Brother in2840 driver download

The acronym VATER/VACTERL association (MIM# 192350) refers to the rare, non-random co-occurrence of the following component features (CFs): vertebral defects (V), anorectal malformations (A), cardiac defects (C), tracheoesophageal fistula with or without esophageal atresia (TE), renal malformations (R), and limb defects (L) [Solomon et al., 2014]. Patients may present with additional congenital anomalies however, the clinical diagnosis requires the presence of at least three CFs [Solomon et al., 2014]. Patients presenting with two CFs have been termed VATER/VACTERL-like [Solomon et al., 2014]. The involvement of genetic factors in the development of this rare association is suggested by reports of familial occurrence, the increased prevalence of component features among first-degree relatives, high concordance rates among monozygotic twins, and chromosomal (micro-)aberrations or single gene mutations in affected individuals [Solomon et al. 2014].

Human ciliopathies such as McKusick-Kaufman syndrome (MIM# 236700), Meckel syndrome Type 1-11, and Kartagener syndrome (MIM# 244400) are clinically closely related to the VATER/VACTERL association as they may present with congenital heart defects (C), congenital anomalies of the kidney (R), and limb (L) defects. Additionally, a VATER/VACTERL-like phenotype was reported in the ciliary mutant mouse model of Ift172 (avc1) [Bujakowska et al., 2015]. Ift172 encodes a component of the intraflagellar transport (IFT) within the cilium.

Based on these observations, we hypothesized that ciliary candidate genes, notably members of the IFT complex, may also be defective in patients with a VATER/VACTERL phenotype. We therefore carried out targeted re-sequencing and mutation analysis of 25 ciliary candidate genes (Supp. Table S1) comprising two genes encoding proteins of intraflagellar transport subcomplex A (IFT-A) (IFT139/TTC21B, IFT144/WDR19), 14 genes encoding IFT proteins of the subcomplex B (IFT-B) (IFT172/SLB, IFT81/CDV1, IFT57/HIPPI, IFT52, IFT54/TRAF3IP1, IFT46, IFT74/CCDC2, IFT22/RABL5, IFT20, IFT25/HSPB11, IFT27, TTC30B/IFT70, TTC30A, IFT88/TG737), and one gene (TTC26) known for its association with IFT-B. Since the avc1 mutant mouse, representing a hypomorphic Ift172 mutation, mechanistically presents with perturbed sonic hedgehog (Shh) signalling [Bujakowska et al., 2015], we included eight ciliary candidate genes (TULP3, SDCCAG8, GPR161, KIF3A, KIF3B, KIFAP3, WDR60, and KIF17) for their association with the Shh-pathway. As human ciliopathies usually follow an autosomal-recessive mode of inheritance [Tobin and Beales, 2009], we assumed an underlying autosomal-recessive disease model for the selected ciliary candidate genes.

To investigate the contribution of previously disease-associated candidate genes, we included HOXD13, PTEN, FOXF1, and ZIC3 in our analysis. Whereas ZIC3 is X-linked, HOXD13, PTEN, and FOXF1 have an underlying autosomal-dominant mode of inheritance. For both PTEN and HOXD13, a single point mutation was previously described in a VATER/VACTERL-like phenotype [Reardon et al., 2001; Garcia-Barceló et al., 2008]. While HOXD13 mutation carriers usually exhibit limb defects (MIM# 186300) representing a VACTER/VACTERL CF, PTEN mutation carriers (MIM# 153480) usually do not exhibit any VATER/VACTERL CFs. Nevertheless, in order to re-evaluate these findings, we added both genes to our re-sequencing analysis. Four heterozygous point mutations were previously reported in FOXF1 in four unrelated patients with alveolar capillary dysplasia and misalignment of the pulmonary veins (ACD/MPV) and additional malformations in VATER/VACTERL organ systems [Stankiewicz et al., 2009]. Several ZIC3 mutations, including point mutations, deletions, and, in one instance, an intragenic polyalanine expansion, were reported for both VATER/VACTERL and VATER/VACTERL-like phenotypes with or without hydrocephalus [Wessels et al., 2010; Chung et al., 2011]. We did not include FGF8 or TRAP1 in our analysis, since both genes had been previously investigated by our group [Zeidler et al., 2014; Saisawat et al., 2014]. Furthermore, we did not include LPP in our analysis, since Hernandez-Garcia et al. [2011] when investigating 170 patients with isolated TE or TE as part of their VATER/VACTERL association by microarray, failed to detect any LPP deletions or disease-causing mutations in the Sanger sequencing analyses of 42 of these patients. Finally, we did not include disease-causing genes for Fanconi anemia, as VATER/VACTERL-like phenotypes have been described in less than 5% of patients with secured Fanconi anemia [Faivre et al., 2005].

