RFLP analysis reveals more isolates belonging to new Neurospora species Christopher F. Villalta1, David J. Jacobson2, John W. Taylor3

Department of Plant and Microbial Biology, UC Berkeley 1. cvillalta@berkeley.edu 2. djjacob@berkeley.edu 3. jtaylor@nature.berkeley.edu

Abstract Recent studies using phylogenetic species recognition (PSR) and biological species recognition (BSR) found three new Neurospora species, phylogenetic species (PS) 1, 2, and 3 (Dettman 2003a). However, in that study there were too few individuals found for each species to provide a confident formal description. A more comprehensive search of the existing Perkins culture collection was made to find more individuals of each of these species. A method using RFLPs of the markers used for PSR was devised to easily distinguish members of the new species (PS1-3) from the closely related species N. crassa and N. intermedia. The RFLP analysis was performed on isolates that were either not clearly identifiable by BSR (crosses with tester strains), or were from the same geographic locations as PS1-3. Additional members of all these species were identified: 1 new PS1, 9 new PS2, and 1 new PS3. BSR in and among PS1-3 was investigated by performing intraspecific and interspecific crosses. PS1 and PS2 appear reproductively isolated, successfully mating intraspecifically while not mating successfully with other species of Neurospora. The additional PS3 strain successfully crossed with other PS3 isolates, however, it also successfully crossed with N. crassa confirming PSR but not BSR between PS3 and N. crassa (Dettman 2003b). We are confident that PS1-3 are indeed new species, and they will be named N. hispaniola (PS1), N. metzenbergi (PS2), and N. perkinsi (PS3).

Introduction • Originally there were believed to be five conidiating species of Neurospora, but that number has now grown to 15

species (Dettman 2003a, 2006).

• The heterothallic species have traditionally been identified by crossing unknown specimens with known tester strains and measuring the success of mating (BSR) while the pseudohomothallic N. tetrasperma is identified morphologically.

• A thorough comparison in 2003 of species recognition by genetic isolation (PSR) with species recognition by reproductive isolation (BSR) found three new species, all of which were genetically isolated and two of which were also reproductively isolated (Dettman 2003a,b).

• The isolates used were from the David Perkins culture collection (Turner 2001) and the Fungal Genetics Stock Center (FGSC). New acquisitions to the collection, from nature, were identified by mating them to tester strains (BSR).

• Several of the specimens collected were never properly identified because they did not mate well with tester strains or show the expected mating patterns (Turner 2001).

• Since Neurospora PS1-3 contained a disproportionate fraction of what had been thought to be hybrid individuals (Turner 2001) and because the new species were narrowly endemic we decided to go back to the collections of natural specimens to search for more individuals of the new species.

Figure 1. Geographic locations where the new species PS1 (Hispaniola), PS2 (Yucatan, Madagascar), and PS3 (Congo) are endemic.

PS2 (Yucatan)

PS1 (Hispaniola)

PS2 (Madagascar) PS3 (Congo)

• We identified 188 natural isolates that were either difficult to assign to species due to equivocal crosses to mating testers or that were collected in the geographic areas occupied by Neurospora PS1-3, or both.

• We developed a PCR and RFLP screen to rapidly exclude genuine members of N. crassa or N. intermedia.

• Isolates passing the screen were then subjected to phylogenetic analysis for assignment to species.

• To further understand the relationship of genetic isolation to reproductive isolation, isolates found to belong to Neurospora PS 1-3 were mated among themselves and to individuals of N. crassa and N. intermedia.

• The information on genetic and reproductive isolation will be used to describe Neurospora PS 1-3 as new species.

Methods

Table 1. Two separate RE (restriction enzyme) digests were used in the RFLP analysis to differentiate PS 1-3 from N. crassa and N. intermedia specimens. NciI (NEB) does not cut N. crassa at the 102 bp TMI position allowing for the discarding of N. crassa specimens from analysis. BciVI (NEB) applied to remaining non-N. crassa samples does not digest N. intermedia at the 240 bp TMI position leaving behind possible PS1-3 specimens.

PCR and Restriction enzyme digest •  TMI locus PCR products were used in the RFLP screen.

Sequencing of Informative Loci • Possible Neurospora PS 1-3 candidates from the screen were sequenced for the DNA regions used to

recognize the phylogenetic species (DMG, QMA, TMI, and TML loci) (Dettman 2003a).

Tree building and species identification • Sequence data from all four loci were aligned together in one consensus file.

