| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Plant Pathology, Agronomy College, Shanxi Agricultural University, Taigu, Shanxi 030801, China
Irina S. Druzhinina
Section of Applied Biochemistry and Gene Technology, Institute of Chemical Engineering, TU Wien, Getreidemarkt 9-166.5, A-1060 Wien, Austria
Payam Fallah 2
Priscila Chaverri 3
Pennsylvania State University, Department of Plant Pathology, 301 Buckhout Laboratory, University Park, Pennsylvania 16802
Cornelia Gradinger 4
Christian P. Kubicek
Section of Applied Biochemistry and Gene Technology, Institute of Chemical Engineering, TU Wien, Getreidemarkt 9-166.5, A-1060 Wien, Austria
Gary J. Samuels 5
USDA-ARS, Systematic Botany and Mycology Laboratory, Room 304, B-011A, BARC-West, Beltsville, Maryland 20705-2350
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONDESCRIPTIONS OF THE SPECIESKEY TO HYPOCREA/TRICHODERMA...LITERATURE CITED |
|---|
We describe or redescribe species of Hypocrea/Trichoderma (Ascomycetes, Hypocreales) having hyaline ascospores and pachybasium-like conidiophores. Teleomorphs are reported for Trichoderma minutisporum (Hypocrea minutispora sp. nov.) and T. polysporum (H. pachybasioides). Hypocrea pilulifera/T. piluliferum is redescribed. Trichoderma croceum is synonymized with T. polysporum. The new species H. parapilulifera, H. stellata and H. lacuwombatensis are described. All of these species fall within the morphological concept of Trichoderma sect. Pachybasium and within the phylogenetic group pachybasium B5 of Kullnig-Gradinger et al (2002)
. Parsimony analysis of nucleotide sequences from three unlinked loci—ITS1 and 2, endochitinase (ech42) and translation elongation factor 1-alpha (tef1)—detects two distinct phylogenetic lineages within the group pachybasium B5. One comprises H. pachybasioides/T. polysporum, H. pilulifera/T. piluliferum, H. parapilulifera and H. stellata; this group, the "polysporum" lineage, is characterized by having conidia that are white in mass and is the only lineage within Hypocrea characterized by such conidia. The second group includes the green conidial T. minutisporum and H. lacuwombatensis. The partition homogeneity test reveals significant recombination within the "polysporum" lineage but not within the "minutisporum" lineage.
Key words: anamorph-teleomorph connection, Ascomycetes, endochitinase gene, Hypocreaceae, Hypocreales, ITS, molecular phylogenetics, systematics, translation elongation factor tef1
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONDESCRIPTIONS OF THE SPECIESKEY TO HYPOCREA/TRICHODERMA...LITERATURE CITED |
|---|
included most of the described species of the genus. These fungi were characterized by having relatively short and broad phialides arising from, and often clustered on, wide branches of the conidiophores. The section included species that produce sterile or fertile elongations of their conidiophores along with a few species that lack any elongations of the conidiophores. It also included species that have green conidia as well as species in which conidia are white or yellow in mass. Despite morphological similarity among the species originally included in the section (Bissett 1991b), DNA sequence analysis has shown it to be paraphyletic (Kindermann et al 1998, Kullnig-Gradinger et al 2002) and separable into two major phylogenetic groups that have been termed clade A and clade B (Kullnig-Gradinger et al 2002). Clade A, which includes the neotype of the type species of sect. Pachybasium, T. hamatum (Bon.) Bain., is derived from within sect. Trichoderma. Most of the pachybasium-like species are placed in clade B and form a highly diverse but monophyletic sister group of sect. Trichoderma. This large group of species has not been given formal taxonomic status within Hypocrea/Trichoderma and is referred to here as pachybasium B. The genetic diversity of Trichoderma sect. Pachybasium not withstanding, the morphological descriptor "pachybasium-like" is informative and used throughout the present work to describe a characteristic type of conidiophore.
Several taxa belonging to pachybasium B are characterized by green conidia and conidiophores from which sterile or fertile elongations arise (Bissett 1991b). Chaverri et al (2003) recently reviewed the taxonomy of these species. They described Hypocrea strictipilosa Chaverri & Samuels as the teleomorph of T. strictipile Bissett and described or redescribed but did not name pachybasium-like anamorphs for the new and closely related species H. cremea Chaverri & Samuels, H. cuneispora Chaverri & Samuels, H. estonica Chaverri & Samuels and H. surrotunda Chaverri & Samuels. All of those species were included in one clade in pachybasium B.
Apart from the species mentioned above, other taxa from pachybasium B have green conidia but lack elongations of conidiophores. In addition, some species have hyaline (white in mass) or yellow conidia; depending on the species, their conidiophores may have sterile elongations. Four of these species (T. polysporum Rifai, T. croceum Bissett, T. minutisporum Bissett and H. pilulifera Rifai & J. Webster) previously have been shown to form one strongly supported clade (subclade B5) in pachybasium B in a multigene phylogenetic analysis. In this paper, we have investigated this clade in detail. We synonymize T. croceum under T. polysporum Rifai and link that species to Hypocrea pachybasioides Doi. We link T. minutisporum Bissett to the new Hypocrea species H. minutispora. We redescribe the teleomorph and anamorph of H. pilulifera and describe the new species Hypocrea parapilulifera, H. stellata and H. lacuwombatensis and their Trichoderma anamorphs.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONDESCRIPTIONS OF THE SPECIESKEY TO HYPOCREA/TRICHODERMA...LITERATURE CITED |
|---|
|
Growth and colony characterization. –
Growth trials were performed to determine the growth rate and optimum temperature for growth following the protocol of Samuels et al (2002) on PDA (Difco potato-dextrose agar) and synthetic low-nutrient agar (SNA, Nirenberg 1976). The isolates were grown in the dark and the colony radius was measured at 24, 48, 72 and 96 h at 15, 20, 25, 30 and 35 C. Each growth-rate experiment was repeated three times and the results averaged for each isolate. The time of first appearance of conidia, the presence of yellow pigmentation of young conidia, the presence of diffusing pigment in the agar, odor and colony appearance also were noted.
