18. J. Jonsson, L. Carlsson, T. Edlund, H. Edlund, Nature 371, 606 (1994). 19. M. F. Offield et al., Development 122, 983 (1996). 20. J. Nichols et al., Cell 95, 379 (1998). 21. M. Schuldiner, O. Yanuka, J. Itskovitz-Eldor, D. A. Melton, N. Benvenisty, Proc. Natl. Acad. Sci. U.S.A. 97, 11307 (2000). 22. K. Gerrish et al., J. Biol. Chem. 275, 3485 (2000). 23. D. A. Stoffers, R. S. Heller, C. P. Miller, J. F. Habener, Endocrinology 140, 5374 (1999). 24. Cells were fixed in 4% paraformaldehyde/0.15% picric acid in phosphate-buffered saline (PBS). Immunocyto- chemistry was carried out with the use of standard protocols. The following primary antibodies were used at following dilutions: nestin rabbit polyclonal 1: 500 (made in our laboratory), TUJ1 mouse monoclonal 1: 500 and TUJ1 rabbit polyclonal 1: 2000 (both from Babco, Richmond, CA), insulin mouse monoclonal 1:1000 (Sigma, St. Louis, MO), insulin guinea pig poly- clonal 1:100 (DAKO, Carpinteria, CA), glucagon rabbit polyclonal 1: 75 (DAKO), somatostatin rabbit polyclonal 1:100 (DiaSorin, Stillwater, MN), GFP 1: 750 polyclonal (Molecular Probes, Eugene, OR), and BRDU rat mono- clonal 1:100 (Accurate, antibodies, Westbury, NY ). For detection of primary antibodies, fluorescently labeled secondary antibodies ( Jackson Immunoresearch Labo- ratories, West Grove, PA, and Molecular Probes) were used according to methods recommended by the man- ufacturers. Histochemical staining for alkaline phospha- tase cells was carried out using commercially available kit (Sigma) following manufacturer’s recommendations. 25. A. Ferreira, A. Caceres, J. Neurosci. Res. 32, 516 (1992). 26. W. W. Kao et al., Ophthalmol. Vis. Sci. 37, 2572 (1996). 27. K. Hadjantonakis, M. Gertsenstein, M. Ikawa, M. Ok- abe, A. Nagy, Mech. Dev. 76, 79 (1998). 28. For counting of insulin-positive cells, stage 5 ES cell cultures 6 days after bFGF withdrawal were dislodged from tissue culture surface in Ca 21 -free 3 mM EDTA containing Krebs-Ringer buffer after incubation for 15 min at 37°C. The clusters were dissociated into single cells, as previously described (29). Dissociated cells were allowed to adhere to poly-ornithine/laminine-coated glass coverslips for 5 hours, fixed, immunostained for insulin, and counterstained with a nuclear stain DAPI (49,69diamidino-2-phenylindole). This was followed by determination of the percentage of insulin-positive cells in the total cell population. 29. C. B. Wollheim, P. Meda, P. A. Halban, Methods En- zymol. 192, 188 (1990). 30. Stage 3 GFP-labeled B5 ES cells were co-cultured at clonal density on poly-ornithine/fibronectin treated–96- well plates [Costar 3603 (Fisher Scientific, Pittsburgh, PA) black plate with clear and thin bottom] with stage 3 wild-type E14.5 ES cells at a final concentration of 1 B5 cell per 40,000 E14.5 cells per well. B5 cells were plated at a concentration of 1 B5 cell per well using serial dilution. The seeding efficiency and the growth of the plated B5 cells were monitored by immunocytochemical staining for GFP. Two hours after seeding, GFP-positive cells were present in 65 6 5% of the wells. Counter- staining with a nuclear dye (DAPI) confirmed presence of only one cell per well. After 6 days of differentiation, single GFP-labeled clones derived from a single B5 cell were present in 14.8 6 1.7% of the wells. Cells were cultured through stages 4 and 5, as shown in Fig. 1A. On day 6 of stage 5, cells were fixed with 4% paraformal- dehyde and subjected to triple immunocytochemistry and laser confocal microscopy analysis. For immunocy- tochemistry, after blocking with a solution of 90% PBS, 10% normal goat serum, and 0.3% Triton-X100, cells were stained with antibodies against insulin [mouse immunoglobulin G1 (IgG1)], GFP (mouse IgG2a), and TUJ1 (rabbit). Cy5, FITC, and Cy3-conjugated goat anti- bodies to IgG1 mouse, IgG2a mouse and IgG rabbit, respectively, were used as secondary antibodies. 31. S. H. Orkin, Nature Med. 6, 1212 (2000). 32. The cells were labeled with BrdU (Boehringer Mann- heim, Indianapolis, IN) at final concentration 10 mm for 24 hours. After the labeling, depending on the specific experiment, the cells were either fixed immediately in 4% paraformaldehyde/0.15% picric acid, treated with 95% ethanol/5% glacial acidic acid for 15 min at room temperature, and subjected to immunocytochemistry, or they were cultured for various lengths of time and then analyzed by immunocytochemistry. 33. A. Vinik, R. Rafaeloff, G. Pittenger, L. Rosenberg, W. Duguid, Horm. Metab. Res. 29, 278 (1997). 