Studies on the secondary metabolism of Rosellinia and Dematophora strains (Xylariaceae) from Iran

The xylariaceous genus Dematophora has recently been resurrected and segregated from Rosellinia based on a molecular phylogeny and morphological characters. This was an important taxonomic change because Dematophora in the current sense contains several important pathogens, while Rosellinia is limited to mainly saprotrophic species that have an endophytic stage in their life cycle and may even have beneficial effects on the host plants. During our ongoing work on the functional biodiversity of the Xylariales, we have encountered new strains of rosellinoid Xylariaceae from Iran and have studied their mycelial cultures for secondary metabolites in an attempt to establish further chemotaxonomic affinities. In the process, we isolated and identified 13 compounds, of which rosellisteroid (1), the cichorine derivative 2, and the alkaloid 3 are new. Out of these, nine were tested for their antimicrobial affinities with cytochalasin E (6) exhibiting weak activity against Schizosaccharomyces pombe. The cytotoxicity of three cytochalasin derivatives was examined and their effects on the F-actin cytoskeletal organization studied by fluorescence microscopy using fluorescent phalloidin. Cytochalasin E (6) and Δ6,12-cytochalasin E (7) showed strong and irreversible action on actin, while cytochalasin K (8) exhibited weaker, reversible effects.


Introduction
The Xylariaceae is one of the largest families of the ascomycete class Sordariomycetes and is phenotypically characterized by geniculosporium-like anamorphs and the absence of stromatal pigments according to the recent classification by Wendt et al. (2018). Most species in this family are saprobes or endophytes, while some species are well-known plant pathogens (Helaly et al. 2018;. Hyde et al. (2020) accepted 32 genera; however, it can be expected that this number will increase until a conclusive classification system has been established (e.g., Konta et al. 2020;Heredia et al. 2020). Furthermore, ongoing large-scale phylogenomic studies based on 3rd-generation genome sequencing techniques will strengthen future classification frameworks. This work has stabilized the phylogenetic backbone and has revealed that the genomes of the Xylariales contain a complex and extensive diversity of biosynthetic gene cluster encoding for secondary metabolites (Wibberg et al. 2021;Kuhnert et al. 2021). Especially taxa from tropical and subtropical areas are known to produce a high diversity of secondary metabolites, many of which are biologically active (Kuephadungphan et al. 2021). For instance, these fungi produce volatile antibiotics (Samarakoon et al. 2020), cytotoxic agents , and even inhibitors of biofilm formation in bacterial pathogens (Yuyama et al. 2017). In the sister family Hypoxylaceae, the use of chemotaxonomic methodology such as HPLC profiling has contributed greatly to the current, polyphasic taxonomic concepts, since secondary metabolite production was found to be widely in agreement with concurrent multi-locus genealogies and even phylogenomic reconstructions (Wendt et al. 2018;Wibberg et al. 2021).
One of the largest, but poorly studied, genera of the Xylariales is Rosellinia, which belongs to the Xylariaceae . Recently, the fungus producing PF1022A, a cyclooctadepsipeptide, which is the precursor of the anthelmintic drug emodepside, was assigned to Rosellinia (Wittstein et al. 2020). Based on a multi-locus molecular phylogeny that was conducted in the same study, the genus Dematophora was resurrected for those Rosellinia species that have a characteristic synnematal anamorph. Dematophora now includes all important plant pathogens that were formerly placed in Rosellinia. However, the species segregation in these genera is based on morphology, and more than 100 of the ca. 160 morphologically described species remain uncultured and unassessed for secondary metabolite production or their phylogenetic affinities.
In the present study, we studied the secondary metabolites of Rosellinia and Dematophora species from Iran for previously undescribed secondary metabolites, complementing previous work by Wittstein et al. (2020). This work was also conducted as a starting point to obtain standards of the prevailing secondary metabolites for a subsequent chemotaxonomic study of these genera. Moreover, we aimed to isolate cytochalasins, F-actin-binding molecules, which are known to disrupt the eukaryotic actin cytoskeleton (Cooper 1987) to study their poorly understood structure activity relationships (Scherlach et al. 2010).

