The fungi of the genus Paraconiothyrium have been shown to produce different classes of secondary metabolites, including sesquiterpenoids, diterpenoids, isoprenoid decalins, furanones, polyketides, and dihydrocoumarins, of which some compounds displayed cytotoxic and antibacterial activities. In the present study, we report the isolation and structural elucidation of five new secondary metabolites and three known compounds, modiolides A (5) and B (6) [1] and 1-(2,5-dihydroxyphenyl)-2-buten-1-one (9) [2] (Fig. 1a) from the EtOAc extract of the fungal strain Paraconiothyrium sp. VK-13 along with their in vitro anti-inflammatory activity.

Fig. 1
figure 1

a Chemical structures of compounds 1−9; b the selected HMBC and COSY correlations of compounds 1−4, 7, and 8; c the selected NOESY correlations of compounds 1 and 3; d 1H NMR chemical shift differences of MTPA esters of compounds 1−4

The molecular formula of 1 was established as C12H16O5 by a sodium adduct ion [M+Na]+ at m/z 263.0890 in the HRESIMS. The 1H NMR spectrum of 1 contained signals for one cis- and one trans-configured double bond, two oxymethine protons, and two methyl groups. The 13C NMR and DEPT spectra showed 12 carbon signals, including two carbonyl carbons at δC 169.8 and 171.6, four sp2 methine carbons, three sp3 oxygenated methines, one sp3 methylene, and two methyl carbons (Table 1). Based on these observations, compound 1 was suggested to possess the ten-membered macrolide skeleton [1]. By HSQC and COSY analysis, the spin system ranging from H-2 to H3-10 was established (Fig. 1b). Comparison of the 13C NMR data of 1 with those of the reported ten-membered macrolide derivative, modiolide A revealed that the structures of both compounds are nearly identical, except for the additional presence of an acetyl group in 1 [1]. The location of the acetyl group at C-4 was established by an HMBC correlation from 5.81 (H-4) to 171.6 (Fig. 1b). In the NOESY spectrum, H-9 showed NOE correlation with H-7, suggesting that H-7 and H-9 have the same orientation (Fig. 1c). The opposite orientations of Ha-8 and H3-10 were deduced by NOE cross-peaks between these protons. Furthermore, H-7 showed an NOE correlation with H-5, while H-6 correlated with H-4, revealing that H-4 and H-7 have opposite orientations. NOE correlations of 1 were found to be in a good agreement with those of the reported analog, modiolide A, suggesting that both compounds have the same relative configuration [1]. Finally, the absolute configuration of 1 was determined using the modified Mosher’s method [3]. Calculation of Δδ values for protons neighboring C-7 led to identification of 7S configuration (Fig. 1d). Thus, the gross structure of 1 was established as (2Z,4R,5E,7S,9R)-4-O-acetyl-7-hydroxy-9-methyl-2,5-nonadien-9-olide, named modiolide D.

Table 1 1H and 13C NMR data for compounds 14

The molecular formula of 2, C12H16O5 was deduced by the HRESIMS: m/z 263.0892 [M+Na]+. Comparative analysis of the 1H and 13C NMR data of 2 with those of 1 suggested that the structures of these compounds are nearly identical, except for the different location of the acetyl functional group (Table 1). The location of the acetyl group at C-7 was deduced by an HMBC correlation from δH 5.22 (H-7) to δC 171.6 (Fig. 1b). NOESY correlations of 2 were shown to be similar with those of 1, suggesting that both compounds have the same relative configuration. The absolute configuration of C-4 was identified as R (Fig. 1d) by the modified Mosher’s method. Therefore, the structure of 2 was established as (2Z,4R,5E,7S,9R)-4-hydroxy-7-O-acetyl-9-methyl-2,5-nonadien-9-olide, named modiolide E.

The HRESIMS of 3 showed an ion peak [M+Na]+ at m/z 263.0893, corresponding with the molecular formula C12H16O5Na. Comparison of the 13C NMR values of 3 with those of 1 showed that the chemical shifts of the C-7, C-8, and C-9 of 3 are shifted to further upfield regions, suggesting that these compounds have different configurations at C-7 and C-9 positions (Table 1). Indeed, the absolute configuration of C-7 was determined to be R by the modified Mosher’s method (Fig. 1d). In the NOESY spectrum, NOE correlations of H-7 with H-5 and H-9 revealed that these protons are oriented to the β-face of the molecule (Fig. 1c). NOE correlations between H-6 and Ha-8 and between Ha-8 and H3-10 indicated that Ha-8 and H3-10 are both α-oriented. Thus, the structure of 3 was identified as (2Z,4R,5E,7R,9S)-4-O-acetyl-7-hydroxy-9-methyl-2,5-nonadien-9-olide, named modiolide F.

Compound 4 has the molecular formula C10H14O4 as deduced by the HRESIMS: m/z 221.0772 [M+Na]+. The 1H and 13C NMR spectroscopic data revealed that 4 is a ten-membered macrolide. Comparison of the 1H and 13C NMR data of 4 with those of the reported ten-membered macrolide, modiolide A resulted in the close similarity, except that the chemical shifts of C-7, C-8, and C-9 in 4 are more upfield shifted [1], suggesting that these compounds are diastereomers at the C-7 and C-9 positions. This suggestion was supported by comparing the carbon chemical shifts of C-7, C-8, and C-9 between 4 and 3 (Table 1). The relative configuration of 4 was shown to be the same as that of 3, as deduced by NOESY spectroscopic analysis. In addition, considering the chemical shift differences observed between 4 and modiolide A, the overall structure of 4 was suggested to be a diasteromer of modiolide A, having inverted configuration at C-7 and C-9 because the absolute configuration of C-4 was determined to be R by application of the modified Mosher’s method (Fig. 1d). Although the stereochemistry of 4 was shown to be similar with that of the previously reported macrolide, fusanolide B [4], the proton and carbon chemical shifts of C-7, C-8, and C-9 of these compounds were significantly different [4 (in CD3OD): δH 4.48 (H-7)/ δC 67.9 (C-7), δH 1.83 and 1.81 (H2-8)/ δC 41.3 (C-8), δH 5.54 (H-9)/ δC 67.6 (C-9) vs. fusanolide B (in CD3OD): δH 4.12 (H-7)/ δC 73.7 (C-7), δH 1.90 and 1.73 (H2-8)/ δC 44.8 (C-8), δH 5.25 (H-9)/ δC 71.0 (C-9)]. Furthermore, the 1H and 13C NMR data of the reported fusanolide B were identical with those of modiolide A, revealing that both compounds have the same configuration [1, 4]. This evidence allowed to revise the configuration of fusanolide B to be the same as that of modiolide A and conclude that 4 is a diastereomer of both fusanolide B and modiolide A. Thus, the structure of 4 was determined to be (2Z,4R,5E,7R,9S)-4,7-dihydroxy-9-methyl-2,5-nonadien-9-olide, named modiolide G.

Analysis of the NMR and HRESIMS (m/z 219.0633 [M+Na]+) data of 7 suggested that 7 is a phenyl-1-butanone derivative [5]. Assignment of all positions of the structure of 7 was done by comparison of the 1H and 13C NMR data of 7 with those of the reported analogs [5,6,7,8], as well as analysis of HSQC and HMBC spectra (Fig. 1b). The stereochemistry of C-3′ was proposed to be R by comparing the proton and carbon chemical shifts of C-3′ and the specific optical rotation of 7 with those of the reported compounds [8]. Thus, the structure of 7 was established as 1-(2,5-dihydroxyphenyl)-3-hydroxybutan-1-one.

Compound 8 possesses the molecular formula C11H14O4 as determined by the HRESIMS: m/z 233.0794 [M+Na]+. The 1H and 13C NMR data of 8 were found to be almost identical with those of 7, except for the addition of a methoxy group [δH 3.32 (s)/δC 56.7] in 8 (Table 2). The location of the methoxy group at C-3′ was deduced by an HMBC correlation observed from δH 3.32 to δC 74.9 (C-3′) in the HMBC spectrum (Fig. 1b). Therefore, the structure of 8 was identified as 1-(2,5-dihydroxyphenyl)-3-methoxy-butan-1-one. The absolute configuration 8 was proposed to be analogous to that of 7.

Table 2 1H and 13C NMR data for compounds 7 and 8

Excessive production of NO during acute and chronic inflammation is associated with the development of many inflammatory disorders and cancer [9,10,11,12]. Therefore, we initially evaluated the inhibitory effects of the isolated compounds on the NO overproduction in LPS-stimulated RAW264.7 cells at 20 µM [13]. The result showed that only 7 and 9 inhibit the nitrite production among the compounds tested (data not shown). Based on the screening data, 7 and 9 were chosen to examine their effects on overproduction of nitrite and PGE2 in LPS-stimulated RAW264.7 cells at different concentrations [14]. Firstly, to exclude the possibility of cytotoxicity, the viability of RAW264.7 cells was evaluated in the presence of 7 and 9 by MTT assay [15]. The result revealed that 7 and 9 are both nontoxic towards RAW264.7 cells at 20 µM. The cells were then pretreated for 12 h with different concentrations of 7 and 9, followed by treatment with LPS (1 μg/mL) for 18 h (see Supplementary information). Figure 2a showed that nitrite concentration in only LPS-treated group is increased nine-fold compared with that of the untreated group. However, pretreatment of the cells with 7 and 9 decreased nitrite production in a dose-dependent manner (Fig. 2a), with IC50 values of 12.5 and 3.9 µM, respectively (Table S1). Similarly, the production of PGE2 in LPS-stimulated RAW264.7 cells was suppressed by 7 and 9 (Fig. 2b), with IC50 values of 9.5 and 6.9 µM, respectively (Table S1).

Fig. 2
figure 2

Effects of compounds 7 and 9 on nitrite (a) and PGE2 (b) production and iNOS and COX-2 protein expression (c) in LPS-stimulated RAW264.7 macrophages. Data represent the mean values of three experiments (±SD). #p < 0.01 versus control groups; *p < 0.05 and **p < 0.01 compared to the group treated with LPS only

During the inflammatory process, the proinflammatory mediators NO and PGE2 are released by the activities of their inducible enzymes, iNOS and COX-2, respectively [16, 17]. Accordingly, downregulation of iNOS and COX-2 protein expression decreases the secretion of the proinflammatory and cytotoxic mediators [14, 18]. Thus, we next examined the effects of 7 and 9 on iNOS and COX-2 protein expression in LPS-stimulated RAW264.7 cells using western blot analysis. When treating with LPS, the expression of iNOS and COX-2 protein levels in the cells were shown to be remarkably upregulated compared with the untreated groups (Fig. 2c). However, 7 and 9 both attenuated LPS-induced iNOS and COX-2 expression in a dose-dependent manner, suggesting that 7 and 9 inhibit the NO and PGE2 production through downregulation of iNOS and COX-2 protein expression, respectively. Next, we investigated the effects of 7 and 9 on TNF-α, IL-1β, IL-6, and IL-12 mRNA expression in LPS-stimulated RAW264.7 macrophages using quantitative real-time reverse transcriptase-polymerase chain reaction (qRT-PCR). As the result, 9 inhibited TNF-α, IL-1β, IL-6, and IL-12 mRNA expression, with IC50 values of 4.2, 2.4, 2.6, and 3.3 µM, respectively; while 7 attenuated TNF-α, IL-1β, IL-6, and IL-12 mRNA expression, with IC50 values of 12.5, 11.7, 10.6, and 13.5 µM, respectively (Table S1). These results revealed that both 7 and 9 inhibit the proinflammatory cytokine gene expression at the transcriptional level.