Are there clusters of growth-related genes in maize?
--Emil E. Khavkin and Ed Coe

In the maize genome, genes with phenotypical expression related to growth and development appear to form clusters about 10 to 30 cM long distributed nonrandomly along the chromosomes. A typical cluster includes mutants expressing retarded stem growth, changed attitude and disturbed growth of leaves, stems and roots, or their components, reduction and various malformations of inflorescences, and vivipary. This pattern is repeated , with considerable consistency, in different regions of the genetic map. Admittedly our identification of clusters and their boundaries is being done arbitrarily, but the number of clusters that can be identified may be as many as 16. The combined length of the clusters is about 30% of the total map length.

At first glance at the maize genetic map, we observe: (1) an apparently uneven distribution of growth-regulating genes (GRGs) in the maize genome, and (2) an obvious regularity of gene constellations in different chromosomes. Clusters of closely mapped GRGs comprise much the same categories of (a) genes governing hormone-sensitive changes in plant growth and development, complemented in most constellations with (b) genes apparently related to hormone metabolism and sensing, and (c) master genes that manifest profound influence on spatial and temporal pattern of cell and tissue differentiation. An example that illustrates the point is the region from an1 and id1 through kn1 and lw1 on chromosome 1, approx. 40 cM long, where these categories are represented. The 15-cM region including rd1, py2, vp8, tls1, and ts6 might be considered to be part of the same cluster, or separate, but the criteria by which we have done this first-approximation suggest that it would more likely be separate.

The majority of already mapped QTLs for plant growth, architecture and productivity, and master genes apparently related to transcription factors, participating in spatial and temporal control over plant development and /or hormone-response functions, map within these clusters. The an1...kn1 region of chromosome 1, for example, displays QTLs for a number of relevant traits studied by Doebley and Stec (1991) and by Stuber et al. (1992).

We suggest that these clusters are functional units comprising genes for environmental sensors and signal transducers, receptor sites to translate environmental and hormonal signals to growth machinery, and master genes to govern critical spatial and temporal transitions in cell growth and differentiation. When clustered in such a functional unit, genes expressed in concert gain more efficient short-distance cis-control or proximity control by transcription factors engaged in protein-protein and DNA-protein interactions. The interactions with different factors may provide a great diversity of growth and developmental reactions to a limited number of environmental stimuli. Some clusters quite distant on the map might also interact in trans if clusters come into spatial proximity in the interphase nucleus, and if some clusters can be identified as "incomplete" they are candidates for such trans-complementation.

Many heritable traits concerning plant form, growth and development are well-documented and mapped (Coe et al., in: Corn and Corn Improvement, 1988; Sheridan, Annu. Rev. Genet. 22: 353-385, 1988), and recently rapid progress has been made in the cloning and sequencing of several GRGs (Freeling et al., BioEssays 14:227-236, 1992). Expression of GRGs provides for hormonal regulation, i.e., production, degradation and interaction of endogenous hormones as well as response to endogenous or exogenous hormonal signals, including transduction of signals and such loosely defined processes as commitment, competence, determination, evocation or sensitivity (Trewavas and Cleland, Trends Bioch. Sci. 8:354-357, 1983). While environmental changes induce profound effects on hormone content and distribution, some environmental effects are not mediated by hormones. Yet in both cases there must be genes for sensors to translate environmental signals into differential gene expression. On the opposite end of this GRG chain displayed as a sequence of growth events we presume to find master genes channeling differential gene expression into specific patterns of cell division, enlargement and specialization.

A working hypothesis: We suggest that the maize genome contains functionally significant units of clustered genes for plant growth and development. Such a unit must comprise:
    - sensors for environmental signals, e.g., daylength, light quality, gravity, temperature, etc., capable of transforming these signals into primary (hormones) or secondary (Ca - calmodulin, transmembrane potential ) messages to genes;
    - receptor sites or independent transmitters to translate the message into the growth machinery within a particular cell (cell-autonomous trait) or in a wider context (non-cell-autonomous trait);
    -service pathways to control these chains of events by producing low- and high-molecular-weight products that are signal transducers or modulators;
    -master genes presumably operating in cascade fashion to govern cell and tissue differentiation at the critical points of development; apparently some of these master genes could also play the role of receptor sites for environmental messages.

Notably, the most prominent GRG clusters seem to contain all the listed components. Mutations at the genes comprising clusters produce several classes of disturbances in plant growth and development associated with specific hormones: ABA-related vivipary of embryos, dwarfism usually related to deficiency in gibberellin metabolism, transport or sensing, auxin-related alteration of apical dominance leading to changes in the branching pattern, and various malformations, including developmental displacements, as ectopic effects of hormone interplay. Some of the clusters include genes that could be environmental sensors (Phy1, Phy2), or hormone sensors (Abp1, D8/D9, Rab, Vp). Finally, in most clusters we find genes regulating cell fates in development; these genes usually contain sequences related to DNA transcription factors (Kn being the best example).

The advantages of functional gene clustering are intuitively attractive. Compartmentalization within a nucleus of signal molecules, transcription factors and co-factors of transcription can facilitate temporal regulation of gene expression and amplify regulatory cascades. Multienzyme complexes are the most extensively studied example of such compartmentalization of functional coordination and control signals. 

Return to the MNL 68 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage