A multigene molecular phylogenetic assessment of true morels (Morchella) in Turkey

Fungal Genetics and Biology 47 (2010) 672–682

Contents lists available at ScienceDirect

Fungal Genetics and Biology

journal homepage: www.elsevier .com/ locate/yfgbi

A multigene molecular phylogenetic assessment of true morels (Morchella)in Turkey

Hatıra Tas�kın a, Saadet Büyükalaca a, Hasan Hüseyin Dogan b, Stephen A. Rehner c, Kerry O’Donnell d,*

a Faculty of Agriculture, Department of Horticulture, University of Çukurova, 01330 Adana, Turkeyb Faculty of Science, Department of Biology, University of Selçuk, 42079 Konya, Turkeyc Systematic Mycology and Microbiology Laboratory, Agricultural Research Service, United States Department of Agriculture, Beltsville, MD 20705, USAd Bacterial Foodborne Pathogens and Mycology Research Unit, National Center for Agricultural Utilization Research, Agricultural Research Service,United States Department of Agriculture, Peoria, IL 61604, USA

a r t i c l e i n f o a b s t r a c t

Article history:Received 30 September 2009Accepted 9 May 2010Available online 24 May 2010

Keywords:ConservationEF-1aGCPSRITS rDNALSU rDNARPB1RPB2Species limits

1087-1845/$ - see front matter Published by Elsevierdoi:10.1016/j.fgb.2010.05.004

* Corresponding author. Address: Bacterial FoodboResearch Unit, National Center for Agricultural UtiliResearch Service, United States Department of AgricuStreet, Peoria, IL 61604-3999, USA. Fax: +1 (309) 681

E-mail address: [email protected] (K. O

A collection of 247 true morels (Morchella spp.) primarily from the Mediterranean and Aegean Regions ofSouthern Turkey, were analyzed for species diversity using partial RNA polymerase I (RPB1) and nuclearribosomal large subunit (LSU) rDNA gene sequences. Based on the result of this initial screen, 62 collec-tions representing the full range of genetic diversity sampled were subjected to multigene phylogeneticspecies recognition based on genealogical concordance (GCPSR). The 62-taxon dataset consisted of partialsequences from three nuclear protein-coding genes, RNA polymerase I (RPB1), RNA polymerase II (RPB2),translation elongation factor (EF1-a), and partial LSU rDNA gene sequences. Phylogenetic analyses of theindividual and combined datasets, using maximum parsimony (MP) and maximum likelihood (ML),yielded nearly fully resolved phylogenies that were highly concordant topologically. GCPSR analysis ofthe 62-taxon dataset resolved 15 putative phylogenetically distinct species. The early diverging Elata(black morels) and Esculenta Clades (yellow morels) were represented, respectively, by 13 and two spe-cies. Because a Latin binomial can be applied with confidence to only one of the 15 species (Morchellasemilibera), species were identified by clade (Mel for Elata and Mes for Esculenta) followed by a uniqueArabic number for each species within these two clades. Eight of the species within the Elata Clade appearto be novel, including all seven species within the Mel-20-to-31 subclade and its sister designated Mel-25.Results of the present study provide essential data for ensuring the sustainability of morel harveststhrough the formulation of sound conservation policies.

Published by Elsevier Inc.

1. Introduction

Species of true morels (Morchella spp.) are one of the mosthighly prized and easily identified epigeous macrofungi collectedby mycophiles during the Spring in temperate regions of theNorthern hemisphere (Weber, 1995; Kuo, 2005). Factors that con-tribute to their economic importance include their extraordinarydemand among gourmets and foodies alike, coupled with theirscarcity due to a sporadic and short fruiting season restricted toa few weeks each spring. With their ever-increasing popularity,harvest of wild morels has become a commercially successful cot-tage industry in morel-rich regions of countries such as China, In-dia, Mexico, Turkey and the United States in the Northern

Inc.

rne Pathogens and Mycologyzation Research, Agriculturallture, 1815 North University6672.

’Donnell).

Hemisphere (Pilz et al., 2007). As a result, it is possible to purchasedried morels in local supermarkets in numerous countriesthroughout the year. In addition, commercial cultivation of morelshas made it possible to enjoy fresh morels year round as well(Ower et al., 1986).

Synapomorphies that united Morchella, together with two otherepigeous genera, Verpa and Discioitis, within the Morchellaceae in-cluded multinucleate, eguttulate ascospores with a crown of epi-plasmic polar granules (Berthet, 1964). Molecular phylogeneticanalyses using nearly complete SSU rDNA and partial LSU rDNA se-quences confirmed the monophyly of this family (O’Donnell et al.,1997) and expanded its circumscription to include two hypogeousgenera, Fischerula and Leucangium (Hansen and Pfister, 2006).Although these studies provided a robust hypothesis of evolution-ary relationships within the Morchellaceae, and identified the Dis-cinaceae as its sister, surprisingly few studies have focused onelucidating species limits within Morchella, and none has usedmultilocus phylogenetic species recognition based on genealogicalconcordance (GCPSR; Taylor et al., 2000). To date, species-level

H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682 673

molecular systematic markers employed within this genus havebeen limited to isozymes (Gessner et al., 1987; Royse and May,1990; Wipf et al., 1996), restriction fragment polymorphisms ofthe LSU rDNA (Bunyard et al., 1994, 1995) and internal transcribedspacer (ITS) region of the nuclear rDNA (Buscot et al., 1996), andphylogenetic analyses of the ITS rDNA sequence data (Wipf et al.,1999; Kellner et al., 2005). However, due to its length variability,ITS rDNA sequences cannot be aligned among members of theblack (Elata Clade) or yellow (Esculenta Clade) morels unlessextensive regions are excluded from the analyses (Kellner et al.,2005).

Although export of fresh morels from Turkey quadrupled overthe past 5 years where in 2008, 238,106 kg averaging $18 US dol-lars per kilogram were exported primarily to France and Germany,knowledge of Turkish morels is limited to a single morphologicalsurvey of macrofungi (Solak et al., 2007). Understanding their spe-cies diversity, however, is critical to elucidating their ecology andsystematics, and for formulating sound conservation policies to en-sure sustainability of morel harvests (Pilz et al., 2007). Towardsthis end, 247 specimens of Morchella, collected in the spring of2007 and 2008 primarily in Mediterranean and Aegean provi-dences of Southern Turkey, were analyzed phylogenetically usingmultilocus DNA sequence data. The primary objectives of the pres-ent study were threefold: (1) compare the utility of partial LSUrDNA, RPB1, RPB2 and EF-1a gene sequences for resolving specieslimits of Turkish morels using GCPSR (Taylor et al., 2000); (2) as-sess the utility of ITS rDNA sequences for species-level phylogenet-ics within the Elata Mel-20-to-31 subclade, and (3) use thephylogenetic data to better understand the geographic distributionof Morchella spp. within Turkey.

2. Methods

2.1. Material studied

The 247 Morchella specimens included in the present studywere collected in the spring of 2007 and 2008 in 10 Turkish

Fig. 1. The 10 provinces and five regions (in p

provinces (Fig. 1; Supplemental Table 1) and then air-dried for sub-sequent analysis. After the specimens were typed molecularlyusing partial RPB1 and LSU rDNA gene sequence data, a subset of62 collections (Table 1), chosen to represent the full range of phy-logenetic diversity sampled, were analyzed phylogenetically usinga four-locus dataset comprising portions of the RPB1, RPB2, EF-1agenes and domains D1 and D2 of the nuclear large subunit (LSU)rDNA. In addition, sequence of the nuclear ribosomal ITS rDNA re-gion was obtained for members of the Mel-20-to-31 subclade. Phy-logenetic species identity, geographic origin including GPScoordinates and dominant vegetation type were recorded for eachcollection (Table 1; Supplemental Table 1). Twenty-seven culturesrepresenting 13 of the 15 species sampled in this study were madeto preserve this novel fungal genetic diversity to insure that itwould be available to the scientific and biotechnological communi-ties (see Supplemental Table 1; Hawksworth, 2004). Pure culturesobtained from germinated ascospores were typed using a partialRPB2 gene sequence to insure their authenticity, prior to beingstored in the ARS Culture Collection (NRRL) at �175 �C in a cryogenconsisting of 10% skim milk and 1% DMSO.