Fungal sources/morphological observations
The fungal specimens were collected in Northern Iran (Guilan, Mazandaran, and Golestan provinces) and Isfahan Province. Dried specimens were deposited in the University of Guilan Mycological Herbarium (GUM). Living cultures were deposited in the culture collection MUCL (Louvain-la Neuve, Belgium) and Iranian Research Institute of Plant Protection (IRAN C), Tehran, Iran. For light microscopy, fresh collections were examined according to Petrini (2013). Single ascospore isolation and examination of culture macromorphology were examined according to Pourmoghaddam et al. (2020). Observations and taxonomical diagnoses can be found in the SI.
DNA extraction, PCR, and sequencing DNA extraction of fresh cultures and amplification of the ITS (nuc rDNA internal transcribed spacer region containing ITS1-5.8S-ITS2), LSU (5′ 1200 bp of the large subunit nuc 28S rDNA), RPB2 (partial second largest subunit of the DNAdirected RNA polymerase II), and TUB2 (partial β-tubulin) loci were carried out as described by Wendt et al. (2018). DNA sequences were generated by an in-house Sanger capillary sequencing solution and processed with Geneious® 7.1.9 (http://www.geneious.com).

Molecular phylogenetic analyses
Alignments and phylogenetic analyses of the combined multilocus matrix of ITS, LSU, TRPB2, and TUB2 DNA sequences using maximum parsimony and maximum likelihood were constructed as described in Lambert et al. (2021). For the combined matrix used for phylogenetic analyses, 59 accessions of 45 species of Xylariaceae and four outgroup taxa from Graphostromataceae (Biscogniauxia nummularia, Graphostroma platystomum) and Hypoxylaceae (Hypoxylon fragiforme, H. howeanum) were included. A table summarizing all used strains, their corresponding sequences, and GenBank accession numbers can be found in the SI (Table S1).

Cultivation, extraction, and isolation procedures
The fungal strains (see Table S1 in the SI, marked in bold) were grown on CS and CSA (cotton seed with and without agar supplement) (maltose 40 g/L, cotton seed flour 20 g/L, CaCO 3 3 mg/L, soy peptone 2 g/L, MgSO 4 × 7 H 2 O 2 g/L, NaCl 2 g/L, agar 20 g/L, pH adjusted to 6.3) media for 7 d at 23°C in darkness. Fungal colonies grown on 90-mm agar plates were cut into pieces (1 × 1 cm) and regularly transferred to new CSA plates during the chemotaxonomic study. To compare their secondary metabolite production, the strains were inoculated in a 500-ml Erlenmeyer flask containing 200 ml of CS medium and shaken at 150 rpm in the dark until the glucose level was depleted (measured by using glucose test stripes) (Wittstein et al. 2020). In addition, the strains were inoculated on CSA plates and incubated in the dark at 23°C until the mycelia covered twothirds to the full agar plate (14-21 d). Extraction of the secondary metabolites followed Wittstein et al. (2020) with minor adaptations; briefly, the supernatant and biomass of submerged cultures were separated by vacuum filtration instead of gauze filtration, while the supernatant-ethyl acetate and solid-phase extraction (biomass harvested from submerged cultures, acetone; agar plates, ethyl acetate) procedures remained unchanged. The crude extract was analyzed by high-performance liquid chromatography coupled with diode array detection and mass spectrometry (HPLC-DAAD/MS) as described previously (Wittstein et al. 2020). The resulting chromatograms were analyzed using the Bruker Compass Data Analysis 4.4 software (Bruker Daltonik GmbH) and evaluated for the presence of known and unknown secondary metabolites.

Isolation of compounds 3-13
The cultivation of Rosellinia cf. akulovii (MUCL 57710) on 230 CSA plates provided 2.12 g of crude extract. One hundred twelve milligrams of the crude extract were subjected to normal-phase flash chromatography (GRACE Reveleris X2 flash system; Reveleris silica cartridge 12 g, solvent A: dichloromethane, solvent B: acetone; gradient 0% B to 100% B in 40 min, followed by 100% MeOH in 5 min). UV-Vis detection was carried out at 210 nm, 265 nm, and 365 nm. Fractions were combined according to the major UV-absorbing peaks to yield 1.3 mg of 4.
In a second attempt, 1.96 g of crude extract was subjected to normal-phase flash chromatography (GRACE Reveleris X2 flash system), using a Reveleris 24 g silica cartridge with the same elution gradient to afford eight main fractions (F1-F8). The resulting fractions were further subjected to preparative HPLC (GX-271/172 DAD/ pumps 305/306, Gilson Inc., Middleton, USA). Deionized water (Milli-Q, Millipore, Schwalbach, Germany) with 0.1% formic acid (solvent A) and acetonitrile with 0.1% formic acid (solvent B) were used as mobile phase. Prior to the preparative HPLC, the fractions were filtered using a SPME Strata-X 33 u Polymeric RP cartridge (Phenomenex, Inc., Aschaffenburg, Germany). UV spectra of eluents were monitored at 210 nm, 254 nm, and 280 nm.
Using the same column and elution gradient, fraction F4 (94 mg) was further purified to give 1.6 mg of 5.
The cultivation of D. necatrix (MUCL 57709) on 260 CSA plates resulted in 2.98 g of crude extract. This crude extract was first subjected to normal phase flash chromatography (GRACE Reveleris X2 flash system) [Reveleris cartridge 24 g, solvent A: dichloromethane; solvent B: acetone (0% B to 100 B in 40 min) and 100% MeOH in 5 min]. UV detection was conducted at 210 nm, 265 nm, and 365 nm. Compound 11 (15.7 mg) and seven fractions (F1-F7) were collected according to the observed peaks.

