endomelanconiopsis, a new anamorph genus in the botryosphaeriaceae
TRANSCRIPT
Endomelanconiopsis, a new anamorph genus in the Botryosphaeriaceae
Enith I. RojasEdward Allen Herre
Smithsonian Tropical Research Institute, ApartadoPostal 0843-03092, Balboa, Ancon, Republic ofPanama
Luis C. MejıaDepartment of Plant Biology and Pathology, RutgersUniversity, 59 Dudley Road, Foran Hall, NewBrunswick, New Jersey 08901
A. Elizabeth ArnoldDivision of Plant Pathology and Microbiology,Department of Plant Sciences, University of Arizona,Tucson, Arizona 85721
Priscila Chaverri1
Department of Biology, Howard University, 415 CollegeStreet NW, Washington D.C. 20059
Gary J. Samuels2
United States Department of Agriculture, AgricultureResearch Service, Systematic Mycology & MicrobiologyLaboratory, B-011A, Room 304, 10300 Baltimore Ave.,Beltsville, Maryland 20705
Abstract: A new lineage is discovered within theBotryosphaeriaceae (Ascomycetes, Dothideomycetes,incertae sedis). Consistent with current practice ofproviding generic names for independent lineages,this lineage is described as Endomelanconiopsis gen.nov., with the anamorphic species E. endophytica sp.nov. and E. microspora comb. nov. (5 Endomelanco-nium microsporum). Endomelanconiopsis is character-ized by eustromatic conidiomata and holoblasticallyproduced, brown, nonapiculate, unicellular conidia,each with a longitudinal germ slit. Phylogeneticanalysis of partial sequences of LSU, ITS andtranslation elongation factor 1 alpha (tef1) indicatethat E. endophytica is sister of E. microspora and thatthey are nested within the Botryosphaeriaceae.However because there is no support for the‘‘backbone’’ of the Botryosphaeriacae we are notable to see the interrelationships among the manygenera in the family. Neither species is known to havea teleomorph. Endomelanconiopsis differs from En-domelanconium because conidia of the type species ofEndomelanconium, E. pini, are papillate at the base,
conidiogenous cells proliferate sympodially and thepycnidial wall is thinner; we postulate that theteleomorph of E. pini as yet unknown is aninoperculate discomycete. Endomelanconiopsis endo-phytica was isolated as an endophyte from healthyleaves of Theobroma cacao (cacao, Malvaceae) andHeisteria concinna (Erythroplaceae) in Panama. En-domelanconiopsis microspora was isolated from soil inEurope.
Key words: Austrocenangium, Endomelanconium,endophytic fungi, Heisteria concinna, systematics,Theobroma cacao
INTRODUCTION
Fungal endophytes live inside plant tissues for all orpart of their lives without causing apparent diseasesymptoms (Wilson 1995). Endophytic fungi havebeen found within host tissues at high density anddiversity in all tropical and temperate ecosystemswhere endophytes have been sought (e.g. Arnold andLutzoni 2007, Clay and Holah 1999, Hawksworth2001). Foliar fungal endophytes have been found tobe abundant and diverse in intact tropical forests withhigh plant diversity and low density of individualspecies (Dreyfuss and Petrini 1984, Lodge et al 1996,Rodrigues and Petrini 1997, Arnold et al 2000, vanBael et al 2005), and in mangroves and Theobromacacao agroforestry systems (cacao, Malvaceae, Mal-vales) with low plant diversity and high density ofindividual plant species (Gilbert et al 2002, Suryanar-ayanan et al 1998, Arnold et al 2003).
In most tropical tree species leaves are flushed in alargely endophyte-free (E2) condition (Arnold et al2003, Herre et al 2007). A taxonomically diverseassemblage of airborne spores land on leaf surfaces,some of which initiate endophytic associations. Afterthe wetting of spore-laden leaf surfaces some sporesgerminate and penetrate directly through the cuticleinto the leaf tissue, where hyphae grow between cells(Herre et al 2005, 2007). With time the initiallyendophyte-free leaf tissues become saturated withendophytes, with ca. 100% of sampled leaf fragments(2 3 2 mm) containing cultivable fungi (Arnold et al2003). Data from cohorts of leaves suggest that overtime the species composition of fungal endophyteswithin a leaf changes and that endophyte diversity(number of species per isolate recovered) usuallydeclines as leaves pass from maturity toward senes-cence. Within the leaf the distribution of fungalspecies resembles a quilt-like patchwork: different
1 Current address: AGNR-Plant Science & Landscape Architecture,2112 Plant Sciences Building, University of Maryland, College Park,MD 20742-44522 Corresponding author. E-mail: [email protected]
Mycologia, 100(5), 2008, pp. 760–775. DOI: 10.3852/07-207# 2008 by The Mycological Society of America, Lawrence, KS 66044-8897
760
species usually abutting, producing an extremelyheterogeneous mix of fungal species and genotypesat fine scales (Brayford 1990a, b, Lodge et al 1996,Herre et al 2007). Colonization by endophytic fungihas been shown to reduce pathogen susceptibility inthe host plant and affect host photosynthetic andhydraulic physiology in various ways, but general rulesfor tropical plant-endophyte interactions are not yetestablished (Arnold et al 2003, Herre et al 2005, 2007,Arnold and Engelbrecht 2007, Mejıa et al 2003,2008).
Arnold et al (2003) reported 344 endophyticmorphospecies (determined by colony appearancein agar culture) from 126 naturally infected leaves ofT. cacao collected during 2 y from five sites inPanama. Common among these fungal morphospe-cies was one that produced a dark gray or green, tonearly black, colony on potato-dextrose agar. Largeeustromatic pycnidial structures developed in agarcultures; these contained dark brown, holoblasticallyproduced conidia with a longitudinal germ slit.Several representatives of this morphospecies couldbe identified as an Endomelanconium (Corda) Petrakas characterized by Sutton (1980). What appeared tobe the same species was isolated also as an endophytefrom leaves of Heisteria concinna (Erythroplaceae,Santalales), a tree that often grows in association withcacao in lowland moist forests in Panama. Thepublished description of E. microsporum Verkley &van der Aa (Verkley and van der Aa 1997), a speciesisolated from soil in Europe, indicated that it was thesame or very closely related to the endophyte. Thecultures and anamorph morphology of these speciesare strongly suggestive of Botryosphaeria and itsrelatives (Luque et al 2005, Crous et al 2006, Phillips2008). At least some described Endomelanconiumspecies are anamorphs of inoperculate discomycetes,although the teleomorph of the type species, E. pini(Corda) Petrak (Petrak 1940), is not known. Possibleteleomorph relationships and subtle morphologicaldifferences among these various described Endome-lanconium species lead us to suspect that the genus isparaphyletic.
In the present study we examine the phylogeneticposition of the endomelanconium-like endophytes bymeans of molecular sequence data from three loci.We compare them to described species of Endome-lanconium.
MATERIALS AND METHODS
Collection and cultures.—The endophytic cultures reportedhere were obtained from two surveys of endophytic fungi inthe Republic of Panama. The first was carried out in 1999and 2000 in five lowland sites with mixed forest cover across
the Isthmus of Panama, including Barro Colorado Island(BCI, Canal Zone, 09u99350N, 79u509330W) and Nombre deDios (ND, Colon, 09u349N, 79u289W). The mean annualrainfall in ND is 4041 mm, and the average annualtemperature is approximately 26 C (Tosi 1971); at BCIaverage annual temperature is 27 C and rainfall is approx-imately 2600 mm (Leigh et al 1996). In the first surveyleaves of three different ages were sampled from adultTheobroma cacao trees during two seasons (dry and rainy) in1999 and 2000. In the second year of the first survey samplesalso were taken from two co-occurring species, Heisteriaconcinna and Ouratea lucens (Ochnaceae, Malpighiales)(Arnold et al 2003).
The second survey was carried during the rainy season(Jul 2000) at ND and samples were taken only from a singletree of Theobroma cacao. In this survey leaves of five ageclasses (young, 1 wk old leaves; young-mature, 2 wk oldleaves; mature, 4 wk old leaves; mature–old, 8 wk old leaves;and old, . 12 wk old leaves) were sampled to addressquestions regarding colonization patterns. The leaves werecollected at the same time and were processed the day thatthey were collected.
