Citation:
Parker, C.W.; Teixeira,
M.d.M.; Singh, N.K.; Raja, H.A.;
Cank, K.B.; Spigolon, G.; Oberlies,
N.H.; Barker, B.M.; Stajich, J.E.;
Mason, C.E.; et al. Genomic
Characterization of
Parengyodontium
torokii
sp. nov., a Biofilm-Forming
Fungus Isolated from Mars 2020
Assembly Facility.
J. Fungi
2022
,
8
, 66.
https://doi.org/10.3390/jof8010066
Academic Editors: Samantha
C. Karunarathna and
Saowaluck Tibpromma
Received: 19 November 2021
Accepted: 20 December 2021
Published: 9 January 2022
Publisher’s Note:
MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright:
© 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Fungi
Journal of
Article
Genomic Characterization of
Parengyodontium torokii
sp. nov.,
a Biofilm-Forming Fungus Isolated from Mars 2020
Assembly Facility
Ceth W. Parker
1, †
, Marcus de Melo Teixeira
2,3, †
, Nitin K. Singh
1
, Huzefa A. Raja
4
, Kristof B. Cank
4
,
Giada Spigolon
5
, Nicholas H. Oberlies
4
, Bridget M. Barker
2
, Jason E. Stajich
6
, Christopher E. Mason
7
and Kasthuri Venkateswaran
1,
*
1
Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USA;
ceth.w.parker@jpl.nasa.gov (C.W.P.); nitin.k.singh@jpl.nasa.gov (N.K.S.)
2
Pathogen and Microbiome Institute, Northern Arizona University, Flagstaff, AZ 86011, USA;
marcus.teixeira@gmail.com (M.d.M.T.); Bridget.Barker@nau.edu (B.M.B.)
3
School of Medicine, University of Brasilia, Brasilia 70910-900, Brazil
4
Department of Chemistry and Biochemistry, University of North Carolina at Greensboro,
Greensboro, NC 27412, USA; haraja@uncg.edu (H.A.R.); k_cank@uncg.edu (K.B.C.);
n_oberli@uncg.edu (N.H.O.)
5
Biological Imaging Facility, California Institute of Technology, Pasadena, CA 91125, USA; giadas@caltech.edu
6
Department of Microbiology and Plant Pathology, University of California—Riverside,
Riverside, CA 92521, USA; jason.stajich@ucr.edu
7
WorldQuant Initiative for Quantitative Prediction, Weill Cornell Medicine, New York, NY 10065, USA;
christopher.e.mason@gmail.com
*
Correspondence: kjvenkat@jpl.nasa.gov; Tel.: +1-(818)-393-1481; Fax: +1-(818)-393-4176
†
These authors contributed equally to this work.
Abstract:
A fungal strain (FJII-L10-SW-P1) was isolated from the Mars 2020 spacecraft assembly
facility and exhibited biofilm formation on spacecraft-qualified Teflon surfaces. The reconstruction
of a six-loci gene tree (ITS, LSU, SSU,
RPB1
and
RPB2
, and
TEF1
) using multi-locus sequence
typing (MLST) analyses of the strain FJII-L10-SW-P1 supported a close relationship to other known
Parengyodontium album
subclade 3 isolates while being phylogenetically distinct from subclade
1 strains. The zig-zag rachides morphology of the conidiogenous cells and spindle-shaped conidia
were the distinct morphological characteristics of the
P. album
subclade 3 strains. The MLST data
and morphological analysis supported the conclusion that the
P. album
subclade 3 strains could be
classified as a new species of the genus
Parengyodontium
and placed in the family Cordycipitaceae.
The name
Parengyodontium torokii
sp. nov. is proposed to accommodate the strain, with FJII-L10-SW-
P1 as the holotype. The genome of the FJII-L10-SW-P1 strain was sequenced, annotated, and the
secondary metabolite clusters were identified. Genes predicted to be responsible for biofilm formation
and adhesion to surfaces were identified. Homology-based assignment of gene ontologies to the
predicted proteome of
P. torokii
revealed the presence of gene clusters responsible for synthesizing
several metabolic compounds, including a cytochalasin that was also verified using traditional
metabolomic analysis.
Keywords:
biofilm; fungi; genomics; mars 2020 mission; metabolomics; morphological analysis;
phylogenetic analysis
1. Introduction
NASA microbial burden assessment of the spacecraft-associated surfaces is biased
toward detecting endospore-forming bacteria as a primary planetary protection (PP) con-
cern [
1
,
2
]. The extreme hardiness of bacterial endospores allows them to tolerate inhos-
pitable conditions for long periods, making them particularly good candidates for surviving
J. Fungi
2022
,
8
, 66. https://doi.org/10.3390/jof8010066
https://www.mdpi.com/journal/jof
J. Fungi
2022
,
8
, 66
2 of 27
the journey to planetary bodies that may support life [
3
]. However, while fungal species
also produce protective structures (spores, conidia, or cysts) as both part of their life cycle
and as a response to environmental stress, few studies have examined their presence on
the spacecraft-associated surfaces or their survival under simulated space conditions [
4
,
5
].
As a result, several reports on the description of novel bacterial species associated with
spacecraft environments were published [
6
]. Still, systematic characterizations of fungal
strains associated with spacecraft environments for their phylogenetic novelty are yet to be
conducted.
In an ongoing microbial surveillance study of NASA Mars 2020 mission-associated
spacecraft assembly environments, a novel fungal strain (FJII-L10-SW-P1) belonging to
the genus
Parengyodontium
was isolated. The internal transcribed spacer (ITS) region-
based phylogenetic analysis demonstrated that the Mars 2020 strain (FJII-L10-SW-P1)
and three other isolates (LEC01, CBS 368.72, and UAMH 9836) were affiliated with the
Parengyodontium album
subclade 3. Subsequent whole-genome sequencing (WGS) analysis
revealed that the strain FJII-L10-SW-P1 exhibited a strong phylogenetic relationship with
the strain LEC01 that was isolated from a hydrocarbon gas turbine fuel sample, which
was misidentified as
Lecanicillium
sp. using 18S rRNA gene phylogeny [
7
].
P. album
strains
were also isolated from a variety of ecosystems, including marine sediments [
8
], plant
materials [9], soil [10], and walls/paintings [11,12].
