Proceedings of the XLVII Italian Society of Agricultural Genetics - SIGA Annual Congress

Verona, Italy - 24/27 September, 2003

ISBN 88-900622-4-X

Oral Communication Abstract - S1e

DETECTION OF CIS-ACTING REGULATORY VARIATION THAT AFFECTS MAIZE GENE EXPRESSION

F. CATTONARO, S. AMADEO, M. MORGANTE

Dipartimento di Produzione Vegetale e Tecnologie Agrarie, Università degli Studi di Udine, Via delle Scienze 208, 33100 Udine, Italy

maize, gene expression, SNPs, RT-PCR, heterosis

Variation in coding sequence is classified as synonymous and non-synonymous, reflecting nucleotide variants with and without an effect on phenotype at the level of protein primary structure and can be recognized directly from DNA sequence. Variation in non-coding cis-regulatory DNA sequences, which affects gene expression levels, has also been proposed as a major component of the genetic basis for phenotypic evolution. In humans it has been proposed to influence disease susceptibility. In contrast to the former, the latter type of variation can not be recognised from nucleotide sequences alone.

A method to detect changes in transcript levels, due to cis-acting sequence differences, without the necessity to recognize specific regulatory variants (which can be hundreds or even thousands of bases upstream from the transcription unit), has been developed in maize, based upon recent papers about methods for measurement of allelic levels of gene expression in humans and mouse (Yan et al., Science, 2002; Cowles et al., Nature Genetics, 2002).

The method involves the study of two alleles of a gene under identical circumstances and the comparison of the transcript expression associated with each. This is done by comparing the expression of alleles from two maize inbred lines (B73 and H99) in the F1 hybrid lines (B73XH99 and H99XB73). To distinguish between the transcripts derived from each of the two parental alleles we used SNP markers (Single Nucleotide polymorphisms) in the transcript itself.

The SNP assay we developed, involves RT-PCR amplification of the region surrounding the SNP marker, Single Base Extension (SBE) of a primer adjacent to the variant base in the presence of fluorescently labeled nucleotides and detection on a capillary DNA sequencer. The ratio of the two alleles is inferred by comparison with known mixtures of B73 and H99 genomic DNA used as reference standard, which, after PCR amplification, are subjected to the same SNP assay.

Preliminary results on genes expressed at medium and high levels in seedlings revealed abundant cis-acting variation in maize: 3 of 5 (60%) genes tested showed greater than 1.5 fold (range: 2.1-4.0) differences in expression among alleles, with no difference among reciprocal hybrids. This is in contrast with what was found in mouse genes where 4 out of 69 genes (6%) showed allelic differences in gene expression (Cowles et al., Nature Genetics, 2002).

The high levels of cis-acting variation in maize genes we are finding agree with recent discoveries about very high levels of sequence diversity in maize transcribed and promoter regions (Ching et al., BMC Genetics, 2002), lack of sharing of repeats/intergenic regions and genes in maize inbred lines (Fu and Dooner, PNAS, 2002) and influence of (retro)transposons insertion/deletions on expression of flanking genes in maize and wheat (Selinger and Chandler, PNAS, 1999; Selinger and Chandler Plant Physiology, 2001; Kashkush et al., Nature Genetics, 2003). Beside representing an important source of phenotypic and quantitative variation, regulatory variation may also provide a possible molecular explanation of the overdominance genetic hypotesis, which at least partly explains the hybrid vigor (heterosis) phenomenon in maize.