You are on page 1of 9

crossmark

Population Genetics of Hirsutella rhossiliensis, a Dominant Parasite of


Cyst Nematode Juveniles on a Continental Scale
Niuniu Wang,a,b Yongjie Zhang,c Xianzhi Jiang,a Chi Shu,a,b M. Imran Hamid,a Muzammil Hussain,a,b Senyu Chen,d Jianping Xu,e
Meichun Xiang,a Xingzhong Liua
State Key Laboratory of Mycology, Institute of Microbiology, Chinese Academy of Sciences, Beijing, Chinaa; University of Chinese Academy of Sciences, Beijing, Chinab;

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


School of Life Sciences, Shanxi University, Taiyuan, Chinac; Southern Research and Outreach Center, University of Minnesota, Waseca, Minnesota, USAd; Department of
Biology, McMaster University, Hamilton, ON, Canadae

ABSTRACT
Hirsutella rhossiliensis is a parasite of juvenile nematodes, effective against a diversity of plant-parasitic nematodes. Its global
distribution on various nematode hosts and its genetic variation for several geographic regions have been reported, while the
global population genetic structure and factors underlying patterns of genetic variation of H. rhossiliensis are unclear. In this
study, 87 H. rhossiliensis strains from five nematode species (Globodera sp., Criconemella xenoplax, Rotylenchus robustus, Het-
erodera schachtii, and Heterodera glycines) in Europe, the United States, and China were investigated by multilocus sequence
analyses. A total of 280 variable sites (frequency, 0.6%) at eight loci and six clustering in high accordance with geographic popu-
lations or host nematode-associated populations were identified. Although H. rhossiliensis is currently recognized as an asexual
fungus, recombination events were frequently detected. In addition, significant genetic isolation by geography and nematode
hosts was revealed. Overall, our analyses showed that recombination, geographic isolation, and nematode host adaptation have
played significant roles in the evolutionary history of H. rhossiliensis.

IMPORTANCE
H. rhossiliensis has great potential for use as a biocontrol agent to control nematodes in a sustainable manner as an endopara-
sitic fungus. Therefore, this study has important implications for the use of H. rhossiliensis as a biocontrol agent and provides
interesting insights into the biology of this species.

I t is well known that fungi possess diverse strategies to obtain


nutrients, as saprobes, pathogens (of plant and animal hosts),
and symbionts (with other microbes, plants, or animals) (1). One
analyses of two housekeeping genes, internal transcribed spacer
(ITS) of ribosomal DNA (rDNA) and the mitogen-activated pro-
tein kinase (MAPK) gene (13). Because of its global distribution
of the important living strategies is parasitoidism, where parasi- and wide host range, H. rhossiliensis provides an ideal model to
toids sterilize and/or kill their hosts before the hosts reach repro- study the effects of geography and host nematodes on fungal pop-
ductive age. Plant-parasitic nematodes are worldwide agricultural ulation genetic divergence and factors driving genetic differentia-
pests and are responsible for global agricultural losses amounting tion of fungi.
to an estimated $157 billion each year (2). Endoparasitic fungi are Investigating the evolutionary patterns and processes of nem-
predominantly soil-dwelling organisms that have the ability to atode endoparasitic fungi over their geographical and ecological
sterilize plant-parasitic nematodes (3). They produce characteris- distribution provides valuable information on the radiation be-
tic spores that adhere to, penetrate, and consume nematode bod- tween nematode endoparasitic fungi and hosts (17). While phe-
ies (4, 5). notypic plasticity and genomic heterogeneity can lead to broad
Hirsutella rhossiliensis (Ophiocordycipitaceae, Hypocreales, geographic and ecological distribution, low levels of gene flow
Ascomycota) (6) lives in the soil, and it has been mainly isolated often result in microbial speciation (18, 19). Population genetic
from five nematodes including Criconemella xenoplax (7), Het-
erodera schachtii (8), Heterodera glycines (9), Rotylenchus robustus
(10), and Globodera (11). Though a dominant parasitoid of H. Received 6 June 2016 Accepted 4 August 2016
glycines J2s (second-stage juveniles) in the United States, mainly Accepted manuscript posted online 19 August 2016
from Minnesota, H. rhossiliensis occurs less frequently in north- Citation Wang N, Zhang Y, Jiang X, Shu C, Hamid MI, Hussain M, Chen S, Xu J,
east China (12). In the agricultural fields, the plants, nematodes, Xiang M, Liu X. 2016. Population genetics of Hirsutella rhossiliensis, a dominant
parasite of cyst nematode juveniles on a continental scale. Appl Environ Microbiol
and endoparasitic fungi constitute a tight tripartite system with 82:63176325. doi:10.1128/AEM.01708-16.
the plants providing food to nematodes and the nematodes serv- Editor: A. A. Brakhage, HKI and University of Jena
ing as nitrogen sources to endoparasitic fungi. This existing natu- Address correspondence to Meichun Xiang, xiangmc@im.ac.cn.
ral interaction suggests that endoparasitic fungi could have great N.W. and Y.Z. contributed equally to this work.
potential for use as biocontrol agents to control nematodes in a C.S. met authorship criteria but was unreachable for final approval of the byline
sustainable manner (1315). and article.
To date, H. rhossiliensis has been detected in many localities in Supplemental material for this article may be found at http://dx.doi.org/10.1128
Europe, the United States, and China (11, 12). Genetic variation /AEM.01708-16.
among different isolates of H. rhossiliensis has been detected using Copyright 2016, American Society for Microbiology. All Rights Reserved.
randomly amplified polymorphic DNA (RAPD) markers (16) and

November 2016 Volume 82 Number 21 Applied and Environmental Microbiology aem.asm.org 6317
Wang et al.

TABLE 1 Summary information for the Hirsutella rhossiliensis populations analyzed in this study
Geographic Host nematode- Host nematode No. of
populationa associated populationb Geographic location species isolates Latitude (north) Longitude (east)c
NL-P NL-GL Netherlands Globodera sp. 5 52.28 to 52.78 6.59 to 6.62
USA-P USA-CX California, USA Criconemella xenoplax 5 39.02 to 39.98 (121.18 to 121.98)
USA-P USA-CX Pennsylvania, USA Criconemella xenoplax 2 41.12 to 41.35 (78.28 to 78.82)
USA-P USA-RR California, USA Rotylenchus robusta 6 34.17 to 34.95 (115.21 to 115.39)
USA-P USA-RR Pennsylvania, USA Rotylenchus robusta 1 40.98 79.89
USA-P USA-HS California, USA Heterodera schachtii 6 36.14 to 36.89 (118.73 to 118.96)
USA-P USA/CN-HG Minnesota, USA Heterodera glycines 28 43.6 to 44.9 (92.63 to 94.93)