The taxonomy of
Parengyodontium
is complex, as its members were originally assigned
to the genus
Beauveria
[
13
], then to
Tritirachium
[
14
], and as
Engyodotium
[
15
]. Finally,
phylogenetic analyses targeting the ITS region, 28S nuclear ribosomal DNA, and
β
-tubulin
gene as well as matrix-assisted laser desorption ionization–time of flight mass spectrometry
(MALDI-TOF-MS) profiles, resulted in transferring members of the
Engyodotium
species
to a novel genus,
Parengyodontium
, within the family Cordycipitaceae [
16
]. At the time
of writing, the genus
Parengyodontium
consists of
P. album
[
16
] and
P. americanum
[
17
].
The ITS-based phylogenetic analysis and MALDI-TOF profiles of several
P. album
strains
displayed three distinct subclades, whereas the 28S rDNA-based phylogeny could not
separate subclades 1 and 2 [
16
]. The cryptic species associated with subclades 1 and 2 need
more study, but during this study, the WGS-based phylogeny and multi-locus sequence
type (MLST) analyses revealed that strains belonging to subclade 3 should be classified as
a novel species of the genus
Parengyodontium
.
The formation of microbial biofilms on surfaces, with consequent biofouling/bio-
corrosion of space hardware and life-support systems, is a significant concern to NASA
and will also be of interest to commercial companies. In addition, the biofilm-suppressing
materials will be helpful to several industries, including the health, medical instruments, oil,
and water pipe industries. Hence, one of the objectives of this study is to isolate fungi from
Mars 2020 assembly facility cleanroom and study on their biofilm formation by the Mars
2020 strain (FJII-L10-SW-P1) on space-qualified material surfaces. Although it is extremely
unlikely to find significant biomass (much less biofilms) on flight hardware associated with
robotic exploration, many of the materials used to fabricate robotic systems are also used in
the construction of crewed spaceflight hardware. Thus, an attempt was made to understand
whether a commercially available antimicrobial compound coated on metal surfaces or
Teflon materials could resist biofilm formation by this novel fungal strain. The second
objective of this study is to define the phylogenetic placement of the NASA Mars 2020 strain
(FJII-L10-SW-P1) using microscopy and taxonomic affiliation based on MLST analyses,
including six-loci (ITS, LSU, SSU,
RPB1
and
RPB2
, and
TEF1
) [
18
]. The third objective
is to compare the WGS of FJII-L10-SW-P1 strain with closely related species and other
cordycipitaceous fungi, and annotate the genomes using various bioinformatics pipelines,
which might aid in the identification of genetic determinants related to, for example, biofilm
formation and survival in harsh extraterrestrial environments. Furthermore, we predicted
a wide range of secondary metabolite clusters from the genomes that are biotechnologically
relevant, and fungal metabolites were also confirmed with metabolomics approaches that
are common to the field of natural products research [19,20].
J. Fungi
2022
,
8
, 66
3 of 27
2. Materials and Methods
2.1. Sample Collection
In general, NASA cleanroom facilities are maintained with cleaning regimens at fre-
quencies appropriate to the current level of activity in each cleanroom. For example, during
the sampling at the Jet Propulsion Laboratory (JPL) spacecraft assembly facility (SAF) on 25
September 2018, a significant amount of assembly activity of critical Mars 2020 spacecraft
hardware was present in the cleanroom; therefore, JPL-SAF was cleaned daily, including
vacuuming and mopping with a cleaning solution (Kleenol 30, Cleancraft Industries, Inc.,
Commerce, CA, USA). Daily cleaning regimens include replacing tacky mats, wiping
surfaces, and vacuuming/mopping floors using cleanroom-certified sanitizing agents (dis-
infectants, alcohol, or ultrapure water). All personnel who enter these cleanrooms must
follow good manufacturing practice procedures to minimize the influx of particulate matter.
Specific entry procedures vary depending on the certification level of the cleanroom and
the presence or absence of mission hardware. General precautions include the donning of
cleanroom-certified garments to minimize exposure of skin, hair, and the regular clothing
of technicians and engineers. In addition, the use of cosmetics, fragrances, body spray, and
hair gels were prohibited before entry into the cleanroom. Additionally, the air supplied to
both facilities was filtered through high-efficiency particle arrestance (HEPA) filters.
Samples were collected from the cleanroom with 12 inch
×
12 inch premoistened
polyester wipes (Sterile TexTra10 TX3225; Texwipe, Kernersville, NC, USA) from 10 different
locations (1 m
2
each) as previously reported [
21
]. After sampling, polyester wipes were
placed in a 500 mL bottle containing 200 mL of sterile PBS and vigorously shaken for one
minute to dislodge microbial cells. These environmental samples were then concentrated
using a CP-150 InnovaPrep concentrating pipette (Innova Prep LLC, Drexel, MO, USA) to a
final volume of ~6 mL [22].
2.2. Isolation of Fungi
Most fungal species resist chloramphenicol, and hence it is used to suppress bacterial
proliferation and allow for the isolation of fungi. Therefore, aliquots of the concentrated
samples as mentioned above were treated with chloramphenicol (100
μ
g/mL) and incu-
bated overnight at 25
◦
C. After 18 to 24 h of incubation, both chloramphenicol-treated and
untreated samples were processed for the isolation of fungal species [
23
]. Subsequently,
samples enriched in chloramphenicol for overnight incubation were 10-fold diluted, and
100
μ
L was added in duplicate to potato dextrose agar (PDA, Difco, Thermo Fisher Sci-
entific, Irwindale, CA, USA) containing chloramphenicol (25 mg/L) and grown at room
temperature (~25
◦
C). After 7 days, 75 of the colonies that grew on PDA were collected and
stored as stab cultures and glycerol stocks for further analysis.
2.3. Morphological Analysis
For phenotypic/morphological characterization, the fungal strain was transferred to
PDA and oatmeal agar (OMA, Difco), incubated at room temperature at 23
◦
C; colony size
(in mm), structure, pigmentation, and characteristics were recorded after 21 days. PDA was
utilized to determine microscopic traits, and cultures were allowed to grow for 7–9 days.
The slide culture technique [
24
] was utilized to observe the microscopic morphology of the
fungal strain. Briefly, a small block of agar was placed in the center of a sterile slide, and
all four sides of the agar were inoculated with the fungus. Subsequently, a sterile cover
slip is gently placed on the top of the block. The slide was kept in a moist chamber, made
of a Petri dish lined with filter paper soaked in sterile water. After 3–4 days, the fungus
grew out on the coverslip as well as the slide. The cover slip was gently picked up with
sterile forceps and placed on a clean slide with a drop of water for observing details of
conidiophores, conidia, and other microscopic structures, such as the width of hyphae.
