Knoxdaviesia capensis: dispersal ecology and population genetics of a flower-associated fungus

Fungal Ecology 26 (2017) 28e36

Contents lists available at ScienceDirect

Fungal Ecology journal homepage: www.elsevier.com/locate/funeco

Knoxdaviesia capensis: dispersal ecology and population genetics of a flower-associated fungus
anne L. Dreyer a, Tessa Laas a, Lelani Smit a, Francois Roets b Janneke Aylward a, *, Le a b

Department of Botany and Zoology, Stellenbosch University, Matieland 7602, South Africa Department of Conservation Ecology and Entomology, Stellenbosch University, Matieland 7602, South Africa

a r t i c l e i n f o

a b s t r a c t

Article history: Received 23 June 2016 Received in revised form 1 November 2016 Accepted 19 November 2016

Protea-associated fungi are dispersed between flower heads by mites, beetles and possibly birds. For the ophiostomatoid fungus, Knoxdaviesia proteae, these vectors offer regular dispersal between distant floral hosts. Unlike K. proteae, Knoxdaviesia capensis occupies multiple Protea host species. In this study, we aimed to determine whether the generalist K. capensis shares the long-distance dispersal pattern with specialist K. proteae and whether it moves freely between different host species. We evaluated the genetic structure of K. capensis from five populations of a wide-spread host and between sympatric hosts. Twelve K. capensis-specific microsatellite markers were developed and applied to 90 individuals. K. capensis showed high genetic diversity and almost maximal genotypic diversity. All populations were poorly differentiated, indicating the presence of long-distance dispersal. No differentiation could be detected between sympatric host populations, suggesting free dispersal between different hosts. This implies that the beetle and bird vectors that pollinate Protea species show the same non-specific movement. © 2016 Elsevier Ltd and British Mycological Society. All rights reserved.

Corresponding Editor: Karina Engelbrecht Clemmensen Keywords: Dispersal Fynbos Genetic diversity Infructescences Knoxdaviesia Microsatellites Ophiostomatoid Pollination Protea Sympatry

1. Introduction The Greater Cape Floristic region (GCFR; Born et al., 2007) harbors a temperate flora with extreme levels of plant beta- and gamma-diversity and endemism (Linder, 2003). It is characterized by winter rainfall and includes both the Cape Floristic Region (CFR; Manning and Goldblatt, 2012) and the Extra Cape Flora (Snijman, 2013). Within the CFR, the shrubland vegetation known as ‘fynbos’ is dominant and comprises numerous species of the Proteaceae family (Manning and Goldblatt, 2012). Protea is one of the bestknown Proteaceae genera and its members are essential to maintaining diversity and organization of numerous other organism communities within the fynbos. The genus carries its flowers in large inflorescences surrounded by involucral bracts that vary in color, depending on the species and its mode of pollination (Carlson

* Corresponding author. E-mail address: [email protected] (J. Aylward). http://dx.doi.org/10.1016/j.funeco.2016.11.005 1754-5048/© 2016 Elsevier Ltd and British Mycological Society. All rights reserved.

and Holsinger, 2010). In the CFR, the main pollinators of the treelike Protea species include various beetles (e.g. Genuchus and Tricostetha spp.) and birds (e.g. Cape sugarbirds and Orange-breasted sunbirds) (Broekhuysen, 1963; Coetzee and Giliomee, 1985). Many Protea species enclose their seeds in canopy-retained infructescences subsequent to pollination and fertilization, a crucial strategy in fire-prone vegetation like the fynbos (Wright, 1994). Apart from its storage and protective role, infructescences inherently provide micro-niches for a large diversity of organisms such as mites, insects and fungi (Marais and Wingfield, 1994; Roets et al., 2007). Fungal diversity of the fynbos has received little attention in comparison to floral diversity, even though Crous et al. (2006) estimated it contains as many as 200 000 species. Some species of ophiostomatoid fungi in the genera Sporothrix and Knoxdaviesia exclusively occur in the infructescences of serotinous Protea species (Roets et al., 2013), where they are often the dominant inhabitants (Lee et al., 2005; De Beer et al., 2016). As evidenced by the structure of their ascomata, these fungi are adapted to entomochoric dispersal, mediated primarily by mites

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

(Malloch and Blackwell, 1993; Cassar and Blackwell, 1996; Roets et al., 2009a). Mites deposit ascospores of ophiostomatoid fungi in Protea flower heads, where they grow vegetatively until the inflorescence matures into an infructescence. In the infructescences, these fungi switch to sexual reproduction, as evidenced by the abundance of ascomata at this stage (Wingfield et al., 1988; Wingfield and Van Wyk, 1993). Recent analyses of the genomes of two Protea-associated Knoxdaviesia species have revealed that they are heterothallic and that sexual reproduction will, therefore, always create genetic diversity via recombination between two different haploid individuals (Aylward et al., 2016b). Of the nine described Protea-associated members, only Knoxdaviesia proteae has received attention in terms of population genetic structure (Aylward et al., 2014b, 2015b). This species is restricted to a single widespread host, Protea repens, and displays high genetic diversity due to frequent gene flow and outcrossing. Within a stand of P. repens trees, the K. proteae population was found to be panmictic (Aylward et al., 2014b). Between distantly separated P. repens stands, near-panmixia was observed, likely due to long-distance dispersal (Aylward et al., 2015b). An assortment of Protea-visiting insects, such as the Protea beetle pollinators Genuchus hottentottus and Trichostetha fascicularis, are implied as midto long-distance vectors of the mites that carry the ophiostomatoid fungi between Protea trees (Roets et al., 2007, 2009a). Birds have been suggested as additional dispersal agents of spore-carrying mites in other systems (Proctor and Owens, 2000), leading Aylward et al. (2015b) to speculate that birds may also play a part in facilitating long-distance dispersal of K. proteae. Preliminary observations by workers on Protea ecology (T. Rebelo), and Proteaassociated mites (N. Theron) and birds (A. Lee) seem to corroborate this, as Protea-associated mites have been observed on Cape sugarbirds (bluehillescape.blogspot.co.za/2012_11_01_archive.html, accessed July 2016). Unlike K. proteae, Knoxdaviesia capensis colonizes a diversity of Protea hosts often present and flowering in sympatry (Marais and Wingfield, 1994; Roets et al., 2009b). Confirmed hosts include Protea burchellii, Protea coronata, Protea laurifolia, Protea lepidocarpodendron, Protea longifolia, Protea magnifica, Protea neriifolia and P. repens (Wingfield et al., 1993; Marais and Wingfield, 1994; Roets et al., 2005, 2011a; Aylward et al., 2015a). These hosts vary in distribution range, flowering time and flower morphology e all factors that will affect visitation by the Protea pollinators that also act as secondary fungal spore vectors. The movement of these vectors between different Protea species will, therefore, have a direct impact on the dispersal of K. capensis within and between its various hosts. The extent of gene flow in K. capensis will depend on the distance over which its spores are transported as well as whether its vectors show host-specificity (or flower consistency) as Protea pollinators. If they tend to move consistently between inflorescences of the same Protea species only, K. capensis populations on different hosts species will become islands without means of connection. Conversely, pollinators may engage in flights between patches of different species, potentially producing one continuous fungal metapopulation (Levin, 1978). The degree to which the Protea pollinators regularly move between different Protea species that flower in sympatry is, however, currently unknown. Knowledge about the population genetics of the organisms that travel with these Protea-pollinators may shed some light on the possible inter-species visitation of some pollinators. The aim of this study was to investigate the population structure of K. capensis across multiple Protea hosts in the CFR. For this purpose, we developed microsatellite markers specific to K. capensis according to the protocol of Aylward et al. (2014a). Subsequently, these markers were employed to assess population structure at two

