| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 U.S. Geological Survey, National Wildlife Health Center, 6006 Schroeder Road, Madison, Wisconsin 53711, USA
2 Wildlife Health Associates, Inc., PO Box 109, Dillon, Montana 59725, USA
3 Southeastern Cooperative Wildlife Disease Study, College of Veterinary Medicine, University of Georgia, Athens, Georgia 30602, USA
4 U.S. Fish and Wildlife Service, Ecological Services, Raleigh, North Carolina 27636-3726, USA
5 Corresponding author (email: tonie_rocke{at}usgs.gov)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key words: American coot, avian vacuolar feeding trials, Fulcia Americana, Hydrilla verticillata, myelinopathy.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Coots clinically affected with AVM exhibit profound motor dysfunction and in-coordination (Thomas et al., 1998; Larsen et al., 2002); they are reluctant to fly, are ataxic on land, and might swim in circles or on their backs. Histologically, the disease is characterized by diffuse, spongy degeneration throughout the white matter of the central nervous system of affected birds, but not all birds with brain lesions have evident clinical signs (Rocke et al., 2002; Rocke, unpubl. data). A diagnosis of AVM requires the observation of generalized white matter vacuolation which is most prominent in the optic tectum and occurs in other regions of the brain and spinal cord when viewed with light microscopy (Thomas et al., 1998), whether the animal has apparent clinical signs or not.
Despite extensive diagnostic and field investigations, the causative agent of AVM is still unknown. Recently, the brain lesion characteristic of AVM (but not clinical signs of disease) was experimentally reproduced in red-tailed hawks (Buteo jamaicensis) following ingestion of tissues from AVM-affected coots, providing evidence that eagles contract the disease by consuming affected coots or ducks (Fischer et al., 2003). A subsequent experiment in which chickens were fed tissues from affected coots demonstrated that the causative agent was present in gastrointestinal (GI) contents, but not in brain, fat, kidney, liver, or muscle (Lewis-Weis et al., 2004). It has been hypothesized that the route of exposure for coots and waterbirds is also via ingestion of a contaminated food source, but previous attempts failed to reproduce the disease in mallards fed sediment, surface water, and Hydrilla verticillata collected during an ongoing AVM outbreak (Larsen et al., 2003).
In this paper, we summarize numerous attempts to identify the source of AVM for waterbirds by feeding test animals a variety of materials collected from various lakes during confirmed outbreaks of AVM. These simple bioassays were conducted primarily to identify the material most likely to contain the causative agent prior to more detailed analyses.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Six species were used in feeding trials in attempts to identify the source of AVM as well as to find a suitable animal model. Animals used in these trials included 30 Japanese quail (Coturnix japonica), 54 American coots, five American kestrels (Falco sparverius), 228 mallards or pekin ducks, 15 American crows (Corvus brachyrhynchos), and 55 laboratory mice. All animals were individually identified with wing tags, leg bands, or dye marks (mice) and held in the U.S. Geological Survey National Wildlife Health Center (NWHC, Madison, Wisconsin, USA) isolation animal facility. Japanese quail obtained from the University of Wisconsin, Poultry Science Department (Madison, Wisconsin) were housed individually in cages (76x61x41 cm). They were provided poultry feed (Purina Mills, St. Louis, Missouri, USA) and water ad libitum.
American coots of mixed sex and age were captured by night lighting at Reelfoot National Wildlife Refuge (NWR, Tennessee, USA, 36°26'N, 89°3'W) in 1997 (24 birds); Mingo NWR (Missouri, USA, 36°58'N, 90°8'W) in 2002 (28 birds); and Horicon NWR (Wisconsin, 43°30'N, 88°38'W) in 2002 (11 birds), and transported to NWHC. These sites have never had AVM to our knowledge. Several coots from each group were randomly chosen and euthanized to confirm absence of brain lesions using light microscopy. In addition, any birds that died prior to treatment were necropsied and their brains examined for lesions. Coots were housed communally on the padded floor and provided with a large swimming pool filled with water and crate in which to hide. Coots were fed ad libitum a combination of Sea Duck pellets (Purina Mills), Waterfowl Maintenance (Purina Mills), mealworms (Rainbow Mealworms, Inc., Compton, California, USA), and greens such as cilantro, parsley, and bean sprouts.
