| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Plant Pathology, University of California at Davis, Davis, California 95616
James M. Trappe
Department of Forest Science, Oregon State University, Corvallis, Oregon 97331-5752
David M. Rizzo
Department of Plant Pathology, University of California at Davis, Davis, California 95616
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
Genea and Genabea are considered ectomycorrhizal (EM) symbionts of higher plants, but because of their hypogeous habit, dark coloration and the small size of their ascomata, relatively little is known about these genera. Ascomata of six morphological species of Genea and one of Genabea were frequently collected at a single site in xeric Quercus woodlands of California’s Sierra Nevada foothills. While most collections were easily referred to known species, those putatively identified as Genea harknessii and Genea arenaria were problematic. Genea harknessii collections appeared relatively homogenous based on morphology, but significant ITS variation revealed by rDNA sequencing suggested cryptic species diversity. Specimens of G. arenaria approximated the brief, original species description except for abundant clumps of septate setae formed at the apex of peridial warts. To verify the identity of this species we reexamined the holotype and analyzed morphology and ITS sequences of G. arenaria ascomata from a wide geographic range. To authenticate the EM status of Genea and Genabea with Quercus we collected healthy EM of Quercus douglasii and Quercus wislizenii and compared their ITS sequences to those from ascomata. We detected nine distinct ITS types of Genea and Genabea on roots. Two new species described here as Genea bihymeniata sp. nov. and Genea cazaresii sp. nov., were discovered during study of herbarium specimens. A phylogenetic analysis of 28 s rDNA from Genea and Genabea indicated three distinct lineages: Genea, Genabea and a third represented by Genea intermedia. For the latter we propose Gilkeya gen. nov. to accommodate the single known species, Gilkeya compacta comb. nov. A dichotomous key to all known Genea, Genabea and Gilkeya spp. from western North America is presented.
Key words: Ascomycota, cryptic species, Mediterranean climate, mycorrhiza, hypogeous fungi, Pezizales, Pyronemataceae
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
, Maia et al 1996 and references therein, Tedersoo et al 2006). Numerous taxa within the Pezizales form EM; yet paradoxically community studies generally report low numbers of pezizalean EM, in terms of both species diversity and biomass (Avis et al 2003, Richard et al 2005, Walker et al 2005). Current evidence suggests that Pezizales EM are uncommon in mesic, undisturbed forest ecosystems but are well adapted to xeric or otherwise extreme environments and proliferate after disturbance (Warcup 1990a, Taylor and Bruns 1999, Horton and Bruns 2001, Fujimura et al 2005).
EM fungi within the Pyronemataceae seem particularly well adapted to harsh environments. They are abundant colonists of EM roots in seasonally dry, high-altitude forests (Bidartondo et al 2001, Izzo et al 2005a) and several forest types after wildfire (Warcup 1990a, Vrålstad et al 1998, Grogan et al 2000, Fujimura et al 2005). Within the Pyronemataceae both epigeous (Wilcoxina, Nothojafnea, Geopyxis, Sphaerosporella) and hypogeous taxa (Genea, Genabea including Myrmecocystis, Geopora cooperi, Hydnocystis) have been reported as EM symbionts with a variety of host plants (Warcup 1990a, Maia et al 1996, Tedersoo et al 2006).
Two hypogeous genera, Genea and Genabea, were once assigned to the Geneaceae (Trappe 1979). However morphological analysis by Pfister (1984) detected affinities among the hypogeous Genea verrucosa Vittad. and the epigeous Jafneadelphus echinatus Gamundi and J. argentinus Rifai (Pyronemataceae). Pfister (1984) transferred Genea and Genabea to the Pyronemataceae. These two genera have been widely accepted since (Trappe 1975) differentiated them by morphology. Genea has verrucose, uniserate spores, cylindrical asci, and hymenia only occasionally interrupted by sterile zones of paraphyses whereas Genabea has echinulate, unise-rate, biserate or randomly arranged spores, asci that are clavate to ellipsoid, and hymenia regularly separated into pockets by sterile zones of paraphyses or isodiametric cells (TABLE I). Upon discovery of a putative intermediate species, Genea variabilis, Zhang (1991) synonymized Genabea as a subgeneric ranking within Genea.
|
). Both form EM with woody plants and have been found with a wide variety of putative EM hosts including Abies, Larix, Pinus, Tsuga, Pseudotsuga, Betula, Fagus, Quercus, Carpinus, Cistus, Nothofagus, Lithocarpus, and Corylus (Ceruti et al 1967, Fontana and Centrella 1967, Froidevaux 1977, Stewart and Heblack 1979, Kropp and Trappe 1982, Zhang 1991, Alvarez et al 1993). Genea and Genabea have lost the ability to forcibly discharge spores and their hymenia are enclosed in an epithecium, a layer of tissue created by the fusion of paraphyses’ distal tips (Gilkey 1954). Their ascomata produce volatile aromatics at maturity, suggesting dependence on animal mycophagy for spore dispersal (Parks 1921, Maser et al 1978). Their spores have been reported from the intestines and scat of a wide array of small mammals including flying squirrels (Glaucomys), voles (Clethrionomys and Microtus), and mice (Per-Peromyscus, Apodemus and Chelemys). (Maser et al 1978, Maser et al 1985, Perez Calvo et al 1989, Blaschke and Baumler 1989, Maser and Maser 1988)
Genea and Genabea have long been considered ectomycorrhizal (Fontana and Centrella 1967) but only recently have been confirmed as EM symbionts by ITS sequencing from colonized roots (Izzo et al 2005b, Tedersoo et al 2006). Although Genabea EM have not been described, all known Genea EM share important morphological features including (i) orange-brown to dark brown color (growing darker in age and at the tips), (ii) thick-walled, septate, emanating hyphae lacking clamp connections and (iii) densely branching monopodial-pinnate to pyramidal root tips (Fontana and Centrella 1967, Jakucs 1998, Brand 1991, Tedersoo et al 2006). The mantle tissues of Genea EM, with their polygonal, pigmented, thick-walled cells, share a strong morphological resemblance with the peridial tissues of Genea ascomata.
