Keywords:Claviceps,Epichloë,Periglandula, Clavicipitaceae, gene clusters, chanoclavine, ergopeptine, subterminal, natural products, secondary metabolism

2.1. Genes Encoding the Early Pathway Steps

The four conserved early pathway steps for all natural ergot alkaloids produce chanoclavine I (CC) (recently reviewed [20]), and are catalyzed by enzymes encoded in the four genes:dmaW, encoding dimethylallyltryptophan synthase,easF, encoding dimethylallyltryptophanN-methyltransferase,easC, encoding a catalase, andeasE, encoding chanoclavine-I synthase. Some strains, such asE. elymi E56, produce CC as the pathway end-product because they contain functional copies only of these four genes in anEAS cluster, which we designate asEAS^CC (Figure 2;Table 1). Based on similar complements in their genomes, we would predict thatE. brachyelytri E4804 andAtkinsonella hypoxylon B4728 are also CC producers, though CC has not been detected in plants infected with B4728. Strains capable of making complex ergot alkaloids also can generate substantial levels of CC and other intermediates and spur products as a result of inherent pathway inefficiency, a property suggested to have selective advantage [21,22].

2.2. Diversification of the EAS Pathways

Beyond the production of CC, multiple biosynthetic pathways begin to branch and diverge (Figure 1), and the chemotypic variation between and even within species is reflected in the gene content of eachEAS locus known or predicted to direct biosynthesis of such pathway end-products as elymoclavine (EC), lysergic acid α-hydroxyethylamide (LAH), ergonovine (EN), and ergopeptines such as ergovaline (ERV), ergotamine (ERA), ergocryptine (ERK) or ergobalansine (ERB). Where clarification is needed, we will designate the variousEAS clusters with superscripts reflecting end products, asEAS^EC,EAS^LAH orEAS^EN, as well asEAS^ERP for ergopeptine producers,EAS^EN/ERP for producers of EN and ergopeptines, andEAS^LAH/ERP for producers of LAH and ergopeptines. Compared toEAS^EN,EAS^LAH has two additional genes,easO andeasP, suggesting that EN may be the LAH precursor. Fungi withEAS^LAH clusters also tend to accumulate substantial levels of ergine, probably by spontaneous hydrolysis of LAH [23]. TheEAS^EN/ERP cluster in the most infamous ergot fungus,Claviceps purpurea, and theEAS^LAH/ERP cluster in the morning-glory symbiont,Periglandula ipomoeae, are the only ones identified to date that determine synthesis of three different ergot alkaloid subclasses [24].

2.3. Contents of the EAS Loci

The EAS pathway specificity mentioned above is reflected in the structural content ofEAS loci across the Clavicipitaceae, which varies based on presence or absence of genes that correspond to ergot alkaloid biosynthetic capability within a given strain. The mostEAS genes (14) are present inP. ipomoeae, which produces ergobalansine (ERB), EN and LAH, whereas only fourEAS genes are common to all strains that are capable or predicted to produce CC [24]. Interestingly, those four early-pathway genes are grouped together in theN. fumigata cluster, whereas the mid-pathway genes for festuclavine biosynthesis are interspersed with late-pathway genes for fumigaclavine [32,33] (Figure 2). The gene arrangements differ betweenN. fumigata and the Clavicipitaceae, and also differ considerably betweenEpichloë spp. and other Clavicipitaceae. InAt. hypoxylon,Balansia obtecta, theClaviceps spp.,Epichloë inebrians (formerlyEpichloë gansuensis var.inebrians; [34]) andP. ipomoeae,early and mid-pathway genes (except foreasA) are interspersed, whereas the late genes (lps genes,easH,easO andeasP) are at theEAS-cluster periphery, separated from the mainEAS cluster, fatally mutated or missing. TheeasA gene is the exception among mid-pathway genes, being located between late-pathway geneslpsB andlpsC in these species. In the otherEpichloë species, there is similar variation in gene content and functionality, but greater variation in gene arrangement even for early- and late-pathway genes. Clearly, strains that produce only CC are derived by losses of mid-and late-pathway genes, because remnants and pseudogenes often remain in their genomes.

PutativeEAS genes can be identified in fungal genome sequences using bioinformatics pipelines developed to identify core secondary metabolite genes encoding NRPSs, polyketide synthases (PKS), prenyltransferases (dmaW homologues) or terpene cyclases and mixed or hybrid versions [35,36,37]. TheEAS gene clusters such asEAS^CC can sometimes be identified within the prenyltransferase or terpene cyclase group, whereas a more complexEAS cluster, such asEAS^ERP, will fall into a hybrid or mixed category since they contain sequences for both prenyltransferases and NRPSs [38]. In genomes of severalTrichophyton andArthroderma species (Arthrodermataceae), apparent orthologues ofdmaW,easF,easC,easE andeasD are identifiable in a cluster [18]. Genes flanking this cluster match signatures of various biosynthetic functions, so it may well be that Arthrodermataceae produce one or more alkaloid subclasses yet to be identified. TheEAS gene complements and structural arrangements identified inMetarhizium robertsii [39] are very similar toE. inebrians and, as such, we would predict thatM. robertsii strain ARSEF 23 could be capable of producing EN and LAH, assuming there is similar specificity of LpsB with LA and LpsC with alanine.

3.1. Comparison ofEAS Gene Phylogenies

We have inferred phylogenetic trees forEAS genes of known and suspected ergot alkaloid-producing Clavicipitaceae, plusN. fumigata, which we presumed to be an outgroup in keeping with housekeeping gene relationships. Phylogenies of the core genes for the first four steps in ergot alkaloid biosynthesis—namely,dmaW,easF,easC andeasE—were congruent, so a concatenated data set (WFCE) was constructed, from which a maximum likelihood (ML) tree was inferred with PhyML at phylogeny.fr [40] (Figure 3). ThelpsB tree (Figure 4) was also congruent with the correspondingWFCE subtree (i.e., the tree pruned of taxa lackinglpsB). TheWFCE andlpsB phylogenies were significantly supported at all nodes. TheWFCE phylogeny grouped sequences from seven out of eightEpichloë species in a clade that was basal in the Clavicipitaceae, whereas the sequences from the eighth representative,E. inebrians, appeared as the sister to theClaviceps clade. Comparing this phylogeny to that of the housekeeping gene,tefA (Figure 5), the only significant disparity was the basal placement of the main clade ofEpichloë EAS genes, which contrasted with thetefA phylogeny that placed all of theEpichloë species together in a clade with a sister relationship to theClaviceps clade. There was also a possible disparity in placement ofP. ipomoeae, but this cannot be considered significant because the relevant branch in thetefA tree lacked statistical support. Therefore, with the glaring exception of the mainEpichloë EAS clade, evolution ofEAS genes in the Clavicipitaceae appears to have been by direct decent without duplication (paralogy) or horizontal gene transfer.

