Isolation and Identification of Secondary Metabolites From Streptomyces sp. SP301

Abstract

Seven compounds, namely aminobenzoates A to D (1-4), naphthalenecarboxylates A and B (5-6), and glycosylatelactone A (7), were isolated from the fermentation medium of Streptomyces sp. SP301. Of these, aminobenzoates C and D (3-4), naphthalenecarboxylate B (6), and glycosylatelactone A (7) are new compounds. Aminobenzoates A to D (1-4) shared a common aromatic starter unit, para-aminobenzoic acid , and biosynthesis involving a different pathway. The structures were elucidated on the basis of 1D- and 2D-Nuclear Magnetic Resonance (NMR) spectroscopy and HR-ESIMS analysis.

Keywords

aminobenzoates naphthalenecarboxylates glycosylatelactone

para-Aminobenzoic acid (PABA), a vital intermediate product in the folate synthesis of bacteria, fungi, and plants, is essential for cell growth and survival.¹ In 1940, Woods discovered that PABA was able to reverse the bacteriostatic action of sulfonamides, and this finding leads to the recognition of the biological importance of PABA.^2,3 Owing to the structure similarity and higher concentration, sulfonamides can replace PABA in the synthesis of dihydrofolate in vivo, which can block the synthesis of dihydrofolate and folate in microorganisms and terminate life prolongation.⁴ The potassium salt of PABA (under the trade name Potaba) was approved to treat certain fibrotic skin disorders by increasing oxygen consumption in the skin, such as scleroderma and Peyronie’s disease.^5,6 Therefore, it is necessary to continue exploring PABA analogs for adding more therapeutical options.

Strain SP301 was constructed for an initial purpose to obtain novel ansamycins from Streptomyces sp. S010 by ansa gene heterologous expression in Streptomyces sp. LZ35.⁷ Streptomyces sp. S010 and LZ35 were 3-amino-5-hydroxy benzoic acid positive strains which have the potential to produce new ansamycins. Our research group did a thorough study of strain LZ35, and a series of novel compounds were isolated from it.^8

-13 However, instead of obtaining new ansamycins from this effort, 7 compounds were isolated from Streptomyces sp. SP301 at the first time. Here, we showed the fermentation, isolation, and structure elucidation of 7 compounds, namely aminobenzoates A to D (1-4), naphthalenecarboxylates A and B (5-6), and glycosylatelactone A (7) from Streptomyces sp. SP301 (Figure 1). Their structures were elucidated on the basis of 1D- and 2D-Nuclear Magnetic Resonance (NMR) spectroscopic and High Resolution-Electrospray Ionization Mass Spectrometry (HR-ESIMS) analysis. A proposed biosynthetic pathway for the PABA derivatives is provided. Further, the antibacterial activities of compounds 1 to 7 were also evaluated.

Figure 1

Structures of compounds 1 to 7.

According to the literatures search and database comparison, we found that aminobenzoate A (1) was a known compound (named p-amino-α-ethyl-hydrocinnamic acid) used as a synthetic intermediate.¹⁴ Aminobenzoate B (2) and naphthalenecarboxylate A (5) were isolated from Streptomyces vinaceusdrappus NRRL2363¹⁵ and Streptomyces caniferus GUA-06-05-006A,¹⁶ respectively.

The high-resolution Mass Spectrometry of aminobenzoate C (3) gave a quasi-molecular ion at m/z 322.1649 [M+H]⁺ which is consistent with the molecular formula C₁₇H₂₃NO₅, and unsaturation of 7. Interpretation of the ¹H-, ¹³C-, and Heteronuclear Single Quantum Correlation (HSQC)-NMR data (Tables 1 and 2) reveals that 3 contains 17 carbon signals (2 methyls, 3 methylenes, 7 methines, and 5 quaternary carbons). The analysis of ¹H-¹H Correlation Spectroscopy (COSY) signals showed the existence of fragments [a: -CH (5)-CH (6)-; b: -CH₂ (7)-CH (8)-CH (9)-CH₂ (10)-CH (11)-CH (12)-; and c: -CH₂ (8a)-CH₃ (8b)]. According to the NMR data, 3 has the same original core structure as that of 2, except the substitution of acetyl and 2-ethyl-3-hydroxyl-5-ene-heptylic acid at C-4-N and C-1, respectively. The coupling constant (J > 15.0 Hz) between H-11 and H-12 demonstrates the configuration of the double bond between C-11 and C-12 as (E)-1-(8-ethyl-9-hydroxyl-11-ene-heptylic acid)-3-hydroxyl-4-acetyl-amidobenzene. The NOE correlations from H-9 to H-8a, and H-8b reveal that C-8-ethyl and C-9-OH are in the opposite side, thus we determined the threo stereochemistry of 3 at C-8/C-9. According to NMR data and molecular weight of 3, we determined the structure of 3.

