Studies On The Occurrence And Diversity Of Bacteria In Fish Culture Systems In Ghana With Special Reference To Species Pathogenic To Fish And Humans

ABSTRACT

The occurrence and diversity of bacteria in selected fish culture ponds from three regions

in Ghana, Aduabenba Farms, Agyeman Farms, Aheto Farms, Akuse Kpong Farms,

ARDEC Station, Asare Farms, Boadi Farm, Boahen Farms, Boateng Farms, Frimpong

Farms, K.K. Farms, Pacific Farms and Sagoe Farms and three open systems, Kpong Head

Pond, Volta River and Weija Dam were studied over a period of three years, 1996 - 1999.

Bacterial populations from the different fish culture systems studied consisted essentially

of about the same bacterial species, but the predominating species varied with each culture

system. Subtyping of the bacterial strains recovered from the different environments, and

comparing the species present was a valuable tool to determine the correlations between

the flora of the different environments. Since the studies involved large numbers of

isolates, simple laboratory methods combined with automated data evaluation and

presentation were used.

The Phene-Plate (PhP) system for biochemical finger printing of bacteria, which is based

on measurements of the kinetics of biochemical tests, performed in microplates, was

employed. The system included mathematical models to calculate the diversity (Di) of the

bacterial flora within each of the different environments, as well as the similarities

between bacterial populations in the different environments as the population similarity

coefficient (Sp).

The PhP system was used to type 80 colonies at a time from each of the cultured systems,

open systems and the sewage fed ponds. The PhP system had the advantage of calculating

the diversity index of the bacterial population present in each of the farm, as well as

calculating the similarity coefficient between the population of bacteria in the different

environments. From each one of the six different cultured ponds receiving different types

of fertilizers, three rivers representing three open systems and the four serial sewage-fed

ponds, a total of 7990 isolates were typed.

The results indicated that the diversity among the bacterial populations from the fish

culture systems were generally high with mean diversity above 0.95 in all cases. Mean Di

iv

for the cow manure-fertilized ponds was 0.977, the mean for poultry manure-fertilized

ponds was 0.981, and that for pig manure-fertilized ponds was 0.952. The rest were,

blood waste-fertilized ponds, 0.989; chemically fertilized ponds, 0.972; ponds with no

fertilization, 0.989; the open systems, 0.950; and the sewage-fed ponds, 0.971. These

results indicated that the bacterial populations were very high in each community and

consisted of many different bacterial types or strains.

Comparison of results obtained for the populations of bacteria present in the different

fertilization type ponds to that of the open systems showed mean population similarity

(Sp) coefficient values below 0.50. This was an indication of different populations with

high diversities that were similar in their pattern of distribution. Similarly, comparison of

the results of all the serial ponds of the sewage-fed ponds showed mean population

similarity (Sp) coefficient with values below 0.50. Comparison of the results of ponds I

and III, however, gave Sp value above 0.50, indicating that there were certain periods of

sampling when the populations were not related. Biochemical and morphological tests, as

well as the API 20NE and 2IE, were used to identify some selected bacteria picked

randomly from the plates and used as the reference data for the PhP procedure.

Twenty-five species of bacteria were identified as associated with the fish culture systems

in this study. The identified bacteria included one genus of spiral and curved bacteria,

Campylobacter sp.; one genus of Gram-negative aerobic rod, Pseudomonas sp.; sixteen

genera of Gram-negative facultative anaerobic rods, Actinobacillus sp., Aeromonas sp.,

Citrobacter sp., Edwardsiella sp., Enterobacter sp., Escherichia sp., Flavobacterium sp.,

Hafnia sp., Klebsiella sp., Pasteurella sp., Proteus sp., Salmonella sp., Serratia sp.,

Shigella sp., Vibrio sp. and Yersinia sp.; one Gram-negative anaerobic bacterium,

Bacteroides sp.; three Gram-positive cocci, Micrococcus sp., Staphylococcus sp. and

Streptococcus sp.; two endospore-forming rods, Bacillus sp. and Clostridium sp.; and one

Actinomycete, Corynebacterium sp.

Pseudomonas sp. was the most dominant species of most ponds; cow manure-fertilized

ponds, blood waste-fertilized ponds, chemically fertilized ponds, sewage-fed ponds, ponds

with no fertilization and in the open systems. Poultry manure-fertilized ponds and pig

manure-fertilized ponds had Salmonella sp. and Streptococcus sp., respectively, as the

most dominant species.

V

Physico-chemical parameters found to have positive correlation with the bacterial

populations in the fish culture systems were acidity, air temperature, alkalinity,

ammonium, biochemical oxygen demand, calcium ion, chloride, conductivity, magnesium

ion, pH, phosphate ion, silicon dioxide, sulphate ion, suspended solids, total dissolved

solids, total hardness, turbidity and water temperature.

Fish from the various culture systems harboured, bacteria belonging to the 25 genera

isolated from the different culture systems in the blood, gut and muscles and on the gills

and skin. Generally, higher bacterial populations were associated with the gills, gut and

skin than the blood and muscle. Although Pseudomonas sp. was most abundant species in

the sewage-fed ponds and was present in considerable quantities in the five fish tissues,

Salmonella sp. was the most important isolate of the gills, muscle and skin. Pathogenic

bacteria were isolated from inhabitants of communities associated with ponds that

received all sorts of fertilizers and even with ponds that were not fertilized.

All the organic manures mostly used to fertilize fish ponds, cow manure, poultry manure,

pig manure, blood waste and sewage water, supported growth to varying degrees of eight

experimentally introduced pathogenic bacteria, Pseudomonas sp. KI-MTC-001K, Shigella

sp. KI-MTC-002K, Enterobacter sp. KI-MTC-003K, Klebsiella sp. KI-MTC-004K,

Citrobacter sp. KI-MTC-005K, Proteus sp. KI-MTC-006K, Salmonella sp. KI-MTC-007K,

and Vibrio parahaemolyticus KI-MTC-008K. The best growth was recorded in the

Pseudomonas sp. KI-MTC-001K tests and the poorest in the Vibrio parahaemolyticus KIMTC-

008K tests.

Resident bacteria were detected in all eleven feed types commonly used in feeding the

fish. Highest count of 4.28 logio CFU g'1 heterotrophic bacteria count of feed was

recorded for fufu waste while banana waste recorded the least count of 2.85 logio CFU g"1.

Enormous quantities of heterotrophic bacteria, from 6.17 to 6.86 logio CFU g"1, occurred

in blood waste, cow manure and poultry manure, used as organic manure.

The bacterial population of the water, and, level of bacterial contamination of the fish of

sewage-fed ponds was not remarkably different from those of ponds fertilized with blood

waste, cow manure, chemical fertilizer, pig manure and poultry manure, and, ponds and

open systems which were not fertilized. It was concluded that sewage fertilization could

serve two purposes. First, it could be a means of sewage disposal for small rural

communities, and secondly, increasing the potential yields of fish ponds.

If four or more linearly arranged and inter-connected ponds were prepared with slowflowing

water moving through the system, sufficient solanization would purify the water

to render the last pond bacteriologically safe for fish culture.