Morphological observations. – Morphological observations of the anamorph were taken from cultures grown on CMD in 9 cm diam vented plastic Petri plates in an incubator at 20 C with alternating 12 h fluorescent light and 12 h darkness within 14 d. These standard characters were measured from 3% KOH or distilled water: width of phialide at the base, phialide width at the widest point, phialide length, cells supporting phialides, presence of chlamydospores and chlamydospore width. Measurements of continuous characters were taken from images using the beta 4.0.2 version of Scion Image (Scion Corp., Frederick, Maryland). Measurements of asci, ascospores and anamorph characters in species descriptions are reported as maxima and minima in parentheses and the mean plus and minus the standard deviation of a minimum of 30 measurements. Four types of compound microscopy were used, viz. bright field (BF), phase contrast (PC), differential interference contrast (DIC) and fluorescence (FL). Preparations studied for fluorescence microscopy were prepared by flooding preparations that had been used for measurements with Calcofluor (Sigma Fluorescent Brightener 28, C.I. 40622 Calcofluor white M2 in a 2 molar phosphate buffer at pH 8.0). Colony appearance was described from CMD at 20 C and PDA at 25 C with alternating 12 h fluorescent light and 12 h darkness, including formation and shape of tufts or pustules. The presence of chlamydospores was recorded by examining the reverse of a colony grown on CMD after ca 1 wk at 20 C under 12 h darkness and 12 h cool white fluorescent light with 40x objective of a compound microscope.
Illustrations of conidiophores and conidia were taken from colonies grown on CMD at 20–21 C under 12 h darkness/12 h cool white fluorescent light for 7–10 days, unless otherwise noted.
The herbarium specimens of Hypocrea were rehydrated briefly in 3% KOH. Rehydrated stromata were supported by Tissue-Tek O.C.T. Compound 4583 (Miles Inc., Elkhart, Indiana) and sectioned at a thickness of ca. 15 µm with a freezing microtome. Permanent preparations of the sections were made following Volkmann-Kohlmeyer and Kohlmeyer (1996). These teleomorph characteristics were evaluated: diameter, height, color and shape of the stroma; texture of surface of the stroma; perithecium shape, length and width; reaction to 3% KOH, color, width of perithecium wall; ostiolar canal length; color and 3% KOH reaction of stroma outer region; shape, diameter and wall thickness of cells of the outer, middle (immediately below the outer region) and inner region (below perithecia) of the stroma; ascus length and width; distal and proximal partascospore length and width. Measurements of continuous characters were obtained with Scion Image beta 4.0.2. Statistical analyses were performed using Systat 10.0 (SPSS Inc., Chicago, Illinois).
DNA extraction, PCR amplifications and sequencing. –
DNA was isolated from fresh mycelium as described previously (Turner et al 1997). A region of nuclear rDNA, containing the ITS1 and 2 and the 5.8S rRNA gene, was amplified by PCR using the primer combinations SR6R and LR1 in 50 µL volumes (White et al 1990) in an automated temperature-cycling device (Biotron, Biometra, Göttingen), using the protocol described in Kullnig-Gradinger et al (2002). A 0.2 kb fragment of tef1 was amplified by the primer pair tef1fw and tef1rev. A 0.4 kb fragment of ech42 was amplified by the primer pair Chit42-1a and Chit42-2a (Kullnig-Gradinger et al 2002). Template DNA (100 µL) was prepared directly from PCR products by purifying it with a commercial kit (Cleanmix; Talent s.r.l., Trieste) and sequenced with the aid of a LI-COR 4000L automatic sequencing system, using cycle-sequencing (Robocycler 40; Stratagene, La Jolla, California) with the ThermoSequenasekit (Amersham Biosciences Inc. Piscataway, New Jersey) as described in Kindermann et al (1998).
The sequences of isolates submitted to GenBank, as indicated in TABLE I, are the sequences used in all phylogenetic analyses. Sequences not obtained during this work are given by their respective GenBank accession numbers. The nucleic acid matrix—ITS1 and 2, tef1, and ech42 combined tree FIG. 2—has TreeBase ID number SN1486.
|
), using M13 as a primer.
Molecular phylogenetic analysis. –
DNA sequences were aligned first with ClustalX 1.81 (Thompson et al 1997) and then visually adjusted, based on the algorithm of Waterman (1986) using Gendoc 2.6.002 (Nicholas and Nicholas 1997). Gaps were treated either as missing or as a fifth character. Phylogenetic analyses were performed in PAUP* 4.0b10 using Trichoderma hamatum DAOM 167057 as outgroup. Parsimony analysis was performed using a heuristic search, with a starting tree obtained via stepwise addition, with random addition of sequences with 1000 replicates, tree-bisection-reconnection as the branch-swapping algorithm, Multrees in effect. Stability of clades was assessed with 500 bootstrap replications. The Partition Homogeneity Test (PHT) integrated in PAUP was used to test the congruence among different gene datasets (Cunningham 1997). For this test parsimony-uninformative characters were excluded, gaps were treated as missing, and 10 000 repetitions were performed. A maximum of 100 trees were saved to conserve memory.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONDESCRIPTIONS OF THE SPECIESKEY TO HYPOCREA/TRICHODERMA...LITERATURE CITED |
|---|
) and of ech42 (encoding the 42-kDa endochitinase, Lieckfeldt et al 2000a). The ITS1 and 2 tree had the fewest parsimony-informative polymorphic sites (30; corresponding to 7.4%; TABLE II). Both tef1 and ech42 produced a high number of parsimony-informative characters (41 and 117, corresponding to 36.2 and 24.3%). Phylogenetic analysis using gaps as a fifth base did not give strongly different tree topologies but resulted in higher bootstrap support for the clades. Both trees obtained from the analysis of single genes only (FIG. 1a, b), as well as that from a combined analysis of ITS1 and 2, tef1 and ech42; FIG. 2), and using T. hamatum (sect. Trichoderma) as outgroup, grouped these species in a strongly supported (98% when gaps were treated as missing data) monophyletic group, thus confirming the previously defined cluster B5 in pachybasium B (Kullnig-Gradinger et al 2002).
|
|
The minutisporum clade received high bootstrap support both in the combined (ITS1+2, ech42 and tef1) as well as the individual (tef1, ech42) trees and includes the named species T. minutisporum ATCC 28012, which previously was misidentified (Kuhls et al 1996) as T. hamatum, and several green-spored Hypocrea species. A new Hypocrea species, G.J.S. 99-198 from New Zealand, described here as H. lacuwombatensis, always occurred basal to this cluster.