29

levels: (1) within a single host, P. coronata, with a widespread, though patchy distribution, and (2) within pairs of sympatric hosts with synchronous flowering times. This strategy enabled us to evaluate both the extent of K. capensis dispersal and to test for its movement between different hosts. 2. Material and methods 2.1. Sampling and fungal isolations Infructescences from different Protea species were sampled from randomly selected trees at seven locations in the Western Cape Province, South Africa (Fig. 1), during January and August 2015. P. neriifolia, P. lepidocarpodendron and P. longifolia infructescences were collected from the Kogelberg Biosphere Reserve, an area where multiple K. capensis hosts occur in sympatry or near sympatry. A fourth species, P. coronata, displays a patchy distribution throughout the Western Cape Province (Rebelo, 2001) and was collected from five sites across its westerly distribution, including a site within the Kogelberg Biosphere Reserve (Fig. 1). Fungal isolations and DNA extraction followed methods described by Aylward et al. (2014a). A sterile needle was used to isolate ascospores of K. capensis from the tip of flask-shaped Knoxdaviesia ascomatal necks in infructescences. Ascospores were germinated at room temperature, sub-cultured onto malt extract agar (MEA; Merck, Wadeville, South Africa) and grown at 25
C. Individual K. capensis strains were purified by sub-culturing a single hyphal tip from water agar (15 g L1 agar) onto fresh MEA. To avoid repeated isolation of the same individual, only a single fungal isolate was maintained per infructescence. To verify the species identity of the fungal isolates, the rRNA Internal Transcribed Spacer (ITS) region was amplified in a subset of representative isolates. The 25 ml PCR reaction consisted of 12 ml KAPA Taq ReadyMix (Kapa Biosystems, Inc., Boston, USA), 2.5 mM additional MgCl2, 0.25 mM of the ITS1F (Larena et al., 1999) and ITS4 (White et al., 1990) primers and ca. 100 ng template DNA. Cycling conditions were 3 min at 95
C followed by 40 cycles of 30 s at 94
C, 1 min at 50
C and 50 s at 72
C. The final extension was 7 min at 72
C. PCR products were sequenced at the Central Analytical Facility (CAF), Stellenbosch University, using ITS1F and the BigDye Cycle sequencing kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions, after which BLAST (Basic Local Alignment Search Tool, Camacho et al., 2009) searches were performed on the NCBI nucleotide data base (www.ncbi.nlm.nih.gov). 2.2. Microsatellite identification and amplification Microsatellites were extracted from the K. capensis genome (GenBank accession: LNGK00000000.1; Aylward et al., 2016a) using the program Msatfinder 2.0.9 (Thurston and Field, 2005). The default search engine and search parameters of this program were used to identify perfect tandem repeat microsatellites. Microsoft Office Excel 2010 (Microsoft Corp., Redmond, WA, USA) was used to filter and analyze the data. Only microsatellites consisting of more than five repeat units were considered, since microsatellites with higher-than-average repeat numbers are more likely to be polymorphic (Dettman and Taylor, 2004; Dutech et al., 2007). The relevant microsatellites were subjected to BLAST searches using BLASTtx (www.ncbi.nlm.nih.gov), the least conservative engine, to ensure that the identified tri- and hex-nucleotide loci were not located in protein-coding regions. Twenty microsatellite loci were initially selected and primers that flank these loci were designed with Primer3Plus 2.4.0 (Untergasser et al., 2007). Sequence analysis was performed in BioEdit 7.2.5 (Hall, 1999). Polymorphism of the markers was

30

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

Fig. 1. Sampling localities of Protea host species in the Western Cape Province, South Africa. Five P. coronata populations (circles) and two sympatric populations of P. neriifolia and P. lepidocarpodendron (diamond), and P. neriifolia and P. longifolia (triangle) were collected. The Kogelberg Biosphere Reserve is indicated in a darker shade of grey.

assessed by amplifying and sequencing all loci in four representative isolates from the K. capensis populations, using the same PCR and sequencing protocols as for the ITS amplification described above. Polymorphic microsatellite loci were divided into two multiplex reactions and the forward primer for each locus was labeled with one of the four ABI fluorescent dyes (Thermo Fisher Scientific Inc., Wilmington, USA). Subsequently, 12 microsatellite markers were amplified on each K. capensis isolate in two panels (Table 1). Multiplex PCRs were performed with the KAPA2G Fast Multiplex PCR Kit (Kapa Biosystems, Inc., Boston, USA). The 25 ml reactions contained 12.5 ml KAPA2G, 1 mM additional MgCl2, 30 ng DNA and a variable concentration of primers (Table 1). PCR conditions were 3 min at 95
C followed by 30 cycles of: 15 s at 95
C, 30 s at 55
C and 20 s at 72
C. Final extension was 7 min at 72
C. The products were resolved on a 96-capillary Applied Biosystems 3730xl DNA Analyzer and sized using a GeneScan 500 LIZ Size Standard (GS500, Applied Biosystems, Carlsbad, CA). GeneMarker 2.6.4 (Softgenetics LLC, State College, PA, USA) was used for allele calling and the estimated size of every fragment was inspected manually. 2.3. Data analyses Microsatellites may be unreliable due to mutations at primerbinding sites that cause insufficient annealing for fragment amplification, often called null alleles, and linkage disequilibrium. Null alleles are obvious in haploid organisms such as K. capensis, but linkage between the microsatellite loci was tested by mapping the loci back onto the 29 scaffolds of the K. capensis genome to ensure that they do not lie close together. Pairwise linkage disequilibrium was calculated between each of the loci using Multilocus 1.3b (Agapow and Burt, 2001). Descriptive diversity indexes were computed using GenAlEx 6.5.02 (Peakall and Smouse, 2006, 2012). The genetic diversity was