34. Insulin secretion was measured using static incubations in Krebs-Ringer with bicarbonate buffer as follows: 120 mM NaCl, 5 mM KCl, 2.5 mM CaCl 2 , 1.1 mM MgCl 2 , 25 mM NaHCO 3 , and 0.1% bovine serum albumin at 37°C. Inhibitors of insulin secretion (nifedipine and diazoxide) were added to the incubation buffer 30 min before addition of glucose. For determination of total cellular insulin content, insulin was extracted from cells with acid ethanol (10% glacial acetic acid in absolute etha- nol) overnight at 4°C, followed by cell sonication. Total cellular and secreted insulin was assayed using insulin enzyme-linked immunosorbence assay (ELISA) kit (ALP- CO, Windham, NH). Protein concentrations were deter- mined using DC protein assay system (Bio-Rad, Her- cules, CA). 35. V. J. Csernus, T. Hammer, D. Peschke, E. Peschke, Cell. Mol. Life Sci. 54, 733 (1998). 36. N. H. McClenaghan, P. R. Flatt, J. Mol. Med. 77, 235 (1999). 37. G. Trube, P. Rorsman, T. Ohno-Shosaku, Pflugers Arch. 407, 493 (1986). 38. W. Montague, J. R. Cook, Biochem. J. 122, 115 (1971). 39. B. Ahren, S. Karlsson, S. Lindskog, Prog. Brain Res. 84, 209 (1990). 40. E. Rojas, J. Hidalgo, P. B. Carroll, M. X. Li, I. Atwater, FEBS Lett. 26, 265 (1990). 41. All animal studies were in accordance with NIH in- stitutional guidelines. Experimental diabetes was in- duced in 10- to 12-week-old male 129/sv mice (Ta- conic, Germantown, NY ) by a single intraperitoneal injection (150 mg/kg of body weight) of streptozo- tocin (Sigma) freshly dissolved in 0.1 M of citrate buffer, pH 4.5. Stable hyperglycemia (blood glucose levels 300 to 600 mg/dl) usually developed 48 to 72 hours after the streptozotocin injections. Blood glucose level was determined using Glucometer Elite XL blood glucose meter (Bayer Corp., Elkhart, IN). The animals were grafted with cells or with a buffer vehicle 24 to 48 hours after the establishment of stable hyperglycemia. Injected into each animal were 1 to 2 3 10 7 cells in the form of cluster suspension. In most experiments, day 6 stage 5 cells were used, suspended in Krebs-Ringer– bicarbonate buffer without Ca 21 , and injected subcutaneously under isofluorine anesthesia in the shoulder area through a 19-gauge hypodermic nee- dle. To prepare the cluster suspension, the cells cultured on 60-mm tissue culture dishes were carefully dis- lodged by treatment with Krebs-Ringer– bicarbonate buffer without Ca 21 and with 3 mM EDTA for 5 min at 37°C. Each experimental group consisted of five to eight animals per group. The animals were killed at different times, and the grafts were excised and fixed in 4% paraformaldehyde/0.15% picric acid in PBS. Sections of the grafts (15 to 20 mm) were analyzed by immunohistochemistry. 42. P. A. Halban, S. L. Powers, K. L. George, S. Bonner- Weir, Diabetes 36, 783 (1987). 43. B. Soria et al., Diabetes 49, 157 (2000). 44. B. E. Reubinoff, M. F. Pera, C. Y. Fong, A. Trounson, A. Bongso, Nature Biotechnol. 18, 399 (2000). 45. V. K. Ramiya et al., Nature Med. 6, 278 (2000). 46. S. Bonner-Weir et al., Proc. Natl. Acad. Sci. U.S.A. 97, 7999 (2000). 47. A. M. Shapiro et al., N. Engl. J. Med. 27, 230 (2000). 48. We thank T. Doetschman for gift of E14.1 ES cells and A. Nagy for gift of B5 ES cells. We also thank C. Smith for her help with confocal microscopy and J.-H. Kim for alkaline phosphatase analysis of ES cells. I.V. was supported by a fellowship from the PEW charitable Trust. We thank the National Parkinsons Foundation for their support of our work and C. Gerfen for many positive contributions. 8 January 2001; accepted 13 April 2001 Published online 26 April 2001; 10.1126/science.1058866 Include this information when citing this paper. Autosomal Recessive Hypercholesterolemia Caused by Mutations in a Putative LDL Receptor Adaptor Protein Christine Kim Garcia, 1 Kenneth Wilund, 1,2 Marcello Arca, 3 Giovanni Zuliani, 4 Renato Fellin, 4 Mario Maioli, 5 Sebastiano Calandra, 6 Stefano Bertolini, 7 Fausto Cossu, 8 Nick Grishin, 9 Robert Barnes, 1 Jonathan C. Cohen, 1 Helen H. Hobbs 1,2* Atherogenic low density lipoproteins are cleared from the circulation by hepatic low density lipoprotein receptors (LDLR). Two inherited forms of hypercho- lesterolemia result from loss of LDLR activity: autosomal dominant familial hypercholesterolemia (FH), caused by mutations in the LDLR gene, and auto- somal recessive hypercholesterolemia (ARH), of unknown etiology. Here we map the ARH locus to a ;1-centimorgan interval on chromosome 1p35 and identify six mutations in a gene encoding a putative adaptor protein (ARH). ARH contains a phosphotyrosine binding (PTB) domain, which in other proteins binds NPXY motifs in the cytoplasmic tails of cell-surface receptors, including the LDLR. ARH appears to have a tissue-specific role in LDLR function, as it is required in liver but not in fibroblasts. The liver is the major site of synthesis and clearance of cholesteryl ester–rich lipopro- teins. More than 70% of circulating LDL is removed from the blood via hepatic LDLR- mediated endocytosis (1). In individuals with two mutant LDLR alleles (homozygous FH), R EPORTS 18 MAY 2001 VOL 292 SCIENCE www.sciencemag.org 1394