Phytochemical Characterization, Antioxidant and Anti-Proliferative Properties of Rubia cordifolia L. Extracts Prepared with Improved Extraction Conditions

Rubia cordifolia L. (Rubiaceae) is an important plant in Indian and Chinese medical systems. Extracts prepared from the root, stem and leaf have been used traditionally for the management of various diseases. Some of the known effects are anti-inflammation, neuroprotection, anti-proliferation, immunomodulation and anti-tumor. A comparative account of the extracts derived from different organs that lead to the identification of the most suitable solvent is lacking. We explored the presence of phytochemicals, antioxidant activity and anti-proliferative properties of a variety of solvent-based extracts of root, and methanol extracts of stem and leaf of R. cordifolia L. The antioxidant potential was determined by DPPH, hydrogen peroxide, nitric oxide and total antioxidant assays. The anti-proliferative nature was evaluated by MTT assay on HeLa, ME-180 and HepG2 cells. The composition of the extracts was determined by UPLC-UV-MS. We found that the root extracts had the presence of higher amounts of antioxidants over the stem and leaf extracts. The root extracts prepared in methanol exhibited the highest cytotoxicity in HepG2 cells. The main compounds identified through UPLC-UV-MS of the methanol extract give credibility to the previous results. Our comprehensive study corroborates the preference given to the root over the stem and leaf for extract preparation. In conclusion, we identified the methanol extract of the root to be the most suited to have bioactivity with anti-cancer potential.


Introduction
Various chemotherapeutic drugs inherently induce side effects due to a lack of nonspecificity towards cancer cells. The search for newer molecules has led to a refreshed look at complementary and alternative medicinal practices [1,2]. A plant-derived anti-cancer molecule is expected to provide a solution owing to its natural source. Hence, many plants are being investigated in view of this vital necessity. Extracts from traditional medicinal plants, such as Rubia cordifolia L. may be one of the alternatives available to fill this lacuna. Quinones, terpenoids, alkaloids and their derivatives form a major class of compounds with considerable bioactivities. These components are responsible for the various antioxidation, anti-inflammation and anti-proliferative bioactivities, among others. Mollugin (derivative of anthraquinone) inhibits pro-inflammatory chemocytokine production [12]. Purpurin is another anthraquinone that gives R. cordifolia L antioxidant properties [13]. Alizarin, 6-hydroxyrubiadin, purpurin and rubiadin are expected to be key constituents responsible for analgesic and anti-inflammatory properties [14]. The mode of action of the exhaustive list of compounds has not been elucidated completely as many compounds are solvent-specific and are not available in large quantities.
Quantified research that directs to the therapeutic usage of specific extraction solvents for different plant organs is still lacking. Further, a comparison among the different extracts prepared from different R. cordifolia plant organs remains unattempted so far. Within this frame of reference, we have focused our attention on the antioxidant and anti-proliferative activities of various extracts prepared from R. cordifolia and have identified methanol as the most suitable solvent [15]. An in vitro analysis on the cancer cell lines confirmed the methanol extract of the root as the most suitable for pertinent pre-clinical studies.

Plant Collection
The stems and leaves of R. cordifolia were freshly collected from Torna fort (18 • 16 33.86 N 73 • 37 21.78 E) and Mahabaleshwar Forest (17 • 55 51 ' N 73 • 38 52 E) located in Maharashtra State, India. Air-dried leaves and stems were separated. The dried samples were pulverized into a coarse powder and stored for further use. The plant was authenticated at Botanical Survey of India, Pune, India center with specimen number mgJRC-1 and a voucher specimen is deposited at the BSI herbarium.

