Rutin Isolated from Chrozophora tinctoria Enhances Bone Cell Proliferation and Ossification Markers

Osteoporosis is a chronic disease in which the skeleton loses a weighty proportion of its mineralized mass and mechanical pliability. Currently available antiosteoporotic agents suffer adverse effects that include elevated risk of thrombosis and cancer. Phytochemicals may constitute a safer and effective option. In the current work, six flavonoids were obtained from Chrozophora tinctoria and identified as amentoflavone (1), apigenin-7-O-β-d-glucopyranoside (2), apigenin-7-O-6′′-E-p-coumaroyl-β-d-glucopyranoside (3), acacetin-7-O-β-d-[α-l-rhamnosyl(1→6)]3′′-E-p-coumaroyl glucopyranoside (4), apigenin-7-O-(6′′-Z-p-coumaroyl)-β-d-glucopyranoside (5), and rutin (6). An extensive review of the literature as well as NMR and mass spectral techniques was employed in order to elucidate the compound structures. Proliferation was enhanced in MCF7, MG-63, and SAOS-2 cells after exposure to subcytotoxic levels of the tested flavonoids. Rutin was chosen for subsequent studies in SAOS-2 cells. Rutin was not found to cause any alteration in the index of proliferation of these cells, when examining the cell cycle distribution by DNA flowcytometric analysis. Rutin was, however, found to increase osteocyte and osteoblast-related gene expression and lower the expression of RUNX suppressor and osteoclast genes. When examining the influence of rutin on vitamin D levels and the activity of alkaline phosphatase enzyme, it was found to enhance both, while decreasing acid phosphatase which is a marker of osteoporosis. Thus, rutin enhances proliferation and ossification markers in bone cells.


Introduction
Osteoporosis is a chronic disease in which the skeleton loses a weighty proportion of its mineralized mass and mechanical pliability [1]. It occurs when bone resorption surpasses bone formation, causing an imbalance [2]. As a result, the bones tend to become more fragile and more susceptible to fractures [3]. Studies have shown that 50% of women and 20% of men are likely to have a fracture resulting from osteoporosis during their lifetime [4]. Such fractures impose a heavy health and economic burden worldwide [5,6]. The risk of developing osteoporosis has been shown to be directly linked to diet. Studies have reported that people eating healthy diets with a high fruit and vegetable content tend to have lower bone resorption than their counterparts eating poor diets rich in processed foods [7]. Pharmacological management of osteoporosis involves the use of bisphosphonates and estrogen replacement therapy. However, these medicines suffer adverse effects that may range from gastric irritation to increased thromboembolic and cancer risks [8][9][10]. Therefore, it is imperative that we look for safer and effective alternatives. In this regard, medicinal herbs and plant-derived molecules have gained wide acceptance by the public and scientific communities [11].
Flavonoids, which are widely found in fruit and vegetables, are bioactive polyphenols with anti-inflammatory and antioxidant properties. Bone health has been associated with the intake of flavonoids. Intake of flavonoids increases bone mass density (BMD) in the neck and spine and decreases bone resorption in perimenopausal women [12]. Moreover, catechins and flavanones were found to associate with markers of bone resorption negatively. At the hip and spine, anthocyanins were found to be strongly linked with BMD [13]. It has been postulated that the reduction of low-grade inflammation and oxidative stress by flavonoids is the hallmark of protecting bone loss. In addition, flavonoids are thought to promote the upregulation of signaling pathways that increase the activity of osteoblasts [14].
Chrozophora tinctoria (L.) A. Juss. has several other names including turnsole, Dyer's croton or giradol. It is an annual plant that belongs to the Euphorbiaceae family [15]. Experimentally, it exhibited antioxidant and wound healing effects [16,17]. It exerts pronounced anti-inflammatory activities which involve inhibition of TNF-α, PGE 2 , IL-1β, and IL-6 [18]. Phytochemically, this plant is rich in biflavones, such as amentoflavone in addition to many flavonoids including apigenin, rutin, quercetin, and acacetin [15,[18][19][20]. The current work aimed at isolating, identifying, and assessing the activity of major flavonoids isolated from C. tinctoria on markers of ossification and proliferation of bone cells.

Phenolic Compound Extraction.
