Anacardic Acid Derivatives Isolated from Fungal Species Tyromyces fissilis as New Histone Acetyltransferase Inhibitors

Abstract

Anacardic acid (AA) was first detected in the shells of cashew nuts, Anacardium occidentale, and is known to possess inhibitory activity against acetyltransferases. Recently, several anacardic acid derivatives (AAds) were isolated from the wild fungus, Tyromyces fissilis, which has been reported as xanthine oxidase inhibitors. In the present study, we investigated whether nine AAds function as acetyltransferase inhibitors. Screening analyses were performed by incubating the enzyme protein (P300/CBP-associated factor; PCAF) and the substrate protein (histone H1) with radioisotope-marked acetyl-CoA, showing that two of the nine derivatives, namely, AAd7 and AAd11, inhibited the acetyltransferase activity of PCAF at concentrations of 50 and 100 µM, respectively. The inhibition intensities were similar to those of the original compound, AA, and the inhibitory effects of these derivatives increased in a concentration-dependent manner. Docking simulations suggested the possibility that AA, AAd7, and AAd11 might bind the same active pocket of PCAF. These results suggest that the AAds can be used as acetyltransferase inhibitors. In contrast, there were no significant differences in the molecular structure of AA and its derivatives; however, these small differences in the functional groups on the alkyl side chain on salicylic acid reduced the acetyltransferase inhibitor activity or newly produced proteolytic activity.

INTRODUCTION

The acetylation of proteins by histone acetyltransferases (HATs) plays a crucial role in modifying protein functions. These enzymes acetylate conserved lysine residues on proteins by transferring an acetyl group from acetyl-CoA to form ε-N-acetyl-lysine.¹⁾ Histones are major targets of HATs is histones, resulting in the modification of the chromatin structure and the epigenetic modulation of gene transcription machinery.¹⁾ Acetylation targets of HATs are histones as well as some types of non-histone proteins, and HATs modulate their functions, localizations and structures.²⁾ HATs are sorted into at least seven families; GCN5/PCAF, MYST, TAF_II250, CBP/p300, SRC, HAT1, and ATF-2 families; based on their structural features, functional roles and sequence conservation.³⁾ Of these HAT families, the GCN5/PCAF family, also referred to as the GNAT (Gcn5-related N-acetyltransferase) family, includes GCN5 and PCAF. These enzymes catalyze the transfer of an acetyl group from a donor molecule (acetyl CoA) to the lysine residues of target proteins.⁴⁾ Some target proteins are histones H2B, H3, and H4.

The HAT activities can be significantly antagonized by several types of natural products, such as anacardic acid (AA) from shell of cashew nut (Anacardium occidentale),⁵⁾ curcumin from turmeric (Curcuma longa),⁶⁾ garcinol from fruit of Garcinia indica,⁷⁾ plumbagin the plant genus Plumbago,⁸⁾ and epigallocatechin gallate from green tea (Camellia sinensis).⁹⁾ Of these inhibitors, interaction between AA and the PCAF bromodomain was simulated by in-silico analyses of the induced fit docking and the molecular dynamics simulation.^10,11) In a recent study, we showed that AA blocked (1) the interaction between influenza virus PB2 and acetyl-CoA,¹²⁾ (2) the acetylation of influenza virus nucleoprotein by PCAF,¹³⁾ and (3) the acetylation of histone H1 by the molluscan CBP.¹⁴⁾ In addition, several types of anacardic acid derivatives (AAds) have been isolated from natural products or synthesized, and they have been shown to possess biological, biochemical and medical activities.^15–20)

Collaborative researcher Mitomo et al. reported eight AAds from the fruiting bodies of Tyromyces fissilis based on their inhibitory activity against xanthine oxidase.²¹⁾ 2-Hydroxy-6-pentadecylbenzoic acid showed the highest inhibiting activity on xanthine oxidase among the isolated compounds. Therefore, we focused on AAds, which contain a salicylic acid moiety with an aliphatic group with one or more substituents at the C-6 position, isolated from the fungus T. fissilis. Six known compounds (AAd6, AAd7, AAd9, AAd10, AAd14, and AAd15)²¹⁾ and isolated three compounds, AAd4, AAd11, and AAd22 from other fractions of the same MeOH extract of fruiting bodies of T. fissilis. In this study, we have examined whether these nine derivatives including the new compounds, AAd11, and AAd22 inhibited the HAT activity of PCAF using a series of biochemical experiments. To our knowledge, this is the first report demonstrating the HAT inhibitory activity of AAds isolated from T. fissilis.

