Isn't rotation inherently non-isotropic? (And, yes, a hypersphere is a topological oddity, but even so I would have thought that a cosmological rotation implied some sort of distinguishable direction.)