GAINESVILLE, FLORIDA
University of Florida

Corn Rab2 homologs
--Laughner, BH; Ferl, RJ

We cloned three Rab2- homologous cDNAs, comprising two classes that share 83% nucleotide identity and 97% amino acid identity (GenBank Accession #s U22432 and U22433). RFLP analysis performed by Helentjaris's group at Arizona mapped only a single specific locus, near the centromere region of chromosome 10. These clones may then represent alleles of a single ZmRab2 gene. Indeed, other Rab subfamily members that are separate genes in other species typically do not share such identities. For example, human Rab5a and human Rab5b share 81% amino acid identity but are not able to cross hybridize since their nucleotide identity is 66% (Wilson and Wilson, J. Clin. Invest. 89:996-1005, 1992). Consequently, these corn clones are referred to as ZmRab2-A and ZmRab2-B. It is possible, however, that these clones represent distinct, but closely linked genes, as is apparently the case in two Arabidopsis Rab2 genes.

Alignments of Rab2 plant GTP binding proteins with animal homologs shows 82% identity with both ZmRab2s, with dissimilarities clustering primarily at the carboxyl ends. Since the first proteins identified in this subfamily were from yeast, some investigators prefer the designation "Ypt" rather than the "Rab" designation derived from the animal counterparts. Indeed, a Ypt designee from Zea mays (Ypt-Z3) has been reported to have approximately 80% amino acid identity with Rab2; however, the sequence information for this clone apparently is not deposited in any database (Fabry et al., Gene 118:153-162, 1992). Using available database matches and truncating the highly variable carboxyl ends, a PAUP analysis confirmed specific homology of our maize clones to the Rab2 group at a confidence level above 90% through bootstrap analysis with 100 replicates. Hence we retain the Rab2 designation.

The family of small GTP binding proteins, more correctly referred to as the "ras superfamily" (Kahn et al., FASEB J. 6:2512-2513, 1992), is characterized by a common mechanism for binding GTP which insures activation with subsequent inactivation rendered by GTP hydrolysis. This family is often divided into three major groups represented by Ras, Rho and Ypt/Rab proteins. Extensive biochemical analysis on the h-ras gene product p21ras has served as a model for other proteins in this family (Pai, EF et al., Nature 341:209-214, 1989). When this model is used to more closely scrutinize the GTP binding motifs as delimited essentially by Valencia et al. (Biochemistry 30:4637-4648,1991), our ZmRab2 homologs exhibit extreme conservation in the GTP binding motifs. None of the substitutions that do exist affect the putative contact residues within the GTP binding motifs. The model predicts our ZmRab2s to have similar functions to the animal forms, and specifies the proteins to be in an active form when GTP is bound.

Further work is necessary, however, to demonstrate GTP binding and any purported role in vesicular trafficking. Even with the similarities to the animal Rab2s, important functional differences may exist. For example, animal Rab2s have been reported to be localized to a specific region of the cell between the endoplasmic reticulum and the Golgi apparatus, but differences in the carboxy-terminal modification signal could possibly change this intracellular targeting for these ZmRab2 homologs. 


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

Return to the MNL 71 On-Line Index
Return to the Maize Genome Database Page