Six New Cyclic Peptides from the Roots of Gypsophila oldhamiana

Abstract

Six new cyclic peptides, gypsophin A–F (1–6), were isolated from Gypsophila oldhamiana. Their structures were elucidated by extensive NMR and chemical degradation. Compound 3 exhibited moderate activity of antiplatelet aggregation in vitro.

Plants of the Caryophyllaceae family are a rich source of cyclopeptide alkaloids. These Caryophyllaceae-type cyclopeptides showed interesting biological properties, including cytotoxic,^1–3) antimalarial,^4,5) antiplatelet,⁶⁾ immunosuppressive,⁷⁾ immunomodulating,⁸⁾ Ca²⁺ antagonistic,⁹⁾ inhibiting cyclooxygenase^10,11) and tyrosinase, enhancing rotamase,^10,11) and estrogen-like activity.^12,13)

Gypsophila oldhamiana (Miq.) (Caryophyllaceae) is widely distributed in the north of China, and its roots have been used as a traditional folk medicine Shan-Yin-Chai-Hu for the treatment of fever, consumptive disease and infantile malnutrition syndrome.¹⁴⁾ Previous phytochemical studies on the roots of this plant have afforded triterpenoid saponins, sterols, and glycosides.^15–17) We also have reported gypsophin,¹⁸⁾ an unusual new cyclic peptide with a pyrrolidine-2,5-dione unit isolated from this plant. Further investigations on this plant led to the isolation of six new cyclic peptides, gypsophin A–F (1–6) (Fig. 1). Herein we describe the isolation and structure elucidation of 1–6. In addition, the isolates were tested for antiplatelet aggregation activity and cytotoxicity in Bel7402, SMMC-7721, HCT 116 and H460 cancer cell lines.

Fig. 1. Structures of Compounds 1–6

Results and Discussion

Gypsophin A (1), obtained as a white powder, has a molecule formula of C₅₁H₆₇N₉O₁₁ (23 unsaturations) based on its positive high resolution-electrospray ionization (HR-ESI)-MS, showing a [M+H]⁺ ion peak at m/z 982.5034 (Calcd for 982.5033). The IR absorptions at 3408, 1651, and 1516 cm⁻¹ were ascribable to amino, amide carbonyl, and aromatic groups, respectively. The ¹H- and ¹³C-NMR data of 1 (Table 1) displayed signals of nine amide carbonyls (δ_C 174.7, 171.3, 171.2, 171.1, 171.0, 170.9, 170.6, 170.0, 169.7), eight normal α-amino acid carbon resonances (δ_C 59.2, 58.3, 56.8, 56.4, 56.0, 56.0, 53.7, 50.2), and nine amide protons (δ_H 8.51, 8.46, 8.05, 8.04, 8.01, 7.85, 7.12, 7.07, 6.89), indicating that compound 1 is a peptide. HPLC analysis of the acid hydrolysates and the ¹H-NMR data showed that 1 consisted proline (Pro), tyrosine (Tyr), serine (Ser), phenylalanine (Phe), valine (Val), isoleucine (Ile) and glutamine (Gln). Eight amino acid residues of Pro¹, Tyr², Ser³, Phe⁴, Phe⁵, Val⁶, Ile⁷ and Gln⁸ were assigned by analysis of the total correlation spectroscopy (TOCSY), ¹H-detected heteronuclear multiple quantum coherence (HMQC) and heteronuclear multiple bond correlation (HMBC) experiments (Table 1). The HMBC correlations between the carbonyls and the α- or β-protons of the same amino acid residue, the carbonyl resonances of the amino acid residues were determined as δ_C 171.3 (Pro¹), 171.0 (Tyr²), 170.0 (Ser³), 169.7 (Phe⁴), 171.1 (Phe⁵), 170.6 (Val⁶), 170.9 (Ile⁷) and 171.2 (Gln⁸), respectively. In addition, another carbonyl signal at δ174.7 was assigned as δ-carbon of glutamine residue (Gln), which was illuminated by HMBC cross-peaks for this carbon with Hγ (2.12) of glutamine residue (Fig. 2). All the amino acid residues accounted for 22 degrees of unsaturation, indicating that 1 is a cyclic octapeptide. The sequence of the amino acid residues was deduced with HMBC and rotating frame nuclear Overhauser effect correlation spectroscopy (ROESY). The HMBC spectrum showed correlations of Tyr²-NH (δ_H 8.05)/Pro¹-CO (δ_C 171.3), Phe⁵-NH (δ_H 8.51)/Phe⁴-CO (δ_C 169.7), Ile⁷-NH (δ_H 7.12)/Val⁶-CO (δ_C 170.6) and Gln⁸-NH (δ_H 8.46)/Ile⁷-CO (δ_C 170.9) (Fig. 2), corresponding to the peptide fragments Pro¹-Tyr² and Val⁶-Ile⁷-Gln⁸. The observed ROESY correlations of Ser³-Hβ (δ 3.98) and the Phe⁴ o-methylidyne resonance at δ_H 7.28,¹⁹⁾ Tyr²-Hα/Ser³-Hα, Phe⁴-Hα/Phe⁵-Hα and Val⁶-NH/Phe⁵-NH, suggested the fragment Ser³-Phe⁴, Tyr²-Ser³, Phe⁴-Phe⁵-Val⁶, respectively (Fig. 2). Thus, 1 was determined as cyclo-(Pro¹-Tyr²-Ser³-Phe⁴-Phe⁵-Val⁶-Ile⁷-Gln⁸) and named gypsophin A. The geometry of the amide bond of Pro¹ was cis, on the basis of the difference in the ¹³C-NMR chemical shifts of Pro¹ (Δδ_Cβ–Cγ=9.0 ppm).^20,21)

Fig. 2. The Key Selective HMBC and ROESY Correlations of Compounds 1, 2, 3 and 6

Table 1. ¹H- and ¹³C-NMR Signal Assignments of Compounds 1–3 in DMSO-d₆ at 298 K

1			2			3
Residue	δH [int., mult., J (Hz)]	δC	Residue	δH [int., mult., J (Hz)]	δC	Residue	δH [int., mult., J (Hz)]	δC
Pro1			Pro1			Pro1
α-CH	4.68 (1H, d, 6.5)	58.3	α-CH	4.03 (1H, d, 8.0)	59.8	α-CH	4.42 (1H, m)	61.4
β-CH2	1.99 (1H, m)	30.7	β-CH2	1.98 (1H, m)	31.0	β-CH2	1.71 (1H, m)	28.9
	2.03 (1H, m)			2.03 (1H, m)			2.16 (1H, m)
γ-CH2	1.80 (1H, m)	21.7	γ-CH2	1.01 (1H, m)	21.4	γ-CH2	1.82 (1H, m)	24.6
	1.95 (1H, m)			1.61 (1H, m)			1.92 (1H, m)
δ-CH2	3.41 (1H, m)	46.2	δ-CH2	3.15 (1H, m)	46.4	δ-CH2	3.48 (1H, m)	47.2
	3.41 (1H, m)			3.30 (1H, m)			3.56 (1H, m)
CO		171.3	CO		171.7	CO		173.7
Tyr2			Phe2			Gly2
α-CH	4.03 (1H, m)	56.4	α-CH	4.28 (1H, m)	55.1	α-CH2	3.59 (1H, m)	42.9
β-CH2	2.96 (1H, m)	34.5	β-CH2	2.63 (1H, d, 12.0)	34.2		3.62 (1H, m)
	3.05 (1H, m)			2.86 (1H, t, 13.5)		NH	8.70 (1H, s)
i-C		127.9	i-C		138.8	CO		168.3
o-CH	6.99 (2H, d, 9.0)	129.3	o-CH	7.35 (2H, d, 8)	129.1	Ile3
m-CH	6.66 (2H, d, 9.0)	115.0	m-CH	7.24 (2H, t, 8)	127.7	α-CH	4.32 (1H, m)	50.2
p-CH		155.8	p-CH	7.16 (1H, t, 8)	126.1	β-CH	1.48 (1H, m)	39.2
NH	8.05 (1H, m)		NH	8.73 (1H, d, 8.0)		γ-CH2	1.44 (1H, m)	24.1
CO		171.0	CO		168.7		1.65 (1H, m)
Ser3			Pro3			γ-CH3	0.85 a)	20.9
α-CH	4.20 (1H, m)	56.0	α-CH	4.63 (1H, t, 7.0)	57.4	δ-CH3	0.79 (3H, d, 6.5)	20.9
β-CH2	3.83 (1H, d, 7.5)	61.7	β-CH2	2.02 (1H, m)	28.5	NH	7.69 (1H, d, 9.5)
	3.98 (1H, d, 7.5)			2.32 (1H, m)		CO		171.4
NH	8.04 (1H, m)		γ-CH2	1.97 (1H, m)	25.5	Phe4
CO		170.0		2.14 (1H, m)		α-CH	3.96 (1H, m)	55.8
Phe4			δ-CH2	3.91 (1H, m)	46.8	β-CH2	3.18 (1H, m)	33.9
α-CH	4.49 (1H, s)	53.7		3.62 (1H, m)			3.30 (1H, m)
β-CH2	2.83 (1H, t, 12.5)	36.8	CO		173.7	i-C		139.0.
	3.25 (1H, t, 12.5)		Pro4			o-CH	7.15 (2H, d, 8.0)	128.7
i-C		138.0	α-CH	5.02 (1H, t, 9.0)	61.4	m-CH	7.25 (2H, t, 8.0)	128.1
o-CH	7.28 (2H, m)	128.1	β-CH2	1.82 (1H, m)	29.1	p-CH	7.17 (1H, t, 8.0)	125.9
m-CH	7.28 (2H, m)	128.0		2.22 (1H, m)		NH	7.92 (1H, s)
p-CH	7.18 (1H, m)	126.3	γ-CH2	1.92 (1H, m)	24.9	CO		170.3
NH	8.01 (1H, m)			1.97 (1H, m)		Thr5