described by computing the effective number of alleles (Ne; Kimura and Crow, 1964), number of private alleles (Np), number of multilocus haplotypes and Nei's unbiased estimate of expected heterozygosity (HE). The latter is the conventional measure of genetic diversity and describes the probability that two randomly sampled alleles will be different (Nei, 1973). Genotypic diversity was calculated as the number of distinct microsatellite multilocus genotypes occurring x times (Stoddart and Taylor, 1988) and this index was subsequently used to calculate the maximum percentage of geno^ McDonald et al., 1994). typic diversity (G; Isolates sampled from each Protea host at individual localities were assumed to represent subpopulations of K. capensis, while isolates from a single host form a population (Fig. 2). SMOGD 1.2.5 (Crawford, 2010) was used to assess population differentiation by calculating the diversity between different populations. Subpopulations were also grouped according to their geographic location, i.e. whether they originate from the Kogelberg Biosphere Reserve and whether they are sympatric (Fig. 2). The effective number of subpopulations (DST) is represented by the diversity between subpopulations, and is the ratio of true diversity (DT; effective number of alleles in the total population) to the withinsubpopulation diversity (DS). The proportion of diversity contained within the average subpopulation is the inverse of this ratio, DS/DT, a measure of similarity that will decrease with the increase of differentiation (Jost, 2008). Theta (q), an estimate analogous to the conventional measure of population differentiation, FST (Weir and Cockerham, 1984), was calculated with Multilocus. Arlequin 3.5.2.2 (Excoffier and Lischer, 2010) was used to calculate an analysis of molecular variance (AMOVA) between the two sites that had sympatric Protea species. Significance was tested by using a FST-like distance matrix and 10 000 permutations. The number of genetic clusters (K) in the metapopulation was estimated using STRUCTURE 3.2.4 (Pritchard et al., 2000; Falush et al., 2003; Hubisz et al., 2009), a program that employs a

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

31

Table 1 Polymorphic microsatellite markers developed for Knoxdaviesia capensis. Locus

Motif

Size range (bp)

Primers

Sequence

Fluorescent tag

Primer concentration (nm)

Reaction 1 Kc-1

(caa)8

203e275

(accc)7

131e155

Kc-8

(cgtt)7

173e215

Kc-9

(ctgt)6

151e212

Kc-11

(aagcag)5

202e249

Kc-12

(gagtcg)6

226e255

AGAGTCCGTCCATAGCAAGC GGCAACAGTCAGTACCAACG GTTTGCCATTCTCAGCTTCC CCTGCTTAGGTGACGGTTGT GCTGCTCCTTTTCCATTCTG GCGTAAGCCACAGACAGACA CGTGGCTAAGGGTAGATTGC TAGTAAACCCGCCGAGAATG GAAACACCCAAGTCCGTCTC CATGTGCGGATTTTCATCTG CTTGCTGCCCTTCTTAGTCG GGATTCGTCCGACTCTAGCTC

PET

Kc-7

Kc1-1 Kc1-2 Kc7-1 Kc7-2 Kc8-1 Kc8-2 Kc9-1 Kc9-2 Kc11-1 Kc11-2 Kc12-1 Kc12-2

40 40 40 30 20 40 20 30 40 30 40 30

Reaction 2 Kc-2

(cat)9

199e230

(aggg)5

202e258

Kc-4

(tcca)5

158e176

Kc-5

(cagc)6

220e264

Kc-6

(ggct)6

177e193

Kc-10

(ctgtg)5

230e262

CACCCTCGTAGATATCCAGGTC CAAGTACATTTGGGGCGACT ACCCATTCACCGTCAACAAC CCTGCCTTCGTTTCTCAAAG CATGCGTGAATCACAGTTTTG GCACTGCAGTAGTGGGTGAA CTTGATGCCCTCGTGTAGGT GCTCTTCTTCCCTGCTTTCA GACCTGCAATGGTTCCTTTC GCCGGATCACATTCTCTTGT GTCTCACAGCAAGCAACCAG CCTGCAGACAGTACGCAGAC

NED

Kc-3

Kc2-1 Kc2-2 Kc3-1 Kc3-2 Kc4-1 Kc4-2 Kc5-1 Kc5-2 Kc6-1 Kc6-2 Kc10-1 Kc10-2

PET NED 6-FAM VIC 6-FAM

20 20 12 20 60 40 40 40 40 40 40 30

VIC VIC PET PET 6-FAM

2011) was used to construct a minimum spanning tree and minimum spanning network. The null hypothesis of random recombination was tested by calculating ṝd (Brown and Weir, 1983) in Multilocus (Agapow and Burt, 2001) and comparing the observed value to the values calculated for 1000 random datasets.

3. Results

Fig. 2. Population hierarchy of Knoxdaviesia capensis isolates for testing population differentiation. Isolates sampled from different sampling localities and hosts are considered subpopulations. Subpopulations from a single Protea host form a population (solid circles). Subpopulations were also grouped according to geographic location (dashed squares).