MATERIALS AND METHODS

Anacardic Acid Derivatives

Six AAds previously reported in the reference²¹⁾ by Mitomo et al.; AAd6 as “Compound 3,” AAd7 as “Compound 4,” AAd9 as “Compound 5,” AAd10 as “Compound 6,” AAd14 as “Compound 8,” and AAd15 as “Compound 7” were used in this assay. Three AAds, AAd4, AAd11, and AAd22 were isolated from the same MeOH extract of fruiting bodies of Tyromyces fissilis.²¹⁾

Isolation of AAd4, AAd11, and AAd22

The general experimental methods and procedures were as described in the previous paper.²¹⁾ A part of the EtOAc extract (4.8 g) obtained from the MeOH extract of fruiting bodies of T. fissilis²¹⁾ was subjected to silica gel column chromatography. Elution was performed using a stepwise mixture of n-hexane : EtOAc (100 : 0, 90 : 10, 80 : 20, 70 : 30, 60 : 40, 50 : 50, 40 : 60, 30 : 70, 20 : 80, 10 : 90, and 0 : 100, v/v; 400 mL each) and MeOH, resulting in 12 fractions (TFE1–TFE12). Fraction TFE3 (part of the 20% n-hexane/EtOAc eluate, 463 mg) was further subjected to octadecyl silica (ODS) column chromatography using a stepwise mixture of H₂O : acetonitrile (40 : 60, 35 : 65, 30 : 70, 25 : 75, 20 : 80, and 15 : 85, v/v; 40 mL ×8 fractions each) and MeOH as the eluent, resulting in 56 fractions (TFE3-1 to TFE3-60). Fraction TFE3-27 (a part of the 85% H₂O : acetonitrile eluate, 16 mg) was further subjected to preparative HPLC using an InertSustain C18 column (ϕ10 × 250 mm) with UV detector at 210 nm. The mobile phase was composed of H₂O-acetonitrile (15 : 85, v/v) and the flow rate of 5.0 mL/min was used to yield AAd4 (2.5 mg). Fraction TFE4 (a part of the 30% n-hexane/EtOAc eluate, 2.76 g) was further subjected to silica gel column chromatography with a stepwise mixture of n-hexane : EtOAc (100 : 0, 90 : 10, 80 : 20, 70 : 30, and 60 : 40, v/v; 24 mL ×10 fractions each) and MeOH (10 fractions) as the eluent, resulting in 60 fractions (TFE4-1 to TFE4-60). The supernatant (136 mg) of the 30% EtOAc eluate, TFE4-32,33 (457 mg) yielded by crystallizing with acetonitrile at −30 °C was further separated using the ODS column with H₂O : acetonitrile (40 : 60, 35 : 65, 30 : 70, 25 : 75, and 20 : 80, v/v; 4 mL ×10 fractions each) as the eluent to give crude AAd11 (79 mg). The crude AAd11 subsequently purified by preparative HPLC using InertSustain C18 column (ϕ10 × 250 mm) eluted with H₂O-acetonitrile (30 : 70) with UV detector at 210 nm. The mobile phase was composed of H₂O-acetonitrile (15 : 85, v/v) and the flow rate of 5.0 mL/min was used to give AAd11 (11.3 mg). Fraction TFE6 (a part of the 50% EtOAc/n-hexane eluate, 288 mg) was cooled to −30 °C precipitated in acetonitrile, and the precipitate was collected twice to give AAd22 (62 mg) as diastereomixture.

2-((8Z, 11Z)-Heptadeca-8,11-dien-1-yl)-6-hydroxybenzoic Acid (AAd4)