Using Mr.Modeltest 3.1.7 (Nylander 2004) with Paup 4.0 (Swofford 2003) the appropriate nucleotide substitution model was chosen.

•  Phylogenetic trees were built using Mr. Bayes (Bayesian Inference) (Huelsenback 2001) and Garli (Maximum Likelihood) (Zwickl 2006).

•  Unknown specimens were identified by their placement in relation to the known isolates.

Design of crossing matrix and mating of isolates • Matings (133 crosses) and evaluation of reproductive success were done following previously

published protocols from Dettman et al. 2003b.

Figure 2. Screen to exclude N. crassa. (a) and (b) displays how the N. crassa (D11, D12), N. intermedia (D7, D31), PS1 (D57), PS2 (D93), and PS3 (D77) controls were digested with the Nci I. All TMI loci PCR products were digested as expected. Both (a) and (b) also include digests of experimental samples. In (a) CV113 and CV146 would have been identified as N. crassa and be removed from further analysis.

Figure 3. Screen to exclude N. intermedia. BciV I digest of the N. crassa (D11, D12), N. intermedia (D7, D31), PS1 (D57), PS2 (D93), and PS3 (D77) controls. Experimental samples are also displayed in the figure. CV95, CV96, CV97,CV99, and CV100 were not digested by BciV I and were thus identified as putative N. intermedia. The remaining samples were identified as putative PS 1-3. There were expected to be no N. crassa isolates among the experimental samples because those were filtered out of the analysis in the first digest.

(a) (b)

Results RFLP Screen for Phylogenetic Species •  After analyzing Dettman et al. 2003a alignments we found that the TMI locus here and throughout had several sites

that were uniform within the species of N. crassa, N. intermedia, and PS1-3 (treated as one uniform group in RFLP analysis), but that differed interspecifically.

• Using NEBCutter 2.0 (Vincze 2003) we searched for variable sites among the three groups of interest that could be digested by restriction enzymes and found that NciI and BciVI could be used for our purposes.

Phylogenetic Species Recognition • Of the 21 putative PS, 10 were identified as PS2 and one isolate each was identified as PS1 and PS3 (Figure 5).

• The nine remaining individuals proved to be N. sitophila (false positives), because our screen was not designed to screen for N. sitophila.

• To search for false negatives from our screen, we sequenced the TMI loci of three N. crassa isolates and seven N. intermedia isolates. None of the excluded individuals proved to be PS1, PS2, or PS3, however one was identified as belonging to N. discreta sensu lato.

188 unknown

+

NciI

42 N. crassa

146 unknown

+

BciVI 21 putative PS 1-3

125 N. intermedia

Figure 4. Work flow and results of the RFLP screen.

Phylogenetic Species BSR Matrix •  A matrix containing the 130 new crosses and 68 crosses taken from Dettman et al. 2003b is shown in Table 2. The

new PS 1-3 all mated well intraspecifically, but while PS1 and PS2 still appears to be reproductively isolated, PS3 is not because it still mates successfully with N. crassa.

Figure 5. The phylogeny above represents the phylogenetic relationships between the 12 specimens found to be putative PS1-3 in the RFLP analysis and 145 specimens already identified by PSR from Dettman et al. 2003a. The tree was made using Mr. Bayes and the major branches defining each PS are marked in bold. The numbers above each major branch display our confidence that the demarcation is real and are all > 0.50 (Bayesian Posterior Probability/Maximum Likelihood Bootstrap Proportions). Specimens in bold are those that were found to be putative PS1-3 with the RFLP analysis. The new specimens in bold and denoted with a (H) were those found to belong to PS1, PS2, and PS3.

Table 2. BSR matrix of crosses between the new PS and previously identified Neurospora specimens. The grading criteria are the same as the criteria used in Dettman et al. 2003b. Cells for intraspecific crosses are bordered by thick lines and have bold face type. The majority of intraspecific crosses were successful displaying cohesion between BSR and PSR. Crosses with an (*) are those taken from Dettman et al. 2003b. Reciprocal crosses were performed in every cell and the data are ordered as follows: mat a (parent)/ mat A (parent).

