A time course study was performed under SSF using two substrates (wheat bran and saw dust) and by two white rot fungi (G. gibbosum and L. sajor – caju), in order to select the best substrate for further studies of polycyclic aromatic hydrocarbon (PAHs) degradation. Both strains shows variable amount of lignin, cellulose and hemicellulose degradation. Lignin, cellulose and hemicellulose loss on both the substrates was found to be slow initially but was found to increase gradually and then remaining more or less constant thereafter. When the lignin loss was maximum both the fungi exhibited maximum laccase activity. Maximum laccase production (10258 IU/g) was found on wheat bran by G. gibbosum and on saw dust by L.sajor – caju (8257IU/g). MnP production was found to be very low in G. gibbosum on wheat bran and Lentinus sajor – caju on both substrates. Maximum MnP (3.67 IU/g) was detected in Ganoderma gibbosum on 10th day of fermentation on saw dust. Maximum LiP (4.07 IU/g) was detected in Ganoderma gibbosum on wheat bran followed by Lentinus sajor – caju on wheat bran. Growth of fungi resulted into gradual degradation of lignocellulosic composition starting from 3rd day to the end of experiment. In the present study cellulase degradation by both the strains on saw dust was found to be 3%, and 2% on wheat bran starting from 3rd day to end of experiment (day 13). Degradation of hemicellulose was found to be maximum by L. sajor caju on saw dust which was around 5.5% at the end of experiment, followed by G. gibbosum on wheat bran (4%). For hemicellulose degradation L. sajor – caju turns to be more potent. Starting lignin content in saw dust was 38% and in wheat bran it was 11% (w/w) (Fig). The lignin content was reduced by fungal treatment to 31% (w/w) after treating with G. gibbosum and 32% after treatment with L. sajor-caju in saw dust (Fig). Further, 8% lignin left after treatment with by G. gibbosum and 9% by L. sajor-caju in wheat bran (Fig). Saw dust bio treated with G. gibbosum showed a slightly higher percentage of lignin loss (7%) than saw dust treated with L. sajor-caju (6%) (Fig). In case of wheat bran lignin loss was 3% and 2% by G. gibbosum and L. sajor-caju respectively, after 13th day of fermentation (Fig).
Laccase activity was found to be highest in Ganoderma gibbosum compared with Lentinus sajor – caju and was found to be apparent during colonization with lower level detected after 6th day. SSF conditions of wheat bran and saw dust has also been shown by both the strains. G.gibbosum on saw dust shows synergistic action of all three ligninolytic enzymes which is similar with the finding of Vares et al., 1995 who also reported synergistic mechanism of T. versicolor. Present study shows higher laccase activity between 5th – 6th days of lignin degradation and subsequently decreases to a steady level while lignin degradation rate remains constant. The low selectivity of G. gibbosum and L. Sajor-Caju during the fermentation stage showed that both species were selective lignin degrades. Slight decrease in lignin percentage was observed from 3rd to 13th day by both isolates; however, it is not very distinct as the lignin loss is more or less stable up to the day 13th of fermentation. However, there is no significant difference shown in the lignin degradation by both the white rots up to 13th day.
