In Vivo Hepatoprotective and Nephroprotective Activity of Acylated Iridoid Glycosides from Scrophularia hepericifolia

Simple Summary Chronic liver disease is a major life-threating problem worldwide. Deaths due to liver disease constituted 2.4% of total deaths globally in 2017. Yet most of the available drugs for improving liver conditions are of natural origin. More research is necessary to discover agents that are more effective. Previously, the extract of Scrophularia hypericifolia showed promising protection of liver and kidney tissues against induced toxicity in experimental animals. The current phytochemical study aimed to identify the active molecules in the extract. Nine iridoid glycoside derivatives were identified from the plant extract, including four new compounds after extensive chromatographic purification. The structures were identified by applying various spectroscopic methods. Biological evaluation for protective effect on liver and kidney tissues was conducted on compounds isolated with a high enough yield. Serum and tissue parameters as well as histopathological studies were conducted for efficacy evaluation. Two of the new compounds showed the best protection of liver and kidney tissues as indicated by the studied parameters. These findings indicated that natural products could provide solutions to health problems. The role of nutraceuticals in managing liver problems is a promising field for further studies. Abstract Phytochemical investigation of the chloroform fraction obtained from Scrophularia hypericifolia aerial parts led to the isolation of nine acylated iridoid glycosides. The new compounds were identified as 6-O-α-L(2″-acetyl, 3″,4″-di-O-trans-cinnamoyl) rhamnopyranosyl-6′-acetyl catalpol (6′-acetyl hypericifolin A) (1), 6-O-α-L(2″, 4″-diacetyl, 3″-O-trans-cinnamoyl) rhamnopyranosyl-6′-acetyl catalpol (6′-acetyl hypericifolin B) (2), 6-O-α-L(2″-acetyl, 3″,4″-di-O-trans-cinnamoyl) rhamnopyranosyl catalpol (hypericifolin A) (3) and 6-O-α-L(2″, 4″-diacetyl, 3″-O-trans-cinnamoyl) rhamnopyranosyl catalpol (hypericifolin B) (4). Previously reported compounds were identified as laterioside (5), 8-O-acetylharpagide (6), 6-O-α-L(4′-O-trans-cinnamoyl) rhamnopyranosyl catalpol (7), lagotisoside D (8) and harpagoside (9). Identification achieved via analyses of physical and spectral data including 1D, 2D NMR and High Resolution Electrospray Ionization Mass spectroscopy (HRESIMS). Compounds 2–4 and 6 were subjected to biological evaluation against paracetamol-induced toxicity. The biochemical parameters aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP) and gamma glutamyl transpeptidase (GGT) as well as total bilirubin were used to access the liver condition. Measurement of serum levels of urea, creatinine, sodium and potassium cations were indicators for kidney condition. Liver and kidney samples were subjected to histopathological study. The best protection was found in the group treated with 3 followed by 4 and 6, while 2 was almost inactive.

Scrophularia species are rich in iridoid glycosides. The hepatoprotective effect of S. koelzii was traced to the iridoid glycosides scropolioside A, harpagoside and koelzioside [12]. Scropolioside D2 and harpagoside B isolated from S. deserti showed significant anti-inflammatory and antidiabetic effects [9]. Iridoid glycosides from S. umbrosa were reported to have inhibitory activity on nitric oxide production [15]. Two iridoid glycosides with analgesic activity were reported from S. kotscyhana [16].
Chronic liver diseases are one of the major causes of morbidity and mortality globally. In 2017, liver diseases caused more than 1.32 million deaths in females and 883,000 in males, compared with less than 899,000 deaths in 1990 worldwide. Deaths due to liver disease constituted 2.4% of total deaths globally in 2017 compared with 1.9% in 1990 [17]. Nonalcoholic fatty liver disease (NAFLD) is characterized by inflammation and progressive tissue degeneration. It affects about 5% of the general adult population, and 20% of obese people [18]. In the absence of specific approved drugs on the market for the treatment of NAFLD, some nutraceuticals could contribute to the improvement of lipid infiltration and/or biochemical parameters. A few nutraceuticals were studied clinically for their effects against NAFLD, including silymarin (Sil), vitamins E and D, curcumin, resveratrol and the total extracts of Salvia milthiorriza [19].
Scrophularia hypericifolia growing in Saudi Arabia was reported to have hepatoprotective and nephroprotective activities [20]. In the current study, the secondary metabolites of the plant were isolated and identified by various spectroscopic methods. Pure compounds isolated in good yield were evaluated for their hepatoprotective and nephroprotective activities.

