Limb Patterning: A Study in Polydactyly

Simpson-Golabi-Behmel syndrome (SBGS) is an X-linked condition, which, when found in humans, results in both pre- and post-natal overgrowth. Characteristically, humans with this disorder display a wide range of phenotypes, including distinct facial appearance, cleft palate, congenital heart and renal defects, supernumerary nipples, vertebral and rib defects, and postaxial polydactyly.

It has been shown that these malformations are caused by spontaneous mutations in the glypican-3 (Gpc3) gene. Glypicans represent a family of six cell surface heparan sulfate proteoglycans in vertebrates that perform a variety of functions, including binding of growth factors. To further study glypicans' role in this disorder, Dr. Scott Saunder's lab created a Gpc3 knockout mouse. Although this knockout mouse did display a wide range of the phenotypes associated with SGBS in humans, postaxial polydactyly was never seen. Because of the absence of the sixth digit, it was thought that another protein might be associated with polydactyly in mice. Previous studies in Drosophila had implicated a connection between glypicans and bone morphogenetic proteins. To explore this hypothesis in vertebrates, Gpc3 mutant animals were mated to mice bearing a mutation in a specific BMP, BMP4. The resulting offspring displayed phenotypes similar to the Gpc3 knockout mice. The new mice also displayed rib malformations and postaxial polydactyly, the focus of our study.

After breeding this new strain, subtractive hybridization was used as a means of detecting genetic differences between Gpc3 hemizygous knockout, heterozygous BMP4 mice (Gpc3/y; BMP4/+) and Gpc3 hemizygous wildtype, BMP4 heterozygous mice (+/y; BMP4/+). RNA from the limbs of 11.5-day-old embryos was isolated from both populations and converted into cDNA. Through a series of reactions, alike sequences were annealed to one another, and differential strands were amplified exponentially through PCR. Amplified sequences were then subcloned into the pCR 2.1 vector. By transforming these constructs into competent bacteria, a library of possible differential clones was created.

Unsubtracted Gpc3/y; BMP4/+ and +/y; BMP4/+ cDNA was then run out on an electrophoresis gel. After transferring the DNA from these gels onto a nylon membrane, probes created from the screening library were used to identify truly differential sequences. When using a specific probe, if a band was visible in either the Gpc3/y or the +/y lane, but not in the parallel lane, then the sequence was in fact differential. This method allowed for positive identification of differential clones and demonstrated to which cDNA population the clone was specific. These differential bands were then sequenced. Using the Blast database on the Internet, searches were run to identify possible human or mouse EST's or non-redundant sequences for each differential sequence.

As of now, nine differential sequences have been amplified using subtractive hybridization. Six of these nine have been sequenced and searched for on the Blast database. Of these six, one has been found to match exactly to a previously identified sequence, high mobility group protein-l. Two sequences were found to be similar to mouse EST's, and one was related to the knockout construct. The latter was evidence that the hybridization and screening procedures worked because it is known that the knockout construct is specific to the knockout mouse. The procedure positively identified this differentiation. Remaining searches will be needed to identify if the remaining sequences have any positive matches, and further tests must be run to ensure the remaining clones are differential. Ultimately then, experiments will be done in hopes of identifying how this differential expression causes polydactyly in vertebrates.