Abietane-Type Diterpenoids from the Resin of Pinus yunnanensis and Their Potential Anti-Renal Fibrosis Activities

TL;DR

This study identifies new diterpenoids from pine resin that may help treat kidney disease by blocking harmful cell signals.

Contribution

Discovery of seven new abietane-type diterpenoids with anti-renal fibrosis potential and insights into their mechanism of action.

Findings

Seven new abietane-type diterpenoids (pinusyunins A–G) and three known analogues were isolated from Pinus yunnanensis resin.

The compounds inhibit collagen I, fibronectin, and α-SMA expression in TGF-β1-induced kidney cells.

Compound 10 showed superior α-SMA inhibition in NRK-52E cells and specificity in blocking p-Smad3 phosphorylation.

Abstract

Chronic kidney disease (CKD) has emerged as a pressing global public health concern, making the identification of renal fibrosis inhibitors a key research focus. In this study, seven undescribed abietane-type diterpenoids, pinusyunins A–G (1, 2, 4, and 7–10) and three known analogues (3, 5, and 6), were isolated from Pinus yunnanensis resin, which were identified by spectroscopic analyses and quantum computational chemistry methods. Biological evaluation showed that all the isolates exhibited inhibitory activity against the expression of collagen I, fibronectin, and α-SMA in transforming growth factor-β1 (TGF-β1)-induced NRK-52E and NRK-49F cells. Specifically, compounds 1–10 reduced the expression of α-SMA at 40 μM in both cell lines, while compounds 6–8 and 10 decreased the expression of these three markers at 40 μM in both cell lines with the potency of compound 10 superior to the others in α-SMA inhibition in NRK-52E cells. Variations in activity are associated with differences in substituents at the C-13 position. Further studies demonstrated that these abietane-type diterpenoids block the TGF-β/Smad signaling pathway by inhibiting the phosphorylation of Smad2/3. In particular, compounds 1, 3, 6, and 7 suppressed only p-Smad3 other than p-Smad2, indicating their specificity. The research on these abietane-type diterpenoids provides novel candidate molecules and a scientific underpinning for developing anti-renal fibrosis drugs.

Figures

Full text

1. Introduction

Pine resin constitutes the principal secondary metabolite of Pinaceae plants, comprising turpentine and rosin. Turpentine is chiefly composed of monoterpenes and sesquiterpenes, while rosin is primarily composed of diterpenes [1,2]. Diterpenes predominantly exist in the form of diterpene resin acids, chiefly comprising pimaric acid, sandarapimaric acid, isopimaric acid, abietic acid, neoabietic acid, and dehydroabietic acid, constituting the primary components of pine resin [3]. Pine resin, as a traditional Chinese medicinal substance, possesses multiple pharmacological effects, including antibacterial [4], anti-inflammatory [5], antitumor [6], and analgesic [7] properties.

Chronic kidney disease (CKD) has emerged as a pressing global public health concern, affecting 10–14% of the world’s population with a persistently rising prevalence [8]. Renal fibrosis represents a critical pathological stage in the progression of CKD towards end-stage renal disease [9]. TGF-β1 plays a critical role in the development of fibrosis. It drives the progression of fibrosis by promoting excessive extracellular matrix (ECM) deposition, which leads to the destruction of renal tissue structure and impairment of renal function through mechanisms including the induction of epithelial-mesenchymal transition (EMT) and fibroblast-to-myofibroblast transition (FMT) [10,11]. Currently, known anti-fibrotic agents such as pirfenidone and nintedanib are primarily synthetic compounds, which usually contain nitrogen-containing heterocyclic structures but are limited by insufficient efficacy, single-target mechanisms, and severe side effects [12]. Therefore, the development of novel anti-fibrotic drugs is urgently needed.

Natural products are increasingly reported to exhibit anti-fibrotic activities, such as the diterpenoid triptolide [13]. Thus, investigating the chemical constituents and biological activities of diterpenoids from P. yunnanensis resin is warranted. In this study, seven undescribed abietane-type diterpenoids, named pinusyunins A–G (1, 2, 4, and 7–10), and three known analogues (3, 5, and 6) (Figure 1) were isolated from P. yunnanensis resin and identified by spectroscopic analyses and quantum computational chemistry methods. Their potential anti-renal fibrosis activities were evaluated in in vitro cellular assays. The TGF-β1-induced renal tubular epithelial cells (e.g., NRK-52E, HK-2) and renal fibroblasts (e.g., NRK-49F, HRF) can, respectively, mimic fibrosis mediated by EMT and FMT, both of which are widely used in renal fibrosis research [14,15]. The combined use of both cell types enhances the reliability of biological evaluation. Given the rapid growth and low culture cost of rat-derived NRK-52E and NRK-49F cells, this study employed TGF-β1-induced NRK-52E and NRK-49F cells to screen for potential anti-renal fibrosis effects, with GW788388 (GW), a well-characterized TGF-βRI inhibitor that blocks TGF-β1-mediated fibrotic signaling, used as a positive control [16].