Table 1

¹H NMR Spectroscopic Data of Compounds of 1 to 7 in CD₃OD (δ in ppm, J in Hz).

Pos.	1	2	3	4	5	6	7
2	7.01 (d, 8.5)	7.50 (m)	6.71 (d, 1.7)	7.71 (dd, 8.4, 1.6)	8.60 (d, 1.5)	8.88 (d, 1.4)
3	6.68 (d, 8.5)			6.30 (d, 8.5)
4					6.94 (m)	6.70 (s)
5	6.68 (d, 8.5)	7.98 (d, 8.8)	7.44 (d, 8.1)				6.37 (s)
6	7.01 (d, 8.5)	7.49 (m)	6.67 (dd, 8.2, 1.8)	7.57 (d, 1.4)
7	2.62 (m)2.78 (m)		2.47 (m)2.71 (dd, 13.8, 5.6)		8.17 (d, 8.0)	8.45 (d, 8.9)	2.51 (m)
8	2.43 (m)		1.65 (m)	3.13 (m)	8.38 (dd, 8.0, 1.7)	7.97 (dd, 8.9, 1.7)	1.08 (t, 7.4)
9			3.75 (m)	4.17 (dd, 8.4, 5.6)			2.60 (m)
10			2.40 (m)		2.19 (d, 1.6)	2.47 (s)	1.62 (m)1.74 (m)
11			7.04 (m)	α:2.05 (m)β:1.61 (m)			0.92 (t, 7.4)
12			5.90 (d, 15.7)	1.80 (m)2.13 (m)			1.24 (d, 6.9)
13				3.47 (d, 6.9)
15				1.25 (s)
16				1.30 (s)
17				0.98 (s)
8a	1.66 (m)		1.29 (m)1.51 (m)
8b	0.94 (t, 7.4)		0.91 (t, 7.4)
2-Ac-NH-4		2.20 (s)	2.16 (s)
1′		7.50 (m)				4.77 (d, 7.8)	5.15 (m)
2′						3.71 (m)	3.52 (m)
3′						3.50 (t, 9.0)	3.51 (m)
4′						3.64 (m)	3.63 (m)
5′						3.59 (d, 9.7)	4.01 (d, 9.6)

Table 2

¹³C NMR Spectroscopic Data of Compounds 1 to 7 in CD₃OD.

Pos.	1	2	3	4	5	6	7
1	131.9 s	127.9 s	140.8 s	117.8 s	136.0 s	120.4 s
2	130.9 d	117.4 d	118.1 d	132.7 d	128.1 d	126.9 d	167.8 s
3	116.0 d	148.4 s	149.8 s	104.1 d	185.6 s	152.5 s	110.1 s
4	156.7 s	131.9 s	124.8 s	157.5 s	136.8 d	112.3 d	166.2 s
5	116.0 d	122.0 d	124.0 d	128.7 s	150.1 s	133.2 s	97.7 d
6	130.9 d	122.5 d	121.4 d	126.7 d	186.1 s	143.7 s	169.1 s
7	39.1 t	169.8 s	35.6 t	170.6 s	127.7 d	123.9 d	17.7 t
8	51.9 d		48.7 d	29.3 t	135.2 d	126.4 d	12.8 q
9	181.1 s		72.0 d	67.6 d	168.3 s	170.4 s	41.4 d
10			37.9 t	71.7 s	16.4 q	17.9 q	28.5 t
11			148.4 d	38.1 t			11.9 q
12			124.4 d	22.2 t			18.4 q
13			169.8 s	58.6 d
14				76.8 s
15				27.9 q
16				28.4 q
17				19.4 q
2a					133.7 s	124.9 s
6a					136.8 s	132.3 s
8a	26.6 t		23.0 t
8b	12.3 q		11.9 q
1-Ac-NH-4		172.2 s	172.2 s
2-Ac-NH-4		23.9 q	23.3 q
1′						106.6 d	100.8 d
2′						75.5 d	74.4 d
3′						77.5 d	77.4 d
4′						73.3 d	72.8 d
5′						76.9 d	76.6 d
6′						173.0 s	173.1 s

