Domain Structure and Lipid Interaction in Human Apolipoproteins A-I and E, a General Model

Hiroyuki Saito, Padmaja Dhanasekaran, David Nguyen, Paul Holvoet, Sissel Lund‐Katz, Michael C. Phillips

Journal of Biological Chemistry · 2003 · 180 citations · 34 references

DOIFull text

Open access

Concepts

Abstract

Detailed structural information on human exchangeable apolipoproteins (apo) is required to understand their functions in lipid transport. Using a series of deletion mutants that progressively lacked different regions along the molecule, we probed the structural organization of lipid-free human apoA-I and the role of different domains in lipid binding, making comparisons to apoE, which is a member of the same gene family and known to have two structural domains. Measurements of α-helix content by CD in conjunction with tryptophan and 8-anilino-1-naphthalenesulfonic acid fluorescence data demonstrated that deletion of the amino-terminal or central regions disrupts the tertiary organization, whereas deletion of the carboxyl terminus has no effect on stability and induces a more cooperative structure. These data are consistent with the lipid-free apoA-I molecule being organized into two structural domains similar to apoE; the amino-terminal and central parts form a helix bundle, whereas the carboxyl-terminal α-helices form a separate, less organized structure. The binding of the apoA-I variants to lipid emulsions is modulated by reorganization of the helix bundle structure, because the rate of release of heat on binding is inversely correlated with the stability of the helix bundle. Based on these observations, we propose that there is a two-step mechanism for lipid binding of apoA-I: apoA-I initially binds to a lipid surface through amphipathic α-helices in the carboxyl-terminal domain, followed by opening of the helix bundle in the amino-terminal domain. Because apoE behaves similarly, this mechanism is probably a general feature for lipid interaction of other exchangeable apolipoproteins, such as apoA-IV. Detailed structural information on human exchangeable apolipoproteins (apo) is required to understand their functions in lipid transport. Using a series of deletion mutants that progressively lacked different regions along the molecule, we probed the structural organization of lipid-free human apoA-I and the role of different domains in lipid binding, making comparisons to apoE, which is a member of the same gene family and known to have two structural domains. Measurements of α-helix content by CD in conjunction with tryptophan and 8-anilino-1-naphthalenesulfonic acid fluorescence data demonstrated that deletion of the amino-terminal or central regions disrupts the tertiary organization, whereas deletion of the carboxyl terminus has no effect on stability and induces a more cooperative structure. These data are consistent with the lipid-free apoA-I molecule being organized into two structural domains similar to apoE; the amino-terminal and central parts form a helix bundle, whereas the carboxyl-terminal α-helices form a separate, less organized structure. The binding of the apoA-I variants to lipid emulsions is modulated by reorganization of the helix bundle structure, because the rate of release of heat on binding is inversely correlated with the stability of the helix bundle. Based on these observations, we propose that there is a two-step mechanism for lipid binding of apoA-I: apoA-I initially binds to a lipid surface through amphipathic α-helices in the carboxyl-terminal domain, followed by opening of the helix bundle in the amino-terminal domain. Because apoE behaves similarly, this mechanism is probably a general feature for lipid interaction of other exchangeable apolipoproteins, such as apoA-IV. Exchangeable apolipoproteins (apo) 1The abbreviations used are: apo, apolipoprotein; ANS, 8-anilino-1-naphthalenesulfonic acid; CD, circular