Comments
Abstract
Rationale: Complete tracheal ring deformity (CTRD) is a rare congenital abnormality of unknown etiology characterized by circumferentially continuous or nearly continuous cartilaginous tracheal rings, variable degrees of tracheal stenosis and/or shortening, and/or pulmonary arterial sling anomaly.
Objectives: To test the hypothesis that CTRD is caused by inherited or de novo mutations in genes required for normal tracheal development.
Methods: CTRD and normal tracheal tissues were examined microscopically to define the tracheal abnormalities present in CTRD. Whole-exome sequencing was performed in children with CTRD and their biological parents (“trio analysis”) to identify gene variants in patients with CTRD. Mutations were confirmed by Sanger sequencing, and their potential impact on structure and/or function of encoded proteins was examined using human gene mutation databases. Relevance was further examined by comparison with the effects of targeted deletion of murine homologs important to tracheal development in mice.
Measurements and Main Results: The trachealis muscle was absent in all of five patients with CTRD. Exome analysis identified six de novo, three recessive, and multiple compound-heterozygous or rare hemizygous variants in children with CTRD. De novo variants were identified in SHH (Sonic Hedgehog), and inherited variants were identified in HSPG2 (perlecan), ROR2 (receptor tyrosine kinase–like orphan receptor 2), and WLS (Wntless), genes involved in morphogenetic pathways known to mediate tracheoesophageal development in mice.
Conclusions: The results of the present study demonstrate that absence of the trachealis muscle is associated with CTRD. Variants predicted to cause disease were identified in genes encoding Hedgehog and Wnt signaling pathway molecules, which are critical to cartilage formation and normal upper airway development in mice.
Complete tracheal ring deformity (CTRD) is a rare congenital disorder of the respiratory tract characterized by abnormal tracheal development resulting in circumferentially continuous cartilaginous rings, variable degrees of tracheal stenosis, tracheal shortening, and/or various anatomic anomalies of the pulmonary arteries referred to as “pulmonary artery sling.” The occurrence of CTRD in twins suggests that CTRD may have a genetic basis; however, the molecular pathogenesis of CTRD is unknown.
Although targeted gene ablation has identified genes critical to tracheal development in mice, to our knowledge, this is the first study to examine the pathogenic basis of CTRD in humans. The trachealis muscle was absent in all of five patients with CTRD. The exome sequences of children with CTRD and their biological parents (“trio analysis”) identified somatic and inherited gene variants in the patients with CTRD comprising a molecular signature of altered Hh (Hedgehog) and Wnt signaling in patients with CTRD. Targeted genetic manipulation of murine homologs required for tracheal cartilage development in mice was used to validate the relevance of the human gene variants identified in patients with CTRD.
Cartilaginous defects of the upper airways are relatively common and are associated with high morbidity. For example, tracheomalacia, which is characterized by incompletely formed or absent cartilaginous rings in the trachea and bronchi (Figure 1), occurs in 1 in 3,000 live births. In contrast, complete tracheal ring deformity (CTRD) is a rarer abnormality involving excessive cartilage formation comprising circumferentially continuous or nearly continuous cartilaginous tracheal rings as well as variable involvement of other developmental anomalies, including tracheal stenosis, tracheal shortening, pulmonary artery anomalies, and reduced mucociliary clearance, in infants and children (Figure 1). CTRD occurs without associated malformations in only 10–25% of patients (1) and in association with cardiovascular anomalies in 50% of the patients. Left pulmonary artery sling is the most common anomaly in CTRD (2), and 70% of patients with pulmonary artery sling have associated CTRD (3). The strong association between vascular sling and CTRD anomalies suggests that a genetic basis underlies these anomalies. The presence of long tracheal segment stenosis and associated cardiovascular and other anomalies in these patients increases the requirement for hospitalization, repeated bronchoscopies, and surgeries (4, 5).
Figure 1. Anterior (top) and cross-sectional (bottom) schematic views of the trachea, proximal airways, and pulmonary arteries illustrating normal anatomy and defects seen in complete tracheal ring deformity (CTRD) and other congenital tracheal deformities with defects in tracheal cartilage development. Cross-sectional views correspond to the level indicated by the red line in the illustration directly above. Illustrations are based on results from the present study and published reports on genetically modified mice. Pulmonary artery sling is a vascular anomaly often seen in CTRD in which the left pulmonary artery (LPA) originates from the right pulmonary artery (RPA) and encircles the distal trachea and right primary bronchus before entering the hilum of the left lung (not shown). See text for further description and details.
