycosides and aglycones. Examples include coumaroyl hydroxy-palmitic acid, quercetin galloyl pentoside, acetylated kaempferol hexoxyl-rhamnoside and quercetin rhamnoside. In addition to 20-HE as the major identified compound, sterols, steroids and triterpenoids such as β-sitosterol, stigmasterol and β/α-amyrin were also identified. This comes in context with previous reports which refers to the occurrence of different types of polyphenols and steroids previously identified from different Atriplex species [36,43,45,49,50,67,73]. However, the current study is the first to report these compounds in A. dimorphostegia.
Several studies reported that polyphenols such as apigenin quercetin and phenolic acids are highly effective in attenuating muscle atrophy and enhancing muscle health. Apigenin reduced oxidative stress and inhibited hyperactive autophagy and apoptosis, alleviating age-related skeletal muscle atrophy in rats [77,78]. Kanzaki et al. [79] reported that quercetin glycosides increased the wet weights of the quadratus femoris, gastrocnemius, tibialis anterior and soleus muscles in mice. While in the tibialis anterior and soleus muscles of rats, ferulic acid increased the production of mechano-growth factor and the antioxidant enzymes [80]. Moreover, Ommati et al. [81] proved that chlorogenic acid improved muscular strength in resistance-trained rats through improving mitochondrial activity and cellular energy metabolism. Phytosterols, such as β-sitosterol, according to Verma et al., by having an activity similar to that of 5-α-reductase, the enzyme that catalyzes the conversion of testosterone to 5-α-dihydrotestosterone, may be involved in the reported anabolic effect in the form of weight gain and muscle mass [82]. Moreover, ursolic acid was found to reduce muscle atrophy and stimulate muscle hypertrophy in mice by enhancing skeletal muscle insulin/IGF-I signaling and inhibiting atrophy-associated skeletal muscle mRNA expression [83].
To understand the practical in vitro selectivity pattern of 20-HE towards ERβ over ERα and the AR, molecular docking experiments were conducted using all three receptors. In these experiments, the protein data bank (PDB) entries 3OLL, 3UUD and 2AM9 were used for ERβ, ERα and AR, respectively [84][85][86]. These crystal structures were specifically selected as they are in ligand-bound states and therefore their conformations would allow ligand docking into the ligand-binding site. The docked poses of 20-HE revealed that the steric clashes negatively influence the binding stability of 20-HE in the binding sites of ERα and AR, and indirectly improve selectivity towards ERβ. These findings were in agreement with previous studies [8,87] The anabolic efficacy of TEAD and the isolated 20-HE was evaluated using testosterone enanthate and methandienone as positive controls in orchidectomized immature male Wistar rats [6]. The groups administered 20-HE and TEAD showed a significant increase (p = 0.0006 and p < 0.0001) in the net mass of muscles compared to the negative control, while the group receiving the TEAD showed the highest percentage (0.64%) among all groups. Tóth et al. proved that 20-HE increased fiber cross-sectional area in the regenerating soleus muscle, suggesting a beneficial effect on muscle regeneration without exhibiting the androgenic effects of steroids [19]. The anabolic to androgenic ratio of groups receiving 20-HE and TEAD is very comparable to methandienone (p > 0.99). An effect even exceeding methandienone was found in vitro [6], and Chermnykh et al. reported that ecdysterone showed an anabolic effect stronger than that of methandienone [76]. A 10-week intervention study performed on young men investigated the effect of ecdysterone on human sport exercise. A significant increase in muscle mass was observed without an increase in biomarkers for liver or kidney toxicity [7].
It could be concluded that TEAD is a potential anabolic agent containing 20-HE together with other secondary metabolites which previously reported to have anabolic effects and activities that promote muscle health [7,[73][74][75][76][77][78]. Moreover, the preparation of a standardized extract of TEAD with potential anabolic activity is of low cost, is time saving and is a simple method compared to the isolation of pure 20-HE [88] Furthermore, TEAD and 20-HE did not affect the cell viability of SkMC with IC 50 (>100 µg/mL).

