| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Horticulture and Crop Sci. Dep., California Polytechnic State Univ., San Luis Obispo, CA
b Dep. of Crop Sci., Univ. of Illinois at Urbana-Champaign, Urbana, IL 61801
c Newe Ya'ar Research Center, ARO, Ramat Yishay, Israel
* Corresponding author (trochefo{at}uiuc.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-Tocopherol, 
-tocopherol, and 
-tocopherol are the most abundant isomers present in maize. -Tocopherol has the highest biological activity for the tocopherols; however, -tocopherol is generally present in much higher concentrations than the other tocopherols in maize kernels. The objectives of this research were to determine levels of tocopherol isomers in maize kernels from two related maize mapping populations and to map quantitative trait loci (QTL) controlling accumulation of tocopherol isomers. Two sets of materials were assayed in this research, a (W64a x A632) F2:4 mapping population and testcross progeny with AE335. Molecular markers were evaluated on the F2:4 mapping population and a linkage map created. Each population was analyzed by high performance liquid chromatography (HPLC) for -, -, and -tocopherol. Composite interval mapping (CIM) identified QTL for individual tocopherols and total tocopherols, notably on chromosomes 1 and 5. The ratio of -/-tocopherol was analyzed to identify QTL influencing the conversion of -tocopherol to -tocopherol. The QTL associated with the largest percentage of phenotypic variation in the study was detected on chromosome 5 for ratio (/) tocopherol. The QTL identified in this study have potential for use in a marker assisted selection program to facilitate increasing levels and ratios of tocopherols in maize grain.
Abbreviations: CIM, composite interval mapping • cM, centimorgan • EST, expressed sequence tag • HPLC, high performance liquid chromatographer • LOD, likelihood of the odds • QTL, quantitative trait loci • SSR, simple sequence repeat
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-tocopherol activity (Bramley et al., 2000; Eitenmiller, 1997). The eight compounds can be divided into two distinct groups, tocopherols and tocotrienols. The two groups differ in saturation of the side chain. Both tocopherols and tocotrienols have four derivatives: alpha (), beta (ß), delta (), and gamma (), which differ in the number and location of methyl groups (Fig. 1). Vitamin E was first discovered in 1922 as a macronutrient essential to reproduction in rats (Brigelius-Flohe and Traber, 1999). Since then other important roles of vitamin E compounds in plants and animals have been found.
|
The function of tocopherols in animal systems is generally related to the level of -tocopherol activity. Tocopherols have the ability to quench free radicals in cell membranes, protecting polyunsaturated fatty acids from damage. An imbalance in the production of free radicals and the natural protective system of antioxidants may lead to oxidized products, capable of harming tissues (Food and Nutrition Board and Institute of Medicine, 2000). Tissue damage from free radicals is considered to be related to chronic diseases such as cardiovascular disease, neurological disorders, cancer, cataracts, inflammatory diseases, and age-related macular degeneration (Bramley et al., 2000).
Natural -tocopherol has a higher biological activity than other tocopherols as well as the synthetic form of -tocopherol (Brigelius-Flohe and Traber, 1999; Food and Nutrition Board and Institute of Medicine, 2000). -Tocopherol has the highest activity in mammalian tissues; one -tocopherol molecule is estimated to be effective in protecting 2000 phospholipids (Bramley et al., 2000). The high level of -tocopherol activity may be due to the method of absorption of vitamin E by the body. Vitamin E is absorbed with the lipids in the intestine and transported to the liver. In the liver, -tocopherol is specifically absorbed into the body by the hepatic -tocopherol transfer protein. Because of the efficiency of the transfer protein, -tocopherol is present in plasma at 10x the level of -tocopherol.
The tocopherol biosynthetic pathway is not well characterized, but a probable model pathway has been formulated (Fig. 1). Most research has been performed on Arabidopsis with three of the six genes in the tocopherol biosynthetic pathway cloned and functionally characterized (Schledz et al., 2001). One gene cloned in Arabidopsis may prove particularly important for maize. Because of the high-level of -tocopherol versus -tocopherol in maize, it may be useful to convert forms, giving a higher level of the more biologically active -tocopherol. Shintani and Dellapenna (1998) cloned -tocopherol methyl transferase (-TMT), a gene from Arabidopsis that converts -tocopherol to -tocopherol. When -TMT was transgenically overexpressed in the Arabidopsis plant, the amount of -tocopherol present increased 80 times (Shintani and Dellapenna, 1998).