CO		169.7	δ-CH2	4.28 (1H, m)	48.1	α-CH	4.70 (1H, d, 7.5)	56.4
Phe5				3.82 (1H, m)		β-CH	4.40 (1H, m)	67.7
α-CH	4.40 (1H, s)	56.0	CO		174.4	γ-CH3	1.05 (3H, d, 6.0)	19.3
β-CH2	2.93 (1H, m)	36.4	Ser5			NH	7.25 (1H, s)
	3.20 (1H, m)		α-CH	3.94 (1H, m)	58.1	CO		170.5
i-C		138.2	β-CH2	3.74 (2H, br s)	39.7	Ile6
o-CH	7.23 (2H, m)	128.9	NH	8.29 (1H, s)		α-CH	3.83 (1H, m)	59.5
m-CH	7.23 (2H, m)	128.8	CO		170.6	β-CH	1.79 (1H, m)	35.3
p-CH	7.18 (1H, m)	126.1	Thr6			γ-CH2	1.18 (1H, m)	24.8
NH	8.51 (1H, d, 7)		α-CH	4.02 (1H, m)	59.9		1.48 (1H, m)
CO		171.1	β-CH	4.15 (1H, m)	67.9	γ-CH3	0.85a)	11.1
Val6			γ-CH3	1.25 (3H, d, 6.5)	19.5	δ-CH3	0.85a)	15.3
α-CH	3.95 (1H, m)	59.2	NH	7.78 (1H, d, 7.5)		NH	7.98 (1H, s)
β-CH	1.67 (1H, m)	30.3	CO		171.1	CO		171.1
γ-CH3	0.45 (3H, d, 6.5)	19.3	Gly7			Ile7
γ-CH3	0.56 (3H, d, 6.5)	18.0	α-CH2	3.25 (1H, m)	42.1	α-CH	4.05 (1H, m)	52.7
NH	7.85 (1H, d, 6)			4.03 (1H, m)		β-CH	1.62 (1H, m)	39.3
CO		170.6	NH	7.85 (1H, d, 7.5)		γ-CH2	1.47 (1H, m)	24.3
Ile7			CO		167.5		1.49 (1H, m)
α-CH	4.12 (1H, t, 7.5)	56.8	Leu8			γ-CH3	0.83a)	22.9
β-CH	1.62 (1H, m)	36.4	α-CH	4.45 (1H, t, 10.5)	48.0	δ-CH3	0.90 (3H, d, 6.0)	22.7
γ-CH2	1.07 (1H, m)	24.5	β-CH2	1.02 (1H, m)	39.7	NH	7.47 (1H, s)
	1.50 (1H, m)			1.25 (1H, m)		CO		171.7
γ-CH3	0.83 (3H, t, 7)	10.9	γ-CH	1.25 (1H, m)	23.7	Thr8
δ-CH3	0.80 (3H, d, 7.5)	16.0	δ-CH3	0.18 (3H, d, 6.5)	22.1	α-CH	4.44 (1H, m)	54.8
NH	7.12 (1H, d, 6)		δ-CH3	0.55 (3H, d, 6.5)	20.5	β-CH	3.89 (1H, q, 5.5)	66.3
CO		170.9	NH	6.92 (1H, d, 9.0)		γ-CH3	0.98 (3H, d, 6.5)	20.0
Gln8			CO		168.7	NH	7.18 (1H, s)
α-CH	4.22 (1H, m)	50.2	Pro9			CO		170.1
β-CH2	1.69 (1H, m)	27.3	α-CH	4.40 (1H, dd, 4.0, 9.0)	57.9
	1.82 (1H, m)		β-CH2	2.03 (1H, m)	29.1
γ-CH2	2.12 (1H, m)	31.1		2.14 (1H, m)
	2.27 (1H, m)		γ-CH2	1.82 (1H, m)	24.4
δ-CO		174.7		1.92 (1H, m)
NH	8.46 (1H, s)		δ-CH2	3.35 (1H, m)	46.3
NH2	6.89 (1H, s)			3.52 (1H, m)
	7.07 (1H, s)		CO		172.3
CO		171.2	Ile10
			α-CH	4.60 (1H, d, 11.0)	51.2
			β-CH	1.49 (1H, t, 11.5)	38.4
			γ-CH2	1.81 (1H, m)	24.4
				1.91 (1H, m)
			γ-CH3	0.92 (3H, t, 5.0)	23.5
			δ-CH3	1.25 (3H, d, 6.5)	20.4
			NH	8.10 (1H, d, 3)
			CO		171.0