dichroism; DMPC, dimyristoylphosphati-dylcholine; GdnHCl, guanidine hydrochloride; ITC, isothermal titration calorimetry; PC, phosphatidylcholine. play critical roles in lipid transport and lipoprotein metabolism. This class of proteins (which includes human apoA, -C, and -E) have the same genomic structure and are members of a multigene family that probably evolved from a common ancestral gene (1Li W.H. Tanimura M. Luo C.C. Datta S. Chan L. J. Lipid Res. 1988; 29: 245-271Google Scholar). The last exon codes for a primary structure of 22-amino acid tandem repeats that are often separated by a proline residue. The 22-residue segments form amphipathic α-helices that mediate the interaction of these proteins with lipids (2Segrest J.P. Garber D.W. Brouillette C.G. Harvey S.C. Anantharamaiah G.M. Adv. Protein Chem. 1994; 45: 303-369Google Scholar). Knowledge of apolipoprotein tertiary structure in the absence and presence of lipid is needed to understand their functions at the molecular level. Given the common evolution, it might be expected that apolipoprotein molecules containing multiple 22-residue repeats (e.g. apoA-I and apoE) fold similarly to adopt a common domain structure. However, this concept has not been established because, as outlined below, current knowledge of apolipoprotein tertiary structure is rudimentary. High resolution structures for intact human apolipoprotein in the lipid-free state are not available. ApoE is perhaps the best defined because the crystal structure for the 22-kDa amino-terminal fragment was solved over a decade ago (3Wilson C. Wardell M.R. Weisgraber K.H. Mahley R.W. Agard D.A. Science. 1991; 252: 1817-1822Google Scholar). This domain forms a globular α-helix bundle that is similar to the structure of insect apolipophorin III (4Breiter D.R. Kanost M.R. Benning M.M. Wesenberg G. Law J.H. Wells M.A. Rayment I. Holden H.M. Biochemistry. 1991; 30: 603-608Google Scholar). In apoE, there are two independently folded domains: the 22-kDa amino-terminal domain contains the low density lipoprotein receptor binding region and the 10-kDa carboxyl-terminal domain has a high affinity for lipid and is responsible for lipoprotein binding (5Weisgraber K.H. Adv. Protein Chem. 1994; 45: 249-302Google Scholar). Recently, we showed that these two domains in apoE4 lead to two different lipid-bound conformations, providing a structural rationale for the variable receptor binding activity displayed by lipoprotein-associated apoE (6Saito H. Dhanasekaran P. Baldwin F. Weisgraber K.H. Lund-Katz S. Phillips M.C. J. Biol. Chem. 2001; 276: 40949-40954Google Scholar). In comparison, the tertiary structure of apoA-I is relatively poorly defined. Thus, various spectroscopic and proteolysis measurements are consistent with a helix bundle structure for apoA-I in solution (7Davidson W.S. Hazlett T. Mantulin W.W. Jonas A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13605-13610Google Scholar, 8Roberts L.M. Ray M.J. Shih T.W. Hayden E. Reader M.M. Brouillette C.G. Biochemistry. 1997; 36: 7615-7624Google Scholar, 9Kiss R.S. Kay C.M. Ryan R.O. Biochemistry. 1999; 38: 4327-4334Google Scholar). However, the α-helices are in a linear conformation in the crystal structure of an amino-terminal-truncated apoA-I molecule (10Borhani D.W. Rogers D.P. Engler J.A. Brouillette C.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12291-12296Google Scholar). Also, in contrast to apoE (11Morrow J.A. Segall M.L. Lund-Katz S. Phillips M.C. Knapp M. Rupp B. Weisgraber K.H. Biochemistry. 2000; 39: 11657-11666Google Scholar), guanidine hydrochloride (GdnHCl)-induced denaturation of apoA-I is monophasic (12Reijngoud D.J. Phillips M.C. Biochemistry. 1982; 21: 2969-2976Google Scholar), giving no evidence for separate unfolding of two domains. Indeed, the apoA-I molecule may not have a defined tertiary structure, because it exhibits characteristics of a molten globule (13Gursky O. Atkinson D. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 2991-2995Google Scholar). The structural plasticity (14Rogers D.P. Roberts L.M. Lebowitz J. Engler J.A. Brouillette C.G. Biochemistry. 1998; 37: 945-955Google Scholar) facilitates lipid-binding with the carboxyl-terminal domain playing a key role (for reviews, see Refs. 15Narayanaswami V. Ryan R.O. Biochim. Biophys. Acta. 2000; 1483: 15-36Google Scholar, 16Frank P.G. Marcel Y.L. J. Lipid Res. 2000; 41: 853-872Google Scholar, 17Brouillette C.G. Anantharamaiah G.M. Engler J.A. Borhani D.W. Biochim. Biophys. Acta. 2001; 1531: 4-46Google Scholar). We further examined the domain structure of apoA-I so that a systematic comparison with apoE could be made. To this end, we characterized six deletion mutants that progressively lack different regions along the molecule, amino-terminal (Δ1–43, Δ44–65, and Δ44–126), central (Δ123–166), or carboxyl-terminal (Δ190–243 and Δ223–243) regions of apoA-I. The results show that the two-domain model of apoE structure applies to apoA-I and that both proteins undergo a similar two-step mechanism of binding to a lipid surface. These findings support the notion that the tertiary structures of the larger members of the human exchangeable apolipoprotein gene family are related. Expression and Purification of Proteins—To express human apoA-I and engineered variants, the cDNA of interest was cloned into multiple cloning sites of the pET32a (+) vector, and the target protein was expressed as a fusion protein with the 109-amino acid thioredoxin at the amino terminus. The pET32a (+) vector contains the ampicillin resistance gene for clonal selection and a His-tag at the fusion linker region to facilitate purification of the expressed protein using a nickel affinity column. In addition, a thrombin recognition site was engineered at the fusion junction in such a way that thrombin cleavage of the thioredoxin leaves the target protein with two extra amino acids, Gly-Ser, at the amino terminus (18Morrow J.A. Arnold K.S. Weisgraber K.H. Protein Expr. Purif. 1999; 16: 224-230Google Scholar). The transformed expression host, Escherichia coli strain BL21–DE3, was grown in Luria Bertani medium, and expression was induced with 1 mm isopropyl-β-d-thiogalactopyranoside for 3 h. After the cell pellet was sonicated and centrifuged to sediment the cell debris, the supernatant was loaded onto a nickel-chelating, binding column. The fusion protein to the was mm with and with The apoA-I was and in GdnHCl, and separated from the fusion protein thioredoxin by and (18Morrow J.A. Arnold K.S. Weisgraber K.H. Protein Expr. Purif. 1999; 16: 224-230Google Scholar). The apoA-I at as by apoA-I was from human high density lipoprotein M. Lund-Katz S. Anantharamaiah G.M. P. A. J.P. Weisgraber K.H. Phillips M.C. J. Biol. Chem. 1999; Scholar). The human apoE4 and 22-kDa fragment expressed and as (6Saito H. Dhanasekaran P. Baldwin F. Weisgraber K.H. Lund-Katz S. Phillips M.C. J. Biol. Chem. 2001; 276: 40949-40954Google Scholar). structures of the apoA-I variants by CD at using a The proteins in mm and to for CD The α-helix content was from the at as Lund-Katz S. Phillips M.C. J. Biol. Chem. similar for the α-helix content by a of the CD R.W. Scholar). The denaturation was from the in at over the as P. Segall M.L. M. J. Weisgraber K.H. Phillips M.C. Lund-Katz S. J. Biochim. Biophys. Acta. Scholar). The the of the denaturation was by the and are the the of the unfolding are and The of at from the at and the of the and forms of the as Lund-Katz S. Phillips M.C. J. Biol. Chem. Scholar). The was from the of the by linear to the is the and is measurements with a fluorescence at In the apoA-I was from solution into mm mm 1 mm To the of fluorescence of the from to using a to denaturation by proteins at of at with at various at a with being used of in the for Lund-Katz S. Phillips