Batch production: The enzyme production was carried out under batch, repeated batch and packed bed fermentation conditions. The production media for immobilized cells of Bacillus circulans consisted (g l^-1) of cotton deoiled meal (CDM), 15.0; Yeast Extract, 5.0 and salt solution (gl^-1) of K₂HPO₄, 1.0; MgSO₄.7H₂O, 0.20 and Na₂CO₃, 10.0 with a pH of 10 (Jaswal and Kocher, 2008). The effect of different cultural variables (agar agar/ calcium alginate concentration, bead size and inoculum size) was optimized under batch conditions. In case of packed bed experiemts, column length and flow rate were optimized.

The agar-agar and calcium alginate entrapped cells were used as inoculum for alkaline protease production medium (50ml in 250ml capacity erlenmeyer flasks). The flasks were incubated at 28°C for 120 h. Samples were withdrawn at regular intervals of 24 h and assayed for cell leakage and alkaline protease activity.

Production by repeated batch process: The reusability of Bacillus circulans entrapped cells was examined in shake flasks. After attaining the maximum production of alkaline protease, the spent medium was replaced with fresh production medium (50 ml) and the process was repeated for several batches until the enzyme activity remained constant.

Packed bed production: Burettes (50ml capacity) were used as column reactors to which beads were filled upto different lengths (5-30 cm) to optimize the column length. The packed columns were filled with enzyme production medium upto 50 ml mark. The column were activated at room temperature for overnight (16 h), followed by flushing the spent production medium and fresh medium was then poured over the beads. The flow rate was regulated and spent medium was collected every hour and used as crude enzyme extracts. The spent medium was replaced with fresh medium continuously.

All the enzyme production experiments were planned statistically using Random Block Design where atleast three replications of each treatment were taken.

Extracellular Protease Activity: The protease activity of the periodic samples was determined in a reaction mixture consisting of 0.1 ml of crude enzyme, 2 ml of 0.5% azocasein (in carbonate-bicarbonate buffer, 0.1M, pH 9.5) and 0.9 ml of distilled water and incubated at 60°C for 15 minutes. The proteins were precipitated out thereafter by adding 3 ml of 5% ice-cold trichloroacetic acid (TCA) and free amino acids released by crude protease from casein hydrolysis were estimated by Lowry’s method. The protease activity was defined as Mol of tyrosine released per minute per ml of crude enzyme.

Study of cell leakage: The cell leakage from the production media were determined by viable cell plate count method. Periodic samples were serially diluted and plated on inoculum medium and incubated at 28±2°C for 48 h for determining the viable cell count.

Scanning Electron Microscopy: The B. circulans cells immobilized in calcium alginate and agar agar were examined for their distribution pattern on the immobilizing material surface. The samples were fixed with 2% glutaraldehyde and dehydrated in graded series of ethanol. Finally, the samples were dried by critical point drier, coated with gold and observed on a Scanning Electron Microscope (Hitachi S-3400 N, Japan) under different magnifications.

Effect of bead size: The effect of bead size was studied by approximately 150 beads each of agar agar and calcium alginate. And the data was fitted into TWO WAY ANOVA classification. The results presented in Table 1 reveal a better protease yield with low bead size of 6.25, 5.2 mm³ having statistically significant values of 16.6 and 18.4 IU having relative activity of 115.9 and 118.8% over free cells in case of agar agar and calcium alginate entrapped cells respectively. The better performance of entrapped cells over free cells has also been reported elsewhere (Beshay 2003, Adinaryana et al, 2005) and small sized beads have more protease activity due to higher surface area (El-Katany et al, 2003).

Effect of entrapment material concentrations: The concentration of entrapment material is an important parameter and the concentration range of 2-4% studied in this study revealed 2% concentration of agar agar and 3% concentration of calcium alginate as the best for protease yield of 18.2 and 19.5 IU respectively (Table 1). Further increase in concentration of entrapment material decreased activity that may be attributed to diffusion limitation arising out of reduced pore size of beads at increased concentration of entrapment material (Elibol and Moreria ,2003; Rao et al, 2008). It is important to mention that both entrapment materials used behaved similarly as the relative protease activity improved to 127.3 and 125.5 respectively. Rao et al

Figure 1

Table 1. Effect of cultural and fermentation conditions on alkaline protease production by immobilized cells of

(1986) infact reported a better ethanol production with agar immobilized cells than alginate and -carrageenan.

