A survey of the indigenous microbiota (bacteria) in three species of mussels from the Clinch and Holston Rivers, Virginia.
Article Type: Abstract
Subject: Mussels, Fresh-water (Protection and preservation)
Microbiota (Symbiotic organisms) (Research)
Authors: Starliper, Clifford E.
Neves, Richard J.
Hanlon, Shane
Whittington, Pamela
Pub Date: 12/01/2008
Publication: Name: Journal of Shellfish Research Publisher: National Shellfisheries Association, Inc. Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Zoology and wildlife conservation Copyright: COPYRIGHT 2008 National Shellfisheries Association, Inc. ISSN: 0730-8000
Issue: Date: Dec, 2008 Source Volume: 27 Source Issue: 5
Topic: Event Code: 970 Government domestic functions; 310 Science & research
Geographic: Geographic Scope: United States; Virginia Geographic Name: Virginia Geographic Code: 1USA United States; 1U5VA Virginia
Accession Number: 191646321
Full Text: ABSTRACT Freshwater mussel conservation efforts by many federal and state agencies have increased in recent years. This has led to a greater number of stream surveys, in which mussel die-offs involving high numbers of dead and moribund animals are being observed and reported with greater frequency. Typically, die-offs have been incidentally observed while research was being done for other purposes, therefore, accurate mortality data have been difficult to obtain. Specifically, seasonal die-offs were noted in localized areas of the Clinch and Holston Rivers, Virginia, and to lesser degrees, in neighboring rivers in this geographic region, including southeast Virginia. The observed mussel species affected were primarily the slabside pearlymussel (Lexingtonia dolabelloides) and to lesser extents, the pheasantshell (Actinonaias pectorosa), rainbow mussel (Villosa iris), and the endangered shiny pigtoe (Fusconaia cor). To determine if a bacterial pathogen might be involved in these recurring mussel die-offs, this study examined characteristics of the indigenous microbiota (bacteria) from healthy mussels from sites on the Clinch and Holston Rivers where die-offs were previously observed. These baseline data will allow for recognition of bacterial pathogens in future mussel die-offs. Means for total bacteria from soft tissues ranged from 1.77 x [10.sup.5] to 3.55 x [10.sup.6] cfu/g, whereas, the range in means from fluids was 2.92 x [10.sup.4] to 8.60 x [10.sup.5] cfu/mL. A diverse microbiota were recovered, including species that are common in freshwater aquatic environments. The most common bacterial groups recovered were motile Aeromonas spp. and nonfermenting bacteria. Flavobacterium columnare, a pathogen to cool- and warm-water fishes was recovered from one specimen, a Villosa iris from the Clinch River.

KEY WORDS: freshwater mussels, bacteria, microbiota, healthy, rivers, disease, prevention

INTRODUCTION

Freshwater mussels native to the United States are one of the more imperiled fauna (Lydeard et al. 2004, Williams et al. 1993). In light of this, many federal and state agencies have implemented or increased conservation efforts towards imperiled mussel species in recent years. In conjunction with the greater numbers of stream surveys or similar fieldwork being conducted, an increased frequency of apparently natural mussel die-offs have been observed in rivers and streams across the United States (Neves 1987). These events are characterized by high numbers of dead and moribund mussels observed in their natural environments, and typically are limited to a relatively localized geographic area. Anecdotal observations indicate that the mortality occurs within a time period ranging from about 2-4 days up to about two weeks, which may be indicative of an infectious agent-induced disease. This death rate generally differs from a rate that might be anticipated from a noninfectious cause, such as critically low dissolved oxygen or toxic substance exposure, in which an acute die-off might occur within a much shorter timeframe, perhaps within one day. Because these natural die-offs were typically noted incidentally and whereas related mussel field research was being conducted, accurate mortality data have been difficult to collect. Often, by the time an epizootic was noted, explanatory evidence towards determining a cause was likely to have been diluted and carried away with the flowing water. Furthermore, the gaped valves of dead animals expose the soft tissues and this renders them useless for any attempt to isolate a culturable pathogen.

Diagnosing the cause(s) of mussel die-offs among wild populations is difficult because of the nature of observing mortality in localized areas of large open waters and the relative lack of knowledge about pathogens to freshwater mussels. This contrasts with the diseases infecting some marine bivalves. Certain marine species are an important aquaculture commodity and have a sustained commercial fishery to support it. With captive rearing, diagnosticians have the opportunity to observe behavioral changes in animals and possibly recognize diseases in their early stages. The mortality can be observed and fresh specimens are readily obtained for laboratory diagnostics. Subsequent descriptions of causes for diseases and mortality may be made. Refuge propagation of imperiled freshwater mussels is likely to continue increasing in future years. Infectious agent-induced diseases could become problematic, similar in the way they have with other organisms propagated at culture facilities; for example, among cultured fishes (Noga 1996, Woo & Bruno 1999, Wedemeyer 2001). Currently, about 15 federal and state facilities (e.g., fish hatcheries) are maintaining and propagating a suite of freshwater mussel species for use in restoration or augmentation of wild populations. Effective disease preventative measures are likely to become essential for the success of these programs.

