BERKELEY, CALIFORNIA
University of California

The order of liguleless gene function
--Lisa Harper and Michael Freeling

Phenotype of lg1 and lg2 homozygotes. Recessive mutants in two unlinked genes remove the ligule and auricle of the maize leaf: lg1 and lg2. In a Mo17 background, they have different phenotypes which are distinguishable in seedlings as well as in adult plants. In lg1 mutant plants, no auricle is found on any leaf, and ligule only on the last few adult leaves before the tassel. In lg2 plants, only the first and sometimes second leaves are completely liguleless and auricleless. On juvenile leaves, the ligule and auricle appear as small wedges at the margin, making the adaxial side appear to have a broad midrib. Each subsequently formed leaf has more ligule and auricle than the previous. The two auricle wedges on either side of the broad midrib area can also be displaced relative to each other. Ligule is never formed at the midrib.

Mosaic analysis results. Earlier results of mosaic analysis on lg1 (Becraft et al., Dev. Bio. 141:220-232, 1990) indicate that Lg1+ product acts in a cell autonomous way. Borders of lg1 sectors in a wildtype leaf correspond exactly to the cell autonomous marker used. In addition, the progression of ligule maturation from midrib to margin is aborted by the presence of a lg1 sector. Ligule and auricle reinitiate on the marginal side of the mutant sector, as if they were starting new from the midrib. This suggests that Lg1+ is involved in the progression or propagation of a "make ligule- make auricle" signal.

Preliminary results of mosaic analysis of lg2 suggest a different role. Inv3a/lg2 y10 plants were X-rayed to induce segmental monosomic sectors on a wildtype leaf. The presence of a lg2 sector on an otherwise wildtype leaf sometimes causes the disruption or lack of ligule. However, the presence or absence of ligule in a mutant sector shows no correlation with the presence or absence of Lg2+ in any particular layer or layers of the leaf at the blade-sheath boundary in 39 sectors studied. No correlation was found between sector width, or distance from the midrib, with the presence or absence of ligule. Since no obvious correlative factor was found, other factors, such as a signaling pathway acting in the longitudinal dimension, cannot be ruled out. The data suggest that Lg2+ function does not confine itself to the cells in which Lg2+ product is made.

In addition, the progression of ligule and auricle on the marginal side of the mutant sector is not disrupted by the presence of a lg2 sector. Thus, the presence of a lg2 mutant sector does not impede the "make ligule-make auricle" signal.

Aneuploid analysis. TB-2Sb was crossed as a male to lg1-0/lg1-0 (the Coop reference allele), and to a dwarf tester. Liguleless and dwarf progeny were observed, respectively, in non-Mendelian ratios. The lg1 plants appeared no different than their mothers with respect to ligule and auricle. Likewise, TB-3Lg, lg2/+ was crossed as a male to lg2-0/lg2-0 (the Coop reference allele). The cross was set up this way in order to compare lg2/lg2 sibs to lg2/-. The progeny were >50% lg2. Hypoploids of TB-3Lg are runts, so lg2/lg2 could be distinguished from lg2/-. There was no apparent difference in the ligule and auricles of the 2 classes of lg2 plants. Thus, both lg1-0 and lg2-0 mutants are acting genetically as nulls.

Double mutant analysis. Double mutants of lg1-0 and lg2-0 were constructed by the following crosses: ((lg1 gl2/lg1 gl2 X lg2/lg2) X lg1 gl2/lg1 gl2) self, and testcross to lg2/lg2, selecting for liguleless and glossy phenotype after the second cross. Of the families that segregate 1:1 for lg2 in the testcross, all progeny of the self showed the lg1 phenotype; i.e., they appeared totally liguleless and auricleless when scored at the three leaf stage. For comparison, sibs of the lg2/lg2 test plants show auricle and ligule on the margins of the second leaf, at the approximate place where the ligule is expected. This is in agreement with segregation of the (lg1 gl2/lg1 gl2 X lg2/lg2) self, where 104 adult plants were scored, and segregated 67 wildtype to 24 lg1-like phenotype to 13 lg2-like phenotype. Statistical analysis suggests this is more likely a 9:4:3 segregation than a 9:3:3:1 (chi squared equals 3.57 versus 10.94 respectively). No novel phenotypes were observed, and there was no significant loss of germination in either planting of double mutants. In a strict sense, Lg1 is epistatic to Lg2, however, the nature of the two phenotypes does not allow us to conclude that Lg1+ and Lg2+ are acting in the same biochemical pathway. Since both behave as nulls, we can conclude that they are both involved in the same biological developmental program.

Conclusion. Mutants of lg1 have a more severe phenotype than mutants of lg2. One might expect the double mutant to display the more severe phenotype, and this is indeed the case. In terms of developmentally ordering gene function, however, Lg1+ cannot act earlier in development than Lg2+. Several observations point to this conclusion. Lg1+ acts in a cell autonomous manner. This precludes having a non-cell-autonomously acting gene downstream. Mosaic analysis has shown that Lg1+ is responsible for reception and/or propagation of a "make ligule-make auricle" signal, but is not the signal itself. Scanning electron microscopy and sectioning have shown that the lg1 mutation blocks the epidermal periclinal and longitudinal anticlinal divisions in the pre-ligule, pre-auricle area (Sylvester et al., Development 110:985-1000, 1990; Becraft et al., Dev. Bio. 141:220-232, 1990). This is relatively late in development.

Lg2+ gene function acts in a non-cell autonomous way, suggesting that the Lg2+ gene product is itself mobile, or causes another product to be mobile. This mobile product is not likely to be the "make ligule-make auricle" signal itself, because lg2 mutants, even in embryonic leaves, make ligule and auricle. The lesion in lg2 mutants affects the placement of ligule and auricle on the leaf, and the coordination of the auricle wedges on either side of the midrib. This suggests that Lg2+ is involved in making the region competent to receive the "make ligule-make auricle" signal. 


Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page