Greater than the sum of its parts: the role of VARP-SNX27 binding in endosomal “supercomplex” formation

Posted by daviskd2 on Friday, February 14, 2025 in News.

In cells, multi-subunit coat protein complexes are responsible for coordinating the transport of transmembrane proteins and lipid cargo between membranes. At the endosome, cargo trafficking is mediated by retromer, a heterotrimer composed of VPS26, VPS29, and VPS35. Retromer serves as a “master regulator”, recruiting machinery to help deform the membrane and sort cargo into vesicles or tubules. Sorting nexin (SNX) proteins interact with protein coats and are characterized by the presence of a PX domain which recognizes phosphoinositides in membranes. Retromer specifically interacts with two SNX-BAR heterodimers (SNX1/SNX5 and SNX2/SNX6), which aid in membrane curvature through the presence of a BAR domain, and SNX27, which contains a PDZ and FERM domain used for recognizing cargo and GTPases, respectively.

In addition to SNX proteins, Retromer also associates with other accessory proteins that regulate endosomal trafficking. One example is VARP, a Rab effector that also displays GEF activity and can bind R-SNAREs. In addition to binding the VPS29 subunit of Retromer, an MS-proteomics study identified a potential interaction between VARP and SNX27. However, a biochemical basis for direct SNX27-VARP binding has not been established and it is unclear how VARP, SNX27, and Retromer work in conjunction to coordinate endosomal trafficking. In this study Mintu Chandra, research assistant professor, and Amy Kendall, lab manager, from the Jackson laboratory, use a combination of biochemical, biophysical, and computational modeling techniques to investigate these endosomal trafficking players.

The authors first set out to determine if SNX27 and VARP interact in vitro. Pulldowns between GST-tagged full-length SNX27 and full-length VARP demonstrated robust binding, and subsequent bio-layer interferometry (BLI) calculated sub-micromolar binding affinity between the two proteins. Computational models of SNX27-VARP binding generated by AlphaFold Multimer predicted an interaction specifically between the N-terminal region of VARP (N-VARP) and the PDZ domain of SNX27. The models further identified a short sequence in N-VARP with similarity to classical PDZ binding motifs which bound the conserved PDZ binding motif pocket in SNX27. When the identified N-VARP sequence was mutated, binding to the SNX27 PDZ domain was no longer observed by BLI.

In order to assess how these proteins work on membranes, the authors next turned to liposome pelleting assays and negative stain electron microscopy. SNX27 was shown to bind membranes enriched with PI(3)P and PDZ binding motif cargo and recruit Retromer, which cannot bind membranes on its own. Further, both SNX27 alone and SNX27/Retromer were capable of generating tubules as seen by negative stain electron microscopy.

Conversely, SNX2/SNX6 bound exclusively to Folch I membranes enriched in bis-phosphoinositides in the presence of its cargo, CI-MPR, and failed to recruit Retromer. The authors then explored the possibility of adding SNX27 to facilitate complex formation as SNX2 is capable of binding the FERM domain of SNX27. However, SNX2/6 was still only found on Folch I membranes and SNX27/Retromer only on PI(3)P membranes. Finally, the authors thought to include VARP, which resulted in the recruitment of Retromer, SNX2/SNX6, and SNX27 to PI(3)P membranes in the presence of PDZ binding motif cargo. This reconstitution is the first evidence of an endosomal “supercomplex” and reveals a vital role for VARP in mediating coat assembly.

Importantly, VARP-GFP also co-immunoprecipitated with SNX2 and VPS26 in cells, indicating the physiological relevance of this newfound complex. Make sure to check out the full paper in Science Advances and learn more about the future of efficient endosomal cargo sorting! ~ Cameron I. Cohen