Microbial Biotechnology

Biotechnology Industry

Electronic Journal of Biotechnology ISSN: 0717-3458  
© 2007 by Pontificia Universidad Católica de Valparaíso -- Chile  
BIP RESEARCH ARTICLE

Application of statistical experimental methods to optimize medium for exopolymer production by newly isolated Halobacterium sp. SM5

Patcharee Lungmann
Department of Industrial Biotechnology
Faculty of Agro-Industry
Prince of Songkla University
Hat Yai, Songkhla 90112 Thailand
Tel: 66 44 611221
Fax: 66 44 612 858
E-mail: plungmann@yahoo.com

Wanna Choorit*
Biotechnology Program
School of Agricultural Technology
Walailak University
Nakhonsithammarat 80160, Thailand
Tel: 66 075 672 355
Fax: 66 075 672 302
E-mail: cwanna@wu.ac.th

Poonsuk Prasertsan
Department of Industrial Biotechnology
Faculty of Agro-Industry
Prince of Songkla University
Hat Yai, Songkhla 90112 Thailand
Tel: 66 74 286 361
Fax: 66 74 446727
E-mail: poonsuk918@yahoo.com

*Corresponding author

Financial support: This research was supported by a research grant of Prince of Songkla University.

Keywords: exopolymer, halophilic bacteria, medium optimization, mixture design, response surface method.

Abbreviations:

CI: confidence interval
KS medium: Kamekura and Seno medium
PHB: Poly-β-hydroxybutyrate
RSM: response surface methodology

Abstract

Reprint (PDF)

Optimization of medium for exopolymer production by a newly isolated bacterium, Halobacterium sp. SM5 using the mixture design and response surface methodology method was investigated. The suitable medium recipe for enhancement the exopolymer production was 7.43 g/l glucose, 12.38 g/l yeast extract, 17.33 g/l MgSO4·7H2O, 9.9 g/l vitamin casamino acid and 2.48 g/l KCl. The exopolymer to be produced by the strain SM5 was 2.25 g/l, which was higher than that obtained in the original medium (1.3 times). The yield of exopolymer was 2.13 g/l to be obtained in medium containing 7.42 g/l glucose, 11.37 g/l yeast extract, 22.28 g/l MgSO4·7H2O, 7.44 g/l vitamin casamino acid and 0.99 g/l KCl which was predicted by response surface methodology. However, under an experiment, the yield of exopolymer was 2.08 ± 0.0020 g/l.

Reprint (BIP)

In general, media are traditionally optimized by the one-at-a-time strategy. This strategy is simple and easy to apply without the need for statistical analysis. In contrast, mixture design, full factorial design, and fractional factorial design are more efficient approaches that can deal with a large number of variables simultaneously. Moreover, the interaction among different variables can be estimated. To date, these techniques have been used by many researchers to optimize medium components, such as clavulanic acid from Streptomyces clavuligerus (Wang et al. 2005) and micrococcin GO5 from Micrococcus sp. GO5 (Kim et al. 2006). In the screening process for the exopolymer, one isolated halophilic bacterial strain, SM5, was found to produce an exopolymer which has potential medical and nanotechnological applications. Thus the objective of this work was to apply a mixture design followed by a response surface methodology to investigate and optimize a medium which might positively affect exopolymer production by the newly isolated Halobacterium sp. SM5.

Materiasl and Methods

Cultivation and exopolymer measurement

In order to prepare a starter culture, the bacterium was transferred into the fresh medium and incubated at 37ºC. Then 10% of the starter culture (optical density, OD660 = 0.5) was transferred into the fresh medium. After the cell pellet was removed by centrifugation, the supernatant was precipitated with cold ethanol. The precipitated exopolymer was then re-dissolved in distilled water and dialysed against distilled water overnight. The lyophilized exopolymer obtained was measured and calculated as g/l.

Medium optimization using mixture design

The total weight of these 5 components, 10 g/l glucose, 10 g/l yeast extract, 20 g/l MgSO4·7H2O, 7.5 g/l vitamin casamino acid and 2 g/l KCl of the modified KS medium, were calculated into percent of the total and displayed in decimal fraction. The pseudo components of five factors (A = glucose, B = yeast extract, C = MgSO4·7H2O, D =  vitamin casamino acid and E = KCl) in the ranges from the lowest to highest levels of their volume fractions were obtained after entered the data into the JMP version 3.2.6 program (software of SAS Institute Inc). These 30 pseudo components were converted to actual corresponding weights by using percent solids.

First model. The linear model for predicting the optimal point was expressed according to the equation:

Y = β1A + β2B + β3C + β4D + β5E    [1]

Where Y is the predicted response; β1, β2, β3, β4 and β5 are the coefficients estimate, and A, B, C, D and E are the studied factors (Hu, 1999).

