Tropical Ecosystems: Structure, Diversity and Human Welfare.Proceedings of the Intematlonai Conference on Tropical EcosystemsK. N: Ganeshalah, R. Uma Shaanker and K. S. Bawa (ads)Published by Oxford-ISH, New Deihl. 2001. pp.739-743.

Fungal endophytes in neotropical trees:Abundance, diversity, and ecological

implications

A. Elizabeth Arnold

Department of Ecology and Evolutionary Biology, 1041 E. Lowell, BSW 310,University of Arizona, Tucson, AZ 85721, USA

e-mail: amold@ccit.arizona.edu

Although fungi comprise only ca. 72,000described species, many authorsrank them among hyperdiversetaxa, comparing fungi with nematodes, mitesand insects in terms of global species richness (Cqlwell and Coddington,1994;Hawksworth et al., 1995).However, the true scale of fungal diversity isa matter of open debate, with estimates ranging from hundreds of thousandsof species (Aptroot, 1997)to values as high as 1.5 million species (Hawks-worth, 1991)and greater (e.g., Cannon, 1997).Recent assessments of fungalbiodiversity in tropical forests, where fungal richness is ~ought to begreatest, have begun to refine richness est:iri1ates,and in the last decade havelent credence to higher values (e.g., Frohlich and Hyde, 1999).However,severa1 authors argue that such extrapolations either represent dramaticoverestimatesof fungal diversity (May, 1991),or are suspicious due to a lackof relevant data (Hammond, 1992).Resolution of the fungal diversity debatebears directly upon such human enterprises as medicine, agriculture, andindustry, and upon our understanding of the structure and function ofterrestrial ecosystems (e.g., Fox, 1993;Lodge, 1997).Mycologistsagree thatfurther exploration of tropical fungi is critical for understanding both thescaleof fungal diversity (Hawksworth,1993)and its ecologicalimplications.

Keywords: Abundance, Ba.rro Colorado Island, biodiversity, fungal endophytes, similarityindices.

739

Tropical flmgi are traditionally understudied (Cannon, 1997), and theirtaxonomic placement has been confounded, at times, by misidentificationaccording to temperate mycota (Lodge, pers. comm.). Extensive trainingneeded for sampling complex tropical habitats (Shivas and Hyde, 1997),inconsistenciesin quantitative methods among studies (Arnold et al.,2000),npaucity of newly trained systematists specializing in tropical mycology(Rossman,1997),and traditional difficultiesin delineating species boundarlo,(Hawksworth, 1991)further inhibit both the accuracy and precision of fungaldiversity estimates. Especially challenging to enumerate are tropicAlmicrofungi: those cryptic flmgi that produce only microscopic fruitingbodies. Rossman (1997) has suggested that as many as 700,000-900,000microfungal species may exist, the majority of which are likely to occur Intropical regions. Data from several authors (e.g., Lodge et al., 1996;Frohlichand Hyde, 1999)have suggested that many of these microfungi may occur II'flmgal endophytes inhabiting tropical woody angiosperms; however, studiesof endophytic flmgi in tropical forests are yet in their infancy.

Endophytes are microorganisms that colonize and cause asymptomaticinfections in healthy plant tissues (Wilson, 1995). Fungal endophytes arcconsidered to be at least as ubiquitous as mycorrhizal associations amongtemperate-zone plants (Carroll, 1988), having been found in algae(Hawksworth, 1988),mosses (Schulzet al.,1993),ferns (Fisher,1996),conifers(Legault et al., 1989),and both monocotyledonous (Clay, 1988)and dicotyle-donous angiosperms (Petrini et al., 1982).Saikkonen et al. (1998)noted thatindividual plants in the temperate zone may harbor dozens of endophytespecies, and several recent, quantitative surveys of tropical angiosperms havedocumented remarkable endophyte richness in individual leaves and treet(e.g., Lodge et al., 1996;Frohlich and Hyde, 1999;Arnold et al., 2000).Thesesurveys suggest that tropical endophytes may contribute substantially toflmgal diversity, but both an ecological approach to quantitative sampling,and consistency among methods at a variety of scales,have been lacking.Toaddress these issues, I have explored patterns of endophyte abundanco,diversity, host preference, and spatial heterogeneity in the lowland, moistforest of Barro Colorado Island, Panama (BCI).

