Tucson, Arizona
University of Arizona
Functional Genomics of Chromatin: a progress report for maize --Plant Chromatin Consortium*
For information contact Vicki L. Chandler, University of Arizona, Tucson, AZ (chandler@ag.arizona.edu)
Overview. The goal of our NSF-funded project (DBI-9975930) is to identify and functionally analyze the entire complement of genes in maize and Arabidopsis that contribute to chromatin-based control of gene expression. This project will result in the generation and classification of a large set of mutations to facilitate investigations of chromatin-level gene regulation in plants, facilitating a deeper understanding of the complex mechanisms by which plants control the expression of their genes. In this report we summarize the genes that have been identified to date in both Arabidopsis and maize and summarize our progress with the maize experiments.

There are three principle objectives within this project. First, mutations are being generated in genes that can be identified in (a) the Arabidopsis genome sequence and (b) EST collections from maize and other plants by sequence similarity to known chromatin genes, especially those thought to play a role in the control of gene expression. Because certain tests of chromatin gene function require or are more efficiently carried out with dominant mutations, we are producing dominant negative mutations for each target chromatin gene, using RNA silencing triggered by transgenes producing double-stranded RNA molecules homologous to target genes (Smith et al., Nature 407:319, 2000; Wesley et al., Plant J. 27:581, 2001). In the case that dominant negatives are lethal or deleterious, dexamethasone-inducible dominant negative mutations are being generated in Arabidopsis (McNellis et al., Plant J. 14:247, 1998). T-DNA insertional mutations are being generated in Arabidopsis for genes encoding histones, histone acetyltransferases, and histone deacetylases, and other chromatin genes to the extent possible (Krysan et al., Proc Natl Acad Sci U S A 93:8145, 1996; McKinney et al., Plant J. 8:613, 1995). In the case of lethal or deleterious dominant negatives in maize, TILLING will be used to generate mutations (McCallum et al., Plant Physiol. 123:439, 2000). Second, all mutations will be characterized to determine their effects on genetic transmission, plant growth and development, and a comprehensive battery of biochemical and epigenetic tests. These tests include DNA methylation, the processes of epimutation and paramutation, and reactivation of silenced transgenes and transposons. Third, a Plant Chromatin Database is being developed to facilitate dissemination of information on chromatin level control in plants and other organisms.

Chromatin Genes in Arabidopsis and Maize. Extensive sequence similarity searches were performed in Arabidopsis using a variety of queries from well-characterized yeast, Drosophila, mammalian and plant chromatin genes and protein domains. To date 225 genes have been identified and 178 of these genes are targets for RNA silencing (the 45 core histones and two TBP homologs, GTF1 and GTF2 are targets for T-DNA insertions). After accounting for genes sharing greater than 80% nucleotide sequence identity, a minimum of 137 genes or gene pairs need to be targeted for RNA silencing in Arabidopsis. Using Arabidopsis genes as queries, we identified 1,356 maize ESTs that cluster into 184 tentative contigs. One hundred seventeen represent candidates for RNA silencing (core histone genes and several other contigs are not being targeted). After accounting for closely related genes (>80% sequence identity), 109 genes or gene pairs are potential targets for RNA silencing. The numbers of genes identified and their categories are summarized in Table 1.

Table 1. Arabidopsis and Maize Chromatin Genes Identified to Date
Categories of Chromatin Genes 
Histone acetyl-transferases    
(GNAT, MYST, CBP, TafII250 homologs) 
Histone acetyl-transferase complex components
(ADA2 homologs)    
Histone deacetylases (RPD3, HD2, SIR2 homologs)
SNF2 homologs (Chromatin remodelers, not including
DNA repair-recombination proteins)    
Components of remodeling complexes
DNA methyltransferases
Methyl binding domain proteins 
Global Transcription Factors
Nucleosome/chromatin Assembly Factors
(NAP, CAF, HMG proteins)    
Silencing Genes (ASF1, ARD1 homologs) 
MAR binding filament-like protein
Core histones 
Linker histones 
SET domain proteins 
Proteins similar to bromodomain proteins belonging to
Ring3 /BDF1/ FSH group of proteins    
Other Bromodomain proteins 
Other Chromodomain proteins
(not including retroelement polyproteins)    

Sequencing, Mapping and Expression Studies With Maize Genes. As there is no genomic sequence currently available for maize, sequences for complete open reading frames will be determined to facilitate phylogenetic comparisons with other organisms. We are isolating full-length cDNA from inbred line B73 for each of the genes being targeted using a combination of sequencing ESTs and RACE/PCR. We continue to acquire and sequence ESTs primarily from the Stanford Maize Gene Discovery project. New ESTs are first verified for insert presence, and then the EST with the largest insert is sequenced to confirm its identity. If the EST is correct, it is sequenced completely. The sequence that is obtained is then immediately used to design primers for PCR amplification. High-quality sequence is being obtained for the full-length cDNA clones isolated by RACE and RT-PCR at Wisconsin. Three independent clones for each full-length cDNA are sequenced on both strands. Sequences are searched for a complete open reading frame that is then compared with the protein sequence in ChromDB. To date full length B73 clones have been obtained for 33 maize chromatin genes and 17 more are in the pipeline. Full-length sequences for 27 clones have been submitted to GenBank.