Photomicrographs were captured using phase and Nomarski contrast on an Olympus
BX53 microscope with Olympus DP25 camera and Olympus cellSens software Version 1.7.
J. Fungi
2022
,
8
, 66
4 of 27
Measurements of micromorphological characters were made with the Olympus cellSens
software. Photographs of the colonies were taken with a Canon Power shot SD1300 IS.
2.4. Scanning Electron Microscopy
Following fresh fungal sample collection, cells were immersed in chilled 2.5% glu-
taraldehyde (Ted Pella Inc.; Redding, CA, USA) in 0.1 M sodium cacodylate buffer (Sigma–
Aldrich, St. Louis, MO, USA) and incubated at 4
◦
C for 1 h before being washed 3 times in
0.1 M sodium cacodylate buffer. Cells then underwent isopropyl alcohol (IPA) dehydration
via a series of incremental IPA steps from 50% to 100% (50%, 70%, 80%, 90%, 95%, and
3 times 100%) and stored at 4
◦
C in 100% IPA. Samples were critically point dried in an
Automegasamdri 915B critical point dryer (Tousimis, Rockville, MD, USA). Samples were
attached to scanning electron microscopy (SEM) stubs with carbon tape (Ted Pella Inc., Red-
ding, CA, USA), followed by carbon coating with a Leica EM ACE600 Carbon Evaporator
(Leica, Wetzlar, Germany) to a thickness of ~12 nm. SEM analysis was performed with an
FEI Quanta 200F scanning electron microscope (Thermo Fisher, Waltham, MA, USA).
2.5. Biofilm Formation
Commercially available and patented organosilane, a surface-penetrating compound,
was used as an antimicrobial coating on the tested surfaces. The active ingredient in the
antimicrobial compound tested is 3-(trihydroxysilyl) propyldimethyloctadecyl ammonium
chloride and previously assessed to control bacterial biofilm formation [
25
]. In the presence
of hydroxyl groups at the surface of the glass, minerals, or metals (e.g., aluminum, steel),
silanols formed a stable Si bond. This chemistry allowed silanes to function as valuable
surface-treating/protecting or coupling agents. In addition, the antimicrobial compound
breaks down the interfacial tension on surfaces, permitting the active ingredient to cova-
lently bond (non-polar covalent bond) more quickly and evenly, resulting in better efficacy
and protection. The spacecraft qualified materials tested during this study were purchased
and precision cleaned at the JPL following NASA standard practices developed for cleaning
spacecraft components, as previously described [
26
]. This cleaning step assured sterility, as
no microorganisms were grown after cleaning when test coupons were placed in sterile
nutrient media [
26
]. Once precision cleaned, the antimicrobial compound was applied and
used.
Suitable quantities (10
6
conidia) of conidial suspension were added to 10 mL of potato
dextrose broth (PDB, Difco) in 50 mL polypropylene conical vials. Sterile Centers for Disease
Control and Prevention (CDC) circular disc coupons (12.5 mm diameter
×
1 mm thickness;
BioSurface Technologies Corporation, Bozeman, MT, USA) composed of Inconel (aerospace
nickel alloy) and Teflon were added separately to vials to serve as biofilm substrate, along
with uninoculated controls. In addition, treated Teflon coupons coated in an antimicrobial
coating were also added to separate vials to serve as substrates to test biofilm mitigation.
All experiments were carried out in triplicate but SEM studies were conducted from a
single coupon. Vials containing Inconel or Teflon were incubated at 25
◦
C in an orbital
shaker at 25 RPM for at least 21 days prior to harvesting and analysis. Additionally, 10 mL
of PDB was added to microscope compatible Nunc Glass Bottom Dishes (150680, Thermo
Fisher Scientific), inoculated with fungal suspension, and incubated for 21 days. After
suitable incubation periods, the coupons were tested for biofilm formation using confocal
microscopic analyses.
2.6. Confocal Microscopy
Biofilm samples were analyzed via confocal microscopy. CDC coupons were removed
from growth vials and immediately submerged into 4% paraformaldehyde (PFA; Sigma–
Aldrich) in 1
×
phosphate-buffered saline solution (PBS; Sigma–Aldrich) in a 24-well COTS
plate. Coupons were then rinsed in PBS 3 times to ensure the removal of PFA. The cell walls
of the samples were first stained with 15
μ
M Calcofluor White (Sigma–Aldrich) for 1 h at 37
◦
C, followed by rinsing with deionized-H
2
O 3 times. Samples were then stained with 1
μ
M
J. Fungi
2022
,
8
, 66
5 of 27
TO-PRO-3 (Sigma–Aldrich) for nucleus staining at room temperature for 30 min, followed
by rinsing with deionized-H
2
O for 3 times. Samples were then stored at 4
◦
C. Samples
were protected from light throughout the staining process by wrapping the 24-well COTS
plate with aluminum foil.
Confocal imaging was performed on ZEISS LSM 980 with Airyscan 2 (Zeiss, Jena,
Germany) at the California Institute of Technology Biologic Imaging Facility using a Zeiss
63X Plan-Apochromat 1.4 NA (Zeiss). Image processing and analysis were performed
with the Imaris software package (Version 9.01.05). Surface filling was used to generate
a composite of TO-PRO-3 and Calcofluor White model. Due to the auto-fluorescence of
both the background Teflon surface and the coverslip, both surfaces were cropped out of
the final confocal image to provide higher clarity and more accurate analysis. The biofilm
density was quantified by determining the number of voxels per Z slice that contained
the surface-filling model. We used these data to determine the median height of each
biofilm. The proportionate biomass within each biofilm was calculated by multiplying the
number of voxels containing surface-filled biofilm by the voxel size (0.132
μ
m L
×
0.132
μ
m
W
×
1.25
μ
m H).
2.7. ITS-Based Fungal Identification
Among the 75 fungal isolates, 12 strains were novel based on ITS-sequence analyses;
however, we have performed detailed phylogenetic analyses only for the FJII-L10-SW-P1
strain due to its capability in forming biofilm on spacecraft qualified materials. DNA was
extracted from all fungal strains (n = 75 isolates) using the Maxwell-16 MDx automated
system following the manufacturer’s instructions (Promega, Madison, WI, USA). Initial
identification of the fungus was performed by amplicon sequencing targeting the internal
transcribed spacers (ITS) region using primers ITS 1F (5
′
-CTT GGT CAT TTA GAG GAA
GTA A-3
′
) [
27
] and Tw13 (5
′
-GGT CCG TGT TTC AAG ACG-3
′
) [
28
]. PCR conditions and
sample preparation steps for sequencing were performed as described elsewhere [29].