2. Results and Discussion

2.1. Structure Elucidation of the Compounds

Pinusyunin A (1), obtained as a white powder, possesses the molecular formula of C_22_H_32_O_4_, as suggested by analysis of its HRESIMS (m/z 383.2180 [M + Na]^+^, calculated for C_22_H_32_O_4_Na, 383.2193), ^13^C NMR, and DEPT spectra data. The ^1^H NMR spectrum (Table 1) of 1 contains a typical ABX spin system [δH 7.29 (1H, d, J = 8.2 Hz, H-11), δH 7.30 (1H, dd, J = 8.2, 1.8 Hz, H-12), δH 7.27 (1H, d, J = 1.8 Hz, H-14)], two methoxy groups (δH 3.42, s; 3.05, s), and four methyl singlets. The ^13^C NMR and DEPT spectra (Table 1) show 22 signals attributed to six methyls (two methoxy groups), four sp^3^ methylenes, five methines (three sp^2^ and two sp^3^ with one oxygenated), and seven non-protonated carbons (including three sp^3^ with one oxygenated, one ester carbonyl, and three aromatic carbons). The 1D and 2D NMR data show a high similarity to those of 7α,15-dihydroxydeydroabietic acid [17] with the exception that the hydroxy groups at C-7 and C-15 are replaced by the methoxy groups. The HMBC correlations (Figure 2) of H_3_-21 (δH 3.42)/C-7 (δC 78.6) and H_3_-22 (δH 3.05)/C-15 (δC 78.1) supported the differences between 1 and 7α,15-dihydroxydeydroabietic acid.

The 1D selective gradient NOESY spectra (Figure 3) and NMR calculation methods enabled the determination of the relative configuration of 1. After irradiating H_3_-20 and H-5, the correlations of H_3_-20/H_3_-19, Ha-1 and H-5/Hb-1 were observed, implying that H_3_-20, H_3_-19 and Ha-1 are on the same face of the bicyclic ring, while H-5 and Hb-1 are on the other side. However, it is difficult to determine the relative configuration of C-7 using NOESY data. Therefore, the NMR chemical shifts of (4R**,5R**,7S**,10S**)-1 (1A) and (4R**,5R**,7R**,10S**)-1 (1B) (Figure S57) were calculated using the PCM/mPW1PW91/6-31G(d,p) methods. The results revealed that the data for 1B (Table S1) exhibited the highest probability (100%) via DP4+ analysis, indicating the relative configuration of 1 as 4R**,5R**,7R**,10S**. Subsequently, 1’s absolute configuration was determined through electronic circular dichroism (ECD) calculations. The results showed that the calculated ECD spectrum of (4R,5R,7R,10S)-1 (Figure 4) exhibits high agreement with that of the experiment, confirming the absolute configuration of 1 as 4R,5R,7R,10S. In this way, the structure of 1 was established and named pinusyunin A.

Pinusyunin B (2) was obtained as a white powder. Its molecular formula was determined as C_21_H_30_O_4_, indicating seven unsaturations, by analysis of its HRESIMS ion peak at m/z 369.2032 [M + Na]^+^ (calculated for C_21_H_30_O_4_Na, 369.2036), ^13^C NMR, and DEPT spectra. After comparing the 1D and 2D NMR data (Table 1) of 2 with those of inumakiol A [18], it was found that 2 is a structural analogue of inumakiol A. The difference between 2 and inumakiol A is one of the hydrogens at C-7 being replaced by a methoxy group in 2. The HMBC correlation (Figure 2) of H_3_-21 (δH 3.38)/C-7 (δC 78.4) confirmed the aforementioned conclusion. The 1D selective gradient NOESY correlations between H_3_-19/H_3_-20, Ha-1/H_3_-20, and Hb-1/H-5 by irradiating H_3_-19, Ha-1, and Hb-1 indicated that these two methyl groups are oriented in the same direction and opposite to H-5. For the relative configuration at C-7, NMR chemical shift calculations were performed. The results showed that the calculated data for (4R**,5R**,7R**,10S**)-2 (2B) exhibited high agreement with the experimental data, and DP4+ analysis (Table S2) indicated the highest confidence level (100%). These data enabled the inference that the relative configuration of 2 (Figure S59) was 4R**,5R**,7R**,10S**. Similar to compound 1, its absolute configuration was determined through ECD calculations. The results showed that the calculated ECD spectrum of (4R,5R,7R,10S)-2 (Figure 4) matches well with the experimental one, confirming the absolute configuration of 2 as 4R,5R,7R,10S and named pinusyunin B accordingly.