Most research on vitamin E has concentrated on -tocopherol, but more recently there is growing interest in the benefits of a high -tocopherol diet. Initial studies showed -tocopherol levels in plasma were lower in subjects with coronary heart disease (Kontush et al., 1999). Low plasma level of -tocopherol could be overcome with a supplement of corn oil, which increased serum -tocopherol levels in healthy women (Lemcke-Norojarvi et al., 2001). The antioxidant capacity of -tocopherol in corn oil may possess a high antioxidant capacity relative to -tocopherol (Tomasch et al., 2001). Another crucial role of -tocopherol is its ability to quench peroxynitrite effectively, an electrophilic mutagen capable of damaging lipids, DNA, and proteins (Brigelius-Flohe and Traber, 1999; Kaur and Kapoor, 2001).
Tocopherol supplements in livestock rations have been shown to affect meat quality. High tocopherol levels in poultry, hogs, and beef cattle tissues have been associated with positive effects with economic ramifications. In poultry, vitamin E supplements increased the stability of poultry meat (Dewinne and Dirinck, 1996; Williams, 1997). In hogs, vitamin E protects against rancid flavor, odor, and discoloration and plays a part in increasing shelflife of packaged meat (Buckley et al., 1995; Dirinck et al., 1996). Vitamin E supplements in beef cattle increased color stability of beefsteaks, which increased visual acceptance (Chan, 1995). Tocopherols, in particular -tocopherol, have also been demonstrated to prolong shelf life of oils (Tomasch et al., 2001).
Maize kernels have been evaluated for their different levels and isoforms of tocopherols with early studies focused on -tocopherol, -tocopherol, and -tocopherol. -Tocopherol was generally found in the highest concentration (Grams et al., 1970; Weber, 1984). The germ contains 70 to 86% of total tocopherols; with endosperm having 11 to 27%, though levels of tocopherol storage are genotype dependent (Grams et al., 1970). A high amount of variation for tocopherol levels is present within different inbreds, as well as for different ratios of tocopherols (Galliher et al., 1985; Weber, 1984). Heritability estimates for and -tocopherol and total tocopherols should allow for increases with selection (Galliher et al., 1985).
We used two mapping populations to quantify the levels and forms of tocopherols present and to locate chromosomal regions with QTL associated with variability in tocopherol levels in maize kernels. Study of QTL related to a biosynthetic pathway that has been characterized, as has been done for the flavonoid pathway and maysin production (McMullen et al., 2001), provides advantages in efforts to characterize the genetic basis underlying quantitative variation for a trait. Through the identification of QTL with effects on tocopherol biosynthesis we hope to gain a better understanding of the genetic controls that regulate the levels and forms of tocopherol in the maize kernel. This information should be useful in efforts to increase overall levels of tocopherols and alter the ratio of and -tocopherol in maize grain, depending on the desired end usage of the grain.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-tocopherol and W64a had high levels of -tocopherol (Weber, 1987). The two parents were crossed to create an F1, the F1 was self pollinated to produce the F2, and F2 plants were self pollinated to create a mapping population consisting of 200 F2:3 families. The 200 families were evaluated as lines and in testcrosses. The lines were crossed to AE335 (Alexander Elite) to produce testcrosses. AE335 was chosen for its high oil level in the kernel. Five to six plants in a row were hand pollinated with the tester, with the tester as the male and the lines as females. Pollinated ears were harvested and seed was bulked within lines at shelling. A total of 185 testcross families were used for planting of testcross evaluations. All plots were grown at the University of Illinois Crop Sciences Research and Education Center at Urbana, IL.
Field Evaluation
The 200 F2:3 families and six inbred checks were grown in 1996, and 1997. Checks were the two parents A632, W64a, and four lines chosen for differences in carotenoids and tocopherols: B37, A619, and two experimental lines. The F2:3 families plus checks were planted in an (0,1)-design, with 42 blocks and five families per block with two replications each year. Families were planted in 5-m rows with 76 cm between rows. Each plot was over planted and thinned to a plant density of approximately 43 000 plants ha-1. Five to seven F2:3 plants per plot were self-pollinated. F4 seed from F2:3 plants within each family was harvested, and bulked at shelling. An aliquot of approximately 15 mL of seed was ground and analyzed for carotenoid concentration.
