Nutrient availability
Although the lime-cooking process to convert maize into tortillas induces some important losses in nutrients, the process also causes important changes in nutrient availability.
Calcium. Because of the use of calcium hydroxide in converting maize into tortillas, the calcium content of the product increases significantly, up to about 400 percent. Bioavailability studies conducted by Braham and Bressani (1966) with animals showed that somewhat less of the calcium was available in lime-treated maize (85.4 percent) than in skim milk (97 percent). Calcium bioavailability increased when lime-treated maize was supplemented with its limiting amino acids, lysine and tryptophan. Recently, Poneros and Erdman (1988) confirmed the high bioavailability of calcium from tortillas with or without the addition of ascorbic acid. As pointed out in a previous section, the use of calcium hydroxide improves the calcium-to-phosphorus ratio in tortillas, which possibly favours the utilization of the calcium ion by the animal. This is an important finding for populations who do not consume diets high in this essential mineral. Furthermore, the finding that better quality in maize protein favours calcium big-utilization is of nutritional significance and provides an additional reason for the commercial production of QPM for people who depend on maize for their nutrition.
Amino acids. Bressani and Scrimshaw (1958) carried out studies using in vitro enzymatic digestions with pepsin, trypsin and pancreatic. At the end of the pepsin digestion, the amount of alpha-amino nitrogen as a percentage of total digested nitrogen was twice as high from tortilla (43.1 percent) as from maize (21.4 percent) and levels of histidine, isoleucine, leucine, lysine methionine, phenylalanine, threonine and tryptophan were higher from the tortilla hydrolysate than from maize, suggesting a faster release from the proteins. These authors proposed that the difference in rate of release could derive from the significant decrease in the solubility of the prolamine protein fraction in tortillas, as compared with maize. Serna-Saldivar et al. (1987), however, working with ileum-cannulated pigs, found that at this level in the intestinal tract the digestibility of most of the essential amino acids was somewhat higher from water-cooked maize than from limecooked maize. Digestibility of the protein decreased slightly, possibly because of the heat treatment involved (Bressani et al., 1990). Other researchers have suggested that during maize processing, hydrophobic interactions, protein denaturation and cross-linking of proteins are probably responsible for changes in the solubility of these components, which could affect amino acid release during enzymatic digestion.
Niacin. The alkaline treatment of maize has been reported to destroy its pellagragenic factor. Evidence from a large number of researchers has suggested that pellagra results from an imbalance of the essential amino acids, increasing the niacin requirement of the animal. This point has been extensively debated between those who claim that niacin in maize is bound and not available to the animal and those who favour the theory of improved amino acid balance induced by the alkaline-cooking process, as lime treatment results in release of the bound niacin. Pearson et al. (1957) have shown that boiling maize in water has the same effect (that is, it increases niacin availability). Bressani, Gómez-Brenes and Scrimshaw (1961) found that in vitro enzymatic digestion liberated all the niacin from raw maize as from tortillas and reached the conclusion that differences in amino acid balance rather than in bound niacin were responsible for the differences between raw and lime-processed maize in biological activity and pellagragenic action. Lime treatment of maize improves amino acid balance, as demonstrated by Cravioto et al. (1952) and Bressani and Scrimshaw (1958). Other workers have shown that experimental animals grow better when fed lime-treated rather than raw maize. Using cats which cannot convert tryptophan into niacin - Braham, Villareal and Bressani (1962) showed that niacin from raw and lime-treated maize was utilized to an equal extent, suggesting its availability is not affected by processing.
Dietary Fibre. It has been shown above that when maize was processed into tortillas by lime-cooking, total dietary fibre (TDF) decreased at the dough stage and increased in the tortilla to levels only slightly below those found in raw maize. In these studies the levels of TDF in tortillas averaged 10 percent on a dry weight basis. If a person consumed about 400 g of tortilla (dry-weight), the TDF intake would be 40 g, a value significantly higher than the recommended intake. Even small children can consume relatively large amounts of dietary fibre, which can affect the availability of iron. Hazell and Johnson (1989), however, indicated that maize-based snack foods prepared by extrusion cooking have a higher iron availability than raw maize. These authors indicated that refining of raw maize, product formulation, extrusion cooking and addition of flavourings were responsible to different degrees. Likewise zinc intake could be affected. The other mineral that could be affected would be calcium; however, Braham and Bressani (1966) and Poneros and Erdman (1988) showed that calcium is relatively well available from tortillas and that its availability is increased when the protein quality is improved through addition of the limiting amino acids. An excess of calcium rather than dietary fibre could be responsible for zinc availability, as has been indicated in a number of studies.
Protein quality of maize and nutrient bioavailability
Growing rats retained calcium from tortillas better when it was supplemented with lysine its limiting amino acid, and with a mixture of amino acids. Protein quality is an important factor in bioavailability of nutrients from maize and its lime-treated products. As already stated, niacin availability also improves when protein quality is improved, and studies with QPM have shown better utilization of niacin. The same observation has been made on the utilization of carotene, which is higher in Iysine-supplemented yellow maize than in the unsupplemented product.
Changes in quality. Changes in nutritional value, particularly that of protein, during the transition from raw maize to tortillas have been studied mainly in animals. Even though chemical losses in some nutrients take place upon limecooking of maize, protein quality is slightly but consistently better in tortillas than in raw maize. Table 22 summarizes the results of various studies where raw maize and the tortillas made from it have been evaluated. The protein efficiency ratio of the tortillas is in general somewhat higher than that of the raw maize, although some studies have reported otherwise. The difference may be attributed to processing conditions, particularly the concentration of lime added, which is lower in rural home cooking than at the industrial level. The chemically determined amino acid pattern of the tortillas is no better than the pattern in raw maize. The only explanation is that the process increases the availability of key amino acids. This is indicated by the results of feeding studies with young rats (Bressani, Elías and graham, 1968). Both raw maize and the lime-cooked dough were supplemented with increasing levels of lysine alone (from 0 to 0.47 percent of the diet). Maximum PER for maize was obtained with an addition of 0.31 percent and for the lime-cooked dough with 0.16 percent. At all levels of supplemental lysine the dough gave higher PER values than the raw maize.
Tryptophan supplementation alone was also tested, and in this case 0.025 percent addition gave the highest PER for maize, with no response for the dough. The addition of the two amino acids at the level of 0.41 percent lysine with tryptophan varying from 0.05 to 0.15 percent improved the quality of both materials, although it was higher for the dough.
TABLE 22 - Protein quality of maize and tortillas
Type of maize | Protein quality (PER) |
||
Raw maize | Tortillas | Casein | |
Common | 1.13 ± 0.26 | 1.27 ± 0.27 | |
Common | 1.49 ± 0.23 | 1.55 ± 0.23 | 2.88 ± 0.20 |
QPM (Opaque-2) | 2.79 ± 0.24 | 2.66 ± 0.14 | 2.88 ± 0.20 |
Common | 1.38 | 1.13 | 2.50 |
Common Tropical | 0.99 ± 0.25 | 1.41 ± 0.11 | 2.63 ± 0.17 |
Common Highland Xetzoc | 0.96 ± 0.19 | 1.41 ± 0.20 | 2.63 ± 0.17 |
Common Highland Azotea | 1.02 ± 0.19 | 1.41 ± 0.17 | 2.63 ± 0.17 |
Common Highland Sta. Apolonia | 0.71 ± 0.20 | 0.98 ± 0.17 | 2.63 ± 0.17 |
QPM Nutricta | 1.91 ± 0.23 | 2.12 ± 0.12 | 2.63 ± 0.17 |
Biological value of common maize | 59.5 | 59.1 | 69.4 |
Net protein utilization of common maize | 51.2 | 49.4 | 64.5 |
These results were interpreted to mean that the quality of lime-treated maize is superior to that of raw maize. This explanation is supported by in vitro studies showing a greater release of essential amino acids (EAA) from tortillas than from maize, even though Ortega, Villegas and Vasal (1986) reported in vitro protein digestibility in maize, dough and tortillas to be 88, 91 and 79 percent respectively. For QPM the respective values were 82, 80 and 68 percent. Recently, Serna-Saldivar et al. (1987) reported on dry matter, gross energy and nitrogen digestibilities of maize cooked with and without lime. No differences in dry matter or gross energy digestibility values were found between the different processing treatments. Cooking maize with lime, however, reduced nitrogen digestibility from 76.5 to 72.8 percent. These values were measured near the end of the small intestine in pigs. Values for dry matter, gross energy and nitrogen digestibility increased when measured over the pigs' total digestive tract. From nitrogen balance studies, the same authors reported a retention of intake nitrogen of 45.8 percent for maize cooked without lime and 41.2 percent for lime-cooked maize. Retention of absorbed nitrogen was 48.2 percent for the lime-cooked maize and 52.9 percent for the maize cooked with water alone. Digestible and metabolizable energy were similar in maize processed with and without lime. The authors concluded that the lime-cooking process decreased the nutritive value of maize.