Materials
The

Extraction and Fractionation of A. dimorphostegia
Two kilograms of the air-dried A. dimorphostegia aerial parts were extracted by maceration in 5 L of absolute ethanol for 24 h at room temperature, followed by filtration. The same process was repeated for four more days until exhaustion. The extracts were collected and dried under reduced pressure at 45 • C yielding 100 g residue. Ten grams of the dried alcohol extract was reserved for further analysis, labeled as TEAD and freeze-dried. The remaining 90 g of the dried extract was suspended in 500 mL distilled water. Fractionation was carried out successively using solvents of increasing polarities: n-hexane followed by ethyl acetate. Each fraction was dried completely under reduced pressure to yield n-hexane (38 g) and ethyl acetate (14 g) residues, respectively. The residues were then freeze-dried for further investigations.

Isolation of 20-HE from EFAD
EFAD was subjected to column chromatography. EFAD (4 g) was dissolved in 20 mL of distilled water and chromatographed over a 160 g Diaion HP-20 (Merck, Germany) column (2.5 × 100 cm). Fractions were eluted using water and methanol (H 2 O: MeOH): 0-100% step-gradient with an increment of 20% methanol every 1000 mL. The column was flushed with 200 mL of 100% methanol for the last three fractions. Elution was monitored using TLC and (DCM: MeOH) (9:1) as the mobile phase. Eluents from 0-40%, 60-80% and 100% methanol were pooled into fractions 1, 2 and 3, respectively. Fraction 2 (60-80%) was further refined using column chromatography. The fraction weighing 1 g was dissolved in 5 mL DCM and chromatographed using 50 g Silica gel 60 (Merck, Germany) column (1.5 × 100 cm). DCM and MeOH were used as eluents with a ratio of 9.5:0.5. Elution was monitored using TLC in a solvent system of DCM and MeOH (9:1). Fractions 101-153 showed a single green band after spraying with p-anisaldehyde/H 2 SO 4 , suggesting the presence of triterpenoid skeleton. Thus, it was pooled and dried under reduced pressure at 45 • C, yielding 55 mg white crystalline solid.

NMR Spectrometrical Analysis of 20-HE
NMR spectra were conducted using a Bruker Avance 400 MHz NMR spectrometer using DMSO-d 6 as a solvent. One-dimensional 1 H and 13 C spectra were obtained using standard pulse sequences and parameters. 1 H chemical shifts (δ) were measured in ppm, relative to TMS and 13 C NMR chemical shifts to DMSO-d 6 , and were converted to the TMS scale by adding 39.49.

LC/ESI/MS Spectrometrical Analysis of 20-HE
LC-ESI-MS spectrometrical analysis of 20-HE Waters ™ ACQUITY Xevo ™ TQD system (Waters Corp., Milford, MA, USA). A C 18 column with dimensions of 2.1 mm × 50 mm, 1.7 µm (Waters, Ireland) was used. 20-HE was dissolved in methanol at a concentration of 100 µg/mL, then filtered using a 0.2 micropore filter. A gradient elution program was conducted using eluent A (0.1% formic acid in water) and eluent B (0.1% formic acid in methanol). The total run time was 32 min and a flow rate was kept at 0.2 mL/min. For mass spectrometric detection, a XEVO TQD triple quadrupole mass spectrometer equipped with an electro-spray ionization (ESI) source (Waters Corp., Milford, MA, USA) was used. Mass spectra were detected in the ESI between m/z 100-1000. The peaks and spectra were processed using the Maslynx 4.1 software (Waters, Milford, MA, USA).

UPLC-PDA-ESI-MS/MS Analysis of TEAD
The profiling of TEAD was performed on a Waters™ ACQUITY Xevo ™ TQD system (Waters Corp., Milford, MA, USA). A C 18 column with dimensions of 100 mm × 2.1 mm (p.s., 1.7 µm) (Waters, Ireland) was used. The sample was dissolved in methanol at a concentration of 1 mg/mL and was then filtered using a 0.2 µm micropore filter. The instrument settings and the gradient elution program were conducted using the method described earlier by El Zahar et al. using eluent A (0.1% formic acid in water) and eluent B (0.1% formic acid in acetonitrile) [89]. The flow rate was kept at 200 µL/min with an injection volume of 10 µL.

