Trimeric and Tetrameric A-Type Procyanidins from Peanut Skins

Article pubs.acs.org/jnp

Trimeric and Tetrameric A‑Type Procyanidins from Peanut Skins Marta K. Dudek,*,†,‡ Vitold B. Gliński,§ Matthew H. Davey,§ Daniel Sliva,⊥ Sławomir Kaźmierski,† and Jan A. Gliński*,§ †

Centre of Molecular and Macromolecular Studies, Polish Academy of Sciences, Sienkiewicza 112, 90-363 Lodz, Poland Physical Chemistry Department, Medical University of Warsaw, Banacha 1, 02-097 Warsaw, Poland § Planta Analytica LCC, 461 Danbury Road, New Milford, Connecticut 06776, United States ⊥ DSTest-Laboratories LLC, Purdue Research Park, 5225 Exploration Drive, Indianapolis, Indiana 46241, United States ‡

S Supporting Information *

ABSTRACT: Peanut skins are a rich source of oligomeric and polymeric procyanidins. The oligomeric fractions are dominated by dimers, trimers, and tetramers. A multistep chromatographic fractionation led to the isolation of four new A-type procyanidins of tri- and tetrameric structures. The structures of the new trimers were defined by NMR, electronic circular dichroism, and MS data as epicatechin-(4β→8,2β→ O→7)-epicatechin-(4β→8,2β→O→7)-catechin, peanut procyanidin B (3), and epicatechin-(4β→8,2β→O→7)-epicatechin-(4β→6)-catechin, peanut procyanidin C (4). The new tetramers were defined as epicatechin-(4β→8,2β→O→7)epicatechin-(4β→6)-epicatechin-(4β→8,2β→O→7)-catechin, peanut procyanidin E (1), and epicatechin-(4β→8,2β→O→7)-epicatechin-(4β→6)-epicatechin-(4β→8,2β→O→7)-epicatechin, peanut procyanidin F (2). In addition, both A-type dimers A1, epicatechin-(4β→8,2β→O→7)-catechin, and A2, epicatechin(4β→8,2β→O→7)-epicatechin, as well as two known peanut trimers, ent-epicatechin-(4β→6)-epicatechin-(4β→8,2β→O→7)catechin, peanut procyanidin A (5), and epicatechin-(4β→8)-epicatechin-(4β→8,2β→O→7)-catechin, peanut procyanidin D (6), were also isolated. Dimer A1, the four trimers, and two tetramers were evaluated for anti-inflammatory activity in an in vitro assay, in which LPS-stimulated macrophages were responding with secretion of TNF-α, a pro-inflammatory cytokine. Tetramer F (2) was the most potent, suppressing TNF-α secretion to 82% at 8.7 μM (10 μg/mL), while tetramer E (1) at the same concentrations caused a 4% suppression. The results of the TNF-α secretion inhibition indicate that small structural differences, as in peanut procyanidin tetramers E and F, can be strongly differentiated in biological systems.

P

procyanidins, especially of the A-type, are rare, reflecting mainly the poor commercial availability of these compounds. However, the available data indicate that biological systems can differentiate between structurally similar procyanidins.8−12 Thus, it appears that A-type procyanidins, as a result of their rigid 3D shape, should interact with biological macromolecules more specifically, leading to potential therapeutic applications. Among the documented pharmacological effects, procyanidins were shown to exert cardiovascular,13,14 neuroprotective,15 antidiabetic,9,16 and hypotensive17 activities. A protective effect of these polyphenols in ischemia-induced cell swelling was also observed.18 Recent advances in medical sciences increasingly indicated that many degenerative diseases are associated with increased oxidative stress levels. Thus, a higher intake of procyanidins19,20 could be beneficial by lowering excessive levels of harmful reactive oxygen species generated during oxidative stress. This postulate seems to be true as more therapeutic applications of procyanidins are explored. Besides

rocyanidins are common phenolic plant metabolites with antioxidant properties and comprise epicatechin (EC) and catechin (C) constituent units. In the more common B-type procyanidins, the flavan-3-ol units are connected through a single bond between C-4 of the upper unit and C-6 or C-8 of the lower unit.1 The B-type procyanidins dominate in dietary important sources such as apples, pears, and cocoa-containing foods. The A-type procyanidins differ from the B-types by having an additional bond between adjacent flavan-3-ol units that connects C-2 of the upper unit via an oxygen atom to C-7 of the lower unit. Such a “bridge” imposes conformational rigidity to both units. In addition, in the A-type procyanidins, the crowding around the B-type connected units often produces rotational restrictions favoring one conformation (rotamer) over another, as is evidenced by the 1H NMR spectroscopic analysis.2,3 The presence of A-type procyanidins has been well established in cranberries, cinnamon, and peanut skins.3−5 To a significant degree, the position of the A-type bond is characteristic for the plant source. For example, in cinnamon it involves the top unit,3 but in cranberries the bottom unit.6,7 Data on the biological activity of individual © 2017 American Chemical Society and American Society of Pharmacognosy

Received: October 14, 2016 Published: February 13, 2017 415

DOI: 10.1021/acs.jnatprod.6b00946 J. Nat. Prod. 2017, 80, 415−426

Journal of Natural Products

Article

Figure 1. Reversed-phase HPLC profile of the peanut skin procyanidins.

the general antioxidant activity of procyanidins, it may be anticipated that other, more specific biological effects that would depend on their 3D structure may be discovered.11,21 The primary determinants of the 3D structure are the linkage location between particular units (i.e., 4→6 or 4→8), the presence of the A-type bond, and the presence of epicatechin or catechin units. Peanut skins are a byproduct of the peanut industry, which processes approximately 29 million metric tons of peanuts every year. 22 The skins are a rich source of A-type procyanidins.23 The extract was shown to exhibit hypocholesterolemic,24 anti-inflammatory,25 and antiallergic26 activities. Recently carried out LC-MS studies have documented the presence of the A-type linkage in 16 trimers and 27 tetramers.27 However, only three A-type trimers and one A-type tetramer have been purified from peanut skin extract thus far.4,10,28 This paper deals with the structure elucidation of four new naturally occurring A-type procyanidins from peanut skins. The approach to structural analysis proves that establishing the location of the interflavan linkage in addition to the C-6/C-8 chemical shift criterion benefits from the ROESY correlations of hydroxy protons. Thus, the necessity of developing an unambiguous and reliable analytical protocol for accurately defining the chemical structure of the A-type procyanidins is

underlined. The isolated procyanidins were shown to suppress the secretion of a pro-inflammatory cytokine, TNF-α, from stimulated macrophages.

