An anaerobic gene, which encodes an apparently non-glycolytic protein, shares sequence homology with Mu1.7 and Mu related sequence-A

--Julie Vogel and Michael Freeling

We would like to provide further information about the characterization of an anaerobically expressed gene in maize roots, upon which our laboratory previously reported in this newsletter (MNL 63:38). Unfortunate DNA sequencing errors led to incorrect conclusions in the previous MNL report about the size, amino acid composition, and charge of the polypeptide encoded by this anaerobically expressed gene. With new, corrected DNA sequence data, we now wish to clarify the primary structural features of this gene and its predicted encoded protein.

This full-length cDNA, whose encoding gene we now denote anp27, hybridizes to a message of approximately 1.1kb that is induced 2-3 fold in anaerobically treated seedling roots and shoots. No transcripts can be detected in any tissue of aerobically grown maize plants that we examined, and no message is present in either cold- or heat-shocked maize seedlings. Thus, expression of the anp27 gene appears to be under specific transcriptional control; message is produced only under anaerobic conditions. This cDNA contains an open reading frame of 765 nucleotides, encoding a polypeptide of 255 amino acids with a predicted molecular mass of 27.5kDa. A cluster of proteins of this approximate size has been observed on two-dimensional gel fluorographs of proteins that have been in vivo-labelled under anaerobic conditions (unpublished data from this laboratory). We are currently attempting to identify which one of these low molecular weight polypeptides is encoded by the anp27 gene. Although we haven't yet demonstrated, it is likely that ANP27 expression is also controlled at the level of translation, as has been seen for other maize anaerobic proteins.

The ANP27 protein appears to be both proline- (10%) and arginine- (10%) rich; yet, it does not show obvious sequence homology to known proteins rich in these amino acids. Although predicted to carry a net charge of -2, the distribution of charges within the molecule indicates two distinct charge domain regions: an acidic N-terminus (~60aa) and a highly basic middle region (~100aa). Protein sequence comparisons using current EMBL/Genbank databases reveal no obvious similarities to any other known proteins. Thus, this anaerobic protein still does not appear to represent a known glycolytic enzyme, and its identity (and function) remains unknown. Its small size and unusual primary structure are consistent with the intriguing possibility that this protein has a regulatory function during the anaerobic response.

Surprisingly, we found that the ANP27 cDNA shows high DNA sequence similarity to two other maize genomic sequences previously characterized: Mu1.7 (Taylor and Walbot, V, Genetics 117:297, 1987) and Mu related sequence-A [denoted MRS-A by Talbert and Chandler, but which we denote herein as MrsA] (Talbert and Chandler, Mol. Biol. Evol. 5:519, 1988). Surprisingly, the ANP27 cDNA shares 60.8% DNA sequence identity with MrsA, within 716 nucleotides of overlap, and 60.7% identity with Mu1.7, within 606 nucleotides of overlap. This region of similarity encompasses nearly the entire protein-encoding portion of the ANP27 cDNA, but not the untranslated regions. The part of Mu1.7 containing the homologous sequences encompasses the region from the beginning of the second internal direct repeat, through the 385bp “insertion” relative to the Mu1 element, and ending 100bp beyond the Mu1.7-unique region (but before the inverted repeat end of the element). The region of sequence similarity within MrsA begins at the same point within the second internal direct repeat, extends contiguously through the Mu1.7-specific segment, and ends 200bp further downstream. Although the ANP27 cDNA is extremely [G+C]-rich (overall, it is 66% [G+C], and greater than 80% [G+C] in its middle region), the sequence similarity to the Mu1.7 and MrsA elements (each ~70% [G+C] in the Mu1.7-unique segment) is mediated by true sequence homology, rather than just [G+C] content. From these comparisons, it is reasonable to propose that anp27 is evolutionarily related to the Mu1.7 element and to MrsA, with anp27 having the most diverged sequence. What this reveals about genomic DNA rearrangements during maize evolution is not yet entirely clear.

The Mu1.7-unique segment of DNA has been shown to hybridize specifically to a 2.2-2.3kb transcript in several different (aerobic) tissues of both Mu-active and Mu-inactive lines (Taylor and Walbot, V, 1987; Talbert and Chandler, 1988), and this transcript is likely to be encoded by the MrsA locus (Talbert and Chandler). Although a cDNA for this transcript has not yet been isolated, and therefore the encoded polypeptide is not known, it is intriguing that MrsA might encode a product related to ANP27. The region of these Mu elements with sequence similarity to anp27 neatly encompasses two oppositely oriented reading frames (ORF3 and ORF4 as per Taylor and Walbot). In fact, ORF3, when translated, is predicted to encode a polypeptide that would be ~60% (for MrsA) and 50% (for Mu1.7) conserved with the ANP27 polypeptide.

We are currently further characterizing the anp27 gene in order to better understand its function during the anaerobic response as well as to clarify its relationship to the Mu1.7 and MrsA sequences in the maize genome.

Return to the MNL 66 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage