Two Satellite Gene Clusters Enhance Ergot Alkaloid Biosynthesis Capacity of Aspergillus leporis

ABSTRACT

Ergot alkaloids are fungal specialized metabolites that are important in agriculture and serve as sources of several pharmaceuticals. Aspergillus leporis is a soil saprotroph that possesses two ergot alkaloid biosynthetic gene clusters encoding lysergic acid amide production. We identified two additional, partial biosynthetic gene clusters within the A. leporis genome containing some of the ergot alkaloid synthesis (eas) genes required to make two groups of clavine ergot alkaloids, fumigaclavines and rugulovasines. Clavines possess unique biological properties compared to lysergic acid derivatives. Bioinformatic analyses indicated the fumigaclavine cluster contained functional copies of easA, easG, easD, easM, and easN. Genes resembling easQ and easH, which are required for rugulovasine production, were identified in a separate gene cluster. The pathways encoded by these partial, or satellite, clusters would require intermediates from the previously described lysergic acid amide pathway to synthesize a product. Chemical analyses of A. leporis cultures revealed the presence of fumigaclavine A. However, rugulovasine was only detected in a single sample, prompting a heterologous expression approach to confirm functionality of easQ and easH. An easA knockout strain of Metarhizium brunneum, which accumulates the rugulovasine precursor chanoclavine-I aldehyde, was chosen as expression host. Strains of M. brunneum expressing easQ and easH from A. leporis accumulated rugulovasine as demonstrated through mass spectrometry analysis. These data indicate that A. leporis is exceptional among fungi in having the capacity to synthesize products from three branches of the ergot alkaloid pathway and for utilizing an unusual satellite cluster approach to achieve that outcome.

IMPORTANCE Ergot alkaloids are chemicals produced by several species of fungi and are notable for their impacts on agriculture and medicine. The ability to make ergot alkaloids is typically encoded by a clustered set of genes that are physically adjacent on a chromosome. Different ergot alkaloid classes are formed via branching of a complex pathway that begins with a core set of the same five genes. Most ergot alkaloid-producing fungi have a single cluster of genes that is complete, or self-sufficient, and produce ergot alkaloids from one or occasionally two branches from that single cluster. Our data show that Aspergillus leporis is exceptional in having the genetic capacity to make products from three pathway branches. Moreover, it uses a satellite cluster approach, in which gene products of partial clusters rely on supplementation with a chemical intermediate produced via another gene cluster, to diversify its biosynthetic potential without duplicating all the steps.

INTRODUCTION

Several different fungi in the Eurotiales and Hypocreales produce specialized metabolites known as ergot alkaloids (1–4) (Fig. 1). These fungi occupy a wide variety of niches, including soil saprophytes, entomopathogens, and plant endophytes and parasites (5–9). Ergot alkaloids have historically been associated with negative impacts on humans and other animals due to their potent toxicity and ability to cause convulsive and gangrenous symptoms when ingested (10). Exposure to some ergot alkaloids can cause psychotropic and narcotic effects, a property that has been used by both Native American cultures and, in more recent history, recreational drug users (11, 12). The ability of these compounds to exhibit a wide array of pharmacological activities is due to their similarity to neurotransmitters and high binding affinity for several neurological receptors, including dopamine, adreno, and 5-hydroxytryptamine receptors (13, 14). The activities of clavine-derived ergot alkaloids (Fig. 1) are less understood compared to those of ergopeptines and lysergic acid amides, though some have been shown to possess anticancer, antimicrobial, anti-inflammatory, and insecticidal activities (14–16).

FIG 1.

Simplified pathways to different ergot alkaloids. All pathways begin with the same core set of genes (dmaW, easF, easE, easC, easD) and branch depending on the combination of other eas genes present. The pathway branches involving the isomerase easA allele can lead to production of lysergic acid amides, as are found in M. brunneum (8) (labeled in red) and ergopeptines as found in C. purpurea and several Epichloë species (2, 4, 10) (labeled in orange). The pathway branches involving the reductase easA allele can lead to production of fumigaclavines, as are found in A. fumigatus (5) (labeled in blue) and cycloclavine, which is found in A. japonicus (48, 50) (labeled in purple). The pathway leading to production of rugulovasines A/B in P. biforme involves keto-enol tautomerization of chanoclavine-I aldehyde and the activity of the protein products of easQ and easH (56) (labeled in green). DMAPP, dimethylallylpyrophosphate; red, reductase; iso, isomerase. Lysergyl peptide synthetases 1 (LPS1), 2 (LPS2), and 3 (LPS3) are encoded by lpsA. lpsB, and lpsC, respectively.

Ergot alkaloids are a complex family of chemicals originating from a shared pathway (Fig. 1) that branches at different points depending on the combination of accessory ergot alkaloid synthesis (eas) genes present in a particular organism’s biosynthetic gene cluster (9, 17–21). Synthesis of chanoclavine-I aldehyde, the common intermediate shared among several major branches of the ergot alkaloid pathway, requires a set of five shared eas genes (dmaW, easF, easE, easC, and easD). The products of dmaW and easF prenylate tryptophan at position 4 (21) and methylate the amino group (22), respectively, before decarboxylative closure of the third ring is carried out by the products of easE and easC, together generating the simplest clavine–chanoclavine-I (23–25). The product of easD then oxidizes the primary alcohol of chanoclavine-I to an aldehyde (26), generating chanoclavine-I aldehyde, which serves as a major branchpoint for several other ergot alkaloid pathways. Branches to most major classes of ergot alkaloids depend on whether the reductase or isomerase allele of easA is present in an organism (27–30). The isomerase and reductase products of easA catalyze closure of the ergoline D ring before reduction via the product of easG occurs, generating either festuclavine or agroclavine, respectively (30, 31). Production of lysergic acid amides, such as lysergic acid α-hydroxyethylamide (LAH), is then dependent upon the presence of other tailoring enzymes in an organism’s eas cluster (32–39). Ergopeptines consist of a tripeptide chain linked to d-lysergic acid and are synthesized by the combination of two nonribosomal peptide synthetases (LPS1 and LPS2) together with the Fe²⁺/2-ketoglutarate-dependent dioxygenase encoded by easH found in Claviceps purpurea (40). EasH is responsible for hydroxylating the α-carbon of the amino acid adjacent to d-lysergic acid, driving cyclolization of the α-carbon with the terminal lactam carbonyl group generated by LPS1/LPS2 (40).

