Basically there are two main problems with using CO2 as a solvent for modified Earth-biology.
The first is that CO2 isn't a polar solvent and won't form hydrogen bonds in the same way that water does. This means, I'm fairly certain, that proteins immersed in supercritical CO2 will not fold, or if they do fold they'll fold in completely different ways to the way they fold in water. This means, most likely, that you'd have to find a whole other class of molecules to perform all the roles that proteins play in the cell, which include catalysing all of the cell's reactions as well as structural roles. Swapping out proteins for something else would require changing virtually every molecule in the cell.
The different properties of CO2 will also cause big issues with the cell's membrane. The membrane is made of lipids, which have a hydrophobic and a hydrophilic end. This is possible because of the polar nature of water, so it will be difficult if not impossible to form a similar kind of membrane in supercritical CO2.
The other issue you face is that water doesn't just play the role of a solvent in biochemistry, it's also a very important reagent. Many of the most important biochemical reactions either produce water as a side-product or use it up. So if you were using CO2 as a solvent but still using chemistry similar to Earth-life, you would still need a source of water. This means there would need to be a lot of water around anyway, presumably dissolved in the supercritical CO2.
Although nobody knows for sure, I like to think that life is possible in solvents other than water. But in order to exist it would surely need to be very, very different from the life we know. In your case we would not be talking about a simple case of changing a few genes, but about completely re-designing the molecular architecture of the cell from the bottom up. This might not be impossible, but it's a long way beyond what we're currently capable of.