The ITS sequences were characterized through the Basic Local Alignment Search Tool
(BLAST) algorithm [
30
] using the National Center for Biotechnology Information (NCBI)
and UNITE database(s) to find the type strains with the closest percent similarity to the
fungal strain. Sequences from all taxa were obtained from the two previous taxonomic
studies on the genus
Parengyodontium
[
16
,
17
] and sequence data for other closely related
taxa from the family Cordycipitaceae were downloaded from NCBI. The ITS dataset
comprising 24 sequences was used for an initial phylogenetic analysis, among which
17 sequences belonged to the genus
Parengyodontium
. The sequences were aligned using
ClustalW followed by generation of the Maximum-Likelihood (ML) tree using MEGA
7.0.26 [31]. One thousand bootstraps were performed to test branch fidelity.
2.8. MLST-Based Phylogenetic Analyses
Two different MLST analyses were utilized due to a lack of resolution from the ITS
region alone to resolve the phylogenetic affiliations of certain fungi and are detailed below.
Since the number of genetic markers are uneven among Cordycipitaceae taxa (especially
beta-tubulin), we decided to split the MLST scheme into two concatenated datasets, as
follows:
(a)
Three-gene MLST analyses. Sequences from ITS, 28S nrDNA, and
β
-tubulin genes
were used in a dataset comprised of 22 fungi, including the outgroup. The out-
group selection was based on [
16
]. Multiple sequence alignments were generated
using MAFFT default settings using PhyloSuite v.1.2.1 [
32
]. The alignments were
trimmed to remove ambiguous characters using GBlocks [
33
,
34
]. For the concate-
nated dataset, PartitionFinder 2 [
35
] was used to select the best-fit model according to
the Akaike Information Criterion corrected (AICc) [
36
]. The best-fitting substitution
models according to AICc were: ITS and
β
-tubulin: GTR+I+G and LSU: TRN+I. Mod-
elFinder [
37
] was used for the ITS dataset to select the best-fit model using the AICc
criterion. The best-fit model according to AICc was TIM2+F+R2. The trimmed align-
J. Fungi
2022
,
8
, 66
6 of 27
ment was then used to construct a ML tree using IQ-TREE implemented in PhyloSuite.
Ultrafast bootstrapping was done with 5000 replicates [
38
]. Nodes with UFBoot
≥
90%
are shown on the clades, but only nodes
≥
95% were considered strongly supported.
Bayesian inference phylogenies were inferred using MrBayes 3.2.6 [
39
] under partition
model (2 parallel runs, 10 million generations), using PhyloSuite v. 2.1. Four inde-
pendent chains of Metropolis-coupled MCMC were run for 10 million generations
with trees sampled every 1000th generation, resulting in 10,000 trees. The first 25% of
the trees were discarded as a burn-in parameter. The average standard deviation of
split frequencies value approaching 0.001 was used to estimate that the two runs had
converged closer to the stationary phase (10 million generations). Consensus trees
were generated and viewed in PAUP* v.4.0a (build 166) [
40
]. Clades with a posterior
probability (PP)
≥
95% were considered significant and strongly supported.
(b)
Six-loci MLST analyses. Gene sequences utilized were: ITS region rRNA gene, D1/D2
domain of large subunit (LSU or 26S) rRNA gene, small subunit (SSU or 18S) rRNA
gene, and housekeeping genes including two subunits of RNA polymerase II (
RPB1
and
RPB2
) and the translation elongation factor 1-
α
(
TEF1
). These six-loci have al-
ready been established for differentiating Cordycipitaceae species [
41
]. Sequences of
58 fungal strains available were downloaded, and sequences were manually concate-
nated (representative sequences are available at [
41
]. The respective gene sequences
that were available on NCBI for different
Parengyodontium
species (n = 8 isolates) were
included in the phylogenetic analysis except for the Mars 2020 strain (FJII-L10-SW-P1),
which was generated during this study. For MLST, sequences were aligned using
MAFFT v7 [
42
], concatenated manually, trimmed using the ClipKit tool, smart-gap
function [
43
] and a ML Tree was generated using the using IQTREE2 v2.0.6 [
31
,
44
].
The best substitution model was calculated using the ModelFinder algorithm [
37
] and
1000 ultrafast bootstraps [
45
] and SH-like approximate likelihood ratio test (aLRT)
were used to test branch support [
46
]. Finally, the trees were visualized using the
FigTree v 1.4.4 software (http://tree.bio.ed.ac.uk/software/figtree/, accessed on 12
November 2021).
2.9. Whole-Genome Sequencing Analyses
A pure and well-isolated colony was picked after streaking onto PDA plates, and
approximately 1-g wet weight mycelia were collected for DNA extraction. DNA extraction,
WGS, and processing were followed, as described elsewhere [
47
]. Briefly, the total nucleic
acid extraction was carried out using ZymoBIOMICS 96 MagBead DNA kit (Lysis tubes)
(Zymo Research, Irvine, CA, USA) after bead beating with Precellys homogenizer (Bertin,
Rockville, MD, USA). This was followed by library preparation using the Illumina Nextera
Flex Protocol as per Illumina document number 1000000025416 v07. The initial amount
of DNA for library preparation was quantified, and 5 to 12 cycles of PCR were carried
out to normalize the output depending on the input DNA concentration. The amplified
genomic DNA fragments were indexed and pooled in a 384-plex configuration with dual-
index adapters. Whole-genome shotgun sequencing was performed on a NovaSeq 6000
S4 flowcell PE 2
×
150 platform with a paired-end module. The data was filtered with
NGS QC Toolkit v2.3 [
48
] for high-quality (HQ) vector and adaptor-free reads for genome
assembly (cutoff read length for HQ, 80%; cutoff quality score, 20). The number of filtered
reads obtained were used for assembly with SPAdes 3.14.0 [
49
] genome assembler (k-mer
size: 32 to 72 bases) using default parameters. The resulting assembly was curated next
using the AAFTF pipeline [
50
], as follows: (1) Mitochondrial and contaminant contigs were
identified and removed with the “vecscreen” and “sourpurge” functions; (2) the “rmdup”
function was used to remove duplicated contigs identified by the minimap2 [
51
] algorithm;
(3) the final assembly was polished using the “pilon” function in order to correct bases, fix
misassembled contigs and fill potential gaps [
52
]; (5) the contigs smaller than 1000 bp were
purged and sorted by size using the “sort” function; (6) finally, the assembly statistics were
obtained using the “assess” tool. The genome completeness of the FJII-L10-SW-P1 assembly
J. Fungi
2022
,
8
, 66
7 of 27
was evaluated using the BUSCO v5.1.2, sordariomycetes_odb10 library [
53
]. Genomic DNA
sequences of all available strains used in this study were downloaded from NCBI, and
comparative analyses were performed (Table S1).