The total cost of the culture medium, equipment and operation with maximum laccase obtained in each culture and the time of cultivation needed. The addition of Cu2+ and surfactant (tween 20) increased the laccase activity in SSF cultures. By simply considering the increment on laccase activity the use of inducers is coherent and reasonable. However when analyzing the final price of produced enzyme, the use of inducers is not always suitable. Cultures under SSF conditions using copper ion as inducer and Polysorbate 20 presented a final price of 1.57 Euro per 1000 units of laccase produced. The maximum laccase activity in this type of culture was 917.54 U/L after 7 days of cultivation. The addition of inducer and surfactant resulted around 3.4 fold increase in laccase activity in relation to the culture with no inducers in about the same time of cultivation. In this way with increase in laccase activity decrement in final price was obtained thus, cultures with induced cultures presented an average final price of 1.57€/1000U. Laccase activity and its cost depend on composition of culture media such as carbon/ nitrogen ratio is one of the most important factors. Jing et al 2010 reported that initial low carbon/nitrogen ratio not only stimulate simultaneous production of laccase from Streptomyces lavendulae but also lowers the cost. But contradictory to this Eggert et al., 1996 reported that ligninolytic enzymes including laccase stimulated and triggered under N – limitation conditions i.e. high C/N ratio. Cost calculated in our study does not include industrial cost, packaging and distribution of enzyme. In this sense three main terms were taken into consideration; cost of the culture medium, cost of equipment and operating cost. In comparision to commercially available laccase the price of laccase produced under SSF using flask was lower than those commercialized by Sigma Aldrich (St. Louis, USA) and Jena Bioscience (Osma et al., 2011). In order to reduce the cost it is essential to explore more agro waste materials for using as substrate during fermentation, leading to more reduction in the final cost of laccase. G. gibbosum was cultured under SSF conditions using various culture medium compositions. Various variables of the culture medium were taken into consideration such as carbon source, nitrogen source, detergent and copper ions. The laccase activity of culture was measured and cost of production was estimated. The cost of crude laccase in the present study was lowest in comparision to the cost of laccase reported in previous results reported by Osma et al., 2011. In the present study similarity between predicted R2 and the adjusted R2 confirms the adequacy of the model to predict the response. Reliability of the experiment depends greatly on lower value of coefficient of variation our value of 2.74% indicates that the model is reliable. The ”lack of fit”F value”of 0.7346 tells that the ”lack of fit” was not significant and hence confirming that the model was fit (Table). Adequate precision is an indicator of signal to noise ratio and should be greater than 4.0 and our result 8.655 depicts a satisfactory signal. Further it is extremely important for a model to give an adequate fit, which otherwise can lead to prediction of false result. In our study as the normality assumption plot deduced along a straight line, hence adequate fit was satisfied and giving satisfactory approximation to the experiment (Fig)Box – Behnken design generally aims to select only those variables that are important for enzyme production. Each variable varies for a desirable response represented at high and low levels. For a good prediction of results, calculated F values should be several times more than tabulated value. Akhnazarova et al., 1982 reported that at higher F value and very low probability P>F=0.0001, indicates that the present model is in good prediction of experimental result. In our study also higher F value (11.46) implies that model is significant. From the optimum value of three factors obtained from the BB analysis, a media was formulated and was inoculated with G. gibbosum and incubated for 7 days at 25°C. The laccase enzyme was extracted and separated and was assayed by method of Diwaniyan et al., 2010. The result showed an increase of 3.4 fold in laccase enzyme production. Max yield was 45 877 IU/g. A response surface methodology allowed calculation of maximum production based on a few sets of experiments in which all the factors were varied within the chosen ranges. This method has been successfully applied in the optimization of the fermentation process (Rosi et al., 1987; Sonia et al., 2005). The variables with low probability level levels contribute to the model, whereas the other can be neglected and eliminated from the model. Values of prob >F less than 0.0500 indicates that model terms are significant model terms are significant model terms (455). High F- values and non significant lack of fit indicated the model was a good fit. “The predicted R-squared” of 0.6715 is in reasonable agreement with the “Adj R squared” of 0.8547. Adeq Precision measures the signal to noise ratio. A ratio greater than 4 is desirable. Ratio of 8.655 indicates an adequate signal, means the model can be used to navigate the design. The corresponding analysis of variance (ANOVA) is presented in the table. The F value is a measure of the variation of the data about the mean. The goodness of fit was checked by determination coefficient (R2) and the determination coefficient (R2 = 93.6%) indicates that only 6.4% was not explained by the model. The adjusted coefficient (Adj R2 = 85.4%) was also very high, which indicates a high significance of the experiment (Box et al 1951; Cochran et al.,1957; Akhnazarova et al., 1982; Khuri et al 1987).