This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.
Article Cite This: J. Agric. Food Chem. 2017, 65, 8311-8318
pubs.acs.org/JAFC
Concentration of Beneficial Phytochemicals in Harvested Grain of U.S. Yellow Dent Maize (Zea mays L.) Germplasm Carrie J. Butts-Wilmsmeyer, Rita H. Mumm, and Martin O. Bohn* Department of Crop Sciences, University of Illinois at Urbana−Champaign, 1102 South Goodwin Avenue, Urbana, Illinois 61801, United States ABSTRACT: Although previous studies have examined the concentration of various nutritional compounds in maize, little focus has been devoted to the study of commercial maize hybrids or their inbred parents. In this study, a genetically and phenotypically diverse set of maize hybrids and inbreds relevant to U.S. commercial maize germplasm was evaluated for its variability in phytochemical content. Total protein, unsaturated fatty acids, tocopherols, soluble phenolics, and insoluble-bound phenolics were evaluated in this study. Of these compounds, only soluble and insoluble-bound phenolic acids exhibited means and variances that were at least as large as the means and variances reported for other sets of germplasm. This suggests that selection for high phenolic acid content is possible in current U.S. commercial germplasm. In contrast, while the total protein, unsaturated fatty acid, or tocopherol content could possibly be improved using current U.S. commercial germplasm, the results of this study indicate that the incorporation of more diverse sources of germplasm would most likely result in quicker genetic gains. KEYWORDS: maize, phenolics, tocopherols, protein, plant breeding
■
INTRODUCTION Over the past decade, consumers have become increasingly interested in the natural nutritional value of their food products.1 This increased interest is not surprising. High obesity rates are common in affluent nations. The incidence of aging-related diseases in developed countries is rising.2 Furthermore, there is greater public concern that processed foods are less nutritious.3 Therefore, food companies must consider not only the calorie and vitamin content of their food products, but also the content of beneficial phytochemicals that occur naturally in their food products. In 2016, 2.8 billion bushels (71.12 billion kg) of maize were devoted to human food consumption,4 thereby highlighting the importance of maize in the American diet. For this study, the variability in the concentration of total protein content, fatty acids, tocopherols, and phenolics was examined in a set of maize lines, which are representative of the commercial maize germplasm grown in the U.S. Corn Belt. 5,6 All these phytochemicals are beneficial to human health. Diets higher in total protein result in satiety and weight maintenance. Since satiety is maintained for longer periods of time, people tend to eat less frequently.7 This reduced food consumption contributes to a lower overall caloric intake and a decreased chance of developing an obesity-related disease. Unsaturated fatty acids play important roles in blood pressure regulation, cell membrane integrity, and improving cholesterol levels.8,9 Oleic acid and linoleic acid are fatty acids important in the human diet and are found in maize. In contrast to the health benefits of protein and unsaturated fatty acids, tocopherols and phenolics exhibit antioxidant activity, prebiotic activity, and other chemopreventive properties.10−13 α-, γ-, and δ-Tocopherol are typically found in maize, whereas βtocopherol is rarely present.14 Most studies have focused on αtocopherol because it is the most efficient antioxidant of the three tocopherols present in maize.13 Like the tocopherols, many of the soluble phenolics exhibit antioxidant activity in vitro, but in © 2017 American Chemical Society