Aminobenzoate D (4), with a quasi-molecular ion at m/z of 306.1700 [M+H]⁺, was determined to have the molecular formula C₁₇H₂₃NO₄, and unsaturation of 7. Interpretation of the ¹H NMR data (Tables 1 and 2) reveals that 4 contains 3 aromatic protons (δ_H = 6.30 (d, J = 8.5 Hz), 7.57 (d, J = 1.4 Hz), and 7.71 (dd, J = 8.4, 1.6 Hz)). Combined with ¹³C NMR, HSQC, and Heteronuclear Multiple Bond Correlation (HMBC) related signals, it could be determined that the carbon signals (δ_C= 104.1 d, 126.7 d, 132.7 d) connect with the above-mentioned aromatic protons. This evidence proves the existence of a trisubstituted benzene ring in compound 4. Three fragments [a: -CH₂ (8)-CH (9)-; b: -CH (2)-CH (3)-; and c: -CH₂ (11)-CH₂ (12)-CH (13)] were observed from the ¹H-¹H COSY signals of 4. Additionally, the HMBC correlations from H-3 to C-1 and C-5; H-6 to C-2, C-4, and C-7; H-8 to C-4, C-5, C-9, and C-10; H-13 to C-4, C-9, C-11, and C-12; H-17 to C-9, C-10, and C-11; and H-15 to C-13, C-14, and C-16 reveal the connection of each fragment. On the basis of its NMR data and molecular weight, the structure of compound 4 was determined. As we all know, the stable conformation of cyclohexane is chair conformation, and it is more stable when the substituents are at the e-bond. First, we defined the hydrogen at C-11 as H-11α (2.05, m) and H-11β (1.61, m). According to the NOE correlation signals H-9 to H-11α (2.05, m), H-15, H-16 and H-11β (1.61, m) to H-17 reveal that C-10-OH and C-12-CHOH(CH₃)₂ are at the same side. Also, on the base of chair-like conformation, it can determine that C-10-CH₃ and C-12-CHOH(CH₃)₂ are at the e-bond with stabilization. Thus, we confirmed the stereochemistry of 4 in Figure 1. Analyzing the structure of compound 4, we found that the biosynthesis of 4 is proposed to be produced by type I Polyketide Synthase (PKS) using PABA and monoterpene (Figure 1, green lines).

The HR-ESIMS of naphthalenecarboxylate (6) gave a quasi-molecular ion at m/z 395.0973 [M+H]⁺, consistent with the molecular formula C₁₈H₁₈O₁₀, and unsaturation of 10. From the ¹H NMR (δ_H = 4.77 (d, J = 7.8 Hz), 3.71 m, 3.50 (t, J = 9.0 Hz), 3.64 (m), 3.59 (d, J = 9.7 Hz)) and ¹³C NMR signals (δ_C = 106.6 d, 75.5 d, 77.5 d, 73.3 d, 76.9 d, 173.0 s), we could speculate the presence of glycosyl in 6. To identify the sugar part of 6, we carried out the acid hydrolysis reaction of compound 6 (methanol as solvent, 10% HCl as the catalyst, stirred at 80°C for 2 hours, detected by Thin Layer Chromatography [TLC]), but it was difficult to determine the product because of the limited yield of it. According to NMR data, we can confirm the structure of glycosyl. The correlation signals of ¹H-¹H COSY and HMBC further prove the existence of β-glucuronic acid. The Nuclear Overhauser Effect Spectroscopy correlations from H-1′ to H-3′, and H-5′ reveal the relative configurations of β-glucuronic acid in 6. According to the Haworth configuration of β-glucuronic acid, it can be converted into the Fisher projection. With reference to the configuration of glyceraldehyde, we can confirm the configuration of β-glucuronic acid in 6 as β-d-glucuronic acid. Simultaneously, the HMBC signal between terminal proton H-1′ and C-6 indicates that β-glucuronic acid is attached to C-6. Additionally, the HMBC correlations from H-2 to C-3, C-8, C-9, and C-6a; H-4 to C-3, C-2a, and C-6; H-7 to C-6, C-8, and C-2a; and H-10 to C-4, C-5, and C-6 revealed the connection of each fragment. According to the ¹H NMR, ¹³C NMR, ¹H-¹H COSY, HMBC, and NOE-NMR data and molecular weight, the complete structure of 6 is confirmed.