Along with the apparent death rate, additional observations from die-offs lead to suspicions of pathogens as the causes of death. At certain locations on the Clinch and Holston Rivers, Virginia, and to lesser degrees in neighboring rivers, die-offs have been observed during the same timeframes of different years, coinciding with seasonal increases in water temperatures that likely favor optimal bacterial pathogen growth and infectivity. Host specificity, or predilection for certain mussel species is also thought to be a contributing factor to disease. Mussel die-offs in the Middle Fork Holston and Clinch Rivers, Virginia, have occurred in recent years during late spring and summer. This encompasses spawning and glochidia release times for some mussel species, including the endangered shiny pigtoe (Fusconaia cor), slabside pearlymussel (Lexingtonia dolabelloides), a candidate species for federal protection, the pheasantshell (Actinonaias pectorosa), and rainbow mussel (Villosa iris). Large numbers of empty shells are common in the affected stretches of the rivers. At this time, additional specimens observed were apparently moribund as evidenced by delayed and weakened hinge/valve closing in response to stimuli.

In an effort to delineate the possible role of a bacterial pathogen in these die-offs, the approach taken in the current study has been to describe the normative microbiota of healthy mussels from the affected regions of these rivers. These baseline data, including total bacterial loads in mussels and bacterial species profiles, will aid in the recognition of suspected bacterial pathogens isolated from mussels during subsequent die-offs. Describing the pathogens and diseases of freshwater mussels is essential to developing disease prevention strategies and procedures. This forms the basis for an effective health management program designed to eliminate or greatly reduce the risk for transmission of pathogens and spread of diseases among captive populations at refuges, and to feral populations of mussels through the stocking of refuge-reared individuals.

MATERIALS AND METHODS

Three species of mussels were collected by snorkeling from the Holston and Clinch Rivers, Virginia: pheasantshell (Actinonaias pectorosa), slabside pearlymussel (Lexingtonia dolabelloides), and rainbow mussel (Villosa iris). These species and the collection sites were selected for this study because of observed mussel die-offs at these locations in recent years, particularly affecting populations of L. dolabelloides, and to lesser degrees, the other two species. The A. pectorosa and L. dolabelloides were collected from the Middle Fork Holston River, river mile 10.1, Washington County, Virginia. The A. pectorosa and V. iris from the Clinch River were collected at Nash's Ford, river mile 279.8, Russell County, VA. Each location was sampled in early summer (a die-off was not observed during this year) and early fall within the same calendar year. After collection, the mussels were kept alive and fresh during overnight shipment to the National Fish Health Research Laboratory (Kearneysville, West Virginia) by wrapping them in wet burlap and placing them in coolers at the collection site. The mussels were sampled for bacteriological analyses immediately upon arrival at the laboratory. A total of 80 mussels were examined for the presence of indigenous bacterial.

Primary isolation of bacteria from each mussel was completed using the methods described in Starliper et al. (1998) and Starliper (2001). External surfaces of the valves were cleaned and disinfected by gentle scrubbing with a brush and a solution of 200 mg/L sodium hypochlorite. The following morphometric data were recorded for each specimen: length, width, depth, total weight, and the area inside the valves (determined by measuring the volume of water necessary to fill the valves where 1 mL = 1 cc area). Aseptic techniques were used to the extent possible to collect the fluids and soft tissues used for bacterial sampling. The valves were pried 12-15 mm apart so the adductor muscles could be cut. The fluid (liquid inside the valves) was poured into a sterile Petri dish, and measured volumetrically. The soft tissues were excised and treated as one sample (per animal). The outer surfaces of the soft tissues were disinfected by submersing them in 1L of 200 mg/L sodium hypochlorite and gently keeping them in motion using a sterile pipette for 30 s. They were immediately rinsed for 5-10 s in sterile pep-ye (0.1% peptone, 0.05% yeast extract; Difco, Becton, Dickinson and Company, Sparks, MD) and placed in a sterile sampling bag (Fisher Scientific, www.fishersci.com). The tissues were then homogenized in an equal amount (w/v; 1:2 dilution) of sterile pep-ye for two min using a Model 80 Laboratory Blender (Seward Medical, London, UK). Three serial 10-fold dilutions in pep-ye were made from each fluid sample and from each tissue homogenate. Standard volumes (three 0.025 mL drops) of the fluid samples, tissue homogenates, and all dilutions were used to drop-inoculate the surface of two bacteriological media poured in Petri dishes, brain heart infusion agar (BHIA; Difco) and R2A agar (Difco). The plates were incubated until the growing bacterial colonies were large enough to comfortably enumerate and select for transfers. Incubation was for 24-72 h; BHIA plates were incubated at 20[degrees]C to 22[degrees]C, and R2A plates were incubated at 16[degrees]C. The total numbers of colony-forming units cultured from each tissue (cfu/ g) sample and fluid (cfu/mL) were calculated by averaging the numbers of colonies enumerated (from three 0.025 mL standard volumes) from the lowest sample dilution that yielded single, isolated colonies, which were easily and accurately scored. The mean was multiplied by the dilution factor to convert to the number in standard units (per mL or per g). Bacterial colonies from BHIA and R2A media representing all colony morphologies were selected and transferred to fresh media for growth. These resulting cultures were checked for purity by streak-plating and each was identified using standard biochemical and characterization methods (Griffin 1992, Holt et al. 1994, Janda & Abbott 1998, Koneman et al. 1992, MacFaddin 2000, Murray et al. 1999) and the API identification system (bio-Merieux, Inc., Hazelwood, MO). Significant differences ([alpha] = 0.05) among the morphometric and bacterial cfu data were identified using analysis of variance (ANOVA) and the z-statistic. The morphometric data were evaluated for differences associated with collection sites (Holston River versus Clinch River) and collection dates (summer versus fall).

