Nutritional composition and glycemic index analyses of vitamin A‐biofortified maize in healthy subjects

Abstract Besides being a veritable tool for easing the problem of vitamin A deficiency (VAD), this study sought to explore another potential health benefit of vitamin A‐biofortified maize (VABM). In the present study, the nutritional composition and glycemic index (GI) of tuwo masara (a nonfermented maize‐based dumpling), made from VABM and the indigenous white maize (IWM) genotype, were evaluated. VABM showed significantly (p < 0.05) lower fat (4.38 ± 0.46%) and crude protein (6.58 ± 0.13%) but higher crude fiber (5.29 ± 0.0%) contents compared to 5.22 ± 0.25% crude fat, 7.28 ± 0.11% crude protein, and 4.69 ± 0.00% crude fiber in the IWM. The phytic acid content in the IWM (2.77 mg/100 g) was 39% higher than the level (2.0 ± 0.04 mg/100 g) in VABM. The major provitamin A carotenoid in the VABM were lutein (7.37 ± 0.52 μg/g), zeaxanthin (1.65 ± 0.01 μg/g), cryptoxanthin (1.29 ± 0.02 μg/g), and all‐trans‐β‐carotene (0.83 ± 0.02 μg/g), while the IWM contained only lutein (1.52 ± 0.32 μg/g). The total carotene concentration, 12.74 ± 1.13 μg/g dry weight in the VABM, was over eight times higher than that observed for the IWM, 1.52 ± 0.32 μg/g dry weight. The VABM tuwo masara showed a significantly lower GI value (70.3%) compared to the IWM tuwo masara (87.7%). Data obtained from the study further attest to the positive nutritional and health benefits of VABM.


| INTRODUC TI ON
Vitamin A deficiency (VAD) is a major public health concern among preschool children and women in developing countries (Arlappa et al., 2011;Bowley, 2008). Consequently, an estimated average of one out of every three preschool children and one quarter of pregnant women in Nigeria are vitamin A deficient (Maziya-Dixon et al., 2006). Therefore, several intervention strategies have been posited to address the devastating consequences of VAD on children (Bouis, Holz, McClafferty, Meenakshi, & Pfeiffer, 2011).
Using traditional plant-breeding and bioengineering strategies, high provitamin A, yellow maize genotypes have been genetically modified to accumulate high levels of pro-vitamin A and other carotenoids, as well as other micronutrients (Ortiz-Monasterio et al., 2007). Since its introduction in 2005, the seeds of this improved maize variety are continuously being distributed to local farmers who in turn plant and share these seeds over generations as a costeffective, renewable means to reduce micronutrient deficiencies (Pfeiffer & McClafferty, 2007). Consequently, it is expected that enhanced provitamin A contents in the maize kernel will have a farreaching health impact in Africa given its level of consumption by the different strata of the society, including children and women who are most susceptible to VAD (Bouis et al., 2011).
Tuwo masara (a non-fermented maize-based dumpling) is a popular food product made from maize in Nigeria. It is consumed in the northern part of Nigeria and other Hausa-speaking West African communities (Boladale, Usman, Rasheed, Benson, & Salifou, 2002). It is a gel-like food product made from a combination of maize flour, water, and heat (Boladale & Adeyemi, 2014). It is eaten with different type of soups, including vegetable, okra, or ogbono soup. In spite of its nutritional and health-promoting potentials, the current trend in terms of acceptability and production of vitamin A-biofortified maize (VABM) is low when compared with the indigenous white maize (IWM) varieties in Nigeria. This could be attributed to the low level of awareness of the nutritional attributes of these new maize varieties among the masses.
Thus, in addition to the current efforts at promoting the cultivation and consumption of VABM through aggressive extension services, experimental research aimed at validating the acclaimed nutritional potential of these new maize varieties could further provide a boost to the current level of its consumer acceptability. Hence, this study was aimed at carrying out a comparative nutritional analysis in terms of proximate composition, antinutritional composition, and total carotenoid level between maize flour made from VABM and IWM varieties, and, in addition, evaluates the glycemic index of tuwo masara made from these maize flour samples, respectively, in healthy subjects.