The molecular formula of glycosylatelactone A (7) was assigned as C₁₇H₂₄O₉ on the basis of high-resolution ESIMS data (m/z 373.1499 for [M+H]⁺) (ESI, Supplemental Figure S53). Interpretation of the NMR data (Tables 1 and 2) reveals that 7 contains 3 methyls, 2 methylenes, 7 methines (including 5 O-methines) and 5 quaternary carbons. The presence of glycosyl in this compound is determined from the ¹H NMR (δ_H = 5.15 m, 3.52 m, 3.51 m, 3.63 m, 4.01 (d, J = 9.6 Hz)) and ¹³C NMR signals (δ_C = 100.8 d, 74.4 d, 77.4 d, 72.8 d, 76.6 d, 173.1, s). We carried out the acid hydrolysis reaction of 7 as that of 6, but the product was present in low quantity. The correlation signals of ¹H-¹H COSY and HMBC further prove the existence of β-glucuronic acid. The HMBC signal between terminal proton H-1′ and C-4 indicates that β-glucuronic acid is attached to C-4. Two fragments [a: -CH₂ (7)-CH₃ (8) and b: CH₃ (12)-CH (9)-CH₂ (10)-CH₃ (11)] can be observed from ¹H-¹H COSY signals of the aglycone part of compound 7. The HMBC correlations from H-7 to C-2, C-3, C-4, and C-8; H-5 to C-3, C-4, and C-9; H-11 to C-9 and C-10; and H-12 to C-6, C-9, and C-10 reveal the connection of each fragment. The relative configuration of β-glucuronic acid is determined by the NOE correlation signals H-1′ to H-3′ and H-5′. Similarly, the configuration of 7 can be determined as β-d-glucuronic acid. The NOE correlation signal H-3′ to H-9 reveals the configuration at C-9. All of the above evidences confirmed the structure of compound 7.

The antimicrobial activities of compounds 1 to 7 against Staphylococcus aureus ATCC 25923, Mycobacterium smegmatis mc² 155, Pseudomonas aeruginosa PA01, and Proteus bacillus vulgaris CPCC 160013 were measured with a paper disc diffusion assay.¹⁷ Kanamycin was used as positive control. Unfortunately, compounds 1 to 7 did not show any antimicrobial activity.

In summary, we isolated and characterized 7 compounds from Streptomyces sp. SP301 strain, including aminobenzoates A to D (1-4), naphthalenecarboxylates A and B (5-6), and glycosylatelactone A (7). Among them, 4 compounds were first isolated (aminobenzoates C-D (3-4), naphthalenecarboxylate B (6), and glycosylatelactone A (7)).

Aminobenzoates A to D (1-4) are PABA derivatives with different extender units and various postmodifications. According to their structures, we proposed the biosynthetic pathway for PABA derivatives (Figure 2).

Figure 2

Proposed biosynthetic pathway for compounds 1 to 4.

Shikimic acid can be turned into PABA via the shikimate pathway, which is a significant metabolic pathway in plants, fungi, and microorganisms that can be used for the biosynthesis of aromatic compounds.¹⁸ Chorismate, the anionic form of chorismic acid, plays an important role as a branch point and synthetic precursor for the generation of most aromatic compounds in plants and microorganisms, such as various aromatic amino acids, quinones, salicylic acids, the folate precursor para-aminobenzoate, polyketones, polyenes, and terpenoids.¹⁹ Chorismate is transformed into PABA by the enzymes 4-amino-4-deoxychorismate synthase and 4-amino-4-deoxychorismate lyase.²⁰ para-Aminobenzoic acid is used as a common starter unit by aminobenzoates A to D (1-4). The formation of compounds 1 and 3 is proposed to be assembled by type I PKS with different extender units (ethylmalonyl CoA and malonyl CoA) and various postmodifications (eg, hydroxylation, acetylation, oxidation, and reduction). The combination between PABA and monoterpene is proposed to cause the formation of compound 4. As a starter unit, PABA can participate in different secondary metabolisms and form a variety of novel compounds.