The polysporum clade was less strongly but still reasonably well supported by bootstrap coefficients and includes the described species T. croceum, H. pilulifera/T. piluliferum and T. polysporum. The ex-type culture of T. croceum and the epitype culture of T. polysporum clustered in subclades within the larger polysporum clade, but neither subclade received strong support. Two other branches, each leading to species doublets (viz. G.J.S. S 90-63 and G.J.S. 90-116; G.J.S. 86-540 and G.J.S. 90-126) always received strong support within this cluster. However, both are morphologically indistinguishable from H. pachybasioides (see below) and thus might be at the stage of clonal isolation only.
Three sets of Hypocrea isolates (viz. H. pilulifera, isolate G.J.S. 91-60/G.J.S. 99-188, and isolate G.J.S. 99-222) always clustered at a basal position of the polysporum clade. Low bootstrap support for their branches indicates that these three sets represent separate lineages.
The partition homogeneity test (Huelsenbeck et al 1996) was used to examine the null hypothesis of recombination (Koufoupanou et al 1997). The actual summed tree length of 576 steps was exactly at the lowest limit of that produced by any of the 10 000 artificial datasets (P-value 0.002; FIG. 3), and four steps shorter than >95% of them, thus indicating incongruence among the different gene trees. However, the P-value (gene partition incongruence value) was 0.0549, which is slightly more than P = 0.05, below which incongruence is considered significant. We therefore have investigated whether this would be due to recombination ocurring only within one of the clusters in pachybasium B5. In addition, we removed the ITS sequences from the analysis because of the unequal and biased mutation rate of rDNA genes (Maynard-Smith and Smith 1998). Thus the partition homogeneity test was applied separately to each of the two clades, with and without using ITS1 and 2 sequences (TABLE IV). The corresponding data (FIG. 3, TABLE IV) clearly show that recombination is only apparent within the polysporum clade, whereas it appears to be absent from the minutisporum clade. Within the polysporum clade recombination only occurs within T. polysporum/T. croceum/H. pachybasioides but not between them and H. stellata, H. parapilulifera and H. pilulifera.
|
|
|
). In contrast to the polysporum clade, where several isolates were collected in Australia and New Zealand, only one strain (G.J.S. 99-198) in the minutisporum clade originated in Australasia. Most T. minutisporum isolates came from North America (eastern U.S.A. and Canada), with only one isolate collected outside of North America (CBS 901.72, Germany).
Phenotype analyses. –
All isolates included in the present study have a pachybasium morphology as was broadly defined by Bissett (1991b) in which phialides tended to be short, wide and more or less densely clustered on broad cells. The phialides of T. piluliferum differs from the typical pachybasium morphology in being more divaricate. The phialides typically were slightly narrower than the cell from which they arose and on average 1.5 times as wide. Each of the two major clades, revealed by DNA sequence analysis, was characterized by conidial color (viz. green in the minutisporum clade and white in the polysporum clade). There were no green-conidial strains in the polysporum clade and no white-conidial strains in the minutisporum clade. The clades also differed in growth rate. Although all the isolates grew slower on PDA at 25 C than is usual of Trichoderma, isolates in the polysporum clade rarely reached a colony radius of 25 mm after 72 h at 25 C in in the dark, whereas the isolates of the minutisporum clade reached a radius of approximately 35 mm under those conditions.
Stromata of all collections studied were typical of Hypocrea in being lightly to brightly pigmented and fleshy. Perithecia were completely immersed. Asco-spores were bicellular but disarticulated early in their development to give 16 unicellular part-ascospores in each ascus. Part-ascospores of all collections were hyaline, dimorphic and finely spinulose. The stromata of all collections tended to be less than 5 mm diam, solitary and sharply delimited, circular to elliptic in outline, with margins free from or united with the substratum. Stromata of the Japanese specimens of H. pachybasioides, which were isolated from Quercus logs infected with shiitake (Lentinula edodes, TABLE I), tended to be aggregated or gregarious and often were effused and the margins were not free from the substratum. In all cases the stromatal surface was plane and the apices of individual perithecia were not evident or barely rose above the stroma surface. Each stroma contained many perithecia, the openings of which were easily seen as dark dots against the lighter surface of the stroma. The youngest, developing stromata tended to be vinaceous and to have a lighter-colored margin, but as stromata aged the tendency was to become a light brown or buff, or to remain vinaceous. The color variation was not associated with any of the subclades of either T. minutisporum or T. polysporum. There was no reaction to 3% KOH.
The stromata of all the Hypocrea specimens were generalized to the extent that one could not predict which anamorph any collection would give. There were no single or combined gross morphological, pigmentation or anatomical features of the teleo-morph that characterized any of the clades. The only exception to this is the stroma of H. pilulifera, the type of which formed on an herbaceous substratum (Juncus effusus). However, the ex-type culture of its anamorph, T. piluliferum, was isolated from wood, as were all the other ascospore collections (but not the conidial isolates, which came from a diversity of sources but mainly soil). Apparently there are no extant ascospore isolates of T. piluliferum.
There were no anatomical differences in the stromata or perithecia in any of the species included in this study. The cells of the stroma surface were typically angular, 3–7 x 2.5–6.5 µm, walls <1 µm thick and easily seen. All stromata were divided into three regions—surface, subsurface and tissue below perithecia. The surface region of the stroma was 20–35 µm thick and was composed of pigmented cells which, in face view, were 2.5–5.0 x 2–3 µm in cross section. In all cases the subsurface region was composed of intertwined, ca. 3 µm wide, septate, thin-walled hyphae. The cells below the perithecia tended to be hyphal or at least of textura intricata. Perithecia were 200–300 µm tall, 100–170 µm wide and the ostiolar canal was 40–90 µm long. The perithecial papilla only rarely protruded through the stroma surface, and the cells of the perithecial apex were not different from the surrounding cells of the stroma surface.