Colorless oil; [α]¹⁹_D −2.9° (c 0.06, MeOH); UV (EtOH) λ_max (log ε) 243 (3.79) and 310 (3.61) nm; IR (NaCl) ν_max 3500–2500, 3009, 2926, 2855, 1653, 1606, 1576, 1452, 1299, 1245, 1216, 1166, 1023, 979, 823, 738, and 708 cm⁻¹; high resolution (HR)-FAB-MS m/z 371.2582 [M − H]⁻ (calcd. for C₂₄H₃₅O₃, 371.2592); ¹H-NMR (500 MHz, CDCl₃) δ_H 0.88 (3H, t, J = 6.9 Hz, H-17′), 1.23–1.40 (14H, m, H-3′–6′, 14′–16′), 1.59 (2H, m, H-2′), 2.05 (4H, m, H-7′, 13′), 2.77 (2H, t, J = 6.8 Hz, H-10′), 2.96 (2H, dd, J = 7.7, 7.9 Hz, H-1′), 5.33 (2H, m, H-9′, 11′), 5.39 (2H, m, H-8′, 12′), 6.74 (1H, d, J = 7.5 Hz, H-5), 6.85 (1H, d, J = 8.3 Hz, H-3), 7.33 (1H, dd, J = 7.7, 8.1 Hz, H-4), and 11.44 (1H, m, 1-COOH); ¹³C-NMR (125 MHz, CDCl₃) δ_C 14.1 (C-17′), 22.6 (C-16′), 25.6 (C-10′), 27.2 (C-7′ or 13′), 27.2 (C-7′ or 13′), 29.3, 29.3, 29.4, 29.7, 29.8 (C-3′–9′), 31.5 (C-15′), 32.1 (C-2′), 36.5 (C-1′), 110.9 (C-1), 115.6 (C-3), 122.5 (C-5), 127.9 (C-9′ or 11′), 128.0 (C-9′ or 11′), 130.2 (C-8′ or 12′), 130.2 (C-8′ or 11′), 134.8 (C-4), 147.3 (C-6), 163.5 (C-2), and 174.3 (1-COOH). These spectral data were identical with those of 6-(8′Z, 11′Z-heptadecadienyl)-salicylic acid previous reported by Corthout et al.²²⁾

2-((E)-14-Oxoheptadeca-12-en-1-yl)-6-hydroxybenzoic Acid (AAd11)

Colorless oil, [α]¹⁹_D −2.9° (c 0.06, MeOH); UV (EtOH) λ_max (log ε): 210 (4.51), and 311 (3.54) nm; IR (NaCl) ν_max 3500–2500, 2926, 2854, 1662, 1607, 1577, 1452, 1297, 1243, 1210, 1166, 1119, 979, 823, 738, and 708 cm⁻¹; HR-FAB-MS m/z 387.2538 [M − H]⁻ (calcd. for C₂₄H₃₅O₄, 387.2541); ¹H-NMR (500 MHz, CDCl₃): δ_H 0.94 (3H, t, J = 7.4 Hz, H-17′), 1.23–1.40 (14H, m, H-3′–9′), 1.46 (2H, m, H-10′), 1.60 (2H, m, H-2′), 1.65 (2H, m, H-16′), 2.21 (2H, dt, J = 7.1, 7.2 Hz, H-11′), 2.54 (2H, t, J = 7.4 Hz, H-15′), 2.98 (2H, dd, J = 7.7, 7.9 Hz, H-1′), 6.11 (1H, d, J = 15.8 Hz, H-13′), 6.77 (1H, d, J = 7.5 Hz, H-5), 6.86 (1H, dt, J = 6.9, 15.3 Hz, H-12′), 6.86 (1H, d, J = 9.1 Hz, H-3), 7.35 (1H, t, J = 7.9 Hz, H-4), ¹³C-NMR (125 MHz, CDCl₃): δ_C 13.8 (C-17′), 17.8 (C-16′), 27.9 (C-10′), 29.0, 29.2, 29.3, 29.4, 29.4, 29.5, 29.8 (C-3′–9′), 32.1 (C-2′), 32.4 (C-11′), 36.5 (C-1′), 41.8 (C-15′), 110.7 (C-1), 115.7 (C-3), 122.6 (C-5), 130.3 (C-13′), 135.1 (C-4), 147.6 (C-12′), 148.1 (C-6), 163.5 (C-2), 175.0 (1-COOH) and 201.8 (C-14′). The spectrum data for AAd11 in CDCl₃ obtained by ¹H-NMR, ¹³C-NMR, ¹H–¹H correlation spectroscopy (COSY), heteronuclear multiple quantum coherence (HMQC) and heteronuclear multiple bond connectivity (HMBC) were shown in the supplementary materials (Supplementary Figs. S1–S5).