One hundred eighty-five hybrids and five checks were grown in 1999 and 2000 at the University of Illinois Research and Education Center at Urbana, IL. The testcrosses and checks were planted in a randomized complete block design with two replications per year. The hybrids were planted in two-row plots, with each plot 5.3 m long and 76 cm between rows. Each plot was over planted and then thinned to a final plant density of approximately 57 000 plants ha-1. Plants were allowed to open pollinate. Plots were harvested with a research plot combine. Moisture and plot weight were measured electronically. Data were used to calculate grain yield for every plot. An approximately 100-g sample of grain was kept from each plot for measurement of kernel composition.
Tocopherol Extraction
Aliquots of harvested seed from the F2:3 and testcross families were placed in 15-mL containers and ground in an M-2 Stein Mill (Stein Lab) for 90 s. Following grinding, the fine powder was collected and stored in darkness at 6.0°C until a 600-mg sample was weighed out. The 600-mg sample was stored in darkness at 6.0°C until extraction. The vitamin extraction procedure used (Kurilich and Juvik, 1999) is a modification of a procedure initially described by Weber (1987). All extractions were done under safe light, and plastic was never in contact with extract. Vitamin extracts were dried in a Savant SpeedVac Plus SC110A and Savant Universal Vacuum System under medium heat for approximately 1 h. Dried samples were resuspended with 200 mL of acetonitrile, methanol, and methylene chloride (45:20:35). For the testcrosses the same procedure was used, except to reduce differences caused by the varying level of supernatant that was removed from samples, a set volume (7 mL) of extract was used for quantification.
Tocopherol Quantification
Tocopherol concentrations for F2:3 families were determined with a high performance liquid chromatographer (HPLC). The procedure is a modification of a method described previously (Hart and Scott, 1995; Kurilich and Juvik, 1999; Weber, 1987). A Waters HPLC system (model 510 pump, Wisp 710B autoinjector, and 490E programmable multiwavelength detector; Waters Corporation, Milford, MA) was used. The solvent system was attached to ERC-3510 degasser (Erma Optical Works LTD, Tokyo). Data processing was handled with Digital Venturis FX computer running Waters Millennium 2010 Chromatography Software (Waters Corporation Version 2.15.01). Two columns were connected in series, a Vydac 4.6 x 150-mm reverse phase column, and a Waters Nova–Pak C-18 3.9- x 150-mm column. An Alltech Adsorbosphere C18 5-µm guard column was used to protect the columns. The mobile phase was acetonitrile, methanol, methylene chloride (75:20:5) with triethylamine (0.05% of total volume) and butylated hydroxytoluene (BHT).
The multiwavelength detector was set at 290 nm with AUFS = 0.8. Flow rate was 1.8 mL min-1, with each run taking approximately 8 to 10 min. Fifty microliters of sample was loaded into WISP 1-mL brown injection vials with 200-µL limited volume inserts with self-centering springs (Alltech and Associates, State College, PA). All HPLC analysis was run within 24-h of extraction to reduce degradation.
Quantification of compounds was accomplished by standard regression with external standards. The standards for -tocopherol, -tocopherol, and -tocopherol were purchased from Sigma (St. Louis, MO). Standards were brought to the correct concentration following a method from National Institute of Standards and Technology (National Institute of Standards and Technology, 1994). Four dilutions were made allowing for the calculation of the standard regression, and quantification of the individual tocopherols.
A new HPLC system was used for the quantification of tocopherols in testcross grain samples, allowing for the utilization of newer technologies as well as better control of system parameters (temperature of column and samples, better flow rate control). A Waters Alliance 2690 separation module (system includes solvent delivery system, inline degasser, column heater, and sample cooler) attached to Waters 996 Photodiode Array detector (PDA) was used. The PDA detector used for the testcrosses was more sensitive than the detector used for analysis of F2:3 families materials, and allowed for greater flexibility in wavelength selection. The HPLC system was managed by Waters Millennium 2010 software run on a Compaq Deskpro EP/SB Series with a Pentium III Processor. The column used for quantification was a YMC Carotenoid C-30 column (5 µm, 4.6 x 100 mm) that has been shown to separate tocopherol better than other columns (Darnoko et al., 2000; Sander et al., 2000). The mobile phase was acetonitrile, methanol, methylene chloride (75:20:5) with triethylamine (0.05% of total volume) and BHT, mixed by the Alliance 2690 solvent delivery system to help reduce differences between runs.
