Key words: enzymes, developing countries, agro-industry; food processing

 Summary

The use of enzymes has the potential to increase productivity, efficiency and quality output in agro-industrial processing operations in many developing countries. Enzyme-catalyzed processes generally have requirements for a simple manufacturing base, low capital investment and consume relatively small amounts of energy, when compared to other methods of food processing.

 This review presents an inventory of current and potential areas in which the use of enzymes may expand and diversify markets for agricultural products, facilitate agro-industrial development, improve nutrition, and reduce toxicity in foods produced and consumed in developing countries.

Enzymes varying in quality from highly purified commercial preparations to relatively crude preparations in the form of leaves, plant exudates, chopped fruit and pounded grains, are widely applied in agro-processing in developing countries. Enzymes enhance extraction and separation processes, eliminate toxic and anti-nutritional factors, catalyze carbohydrate, protein and lipid conversions, and through their antioxidant and biocatalytic activities, are applicable as non-thermal food preservatives. Enzyme-catalyzed processes are scaleable with requirements for a simple manufacturing base, and low energy and capital investment requirements. Through the use of enzymes, agro-industrial processes may be upgraded, energy costs associated with processing reduced, the nutritional quality and safety of foods improved, processing times shortened, new products may be generated and alternative applications for several agricultural products may be realized.

Enzymes may either be directly incorporated into food systems, or may be immobilized on inert support matrices (immobilized enzymes) and allowed to interact with food systems during processing. High costs associated with the use of immobilized enzymes and their susceptibility to microbial contamination, precludes their applicability in processing in most developing countries.

Enzymatic reactions in foods may be controlled through the manipulation of temperature, pH, water activity and oxygen levels, or through the use of chemical inhibitors or inactivators. Novel technologies such as the use of ionizing radiation, protein inhibitors and "killer enzymes" which are enzymes capable of inactivating other enzymes, are all mechanisms which are being increasingly used for controlling enzymatic reactions in food systems. Mechanisms for the control of enzymatic reactions in foods were recently reviewed by Ashie et al. (1996).

Food fermentation processes are reliant on both endogenous and microbial enzymatic activities for the degradation of starches, lipids, proteins, anti-nutritional and toxic factors. In some cases microbial enzymatic activities associated with indigenous fermentation processes exhibit unique properties. Micro-organisms are currently the primary source of industrial enzymes: 50 % originate from fungi and yeast; 35 % from bacteria, while the remaining 15 % are either of plant or animal origin (Boopathy 1994). Microbial enzymes are commercially produced either through submerged fermentation (SMF) or solid substrate fermentation (SSF) techniques. SMF techniques for enzyme production, are generally conducted in stirred tank reactors under aerobic conditions using batch or fed batch systems. High capital investment and energy costs, and the infrastructural requirements for large scale production make the application of SMF techniques in enzyme production, impractical in a majority of developing country environments.

SSF on the other hand is a low cost, low technology fermentation technique applicable in use in the small scale production of enzymes. SSF incorporates microbial growth and product formation on or within particles of a solid substrate (Mudgett 1986) under aerobic conditions, in the absence or near absence of free water, and does not generally require aseptic conditions for enzyme production. According to the Central Food Technological Research Institute (CFTRI) in India, enzyme production by SSF accomplishes higher productivity per unit volume of fermentor space than SMF techniques. Processing wastes such as soybean hulls (Jha et al. 1995), wheat bran (Babu & Satyanarayan 1995) and cassava peels (Ofuya & Nwajuba 1990), may be upgraded through the production of enzymes by SSF techniques.

This review presents an inventory of current and potential enzyme applications which offer opportunities for improving nutrition, reducing toxicity and enhancing efficiency and productivity in agro-industrial processing in developing countries.

Energy Density. In a number of developing countries, weaning foods for infants consist of starchy liquid gruels of low energy density. Hydrolysis of the starchy component of these liquid gruels by a -amylases, results in the production of maltose and low molecular weight dextrins which are of low water-binding capacity, thus causing a reduction in viscosity. This reduced viscosity facilitates the incorporation of increased amounts of dry starchy material in the preparation of these gruels, thus increasing their energy densities. The use of amylase rich flour (ARF) prepared from germinated cereal grains (Gopaldas et al. 1988; Anon. 1991; Wahed et al. 1994; Darling et al. 1995) in improving the energy density of cereals is well documented. ARF is generally added to the gruel subsequent to cooking, and has been shown to lower the viscosity of traditional cereal preparations to levels greater than 90 % (Wahed et al. 1994). Improvement of the energy density of starch-based diets through amylase activity is very easily applied and adoptable at the household level in developing countries.

