Phytochemical Analysis and Anti-Inflammatory and Anti-Osteoarthritic Bioactive Potential of Verbascum thapsus L. (Scrophulariaceae) Leaf Extract Evaluated in Two In Vitro Models of Inflammation and Osteoarthritis

Osteoarthritis (OA) is a complex disease, source of pain and disability that affects millions of people worldwide. OA etiology is complex, multifactorial and joint-specific, with genetic, biological and biomechanical components. Recently, several studies have suggested a potential adjuvant role for natural extracts on OA progression, in terms of moderating chondrocyte inflammation and following cartilage injury, thus resulting in an overall improvement of joint pain. In this study, we first analyzed the phenylethanoid glycosides profile and the total amount of polyphenols present in a leaf aqueous extract of Verbascum thapsus L. We then investigated the anti-inflammatory and anti-osteoarthritic bioactive potential of the extract in murine monocyte/macrophage-like cells (RAW 264.7) and in human chondrocyte cells (HC), by gene expression analysis of specifics inflammatory cytokines, pro-inflammatory enzymes and metalloproteases. Six phenylethanoid glycosides were identified and the total phenolic content was 124.0 ± 0.7 mg gallic acid equivalent (GAE)/g of extract. The biological investigation showed that the extract is able to significantly decrease most of the cellular inflammatory markers, compared to both control cells and cells treated with Harpagophytum procumbens (Burch.) DC. ex Meisn, used as a positive control. Verbascum thapsus leaf aqueous extract has the potential to moderate the inflammatory response, representing an innovative possible approach for the inflammatory joint disease treatment.


Introduction
Osteoarthritis (OA), also known as degenerative arthritis, is the most common chronic condition of joint disease that affects up to 15% of the adult population. It is a disorder caused in part by damage of cartilage component and function, and deregulation of proinflammatory and anti-inflammatory pathways [1][2][3]. OA is mainly characterized by the breakdown of the articular cartilage and often accompanied by subchondral bone injuries, deterioration of tendons and ligaments and several levels of inflammation of the synovium, leading to pain, swelling and disability. OA is an incurable disease because once cartilage breaks down, it can hardly be repaired, but some therapies can slow down their breakdown, alleviate pain and improve joint mobility [4]. In fact, pharmacological therapies, with anti-inflammatory and chondroprotective effects, aim at reducing pain and

Phenylethanoid Glycosides Profile and Total Phenolic Content
The trace recorded at 330 nm ( Figure 1 and Figure S1) showed four predominant constituents (peaks 3, 4, 5 and 6), accompanied with at least another two minor peaks (peaks 1 and 2), six compounds were identified. Peak 1 was identified as samioside (52.3 mg/mL), peak 2 was identified as echinacoside (42.1 mg/mL), peak 3 was identified as forsythoside B (108.1 mg/mL), peak 4 was identified as verbascoside (254.8 mg/mL), peak 5 was identified as iso-verbascoside (142.9 mg/mL) and peak 6 was identified as martynoside (118.4 mg/mL) ( Table 1). The total PhEGs content determined by HPLC was 718.6 (mg/mL) calculated using calibration curves with the closest appropriate standard. Otherwise, all the other peaks not identified are mostly attributable to different classes of polyphenols. Therefore, the total phenolic content determined spectrophotometrically by the Folin-Ciocalteu method, was estimated in 124.0 ± 0.7 mg gallic acid equivalent (GAE)/g of extract.  Table 1.

Evaluation of the Anti-Inflammatory Property of V. thapsus Extract on RAW 264.7 Cells
To assess the anti-inflammatory property of V. thapsus compared to H. procumbens the two selected concentrations of extract (50 and 100 µg/mL) were added on LPS-stimulated RAW 264.7 and NO• levels measured after 24 h and 6 days of treatment. Our data showed that after 24 h of treatment, only V. thapsus at the concentration of 50 µg/mL and H. procumbens at the concentration of 100 µg/mL were able to reduce the NO• level in a statistically significant manner. On the contrary, after 6 days of treatment, V. thapsus and H. procumbens, at both concentrations, were able to significantly decrease the NO• level compared to the control (** p< 0.01), even if V. thapsus at the concentration of 50 µg/mL showed a greater NO• level reduction than H. procumbens at the same concentration of 50 µg/mL (** p< 0.01) ( Figure 3).