The endomelanconium-like isolates were found duringthe rainy season in asymptomatic foliage of T. cacao and H.concinna at BCI in our first survey; it was the most commonfungus in healthy leaves of cacao for the second survey. Atotal of 15 isolates of endomelanconium-like endophyteswere selected for DNA sequence analyses.
Characterization of colonies and morphology.—After initialisolation on 2% malt extract agar (MEA, Difco, Detroit,Michigan), isolates were cultured on potato-dextrose agar(PDA; BBL, Sparks, Maryland) and simple nutrient agar(SNA, Nirenberg 1976) at 25 C with a 12 h cool whitefluorescent light/12 h dark photoperiod. In addition apiece of sterile filter paper was put on the surface of thesolidified SNA to induce sporulation. Pycnidia typicallyformed within 3 wk on both media.
Growth rates were determined in darkness (with fluores-cent light briefly introduced when the colonies weremeasured) on PDA and SNA (without filter paper) at fivetemperatures (15, 20, 25, 30 and 37 C). Inoculum forgrowth rate studies was taken from young, actively growingcolonies on cornmeal-dextrose agar (Difco cornmeal agar +2% w/v dextrose). A 5 mm diam plug was cut from the edgeof the colony and placed in the middle of the test medium(PDA or SNA). Measurements of colony radius from theedge of the inoculum plug to the edge of the colony weretaken at 24 h intervals. The growth trial was repeated twiceon succeeding weeks and the two measurements wereaveraged.
Morphological descriptions were made from isolatessporulating on PDA or SNA (with filter paper) at 25 C forca. 10 d with photoperiod 12 h darkness/12 h cool whitefluorescent light. Thirty conidia and up to 30 conidiogen-ous cells were measured for each isolate. Sections ofpycnidia for microscopic examination were prepared witha freezing microtome (IEC-CTF Microtome-Cryostat, Inter-national Equipment Co., Needham Heights, Massachu-setts).
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 761
Images were captured with a Nikon DXM1200 digitalcamera and Nikon ACT 1 software or with a Nikon Ds-Fi1camera and NIS Elements Basic Research 2.30, SP4(Nikon). Some composite images were made with HeliconFocus version 4.21.5 Pro (Helicon Soft, www.heliconfocus.com). Measurements were made with Scion Image (releaseBeta 4.0.2; Scioncorp, Frederick, Maryland). Data wereanalyzed statistically with Systat 10.0 (SPSS, Chicago,Illinois). Representative cultures of the endophytic endo-melanconium-like species were deposited in the Centraal-bureau voor Schimmelcultures (CBS), Utrecht, the Nether-lands.
DNA extraction, PCR amplification and sequence assembly.—Mycelium for DNA extraction was prepared in potato-dextrose broth (Difco). DNA was extracted following themanufacturer’s protocol for fresh plant tissue with thePureGene genomic DNA isolation kit (Gentra Systems,Minneapolis, Minnesota).
For each isolate 25 mL PCR reactions were employed toamplify sections of the nuclear ribosomal large subunit(LSU), internal transcribed spacers and 5.8 s gene (ITS),and translation elongation factor 1 alpha (tef1). PCRmixtures contained 2.5 mL 103 PCR buffer (New EnglandBiolabs, Ipswich, Massachusetts), 200 mM dNTPs, 25 pmoleeach primer, 1.25 units Taq DNA Polymerase and 10–50 ngtemplate DNA. Primers LROR and LR7 were used toamplify ca. 1100 base pairs of LSU, followed by sequencingof ca. 600 bp with LROR and LR3 (Rehner and Samuels1994). Primers ITS5 and ITS4 (White et al 1990) were usedto amplify the approximately 540 bp ITS, followed bysequencing with the same primers. Primers EF1-728F(Carbone and Kohn 1999) and tef1 rev (Samuels et al2002) were used to amplify and sequence ca. 600 bp of tef1.PCR was carried out on a PT-200 PCR thermal cycler (MJResearch, Waltham, Massachusetts) according to theseconditions: 95 C for 5 min; 35 cycles denaturation at 95 Cfor 30 s, annealing at 51 C for 30 s, extension at 72 C for1 min; and a final extension at 72 C for 10 min. Theconcentration of PCR products in ng/mL was determinedon 1% agarose gel via electrophoresis with Lambda Hind IIIDNA as a marker. PCR products were purified with ExoSAP-ITH reagent following the manufacturer’s instructions (USBCorp., Cleveland, Ohio). Bidirectional sequences wereobtained with BigDye Terminator cycle sequencing kit(Applied Biosystems, Foster City, California). Products wereanalyzed on a 3100 DNA sequencer (Applied Biosystems).Eight representative cultures were sequenced for LSU and15 cultures were sequenced for both ITS and tef1.
Phylogenetic analyses.—Sequences were edited and assem-bled with Sequencher 4.1 (Gene Codes, Wisconsin) andClustal X version 1.8 (Thompson et al 1997) was used toalign sequences, followed by manual adjustment in Mac-Clade 4 (Maddison and Maddison 2000). New DNAsequences were deposited in GenBank (TABLE I).
Two datasets were prepared. One alignment includedLSU sequences from representatives of the major lineageswithin the Botryosphaeriaceae (Crous et al 2006), twospecies representing the endomelanconium-like group, andspecies of Guignardia as outgroup, for a total of 56
terminals. The second dataset included a concatenatedalignment of ITS and tef1 sequences for 15 endomelanco-nium-like isolates from T. cacao and H. concinna,Endomelanconium microsporum (ex type culture, CBS353.97), representative species of the Botryosphaeriaceae,and two species of Guignardia, identified as closely relatedto Botryosphaeria (Schoch et al 2006) as outgroup taxa (totalof 50 terminals).
The LSU dataset was analyzed with PAUP* v. 4.0b10(Swofford 2002) using the neighbor joining (NJ) methodwith Kimura 2-parameter distance calculation and maxi-mum parsimony (MP) using the heuristic search optionwith simple taxa additions and tree bisection and recon-nection (TBR) as the branch-swapping algorithm. Robust-ness of the resulting topology was assessed with 1000bootstrap replicates in both neighbor joining and parsimo-ny analyses.
The ITS and tef1 datasets were analyzed separately and inthe absence of phylogenetic conflict combined for analysiswith NJ and MP as described above and with Bayesianmethods (Rannala and Yang 2005) as implemented with theprogram MrBayes v3.1.2 (Huelsenbeck and Ronquist 2001).MrModeltest 2.2 (Nylander 2004) was used to select the bestmodel, taking into account both the hierarchical likelihoodratio test (LRT) and Akaike information criteria (AIC).Preference was given to the latter (AIC). Bayesian analysiswas performed with models specific to each partition of thedata (GTR + I + G for ITS, and GTR + G for tef). TheBayesian MCMC (Markov chain Monte Carlo) analysis wasrun with three hot chains and one cold chain for 1 500 000generations, sampling every 500 generations. The first 400trees were discarded as burn-in on the basis of likelihoodscores, and PAUP* was used to construct a 50% majorityrule consensus tree from the remaining 5600 trees.
RESULTS
The endomelanconium-like endophyte was foundonly during the rainy season. At Nombre de Dios itwas not found in the first survey, but it was isolatedfrom Theobroma cacao and Heisteria concinna at BCI.It was the most abundant species isolated from thesingle tree (Theobroma cacao) that was the subject ofthe second survey. Of 16 cultures identified as beingthe morphospecies that included the endomelanco-nium-like fungus, nine were confirmed by microscopyto be endomelanconium-like. The remaining sevencultures either had died before they could be fullyidentified or were a different fungus. The endome-lanconium-like fungus was isolated from mature andold leaves; it never was found in leaves younger than1 wk.