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


CN-P USA/CN-HG Heilongjiang, Heihe, China Heterodera glycines 8 49.13 to 50.13 127.16 to 127.25
CN-P USA/CN-HG Heilongjiang, Jixi, China Heterodera glycines 11 45.18 to 45.38 131.11 to 131.25
CN-P USA/CN-HG Heilongjiang, Jiamusi, China Heterodera glycines 6 46.5 130.26 to 130.28
CN-P USA/CN-HG Heilongjiang, Suihua, China Heterodera glycines 7 46.13 to 46.48 125.28 to 126.30
CN-P USA/CN-HG Jilin, China Heterodera glycines 1 43.63 122.46
CN-P USA/CN-HG Heilongjiang, Nehe, China Heterodera glycines 1 48.49 124.89
Total samples 87 36.14 to 52.78 6.59 to 131.25;
(78.28 to 121.98)
a
NL-P, population from Netherlands; USA-P, population from USA; CN-P, population from China.
b
NL-GL, strains isolated from Globodera sp.; USA-CX, strains isolated from C. xenoplax; USA-RR, strains isolated from R. robusta; USA-HS, strains isolated from H. schachtii; USA/
CN-HG, strains isolated from H. glycines.
c
Negative values (with parentheses) represent west longitude.

analyses based on genotyping and genealogical analyses can help population structure and their modes of reproduction. In this
to infer past evolutionary events and evaluate the potential pro- study, we used a multilocus sequence analysis (MLSA) scheme
cesses that generated the patterns of genetic variation (20, 21). An based on SNP markers to analyze population genetics of H.
analysis of the population genetics of Hirsutella minnesotensis in- rhossiliensis by sampling from three geographic populations
dicated that this dominant species in China had a clonal popula- and five host nematode-associated populations. We aimed to
tion structure and likely experienced a founder effect after coloni- understand (i) the level and partitioning of genetic variation
zation of China but without a significant bottleneck (22). within H. rhossiliensis, (ii) genetic structure and modes of re-
However, the population genetic structure, possible geographic production in H. rhossiliensis, and (iii) whether geographic isola-
origins, and host nematode adaptation of H. rhossiliensis are tion and host nematode adaptation lead to rapid differentiation
largely unknown. among populations of H. rhossiliensis. This study not only pro-
In recent years, single-nucleotide polymorphism (SNP) analy- vides essential insights into the population differentiation and
ses have been widely used to detect intraspecific variation in mi- host adaptation of H. rhossiliensis but also helps scientists to
croorganisms, including endoparasitic fungi (23, 24). When ana- better understand the ecological mechanism underlying the
lyzing the genetic diversities of human fungal pathogens, for natural control of H. rhossiliensis in its role as a biocontrol agent
instance, Histoplasma capsulatum (25), Candida albicans (26), against nematodes.
Aspergillus fumigatus (23), and Cryptococcus neoformans (27), it
has been demonstrated that SNP analyses were highly discrimina- MATERIALS AND METHODS
tive in these species. Based on multilocus sequence analyses (ML- Sample collection and DNA extraction. We employed all available H.
SAs), genetic diversity in plant-pathogenic fungi was also reported rhossiliensis isolates (87 in total) around the world for this study. They
for Magnaporthe grisea and Ustilaginoidea virens (28, 29). were classified into three geographic populations (Europe, the United
States, and China) and five host nematode-associated populations (Glo-
Field experiments demonstrated that H. rhossiliensis was effec-
bodera, H. schachtii, C. xenoplax, R. robusta, and H. glycines). Except for 8
tive for control of H. glycines in greenhouses (30). Current biolog- isolates from the CBS-KNAW Fungal Biodiversity Centre and 17 from the
ical control applications using H. rhossiliensis, however, have USDA-ARS Collection of Entomopathogenic Fungal Cultures (ARSEF),
shown relatively limited success in agricultural fields. We think the other 62 isolates were all recovered by us from parasitized H. glycines in
that an effective application of H. rhossiliensis or other fungi for Heilongjiang Province, China, or northern Minnesota, USA (Table 1 and
biological control requires a thorough understanding of their Fig. 1; see also Table S1 in the supplemental material).

FIG 1 Map of the geographic distribution of H. rhossiliensis: United States, Europe, and China.

6318 aem.asm.org Applied and Environmental Microbiology November 2016 Volume 82 Number 21
H. rhossiliensis Population Differentiation

To isolate H. rhossiliensis from parasitized H. glycines, we collected simple stepwise-addition heuristic searches with 1,000 replicates. A P
rhizosphere soil samples from soybean fields in the main soybean-pro- value lower than 0.001 indicated statistically significant differences (35).
ducing regions across northeast China from 2007 to 2008 and across Min- After identification of loci with no significant conflict, the phylogenic
nesota (USA) from 2010 to 2011. The sucrose flotation and centrifugation relationship among 87 strains was constructed using two methods: Bayes-
method (31) was used to extract the second-stage juveniles (J2s) that ian inference (BI) and maximum likelihood (ML). To conduct phyloge-
naturally occur in the soil, and these J2s were then transferred to wells of a netic analyses, we used the program PartitionFinder 1.1.1 (36) to evaluate
6-well tissue culture plate. The extracted J2s were examined with an in- the best partitioning scheme and determine suitable substitution models
verted microscope (Olympus, Japan) at 40 magnification, washed twice under the Bayesian information criterion (see Table S5 in the supplemen-
with sterilized water, and then placed on Difco potato dextrose agar tal material). For partitioned data sets, model parameters across different
(PDA) plates at 25C for 2 weeks. All of the filamentous fungi isolated partitions were unlinked, and the overall rate of evolution among parti-