6 >50% black ascospores5 15-50% black ascospores

3&4 <1% black ascospores; & 1-15% black ascospores 2 perethecia developed ostioles, no ascospores ejected

0&1 sterile, no perethecia produced; & barren perethecia, no ostiole developed

PS1 PS2 PS3 Nc NimatA ---> CV55 D58 CV152 CV155 CV156 D120 D78 D107 D62 D122 D36

mataPS1 D55 6/6 6/6* 1/1 6/1 1/5 0/4* -* 0/3 -* 0/0* -PS2 CV89 6/0 6/1 6/6 6/5 6/5 6/6 5/5 5/6 4/5 4/1 1/5

CV119 1/1 1/2 2/5 5/4 4/5 5/6 6/6 6/5 6/6 1/1 6/5CV148 0/5 5/1 6/6 6/6 6/6 5/6 5/5 4/4 4/4 1/5 5/6D92 1/5 3/0* 5/5 6/5 5/4 5/5* 3/3* 4/4* 3/3* 4/1 -*

PS3 CV79 1/1 1/1 6/6 6/1 2/5 5/6 6/6 6/6 6/6 1/1 6/6D77 1/4 3/0* 5/5 5/4 5/4 3/3* 6/6* 5/5* -* 3/0* -*

Nc D100 4/1 3/0* 5/4 5/4 2/6 3/3* 5/5* 6/6* 4/4* 3/2* -*D116 1/4 3/0* 6/6 6/5 4/1 3/3* 6/6* 6/5* 6/6* 0/0* 3/3*

Ni D127 0/1 1/0* 2/2 2/1 1/2 3/3* 3/3* 3/1* 1/2* 5/6* 6/6*D2 0/3 -* - 2/1 1/1 -* -* 3/3* -* 6/6* 6/0*

Discussion We found additional members of Neurospora PS 1-3. Biasing our sampling toward individuals for which mating tests were equivocal or individuals that had been collected in endemic areas and then screening to exclude N. crassa or N. intermedia individuals saved labor and supplies.

Including the new individuals of Neurospora PS 1-3 in intra- and interspecific matings showed results similar to crosses using the original members of Neurospora PS 1-3 (Dettman 2003b). Neurospora PS 1 and 2 are reproductively isolated from each other and described Neurospora species, Neurospora PS 3 is not isolated from N. crassa.

We found that nine of the putative new members of PS 1-3 identified by our screen were truly N. sitophila. Our screen could be improved to account for N. sitophila and, thereby, reduce the number of false positives. We were unable to detect any false negatives.

There is room to discover more individuals of these Neurospora species, particularly from Neurospora PS1 and PS3. Neurospora PS2 has the most interesting geographic distribution, with all individuals coming from either Yucatan or Madagascar. Genetic isolation in this PS correlates with geographic range, but our data show that individuals from the two geographic areas are not reproductively isolated. Finding more Madagascar individuals is a priority for understanding Neurospora PS2.

Adding to the existing phylogeny the newly discovered Neurospora PS 1-3 individuals did not change the tree topology. As a result, we now feel confident in providing Neurospora PS 1-3 with proper names. We hope that the descriptions of three new Neurospora PS will add to the attractiveness of Neurospora as a model for evolutionary biology. The names we have chosen for the new species are N. hispaniola (PS1), N. metzenbergi (PS2), and N. perkinsi (PS3).

Literature Cited Dettman, J. R., D. J. Jacobson, et al. (2003). "A multilocus genealogical approach to phylogenetic species recognition in the model eukaryote Neurospora." Evolution Int J Org Evolution 57(12): 2703-20. Dettman, J. R., D. J. Jacobson, et al. (2006). "Multilocus sequence data reveal extensive phylogenetic species diversity within the Neurospora discreta complex." Mycologia 98(3): 436-46. Dettman, J. R., D. J. Jacobson, et al. (2003). "Reproductive isolation and phylogenetic divergence in Neurospora: comparing methods of species recognition in a model eukaryote." Evolution Int J Org Evolution 57(12): 2721-41. Huelsenbeck, J. P., F. Ronquist, et al. (2001). "Bayesian inference of phylogeny and its impact on evolutionary biology." Science 294: 2310-2314. Nylander, J. A. A. (2004). MrModeltest Uppsala, Sweeden, Evolutionary Biology Centre, Uppsala University. Turner, B. C., D. D. Perkins, et al. (2001). "Neurospora from natural populations: a global study." Fungal Genet Biol 32(2): 67-92. Vincze, T., J. Posdai, et al. (2003). "NEBcutter: a program to cleave DNA with restriction enzymes." Nucleic Acids Research 31: 3688-3691. Zwickl, D. J. (2006). Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Biology. Austin, The University of Texas at Austin. Ph.D.