To assess whether ascospore and ascus dimensions could beused for morphological species recognition, detailed measure-ments were made of ascospore length and width (N = 50) and ascuslength and width (N = 25) from two independent collection of11/15 Morchella species detected in Turkey; measurements of4/15 species (Mel-9, Mel-26, Mel-30 and Mes-8) were based onsingle collections (Supplemental Table S2).

2.2. Molecular biology

Total genomic DNA was extracted following a CTAB (hexadecyl-trimethyl-ammonium bromide (Sigma Chemical Co., St. Louis, MO)protocol (O’Donnell et al., 1997) from approximately 50–100 mg ofeach dried specimen, which had been pulverized with a pipette tipin a 1.5 ml eppendorf tube. Once the mycelium was resuspended in700 ll of the CTAB extraction buffer (100 mM Tris–Cl pH 8.4, 1.4 MNaCl, 25 mM EDTA pH 8.0, 2% CTAB), it was incubated at 60 �C for

arentheses) where morels were collected.

Table 1Species subjected to multilocus phylogenetic analyses.

Speciesa HT no.b Provincec Region GPS coordinated Dominant vegetatione Herbarium numberf

Mel-2 176 Adana (F) Mediterranean 37�5205100N–035�4801800E-1373 m Pine (Pb, Pn), Cedar, Fir ANK Taskin 39Mel-2 179 Adana (F) Mediterranean 37�5205100N–035�4801800E-1373 m Pine (Pb, Pn), Cedar, Fir ANK Taskin 40Mel-2 130 Kahramanmaras� (G) Mediterranean Unknown Pine (Pb, Pn), Cedar, Fir ANK Taskin 49Mel-2 144 Kahramanmaras� (G) Mediterranean Unknown Pine (Pb, Pn), Cedar, Fir ANK Taskin 50Mel-2 147 Kahramanmaras� (G) Mediterranean Unknown Pine (Pb, Pn), Cedar, Fir ANK Taskin 51Mel-2 197 Mugla (B) Agean 36�5003800N–029�0701400E-1048 m Pine (Pb), Oak ANK Taskin 28Mel-2 199 Mugla (B) Agean 36�5105400N–029�1002800E-1413 m Pine (Pb), Oak ANK Taskin 29Mel-2 200 Mugla (B) Agean 36�5105400N–029�1002800E-1413 m Pine (Pb), Oak ANK Taskin 30Mel-3 213 Antalya (D) Mediterranean 37�0504600N–031�3601400E-999 m Chestnut ANK Taskin 10Mel-3 215 Antalya (D) Mediterranean 37�0504600N–031�3601400E-999 m Chestnut ANK Taskin 11Mel-3 216 Antalya (D) Mediterranean 37�0504600N–031�3601400E-999 m Chestnut ANK Taskin 12Mel-7 235 Antalya (D) Mediterranean 36�1902100N–029�4502100E-180 m Pine (Pb), Oak (Burned area) ANK Taskin 16Mel-7 239 Antalya (D) Mediterranean 36�1902100N–029�4502100E-180 m Pine (Pb), Oak (Burned area) ANK Taskin 22Mel-7 240 Antalya (D) Mediterranean 36�1902100N–029�4502100E-180 m Pine (Pb), Oak (Burned area) ANK Taskin 17Mel-7 242 Antalya (D) Mediterranean 36�1902100N–029�4502100E-180 m Pine (Pb), Oak (Burned area) ANK Taskin 19Mel-7 244 Antalya (D) Mediterranean 36�1902100N–029�4502100E-180 m Pine (Pb), Oak (Burned area) ANK Taskin 20Mel-7 245 Antalya (D) Mediterranean 36�1902100N–029�4502100E-180 m Pine (Pb), Oak (Burned area) ANK Taskin 21Mel-7 93 Unknown Unknown Unknown ANK Taskin 08Mel-9 166 Denizli (C) Agean Unknown Pine (Pb, Pn) ANK Taskin 33Mel-10 114 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 54Mel-10 85 Mersin (E) Mediterranean 36�1802000N–033�2505700E-780 m Pine (Pb) ANK Taskin 15Mel-10 238 Antalya (D) Mediterranean 36�1902100N–029�4502100E-180 m Pine (Pb), Oak (Burned area) ANK Taskin 18Mel-20 123 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 58Mel-20 260 Sivas (I) Central Anatolian 40�2100500N–037�5405600E-1963 m Pine (Pb, Pn), Fir ANK Taskin 63Mel-20 42 Adana (F) Mediterranean 37�5105600N–035�4800500E-1325 m Pine (Pb, Pn), Fir ANK Taskin 02Mel-20 46 Adana (F) Mediterranean 37�5105600N–035�4800500E-1325 m Pine (Pb, Pn), Fir ANK Taskin 60Mel-20 165 Denizli (C) Agean Unknown Pine (Pb, Pn) ANK Taskin 32Mel-20 168 Denizli (C) Agean Unknown Pine (Pb, Pn) ANK Taskin 34Mel-20 181 Adana (F) Mediterranean 37�5205100N–035�4801800E-1373 m Pine (Pb, Pn), Cedar, Fir ANK Taskin 41Mel-20 182 Adana (F) Mediterranean 37�5205100N–035�4801800E-1373 m Pine (Pb, Pn), Cedar, Fir ANK Taskin 42Mel-20 183 Adana (F) Mediterranean 37�5205100N–035�4801800E-1373 m Pine (Pb, Pn), Cedar, Fir ANK Taskin 43Mel-20 184 Adana (F) Mediterranean 37�5205100N–035�4801800E-1373 m Pine (Pb, Pn), Cedar, Fir ANK Taskin 44Mel-20 187 Mugla (B) Agean 36�5003800N–029�0701400E-1048 m Pine (Pb), Oak ANK Taskin 01Mel-20 191 Mugla (B) Agean 36�5003800N–029�0701400E-1048 m Pine (Pb), Oak ANK Taskin 24Mel-20 195 Mugla (B) Agean 36�5003800N–029�0701400E-1048 m Pine (Pb), Oak ANK Taskin 26Mel-20 92 Unknown Unknown Unknown ANK Taskin 07Mel-25 157 Aydın (A) Agean 37�3701300N–028�1901500E-819 m Pine (Pb, Pn) ANK Taskin 36Mel-25 65 Mugla (B) Agean 36�5003800N–029�0701400E-1048 m Pine (Pb), Oak ANK Taskin 31Mel-25 188 Mugla (B) Agean 36�5003800N–029�0701400E-1048 m Pine (Pb), Oak ANK Taskin 27Mel-26 156 Aydın (A) Agean 37�3701300N–028�1901500E-819 m Pine (Pb, Pn) ANK Taskin 35Mel-26 120 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 56Mel-26 122 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 57Mel-26 124 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 59Mel-27 107 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 53Mel-27 118 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 55Mel-27 25 Mersin (E) Mediterranean 36�4200700N–033�5905600E-1335 m Pine (Pb), Juniper, Spruce ANK Taskin 47Mel-27 207 Kahramanmaras� (G) Mediterranean 38�0800000N–036�3500400E-1685 m Cedar, Fir, Juniper ANK Taskin 37, ANK Taskin 38Mel-28 251 Sivas (I) Central Anatolian 40�2100500N–037�5405600E-1963 m Pine (Pb, Pn), Fir ANK Taskin 62Mel-28 50 Adana (F) Mediterranean 37�5105600N–035�4800500E-1325 m Pine (Pb, Pn), Fir ANK Taskin 03Mel-28 201 Kahramanmaras� (G) Mediterranean 38�0801500N–036�3402000E-1630 m Cedar, Fir, Juniper ANK Taskin 45Mel-29 155 Aydın (A) Agean 37�3701300N–028�1901500E-819 m Pine (Pb, Pn) ANK Taskin 04Mel-29 45 Adana (F) Mediterranean 37�5105600N–035�4800500E-1325 m Pine (Pb), Cedar, Fir ANK Taskin 23Mel-29 82 Mersin (E) Mediterranean 36�1802000N–033�2505700E-780 m Pine (Pb) ANK Taskin 05Mel-30 193 Mugla (B) Agean 36�5003800N–029�0701400E-1048 m Pine (Pb), Oak ANK Taskin 25Mel-31 106 Kastamonu (J) Black Sea Unknown Unknown ANK Taskin 52Mel-31 202 Kahramanmaras� (G) Mediterranean 38�0801500N–036�3402000E-1630 m Cedar, Fir, Juniper ANK Taskin 46Mel-31 203 Kahramanmaras� (G) Mediterranean 38�0801500N–036�3402000E-1630 m Cedar, Fir, Juniper ANK Taskin 61Mel-31 212 Kahramanmaras� (G) Mediterranean 38�0705800N–036�3500500E-1706 m Cedar, Fir, Juniper ANK Taskin 48Mes-8 218 Antalya (D) Mediterranean 37�0402600N–031�3804800E-730 m Pine (Pb) ANK Taskin 13Mes-17 225 Antalya (D) Mediterranean 37�050380 0N–031�400230 0E-470 m Pine (Pb) ANK Taskin 14Mes-17 98 Diyarbakır (H) Southeastern Anatolian Unknown Unknown ANK Taskin 06Mes-17 95 Unknown Unknown Unknown ANK Taskin 09

a Fifteen putatively phylogenetically distinct species were resolved employing GCPSR (Fig. 3; Taylor et al., 2000).b Ht no., Hatira Tas�kin’s collection number.c The letter in parentheses identifies each of the 10 provinces where morels were collected in Fig. 1.d GPC coordinates for 18 of the collections are unknown because they were purchased from collectors.e Pine, Pb = Pinus brutia Tenore, Pn = Pinus nigra J.F. Arnold; Cedar, Cedrus libani A. Rich.; Fir, Abies cilicica (Ant. & Kotschy)Carr.; Oak, Quercus coccifera L.; Chestnut, Castanea

sativa Mill.; Juniper, Juniperus sp.; Spruce, Picea orientalis (L.)Link.f ANK, Ankara University Herbarium, Ankara, Turkey.

674 H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682

1–24 h. Afterwards 700 ll of chloroform was added to each tube,samples were vortexed briefly and then spun for 10 min at12,300g in a microcentrifuge (Savant, Holbrook, NY). Next 350 ll

of the upper phase was carefully removed to a new 1.5 ml eppen-dorf tube to which an equal volume of �20 �C isopropanol wasadded to precipitate the DNA. After the DNA was pelleted at

H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682 675

12,300g for 5 min, the supernatant was discarded and the DNA pel-let was gently washed with 70% ethanol and resuspended in 200 llddH2O by heating at 65 �C for 1–2 h. Once the DNA pellet was com-pletely dissolved, 8 ll of the genomic DNA prep was added to192 ll of sterile double-distilled H2O in a 96-well plate and usedas a template for amplification by the polymerase chain reaction(PCR) (O’Donnell et al., 1998). DNA extraction from pure culturesgrown in yeast–malt broth (O’Donnell et al., 1997) followed theprotocol outlined above, with the exception that mycelium wasfreeze-dried before the CTAB extraction step.

All PCR and sequencing primers are listed in Table 2. PlatinumTaq DNA polymerase (Invitrogen Life Technologies, Carlsbad, CA)was employed in all PCR amplifications. Amplifications were con-ducted using an Applied Biosystems (ABI) 9700 thermocycler(Emeryville, CA) using the following program: 1 cycle of 90 s at94 �C; 40 cycles of 30 s at 94 �C, 90 s at 55 �C, and 3 min at 68 �C;followed by 1 cycle of 5 min at 68 �C and a 4 �C soak. Gel electro-phoresis in 1.5% agarose gels (Invitrogen) run in TAE buffer (Sam-brook et al., 1989) was used to size-fractionate amplicons.Subsequently gels were stained with ethidium bromide and visual-ized over a UV trans-illuminator. Montage96 filter plates (MilliporeCorp., Billerica, MA) were used to purify amplicons prior to cyclesequencing. All sequencing reactions were conducted in a 10-llvolume containing 2–4 pmol of a sequencing primer, 2 ll of ABIBigDye version 3.1-terminator reaction mix, and approximately50 ng of amplicon as described previously (O’Donnell et al.,1998). DNA sequencing reactions were purified with ABI Big-DyeXTerminator and then run on an ABI 3730 48 capillary automatedsequencer.

2.3. Molecular phylogenetics

Raw ABI sequence chromatograms were edited and alignedwith Sequencher ver. 4.1.2 (Gene Codes, Ann Arbor, MI, USA), priorto manual improvement of the alignments using TextPad ver. 5.1.0for windows (http://www.textpad.com/). Sequences from the indi-vidual partitions were concatenated in a single nexus file and thenassessed for combinability using PAUP* 4.0b10 (Swofford, 2002).

Table 2PCR and sequencing primers.

Locus Primer References

RPB1RPB-1A Matheny et al. (2002)RPB-1C Matheny et al. (2002)RPB-C2 This studyRPB-A2 This study

RPB2RPB2-7cf Liu et al. (1999)RPB2-3053r Reeb et al. (2004)RPB2-9f This studyRPB2-3r This study

EF-1a (50 fragment)EF-526F Rehner and Buckley (2005)EF-1567R Rehner and Buckley (2005)EF-3AR This study

EF-1a (30 fragment)EF-2F Rehner and Buckley (2005)EF-2218R Rehner and Buckley (2005)EF-4AF This studyEF-5AR This study

rDNAITS4 White et al. (1990)ITS5 White et al. (1990)NL1 O’Donnell et al. (1997)NL4 O’Donnell et al. (1997)

a IUPCA degenerate nucleotides: D, AGT; H, ACT; M, AC; N, ACGT; R, AG; S, CG; Y, CT.

Comparisons based on maximum parsimony bootstrap valuesP70% indicated that they were topologically congruent. Thereforethe individual partitions were analyzed phylogenetically as a com-bined dataset, using MP in PAUP* (Swofford, 2002) and by maxi-mum likelihood (ML) in GARLI ver. 0.951 (Zwickl, 2006). Thehierarchical likelihood ratio tests in MrModeltest ver. 2.2 (Posadaand Crandall, 1998), using PAUP*, identified the general-time-reversible model with a proportion of invariant sites and gammadistributed rate heterogeneity (GTR + I + C) as the best-fit modelof nucleotide substitution for the combined dataset. MP searchesfor the shortest trees employed tree-bisection and reconnectionbranch swapping (TBR), and 1000 random sequence addition rep-licates. Nonparametric bootstrapping, employing 1000 pseudo-replicates of the data, 10 random addition sequences per replicate,and TBR branch swapping was used to assessed MP clade support.Nonparametric bootstrap analyses of the 62- and 36-taxon data-sets were conducted in GARLI (Zwickl, 2006) using the 5000 gener-ations without improving the topology parameter and 500 MLpseudo-replicates of the data. DNA sequences generated in thisstudy have been deposited in GenBank as accessions HM056248-HM056530.

3. Results

Although the 247 morels studied were obtained from 10 prov-inces in five regions (Fig. 1) during the spring of 2007 and 2008,81% of the specimens were collected in three Aegean (N = 48)and four Mediterranean (N = 152) provinces (Supplemental Table1). The remaining collections were made in the provinces of Kasta-monu (N = 22; Black Sea region), Sivas (N = 18; Central Anatolia re-gion), Diyarbakır (N = 1; Southeastern Anatolia region) or theprovince and region were not documented because the collectionswere purchased from exporters (N = 6). GPS coordinates were re-corded for all collections except for 57 purchased from regionalor local collectors. In addition, the dominant vegetation type wasrecorded, except for 30 collections purchased from exporters orcollectors (Supplemental Table 1). An initial assessment of Morch-

Sequence (50–30)a PCR Sequencing

GARTGYCCDGGDCAYTTYGG XCCNGCDATNTCRTTRTCCATRTA XGMAGAACMGTAATCACCATCC XGTTAGATGAAGTGAGACACAC X

ATGGGYAARCAAGCYATGGG XTGRATYTTRTCRTCSACCATRTG XCAAATGGGCRATTGTCATACG XGCATYGGTATGCAGGTTGTGG X

GTCGTYGTYATYGGHCAYGT XACHGTRCCRATACCACCRATCTT XGAAACGRTCCTCRGACCAC X

AACATGATSACTGGTACYTCC X XATGACACCRACRGCRACRGTYTG XTCAAGTCCGTCGARATGCACC XCCAGCAACRTTACCACGACG X

TCCTCCGCTTATTGATATGC XGGAAGTAAAAGTCGTAACAAGG X XGCATATCAATAAGCGGAGG XGGTCCGTGTTTCAAGACGG X X

676 H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682

ella genetic diversity in Turkey was obtained from maximum par-simony (MP) phylogenetic analyses of aligned partial RPB1(852 bp) and LSU rDNA (535 bp) gene sequences obtained fromthe 247 collections (Supplemental Table 1; Fig. 2). Based on thispreliminary genetic screen, 62 collections were selected for furtherstudy to represent the full range of phylogenetic diversity withinthe 247-taxon collection (Table 1). For this 62-taxon matrix, por-tions of two additional nuclear genes were sequenced, and they in-cluded RPB2 (956 bp) and EF1-a (1299 bp). The 956 bp region ofRPB2 sequenced does not contain any introns. The A–C region ofRPB1 sequenced (852 bp alignment) was interrupted by two smallintrons representing 133 bp of aligned sequence in the 62-taxondataset. Collectively, the RPB1 introns contributed only 44 parsi-mony-informative characters (PIC) compared with 134 from theexons. By contrast, the EF-1a alignment (1299 bp) for the 62-taxondataset contained four introns whose aligned length totaled274 bp. These contributed 110 parsimony-informative characters(PIC), the same number as the four exons whose length totaled1025 bp.

MP bootstrap (BS) analyses of the four individual partitions re-vealed strong gene-gene concordance. Topological discord involv-ing nodes receiving P70% bootstrap support was not detected.Relative to the LSU rDNA, which provided little phylogenetic signal(5.98% PIC) and nodal support (Tables 3 and 4), the RPB1, RPB2 andEF1-a protein-coding genes possessed between 16.9% and 21.8%parsimony-informative characters and these three partitions indi-vidually provided bootstrap support for the monophyly of 9-to-10of the 12 species represented by two or more collections (Table 4).Strong support for the monophyly of all 12 species (i.e., P92% BS),however, was only obtained by analyses of the combined dataset.Based on the results of a conditional combination approachemploying P70% bootstrap values as the threshold for topologicalconcordance, sequence data from the four individual loci werecombined and analyzed using MP in PAUP* (Swofford, 2002) andML in GARLI (Zwickl, 2006). The 62 sequences in the combineddataset totaled 3642 bp of aligned sequence of which 17.4% wereparsimony-informative (Table 3). Unweighted MP analysis of thecombined dataset identified 384 most-parsimonious trees (MPTs)1029 steps in length with a CI of 0.751 and RI of 0.971. ML boot-strap analysis using the GTR + I + G model of DNA substitution inGARLI recovered clade support nearly identical to MP bootstrap-ping with the exception that one weakly supported node in theMP analysis (71% BS) received less than 50% BS in the ML analysis(Fig. 3). Sequences of the two Esculenta Clade (yellow morels) spe-cies, Mes-17 and Mes-8, were used to root the trees based on theresults of more inclusive analyses. Of the 15 phylogenetically dis-tinct species resolved by MP and ML analyses of the combineddataset (Figs. 2 and 3), a Latin binomial can be confidently appliedonly to Mel-3, Morchella semilibera. Twelve of the species in our col-lection from Turkey, based on a more inclusive sampling world-wide (O’Donnell et al., in review), do not appear to have beendescribed formally. The remaining two species (Mel-20 and Mes-8) very likely have names, because they are common in Europe,but we do not know what names to apply because there are manyvalidly published names for black and yellow morels in Europe.Even when type species exist, most are too old to yield usefulmolecular systematic data. In the absence of Latin binomials for14 of the 15 terminals, species were identified by clade (Mes forEsculenta and Mel for Elata) followed by an Arabic number to dis-tinguish species.

Except for Mel-9 and Mel-10, which formed an unresolved poly-tomy in the bootstrap analyses, evolutionary relationships amongthe 13 other putatively phylogenetically distinct species werenearly fully resolved. Four of the five basal-most species withinthe Elata Clade (i.e., Mel-2, Mel-7, Mel-9 and Mel-10), representing35.6% of the collection (88/247), appear to have been introduced to

Turkey given that they phylogenetically match species that appearto be endemic to the Pacific Northwest of North America (O’Don-nell et al., in review). Of the four introduced species, Mel-2 wasthe most common (N = 59) and widespread with collections fromthree regions and seven provinces (Table 5), followed by Mel-10represented by 17 collections from two regions and three prov-inces. Phylogenetic species Mel-7 and Mel-9 represented by 11and 1 collection were restricted, respectively, to the Mediterraneanprovince of Antalya and the Aegean province of Denizli (Supple-mental Table 1). The dominant vegetation for Aegean collectionsof Mel-2 included Pinus brutia (Turkish pine), Pinus nigra (Austrianpine) and Quercus coccifera (Kermes oak), whereas it consisted ofP. brutia, P. nigra together with Abies cilicica (Cicilian fir) and Cedruslibani (Lebanon cedar) from specimens collected from the Mediter-ranean region. By way of contrast, Mel-7 and Mel-10 were collectedon burn sites dominated by P. brutia and Q. coccifera primarily inthe Mediterranean province of Antalya. With the exception ofP. nigra, which was introduced to and is widely grown within theUnited States, the dominant tree species at the sites where theintroduced morel species were collected are not planted widelywithin the US.

In addition to Mes-17 from Anatalya and Diyarbakır, seven ofthe eight terminal-most Elata Clade species resolved as phyloge-netically distinct in the present study have only been collected inTurkey (Fig. 3; Mel-20 is found in Europe), based on more inclusiveanalyses (O’Donnell et al., in review). Six of these species (i.e., Mel-26 through Mel-31) are members of the Mel-20-to-31 subclade, aninformally named lineage representing a relatively recent speciesradiation within the Elata Clade. The sister species of the Mel-20-to-31 subclade, Mel-25, was collected widely in the Aegean andMediterranean regions at sites dominated by P. brutia and Q. coccif-era. Mel-25 was highly divergent from the Elata Clade Mel-20-to-31subclade based on analyses of the four individual and combineddataset. Because it was resolved as a sister to this subclade in allof the analyses, we felt justified in treating it as an outgroup.Mel-27 (N = 44) and Mel-20 (N = 55), next to Mel-2, comprisedthe second and third most widespread species in the present study,being sampled respectively from three regions and five provincesand four regions and seven provinces from the Black to the Medi-terranean Sea (Table 5; Supplemental Table 1). Excluding the threespecies represented by a single collection (Mel-9, Mel-30 and Mes-8), 10 of the 12 species were collected in two or more regions(Table 5). Only Mel-3 (N = 5) from sites dominated primarily byCastanea sativa (Chestnut) and Mel-7 (N = 11) from P. brutia andQ. coccifera post-fire sites were restricted to a single region, theMediterranean (Table 5).