a) Signal pattern unclear due to overlapped.

Table 2. ¹H- and ¹³C-NMR Signal Assignments of Compounds 4–6 in DMSO-d₆ at 298 K

4			5			6
Residue	δH [int., mult., J (Hz)]	δC	Residue	δH [int., mult., J (Hz)]	δC	Residue	δH [int., mult., J (Hz)]	δC
Pro1			Pro1			Pro1
α-CH	4.40 (1H, t, 9.0)	61.4	α-CH	4.08 (1H, m)	60.9	α-CH	4.18 (1H, m)	60.8
β-CH2	1.75 (1H, m)	28.9	β-CH2	1.75 (1H, m)	28.7	β-CH2	1.79 (1H, m)	28.8
	2.16 (1H, m)			2.12 (1H, m)			2.13 (1H, m)
γ-CH2	1.85 (1H, m)	24.7	γ-CH2	1.86 (1H, m)	24.5	γ-CH2	1.88 (1H, m)	24.7
	1.95 (1H, m)			1.93 (1H, m)			2.04 (1H, m)
δ-CH2	3.55 (1H, m)	47.3	δ-CH2	3.50 (1H, m)	45.7	δ-CH2	3.57 (1H, m)	47.6
	3.59 (1H, m)			3.30 (1H, m)			3.88 (1H, m)
CO		173.5	CO		171.6	CO		171.4
Gly2			Gly2			Gly2
α-CH2	3.64 (1H, m)	43.0	α-CH2	3.55 (1H, m)	42.3	α-CH2	3.25 (1H, m)	42.3
	3.75 (1H, m)						3.97 (1H, m)
NH	8.80 (1H, s)		NH	8.49 (1H, m)		NH	8.76 (1H, s)
CO		168.5	CO		168.3	CO		169.2
Leu3			Leu3			Phe3
α-CH	4.49 (1H, m)	50.3	α-CH	4.42 (1H, m)	51.0	α-CH	4.53 (1H, m)	55.3
β-CH2	1.49 (1H, m)	39.1	β-CH2	1.28 (1H, m)	40.0	β-CH2	2.78 (1H, m)	39.1
	1.69 (1H, m)			1.53 (1H, m)			2.97 (1H, m)
γ-CH	1.48 (1H, m)	24.3	γ-CH	1.42 (1H, m)	24.5	i-C		137.2
δ-CH3	0.81 (3H, d, 8.5)	21.1	δ-CH3	0.79 (3H, d, 7.5)	21.8	o-CH	7.20–7.26 (1H, m)	128.1
δ-CH3	0.83 (3H, d, 8.5)	23.0	δ-CH3	0.85 (3H, d, 7.5)	23.2	m-CH	7.20–7.26 (1H, m)	128.0
NH	7.73 (1H, d, 9.5)		NH	7.43 (1H, d, 7)		p-CH	7.17–7.25 (1H, m)	126.3
CO		170.5	CO		171.6	NH	8.08 (1H, d, 10.0)
Ser4			Val4			CO		169.9
α-CH	3.83 (1H, m)	56.8	α-CH	4.05 (1H, m)	58.0	Asp4
β-CH2	3.84 (1H, m)	59.5	β-CH	1.98 (1H, m)	29.4	α-CH	4.16 (1H, m)	50.3
	3.87 (1H, m)		γ-CH3	0.88 (3H, d, 6.5)	18.3	β-CH2	2.72 (1H, m)	35.1
NH	7.91 (1H, s)		γ-CH3	0.99 (3H, d, 6.5)	19.0		2.98 (1H, m)
CO		170.0	NH	8.90 (1H, d, 3.5)		NH	8.47 (1H, d, 5.0)
Thr5			CO		170.4	γ-CO		172.8
α-CH	4.45 (1H, m)	56.8	Pro5			CO		170.0
β-CH	4.33(1H, m)	67.3	α-CH	4.74 (1H, d, 7.5)	60.6	Phe5
γ-CH3	1.03 (3H, d, 6.5)	19.5	β-CH2	1.74 (1H, m)	29.6	α-CH	4.22 (1H, m)	56.6
NH	7.34 (1H, d, 8.5)			2.42 (1H, dd, 6.5, 12)		β-CH2	2.98 (1H, m)	36.1
CO		171.0	γ-CH2	1.50 (1H, m)	24.4		3.05 (1H, m)
Ile6				1.87 (1H, m)		i-C		137.7
α-CH	3.88 (1H, m)	59.1	δ-CH2	3.18 (1H, m)	45.3	o-CH	7.15–7.25 (1H, m)	128.8
β-CH	1.79 (1H, m)	35.7		3.56 (1H, m)		m-CH	7.15–7.25 (1H, m)	129.0
γ-CH2	1.12 (1H, m)	24.6	CO		170.1	p-CH	7.17–7.25 (1H, m)	126.3
	1.47 (1H, m)		Ile6			NH	7.93 (1H, s)
γ-CH3	0.85 (3H, t, 6.5)	11.4	α-CH	3.97 (1H, m)	59.6	CO		170.8
δ-CH3	0.82 (3H, d, 8)	15.4	β-CH	2.02 (1H, m)	35.5	Ile6
NH	7.78 (1H, s)		γ-CH2	1.12 (1H, m)	24.9	α-CH	3.85 (1H, m)	59.6
CO		170.0		1.42 (1H, m)		β-CH	1.90 (1H, m)	35.4
Leu7			γ-CH3	0.77 (3H, d, 7.0)	10.5	γ-CH2	1.01 (1H, m)	24.7
α-CH	4.03 (1H, m)	52.7	δ-CH3	0.87 (3H, d, 6.5)	15.7		1.34 (1H, m)
β-CH2	1.46 (1H, m)	39.0	NH	8.47 (1H, m)		γ-CH3	0.79 (3H, t, 6)	10.6
	1.60 (1H, m)		CO		170.9	δ-CH3	0.80 (3H, 5.5)	15.3
γ-CH	1.59 (1H, m)	24.1	Gly7			NH	7.70 (1H, m)
δ-CH3	0.80 (3H, d, 6.5)	21.1	α-CH2	3.93 (1H, m)	42.5	CO		170.1
δ-CH3	0.90 (3H, d, 6.5)	22.7		3.98 (1H, m)		Leu7
NH	7.43 (d, 6.5)		NH	7.60 (1H, s)		α-CH	4.79 (1H, m)	47.8
CO		171.0	CO		167.5	β-CH2	1.42 (1H, m)	40.8
Thr8							1.58 (1H, m)
α-CH	4.45 (1H, m)	55.1				γ-CH	1.57 (1H, m)	23.9
β-CH	3.87 (1H, m)	66.5				δ-CH3	0.78 (3H, d, 6)	22.0
γ-CH3	1.00 (3H, d, 6.5)	19.9				δ-CH3	0.82 (3H, d, 6)	22.6
NH	7.25 (1H, d, 8.5)					NH	7.65 (1H, m)
CO		171.0				CO		170.0