■

RESULTS AND DISCUSSION A reversed-phase HPLC chromatogram of the 70% aqueous acetone extract of peanut skins, showing individual procyanidins, is shown in Figure 1. Compound 1 (Figure 2) was isolated as an off-white solid with a molecular formula of C60H46O24, which was established on the basis of 13C NMR data and the presence of deprotonated and protonated molecular ions at m/z 1149.5 [M − H]− and 1151.5 [M + H]+, respectively. The HRMS data gave an [M + H]+ ion at m/z 1151.2463, which is in agreement with the calculated mass for C60H47O24 of 1151.2457, being consistent with the molecular formula. These data, together with the 1H NMR spectrum, indicated that 1 is a tetrameric flavan-3-ol. The 1H NMR signals assignable to H-3 and H-4 in units I−III (C-, F-, and I-ring, respectively) appear as singlets or doublets with small J-couplings (ca. 3.5 Hz), whereas H-2 and H-3 in the terminal (IV) unit (L-ring) resonate as doublets with J-couplings of 9.0 and 5.9 Hz, respectively, and H-4α and H-4β (unit IV, L-ring) as doublet of doublets with J-values of 16.5 and 9.0 Hz and 16.5 and 5.9 Hz, respectively. The molecular 416



Article

Figure 2. Molecular structures of 1−6.

criterion there are C-4→C-8 linkages between units I and II and III and IV [C-8 (unit II, D-ring) and C-8 (unit IV, J-ring) δC 107.17 and 106.84, respectively] and a C-4→C-6 linkage between units II and III [C-6 (unit III, G-ring) δC 111.69 ppm].29,30 Taking into account that the chemical shift criterion is not fully reliable in the differentiation between the linkage locations, the ROESY correlations were carefully analyzed. The C-4→C-8 linkage results in the proximity of H-2′ and H-6′ (unit I, B-ring) and H-6 (D-ring) of unit II. Therefore, the cross-signals between H-6 (unit II, D-ring) and H-2′/H-6′ (unit I, B-ring), as well as between H-6 (unit IV, J-ring) and H-

mass, as well as the presence of two ketal carbons (C-2) with characteristic chemical shifts of 100.52 and 100.59 ppm in the 13 C NMR spectrum, indicates that there are two A-type and one B-type bond between the flavan-3-ol units. The HMBC cross-peaks between H-4 (unit II, F-ring) and C-6/8 and C5/7 (unit III, G-ring), as well as COSY correlation between H-2 (unit II, F-ring) and H-4 (unit II, F-ring), locates the B-type bond between units II and III. The locations of the interflavan linkages were established on the basis of ROESY correlations, as well as the chemical shifts of the respective C-6 or C-8 signals. On the basis of the latter 417



Article

Compound 3 has a molecular formula of C45H34O18. The ESIMS deprotonated and protonated molecular ions at m/z 861.4 and 863.4, respectively, indicate a procyanidin trimer with two A-type interflavanyl bonds. The HRMS data show an [M + H]+ ion at m/z 863.1834, which agrees with the calculated mass of m/z 863.1823 for a molecular formula of C45H35O18. This formula is further supported by two ketal carbon (C-2) resonances at δC 100.15 and 100.92 in the 13C NMR spectrum. Signals originating from units I and II have chemical shifts characteristic for an epicatechin moiety, which is also supported by the small J-couplings of the H-3 and H-4 resonances. Signals at δC 84.9 and δH 4.65 (d, J = 8.3 Hz), assignable to the lower unit of the trimer, indicate that this is a catechin unit. The same is also evident from the characteristic J-couplings and chemical shifts of the H-4 (unit III, I-ring) protons (doublet of doublets with J-couplings of 16.5 and 5.8, and 16.5 and 8.8 Hz, respectively; see Table 2). The interflavanyl linkage between units I and II is located between C-4 (unit I, C-ring) and C-8 (unit II, D-ring), based upon the ROESY correlations between H-6 (unit II, D-ring) and H-2′/H-6′ (unit I, B-ring), and H-4 (unit I, C-ring) and H2′/H-6′ (unit II, E-ring), observable only for the C-4−C-8 linkage (Figure 4). The C-8 (unit II, D-ring) chemical shift (δC 108.95) might indicate that there is a C-4−C-6 linkage, as was earlier suggested,29 but the ROESY correlations exclude this possibility. Additionally, the chemical shifts of C-8 (unit II, Dring) in pavetannins B7 and B8,33 which also possesses C-4→ C-8 linkages, are δC 108.9 and 107.8, respectively. Clearly, in the case of 3 the “chemical shift criterion” does not suffice to establish the linkage location. Defining the second linkage location was more complicated. The chemical shift of the C-6 or C-8 atom engaged in the linkage is δC 107.79, favoring rather the C-4−C-8 connectivity. However, because of the notorious ambiguity of this criterion, additional evidence was needed. The ROESY spectra recorded in methanol-d4 are not helpful in showing cross-signals possible for both linkage locations, only slightly favoring the C-4−C-8 linkage through a weak correlation between H-4 (unit II, Fring) (d, δH 4.32) and H-2 (unit III, I-ring) (d, δH 4.65). The literature provides one more criterion for establishing linkage locations.4 It is based on the HMBC cross-peaks of H-4 with C7 and C-9 of the second unit for a C-4−C-8 linkage or with C-5 and C-7 of this unit for a C-4−C-6 linkage. However, the assignment of the C-5, C-7, and C-9 signals is based on the same correlations that are used for the determination of the linkage location. By reversing this assignment, the same correlations could be used as proof of an alternative linkage location. Therefore, in order to properly elucidate linkage location based on the HMBC cross-peaks, first the C-5 and C-9 signals have to be unambiguously assigned. For C-5 the chemical shift is usually at lower field than C-9 [e.g., for procyanidins A1 and A2, C-5 (unit II, D-ring) resonates at ca. δC 156, whereas C-9 (unit II, D-ring) resonates at ca. δC 151− 152 ppm].30 In 3, H-4 (unit II, F-ring) interacts in the HMBC spectrum with carbon signals at δC 149.56, 151.64, 153.29, and 156.43. The signals at δC 149.56 and 156.43 are assignable to C-5 (unit II, D-ring) and C-9 (unit II, D-ring) carbons, respectively, whereas the signal at δC 153.29 originates from C7 (unit III, G-ring). The remaining signal at δC 151.64 is therefore assignable to C-5 (unit III, G-ring) or C-9 (unit III, G-ring), depending on the assumed linkage location. According to the assignments for procyanidins A1/A2, the δC 151.64 resonance represents C-9 (unit III, G-ring), suggesting the C-

2′ (unit III, H-ring), confirm the C-4→C-8 linkage between units I→II and III→IV. The C-4→C-6 linkage between units II→III is supported by the ROESY correlation between H-4 (unit II, F-ring) and H-6′ (unit IV, K-ring). The chemical shifts of 1 were also compared with those published earlier for similar compounds. Units II−IV of 1 have highly similar chemical shifts compared to the EC-(4β→6)-EC(4β→8,2β→O→7)-C trimer,9 supporting previous evidence for the location of the interflavan linkages. The β-orientation of the interflavan linkages was confirmed by electronic circular dichroism (ECD) spectroscopy, with 1 exhibiting a high-amplitude positive Cotton effect at 226 nm (Figure 3), characteristic for β-substituted flavan-3-ol moi-

Figure 3. ECD spectra of 1 and 2 in methanol.