vivo, they are more likely to activate gene cascades involved in immunoprotection and chemoprevention.11,12,15 The predominant phenolics in maize are ferulic acid and p-coumaric acid, both of which are hydroxycinnamic acids.16,17 In cereals, these are mainly bound to the cell wall via ester bonds. Thus, insolublebound phenolics are not readily available for the activation of gene cascades. However, in conjunction with the arabinoxylan fibers to which they are bound, the insoluble-bound phenolics act as prebiotics by creating an environment beneficial for gut bacteria, including Lactobacillus and Bif idobacterium species.18 A precursor to the phenolics in maize is the carboxylic acid cinnamic acid. Cinnamic acid is stored in the vacuoles of plants19 prior to being modified in, most likely, the Golgi bodies. Cinnamic acid-4 hydroxylase acts on cinnamic acid, producing pcoumaric acid. p-Coumaric acid is acted upon by p-coumaric hydroxylase and then o-methyltransferase to produce ferulic acid. p-Coumaric acid and ferulic acid most likely are synthesized before esterification to arabinoxylans.20 It also appears that hydroxycinnamic acids are esterified to arabinoxylans in the Golgi bodies.21 Therefore, to understand the variability in the composition of the phenolics in maize, an understanding of cinnamic acid is useful. Our overall objective was to determine the potential for improving the nutritional content of maize hybrids that are representative of U.S. Corn Belt germplasm. In this study, we determined the average content and variability of total protein, unsaturated fatty acids, α-, γ-, and δ-tocopherol, and phenolics in U.S. Corn Belt maize inbred parents (i.e., inbreds) and a subset of their hybrids. Using the estimated means and variability in the phytochemicals of interest, we determined if this germplasm is Received: Revised: Accepted: Published: 8311
May 3, 2017 August 16, 2017 September 5, 2017 September 5, 2017 DOI: 10.1021/acs.jafc.7b02034 J. Agric. Food Chem. 2017, 65, 8311−8318
Article
Journal of Agricultural and Food Chemistry Table 1. Inbred Maize Parents Used in This Study and Additional Information Regarding Their Pedigree Background line
group
assignee
background
B73 LH1 PHG39 PHJ40 4676A LH123HT LH82 Mo17 PH207 PHG47 PHG84 PHZ51
SSS SSS SSS SSS SSS NSS NSS NSS NSS NSS NSS NSS
None (Public) Holden Foundation Seeds Pioneer Hi-Bred International Pioneer Hi-Bred International Dekalb Genetics Corporation Holden Foundation Seeds Holden Foundation Seeds None (Public) Pioneer Hi-Bred International Pioneer Hi-Bred International Pioneer Hi-Bred International Pioneer Hi-Bred International
Iowa Stiff Stalk Synthetic Iowa Stiff Stalk Synthetic; B37 type Maiz Amargo/Iowa Stiff Stalk Synthetic; B37/B14 type Iowa Stiff Stalk Synthetic 1067−1/B-Line Composite; B14 type Pioneer Hybrid 3535 Krug/W153 Lancaster Iodent/Long Ear OPV/Minn13 Oh43/Iodent*Wf9/MKSDTA C10 Synthetic Oh07-Midland/Minn13/Iodent/Reid YD/Osterland YD/Lancaster/Pioneer Female Composite OPV Minn13/Iodent/Reid YD/Osterland YD/Lancaster/South US Land Race Synthetic/Funks G4949/Midland
based on dry milling efficiency alone had PH207 as a parent. Considering that the content of many nutritional compounds in the maize grain, like starch, protein, and oil, is largely influenced by additive gene action,23 it was important that hybrids with PH207 as a parent were included as part of this study and that inferences were made on as diverse of a subset as possible. In addition, PH207 represents an important heterotic subgroup in commercial maize breeding germplasm. Reagents. Ethyl acetate, α-amylase, and methanol were purchased from Sigma-Aldrich (St. Louis, MO). Sodium hydroxide, sodim choloride, hydrochloric acid, and pyridine were purchased from Fisher Scientific (Pittsburgh, PA). N,O-bis(Tremethylsilyl)trifluoroacetamide (BSTFA) was obtained from Thermo Scientific (Waltham, MA). Ferulic acid, p-coumaric acid, sinapic acid, and cinnamic acid was purchased from both Sigma-Aldrich (St. Louis, MO) and Fisher Scientific (Pittsburgh, PA). Determination of Protein Content. Whole grain samples (8% moisture) were measured using nondestructive NIR techniques. Calibration curves were built based on known standards. The remainder of the procedure was conducted following the direction of the DA 7200 Diode Array Analyzer Operation Manual24 and Bubert.25 Determination of Fatty Acid Content. The total oil content of ten random kernel maize samples and their corresponding ground samples was measured, and these data were used for calibration of the Nuclear Magnetic Resonance (NMR) equipment.26 Ground samples underwent a hexane wash. The hexane layer was removed, placed in a separate vial, and evaporated, which