RESULTS

The means and ranges for the morphometric data from the three species of mussels, A. pectorosa, L. dolabelloides, and V. iris are presented in Table 1. The largest of the three species was A. pectorosa; the mean total weights for 10 specimens ranged between 208.4 g for the Clinch River summer collection and 294.5 g for the fall group from Holston River. The smallest A. pectorosa of the study weighed 142.6 g, whereas the heaviest was 447.5 g. From the summer and fall collection dates, the means for the total weights of A. pectorosa groups from the Holston River were greater than those for groups from the Clinch River. However, the means for percent of the mussels' total weight comprised of soft tissues or fluids from the Clinch River groups were larger than cohort groups from the Holston River. Regardless of the river of origin and collection date, the combined weight of soft tissues and fluid comprised about one fifth of the mussels' total weight. The mean total weights of A. pectorosa collected in the fall, from the Holston and Clinch Rivers were greater than those from the summer collections. This contrasted with the other two mussel species in which the means for total weights were greater in summer. The total weight of L. dolabelloides, comprised of soft tissues and fluids, was greater in fall (23.0%) compared with summer (15.8%). This contrasted with V. iris where the percentage in summer (28.9%) was greater than that for fall (27.2%). The smallest of the three mussel species in total weight was V. iris; however, the percentages of their total weight comprised of soft tissues and fluids, combined, were greater than those for A. pectorosa and L. dolabelloides.

Statistical comparisons of morphometric data by river and collection date are given in Table 2. Of six possible pairings of river and date for comparison, only one differed by total weight; A. pectorosa collected in fall from the Holston River were larger (P = 0.006) than those collected from the Clinch River. Of the 42 individual comparisons listed in Table 2, significant differences were noted in 12 instances. Only one of these was a comparison that did not involve A. pectorosa; the percent of the total weight of L. dolabelloides comprised of fluid was greater (P < 0.001) in fall (12.3%; Table 1) than in summer (4.2%). Eight of the 12 significant pairings were attributed to seasonal (summer versus fall) comparisons. The summer versus fall collection dates for A. pectorosa from the Clinch River accounted for the largest number (4) of significant differences, with differences in length (P = 0.030), depth (P = 0.025), tissue percent weight (P < 0.001), and fluid percent weight (P = 0.002). Furthermore, the means for length, depth, and tissue percent weight were greater in fall, whereas the mean for fluid percent weight was greater in summer, 10.2% relative to 7.5 % (Table 1). Collection dates for A. pectorosa from the Holston River accounted for three significant differences; the means of each were greater from the fall collection, length (P = 0.001), width (P = 0.007) and depth (P = 0.020) than from the spring collection. There were no significant differences in morphometric data for area, and for V. iris from summer versus fall collections from the Clinch River.

The means and ranges for quantities of total bacteria isolated from the mussels' fluids and tissues are given in Table 3. There were no significant differences between the two bacteriological media (BHIA and R2A) for quantitative recovery of total bacteria isolated from fluids (P = 0.682) or tissue homogenates (P = 0.556). Each of the means for total bacteria from the tissue homogenates was greater than the means from paired fluids; seven of the eight tissue homogenate means were significantly greater (P values ranged from <0.001-0.015), and although the mean from A. pectorosa tissues from the (Clinch River) fall sampling was greater than in summer, it was not significantly different (P = 0.164). The highest total bacterial recovery was from the tissues and fluids from A. pectorosa and V. iris during the Clinch River summer collections with a minimum of 2.41 X [10.sup.6] cfu/g from tissues and at least 1.63 X [10.sup.5] cfu/mL from fluids. The largest difference between tissue and fluid cfu was from V. iris in summer, the differences were 2.25 x [10.sup.6] cfu with BHIA, and 3.32 x [10.sup.6] cfu using R2A.

The means for bacteria isolated from tissues ranged from 1.77 x [10.sup.5] cfu/g to 3.55 X [10.sup.6] cfu/g. The range from fluids was 2.92 x [10.sup.4] cfu/mL to 8.60 x [10.sup.5] cfu/mL. The highest total bacterial recovery from a tissue homogenate from a specimen, a V. iris, was 1.52 x [10.sup.7] cfu/g, and the highest from a fluid sample was nearly 10-fold less at 4.80 x [10.sup.6] cfu/mL from an A. pectorosa. Comparisons of the means for bacteria recovered from mussels to determine significant differences because of river of origin or collection date are presented in Table 4. During summer collections of A. pectorosa from the two rivers, bacterial recovery from tissues (P = 0.007) and fluids (P < 0.001) was greater from the Clinch River specimens. Total bacterial loads from A. pectorosa collected in summer from the Clinch River were also significantly higher (tissues: P = 0.002; fluids: P < 0.001) than the means in bacteria from A. pectorosa collected from the same river in fall. From the Holston River, A. pectorosa tissues from the fall collection yielded significantly higher cfu/g (P = 0.019) compared with summer, and also produced a greater mean for total bacteria from tissues relative to cohorts from the Clinch River (P = 0.003). Tissue homogenates from L. dolabelloides resulted in significantly higher cfu/g (P = 0.002) in the fall, with no difference (P = 0.829) in bacterial cfu/mL from the paired fluids. There were no significant differences in the total bacteria isolated from tissues (P = 0.949) or fluids (P = 0.180) from V. iris between the two collection dates.

