Triterpenoids from the Leaves of Diospyros digyna and Their PTP1B Inhibitory Activity

Abstract

Six new 2α-hydroxy ursane triterpenoids, 3α-cis-p-coumaroyloxy-2α,19α-dihydroxy-12-ursen-28-oic acid (1), 3α-trans-p-coumaroyloxy-2α,19α-dihydroxy-12-ursen-28-oic acid (2), 3α-trans-p-coumaroyloxy-2α-hydroxy-12-ursen-28-oic acid (3), 3β-trans-p-coumaroyloxy-2α-hydroxy-12,20(30)-ursadien-28-oic acid (4), 3β-trans-feruloyloxy-2α-hydroxy-12,20(30)-ursadien-28-oic acid (5), and 3α-trans-feruloyloxy-2α-hydroxy-12,20(30)-ursadien-28-oic acid (6), along with eleven known triterpenoids (7–17), were isolated from the leaves of Diospyros digyna. Their chemical structures were elucidated by comprehensive analysis of UV, IR, HRESIMS, and NMR spectra. All the isolated compounds were evaluated for their PTP1B inhibitory activity. 3β-O-trans-feruloyl-2α-hydroxy-urs-12-en-28-oic acid (13) showed the best inhibition activity with an IC₅₀ value of 10.32 ± 1.21 μM. The molecular docking study found that the binding affinity of compound 13 for PTP1B was comparable to that of oleanolic acid (positive control).

1. Introduction

Type 2 diabetes (T2DM), a chronic metabolic disease, primarily results from an impaired insulin receptor signaling pathway [1]. T2DM accounts for approximately 90% of cases of diabetes. The multifactorial etiology of T2DM, including genetic, environmental, and lifestyle factors, has made its management and prevention a significant challenge for healthcare systems worldwide [2]. Protein tyrosine phosphatase 1B (PTP1B), an intracellular enzyme, has been implicated in the negative regulation of insulin signaling [3]. The overexpression of PTP1B in various tissues of T2DM patients highlights its pivotal role in the pathogenesis of insulin resistance [4]. Recent studies have demonstrated that genetic deletion or the pharmacological inhibition of PTP1B enhances insulin sensitivity and protects against diet-induced obesity [5,6,7,8], thereby underscoring PTP1B as a crucial therapeutical target for T2DM. Several PTP1B inhibitors have shown promise in preclinical models by improving glycemic control and insulin sensitivity [9]. Natural products are a rich source of new drug candidates [10] and an important source of PTP1B inhibitors [11,12].

The Diospyros genus comprises over 500 species of evergreen trees and shrubs, belonging to the family of Ebenaceae [13]. Diospyros plants are widely distributed in pantropical regions, and some species (particularly Diospyros digyna Jacq.) are edible fruit-yielding plants [14]. The chemical constituents of the plants, including triterpenoids, flavonoids, tannins, sugars, and phenolic acids, show antioxidant, anti-inflammatory, antiviral, antitumor, and PTP1B inhibitory activities [15,16,17]. In the search for natural PTP1B inhibitors, a phytochemical investigation on the leaves of D. digyna was carried out. Six new 2α-hydroxy ursane triterpenoids (1–6), along with eleven known ones were isolated from this plant (Figure 1). In addition, the PTP1B inhibitory activity of these triterpenoids was tested. Herein, the isolation, structural elucidation, and bioactivity of the isolates are presented.

2. Results

Compound 1 was isolated as a white amorphous powder. Its molecular formula was deduced as C₃₉H₅₄O₇ based on its HRESIMS ion at m/z 635.3934 [M + H]⁺. The UV spectrum showed the absorption maxima at 206, 228, and 310 nm. The IR spectrum suggested the presence of hydroxy (3425 cm⁻¹), carbonyl (1694 cm⁻¹), and aromatic (1605, 1513, and 1454 cm⁻¹) groups. The ¹H and ¹³C NMR spectra showed a cis-p-coumaroyl group [δ_H 7.66 (2H, d, J = 8.7 Hz), 6.88 (1H, d, J = 13.0 Hz), 6.75 (2H, d, J = 8.7 Hz), 5.87 (1H, d, J = 13.0 Hz); δ_C 168.6, 160.0, 144.8, 133.8 (×2), 127.9, 117.5, 116.0 (×2)], an olefinic bond [δ_H 5.32 (1H, m); δ_C 140.2, 129.4], two oxymethines [δ_H 5.00 (1H, d, J = 4.1 Hz), 4.11 (1H, dt, J = 11.0, 4.1 Hz); δ_C 81.4, 66.2], and seven methyls [δ_H 1.36 (3H, s), 1.23 (3H, s), 1.04 (3H, s), 1.00 (3H, s), 0.95 (3H, d, J = 6.7 Hz), 0.91 (3H, s), 0.81 (3H, s); δ_C 28.7, 27.2, 25.1, 22.4, 17.7, 17.0, 16.8] (

Table 1. ¹H (400 MHz) and ¹³C NMR (100 MHz) data of 1–3 (δ in ppm, J in Hz) ^a.