| Proximate composition
Proximate composition analysis of the respective flour of the two maize varieties (VABM and IWM) was analyzed following standard methods (AOAC, 1995). Percentage amylose, sugar, and starch contents were determined following the procedures of Juliano et al. (1981).

| Anti-nutritional factor
Phytate, tannin, and cyanogenic potential levels in the respective maize flour were determined using the method of AOAC (1995).
The colorimetric method of Vaintraub and Lapteva (1988) and modified by Gao et al. (2007) was employed in the quantification of the phytate content using phytic acid solution to obtain the standard curve. The total polyphenols content and tannins were determined by the method of Marigo (1973). The optical densities were read at 725 nm. Total dietary fiber content was estimated using the enzymatic-gravimetric method described by Asp, Johansson, Hallmer, and Siljestom (1983).

| Total carotenoids
Carotenoid extraction was done according to the procedure previously described by Rodriguez-Amaya and Kimura (2004) with little modifications and the individual carotenoid quantified against a calibration curve obtained using standard carotenoids based on the observed peak areas.

| Preparation of tuwo masara
Tuwo masara (i.e., maize tuwo) was prepared according to the procedure previously described by Boladale et al. (2002). The tuwo was prepared from dried maize in which the outer coat (testa) has been removed by grinding gently inside mortar with pestle. A small quantity of water was added to the grains to facilitate the testa removal process. Thereafter, the grains in which the testa has been removed were sun-dried, milled using a mechanical grinder to obtain smooth, whitish flour. The flour was sieved and added to hot boiling water with continuous stirring to form a thickened gel-like paste herewith referred to as "tuwo masara."

| Study population
The study was undertaken at Joseph Ayo Babalola University, Ikeji Arakeji (Nigeria), a fully residential private University. Sixty-five apparently healthy, nondiabetic human subjects (23 males and 42 females) aged between 17 and 39 years were recruited for the feeding trial upon obtaining their informed consent.

| Glycemic index analysis
The glycemic index (GI) was calculated according to the formula described by Jenkins et al. (1981) as follows:

| Statistical analysis
Data are expressed as mean ± SD or SEM as applicable. Statistical comparisons between samples/subjects were made by paired

GI (%) =
Incremental area under the 2-h glucose response curve for a 50-g carbohydrate equivalent of the test food Incremental area under the 2-h glucose response curve for a 50-g glucose Student's t tests or one-way ANOVA as applicable. The level of statistical significance was set at p < 0.05.

| Proximate composition
The proximate composition profile for the VABM and IWM used in this study as presented in Table 1 showed a nonsignificant difference in terms of moisture (8.66 ± 0.25%, 8.33 ± 0.23%) and ash (1.23 ± 0.11%, 1.34 ± 0.04%), respectively. However, percentage fat and crude protein contents were significantly higher in the IWM compared to the VABM. Crude fiber content was significantly lower in the IWM (4.69 ± 0.00%) compared to the VABM (5.29 ± 00%). There was no significant variation in percentage total carbohydrate, sugar, starch, amylose, and amylopectin contents in the two maize genotypes. In addition, the percentage amylose-toamylopectin ratio was not significantly different in the two maize samples.

| Anti-nutritional factors
The anti-nutritional profile of the IWM and VABM is shown in Table 2. The results showed that the phytate concentration in VABM (2.0 ± 0.04 mg/100 g) was significantly lower compared with the value recorded for IWM sample (2.77 ± 0.09 mg/100 g). The observed total polyphenolic content and cyanide cyanogenic potential level in the two maize samples were not statistically significant.
Total dietary fiber content in the VABM was significantly higher (9.62 ± 0.12) compared to the level in the IWM (6.80 ± 0.25).

| Total carotenoids
The carotenoid content of the two maize samples is shown in Table 3.
The total carotene concentration, 12.74 ± 1.13 μg/g dry weight in the VABM, was over eight times higher than that observed for the IWM, 1.52 ± 0.32 μg/g dry weight. The chromatogram showing the carotenoid arrays for the IWM and VABM samples is shown in Figure 1a,b, respectively.

| Glycemic index
The mean glycemic responses to glucose solution and the test foods are shown in Figure 1. The blood glucose response curve to glucose solution and tuwo masara prepared with the IWM and VABM genotypes peaked at 60 min. The postprandial blood glucose responses

| D ISCUSS I ON
The data obtained in this study showed that the VABM flour is lower in crude fat, crude protein, and phytic acid but higher in fiber compared to the IWM. In addition, the VABM contained the major provitamin A carotenoids including β-cryptoxanthin, 13-cisβ-carotene, 9-cisβ-carotene, and all-transβ-carotene unlike the IWM which is devoid of the major provitamin A carotenoids but for a low amount of zeaxanthin and β-cryptoxanthin. The GI of the VABM tuwo masara was observed to be lower compared to the IWM tuwo masara. ). Theoretically, amylose should be easier to digest due to the absence of branched points within its structure and so only one enzyme, amylase, is required for its hydrolysis. However, amylose often forms a very compact physical structure, which inhibits digestion (El-Harith et al., 1976;Riley et al., 2004). Plasma glucose response has been reported to significantly decrease following consumption of high-amylose but low-amylopectin starch diet (Juliano & Goddard, 1986). Earlier investigations by Goddard, Young, and Marcus (1984) showed that plasma glucose response was significantly lower when a high-amylose rice (Labelle variety, 24% amylose) was fed compared to sweet rice (Mochigome variety, 1% amylose) or glucola drinks.
The presence of antinutritional factors in the maize samples in this study is of significance due to their reported negative health impacts on humans and animals. Experimental evidences from animal studies have shown that phytic acid, when present in plant food, forms complexes with dietary essential minerals such as calcium, zinc, iron, and magnesium and makes them biologically unavailable for absorption (Cheryan & Rackis, 1980;Zhou & Erdman, 1995). High concentration of phytic acid in food has been reported to induce hypocalcemia (Checke & Shull, 1985). The phytic acid level in the two maize genotypes used in the present study is low and well below the acceptable range for seeds and legumes. Studies have shown that the glycemic response of foods varies inversely with the level of phytic acid in the food (Yoon, Thompson, & Jenkins, 1983). The observed total polyphenol content in both the IWM and VABM in this study (1.33-1.43 mg/100 g) was found to be relatively low in comparison with levels found in some literature (Akinyede, Amoo, & Eleyinmi, 2005;Enujiugha, 2003;Umoren, Essien, Ukorebi, & Essien, 2005).
The presence of polyphenols in plant foods has been reported to TA B L E 4 Plasma glucose response indices following consumption of indigenous white maize (IWM) tuwo and vitamin A-biofortified yellow maize (VABM) tuwo F I G U R E 2 Mean glycemic responses of subjects fed 50 g portion of tuwo prepared with the indigenous white maize (white maize tuwo) and the vitamin A-biofortified yellow maize (yellow maize tuwo) interfere with the activities of some starch-digesting enzymes such as glucosidases and maltase (Hanhineva et al., 2010). A number of intervention studies have reported considerable improvement in glycemic status in both rats and humans following supplementation with polyphenol-rich diets (Abesundara, Matsui, & Matsumoto, 2004;Almoosawi, Tsang, Ostertag, Fyfe, & Al-Dujaili, 2012). The HCl released as a by-product in the metabolism of cyanogenic glycosides has been implicated in central nervous system dysfunction, respiratory failure, and cardiac arrest (D'Mello, 2000). In addition, the presence of cyanogenic glycosides in animal feeds has also been shown to reduce the amount of metabolizable energy in such feeds (Kumar, 1992). The cyanide content of the two maize varieties used in this study is low and falls within the range considered safe for both humans and other animals.
