Andrew Kane is an Associate Research Scientist at the Aquatic Pathobiology Center,
VA-MD Regional College of Veterinary Medicine, College Park campus (UMCP).

David Oldach is a clinician and researcher at University of Maryland School of Medicine, Department of Medicine
Baltimore, MD.

Renate Reimschuessel is a researcher at the U.S. Food and Drug Administration, Center for Veterinary Medicine,
Laural, MD.

Correspondence may be addressed to:

A.S. Kane
University of Maryland
Department of Veterinary Medicine
8075 Greenmead Drive
College Park, MD 20742
akane@umaryland.edu

ABSTRACT

Ulcerative lesions and mass mortalities of Atlantic estuarine fish, particularly menhaden (Brevoortia tyrannus), have been associated with exposure to Pfiesteria-like dinoflagellates and their toxins. We collected fish from the Chicamacomico River, Maryland, and observed solitary ulcerative lesions on the majority of menhaden sampled. One striped bass (Morone saxatilis) had an area of reddening around the base of the dorsal fin. Bluegill (Lepomis macrochirus), channel catfish (Ictalurus punctatus), yellow perch (Perca flavescens) and carp (Cyprinus carpio) were externally non-remarkable. Histologically, ulcerative menhaden lesions demonstrated a marked chronic inflammatory infiltrate in large areas of exposed necrotic muscle. The ulcers contained granulomata with fungal hyphae in the necrotic tissue. Gram negative rod-shaped bacteria were also observed in the lesions, a common finding in ulcers of aquatic organisms. Our data suggest that "typical" ulcerative lesions observed on fish from areas of Pfiesteria-like dinoflagellate blooms are reflective of dermatosis which may be related to a variety of individual or combined environmental stressors. Exposure to dinoflagellate toxin(s) potentially represents one such stressor. The role of Pfiesteria-like dinoflagellate toxin in fish primary lesion development is currently under investigation.

BACKGROUND

Fish, like mammals, are susceptible to a variety of environmental stressors which may directly cause, or indirectly predispose, them to develop different types of lesions. These stressors may be biological (bacterial, viral, fungal, parasitic), chemical (pollutants, toxins, suboptimal water quality, hormonal changes due to photoperiod or breeding) and/or physical (rapid water temperature change, trauma). Disease outbreaks and mortality occur naturally in all wild populations. What causes concern, however, is when huge numbers (i.e., hundreds of thousands) of fish exhibiting lesions, morbidity, or death occur in a relatively short time period. Such is the case for fish kills co-occurring during toxic blooms of Pfiesteria-like dinoflagellates. A toxic dinoflagellate (or algal) bloom is a notable increase in cell numbers (density) and predominance of at least one harmful dinoflagellate species. Since Pfiesteria-like dinoflagellates do not always produce toxin (1), the mere presence of these dinoflagellates may not necessarily be harmful.

Pfiesteria-like dinoflagellates have been observed in tributaries of the lower Chesapeake Bay and the Tar-Pamlico River Estuary (the two largest estuaries in North America), as and far south as St. Johns River and Pensacola Bay in Florida (2,3,4). Table 1 lists a recent chronology of toxic dinoflagellate observations and associated fish health problems in Maryland. The types of lesions noted on many fish sampled from these waters are sizable, deep ulcers.

The organism implicated with these lesions and the fish kills is a dinoflagellate, Pfiesteria piscicida. P. piscicida is a newly identified genus and species in a newly recognized dinoflagellate family, Pfiesteriaceae (5), named after the late Dr. Lois A. Pfiester. Dr. Pfiester was an experimental and field phycologist who described and unraveled some of the many complex sexual cycles of freshwater dinoflagellates, e.g., Pfiester and Popovsky (6). The specific epithet, piscicida (from Latin), means "fish killer." Close relatives of Pfiesteria, yet to be taxonomically confirmed, have also been found in the Chesapeake Bay and other areas, and have also been implicated with fish lesions and mortalities; hence the term "Pfiesteria-like" dinoflagellates.