Bayesian, model-based approach to assign individuals to clusters based on their allelic frequencies. Ten replicate runs were conducted for K values ranging from one to 10, using 100 000 burn-in and 250 000 Markov Chain Monte Carlo repetitions. Runs were conducted assuming correlated allele frequencies with the admixture as well as the no admixture model (independently). Runs were initially conducted without supplying sampling information. Thereafter, the information was included with the LOCPRIOR model. The optimal number of clusters was determined by computing L(K) and DK with the online platform STRUCTUREHARVESTER (Earl and von Holdt, 2012). The molecular-variance parsimony technique was used to investigate the relatedness of K. capensis individuals. Pairwise distances between the microsatellite haplotypes were calculated in Arlequin (Excoffier et al., 1992). HapStar 0.7 (Teacher and Griffiths,

Ninety K. capensis isolates were obtained from the nine sites (Table 2). Sample sizes were inconsistent between subpopulations even after resampling some subpopulations in an attempt to increase the number of isolates. This was partially due to the small size (ca. 2e10 individuals) of the host populations sampled and the difficulty of isolating K. capensis from infructescences. Sixteen of the designed primer pairs amplified in K. capensis and 12 generated polymorphic fragments (Table 1). Linkage disequilibrium tests revealed no significant pairwise linkage at a critical value of P ¼ 0.05. Furthermore, annotation of the microsatellites on the genome revealed that eight of the microsatellites occur on different scaffolds. The remaining four were located on scaffolds already occupied by one of the eight microsatellites. These were, however, separated by at least 1.2 million bases (Table S1), far more than the

Table 2 Knoxdaviesia capensis isolates sampled in the Western Cape Province, South Africa. Sampling locality

Ataraxia Du Toits Kloof Greyton Kleinmond* Helderberg Betty's Bay 1* Betty's Bay 2* Kogelberg 1* Kogelberg 2*

Protea species

Protea coronata Protea coronata Protea coronata Protea coronata Protea coronata P. neriifolia P. lepidocarpodendron P. longifolia P. neriifolia

Isolates

16 10 9 9 9 8 8 10 11

*Located within the Kogelberg Biosphere Reserve.

Sampling coordinates Latitude

Longitude

34.33493 33.69525 34.03848 34.33205 34.03194 34.35495 34.35495 34.28858 34.28858

19.312217 19.0896 19.6141 19.007383 18.876389 18.90135 18.90135 19.109683 19.109683

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

80e185 kb estimated to be necessary for meiotic recombination in Saccharomyces cerevisiae (Kaback and Guacci, 1992). Nine of the loci had a low proportion of null alleles (between zero and 1.12%) and these were treated as missing data in subsequent analyses. A high percentage of null alleles were observed at three loci (Kc-5, Kc-9, and Kc-11) and these were excluded from further analyses. The exclusion of these three loci did not significantly affect other diversity indices (Table 3). Furthermore, a plot of the number of sampled loci against the number of multilocus genotypes and the genotypic diversity (calculated in Multilocus) reaches a plateau (Fig. 3; Table S2), indicating that the remaining nine loci are sufficient to capture the diversity of the metapopulation.

1

0.8

Mean value

32

0.6

0.4

0.2

0

3.1. Genetic diversity

0

A total of 107 alleles were detected across the nine loci, with an average of 11.90 ± 2.25 alleles per locus. Allele frequencies ranged from 0.01 to 0.80. The expected heterozygosity of the entire population across the nine loci was 0.64 ± 0.08 and the genetic diversity as described by the number of effective alleles (Ne) was 4.49 ± 1.15 (Table 3). Eighty-seven unique haplotypes were observed among the 90 K. capensis isolates, resulting in a high ^ ¼ 95.10%, G ¼ 82.74). Three pairs maximum genotypic diversity (G of isolates, therefore, shared haplotypes, but identical pairs originated from the same location. Two originated from the Ataraxia (P. coronata) subpopulation and the third from the Kogelberg 1 (P. longifolia) subpopulation. The duplicate haplotypes were excluded from subsequent calculations of population differentiation and structure. There were no apparent differences between the diversity measures and, therefore, the genetic compositions, of isolates from different localities, although Kogelberg 1 had fewer private alleles than the other subpopulations (Fig. 4).

1

2

3

4

5

6

7

8

9

Number of loci Genotypic diversity

Number of genotypes/100

Fig. 3. Genotypic diversity and the mean number of genotypes plotted against the number of sampled loci. Both graphs reach a plateau.

3.2. Population differentiation Population differentiation (Table 4) was calculated between subpopulations within Protea hosts, between sympatric subpopulations within the Kogelberg Biosphere Reserve, between all sampling localities (subpopulations) and between all hosts (populations). For all cases, DS/DT indicates that the average subpopulation contains 92%e96% of the genetic diversity and, conversely, DST suggests that each population effectively comprises only slightly more than one subpopulation. Pooling all subpopulations would, therefore, not significantly increase the observed diversity, but only

Fig. 4. Comparison of diversity indices among the nine sampling localities. Bars represent the mean and standard error of the mean across the nine microsatellite loci. HE ¼ expected heterozygosity, Np ¼ private alleles, Ne ¼ effective alleles, Na ¼ total alleles.

add the few private alleles present in each. Theta (q) consistently detected low (0.03e0.16), but highly significant differentiation

Table 3 Diversity indices of Knoxdaviesia capensis calculated from the 12 microsatellite loci. Locus

Naa

Null alleles (%)

Neb

HEc

Kc-1 Kc-2 Kc-3 Kc-4 Kc-5 Kc-6 Kc-7 Kc-8 Kc-9 Kc-10 Kc-11 Kc-12 Mean ± SEMd Excluding Kc-5, Kc-9, Kc-11

25 16 8 6 10 8 7 19 13 12 17 6 12.25 ± 1.72 11.90 ± 2.25

1.12 0.00 0.00 0.00 58.43 1.12 1.12 2.25 22.47 0.00 21.35 0.00 8.99 ± 0.05 0.62 ± 0.00

8.490 8.635 1.685 1.965 5.730 1.764 1.551 9.778 4.667 3.453 7.673 3.068 4.87 ± 0.89 4.49 ± 1.15

0.892 0.894 0.411 0.497 0.848 0.438 0.359 0.908 0.797 0.718 0.882 0.682 0.69 ± 0.06 0.64 ± 0.08

a b c d

Na ¼ number of alleles. Ne ¼ number of effective alleles (Kimura and Crow, 1964). HE ¼ Nei's unbiased expected heterozygosity (Nei, 1973). SEM: Standard error of the mean.

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

33

Table 4 Descriptive measures of population differentiation for the different Knoxdaviesia capensis subpopulation groupings. Mean values and the standard error of the mean across 9 loci are reported.