Effect of inoculum size: The effect of inoculum size studied by varying the number of beads in production experiments showed increase in protease activities with increase in number of beads/flask at all the 5 levels (40-200 beads/flask) tested. While a significantly higher value of 19.9 IU was obtained with 200 beads/flask of agar agar, a maximum protease activity of 19.7 IU was achieved with similar number of calcium alginate beads. Thus, by batch production the protease activity was improved by 139.2 and 128.5 % over free cells in case of agar agar and calcium alginate beads respectively. Elsewhere, Beshay (2003) and Rao et al (2008) reported 400 and 180 beads/flask of alginate beads for optimum protease production by T.turnirae and B. circulans respectively.

The reusability of beads was studied to determine their strength and performance. The results presented in Fig.1 reveal that agar agar beads behaved consistently over 9 consecutive cycles while calcium alginate beads were stable for 4 cycles only before the respective beads started disintegrating. The low stability of calcium alginate beads is attributed to the presence of potassium phosphate in the enzyme production medium that tends to dissolve the beads (Bajpai and Sharma, 2004). On the other hand, agar agar beads had no effect of this salt. Elsewher, Beshay (2003) and Elibol and Moreira (2003) reported maximum alkaline protease production in 3rd, and 4th cycle while Adinarayana et al (2005) and Rao et al (2008) reported 9 and 11 repeated batch cycles with alginate entrapped cells. Another observation in repeated batches was that maximum enzyme production was decreased from 100h with free cells to 48h in 4^th cycle in agar agar and 40h in 3^rd cycle in alginate beads (unpublished data). The cell leakage was studied from the periodic samples revealed a mean loss of 28910⁴ cfu/ml at the end of respective cycles.

Figure 2

Fig.1 Alkaline protease production by entrapped in agar-agar and calcium alginate under repeated batch conditions

Earlier workers have suggested use of immobilized cells for continuous production (Beshay, 2003 and Adinaryana et al, 2005). As the reusability of beads confirmed consistent protease production over successive cycles an effort was made to improve protease production and provide a more handy alkaline protease production by using continuous fermentation. Among the different methods of continuous fermentation, packed bed reactors have become very popular in recent years due to their low manufacturing and operating costs and also due to the ease of process automation in these reactors (Goksungur and Zorlu, 2001). In the present study, we employed plastic burrettes of 50ml capacity as columns for packing beads as explained earlier in materials and methods. Incubation time, column/bed length and flow rate were the 3 important parameters optimized for packed bed operation. The incubation time was reduced to 36h in both agar agar and calcium alginate beads (Table 1) and the relative protease activity was improved to 165.7 with agar agar beads and 123% with alginate beads compared to free cells. Although, to the best of our knowledge, there are no reports of continuous alkaline protease production by immobilized cells, Adinaryana et al (2007) reported a high CGTase production by Bacillus sp. under continuous conditions than batch fermentation.

A column/packed bed length of 30 cm was found to be optimum resulting in an overall increase in relative activity of 165.7 and 144.9% with agar gar and alginate entrapped cells respectively (Table 1). An optimum time of 36h and bed length of 30 cm used to standardize flow rate revealed a recovery of 10ml/h crude enzyme having 28 and 27 IU respectively with agar agar and alginate beads packed beds (Table 1). Thus, by using packed beds we were able to further enhance the protease activity by about 50% and reduce the production time from 100 to 36h. The study also revealed use of agar agar as an alternate equally effective entrapment material in place of calcium alginate the beads of which dissolve over the period of time in phosphate containing media.

The surface structure of entrapped B. circulans cells were studied using SEM. Both calcium alginate as well as agar-agar beads depicted a random distribution of B. circulans rods or vesicular shaped cells on the surface though cells appeared more as vesicles in agar agar beads (Fig. 2b, 2d). There were pockets of high cell density at a scale of 10 µm in agar agar and alginate beads that were the areas of cell growth, (Fig 2a, 2c) though in case of alginate beads, cells were more relatively uniformly distributed on the surface (Fig 2c). Thus, both in alginate and agar agar beads, the distribution of cells showed almost similar clustering in the areas of growth in the furrows of beads as has been discussed earlier by Rao et al (2008). To the best of our knowledge, we have not come across of any report of SEM of alkaline protease producing agar agar entrapped cells. Though in literature, SEM of alginate entrapped cells is available showing similar observations of random cell distribution (Beshay 2003, Pai et al 1995, Rao et al, 2008).

Figure 3

Fig 2. SEM observation of calcium alginate (Fig 2a, 2c) and agar-agar entrapped cells of (Figs 2b, 2d). 2a, 2c, 10 m and 2b, 2d, 5m.

The author are thankful to Dr. (Ms.) Rabinder Kaur of Nanoscience Laboratory, PAU, Ludhiana for help during Scanning Electron Microscopy (SEM) studies.