The well-supported position ofEAS genes from mostEpichloë species, being basal among the Clavicipitaceae (Figure 3 andFigure 4) indicates a deviation from strict orthology because it differs dramatically from housekeeping gene phylogenies (Figure 5). Possible causes of this deviation are trans-species polymorphism, paralogy or horizontal gene transfer. In a BLASTp search ofdmaW against available sequences at GenBank, no homologues were closely related to this clade except those of otherEpichloë species, so there was no obvious source for a horizontal gene transfer event. Comparing the genomic context, sequences nearest theEAS clusters differed betweenE. festucae andE. inebrians (assemblies of otherEpichloë species did not linkEAS clusters with other genes), so trans-species polymorphism also was unsupported. This leaves, as our favored possibility, that theE. inebrians and otherEpichloë EAS clusters were derived from paralogous copies that arose from duplication of theEAS cluster in an ancestor to most or all of the Clavicipitaceae. In this regard, thetefA phylogeny (Figure 5) placedE. inebrians basal in genusEpichloë, supporting the possibility that the cluster common to mostEpichloë species was lost on that basal branch toE. inebrians. However, a puzzle remains in that no genus other thanEpichloë showed indications of paralogousEAS clusters. This may be just a matter of limited sampling, and we predict that a wider and deeper survey of the Clavicipitaceae will reveal paralogs related to the basal group ofEpichloë EAS sequences.

ParalogousdmaW genes are in fact evident in some of the Clavicipitaceae. Specifically,C. purpurea 20.1 has twodmaW copies flanking a paralogouseasF and located 94 kb and 98 kb from theEAS-clusterdmaW.Epichloë mollis also has a paralogousdmaW, which appears to be a pseudogene. However, these paralogues are due to relatively recent duplications and group with the respectiveClaviceps andEpichloë dmaW genes in phylogenetic analysis [43].

3.2. Mapping EAS Gene Gains and Losses

Mapping genomic alternations onto theWFCE phylogeny (Figure 3) revealed repeated instances in which multipleEAS genes have been lost. The most extensive losses were eight genes in two separate instances resulting in gene sets for CC production: The branch toE. elymi andE. brachyelytri, and the branch toAt. hypoxylon. The identification of remnant or pseudogene copies of otherEAS genes supported the scenario of extensive gene loss, as inferred from the phylogeny. Losses of multipleEAS genes on numerous lineages have given rise to at least four distinct chemotypes in addition to the variations in ergopeptines. These gene losses add to the potential for ergot alkaloid diversification, together with neofunctionalization oflpsA (Table 2), and witheasA variations to give agroclavine or festuclavine as precursors of ergot alkaloids and dihydroergot alkaloids, respectively, andeasG variations to yield 8(S)-dihydroergot alkaloids (Figure 1) [17].

Interestingly, in almost all cases ofEAS gene loss (Figure 3), all genes were lost or inactivated for a branch of the pathway, leaving only thoseEAS genes required for biosynthesis of the observed metabolites. In two instances, this process has given CC as the end product (or presumed end product), in one instance it has given EN rather than LAH, and in three instances it has eliminated the ergopeptine pathway. The only exception to this pattern is inC. fusiformis, which has apparently retainedeasH despite losing all other genes for late pathway steps both to ergopeptines and to EN and LAH. However, whethereasH is transcribed or gives an active (but presumably useless) product inC. fusiformis is unknown. Among many natural strains, gene losses are similarly evident in indole-diterpene (IDT) and loline alkaloid (LOL) gene clusters with loss or inactivation of all other genes for downstream steps of the affected pathway branch [24,44,45]. Such common patterns suggest that there is selection against expression of enzymes in strains that lack their normal substrate. We speculate that expression of such enzymes may be directly harmful because the enzymes may catalyze reactions with substrates other than the missing natural substrate, thereby generating toxic products.

3.3. Positional Changes of Clusters with Respect to Telomeres

Also mapped to theWFCE phylogeny (Figure 3) are three changes in theEAS cluster positions relative to telomeres. Considering that theN. fumigata and basalEpichloëEAS genes are near telomeres (“subterminal”), it is parsimonious to propose this as the ancestral state. If so, internalization would have occurred on the branch that separates most of the Clavicipitaceae from the basalEpichloë EAS genes. An interesting translocation event is evident on the branch toE.inebrians, whereby theEAS cluster was divided and at least one of the two resulting portions returned to a subterminal position. Interestingly, thecloA gene (designatedB on the map inFigure 2) is situated such that its stop codon is a single base away from the telomere repeat array, implying that the telomere serves as a transcription terminator for that gene. The translocation event on theE. inebrians branch nicely illustrates the dynamics of subterminal regions of the chromosomes. Recalling that theIDT cluster—directing biosynthesis of the indole-diterpene, paxilline—is subterminal in the closely relatedE. gansuensis strain E7080 [24], it is particularly interesting to find two remnantIDT genes flanking the centromeric side of the subterminalEAS cluster inE. inebrians. It appears likely that theEAS cluster displaced most of theIDT cluster as it moved into the subterminal position. This supports an important role of subterminal regions in arrangements and rearrangements of secondary metabolism gene clusters. With that in mind, we would predict that mostEAS cluster rearrangements occur in association with subterminal clusters, whereas internalEAS clusters are more stable. Evidence pertinent to that prediction is discussed later, inSection 5.

3.4. Evolution of LpsC

Assuming that the common ancestor of Clavicipitaceae and Trichocomaceae possessed the seven genes for early and intermediate biosynthetic steps, a parsimonious scenario would identify three branches for gene gains: one toN. fumigata, one to all Clavicipitaceae, and the third to Clavicipitaceae after splitting from the basalEpichloë EAS clade (Figure 3). In the context of our proposed paralogy, the last of these would have involved acquisition of the new genes,lpsC,easO andeasP, on one of the branches after that split in the Clavicipitaceae. Interestingly, LpsC serves as an alternative to the LpsA subunit in interacting with LpsB as a component of a lysergyl peptide synthetase complex (Figure 1). A phylogenetic analysis of Lps subunit A domains (Figure 6) indicates thatlpsC may be derived from a copy of the first module oflpsA fused to an R* (reductase) coding sequence. Thus, biosynthesis of EN and LAH seems to have evolved later than biosynthesis of the much more complex ergopeptines.

3.5. Evolution of Module Specificity in LpsA

The evolution of module specificity in LpsA subunits can also be mapped onto theWFCE (Figure 3) andlpsA A-domain (Figure 6) phylogenies. However, a crucial difference betweenWFCE andlpsA phylogenies is thatlpsA A-domains consistently groupP. ipomoeae andB. obtecta in a terminal clade, whereasWFCE groupsP. ipomoeae withClaviceps spp. and placesB. obtecta in a relatively basal position. This may be the result oflpsA paralogy due to duplications and losses (a scenario reminiscent of the tandemly duplicatedlpsA genes inC. purpurea), or may simply be due to lineage sorting effects in the evolution of these lineages (as suggested by the short and poorly supported branch in thetefA phylogeny). To discuss LpsA module evolution, we now use single-letter abbreviations for the three specified amino acids in order of the modules 1, 2 and 3. The ancestral state could have been ALP or AVP, with module A2 changing specificity either on the basalEpichloë branch or on the lineage leading to theClaviceps/Periglandula/Balansia clade, respectively. During evolution of thePeriglandula/Balansia clade, specificity of module 3 switched from P to A. What is especially interesting is that, withinClaviceps, there is an accelerated diversification of modules 1 and 2, such that the variant LpsA subunits specify the substrate combinations for at least 19 different ergopept(in)ines (Table 2).

4.1. Distribution of EAS Genes across Epichloë Species

The distribution of theEAS locus is well understood withinEpichloë species due to the availability of genomic and genotyping data (www.endophyte.uky.edu). Genome sequences (including draft genomes) for 54 isolates representing 10 describedEpichloë species, two varieties, and five undescribed species have enabled comparisons of theEAS locus across a diverse collection of isolates with varying capabilities to produce ergot alkaloids [24,34,38,43,45,46]. Moreover, in recent studies, markers developed from the genome sequences have been used for PCR-based genotyping of endophyte-infected grass collections for the presence of genes encoding key alkaloid biosynthesis pathway steps [10,47,48,49,50,51,52]. Compared to genome sequencing, PCR-based genotyping is less reliable in determining if a gene is functional, but the presence of full-length genes typically associated withEAS clusters for the metabolites CC, ERV or EN usually corresponds well to the production of the corresponding metabolite. The predictive power of this method is helped by the tendency for complete losses or large deletions in the nonfunctional alkaloid biosynthesis genes, and for the downstream genes to be partly or completely deleted as well [24].

The presence ofEAS genes withinEpichloë species has been observed in strains of at least 19 (9 nonhybrid and 10 hybrid species) of the 35 taxa tested (including varieties and undescribed species) (Table 3). TheEAS locus has a discontinuous distribution withinEpichloë species. For example, not allE. festucae isolates are capable of producing ergovaline, since some isolates (e.g.,E. festucae E434) lack theEAS locus. However, in addition to their discontinuous distribution, theEAS loci can differ in structural content. Strains capable of producing CC (e.g.,E. elymi E56) only contain functional genes encoding the first fourEAS pathway steps (DmaW, EasF, EasC and EasE) (Figure 1;Table 1), whereas ergovaline producers (e.g.,E. festucae Fl1) have more complexEAS loci (EAS^ERP) with at least 11 functional genes present (Table 1).

Many of the chemotype differences identified within theEpichloë species can be attributed to the presence or absence of genes encoding the key pathway steps, but there are some anomalies. Isolates, such asE. festucae strains E2368 and E189,E. amarillans E4668 andE. glyceriae E2772 appear to have completeEAS^ERP clusters with no obvious deleterious mutations, yet ergot alkaloids have not been detected in host plants symbiotic with these isolates. Transcriptome (RNA-seq) analysis of E2368-infected meadow fescue (Lolium pratense) and tall fescue show that theEAS genes are not expressed in this strain, thus providing a clue to why the predicted alkaloid, ERV, is not produced [24].

Many endophytes can readily produce ergot alkaloidsin planta but have a more limited and unreliable production in non-symbiotic culture conditions [53,54]. Histone methylation apparently helps repress expression ofEAS and other alkaloid gene clusters whenEpichloë species are grown in culture, since theEAS andLTM genes ofE. festucae strain Fl1 were de-repressed under non-symbiotic culture conditions when two genes that encode histone H3 methylases were deleted [55].

4.2. Pseudogenes and Gene Remnants within the EAS Locus

Pseudogenes and gene remnants have been identified within or adjacent to many of the describedEAS clusters within the Clavicipitaceae. MostEpichloë species that are unable to produce ergot alkaloids lack functionalEAS genes or only retain remnants of someEAS genes. For example, a remnantlpsA sequence is present in the genome sequences ofE. amarillans strain E57,E. bromicola strain AL0434, andEpichloë coenophiala strains e4509, AR542 and AR584. Typically theselpsA remnants have multiple frameshifts, stop codons or both, and are flanked or even disrupted by AT-rich repeat sequences; or, in the case of E57, the gene is truncated at the telomere [24]. In comparison toE. bromicola strain AL0434, which has only anlpsA remnant, strain AL0426/2 has retained a greater number ofEAS genes (Figure 7). TheE. bromicola AL0426/2 genome assembly (www.endophyte.uky.edu) containslpsB,easE,easF andeasG (contig 634), buteasE is a pseudogene in which part of the coding sequence is missing. SincedmaW andeasC are also missing, the AL0426/2 isolate is not expected to produce ergot alkaloids. The identification of two independentEAS gene loss events inE. bromicola (strains AL0434 and AL0426/2) suggests that this species may have greaterEAS diversity than is currently recognized, butE. bromicola strains capable of producing ergot alkaloids have not yet been identified [61].

4.3. Hybrids: EAS Gene Cluster Variations

Compared to sexual strains or other haploids, the hybridEpichloë species have a greater potential to containEAS genes because each of the ancestors contributes genetic information. In addition, variations of theEAS locus within a hybrid can be due to copy number or differences within the inheritedEAS cluster (EAS^CC vsEAS^ERP clusters). Some, but not all isolates from the hybrid species,Epichloë canadensis,E. coenophiala, andE. sp. FaTG-2, contain twoEAS copies (Figure 8). TheE. canadensis isolate e4815 contains twoEAS clusters—EAS^ERP for ERV andEAS^CC—that are representative of the two contributing ancestors,E. amarillans andE. elymi, respectively. In contrast,E. canadensis isolate CWR34 contains a singleEAS^CC cluster, contributed by itsE. elymi ancestor. Mating-type gene differences betweenE. canadensis isolates e4815 and CWR34 clearly indicate that they are the result of independent hybridizations, but it is unclear if CWR34 subsequently lost theE. amarillans EAS cluster, or if its particular ancestral strain ofE. amarillans lackedEAS genes.