Evaluation of the Anti-Inflammatory Property of V. thapsus Extract on RAW 264.7 Cells
To assess the anti-inflammatory property of V. thapsus compared to H. procumbens the two selected concentrations of extract (50 and 100 µg/mL) were added on LPS-stimulated RAW 264.7 and NO• levels measured after 24 h and 6 days of treatment. Our data showed that after 24 h of treatment, only V. thapsus at the concentration of 50 µg/mL and H. procumbens at the concentration of 100 µg/mL were able to reduce the NO• level in a statistically significant manner. On the contrary, after 6 days of treatment, V. thapsus and H. procumbens, at both concentrations, were able to significantly decrease the NO• level compared to the control (** p < 0.01), even if V. thapsus at the concentration of 50 µg/mL showed a greater NO• level reduction than H. procumbens at the same concentration of 50 µg/mL (** p < 0.01) ( Figure 3). s 2021, 26, x FOR PEER REVIEW 5 of 13 ANOVA test p-values were reported (p < 0.0001), and (**, ++, °° p < 0.01) indicate significant differences between the two groups, as reported by the Holm post hoc test.
To assess whether IL-1β was able to induce activation of the inflammatory response in chondrocytes stimulated for 24 h, we performed a gene expression profile by using specific inflammatory mediators and metalloproteases. The results, reported in Figure  4B,C, highlight that IL-1β-stimulated HC show a significant expression level increase of all analyzed genes compared to not stimulated cells. Precisely, IL-1β-stimulated HC show an increase of 13.5-fold for COX2, 11.6-fold for IL-1β, 9.4-fold for IL-6, 2.5-fold for I-NOS, 17.1-fold for MMP1, 26.8-fold for MMP3, 34.5-fold for MMP9 and 75.6-fold for MMP13.
To assess whether IL-1β was able to induce activation of the inflammatory response in chondrocytes stimulated for 24 h, we performed a gene expression profile by using specific inflammatory mediators and metalloproteases. The results, reported in Figure 4B,C, highlight that IL-1β-stimulated HC show a significant expression level increase of all analyzed genes compared to not stimulated cells. Precisely, IL-1β-stimulated HC show an increase of 13.5-fold for COX2, 11.6-fold for IL-1β, 9.4-fold for IL-6, 2.5-fold for I-NOS, 17.1-fold for MMP1, 26.8-fold for MMP3, 34.5-fold for MMP9 and 75.6-fold for MMP13.

Anti-Osteoarthritis Effect of V. thapsus Extract
To evaluate the anti-osteoarthritis effect of V. thapsus on IL-1β-stimulated HC, we performed a gene expression profile with specifics inflammatory mediators ( Figure 5) and MMPs ( Figure 6) after 24 h and 6 days of treatment. Our results showed that after 24 h and 6 days of V. thapsus treatment, at both concentrations (50 and 100 µg/mL), it was possible to observe a significant decrease (* p < 0.05 and ** p < 0.01) in the expression levels of all cytokines, pro-inflammatory enzymes and metalloproteases, even if the higher concentration displayed a greater effect ( Figure 5) compared to both the control (IL-1βstimulated HC) and HC cells treated with H. procumbens.

Anti-Osteoarthritis Effect of V. thapsus Extract
To evaluate the anti-osteoarthritis effect of V. thapsus on IL-1β-stimulated HC, we performed a gene expression profile with specifics inflammatory mediators ( Figure 5) and MMPs ( Figure 6) after 24 h and 6 days of treatment. Our results showed that after 24 h and 6 days of V. thapsus treatment, at both concentrations (50 and 100 µg/mL), it was possible to observe a significant decrease (* p < 0.05 and ** p < 0.01) in the expression levels of all cytokines, pro-inflammatory enzymes and metalloproteases, even if the higher concentration displayed a greater effect ( Figure 5) compared to both the control (IL-1β-stimulated HC) and HC cells treated with H. procumbens.   . Effects of V. thapsus and H. procumbens treatment on IL-1β-induced expression levels of inflammatory cytokines and enzymes. Data are represented as the means ± SD of three independent experiments. ANOVA test p-values were reported (p < 0.0001) and (**, ## p < 0.01 and *, # p < 0.05) indicate significant differences between the two groups, as reported by the Holm post hoc test.

Discussion
Osteoarthritis (OA) is a complex disease, source of pain and disability that affects millions of people worldwide. There is no single cause for OA, and the exact etiology is Figure 6. Effects of V. thapsus and H. procumbens treatment on IL-1β-induced expression levels of inflammatory cytokines and enzymes. Data are represented as the means ± SD of three independent experiments. ANOVA test p-values were reported (p < 0.0001) and (**, ## p < 0.01 and # p < 0.05) indicate significant differences between the two groups, as reported by the Holm post hoc test.

Discussion
Osteoarthritis (OA) is a complex disease, source of pain and disability that affects millions of people worldwide. There is no single cause for OA, and the exact etiology is complex, multifactorial and joint-specific, with genetic, biological and biomechanical components [18]. Recent in vitro and preclinical studies suggest the protective effects of polyphenols and a possible key role in the prevention and treatment of the early stages of OA, through the mitigation of chondrocyte inflammation and joint-associated tissues [6,19,20]. In vitro and clinical studies showed that both H. procumbens and its bioactive secondary metabolites, the iridoid harpagoside, exert some anti-inflammatory effects and can improve pain and movement limitation in subjects with OA in the lower extremities [7,8]. Several evidences demonstrated that phenylethanoids can modulate different molecular pathways underlying inflammatory responses in human cells in vitro and inhibit inflammation in different tissues in vivo [21,22]. Based on this evidence, in this study, we demonstrated the inhibitory effects against inflammation of a leaf extract of V. thapsus, compared to a standardized extract of H. procumbens (harpagoside 12%), on the major inflammation signaling pathways in LPS-activated RAW 264.7 and in IL-1β activated HC cells [23]. After characterizing the profile and amount of PhEGs and the total polyphenol content of the extract, which were estimated in 9.92% (W/W) and 124.0 ± 0.7 mg gallic acid equivalent (GAE)/g of extract, respectively, we found that both extracts were not toxic at all concentrations tested after 24 h, but rather, they seemed to stimulate cell proliferation. Conversely, after 6 days of treatment, H. procumbens exerted a dose dependent cytotoxicity, while V. thapsus showed only a slight cytotoxic effect (5.8 ± 0.03) % at the higher dosage of 200 µg/mL in RAW 264.7 cells (Figure 2A,B). Macrophages play an important role in the initiation, maintenance and resolution of inflammation. In fact, during inflammation, infiltrating macrophages and neutrophils release nitric oxide (NO•), which is known to mediate the inflammatory response in tissues [24,25]. Inflammatory responses in macrophages induced by LPS also include increased expression of iNOS with consequent production of NO• [26]. Subsequently, we first focused on the LPS-induced NO• production, demonstrating that both extracts were able to significantly decrease NO• production after 6 days of pre-treatment (about 40%), while after 24 h, no significant change in NO• levels were detected in RAW 264.7 cells (Figure 3). Therefore, V. thapsus anti-inflammatory properties were similar to those of H. procumbens, but without concomitant cytotoxic effects, according to previous results about V. thapsus extracts or PhEGs compounds, performed in different experimental models [12,14,17,27,28]. After confirming the anti-inflammatory effect of the extract on macrophages, we proceeded to assess the inhibitory action on the cartilage inflammatory process in an in vitro model. In osteoarthritic condition, chondrocytes express IL-1, which induces the expression of MMPs and TNF-α. IL-1 and TNF-α increase the prostaglandin E 2 (PGE 2 ) synthesis by stimulating the COX-2 gene expression and upregulate the production of nitric oxide via iNOS [2,18,19]. In addition, IL-1β and TNF-α can also induce other proinflammatory cytokines, including IL-6 [29]. Based on this evidence, we evaluated the expression levels of cytokines and pro-inflammatory enzymes and metalloproteases involved in the inflammatory process of OA. Our qRT-PCR data showed that V. thapsus and H. procumbens extracts, at the used concentrations (50 and 100 µg/mL) and at both time points (24 h and 6 days), are able to decrease the expression levels of IL-1β and IL-6 pro-inflammatory cytokines, together with the inflammatory enzymes COX-2 and iNOS, compared to IL-1β-induced HC. However, V. thapsus extract exhibited a greater antiinflammatory effect on IL-6, COX-2 and iNOS at the concentration of 100 µg/mL for 6 days, compared to H. procumbens extract at all concentrations and time points analyzed ( Figure 5). The present study's obtained data agreed with the several pieces of evidence reported in the scientific literature, showing that PhEGs can differently modulate MAPK, NF-κB, JAK-STATs and Nrf2 pathways, providing inhibitory effects on different inflammatory mediators, such as TNF-α, IL-6, NO and ROS generation [28]. Furthermore, it was observed that the presence of two adjacent hydroxide groups in PhEGs molecular structures, as it is for most PhEGs compounds present in our extract (Table S1), is probably linked to a more marked anti-inflammatory activity [30]. Matrix metalloproteinases (MMPs) are well established to play key roles in OA through degradation of the extracellular matrix, inducing degenerative changes in joint cartilage [31,32]. In our in vitro experimental model of OA, both extracts decreased mRNA expression levels of the examined MMPs induced by the IL-1β exposure, but in particular, the PhEGs compounds present in the extract at the concentration of 100 µg/mL for 6 days, markedly reduced mRNA levels of MMP1, MMP8 and MMP9 compared to H. procumbens extract at the same experimental condition ( Figure 6). These preliminary results, although they require further deep studies, suggest for the first time that V. thapsus leaf extract phytocomplex, with its anti-inflammatory activity, has the potential to counteract the response of HC to the inflammatory insults and tissue degeneration that typically characterize the osteoarthritic disease.