American kestrels (adults, captive-reared) were obtained from the Patuxent Wildlife Research Center (Laurel, Maryland, USA). They were housed in wire mesh flight cages (0.9 x 1.8 m) provided with perches. Birds were fed 30 g meatballs or euthanized mice (Harlan Sprague Dawley, Indianapolis, Indiana, USA).
One-day-old mallard (n = 133) or pekin ducklings (n = 40) were purchased from Whistling Wings (Hanover, Illinois, USA) or Abendroths (Waterloo, Wisconsin, USA). Ducklings were housed communally in cages and provided with Duck Grower mash (Purina Mills) and water ad libitum. Feeding trials began when the birds were 1 wk old. Adult mallard ducks (n = 55) purchased from Whistling Wings were wing-clipped, housed communally on a padded floor in groups no larger than 25 birds, and provided access to a swimming pool. Duck Maintenance Diet (Purina Mills) and water were provided ad libitum.
Adult American crows were captured using a drop net near Wichita, Kansas (USA, 37°41'N, 97°20'W). Two were euthanized prior to treatment to confirm absence of AVM lesions. Crows were housed at NWHC, two to three per cage (76x122x175 cm), and provided perches and toys such as ping-pong balls. They were fed Science Diet Adult Canine Maintenance (Hills Pet Nutrition, Topeka, Kansas, USA), broccoli, and hard-boiled eggs.
Mice (6–8 wk old, ICR strain) were purchased from Harlan Sprague Dawley (Indian-apolis, Indiana, USA) and housed in groups of four in standard laboratory mouse cages. Pelleted mouse food (PMI Nutrition International, Brentwood, Missouri, USA) and water were provided ad libitum.
Test material
Test material was collected opportunistically during AVM outbreaks from Woodlake (WL, also known as Lake Surf), North Carolina; DeGray Lake (DGL), Arkansas; and Strom Thurmond Lake (STL), on the Georgia and South Carolina border. The most extensive collections were from WL, a site where AVM has been documented nearly every year since 1997 and where we previously documented the disease in sentinel mallards and coots (Rocke et al., 2002). Test material collected from WL included lake water, submerged aquatic vegetation, sediment, plankton, small fish, and invertebrates. Samples were collected at depths <1 m about twice weekly between early October and mid-December, 2001 at three locations on the lake where waterfowl use is concentrated and affected coots have been found (Rocke et al., 2002). At the same time, mortality in wild birds was monitored and several carcasses were submitted to NWHC for diagnostic evaluation (NWHC, unpubl. data). Approximately 3 l of water and vegetation, primarily Hydrilla were collected in plastic freezer bags; no attempts were made to remove any of the associated epiphytes, plankton, or external debris (e.g., silt) on the vegetation. Using a shovel, about 250 ml of surface sediments (representing the top 5 cm of material) were collected. Suspended and loosely attached plankton (algae, zooplankton, phytoplankton, and other material hereafter referred to as planktonic filtrate) were obtained from areas of dense aquatic vegetation using a 153 µm mesh net. Manual net sweeps were conducted after agitation of the vegetation until about 100–200 ml of filtrate was obtained. Small fish and aquatic invertebrates were collected at one location (marina) using a seine until a sufficient sample size (approximately 100 g invertebrate material and 200 g small fish) or the sampling effort goal (two person-hr) was met. Fish collected, primarily juvenile bluegill (Lepomis macrochirus), represented sizes of prey items that could be consumed by coots and waterfowl and were typically <8 cm in length. Test material from DGL was collected in December 1996 and included Elodea (the dominant aquatic vegetation), water, duckweed (Lemna spp.), planktonic filtrate, and tissues from unidentified species of catfish, bass, and shad. Hydrilla was the only test material obtained from STL. Hydrilla was also collected from Harris Lake (HL; New Hill, North Carolina, USA), a site where AVM has not been diagnosed. All test materials were held on ice in the field and were stored at –20 C. Only those materials collected during confirmed AVM outbreaks in wild birds were used in the feeding studies. A portion of the material was also stored frozen for future chemical analyses if warranted by the outcome of the trial.