4.4. Gene Losses in Hybrids with Multiple Copies

Some isolates ofE. coenophiala are unable to produce ergot alkaloids because they only contain a remnantEAS locus (e.g., e4309, AR542 and AR584;Table 3) [51]. TheE. coenophiala isolates known to produce ERV, such as isolates e19 and e4163, have two copies of theEAS clusters from two of the three contributing ancestors:E. festucae and theLolium-associatedEpichloë (LAE) subclade. Interestingly, theEAS clusters of bothE. coenophiala isolates e19 and e4163 are structurally very similar with respect to AT content and repeat sequences, though they differ in whichEAS genes are absent or inactive pseudogenes in eachEAS cluster (Figure 9). ThelpsB2 gene in e19 (from theE. festucae ancestor) is nonfunctional due to a frame shift within the coding region. The situation in e4163 is more complex in that neitherEAS cluster is complete;EAS1 (from LAE) lackslpsB1 andeasE1, andEAS2 (fromE. festucae) includes a nonfunctionaleasG2. Therefore, since e4163 is able to produce ERV, eachEAS cluster must functionally complement the genetic deficiencies of the other cluster.Epichloë sp. FaTG-2 isolates NFe45115 and NFe45079 are both capable of producing ERV, butEAS copy numbers differ; NFe45115 has one copy, whereas NFe45079 has two copies [49]. Recent genome sequence data have also revealed that NFe45079 contains only a single copy oflpsA, but all other genes are present in two copies.

4.5. Endophyte Genetic Variation within a Single Host Species

The associations of endophytes and hosts represent co-evolving associations as evidenced by the tendency for evolution of theEpichloë spp. to track evolution of their hosts [62]. It is interesting to note that sometimes a single host species has symbiotic associations with more than one endophyte species, but each individual plant will only host one endophyte genet. Tall fescue can be symbiotic withE. coenophiala,Epichloë sp. FaTG-2, FaTG-3 and FaTG-4, and strains representing each of these endophytes have different alkaloid profiles [43,51,60].Elymus canadensis can be symbiotic withE. elymi andE. canadensis, and each of these endophytes also exhibits chemotype variation [47]. As we develop and refine high throughput methods to explore large host collections more thoroughly, more endophyte variation may be identified within a single host species.

RecentlyBromus laevipes, a bunchgrass native to California, was found to have independently formed symbiotic associations with threeEpichloë species, the nonhybridE. typhina subsp.poae (designatedBromuslaevipes Taxonomic Group 1; BlaTG-1), a hybrid designated BlaTG-2, which appears to be phylogenetically similar toE. cabralii fromPhleum alpinum, and another hybrid designated BlaTG-3 [50]. Endophyte diversity within this host collection was identified by PCR-based genotyping of theEAS,LOL, indole-diterpene/lolitrem (IDT/LTM) and peramine (PER) loci. The BlaTG-3 isolates could be separated into three genotypes (G1, G2, and G3) based on theEAS and mating type complement. Isolates considered BlaTG-3 G1 only containeddmaW, whereas BlaTG-3, G2, and G3, had different mating-type genes, but shared the sameEAS gene complement (EAS^CC) in keeping with its ability to produce CC. In contrast, the completeEAS gene complement required for ERV production (EAS^ERP) was identified within BlaTG-2, yet ergot alkaloids were not detected in plants with BlaTG-2. RT-PCR expression analyses of theEAS genes from BlaTG-2-infected plants indicated they were not expressed. Interestingly, although BlaTG-2 andE. cabralii fromPhleum alpinum are hybrids with the same two ancestralEpichloë species, noEAS genes are present in characterizedE. cabralii strains.

Sleepygrass plants growing in the vicinity of Cloudcroft, New Mexico, U.S.A., are renowned for their narcotic effects on livestock because they can have high levels of the ergot alkaloids ergine and EN [63]. It is now clear that two endophyte species can be symbiotic with sleepygrass,E. funkii [64] and a so-far undescribed hybridEpichloë species designated AroTG-2 [10]. Each of these endophytes has theEAS complement and ability to produce different ergot alkaloids, either CC (EAS^CC) or ergine and EN (EAS^EN), respectively [10]. Phylogenetic analysis ofdmaW groups theE. funkii gene in a clade withdmaW of theE. festucae clade (Figure 8). In contrast, AroTG-2 has admaW sequence more similar to that ofE. inebrians from drunken horse grass than to that of otherEpichloë species (data not shown), which is in keeping with the similarity of the alkaloids produced by these two endophyte species and their strong stupor-inducing effects on grazing livestock [11,63].

5.1. Syntenic Regions of theEAS Loci

Although functionalEAS genes are always clustered, they are not always in a single cluster, arrangements of the genes are not highly conserved, and locations of the clusters can vary, some being subterminal (near chromosome ends), and others being internal and flanked on both sides by long regions rich in housekeeping genes (Figure 10). With the exception of theEpichloë species (discussed below), the degree of divergence inEAS gene arrangements, gene contents, and the pathway end-products generally relates to the degree of divergence between species. Thus, it is unsurprising that theEAS cluster arrangements and gene contents differ greatly in comparison ofN. fumigata to the Clavicipitaceae. A surprisingly consistent feature is the arrangement and close linkage ofeasE andeasF in all exceptE. elymi E56; even theirAr. benhamiae orthologues, respectively designated ARB_4648 and ARB_4647, are adjacent but arranged tail to tail [18]. However,EAS gene arrangements and orientations are very similar in what we will call the “crownEAS clade” (Figure 3):Metarhizium spp. (includingMetarhizium acridum, which is not shown),P. ipomoeae,B. obtecta,At. hypoxylon,Claviceps spp. andE. inebrians. Differences within the crownEAS clade were as follows: (1) an inversion oflpsB-easA segment inC. fusiformis relative to the others; (2) separation of theeasH-lpsA segment from others inB. obtecta, due to an event that (based on remnant genes) occurred in a common ancestor ofB. obtecta andAt. hypoxylon; (3) breakage of the cluster inE. inebrians by a telomere introduced immediately downstream ofcloA; and (4) several gene losses or inactivations that have resulted in changes in ergot alkaloid profiles, as discussed above.

