TCTEX1D2 mutations underlie Jeune asphyxiating thoracic dystrophy with impaired retrograde intraflagellar transport

The analysis of individuals with ciliary chondrodysplasias can shed light on sensitive mechanisms controlling ciliogenesis and cell signalling that are essential to embryonic development and survival. Here we identify TCTEX1D2 mutations causing Jeune asphyxiating thoracic dystrophy with partially penetrant inheritance. Loss of TCTEX1D2 impairs retrograde intraflagellar transport (IFT) in humans and the protist Chlamydomonas, accompanied by destabilization of the retrograde IFT dynein motor. We thus define TCTEX1D2 as an integral component of the evolutionarily conserved retrograde IFT machinery. In complex with several IFT dynein light chains, it is required for correct vertebrate skeletal formation but may be functionally redundant under certain conditions.

Visualisation using Integrated Genome Viewer of the TCTEX1D2 deletion encompassing exon 1-2 and exon 2-5 of the adjacent gene TM4SF19, exons numbered below. Input BAM files of individuals UCL4 II.6 and II.8 (upper and middle panel) and a control (lower panel) were generated from whole exome sequencing (UK10K project). Note that exon 1 of TM4SF19 was found to be generally absent from all UK10K exomes, probably due to capture failure by the Agilent V2 kit. Supplementary Fig. 4. X-rays of individuals UCL 4 II.1 and II.5. Mild brachydactyly (a-d) and clinodactyly (c,d) but no rib shortening or handlebar clavicles (e,f) and no signs of trident acetabulum with spurs (g,h).  Figure  6d,e are highlighted in the black boxes (two different exposures of the same blot were used for D1bIC2 and D1bLIC in 6d, as shown).

Whole Exome Sequencing (WES)
Exomes (69) generated at UCL were prepared using the Agilent V2 (UK10K project) or Truseq whole exome kit with sequencing of enriched libraries performed as 75 base paired-end reads (Illumina HiSeq) [11][12][13] . Exomes (154) generated at UCLA were sequenced at the University of Washington Center for Genome Sciences, captured using the NimbleGen SeqCap EZ Exome Library v2.0 probe library and sequenced on the Illumina GAIIx platform with 50 bp reads. For both sets of data, passed reads were mapped to human genome reference hg19 using BWA v0.5.9r16 14 with realignment around known indels from the 1000 Genomes Pilot study and recalibration of base quality scores performed using GATK 1.1.5 15 . Then variant calling and independent filtering was performed using SAMtools mpileup v0.1.17 16 and GATK UnifiedGenotyper v1.3.31 17 and the callsets were merged. Exome variant profiles were filtered by standardised protocol using EVAR software tool vs 0.2.2 beta (www.exome.info) [11][12][13] . They were filtered by quality score using a standard cut-off value that implicates a base error rate < 0.1%, and we removed all variants occurring in dbSNP132, the 1000 Genomes and NHLBI Exome project with a minor allele frequency >0.1%. Remaining variants were filtered for affecting protein coding regions and known splice sites (up to 15 basepairs away from the exon-intron boundary), and synonymous changes were removed. Variants were then filtered against an in-house database to remove any occurring with a frequency >0.1%. Remaining biallelic variants were prioritised according to their presence in cilia proteome databases and mutation type. Coverage of known disease causing genes was manually analysed using Integrated Genome Viewer. Copy number variant analysis was performed using ExomeDepth 18 . Functional effects of variants were assessed using Polyphen2 and Mutationtaster.

Linkage analysis
Genome-wide homozygosity mapping was performed using GeneChip Human Mapping 250K NspI arrays (Affymetrix) on DNA samples from parents and 8 children from family UCL4 (I.1, I.2, II.1, II.2, II.3, II.4, II.5, II.7, II.8, II.9). Linkage data were generated by the Bioinformatics platform (Université Paris Descartes, Paris) and visualized using the interface created by the Bioinformatic platform. Linkage analysis was performed using MERLIN 19 considering a threshold of 1.0 and a minimum chromosomal region length of 2 Mb.

Antisense-morpholino knockdown in zebrafish
Zebrafish husbandry and morpholino techniques were performed according to standard procedures 20-23 with licenced approval from the Home Office (UK) under the Animal (Scientific Procedures) Act 1986. MO sequences were p53 MO: 5'-GCGCCATTGCTTTGC-3'; control MO: 5'-TAGTGCAAAGCTTA-3'; tctex1d2 MO1 (exon2-intron2 splice site): 5′-CTAAATCGTACTGTGCTGTTACCTT-3′; tctex1d2 MO2 (intron2-exon3 splice site) 5'-CATGTCTGAGGAAGAGCAAATATAC-3'. tctex1d2 morpholinos were designed by Genetools LCC on ENSDART00000076424 (transcript zebrafish tctex1d2-001). MOs were dissolved in ddH2O and injections were performed at the 1-4 cell stage. Controls were injected with non-silencing control morpholino not targeting any known zebrafish gene to control for unspecific effects due to the injection and morpholino doses for sub-phenotypic co-injection experiments were balanced with this non-silencing control morpholino. To rule out phenotypic effects due to morpholinodependant p53 activation, all embryos were co-injected with 5ng of p53 Morpholino 20 . Morpholino effects on transcript level were assessed by reverse transcription Omniscript kit (Qiagen) using Trizol-Chloroform extracted total RNA obtained from 3 dpf embryos and primers listed in Supplementary Table 1. Light microscopy imaging was performed for whole mount embryos fixed in 4% PFA and mounted in methylcellulose, with cartilage stained using Alcian-Blue 20,22 .

Immunofluorescence in Chlamydomonas
Cells were fixed with cold methanol and stained for immunofluorescence microscopy as adapted from Sanders and Salisbury 24 . Cells were allowed to adhere to the coverslips for about 3 minutes before fixation. Fixed and dried cells on coverslips were rehydrated in PBS buffer for 1 hour and blocked in blocking buffer (5% BSA, 1% fish skin gelatin, 10% goat serum in 1×PBS) for 1 hour. Cells were incubated with the primary antibody diluted in blocking buffer overnight at 4°C. Incubation with the secondary antibody was done at room temperature for 2 hours. The coverslips were mounted onto the slides with ProLong antifade reagent (Molecular Probes, Eugene, OR). Images were acquired with an AxioCam camera, AxioVision 3.1 software, and an Axioskop 2 plus microscope (Zeiss). Images were processed using Adobe Photoshop (Adobe Systems Incorporated, San Jose, CA).

Chlamydomonas protein biochemistry
To prepare Chlamydomonas whole cell lysates, cell pellets were re-suspended in water. An equal volume of SDS-PAGE sample loading buffer (50 mM Tris, pH 8, 160 mM DTT, 5 mM EDTA, 50% sucrose, 5% SDS, 0.04% Bromophenol Blue) was added to lyse the cells. When necessary, cells were  Supplementary Table 6. When needed, protein samples were concentrated using Ultracel®-3k filters (Merck kGaA). Signals on blots were detected by use of x-ray film. Contrast was adjusted using Adobe Photoshop (Adobe Systems Inc.). Western blots were quantitated by comparing the intensity of a protein band in the mutant fraction to that of the same protein in a serial dilution of the wildtype fraction on the same western blot.

Proteomics in mammalian cells
HEK293T cells were transfected with constructs expressing SF-TAP-TCTEX1D2, or SF-TAP-RAF1 as a control, using polyethyleneimine (PEI, Polysciences), cultured for 48 hours and then lysed in lysis buffer containing 30 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.5% Nonidet-P40 (NP40), freshly supplemented with protease inhibitor cocktail (Roche), phosphatase inhibitor cocktail 2 (Sigma) and PhosphataseArrest III (GBiosciences) for 20 minutes at 4°C. Streptavidin-and FLAG-based tandem affinity purification were performed using standard protocols 25,26 . 5% of the finale eluate was evaluated by PAGE followed by silver staining according to standard protocols, while the remaining 95% was subjected to protein precipitation with chloroform and methanol. Protein precipitates were subsequently subjected to mass spectrometry analysis and peptide identification 27 28 . Protein identifications were accepted if they could be established at >95.0% probability and contained at least 2 identified peptides. Protein probabilities were assigned using Protein Prophet 29 . Proteins containing similar peptides that could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony.

Human immunoprecipitation protocol
HEK293T cells were transfected with the corresponding constructs using Lipofectamine 2000 (Life Technologies Ltd), according to the manufacturer's protocol. Cells were harvested 48 hours after transfection in 500 ml of lysis buffer (1% NP40, 20 mM Tris-HCl pH 7.4, 50 mM NaCl + Roche Complete Protease inhibitor cocktail), homogenized with a 27G needle 10 times and then centrifuged at 4°C for 5min, 15800 g. Supernatant was collected and used for downstream applications or stored short-term at -20°C. 20 μl of Protein G Dynabeads were washed 3 times with 200 μl PBTw (PBS + 0.05% Tween-20), and then resuspended in 200 μl of PBTw and 1 μg of the corresponding antibody was added. The solution was rotated at RT for 90 mins, then the beads:antibody complexes were washes 3 times in 500 μl PBTw. After the last wash, complexes were resuspended in 200 μl IP buffer (0.5% NP40, 20 mM Tris-HCl pH 7.4, 50 mM NaCl + Roche Complete Protease inhibitor cocktail). Cell lysate was combined with resuspended beads:antibody complex, and IP buffer was added to a final volume of 1 ml. Mix was rotated overnight at 4°C. Beads were washed 3 times with 1 ml IP buffer, then resuspended in 50 ul of 1x Sample Buffer (5x sample buffer, 250 mM Tris-HCl pH 6.8, 10% SDS, 30% glycerol, 5% beta-mercaptoethanol, 0.02% bromophenol blue, diluted to 1x in lysis buffer). Samples were boiled for 5 min then loaded on acrylamide gels for western blot or stored short-term at -20°C.