A diverse microbiota was recovered from the mussels from the Holston and Clinch Rivers (Table 5). There were at least 30 different bacteria present in mussels throughout this study; 13 of which were recovered from both rivers. This broad range in bacteria was noted in summer and fall collections and from both tissues and fluids (data not shown). The predominant bacteria in animals from both rivers were motile Aeromonas spp. and various glucose-nonfermenting bacteria, including Acinetobacter species, Brevundimonas vesicularis, Chryseobacterium indologenes, Pseudomonas fluorescens, and Sphingomonas paucimobilis. A variety of enteric bacteria were also identified, including Citrobacter koseri, Enterobacter intermedius, Hafnia alvei, Proteus vulgaris, Providencia rettgeri, and Serratia spp. Flavobaeterium columnare, a pathogen to many warm- and cool-water fishes (Noga 1996, Wedemeyer 2001), was isolated from a single V. iris from the Clinch River during the summer collection. This Gram-negative bacterium was isolated on R2A medium from the fluid sample, bur not from the paired soft tissues homogenate. Furthermore, this specimen was one of the smaller mussels (total weight: 4.75 g) collected during the study. On the R2A primary isolation plate, 4.00 x [10.sup.3] F. columnare per mL was cultured, and 0.3 mL of fluid was recovered, therefore, a total of 1.20 x [10.sup.3] viable cfu F. columnare was calculated to be present in that specimen.

DISCUSSION

Historic numbers of many species and populations of freshwater mussels native to the United States have declined, and continue to decline, because of causes that either directly affect the mussels or impact hosts for transformation of glochidia (Lydeard et al. 2004; Williams et al. 1993). Mussel conservation efforts to salvage or sustain many of the imperiled species are ongoing. One strategy involves collecting at-risk individuals from impacted rivers and relocating them to refugia for propagation. Success at propagation will depend on the development and implementation of techniques that ensure good mussel husbandry, including proper diet and feeding, adequate water quality and flow, proper substrate, and identification of hosts for transformation. The objective for propagation is to provide healthy individuals for future strategic stockings into impacted rivers for augmentation or restoration of affected species or populations.

Maintaining excellent health among captive individuals is imperative for a successful mussel propagation program. Healthy mussels will reflect the historic genetic integrity of wild populations, maintain a condition factor indicative of good physical health, and remain pathogen- and disease-free (Villella et al. 1998, Jones et al. 2006). Optimal condition factor will likely enhance resistance to diseases. In addition, regularly scheduled pathogen and disease examinations, or inspections, of refuge-reared mussels could become a useful tool.

Federal and state fishery agencies rely on periodic health inspections of cultured fishes as an element of disease prevention programs. For example, the United States Fish and Wildlife Service has regional fish health units that specialize in proper husbandry and recognition of common and emerging fish pathogens and diseases. Routine health inspections are essential for controlling pathogen introductions to production facilities and to naive, feral populations through stockings into streams. These Fish Health Units also conduct extensive health evaluations on wild fish populations (National Wild Fish Health Survey; www.fws.gov/wildfishsurvey) to determine the prevalences of select pathogens, and the host and geographic ranges of diseases. This important information allows resource managers to make well-informed decisions about relocating or stocking fishes between rearing facilities, and to feral streams. A program similar to this for mussels could provide important information to mussel resource managers.

Because mussels continue to be collected from open waters and introduced to refuges, this presents a continuing risk of pathogens to resident mussels at refuges. The consequences for a pathogen introduction scenario to hatchery-reared fish are well documented (Noga 1996, Wedemeyer 2001), and a similar process could occur for captive mussels. Another complicating factor is many of the host fishes required for glochidia transformation are not commonly propagated. Therefore, these fish must also be wild-caught and placed at refuges, which heightens the risk of introducing pathogens, to resident fish and perhaps, to mussels.

The profile of bacteria within mussels apparently changes quite rapidly, which can be useful to prevent pathogen introductions. Nichols et al. (2001) identified if endosymbiotic microbes (i.e., bacteria) were present in mussels; they were not present. They described bacteria in mussels to be transient, which can imply that the microbiota is subject to relatively quick change in response to a change in environment. Furthermore, they showed that the bacteria in mussels varied by season and habitat, and not by mussel species. If mussels collected from rivers for refugia are first quarantined (e.g., 30 d; Chaffee 1997, Gatenby et al. 1998) to guard against the introduction of zebra mussels, this quarantine could also offer the opportunity for mussels to depurate pathogens. Previous studies with three-ridge (Amblema plieata) and ebonyshell (Fuseonaia ebena) have also shown that the microbiota of mussels responds rapidly to changes in their water supply (Starliper et al. 1998). Using a model system and the fish pathogenic bacterium Aeromonas salmonicida, it was shown that the percentage of A. plicata that were A. salmonicida-positive was reduced from 100% to 0% within 30 d in a flow-through water system (Starliper 2001). This demonstrated that quarantine for zebra mussels might also be effective in greatly minimizing the risk of transmitting pathogens into the captive population. However, extended periods of quarantine can result in a reduced condition factor (Patterson et al. 1997, 1999), which might serve to predispose mussels to diseases and minimize the effect of the quarantine.

It is important to develop knowledge on the characteristics of the indigenous microbiota in healthy mussels, because this will serve as the expected, or baseline, when diseased animals are examined. Whereas the microbiota from the mussels of a certain area may typify that particular geographic locale, some generalized observations can be made for nondiseased mussels. There is typically one-half to 10-fold greater total bacterial load (cfu) present within tissues than is isolated from the paired fluid samples. However, the microbiota present in fluids reflects that which is isolated from the soft tissues. As would be expected from other healthy aquatic animals such as fishes, the microbiota in mussels is also quite diverse, usually at least several bacterial species may be readily isolated from healthy mussels. A wide range in bacterial species was isolated in the present study, and has been noted in other mussel species, including several species from the Ohio River (Starliper et al. 1998), and Elliptio complanata (Chittick et al. 2001). An expected diverse microbiota is likely a useful criterion to keep in mind when examining moribund or freshly dead mussels from a die-off having a suspected bacterial etiology. In an epizootic offish, for example, recovery of the bacterial pathogen causing the disease can be expected from many of the individuals examined, and the pathogen cfu recovery will be high relative to that from a normative microbiota. Primary bacterial growth on bacteriological media will often show pure or nearly pure cultures from diseased tissues, which contrasts with the variety in bacterial species isolated from healthy individuals. High cfu recovery of nearly pure cultures on the appropriate medium will be indicative of a pathogen.

