Endoreduplication of the nuclear DNA is a prominent feature of maize endosperm development. Following the peak mitotic period, at 8 to 10 days after pollination (DAP), an increase in DNA content per nucleus is observed. This increase is due to poly-tenization of the chromosomes caused by endoreduplication of the DNA. During the endoreduplication process, two different cell cycles are occurring in the endosperm simultaneously, the mitotic cell cycle and the endoreduplicating cell cycle. The mitotic cell cycle consists of G1, S, G2, and M-phases. The endoreduplicating cell cycle consists only of alternating S and G-phases; thus, DNA synthesis occurs but there is no chromosome condensation, strand separation, karyokinesis nor cytokineses. The endosperm may be divided into three regions on the basis of these cycles. In the central region of the endosperm at 10 to 12 DAP, the cells begin to undergo the endoreduplicating cell cycle. These cells are not synchronized in their position in the cell cycle. Once a nucleus has undergone the endoreduplication process, it does not become mitotic again. Thus cells which are 48C can be found near to other cells which are 12C or 96C. The second region occurs around the peri-phery of the endosperm. Here the conventional mitotic cell cycle is the rule. Although the mitotic index peaks between 8 and 10 DAP, cells in this peripheral region continue to be mitotic once endoreduplication has begun in the cells of the central region of the endosperm. The third region is a transitional region located between the first two regions. It contains both mitotic and endoreduplicating cells.

Using flow cytometry, it is possible to monitor both the mitotic and endoreduplicating cell cycles at the same time. Mitotic cells are represented by nuclei with 3C to 6C DNA content where C is the haploid DNA content per nucleus. Endoreduplicating cells comprise the nuclei with DNA content greater than 6C. Among the endoreduplicating nuclei, cells in the G-phase appear as peaks at the 12C, 24C, 48C, 96C, ..., DNA contents, while those in the S-phase appear between these peaks on output from the flow cytometer. By analysis of nuclear preparations from maize endosperm at different DAP, it is possible to follow the progression of the mitotic and endoreduplicating cell cycles. Different inbred lines display characteristic patterns of endoreduplication. These differences may be the percentage of cells in the S- vs. G-phase, as well as in the number of rounds of endoreduplication that they undergo. These differences in pattern of endoreduplication are controlled by a maternal effect, that is, a protein or transcript encoded by the nuclear genome of the female parent which controls a developmental function in the endosperm.

The purpose of this research was to identify a number of mutants in various inbred backgrounds which represent different types of alterations in the endoreduplicating cell cycle in maize endosperm.

Normal seed of seventy-two lines segregating for defective kernel phenotype were planted in the field in 1991. Among the 72 lines, seventeen lines were produced by ethylmethylsulfonate (EMS) mutagenesis of inbred W23, twenty-four lines were derived by EMS mutagenesis of inbred A188, fourteen lines were generated by EMS mutagenesis of a hybrid, and seventeen lines were induced by transposition of Mu. Seed of the EMS mutagenized hybrid was provided by Dr. G. Neuffer, University of Missouri, Columbia. Seed of the Mu-induced lines was provided by Dr. Martha James, Iowa State University. Other lines were available from previous EMS experiments by Dr. R. L. Phillips. In all of the mutant lines used, either one or two loci were responsible for the defective kernel phenotype.

Several plants within a line were self-pollinated. Kernel samples for W23 lines and Mu lines were collected at 18 DAP. Those for the A188 lines and the hybrid lines were collected at 16 DAP. Both defective and normal kernels from a segregating ear were removed using a scalpel and placed in 3:1 (95% ethanol: glacial acetic acid). The following day, the samples were placed in 70% ethanol and stored at -20 C. Pairs of nuclear preparations of normal and defective kernels from a segregating ear were made according to the method of Kowles and Phillips (Dev. Genet. 11:125-132, 1990). Nuclear preparations of normal kernels were made from a single endosperm. Those of defective kernels were made using one to three endosperms due to the reduced size of the material. Endosperm nuclei from all pairs of normal and defective kernels from one genetic background were prepared and stained the same day. Nuclear preparations of W23, Mu and hybrid lines were analyzed using an Ortho Diagnostics flow cytometer. A Becton Dickinson flow cytometer was utilized to evaluate the A188 lines. Mean DNA content per nucleus was determined based on the fluorescence values obtained from analysis of the Mithramycin stained nuclei. Cell counts were made using a hemocytometer. Three to five counts were made from a nuclear preparation. DNA content and cell number data within a pair of normal and defective kernels from the same segregating ear were compared using a standard t-test. Comparisons between lines are invalid due to differences in pollination date.

Ten classes of mutants were identified among the materials studied. They are lines with: 1) lower DNA content per nucleus due to no or almost no endoreduplication and lower cell number in the dek vs. the normal endosperm, 2) lower DNA content per nucleus due to fewer rounds of endoreduplication and lower cell number in the dek vs. the normal endosperm, 3) lower DNA content having the same number of rounds of endoreduplication but fewer nuclei in the endoreduplicating peaks and lower cell number in the defective endosperm when compared to normal, 4) lower average DNA content per nucleus and equivalent cell number in the dek vs. the normal endosperm, 5) decreased DNA content per nucleus and increased cell number in the dek compared to the normal endosperm, 6) equivalent DNA content per nucleus and reduced cell number in the dek compared to the normal, 7) higher average DNA content per nucleus and higher cell number in defective endosperm compared to normal, 8) equivalent DNA content per nucleus and cell number in the dek compared to the normal, 9) increased DNA content per nucleus and reduced cell number in dek compared to normal lines, and 10) higher average DNA content per nucleus and equivalent cell number in the dek and the normal endosperm. Table 1 lists the number of mutants for each class identified within each background.

Table 1. Frequency of ten types of mutants in DNA content and cell number in four genetic backgrounds.
 

Class	W23 EMS lines	A188 EMS lines	Hybrid EMS lines	Mutator lines
1	1	4	2	0
2	7	12	3	7
3	0	3	0	2
4	8	1	5	5
5	0	0	0	2
6	1	0	0	0
7	0	1	0	0
8	0	1	0	0
9	0	0	0	1
10	0	0	2	0

These mutants should provide the necessary materials for use in further elucidating the biochemistry, molecular, and cell biology of the endoreduplication process. 

Return to the MNL 67 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage