pubs.acs.org/jnp Article Anti-inflammatory Flavanones and Flavones from Tephrosia linearis Richard Oriko Owor, Kibrom Gebreheiwot Bedane, Sebastian Zühlke, Solomon Derese, George Otieno Ong’amo, Albert Ndakala,* and Michael Spiteller* Cite This: https://dx.doi.org/10.1021/acs.jnatprod.9b00922 Downloaded via UPPSALA UNIV on March 11, 2020 at 01:12:18 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. ACCESS Metrics & More Read Online Article Recommendations sı Supporting Information * ABSTRACT: Phytochemical analysis of a methanol−dichloromethane (1:1) extract of the aerial parts of Tephrosialinearis led to the isolation of 18 compounds. Seven of these, namely, lineaflavones A−D (1−4), 6-methoxygeraldone (5), 8″-acetylobovatin (6), and 5-hydroxy-7-methoxysaniculamin A (7) are new compounds. The compounds were characterized based on their NMR and HRMSn data. The anti-inflammatory effects of the crude extract and isolated compounds were evaluated by measuring the levels of interleukins (IL-1β, IL-2, and IL-6), granulocyte-macrophage colony-stimulating factor (GM-CSF), and tumor necrosis factor-α (TNF-α) in lipopolysaccharide (LPS)-stimulated peripheral blood mononuclear cells (PBMCs). The crude extract inhibited the release of all cytokines except IL-1β, which slightly increased in comparison to the LPS control. All the tested compounds suppressed the production of IL-2, GM-CSF, and TNF-α. Whereas compounds 1, 2, 4−8, 10−15, 17, and 18 decreased production of IL-6, compounds 1, 2, 4, 7, 10, 13−15, and 17 inhibited the release of IL-1β. It is worth noting that most of the compounds tested showed a superior reduction in cytokines release compared to the reference drug ibuprofen. P to investigate Tephrosia species, the aerial parts of T. linearis were investigated phytochemically leading to the isolation and identification of 18 compounds, of which seven are new. Herein, are discussed the isolation, structure elucidation, and antiinflammatory effects of these compounds. To evaluate the antiinflammatory effects of the crude plant extract and the isolated compounds, lipopolysaccharide (LPS)-stimulated cytokine release of peripheral blood mononuclear cells (PBMCs) was quantified through measurement of the levels of IL-1β, IL-2, IL6, GM-CSF, and TNF-α. lants of the genus Tephrosia Pers. (Fabaceae) mainly inhabit tropical and subtropical regions with over 30 species occurring in Kenya.1 Many of the species have been used ethnomedicinally to alleviate diverse illnesses.2,3 Phytochemical investigations of some of these plants by our research group have led to the isolation of a chalcone,4 rotenoids,5 flavanonols,6 and flavones7 that are biologically active. Also, the genus Tephrosia is reported to elaborate other bioactive flavonoids such as flavanones, isoflavones, and pterocarpans.8,9 Some of these flavonoids also exhibit anti-inflammatory properties. For instance, genistein found in Tephrosia toxicaria10 reduces peripheral and central nuclear factor-κB (NF-κB) and the nitric oxide system as well as pro-inflammatory cytokine overactivation,11 while naringenin, common in the family Fabaceae, decreases the production of TNF-α,12 and apigenin inhibits TNF-α-induced NF-κB.13 In Kenya, the juice of boiled leaves of Tephrosialinearis (Willd.) Pers. is used traditionally to treat a broad spectrum of ailments in infants.14 An earlier phytochemical investigation on the roots of this plant led to the isolation of rotenone, deguelin, tephrosin, and 12a-hydroxyrotenone.15 In our continued effort © XXXX American Chemical Society and American Society of Pharmacognosy RESULTS AND DISCUSSION The crude extract of the aerial parts of T. linearis was subjected to silica gel and Sephadex LH-20 column chromatography, ■ Received: September 24, 2019 A https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp Article Table 1. 1H NMR (600 MHz) [δ, mult. (J in Hz)] Spectroscopic Data for Compounds 1−7 in Acetone-d6 1 2 3 5 6 8 2′/6′ 3′/5′ 4′ 5′ 6′ 1″ 2″ 3″ 4″ 5″ 1‴ 2‴ 4‴ 5‴ OMe-5 OH-5 OMe-6 OMe-7 OMe-3′ OH-4′ Me-2″ CH2-2″ Me2-3′″ OMe-3‴ C(O)CH3 6.69, s 2 6.67, s 3 6.89, s 4 5 6.70, s 6.66, s 6 7 5.68, dd (12.7, 3.1) 2.90, dd (17.1, 3.2, Heq) 3.22, dd (17.1, 12.7, Hax) 5.47, dd (13.2, 2.5) 2.73, m (Heq) 3.22, dd (17.2, 13.0, Hax) 7.46, s 5.91, s 8.05, m 7.60, m 7.61, m 8.06, m 7.56, m 7.57, m 8.11, m 7.65, m 7.64, m 7.11, s 7.62, d (2.1) 8.05, m 7.61, m 7.60, m 7.60, m 7.47, m 7.42, m 6.15, s 7.41, d (8.5) 6.91, d (8.5) 7.00, d (8.3) 7.58, dd (8.3, 2.1) 2.77, m 2.86, m 4.29, m 5.93, d (10.1) 6.77, d (10.1) 5.87, d (10.1) 6.75, d (10.1) 5.84, d (10.0) 6.72, d (10.0) 5.94, d (10.1) 6.77, d (10.1) 6.86, d (16.7) 6.63, d (16.7) 6.13, d (12.1) 6.09, d (12.1) 6.93, d (16.5) 7.45, d (16.5) 5.13, br s 5.14, br s 2.07, s 6.99, d (16.5) 7.49, d (16.5) 5.17, br s 