Our study sample comprised 123 patients (78 males and 45 females). A VATER/VACTERL phenotype with at least three CFs was present in 80, a VATER/VACTERL-like phenotype with two CFs in 43 patients. Due to the recruitment procedure, all of our patients had anorectal malformations [Schramm et al., 2011]. Within our cohort was one affected sib pair, brother and sister, who were not previously reported. Here, only the sister was included in the initial study sample.

All 29 candidate genes were subjected to our re-sequencing approach using microfluidic Access-Array technology (Fluidigm). In two of these genes, namely KIF17 and SDCCAG8, re-sequencing was not successful and no results could be obtained. In the other 27 candidate genes, PCR-based multiplex amplification of genomic target regions was successful and followed by molecular barcoding and next-generation re-sequencing on an Illumina HiSeq platform, as outlined in detail elsewhere [Halbritter et al., 2012]. In total, all 406 coding exons and splice sites were sequenced with this method (Supp. Table S1).

Read alignment and variant detection were carried out using CLC Genomics Workbench software (CLC-bio, Aarhus, Denmark) as described previously [Halbritter et al., 2012]. For variant calling, standard filtering criteria were used (Supp. Materials and Methods). The study was conducted in accordance with the Declaration of Helsinki, and ethical approval was obtained from the respective ethics committees at the recruitment centers. Written informed consent was obtained from all patients and parents prior to study entry. Mutations, unclassified variants, and phenotype data were submitted to the ClinVar, NCBI database (http://www.ncbi.nlm.nih.gov/clinvar/).

Mutation analysis of all 25 ciliary candidate genes did not identify homozygous or compound heterozygous disease causing variants in any of the investigated 123 patients. However, we identified 55 heterozygous variants of unknown significance in two ITF-A associated, nine IFT-B associated, and five Shh-pathway associated candidate genes (Supp. Table S1). Among these variants, 16 were found to be novel according to NCBI dbSNP (National Center for Biotechnology database for single nucleotide polymorphism) (Supp. Table S1). In nine patients, two mutations in different genes were identified, suggesting a possible digenic disease pathway. Nevertheless, since none of the combination of genes was found in more than one patient, these combinations remain of uncertain significance.

Mutation analysis of FOXF1 (NM_001451.2), HOXD13, and PTEN, identified one de novo mutation in FOXF1, c.658G>T (p.Gly220Cys), in male patient no. 400. This patient presented with an anorectal malformation (A), left sided renal agenesis (R) and glandular hypospadias.

Mutation analysis of ZIC3 (Ensembl ENST00000370606; NM_003413.3) identified a recurrent disease-causing mutation (c.49G>T, p.Gly17Cys) in three patients, including the sister of the above mentioned sib pair. Familial segregation analyses using Sanger sequencing revealed one further mutation in her affected brother. The male patient no. 347 (Supp. Table S2) presented with three CFs namely recto-vesical fistula (A), atrial septal defect (C), and right renal agenesis, grade IV-V vesicoureteral reflux (left) (R). Additional features were cryptorchidism and penoscrotal transposition. He inherited the ZIC3 p.Gly17Cys mutation from his unaffected mother. In addition, this patient showed a heterozygous p.Asn411Ser variant in the IFT57 gene, encoding the protein intraflagellar transport 57, Clamydomonas homolog. This variant was not detected in his mother and no paternal sample was available for testing.

