Traits such as fruit size, flowering time, plant morphology, and light responsiveness are under complex genetic control and show continuous (quantitative) variation among genotypes of a particular species. In recent years, a number of these quantitative traits have been successfully analyzed genetically by exploiting natural genetic variation. Three critical tools for doing so are numerous polymorphic molecular markers, large segregating populations, and programs to analyze recombination frequencies, which can then be used to "map" quantitative trait loci (QTLs). A number of genes corresponding to QTLs have been successfully isolated (Doebley et al., 1997; Frary et al., 2000; El-Assal et al., 2001; Maloof et al., 2001; Yano et al., 2001). Contrary to some expectations, the QTLs responsible for natural variation have turned out to be not simply ‘minor, modifier’ genes, but variant, functional alleles of fundamental genes (Millar et al., 2001).

We believe that the analysis of natural variation is a feasible approach to the identification of the genes underlying the complex process of hemicellulose biosynthesis. Genetic analysis of natural variation is particularly well-suited to the study of processes that are so central to the life of the organism that only subtle variations are tolerated. Since hemicelluloses are present in all plant cells at all stages of development, null mutations or ‘strong’ alleles in genes affecting hemicellulose biosynthetic enzymes are likely to be lethal. On the other hand, working with natural variation can be difficult because the genetic differences, and therefore the phenotypic differences, are typically ‘weak’. Therefore, the influences of other genes and of the environment can make phenotypic analysis difficult, and successful mapping, a prerequisite to gene identification, requires special tools such as large populations and many mapping markers.

The chemical properties of cereal hemicelluloses affect many economically important processes, such as silage digestibility, bread making quality, insect resistance, and brewing properties (e.g., Brice and Morrison, 1982; Courtin and Delcour, 1998; Lundvall et al., 1994; Vinkx and Delcour, 1996). Many of these traits have been studied genetically in wheat, maize, and other cereals (e.g., Lempereur et al., 1997; Saulnier et al., 1995). Not surprisingly, these traits have complex patterns of inheritance indicative of control by many genes. QTLs affecting mixed-linkage glucan (MLG) content in oat grain and digestibility and fiber content in maize have been detected (Falkner et al., 2000; Kianian et al., 2000). In a unique study, a single major QTL that accounts for 35-42% of the variation in the ratio of arabinose to xylose in wheat flour arabinoxylan was identified (Martinant et al., 1998). This study used two populations of 106 and 115 recombinant inbred lines (RILs) with 266 and 1125 molecular markers, respectively.

In order to succeed in the identification of one or more genes that affect hemicellulose content or composition based on natural variation, the following resources are needed: (1) demonstration of statistically significant genetic variation between at least two cultivars or genotypes, (2) a segregating population with a large number of mapped markers, (3) the capacity to take the results from the mapping species into a species whose entire genome sequence is available. These criteria are fulfilled for maize and rice together.

In order to test for genetic differences among maize accessions, they were grown under uniform conditions and repeatedly sampled. Differences measured among inbreds were assumed to be caused by genetic effects barring effects caused by variable seed quality. Analyses of variance were performed using a general linear model with inbred as the single fixed effect (SAS, 2000). All pairs of inbred lines were tested for equality at a probability of 0.05 using single degree of freedom contrast.

Six types of tissue from ten maize inbred lines were sampled at three growth stages. For plants grown in the greenhouse, six types of tissue were sampled from a subset of four inbred lines at approximately the time when the sixth above-ground internode began to expand. Each tissue was replicated three times. Total cell wall monosaccharides were extracted and analyzed as the alditol acetates by gas chromatography. These experiments indicated that there are reproducible differences in monosaccharide composition among tissues and among inbred lines. Levels of rhamnose, fucose, and mannose (collectively less than 7%) were too low to be accurately quantitated. Xylose, arabinose, and galactose accounted for an average of 49, 29, and 18% of the total monosaccharide content, respectively. Significant differences (p < 0.05) were detected for arabinose, xylose, and galactose in eleven of the eighteen seedling tissue types and six of eighteen juvenile tissue types.

The Missouri Maize Mapping Project (http://www.cafnr.missouri.edu/mmp/genetic1.htm) has developed a high density linkage map using a segregating population (known as IBM) that was created from a cross of B73 and Mo17. The IBM population has a map size ~3-fold larger than normal F₂ populations or RILs, and hence greater utility for QTL mapping, due to four cycles of random intermating before selfing (Beavis et al., 1992; Davis et al., 2000). Over 17,000 molecular markers of several kinds have been placed on the IBM map, and it will be integrated with a physical BAC map (G. Davis, University of Missouri, personal communication).

Several criteria were considered for defining a model plant and tissue for the analysis of natural genetic variation of cell wall composition. A major consideration is the existence of rich genetic and genomic resources, which is satisfied for maize in light of recent advances in the development of recombinant inbred lines (RILs) such as the IBM population derived from B73 and Mo17, a large number of genetic markers, and synteny with the completely sequenced genome of rice (Davis et al., 1999, 2001; Goff et al., 2002; Lee et al., 2002; Song et al., 2002). The tissue to be sampled should be easy to collect, have a low level of starch, and its composition be insensitive to small differences in rates of growth and development among accessions or ecotypes. The maize pericarp satisfies these criteria. It is predicted to have low starch content and, if sampled from a fully mature cob, variation in growth and development should be minimal. Sufficient amounts of tissue can be sampled from a single ear. The pericarp can be easily sampled from soaked seeds. The material is free of aleurone and endosperm (Fig. 1).

Figure 1. Fluorescent microscopy of cross-sections of intact seeds (caryopses) (a and b) and excised pericarp (c and d). (a) Mo17, (b) B73, (c) RIL 356, and (d) RIL 338. The three layers of cells visible in the whole caryopsis cross section, from inside to outside, are endosperm, aleurone, and pericarp. Only pericarp is present in (c) and (d). Bar equals 100 µm. (From Hazen et al., Plant Physiology, in press).

Results to Date:

Analysis of the sugar monomer composition of the pericarp of the IBM population revealed 13 QTLs. A paper on this work is in press (Hazen SP, RM Hawley, GL Davis, B Henrissat, and JD Walton (2003) Plant Physiology). An example of four QTLs found on chromosome three is shown in Fig. 2. 

Figure 2. QTLs for sugar monomer composition on chromosome 3 of maize, based on analysis of the IBM population (Hazen et al., in press).

Figure 3. Example of using synteny between rice and maize to identify candidate rice genes corresponding to a maize QTL (Hazen et al., in press).