3.87, s 3.86, s 5.58, d (10.2) 6.68, d (10.2) 4.69, br s 4.63, br s 1.79, s 2.07 3.88, s 13.55, s 12.24, s 12.35, s 3.97, s 3.88, s 4.00, s 8.38, s 1.53, s 1.49, s 1.42, s 3.26, s 1.21, s 1.56, s 1.55, s 8.57, s 1.45, s 4.09, d (11.7) 4.25, d (11.7) 1.97, s followed by preparative reversed-phase HPLC, to give five new flavones (1−5) and two new flavanones (6 and 7) in addition to 11 known flavanones (8 and 9) and flavones (10−18). By comparison of their NMR data with those described in the literature, the known compounds were identified as 7-O-methyl6-prenylnaringenin (8),16 erylivingstone I (9),17 6-C-prenylapigenin (10),18 5,7,4′,2″-tetrahydroxy-6-[3-methyl-3-butenyl]flavone (11),19 apigenin (12),20 luteolin (13),21 velutin (14),22 atalantoflavone (15),23 geraldone (16),24 patuletin 3O-rhamnoside (17),25 and eupatolitin 3-O-rhamnoside (18).26 Compound 1 was isolated as a pale yellow residue. Its molecular formula was established as C27H28O5 from the HRESIMS molecular ion [M + H]+ at m/z 433.2011 (calcd 433.2010) and sodiated molecular ion [M + Na]+ at m/z 455.1828 (calcd 455.1829), together with NMR data (Tables 1 and 2 and Figures S1−S9, Supporting Information). The presence of a flavone moiety was evident from the UV (λmax 252, 300, and 336 nm), 1H NMR [δH 6.69, s, H-3), and 13C NMR (δC 161.4 (C-2), 109.0 (C-3), and 176.6 (C-4)] spectra27 (HSQC correlation, Figure S7, Supporting Information). The NMR data showed the presence of a methoxy substituent (δH 3.87, δC 62.9), a 2″,2″-dimethylpyran ring [(cis-olefinic protons at δH 5.93, 6.77 (d, J = 10.1 Hz) and the methyl protons at δH 1.42 (s, 6H)] and a trans-olefinic 3-methoxy-3-methylbut-3-enyl substituent [olefinic protons at δH 6.86, 6.63 (d, J = 16.7 Hz)7 and a methoxy group at δH 3.26]. From the MS data (Figure S3, Supporting Information), compound 1 showed neutral losses of 32 Da (CH3OH) and 64 Da (2 × CH3OH) in the positive-ion mode, which confirmed the presence of two methoxy groups in the compound. The observation of fragments associated with neutral losses of 42 Da (C3H6), 54 Da (C4H6), and 66 Da (C5H6) when MS3 was performed on the [M + H − 64]+ ion confirmed the presence of prenyl and 2,2-dimethylpyran moieties.28 The compound also exhibited typical patterns for an unsubstituted flavone ring B sets of protons resonating at δH B https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp Article Table 2. 13C NMR (150 MHz) Spectroscopic Data for Compounds 1−7 in Acetone-d6 2 3 4 5 6 7 8 9 10 1′ 2′/6′ 3′/5′ 4′ 5′ 6′ 1″ 2″ 3″ 4″ 5″ 1‴ 2‴ 3‴ 4‴ 5‴ OMe-5 OMe-6 OMe-7 OMe-3′ CH2−2″ Me-2″ Me-3‴ OMe-3‴ C(O)CH3 C(O)CH3 1 2 3 4 5 6 7 161.4 109.0 176.6 154.7 113.7 156.1 111.4 156.4 113.3 132.8 127.0 130.0 132.2 161.2 100.4 176.9 154.6 113.0 155.0 113.2 156.3 112.9 132.6 127.0 129.8 132.1 164.1 106.4 183.8 156.3 106.2 158.0 106.3 155.1 106.1 132.4 127.4 130.1 132.9 161.5 109.0 176.7 154.7 113.7 156.3 111.5 156.5 113.4 132.7 127.0 130.0 132.3 163.4 105.6 177.0 105.3 147.3 153.1 104.1 153.3 117.7 124.4 110.2 148.8 150.8 116.3 120.8 80.2 43.4 197.4 164.8 97.6 162.4 102.3 157.9 103.8 139.8 127.2 129.5 129.6 79.3 42.7 196.9 160.9 106.8 166.1 90.9 162.1 102.5 129.9 128.1 115.2 157.9 78.8 131.7 116.9 78.7 131.6 116.9 79.5 129.3 115.9 79.0 131.7 116.8 80.0 123.1 118.5 28.6 74.5 148.4 109.3 16.6 118.1 142.1 76.0 116.0 144.6 71.8 118.2 136.0 143.9 117.3 18.3 62.9 62.8 118.5 136.8 143.9 118.0 18.3 62.9 56.5 55.5 56.5 28.4 26.4 50.6 28.7 30.2 28.5 28.4 68.9 23.9 170.6 20.6 Supporting Information). Compound 2 showed close similarities with compound 1 in its 1D- and 2D-NMR data. Like compound 1, its ring B is unsubstituted and there are no ring A protons. Also, present were a methoxy group and a 2″,2″dimethylchromene ring, which were considered to be the substituents. The only difference between the two compounds is in the nature of the prenyl substituent at C-8. Compound 2 has a hydroxy group instead of a methoxy group at C-3‴ and a cisolefinic double bond (J = 12.1 Hz) rather than trans.30,31 The NOESY correlation between H-1‴ and H-2‴ confirmed the cisconfiguration (Figure S19, Supporting Information). Consistent with this, its MS showed neutral losses of 18 Da (H2O) and 32 Da (CH3OH) confirming the presence of the hydroxy group and one methoxy group, instead of two methoxy groups as in compound 1. Their HMBC correlations (Figure S17, Supporting Information) were also similar. Therefore, compound 2 was characterized as 5-methoxy-2″,2″-dimethylpyrano[5,6:6,7]-8(Z-3-hydroxy-3-methylbut-1-enyl)flavone. Compound 3 (lineaflavone C) was assigned the molecular formula, C25H22O4, based on its HRESIMS molecular ion [M + H]+ at m/z 387.1591 (calcd 387.1591) and the [M + Na]+ peak at m/z 409.1365 (calcd 409.1410) along with the NMR data (Tables 1 and 2 and Figures S20−S28, Supporting Information). It