~a N

DSTb DS/DTc qd

P. coronata

P. neriifolia

Kogelberg Biosphere Reserve

Sympatric (1) Kogelberg

Sympatric (2) Betty's Bay

All subpopulations

All populations (hosts)

9.91 1.08 ± 0.02 0.93 ± 0.02 0.06**

9.26 1.09 ± 0.04 0.93 ± 0.03 0.16*

8.88 1.10 ± 0.03 0.92 ± 0.02 0.08**

9.90 1.04 ± 0.02 0.96 ± 0.02 0.02

8.00 1.05 ± 0.02 0.96 ± 0.02 0.03

9.40 1.10 ± 0.03 0.91 ± 0.02 0.07**

12.97 1.05 ± 0.01 0.95 ± 0.01 0.02*

*P ¼ 0.002; **P < 0.001. a ~ N ¼ harmonic mean of the sample sizes. b DST ¼ effective number of subpopulations (diversity between subpopulations). c DS/DT ¼ proportion of diversity in a subpopulation. d q ¼ conventional measure of relative differentiation.

(P < 0.001), except at the two sympatric sites where differentiation was not significant. The two undifferentiated sympatric groupings also showed the greatest average subpopulation genetic diversity, supporting their cohesion. The AMOVA (Table 5) indicated that more than 86% of the distributed molecular variation was found between individuals within the different hosts at the sympatric sites. Fixation indices were only significant at the highest population hierarchy (comparing fungi from the plant hosts within the total population), congruent with the differentiation indices that show low differentiation between sympatric sites, but not within. 3.3. Population structure The STRUCTURE runs without sampling information did not indicate separate clustering of the isolates. Assuming the admixture model, L(K) was greatest at K ¼ 1. In the no admixture model, L(K) had its greatest value at K > 1, but the histogram shows that each genetic cluster contains similar proportions of each isolate's genome (Fig. S1). When sampling localities were considered, however, L(K) and DK were greatest at K ¼ 2 in both the admixture and the no admixture model. The histograms indicate that some individuals have a proportion of diversity that is not contained in the rest of the sampled isolates. Nevertheless, the majority of each individual's genetic structure is attributed to one cluster and the second cluster is only detected once sampling information is supplied. The STRUCTURE results are, therefore, congruent with the population differentiation statistics that indicate low genetic differentiation between sampling hosts and localities. Three subpopulations, Ataraxia and the two Betty's Bay sites, had rd values that suggest a significant departure from random recombination (Table S3). The remaining six subpopulations have low rd values indicating that individuals are randomly recombining. Random recombination and clonal populations are considered as the two extremes, with an rd value of zero indicating total recombination and a value of one indicating clonal populations. The minimum spanning tree (Fig. 5) illustrates that the relationship between the K. capensis genotypes is not defined by host population or sampling locality. Extension of this tree into a minimum spanning network (not shown) generates numerous interconnections between genotypes that portray the diversity and cohesion of the metapopulation. Due to the diversity of genotypes

and the lack of structure in these subpopulations, we conclude that recombination is more prevalent in all the K. capensis subpopulations than clonal reproduction, although the latter cannot be completely disregarded. 4. Discussion Although the floral and faunal diversity of the CFR has been well-studied, much less is known about the fungal diversity of this iconic region (Crous et al., 2006). A large gap also exists in knowledge on the biology of these microbes. The first research on the genetic structure and diversity of any fungus in the CFR was conducted by Aylward et al. (2014b; 2015a). Consequently, the present study aimed to broaden the scope of this knowledge by investigating the population genetics of K. capensis, a species closely related to the previously studied K. proteae. Twelve microsatellite markers were developed for K. capensis and nine were effectively applied to investigate the genetic diversity and structure of this arthropod-vectored species. Microsatellite markers previously developed for K. proteae were poorly transferable to K. capensis (Aylward et al., 2014a) and the markers developed in this study, therefore, enabled comparison of genetic diversity between these two CFR species. The genotypic diversity of ^ ¼ 95.10%) compared to K. capensis was found to be very high (G those reported in previous studies of ophiostomatoid fungi (Zhou ^ ¼ 46.5%; Nkuekam et al., 2009; G ^ ¼ 34.8%). Howet al., 2007; G ever, studies considering flower-living yeasts in the genus Metschnikowia, have also observed that the majority of multilocus genotypes are unique (Wardlaw et al., 2009; Herrera et al., 2011). This diversity is despite prevalent clonal reproduction, implying that the flower environment imposes diversifying selection (Herrera et al., 2011). The high genotypic diversity in K. capensis is the result of recombination in infructescences mediated by a heterothallic reproductive strategy (Aylward et al., 2016b). This diversity is also comparable to that found by Aylward et al. (2014b) for K. proteae ^ ¼ 97.9%). The diversity observed in both of these Protea-associ(G ated Knoxdaviesia species implies substantial effective population sizes (Charlesworth, 2009). Although this study has not considered the genetic diversity within infructescences, ascospore droplets produced by Knoxdaviesia carry numerous genetically novel

Table 5 Analysis of Molecular Variance (AMOVA) results indicating the variance attributable to each hierarchy between the two sites containing sympatric Knoxdaviesia capensis host populations. Variance component

dfa

Variance

%total

P

Fixation

FCT ¼ within sympatric sites (groups) FSC ¼ between sympatric sites FST ¼ among all host species (populations)

1 2 33

0.33 0.09 2.62

10.97 2.82 86.22

>0.05 >0.05 0

0.11 0.03 0.14

a

Degrees of freedom.

̅

34

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

Fig. 5. Minimum spanning tree depicting the relationship between the 87 Knoxdaviesia capensis genotypes (colored circles) identified in this study. Black circles represent mutation events. Protea hosts are distinguished by color. The different sampling localities of P. coronata and P. neriifolia are indicated with numbers that follow the sequence of populations as presented in Table 2.