The clavine branches of the ergot alkaloid pathway include those ending in production of fumigaclavines, cycloclavine, and rugulovasines. Production of fumigaclavines A, B, and C in Aspergillus fumigatus (synonym Neosartorya fumigata) involves the activities of EasM, EasN, and EasL acting on festuclavine in the absence of a CloA allele (41–43). Fumigaclavine C has anti-inflammatory, antitumor, and anti-atherogenic properties, highlighting the potential for clavine-derived medicines (15, 44–46). Cycloclavine, originally isolated from seeds of Ipomoea hildebrandtii (47) and later the fungus Aspergillus japonicus (48), possesses insecticidal and antiparasite activities (49). Cycloclavine production in A. japonicus is dependent upon a version of the dioxygenase EasH that, together with the reductase version of EasA and EasG, catalyzes the formation of a cyclopropyl ring moiety attached to the clavine ring system (50, 51). The A. japonicus version of EasH has also been shown to possess the ability to catalyze the asymmetric hydroxylation of other clavines, such as festuclavine and elymoclavine, a result that displays the promiscuity of this enzyme (52). Rugulovasines A and B, originally isolated from several Penicillium species, including P. biforme (53, 54), are a stereoisomeric pair of clavines with toxic and hypotensive properties (55). Rugulovasine biosynthesis branches from the ergot alkaloid pathway at chanoclavine-I aldehyde and relies upon easQ, encoding an aldehyde dehydrogenase, and a version of EasH hypothesized to hydroxylate the same carbon oxidized by EasH from A. japonicus during cycloclavine synthesis (56).

Jones et al. (57) identified two eas clusters associated with accumulation of the lysergic acid amide LAH in Aspergillus leporis, a member of Aspergillus section Flavi originally isolated from jackrabbit dung (58). Aspergillus leporis also is capable of infecting the model lepidopteran insect Galleria mellonella and produces relatively high concentrations of LAH during infection (59). Phylogenetic studies indicated the eight genes required for synthesis of lysergic acid in A. leporis were orthologous to those of the ergot alkaloid producers of the fungal family Clavicipitaceae, but the genes required to put lysergic acid into LAH evolved independently in A. leporis and two other species of Aspergillus compared to those of the Clavicipitaceae (57). Further analysis of the genome of A. leporis has led us to two additional clusters of genes: one containing homologs of some of the genes required for synthesis of fumigaclavines in A. fumigatus, and the other containing two genes associated with rugulovasine biosynthesis in P. biforme. The objectives of this present study were to characterize these gene clusters and more fully investigate the ergot alkaloid biosynthesis capacity of A. leporis.

RESULTS

Investigation of two partial, clavine-type eas clusters in Aspergillus leporis.

Through investigation into alleles of easA, easD, and easG that were observed in A. leporis but not associated with the two lysergic acid amide gene clusters described by Jones et al. (57), we identified a third eas cluster containing these three alleles and several other genes associated with fumigaclavine production (Fig. 2A). The A. leporis contig (SWBU01000167.1) containing homologs of easA, easD, and easG was also found to contain apparently intact copies of easM and easN along with a copy of easL that appeared to be a pseudogene due to a 1,990-bp insertion (nucleotides 93,081 to 95,071 of this contig), two deletions, and multiple nonsense mutations (Fig. S1). The cluster also contains a copy of a gene called easK which is found in the eas cluster of A. fumigatus but not in the fumigaclavine-associated eas cluster of P. commune (1). No role for the product of easK has been found or proposed, and the copy in the A. leporis cluster appears to be a pseudogene due to the presence of several nonsense mutations. Accumulation of fumigaclavine A should occur if the copies of easA, easD, easG, easM, and easN present in this cluster are functional and active at the same time as products of dmaW, easF, easE, and easC from one of the lysergic acid amide clusters of A. leporis.

FIG 2.

Clusters containing genes potentially associated with fumigaclavine (A) or rugulovasine (B) synthesis in A. leporis. Black arrows indicate functional gene copies and red arrows with Ψ in the gene name indicate hypothesized pseudogenes. The role for easK has not been characterized, but it has been identified in fumigaclavine clusters of other fungi (4, 5). Scale bar included for approximate sizes of genes and intergenic regions.

The area upstream of the fumigaclavine cluster was analyzed in three increments of 5,000 bp at a time by BLASTx before it was concluded that no other eas genes were present nearby. The remaining 3,251 bp of the contig following the easA termination codon were also analyzed by BLASTx and did not possess additional eas genes. The area flanking this cluster was found to contain elements related to transposon activity with putative products resembling a transposase (XP_013329072.1), a pogo transposable element (RJE23005.1), and a reverse transcriptase (XP_001395701.2) having been identified nearby. Apart from the copies previously characterized in the lysergic acid amide synthesis clusters of A. leporis (57), no additional copies of dwaW, easF, easE, or easC (which would encode the steps preceding those predicted in the fumigaclavine cluster) meeting minimum criteria of 30% amino acid sequence identity and 70% query coverage could be found within the genome. This observation indicates the fumigaclavine cluster is isolated from, and devoid of, other eas genes.

The discovery of a partial fumigaclavine cluster in A. leporis prompted additional genome mining for other potential eas genes. Through this approach, a copy of easH encoding a product with 61% identity to EasH encoded by the P. camemberti allele of that gene was identified (Fig. 2B). Another gene, encoding an aldehyde dehydrogenase, was found ~2 kb away and encodes a product with 75% identity with the product of the P. camemberti allele of easQ. The genomic context of the A. leporis easQ and easH homologs was investigated by searching the areas upstream and downstream to test for other potential eas genes. Increments of 5,000 bp at a time were analyzed by BLASTx, with the process being repeated a total of three times in both directions. Putative genes encoding a metalloprotease (KAB8075085.1), major facilitator transporter (KAE8380494.1), NADH-ubiquinone oxidoreductase (XP_033419163.1), RNA-directed DNA polymerase (GFF18983.1), kinesin (KAB8075089.1), and cytochrome P450 oxioreductase (RAQ58036.1) were identified nearby. However, no other eas genes were detected, indicating that this partial cluster of easQ and easH is isolated from other genes needed for ergot alkaloid production.

Fumigaclavine A accumulation in A. leporis NRRL 3216.