eties. 31,32 The ECD spectrum also supports the C-2 configurations for units II and IV in 1, displaying a weak negative Cotton effect at 272 nm [a weak positive Cotton effect at ca. 270 nm is characteristic for ent-(epi)catechin subunits]. Therefore, the structure of 1 was established as EC-(4β→ 8,2β→O→7)-EC-(4β→6)-EC-(4β→8,2β→O→7)-C. The ESIMS spectra of 2 exhibit a similar molecular pattern and molecular ions at the same m/z values compared to 1. The HRMS data showed an [M + H]+ ion at m/z 1151.2468, which agrees with the calculated m/z 1151.2457 for the molecular formula, C60H47O24. The chemical shifts of 2 are also similar to those of 1, suggesting that 2 has an analogous tetrameric structure. The differences in the NMR spectra involve the chemical shifts assignable to protons and carbons of the terminal unit. The 1H NMR chemical shifts of H-4 (unit IV, Lring) (δH 2.84 and 2.97), as well as the small J-coupling of H-4β (4.4 Hz), are characteristic for an epicatechin moiety. The ECD spectrum of 2 (Figure 3) is similar to that of 1, hence confirming the all-β configurations of both the A-linkages and the single B-type linkage. Therefore, the structure of 2 is defined as EC-(4β→8,2β→O→7)-EC-(4β→6)-EC-(4β→ 8,2β→O→7)-EC. The 1H NMR spectra of the tetramers recorded in methanold4 revealed the presence of rotational isomers in the ratio of 3.3:1 and 5:1 for 1 and 2, respectively. The presence of rotamers results from the hindered rotation along the B-type bond in the procyanidin structure. Table 1 shows a collation of the 1H and 13C NMR chemical shifts of 1 and 2 and their rotamers. 418



Article

Table 1. Chemical Shifts (ppm) for 1 (600 MHz, T = 273 K) and 2 (500 MHz, 273 K) in Methanol-d4 (J Values in Hz are Given in Parentheses) 1

2

major rotamer δC, type

ring

no.

C

2 3 4 5 6 7 8 9 10 1′ 2′ 3′ 4′ 5′ 6′ 2 3 4 5 6 7 8 9 10 1′ 2′ 3′ 4′ 5′ 6′ 2 3 4 5 6 7 8 9 10 1′ 2′ 3′ 4′ 5′ 6′ 2 3 4

100.5, 68.8, 29.8, 157.6, 98.6, 158.5, 96.1, 154.6, 104.8, 133.0, 116.1, 146.1, 147.2, 116.0, 112.2, 79.4, 72.9, 38.8, 157.2, 96.7, 156.6, 107.2, 152.6, 106.5, 132.1, 116.7, 146.3, 146.6, 116.3, 121.1, 100.7, 68.2, 30.2, 154.5, 111.7, 157.6, 98.0, 152.8, 104.6, 132.5, 116.1, 146.1, 147.3, 116.0, 120.3, 85.7, 68.7, 29.9,

C CH CH C CH C CH C C C CH C C CH CH CH CH CH C CH C C C C C CH C C CH CH C CH CH C C C CH C C C CH C C CH CH CH C CH2

5 6 7 8 9 10 1′ 2′ 3′ 4′

156.7, 97.0, 152.6, 106.8, 151.8, 103.6, 130.4, 116.8, 146.6, 147.4,

C CH C C C C C C C C

A

B

F

D

E

H

G

I

L

J

K

δH (J in Hz) 4.08, d (3.5) 4.43, d (3.5)

minor rotamer δC, type 100.5,a C 68.4, CH 30.1, CH

major rotamer

δH (J in Hz) 4.05, d (3.6) 4.71,m

6.03, m

98.6, CH

6.07, m

6.07, d (2.3)

97.1, CH

6.09, d (2.4)

105.1, C 7.13, d (2.2)

6.80, d (8.2) 7.015, dd (8.2, 2.2) 5.33, s 3.93, d (1.5) 4.73, m

80.2, CH 74.1, CH 38.2, CH

5.76, s 3.98, bs 4.62, bs

6.05, s

97.0, CH

5.96, s

107.1, C

7.105, d (1.9)

6.76, d (8.1) 6.92, dd (8.1, 1.9) 4.10, d (3.5) 4.19, d (3.5)

107.0, C 132.5, C 116.1, CH

120.7, CH 100.6,a C 68.6, CH 29.7, CH

7.15, d (1.6)

6.98, m 4.12, m 4.11, m

111.5, C 6.03, m

97.0, CH

6.22, s

104.3, C 7.109, d (2.2)

6.80, d (8.2) 7.011, dd (8.2, 2.2) 4.72, d (9.0) 4.16, bd (5.9) 2.56, dd (16.5, 9.0) 3.02, dd (16.5, 5.9) 6.11, s

85.6, CH 68.6, CH 30.3, CH2

4.53, 4.15, 2.54, 3.07,

97.3, CH

6.13, s

107.0, 149.9, 103.9, 130.7,

d (8.6) d (5.9) m dd (16.4, 5.9)

δC, type 100.5,a C 68.8, CH 29.9, CH 157.6, C 98.7, CH 158.6, C 97.1, CH 154.6, C 104.7, C 132.9, C 116.1, CH 146.2, C 147.2, C 116.0, CH 120.3, CH 79.2, CH 72.9, CH 39.0, CH 157.5, C 96.5, CH 156.8, C 107.4, C 152.7, C 106.1, C 131.9, C 116.7, CH 146.4, C 146.5, C 116.4, CH 121.1, CH 100.6,a C 68.2, CH 30.1, CH 155.0, C 111.1, C 152.9, C 97.9, CH 152.7, C 104.5, C 132.6, C 116.1, CH 146.1, C 146.4, C 116.0, CH 120.3, CH 83.2, CH 67.4, CH 31.1, CH 157.5, 97.2, 152.9, 107.2, 152.7, 102.6, 130.9, 117.0, 146.3, 147.3,

C C C C

6.98, d (2.0)

419

C CH C C C C C CH C C

δH (J in Hz)

minor rotamer δC, type

δH (J in Hz)

4.08, d (3.3) 4.43, d (3.3)

100.1,a C 68.7, CH 30.0, CH

4.04, m 4.68, m

6.03, d (2.3)

98.8, CH

6.06, m

6.08, d (2.3)

97.0, CH

6.09, d (2.4)

104.7, C 7.14, d (2.1)

6.80, 7.02, 5.23, 4.00, 4.69,

d (8.3) m s d (1.6) s

6.07, s

80.8, CH 73.7, CH 38.2, CH

5.76, s 4.04, m 4.62, bs

97.3, CH

5.97, s

107.3, C 107.2, C 7.116, d (2.1)

6.77, d (8.1) 6.93, dd (8.1, 2.1) 4.12, d (3.5) 4.32, d (3.5)

100.4,a C 68.6, CH 29.7, CH

4.17, m 4.21, d (3.7)

111.7, C 6.04, s

96.8, CH

6.22, s

104.2, C 7.123, d (2.3)

6.81, 7.02, 4.96, 4.18, 2.84, 2.97,

d (8.3) m s m bd (17.1) dd (17.1, 4.4)

6.13, s

82.7, CH 67.0, CH 30.6, CH2

4.85, 4.12, 2.82, 2.93,

m m m dd (17.8, 4.8)

97.4, CH

6.13, s

102.7, C 7.27, d (1.8)

116.9, CH

7.22, d (1.5)



Article

Table 1. continued 1

2

major rotamer ring

no. 5′ 6′

a

δC, type 117.3, C 121.8, C

δH (J in Hz) 6.87, d (8.1) 6.94, dd (8.1, 2.0)

minor rotamer δC, type 118.5, C 119.7, C

major rotamer

δH (J in Hz)