Phylogenetic analysis.—BLAST analyses of the NCBIGenBank database with LSU sequences of theendomelanconium-like cultures showed that theclosest matches (maximum identity 97% and se-quence coverage of 100%) were with members of
762 MYCOLOGIA
the Botryosphaeriaceae. We aligned our sequences ofthe endomelanconium-like endophyte with the ex-type culture of E. microsporum CBS 353.97 and withone representative taxon selected from each clade inthe Botryosphaeriaceae tree of Crous et al (2006,TreeBase M2707); we added to this tree Aplosporellaprunicola (Damm et al 2007). Because no variationwas observed among the eight endophytic isolatessequenced for the LSU locus, only two were selectedfor the analysis. The two sequences of the endophytein the tree (FIG. 1) are from cultures that representthe phenotype-groups described below. The LSUdataset had 933 characters, 102 of which wereparsimony informative, 47 were parsimony uninfor-mative and 784 were constant when gaps were treatedas missing characters. A heuristic search found 100equally parsimonious trees (TL 5 339 steps; CI 5
0.522; RI 5 0.787; RC 5 0.411). Neighbor joininganalysis with Kimura 2-parameter distance calculationon the sequence data yielded trees with similartopology (see FIG. 1). The two isolates of theendomelanconium-like endophyte clustered with E.microsporum in a highly supported clade (FIG. 1),indicating that they are congeneric. Although therewas high BS support (FIG. 1) for Botryosphaeria s. lat.,(including Pseudofusicoccum), only internal cladesreceived significant support, and there was nosupport for a backbone. Thus it was impossible todetermine relationships among the clades.
The manually adjusted alignment of 15 isolates ofthe endomelanconium-like species with species of theBotryosphaeriaceae for the concatenated ITS and tef1datasets contained 34 ambiguously aligned charactersin the ITS region (25–42 bp, 54–69 bp insertionfound only in G. mangiferae and G. bidwellii) thatwere excluded, leaving 892 characters. Of these 385were parsimony informative, 46 were parsimonyuninformative and 461 were constant when gaps weretreated as missing characters. A heuristic search foundsix equally parsimonious trees (TL 5 967 steps; CI 5
0.698; RI 5 0.908; RC 5 0.634) and same topology wasfound with Kimura 2-parameters distance calculationsin neighbor joining and Bayesian analyses.
Together these analyses placed the endomelanco-nium-like endophyte isolates and E. microsporum inthe Botryosphaeriaceae (FIG. 2). The 15 endophyticcultures form a highly supported (100% bootstrap,1.0 posterior probability) clade. There was nodistinction between endophytic isolates from T. cacaoand those from H. concinna, indicating that theendomelanconium-like endophytes collected fromboth hosts likely are one species. The endophytesand the ex-type culture of E. microsporum formed astrongly supported clade but E. microsporum differedmarkedly (FIG. 2). The closest morphological com-
parison within the Botryosphaeriaceae is with Sphaer-opsis subglobosa, the anamorph of Neodeightoniasubglobosa Booth in Punithalingham (1969)and which Crous et al (2006) referred to as‘Botryosphaeria’ subglobosa (FIG. 1). It is not closelyrelated to E. microsporum or the endomelanconium-like endophyte (FIG. 1).
Although ITS sequences for all the endophyticcultures were identical, there was slight variation inthe tef1 gene. Two groups were formed with . 0.95Bayesian posterior probability and high NJ and MPbootstrap support, and one isolate (culture 3370 fromT. cacao, BCI, 1999) formed a third group thatdiffered from the other two groups by two bp in thesame locus. These three groups correlated with thesame three groups that were observed based on themorphological characters of their colonies (FIGS. 3–5,see below). There was no obvious relationshipbetween observed groups and their occurrence inparticular sites or host species.
Morphological analyses.—The endomelanonium-likeendophyte was not observed sporulating in nature.Pycnidia (FIGS. 6–9) formed in culture were eustro-matic (Sutton 1980), tending to be cylindrical, largeand the wall was formed of pseudoparenchymatouscells (FIGS. 6–9). The locule was convoluted (FIGS. 8,9) and was lined entirely with conidiogenous cells(FIGS. 8–10). The conidiogenous cells arose directlyfrom the cells of the pycnidial wall (FIG. 11) andpossessed a single, terminal, holoblastic conidiogen-ous locus (FIGS. 11, 12). Macroconidia (FIGS. 13–15)were opaque, brown, unicellular and bilaterallysymmetrical, elliptical in face view and flattened witha narrow edge; a germ slit was positioned on theflattened edge. Often conidia were flattened at thebase at the point of abscission from the conidiogen-ous cell. In addition to these macroconidia, microco-nidia (spermatia?, FIG. 18) formed either in the samepycnidium as the macroconidia or in separatepycnidia (FIG. 16) on SNA and PDA. Microconidiawere variable in shape, ellipsoidal to subcylindrical,hyaline, and arose from phialidic conidiogenous cells(FIG. 17). Two size classes were observed: those withmicroconidia 2.0–4.5 3 1–2 mm (Mean 5 2.9 3
1.3 mm) and those with microconidia 3–10 3 1.0–2.7 mm (Mean 5 6.9 3 1.7 mm).
Two size classes of macroconidia were observed(FIG. 19, TABLE II), designated respectively as Groups1 and 2. Macroconidia of Group 1 were shorter (5.7 6
0.7 3 3.8 6 0.3 mm; 95% confidence interval (CI) 5.6–5.9 3 3.7–3.8 mm), macroconidia of Group 2 werelonger (7.2 6 0.9 3 4.4 6 0.45 mm; 95% CI 7.1–7.3 3
4.3–4.4 mm), the macroconidia of a single culture(3370), referred to below as Group 3 because of its
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 763
TA
BL
EI.
Cu
ltu
res
use
din
ph
ylo
gen
etic
anal
ysis
,th
eir
ori
gin
and
Gen
Ban
kn
um
ber
s
Spec
ies
Cu
ltu
res1
Co
llec
tor
Geo
grap
hy2
Sub
stra
tum
Gen
Ban
kac
cess
ion
nu
mb
er3
ITS1
+2
tef1
LSU
En
dom
elan
con
iopsi
sm
icro
spor
aC
BS
35
3.9
7H
.A.
van
der
Aa
Pap
ua
New
Gu
inea
soil
EU
6836
55E
U68
3636
EU
6836
28
E.
endop
hyt
ica
CB
S1
20
39
7=
74
63
E.
Ro
jas,
L.
Mej
ıa&
Z.
May
nar
dP
anam
a(N
D)
Theo
brom
aca
cao,
leaf
EU
6836
56E
U68
3637
EU
6836
29E
.en
dop
hyt
ica
CB
S12
2550
573
94E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(ND
)T
.ca
cao,
leaf
EU
6836
64E
U68
3645
EU
6836
34E
.en
dop
hyt
ica
CB
S12
2549
533
70E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BC
I)T
.ca
cao,
leaf
EU
6836
59E
U68
3640
—E
.en
dop
hyt
ica
7473
E.
Ro
jas,
L.
Mej
ıa&
Z.
May
nar
dP
anam
a(N
D)
T.
caca
o,le
afE
U68
3660
EU
6836
41E
U68
3630
E.
endop
hyt
ica
CB
S12
2548
577
59E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BC
I)T
.ca
cao,
leaf
EU
6836
57E
U68
3638
—E
.en
dop
hyt
ica
7329
E.
Ro
jas,
L.
Mej
ıa&
Z.
May
nar
dP
anam
a(N
D)
T.
caca
o,le
afE
U68
3662
EU
6836
43E
U68
3632
E.
endop
hyt
ica
CB
S12
2547
573
70E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(ND
)T
.ca
cao,
leaf
EU
6836
66E
U68
3647
EU
6836
35E
.en
dop
hyt
ica
CB
S12
2546
573
31E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(ND
)T
.ca
cao,
leaf
EU
6836
61E
U68
3642
EU
6836
31E
.en
dop
hyt
ica
CB
S12
2545
573
26E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(ND
)T
.ca
cao,
leaf
EU
6836
63E
U68
3644
EU
6836
33E
.en
dop
hyt
ica
CB
S12
2544
581
45E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BC
I)T
.ca
cao,
leaf
EU
6836
58E
U68
3639
—E
.en
dop
hyt
ica
CB
S12
2551
577
52E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BC
I)T
.ca
cao,
leaf
EU
6836
65E
U68
3646
—E
.en
dop
hyt
ica
CB
S12
2543
590
90E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BC
I)H
eist
eria
con
cin
na
,le
afE
U68
3670
EU
6836
51—
E.
endop
hyt
ica
CB
S12
2542
591
85E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BC
I)H
.co
nci
nn
a,
leaf
EU
6836
69E
U68
3650
—E
.en
dop
hyt
ica
9188
E.
Ro
jas,
L.
Mej
ıa&
Z.