No.		1 b		2 b		3 c
		δH	δC	δH	δC	δH	δC
1	β	1.65	43.2	1.71	43.5	2.05	44.1
	α	1.37		1.38		1.86, m
2		4.11, dt (11.0, 4.1)	66.2	4.12, dt (10.4, 4.1)	66.2	4.50, dt (11.4, 4.3)	65.2
3		5.00, d (4.1)	81.4	5.03, d (4.1)	81.7	5.62, d (4.3)	81.4
4			39.6		39.8		39.2
5		1.19, m	51.1	1.30	51.3	1.56, m	50.9
6	α	1.47	19.3	1.50	19.3	1.49, m	18.8
	β	1.47		1.45		1.34
7	α	1.59, m	34.1	1.60	34.2	1.67, m	33.8
	β	1.34		1.35		1.41
8			41.4		41.4		40.6
9		1.82	48.6	1.94, m	48.8	2.00	48.6
10			39.6		39.7		39.2
11	α	2.04, m	24.8	2.06, m	24.9	2.12	24.2
	β	1.45		1.36		2.02
12		5.32, m	129.4	5.34, m	129.4	5.51, m	125.9
13			140.2		140.3		139.9
14			42.8		42.8		43.0
15	β	1.83	29.7	1.83	29.7	2.37, m	29.1
	α	1.03		1.02		1.21, m
16	α	2.60, m	26.7	2.61, m	26.7	2.11	25.4
	β	1.53, m		1.54		1.96
17			48.8		49.0		48.5
18		2.54, s	55.2	2.54, s	55.2	2.67, m	54.0
19			73.8		73.8	1.45	39.9
20		1.31	43.3	1.33	43.2	1.04	39.9
21	α	1.73	27.4	1.73	27.4	1.40	31.6
	β	1.27		1.23		1.48
22	β	1.74	39.2	1.75	39.2	2.00	38.0
	α	1.64		1.65		2.00
23		0.91, s	28.7	0.92, s	28.7	1.13, s	28.9
24		1.00, s	22.4	1.01, s	22.3	0.96, s	21.9
25		1.04, s	17.0	1.07, s	17.0	1.00, s	17.1
26		0.81, s	17.7	0.84, s	17.7	1.08, s	18.0
27		1.36, s	25.1	1.44, s	25.1	1.25, s	24.5
28			182.4		182.4		180.6
29		1.23, s	27.2	1.23, s	27.2	1.00	22.5
30		0.95, d (6.7)	16.8	0.96, d (6.6)	16.8	1.00	18.0
1’			168.6		169.5		168.4
2’		5.87, d (13.0)	117.5	6.40, d (15.9)	116.0	6.78, d (15.9)	116.6
3’		6.88, d (13.0)	144.8	7.63, d (15.9)	146.4	8.03, d (15.9)	145.5
4’			127.9		127.4		126.7
5’		7.66, d (8.7)	133.8	7.47, d (8.6)	131.3	7.52, d (8.6)	131.1
6’		6.75, d (8.7)	116.0	6.82, d (8.6)	117.0	7.15, d (8.6)	117.2
7’			160.0		161.3		161.8
8’		6.75, d (8.7)	116.0	6.82, d (8.6)	117.0	7.15, d (8.6)	117.2
9’		7.66, d (8.7)	133.8	7.47, d (8.6)	131.3	7.52, d (8.6)	131.1

^a Overlapped signals were reported without designating multiplicity. ^b NMR data were recorded in CD₃OD. ^c NMR data were recorded in C₅D₅N.

). Comparison of the 1D NMR spectra of 1 with those of 3-O-cis-p-coumaroyltormentic acid [18] indicated similar planar structures, which was further verified by the 2D NMR spectra of 1.