In surveys of 24 common host species representing 23 families of woodyangiosperms, I have shown that endophytes are highly abundant in healthy,mature leaves of understorey saplings, where they typically colonize > 80%of leaf tissue. Unsuccessful attempts to isolate endophytic flmgi from bothseeds and seedlings grown under sterile conditions, and observations thatendophyte-free seedlings quickly accumulate endophytes under fieldconditions, strongly suggest that endophytic fungi assoQated with tropicaJtrees are horizontally transmitted. A consistent and significant increase Inendophyte abundance with increasing leaf age further corroborates thJapattern and appears to result from differences in leaf exposure time, ratherthan from differences in leaf chemistry between young and mature leavCl,

740

Similarly, high endophyte abundance among forest-grown seedlings relativeto clearing-grown seedlings appears to be a common pattern, and can beexplained by differences in inoculum level, rather than by differences in leafchemistry between exposed and shaded hosts. Thesedata, and emerging datafrom other tropical sites (e.g., Puerto Rico: Lodge et al., 1996;Gamboa andBayman, in press; Borneo and Australia: Frohlich and Hyde, 1999)suggestthat flmgal endophytes may represent a ubiquitous, cryptic, and ecologicallyinteresting component of tropical forests.

Similarly, endophytic flmgi appear to be consistently diverse among hostspecies at BCI:mature leaves typically contain> 10 morphospecies of fungalendophytes, and individual host species may harbor hundreds of endophytemorphospecies. However, endophyte assemblages differ with respect to bothsampling site and host taxon: using frequency-based similarity indices andrandomization tests based on isolation frequencies, I have found strongevidence for both spatial heterogeneity and host preference amongnonsingleton morphospecies. In ongoing work, I am exploring the scale ofspatial heterogeneity within the forest at BCI, and am empirically assessingendophytes for a mechanism of host preference, concentrating on host-specific growth rates. Concurrently, I am seeking to reconcile endophytemorphospecies with genetically delimited species using sequence data fromthe nrDNA internal transcribed spacer regions 1 and 2 (ITS1, ITS2), andintrons associated with translation elongation factor 1a (EF-1a).Preliminarydata suggest general concordance between ITS-delimited species andmorphospecies, and demonstrate that endophytes are diverse at hightaxonomic levels within the Ascomycota; however, further sampling isneeded to assess the number of genetically distinct units recovered fromindividual leaves, trees, sites, and host species.

Studies of endophyte abundance, diversity, spatial heterogeneity, and hostpreference are critical for assessing the contribution of tropical endophytes tofungal diversity estimates. Based on evidence from preliminary surveys ofendophytes in tropical forests, several authors have suggested that the mostcommonly accepted estimate for flmgal diversity (1;5 million species:Hawksworth, 1991) represents a marked underestimate (e.g., Dreyfuss andChapela, 1994;Frohlich and Hyde, 1999;Arnold et al., 2000).Similarly, dataregarding endophyte abundance and diversity have important implicationsfor understanding plant-fungus interactions in tropical forests, andaccumulating data present several questions of ecological importance: forexample, given that endophytes are obligately heterotrophic, and that plantsin the shaded tropical understorey are often carbon-limited, why shouldtropical trees harbor such a high abundance of endophytic flmgi within theirleaves? Similarly, given that fungal diseases represent a significant cause ofplant mortality in lowland tropical forests, that many tropical trees arerelatively well defended against a variety of fungal pathogens, and that many,fungal endophytes are congeneric with a taxonomically diverse array of

741

- - --- - - ---.

References

Hammond, P. M Species inventory. in Glomi Diversity: Status of the Earth's Living Resources (ed.Groombridge, B.) 17-39 (Chapman &<Hall, London, UK, 1992).

Hawksworth, D. L. The variety of fungal-algal symbioses, their evolutionary significance, andthe nature of lichens. Bot.T. Linn. Soc. 96, 3-20 (1988).

Hawksworth, D. L. The fungal dimension of biodiversity: magnitude, significance, andconservation. Mycol. Res. 95, 641~55 (1991).

Hawksworth, D. L. The tropical fungal biota: Census, pertinence, prophylaxis, and prognosis. in

AspectsofTropicalMycology(eds Isaac, S., Frankland, J. c., Watling, R. &<Whalley, A. J. S.)265-293 (Cambridge University Press, Cambridge, UK, 1993).

Hawksworth, D. L., Kirk, B. c., Sutton, B. C. &<Pegler, D. N. Ainsworth & BisUy's Dictionary of the

Fungi. 8th edition. 359 pp. (CAB International, Wallingford, UK, 1995).

Legault, D., Dessureault, M &<Laflamme, G. Mycoflore des aiguilles de Pinus mnksiana et Pinusresinosa 1.Champignons endophytes. Can.T. Bot. 61, 2052-2060 (1989).

Lodge, D. J., Fisher, P. J. &<Sutton, B. C. Endophytic fungi of Manilkara bidentata leaves in PuertoRico. Mycologia 88, 733-738 (1996).

Lodge, D. J. Factors related to diversity of decomposer fungi in tropical forests. Biodiver. Conserv.6, 681~ (1997).

May, R. M A fondness for fungi. Nature 352, 475-476 (1991).Petrini, 0., Stone, J. &<Carroll, F. E. Endophytic fungi in evergreen shrubs in western Oregon: A

preliminary study. Can.T. Bot. 60, 789-796 (1982).Rossman, A. Y. Biodiversity of tropical microfungi: An overview. in Biodiversity of Tropical

Microfungi (ed. Hyde, K. D.) 1-10 (Hong Kong University Press, Hong Kong SAR, China,1997).

