Defining the mechanistic and functional details of an evolutionarily conserved non-canonical retromer pathway.

Lead Research Organisation: University of Bristol
Department Name: Biochemistry


Mammalian cells are composed of a variety of interconnected membrane compartments each composed of a unique combination of proteins and lipids. For cells to function correctly, proteins and lipids are required to be transported to specific compartments within this maze of membranes. Understanding how cells achieve this is a major challenge in modern cell biology. Our research focuses on the role of two lipids, called PI3P and PI3,5P2 in the regulation of transport within a specific membrane maze called the endocytic network. Altered levels of these lipids, and resulting defects within the network can lead to various diseases including neurodegenerative diseases such as Alzheimer's. In the current proposal, we outline a multidisciplinary programme of research that aims to describe how one family of PI3P-binding proteins - the sorting nexins - regulate a specific transport step within the endocytic network. We focus on a multiprotein complex called the retromer. Previous research has established that retromer is evolutionarily comserved - that is, it is present in relatively primative (e.g. yeast) as well as complex organisms (e.g. humans). Classically the human retromer contains the following proteins: VPS26, VPS29 and VPS35 and the sorting nexins, SNX1, SNX2, SNX5 and SNX6. Importantly, research has implicated retromer in various disease states. For example, perturbed retromer function may be involved in Alzheimer's disease, and various pathogens (e.g. Salmonella) may also require retromer function for their pathology. Characterising the detail of retromer function is therefore important if we are to fully understand certain human diseases. Through an international collaboration with the laboratory of Dr Rik Korswagen (Hubrecht Institute, The Netherlands), we have recently established the presence of a 'non-classical' retromer. While this shares the classic retromer proteins, VPS26, VPS29 and VPS35, it does not contain SNX1, SNX2, SNX5 or SNX6 but instead contains an entirely distinct sorting nexin, called SNX3. Like the classical retromer, the non-classical retromer is also evolutinarily conserved. This is an exciting discovery since it has established that when examining retromer and its role in human diseases, one needs to also consider the non-classical retromer pathway. In the current project we propose to define in more detail the molecular assembly of the non-classic retromer, and elucidate how it functions alongside the classical retromer in regulating transport of specific proteins through the endocytic network. Biochemical, molecular cell biology and experiments in whole organism genetics will be performed in order to obtain a broad picture of the non-canonical retromer function, from the molecular components and interactions through to physiological consequences for the whole organism. Successful completion of the proposed research, will generate invaluable, fundamental information on this new pathway that will add significantly to our understanding of retromer function in normal and disease-related contexts.

Technical Summary

The canonical retromer is an evolutionarily conserved multiprotein complex, which regulates endosome-to-Golgi transport. In mammalian cells, it is composed of a cargo-selective trimer of VPS26:VPS29:VPS35 and a membrane-bound coat containing the sorting nexins (SNXs), SNX1, SNX2, SNX5 and SNX6. These SNXs possess two membrane binding domains: a phox homology (PX) domain that associates with the endosome-enriched phospholipid, phosphatidylinositol 3-monophosphate (PI(3)P), and a BAR domain which, through forming a rigid 'banana-shaped' structure associates with and drives the formation of highly curved endosomal tubules. In collaboration with Dr Rik Korswagen (Hubrecht Institute, Utrecht, The Netherlands), we have recently characterised an evolutionarily conserved non-canonical retromer. Utilising a variety of mutlidisciplinary procedures, we have defined that while this pathway shares the cargo-selective VPS26:VPS29:VPS35 trimer, it requires a distinct SNX coat. This comprises SNX3 - a SNX that only contains one recognised domain, the PI(3)P-binding PX domain. Importantly, we have established that in C. elegans, Drosophila and mammalian cells, it is the non-canonical, and not the canonical retromer, that is required for the correct endosome-to-Golgi sorting of Wntless, a Wnt-binding protein vital for secretion of Wnt developmental morphogens. Here we aim to address two major objectives: (1). Define the assembly of the non-canonical retromer complex, establishing how this relates to the canonical retromer. In addition, build on preliminary proteomic data revealing an association of SNX3 with clathrin, addressing whether in the absence of a membrane tubulating BAR domain, this assist the formation of vesicular-based carriers. (2). By studying Wntless trafficking in cultured mammalian cells and genetically within C. elegans, elucidate how assembly of the non-canonical retromer relates to its function in endosome-to-TGN sorting of Wntless.