was found to be a flavone derivative based on its UV (λmax 230 8.05 (H-2′/6′), 7.60 (H-3′/5′), and 7.61 (H-4′) with a fully substituted ring A. The absence of substitution on the ring B was further evident from the retro-Diels−Alder MS fragmentations of the ring C,29 yielding fragment ions at m/z 249 indicating a loss of C8H6 and also m/z 233 for loss of C9H6O2 (Figure S3, Supporting Information). This implied that the three substituents are in ring A. The HMBC correlations of H-4″ (δH 6.77) with C-5 (δC 154.7) and C-7 (δC 156.1) and H-3″ (δH 5.93) with C-6 (δC 113.7) allowed the placement of the 2″,2″dimethylpyran ring at C-6/7 (Figure S8, Supporting Information). HMBC correlations of H-1‴ (δH 6.86) with C-7 (δC 156.1) and C-9 (δC 156.4) supported the location of the transolefinic 3-methoxy-3-methylbut-3-enyl substituent at C-8 (δC 111.4). In turn, the HMBC correlation of the methoxy group protons with C-5 (δC 154.7) was consistent with its placement at C-5. Therefore, compound 1 was characterized as 5-methoxy2″,2″-dimethylpyrano[5″,6″:6,7]-8-(E-3-methoxy-3-methylbut-1-enyl)flavone and was given the trivial name lineaflavone A. Compound 2 (lineaflavone B), UV (λmax 238, 286, and 328 nm), was isolated as a pale-yellow residue with a molecular formula of C26H26O5 as established from the HRESIMS molecular ion [M + H]+ at m/z 419.1853 (calcd 419.1853) and the [M + Na]+ ion at m/z 441.1673 (calcd 441.1672), along with the NMR data (Table 1 and 2 and Figures S10−S19, C https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp and 280 nm) and NMR data. The NMR data of this compound were closely comparable to those of compound 1. The only differences were in the presence of a hydrogen-bonded hydroxy group at δH 13.55 attached to C-532 and the nature of prenyl substituent at C-8, which, instead of a methoxy group at C-3‴, had olefinic methylene (δH 5.13, 5.14 (br s), δC 117.3). The MS2 of the compound in the positive-ion mode was dominated by neutral losses of 42 Da (C3H6), 54 Da (C4H6) and 70 Da (C3H6 + CO), which confirmed the presence of prenyl and dimethylpyran substituents (Figure S22, Supporting Information). Thus, compound 3 was characterized as 5-hydroxy-2″,2″dimethylpyrano[5,6:6,7]-8-(E-3-methylbuta-1,3-dienyl)flavone. Compound 4 (lineaflavone D) was isolated as a pale-yellow residue with a molecular ion [M + H]+ at m/z 401.1747 (calcd 401.1747) and a [M + Na]+ peak at m/z 423.1567 (calcd 423.1567), consistent with a molecular formula of C26H24O4. The UV profile (λmax at 232, 280, and 342 nm) and NMR (Tables 1 and 2 and Figures S29−S38, Supporting Information) data suggested that the compound is a flavone derivative. The 1 H and 13C NMR spectroscopic patterns for compound 4 were closely related to those of compound 3, with the only differences being the presence of a methoxy group (δH 3.88, δC 62.9) at C-5 instead of the hydroxy group. The MS2 of the compound in the positive-ion mode was dominated by a neutral loss of 32 Da (CH3OH) confirming the presence of a methoxy group. When MS3 was performed on the [M + H − 32]+ ion (at m/z 369.1485), neutral losses of 42 Da (C3H6), 54 Da (C4H6), and 66 Da (C5H6) were observed consistent with the presence of prenyl and dimethylpyran moieties28 (Figure S31, Supporting Information). Therefore, compound 4 was characterized as 5methoxy-2″,2″-dimethylpyrano[5,6:6,7]-8-(E-3-methylbuta1,3-dienyl)flavone. The molecular formula of compound 5 was established as C17H14O6, based on the HRESIMS molecular ion [M + H]+ at m/z 315.0863 (calcd 315.0863) and the [M + Na]+ peak at m/z 337.0684 (calcd 337.0683), together with its NMR data (Tables 1 and 2 and Figures S39−S47, Supporting Information). The UV profile (λmax 218, 224, and 342 nm), 1H NMR [δH 6.66 (s, H-3)] and 13C NMR [δC 163.4 (C-2), 105.6 (C-3), and 177.0 (C-4)] data were all typical of a flavone.27 The NMR data showed the presence of two methoxy groups (δH 3.97, δC 56.5 and δH 4.00, δC 56.5). The MS3 spectrum (Figure S41, Supporting Information) of the compound was dominated by losses of 15 Da (CH3) and 28 Da (CO), supportive of the presence of methoxy groups. In the aromatic region, the 1H NMR showed two singlets at δH 7.46 and 7.11 and an ABX spin system [δH 7.00 (d, J = 8.3 Hz), 7.58 (dd, J = 8.3, 2.1 Hz and 7.62 (d, J = 2.1 Hz)]. HMBC correlations of the singlet proton at δH 7.46 with C-7 (δC 153.1), C-9 (δC 153.3), and C-4 (δC 177.0) allowed its placement at C-5, while its NOESY correlation with the methoxy protons (δH 3.97) permitted the location of the methoxy group at C-6 (Figure S47, Supporting Information). HMBC correlation of the singlet at δH 7.11 with C-7 (δC 153.1), C-9 (153.3), C-6 (147.3), and C-10 (117.7) allowed its placement at C-8, consistent with a disubstituted ring A (Figure S46, Supporting Information). HMBC