In family no. 606, the p.Gly17Cys mutation in ZIC3 was detected in the sister (no. 603b) and her brother (no. 603a; Supp. Table S2). The girl presented with vestibular fistula (A), higher grade vesicoureteral reflux (R) and 13 ribs on both sides. Her brother presented with recto prostatic fistula (A) and atrial septal defect (C). Because both siblings were affected to similar degree, we tested the hypothesis that skewed X chromosome inactivation in the sister led to expression of only the affected allele. Testing for the skewedness of X-chromosome inactivation in the family (Supp. Materials and Methods) showed indeed the allele transmitted from the mother to the daughter was the main active allele (only 9% inactivated) while the other allele was nearly 91% inactivated (Supp. Fig. S1). The same active allele in the sister was shared by her brother, which explains why both siblings were equally affected.

The male patient no. 633 (Supp. Table S2) presented with two CFs, namely recto-urethral fistula (A) and right sided ectopic kidney with grade I vesicoureteral reflux (R). In addition, he displayed penoscrotal transposition and glandular hypospadias. In this patient, aside from the ZIC3 p.Gly17Cys mutation, an additional heterozygous variation p.Arg1311His in the TTC21B gene, encoding the protein tetratricopeptide repeat domain 21B, was found. TTC21B is a retrograde intraflagellar transport protein. Homozygous or heterozygous TTC21B mutations have been causally related to nephronophthisis-12 and/or the Joubert syndrome 11 (JBTS11; MIM# 613820). In our patient, both missense mutations were transmitted from the mother, affected with cleft lip and palate, an anomaly not associated with ZIC3 or TTC21B.

To better describe the significance of the identified mutations and variants we used various online prediction tools (Mutation Taster (http://www.mutationtaster.org/), MutPred (http://mutpred.mutdb.org/), PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/), SIFT (http://sift.jcvi.org/) and PROVEAN (http://provean.jcvi.org/index.php). In patient no. 400, the p.Gly220Cys (c.658G>T) mutation in FOXF1 substitutes a highly conserved small non-polar glycine with a polar cysteine. The introduction of a further cysteine residue may also allow formation of illegitimate disulfide bridges. This nucleotide transition is not listed in SNP database (build 141); however, only Mutation Taster classified it as disease-causing (Supp. Table S3). Previous FOXF1 mutations have been described in patients with VATER/VACTERL-like phenotypes [Stankiewicz et al. 2009]. The authors suggested, that heterozygous invalidation of FOXF1 could be considered in VATER/VACTERL cases for which there is a severe respiratory distress and/or neonatal lethality. Our patient with a probably disease-causing de novo mutation did not show any cardiac defects, MPV or any other large vessel anomaly, and his neonatal period was not complicated by respiratory distress. The Variant was submitted to ClinVar (http://www.ncbi.nlm.nih.gov/clinvar/) (Supp. Table S3).

The ZIC3 p.Gly17Cys mutation affects a highly conserved amino acid residue, substituting a small non-polar glycine with a polar cysteine. In accordance with most recent reports [D'Alessandro et al., 2013], four out of five publically available mutation prediction programs (Mutation Taster, MutPred, PolyPhen-2, SIFT and PROVEAN) predicted this mutation to be deleterious (Supp. Table S4). D'Alessandro et al. [2013] and De Luca et al. [2010] did not observe p.Gly17Cys in altogether 1129 controls, however, it is listed with database no. rs147232392 in the NCBI SNP database (Build 141), referring to data from the NHLBI ESP cohort. Here, a minor allele frequency (MAF) of 0.002 was observed in 2840 samples with no further data given. As outlined recently [D'Alessandro et al., 2013], p.Gly17Cys has also been deposited in the Exome Variant Server (EVS) [2012]; here, its presence was reported in 19 heterozygous females out of 4,053 individuals and 12 hemizygous males (total no. 2,414), resulting in a MAF of 0.0029. However, no phenotypic data were provided for these mutation carriers. Furthermore, the EVS data show that p.Gly17Cys is found slightly more often in European Americans (MAF 0.0039) than in African Americans (MAF 0.0012). This unequal distribution of the frequency might be due to a European founder mutation. The additional IFT57 p.Asn411Ser (c.1232A>G) substitution observed in patient 347 represents a conservative exchange and affects a highly conserved carboxyterminal residue. There is no SNP database (Build 141) or Deep Catalog of Human Genetic Variation (www.1000genomes.org) entry and this variant is predicted to be possibly damaging by two (Mutation Taster, MutPred) out of five publically-available mutation prediction programs. The Variant was submitted to ClinVar (Supp. Table S3).