Solid (combination of hyphae, conidiophores, and conidia) and liquid (culture fluids) phases of independent cultures of A. leporis NRRL 3216 grown in sucrose-yeast extract medium for 14 days accumulated fumigaclavine A (Fig. 3). Extracts from both phases were analyzed by high-performance liquid chromatography (HPLC) with fluorescence detection and compared to a fumigaclavine A standard of known concentration. A peak eluting at 43.8 min matched the peak representing the fumigaclavine A standard in retention time and fluorescence properties. Percent secretion of fumigaclavine A was calculated from seven technical replicates analyzed by HPLC and found to have a mean of 35 (range 9 to 57, median 35), which is approximately 3-fold higher than the secretion of fumigaclavine A observed in A. fumigatus (60). Extracts from the solid phase of the cultures were analyzed by liquid chromatography-high resolution mass spectrometry (LC-HRMS) and compared to a fumigaclavine A standard. An analyte of m/z = 299.1747 with a single peak occurring at 5.29 min was detected (Fig. 4A), which lines up with the peak corresponding to the fumigaclavine A standard occurring at 5.33 min (Fig. 4B). Fragmentation analyses of the m/z 299.1747 analytes revealed matching spectra (Fig. 4C to D).

FIG 3.

HPLC chromatogram of wild-type A. leporis solid and liquid extracts compared to fumigaclavine A standard. Hyphal and medium extracts were from the same culture, which is representative of the multiple cultures analyzed for Table 1 and for the secretion study summarized in the text. Fluorescence was detected at 372 nm after excitation at 272 nm. Peaks corresponding to characterized ergot alkaloids are indicated.

FIG 4.

LC-HRMS of A. leporis hyphal extract compared to fumigaclavine A standard. Extracted ion chromatograms of (A) wild-type A. leporis NRRL 3216 hyphal extracts and (B) 400 ng/μL fumigaclavine A standard analyzed at m/z 299.1747 ± 5 ppm. (C and D) High resolution MS spectra resulting from fragmentation of parent ions shown in panels A and B. Instrument resolution accurate to < 5 ppm.

A strain of A. leporis that had been previously modified to disrupt the sole functional copy of easD present in the LAH-associated eas clusters (59) was grown and analyzed alongside the wild-type to test whether the availability of increased chanoclavine-I would result in increased accumulation of fumigaclavine A. The easD knockout strain of A. leporis accumulated significantly more chanoclavine-I (P = 0.0073) and fumigaclavine A (P = 0.0014) than did wild-type A. leporis (Table 1). The rate of conversion of chanoclavine-I to fumigaclavine A did not differ significantly between the two strains (P = 0.659), suggesting the observed increase in fumigaclavine A resulted from the availability of additional chanoclavine-I in the easD knockout. This observation supports the hypothesis that the fumigaclavine cluster is fed chanoclavine-I intermediate derived from the lysergic acid amide cluster and thus can be considered a satellite cluster relative to the lysergic acid amide cluster. The relative timing of accumulation of LAH and fumigaclavine A is relevant in this context. LAH accumulation peaks sharply at 6 days, then declines markedly by day 12 under the conditions used (57) (Fig. S2). Fumigaclavine A accumulation increased gradually over time, reaching a maximum at 15 days postinoculation before declining gradually (Fig. S2). Chanoclavine-I, which is a required intermediate in both LAH and fumigaclavine pathways, reached maximum concentration in the 6- to 9-day range and then declined only slightly over the remainder of the 3-week culture period. Any chanoclavine-I not converted to LAH during what is typically the optimal period of LAH accumulation in the wild-type was then likely available to the enzymes of the fumigaclavine pathway during peak production of that metabolite. The difference in timing of the two pathways also provides a plausible explanation for why the copy of EasD encoded in the fumigaclavine pathway was not observed to complement the easD mutation in the LAH pathway (59).

TABLE 1.

Accumulation (mean ± standard error; n = 7) of chanoclavine-I and fumigaclavine A in cultures of A. leporis grown in sucrose-yeast extract medium for 14 days at 30°C

Strain	Chanoclavine-I (nmol/g fungus)	Fumigaclavine A (nmol/g fungus)	Percent conversiona
A. leporis NRRL 3216	92 ± 7 Ab	84 ± 21 A	43 ± 7 A
A. leporis easD ko	454 ± 91 B	274 ± 41 B	40 ± 3 A

^aPercent of chanoclavine-I converted to fumigaclavine A.

^bMeans within a column that differ significantly (P < 0.05) are indicated with different letters.

Rare rugulovasine accumulation in A. leporis NRRL 3216 under conditions that favor ergot alkaloid accumulation.

Rugulovasine A spontaneously stereoisomerizes over time to rugulovasine B; the two stereoisomers then readily interconvert (61). The LC-HRMS approach we used does not allow distinction of the two stereoisomers, so we will refer to the observed analyte demonstrating properties of rugulovasines A/B as simply rugulovasine from here on. Cultures of P. biforme NRRL 885 were grown on malt extract agar for 7 days to provide a reference analyte for rugulovasine. Samples were analyzed by LC-HRMS and found to accumulate the m/z 269.1278 ion typical of rugulovasine with a single peak occurring at 5.91 min (Fig. 5A). Cultures of A. leporis NRRL 3216 were grown under similar conditions, and among a total of 26 samples ranging in age from 3 to 18 days (including 14 in the 12-day to 18-day range), only one culture at 14 days old was found to accumulate an analyte corresponding to rugulovasine (m/z 269.1278, eluting at 5.88 min) (Fig. 5B). Fragmentation analyses of the parent ions from P. biforme and A. leporis extracts revealed matching spectra, indicating A. leporis is capable of rugulovasine production (Fig. S3). Larvae of G. mellonella were inoculated with A. leporis to test if insect infection would more consistently elicit rugulovasine production. Extracts from 8-day postinfection larvae were analyzed by LC-HRMS and were found to lack rugulovasine. Extracted ion chromatograms for m/z 269.1278 from A. leporis cultures as well as larvae infected with A. leporis displayed peaks eluting at approximately 4.99 min and 5.35 min (Fig. 5C to D). The identities of the m/z 269.1278 peaks occurring in panels B to D of Fig. 5 were confirmed to be lysergic acid and its stereoisomer by comparison to the lysergic acid previously documented to accumulate in G. mellonella larvae infected with a lpsB knockout of M. brunneum (Fig. 5E) (37). The lysergic acid stereoisomers appeared in this extracted ion chromatogram because lysergic acid has the same molecular formula as rugulovasine. Fragmentation analyses of the parent ions corresponding to these earlier-eluting m/z 269.1278 ion peaks further confirmed their identity as diastereoisomeric lysergic acid and isolysergic acid (Fig. S4).

FIG 5.

