A large-scale effort, termed the Secreted Protein Discovery Initiative (SPDI), was undertaken to identify novel secreted and transmembrane proteins. In the first of several approaches, a biological signal sequence trap in yeast cells was utilized to identify cDNA clones encoding putative secreted proteins. A second strategy utilized various algorithms that recognize features such as the hydrophobic properties of signal sequences to identify putative proteins encoded by expressed sequence tags (ESTs) from human cDNA libraries. A third approach surveyed ESTs for protein sequence similarity to a set of known receptors and their ligands with the BLAST algorithm. Finally, both signal-sequence prediction algorithms and BLAST were used to identify single exons of potential genes from within human genomic sequence. The isolation of full-length cDNA clones for each of these candidate genes resulted in the identification of >1000 novel proteins. A total of 256 of these cDNAs are still novel, including variants and novel genes, per the most recent GenBank release version. The success of this large-scale effort was assessed by a bioinformatics analysis of the proteins through predictions of protein domains, subcellular localizations, and possible functional roles. The SPDI collection should facilitate efforts to better understand intercellular communication, may lead to new understandings of human diseases, and provides potential opportunities for the development of therapeutics.

A series of mutations in the homeo domain of the yeast a2 protein were constructed to test, both in vivo and in vitro, predictions based on the oL2-DNA cocrystal structure described by Wolberger et al. (1991). The effects of the mutations were observed in three different contexts using authentic target DNA sequences: a2 binding alone to specific DNA, ,v2 binding cooperatively with MCM1 to specific DNA, and a2 binding cooperatively with al to specific DNA. As expected, changes in the amino acid residues that contact DNA in the X-ray structure severely compromised the ability of a2 to bind DNA alone and to bind DNA cooperatively with MCM1. In contrast, many of these same mutations, including a triple change that altered all the "recognition" residues of helix 3, had little or no effect on the cooperative binding of a2 and al to specific DNA, as determined both in vivo and in vitro. These results show that the ability of a homeo domain protein to correctly select and repress target genes does not necessarily depend on the residues commonly implicated in sequence-specific DNA binding.

In diploid cells of the yeast Saccharomyces cerevisiae, the alpha2 and a1 homeodomain proteins bind cooperatively to sites in the promoters of haploid cell-type-specific genes (hsg) to repress their expression. Although both proteins bind to the DNA, in the alpha2 homeodomain substitutions of residues that are involved in contacting the DNA have little or no effect on repression in vivo or cooperative DNA binding with a1 protein in vitro. This result brings up the question of the contribution of each protein in the heterodimer complex to the DNA-binding affinity and specificity. To determine the requirements for the a1-alpha2 homeodomain DNA recognition, we systematically introduced single base-pair substitutions in an a1-alpha2 DNA-binding site and examined their effects on repression in vivo and DNA binding in vitro. Our results show that nearly all substitutions that significantly decrease repression and DNA-binding affinity are at positions which are specifically contacted by either the alpha2 or a1 protein. Interestingly, an alpha2 mutant lacking side chains that make base-specific contacts in the major groove is able to discriminate between the wild-type and mutant DNA sites with the same sequence specificity as the wild-type protein. These results suggest that the specificity of alpha2 DNA binding in complex with a1 does not rely solely on the residues that make base-specific contacts. We have also examined the contribution of the a1 homeodomain to the binding affinity and specificity of the complex. In contrast to the lack of a defective phenotype produced by mutations in the alpha2 homeodomain, many of the alanine substitutions of residues in the a1 homeodomain have large effects on a1-alpha2-mediated repression and DNA binding. This result shows that the two proteins do not make equal contributions to the DNA-binding affinity of the complex.

The dose response curve is the gold standard for measuring the effect of a drug treatment, but is rarely used in genomic scale transcriptional profiling due to perceived obstacles of cost and analysis. One barrier to examining transcriptional dose responses is that existing methods for microarray data analysis can identify patterns, but provide no quantitative pharmacological information. We developed analytical methods that identify transcripts responsive to dose, calculate classical pharmacological parameters such as the EC50, and enable an in-depth analysis of coordinated dose-dependent treatment effects. The approach was applied to a transcriptional profiling study that evaluated four kinase inhibitors (imatinib, nilotinib, dasatinib and PD0325901) across a six-logarithm dose range, using 12 arrays per compound. The transcript responses proved a powerful means to characterize and compare the compounds: the distribution of EC50 values for the transcriptome was linked to specific targets, dose-dependent effects on cellular processes were identified using automated pathway analysis, and a connection was seen between EC50s in standard cellular assays and transcriptional EC50s. Our approach greatly enriches the information that can be obtained from standard transcriptional profiling technology. Moreover, these methods are automated, robust to non-optimized assays, and could be applied to other sources of quantitative data.

The yeast MAT alpha 2 and MATa1 homeodomain proteins bind cooperatively as a heterodimer to sites upstream of haploid-specific genes, repressing their transcription. In the crystal structure of alpha 2 and a1 bound to DNA, each homeodomain makes independent base-specific contacts with the DNA and the two proteins contact each other through an extended tail region of alpha 2 that tethers the two homeodomains to one another. Because this extended region may be flexible, the ability of the heterodimer to discriminate among DNA sites with altered spacing between alpha 2 and a1 binding sites was examined. Spacing between the half sites was critical for specific DNA binding and transcriptional repression by the complex. However, amino acid insertions in the tail region of alpha 2 suppressed the effect of altering an a1/alpha 2 site by increasing the spacing between the half sites. Insertions in the tail also decreased DNA bending by a1/alpha 2. Thus tethering the two homeodomains contributes to DNA bending by a1/alpha 2, but the precise nature of the resulting bend is not essential for repression.