2-(12,14-Dihydroxyheptadecanyl)-6-hydroxybenzoic Acid (AAd22)

Colorless paste, [α]²¹_D −0.8° (c 0.59, MeOH), IR (film) ν_max 3504, 3350, 2931, 1667, 1604, 1455, 1266, 1095, 1030, 957, 824, 741, 590, and 551 cm⁻¹, HR-FAB-MS m/z 407.2789 [M−H]⁻ (calcd. for C₂₄H₃₉O₅, 407.2803); ¹H-NMR (pyridine-d₅): δ_H 0.95 (3H, t, J = 7.0 Hz), 1.2–1.5 (14H, m), 1.6–1.8 (10H, m), 1.77(1H, m), 1.90 (1H, m), 3.27 (2H, t, J = 7.6 Hz), 4.18 (2H, m), 6.91 (1H, d, J = 7.5 Hz), 7.14 (1H, d, J = 8.2 Hz), and 7.39 (1H, t, J = 7.8 Hz), ¹³C-NMR (pyridine-d₅): δ_C 14.4 (C-17′), 19.1 (C-16′), 26.0, 29.8, 29.9, 29.9, 29.9, 30.0 30.1, 30.2, 32.5, 36.1, 38.9, 41.0, 44.7, 71.4 (C-12′ or 14′), 71.7, (C-14′ or 12′), 115.4 (C-1), 116.8 (C-3), 121.8 (C-5), 132.9 (C-4), 146.2 (C-6), 162.3 (C-2), and 174.8 (1-COOH). The spectrum data for AAd22 in Pyridine-d₅ obtained by ¹H-NMR and ¹³C-NMR were shown in the supplementary materials (Supplementary Figs. S6, S7).

In-Vitro HAT Assays

The procedures used in this study were modified from previous reports.^13,14,23) Briefly, the histone H1 purified from calf thymus (Sigma-Aldrich, St. Louis, MO, U.S.A.; 1 µg) were incubated with the partial recombinant protein of PCAF HAT domain (1 µg), and 7.4 kBq of [¹⁴C]-acetyl-CoA (PerkinElmer, Inc., Waltham, MA, U.S.A.) at 37 °C for 2 h in buffer containing 10 µM sodium butyrate. The reaction products were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The gels were stained with Coomassie brilliant blue (CBB), dried on piece of filter papers, and exposed to imaging plates (FUJIFILM, Tokyo, Japan) for several days. The signals were detected using a fluoro image analyzer. The intensities corresponding to the acetylation levels of histone H1 were quantified using ImageJ software (https://imagej.nih.gov/ij/index.html).

In Silico Docking Simulation Analyses

Various AAds conformations were generated using a stochastic search program. The X-ray crystallographic structure of the endonuclease domain of PCAF (Protein Data Bank (PDB) ID: 4NSQ) was obtained from the PDB. This enzyme was prepared for docking studies in which (i) hydrogen atoms were added to the structure with a standard geometry using Protonate 3D program; and (ii) Molecular Operating Environment (MOE) Alpha Site Finder was used for active site searches within the enzyme structure and dummy atoms were created from the obtained alpha spheres. Docking simulations between AAds and the enzyme, PCAF (PDB ID code: 4NSQ) were performed using the MOE (Chemical Computing Group, QC, Canada).

RESULTS AND DISCUSSION

The structures of the AAds, six known compounds (AAd6, AAd7, AAd9, AAd10, AAd14, and AAd15)²¹⁾ and isolated three compounds, AAd4, AAd11, and AAd22 from the same MeOH extract of fruiting bodies of T. fissilis, prepared in this experiment are shown in Fig. 1.

Fig. 1. Structures of AA and Nine AAds

Structure Determination of AAd11 and AAd22

The negative HR-FAB-MS of AAd11 showed a deprotonated molecular ion at m/z 387.2538 [M−H]⁻ (calcd. 387.2541), indicating that the molecular formula is C₂₄H₃₆O₄ with five unsaturations. The ¹H- and ¹³C-NMR spectral data of compound AAd11 were in good agreement with those of the previously reported 2-hydroxy-6-(14-oxoheptadec-1-yl)-benzoic acid²¹⁾ except for the presence of one double bond [δ_C 130.3 and 147.6, and δ_H 6.11 (1H, d, J = 15.7 Hz) and 6.86 (1H, dt, J = 6.9, 15.7 Hz)] determined E-configuration by the coupling constants and the chemical shifts. The position of the carbonyl group was confirmed to be the fourth carbon from the end of the alkyl side chain based on the combined analysis of ¹H–¹H COSY and HMBC spectral data. Additionally, HMBC correlations observed from olefinic protons at δ_H 6.11 and 6.86 to carbonyl carbon at δ_C 201.8 and the second methylene proton from the end of the alkyl side chain at δ_H 1.60 to the same carbonyl carbon. These observations indicate that AAd11 has 14(E)-oxoheptadeca-12-en-1-yl side chain at C-2 on the salicylic acid. Thus, AAd11 was determined to be as 2-hydroxy 6-[(E)-14-oxoheptadeca-12-en-1-yl]-benzoic acid (AAd11).