In patient no. 633, the additional p.Arg1311His (c.3932G>A) substitution also resides in a highly conserved carboxyterminal position in the TTC21B protein. A deleterious effect is predicted by four out of five publically available mutation prediction programs (Mutation Taster, Polyphen2, SIFT and PROVEAN). The variant has been deposited in the SNP database (Build 141) with ref. no. rs139327086 and was observed once in the NHLBI ESP cohort in 4542 samples (MAF: 0.001) with no further data available.

Based on their strict conservation and pathogenic prediction it could be that both additional variants: the IFT57 p.Asn411Ser (c.1232A>G) substitution observed in patient no. 347, and the TTC21B p.Arg1311His (c.3932G>A) substitution in patient no. 633, might affect protein function. Although TTC21B has been implicated in nephronophthisis 12 (MIM# 613820) only in the mutated homozygous/compound heterozygous state, heterozygous TTC21B mutations have been observed in some patients with Joubert syndrome 11 [Davis et al., 2011], an allelic ciliopathy. Hence, as suggested by Davis et al. [2011], it could be that the additional variants in TTC21B and IFT57 might have contributed to a mutational load in the described patients.

To attribute a functional effect to the p.Gly17Cys mutation, we performed a PROSITE motif search (www.Expasy.ch) which identified a potential N-myristoylation site in ZIC3, comprising amino acids 13-18 (GLgvGS). N-myristoylation refers to a co-translational modification involving the covalent attachment of the 14-carbon saturated fatty acid myristate to the N-terminal glycine of proteins [Farazi et al., 2001]. This irreversible acylation promotes protein-membrane and protein-protein interactions, thereby regulating protein targeting and function. Although other small uncharged residues, like Cys17, seem to be tolerated [Towler et al., 1988], a substitution of Gly17 might impair the covalent addition of myristate. In eukaryotic proteins, N-myristoylation has been shown to be essential for cell growth and survival [Farazi et al., 2001], and further studies should elucidate whether recruitment of this motif also governs ZIC3 function, thereby providing evidence for the functional consequences of this Glycine17 missense mutation, now observed in 11 (9 unrelated) patients (Supp. Table S2).

Overall the phenotypic spectrum of ZIC3 mutation carriers is quite variable, with incomplete penetrance in males shown by D'Alessandro et al. [2013], who reported transmission of the mutation from unaffected males. Upregulation of ZIC3 is detected in around 8% of female carriers [Chhin et al., 2007; de Luca et al., 2010]. This has, at least in one case, been attributed to a skewed (97:3) X-inactivation pattern [Chhin et al., 2007], a mechanism which also occurred in our female patient 606b. Penetrance variability has also been observed in the mouse, indicating that the penetrance of the Zic3 null allele is sensitive to genetic background differences [Purandare et al., 2002]. To date, eight patients fulfilling the criteria of VATER/VACTERL association (at least three CFs being present) have been reported with ZIC3 mutations (Supp. Table S4). None of these patients displayed hydrocephalus. Situs abnormalities were detected in four. Interestingly, aside from cardiac defects, ARM have been reported in at least 17 patients with ZIC3 involvement [Purandare et al., 1999; Wessels et al., 2010; Chung et al., 2011]. Although knockout mice recapitulate the human heterotaxy and cardiac phenotype with occasional vertebral/rib anomalies, ARM or other VATER/VACTERL CFs were not reported in Zic3 deficient mice [Purandare et al., 2002; Sutherland et al., 2013].

To assess the potential role of ZIC3 and FOXF1 for the VATER/VACTERL association, we investigated the expression of Zic3 and Foxf1 by in situ hybridization (ISH) on mouse embryos at embryonic days (E) 9.5 to 14.5, the equivalent of human gestational weeks 4-6 (postulated time of VATER/VACTERL organogenesis in humans) [Stevenson and Hunter, 2013]. In situ hybridization on mid-sagittal sections of E12.5 and E13.5 mouse embryos (corresponding to approximately week 5 of human gestation) was performed to examine Foxf1 expression (Fig. 1). Strong expression of Foxf1 could be detected in various VATER/VACTERL associated tissues, including trachea and esophagus, pre-vertebrae, and areas surrounding the urethra and rectum/anus within the genital tubercle. Staining was also apparent in the developing lungs, and in the tooth buds.