In this study we combine morphological and molecular data obtained from field collections and herbarium specimens to clarify taxonomic status and substantiate the EM habit in several species of Genea and Genabea. Specifically, our objectives were to (i) verify the EM status of Genea and Genabea with Quercus spp. by matching ITS sequences from ascomata with those of EM roots, (ii) clarify species boundaries within field collections of Genea and Genabea and (iii) examine phylogenetic relationships within Genea and between Genea and Genabea.
In addition phylogenetic and morphological data suggest that Genea intermedia Gilkey is distinct from Genea and Genabea and is best accommodated by placement in a new genus, Gilkeya. Moreover examination of herbarium specimens revealed two previously unknown species described here as Genea bihymeniata sp. nov. M.E. Sm. & Trappe and Genea cazaresii sp. nov. M.E. Sm. & Trappe.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
Ascomata of Genea and Genabea were collected under Q. douglasii and Q. wislizenii Jan 2001–Apr 2005. A garden cultivator was used to carefully remove litter and soil at random locations beneath mature host canopies. Ascomata were described, photographed and taken to the laboratory for tissue sampling and drying on the same day. All herbarium specimens were obtained from the Oregon State University Mycological Collection (OSC).
We sampled EM from 126 separate soil cores (Q. douglasii n = 94, Q. wislizenii, n = 32) on four dates during winter and spring 2003 and 2004. Litter was removed and soil cores of ca. 900 cm3 were collected under canopies of mature trees, stored at 4 C and processed within 15 d of sampling. Soil was sieved and washed by hand. After cleaning, 100 healthy EM roots per core were selected randomly, pooled and lyophilized for later DNA extraction and molecular analysis. Because no morphotyping was used to sort EM root tips before DNA extraction, a diverse array of EM fungi besides Genea, Genabea and Gilkeya was present on roots. These data will be presented in future publications (Smith 2006). Most authors consider the presence of a Hartig net as essential to the definition of an EM (Smith and Read 1997) but others regard the presence of a fungal mantle on healthy roots as sufficient to indicate EM status (Warcup 1990b). For this study roots were considered healthy and colonized by EM fungi when they lacked root hairs and displayed signs of tip swelling or a fungal mantle (color change and/or emanating hyphae). Roots with signs of decay such as dark cortex tissue or dry, flaking exteriors were discarded.
Morphological examination of ascomata.— –
Macroscopic characters were described from fresh specimens when possible. In the case of herbarium specimens, macroscopic descriptions are based on photographs, field descriptions when available and the appearance of dried ascomata. Colors are designated by ISCC-NBS terminology (Kelly and Judd 1955). Microscopic characters of all specimens were determined from hand-cut sections stained with cotton blue, heated over an alcohol lamp, rinsed and mounted in deionized water. KOH was not used because alkaline solutions dissolve the spore ornaments of Genea and Genabea (J. Trappe pers obs).
Molecular techniques for ascomata.— – We sequenced the internal transcribed spacer region of the rDNA (ITS1/ 5.8 s/ITS2) for several representative specimens of each morphological species. The Genea harknessii Gilkey morphological species group showed significant divergence in the ITS, so we attempted to sequence as many ascomata as possible to ascertain ITS diversity within the species complex. For herbarium material, specimens were selected for ITS sequencing based on a combination of morphological characters and location. Specifically, ITS sequences were obtained for ascomata of putative G. arenaria from a wide geographic range and exhibiting a wide diversity in development of peridial setae. In addition ITS sequences were obtained from specimens that were identified as putatively undescribed species (G. bihymeniata and G. cazaresii). Sequences of the ITS rDNA from various Genea and Genabea lineages could not be unambiguously aligned, so we also sequenced approximately 550 bp of the adjoining 28s rDNA for one to four representative ascomata and EM of each species group for phylogenetic analysis.
Small pieces of clean, dry tissue were cut or shaved from ascomata with a sterilized scalpel. Fruiting body material was ground with a micropestle in either CTAB extraction buffer or TE buffer. DNA was extracted by a modified CTAB method (Gardes and Bruns 1993) or QIAGEN Stool Kit (QIAGEN Inc., Valencia, California). PCR was performed with one or more of three primer combinations: ITS1F and ITS4, ITS1F and LR3, or LROR and LR3 (Gardes and Bruns 1993, Hopple and Vilgalys 1994, White et al 1990). PCR conditions were optimized for each primer combination, with the general reaction protocol as follows: initial denaturation of 94 C for 5 min followed by 25 cycles of 1 min (94 C), 1 min (55 C), 2 min (72 C) with a final extension of 72 C for 7 min.
PCR products were viewed on 1.5% agarose gels stained with SYBR Green I (Molecular Probes, Eugene, Oregon). When multiple bands occurred on a gel, PCR products of appropriate molecular weight were gel purified with BioRad Freeze ‘n’ Squeeze Tubes (BioRad, Hercules, California) and re-amplified with appropriate primers. PCR products were cleaned before sequencing with Microcon-PCR Spin Tubes or Montage PCR96 Filter Plates (Millipore, Billerica, Massachusetts). DNA sequencing was performed with the same primers as above with the ABI Big Dye Terminator Sequencing Kit (v3.1) and read with an ABI13730xl capillary sequencer (Applied Biosystems, Foster City, California) at the College of Agricultural and Environmental Sciences Genomics Facility, University of California, Davis. Sequences were edited with Sequencher v.4.1 (Gene Codes Inc., Ann Arbor, Michigan), exported to ClustalX (Chenna et al 2003) for initial alignment, and final alignment was performed by hand in MacClade (Swofford 2001).