M.C. J. Biol. Chem. Scholar). The of the of and which the of denaturation in the by the linear In fluorescence the of proteins at of using a solution containing 1 mm to the of After for the fluorescence to the and are the fluorescence in the absence and presence of is the of to the and is the 8-anilino-1-naphthalenesulfonic acid fluorescence from to at an of in the presence of protein and an of of of by the apoA-I variants by the in at as M.L. Dhanasekaran P. Baldwin F. Anantharamaiah G.M. Weisgraber K.H. Phillips M.C. Lund-Katz S. J. Lipid Res. Scholar). The of in with to a and the in of the of at to the for apoA-I. by measurements was by and by as (6Saito H. Dhanasekaran P. Baldwin F. Weisgraber K.H. Lund-Katz S. Phillips M.C. J. Biol. Chem. 2001; 276: 40949-40954Google Scholar). of apoA-I binding to emulsions with a isothermal titration at To that the protein to the the to protein was over The rate for the of binding from the titration to a CD of CD of apoA-I variants used to the α-helix content in the lipid-free structures with apoA-I content of and in the of in deletion mutants for The that the other a in the of that that no in conformation of the amino In the of the that of the are that this region of the apoA-I molecule forms an structure (7Davidson W.S. Hazlett T. Mantulin W.W. Jonas A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13605-13610Google Scholar, 8Roberts L.M. Ray M.J. Shih T.W. Hayden E. Reader M.M. Brouillette C.G. Biochemistry. 1997; 36: 7615-7624Google Scholar). The central region contains a of structure T. Atkinson D. Biochemistry. 41: Scholar) because of the in are the carboxyl-terminal deletion the of the larger carboxyl-terminal deletion results in an of in a This that the carboxyl-terminal region is a structure (7Davidson W.S. Hazlett T. Mantulin W.W. Jonas A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13605-13610Google Scholar, 8Roberts L.M. Ray M.J. Shih T.W. Hayden E. Reader M.M. Brouillette C.G. Biochemistry. 1997; 36: 7615-7624Google Scholar) in which there is in M.A. Jonas A. Biochemistry. 2000; 39: Scholar, O. Atkinson D. Biochemistry. 2000; 39: content and of apoA-I deletion mutants by and from at six of in show in the of to in is for it is as is to to from at six in show in the of to The in is for it is as is to apoA-I. in a 1 the unfolding of and apoA-I by the at The of the that the deletion the of whereas the deletion the with apoA-I. the of the the and the of of Δ44–65, or a in that of these regions the protein structure. of the and that of the amino-terminal or central regions in a in the of whereas the carboxyl-terminal deletion the 1 the unfolding of apoE4 and 22-kDa fragment for comparison (11Morrow J.A. Segall M.L. Lund-Katz S. Phillips M.C. Knapp M. Rupp B. Weisgraber K.H. Biochemistry. 2000; 39: 11657-11666Google Scholar). of the carboxyl-terminal domain in apoE in an in the α-helix content and an in the similar to the of deletion of in apoA-I. of the structure and stability of the amino-terminal region the apoA-I variants, we fluorescence because the are in the amino-terminal of apoA-I and Because of the in similar in a folded protein structure W.S. A. J. Hazlett Jonas A. Biochemistry. 1999; 38: Scholar), fluorescence is to the conformation of the amino-terminal of the deletion mutants are denaturation of the apoA-I variants by of fluorescence of The the of and from the linear of the are in and apoA-I a of denaturation of consistent with results using the at (12Reijngoud D.J. Phillips M.C. Biochemistry. 1982; 21: 2969-2976Google Scholar, Lund-Katz S. Phillips M.C. J. Biol. Chem. Scholar). of the amino-terminal (Δ1–43, Δ44–65, or or central regions lead to a in and that of these regions the stability and the unfolding of the amino-terminal of apoA-I. In of these for and are to that carboxyl-terminal deletion has no effect on the tertiary structure of the amino-terminal of the of and binding of apoA-I deletion mutants from fluorescence at of by of fluorescence of to fluorescence are to apoA-I. is to to to to to to to to to to to to to to of fluorescence of to are to apoA-I. is to apoA-I. in a To the of to the in similar of the for Δ44–65, and apoA-I similar of to in these in for and an in for with apoA-I that in are less to whereas in is more The fluorescence of is known to to an surface or a of a protein L. J. Biol. Scholar). with ANS, a in fluorescence was in the presence of in contrast to the of a folded globular protein R.S. Kay C.M. Ryan R.O. Biochemistry. 1999; 38: 4327-4334Google Scholar, D.P. Brouillette C.G. Engler J.A. Roberts L. Anantharamaiah G.M. Lund-Katz S. Phillips M.C. Ray M.J. Biochemistry. 1997; 36: Scholar). in of the amino-terminal or central regions a in fluorescence with that there is more surface to in these In a in fluorescence for and that carboxyl-terminal deletion the surface in apoA-I. similar effect of carboxyl-terminal deletion on fluorescence was in apoE4 these results the of in the amino-terminal or the carboxyl terminus of the apoA-I structure that the lipid-free apoA-I molecule is organized into two structural domains similar to The amino-terminal and central parts form a helix bundle, whereas the carboxyl-terminal α-helices form a separate, less organized structure (7Davidson W.S. Hazlett T. Mantulin W.W. Jonas A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13605-13610Google Scholar, 8Roberts L.M. Ray M.J. Shih T.W. Hayden E. Reader M.M. Brouillette C.G. Biochemistry. 1997; 36: 7615-7624Google Scholar, D.P. Roberts L.M. Lebowitz J. Engler J.A. Brouillette C.G. Biochemistry. 1998; 37: 945-955Google Scholar, D.P. Brouillette C.G. Engler J.A. Roberts L. Anantharamaiah G.M. Lund-Katz S. Phillips M.C. Ray M.J. Biochemistry. 1997; 36: Scholar). Lipid of used a to the of the two structural domains in apoA-I to lipid in and of that the carboxyl-terminal domain of apoA-I a key role in lipid binding (7Davidson W.S. Hazlett T. Mantulin W.W. Jonas A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13605-13610Google Scholar, Jonas A. J. Biol. Chem. Scholar, P. B. E. A. J. E. E. M. D. Biochemistry. Scholar), the carboxyl-terminal domain of apoE M.L. Dhanasekaran P. Baldwin F. Anantharamaiah G.M. Weisgraber K.H. Phillips M.C. Lund-Katz S. J. Lipid Res. Scholar). of the helix bundle by the amino-terminal or central because of more of these proteins M.L. Dhanasekaran P. Baldwin F. Anantharamaiah G.M. Weisgraber K.H. Phillips M.C. Lund-Katz S. J. Lipid Res. rate and of binding of apoA-I deletion are to the rate for apoA-I is of binding from are to is for it is The rate for apoA-I is to apoA-I. to to apoA-I. to apoA-I. to to apoA-I. to apoA-I. to to apoA-I. to are to the rate for apoA-I is are to is for it is The rate for apoA-I is to to apoA-I. in a To further the of the two domains on lipid we the of binding of the apoA-I variants using (6Saito H. Dhanasekaran P. Baldwin F. Weisgraber K.H. Lund-Katz S. Phillips M.C. J. Biol. Chem. 2001; 276: 40949-40954Google Scholar). Because there is no in apoA-I binding, the heat on apoA-I binding to emulsions contains from the interaction and from the of protein A. D. Biophys. J. 1996; Scholar). The binding at a low surface of protein are in of the amino-terminal or central regions have on the binding for the in which the tertiary structure of the carboxyl-terminal domain is to be (10Borhani D.W. Rogers D.P. Engler J.A. Brouillette C.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 12291-12296Google Scholar, D.P. Brouillette C.G. Engler J.A. Roberts L. Anantharamaiah G.M. Lund-Katz S. Phillips M.C. Ray M.J. Biochemistry. 