A short background on Pfiesteria: this organism has a complex life cycle consisting of amoeboid, flagellated, and encysted forms. Further, there is a high degree of pleomorphism within each of these forms; Burkholder and Glasgow (1) suggest over 20 distinct stages. P. piscicida also varies greatly in size: flagellated forms range from 5-18 µm; amoeboid forms range from <5 to 250 µm, and cysts have been observed from 10 to 33 µm (5). A variety of environmental cues are responsible for the dinoflagellate's growth, sexual reproduction, encystation, and toxin production. The presence of live fish stimulates exotoxin (1). Recent evidence suggests that there are multiple toxin forms (lipophilic and hydrophilic), each having different effects on fish and possibly humans. Pfiesteria is known to feed on algal prey as well as dead or dying fish (1); a heterotrophic feeding ecology. However, Pfiesteria is capable of ingesting algal cells, emptying their contents and "stealing" their chloroplast (the light sensitive, energy-producing organelle), a phenomenon termed cleptochloroplasty (5). Pfiesteria is a relatively rugged dinoflagellate covered with a series of armored plates. The plate configuration is used for electron microscopic species-level identification.

Although there are many ichthyotoxic dinoflagellates, readers may be most familiar with Gymnodinium breve, a toxic "red tide" dinoflagellate which produces brevetoxins. Brevetoxin is the causative agent of neurotoxic shellfish poisoning, which effects fish, shellfish, birds, marine mammals and humans (7). In contrast to P. piscicida, G. breve is a naked dinoflagellate, without an array of external armored plates. G. breve produces an endotoxin, which is released upon rupture of its soft outer membrane. Rupture can occur at the water surface due to wind action (which is why toxin aerosols affect humans), or when the fragile cells pass through the processes of fish gills, whereupon the toxin diffuses through the gill membrane and causes lethality if the cell concentration is sufficiently high (8). Fish mortality caused by acute brevetoxin exposure occurs without pathologic lesions (9 as cited in 8).

METHODS

Water temperature was measured using a hand held thermometer and salinity was measured using a refractometer. Fish were collected in September 1997 from the Chicamacomico River, Maryland, using a 4 meter diameter monofilament cast net. Repeated net casts were made over a period of several hours. Fish were photographed and subsequently sacrificed with buffered MS222. Skin scrapes and gill biopsies were performed and examined in the field from representative specimens. Tissue samples were preserved in 10% neutral buffered formalin and submitted for routine histological processing (10).

RESULTS

Water quality variables evaluated at the collection site included temperature =14° C and salinity = 3 parts per thousand (ppt); the water color was bluish-green with moderate turbidity. Fish species collected during the Chicamacomico River sampling included striped bass (Morone saxatilis), yellow perch (Perca flavescens), Atlantic menhaden (Brevoortia tyrannus), channel catfish (Ictalurus punctatus), carp (Cyprinus carpio), and bluegill (Lepomis macrochirus). Additional samples were obtained from Kings Creek and the Pocomoke River. Skin scrapes and gill biopsies for parasites from representative specimens were negative. No gross lesions were noted on the yellow perch (n=3), channel catfish (n=1), carp (n=1) or bluegill (n=5). The striped bass specimen (n=1) had an annular reddening on the dorsum, circumscribing the base of the dorsal fin. External lesions were present on the majority of the menhaden (n=22). Several of the menhaden were weak or moribund at the time of capture. Solitary lesions on menhaden were noted on the ventrum, around the vent, on the side of the abdomen, on the dorsum, or on the caudal peduncle (Figure 1). The gross lesions consisted of focal erthythema (Figure 1A), well-circumscribed ulcers with necrotic centers (Figure 1B-E), or round raised, friable red nodules (Figure 1F). There was a similar presentation of ulcerative lesions on other species examined (Figure 2).