The ¹H–¹H COSY spectrum of 1 suggested spin-coupling systems of H₂-1/H-2/H-3, H-5/H₂-6/H₂-7, H-9/H₂-11/H-12, H₂-15/H₂-16, H₃-30/H-20/H₂-21/H₂-22, H-2′/H-3′, H-5′/H-6′, and H-8′/H-9′ (Figure 2). In the HMBC spectrum, the correlations from H₂-1/H-3/H₂-7/H-9/H₃-25 to C-5, from H-9/H-12/H-18 to C-14, from H-18 to C-12/C-14/C-16/C-28/C-29, from H-20/H-22 to C-18, from H₃-23 to C-3/C-5/C-24, from H₃-25 to C-1/C-5/C-9, from H₃-26 to C-7/C-9, from H₃-27 to C-8/C-13/C-15, and from H₃-30 to C-19/C-21, established an ursane-type triterpenoid skeleton (Figure 2). Moreover, the HMBC correlation from H-3 to C-1′ located the cis-p-coumaroyl group at C-3. In the NOESY spectrum, the correlations between H-9/H₃-24 and H-5, between H-9/H-16α and H₃-27, and between H-16α and H₃-30 suggested H-5, H-9, H₃-24, and H₃-27 were α-oriented. The NOE correlations between H-2/H₃-23/H₃-26 and H₃-25, between H-3 and H₃-23, and between H-20/H₃-29 and H-18 suggested H-2, H-3, H-18, H-20, H₃-23, H₃-25, H₃-26, and H₃-29 were β-oriented (Figure 3). Thus, the structure of 1 was elucidated and named as 3α-cis-p-coumaroyloxy-2α,19α-dihydroxy-12-ursen-28-oic acid.

The molecular formula of 2 was determined as C₃₉H₅₄O₇ by the HRESIMS at m/z 635.3938 [M + H]⁺ (calcd for C₃₉H₅₅O₇, 635.3942). The UV spectrum showed the absorption maxima at 205, 226, and 312 nm. The IR spectrum suggested the presence of hydroxy (3417 cm⁻¹), carbonyl (1693 cm⁻¹), and aromatic (1608, 1515, and 1453 cm⁻¹) groups. The 1D NMR data of 2 were similar to those of 1, except for the presence of a trans-p-coumaroyl group [δ_H 7.47 (2H, d, J = 8.6 Hz), 7.63 (1H, d, J = 15.9 Hz), 6.82 (2H, d, J = 8.6 Hz), 6.40 (1H, d, J = 15.9 Hz); δ_C 169.5, 161.3, 146.4, 131.3 (×2), 127.4, 117.0 (×2), 116.0], and the absence of a cis-p-coumaroyl group (Table 1). The structure of 2 was verified by its 2D NMR spectra. In the NOESY spectrum, the correlations between H-3/H₃-25 and H₃-23 indicated that H-3 was β-oriented (Figure 3). Thus, the structure of 2 was elucidated and named as 3α-O-trans-p-coumaroyloxy-2α,19α-dihydroxy-12-ursen-28-oic acid.

The molecular formula of 3 was determined as C₃₉H₅₄O₆ by the HRESIMS at m/z 619.3986 [M + H]⁺ (calcd for C₃₉H₅₅O₆, 619.3993). The UV spectrum showed the absorption maxima at 208, 228, and 312 nm. The IR spectrum suggested the presence of hydroxy (3449 cm⁻¹), carbonyl (1695 cm⁻¹), and aromatic (1599, 1519, and 1458 cm⁻¹) groups. The 1D NMR data of 3 were similar to those of jacoumaric acid (11) [19,20], except for the chemical shifts of Δδ_C −5.0 (C-1), −1.7 (C-2), −4.1 (C-3), −1.6 (C-4), −5.1 (C-5), −0.6 (C-23), and +3.7 (C-24), which indicated 3 might be a 3-epimer of 11. In the NOESY spectrum, the similar correlations between H-3/H₃-25 and H₃-23 suggested that H-3 was β-oriented. Based on the above analysis, compound 3 was determined to be 3α-O-trans-p-coumaroyloxy-2α-hydroxy-12-ursen-28-oic acid.