Prior to feeding trials, vegetation, sediment, invertebrates, and small fish were homogenized using a Waring blender. Only the muscle and GI contents were collected from larger fish (bass, shad, and catfish from DGL) and also homogenized with a Waring blender. The blended material was loaded into syringes and administered to test animals by gavage into the esophagus and/or oral cavity. In a few cases where animals would feed on the material voluntarily (e.g., mallards and Hydrilla), it was provided ad libitum as indicated in Table 1
.
|
Table 1
summarizes the experiments. Amount of material fed was dependent on the size of the test animals and duration of a trial was dependent on the amount of material available from collections by field personnel. Typically, animals were fed test material until the amount available was nearly completely used; this was as long as 43 days in a few trials. Control animals, co-housed in the same room with the test animals, received either the negative control samples (HL Hydrilla, tissues from AVM negative coots) or a placebo (water or saline) for the same length of time as the test animals.
Test animals were visually evaluated at least once daily for clinical signs consistent with AVM (Larsen et al., 2002), for example, inco-ordination, ataxia, knuckling, or limb weakness in one or both hind limbs. Complete necropsies were performed on any animal that died or became sick and was euthanized (via CO2 asphyxiation). Brains and other tissues were placed in 10% buffered formalin for histologic examination and tissues were collected for laboratory analyses if warranted by gross signs observed at necropsy. At the end of a trial, test and control animals were euthanized and their brains were removed for histologic examination. Brain sections were processed routinely for paraffin embedment, sectioned at 4–5 µm, and stained with hematoxylin and eosin. White matter in four regions, optic tectum, optic chiasm, medulla, and cerebellar folia, of each brain was examined by light microscopy. Our case definition for AVM included diffuse white matter vacuolation in the optic tectum and at least one other region under light microscopy as described in Thomas et al. (1998). Birds with this specific brain lesion were considered positive even in the absence of clinical signs.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In the earliest trials, conducted in 1997, no clinical signs were evident in any of the birds tested (quail, coots, and kestrels) and no histologic lesions were noted in the quail brains (Table 1). Unfortunately, at the time this experiment was conducted, it was not recognized that brain lesions could occur in animals in the absence of clinical signs (Fischer et al., 2003) and brains were not collected from the clinically normal coots, kestrels, or control animals.
In the subsequent trials conducted in 1999, 2000, and 2002, brains were collected from all test animals in every trial and examined by light microscopy for lesions. Clinical signs and brain lesions were not evident in the majority of animals tested (Table 1) with a few exceptions. Two of four mallards fed Hydrilla collected on 20 November 2001 from STL ad libitum for 24 days were found to have moderate, but distinct white matter vauolation consistent with AVM using light microscopy. Widespread vacuoles were evident in the inner white stratum of the optic lobe and optic chiasma, and although fewer, vacuoles were also present in the medulla and cerebellar folia. No clinical signs were evident in these birds. Two mallards fed Hydrilla from Harris Lake (HL), a site with no documented AVM outbreaks, had no signs of illness or histologic brain lesions after ingesting the vegetation for 28 days.
Two coots fed a sample of Hydrilla collected from one site at WL (the marina) on 6 November 2001 and used in a feeding trial in 2002 developed neurologic signs within 9 days (ataxia, limb weakness, and incoordination); one died on day 10 and the other was euthanized the following day. A subsequent sample from the same location collected 13 days later on 19 November 2001 was fed to two other coots and one of them became sick with similar neurologic signs and died on day 38. Two additional coots were fed samples of Hydrilla collected from the same site on 27 November 2001 for 42 days, and neither of them showed any signs of illness.
Upon necropsy of these birds, no gross lesions or abnormalities were observed, and no lesions were evident in the heart, liver, lung, kidney, or GI tract upon examination by light microscopy. Very mild, but inconclusive, vacuolation was observed in two of the three birds that showed clinical signs but not in the other four. Two coots fed Hydrilla collected from HL, the lake with no documented AVM outbreaks, had no signs of illness or brain lesions after ingesting the vegetation for 29 days.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Reproduction of AVM is inconsistent. Two of four mallards fed the same sample of Hydrilla from STL for 23 days developed brain lesions consistent with AVM, even though clinical signs were not observed in these birds; the other two birds had no brain lesions. The agent or agents responsible for AVM may not be evenly distributed throughout the vegetation. Non-uniform distribution of the agent(s) in the field could explain the failure to reproduce the disease in mallards fed Hydrilla from WL in a previous study (Larsen et al., 2003) and from some of the Hydrilla samples collected in our experiments. Another possible explanation is that only a low dose of the causative agent(s) was present in these samples and a threshold level must be ingested before brain lesions or clinical signs are evident. Individual variations between birds in their nutritional or physiologic condition might also play a role as well as storage, transport, and freezing of samples, which might affect the agent or its potency.