To improve phylogenetic resolution within the Mel-20-to-31subclade, 542 bp of aligned sequence from the ITS rDNA regionwas added to the four-locus dataset. This 36-taxon dataset com-prised 4163 bp of aligned DNA sequence of which 5.9% was parsi-mony-informative (Table 6). Alignment of the ITS rDNA partitionrequired the insertion of one 1-bp indel within the ITS1 and three1-to-2-bp indels within the ITS2. Analyses of the 36-taxon datasetrevealed that the ITS and LSU rDNA partitions possessed the mostand least phylogenetic signal, respectively, with 7.56% and 0.19%parsimony-informative characters. Addition of the ITS rDNA parti-tion to the four-locus dataset, however, did not improve phyloge-netic resolution within this subclade (Fig. 4). Consistent with theresults obtained with the 62-taxon dataset, support for the mono-phyly of the most species was obtained using partial sequences fromthe three protein-coding genes (Table 7). Of these, RPB2 performedthe best by providing support for six of the seven species; however,strong support (i.e., P97% BS) for the monophyly of all seven specieswas only obtained through analyses of the combined dataset.

We conducted an a posteriori morphological analysis to assesswhether ascospore and ascus dimensions could be used for mor-

Fig. 2. Morphological diversity of Morchella Elata Clade ascocarps: (A) Mel-2, (B) Mel-7, (C) Mel-9, (D) Mel-10, (E) Mel-20, (F) Mel-25, (G) Mel-26, (H) Mel-27, (I) Mel-28, (J) Mel-29, (K) Mel-30, and (L) Mel-31. Note that Mel-26 through Mel-31 are members of the Mel-20-to-31 subclade.

H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682 677

Table 362-Taxon tree statistics and summary sequence (see Fig. 3).

Locus # Characters # MPTsa MPT length CIb RIc PIC/bpd (%)

LSU rDNA 535 894 44 0.818 0.962 5.98RPB1 852 108 263 0.806 0.967 20.77RPB2 956 8 329 0.745 0.953 21.77EF-1a 1299 511 376 0.745 0.954 16.94Combined 3642 384 1029 0.751 0.971 17.43

a MPTs, most-parsimonious trees.b CI, consistency index.c RI, retention index.d PIC/bp, parsimony-informative characters/base pair.

Table 4Bootstrap support for individual and combined 62-taxon dataset.

Speciesa LSU rDNA RPB1 RPB2 EF-1a Combined

Mel-2 83 100 100 100 100Mel-3 83 100 100 100 100Mel-8 61 97 100 100 100Mel-9 NA NA NA NA NAMel-10 93 100 100 100 100Mel-20 – – 73 90 98Mel-25 86 100 100 100 100Mel-26 – 61 96 77 100Mel-27 – 100 96 97 100Mel-28 – 88 92 57 100Mel-29 – 83 – 64 94Mel-30 NA NA NA NA NAMel-31 – – 62 96 92Mes-8 NA NA NA NA NAMes-17 96 98 100 100 100

a Of the 13 species within the Elata Clade (Mel) and two species within EsculentaClade (Mes), Latin binomial can be applied with confidence to only Mel-3 (Morchellasemilibera). As indicated by NA, bootstrap support could not be obtained for Mel-9,Mel-30 and Mes-8 because they were each represented by a single collection.

678 H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682

phological species recognition (Supplemental Table S2). Becausevariability of ranges was large, ascospore and ascus length andwidth cannot be used for species differentiation.

4. Discussion

The primary objective of the present study was to investigatespecies limits and evolutionary relationships among 247 true mor-els (Morchella) collected in 10 different provinces from five regionswithin Turkey during 2007 and 2008. Towards this end, partialRPB1 and LSU rDNA sequence data were collected and analyzedphylogenetically to obtain an initial estimate of Morchella speciesdiversity in Turkey. All of the sequence data was obtained using to-tal genomic DNA extracted directly from dried specimens as a tem-plate for PCR amplifications. The RPB1 primers designed byMatheny et al. (2002) for Inocybe (Agaricales) were used to amplifythe 0.9 kb A–C region from all 247 collections. We attribute thesuccessful amplification of this single-copy gene in part becausethe specimens were only three to 15 months old at the time DNAwas extracted and because they had been dried as soon as possibleafter they had been collected. By designing Morchella-specificinternal sequencing primers, we were able to increase the qualityof RPB1 sequence data substantially compared to that using thePCR primers.

To further elucidate species boundaries using GCPSR, portionsof two additional protein-coding genes, RPB2 and EF1-a, were cho-sen to increase phylogenetic resolution based on several recent re-ports documenting their utility at low taxonomic levels in diversefungal groups such as the Ascomycota (Hansen et al., 2005; Hof-stetter et al., 2007), Basidiomycota (Frøslev et al., 2005; Matheny,

2005), and Mucorales (O’Donnell et al., 2001). A forward primer de-signed by Liu et al. (1999) and a reverse primer from Reeb et al.(2004) successfully amplified an approximately 1 kb fragmentspanning the 7–11 region of RPB2 from all 247 collections (Table3). For the partial EF1-a gene sequence, two pairs of PCR primersdesigned by Rehner and Buckley (2005) for Beauveria were usedto amplify this region as two overlapping fragments. Using thesame strategy for sequencing the RPB1 amplicon, we designedinternal primers for sequencing the RPB2 and EF1-a amplicons,thereby greatly increasing the quality of sequence data from thesetwo genes. EF1-a PCR primer EF-2F worked well as an internalsequencing primer as did EF-4AF and EF-5AR. These three primersgenerated approximately 1.3 kb of contiguous sequence for the 62-taxon dataset. In contrast to the three single-copy protein-codinggenes, the rDNA PCR primers generally worked well for DNAsequencing (White et al., 1990; O’Donnell, 1996). Consistent withthe results of several multilocus phylogenetic analyses of diversefungal groups (O’Donnell et al., 1998; Frøslev et al., 2005; Hofstet-ter et al., 2007; Matheny, 2005), partial sequence data from the LSUrDNA contributed relatively little phylogenetic signal and nodalsupport relative to the RPB1, RPB2 and EF1-a protein-coding genessampled (Tables 3 and 6). Given the small number of parsimony-informative characters within the partial LSU rDNA sequence, itis not surprising that phylogenetic inferences within the Morchell-aceae based on RFLP analyses of this region failed to resolve themonophyly of Morchella (Bunyard et al., 1994, 1995). In addition,these analyses were also unable to distinguish it from Gyromitrain its sister family, the Discinaceae (Hansen and Pfister, 2006;O’Donnell et al., 1997). To date, published DNA sequenced-basedphylogenetic analyses of Morchella have been limited to sequencesfrom the nuclear ITS rDNA region (Kellner et al., 2005; Wipf et al.,1999). However, because sequences from this locus cannot bealigned to include all of the taxa within the Elata or EsculentaClades, we limited its use in the current study to phylogeneticanalysis of the Mel-20-to-31 subclade, where the insertion of onlythree 1–2 bp indels were required to establish positional homologyat this low taxonomic level.

Results of the present study add to a growing number of GCPSR-based studies that have employed concordance among multigenegenealogies to investigate species limits within agriculturally(Geiser et al., 1998; O’Donnell et al., 2000; Rehner and Buckley,2005) and medically important fungi (Koufapanou et al., 1997;Pringle et al., 2005), including model system fungi (Dettmanet al., 2003a,b). Phylogenetic species were delimited herein basedon the following two criteria, after excluding the LSU rDNA parti-tion because it was nearly devoid of phylogenetic signal: (1) strongmonophyly bootstrap support from a majority of the individualand combined partitions and (2) none of the genes contradicted aspecies genealogical exclusivity (i.e., there was no conflict betweenwhat species lineages were supported by bootstrap analyses of thefour single-gene partitions; Dettman et al., 2003a,b). As almostuniversally noted in multigene phylogenetic studies, bootstrap

25steps

28S rDNA, RPB1RPB2, EF-1α3642 bp635 PIC1 of 384 MPTs1029 stepsCI = 0.751RI = 0.954

106202203

212193

155458220125150123182

424618118318426016816519592

187191

10711820725

120122124156

15718865

11423885

166235244

93239240242245

213216215

130179

144147197199

200176

2182259598

Mel 20-to-31 Subclade

ElataClade

Esculenta Clade(outgroup)

Mel-31 (4)

Mel-30 (1)

Mel-29 (5)

Mel-28 (5)

Mel-20 (55)

Mel-27 (44)

Mel-26 (5)

Mel-25 (30)

Mel-10 (17)

Mel-9 (1)

Mel-7 (11)

Mel-3 (5)

Mel-2 (59)

Mes-8 (1)

Mes-17 (4)100

100

100

100

100

100

71

82

91

88

98

94

95

<50

100

100

100

100

100

74

91

92

100

100

100

100

100

98

79

Fig. 3. One of 384 most-parsiminous trees inferred from MP analysis of the combined four-locus dataset for 62 specimens of Morchella collected primarily from the Aegeanand Mediterranean regions of Turkey. Collections are identified by Hatira Tas�kin’s collection numbers. The 15 phylogenetically distinct species are identified by three Romanletters, identifying clade affiliation (i.e., Mel for Elata and Mes for Esculenta), followed by an Arabic number. Mel-3 = M. semilibera is the only species for which a Latin binomialcan be applied confidently. Four species highlighted in gray (Mel-2, Mel-7, Mel-9 and Mel-10) appear to have been introduced into Turkey from the Pacific Northwest of NorthAmerica. Numbers above internodes indicate MP bootstrap values based on 1000 pseudo-replicates of the data. ML bootstrap values based on 500 pseudo-replicates of thedata are indicated below internodes only when they differed by P5% of the MP value.

H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682 679

Table 5Geographic distribution of 247 Morchella collections.

Speciesa # Collectionsb # Regions/# provincesc Regions (# collections)d

Mel-2 59 3/7 Agean (12), Black Sea (1), Mediterranean (43), ? (3)Mel-3 5 1/2 Mediterranean (5)Mel-7 11 1/1 Mediterranean (11)Mel-9 1 1/1 Agean (1)Mel-10 17 2/3 Black Sea (3), Mediterranean (14)Mel-20 55 4/7 Agean (7), Black Sea (8), Central Anatolian (16), Mediterranean (23), ? (1)Mel-25 30 2/5 Agean (22), Mediterranean (8)Mel-26 5 2/2 Agean (1), Black Sea (4)Mel-27 44 3/5 Agean (2), Black Sea (5), Mediterranean (37)Mel-28 5 2/3 Central Anatolian (2), Mediterranean (3)Mel-29 5 2/5 Agean (2), Mediterranean (3)Mel-30 1 1/1 Agean (1)Mel-31 4 2/2 Black Sea (1), Mediterranean (3)Mes-8 1 1/1 Mediterranean (1)Mes-17 4 2/2 Mediterranean (2), SE Anatolian (1), ? (1)

a Fifteen putatively phylogenetically distinct species resolved by GCPSR (Taylor et al., 2000).b See Supplemental Table 1 for data on all 247 collections.c Collections were obtained from five regions and 10 provinces (Fig. 1).d Species distribution by region. ? Indicates the region is unknown because it was purchased from a collector.

Table 636-taxon tree statistics and summary sequence (see Fig. 4).

Locus # Characters # MPTsa MPT length CIb RIc PIC/bpd (%)

ITS rDNA 542 16 51 0.9804 0.992 7.56LSU rDNA 534 1 9 1 1 0.19RPB1 848 61 61 0.983 0.994 6.6RPB2 956 4 62 0.984 0.994 5.96EF-1a 1283 8 103 0.884 0.965 6.55Combined 4163 88 295 0.919 0.971 5.91

a MPTs, most-parsimonious trees.b CI, consistency index.c RI, retention index.d PIC/bp, parsimony-informative characters/base pair.

680 H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682

analyses of the combined dataset provided stronger nodal supportthan the individual partitions (Hofstetter et al., 2007; Matheny,2005; O’Donnell et al., 2000; Reeb et al., 2004).

The most important finding to emerge from the present studywas the discovery of 15 putatively phylogenetically distinct butmorphologically cryptic Morchella species (Fig. 2), based on oursurvey conducted during two consecutive years within five regionsof Turkey. This finding contrasts with regional field guides that listbetween three and seven species in North America (Miller, 1972;Arora, 1979; Weber, 1995), eight species in Japan (Imazeki et al.,1988), and between four (Breitanbach and Kränzlin, 1984) and30 species in Europe (Jacquetant, 1984). The latter study, however,stands in sharp contrast to a molecular phylogenetic assessment ofMorchella in Europe where only eight species were detected(O’Donnell et al., in review). Although Solak et al. (2007) recorded21 Morchella species in Turkey, all of the names used in this check-list were ones proposed for species from Europe and all of the iden-tifications were conducted using morphological data. Furthermore,it is impossible to determine to what extent our sampling overlapswith that of Solak et al. (2007) because the latter represents a sim-ple checklist without specimens or morphological data. Consistentwith the preliminary findings of Breitanbach and Kränzlin (1984),conducted on a small number of Esculenta and Elata Clade speciesfrom Switzerland, our analyses show that ascus and ascosporelength and width are too conserved to distinguish any of the spe-cies within Morchella included in our study (Supplemental Table2). Macromorphological data was not collected in the presentstudy because our phylogenetic results identified a number of mor-phologically cryptic species (see Fig. 2), suggesting that ascocarpmorphology is too homoplastic for distinguishing many of the spe-

cies included in our study. Nevertheless, a posteriori morphologicalanalyses of a broader sampling of Esculenta and Elata Clade speciesmay reveal morphological apomorphies that can be used to distin-guish some of the species.

One of the surprising results of the present study was that onlytwo Esculenta Clade species represented by five specimens wereamong the 247 collections, suggesting that members of this clademay be better adapted to northern temperate environments. Con-versely, the putatively high endemicity of members of the Mel-20-to-31 subclade and its sister Mel-25 in Turkey, suggests that theseeight species may have evolved novel adaptations associated withspeciation within the Mediterranean region. In a related study,multilocus typing of fusaria in Sardinian soils also revealed surpris-ingly high levels of endemics in this Mediterranean island notedfor its high biodiversity (Balmas et al., 2010). Our preliminary re-sults also suggest that the majority of Morchella species we sam-pled may not exhibit provincialism within Turkey, given that 10of the 12 species represented by two or more collections werefound in two or more regions. However, this hypothesis needs tobe tested by sampling in regions not included in the present study.Another surprising result was the scarcity of Mel-3 (=M. semilibera),the only species for which a Latin binomial could be applied withconfidence. Although this species has been treated within the sep-arate genus Mitrophora by some authors (Breitanbach and Kränz-lin, 1984; Wipf et al., 1999), our results show that it is deeplynested within the Elata Clade of Morchella. We are confident thatour Turkish collections represent M. semilibera because they matchcollections of this species made in Europe, which is where this spe-cies was initially described (O’Donnell et al., in review). Because M.semilibera is the only half-free morel found in Europe thus far, we

25steps

ITS + 28S rDNARPB1, RPB2, EF-1α4163 bp246 PIC1 of 88 MPTs295 stepsCI = 0.919RI = 0.971

106202203

212193

155458220125150

12342182

4618118318426016816519592

187191

10720725

118120122124156

15718865

Mel-31

Mel-30

Mel-29

Mel-28

Mel-20

Mel-27

Mel-26

Mel-25(outgroup)

100

100

100

100

84

69

99

99

98

98

90

97

95

100

76

Fig. 4. Phylogenetic relationships among seven members of the Elata Mel-20-to-31subclade inferred from MP analysis of the combined five-locus dataset for 36terminals, using sequences of Mel-25 to root the phylogeny. Collections areidentified by Hatira Tas�kin’s collection numbers. This dataset differs from the oneanalyzed for Fig. 3 by the inclusion of ITS rDNA sequence data (see Table 3). MPbootstrap values based on 1000 pseudo-replicates of the data are indicated aboveinternodes; ML bootstrap values from 500 pseudo-replicates of the data areindicated below internodes only when they differed by P5% of the MP value.

Table 7Bootstrap support for individual and combined 36-taxon Elata Clade Mel-20-to-31subclade dataset.

