An unusual fungus producing minute orange stilboid sporocarps was found on a palm leaf mid-rib in a Neotropical forest. Morphological observations could not place this collection into any previously described species or genus and, due to an absence of sexual structures, even higher level placement was uncertain. Phylogenetic analysis of a portion of the large subunit and the internal transcribed spacer of the nuclear ribosomal DNA indicated that this fungus is related to Heterogastridium pycnidioideum and belongs to Heterogastridiales, Microbotryomycetes (Pucciniomycotina). A new genus and species, Pycnopulvinus aurantiacus, are proposed here to accommodate this fungus.

Ceratocystis , Fungal biodiversity, litter fungi, palm fungi, Pycnobasidium , tropical mycology

The majority of Pucciniomycotina (Basidiomycota) species have life cycles that include the production of microscopic fruiting structures (e.g., spermogonia of rust fungi), but only a few species within the subphylum form macroscopic fruiting bodies (Swann et al. 2001; Aime et al. 2014). Although these sporocarps vary in form (e.g., see Aime et al. 2014), many are stilboid or pycnidioid with spore masses produced at the base of the sporocarp and exiting through the tip of the neck. Fungi with such fruiting bodies can be found mostly in Atractiellomycetes (Oberwinkler and Bandoni 1982) and, as is true for many other members of Pucciniomycotina, little is known about their biology or biodiversity.

Heterogastridium pycnidioideum Oberw. & R. Bauer (anamorph Hyalopycnis blepharistoma (Berk.) Seeler) is one such fungus that produces stilboid fruiting bodies. It is a strictly filamentous species with simple septal pores and specialized organelles–colacosomes–associated with mycoparasitism (Oberwinkler et al. 1990). Heterogastridium is placed in Heterogastridiaceae and Heterogastridiales (Oberwinkler et al. 1990), which are now placed in Microbotryomycetes (Weiss et al. 2004). Bauer et al. (2006) placed three genera in Heterogastridiaceae in addition to Heterogastridum–Atractocolax R. Kirschner, R. Bauer & Oberw., Colacogloea Oberw. & R. Bauer, and Krieglsteinera Pouzar. Of these, neither sequence data nor reference cultures are available for two of the genera, Atractocolax and Krieglsteinera, both of which are monotypic. Colacogloea, on the other hand, has been demonstrated several times to form a separate clade of Microbotryomycetes, distant from Heterogastridiales (e.g. Aime et al. 2006; Kurtzman et al. 2011). Thus, at present, only Heterogastridium and its sole species, H. pycnidioideum, can be confidently assigned to Heterogastridiales with molecular data.

An unusual stilboid fungus was discovered on the mid-rib of a palm leaf in litter that could not be confidently assigned to any previously described genus. DNA sequence data and phylogenetic analyses indicated that the collection represents a new member of Heterogastridiales. Herein we describe and illustrate Pycnopulvinus aurantiacus gen. et sp. nov. and provide a phylogenetic analysis of Heterogastridiales.

The specimen (PUL F2679) was collected near Bilsa Biological Station in Ecuador, in the vicinity of N0.350444, W79.732075, on 3 May 2004, where it was growing on the mid-rib of a dead palm leaf in the litter. The fungus was photographed and described in the field after which small pieces of substrate bearing the fruiting bodies were dried on an herbarium drier. Color designations refer to Kornerup and Wanscher (1978). Duplicates are deposited in the Kriebel Herbarium at Purdue University (PUL) and the herbarium of the Pontificia Universidad Catolica del Ecuador (QCA).

Morphological characters of the fruiting bodies were observed first with an Olympus SZ61 dissecting microscope. With the aid of the dissecting microscope, a few fruiting bodies were carefully removed from the substrate and placed in a sterile water droplet on a microscope slide, permitting five minutes of rehydration before further preparation. Thereafter microscopic characters were examined with a Nikon Eclipse 80i microscope with standard differential interference contrast (DIC) settings and with 10×, 20×, 40× and 100× objectives. The length and width of 20 spores was measured from three different fruiting bodies with an ocular micrometer using 100× oil-immersion objective. Images were taken with Nikon Digital Sight DS-Fi1 camera setup and measurements were calibrated with a stage micrometer.

For molecular characterization, five dry fruiting bodies were carefully removed from the leaf surface to avoid the inclusion of leaf material and potential contaminants. These were used for DNA extraction with the E.Z.N.A. High Performance DNA Kit (Omega Bio-Tek Inc., Norcross, GA, USA), following the manufacturers’ instructions for samples with lower DNA content (protocol 3). PCR reactions were carried out in 25 μL reactions that contained 12.5 μL of Apex Taq RED Master Mix (Genesee Scientific, San Diego, CA, USA), 1.25 µL of each primer (10 µM), 5 µL of molecular grade water and 5 µL of temfig DNA. Amplification of the ITS region was conducted with primer pair ITS1F (Gardes and Bruns 1993) and ITS4 (White et al. 1990), and the LSU region was amplified with LR0R and LR5 (Vilgalys and Hester 1990). Amplification conditions followed Toome et al. (2013). Sequencing of amplified fragments was performed by Beckman Coulter, Inc. (Danvers, MA), using the same primers that were used for amplification. Sequences were edited with Sequencher 5.2.3 (Gene Codes Corporation, Ann Arbor, MI, USA) and are deposited in GenBank.