Gypsophin B (2) was isolated as an amorphous solid. It showed a quasimolecular ion at m/z 1029.5385 [M+Na]⁺ in HR-ESI-MS spectrum, in conjunction with the NMR experiments, suggesting the molecular formula C₅₀H₇₄N₁₀O₁₂Na (Calcd for 1029.5380) with 19 degrees of unsaturation. The IR spectrum showed absorption bands of amine (3419 cm⁻¹), amide carbonyl (1644 cm⁻¹) and aromatic (1546 cm⁻¹) groups. The ¹H-NMR spectrum showed six amide NH resonances, and the ¹³C-NMR spectrum displayed ten amide carbonyl carbons. These data were indicative of a peptide. Using ROESY, TOCSY, HMQC, and HMBC spectra, ten amino acid residues were identified as Phe, Ser, Leu (leucine), Thr (threonine), Gly (glycine), Ile and Pro (×4) (Fig. 1, Table 1), which was further confirmed by HPLC analysis of the acid hydrolysates. These residues accounted for 18 degrees of unsaturation, suggesting that 2 is also a cyclic decapeptide. The sequence of 2 was assembled with HMBC and ROESY experiments. The HMBC spectrum showed correlations of Phe²-NH (δ_H 8.73)/Pro¹-CO (δ_C 171.7), Ser⁵-NH (δ_H 8.29)/Pro⁴-CO (δ_C 174.4), Thr⁶-NH (δ_H 7.78)/Ser⁵-CO (δ_C 170.6), Gly⁷-NH (δ_H 7.85)/Thr⁶-CO (δ_C 171.1), Leu⁸-NH (δ_H 6.92)/Gly⁷-CO (δ_C 167.5) and Ile¹⁰-NH (δ_H 8.10)/Pro⁹-CO (δ_C 172.3) (Fig. 2), corresponding to the peptide fragments Pro¹-Phe², Pro⁴-Ser⁵-Thr⁶-Gly⁷-Leu⁸ and Pro⁹-Ile¹⁰. The observed ROESY correlations of Phe²-Hα/Pro³-Hα, Pro³-Hα/Pro⁴-Hγ, Leu⁸-Hα/Pro⁹-Hγ, Ile¹⁰-Hα/Pro¹-Hα and Pro⁴-Hα/Ser⁵-NH indicated the three peptide units Phe²-Pro³-Pro⁴, Leu⁸-Pro⁹ and Ile¹⁰-Pro¹ (Fig. 2). Therefore, the structure of 2 was elucidated as cyclo-(Pro¹-Phe²-Pro³-Pro⁴-Ser⁵-Thr⁶-Gly⁷-Leu⁸-Pro⁹-Ile¹⁰) and named gypsophin B. The differences in the ¹³C-NMR chemical shifts of Pro¹ (Δδ_Cβ–Cγ=9.6 ppm), Pro³ (Δδ_Cβ–Cγ=3.0 ppm), Pro⁴ (Δδ_Cβ–Cγ=4.2 ppm) and Pro⁹ (Δδ_Cβ–Cγ=4.7 ppm) suggested that the amide bonds in the Pro³, Pro⁴ and Pro⁹ residues were trans, and that in the Pro¹ was cis.^20,21)

Gypsophin C (3), a colorless solid, had a molecular formula C₄₂H₆₆N₈O₁₀, established by positive HR-ESI-MS (m/z 865.4814 [M+Na]⁺; Calcd 865.4818). It was identified as a peptide on the basis of the presence of seven amide protons (δ_H 7.18–8.70), eight carbonyls (δ_C 173.7–168.3) and eight α-amino acid carbon resonances (δ_C 61.4–42.9) in the ¹H- and ¹³C-NMR spectrum. A negative response to the ninhydrin reagent but a positive response after its hydrolysis with 6 n aqueous HCl implied that 3 is a cyclic peptide. Eight amino acid residues of 3 were determined as Pro¹, Gly², Ile³, Phe⁴, Thr⁵, Ile⁶, Ile⁷ and Thr⁸ on the basis of detailed analysis of one-dimensional (1D)- and two-dimensional (2D)-NMR, and their proton and carbon resonances were completely assigned (Fig. 1, Table 1). The amino acid sequence of 3 was established by HMBC cross-peaks: Gly²-NH (δ_H 8.70)/Pro¹-CO (δ_C 173.7), Ile³-NH (δ_H 7.69)/Gly²-CO (δ_C 168.3), Phe⁴-NH (δ_H 7.92)/Ile³-CO (δ_C 171.4), Thr⁵-NH (δ_H 7.25)/Phe⁴-CO (δ_C 170.3), Ile⁶-NH (δ_H 7.98)/Thr⁵-CO (δ_C 170.5), Ile⁷-NH (δ_H 7.47)/Ile⁶-CO (δ_C 171.1) and Thr⁸-NH (δ_H 7.18)/Ile⁷-CO (δ_C 171.7). The gross structure was further supported by ROESY data, as indicated in Fig. 2. Thus, 3 was determined as cyclo-(Pro¹-Gly²-Ile³-Phe⁴-Thr⁵-Ile⁶-Ile⁷-Thr⁸) and named gypsophin C. Furthermore, the ¹³C-NMR chemical shifts difference of Pro¹ (Δδ_Cβ–Cγ=4.3 ppm) indicated that the amide bond in the Pro¹ residue was trans.^20,21)

Gypsophin D (4), a colorless solid, had a molecular formula of C₃₆H₆₂N₈O₁₁, established by HR-ESI-MS (m/z 805.4423 [M+Na]⁺; Calcd 805.4430). The ¹H- and ¹³C-NMR data were similar to those of compound 3. The octapeptide skeleton was suggested by the presence of seven amide protons (δ_H 7.25–8.80) and eight amide carbonyl resonances (δ_C 173.5–168.5) in the ¹H- and ¹³C-NMR spectra. Eight amino acid residues were determined as Pro¹, Gly², Leu³, Ser⁴, Thr⁵, Ile⁶, Leu⁷ and Thr⁸ on the basis of detailed analysis of 1D- and 2D-NMR. The amino acid sequence of 4 was established by HMBC and ROESY data. HMBC correlations were observed between Leu³-NH (δ_H 7.73)/Gly²-CO (δ_C 168.5), Thr⁵-NH (δ_H 7.34)/Ser⁴-CO (δ_C 170.0), Thr⁸-NH (δ_H 7.25)/Leu⁷-CO (δ_C 171.0), indicating the peptide fragments Gly²-Leu³, Ser⁴-Thr⁵ and Leu⁷-Thr⁸. Correlations in the ROESY spectra of Thr⁸-Hα/Pro¹-Hγ, Pro¹-Hα/Gly²-NH, Leu³-NH/Ser⁴-NH and Thr⁵-Hα/Ile⁶-NH indicated the three peptide units Thr⁸-Pro¹-Gly², Leu³-Ser⁴ and Thr⁵-Ile⁶. Thus, 4 was determined as cyclo-(Pro¹-Gly²-Leu³-Ser⁴-Thr⁵-Ile⁶-Leu⁷-Thr⁸) and named gypsophin D. The trans configuration of the amide bond in the Pro¹ residue was deduced from chemical shift differences in Δδ_Cβ–Cγ value (4.2 ppm).^20,21)