Structure Elucidation
The total extract of the aerial parts of S. hypericifolia was partitioned by liquid-liquid fractionation to a petroleum ether-soluble fraction, chloroform-soluble fraction and nbutanol-soluble fraction. Extensive chromatographic fractionation and purification of the chloroform-soluble fraction resulted in the isolation of compounds 1-9 ( Figure 1) in addition to cinnamic acid. The known compounds were identified by comparison of their measured spectra with the literature data. All the assignments of 1 H and 13 C values were based on DEPT135, COSY, HSQC, HMBC, H2BC, NOESY and/or ROESY experiments. Compound 5 was identified as laterioside, an iridoid glycoside first identified from S. lateriflora [21] then reported from S. deserti [22] as well as many Verbascum species [23]. Compound 6 was identified as 8-O-acetylharpagide common in Scrophularia species [24]. 8-O-Acetylharpagide is reported to have antibacterial, anti-inflammatory, vasoconstrictor and cancer chemopreventive activities [25][26][27][28]. Compound 7 was identified as 6-O-α-L(4"-O-trans-cinnamoyl) rhamnopyranosyl catalpol. Compound 7 is reported in this work for the first time from Scrophularia species. Previously, it was characterized from Gmelina arborea (Lamiaceae) and Buddleja polystachya (Scrophulariaceae) [29,30]. The data of 8 enable the identification of the structure as lagotisoside D previously isolated from Lagotis yunnanensis and S. dentata [31,32]. Compound 9 data were identical to those reported for harpagoside [33]. Harpagoside was first isolated from Harpagophytum procumbens F. Pedaliaceae (devil's claw) [34]. Harpagoside proved to have potent anti-rheumatic, anti-inflammatory and analgesic effects [35].
(devil's claw) [34]. Harpagoside proved to have potent anti-rheumatic, anti-inflammatory and analgesic effects [35].    Figures S1-S7). 13 C NMR of 1 showed ten carbon resonances sorted by DEPT 135 experiments into 7 X CH including two oxygenated, two vinyls and one di-oxygenated methins, one CH2O and one oxygenated quaternary carbon. Such an array (Tables 1 and 2) pointed to an iridoid skeleton of a catalpol skeleton [36]. Another set of carbons, including four CHO, CH2O and one di-oxygenated CH were assigned to glucose moiety attached to C-1 of the iridoid aglycone. The relative stereochemistry of the iridoid aglycone was confirmed from the NOE correlations between the protons H-1 and H-6; H-6 and H-7; H-6 and H-10, all with α-orientation (Figures S17-S20). The chemical shift of C-6 at C 83.74 was diagnostic for a substituted position. The six carbon resonances at C 96. 41, 71.05, 69.92, 69.45, 66.89 in addition to the methyl carbon at C 16.53 correlated by the HSQC experiment with the proton doublets at C 1.27 in 1 H NMR spectra were assigned to a rhamnosyl substituent ( Figure S9). HMBC correlations between the H-6 doublet at H 4.03 and C-1" of rhamnose at C 96.41C as well as the NOE correlation between H-6 and H-1" singlet at H 5.13 all support the position of a rhamnosy moeity at C-6. Many iridoid glycosides including 7 and 8 bearing a rhamnose moiety at C-6 were previously reported (Figures S10-S14) [29][30][31][32]. Compared with compound 7, 1 bears an additional two acetyls and one trans-cinnamoyl (Tables 1 and 2). The H-6' and C-6' of the glucose moiety in 1 were downfield shifted to H 4.28 (dd, J = 5, 12 Hz), 4.47(d, J = 6.8 Hz) and C 62.74 compared with normal values of glucose [33], indicating acylation at that position. NOE correlations were observed between the acetyl protons at H 2.07 (s) and H-6' at H 4.28 (dd, J = 5, 12 Hz), 4.47(d, J = 6.8 Hz), H-5' at H 3.52 (m) and H-4' at 3.43 (m) (Figures S17-S20). Further evidence for C-6' acetylation was obtained from the HMBC correlations between both H-6' protons, and the methyl protons at H 2.07 with the acetyl carbonyl at C 171.40 (Figures S10-S14). The H2BC experiment was extensively used to assign protons and carbons, especially those of the rhamnose moiety (Figures S15 and S16). The chemical shifts of H-2'', H-3'', H-4'' at H 5.43 (bs), 5.53 (dd, J = 2.7, 10 Hz) and 5.32 (t, J = 10 Hz) indicated that the three positions are acylated. The assignments of the two cinnamoyl and the left acetyl substituents to their exact positions on C-2''-C4'' were achieved via detailed analyses of the HMBC correlations. The broad singlet assigned for H-2'' at H 5.43 as well as the acetyl methyl singlet at H 2.19 showed three bond correlations with the acetyl carbonyl at C 170.32. HMBC correlations were observed between H-3'' at 5.53 (dd, J = 2.7, 10 Hz), the two trans-oriented cinnamoyl doublets at H 6.40 and 7.61 and the carbonyl signal   Figures S1-S7). 13 C NMR of 1 showed ten carbon resonances sorted by DEPT 135 experiments into 7 X CH including two oxygenated, two vinyls and one di-oxygenated methins, one CH 2 O and one oxygenated quaternary carbon. Such an array (Tables 1 and 2) pointed to an iridoid skeleton of a catalpol skeleton [36]. Another set of carbons, including four CHO, CH 2 O and one di-oxygenated CH were assigned to glucose moiety attached to C-1 of the iridoid aglycone. The relative stereochemistry of the iridoid aglycone was confirmed from the NOE correlations between the protons H-1 and H-6; H-6 and H-7; H-6 and H-10, all with α-orientation (Figures S17-S20). The chemical shift of C-6 at δ C 83.74 was diagnostic for a substituted position. The six carbon resonances at δ C 96.41, 71.05, 69.92, 69. 45,66.89 in addition to the methyl carbon at δ C 16.53 correlated by the HSQC experiment with the proton doublets at δ C 1.27 in 1 H NMR spectra were assigned to a rhamnosyl substituent ( Figure S9). HMBC correlations between the H-6 doublet at δ H 4.03 and C-1 of rhamnose at δ C 96.41δ C as well as the NOE correlation between H-6 and H-1 singlet at δ H 5.13 all support the position of a rhamnosy moeity at C-6. Many iridoid glycosides including 7 and 8 bearing a rhamnose moiety at C-6 were previously reported (Figures S10-S14) [29][30][31][32]. Compared with compound 7, 1 bears an additional two acetyls and one trans-cinnamoyl (Tables 1 and 2). The H-6 and C-6 of the glucose moiety in 1 were downfield shifted to δ H 4.28 (dd, J = 5, 12 Hz), 4.47(d, J = 6.8 Hz) and δ C 62.74 compared with normal values of glucose [33], indicating acylation at that position. NOE correlations were observed between the acetyl protons at δ H 2.07 (s) and H-6 at δ H 4.28 (dd, J = 5, 12 Hz), 4.47(d, J = 6.8 Hz), H-5 at δ H 3.52 (m) and H-4 at 3.43 (m) (Figures S17-S20). Further evidence for C-6 acetylation was obtained from the HMBC correlations between both H-6 protons, and the methyl protons at δ H 2.07 with the acetyl carbonyl at δ C 171.40 (Figures S10-S14). The H2BC experiment was extensively used to assign protons and carbons, especially those of the rhamnose moiety (Figures S15 and S16). The chemical shifts of H-2", H-3", H-4" at δ H 5.43 (bs), 5.53 (dd, J = 2.7, 10 Hz) and 5.32 (t, J = 10 Hz) indicated that the three positions are acylated. The assignments of the two cinnamoyl and the left acetyl substituents to their exact positions on C-2 -C4 were achieved via detailed analyses of the HMBC correlations. The broad singlet assigned for H-2 at δ H 5.43 as well as the acetyl methyl singlet at δ H 2.19 showed three bond correlations with the acetyl carbonyl at δ C 170.32. HMBC correlations were observed between H-3 at 5.53 (dd, J = 2.7, 10 Hz), the two trans-oriented cinnamoyl doublets at δ H 6.40 and 7.61 and the carbonyl signal at δ C 165.93 assigned to C-9 , while the H-4 triplet at δ H 5.32, and the two trans-oriented cinnamoyl doublets at δ H 6.50 and 7.70 were correlated to the carbonyl signal at δ C 166.31 assigned to C-9 (Figures S10-S14). Based on the provided evidence, 1 was identified as 6-O-α-L(2 -acetyl, 3 ,4 -di-O-trans-cinnamoyl) rhamnopyranosyl-6 -acetyl catalpol and was given the name 6 -acetyl hypericifolioside A.  (Figures S80  and S81). The appearance of H-6 and C-6 of glucose at δ H 3.67 (dd, J = 6.5, 11.9 Hz), 3.96 (d, J = 10.9 Hz) and δ C 61.58 in 3 indicated that C-6 hydroxyl is free ( Figures S50-S55). Spectra of 3 were repeated in pyridine-d6 to resolve overlapping peaks in the CD 3 OD spectra, such as H-1, H-4 and H-1" (Table 1) ( Figures S66-S70). The assignment of the two cinnamoyl moieties at 3", 4" and the acetyl group at 2" of the rhamnose moiety followed the same spectral evidence explained for 1. The broad singlet assigned for H-2" at δ H 5.43 as well as the acetyl methyl singlet at δ H 2.19 showed three bond correlations with the acetyl carbonyl at δ C 170.31. HMBC correlations were observed between H-3" at δ H 5.55 (dd, J = 3.2, 10.2 Hz), the two trans-oriented cinnamoyl doublets at δ H 6.39 and 7.60 and the carbonyl signal at δ C 165.89 assigned to C-9 , while the H-4" triplet at δ H 5.33 and the two trans-oriented cinnamoyl doublets at δ H 6.49 and 7.71 correlated to the carbonyl signal at δ C 166.30 assigned to C-9 (Figures S58-S60 and S106). Consequently, 3 was identified as 6-O-α-L(2"-acetyl, 3",4"-di-O-trans-cinnamoyl) rhamnopyranosyl catalpol and was given the trivial name hypericifolioside A.

