ABSTRACT: The control of gene expression in Toxoplasma is not well understood. As we and others have shown, this organism does contain bipartite promoters that are sufficient to induce gene expression. This information can be used in downstream experiments to identify proteins that bind to experimentally confirmed cis elements. One of these methods, yeast one-hybrid is currently being employed to identify proteins which bind to the BAG1 and B-NTPase promoter elements. Another powerful method that can be used to identify genomic regions containing controllers of gene expression takes advantage of genetic linkage mapping provided by genetic crosses, called expression quantitative trait locus (eQTL) mapping. In the eQTL approach, one uses gene expression as quantitative traits, or phenotypes, in QTL mapping. QTL mapping is a proven method in Toxoplasma that has recently identified apicomplexan-specific virulence factors in both the Type II X III and Type I X III crosses (Saeij et al., 2006, Taylor et al., 2006, Su et al., 2002). With the recent design of the Affymetrix Toxoplasma GeneChip one can gather gene expression profiles for all predicted genes in the Toxoplasma genome in a rapid and reproducible way. Combining the power of Affymetrix gene expression technology and QTL mapping can lead to the identification of genes with differential expression in the parents and progeny of a genetic cross and then to the identification of the genomic regions that contain mechanisms that have an effect on expression. The first description of combining global gene expression profiles with QTL mapping was published by Brem et al. using the model organism Saccharomyces cerevisiae (Brem et al., 2002). Since that time, eQTL projects have been described for several other organisms including human, mouse, rat, maize, Arabidopsis, barley, eucalyptus, and wheat (Gibson & Weir, 2005). There are two mapping location types expected when using gene expression profiles as phenotypes. Mapping of gene expression profiles can lead to linkage of a locus (region) containing the gene being mapped, thus cis to the gene. Although the boundaries of the locus will contain many genes, it is presumed that for most cis QTLs sequence polymorphisms in the regulatory region(s) of the gene being mapped are the cause of the observed expression differences. The other mapping location type expected is one where the expression profile maps to a locus not containing the gene being mapped, or trans to the gene (Ronald et al., 2005). The most interesting trans loci are those that have multiple genes mapping to the same locus in trans, or to a trans hotspot. These hotspots likely contain factor(s) that regulate many genes, such as transcription factors or factors involved in upstream signal transduction pathways that can eventually lead to a number of transcriptional changes (Yvert et al., 2003). Here we describe the results of an eQTL project using the gene expression profiles for the parents and 34 informative progeny of the type I-GT1-Fudr X type III-CTGara cross for both the tachyzoite and bradyzoite stages. This is the first time global gene expression profiles have been used to determine linkage to regions presumed to be controllers of transcription using the Affymetrix Toxoplasma GeneChip. The dominant genomic associations found were those mapping cis and there were several regions, or hotspots, where multiple genes map in trans. In addition, we observed an unexpected finding suggesting there are several duplication events in the parasites of the Type I X III cross which may also have consequences on gene expression levels. The ToxoGeneChip microarray (http://ancillary.toxodb.org/docs/Array-Tutorial.html) was used to measure both tachyzoite and bradyzoite mRNA expression in the parents and progeny of the I X III cross. For each of the two developmental expression sets, genome-wide eQTL scans were run for each gene specific probeset on the microarray. The parents of the cross are CTGara and GT1-Fudr. The rest comprise the informative progeny.