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


from these J2s were cultured on PDA plates covered with a piece of cello- tions was allowed to vary. The Bayesian analysis, which was performed in
phane paper, and 0.05 to 0.1 g of mycelia was harvested and used to extract MrBayes 3.2.2 (37), was initiated with random starting trees, run for 100
genomic DNA with the cetyltrimethylammonium bromide (CTAB) million generations with four incrementally heated chains (Metropolis-
method (32). Fungal cultures were identified based on their morphology coupled Markov chain Monte Carlo [38]), and sampled at intervals of
(30) and nonribosomal DNA (nrDNA) ITS sequences (13). 10,000 generations. To avoid entrapment in the local optima, two inde-
Molecular marker selection. We initially chose three housekeeping pendent Bayesian analyses were run, and the log-likelihood scores were
genes as molecular markers: the nrDNA ITS region, the beta-tubulin gene compared for convergence (38). An ML analysis was performed in
(tub), and the translation elongation factor 1 gene (tef). However, the RAxML version 8.0.0 (39) and conducted under the GTRGAMMA
number of phylogenetically informative sites provided by the three frag- model, and the ML support was assessed by 1,000 bootstrap replicates.
ments was insufficient. Therefore, novel DNA markers were determined Two H. minnesotensis strains, CN3608 and CBS113353, were used as the
by searching the genome data (15). Specifically, 27 fragments of genes outgroup.
putatively involved in the infection processes of H. rhossiliensis into nem- A phylogenetic network was constructed using the median-joining
atodes and 15 fragments of interest as a result of a genomic comparison method implemented in the program SplitsTree 4.12 (40) to further assess
between H. minnesotensis and H. rhossiliensis were chosen as candidate the relationships among worldwide H. rhossiliensis haplotypes. This model-
markers. PCR primers were designed for each of the 42 fragments using free method uses a parsimony approach based on pairwise differences to
the online software Primer3 (http://frodo.wi.mit.edu/primer3/). From 12 connect each sequence to its closest neighbor and allows for the creation
randomly chosen isolates of different origins (CBS104.94, CBS105.94, of internal nodes (median vectors), which are interpreted as unsampled or
ARSEF2006, ARSEF3746, ARSEF2788, ARSEF2789, ARSEF3755, extinct ancestral genotypes to link the existing genotypes in the most
ARSEF3756, CN-13-1, CN-30-1, USA-87-2, and USA-92-1) (see Table S1 parsimonious way (41).
in the supplemental material), we amplified and sequenced each of the 42 Population genetic analyses. To determine whether the number of
DNA fragments. Fragments with a high frequency of nucleotide variation loci was sufficient to represent the genotypic diversity of the populations,
were finally chosen as markers used in this study. A neutrality test for the we plotted the relationship between the number of loci and the genotype
selected markers was performed by DnaSP 5.10 (33). diversity using MultiLocus1.3b (42). A population genetic analysis was
PCR amplification and sequence determination. Nine molecular also performed within and between the three geographic populations and
markers, including ITS, tub, tef, and six newly identified markers, were within and between the five host nematode-associated populations. For
amplified from each of the 87 H. rhossiliensis isolates. Each PCR mixture the within-population genetic analysis, we analyzed the genetic variation
(50 l) was composed of 5 l of 10 EasyTaq buffer, 5 l of deoxynucleo- patterns with DnaSP 5.10, including the number of haplotypes, nucleo-
side triphosphate (dNTP) mixture (2.5 mM), 2 l of each primer (10 tide diversity, and haplotype diversity. For the cross-population genetic
M), 1 l of genomic DNA, 0.5 l of EasyTaq DNA polymerase (Trans- analysis, pairwise differentiation coefficients (Fst) (43) were calculated
Gen Biotech, Beijing, China), and 36.5 l of double-distilled water and tested for significance against 1,000 bootstrap replicates using Arle-
(ddH2O). The PCRs were performed in a TGradient thermocycler (Bi- quin 3.1 (44), and the average gene flow (Nm) (45) was calculated using
ometra, Germany) under the following conditions: denaturation at 94C DnaSP 5.10.
for 3 min; 35 cycles of denaturation at 94C for 40 s, annealing at variable Tests for recombination. Sexual reproduction generates random as-
temperatures for 50 s, and elongation at 72C for 1 min; and a final 10-min sociations between alleles at different loci because genes from different
elongation step at 72C. After confirmation of the PCR products by aga- individuals are mixed in a given population by sexual reproduction.
rose gel electrophoresis, the amplicons were cleaned using the 3S Spin Therefore, when recombination occurs among populations, random as-
PCR product purification kit (Biocolor Bioscience & Technology Com- sociations between alleles may be observed at different loci, even if sexual
pany, China). The purified PCR products were then sequenced using an reproduction has not been observed in the natural populations (46). To
ABI 3730 XL DNA sequencer (Applied Biosystems, USA) with Sanger identify evidence for recombination in H. rhossiliensis, two methods were
dideoxy terminator sequencing. In the case of poor sequencing quality used: the proportion of compatible pairs of loci (PrCP) (47) and the w
resulting from the direct sequencing of PCR products, the representative test (or pairwise homoplasy index [PHI]) (48). Tests for recombination
PCR fragments were cloned, and individually cloned fragments were then based on the principle of compatibility have been proven to be the most
sequenced again. powerful (49, 50). Two loci are compatible (PrCP 1) if it is possible to
Nucleotide variations. Sequences of the fragments were aligned with account for all the observed genotypes by mutations without inferring
the Muscle algorithm of MEGA 6.0 (34) and trimmed to remove ambig- homoplasy (reversals, parallelisms, or convergences) or recombination;
uously aligned regions. The nucleotide diversity (), haplotype diversity, otherwise, the loci are incompatible (PrCP 1). The w test means that
number of polymorphic sites, and allele numbers were estimated for both phylogenetic incompatibility is an indicator of recombination at the
individual fragment and combined sequences using the program DnaSP population level; the lack of phylogenetic incompatibility, in contrast,
5.10 (33). By employing DnaSP 5.10, alleles at each locus were assigned implies asexual reproduction. The tests for PrCP were calculated using
and combined into an allelic profile designated a haplotype (HA) for each MultiLocus1.3b (42) with 1,000 randomizations; the w test was calcu-
strain. lated by SplitsTree 4.12.
Phylogenetic and network analyses. We performed a partition ho- Tests for genetic isolation according to geography and host nema-
mogeneity test (PHT) to detect potential phylogenetic conflicts between todes. An analysis of molecular variance (AMOVA) was performed to
different loci using PAUP version 4.0b10 (Sinauer Associates, Sunder- determine the relative contribution of genetic variation among and within
land, MA, USA; D. Swofford, 2003). PHT used informative characters and populations using GenAlEx 6.5 (51). The AMOVA was conducted for

November 2016 Volume 82 Number 21 Applied and Environmental Microbiology aem.asm.org 6319
Wang et al.

TABLE 2 Patterns of genetic variation at the individual loci and combined-locus data sets
No. of sites
Aligned Nucleotide Haplotype
Gene length Parsimony Singleton diversity, diversity No. of Significance of
Fragment name identifier (bp) Variableb informativec variable Indel (102) (Hd) allelesd Tajimas D
nrDNA ITS 476 3 3 0 0 0.124 0.503 4 P 0.1
Beta-tubulin (tub) HIR_02075 284 8 3 0 5 0.072 0.194 4 P 0.1
Translation elongation HIR_06488 803 2 2 0 0 0.024 0.191 3 P 0.1
factor (tef)
Glycoside hydrolase HIR_00714 715 67 15 0 52 0.419 0.469 7 0.1 P 0.05

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


SignalP HIR_06686 807 191 18 2 171 0.281 0.531 8 P 0.1
Abhydrolase HIR_08121 754 7 6 1 0 0.363 0.487 7 P 0.1
Fucose-specific lectin HIR_08679 711 11 8 3 0 0.213 0.515 11 P 0.1
Nucleoside triphosphate HIR_01569 350 11 6 5 0 0.405 0.598 8 P 0.1
hydrolases
Exo_endo_phos HIR_09423 473 47 47 0 0 3.269 0.587 10 0.1 P 0.05
Combined 9-locus data set 5,373 347 108 11 228 0.497 0.921 44
Combined 8-locus data set 4,658 280 93 11 176 0.493 0.921 44
excluding HIR_00714a
a
The sequences of HIR_00714 are excluded due to the partition homogeneity test results (see Table S3 in the supplemental material). The fungal combined data set without
HIR_00714 was used in all the following analyses.
b
The variable sites consist of substitution sites (phylogenetically informative or uninformative) and indel (insertion/deletion) sites.
c
At least two mutations occurred no less than twice after sequence alignment.
d
The number of individuals or multiple loci is represented by the most dominant allele or multilocus haplotypes.