The isolation process was performed as previously reported [18]. In brief, two kilograms of the aerial parts of C. tinctoria were dried and methanol was used as an extraction solvent till exhaustion to give a 150 g residue. The total extract was suspended in the least amount of water and extracted with chloroform leaving flavonoid-rich mother liquor that was separated using a Diaion HP-20 column starting with water up to 100% methanol to give three fractions (A-C). Fraction A was free from any phenolic compounds. Silica gel column chromatography was employed to separate fraction B (50 × 5 cm, 180 g). CHCl 3 : MeOH was employed with gradient elution resulting in three fractions, I, II, and III. The first fraction (0.5 g) was separated on CC Sephadex LH-20 using the eluent MeOH to give compound 1 (50 mg). The second fraction (1.5 g) was subjected to chromatography with reversed phase Silica gel 100 C 18 -column and MeOH : H 2 O, 3 : 7 as an eluent to give compound 2 (40 mg). The third fraction (2 g) was repeatedly fractionated on Sephadex LH-20 using MeOH as an eluent; followed by CC on reversed phase Silica gel 100 C 18 using a system of MeOH : water, 3 : 7; and finally purification was performed on HPLC using a Zorbax SB-C18 column (9.4 × 250 mm), flow rate 5 ml/min to give to compounds 3 (20 mg), 4 (35 mg), and 5 (45 mg). Fraction C was chromatographed on Sephadex LH-20 using MeOH as an eluent to give compound 6 (20 mg).  2.6. Assessment of Cytotoxicity. The cytotoxic effects of the compounds obtained from C. tincturia were examined in MCF-7, SAOS-2, and MG-63 cells using SRB assay as described in the previous work [21]. Trypsin-EDTA (0.25% w/v) was utilized to briefly detach exponentially growing cells. Subsequently, cells were transferred to 96-well plates (10 3 cells per well). The test compounds were applied for 72 h to the cells. The cells were then fixed using TCA (10% w/v) at 4°C for 1 h. Subsequently, the cells were washed three times and SRB solution (0.4% w/v) was added in a dark room for 10 min after which glacial acetic acid (1% v/v) was applied for a final wash. Cells were dried and the SRB-stained cells were dissolved by Tris-HCl (0.1 M). A microplate reader was utilized to record the color intensity at 540 nm.

Determination of Doubling Time Using the Proliferation
Assay. SRB assay was employed to calculate the doubling time (as a measure for the proliferative effect) of MCF-7, SAOS-2, and MG-63 cells in the presence of and absence of incubation with the compounds isolated from C. tincturia. A subcytotoxic concentration (1 μM) of the flavonoids obtained from C. tincturia was shortly applied to cells growing exponentially in media free of phenol red for 96 h. SRB solution was used to stain cells in order to carry out their quantification and to calculate doubling time using the best fit linear regression analysis curve [22].

Cell Cycle Distribution Study.
In order to determine the effects of rutin on the cell cycle distribution, 1 μM rutin was applied to SAOS-2 cells for 48 h and a comparison was made relative to 0.1 μM estradiol. Trypsin was used to detach and collect the cells, after which PBS was added to wash the cells twice at 4°C. The cells were then resuspended in PBS (0.5 ml). Ethanol (ice-cold 70% v/v) (2 ml) was applied while shaking. This was followed by an incubation of the cells for 1 h at 4°C to allow fixation to occur. Cells were then analyzed, washed, and resuspended in PBS (1 ml) containing 50 μg/ml of RNAase-A and 10 μg/ml of PI. The cells were kept in the dark (20°C) for 20 minutes and the DNA contents were determined. After being passed into an ACEA Novocyte™ flow cytometer (ACEA Biosciences Inc., San Diego, CA, USA), the cells were analyzed using a FL2 signal detector (λex/em 535/617 nm) for PI positive events. Phases of the cell cycle were determined by ACEA NovoExpress™ software (ACEA Biosciences Inc., San Diego, CA, USA) for every sample (12,000 events per sample) subsequent to defining cell fragment-free fluorescent gate. Cells in the supra-G 2 /M phase were identified using ungated events. Total cells in S-and G 2 /M phases were divided by cells in G 0 /G 1 phases in order to calculate the proliferation index.
2.9. Gene Array Studies. Expression of osteogenic-and osteolytic-related genes was studied by exposing the cells (1 × 10 6 ) to either 1 μM rutin or 0.1 μM estradiol for 48 h. The latter served as a positive control. RNeasy Mini Kit® (Qiagen Inc., Valencia, CA, USA) was used for RNA extraction. cDNA was obtained by reverse transcription with a cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). PCR was performed in real time on the cDNA via GeneQuery™ Human Osteogenic Differentiation qPCR Array (Science Cell Research Laboratories Inc., Carlsbad, CA, USA) according to the manufacturer's instructions [23]; β-actin was selected as the housekeeping gene. The formula: 2 −ΔΔCq was employed to detect the normalized change in the gene expression of the studied genes after treatment with rutin and estradiol. cells for 48 h. The latter served as a positive control. After collecting the media, assays of vitamin D, alkaline phosphatase (ALP), osteocalcin (OC), and acid phosphatase (ACP) were performed. A direct HTS-ready colorimetric kit (Abcam, Cambridge, UK) was used to determine ALP as well as tartrate-resistant ACP [24]. Uscan® immunoassay ELISA Kit (Life Science Inc., Wuhan, China) was utilized to detect osteocalcin [25]. Human Total 25-OH Vitamin D IVD ELISA Kit (R&D Systems, Inc., Minneapolis, MN, USA) was used to detect vitamin D in its active form.
