Melting points were determined on an electrothermal melting point apparatus. Measurements of optical rotation were performed on a Perkin Elmer model 241 polarimeter. UV spectra were recorded on a Perkin Elmer Lambda 2 spectrophotometer. NMR spectra were measured on a Bruker AV 600 spectrometer (600 and 150 MHz for ¹H and ¹³C NMR, respectively) and the residual solvent signals served as internal references (δ_H 3.31 and δ_C 49.0 for methanol-d₄). High-resolution mass spectra were obtained on a Bruker micrOTOF II spectrometer using ESI ion source in negative mode. Analytical HPLC-UV/MS was conducted on a VWR-Hitachi LaChrom Elite system (pump L-2130, diode array detector L-2450, autosampler L-2200 and column oven L-2300) with a Phenomenex Onxy Monolithic column (C18, 100 × 3.00 mm) applying a gradient of 0.1% formic acid in H₂O (A) and 0.1% formic acid in acetonitrile (B): 0 min 5% B, 4 min 60% B, 6 min 100% B; flow 2 ml min⁻¹. Coupling of the HPLC system to a Bruker esquire4000 ESI-ion trap allowed mass detection.

Strain KF501 was isolated from a water sample taken in the German Wadden Sea. DNA extraction, amplification of the internal transcribed spacer region (ITS) and sequencing were performed as described by Wiese et al. (2011) with slight modifications, centrifugation of the DNA at 8000 × g and 35 cycles of DNA amplification. The DNA sequence was deposited in GenBank under the accession number KF923800. Cryo-conserved stock cultures of strain KF501 were kept at −100°C using the Microbank system (Pro-Lab).

The fungal isolate was cultivated in the following media: casamino acids glucose medium (casein hydrolysate 0.25%, glucose × H₂O 4%, MgSO₄ × 7H₂O 0.01%, KH₂PO₄ 0.18%, pH 6.8) (Stevens, 1974), Czapek medium (sucrose 3%, NaNO₃ 0.3%, K₂HPO₄ 0.1%, KCl 0.05%, MgSO₄× 7H₂O 0.05%, FeSO₄× 7H₂O 0.001%, pH 6.2) (Samson et al., 2000), modified malt extract medium (malt extract 3%, NaCl 1.5%, pH 5.5) (Samson et al., 2000), potato-carrot medium (potatoes 40 g, carrots 40 g, each boiled in 1 l of H₂O and filtered off, 250 ml of potato extract and 250 ml of carrot extract were filled up with 500 ml of distilled H₂O) (Samson et al., 2000), and modified Wickerham medium (malt extract 0.3%, yeast extract 0.3%, peptone from soymeal 0.5%, glucose × H₂O 1%, NaCl 3%, agar 1.5%, pH 6.25) (Wickerham, 1951).

For isolation of the malettinins, Cladosporium sp. KF501 was cultivated in 9.75 l of casamino acids glucose medium. The cultivation experiments were performed as surface cultures at 20°C in the dark in 2-l Erlenmeyer flasks, each containing 750 ml medium. The cultures were inoculated with an agar slant (2.6 cm in diameter) of a 7 days old preculture which was grown on solid, modified Wickerham medium at room temperature in the dark. After 40 days of incubation, the mycelia of strain KF501 were harvested. For extraction, the mycelium of each flask was mixed with 150 ml of 96% EtOH, homogenized and filtered. As the obtained extract was very greasy, fats were frozen out at −20°C yielding 1.2 g of extract.

The malettinins were purified from the mycelium extract by preparative HPLC. The separations were performed on a VWR LaPrep system (pump P110, UV detector P311 UV, fraction collector Labocol Vario-2000 from LABOMATIC, autosampler Smartline 3900 from Knauer) using a Phenomenex Gemini-NX column (10 μ C18, 100A, Axia, 100 × 50.00 mm). A gradient of 0.1% formic acid in H₂O (A) and 0.1% formic acid in acetonitrile (B) (0.0 min 20% B, 17.5 min 40% B, 21.0 min 60% B; flow 100 ml min⁻¹) was applied to yield 17.5 mg of 2 (t_R 8.6–9.0 min), 6.1 mg of 3 (t_R 9.2–9.6 min), 4.6 mg of 4 (t_R 9.9–10.3 min) and 15.1 mg of 1 (t_R 12.2–18.0 min).