Standardisation of TEAD Using UPLC Analysis
Chromatographic analysis of alcohol extract was performed using an UPLC system Thermo Fisher UHPLC Dionex Ultimate 3000 (Germering, Germany). The system is equipped with a (ISO-3100SD) pump, (WPS 3000 SL) autosampler, (TCC-3000SD) column thermostat and (DAD-3000 RS) diode array detector (Germering, Germany). The software used was Chromeleon 6.8 (Germering; Germany). A column of Hypersil Gold was used with a 5 µm particle size and dimensions of 250 × 4.6 mm. The solvents used were 1% orthophosphoric acid in water (A) and acetonitrile (B). The separation was performed for a total of 70 min using gradient elution from 10% to 75% B in A for the first 35 min at a flow rate 0.5 mL/min, and then increased to 0.8 mL/min for the whole run. Then, the elution went from 75% to 100% B in A for 15 min and was kept isocratic for another 15 min. For the next 15 min, the elution was from 100% to 40% B in A and for the last 10 min kept isocratic from 10% B in A. UV detection was at 213 nm, column temperature at 30 • C and the injection volume was 20 µL. The isolated 20-HE (1 mg) was dissolved in 1 mL methanol, serial dilutions were performed using five concentrations (1, 10, 20, 40 and 80 µg/mL) and a calibration curve was generated using the same method [90].

In Silico Modelling of Steroid Receptor Binding
The molecular docking experiments were carried out using protein crystal structures obtained from the Protein Data Bank (PDB) [91]. The structures used were 3OLL, 3UUD and 2AM9 for ERβ, ERα and AR, respectively [84][85][86]. The PDB structures were prepared using the Dock Prep protocol embedded in Chimera [92]. The docked ligands were prepared using the same protocol. Molecular docking was then carried out using the AutoDock Vina extension in Chimera [93]. A docking grid was generated with dimensions 25 × 25 × 25 Å for all the docking experiments. The position of the docking grid center is (−26, 25, −12), (20,7,5) and (27,3,5) for ERβ, ERα and AR, respectively, using an (x, y, z) format. A default exhaustiveness value of 8 and an energy range of 3 kcal/mol were used for generating binding poses. The resulting poses were evaluated using Maestro Visualizer (Maestro, Schrödinger, LLC, New York, NY, USA, 2021) and figures and graphics were generated using the same software. To validate the docking protocol, the co-crystallized ligands were re-docked using the same docking parameters. The generated docking poses were identical to the co-crystallized poses for ERβ, ERα and AR with RMSD values of 0.508, 0.643 and 0.751, respectively. This indicates a valid docking protocol.

Evaluation of the Cytotoxic Activity of 20-HE and TEAD (SRB-U Assay)
The human skeletal muscle cell line was cultivated as adherent cultures on Petri dishes and kept at 37 • C. The SRB-U assay was used to assess cell viability [94,95] Three independent experiments with three internal replicates were performed. All data were blank adjusted prior to further interpretation.

Orchidectomy
Orchidectomy was performed using the technique described earlier [1]. General anesthesia was performed with the association of Ketamine (120 mg/kg) and Xylazine (16 mg/kg) administered in the same syringe as 1.0 mL. The rat was restrained in dorsal recumbency, and the surrounding scrotal area was prepared for aseptic surgery. Using a scalpel blade, a 1 cm incision was made through the skin ventrally on both sides of the scrotum. The testis was then explored and retracted to expose a section of the vas deferens and the spermatic cord's vasculature, ligated, then tied in an overhand knot. The cord was then excised, and the process was repeated on the other side. Finally, muscle and skin closure was performed with two stitches using synthetic absorbable sutures and then observed for a few hours for signs of hemorrhage. Rats were then left to recover for one week; two rats died at days 2 and 5.

Experimental Design
Twenty-three rats with an average weight of 62.25 ± 10.26 g were randomly divided into five groups and treated with test compounds by gavage and subcutaneous (SC) injection ( Figure 16). Group A received only the vehicle (20% ethanol and 80% corn oil). Groups B to E received methandienone, testosterone enanthate, 20-HE and TEAD with concentrations of 5 mg/kg, 0.1 mg/kg, 5 mg/kg and 185 mg/kg, respectively, each diluted in a solution of 20% ethanol and 80% corn oil. The rats were administered once daily with the respective substances for ten consecutive days.

