Lowell P. Bush

Bioprotective Alkaloids of Grass-Fungal Endophyte Symbioses

Symbiotic interactions of C3 grasses with fungal endophytes, Epichloe species and their asexual relatives Neotypkodium, often provide the grass hosts with major fitness enhancements (for review, see Siegel and Bush [1994]). The endophytes protect host plants from both biotic and abiotic environmental stresses. Documented enhancements to host fitness include greater resistance to mammalian and insect herbivores, pathogens, and nematodes, as well as increased drought tolerance and competitiveness. Our understanding of the mechanisms responsible for a11 of these effects on host fitness is not well established. However, the antiherbivore fitness enhancements are largely attributable to the accumulation of four groups of alkaloids: lolines, peramine, ergot alkaloids, and lolitrems. The objective of this Update is to present the current status of knowledge on these alkaloids, focusing on their biological activities and the potential for genetically manipulating their expression in grasslendophyte symbiota.

Fungal endophyte-infected grasses: Alkaloid accumulation and aphid response.

The occurrence of the alkaloidsN-formyl andN-acetyl loline, peramine, lolitrem B, and ergovaline and the response of aphids to plants containing these compounds were determined in species and cultivars ofFestuca,Lolium, and other grass genera infected with fungal endophytes (Acremonium spp., andEpichloe typhina). Twenty-nine of 34 host-fungus associations produced one or more of the alkaloids, most frequently peramine or ergovaline. Three alkaloids (lolines, peramine, and ergovaline) were found in tall fescue and in perennial ryegrass infected withA. coenophialum, while peramine, lolitrem B, and ergovaline were present in perennial ryegrass and in tall fescue infected withA. lolii and inF. longifolia infected withE. typhina. WhileA. coenophialum andA. lolii produced similar patterns of alkaloids regardless of the species or cultivar of grass they infected, isolates ofE. typhina produced either no alkaloids or only one or two different alkaloids in the grasses tested. Aphid bioassays indicated thatRhopalosiphum padi andSchizaphis graminum did not survive on grasses containing loline alkaloids and thatS. graminum did not survive on peramine-containing grasses. Ergovaline-containing grasses did not affect either aphid.

Contribution of fungal loline alkaloids to protection from aphids in a grass-endophyte mutualism.

Fungal endophytes provide grasses with enhanced protection from herbivory, drought, and pathogens. The loline alkaloids (saturated 1-aminopyrrolizidines with an oxygen bridge) are fungal metabolites often present in grasses with fungal endophytes of the genera Epichloë or Neotyphodium. We conducted a Mendelian genetic analysis to test for activity of lolines produced in plants against aphids feeding on those plants. Though most loline-producing endophytes are asexual, we found that a recently described sexual endophyte, Epichloë festucae, had heritable variation for loline alkaloid expression (Lol+) or nonexpression (Lol-). By analyzing segregation of these phenotypes and of linked DNA polymorphisms in crosses, we identified a single genetic locus controlling loline alkaloid expression in those E. festucae parents. We then tested segregating Lol+ and Lol- full-sibling fungal progeny for their ability to protect host plants from two aphid species, and observed that alkaloid expression cosegregated with activity against these insects. The in planta loline alkaloid levels correlated with levels of anti-aphid activity. These results suggested a key role of the loline alkaloids in protection of host plants from certain aphids, and represent, to our knowledge, the first Mendelian analysis demonstrating how a fungal factor contributes protection to plant-fungus mutualism.

Chemistry, occurrence and biological effects of saturated pyrrolizidine alkaloids associated with endophyte-grass interactions

