MATERIALS AND METHODS

Collection of plant
Cordyline terminalis Kunth. (Liliaceae) was collected from the district of Narail during the month of January’ 2003 in its flowering stage and was identified by the National Herbarium of Bangladesh (accession No. 29752).

Extraction
The collected plant parts (leaves) were washed with water, separated from undesirable materials or plants or plant parts. They were sun-dried for one week after cutting into small pieces and were ground into a fine powder with the help of a suitable grinder (Capacitor start motor, Wuhu motor factory, China). The powder was stored in an airtight container and kept in a cool, dark and dry place until analysis commenced.

About 400 gm of powered material was taken in a clean, flat bottomed glass container (4l) and soaked in 1300 ml of 90% methanol. The container with its contents was sealed and kept for a period of 7 days accompanying occasional shaking and stirring. The whole mixture then underwent a coarse filtration by a piece of clean, white cotton material followed by a filtration through Whatmann filter paper and the filtrate thus obtained was concentrated by using a rotary evaporator (Bibby RE200, Sterilin Ltd., UK) to get the crude methanolic extract. 5 g of this extract was triturated with 100 ml methanol (90%) to prepare a solution of the crude extract. This was used as mother solution which was partitioned off successively by three solvents of different polarity as for example, n-hexane, acetone and chloroform (Rahman, 1994).

Test of antibacterial activity
Antimicrobial activity of the crude extract as well as different solvent fractions were determined by disk diffusion method (Bauer et al., 1966 and Ahmed et al., 2001).

Preparation of disks
Three types of disks were used for antibacterial screening.

Sample disks
Sterile filter paper disks (5 mm in diameter) were taken in a blank petridish. 5 μl of sample solution (prepared by dissolving 1 g of the extract in 10 ml of methanol) of the desired concentration (100 μgμl^-1) was applied on the disks with the help of a micropipette in an aseptic condition. These discs were left for few minutes in aseptic condition for complete removal of solvent.

Standard disks
These were used as positive control to ensure the activity of standard antibiotic against the test organisms as well as for comparison of the response produced by the known antimicrobial agent with that produced by test samples. In this investigation standard kanamycin (30 μgdisk^-1) disks(Oxoid, U.K.) were used as the reference.

Blank disks
These were used as negative control. They ensure that the residual solvents (left over the disks even after air drying) and the filter paper were not active themselves. 5 μl of methanol was applied on the sterile filter paper disk with the help of a micropipette and left for few minutes for complete removal of solvent.

Preparation of media
14 g dried Nutrient Agar Media (Oxoid, UK) was dissolved in 500 ml of distilled water and a clear medium was obtained by thorough shaking and heating in a water bath. The media was then sterilized in an autoclave at a temperature of 121°c and pressure of 15 lbs.sq-inch^-1 for 20 min.

Selection of the test organisms
Following bacteria were used as test organisms for the antibacterial activity test (Table 1).

Table 1:	List of bacteria used for the test

Preparation of the seeded test plates
16 ml of the sterilized medium was poured to each (sterilized) test tubes aseptically, under laminar air hood. Each of the test organisms was transferred from the subculture to the test tube with the help of the sterilized inoculating loop at 45°C under laminar air hood. The test tubes were shaken by rotation to get a uniform suspension of organisms. The bacterial suspensions were immediately transferred to the sterile petri-dishes and the petri-dishes were rotated several times, first clockwise and then anticlockwise, to assure homogeneous distribution of the test organisms to give a uniform layer of depth of approximately 4 mm. After the medium became cooled to room temperature, it was stored in a refrigerator (4°C) for 2 h.

Table 2:	Antibacterial activity of methanolic extract of C. terminalis

Table 3:	Antibacterial activity of different solvent extracts of C. terminalis
(--): No inhibition

All of the three disks( sample, standard and blank) were then placed in the seeded test plates using sterile transfer loop for antibacterial screening . The plates were then kept at 4-8°C facilitating maximum diffusion. The plates are then kept in an incubator at 37°C for 12-18 h to allow the growth of bacteria. The experiments were carried out more than twice and the mean of the reading were recorded.

RESULTS AND DISCUSSION

The methanolic extract of C. terminalis showed a moderate antibacterial activity against Escherichia coli( zone of inhibition 12 mm), Shigella boydii (14mm), Staphylococcus epidermis(12 mm) and Streptococcus pyogenes (13 mm) (Table 2). As it showed a moderate activity against E. coli and S. boydii and mild activity against S. boydii, S. dysenteriae and S. flexneri, the results support the traditional use of this plant as a remedy of diarrhoea, dysentery and systemic shigellosis. Table 3 shows the result of antibacterial activity of different solvent fractions of C. terminalis. The n-hexane soluble fraction showed a mild antibacterial activity against E. coli (8 mm), Salmonella typhi (9 mm), S. boydii (8 mm) and S. dysenteriae (9 mm) that also support its traditional use. The acetone and chloroform fraction did not show any significant antibacterial activity.