The extensive list of bacteria from mussels in the present study (Table 5) is generally representative of common aquatic environmental bacteria from freshwater, and some or most would likely be cultured from the fluids and tissues of mussels from many different waters. Acinetobacter spp., motile Aeromonas spp., Pseudomonas spp., Moraxella sp., and enteric bacteria are noteworthy examples, because these genera are also isolated from marine bivalves. Iida et al. (2000) isolated an average of 1.00 x [10.sup.5] cfu total bacteria per gram from the digestive tract tissues of Pacific oysters Crassostrea gigas. Some of the same genera listed in Table 5 were isolated as part of the microbiota from other marine bivalves including Mytilus galloprovincialis, Perna viridis, and Seapharca cornea (Kueh & Chan 1985, Salati et al. 1999). Also, these bacteria are occasionally isolated from the external (mucus) and internal tissues of apparently healthy fishes, and some, in particular motile Aeromonas spp., P. fluoreseens, S. liquefaciens, and S. putrefaciens, are recognized as opportunistic or secondary pathogens to some species of cultured fishes, including salmonid and catfish species. At present, none of these bacteria are known to be the cause of any diseases to freshwater mussels. In the present study, some of the bacteria were isolated from the Holston and Clinch Rivers (Table 5), whereas others (e.g., Acinetobacter lwoffii in the Holston or Hafnia alvei in the Clinch) were isolated from only one site. The microbiota can also be anticipated to vary between origins and collection dates. For example, although A. lwoffii was not isolated from the Clinch River during this study, its presence in the Clinch River in future isolations would not necessarily identify it as a pathogen unless other criteria were met. Heavy growth of pure, or nearly pure cultures on primary bacterial isolation plates from a significant number of moribund or fresh-dead individuals is one characteristic that may lead to suspicion of the bacterium as a pathogen.

The isolation of F. columnare from V. iris was the second such reported isolation of this bacterium from a riverine mussel. Previously, F. columnare was isolated from tissues of a three-ridge mussel Amblema plicata from the Ohio River; however, it was not recovered from cohort A. plicata after one day or more of depuration in a pathogen-free water supply (Starliper et al 1998). There was a total of 1.20 x [10.sup.3] cfu of F. columnare present in the positive V. iris in the current study. This negligible quantity of viable cells would likely have been quickly depurated if this mussel had been subjected to similar quarantine parameters applied to the A. plicata in the previous study (Starliper et al. 1998).

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the United States Government.

ACKNOWLEDGMENTS

This research was funded through the Quick Response program of the United States Geological Survey and the United States Fish and Wildlife Service, Department of the Interior. Janet Clayton and Craig Stihler of the West Virginia Division of Natural Resources provided guidance on mussel collection and importation permits. The authors thank Dr. Christine Densmore and Dr. Barnaby Watten for their insightful reviews.

LITERATURE CITED

Chaffee, C. 1997. Ohio River valley ecosystem mollusk conservation plan: Strategic action plan-phase II. Appendix H: Draft freshwater mussel quarantine protocol. Bloomington, IN: United States Fish anal Wildlife Service. 13 pp.

Chittick, B., M. Stoskopf, M. Law, R. Overstreet & J. Levine. 2001 Evaluation of potential health risks to Eastern EUiptio (Elliptio complanata) (Mollusca: Bivalvia: Unionida Unionidae) and implications for sympatric endangered freshwater mussel species. J. Aquat. Ecosyst. Stress Recovery 9:35-42.

Gatenby, C. M., M. A. Patterson, B. C. Parker, P. Morrison & R. J. Neves. 1998. A protocol for the salvation and quarantine of mussels from zebra-infested waters. Conservation, captive care, & propagation. Freshwater Mussel Symposium. March 6-8. Columbus, OH. pp. 18-19.

Griffin, B. R. 1992. A simple procedure for identification of Cytophaga columnaris. J. Aquat. Anim. Health 4:63-66.

Holt, J. G., N. R. Krieg, P. H. A. Sneath, J. T. Staley & S. T. William: Eds. 1994. Bergey's Manual of Determinative Bacteriology, 9th Edition. Baltimore, MD: Williams and Wilkins. 787 pp.

Iida, Y., R. Honda, M. Nishihara & K. Muroga. 2000. Bacterial flora i the digestive tract of cultured Pacific Oyster. Fish Pathol 35:173-177.

Janda, J. M. & S. L. Abbott. 1998. The Enterobacteria. Philadelphia, PA: Lippincott-Raven Publishers. 387 pp.

Jones, J. W., E. M. Hallerman & R. J. Neves. 2006. Genetic management guidelines for captive propagation of freshwater mussels (Unionoidea). J. Shellfish Res. 25:527-535.

Koneman, E. W., S. D. Allen, W. M. Janda, P. C. Schreckenberger & W. C. Winn, Jr. 1992. Color atlas and textbook of diagnostic microbology, 4th ed. Philadelphia, PA: J. B. Lippincott Company. 1154 pp.

Kueh, C. S. & K. Y. Chan. 1985. Bacteria in bivalve shellfish with special reference to the oyster. J. Appl. Bacteriol. 59:41-47.