Compound 4 was isolated as an amorphous powder. Its HRESIMS at m/z 639.3637 [M + Na]⁺ exhibited the molecular formula of C₃₉H₅₂O₆. The UV spectrum showed the absorption maxima at 208, 228, and 312 nm. The IR spectrum suggested the presence of hydroxy (3204 cm⁻¹), carbonyl (1695 cm⁻¹), and aromatic (1602, 1515, and 1453 cm⁻¹) groups. The ¹H and ¹³C NMR spectra showed a trans-p-coumaroyl group [δ_H 8.02 (1H, d, J = 15.9 Hz), 7.57 (2H, d, J = 8.5 Hz), 7.18 (2H, d, J = 8.5 Hz), 6.70 (1H, d, J = 15.9 Hz); δ_C 168.4, 161.8, 145.3, 131.1, 131.1, 126.7, 117.3, 117.3, 116.6], two olefinic bonds [δ_H 5.47 (1H, m), 4.84 (1H, br s), 4.79 (1H, br s); δ_C 154.2, 139.5, 126.2, 105.6], two oxymethines [δ_H 5.28 (1H, d, J = 10.8 Hz), 4.32 (1H, ddd, J = 10.8, 4.3, 3.7 Hz); δ_C 85.5, 66.8], and six methyls [δ_H 1.22 (3H, s), 1.13 (3H, d, J = 6.4 Hz), 1.09 (3H, s), 1.06 (3H, s), 1.03 (3H, s), 1.00 (3H, s); δ_C 29.5, 24.2, 18.7, 17.9, 17.4, 17.1] (

Table 2. ¹H (400 MHz) and ¹³C NMR (100 MHz) data of 4–6 (C₅D₅N, δ in ppm, J in Hz) ^a.

No.		4		5		6
		δH	δC	δH	δC	δH	δC
1	β	2.33	49.0	2.33	49.1	2.03	44.1
	α	1.42, m		1.41, m		1.86, m
2		4.32, ddd (10.8, 4.3, 3.7)	66.8	4.32, ddd (10.8, 4.3, 3.7)	66.9	4.49, dt (10.8, 4.5)	65.1
3		5.28, d (10.8)	85.5	5.29, d (10.8)	85.5	5.64, br s	81.4
4			40.4		40.3		39.2
5		1.10, m	56.0	1.11, m	56.1	1.53	50.9
6	α	1.51	19.1	1.53	19.1	1.47	18.7
	β	1.37		1.38		1.33
7	α	1.57	33.7	1.56	33.8	1.62, m	33.7
	β	1.35		1.37		1.39
8			40.3		40.5		40.6
9		1.74, m	48.4	1.76, m	48.5	1.95	48.5
10			38.7		38.8		39.2
11	α	1.98	24.2	1.99	24.2	2.06	24.1
	β	1.98		1.27		1.27
12		5.47, m	126.2	5.48, m	126.2	5.49, m	126.3
13			139.5		139.5		139.5
14			43.1		43.1		43.0
15	β	2.31	29.1	2.32	29.1	2.33	29.0
	α	1.24, m		1.25, m		1.22, m
16	α	2.30	25.3	2.32	25.4	2.30	25.3
	β	2.12		2.10		2.09
17			48.7		48.8		48.7
18		2.77, d (11.8)	56.0	2.78, d (11.8)	56.1	2.76, d (11.6)	56.0
19		2.48, m	38.2	2.48, m	38.2	2.44, m	38.2
20			154.2		154.3		154.2
21	α	2.27	33.2	2.30	33.2	2.28	33.2
	β	2.43, m		2.45, m		2.40
22	β	2.14	40.1	2.15	40.2	2.12, m	40.1
	α	2.03		2.02		2.02
23		1.09, s	29.5	1.10, s	29.5	1.14, s	28.9
24		1.06, s	18.7	1.05, s	18.8	0.96, s	22.5
25		1.00, s	17.4	1.01, s	17.4	0.98, s	17.1
26		1.03, s	17.9	1.04, s	17.9	1.05, s	17.9
27		1.22, s	24.2	1.23, s	24.3	1.17, s	24.2
28			179.8		179.9		179.8
29		1.13, d (6.4)	17.1	1.14, d (6.1)	17.1	1.11, d (6.1)	17.1
30	a	4.84, br s	105.6	4.84, br s	105.6	4.83, br s	105.6
	b	4.79, br s		4.79, br s		4.79, br s
1′			168.4		168.4		168.4
2′		6.70, d (15.9)	116.6	6.75, d (15.8)	116.7	6.89, d (15.9)	116.7
3′		8.02, d (15.9)	145.3	8.04, d (15.8)	145.6	8.07, d (15.9)	146.0
4′			126.7		127.2		127.1
5′		7.57, d (8.5)	131.1	7.31, m	112.0	7.24, m	111.8
6′		7.18, d (8.5)	117.3		149.5		149.4
7′			161.8		151.5		151.5
8′		7.18, d (8.5)	117.3	7.23, m	117.3	7.20, m	117.2
9′		7.57, d (8.5)	131.1	7.23, m	124.1	7.20, m	124.5
Ome				3.80, s	56.4	3.72, s	56.3

^a Overlapped signals were reported without designating multiplicity.