A test for significant sequential models was performed by employing a statistical methodology called an analysis of variance (ANOVA). The response surface methodology (RSM), was also utilized to predict the optimization of medium for exopolymer production. The DESIGN-EXPERT 6.0.1 software was used for the RSM.

Results and Discussion

Mixture design

The mixture design experiment showed that the exopolymer productivity values varied within the range of 1.22-2.25 g/l (Table 1). To select an appropriate type of model for the response exopolymer production, the DESIGN-EXPERT software was applied. Sequential F-tests are performed, starting with a linear model and adding terms (quadratic, and higher if appropriate). In this result, the linear model was high F- value (34.67) as significant (Prob > F is less than 0.0001). Therefore, the linear model was appropriated for the response exopolymer production. The R-Squared value indicates that five components altogether would explain about 81% of the variability in the responses whereas about 19% of the variability in the responses remains unexplained.

Table 2 shows the results of the analysis of the pseudo components experiment as follows: The regression coefficient estimates of glucose, yeast extract, MgSO4·7H2O, vitamin casamino acid and KCl were 0.66, 2.11, 2.18, 2.10, and 1.45, respectively. Thus, the estimated regression equation for the pseudo components is

Y = 0.66A + 2.11B + 2.18C + 2.10D + 1.45E    [2]

The DESIGN-EXPERT program gave the output of the final equation in terms of actual components as follow:

Y = -0.069A + 0.067B + 0.063C + 0.066D + 0.011F   [3]

Where the variables take their coded values, represents exopolymer yield (Y) as a function of glucose (A), yeast extract (B), MgSO4·7H2O (C), vitamin casamino acid (D) and KCl (E) concentrations.            

In the above equation, glucose has a negative effect on the exopolymer production. This indicates that for every one unit increase in glucose, the exopolymer production will decrease by 0.069 unit when other components hold constant.

The results from DESIGN-EXPERT program revealed the adjusted components affected to the exopolymer production. The  Prob > | t | was less than 0.01 for glucose, yeast extract, MgSO4·7H2O, and vitamin casamino acid, it can also be concluded that these four components contribute to the model. It can also conclude that KCl did not effect to the exopolymer production as can be seen that its Prob > | t | is as big as 0.3392.

Triaxial diagram

Figure 1 shows the level curves of exopolymer yield as a function of the compositions, obtained from a linear regression. Combination of the three components (MgSO4·7H2O, yeast extract, and vitamin casamino acid) lead to higher yield. A high synergetic effect can be observed in the central area of this graph. The lowest component for glucose and KCl, 10.01-11.37 (g/l) for yeast extract, 21.17-22.28 (g/l) for MgSO4·7H2O, 7.44-9.90 (g/l) for vitamin casamino acid correspond, approximately, to the maximum of yield. In this case, the predicted maximum value of exopolymer was around 2.13 g/l whereas the actual experiment gave the maximum exopolymer yield of about 2.08 ± 0.0020 g/l from using the medium containing 7.43 g/l glucose, 11.37 g/l yeast extract, 22.28 g/l MgSO4·7H2O, 7.44 g/l vitamin casamino acid and 0.99 g/l KCl. 

Concluding Remarks

The data obtained from our experiments demonstrated the strategies for enhancing exopolymer and analysing the factors that affected exopolymer production by the bacterium strain SM5. The medium No. 23 formulated by mixture design was superior to other media in terms of original compositions for enhancing exopolymer production. The results of linear model for mixture design experiments showed that yeast extract, MgSO4·7H2O and vitamin casamino acid gave positive effect while glucose gave negative effect for exopolymer production. The KCl exhibited no effect on exopolymer production.

References

HU, Ruguo. Food Product Design: A computer-Aided Statistical Approach. Technomic Publishing CO. Ltd. Pennsylvania, 1999, 240 p. ISBN 1-56-676743-1.

KIM, Mi-Hee; KONG, Yoon-Jung; BAEK, Hong and HYUN, Hyung-Hwan. Optimization of culture conditions and medium composition for the production of micrococcin GO5 by Micrococcus sp. GO5. Journal of Biotechnology, January 2006, vol. 121, no. 1, p. 54-61. [CrossRef]

WANG, Yong-Hua; YANG, Bo; REN, Jie; DONG, Mei-Ling; LIANG, Dong and XU, An-Long. Optimization of medium composition for the production of clavulanic acid by Streptomyces clavuligerus. Process Biochemistry, March 2005, vol. 40, no. 3-4, p. 1161-1166. [CrossRef]

Note: Electronic Journal of Biotechnology is not responsible if on-line references cited on manuscripts are not available any more after the date of publication.

Supported by UNESCO / MIRCEN network
Home | Mail to Editor | Search | Archive