An innovative approach for microbial production of pyruvate using agro-industrial waste
Loading...
![Thumbnail Image](assets/images/Item.jpg)
Date
2021-02
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand)
Abstract
Pyruvic acid (pyruvate) is a cellular metabolite found in the biochemical link between glycolysis and the tricarboxylic acid cycle. The microbial production of pyruvate from yeasts or bacteria is based on limiting the natural catabolism of pyruvate and limiting the accumulation of its many potential by-products. The industrial pyruvate production methodology involves the use of organic chemicals along with standardised protocol which increases the cost of the end product. In this process, pyruvic acid is distilled from a mixture of tartaric acid and potassium hydrogen sulphates at 220°C; the crude acid obtained is then distilled under vacuum. This process is simple to realize but not cost-effective. Hence, to realise a cheaper and efficient methodology alternative nutrient sources are explored. Development of media formulations is key in any bio-transformation involving micro-organisms. Establishment of optimal culture constraints and process development is considered crucial in this regard. Since, microbial enzyme activity is important to pyruvate accumulation in isolated strains, advances in pyruvate production can be achieved by media optimization. In this study, the media formulation involved various carbon sources viz. glycerol, rice straw and
jackfruit rind with various proportions of nitrogen source, corn steep liquor (diluted with distilled water). The amount of nutrient sources is also very critical for formulation of standard media composition. Thus, the screening experiments were focused on standardizing the carbon and nitrogen levels for final experiments. In screening experiments, only incubation time and incubation temperature had significant effect on pyruvate production at a confidence level p<0.05. Finally, the factors compromised for main optimization experiments were screened as follows: carbon source 25%g/g, CSL 12%v/v, pH 5.0, agitation speed 220rpm, KH2PO4 1.1%g/g, thiamine 1.3%μg/g, biotin 1.4%μg/g, MgSO4.7H2O 0.3%g/g and CaCO3 43%g/g. The final experiments were based on full factorial design on different levels of independent variables. With glycerol as carbon source, the highest effect of CSL concentration on pyruvate concentration was observed and was highly significant (p<0.01) because it had high calculated F-value (519.60). The effect of temperature (340.32) followed by time (324.60) was also found significant (p<0.01). As per the results, the optimum solution was obtained when the CSL concentration was 0.992601, time was 0.509991, and temperature was 0.317417. Similarly with rice straw as carbon source, the highest effect of CSL concentration on pyruvate concentration was observed and was highly significant (p<0.01) because it had high calculated F-value (444.80). The effect of time (252.25) followed by temperature (143.40) was also found significant (p<0.01). As per the results, the optimum solution was obtained in terms of coded values when the CSL concentration was 0.999997, time was 0.882823, and temperature was 0.363031. Similarly with jackfruit rind as carbon source, the highest effect of CSL concentration on pyruvate concentration was observed and was highly significant (p<0.01) because it had high calculated F-value (1199.93). The effect of time (925.66) followed by temperature (255.70) was also found significant (p<0.01). As per the results, the optimum solution was obtained when the CSL concentration was 0.999986, time was 0.851276, and temperature was 0.438559. In supervised learning approach, the highest accuracy corresponds to 68.8889% for multilayer perceptron under 10X cross validation fold-maker, and the lowest is 0% for IBk. In fact, in this experimental comparison, we can say that multilayer perceptron was the best scheme in all applicable classifiers, with highest accuracy. Moreover, in unsupervised learning approach feature selection preprocessing was considered essential and principal component analysis was performed prior to each clustering algorithms application. The paired t-test analysis of the three carbon sources show a draw when compared with glycerol as a standard source. This sustains the fact that, organic replacements instead of conventional organic source i.e. glycerol can be utilized for pyruvate production with no random aberrations in pyruvate yield. The SEM results show that the diametric dimensions of the pyruvate produced from carbon sources were in the range of 30μm to 300 μm and is comparable with that of industrially produced sample.