The asci of all species were cylindrical and sessile; ascospores were uniseriate or at most partially biseriate. The ascus apex of all collections was slightly thickened and possessed an obscure pore. Part-ascospores were more or less dimorphic. The distal part-ascospores were subglobose to ellipsoidal, rarely conical. The proximal part-ascospores were oblong to wedge-shaped, tapering slightly from the middle to the end. The distal part-ascospores of members of most subclades of the polysporum clade were 3.5–4.5 µm long and ca. 3.5 µm wide and the proximal part-ascospores were 4.0–5.5(–6.0) x 2.5–3.2 µm. The part ascospores of the subclade that includes G.J.S. 86-540 (USA: New York), G.J.S. 90-126 (USA: Virginia), and TMI 8084 ( Japan) were significantly smaller than all others in all regards (distal part-ascospores: 3.3 ± 0.4[2.5–4.6] x 3.0 ± 0.3[2.1–3.7] µm; proximal part-ascospores 3.8 ± 0.5[2.5–5.2] x 2.6 ± 0.3[1.9–3.3] µm).
Cultures derived from ascospores could not be distinguished from conidial cultures of, respectively, T. polysporum and T. minutisporum. The Hypocrea specimens that gave T. polysporum anamorphs in culture could not be distinguished from the isotype specimen of H. pachybasioides.
Most of the species in the polysporum clade had a more or less strong tendency to form discrete, white pustules on CMD or SNA. On PDA, conidial production tended to be continuous and not in pustules. The pustules of H. pachybasioides/T. polysporum were dense, and spiraled hairs were produced abundantly from each pustule. Within the polysporum clade the pustules of H. pilulifera/T. piluliferum, G.J.S. 91-60 and G.J.S. 99-222 were constructed more loosely and conidiophores with more or less long, fertile axes were conspicuous. This was especially true of G.J.S. 99-222, in which long, narrow, plumose conidiophores with short lateral branches were conspicuous (FIGS. 18, 75–79)
|
|
|
|
Conidia of most species were ellipsoidal with a length/width ratio of ca. 1.5. Conidia of H. pilulifera/T. piluliferum were subglobose, L/W = ca. 1.1. Conidia of most species had a mean of size of ca. 3.0 x 2.1 µm. Of the H. pachybasioides/T. polysporum subclades, only G.J.S. 99-207 (New Zealand) had overall larger conidia (mean = 3.5 x 2.3 µm) than other members of the clade. Conidia of H. pilulifera/T. piluliferum were wider than those of most species studied, but as was already mentioned conidia of that species were subglobose. In the minutisporum clade, conidia of G.J.S. 99-198 (New Zealand) were slightly longer (95% CI = 3.4–3.6 µm) than conidia of other members of the clade (95% CI = 3.2–3.3), all of which originated in temperate eastern North America and Europe.
Conidiophores of H. pachybasioides (FIGS. 24–27, 46, 47) and G.J.S. 91-60/G.J.S. 99-188 (FIGS. 65, 66) comprised a spiralled, thin-walled-septate, sterile hair with subacute to rounded tips, from the base of which arose fertile branches. The hairs of H. pachybasioides tended to be rugose. All other cultures produced conidiophores that had a discernible main axis that bore phialides at the tip and produced fertile lateral branches along the length. The branches tended to increase in length with distance from the tip and often were paired.
|
|
|
Chlamydospores were produced within 10 d on CMD by some collections of all clades. They were terminal or intercalary and subglobose.
Within the polysporum clade four ascospore isolates (G.J.S. 86-540, G.J.S. 90-126, G.J.S. 93-38, G.J.S. 99-244) formed a well-supported sister clade of the clade that included the ex-type isolates of T. polysporum and T. croceum. These ascospore collections had slightly but significantly smaller ascospores than is typical of T. polysporum and slightly but significantly slower growth (after 72 h on PDA maximum = 20 mm and 25–30 mm, respectively). These ascospore isolates originated from widely separated geographic stations (United States, New Zealand). However, there was continuous variation in growth rate in this clade with each subclade having a slightly different rate of growth. The optimum temperature for all isolates except G.J.S. 99-207 was 25 C; the optimum for G.J.S. 99-207 was 20 C.
The ex-type isolate of T. croceum produced yellow conidia (FIGS. 15, 16) but it was almost identical to several strains in RAPD and ITS 1+2 sequences, all of which had hyaline conidia.
The isolate G.J.S. 99-222, which was alone in a basal position in the polysporum clade, grew more slowly than any of the isolates that were studied, reaching a maximum radius of approximately 15 mm after 72 h on PDA at 20 and 25 C.
Within the minutisporum clade most isolates genetically were identical or very close to the ex-type isolate of T. minutisporum (DAOM 167069) and there was very little divergence in their respective growth rates, the optimum temperature being 25 C on PDA. The only other subclade that included more than one culture was that including G.J.S. 90-82, and all five isolates in that clade had the same growth characteristics. G.J.S. 90-112 and ATCC 28012 had slightly different growth rates from the rest, the former being slower and the latter being faster. Greater differences were seen at 30 C, where G.J.S. 90-198 was slowest (10 mm), T. minutisporum ex-type group, G.J.S. 90-112 the next slowest (ca. 15 mm), and all the rest reached a radius of 27–30 mm on PDA after 72 h.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONDESCRIPTIONS OF THE SPECIESKEY TO HYPOCREA/TRICHODERMA...LITERATURE CITED |
|---|
). This section is now known to be paraphyletic (Kindermann et al 1998, Kullnig-Gradinger et al 2002). Our work demonstrates that the white-conidial species of Trichoderma/Hypocrea and the green-spored species without apical elongations form a strongly supported clade (B5) in pachybasium B in which at least six different species can be distinguished. Although evidence for this clade confirms earlier investigations by Kindermann et al (1998) and Kullnig-Gradinger et al (2002), support for its fine structure largely came from the analysis of single-copy loci (viz. tef1, ech42). It is interesting to note that sequence analysis of the internally transcribed spacer regions 1 and 2 of the rDNA cluster, which have been shown to result in a reliable phylogeny in earlier studies (Kuhls et al 1996; Lieckfeldt et al 1998, 1999), failed to resolve the phylogeny of subclade B5. A closer investigation of the ambiguous nucleotide area revealed that it was due mainly to the occurrence of an extended AT-rich stretch within a GC-rich sequence in the 5' end of ITS1. This does not occur in other species of Trichoderma and not in phylogenetically close species in genera such as Hypomyces, Nectria or Gibberella (C.P. Kubicek, unpubl) and thus is specific for subclade B5. The origin of this sequence probably is due to an initial T
A exchange. Such an exchange can be found in some members of pachybasium B, such as T. oblongisporum Bissett, T. tomentosum Bissett or H. semiorbis Berk., which is an invariable hallmark sequence (TCAT) in species belonging to sect. Trichoderma. In a phylogenetic tree based on ITS1 sequence analysis only, subclade B5 therefore occurs as a sister clade of sect. Trichoderma (Kindermann et al 1998). Species of polysporum subclade B5 bear a duplication of this region, with CT exchanges in some isolates. On the other hand, a variable (n = 2–4) stretch of As adjacent to the G-rich stretch occurs in the minutisporum clade, which can be explained by preferential insertion of A in depurinated places (Dianov et al 1992). All in all, this indicates that this region is a major hot spot for mutations in the B5 subclade of Trichoderma/Hypocrea species and in fact there is little nucleotide variation in the remaining sequence of ITS1 and 2. The reason for this is unknown. However, Moore et al (1991) demonstrated that various families of AT-rich triplet repeats form secondary structures that escape DNA repair, which, because of the proliferated secondary structure of the internally transcribed spacer regions of rDNA, might be an explanation. Recombination might be a further reason for the variability of this region. A much higher variability was observed in the polysporum clade than in the minutisporum clade, which agrees with the results from the partition homogeneity test. These data reveal that there is considerable genetic exchange within H. pachybasioides but not among H. pachybasioides, H. pilulifera, H. parapilulifera or H. stellata, thus justifying our recognition of the latter three as species distinct from H. pachybasioides.