Preparation of Extracts
All solvents, reagents and standards used were of analytical grade (HiMedia, Mumbai, India). Extracts of powders were prepared in methanol, ethanol or distilled water as described previously [16]. In brief, powders of different plant parts of R. cordifolia were extracted with solvent individually by conventional Soxhlet apparatus (Goel Scientific, Vadodara, India) extraction procedure. After the exhaustive extraction, each extract was evaporated to dryness by rotary evaporator (Aditya Scientific, Hyderabad, India). We quenched the polyphenols using polyvinylpolypyrrolidone (PVPP) to determine if antioxidant activity is exclusive to the polyphenols present in the extract. To remove polyphenols from the extracts, they were treated with 10% (w/v) PVPP made in respective solvents and kept on a shaking incubator (238019, Thermo Fisher, Waltham, MA, USA) at 37 • C Antioxidants 2022, 11, 1006 6 of 20 overnight. The polyphenols bind with PVPP and settle at the bottom, while the supernatant contains the polyphenol-free extract [17].

Qualitative Phytochemical Screening of R. cordifolia Constituents
The preliminary screening of different classes of natural plant constituents was performed. The presence of secondary metabolites viz. alkaloids, saponins, tannins, phenols, glycosides, terpenes, carotenoids and quinones was detected using the standard tests as described below [16,18].

Alkaloid Detection
Mayer's test for alkaloids was performed by treating equivalent volumes of extract with Mayer's reagent (in-house prepared by dissolving 1.36 g of mercuric chloride (GRM1067, HiMedia, Mumbai, India) and 5 g of potassium iodide (GRM252, HiMedia, Mumbai, India) in 100 mL distilled water), and the subsequent development of creamcolored precipitate implied existence of alkaloid. Dragenforff's reagent was prepared by dissolving 8 g of bismuth nitrate (RM1221, HiMedia, Mumbai, India) in 20 mL of concentrated nitric acid (GRM6105, HiMedia, Mumbai, India) and 27.2 g of potassium iodide (KI) in 50 mL of distilled water. Both the solutions were kept standing till KIO 3 crystallized out. The supernatant was decanted, and final volume was adjusted to 100 mL with distilled water. Dragendorff's test for alkaloids was accomplished by treating equivalent volumes of extract with Dragendorff's reagent. Subsequent generation of red-colored precipitate suggested presence of alkaloid. Wagner's reagent was prepared by dissolving 2 g of iodine (GRM1064, HiMedia, Mumbai, India) and 6 g of potassium iodide in 100 mL of distilled water. Wagner's test for alkaloids was performed by treating equivalent volumes of extract with Wagner's reagent. Subsequent development of reddish-brown-colored precipitate indicated existence of alkaloid. Hager's reagent was prepared by dissolving 1 g of picric acid (S026, HiMedia Mumbai, India) in 100 mL of distilled water. Hager's test for alkaloids was performed by treating equivalent volumes of extract with Hager's reagent. Subsequent development of yellow-colored precipitate suggested presence of alkaloid.

Saponin Detection
Saponin was detected by dissolving equivalent quantity of extract in water followed by vigorous shaking. Formation of honeycomb-shaped persistent froth indicated the existence of saponins in the sample.

Tannin Detection
Tannins were determined by mixing extract with 0.5% aqueous ferric chloride (GRM165-500G, HiMedia, Mumbai, India), and dark green/bluish-green coloration of the sample indicated presence of tannins.

Phenol Detection
Phenols were determined by adding equivalent volumes of extract to Folin-Ciocalteu reagent (RM10822, HiMedia, Mumbai, India), and blue coloration of sample indicated presence of phenols.

Glycoside Detection
Glycosides were identified by treating equivalent volumes of extract with glacial acetic acid (AS001, HiMedia, Mumbai, India) and some drops of 5% aqueous ferric chloride (FeCl 3 ) and concentrated sulphuric acid (H 2 SO 4 ) (AS016-500ML, HiMedia, Mumbai, India). This is known as Keller-Kiliani test. Reddish-brown coloration at the confluence and bluish-green color in top layer solution indicated presence of glycosides.

Flavonoids Detection
Flavonoids were detected by Shinoda test when to 1 mL of extract, few mg turnings were added followed by a few drops of concentrated hydrochloric acid (HCl). Development of reddish pink coloration indicated presence of flavonoids.

Terpene Detection
Terpenes were detected by mixing equivalent volumes of extract with chloroform and concentrated sulphuric acid. Reddish-brown coloration at the junction of two solutions suggested the occurrence of terpenes.

