Supporting Information

Files in this Data Supplement:

Table 4. Logistic regression analysis

For the purposes of this analysis, APOE variable was coded according to the presence or absence of APOE-e4. For LRP6-18e, each variable was coded by genotype. B, regression coefficient; df, degrees of freedom; OR, odds ratio; CI, confidence interval.

Table 5. Single locus and haplotype test for SNPs 14e and 18e in the whole multicenter case-control series

SI Methods

. PCR primers (Operon-Qiagen) were designed to amplify each exon plus at least 40 bp upstream and downstream from the genomic sequence flanking each exon to yield products between 200 and 400 bp in length by using Primer 3 (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi). PCR amplification was performed with Thermo-Start DNA Polymerase mix (ABgene), 5 mM each primer, and 50 ng of human genomic DNA in a 15-ml reaction. Cycling was conducted on a Hybaid-MBS Satellite Thermal Cycler (Thermo Electron Corporation) with an initial denaturation of 94°C for 15 min, followed by 35 cycles of 94°C for 30 s, appropriate annealing temperature for 30 s and 68°C for 45 s, with a final extension step of 68°C for 10 min. Sized PCR products were confirmed by electrophoresis on 2% agarose gels. PCR purification was achieved by using the Montage PCR96 Cleanup kit (Millipore). Purified products were then bidirectionally sequenced on an ABI3700 DNA Sequencer (Applied Biosystems) with the Big Dye Terminator (v3.1) Cycle Sequencing kit (Applied Biosystems). Sequence traces were analyzed on Sequencher 4.0.5 (Gene Codes Corporation) to characterize polymorphisms.

. We genotyped nonsynonymous coding sequence SNPs: in exon 7 (7e; Ile-483 → Val, rs7975614); in exon 11 (11e; Cys-817 → Ser, rs2302686); and in exon 14 (14e; Ile-1062 → Val, rs2302685), a synonymous coding sequence SNP in exon 18 (18e; Cys-1270 → Cys, rs1012672), a SNP within the boundaries of a putative splicing site in the 5' UTR of exon 16 (16e, rs2075241) and five noncoding SNPs distributed throughout the LRP6 gene (rs7308022, rs2417086, rs7294695, rs2284396, and rs7316466). To ensure consistent and high-quality genotyping data, negative control samples were included in every genotyping experiment. In addition, near 5% of the samples were genotyped in duplicate, and the genotypes of the duplicate samples were checked for consistency. Primer design and PCR cycling conditions are available upon request.

Plasmid pCDNA3.1 containing human LRP6 corresponding to allele 14e-Ile (determined by sequencing) was a gift from J. F. Hess (Merck Research Laboratories, West Point, PA). The COOH-terminal hemagglutinin (HA)-tagged hLRP6 (hLRP6-HA), was generated by ligating hybridized oligonucleotides to the 3' terminus BsrGI site in hLRP6 and the XbaI site in the pcDNA3.1 polylinker. The hybridized oligonucleotides contained a BsrGI overhang, followed by the remaining 3' hLRP6 sequence (minus stop codon), an HA epitope, a stop codon, and an XbaI overhang. Mutant hLRP6 (14e-Val)-HA was generated by replacement of an XhoI-XhoI restriction fragment in which the corresponding mutation was engineered by overlap extension PCR-based mutagenesis. All constructs were verified through sequencing.

. HEK293T cells, HT22 cells, and mouse L cells or L cells (L-Wnt3a) that stably express Wnt3a (a gift from R. Nusse, Stanford University, Stanford, CA) were maintained in DMEM, 10% FBS (Invitrogen,) and 100 units/ml pen/strep (Invitrogen). HEK293T/STF and HT22/STF stably expressing the b-catenin-responsive superTOPFLASH luciferase reporter were generated by Lentiviral transduction of HEK293T or HT22 cells respectively. HEK293T/STF cells or HT22/STF were transiently transfected with the appropriate constructs by using Lipofectamine Plus (Invitrogen) for Western blot and luciferase reporter assays. Wnt3a conditioned media was prepared as described [Willert K, Brown JD, Danenberg E, Duncan AW, Weissman IL, Reya T, Yates JU, III, Nusse R (2003) Nature 423:448-452].