Recombination within the minutisporum clade appears to be rare, and this cluster is characterized by a significant clonal element despite the inclusion of a significant number of ascospore isolates of T. minutisporum in the analysis. One possible explanation would be that T. minutisporum is homothallic. Few species of Hypocrea undergo sexual reproduction in vitro. Of those, bipolar heterothallism has been demonstrated only in H. jecorina Berk. & Broome/T. reesei E.G. Simmons (Lieckfeldt et al 2000b). The more usual situation, first exemplified (Mathieson 1952, Perkins 1987) by Hypocrea spinulosa Fuckel (= Chromocrea spinulosa [Fuckel] Petch), has been described as "mating-type shifting" wherein half the spores in an ascus are self fertile and the other half are self sterile but cross compatible with the self-fertile cultures. This system is strongly suggested for H. citrina (Pers. Fr.) Fr. (= H. pulvinata Fuckel) (Canham 1969) and H. poronioidea A. Möller (Samuels and Lodge 1996). However, we do not know of any strictly homothallic species of Hypocrea.
Hyaline conidia are uncommon in Trichoderma. Species related to H. poronioidea (Samuels and Lodge 1996), H. pulvinata Fuckel (Rifai and Webster 1966) and H. sulphurea (Schw.) Fr. (Samuels unpublished) produce acremonium-like anamorphs that have hyaline conidia held in drops of clear liquid. These are never formed en mass and never appear white; they are not morphologically pachybasium-like. Bissett (1991a) proposed Trichoderma sect. Hypocreanum Bissett to accommodate them. The H. citrina group of species falls within a large clade that is sister of Pachybasium B5, where they form a subclade (termed "Pachybasium B1" by Kullnig-Gradinger et al 2002) that is sister of several species that have a typical pachybasium-like morphology. Hypocrea poronioidea produces a green-conidial Trichoderma in addition to the hyaline-conidial, acremonium-like synanamorph. The phylogenetic position of H. poronioidea is not known. Samuels and Lodge (1996) suggested that these acremonium-like anamorphs could be spermatial because species of the H. citrina group and H. poronioidea form stromata in culture and incipient stromata of H. poronioidea are associated with such conidia in culture.
The species of Trichoderma that have white conidia all have a pachybasium-like morphology. Green conidial pigment, coupled with the pachybasium-like morphology, seems to have been lost only once in Hypocrea because the species that we have included in this study appear to share a common ancestor. We know of only one species with "white" conidia that was not included in this study, viz. H. placentula Grove (Spooner and Williams 1990). On the basis of the morphology of its teleomorph and anamorph, we predict it will be closely related to H. pachybasioides and H. pilulifera.
The "bright greenish-yellow" or "rosy-buff" conidiogenous pustules and somewhat narrower conidia of T. croceum Bissett distinguished it from T. polysporum (Bissett 1991b). Unfortunately, only a single isolate (DAOM 167068) was cited in the original description of T. croceum. Widden (1979) isolated Trichoderma species from soil particles. He identified a number of morphological species, including LP63. Although each morphospecies comprised more than one isolate, LP63 is given in the DAOM catalogue of cultures as the only available isolate of T. croceum. The morphospecies LP63 was isolated rarely from soil particles in a deciduous forest of southern Ontario, although the protologue of the new species T. croceum indicates that it was isolated from soil under Pinus, according to Widden (1979). Thus the phenotypic variability of this unusual yellow-conidial form is not known and its provenance is suspect. Nonetheless, conidia of the ex-type culture of T. croceum, in addition to being yellow, are slightly but significantly longer and wider than those of the several collections that are morphologically T. polysporum. However, comparison of all isolates identical to DAOM 167068 in ITS 1 + ITS 2 sequences and >80% similar in RAPD profiles (TABLE II) revealed that the conidia of the ex-type culture of T. croceum fell within the range of all of the isolates. We conclude that T. croceum is synonymous with T. polysporum, the older name. The constancy of the yellow conidia suggests that the ex-type strain may represent a clonally isolated population of T. polysporum. Synonymy of T. polysporum and T. croceum also has been suggested by Lee and Hseu (2002) on the basis of UP-PCR and ITS1 sequence analysis.
Yellow conidia occur in at least two additional and unrelated species of Trichoderma, viz. T. crassum Bissett and T. fasciculatum Bissett. Chaverri et al (2003) found that Trichoderma crassum is closely related to T. virens ( J. Miller et al) Arx, which has green conidia, and that T. fasciculatum is a synonym of the green-conidial Hypocrea strictipilosa Chaverri & Samuels/T. strictipile Bissett. Conidia of T. stromaticum Samuels et al remain white or yellow for a long time and often do not become green (Samuels et al 2000). In a genus wanting in useful phenotypic characters, the taxonomically seductive trait of yellow conidia unfortunately may be consistent only at the strain level.