Steroid Detection
Steroids were detected by formation of orange color in solution consisting of equivalent volumes of extract with chloroform, glacial acetic acid and concentrated sulphuric acid.

Quinone Detection
Presence of quinone was determined by formation of green color upon addition of concentrated hydrogen chloride (RM5955-500ML, HiMedia, Mumbai, India) to the extract [19].

Carotenoids Detection
Carotenoids were detected by formation of deep blue color in solution consisting of equivalent volumes of extract with concentrated sulphuric acid (H 2 SO 4 ) and a few crystals of iodine.

Quantification of Phenols
Phenolic content was determined according to the method reported earlier [17]. Briefly, 1 mL of 1 mg/mL extract and gallic acid with the concentrations of 20, 40, 60, 80 and 100 µg/mL was mixed with 0.5 mL of 1N Folin-Ciocalteu reagent and incubated for 5 min, followed by addition of 1 mL of 20% sodium carbonate. After 10 min incubation at room temperature, absorbance was measured at 730 nm. Gallic acid was used as the standard and the phenolic content was expressed as gallic acid equivalent (GAE). The equation of the curve: y = mx + c with R 2 > 0.99. The limit of detection (LOD) and limit of quantification (LOQ) were based on the standard deviation of the blank and calculated using following equations: where σ is the standard deviation of y-intercepts of the regression line, and S is the slope of the calibration curve.

Quantification of Flavonoids
Flavonoid content in the extract was determined in accordance with the reported method [20]. In brief, 1 mL of extract and quercetin with the concentration of 100, 200, 300, 400 and 500 µg/mL was mixed with 1.25 mL of distilled water and 75 µL of 5% of sodium nitrite solution incubated for 5 min; subsequently, 150 µL of 10% aluminum chloride (Sigma-Aldrich, Burlington, MA, USA) solution was added. After incubation for 6 min, 500 µL of 1 M sodium hydroxide and 275 µL of distilled water were added to prepare the mixture. The absorbance was recorded at 510 nm. Quercetin was used as the standard, and the flavonoid content is expressed as quercetin equivalent (QE). The equation of the curve: y = mx + c with R 2 > 0.99. The LOD and LOQ were based on the standard deviation of the blank and calculated as described by equations 1 and 2, respectively.
2.6. Antioxidant Assays 2.6.1. DPPH Free Radical Scavenging Assay DPPH (2,2-diphenyl-1-picrylhydrazyl) scavenging activity was measured with spectrophotometric method as described previously [21]. To 0.5 mL extract solution made in respective solvents of concentration ranging from 20 to 100 µg/mL, 1 mL of 0.2 mM DPPH (RM2798, HiMedia, Mumbai, India) made in methanol was added and volume was made up to 2 mL with methanol and incubated for 30 min at room temperature. The absorbance was measured at 517 nm against blank. Ascorbic acid was used as the standard control. The antioxidant activity was presented as IC 50 value (µg/mL) based on percentage of inhibition of DPPH as calculated in accordance with Equation (3).

Hydrogen Peroxide Scavenging Assay
The scavenging effect of hydrogen peroxide was determined as described earlier [22]. Briefly, 1 mL of extract solution of concentration ranging from 20 to 100 µg/mL was treated with 0.6 mL, 40 mM of hydrogen peroxide (Fisher Scientific, Pittsburgh, PA, USA) prepared in phosphate buffer (pH 7.4) for 10 min. The absorbance was read at 230 nm against blank of hydrogen peroxide. Ascorbic acid was used as standard, and the antioxidant activity was presented as IC 50 value (µg/mL) based on percentage of inhibition of hydrogen peroxide (Equation (3)).