The negative HR-FAB-MS of AAd22 showed a deprotonated molecular ion at m/z 407 [M−H]⁻, thus we estimated that the molecular formula was C₂₄H₄₀O₅ with four unsaturations. The ¹H- and ¹³C-NMR spectral data of compound AAd22 were in good agreement with those of the previously isolated 2-hydroxy-6-(12,14-dihydroxyheptadec-1-yl)-benzoic acid (unpublished compound) determined by Mr. Mizuki Nomura in his Master’s thesis at the Graduate School of Agriculture, Shinshu University.²⁴⁾ Therefore, the structure of AAd22 was determined as shown in Fig. 1. Note that the stereochemistry of two hydroxyl groups at C-12′ and 14′ of AAd22 is undetermined.

Screening of AAds as HAT Inhibitors

We performed the in-vitro HAT assays to screen AA and seven AAds described in the cited literature²¹⁾ isolated from the MeOH extract of T. fissilis and two newly obtained compounds AAd11 and AAd22 for HAT inhibitors (Fig. 2). The final concentrations of the AAds with different functional groups on the alkyl side chain of salicylic acid were determined at 50 and 100 µM, respectively, because we previously reported that AA inhibited the acetylation of the influenza virus nucleoprotein (NP) by PCAF at 50 and 100 µM.¹³⁾ In this assay, AA inhibited the acetylation of histone H1 as well as the influenza virus NP at 50 and 100 µM as previously reported (Figs. 2A, B, lower panels; n = 3).¹³⁾ The AAd7 and AAd11 blocked acetylation of histone H1 more than 50 µM (Fig. 2A), and AAd4 and AAd14 also inhibited at 100 µM (Figs. 2A, B). We especially focused on AAd7 and AAd11, which could inhibit acetylation at 50 µM. In the case of AAd7, there were no changes in the band intensity of the CBB staining for either histone H1 or PCAF (Fig. 2A, upper panel). In contrast, incubation with AAd11 decreased the band intensity of the CBB staining for both histone H1 and PCAF. These results suggest that AAd7 and AAd11 could block the acetylation activity of histone H1.

Fig. 2. Results of in-Vitro Acetylation Assays Using [¹⁴C]-Acetyl-CoA

Assays were performed with nine AAds divided into two groups of seven AAds each; (A) AAd4, 6, 7, 9, 10, and 11 were allocated to the 1st group, and (B) AAd14, 16 and 22 were allocated to the 2nd group. The upper and lower panels show the proteins stained with CBB and the acetylated proteins visualized using autoradiography, respectively. Concentrations of AA and AAds were 50 and 100 µM, respectively.

Concentration Dependence of Inhibition

We investigated the concentration-dependent inhibition by AA, AAd7, and AAd11 using in in-vitro HAT assays (Fig. 3; n = 3). The maximum concentration was determined to be 50 µM. CBB staining showed that the band intensity of histone H1 was similar at all concentrations of AA and AAd7 (Figs. 3A, B, upper panels). The band intensity was measured using the image processing and analysis software ImageJ, suggesting that the histone H1 recombinant protein was stable [p = 0.99 and 0.75, respectively for AA (Fig. 3D) and AAd7 (Fig. 3E), calculated by one-way ANOVA]. However, the intensity of CBB staining gradually decreased with increasing concentration of AAd11 (Fig. 3F; p < 0.05 calculated by one-way ANOVA as shown in lower panel of Fig. 3C), and there was a significant difference between 0 and 50 µM (* p < 0.05, Tukey’s post hoc test; n = 3). These results suggest the possibility that AAd11 possesses proteolytic activity. However, cases of chemical degradation are rare and must be examined carefully.²⁵⁾ Most protein degradation is dependent on proteases. Although no proteins with protease activity were mixed into the experimental samples in this study, PCAF was prepared by us using E. coli in the laboratory, and trace amounts of bacterial proteases may have been contaminated during the expression and purification process. Further experiments are required to prove the possibility of proteolytic activity of AAd11.

