Evaluation of the dissertation submitted by Camilla Raiborg for the degree of Doctor of Philosophy Title: Hrs makes receptors silent. A key to endosomal protein sorting.

This dissertation is based on five manuscripts which have all been published in major, international, refereed, cell biology journals. The candidate is first author on three of the papers and is second author on the two other papers. Taken as a whole, the dissertation is a detailed and penetrating study of the role Hrs in membrane traffic and it provides new insights into the molecular mechanisms whereby sorting of transmembrane proteins takes place in the endocytic pathway. The Dissertation contains, in addition to the five papers, an introduction and a general discussion. The introduction gives a brief and adequate overview of endocytic trafficking with emphasis on protein sorting and the main players in the investigation: clathrin, ubiquitin and Hrs. Overall, the introduction is well written and includes appropriate references to the wide literature on endocytic membrane traffic and protein sorting. An important part of the dissertation is the General Discussion of the data published in the five papers. This part of the thesis is not only an excellent discussion/evaluation of the data obtained in relation to earlier reports from the group of the candidate and from the many other groups working in the field. It also points to a number of experiments that may be done to reveal novel molecular mechanisms involved in the complex network that controls sorting of endosomal transmembrane receptors.

The main objective of the doctoral project was to determine the molecular mechanisms whereby Hrs regulates trafficking of receptors in the endocytic pathway. At the outset of the work it had been shown that Hrs belongs to the group of FYVE-proteins, and from the work of the group of Harald Stenmark it was reasonable to assume that Hrs could be associated with early endosomes by binding to PI(3)P, similar to EEA1. This turned out to be a fruitful working hypothesis.

The initial studies (presented in paper I) showed that the FYVE-domain was necessary but not sufficient for binding to early endosomes. By transient overexpressing different parts of the Hrs-protein and analyzing their subcellular localization by confocal immunofluorescence microscopy it was shown that in addition to the FYVE-domain the second coil-coil-domain of Hrs is necessary for endosomal targeting of Hrs. Whereas EEA1 binds to rab5 (in addition to PI(3)P) Hrs did not intereact with rab5.

To determine to what extent the two FYVE-proteins colocalize on endosomes, confocal immunofluorescence microscopy was used (Paper II). In this connection the authors took advantage of the earlier observation that transfection of cells with a constitutive active form of rab5 leads to formation of very large early endosomes. Using this approach it was possible to demonstrate that microdomains in the endosomal membrane containing EEA1 were separated from those containing Hrs, a very interesting observation that suggested that the two FYVE-proteins could be involved in different steps in the endocytic pathway. This notion was further supported by the finding that even clathrin, dynamin and Eps15 were associated with Hrs.

The distinct localizations of Hrs and EEA1 were confirmed in untransfected cells using electron microscopy (Paper III). A natural next question was to determine the function of the complex containing Hrs in the endocytic pathway. Earlier studies had suggested that Hrs might contain a ubiquitin-interacting motif (UIM) and that ubiquitin is a prerequisite for sorting certain membrane-proteins to lysosomes for degradation. The studies presented in paper III confirmed and extended these suggestions. It was shown that Hrs can interact directly with ubiquitin via its UIM (albeit with very low affinity). A particularly elegant experiment was to fuse the transferrin receptor to ubiquitin and then compare its intracellular transport with the wild-type receptor. This study demonstrated a striking colocalization of the Hrs and the ubiquitinated transferrin-receptor, whereas the wild-type receptor was not seen in the Hrs-containing microdomains. The results of paper III clearly indicated that Hrs interacts with ubiquitinated cargo, and that this may be a step in the transport of ubiquitinated proteins to lysosomes.

Paper IV presents important additional information about the central role played by Hrs in the endocytic pathway and in particular its role in sorting of membrane protein for lysosomal degradation. The data shows that Hrs forms a complex with STAM and the endocytic protein Eps15. All three proteins in this complex bind ubquitinated proteins, and this suggests that Hrs, Eps15 and STAM are subunits of a larger multivalent ubiquitin-interacting complex that may bind proteins (e.g. EGFR) that are ubiquitinated on several lysine-residues. Binding of ubiquitin to UIM is of low affinity. In this study small interfering RNA was used to deplete Hrs and this approach gave results that underlined the central role played by Hrs in forming the complex between STAM, Eps15 and Hrs on early endosomes. An additional finding in this study was that Vps4 may regulate dissociation of the Hrs/STAM/Eps15-complex from endosomes.

The purpose of the last paper in the thesis was to determine whether Hrs were involved in downregulation of the G-protein-coupled receptor CXCR4. The central role of Hrs in downregulation of EGFR had been clearly demonstrated in earlier studies. CCCR4 was known to be ubiquitinated as a prerequisite for lysosomal degradation. The work resulted in several interesting/important findings: First, the CXCR4 was found to be ubiquitinated specifically by a member of the Nedd4-family of E3 ubiquitin ligases, AIP4. This ligase does not ubiquitinate the EGFR. Further experiments showed that CXCR4 is sorted via the Hrs-pathway and that AIP4-ligase also ubiquitinates Hrs. Using siRNA to deplete Hrs it was found that the lack of Hrs impaired the degradation of CXCR4. Again, Vps4 was needed for degradation of the CXCR4.

Conclusions: The Dissertation focuses on a central theme in cell biology, the sorting of endocytosed proteins. The common denominator in the thesis is the Hrs and during the project several components were shown to cooperate with Hrs in the sorting process: clathrin, STAM, Eps15, AIP4 etc. The work is well planned and the methods used are excellent and always up-to-date and adapted to the problems investigated. The results presented in the five articles give novel, solid and detailed information about molecular mechanisms involved in sorting and targeting of endocytosed proteins. The dissertation is a major contribution to novel insight into receptor-downregulation. The work follows a logical path, questions raised in one paper are subject to new investigations in the next. As a consequence, the dissertation gives a rounded overview of the role of Hrs and its fellow players in the endocytic pathway/membrane traffic.The thesis-work brings the field a major step forward, and at the same time it raises a number of questions. Judged from the present thesis, the candidate may certainly be an important future contributor in the field.

Overall this dissertation is of an appropriate standard for a PhD. The aims of the work described are clearly expressed and high quality experiments have been carried out and properly interpreted. The reprints/manuscript included in the dissertation, together with the general discussion, demonstrate that the candidate has an excellent scientific grasp together with a mature understanding of the field of research.

In summary, this dissertation is worthy of being defended in its present form.

Cambridge/Oslo February 1st 2004

Elizabeth Smythe, Trond Berg and Henrik Huitfeldt
Page visits: 2969