In another study by Serna-Saldivar et al. (1988b), this time conducted with rats, the authors noted an increase in percentage of dry matter and gross energy digestibilities from maize to nixtamal (dough) and to tortillas; however, protein digestibility decreased. In vitro studies correlated with in vivo values. graham, Bressani and Guzmán (1966) showed better weight gain in DurocJersey pigs fed lime-treated maize than in those fed raw maize, with better feed efficiency. In studies with dogs, lysine and tryptophan added to lime-cooked maize improved nitrogen balance to the value obtained with skim milk (Bressani and de Villareal, 1963; Bressani and Marenco, 1963). It was further shown that after these two amino acids, isoleucine, threonine, methionine and valine increased nitrogen retention above values measured with lysine and tryptophan. Lime-treated maize has also been evaluated in children (see Chapter 6). Nitrogen balance results have shown a high response to lysine and tryptophan addition, which in turn is dependent on the level of protein intake. At low levels, only lysine improved quality, but as nitrogen intake increased, the addition of tryptophan with lysine became important. All studies suggest that in limetreated maize, lysine is slightly more deficient than tryptophan, and the contrary seems to be the case for raw maize. Nevertheless, for a significant improvement in protein nutritional quality of lime-treated maize, both of these amino acids are required.
Use of QPM. Nutritionally improved (QPM) maize shows the same changes in protein quality and bioavailability after lime-cooking and conversion to tortillas as observed in normal maize. The difference is that QPM tortillas and products are nutritionally superior to those made from common maize. They are as acceptable to consumers.
Other effects of lime-cooking
Lysinoalanine formation. In 1969, De Groot and Slump demonstrated that alkali treatment of proteins gave rise to peptides such as Iysinoalanine (LAL), lanthionine and ornithine which had negative effects on animals. They were not biologically available and had detrimental effects on protein quality. Consequently, the effect of the alkaline-cooking process to convert maize into tortillas has received attention from various researchers. Sternberg, Kim and Schwende (1975) reported that commercial samples of masa flour, tortillas and taco shells contained 480,200 and 170 µg LAL per gram. Sanderson et al. (1978) also found that lanthionine and omithine were formed during alkaline cooking of maize. These authors found no LAL in common or in high-lysine raw maize; however, these products contained 0.059 and 0.049 g percent protein respectively after alkali treatment. A commercial masa contained 0.020 percent, and tortillas 0.081 percent protein. These authors also reported lanthionine and ornithine values in the masa prepared from the two types of maize. Chu, Pellet and Nawar (1976) reported values of 133.2 µg of LAL per gram protein when maize was processed with 4.1 mol per kg of lime for 30 minutes at 170°F (76.6°C). The use of sodium hydroxide under equal conditions yielded higher levels of LAL. Since higher levels of LAL were obtained with NaOH and KOH, the authors suggested that calcium ions may in some way interfere with the mechanism of LAL formation. It is difficult to evaluate the significance of LAL formation during tortilla-making for people who eat relatively large amounts of this food daily. Since this has been practiced for a long time, the small amounts may not interfere with nutritive value or cause any pathological effects. Studies on the effect of lime level on the protein quality of maize have shown, however, that levels above 0.5 percent of grain weight reduce protein quality. The type of maize used and its size are of importance in this respect. Softer types of grains are more affected than hard grains cooked under similar conditions (Bressani et al., unpublished data).