Determination of absolute amino acid stereochemistry in xylarotide (4)
Amino acid absolute stereochemistry of xylarotide A (4) was determined according to Primahana et al. (2021). Briefly, xylarotide A (4; 100 μg) was acid hydrolyzed using HCl (100 μL, 6n) at 110°C for 14 h. The acid hydrolysate was dried under N 2 stream and divided into two separated vials. The hydrolysates were treated with Marfey reagent (D or L FDVA) as described above.
Subsequently, the compounds were examined in a eukaryotic F-actin network disruption assay using adherent mammalian osteosarcoma (U2-OS, ATCC HTB-96) cells. This cell line has already been used to describe effects of cytochalasins on the actin organization in previous publications (Kretz et al. 2019;Moussa et al. 2020;Lambert et al. 2021). Briefly, 20,000 cells of a maintained cell culture were sown on fibronectin-coated glass coverslips and expanded overnight. The growth medium for the assay was equilibrated in an incubator overnight. Cells were treated with full medium supplied with a concentration gradient ranging from 0.01 to 10 μg/ml (0.01 μg/ml, 0.03 μg/ml, 0.1 μg/ml, 0.3 μg/ml, 1 μg/ml, 3 μg/ml, 10 μg/ml) and subsequently, varying concentrations for compounds 6 (0.13 μg/ml, 0.65 μg/ml) and 7 (0.084 μg/ml, 0.42 μg/ml) and compound 8 (7.2 μg/ml, 36 μg/ml) dissolved in DMSO. Cells treated with 10 μl/ml DMSO served as a vehicle control. After 1 h, cells were washed once with PBS and fixed with PBS containing 4% para-formaldehyde (PFA). In a separate experiment, the reversibility of the cytochalasins' impact on the F-actin organization was examined by wash-out after 1 h of treatment with PBS (3×) and addition of equilibrated full medium, allowing cells to recover for 1 h. Cells were then fixed using 4% PFA dissolved in PBS. After permeabilization with 0.1% Triton X-100 at room temperature for 1 min, all cells fixed on cover slips were washed with PBS and stained for the actin cytoskeleton by administering phalloidin coupled to a fluorescent probe (ATTO-594, ATTO-Tec, Siegen, Germany) for 1 h. The cover slips were embedded using Pro-long Diamond Antifade (Invitrogen, Carlsbad, CA, USA) mountant supplied with DAPI to stain for the nucleus. Epifluorescence was detected and images captured using either an inverted microscope (Axio Vert 135 TV, Zeiss, Jena, Germany) with a 63× Zeiss oil immersion objective (Neofluar, 1.25 NA) equipped with a PCOedge 4.2 (sCMOS, PCO, Germany) camera o p e r a t e d b y t h e M e t a m o r p h s o f t w a r e p a c k a g e (molecularDevices, San Jose, CA, USA) or a Nikon Ti eclipse 2 microscope with a 60× Nikon oil immersion objective (plan apo, NA 1.4) equipped with a pE-4000 LED light source (CoolLED, Andover, UK) and PCOedge 4.2 camera and operated by NIS elements (Nikon, Düsseldorf Germany). Images were processed by Image J (NIH, Bethesda, MD, USA).