). The above NMR data of 4 were similar to those of 3α-trans-coumaroyloxy-2α-hydroxy-12,20(30)-dien-28-ursolic acid (15) [21], except for the chemical shifts of Δδ_C +5.0 (C-1), +1.7 (C-2), +4.1 (C-3), +1.3 (C-4), +5.1 (C-5), +0.7 (C-23), and −3.7 (C-24), which indicated 4 might be a 3-epimer of 15. In the NOESY spectrum, the correlations between H-2 and H₃-25, and between H-3 and H-5, suggested that H-2 was β-oriented, while H-3 was α-oriented (Figure 4). Thus, the structure of 4 was elucidated and named as 3β-trans-p-coumaroyloxy-2α-hydroxy-12,20(30)-ursadien-28-oic acid.

The molecular formula of 5 was determined as C₄₀H₅₄O₇ by the HRESIMS at m/z 647.3944 [M + H]⁺ (calcd for C₄₀H₅₅O₇, 647.3942). The UV spectrum showed the absorption maxima at 208, 238, and 325 nm. The IR spectrum suggested the presence of hydroxy (3301 cm⁻¹), carbonyl (1699 cm⁻¹), and aromatic (1598, 1514, and 1447 cm⁻¹) groups. Comparison of NMR data of 5 to those of 4 (Table 2) revealed similar structures, except for the presence of a methoxy group [δ_H 3.80 (3H, s); δ_C 56.4] in 5. In the HMBC spectrum, the correlation from CH₃O- to C-6′ located CH₃O- at C-6′ (Figure 2). Thus, the structure of 5 was elucidated and named as 3β-trans-feruloyloxy-2α-hydroxy-12,20(30)-ursadien-28-oic acid.

The molecule formula of 6 was identical to that of 5 based on the HRESIMS at m/z 647.3945 [M + H]⁺ (calcd for C₄₀H₅₅O₇, 647.3942). The UV spectrum showed the absorption maxima at 208, 242, and 325 nm. The IR spectrum suggested the presence of hydroxy (3450 cm⁻¹), carbonyl (1699 cm⁻¹), and aromatic (1597, 1517, and 1460 cm⁻¹) groups. The 1D NMR data of 6 were similar to those of 5, except for the chemical shifts of Δδ_C +5.0 (C-1), +1.8 (C-2), +4.1 (C-3), +1.1 (C-4), +5.2 (C-5), +0.6 (C-23), and −3.7 (C-24), which indicated 6 might be a 3-epimer of 5. In the NOESY spectrum, the correlations between H-3/H₃-25 and H₃-23 suggested that H-3 was β-oriented. Thus, the structure of 6 was identified and named as 3α-trans-feruloyloxy-2α-hydroxy-12,20(30)-ursadien-28-oic acid.

Apart from the above six new 2α-hydroxy ursane triterpenoids (1–6), eleven known triterpenoids were isolated and identified as 3α-trans-feruloyloxy-2α-hydroxyurs-12-en-28-oic acid (7) [22], 3-O-trans-feruloyl euscaphic acid (8) [23], colosolic acid (9) [24], 3β-O-cis-p-coumaroyl-2α-hydroxy-urs-12-en-28-oic acid (10) [20,25], jacoumaric acid (11) [20], 3β-O-cis-feruloyl-2α-hydroxy-urs-12-en-28-oic acid (12) [20], 3β-O-trans-feruloyl-2α-hydroxy-urs-12-en-28-oic acid (13) [20], (2α,3α)-2-hydroxy-3-[(2Z)-3-(4-hydroxyphenyl)-1-oxo-2-propenyl]oxy]ursa-12,20(30)-dien-28-oic acid (14) [21], 3α-trans-coumaroyloxy-2α-hydroxy-12,20(30)-dien-28-ursolic acid (15) [21], 2α-hydroxymicromeric acid (16) [26], and 3β-cis-p-coumaroyloxy-2α-hydroxyursa-12,20(30)-dien-28-oic acid (17) [27].

All the isolated compounds were evaluated for their PTP1B inhibitory activity. As a result, compounds 4–6, 10–13, and 15 showed PTP1B inhibition with IC₅₀ values in the range of 10.32–48.67 μM (

Table 3. PTP1B inhibitory activity.