We are finding many highly similar maize chromatin genes. Mapping the genes is one way to determine whether different contigs and ESTs come from the same or similar genes. Having map positions for chromatin genes may also facilitate matching candidate genes to epigenetic mutants. Furthermore, mapping gives us information about whether similar, recently duplicated genes are parts of segmental duplications. These data are of interest in determining the evolutionary history of plant chromatin genes, and when combined with phenotypic data, will allow us to determine the range of functions performed by different members of chromatin gene families. As of this report, EST clones for nearly all targeted genes have been used as hybridization probes on DNA blots to determine gene copy number and screen for polymorphisms amenable for gene mapping. Images of these screening blots are posted on the project website. All polymorphic bands are being mapped on an abbreviated set (94 lines) of the intermated B73-Mo17 (IBM) recombinant inbred mapping population used in the Maize Mapping Project. When completed, results will be posted on MaizeDB, with links from ChromDB.

Expression of chromatin ESTs is evaluated by hybridization of cDNA clones to a panel of maize RNAs. Chromatin clones are used to probe Northern blots of immature ear, tassel, kernel, seedling leaf (9 days), mature leaf (8-leaf stage), and callus RNA. These experiments have been completed for more than a third of the targeted genes. Results can be viewed at the ChromDB website.

RNA Silencing Mutants in Maize. Ninety-one genes are being directly targeted for dsRNA constructs at this time. Because of the high sequence identity between closely related genes in maize these 91 constructs are expected to silence the 109 maize genes targeted to date. PCR primers have been designed to amplify ~700 bp fragments from the EST in the mid-region of each gene’s coding sequence. These products are then each inserted twice into the vector pMCG161 in opposite orientations and flanking an intron. Transcription of this construct will give rise to a dsRNA that should direct the degradation of the target mRNA. The website has details on the vector. Sixty-five dsRNA constructs have been produced and 55 bombarded into immature maize embryos. Multiple transformation events (12-20/construct) have been obtained for 26 constructs. Regenerated plants for six different constructs are currently growing in the greenhouse. Fertile plants will be crossed to B73 to bulk seed, the transgene structure determined via DNA blots and silencing assayed by RNA blots and RT-PCR.

Planned Biochemical and Genetic Assays. The chromatin mutant collection will be characterized functionally with respect to genetic transmission, plant growth and development, DNA methylation, stability of epimutations and paramutation, and stability of silent states of transgenes and transposons.

ChromDB. ChromDB is both a repository for data produced by this project and a research tool to allow users to perform queries of the data contained within it. ChromDB is a relational database built with MySQL and a Perl-CGI interface that displays the contents of the database and allows users to browse through genes in various ways. The public interface to the ChromDB is www.chromdb.org. There are multiple options to access the chromatin genes in ChromDB. Searches can be done using a gene name as query on the welcome page or users can go via a link to a Perl-CGI script that generates a web page with three combinations of choices of how to access data. A local BLAST server is available to search sequences in ChromDB. The query results are displayed with hyperlinks to the genes.

ChromDB holds the following information on the sets of chromatin genes we have identified in Arabidopsis and maize: accession numbers for the sequences in GenBank, the functional class of chromatin gene, sequences, exon models, and corresponding ESTs, as well as experimental data produced by the project that includes the expression profiles and the genomic DNA gel blot images for maize genes to indicate copy number. Arabidopsis exon models incorporate information from EST sequences and project generated RT-PCR sequences so as to provide a better annotation of the gene. Sequence similarity relationships between the chromatin genes in the database have been generated by BLAST comparisons and can be either produced through user selection or viewed along with the detailed information on a gene. The database search for any gene leads to a detailed page containing all the above information for that gene and links from that page go to pages that display the predicted gene, mRNA and protein sequences. Links to GenBank are provided for all sequences. The web site also includes detailed information on the tools and strategies that are being used to silence target genes in Arabidopsis and maize. dsRNA silencing vectors are available for academic research purposes via simple procedures described at the web site.

All data generated in this project will be deposited in the Plant Chromatin Database and mutants will be deposited in public stock centers as soon as they are confirmed to be mutant.

*Plant Chromatin Consortium Participants

University of Arizona: Vicki L. Chandler, Richard Jorgensen, David Mount, Carolyn Napoli, Ross Atkinson, Ed Butler, Raghavendra Guru, Erin Guthrie, Arthur Kerschen, Traci Klein, Andreas Müller, Virginia O’Connell, Ritu Pandey, Robert Sandoval, Jr., Lyudmila Sidorenko, Dave Selinger, Ita Vargas-Lagunes and Qin Zeng.

Johns Hopkins University: Judith Bender, Brandie Rocci and Sharon Wilensky.

University of Missouri, Columbia: Karen Cone, Dean Bergstrom, Ebony Courtney and Miriam Hankins

Purdue University: Stan Gelvin, Hongbin Cao and Xiaowen Zhao

Washington University: Craig Pikaard, Eric Richards, Kari Hesselbach, Richard Lawrence, Todd Smith and Tuya Wulan.

University of Wisconsin: Heidi Kaeppler, Shawn Kaeppler, Anisha Akula, Chakradhar Akula, William Haun, Laura Schmitt, Alan Smith and Nathan Springer.

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

Return to the MNL 76 On-Line Index
Return to the Maize Newsletter Index
Return to the Maize Genome Database Page