Biodiversity of Lecanosticta pine-needle blight pathogens suggests a Mesoamerican Centre of origin

Lecanosticta acicola causes the disease known as brown spot needle blight (BSNB), on Pinus species. The pathogen is thought to have a Central American centre of origin. This was based on the morphological variation between isolates believed to represent L. acicola from native Pinus spp. Two species of Lecanosticta, L. brevispora and L. guatemalensis, have recently been described from Mexico and Guatemala respectively based on morphology and sequence-derived phylogenetic inference. However, the putative native pathogen, L. acicola, was not found in those areas. In this study, the species diversity of a large collection of Lecanosticta isolates from Central America was considered. Phylogenetic analyses of the BT1, ITS, MS204, RPB2 and TEF1 gene regions revealed six species of Lecanosticta, four of which represented undescribed taxa. These are described here as Lecanosticta jani sp. nov. from Guatemala and Nicaragua, L. pharomachri sp. nov. from Guatemala and Honduras, L. tecunumanii sp. nov. from Guatemala and L. variabilis sp. nov. from Guatemala, Honduras, and Mexico. New host and country records were also found for the previously described L. brevispora and L. guatemalensis. Lecanosticta acicola was not found in any of the samples from Central America, and we hypothesize that it could be a northern hemisphere taxon. The high species diversity of Lecanosticta found in Mesoamerica suggests that this is a centre of diversity for the genus. Electronic supplementary material The online version of this article (10.1186/s43008-019-0004-8) contains supplementary material, which is available to authorized users.


INTRODUCTION
Brown spot needle blight (BSNB) or Lecanosticta needle blight is an important needle disease on Pinus species. The disease is characterised by brown spots on necrotic yellow lesions at the points of infection and die-back of the needles from the apex, which often leads to premature defoliation (Ivory 1987). BSNB is caused by the fungal pathogen, Lecanosticta acicola (Siggers 1944). The fungus is a well-known pathogen in the USA and has also been recorded in Central America, Colombia, Europe as well as Asian countries including China, Japan and Korea. Lecanosticta acicola is regarded as an A2 quarantine pathogen in Europe and Colombia where it is present as well as an A1 quarantine pathogen in the rest of South America (COSAVE), Africa (IASPC) and the Eurasian Economic Union countries where it has yet to be recorded (https://gd.eppo.int/taxon/SCIRAC/categorization). Despite its quarantine status, L. acicola has been discovered in various new locations and on new hosts in Europe during the past decade (Jankovsky et al. 2009;Markovskaja et al. 2011;Anonymous 2012;Hintsteiner et al. 2012;Adamson et al. 2015;Janoušek et al. 2016;Ortíz de Urbina et al. 2017;Mullett et al. 2018;Cleary et al. 2019;Sadiković et al. 2019). Siggers (1944) and Evans (1984) summarised the taxonomic and nomenclatural history of Lecanosticta acicola, which was complicated by the former system which allowed asexual and sexual morphs of the same species of fungi to be given separate scientific names (Kais 1971;Evans 1984). From 1972 to 2012, the name Mycosphaerella dearnessii was widely used for the causal agent of BSNB. It was, however, recently recognised that Mycosphaerella is polyphyletic and should be strictly used for could represent cryptic species. The present study emerged from an opportunity to collect pine needles infected with Lecanosticta spp. in Guatemala, Honduras and Nicaragua from 2010 to 2012. Specimens were identified based on DNA sequence comparisons and an attempt was made to confirm whether L. acicola occurs in Central America.

Collections used in the study
Specimens prepared from ex-type cultures and other representatives of all known Lecanosticta species and closely related species  were obtained from the culture collection of the Westerdijk Fungal Biodiversity Institute, Utrecht, The Netherlands (CBS), and from the UK National Fungus Collection maintained by CABI Bioscience (Egham, UK: IMI). Living cultures or DNA of six isolates from Central America examined by Evans (1984), and believed to represent L. acicola, were also acquired from IMI (Table 1). Furthermore, isolates of Dothistroma septosporum, D. pini, Phaeophleospora eugenia, P. gregaria, and Amycosphaerella africana that represent genera in Mycosphaerellaceae closely related to Lecanosticta  were included for comparative purposes. These cultures were obtained from CBS and the culture collection (CMW) of the Forestry and Agricultural Biotechnology Institute (FABI) in Pretoria, South Africa (Table 1).