2.11. Statistical Analysis. Data are presented as mean ± SD. Analysis of variance (ANOVA) and Tukey's post hoc test were used to calculate statistical significance by SPSS® for Windows, version 17.0.0. p < 0 05 was regarded as being statistically significant.
The cytotoxicity of these compounds was studied using the SRB viability assay in various estrogenic cell lines, namely, MCF-7, SAOS-2, and MG-63 cells. We have previously used these cell lines to study the effects of paradol on A. meleguea seeds, where we noted an accelerated proliferation [27]. In the present study, when looking at the effects on MCF-7 cells, the compounds showed a maximum of 20% alteration in the viability of cells, after being applied to the compounds for 72 h in 10 μM concentrations. Exposing the cells to a higher concentration (100 μM) of amentoflavone for 72 h (1) lead to a 78.8% viability of the control untreated cells (Table 1). As in the MCF-7 cells, exposure of SAOS-2 cells to the compounds at 10 μM concentration for 72 h showed a maximum of 20% change in viability. However, increasing the concentration of the compounds 1, 3, 5, and 6 to 100 μM induced a viability drop down to 70.1%, 66.2%, 77.1%, and 76.2% of the untreated control cells, respectively (Table 1) (Table 1). It has been previously reported that several flavonoids and related phytochemicals interact with estrogen receptors [36,37]. The influence of a subcytotoxic concentration (1 μM) from all compounds under investigation on doubling times of MCF-7, SAOS-2, and MG-63 cells was tested. A concentration of 0.1 μM E 2 served as a positive control. Compounds 1, 3, and 6 caused a significant lowering of the doubling times of MCF-7 cells from 16.6 ± 1.4 h to 9.3 ± 0.2 h, 8.6 ± 0.4 h, and 7.2 ± 0.5 h, respectively. The addition of E 2 to MCF-7 cells resulted in a lowering of the doubling time to 8.5 ± 0.4 h (  (Table 2). It is worth mentioning that some flavonoids might be equipotent or ever more potent than E 2 which might be attributed to the higher concentrations used (more than 10-fold). When comparing all tested flavonoids, rutin (6) showed promising proliferative properties with minimal expected mutagenic influence. This is in line with reports by Hyun et al. regarding its osteoclast activating properties [38]. Additionally, the selection of SAOS-2 cells for future studies was based upon their low viability drop when treated with the flavonoids being studied.
In this study, the flavonoids obtained from C. tinctoria displayed clear proliferative effects in all the three studied cell lines. Although several studies have reported that estrogen along with various estrogen metabolites possesses proliferative properties, this was limited by the fact that they cause mutagenicity [22,39]. Therefore, it was of importance to record the impact of rutin on the distribution of the phases of cell cycle in relation to the progression of the cell cycle flow cytometry to determine DNA content. Treatment of SAOS-2 cells with 1 μM rutin for 48 h was carried out. Exposing the cells to 0.1 μM E 2 served as a positive control. Interestingly, rutin produced a significant increase of cells in G 0 /G 1 from 55.2 ± 2.4% to 62.6 ± 1.1% with a reciprocal decrease for cells in S-phase from 30.7 ± 2.9% to 22.5 ± 1.0%. G 2 /M phase was  not affected by treatment with rutin (Figures 2(a), 2(c), and 2 (d)). However, 0.1 μM E 2 produced a significant drop of cells in the S-phase (30.7 ± 2.9% to 26.5 ± 0.9%). Treating SAOS-2 cells with E 2 caused no significant changes of cells in G 0 /G1 or S-phases (Figures 2(a), 2(b), and 2(d)). When studying the proliferating cell fraction balance in SAOS-2 cells, rutin significantly decreased the proliferation index; on the other hand, treatment with E 2 did not alter their proliferation index (Figure 2(e)). In a previous study in our laboratory, paradol showed no influence on the proliferation index of SAOS-2 cells [27]. Our current data are in alignment with our studies on SAOS-2 cells where 1 μM rutin was found to decrease the doubling time. Yet, we tested the nature of the rutin proliferative effect in SAOS-2; accumulation of cells in the supra-G 2 compartment (multiploidy phase) might indicate uncontrolled cell proliferation. Interestingly, after SAOS-2 exposure to rutin, cells in supra-G 2 phase were much lower than control cells (Figure 2(f)). This indicates a punctuated cell division without cell accumulation in multiploidy phase [40]. This gains support by the known chemopreventive and anticarcinogenic properties of rutin [41][42][43].