May
nar
dP
anam
a(B
CI)
H.
con
cin
na
,le
afE
U68
3668
EU
6836
49—
E.
endop
hyt
ica
CB
S12
2552
591
81E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BC
I)H
.co
nci
nn
a,
leaf
EU
6836
67E
U68
3648
—B
otry
osphaer
iari
bis
CB
S12
2553
5P
I006
(E.A
.H
erre
)E
.R
oja
s,L
.M
ejıa
&Z
.M
ayn
ard
Pan
ama
(BT
)T
.ca
cao,
leaf
EU
6836
73E
U68
3654
—
B.
ribi
sC
BS
1154
75G
.H
ud
ler
USA
,N
ewYo
rkR
ibes
sp.
AY2
3693
5A
Y236
877
—B
.ri
bis
C.M
.W.
7773
G.
Hu
dle
rU
SA,
New
York
Rib
essp
.A
Y2
36
93
6A
Y2
36
87
8—
B.
dot
hid
eaC
.M.W
.80
00B
.Sl
ipp
ers
Swit
zerl
and
,C
roci
fiss
oP
run
us
sp.
AY
23
69
49
AY
23
68
98
—B
.dot
hid
eaC
BS
1103
02A
.J.L
.P
hil
lip
sP
ort
uga
l,M
on
tem
or-
o-
No
voV
itis
vin
ifer
aA
Y2
59
09
2A
Y5
73
21
8—
B.
lute
aC
.M.W
.90
76S.
R.
Pen
nyc
oo
kN
ewZ
eala
nd
Mal
us
3d
om
esti
caA
Y2
36
94
6A
Y2
36
89
3—
B.
lute
aC
BS
11
02
99
A.J
.L.
Ph
illi
ps
Po
rtu
gal,
Oei
ras
V.
vin
ifer
aA
Y2
59
09
1A
Y5
73
21
7A
Y928
043
B.
au
stra
lis
CB
S11
2872
F.
Hal
leen
Sou
thA
fric
aV
.vin
ifer
aA
Y3
43
38
8A
Y3
43
34
7—
B.
au
stra
lis
CB
S11
2877
F.
Hal
leen
Sou
thA
fric
aV
.vin
ifer
aA
Y3
43
38
5A
Y3
43
34
6—
Fu
sico
ccu
mvit
icla
vatu
mC
BS
11
29
77
F.
Hal
leen
Sou
thA
fric
aV
.vin
ifer
aA
Y3
43
38
1A
Y3
43
34
1—
F.
vit
ifu
sifo
rme
CB
S1
10
88
0J.
M.
van
Nie
kerk
Sou
thA
fric
aV
.vin
ifer
aA
Y3
43
38
2A
Y3
43
34
4—
B.
parv
aC
.M.W
.9
08
1G
.J.
Sam
uel
sN
ewZ
eala
nd
,T
eP
uke
Pop
ulu
sn
igra
AY
23
69
43
AY2
3688
8—
B.
parv
aC
BS
1103
01A
.J.L
.P
hil
lip
sP
ort
uga
l,P
alm
ela
V.
vin
ifer
aA
Y2
59
09
8A
Y5
73
22
1A
Y928
046
B.
stev
ensi
iC
BS
431.
82H
.A.
van
der
Aa
Net
her
lan
ds,
Maa
rsev
een
Fra
xin
us
exce
lsio
rA
Y236
955
AY
23
69
04
DQ
3778
63
B.
stev
ensi
iC
BS
1125
53A
.J.L
.P
hil
lip
sP
ort
uga
l,M
on
tem
or-
o-
No
voV
.vi
nif
era
AY
25
90
93
AY
57
32
19
—
764 MYCOLOGIA
Spec
ies
Cu
ltu
res1
Co
llec
tor
Geo
grap
hy2
Sub
stra
tum
Gen
Ban
kac
cess
ion
nu
mb
er3
ITS1
+2
tef1
LSU
B.
obtu
saC
BS
1125
55A
.J.L
.P
hil
lip
sP
ort
uga
l,M
on
tem
or-
o-
No
voV
.vin
ifer
aA
Y2
59
09
4A
Y5
73
22
0A
Y928
050
B.
obtu
saC
.M.W
.77
75B
.Sl
ipp
ers
&G
.H
ud
ler
USA
,N
ewYo
rkR
ibes
sp.
AY
23
69
54
AY
23
69
03
—B
.co
rtic
ola
CB
S1
12
54
9A
.A
lves
Po
rtu
gal,
Ave
iro
Qu
ercu
ssu
ber
AY
25
91
00
AY
57
32
27
—B
.co
rtic
ola
CB
S11
2545
M.E
.Sa
nch
ez&
A.
Tra
per
oSp
ain
,C
adiz
Q.
sube
rA
Y2
59
08
9A
Y5
73
22
6—
B.
rhod
ina
C.M
.W.
1013
0J.
Ro
ux
Uga
nd
aV
itex
don
nia
na
AY
23
69
51
AY
23
69
00
—B
.rh
odin
aC
.M.W
.90
74T
.B
urg
ess
Mex
ico
Pin
us
sp.
AY
23
69
52
AY
23
69
01
—D
iplo
dia
por
osu
mC
BS
11
04
96
J.M
.va
nN
ieke
rkSo
uth
Afr
ica,
Stel
len
bo
sch
V.
vin
ifer
aA
Y3
43
37
9A
Y3
43
34
0D
Q37
7894
D.
por
osu
mC
BS
1105
74J.
M.
van
Nie
kerk
Sou
thA
fric
aV
.vi
nif
era
AY
34
33
78
AY
34
33
39
—B
.vit
icol
aC
BS
11
70
09
J.L
uq
ue
&S.
Mar
tos
Spai
n,
Vim
bo
dı
V.
vin
ifer
acv
.G
arn
atxa
Neg
raA
Y9
05
55
4A
Y9
05
55
9D
Q37
7873
B.
vit
icol
aC
BS
1170
08J.
Lu
qu
e&
J.R
eyes
Spai
n,
San
tSa
du
rnı
d’A
no
iaV
.vin
ifer
acv
.X
arel
-loA
Y9
05
55
7A
Y9
05
56
0—
B.
vit
icol
aC
BS
1170
06J.
Lu
qu
e&
R.
Mat
euSp
ain
,V
imb
od
ıV
.vin
ifer
acv
.G
arn
atxa
Neg
raA
Y9
05
55
5A
Y9
05
56
2—
Dip
lodia
sp.
CB
S11
2870
J.M
.va
nN
ieke
rkSo
uth
Afr
ica,
Wes
tern
Cap
eV
.vi
nif
era
AY
34
33
76
AY
34
33
37
—
Dip
lodia
sp.
CB
S11
2869
J.M
.va
nN
ieke
rkSo
uth
Afr
ica,
Wes
tern
Cap
eV
.vi
nif
era
AY
34
33
73
AY
34
33
36
—
B.
sarm
ento
rum
IMI
63
58
1b
E.A
.E
llis
En
glan
d,
War
wic
ksh
ire
Ulm
us
sp.
AY
57
32
12
AY
57
32
35
AY9
2805
2B
.sa
rmen
toru
mC
BS
1150
38A
.J.L
.P
hil
lip
sN
eth
erla
nd
s,D
elft
Malu
spu
mil
aA
Y5
73
20
6A
Y5
73
22
3—
B.
iber
ica
CB
S1
15
04
1J.
Lu
qu
eSp
ain
,A
rago
nQ
.il
exA
Y5
73
20
2A
Y5
73
22
2D
Q37
7853
B.
iber
ica
CB
S11
5040
J.L
uq
ue
Spai
n,
Cat
alo
nia
Q.
ilex
AY
57
32
14
AY
57
32
32
—G
uig
nard
iam
an
gife
rae
1095
E.
Ro
jas,
L.
Mej
ıa&
Z.
May
nar
dP
anam
a(D
I)T
.ca
cao,
leaf
EU
6836
71E
U68
3652
—G
.bid
wel
lii
CB
S11
1645
G.
Car
roll
USA
,M
isso
uri
Part
hen
ocis
sus
quin
quef
olia
EU
6836
72E
U68
3653
DQ
3778
76
1A
bb
revi
atio
ns
of
cult
ure
coll
ecti
on
s:C
BS
5C
entr
aalb
ure
auvo
or
Sch
imm
elcu
ltu
res,
Utr
ech
t,th
eN
eth
erla
nd
s;IM
I5
CA
BI
Gen
etic
Res
ou
rce
Co
llec
tio
n,
Egh
am,
UK
;C
.M.W
.5
C.M
.W
ingf
ield
.A
cces
sio
nn
um
ber
sin
bo
ldfa
cesi
gnif
yex
-typ
ecu
ltu
res.
2G
eogr
aph
icre
gio
ns
inP
anam
aab
bre
viat
edas
ND
5N
om
bre
de
Dio
s,B
CI
5B
arro
Co
lora
do
Isla
nd
,D
I5
Div
isa,
BT
5B
oca
sd
elT
oro
.3Se
qu
ence
nu
mb
ers
init
alic
wer
ere
trie
ved
fro
mG
enB
ank.
All
oth
ers
wer
eo
bta
ined
inth
isst
ud
y.
TA
BL
EI.
Co
nti
nu
ed
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 765
differing growth rate, were not significantly differentin size from others (6.7 6 0.3 3 4.1 6 0.2 mm; 95% CI6.6–6.8 3 4.1–4.2). Macroconidia of E. microsporumwere of intermediate length but wider (FIG. 19; 6.1 6
0.4 3 4.6 6 0.4 mm; 95% CI 6.0–6.3 3 4.4–4.7 mm)than those of either of the endophyte size groups(FIG. 19). No differences were observed in the size ofthe conidiogenous cells of the endophyte groups.
FIG. 1. Parsimony tree based on LSU sequence. The tree was rooted to Guignardia bidwellii, Guignardia sp, and Phyllostictasp. This tree is modified from Crous et al (2006), to which the reader should refer for full discussion of the clades; numbers onthe right side refer to clades in Crous et al (2006). Numbers in the nodes refer to bootstrap support values from 1000 replicatesof NJ and MP respectively.
766 MYCOLOGIA
FIG. 2. Bayesian inference tree (majority rule consensus of 5600 trees) based on the combined ITS and tef1 sequences.Fifteen Endomelanconiopsis isolates from T. cacao and H. concinna were included. Tree was rooted with Guignardia mangiferaeand G. bidwellii. Black lines are $ 0.95 posterior probability coefficients and $ 70% NJ and MP bootstrap support; (*)represents , 0.95 posterior probability coefficients and $ 70% NJ and MP bootstrap support; (**) represents , 0.95 posteriorprobability coefficients and , 70% NJ and MP bootstrap support. Group numbers refer to phenotype-defined groups inEndomelanconiopsis endophytica. Accession numbers in boldface signify ex-type cultures.
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 767
Colonies on PDA become dark gray to black andthe quantity of aerial mycelium was variable. Opti-mum temperature for growth on PDA and SNA was30–35 C. Three growth curves were observed after 5 don PDA (FIG. 20). Two of these curves correspondedto the macroconidium-size groups noted above:Group 1 (n 5 5) colony radius at 30 C the same asthe colony radius at 37 C (ca. 50 mm); Group 2 (n 5
9) colony radius at 30 C (ca. 50 mm) that was lessthan the radius at 37 C (55–60 C); Group 3 (n 5 1)colony radius at 30 C (ca. 45 mm) greater than theradius at 37 C (ca. 35 mm). All endophytic isolatesgrew well at 37 C, although isolate 3370 (Group 3)was significantly slower at 30 and 37 C than were theother cultures. Pycnidia usually formed within 1 wk at25 C under alternating light and dark.
In the phylogram (FIG. 2) Group 2 forms a wellsupported clade that is sister of Group1/Group 3.Thus Group 1 isolates had shorter conidia andrelatively slower growth at 37 C while Group 2 isolateshad longer conidia and faster growth at 37 C. Group3, which comprised a single culture, had conidia ofintermediate size and a significantly slower rate ofgrowth than Groups 1 or 2. Despite the phenotypicdifferences, the phylogenetic distance between thesesister groups is small (only five bp difference in tef).We consider the phenotypic differences to representintraspecific variation, but additional collectionsmight support taxonomic subdivision.
Conidia of the endophytic endomelanconium-likespecies are most similar in size to those of E.
microsporum and smaller than the reported measure-ments of all other Endomelanconium species (TA-
BLE II). As noted above conidia of E. microsporumare wider (4.4–4.7 mm) than the endophytes. Afurther difference between the endophytes and E.microsporum is the absence of chlamydospores in theformer (see Verkley and van der Aa 1997). Microco-nidia have not been reported for any other Endome-lanconium species, but the describing authors couldhave overlooked them. We did not notice them in theex-type culture of E. microsporum.
These differences, combined with the respectivedifferences in habitat (leaf endophyte vs. soil) andgeography (Central America vs. southwestern Pacific)and our phylogenetic results, distinguish the endo-phytes from E. microsporum at the species level.
Comparison with Endomelanconium species.—Onlyfive species have been included in Endomelanconium,including E. microsporum discussed above and theunnamed anamorph of Austrocenangium australe(Speg.) Gamundı (TABLE II). The type species ofEndomelanconium, E. pini (Corda) Petrak, has beencollected from dead trunks and branches of Abies fromSwitzerland, Austria, Germany, the Czech Republicand Slovakia but apparently has not been grown inculture. As a result we did not include it in ourphylogenetic analysis. No teleomorph is known for it.
There are small but significant differences betweenE. pini and the two endomelanconium-like speciesnoted above. We studied one specimen of E. pini
FIGS. 3–5. Endomelanconiopsis endophytica, three groups on PDA after10, 25 C, 12 h cool white fluorescent light/12 h dark.3 Group 1 (7394). 4. Group 2 (7463). 5. Group 3 (3370).
R
FIGS. 6–18. Endomelanconiopsis endophytica/Botryosphaeria anamorph. 6–9. Pycnidia. 6. From SNA. 7. From PDA. 8, 9.Pycnidia showing convoluted hymenium in the locule. 10–12. Conidiogenous cells. 10, 11. Conidiogenous cell arising frompycnidial wall cells. 11, 12. Holoblastic conidiogenous locus. 13–15. Macroconidia, opaque brown, unicellular, often flattened
768 MYCOLOGIA
at the base. 14, 15. Germ slit on the flattened edge marked with arrow. 16. Pycnidium producing microconidia on SNA. 17.Phialidic microconidial conidiogenous cells. 18. Microconidia ellipsoidal to subcylindrical, hyaline. FIG. 6 from 9090; 7, 8, 10,14 from 7463; 9 from 9181; 11, 12 from 9188; 13 from 7331; 15 from 3370; 16–18 from 8145. Bars: 6, 7 5 0.5 mm; 8, 9 5
200 mm; 10 5 20 mm; 11–15 5 10 mm; 16 5 0.5 mm; 17, 18 5 10 mm.
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 769
(Krieger, Fungi Saxonici 1450; BPI 402518) and madethe following observations. Conidiogenous cells of E.pini tend to proliferate by pushing aside the first-formed conidiogenous cell, which remains as a spur,to form up to two more conidiogenous cells(FIGS. 21–24). Conidiogenesis is holoblastic, but thepoint of attachment of the conidium to the con-idiogenous cell is narrow, which results in the conidiahaving a papillate base when they are discharged(FIGS. 25–28). Papillate conidia also were noted byPetrak (1940) when he proposed Endomelanconium.Conidiogenous cells of E. nanum Gamundı &Arambari (Gamundı and Arambari 1983, Suarez etal 2000) and the unnamed Endomelanconium ana-morph of Austrocenangium australe (Gamundı 1997)are illustrated as being narrowed at the point ofdehiscence of conidia and conidia are papillate.Conidia of E. phoenicicola Yanna et al (Yanna et al1999) have a flat, somewhat protuberant base but theconidiogenous cell is not illustrated as being con-stricted at the point of dehiscence. Pycnidia of E. piniare subcortical and have a convoluted hymenium; thewalls are made of small-celled pseudoparenchyma(Petrak 1940, Sutton 1980) but there is minimalstromatic development (see also Sutton 1980), unliketypical botyosphaeriaceous anamorphs. Even thoughour endomelanconium-like fungi are known onlyfrom cultures their pycnidial walls are formed ofthick-walled pseudoparenchymatous cells (see Verkleyand van der Aa 1997, FIG. 2). Conidia of E. pini areextruded in a slimy, black mass that may extendconsiderably beyond the erumpent pycnidia (alsodescribed by Petrak 1940).