δC, type

6.79, m 6.81, m

116.5, CH 121.9, CH

δH (J in Hz) 6.82, d (8.2) 7.00, m

minor rotamer δC, type 121.8, CH

δH (J in Hz) 7.08, m

Signals followed by the same letter within each column are exchangeable. 1

4−C-8 linkage. However, there are cited cases where the assignment of the C-5/C-9 signals is opposite [e.g., for lichitannin A-2 it is δC 149.15 for C-5 (unit II, D-ring) and δC 154.32 for C-9 (unit II, D-ring)].34 In order to resolve these ambiguities, spectra of 3 were recorded in anhydrous DMSOd6, to gain additional information from hydroxy proton correlations. The 5-OH protons of units I and II (rings A and D, respectively) were assigned as singlets at δH 9.05 and 8.51, respectively, on the basis of their ROESY correlations with H-4 signals at δH 4.47 (unit I, C ring) and 4.28 (unit II, F-ring). The assignment is also confirmed by a weak HMBC cross-peak between the δH 9.05 signal and quaternary carbon at δC 102.2 attributable to C-10 (unit I, A-ring). The other ROESY correlations of 5-OH (unit I, A-ring) (singlet at δH 9.05) with H-2′ and H-6′ (unit II, E-ring) (doublet at δH 7.07 and doublet of doublets at δH 7.43, respectively) and 5-OH (unit II, D-ring) (singlet at δH 8.51) with H-2′ and H-6′ (unit III, H-ring) (doublet at δH 6.87 and doublet of doublets at δH 6.86) confirm the C-4−C-8 linkage between both subunits (Figure 5). Full assignment of the 1H and 13C NMR resonances of 3 in DMSOd6, based on the COSY, HSQC, and Impact-HMBC, can be found in the Supporting Information. The ECD spectrum of 3 (Figure 6) confirms the βsubstitution via the strong positive Cotton effect at 228 nm. Collectively these data permitted definition of the structure of 3 as EC-(4β→8,2β→O→7)-EC-(4β → 8,2β→O→7)-C, a compound similar to pavetannin B8;33 however pavetannin B8 possesses an ent-catechin instead of a catechin moiety as the third unit. Compound 4 was isolated as a light tan solid. Its molecular formula of C45H36O18 was determined by ESIMS (ESIMS deprotonated and protonated molecular ions at m/z 863.5 and 865.4, respectively) and 13C NMR data. The HRMS showed an [M + H]+ ion at m/z 865.1991, which agrees with the calculated m/z 865.1980 value for the molecular formula of C45H37O18. The observed mass, as well as the NMR spectra, indicates that 4 is a procyanidin trimer with one A-type and one B-type bond. The 1H NMR spectrum of 4 shows two metacoupled protons at δH 6.03 and 6.08 originating from the Aring. This upper unit is a C-2/C-4 doubly linked epicatechin moiety, as may be concluded from the chemical shifts of H-3 and H-4 (unit I, C-ring) of δH 4.08 and 4.53, respectively, and their small J-couplings of ca. 3 Hz, along with the presence of a ketal carbon at δC 100.05. The middle unit of 4 is an epicatechin moiety [H-2 and H-3 (unit II, F-ring), singlets, at δH 5.10 and 4.04, respectively; see Table 2], while the lower unit is a catechin moiety [J values of H-2 and H-4 (unit III, Iring) ca. 7 and 5 Hz]. The low-amplitude negative Cotton effect near 280 nm is reminiscent of the α-orientation of the C2 substituents of the constituent flavan-3-ol moieties and hence of the presence of a catechin GHI-moiety in 4 (Figure 6).35 The NMR data of units I and II are similar to those of procyanidin A2,30 whereas the terminal unit has highly similar

H NMR data compared to those of EC-(4β→8)-EC-(4β→6)C,36 suggesting that the structure of 4 is EC-(4β→8,2β→O→ 7)-EC-(4β→6)-C. The ROESY correlations between H-4 (unit I, C-ring) (δH 4.53) and H-2′ (unit II, E-ring) (δH 7.14), as well as a weak correlation between H-6 (unit I, A-ring) (δH 6.03) and H-2′ (unit II, E-ring), strongly support the C-4−C-8 linkage between units I and II. Also, the chemical shift of C-8 (unit II, D-ring, δC 107.80) favors the proposed linkage. Similarly, the second interflavan linkage can be established on the basis of ROESY correlations. The C-4−C-8 linkage would result in the proximity of H-3 and H-4 (unit I, C-ring) and H2′/H-6′ (unit III, H-ring). The absence of the respective ROESY correlations supports the C-4−C-6 linkage. The C-6 chemical shift at δC 109.14 (unit III, H-ring) is in agreement with these observations. Additionally, the 1H and 13C NMR chemical shifts of 4 closely resemble those of the less-abundant cinnamtannin D-1 rotamer.37 This similarity further confirms the C-4−C-6 linkage between units II→III, as the 3D structure of 4 with this linkage is similar to that of the less-abundant rotamer of cinnamtannin D-1 (Figure 7). The β-configuration of the interflavan linkages of 4 is supported by the strong positive Cotton effect at 228 nm in the ECD spectrum (Figure 6), which originates from the interaction of the two βconfigured aryl chromophores.38 All of these data confirm the proposed structure of 4. The procedures of structural elucidation for the compounds 1−4 indicate the complexity of procyanidin structure determination, especially in terms of establishing the position of the interflavan linkage. The view persists that the linkage location can be determined based on the C-6 and C-8 chemical shifts. The chemical shift range for a substituted C-8 (ca. δC 107) is thought to be always lower than that of C-6 (ca. δC 109).29,30 This, indeed, is often true, and many researchers place the linkage location solely on that assumption. There are however cases when the linkage was determined to be C-4−C-8 (e.g., with derivatization methods) with the C-8 chemical shift even higher than δC 109.4 Evidently, this assumption does not always lead to a correct structure, which is the likely reason behind ambiguous structural NMR assignments occasionally found in the literature.39 Our results emphasize the need to incorporate into the analysis of an interflavan linkage other experimental features, such as ROESY correlations, preferably in a solvent in which hydroxy proton correlations are observable. The 1H and 13C NMR spectra of compounds 5 and 6 (Table 2), as well as their ESIMS spectra, indicate that they are both trimeric procyanidins with one A-type and one B-type interflavan bond. According to the HMBC cross-peaks, the Atype bond is located between the middle and lower units in both compounds, and the B-type between the upper and middle units. The order of units in 5 was established as ent-epicatechin, epicatechin, and catechin on the basis of their respective Jcouplings of H-2−H-4 (unit I, C-ring), as well as an unusual 420


421

G

H

E

D

F

B

5 6 7 8 9 10

2 3 4 5 6 7 8 9 10 1′ 2′ 3′ 4′ 5′ 6′ 2 3 4 5 6 7 8 9 10 1′ 2′ 3′ 4′ 5′ 6′ 2 3 4

C

A

no.

ring

δC, type

151.6, 96.5, 153.3, 107.8, 156.6, 103.6,

C CH C C C C

100.3,C 69.1, CH 30.2, CH 158.5, C 97.9, CH 158.6, C 96.6, C 154.8, C 104.0, C 133.1, C 115.8, CH 146.0, C 147.1, C 116.0, CH 120.1, CH 101.2, C 68.4, CH 30.1, CH 149.5, C 98.4, CH 154.0, C 109.2, C 156.4, C 105.7, C 132.9, C 115.8, CH 146.0, C 147.3, C 116.0, CH 120.9, CH 84.9, CH 69.0, CH 30.2, CH2

δH (J in Hz)

d (8.4) dd (8.4; 2.1) d (8.8) m dd (16.5; 5.7) dd (16.5; 8.8)

6.08, s

6.88, 7.46, 4.65, 4.07, 3.01, 2.56,

7.37, d (2.1)

6.12, s

4.06, d (3.0) 4.32, d (3.0)