2.10. De Novo and Functional Genome Annotation
The resulting assembly of the FJII-L10-SW-P1 strain was annotated using the funan-
notate v 1.8 pipeline, called the “annotate” function [
54
]. We re-annotated the genomes of
other Cordycipitaceae fungi (Table 1 and Table S1) in order to avoid annotation bias for
comparative genomic analysis. Initially, repetitive sequences were soft-masked using the
tantan algorithm [
55
]. We generated and curated the library of repeats identified de novo by
RepeatModeler v2.0.1 [
56
] combined with well-characterized transposable elements from
RepBase [
57
] for fungi. The genome was masked with RepeatMasker v4.0.7 [
58
] to identify
these interspersed repeats based on the library and low complexity DNA sequences. Gene
prediction using ab initio gene predictors was performed using the funannotate “predict”
function on the masked genome; GeneMark-ES v4.62 was used to predict genes using the
self-training algorithm [
59
] while AUGUSTUS v3.3.3 [
60
], Glimmerhmm v3.0.4 [
61
], and
SNAP v 0.15.4 [
62
] predicted the gene structures based on a combination of high-fidelity
transcripts of
Beauveria bassiana
ARSEF 2860 and
Cordyceps militaris
CM01 as well as using
the BUSCO lineage-specific sordariomycetes_odb10 library [
53
]. The consensual gene pre-
diction was achieved using the EVidence Modeler (EVM v1.1.1), where a weight of 2 was
attributed to high quality AUGUSTUS predictions, while in the other ab initio predictions,
the weight was set to 1 [
63
]. Gene models with less than 50aa, spanning gaps, or containing
transposable elements were removed. The tRNAs were identified using tRNAscan-SE v
1.3.1 [
64
]. Genome annotation was performed using the “annotate” function of funannotate.
This was achieved with searches of the predicted proteins by diamond blastp [
65
] to the
UniProt DB version 2021-03. Further alignments to Pfam v31.0 [
66
] and InterPro5 [
67
]
domains, carbohydrate-active enzymes (CAZymes) [
68
], secreted proteins [
69
], transmem-
brane proteins [
70
], proteases (MEROPS) [
71
], fungal transcription factors [
67
], and BUSCO
groups [
53
] were found. The GO terms were assigned based on matches to the InterPro
database searches. Secondary metabolite gene clusters were identified with antiSMASH
v6.0.0 [
72
]. Matches to Eggnog and the cluster of orthologous groups of proteins (COGs)
orthologous groups were further added to the functional annotation. The functional de-
scriptions from UniProtKB/SwissProt best matches at 80% alignment and 80% identity
were combined with descriptions from Eggnog-mapper searches to generate gene names
and product descriptions. The mitogenome was annotated using the RNAweasel and
MFannot pipelines (https://github.com/BFL-lab/Mfannot, accessed on 12 November
2021). Lastly, the annotations collected from each genome were converted into the GenBank
flat-file format (gbk) for comparative genomic analysis, and the generated.tbl and. sqn files
were submitted to NCBI Genomes.
J. Fungi
2022
,
8
, 66
8 of 27
Table 1.
Summary of the draft whole-genome sequences of
Parengyodontium
and closely related
species belonging to the family Cordycipitaceae.
Parengyod-
ontium
toroki
FJII-
L10-SW-P1
Parengyod-
ontium
toroki
LEC01
Parengyod-
ontium
americanum
AZ2
Akanthomyces
lecanii
RCEF 1005
Simplicillium
aogashi-
maense
72-15 1
Lecanicillium
fungicola
150-1
Lecanicillium
psalliotae
HWLR35
Beauveria
bassiana
ARSEF 2860
Samsoniella
hepiali
FENG
Assembly
# of contigs
440
352
295
131
20
782
197
239
222
Genome size
30,424,506
31,084,693
32,962,623
35,580,375
29,244,117
44,547,425
36,133,949
33,693,821
34,650,604
Largest
contig
718,708
1,200,404
821,300
5,461,016
4,930,463
493,648
4,365,396
2,084,429
1,229,925
Repetitive
DNA (%)
2.83%
7.31%
9.98%
11.29%
7.38%
8.44%
11.34%
11.78%
11.91%
GC (%)
50.45
50.28
52.88
53.10
49.01
49.87
52.73
51.36
53.89
N50
122,374
310,369
345,622
3,613,853
3,162,613
154,124
2,330,369
724,305
576,310
L50
70
27
31
4
4
92
6
13
20
Annotation
tRNA
70
77
95
115
85
144
121
111
115
intron
16,303
16,562
16,709
14,195
15,835
19,694
13,365
15,912
13,429
Exons
25,899
26,363
27,008
24,306
25,962
32,964
23,566
25,710
23,649
average
exon length
478
482
472
494
483
476
502
476
518
mRNA
9596
9801
10,299
10,111
10,127
13,270
10,201
9798
10,220
CDS
9596
9801
10,299
10,111
10,127
13,270
10,201
9798
10,220
gene
9666
9878
10,394
10,226
10,212
13,414
10,322
9909
10,335
average
gene length
1658
1642
1570
1565
1579
1536
1521
1627
1560
average
protein
length
496
501
479
484
489
470
472
498
482
Functional
go_terms
2913
5980
6243
1832
3109
2750
2306
2301
2947
interproscan
3915
8073
8454
2523
4184
3879
3206
3118
4061
eggnog
9330
9455
9892
9736
9780
12,457
9749
9461
9793
pfam
6961
7187
7476
7269
7530
9092
7160
7001
7141
cazyme
355
370
399
380
459
507
403
331
364
merops
412
418
438
472
498
549
471
402
447
busco
3685
3742
3739
3661
3748
3755
3596
3747
3573
secretion
826
899
995
1060
1144
1403
1144
1008
1009
3. Comparative Analysis of Fungal Genomes
Comparative genomic analyses were performed using the “compare” function of fu-
nannotate pipeline using nine Cordycipitaceae fungi. Initially, to generate a phylogenomic
informed species tree, a set of single-copy orthologs between the nine genomes analyzed
were identified using the Proteinortho v6.0.20 [
73
]. Individual single-copy orthologs were
aligned using the MAFFT v7 pipeline [
42
] and trimmed using the ClipKit tool and smart
gap function [
43
]. Therefore, each unique alignment was submitted for ML analysis using
the IQTREE2 software [
44
]. We set the species
Simplicillium aogashimaense
72–15.1 as the
outgroup, and each individual best protein substitution model was calculated using the
ModelFinder method [
37
]. The concatenated tree and individual trees were submitted
to branch fidelity using two different approaches in IQTREE2 software: Ultrafast boot-
straps [
45
] and Gene Concordance Factors [
74
]. The phylogenomic tree was visualized
using the FigTree v1.4.4 pipeline (http://tree.bio.ed.ac.uk/software/figtree/, accessed on