Pinusyunin C (4) was isolated as a white powder. Its molecular formula was deduced as C_21_H_32_O_4_ by HRESIMS ion peak at m/z 371.2187 [M + Na]^+^ (calculated for C_21_H_32_O_4_Na, 371.2193), ^13^C NMR, and DEPT spectra, indicating six degrees of unsaturation. Analyses of the 1D and 2D NMR spectra (Table 2) and comparison with the literature suggested that compound 4 possesses a similar structure to bieta-7,13-diene-12amethoxy-18-oic acid [19]. The difference lies in the presence of an additional OH substituent at the C-3 in compound 4, which was supported by the ^1^H-^1^H COSY correlation of H-3/H_2_-2 and the HMBC correlations (Figure 2) of H_3_-19 (δH 1.19)/C-3 (δC 76.4) and H-3 (δH 4.02)/C-1 (δC 38.2). The relative configuration of 4 was deduced from the 1D selective gradient NOESY spectra and NMR calculations. The selective NOESY correlations by irradiating H_3_-20 show H_3_-20/Ha-1, H_3_-19, and Hb-11, indicating that these protons are at the same orientation and opposite to H-9. In addition, the correlations of H-3/Hb-1 and H-5 by the irradiation of H-3 indicated that H-3, Hb-1, and H-5 are at the same orientation. The relative configuration at the C-12 of 4 was determined using the above-mentioned NMR calculations. The results indicated that the relative configuration of 4 was 3S**,4S**,5R**,9R**,10R**,12S**. Furthermore, the absolute configuration of 4 was determined as 3S,4S,5R,9R,10R,12S by ECD calculations (Figure 4). The structure of 4 was definitively determined and named as pinusyunin C.

Pinusyunin D (7) was acquired as yellowish gums and found to have the molecular formula C_20_H_28_O_4_, indicating seven degrees of unsaturation derived from its HRESIMS data at m/z 355.1870 [M + Na]^+^ (calculated for C_20_H_28_O_4_Na, 355.1880), ^13^C NMR, and DEPT data. Careful comparison of the NMR data (Table 2) of 7 with those of 14-oxoabieta-7,12-dien-18-oic acid [20] indicated that they are similar. The only difference lies is that the presence of an additional OH substituent at the C-15 in 7, which was confirmed by the HMBC correlations (Figure 2) of H_3_-16/C-15 (δC 72.9), H_3_-17/C-15, and H-12/C-15. The selective NOESY correlates by irradiating H_3_-20 (Figure 3) of H_3_-20/Ha-1, Hb-11, and H_3_-19, suggesting 10-CH_3_ and 4-CH_3_ are at the same orientation. The same methods, by irradiating Hb-1, revealed the correlations of Hb-1/H-5 and H-9, indicating that H-5 and H-9 are on the same side. The absolute configuration of 7 was identified as 4R,5R,9R,10R according to ECD calculations by comparison with their accordance (Figure 4).

Pinusyunin E (8), collected as yellowish gums, has a molecular formula of C_20_H_28_O_5_ as deduced from the HRESIMS ion at m/z 371.1823 [M + Na]^+^ (calculated for C_20_H_28_O_5_Na, 371.1829) and ^13^C NMR spectrum. The ^1^H NMR data (Table 2) of 8 exhibit four methyls (δH 1.44, s; 1.41, s; 1.24, s; 1.23, s) and two olefinic proton signals (δH 6.76, s; 6.02, s). The ^13^C NMR and HSQC spectra (Table 2) reveal 20 carbons, including 4 methyls, 5 sp^3^ methylenes, 3 methines (2 olefinic), and 8 quaternary carbons (including 4 sp^3^ with 2 oxygenated, 1 ketone carbonyl, 1 ester carbonyl, and 2 olefinic). Analysis of its 1D and 2D NMR spectra suggested that the structure of 8 belongs to an abietane diterpenoid. The ^1^H-^1^H COSY correlations of H-1/H-2/H-3 and the HMBC correlations of (Figure 2) H_3_-19/C-3, C-4, C-5, C-18 (δC 181.2) and H_3_-20/C-1, C-2, C-9, C-10 indicated that two methyl groups are attached to C-4 and C-10, and the two fragments are combined to form a six-membered ring. Moreover, the ^1^H-^1^H COSY correlations of H-5/H-6/H-7 and the HMBC correlations of H_2_-7/C-8, C-9, C-14, H-11 (δH 6.02)/C-8, C-9, C-13, and H-14 (δH 6.76)/ C-7, C-9, C-12, in consideration of the coupling pattern of H-11 (s) and H-14 (s) and the chemical shifts of C-8 (δC 68.5), C-12 (δC 189.7), and C-13 (δC 141.9), revealed the presence of a six-membered ring in which two pairs of double bonds are conjugated with the carbonyl group, and another carbon atom is substituted by a hydroxyl group. Based on the above information, it can be further inferred that these two six-membered rings are linked via C-5–C-6–C-7–C-8 and C-9–C-10 bonds to form a tricyclic diterpene skeleton. Further, HMBC correlations of H_3_-16/C-13, C-15, C-17 and H_3_-17/C-13, C-15, C-16 and the chemical shifts of C-15 (δC 71.9) revealed that C-13 is substituted by the isopropyl and that C-15 on the substituent is substituted by OH. The selective NOESY correlates by irradiating Ha-1 and H-5 of Ha-1/H_3_-20 and H-5/Hb-1, indicating that H_3_-20 and H-5 are on the opposite side. The relative configurations of C-4 and C-8 were determined by NMR calculations. The NMR chemical shifts of (4R**,5R**,8R**,10S**)-8 (8A), (4R**,5R**,8S**,10S**)-8 (8B), (4S**,5R**,8S**,10S**)-8 (8C), and (4S**,5R**,8R**,10S**)-8 (8D) (Figure S63) were calculated at the mPW1PW91/6-31G(d,p) level. DP4+ analysis yielded the highest probability (100%) for the 8B data, indicating the relative configuration of 8 as 4R**,5R**,8S**,10S**. Subsequently, ECD calculations were performed to determine 8’s absolute configuration. The calculated ECD spectrum for (4R,5R,8S,10S)-8 (Figure 4) shows good agreement with the experimental one, thereby confirming the absolute configuration of 8 as 4R,5R,8S,10S.

Pinusyunin F (9) was afforded as colorless needle crystals (MeOH). Its molecular formula was determined as C_21_H_30_O_5_ by the HRESIMS ion at m/z 385.1958 [M + Na]^+^ (calculated for C_21_H_30_O_5_Na, 385.1985) and ^13^C NMR spectrum. The ^1^H NMR spectrum of 9 (Table 3) indicates the presence of two methyl signals, one methoxy group (δH 3.76, s), one olefinic proton signal (δH 6.40, s), and one hydroxymethyl group (δH 4.43, s). The ^13^C NMR and HSQC spectra (Table 3) show 21 signals attributed to three methyls (one methoxy group), eight sp^3^ methylenes, three methines (two sp^2^ and one sp^3^), and seven non-protonated carbons (including two sp^3^, three olefinic carbons, and two ester carbonyls). A comparison of NMR data of 9 with those of 7-oxo-12α-hydroxy-abieta-8(14),13(15)-dien-18-oic acid [21] revealed their resemblance. Four changes are observed: (a) the carbonyl group at the C-7 is converted to a methylene group; (b) the OH group attached at the C-12 disappears; (c and d) the methyl groups at the C-16 and C-17 are oxidized to an ester group and a hydroxymethyl group, respectively. The observed ^1^H-^1^H COSY correlations of H-5/H_2_-6/H_2_-7 and H-9/H_2_-11/H_2_-12 and HMBC correlations of H_2_-7 (δH 2.44, 2.29)/C-5, C-8, C-9, C-11 and H_2_-12 (δH 3.28, 2.15)/C-9, C-13, C-15 combined with C-7 (δC 35.9) and C-12 (δC 27.6) chemical shifts, indicated the changes of C-7 and C-12. Moreover, the HMBC correlations of H-14 (δH 6.40)/C-12, C-13, C-15 andH_2_-17 (δH 4.43)/C-13, C-15, C-16 (δC 169.5), in combination with the chemical shifts of C-16 and C-17, showed that the methyl groups thereof are oxidized to an ester group and a hydroxymethyl group, respectively. Furthermore, the HMBC correlation of H_3_-21/C-16 suggested that C-7 and C-21 form an ester bond. Thus, the planar structure of 9 was determined.

The relative configuration of 9 was deduced from the 1D selective gradient NOESY spectra of H_3_-20/Ha-1, H_3_-19 by irradiating H_3_-20, indicating that two methyl groups are at the same orientation. By irradiating H-9, the correlations of H-9/Hb-1 and H-5 indicated that H-5 and H-9 are on the same side. As a result, the relative configuration of 9 was assigned. As for the geometry of the Δ^13(15)^ double bond, it was determined to be in the E-configuration through X-ray diffraction. Likewise, the absolute configuration of 9 was clarified by ECD calculations. It was found that the calculated ECD spectrum was in good accordance with the experimental one, suggesting the absolute configuration of 9 to be 4R,5R,9S,10R. Such a conclusion was further confirmed by the X-ray diffraction analysis using CuKα radiation with a Flack parameter of −0.02(9) (Figure 5).

Pinusyunin G (10), obtained as yellowish gums, presented the molecular formula of C_21_H_30_O_5_ as inferred from the HRESIMS ion peak at m/z 385.1972 [M + Na]^+^ (calculated for C_21_H_30_O_5_Na, 385.1985), ^13^C NMR, and DEPT spectra. A detailed comparison of ^1^H and ^13^C NMR data (Table 3) for 10 and 6,13β-dihydroxy-7-oxoabieta-5,8(14)-dien-19-al [22] found that their structures are similar. The only difference lies in the conversion of the aldehyde group at the C-18 to an ester group. The HMBC correlations (Figure 2) of H_3_-19/C-3, C-4, C-5, C-18 (δC 177.9) and H_3_-21/C-18 confirmed the aforementioned conclusion. The 1D selective gradient NOESY correlations of H_3_-20/Ha-1, Hb-11, H_3_-19, and H-9/Hb-1 by irradiating H_3_-20 and H-9, suggesting 10-CH_3_ and 4-CH_3_ are at the same orientation and opposite to H-9. The relative configuration of C-13 was determined through NMR calculations followed by DP4+ analysis. The results showed that the calculated NMR chemical shifts of (4R**,9R**,10R**,13S**)-10 other than (4R**,9R**,10R**,13R**)-10 exhibited a 100% probability by DP4+ analysis. With this information in hand, the absolute stereochemistry of 10 was accordingly assigned as 4R,9R,10R,13S by comparison of the calculated and experimental ECD curves.

Three known compounds were identified as (4R,5R,10S)-12-hydroxy-15-methoxy-7-oxo-abieta-8,11,13-trien-18-oic acid (3) [23], totara-8-en-7,13-dioxo-18-oic acid (5) [23], and macrophypene I (6) [24] by comparison with the literature data.

2.2. Biological Activity

The search for anti-fibrotic agents from natural sources has become our research focus in recent years. Within this context, the effects of compounds 1–10 on α-SMA, fibronectin, and collagen I were assessed using TGF-β1-induced NRK-52E and NRK-49F cells. To achieve this purpose, cell viability assays were first performed via Cell Counting Kit-8 (CCK-8) to exclude the effects arising from the cytotoxicity. The results showed that all of them were nontoxic at 40 μM in both cell lines (Figure 6A). Given that natural products generally exhibit relatively weak bioactivity, 40 μM was used in our bioactivity evaluations to avoid missing potential active compounds [14,15].

Western blot analysis demonstrated that all the compounds exerted significant inhibitory effects on the expression of α-SMA, fibronectin, and collagen I at 40 μM in TGF-β1-induced NRK-52E and NRK-49F cells. Of note, it was found that compounds 1–10 suppressed the expression of α-SMA in both cell lines, and compounds 1, 3, 6–8, and 10 decreased the expression of collagen I and fibronectin in NRK-52E cells (Figure 6B). In NRK-49F cells, all the isolates were found to reduce the expression of fibronectin. In contrast, only compounds 6–8 and 10, unlike the other analogues, downregulated collagen I expression in NRK-49F cells (Figure 6C), despite the fact that all the isolates are structural analogues. Finally, the results of all the three fibrosis-related proteins being sensitive to compounds 6–8 and 10 imply their beneficial effects for kidney protection. Furthermore, quantitative analysis in Figure 6B,C revealed that compound 10 exerted the most potent inhibitory effects on these three fibrosis-related proteins across both cell lines, confirming its superior activity against the tested fibrotic markers.

Moreover, we performed the structure–activity relationship (SAR) analysis to guide potential future structural optimization. The fact that all these abietane-type diterpenoids exhibited inhibitory activity against α-SMA in both cell lines and fibronectin in NRK-49F cells, suggests that their tricyclic rigid skeleton may contribute to the anti-renal fibrosis potential. This scaffold is completely distinct from those of known anti-fibrotic compounds, such as pirfenidone, nintedanib, GW, and SIS3 [12,16]. However, their effects on fibronectin in NRK-52E cells were variable, which may be attributed to differences in the regulatory mechanisms of fibronectin production between the two cell lines. The differential inhibitory activities on collagen I in both cell lines may be significantly associated with the different substituents at the C-13 position. Specifically, compound 4, with a native isopropyl substituent at C-13, exhibited better activity than the analogs with isopropyl migration (2 and 5) and isopropyl oxidation (9) at this position. In comparison, the substitution of a hydroxyl group at the C-15 of the isopropyl moiety attached to C-13 (7 and 8) exhibited more potent activity than the methoxylation at the same C-15 (1 and 3). Furthermore, the simultaneous introduction of isopropyl and hydroxyl groups at C-13 (1 and 10) yielded the most excellent activity, and 1, 3, 6–8, and 10 all exhibited superior anti-renal fibrosis activity to that of 4 with a native isopropyl substituent at C-13. Based on the SAR analysis, these active compounds could be further optimized via chemical synthesis to enhance their potency, laying a solid foundation for their potential clinical translation.

Considering that compounds 1, 3, 6–8, and 10 showed pronounced inhibitory effects on α-SMA, fibronectin, and collagen I, dose-response experiments in both NRK-52E and NRK-49F cells were subsequently performed by Western blotting. The results revealed that all of them could dose-dependently decrease fibronectin, collagen I, and α-SMA levels in both cell lines (Figure 7). Among the 10 compounds, only compound 10 retained the ability to reduced α-SMA expression in NRK-52E cells even at 10 μM. These results further demonstrate that compound 10 exhibits stronger activity against TGF-β1-induced fibrotic marker expression than the other tested compounds.

