Material and Methods

WES
We used standard techniques6 to collect blood samples of different family members (Figure 1A) and isolate DNA from blood or saliva. All subjects or their legal representatives provided written informed consent for this study, which was performed in accordance with the Declaration of Helsinki protocols and approved by the local institutional review boards.
WES was undertaken for one or more individuals from families 1–5, 7, and 8. DNA enrichment for WES was achieved with the Nextera Rapid Capture Exome 38 Mb Kit (Illumina) for family 1 and with the SureSelect Human All Exon Kit v.5 (Agilent) for families 2–5, 7, and 8. Paired-end sequencing (100 bp) was run on an Illumina HiSeq 1500 or 2500. A minimum of 4.5 Gb of sequence was generated for each individual, yielding a mean depth of coverage ranging from 75× to 163× and 88.3%–98% of target bases sequenced at 20× coverage. The sequence data were mapped to the human reference genome (UCSC Genome Browser hg19) with the Burrows Wheeler Aligner (BWA).7 Variant calling was performed with the Genome Analysis Toolkit (GATK) v.2.4.7.8 VariantDB was used for variant annotation and filtering. Variant annotation was based on information from GATK, SNPeff,9 ANNOVAR, and Gene Ontology.10 Quality-based filtering was performed according to the following parameters: (1) mapping quality above 50, (2) quality by depth above 4.8, (3) mapping-quality rank sum between −3 and 3, and (4) Fisher-scaled strand bias smaller than 20. Next, common variants with a minor allele frequency (MAF) above 1% were filtered out on the basis of dbSNP (v.137), 1000 Genomes (April 2012), and the NHLBI Exome Sequencing Project Exome Variant Server (ESP6500). Variants present in our in-house control database, including 770 exomes, were filtered out. Only non-synonymous, frameshift, nonsense, and splice-site variants and genes containing biallelic variants were selected. The effect of the variant on protein function was predicted by PolyPhen-2, SIFT, and MutationTaster with dbNSFP11 for non-synonymous variants and with multiple tools for potential splice-site mutations. Confirmation of the putative disease-causing variants and their cosegregation with the disease phenotype and analysis of PEX6 for variants in family 64 were performed by standard Sanger dideoxy sequencing on an ABI 3130XL or ABI 3730 DNA sequencer (Applied Biosystems).

Analysis of STRs and Healthy Control Population for Family 1
For family 1, we amplified, pooled, and analyzed short tandem repeat (STR) markers via capillary electrophoresis on an ABI 3130XL DNA sequencer (Applied Biosystems). We analyzed data with GeneMapper v.3.7 (Applied Biosystems). We used NCBI Map Viewer (annotation release 105) to search for STRs in a 10 Mb region surrounding the PEX1 mutation.
We collected blood samples from umbilical cords of 250 unrelated newborns originating from different regions of Morocco. The Moroccan origin of their parents and grandparents was confirmed. We obtained informed consent for DNA analysis from the parents. We used a standard salting-out method to extract genomic DNA from 3 ml blood. We developed a real-time PCR (Applied Biosystems 7500 Fast Real-Time PCR Systems) assay by using TaqMan probes for the PEX1 c.3750G>A (p.Trp1250∗) nonsense mutation (Table 2) and validated the assay by using homozygous and heterozygous members of family 1.

Electron Microscopy
We cultured fibroblasts to confluence and fixed them with 4% formaldehyde and 2.5% glutaraldehyde in 0.1 M HEPES buffer (pH 7.2). We postfixed the cells with 1% osmium tetroxide and 1.5% potassium ferrocyanide in 0.1 M cacodylate buffer (pH 7.2) for 1 hr, then in 1% tannic acid in 0.1 M cacodylate buffer (pH 7.2) for 1 hr, and finally in 1% uranyl acetate in water for 1 hr. The samples were dehydrated in ethanol series, infiltrated with TAAB 812 resin, and polymerized for 24 hr at 60°C. Ultrathin sections were cut with a Reichert Ultracut ultramicrotome and visualized with a FEI Tecnai 12 Biotwin microscope at 100 kV accelerating voltage. Images were taken with a Gatan Orius SC1000 CCD camera.