Speciesa ITS rDNA LSU rDNA RPB1 RPB2 EF-1a Combined

Mel-20 – 68 – 76 78 99Mel-25 100 98 100 100 100 100Mel-26 79 64 62 97 81 100Mel-27 96 – 100 97 95 100Mel-28 – – 86 95 66 99Mel-29 64 – 81 82 63 98Mel-30 NA NA NA NA NA NAMel-31 69 – – 60 98 97

a Mel, Morchella elata clade. As indicated by NA, bootstrap support could not beobtained for Mel-30 because it was represented by a single collection.

H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682 681

can make this connection without trying to obtain DNA data fromthe type specimen, which would probably be unproductive be-cause the type is 195 years old. Twelve of the species in our collec-tion from Turkey, based on available sampling, do not appear tohave been described formally. The remaining two species (Mel-20Elata Clade and Mes-8 Esculenta Clade) very likely have names, be-cause they are common in Europe, but we do not know whatnames to apply because there are many validly published namesfor black and yellow morels in Europe.

The nearly fully resolved multigene phylogeny inferred forTurkish collections of Morchella provides the first robust hypothe-

sis for evaluating traditional morphology-based classificationschemes and species diversity within this economically importantgenus, as well as providing a foundation for formulating soundconservation policies to help ensure sustainable morel harvestsfor this rapidly expanding cottage industry. In the near absenceof morphological apomorphies, advancements in the ecology andsystematics of Morchella will necessarily be heavily reliant on theresults of multigene molecular phylogenetics using GCPSR to iden-tify species limits (Taylor et al., 2000). The ability to discern indi-vidual species makes it possible for the first time to identifywhich morel species if any might have biotrophic or facultativelymycorrhizal associations with vascular plant hosts (Buscot, 1994;Hobbie et al., 2002). In addition, knowledge of their species bound-aries is a prerequisite for ascertaining which species are post-firemorels and whether they are obligatorily so. In this regard, our pre-liminary results suggest that Mel-10 may not be an obligate pyro-phile given that three of the 14 collections were made on non-burned sites dominated by P. brutia (Supplemental Table 1). Clearlymuch needs to be learned about the ecological roles played bypost- and non-post-fire morels with regard to their saprobic and/or mycorrhizal-like associations (Fujimura et al., 2005).

Perhaps the most interesting result of the present study was thediscovery that four putatively western North American endemicscomprised one-third of the Turkish collections. We hypothesizethat these species were introduced inadvertently in soil or wereassociated with the root tips of exotic forestry species such asPseudotsuga menziesii (Douglas fir) or Pinus species for use in Turk-ish forestry (Vellinga et al., 2009). However, whether and howthese morels might have been introduced remains a mystery fortwo reasons. First, no exotic tree species were observed at the siteswhere Mel-2, Mel-7, Mel-9 and Mel-10 were collected. Secondly,based on our preliminary data, Mel-2 and Mel-7 appear to be sim-ilarly diverse in western North America and Turkey. This resultstands in sharp contrast to Amanita phalloides, which is consider-ably more genetically diverse in its native range of Europe, com-pared to its introduced North American population (Pringle et al.,2009). Our working hypothesis is that the four exotic Morchellaspecies were introduced into Turkey multiple times from forestryplantations with a connection to western North America. Interest-ingly, of the four exotic morels, Mel-8 and Mel-10 and Mel-2 andMel-9 were recovered, respectively, from post- and non-post-firesites in Turkey and western North America, suggesting that theirreproductive strategies and ecological roles may be similar in bothcountries. Another area for future study is what impact the exoticmorels might have on native species. Given their successful estab-lishment and apparent ability to spread to regions dominated bynative tree species, studies are needed to assess whether they aredisplacing the native Morchella species. Lastly, the present studyadds to the field of mushroom conservation genetics pioneeredby Hibbett and Donoghue’s (1996) studies of Lentinula edodes. Aspromoted by these authors for shiitake, we strongly advocate usinga phylogenetic species concept based on genealogical exclusivity toidentify the full range of Morchella biodiversity. Such a concept isessential to help insure sustainable harvests of these economicallyimportant and charismatic macrofungi by developing robust man-agement practices and conservation policies within Turkey andelsewhere (Pilz et al., 2007).

Acknowledgments

Special thanks are due the Scientific and Technological ResearchCouncil of Turkey (TUBITAK) and the Çukurova University, Scien-tific Research Projects Coordinating Office (ÇÜ-BAP-ZF2009D41)for supporting the studies of HT at NCAUR, Stacy Sink for excellenttechnical assistance, Deb Palmquist for assistance with the statisti-cal analyses, Don Fraser for preparation of the publication figures,

682 H. Tas�kın et al. / Fungal Genetics and Biology 47 (2010) 672–682

and Nathane Orwig for running the DNA sequences in the NCAURDNA core facility. The mention of trade products or firm namesdoes not imply that the US Department of Agriculture recommendsthem over similar products or other firms not mentioned.

Appendix A. Supplementary material

Supplementary data associated with this article can be found, inthe online version, at doi:10.1016/j.fgb.2010.05.004.

References

Arora, D., 1979. Mushrooms Demystified. Ten Speed Press, Berkeley.Balmas, V., Migheli, Q., Scherm, B., Garau, P., O’Donnell, K., Ceccherelli, G., Geiser,

D.M., 2010. Multilocus molecular phylogenetics indicates high levels ofendemic fusaria inhabit Sardinian soils (Tyrrhenian Islands). Mycologia.doi:10.3852/09-201.

Berthet, P., 1964. Essai biotaxinomique sur les Discomycètes. Ph.D. Dissertation.Université de Lyon at Lyon.

Breitanbach, J., Kränzlin, F., 1984. Fungi of Switzerland. Ascomycetes, vol. 1.Mykologia, Luzern.

Bunyard, B.A., Nicholson, M.S., Royse, D.J., 1994. A systematic assessment ofMorchella using RFLP analysis of the 28S ribosomal RNA gene. Mycologia 86,762–772.

Bunyard, B.A., Nicholson, M.S., Royse, D.J., 1995. Phylogenetic resolution ofMorchella, Verpa, and Disciotis [Pezizales: Morchellaceae] based on restrictionenzyme analysis of the 28S ribosomal RNA gene. Mycol. Res. 19, 223–233.

Buscot, F., 1994. Ectomycorrhizal types and endobacteria associated withectomycorrhizas of Morchella elata (Fr.) Boudier with Picea abies (L.) Karst.Mycorrhiza 4, 223–232.

Buscot, F., Wipf, D., Di Battista, C., Munch, J.C., Botton, B., Martin, F., 1996. DNApolymorphisms in morels I: PCR/RFLP analysis of the ribosomal DNA spacersand microsatellite-primed PCR. Mycol. Res. 100, 63–71.

Dettman, J.R., Jacobson, D.J., Taylor, J.W., 2003a. A multilocus genealogical approachto phylogenetic species recognition in the model eukaryote Neurospora.Evolution 57, 2703–2720.

Dettman, J.R., Jacobson, D.J., Turner, E., Pringle, A., Taylor, J.W., 2003b. Reproductiveisolation and phylogenetic divergence in Neurospora: comparing methods ofspecies recognition in a model eukaryote. Evolution 57, 2721–2741.

Frøslev, T.G., Matheny, P.B., Hibbett, D.S., 2005. Lower level relationships in themushroom genus Cortinarius (Basidiomycota, Agaricales): a comparison ofRPB1, RPB2 and ITS phylogenies. Mol. Phylogenet. Evolut. 37, 602–618.

Fujimura, K.E., Smith, J.E., Horton, T.R., Weber, N.S., Spatafora, J.W., 2005. Pezizaleanmycorrhizas and sporocarps in ponderosa pine (Pinus ponderosa) afterprescribed fires in eastern Oregon, USA. Mycorrhiza 15, 79–86.

Geiser, D.M., Pitt, J.I., Taylor, J.W., 1998. Cryptic speciation and recombination in theaflatoxin-producing fungus Aspergillus flavus. Proc. Natl. Acad. Sci. USA 95, 388–393.

Gessner, R.V., Romano, M.A., Schultz, R.W., 1987. Allelic variation and segregation inMorchella deliciosa and M. Esculenta. Mycologia 79, 683–687.