Samples were loaded into Amber glass containers with 50-µL limited volume inserts. All samples were run within 36 h of extraction. If samples were unable to be loaded into the HPLC immediately, they were stored at 4°C. Tocopherols were detected at 295 nm and were quantified by external standards as described previously. Flow rate was 1.8 mL min-1, with each run taking 6 to 8 min. To control fluctuations in retention time due to temperature, column temperature was set at 30°C. Sample temperature was set at 4°C to control degradation of samples by heat.
DNA Isolation
Equal volumes of F4 seed were taken from each bulk of each replication from both 1995 and 1996 F2:3 family field evaluations. From these bulks of the 200 F2:4 families, 30 random kernels were planted in the greenhouse for tissue sampling. Tissue was cut from the young plants, ground under liquid nitrogen, and stored at -80°C until DNA isolation could be done. DNA was isolated following the procedure described previously (Mikkilineni, 1997).
Genotypic Analysis
Microsatellites (SSR) were used for molecular marker mapping (Senior et al., 1996). The mapping population parents, W64a and A632, were screened with 531 markers from the MaizeDB Public SSR list (Maize Database, 2001). Of the markers screened, 163 were polymorphic for the cross and were assayed on the 200 F2:4 family mapping population, 123 of these markers were reliably scorable for the entire population. An aliquot of DNA from 200 F2:4 families was placed into 96 well plates and diluted 50 times with water. All reactions were run with a PTC-100 with 96 V-bottom well thermocycler (MJ Research, Waltham, MA). The microsatellite procedure is a modification of the procedure outlined previously (Senior et al., 1996). Following amplification, samples were stored at 4°C until the reaction could be evaluated on gels.
Reaction products were separated by gel electrophoresis in 4% (w/v) MetaPhor agarose (FMC Corp., Philadelphia) stained with ethidium bromide at 130 V for approximately 2 to 4 h. Use of Owl gel rigs with five 50-well combs allowed for all progeny assayed with each individual marker to be separated on the same gel. Each gel included one lane of Gibco BRL (Gaithersburg, MD) 50-bp ladder. Gels were viewed by means of a Stratagene Eagle Eye (Stratagene, La Jolla, CA) with a thermal printer attachment or a Kodak DC295 digital camera with an ultraviolet light filter attached to a Gateway 450mhz computer running Adobe Photoshop Professional Edition. If scoring could be performed via the digital photo, then gel photos were uploaded to Excel 2000 for scoring, if bands were not clear in the photo, then the gel was scored directly while on the UV light box. Bands were scored as 0.5 for homozygous A632,-0.5 for homozygous W64a, and 0.0 for heterozygotes.
Statistical Analysis
Phenotypic analyses were performed with the SAS statistical software package (SAS Institute, 1997). The three traits, -, -, and -tocopherol, were analyzed as individual compounds. Two other tocopherol traits were analyzed: total tocopherols and a ratio of tocopherols. The tocopherol compounds were summed to create total tocopherols. The ratio of -tocopherol to -tocopherol (tocopherol ratio) was calculated by dividing the -tocopherol concentration by the -tocopherol concentration on a plot basis. Means, range of means, and variances were calculated for individual and combined years on the complete unadjusted data set, keeping replications separate. Significance of effects for the (0,1) general lattice design (Federer and Wolfinger, 1998) for the F2:3 families and the RCB design of the testcrosses, were tested using the generalized linear model procedure of SAS. All effects for both F2:3 families and testcross progenies were considered random. For the F2:3 families, Proc GLM was used to analyze the effects of year, replications (year), block (year and replications), family, and family x year. For the testcrosses the effects of year, replications (year), family, and family x year were analyzed. Heritabilities were calculated by means of the covariance estimates from Proc Mixed. Best linear unbiased predictor (BLUP) estimates of family values for all traits were used in the marker analysis (Littell et al., 1996). Initial QTL detection was done by single factor analysis of variance.