Microscopic examination of these ulcerative lesions demonstrated a marked chronic inflammatory infiltrate in large areas of exposed necrotic muscle. At the margin of the ulcers, inflammation extended into scales and dermis, and along facial planes into the underlying musculature. In some fish the lesions extended to the backbone and into the peritoneal cavity. In these cases there was an extensive peritonitis surrounding the abdominal organs. The majority of fish had granulomata present in the necrotic tissue, and fungal hyphae were present within the granulomata (Figure 3). Gram negative rod-shaped bacteria were also observed in the lesions, a common finding in ulcers of aquatic organisms.

Discussion

The gross lesions observed on menhaden collected from the Chicamacomico River are consistent with lesions previously described from fish collected from other environmental kills (11,12). Although there was a prevalence of lesions near the anus, the location of lesions on the fish was variable. The extent of the lesions ranged from an area of reddening to deep penetrating ulcers (Figure 1). The annular lesion at the base of the dorsal fin noted on the striped bass in the present study appeared similar to lesions observed on Pfiesteria-exposed hybrid striped bass (M. saxatilis x M. chrysops) as described by Burkholder et al. (3).

Three potentially toxic Pfiesteria-like dinoflagellates (Pfiesteria piscicida, Gyrodinium galatheanum, Cryptoperidiniopsis spp.) have been identified from the Chicamacomico River in 1997 (13). Limited laboratory experiments with striped bass exposed to a sublethal concentration of Pfiesteria toxin demonstrated chronic ulcer initiation (14). So far, however, there has been no field validation to definitively demonstrate ulcerative lesion initiation due to exposure to toxic dinoflagellates.

It is interesting to note the diversity of fish taken during our low-salinity Chicamacomico sampling efforts. In addition to relatively salinity-tolerant menhaden, striped bass and channel catfish, we also sampled freshwater species: yellow perch, carp and bluegill. The portion of the Chicamacomico River which was sampled is tidal, with a salinity ranging from 0.5 - 9.5 ppt (15). This explains the species diversity which was observed. This is also the range of salinity which tends to accompany Pfiesteria-related fish mortality and lesions as seen in Maryland. Burkholder et al. (3) observed that while Pfiesteria tolerates salinities between 0 - 35 ppt, an optimal salinity for Pfiesteria growth and toxin production occurs at 15 ppt.

Ulcerative lesions in wild fish are often a primary or secondary consequence of bacterial, viral or fungal pathogens, or parasites. Examples of biological agents which are known to cause or be associated with primary or secondary ulcerative or hemorrhagic lesions, morbidity and mortality, other than toxic dinoflagellates, include: Hemophilus piscum, Aeromonas hydrophila, A. salmonicida, Pseudomonas fluorescens, Flexibacter spp., Vibrio anguillarum, Edwardsiella tarda (bacteria); Ichthyosporidium hoferi, Aphanomyces spp., Saprolegnia spp. (fungi); Glugea (microsporidian protozoan), Henneguya spp. and Myxobolus spp. (myxosporidean protozoans), Argulus spp. and Lernea elegans (arthropod parasites), and Petromyzon spp. (vertebrate parasites) (16,17,18). Some examples of lesions from Chesapeake Bay fish, not associated with toxic dinoflagellates, are shown in Figure 4.

Although several opportunistic pathogenic bacterial species have been isolated from Chesapeake Bay menhaden lesions, fungal involvement appears universal. In fact, these ulcerative lesions have been associated with the term "ulcerative mycosis" (12,14). Commonly, these ulcers involve at least two genera of fungi: Aphanomyces and Saprolegnia. (12,14). Noga and Dykstra (14) hypothesized that lesion progression may begin as a) small flat red areas which lead to scale and skin loss, exposing muscle, or b) raised masses with a necrotic core which then sloughs off, leaving a crater-shaped cavity. Osmotic stress due to these open lesions contributes to loss of ionic homeostasis and death (19).

Lesions associated with ulcerative lesions are chronic, and are not induced at the time of an acute fish kill. Although the pathogenesis of these ulcerative lesions remains unclear, it is likely that some type of initial injury to the epidermis fosters subsequent invasion by opportunistic pathogens (as described above). Such injury may be caused by trauma, pathogens, parasites, or possibly sublethal exposure to dinoflagellate toxin(s).

Based on the current literature, Atlantic menhaden appear to be most susceptible to the exposure effects of Pfiesteria-like organisms (and/or its toxins). However, other fish species have been shown to be affected, with or without gross pathology (Table 2). In general, menhaden are highly sensitive to most types of environmental stressors (biological, chemical and physical). Species, such as killifish (Fundulus heteroclitus) and hogchoker (Trinectes maculatus), on the other hand, tend to be relatively resistant to stress. The biological reason(s) for these consistent disparities in susceptibility to stress remains unclear, but based on their mass mortalities, menhaden are currently the best biological indicator of toxic Pfiesteria-like dinoflagellate blooms and environmental stress. The presence of toxic dinoflagellates, however, must be definitely confirmed by electron microscopic species-level identification and light microscopic cell density estimates (3,5,20).

To date, there is no evidence that live fish with lesions harbor Pfiesteria-like organisms or their toxin(s). Further, there are no current data to support that these fish lesions are a source of infection for human dermal pathologies (as described by Shoemaker (21) and Lowitt (22)). From a more general perspective, however, whole (healthy) fish have been known to cause acute contact dermatitis in humans and asthma-like symptoms (23,24). It appears that some human hypersensitivity reactions may be related to a glycoprotein component of the outer protective mucus coat covering fish epithelium (25,26).

In general, observations of ulcerative fish lesions in many different waterways worldwide, has appeared to increase over the last half century (16). Although multiple etiologies are likely involved, increased incidence of fish mortalities and lesions is indicative of an increase in environmental stress. These environmental stressors include pollution, non-point source contamination, and nutrient enrichment, all of which foster accelerated eutrophication of aquatic systems. These issues are obviously complex, and multidisciplinary efforts are needed to better understand the elaborate relationships between the biotic and abiotic components of our aquatic systems. Additional information about fish health may be obtained from the "Fish Health in Chesapeake Bay" worldwide web site (http://www.mdsg.umd.edu/fish-health).

ACKNOWLEDGMENTS

The authors thank the U.S. Army Garrison Aberdeen Proving Ground, Installation Restoration Program for support in portions of this study, Maryland Department of Natural Resources for their assistance and cooperation with field sampling, and Maryland Sea Grant for their continued support. We also thank Dr. JH Landsberg, Florida Department of Environmental Protection, for constructive review this manuscript.

REFERENCES

1. Burkholder JM, Glasgow HB. Trophic controls on stage transformations of a toxic ambush-predator dinoflagellate. J Euk Microbiol 1997;44(3):200-205.

2. Lewitus AJ, Jesien RV, Kana TM, Burkholder JM, Glasgow HB, May E. Discovery of the "phantom" dinoflagellate in Chesapeake Bay. Estuaries 1995;18(2):373-378.

3. Burkholder JM, Glasgow HB, Hobbs CW. Fish kills linked to a toxic ambush-predator dinoflagellate: distribution and environmental conditions. Mar Ecol Prog Ser 1995;124:43-61.

4. Matuszak DL, Sanders M, Taylor JL, Wasserman MP. Toxic Pfiesteria and human health. MMJ 1997; 46(10):515-520.

5. Steidinger KA, Burkholder JM, Glasgow HB, Hobbs CW, Garrett JK, Truby EW, Noga EJ, Smith SA. Pfiesteria Piscicida gen. Et sp. Nov. (Pfiesteriaceae fam. Nov.), a new toxic dinoflagellate with a complex life cycle and behavior. J Phycol 1996;32:157-164.

6. Pfiester LA, Popovsky J. Parasitic, amoeboid dinoflagellates. Nature 1979;279:421-424.

7. Steidinger, KA. 1993. Some taxonomic and biologic aspects of toxic dinoflagellates. Pages 1-28 in Algal Toxins in Seafood and Drinking Water. I.R. Falconer, editor. Academic Press, London.

8. WHO 1984. Aquatic (Marine and Freshwater) Biotoxins. Environmental Health Criteria 37. World Health Organization, Geneva.