To investigate whether these compounds exert their inhibitory effects on fibrotic marker expression by attenuating the activation of the TGF-β/Smad signaling pathway, we detected the effects of compounds 1, 3, 6–8, and 10 on the expression of p-Smad2 and p-Smad3 at 40 μM in NRK-52E cells. It was found that compounds 8 and 10 exerted significant inhibitory effects on the expression of p-Smad2 and p-Smad3, while compounds 1, 3, 6, and 7 could only inhibit the expression of p-Smad3 (Figure 8). So far, it is known that the phosphorylation of Smad3 rather than Smad2 is involved in organ fibrosis. The selective inhibition of Smad3 activation by compounds 1, 3, 6, and 7 can precisely block canonical TGF-β/Smad signaling pathway-induced fibrosis, alleviate inflammation, and protect renal cells. Meanwhile, it retains the natural antagonistic effect of Smad2 on Smad3, thereby avoiding side effects caused by non-specific inhibition, making it an ideal strategy for the treatment of tissue fibrosis-related diseases such as renal and hepatic fibrosis [25,26]. Therefore, the structure–selectivity relationship of these diterpenoids in Smad2/3 inhibition cannot be summarized at present, and further investigation is warranted.

These active diterpenoids possess anti-renal fibrotic potential and may serve as effective, low-toxicity agents by selectively inhibiting Smad3. However, this study has limitations. On the one hand, all findings in this study were obtained from in vitro experiments using rat-derived NRK-52E and NRK-49F cells. In vitro assays fail to recapitulate the complex in vivo microenvironment and biological effects of fibrotic processes, and species differences further increase the challenges of clinical translation. On the other hand, the high lipophilicity and low bioavailability of diterpenoids exacerbate the difficulties in their clinical translation. Therefore, the bioactive compounds identified in this study need to be optimized by retaining the active moieties, introducing polar functional groups, or utilizing delivery systems such as nanoparticles to enhance their bioavailability. Meanwhile, further activity assays and pharmacokinetic studies should be performed in human-derived cells and unilateral ureteral obstruction (UUO) animal models to improve the potential for clinical translation.

3. Materials and Methods

3.1. General Procedures

Measurements of optical rotations were obtained using an Anton Paar MCP 100 polarimeter (Anton Paar GmbH, Graz, Austria). The UV and CD spectra were obtained on a Jasco J−815 circular dichroism spectrometer (JASCO, Tokyo, Japan). In addition, 1D and 2D NMR spectra were recorded on a Bruker Avance III 600 MHz spectrometer (Bruker, Karlsruhe, Germany), with TMS (tetra methyl silane) as an internal standard. Measurements were performed using SCIEX X500R QTOF MS spectrometer (AB Sciex LLC, Framingham, MA, USA) to obtain HRESIMS data. MCI gel CHP 20P (75–150 μm, Mitsubishi Chemical Industries, Tokyo, Japan), silica gel (200–300 mesh; Qingdao Marine Chemical Inc., Qingdao, China), reversed-phase C-18 silica gel (40–60 μm; Daiso Co., Ltd., Daito, Osaka, Japan) and Sephadex LH-20 (Amersham Pharmacia, Uppsala, Sweden) were used for column chromatography (CC). Semi-preparative HPLC was carried out by a Saipuruisi (LC-52, SEP, Beijing, China) chromatograph with a YMC-Pack ODS-A column (250 mm × 10 mm, i.d., 5 µm). Preparative HPLC was conducted on a Saipuruisi (LC-52, SEP, Beijing, China) chromatograph equipped with a YMC-Actus ODS-A column (250 × 20 mm, i.d., 5 μm).

3.2. Resinous Material

The source and authentication of P. yunnanensis resin was identical with our previous study [4] and the voucher specimen (CHYX0708) of P. yunnanensis resin has been deposited at the Institute for Inheritance-Based Innovation of Chinese Medicine, School of Pharmacy, Shenzhen University Medical School, Shenzhen University, Shenzhen, Guangdong, China.

3.3. Extraction and Isolation

P. yunnanensis resin (25.0 kg) was extracted with petroleum ether under ambient conditions (3 × 25 L, 3 h). The residue was then extracted with 95% EtOH, followed by concentration under reduced pressure to yield P. yunnanensis resin extract (6.8 kg). The crude extract was gradient eluted using an aqueous MeOH/^i^PrOH (10:1) (60–100%) on an MCI gel CHP 20P column, yielding eight fractions (Fr.A–Fr.H). Fr.E (800.0 g) was separated into eight fractions (Fr.E.A–Fr.E.H) by silica gel CC washed with petroleum ether–acetone (10:1, 8:1, 5:1, 3:1, 2:1, 1:1). The detailed purification process is shown in the Supporting Information section.

3.4. Compound Characterization Data

Pinusyunin A (1): white powders; [α]D^20^ −14 (c 0.05, MeOH); UV (MeOH) λmax (logε) 215 (2.56), 200 (3.02) nm; CD (MeOH) Δε_248_ −0.10, Δε_219_ +1.54, Δε_206_ +0.94, Δε_200_ +4.39; HRESIMS m/z: 383.2180 [M + Na]^+^ (calcd for C_22_H_32_O_4_Na, 383.2193); ^1^H and ^13^C NMR data, see Table 1.

