USE OF CASSIA ALATA AQUEOUS EXTRACT AS A BATH TREATMENT TO CONTROL PSEUDOMONAS ANGUILLISEPTICA INFECTION IN TILAPIA ( OREOCHROMIS NILOTICUS )

The aqueous extracts of six plants, Andrographis paniculata, Cassia alata, Centella asiatica, Garcinia mangostana, Punica granatum and Psidium guajava, were investigated for their antimicrobial activity and mode of action against Pseudomonas anguilliseptica, an important fish pathogenic bacterium, which is responsible for economic losses in aquaculture worldwide. Among the tested plant extracts, the C. alata aqueous extract had the strongest inhibitory effect and exhibited a bactericidal mode of action against the pathogenic bacterium. When an infection of tilapia (Oreochromis niloticus) with P. anguilliseptica was induced by intraperitoneal, the median lethal dose (LD50) was determined to be 1.59 x 10 5 CFU/ml. For the in vivo trial, four different concentrations (25, 50, 75 and 100 ppm) of C. alata aqueous extract were used as bath treatment to remedy the infection. The effect of the extract on the infection was dose-dependent and an extract with the concentration of 100 ppm eliminated mortality of the infected fish without producing any adverse effects on the animals. This study suggests that C. alata aqueous extract has the potential to control fish disease caused by P. anguilliseptica.

Additional signs include a red peritoneum, congested and hyperemic liver, a soft, swollen kidney, and dark, swollen spleen.
The most common treatment of Pseudomonas septicemia is the use of antibiotics.Many drugs, especially ciprofloxacin, have proved to be very useful in controlling the disease (Daly, 1999).However, the use of such therapeutic agents is problematic.Their accumulation in the environment and in fish can be potentially harmful to consumers and other organisms (Aldermann and Hastings, 1998).Furthermore, prolonged use of antibiotics will favor the development of bacteria resistant to antibiotics, thereby reducing drug efficacy (Aldermann and Hastings, 1998).With the increasing demand for organic aquaculture, researchers have been focusing on plants and plant products as a source of natural antimicrobial substances to be used in aquaculture.Many herbs have been shown to have the potential to control fish diseases caused by bacterial infections, including Stella aquatica, Impatiens biflora, Oenothera biennis, Artemisia vulgaris and Lonicera japonica (Shangliang et al., 1990), Rosmarinus officinalis (Abutbul et al., 2004), Psidium guajava (Adithepchaikarn et al., 2008), and Andrographis paniculata (Rattanachaikunsopon and Phumkhachorn, 2009).Because of their nontoxicity to fish farmers, consumers and the environment, as well as their recognized antimicrobial activity (Cowen, 1999), plants have been suggested as possible alternatives to antibiotics to control P. anguilliseptica infection in aquaculture.
The aim of this study was to investigate the potential of using a plant aqueous extract to control P. anguilliseptica infection in fish.The experiments were designed to examine the antimicrobial activity and mode of action of six plant species: Andrographis paniculata, Cassia alata, Centella asiatica, Garcinia mangostana, Punica granatum and Psidium guajava, against P. anguilliseptica.This work also examined the use of C. alata aqueous extract as a therapeutic agent against experimental P. anguilliseptica infection in tilapia.

Bacteria and culture conditions
The fish pathogen used in this study was P. anguilliseptica FT421, which was kindly donated by the Department of Fisheries, the Ministry of Agriculture and Cooperatives, Thailand.The bacterial pathogen was isolated from tilapia naturally infected by P. anguilliseptica in a farm in Ayutthaya Province, Thailand, in January 2004.The identity of the bacterium was confirmed by 16S rDNA analysis, following the protocol of Anzai et al. (2000).The type strain of the species P. anguilliseptica NCIMB 1949 T was also included in the identification test as a reference strain.Trypticase soy medium (agar and broth) was used to culture the bacterial strain.The growth temperature was 25ºC.The stock organism was kept in the culturing broth supplemented with 15% glycerol at -80 0 C.