correlation of H-6′ with C-2 (δC 163.4), C-4′ (δC 150.8), and C-2′ (δC 110.2) and also the hydroxy group proton (δH 8.38, OH-4′) with C-5′ (δC 116.3), C-4′ (δC 150.8), and C-3′ (δC 148.8) allowed assignment of the methoxy group (δH 4.00, δC 56.5) at C-3′. This was corroborated by the NOESY correlation of H-2′ (δH 7.62) with the methoxy group protons (δH 4.00, OMe-3′). Thus, compound 5 was characterized as 7,4′-dihydroxy-6,3′-dimethoxyflavone and assigned the trivial name, 6-methoxygeraldone, by comparison with geraldone (16).24 Compound 6 was isolated as an off-white residue with [α]25 D −0.15 (c 0.075, MeOH). The molecular formula was established as C22H20O6, based on the HRESIMS of the compound that gave a molecular ion [M + H]+ at m/z 381.1333 (calcd 381.1333) and the [M + Na]+ peak at m/z 403.1161 (calcd 403.1152), together with its NMR data (Tables 1 and 2 and Figures S48−S56, Supporting Information). The presence of a 5-hydroxyflavanone skeleton was evident from the UV profile (λmax 224, 272, 296, and 358 nm) and the NMR data [AMX spin system of δH 5.68 (dd, J = 12.7, 3.1 Hz, H-2), 2.90 (dd, J = 17.1, 3.2 Hz, H-3eq), 3.22 (dd, J = 17.1, 12.7 Hz, H-3ax) and 12.24 (OH-5); δC 80.2 (C-2), 43.4 (C-3), and 197.4 (C-4)].27 The presence of a modified 2″,2″-dimethylpyran ring was evident in the NMR spectra, which showed signals for cis-olefinic protons at δH 6.68, 5.58 (d, J = 10.2 Hz, 1H), a methyl group δH 1.45 (s, 3H) and oxymethylene protons δH 4.09, 4.25 (d, J = 11.7 Hz, 2H). The NMR spectra further showed the presence of an acetyl substituent [δH 1.97 (s, 3H), δC 20.6, 170.6 (CO)]. Loss of 18 Da (H2O), 28 Da, (CO), 60 Da (CH3COOH), 70 Da (C3H6, CO), and 104 Da (C8H8) were observed when MS3 was performed on the [M + H − C2H2O]+ ion, confirming the presence of an acetyl group and a pyran ring.28 HMBC correlation between the oxymethylene protons (δH 4.09, 4.25) and the acetyl carbonyl carbon (δC 170.6) allowed the placement of the acetyl group on the oxymethylene carbon. In the aromatic region, the 1H NMR spectrum showed signals for an unsubstituted ring B (δH 7.60 (H-2′/6′), 7.47 (H-3′/5′) and 7.42 (H-4′)) and a singlet at δH 5.91. HMBC correlation between the hydroxy group (δH 12.24, OH-5) and δC 97.6 (C-6, δH 5.91) allowed the placement of the modified 2″,2″dimethylpyran ring at C-7/8. The ECD spectrum of 6 (Figure S51, Supporting Information) showed positive and negative Cotton effects at 317 and 295 nm, respectively, which were consistent with a (2S)-configuration.31 Therefore, compound 6 was characterized as (2S)-5-hydroxy-2″-methyl-2″acetoxymethylpyrano[5,6:7,8]flavanone and given the trivial name 8″-acetylobovatin, by comparison with obovatin.33 Compound 7 was isolated as an off-white residue with a specific optical rotation of [α]25 D −0.01 (c 0.075, MeOH). Its molecular formula was deduced as C21H22O6, from the HRESIMS molecular ion [M + H]+ at m/z 371.1488 (calcd 371.1489) and the [M + Na]+ peak at m/z 393.1309 (calcd 393.1309) together with its NMR data (Tables 1 and 2 and Figures S61−S65, Supporting Information). The presence of a 5-hydroxyflavanone skeleton was evident from the UV profile (λmax 224, 230, 292, and 336 nm) and the NMR spectroscopic data [AMX spin system of δH 5.47 (dd, J = 13.2, 2.5 Hz, H-2), 2.73(m, H-3eq) and 3.22 (dd, J = 17.2, 13.0 Hz, H-3ax), 12.35 (OH-5); δC 79.3 (C-2), 42.7 (C-3), and 196.9 (C-4)].34 The NMR spectra showed the presence of a methoxy group (δH 3.88, δC 55.5) and a 2″-hydroxy-3″-methylbut-3″-enyl [δH 2.86, 2.77 (m) for H-1″, 4.29 (dd, J = 7.0, 3.8 Hz) for H-2″, 4.69 and 4.63 (br s) for H-4″, and 1.79 (s) for H-5″] substituent. Its MS fragments were dominated by losses of 72 Da (C4H8O) and 18 Da (H2O), confirming the presence of the 2-hydroxy-3methylbut-3-enyl substituent28 (Figure S59, Supporting Information). The 1H NMR spectrum further showed an AA′XX′ spin system at δH 7.41 and 6.91 (J = 8.5 Hz, d, 2H) assigned to a 4′-substituted ring B and a singlet at δH 6.15 for a ring A proton. HMBC correlations of the methylene protons (δH4.25, 4.09) of D Article https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp Article Figure 1. Results of controls (mean ± SD, n = 3 for the medium and n = 4 for LPS and ibuprofen). and TNF-α in LPS stimulated PBMCs. The assays are premised on the fact that when inflammation occurs, cytokines are produced and released from PBMCs as part of the immune response.35 This situation can be mimicked by incubating PBMCs with bacterial LPS, a major structural component of the outer wall of Gram-negative bacteria, and considered to be a potent initiator of inflammatory responses.36 The antiinflammatory effects were assessed using this in vitro model with the anti-inflammatory drug ibuprofen as a reference control. As shown in Figure 1, cells treated