LC-HRMS analyses of P. biforme and single rugulovasine-producing A. leporis culture compared to typical A. leporis culture, A. leporis-infected G. mellonella, and G. mellonella infected with lpsB knockout of M. brunneum. Extracted ion chromatograms of (A) P. biforme NRRL 885 malt extract culture, (B) single rugulovasine-producing A. leporis NRRL 3216 malt extract culture, (C) typical rugulovasine-nonproducing A. leporis NRRL 3216 malt extract culture, (D) A. leporis-infected G. mellonella, and (E) lpsB knockout M. brunneum-infected G. mellonella analyzed at m/z 269.1748 ± 5 ppm. Instrument resolution accurate to < 5 ppm.

Phylogenetics of A. leporis easQ and easH associate them with ergot alkaloid synthesis versions of these genes.

Given that rugulovasines were only detected in single A. leporis sample, we examined the versions of EasQ and EasH encoded in the A. leporis gene cluster more closely. We conducted phylogenetic analysis of the predicted EasQ along with homologs from P. biforme and P. camemberti eas clusters as well as non-eas cluster homologs found in A. leporis, P. biforme, P. camemberti, A. fumigatus, and M. brunneum. Homologs were defined as having >30% amino acid sequence identity and 70% query coverage with the version of EasQ encoded in that organism’s eas cluster. The results showed that the version of EasQ encoded from the allele in the easQ/easH satellite cluster of A. leporis formed a clade with those encoded within the P. biforme/P. camemberti rugulovasine eas gene clusters (Fig. 6), though bootstrap support for this clade was marginal. Two non-eas cluster homologs of EasQ encoded elsewhere in the A. leporis genome met the criteria for inclusion in this data set and were found to form distinct clades with their orthologous, non-eas cluster counterparts in the other species.

FIG 6.

Maximum likelihood tree for amino acid sequences of EasQ encoded in eas clusters and homologs from rugulovasine producers and species used in this study. Homologs had at least 30% amino acid sequence identity over 70% query coverage and are identified by species and NCBI accession number. The trees presented have the greatest log likelihood of 1,000 bootstrapped trees calculated with the Le and Gascuel (LG) model (79) using a discrete Gamma distribution (+G) in MEGA X (78); bootstrap values of 50% or higher are indicated at the corresponding nodes. Tree is unrooted. The bar represents changes per site.

Phylogenetic analysis of the predicted A. leporis EasH protein sequence compared to homologs from rugulovasine, cycloclavine, and ergopeptine producers showed that the version of EasH encoded in the A. leporis satellite cluster forms a clade with EasH encoded in eas clusters from rugulovasine producers (Fig. 7). Additionally, these phylogenetic data show strong bootstrap support for the EasH versions required for rugulovasine, cycloclavine, and ergopeptine production having diverged from a recent common ancestor.

FIG 7.

Maximum likelihood tree for amino acid sequences of EasH encoded in eas clusters and homologs from species producing differing branches of ergot alkaloids. Homologs had at least 30% amino acid sequence identity over 70% query coverage and are identified by species and NCBI accession number. The tree presented has the greatest log likelihood of 1,000 bootstrapped trees calculated with the Le and Gascuel (LG) model (79) using a discrete Gamma distribution (+G) in MEGA X (78); bootstrap values of 50% or higher are indicated at the corresponding nodes. Clades possessing ergot alkaloid pathways for rugulovasine, cycloclavine, and ergopeptine production are highlighted in green, purple, or orange, respectively. Tree is unrooted. The bar represents changes per site.

Rugulovasine production by heterologous expression of easQ and easH from A. leporis in M. brunneum.

To more thoroughly test the ability of easQ and easH from A. leporis to catalyze formation of rugulovasine, both genes were placed under the control of the bi-directional easG/dmaW promoter of M. brunneum and transformed into an easA knockout strain of M. brunneum (37). Fusion PCR was used to generate the expression construct which was then introduced into M. brunneum as part of a construct based on pBChygro, a selectable marker conferring hygromycin resistance (62). Presence of the fusion construct and hygromycin resistance gene was confirmed through PCR (Fig. S5). The strain of M. brunneum expressing A. leporis easQ and easH genes was analyzed by LC-HRMS and compared to the easA knockout background strain and wild-type P. biforme, a natural rugulovasine producer (54). There was no evidence of rugulovasine in the background, easA knockout strain of M. brunneum (Fig. 8A). The M. brunneum strain expressing both A. leporis genes was found to produce an analyte of m/z = 269.1278 with a single peak that aligns with the peak corresponding to rugulovasine in P. biforme (Fig. 8). Fragmentation analyses of the m/z 269.1278 analyte from the M. brunneum mutant revealed a spectrum consistent with that of rugulovasine from P. biforme (Fig. 8D to E).

FIG 8.

LC-HRMS analyses of M. brunneum mutant expressing easQ and easH from A. leporis compared to the background strain of M. brunneum (easA knockout) and P. biforme NRRL 885. Extracted ion chromatograms of (A) easA knockout strain of M. brunneum, (B) M. brunneum mutant expressing easQ and easH from A. leporis, and (C) P. biforme analyzed at m/z 269.1278 ± 5 ppm. (D-E) High resolution MS spectra resulting from fragmentation of parent ions shown in panels B and C. Instrument resolution accurate to < 5 ppm.

Lack of cycloclavine production upon expression of easH from A. leporis in a mutant strain of A. fumigatus.

Because versions of EasH involved in cycloclavine synthesis (51, 52) oxidize the same carbon likely to be oxidized by the version of EasH involved in rugulovasine synthesis (56), we tested the ability of EasH from A. leporis, in conjunction with the products of the required early pathway genes from A. fumigatus, to catalyze formation of cycloclavine. In this study, easH from A. leporis was placed under the control of the A. fumigatus easA promoter and expressed in an easM knockout strain of A. fumigatus (41). Fusion PCR was used to generate this construct which was then introduced into A. fumigatus as part of a construct prepared in pBCphleo (62). Presence of the fusion construct and phleomycin resistance gene were confirmed through PCR (Fig. S5). The strain of A. fumigatus expressing A. leporis easH was analyzed by LC-MS and compared to the easM knockout strain of A. fumigatus and wild-type A. japonicus Saito NRRL 360, a documented cycloclavine producer (48). An analyte of m/z 239.1, corresponding to cycloclavine (as confirmed by LC-HRMS, Fig. S6), was detected in extracts of A. japonicus, but no corresponding metabolite was detected in any of the A. fumigatus colonies expressing easH from A. leporis (Fig. S6). Reverse transcription-PCR (RT-PCR) analyses of triplicate malt extract cultures of this mutant demonstrated the presence of easH mRNA (Fig. S7), indicating that failure to generate cycloclavine was not due to lack of easH expression.

