Subjects and Methods

Subjects
The clinical studies were approved by Oxfordshire Research Ethics Committee B (reference C02.143), London Riverside Research Ethics Committee (reference 09/H0706/20), and the Medical Ethical Committee of the Erasmus University Medical Center Rotterdam (MEC-2012-140 and MEC-2013-547). Written informed consent to obtain samples for genetics research was obtained from each child’s parent or guardian. Venous blood was used for DNA extraction and fibroblast cultures were established from skin biopsies taken from scalp incisions during surgical intervention. Intracranial pressures in subject 1 were documented by 24–48 hr direct recording with an intraparenchymal Codman Microsensor.21 The screening panel comprised samples from 307 individuals with syndromic or non-syndromic craniosynostosis. All DNA samples were previously tested for mutation hotspots in FGFR2, FGFR3, TWIST1, and TCF12.3,4 Significant chromosome aneuploidy in individuals with ZIC1 mutations was excluded by karyotyping and/or array comparative genomic hybridization. Where necessary, correct biological relationships were confirmed by segregation analysis of a panel of 13 microsatellites (D1S2868, D3S1311, D4S403, D5S2027, D6S1610, D7S519, D9S158, D10S548, D11S898, D13S1265, D14S280, D16S415, and D18S474).

Whole Genome/Exome Sequencing and Mutation Screening of ZIC1
Whole genome sequencing (WGS) of the male proband subject 1 and his parents was performed as part of the WGS500 clinical genome sequencing initiative.22 In brief, 3–5 μg DNA was used to prepare libraries for 100 bp paired-end sequencing to generate a mean coverage of 30× using the Illumina HiSeq2000 platform. Sequence reads were mapped to the human reference GRCh37d5 using Stampy (v1.0.12–1.0.22) and variants called with Platypus (v.0.2.4).23,24 To identify de novo mutations, we prioritized variants within coding regions that were called as absent in both parents and in dbSNP135, generating a list of 203 variants in 177 genes, of which 39 were classified as protein altering. Visualization of the trio read alignments revealed a single bona fide change in ZIC1 (12 of 27 reads), which was absent in both the paternal (19 reads) and maternal (31 reads) samples; the other 38 variants were either in fact present in one of the parents or were artifactual (Table S1). Possible recessive inheritance was analyzed with an in-house perl script to list homozygous, compound heterozygous, and hemizygous X chromosomal variants in subject 1, with a frequency cut-off of 0.003 in either 1000 Genomes or Exome Variant Server; variants in two genes fitted the criteria (Table S1). Whole genome sequencing of genomic DNA from four subjects in family 5 (affected: 5:II.2, 5:III.3, 5:III.6; unaffected: 5:II.3) was performed by BGI Complete Genomics.25,26 Filtering based on a list of genes mutated in craniosynostosis identified a predicted missense substitution encoded by ZIC1, present only in the three affected individuals. Exome sequencing of subject 3 was performed on genomic DNA (extracted from whole blood) using an Agilent SureSelect Human All Exon Kit (v.5; 50 Mb) on the Illumina HiSeq2000 platform. Reads were mapped to hg19 with Novoalign (Novocraft Technologies) and variants called with SAMtools and annotated by ANNOVAR.
To investigate further the significance of the ZIC1 mutations, primers were designed for amplification of genomic DNA (GenBank: NT_005612.17) and cDNA (GenBank: NM_003412.3), for multiplex ligation-dependent probe amplification (MLPA) analysis, and for deep sequencing (Table S2, which provides details of experimental conditions). Variant screening of all three exons of ZIC1 was performed by dideoxy sequencing on PCR amplification products from genomic DNA by BigDye Terminator v3.1 (Applied Biosystems). Copy-number variation was analyzed by MLPA using probes to each exon, according to the manufacturer’s instructions (MRC Holland). RNA was extracted from fibroblasts (Trizol, Invitrogen), cDNA synthesized with RevertAid first strand cDNA kit (Thermo Scientific), and the samples analyzed by agarose gel electrophoresis after digestion with BfaI. To quantify the proportions of wild-type to mutant allele in cDNA, an amplification product spanning exons 2–3 was used as a template for PCR to add Ion Torrent P1 and A adapters, and the resulting product was purified with AMPure beads (Beckman Coulter). Emulsion PCR and enrichment were performed with the Ion PGM Template OT2 200 Kit (Life Technologies) according to the manufacturer’s instructions and sequencing of enriched templates performed on the Ion Torrent PGM (Life Technologies) for 125 cycles with the Ion PGM Sequencing 200 kit v2. Data were processed with Ion Torrent platform-specific pipeline software v.4.2.1.

Xenopus Assays
All experiments using Xenopus were approved by the Institutional Animal Care and Use Committee of Montana State University. Xenopus full-length zic127 and zic1ΔC28 cDNA constructs were described previously (zic1ΔC was originally termed oplΔC). The zic1ΔC2 construct was made by PCR amplification of the portion of zic1 cDNA encoding the N terminus and zinc finger domains, including four amino acids of the C-terminal region, followed by cloning into the EcoR1 and Xba1 sites of pCS2+ATG.27 The human ZIC1 cDNA pCR4-Topo-ZIC1 (ThermoFisher) was subcloned into pcDNA3 and six different nucleotide substitutions—c.895G>T (p.Glu299∗), c.1163C>A (p.Ser388∗), c.1198G>C (p.Gly400Arg), c.1204G>T (p.Glu402∗), c.1240A>G (p.Thr414Ala), and c.1309_1310GC>TA (p.Ala437∗)—were introduced by PCR mutagenesis using the primer sequences and experimental conditions provided in Table S2. The human ZIC1 constructs were subsequently digested with EcoR1 and Xba1 and ligated into the pCS2+ plasmid. Capped sense RNAs for microinjection were synthesized from the Xenopus and human pCS2+ constructs by SP6 transcription of NarI linearized plasmids. Xenopus laevis eggs were collected and fertilized as previously described28 and embryos were staged according to Nieuwkoop and Faber.29 Embryos at the two-cell stage were injected into a single cell with 200 pg sense RNA synthesized from cDNA constructs, together with 25 pg lacZ RNA as tracer. After β-galactosidase staining,30 wild-type embryos were bleached by exposing the embryos to fluorescent light in hybridization buffer containing 1% H2O2. Expression of en-2 was determined in neurula stage 15–17 albino and wild-type embryos by in situ hybridization31 with digoxygenin-labeled antisense RNA en-2 probe as described.32 An anti-digoxygenin alkaline phosphatase-conjugated antibody (Roche) and the alkaline phosphatase substrate NBT/BCIP (Fisher Scientific) were used for color detection. Embryos were scored double-blind to determine changes in en-2 expression in comparison to the uninjected side. Results using wild-type and mutant constructs were compared by Fisher’s exact tests with Bonferroni correction for multiple comparisons (n = 9).

RNA In Situ Hybridization of Mouse Embryos
Experimental procedures were performed in accordance with UK Animals (Scientific Procedures) Act, 1986 (PPL 70/7194). For whole-mount embryo in situ hybridization, embryos were dissected, fixed overnight in 4% paraformaldehyde in phosphate-buffered saline, and dehydrated through graded methanol solutions. Non-radioactive RNA in situ hybridization was performed as described33 before vibratome sectioning. RNA probes for Zic134 and En135 were digoxygenin labeled with the In Vitro Transcription kit (Roche Applied Science) followed by anti-digoxygenin-AP antibody (1:1,000) (Roche Applied Science) and NBT/BCIP (Sigma) staining to detect the hybridization signals.