Fig. 3. Concentration-Dependent Inhibition of AA, AAd7, and AAd11

In-vitro acetylation assays were performed with gradually diluted inhibitors (0, 12, 18.5, 25, 37.5, and 50 µM). Left, middle and right panels indicate AA, AAd7 and AAd11, respectively. Upper and lower panels of (A), (B), and (C) show the results of CBB staining and autoradiography, respectively. (D–F) Statistical analyses of band intensity of CBB staining of histone H1 (n = 3; * p < 0.05). (G–I) Statistical analyses of acetylation levels of histone H1 investigated using autoradiography (n = 3; * p < 0.05, ** p < 0.01).

We observed changes in the signal intensity of acetylation with increasing concentrations of AA, AAd7, and AAd11 by autoradiography (Figs. 3A–C, lower panels) and quantified the signal intensity using the image processing and analysis software, ImageJ (Figs. 3G–I). Interestingly, 12 µM of AA, AAd7, and AAd11 tended to increase the acetylation levels, and especially in AAd11, there was significant difference between 0 and 12 µM (Figs. 3G–I). To investigate the effect of low AA concentrations of histone H1 acetylation, similar experiments were performed using AA concentrations of 0, 3, 6, 9, and 12 µM. The results showed no change from 0 to 9 µM; however, at 12 µM, the acetylation level tended to increase, although no significant difference was observed (n = 3; Supplementary Fig. S8). We previously reported that the acetylation level of influenza virus NP increased at low concentration (5 and 10 µM) of AA.¹³⁾ At concentrations above 12 µM, the acetylation level of histone H1 gradually decreased with increasing concentration (n = 3; Figs. 3G–I). These results suggest that AA, AAd7, and AAd11 inhibit the acetyltransferase activity of PCAF. However, there were no significant differences between acetylation levels in the presence and absence of AA, AAd7, or AAd11 (n = 3; Figs. 3G–I). The reason for these interesting results may be resolved by three-dimensional structural analysis of the complexes. We did not perform a three-dimensional structural analysis of the complex between PCAF and the AA of AAds; therefore, and we cannot make any definitive statements here.

The structural differences between AA and AAd11 include (1) a longer hydrocarbon chain than AA, (2) an α, β-unsaturated carbonyl group in the alkyl side chain (Fig. 1). Although other derivatives also possess one of the similar structural properties. It might be important to have all these structural characteristics to express this activity.

Docking Simulation of AA, AAd7, and AAd11 with PCAF Acetyltransferase Domain

First, we focused on AA and AAd7, which have been suggested to inhibit the acetyltransferase activity of PCAF, and predicted the molecular mechanisms of inhibition using a docking simulation (Fig. 4). We used a partial tertiary structure of the PCAF acetyltransferase domain (PDB ID: 4NSQ). The structural differences between AA and AAd7 are (1) the length of the hydrocarbon chain and (2) the addition of a hydroxyl group on the alkyl side chain (Fig. 1). The docking simulations suggested that AA and AAd7 share the same binding pockets as the PCAF active sites (Figs. 4A, B), suggesting that AA and AAd7 inhibit enzyme activity through the same mechanism.

Fig. 4. Docking Simulation of AA, AAd7 and AAd11 with PCAF Acetyltransferase Domain

3-D and 2-D views showing interaction between PCAF active sites and AA (A), AAd7 (B), and AAd11 (C). AA and AAds share the same binding pockets. In 2-D views, navy shades around carbon atoms in AA, AAd7, and AAd11 indicated hydrophobic interaction. Green and pink circles indicate hydrophobic and hydrophilic amino acid residues, respectively. Pink circles with red or blue borders indicate acidic and basic amino acid residues, respectively.

The structural differences between AA and AAd7, such as the length of the hydrocarbon chain and presence or absence of a hydroxyl group, were very small (Fig. 1). Furthermore, it was suggested that AA and AAd7 bind to the same pocket on the PCAF surface, and 13 amino acids were shared between the two inhibitors (Figs. 4A, B). However, the parts of AA and AAd7 used for interaction with PCAF were different in that they were a hydrocarbon chain and a benzene ring, respectively (Figs. 4A, B, right panel). To elucidate the detailed mechanism of acetyltransferase activity inhibition, 3-D structural analysis of this complex is necessary. We performed a docking simulation between AAd11 and PCAF (Fig. 4C). These results suggest that AAd11 binds to the same pocket as AA and AAd7, and that the types of amino acid residues that are the source of the interaction are almost the same. It is necessary to perform more detailed experiments such as 3-D structure analysis to determine the actual tertiary structure of the complexes.

