The most direct approach to study the physiological role of a protein of unknown function (Functional Genomics) is to change its expression pattern in an intact organism and analyze the phenotypic consequences of this manipulation. The introduction of a method to generate stably transgenic Xenopus laevis has paved the way to the use of tissue / cell- and developmental stage-specific promoters allowing to study the physiological function of proteins in a defined set of fully differentiated cells. Whereas stable (over)expression of proteins in Xenopus is now within reach, stable inhibition of protein expression can only be accomplished randomly, by gene trap approaches. We here report our efforts to induce stable RNA interference (RNAi) in X. laevis via transgene-driven expression of inverted repeats. Stable, and muscle- and neuron-specific knock-down of expression of exogenous green fluorescent protein (GFP) reporter was achieved via RNA polymerase II promoter-driven expression of long GFP RNA duplexes. Unfortunately, our attempts to induce RNAi directed against various endogenous targets, based on the use of RNA polymerase II and III promoters, and long and short inverted repeats have not resulted in a reliable protocol for stable, transgene-driven RNAi in Xenopus. In the second part, we present an example of the use of a cell-specific promoter for functional studies. Cell-specific transgene overexpression of a GFP-tagged member of the p24 family thought to be involved in intracellular protein transport was achieved and this manipulation of the intermediate pituitary melanotrope cell had a phenotypic consequence at its physiological target, the skin melanophore. Thus, the traditional experimental advantages of X. laevis combined with the recently developed technique of stable, non-mosaic Xenopus transgenesis make this lower vertebrate an attractive model organism for Functional Genomics.