Biological Evaluation
The total extract of S. hypericifolia showed promising hepatoprotective and nephroprotective activities [20]. Compounds 2-4 and 6 isolated in good yield were subjected to biological testing against paracetamol (Pa)-induced liver and kidney toxicities. Toxic doses of Pa produce fatal hepatic necrosis in humans and other mammals, including rats and mice [37]. Toxic doses of Pa cause saturation of the sulfation and glucoronidation routes of metabolism. As an alternative to get rid of the extra Pa, the cytochrome P450 enzymes are enhanced to oxidize a higher percentage of Pa molecules to the highly reactive N-acetyl-p-benzoquinone imine (NAPQI) species. NAPQI's loss of one electron results in the formation of semi-quinone radicals with an extremely short half-life. It is then rapidly conjugated with the sulphydryl donor glutathione (GSH), resulting in the depletion of the liver GSH pool [38]. Excessive formation of NAPQI as well as glutathione store depletion leads to covalent binding of NAPQI to vital proteins and the lipid bilayer of hepatocyte membranes and enhances lipid peroxidation. These consequences lead to hepatocellular death and centrilobular liver necrosis [39]. The transport system of the hepatocytes was impaired, leading to the leakage of plasma membrane [40], reflected by increases in serum enzyme levels. Treatment of animals with Pa resulted in a dramatic elevation of transaminases (aspartate aminotransferase (AST), alanine aminotransferase (ALT)) and alkaline phosphatase (ALP) levels. Severe jaundice is expressed by the elevation of serum bilirubin levels ( Figure 2 and Table S1) [41].  Figures S104 and S105). The absence of one acetyl group was clear in the NMR data and fully supported by the MS data ( Figures S82-S89). In the spectra of 4, the appearance of H-6' and C-6' of glucose at H 3.65 (dd, J = 6.8, 12 Hz), 3.94 (d, J = 12 Hz) and C 61.55 indicated that 4 is the C-6' deacetylation derivative of 2. Similar to 1-3, the acylations in 4 were located at H-2'' (H 5.88, s), H-3'' (H 5.94, bd, J = 9.7 Hz) and H-4'' (H 5.71, bd, J = 9.7 Hz) based on their downfield shift values. The trans-cinnamoyl moiety was assigned to position 3'' based on the HMBC correlation between H-3'' at H 5.94 and the cinnamoyl carbonyl at C 165.88 (Figures S92 and S93). Compound 4 was identified as 6-O-α-L(2", 4"diacetyl, 3"-O-trans-cinnamoyl) rhamnopyranosyl catalpol and was given the trivial name hypericifolioside B.