DISCUSSION

Our data show that A. leporis is an unusual species of ergot alkaloid producing fungus as it possesses two partial gene clusters encoding some of the genes necessary for fumigaclavine A and rugulovasine production, in addition to two gene clusters encoding for LAH biosynthesis one of which is complete, and the other lacks a functional copy of easD (57). We hypothesize the fumigaclavine and rugulovasine clusters act as satellite clusters of the LAH cluster because they lack steps required for production of early steps, thus requiring intermediates chanoclavine-I or chanoclavine-I aldehyde from one of two previously characterized LAH gene clusters (57). We identified fumigaclavine A in extracts of A. leporis, with 65% of the total amount accumulating being retained within the hyphae and/or conidiophores and conidia. An easD knockout strain of A. leporis (59) was found to accumulate fumigaclavine A at more than three times the rate of the wild-type fungus. This result is consistent with the reliance of this pathway on intermediates from the LAH pathway and the observed activation of this satellite pathway at a later point in time relative to LAH accumulation.

We were only able to detect rugulovasine accumulation in a single sample of A. leporis, prompting the need to better test functionality of easQ and easH. Heterologous expression of these genes in an easA knockout strain of M. brunneum resulted in reliable rugulovasine accumulation in the engineered mutants. This result indicates that the products of easQ and easH from the satellite cluster are functional and encode the capacity to convert chanoclavine-I aldehyde to rugulovasine. The rarity with which we detected rugulovasine in cultures of A. leporis suggests the fungus requires some specific nutritional or environmental conditions to accumulate rugulovasine. Determination of these conditions is beyond the scope of this present study; however, one factor that may have varied tube-to-tube was gas exchange, because tubes were covered with loosely closed screw caps. The A. fumigatus mutant expressing A. leporis easH alone failed to produce cycloclavine, indicating specialization among versions of EasH from different pathway branches.

Aspergillus leporis is the only fungus known to possess at least four biosynthetic gene clusters encoding three different branches of the ergot alkaloid pathway. Two of these branches requiring activity of the main branch for early pathway steps. These findings are significant as they support the idea that different offshoots of the ergot alkaloid pathway have evolved to shunt products into various pathway branches depending on specific environmental conditions. While A. leporis is unique among ergot alkaloid producers for possessing multiple satellite gene clusters for core pathway modifications, other fungi have also been found to utilize some discontiguous genetic loci to complete or modify specialized metabolites. Biosynthesis of different trichothecenes in Fusarium species is encoded by a set of up to 12 Tri genes clustered on chromosome 2 (63). A two-gene satellite cluster has been identified on chromosome 1 (64, 65), while at least two other single-gene satellites have been found elsewhere in the genome (66–68). Aspercryptin biosynthesis in Aspergillus nidulans relies on the combinatory action of two different biosynthetic gene clusters (69). The utilization of satellite gene clusters for specialized metabolite biosynthesis has also been characterized in plants, as is seen with α-tomatine in tomato (70), cucurbitacins in cucumber (71), and morphine in opium poppy (72). Our results provide further evidence that fungi may increase the diversity of their specialized metabolites through multiple, related biosynthetic gene clusters. Potential products of physically separated gene clusters should be considered when taking approaches to mine fungal genomes for novel specialized metabolites or when modifying characterized clusters in mutant strains.

Fumigaclavine A was the fumigaclavine we detected in A. leporis, but the genome of the fungus indicates it, or a recent ancestor, once had the capacity to synthesize fumigaclavine C. The final step of fumigaclavine C production in Aspergillus species is controlled by the reverse prenyltransferase gene easL, or fgaPT1 (43). This step is absent from Penicillium species that produce fumigaclavines, resulting in cessation of the pathway at fumigaclavine A (73). Previous studies of Penicillium commune have shown that this species produces fumigaclavine A as a pathway end product due to lack of easL (74, 75). Here, we show that A. leporis also accumulates fumigaclavine A (as opposed to fumigaclavine C) due to inactivation of easL by an insertion that alters the reading frame and prevents its product from being translated correctly. Interestingly, the same set of genes is also found in Penicillium roqueforti, which produces isofumigaclavines A and B (stereoisomers of fumigaclavines A and B), but the genes are split into two clusters encoded by separate areas of the genome (76).

Cycloclavine is an uncommon ergot alkaloid found only in a few species of fungi that possess a specific version of EasH (50, 51). Our phylogenetic data indicate that the EasH responsible for cycloclavine production shares a common ancestor with the versions required for rugulovasine production and those involved in ergopeptine biosynthesis. While we demonstrated the ability of A. leporis EasH (together with EasQ) to catalyze rugulovasine production, it was unable to catalyze formation of cycloclavine when engineered alone in an appropriate mutant background in A. fumigatus. This result indicates that a degree of specialization exists among the different versions of EasH. Investigations into any differences in their crystal structures as well as their substrate-binding specificity may provide clues regarding differences in their mechanisms.

Evolutionary data across fungal species suggest that the ancestor to all ergot alkaloid producers may have been related to extant Aspergillus species and that neofunctionalization of easH coupled with entire cluster duplications played an important role in ergot alkaloid diversification (2). Our results show that A. leporis retains the capacity for production of two clavine branches of the pathway that rely on obtaining an intermediate from one or both of the duplicated gene clusters encoding lysergic acid amide production. Jones et al. (57) previously proposed that genes encoding the final three steps of the LAH pathway in A. leporis evolved independently from those catalyzing similar steps in the Clavicipitaceae and that lysergic acid amides provide an advantage to A. leporis during infection of a model insect (59). The persistence of multiple ergot alkaloid synthesis clusters suggests that a variety of ergot alkaloids confers a benefit to the fungus, perhaps depending on the environment the fungus finds itself in. Further studies into the ecology of the fungus may provide clues as to the nature of these specific benefits.

MATERIALS AND METHODS

Identification of novel A. leporis eas clusters.

The variants of EasD (KAB8071224.1), EasA (KAB8071228.1), and EasG (KAB8071227.1) noted by Jones et al. (57) as encoded by homologs not associated with either of the two characterized LAH-associated gene clusters were analyzed by tBLASTn against the whole-genome shotgun database for A. leporis. The single contig that each of these genes was found to originate from (GenBank accession number SWBU01000167.1) was downloaded and annotated for easD, easG, and easA. The areas adjacent to these genes were analyzed in 5-kb fragments by BLASTx to determine presence of other potential ergot alkaloid genes. Copies of genes with products resembling EasM and EasN were noted along with copies of easL and easK that appear to be pseudogenes. The areas flanking this cluster of eas genes were searched in successive 5 kb fragments three times in either direction before it was concluded no other eas genes were present nearby.