both the three geographic populations and the five host nematode-asso- Phylogenetic and population structure analyses of H. rho-
ciated populations. ssiliensis. Bayesian (BI) and maximum likelihood (ML) phyloge-
Mantel tests were performed in GenAlEx 6.5 for three groups of cor- netic trees were constructed based on the 8-locus concatenated
relations: between the fungal genetic distance and longitudinal-latitudinal data set. H. rhossiliensis strains tend to cluster by geography and
coordinates of geographic distance, between the fungal genetic distance
nematode hosts, and each of the two phylogenetic approaches
and longitudinal distance, and between the fungal genetic distance and
latitudinal distance. For the genetic distances, we used the pairwise pop- identified six clusters: European strains from Globodera sp. (NL-
ulation PhiPT values, which represent genetic distances among all pairs of GL), American strains from C. xenoplax (USA-CX), American
populations as a tri-matrix, and these distances were then compared with strains from R. robusta (USA-RR), American strains from H.
the geographic distances between populations. schachtii (USA-HS), American strains from H. glycines (USA-
Accession number(s). All of the nucleotide sequences obtained in this HG), and Chinese (northeastern) strains from H. glycines (CN-
study have been submitted to the GenBank database, and the accession HG). However, there were subtle topological differences between
numbers are KP885720 to KP886014 (see also Table S2 in the supplemen- the two approaches. For example, two American strains from H.
tal material).
schachtii (CBS567.92 and ARSEF3761) did not group into the
RESULTS USA-HS subclade, and the positions of two Chinese strains from
H. glycines (CN-30-2-JX and CN-3-10-JL) and one American
Nucleotide variations. To determine appropriate markers, we
strain from C. xenoplax (ARSEF3746) had inconsistent place-
successfully amplified the 42 candidate fragments from each of
ments between the BI and ML trees (Fig. 2; see also Fig. S2 in the
the 12 test isolates. Six fragments (HIR_00714, HIR_06686,
HIR_08121, HIR_08679, HIR_01569, and HIR_09423) showed a supplemental material).
high frequency of nucleotide variation and existed as single copies The network diagram showed a linear structure among the 87
as indicated by BLAST searches against the H. rhossiliensis ge- strains (Fig. 3), with Chinese strains from H. glycines, American
nome. The six fragments, together with three commonly used strains from H. glycines, European strains from Globodera sp.,
markers (nrDNA ITS, tub, and tef), were finally chosen to amplify American strains from H. schachtii, American strains from R. ro-
from each of the 87 isolates (Table 2). Direct sequencing of PCR busta, and American strains from C. xenoplax each forming an
products from any of the nine fragments indicated no heteroge- individual cluster.
neity in sequencing chromatograms. Three to 11 alleles were iden- Genetic differentiation and gene flow among geographic or
tified in our samples depending on the locus. Moreover, the result host nematode-associated populations. A high value for the ge-
of a neutrality test showed that all nine markers were neutral netic differentiation coefficient Fst (P 0.01) revealed that signif-
(Table 2). icant genetic differentiation occurred among all of the geographic
A significant conflict was detected by PHT between HIR_00714 or host nematode-associated populations except between NL-P
and the other loci (see Table S4 in the supplemental material). and USA-P, the value for which was not significant (P 0.01)
Therefore, HIR_00714 was excluded from the concatenated data (Tables 3 and 4). Consistently, the estimated gene flow was low
set in the following analyses. From the concatenation of the re- among the three geographic populations (Nm 0.14) and among
maining eight loci, a total of 280 variable sites within the com- the five host nematode-associated populations (Nm 0.28).
bined 4,658-bp sequence and 44 multilocus haplotypes were iden- Evidence for recombination. The PrCP index provided evi-
tified from the 87 isolates (Table 2). dence for recombination except for two populations, American

6320 aem.asm.org Applied and Environmental Microbiology November 2016 Volume 82 Number 21
Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY

FIG 2 Bayesian tree of the H. rhossiliensis strains based on the concatenated 8-locus data set. The strict consensus tree from phylogenetic analysis was constructed
using a Bayesian inference method. The numbers above the branches indicate the posterior probability (PP) support for each node according to the Bayesian
inference (nodes with PPs of 0.5 are shown).

November 2016 Volume 82 Number 21 Applied and Environmental Microbiology aem.asm.org 6321
Wang et al.

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


FIG 3 Median joining of phylogenetic network inferred with H. rhossiliensis strain concatenated 8-locus data set. This network includes all of the most
parsimonious trees linking the isolates. Each unique haplotype is represented by a block, with the blocks size indicating the haplotypes frequency in the data set.
Red blocks denote haplotypes parasitizing H. glycines in China, blue blocks denote haplotypes from H. glycines in the United States, green blocks denote
haplotypes parasitizing Globodera, yellow blocks denote haplotypes parasitizing R. robusta, black blocks denote haplotypes parasitizing H. schachtii, and pink
blocks denote haplotypes parasitizing C. xenoplax.

strains from C. xenoplax (USA-CX) and American strains from R. DISCUSSION


robusta (USA-RR). When the 87 strains were analyzed together, We collected 87 H. rhossiliensis strains from five host nematodes in
recombination signals were found in 3/5 locus pairs (Table 3; see Europe, the United States, and China. Although a large number of
also Table S6 in the supplemental material). strains were obtained from H. glycines, at least three strains were
Similarly, the w test found statistically significant evidence for included from each of the other nematode hosts (Table 1; see also
recombination (P 0.001). However, there was a slight difference Table S1 in the supplemental material). We would explain that
in the w results for recombination among the three geographic our samples from four nematodes (Globodera sp., H. schachtii, C.
populations and five host nematode-associated populations com- xenoplax, and R. robusta) were isolated in the 1980s and 1990s,
pared with the PrCP results. For example, the w test did not find
while samples from H. glycines were isolated during 2007 to 2011.
statistically significant evidence for recombination in the USA-P,
Because some nematodes did not occur in China or we did not
USA-CX, USA-HS, or USA-RR population (Table 3).
find this fungus from them (11, 5254), we failed to isolate H.
Genetic isolation by geography or host nematodes. The
rhossiliensis from other host nematodes except from H. glycines. It
AMOVA results showed that 44% of the genetic variation could be
attributed to variation among the three geographic populations was also difficult to obtain strains from the above four host nem-
and 56% could be attributed to variation among the individual atodes (except H. glycines) in the United States. The cultures from
isolates within populations (Table 5). For the five host nematode- culture collection centers (CBS and ARSEF) were preserved in
associated populations, the AMOVA results showed that 61% and lyophilized forms, which largely decreased strain mutations over
39% of the genetic variation could be attributed to variation the past 20 to 30 years (see Table S1 in the supplemental material).
among the populations and among individual isolates within pop- Although there might be some genetic variations of fungus in
ulations, respectively. Both of these sources of variation were sig- nature, we believe that the variation during 20 to 30 years could
nificant (P 0.01) (Table 6). not affect the main results obtained from this study.
The Mantel tests revealed significant genetic isolation by dis- In this study, recombination was found in most of the H. rho-
tance irrespective of the horizontal geographical distances (based ssiliensis populations and when the whole samples were consid-
on longitudinal-latitudinal coordinates, along a longitudinal gra- ered. It can be inferred and supported by two lines of evidence, the
dient, or along a latitudinal gradient) or population divisions PrCP values and the w test (Tables 3 and 4). The PrCP test mea-
(three geographic populations) employed (Table 7). sures whether two loci are compatible, and the w test examines