Mycotoxins and alkaline-cooking of maize. The presence of mycotoxins in a variety of cereal grains and other foods and foodstuffs widely recognized, and maize is no exception. In Central America, where maize is such an important food, the grain is harvested twice a year in the tropical areas. One harvest is in August, when rain and temperature conditions are ideal for the growth of fungi. Martínez et al. (1970b) reported the presence of six different fungi in maize samples obtained from different markets throughout Guatemala. The frequency of Aspergillus versicolor was 57.1 percent; of Aspergillus wentii, 32.1 percent; of Aspergillus ruber, 26.8 percent; of Aspergillus echinulatus, 25.0 percent; of Aspergillus flavus, 25.0 percent; and of Chaedosporium spp., 26.8 percent.
Because of the significance of the presence of mycotoxins in cereal grains, a number of studies have been conducted to assess the degree of retention of mycotoxins during grain processing. The effect of calcium hydroxide cooking of maize has received some attention. Martínez-Herrera (1968) fed infected maize, raw and alkali-processed, to chickens and rats. The maize was infected with Fusarium sp., Penicillium spp., Aspergillus niger and A. flavus. The author found high mortality among birds fed on the raw infected maize, but none in the group of chickens fed the same maize processed with calcium hydroxide. In young rats, the raw and infected grain reduced weight gain and caused some mortality. Infected grain processed with lime induced no mortality, however, and weight gain as well as feed efficiency were like those in the control. Adult rats were also affected by the infected maize, but not by infected maize processed with lime. The study did not report levels of mycotoxins before and after processing.
Martinez (1979) reported on studies of tortilla samples collected in Mexico City in different seasons. He found that 15 to 20 percent of the samples collected in spring 1978 and in the rainy season of 1977-1978 contained aflatoxins. Furthermore, he found that concentrations of aflatoxins B1 varied from 50 to 200 ppb. He also indicated that lime-cooking of maize reduced aflatoxin concentrations by 50 to 75 percent. Martínez and also de Campos, Crespo-Santos and Olszyna-Marzys (1980) reported that lime concentrations of up to 10 percent were no more effective in reducing aflatoxins than a 2 percent concentration.
Ulloa-Sosa and Schroeder (1969) reported that the tortilla-making process was not effective in removing aflatoxins from contaminated maize. Nevertheless, others have obtained different results. For example, SolorzanoMendizabal (1985) found that maize inoculated with A. flavus and Aspergillus parasiticus produced high levels of aflatoxins which were reduced by lime-cooking, completely in some cases, but most often by up to 80 percent. Lime concentration varied from 0.6 to 8 percent, and analyses were done on maize, masa, tortillas and cooking waters. In another study, de Arriola et al. (1987, 1988), using QPM Nutricta, found that the lime levels at which nixtamal is normally prepared in Guatemala do not reduce aflatoxin in contaminated grain sufficiently to make it safe for human consumption.
Lime levels of 2 percent and above gave high aflatoxin reduction, but the tortillas were not acceptable. Aflatoxin B1 was reported to be reduced the most. Torreblanca, Bourges and Morales (1987) found relatively high aflatoxin levels in both maize and tortillas in a study conducted in Mexico City. Aflatoxin B1 was found in 72 percent of the maize tortilla samples tested; furthermore, 24 percent of the samples gave positive reactions for zearalenone. Carvajal et al. (1987) found mycotoxins in maize and tortillas in Mexican samples and indicated that aflatoxins, zearalenone and deoxynivalenol (DON) were not destroyed by the lime treatment or by temperatures of 110°C.
Price and Jorgensen (1985) found that the alkaline cooking process reduced aflatoxin levels from 127 µg per kg in raw maize to 68.6 µg per kg in tortillas. The authors concluded that the process was poorly effective, since the lower value obtained was still much above the value established as acceptable (about 20 mg per kg). These authors found that acidification - as it occurs in the intestinal tract - increased aflatoxin levels. Abbas et al. (1988) reported on the effect of 2 percent lime-cooking of maize on the decomposition of zearalenone and DON. They found significant reductions, i.e. 58 to 100 percent for zearalenone and 72 to 82 percent for DON. Furthermore, 15-acetyl-DON was completely destroyed.
Results obtained by various authors are somewhat conflicting, since some of them report partial reduction in some mycotoxins while others note total reduction. In many studies the mycotoxin levels were relatively high, necessitating stronger processing conditions in terms of lime concentration and cooking time. The problem warrants further study. Grain quality is probably the best means of ensuring the absence of mycotoxins rather than dependence on lime to reduce them partly or eliminate them in the final product.
Microbiological aspects of tortillas and tortilla flour. Studies on the microflora in lime-cooked maize tortillas are very limited. Capparelli and Mata (1975) showed that the main contaminants of tortillas as made in the highlands of Guatemala were coliforms, Bacillus cereus, two Staphylococcus species and many types of yeasts. When tortillas are first cooked, bacterial counts are about 103 or fewer organisms per gram, which is a safe level for consumption. After they are cooked for about five minutes on a hotplate they are placed hot in a basket, often covered with a cloth. This captures the vapour from the tortillas, creating an environment appropriate for microbial growth. After some ten hours under these conditions the surfaces of stacked tortillas become slimy and they are not acceptable for consumption.
Although there are many opportunities in rural areas for contamination during processing from maize to tortillas, the factors that possibly contribute the most are the water used during conversion of cooked maize to dough and the mill used to grind the cooked maize. Molina, Baten and Bressani (1978) reported a greater increase in bacteria counts in tortillas fortified with soybean flour and vitamins than in unfortified tortillas. In this case the mill used to grind the cooked maize to make the dough was chlorinated, which helped in lowering the bacteria count in the soy-supplemented maize. The tortillas made from it also had a lower bacteria count. The rate of increase in bacterial number decreased as well. Higher bacteria counts were reported by Valverde et al. (1983) in the cough and tortillas made from QPM Nutricta than in those from common maize, showing the effect of nutritional quality on bacterial growth.
The relatively high moisture content which is responsible for a very short shelf-life has limited marketing of tortillas. Nevertheless, there is a demand for them in urban areas, where they are marketed under refrigerated conditions. A number of attempts have been made to lengthen their shelf life. Rubio (1972a, 1972b, 1973, 1974a, 1974b, 1975) patented a number of methods which included various additives: epichlorohydrin and polycarboxylic acid and their anhydrides; hydrophilic inorganic gels; sorbic acid and its salts as well as the methyl, ethyl, butyl and propyl esters of para-hydroxy benzoic acid; and acetic and propionic acids. Pelaez and Karel (1980) developed an intermediatemoisture tortilla with a stable shelf-life. It was free from microbial growth, including Staphylococcus aureus, yeasts, moulds and enterotoxin. This was achieved through the use of glycerol, corn solids DE-42 and salt, as well as the mycostatic agent potassium sorbate. Protection with appropriate packaging was claimed for at least 30 days and the appearance, texture and other characteristics were similar to those of regular tortillas with a water activity of 0.97. Hickey, Stephens and Flowers (1982) reported relatively good protection of tortillas with low levels of sorbates or propionates added to the dough, and with a spray of sorbate on the surface (both sides) after cooking on the hot plate. More recently, Islam, Lirio and Delvalle (1984) claimed that using calcium propionate extended the shelf-life of tortillas at room temperature to 2 to 5 days; with dimethyl fumarate shelf-life was 2 to 11 days under the same storage conditions and using polythene bags. Although advances have been made in extending shelf-life, it still constitutes a problem for people who buy food in supermarkets.
Reports on the microbiology of tortilla flour and the tortillas made from it are not available. Lower total bacteria counts would be expected, however, because of the process employed to prepare the flour and use it at home.