The data obtained from this study showed that the IWM is devoid of β-carotene but contained a significantly low amount of zeaxanthin and β-cryptoxanthin compared to the vitamin A-biofortified yellow maize genotype. On the other hand, the VABM contained the major provitamin A carotenoids including β-cryptoxanthin, 13-cisβ-carotene, 9-cisβ-carotene, and all-transβ-carotene.
The total provitamin A carotenoid in the VABM maize reported in this study though higher than the earlier reported range of value (0.25-2.5 μg/g dry weight) reported for a typical yellow maize variety (Berardo, Mazzinelli, Valotti, Lagianna, & Redaelli, 2009;Nuss & Tanumihardjo, 2010) falls short of the expected breeding target of 15 μg/g dry weight for biofortified maize by HarvestPlus (Ortiz-Monasterio et al., 2007). Naturally, maize has been reported to exhibit variation in their total carotenoid content, with some genotypes reported to contain total carotenoids level as high as 80 μg total carotenoids/g dry weight (Menkir, Liu, White, Maziya-Dixon, & Rocheford, 2008). The fraction of provitamin A carotenoid of the total carotenoid is usually between 10% and 20%; however, zeaxanthin and lutein each most time represent 30%-50% of total carotenoids in maize (Ortiz-Monasterio et al., 2007). However, majority of the yellow maize varieties grown and consumed throughout the world contain less than 2 μg provitamin A carotenoids/g dry weight.
Vitamin A protects the body against diet-related chronic diseases as well as reduces the body's predisposition to cataracts, age-related macular degeneration, and other degenerative diseases (Bertram, 1999;Johnson, 2002 Evidence from experimental studies showed that enhanced intake of dietary fibers effectively neutralized the abnormal increase in blood glucose level following a high-carbohydrate diet (Wenk, 2001). Several mechanisms have been proposed to explain the hypoglycemic action of dietary fibers. Dietary fibers may slow the rate of transit of sugars from the stomach to the small intestine, thus working against a rapid increase in blood glucose concentration following sugar ingestion (Schulze et al., 2004). Increased consumption of fibers has been shown to effectively flatten the sudden surge in blood glucose level following a meal (Salmeron et al., 1997;Schulze et al., 2004). Thus, the high content of total dietary fibers observed in the VABM in this study could be of significance in blood glucose control both in healthy and diabetic subjects.
From the foregoing, the lower GI value observed for the VABM tuwo as against tuwo made from the IWM in this study could be attributed to their varying contents of fiber, phytic acid, and provitamin A carotenoids. Reports from previous experimental studies have shown that increased consumption of carotenoid resulted in significant reduction in plasma glucose concentration. Ford, Will, Bowman, and Narayan (1999) in a population-based US third National Health and Nutrition Examination Survey observed that serum total carotenoids exhibited an inverse relationship with serum glucose as well as insulin concentrations. In another separate study by Facchini, Humphreys, Donascimento, Abbasi, and Reaven (2000), serum carotenoid concentrations were observed to negatively correlate with insulin resistance and blood glucose concentrations. Furthermore, increased carotenoid intake has been reported to lower the risk of chronic diseases as well as cardiovascular diseases (Bertram, 1999;Fawzi et al., 2000;Johnson, 2002).

| CON CLUS ION
Based on the GI value obtained in this study, the tuwo made with VABM could be considered a moderate glycemic index product, based on Englyst and Hudson (1996) classification. Thus, in addition to being a veritable tool for alleviating VAD, the VABM could be considered an alternative energy source for subjects undergoing weight reduction and glycemic control.

ACK N OWLED G M ENTS
The generous gift of the two maize genotypes (TZL COMP4 C2 and BRY 9928 DMR SR) used in this study from the Crop Multiplication Unit of the International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria, is well acknowledged.