Lysine desert

Lysine desert is an extended lysine-free region of a protein. Because lysine side chains are common sites of ubiquitin attachment and related protein modifications, lysine deserts have been studied in the context of ubiquitination, pupylation and proteasome-dependent protein degradation.

The term lysine desert was introduced by Sharma and colleagues in a study of a lysine-free region in the yeast SUMO-targeted ubiquitin ligase complex Slx5-Slx8.

Biological significance

Lysine depletion was first discussed as a possible way for some proteins to avoid ubiquitin-mediated proteasomal degradation. Hazes and Read proposed that several AB toxins enter cells, travel to the endoplasmic reticulum, and then exploit endoplasmic reticulum-associated degradation, a cellular pathway that moves misfolded proteins from the endoplasmic reticulum into the cytosol for degradation. They noted that the catalytic A chains of several endoplasmic reticulum-directed AB toxins contain few lysine residues, unlike their B chains or toxins that enter the cytosol by other routes. They suggested that this low lysine content could help the AB toxins escape ubiquitin-mediated degradation after reaching the cytosol.

Experimental studies later supported this hypothesis for ricin A chain and abrin A chain. Introducing additional lysines into these proteins reduced their toxic potency and made them more sensitive to ubiquitin- and proteasome-dependent degradation. This suggested that low lysine content can help some toxins avoid degradation after entering the cytosol from the endoplasmic reticulum.

A related mechanism was described in the yeast quality-control ubiquitin ligase San1, which targets misfolded nuclear proteins for proteasomal degradation. San1 contains intrinsically disordered substrate-binding regions that lack lysine residues. Introducing single lysines into these regions was sufficient to destabilize San1, supporting the idea that local lysine depletion protects this ubiquitin ligase from self-ubiquitination and degradation. A later bioinformatic analysis found San1-like features in several yeast and human ubiquitin ligases, suggesting that lysine suppression may occur more broadly among proteins involved in the ubiquitin–proteasome pathway.

Another yeast example was described in Slx5, a subunit of the Slx5-Slx8 SUMO-targeted ubiquitin ligase complex. Lysine-introducing mutations within an extended lysine-free region of Slx5 increased its self-ubiquitination and partial proteolysis, supporting the idea that lysine deserts can protect ubiquitin ligases from excessive degradation.

Further studies broadened the view of lysine deserts beyond individual yeast ubiquitin ligases. Long and evolutionarily conserved lysine-free regions have been reported in several human ubiquitin–proteasome system components, including ubiquitin ligases such as RNF115 and RNF126, and proteasomal shuttle factors such as BAG6, RAD23A, UBQLN1, and UBQLN2. In selected cases, introducing lysines into these regions increased ubiquitination and reduced protein abundance or altered protein function.

Studies of VHL and SOCS1, lysine-poor substrate receptors of cullin-RING ubiquitin ligase complexes, suggest that lysine deserts do not necessarily prevent ubiquitination altogether. Even when deprived of lysines, both proteins could still undergo non-lysine ubiquitination, but their degradation mechanisms differed: VHL remained dependent on ubiquitin and the proteasome, whereas SOCS1 was degraded by the proteasome independently of ubiquitin.

A study of Caenorhabditis elegans EEL-1, the nematode ortholog of human HUWE1 ubiquitin ligase, also suggests that lysine deserts can influence ubiquitin–proteasome system in ways beyond simply controlling protein abundance. Introducing lysines into conserved lysine-deficient regions of EEL-1 impaired EEL-1-dependent degradation of a ubiquitin-fused GFP reporter. For one tested position, this functional defect was not accompanied by a detectable change in EEL-1 protein level or localization.

Outside eukaryotes, lysine deserts have also been reported as a substantial feature of proteomes in bacteria with pupylation-dependent proteasomal degradation systems, including Mycobacterium tuberculosis and Mycolicibacterium smegmatis, as well as in mycobacteriophages, suggesting an early evolutionary emergence and possible relevance to bacterial ubiquitin-like modification by Pup.