individuals (Aylward et al., 2016b). These droplets are picked up by the primary mite vectors that may disperse them to younger flower heads on the same plant or may attach to beetles and birds for dispersal between plants (Roets et al., 2009a). Numerous mites typically occur within one flower head (Theron et al., 2012), suggesting that they would seed multiple Knoxdaviesia individuals. Continuous recombination and dispersal, therefore, underlies the diversity of these species. 4.1. Cohesive K. capensis subpopulations across P. coronata The low population differentiation (q ¼ 0.07) on the P. coronata host indicates that K. capensis commonly disperses between distant host populations. This is in contrast to our expectation of geographic barriers and discontinuous host populations imposing limitations to extensive dispersal. The high level of genetic diversity that is shared between the K. capensis subpopulations on P. coronata highlights the essential role of pollinators as medium-to long-distance dispersers of Protea-associated ophiostomatoid fungi. Additionally, the generalist nature of K. capensis may enable it to maintain dispersal over great distances by interchanging between Protea host species. Protea host species interspersed between P. coronata stands may, therefore, act as fungal ‘stepping stones’. However, host-switching is unlikely to be ubiquitous as it will be dependent on various factors, including host availability, flower phenology, vector identity and vector flower consistency. The beetles that vector ophiostomatoid fungi are well-known Protea-flower visitors and pollinators of these plants (Roets et al., 2007, 2009a; Steenhuisen and Johnson, 2012) and consequently visit numerous inflorescences, transmitting fungus-carrying mites

and pollen from one plant to another. Beetles are believed to be primarily responsible for shared diversity between K. proteae populations in close proximity (Aylward et al., 2014b). The low differentiation between distantly separated subpopulations, such as these K. capensis subpopulations on P. coronata, may be difficult to attribute to beetles alone. A possible role for birds to facilitate gene flow of K. proteae across greater distances has been implicated for P. repens (Aylward et al., 2015b; Theron, pers. com.) and may also hold for K. capensis on P. coronata. The primary bird pollinator of P. coronata is the Cape sugarbird (Promerops cafer), which is known to travel hundreds of kilometers each year (Calf et al., 2003; Hockey et al., 2005). If birds indeed carry spore-laden mites and are in this way involved in the dispersal of K. capensis, we expect the relatively low estimates of genetic differentiation found here to be due to the movement of birds maintaining significant gene flow between K. capensis subpopulations. 4.2. Free dispersal between sympatric hosts The lack of population differentiation between K. capensis subpopulations at sympatric sites indicates that K. capensis moves freely between different hosts. All the Protea species studied here are primarily pollinated by Cape sugarbirds and beetles (Littlejohn, 2001; Calf et al., 2003), the same organisms implicated in fungal spore transmission. Even though no information currently exists on the flower consistency of Protea pollinators, our study provides compelling evidence that they often move between sympatric species. In addition, our results may also indicate that pollination facilitation is likely to occur among different Protea species when they have synchronous flowering times (Rathcke and Lacey, 1985;

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

Rebelo, 2001). However, in terms of Protea population dynamics, this could lead to increased hybridisation between species, a phenomenon that is often observed in nature (Coetzee and Littlejohn, 2001). Protea-pollinating beetles and Cape sugarbirds are also known to pollinate P. repens, yet K. capensis is very rarely isolated from this host (Roets et al., 2011a; Aylward et al., 2015a). In addition, the mite species involved with the transmission of the fungal spores are also known from both P. repens and the hosts of K. capensis (Roets et al., 2011b). Differences in host preferences of these fungi may be attributed to the unique phylogenetic position of P. repens relative to the other Knoxdaviesia host plants (Valente et al., 2010). All the hosts of K. capensis are fairly closely related and reside in the ‘bearded sugarbush’ group (sensu Rebelo, 2001), while P. repens, the only known host for K. proteae, is fairly distantly related to these and groups within the ‘true sugarbush’ group. Again, the flower consistency of Protea pollinators in sympatric populations of P. repens and other Protea species is unknown, but the presence of K. capensis on this host, albeit in very low numbers, indicates some movement of the potential pollinators and their mites between these hosts. The preferences of K. capensis and K. proteae towards different hosts may consequently be attributed to traits particular to these different fungal species. This implies that K. capensis is not able to thrive in conditions within P. repens infructescences and K. proteae will also not thrive in bearded Protea hosts due to adaptations to host specific parameters, such as different host chemistries, rather than limitations posed by dispersal. Roets et al. (2011a) investigated the role of host chemistry in shaping these communities, but concluded that it only partially explains host exclusivity and that other factors such as differential competitive abilities of the fungi on different hosts may be important. 5. Conclusions K. capensis in the CFR harbors remarkable genetic and genotypic diversity. A high level of gene flow coupled with outcrossing seems to maintain the high level of diversity and prevent population differentiation, consequently limiting the potential effects of genetic drift. The role of beetles and possibly birds in the medium-to long-distance dispersal of K. capensis on P. coronata would be primarily responsible for the observed high level of shared diversity. Sympatric K. capensis subpopulations occurring on different Protea hosts also exhibited high genetic and genotypic diversity indicating that Protea pollinators likely do not discriminate between the different Protea species studied. The lack of population structure observed within CFR populations of K. capensis and K. proteae prompts further questions on the population genetic structure of ophiostomatoid fungi over the entire distribution range of Protea species in Africa. Moreover, the role of avian vectors in longdistance dispersal of Knoxdaviesia and Sporothrix has not yet been elucidated and remains to be explored. Acknowledgments We thank the National Research Foundation (NRF) and the Department of Science and Technology (DST)-NRF Centre of Excellence in Tree Health Biotechnology (CTHB) for financial support. We are grateful to the Western Cape Nature Conservation Board for supplying the necessary collection permits and to two anonymous reviewers for their comments on this manuscript. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.funeco.2016.11.005.

35

References Agapow, P.-M., Burt, A., 2001. Indices of multilocus linkage disequilibrium. Mol. Ecol. Notes 1, 101e102. Aylward, J., Dreyer, L.L., Steenkamp, E.T., Wingfield, M.J., Roets, F., 2014a. Development of polymorphic microsatellite markers for the genetic characterisation of Knoxdaviesia proteae (Ascomycota: Microascales) using ISSR-PCR and pyrosequencing. Mycol. Prog. 13, 439e444. Aylward, J., Dreyer, L.L., Steenkamp, E.T., Wingfield, M.J., Roets, F., 2014b. Panmixia defines the genetic diversity of a unique arthropod-dispersed fungus specific to Protea flowers. Ecol. Evol. 4, 3444e3455. Aylward, J., Dreyer, L.L., Steenkamp, E.T., Wingfield, M.J., Roets, F., 2015a. Knoxdaviesia proteae is not the only Knoxdaviesia-symbiont of Protea repens. IMA Fungus 6, 471e476. Aylward, J., Dreyer, L.L., Steenkamp, E.T., Wingfield, M.J., Roets, F., 2015b. Longdistance dispersal and recolonization of a fire-destroyed niche by a miteassociated fungus. Fungal Biol. 119, 245e256. Aylward, J., Steenkamp, E.T., Dreyer, L.L., Roets, F., Wingfield, B.D., Wingfield, M.J., 2016a. Genome sequences of Knoxdaviesia capensis and K. proteae (fungi: Ascomycota) from Protea trees in South Africa. Stand. Genomic Sci. 11, 1e7. Aylward, J., Steenkamp, E.T., Dreyer, L.L., Roets, F., Wingfield, M.J., Wingfield, B.D., 2016b. Genetic basis for high population diversity in Protea-associated Knoxdaviesia. Fungal Genet. Biol. 96, 47e57. Born, J., Linder, H., Desmet, P., 2007. The greater cape floristic region. J. Biogeogr. 34, 147e162. Broekhuysen, G., 1963. The breeding biology of the orange-breasted sunbird Anthobaphes vzolacea (Linnaeus). Ostrich 34, 187e234. Brown, A.H.D., Weir, B.S., 1983. Measuring genetic variability in plant populations. In: Tanksley, S.D., Orton, T.J. (Eds.), Isozymes in Plant Genetics and Breeding, Part A. Elsevier Science Publishers, Amsterdam, pp. 219e239. Calf, K., Downs, C., Cherry, M., 2003. Territoriality and breeding success in the Cape sugarbird (Promerops cafer). Emu 103, 29e35. Camacho, C., Coulouris, G., Avagyan, V., Ma, N., Papadopoulos, J., Bealer, K., Madden, T.L., 2009. BLASTþ: architecture and applications. BMC Bioinform. 10, 421-421. Carlson, J.E., Holsinger, K.E., 2010. Natural selection on inflorescence color polymorphisms in wild Protea populations: the role of pollinators, seed predators, and intertrait correlations. Am. J. Bot. 97, 934e944. Cassar, S., Blackwell, M., 1996. Convergent origins of Ambrosia fungi. Mycologia 88, 596e601. Charlesworth, B., 2009. Effective population size and patterns of molecular evolution and variation. Nat. Rev. Genet. 10, 195e205. Coetzee, J.H., Giliomee, J.H., 1985. Insects in association with the inflorescence of Protea repens (L.) (Proteaceae) and their role in pollination. J. Entomol. Soc. S. Afr. 48, 303e314. Coetzee, J.H., Littlejohn, G.M., 2001. Protea: a floricultural crop from the Cape floristic Kingdom. Hortic. Rev. 26, 1e48. Crawford, N.G., 2010. SMOGD: software for the measurement of genetic diversity. Mol. Ecol. Resour. 10, 556e557. Crous, P.W., Rong, I.H., Wood, A., Lee, S., Glen, H., Botha, W., Slippers, B., de Beer, W.Z., Wingfield, M.J., Hawksworth, D.L., 2006. How many species of fungi are there at the tip of Africa? Stud. Mycol. 55, 13e33. De Beer, Z.W., Duong, T.A., Wingfield, M.J., 2016. The divorce of Sporothrix and Ophiostoma: solution to a problematic relationship. Stud. Mycol. 83, 165e191. Dettman, J.R., Taylor, J.W., 2004. Mutation and evolution of microsatellite loci in Neurospora. Genetics 168, 1231e1248. s, B., Carlier, J., Tharreau, D., Dutech, C., Enjalbert, J., Fournier, E., Delmotte, F., Barre Giraud, T., 2007. Challenges of microsatellite isolation in fungi. Fungal Genet. Biol. 44, 933e949. Earl, D.A., von Holdt, B.M., 2012. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 4, 359e361. Excoffier, L., Lischer, H.E., 2010. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol. Ecol. Resour. 10, 564e567. Excoffier, L., Smouse, P.E., Quattro, J.M., 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131, 479e491. Falush, D., Stephens, M., Pritchard, J.K., 2003. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics 164, 1567e1587. Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95e98. Herrera, C.M., Pozo, M.I., Bazaga, P., 2011. Clonality, genetic diversity and support for the diversifying selection hypothesis in natural populations of a flower-living yeast. Mol. Ecol. 20, 4395e4407. Hockey, P., Dean, W., Ryan, P., 2005. Roberts Birds of Southern Africa, seventh ed. The Trustees of the South African Bird Book Fund, Cape Town. Hubisz, M.J., Falush, D., Stephens, M., Pritchard, J.K., 2009. Inferring weak population structure with the assistance of sample group information. Mol. Ecol. Resour. 9, 1322e1332. Jost, L., 2008. GST and its relatives do not measure differentiation. Mol. Ecol. 17, 4015e4026. Kaback, D.B., Guacci, V., 1992. Chromosome size-dependent control of meiotic

36

J. Aylward et al. / Fungal Ecology 26 (2017) 28e36

recombination. Science 256, 228. Kimura, M., Crow, J.F., 1964. The number of alleles that can be maintained in a finite population. Genetics 49, 725e738. Larena, I., Salazar, O., Gonz
alez, V., Juli
an, M.a.C., Rubio, V., 1999. Design of a primer for ribosomal DNA internal transcribed spacer with enhanced specificity for ascomycetes. J. Biotechnol. 75, 187e194. Lee, S., Roets, F., Crous, P.W., 2005. Biodiversity of saprobic microfungi associated with the infructescences of Protea species in South Africa. Fungal Divers. 19, 69e78. Levin, D., 1978. Pollinator Behaviour and the Breeding Structure of Plant Populations. Academic Press, London. Linder, H.P., 2003. The radiation of the Cape flora, southern Africa. Biol. Rev. 78, 597e638. Littlejohn, G., 2001. The challenges of breeding wild flower cultivars for use in commercial floriculture: African Proteaceae. In: XX International Eucarpia Symposium, Section Ornamentals, Strategies for New Ornamentals-part I 552, pp. 25e38. Malloch, D., Blackwell, M., 1993. Dispersal biology of the ophiostomatoid fungi. In: Wingfield, M.J., Seifert, K.A., Webber, J.F. (Eds.), Ceratocystis and Ophiostoma: Taxonomy, Ecology and Pathology. APS Press, St. Paul, pp. 195e206. Manning, J., Goldblatt, P., 2012. Plants of the Greater Cape Floristic Region. 1: the Core Cape Flora, Strelitzia 29. South African National Biodiversity Institute, Pretoria. Marais, G.J., Wingfield, M.J., 1994. Fungi associated with infructescences of Protea species in South Africa, including a new species of Ophiostoma. Mycol. Res. 98, 369e374. McDonald, B.A., Miles, J., Nelson, L.R., Pettway, R.E., 1994. Genetic variability in nuclear DNA in field populations of Stagonospora nodorum. Phytopathology 84, 250e255. Nei, M., 1973. Analysis of gene diversity in subdivided populations. Proc. Natl. Acad. Sci. 70, 3321e3323. Nkuekam, G.K., Barnes, I., Wingfield, M.J., Roux, J., 2009. Distribution and population diversity of Ceratocystis pirilliformis in South Africa. Mycologia 101, 17e25. Peakall, R., Smouse, P.E., 2006. GenAlEx 6: genetic analysis in Excel. Population genetic software for teaching and research. Mol. Ecol. Notes 6, 288e295. Peakall, R., Smouse, P.E., 2012. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research e an update. Bioinformatics 28, 2537e2539. Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure using multilocus genotype data. Genetics 155, 945e959. Proctor, H., Owens, I., 2000. Mites and birds: diversity, parasitism and coevolution. Trends Ecol. Evol. 15, 358e364. Rathcke, B., Lacey, E.P., 1985. Phenological patterns of terrestrial plants. Annu. Rev. Ecol. Syst. 179e214. Rebelo, T., 2001. Proteas: a Field Guide to the Proteas of Southern Africa. Fernwood Press, Vlaeberg, South Africa. Roets, F., Crous, P.W., Wingfield, M.J., Dreyer, L.L., 2009a. Mite-mediated hyperphoretic dispersal of ophiostoma spp. from the infructescences of South African Protea spp. Environ. Entomol. 28, 143e152. Roets, F., Dreyer, L.L., Crous, P.W., 2005. Seasonal trends in colonisation of Protea infructescences by Gondwanamyces and Ophiostoma spp. South Afr. J. Bot. 71, 307e311. Roets, F., Theron, N., Wingfield, M.J., Dreyer, L.L., 2011a. Biotic and abiotic constraints that facilitate host exclusivity of Gondwanamyces and Ophiostoma on Protea. Fungal Biol. 116, 49e61.