Neurologic signs similar to those in wild birds with confirmed AVM were observed in three test coots fed samples of WL Hydrilla. Light microscopy of the brains from the three sick birds was inconclusive. No lesions were noted in other tissues from these birds either by gross or histologic examination, and no other cause of death could be determined. Furthermore, none of the coots co-housed with these birds developed similar signs of illness. We believe that these three coots died as direct result of ingestion of the Woodlake Hydrilla (and its associated biotic and abiotic material), but because the lesions were not conclusive by light microscopy of the brain, they did not meet our case definition for AVM. Transmission electron microscopy was attempted (data not shown), but the results were inconclusive because of formalin-fixation artifact. Perhaps myelin vacuolation was occurring primarily on an ultrastructural level or alternatively, perhaps a separate neurotoxic agent caused these deaths.
Interestingly, the occurrence of clinical signs in our test coots fed Hydrilla coincides with the observed morbidity of wild birds at WL in 2001. The first impaired wild coot with evident neurologic signs was found at WL on 26 October 2001, but light microscopic evaluation of its brain was inconclusive. Three other moribund coots were collected at WL on 9 November 2001, and all had brain lesions consistent with AVM. Moribund and dead coots and mallards continued to be observed from this date through mid-December. In our feeding studies, one Hydrilla sample collected on 6 November at WL caused neurologic signs in two of two test coots within 9 days of consumption. A second Hydrilla sample collected at the same site at WL on 19 November 2001 caused neurologic signs in one of two test coots within 38 days of consumption. Neurologic signs in two test coots were not evident upon ingestion of a Hydrilla sample collected from the same site in late November. In another study at WL conducted the year before, serial releases of sentinel mallards during the summer, fall, and winter demonstrated that exposure to the causative agent of AVM at a threshold sufficient to manifest disease (clinical signs and/or brain lesions) was seasonal and occurred over about a 2 mo period, during November and December (Rocke et al., 2002).
The experimental documentation of aquatic vegetation as a link in AVM transmission is consistent with coot feeding behavior in the wild. Coots primarily consume aquatic vascular plants and algae, with lesser amounts of grasses and other terrestrial vegetation, fish, tadpoles, crustaceans, mollusks, aquatic and terrestrial insects, and other invertebrates (Allen, 1985; Brisbin et al., 2002). They generally feed on or under shallow water where submerged or emergent macrophytes, such as Hydrilla, are most abundant (Brisbin et al., 2002). The range of food items for coots in the freshwater wintering grounds in North Carolina is not definitively known, but the preferred food item at WL appears to be Hydrilla. Coots are commonly observed dabbling and diving for this plant and consuming it in large quantities at WL, and Hydrilla is a common food item for coots elsewhere in the southeastern US (Brisbin et al., 2002). Coots at WL have also been observed to be stripping material, presumably algae, from the surface of Hydrilla, a foraging behavior noted by others (Brisbin et al., 2002). The experimental reproduction of AVM in mallards fed Hydrilla from STL is also relevant to the field scenario. Mallards are omnivorous and opportunistic with more reliance on aquatic vegetation in autumn and winter (Drilling et al., 2002).
In our experiments, clinical signs and brain lesions were not evident in animals that received other test materials, including sediments, fish, plankton, and invertebrates. Although further work might be necessary once the agent is identified, we believe that it is unlikely that sediments, fish, and invertebrates are commonly associated with the disease. It is interesting that the planktonic filtrate associated with WL Hydrilla did not result in clinical signs or brain lesions of AVM, even though it was collected at the same time and from the same location as the Hydrilla samples that caused clinical signs in coots. Either this material alone is not the source of AVM or the duration of feeding and/or dose was insufficient to induce disease, although the volume fed to the birds would far exceed that ingested incidentally with unwashed Hydrilla. Perhaps a more deliberate approach to stripping epibiotic material from the surface of Hydrilla should be explored.