with LPS showed an increase in inflammatory cytokine release compared to untreated control cells. Except for IL-1β, the tested proinflammatory cytokine release was suppressed in the presence of ibuprofen in comparison to the LPS control. The percentages of the 2-hydroxy-3-methylbut-3-enyl substituent with C-7 (δC 166.1) and C-5 (δC 160.9) were used to place it at C-6 (δC 106.8). The ECD spectrum of 7 (Figure S60, Supporting Information) showed positive and negative Cotton effects at 335 and 291 nm, respectively, which were consistent with a 2Sconfiguration.17 However, the configuration at 2″ remains undetermined. On the basis of the above data, compound 7 was characterized as (2S)-5,4′-dihydroxy-7-methoxy-6-(2-hydroxy3-methylbut-3-enyl)flavanone and given the trivial name 5hydroxy-7-methoxysaniculamin A, by comparison with saniculamin A.34 The anti-inflammatory effects of the crude extract of the aerial parts of T. linearis, as well as the isolated compounds, were evaluated by measuring the levels of IL-1β, IL-2, IL-6, GM-CS, E https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp Article Figure 2. Results of cytokine release of PBMCs after incubation with test compounds (n = 1). C = plant crude extract. 622.32% and 238.54%, respectively, the remaining compounds suppressed the production of IL-6. However, there was no difference in IL-6 secretion compared to the LPS control for compounds 9 and 16. Therefore, this study not only complements previous reports on the genus Tephrosia5−7 but also identifies a number of flavonoids from the plant T. linearis with anti-inflammatory potential. release in comparison to the LPS control were 80.3%, 46.4%, 47.7%, and 12.2% for IL-2, IL-6, GM-CSF, and TNF-α, respectively. As shown in Figure 2 and Table 3, the crude plant extract inhibited the release of all cytokines except IL-1β that slightly increased to about 107.67% in comparison to the LPS control. All the tested compounds resulted in decreased release of IL-2, GM-CSF, and TNF-α. Compounds 1, 2, 4, 7, 10, 13−15, and 17 suppressed the release of IL-1β by levels varying from 0.67 to 75.85% as compared to the LPS control. The strongest inhibition occurred in the presence of compound 13 while 9 resulted in an increased cytokine release of about 220.43% as compared to the LPS control. Except for compounds 3 and 11, which provoked an increase in IL-6 secretion of up to ■ EXPERIMENTAL SECTION General Experimental Procedures. The optical rotations were determined in methanol using a Kruss P8000-T polarimeter while ECD measurements were run on a JASCO J-715 spectrometer. IR spectra F https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp Table 3. Percentage of Cytokine Release in Comparison to LPS Controla,b Fractions from 30% EtOAc in cyclohexane were also purified on Sephadex LH-20 followed by preparative HPLC to give 1 (24.3 mg), 2 (5.3 mg), 4 (1.9 mg), 5 (1.0 mg), 7 (1.0 mg), 8 (2.3 mg), 9 (1.5 mg), 11 (8.8 mg), 12 (1.2 mg), 15 (4.9 mg), and 16 (1.0 mg). Purification of the 100% EtOAc fractions by preparative HPLC gave 13 (3.3 mg), 14 (1.2 mg), 17 (1.0 mg), and 18 (16.5 mg). Lineaflavone A (1). Pale yellow residue; LC-UV [acetonitrile (aq) in H2O (0.1% formic acid)] λmax 252, 300, and 336 nm; IR (neat) νmax 3065, 2974, 1645, 1581 cm−1; 1H and 13C NMR, see Tables 1 and 2 and Figure S4−S9, Supporting Information; HRESIMS m/z 433.2011 [M + H]+ (calcd for C27H29O5, 433.2010) and m/z 455.1828 [M + Na]+ (calcd for C27H28O5Na, 455.1829). Lineaflavone B (2). Pale yellow residue, LC-UV [acetonitrile (aq) in H2O (0.1% formic acid)] λmax 238, 286, 328 nm; IR (neat) νmax 3392, 2930, 1711, 1630 cm−1; 1H and 13C NMR, see Tables 1 and 2 and Figures S13−19, Supporting Information; HRESIMS m/z 419.1853 [M + H]+ (calcd for C26H27O5, 419.1853) and m/z 441.1673 [M + Na]+ (calcd for C26H26O5Na, 441.1672). Lineaflavone C (3). Pale yellow residue; LC-UV [acetonitrile (aq) in H2O (0.1% formic acid)] λmax 230, 280 nm; IR (neat) νmax 3421, 2928, 1651, 1613 cm−1; 1H and 13C NMR, see Tables 1 and 2 and Figure S23−S28, Supporting Information; HRESIMS m/z 387.1591 [M + H]+ (calcd for C25H23O4, 387.1591) and m/z 409.1365 [M + Na]+ (calcd for C25H22O4Na, 409.1410). Lineaflavone D (4). Pale yellow residue, LC-UV [acetonitrile (aq) in H2O (0.1% formic acid)] λmax 232, 280, 342 nm; 1H and 13C NMR, see Tables 1 and 2 and Figures S32−S38, Supporting Information; HRESIMS m/z 401.1747 [M + H]+ (calcd for C26H25O4, 401.1747) and m/z 423.1567 [M + Na]+ (calcd for C26H24O4Na, 423.1567). 6-Methoxygeraldone (5). Yellow residue, LC-UV [acetonitrile (aq) in H2O (0.1% formic acid)] λmax 218, 224, 342 nm; IR (neat) νmax 2873, 1624, 1504 cm−1; 1H and 13C NMR, see Tables 1 and 2 and Figures S42−S47, Supporting Information; HRESIMS m/z 315.0863 [M + H]+ (calcd for C17H15O6, 315.0863) and m/z 337.0684 [M + Na]+ (calcd for C17H14O6Na, 337.0683). 8″-Acetylobovatin (6). Off-white residue; [α]25 D −0.15 (c 0.075, MeOH); ECD (c 0.1, MeOH) [θ]317 + 5.55, [θ]295 −19.14; LC-UV [acetonitrile (aq) in H2O (0.1% formic acid)] λmax 224, 272, 296, 358 nm; IR (neat) νmax 3566, 2975, 1745, 1644 cm−1; 1H and 13C NMR see Tables 1 and 2 and Figures S52−S56, Supporting Information; HRESIMS m/z 381.1333 [M + H]+ (calcd for C22H21O6, 381.1333) and m/z 403.1161 [M + Na]+ (calcd for C22H20O6Na, 403.1152). 5-Hydroxy-7-methoxysaniculamin A (7). Off-white solid; [α]25 D −0.01, (c 0.075, MeOH); ECD (c 0.1, MeOH) [θ]355 + 9.80, [θ]291 −42.62; LC-UV [acetonitrile (aq) in H2O (0.1% formic acid)] λmax 224, 230, 292, 336 nm; IR (neat) νmax 3336, 2972, 1637, 1617 cm−1; 1H and 13C NMR see Tables 1 and 2 and Figures S61−S65, Supporting Information; HRESIMS m/z 371.1488 [M + H]+ (calcd for C21H23O6, 371.1489) and m/z 393.1309 [M + Na]+ (calcd for C21H22O6Na, 393.1309). Biological Assays. The anti-inflammatory tests were conducted by Pharmacelsus, Saarbrücken, Germany. The levels of inflammatory cytokines were assessed using peripheral blood mononuclear cells from Immunospot (ePBMC-Uncharacterized Cryopreserved Human PBMC) purchased from Cellular Technologies Limited (Shaker Heights, OH, USA; Figure S66, Supporting Information) (http:// www.immunospot.com/CatalogueRetrieve.aspx?ProductID= 10537096&A=SearchResult&SearchID=10324581&ObjectID= 10537096&ObjectType=27). PMBCs were isolated from blood obtained from three healthy donors with the ethnicity AfricanAmerican (male, 38 years old), Pacific Islander (male, 38 years old), and Hispanic-Latino (female, age 34). The pure compounds were dissolved in dimethyl sulfoxide (DMSO) to obtain 20 mM stock solutions while the crude plant extract was constituted as a 20 mg/mL stock solution. Ibuprofen was prepared as a stock solution of 20 mM in DMSO. LPS was dissolved in cell culture medium with a concentration of 1 mg/mL. Pure compounds were tested at a concentration of 100 μM while the crude extract was used at a concentration of 100 μg/mL. The final DMSO concentration was 0.5% in all the samples. The positive control, ibuprofen, was also used at 100 μM and all samples were cytokine release [percentage of LPS control] compound IL-1β IL-2 IL-6 GM-CSF TNF-α ibuprofen 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 crude extract 199.29 68.27 1.18 169.84 75.85 170.51 157.09 53.77 102.34 220.43 40.83 103.28 105.17 0.67 0.80 48.35 109.22 4.81 148.44 107.67 80.33 57.60 57.60 17.72 57.60 17.72 17.72 17.72 57.64 57.60 17.72 57.64 57.60 57.60 57.60 57.64 57.64 57.60 17.72 17.72 46.46 10.89 0.00 622.32 97.43 43.11 43.39 25.79 19.57 100.00 24.72 238.54 48.33 0.04 0.09 7.23 100.00 0.64 24.94 13.91 47.76 17.39 8.63 78.39 83.73 17.39 40.97 32.47 46.89 35.76 48.79 38.92 32.47 8.63 8.63 4.87 92.58 8.63 48.79 36.83 12.23 0.92 0.17 1.89 3.15 4.03 2.66 4.03 6.38 12.55 5.01 5.20 4.22 0.17 0.92 1.12 3.64 0.54 3.83 4.61 a Test compounds were evaluated at 100 μm. bThe plant crude extract was evaluated at 100 μg/mL. were recorded on a Bruker Tensor-27 FT-IR spectrometer (Cricket, Harrick Scientific, USA). NMR spectra were recorded on a Bruker Advance III spectrometer using standard pulse sequences and referenced with the residual solvent signal. High-resolution electrospray ionization mass spectrometry (HRESIMS) was done on an LTQ orbitrap spectrometer (Thermo Scientific, USA) with a HESI-II source. The spectrometer was equipped with an Agilent 1200 HPLC system (Santa Clara, CA, USA) with pump, PDA detector and autosampler. All MSn experiments were performed with collision-induced dissociation with 35 arbitrary units. Column chromatography was performed using Merck silica gel 60 (70−230 mesh) and Sephadex LH-20. TLC was carried out on precoated silica gel 60 PF254+365 plates. Preparative HPLC was performed on a Shimadzu LC-20AP system equipped with DGU 20A5R degassing unit, an SPD-M20A detector, SIL-20ACHT autosamplers, and a Nucleodur Polartec 5 μm RP column (10 × 125 mm) using LabSolution software system. The separation was achieved using a gradient MeOH-H2O (0.1% formic acid) solvent system. Plant Material. Tephrosia linearis was collected from Gongoni (S04°23′57.3″, E039°27′17.4″; El. 39 m) in Kwale County, Kenya, in May 2018. The plant was authenticated by Mr. Patrick Mutiso of the University Herbarium, School of Biological Science, the University of Nairobi, where a voucher specimen, ORO-2018/11, was deposited. Extraction and Isolation. The air-dried and ground aerial parts of T.linearis (890 g) were extracted three times using dichloromethane− methanol (1:1, 3 L) at room temperature. The extract was concentrated under reduced pressure on a rotary evaporator to provide a dark brown paste (72.5 g). This concentrated extract was partitioned between water and n-hexane to remove lipid materials. The aqueous layer was further partitioned in ethyl acetate (EtOAc). The ethyl acetate portion was concentrated to provide a brown paste (18.4 g) and then subjected to column chromatography on silica gel (250 g), eluting with cyclohexane−ethyl acetate mixtures of increasing polarity. A fraction that eluted with 5% EtOAc in cyclohexane was purified on Sephadex LH-20 using dichloromethane−methanol (1:1) followed by preparative HPLC (20:80, MeOH/H2O-100% MeOH gradient elution for 50 min at a flow rate of 4 mL/min) to give 3 (1.1 mg). Similarly, the 10% EtOAc in cyclohexane fraction was purified using Sephadex LH-20 followed by preparative HPLC to give 6 (1.0 mg). The 20% EtOAc in cyclohexane fraction was purified in a similar way to give 10 (18.2 mg). G Article https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp coincubated with 10 μg/mL LPS. PBMCs are the main source of cytokines within the circulating blood. Due to the small amounts of cytokines released by PBMCs into the supernatant, a bead-based assay (ProcartaPlex, Luminex, ThermoFisher Scientific, Vienna, Austria; Figure S66−S68, Supporting Information) was used to quantify the five cytokines in parallel within a 50 μL sample using appropriate calibration standards. Human cryopreserved PBMCs were thawed according to the manufacturer’s instructions. Three vials of the cells from different donors were pooled. Cells were washed, resuspended in RPMI 1640 containing 10% FBS, plated in 96-well round-bottomed plates at 100 000 PBMCs/well and exposed to the test items at the concentrations stated above. Therefore, subsequent dilutions of stock solutions of the test items were prepared in a 96-well plate and transferred to the PBMCs-containing wells. The cells were incubated for 24 h at 37 °C and 5% CO2. The plates were then centrifuged for 3 min at 350 gyrations, and the cell-free supernatant was collected and forwarded to cytokine bead-array assay. The cytokine bead-array assay was conducted according to the manufacturer’s instructions and read in a MagPix reader. For the dose−response relationship, absolute concentrations were calculated by the MagPix software using two separate calibration series as provided by the manufacturer. For the negative control, cells were incubated only with cell culture medium. As a positive control for inflammation, cells were incubated with 10 μg/mL LPS while as a positive control for anti-inflammation, cells were coincubated with 10 μg/mL LPS and 100 μM ibuprofen. https://pubs.acs.org/10.1021/acs.jnatprod.9b00922 ■ Notes The authors declare no competing financial interest. ACKNOWLEDGMENTS R.O.O. is grateful to the German Academic Exchange Services (DAAD) for a doctoral scholarship which was offered through the Natural Products Research Network for Eastern and Central Africa (NAPRECA) and for supporting his research stay at the TU Dortmund. Mr. Patrick Mutiso (the Herbarium, School of Biological Sciences, University of Nairobi) is acknowledged for authentication of the plant. We are thankful to Dr. Wolf Hiller (Faculty of Chemistry and Chemical Biology, TU Dortmund) for NMR measurements, Dr. Souvik Kusar (INFU, TUDortmund) for valuable discussions on the biological assays, Mr. Michael Kubicki for acquisitions of HRESIMS, and Mrs. G. Hardes (INFU, TU Dortmund) for technical assistance. ■ ■ REFERENCES (1) Atilaw, Y.; Duffy, S.; Heydenreich, M.; Muiva-Mutisya, L.; Avery, V.; Erdélyi, M.; Yenesew, A. Molecules 2017, 22, 318. (2) Smalberger, T. M.; Vleggaar, R.; Weber, J. C. Tetrahedron 1974, 30, 3927−3931. (3) Samuel, V. J.; Mahesh, A. R.; Murugan, V. J. Appl. Pharm. Sci. 2019, 9 (3), 117−125. (4) Muiva, L. M.; Yenesew, A.; Derese, S.; Heydenreich, M.; Peter, M. G.; Akala, H. M.; Eyase, F.; Waters, N. C.; Mutai, C.; Keriko, J. M.; Walsh, D. Phytochem. Lett. 2009, 2, 99−102. (5) Muiva-Mutisya, L.; Macharia, B.; Heydenreich, M.; Koch, A.; Akala, H. M.; Derese, S.; Omosa, L. K.; Yusuf, A. O.; Kamau, E.; Yenesew, A. Phytochem. Lett. 2014, 10, 179−183. (6) Muiva-Mutisya, L. M.; Atilaw, Y.; Heydenreich, M.; Koch, A.; Akala, H. M.; Cheruiyot, A. C.; Brown, M. L.; Irungu, B.; Okalebo, F. A.; Derese, S.; Mutai, C.; Yenesew, A. Nat. Prod. Res. 2018, 32, 1407−1414. (7) Atilaw, Y.; Muiva-Mutisya, L.; Ndakala, A.; Akala, H.; Yeda, R.; Wu, Y.; Coghi, P.; Wong, V.; Erdélyi, M.; Yenesew, A. Molecules 2017, 22, 1514. (8) Touqeer, S.; Saeed, M. A.; Ajaib, M. Phytopharmacology 2013, 4, 598−637. (9) Chen, Y.; Yan, T.; Gao, C.; Cao, W.; Huang, R. Molecules 2014, 19, 1432−1458. (10) Jang, D. S.; Park, E. J.; Kang, Y.-H.; Hawthorne, M. E.; Vigo, J. S.; Graham, J. G.; Cabieses, F.; Fong, H. H. S.; Mehta, R. G.; Pezzuto, J. M.; Kinghorn, A. D. J. Nat. Prod. 2003, 66, 1166−1170. (11) Valsecchi, A. E.; Franchi, S.; Panerai, A. E.; Sacerdote, P.; Trovato, A. E.; Colleoni, M. J. Neurochem. 