Biological Evaluation
The total extract of S. hypericifolia showed promising hepatoprotective and nephroprotective activities [20]. Compounds 2-4 and 6 isolated in good yield were subjected to biological testing against paracetamol (Pa)-induced liver and kidney toxicities. Toxic doses of Pa produce fatal hepatic necrosis in humans and other mammals, including rats and mice [37]. Toxic doses of Pa cause saturation of the sulfation and glucoronidation routes of metabolism. As an alternative to get rid of the extra Pa, the cytochrome P450 enzymes are enhanced to oxidize a higher percentage of Pa molecules to the highly reactive N-acetyl-p-benzoquinone imine (NAPQI) species. NAPQI's loss of one electron results in the formation of semi-quinone radicals with an extremely short half-life. It is then rapidly conjugated with the sulphydryl donor glutathione (GSH), resulting in the depletion of the liver GSH pool [38]. Excessive formation of NAPQI as well as glutathione store depletion leads to covalent binding of NAPQI to vital proteins and the lipid bilayer of hepatocyte membranes and enhances lipid peroxidation. These consequences lead to hepatocellular death and centrilobular liver necrosis [39]. The transport system of the hepatocytes was impaired, leading to the leakage of plasma membrane [40], reflected by increases in serum enzyme levels. Treatment of animals with Pa resulted in a dramatic elevation of transaminases (aspartate aminotransferase (AST), alanine aminotransferase (ALT)) and alkaline phosphatase (ALP) levels. Severe jaundice is expressed by the elevation of serum bilirubin levels ( Figure 2 and Table S1) [41].