Despite the conserved arrangement ofEAS genes in the aforementioned crownEAS clade, there is very limited synteny of flanking genes (Figure 2), except within theClaviceps subclade and, separately, within theB. obtecta/At. hypoxylon subclade. The only group of orthologous genes that flanks most members of the crownEAS clade is represented inC. purpurea 20.1 byAET79176 (GenBank accession number; labeled with asterisks inFigure 2) and is similarly positioned in all exceptE. inebrians, which has itsAET79176 orthologue at the opposite end of the cluster. With the possible exceptions ofAt. hypoxylon B4728 andE. festucae E2368, theAET79176 orthologue was linked toEAS in every strain that had one. In B4728, it is also possible that the gene is linked to theEAS cluster, but at >75 kb fromeasC (theAET79176 orthologue being near the middle of the 123,479-bp contig00086, which is otherwise very AT-rich and lacks other identifiable genes). So far, there is no putative function or conserved signature domain forAET79176, but it might warrant future investigation for a possible role in the regulation ofEAS genes or in a biosynthetic role yet to be identified.

5.2. Epichloë Species Have MoreEAS Loci Rearrangements

TheEpichloë species show the greatest variation in cluster organization (Figure 10). EvenEAS^ERP clusters, which possess apparently functional forms of 11EAS genes, show extreme rearrangements of gene positions relative to each other and to the telomeres. (The actual linkage and order of the contigs containingEAS genes was established forE. festucae strains E2368 and Fl1 [24] but could not be determined for other genome sequences when not assembled into scaffolds.) Also highly variable amongEpichloë EAS clusters was the organization of their extensive, AT-rich, repetitive sequences, which may directly facilitate cluster instability and rearrangements, and in some cases cause partial or entire deletion of genes giving, for example,EAS^CC clusters (Figure 2 andFigure 9) [24]. In addition, within theEpichloë species, there is a strong tendency for theEAS locus to be retained at the subterminal region, even though the order of genes relative to the telomere (chromosome end) can differ.

Miniature inverted-repeat transposable elements (MITEs) are prevalent inEpichloë species [57], and have been identified in the promoters of someEAS genes with expansion in theEAS^CC clusters associated withdmaW andeasC. The repetitive AT-rich sequences and prevalence of MITEs are also associated with the gene clusters for biosynthesis of other alkaloid classes; namely,IDT/LTM (indole-diterpenes) andLOL (lolines) [24]. The presence of long transposon-derived repeat blocks seems consistent with the subterminal location of telomere-linked clusters likeEAS andIDT/LTM, but the fact that they also feature prominently in theEpichloë LOL clusters, which are typically internal with extensive genic regions flanking both ends, suggests that this is a more general feature of alkaloid clusters [24].

5.3. The Complex History ofEAS Loci

Although not by any means restricted to subtelomeric and subterminal regions, blocks of transposon-derived repeats are features of these genomic regions in fungi [65,66], and such regions are prone to considerable instability and gene duplication events [67,68]. A subterminal location appears to be the ancestral state of theEAS cluster, considering that it is a shared feature ofN. fumigata and the basalEAS clade in Clavicipitaceae; namely, the clade comprised of the majority ofEpichloë EAS clusters (Figure 3). Thus, increased stability, particularly of theEAS core containing early- and mid-pathway genes, seems to be in keeping with the shift from subterminal to internal location in the common ancestor of the crownEAS clade. Nevertheless, the differences in flanking housekeeping genes between theClaviceps subclade and theB. obtecta/At. hypoxylon subclade, plus indications of gene duplication inC. purpurea, indicate additional complexity in the history of theEAS clusters. InC. purpurea, the duplication oflpsA has enabled its neofunctionalization to greatly enhance the diversity of ergopeptine products (Figure 10). Interestingly, 55-73 kb downstream oflpsC in theC. purpurea genome are two additional copies ofdmaW andeasF (Figure 11), suggesting that gene duplication has been a particularly dynamic evolutionary process inC. purpurea in addition to providing an attractive explanation for the fact that most of the known ergopeptin(in)es are reported from this species. Furthermore, thedmaW andeasF duplications are close to a recQ helicase pseudogene, and reflecting their typical locations, recQ helicase genes are also called telomere-linked helicase genes (TLH). (For example, a recQ helicase gene is located near theEAS-linked telomere ofN. fumigata, as shown inFigure 2 andFigure 9). InC. purpurea, the association of a recQ helicase pseudogene with duplicatedEAS genes and in the vicinity of theEAS cluster suggests more recent evolutionary history associated with a chromosome end than implied in our phylogenetic inferences (Figure 3). Thus, repeated shifts between subterminal and internal locations may have characterizedEAS clusters in the Clavicipitaceae, and perhaps especially inC. purpurea as a driver of ergopeptine diversification.