Pine needles, showing symptoms of brown spots or bands, were collected from Pinus species native to Central America from 2010 to 2012 in Guatemala, as well as from Honduras and Nicaragua in 2011 (Table 1). Conidiomata formed on the needles were aseptically excised, rolled onto 2% Dothistroma Sporulating Media (DSM: 5 g yeast extract (Biolab, Merck, Modderfontein, South Africa), 20 g malt extract (Biolab) and 15 g agar (BD Difco™, Sparks, MD) per litre of distilled water) with 100 mg/L streptomycin (Sigma-Aldrich, St Louis, MO) in order to release conidia from the conidiomata as described by Barnes et al. (2004). The isolated conidiomata were incubated for one to two days at 23°C. The plates were examined using a dissection microscope and single germinating conidia were selected and replated onto 2% DSM. The single conidial isolates were grown for 4-6 wk. on a natural day light cycle, at 23°C.

DNA extractions and sequencing
Fungal tissue was scraped from the surface of the cultures on 2% DSM with a sterile scalpel blade and lyophilized. The freeze-dried mycelium was homogenized using a Retsch MM301 mixer mill (Haan, Germany) and approximately 20 ng of the crushed mycelium was used as starting material for DNA extractions. DNA was extracted using a    Zymo Research ZR Fungal/Bacterial DNA MiniPrep™ kit (Irvine, CA) and eluted into a final volume of 50 μl. The quality and quantity of the extracted DNA was determined using a NanoDrop ND-1000 spectrophotometer (Thermo Fischer Scientific, Waltham, MA). DNA concentrations were diluted to 20 ng/μl working stock for polymerase chain reaction (PCR) amplifications and stored at − 20°C until further use. The nuclear rDNA region encompassing the internal transcribed spacers (ITS) 1 and 2, along with the 5.8S rDNA region was amplified using primers ITS1 and ITS4 (White et al. 1990) and a portion of the translation elongation factor 1-α gene (TEF1) using primers EF1-728F (Carbone and Kohn 1999) and EF2 (O'Donnell et al. 1998) for all the isolates. The Beta-tubulin-2 gene region (BT2) was amplified using the primer pair T1 (O'Donnell and Cigelnik 1997) and β-Sandy-R (Stukenbrock et al. 2012) or the primers Bt2A and Bt2B (Glass and Donaldson 1995). The Beta-tubulin-1 gene region (BT1) was amplified using primers Bt1A and Bt1B (Glass and Donaldson 1995), the RNA polymerase II second largest subunit (RPB2) gene region using primers RPB2-5f2 (Sung et al. 2007) and RPB2-7cR (Liu et al. 1999) and the guanine nucleotide-binding protein subunit beta (MS204) using primers MS204F.cerato and MS204R.cerato (Fourie et al. 2015).