Lydeard, C., R. H. Cowie, W. F. Ponder, A. E. Bogan, P. Bouchet, S. A. Clark, K. S. Cummings, T. J. Frest, O. Gargominy, D. G. Herber R. Hershler, K. E. Perez, B. Roth, M. Seddon, E. E. Strong & F. G. Thompson. 2004. The global decline of nonmarine mollusks. Bioscience 54:321-330.

MacFaddin, J. F. 2000. Biochemical tests for identification of medico bacteria, 3rd ed. Philadelphia, PA: Lippincott Williams and Wilkin 912 pp.

Murray, P. R., E. J. Baron, M. A. Pfaller, F. C. Tenover & R. H. Yolken. 1999. Manual of Clinical Microbiology, 7th ed. Washington, D.C: American Society for Microbiology Press. 1773 pp.

Neves, R. J. 1987. Recent die-offs of freshwater mussels in the United States: An overview. In: R. J. Neves, editor. Proceedings of the Workshop on Die-offs of Freshwater Mussels in the United States. Blacksburg, VA: Virginia Tech Press. pp. 7-18.

Nichols, S. J., J. Allen, G. Walker, M. Yokoyama & D. Garling. 2001. Lack of surface-associated microorganisms in a mixed species community of freshwater Unionidae. J. Shellfish Res. 20:329-335.

Noga, E. J. 1996. Fish Disease: Diagnosis and Treatment. St. Louis, MO: Mosby-Year Book, Inc. 367 pp.

Patterson, M. A., B. C. Parker & R. J. Neves. 1997. Effects of quarantine rimes on glycogen levels of native freshwater mussels (Bivalvia:Unionidae) previously infested with zebra mussels. Am. Malacol. Bull. 14:75-79.

Patterson, M. A., B. C. Parker & R. J. Neves. 1999. Glycogen concentration in the mantle tissue of freshwater mussels (Bivalvia: Unionidae) during starvation and controlled feeding. Am. Malacol. Bull. 15:47-50.

Salati, F., A. Meloni, E. Marongiu & R. Kusuda. 1999. Microbiological study on the flora of mussel, Mytilus galloprovincialis, cultured in Tortoli Lagoon, South-East Sardinia, Italy. Fish. Sci. 65:657-658.

Starliper, C. E. 2001. The effect of depuration on transmission of Aeromonas salmonicida between the freshwater bivalve Amblema plicata and Arctic char. J. Aquat. Anim. Health 13:56-62.

Starliper, C. E., R. Villella, P. Morrison & J. Mathias. 1998. Studies on the bacterial flora of native freshwater bivalves from the Ohio River. Biomed. Lett. 58:85-95.

Villella, R., T. L. King & C. E. Starliper. 1998. Ecological and evolutionary concertas in freshwater bivalve relocation programs. J. Shellfish Res. 17:1407-1413.

Wedemeyer, G. A. Ed. 2001. Fish hatchery management 2nd ed. Bethesda, MD: American Fisheries Society. 733 pp.

Williams, J. D., M. L. Warren, Jr., K. S. Cummings, J. L. Harris & R. J. Neves. 1993. Conservation status of freshwater mussels of the United States and Canada. Fisheries 18:6-22.

Woo, P. T. K. & D. W. Bruno. Eds. 1999. Fish Diseases anal Disorders, Volume 3: Viral, Bacterial and Fungal Infections. New York, NY: CABI Publishing. 874 pp.

CLIFFORD E. STARLIPER, (1) * RICHARD J. NEVES, (2) SHANE HANLON (3) AND PAMELA WHITTINGTON (1)

(1) USGS Leetown Science Center, National Fish Health Research Laboratory, 11649 Leetown Road, Kearneysville, West Virginia 25430; (2) USGS Virginia Cooperative Fish and Wildlife Research Unit, Department of Fisheries and Wildlife Sciences, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061; (3) US Fish and Wildlife Service, Southwest Virginia Field Office, 330 Cummings Street Abingdon, Virginia 24210

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

* Corresponding author. E-mail: cstarliper@usgs.gov
TABLE 1.
Morphometric characteristics of Actinonaias pectorosa, Lexingtonia
dolabelloides and Villosa iris from the Holston and Clinch rivers,
VA. Means and ranges for 10 mussels of each species from each
collection date.

Species             Origin       Date            Weight (a)

A. pectorosa        Holston      Summer (July)   239.3 (150.9-315.7)
L. dolabelloides    Holston      Summer (July)   110.6 (81.4-160.9)
A. pectorosa        Clinch       Summer (June)   208.4 (152.8-245.4)
V. iris             Clinch       Summer (June)   8.8 (4.8-12.3)
A. pectorosa        Holston      Fall (October)  294.5 (220.9-447.5)
L. dolabelloides    Holston      Fall (October)  103.8 (37.7-141.7)
A. pectorosa        Clinch       Fall (Sept.)    214.4 (l42.0-349.4)
V. iris             Clinch       Fall (Sept.)    8.2 (4.0-14.1)

Species             Length (a)        Width (a)           Depth (a)

A. pectorosa        105.7 (94-120)    66.7 (60-75)        43.7 (37-48)
L. dolabelloides    75.1 (69-85)      60.7 (53-74)        30.5 (27-34)
A. pectorosa        108.0 (103-116)   68.4 (65-74)        42.5 (38-46)
V. iris             38.5 (33-42)      23.7 (20-27)        14.0 (11-16)
A. pectorosa        118.6 (107-127)   72.6 (68-80)        47.8 (43-59)
L. dolabelloides    74.5 (56-86)      58.9 (50-66)        32.3 (28-38)
A. pectorosa        113.5 (106-130)   69.8 (62-78)        45.3 (43-53)
V. iris             39.2 (32-47)      24.0 (20 -28)       13.4 (11-16)