[More] [Minimize]The pathogenesis of congenital tracheal stenosis is unknown, but its occurrence in monozygotic twins suggests a genetic etiology (6). In mammals, the signaling pathways that direct “patterning” (development of specific tissue types) in the large airways during embryogenesis converge on SOX9 (SRY [sex-determining region Y]-box 9), a transcription factor required for development of cartilage in the ventrolateral–tracheal mesenchyme (7–9). Wnt ligands and β-catenin are proteins that promote expression of Sox9 and interact with SOX9 to regulate chondrocyte differentiation (10). BMP4 (bone morphogenetic protein 4), via SOX9, induces mesenchymal cells to differentiate into chondroblasts (11), and SHH (Sonic Hedgehog), via SOX9, stimulates tracheal cartilage formation (8). FGF (fibroblast growth factor) family members (FGF4, 8, 10, and 18) also participate in cartilage formation (12). In mice, disruption of genes encoding factors in the Wnt, Bmp, Fgf, RA (retinoic acid), and Hh (Hedgehog) pathways disrupt tracheal cartilage development (13) and cause tracheal cartilage anomalies ranging from absence to a continuous rostrocaudal cartilaginous ring known as “tracheal sleeve deformity” (Figure 1). The tracheal epithelium participates in tracheal patterning through its critical role in organizing the tracheal mesenchyme along the dorsoventral axis, which specifies the distribution of cartilage and muscle cells that form the cartilaginous rings and trachealis muscle (14). In mice, tracheomalacia, tracheal sleeve, and other upper airway anomalies are caused by disruption of specific genes (15). Although no mutations have been associated with CTRD in humans and the pathogenesis of CTRD is unknown, disproportionate growth of cartilage along the posterior tracheal membrane (16) and an intrinsic defect in patterning by cervical splanchnic mesenchyme have been proposed (17). A genetic basis is also suggested by association of CTRD with other mediastinal and cervical chondrogenic anomalies such as shortening of the neck and trachea, pulmonary agenesis, and vascular anomalies, all involving tissues that require patterning signals from splanchnic mesenchyme during embryogenesis.
During human embryogenesis, the trachea and esophagus separate at approximately the fourth week of gestation and thereafter extend caudally as tracheal epithelial cells differentiate and submucosal glands form. Mesenchymal cells surrounding the tracheal tube form cartilage, muscle, and connective tissue by Week 8. Normal dorsal–ventral patterning of the tracheal mesenchyme leads to formation of cartilage in the ventrolateral mesenchyme and muscle in the dorsal mesenchyme (Figure 1). Although the critical cellular and molecular events directing dorsal–ventral patterning of tracheal mesenchyme are unknown, proliferation, differentiation, and migration of muscle and cartilage precursor cells are all likely to be involved (18).
Because rare congenital diseases are frequently genetic in etiology and exome sequencing of a small number of unrelated patients is a powerful strategy for identifying genes causing rare Mendelian disorders (19–21), we sought to identify the gene mutation(s) causing CTRD. We used exome sequencing of patient “trios” (i.e., the patient and both biological parents) to identify potential disease-causing mutation(s) in CTRD. Clinical, histologic, and gene mutation analysis of CTRD trios were integrated with data from mice with gene disruption–related developmental tracheal cartilage and muscle abnormalities to identify the signaling pathways and gene mutations associated with CTRD.
The clinical protocol was approved by Cincinnati Children’s Hospital Institutional Review Board, and written informed consent was obtained from the parents of all study subjects. Between May and September 2012, five children with CTRD (probands) receiving medical care at Cincinnati Children’s Hospital Medical Center (CCHMC) were identified and enrolled by their treating physician (M.J.R.) together with both biological parents (proband, mother, and father = family trio). All probands had CTRD involving at least 50% of the trachea (none had short tracheal segment involvement) with or without left pulmonary artery sling based on bronchoscopic and intraoperative examination. None had any other genetic syndrome or congenital anomaly, based on clinical history and examination. Primary human airway epithelial cells were obtained from healthy, nonsmoking adults. Passage 1 cells were obtained from the Tissue Procurement and Cell Culture Core (CFF BOUCHE15R0) and the University of North Carolina Cystic Fibrosis Research and Translation Core Center Cell Models Core (NIHP30DK065988), maintained as air–liquid cultures, and evaluated as described previously (22).