Single crystals of 4 suitable for X-ray data collection were obtained by recrystallization from a methanol solution. The data were measured using an Imaging Plate Diffraction System (IPDS-2) from STOE and CIE (Table 1). All non-hydrogen atoms were refined anisotropically. The H atoms were positioned with idealized geometry (O-H H atoms allowed to rotate but not to tip) and refined isotropically with U_iso(H) = 1.2·U_eq(C) (1.5 for methyl and O-H H atoms) using a riding model. There are two crystallographically independent molecules in the asymmetric unit, which exhibit the same relative configuration. In one of these molecules the H atoms of one methyl group are disordered and were refined using a split model with two orientations rotated by 60°. Because no strong anomalous scattering atoms are present, the absolute configuration cannot be determined. Therefore, Friedel opposites were merged in the refinement.

Antibacterial bioassays with the test strains Xanthomonas campestris and Staphylococcus epidermidis were performed as described by Silber et al. (2013). Bacillus subtilis and Candida albicans were tested as described by Ohlendorf et al. (2012). For antifungal assays using Trichophyton rubrum, cultivation was carried out in modified Sabouraud medium (peptone 1%, glucose 2%, pH 5.6). A spore solution containing 5 × 10⁴ spores of T. rubrum per ml medium was prepared. 10 mM DMSO solutions of compounds were diluted with medium to achieve the desired test concentrations. For assaying, 200 μl of the spore solutions were mixed with 10.5 μl compound solutions in 96-well microtiter plates and incubated for 72 h at 28°C in the dark. The cell growth of T. rubrum was evaluated by measuring the optical densities with a Tecan Infinite M200 plate reader. The resulting values were compared with a positive control (0.1 and 0.5 μM clotrimazole) and a negative control (no compound). IC₅₀ values were determined according to the inhibition of metabolic activity or cell growth in the presence of a compound as compared to the non-inhibited negative control.

Strain KF501 was isolated from water samples taken in the German Wadden Sea. The strain grew well in casamino acids glucose, Czapek and malt extract medium, slowly in potato-carrot medium and did not grow in liquid modified Wickerham medium. In general, the growth of strain KF501 was relatively slow by comparison with other fungal isolates.

Macroscopic and microscopic morphological features, colony color and structure of conidiophores were characteristic of Cladosporium spp. (Samson et al., 2000). This classification was corroborated by the comparison of the ITS DNA sequence of strain KF501 with sequences available at GenBank employing the Basic Local Alignment Search Tool (BLAST) (Altschul et al., 1990). The search resulted in 100% similarity to Cladosporium cladosporioides, C. pseudocladosporioides, C. uredinicola, C. bruhnei and C. colombiae. Hence, strain KF501 could be identified as a member of the genus Cladosporium, but identification to the species level was not possible.

The new malettinin E (4) along with the known malettinins A–C (1–3) were obtained from the mycelia extracts of Cladosporium sp. strain KF501 grown in casamino acids glucose medium. The compounds were isolated by preparative reversed-phase HPLC. Known compounds 1–3 were readily identified by comparison of their spectroscopic (¹H NMR, UV and MS) data with literature values (Angawi et al., 2003, 2005).