Abstract Saturated amino pyrrolizidine alkaloids with an oxygen bridge between C-2 and C-7 have been isolated from grasses and the chemistry and biology of these substances are reviewed in this paper. These substances are commonly referred to as loline alkaloids as they are derivatives of loline and originally were isolated from Lolium spp. The saturated amino pyrrolizidine alkaloids are not hepatotoxic and must be distinguished from the 1,2-unsaturated pyrrolizidine alkaloids which are very significant animal and human toxins and carcinogens of plant origin. Chemical synthesis of loline has been achieved based on starting materials of a nitrone and methyl 4-hydroxycrotonate. N-formylloline is made by addition of ethyl formate to loline and removal of excess ethyl formate and ethanol formed. N-acetylloline is synthesized by addition of acetyl chloride to loline in chlorofirm. N-acetylloline is partitioned into a dilute acid, then from a dilute base with CHCl3, and the CHCl3 removed to yield N-acetylloline. N-methylloline, norloline, N-formynorloline, N-acetylloline, and N-propionylnorloline are also readily formed from loline. Little is known about the biosynthesis of the loline alkaloids but from biosynthesis of other pyrrolizidine alkaloids a proposed biosynthesis is ornithine → putrescine → spermidine → → dialdehyde → → norloline → → loline. Loline was first isolated from Lolium temulentum L. (Lolium cuneatum Nevski) and later several derivatives were identified in tall fescue (Festuca arundinacea Schreb.). Chromatographic methodologies have been used to separate the pyrrolizidine alkaloids but 50-m capillary columns (0.32 mm i.d.) or 10-m wide-bore columns (0.53 mm i.d.) provide the greatest resolution of the many loline derivatives. Loline alkaloids have been found primarily in plants known to be infected witn Acremonium coenophialum Morgan-Jones and Gams. Lolines have been found in Lolium perenne L., F. arundinacea, and Poa autumnalus Muhl. ex Ell. infected with A. coenophialum and in Festuca gigantea (L.) Vill. infected with an unidentified Acremonium spp. Loline alkaloids were present in infected tall fescue in greatest amounts in seed, followed in decreasing amounts in the rachis, stem, leaf sheath, and leaf blade. Site of synthesis, plant and/or endophyte, is not known nor is the translocation tissue known. During the main part of the growing season there is little change in the accumulation of the loline alkaloids in leaf blade. Alkaloid levels increased with plant age and increased in regrowth tissue. The concentration of endophyte mycelium in leaf sheaths is positively associated with loline alkaloid accumulation. N-formylloline is toxic to several different insects from ingestion, topical and injected applications. Larger animal bioassay have not been conducted with pure compounds but the observations suggest some pharmacological activity in these systems.

Different Levels of Protective Alkaloids in Grasses with Stroma-forming and Seed-transmitted Epichloë/Neotyphodium Endophytes

The three alkaloid groups—lolines, ergopeptides, and peramine— are typically associated with endophyte infection of grasses, with the main function to protect hosts against herbivores. We determined levels of N-formylloline, N-acetylloline, ergovaline, and peramine in 18 European grasses naturally infected with seed-transmitted Neotyphodium endophytes or sexual Epichloë species. Peramine was the most common alkaloid, whereas lolines and ergovaline were only detected in Festuca hosts infected with E. festucae, N. coenophialum, or N. uncinatum. Only ten of the grass species analyzed contained detectable amounts of one or more of these alkaloids. There was a clear tendency for plants associated with stroma-forming Epichloë species to be free of alkaloids, and those that did produce alkaloids contained only small levels of peramine. In contrast, plants infected with seed-transmitted Neotyphodium endophytes often contained extremely high levels of lolines. Lolines enhance host survival through increased protection from herbivores and, thus, may be particularly favored in asexual endophytes that depend on host seed-production for their dispersal.

Production of loline alkaloids by the grass endophyte, Neotyphodium uncinatum, in defined media.

Lolines (saturated 1-aminopyrrolizidines with an oxygen bridge) are insecticidal alkaloids produced in symbioses of certain Epichloë (anamorph-Neotyphodium) species (fungal endophytes) with grasses, particularly of the genera Lolium and Festuca. Prior to the present study, it was unknown whether lolines were of plant or fungal origin. Neotyphodium uncinatum, the common endophyte of meadow fescue (Lolium pratense=Festuca pratensis) produced loline, N-acetylnorloline, and N-formylloline when grown in the defined minimal media at pH 5.0-7.5, with both organic and inorganic nitrogen sources and sugars as carbon sources. In contrast, lolines were not detected in complex medium cultures. GC-MS and 13C NMR spectroscopic analyses confirmed the identity of the alkaloids isolated from the defined medium cultures. Lolines accumulated to ca. 700 mg/l (4 mM) in cultures with 16.7 mM sucrose and 15-30 mM asparagine, ornithine or urea. Kinetics of loline production and fungal growth were assessed in defined medium with 16.7 mM sucrose and 30 mM ornithine. The alkaloid production rate peaked after the onset of stationary phase, as is common for secondary metabolism in other microbes.

Defensive chemicals in grass-fungal endophyte associations.

The ability of organisms to form long term intimate and diverse relationships with each other (symbiosis) is now recognized as a common ecological phenomenon. Symbiosis, as a general term, does not imply detriment or benefit, but rather that the outcome (net effect) of species interaction exists within a symbiotic continuum or “species interaction grid” that includes agonism (predation and disease) and mutualism (benefits for both species).1,2 The continuum also includes pleotropic symbiosis, where net effects of species interaction vary spatially or temporally in relative agonism or mutualism.3 Grasses systematically infected with specific clavicipitaceous fungi are examples of species interactions that span the symbiotic continuum, profoundly affecting the ecological fitness of the hosts.4–8

Effects of multiple climate change factors on the tall fescue–fungal endophyte symbiosis: infection frequency and tissue chemistry