Hepatoprotective Effect
Administration of Sil, at a dose of 10 mg/kg (20.7 µmol/kg) prior to Pa resulted in a significant correction (p < 0.001) in the elevated AST (37.74%), ALT (43.29%), gamma glutamyl transpeptidase (GGT) (37.53%), ALP (27.31%) and bilirubin (54.15%) levels in the corresponding group of rats ( Figure 2 and Table S1). Sil acts by several mechanisms including an antioxidant effect by scavenging prooxidant free radicals and via restoring the concentration of GSH. Sil also restores the normal cellular membrane function, resulting in protection against xenobiotic injury. Sil also initiates the synthesis of ribosomal RNA via activation of DNA polymerase-I and steroid-like action in regulating DNA transcription and enhancement of protein synthesis necessary for the regeneration of liver cells [42,43].
Treatment of rats with 3 at 20.7 µmol/kg doses prior to Pa showed a significant (p < 0.01; 0.001) reduction by 23.26, 33.71, 37.95, 16.67 and 27.70% in the elevated levels of AST, ALT, GGT, ALP and bilirubin. Compound 4 showed less protection, expressed as 13.86, 26.72, 36.14, 13.99 and 25.00% reductions in the levels of AST, ALT, GGT, ALP and bilirubin. Compound 6 showed weaker effects on the serum biochemical parameters, while 2 was almost inactive (Figure 2). The impact of the tested compounds was also evaluated on total protein (TP) and non-protein sulfhydryl groups (NP-SH) levels in liver cells ( Figure 3A, Figure 4 and Table S2). Compound 3 restored TP contents to about 50% of that of Sil. The effect of 3 restoring NP-SH (3.43 ± 0.30) was slightly less than the standard drug Sil (3.37 ± 0.28). The effects of 4 were less than 3, followed by 6. The results of the histopathological study were in support of the serum biochemical and tissue parameters obtained. Compared with the normal hepatocytes ( Figure 5A), the liver samples of the group only treated with Pa ( Figure 5B) showed severe damage, expressed as portal vessel congestion, necrosis and infiltration. The Sil-treated group indicated that Sil restores the liver cell architecture closer to the normal state ( Figure 5C) with little congestion. The group treated with 3 expressed a great level of protection ( Figure 5D) where the appearance of cells was almost normal. Mild focal necrosis and portal tract congestion were observed in the liver specimen of the group treated with 4 ( Figure 5E), indicating a moderate level of protection. Less protection was evident from treatment with 6 as the cells suffered more from degeneration, portal congestion and inflammatory cell infiltration.
Biology 2021, 10, x 3 of 18 cloudy swelling in tubules, while the group treated with 3 ( Figure 7D) expressed moderate cortical atrophy and moderate reduction in glomerular size. More damage was observed in groups treated with 4 ( Figure 7E) and 6 ( Figure 7F).