Several studies have reported that targeting ossification is a more fruitful approach than targeting proliferation, when looking at developing new drug targets to treat osteoporosis. [44,45]. The effects of rutin on osteogenic marker gene expression were quantitatively studied using RT-PCR gene array battery kit that comprises various osteogenic gene families. These include osteocyte, osteoclast, and osteoblast activity markers, as well as RUNX suppressor genes. When studying the osteocyte activity markers (BGN, FGF23, PDPN, HYOU1, and SOST), rutin and E 2 caused an increase in expression of all the tested osteocyte activity gene markers by 1.9-to 2.9-folds and 2.7-to 3.5-folds, respectively (Figure 3(a)). The role of osteocytes in calcium and phosphate homeostasis has been previously reported. Osteocytes cloned with BGN caused acceleration of osteoblast differentiation in vitro and an increase in the area of lamellar bone-like matrices in vivo [46]. Phosphate levels are known to decline due to the presence of the FGF23 protein [47]. Dividing osteocytes are known to express PDPN, which is regarded as a marker of activity [48]. They also secrete SOST causing a negative effect on the formation of bones [49]. Also, rutin increased the expression level of GNL3, CD44, MME, and SCUBE3 osteoblast activity markers, by 1.8 ± 0.2-, 1.9 ± 0.3-, 2.5 ± 0.7-, and 3.5 ± 0.9-folds, respectively. OMD expression after rutin treatment was not changed. The five tested gene markers of osteoblast activity were augmented by 2.1-to 4.0-folds after the application of E 2 (Figure 3(b)). Compared to our previous work on paradol [27], rutin significantly increased HYOU1 but not SOST. Further, rutin could significantly enhance expression of the osteoblast marker CD44 while paradol failed to exert a similar effect. With respect to osteoclast activity markers, both rutin and E 2 downregulated CA2, CALCR, and CTSK expression to 0.4-to 0.5-folds and 0.2-to 0.4-folds, respectively. The expression of MMP9 and TNFRSF11A genes was unaltered after treatment with either rutin or E 2 (Figure 3(c)). These findings confirm the potential favorable effects of rutin. This is further supported by studies in the bone marrow that showed rutin inhibits osteoclastogenesis [50]. When studying the RUNX suppressor gene family (GLB, HES1, STAT1, TWIST1, and HAND2), rutin downregulated the expression of HES1and TWIST1 mRNA to 0.7 ± 0.1-and 0.4 ± 0.07-folds of the control cells, respectively. On the other hand, E 2 was found to suppress HES1, STAT1, TWIST1, and HAND2 gene expression to 0.4-to 0.6-folds of control (Figure 3(d)). Studies have reported that RUNX regulates RANKL leading to the maturation and differentiation of osteoblasts [51]. The lowered expression of HES1and TWIST1 (RUNX suppressors) by rutin supports its osteogenic effects. All these data are in alignment with our previous publication on paradol [27]. However, rutin showed superior activities with regard to suppressing CALCR expression.
In the present study, we determined the effects of treating SAOS-2 cells with 1 μM rutin for 48 h on the levels of four essential osteoporosis-related markers. The results were compared with treatment of the cells with 0.1 μM E 2 as a positive control. Rutin produced a significant increase in the activity/concentrations of all ossification markers, ALP   Figure 3: Effect of rutin on mRNA expression of some ossification related genes in SAOS-2 cell line. Cells were incubated with rutin (1 μM) or E 2 (0.1 μM) for 48 h. Total RNA was extracted and subjected to RT-PCR. Data were normalized to β-actin; fold changes were calculated and expressed as mean ± SD; n = 3. * Significantly different from the control untreated cells; p < 0 05. enzyme, OCN hormone, and active Vit-D3 concentration, by 7.1-, 3.4-, and 1.1-folds, respectively. Similarly, E 2 increased the activity/concentrations of ALP enzyme, OCN hormone, and active Vit-D3 concentration, by 6.9-, 3.9-, and 1.2-folds, respectively (Figures 4(a)-4(c)). On the other hand, rutin produced a statistically significant drop in ACP enzyme activity (bone resorption marker) from 4.7 ± 0.4 IU/ml to 1.6 ± 0.4 IU/ml compared to 1.5 ± 0.4 IU/ml for E 2 -treated cells (Figure 4(d)). The biochemical markers chosen in this study are well documented in the literature as reliable markers having antiosteoporotic properties. [52]. In conclusion, the analysis of C. tinctoria extract led to the isolation of various flavonoids showing antiosteoporosis influence. Rutin was especially promising as it showed ossification in bone cells as well as possessing punctuate proliferative activity with minimal influence to cell cycle distribution.
In other words, rutin might be further studied as a potential antiosteoprotic agent with minimal expected mutagenic effects.

Conflicts of Interest
The authors declare no conflicts of interest.