Hansen, K., Pfister, D.H., 2006. Systematics of the Pezizomycetes – the operculatediscomycetes. Mycologia 98, 1029–1040.

Hansen, K., LoBuglio, K.F., Pfister, D.H., 2005. Evolutionary relationships of the cup-fungus genus Peziza and Pezizaceae inferred from multiple nuclear genes: RPB2,b-tubulin, and LSU rDNA. Mol. Phylogenet. Evol. 36, 1–23.

Hawksworth, D.L., 2004. Fungal diversity and its implications for genetic resourcecollections. Stud. Mycology 50, 9–18.

Hibbett, D.S., Donoghue, M.J., 1996. Implications of phylogenetic studies forconservation of genetic diversity in shiitake mushrooms. Conserv. Biol. 10,1321–1327.

Hobbie, E.A., Weber, N.S., Trappe, J.M., van Klinken, G.J., 2002. Using radiocarbon todetermine the mycorrhizal status of fungi. New Phytol. 156, 129–136.

Hofstetter, V., Miadlikowska, J., Kauff, F., Lutzoni, F., 2007. Phylogenetic comparisonof protein-coding versus ribosomal RNA-coding sequence data: a case study ofthe Lecanoromycetes (Ascomycota). Mol. Phylogenet. Evol. 44, 412–426.

Imazeki, R., Otani, Y., Hongo, T., 1988. Fungi of Japan. Yama-Kei Publishers Co., Ltd.,Tokyo (in Japanese).

Jacquetant, E., 1984. Les Morilles. Plantanida, Lausanne.Kellner, H., Renker, C., Buscot, F., 2005. Species diversity within the Morchella

esculenta group (Ascomycota: Morchellaceae) in Germany and France. Organ.,Divers. Evol. 5, 101–107.

Koufapanou, V., Burt, A., Taylor, J.W., 1997. Concordance of gene genealogies revealsreproductive isolation in the pathogenic fungus Coccidioides immitis. Proc. Natl.Acad. Sci. USA 94, 5478–5482.

Kuo, M., 2005. Morels. University of Michigan Press, Ann Arbor.

Liu, Y.J., Whelen, S., Hall, B.D., 1999. Phylogenetic relationships among ascomycetes:evidence from an RNA polymerase II subunit. Mol. Biol. Evol. 16, 1799–1808.

Matheny, P.B., 2005. Improving phylogenetic inference of mushrooms with RPB1and RPB2 nucleotide sequences (Inocybe; Agaricales). Mol. Phylogenet. Evol. 35,1–20.

Matheny, P.B., Liu, Y.J., Ammirati, J.F., Hall, B.D., 2002. Using RPB1 sequences toimprove phylogenetic inference among mushrooms (Inocybe, Agaricales). Am. J.Bot. 89, 688–698.

Miller, O.K., 1972. Mushrooms of North America. E.P. Dutton and Co., Inc., New York.O’Donnell, K., 1996. Progress towards a phylogenetic classification of Fusarium.

Sydowia 48, 57–70.O’Donnell, K., Cigelnik, E., Weber, N.S., Trappe, J.M., 1997. Phylogenetic relationships

among ascomycetous truffles and the true and false morels inferred from 18Sand 28S ribosomal DNA sequence analysis. Mycologia 89, 48–65.

O’Donnell, K., Cigelnik, E., Nirenberg, H., 1998. Molecular systematics andphylogeography of the Gibberella fujikuroi species complex. Mycologia 90,465–493.

O’Donnell, K., Kistler, H.C., Tacke, B.K., Casper, H.H., 2000. Gene genealogies revealglobal phylogeographic structure and reproductive isolation among lineages ofFusarium graminearum, the fungus causing wheat scab. Proc. Natl. Acad. Sci. USA97, 7905–7910.

O’Donnell, K., Lutzoni, F.M., Ward, T.J., Benny, G.L., 2001. Evolutionary relationshipsamong mucoralean fungi (Zygomycota): evidence for family polyphyly on alarge scale. Mycologia 93, 286–296.

O’Donnell, K., Rooney, A.P., Mills, G.L., Kuo, M., Weber, N.S., Rehner, A.S., in review.Phylogeny and historical biogeography of morel mushrooms (Morchella) revealsan early Cretaceous origin and high continental endemism and provincialism inthe Holarctic.

Ower, R., Mills, G.L., Malachowski, J.A., 1986. Cultivation of Morchella. US Patent4594,809.

Pilz, D., McLain, R., Alexander, S., Villarreal-Ruiz, L., Berch, S., Wurtz, T.L., Parks, C.G.,McFarlane, E., Baker, B., Molina, R., Smith, J.E., 2007. Ecology and Managementof Morels Harvested from the Forests of Western North America. GeneralTechnical Report PNW-GTR-710. Portland: US Department of Agriculture, ForestService, Pacific Northwest Research Station, 161 pp.

Posada, D., Crandall, K.A., 1998. MODELTEST: testing the model of DNA substitution.Bioinformatics 14, 817–818.

Pringle, A., Baker, D.M., Platt, J.L., Wares, J.P., Latge, J.P., Taylor, J.W., 2005. Crypticspeciation in the cosmopolitian and clonal human pathogenic fungus,Aspergillus fumigatus. Evolution 59, 1886–1899.

Pringle, A., Adams, R.I., Cross, H.B., Bruns, T.D., 2009. The ectomycorrhizal fungusAmanita phalloides was introduced and is expanding its range on the west coastof North America. Mol. Ecol. 18, 817–833.

Reeb, V., Lutzoni, F., Roux, C., 2004. Contribution of RPB2 to multilocus phylogeneticstudies of the euascomycetes (Pezizomycotina, Fungi) with species emphasis onthe lichen-forming Acarosporaceae and evolution of polyspory. Mol.Phylogenet. Evol. 32, 1036–1060.

Rehner, S.A., Buckley, E., 2005. A Beauveria phylogeny inferred from nuclear ITS andEF1-a sequences: evidence for cryptic diversification and links to Cordycepsteleomorphs. Mycologia 97, 84–98.

Royse, D.J., May, B., 1990. Interspecific allozyme variation among Morchella spp. andits inferences for systematics within the genus. Biochem. Syst. Ecol. 18, 475–479.

Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning. A LaboratoryManual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor.

Solak, M.H., Is�iloglu, M., Kalmıs�, E., Alli, H., 2007. Macrofungi of Turkey Checklist,vol. 1. Üniversiteliler Ofset, _Izmir.

Swofford, D.L., 2002. PAUP*. Phylogenetic analysis using parsimony (*and othermethods). Version 4.0b10. Sinauer Associates, Sunderland.

Taylor, J.W., Jacobson, D.J., Kroken, S., Kasuga, T., Geiser, D.M., Hibbett, D.S., Fisher,M.C., 2000. Phylogenetic species recognition and species concepts in fungi.Fungal Genet. Biol. 31, 21–32.

Vellinga, E.C., Wolfe, B.E., Pringle, A., 2009. Global patterns of ectomycorrhizalintroductions. New Phytol. 181, 960–973.

Weber, N.S., 1995. A Morel Hunter’s Companion: A Guide to the True and FalseMorels of Michigan. Thunder Bay Press, Lansing.

White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. Amplification and direct sequencing offungal ribosomal RNA genes for phylogenetics. In: Innis, M.A., Gelfand, H.H.,Sninsky, J.J., White, T.J. (Eds.), PCR Protocols: A Guide to Methods andApplications. Academic Press, San Diego, pp. 315–322.

Wipf, D., Bedell, J.P., Munch, J.C., Botton, B., Buscot, F., 1996. Polymorphisms inmorels: isozyme electrophoretic analysis. Can. J. Microbiol. 42, 819–827.

Wipf, D., Fribourg, A., Munch, J.C., Botton, B., Buscot, F., 1999. Diversity of theinternal transcribed spacer of rDNA in morels. Can. J. Microbiol. 45, 769–778.

Zwickl, D.J., 2006. Genetic algorithm approaches for the phylogenetic analysis oflarge biological sequence datasets under the maximum likelihood criterion.Ph.D. Dissertation, The University of Texas at Austin.