Nephroprotective Effect
The kidney is a vital organ responsible for the regulation of plasma ionic composition and the elimination of nitrogenous end metabolic waste products such as urea, creatinine and uric acid [44]. Elevations of these parameters are reliable parameters for nephrotoxicity [45]. Pa nephrotoxicity is reflected by elevated serum levels of creatinine, urea, sodium and potassium (Table S3). The levels of NP-SH and TP were diminished (Table S2). Treatment with Sil resulted in marked improvement of all these parameters. The group treated with 3 showed the most effective degree of protection among the tested compounds. Prior treatment with Sil resulted in a 39.87% reduction in urea, a 49.88% reduction in creatinine, a 57.46% reduction in sodium and a 46.03% reduction in potassium levels compared with 24.61%, 36.00%, 33.88% and 35.47% reductions in the same parameters, respectively, as a result of the administration of 3 ( Figure 6, Table S3). Compound 3 was as effective as Sil in restoring NP-SH groups and slightly less in restoring TP (Figures 3B and 4, and Table S2). Less improvement was observed with treatment with 4, followed by 6, while 2 was totally ineffective. The normal kidney cells structure is presented in Figure 7A. Histopathological study revealed that the kidney cells of rats treated with Pa expressed severe changes ( Figure 7B), such as degenerated corpuscle, reduced size of glomeruli, shrinkage of Bowman capsule, cloudy swelling in tubules and inflammatory infiltrate. Pre-administration of Sil ( Figure 7C) resulted in nearly normal glomeruli and mild cloudy swelling in tubules, while the group treated with 3 ( Figure 7D) expressed moderate cortical atrophy and moderate reduction in glomerular size. More damage was observed in groups treated with 4 ( Figure 7E) and 6 ( Figure 7F).
Many nutraceuticals containing well-known traditional plants are utilized for their liver and/or kidney health benefits. Extracts of Nigella sativa [46], Chenopodium ambrosioides [47], Hemidesmus indicus [48], Bryophyllum pinnatum [49] and Verbascum thapsus [50] are reported to have hepatoprotective and/or nephroprotective effects. Interestingly, Verbascum thapsus is rich in iridoid contents [50]. One of its components, the iridoid glycoside aucubin, was reported to have wide range of pharmacological activities, including hepatoprotective and antioxidant effects [51]. The hepatoprotective effects of related iridoid glycosides were reviewed [52]. Molecular docking analysis was performed to assess the binding efficiencies of iridoids to the active sites of human CYP3A4 enzyme. Selected iridoids were evaluated for their hepatoprotective effects using quantitative real-time PCR to measure the expression levels of CYP3A4 mRNA in HepG2 cells. The tested iridoids showed an inductive effect on CYP3A4 mRNA levels in the HepG2 cells. The study results indicated that the hepatoprotective effects were caused by the facilitation of drug metabolism and the reduction of both mitochondrial dysfunction and oxidative stress [53].

General Experimental Procedures
Melting points were determined in open capillary tubes using a Thermosystem FP800 Mettler FP80 central processor supplied with an FP81 MBC cell apparatus and are shown uncorrected. Ultraviolet absorption spectra were measured on a Unicum Heyios a UV-Visible spectrophotometer. A Jasco P-2000 Polarimeter was used to measure the optical rotations. 1 H, 13 C-NMR and 2D-NMR data were measured on a Bruker UltraShield Plus 500 MHz spectrometer at NMR Unite, College of Pharmacy, Prince Sattam Bin Abdulaziz University, operated at 500 MHz for protons and 125 MHz for carbon atoms. Chemical shift values were reported in δ (ppm) relative to the residual solvent peaks. Coupling constants (J) are reported in Hertz (Hz). 2D-NMR experiments (COSY, HSQC, HMBC, H2BC, NOESY and/or ROESY) were performed utilizing the standard Bruker program. HRMS were determined by direct injection using a Thermo Scientific UPLC RS Ultimate 3000 Q Exactive hybrid quadrupole-Orbitrap mass spectrometer combining high-performance quadrupole precursor selection with high-resolution, accurate mass (HR/AM) Orbitrap™ detection. Direct infusion of isocratic elution acetonitrile/methanol (70:30) with 0.1% formic acid was used to flush the samples. Runtime was 1 min using nitrogen as auxiliary gas with a flow rate of 5 µL/min. A scan range from 160-1500 m/z was used.

