Human LDLR is a chain of 839 amino acids organized into several modular domains. The "ligand-binding domain," crucial to binding LDL, consists of seven repeated amino-acid sequences (repeats R1 to R7). Next is a section referred to as the "epidermal growth factor (EGF) precursor homology domain," because it is analogous to a precursor protein that stimulates cell growth. The EGF precursor homology domain in LDLR includes the EGF-like domains A, B, and C as well as a distinctive protein structure called a "b propeller.

Model of LDLR "b propeller" as described in T. A. Springer, J. Mol. Biol. 283, 837 (1998), and further verified by H. Jeon et al., Nat. Struc. Biol. 8, 499 (2001). As described by Rudenko et al., the propeller plays a key role in displacing LDL and promoting its release within cells. Illustration rendered by PyMOL (www.pymol.org).

The remainder of the protein contains a highly sugar-linked region, a membrane-spanning region, and a cytoplasmic domain. In this study, the researchers focused on a fragment of LDLR that contains the ligand-binding and EGF precursor homology domains. They wanted to determine the organization of and interactions between the domains and shed light on the mechanism by which LDLR releases its ligand when the pH changes from 7.5 (extracellular) to less than 6 (in endosomes inside the cell, where the LDL is released).

"Have Your Steak and Live to Enjoy It Too"

Most people these days are aware that a high blood cholesterol level can clog arteries and increase the risks of stroke and heart attack. For most people, a healthy, low-animal-fat diet along with moderate exercise goes a long way toward reducing these risks. However, about one in 500 people are affected by familial hypercholesterolemia (FH), an inherited genetic condition marked by elevated LDL cholesterol levels beginning at birth and resulting in heart attacks at an early age. In men with FH, heart attacks typcally occur at 40 to 50 years of age, and 85% experience a heart attack by age 60. For women with FH, there is a 10-year delay in onset but a similar incidence rate. It is known that FH is caused by genetic mutations that disrupt the body's ability to metabolize cholesterol, but how exactly do these mutations interfere with the mechanics of the process? The mechanism proposed by Rudenko et al. adds one more piece to the puzzle and brings us one step closer to new treatments and drug discovery. As Brown and Goldstein, two of this work's Nobel-laureate authors, wrote, "it may one day be possible for many people to have their steak and live to enjoy it too."

Capture and release of LDL. Spherical LDL particles attach to LDLR anchored to the cell membrane. The cell membrane folds inwards and pinches off into a cavity within the cell (vesicle). Fusion of several vesicles gives rise to an endosome, an acidic "compartment" where the LDL particle is released. The LDLR is then recycled to the cell surface. Reprinted with permission from T. Innerarity, Science 298, 2337 (2002). Illustration: K. Sutliff. © 2002 AAAS. http://www.sciencemag.org/

Crystals of the human LDLR fragment were grown at pH 5.3. They diffracted x rays very weakly with a best resolution of 7.5 Å, obtained with intense sources such as ALS Beamlines 5.0.2 and 8.2.1 and APS Beamline 19-ID. Tungsten clusters soaked into the crystals of a mutant LDLR dramatically increased the resolution to 3.7 Å and also provided anomalous scatterers for the multiwavelength anomalous diffraction experiments that were used to solve the phase problem. Using known high-resolution structures of smaller fragments of the LDLR, electron density maps could be interpreted. The resulting model of human LDLR at pH 5.3 shows that, while repeat R1 is disordered in the crystal, repeats R2 to R7 are arranged in an arch covering one side of the EGF precursor homology domain; in this domain, modules A, B, C, and the b propeller form an apparently rigid entity. Repeats R4 and R5 interact extensively with the b propeller; this interaction would preclude the binding of a ligand such as LDL.

Model of human LDLR at 3.7 Å from data obtained at ALS Beamlines 5.0.2 and 8.2.1 and APS Beamline 19-ID. Repeats R4 and R5 (critical for LDL binding) interact extensively with the b propeller.

The structure offers a plausible hypothesis for the mechanism of ligand release upon a change in pH. While at neutral pH, the LDLR probably adopts a flexible, extended conformation and can bind ligands; at low pH it develops a binding site for the central part of its own ligand-binding domain. This new binding site can compete with the ligand, which is then released. This hypothesis can explain why LDLR mutants that lack the b propeller can bind, but not release ligands. The model can also serve as a basis for explaining the effects of many mutations in LDLR that cause familial hypercholesterolemia, one of the most common single human gene disorders.

Research conducted by G. Rudenko, K. Ichtchenko, M.S. Brown, J.L. Goldstein (University of Texas Southwestern Medical Center); K. Henderson (Berkeley Lab); and L. Henry and J. Deisenhofer (University of Texas Southwestern Medical Center and Howard Hughes Medical Institute).

Research funding: Howard Hughes Medical Institute, National Institutes of Health, Perot Family Foundation. Operation of the ALS is supported by the U.S. Department of Energy, Office of Basic Energy Sciences (BES).

Publication about this research: G. Rudenko, L. Henry, K. Henderson, K. Ichtchenko, M.S. Brown, J.L. Goldstein, and J. Deisenhofer, "Structure of the LDL Receptor Extracellular Domain at Endosomal pH," Science 298, 2353 (2002).