Animals were housed in pathogen-free conditions and handled according to protocols approved by the local institutional animal care and use committee. To examine the effects of Wnt signaling on tracheal development, embryos with conditional Wls gene deletion were created by first crossing Wlsf/f and ShhCre/wt mice and then backcrossing their F1 progeny to Wlsf/f mice (23). The resulting (Wlsf/f ShhCre/wt) embryos undergo deletion of Wls in tracheal epithelial cells under regulation by the temporospatial expression of Shh in the developing trachea. To improve readability, Wls conditional knockout embryos are referred to as “Wls-cKO.” To examine Wnt/β-catenin signaling in tracheal development, embryos expressing a reporter gene (LacZ) under control of either Axin2 or the TCF (T-cell factor) promoter were created by first crossing Wlsf/wt/ShhCre/wt and Axin2-LacZ knock-in mice (or Wlsf/wt/ShhCre/wt and TopGal mice) and then crossing F1 progeny with Wlsf/f (14, 24). These embryos are referred to as “Wls-cKO/Axin2-LacZ” (or “Wls-cKO/TopGal”) embryos.
Human tracheal specimens and mouse tissues were prepared and examined using established methods as described in the online supplement.
Human adult tracheobronchial smooth muscle cells were obtained from Lifeline Cell Technology. Primary airway epithelial cells were obtained from nonsmoking human donors. RNA isolation, cDNA synthesis, and qRT-PCR amplification were performed using TaqMan primers on an Applied Biosystems StepOnePlus Real-Time PCR System (Thermo Fisher Scientific) as previously described (25).
Genomic DNA was obtained from peripheral blood using routine methods, and exon-specific next-generation sequencing was performed at PerkinElmer Inc. Briefly, samples were processed using the Epicentre Nextera DNA Sample Prep Kit (Illumina) and SureSelect All Exon 50Mb (Agilent Technologies) protocols and sequenced on an Illumina HiSeq 2000 system. Next-generation sequencing and sequence data were aligned to the human genome (HG19) using Burrows-Wheeler Aligner software (bio-bwa.sourceforge.net) (see online supplement).
Exome sequence data for the proband and both biologic parents from each family (trio) were used to identify variants in probands by filtering and annotating results using Golden Helix SVS software and data from the Golden Helix website (www.goldenhelix.com) and the CCHMC exome allele frequency database. An allele was defined as rare if it occurred in less than 1% of the population (including all races) or if it was novel by comparison with data from the 1000 Genomes Project (phase 1, version 3), the NHLBI Exome Sequencing Project (ESP6500), and CCHMC whole-exome sequencing databases (an allelic frequency table derived from 312 whole exomes determined at CCHMC). De novo variants were new hemizygous polymorphisms present in the proband but not in either parent. Homozygous recessive variants were polymorphisms present in both alleles in the proband and only one allele in each parent. Compound-heterozygous variants were two (or more) different polymorphisms present in the proband that were each present in only one allele of each biological parent. By definition, in compound-heterozygous variants, one polymorphism was required to have a frequency of ≤5% in the general population and the other one had to be rare (<1%). Functional predictions for variants were made using the dbNSFP functional predictions database (version 2).
Measurement of mRNA concentrations by in situ hybridization was performed as previously described (26) (online supplement).
The five children with CTRD studied had a median age of 5.6 years (range, 2–12 yr). All were white, and 40% were Hispanic. All were residents of the United States or the Caribbean islands; all had CTRD involving more than 50% of the trachea; four had pulmonary artery sling; and four had undergone surgical correction of CTRD (Table 1).
| Trio ID | Participant Age (yr) | Race | Ethnicity | Vascular Anomaly | CTRD-related Surgical Therapy | ||
|---|---|---|---|---|---|---|---|
| Proband | Mother | Father | |||||
| 1 | 4 | 34 | 35 | White | Hispanic | PAS | PAS repair, pericardial patch, STP |
| 2 | 2 | 30 | 30 | White | Unknown | PAS | STP, PAS repair |
| 3 | 4 | 36 | 41 | White | Not Hispanic | PAS, PDA | STP, PAS repair |
| 4 | 12 | 46 | 48 | White | Not Hispanic | PAS | PAS repair, tracheal resection, homograft patch |
| 5 | 6 | 31 | 43 | White | Hispanic | None | None |
Paraffin-embedded tracheal specimens were obtained from patients with CTRD (n = 5) who had undergone surgical correction of CTRD at CCHMC. Normal tracheal specimens (n = 3) from cadaveric donors were obtained from the CCHMC Biobank Core Facility. Tracheas from patients with CTRD all had circumferentially continuous cartilaginous rings, undetectable trachealis muscle, widening of the subepithelial lamina propria, increased submucosal glands, and luminal narrowing indicating tracheal stenosis (Figures 1–3). Tracheas from individuals without CTRD had C-shaped cartilaginous rings located in the ventrolateral trachea and readily identifiable trachealis muscle in the dorsal trachea (Figures 1 and 2).
