3.A.25 The Symbiont-specific ERAD-like Machinery (SELMA) Family

SELMA is a pre-protein translocase residing in the second outermost membrane of many four-membrane-bound plastids (Bolte et al., 2011). These complex plastids are found in ecologically important algal groups, such as cryptophytes, haptophytes and heterokontophytes, but also in pathogenic human parasites like Plasmodium falciparum and Toxoplasma gondii, which belong to the group apicomplexa. These complex plastids were acquired by engulfment and subsequent reduction of an ancestral algal cell by a eukaryotic host cell - a process termed secondary endosymbiosis. Since many genes of the endosymbiont were transferred to the nucleus of the host cell during evolution, mechanisms inevitably evolved for transporting such nucleus-encoded pre-proteins back into the plastid across up to four membranes: (i) the outermost membrane, which is continuous with the ER of the host cell or, rather, stays in vesicular contact with the ER in the case of apicomplexans, (ii) the second outermost membrane, which may correspond to the plasma membrane of the eukaryotic endosymbiont, and (iii) membranes three and four, the envelope of the primary plastid, which presumably still harbour TOC/TIC-like translocation machineries (TC#3.A.9).

An ERAD-like transport machinery mediates pre-protein transport across the second outermost plastid membrane. Core components include the Der1 proteins, Cdc48 with its co-factor Ufd1, and a complete set of proteins needed for ubiquitination (E1, E2, E3) in cryptophytes, heterokontophytes, haptophytes and apicomplexa. One set of these ERAD factors is specific for the host cell, whereas the second version originates from the red algal endosymbiont. As the symbiont's ER is not retained within the periplastidal compartment (PPC), which is derived from the former cytosol between membranes two and three, those symbiont-specific factors might play an ERAD-independent role, such as that of a protein translocation system across the second outermost plastid membrane (Bolte et al., 2011).

In a heterokontophyte, the diatom Phaeodactylum tricornutum. The symbiont-specific membrane proteins sDer1-1 and sDer1-2 are localised within the second outermost plastid membrane and exhibit a similar topology to Der1 proteins of the conventional ERAD-L system (3.A.16). The symbiont-specific Der1 proteins are able to form homo- and hetero-oligomers, a feature that supports the idea that these proteins might form a translocation channel. The symbiont-specific Der1 proteins might also play a role in discrimination and sorting of periplastidal and stromal pre-proteins since only periplastidal pre-proteins show interaction with the sDer1 complex. A conditional knock-out of one of the symbiont-specific Der1 proteins in Toxoplasma gondii blocks pre-protein transport into the apicoplast, providing strong evidence that SELMA has an ERAD-independent function in pre-protein transport to the complex plastid.

Analogous to protein translocation in the conventional ERAD-L system (3.A.16), substrates are inserted into a membrane, to be ubiquitinated on the cytosolic side by E1, E2 and E3 enzymes and, subsequently, to be extracted by an AAA-ATPase providing the energy for translocation. Since the SELMA machinery functions in a different context than ERAD-L, several modifications must have occurred during evolution. Most obviously, the recognition process on the extra-plasmatic side (the ER-lumen) was adapted as, in the case of SELMA, plastidal pre-proteins are the substrates instead of misfolded proteins. Accordingly, a homolog of the Hrd3p protein - a core receptor component of the ERAD membrane complex - has not been identified for SELMA. In fact, the transit peptide of plastid pre-proteins is essential for transport across the second outermost membrane and, hence, some kind of transit peptide receptor is presumably involved in the recognition process. Subsequent to substrate recognition, pre-proteins have to be inserted into a translocation channel, which might be formed by the oligomeric sDer1-1/sDer1-2 membrane complex (Hempel et al. 2009). In the conventional ERAD-L system, the Sec61 complex, along with the Der1p protein and the E3 ligase, Hrd1, are the prominent candidates for building the channel. When emerging on the periplastidal side, i.e. the former cytosol of the red algal endosymbiont, pre-proteins are ubiquitinated by the symbiont-specific ubiquitination machinery comprising the proteins sUba1 (ubiquitin activating protein), sUbc4 (ubiquitin conjugating protein) and the ubiquitin ligase ptE3P (P. tricornutum E3 enzyme of the PPC). Ubiquitinated pre-proteins are subsequently bound by the symbiont-specific AAA-ATPase sCdc48 and its co-factor sUfd1, which might provide the energy to pull the proteins out of the membrane. When substrate translocation is completed, pre-proteins are not targets of degradation as in the ERAD system, but still have to be de-ubiquitinated to guarantee correct maturation, functionality and, probably, further transport in the case of stromal pre-proteins. This final step might be carried out by the periplastidal de-ubiquitinating enzyme ptDUP (P. tricornutum deubiquitinating enzyme of the PPC), which was recently identified.

The reaction catalyzed by the SELMA system is:

translocation across
pre-protein                         →                                      pre-protein
(endoplasmic        the second outermost membrane         (periplastidal
reticular lumen)           of 4-membrane plastids            compartment (PCC))