Plant aqueous extracts preparation
Six plants were used in this study: A. paniculata, C. alata, C. asiatica, G. mangostana, P. granatum, and P. guajava.All were purchased from the herb shop "Ban Samunprai" in Bangkok, Thailand.They were sold with identification certificates approved by the Royal Forest Department, Bangkok, Thailand.The samples were dried at 70ºC for 72 h and sold as dried materials.They were subjected to aqueous extraction.The dried samples were finely ground and then thoroughly mixed with distilled water in a ratio of 1:10 (w/v) for 24 h at room temperature.The mixtures were centrifuged at 10000 g for 10 min at room temperature and the supernatants were collected for filtration.The filtrates were frozen at -80ºC for 48 h before being freeze-dried.The extracts were kept at -20ºC in glass vials until use.Before use, each herb extract was dissolved in distilled water and its concentration was adjusted to 10 mg (of dry weight)/ml.

Detection of antimicrobial activity
Each plant extract was tested for its antimicrobial activity against P. anguilliseptica using the swab paper Phumkhachorn and Rattanachaikunsopon disc method (Rattanachaikunsopon and Phumkhachorn 1998).The bacterial culture (10 8 CFU/ml) was spread with a sterile swab on trypticase soy agar before placing sterile filter paper discs, 6 mm in diameter (Schleicher & Schuell, USA) on the agar.Forty μl of each plant extract were spotted on the paper disc.Each extract was tested in five replicates.Ciprofloxacin (5 μg/disc) was used as a positive control while distilled water was used as a negative control.After incubation at 25ºC for 24 h, the diameters of the growth-free zones around the discs were measured and subtracted from the diameter of the paper disc, giving the sizes of inhibition zones beyond the paper disc.

Determination of minimal inhibitory concentration (MIC)
Each plant extract (10 mg/ml) was subjected to twofold serial dilutions using sterile distilled water as a diluent.Five hundred µl of each dilution were added into 4.5 ml of P. anguilliseptica culture (approximately 10 3 CFU/ml).Sterile distilled water (500 µl) was used as control.The mixture was incubated at 25ºC for 24 h before examination of bacterial growth by measuring the optical density at 600 nm (OD 600nm ).The MIC value of each extract (µg/ml) represented the lowest concentration of the extract showing inhibition of the tested organism.The MIC value of ciprofloxacin was also determined.Each MIC was determined from five experiments.

Examination of mode of action
To examine the mode of action of the plant extracts against P. anguilliseptica, 500 μl of each plant extract (at the final concentration equal to the MIC value) was added to 4.5 ml culture of P. anguilliseptica (approximately 10 3 CFU/ml).For the control, sterile distilled water (500 μl) was used instead of the herb extract.After incubation at 25ºC for 24 h, 100 μl of each mixture was inoculated into 4.9 ml of fresh trypticase soy broth and incubated at 25ºC.The OD 600nm of the culture was determined at the time of inoculation in the fresh culturing broth and every 3 h after incubation for 24 h.Each extract was tested in five replicates.

Preparation of experimental animals
Tilapia (O.niloticus) were purchased from "Nong Khon" Farm (Ubon Ratchatnai, Thailand).They were maintained in plastic tanks (150 l) for two weeks to acclimatize to laboratory conditions prior to the experiments.Throughout the acclimation period and subsequent experiments, fish were held at 25ºC with 12-h light/12-h dark photoperiod and fed 5% body weight twice a day with a commercial fish feed "C.P. CLASSIC" (S.W. T. Co., Ltd., Bangkok, Thailand).All experiments were conducted in 50 l aquaria.Fish weighing 10±1 g were stocked in the aquaria (10 fish per aquarium) 24 h prior to the experiments.

Median lethal dose (LD 50 ) determination
Groups of 10 fish with an average weight of 10±1 g were used to determine the LD 50 of P. anguilliseptica FT421 for tilapia.Each group of fish was kept in a 50-l aquarium at 25ºC throughout the experiments.Tenfold serial dilutions were prepared from a bacterial culture with concentration of 10 8 CFU/ml by using phosphate buffer saline as a diluent.One hundred μl of each dilution was injected intraperitoneally into each fish.The same volume of physiological saline was used instead of the bacterial suspension in the controls.Each dilution trial was performed in five replicates.Mortalities were recorded daily for 2 weeks.Dead fish were removed from the aquaria daily.Their livers and kidneys were aseptically streaked on trypticase soy agar.After incubation at 25ºC for 24 h, colonies grown on the agar were confirmed to be P. anguilliseptica as described earlier.The median lethal dose (LD 50 ) was calculated by the method of Reed-Muench (1938).