Saikkonen, K., Faeth, S. H., Helander, M. &<Sullivan, T. J. Fungal endophytes: A continuum ofinteractions with host plants. Annu. Rev.£COl.Syst.29, 319-343 (1998).

Schulz, B., Wanke, U., Draeger, S. &<Aust, H.-J. Endophytes from herbaceous plants and shrubs:Effectiveness of surface sterilization methods. Mycol. Res. 91, 1447-1450 (1993).

Shivas, R. G. &<Hyde, K. D. Biodiversity of plant pathogenic fungi in the tropics. in Biodiversity ofTropicalMicrofungi(ed. Hyde, K.D.) 47-56 (Hong Kong University Press,Hong Kong, SAR,China, 1997).

Wilson, D. Endophyte - the evolution of a term, and clarification of its use and definition. Oikas13,274-276 (1995).

pathogens, why do tropical trees harbor such a diversity of fungalendophytes? Using these questions to guide my work, I am assessing thecosts and benefits of endophyte infection, and am testing the hypothesis thatfungal endophytes may influence host resistance to invasive pathogens.

The ecological importance of fungal endophytes in tropical trees is not yetknown, but endophytes appear to be both ubiquitous and highly diverse intropical forests.Further explorationof tropical endophytes will help to clarifythe fungal diversity debate, and will likely lend support to higher estimatesof global fungal diversity. At the same time, the study of tropical endophytesseems promising as a means to enrich our understanding of' plant-fungusinteractions in tropical forests, tropicalbiodiversity,and tropicalecology.

Acknowledgements

I sincerely thank L. McDade for valuable insights and guidance; E. A. Herre for furtheringtropical endophyte research; G. Gilbert for training and advice; Z. Maynard and J. Barnard fortechnical assistance; and the Smithsonian Tropical Research Institute for logistical support.Funding support from an NSF Graduate Fellowship, NSF-DEB 9902346 (to L. McDade andABA), the Research Training Group in Biological Diversification at the University of Arizona(NSF-DIR-9113362, BIR-9602246), the American Cocoa Research Institute, the SmithsonianTropical Research Institute, and the Hoshaw family are gratefully acknowledged. For travelsupport, I thank the Department of Ecology and Evolutionary Biology at the University ofArizona and the Association for Tropical Biology. Special thanks are given to P. D. Coley and T.A. Kursar for an introduction to tropical ecology.

Aptroot, A. Species diversity in tropical rainforest ascomycetes: Lichenized vs non-lichenized;follicolous vs corticolous. Abstr. Bot. 21, 37-44 (1997).

Arnold, A. E., Maynard, Z., Gilbert, G. S., Coley, P. D. &<Kursar, T. A. Are tropical fungalendophytes hyperdiverse? Ecol.Lett. 3, 267-274 (2000).

Cannon, P. Diversity of the Phyllachoraceae with special reference to the tropics. in Biodiversity ofTropical Microfungi (ed. Hyde, K. D.) 255-218 (Hong Kong University Press, Hong KongSAR, China, 1997).

Carroll, G. Fungal endophytes in stems and leaves: from latent pathogen to mutualisticsymbiont. Ecology 69, 2-9 (1988).

Clay, K. Fungal endophytes of grasses: A defensive mutualism between plants and fungi. Ecology69, 10-16 (1988).

Colwell, R. K. &<Coddington, J. A. Estimating terrestrial biodiversity through extrapolation.

Philos. Trans. Roy. Soc. London, B 345, 101-118 (1994).

Dreyfuss, M M &<Chape1a, I. H. Potential of fungi in the discovery of novel, low-molecularweight pharmaceuticals. in The Discovery of Natural Products with Therapeutic Potential (ed.Gullo, V. P.) 49-80 (Butterworth-Heinemann, London, UK, 1994).

Fisher, P. J. Survival and spread of the endophyte Stagonospora pteridiicola in Pteridium aquilinum,

other ferns and some flowering plants. New Phytol. 132, 119-122 (1996).

Fro~ch, J. &<Hyde, K. D. Biodiversity of palm fungi in the tropics: Are global fungal diversityestimates realistic? Biodiver. Conserv. 8, 97Z-1004 (1999).

Fox, F. M Tropical fungi: their commercial potential in Aspects of Tropical Mycology (eds Isaac, S.,

Frankland, J. C., Watling, R. &< Whalley, A. J .S.) 253-264 (Cambridge University Press,

Cambridge, UK, 1993).

Gamboa, M A. &<Bayman, P. Communities of endophytic fungi in leaves of a tropical timber

tree (Guarea guidonia: Meliaceae) (In press).

742743