Pinusyunin B (2): white powders; [α]D^20^ −108 (c 0.05, MeOH); UV (MeOH) λmax (logε) 386 (2.24), 221 (2.33), 201 (2.76) nm; CD (MeOH) Δε_248_ −0.03, Δε_223_ +1.01, Δε_216_ +0.74, Δε_208_ +1.39, Δε_201_ +0.75; HRESIMS m/z: 369.2032 [M + Na]^+^ (calcd for C_21_H_30_O_4_Na, 369.2036); ^1^H and ^13^C NMR data, see Table 1.

Pinusyunin C (4): white powders; [α]D^20^ −18 (c 0.05, MeOH); UV (MeOH) λmax (logε) 242 (2.82), 200 (2.39) nm; CD (MeOH) Δε_240_ −2.78, Δε_203_ +0.98; HRESIMS m/z: 371.2187 [M + Na]^+^ (calcd for C_21_H_32_O_4_Na, 371.2193); ^1^H and ^13^C NMR data, see Table 2.

Pinusyunin D (7): yellowish gums; [α]D^20^ −10 (c 0.05, MeOH); UV (MeOH) λmax (logε) 260 (2.91), 203 (2.82) nm; CD (MeOH) Δε_373_ +1.03, Δε_317_ +0.07, Δε_286_ +1.26, Δε_247_ −4.40, Δε_212_ +3.15; HRESIMS m/z: 355.1870 [M + Na]^+^ (calcd for C_20_H_28_O_4_Na, 355.1880); ^1^H and ^13^C NMR data, see Table 2.

Pinusyunin E (8): yellowish gums; [α]D^20^ +38 (c 0.05, MeOH); UV (MeOH) λmax (logε) 241 (2.41), 200 (2.27) nm; CD (MeOH) Δε_255_ +2.67, Δε_224_ −1.02; HRESIMS m/z: 371.1823 [M + Na]^+^ (calcd for C_20_H_28_O_5_Na, 371.1829); ^1^H and ^13^C NMR data, see Table 2.

Pinusyunin F (9): colorless needle crystals (MeOH); [α]D^20^ +34 (c 0.05, MeOH); UV (MeOH) λmax (logε) 276 (2.77), 201 (2.62) nm; CD (MeOH) Δε_369_ +2.44, Δε_232_ +0.58, Δε_221_ +0.72, Δε_206_ −0.36; HRESIMS m/z: 385.1958 [M + Na]^+^ (calcd for C_21_H_30_O_5_Na, 385.1985); ^1^H and ^13^C NMR data, see Table 3.

Pinusyunin G (10): yellowish gums; [α]D^20^ −52 (c 0.05, MeOH); UV (MeOH) λmax (logε) 315 (2.72), 269 (2.51), 200 (2.72) nm; CD (MeOH) Δε_354_ −3.81, Δε_311_ +2.68, Δε_239_ +0.10, Δε_225_ −1.39, Δε_200_ +5.36; HRESIMS m/z: 385.1972 [M + Na]^+^ (calcd for C_21_H_30_O_5_Na, 385.1985); ^1^H and ^13^C NMR data, see Table 3.

3.5. Crystal Structure Determination of 9

Crystal Data for 9 C_21_H_30_O_5_ (M =362.45 g/mol): orthorhombic, space group P2_1_2_1_2_1_ (no. 19), a = 7.02300(10) Å, b = 13.3132(3) Å, c = 20.9135(4) Å, V = 1955.38(6) Å^3^, Z = 4, T = 100.00(10) K, μ(Cu Kα) = 0.701 mm^−1^, Dcalc = 1.231 g/cm^3^, 18587 reflections measured (7.872° ≤ 2Θ ≤ 150.094°), 3886 unique (Rint = 0.0424, R_sigma_ = 0.0236) which were used in all calculations. The final R1 was 0.0520 (I > 2σ(I)), and wR2 was 0.1475 (all data). The crystal structure data for 9 were deposited in the Cambridge Crystallographic Data Centre (CCDC 2516890).

3.6. ECD Calculations of 1, 2, 4, and 7–10

The absolute configurations of 1, 2, 4, and 7–10 were determined by comparing the corresponding theoretical ECD spectra with the experimental spectra using the SpecDis (v1.62) program. For performing Molecular Mechanics Free Field (MMFF) and DFT/TDDFT calculations, the Spartan (v’14) software package (Wolfram Inc., Irvine, CA, USA) and the Gaussian (v09) software package were utilized [27,28,29] (References [28,29] are cited in Supplementary Information). MMFF conformation searches generated low-energy conformations within the 10 kcal mol^−1^ energy range, which were subsequently geometrically optimized using the DFT method at the B3LYP/6-311g(d,p) level.