Scavenging Activity of Nitric Oxide
Nitric oxide was generated from sodium nitroprusside, and its scavenging effect was determined as described previously [16]. Briefly, different concentrations ranging from 20 to 100 µg/mL of 1 mL of extract solution and 1 mL (pH 7.4) phosphate buffer were used to prepare 0.5 mL of 10 mM sodium nitroprusside. After incubation for 5 h at 25 • C, 0.5 mL of supernatant liquid was removed and 0.5 mL of Griess reagent (G7921, Thermo Fisher, Waltham, MA, USA) (1 mM) prepared in distilled water was added. The absorbance of the chromophore formed during diazotization of nitric oxide with sulphanilamide and its subsequent coupling with N-(1-naphthyl) ethylene-diamine was determined at 546 nm. Ascorbic acid was used as standard, and the antioxidant activity was presented as IC 50 value (µg/mL) based on percentage of inhibition of nitric oxide (Equation (3)).

Determination of Total Antioxidant Capacity
The total antioxidant capacity was determined by phosphomolybdate assay [23]. In brief, 1 mL of extract of concentrations ranging from 20 to 100 µg/mL prepared in respective solvents was taken and mixed with 1 mL of reagent containing 0.6 M sulphuric acid, 28 mM sodium phosphate (MB047-250G, Thermo Fisher, Waltham, MA, USA) and 4 mM ammonium molybdate (A7302-100G, Sigma-Aldrich, Burlington, MA, USA). The solution formed was incubated at 95 • C for 90 min, cooled to room temperature and absorbance was noted at 695 nm. Ascorbic acid was used as the standard, and the total antioxidant capacity was calculated as percentage scavenging activity (refer Equation (3)).

Principal Component Analysis
Principal component analysis (PCA) was performed to point out the clustering of data into two separated groups, namely PVPP untreated (−PVPP) and treated (+PVPP) extracts. The PCA is a procedure aiming at reducing the dimensionality of the data and allowing the visualization of a large number of variables in a two-dimensional plot [24]. The input data were obtained from quantification of phenol and flavonoid and antioxidant activity (phenol content expressed as mg GAE/g of plant extract; flavonoid content expressed as mg QE/g of plant extract; antioxidant potential by DPPH free radical scavenging expressed as IC 50 ; antioxidant potential by hydrogen peroxide scavenging expressed as IC 50 ; antioxidant potential by nitric oxide scavenging assay expressed as IC 50 and total antioxidant capacity expressed as IC 50 ) in root-methanol, root-ethanol, root-aqueous, leaf-methanol and stemmethanol extracts. A diagram of the values obtained from each treatment condition was plotted in the bidimensional space, defined by the 1st and 2nd principal component functions (PC1 and PC2, respectively).

Cell Culture and Cytotoxicity
Authenticated cell lines ME-180, HeLa and HepG2 were procured from National Centre for Cell Science, Pune, India. The cells were grown in Roswell Park Memorial Institute-1640, Eagle's minimal Essential Medium and Dulbecco's Modified Eagle Medium media, respectively, and 10% FBS (16000044, Thermo Fisher, Waltham, MA, USA) and 1% antibiotic solution were used for supplementation. Cells were grown in T-25 flasks and were passaged upon confluence using trypsin-EDTA [16]. Nearly 5000 cells were seeded per well in 96-well plate and incubated at 37 • C in 5% CO 2 incubator and left overnight to enable surface attachment. Cells were treated with extracts (methanol, ethanol and aqueous) with concentrations of 50, 25, 10, 5, 1, 0.5, 0.1, 0.05 mg/mL and left overnight in incubator. 5 mg/mL of MTT per well was added and incubated for 2 h at 37 • C. Formazan crystals were solubilized with 100 µL DMSO and incubated for 10 min. The absorbance was measured at 570 nm and reference at 630 nm.

Statistical Analysis
All experiments were performed in triplicate and the values were expressed as mean ± standard error of mean (SEM). The data were analyzed by Student-Newman-Keuls test using Sigma Plot version 14 (Systat Software Inc., Palo Alto, CA, USA), and IC 50 values were calculated using OriginPro, version 2021 (OriginLab Corporation, Northampton, MA, USA).