The intimate juxtaposition of E. nanum (Cash)Gamundı and the unnamed Endomelanconium ana-morph with their respective Austrocenangium Ga-mundı teleomorphs (Helotiales) leave no doubt oftheir respective conspecficity (Gamundı et al 1983,
Gamundı 1997). Based on the observations notedabove, the endomelanconium-like fungi that wediscussed are unlikely to be phylogenetically relatedto E. pini, which appears to be more similar to thehelotiaceous anamorphs of Austrocenangium.
The anamorph genus Endomelanconium Petrak isparaphyletic. There are significant differences inpycnidial and conidial morphology and in prolifera-tion of conidiogenous cells among the known‘Endomelanconium’ species (TABLE II). Endomelanco-nium nanum and the unnamed Endomelanconiumanamorph of A. australe (Gamundı and ArambarriGamundı 1997), and most likely E. pini, are helotiac-eous anamorphs. Whether the Austrocenangiumanamorphs are congeneric with E. pini remains tobe proven, and proliferating conidiogenous cells arenot reported for either anamorph. Our endomelan-conium-like endophyte and E. microsporum areobviously Dothideomycetes and not likely to becongeneric with E. pini. However, apart from abroadly circumscribed Endomelanconium, we do notknow of any genus that can accommodate them.Consequently we propose a new genus, Endomelanco-niopsis. The illustrations of E. phoenicicola (Hyde et al1999) suggest that it is also botryosphaeriaceous,possibly a species of Endomelanconiopsis; the describ-ing authors did not report the deposit of cultures ofthis species.
TAXONOMY
Endomelanconiopsis Rojas et Samuels, gen. nov.Genus Endomelanconii Petrak simile sed ad Botryosphaer-
iaceas pertinens et conidia non papillata.Species typica: Endomelanconiopsis endophytica Rojas et
Samuels
Endomelanconiopsis endophytica Rojas et Samuels,sp. nov. FIGS. 3–18Endomelanconio microsporo Verkley & van der Aa simile,
sed conidia (4.7–)5.5–7.5(–10.0) 3 (3.0–)3.5–4.5(–6.2) mmproducens et chlamydosporae absunt; species endophytica.
Holotypus: Ex culto agaro sicco BPI 878370.
Mycobank 11838Optimum temperature at 30–37 C, colony radius
43–55 mm after 5 d (n 5 15) on PDA. Colonies at firstcolorless with hyaline immersed hyphae, after 4 dcolonies olivaceous in center and concentric ringswith irregular shape, after 10 d aerial mycelium densedark olivaceous or gray (Groups 1, 3) or shiny blackwith little aerial mycelium (Group 2). Conidiomataeustromatic, scattered throughout colony, varyingfrom globose to cylindrical, 1–3 cylindrical necks,superficial or immersed in the agar; often cylindricalpapillae protruding from the agar in groups of a few.
FIG. 19. Endomelanconiopsis endophytica. Box plot show-ing conidium length (above, black line) and width (below,gray line) of three phenotype groups. Group 4 isE. microsporum.
770 MYCOLOGIA
TA
BL
EII
.C
om
par
iso
no
fE
ndom
elan
con
iopsi
san
dE
ndom
elan
con
ium
spec
ies
Ch
arac
ter
En
dom
elan
con
iopsi
sen
dop
hyt
ica
1
En
dom
elan
con
iopsi
sm
icro
spor
a1
En
dom
elan
con
ium
nan
um
Gam
un
dı
and
Ara
mb
arri
1983
E.
phoe
nic
icol
aYa
nn
aet
al19
99E
.pin
iSu
tto
n19
80
En
dom
elan
con
ium
ana-
mo
rph
,A
ust
roce
nan
giu
mau
stra
leG
amu
nd
ı19
97
Tel
eom
orp
hU
nkn
ow
n,
Bo
tryo
sph
aeri
acea
eU
nkn
ow
n,
Bo
tryo
sph
aeri
acea
eA
ust
roce
nan
giu
m,
Hel
oti
ales
Un
kno
wn
Un
kno
wn
Au
stro
cen
an
giu
m,
Hel
oti
ales
Co
nid
ium
,le
ngt
h( m
m)
4.7–
8.7
(Gro
up
1)5.
4–6.
47.
2–10
10–1
2(a
v5
11.2
3)11
.5–1
3.5
12–1
36.
1–7.
4(G
rou
p2)
5.1–
9.9
(Gro
up
3)C
on
idiu
m,
wid
th( m
m)
3.1–
4.9
(Gro
up
1)4–
4.6
3.8–
5.8
9–11
(av
510
.33)
6.5–
7.5
9–10
3.7–
4.5
(Gro
up
2)3.
3–6.
2(G
rou
p3)
Co
nid
ium
bas
efl
atto
rou
nd
edfl
atto
rou
nd
edp
apil
late
flat
pap
illa
tep
apil
late
Co
nid
ioge
no
us
cell
,le
ngt
h( m
m)
7.8–
23.4
6–10
7.9–
10.5
14–2
5(a
v5
19)
7–16
9–12
Co
nid
ioge
no
us
cell
,ap
exw
idth
( mm
)0.
7–3.
5—
—2–
4(a
v5
2.88
)—
Co
nid
ioge
no
us
cell
,b
ase
wid
th(m
m)
1.3–
4.1
5–7
4.1–
4.7
2–5
(av
53.
25)
3–7
2.5–
4.5
Mic
roco
nid
ia,
len
gth
( mm
)1.
8–4.
4(G
rou
p1)
——
——
—2.
9–10
(Gro
up
3)M
icro
con
idia
,w
idth
( mm
)0.
9–1.
9(G
rou
p1)
——
——
—0.
9–2.
8(G
rou
p3)
Ch
lam
ydo
spo
res
( mm
)—
9–17
36–
10—
——
—H
ost
Theo
brom
aca
cao
and
Hei
ster
iaco
nci
nn
a,
Lea
fen
do
ph
ytes
Soil
Pla
nt
deb
ris
Phoe
nix
han
cean
a,
dea
dp
etio
leA
bies
pec
tin
ata
,A
.alb
a,
A.
exce
lsa
;d
ead
tru
nks
and
bra
nch
es
Not
hof
agu
sbe
tulo
ides
,fa
llen
bra
nch
es
1U
pp
eran
dlo
wer
95%
con
fid
ence
inte
rval
s.
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 771
Wall comprising pale brown and black angular cells,becoming hyaline and more hyphal toward theconidiogenous cells. Pycnidial locule convoluted, com-pletely lined with conidiogenous cells. Conidiogenouscells formed from the inner cells all over theconidiomata wall, discrete, determinate, cylindrical,tapered toward the apex, hyaline, holoblastic, rarelywith a single percurrent proliferation, (7.8–23.4 3
0.8–3.5 [apex] 1.3–4.1 [base] mm [Mean 5 14.2 mm 3
1.6 mm apex 3 2.2 mm base]). Conidia ellipsoidal tolimoniform, apex rounded, base flat to rounded,aseptate, hyaline when immature, dark brown with asingle longitudinal slit three-quarters of the length ofthe conidia when mature, (4.7–)5.5–7.5(–10.0) 3
(3.0–)3.5–4.5(–6.2) mm (n 5 399). Microconidiaforming in the same locules as macroconidia fromdensely arranged, enteroblastic, phialidic conidiogen-ous cells, appearing to arise from the inner cells ofthe pycnidial wall. Microconidia ellipsoidal to allan-toid, 2.0–7.0(–10.0) 3 (1.0–2.0(–3.0) mm (n 5 80),formed on PDA and SNA. Chlamydospores notobserved.
Etymology. endophytica, referring to the occurrenceof this fungus as an endophyte.
HOLOTYPE: PANAMA. Nombre de Dios, isolated fromleaves of Theobroma cacao, 2000, E. Rojas, L. Mejıa & Z.Maynard 7463 (BPI 878370, ex-type culture CBS 120379).
Commentary. Additional collections are cited (TA-
BLE I). The description given above is based on severalof these cultures grown on PDA, and it amalgamatesthe diverse measurements of the groups discussed inRESULTS. On SNA there was little mycelial productionand colonies were transparent. Pycnidia formed oftenat the edge of the colony at the interface between theagar and plastic of the Petri plate; they were muchsmaller than on PDA, nearly flask-shaped with one ormore papillae.