Experimental

General

Optical rotations were carried out using an Anton Paar MCP200 automatic polarimeter. The UV spectra were obtained on a TU-1810 spectrophotometer (Beijing Purkinje General Instrument Co., Ltd.). NMR spectra were recorded on a Bruker DRX-600 NMR spectrometer (Bruker Daltonics Inc., Billerica, MA, United States) with tetramethylsilane as an internal standard. HR-ESIMS were measured on an LTQ-Orbitrap XL. Sephadex LH-20 was obtained from GE Amersham Biosciences (25-100 μm; Piscataway, NJ, United States) and LiChroprep RP-18 for column chromatography (CC) from Merck (40-63 μm; Darmstadt, Germany). Semipreparative High Performance Liquid Chromatography (HPLC) was performed on an Agilent 1200 equipped with a ZORBAX Eclipse XDC₁₈ 5 µm column (9.4 × 250 mm).

Strain, Fermentation, and Extraction

Streptomyces sp. SP301 was separated from No.1 soil of Nanjing arboretum, China. The mutant strain SP301 was cultured in 30 L ISP3 medium (oatmeal 30 g, saline salt 1 mL, agar 20 g, pH 7.2) for 10 days at 28°C. The cultures (a total volume of 30 L) were chopped, diced, and extracted 3 times overnight with an equal volume of EtOAc/MeOH/AcOH 80:15:5 (v/v/v) at room temperature and partitioned between ddH₂O and EtOAc until the EtOAc layer was colorless. Then, the EtOAc extract was dried with Na₂SO₄, and the solvent was removed under vacuum at 38°C. The EtOAc extract was partitioned with light petroleum (PE) and MeOH until the PE layer was colorless. The MeOH solution was concentrated under vacuum at 38°C to obtain the MeOH extract (7.5 g).

Isolation and Purification of Compounds 1 to 7

The MeOH extract (7.5 g) was subjected to MPLC over RP-18 silica gel (140 g), and 24 subfractions with 200 mL for each gradient were obtained from the elution with 30%, 50%, 70%, and 100% MeOH, respectively. In accordance with TLC results, 10 fractions, Frs. A to J, were obtained.

Fr. B (155 mg) was subjected to CC over Sephadex LH-20 (80 g) eluted with MeOH to obtain 7 fractions (Frs. B1-B7). Fr. B4 (38 mg) was purified by preparative HPLC (eluted with 12% acetonitrile, 4 mL/min, UV 320 nm) to yield 2 (2.2 mg). Fr. B5 (20 mg) was purified by preparative HPLC (eluted with 15% acetonitrile, 4 mL/min, UV 320 nm) to yield 6 (2.4 mg). Fr. C (330 mg) was subjected to CC over Sephadex LH-20 (80 g) eluted with MeOH to obtain 7 fractions (Frs. C1-C7). Fr. C2 (20 mg) was purified by preparative HPLC (eluted with 18% acetonitrile, 4 mL/min, UV 320 nm) to yield 7 (2.8 mg). Fr. C4 (32 mg) was purified by preparative HPLC (eluted with 15% acetonitrile, 4 mL/min, UV 320 nm) to yield 1 (2.9 mg). Fr. E+F (576.8 mg) was subjected to CC over Sephadex LH-20 (100 g) eluted with MeOH to obtain 6 fractions (Frs. E1-E6). Fr. E3 (95 mg) was purified by preparative HPLC (eluted with 20% acetonitrile, 4 mL/min, UV 320 nm) to yield 4 (2.1 mg) and 3 (2.2 mg). Fr. E5 (20 mg) was purified by preparative HPLC (eluted with 35% acetonitrile, 4 mL/min, UV 320 nm) to yield 5 (2.5 mg).