12 November 2021).
The copy number counts Pfam, CAZymes, MEROPS, transmembrane proteins, se-
creted proteins, COGs, secondary metabolites, and fungal transcription factors and plotted
the categories with a standard deviation >1 in a heat map. We also compared the GO-
enriched terms for each species by taking into account those with an FDR GO-enrichment
p
-value < 0.05. To identify specific GO categories enriched for the
Parengyodontium
lineage
as well for each individual of this genus, we used the OrthoVenn2 approach [
75
]. The
predicted proteomes from each individual were submitted to OrthoVenn2 for visualization
and comparison of gene content, and the enriched GO categories with a
p
-value < 0.05
were retrieved. We also looked for the presence of genes related to biofilm formation and
adhesion, melanin biosynthesis, radioresistance, and microgravity resistance in the genome
of the strain FJII-L10-SW-P1.
J. Fungi
2022
,
8
, 66
9 of 27
Metabolomics
The cultures of fungal strain FJII-L10-SW-P1 were maintained on potato dextrose agar
(PDA; Difco). An agar plug from the leading edge of the PDA culture was transferred to
a sterile tube with 10 mL of liquid PDA. The culture was grown for 12 days on an orbital
shaker (100 rpm) at room temperature (rt; ~23
◦
C) and then used to inoculate the solid
fermentation media, as described below. Solid-state fermentations were carried out in
250-mL Erlenmeyer flasks. To prepare the two media, 10 g of rice and oatmeal were added
to separate flasks (two rice and two oatmeal flasks) with 50 mL of deionized water. After
autoclaving these samples at 120
◦
C for 20 min, the flasks were inoculated with the fungal
culture (10 mL of 7 day grown cultures) to be tested and incubated at room temperature for
2 weeks. Subsequently, each of the two solid-state fermentation cultures of FJII-L10-SW-P1
were chopped up into small pieces using a spatula, and 60 mL of 1:1 MeOH-CHCl
3
were
added. The fungal cultures were then shaken using a rotary shaker overnight (~16 h) at
~125 rpm at room temperature. The cultures were filtered in vacuo and then pooled to form
a combined filtrate, and the solid residue was rinsed with a small volume of 1:1 CH
3
OH-
CHCl
3
. To the filtrate, 90 mL of CHCl
3
and 150 mL of H
2
O were added; the solution was
stirred for 20 min and transferred to a separatory funnel. The organic layers were collected
and evaporated to dryness under vacuum using a rotary evaporator. The resulting organic
layers were partitioned between 100 mL of 1:1 CH
3
OH-CH
3
CN and 100 mL of hexane. The
CH
3
OH-CH
3
CN layers were collected and evaporated to dryness under vacuum to yield
two organic extracts, one from rice (77.12 mg) and one from oatmeal media (253.87 mg).
High-resolution electrospray ionization mass spectrometry (HRESIMS) was performed
on a Thermo LTQ Orbitrap XL mass spectrometer (Thermo Fisher, San Jose, CA, USA)
equipped with an electrospray ionization source. Source conditions in the positive-ioni-
zation mode were set at 275
◦
C for the capillary temperature, 4.5 kV for the source voltage,
20 V for capillary voltage, and 95 V for the tube lens. Nitrogen was utilized for the sheath
gas and set to 25 and 20 arb for the positive and negative modes, respectively. For the
negative-ionization mode, nitrogen was also used as an auxiliary gas and set at 10 arb.
Scan events were carried out, with full-scan (
m
/
z
of 100–2000) and ion-trap MS/MS of the
most intense ion from the parent mass list utilizing CID with a normalized collision energy
of 30. Thermo Scientific Xcalibur 2.1 software was used for instrument control and data
analysis. Ultra-performance liquid chromatography (UPLC) was carried out on a Waters
Acquity system using a BEH C18 (2.1
×
50 mm, 1.7
μ
m) column (Waters Corp., Milford,
MA, USA) equilibrated at 40
◦
C. A mobile phase consisting of CH
3
CN–H
2
O (acidified with
0.1% formic acid) was used, starting with 15:85 then increasing linearly to 100% CH
3
CN
within 8 min, holding for 1.5 min, and then returning to the starting conditions within
0.5 min. An Acquity UPLC photodiode array detector was used to acquire PDA data, which
were collected from 200 to 500 nm with a 4 nm resolution.
4. Results
4.1. Taxonomy of the Strain FJII-L10-SW-P1
Parengyodontium torokii
, N.K. Singh and K. Venkateswaran, sp. nov.
MycoBank number
: MB841139.
Etymology
: Torokii refers to name Dr. Tamas Torok, an American mycologist conduct-
ing research on extremophiles).
Diagnosis
: Similar to
Parengyodontium album
but phylogenetically unique and mor-
phologically distinguished by its subcylindrical to ellipsoidal conidia.
Holotype:
USA: Pasadena, CA, 34.1478
◦
N, 118.1445
◦
W, JPL-SAF cleanroom floor
where the Mars 2020 mission components were assembled, 25 September 2018. Nitin K.