Morphological analysis and molecular identification
The specimens were identified by comparison of their morphological characteristics and sequencing of four loci (ITS, LSU, RPB2, and TUB2) (see molecular phylogenetic reconstruction and taxonomy, Figure S1, Figure S2, Figure S3, Figure S4, and Table S1 in Supplementary Information (SI)). Two well-known widely distributed species (Dematophora necatrix and Rosellinia corticium) were identified from Iran. Moreover, another taxon was found that may either constitute an undescribed taxon or be conspecific with Rosellinia akulovii. However, the clarification of the identity and status of the latter specimen will require additional work that is unrelated to the main objective of this manuscript. Additional specimens need to be made available and studied. For the time being, this fungus is referred to R. cf. akulovii, and the preliminary data are being reported in the Supplementary Information.

Secondary metabolites
All strains were screened for secondary metabolite production using submerged shake cultures in cotton seed medium and in static condition on cotton seed medium supplemented with agar (Wittstein et al. 2020). Submerged cultures were harvested after glucose was consumed and plates after the fungal strains covered approximately the whole plate. Shake cultures were divided into supernatant and mycelial biomass. Extracts were screened for peaks potentially representing novel compounds and cytochalasins. Briefly, compounds 1 and 2 were isolated from pooled extracts derived from the supernatant of R. corticium and R. cf. akulovii, compounds 12 and 13 from combined agar extracts of R. corticium (MUCL 57712 and MUCL 57713), compounds 3-5 from agar extracts derived from strain R. cf. akulovii (MUCL 57710), and compounds 6-11 from a D. necatrix (MUCL 57709)-derived extract grown on agar plates. A complete list of the occurrence of each isolated metabolite in all strains is given in the SI (Table S2).
The HRESIMS data of metabolite 2 supported a molecular formula of C 15 H 19 NO 5 , implying 7 degrees of unsaturation. 1 H and HSQC NMR data accounted for four methyls, two methylenes, and one oxymethine. The 13 C NMR spectrum contained signals for six further aromatic carbons devoid bound protons and one carboxyl atom. HMBC correlations from 3-H 2 , 5-Me, and 4-OMe connected the indole core, which was connected to a 3-methylbutanetriolyl subunit (see Fig. S60 for details).
The molecular formula C 12 H 17 NO 4 of the pyrrol alkaloid 3 was deduced from its HRESIMS data, implying five numbers of unsaturation. 1 H and HSQC NMR data accounted for one oxymethyl, five methylenes, two olefinic methines, and a conjugated aldehyde. The 13 C NMR spectrum showed signals for a carboxylic acid two additional aromatic carbons.

Bioactivity assays and actin disruption assays
Compounds 1-13 were evaluated by a set of serial dilution assays to determine MIC that are employed as standard procedure for characterization growth inhibition. Not all compounds were tested because for some, only small quantities were available after structure elucidation. For compounds 3, 4, 6-11, and 13 ( Supplementary Information, Table S3), only compound 6 inhibited the growth of Schizosacharomyces pombe at 16.8 μM. A summary of tested organisms and effects of compounds is given in the Supplementary Information, Table S3. Next, we determined 50% inhibition of cellular survival by assessing metabolic reduction of MTT (compounds 6-8) with a panel of tested eukaryotic cell lines.
The cytochalasins (6-8) belong to a class of molecules that are known as potent F-actin disrupting toxins (Scherlach et al. 2010) and were evaluated for their cytotoxicity against a panel of eukaryotic cell lines ( Fig. 2A). All exhibited moderate cytotoxicity against all tested cell lines. Notably, compound 8 was about 80 times less toxic than its close derivatives 6 and 7, which only differ by the presence of a double bond in the 6ring system (7), putting the attached methyl groups into a planar position and an epoxide at C-12 (compound 6). As cytochalasans are well known F-actin network disruptors, the cellular bioactivity against the cellular target actin was evaluated in a functional assay in U2-OS cells using fluorescence microscopy. We examined the effect of the three cytochalasins on the F-actin network in cells fixed and stained for F-actin using fluorescent phalloidin upon cytochalasin treatment and DMSO treatment as vehicle control. A titration experiment ranging from 0.01 to 10 μg/ml (Fig. 2B) showed different degrees of F-actin network disruption severity, which was accompanied by a substantial reduction of Factin bundles in the cell body. Higher concentrations of compounds 6 and 7 (> 0.3 μg/ml) lead to a total collapse of the Factin network, while the F-actin network antagonizing effect of compound 8 was only observed at a higher concentration (10 μg/ml). Interestingly, concentrations of compounds 6, 7, and 8 that lead to significant alterations in the F-actin network approximately corresponded to the determined IC 50 value determined in mouse fibroblasts L929. This prompted us to test the toxins at a "low-dose" (1 × IC 50 ) and a "high dose" (5 × IC 50) , resulting in partial (Fig. 2C, panels a, b, c) or total (panels d, e, f) actin cytoskeletal network disruption. Moreover, a wash-out experiment upon high-dose treatment demonstrated the irreversible impact of compounds 6 and 7 on the F-actin cytoskeletal organization after a recovery time of 1 h (Fig. 2C; compare d and g; e and h), while the F-actin network partially recovered from a high dosage of compound 8 (Fig. 2C; compare f and i).

