Theoretical and Applied Genetics

, Volume 120, Issue 4, pp 753–763

Cloning and characterization of a putative GS3 ortholog involved in maize kernel development

  • Qing Li
  • Xiaohong Yang
  • Guanghong Bai
  • Marilyn L. Warburton
  • George Mahuku
  • Michael Gore
  • Jingrui Dai
  • Jiansheng Li
  • Jianbing Yan
Original Paper

DOI: 10.1007/s00122-009-1196-x

Cite this article as:
Li, Q., Yang, X., Bai, G. et al. Theor Appl Genet (2010) 120: 753. doi:10.1007/s00122-009-1196-x

Abstract

The GS3 gene was the first identified gene controlling the grain size in rice. It has been proven to be involved in the evolution of grain size during domestication. We isolated the maize ortholog, ZmGS3 and investigated its role in the evolution of maize grain size. ZmGS3 has five exons encoding a protein with 198 amino acids, and has domains in common with the rice GS3 protein. Compared with teosinte, maize has reduced nucleotide diversity at ZmGS3, and the reduction is comparable to that found in neutrally evolving maize genes. No positive selection was detected along the length of the gene using either the Hudson–Kreitman–Aguadé or Tajima’s D tests. Phylogenetic analysis reveals a distribution of maize sequences among two different clades, with one clade including related teosinte sequences. The nucleotide polymorphism analysis, selection test and phylogenetic analysis reveal that ZmGS3 has not been subjected to selection, and appears to be a neutrally evolving gene. In maize, ZmGS3 is primarily expressed in immature ears and kernels, implying a role in maize kernel development. Association mapping analysis revealed one polymorphism in the fifth exon that is significantly associated with kernel length in two environments. Also one polymorphism in the promoter region was found to affect hundred kernel weight in both environments. Collectively, these results imply that ZmGS3 is involved in maize kernel development but with different functional polymorphisms and thus, possibly different mechanisms from that of the rice GS3 gene.

Supplementary material

122_2009_1196_MOESM1_ESM.doc (81 kb)
Supplementary material 1 (DOC 81 kb)
122_2009_1196_MOESM2_ESM.tif (895 kb)
Supplementary Fig. S1 Genetic and physical location of yield-related QTL. Genetic coordinates are based on IBM2 2008 Neighbors. The physical locations are from www.maizegdb.org. The two markers that have location discrepancy in genetic and physical map are underlined. ZmGS3 cloned in this study is in italics and in open box. For detailed QTL information, see Supplementary Table S1. (TIFF 894 kb)
122_2009_1196_MOESM3_ESM.tif (290 kb)
Supplementary Fig. S2 a PCR amplification product of the gap1 primer pair using B73 genomic DNA. b Tertiary product of TAIL-PCR using primers listed in Supplementary Table S2. c PCR amplification product of the CDS2 primer pair (Supplementary Table S2) using cDNA isolated from immature ears (TIFF 290 kb)
122_2009_1196_MOESM4_ESM.tif (632 kb)
Supplementary Fig. S3 Schematics of the domain structure of ZmGS3 and OsGS3 protein. Because of two (one) amino acids change, the tumor necrosis factor receptor (TNFR)/nerve growth factor receptor (NGFR) domain in rice (the von Willebrand factor type C (VWFC) domain in maize) can not be identified by InterProScan (http://www.ebi.ac.uk/Tools/InterProScan/), they were identified manually. The phosphatidylethanolamine-binding protein (PEBP)-like domain can not be identified by InterProScan either, thus the rice domain was presented according to the results of Fan et al. (2006) and the corresponding maize domain was identified by the Blast program, MUSCLE (Edgar 2004). Only three amino acid substitutions were observed between OsGS3 and ZmGS3 in the PEBP-like domain. The amino acid change in the rice protein that leads to altered grain characteristics is marked by an asterisk (TIFF 631 kb)
122_2009_1196_MOESM5_ESM.tif (390 kb)
Supplementary Fig. S4 Plots comparing identity of GS3 protein sequences from higher plants, including maize (this study), sorghum (assembly of BI098398 and BI098087), rice (CT835094), tomato (assembly of BI210240 and AW217519), potato (BQ116994), populus (DT488475), Arabidopsis (AK221695), and Brassica (DY004631 and AC189411). The similarity was averaged over every 5 aligned amino acids. Relative position of amino acids is given based on the aligned maize protein sequence, which is indicated by five open boxes at the bottom (TIFF 390 kb)
122_2009_1196_MOESM6_ESM.tif (2.4 mb)
Supplementary Fig. S5 Multiple sequence alignment (a) and phylogenetic analysis (b) of GS3 and DEP1 protein sequences in maize and rice. a Identical residues are indicated by dark black boxes, and similar residues by light gray boxes. Asterisk means the amino acids are completely conserved in the five protein sequences. The sequences were aligned using MUSCLE (Edgar 2004). b The phylogeny was generated using MEGA, version 3.1 (Kumar et al. 2004). Numbers at the branches are percentages based on 1,000 bootstrap repetitions, bootstrap values >50% are given. The scale bar indicates the number of amino acid substitutions per position. GenBank accession numbers are as follows: OsGS3, DQ355996. ZmGS3, FJ797616. OsDEP1, FJ039905. ZmDEP1-7, AC187408. ZmDEP1-2, AC190873 (TIFF 2490 kb)

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  • Qing Li
    • 1
  • Xiaohong Yang
    • 1
  • Guanghong Bai
    • 1
    • 2
  • Marilyn L. Warburton
    • 3
  • George Mahuku
    • 4
  • Michael Gore
    • 5
  • Jingrui Dai
    • 1
  • Jiansheng Li
    • 1
  • Jianbing Yan
    • 1
    • 4
  1. 1.National Maize Improvement Center of China, Key Laboratory of Crop Genomics and Genetic Improvement (Ministry of Agriculture)China Agricultural UniversityBeijingChina
  2. 2.College of AgricultureXinjiang Agricultural UniversityUrumqiChina
  3. 3.USDA-ARS, Corn Host Plant Resistance Research UnitMississippi StateUSA
  4. 4.International Maize and Wheat Improvement Center (CIMMYT)Mexico D.F.Mexico
  5. 5.Department of Plant Breeding and GeneticsCornell UniversityIthacaUSA