Malettinin E (4) has the molecular formula of C₁₆H₂₀O₅ as determined by HRESIMS which requires seven degrees of unsaturation. The physico-chemical properties of 4 are summarized in Table 2. The molecular formula of 4 being the same as for 2 and 3 suggested the compounds to be isomeric. In addition, the one- and two-dimensional NMR spectra of 4 (¹H, ¹³C, DEPT, COSY, HSQC, and HMBC; see Table 3 and Supplementary Information) revealed the presence of considerable structural similarities with 2 and 3 (Angawi et al., 2005). The ¹H and ¹³C NMR data of 4 showed resonances for three methyl, two methylene and two aliphatic methine groups, one oxygenated methane, one ketalic carbon and one carbonyl group as well as signals for six aromatic carbons one of which appeared to be oxygenated. Thus, all structural groups and atoms present in 2 or 3 were also observed for 4. Correlations of the aromatic protons in the HMBC NMR spectrum (Figure 2) established the same tropolone moiety for 4 as found in the known malettinins. In addition to the tropolone unit, 4 contained a dihydropyran ring identical to that of 2 and 3. The proton and carbon chemical shifts of the corresponding signals were in good agreement with those reported for 2 and 3, with the proton signals agreeing particularly well to the structure of 3. HMBC couplings from H₂−7 to C-5, C-6, and C-15 gave evidence for the attachment of the dihydropyran ring to the tropolone moiety. The NMR data determined the remaining structural part of 4 to be a tetrahydrofuran ring connected to the dihydropyran unit via C-9 (Figure 2). As a result, compound 4 was proven to have the same planar structure as described for 2 and 3. The vicinal J_H7–H8 values of 4 (5.7 and 10.3) showed better correlations with those of 3 (5.9 and 13) than with those of 2 (6.2 and 3.8), suggesting a configuration like in 3 for the stereogenic center at C-9. However, the structures were not identical as considerable deviations in the chemical shifts of the proton and carbon signals of CH-12 (δ_H 2.23, δ_C 43.0) and CH-13 (δ_H 3.96, δ_C 86.0) in 4 were observed when compared to the reported shifts of the respective signals in 3 (δ_H 2.5, δ_C 34.7 and δ_H 4.32, δ_C 73.0, respectively) (Angawi et al., 2005). Taken the NMR data together, it seemed most likely that 4 was a stereoisomer of malettinin C (3), differing in the configuration at C-12 and/or C-13.

In order to determine the relative configuration of 4 conclusively, a single-crystal X-ray diffraction analysis was performed. It showed that the compound consisted of two crystallographically distinct molecules (4a and 4b) with identical relative configuration, exhibiting slight conformational differences between the molecules (Figure 3). The X-ray crystal structure of 4 confirmed the assumptions made on the basis of the NMR data as it was found to be identical to 3 except for the configuration at C-13. Thus, 4 was identified to be the C-13 epimer of 3 as shown in Figure 1.

Compounds 1–4 were tested for biological activities against a set of bacterial and fungal test strains. The malettinins were found to be bioactive against the plant pathogenic bacterium Xanthomonas campestris and the human pathogenic dermatophyte Trichophyton rubrum (Table 4). Compounds 2–4 showed inhibition of X. campestris, whereas 1 did not inhibit the bacterial test strain notably. The IC₅₀ values of 2–4 were in approximately the same concentration range (28.3–37.9 μM). With regard to T. rubrum, all tested malettinins 1–4 exhibited inhibitory activities. However, the strength of inhibition differed between the compounds with the new malettinin (4) and 1 showing highest activities (IC₅₀ of 30.7 and 33.1), while 2 was half as active (IC₅₀ of 60.6 μM) and 3 was even less active (IC₅₀ of 83.2 μM).

Beside the reported antifungal and antibacterial properties in Table 4, malettinins B and C (2 and 3) showed weak inhibition (<80%) of Staphylococcus epidermidis, Bacillus subtilis and Candida albicans at a test concentration of 100 μM. At the same concentration malettinin E (4) exhibited weak inhibitory effects on the metabolic activity of S. epidermidis and C. albicans. Malettinin A did not show antibacterial effects, but also inhibited C. albicans (81% inhibition at 100 μM). As these activities were poor, IC₅₀ values were not determined.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.