Biochemical and Enzyme-Activity Assays
We measured peroxisomal parameters in plasma (very-long-chain fatty acids [VLCFAs], bile acid intermediates, pipecolic acid, phytanic acid, and pristanic acid),12 in erythrocytes (plasmalogens),13 and in skin fibroblasts (VLCFAs,14 C26:0 and pristanic acid β-oxidation,15 phytanic acid α-oxidation,16 and dihydroxyacetonephosphate acyltransferase [DHAPAT] activity17). Immunoblot analyses assessed the processing of thiolase and acyl-CoA oxidase I (ACOX1) in fibroblasts.16,18

Molecular Cloning
We introduced the different PEX1 or PEX6 variants identified in the individuals with HS in the mammalian expression vector pcDNA3 containing full-length PEX1 or PEX6 cDNA, respectively, by site-directed mutagenesis (QuikChange Site-Directed Mutagenesis Kit, QIAGEN) according to the manufacturer’s instructions. We obtained the pcDNA3 vector containing PEX6 c.1930C>T (p.Arg644Trp) by amplifying PEX6 cDNA spanning nucleotides c.1856 to c.∗61 (with flanking restriction sites for NheI and KpnI) from total RNA isolated from the fibroblasts of family 5 individual II:2 (F5-II:2). We subsequently subcloned the amplicons into the pcDNA3 vector containing full-length wild-type PEX6 cDNA. We confirmed successful introduction of the variants by sequence analysis of the cDNAs. To exclude unintentional mutations in the vector backbone during site-directed mutagenesis, we either used several clones per construct for further analyses or recloned the mutated cDNA into pcDNA3 plasmids.

Cell Culture and Transfection
We used primary skin fibroblasts from individuals with HS and primary skin fibroblast cell lines completely deficient of PEX1 (compound heterozygous for p.[Thr263Ilefs∗6];[Ile700Tyrfs∗42], c.[788_789del];[2097dup])19 or PEX6 (homozygous for p.Gly135Aspfs∗23 [c.402del]).20 Cells were cultured in DMEM with L-glutamine (Bio-Whittaker) supplemented with 10% fetal bovine serum (Bio-Whittaker), 25 mM HEPES buffer (BioWhittaker), 100 U/ml penicillin, 100 μg/ml streptomycin (Life Technologies), and 250 ng/ml Fungizone (Life Technologies) in a humidified atmosphere of 5% CO2 at 37°C or 40°C. Transfections were performed with the AMAXA NHDF Nucleofector Kit (Lonza) according to the manufacturer’s instructions (program U23). The medium was changed 24 hr after transfection, and the cells were imaged 72 hr after transfection.

Immunofluorescence Assays
We analyzed peroxisomal appearance in skin fibroblasts from HS individuals by immunofluorescence microscopy. The cells were cultured on glass slides to a confluency of 50%–70%. For fixation, we treated the cells with 2% paraformaldehyde (Merck) in PBS for 20 min at room temperature and permeabilized them with 0.5% Triton X-100 (BioRad) for 5 min. The peroxisomal matrix protein catalase was labeled with the monoclonal antibody α-catalase (Map 17E10, own production), biotinylated α-mouse antibodies (E 433, Dako), and streptavidine-FITC (F 422, Dako). Peroxisomal membranes were labeled with antibodies against PMP70 (ABCD3) (PMP70, no. 718300, Zymogen) and Alexa Fluor 555 goat anti-rabbit (Invitrogen). The slides were fixed on mounting medium Vectashield H1000 (Brunschwig). Images were taken with the Leica TCS SP8 filter-free spectral confocal microscope.

Assays of Genetic and Functional Complementation
We performed genetic complementation of fibroblasts by transfecting the cells from HS individuals with PEX cDNA as described in Ebberink et al.19 To test the functionality of the PEX variants, we co-transfected pcDNA3-PEX1 or -PEX6 plasmids with the peroxisomal matrix marker pEGFP-SKL21 into skin fibroblasts deficient in PEX1 or PEX6. Cells transfected with only pEGFP-SKL served as negative controls, whereas co-transfections of the marker with pcDNA3 vectors containing the respective wild-type PEX cDNA served as positive controls. We subsequently analyzed the localization of the fluorescent signal 3 days after transfection by using the fluorescence microscope Zeiss Axio Observer A1.
To evaluate the effect of the variants found in the affected individuals, we determined per transfection the percentage of cells showing a punctate GFP signal (indicating “peroxisome-positive” or “complemented” cells) of the total number of 100–200 transfected cells. These ratios were normalized to the complementation efficiency of the positive control (set as 100%) and averaged per construct (n = 5–7). We used the one-sample Wilcoxon signed-rank test to test the statistical significance of deviations from the positive control.