In situ hybridization of mouse embryo sections for Foxf1. At E12.5 and E13.5 Foxf1 could be detected in various tissues affected in VATER/VACTERL association, including trachea and esophagus, vertebrae, and urorectal structures. Staining was detected in mesenchyme surrounding the urethra and anus at both stages. Foxf1 was also present in the lung and tooth buds, but was absent from the heart (a, anus; es, esophagus; gt, genital tubercle; ht, heart; lu, lungs; tr, trachea; ve, vertebra).

These ISH data for Foxf1 in developing mouse embryos suggests a role for this gene in relevant VATER/VACTERL tissues, although whole-mount expression data from urogenital tissues from the GUDMAP database (http://www.gudmap.org) does not show significant staining in the kidneys. Nevertheless, Foxf1 heterozygotes have been shown to display tracheo-esophageal atresia and fistulas (Mahlapuu et al., 2001), one hallmark of VATER/VACTERL association.

In situ hybridization for Zic3 was performed on mouse embryos at gestational days (E) 9.5-14.5, with particular emphasis on the uro-rectal region (Fig. 2). Since Zic3 is located on the X-chromosome, male and female embryos were examined separately to determine whether sex-specific differences in expression were present. No differences were noted. From E9.5 to E13.5 Zic3 could be detected in discrete regions within the developing brain, eye, spinal column, and limbs. A stripe of Zic3 expression was present at all stages within the anterior pre-somitic mesoderm (Fig. 2, top). Expression in the limbs was present in the autopod and apical ectodermal ridge at early stages, and then became confined to the presumptive wrist region and digital tips. A close-up examination of staining in the uro-rectal region of male and female embryos between E12.5 and E14.5 again revealed no sex-specific differences in expression. Staining was undetected in the cloaca/genital tubercle up to E12.5. From E13.5, there was weak expression in the lateral aspects of the genital tubercle, but Zic3 could not be detected in the developing anal region (Fig. 2, bottom).

Whole-mount in situ hybridization of mouse embryos for Zic3. (top) From E9.5 to E13.5 Zic3 could be detected in discrete regions within the developing brain, eye, spinal column, and the developing limbs. A stripe of Zic3 expression is present at all stages at the anterior pre-somitic mesoderm. No differences were noted in the expression pattern between males and females, and male embryos are shown. (bottom) Close-up examination of staining in the uro-rectal region of male and female embryos between E12.5 and E14.5 revealed weak expression in the lateral aspects of the genital tubercle, but Zic3 was not detected in the developing anal region. Note that the tails have been removed at E12.5 and E13.5 to reveal the genital tubercle (flb, forelimb bud; hlb, hindlimb bud; fl, forelimb; hl, hindlimb; gt, genital tubercle; tl, tail).

Taken together our WISH data do not suggest a role for Zic3 in renal and urogenital development, since there was lack of expression in these mouse tissues at the relevant stages. The failure to detect Zic3 expression in heart at the stages examined (E9.5-E13.5), is in agreement with the study from Sutherland et al. [2013], who found absent Zic3 expression in mouse heart when using either WISH or a Zic3-LacZ-BAC reporter mouse during heart looping morphogenesis. These authors defined the temporal requirement of Zic3 for cardiac development between E7.0 and E8.0. However, Zic3 expression could be detected in murine E10.5 heart tissue by RT-PCR [Bedard et al., 2011]. Nevertheless, we detected Zic3 expression in the limbs and prevertebral structures, all tissues affected in the VATER/VACTERL association. Failure of staining renal and urogenital tissues might depend on low but relevant Zic3 expression only detectable by RT-PCR.

In summary, we present novel families with VATER/VACTERL or VATER/VACTERL-like phenotypes with mutations in ZIC3 and FOXF1, respectively. WISH/ISH expression studies in mouse embryos revealed Zic3 and Foxf1 expression in the presomitic mesoderm and limb, esophageal, tracheal, vertebral, anal and genital tubercle tissues, all belonging to VATER/VACTERL organ systems, providing independent corroboration for ZIC3 and FOXF1 as disease-causing genes for human VATER/VACTERL association.

Acknowledgments

Footnotes

References