• 1.Billings M. The Crusades: Five Centuries of Holy Wars. Sterling Publishing Company; New York, NY, USA: 1996. [Google Scholar]
• 2.Alm T. The witch trials of Finnmark, northern Norway, during the 17th century: Evidence for ergotism as a contributing factor. Econ. Bot. 2003;57:403–416. doi: 10.1663/0013-0001(2003)057[0403:TWTOFN]2.0.CO;2. [DOI] [Google Scholar]
• 3.Caporael L.R. Ergotism: the Satan loosed in Salem? Science. 1976;192:21–26. doi: 10.1126/science.769159. [DOI] [PubMed] [Google Scholar]
• 4.Woolf A. Witchcraft or mycotoxin? The Salem witch trials. J. Toxicol.: Clin. Toxicol. 2000;38:457–460. doi: 10.1081/CLT-100100958. [DOI] [PubMed] [Google Scholar]
• 5.Zadoks J.C. On the Political Economy of Plant Disease Epidemics: Capita Selecta in Historical Epidemiology. Wageningen Academic Publishers; Wageningen, The Netherlands: 2008. [Google Scholar]
• 6.Tribble M.A., Gregg C.R., Margolis D.M., Amirkhan R., Smith J.W. Fatal ergotism induced by an HIV protease inhibitor. Headache. 2002;42:694–695. doi: 10.1046/j.1526-4610.2002.02163.x. [DOI] [PubMed] [Google Scholar]
• 7.Urga K., Debella A., W’Medihn Y., Bayu A., Zewdie W. Laboratory studies on the outbreak of gangrenous ergotism associated with consumption of contaminated barley in Arsi, Ethiopia. Ethiop. J. Health Dev. 2002;16:317–323. [Google Scholar]
• 8.Scott P. Ergot alkaloids: Extent of human and animal exposure. World Mycotoxin J. 2009;2:141–149. doi: 10.3920/WMJ2008.1109. [DOI] [Google Scholar]
• 9.Lyons P.C., Plattner R.D., Bacon C.W. Occurrence of peptide and clavine ergot alkaloids in tall fescue grass. Science. 1986;232:487–489. doi: 10.1126/science.3008328. [DOI] [PubMed] [Google Scholar]
• 10.Shymanovich T., Saari S., Lovin M., Jarmusch A., Jarmusch S., Musso A., Charlton N., Young C., Cech N., Faeth S. Alkaloid variation among epichloid endophytes of sleepygrass (Achnatherum robustum) and consequences for resistance to insect herbivores. J. Chem. Ecol. 2015;41:93–104. doi: 10.1007/s10886-014-0534-x. [DOI] [PubMed] [Google Scholar]
• 11.Miles C.O., Lane G.A., di Menna M.E., Garthwaite I., Piper E.L., Ball O.J.P., Latch G.C.M., Allen J.M., Hunt M.B., Bush L.P., et al. High levels of ergonovine and lysergic acid amide in toxic Achnatherum inebrians accompany infection by an Acremonium-like endophytic fungus. J. Agric. Food Chem. 1996;44:1285–1290. doi: 10.1021/jf950410k. [DOI] [Google Scholar]
• 12.De Groot A.N., van Dongen P.W., Vree T.B., Hekster Y.A., van Roosmalen J. Ergot alkaloids. Current status and review of clinical pharmacology and therapeutic use compared with other oxytocics in obstetrics and gynaecology. Drugs. 1998;56:523–535. doi: 10.2165/00003495-199856040-00002. [DOI] [PubMed] [Google Scholar]
• 13.Burn D. Neurology (2) Parkinson’s disease: Treatment. Pharm. J. 2000;264:476–479. [Google Scholar]
• 14.Crosignani P.G. Current treatment issues in female hyperprolactinaemia. Eur. J. Obstet. Gynecol. Reprod. Biol. 2006;125:152–164. doi: 10.1016/j.ejogrb.2005.10.005. [DOI] [PubMed] [Google Scholar]
• 15.Micale V., Incognito T., Ignoto A., Rampello L., Sparta M., Drago F. Dopaminergic drugs may counteract behavioral and biochemical changes induced by models of brain injury. Eur. Neuropsychopharmacol. 2006;16:195–203. doi: 10.1016/j.euroneuro.2005.08.003. [DOI] [PubMed] [Google Scholar]
• 16.Hofmann A. LSD: Mein Sorgenkind. Die Entdeckung einer “Wunderdroge”. Deutscher Taschenbuch Verlag; München, Germany: 2006. [Google Scholar]
• 17.Robinson S.L., Panaccione D.G. Diversification of ergot alkaloids in natural and modified fungi. Toxins. 2015;7:201–218. doi: 10.3390/toxins7010201. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 18.Wallwey C., Heddergott C., Xie X., Brakhage A.A., Li S.-M. Genome mining reveals the presence of a conserved biosynthetic gene cluster for the biosynthesis of ergot alkaloid precursors in the fungal family Arthrodermataceae. Microbiology. 2012;158:1634–1644. doi: 10.1099/mic.0.056796-0. [DOI] [PubMed] [Google Scholar]
• 19.Vazquez M.J., Roa A.M., Reyes F., Vega A., Rivera-Sagredo A., Thomas D.R., Diez E., Hueso-Rodriguez J.A. A novel ergot alkaloid as a 5-HT(1A) inhibitor produced by Dicyma sp. J. Med. Chem. 2003;46:5117–5120. doi: 10.1021/jm0341204. [DOI] [PubMed] [Google Scholar]
• 20.Gerhards N., Neubauer L., Tudzynski P., Li S.-M. Biosynthetic pathways of ergot alkaloids. Toxins. 2014;6:3281–3295. doi: 10.3390/toxins6123281. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Panaccione D.G. Origins and significance of ergot alkaloid diversity in fungi. FEMS Microbiol. Lett. 2005;251:9–17. doi: 10.1016/j.femsle.2005.07.039. [DOI] [PubMed] [Google Scholar]
• 22.Panaccione D.G., Schardl C.L., Coyle C.M. Pathways to diverse ergot alkaloid profiles in fungi. In: Romeo J.T., editor. Recent Advances in Phytochemistry. Volume 40. Elsevier; Amsterdam, The Netherlands: 2006. pp. 23–52. [Google Scholar]
• 23.Flieger M., Sedmera P., Vokoun J., R̆ic̄icovā A., R̆ehác̆ek Z. Separation of four isomers of lysergic acid α-hydroxyethylamide by liquid chromatography and their spectroscopic identification. J. Chromatogr. A. 1982;236:441–452. doi: 10.1016/S0021-9673(00)84895-5. [DOI] [Google Scholar]
• 24.Schardl C.L., Young C.A., Hesse U., Amyotte S.G., Andreeva K., Calie P.J., Fleetwood D.J., Haws D.C., Moore N., Oeser B., et al. Plant-symbiotic fungi as chemical engineers: Multi-genome analysis of the Clavicipitaceae reveals dynamics of alkaloid loci. PLoS Genet. 2013;9:e1003323. doi: 10.1371/journal.pgen.1003323. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 25.Coyle C.M., Cheng J.Z., O’Connor S.E., Panaccione D.G. An old yellow enzyme gene controls the branch point between Aspergillus fumigatus and Claviceps purpurea ergot alkaloid pathways. Appl. Environ. Microbiol. 2010;76:3898–3903. doi: 10.1128/AEM.02914-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 26.Ortel I., Keller U. Combinatorial assembly of simple and complex d-lysergic acid alkaloid peptide classes in the ergot fungus Claviceps purpurea. J. Biol. Chem. 2009;284:6650–6660. doi: 10.1074/jbc.M807168200. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 27.Havemann J., Vogel D., Loll B., Keller U. Cyclolization of D-lysergic acid alkaloid peptides. Chem. Biol. 2014;21:146–155. doi: 10.1016/j.chembiol.2013.11.008. [DOI] [PubMed] [Google Scholar]
