Human Forkhead Genes

Doctoral thesis, 1996

The forkhead family constitutes a group of genes which encodes transcription factors that share a common DNA-binding domain, the forkhead domain. The first member to be discovered was the Drosophila melanogaster gene fork head. Members of this family have been found in a variety of species from bakers' yeast to man. We have isolated and characterised the DNA binding motif of eight, previously unknown, human members of the forkhead family. They were isolated by low stringency hybridisation of human genomic and cDNA libraries. The newly identified genes were named FREAC genes, short for Forkhead Related Activators, to indicate their homology to the Drosophila melanogaster gene fork head and their ability to activate transcription. In order to determine their DNA binding specificity, a PCR-based selection method was developed. Fusion proteins between glutathione-S-transferase (GST) and the DNA binding domain of four different FREAC proteins were expressed in bacteria. For each protein high affinity binding sites were selected from a pool of double-stranded oligonucleotides with a randomised sequence. All four proteins had a similar core motif in their recognition site consensus described as RTAAAYA. By swapping subdomains between the DNA binding domains of FREAC-3 and FREAC-4, the different binding site specificities could be mapped to a couple of basic residues in the C-terminal end and to a short stretch of amino acids in the N-terminal of the recognition helix. The expression patterns of the FREAC genes were either restricted to a few tissues or ubiquitous. FREAC-1 and -2 were solely expressed in placenta and lung, while FREAC-6 was restricted to kidney. FREAC-4 mRNA was found in kidney, testis, monocytes, fourth ventricle of the brain and in the spinal chord. Expression of FREAC-3 and -5 were ubiquitous while FREAC-7 and -8 could not be found in any tissue examined. The degree of DNA distortion as a result of DNA binding could be estimated using a circular permutation assay. The mobility ratio between the slowest and the fastest of the permutated complexes was interpolated to standard curves formed by A-tract DNA. This interpolation suggested that the FREAC proteins bend DNA to an angle of approximately 90°. The chromosomal localisations of six FREAC genes were determined. This was done by fluorescence in situ hybridisation (FISH) with cosmids, each containing a FREAC gene, to human metaphase chromosomes. The results were verified by hybridising genomic DNA from somatic cell hybrid panels of mouse hybrid cell lines, containing individual human chromosomes, to specific FREAC probes. The results were: FREAC-1, 16q24; FREAC-3, 6p25; FREAC-4, 5q12-q13; FREAC-5, 9pericen; FREAC-6, 5q34, and FREAC-8, 1p32. The full length c-DNA and the corresponding genomic DNA of FREAC-4 were cloned and the transcription start was determined. The gene contains no introns, and codes for a 2.5 kb transcript. The 465 amino acid long sequence harbours a hyper-acidic domain in the N-terminus and in the C-terminal region there are homopolymeric runs of prolines. Co-transfection assays showed that the FREAC-4 promoter is regulated by the Wilm's tumour protein (WT-1) and p53.