Molecular techniques for EM.— –
Lyophilized EM root tips were ground with a micropestle and DNA extracted by a modified CTAB method (Gardes and Bruns 1993) followed by purification with a MO-BIO Soil DNA kit (MO-BIO Laboratories, Solana Beach, California). PCR was performed with primers ITS1F (Gardes and Bruns 1993) and LR3 (Hopple and Vilgalys 1994). The reaction protocol began with initial denaturation of 94 C for 5 min followed by 20 cycles of 1 min (94 C), 1 min (55 C), 4 min (72 C) with a final extension of 72 C for 7 min. PCR products were cloned with a TOPO–TA kit (Invitrogen, Carlsbad, California). At least 48 successful clones per reaction were grown overnight in Luria-Bertani (LB) media amended with 100 µg/mL of ampicillin. Cloned fragments were re-amplified in a PCR reaction with approximately 0.5 µL of the bacterial suspension as template. Amplicons were digested with the restriction enzymes AluI and Hinf I and electrophoresed through a 1.5% agarose gel and stained with SYBR Green I. The first 48 clones were inspected visually and scored to determine the number of different restriction fragment length polymorphism (RFLP) types for each core. One to four representative clones of each type were sequenced with ITS1F and in some cases with LR3 and/or ITS4. Sequences were processed and analyzed as above. In all we sequenced 600–700 bp of the ITS region from 33 representative ascomata and 45 EM root tip clones of Genea, Genabea and Gilkeya.
Phylogenetic analysis.— –
All phylogenetic analyses were performed with PAUP* 4.0b10 (Swofford 2001). Gaps were treated as missing data. The 28s rDNA data set of 550 bp was analyzed with maximum parsimony, maximum likelihood and neighbor joining. Neighbor joining analysis was performed with PAUP* 4.0b10 default settings, except that the Kimura-2 parameter was selected for calculating DNA/ RNA differences. Maximum parsimony was performed with a full heuristic search with 1000 random stepwise addition replicates and TBR (tree bisection-reconnection) branch swapping and equal weighting of all characters. For maximum likelihood analysis, the neighbor joining tree generated from the above analysis was used as the starting tree for a full heuristic search with TBR branch swapping. An appropriate model of nucleotide substitution was selected with the akaike information criterion (AIC) in Mr.Modeltest (Nylander 2004). The general time reversible (GTR) with rates that vary over sites according to the invariable sites plus gamma (I + G) model was selected. The relative robustness of individual branches and clade stability were estimated for the neighbor joining, parsimony, and maximum likelihood analyses by bootstrapping (Felsenstein 1985). Bootstrap values were calculated with 1000 replicate searches on all characters with 10 random sequence addition replications and TBR branch swapping. Sequences of the 28s rDNA of four species of Pyronemataceae were retrieved from GenBank and used as outgroups (Melastiza contorta [AY500539], Aleuria aurantia [AY544654] Byssonectria terrestris [AY500531], Cheilymenia stercorea [AY544661]). To elucidate the relationships within the Genea harknessii group we performed neighbor joining (NJ) analysis on ca. 450 bp of the ITS rDNA with the same settings as above and Genea gardneri Gilkey (Trappe 13007, DQ206860
[GenBank]
) as outgroup. The level of variation between ITS sequences within the Genea harknessii group and among geographically isolated specimens of Genea arenaria were calculated with the pairwise base differences calculator in PAUP* 4.0b10.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
). Ascomata were found between January and May, but specimens collected early in the season sometimes lacked mature spores. Ascomata were sorted into five species groups after examination of macroscopic and microscopic morphological features: Genabea cerebriformis (Harkn.) Trappe, Gilkeya compacta (Harkn.) M.E. Sm. & Trappe, comb. nov., G. arenaria Harkn., G. gardneri Gilkey, and G. harknessii Gilkey. Specimens in the G. harknessii morphological group were collected most often (n = 13), while Gilkeya compacta was least common (n = 2) (TABLE II). The number of ascomata per collection varied among species. Genea arenaria and Genabea cerebriformis commonly were found in clusters of 3–10 ascomata, whereas Gilkeya compacta and G. gardneri were found as isolated specimens or in small clusters of 2–3 ascomata.
|
We sequenced ca. 550 bp of 28 s rDNA from 13 representative ascomata and seven representative EM for phylogenetic analysis. Unambiguous alignment of 28 s rDNA sequences included 583 characters of which 123 were phylogenetically informative. Neighbor joining, maximum parsimony and maximum likelihood analyses generally gave similar results. The single most parsimonious tree of 315 steps is shown (retention index [RI] = 0.8079, consistency index [CI] = 0.6762, FIG. 1). Clades resolved in all three analyses and receiving high bootstrap support (70 or greater) are designated with an asterisk. All three analyses suggest that Genabea is distinct from Genea. The Genea harknessii group forms a well supported clade. G. bihymeniata consistently forms a sister group with G. gardneri, as suggested by overall morphology and similarities in spore ornamentation, but support for this clade is moderate (FIG. 1). Genea sp. (src680) (a single immature ascoma) and Genea sp. (src680-like) (EM roots) form a well supported clade that is distinct from the other dark-colored Genea spp. from western North America (G. harknessii, G. gardneri and G. bihymeniata). The position of G. cazaresii was not stable between the three different analyses but G. cazaresii always fell within Genea and was distant from the other brown-colored species, G. arenaria. The phylogenetic position of Gilkeya compacta was similarly unresolved, but this species was never clustered within Genea or Genabea (FIG. 1).
|
). Careful re-examination of microscopic morphological features such as spore dimensions, spore ornamentation and peridium thickness revealed notable variation among the G. harknessii specimens, but none of the morphological traits we measured could be used to consistently separate the various lineages.
|
, FIGS. 1, 2). We examined the dynamics of Genea and Genabea on EM roots via two measures: (i) the frequency of root cores where Genea/Genabea was detected (the "core view") and (ii) the abundance of Genea/ Genabea clones from each core (the "clone view").
Individual Genea and Genabea lineages differed in frequency of occurrence on EM roots. Genea harknessii (src616) was most frequently encountered lineage, with 58 clones in nine root cores (7.1% of all cores, 1% of all clones) whereas Genea sp. (src680-like) was the least frequently found with only four clones in three root cores (2.4% of all cores, 0.1% of all clones). Genea and Genabea lineages were encountered 48 times in 45 separate root cores (three cores had two Genea/Genabea lineages); 35.7% of all root cores that we sampled contained at least one Genea or Genabea lineage. A total of 195 Genea and Genabea clones were detected out of the 6101 clones that were screened with RFLP, accounting for approximately 3.2% of all clones.