Qualitative Analysis of Secondary Metabolites of R. cordifolia Extracts
The methanol extract of R. cordifolia root had alkaloids, tannins, phenols, flavonoids and terpenes (Table 2). In comparison, while the ethanol extract lacked tannins, the aqueous extract had saponins and glycosides. Considering the maximally reported usage of methanol extracts for roots, we evaluated methanol extracts of leaves and stems in the same way. In contrast to the methanol extracts of roots, the methanol extracts of leaves had glycosides and quinones, while the stem-methanol extracts had quinones and carotenoids.

Quantification of Phenols and Flavonoids in Extracts
Standard calibration curves were plotted for the quantification of phenols in extracts. The plot for standard gallic acid for both PVPP-untreated and -treated was linear, with correlation coefficients (R 2 ) equal to 0.9916 and 0.99, respectively. The regression equations for PVPP-untreated and -treated were y = 0.0093x + 0.0436 and y = 0.0062x + 0.0335, respectively, with an LOD under 10 mg/g and LOQ under 30 mg/g for both. Similarly, standard quercetin calibration plots were obtained as linear with R 2 of 0.9986 and 0.991, and regression equations of y = 0.0014x + 0.0067 and y = 0.0012x + 0.0308 for PVPP-untreated and -treated, respectively. The LODs were under 20 mg/g and LOQs were under 40 mg/g for both.
Significant levels of difference were observed in all the root extracts post-PVPP treatment for the phenols and flavonoids. The ethanol and methanol extracts of roots had the highest phenol and flavonoid content, respectively, compared to the other extracts for 1 mg/mL concentrations of extracts ( Table 3). The roots had the highest phenol and flavonoid content among the methanol extracts of different organs of R. cordifolia L.

Root Extracts Have Better Antioxidant Activity Than Leaf and Stem Extracts
The percentage of scavenging activity of the root-ethanol extract in 2,2-diphenyl-1-picrylhydrazyl (DPPH) and hydrogen peroxide assays is less in absence of the PVPP treatment, while higher IC 50 values were obtained in the presence of PVPP in nitric oxide and total antioxidant assays ( Figure 1 and Table 4). With the post-treatment of root extracts by PVPP, the aqueous extract was found to be 84%, 81% and 84% more potent in DPPH, hydrogen peroxide and total antioxidant assay, respectively. The methanol extracts of leaf and stem showed higher IC 50 in all the assays. Considering the absence of significant levels of antioxidant activities in the methanol extracts of leaves and stems, we continued with the extracts of root for further assays.  Table 4. IC50 values of DPPH, hydrogen peroxide, nitric oxide and total antioxidant assay of R. cordifolia. Results were expressed as the mean ± SD of three independent experiments. Significant difference between without PVPP and with PVPP representing p < 0.001, p < 0.01, p < 0.05 is by ***, ** and *, respectively. R. cordifolia extracts were tested at concentrations of 20, 40, 60, 80, 100 µg/mL.  Results were expressed as the mean ± SD of three independent experiments, test of significance among PVPP untreated and treated extracts by ANOVA, wherein *, ** and *** represent statistical significance of p < 0.05, p < 0.01 and p < 0.001, respectively.

Principal Component Analysis of R. cordifolia Phenol, Flavonoid and Antioxidant Levels in PVPP Untreated and Treated Extracts
The data obtained by the quantification of the phenols and flavonoids with antioxidant levels of R. cordifolia among PVPP-untreated and -treated extracts have been used to perform a principal components analysis (PCA) ( Table 5). As shown in Table 6, the first principal component was highly correlated with flavonoid content and antioxidant levels by H 2 O 2 scavenging activity (Antioxidant_H 2 O 2 ) variables, while the second principal component was highly correlated with antioxidant levels by NO scavenging activity (Antioxidant_NO) variable (highest score coefficients in absolute value). Table 6. Component score coefficient matrix (coefficients by which variables are multiplied to obtain factor scores). The highest score coefficients in absolute value are marked in bold. The first and the second principal components explain together 77.44% of the total observed variance, which is a considerable value. The PCA showed a clear separation between the -PVPP and +PVPP data (Figure 2), as better evidenced by the dotted line added in the plot.