Penicillium biforme EasH was analyzed by tBLASTn against the whole-genome shotgun database for Aspergillus species. From the returned matches, A. leporis was chosen for further analysis. The matching contig (GenBank accession number SWBU01000070.1) was downloaded and annotated as described above, revealing the presence of a homolog of easQ adjacent to easH. The areas adjacent to these genes were analyzed in successive 5-kb fragments by BLASTx to investigate the presence of other potential ergot alkaloid genes. The process was repeated three times in either direction before we concluded no other ergot alkaloid synthesis genes were detected.

Growth and maintenance of fungi.

Cultures of A. leporis NRRL 3216 and an easD knockout of A. leporis (59) were maintained on sucrose-yeast extract agar medium (per L, 20 g sucrose, 10 g yeast extract, 1 g magnesium sulfate-heptahydrate, and 15 g agar). Cultures were grown at 30°C for at least 7 days before testing for ergot alkaloids. Cultures of M. brunneum ARSEF 9354 and its mutant derivatives were maintained on sucrose-yeast extract agar medium and were grown for at least 10 days prior to collecting conidia for injection into G. mellonella larvae. Cultures of the previously engineered easM knockout strain of A. fumigatus (41) were maintained at 37°C on malt extract agar medium (per L, 6.0 g malt extract, 1.8 g maltose, 6.0 g dextrose, 1.2 g yeast extract, and 15 g agar) and were grown for at least 7 days before testing for ergot alkaloids.

Sample preparation for alkaloid analyses.

To measure the number of moles of fumigaclavine A retained in the hyphae compared to the quantity secreted into the medium liquid, seven cultures of A. leporis NRRL 3216 were grown, prepared, and harvested as described by Jones et al. (57). For the comparison of fumigaclavine A accumulation in wild-type versus the easD knockout, seven replicate cultures of each strain were compared. Replicate cultures were inoculated with a common spore suspension (30,000 conidia/tube) and grown in 0.5 mL of sucrose-yeast extract medium in 2-mL screw cap tubes (part number 02–681-343; Thermo Fisher Scientific, Pittsburgh, PA) with the lids slightly ajar to allow gas exchange. For malt extract agar cultures of P. biforme NRRL 885 and A. japonicus NRRL 360, alkaloids were extracted from ~400 μL samples of colonized medium with 400 μL of methanol. Samples were rotated end-over-end (40 rpm) at room temperature for 1 h and clarified by centrifugation. The P. biforme extract served as reference for rugulovasines as it has been characterized previously as containing these compounds, whereas the A. japonicus extract provided a reference for cycloclavine (38, 54, 56). Extracts from these reference materials were prepared at least two times for each isolate. Mutant cultures of A. fumigatus generated during this study were grown on malt extract agar medium for 7 days before the conidiating surface of the entire Petri dish was extracted by repeatedly washing with 2 mL methanol. The entire volume of each A. fumigatus extract was concentrated to ~200 μL in a vacuum concentrator prior to analysis. Wild-type A. leporis and mutants of M. brunneum generated during this study were grown on sucrose-yeast extract agar medium for 10 days prior to injection of their conidia into G. mellonella larvae. After 8 days, the larval cadavers were bead beaten with 10 3-mm diameter glass beads in 1 mL methanol at 6 m/s for 30 s. The resulting extracts were then rotated end-over-end at 40 rpm for 1 h before being clarified by centrifugation. For HPLC analyses, 20 μL of each sample was injected, and for LC-HRMS analyses, 10 μL of each sample was injected.

HPLC analyses.

Samples were analyzed for ergot alkaloids by HPLC with fluorescence detection by methods described in detail previously (77). The column was a 150- by 4.6-mm inner-diameter, 5-μm particle size Prodigy C₁₈ column (Phenomenex, Torrance, CA), and the mobile phase was a 55-min, binary, multilinear gradient of 5% acetonitrile to 75% acetonitrile in 50 mM aqueous ammonium acetate. Chanoclavine and fumigaclavine A were detected using excitation and emission wavelengths of 272 nm and 372 nm, respectively, and LAH was detected by excitation at 310 nm and recording emission at 410 nm. For secretion analyses, alkaloids were quantified according to Davis et al. (37). Chanoclavine-I was quantified by comparing peak areas to an external standard curve prepared from chanoclavine-I standard (Alfarma, Prague, Czech Republic). Fumigaclavine A was relative to an external standard curve prepared from fumigaclavine A (Alexis Biochemicals, San Diego, CA). LAH was quantified by comparing peak areas to an external standard curve prepared from ergonovine (Sigma-Aldrich, St. Louis, MO), which contains the identical fluorophore; thus, LAH values must be considered relative to ergonovine as opposed to absolute.

LC-HRMS analyses.

Initial LC-MS analyses were conducted on a Thermo LCQ Deca XP plus mass spectrometer connected to a Thermo Surveyor HPLC system (Thermo Fisher Scientific, Waltham, MA) as described previously (56). High-resolution mass spectra were collected on a Thermo Scientific Q Exactive mass spectrometer coupled to a Thermo Accela 1250 ultra-high-performance liquid chromatography (UHPLC) system. Separations were performed on a 150-mm by 4.6-mm inner-diameter, 2.6-μm particle size Kinetex Evo C₁₈ column (Phenomenex, Torrance, CA) subjected to a gradient prepared by combining mobile phase A (5% acetonitrile, 0.1% formic acid) and mobile phase B (75% acetonitrile, 0.1% formic acid). The sample was loaded at 95% A plus 5% B and held for 1 min before ramping linearly to 40% A plus 60% B at 10 min at a flow rate of 300 μL/min. The mass spectrometer was operated in positive ion mode and programmed with data-dependent acquisition settings. Analytes were electrospray ionized with a spray voltage of 3.5 kV and a capillary temperature of 300°C. Precursor scans were acquired at 70,000 resolution (at m/z 200) over an m/z range of 150 to 400. Ions of m/z = 239.1535, 269.1278, and 299.1747 were selected for higher-energy collisional-dissociation (HCD) fragmentation (normalized collisional energy [NCE] = 30) and were analyzed at 35,000 resolution.

Phylogenetic analyses.