Endomelanconiopsis microspora (Verkley & van der Aa)Rojas et Samuels, comb. nov.; Endomelanconium microsporum Verkley & van der Aa,
Mycologia 89:967. 1997.
Mycobank 5111839This species is described and illustrated in Verkley
and Van der Aa (1997).
DISCUSSION
It comes as no surprise that Endomelanconiopsis isderived from within the Botryosphaeriaceae. Rapidlyspreading dark colonies on PDA that characterize E.endophytica and E. microspora (Verkley and van der Aa1997) are typical of Botryosphaeria s.l. (see Phillips2008 and colony descriptions in Punithalingham1978, Samuels and Singh 1986, Alves et al 2004,Luque et al 2005). The eustromatic pycnidia withoften convoluted locules (appearing multiloculate)and holoblastic conidiogenous cells arising directlyfrom cells of the pycnidial wall are further character-istics as are the rather large, unicellular conidia thatare often truncate at the base. Microconidia (sper-matia) usually are not reported for ‘Botryosphaeria’anamorphs (e.g. Pennycook and Samuels 1985,Samuels and Singh 1986, Phillips et al 2005), butthey probably are produced by all species.
Endomelanconiopsis is unusual in having a germ slitin the conidia. In the Botryosphaeriaceae we areaware only of Neodeightonia subglobosa C. Booth(teleomorph Botryosphaeria subglobosa [C. Booth]Arx. & E. Muller) as having similar conidia (seePunithalingham 1969). Crous et al (2006) observedgerm slits in conidia of this species and compared it toLasiodiplodia theobromae, the striations of which werepostulated to be germ slits. Despite the similarity ofconidia of E. endophytica and N. subglobosa, the twospecies are not closely related (FIG. 1).
Botryosphaeria and the Botryosphaeriaceae haveattracted considerable attention in recent years, withthe result that the number of identified species isincreasing rapidly along with the number of associat-ed anamorph generic names (see Crous et al 2006,Damm et al 2007). A process of dismemberingBotryosphaeria based in part on the anamorphphenotype and in part on phylogenetic analysis isunder way (Crous et al 2006, Shenoy et al 2007).There are challenges to defining genera on thesebases. First, as can be seen in (FIG. 1) there is nosupport for the backbone of the Botryosphaeriaceaewhen species of the closely related Guignardia areused as outgroup. The strongly supported lineages(FIG. 1), which is derived from Crous (2006) but witha different outgroup, could be interpreted equally asgenera or as species of Botryosphaeria. That being the
FIG. 20. Endomenanconiopsis endophytica. Growth curvesof three groups on PDA after 72 h in intermittent light.Lines 5 standard error. A total of 5, 9 and 1 isolates weremeasured respectively in Groups 1, 2 and 3.
772 MYCOLOGIA
case the argument for recognizing the lineages mustdepend on phenotype and/or biology. There is auniform biology in the Botryosphaeriaceae (seePhillips 2008); they cause cankers, root diseases andpostharvest fruit rot. Teleomorphs are very similar.Most generic differences are attributed to conidialmorphology, but while there is broad agreement ofanamorph phenotype with clades (see Crous et al2006) this is not consistent. Two primary examplesinclude the distantly related yet apparently morpho-logically indistinguishable Endomelanconiopsis, de-scribed here, and Neodeightonia subglobosa. Similarlythe highly distinctive Lasiodiplodia morphology oc-curs in two or more clades (Crous et al 2006). Thus
whether these lineages represent genera or speciescan be questioned reasonably. We have adopted thenomenclature summarized in Crous et al (2006) andin the Botryosphaeria Website (Phillips 2008) but notwithout misgivings.
Endomelanconiopsis endophytica was the dominantendophytic fungus isolated from mature to old cacaoleaves taken from the one tree that was the subject ofthe second survey at Nombre de Dios in Jul 2000.However it was not encountered in any other tree atNombre de Dios during the first survey, 1999 and2000. Thus it appears to vary greatly in its occurrenceover time, across sites and possibly among hosts. Inthe present work we discuss only endophytic isolates
FIGS. 21–28. Endomelanconium pini. 21. Two immersed pycnidia marked by glistening black masses of extruded conidia. 22–26. Holoblastic conidiogenous cells. 23. A spur-like sympodial proliferation (arrow). 24. A chain of conidiogenous cells. 23, 25,26. Constriction at the point of attachment of the conidium to the conidiogenous cell. 27–28. 27. Conidia, germ slits at arrow. 28.A papillate conidial base at arrow. From Krieger, Fungi Saxonici 1450; BPI 04024518. Bars: 21 5 0.5 mm, 22–28 5 10 mm.
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 773
from Theobroma and Heisteria. However isolates withidentical ITS sequences and consistent colony mor-phology have been recovered from additional treesand shrubs representing the Flacourtiaceae, Melia-ceae and Ochnaceae at Barro Colorado Island inPanama (Arnold unpubl). It is likely that E.endophytica has a much wider host and geographicrange than we record here. It also is strongly possiblethat reports of endophytic isolates of species such asDiplodia juglandis, Botryosphaeria sarmentorum, B.viticola or B. iberica, all of which give 97% maximumidentity and 100% sequence coverage (LSU locus),actually could refer to E. endophytica. The wide hostrange is reminiscent of Guignardia mangiferae (anam:Phyllosticta capitalensis), a ubiquitous, cosmopolitanendophyte of many genera of woody plants (Baayen etal 2002), and a species that we also have recovered asan endophyte from leaves of cacao. These fungi areknown only as endophytes in a wide range of hosts.
ACKNOWLEDGMENTS
The authors thank Amy Rossman for comments on thismanuscript; Adnan Ismaiel (BPI) for valuable advice duringthe sequencing phase of this project; Zuleyka Maynard, LuisRamirez and Terri Shirshac for help with the culturecollection; and USDA-ARS-SMML and the SmithsonianTropical Research Institute for research infrastructure andlogistical support. Walter Gams and Orlando Petrinicorrected the Latin diagnoses. Financial support wasprovided by World Cocoa Foundation (WCF), the JohnClapperton Fellowship of Mars Inc., the Andrew W. MellonFoundation and the Smithsonian Tropical Research Insti-tute. ER was the recipient of a Norman E. BorlaugFellowship in support of this work.
LITERATURE CITED
Arnold AE, Engelbrecht BMJ. 2007. Fungal endophytesnearly double minimum leaf conductance in seedlingsof a tropical tree. J Tropic Ecol 23:369–372.
———, Lutzoni F. 2007. Diversity and host range of foliarfungal endophytes: Are tropical leaves biodiversityhotspots? Ecology 88:541–549.
———, Mejıa L, Kyllo D, Rojas E, Maynard Z, Herre EA.2003. Fungal endophytes limit pathogen damage in atropical tree. Proc Nat Acad Sc USA 100:15649–15654.
———, Maynard Z, Gilbert GS, Coley PD, Kursar TA. 2000.Are tropical fungal endophytes hyperdiverse? Ecol Lett3:267–274.
Baayen RP, Bonants PJM, Verkley G, Carroll GC, van der AaHA, de Weerdt M, van Brouwershaven IR, Schutte GC,Maccheroni W Jr, de Blanco CG, Azevedo JL. 2002.Nonpathogenic isolates of the citrus black spot fungus,Guignardia citricarpa, identified as a cosmopolitanendophyte of woody plants, G. mangiferae (Phyllostictacapitalensis). Phytopathology 92:464–467.
Brayford D. 1990a. Vegetative incompatibility in Phomopsisfrom elm. Mycol Res 94:745–732.
———. 1990b. Variation in Phomopsis isolates from Ulmusspecies in the British Isles and Italy. Mycol Res 94:691–697.
Carbone I, Kohn LM. 1999. A method for designing primersets for speciation studies in filamentous ascomycetes.Mycologia 91:553–556.
Clay K, Holah J. 1999. Fungal endophyte symbiosis andplant diversity in successional fields. Science 285:1742–1744.
Crous PW, Slippers B, Wingfield MJ, Rheeder J, MarasasWFO, Philips AJL, Alves A, Burgess T, Barber P,Groenewald JZ. 2006. Phylogenetic lineages in theBotryosphaeriaceae. Stud Mycol 55:235–253.
Damm U, Fourie PH, Crous PW. 2007. Aplosporellaprunicola, a novel species of anamorphic Botryosphaer-iaceae. Fung Divers 27:35–43.