TABLE 3 Population genetic parameters for each of three geographic populations of H. rhossiliensis
Fst valueb
Geographic No. of Nucleotide Haplotype
population haplotypes diversity, (102) diversity (Hd) PrCPa (P value) P value (w) NL-P USA-P
NL-P 4 0.053 0.9 0.964 (0.271) 0.042
USA-P 23 0.195 0.804 0.750 (0.001) 0.192 0.007
CN-P 17 0.378 0.868 0.857 (0.933) 0.005 0.607** 0.399**
Total 44 0.393 (0.001) 0.001
a
PrCP, proportion of phylogenetically compatible pairs of loci.
b
Fst, fixation index. **, P 0.01

6322 aem.asm.org Applied and Environmental Microbiology November 2016 Volume 82 Number 21
H. rhossiliensis Population Differentiation

TABLE 4 Population genetic parameters for each of five host nematode-associated populations of H. rhossiliensis
Fst valuea
Host nematode-associated No. of Nucleotide diversity, Haplotype P value
population No. haplotypes (102) diversity (Hd) PrCP (P value) (w) NL-GL USA-CX USA-RR USA-HS
NL-GL 5 4 0.053 0.9 0.964 (0.271) 0.042
USA-CX 7 7 0.04 1 1 (1) 0.248 0.532**
USA-RR 7 3 0.038 0.524 1 (1) 0.776** 0.634**
USA-HS 6 6 0.125 1 0.892 (0.717) 0.24 0.173* 0.542** 0.763**
USA/CN-HG 62 24 0.445 0.849 0.785 (0.261) 0.001 0.325* 0.749** 0.746** 0.341*
a
*, 0.01 P 0.05; **, P 0.01.

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


phylogenetic incompatibility. So, it is reasonable that there is con- observed from high Fst values and low Nm values. We suggest
flict between the two results (49). Indeed, the sexual cycles of most that the North Pacific Ocean may be a boundary separating the
fungi are difficult to observe in nature (55), and inferences of the Chinese population and the American population of H. rhossil-
potential for a sexual cycle to occur have largely relied on analyses iensis because of limited opportunities to disperse via wind or
of gene and genotype frequencies in natural populations. The sex- soil. A similar geography-based separation has been reported in
ual stage of H. rhossiliensis has not been discovered in nature or in Colletotrichum truncatum from chili peppers in China that were
the laboratory, although we detected mating type genes in its ge- genetically differentiated into southern and northern popula-
nome (unpublished data). In plant fungal pathogens, a high level tions (61). Previously, we found significant differences among
of recombination provides an advantage in rapidly generating local populations of H. minnesotensis from China and the United
many new combinations of virulence genes to counterbalance States, where isolation by geography plays a key role in genetic
corresponding resistance genes in the host (56, 57). In arbuscular differentiation (22). Based on these results, we conclude that geo-
mycorrhizal fungi, recombination or recombination-like events graphic distance contributes to driving population genetic differ-
in addition to clonality have greatly contributed to genetic diver- entiation of H. rhossiliensis.
sity (58). This study has demonstrated evidence of recombination Among the five host nematode-associated populations studied
among H. rhossiliensis strains, and recombination appears to be here, we identified positive correlations between H. rhossiliensis
one of the factors shaping the evolution of H. rhossiliensis. Similar genetic distance at different latitudinal and longitudinal distances.
results were also observed in Candida albicans, which showed both We observed 61% of the molecular genetic variance among those
clonality and recombination, even though a complete sexuality nematode-associated populations. This might result from host-
stage is not known to exist in this fungus (59). However, sexual related selection driving population genetic differentiation at neu-
recombination may not be the sole factor because we cannot ex- tral and selected loci (62). Another explanation was that host shifts
clude the possibility that parasexual recombination has contrib- might occur frequently between neighboring nematodes, which
uted to the observed linkage equilibrium and phylogenetic incom- increased the mating probability of H. rhossiliensis across the
patibility (60). nearby hosts and also increased the genetic variation within and
Our population genetics analyses of H. rhossiliensis were based among populations.
on two types of population divisions: three geographic popula- Our analyses revealed that both nematode hosts and geography
tions (NL-P, USA-P, and CN-P) and five host nematode-associ- contributed to genetic differentiation among H. rhossiliensis pop-
ated populations (NL-GL, USA-CX, USA-RR, USA-HS, and USA/ ulations (Tables 5 to 7). We, however, could not perform a two-
CN-HG) (Table 1). The significance of geography and the host way AMOVA to further look at their relative effects because pop-
nematodes in shaping the structure of H. rhossiliensis populations ulation assignments of our samples did not meet the requirement
on a large scale was supported by the strong divergent signals in for performing such an analysis. In this study, we confirmed that
the phylogenetic and haplotype network analyses (Fig. 2 and 3). In difference in host nematode was a key factor in regulating genetic
this study, a significant positive correlation was observed between divergence of nematode parasitic fungus from the phylogenetic
H. rhossiliensis genetic distance and latitudinal and longitudinal approaches and the haplotype network (Fig. 2 and 3). Similarly,
geographic distance, thus indicating that geography is an im- the strong association of genetic groups with different geographic
portant barrier to gene flow and has promoted the genetic di- locations suggested the important role of geographic isolation.
vergence within fungal species (Table 7). Moreover, differenti- Previous studies on microorganisms such as Bacillus simplex con-
ation among the three geographic populations was also cluded that genetic differentiation was strongly attributed to geo-

TABLE 5 Summary results of the analysis of molecular variance TABLE 6 Summary results of the analysis of molecular variance
(AMOVA) within and among three geographic populations of H. (AMOVA) within and among five host nematode-associated
rhossiliensisa populations of H. rhossiliensisa
Estimated Estimated
Source df SS MS variance % Statistic Value P Source df SS MS variance % Statistic Value P
Among populations 2 48.571 24.286 0.981 0.44 PhiPT 0.436 0.001 Among populations 4 71.073 17.768 1.634 61 PhiPT 0.615 0.001
Within populations 84 106.486 1.268 1.268 0.56 PhiPT 0.505 0.001 Within populations 84 83.985 1.024 1.024 39 PhiPT 0.527 0.002
Total 86 155.057 2.248 1 Total 86 155.057 2.658 100
a a
There were 3 geographic populations and 87 isolates. Abbreviations: df, degree of There were 5 host-nematode associated populations and 87 isolates. Abbreviations: df,
freedom; SS, sum of squared observations; MS, mean of squared observations. degree of freedom; SS, sum of squared observations; MS, mean of squared observations.