• Climate change (altered CO(2) , warming, and precipitation) may affect plant-microbial interactions, such as the Lolium arundinaceum-Neotyphodium coenophialum symbiosis, to alter future ecosystem structure and function. • To assess this possibility, tall fescue tillers were collected from an existing climate manipulation experiment in a constructed old-field community in Tennessee (USA). Endophyte infection frequency (EIF) was determined, and infected (E+) and uninfected (E-) tillers were analysed for tissue chemistry. • The EIF of tall fescue was higher under elevated CO(2) (91% infected) than with ambient CO(2) (81%) but was not affected by warming or precipitation treatments. Within E+ tillers, elevated CO(2) decreased alkaloid concentrations of both ergovaline and loline, by c. 30%; whereas warming increased loline concentrations 28% but had no effect on ergovaline. Independent of endophyte infection, elevated CO(2) reduced concentrations of nitrogen, cellulose, hemicellulose, and lignin. • These results suggest that elevated CO(2) , more than changes in temperature or precipitation, may promote this grass-fungal symbiosis, leading to higher EIF in tall fescue in old-field communities. However, as all three climate factors are likely to change in the future, predicting the symbiotic response and resulting ecological consequences may be difficult and dependent on the specific atmospheric and climatic conditions encountered.

Effects of selected combinations of tall fescue alkaloids on the vasoconstrictive capacity of fescue-naive bovine lateral saphenous veins.

Vasoconstriction is a response associated with consumption of toxic endophyte-infected tall fescue. It is not known if endophyte-produced alkaloids act alone or collectively in mediating the response. Therefore, the objective of this study was to examine the vasoconstrictive potentials of selected ergot alkaloids, individually or in paired combinations, using bovine lateral saphenous veins biopsied from fescue-naïve cattle. Segments (2 to 3 cm) of vein were surgically biopsied from healthy crossbred yearling heifers (n = 22; 330 +/- 8 kg of BW). Veins were trimmed of excess fat and connective tissue, sliced into 2- to 3-mm sections, and suspended in a myograph chamber containing 5 mL of oxygenated Krebs-Henseleit buffer (95% O(2)/5% CO(2); pH = 7.4; 37 degrees C). Increasing doses of ergovaline, lysergic acid, and N-acetylloline individually or in combination were evaluated. Contractile data were normalized as a percentage of the contractile response induced by a reference dose of norepinephrine (1 x 10(- 4) M). Increasing concentrations of lysergic acid did not result in an appreciable contractile response until the addition of 1 x 10(- 4) M lysergic acid. In contrast, the vascular response to increasing concentrations of ergovaline was apparent at 1 x 10(- 8) M and increased to a maximum of 104.2 +/- 6.0% with the addition of 1 x 10(- 4) M ergovaline. The presence of N-acetylloline did not alter the onset or magnitude of vascular response to either lysergic acid or ergovaline. The presence of 1 x 10(- 5) M lysergic acid with increasing concentrations of N-acetylloline and ergovaline generated an increased contractile response during the initial additions compared with the responses of N-acetylloline and ergovaline alone. In the presence of 1 x 10(- 7) M ergovaline, the contractile response increased with increasing concentrations of N-acetylloline and lysergic acid. Neither N-acetylloline nor lysergic acid elicited an intense contractile response individually (maximum contractile responses of 1.9 +/- 0.3% and 22.6 +/- 4.1%, respectively), suggesting that this was the result of the repetitive addition of 1 x 10(- 7) M ergovaline. These data indicate that ergovaline is a more potent vascular toxicant than lysergic acid or N-acetylloline. The contractile responses of the ergovaline and lysergic acid combinations appeared to differ from the individual dose responses. These data support the possibility that an additive alkaloid exposure effect may exist and should be considered during evaluations of ergot alkaloids.

Peramine alkaloid variation in Neotyphodium-infected Arizona fescue: effects of endophyte and host genotype and environment.

We determined concentrations of peramine, the only alkaloid produced by Neotyphodium-infected (E+) Arizona fescue plants (of the four major types typically assayed in infected grasses), in a long-term field experiment. Four plant genotypes with (E+) and without (experimentally removed, E−) their respective haplotypes (two haplotypes in two plant genotypes) of Neotyphodium were grown in the field under manipulated soil moisture and nutrients. Peramine production required the presence of the endophyte; plants without their endophytes did not contain peramine. Whereas the endophyte was necessary for peramine production, levels of peramine did not vary by Neotyphodium haplotype within plant genotypes. Furthermore, peramine levels did not differ among soil moisture and nutrient treatments, and growth and reproductive measures of the host grass explained little of variation in peramine levels. Instead, peramine levels differed significantly between plant genotypes harboring the same endophyte haplotype, suggesting that plant genotype, rather than endophyte haplotype or environment, largely determines levels of peramine in Arizona fescue. The results suggest that whereas the endophyte is required for peramine production, the plant genotypic background in which the endophyte is embedded, rather than endophyte haplotype or environmental factors, mostly influences peramine levels within this population of Arizona fescue.