6.77, d (8.3) 6.98, dd (8.3; 2.1)

7.08, d (2.1)

6.04, d (2.3)

6.00, d (2.3)

4.02, d (3.1) 4.71, d (3.1)

3

155.3, 109.1, 156.4, 96.4, 156.9, 101.6,

C C C CH C C

C CH CH C CH C CH C C C CH C C CH CH CH CH CH C CH C C C C C CH C C CH CH CH CH CH2

δC, type 100.6, 68.7, 29.9, 157.8, 98.6, 158.6, 96.9, 154.6, 104.4, 132.8, 116.0, 146.1, 147.2, 116.0, 120.2, 79.4, 72.9, 38.8, 158.0, 96.9, 154.1, 107.8, 153.1, 103.6, 131.8, 116.4, 146.6, 146.3, 116.4, 120.8, 82.9, 69.2, 28.8,

δH (J in Hz)

d (8.3) dd (8.3; 1.8) s bs m

d(8.1) m d (7.8) m dd (16.5; 7.8) dd (16.5; 4.9)

6.04, s

6.80, 6.97, 4.61, 4.01, 2.45, 2.72,

7.14, m

6.13, s

6.81, 7.02, 5.10, 4.04, 4.60,

7.13, d (1.8)

6.08, d (2.3)

6.03, d (2.3)

4.08, d (3.1) 4.53, d (3.1)

4

156.7, 97.0, 152.7, 107.4, 151.9, 103.7,

77.7, 73.8, 37.8, 158.7, 97.0, 158.4, 96.6, 156.7, 102.3, 133.1, 116.4, 146.1, 146.2, 116.1, 120.6, 101.1, 68.0, 30.0, 155.9, 99.8, 157.6, 108.9, 151.7, 104.5, 132.8, 116.3, 146.3, 147.3, 115.7, 119.9, 85.1, 68.7, 29.8, C CH C C C C

CH CH CH C CH C CH C C C CH C C CH CH C CH CH C CH C C C C C CH C C CH CH CH CH CH2

δC, type

Table 2. 1H and 13C NMR Chemical Shifts [ppm] for 3−6 (in Methanol-d4, at 273 K)

δH (J in Hz)

d (8.4) dd (8.4; 2.0) d (8.1) ddd (8.8; 8.1; 5.6) dd (16.3; 8.8) dd (16.3; 5.6) 6.03, s

6.79, 7.17, 4.71, 4.15, 2.57, 3.00,

7.29, d (2.0)

5.93, s

4.13, d (3.4) 4.29, d (3.4)

6.67, d(8.2) 6.62, dd (8.2; 1.7)

6.82, d (1.7)

5.98, bs

5.95, bs

5.04, s 3.90, bs 4.77, bs

5 δC, type

156.8, 97.1, 152.7, 106.9, 151.7, 104.3,

C CH C C C C

77.5, CH 73.3,CH 38.0, CH 157.4, C 96.6, CH 158.5, C 96.2, CH 158.4, C 102.6, C 133.1, C 115.6, CH 146.3, C 146.0, C 116.2, CH 119.7, CH 100.7, C 68.2, CH 30.1, CH 154.5, C 111.1, C 157.6, C 98.0, CH 152.6, C 104.5, C 132.6, C 116.1, CH 146.1, C 147.2, C 116.0, CH 120.3, CH 85.7, CH 69.3, CH 29.7, CH2 d (8.0) m d (8.4) m dd (16.4; 8.4) dd (16.4; 5.4) 6.11, s

6.80, 7.01, 4.75, 3.90, 2.58, 3.00,

7.12, m

6.03, s

4.10, d (3.4) 4.20, d (3.4)

6.69, d (8.2) 6.65, m

6.87, m

5.98, d (1.8)

5.92, d (1.8)

5.05, s 3.90, m 4.65, s

δH (J in Hz)

major rotamer

6

156.8, 97.3, 152.7, 107.4, 152.3, 103.6,

C CH C C C C

CH CH CH C CH C CH C C C CH C C CH CH C CH CH C C C CH C C C CH C C CH CH CH CH CH2

δC, type 78.3, 74.0, 37.3, 158.8, 96.4, 158.4, 96.5, 158.3, 103.9, 133.4, 115.7, 146.4, 146.1, 116.3, 119.9, 100.4, 68.2, 29.7, 156.8, 111.8, 155.5, 97.0, 152.5, 104.8, 132.6, 116.1, 146.1, 147.3, 116.0, 120.3, 86.2, 68.7, 31.6,

d (8.2) m d (9.6) m dd (16.4; 9.6) dd (16.4; 6.0) 6.11, s

6.81, 7.01, 4.36, 4.20, 2.48, 3.07,

7.10, bs

6.20, s

4.05, d (3.1) 4.07, d (3.1)

6.72, d (8.2) 6.61, m

6.93, m

6.18, d (1.7)

5.86, d (1.7)

5.45, s 3.81, bs 4.64, s

δH (J in Hz)

minor rotamer

Journal of Natural Products Article


7.00, m 6.78, m

C CH C C CH CH

6.74, bs

δH (J in Hz)

ROESY correlation between H-3 and H-4 (unit I, C-ring), indicating that H-3 and H-4 (unit I, C-ring) are in a cis configuration. The ECD spectrum of 5 shows a strong positive Cotton effect at 223 nm, which originates from 4β-configured aryl rings. As a result H-4 (unit I, C-ring) is in an α-position, together with H-2 and H-3, which were earlier determined to be cis-oriented. Together these data indicate that unit I is an ent-epicatechin linked to unit II by a 4β-linkage. This very unusual configuration was described for the first time in 20159 and said to be exceptional, as the formation of 2,3-cis-3,4-trans linkages is more favorable.40 The order of flavan-3-ol units of 6 is epicatechin, epicatechin, and catechin. The main difference between 5 and 6 is the location of the B-type interflavan bond. In 5 the C-8 (unit II, D-ring) chemical shift (δC 108.9) is at higher field than in 6 (δC 111.1), indicating a C-4−C-8 linkage in 5 and a C-4−C-6 linkage in 6. The 4β-configurations in both trimers were confirmed by the ECD spectra (Figure S23). Therefore, the structure of 5 was defined as ent-EC-(4β→8)EC-(4β→8,2β→O→7)-C and 6 as EC-(4β→6)-EC-(4β→ 8,2β→O→7)-C. Both compounds 5 and 6 are known,9,10,28,41 but only 6 was reported to be present in peanut skins.10,28 Note that in ref 28 it was erroneously concluded that 6 has a chemical structure of 5, probably due to the poor resolution of the NMR spectra of 5 at room temperature. The published 1H NMR spectrum of the compound describes it as EC-(4β→8)EC-(4β→8,2β→O→7)-C, but its chemical shifts closely resemble the 1H NMR spectrum and chemical shifts of 6, i.e., EC-(4β→6)-EC-(4β→8,2β→O→7)-C, at room temperature (see Figure 24S and Supporting Information in ref 28). Interestingly, compound 6 exhibits the presence of two rotational isomers in a 1:0.9 ratio, whereas 5, despite structural similarity, exists as only one rotamer, even at 250 K. The ability to adopt two different conformations may influence potential biological activity of 5 and 6, especially in terms of interaction with proteins. Such differences in procyanidin−protein interactions were reported by Pianet and co-workers.21 Suppression of TNF-α Secretion by the Peanut Skin Procyanidins. Procyanidin A1, the trimers 3, 4, 5, and 6, and the tetramers 1 and 2 were evaluated for anti-inflammatory activity in an in vitro assay, in which lipopolysaccharide (LPS)stimulated macrophages responded with secretion of TNF-α, a pro-inflammatory cytokine (Table 3). They were tested at 100, 50, and 10 μg/mL. Tetrameric peanut procyanidin F (2) was the most potent, suppressing TNF-α secretion by 82% at 8.7 μM (10 μg/mL) and by 100% at 43.4 μM (50 μg/mL), while tetramer E (1) at the same concentrations showed 4% and 71% suppression, respectively. The most potent among the trimers was peanut procyanidin C (4), suppressing TNF-α secretion by 28% at 11.6 μM (10 μg/mL) and by 99% at 57.9 μM (50 μg/ mL). Dimer A1 and peanut procyanidin A (5) were inactive at the highest tested concentration of 115 μM (100 μg/mL). The results of inhibition of the TNF-α secretion indicate that apparently small structural differences like between tetramers E (1) and F (2) can be strongly differentiated in biological systems. The TNF-α suppressing activity could be potentially exploited for therapeutic applications, providing the delivery of a procyanidin bypasses the oral route, which is not effective for absorption.