JoinMap Version 3 (Van Ooijen and Voorrips, 2001) was used to construct a linkage map for the molecular markers used. Before mapping, JoinMap data analysis tools were used to evaluate quality of molecular marker data. Data were screened for missing data points, segregation distortion, and similarity between markers or individuals. Markers were removed for high level of segregation distortion and missing values. Initial linkage grouping of markers was done at a LOD threshold of 5.5. Groups were joined together based on previous mapping data for these markers (Maize Database, 2001). Maps were created from each individual group by means of a LOD of 0.001 and recombination frequency of 0.49. Haldane's mapping function was used so the map could be used in PLABQTL (Utz and Melchinger, 1996). The low stringency mapping procedure allowed as many markers as possible to map onto the population. The final map included 111 out of the 123 markers assayed on the entire population, with a total genome length of 1645.9 centimorgans (cM), and an average distance between markers of 16.3 cM (Fig. 2).
|
0.25 was calculated as having a critical LOD ranging from 2.76–2.83 and 2.81–2.85 for the per se and testcross progenies, respectively. QTL with a LOD value greater than 3.0 will be presented. Cofactors used for calculation of CIM were selected by the program using stepwise regression using the cov SELECT option. LOD curves were created by scanning every 2 cM of all possible linkage groups. QTL in adjacent intervals were considered to be the same QTL because of limitations of the PLABQTL program (F. Utz, personal communication). Digenic epistatic interactions were tested in the initial models with the model AA commands in PLABQTL. For the F2:3 families dominant effects were also included with the model D command. The phenotypic variation accounted for by an individual QTL (R2) is calculated as the square of the partial correlation coefficient from the final multiple regression model. This value is the coefficient of determination of the specified QTL and the phenotypic variation keeping all other QTL detected fixed (Utz and Melchinger, 1996). The proportion of the genetic variance accounted for by the marker model was calculated as R2 divided by heritability (Dudley, 1994). |
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-, -, and -tocopherol were 10.92, 1.64, and 34.94 µg g-1 (Table 1). Concentrations of and -tocopherol were greater in 1997 than in 1996. Conversely, -tocopherol had a greater mean in 1996 versus 1997. Testcross progenies had combined year means for -, -, and -tocopherol concentration of 44.66, 1.95, and 178.71 µg g-1, respectively. For individual year means, - and -tocopherol showed a greater concentration in 2000 than 1999 and -tocopherol had a greater concentration in 1999 than in 2000.
|
- and -tocopherol levels.
Tocopherol ratio had means of 0.44 and 0.26 for the F2:3 families and testcross progenies, respectively. The F2:3 families had a greater mean ratio of 0.53 in 1996 versus 0.34 in 1997. The greater ratio in 1996 than in 1997 was attributable to the larger proportion of families with -tocopherol concentrations greater than -tocopherol concentrations. Earlier studies have detected only a few lines with -tocopherol concentrations greater than -tocopherol (Galliher et al., 1985; Weber, 1984).
-Tocopherol was detected in the largest concentration and -tocopherol in the smallest concentration in both combined and individual years for the F2:3 families and testcross progenies. This agrees with earlier reports on tocopherol concentrations in corn (Weber, 1984). The testcross progenies exhibited greater concentrations of -, -, and total tocopherols than the F2:3 families. -Tocopherol did not differ between the F2:3 families and testcross progenies. The increased values for -, -, and total tocopherols in the testcross progenies may be attributable to the use of the high oil tester, which increased germ size and oil concentration. Since tocopherols are lipid soluble, the greater levels of oil in the testcrosses may have created a larger sink for tocopherols. The amount of - and -tocopherol are inversely related to the amount of -tocopherol. In both the F2:3 families and testcross progenies when - and -tocopherol levels were relatively high, the -tocopherol levels were relatively low. This observation suggests an interrelationship between the individual tocopherols, which is supported by the proposed biosynthetic pathway model whereby - and -tocopherol are converted to -tocopherol (Fig. 1).
Heritability estimates for the individual tocopherols for the F2:3 families were 93, 73, 93, 89, and 86% for -tocopherol, -tocopherol, -tocopherol, total tocopherols, and tocopherol ratio, respectively. For the testcross progenies the estimates were 85, 64, 76, 67, and 82 for -tocopherol, -tocopherol, -tocopherol, total tocopherols, and tocopherol ratio, respectively. Thus heritabilities were higher in the F2:3 families than in the testcross progenies.