Compounds	IC50 (μM)	Compounds	IC50 (μM)
4	48.67 ± 5.17	11	26.20 ± 1.31
5	23.74 ± 0.73	12	16.20 ± 0.57
6	34.01 ± 4.88	13	10.32 ± 1.21
10	19.15 ± 0.22	15	17.12 ± 1.67
Oleanolic acid a	10.19 ± 0.12

^a Positive control.

), while other compounds were over 50 μM. Compounds 12, 13, and 15 showed better inhibition activity with IC₅₀ values of 16.20, 10.32, and 17.12 μM, respectively. To know more about the binding and interaction mode between PTP1B and compounds 12, 13, 15, and oleanolic acid, a molecular docking study was conducted by AutoDock Vina. The binding energies of compounds 12, 13, and 15 to PTP1B are −7.1, −7.8, and −7.5 kcal/mol, respectively. Compound 13 is slightly better than that of oleanolic acid (−7.6 kcal/mol), suggesting comparable binding affinity to PTP1B. As shown in Figure 5, these active compounds can dock into the same hydrophobic pocket and bind to the catalytic residues (Gln262, Ala217, and Tyr46) by different interactions as the positive control.

3. Experimental Section

3.1. General Experimental Procedures

UV, IR, ECD, and optical rotations were recorded on JASCO V550 UV/VIS, JASCO FTR-4600, JASCO-180, and JASCO P-2000 spectrometers (JASCO, Tokyo, Japan), respectively. HRESIMS were obtained using an Agilent 6210 LC/MSD TOF mass spectrometer (Agilent Technologies, Inc., Santa Clara, CA, USA). NMR data were measured by a Bruker AV-400 NMR spectrometer (Bruker, Fällanden, Switzerland). The preparative HPLC was carried on an Agilent 1200 HPLC (Agilent Technologies, Inc., Santa Clara, CA, USA) with a Cosmosil 5C18-MS-II (250 mm × 10 mm, 5 μm). The GF₂₅₄ silica gel plates were purchased from the Yantai Institute of Industrial Chemistry, Yantai, China. Silica gel (80–100 mesh, 100–200 mesh, and 200–300 mesh; Qingdao Marine Chemical, Ltd., Qingdao, China), ODS (C₁₈, Merck, Darmstadt, Germany), and Sephadex LH-20 (Pharmacia, Kalamazoo, MI, USA) were used for column chromatography.

3.2. Plant Material

The leaves of Diospyros digyna Jacq. were collected from Zhongshan Haizaoye Agricultural Technology Co., Ltd., Guangdong, China, in July 2018. The plant was identified by Prof. Guangxiong Zhou, College of Pharmacy, Jinan University. A voucher specimen (No. CP2018070903) was deposited in the herbarium of Jinan University.