6.76, d (8.1) 6.72, m

6.84, m

C CH C C CH CH 6.97, d (1.9)

6.82, d (8.2) 6.90, dd (8.2; 1.9)

δC, type

132.5, 115.4, 146.7, 146.7, 116.5, 120.3,

δH (J in Hz)

C CH C C CH CH

δC, type

131.2, 116.4, 146.7, 147.2, 116.6, 121.1, 1′ 2′ 3′ 4′ 5′ 6′

no.

■

EXPERIMENTAL SECTION

General Experimental Procedures. 1D and 2D NMR spectra were recorded on a Bruker Avance III 500 or 600 spectrometer (Bruker BioSpin, Rheinstetten, Germany). The Bruker Avance III 500

I

ring

6.83, d (8.1) 6.87, dd (8.2; 1.8)

6.95, d (1.7)

C CH C C CH CH 131.0, 116.8, 146.9, 147.4, 116.4, 121.3,

δH (J in Hz)

δC, type

5 4 3

Table 2. continued

6.86, m 6.90, m

7.00, m

δH (J in Hz)

C CH C C CH CH

δC, type δH (J in Hz)

130.6, 116.4, 146.7, 147.4, 117.2, 121.7,

major rotamer

6

δC, type

Article

130.2, 118.4, 145.9, 147.0, 118.7, 120.2,

minor rotamer


422



Article

Figure 4. 3D structures of EC-(4β→8,2β→O→7)-EC-(4β→8,2β→O→7)-C (3) (left) and EC-(4β→6,2β→O→7)-EC-(4β→8,2β→O→7)-C (right), with marked H4−H2′ distances as visualized by GaussView04 software.

Figure 5. Key ROESY correlations for 3 in anhydrous DMSO-d6. MHz spectrometer (operating at 500.13 MHz for 1H and 125.76 MHz for 13C) was equipped with a 5 mm BBI probe head with an actively shielded Z-gradient coil connected to a GAB/2 gradient unit capable of producing B0 gradients with a maximum strength of 50 G/cm. During all measurements, the temperature was set at and was controlled and stabilized with a BCU 05 cooling unit controlled by a BVT3200 variable-temperature unit. The low-temperature spectra were recorded using a liquid nitrogen evaporator, controlled by a BVT3200 unit. The Bruker Avance III 600 MHz spectrometer (operating at 600.14 and 150.92 MHz for 1H and 13C, respectively) was equipped with a 5 mm BBFO probe head with an actively shielded Z-gradient coil connected to a GAB/2 gradient unit capable of producing B0 gradients with a maximum strength of 50 G/cm. During all measurements, the temperature was controlled and stabilized with the BCU X-treme cooling unit controlled by a BVT3200 variabletemperature unit. All spectra were recorded using 3 mm NMR tubes (Norell) and methanol-d4 or DMSO-d6 (Armar Chemical, Switzerland) as a solvent. Chemical shifts were referenced to the residual solvent signals: 3.31 and 49.50 ppm for 1H and 13C in methanol-d4 and 2.50 and 39.51 ppm for 1H and 13C in DMSO-d6, respectively. All spectra, except long-range 1H−13C couplings, were acquired with the original Bruker pulse sequence. For observation of the long-range correlations the Impact-HMBC pulse sequence42 was used.

On both spectrometers, the spectra were acquired and processed using the TopSpin 3.1 program (Bruker BioSpin) running under Windows 7 (64 bit) OS on an HP Z700 workstation, used for operating and controlling the spectrometers. The experimental ECD spectra were recorded using a Jasco J-815 spectrometer at room temperature in spectroscopic grade MeOH. Solutions with a concentration of 6.2 × 10−4 M were measured in a quartz cell with a path length of 1−0.1 cm. All spectra were recorded between 600 and 190 nm. Mass spectra were recorded on a 4000-QTRAP Applied Biosystems spectrometer, using positive and negative electrospray ionization. Isolation of Peanut Procyanidins. Analysis of the procyanidin content in peanut skins was done via HPLC analysis (Agilent 1050 w/ diode array detector) using a Jupiter Proteo 4 μm 90 Å 4.6 × 150 mm column with a gradient from 10% to 34% MeCN in H2O with 0.1% trifluoroacetic acid over 22 min. The concentration of procyanidins in the skin was determined to be 0.33 (1), 0.28 (dimer A1), 0.14 (4), 0.13 (6), 0.058 (dimer A2), 0.031 (3), 0.030 (5), and 0.028% (2). The TLC analysis of the fractions was performed on Kieselgel 60 Merck F254 plates developed using a solution comprising EtOAc−HOAc− H2O (5:1:1) and visualized by spraying with CeSO4 dissolved in 10% H2SO4 followed by heating with a heat gun until optimum color development. 423



Article

Table 3. Inhibition of TNF-α Secretion from the LPSStimulated Macrophages by Tetrameric and Trimeric Peanut Procyanidins tetramer 1 2 trimer 3 4 6

10 μg/mL (8.69 μM) 4.26 ± 11.13 82.25 ± 3.37 10 μg/mL (11.6 μM) 6.28 ± 16.58 28.60 ± 23.62 0.00 ± 11.93

50 μg/mL (43.5 μM)

100 μg/mL (86.9 μM)

71.18 ± 4.66 100.00 ± 0.00 50 μg/mL (57.9 μM)

98.64 ± 0.65 100.00 ± 0.00 100 μg/mL (115.7 μM)