Figure 2. Representative photomicrographs of tracheal pathology specimens from children with complete tracheal ring deformity (CTRD 1, 2, and 3) and without CTRD (non-CTRD controls 1 and 2). (A–F) Representative images of tracheal specimens immunostained to identify and locate ACTA2 (αSMA; red) or Sox9 (green). (G–L) Representative images of tracheal specimens stained with Alcian blue to identify and locate cartilage (blue). Regions in upper panels (A–C and G–I) examined at higher power in lower panels (D–F and J–L) are indicated. In the CTRD specimens, note absence of the trachealis muscle (B and C), widening of the lamina propria (LP) (K and L), and increased numbers of submucosal glands (E and F).
[More] [Minimize]
Figure 3. Validation of human complete tracheal ring deformity (CTRD) candidate genes by analysis of their expression in human or mouse tracheal cells and tissues. (A) qRT-PCR analysis showing expression of WNT5A, ROR2, ACTA2, SHH, and NKX2.1 in human primary epithelial cells from nonsmoking adults (n = 3) and expression of WNT5A, ROR2, and ACTA2 but not SHH or NKX2.1 in a human smooth muscle cell line (n = 3 replicates). (B) In situ hybridization analysis showing Shh expression in the developing trachea of mice embryos at the indicated ages. Brown color indicates Shh expression (arrowheads). Note that Shh expression is abundant during tracheal patterning early in gestation (Embryonic Days 12.5 and 14.5 [E12.5 and E14.5, respectively]) and diminished at later times after induction of tracheal cartilage formation (E17.5). (C) Representative photomicrographs of tracheal tissues from a stillborn human fetus (at 20 wk of gestation) immunostained to demonstrate that HSPG2 (green) is present in tracheal mesenchyme. Tissues were also immunostained with NKX2.1 (orange) and counterstained for DNA with DAPI (blue). (D) Immunofluorescence staining of Muc5B (red) in tracheal specimens from individuals without or with CTRD (as indicated). Note the presence of Muc5b in submucosal glands (sg), which are increased in number in CTRD. ACTA2 = α-smooth muscle actin; HSPG2 = heparin sulfate glycoprotein 2; ND = not detected; NKX2.1 = NK2 homeobox 1; ROR2 = receptor tyrosine kinase–like orphan receptor 2; SHH = Sonic Hedgehog; T = trachea; WNT5A = Wnt family member 5A.
[More] [Minimize]To identify gene mutations associated with CTRD, genomic DNA was obtained, and exon sequencing was performed for the proband and both parents from each trio (Data File E1 in the online supplement). De novo homozygous, compound heterozygous, and hemizygous variants were identified in CTRD probands (Tables 2 and E2). Six of the nine de novo variants identified were validated by PCR amplification and Sanger sequencing of DNA encompassing the variant for all members of the relevant trio (not shown). Compound and rare heterozygous variants were not sequenced, because they were confirmed by their presence in the proband’s mother and father.