QTL Analysis–F2:3 Families
The QTL on chromosome one for - and -tocopherol were clustered near one another (Fig. 2). Chromosome one had one interval accounting for 2.8 and 5.4% of the total phenotypic variation for - and -tocopherol, respectively (Table 2).
|
-, -, and -tocopherol respectively. The intervals on chromosome five accounted for the largest proportion of the phenotypic variation for all QTL detected for each of the individual tocopherols. The intervals are spread across the chromosome and likely represent independent QTL. Six other intervals were identified with significant QTL for -, -, or -tocopherol on chromosomes 6, 7, or 8 (Table 2). These intervals each accounted for a relatively small proportion (2.5–7.7%) of the total phenotypic variation.
QTL Analysis–Testcross Progenies
Four intervals on chromosome five (Fig. 2) were significant for one or more individual tocopherols (Table 2). Three adjacent intervals contained QTL for -tocopherol accounting for 2.3, 0.4, and 4.7% of the total phenotypic variation, respectively. One interval accounted for 8.8% of the total phenotypic variation for -tocopherol. Two intervals significant for -tocopherol accounted for 4.0 and 7.6% of the total phenotypic variation, respectively. Interval umc1155–phi085 was significant for all individual tocopherols, and accounted for the largest proportion of phenotypic variation for -tocopherol, but accounted for a low proportion of phenotypic variation for -tocopherol.
Three other chromosomes had QTL for individual tocopherols in the testcross progenies. Chromosome three and six each had significant intervals accounting for 3.5 and 4.2% of the phenotypic variation of QTL detected for -tocopherol. Chromosome seven had one interval accounting for the largest proportion of phenotypic variation of the QTL detected for -tocopherol with 11.1%.
Comparison of Per Se and Testcross Results
Five intervals on chromosome five were significant for tocopherols in both the F2:3 families and testcross progenies. When the two populations are considered together, the individual tocopherol QTL within these five intervals tend to cluster. -Tocopherol in the per se and testcross progenies have QTL located in two adjacent intervals on chromosome five, phi113–dupssr10 and dupssr10–umc1221. The peaks of the QTL are located at positions 20 and 36 for both populations (Fig. 2). Interval bmc1305–dupssr9 on chromosome seven, was significant for -tocopherol in both populations, with the QTL in both populations at position 112. The amount of phenotypic variation accounted for by this QTL was higher in the testcross than the F2:3 families accounting for 11.4 versus 4.5% of the total phenotypic variation.
-Tocopherol had two QTL in the F2:3 families and three QTL in the testcross progenies on chromosome five. Interval mmc0081–umc1155 and umc1155–phi085 contained QTL in both populations. The QTL in interval mmc0081–umc1155 for both populations were at position 52. Interval umc1155–phi085 had QTL at positions 80 and 88 in the F2:3 families and testcross progenies, respectively. The QTL in interval umc1155–phi085 for the testcross progenies accounted for a very low proportion of the phenotypic variation (0.5%). The QTL in the adjacent interval phi085–bmc1346 did account for a higher amount of variation (3.9%), at position 102 (Fig. 2).
QTL for -tocopherol in the F2:3 families and testcross progenies were located in adjacent intervals near marker umc1155. The QTL in the F2:3 families was located in interval mmc0081–umc1155 at position 56, and in the testcross progenies the QTL was in umc1155–phi085 at position 66 (Fig. 2).
The AE335 tester had tocopherol concentrations of 32.68, 2.11, and 103.19 µg g-1 for -, -, and -tocopherol, respectively (Data not shown). The AE335 values for and -tocopherol are much larger than the values for the two inbred parents and the mean concentration values for the F2:3 families. The -tocopherol concentration did not differ much between the inbred parents of the population, the mean of the F2:3 families and AE335. The testcross progenies mean is higher for both - and -tocopherol than for AE335, but lower for -tocopherol.
The amount of phenotypic and genotypic variation for tocopherols accounted for by the final models for the per se and testcross progenies is presented in Table 2. The phenotypic variation accounted for by the final model in the F2:3 families was 35.1, 12.3, and 29.3% for -, -, and -tocopherol, respectively. The model for testcross progenies accounted for 42.7, 7.8, and 26.5% of the phenotypic variation for -, -, and -tocopherol.
QTL Analysis–Total Tocopherols
One interval was significant for total tocopherols on chromosomes five for the F2:3 families (Table 2, Fig. 2), on chromosome five, accounted for 10.9% of the total phenotypic variation. Chromosomes five and seven were identified as having significant intervals for total tocopherols in the testcross progenies. Chromosome five accounted for the largest proportion of the phenotypic variation of the QTL detected with 6.4% of the phenotypic variation. Two intervals on chromosome seven accounted for 5.1 and 3.9% of the total phenotypic variation, respectively.
Chromosome five was significant in both F2:3 families and testcross progeny for total tocopherols. The two significant intervals are adjacent to one another, and the location of the QTL in the two populations are 22 cM apart. The QTL in the F2:3 families was located at position 16 and in the testcross progenies the QTL was located at position 38. In both the per se and testcross progenies the intervals on chromosome five accounted for the largest proportion of the phenotypic variation of all QTL detected. However these values are relatively small.
The final model for total tocopherols in the F2:3 families included one QTL and accounted for 10.0% of the total phenotypic variation. Three QTL accounted for 14.9% of the total phenotypic variation in the testcross progenies. The amount of variation accounted for by the final model in the two populations is not very different. The amount of genotypic variation accounted for in the per se and testcross progenies showed a greater difference, with 11.2 and 22.2% for the per se and testcross progenies, respectively (Table 2).
QTL Analysis–Tocopherol Ratio
The two inbred parents W64a and A632 had / tocopherol ratios of 0.09 and 0.53, respectively. This magnitude of difference in ratio of - to -tocopherol between parents may have enhanced the identification of QTL controlling this ratio. Chromosome five contained a QTL for tocopherol ratio accounting for 32.4% of the total phenotypic variation.
The AE335 tester had a tocopherol ratio of 0.32, which was only slightly greater than the 0.26 mean for the tocopherol ratio in the testcross progenies. Composite interval mapping for the testcross progenies identified two adjacent intervals on chromosome five which explained 30.5 and 12.8% of the total phenotypic variation. Interval umc1155–phi085 on chromosome five contained a QTL at position 92 which accounted for the largest proportion of phenotypic variation for the F2:3 families and testcross progenies.
The F2:3 families had one QTL accounting for 32.4% of the phenotypic variation and 37.7% of the genotypic variation for / tocopherol (Table 2). The testcross progenies had two QTL accounting for 52.2% of the phenotypic variation and 63.7% of the genotypic variation. The model for the testcross progeny accounted for a higher proportion of the phenotypic and genotypic variation then the model for the F2:3 families. Noteworthy, AE335 did not have a / ratio much different than the ratio of the mean for the F2:3 families. Therefore the AE335 genome may not be masking the affect of genes in F2:3 families in the testcross progenies.
Single Factor Analysis of Variance
Marker phi037 was significant in single factor analyses of variance for all individual tocopherols and total tocopherols in F2:3 families, and -, -, and total tocopherols in the testcross progenies (Data not shown). Marker phi037 maps to bin 1.08 on the Missouri Consensus Map, but was unlinked to any markers in the F2:3 mapping population. No other polymorphic markers were identified in the population on the long arm of chromosome one, near marker phi037. A linkage map could not be created that included marker phi037, therefore CIM could not be used to detect QTL in this region.
Interpretation and Future Study
One notable aspect of this investigation is the study of QTL influencing levels of compounds that have interrelationships because of the common tocopherol biosynthetic pathway, with - and -tocopherol precursors to -tocopherol. Because of the interrelationship of the tocopherol biosynthetic pathway, QTL with effects on one tocopherol compound may also affect other tocopherol compounds. The major QTL do affect more than one tocopherol compound in both the F2:3 families and testcross progenies. Most of the major QTL detected for the different tocopherol compounds are in the same or adjacent intervals. However, PLABQTL located the peaks of QTL for individual tocopherols at slightly different locations within intervals. The discrepancy between locations of peaks may be attributable to the differences in experimental error in the phenotypic and genotypic data as well as the limitations of the program.
PLABQTL may tend to locate peaks of QTL somewhere near the center of the interval, because of the method used to calculate position. The center of the interval allows more error to be explained by possible recombination (F. Utz, personal communication). The tendency to place the peak in the middle of an interval could be a source of error when comparing peak positions between two interrelated traits. It is possible that the larger the distance between flanking markers of an interval, the greater chance of having the position of the peak moved several centimorgans in either direction. One example of this is the 41-cM distance between markers umc1155 and phi085. Within this large interval are located a number of QTL for individual tocopherols, combined tocopherols and ratio of - to -tocopherols in the F2:3 families and testcross progenies. Because the distance is so great between the two flanking markers, the position of the QTL for different compounds may not be very precise. Placement of additional markers within this interval might reduce the map position discrepancies among QTL for different compounds within and across progenies. We found a similar relationship for the peak of QTL for four individual carotenoid compounds as well as the combined compounds (Wong et al., 2003).
We identified QTL on chromosomes one and five in both F2:3 families and testcross progenies. The detection of QTL in both populations provides greater confidence to the presence of genes controlling tocopherol accumulation in these chromosomal regions. More noteworthy perhaps are the QTL for tocopherol ratio. The largest QTL for tocopherol ratio in both F2:3 families and testcross progenies is located at position 92 on chromosome five. This QTL may be helpful in altering the levels of -tocopherol relative to -tocopherol in a marker assisted selection program. The maize EST with similarity to the Arabidopsis -tocopherol methyl transferase should be mapped onto the population and evaluated for possible linkage to the QTL for tocopherol ratio at position 92 on chromosome five, or QTL in other positions influencing individual tocopherols, total tocopherols, or tocopherol ratio.
Although we have identified QTL that are associated with levels of tocopherols, there are likely limits to the extent that these QTL could be used to modify levels of -tocopherol in maize grain. Another approach would be the overexpression of the -tocopherol methyl transferase gene through transgenic technologies. This may result in very high levels of -tocopherol, as has been demonstrated in Arabidopsis (Shintani and Dellapenna, 1998). Conventional and/or marker assisted selection could be performed to increase levels of -tocopherol in breeding lines, and then the transgenic approach using -TMT could be used to convert the high levels of -tocopherol to the more biologically active -tocopherol. This may produce levels of -tocopherol that are more useful from both an economic and nutritional standpoint. However, if it is determined that -tocopherol is nutritionally as important as -tocopherol, the transgenic approach with -TMT may not be needed. With interest in increasing levels of tocopherols, it would be helpful to have a better understanding of the relationship between tocopherols and levels of oil in the kernel since tocopherols are fat soluble and predominantly present in the oil of the kernel.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 5, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-tocopherol. J. Food Sci. 60:966–971.
-carotene desaturase associated with accumulation of carotenoids in maize. Theor. Appl. Genet. (in press).This article has been cited by other articles:
|
|
 |
|
  E. W. Jackson, M. Wise, J. M. Bonman, D. E. Obert, G. Hu, and D. M. Peterson QTLs Affecting {alpha}-Tocotrienol, {alpha}-Tocopherol, and Total Tocopherol Concentrations Detected in the Ogle/TAM O-301 Oat Mapping Population Crop Sci., November 24, 2008; 48(6): 2141 - 2152. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. J. Wassom, V. Mikkelineni, M. O. Bohn, and T. R. Rocheford QTL for Fatty Acid Composition of Maize Kernel Oil in Illinois High Oil x B73 Backcross-Derived Lines Crop Sci., January 16, 2008; 48(1): 69 - 78. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. U. Gilliland, M. Magallanes-Lundback, C. Hemming, A. Supplee, M. Koornneef, L. Bentsink, and D. DellaPenna Colloquium Paper: Genetic basis for natural variation in seed vitamin E levels in Arabidopsis thaliana PNAS, December 5, 2006; 103(49): 18834 - 18841. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Dudley, A. Dijkhuizen, C. Paul, S. T. Coates, and T. R. Rocheford Effects of Random Mating on Marker-QTL Associations in the Cross of the Illinois High Protein x Illinois Low Protein Maize Strains Crop Sci., July 1, 2004; 44(4): 1419 - 1428. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
Significant in the per se progenies. 
Significant in the testcross progenies. 
Significant by single factor analysis of variance; Estimated distance between markers when phi037 is forced onto the final map.