Singh and Kasthuri Venkateswaran, (HOLOTYPE is stored in a metabolically inactive
state as a lyophilized culture at the Northern Regional Research Laboratory [NRRL],
Agricultural Research Service, USA; ex-holotype culture, FJII-L10-SW-P1 = NRRL 64203,
conidial isolate from HOLOTYPE). GenBank accession numbers of the type strain (FJII-
J. Fungi
2022
,
8
, 66
10 of 27
L10-SW-P1):
ITS = MT704894
, Draft genome = JADQAY000000000. The genome size is
30.4 Mbp and the G + C content is 50.45 mol%.
Description
: Colonies on PDA and OMA are white, floccose, cottony, velvety, opaque,
with colorless exudates on the colony surface of PDA without diffusible pigments, and are
reverse pale yellow with or without ridges (Figure 1A,B). After three weeks incubation
at room temperature (23
◦
C), colonies on PDA have a 40 mm diameter, and colonies on
OMA have a ~50 mm diameter. Vegetative hyphae are smooth-walled, hyaline, and septate.
Conidiophores are erect, arising from hyphae at right angles, tapering to subcylindrical,
slightly swollen at base, occasionally biverticillate, and bearing one to numerous whorls of
conidiogenous cells (one to
≥
five). Basal portion of the conidiogenous cell is elongated,
tapering, 16–22
μ
m
×
1–2
μ
m, terminating in fertile zigzag-shaped rachides, and bearing
conidia (Figure 1C). Conidia are one-celled, smooth, thin-walled, hyaline, sub cylindrical to
ellipsoidal, aseptate, apiculate, 2–3
μ
m
×
1–2
μ
m, and arising from alternating points with
butt-shaped denticles on zigzag-shaped and genticulate rachides (Figure 1D). No sexual
state observed.
Ecology/Substrate/Host
: Cleanroom floor where spacecraft components are assem-
bled.
Other materials examined
: Three other strains belong to
P. album
subclade 3 CBS
368.72, UAMH 9836, and LEC01 were isolated from turbine fuel sample, Dayton, Ohio, USA
(LEC01); from fresco, Romania (CBS 368.72); and from a human bronchoscopy specimen,
Canada.
Notes:
The new species,
P. torokii
is both morphologically and phylogenetically unique
from other described members of the genus
Parengyodontium
[
16
,
17
]. Phenotypically,
P. torokii
can be readily distinguished from
P. album
based on conidial shape.
P. torokii
produces subcylindrical to ellipsoidal conidia, whereas
P. album
conidia are globose, smooth,
hyaline, oval, and apiculate [
16
].
P. torokii
differs from
P. americanum
as the former produces
terminal fertile zigzag shaped rachides, but the latter lacks them and produces conidia on
right-angled phialides or aphanophialides. In addition, the conidia of
P. torokii
differ from
those of
P. americanum
in that they are sub cylindrical to ellipsoidal vs. cylindrical to globose
in
P. americanum
[
17
]. Interestingly, the strain CBS 368.72 (subclade 3, sensu Tsang [
16
]) was
morphologically similar to
P. torokii
in conidial shape and zigzag rachides based on SEM
(Figure S1). Based on Maximum Likelihood molecular phylogenetic analyses of the ITS
region (Figure S2) as well as three loci analysis (Figure S3),
P. torokii
is a distinct species
as it occurs on a unique clade (subclade3 sensu Tsang [
16
], Figures S2 and S3). Further, in
the six-loci analysis,
P. torokii
,
P. album
, and
P. americanum
are seen as distinct clades with
moderate to significant statistical support (see below for details).
4.2. Biofilm Formation of the Strain FJII-L10-SW-P1
The SEM of vegetative cells of the FJII-L10-SW-P1 strain revealed thin membranous
white layers surrounding the conidia and were presumed to be composed of molecules
such as extracellular polymeric substances (EPS), which enabled biofilm formation
(Figure 1E and Figure S4). When the biofilm formation was characterized on three different
materials as substrate, the FJII-L10-SW-P1 strain was able to form biofilms on the Teflon
(tetrafluoroethylene) coupons as well as on the plastic (polypropylene) walls of the conical
Falcon tubes and the plastic (polystyrene) of the sides of the glass (borosilicate) bottomed
Petri dish. Confocal imaging of the Teflon coupons (Figure 2) indicated that there was
biomass present on both the uncoated and the antimicrobial-coated coupons. The uncoated
coupon had patches of high-density biofilm and regions of no fungal mycelium, while the
antimicrobial coated coupon had a much more distributed density of mycelium across the
area that was imaged. Both antimicrobials coated and uncoated coupons showed lower
amounts of biofilm at the substrate surface of the coupon (purple and blue colors, Figure 2)
and more biofilm near the top of the biofilm (orange and red colors). Quantification of
surface-filling voxels indicated that the biofilm formed on the uncoated Teflon is smaller
(1767
μ
m
3
) and the median height is closer to the surface (94
μ
m) while the biofilm formed
J. Fungi
2022
,
8
, 66
11 of 27
the coated Teflon coupon is larger (3326
μ
m
3
) and has a higher median height (107
μ
m).
The shape of both biofilms resembles canopy morphology. Biofilms were not formed on
either the glass-bottomed Petri dishes or on the Inconel coupons. Both borosilicate glass
surfaces and Inconel (a nickel-chrome superalloy) are smooth surfaces, while Inconel is
additionally resistant to corrosion and oxidation.
J. Fungi
2022
,
8
, 66
11 of 29
Figure 1. Macro and micromorphology of
Parengyodontium torokii
.
Colony surface of FJ11-L10-
SW-P1 after 21 days of incubation at room temperat
ure (23 °C) in standard 9 cm petri dishes on (
A
)
PDA media and (
B
) OMA media. (
C
) Conidia produced at each bent
point of the zigzag rachides of
the fertile conidiogenous cells. (
D
) Whorl of two conidiogenus cells with conidia attached at the
Figure 1. Macro and micromorphology of
Parengyodontium torokii
.
Colony surface of FJ11-L10-
SW-P1 after 21 days of incubation at room temperature (23
◦
C) in standard 9 cm petri dishes on
(
A
) PDA media and (
B
) OMA media. (
C
) Conidia produced at each bent point of the zigzag rachides
of the fertile conidiogenous cells. (
D
) Whorl of two conidiogenus cells with conidia attached at the
zigzag rachides. Scale Bars (
C
–
E
) = 20
μ
m. (
E
) Scanning electron microscopy images of
Parengy-
odontium torokii
from ex-type strain FJ11-L10-SW-P1 with whorl of conidiogenous cells showing
butt-shaped denticles and (
F
) subcylindrical to ellipsoidal, hyaline single-celled conidia. Scale bar for
all microscopy is 10
μ
m.
J. Fungi
2022
,
8
, 66
12 of 27
J. Fungi
2022
,
8
, 66
13 of 29
Figure 2. Confocal analys
is of biofilm formation.