| Gene | Trio ID | Variant | Mutation Classification | Functional Effect Prediction* | ||
|---|---|---|---|---|---|---|
| SIFT | Polyphen-2 | MutationTaster | ||||
| RIF1 | 5 | 2:152319736 | De novo frameshift Ins | ND | ND | ND |
| NT5DC2 | 3 | 3:52561731 | De novo frameshift Ins | ND | ND | ND |
| SHH | 5 | 7:155596148 | De novo NS SNV | Damaging | Possibly damaging | Disease causing |
| NPAS1 | 5 | 19:47525028 | De novo frameshift Del | ND | ND | ND |
| FAM83C | 1 | 20:33874884 | De novo frameshift Del | ND | ND | ND |
| MED14 | 5 | X:40551442 | De novo NS SNV | Tolerated | Benign | Disease causing |
| HSPG2 | 4 | 1:22160001-SNV | CH NS SNV | Tolerated | Probably damaging | Polymorphism |
| HSPG2 | 4 | 1:22179223-SNV | CH NS SNV | Tolerated | Benign | Disease causing |
| HSPG2 | 4 | 1:22203106-SNV | CH NS SNV | Tolerated | Probably damaging | Disease causing |
| TTN | 4 | 2:179470001-SNV | CH NS SNV | Damaging | Possibly damaging | — |
| TTN | 4 | 2:179579935-SNV | CH NS SNV | Damaging | Benign | — |
| TTN | 4 | 2:179588996-SNV | CH NS SNV | Damaging | Benign | — |
| TTN | 4 | 2:179593449-SNV | CH NS SNV | Damaging | Benign | — |
| TTN | 4 | 2:179615060-SNV | CH NS SNV | Damaging | Probably damaging | Disease causing |
| TTN | 4 | 2:179615386-SNV | CH NS SNV | Tolerated | Benign | Polymorphism |
| TTN | 2 | 2:179401074-SNV | CH NS SNV | Damaging | Probably damaging | — |
| TTN | 2 | 2:179480163-SNV | CH NS SNV | Damaging | Benign | — |
| TTN | 2 | 2:179606172-SNV | CH NS SNV | Damaging | Benign | — |
| ROR2 | 5 | 9:94486491-SNV | CH NS SNV | Damaging | Probably damaging | Disease causing |
| ROR2 | 5 | 9:94487134-SNV | CH NS SNV | Damaging | Probably damaging | Disease causing |
| MYH13 | 1 | 17:10215944-SNV | CH NS SNV | Damaging | Probably damaging | Polymorphism |
| MYH13 | 1 | 17:10236464-SNV | CH NS SNV | Damaging | Probably damaging | Disease causing |
A SHH gene polymorphism (7:155596148) was identified in the proband of trio 5 and determined to be a de novo nonsynonymous variant (i.e., present in the proband but not in either parent) and predicted to disrupt the structure and function of the encoded SHH protein (Table 2). Interestingly, this variant was present in only approximately 30% of nucleated peripheral blood cells. The mosaicism observed for this mutation is consistent with 1) the observation that Shh gene ablation in mice is lethal in the perinatal period owing to developmental defects in the lungs and other endoderm-derived organs (27) and 2) the prediction of its deleterious effects on protein structure and function. Notwithstanding this, the degree of mosaicism may vary in other tissues, including the trachea. To evaluate the relevance of the SHH mutation, we quantified expression of SHH and NKX2.1 (NK2 homeobox 1) in adult primary human airway epithelial cells. Although detectable, mRNA for both was low (Figure 3A), consistent with reduced expression of SHH and other developmental modulators beyond embryogenesis (28). As controls, adult airway muscle cells expressed WNT5A, ROR2, and ACTA2 but not SHH or NKX2.1 (Figure 3A). Accordingly, to further evaluate relevance, we quantified expression of the mouse homolog in vivo during embryogenesis and found that Shh was expressed transiently in tracheal epithelium during development. Expression was high early (embryonic day 12.5 [E12.5]) during patterning of tracheal mesenchyme, less abundant later (E14.5), and absent after cartilaginous rings had formed (E17.5) (Figure 3B).
De novo mutations were also identified in other genes, including RIF1 (Ras-proximate-1–interacting factor homolog), NT5DC2 (5′-nucleotidase domain containing 2), NPAS1 (neuronal pulmonary artery sling domain protein 1), FAM83C (family with sequence similarity 83, member C), and MED14 (mediator complex subunit 14) (Table 2). The variants in RIF1, NT5DC2, NPAS1, and FAM83C were frameshift mutations and thus assumed to disrupt protein structure and function, and the MED14 variant was predicted to be disease causing (Table 2). These variants exhibited mosaic expression suggesting intolerance to loss of function, as predicted by their effects on protein structure/function. Notwithstanding this, no role in tracheal cartilage or trachealis muscle development has been reported for any of these genes.
Among the gene variants identified, several may alter the expression or function of modulators of tracheal cartilage and/or muscle development via the Hedgehog signaling pathway. Compound heterozygous variants in HSPG2 (perlecan) were identified in the proband of trio 4, and rare hemizygous HSPG2 variants were found in the proband of trio 4 (Table 2). HSPG2 encodes a protein present in the extracellular matrix of the trachea (Figure 3C) that has been identified as a component of the Hedgehog signaling pathway (29).