A BLASTn analysis in GenBank (http://www.ncbi.nlm.nih.gov) was used to locate similar sequences for phylogenetic analyses. For the LSU dataset, sequences sharing >92% identity with PUL F2679 were selected and for the ITS dataset sequences sharing >85% identity with PUL F2679 were included. Sequences from Rhodotorula hylophila and R. javanica were added based on their relatedness according to the analyses of Kurtzman et al. (2011) and R. yarrowii was included for rooting purposes. Only the sequences from type strains were retained for previously described species. Accession numbers for sequences used are indicated on Figure 1.

Figure 1. 

Maximum likelihood tree illustrating the placement of Pycnopulvinus aurantiacus in relation to other known members of Heterogastridiaceae. The analysis was performed with combined LSU and ITS sequence data and the topology was rooted with Rhodotorula yarrowii. The numbers above and below branches show the bootstrap and posterior probability values, respectively. LSU and ITS GenBank accession numbers of each used strain are given in the brackets. Superscript letter T indicates sequence data that originate from the type.

Both datasets were aligned separately using the Muscle algorithm in MEGA 5.2 (Kumar et al. 2008), yielding 695 bp and 605 bp alignments for the ITS and the partial LSU region, respectively. These two alignments were subsequently concatenated for phylogenetic analyses. The final alignment is available in TreeBASE (http://treebase.org) under accession number 15679.

Phylogenetic analyses were performed via the CIPRES Science Gateway (Miller et al. 2010). The maximum likelihood (ML) analyses were conducted in RAxML-HPC2 7.6.3 using the -k option for bootstrap analysis. Bayesian posterior probability analyses were conducted with MrBayes 3.2.2 with parameters set to 10 000 000 generations, two runs and four chains. The resulting phylogenetic tree was edited in Incscape (http://www.inkscape.org).

Sporocarps of PUL F2679 had a swollen cushion-like basal region, measuring 0.2 to 1 mm in diameter after drying. The basal region was pigmented, ranging from light to dark orange when fresh and appearing orange-brown after drying. This base supports a narrow synnemata-like structure with a 1 to 2 mm long neck. At the apex of the neck, a light yellow to orange mucous droplet of spores is formed (Figure 2a–f). While the size of sporocarps was variable, all of them produced a spore-containing droplet. Although it could not be definitely determined, the spores appear to be asexually produced. Also, despite numerous attempts to isolate the fungus into pure culture from revived spores, no isolate was obtained. Only an isolate of Ceratocystis paradoxa (deposited in GenBank as AY821864; CBS 116770) was frequently recovered during these attempts.

Figure 2. 

Pycnopulvinus aurantiacus (holotype PUL F2679). a Field photo of fresh sporocarps on palm leaf. Note the variable size and color of the sporocarps. Bar = 2 mm b–d Sporocarps of various stages and sizes after drying. Bars = 0.5 mm e Sporocarp as seen under the light microscope. Note the swollen pigmented base with surrounding large globose cells and the long tubular neck with a widening tip. Bar = 200 μm f Tip of the neck with spore mass exiting the sporocarp. Bar = 25 μm g Ostiolar hyphae at the tip of the neck. Bar = 10 μm h Outer surface of the neck, wide hyphae are visible, septa are marked with arrows. Bar = 10 μm i Globose cells (marked with asterisk) surrounding the base of the sporocarp. Bar = 25 μm j Multicellular spores produced inside the sporocarps. Note the four-celled spores on the left breaking into smaller compartments. Bar = 10 μm.

Of previously sequenced species, PUL F2679 shared the most sequence identity with H. pycnidioideum (= H. blepharistoma) at 92% (414 bp of 452 bp) in the LSU region and 86% (515 bp of 596 bp) in the ITS region. The only other close match was to a previously sequenced but undescribed isolate, CBS 196.95 (GenBank no. LSU–AY323905; ITS–AY323904), which shared 98% identity (446 bp of 453 bp) in the LSU region and 99% identity (609 bp of 616 bp) in the ITS region. Results of phylogenetic analyses are presented in Figure 1.

We are grateful to Prof. Rosario Briones of Pontificia Universidad Catolica del Ecuador, Quito for facilitating specimen deposition and exchanges at Pontificia Universidad Catolica del Ecuador (QCA), to Drs. Harry Evans and Gary Samuels for facilitating permitting and collecting logistics in Ecuador and companionship in the field, and the staff and volunteers at Bilsa. Cindy Park is gratefully acknowledged for laboratory assistance at the USDA. We are thankful to Tom Creswell, Gail Ruhl and Anna Meier from the Purdue University Plant and Pest Diagnostic Laboratory for use of microscopy facilities and to Mehrdad Abbasi assisting with submission to Kriebel Herbarium at Purdue. We are also thankful to the two anonymous reviewers for constructive comments and suggestions on an earlier version of this manuscript.