Gypsophin E (5) was isolated as amorphous powder with a molecular formula of C₃₁H₅₁N₇O₇ based on the pseudomolecular ion peak at m/z 656.3745 [M+Na]⁺ (Calcd 656.3742) in HR-ESI-MS spectrum. The peptidic nature was suggested by the presence of five amide protons and seven amide carbonyl signals in the ¹H- and ¹³C-NMR spectra. Analysis of 2D-NMR data (TOCSY, HMQC, HMBC and ROESY spectra) permitted the assignment of the ¹H- and ¹³C-NMR resonances and also allowed the diagnosis of the seven amino acid residues as Pro (×2), Gly (×2), Leu, Ile and Val. The HMBC spectrum showed correlations of Gly²-NH/Pro¹-CO, Leu³-NH/Gly²-CO, and Ile⁶-NH/Pro⁵-CO, corresponding to the peptide fragment Pro¹-Gly²-Leu³ and Pro⁵-Ile⁶, respectively. The observed ROESY correlations of Pro¹-Hα (δ_H 4.08) and the Gly⁷ α-methylene resonance at δ 3.98,²²⁾ Val⁴-Hα/Pro⁵-Hα, Ile⁶-NH/Gly⁷-NH suggested the fragment Pro¹-Gly⁷-Ile⁶, Val⁴-Pro⁵, respectively. These data implied that the sequence of 5 is cyclo-(Pro¹-Gly²-Leu³-Val⁴-Pro⁵-Ile⁶-Gly⁷), named gypsophin E. The trans configuration of the amide bonds in the two Pro residues was deduced from their chemical shift differences in Δδ_Cβ–Cγ values (4.2 ppm for Pro¹ and 5.2 ppm for Pro⁵).^20,21)

Gypsophin F (6) was isolated as amorphous powder. Its molecular formula was determined to be C₄₁H₅₅N₇O₉ by the HR-ESI-MS at m/z 788.3999 [M−H]⁻ (Calcd 788.3988), indicating 18 degrees of unsaturation. The ¹H- and ¹³C-NMR were similar to those of gypsophin.¹⁸⁾ However, the HR-ESI-MS showed that the molecular weight of compound 6 was 18 amu more than that of gypsophin. The C–H long range correlation of Hα of Phe⁵ to the γ-carbonyl carbon of Asp⁴ (aspartic acid) observed in HMBC spectrum of gypsophin was not detected in compound 6. Alternatively, the ¹H-NMR spectrum of compound 6 in DMSO-d₆ revealed an additional NH proton, which was attributed to Asp residue by HMBC experiment. These data indicated that the dehydrated Asp unit in gypsophin was liberated in compound 6. Therefore, the structure compound 6 was elucidated as cyclo-(Pro¹-Gly²-Phe³-Asp⁴-Phe⁵-Ile⁶-Leu⁷) and named gypsophin F.

The absolute configurations of all the amino acids in 1–6 were determined to be l-configuration by slightly modified Marfey’s method.²³⁾ The hydrolysates were treated with Marfey’s reagent, 1-fluoro-2,4-dinitrophenyl-5-l-alaninamide (FDAA), and the resulting derivatives were analyzed by reversed-phase HPLC. The peaks on the chromatogram were identified by comparing the retention times with those of the FDAA derivatives of the authentic amino acids.

Since some cyclopeptides exhibited antiplatelet aggregation activity, the isolates were evaluated for their effects against ADP-induced platelet aggregation. As shown in Table 3, compound 3, gypsophin C, showed moderate inhibition with 30.80% inhibitory rate at the concentration of 100 µg mL⁻¹. The other compounds exhibited weak inhibitory rate between 8% and 15%. Aspirin was used as a positive control with 38.78% inhibitory rate at the same concentration. In addition, the cytotoxicity of compounds 1–6 against four human cancer cell lines Bel7402 (human liver cancer), SMMC-7721 (human liver cancer), HCT 116 (human colon cancer) and H460 (human lung cancer) was tested using the methyl thiazol tetrazolium (MTT) method.²⁴⁾ All of the compounds showed no cytotoxic (IC₅₀ values >100 µm) against the tested tumor cells.

Table 3. Antiplatelet Aggregation Activity of Compounds 1–6 (x̄±s, n=3)

Compounds	Dosage (µg mL−1)	Inhibition ratio (%)
1	100	8.07±0.87
2	100	14.10±0.85
3	100	30.80±0.47
4	100	12.91±1.25
5	100	12.17±0.88
6	100	15.32±0.83
Aspirin (positive control)	100	38.78±0.45

Experimental

General Experimental Procedures

IR (KBr-disc) spectra were recorded by Bruker Tensor 27 spectrometer. NMR spectra were recorded on Bruker Avance-III NMR instrument (¹H: 500 MHz, ¹³C: 125 MHz) with TMS as internal standard. Chemical shifts were given in values of ppm and coupling constants in Hertz. HR-ESI-MS were obtained on an Agilent 6520B Q-TOF spectrometer (Agilent, CA, U.S.A.). Semi-preparative HPLC was carried out on a Shimadzu LC-8A instrument (flow rate set at 10 mL/min) with a SPD-20A detector using a Shim-pack PRC-ODS column (5 µm, 250×20 mm). Silica gel (100–200 mesh and 200–300 mesh, Qingdao Haiyang Chemical Co., Ltd., China), Sephadex LH-20 (Pharmacia, U.S.A.), and RP-C₁₈ (40–63 µm, YMC, Japan) were used for column chromatography. All solvents used were of analytical grade (Jiangsu Hanbon Science and Technology Co., Ltd., China).