Rare and novel gene variants were also identified with potential to affect tracheal cartilage development because of their involvement in Hedgehog signaling (Tables 3 and E2). These variants were predicted to disrupt function of the encoded gene product. The probands in this study each carried between two and six variants in genes related to Hedgehog signaling (Table 3).
| Gene | Role in Hedgehog Signaling Pathway | Diseases Associated with Gene Mutation | Trio ID | Mutation |
|---|---|---|---|---|
| SHH | Critical in embryonic patterning and tracheal cartilage formation; deletion causes tracheomalacia | Vertebral defects, anal atresia, tracheoesophageal fistula/esophageal atresia, renal dysplasia, and cardiac and limb abnormalities | 5 | NS, SNV |
| GLI2 | Mediates Shh signaling, localized in cytoplasm, and activates PTCH expression; deletion causes tracheal stenosis | Greig cephalopolysyndactyly syndrome, Pallister-Hall syndrome, preaxial polydactyly type IV, and postaxial polydactyly types A1 and B | 5 | NS, SNV |
| BOC | Mediates muscle precursor cell interactions and promotes myogenic differentiation | 5 | NS, SNV | |
| ADAM17 | Implicated in cell–cell and cell–matrix interactions, including muscle development, and neurogenesis | Likely involved in autoimmune diseases, including psoriasis, rheumatoid arthritis, multiple sclerosis, and Crohn’s disease | 5 | NS, SNV |
| SMO | Transduces signals to other proteins after activation by a Hedgehog protein–patched protein complex | No known associated disease | 4 | NS, SNV |
| GLI3 | Localized to cytoplasm and activates PTCH expression; deletion of Gli2 and Gli3 leads to stenotic trachea | Greig cephalopolysyndactyly syndrome, Pallister-Hall syndrome, preaxial polydactyly type IV, and postaxial polydactyly types A1 and B | 4 | NS SNV |
| GLI1 | Activated by Shh pathway and regulates stem cell proliferation | No known associated disease | 3 | NS, SNV |
| PTCH1 | Receptor for Shh ligands; Hedgehog binding relieves inhibition of G protein–coupled receptor “smoothened” and initiates signaling | Basal cell nevus syndrome and holoprosencephaly | 1 | NS, SNV |
| ADCY9 | Membrane-bound enzyme catalyzing cAMP formation; regulated by G protein–coupled receptors, protein kinases, and calcium | No known associated disease | 1 | NS, SNV |
| DISP2 | May be required for normal Hedgehog signaling during embryonic pattern formation | No known associated disease | 5 | NS, SNV |
| EFCAB7 | Encodes a calcium ion binding protein that mediates signaling by Hedgehog and by G protein–coupled receptors | Ellis-Van Creveld syndrome | 2 | NS, SNV |
| LRP2 | Multiligand endocytic receptor with role in cell signaling; extracellular ligands include Shh | Donnai-Barrow syndrome and facio-oculo-acoustico-renal syndrome | 2, 5 | NS, SNV |
| TTC21B | Localized to cilium axoneme; may participate in retrograde intraflagellar transport in cilia | Various ciliopathies, nephronophthisis 12, and asphyxiating thoracic dystrophy 4 | 3 | NS, SNV |
| PSMB6 | Member of the proteasome B-type family; is a 20S core β-subunit in the proteasome | No known associated disease | 1 | NS, SNV |
| DYNC2H1 | Cytoplasmic dynein protein; involved in retrograde transport and intraflagellar transport, which is required for ciliary/flagellar assembly | Various primary ciliary dysfunction disorders that often involve polydactyly, abnormal skeletogenesis, and polycystic kidney disease | 1 | Stop-gain SNV |
Variants were also identified in genes involved in Wnt signaling. In the proband of trio 5, compound heterozygous and nonsynonymous single-nucleotide variants were identified in ROR2 that were predicted to be deleterious (Tables 2 and E2 and Data File E2). ROR2, which is expressed in tracheobronchial cells (Figure 3A), encodes a receptor mediating Wnt/β-catenin–independent (i.e., noncanonical) signaling that is activated by WNT5a ligand (30). Interestingly, ROR2 mutations are associated with skeletal and craniofacial malformations (Table 4). In the proband of trio 2, variants were identified in WLS (Data File E2), a gene required for secretion of Wnt family members. In the probands of trios 3, 4, and 5, variants were identified in LRRC7 (leucine-rich repeat containing 7) (Data File E2), a target of Wnt signaling that interacts with β-catenin (i.e., canonical signaling) and actin. To address the relevance of variants with potential to alter tracheal development by disrupting Wnt/β-catenin signaling, we tested nonphosphorylated β-catenin in human neonatal tracheal tissues (Figure E2).