November 2016 Volume 82 Number 21 Applied and Environmental Microbiology aem.asm.org 6323
Wang et al.

TABLE 7 Mantel test results between genetic distances and geographic tial, and soil solutions on parasitism of Criconemella xenoplax by Hirsu-
distances among three geographic populations tella rhossiliensis. Phytopathology 73:544 546. http://dx.doi.org/10.1094
/Phyto-73-544.
Genetic distance 8. Mller J. 1984. The influence of two pesticides on fungal parasites of
Geographic distance r2 P Heterodera schachtii. Colloq INRA 31:225231.
9. Chen SY, Reese CD. 1999. Parasitism of the nematode Heterodera glycines
Latitudinal and longitudinal 0.315 0.01 by the fungus Hirsutella rhossiliensis influenced by crop sequence. J Nema-
Latitudinal 0.351 0.01 tol 31:437 444.
Longitudinal 0.316 0.02 10. Tedford EC, Jaffee BA. 1995. In vitro parasitism of Rotylenchus robustus
by isolates of Hirsutella rhossiliensis. J Nematol 22:486 489.
11. Dobosz R, Obrepalska-Steplowska A, Kornobis S. 2006. Globodera ar-

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


temisiae (Eroshenko et Kazachenko, 1972) (Nematoda: Heteroderidae)
graphic locations (63). Numerous studies found that geographic from Poland. J Plant Prot Res 46:403 407.
region and host species contribute equally to the population ge- 12. Ma R, Liu XZ, Jian H, Li SD. 2005. Detection of Hirsutella spp. and
netic differentiation, for instance, Grosmannia clavigera (64), Pasteuria sp. parasitizing second-stage juveniles of Heterodera glycines in
soybean fields in China. Biol Control 33:223229. http://dx.doi.org/10
Claviceps purpurea (65), Venturia inaequalis (66), Arthrobotrys oli- .1016/j.biocontrol.2005.03.004.
gospora (67), Colletotrichum gloeosporioides (68), and Metarhiz- 13. Xiang MC, Yang XH, Wang ZX, Liu XZ, Chen SY, Xiao QM. 2007.
ium anisopliae (69). Variability of morphology, parasitism, and nucleotide sequences among
In conclusion, our study identified the population genetic struc- isolates and species of nematophagous Hirsutella. Biol Control 41:110
119. http://dx.doi.org/10.1016/j.biocontrol.2006.12.016.
ture of the nematode endoparasitic fungus H. rhossiliensis and found 14. Timper P. 2011. Uitilization of biological control for managing plant-
a high level of genetic variation among its populations worldwide. parasitic nematodes, p 259 289. In Davies K, Spiegel Y (ed), Biological
Using multiple sources of evidence, we demonstrated that recombi- control of plant-parasitic nematodes: building coherence between micro-
nation among strains, isolation by geography, and isolation based on bial ecology and molecular mechanisms, vol 11. Springer, Dordrecht,
the host nematode were factors driving the genetic differentiation of Netherlands.
15. Lai YL, Liu KK, Zhang XY, Zhang XL, Li K, Wang NN, Shu C, Wu YP,
H. rhossiliensis. Therefore, the findings of this study on the popula- Wang CS, Bushley KE, Xiang MC, Liu XZ. 2014. Comparative genomics
tion structure of H. rhossiliensis and the associated multiple driving and transcriptomics analyses reveal divergent lifestyle features of nema-
factors affecting its genetic diversity provide an essential understand- tode endoparasitic fungus Hirsutella minnesotensis. Genome Biol Evol
ing of the mechanisms of evolution and differentiation of the fungus 6:30773093. http://dx.doi.org/10.1093/gbe/evu241.
16. Tedford EC, Jaffee BA, Muldoon AE. 1994. Variability among isolates of the
on a global scale. These findings are expected to provide important nematophagous fungus Hirsutella rhossiliensis. Mycol Res 98:11271136.
insights into nematode control by this fungus. 17. McDonald BA, Linde C. 2002. Pathogen population genetics, evolution-
ary potential and durable resistance. Annu Rev Phytopathol 40:349 379.
ACKNOWLEDGMENTS http://dx.doi.org/10.1146/annurev.phyto.40.120501.101443.
This study was supported by the National Natural Science Foundation of 18. Stukenbrock E. 2013. Evolution, selection and isolation: a genomic view
of speciation in fungal plant pathogens. New Phytol 199:895907. http:
China (grant no. 31430071) and the National Basic Research Program of
//dx.doi.org/10.1111/nph.12374.
China (973 program, grant no. 2013CB127506). 19. Cheng K, Rong XY, Pinto-Toms AA, Fernndez-Villalobos M,
We thank Richard A. Humber, who kindly provided 15 strains from Murillo-Cruz C, Huang Y. 2015. Population genetic analysis of Strepto-
the USDA-ARS Collection of Entomopathogenic Fungal Cultures myces albidoflavus reveals habitat barriers to homologous recombination
(ARSEF) database at the USDA-ARS Biological Integrated Pest Manage- in the diversification of streptomycetes. Appl Environ Microbiol 81:966
ment Research Unit, Cornell University, New York, USA. 3975. http://dx.doi.org/10.1128/AEM.02925-14.
20. Giraud T, Refrgier G, Le Gac M, de Vienne DM, Hood ME. 2008.
FUNDING INFORMATION Speciation in fungi. Fungal Genet Biol 45:791 802. http://dx.doi.org/10
This work, including the efforts of Xingzhong Liu, was funded by National .1016/j.fgb.2008.02.001.
Basic Research Program of China (2013CB127506). This work, including 21. Tittes S, Kane SC. 2014. The genomics of adaptation, divergence and
speciation: a congealing theory. Mol Ecol 23:3938 3940. http://dx.doi.org
the efforts of Xingzhong Liu, was funded by National Natural Science
/10.1111/mec.12855.
Foundation of China (NSFC) (31430071). 22. Shu C, Jiang XZ, Cheng XL, Wang NN, Chen SY, Xiang MC, Liu XZ.
2015. Genetic structure and parasitization-related ability divergence of a
REFERENCES nematode fungal pathogen Hirsutella minnesotensis following founder ef-
1. Cantrell SA, Claribel BF. 2010. Fungal molecular diversity of a Puerto fect in China. Fungal Genet Biol 81:212220. http://dx.doi.org/10.1016/j
Rican subtropical hypersaline microbial mat. Fungal Ecol 3:402 405. .fgb.2015.02.005.
http://dx.doi.org/10.1016/j.funeco.2010.04.001. 23. Bain JM, Tavanti A, Davidson AD, Jacobsen MD, Shaw D, Gow NA,
2. Abad P, Gouzy J, Aury J, Castagnone-Sereno P, Danchin E, Deleury E, Odds FC. 2007. Multilocus sequence typing of the pathogenic fungus
Perfus-Barbeoch L, Anthouard V, Artiguenave F, Blok VC, Caillaud Aspergillus fumigatus. J Clin Microbiol 45:1469 1477. http://dx.doi.org
MC, Coutinho PM, Dasilva C. 2008. Genome sequence of the metazoan /10.1128/JCM.00064-07.
plant-parasitic nematode Meloidogyne incognita. Nat Biotechnol 26:909 24. Nunney L, Vickerman DB, Bromley RE, Russell SA, Hartman JR,
915. http://dx.doi.org/10.1038/nbt.1482. Morano LD, Stouthamer R. 2013. Recent evolutionary radiation and host
3. Li TF, Zhang KQ, Liu XZ. 2000. Taxonomy of nematophagous fungi. plant specialization in the Xylella fastidiosa subspecies native to the United
China Science and Technology Press, Beijing, China. States. Appl Environ Microbiol 79:2189 2200. http://dx.doi.org/10.1128
4. Liu XZ, Chen SY. 2001. Screening isolates of Hirsutella species for bio- /AEM.03208-12.
control of Heterodera glycines. Biocontrol Sci Technol 11:151160. http: 25. Kasuga T, White TJ, Koenig G, McEwen J, Restrepo A, Castaeda E, Da
//dx.doi.org/10.1080/09583150020029826. Silva Lacaz C, Heins-Vaccari EM, De Freitas RS, Zancop-Oliveira RM,
5. Liu XZ, Xiang MC, Che YS. 2009. The living strategy of nematopha- Qin Z, Negroni R, Carter DA, Mikami Y, Tamura M, Taylor ML, Miller
gous fungi. Mycoscience 50:20 25. http://dx.doi.org/10.1007/S10267 GF, Poonwan N, Taylor JW. 2003. Phylogeography of the fungal patho-
-008-0451-3. gen Histoplasma capsulatum. Mol Ecol 12:33833401. http://dx.doi.org
6. Minter DW, Brady BL. 1980. Mononematous species of Hirsutella. Trans /10.1046/j.1365-294X.2003.01995.x.
Br Mycol Soc 74:271282. http://dx.doi.org/10.1016/S0007-1536(80) 26. Li J, Fan SR, Liu XP, Li DM, Nie ZH, Li F, Lin H, Huang WM, Zong
80157-4. LL, Lei H, Bai FY. 2008. Biased genotype distributions of Candida albi-
7. Jaffee BA, Zehr EI. 1983. Effects of certain solutions, osmotic poten- cans strains associated with vulvovaginal candidosis and candidal balano-