• 28.Haarmann T., Lorenz N., Tudzynski P. Use of a nonhomologous end joining deficient strain (∆ku70) of the ergot fungus Claviceps purpurea for identification of a nonribosomal peptide synthetase gene involved in ergotamine biosynthesis. Fungal Genet. Biol. 2008;45:35–44. doi: 10.1016/j.fgb.2007.04.008. [DOI] [PubMed] [Google Scholar]
• 29.Uhlig S., Petersen D., Rolèn E., Egge-Jacobsen W., Vrålstad T. Ergosedmine, a new peptide ergot alkaloid (ergopeptine) from the ergot fungus, Claviceps purpurea parasitizing Calamagrostis arundinacea. Phytochem. Lett. 2011;4:79–85. doi: 10.1016/j.phytol.2010.09.004. [DOI] [Google Scholar]
• 30.Schardl C.L., Panaccione D.G., Tudzynski P. Ergot alkaloids—Biology and molecular biology. Alkaloids. 2006;63:45–86. doi: 10.1016/s1099-4831(06)63002-2. [DOI] [PubMed] [Google Scholar]
• 31.Matuschek M., Wallwey C., Wollinsky B., Xie X., Li S.-M. In vitro conversion of chanoclavine-I aldehyde to the stereoisomers festuclavine and pyroclavine controlled by the second reduction step. RSC Adv. 2012;2:3662. doi: 10.1039/c2ra20104f. [DOI] [Google Scholar]
• 32.Unsöld I.A., Li S.M. Overproduction, purification and characterization of FgaPT2, a dimethylallyltryptophan synthase from Aspergillus fumigatus. Microbiol. SGM. 2005;151:1499–1505. doi: 10.1099/mic.0.27759-0. [DOI] [PubMed] [Google Scholar]
• 33.Coyle C.M., Panaccione D.G. An ergot alkaloid biosynthesis gene and clustered hypothetical genes from Aspergillus fumigatus. Appl. Environ. Microbiol. 2005;71:3112–3118. doi: 10.1128/AEM.71.6.3112-3118.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 34.Chen L., Li X.Z., Li C.J., Swoboda G.A., Young C.A., Sugawara K., Leuchtmann A., Schardl C.L. Two distinct Epichloë species symbiotic with Achnatherum inebrians, drunken horse grass. Mycologia. 2015 doi: 10.3852/15-019. in press. [DOI] [PubMed] [Google Scholar]
• 35.Blin K., Medema M.H., Kazempour D., Fischbach M.A., Breitling R., Takano E., Weber T. antiSMASH 2.0—A versatile platform for genome mining of secondary metabolite producers. Nucl. Acids Res. 2013;41:W204–W212. doi: 10.1093/nar/gkt449. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 36.Khaldi N., Seifuddin F.T., Turner G., Haft D., Nierman W.C., Wolfe K.H., Fedorova N.D. SMURF: Genomic mapping of fungal secondary metabolite clusters. Fungal Genet. Biol. 2010;47:736–741. doi: 10.1016/j.fgb.2010.06.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 37.Keller N.P., Turner G., Bennett J.W. Fungal secondary metabolism—From biochemistry to genomics. Nat. Rev. Microbiol. 2005;3:937–947. doi: 10.1038/nrmicro1286. [DOI] [PubMed] [Google Scholar]
• 38.Schardl C.L., Young C.A., Moore N., Krom N., Dupont P.-Y., Pan J., Florea S., Webb J.S., Jaromczyk J., Jaromczyk J.W., et al. Genomes of plant-associated Clavicipitaceae. Adv. Bot. Res. 2014;70:291–327. [Google Scholar]
• 39.Gao Q., Jin K., Ying S.-H., Zhang Y., Xiao G., Shang Y., Duan Z., Hu X., Xie X.-Q., Zhou G., et al. Genome sequencing and comparative transcriptomics of the model entomopathogenic fungi Metarhizium anisopliae and M. acridum. PLoS Genet. 2011;7:e1001264. doi: 10.1371/journal.pgen.1001264. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 40.Dereeper A., Guignon V., Blanc G., Audic S., Buffet S., Chevenet F., Dufayard J.-F., Guindon S., Lefort V., Lescot M., et al. Phylogeny.fr: Robust phylogenetic analysis for the non-specialist. Nucl. Acids Res. 2008;36:W465–W469. doi: 10.1093/nar/gkn180. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 41.Edgar R.C. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucl. Acids Res. 2004;32:1792–1797. doi: 10.1093/nar/gkh340. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 42.Anisimova M., Gascuel O. Approximate likelihood-ratio test for branches: a fast, accurate, and powerful alternative. Syst. Biol. 2006;55:539–552. doi: 10.1080/10635150600755453. [DOI] [PubMed] [Google Scholar]
• 43.Schardl C.L., Young C.A., Pan J., Florea S., Takach J.E., Panaccione D.G., Farman M.L., Webb J.S., Jaromczyk J., Charlton N.D., et al. Currencies of mutualisms: Sources of alkaloid genes in vertically transmitted epichloae. Toxins. 2013;5:1064–1088. doi: 10.3390/toxins5061064. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 44.Pan J., Bhardwaj M., Faulkner J.R., Nagabhyru P., Charlton N.D., Higashi R.M., Miller A.-F., Young C.A., Grossman R.B., Schardl C.L. Ether bridge formation in loline alkaloid biosynthesis. Phytochemistry. 2014;98:60–68. doi: 10.1016/j.phytochem.2013.11.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 45.Pan J., Bhardwaj M., Nagabhyru P., Grossman R.B., Schardl C.L. Enzymes from fungal and plant origin required for chemical diversification of insecticidal loline alkaloids in grass-Epichloë symbiota. PLoS One. 2014;9:e115590. doi: 10.1371/journal.pone.0115590. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 46.Schardl C.L., Farman M.L., Jaromczyk J.W., Young C.A., Webb J.S., Jaromczyk J., Moore N., Bec S., Calie P.J., Charlton N., et al. Genome sequences of 56 plant-associated Clavicipitaceae. 2015 unpublished. [Google Scholar]
• 47.Charlton N.D., Craven K.D., Mittal S., Hopkins A.A., Young C.A. Epichloë canadensis, a new interspecific epichloid hybrid symbiotic with Canada wildrye (Elymus canadensis) Mycologia. 2012;104:1187–1199. doi: 10.3852/11-403. [DOI] [PubMed] [Google Scholar]