Plant Material
The plants of Scrophularia hypericifolia Wydler (aerial parts) were described earlier [20] and collected in March, 2019 from Al-Qassim Province, Saudi Arabia. The plants were identical to the voucher specimen (# 13274) preserved at the herbarium of the Medicinal, Aromatic and Poisonous Plants Research Center (MAPPRC).

Extraction and Isolation
Air-dried powdered aerial parts (1000 g) were repeatedly extracted with 95% ethanol by percolation at room temperature till exhaustion. The solvent was distilled off under reduced pressure using a rotary vacuum evaporator at 40 • C to give 130 g residue. Part of the obtained dried extract (35 g) was dissolved in 800 mL of 40% aqueous ethanol and fractionated with light petroleum (500 mL × 3) to yield 9.25 g of light petroleum-soluble fraction, chloroform (500 mL × 4) to yield 15.32 g of chloroform-soluble fraction and n-butanol (400 mL × 2) to yield 3.15 g of n-butanol-soluble fraction.
The chloroform-soluble fraction was fractionated chromatographically using a silica gel column (150 × 7 cm i.d., 450 g) eluting with chloroform, followed by chloroform/methanol mixtures with a gradual increase in methanol content. Fractions of 200 mL each were collected, screened by TLC and similar fractions were pooled to yield 6 major fractions (A-F). Fraction A (1.51 g) eluted with 5% methanol in chloroform was further purified using an RP18 column (45 × 2 cm i.d., 40 g) eluting with 20% water in methanol to afford 85 mg of 1 and 328 mg of 2 after crystallization from methanol. Fraction B (0.93 g) eluted with 10% methanol in chloroform afforded 325 mg of 3 after purification on an RP18 column (45 × 2 cm i.d., 30 g) eluting with 20% water in methanol. Fraction C (2.73 g) eluted with 10% methanol in chloroform was chromatographed on an RP18 column (90 × 2.5 cm i.d., 90 g) eluting with 20% water in methanol to afford another 200 mg of 3. Further purification on a Sephadex LH20 eluting with 10% water in methanol afforded 272 mg of 4 from fractions 4-7. Fractions 11-14 were further purified using a Chromatotron equipped with 2 mm silica gel plates and 2% methanol in acetonitrile as the mobile phase to afford 17 mg of cinnamic acid and 92 mg of 5. Fraction D (1.43 g) eluted with 10% methanol in chloroform was chromatographed on an RP18 column (90 × 2.5 cm i.d., 45 g) eluting with 20% water in methanol to afford 439 mg of 6. Fraction E (0.67 g) eluted with 20% methanol in chloroform was further purified on a Sephadex LH20 eluting with 20% water in methanol followed by separation on a Chromatotron equipped with 2 mm silica gel plates and 15% methanol in chloroform to afford 32 mg of 7 and 58 mg of 8. Fraction F (0.43 g) eluted with 20% methanol in chloroform was chromatographed on an RP18 column (45 × 2 cm i.d., 90 g) eluting with 30% water in methanol to afford another 130 mg of 9.

Animals
Male Wistar albino rats of average weight (160-180 g) (age 8-10 weeks) were secured by the Experimental Animal Care Center, College of Pharmacy, Prince Sattam Bin Abdulaziz University, Al-Kharj, KSA. The animals were kept under controlled conditions of temperature (22 ± 2 • C), humidity (55%) and light/dark cycle (12/12 h). The animals were provided with Purina chow and free access to drinking water ad libitum [20]. The experimental protocol was approved by the Bioethical Research Committee at Prince Sattam Bin Abdulaziz University.

