CD-MPR: Quo Vadis?

Lysosomes are membrane-bound organelles that serve in the degradation of
many extracellular and intracellular macromolecules. Lysosomal biogenesis depends
on the delivery of newly synthesized lysosomal hydrolases. This process requires the
acquisition of the lysosomal targeting signal, the mannose 6-phosphate tag that is
specifically recognized by mannose 6-phosphate receptors (MPRs) in the TGN. The
receptor-ligand complex is subsequently packaged into clathrin-coated vesicles and
transported to early endosomes. The lower pH in the endosomal compartment causes
the dissociation of the MPR and the ligand. The lysosomal enzymes are transferred to
the lysosome, where they are activated, whereas the MPRs are transported from
endosomes back to the TGN where they mediate another round of transport. Two
distinct MPRs were identified and characterized - the 46 kDa cation-dependent (CD)
MPR and the ~300 kDa cation-independent (CI) MPR. This study concentrates on the
CD-MPR.
The intracellular trafficking of the CD-MPR is mediated by sorting signals
located in its cytoplasmic tail of 67 amino acids. The sorting motifs are recognized by
specific adaptor proteins that mediate the vesicular transport of the receptor. Although
several motifs and their interacting partners were identified in the CD-MPR, the
various trafficking steps are not yet fully understood. In this study we focused on the
characterization of two motifs of the receptor - the cysteine C30 and C34 which
undergo reversible palmitoylation and the acidic cluster of the casein kinase 2 (CK2)
phosphorylation site (E55-E56-S57-E58-E59).
The CD-MPR is transported efficiently from late endosomes back to the TGN
since only a very small percent of receptors are missorted to lysosomes where they
are rapidly degraded. This transport step depends on the palmitoylation of C34, and
additionally on the diaromatic motif F18W19. The membrane anchoring mediated by
the palmitate, 34 amino acids away from the trans-membrane domain, implies a
drastic conformational change on the cytoplasmic tail of the CD-MPR. The
diaromatic motif is likely to be better exposed to the interacting protein in the
palmitoylated than in the non-palmitoylated CD-MPR. Our hypothesis suggests that
the reversible palmitoylation regulates the sorting signals in the cytoplasmic tail of
the receptor. This would require that the palmitoylation occur enzymatically. In Part I, we show that indeed the palmitoylation depends on a membrane-bound
enzyme. This palmitoyltransferase cycles between the plasma membrane and
endosomes. Close proximity of the palmitoyltransferase to the site where the
palmitoylation of the CD-MPR is required is optimal to ensure the presence of the
palmitoylated C34 in late endosomes. Thus, the localization of the
palmitoyltransferase supports our hypothesis of palmitoylation as a regulatory
mechanism for the sorting signals in the cytoplasmic tail of the receptor.
Correct sorting of the CD-MPR from the TGN to endosomes depends on the
D61-X-X-L64-L65 sequence, which interacts with GGA (Golgi-localizing, γ-earcontaining,
ARF-binding protein), a monomeric adaptor protein that mediates the
formation of clathrin-coated vesicles at the TGN. Several substrates of GGA have a
CK2 site upstream of the DXXLL motif and in two cases, phosphorylation by CK2
was shown to increase the affinity of GGA1 to cargo. The CD-MPR also contains a
CK2 site upstream of the DXXLL motif, but its involvement in GGA1 binding has
not been investigated so far. The CK2 site of the CD-MPR was shown to interact with
the adaptor protein 1 (AP-1), another protein involved in the sorting of cargo in the
TGN, possibly in cooperation with GGA. Previous reports on the requirement of
phosphorylation of the CD-MPR for binding to AP-1 were controversial. In Part II,
we analyzed the influence of the CK2 phosphorylation site of the CD-MPR in binding
to GGA1 and AP-1 and thus, in sorting in the TGN. A mutational analysis revealed
that high affinity binding between CD-MPR and GGA1 was dependent on the acidic
amino acid E59 and to a lesser extent on E58, while the phosphorylation of the S57 had
no influence, indicating that the GGA1 binding site in the CD-MPR extends to
E58-E59-X-D61-X-X-L64- L65. In contrast, AP-1 depended on all glutamates
surrounding the serine E55, E56, E58, E59 in the CD-MPR for binding, but was also
independent of the phosphorylation of S57. Therefore, we revealed that the
phosphorylation of S57 is not required for sorting in the TGN. Interestingly, the
binding affinity of GGA1 to the CD-MPR was 2.4-fold higher than that of AP-1 to
the partially overlapping binding site in the CD-MPR. Thus, we present a modified
model for the sorting process in the TGN, involving both GGA1 and AP-1, where the
different binding affinities, determine the order of binding to the partially overlapping
binding sites in the CD-MPR. First, GGA1 binds to the CD-MPR due to its higher
affinity and is subsequently released from the CD-MPR as a result of its autoinhibition caused by phosphorylation. This allows the AP-1 to bind and recruit
the remaining components for correct sorting of the CD-MPR in the TGN.
With our work we contributed to the understanding of specific transports steps
of the CD-MPR and thereby we are advancing towards the goal of fully elucidating
the trafficking of the receptor.