1997; 36: Scholar). of the carboxyl-terminal domain in a in the binding further the critical role of the carboxyl-terminal domain in lipid In to the binding we the of release of heat because in the of isothermal titration the apoA-I variants in this rate is by of the amino-terminal or central whereas the carboxyl-terminal have a effect on we an by fluorescence and rate that the stability of the amino-terminal helix bundle structure the of apoA-I on the surface. in of denaturation of the apoA-I variants that of the amino-terminal or central regions lead to a more and structure, whereas the carboxyl-terminal deletion mutants similar stability and unfolding to apoA-I. This that the amino-terminal of the molecule has a folded domain structure, independently from the carboxyl terminus. In this an in fluorescence by of the amino-terminal or central regions to from the of this domain structure. However, fluorescence show that the in the amino-terminal of apoA-I are from and of the in the or carboxyl-terminal regions a in to this that apoA-I variants have of folded tertiary structure. The on the and the by the or carboxyl-terminal in the presence of two structural domains in apoA-I. The carboxyl-terminal domain to be relatively because deletion more cooperative unfolding with In in the amino-terminal domain the and of consistent with of a helix bundle. The in may be central to the bundle because their deletion the T. Atkinson D. Biochemistry. 41: Scholar). these data are consistent with the domain model of lipid-free apoA-I in which the amino-terminal of the domain is organized into a helix bundle, whereas the carboxyl-terminal domain as a less organized structure (7Davidson W.S. Hazlett T. Mantulin W.W. Jonas A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13605-13610Google Scholar, 8Roberts L.M. Ray M.J. Shih T.W. Hayden E. Reader M.M. Brouillette C.G. Biochemistry. 1997; 36: 7615-7624Google Scholar). The concept of two structural domains is established for apoE (5Weisgraber K.H. Adv. Protein Chem. 1994; 45: 249-302Google Scholar), and a helix bundle is consistent with the crystal structure of the amino-terminal fragment of apoE (3Wilson C. Wardell M.R. Weisgraber K.H. Mahley R.W. Agard D.A. Science. 1991; 252: 1817-1822Google Scholar). In this the tertiary structure of apoA-I is similar to apoE that the amino-terminal helix bundle is less organized and less that in Indeed, denaturation of apoE that the 22-kDa amino-terminal and 10-kDa carboxyl-terminal domain independently (11Morrow J.A. Segall M.L. Lund-Katz S. Phillips M.C. Knapp M. Rupp B. Weisgraber K.H. Biochemistry. 2000; 39: 11657-11666Google Scholar), whereas denaturation of apoA-I is monophasic (12Reijngoud D.J. Phillips M.C. Biochemistry. 1982; 21: 2969-2976Google Scholar). The that the amino-terminal helix bundle domains in apoA-I and different of amphipathic helix in which the are class in apoE and class in apoA-I J.P. H. Brouillette C.G. Anantharamaiah G.M. J. Lipid Res. Scholar) may the in Also, the proline is more in apoA-I in apoE, the in the bundle from as organized as are in apoE (3Wilson C. Wardell M.R. Weisgraber K.H. Mahley R.W. Agard D.A. Science. 1991; 252: 1817-1822Google Scholar, K.H. Adv. Protein Chem. 1994; 45: 249-302Google Scholar). In addition, the of on helix bundle stability could be a J. Ryan R.O. Proc. Natl. Acad. Sci. U. S. A. Scholar). and are in the same gene family (1Li W.H. Tanimura M. Luo C.C. Datta S. Chan L. J. Lipid Res. 1988; 29: 245-271Google Scholar) and exon codes for the amphipathic repeats in both in apoA-I and in apoE J.P. H. Brouillette C.G. Anantharamaiah G.M. J. Lipid Res. Scholar). The carboxyl-terminal domain is critical for lipid binding in both molecules in apoE, this to the 10-kDa fragment (5Weisgraber K.H. Adv. Protein Chem. 1994; 45: 249-302Google Scholar), which of the repeats (1Li W.H. Tanimura M. Luo C.C. Datta S. Chan L. J. Lipid Res. 1988; 29: 245-271Google Scholar). from the that the carboxyl-terminal repeats the lipid binding domain in this domain to (1Li W.H. Tanimura M. Luo C.C. Datta S. Chan L. J. Lipid Res. 1988; 29: 245-271Google Scholar), consistent with on the In proteolysis of lipid-free apoA-I cleavage sites in the carboxyl L.M. Ray M.J. Shih T.W. Hayden E. Reader M.M. Brouillette C.G. Biochemistry. 