Growth of CR200 on potato dextrose agar (PDA), cornmeal agar (CMA) and Leonian’s agar (LA) were examined. The organism grew rather well on all three media, but the most differentiation and texture occurred on PDA (Table 1). CR200 grew robustly on PDA at 22°C reaching a diameter of 74 mm after only 7 days of growth and a maximum of 81 mm by 14 days (Figure 2A). Surface morphology was generally floccose to felty throughout the incubation period. Colony color progressed from white to taupe and buff with some yellow-brown at the colony center. Conidiomata (pycnidia) were visible by day 14 as white to light brown dots partially submerged into the agar (Figure 2A). By the 21^st to 28^th days, some conidiomata appeared dark brown (Figures 2B). About five to ten percent of them oozed a spore mass as yellow cirrhi in globules (Figures 2C) while others leaked a clear to brownish exudate. The conidiomata displayed elongated necks/beaks, some of them with extensive branching after an extended incubation of six weeks to several months.

CR200 grew most rapidly on LA, but remained flat or submerged in the agar. Mycelia were colorless or translucent becoming light olive-gray by day 14. Conidiomata also formed on LA by day 14, but were completely submerged, olive-gray and larger than those that formed on PDA.

CR200 grew more slowly on all three media at 33°C. The growth stress at the higher temperature was indicated by altered morphologies. On PDA, a red-brown to brown soluble pigment was produced, while on LA both the surface and reverse displayed red-brown to brown pigmentation. Conidiomata did not form on any media at 33°C. Conidiophores were hyaline, unbranched to highly branched at the base and/or mid-height and formed a continuous layer (Figure 2D, E). Conidiogenous cells were enteroblastic, phialidic, tapering to the apices with collerettes up to 2 μm long (Figure 2E). Conidia were hyaline, allantoid and measured about 5 to 6 μm long and 1.5 μm wide (Figure 2E).

The ITS of CR200 was found to be 517 bp in length. A BLAST search of the ITS of CR200 revealed that it is most similar to C. acaciae (96%). A phylogenetic tree (Figure 2) of the ITS of 22 Cytospora spp. placed CR200 in a clade with C. acaciae, C. rhizophorae, C. nitschkii, and C. abyssinica. However, CR200 branched deeply from C. acaciae (supported by 69 bootstrap replicates), indicating that it is a distinct species.

Production of cytoskyrin A as measured by BIA activity was quite heterogeneous in the 100-flask fermentation. Approximately 87% of the flask fermentations were active in the BIA, with 42% displaying very good activity and 45% with fair to poor activity. The remaining 13% of the flask fermentations did not exhibit activity in the BIA, suggesting no production of cytoskyrin A. Separation of the cells from the broth and extraction of both with ethyl acetate revealed that the majority of cytoskyrin A was associated with the cells, but is also present in the fermentation broths. The higher shaking speed of 200 rpm produced activity detected by the BIA sooner than the lower shaking speed of 100 rpm, whether incubated at 22°C or 28°C. Interestingly, BIA activity was not detected from the whole broths of CR200 fermented at 22°C and 100 rpm, even after 21 days of incubation.

DNA-damaging activities of the CR200 fermentation broth and pure compounds were tested by the biochemical induction assay (BIA). Cytoskyrin A demonstrated a minimum inducing concentration of 12 ng/spot in this assay, suggesting this compound as a potent DNA-damaging agent in bacteria (Table 2). Related bisanthraquinones such as rugulosin and luteoskyrin, known DNA intercalating agents, were toxic but did not induce the positive BIA response under the used assay conditions. These agents exhibit induction in the BIA only after a prolonged preincubation period. Cytoskyrin B, also produced by CR200, was inactive in the BIA. Cytoskyrin A appears to have the structural requirements to exhibit potent DNA-damaging activity in bacteria. Cytoskyrin A also exhibited excellent activity against the Gram-positive bacteria and the E. coli imp. However, it did not have activity against other Gram-negative bacteria, indicating that it may have difficulty crossing the outer membrane barrier (Table 3). Cytoskyrin B was inactive against the bacterial isolates tested (Tables 2 and 3). Minor differences in the chemical structures appear to affect the bioactivity of this class of compound (Figure 1). Among the five novel cytosporones A-E isolated from the CR200 fermentations, only D and E were active against Gram-positive bacteria and yeast (Table 3).

Inhibition of DNA, RNA, and protein synthesis was determined by measuring the incorporation of ³H-Tdr, ³H-Udr, and ³H-AA, respectively, into TCA-precipitable material of a logarithmic-phase culture of E. coli imp. Control drugs affected the anticipated macromolecular processes (Table 4). Cytoskyrin A specifically inhibited the incorporation of ³H-thymidine into DNA (Table 4). Incorporation of amino acids into protein was also affected at higher concentrations. Based on the BIA activity and the specific inhibition of DNA synthesis in E. coli imp, cytoskyrin A appears to be a potent DNA-damaging agent. Since DNA-damaging agents also constitute an important class of anticancer drugs, cytoskyrin A was tested for antiproliferative activity against four human tumor cell lines and it was found to have moderate cytoxicity with IC₅₀ values in the range of 4 to 24 μg/mL (Table 5).