PCR reactions for each of the six regions contained 20 ng DNA, 2.5 μl 10x PCR reaction buffer, 2.5 mM MgCl 2 , 400 nM of each primer, 200 μM of each dNTP and 1 U Faststart Taq DNA Polymerase (Roche Diagnostics, Indianapolis, IN). Reaction volumes were adjusted to 25 μl with sterile SABAX water (Adcock Ingram, Midrand, South Africa). PCR reactions were carried out on an Applied Biosystems® Veriti® 96 well Thermal cycler (Thermo Fisher Scientific, Waltham, MA). The cycling conditions for all six gene regions included an initial denaturation step at 95°C for 4 min, 10 cycles consisting of 94°C for 20 s (denaturation), a 45 s annealing step according to the primer pair annealing temperature (Table 2) and an elongation step of 45 s at 72°C. This was followed by a further 25 cycles of 94°C for 20 s, 45 s with a 5 s extension step per cycle at the indicated annealing temperature, a 72°C extension for 45 s and a final step of 72°C for 10 min. The annealing temperature was set at 56°C for ITS, 52°C for TEF1, 50°C for BT1, 52°C for BT2, 55°C for MS204 and 56°C for RPB2. To visualise amplified products, 5 μl PCR products were stained with 1 μl GelRed™ nucleic acid gel stain (Biotium, Fremont, CA) and separated on 2% SeaKem® LE agarose gel (Lonza, Rockland, ME) for 20 min at 100 V and viewed under a UV light using the GelDoc™ EZ Imager (BioRad, Hercules, CA). PCR products were cleaned with a 6.65% G-50 Sephadex solution (Sigma-Aldrich, St Louis, MO) following the manufacturer's instructions using Centri-sep spin columns (Princeton Separations, Freehold, NJ).
The concentrations of the cleaned PCR products were determined using a NanoDrop ND-1000 spectrophotometer and 60-100 ng of DNA and products were sequenced in both the forward and reverse direction using the BigDye Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific) on an ABI PRISM 3500xl capillary auto sequencer (Thermo Fisher Scientific).

Data analyses
Five datasets (BT1, ITS, MS204, RPB2 and TEF1) were generated and analysed individually. A partition homogeneity test (PHT) was performed with the software package PAUP* 4.0b10 (Swofford 2003) to test congruence between the five gene regions and a sixth dataset, where sequences were available for all five gene regions, was compiled and analysed. The BT1, ITS, MS204 and RPB2 datasets included all of the sequences generated in this study and additional sequences available from Gen-Bank (Table 1). The TEF1 dataset included all of the sequence data generated in this study as well as additional sequences representing 14 different TEF1 haplotypes of L. acicola (including possible cryptic species) (Janoušek et al. 2016) that were downloaded from GenBank (Table 3). Sequences for all datasets were aligned with the online version of MAFFT Version 7 (Katoh and Standley 2013; http://mafft.cbrc.jp/alignment/server/) using default settings. Aligned data were imported into MEGA 7.0.14 (Kumar et al. 2016) and manually checked and adjusted. Three separate analyses were performed for each of the six datasets: Maximum Parsimony (MP), Maximum Likelihood (ML) and Bayesian inference (BI). The MP analysis were performed with the software package PAUP* 4.0b10 (Swofford 2003). Gaps were treated as a fifth character state. One thousand random stepwise addition heuristic searches were performed with tree-bisection-reconnection (TBR) as the branch-swapping algorithm. Uninformative characters were excluded and the consistency index (CI), homoplasy index (HI), rescaled consistency index (RC), retention index (RI) and tree length (TL) were determined for the resulting trees (Table 4). The confidence levels were estimated by performing 1000 bootstrap replicates.
In order to determine the ML and BI, the best fit substitution model for each of the data sets were determined using jModelTest 0.1.1 (Posada 2008). Maximum likelihood analysis was performed with the program PhyML 3.0 (Guindon et al. 2010). The confidence levels were estimated with 1000 bootstrap replicates.
MrBayes 3.1.2 (Ronquist et al. 2012) was used to determine the BI for each data set by applying the Markov Chain Monte Carlo (MCMC) method. For each dataset, four independent MCMC chains were randomly started and run for six million generations, applying the best substitution model determined by jModelTest 0.1.1. Trees were sampled every 100 generations. Burn-in values were determined using Tracer 1.6 (Rambaut et al. 2014) by comparing the log likelihoods. Trees sampled in the burn-in phase were discarded. The remaining trees were used to construct majority rule consensus trees and to determine posterior probabilities for the tree topology.