• 48.Schardl C.L., Young C.A., Faulkner J.R., Florea S., Pan J. Chemotypic diversity of epichloae, fungal symbionts of grasses. Fungal Ecol. 2012;5:331–344. doi: 10.1016/j.funeco.2011.04.005. [DOI] [Google Scholar]
• 49.Takach J.E., Mittal S., Swoboda G.A., Bright S.K., Trammell M.A., Hopkins A.A., Young C.A. Genotypic and chemotypic diversity of Neotyphodium endophytes in tall fescue from Greece. Appl. Environ. Microbiol. 2012;78:5501–5510. doi: 10.1128/AEM.01084-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 50.Charlton N.D., Craven K.D., Afkhami M.E., Hall B.A., Ghimire S.R., Young C.A. Interspecific hybridization and bioactive alkaloid variation increases diversity in endophytic Epichloë species of Bromus laevipes. FEMS Microbiol. Ecol. 2014;90:276–289. doi: 10.1111/1574-6941.12393. [DOI] [PubMed] [Google Scholar]
• 51.Takach J.E., Young C.A. Alkaloid genotype diversity of tall fescue endophytes. Crop. Sci. 2014;54:667–678. doi: 10.2135/cropsci2013.06.0423. [DOI] [Google Scholar]
• 52.Young C.A., Charlton N.D., Takach J.E., Swoboda G.A., Trammell M.A., Huhman D.V., Hopkins A.A. Characterization of Epichloë coenophiala within the US: Are all tall fescue endophytes created equal? Front. Chem. 2014;2:95. doi: 10.3389/fchem.2014.00095. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 53.Bacon C.W. Procedure for isolating the endophyte from tall fescue and screening isolates for ergot alkaloids. Appl. Env. Microbiol. 1988;54:2615–2618. doi: 10.1128/aem.54.11.2615-2618.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 54.Blankenship J.D., Spiering M.J., Wilkinson H.H., Fannin F.F., Bush L.P., Schardl C.L. Production of loline alkaloids by the grass endophyte, Neotyphodium uncinatum, in defined media. Phytochemistry. 2001;58:395–401. doi: 10.1016/S0031-9422(01)00272-2. [DOI] [PubMed] [Google Scholar]
• 55.Chujo T., Scott B. Histone H3K9 and H3K27 methylation regulates fungal alkaloid biosynthesis in a fungal endophyte–plant symbiosis. Mol. Microbiol. 2014;92:413–434. doi: 10.1111/mmi.12567. [DOI] [PubMed] [Google Scholar]
• 56.Fleetwood D.J., Scott B., Lane G.A., Tanaka A., Johnson R.D. A complex ergovaline gene cluster in epichloë endophytes of grasses. Appl. Environ. Microbiol. 2007;73:2571–2579. doi: 10.1128/AEM.00257-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 57.Fleetwood D.J., Khan A.K., Johnson R.D., Young C.A., Mittal S., Wrenn R.E., Hesse U., Foster S.J., Schardl C.L., Scott B. Abundant degenerate miniature inverted-repeat transposable elements in genomes of epichloid fungal endophytes of grasses. Genome Biol. Evol. 2011;3:1253–1264. doi: 10.1093/gbe/evr098. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 58.Panaccione D.G., Johnson R.D., Wang J.H., Young C.A., Damrongkool P., Scott B., Schardl C.L. Elimination of ergovaline from a grass-Neotyphodium endophyte symbiosis by genetic modification of the endophyte. Proc. Natl. Acad. Sci. USA. 2001;98:12820–12825. doi: 10.1073/pnas.221198698. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 59.Ghimire S.R., Rudgers J.A., Charlton N.D., Young C., Craven K.D. Prevalence of an intraspecific Neotyphodium hybrid in natural populations of stout wood reed (Cinna arundinacea L.) from eastern North America. Mycologia. 2011;103:75–84. doi: 10.3852/10-154. [DOI] [PubMed] [Google Scholar]
• 60.Christensen M.J., Leuchtmann A., Rowan D.D., Tapper B.A. Taxonomy of Acremonium endophytes of tall fescue (Festuca arundinacea), meadow fescue (F. pratensis), and perennial rye-grass (Lolium perenne) Mycol. Res. 1993;97:1083–1092. doi: 10.1016/S0953-7562(09)80509-1. [DOI] [Google Scholar]
• 61.Leuchtmann A., Schmidt D., Bush L.P. Different levels of protective alkaloids in grasses with stroma-forming and seed-transmitted Epichloë/Neotyphodium endophytes. J. Chem. Ecol. 2000;26:1025–1036. doi: 10.1023/A:1005489032025. [DOI] [Google Scholar]
• 62.Schardl C.L., Craven K.D., Speakman S., Stromberg A., Lindstrom A., Yoshida R. A novel test for host-symbiont codivergence indicates ancient origin of fungal endophytes in grasses. Syst. Biol. 2008;57:483–498. doi: 10.1080/10635150802172184. [DOI] [PubMed] [Google Scholar]
• 63.Petroski R., Powell R.G., Clay K. Alkaloids of Stipa robusta (sleepygrass) infected with an Acremonium endophyte. Nat. Toxins. 1992;1:84–88. doi: 10.1002/nt.2620010205. [DOI] [PubMed] [Google Scholar]
• 64.Moon C.D., Guillaumin J.-J., Ravel C., Li C., Craven K.D., Schardl C.L. New Neotyphodium endophyte species from the grass tribes Stipeae and Meliceae. Mycologia. 2007;99:895–905. doi: 10.3852/mycologia.99.6.895. [DOI] [PubMed] [Google Scholar]
• 65.Farman M.L. Telomeres in the rice blast fungus Magnaporthe oryzae: The world of the end as we know it. FEMS Microbiol. Lett. 2007;273:125–132. doi: 10.1111/j.1574-6968.2007.00812.x. [DOI] [PubMed] [Google Scholar]
• 66.Starnes J.H., Thornbury D.W., Novikova O.S., Rehmeyer C.J., Farman M.L. Telomere-targeted retrotransposons in the rice blast fungus Magnaporthe oryzae: Agents of telomere instability. Genetics. 2012;191:389–406. doi: 10.1534/genetics.111.137950. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 67.Wu C., Kim Y.-S., Smith K.M., Li W., Hood H.M., Staben C., Selker E.U., Sachs M.S., Farman M.L. Characterization of chromosome ends in the filamentous fungus Neurospora crassa. Genetics. 2009;181:1129–1145. doi: 10.1534/genetics.107.084392. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 68.Stajich J.E., Wilke S.K., Ahren D., Au C.H., Birren B.W., Borodovsky M., Burns C., Canback B., Casselton L.A., Cheng C.K., et al. Insights into evolution of multicellular fungi from the assembled chromosomes of the mushroom Coprinopsis cinerea (Coprinus cinereus) Proc. Natl. Acad. Sci. USA. 2010;107:11889–11894. doi: 10.1073/pnas.1003391107. [DOI] [PMC free article] [PubMed] [Google Scholar]