78.75 ± 1.36 99.36 ± 0.57 89.72 ± 6.70

100.00 ± 0.00 100.00 ± 0.00 100.00 ± 0.00

grade silica gel 0.063-02 mm (Machery-Nagel) and fractionated on a 50 × 500 mm glass column loaded with 700 g of the same grade silica gel with an approximate column volume (CV) of 0.2 L. The solvent was delivered at 10 mL/min using an SSI Flash-300 pump. The initial solvent system, comprising n-hexane (62%), EtOAc (15%), and MeOH (23%) with an addition of 0.3% v/v HOAc, was pumped through the system for 3.6 CV, followed by a gradient of n-hexane (42%), EtOAc (18%), and MeOH (40%) over 8 CV. The 11.2 L of eluent was collected in 200 mL fractions and combined into six cuts based on the TLC analysis as follows: S1 [21−28 (0.8 g)], S2 [30−33 (0.47 g)], S3 [34−37 (0.31 g)], S4 [38−45 (2.08 g)], S5 [46−49 (1.48 g)], and S6 [50−54 (0.8 g)]. A second purification step was performed using centrifugal partition chromatography (CPC) on a Kromaton A100 FCPC (Rousselet Robatel, Annonay, France) with a 1 L rotor. Peanut tetrameric procyanidin 1 was purified from fraction S1 in a solvent system comprising heptane−EtOAc−MeOH−H2O (1:6:2:5), v/v. An 800 mg amount of fraction S1 was dissolved in 50 mL of stationary phase and pumped onto the rotor at 200 rpm, and then 3.6 L of mobile phase was eluted in ascending mode at 1500 rpm (back pressure 800 psi). The eluent was collected at 25 mL/test tube, and from the HPLC analysis 1 (314 mg) was present in tubes 100−136 with 98% purity. The enriched procyanidin 2 was found in test tubes 57−71 (98 mg) and required additional purification by preparative HPLC. Tetramer 2 eluted from 4 to 4.5 CV giving 48 mg at 98% purity. Fraction S2, containing 10% of 3, was dissolved in 50 mL of stationary phase of the solvent system n-hexane−EtOAc−MeOH−H2O (1.5:5:1:5) in ascending mode, pumped into the rotor at 200 rpm, and then eluted with 1.8 L of mobile phase at 1250 rpm (800 psi back pressure) collected at 25 mL/test tube. Based on the HPLC analysis, the contents of test tubes 45−63 were combined, yielding approximately 24 mg of 3 at 90% purity. Peanut procyanidins 4 and 5 present in fraction S4 were enriched by CPC using the solvent system n-hexane−EtOAc− MeOH−H2O (1.5:5:1.5:5), and fractions were collected from test tubes 138−194 and 77−98, respectively. A CPC fractionation of S5 in the aforementioned solvent system produced a highly enriched fraction of 6 in test tubes 81−101 (318 mg). The final HPLC purifications for peanut procyanidins 4, 5, and 6 were performed on a YMC ODS-AQ 5 μm 120 Å 30 × 150 mm column in a gradient from 10% to 22% MeCN in H2O acidified with 0.5% HOAc over 6.6 CV, with collection ranging from 4.2 to 6 CV yielding 63 mg of 4, 42 mg of 5, and 236 mg of 6. Peanut procyanidin E (1): off-white solid; ECD (6.2 × 10−4 M, MeOH) λmax (Δε) 209 (−89.41), 226 (+76.96), 273 (−6.98) nm; 1H and 13C NMR data, see Table 1; HRESIMS m/z 1151.2463 [M + H]+ (calcd for C60H47O24, 1151.2457). Peanut procyanidin F (2): light brown solid; ECD (6.2 × 10−4 M, MeOH) λmax (Δε) 209 (−113.58), 226 (+79.82), 275 (−7.38) nm; 1H and 13C NMR data, see Table 1; HRESIMS m/z 1151.2468 [M + H]+ (calcd for C60H47O24, 1151.2457). Peanut procyanidin B (3): light brown solid; ECD (6.2 × 10−4 M, MeOH) λmax (Δε) 208 (−122.17), 228 (+133.48), 273 (−11.97) nm; 1 H and 13C NMR data, see Table 2; HRESIMS m/z 863.1834 [M + H]+ (calcd for C45H35O18, 863.1823).

Figure 6. ECD spectra of compounds 3, 4, and lindetannin in MeOH.

Figure 7. 3D structures of 4 (left) and the lower-abundant rotamer of cinnamtannin D-1 (right) as visualized by GaussView04 software. Dried peanut skins (2.5 kg) obtained from Universal Blanchers (Sylvester, GA, USA) were extracted by percolation using 6 L of 30% MeOH and then 6 L of 70% aqueous acetone. The combined extracts after solvent removal yielded approximately 240 g of a solid residue. Fifty grams of this solid was dissolved in 300 mL of H2O at 40 °C and partitioned three times against 300 mL of EtOAc containing 5%, 10%, and 15% of EtOH. The 10% EtOH extract was loaded onto coarse424



■

Peanut procyanidin C (4): light tan solid; ECD (6.2 × 10−4 M, MeOH) λmax (Δε) 206 (−79.76), 229 (+74.75), 273 (−5.22) nm; 1H and 13C NMR data, see Table 2; HRESIMS m/z 865.1991 [M + H]+ (calcd for C45H37O18, 865.1980). Peanut procyanidin A (5): off-white solid; ECD (6.2 × 10−4 M, MeOH) λmax (Δε) 204 (−42.48), 223 (+55.11), 274 (−6.62) nm; 1H and 13C NMR data, see Table 2; ESIMS m/z 865.3 [M + H]+ (calcd for C45H35O18, 865.2). Peanut procyanidin D (6): white solid; ECD (6.2 × 10−4 M, MeOH) λmax (Δε) 196 (−15.56), 217 (+21.39), 239 (+24.31), 273 (−11.97) nm; 1H and 13C NMR data, see Table 2; ESIMS m/z 865.4 [M + H]+ (calcd for C45H35O18, 865.2). Anti-inflammatory Activity of Peanut Procyanidins. RAW264.7 cells, a murine macrophage cell line (American Type Culture Collection (ATCC), Manassas, VA, USA), were cultured in Dulbecco’s modified Eagle’s medium (ATCC), containing 10% fetal bovine serum (Hyclone, Logan, UT, USA). RAW264.7 cells were plated at a density of 5 × 105 and cultivated at 37 °C in a humidified atmosphere containing 5% CO2. For all experiments, cells were grown to 80−90% confluence. Cells were preincubated with 0−100 μg/mL of peanut procyanidins dissolved in sterile water for 1 h followed by an additional 23 h incubation with LPS (1 μg/mL, LPS, Escherichia coli 0111:B4; Sigma-Aldrich, St. Louis, MO, USA), and the media were collected and stored at −20 °C. The production/secretion of TNF-α in the culture medium was measured by ELISA according to the manufacturer’s protocol (Peprotech, Rocky Hill, NJ, USA). The antiinflammatory activity is expressed as the inhibition of LPS-dependent production of TNF-α from activated RAW264.7 cells (Table 3). The statistical analysis was performed by using SigmaPlot 11.2 (Systat Software, San Jose, CA, USA).

