Unit: Nematode Genetics
Director: Bernard LAKOWSKI
Mutations in the human presenilin genes can cause Familial Alzheimer's disease. Using the Nematode Caenorhabditis elegans we have identified several suppressor of presenilin' (spr) genes that regulate presenilin transcription. The human homologues of these genes repress the expression of many neuronal genes in non-neuronal cell types. We are also characterising dog-1, a C. elegans DNA helicase involved in genomic stability, in order to use it better as a tool for genetics.
1) Presenilin suppressors
The Nematode Caenorhabditis elegans is a powerful model system to study neurobiology and development. We are using C. elegans to study the genetics of presenilin genes. Presenilins are polytopic proteins that are found as part of a high molecular weight complex that cleaves certain types of type I transmembrane proteins in the middle of the transmembrane domain. Mutations in the human presenilin genes lead to Familial Alzheimer's Disease by affecting the processing of the Amyloid Precursor Protein (APP). In all animals, presenilin genes are essential for normal development as they are necessary for the activation of Notch-type receptors.
To try to understand more about this important class of proteins, we have been using genetic approaches in C. elegans to identify factors that can modify the effect of mutations in presenilin genes. Loss of sel-12 presenilin activity in C. elegans results in a strong egg-laying defect. In screens for mutations that suppress this defect, five spr genes (for suppressor of presenilin) were identified by us and others. These genes encode components of a putative transcriptional repressor complex that normally represses the transcription of a second presenilin gene, hop-1, and presumably other targets. The SPR proteins are similar to components of the human REST-CoREST complex, which represses the expression of certain neuronal genes in non-neuronal cell types. spr-3 and spr-4 encode C2H2 zinc finger proteins that weakly resemble one another, as well as the human REST (RE1 Silencing Transcription Factor, also known as NRSF) transcriptional repressor. spr-1 and spr-5 encode C. elegans homologues of two core components of the CoREST co-repressor complex. SPR-5 is a polyamine oxidase-like protein, while SPR-1 is homologous to CoREST itself. spr-2 encodes an nucleosome assembly factor protein and is most similar to the human oncogene SET. We have been doing new screens to find additional components of this complex and other types of genes that can suppress sel-12. In two new screens we have recovered over 50 new spr mutations and we are in the process of characterizing these mutations. As expected, we have recovered new alleles of all previously identified spr genes, but we have also recovered several mutations that may define additional spr genes.
The CoREST complex has recently been shown to link local modification of core histones to regional chromatin silencing. Silencing of different genomic regions is essential for the maintenance of a cell's fate and defects in silencing are associated with cancer. The human homologues of the spr genes are implicated in several types of cancer. It has also been suggested that the mislocalization of REST may contribute to the progression of Huntington's disease. We are using the C. elegans spr genes to address the biological and biochemical function of CoREST-like complexes and their associated transcription factors in order to better understand the role of these complexes in transcriptional repression and disease.
2) The mutator dog-1
The Caenorhabditis elegans gene dog-1 (for deletions of guanine rich DNA) encodes a DEAH box DNA helicase most similar to the human BACH1 gene (Brca1 associated helicase). It has been reported that loss of dog-1 function leads to genomic instability and causes a high frequency of small deletions at poly-guanine (poly-G) stretches longer than 17 nucleotides. Based on the spectrum of dog-1 induced mutations, it was suggested that DOG-1 may be required to resolve three-dimensional structures formed by poly-G stretches during lagging strand DNA synthesis. This, along with the fact that BACH1 binds to the breast cancer susceptibility gene Brca1, suggests that BACH1 may have a role in genomic instability and cancer in humans.
We are trying to use the extremely unusual and restricted spectrum of dog-1 induced mutations to advance forward and reverse genetics in C. elegans. To better characterise the spectrum of mutations induced by dog-1, we have carried out several screens. So far we have identified dog-1 induced mutations in the presenilin suppressor genes spr-3 and spr-4, as well as in dpy-1 and dpy-13, two components of the worm's cuticle. We have also identified mutations in unc-34, the C. elegans enabled homologue that is involved in axonal guidance, and unc-10, the Rim homologue that is involved in synaptic transmission. Our preliminary results confirm that dog-1 does induce a high frequency of deletions at poly-G stretches but that it is not totally specific for such sites. We also find that the deletions are often larger than previously reported and that DOG-1 may not be functioning exactly as proposed. However, the restricted spectrum of dog-1 induced mutations can indeed aid in cloning genes identified by mutation (forward genetics) and could be used to target deletions in selected gene sequences (reverse genetics). This is potentially very interesting, as it is not currently possible to knock out genes in C. elegans by homologous recombination. We hope that a better characterisation of the dog-1 mutational spectrum will help us better use dog-1 as a genetic tool and give insight into the normal function of this gene and its human homologue.
Keywords: suppressors, transcriptional repression, Caenorhabditis elegans, Alzheimer~Rs disease, mutator, DNA helicase