The respective protein sequence translated for EasQ from eas clusters of A. leporis, P. biforme, and P. camemberti were used as the query in a BLASTp search for homologous proteins in the NCBI database for each organism. A cutoff of at least 30% identity over 70% query coverage was used to select matches from the returned list to be included in the data set for phylogenetic analysis. The objective of these searches was to find the two closest homologs of a particular eas gene in an organism’s genome. We set lower limits based in part on values discussed by Pearson (78) and on empirical experience from a similar study (57). Comparisons that included sequences not meeting these criteria resulted in alignments that trimmed off numerous informative amino acids from other well-aligning homologs. If more than two proteins from a given species met those criteria, only the top two matches were used. Because M. brunneum and A. fumigatus lack an eas cluster-associated version of EasQ, the EasQ sequences from A. leporis and P. camemberti were both used to search M. brunneum and A. fumigatus databases and the top three matches from both species were chosen (provided they met the criteria delineated above). When the BLASTp search resulted in fewer matches than outlined above, a tBLASTn search was performed against the whole-genome shotgun database for that organism. If hypothetical proteins queried in this manner met the cutoff criteria described above, the protein sequences were deduced by BLASTx comparison of the appropriate regions of the identified contigs. Accession numbers for homologs not already identified as a functional eas cluster-associated EasQ can be found in Fig. 6. This process was repeated for EasH alleles with the only change being the inclusion of organisms having versions of EasH shown to catalyze key steps in the synthesis of cycloclavine (A. japonicus and Byssochlamys spectabilis) or ergopeptines (C. purpurea and Periglandula ipomoeae). Accession numbers for homologs not already identified as a functional EasH and associated with eas clusters can be found in Fig. 7. Alignments and phylogenetic analyses were performed in MEGA X (79). Once both sets of proteins for EasQ and EasH were assembled, sequences were aligned by MUSCLE as contained in MEGA X, accepting the defaults, and aligned proteins were trimmed by eye. Model tests of both data sets indicated that the Le and Gascuel (LG) model using a discrete Gamma distribution (+G) was most appropriate, so this was used for maximum likelihood analyses of both the EasQ and EasH data sets (80). At total of 1,000 bootstrap analyses were performed for both data sets, and bootstrap values of 50% or higher were included at the relevant nodes. Neighbor-joining trees constructed with 1,000 bootstrap analyses were consistent with the data derived from maximum likelihood analyses (not included).

Preparation of transformation constructs.

Genomic DNA (gDNA) was extracted from wild-type A. leporis, wild-type M. brunneum, and easM ko A. fumigatus samples according to the GeneClean spin protocol (MP Biomedicals, Solon, OH). The A. leporis gDNA was used as template in PCRs with primer combinations 1 and 3 (Table 2) to generate the A. leporis easQ and A. leporis easH fragments with their native 3′ untranslated regions (UTRs) (nucleotides 47,107 to 48,732 and nucleotides 51,683 to 51,702, respectively, in the record under GenBank accession number SWBU01000070.1) with 5′ overlaps for the M. brunneum bidirectional easG/dmaW promoter (37) (Fig. S8). These and all subsequent PCRs were performed with Phusion green hot start II high-fidelity PCR master mix (Thermo Fisher Scientific, Waltham, MA) and followed similar protocols as described previously by Davis et al. (37) and with primers and reaction conditions detailed in Table 2. All products were gel purified using a ZymoClean gel DNA recovery kit (Zymo) prior to performing fusion PCR. The M. brunneum gDNA was used as template with primer combination 2 to amplify the bidirectional easG/dmaW promoter with overlaps for A. leporis easQ and A. leporis easH (Fig. S8). An equimolar ratio of the A. leporis easQ, M. brunneum bidirectional promoter, and A. leporis easH fragments were fused using primer combination 6 (Fig. S9). The final product included unique restriction sites for MluI and XhoI near their termini allowing for double digestion, gel purification, and ligation into pBChygro (Fungal Genetics Stock Center, Manhattan, KS) (62) generating A. leporis easQ-A. leporis easH-pBChygro. Competent Escherichia coli cells were transformed with this plasmid and plated on LB medium (per L, 10 g tryptone, 5 g yeast extract, 5 g NaCl, and 15 g agar) supplemented with chloramphenicol (25 g/mL). Plasmid products were harvested from selected colonies and purified using a Zyppy plasmid miniprep kit (Zymo). Correct assembly was verified through double digestion with MluI/XhoI. The plasmid was linearized using SnaBI and purified with a PCR purification kit (Zymo) prior to fungal transformation.

TABLE 2.

Primers and PCR protocol information

Primer pair	Primer sequences (5′ to 3′)a	Product (length in base pairs)	Annealing temp (°C), Extension time (s)
1	CTCAACGCGTGAACGAATAGACGTCGACCC + CCATCTCGGAAAAGAAAAATGTCGGGATTCGTGGTCCA	A. leporis easQ + overlap M. brunneum easG/dmaW promoter (1877 bp)	63, 60
2	TGGACCACGAATCCCGACATTTTTCTTTTCCGAGATGG + CGGTGGTGGGAACCGTCATCGTAAACCAGAGTATTATG	M. brunneum easG/dmaW promoter + overlaps for A. leporis easQ & A. leporis easH (1325 bp)	53, 60
3	CATAATACTCTGGTTTACGATGACGGTTCCCACCACCG + CGTACTCGAGCAGCAAGGGGTAGGGAATGA	A. leporis easH + overlap M. brunneum easG/dmaW promoter (1140 bp)	64, 60
4	CGTAACGCGTGCTTCTAATCCACCAAGTACTTG + CGGTGGTGGGAACCGTCATGGTGCGGAGTGCCTACTCTA	A. fumigatus easA promoter + overlap A. leporis easH (827 bp)	65, 30
5	TAGAGTAGGCACTCCGCACCATGACGGTTCCCACCACCG + CGTATCTAGACAGCAAGGGGTAGGGAATGA	A. leporis easH + overlap A. fumigatus easA promoter (1140 bp)	64, 60
6	CTCAACGCGTGAACGAATAGACGTCGACCC + CGTACTCGAGCAGCAAGGGGTAGGGAATGA	A. leporis easQ – M. brunneum easG/dmaW promoter – A. leporis easH fusion (4281 bp)	72, 150
7	CGTAACGCGTGCTTCTAATCCACCAAGTACTTG + CGTACTCGAGCAGCAAGGGGTAGGGAATGA	A. fumigatus easA promoter – A. leporis easH fusion (1916 bp)	72, 60
8	TCCGCCAATTAGGGTTCGAC + CAGCGTACCGACCTCCATC	A. leporis easQ – M. brunneum easG/dmaW promoter – A. leporis easH-pBChygro mutant verification (3914 bp)	65, 120
9	TCCGTCTCCATTGGCTCTTG + CTATTCCTTTGCCCTCGGAC	Verification of hygromycin resistance gene in M. brunneum mutants (1835 bp)	64, 60
10	GAAGAGACTGAGGGTGGTGG + AACAAGCAAGCCCAAGCAAC	A. fumigatus easA promoter – A. leporis easH fusion – pBCphleo mutant verification (1309 bp)	64, 60
11	GTACCCGGGGATCTTTCGAC + ACCAAAGGCCATCTTGGTAC	Verification of phleomycin resistance gene in A. fumigatus mutants (2644 bp)	63, 90
12	CAATCCTCAGTGCGCTACAAG + GACGCATGTTGCGCCATTC	A. leporis easH fragment (446 bp cDNA and genomic DNA)	64, 30
13	GCTGGAGCGTATGAACGTC + GTTGTTACCAGCACCGGAC	A. fumigatus βtub fragment (172 bp cDNA and 237 bp genomic DNA)	65, 30