3.3. Extraction and Isolation

The dried leaves of D. digyna (17.0 kg) were extracted with 95% ethanol at room temperature to give the crude extract (4.3 kg). Then, the crude extract was suspended into H₂O, and partitioned successively with petroleum ether (PE), ethyl acetate (EA), and n-butanol, respectively. The ethyl acetate extract (803 g) was subjected to a silica gel column and eluted with CH₂Cl₂/CH₃OH (100:0→0:100, v/v) to obtain eight main fractions (Fr. A–Fr. H). Fr C (20 g) was subjected to a silica gel column eluted with PE/EA (10:1→1:1, v/v) to afford six subfractions (Fr. C1–Fr. C6). Fr. C3 (5.0 g) was purified by Sephadex LH-20 columns (CHCl₃:CH₃OH = 1:1, v/v) and preparative HPLC to afford 3 (8.5 mg, CH₃OH:H₂O:HCOOH = 83:17:0.1, v/v/v, t_R = 32.7 min), 9 (8.0 mg, CH₃OH:H₂O:HCOOH = 98:2:0.1, v/v/v, t_R = 32.5 min), 11 (9.0 mg, CH₃OH:H₂O:HCOOH = 85:15:0.1, v/v/v, t_R = 11.0 min), 13 (6.0 mg, CH₃OH:H₂O:HCOOH = 83:17:0.1, v/v/v, t_R = 31.5 min), and 14 (13.0 mg, CH₃OH:H₂O:HCOOH = 83:17:0.1, v/v/v, t_R = 39.3 min). Fr. D (15 g) was subjected to an ODS column eluted with CH₃OH/H₂O (100:0→0:100, v/v) to obtain seven subfractions (Fr. D1–Fr. D7). Fr. D4 (5.8 g) was purified by Sephadex LH-20 columns (CH₃OH) and preparative HPLC to afford 1 (26.0 mg, CH₃CN:H₂O:HCOOH = 60:40:0.1, v/v/v, t_R = 22.1 min), 2 (10.0 mg, CH₃CN:H₂O:HCOOH = 60:40:0.1, v/v/v, t_R = 25.2 min), 10 (10.0 mg, CH₃OH:H₂O:HCOOH = 80:20:0.1, v/v/v, t_R = 15.5 min), 12 (20.0 mg, CH₃CN:H₂O:HCOOH = 75:25:0.1, v/v/v, t_R = 17.5 min), 16 (2.0 mg, CH₃CN:H₂O:HCOOH = 65:35:0.1, v/v/v, t_R = 20.5 min), and 17 (6.5 mg, CH₃CN:H₂O:HCOOH = 65:35:0.1, v/v/v, t_R = 22.5 min). Fr. F (10 g) was subjected to a silica gel column eluted with PE/EA (9:1→0:10, v/v) to obtain six subfractions (Fr. F1–Fr. F6). Fr. F4 (2 g) was purified by a MCI column and preparative HPLC to afford 4 (10.0 mg, CH₃CN:H₂O:HCOOH = 60:40:0.1, v/v/v, t_R = 19.5 min), 5 (12.0 mg, CH₃CN:H₂O:HCOOH = 60:40:0.1, v/v/v, t_R = 14.5 min), 6 (9.0 mg, CH₃CN:H₂O:HCOOH = 70:30:0.1, v/v/v, t_R = 32.0 min), 7 (28.5 mg, CH₃CN:H₂O:HCOOH = 70:30:0.1, v/v/v, t_R = 24.3 min), 8 (19.0 mg, CH₃CN:H₂O:HCOOH = 60:40:0.1, v/v/v, t_R = 14.0 min), and 15 (12.0 mg, CH₃OH:H₂O:HCOOH = 80:20:0.1, v/v/v, t_R = 12.5 min).

Compound 1: white amorphous powder; ${[\mathsf{\alpha }]}_D^{25}$ + 16 (c 1.3, CH₃OH); UV (CH₃OH) λ_max (log ε) 206 (3.59), 228 (3.43), 310 (3.58) nm; IR (KBr) v_max 3425, 2937, 1694, 1605, 1513, 1454, 1387, 1273, 1173, 1041, 979, 937, 839 cm⁻¹; HRESIMS m/z: 635.3934 [M + H]⁺ (calcd for C₃₉H₅₅O₇, 635.3942); ¹H NMR (400 MHz, CD₃OD) and ¹³C NMR (100 MHz, CD₃OD) (Table 1 and Figures S1–S6, Supplementary Materials).

Compound 2: white amorphous powder; ${[\mathsf{\alpha }]}_D^{25}$ + 24 (c 1.3, CH₃OH); UV (CH₃OH) λ_max (log ε) 205 (3.62), 226 (3.46), 312 (3.64) nm; IR (KBr) v_max 3417, 2938, 1693, 1608, 1515, 1453, 1386, 1272, 1198, 1177, 1041, 960, 936, 837 cm⁻¹; HRESIMS m/z: 635.3938 [M + H]⁺ (calcd for C₃₉H₅₅O₇, 635.3942); ¹H NMR (400 MHz, CD₃OD) and ¹³C NMR (100 MHz, CD₃OD) (Table 1 and Figures S10–S15, Supplementary Materials).

Compound 3: white amorphous powder; ${[\mathsf{\alpha }]}_D^{25}$ + 35 (c 0.3, CH₃OH); UV (CH₃OH) λ_max (log ε) 208 (3.46), 228 (3.31), 312 (3.53) nm; IR (KBr) v_max 3449, 2935, 1695, 1599, 1519, 1458, 1268, 1178, 1039, 984, 819 cm⁻¹; HRESIMS m/z: 619.3986 [M + H]⁺ (calcd for C₃₉H₅₅O₆, 619.3993); ¹H NMR (400 MHz, C₅D₅N) and ¹³C NMR (100 MHz, C₅D₅N) (Table 1 and Figures S19–S24, Supplementary Materials).

