Evidence That Clusterin Has Discrete Chaperone and Ligand Binding Sites ² Johnathon N. Lakins, ‡ Stephen Poon, § Simon B. Easterbrook-Smith, | John A. Carver, ⊥ Martin P. R. Tenniswood, # and Mark R. Wilson* ,§ Institute for Medicine and Engineering, UniVersity of PennsylVania, 1170 Vagelos Research Labs, 3340 Smith Walk, Philadelphia, PennsylVania 19104-6383, Department of Biological Sciences and Department of Chemistry, UniVersity of Wollongong, Northfields AVenue, Wollongong, NSW 2522, Australia, School of Molecular and Microbial Biosciences, UniVersity of Sydney, Sydney, NSW 2006, Australia, and Department of Biological Sciences, UniVersity of Notre Dame, Notre Dame, Indiana 46556-0369 ReceiVed September 12, 2001; ReVised Manuscript ReceiVed October 31, 2001 ABSTRACT: Clusterin is the first identified extracellular mammalian chaperone and binds to a wide variety of partly unfolded, stressed proteins.Clusterin also binds to many different unstressed ligands including the cell surface receptor low-density lipoprotein receptor-related protein-2 (LRP-2). It is unknown whether clusterin binds to all of these many ligands via one or more binding sites. Furthermore, the region(s) of clusterin involved in these many binding interactions remain(s) to be identified. As part of an investigation of these issues, we expressed recombinant human clusterin in the yeast Pichia pastoris. The resultant protein had variable proteolytic truncations of the C-terminal region of the R-chain and the N-terminal region of the -chain. We compared the chaperone and ligand binding activities of this recombinant product with those of clusterin purified from human serum. We also tested whether the binding of clusterin to ligands could be inhibited by competitive binding with other clusterin ligands or by anti-clusterin monoclonal antibodies. Collectively, our results indicate that (i) clusterin has three independent classes of binding sites for LRP-2, stressed proteins, and unstressed ligands, respectively, and (ii) the binding sites for LRP-2 and stressed proteins are likely to be in parts of the molecule other than the C-terminal region of the R-chain or the N-terminal region of the -chain. It has been suggested that, in vivo, clusterin binds to toxic molecules in the extracellular environment and carries these to cells expressing LRP-2 for uptake and degradation. This hypothesis is supported by our demonstration that clusterin has discrete binding sites for LRP-2 and other (potentially toxic) molecules. Clusterin is a widely distributed, 70-80 kDa disulfide- linked heterodimeric protein of uncertain function. It was first described in 1983 as a secreted glycoprotein in ram testis fluid that enhanced aggregation (“clustering”) of various cells in vitro. Many homologues in other species have subse- quently been discovered. Clusterin is encoded by a single gene, and the translated product is (i) internally cleaved to produce its R- and -chains, which are linked by five interchain disulfide bonds, and (ii) extensively glycosylated, such that 30% of its final mass is N-linked carbohydrate ( 1). In most cases, mature clusterin is secreted from the cell; however, it remains intracellular in chickens (2). In humans, clusterin is found in blood plasma at about 100 µg/mL and in seminal plasma at about 400 µg/mL (3). Clusterin expression is increased in many experimental models of stress and disease states (1). Many studies suggest that clusterin functions in vivo as a protective molecule. We recently reported that clusterin has an in vitro chaperone action that is similar to but more potent than that of the intracellular small heat shock proteins (4). Furthermore, we recently demonstrated that clusterin inhibits stress-induced precipitation of proteins in undiluted human serum, suggesting that the levels of clusterin in biological fluids such as plasma may affect the rate or extent of progression of diseases associated with abnormally high levels of protein precipitation (e.g., Alzheimer’s, Creutzfeldt- Jakob, and Parkinson’s diseases) (5). Other studies have shown that cells are protected from stresses by overexpres- sion of clusterin, or by its addition to the medium surrounding cells (1). Recent studies with clusterin knock-out mice suggest that clusterin may protect cells from damage resulting from inflammatory reponses (6) but, in contrast, may enhance neurotoxicity in a model of hypoxic-ischemic injury (7). Collectively, these observations indicate that the interactions of clusterin with other biological molecules are likely to be of clinical importance. The history of clusterin research has been notable for the rapid succession of reports describing new “specific” inter- actions of clusterin with a diverse array of native biological ² Parts of this work were supported by grants to J.A.C. from the NHMRC (980497), to S.B.E.-S. from the ARC (DP0208752), the Australian Brain Foundation, and the Rebecca L. Cooper Medical Research Foundation, to M.P.R.T. from the USPHS (CA692331), and to M.R.W. from the ARC (X00106477 and DP0211310) and the University of Wollongong. J.N.L. was supported by an MRC Predoc- toral Fellowship, and S.P. was supported by an Australian Com- monwealth Postgraduate Scholarship. * Corresponding author. Fax: (61)-242-214135. Telephone: (61)- 242-214534. Email: mrw@uow.edu.au. ‡ University of Pennsylvania. § Department of Biological Sciences, University of Wollongong. | University of Sydney. ⊥ Department of Chemistry, University of Wollongong. # University of Notre Dame. 282 Biochemistry 2002, 41, 282-291 10.1021/bi0157666 CCC: $22.00 © 2002 American Chemical Society Published on Web 12/08/2001