■

ASSOCIATED CONTENT

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.6b00946. 1 H, 13C, and 2D NMR spectra of 1−4 in methanol-d4 (Figures S1−S22), ECD spectra of 5 and 6 (Figure S23), 1 H NMR spectra of 6 at 295 and 273 K in methanol-d4 (Figures S24 and S25), and 1H NMR spectrum of 5 in methanol-d4 (Figure S26) (PDF)

AUTHOR INFORMATION

Corresponding Authors

*Tel (M. Dudek): +48 42 6803306. E-mail: mdudek@cbmm. lodz.pl. *Tel (J. Glinski): (860) 799-5356. E-mail: jan@plantaanalytica. com. ORCID

Marta K. Dudek: 0000-0003-3412-0177 Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Ferreira, D.; Bekker, R. Nat. Prod. Rep. 1996, 13, 411−433. (2) Khan, M. L.; Haslam, E.; Williamson, M. P. Magn. Reson. Chem. 1997, 35, 854−858. (3) Killday, K. B.; Davey, M. H.; Glinski, J. A.; Duan, P.; Veluri, R.; Proni, G.; Daugherty, F. J.; Tempesta, M. S. J. Nat. Prod. 2011, 74, 1833−1841. (4) Lou, H.; Yuan, H.; Ma, B.; Ren, D.; Ji, M.; Oka, S. Phytochemistry 2004, 65, 2391−2399. (5) Xie, D.-Y.; Dixon, R. A. Phytochemistry 2005, 66, 2127−2144. (6) Foo, L. Y.; Lu, Y.; Howell, A. B.; Vorsa, N. J. Nat. Prod. 2000, 63, 1225−1228. (7) Krueger, C. G.; Reed, J. D.; Feliciano, R. P.; Howell, A. B. Anal. Bioanal. Chem. 2013, 405, 4385−4395. (8) Chen, L.; Yang, Y.; Yuan, P.; Yang, Y.; Chen, K.; Jia, Q.; Li, Y. Evid. Based Complement. Alternat. Med. 2014, 2014, 365258. (9) Nam, J.-W.; Phansalkar, R. S.; Lankin, D. C.; Bisson, J.; McAlpine, J. B.; Leme, A. A.; Vidal, C. M. P.; Ramirez, B.; Niemitz, M.; Bedran-Russo, A.; Chen, S.-N.; Pauli, G. F. J. Org. Chem. 2015, 80, 7495−7507. (10) Tamura, T.; Inoue, N.; Ozawa, M.; Shimizu-Ibuka, A.; Arai, S.; Abe, N.; Koshino, H.; Mura, K. Biosci., Biotechnol., Biochem. 2013, 77, 1306−1309. (11) Leite da Silva; Porto, P. A.; Nave Laranjinha, J. A.; Pereira de Freitas, V. A. Biochem. Pharmacol. 2003, 66, 947−954. (12) Lin, H.-C.; Lee, S.-S. J. Nat. Prod. 2010, 73, 1375−1380. (13) Khan, N. Q.; Patel, B.; Kang, S. S.; Dhariwal, S. K.; Husain, F.; Wood, E. G.; Pothecary, M. R.; Corder, R. Tetrahedron 2015, 71, 3059−3065. (14) Kirch, N.; Ellinger, S. Ernahrungs Umschau 2014, 61, 144−151. (15) Mandel, S.; Youdim, M. B. H. Free Radical Biol. Med. 2004, 37, 304−317. (16) Anderson, R. A.; Broadhurst, C. L.; Polansky, M. M.; Schmidt, W. F.; Khan, A.; Flanagan, V. P.; Schoene, N. W.; Graves, D. J. J. Agric. Food Chem. 2004, 52, 65−70. (17) Ottaviani, J. I.; Actis-Goretta, L.; Villordo, J. J.; Fraga, C. G. Biochimie 2006, 88, 359−365. (18) Panickar, K. S.; Qin, B.; Anderson, R. A. Nutr. Neurosci. 2015, 18, 297−306. (19) Buijsse, B.; Feskens, E. J. M.; Kok, F. J.; Kromhout, D. Arch. Intern. Med. 2006, 166, 411−417. (20) Murphy, K. J.; Chronopoulos, A. K.; Singh, I.; Francis, M. A.; Moriarty, H.; Pike, M. J.; Turner, A. H.; Mann, N. J.; Sinclair, A. J. Am. J. Clin. Nutr. 2003, 77, 1466−1473. (21) Cala, O.; Pinaud, N.; Simon, C.; Fouquet, E.; Laguerre, M.; Dufourc, E. J.; Pianet, I. FASEB J. 2010, 24, 4281−4290. (22) American Peanut Council, 2016, http://www.peanutsusa.com/, accessed Jan 10, 2016. (23) Karchesy, J. J.; Hemingway, R. W. J. Agric. Food Chem. 1986, 34, 966−970. (24) Blade, C.; Arola, L.; Salvado, M.-J. Mol. Nutr. Food Res. 2010, 54, 37−59. (25) Lewis, W. E.; Harris, G. K.; Sanders, T. H.; White, B. L.; Dean, L. L. Food Nutr. Sci. 2013, 4, 22−32. (26) Takano, F.; Takata, T.; Yoshihara, A.; Nakamura, Y.; Arima, Y.; Ohta, T. Biol. Pharm. Bull. 2007, 30, 922−927. (27) Appledoorn, M. M.; Vincken, J.-P.; Sanders, M.; Hollman, P. C. H.; Gruppen, H. J. Agric. Food Chem. 2009, 57, 6007−6013. (28) Zhang, H.; Yerigui; Yang, Y.; Ma, C. J. Agric. Food Chem. 2013, 61, 8814−8820. (29) Kamiya, K.; Watanabe, C.; Endang, H.; Umar, M.; Satake, T. Chem. Pharm. Bull. 2001, 49, 551−557. (30) Lou, H.; Yamazaki, Y.; Sasaki, T.; Uchida, M.; Tanaka, H.; Oka, S. Phytochemistry 1999, 51, 297−308. (31) Botha, J. J.; Ferreira, D.; Roux, D. G. J. Chem. Soc., Chem. Commun. 1978, 700−702. (32) Korver, O.; Wilkins, C. K. Tetrahedron 1971, 27, 5459−5465. (33) Balde, A. M.; de Bruyne, T.; Pieters, L.; Kolodziej, H.; Berghe, D. V.; Claeys, M.; Vlietinck, A. Phytochemistry 1995, 38, 719−723.

S Supporting Information *

■

Article

ACKNOWLEDGMENTS

The authors are grateful to Dr. M. Górecki and Prof. J. Frelek for ECD measurements. Part of these studies were conducted during a postdoctoral fellowship financed by the Polish National Science Centre (NCN) on the basis of decision no. DEC-2015/16/S/ST4/00466. We thank B. Paulk, Universal Blanchers, Sylvester, GA, USA, for generously providing peanut skins for this study, and Mr. Slawomir Gorecki for logistical help important for conducting the experiments. 425



Article

(34) Xu, X.; Xie, H.; Wang, Y.; Wei, X. J. Agric. Food Chem. 2010, 58, 11667−11672. (35) Slade, D.; Ferreira, D.; Marais, J. P. J. Phytochemistry 2005, 66, 2177−2215. (36) Shoji, T.; Mutsuga, M.; Nakamura, T.; Kanda, T.; Akiyama, H.; Goda, Y. J. Agric. Food Chem. 2003, 51, 3806−3813. (37) Dudek (Jamróz), M. K.; Kaźmierski, S.; Stefaniak, K.; Gliński, V. B.; Gliński, J. A. Org. Biomol. Chem. 2014, 12, 9837−9844. (38) Barrett, M. W.; Klyne, W.; Scopes, P. M.; Fletcher, A. C.; Porter, L. J.; Haslam, E. J. Chem. Soc., Perkin Trans. 1 1979, 1, 2375−2377. (39) Lin, L.-C.; Kuo, Y.-C.; Chou, C.-J. J. Nat. Prod. 2002, 65, 505− 508. (40) Coetzee, J.; Malan, E.; Ferreira, D. Tetrahedron 1999, 55, 9999− 10004. (41) Morimoto, S.; Nonaka, G.-I.; Nishioka, I. Chem. Pharm. Bull. 1988, 36, 33−38. (42) Furrer, J. Chem. Commun. 2010, 46, 3396−3398.

426


Trimeric and Tetrameric A-Type Procyanidins from Peanut Skins

Recommend Documents