Based on the "core view," Genea and Genabea were frequently encountered symbionts of Quercus; 35.1% of cores from under Quercus douglasii had at least one representative lineage, 37.5% from Q. wislizenii. Individual Genea and Genabea lineages varied considerably in EM colonization of the two Quercus spp. For example Genea harknessii group (lineage 2) was found seven times (7.4% of cores) and Genea sp. (src680-like) was found three times (3.2% of cores) on Q. douglasii, but these lineages were never found on Q. wislizenii. In contrast Genea arenaria was found only once (1.1% of cores) on Q. douglasii but three times (9.4% of cores) on Q. wislizenii. Whereas the "core view" indicates that Genea and Genabea are present in many root cores, the "clone view" suggests that these fungi usually account for only a small amount of the total fungal DNA on EM roots in any given core. Genea and Genabea lineages never accounted for more than 21 out of 48 clones (43.8%) in any root core and were often represented by only a single clone. Genea and Genabea sequences obtained from EM roots can be found in GenBank under these accessions: DQ206853 [GenBank] –DQ206855 [GenBank] , DQ206865 [GenBank] –DQ206867 [GenBank] , and DQ206969 [GenBank] –DQ206971 [GenBank] .
|
TAXONOMY |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
A Genea sporis subglobosis globosis, peridio vinaceo et absentia caespitis basalis hypharum differt.
Differing from Genea by its globose spores, vinaceous peridium, and lack of a basal tuft of hyphae. Type species: Hydnocystis compacta Harkn., Proc Cal Acad Sci. 3:214. 1899.
Etymology: in honor of Professor Helen Gilkey (1886–1972), pioneering taxonomist at Oregon State University and internationally recognized expert on taxonomy of hypogeous Ascomycota.
Gilkeya compacta (Harkn.) M.E. Sm. & Trappe, comb. nov.
Hydnocystis compacta Harkn. Proc Cal Acad Sci 3:214. 1899.
Genea intermedia Gilkey Univ Calif Publ Bot 6:303. 1916.
Myrmecocystis compacta (Harkn.) Gilkey N Amer Fl., Ser. II. 1:7. 1954.
Ascomata 5–20 mm broad, subglobose to convoluted, hollow, with a small, hairless apical opening; base lacking a tuft of hyphae. Peridium pink to vinaceous, verrucose, lacking hairs, of inflated, thick-walled cells; interior of ascomata lined with an epithecium similar to the peridium. Asci 250–280 x 31–45 µm, the walls up to 2 µm thick in youth but thinning to ±0.5 µm by maturity.
Spores subglobose to globose, 28–43 x 25–38 µm excluding the ornamentation of hyaline, rounded papillae 2–3(–7) µm tall and 2–6(–15) µm broad, the interpapillary spore surface with smaller papillae and granules.
Etymology: – Latin compacta (compressed or compacted together), possibly in reference to the tendency of ascomata to fruit in groups.
Distribution, mycorrhizal hosts and season.— – Northern Oregon to southern California, rare in Idaho, apparently disjunct in central Mexico, from near sea level to ca. 1600 m elev. in the Cascades, to ca1900 m in the Sierra and San Gabriel Mountains, and up to 3200 m in Mexico; associated with Quercus, Abies, Pinus and Pseudotsuga spp. in pure or mixed stands; March through August but mostly May and June.
Collections examined:. HOLOTYPE—USA. CALIFORNIA: Placer County, Alta, May, Harkness 98, H.W. Harkness, (BPI, isotype OSC 81199). EPITYPE HERE DESIGNATED—USA. CALIFORNIA: Yuba County, University of California Sierra Research and Extension Center, Koch Natural Area, 29 Apr 2003, src718, M.E. Smith, (OSC 111705, GenBank DQ206862 [GenBank] ). OTHER COLLECTIONS—CALIFORNIA: Calaveras County, N of Melones, 26 Mar 1985, Trappe 8421, M. Castellano & Y. Want, (OSC). Del Norte County, Head of E Fork of Illinois River, 8 May 1985, Trappe 8585, M. Castellano, (OSC 45318). El Dorado County, Univ. Cal. Blodgett Forest, 6 May 1983, Trappe 7295, J. Trappe, (OSC). Fresno County, Sierra National Forest, Ross Creek Watershed, Turtle Creek, 25 Jun 1997, Trappe 28053, R. Denton, M. Castellano & K. Pendleton, (OSC). Los Angeles County, San Gabriel Mountains, Charlton Flats, 4 Jun 1983, Trappe 7365, H. Nadel, (OSC). Mendocino County, Mountain View Road (Mendocino 510), 17 miles east of Point Arena, 6 Mar 1989, Trappe 11071, M. Castellano & H. Massicotte, (OSC). Monterey County, UC Hastings Natural History Reserve, Upper Big Canyon, 5 May 1945, Linsdale 10, J. M. Linsdale, (OSC 80001). Plumas County, Lassen National Forest, Jennie Springs, 9 Jun 1994, Trappe 16499, J. Waters, (OSC). Placer County, Alta, 18 May 1974, Trappe 3904, E. Butler, (OSC). Riverside County, San Jacinto Mountains, milepost 15.5 of California 243, 24 Apr 1995, Trappe 15473, J. Klironomos, (OSC). Tuolumne County, E of Groveland, 27 Mar 1985, Trappe 8452, D. Luoma, (OSC). Ventura County, Los Padres National Forest, milepost 40 on California 33, 23 Apr 1996, Trappe 18020, M. Castellano, (OSC 61552). Yuba County, University of California Sierra Research and Extension Center, Koch Natural Area, 14 May 2003. src729, M.E. Smith, (OSC 111865). IDAHO: Bonner County, Near Blanchard, 27 Jun 1997, Trappe 20743, A. Jumpponen, (OSC 61560). OREGON: Benton County, Vinyard Mountain, 19 Apr 1971, Trappe 2622, J. Trappe, (OSC). Douglas County, Bureau of Land Management Beatty Creek Research Natural Area near summit, 18 Apr 1997, Trappe 20411, J. Trappe, (OSC 59964). Jackson County, Wellington Butte, 20 Apr 1998, Trappe 22798, R. Young, (OSC 61954). Josephine County: Lower Illinois River Road, Oak Flat, 29 Jun 1971, Trappe 2782, J. Trappe & E. Stewart, (OSC 80002). Lane County, Willamette National Forest, Mill Creek, 25 May 1993, Trappe 13726, J. Smith & D. McKay, (OSC 44569). Linn County, 3 mi W of Roaring River Fish Hatchery, 7 Apr 1971, Trappe 2601, J. Trappe, (OSC). Marion County, Turner, 10 Jul 1980, Trappe 5874, O. Bynum, (OSC 40654). Polk County, Dallas, Liberty Road, Dawson Tree Farm, 31 Mar 1986, Trappe 8873, W. Bushnell, (OSC 46820). Tillamook County, Tillamook State Forest, Ben Smith Block, 29 May 1997, Trappe 22757, D. Gomez, (OSC 61439). Wasco County, Bear Springs, 14 May 1983, Trappe 7305, F. Evans, (OSC). MEXICO: ESTADO MEXICO: Camino Amecameca-Tlamacas, 14 Aug 1972, Trappe 3404, J. Trappe, (OSC 34424).