In Conclusion, even with minor differences in the molecular structures of AA and its derivatives, our results clearly demonstrated that these structural differences reduced the acetyltransferase inhibitory activity or the newly generated proteolytic activity. New AAds with unknown functions were also identified.

Acknowledgments

This work was partially supported by grants from the Japan Society for the Promotion of Science (JSPS), Grants-in-Aid for Scientific Research (C) (17K08867 and 21K07058), Takeda Science Foundation, Japan Foundation for Applied Enzymology, Waksman Foundation of Japan, and Tokushima Bunri University for Educational Reform and Collaborative Research (TBU2022-2-1) to D.H. This work was partially supported by Grants from the Japan Society for the Promotion of Science (JSPS), Grants-in-Aid for Scientific Research (C) (15H01751) to T.F., and the Joint Usage and Joint Research Programs of the Institute of Advanced Medical Sciences of Tokushima University to T.K.

Author Contributions

D.H., H.T., E.N., M.T., T.H., N.E., N.A. and Y.S. performed biochemical experiments; S.M., A.T. and T.F. purified anacardic acid derivatives; This manuscript was written by D.H. and edited by D.H., Y.S., A.T., T.F. and T.K.; D.H. and T.K. directed this research.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1) Drazic A, Myklebust LM, Ree R, Arnesen T. The world of protein acetylation. Biochim. Biophys. Acta. Proteins Proteom., 1864, 1372–1401 (2016).
• 2) Ononye OE, Downey M. Posttranslational regulation of the GCN5 and PCAF acetyltransferases. PLoS Genet., 18, e1010352 (2022).
• 3) Marmorstein R. Structure and function of histone acetyltransferases. Cell. Mol. Life Sci., 58, 693–703 (2001).
• 4) Lee KK, Workman JL. Histone acetyltransferase complexes: one size doesn’t fit all. Nat. Rev. Mol. Cell Biol., 8, 284–295 (2007).
• 5) Balasubramanyam K, Swaminathan V, Ranganathan A, Kundu TK. Small molecule modulators of histone acetyltransferase p300. J. Biol. Chem., 278, 19134–19140 (2003).
• 6) Balasubramanyam K, Varier RA, Altaf M, Swaminathan V, Siddappa NB, Ranga U, Kundu TK. Curcumin, a novel p300/CREB-binding protein-specific inhibitor of acetyltransferase, represses the acetylation of histone/nonhistone proteins and histone acetyltransferase-dependent chromatin transcription. J. Biol. Chem., 279, 51163–51171 (2004).
• 7) Balasubramanyam K, Altaf M, Varier RA, Swaminathan V, Ravindran A, Sadhale PP, Kundu TK. Polyisoprenylated benzophenone, garcinol, a natural histone acetyltransferase inhibitor, represses chromatin transcription and alters global gene expression. J. Biol. Chem., 279, 33716–33726 (2004).
• 8) Ravindra KC, Selvi BR, Arif M, Reddy BA, Thanuja GR, Agrawal S, Pradhan SK, Nagashayana N, Dasgupta D, Kundu TK. Inhibition of lysine acetyltransferase KAT3B/p300 activity by a naturally occurring hydroxynaphthoquinone, plumbagin. J. Biol. Chem., 284, 24453–24464 (2009).
• 9) Choi KC, Jung MG, Lee YH, Yoon JC, Kwon SH, Kang HB, Kim MJ, Cha JH, Kim YJ, Jun WJ, Lee JM, Yoon HG. Epigallocatechin-3-gallate, a histone acetyltransferase inhibitor, inhibits EBV-induced B lymphocyte transformation via suppression of RelA acetylation. Cancer Res., 69, 583–592 (2009).
• 10) Ghizzoni M, Boltjes A, Graaf C, Haisma HJ, Dekker FJ. Improved inhibition of the histone acetyltransferase PCAF by an anacardic acid derivative. Bioorg. Med. Chem., 18, 5826–5834 (2010).
