The work described in this thesis addresses two classes of mixed biopolymer systems: (a) starches and sodium caseinate ; (b) gelling seaweed polysaccharides (x-carrageenan - both with and without locust bean gum - agar and alginate) and proteins (gelatin and blood plasma proteins). The viscosity and swelling volume of a 1% potato starch paste in distilled, deionised water is markedly reduced in the presence of caseinate. Similar effects were seen with simple electrolytes suggesting that this occurs as a consequence of a non-specific ionic strength effect. In contrast a 4% maize starch paste in distilled, deionized water undergoes a viscosity and swelling volume increase in the presence of caseinate. However, when pasted in a 0.1M, pH 7.0 phosphate buffer caseinate addition has little effect on the viscosity of the fresh paste and at high concentrations appears to prevent retrogradation on ageing. It is suggested that in buffer caseinate prevents the leaching of starch polysaccharides from the swollen granule and therefore maintains amylose in the granular phase. This is attributable to the high ionic strength of the solvent, allowing caseinate and the starch polysaccharides to phase separate. In water the unfavourable entropy, change due to the uneven distribution of the counter-ions, prevents phase separation and results in an interpenetrating network. Studies on the large deformation stress relaxation behaviour and melting points of 2% carrageenan, 0.5% carrageenan/0.5% locust bean gum (LBG) and 2% agar gels in a variety of solvent media indicate that the inclusions of 0-20% gelatin and 0-5% bovine serum albumin (BSA) give different results depending on both protein and polysaccharide. The main points of this study show that agar/gelatin mixed gel undergoes a distinct phase inversion at 4-7% gelatin levels, which is not seen with the carrageenan gels containing gelatin. Even when 20% gelatin is incorporated into a 2% carrageenan gel the melting point of the gel is unaltered from that of carrageenan alone. In the presence of high levels of BSA the carrageenan/LBG gel undergoes a marked increase in melting point. Investigations using locust bean gums of variable protein content suggests a possible (LBG)protein-BSA interaction since the melting point increases with the LBG protein content. It is shown that carrageenan/LBG gels have clear regions when formed by autoclaving in the presence of blood plasma. This supports the idea of an association between the protein in the insoluble husk and the blood plasma proteins. The interaction mixtures of BSA with sodium alginate at the interface and in bulk solution have been studied through the techniques of microelectrophoresis and ultracentrifugation respectively, to further elucidate the association between denatured proteins above their isoelectric point and anionic polysaccharides. Both techniques clearly show that the macromolecules can associate electrostatically at pH's above the pI of the protein.

Deterioration of flavour quality during processing and storage is often brought about by oxidative processes. These typically involve oxygen or an active form of oxygen in effecting transformation of a wide range of volatile and non-volatile compounds, including key quality chemicals, flavour precursors and antioxidants. To investigate the nature of the chemical and biochemical change within vegetables and herbs, unblanched frozen parsley was selected as a suitable tissue. The chemical status of parsley during technological processing was determined using novel analytical protocols (SNCV A/SNCNV A) implemented as part of a unified strategy for the quantitative analysis of volatile and non-volatile species. The analysis utilized a single stabilized solution produced from plant tissue, under a regime which minimized isolation stress and artifact formation. On frozen storage (-10OC) the principal volatiles of parsley, myrcene, beta-phellandrene and menthatriene were extensively degraded to non-volatile products at differential rates. p-Cymenene and the tentatively assigned menthatriene diepoxide were formed as minor volatile oxidation products. Myristicin remained largely unchanged. Under similar frozen storage, chlorophyll 'a' displayed significant degradation with only minor amounts of chlorophyllide 'a', pheophytin 'a' and 13[superscript]2 hydroxychlorophyll 'a' formed. Ascorbic acid was extensively degraded in timescales preceding monoterpene and chlorophyll loss. Thermal blanching of parsley extensively prevented the degradation of the monoterpenes, suggesting that endogenous enzymes were responsible for the changes. Elimination of oxygen, in the absence of blanching, prevented volatile loss, confirming the requirement for oxygen. The hypothesis that peroxidase can operate in a co oxidative couple with the flavonoid, apigenin-7-glucoside and hydrogen peroxide, as proposed by Yamauchi (1985), was investigated to establish its potential role in the degradation of terpenoids and chlorophyll. In model experiments, using horseradish peroxidase, menthatriene and chlorophyll showed extensive degradation only when all components of the couple were present. In addition the requirement for oxygen was also established. Naringenin and umbelliferone have been shown to behave similarly to apigenin, as co-substrates for peroxidase. Lycopene, with some structural similarity to menthatriene, was also susceptible to co-oxidation. Polyphenol oxidase, proposed to operate in a similar fashion to peroxidase with mono- and di-phenols as substrates (Montedoro et al. 1995), in model experiments did not cause the degradation of chlorophyll. The co-oxidative role of lipoxygenase in parsley is believed to be of minor significance, however, it is likely to be responsible for the production of low levels of hexanal observed during thawing of frozen parsley. From this thesis it is concluded that the aroma and colour quality loss in frozen unblanched parsley probably results from the oxidative degradation of the unsaturated monoterpenes and chlorophyll 'a' respectively via an oxidative cascade initiated by the action of peroxidase.

Lipase 3 is a fungal lipase produced industrially for use as a dough-conditioning enzyme in bread making. Mutants of Lipase 3 were designed to improve enzyme specific activity and to prevent N-linked glycosylation, which was found to cause a drop in activity on industrial-scale production. These were based on three-dimensional model structures of Lipase 3 in `open' and `closed' conformational states derived from crystallographic data of fungal lipases sharing high sequence homology. Lipase variants were expressed and secreted by Pichiapastoris yeast and purified by anionexchange chromatography, which allowed the separation of two active isoforms. Analysis of the `glycosylation' mutants by SDS-PAGE and MALDI-TOF mass spectrometry implied that mutation of N-linked glycosylation sites prevented attachment of oligosaccharide groups to these sites. Mutants were characterised by measurement of specific activities with soluble and emulsified substrates and determination of kinetic constants with emulsified substrate. None of the `activity' mutants showed improved activity over wild type, and the significant drop in activity with emulsified substrate on mutation of a `lid' tryptophan residue indicated that this residue was required for interaction with long-chain triacylglycerol substrates. Specific activities and kinetic constants measured for the ‘glycosylation' mutants did not differ significantly from those of the wild type enzyme. Sigmoidal kinetic curves were observed for lipases expressed in Pichia pastoris and Aspergillus niger with emulsified substrate. Co-operativity was measured using the Hill plot and found to be positive. The possibility of a kinetic mechanism, rather than an allosteric mechanism (involving interactions between ligand binding sites), for cooperativity is discussed.

Ogi is a fermented food made from maize, sorghum or millet which serves as complementary food for infants and breakfast for adults in Nigeria, West Africa. This study characterized the microbial diversity of maize and sorghum grains and ogi produced by their natural fermentation in an attempt to understand the roles of the key microbial species and the impact of the population dynamics and selected species on changes in nutritional composition and aroma notes of ogi during fermentation. A combined approach of culture dependent and culture independent methods of analysis was applied to investigate the microbial community of grains and ogi from two different sources. Microbial diversity and viable populations varied with the source of the grain. Bacterial and fungal genera identified with the partial 16S rRNA and 26S rRNA sequence analysis respectively in maize and sorghum were Bacillus, Enterobacter, Micrococcus, Kytococcus, Pantoea, Staphylococcus, Amycolatopsis, Methanoculleus, Aspergillus, Penicillium, Eupenicillium, Acremonium, Schizosaccharomyces, Meyerozyma, Hyphopichia, and Pichia in maize grains; Enterococcus, Enterobacter, Pantoea, Bifidobacterium, Aspergillus, Cladosporum, and Penicillium in maize ogi; Enterococcus, Enterobacter, Pantoea, Aeribacillus, Cyanobacterium, Acinetobacter, Fusarium and Trametes in sorghum grains; and Pediococcus, Lactobacillus, Enterococcus, Bacillus, Cladosporum and Penicillium in sorghum ogi. Similar species were observed in both sources of maize while those of sorghum differed slightly. Predominant microbes included species of Enterobacteriaceae and moulds. Acetic acid bacteria were not identified as part of the diverse community. Following the predominance of moulds during the natural fermentation, preliminary screening was performed by PCR using specific biosynthetic gene primers to test whether they are the mycotoxin producing species. None of the genes tested were detected by PCR thus they may not be the toxin producing species. Starch, non-starch polysaccharide (NSP), phytate and phenolic compounds were determined in the grains and respective ogi to ascertain the levels of these nutritionally important components in the naturally fermented ogi and the impact of the varying microbial populations on the fate of these compounds during fermentation. In the grains, the average starch and NSP contents in each case were 80.35 g/100g and 9.40g/100g in maize and 93.12 g/100g and 8.14 g/100g in sorghum. Out of the total in grain the average percentage recovery of starch and NSP respectively in the ogi showed 63% and 42% in maize and 58% and 27% in sorghum. Maize showed good starch and fibre (NSP) retention than sorghum after fermentation. To further understand the types and levels of polymers in NSP hydrolysis in ogi fermentation, HPLC analysis of the hydrolysed extract was performed. Glucose was entirely present in maize and sorghum ogi which represents the beta-D-glucans while arabinose and xylose (in maize only), mostly lost with the pomace, signify the arabinoxylans. Overall variations in the microbial populations of sorghum seemed causal to the difference in starch and NSP recoveries. Phytate was assessed based on release of total phosphorus in the samples by enzymatic and chemical methods. Recovery of phytate in the naturally fermented ogi ranged from 18-25% in maize and 40-48% in sorghum suggesting greater phytase activity and more nutrient bioavailability in maize ogi than in the sorghum. Greater activity in maize reflects the presence of phytate hydrolysing species such as Aspergillus in the grain. Total phenolic content (TPC) was assessed by Folin-Ciocalteu colorimetric method after direct extraction of samples by saponification. TPC in the original grains ranged from 410–437 mg GAE/100g in maize and 221–247 mg GAE/100g in sorghum. Due to the nutritional significance, the amount of phenolics that are either freely soluble or are covalently bound to the food matrix were assessed. Soluble phenolics in ogi ranged from 16-38% in maize and 32-49% in sorghum based on the total soluble fraction in the original grain. In all cases loss of soluble phenolics with the waste waters accounted for 12-25% and 31-39% with the pomace. Only the LAB population seemed to correlate with the release of phenolics in the natural fermentation. Given the higher value of soluble phenolics, naturally fermented sorghum ogi appeared to have higher antioxidant potential than the maize ogi. Furthermore an attempt was made to ascertain whether the use of selected microbes would improve the antioxidant properties and aroma of ogi while minimizing the incidence of pathogens due to chance inoculation. Thus the impact of selected LAB (Pediococcus pentosaceus) and fungi (T. hirsuta and A. zeae previously shown to have phytase activity) on changes in phytate, phenolics and aroma of ogi was assessed following a parallel experiment to the previous study but using autoclaved grains. Five fermentation treatments of the pure and co-cultures were investigated. Cell populations in all culture fermentations varied and reached the average maximum of log 6-9 cfu/ml. Changes in the distribution of bound and soluble phenolics were observed showing esterase activity. Leaching of phenolics was evident in all cases but was higher in the sorghum fermentations. Higher levels of soluble phenolics were recovered in pure culture fermented ogi using T. hirsuta or P. pentosaceus than in the natural fermentation having 76% and 45% of the original soluble fraction in maize and sorghum respectively. This suggests greater antioxidant potentials than the naturally fermented ogi. Pure culture fermentations using T. hirsuta and co-culture of P. pentosaceus with A. zeae reduced phytate by 97% and 96% in maize and sorghum ogi respectively showing greater phytase activity and more nutrient bioavailabilty in the ogi than in the natural fermentation. The aroma profile of ogi was analysed by solid-phase microextraction and gas chromatography-mass spectrophotometry (SPME GC-MS). Ethyl acetate, butyl acetate and ethyl hexanoate were observed as the key active aroma components in ogi. The ester, methyl thiobutanoate was found to be unique to the naturally fermented ogi suggesting that it may have been generated by species other than the selected starter organisms. Overall in both natural and starter culture fermentations, maize ogi showed high relative abundance of volatile components suggesting good substrate compatibility and utilization during fermentation. Thus compounds with high threshold values may be significant in the aroma notes of maize ogi. P. pentosaceus and T. hirsuta in pure and in co-culture fermentations produced ogi with aroma notes mostly related to the naturally fermented product. In conclusion the diversity and levels of the initial microflora and the structural composition of grain could be major factors contributing to the nutritional compositional changes in ogi fermentation.

There is demand from the food industry for novel savoury seasonings based on low-cost food ingredients, sourced from within the EU. A nucleotide and protein rich spray dried powder which was derived from a Fusarium venenatum fermenter waste stream and food-grade Alphitobius diaperinus with potential as a savoury flavour enhancer was evaluated. Enzymatic digestion of these two raw materials as a source of flavour precursors was evaluated. Serial enzyme combinations, enzyme dosages, sequence of enzyme application, pH, temperature and length of digestion for the liberation of amino acids and nucleotides were optimised for the liberation of taste active compounds. For amino acids, free glutamine (GLN) and glutamic acid (GLU) could be enhanced using a combination of peptidases on both raw materials. Digesting a spray dried powder derived from the fermenter waste stream of Fusarium venenatum with exopeptidase (1% Flavourzyme TM), resulted in an improved yield of GLN (from 0.1 mg/g to 28.9 mg/g powder) and GLU (from 1.71 mg/g to 5.98 mg/g powder). For milled Alphitobius diaperinus, mixed use of exopeptidase (1% Flavourzyme TM) and endopeptidase (2% Alcalase 1.4-fold increased yield of GLU (17.5 mg/g powder) and 1.7-fold increased yield of GLN (1.2 mg/g) as best production was obtained. For nucleotides, digestions of the waste stream with a yeast lytic enzyme (YL-TLTM) followed by a nuclease (RP-1GTM) resulted in the highest 5’-guanosine monophosphate (GMP) and 5’-adenosine monophosphate (AMP) production. Specifically, a 2% and 0.05% treatment by YL-TLTM and RP-1GTM respectively was shown to be optimal, followed by a 0.05% DeamizymeTM treatment for the conversion of AMP to 5’-Inosinic acid (IMP) of 38 mg/g. For the solid digestion of Alphitobius diaperinus, being treated with a nuclease tretment (2% RP-1GTM) followed by a 0.2% DeamizymeTM treatment for the conversion of AMP to IMP, resulted in the highest GMP yield, a 3.5-fold increased (2.6mg/g), and 7.8-fold increased IMP (4.7mg/g) production.