Fungal culture CR200 was isolated from a branch of Conocarpus erecta (Buttonwood tree) in the Guanacaste National Park, Costa Rica (collection permit number 246.94). CR200 was streaked on Difco potato dextrose agar (PDA) and grown for 7–14 days at 22 ºC. Mycelia and conidia were washed with potato dextrose broth (PDB) containing 25 % glycerol, and 1 mL aliquots were transferred into cryovials for long-term storage at -140ºC at Wyeth Research, Pearl River, NY. An additional copy is kept at the repository of the National Institute of Biodiversity, INBio Code 12652.

For macroscopic morphology, an agar plug of CR200 was placed in duplicate onto PDA, cornmeal agar (CMA, Oxoid) and Leonian’s agar (LA, 1.2 g KH₂PO₄, 0.6 g MgSO₄, 6.25 g maltose, 6.25 g Difco malt extract, 15.0 g Bacto agar/L of distilled water) and the culture plates were incubated at 22 °C and 33 °C. Colony morphology and size (in mm) were noted on the 7^th, 14^th, and 21^st days of incubation. For microscopic analysis, CR200 was grown on PDA for 21 days. The conidiomata were observed at 1x to 6.3x magnification under an Olympus brand stereoscope. A smash prep was prepared by aseptic excision of several conidiomata, which were placed on a microscope slide with water (10 μL) and gently flattened with a coverslip. Microscopic observations were performed with an Olympus B071 phase-contrast microscope at 50x and 1,250x magnification.

Isolation of the genomic DNA, PCR amplification and direct sequencing of the nuclear ribosomal ITS1-5.8S-ITS2 region have been described previously [13]. Briefly, mycelia (50–100 mg) were lysed in lysis buffer (0.5 mL, 2% Triton X-100, 1% SDS, 100 mM NaCl, 10 mM Tris pH 8.0, 1 mM EDTA). Proteins were precipitated by the addition of 25:24:1 phenol/chloroform/isoamyl alcohol (0.5 mL, Sigma). The aqueous and organic layers were separated by centrifugation at 14,000 rpm for 5 minutes. The aqueous layer was pipetted into a new microfuge tube and the DNA was precipitated by addition of 3 M sodium acetate (0.1 mL) and isopropanol (0.8 mL). The nuclear ribosomal ITS1-5.8S-ITS2 region was amplified with primers ITS1 and ITS4 [14]. Amplification of the ITS was confirmed by gel electrophoresis. The PCR product was cleaned and sequenced directly with an ABI 3700 sequencer with the ABI Prism DNA sequencing kit and Big Dye terminators version 3.0 (Applied Biosystems).

For this study, the ITS were sequenced for Cytospora sp. CR200, Cytospora cincta ATCC 32673, Cytospora chrysosperma ATCC 56609, Cytospora leucostoma ATCC 42549 (deposited as Leucostoma persoonii), Cytospora rhizophorae ATCC 38475, and Cytospora sacchari ATCC 32322 and assigned the following GenBank accessions: DQ996039-DQ996044.

To determine the most closely related fungi, the ITS of CR200 was compared to other sequences in the GenBank database by BLASTN 2.2.2 analysis. The ITS of CR200 and several Cytospora spp. (Table 6) were aligned in Clustal X 1.81 with the multiple alignment parameters set at default values. A phylogenetic tree (Figure 2) was generated in TREECON 1.3b [15] with the distances calculated according to the Jukes and Cantor [16] method and insertions and deletions not taken into account. Bootstrap values were calculated using 1,000 bootstrap samples, and neighbor-joining was used to infer tree topologies. The phylogenetic tree was rooted with Diaporthe phaseolorum.

Fermentation of CR200 for the production of the cytoskyrins was performed in a 6-liter batch. For seed fermentation, culture mycelia grown for seven days on Bennetts’ agar (pH 7.0, containing per liter: dextrose 10 g, beef extract 0.77 g, yeast extract 1 g, NZ amine 2 g, agar 15 g) were scraped into thirty 25x150 mm Pyrex culture tubes each containing PDB (11 mL). Seed tubes were incubated for four days at 22 °C and 160 rpm. On the fourth day the contents of the seed tubes were combined into a sterile vessel and gently macerated to obtain a homogeneous inoculum. One hundred Erlenmeyer flasks (250 mL), each containing sterile PDB (60 mL) were inoculated with the seed culture (5% v/v), and fermented for 21 days at 24 °C and 200 rpm. Additional flasks (five flasks for each condition) were inoculated to study the effects of temperature (22 °C versus 28 °C) and aeration (shaking at 100 rpm versus 200 rpm) on the production of cytoskyrin A, which was measured by testing whole broths in the BIA. Fermentation of CR200 for the production of the cytosporones was performed in a 10-liter batch. First-stage seed conditions were the same as described above except that only three seed tubes were inoculated. For the second stage, the contents of the seed tubes were added to PDB (500 mL in a 2.8 L Fernbach flask) and incubated at 22°C, 200 rpm for three days. For production fermentation, 10 L of PDB in a 10 L bioreactor (Bioflo 3000, New Brunswick Scientific) were inoculated at 5% v/v with second-stage and incubated at 24 °C, 350 rpm for 15 days.