Plant Material

The roots of G. oldhamiana were collected in Lianyungang City, Jiangsu Province, China, and identified by Prof. Minjian Qin, the Department of Natural Medicinal Chemistry, China Pharmaceutical University, Nanjing, China. A voucher specimen (No. 20110616) was deposited in the Department of Natural Medicinal Chemistry, China Pharmaceutical University.

Extraction and Isolation

The air-dried roots (4.5 kg) were refluxed with 80% EtOH. The residue obtained by concentrating alcohol was partitioned between H₂O and EtOAc, and then the EtOAc-soluble portion (20 g) was subjected to a silica gel column (CHCl₃–MeOH, 25 : 1–1 : 1), in which a fraction eluted with CHCl₃–MeOH (10 : 1) was further separated by RP-C18 column (MeOH–H₂O, 6 : 4–8 : 2). The fraction (MeOH–H₂O, 6 : 4) was further separated by HPLC (MeOH–H₂O, 62 : 38, UV detection at 210 nm), to yield pure 1 (10 mg, t_R=20 min), and 2 (12 mg, t_R=26.2 min), respectively. The fraction (MeOH–H₂O, 8 : 2) was subjected to a silica gel column (200–300 mesh), which was eluted with CHCl₃–MeOH (10 : 1, 5 : 1, 1 : 1) to give three fractions (fractions 1–3). The fraction 1 was further separated by HPLC (MeOH–H₂O, 70 : 30, UV detection at 210 nm), and yielded pure 3 (10 mg, t_R=15 min), 4 (17 mg, t_R=21 min), and 5 (12 mg, t_R=28.5 min), respectively. Fraction 2 was subjected to HPLC (MeOH–H₂O, 80 : 20) to afford 6 (10 mg, t_R=25 min).

Gypsophin A (1): White powder; [α]_D²³ −77.2 (c=0.192, MeOH); IR (KBr) ν_max 3408, 2965, 1651, 1516, 1452, 1244, 701 cm⁻¹; ¹H- and ¹³C-NMR see Table 1; ESI-MS (positive) m/z 982 [M+H]⁺; HR-ESI-MS (positive) m/z 982.5034 [M+H]⁺ (Calcd for C₅₁H₆₈N₉O₁₁, 983.5033).

Gypsophin B (2): Amorphous solid, [α]_D²³ −30.1 (c=0.300, MeOH); IR (KBr) ν_max 3419, 2954, 1644, 1546, 1450, 1204, 1027 cm⁻¹; ¹H- and ¹³C-NMR see Table 1; ESI-MS (positive) m/z 1029 [M+Na]⁺; HR-ESI-MS (positive) m/z 1029.5385 [M+Na]⁺ (Calcd for C₅₀H₇₄N₁₀O₁₂Na, 1029.5380).

Gypsophin D (3): Colorless solid; [α]_D²³ −30.1 (c=0.158, MeOH); IR (KBr) ν_max 3410, 2962, 1645, 1530, 1455, 1224 cm⁻¹; ¹H- and ¹³C-NMR see Table 1; ESI-MS (positive) m/z 843 [M+H]⁺; HR-ESI-MS (positive) m/z 865.4814 [M+Na]⁺ (Calcd for C₄₂H₆₆N₈O₁₀Na, 865.4818).

Gypsophin E (4): Colorless solid; [α]_D²³ −31.3 (c=0.142, MeOH); IR (KBr) ν_max 3414, 2963, 1640, 1530, 1400, 1246 cm⁻¹; ¹H- and ¹³C-NMR see Table 2; ESI-MS (positive) m/z 783 [M+H]⁺; HR-ESI-MS (positive) m/z 805.4423 [M+Na]⁺ (Calcd for C₃₆H₆₂N₈O₁₁Na, 805.4430).

Gypsophin E (5): Amorphous powder; [α]_D²³ −63.3 (c=0.140, MeOH); IR (KBr) ν_max 3406, 2962, 1652 cm⁻¹; ¹H- and ¹³C-NMR see Table 2; ESI-MS (positive) m/z 634 [M+H]⁺; HR-ESI-MS (positive) m/z 656.3745 [M+Na]⁺ (Calcd for C₅₁H₆₆N₈O₁₂Na, 656.3742).

Gypsophin F (6): Amorphous powder; [α]_D²³ −31.5 (c=0.164, MeOH); IR (KBr) ν_max 3442, 2962, 1626, 1527, 1455, 1384 cm⁻¹; ¹H- and ¹³C-NMR see Table 2; ESI-MS (positive) m/z 790 [M+H]⁺; HR-ESI-MS (negative) m/z 788.3999 [M−H]⁻ (Calcd for C₄₁H₅₄N₇O₉, 788.3988).

Marfey’s Derivitization and HPLC Analysis of 1–6

Each compound (0.5 mg) was dissolved in 6 n HCl (1 mL) in a sealed container and heated at 110°C for 24 h. After cooling, the reaction mixture was concentrated in vacuo to dryness. The hydrolysate was added to 20 µL of 1 m NaHCO₃ solution and 100 µL of 1% 1-fluoro-2, 4-dinitrophenyl-5-l-alaninamide in acetone. The solution was reacted at 40°C for 1 h. The Marfey’s derivatives were analyzed by co-injection into an HPLC apparatus (4.6×250 mm C₁₈ column (5 µm); MeCN–H₂O (0.05% TFA)=32–45%; flow rate 1.0 mL/min; UV detection at 340 nm) and compared with the Marfey’s derivatives of authentic amino acids.²³⁾ The results showed that all the amino acids had l-configurations, except Gly. The standards derivatives gave the following t_R values (min): 5.67 (l-Glu-l-FDAA), 5.85 (l-Ser-l-FDAA), 6.40 (l-Asp-l-FDAA), 6.49 (l-Thr-l-FDAA), 8.17 (Gly-l-FDAA), 10.29 (l-Pro-l-FDAA), 12.29 (l-Tyr-l-FDAA), 16.29 (l-Val-l-FDAA), 20.40 (l-Ile-l-FDAA), 21.32 (l-Phe-l-FDAA), 21.70 (l-Leu-l-FDAA).