1997; 36: 7615-7624Google Scholar, Jonas A. J. Biol. Chem. Scholar). Lipid of and to and binding of the apoA-I variants that the carboxyl-terminal domain a key role in the lipid binding of consistent with (7Davidson W.S. Hazlett T. Mantulin W.W. Jonas A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 13605-13610Google Scholar, Jonas A. J. Biol. Chem. Scholar, P. B. E. A. J. E. E. M. D. Biochemistry. Scholar). the other the of the rate by of the amino-terminal or central regions that the stability of the amino-terminal domain is critical to the lipid interaction of apoA-I. this a the unfolding stability of the amino-terminal domain, and the rate of release of heat on binding is These results that the tertiary structure of the amino-terminal domain the lipid binding of the helix of apoA-I. an the stability of the apoE amino-terminal domain and to was M.L. Dhanasekaran P. Baldwin F. Anantharamaiah G.M. Weisgraber K.H. Phillips M.C. Lund-Katz S. J. Lipid Res. Scholar). it has been that a molten state of apolipophorin III Biochemistry. 1998; 37: Scholar) and apoE V. Ryan R.O. J. 2001; Scholar) is with lipid binding activity of these Based on the observations, we propose a two-step mechanism for lipid binding of apoA-I on In this apoA-I initially binds to a lipid surface through amphipathic α-helices in the carboxyl-terminal domain, by an in probably in the region the helix bundle in the amino-terminal domain a to In this the extra in the amino-terminal domain to from T. Atkinson D. Biochemistry. 41: Scholar). mechanism for apoA-I lipid binding to form high density lipoprotein has been by Rogers D.P. Roberts L.M. Lebowitz J. Datta G. Anantharamaiah G.M. Engler J.A. Brouillette C.G. Biochemistry. 1998; 37: Scholar) on structural of lipid-free that the protein both globular bundle and In the we not have information on the of the two domains in the lipid-free apoA-I molecule, we evidence for the different roles in lipid binding of the two domains in apoA-I: an lipid binding through the carboxyl-terminal domain followed by a reorganization of the amino-terminal helix bundle. Using ITC, we showed (6Saito H. Dhanasekaran P. Baldwin F. Weisgraber K.H. Lund-Katz S. Phillips M.C. J. Biol. Chem. 2001; 276: 40949-40954Google Scholar) that two domains in apoE4 lead to two lipid-bound in which the amino-terminal bundle or We that the binding of apoA-I in a was that in a low surface in that apoA-I multiple lipid-bound However, it be that results a conformation in which the helix bundle in apoA-I intact of with lipid from in which from the surface a J.P. H. Brouillette C.G. Anantharamaiah G.M. J. Lipid Res. both are because the helix bundle of apoA-I is more the bundle of Given the structural of apoA-I and other exchangeable apolipoproteins such as and that are of the apolipoprotein multigene family (1Li W.H. Tanimura M. Luo C.C. Datta S. Chan L. J. Lipid Res. 1988; 29: 245-271Google Scholar), the two-step lipid binding mechanism could be a general feature for lipid interaction of exchangeable apolipoproteins of two structural domains. be that this model is not to to the exchangeable apolipoproteins such as that not multiple 22-residue In there is no evidence for two domains as J. Biol. Chem. Scholar), because the helix bundle in is not more the carboxyl-terminal as in apoA-I in this and in contrast to apoE (11Morrow J.A. Segall M.L. Lund-Katz S. Phillips M.C. Knapp M. Rupp B. Weisgraber K.H. Biochemistry. 2000; 39: 11657-11666Google Scholar). In addition, the concept of multiple lipid-bound of apoA-I may into apoA-I such as binding to receptor class in which the of apoA-I to variable receptor binding affinity M.C. L. Jonas A. D.R. J. Lipid Res. 2001; Scholar). We for Protein for with measurements and Baldwin for

References

34