Also, we could not reproduce the clinical signs or brain lesions in ducklings and crows that received tissues from affected birds, even those that were fed GI contents. Other investigators found brain lesions (but no clinical signs) in five of five red-tailed hawks (Fischer et al., 2003) and five of five chickens (Lewis-Weis et al., 2004) fed tissues from AVM-affected coots. These investigators determined that the agent was associated with GI contents from affected coots. Perhaps ducklings and crows are less sensitive to the agent or the amount we fed to test animals was insufficient to cause lesions. The chickens were fed 20 g of GI contents for 28 days (Lewis-Weis et al., 2004) and our crows were fed approximately 20 g for 30 days. Ducklings received 2 ml/day for 5–10 days. It is also possible that the GI content from the wild coots with AVM did not contain the toxic agent at the time of sampling or we did not feed the test material for a long enough period.
In summary, ingestion of several samples of Hydrilla (but not all) from lakes with ongoing outbreaks of AVM resulted in brain lesions in mallards indicative of AVM. These results support our hypothesis that the causative agent of AVM is ingested by waterbirds while consuming aquatic vegetation at affected sites. At WL and STL, Hydrilla is the dominant aquatic vegetation, however, it is not present in all AVM-affected lakes. Although we don’t have definitive data, we suspect the disease is associated with other aquatic vegetation that is dominant in other affected lakes. Based on results of our previous work with sentinel mallards and coots at WL (Rocke et al., 2002), we hypothesize the agent is either seasonally accumulated by aquatic vegetation, such as Hydrilla, or seasonally produced by one or more organisms associated with aquatic vegetation at affected sites. Also, upon ingestion of some Hydrilla samples collected during an AVM outbreak, several coots in our studies became sick and died with neurologic signs similar to those seen in wild birds, but lacking the characteristic brain lesions of AVM.
|
ACKNOWLEDGMENTS |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
AUGSPURGER, T., J. R. FISCHER, N. J. THOMAS, L. SILEO, R. E. BRANNIAN, K. J. G. MILLER, AND T. E. ROCKE. 2003. Vacuolar myelinopathy in waterfowl from a North Carolina impoundment. Journal of Wildlife Diseases 39: 412–417.[Abstract]
BRISBIN, I. L., JR., H. D. PRATT, AND T. B. MOWBRAY. 2002. American coot (Fulica americana) and Hawaiian coot (Fulica alai). In The Birds of North America, No. 697, A. Poole and F. Gill (eds.). The Academy of Natural Sciences, Philadelphia, Pennsylvania, 44 pp.
DRILLING, N., R. TITMAN, AND F. MCKINNEY. 2002. Mallard (Anas platyrhynchos). In The Birds of North America, No. 658, A. Poole and F. Gill (eds.). The Academy of Natural Sciences, Philadelphia, Pennsylvania, 44 pp.
FISCHER, J. R., L. A. LEWIS, T. AUGSPURGER, AND T. E. ROCKE. 2002. Avian vacuolar myelinopathy: A newly recognized fatal neurological disease of eagles, waterfowl and other birds. Transactions of the North American Wildlife and Natural Resources Conference 67: 51–61.
———, ———, AND C. M. TATE. 2003. Experimental vacuolar myelinopathy in red-tailed hawks. Journal of Wildlife Diseases 39: 400–406.[Abstract]
LARSEN, R. S., F. B. NUTTER, T. AUGSPURGER, T. E. ROCKE, L. TOMLINSON, N. J. THOMAS, AND M. K. STOSKOPF. 2002. Clinical features of avian vacuolar myelinopathy in American coots. Journal of the American Veterinary Medical Association 221: 80–85.[Medline]
———, ———, ———, ———, ———, ———, AND ———. 2003. Failure to transmit avian vacuolar myelinopathy to mallard ducks. Journal of Wildlife Diseases 39: 707–711.[Abstract]
LEWIS-WEIS, L. A., R. W. GERHOLD, AND J. R. FISCHER. 2004. Attempts to reproduce myelinopathy in domestic swine and chickens. Journal of Wildlife Diseases 40: 476–484.
ROCKE, T. E., N. J. THOMAS, T. AUGSPURGER, AND K. MILLER. 2002. Epizootiologic studies of avian vacuolar myelinopathy in waterbirds. Journal of Wildlife Diseases 38: 678–684.[Abstract]
THOMAS, N. J., C. U. METEYER, AND L. SILEO. 1998. Epizootic vacuolar myelinopathy of the central nervous system of bald eagles (Haliaeetus leucocephalus) and American coots (Fulica americana). Veterinary Pathology 35: 479–487.[Abstract]
Received for publication 25 November 2003.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