2008, 107, 230−240. (12) Tsai, S.-J.; Huang, C.-S.; Mong, M.-C.; Kam, W.-Y.; Huang, H.Y.; Yin, M.-C. J. Agric. Food Chem. 2012, 60, 514−521. (13) Funakoshi-Tago, M.; Nakamura, K.; Tago, K.; Mashino, T.; Kasahara, T. Int. Immunopharmacol. 2011, 11, 1150−1159. (14) Kokwaro, J. O. Medicinal Plant of East Africa; East African Literature Bureau: Nairobi, 1976. (15) Were, O.; Munavu, R. M.; Lwande, W.; Nyandat, E. Fitoterapia 1990, 61, 372. (16) Intekhab, J.; Aslam, M. J. Saudi Chem. Soc. 2009, 13, 295−298. (17) Bedane, K. G.; Kusari, S.; Masesane, I. B.; Spiteller, M.; Majinda, R. R. T. Fitoterapia 2016, 108, 48−54. (18) Chang, C.-H.; Lin, C.-C.; Kadota, S.; Hattori, M.; Namba, T. Phytochemistry 1995, 40, 945−947. (19) Lee, S.-J.; Wood, A. R.; Maier, C. G. A.; Dixon, R. A.; Mabry, T. J. Phytochemistry 1998, 49, 2573−2577. (20) Parmar, V. S.; Bisht, K. S.; Sharma, S. K.; Jain, R.; Taneja, P.; Singh, S.; Simonsen, O.; Boll, P. M. Phytochemistry 1994, 36, 507−511. (21) Lin, L.-C.; Pai, Y.-F.; Tsai, T.-H. J. Agric. Food Chem. 2015, 63, 7700−7706. ASSOCIATED CONTENT * Supporting Information sı The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jnatprod.9b00922. UV data for compounds 1−7, ECD data for compounds 6 and 7, NMR spectra for compounds 1−7, HRESIMS spectra for 1−7, and certificates of analysis (PDF) ■ Article AUTHOR INFORMATION Corresponding Authors Michael Spiteller − Institute of Environmental Research (INFU), Department of Chemistry and Chemical Biology, Chair of Environmental Chemistry and Analytical Chemistry, TU Dortmund, 44221 Dortmund, Germany; Email: m.spiteller@ infu.tu-dortmund.de Albert Ndakala − Department of Chemistry, University of Nairobi, Nairobi, Kenya; Email: andakala@uonbi.ac.ke Authors Richard Oriko Owor − Department of Chemistry, University of Nairobi, Nairobi, Kenya; Institute of Environmental Research (INFU), Department of Chemistry and Chemical Biology, Chair of Environmental Chemistry and Analytical Chemistry, TU Dortmund, 44221 Dortmund, Germany; Department of Chemistry, Busitema University, Tororo, Uganda; orcid.org/ 0000-0002-0211-8833 Kibrom Gebreheiwot Bedane − Institute of Environmental Research (INFU), Department of Chemistry and Chemical Biology, Chair of Environmental Chemistry and Analytical Chemistry, TU Dortmund, 44221 Dortmund, Germany Sebastian Zühlke − Institute of Environmental Research (INFU), Department of Chemistry and Chemical Biology, Chair of Environmental Chemistry and Analytical Chemistry, TU Dortmund, 44221 Dortmund, Germany Solomon Derese − Department of Chemistry, University of Nairobi, Nairobi, Kenya; orcid.org/0000-0002-2640-3583 George Otieno Ong’amo − School of Biological Sciences, University of Nairobi, Nairobi, Kenya Complete contact information is available at: H https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX Journal of Natural Products pubs.acs.org/jnp Article (22) Das, K. C.; Farmer, W. J.; Weinstein, B. J. Org. Chem. 1970, 35, 3989−3990. (23) Banerji, A.; Luthria, D. L.; Prabhu, B. R. Phytochemistry 1988, 27, 3637−3640. (24) Lopes, J. L. C.; Lopes, J. N. C.; Leitao Filho, H. F. Phytochemistry 1979, 18, 362. (25) Chiung-Sheue Liu, K.; Shi-Lin, Y.; Roberts, M. F.; Phillipson, J.D. Phytochemistry 1989, 28, 2813−2818. (26) Quijano, L.; Malanco, F.; Ríos, T. Tetrahedron 1970, 26, 2851− 2859. (27) Agrawal, P. K. Carbon-13 NMR of Flavonoids; Elsevier: Amsterdam, The Netherlands, 1989; Vol. 39. (28) Xu, M.-J.; Wu, B.; Ding, T.; Chu, J.-H.; Li, C.-Y.; Zhang, J.; Wu, T.; Wu, J.; Liu, S.-J.; Liu, S.-L.; Ju, W.-Z.; Li, P. Rapid Commun. Mass Spectrom. 2012, 26, 2343−2358. (29) Ma, Y. L.; Li, Q. M.; Van den Heuvel, H.; Claeys, M. Rapid Commun. Mass Spectrom. 1997, 11, 1357−1364. (30) Hegazy, M.-E. F.; Abd El-Razek, M. H.; Nagashima, F.; Asakawa, Y.; Paré, P. W. Phytochemistry 2009, 70, 1474−1477. (31) Stevenson, P. C.; Kite, G. C.; Lewis, G. P.; Forest, F.; Nyirenda, S. P.; Belmain, S. R.; Sileshi, G. W.; Veitch, N. C. Phytochemistry 2012, 78, 135−146. (32) Venkataratnam, G.; Rao, E. V.; Vilain, C. Phytochemistry 1986, 25, 1507−1508. (33) Andrei, C. C.; Ferreira, D. T.; Faccione, M.; de Moraes, L. A. B.; de Carvalho, M. G.; Braz-Filho, R. Phytochemistry 2000, 55, 799−804. (34) Xu, G.; Wang, Z.; Zhao, B.; Liu, N.; Yang, S.; Liu, Y.; Wang, J.; Zhou, X. Phytochem. Lett. 2016, 18, 35−38. (35) Davis, J. M.; Knutson, K. L.; Strausbauch, M. A.; Crowson, C. S.; Therneau, T. M.; Wettstein, P. J.; Matteson, E. L.; Gabriel, S. E. J. Immunol. 2010, 184, 7297−7304. (36) O’Bryan, M. K.; Schlatt, S.; Phillips, D. J.; de Kretser, D. M.; Hedger, M. P. Endocrinology 2000, 141, 238−246. I https://dx.doi.org/10.1021/acs.jnatprod.9b00922 J. Nat. Prod. XXXX, XXX, XXX−XXX