Parengyodontium torokii
isolate FJII-L10-SW-P1
was grown in PDB in the presence of untreated (
A
,
C
) and treated (
B
,
D
) Teflon coupons. Each scan
was 1161 μm × 1336 μm and was taken at a representa
tive location on the coupon surface. Compo-
site confocal fungal space-filling structure is color-coded based on relative distance from the coupon
surface with blue/violet being proximal to
the surface and orange
/red being distal. (
A
,
B
) are plan
views of the region that was imaged (100 μm scale bars) while (
C
,
D
) are orthogonal views of the
biofilm looking length-wise through the biomass (100 μm scale bars). The comparative biofilm dis-
tribution as compared to distance from
the coupon surface is presented in (
E
). The biofilm formed
on the uncoated Teflon is smaller and the median
height is closer to the surface while the biofilm
from the coated Teflon coupon is larger and has
a higher median height (94 μm and 107 μm, respec-
tively).
4.3. Phylogenetic Analyses of the Strain FJII-L10-SW-P1
Currently, the Mycobank and CBS
databases documented only two
Parengyodontium
species, specifically
P. album
and
P. americanum.
The ITS sequences available on NCBI for
Parengyodontium
species including the strain FJII-L10-SW-P1 (n = 18 isolates) and other
closely related species (n = 6) were used
in the ML phylogenetic analysis with
Isaria cole-
opterora
(CBS 110.73) as an outgroup. A phylogram of the most likely tree (
−
lnL = 1622.37)
from a ML analysis of 24 sequences based on
the ITS region (568 bp) using IQ-TREE is
shown in Figure S2. Among the 18 strains of
P. album
there were 3 subclades and to con-
firm that four strains that form the
P. album
subclade 3 belong to a novel species, their
phylogenetic affiliations were analyzed. Next, a phylogram of the most likely tree (
−
lnL =
3824.45) from a 3-gene ML analysis of 22 se
quences based on the combined regions of ITS,
LSU, and
β
-tubulin gene (1435 bp) using IQ-TREE, was created (Figure S3). The 3-gene
MLST phylogram also supported the phylogenetic clusters that was noticed in the ITS-
tree, forming a separate branch for four st
rains including FJII-L10-SW-P1 isolate, which
was distinct from
P. album
subclade 1 and 2.
Subsequently, a six-gene MLST analysis was carried out by manually concatenating
ITS, LSU, SSU, RPB1, RPB2, and TEF1 gene sequences. In this analysis, in addition to the
Mars 2020 isolate FJII-L10-SW-P1, three other strains belong to
P. album
subclade 3 (CBS
368.72, UAMH 9836, and LEC01 isolates), two strains of
P. album
(HKU48 and IHEM 4198
isolates), two strains of
P. americanum
(AZ2 and CA11 isolates), one strain of
Lecanicillium
kalimantanense
BTCC-F23 and one strain of
Torrubiella wallacei
CBS 101237 were included,
and the tree was rooted with members of the genus
Simplicillium
. The 5-gene MLST anal-
ysis confirmed that the FJII-L10-SW-P1 strain
and other three strains clustered in a single
clade are distinct from
P. album
and
P. americanum
(Figure 3). Single loci phylogenetic
Figure 2. Confocal analysis of biofilm formation.
Parengyodontium torokii
isolate FJII-L10-SW-P1
was grown in PDB in the presence of untreated (
A
,
C
) and treated (
B
,
D
) Teflon coupons. Each scan
was 1161
μ
m
×
1336
μ
m and was taken at a representative location on the coupon surface. Composite
confocal fungal space-filling structure is color-coded based on relative distance from the coupon
surface with blue/violet being proximal to the surface and orange/red being distal. (
A
,
B
) are plan
views of the region that was imaged (100
μ
m scale bars) while (
C
,
D
) are orthogonal views of the
biofilm looking length-wise through the biomass (100
μ
m scale bars). The comparative biofilm
distribution as compared to distance from the coupon surface is presented in (
E
). The biofilm formed
on the uncoated Teflon is smaller and the median height is closer to the surface while the biofilm from
the coated Teflon coupon is larger and has a higher median height (94
μ
m and 107
μ
m, respectively).
4.3. Phylogenetic Analyses of the Strain FJII-L10-SW-P1
Currently, the Mycobank and CBS databases documented only two
Parengyodontium
species, specifically
P. album
and
P. americanum.
The ITS sequences available on NCBI
for
Parengyodontium
species including the strain FJII-L10-SW-P1 (n = 18 isolates) and
other closely related species (n = 6) were used in the ML phylogenetic analysis with
Isaria coleopterora
(CBS 110.73) as an outgroup. A phylogram of the most likely tree
(
−
lnL = 1622.37
) from a ML analysis of 24 sequences based on the ITS region (568 bp)
using IQ-TREE is shown in Figure S2. Among the 18 strains of
P. album
there were 3 sub-
clades and to confirm that four strains that form the
P. album
subclade 3 belong to a novel
species, their phylogenetic affiliations were analyzed. Next, a phylogram of the most likely
tree (
−
lnL = 3824.45) from a 3-gene ML analysis of 22 sequences based on the combined
regions of ITS, LSU, and
β
-tubulin gene (1435 bp) using IQ-TREE, was created (Figure S3).
The 3-gene MLST phylogram also supported the phylogenetic clusters that was noticed in
the ITS-tree, forming a separate branch for four strains including FJII-L10-SW-P1 isolate,
which was distinct from
P. album
subclade 1 and 2.
Subsequently, a six-gene MLST analysis was carried out by manually concatenating
ITS, LSU, SSU, RPB1, RPB2, and TEF1 gene sequences. In this analysis, in addition to
the Mars 2020 isolate FJII-L10-SW-P1, three other strains belong to
P. album
subclade
3 (CBS 368.72, UAMH 9836, and LEC01 isolates), two strains of
P. album
(HKU48 and
IHEM 4198 isolates), two strains of
P. americanum
(AZ2 and CA11 isolates), one strain of
Lecanicillium kalimantanense
BTCC-F23 and one strain of
Torrubiella wallacei
CBS 101237
were included, and the tree was rooted with members of the genus
Simplicillium
. The
5-gene MLST analysis confirmed that the FJII-L10-SW-P1 strain and other three strains
clustered in a single clade are distinct from
P. album
and
P. americanum
(Figure 3). Single loci
phylogenetic analyses (for example ITS; Figure S2) always placed FJII-L10-SW-P1 strain as