Sample Collection
On day 10, rats were killed by cervical dislocation. The seminal vesicles (S. vesicles), ventral prostate and levator ani were dissected out and weighed to the nearest 0.1 mg. The seminal vesicles were squeezed before weighing to remove any fluids. The soleus and gastrocnemius muscles were then dissected together as a bundle (muscle bundle) from the right and left hind limb of rats, weighed immediately and then stored in 10% neutral buffered formalin for 72 h for further investigation.

Sample Collection
On day 10, rats were killed by cervical dislocation. The seminal vesicles (S. vesicles), ventral prostate and levator ani were dissected out and weighed to the nearest 0.1 mg. The seminal vesicles were squeezed before weighing to remove any fluids. The soleus and gastrocnemius muscles were then dissected together as a bundle (muscle bundle) from the right and left hind limb of rats, weighed immediately and then stored in 10% neutral buffered formalin for 72 h for further investigation.

Histopathological Examination and Determination of Muscle Fiber Cross-Sectional Area
Samples were trimmed and processed in serial grades of ethanol, cleared in Xylene, synthetic paraffin wax infiltration and embedded into Paraplast tissue embedding media. Then, 4 μm thick cross sections were cut by rotatory microtome. The sections were stained with Harris hematoxylin and eosin as a general tissue examination staining method. All methods of tissue samples processing and staining were as outlined before [96]. Nine random non-overlapping fields per sample tissue section were analyzed for determination of muscle fibers cross sectional area. All data and micrographs were obtained by using a Full HD microscopic camera operated by the Leica application module for tissue sections analysis (Leica Microsystems GmbH, Germany, Berlin).

Statistical Analysis
All statistical calculations and generation of graphs and the cumulative data of weight gain in rats, secondary sex organs and muscle fibers, as well as the muscle fiber's cross-sectional area, were compared by using the t-test for unpaired samples and one-way ANOVA analysis followed by Tukey-Kramer as a post hoc test. All tests were performed using GraphPad Prism version 7.00. The data are expressed as means ± SD.

Conclusions
The Atriplex dimorphostegia total ethanol extract proved anabolic potency. This was represented by the increase in net muscle mass of gastrocnemius and soleus muscle, as well as an increase in the cross-sectional area of muscle fibers. Moreover, both 20-HE and

Histopathological Examination and Determination of Muscle Fiber Cross-Sectional Area
Samples were trimmed and processed in serial grades of ethanol, cleared in Xylene, synthetic paraffin wax infiltration and embedded into Paraplast tissue embedding media. Then, 4 µm thick cross sections were cut by rotatory microtome. The sections were stained with Harris hematoxylin and eosin as a general tissue examination staining method. All methods of tissue samples processing and staining were as outlined before [96]. Nine random non-overlapping fields per sample tissue section were analyzed for determination of muscle fibers cross sectional area. All data and micrographs were obtained by using a Full HD microscopic camera operated by the Leica application module for tissue sections analysis (Leica Microsystems GmbH, Germany, Berlin).

Statistical Analysis
All statistical calculations and generation of graphs and the cumulative data of weight gain in rats, secondary sex organs and muscle fibers, as well as the muscle fiber's crosssectional area, were compared by using the t-test for unpaired samples and one-way ANOVA analysis followed by Tukey-Kramer as a post hoc test. All tests were performed using GraphPad Prism version 7.00. The data are expressed as means ± SD.

Conclusions
The Atriplex dimorphostegia total ethanol extract proved anabolic potency. This was represented by the increase in net muscle mass of gastrocnemius and soleus muscle, as well as an increase in the cross-sectional area of muscle fibers. Moreover, both 20-HE and TEAD exhibited a low androgenic effect on sex organs without increasing the weights of prostate and seminal vesicles of rats after administration for ten consecutive days. These findings indicate that A. dimorphostegia is a promising candidate for use as an anabolic agent and warrants further clinical investigation. Moreover, future study of the pharmacokinetics of TEAD and 2-HE in rats, as well as the lifespan, biodistribution, and muscle tissue retention, will be conducted to evaluate the biological effects of indicated doses.