Variables
Negative PC1 values correlated to the flavonoid content and Antioxidant_H 2 O 2 variables were mostly associated with the root samples (blue symbols). On the other hand, PVPP-untreated (empty symbols) and -treated (solid plain symbols) roots were markedly separated by the dotted line, indicating that the flavonoid content, H 2 O 2 and NO antioxidant activities are different. The stem samples (green symbols) have different PC2 values correlated to Antioxidant_NO scavenging activity, and leaf samples (red symbols) have different PC1 and PC2 values since they are separated by the dotted line. The first and the second principal components explain together 77.44% of the total observed variance, which is a considerable value. The PCA showed a clear separation between the -PVPP and +PVPP data (Figure 2), as better evidenced by the dotted line added in the plot.
Negative PC1 values correlated to the flavonoid content and Antioxidant_H2O2 variables were mostly associated with the root samples (blue symbols). On the other hand, PVPP-untreated (empty symbols) and -treated (solid plain symbols) roots were markedly separated by the dotted line, indicating that the flavonoid content, H2O2 and NO antioxidant activities are different. The stem samples (green symbols) have different PC2 values correlated to Antioxidant_NO scavenging activity, and leaf samples (red symbols) have different PC1 and PC2 values since they are separated by the dotted line.
The methanol extracts (circle symbols) have mostly positive PC1 values but different PC2 values, indicating a difference in the Antioxidant_NO activity related to the PVPP treatment. The aqueous extracts (square symbols) have negative PC1 and similar PC2 values, indicating a similar Antioxidant_NO activity independent from the PVPP treatment. The ethanol extracts (triangle symbols) have both PC1 and PC2 values, indicating different flavonoid content, Antioxidant_H2O2 and NO activities related to the PVPP treatment.

Plant Extracts Are Cytotoxic to Cancer Cells
Cancer cell lines ME-180, HeLa and HepG2 were exposed to various concentrations of extracts and standard drug (5-Flurouracil) to determine the cell viability by MTT cell The methanol extracts (circle symbols) have mostly positive PC1 values but different PC2 values, indicating a difference in the Antioxidant_NO activity related to the PVPP treatment. The aqueous extracts (square symbols) have negative PC1 and similar PC2 values, indicating a similar Antioxidant_NO activity independent from the PVPP treatment. The ethanol extracts (triangle symbols) have both PC1 and PC2 values, indicating different flavonoid content, Antioxidant_H 2 O 2 and NO activities related to the PVPP treatment.

Plant Extracts Are Cytotoxic to Cancer Cells
Cancer cell lines ME-180, HeLa and HepG2 were exposed to various concentrations of extracts and standard drug (5-Flurouracil) to determine the cell viability by MTT cell proliferation assay. HeLa and HepG2 cells were susceptible to any of the extracts at similar concentrations (Figures 3 and 4). The root-methanol extract was more potent than other extracts for HeLa (IC 50 of 0.29 ± 0.23 mg/mL) and HepG2 (IC 50 of 0.39 ± 0.26 mg/mL) ( Table 7). 5-Flurouracil (5-FU) was most toxic to HepG2 cells (IC 50 of 1.51 ± 0.38µM), and the levels of toxicity were significantly lower than those in the other cell lines evaluated. proliferation assay. HeLa and HepG2 cells were susceptible to any of the extracts at similar concentrations (Figures 3 and 4). The root-methanol extract was more potent than other extracts for HeLa (IC50 of 0.29 ± 0.23 mg/mL) and HepG2 (IC50 of 0.39 ± 0.26 mg/mL) ( Table  7). 5-Flurouracil (5-FU) was most toxic to HepG2 cells (IC50 of 1.51 ± 0.38µM), and the levels of toxicity were significantly lower than those in the other cell lines evaluated.