Plant Material and Extraction Procedure
The final extract called Verbalief ® (product batch number EXPS-01201907-115), was performed and provided by Bionap (Belpasso, Catania, Italy). Verbascum thapsus L. leaves (medicinal name has been unified using the Kew Medicinal Plant Names Services, https://www.kew.org, accessed on 23 August 2021) were collected at the beginning of the second year of growing before the beginning of the flowering stage, in a cultivated field in the area of Mount Etna (37 • 35 34 N, 14 • 58 27 E, Catania, Italy). The taxonomic identification was confirmed by the pharmaceutical botanist Prof. R. Acquaviva, Department of Drug and Health Sciences, University of Catania (Italy). A voucher specimen (No. 02/19) was deposited in the same department. After harvesting, the plant material was cleaned and dried at 40 • C for 72 h in a ventilated oven. Subsequently, it was ground and blended before the extraction that was performed in ten times the volume of osmotized water at 80 • C under continuous stirring for five hours. The obtained solution was filtered and concentrated; the yield of the leaf extract, compared to 100 g of dried plant material, was about 30.00%. The extract was subsequently resuspended in water and supported on maltodextrin by spray drying.

Determination of Total Phenolic Content
The total phenolic content of V. thapsus leaf extract was determined spectrophotometrically by the Folin-Ciocalteu method, as previously described by Genovese [33], and compared to a calibration curve of a known amount of gallic acid used as standard. The absorbance was measured at 765 nm with a UV-1601 spectrophotometer (Shimadzu, Milan, Italy). The total polyphenols were estimated as gallic acid equivalent (GAE) and expressed in mg GAE/g extract (dw) ± standard deviation (SD). The data were obtained from three independent determinations.

Determination of Phenylethanoid Glycosides by HPLC-DAD
The polyphenolic fingerprinting of the extract was defined by high-pressure liquid chromatography (HPLC). The analyses were performed as above described [34]. Briefly, samples were dissolved in 1 mL of dimethylformamide/water solution (9:1) with a final concentration of 10 mg/mL. HPLC-DAD analyses were carried out in duplicate and performed using a Shimadzu LC 20, equipped with a diode array detector (DAD) and with a 150 × 4.6 mm i.d., 2.7 µm Ascentis Express C 18 column; the mobile phases: H 2 O/H 3 PO 4 (99:1, solvent A), MeOH/CAN/H3PO4 (49,5:49,5:1, solvent B); the gradient used was: concentration of the solvent A of 95% going to 77% (34 min), maintained at 77% (3 min), 74% (60 min), 60% (85 min), 20% (90 min) and 0% (92 min); total time 105 min. The column temperature was maintained at 25 • C. The flow was 1 mL/min and the injection volume was 5 µL. The chromatogram profiles were recorded from 190 to 500 nm and monitored at 280 and 330 nm ± 2 nm and the quantitative analysis was performed at wavelengths of 330 nm ± 2 nm.

Cell Cultures
Two different cell lines, mouse leukemic monocyte-macrophage (RAW 264.7, Sigma-Aldrich, Milan, Italy) and human adult chondrocytes (HC, Sigma-Aldrich) were used in this study. RAW 264.7 cells were cultured in EMEM (Sigma-Aldrich) supplemented with 2 mM L-Glutamine (Euroclone, Milan, Italy), 1% Non-Essential Amino Acids (Sigma-Aldrich), 10% Fetal Bovine Serum (FBS, Sigma-Aldrich) and penicillin/streptomycin/amphotericin (PSA) (Euroclone), while HC cells were grown in Chondrocyte Growth Medium (Sigma-Aldrich) containing PSA. Cells were maintained in a humidified environment at 37 • C and 5% CO 2 /95% air atmosphere and cultured in T75 flasks. The medium was replaced twice a week and cells were split at about 60-80% of confluence. Treatments with V. thapsus and H. procumbens were performed by adding different concentrations of each extract (50, 100 and 200 µg/mL) to the culture medium for 24 h and 6 days before inducing inflammation in cells with LPS (1 µg/mL) (O26:B6 E. coli, Sigma-Aldhric) and Il-1β (10 ng/mL) (recombinant human Il-1ß (PeproTech EC, London, UK)). Next, the medium was removed and further analyses were carried out.