left only the oil extracted from a particular sample in the bottom of the vial. All vials were weighed before extraction and after evaporation. The difference in the mass is equal to the mass of oil extracted for that particular sample. Dividing this mass by the mass of ground maize used in the extraction gives the total oil content in g oil g−1 maize sample. Three technical replicates were used for each of the ten training samples, and the average content was calculated for the sample using these technical replicates. These values and their corresponding whole kernel samples were used to calibrate a prediction curve using the NMR. The calibration was performed by Dow AgroSciences. The total oil concentration of the 225 samples was estimated using NMR. Approximately 0.5 g of each sample was used for fatty acid quantification. The sample was placed in a vial, and 3 mL of hexane was added. The vials were capped, placed in a sonicator at 75 °C for 10 min, and then put on a horizontal shaker for 10 min. Each vial was then vortexed before being centrifuged at 465g for 5 min. The top hexane layer was removed and placed in a clean vial. The hexane wash was repeated two more times. The hexane extract was dried using a centrifugal evaporator. The oil remaining in the vial was reconstituted in 1 mL of heptane, and 300 μL of the heptane/oil mixture was placed in a gas chromatography vial and derivitized using 20 μL of sodium methoxide. Subsequently, the samples underwent a fatty acid methyl ester (FAME) analysis. The proportion of total oil content was recorded for each fatty acid. The following equation was used to calculate the concentration of each fatty acid:
well suited for improving the nutritional quality of yellow dent maize hybrids or if the other germplasm sources must be tapped to access favorable alleles. Using the literature base, we also discuss the potential fate of these phytochemicals during maize food product processing and consider which phytochemicals may be the most feasible to improve using plant breeding techniques.
■
MATERIALS AND METHODS
Plant Materials. The hybrids evaluated in this study were created from a genetically diverse set of inbreds that represent important heterotic subgroups in the current U.S. maize commercial germplasm base (see Table 1).5,6 A heterotic group is a group of parental genotypes whose observed combining ability and heterosis are similar when crossed with parental genotypes from other groups.22 Heterotic groups can be broken into smaller sets called subgroups. Five of the inbreds (B73, LH1, PHG39, PHJ40, and 4676A) were derived from the Iowa Stiff Stalk Synthetic (SSS), which represents the female side of the heterotic pattern used in the U.S. Seven other inbreds (LH123HT, LH82, Mo17, PH207, PHG47, PHG84, and PHZ51) are of Non-Stiff Stalk (NSS) origin and are composed largely of Iodent, Minnesota 13, Oh07, Midland Yellow Dent, Oh43, Lancaster, and broad-based subgroups which generally represent the male side of the heterotic pattern used in the U.S.6 Using a diallel mating design, 66 hybrids were created from the 12 inbreds used in this study. All inbreds and hybrids (NTotal = 78 entries) were grown at the University of Illinois Crop Sciences Research and Education Center in Urbana, IL in the summer seasons of 2009, 2010, and 2011.6 Plots were composed of four rows with length of 5.3 m and which were 0.76 m apart at a plant density of approximately 74 000 plants ha−1; only the center two rows were harvested for grain. Plots were managed in accordance with regional agronomic production practices (i.e., no irrigation, minimum tillage, and standard fertilizer regime). Three replications of the 78 entries were grown in each of the three years. Of the 78 entries, 25 of those entries were chosen for use in this study. Exactly 454 g of harvested grain from each plot was dried to a moisture content of approximately 8%, bagged, and placed in cold storage prior to nutritional analyses. From each bag, a 100 g of grain sample was ground to a fine powder using a Foss Cyclone Mill (1 mm mesh screen). We used this powder in all wet-lab chemistry. Subset Determination. The available resources allowed us to analyze 25 entries, which amounted to nine samples per