Water was collected from three sources in Townsville, North Queensland: deionized water, tap water collected from the chlorinated reticulated supply, and lake water from a 1-hectare lake in the suburb of Hyde Park. All water was then autoclaved at 121°C for 15 min. In duplicate 25-cm² culture flasks (TPP, CSL Biosciences Ltd., Australia) containing each of the three water types, we added approximately 1 x 10⁵ spores/mL of 98-1810/3 of B. dendrobatidis, obtained from a wild adult of Nyctimystes dayi from Tully, Queensland, in 1998, or B. dendrobatidis strain 98-1469/10 isolated from a captive juvenile Limnodynastes dumerilii from the Amphibian Research Centre, Melbourne, Victoria, in 1998. Strains have been maintained by serial passage in TGhL broth (16 g tryptone, 4 g gelatin hydrolysate, 2 g lactose in 1,000 mL distilled water) approximately every week. Flasks containing water were held at 23°C in an incubator and observed during a 10-week period by using an inverted Olympus microscope (Olympus Optical Co., Ltd., Tokyo, Japan). Evidence of growth included attachment of inoculated zoospores, change in size and form of zoosporangia, and release of new zoospores into the water. Signs of viability included movement of free zoospores in the water or movement of zoospores inside the zoosporangia. From weeks 2 to 7, when activity of zoospores was no longer visible and growth appeared to have ceased, 0.5 mL of water from each flask was added into flasks containing TGhL media and incubated at 23°C. These newly inoculated cultures were then observed for growth and activity, with the observations terminating at week 10.

Figure

Figure. Zoosporangia of strain 98-1810/3 visible as transparent spherical bodies growing in lake water on (a) freshwater arthropod and (b) algae. Bars = 30 μM.

The duration of survival of B. dendrobatidis in water varied with the source of the water and the strain of fungus (Table). For all three water samples, zoospores attached to the plastic of the flasks and grew into zoosporangia, but new zoospores appeared to be released only into the lake water. After 1 week in tap and deionized water, no further growth of zoosporangia or activity by zoospores was apparent. However, viability of B. dendrobatidis was demonstrated in tap water until week 3 by growth of both strains on addition into culture media. For deionized water, viability of strains differed with growth on inoculation into culture media up to week 3 and week 4 for 98-1810/3 and 98-1469/10, respectively. The lake water contained dead microscopic algae, arthropods, protozoa, and plant debris, and zoosporangia were attached to these organic bodies as well as to the plastic of the flask (Figure). Growth of zoosporangia was apparent at week 1, and motile zoospores of both strains were present in lake water cultures for 1 to 7 weeks. However, during this period, no activity of zoospores was apparent at times. Subcultures from lake water into TGhL broth showed viability to week 6 and week 3 for the strains 98-1810/3 and 98-1469/10, respectively.

Zoospores of B. dendrobatidis were highly active when placed into all three types of water and progressively attached to the flask to form sessile zoosporangia. With routine culture in nutrient media, zoosporangia release zoospores into the medium mature as they mature. This cycle continues until the culture becomes overgrown and inhibition of growth occurs, possibly as nutrients become limiting or metabolic waste products accumulate. One cycle of normal growth for both these strains of B. dendrobatidis takes up to 5 days at optimum temperatures and nutrient conditions. However, in low nutrient conditions, such as exists in tap, lake, and distilled water, growth is inhibited and the life cycle takes longer to complete. To allow for a slower life cycle, observations were continued for 10 weeks. In our study zoospores developed to zoosporangia in all three water types, but new zoospores were not released in two of the three water types, tap water, and deionized water. Previous research had indicated that B. dendrobatidis did not survive well in distilled water (11). However, this study confirmed that the absence of activity by zoospores, or growth of new zoosporangia, is not a reliable indicator of a nonviable or dead culture. Zoosporangia may be able to survive in a state of arrested or nondiscernible development for a long period in environments not conducive to growth, and on inoculation into nutrient broth, the life cycle may then recommence. Results from inoculation of TGhL broth showed that B. dendrobatidis in tap water remained viable for 3 weeks. In deionized water strain 98-1810/3 survived for 3 weeks and strain 98-1469/10 survived 4 weeks. B. dendrobatidis in lake water was more successful: both strains were able to release new zoospores, or less likely, the original zoospores were capable of surviving for extended periods.

We presume that the longer survival time of B. dendrobatidis in lake water could be due to the higher level of nutrients and possibly the nonliving organic substrate offered by algae and other microorganisms. Attachment and growth of zoosporangia on skeletons of algae and invertebrates support this theory. However, no evidence of digestion of these bodies by the chytrid was visible. Repeated observations (Figure a) did not show any appearance of degradation. We do not know the effect of these microorganisms, if alive, on the growth and survival of B. dendrobatidis. Autoclaved water was used because of the limitations of our current culture system. Although active zoospores of both strains were observed in lake water cultures after 7 weeks, no growth occurred when subcultured into TGhL. Low numbers of zoospores were present. Therefore, when aliquots were removed to inoculate TGhL broth, zoospores may not have been included. Alternatively, zoospores may have been subject to osmotic shock when transferred from water to broth. B. dendrobatidis grows best when clusters of zoospores form (13), and too few may have survived inoculation to successfully establish a new culture.

These results have immediate relevance for disease control and quarantine strategies. Water in contact with amphibians should be regarded as contaminated with B. dendrobatidis for up to at least 7 weeks after last contact with the amphibian. For quarantine purposes, all water, moist soil, and wet fomites imported into a country with amphibians should be regarded as infectious for B. dendrobatidis unless the amphibians are shown to be uninfected. A similar strategy should be adopted when introducing new amphibians into a captive colony or collection. Similarly, water and any items coming into contact with amphibians moved within countries should be regarded as infectious for B. dendrobatidis. In practical terms, storage alone for a period of time should not be used as a means of ensuring water that has been in contact with an amphibian is not contagious. All water and wet soil in contact with an amphibian should be disinfected before discharge into the wastewater system or the natural environment. Amphibians should not be placed into enclosures with water used previously by other amphibians without prior disinfection. Any other wet objects that have been in contact with amphibians should either be disposed of or disinfected before repeat use. Various disinfection strategies have been described (M. L. Johnson et al, unpub. data). The most effective strategies for disinfection are heat (>47°C for 30 min), didecyl dimethyl ammonium chloride at >0.0012% final concentration for 2 min, or sodium hypochlorite (>1% for 1 min). To comply with the intentions of Office Internationale des Epizootes listing, amphibians, when moved between countries, should be placed in a different container on arrival; all water, soil, plants, and litter in contact with the amphibian during transport should be adequately disinfected by using techniques capable of killing B. dendrobatidis.