Comments:.
As is evident from its list of synonyms, Gilkeya compacta has been shuffled among several genera in the past. Harkness (1899) originally assigned this species to the genus Hydnocystis, but Gilkey (1916) quite reasonably regarded it as closer to Genea because Hydnocystis has smooth spores. In transferring it to Genea Gilkey had to coin a new species epithet (G. intermedia) because Harkness (1899) had already used compacta for another Genea sp. However she recognized that Genea intermedia was an "intermediate" species that did not fit well in any of her generic concepts (Gilkey 1916). Later (Gilkey 1954) concluded the species would best be accommodated in the genus Myrmecocystis. Because the epithet compacta had not been used in that genus, she was able to restore Harkness’ original epithet. However Trappe (1975) transferred the type species of Myrmecocystis to the genus Genabea, under which he thereby synonymized the generic name Myrmecocystis. He regarded Myrmecocystis compacta as most closely related to Genea and accordingly restored the name Genea intermedia. Now however additional molecular data suggests that G. intermedia can be recognized as a separate genus, so we designate it as the type species of the new genus Gilkeya. The original epithet of Harkness, which has been lost, regained, and frequently transferred about, is resurrected yet again in the name Gilkeya compacta.
Because the type collection of Harkness is in poor condition, we designate Smith src718 as an epitype with its ITS and partial 28s rDNA sequences submitted to GenBank as DQ206862 [GenBank] .
Genea bihymeniata M.E.Sm. & Trappe, sp. nov. FIGS. 3–5
|
Ascoma hypogeous, subglobose to lobed and furrowed, hollow, 10–20 mm broad as dried, with an apical orifice 2–5 mm broad opening into a single, highly convoluted and irregular chamber, into which intrude ridges of varying heights and configurations. Peridium black, with rounded to subangular warts 50–110 µm tall x 100–200 µm broad, lacking a tomentum or setae. Hymenium lining the chamber, covering both sides of the intruding ridges such that two hymenial palisades often emerge in opposite directions from a single, shared subhymenium, enclosed by an epithecium similar to the peridium but sometimes lighter in color. Mycelial tuft attached at base of fruiting body. Peridium 50–110 µm thick between the warts, up to 350 µm thick at the apex of the warts; cells near surface 10–60 µm broad, the walls ±1 µm thick and dark brown, toward the gleba abruptly grading to a zone of crowded, hyaline cells 10–35(–40) µm broad, transitioning to the trama, a pseudoparenchyma that intergrades with hyphal elements 3–5 µm broad at the septa. Subhymenium poorly differentiated from the trama, up to 25 µm thick, of tightly interwoven hyphae 2–4 µm broad at the septa with occasional cells inflated up to 8 µm.
Spores hyaline, subglobose to globose, Q = 1.0–1.25, 22–25(–28) x (20–)21–24 µm excluding the ornamentation of crowded, rounded papillae 1–2(–3.5) x 3–5.5(–7) µm. Asci hyaline, thin-walled, ca. 200 x 20–25 µm. Paraphyses hyaline, thin-walled, septate, 4–7 µm broad, the tips exceeding the asci and forming a tissue of small isodiametric cells which grade to larger and thicker epithelial cells similar to those of the peridium but lining the ascomatal chamber.
Etymology.. Latin bi (two) and hymeniata (bearing fertile layers) in reference to the frequent occurrence of two hymenial layers that emerge in opposite directions from a shared subhymenium.
Distribution, mycorrhizal hosts and season.. Known only from the holotype from the Santa Margarita Ecological Reserve in Riverside County, California, under Quercus agrifolia. April.
Collection examined:. HOLOTYPE HERE DESIGNATED—USA. CALIFORNIA: Riverside County, Santa Margarita Ecological Reserve, 12 Apr 1993, Trappe17619, M. Allen, (OSC 111669, GenBank DQ206963 [GenBank] ).
Comments.. Genea bihymeniata is distinguished from other Genea spp. by the numerous infoldings into the ascoma with two hymenial layers that emerge in opposite directions from a shared subhymenium. This character can be difficult to observe in cross-sectional mounts under a compound microscope because the fused hymenial layers sometimes break apart at the area of shared subhymenium. However the double hymenia can be seen when large portions of the hymenia are carefully sectioned under a dissecting microscope at high magnification.
Because of its black peridium G. bihymeniata superficially resembles two other black Genea spp. known from California, G. gardneri Gilkey and G. harknessii Gilkey. Genea harknessii is easily distinguished from the other two species by its relatively tall and narrow conical to fork-, anvil- or truncate-tipped spore ornaments that are usually sparsely distributed across the spore surface. Both G. gardneri and G. bihymeniata have rounded spore ornaments that are crowded on the spore surface, but G. gardneri has larger spores (28–34 x 32–36 µm) and lacks the fused hymenial layers found in G. bihymeniata. Sequences of the 28s rDNA further confirm that G. bihymeniata is distinct from both G. harknessii and G. gardneri but is more closely related to G. gardneri.