Discussion
With hyfraxins A, we recently isolated a related steroid to the novel compound 1 from the Ash dieback causal fungus Hymenoscyphus fraxineus (Surup et al. 2018). Rosselisteroid (1) shares the conjugated C-3 ketone and the β-D-mannose moiety attached to C-23. However, the stereo configuration at C-23 is contrary and the π-system of 1 is truncated by the C8/ C9 epoxide.
The new metabolite 2 is closely related with cichorine, a phytotoxin isolated from Alternaria cichorii, a phytopathogen of Russian knapweed (Stierle et al. 1993). From Lycium chinense fruits, another chemically similar derivative has been isolated (Chin et al. 2003), which, however, bears a butanoic acid side chain instead of the pentanoic moiety of the novel pyrol alkaloid 3. This compound showed hepatoprotective effects comparable to the standard, silybin.
The cyclopeptide 4 has been previously encountered in a Xylaria strain (Li et al. 2011), and we confirmed the report by these authors that lacks antimicrobial effects. The steroid 5 was previously isolated from a Nodulisporium sp. grown on rice medium and found to be devoid of cytotoxicity against a panel of human cancer cell lines . Interestingly, nodulisporium-like anamorphs can be observed in all genera in the Hypoxylaceae and even in some Rosellinia species (Wittstein et al. 2020).
The cytochalasin 6 is well known to be produced by D. necatrix ((Aldridge et al. 1972), as Rosellinia necatrix) and discussed as a pathogenicity factor (Kanematsu et al. 1997), while the occurrence of 8 has been discussed to represent either a natural compound, or a degradation product of 6. The isomer 7 has so far only been obtained through acid rearrangement of 6 (Steyn and van Heerden 1982; Kajimoto et al. 1989). Since we used solvents supplemented with TFA, it cannot be formally excluded that 7 and 8 might have been isolated as artifacts. Besides their phytotoxic effect, cytochalasans are well known to interfere with F-actin polymerization (Cooper 1987) and actin cytoskeletal dependent biological phenomena such as cell adhesion (Takamatsu et al. 2002) and is discussed as an antiangiogenetic and tumor growth suppressing agent (Udagawa et al. 2000). Since 6, 7, and 8 only differ by an epoxide (6 versus 7 and 8) and an endo-or exocyclic position of a double bond, influencing the stereochemistry of the methyl groups (compounds 7 and 8), the here isolated compounds are ideal for experiments to gain deeper insights into the structure activity relationship of cytochalasans. Interestingly, 6 and 7 share almost equal cytotoxic properties with comparable F-actin network disrupting activities, while the toxicity and effect of 8 appears strongly reduced. Liu et al. (2006) observed a similar phenomenon while testing the cytotoxicity of 6 and 8 in the P388D1 (mouse macrophage, ATCC: CCL-46) and A549 (human lung epithelial, ATCC: CCL-185) cell lines, suggesting the epoxide moiety to strongly increase the toxicity of 6.
Since we observed 6 and 7 to exhibit a comparable cytotoxicity, this rather speaks for a role of the positioning of the substituents in the 6-ring system through the placing of the double bond. In a previous study, no correlations between differently positioned double bonds on the strength of F-actin network disruptive capabilities could be drawn (compare 19,20-epoxycytochalasin C and 19,20-epoxycytochalasin D, 18 and 19 in Kretz et al. (2019)). This observation highlights that a chemical change in the 6-ring system alone is not sufficient to influence its F-actin binding capabilities by such drastic extent as has been observed in this study. This leads us to the conclusion that the stereochemical properties of the 6-ring have to influence the cytotoxicity and F-actin network disrupting capabilities in the cytochalasin E backbone, if one excludes the possibility of a reduced membrane permeability due to the structural differences in the molecules. Additionally, the effect of the tested compounds on the F-actin organization did not appear to be reversible, in line with previous results suggesting the keto function to be an important structural parameter to influence reversibility . It should be noted that a partial F-actin network re-organization after 1 h recovery time has been observed with compound 8, indicating its weaker affinity for actin.