Morphological characterization
Cultures were grown on 2% Malt Extract Agar (MEA), Oatmeal Agar (OA) and Potato Dextrose Agar (PDA) (Crous et al. 2009b;Quaedvlieg et al. 2012) at 20°C for 2 wk. in darkness in order to examine the morphology and colour of the cultures of each species. Cultures on MEA were used for microscopic measurements of the conidiophores, conidiogenous cells and conidia. Slides were mounted in SABAX water (Adcock Ingram, Midrand, South Africa) for microscopy and examined using a Zeiss Axioskop 2 Plus compound microscope (Zeiss, Oberkochen, Germany). Photographic images were captured with a Nikon DS-Ri2 camera with the NIS Element BR v4.3 software package (Nikon, Tokyo, Japan). Up to 50 measurements of each morphologically characteristic structure was taken for each ex-type isolate and ten measurements were made for each of the paratypes examined. The mean, standard deviation, minimum and maximum were calculated for each morphological structure and the measurements presented as (minimum-) (meanstandard deviation) -(mean + standard deviation) (−maximum) for the conidia and conidiogenous cells. For the conidiophores, the maximum observed length was indicated together with the width as (minimum-) (mean) (−maximum).
Temperature requirements for growth in culture was studied on representative isolates selected for each of the novel species. Four by four millimeter blocks of each culture were plated, in triplicate, onto the centres of 2% MEA plates per temperature (10, 15, 20, 25, and 30°C) and incubated in darkness. The diameters of each colony were recorded weekly along perpendicular axes for 4 wk. The colour and shape of each colony was recorded after 2 wk. of growth at 20°C. Culture colour was determined using Rayner's colour chart (Rayner 1970).

Accession of cultures and types
Holotype specimens of the new species, which are dried cultures, are deposited in the National Mycological Herbarium in Pretoria (PREM). Cultures are deposited in the culture collection (CBS) of the Westerdijk Fungal Biodiversity Institute, Utrecht, The Netherlands, and ex-type cultures, as well as all other isolates included in this study, are maintained in the culture collection (CMW) of the Forestry and Agricultural Biotechnology Institute (FABI) in Pretoria, South Africa (Table 5).

Fungal collections
Twenty-six isolates or DNA samples were obtained from culture collections to include in the study. An additional 127 isolates of putative Lecanosticta species were obtained from symptomatic needles collected from 36 different trees in Guatemala, Nicaragua and Honduras (Table 1). In Guatemala, 22 isolates were obtained from Pinus oocarpa, P. maximinoi, and P. tecunumanii needles that were collected in the Alta Verapaz District, 16 isolates were obtained from P. oocarpa needles collected in Chiquimula, 35 isolates from P. pseudostrobus needles collected in the Chimaltenango District in the Tecpán Municipality, eight isolates from P. tecunumanii needles collected in the Baja Verapaz District, 29 isolates from P. tecunumanii and P. oocarpa needles collected in the Jalapa District, and seven isolates from P. maximinoi needles in Coban and other regions (Table 1). Two isolates were obtained from P. oocarpa needles collected in Honduras and eight isolates were made from P. oocarpa needles collected in Matagalpa, Nicaragua.

DNA extraction and sequencing
The ITS and TEF1 regions were sequenced for all 153 isolates obtained and the BT1, MS204 and RPB2 regions were sequenced for 127 representatives of all monophyletic groups identified in the generated ITS and TEF1 phylogenetic trees. The selected representatives included all of the closely related Mycosphaerellaceae isolates, all the isolates that did not group with known Lecanosticta species, and a selection of isolates that grouped with known Lecanosticta species (Table 1). PCR fragments of approximately 550 bp were generated for ITS, 520 bp for TEF1, 420 bp for BT1, 760 bp for MS204 and 940 bp for RPB2. The amplification success of the TEF1, BT1, MS204 and RPB2 gene regions varied for the isolates that were selected and the amplification success rate of TEF1 was 88.2%, BT1 was 87.4%, MS204 was 71.7 and 82.7% for the RPB2 region ( Table 2). The BT2 region did not amplify well across species of Lecanosticta. The amplification success rate and subsequent sequencing of the BT2 region using the T1 and β-Sandy-R primer pair, as well as Bt2a and Bt2b was very poor and further analysis of the BT2 region was abandoned.

Phylogenetic analyses
For the analyses, the datasets of the ITS region consisted of 153 taxa with 734 aligned nucleotides including gaps; the TEF1 dataset consisted of 147 taxa with 586 aligned nucleotides, the BT1 dataset consisted of 111 taxa with 440 aligned nucleotides; the MS204 dataset consisted of 91 taxa with 785 aligned nucleotides, and the RPB2 dataset consisted of 105 taxa with 929 aligned nucleotides, all including gaps. The PHT test yielded a P value = 0.01 and therefore the five datasets were considered incongruent. However, it was previously argued that a P value > 0.01 did not reduce phylogenetic accuracy (Cunningham 1997) and a combined phylogenetic tree representing the five gene regions ITS, TEF1, BT1, MS204 and RPB2 was constructed for presentation purposes (Fig. 1). The combined dataset consisted of 76 taxa with 3344 aligned nucleotides including gaps. Constant characters, parsimony-uninformative and informative characters, the consistency index (CI), homoplasy index (HI), rescaled consistency index (RC), retention index (RI) and tree length (TL) values for the maximum parsimony analyses are indicated in Table 4. For the parsimony analyses, 108 trees were retained for ITS, 396 for TEF1, 1 for BT1, 2448 for MS204 and 420 for RPB2. The best fit substitution models for ML and BI were selected by Akaike Information Criterion (AIC) and are indicated in Table 4. A 10% burn-in value was selected in the BI analysis for each of the data matrices for each of the analyses. Because the MP, ML and BI analysis all resulted in similar tree topologies, the ML trees were selected and chosen for presentation (Figs. 1 and 2, Additional file 1: Figure S1, Additional file 2: Figure S2, Additional file 3: Figure S3 and Additional file 4: Figure S4). Phylogenetic analyses of the combined dataset ( Fig. 1), ITS (Additional file 1: Figure S1), TEF1 (Fig. 2) and MS204 (Additional file 3: Figure S3) consistently grouped the isolates sequenced in this study into seven distinct clades. The clades in Fig. 2 and Additional file 1: Figure S1, Additional file 2: Figure S2, Additional file 3: Figure S3 and Additional file 4: Figure S4 are labelled according to the clades assigned in Fig. 1. In the case of RPB2 (Additional file 4: Figure S4) Clades 1-4, and 7 were also present but Clades 5 and 6 were not distinct from each other for this particular gene region. In the case of BT1 (Additional file 2: Figure S2), Clades 3, 5 and 6 could not be distinguished from each other. None of the isolates grouped with the types of L. gloeospora or L. longispora.
Forty-two of the isolates from Central America grouped in Clade 1 based on the ITS analysis (Additional file 1: Figure S1) and were identified as Lecanosticta brevispora. This was the most common species identified from the Central American collection and most isolates were from Chimaltenango on Pinus pseudostrobus. The pathogen was also isolated from P. oocarpa needles near Jalapa as well as near Tactíc in Guatemala and in Honduras. This clade was well supported for all five of the gene regions analysed. Twenty-seven isolates grouped into Clade 2 in the ITS analyses (Additional file 1: Figure S1) and represent an undescribed species. Clade 2 resolved into two subclades in the five gene analyses. Subclade 1 was mostly isolated from Chiquimula and Alta Verapaz in Guatemala on P. oocarpa, P. maximinoi and P. tecunumanii as well as from P. oocarpa in Nicaragua. Isolates collected in Jalapa in Guatemala mostly grouped into Subclade 2. However, the topology of isolates CMW 47109 (Subclade 1 on Additional file 1: Figure S1, Additional file 3: Figure S3, Additional file 4: Figure S4; Subclade 2 on Fig. 2), CMW 51059 (Subclade 1 on Additional file 1: Figure S1, Additional file 3: Figure S3, Additional file 4: Figure S4), IB30.2b (Subclade 1 on Additional file 1: Figure S1, Additional file 3: Figure S3; Subclade 2 on Fig. 2) and IB30.2b (Subclade 1 on Additional file 1: Figure S1, Additional file 3: Figure S3, Additional file 4: Figure S4; Subclade 2 on Fig. 2) changed in the two subclades depending on the gene region analysed (Fig. 2, Additional file 1: Figure S1, Additional file 3: Figure S3, Additional file 4: Figure S4). Furthermore, the two subclades were not well supported for the BT1 gene region. Therefore, the two subclades are treated here as representing a single species.
Clade 3 also represented an undescribed Lecanosticta species. This clade included 11 isolates from P. oocarpa in Jalapa, Guatemala, one isolate from P. oocarpa in Honduras, as well as five isolates collected from Baja Verapaz in Guatemala on P. tecunumanii. This clade had high bootstrap support for TEF1, MS204 and RPB2 but was not well supported in the ITS and BT1 gene regions. Three isolates collected from different needles on a single P. tecunumanii tree in Baja Verapaz in Guatemala grouped together in Clade 4 and represent another undescribed species. With the exception of BT1, Clade 4 was statistically well supported in all the gene regions that were analysed. Clade 5 accommodated sequences representing nine of the 14 known TEF1 haplotypes of L. acicola identified by Janoušek et al. (2016). These TEF1 haplotypes represent isolates collected from North America (Canada, USA, and Mexico), South America (Colombia), Europe (Spain, France, Switzerland, Slovenia, Lithuania, Italy, Germany, Estonia, Czech Republic, Croatia, and Austria) and Asia (South Korea, Japan, and China) ( Table 3). This clade was clearly distinct from other clades in the ITS, TEF1, BT1 and MS204 phylogenetic analysis and statistically well supported in the ITS, TEF1, and MS204 analyses. Clade 5 included the ex-type of L. acicola and therefore is that species. None of the isolates from Central America obtained in the present study grouped with this clade in any of the gene regions analysed.
The remaining five assigned L. acicola TEF1 haplotypes considered by Janoušek et al. (2016), grouped together in Clade 6. This was together with an isolate obtained from P. caribaea in Honduras collected in 1983 (Evans 1984), four isolates obtained in the present study from Guatemala on P. oocarpa and P. maximinoi, and an isolate previously identified as L. acicola from Mexico on an unknown Pinus species ). In the present study, Clade 6 is treated as a novel taxon. The ITS, TEF1, BT1 and MS204 gene regions clearly distinguish Clades 5 and 6, however, RPB2 was not effective in resolving these two groups.
The second most abundant species collected in this study was Lecanosticta guatemalensis, represented by Clade 7 in the phylogenetic analyses. This clade was well supported in all five gene regions that were analysed. A total of 37 isolates from our collection grouped together with L. guatemalensis based in the ITS and TEF1 analyses. Lecanosticta guatemalensis was identified on P. maximinoi and P. oocarpa in various regions of Guatemala, as well as on P. oocarpa in Nicaragua. Isolates that had previously been collected in Nicaragua and Honduras and that were identified as L. acicola by Evans (1984) based on morphological characteristics also grouped with L. guatemalensis in the present study.

TAXONOMY
Using phylogenetic analyses, 51 of the Lecanosticta isolates obtained from Guatemala, Honduras and Nicaragua, one isolate obtained from CBS, and one isolate obtained from IMI, were found to include four undescribed species. These are described below as follows: Lecanosticta jani van der Nest, M.J. Wingf. & I. Barnes, sp. nov. MycoBank MB 826875. (Fig. 3) Etymology: The name is derived from Janus, the Roman god of gates and doorways having two faces or sides, and refers to the variable culture morphology ranging from light pink and fluffy to dark olive green and mucoid.
Diagnosis: Lecanosticta jani can be distinguished from the closely related L. brevispora by the distinct globose basal cells on the conidiophores that are mostly observed on MEA.
Culture characteristics: Colonies with two distinct morphologies. One type (Type 1), flat to somewhat erumpent, spreading with flat to fluffy aerial mycelium. A second type (Type 2) erumpent, mucoid and shiny, with irregular form and undulate to filiform edges. On MEA, the surface of Type 1 isolates pale to rosy vinaceous, reverse flesh to peach coloured. Type 2 (See figure on previous page.) Fig. 1 Maximum likelihood tree representing the five known and four novel species of Lecanosticta generated from the combined data of the ITS, TEF1, BT1, MS204 and RPB2 gene regions. MP bootstrap support (> 70%) are indicated first, followed by ML bootstrap values (MP/ML, * = insignificant value). Bold branches indicate BI values > than 0.95. Dothistroma septosporum was used as the outgroup taxa. The indicated clades are referred to in the text. All represented type species are indicated in bold and with a "T" isolates citrine to isabelline, reverse olivaceous to fuscuous black (Fig. 3). On PDA, Type 1 surface rosy vinaceous to peach in centre with dark brown edge, isabelline in reverse. Type 2, surface dark olivaceous with fuscious black centres and tufts of isabelline mycelium at edges, dark isabelline in reverse. On OA, Type 1 surface dirty white to pale vinaceous, fluffy mycelia to flat growth. Type 2 surface flat with smooth edge, fuscious black in centre at the point of inoculation with light apricot surrounding mycelium. Growth characteristics: optimal growth temperature for Type 1 isolates 25°C, after 4 wk., colonies at 10, 15, 20, 25 and 30°C reached maximum of 10.5, 22, 32, 32 and 10 mm respectively, with mean growth rate of 2.1, 5.1, 6.9, 7 and 1.8 mm / wk. respectively. Type 2 isolates optimal growth temperature 20°C, after 4 wk., colonies at 10, 15, 20, 25 and 30°C reached maximum of 12.5, 17, 29.5, 22 and 4.5 mm, with mean growth of 2.1, 3.3, 5.5, 5 and 1 mm / wk. respectively.
Notes: Lecanosticta jani resolved in a distinct clade (Clade 2, Figs. 1 and 2, Additional file 1: Figure S1, Additional file 2: Figure S2, Additional file 3: Figure  S3 and Additional file 4: Figure S4) based on all five gene regions considered. This clade divides into two subclades that were mostly represented by isolates obtained from Alta Verapaz and Chiquimula in Guatemala as well as in Nicaragua in subclade 1 and isolates obtained from Jalapa in Guatemala in subclade 2. Jalapa isolates all had the Type 2 morphology and the dark colour was associated with conidial production. Type 1 isolates produced few spores after 2 wk. The optimal growth temperature and growth rates were different for the two isolate types. However, the topology of some isolates changed between the two subclades depending on the gene region that is analysed and therefore the subclades are treated as one species. The morphological variation suggests that the two types could represent two ecotypes. MycoBank MB 826876. (Fig. 4) Etymology: The epithet refers to the Resplendid Quetzal (Pharomachrus mocinno), which is the national bird of Guatemala and the spirit bird/companion of Tecún Umán; a Guatemalan legend.
Diagnosis: Lecanosticta pharomachri is distinguished from the other taxa in the genus by all five gene regions investigated but especially by sequences of TEF1, MS204 or RPB2. Conidia are also larger than those of L. guatemalensis and similar to L. acicola but differ from these species in that the conidia are frequently surrounded by a thick mucoid sheath and are mostly straight.   Fig. 2 Maximum likelihood tree representing the five known and four novel species of Lecanosticta generated from the TEF1 region. MP bootstrap support (> 70%) are indicated first, followed by ML bootstrap values (MP/ML, * = insignificant value). Bold branches indicate BI values > than 0.95. Dothistroma species were used as the outgroup taxa. All represented type species are indicated in bold and with a "T". Clades indicated on the left correspond with the clades in Fig. 1. Within the L. jani clade a "Δ" next to the isolate indicates that the isolate either exhibits Type 2 morphology and groups with Subclade 1, or, exhibits Type 1 morphology and groups with Subclade 2 van der Nest et al. IMA Fungus (2019) 10:2 well as conidial budding -secondary conidia sometimes produced from apical cell, 0-2-septate.