Dreyfuss M, Petrini O. 1984. Further investigations on theoccurrence and distribution of endophytic fungi intropical plants. Bot Helvet 94:33–40.
Gamundı de Amos I, Arambarri AM. 1983. Bulgaria nanaCash y su anamorfo Endomelanconium (HelotiaceaeOmbrophiloidea). Rev Fac Agron Univ Nac La Plata 59:17–23.
Gamundı IJ. 1997. Austrocenangium gen. nov. from south-ern South America. Mycotaxon 63:261–268.
Gilbert GS, Mejia-Chang M, Rojas E. 2002. Fungal diversityand plant disease in mangrove forest: salt excretion as apossible defense mechanism. Oecologia 132:278–285.
Hawksworth DL. 2001. The magnitude of fungal diversity:the 1.5 million species estimate revisited. Mycol Res105:1422–1432.
Herre EA, van Bael SA, Maynard Z, Robbins N, Bischoff J,Arnold AE, Rojas E, Mejıa LC, Cordero RA, WoodwardC, Kyllo DA. 2005. Tropical plants as chimera: someimplications of foliar endophytic fungi for the study ofhost plant defense, physiology and genetics. In:Burslem DFRP, Pinard MA, Hartley SE, eds. Bioticinteractions in the tropics. Cambridge, UK: CambridgeUniv Press. p 226–237.
———, Mejia LC, Kyllo DA, Rojas E, Maynard Z, Butler A,van Bael SA. 2007. Ecological implications of anti-pathogen effects of tropical fungal endophytes andmycorrhizae. Ecology 88:550–558.
Huelsenbeck JP, Ronquist F. 2001. MrBayes: Bayesianinference of phylogenetic trees. Bioinformatics 17:754–755.
Hyde Y, Hyde KD, Goh TK. 1999. Endomelanconiumphoenicicola, a new coelomycete from Phoenix hanceanain Hong Kong. Fung Div 2:199–204.
Leigh EG Jr, Rand AS, Windsor DM. 1996. The ecology of atropical forest. 2nd ed. Washington, DC: SmithsonianInstitution Press. 503 p.
Lodge DJ, Fisher PJ, Sutton BC. 1996. Endophytic fungi ofManilkara bidentata leaves in Puerto Rico. Mycologia88:733–738.
Luque J, Martos S, Phillips AJL. 2005. Botryosphaeria viticolasp. nov. on grapevines: a new species with a Dothiorellaanamorph. Mycologia 97:1111–1121.
774 MYCOLOGIA
Maddison DR, Maddison WP. 2000. MacClade 4: analysis ofphylogeny and character evolution. Sunderland, Mas-sachusetts: Sinauer Associates.
Mejıa LC, Rojas EI, Maynard Z, Arnold AE, Kyllo D, RobbinsN, Herre EA. 2003. Inoculation of beneficial endophyt-ic fungi into Theobroma cacao tissues. Proc 14th IntCocoa Res Conf, Accra, Ghana. Vol. II. p 669–705.
———, ———, ———, van Bael S, Arnold AE, Hebbar P,Samuels GJ, Robbins N, Herre EA. 2008. Endophyticfungi as biocontrol agents of Theobroma cacao patho-gens. Biol Cont 46:4–14.
Nirenberg HI. 1976. Untersuchungen uber die morpholo-gische und biologische Differenzierung in der Fusari-um-Sektion Liseola. Mitt Biol Bundesanst Land- ForstwBerlin-Dahlem 169:i–v + 1–117.
Nylander JAA. 2004. MrModeltest v2. Program distributedby the author. Uppsala, Sweden: Evolutionary BiologyCentre, Uppsala University.
Pennycook SR, Samuels GJ. 1985. Botryosphaeria andFusicoccum species associated with ripe fruit rot ofActinidia deliciosa (kiwifruit) in New Zealand. Myco-taxon 24:445–458.
Petrak F. 1940. Mykologische Notizen XIII. Ann Mycol 38:205–207.
Phillips AJL, Alves A, Correia A, Luque J. 2005. Two newspecies of Botryosphaeria with brown, 1-septate ascosporesand Dothiorella anamorphs. Mycologia 97:513–529.
———. 2008. The Botryosphaeria site. http://www.crem.fct.unl.pt/botryosphaeria_site/index.htm
Punithalingham E. 1969. Studies on Sphaeropsidales inculture. Mycol Pap 119:1–24 + 5 pl.
———. 1978. Botryodiplodia theobromae. Descriptions ofPathogenetic Fungi and Bacteria 519. CommonwealthAgricultural Bureau.
Rannala B, Yang Z. 2005. Probability distribution ofmolecular evolutionary trees: a new method of phylo-genetic inference. J Mol Evol 43:304–311.
Rehner SA, Samuels GJ. 1994. Taxonomy and phylogeny ofGliocladium analysed from nuclear large subunitribosomal DNA sequences. Mycol Res 98:625–634.
Rodrigues KF, Petrini O. 1997. Biodiversity of endophyticfungi in tropical regions. In: Hyde KD, ed. Biodiversityof tropical microfungi. Hong Kong: Hong Kong UnivPress. p 57–69.
Samuels GJ, Singh B. 1986. Botryosphaeria xanthocephala,cause of stem canker in pigeon pea. Trans Br Mycol Soc86:295–199.
———, Dodd SL, Gams W, Castlebury LA, Petrini O. 2002.Trichoderma species associated with the green moldepidemic of commercially grown Agaricus bisporus.Mycologia 94:146–170.
Schoch CL, Shoemaker RA, Seifert KA, Hambleton S,Spatafora JW, Crous PW. 2006. A multigene phylogenyof the Dothideomycetes using four nuclear loci.Mycologia 98:1041–1052.
Shenoy BD, Jeewon R, Hyde KD. 2007. Impact of DNAsequence-data on the taxonomy of anamorphic fungi.Fung Divers 26:1–54.
Suarez V, Carmaran CC, Sutton BC. 2000. Melanconiopsismicrospora sp. nov from bamboo in Argentina. MycolRes 104:1530–1534.
Suryanarayanan TS, Kumaresan V, Johnson JA. 1998. Foliarfungal endophytes from two species of the mangroveRhizophora. Microbiology 44:1003–1006.
Sutton BC. 1980. The Coelomycetes. Fungi imperfecti withpycnidia, acervuli and stromata. Kew, UK: Common-wealth Mycological Institute. 696 p.
Swofford DL. 2002. PAUP*: phylogenetic analysis usingparsimony (*and other methods). version 4.0b10win32. Sunderland, Massachusetts: Sinauer Associates.
Thompson JD, Gibson TJ, Plewniak F, Jenmougin F,Higgins DG. 1997. The Clustal X Windows interface:flexible strategies for multiple sequence alignmentaided by quality analysis tools. Nuc Acid Res 24:4876–4882.
Tosi JA. 1971. Zonas de vida, una base ecologica parainvestigaciones silvıcolas e inventariacion forestal en laRepublica de Panama. Programa de las NacionesUnidas para el desarrollo, FAO, Roma, Italia. 76, 87 p.
van Bael SA, Maynard Z, Robbins N, Bischoff J, Arnold AE,Rojas E, Mejia LC, Kyllo DA, Herre EA. 2005. Emergingperspectives on the ecological roles of endophytic fungiin tropical plants. In: Dighton J, Oudemans P, White J,eds. The fungal community: its organization and rolein the ecosystem. 3rd ed. Boca Raton, Florida: CRCPress. p 181–191.
Verkley GJM, van der Aa H. 1997. Endomelanconiummicrosporum, a new coelomycete isolated from soil inPapua New Guinea. Mycologia 89:967–970.
Vinale F, Sivasithamparam K, Ghisalberti EL, Marra A, WooSL, Lorito M. 2007. Trichoderma-plant pathogeninteractions. Soil Biol Biochem. doi:10.1016/j.soil-bio.2007.07.002.
White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification anddirect sequencing of fungal ribosomal RNA genes forphylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ,White TJ, eds. PCR protocols: a guide to methods andapplications. San Diego: Academic Press. p 315–322.
Wilson D. 1995. Endophyte—the evolution of a term, andclarification of its use and definition. Oikos 73:274–276.
ROJAS ET AL: ENDOMELANCONIOPSIS GEN. NOV. 775