Genea cazaresii M.E. Sm. & Trappe, sp. nov. FIG. 6
Ascomata hypogaea, subglobosa, ±10 mm lata, orificio apicali 1–2 mm lato, unilocellata. Peridium brunneolum vel rubrobrunneum, verrucis rotundatis vel angulatis 50–150 µm altis x 150–250 µm latis, in verrucis usque ad 550(–650) µm crassum, in sulcis inter verrucis minus crassum (ad 120 µm). Locellus ascomi epithelio laevi, pallidiore quam peridio. Asci 170–220 x 14–20 µm, cylindracei, inamyloidei, 8-spori, in epithelio inclusi. Sporae hyalinae, subglobosae 18–21 x (12–)14–16 µm sine ornamento verrucarum rotundatarum, saepe dispersarum, ±0.5 x 0.3–1 µm, pagina sporarum inter verrucis punctata. Holotypus hic designatus Trappe 18044 (OSC 111670).
Ascomata hypogeous, subglobose, hollow, 10 mm broad as dried, with an apical orifice 1–2 mm broad opening to a single, rounded chamber and a basal tuft of hyphae that is darker than the peridium. Peridium light brown to reddish brown, with rounded to subangular warts 50–150 µm tall x 150–250 µm broad, lacking a tomentum or setae. Epithecium lining the chamber, smooth and paler than the peridium with scattered brown spots when viewed under a dissecting microscope. Peridium highly variable in thickness due largely to variation in the development of peridial warts, mostly 450–550 µm but sometimes as thin as 120 µm in deep troughs between peridial warts and as thick as 650 µm at the apex of peridial warts, with two layers: the outer layer 30–150(–250) µm thick, of several tiers of pigmented, isodiametric cells 10–35(–50) µm broad, gradually to abruptly grading to the inner layer 70–200 µm thick, of tightly constructed pseudoparenchyma with both isodiametric and hyphal elements, abruptly differentiated from a two layered trama: the outer layer 40–100 µm composed of tightly interwoven hyphae 2–5 µm broad with scattered cells inflated up to 10 µm; the inner layer 50–80 µm thick, composed of hyphae 3–8 µm broad at the septa, the cells mostly inflated to 5–15 µm, intergrading with the subhymenium 20–30 µm thick of interwoven hyphae 2–5 µm at the septa, with some cells inflated up to 10 µm.
Spores hyaline, ellipsoid to broadly ellipsoid, Q = 1.2–1.4, 18–21 x (12–) 14–16 µm excluding the ornamentation of widely spaced, rounded, hyaline warts 0.5 µm tall x 0.3–1 µm broad, the intervening spore surface appearing punctate-roughened at x400–1000 magnification; spore walls 1–1.5 µm thick. Asci hyaline, thin-walled, 170–220 x 10–13 µm, 8-spored. Paraphyses hyaline, thin-walled, septate, 2.5–5 µm broad, the tips exceeding the asci and grading into a poorly developed, pseudoparenchymatous epithecium 20–70 µm thick with occasional thick–walled cells 10–20 x 5–10 µm protruding from the surface and patches of pigmented isodiametric cells forming the brown spots visible under the dissecting microscope.
Etymology.. In honor of Efren Cázares, distinguished mycologist and collector of the holotype specimen of G. cazaresii.
Distribution, mycorrhizal hosts and season.. Coastal California between Los Angeles and Monterrey up to ca. 600–700 m elevation; associated with a variety of EM hosts including Quercus agrifolia, Arbutus menziesii, Alnus rhombifolia, Pseudotsuga macrocarpa, and Salix sp.; April–May.
Collections examined:. HOLOTYPE HERE DESIGNATED—USA. CALIFORNIA: Los Angeles County, Angeles National Forest, Prospect Campground, 18 Apr 1993, Trappe18044, E. Cázares, (OSC 111670, GenBank DQ206863 [GenBank] ); Monterey County, UC Hastings Natural History Reserve, 22 May 1945, Linsdale49, J.M. Linsdale. (OSC 111673); 24 May 1945, Linsdale66, J.M. Linsdale. (OSC 111671); 24 May 1945, Linsdale68. J.M. Linsdale. (OSC 111672).
Comments.. Genea cazaresii is easily distinguished from other Genea spp. by its combination of a brown peridium lacking setae and tomentum, small spores with inconspicuous ornamentation, and a poorly developed, hyaline epithecium. The only other brown Genea spp. known from western North America are G. arenaria Harkn. and G. compacta Harkn. Both are easily distinguished from G. cazaresii by their large and obvious spore ornamentation. Sequences of the ITS and 28s regions of the rDNA indicate that G. cazaresii is distinct from G. arenaria, but no suitable material was available for rDNA sequencing of G. compacta.
Genea arenaria Harkn. Proc. Cal. Acad. Sci. 3: 214. 1899. FIGS. 7–10
Ascomata hypogeous, subglobose to lobed and furrowed, hollow, 8–22 mm broad as dried, with an apical orifice having a round to irregularly wavy opening to the single, rounded to irregular chamber, in occasional specimens broadly open (cupulate). Peridium light brown to brown, with rounded to subangular warts 50–100 µm tall x 70–250 µm broad, usually with a sparse brown tomentum in the furrows and scattered to abundant setae tending to cluster on the warts (easily broken off in exposed areas, hence especially concentrated in the furrows). Epithecium lining the ascomatal chamber, similar to the peridium but sometimes lighter in color. Mycelial tuft attached at base of the ascomata, usually darker than the peridium. Peridium 50–100 µm thick between the warts, up to 300 µm thick at the warts; cells near surface 15–40 x 15–30 µm, the walls ± 1 µm thick and brown, toward the gleba abruptly grading to a narrow zone of cells 10–20 µm broad, then abruptly transitioning to the narrow subhymenium of isodiametric cells 5–10 µm broad. Apex of peridial warts with clusters of irregularly septate setae 250–600 (–900) µm long and 8–12 (–15) µm wide at the base, tapering gradually to 3–6 µm wide at the tips. Spores hyaline, ellipsoid to broadly ellipsoid, Q = 1.25–1.67, 20–28 x 15–23 µm excluding the ornamentation of cones 0.5–2.5 (–4) x 0.5–2 (–3) µm with obtuse to truncate, anvil-topped or forked tips. Asci hyaline, thin-walled, ca 200 x 18–22 µm. Paraphyses hyaline, thin-walled, septate, 3–4 µm broad, the tips exceeding the asci and forming a tissue of small isodiametric cells which grade to larger, pigmented, thick-walled cells similar to those of the peridium but lining the ascomatal chamber.