Metabolite 9 is known from a static culture extract of Dematophora arcuata ( (Oppong et al. 2009 Table summarizing half-maximal inhibitory concentrations (IC 50 ; in μM) of the cytochalasin derivatives 6, 7 and 8 measured by inhibititon of metabolic activity with MTT against various eukaryotic fibroblast and human cancer cell lines; B Overlay of the fluorescence signal of the F-actin network (pseudocoloured in red) and nuclear DNA (blue), of U2-OS cells treated for 1 h with increasing concentrations of compounds 6, 7, 8 and DMSO as vehicle control as indicated, stained with fluorescently tagged phalloidin-and DAPI, respectively. Arrowheads represent previously determined IC 50 concentrations, corresponding with the onset of visible perturbations on the actin cytoskeletal organization; C Overlay images of recorded fluorescence signal of visualized F-actin and nuclear DNA as above. Images show U2-OS cells treated for 1 h with concentrations of compound 6, 7, 8 equaling their determined IC 50 (low dose; 0.13 μg/ml, (a) 0.084 μg/ml, (b) 7.2 μg/ml, (c) or 5× the IC 50 (high dose; 0.65 μg/ml, (d) 0.42 μg/ml, (e) 36 μg/ml, f). After 1 h of high-dose treatment, the test compounds were washed-out, the cells supplemented with fresh medium and given 1 h of recovery time prior to fixation and staining for F-actin reorganization (g, h, i). DMSO was used as a vehicle control in a 1 h treatment experiment (j) and 1 h treatment followed by 1 h recovery after washout (k). Scale bars represent 20μm It was reported to possess weak to moderate cytotoxic, antimicrobial and even antimalarial activities, indicating a weak, non-selective mode of action in biological systems. We could not reproduce the antimicrobial and cytotoxic effects reported in our standard assays, which may be due to the use of different parameters and strains.
Metabolite 10 has been isolated from other fungi of the Xylariales belonging to the Hypoxylaceae (Annulohypoxylon stygium (Cheng et al. 2014)) and Graphostromataceae (Biscogniauxia cylindrospora (Cheng et al. 2011)), with this study being the first record from Rosellinia spp. No bioactivity was reported in either of the above-cited papers, and we could also not detect any significant effect.
Metabolite 11 has been previously isolated from an endophytic isolate of Colletotrichum gloeosporioides and reported to exhibit antifungal activity against Cladosporium cladosporioides and Cladosporium sphaerospermum (Chapla et al. 2014;Kim and Shim 2019). The compound was inactive against the tested strains in this study. The small compound 12 has previously been isolated from a Xylaria strain and did not show any antimicrobial or cytotoxic effects (Sun et al. 2017). Like the fatty acid 13, it proved to be devoid of significant activities in our own bioassays, too.

Conclusion
The current study has investigated Dematophora and Rosellinia species from Iran. Representative strains of two known and one potentially new species were characterized taxonomically and intensively examined for secondary metabolites in their cultures including their isolation by means of preparative chromatography and their structure elucidation using NMR spectroscopy and other state-of-the-art methodology. Interestingly, three out of 13 compounds characterized turned out to be unprecedented natural products belonging to several different molecule classes, again indicating the creativity of the Xylariaceae with respect to secondary metabolite production. Most of the major components in the extracts did not show prominent biological activities in a panel of standard assays that is established as a routine procedure in our laboratory. They may possess other biological function that could not be detected by these methods and may be involved in intraspecific communication or the defense against nematodes or other organisms that the fungi are confronted with in their natural environments. The described compounds add towards the repertoire of chemodiversity of Rosellinia and Dematophora. The isolated compounds can serve as standards for our pending, comprehensive studies on the chemotaxonomy of the Xylariaceae as more strains become available. Moreover, the spectral data reported here will easily allow other scientists to compare the metabolite profiles of their strains. In addition, the characterization of three cytochalasins in cell-based assay provided important information on the structure-activity relationships of this compound class on the actin cytoskeleton.