Platelet Aggregation in Rabbit Platelet-Rich Plasma (PRP)

Rabbit common carotid artery was cut off to take a sample of blood, which was mixed with an anticoagulant (3.8% sodium citrate) in the proportion of 9 to 1, then centrifuged at 1000 rpm for 10 min at room temperature to give PRP. Centrifugation of the remaining blood at 3000 rpm for 10 min yielded platelet-poor plasma (PPP). Platelet count was adjusted to 3×10⁵ with PPP. The samples were added to the PRP (0.200 mL), then the mixture was incubated at 37°C with stirring for 1 min before the addition of ADP (10 µm) as inducer of platelet aggregation. Aggregation was measured by a turbidimetric method. Changes in light transmission caused by platelet clotting were measured using an aggregometer. Platelet aggregation was expressed as the percentage change with the difference of light transmittance between PRP and PPP as 100%. The same experiment was done for the positive control drug Aspirin and the blank solvent ethanol. Each analyte was tested three times, and an average value was applied. Antiplatelet activity was expressed as the percent inhibition of the control value.²⁵⁾

Acknowledgment

This research work was financially supported by the National Natural Science Foundation of China (81073009), a Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the Program for Changjiang Scholars and Innovative Research Team in University (PCSIRT-IRT1193).

References

• 1)  Hsieh P. W., Chang F. R., Wu C. C., Wu K. Y., Li C. M., Chen S. L., Wu Y. C., J. Nat. Prod., 67, 1522–1527 (2004).
• 2)  Wele A., Landon C., Labbe H., Vovelle F., Zhang Y. J., Bodo B., Tetrahedron, 60, 405–414 (2004).
• 3)  Morita H., Gonda A., Takeya K., Itokawa H., Bioorg. Med. Chem. Lett., 6, 767–770 (1996).
• 4)  Baraguey C., Blond A., Correia I., Pousset J. L., Bodo B., Auvin-Guette C., Tetrahedron Lett., 41, 325–329 (2000).
• 5)  Auvin-Guette C., Baraguey C., Blond A., Xavier H. S., Pousset J. L., Bodo B., Tetrahedron, 55, 11495–11510 (1999).
• 6)  Hsieh P. W., Chang F. R., Wu C. C., Wu K. Y., Li C. M., Wang W. Y., Gu L. C., Wu Y. C., Helv. Chim. Acta, 87, 57–66 (2004).
• 7)  Matsumoto T., Shishido A., Morita H., Itokawa H., Takeya K., Phytochemistry, 57, 251–260 (2001).
• 8)  Kosasi S., Sluis W. G., Boelens R., Hart L. A., Labadie R. P., FEBS Lett., 256, 91–96 (1989).
• 9)  Wang Y. C., Zhou J., Tan N. H., Ding Z. T., Jiang X., Acta Pharmacol. Sin., 34, 19–23 (1999).
• 10)  Morita H., Shishido A., Kayashita T., Takeya K., Itokawa H., J. Nat. Prod., 60, 404–407 (1997).
• 11)  Morita H., Gonda A., Takeya K., Itokawa H., Iitaka Y., Tetrahedron, 53, 1617–1626 (1997).
• 12)  Itokawa H., Yun Y. S., Morita H., Takeya K., Yamada K., Planta Med., 61, 561–562 (1995).
• 13)  Yun Y. S., Morita H., Takeya K., Itokawa H., J. Nat. Prod., 60, 216–218 (1997).
• 14)  “The Chinese Medicine Dictionary,’’ Shanghai People’s Publishing House, Shanghai, 1977, p. 2170.
• 15)  Liu Z., Li D., Owen N. L., Grant D. M., Cates R. G., Jia Z., J. Nat. Prod., 58, 1632–1635 (1995).
• 16)  Luo J. G., Kong L. Y., Helv. Chim. Acta, 89, 947–953 (2006).
• 17)  Luo J. G., Kong L. Y., Takaya Y., Niwa M., Chem. Pharm. Bull., 54, 1200–1202 (2006).
• 18)  Luo J. G., Wang X. B., Ma L., Kong L. Y., Bioorg. Med. Chem. Lett., 17, 4460–4463 (2007).
• 19)  Ribeir T. A. N., da Silva L. R., de Sousa Júnior P. T., Castro R. N., de Carvalho M. G., Helv. Chim. Acta, 95, 935–939 (2012).
• 20)  Schmidt G., Grube A., Kock M., Eur. J. Org. Chem., 24, 4103–4110 (2007).
• 21)  Siemion I. Z., Wieland T., Pook K. H., Angew. Chem. Int. Ed. Engl., 14, 702–703 (1975).
• 22)  Tong Y., Luo J. G., Wang R., Wang X. B., Kong L. Y., Bioorg. Med. Chem. Lett., 22, 1908–1911 (2012).
• 23)  Zhou J., Tan N. H., Chin. Sci. Bull., 45, 1825–1831 (2000).
• 24)  Lu Y. Y., Luo J. G., Huang X. F., Kong L. Y., Steroids, 74, 95–101 (2009).
• 25)  Iyobe A., Uchida M., Kamata K., Hotei Y., Kusama H., Harada H., Chem. Pharm. Bull., 49, 822–829 (2001).