Quinazolin-4-one alkaloids are a class of natural scaffold which has been proved as a drug-like template in medicinal chemistry and considered a favored structure [1]. The quinazolin-4-one ring system has been consistently recognized as a promising pharmacophore because of its broad spectrum pharmacological activities such as antitumor [2], antitubercular [3], anti-HIV [4], anti-inflammatory [5], antiangiotensin [6], antibacterial [7], and antifungal [8]. Involved in centrosome separation and bipolar mitotic spindle formation, kinesin spindle protein (KSP) plays an important role in cell division [9]. Some quinazolin-4-one compounds are KSP inhibitors. They can arrest cells in mitosis and induce cell death [10], and have proved to be promising candidates for anticancer drugs [11]. The biological and pharmacological activities of quinazolin-4-one derived compounds have been documented not only from synthetic derivatives but also from several naturally occurring alkaloids isolated from families of the plant kingdom, and from microbes such as Streptomycetes and fungi [12,13,14,15,16,17].

During high throughput screening of novel compounds from marine derived microorganisms [18], our group have had identified two pyrone-type polyketides from the marine derived fungus Penicillium aurantiogriseum [19]. By varying the culture media according to the OSMAC (one strain-many compounds) approach [20,21], a drastically altered metabolite profile of the same strain was obtained. With the help of HPLC, we found the UV spectra of some compounds in the crude extract were similar to the specific absorbance of the quinazoline-4-one core. Using UV-guided fractionation, three new quinazolin-4-one derivatives (1−3), together with two known metabolites auranthine (4) [22] and aurantiomides C (5) [23] (Figure 1) were isolated from this marine-derived fungus. The novel quinazolinone derivatives were named as auranomides A, B and C (1−3). Herein, we report the isolation, structure elucidation and bioactivity evaluation of these alkaloids.

Compound 1 was obtained as a white amorphous powder with the specific absorbance for the quinazoline-4-one. The molecular formula of 1 was determined to be C₁₉H₁₆N₄O₃ (fourteen degrees of unsaturation) by analysis of its HRESIMS (m/z 349.1290 [M + H]⁺). The UV spectrum of 1 showed a specific absorbance for the quinazoline-4-one at 216.0, 259.0 and 296.0 nm. The ¹H, ¹³C, and HSQC spectra of 1 (Table 1), showed 19 carbon signals for two methylene group, one sp³ hybrid methine group, eight aromatic methine carbons, five sp² hybrid aromatic quaternary carbons, and three sp² hybrid quaternary carbons at δ_C 161.3 (C-4), 165.7 (C-23) and 171.6 (C-14), suggesting the presence of an amide carbonyl, a carboxylic acid, and a C=N carbon, respectively. The ¹H and ¹³C NMR spectra revealed the presence of an ortho-disubstituted benzene ring corresponding to the anthranilate moieties for 1, which is a fragment of the quinazoline-4-one moiety. Analysis of the ¹H–¹H COSY NMR data led to the identification of the fragment from C-11 through C-12 to C-13. In the HMBC spectrum (Figure 2), correlations from H-15 to C-11, C-12, C-13 and C-14, together with correlations from H₂-N-16 to C-13 and C-14 revealed the pyrrolidin-2-iminium moiety. The HMBC correlations from H-11 and H₂-12 to C-2 indicated that the pyrrolidin-2-iminium moiety was attached to C-2 through C-11. Protons and carbons signals for another ortho-disubstituted benzene ring were observed from the NMR data. Considering the molecular formula, as well as the unsaturation requirements for 1, this ortho-disubstituted benzene ring was attached to N-3 through C-17. In the ¹H spectrum, three exchangeable protons were observed. The proton at δ_H 10.29 was assigned to N-15 by its HMBC correlations to C-11, C-12, C-13 and C-14. The HMBC correlations from the other two exchangeable protons at δ_H 9.25 and 8.75 to C-13 and C-14 indicated the carbon signal (δ_C 171.6) should be an iminium positive ion. On the basis of these data, the structure of compound 1 was established.

High resolution ESIMS(+) analysis of 2 revealed a pseudomolecular ion at m/z 363.1451 [M]⁺, consistent with the molecular formula C₂₀H₁₈N₄O₃, and corresponding to fourteen degrees of unsaturation. Compound 2 had a very similar UV spectrum to that of 1. The ¹H, ¹³C, and HSQC spectra of 2 showed 20 carbon signals which were very similar to those of 1 except for the presence of methoxyl group signals (δ_H 3.36, δ_C 52.5). In the HMBC spectrum, correlation from H₃-24 to C-23 indicated that the methoxyl group was attached to the carboxyl and generated a carboxylic acid methyl ester. The exchangeable proton at δ_H 10.32 was assigned to H-N-15 by the HMBC correlations to C-11, C-12, C-13 and C-14. The other exchangeable protons at δ_H 8.89 and 9.24 were assigned to H-N-15 by the HMBC correlations of H-N-16 to C-13 and C-14. On the basis of these data, the structure of compound 2 was established as the methyl ester of 1.

The molecular formula of 3 was determined to be C₁₉H₁₆N₄O₃ (fourteen degrees of unsaturation) by analysis of its HRESIMS (m/z 371.1132 [M + Na]⁺) and NMR data (Table 2), the same as that of auranomide A (1). It also showed the specific UV spectrum (λ_max 219, 259 and 301 nm) for quinazolin-4-one. The ¹H, ¹³C, and HSQC spectra of 3 showed 19 carbon signals for two methylene groups, one sp³ methine group, eight aromatic methine carbons, five sp² aromatic quaternary carbons, and three sp² quaternary carbons at δ_C 161.0 (C-4), 166.9 (C-23) and 173.7 (C-14), suggesting the presence of three carbonyl carbons. The ¹H and ¹³C NMR spectra revealed the presence of an ortho-disubstituted benzene ring corresponding to two anthranilate moieties in compound 3. Analysis of the ¹H–¹H COSY NMR data led to the identification of the fragment from C-11 through C-12 to C-13. In the HMBC spectrum (Figure 2), correlations from H-N-16 to C-11, C-12 and C-18, as well as from H₂-12 to C-2 and C-11 revealed the connection from C-2 through C-11, N-16 to C-23. On the basis of these data, the structure of compound 3 was established.

Auranomides A and B are a new class of alkaloids which contain the moiety of 2-pyrrolidin-2-iminium quinozoline-4-one. Quinozolin-4-one derivatives have been isolated from several fungi and originated via similar biosynthesis pathways [17,22,23]. Auranthine (4) was biosynthesized by Penicillium aurantiogriseum [24,25]. Scheme 1 shows a plausible biosynthesis pathway for quinozolin-4-ones analogues (1–3). Two molecules of anthranilic acid were incorporated into 2-(2-aminobenzamido)-benzoic acid (6). A subsequent incorporation of glutamine yielded 7. The amino group of anthranilic acid reacted with the carbonyl carbons of glutamine to yield 8. The primary amino group of glutamine could then react with the terminal amide and the carbonyl carbons of anthranilic acid to form 1 and 3, respectively. From the perspective of biosynthesis, the absolute configurations of C-11 for 1–3 were assigned as 11S.

Antitumor activity of auranomides A–C (1–3) was evaluated in several cells lines by the CCK-8 method. As shown in Table 3, auranomide B exhibited the most potent inhibitory effect against human myelogenous leukemia HEPG2 cells, with an IC₅₀ value of 0.097 μmol/mL. These compounds were also assessed for activities against methicillin-resistant Staphylococcus aureus (MRSA, Clinical isolates, Beijing Chao-yang Hospital, Beijing, China), Candida albicans and synergistic antifungal activity with ketoconazole. None of them showed activities at low concentration (MICs > 100 μg/mL).

By changing the fermentation conditions, such as changing the carbon and nitrogen source as well as the marine derived fungus Penicillium aurantiogriseum, completely different natural products were produced including three novel quinazolin-4-ones auranomides A–C. Among them, auranomide B showed moderate antitumor activity. The possible biosynthetic pathway for these novel alkaloids was proposed which begins with two molecules of anthranilic acid and involves one molecule of glutamine. This study proved that the OSMAC approach seems effective in searching for natural products with new structures.