entry and a total number of 225 samples. For this study, we used all 12 parental inbreds and 13 hybrids. The set of 13 hybrids was composed of the five highest and the five lowest performing SSS × NSS hybrids in relation to their dry milling efficiency and three hybrids that have Iodent parent PH207 in common. Dry milling efficiency is the proportion by weight, on a dry basis, of flaking grits obtained from one kg of dry milled grain (for more information see Macke et al.6). Three hybrids with PH207 parentage were included because preliminary analyses (data not shown) indicated that the inbred PH207 displayed considerably high content of some of the phytochemicals of interest, yet none of the ten hybrids selected 8312
DOI: 10.1021/acs.jafc.7b02034 J. Agric. Food Chem. 2017, 65, 8311−8318
Article
Journal of Agricultural and Food Chemistry FAij =
PTOij × TOj mj
×
coumaric acid, cinnamic acid, and sinapic acid were also used in creating the standard curve. Samples were analyzed using a gas chromatograph with a flame ionization detector (Agilent 6890). A DB-200 megabore column (ID = 0.53 mm, length = 30 m) was used. Hydrogen gas was used as the carrier gas with a split inlet and a split ratio of 1:10 (split vent flow = 10 mL/min, column flow = 1 mL/min). Statistical Analyses. An analysis of variance (ANOVA) was performed based on the following linear mixed model:
1000 mg 1g
where FAij is the concentration of the ith fatty acid and the jth observation in mg per g of maize sample, PTOij is the proportion of total oil content for the ith fatty acid and the jth observation, TOj is the total oil content, in g, for the jth observation, and mj is the recorded mass of the jth observation in grams. Determination of Tocopherol Content. Following the removal of 300 μL from the heptane aliquot for the FAME analysis, the heptane/oil mixture was evaporated to dryness. Gas chromatography was used for quantification of tocopherols in the oil extracts. The total mass of tocopherols (μg) remaining was reported by the gas chromatography system. To calculate the concentration in the original sample, the following equation was used:
Tij =
Tij′ 0.7 × mj
×
yijk = μ + Yi + R (i)j + τk + Yτik + εijk where yijk is the observed phenotypic value corresponding to the ith year, the jth replication within the ith year, and the kth genotype, μ is the grand population mean, Yi is the random effect of the ith year, R(i)j is the random effect of the jth field replication nested within the ith year, τk is the fixed effect of the kth genotype, Yτik is the random interaction between the ith year and the kth genotype, and εijk is the random error term associated with the ith year, the jth replication within the ith year, and the kth genotype. The random error effects εijk are assumed to be normally and independently distributed with mean 0 and variance σe2. Inbreds and hybrids were analyzed separately using this statistical model in SAS PROC MIXED (version 9.3). Because of adverse field conditions in 2010 and 2011, a balanced set of inbred samples was not available for analysis. Therefore, the REML method was used to conduct the ANOVA and calculate p-values associated with the genotypic effect as well as the standard error of the differences between inbred means. The REML method was also used to conduct the ANOVA for the tocopherols because several samples had tocopherol contents below the GC detection level, leading to an unbalanced data set for the tocopherol analyses. The assumptions of normality and homogeneous variances were validated by conducting Shapiro-Wilk’s test of normality on the residuals in PROC UNIVARIATE and by using the Brown-Forsythe modification of the Levene test in the MEANS option of PROC GLM. If a highly significant year-by-genotype interaction existed and the average concentration was high enough to warrant further investigation of the compound, multidegree of freedom contrast statements were used to examine the genotype effect in each year. If phytochemicals displayed either a mean or a variance that was at least as large as what is found in other germplasm sources reported in the literature, the relationship between years was measured using Pearson correlation coefficients, which were calculated in PROC CORR.
1 × 106μg g
where Tij is the actual tocopherol content of the ith tocopherol and the jth sample and T′ij is the reported content of the ith tocopherol in the jth sample before the correction. We applied a correction factor of 0.7, assuming that approximately 70% of the originally extracted tocopherols remained in the oil extract. Determination of Soluble Phenolics Content. Following the hexane wash, the maize pellet was dried using a centrifugal evaporator. Then 3 mL of 70% acetone was added to each vial. Vials were placed in a sonicator for 10 min at 75 °C and then placed on a horizontal shaker for 10 min. Vials were then vortexed and centrifuged at 465g for 5 min. The top acetone layer was removed and put in a separate vial. The acetone wash was completed two more times for a total of three acetone washes. The extracted acetone aliquot was then evaporated to dryness using a centrifugal evaporator. Soluble extracts were reconstituted in 2 mL of methanol, vortexted, and filtered using syringe filters. One milliliter of the filtered solution was then placed in an ultra performance liquid chromatography (UPLC) for analysis. A Waters Acquity UPLC H-Class with an Acquity BEH Shield RP18 (2.1 × 150 mm2, 1.7 μm) column was used for the quantification of the soluble phenolics content. The column temperature was maintained at 25 °C. For the first 1.5 min, 60% acetonitrile with 0.1% formic acid and 40% methanol was used for the elution at a flow rate of 0.2 mL/min. For the time interval between 1.5 and 11 min, 50% acetonitrile with 0.1% formic acid and 50% methanol was used for the elution at a flow rate of 0.2 mL/min. For the time interval between 11 and 12 min, 60% acetonitrile with 0.1% formic acid and 40% methanol was used for the elution at a flow rate of 0.2 mL/min. An Acquity UPLC LG 500 nm photodiode array detector was used for UV detection of analytes at 300, 270, and 245 nm. Data signals were recorded and processed using Empower 3 Software (Build 3471) (Waters Corp.). Standards of the typical hydroxycinnamic acids in grains, that is, ferulic acid, p-coumaric acid, sinapic acid, and cinnamic acid, were first analyzed using UPLC to record the retention time of each of these compounds and to build a standard curve for the calculation of concentrations in the maize samples. Our maize samples were then analyzed using UPLC using the protocol as described, and the standard curve was used to calculate the concentration of the hydroxycinnamic acids in each sample using the following equation:
Sij =
■
RESULTS Total Protein Content. The mean total protein content of hybrids and inbreds was found to be 9.12% and 11.51%, respectively (Table 2). The protein content observed in this study ranged between 8.34% and 10.22% in the hybrids and between 9.86% and 13.24% in the inbreds. Unsaturated Fatty Acid Content. In the hybrid group, oleic acid varied between 7.21 and 12.22 mg g−1, whereas linoleic acid ranged from 17.38−21.71 mg g−1. In the inbred group, oleic acid ranged between 4.77 and 9.51 mg g−1, whereas linoleic acid ranged between 15.92 and 22.90 mg g−1. The hybrids examined in this study varied between 21.2% and 33.4% for oleic acid and between 50.4% and 62.7% for linoleic acid content in the oil (Table 2). Tocopherols. The mean contents of α-, γ-, and δ-tocopherol in the hybrids were 5.42 μg g−1, 21.74 μg g−1, and 1.47 μg g−1, respectively (Table 2). The mean contents of α-, γ-, and δtocopherol in the parental inbreds were 5.70 μg g−1, 18.68 μg g−1, and 1.16 μg g−1, respectively (Table 2). The tocopherol values observed in the hybrids differed very little from the values observed in the inbreds. Soluble and Insoluble-Bound Phenolics. Soluble ferulic acid and p-coumaric acid were consistently detected at low concentration levels using UPLC. The average concentrations of
2 × Sij′ mj
where Sij is the content of the ith soluble phenolic in the jth observation and S′ij is the recorded mass of the ith soluble phenolic in the jth observation as calculated from the standard curve, and mj is the recorded mass of the jth observation. S′ij was multiplied by a factor of 2 to account for the reconstitution of the soluble phenolics in 2 mL of methanol. Determination of Insoluble-Bound Phenolics Content. The measurement of the insoluble-bound phenolics content took place using the protocol, including α-amylase digestion, as described by ButtsWilmsmeyer and Bohn27 with the exception that standards of p8313
DOI: 10.1021/acs.jafc.7b02034 J. Agric. Food Chem. 2017, 65, 8311−8318
Article
Minimum, maximum, and average values reported are based off of the LSMEANS of the hybrids and inbreds. bStandard error reported is the standard error of the difference between two entry means. Because of unequal replication, the range of standard errors between the means is reported for the inbreds. cP-values reported are those associated with the genotype effect in the mixed model.
18.68 4.13 34.01 (4.45, 12.39) 0.02 5.70 2.57 10.03 (1.05, 2.46)