Examination of toxicity of C. alata aqueous extract
Groups of 10 fish were kept in aquaria containing 49.5 l of water.In five replicates, 0 (control), 25, 50, 75 and 100 ppm of aqueous extract of C. alata were added to aquaria.The plant extract was predissolved in 500 ml distilled water and then added to the aquaria.Mortality, appearance, feeding response and behavioral alterations of the fish were observed daily for 2 weeks.

Examination of the effect of C. alata aqueous extract against P. anguilliseptica infection
Groups of 10 fish maintained in aquaria containing 49.5 l of water were infected with P. anguilliseptica by intraperitoneal injection of 100 μl of a bacterial suspension at an LD 50 .Two days after fish were exposed to the bacterium, 500 ml of aqueous extract of C. alata were added to aquaria to achieve concentrations of 25, 50, 75 and 100 ppm.There were five replicate aquaria for each concentration.In this trial, groups treated with ciprofloxacin at the final concentration of 25 ppm were also included.Mortality, appearance, feeding response and behavioral alterations of the fish were observed daily for two weeks after fish were exposed to the bacterial pathogen.Dead fish were removed from the aquaria daily.Their livers and kidneys were subjected to bacterial isolation on trypticase soy agar.P. anguilliseptica isolated from the dead fish was confirmed by 16S rDNA analysis as mentioned above.Bacterial isolation was also performed with surviving fish as described above to confirm that they were free of P. anguilliseptica infection.

Antimicrobial activity of plant extracts
Antimicrobial activity of aqueous plant extracts against P. anguilliseptica was qualitatively determined by swab paper disc method and the results are presented in Table 1.All tested extracts, except G. man-gostana and P. granatum, were found to be able to inhibit the growth of the bacterium with different degree of inhibition.Based on the size of inhibition zone, the aqueous extract of C. alata gave the largest inhibition zone, whereas aqueous extract of P. guajava gave the smallest.

MIC of plant extracts
All of the aqueous plant extracts giving positive results in the swab paper disc method were subjected to the determination of MIC values against P. anguilliseptica.The results are shown in Table 1.There is, in general, a good agreement between the findings for the swab paper disc method and the MIC determination assay.Aqueous extracts of C. alata and P. guajava had the lowest and highest MIC values, respectively.

Mode of action of plant extracts
The mode of action of the plant extracts having inhibitory effect on P. anguilliseptica was examined and the results are presented in Fig. 1.After P. anguilliseptica cells inhibited with plant extracts were transferred to fresh trypticase soy broth, the cultures were examined for their absorbance at 600 nm at the time of transfer (time 0) and every 3 h for 24 h after transfer.P. anguilliseptica cells inhibited by A. paniculata and P. guajava extracts were found to be able to re-grow in the fresh culturing broth after transfer for 24 h, indicating the bacteriostatic mode of action of the plant extracts.However, the aqueous extracts of C. alata and C. asi- atica exhibited a bactericidal mode of action against P. anguilliseptica because the bacterial cells inhibited by the extracts could not resume their growth in the fresh broth 24 h after transfer.Growth of the bacterium was still undetectable at 72 h after transfer.

Median lethal dose (LD 50 ) of P. anguilliseptica
To determine the LD 50 of P. anguilliseptica for tilapia, intraperitoneal infections of the fish with different concentrations of the bacterial suspension were performed.Mortality was not observed in the negative control groups that were not exposed to the bacterium.The mortality of tilapia during the 2-week observation period after bacterial infection is shown in Fig. 2. The first dead fish were observed on day 2 or day 3 after exposure to P. anguilliseptica, depending on the applied dose, and no fish died after 10 days postexposure for all tested doses.P. anguilliseptica was isolated from the livers and kidneys of all dead fish.Based on the mortality, the calculated median lethal dose (LD 50 ) of P. anguilliseptica for tilapia was 1.59 x 10 5 CFU/ml.

Toxicity of C. alata aqueous extract
All concentrations of C. alata aqueous extract used in this study (25, 50, 75 and 100 ppm) were tested for toxicity on tilapia.They were shown not to have an adverse effect on the fish.During a 2-week observation period, the fish reared in the aquaria containing all concentrations of the herb extract were as normal as those reared in the control aquaria C. alata extract) in terms of appearance, behavior and feeding response.No mortality was detected in any of the experiments.

Effect of C. alata aqueous extract against P. anguilliseptica infection
The therapeutic effect of C. alata aqueous extract on tilapia experimentally infected with P. anguilliseptica is presented in Fig. 3.In this study, different concentrations of C. alata extract (25, 50, 75 and 100 ppm) and ciprofloxacin with a final concentration of 25   ppm were used as bath treatment 2 days after the fish were challenged with P. anguilliseptica at an LD 50 .The cumulative mortality of fish in control groups (without C. alata extract) in 2 weeks was about 50%.The dose-dependent reduction in mortality of bacterialinfected tilapia was observed in the groups treated with C. alata aqueous extract.No mortality was observed in the groups of fish reared in the aquaria containing the extract with the final concentration of 100 ppm, and the same result was observed in the groups treat with ciprofloxacin (final concentration of 25 ppm).Dead fish were removed from the aquaria every day and their livers and kidneys were subjected to bacterial isolation and identification.It was found that P. anguilliseptica could be isolated from all of the dead fish.However, at the end of the experiments, the bacterial isolation showed that surviving fish were free from the bacterial pathogen.

DISCUSSION
P. anguilliseptica is a bacterial fish pathogen causing Pseudomonas septicemia leading to massive mortalities of fish in the natural environment and fish farming, worldwide (Daly, 1999).Antibiotics are commonly used to control the disease.However, at the global level, people have understood the negative effect of antibiotics and they are now moving towards natural products.Due to their recognized antimicrobial activities, cost effectiveness and safety to fish farmers, consumers and the environment, plants have recently received increasing attention for use as alternatives to antibiotics to control bacterial infectious diseases in fish.In this study, we aimed to find a suitable plant to control P. anguilliseptica infection in tilapia.Because the affected fish usually lose their appetite, we preferred to use plant extracts as bath treatment against Pseudomonas septicemia instead of oral administration.Hence, only aqueous herb extracts, which readily dissolve in water, were investigated.Six plants (A. paniculata, C. alata, C. asiatica, G. mangostana, P. granatum and P. guajava) were selected for this study because they can be easily found throughout Thailand and have been reported to have antimicrobial activity (Sudaram et al., 1983;Jaiarj et al., 1999;Singha et al., 2003;Mamtha et al., 2004;Makinde et al., 2007;Rattanachaikunsopon and Phumkhachorn, 2007;Adithepchaikarn et al., 2008;Pradeep et al., 2008;Rattanachaikunsopon and Phumkhachorn, 2009).
The antimicrobial activities of aqueous plant extracts against bacteria generally vary depending on the sources of the extracts and strains of bacteria.For example, Streptococcus agalactiae was found to be more sensitive to the aqueous extract of A. paniculata than to that of P. guajava (Rattanachaikunsopon and Phumkhachorn, 2009), whereas Aeromonas hydrophila was more sensitive to the aqueous extract of P. guajava than to that of A. paniculata (Adithepchaikarn et al., 2008).Furthermore, a given bacterial strain responded differently to different aqueous plant extracts.A. hydrophila was inhibited by aqueous extracts of A. paniculata, G. mangostana and P. guajava but resistant to those of Alpinia galangal, Cymbopogon citrates, Momordica charantia and tinospora crispa (Adithepchaikarn et al., 2008).Of all six plant aqueous extracts tested in this study, C. alata aqueous extract exhibited the strongest antimicrobial activity against P. anguilliseptica, both in swab paper disc assay and MIC determination.
For using as a bath treatment, a plant with a bactericidal mode of action would be a better choice than one with a bacteriostatic mode of action because the fish pathogen inhibited by the latter may resume its growth and then start to cause the infection again.Therefore, it is of interest to examine the mode of action of the plant aqueous extracts found to have antimicrobial activity against P. anguilliseptica.Of all tested extracts, C. alata and C. asiatica extracts were found to exhibit a bactericidal mode of action, while A. paniculata and P. guajava extracts had a bacteriostatic mode of action.The results from this and previous studies indicate that each plant may have the same or different mode of action against different bacteria.P. guajava was found to have a bacteriostatic mode of action against P. anguilliseptica and many other species of fish pathogenic bacteria, including A. hydrophila, Aeromonas salmonicida subsp.salmonicida, Lactococcus garvieae, Streptococcus agalactiae, Vibrio salmonicida and Flavobacterium columnare (Rattanachaikunsopon and Phumkhachorn, 2007).However, C. asiatica had bactericidal effect on P. anguilliseptica but bacteriostatic effect on E. coli, Vibrio parahaemolyticus, Vibrio cholerae, Pseudomonas aeruginosa, S. typhimurium and A. hydrophila (Mamtha, 2004).
Based on its strong antimicrobial activity and bactericidal mode of action against P. anguilliseptica, the aqueous extract of C. alata was selected to test for its potential to control P. anguilliseptica infection in tilapia.C. alata, also known as Candle Bush or Ringworm Bush, is a shrub in the family Leguminosae.It is widely distributed in tropical countries, stretching from tropical America to India, Fiji, Indonesia and Malaysia.(Smith and Ali, 1979;Ibrahim and Osman, 1995).C. alata leaf has been reported to have a laxative effect and is used against ringworm, scabies, ulcers and skin diseases.It also possesses antibacterial and antifungal activities (Palanichamy and Nagarajan, 1990;Ibrahim and Osman, 1995;Khan et al., 2001;Makinde et al., 2007).For antibacterial activity, C. alata leaf was reported to be able to inhibit both Gram-positive and Gram-negative bacteria, including Actinomyces bovis, Bacillus cereus, Bacillus subtilis, Dermatophilus congolensis, Escherichia coli, Salmonella typhimurium, Staphylococcus aureus and Streptococcus faecalis (Khan et al., 2001;Makinde et al., 2007).The active constituents of C. alata were found to be alquinone, 3,5,7,4-tetrahydroxy flavone and flavonoid glycosides (Gupta, 1991;Yadav and Kalidhar, 1994;Rahaman et al., 2006).Although C. alata has never been used to control diseases in fish, the results from our in vitro studies and its effective medicinal applications encourage the in vivo trial of this plant in P. anguilliseptica-infected fish.
For use of C. alata aqueous extract as bath treatment against experimental P. anguilliseptica infection in tilapia, different concentrations of the extract (0, 25, 50, 75 and 100 ppm) were added into aquaria containing the diseased fish.All of the extracts were shown to be able to reduce the mortality rate of the infected fish in a dose-dependent fashion without adverse effect on the fish.Moreover, when C. alata aqueous extract was used as bath treatment at the final concentration of 100 ppm, no mortality of the diseased fish was observed.The same result was obtained from the groups treated with ciprofloxacin (25 ppm).These results suggest that the aqueous extract of C. alata has therapeutic potential against P. anguilliseptica infection and it may be used to replace antibiotics to control the disease.Intensive experiments on the use of C. alata aqueous extract in tilapia farms against natural P. anguilliseptica infection are underway in order to develop a suitable treatment for the disease in tilapia aquaculture.

Fig. 3 .
Fig. 3. Therapeutic effect of C. alata aqueous extract with different concentrations (ppm) and ciprofloxacin (25 ppm) on cumulative mortality of tilapia challenged with P. anguilliseptica at an LD 50 .Values are mean ±SD of five replicates.

Fig. 1 .
Fig. 1.Growth of P. anguilliseptica cells inhibited by plant aqueous extracts in fresh culturing broth.Values are mean ±SD of five replicates.

Fig. 2 .
Fig. 2. Cumulative mortality of tilapia in the first 2 weeks after intraperitoneal injection of different concentrations of P. anguilliseptica.Values are mean ±SD of five replicates.

Table 1 .
Antimicrobial activity and mode of action of plant aqueous extracts, ciprofloxacin and water against P. anguilliseptica.
a Results are mean ± S.D. values of five replicates.Phumkhachorn and Rattanachaikunsopon