Etymology.. Latin, arenaria ("sandy"), in reference to the soil in which Harkness found the type collection.
Distribution, mycorrhizal hosts and season.. Northern Oregon to southern California in coastal and montane forests up to ca. 1 500 m elevation. Usually associated with Quercus. At lower, drier sites found February–April; at higher, wetter sites found April–June.
Collections examined:. USA. CALIFORNIA: locality and date not recorded, Harkness 42, H.W. Harkness (HOLOTYPE: BPI, isotype OSC 81193); Alameda County, University of California Berkeley Campus, 26 Nov 1901, UC225, N.L. Gardner, (OSC 111674); Los Angeles County, Angeles National Forest, Angeles Forest Highway, 17 Apr 1993, Trappe12975, M. Castellano, (OSC 111685, GenBank DQ206858 [GenBank] ); Angeles National Forest, Lake Hughes Road, Prospect Campground, 18 Apr 1993, Trappe13005, M. Castellano, (OSC 111684, GenBank DQ206840 [GenBank] ); Mariposa County, Greely Hill Road off California 120, 27 Mar 1985, Trappe8470, M. Castellano, D. Luoma, Y. Wang, (OSC 111689, GenBank DQ206838 [GenBank] ); Monterey County, UC Hastings Natural History Reserve near Robertson Creek, 12 Jun 1945, Linsdale91, J.M. Linsdale, (OSC 111675); Riverside County, San Bernardino National Forest, San Jacinto Mountains along Fulmer Mill Creek, 3 Jun 1983, Trappe7345, H. Nadel, (OSC 111678, Genbank DQ206841 [GenBank] ); San Bernardino National Forest, San Jacinto Mountains along Black Mountain Road, 24 Apr 1995, Trappe15483 and Trappe15487, M. Castellano, (OSC 111683 & OSC 111969), GenBank DQ206847 [GenBank] ); San Bernadino County, San Bernardino National Forest, Highway 138 to Crestline, 22 Apr 1996, Trappe18000, M. Castellano, (OSC 111682, GenBank DQ206837 [GenBank] ); San Diego County, Cleveland National Forest, Palomar Mountain, 3 Mar 1990, Trappe 11458, G. Menser, (OSC 111687, GenBank DQ206856 [GenBank] ); Santa Clara County, Almaden Quicksilver County Park, 13 Feb 1985, HS2375, H. Saylor, (OSC 111681, GenBank DQ206836 [GenBank] ); Sierra County, Tahoe National Forest, Wild Plum Campground, 9 Jun 1989, Trappe11171, N. Weber and T. O’Dell, (OSC 111688, GenBank DQ206839 [GenBank] ); Yuba County, UC Sierra Research and Extension Center, Koch Natural Area, 19 Feb 2002, src398. M.E. Smith, (OSC 111691, GenBank DQ206842 [GenBank] ); 9 Mar 2002, src430, M.E. Smith, (OSC111692, GenBank DQ206845 [GenBank] ); 20 Feb 2003, src615, M.E. Smith, (OSC111694, GenBank DQ206844 [GenBank] ); 20 Feb 2003, src634, M.E. Smith, (OSC); 1 Mar 2003, src652, M. Smith, (OSC 111693, Genbank DQ206833 [GenBank] , DQ206966 [GenBank] ); 21 Mar 2003, src681, M. Smith, (OSC 111695); 24 Apr 2003, src699, M.E. Smith, (OSC); 12 Feb 2004, src829, M.E. Smith (OSC). OREGON: Jackson County, Applegate Road, Ramsgate Ranch just E of Josephine County line, 5 Apr 1991, Trappe11737, J. Trappe, (OSC 111680, GenBank DQ206848 [GenBank] , DQ206968 [GenBank] ); Near Emmigrant Lake, 2 Apr 2004, Frank644, J. Frank, (OSC 111690, GenBank DQ206835 [GenBank] ); Josephine County, Applegate Road, Lazy A. Ranch, 30 Mar 1991, Trappe11733, M. Amaranthus, (OSC 111686, Genbank DQ206834 [GenBank] ); Polk County, SE of Buell, 10 May 1971, Trappe2656, J. Trappe, (OSC 82222); Valley Junction, 30 Mar 1966, Trappe520, J. Trappe, (OSC 111677, GenBank DQ206843 [GenBank] ).
Comments..
We examined the macroscopic and microscopic morphological features of 25 collections of putative Genea arenaria from a wide geographic range (FIG. 11) stretching from southern California to northern Oregon and compared them with the holotype collection (Harkness 42). Unfortunately very little information is available for the holotype because Harkness crafted a cursory description of the species based on a single collection with no additional information about associated plant hosts, date, elevation or geographic area (Harkness 1899). The type collection now consists of ascoma fragments in poor condition with little suggestion of tomentum or setae on the peridial surface. Several of our collections had abundant setae clustered at the apex of the peridial warts, leading us to think we had discovered an undescribed species. However other anatomical features, especially the spores, matched those of the type. To add to the confusion (Gilkey 1939) described G. arenaria as "scatteringly hispid with long, brown, septate hairs", but it was unclear whether her description was based on the holotype. Examination of many other collections revealed considerable variation in the abundance of tomentum and setae, from nearly absent to abundant. After viewing the variation among many putative G. arenaria collections, we re-examined permanent mounts of the holotype (Harkness 42) and were able to discern a few broken setae similar to those found on fresh ascomata.
