Endolithic fungi in reef-building corals (Order : Scleractinia) are common, cosmopolitan, and potentially pathogenic

Abstract

Reef-building corals appear to exist in dynamic equilibria with four principal partners: interconnected polyps of a colonial coelenterate, endosymbiotic dinoflagellate zooxanthellae residing in the host’s endoderm, endolithic algae that penetrate coral skeletons, and endolithic fungi that attack both endolithic algae and the polyps. Although reports of fungal and algal-like endoliths in corals date back almost 150 years (1) and evidence of a fossil history extends as far back as the Upper Devonian (;370 ma) (2), most attention has been paid to the structure (3), function (4, 5), and diversity (6) of the coral-zooxanthellae interactions, ignoring the endolithic members of the consortium. Recently, Le Campion-Alsumard et al. (1995) (7, 8) described an interrelationship between endolithic algae and fungi within the massive coral Porites lobata Dana 1846 (Poritidae), on Moorea island near Tahiti, French Polynesia. Fungi were also found to penetrate the most recently deposited skeletal material and to be associated with pearllike skeletal deposits formed by polyps of P. lobata in response to attack by their heterotrophic, endolithic symbionts. Here we extend these observations to the pocilloporid coral Pocillopora eydouxi and to the acroporid corals Acropora cytherea, Acropora humulis, and Montipora cf. studeri, collected at Johnston Atoll, central Pacific Ocean. Our observations suggest that direct coral-fungal interaction is widespread, not only geographically, but taxonomically as well. Thus, fungal endoliths, acting as opportunistic pathogens, may play a greater role in the ecology of coral reef systems than previously recognized. The oligophotic, siphonal green alga Ostreobium quekettii is ubiquitous in skeletons of live corals (9, 10). It is also common in other carbonate substrates, including dead shells and limestone, down to a depth of 300 m in clear waters of the Bahamas (11). Two other phototrophic organisms were reported from skeletons of live corals, the filamentous cyanobacterium Plectonema terebrans and conchocelis stages of bangiacean rhodophytes (12). Endolithic fungi in coral skeletons are equally common. They penetrate the corallum (euendolithic) and are often intermingled with endolithic algae, frequently parasitizing the latter (7). Fungi attack algal filaments by specialized hyphal branches, or haustoria, and often continue to grow inside algal filaments. Dense populations of algal and fungal endoliths have been associated with black-stained bands in specimens of P. lobata (13). Although skeletons of dead corals are bored by a variety of endolithic microorganisms, there has been no evidence that endoliths can penetrate the layer of tissue that covers living coral surfaces, leading to the conclusion that infestation by a limited number of specialized endoliths occurs early in the life of a coral, and that endolithic algae and fungi continue to grow in parallel with the accretion of the corallum (8). Most filaments of O. quekettii and endolithic fungi extend in the direction of the axes of skeletal growth. In P. lobata, borings were detected in newly deposited skeletal spines (pali), demonstrating that the endoliths are able to keep up with the rates of skeletal accretion (7, fig. 2). The pearl-like skeletal deposits are always associated with fungal attacks during the residence of polyps in actively growing calicies. The polyps encapsulate the advancing hypha into dense repair aragonite, forming a distinct skeletal structure referred to as a “cone” (7). The present study has two main objectives: to determine whether the relationship exhibited between P. lobata and endolithic fungi as described from the island of Moorea, in Received 7 July 1999; accepted 29 November 1999. *To whom correspondence should be addressed. E-mail: cbentis @bu.edu Reference: Biol. Bull. 198: 254–260. (April 2000)