UPLC-UV-MS Phytochemical Profiling of Methanol Extract of R. cordifolia
To identify the compounds responsible for anti-proliferative potential, the composition of the root methanol extract was evaluated by UPLC-UV-MS analysis. A number of secondary metabolites were detected (Supplementary Table S1). Out of them, the structures of two of the signature compounds from R. cordifolia L. were used to compare with the existing PubChem database (Figures 5 and 6). In addition, the formula, score, mass and CAS numbers with retention time are listed in Table 8.  Results were expressed as the mean ± SD of three independent experiments, test of significance by ANOVA, wherein * and *** represent statistical significance of p < 0.05 and p < 0.001, respectively.

UPLC-UV-MS Phytochemical Profiling of Methanol Extract of R. cordifolia
To identify the compounds responsible for anti-proliferative potential, the composition of the root methanol extract was evaluated by UPLC-UV-MS analysis. A number of secondary metabolites were detected (Supplementary Table S1). Out of them, the structures of two of the signature compounds from R. cordifolia L. were used to compare with the existing PubChem database (Figures 5 and 6). In addition, the formula, score, mass and CAS numbers with retention time are listed in Table 8.

UPLC-UV-MS Phytochemical Profiling of Methanol Extract of R. cordifolia
To identify the compounds responsible for anti-proliferative potential, the composition of the root methanol extract was evaluated by UPLC-UV-MS analysis. A number of secondary metabolites were detected (Supplementary Table S1). Out of them, the structures of two of the signature compounds from R. cordifolia L. were used to compare with the existing PubChem database (Figures 5 and 6). In addition, the formula, score, mass and CAS numbers with retention time are listed in Table 8.

Discussion
The utility of secondary metabolites for human health has achieved high recognition owing to their promising usage in traditional knowledge-based medication for centuries. R. cordifolia L produces a range of secondary metabolites that have been evaluated for various illnesses. In the present study, we have evaluated three solvent systems for roots and methanol as a solvent for stems and leaves to extract secondary metabolites from R. cordifolia L. The suitability of methanol extracts in antioxidant assays prompted us to evaluate methanol extracts of stems and leaves for phytochemical analysis. Nevertheless, the antioxidant levels of roots were noticed to be higher than stems and leaves.
Phenols are a major antioxidant group present in plants. We detected significant amounts of phenols in the ethanol extract of root, followed by the methanol extract of leaf. Flavonoids are the largest group of natural phenolics that possess tremendous free radical scavenging properties and, hence, antioxidant potential. Our method of Soxhlet extraction led to an increased release of phenols and flavonoids.
The presence of antioxidants in the extract is crucial for usage as an anti-proliferative