The viscosity-reducing ability of a -amylases is also applied on a large scale in the downstream processing of starches (Christensen 1989). Amylases increase extruder throughput in cereal processing through viscosity reduction. Use of thermostable microbial a -amylase in extrusion processing has been shown to result in considerable (1000-fold) viscosity reduction (Likimani et al. 1991), yielding extruded products of high nutrient density. This technology is already applied in a few African countries (Bunders et al. 1994).

Nutrient Bioavailability. Phytic acid (inositol hexaphosphate) is an antinutritional component of many cereal grains, oilseeds and legumes. It occurs in cereals at levels ranging between 0.5 and 6 % (Graf 1983; Erdman 1979), and is a major contributor to poor nutrient bioavailability in cereal based diets owing to it sequestration of indigenous micronutrients such as calcium, iron and zinc.

Endogenous cereal phytases are capable of dephosphorylating phytic acid (Reddy & Pierson 1994), and improving the bioavailability of these micronutrients in the diet. Cereal phytases are relatively easily activated by soaking, germination and fermentation. McKenzie-Parnell & Davies (1986) determined that phytate is lost at all stages of brown and white bread preparation, with the highest rate of loss occurring during the fermentation or proving step. In countries where flat breads are major dietary constituents and minimal proving is incorporated into the baking process, low nutrient bioavailability due to sequestration of phytic acid is a major cause for concern, and exogenous supplementation of flours with microbial phytases is required to enhance micronutrient bioavailability.

Cassava is consumed as a staple in a large number of countries in the developing world. Its consumption is however associated with pathological disorders such as goitres, tropical ataxic neuropathy and cretinism (Ermans et al. 1980; Mlingi et al. 1992; Tylleskare et al. 1992). This is due to the presence in cassava of the cyanogenic glucosides linamarin and lotaustralin, which are degraded through enzymatic activities to produce the toxic factors, cyanohydrins and hydrogen cyanide (HCN) (Fig. 1). Cyanogenic glucoside levels in the cassava root vary widely, and can accumulate to concentrations as high as 500 mg/kg (McMahon et al. 1995). Cyanogenic glucoside levels in cassava must therefore be quantitated in order to assess "bitterness" or the "cyanogenic potential" of cassava varieties. Linamarase enzymes isolated from the cassava peel, are currently applied as diagnostic reagents in the screening of cassava varieties for "bitterness." Methodology used incorporates the enzymatic hydrolysis of cyanogenic glucosides, followed by colorimetric measurement of cyanide released (Cooke et al. 1979). The development of biosensors incorporating linamarase, for linamarin detection in cassava and cassava-based products has also been reported (Yeoh & Truong 1993; Tatsuma et al. 1996). Biosensors of this type are increasingly replacing the use of colorimetric assays for monitoring enzymatic reactions in foods.

Complete detoxification of "bitter" cassava varieties can feasibly be accomplished through the enzyme-catalyzed degradation of both cyanogenic glucosides and the cyanohydrins which result from their degradation. Linamarase enzymes located in the cell walls of the cassava root (Mkpong et al. 1990), catalyze the initial step in the breakdown of the cytoplasmic cyanogenic glucosides linamarin and lotaustralin, resulting in the release of cyanohydrins. These cyanohydrins are relatively stable under low pH conditions but decompose under conditions of high temperature and high pH (pH >5), to produce ketones and HCN. The destruction of cassava cyanogenic glycosides is primarily an endogenous phenomenon, although microbial b-glucosidases assist in linamarase degradation during fermentation processing (Anupe & Brauman 1995). Bacillus species (Amoa-Awau & Jakobsen 1995), lactic acid bacteria (Cohen 1994) Aspergillus and Fusarium strains of fungi, some Penicillium strains and some Trichoderma strains (Yeoh et al. 1995) have been shown to secrete linamarase activity, and are potential sources of microbial linamarases.

Hydroxynitrile lyase (acetone cyanohydrin lyase) (Thayer & Conn 1981; Kojima et al. 1979; Hughes et al. 1994; McMahon et al. 1995; Cheskul & Chulvatnatol 1996; Wajant & Pfizenmaier 1996) an endogenous enzyme of the cassava root, catalyzes the breakdown of cyanohydrins produced as a result of linamarase activity, leading to the release of HCN (Fig. 1). HCN thus released is readily removable by evaporation. Complete cyanohydrin degradation through the combined activities of linamarase and the cyanohydrin-degrading enzymes, followed by evaporative removal of HCN, would preclude the exposure of individuals involved in the processing of "bitter" cassava varieties, to dangerously high HCN levels emitted during roasting or high temperature cooking. Exogenous addition of cyanohydrin-degrading enzymes should certainly be a consideration for upgrading processing of "bitter" cassava roots where financial and technological conditions permit.

 Value addition to Roots, Tubers and Cereals Through The Application of Enzymes

Sweeteners. Starchy substrates may be converted to sweeteners through enzyme catalysis. In India, bacterial a -amylase and fungal glucoamylase are utilized in the production of dextrose and liquid glucose from starch (Lonsane & Ramakrishna 1989). Enzyme-enzyme production of dextrose involves starch liquefaction by bacterial a -amylase, followed by saccharification to 95 % glucose with fungal glucoamylase. Both enzymes have also been used for the production of potable alcohol from tapioca flour, through simultaneous saccharification and fermentation. High-fructose syrup is produced from starch in China (Ruivenkamp 1990), while in Vietnam and Benin, rice and maize seedlings are applied as b -amylase sources in the production of a syrup from cassava starch containing 60 % maltose, 25 % glucose and 15 % of other sugars (Cecil 1995; Quynh & Cecil 1996). Sorghum malt enzymes have also been applied in the bioconversion of other starchy substrates to sugars (Solomon et al. 1994).

The application of enzymes in the production of sweeteners from starches, has the potential to make food processing less dependent on sugar-cane in areas which are unsuitable for sugar production, or where starches are more cheaply accessible than sugars. Furthermore, starch-sweetener bioconversions offer alternative uses for starchy materials (such as breadfruit) which are highly perishable and lack proper preservation techniques. In areas where sugar costs are greater that starch costs, the bioconversion of starches to sweeteners should definitely be considered.

Brewing. Tropical cereals are not particularly suitable for the brewing of beer, owing to their low enzyme levels (Nout & Davis 1982; Bajomo & Young 1992). They are however increasingly utilized in the replacement of imported barley in developing country brewing applications. Replacement of malted barley by these tropical cereals in the brewing industry necessitates supplementation with exogenous enzymes (Bajomo & Young 1992; Agu et al. 1993 ). In Viet Nam, sprouting rice is supplemented with microbial a -amylases from Bacillus subtilus, while in South Africa the brewing of sorghum beer is supplemented with commercial enzyme preparations.

A major hurdle which must be overcome in enzyme-catalysed starch bioconversions is the high energy requirement of the process, since starchy substrates must undergo the gelatinization process in order to be rendered suitable for hydrolysis by amylolytic enzymes. In the production of ethanol from sweet potatoes for example, 30-40 % of energy requirements are consumed by the saccharification of starch (Matsuoka et al. 1982). Gelatinization temperatures of starches vary in accordance with their biological sources. Where gelatinization temperatures exceed optimum temperatures for b -amylase activity, as occurs in the case of sorghum starch, extraction of endogenous sorghum malt enzymes prior to gelatinization is essential (Ilori et al. 1991). Identification of thermostable amylases capable of hydrolyzing starch under such high temperature conditions would therefore be of interest. Thermostable bacterial cyclodextrin glycosyl transferases (CGTase) capable of liquefying starch under low pH conditions have been reported (Nielsen 1991).

A more feasible approach for starch bioconversions in developing countries in the light of high energy requirements of the gelatinization process, is the use of amylases capable of hydrolyzing non-gelatinized "native" forms of starch. Yeast (Ueda & Saha 1983; Iefuji et al. 1996), moulds of the genera Rhizopus (Yamazaki & Ueda 1951), and Aspergillus (Ueda et al. 1984; Okolo et al. 1995), alkali-tolerant Bacillus species (Kelly et al. 1995) and Lactobacillus plantarum A6, (Giraud et al. 1994), are all capable of hydrolyzing non-gelatinized, "native" forms of starch. The energy-saving potential of these enzymes certainly warrants an interest in their applications for starch bioconversions in developing countries.