Hepatoprotective and Nephroprotective Activity
Rats were divided into seven groups of five animals. Group 1 was designated as the control group and received normal saline only. Groups 2-7 received Pa via the intraperitoneal route for two days at 500 mg/kg body weight. Group 2 did not receive any further treatment. Group 3 was treated with 10 mg/kg p.o. (20.7 µmol/kg) of Sil (Sigma-Aldrich, St. Louis, MO, USA) [20]. Groups 4-7 were treated with 20.7 µmole/kg body weight of compounds 2-4, 6. Treatment started 5 days prior to Pa administration and continued to the end of the experimental protocol. After 24 h, following the second dose of Pa, the animals were sacrificed using ether anesthesia. Blood samples were collected by heart puncture and the serum was separated to estimate the biochemical parameters. The liver and kidney tissues were immediately removed and properly treated for the determination of the tissue parameters. Representative pieces were immersed in 10% formalin for fixation to carry out the histopathological study.

Determination of Non-Protein Sulfhydryl Groups and Total Protein
The liver and kidney tissues were homogenized in 0.02 M EDTA using a Potter-Elvehjem type C homogenizer (Sigma-Aldrich, St. Louis, MO, USA). Homogenates equivalent to 100 mg of tissues were used for the measurements. Nonprotein sulfhydryl groups (NP-SH) were quantified by dilution of the homogenates with 4 mL of distilled water and 1 mL of 50% trichloroacetic acid (TCA) followed by shaking for 15 min. The supernatants were then mixed with 2 mL Tris buffer, pH 8.9 and 0.1 mL of 0.01 M of 5,5 -dithio-bis-(2nitrobenzoic acid) (DTNB). The absorbances were measured spectrophotometrically within 5 min at 412 nm against a reagent blank with no homogenate (Figure 4 and Table S3) [57]. TP were quantified by mixing homogenates with 0.7 mL of Lowry's solution followed by incubation in the dark for 20 min at room temperature. Diluted Folin's reagent (0.1 mL) was then added, and samples were incubated at room temperature in the dark for 30 min. The absorbances were then measured at 750 nm ( Figure 5 and Table S3) [58].

Statistical Analysis
An analysis of variance (ANOVA) test was used to evaluate the significance of the difference between the groups. Non-paired samples such as the control and Pa-treated group were compared for significance using Dunnett's test [59]. All the reported values are presented as mean ± S.E.

Histopathology
Water-ethanol mixtures with a gradual increase in ethanol were used to dehydrate liver and kidney samples till reaching absolute ethanol, cleared and infiltrated by immersion in increasing concentrations of ethanol (70-100%) and xylene (3 times, 1 h each) followed by paraffin wax (4 times, 1 h each). The tissues were placed by hot forceps in molds, chilled on cold plates and excess waxes were removed. Thin sections of 3 µm were obtained by a rotary microtome (Leitz 1512) and placed onto clean slides. The slides were drained and stained using Mayer's hematoxylin solution for 15 min after deparaffinization and hydration. The slides were then washed in lukewarm tap water for 15 min then immersed in 80% ethanol for two minutes and counterstained in eosin-phloxine solution for 2 min. The slides were then washed with 95% ethanol, absolute ethanol and xylene (2 min each) and finally mounted in resinous medium [60].

Conclusions
A phytochemical study of the aerial parts of Scrophularia hypericifolia resulted in the isolation of nine acylate iridoid glycosides. Four new compounds were identified as 6 -acetyl hypericifolioside A (1) 6 -acetyl hypericifolioside B (2) hypericifolioside A (3) and hypericifolioside B (4). In addition, five known compounds were identified as laterioside (5) 8-O-acetylharpagide (6) 6-O-L(4"-O-trans-cinnamoyl) rhamnopyranosyl catalpol (7) lagotisoside D (8) and harpagoside (9). Compounds 2-4 and 6 were tested against paracetamol-induced hepatic and renal toxicity. Compound 3 resulted in 23. 26, 33.71, 37.95, 16.67 and 27.70% reductions in the elevated levels of AST, ALT, GGT, ALP and bilirubin. It also reduced the levels of urea, creatinine, sodium and potassium by 24.61%, 36.00%, 33.88% and 35.47%, respectively. Compound 3 restored TP contents and NP-SH in both liver and kidney tissues to levels comparable to Sil. These findings were supported by histopathological study by the almost normal appearance of liver cells, while little degeneration was observed in kidney cells after treatment with 3. In some of the measured parameters, 3 was as protective as the standard drug Sil. Compounds 4 and 6 expressed less protection, while 2 was almost inactive.