• 11) Jaganathan R, Kumaradhas P. Binding mechanism of anacardic acid, carnosol and garcinol with PCAF: a comprehensive study using molecular docking and molecular dynamics simulations and binding free energy analysis. J. Cell. Biochem., 124, 731–742 (2023).
• 12) Hatakeyama D, Shoji M, Yamayoshi S, Hirota T, Nagae M, Yanagisawa S, Nakano M, Ohmi N, Noda T, Kawaoka Y, Kuzuhara T. A Novel functional site in the PB2 subunit of influenza A virus essential for acetyl-CoA interaction, RNA polymerase activity, and viral replication. J. Biol. Chem., 289, 24980–24994 (2014).
• 13) Hatakeyama D, Shoji M, Yamayoshi S, Yoh R, Ohmi N, Takenaka S, Saitoh A, Arakaki Y, Masuda A, Komatsu T, Nagano R, Nakano M, Noda T, Kawaoka Y, Kuzuhara T. Influenza A virus nucleoprotein is acetylated by histone acetyltransferases, PCAF and GCN5. J. Biol. Chem., 293, 7126–7138 (2018).
• 14) Hatakeyama D, Sunada H, Totani Y, Watanabe T, Felletár I, Fitchett A, Eravci M, Anagnostopoulou A, Miki R, Okada A, Abe N, Kuzuhara T, Kemenes I, Ito E, Kemenes G. Molecular and functional characterization of an evolutionarily conserved CREB-binding protein in the Lymnaea CNS. FASEB J., 36, e22593 (2022).
• 15) Chandregowda V, Kush A, Reddy GC. Synthesis of benzamide derivatives of anacardic acid and their cytotoxic activity. Eur. J. Med. Chem., 44, 2711–2719 (2009).
• 16) Zhao Q, Zhang X, Cai H, Zhang P, Kong D, Ge X, Du M, Liang R, Dong W. Anticancer effects of plant derived anacardic acid on human breast cancer MDA-MB-231 cells. Am. J. Transl. Res., 10, 2424–2434 (2018).
• 17) Zafar F, Gupta A, Thangavel K, Khatana K, Sani AA, Ghosal A, Tandon P, Nishat N. Physicochemical and pharmacokinetic analysis of anacardic acid derivatives. ACS Omega, 5, 6021–6030 (2020).
• 18) Umehara E, Costa Silva TA, Mendes VM, Guadagnin EC, Sartorelli P, Tempone AG, Lago JHG. Differential lethal action of C17:2 and C17:0 anacardic acid derivatives in Trypanosoma cruzi—a mechanistic study. Bioorg. Chem., 102, 104068 (2020).
• 19) Saedtler M, Förtig N, Ohlsen K, Faber F, Masota N, Kowalick K, Holzgrabe U, Meinel L. Antibacterial anacardic acid derivatives. ACS Infect. Dis., 6, 1674–1685 (2020).
• 20) Ohta S, Takeda M, Ohta E, Nehira T, Ômura H, Uy MM, Ishihara Y. Janohigenins: long-chain anacardic acid derivatives with neuroprotective activity from Ophiopogon japonicus seeds. Phytochemistry, 191, 112904 (2021).
• 21) Mitomo S, Hirota M, Fujita T. New xanthine oxidase inhibitors from the fruiting bodies of Tyromyces fissilis. Biosci. Biotechnol. Biochem., 83, 813–823 (2019).
• 22) Corthout J, Pieters L, Claeys M, Geerts St, Vanden Berghe D, Vlietinck A. Antibacterial and molluscicidal phenolic acids from Spondias mombin. Planta Med., 60, 460–463 (1994).
• 23) Hatakeyama D, Shoji M, Ogata S, Masuda T, Nakano M, Komatsu T, Saitoh A, Makiyama K, Tsuneishi H, Miyatake A, Takahira M, Nishikawa E, Ohkubo A, Noda T, Kawaoka Y, Ohtsuki S, Kuzuhara T. Acetylation of the influenza A virus polymerase subunit PA in the N-terminal domain positively regulates its endonuclease activity. FEBS J., 289, 231–245 (2022).
• 24) Nomura M. Master’s theses, Graduate School of Agriculture, Shinshu University, Nagano, Japan, 2014.
• 25) Kim K, Rhee SG, Stadtman ER. Nonenzymatic cleavage of proteins by reactive oxygen species generated by dithiothreitol and iron. J. Biol. Chem., 260, 15394–15397 (1985).