|
Harkness (1899) described another brown species, Genea compacta, based on a single specimen from northern California. Genea compacta is distinct from G. arenaria because it lacks peridial setae and has larger spores (32–34 x 24–28 µm) with large, rounded ornaments that are crowded on the spore surface. Harkness referred to G. compacta as "exceedingly rare" and the species is apparently only known from two collections (Harkness 86 and Trappe 1361). Unfortunately both are in poor condition and we were unable to obtain DNA of sufficient quality for ITS sequencing. A potentially related species, G. asperula Trappe, Guzmán & T. Herrera, was described from a Pinus-Quercus forest in southern Mexico (Trappe and Guzmán 1971). It differs from G. arenaria in having a spore ornamentation of low, rounded warts and asci with thickened walls. A similar species found under Quercus spp. in Mexico was ascribed to G. arenaria by Cázares et al (1992), but it has slightly larger spores and longer asci than either G. arenaria or G. asperula and might represent an undescribed species.
Genea arenaria has been collected in mixed forests with diverse EM plants (Arbutus menziesii, Arctostaphylos spp., Quercus agrifolia, Q. chrysolepis, Q. douglasii, Q. garryana, Q. kelloggii, Abies concolor, Pinus coulteri, P. lambertiana, P. ponderosa, Pseudotsuga menziesii, P. macrocarpa, Salix spp., Alnus spp.) but seems always under Quercus or in forests where Quercus is present. This, in conjunction with confirmation of its symbiotic association with Q. douglasii and Q. wislizenii, suggests that G. arenaria may preferentially or obligately form EM with Quercus.
|
OTHER COLLECTIONS EXAMINED IN THIS STUDY |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
|
KEY TO GENEA, GENABEA, AND GILKEYA OF WESTERN NORTH AMERICA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSTAXONOMYOTHER COLLECTIONS EXAMINED IN...KEY TO GENEA, GENABEA,...DISCUSSIONLITERATURE CITED |
|---|
A second possibility is that Genea and Genabea have seldom been found on roots because current methods in EM research are more effective at detecting epigeous Basidiomycota than hypogeous fungi and Ascomycota. Many studies identify fungi from single EM root tips using RFLP matching to fruiting bodies or by direct ITS sequencing, yet hypogeous fungi and EM Ascomycota are underrepresented in both RFLP and public sequence databases (Izzo et al 2005b, M. Smith personal observation). For example three Genea species included in this study (Genea balsleyi nom. prov., Genea arenaria and Genea harknessii –lineage 1) showed strong affinity (ITS BLAST identities greater than 95%) to unidentified pezizalean DNA sequences from EM roots and orchid mycorrhiza root sections (AY634166
[GenBank]
, AY351627
[GenBank]
and AY310834
[GenBank]
) (data not shown). In addition Ascomycota EM tips are morphologically and physiologically distinct from Basidiomycota EM; they are usually smaller, have less emanating hyphae and rarely produce rhizomorphs (Danielson 1984, Berndt et al 1990, Warcup 1990a, Smith and Read 1997). This means that Ascomycota EM likely contain less rDNA template than Basidiomycota EM and thus might be more challenging to amplify with PCR. Because our approach bulks many EM tips together and employs a DNA extraction kit that removes a wide variety of PCR inhibitors, it is possible this method detects Ascomycota EM more effectively than traditional, single tip methods.
A third possibility is that Genea and Genabea have been overlooked on roots because they are common in seasonally dry, angiosperm-dominated woodlands but not in the mesic and boreal coniferous forests that have been the focus of much EM research. Although ascomata of Genea and Genabea have been collected in moist habitats such as temperate rain-forest (Kropp and Trappe 1982) most reports of putative Genea EM roots are from drier, angiosperm-dominated forests (e.g. Fontana and Centrella 1967, Avis and Charvat 2005, de Roman and de Miguel 2005). In addition many of the putative Genea sequences in GenBank are from areas with a pronounced dry season (M. Smith personal observation). Based on taxonomic studies of ascomata both Genea and Genabea appear to be most diverse in low-elevation woodlands and savannas (Gilkey 1916, Fontana and Centrella 1967, Cázares et al 1992, Alvarez et al 1993).
Our surveys of Genea and Genabea at a single site in a Quercus-dominated woodland yielded an unexpectedly high level of diversity, with five morphological species groups from ascomata but 11 distinct ITS types from ascomata and EM roots. The Genea harknessii morphological group was particularly diverse with five related ITS types encountered at a small spatial scale. The ITS region generally shows low variation within species (Horton and Bruns 2001, Horton 2002) and ITS variation is often congruent with variation at other loci (e.g. Douhan and Rizzo 2005, Geml et al 2006). Pairwise comparisons showed 3.8–8.3% variation between the five G. harknessii ITS types and phylogenetic analysis yielded five strongly supported lineages (FIG. 2). Despite significant diversity in the ITS, mature ascomata from three of the G. harknessii lineages could not be morphologically differentiated, suggesting the possibility of five cryptic phylogenetic species. Two additional lineages, src680 & src680-like, appear distantly related to all other black-colored Genea spp. from western North America. However, because these lineages were detected only as a single immature ascoma and an ITS sequence from EM roots, further collections will be needed before they can be adequately characterized.
In contrast to the cryptic diversity within the Genea harknessii group, Genea arenaria showed moderate morphological diversity but limited genetic diversity over a large geographic range. We found a low level of ITS diversity (ca. 1% bp differences in pairwise comparisons) among 17 G. arenaria specimens collected in diverse habitats. This suggests that environmental or developmental, rather than genetic factors, might be responsible for the morphological variation in setae and other characters. In addition old or poorly preserved collections often lacked setae across exposed peridial surfaces but retained rare to scattered clusters of setae within protected furrows. This observation implies that some phenotypic diversity in setae length and abundance is actually due to poor preservation of specimens rather than actual morphological variation.
Examination of herbarium