6324 aem.asm.org Applied and Environmental Microbiology November 2016 Volume 82 Number 21
H. rhossiliensis Population Differentiation

posthitis in China. Clin Infect Dis 47:1119 1125. http://dx.doi.org/10 51. Peakall R, Smouse PE. 2012. GenAlEx 6.5: genetic analysis in Excel.
.1086/592249. Population genetic software for teaching and researchan update. Bioin-
27. Meyer W, Aanensen DM, Boekhout T, Cogliati M, Diaz MR, Esposto formatics 19:25372539.
MC, Fisher M, Gilgado F, Hagen F, Kaocharoen S, Litvintseva AP, 52. Nyczepir AP, Zehr EI, Lewis SA, Harshman DC. 1983. Short life of peach
Mitchell TG, Simwami SP, Trilles L, Viviani MA, Kwon-Chung J. 2009. trees induced by Criconemella xenoplax. Plant Dis 67:507508.
Consensus multi-locus sequence typing scheme for Cryptococcus neofor- 53. Atighi MR, Pourjam E, Pedram M, Cantalapiedra-Navarrete C,
mans and Cryptococcus gattii. Med Mycol 47:561570. http://dx.doi.org Palomares-Rius JE, Castillo P. 2011. Molecular and morphological
/10.1080/13693780902953886. characterisations of two new species of Rotylenchus (Nematoda: Hop-
28. Choi J, Park SY, Kim BR, Roh JH, Oh IS, Han SS, Lee YH. 2013. lolaimidae) from Iran. Nematology 13:951964. http://dx.doi.org/10
Comparative analysis of pathogenicity and phylogenetic relationship in .1163/138855411X571795.
Magnaporthe grisea species complex. PLoS One 8:e57196. http://dx.doi 54. Amiri S, Subbotin SA, Moens M. 2002. Identification of the beet cyst

