Ubiquitin/proteasome system

To the editor

Dantuma and colleagues¹ have recently proposed the use of engineered green fluorescent proteins as substrates for the quantification of the ubiquitin/proteasome-dependent proteolytic system in living cells. Because of the critical involvement of this pathway in many essential processes, this effort is greatly appreciated and the development of new assays urgently needed. However, what Dantuma et al. have really developed is an assay representative of only a limited fraction of the ubiquitin/proteasome system. In fact, it is well known that the ubiquitin–protein ligases (E3s) responsible for substrate recognition can be grouped in at least four subtypes with regard to structure and degradation signal recognition². The fusion proteins described by Dantuma et al. utilize only one of such pathways (the N-end rule) that in addition has a yet unknown function and thus cannot be considered representative enough to be used in high-throughput screening programs of compounds that selectively modify proteolysis in vivo. Furthermore, the UFD (Ubiquitin Fusion Degradation) pathway is also in search of a defined function in mammalian cells. The only possible implication for UFD1 in humans appears to be the DiGiorge syndrome in 22q11.2 deletions³.

Huang and colleagues⁴ have reported that expression of a green fluorescent protein in transgenic mice causes cardiomyopathy. Thus the accumulation of the reporter cannot be considered without consequences for the cell, making the suggested correlation between ubiquitin/proteasome-dependent proteolysis and toxicity difficult to evaluate with confidence. Finally, green fluorescence proteins constructed to be a substrate for the N-end rule pathway seem instead to follow the UFD pathway, at least in Dictyostelium discoideum ⁵.

In conclusion, the green fluorescent protein reporters described by Dantuma et al. are certainly of interest but do not yet provide an informative tool useful in the screening of inhibitors that may interfere with the multiple steps (ubiquitin tagging, unfolding, and proteolysis) of the ubiquitin/proteasome pathway. We believe that the ubiquitin/proteasome pathway is involved in so many cellular processes that it could not be properly described by a single model substrate and that the minimal requirement should be the availability of substrates recognized selectively by the four subtypes of ubiquitin–protein ligases.