3.7. Cell Culture

NRK-52E and NRK-49F cells (Procell Life Science & Technology Co., Ltd., Wuhan, China) were cultured in a high-glucose Dulbecco’s Modified Eagle Medium (DMEM, Shanghai VivaCell Biosciences Ltd., Shanghai, China). Specifically, NRK-52E cells were supplemented with 5% fetal bovine serum (FBS, Sigma-Aldrich LLC., St. Louis, MO, USA), while NRK-49F cells with 10% FBS, plus 1% penicillin/streptomycin in both formulations. All the cells were incubated at 37 °C in an atmosphere containing 5% CO_2_.

3.8. Cell Viability

NRK-52E and NRK-49F cells were plated into 96-well plates. After incubation for 24 h, cells were serum-starved for 6 h and then were treated with compounds 1–10 or DMSO at 40 μM for 48 h. Cell viability was measured by the CCK-8 assay (Shenzhen Laibo Biotechnology Co., Ltd., Shenzhen, China). 100 μL CCK-8 solution (1:9 in DMEM) was added into each well for 25 min at 37 °C, and the absorbance was measured at 450 nm using a BioTek microplate reader (BioTek, Winooski, VT, USA).

3.9. Western Blot

NRK-52E and NRK-49F cells were plated into 12-well plates. After incubation for 24 h, cells were serum-starved for 6 h and then incubated with recombinant human TGF-β1 (10 ng/mL; HY-P70543, MedChemExpress, San Jose, CA, USA) for 48 h with or without different concentrations of compounds. Notably, TGF-β1 was applied for only 2 h to determine the expression of p-Smad2 and p-Smad3. GW788388 (GW) was used as the positive control. After respective treatments, cells were lysed in RIPA lysis buffer (R21237, Shanghai Yuanye Bio-Technology Co., Ltd., Shanghai, China) containing protease inhibitors (DB612A, MIKX Biotechnology Co., Ltd., Wuhan, China) and phosphatase inhibitors (MB12707, Meilun Biotechnology Co., Ltd., Dalian, China). Equal amounts of total protein extracts were separated by 9% SDS-PAGE gels and transferred onto PVDF membranes (10600023, Cytiva, Marlborough, MA, USA). The membranes were blocked with 5% skim milk for 30 min at room temperature and subsequently incubated with primary antibodies (fibronectin, collagen I, α-SMA, Smad2, Smad3, p-Smad2, p-Smad3; GAPDH served as a loading control) overnight at 4 °C. The primary antibodies used were as follows: anti-GAPDH (GB15002, Servicebio Technology Co., Ltd., Wuhan, China); anti-α-SMA (A5228, Sigma-Aldrich, St. Louis, MO, USA); anti-Smad3 (A16913, ABclonal Technology, Wuhan, China); anti-p-Smad3 (#8828, Cell Signaling Technology, Danvers, MA, USA); anti-collagen I (ab270993, Abcam, Cambridge, UK); anti-fibronectin (ab2413, Abcam, Cambridge, UK); anti-Smad2 (ab40855, Abcam, Cambridge, UK); and anti-p-Smad2 (ab280888, Abcam, Cambridge, UK). The membranes were then incubated with horseradish peroxidase (HRP)-conjugated species-specific secondary antibodies (anti-rabbit, #7074; anti-mouse, #7076; Cell Signaling Technology, Danvers, MA, USA) for 1.5 h at room temperature. Protein bands were developed using the ECL kit (MK-S400, MIKX Biotechnology Co., Ltd., Wuhan, China) and then visualized on a gel imaging analysis system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). After that, the intensities of the protein bands were quantified by ImageJ software (v1.51p). The target protein band intensity ratio relative to GAPDH was further normalized to the mean value of the corresponding ratios in the TGF-β1-treated group (TGF-β1 group). These data were subsequently subjected to one-way analysis of variance (one-way ANOVA) and plotted as bar graphs using GraphPad Prism software (v9.5.0).

4. Conclusions

In summary, 10 abietane-type diterpenoids were isolated and structurally identified from P. yunnanensis resin in this study. Seven of them are new abietane-type diterpenoids (1, 2, 4, and 7–10). All the isolates share a characteristic tricyclic rigid skeleton, which is structurally distinct from those of known anti-fibrotic compounds. The biological evaluation demonstrated that all ten diterpenoids inhibited the expression of α-SMA, collagen I, and fibronectin in TGF-β1-induced NRK-52E and NRK-49F cells, supporting that this tricyclic rigid skeleton serves as a key structural scaffold for such inhibitory effects. Among the 10 compounds, compounds 6–8 and 10 demonstrated superior inhibitory activity, with compound 10 being the most potent inhibitory effects. The SAR analysis revealed that substituents at the C-13 position are critical for activity regulation. Abietane-type diterpenoids with both isopropyl and hydroxyl substituents at the C-13 position exhibited significantly enhanced inhibitory activity. Mechanistically, these diterpenoids exert their effects by attenuating the TGF-β/Smad signaling pathway. Compounds 1, 3, 6, and 7 inhibited fibrotic markers by suppressing the activity of the TGF-β/Smad signaling pathway, specifically inhibiting p-Smad3 without affecting the expression of p-Smad2. These findings may provide a potential source for the development of anti-renal fibrosis drugs.