^aUnderlines indicate unique restriction sites used for cloning fusion PCR product: ACGCGT: MluI, CTCGAG: XhoI, and TCTAGA: XbaI.

DNA from the easM knockout of A. fumigatus was used as template with primer combination 4 to amplify the easA promoter (33) with an overlap for A. leporis easH (Fig. S8). Primer combination 5 was used with the A. leporis gDNA template to generate A. leporis easH with a 5′ overlap for the A. fumigatus promoter. Equimolar ratios of A. fumigatus promoter and A. leporis easH fragments were fused together using primer combination 7 (Fig. S9). The A. fumigatus promoter-A. leporis easH final product included restriction sites for MluI/XbaI and was processed as described above with the substitution of pBCphleo (Fungal Genetics Stock Center, Manhattan, KS) (62) as the parent plasmid, generating A. leporis easH-pBCphleo. Correct assembly of the plasmid was confirmed through double digestion with MluI/XbaI. The plasmid was linearized using AleI and purified using a PCR purification kit prior to fungal transformation.

Transformation of M. brunneum and A. fumigatus.

To test the capacity of EasQ and EasH from A. leporis to generate rugulovasines from chanoclavine-I aldehyde, protoplasts from an M. brunneum easA knockout strain were prepared and transformed with a construct containing A. leporis easQ and A. leporis easH cloned into pBChygro according to a previously described protocol (37). This strain was selected as the recipient as it accumulates chanoclavine-I aldehyde, the precursor to rugulovasines A/B. Transformants were plated in TM102 medium (per L, 310 g sucrose, 10 g malt extract, 10 g peptone, 2 g yeast extract, 1 g magnesium sulfate-heptahydrate, 0.5 g monobasic potassium phosphate, 0.5 g dibasic potassium phosphate, 0.5 g potassium chloride, 0.05 g chloramphenicol, and 15 g agar) supplemented with hygromycin (InvivoGen, San Diego, CA) at 600 μg/mL and incubated at 30°C. Upon surfacing, colonies were transferred to sucrose-yeast extract agar medium supplemented with hygromycin at 300 μg/mL and incubated at 30°C to allow further growth under selection. Genomic DNA was extracted from selected colonies and checked by PCR for presence of the introduced constructs. Primer combinations 8 and 10 were used to verify the presence of A. leporis easQ and A. leporis easH portion of the construct and the hygromycin resistance gene, respectively, in the M. brunneum background (Fig. S5).

To test the ability of A. leporis EasH to generate cycloclavine in a conducive genomic context, protoplasts of an easM knockout strain of A. fumigatus (41) were transformed with a construct comprised of A. leporis easH cloned into pBCphleo. This strain was selected as the recipient because it accumulates festuclavine and its precursor chanoclavine-I aldehyde while leaving easA_red intact for cycloclavine production. Transformants were plated in TM102 supplemented with phleomycin at 200 μg/mL (InvivoGen, San Diego, CA) and incubated at 37°C. Upon surfacing, colonies were transferred to malt extract agar medium supplemented with phleomycin at 100 μg/mL and incubated at 37°C to allow further growth under selection. Genomic DNA was extracted from selected colonies and checked by PCR for presence of the introduced constructs. Primer combinations 11 and 12 were used to verify the presence of A. leporis easH and the phleomycin resistance gene, respectively, in the A. fumigatus background (Fig. S5).

RNA extraction and RT-PCR.

Triplicate cultures of the A. fumigatus mutant expressing A. leporis easH were grown in 25 mL malt extract medium (lacking agar) inside deep-dish culture plates and incubated overnight at 37°C. The plates were kept still to ensure formation of a single mat of hyphae on the surface of the broth with sparse conidia beginning to develop. RNA was extracted from approximately 100 mg of a conidiating portion of the mat with the Plant RNeasy kit (Qiagen), treated with DNase I (Qiagen), and reverse transcribed with Superscript IV (Invitrogen, Carlsbad, CA). RNA extraction and cDNA synthesis of replicates were performed concurrently to ensure consistency. Template cDNA from each replicate was diluted 1:100 prior to PCR amplification. Primer combinations 12 and 13 were used to amplify portions of easH and an endogenous β-tubulin encoding gene (GenBank accession number XP_752456.1) from each cDNA sample as well as from gDNA isolated from a malt extract agar culture of the mutant (Fig. S7). Because easH lacks introns, amplification of a β-tubulin encoding gene was carried out with primers flanking an intron to serve as a control for cDNA quality.

Statistical analyses.

Concentrations of chanoclavine-I and fumigaclavine A in cultures (n = 7) of wild-type and easD knockout strains of A. leporis were measured as described above (under HPLC analyses) and used to calculate the percent conversion of precursor to product (Table 1). Data were checked by Brown-Forsythe tests for normality of variances. Data for chanoclavine-I accumulation did not pass (P < 0.05) and were therefore analyzed using a Wilcoxon’s rank sum test. The variances of both fumigaclavine A accumulation and overall percent conversion were found to be normally distributed by Brown-Forsythe tests and were subsequently analyzed using paired t-tests. All statistical analyses were performed with the JMP software package (SAS, Cary, NC).

Data availability.

Quantitative data related to alkaloid accumulation in cultures of A. leporis and trimmed sequences used in phylogenetic analyses are available in files accessible via the following URL: https://datadryad.org/stash/dataset/doi:10.5061/dryad.7pvmcvdzt.