Planned Impact

Communications and Engagement Research papers and awarded grants are made public through the departmental and university websites. Cell biology research at the University of Bristol can also be accessed through the 'Dynamic Cell Biology' website ( As an enthusiastic cell biologists Professor Cullen has always been keen to discuss research with younger scientists. Because of extensive use of live cell imaging and EM their research has always had a visualising appealing element. Every opportunity has therefore been taken to engage with students from the local Bristol area as part of the University's Schools week. This has mainly taken the form of oral presentations describing how by performing research into fundamental cell biology one can better understand human disease, and hence have a more scientific approach to therapeutic intervention. For example, Professor Cullen has delivered a public lecture on his research 'From driving combine harvesters to studying cancer biology'. In addition, movies from his research have been used by science teachers at the nearby Bristol Grammar School to inform their students that cells are not static 2D blobs like they see in their text books, but in reality are highly dynamic, extremely complex, 'cool and funky' structures. Collaboration A major component of the current application is to extend existing collaborations between the Cullen and Collinson (Bristol, UK) and Korswagen laboratories (Hubrecht Institute, The Netherlands). The Cullen labs interest in utilising biochemical and molecular techniques to dissect out the assembly of multiprotein complexes will be greatly enhanced through the internationally recognised expertise of the Collinson lab in this area. In contrast to the reliance of the Cullen lab on mammalian tissue culture, the Korswagen lab utilises model organisms, and in particular C. elegans, to identifiy molecular details that underly the cell biology of Wnt morphogenic signalling. Their laboratory was one of the first to establish, through C. elegans genetics, that the retromer is required for the trafficking of Wntless, a Wnt-binding protein that assists the secretion of Wnt proteins (Coudreuse et al., (2006) Science 312, 921-924; Yang et al., (2008) Dev Cell 14, 140-147). The practical experience of the Cullen and Korswagen laboratories therefore complement one another, allowing their research to probe deeper into how retromer-mediated endosome-to-Golgi sorting impinges on the physiology of a whole organism. For example, proteomic and Y2H screens combined with genetic screens, has allowed the two laboratories to define the role of the dyenin motor protein in retromer function (see Wassmer et al (2009) Dev Cell 17, 110-122). The extensive preliminary data presented in the current proposal, describing the SNX3-dependent non-canonical retromer pathway, has also emerged from this collaboration. Funding of the current application will further strengthen the collaboration, enhancing discussion, exchange of material and preliminary data that will ensure optimal delivery and impact of our research in a mutually beneficial way. Exploitation and Application The proposed research is unlikely to provide any direct commercial exploitation. However, in the event that potential applications are identified that are exploitable, notably in terms of intellectual property or knowledge transfer, then this will be handled by the university's Research and Enterprise department (RED); who engage closely with the BBSRC. Capability Communcation through press releases is undertaken by professional staff, who tailor these to the relevant audience. Further work in this area is undertaken by the PI depending on the technical level required. The university also runs regular courses on Science communication that is open to PI's and post-docs. Resource for the activity No additional resource req
Description We discovered that the SNX3-retromer associates with an evolutionary conserved protein complex that in yeast has been shown to regulate the biogenesis of vesicular transport carriers. In human this complex is composed of three proteins, ATP9A, MON2 and DOPEY1. Using a combination of biochemical and molecular cell biology procedures coupled with in vivo genetic analysis in C. elegans, we have established that the coupling of SNX3 with retromer and subsequent association with the ATP9A:MON2:DOPEY1 complex is essential for the endosomal sorting of Wntless and the generation of Wnt morphogenic gradients. Conceptually, our study established the molecular mechanism by which the SNX3-retromer complex functions to sort Wntless through the endosomal network in order to allow this protein to regulate the secretion of Wnt morphogens.

Additional outcomes:
This study provided part of the concept development behind a successful Wellcome Trust Senior Investigator Award made by Cullen.

The collaboration between the Cullen and Korswagen laboratories was extended and has continued to development post ending of the grant.
Exploitation Route De-regulation of Wnt signaling is an important feature in cancer development and has an increasingly interesting role in neurodegenerative disease. With evidence continuing to emerge defining the role of retromer in neuroprotection and retromer de-regulation in age-related Parkinson's disease and Alzheimer's disease, the new mechanistic understanding that has arisen from the current study is likely to provide important insight into retromer's de-regulation in disease, especially in the context of Wnt signaling. Such fundamental insight provides the necessary foundation on which to build ideas and concepts the underpin the rationale design of therapeutics.
Sectors Pharmaceuticals and Medical Biotechnology