Several collections of Hypocrea produced T. polysporum and one of them (G.J.S. 99-90, from Australia) was identical to a strain of T. polysporum (CBS 820.68, from Germany) considered by Rifai (1969) to be typical of the species. These Hypocrea collections are indistinguishable from H. pachybasioides Doi (1969), and we concur with Komatsu (1976) and Gams and Bissett (1998) that the teleomorph of T. polysporum is H. pachybasioides.
Rifai (1969) reported that T. polysporum is common and cosmopolitan. Some authors have reported it to be more frequent or even restricted to cooler climates and soil around conifers (Danielson et al 1973, Söderström and Bååth 1978, Widden 1979, Widden and Abitol 1980, Smith 1995). However, the collections of teleomorphs were found on a diversity of dicotyledonous trees, never on conifer wood. Our own collections included isolates from temperate Northern America, Europe, Japan and Australia, and we are not aware of any correctly identified isolates from subtropical or tropical areas. In this regard it is intriguing that our own Northern Hemisphere collections were made over several years during which H. pachybasioides was found infrequently, while the Australasian collections reported here all were found during a single two-week collecting trip in 1999. This observation suggests a greater diversity of the species in Australasia. A similar Australasian richness was found for the H. schweinitzii (Fr.) Sacc. complex, with anamorphs in Trichoderma sect. Longibrachiatum (Kuhls et al 1997). These observations suggest that the New Zealand, Australia and the Pacific area could be a center of genetic diversification of Hypocrea.
As emphasized above, most of the genetic diversity within H. pachybasioides/T. polysporum probably is due to recombination. However, for four isolates (G.J.S. 86-540, G.J.S. 93-38 [both New York], G.J.S. 90-126 [North Carolina], and G.J.S. 99-244 [New Zealand]) some phenotypic differences were seen, because their ascospores are significantly smaller than in those of other collections and they grow more slowly on PDA than the other isolates in the clade. They also form a well-supported clade that is sister of the rest of the isolates that are morphologically H. pachybasioides/T. polysporum. In the absence of other phenotypic differences, we do not propose a separate taxon for them, yet consider them to be clonal isolates in the process of speciation.
Trichoderma minutisporum previously has not been linked to a teleomorph. The isolate CBS 901.72 was nearly identical to the type of T. minutisporum (DAOM 167069) in forming a clade with 96% bootstrap support (FIG. 2); additional ascospore isolates from North America were identical phenotypically but genetically slightly different from the ex-type isolate and from each other (FIG. 2). In characters of the teleomorph, this species scarcely can be distinguished from many other Hypocrea species; it is its anamorph that distinguishes it, and no described Hypocrea has been shown to have an anamorph with the morphology of T. minutisporum. The teleomorph is proposed here as the new species H. minutispora. Contrary to H. pachybasioides/T. polysporum, which is widely distributed in north temperate regions, being known from Asia, North America, Europe and Australasia, H. minutispora/T. minutisporum is apparently a species of temperate North America and northern Europe.
The single isolate G.J.S. 99-198 from New Zealand is genetically distinct within the minutisporum clade. Its anamorph is similar to T. minutisporum but it can be distinguished from H. minutispora/T. minutisporum in having larger conidia and in the structure of its conidial pustules. It is described below as the new species H. lacuwombatensis.
The isolates G.J.S. 91-60 and G.J.S. 99-188, and G.J.S. 99-222 are genetically distinct within the polysporum clade. The first two of these apparently are related closely to H. pilulifera/T. piluliferum (FIG. 2) but differ in having ellipsoidal conidia. It is proposed here as the new species H. parapilulifera. The isolate G.J.S. 99-222 is distinguished by pustules that comprise long fertile branches from which short lateral branches arise; many of these branches extend beyond the surface of the pustule giving the pustule a stellate aspect (FIGS. 17, 18). It is proposed here as H. stellata. We do not propose species names for the Trichoderma anamorphs of these Hypocrea species because they are known only from one or two collections and have not been encountered in nature as their anamorphs.
|
DESCRIPTIONS OF THE SPECIES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONDESCRIPTIONS OF THE SPECIESKEY TO HYPOCREA/TRICHODERMA...LITERATURE CITED |
|---|
|
|
Sporotrichum polysporum Link, Mag. Ges. Naturf. Freunde Berl. 7:34. 1815: Fries, Syst. Mycol. 3:424. 1832.
(see Rifai [1969] for extensive anamorph synonymy)
Stromata solitary, sometimes gregarious but rarely cespitose, (0.3–)0.5–2.0(–6.0) x (0.2–)0.5–1.5(–4.5) mm, shape of stromata variable, mostly rounded to elongate, or sometimes irregular in outline, particularly when aggregated or cespitose, pulvinate; usually with margins of the stromata fully attached to the substratum, but sometimes slightly constricted at the base; color variable among different collections and different stages of the same collection, mostly brown to reddish-brown, some collections yellow or light brown particularly when young, collections from Japan purple-brown to black-brown; color of most stromata nearly uniform, but margins of young stromata often white. All tissues KOH–. Stromatal surface variously wrinkled or creased, plane but infrequently slightly tuberculate from perithecial apices. Ostiolar openings mostly visible as purple-brown to blackish-brown flat or slightly raised dots (FIGS. 4–6). Cells of the stroma surface in face view elongate, angular or irregular in outline, (2–)4–8(–18) x 2–5(–8.5) µm, usually reddish-brown, rarely gray to whitish-gray, cell walls 0.5–1.0 µm thick (FIG. 35). Stroma surface region (10–)20–40(–50) µm thick, cells elongate, angular or compressed, occasionally rounded, (2–)3–5(–10) x (1–)2–4(–7) µm, cell walls 0.5–1.0 µm thick (n = 10), brown, KOH– (FIGS. 36, 37). Hairs arising from the stroma surface scattered and inconspicuous, hyphal, (4–)5–7(–9) µm long, (2.0–)2.5–3.5(–6.0) µm wide at the base, hyaline to light brown (FIG. 38). Cells immediately below the stroma surface hyphal, thin-walled, hyaline. Tissue below the perithecia tending to comprise compact textura intricata or intertwined hyphae, less frequently cells pseudo-parenchymatous, (3–)6–10(–22) x (2–)4–5(–8) µm, thin-walled, hyaline, KOH– (FIGS. 39, 40). Perithecia immersed in the stroma, densely disposed, mostly globose to subglobose, laterally compressed and pyriform to clavate when greatly compacted, (75–)150–270(–346) µm high, (53–)100–150(–232) µm wide, ostiolar canal (18–)50–80(–116) µm long, cells of the perithecial wall brown or reddish-brown to hyaline; ostiolar region not sharply delimited from the surrounding tissue of the stromal surface (FIGS. 33, 34). Asci cylindrical, (48–)75–100(–133) x (3.5–)4.5–6.0(–8.5) µm, tip thickened and with a pore (FIGS. 41–43). Part-ascospores hyaline, uniseriate, finely spinulose, dimorphic. Distal part-ascospores globose to subglobose or conical, (2.2–)3.2–4.5(–6.0) x (2.0–) 3.0–4.0(–5.7) µm. Proximal part-ascospores mostly subglobse to oblong, sometimes wedge-shaped or attenuated toward the base, tending to be more oblong toward the base of the ascus, (2.5–)3.7–5.2(–7.2) x (1.7–)2.5–3.5(–4.5) µm (FIG. 44).