Species             Area (b)        Tissues % Wt (c)    Fluid % Wt (c)

A. pectorosa        65.8 (41-94)    14.5 (10.4-20.0)    6.1 (3.4-8.4)
L. dolabelloides    20.6 (9-33)     11.6 (8.8-13.8)     4.2 (2.8-7.0)
A. pectorosa        75.8 (65-92)    11.5 (9.0-14.3)     10.2 (6.4-12.4)
V. iris             4.5 (2.8-6.5)   22.8 (18.4-26.5)    6.1 (3.7-9.9)
A. pectorosa        76.9 (66-89.5)  12.4 (8.3-13.7)     7.5 (5.2-11.6)
L. dolabelloides    24.6 (12.5-35)  10.7 (6.9-17.7)     12.3 (9.3-14.7)
A. pectorosa        72.8 (48-110)   14.3 (12.4-16.3)    7.5 (5.7-8.3)
V. iris             3.8 (1-7)       21.0 (13.5--26.3)   6.2 (2.5- 11.8)

(a) Total weight (g): length, width, depth (mtn).

(b) Area equals the volume of water held; 1 mL - 1 cc.

(c)  Weight (g) of soft tissues or volume (mL) of fluid expressed
as a percent of the total weight: tissues % = w/w; fluid % = v/w.

TABLE 2.
Comparisons of morphometric data from three species
of mussels to determine significant differences
(P values; a = 0.05) because of river of origin
(Holston versus Clinch) or collection date (summer
versus fall).

Pairing (a)          Factor   Weight   Length    Width

Apect-Ho-Su versus
  Apect-Cl-Su         Site     0.108    0.524     0.379
Apect-Ho-Su versus
  Apect-Ho-Fa         Date     0.054    0.001     0.007
Apect-Cl-Su versus
  Apect-Cl-Fa         Date     0.760    0.030     0.416
Apect-Ho-Fa versus
  Apect-Cl-Fa         Site     0.006    0.087     0.164
Ldol-Ho-Su versus
  Ldol-Ho-Fa          Date     0.599    0.858     0.531
Viris-Cl-Su versus
  Viris-Cl-Fa         Date     0.624    0.710     0.800

                                       Tissue    Fluid
Pairing (a)          Depth     Area     % Wt      % Wt

Apect-Ho-Su versus
  Apect-Cl-Su         0.368    0.103    0.010   < 0.001
Apect-Ho-Su versus
  Apect-Ho-Fa         0.020    0.058    0.109     0.116
Apect-Cl-Su versus
  Apect-Cl-Fa         0.025    0.624   -0.001     0.002
Apect-Ho-Fa versus
  Apect-Cl-Fa         0.140    0.487    0.033     0.927
Ldol-Ho-Su versus
  Ldol-Ho-Fa          0.237    0.234    0.400   < 0.001
Viris-Cl-Su versus
  Viris-Cl-Fa         0.437    0.309    0.275     0.911

(a) Apect: Actinonaias pectorosa, Ldol: Lexingtonia
dolahelloides, Viris: Villosa iris, Ho: Holston River,
CI: Clinch River, Su: summer collection, Fa: fall collection.

TABLE 3.
Total bacteria recovered (cfu/g of tissues or cfu/mL of
fluid) from Actinonaias pectorosa, Lexingtonia dolabelloides
and Villosa iris from the Holston and Clinch rivers, VA.
Means and ranges (in parentheses) for ten animals of each
species from each collection date (summer versus fall).

                       Origin        Date

A. pectorosa           Holston       Summer
L. dolabelloides       Holston       Summer
A. pectorosa           Clinch        Summer
V. iris                Clinch        Summer
A. pectorosa           Holston       Fall
L. dolabelloides       H01ston       Fall
A. pectorosa           Clinch        Fall
V. iris                Clinch        Fall

                    Tissues cfu/g BHIA (a)

A. pectorosa        6.39 x [10.sup.5] (5.29 x
                    [10.sup.4] -2.03 x [10.sup.6])

L. dolabelloides    2.40 x [10.sup.5] (8.80 x
                    [10.sup.4] -5.87 x [10.sup.5])

A. pectorosa        2.64 x [10.sup.6] (5.81 x
                    [10.sup.5] -1.16 x [10.sup.7])

V. iris             2.41 X [10.sup.6] (1.57 x
                    [10.sup.5] -1.15 x [10.sup.7])

A. pectorosa        1.72 x [10.sup.6] (4.26 x
                    [10.sup.4] -6.40 x [10.sup.6])

L. dolabelloides    1.34 X [10.sup.6] (6.14 x
                    [10.sup.4] -5.34 x [10.sup.6])

A. pectorosa        1.77 x [10.sup.6] (3.74 x
                    [10.sup.3] -1.44 x [10.sup.5])

V. iris             2.29 x [10.sup.6] (2.66 x
                    [10.sup.5] -5.86 x [10.sup.6])

                    Tissues cfu/g R2A (a)