Roets, F., Wingfield, M.J., Crous, P.W., Dreyer, L.L., 2007. Discovery of fungus-mite Mutualism in a unique niche. Environ. Entomol. 36, 1226e1237. Roets, F., Wingfield, M.J., Crous, P.W., Dreyer, L.L., 2009b. Fungal radiation in the Cape floristic region: an analysis based on gondwanamyces and ophiostoma. Mol. Phylogen. Evol. 51, 111e119. Roets, F., Wingfield, M.J., Crous, P.W., Dreyer, L.L., 2013. Taxonomy and ecology of ophiostomatoid fungi associated with Protea infructescences. In: Seifert, K.A., de Beer, Z.W., Wingfield, M.J. (Eds.), Ophiostomatoid Fungi: Expanding Frontiers. CBS Biodiversity Series. Utrecht, The Netherlands, pp. 177e187. Roets, F., Wingfield, M.J., Wingfield, B.D., Dreyer, L.L., 2011b. Mites are the most common vectors of the fungus Gondwanamyces proteae in Protea infructescences. Fungal Biol. 115, 343e350. Snijman, D.A., 2013. Plants of the Greater Cape Floristic Region. 2: the Extra Cape Flora, Strelitzia 30. South African National Biodiversity Institute, Pretoria. Steenhuisen, S.-L., Johnson, S., 2012. Evidence for beetle pollination in the African grassland sugarbushes (Protea: Proteaceae). Plant Syst. Evol. 1e13. Stoddart, J.A., Taylor, J.F., 1988. Genotypic diversity: estimation and prediction in samples. Genetics 118, 705e711. Teacher, A.G.F., Griffiths, D.J., 2011. HapStar: automated haplotype network layout and visualization. Mol. Ecol. Resour. 11, 151e153. Theron, N., Roets, F., Dreyer, L.L., Esler, K.J., Ueckermann, E.A., 2012. A new genus and eight new species of Tydeoidea (Acari: Trombidiformes) from Protea species in South Africa. Int. J. Acarol. 38, 257e273. Thurston, M.I., Field, D., 2005. Msatfinder: Detection and Characterisation of Microsatellites. CEH Oxford, Mansfield Road, Oxford OX1 3SR. http://dx.doi.org/ 10.5281/zenodo.11066. Available from: https://github.com/knirirr/Msatfinder. Untergasser, A., Nijveen, H., Rao, X., Bisseling, T., Geurts, R., Leunissen, J.A.M., 2007. Primer3Plus, an enhanced web interface to Primer3. Nucleic Acids Res. 35, W71eW74. Valente, L.M., Reeves, G., Schnitzler, J., Mason, I.P., Fay, M.F., Rebelo, T.G., Chase, M.W., Barraclough, T.G., 2010. Diversification of the African genus Protea (Proteaceae) in the Cape biodiversity hotspot and beyond: equal rates in different biomes. Evolution 64, 745e760. Wardlaw, A.M., Berkers, T.E., Man, K.C., Lachance, M.-A., 2009. Population structure of two beetle-associated yeasts: comparison of a New World asexual and an endemic Nearctic sexual species in the Metschnikowia clade. Antonie Leeuwenhoek 96, 1e15. Weir, B.S., Cockerham, C.C., 1984. Estimating F-Statistics for the analysis of population structure. Evolution 38, 1358e1370. White, T.J., Bruns, T., Lee, S., Taylor, J.W., 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J. (Eds.), PCR Protocols: a Guide to Methods and Applications. Academic Press, San Diego, California, pp. 315e322. Wingfield, M.J., Seifert, K.A., Webber, J.F., 1993. In: Ceratocystis and Ophiostoma: Taxonomy, Ecology and Pathogenicity. APS Press, St Paul, MN. Wingfield, M.J., Van Wyk, P.S., 1993. A new species of Ophiostoma from Protea infructescences in South Africa. Mycol. Res. 97, 709e716. Wingfield, M.J., Wyk, P.S.V., Marasas, W.F.O., 1988. Ceratocystiopsis proteae sp. nov., with a new anamorph genus. Mycologia 80, 23e30. Wright, M., 1994. Unpredictable seed-set: a defence mechanism against seed-eating insects in Protea species (Proteaceae). Oecologia 99, 397e400. Zhou, X., Burgess, T.I., De Beer, Z.W., Lieutier, F., Yart, A., Klepzig, K., Carnegie, A., Portales, J.M., Wingfield, B.D., Wingfield, M.J., 2007. High intercontinental migration rates and population admixture in the sapstain fungus Ophiostoma ips. Mol. Ecol. 16, 89e99.