Cultures and anamorph. – Optimum temperature for growth on PDA 25 C, no growth occurring at 35 C. Colony radius on PDA at 25 C after 3 d in darkness (14–)18–34(–46) mm, <15 mm at 30 C. Conidia typically not forming on PDA in darkness within 6 d, or rarely after 4 d at 20 C; yellow pigment sometimes forming in colony reverse after 4 d in darkness at 20 and 25 C. No odor detected on PDA or CMD. Colony radius on CMD at 25 C in light after 6 d 65 mm, within 10 d small white hyphal tufts scattered throughout the colony and conidia forming on the tufts after 13 d. Conidial pustules white to cream-colored, solitary or rarely aggregated, pulvinate, (0.2–) 0.8–1.5(–13.5) mm diam, dense, appearing velvety owing to numerous, projecting, sterile extensions of conidiophores (hairs; FIGS. 13,16, 45–47), production of pustules increasing to 30 d. Hairs flexuous, sinuous to corkscrew-shaped, typically attenuated to a narrow tip, septate, thin-walled, roughened by numerous small warts near the apices, unbranched for a great distance, sometimes producing a single phialide at the tip (FIGS. 24–27, 46, 47). Fertile branches of conidophores arising from the base of the hairs at approximately 90° to the hairs, or conidiophores lacking sterile extensions. 1° conidiophore branches mostly short, comprising one or few cells, and increasing in length with distance from the tip, producing phialides directly and also rebranching to produce 2° branches. 2° branches typically unicellular and producing 1–4 phialides from the tip. Phialides lageniform and more or less constricted to form a neck, (3.0–)3.7–6.7(–13.5) µm long, (1.7–)2.5–3.5 (–5.0) µm at the widest point, L/W = (0.9–)1.5–2.2(–5.0), (1.0–)1.7–2.7(–4.0) µm wide at the base. Cell supporting the phialides (1.7–)2.5–4.5(–9.7) µm wide (FIGS. 48, 50). Conidia ellipsoidal, occasionally oblong or subglobose, (1.2–)2.5–3.5(–5.5) x (1.0–) 1.5–2.5(–3.0) µm, L/W = (0.5–)1.0–2.0(–2.5), hyaline, smooth (FIGS. 26, 49). Chlamydospores terminal or intercalary within hyphae, globose or subglobose, (3.0 –)5.5–9.0(–23.5) x (2.8–)5.5–8.0(–16.0) µm, smooth or somewhat spinulose (FIG. 51).
Known distribution. – Australia, Canada, Germany, Italy, Japan, Korea, New Zealand, Switzerland, United Kingdom (England), United States.
HOLOTYPE. JAPAN. Otsuno, Kochi City, on bark, 3 May 1966, Y. Doi TNS.D-77 (TNS-F-190528, holotype not available, ISOTY PE: NY!).
Additional Hypocrea specimens examined. – AUSTRALIA. NEW SOUTH WALES: Blue Mountains, Morton National Park, vicinity of Bundnadoon, Fairy Bower Track, on bark, 19 Aug 1999, G.J.S. et al 8740 (BPI 74667, culture G.J.S. 99-155); Fairy Bower Track, on bark, 19 Aug 1999, G.J.S. 8741 (BPI 746680, culture G.J.S. 99-159). VICTORIA: vic. Healesville, Toolangi State Forest, Myers Creek Road, Wirrawalla Rainforest Walk and Myrtle Walking track in Myrtle Gully, altitude. 575 m, on bark of recently dead tree, 23 Aug 1999, G.J.S. 8599 (BPI 746812,culture G.J.S. 99-219); Otway Ranges, Melba Gully State Park, Madsen’s Track along Johanna River, altitude 350 m, on bark, 27 Aug 1999, G.J.S. 8636 (BPI 746848; culture G.J.S. 99-221 = BPI 112261); Yarra Ranges National Park, at intersection of road to Donna Buang and Acheron Way, Donna Buang Gallery along stream, altitude ca. 600 m, on bark of Nothofagus sp., G.J.S. 8607 (BPI 746821, culture G.J.S. 99-223). CANADA. QUEBEC: Gatineau State Park, on decorticated wood, possibly on an ascomycete, 20 June 1987, R.P. Korf (NY). ITALY. VITERBO PROV.: Bomarzo, along river Vezza, on wood of Quercus sp., Nov 2002, W. Gams (CBS 111723, specimen and culture). JAPAN. HIKUI PREF.: Konjyocho, Nanjyogun, 18 Jul 1968, M. Komatsu (TMI 8218). HYOGO-PREF.: Fukuzaki-cho, kanzaki-gun, 23 Feb 1965, M. Komatsu (TMI 8215). HYOYO-PREF.: Nahokaichi, Kasumi-cho, Kinosaki-gun, 13 May 1968, Akiyama (TMI 8258). TOTTORI PREF.: 4 Jul 1968, M. Komatsu (TMI 8220, 8221); Bogaki, Tottorishi, 13 Apr 1964, M. Komatsu (TMI 8084); Tottorishi, 4 Jul 1968 (TMI 8217). NEW ZEALAND. BULLER: Lewis Pass, St. James Walkway, altitude ca. 900 m, 42°24'S 172°24'E, on bark of Nothofagus menziesii, 9 Sep 1999, G.J.S. & S. Dodd 8736 (BPI 842229; culture G.J.S. 99-90 = CBS 112257). WESTLAND: vicinity of Franz Josef, trail to Lake Wombat, 43°25'S 170°21'E to 43°21'S 170°9'E, mixed podocarp forest, altitude 200–275 m, on bark, 3 Sep 1999, G.J.S. & S. Dodd 8684 (BPI 746623; culture G.J.S. 99-207 = CBS 112256); WESTLAND: vicinity of Hokitika, Lake Kaniere, Track along S side of lake between Sunny Bight and Hiiker Creek, mixed podocarp with Dacrydium cupressinum (Rimu), Prumnopitys taxifolia (Matai), and Metrosidros umbellata (Southern Rata), altitude 0 m, 43°S 171°7'E, on bark of small branches of recently dead tree, G.J.S. & S. Dodd 8672 (BPI 746612, culture G.J.S. 99-224). SWITZERLAND. KT. GRAUBÜNDEN: vicinity of Davos, Dischmatal, altitude 1500–1700 m, on decorticated wood of Picea sp., 4 Sep 1990, G.J.S. (BPI 1107148; culture G.J.S. 90-28 = BBA 70310 = CBS 112262). UNITED STATES. MARYLAND: Garrett County, 5 miles North of Barton, Little Savage River ravine, on bark, 23 Sep 1989, G.J.S. et al (NY; culture G.J.S. 89-121 = CBS 112258). NEW YORK: Dutchess County, east side of Pawling, Pawling Nature Reserve, Nature Conservancy, on bark, 8 Oct 1990, G.J.S. & C.T. Rogerson (BPI 1107154; culture G.J.S. 90-63 = CBS 112260); Hamilton County, Long Point, Racquette Lake, on decorticated wood of Fagus, 6 Sep 1986, J.H. Haines (NY; culture G.J.S. 86-540 = BBA 70311 = CBS 112267); Schuyler County, Van Etten, Arnot Forest, Cornell University, Bonfield road, vicinity of Bonfield Creek, on decorticated wood, 9 Oct 1993, K.F. Rodrigues (BPI 802508 culture G.J.S. 93-38). NORTH CARO-LINA: Clay County, Standing Indian Campground, off U.S. 64, on Aphyllophorales on Betula, 15 Oct 1990, Y. Doi, A.Y. Rossman & G.J.S. (BPI 1107182; culture G.J.S. 90-116 = CBS 112259); Jackson County, Ellicott Rock Trail, on recently dead tree, 1989, G.J.S. et al (NY, culture G.J.S. 89-135); Macon County, Blue Valley, off Clear Creek Road, along Overflow Creek, 35°00'N, 83°15'W, on fungi on bark, 16 Oct 1990, Y. Doi, A.Y. Rossman & G.J.S. (BPI 1107168, culture G.J.S. 90-126); Swain County, Great Smoky Mountains, National Park, 5 Mile North of Deep Creek Camp Ground, Indian Creek Trail, on bark, 27 Sep 1988, K.F. Rodrigues, C.T. Rogerson, G.J.S., E. Parmasto & R.H. Petersen (NY, G.J.S. 88-44); Swain County, Great Smoky Mountains National Park, 3.6 mile N of Deep Creek Camp ground, Indian Creek Trail, on decorticated wood, 27 Sep 1988, K.F. Rodrigues et al (NY, culture G.J.S. 88-59).
Commentary. – Hypocrea pachybasioides is possibly the easiest species of the genus to identify when its anamorph, T. polysporum, is known. Despite the absence of strongly diagnostic characters in the teleomorph, the anamorph of this common species is almost invariant. Several isolates of T. polysporum are listed in the culture catalogue of the Centraalbureau voor Schimmelcultures, including isolates from Germany, Korea, the Netherlands, and U.K.; we have not studied those isolates but do not doubt their identity.
The shape, degree of aggregation and color of stromata of H. pachybasioides are variable. Typically stromata are rounded to elongate; when aggregated they are irregular in outline. Most of the collections from Japan were cespitose in lines, and stromata of one Japanese collection were somewhat effused over the substratum. The stromata of most collections are reddish-brown (i.e., G.J.S. 99-90, 99-116, 99-155, 99-207, 99-219, 99-220, 99-221, 99-222, 99-223, TMI 8215, TMI 8218); immature stromata tend to be yellow to yellowish-brown (i.e., G.J.S. 88-44, 88-59, 89-135, 90-63, 99-159, TMI 8258) and old stromata are dark brown to rusty (i.e., G.J.S. 99-224, TMI 8217, 8220, 8221). The specimen TMI 8084, from Japan, was unusual in having pale vinaceous stromata; ascospores of this collection were smaller than is typical of the species.
Trichoderma polysporum frequently is cited in the biological control and ecology literature (Domsch et al 1980). It is one of the slowest-growing species in the genus. Its phenotypic similarity to the distantly related, common soil fungus Tolypocladium inflatum W. Gams (= To. niveum [O. Rostrup] Bissett, nom. rejic.) has led to its misidentification (Samuels 1996). The species has been identified as the producer of the immunosuppressant cyclosporin A, but it is now known that cyclosporin A is produced by To. inflatum (Horsburgh et al 1980, Thali 1995). Iida et al (1999) isolated a peptidic immunosuppressant from T. polysporum. Komatsu (1976) found that H. pachybasioides/T. polysporum is common on and inside of bed logs used in cultivation of shiitake mushrooms in Japan and was strongly antagonistic to the mycelium, severely diminishing mycelial growth and mushroom production. Nelson (1982) found T. polysporum to be a normal member of the soil biota of Douglas fir (Pseudotsuga menziesii) and he and his collaborators isolated it from roots and stumps of Douglas fir that were infected with Phellinus weirii (Nelson et al 1987, 1995; Goldfarb et al 1989). Trichoderma polysporum has shown some promise in biological control of Pythium aphanidermatum ( Jackisch-Matsuura and Menezes 1999) and Phytophthora cinnamomi (Kelley 1977). Barbosa et al (2001) found that T. polysporum was highly antagonistic to Cladosporium cladosporioides on passion fruit. A combination of arbuscular mycorrhizal fungi, Rhizobium and T. polysporum slowed fusarium-induced seed rot and damping-off in Dalbergia sissoo while promoting tree growth (Singh et al 2002).
2. Hypocrea pilulifera J. Webster &