Compound 4: white amorphous powder; ${[\mathsf{\alpha }]}_D^{25}$ + 24 (c 1.4, CH₃OH); UV (CH₃OH) λ_max (log ε) 208 (3.49), 228 (3.30), 312 (3.54) nm; IR (KBr) v_max 3204, 2938, 1695, 1602, 1515, 1453, 1385, 1273, 1175, 1042, 961, 938, 840 cm⁻¹; HRESIMS m/z: 639.3637 [M + Na]⁺ (calcd for C₃₉H₅₂O₆Na, 639.3656); ¹H NMR (400 MHz, C₅D₅N) and ¹³C NMR (100 MHz, C₅D₅N) (Table 2 and Figures S28–S33, Supplementary Materials).

Compound 5: white amorphous powder; ${[\mathsf{\alpha }]}_D^{25}$ + 94 (c 1.7, CH₃OH); UV (CH₃OH) λ_max (log ε) 208 (3.60), 238 (3.37), 325 (3.51) nm; IR (KBr) v_max 3301, 2939, 1699, 1598, 1514, 1447, 1369, 1268, 1170, 1102, 1018, 869, 833 cm⁻¹; HRESIMS m/z: 647.3944 [M + H]⁺ (calcd for C₄₀H₅₅O₇, 647.3942); ¹H NMR (400 MHz, C₅D₅N) and ¹³C NMR (100 MHz, C₅D₅N) (Table 2 and Figures S37–S42, Supplementary Materials).

Compound 6: white amorphous powder; ${[\mathsf{\alpha }]}_D^{25}$ + 24 (c 1.4, CH₃OH); UV (CH₃OH) λ_max (log ε) 208 (3.48), 242 (3.26), 325 (3.35) nm; IR (KBr) v_max 3450, 2942, 1699, 1597, 1517, 1460, 1378, 1269, 1175, 1131, 1037, 966, 942, 817 cm⁻¹; HRESIMS m/z: 647.3945 [M + H]⁺ (calcd for C₄₀H₅₅O₇, 647.3942); ¹H NMR (400 MHz, C₅D₅N) and ¹³C NMR (100 MHz, C₅D₅N) (Table 2 and Figures S46–S51, Supplementary Materials).

3.4. PTP1B Inhibition Assay

The inhibitory activity of the isolated compounds against PTP1B (Abcam, Cambridge, UK, human recombinant) was assayed according to the method reported previously [28]. The reagent p-nitrophenyl phosphate (pNPP) was used as the substrate, and oleanolic acid was used as the positive control. In brief, a 100 μL assay mixture containing 1 μg/mL PTP1B, samples, 4 mM pNPP, 55 mM NaCl, 2.2 mM DTT, 1.1 mM EDTA, and 1 mM BSA in 11 mM Tris-HCl, pH 7.5, was incubated at 37 °C for 30 min in a 96-well plate. The absorbance of 405 nm was measured by a microplate reader. Data were analyzed by GraphPad Prism v.10.2.0 software. All data were obtained in triplicate and presented as means ± SD.

3.5. Molecular Docking Analysis

The crystal structure of PTP1B (PDB ID: 8U1E) was obtained from the RCSB Protein Data Bank database. The receptor was prepared by PyMOL 2.5.0 and deposited as a .pdb format. The 3D structures of the ligands were energy optimized by ChemDraw 3D 18.0 and deposited as a mol2 format. The ligands and receptors for molecular docking analysis were conducted by AutoDock Vina 1.2.2 [29]. The grid box parameters (X-center = 0.405, Y-center = 12.132, Z-center = 25.000; x-dimension = 46, y-dimension = 60, z-dimension = 48) were set to cover the binding pocket in the receptor. The docking calculation results were analyzed by PyMOL 2.5.0.

4. Conclusions

In summary, six new 2α-hydroxy ursane triterpenoids (1–6), along with eleven known ursane triterpenoids (7–17), were isolated from the leaves of D. digyna. Compounds 4–6, 10–13, and 15 showed PTP1B inhibitory activity. Notably, compound 13 demonstrated PTP1B inhibition comparable to that of oleanolic acid (positive control). The structure–activity relationships of these triterpenoids were briefly summarized. Compound 13 showed stronger PTP1B inhibitory activity than those of 10–12, indicating that the trans-feruloyl group at C-3 strengthens the activity. For compounds 3, 7, 11, and 13, the α-orientation of the substituents at C-3 weakened the activity. Moreover, the molecular docking study further confirmed the binding affinity between compound 13 and PTP1B. The naturally occurring PTP1B inhibitors might reveal the potential utilization of D. digyn in the treatment of T2DM, and their action mechanisms deserve further investigation.