Culture characteristics: Colonies flat to erumpent, form irregular with undulate edge, spreading with fluffy aerial mycelium at centers. On MEA, surface apricot to cinnamon with isabelline and rosy buff mycelial mat at centers, reverse isabelline to dark brick in centre with cinnamon to apricot edges. On PDA, surfaces rosy to pale vinaceous with light isabelline to greenish white edges, reverse isabelline with cream edges. On OA, surface dirty white to isabelline to dark brown, fluffy mycelium to flat growth. Growth characteristics: optimal growth temperature 20°C, after 4 wk., colonies at 10, 15, 20, 25, and 30°C reaching a maximum of 9, 17, 18.5, 18.5 and 8.5 mm diam, with mean growth rates of 1.9, 3.6, 4.6, 4.4, and 1.9 mm / wk. respectively.
Notes: Some of the isolates, including the ex-type strain, produced a luteus exudate that diffused into MEA after 4-6 wk. Conjugation tubes were reported previously in L. acicola cultures as well as in needles (Siggers 1950;Crosby 1966). Conjugation tubes were also observed in this species (Fig. 4g) in the present study. Endospores as described by Crosby (1966) were also observed in some conidia.  Culture characteristics: Colonies somewhat erumpent, spreading with flat to fluffy aerial mycelium. On MEA, surface olivaceous to isabelline with rosy buff mycelial tufts, reverse isabelline. On PDA, surface rosy vinaceous to peach in centre with a dark brown edge, isabelline in reverse. On OA, surface dirty white to pale vinaceous, fluffy mycelia to flat peach growth. Growth characteristics: optimal growth temperature 25°C, after 4 wk., colonies at 10, 15, 20, 25, and 30°C reached maximum of 9, 15.5, 24, 24, and 4.5 mm, with mean growth of 2.2, 3.8, 5.3, 5.7, and 1.1 mm / wk. respectively. MycoBank MB 826878. (Fig. 6) Etymology: The epithet refers to the variable size and shape of the conidia.
Diagnosis: Lecanosticta variabilis is distinguished from the closely related species, L. acicola, by either ITS, TEF1 or MS204. Morphologically, it is distinguished from other species with the exception of L. acicola by the diffusion of sulphur-yellow to cinnamon metabolite into PDA and a luteus to sienna coloured metabolite Description: Sexual state not observed. Conidiomata olivaceous to vinaceous brown on MEA. Conidiophores cylindrical, extending in densely aggregated palisade, hyaline to honey to pale vinaceous brown, smooth to verruculose, unbranched or branched at base, septate or aseptate, often encased in granular yellow to light brown mucoid sheath, length up to 60 μm, 2.0-5.0 μm diam. Conidiogenous cells terminal, integrated, subcylindrical to cylindrical, hyaline to light brown, smooth to verruculose, proliferating several times percurrently with visible annelations near apex, septate or aseptate, (4.5-)5.5-10.5(− 12.0) × (1.5-)2.0-3.5(− 5.0) μm. Conidia three different conidial types. All three types solitary, smooth to verruculose, subhyaline to honey to light brown, often enclosed in granular light luteus mucoid sheath. Type 1 straight to strongly curved, subcylindrical to cylindrical, subobtusely rounded apex, truncate, 1-4-septate, base ( Culture characteristics: Colonies flat to somewhat erumpent, spreading, with sparse aerial mycelium, surface folded, with smooth, lobate margins. On MEA, surface isabelline with patches of pale luteus to dark olivaceous green, reverse olivaceous to fuscous black. Mucoid yellow to peach to yellow-green exudate present. Luteus to sienna coloured metabolite diffusing into medium. On PDA, surface isabelline in centre, rosy buff in outer region, dark olivacous-brown on edges and isabelline in reverse. Sulphur yellow to cinnamon coloured metabolite diffuses into media. On OA, surface dirty white with diffuse umber outer region. Growth characteristics: optimal growth temperature 25°C, after 4 wk., colonies at 10, 15, 20, 25 and 30°C reached maximum of 11.5, 21, 31, 31.5 and 22.5 mm,