. HEK293T/sTF cells seeded in six-well plates were transfected with the corresponding amounts of each C-terminal HA-tagged hLRP6 allele and 10 ng of the Renilla luciferase construct, pRLTK (Promega), for normalization. Total DNA in all transfections was brought to 2 mg by using pCS2+ vector DNA. Forty-eight hours after transfection, cells were washed three times with PBS, and the cell surface was biotinylated with 30 mg/ml noncleavable sulfo-NHS-biotin in PBS (Pierce Chemical.) for 30 min at room temperature. Unreacted biotin was quenched with three washes of TBS containing 10% glycine. Cells were then lysed with 300 ml of passive lysis buffer (Promega) containing 1× complete protease inhibitor mixture (Roche) for 20 min at room temperature. Luciferase reporter activity was measured in duplicate by using 5 ml of lysate in accordance with the dual luciferase assay specifications (Promega). Thirty microliters of 10× RIPA buffer was added to the remaining lysate for solubilization of membrane proteins. Protein concentration in all samples was determined by Bradford assay (Bio-Rad). Total cellular hLRP6-HA in each sample was determined by SDS/PAGE separation of 10 mg of protein from the whole-cell lysate, followed by Western blot analysis probing with an anti-HA primary antibody (H6908; Sigma). Antibodies were stripped, and blots were reprobed with an anti-b-tubulin primary antibody (Sigma) for normalization. From the remaining lysate, 250 mg of total protein from each sample was subjected to an HA immunoprecipitation by using 5 mg/ml anti-HA antibody (Sigma) and 30 ml of protein G-Sepharose beads (Sigma). Immunoprecipitates were washed five times with 1× RIPA and eluted with 30 ml of 100 mg/ml HA peptide (Roche). Eluates were separated by SDS/PAGE, transferred to nitrocellulose and blotted with anti-biotin monoclonal antibody (Sigma). Antibodies were stripped, and blots were reprobed with anti-HA antibody (Sigma).

HT22 cells seeded in 24-well plates were transfected with the corresponding amounts of each C-terminal HA-tagged hLRP6 allele and 0.25 ng of phRluc (BioSignal Packard) for normalization. Forty-eight hours after transfection, cells were then lysed with 100 ml of passive lysis buffer (Promega) containing 1× complete protease inhibitor mixture (Roche) for 20 min at room temperature. Luciferase reporter activity was measured in duplicate by using 5 ml of lysate as described and in accordance with the dual luciferase assay specifications (Promega). Ten microliters of 10× RIPA was added to the remaining 90 ml of lysate, which was then subjected to SDS/PAGE, transferred to nitrocellulose, and blotted with anti-HA antibody (Sigma).

. An equal volume of 2× sample buffer containing 8% b-mercaptoethanol was added to each sample. Samples were heated for 5 min at 95°C, separated on a 4-12% gradient SDS/PAGE gel (Criterion; Bio-Rad), and transferred to a nitrocellulose membrane (TransBlot transfer media; BioRad). The membrane was blocked with 5% dry milk in TBS/0.1% Tween 20. Blots were probed with primary antibody, washed with TBS/0.1% Tween 20, probed with an HRP-conjugated goat anti-primary species secondary antibody, and developed by ECL (Pierce) according to the manufacturer's specifications. Antibody stripping was achieved by incubating blots in TBS/2% SDS containing 7 ml/ml 2-mercaptoethanol for 30 min, followed by six 5-min washes with TBST.

. Normal human hippocampus polyA-mRNA was purchased from BD Biosciences. cDNA was synthesized with oligodT primers from 500 ng of mRNA by using SuperscriptIII (Invitrogen). Real-time PCR was performed by using the Roche Light Cycler 1.2 and software version 3.5 (Roche Diagnostics). Briefly, intron-spanning primers (LRP6 forward 5'-CAT GTG ATT GGC TTG GAG AA-3', reverse 5'-CGA CTT GAA CCA TCC ATT CC-3'; LRP5 forward 5'-ACTCAGAGACCAACCGCATC-3', reverse 5'-TCC ATT GGG CCA GTA AAT GT-3'; b-actin forward 5'-AGA GCA AGA GAG GCA TCC TC-3', reverse 5'-CTC AAA CAT GAT CTG GGT CA-3') were designed to amplify ~200-bp products to minimize contamination from genomic DNA. Quantitative PCR was performed in triplicate by using the FastStart SyBr green DNA MasterMix Plus, as directed by the manufacturer (Roche Diagnostics). Primers were annealed at 58°C. We determined absolute transcript copy number by comparing the gene amplification in hippocampus cDNA to the amplification of standard samples containing 102-108 copies of the gene. The absolute copy number of LRP6 and LRP5 was calculated by dividing the average copy number of triplicate quantitations to the average copy number of a-actin. The error bars represent the standard error of the ratios, as calculated by using a confidence interval.