CMO Progress Report

Lay Progress Report on the Search for the CMO (Craniomandibular Osteopathy) Gene

Submitted to the Cairn Terrier Club of America, March 7, 2003

Patrick J. Venta, Ph.D.

Small Animal Clinical Sciences

College of Veterinary Medicine

Michigan State University

East Lansing, MI 48824-1314

In our previous progress report we mentioned that other genes that cause Paget‘s disease in humans have been mapped to specific chromosomal intervals on human chromosomes 2 and 18, and now 6 and 10. Paget’s disease is a bone disorder that has been reported in the literature as potentially being homologous to the CMO gene. We last reported that we were attempting to build a map of the canine equivalent of the human chromosome 2 Paget’s interval to determine if it had been disrupted between the two species. We have now completed this map with 19 markers (most of them newly developed by us) and have determined that the canine interval is split into two nearly equal parts compared to the human Paget’s interval. The two parts fall on dog chromosomes 25 and 37. We are currently testing the genetic markers that we simultaneously developed with the map construction on these two chromosomes for variability in the CMO DNA samples that we have. Identification of such rearrangements where a gene has been mapped but not identified in humans is critical to testing intervals for causation so that we do not falsely exclude an interval because of a break. Once we have determined which markers are variable on both canine chromosomes in the CMO samples, we will be able to test for linkage of the unidentified Paget’s gene for causation of CMO.

We have made canine markers in the other recently described Paget’s chromosome intervals (human chromosome 6, 10, and 18) to determine if these intervals are intact or disrupted in the dog as compared to the human. The chromosome 18 region has been determined to be intact and very near one end of dog chromosome 1, although we are still in the process of determining if the chromosome 6 and 10 regions are intact. We are currently preparing to test canine chromosome 1 markers for variability in the CMO samples, and once variable ones have been found we will also test linkage of the unidentified Paget’s gene in this region. The same process will be done for the canine-equivalent of the human chromosome 6 and 10 regions once they have been examined for rearrangements compared to the human interval. We are also examining a canine gene called SQSTM1 that is on chromosome 5 in humans, but so far we have not been fortunate in finding a genetically variable site although we anticipate that one will be found.

We are also continuing the whole genome scan. We are about halfway through developing a new set of about 1000 evenly spaced genetic markers that are very likely to help increase our ability to map all canine genes of interest in addition to the CMO gene. These markers are based on known genes, rather than random chromosomal segments, and have homologues in the completely sequenced human genome so that it is possible to more easily transfer discoveries from the human to the dog. These third-generation markers are also based on genetic variation called single nucleotide polymorphism (SNPs, pronounced snips) which are much easier to automate than the second-generation random markers which are called micro satellites. Most of the markers described in the first two paragraphs of this report are in fact these new markers and have been applied early on to the CMO project. As these markers are developed they will be incorporated into the whole genome scan. The development of these markers is funded by mechanism unrelated to the CMO project, but which we are using in the scan for CMO gene. We are attempting to further refinements the automation for the typing of these markers to increase the rate at which markers can be put through the CMO pedigrees.

In addition to the work that is conducted specifically on CMO, we have also continued to work to develop general approaches to identifying all genes that predispose dogs to a wide variety of diseases. Some diseases, such as PSS (portosystemic shunt) do not appear to follow simple inheritance patterns, that does not depend on knowing whether the necessary gene operates in a dominant or a recessive mode. Assuming we have reasonable evidence that a necessary gene is required for the disease phenotype to appear, we can be certain that affected dogs must carry the mutant gene. We must also assume that there is only one mutant form of the gene that has become relatively widespread in a breed. Although we cannot prove this to be true in any given case until the mutant gene has actually been identified, this assumption has usually held for the canine disease genes that have been identified to date (examples, von Willebrand disease in Scottish terriers, PK deficiency in West Highland White terriers, and globoid cell leukodystrophy in Cairn terriers). For this reason we believe that the assumption is generally reasonable to make, and that all affected dogs will have the same identical mutant gene. Using the improved canine map when it has been completed, the rapid genotyping of the SNP-based markers, and with an estimate of a special property of genes in populations called (linkage disequilibrium) we believe that there is reason to hope that in a few years it may be possible many times to track down genes that do not have simple modes of inheritance using only affected dogs and a control population of unaffected dogs in the same breed.