Seleção de progênies visando à obtenção de indutores de haploides em milho

Abstract

Submitting haploid inducing progenies to selection cycles allows gains in haploid induction rate and aids in the identification and selection of those that present better performance. Haploid identification methods are used to estimate induction rates. The R-navajo morphological marker is the most used but can be inhibited to the presence of inhibiting alleles of the anthocyanin regulating genes, transposable elements, environmental factors, or genotype used as genome donor for the haploid, due to variable expressiveness and incomplete penetrance. With this work, the objective was to recombine haploid-inducing progenies, existing in the Plant Genetics and Breeding Program of UFLA, and select the most adapted to tropical conditions and with the highest percentages of induction. The experiments were conducted at Lavras, MG, Brazil. S2:4 seeds of the 2015/2016 harvest were recombined. The obtained individuals were self-fertilized, and evaluated regarding haploid-induction by crossing with hybrid 30A37PW. Induction was verified for the seeds derived from the cross between the inducers and the donors. There was complete inhibition of R1-navajo, demanding the search for other identification methods and explanation for this inhibition. The method that obtained results was the sowing of 200 seeds per treatment in greenhouse, and the selection of possible haploids by the size of the seedling at 50 days of age. Those presenting size inferior to half the average of the remaining individual of the treatment were selected as possible haploids and submitted to confirmation by flow cytometry. The calculation of the induction rate was done considering the haploids confirmed by flow cytometry. The statistical analyses were done using mixed generalized linear models (MGLM). For better understanding the inhibition of the R1-navajo, molecular analysis with specific markers designed for gene C1 and allele C1-I inhibitor were performed. The DNA of hybrids 30A37PW and DKB390, used as control, were extracted, submitted to PCR, and observed in 1% agarose gel. Two non-expected expressive bands were identified, extracted, and submitted to DNA sequencing for better understanding. The haploid selection strategy by size did not allow a precise assessment of the induction rate, however, it was the most adequate method given the circumstances. There was increase in the induction rate of the previous progenies when compared to the present cycle. It was impossible to, feasibly and practically, distinguish haploid individuals in the cases of total inhibition of R1-navajo. When desiring to use the R1-navajo, it is indicated that tests be conducted to guarantee that the donor genotype does not present coloration inhibitors. Alternate studies and techniques are necessary to identify possible haploids and to calculate induction rates. It was verified that the C1-I inhibiting allele is not present in the 30A37PW hybrid and it has not caused the inhibition of the R1-navajo. However, the sequencing of one of the bands obtained 99% of alignment with a miniature inverted replicate transposable element of the Heartbreaker family, which probably caused the silencing of the CA gene and the coloration inhibition conferred by the R1-navajo.