9. Abbott, BC, Siger, A, Spiegelstein, M. Toxins from the blooms of Gymnodinium breve. In: LoCierco, VR, editor. Proceedings of the First International Conference on Toxic Dinoflagellate Blooms, Wakefield, MA, Massachusetts Science and Technology Foundation, pp. 335-363.

10. Profet, EB, Mills, B, Arrington, JB,. Sobin, LH. 1992 Laboratory Methods in Histotechnology. Armed Forces Institute of Pathology, Washington, D.C., Published by the American Registry of Pathology, Washington, D.C.

11. Ahrenholz DW, Guthrie JF, Clayton RM. Observations of Ulcerative Mycosis Infections on Atlantic Menhaden (Brevoortia tyrannus). NOAA Technical Memorandum, NMFS-SEFC-196 June 1987.

12. Dykstra MJ, Levine JF, Noga EJ, Hawkins JH, Gerdes P, Hargis WJ, Grier HJ, Te Strake D. Ulcerative mycosis: a serious menhaden disease of the southeastern coastal fisheries of the United States. J of Fish Diseases 1989;12:175-178.

13. JH Landsberg and KA Steidinger, Florida Department of Environmental Protection, personal communication.

14. Noga EJ, Dykstra MJ. Oomycete fungi associated with ulcerative mycosis in menhaden, Brevoortia tyrannus (Latrobe). J of Fish Diseases 1986;9:47-53.

15. Speir, H. 1998. Maryland Department of Natural Resources, personal communication.

16. Sinderman, CJ. Epizootic ulcerative syndromes in coastal/estuarine fish. NOAA Technical Memorandum NMFS-F/NEC-54, June 1988.

17. Ferguson, HW. Systematic Pathology of Fish. Iowa State University Press, Ames, IO.

18. Roberts, RJ. Fish Pathology. 2nd. Ed. Bailliere Tindall. London, England, 1989.

19. Noga, EJ, Khoo, L, Stevens, JB, Fan, Z, Burkholder, JM. Novel toxic dinoflagellate causes epidemic disease in estuarine fish. Mar Pollut Bull 1996;32(2):219-224.

20. MMWR (Morbidity and Mortality Weekly Report). Results of the public health response to Pfiesteria Workshop- Atlanta Georgia, September 29-30, 1997.

21. Shoemaker RC. Diagnosis of Pfiesteria-human illness syndrome. MMJ 1997;46(10):521523.

22. Lowett M, this issue of MMJ.

23. Alonso MD, Davila I, Conde Salazar L, Cuevas M, Martin JA, Guimaraens MD, Losada E. Occupational protein contact dermatitis from herring. Allergy 1993;48:349-352.

24. Dominguez C, Ojeda I, Crespo JF, Pascual C, Ojeda A, Martin-Esteban M. Allergic reactions following skin contact with fish. Allergy and Asthma Proc 1996;17:83-87.

25. Chiou, FY, Tschen, JA. Fish scale-induced dermatitis. J Am Acad Dermatol 1993;28(6):962-965.

26. Warpinski JR, Folgert J, Voss M, Bush RK. Fish surface mucin hypersensitivity. J of Wilderness Med 1993;4:261-269.

27. Terlizzi D, Fish kills and harmful algal blooms. Maryland Aquafarmer 1997, Fall:1-2.

28. Breitburg, D. 1997. Academy of Natural Sciences Research Center, personal communication.

29. Leffler, M. 1998. Maryland Sea Grant, personal communication.

------------------------------------------------------------------

Table 1. Chronology of toxic dinoflagellate observations and associated fish health problems in Maryland. Data compiled by Maryland Sea Grant (29).

1992 (summer) Pfiesteria piscicida identified from Jenkins Creek on Choptank River (2).

1994 Pfiesteria associated with laboratory fish mortalities at Benedict Estuarine Research Laboratory on the Patuxent River (28).

1996 Pfiesteria associated with laboratory fish mortalities at Academy of Natural Sciences Research Center on St. Leonard's Creek off the Patuxent River (28).

1996 (August) Pfiesteria piscicida and other dinoflagellate species associated with striped bass mortalities in culture ponds at HyRock Farms on Manokin River off Kings Creek (27).

1997 (summer/fall) Pfiesteria piscicida identified from Pocomoke River / fish kills occurred.

1997 Laboratory fish mortalities at Academy of Natural Sciences Research Center on St. Leonard's Creek off the Patuxent River (28).

1997 (August) Striped bass mortalities in culture ponds at HyRock Farms on Manokin River off Kings Creek - Pfiesteria piscicida not identified while other dinoflagellates were.

1997 (September) Presence of Pfiesteria-like dinoflagellates confirmed on Manokin River off Kings Creek / fish kills occurred.

1997 (September) Presence of Pfiesteria-like dinoflagellates confirmed on Chicamacomico River / fish lesions and mortalities.

------------------------------------------------------

Table 2. Chesapeake Bay region fish species known to be affected by exposure to Pfiesteria-like dinoflagellates or their toxin(s). Compiled from Lewitus et al. (2), Burkholder et al. (3), Noga et al. (19) and Terlizzi (27).

American eel (Anguilla rostrata)
Atlantic croaker (Micropogonias undulatus)
Atlantic menhaden (Brevoortia tyrannus)
black grouper (Mycteroperca bonaci)
channel catfish (Ictalurus punctatus)
clownfish (Amphiprion percula)
goldfish (Carassius auratus)
guppy (Poecilia reticulata)
hogchoker (Trinectes maculatus)
killifish (mummichog, Fundulus heteroclitus)
largemouth bass (Micropterus salmoides)
mosquitofish (Gambusia affinis)
naked goby (Gobiosoma bosc)
pinfish (Lagodon rhomboides)
red drum (Sciaenops ocellatus)
redear sunfish (Lepomis microlophys)
sheepshead (Archosargus probatocephalus)
Southern founder (Paralichthys lethostigma)
spot (Leiostomus xanthurus)
spotted sea trout (Cynoscion nebulosus)
striped bass (Morone saxatilis)
striped bass hybrids (Morone saxatilis x M. chrysops)
striped mullet (Mugil cephalus)
white perch (Morone americana)

--------------------------------------------------------

Figure 1. External lesions observed on menhaden sampled from the Chicamacomico River suspected to be associated with toxic dinoflagellates. (A) flat, reddened lesion centrally located on the side of the fish, (B through E) chronic ulcerative lesions with necrotic centers which penetrate the epidermis, dermis and musculature, at the anus, mid-abdomen, trunk, and dorsal fin area, respectively, and (F) a round raised, friable red nodule at the base of the caudal fin.

----------------------------------------------

Figure 2. External lesions observed on (A) spot (Leiostomus xanthurus), (B) spotted sea trout (Cynoscion nebulosus) and (C and D) flounder (Paralichthys spp.) suspected to be associated with toxic dinoflagellates. The summer flounder lesion penetrates from the ventral surface of the fish (D) through the dorsal surface (C).

-----------------------------------------

Figure 3. (A, top) Photo micrograph of granulomata with fungel hyphae in section of necrotic muscle (fungi not evident), H&E 200x; (B, bottom) Section of muscle stained with GMS in order to visualize fungi (fungal hyphae are the dark stained thin "ribbons").

---------------------------------

Figure 4. Examples of fish lesions on largemouth bass not associated with exposure to toxic dinoflagellates. (A and B) skeletal deformities: lordosis and brachygnathia, respectively, which may have been genetic or caused by nutritional deficiency or parasites, (C) two relatively shallow, small ulcers beneath the left pectoral fin, and (D) parasitism (leech) on the anal fin and erosion at the posterior edge of the caudal fin. Small skin abrasions, erosions or ulcers are not uncommon in wild fish and are often non-specific in origin. These types of lesions may be associated with changes in water temperature, spawning, parasitism, trauma, or poor water quality. These fish were collected during a fish health survey in northern Chesapeake Bay.