The few unfolded substrates to be examined fall into a simple pattern: transmembrane proteins are recognized by one system and soluble, secreted proteins by another. Vashist and Ng (page 41) take a closer look and find that the two transmembrane proteins previously examined had defects in their cytoplasmic domains. When the defects are instead in the lumenal domains of transmembrane proteins, these proteins are treated in the same way as soluble lumenal proteins.
Thus, the two recognition categories depend not on membrane association but on the site of the unfolded domain. Any protein with an unfolded cytoplasmic domain is recognized by one system (ERAD-C), whereas proteins (both transmembrane and soluble) that have unfolded domains in the ER lumen are recognized by a second system (ERAD-L). Proteins subject to ERAD-C are retained in the ER and destroyed rapidly, whereas proteins subject to ERAD-L must travel to the Golgi and return back to the ER before being destroyed. One possible function for the transport step might be the tagging of the unfolded protein with a Golgi-specific sugar modification.
If a protein has unfolded domains in both cytoplasm and lumen, the ERAD-C pathway takes precedence, and ERAD-L pathway proteins are not needed. This suggests either that the two pathways constitute an ordered series of checkpoints or that, when the ERAD-C and ER-to-Golgi recognition machineries are head to head, the former is more aggressive in pinning down its substrates. ▪