Isolation And Characterization Of Antimicrobial Metabolites From Microbial Isolates

Abstract

Antibiotic resistance has become a serious problem in both the developed and developing nations. The incidence of antibiotic resistance among the infectious microorganisms is fairly high the available antibiotics have eventually become ineffective for the treatment of frequently appearing infections. Thus the discovery of new antibiotics via natural and/or synthetic route has become extremely necessary for fighting important microbial infections. As the microorganisms have great potential of adaptation and show presence in almost all ecosystems, they are considered as the major source for discovering novel drug/lead molecules. For clinical purposes compounds from natural sources is highlighted by the fact that 60 % of the drugs in current use are either natural products or have a natural product background. The importance of natural product chemistry has also been maintained by new screening and microbiological methods. The work presented in this thesis deals with the isolation and screening of the microorganisms showing antimicrobial properties. For this, nine soil samples were collected from different stressed agro-ecological niches of northern India and screened by subjecting the soil samples to different physico-chemical treatments (heat, phenol, chloroform and antibiotics). After primary and secondary screening, twenty active isolates belonging to the genus Streptomyces were selected for further study showing greater potential against test organisms such as Bacillus subtilis, Staphylococcus aureus, Escherichia coli, Salmonella typhi, Candida albicans, Candida tropicalis, Cryptococcus terreus, Trichophyton rubrum, Penicillium ochrochloron, Fusarium moniliforme and other multiple drug resistant bacterial and fungal test strains. On comparing the activity profile two most potent strains (AB and M4) were selected for detail study. Initial classification and identification of organisms, were done according to the traditional methods which were based on morphological, physiological, biochemical and nutritional characteristics. Finally for accurate assignment of taxonomic status to the biologically active microbial isolates, bioinformatics methods (16S rRNA homology studies) were applied. Microbial strains, AB and M4 were isolated from soil samples of agricultural fields showed broad spectrum antibacterial and antimicrobial activity respectively. Morphological, physiological, biochemical and with 16S rRNA homology studies, revealed the strain (M4) show maximum closeness with Streptomyces triostinicus (AB184519.1) with gene sequence similarity: 98% whereas strain (AB) was characterized as Streptomyces olivaceus XSD-112 (EU273547.1) having 100% gene sequence similarity. During production, attempts were made to optimize the antibiotic production by standardizing the concentrations of medium components using classical (one factor at a time), statistical (plackett-burman design, response surface methodology) and mathematical methods (artificial neural networking and genetic algorithm). Effects of medium components (soybean meal, glycerol, CaCO3 and DL-alanine) for the optimization of olivanic acid production by Streptomyces olivaceus were investigated with the help of plackett-burman design (PBD). The individual and interaction effect of the studied variables (soybean meal, glycerol and CaCO3) were evaluated by response surface methodology (RSM) using central composite design (CCD). By applying statistical design, antibiotic production was enhanced nearly 8 times (415 mg/l) as compared with the normal production medium (50 mg/l). Apart from the above optimization techniques, artificial neural network (ANN) coupled with genetic algorithm (GA) has been also used in the optimization of medium components for the production of actinomycin V from Streptomyces triostinicus, which is yet not reported to produce this class of antibiotics. Experiments were performed according to five variable central composite designs and the data generated is used to build neural network model. The concentrations of five (MgSO4, NaCl, CaCO3, soybean meal and glucose) medium components served as inputs to the neural network model, and the antibiotic yield served as outputs of the model. A genetic algorithm is used to optimize the input space of the neural network model to find the optimum values for maximum antibiotic yield. Maximum antibiotic yield of 452.0 mg/l was obtained at the GA optimized concentrations of medium components (MgSO4 3.657; NaCl 1.9012; glucose 8.836; soybean meal 20.1976 and CaCO3 13.0842 g/l). The antibiotic yield obtained by the ANN/GA was 36.7 % higher than the yield obtained with the response surface methodology (RSM). For the purification of active compounds from the fermented broth or cell of the culture, various chromatographic techniques such as silica gel and sephadex LH20 column chromatography, thin layer chromatography and high pressure liquid chromatography were used. The compounds purified from S. olivaceus and S. triostinicus were chemically characterized as a new form of olivanic acid and actinomycins (Act V and Act D) respectively with the help of ultraviolet (UV), Fourier transform infrared (FTIR), electro spray ionization spectrometry (ESI) and nuclear magnetic resonance (NMR).