A. pectorosa        4.34 x [10.sup.5] (1.03 x
                    [10.sup.5] -9.36 x [10.sup.5])

L. dolabelloides    2.38 x [10.sup.5] (6.13 x
                    [10.sup.4] -6.40 x [10.sup.5])

A. pectorosa        2.77 x [10.sup.6] (5.93 x
                    [10.sup.5] -1.38 x [10.sup.7])

V. iris             3.55 x [10.sup.6] (1.52 x
                    [10.sup.5] -1.52 x [10.sup.7])

A. pectorosa        1.62 x [10.sup.6] (8.54 x
                    [10.sup.4] -6.66 x [10.sup.6])

L. dolabelloides    1.17 x [10.sup.6] (6.14 x
                    [10.sup.4] -5.06 x [10.sup.6])

A. pectorosa        2.45 x [10.sup.5] (1.60 x
                    [10.sup.3] -2.14 x [10.sup.6])

V. iris             3.50 x [10.sup.6] (2.94 x
                    [10.sup.5] -1.04 x [10.sup.7])

                    Fluid cfu/ml BHIA

A. pectorosa        1.02 x [10.sup.5] (6.00 x
                    [10.sup.3] -2.80 x [10.sup.5])

L. dolabelloides    9.69 x [10.sup.4] (5.33 x
                    [10.sup.3] -5.20 x [10.sup.5])

A. pectorosa        8.60 x [10.sup.5] (2.20 x
                    [10.sup.5] -3.20 x [10.sup.6])

V. iris             1.63 x [10.sup.5] (4.27 x
                    [10.sup.4] -5.47 x [10.sup.5])

A. pectorosa        2.00 x [10.sup.5] (3.20 x
                    [10.sup.4] -7.07 x [10.sup.5])

L. dolabelloides    9.80 x [10.sup.4] (3.20 x
                    [10.sup.3] -3.33 x [10.sup.5])

A. pectorosa        2.92 x [10.sup.4] (3.20 x
                    [10.sup.3] -6.27 x [10.sup.4])

V. iris             3.81 x [10.sup.5] (4.93 x
                    [10.sup.3] -1.33 x [10.sup.6])

                    Fluid cfu/ml R2A

A. pectorosa        5.43 x [10.sup.4] (6.67 x
                    [10.sup.3] -2.27 x [10.sup.5])

L. dolabelloides    7.27 x [10.sup.4] (5.07 x
                    [10.sup.3] -4.60 x [10.sup.5])

A. pectorosa        7.16 x [10.sup.5] (1.40 x
                    [10.sup.5] -1.80 x [10.sup.6])

V. iris             2.28 x [10.sup.5] (3.20 x
                    [10.sup.4] -6.40 x [10.sup.5])

A. pectorosa        5.55 x [10.sup.5] (3.33 x
                    [10.sup.4] -4.80 x [10.sup.6])

L. dolabelloides    5.59 x [10.sup.4] (1.07 x
                    [10.sup.4] -8.00 x [10.sup.4])

A. pectorosa        6.55 x [10.sup.4] (2.40 x
                    [10.sup.3] -3.47 x [10.sup.5])

V. iris             4.21 x [10.sup.5] (1.60 x
                    [10.sup.4] -1.60 x [10.sup.6])

(a) Media for primary bacterial isolation. BHIA, brain heart
infusion agar; R2A. R2A agar.

TABLE 4.
Comparisons of total bacterial cfu/mL recovered from three species
of mussels to determine significant differences (ANOVA;
[alpha] = 0.05) because of river of origin (Holston versus
Clinch) or collection date (summer versus fall).

Pairing (a)                      Factor   Tissues (b)       Fluids (b)

Apect-Ho-Su versus Apect-Cl-Su   River    0.007 (Clinch)    <0.001
                                                              (Clinch)
Apect-Ho-Su versus Apect-Ho-Fa   Date     0.019 (Fall)      0.206
Apect-Cl-Su versus Apect-Cl-Fa   Date     0.002 (Summer)    <0.001
                                                              (Summer)
Apect-Ho-Fa versus Apect-Cl-Fa   Riser    0.003 (Holston)   0.161
Ldol-Ho-Su versus Ldol-Ho-Fa     Date     0.002 (Fall)      0.829
Viris-Cl-Su versus Viris-Cl-Fa   Date     0.949             0.180

(a) Apect: Actinonaias pectorosa, Ldol: Lexinglonia dolubelloides.
Viris: Villosa iris. Ho: Holston River, Cl: Clinch River, Su:
summer collection date, Fa: fall collection date. Bacterial cfu/mL
recovered using BHIA and R2A, combined (i.e.. n = 20 versus 20),
were evaluated by ANOVA.

(b) p value (collection site or date having the greater mean
bacterial cfu).

TABLE 5.
Bacteria isolated from summer and fall collections of Actinonaias
pectorosa, Lexingtonia dolabelloides and Villosa iris from the
Holston (H) and Clinch (C) Rivers, VA.

Acinetobacter junii/johnsonii H.C,A. lwoffii H, A. radioresistens C
(motile) Aeromonas spp. i.e., A. hydrophila, A. caviae H,C
Brevundimonas vesicularis H,C
Chryseobacterium indologenes H,C
Chryseomonas luteola C
Citrobacter koseri C
Comamonas testosteroni H
Enterobacter intermedius H
Flarobacterium columnare C
Gram-positive rods. e.g., Corynebacterium-like H,C
Hafnia alvei C
Klebsiella spp. H
Moraxella spp. H,C
Ochrobactrum sp. C
Pantoca spp. H,C
Pasturella spp. H,C
Plesionronas shigelloides H,C
Proteus vulgaris H
Providencia rettgeri H
Pseudomonas fluorescens H,C
Ralstonia pickettii H
Serratia fonticola H,C,S. liquefaciens H, S. marcescens H,C
Shewanella putrelaciens C
Sphingobacterium multivorum H
Sphingomonas paucimobilis H,C
Stenotrophomonas maltophilia H
Gale Copyright: Copyright 2008 Gale, Cengage Learning. All rights reserved.