Downloaded from http://aem.asm.org/ on May 24, 2017 by INSTITUTE OF MICROBIOLOGY


.org/10.1371/journal.pone.0057196. nematode Heterodera schachtii by PCR. Eur J Plant Pathol 108:497506.
29. Sun XY, Kang S, Zhang YJ, Tan XQ, Yu YF, He HY, Zhang XY, Liu YF, http://dx.doi.org/10.1023/A:1019974101225.
Wang S, Sun WX, Cai L, Li SJ. 2013. Genetic diversity and population 55. Calo S, Billmyre RB, Heitman J. 2013. Generators of phenotypic diver-
structure of rice pathogen Ustilaginoidea virens in China. PLoS One sity in the evolution of pathogenic microorganisms. PLoS Pathog
8:e76879. http://dx.doi.org/10.1371/journal.pone.0076879. 9:e1003181. http://dx.doi.org/10.1371/journal.ppat.1003181.
30. Chen SY, Liu XZ. 2005. Control of the soybean cyst nematode by the 56. Zhan J, Pettway RE, McDonald BA. 2003. The global genetic structure of
fungi Hirsutella rhossiliensis and Hirsutella minnesotensis in greenhouse the wheat pathogen Mycosphaerella graminicola is characterized by high
studies. Biol Control 32:208 219. http://dx.doi.org/10.1016/j.biocontrol nuclear diversity, low mitochondrial diversity, regular recombination,
.2004.09.013. and gene flow. Fungal Genet Biol 38:286 297. http://dx.doi.org/10.1016
31. Jenkin WR. 1964. A rapid centrifugal-flotation technique for separating /S1087-1845(02)00538-8.
nematodes from soil. Plant Dis Rep 48:692. 57. Souza EA, Camargo OA, Jr, Pinto JM. 2010. Sexual recombination in
32. Murray MG, Thompson WF. 1980. Rapid isolation of high molecular Colletotrichum lindemuthianum occurs on a fine scale. Genet Mol Res
weight plant DNA. Nucleic Acids Res 8:4321 4326. 9:1759 1769. http://dx.doi.org/10.4238/vol9-3gmr863.
33. Librado P, Rozas J. 2009. DnaSP v5: a software for comprehensive anal- 58. Vandenkoornhuyse P, Leyval C, Bonnin I. 2001. High genetic diversity
ysis of DNA polymorphism data. Bioinformatics 25:14511452. http://dx in arbuscular mycorrhizal fungi: evidence for recombination events. He-
.doi.org/10.1093/bioinformatics/btp187. redity 87:243253. http://dx.doi.org/10.1046/j.1365-2540.2001.00941.x.
34. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. 2013. MEGA6: 59. Graser Y, Volovsek M, Arrington O, Schonian G, Presber W, Mitchell
molecular evolutionary genetics analysis version 6.0. Mol Biol Evol 30: TG, Vilgalys AR. 1996. Molecular markers reveal that population struc-
27252729. http://dx.doi.org/10.1093/molbev/mst197. ture of the human pathogen Candida albicans exhibits both clonality and
35. Cunningham CW. 1997. Can three incongruence tests predict when data recombination. Proc Natl Acad Sci U S A 93:1247312477.
should be combined? Mol Biol Evol 14:733740. 60. Noguchi MT, Yasuda N, Fujita Y. 2006. Evidence of genetic exchange by
36. Lanfear R, Calcott B, Ho SY, Guindon S. 2012. Partitionfinder: com-
parasexual recombination and genetic analysis of pathogenicity and mat-
bined selection of partitioning schemes and substitution models for phy-
ing type of parasexual recombinants in rice blast fungus, Magnaporthe
logenetic analyses. Mol Biol Evol 29:16951701. http://dx.doi.org/10.1093
oryzae. Phytopathology 96:746 750. http://dx.doi.org/10.1094/PHYTO
/molbev/mss020.
-96-0746.
37. Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Hhna
61. Diao YZ, Zhang C, Xu JP, Lin D, Liu L, Mtunge OG, Liu XL. 2015.
S, Larget B, Liu L, Suchard MA, Huelsenbeck JP. 2012. MrBayes 3.2:
Genetic differentiation and recombination among geographic popula-
efficient Bayesian phylogenetic inference and model choice across a large
tions of the fungal pathogen Colletotrichum truncatum from chili peppers
model space. Syst Biol 61:539 542. http://dx.doi.org/10.1093/sysbio
in China. Evol Appl 8:108 118. http://dx.doi.org/10.1111/eva.12233.
/sys029.
38. Huelsenbeck JP, Bollback JP. 2001. Empirical and hierarchical Bayesian 62. Orsini L, Vanoverbeke J, Swillen I, Mergeay J, Meester LD. 2013.
estimation of ancestral states. Syst Biol 50:351366. http://dx.doi.org/10 Drivers of population genetic differentiation in the wild: isolation by dis-
.1080/106351501300317978. persal limitation, isolation by adaptation and isolation by colonization.
39. Stamatakis A. 2014. RAxML version 8: a tool for phylogenetic analysis and Mol Ecol 22:59835999. http://dx.doi.org/10.1111/mec.12561.
post-analysis of large phylogenies. Bioinformatics 30:13121313. http: 63. Sikorski J, Nevo E. 2005. Adaptation and incipient sympatric speciation
//dx.doi.org/10.1093/bioinformatics/btu033. of Bacillus simplex under microclimatic contrast at Evolution Canyons
40. Huson DH, Bryant D. 2006. Application of phylogenetic networks in I and II, Israel. Proc Natl Acad Sci U S A 102:15924 15929. http://dx.doi
evolutionary studies. Mol Biol Evol 23:254 267. .org/10.1073/pnas.0507944102.
41. Bandelt HJ, Forster P, Rohl A. 1999. Median-joining networks for in- 64. Alamouti SM, Wang V, DiGuistini S, Six DL, Bohlmann J, Hamelin RC.
ferring intraspecific phylogenies. Mol Biol Evol 16:37 48. 2011. Gene genealogies reveal cryptic species and host preferences for the
42. Agapow PM, Burt A. 2001. Indices of multilocus linkage disequilibrium. pine fungal pathogen Grosmannia clavigera. Mol Ecol 20:25812602. http:
Mol Ecol Notes 1:101102. http://dx.doi.org/10.1046/j.1471-8278.2000 //dx.doi.org/10.1111/j.1365-294X.2011.05109.x.
.00014.x. 65. Douhan GW, Smith ME, Huyrn KL, Westbrook A, Beerli P, Fisher AJ.
43. Weir BS, Cockerham CC. 1984. Estimating F-statistics for the analysis of 2008. Multigene analysis suggests ecological speciation in the fungal
population structure. Evolution 38:1358 1370. pathogen Claviceps purpurea. Mol Ecol 17:2276 2286. http://dx.doi.org
44. Excoffier L, Laval G, Schneider S. 2005. Arlequin version 3.0: an inte- /10.1111/j.1365-294X.2008.03753.x.
grated software package for population genetics data analysis. Evol Bioin- 66. Gladieux P, Gurin F, Giraud T, Caffier V, Lemaire C, Parisi L, Didelot
form 1:4750. F, Cam LB. 2011. Emergence of novel fungal pathogens by ecological
45. Slatkin M. 1987. Gene flow and the geographical structure of natural speciation: importance of the reduced viability of immigrants. Mol Ecol
populations. Science 236:787792. 20:4521 4532. http://dx.doi.org/10.1111/j.1365-294X.2011.05288.x.
46. Xu JP. 2006. Fundamentals of fungal molecular population genetic anal- 67. Zhang Y, Qiao M, Xu J, Cao Y, Zhang KQ, Yu ZF. 2013. Genetic
yses. Curr Issues Mol Biol 8:75 89. diversity and recombination in natural populations of the nematode-
47. Liang JF, Xu J, Yang ZL. 2009. Divergence, dispersal and recombination trapping fungus Arthrobotrys oligospora from China. Ecol Evol 3:312325.
in Lepiota cristata from China. Fungal Divers 38:105. http://dx.doi.org/10.1002/ece3.450.
48. Bruen TC, Philippe H, Bryant D. 2006. A simple and robust statistical 68. Gazis R, Rehner S, Chaverri P. 2011. Species delimitation in fungal
test for detecting the presence of recombination. Genetics 172:26652681. endophyte diversity studies and its implications in ecological and biogeo-
49. Posada D, Crandall KA. 2001. Evaluation of methods for detecting recom- graphic inferences. Mol Ecol 20:30013013. http://dx.doi.org/10.1111/j
bination from DNA sequences: computer simulations. Proc Natl Acad Sci .1365-294X.2011.05110.x.
U S A 98:1375713762. http://dx.doi.org/10.1073/pnas.241370698. 69. Bidochka MJ, Kamp AM, Lavender TM, Dekoning J, De Croos JA. 2001.
50. Wiuf C, Christensen T, Hein J. 2001. A simulation study of the reliability Habitat association in two genetic groups of the insect-pathogenic fungus
of recombination detection methods. Mol Biol Evol 18:1929 1939. http: Metarhizium anisopliae: uncovering cryptic species? Appl Environ Micro-
//dx.doi.org/10.1093/oxfordjournals.molbev.a003733. biol 67:13351342. http://dx.doi.org/10.1128/AEM.67.3.1335-1342.2001.

November 2016 Volume 82 Number 21 Applied and Environmental Microbiology aem.asm.org 6325

You might also like