Discussion
The utility of secondary metabolites for human health has achieved high recognition owing to their promising usage in traditional knowledge-based medication for centuries. R. cordifolia L produces a range of secondary metabolites that have been evaluated for various illnesses. In the present study, we have evaluated three solvent systems for roots and methanol as a solvent for stems and leaves to extract secondary metabolites from R. cordifolia L. The suitability of methanol extracts in antioxidant assays prompted us to evaluate methanol extracts of stems and leaves for phytochemical analysis. Nevertheless, the antioxidant levels of roots were noticed to be higher than stems and leaves.
Phenols are a major antioxidant group present in plants. We detected significant amounts of phenols in the ethanol extract of root, followed by the methanol extract of leaf. Flavonoids are the largest group of natural phenolics that possess tremendous free radical scavenging properties and, hence, antioxidant potential. Our method of Soxhlet extraction led to an increased release of phenols and flavonoids.
The presence of antioxidants in the extract is crucial for usage as an anti-proliferative agent. The results of the DPPH assay for the ethanol extract of root reported by Zhang et al. [25] were in the range of 23.88 to 65.23 µg/mL. They used an ultrasonic-assisted extraction process. These values are much lower than the presently reported values in the range of 88.5 to 98.26 µg/mL. We believe the suitability of the extraction method and the mother plant selection are the drivers of differential results. Basu and Hazra [26] reported a range of 153.7 to 310.3 µg/mL for methanol and aqueous extracts of root as evaluated by a nitric oxide assay. They used the filtrate of the directly solubilized extracts in the respective solvents. Our results have a different range, possibly due to our choice of method of the Soxhlet exhaustive extraction process.
The antioxidant activity of the plant extract is attributed to various secondary metabolites, including polyphenols. Studies pertaining to the significance of polyphenols have emphasized their influence on the antioxidant results [27][28][29]. We propose to present the case that the antioxidant activity observed in R. cordifolia L. is not entirely due to polyphenols. To prove that the determined antioxidant activity is not exclusive to the polyphenols present in the extract and is contributed to by other secondary metabolites as well, we quenched the polyphenols using PVPP. Rantunge et al., 2017 have demonstrated the quenching effect of PVPP on different polyphenols, and it clearly shows remarkable differences [17]. The precipitation allows the removal of any complex of PVPP-polyphenols. A comparison of the PVPP-untreated and -treated extracts by the same antioxidant assays proved that there are other compounds responsible for antioxidant properties as well. We are reporting for the first time the results pertaining to R. cordifolia root extracts (ethanol, methanol and aqueous) treated with PVPP for antioxidant assays. Even after the removal of phenols and flavonoids, the antioxidant activity of the extract is not hampered. This suggests the involvement of other non-phenolic secondary metabolites in bringing about the antioxidant potential. The PCA correlated the phenol, flavonoid and antioxidant levels, as evaluated by hydrogen peroxide and nitric oxide scavenging assays. Our evaluation of R. cordifolia leaves and stems demonstrates that the root is more suited to be used for antioxidant properties. The high prevalence of antioxidant compounds in root extracts may be utilized for the anti-proliferative process in certain cancers [30,31]. The anti-proliferative assay corroborated the suitability of the methanol extract of the root for anti-cancer activity. The sensitivity of HepG2 towards 5-Flurouracil as compared to other cell lines was not reflected for plant extracts as similar toxicity was observed in the ME180 and HeLa for cell lines, suggesting its usage for the management of multiple cancers. The cytotoxicity may be mediated by reactive oxygen species, as indicated in laryngeal squamous cell carcinoma HEp-2 cells [7]. However, the apoptotic pathway responsible for cell toxicity needs further elucidation. The isolation of suitable cancer-specific bioactive compounds is necessary, or else it may yield a negative result [10].
Our results of UPLC-UV-MS identified some previously reported compounds and some new compounds from Rubia plants. Pseudopurpurin (anthraquinone) is a characteristic natural red-color compound present in the roots of R. cordifolia and Rubiaceae family members. It is a derivative of purpurin (pseudopurpurin is purpurin 3-carboxylic acid). It improves bone geometry [32] and selectively exhibits tumor inhibitory potential [33]. Morindaparvin A is reported to be an antileukemic anthraquinone and is chemically derived from alizarin (1,2-methylenedioxyanthraquinone by synthesis from alizarin) [34]. We report its presence in R. cordifolia for the first time. It is possible that its presence was not detected previously or was not considered as it is a derivative of alizarin. These preliminary findings require a detailed supplemental study for verification before confirmation.
The presence of multiple compounds in the methanol extract that are established to be cytotoxic to cancer cells supports our results. However, the validation of cytotoxic activities requires independent assays.

Conclusions
R. cordifolia L. is a widely used plant for its significant medicinal value. This is attributed to the presence of unique secondary metabolites in R. cordifolia L. Exhaustive methods of extraction lead to an increase in the retrieval of secondary metabolites, as observed in our research endeavor. This work provides the initial steps required in selecting the suitable solvents for R. cordifolia extract preparations. Our study has revealed the presence of a high quantity of antioxidants in the root, stem and leaf extracts of R. cordifolia. The antioxidant levels in the root, stem and leaf provide a comparative benchmark for further exploration. The results obtained for the antiproliferative assay make the extracts valuable to medicinal practitioners. Identification of different compounds may help in determining a metabolite signature characteristic of R. cordifolia. The individual compounds need to be evaluated to verify the extent of the utility of the antioxidant nature for identifying a potential anti-cancer agent. In summary, the medicinal value imparted by the extracts is comprehensively documented for its usage in anti-cancer research.