Cell Viability Assays
The cytotoxic effect of the V. thapsus was evaluated by colorimetric assay MTT [3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] (Sigma-Aldrich), as previously reported [35]. For MTT 5 × 10 3 cells/well were cultured in 96 wells and after 24 h, a fresh medium containing several concentrations of V. thapsus and H. procumbens was added. Following 24 h and 6 days of incubation, the medium was removed, LPS (1 µg/mL) (O26:B6 E. coli, Sigma-Aldhric) and Il-1β (10 ng/mL) (recombinant human Il-1ß (PeproTech EC, London, UK)) was added for 24 h. After the cells were washed and 200 µL of MTT solution (1 mg/mL in FBS-free medium) was added to each well and incubated for 2 h at 37 • C and 5% CO 2 . Following 2 h of incubation, the medium was removed, each well was washed two times using cold PBS, and the formed crystals were melted using 200 µL of DMSO. The absorbance at 570 nm was read using a synergy HT plate reader (BioTek Instruments, Inc., Winooski, VT, United States).

NO• Production
The inhibitory effect of V. thapsus on NO• production was determined by measuring nitrite levels using the Griess reagent (Sigma-Aldrich) according to manufacturer instructions and as previously described [36,37]. Briefly, LPS-induced RAW 264.7 cells were cultured in a medium containing V. thapsus (50 and 100 µg/mL) for 24 h and 6 days. After 24 h and 6 days of treatment, 100 µL of the culture medium was mixed with an equal volume of Griess reagent and incubated at room temperature for 10 min according to manufacturer instructions. The nitrite content in culture media was determined at 540 nm using a synergy HT plate reader (BioTek Instruments, Inc.).

Quantitative Real-Time PCR (qRT-PCR)
For qRT-PCR analyses, total RNA, from LPS-induced RAW 264.7 and Il-1β-stimulated HC cells treated with V. thapsus, was isolated using Rneasy Mini Isolation Kit (Qiagen, Germantown, MD, USA) and quantified as previously described [38,39]. Three independently isolated and cultured samples were used. cDNA was synthesized from 1 µg of total RNA using ImProm-II Reverse Transcription System (Promega, Milan, Italy). Quantitative PCR was performed using SYBR Green method on a 7900HT Real Time PCR (Applied Biosystems). Specific primers for each of the investigated molecular endpoints were designed using primer blast and selecting exon-exon junctions on mRNA as a target region for annealing. Each sample was tested in triplicate and gene expression was assessed using the 2-∆∆Ct method. RNA from control cells was used as a reference for relative expression quantitation. The following PCR primers were used: IL-1β, IL-6, iNOS, COX2 and MMPs (1, 8, 9 and 13). Oligonucleotide sequences are reported in Table 2. Results were normalized to the levels of Glyceraldehyde 3-Phosphate Dehydrogenase (GAPDH).

Statistical Analysis
Statistical analyses were performed by one-way and two-way ANOVA. Holm methods have been used as post hoc tests when the ANOVA reported statistically significant differences, to evaluate the differences between the individual time-points or treatment groups. For all experiments, * p < 0.05 was considered to be significant.

Conclusions
In conclusion, most of the results obtained in this study, based on an in vitro model of OA, are comparable to those exerted by H. procumbens extract, but with fewer cytotoxic effects. In addition, for some of the considered biomarkers, such as IL-6, COX-2, iNOS and MMPs 1, 8 and 9, the inhibitory effects resulted more remarkably, demonstrating that V. thapsus leaf aqueous extract exhibits a good anti-inflammatory activity and suggesting that this extract, which certainly needs to be supported by further in-depth studies, could be a new potential candidate in the treatment of the early stages of osteoarthritis or mild joint inflammation.
Supplementary Materials: The following are available online. Figure S1: HPLC-DAD of standards phenylethanoids glycosides used in this study; Table S1: Main representative chemical structures of phenylethanoid glycosides of V. thapsus leaf aqueous extract.