Maize Genetics Cooperation Newsletter 80. 2006.

Saint Paul, Minnesota

University of Minnesota

How many maize genes are not in B73?

--Okagaki, RJ; Schmidt, C; Stec, AO; Rines, HW; Phillips, RL

We are finding evidence that the genomic sequences for numerous maize ESTs are not present in the B73 genome. These gene-like sequences can be found in other maize lines, but they appear to be missing in B73. In retrospect, it is not surprising to find deletion polymorphisms for maize genes. Polymorphisms for single base changes and simple sequence repeats are common, and deletions are merely another type of molecular lesion. Our results, based on a study of ESTs from A188, suggest there could be many gene-like sequences in the A188 inbred line that are partially or entirely deleted in B73.

Computational analysis identified gene-like sequences that were likely to be absent in B73; this was followed by a molecular analysis to determine if the sequences were indeed absent in B73 and other maize lines. Approximately 16,000 EST sequences that had been annotated as having come from A188 were downloaded from GenBank. EST sequences longer than 300 nucleotides were then searched against the maize B73 genomic sequences in the TIGR Maize Database, AZM Release 5.0. Approximately 500 EST sequences longer than 300 basepairs did not match a B73 genomic sequence with a probability value of e^-10 or less, and thus may identify a maize gene that is not in the B73 genome. These sequences were matched with the EST contigs developed by TIGR, and the 521 sequences corresponded to 71 singletons and 156 contigs. BLAST searches were then repeated using the entire contig sequences to eliminate additional sequences. PCR primers were developed against the remaining 63 EST sequences, and PCR assays were used to detect the presence of these sequences in genomic DNA from A188, B73 and 12 other maize lines. We obtained results with 53 sets of primers. No products from B73 genomic DNA were detected with 17 primer sets, and 28 primer sets failed to amplify products from at least one maize line. Twenty-five primer sets amplified products from all 14 lines tested. Table 1 presents a sample of the data. We are in the process of confirming these results by Southern blot analysis. To date, Southern blot analysis has confirmed the absence of five sequences in B73 out of the 10 sequences tested. Extrapolation of these results to the approximately 50,000 maize genes suggests there could be many gene-like sequences in A188 that are missing in B73.

Table 1. Detection of A188 ESTs in maize lines.

EST sequence

A188

A632

B37

B73

C103

CML5

CML52

CML91

Mo17

Oh43

Tzi18

Co159

Tx303

W64A

CK700895

CB179401

CB179394

DR906760

CN845215

CF349054

BM660009

CF920121

DN559697

DN559490

DN475032

DN586584

CD052452

CD052345

CF974781

CF273223

CN845203

CV072139

These data are inadequate to estimate accurately the number of A188 gene-like sequences absent in B73. The BLAST analysis used to identify sequences that are present in B73 may underestimate the number of A188 sequences missing in B73. We used a cutoff of e^-10 for identifying hits between A188 EST sequences and B73 GSS sequences. The BLAST threshold permitted hits between an A188 EST sequence and related, but not identical, sequences in B73. Many deletion polymorphisms occurring in duplicated genes could be missed. A study by Lai and co-workers (Genome Res. 14:1924, 2004) estimated that 50% of the genes duplicated in the maize tetraploid ancestor remain duplicated. Our search for A188 sequences missing in B73 may have ignored half of the maize genes. If the frequency of deletion polymorphisms was the same in single-copy and duplicated genes, then estimates of the number of A188 gene-like sequences missing in B73 should be doubled. The assumption that the frequency of deletion polymorphisms is similar in single copy versus duplicated sequences may not be valid; one study of human genes found that gene-deletion polymorphisms were 10-fold more common in segmentally duplicated regions of the human genome (Tuzun et al., Nature Genetics 73:727, 2005). Lastly, the polymorphisms observed here were unlikely to have been produced by the replication and transposition of sequences by helitron elements. Helitron elements do create polymorphisms similar to those observed here (Lai et al., PNAS 102:9068, 2005), however we are looking primarily at single copy sequences rather than the duplicated sequences associated with helitron transposition. In conclusion, it is likely that a full sequence of the B73 inbred line will necessarily miss a large number of the genes present in the species.

_________________________________________________

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

Return to MNL 80 on-line index.

Return to MNL index.

Return to MaizeGDB home page

____________________________________________