Oligosaccharides such as raffinose, stachyose and verbascose, are anti-nutritional factors in that they are not metabolized by humans and cause flatulence, diarrhoea and indigestion. Alpha-D-galactosidic bonds of these oligosaccharides are resistant to cooking and other small scale processing steps, but are hydrolyzable by endogenous a -galactosidases on germination (Chang & Mo 1985; King 1987; Kasai 1987; Suparmo & Markis 1987; Trugo et al. 1993; Abdel Gawad 1993) or soaking (Mulyowidarso et al. 1991; Abdel Gawad 1993). Rhizopus oligosporus moulds used in tempeh fermentations and lactic acid bacteria (Mital & Steinkraus 1975; Shallenberger et al. 1976; Akinyele & Akinlosotu 1991; Garro et al. 1996) are both sources of a -galactosidases. These bacterial a-D-galactosidases can potentially be exploited as food additives in the production of processed legume-based foods containing low levels of carbohydrates. The pressurized infusion of a -galactosidases may also offer potential for degrading legume oligosaccharides.

In recent times, enzymes have received much attention for their applicability as antimicrobial agents in foods. Antimicrobial enzymes offer potential for extending the self-life of unprocessed or minimally processed foods, controlling food fermentation processing, assisting the release of intracellular products from bacteria, and in disinfecting equipment, packaging materials and process water (Nielsen 1991; Teichgraeber et al. 1991; Teichgraeber et al. 1993; Colantuoni et al. 1992). Chitinases are produced by plants as a defence mechanism against invading fungal pathogens. These enzymes are also active against human pathogens such as Listeria monocytogenes, Clostridium botulinum, Bacillus cereus, Staphylococcus aureus and E. coli (Scott 1988). The high level of antifungal chitinase activity present in germinating soybean seeds, and in other legumes, should certainly be exploited in light of its energy-saving potential applicability in the non-thermal preservation of foods in developing countries.

The use of enzymatic pre-treatment for enhancing the recovery of oil from fruits and oilseeds was recently reviewed by Dominguez et al. 1994. Non-lipolytic enzymes enhance extraction processes through digestion of cell walls (Graille et al.1988), and improve vegetable oil yields (McGlone et al. 1986; Bhatnagar & Johari 1987; Bouvier & Entresangle 1992; Sitohy et al. 1993; Dominguez et al. 1995) with the benefits of reduced pollution and processing costs. Phospholipases are applicable in the degumming of vegetable oils during processing (Anon. 1992; Buchold 1993; Dahlke et al. 1995) with similar benefits. These enzymes have recently been incorporated into a commercial vegetable oil-refining process (Dahlke et al. 1995).

 Lipolytic enzymes also add value to oils through quality improvement. Hydrolysis, transesterification of glycerols and esterification of fatty acids are three lipase-catalyzed reactions applied in adding value through vegetable oil modification. Hydrolysis involves the enzymatic cleavage of specific fatty acids; transesterification of glycerols involves rearrangement of the triglyceride esters of fatty acids, with those from an added source, as occurs in the production of cocoa butter equivalents from palm oil and stearic acid, while esterification of fatty acids on the other hand, involves the building of di- and tri-acylglycerols from monoglycerols with the addition of certain fatty acids. The selectivity of lipase activity is used in the production of margarines and fat spreads (Macrae 1989). Lipase-catalyzed transesterification of palm oil to obtain harder fats suitable for many food products is currently being researched at Kasetsart University in Thailand.

Lipase modification of vegetable oils may also offer developing countries the opportunity to convert edible oils into industrial fuels such as biodiesels (vegetable oil fatty acid methyl esters) (Hummel et al. 1994) and to diversify in their utility of vegetable oils.

 Endogenous enzymes play a major role in determining the quality of fruits and vegetables. Lipoxygenase alone, or in combination with aldehyde oxidase and alcohol dehydrogenase is important in flavour development of a number of vegetables. Peroxidase enzymes serve as indicators of adequate heat treatment during blanching, while phenoloxidases play a major role in the discoloration of fruits and vegetables with adverse effects on taste and nutritional quality.

Enzyme infusion techniques are utilized in the peeling and segementation of fruits (Elliot & Tinibel 1993; Baker & Grohmann 1995; McArdle & Culver 1994; Baker & Wicker 1996), and in preserving the texture of processed fruits and vegetables (McArdle & Culver 1994; Baker & Wicker 1996; Kwon et al. 1996). Cellulases, amylases and pectinases facilitate maceration, liquefaction and clarification during fruit juice processing (Sreenath et al. 1994; Pong et al.1996; Pheantaveerat & Ansprung 1993; Faigh 1995) with the benefit of reducing processing costs and improving yields. In India, microbial pectinesterase is used in the fruit-processing industries for the maceration of fruit pulp and the depectinization of juices, (Lonsane & Ramakrishna 1989). Through the use of commercial pectinase enzymes, apple juice yields have increased by 30 % in Mexico (Anon. 1990).

Enzymes are applicable in enhancing the quality attributes of processed products. Rhamnosidases such as naringinase improve the quality attributes of fruit juice, by catalyzing the hydrolysis of bitter components (Baker & Grohmann 1995), while glucose-oxidase catalase systems (Duxbury 1993; Meyer & Isaksen 1995), and fungal phenoloxidases (laccases) (Maier et al. 1990) have an antioxidant effect in juice systems by removing residual oxygen present thus preserving flavour and color.

Exogenous enzymes are applicable in improving the digestibility of milk, while endogenous enzymes improve its keeping quality. Microbial lactase (b -galactosidase) which hydrolyses lactose in milk into glucose and galactose is used to address the problem of lactose intolerance which is prevalent in a number of developing countries. High costs associated with the use of these enzymes can be offset by the application of low enzyme concentrations over long incubation periods, or recouperation of the enzymes for re-use by dialysis or ultrafiltration techniques.

Lactoperoxidase enzymes (LP), which are indigenous to milk, are major contributors to its antibacterial activity (Fuglsang et al. 1995). These enzymes oxidize thiocyanates during removal of hydrogen peroxide present in milk leading to the production of hypothionate ions (OSCN)¯ which are believed to be antibacterial agents, and which react with bacterial membrane proteins. LP improve the storage stability of raw milk by delaying the growth of psychrotrophic organisms (Wolfson & Summer 1993). At low cell concentrations LP are bactericidal against several Gram-negative bacteria, including E. coli, Pseudomonas aeruginosa and S. Typhimurium and show both bacteriostatic and bactericidal activity towards Gram-positive bacteria. Oxidase systems such as glucose oxidase and xanthine oxidase, which are hydrogen peroxide producers, enhance the antibacterial activity of LP in milk and decrease or eliminate bacterial growth (Denis & Ramet 1989). The addition of small, but balanced amounts of thiocyanante (15 ppm) and hydrogen peroxide (8.5 ppm) to milk also activates its natural lactoperoxidase system (Claesson 1994). Exogenous addition of thiocyanate and hydrogen peroxide, for enhancement of the keeping quality of milk has been field tested in Kenya, Mexico, Nicaragua, Pakistan, Sri-Lanka, India, Poland and the People’s Republic of China, with consistently good results (Claesson 1994). With good sanitary practice, the LP-system is applicable in providing a simple and cheap method for preventing spoilage of raw milk during collection and storage in tropical regions.

The digestive tract of marine organisms harbours proteases, lipases and carbohydrases. Of these, the proteases have been most widely applied and studied as a food processing aids. Digestive proteases from aquatic organisms have been used as fish processing aids, for the production of fish sauces and other protein hydrolysates.

Endogenous proteolytic enzymes of the fish gut are responsible for protein hydrolysis during traditional fish fermentations. In recent times, research has focused on the addition of exogenous sources of proteolytic enzymes, such as soy sauce koji (Chae et al. 1989) microbial proteases (Sumangue & Mabesa 1995) and enzyme-rich fruit components (Chuapoehuk & Raksaulthai 1992) to accelerate the rate of protein hydrolysis in traditional fish fermentations. In VietNam, pineapple juice is added as an exogenous enzyme source to fish meat in the production of a special fish sauce product, while in Thailand, chopped pineapple and papaya are used as exogenous sources of proteolytic enzymes for fish sauce fermentation.

In addition to the production of fish sauce, inexpensive fish and fish processing wastes are transformed by endogenous proteolytic enzymes to protein concentrates which are applicable in a variety of food applications as milk replacers, feed supplements and broth flavorants. Similarly, endogenous proteolytic enzymes play a major role in protein hydrolysis during traditional fish fermentations.

A wealth of opportunity exists for the use of enzymes in household, village level, and large-scale agro-processing applications in developing countries. A number of these enzymes and enzyme-catalysed processes however need to more fully explored and exploited in light of the benefits that may be derived from their use.