| Malformation | Mouse Gene | Human Syndrome* | Gene Variant in CTRD? | References |
|---|---|---|---|---|
| Tracheal sleeve | Fgf10 FgfR2 | Apert | No | 47 |
| Reduced number of tracheal rings and discontinuous rings | Fstl1 | — | Yes | 48 |
| Wnt5a | Robinow | No | 31 | |
| Ror2 | Robinow | Yes | 30 | |
| Tbx4 Tbx5 | — | No | 49 | |
| Sox2 | — | No | 50 | |
| Hoxa5 | — | No | 51, 52 | |
| Raldh2 | DiGeorge | Yes | 53 | |
| Discontinuous rings | β-Catenin | — | No | 54, 55 |
| Rspo2 | — | No | 55 | |
| Wnt4 | SERKAL | No | 56 | |
| Foxf1 | VACTERL | No | 57 | |
| Cftr | Cystic fibrosis | No | 58 | |
| miR125b miR30a/c (Snail1) | — | No | 59 | |
| Tmem16 | — | No | 60 | |
| TgfβRII | — | No | 61 | |
| FGF18 | — | No | 7, 62 | |
| Cav3.2 | — | No | 63 | |
| Mek1/Mek2 | — | No | 64 | |
| Osr1 | — | No | 65 | |
| Nearly absent cartilage | Shh | — | Yes | 8, 27 |
| Wls | — | Yes | 14, 25 | |
| Sox9 | Campomelic dysplasia | No | 9, 18 |
Variants were identified in genes involved in muscle development. In the probands of trios 3 and 4, compound heterozygous, nonsynonymous single-nucleotide variants were identified in MYH13 (encoding myosin heavy chain 13) that were judged to be damaging/probably damaging and disease-causing polymorphisms (Table 2). The significance of these variants is uncertain and requires further study. In probands of trios 2 and 4, compound heterozygous and rare hemizygous variants were identified in TTN (Table 2 and Data File E1), which encodes a structural molecule, Titin, involved in assembly and functioning of vertebrate striated muscles.
In summary, patients with CTRD harbored variants in genes associated with the Hedgehog and Wnt signaling pathways, known to be important in tracheal patterning and development in mice.
Patterning and organization of the tracheal mesenchyme are highly conserved in humans and mice; both have C-shaped cartilaginous rings located ventrolaterally and trachealis muscle tissue located dorsally connecting the tips of the cartilaginous rings (Figures 1, 4A, and 4B). We confirmed that Nkx2.1, Sox9, and αSMA (ACTA2) were expressed in the tracheal epithelium, ventrolateral mesenchymal chondrocyte progenitors, and dorsal mesenchymal muscle cells, respectively, in mice (Figure 4C), as they are in the developing trachea in humans. Gene ablation studies have identified a critical role for multiple genes in patterning and development of the trachea, including Shh, Ror2, Wnt5a, Wls, Fgf10, and Fgfr1 (14, 27, 30–33), all of which are essential for formation of tracheal cartilage and muscle in mice or humans (Figures 4D–4I and 5K and Table 4). Mutations in the human orthologs of the murine genes are associated with congenital tracheal anomalies in humans (Figure 1 and Table 4), and our data for patients with CTRD identified variants in SHH and ROR2 (Table 2).
Figure 4. Tissue-specific expression of genes regulating tracheal patterning and development in mice. (A and B) Photograph of trachea and lungs removed en bloc from Embryonic Day 14.5 (E14.5) wild-type mouse embryos and immunostained to identify αSMA (α-smooth muscle actin; green) in the developing trachealis muscle (double arrows) located between the tips of the C-shaped cartilaginous rings of the trachea (T) and in the lung (L) and blood vessels (BV), as well as to identify Sox9 (red) in the cartilaginous rings (single arrow) of the developing trachea. (C) Representative photomicrograph of the trachea and surrounding mesenchyme from a 13.5-day-old wild-type mouse embryo sectioned perpendicular to the rostrocaudal axis and immunostained to identify αSMA (red), Sox9 (green), and Nkx2.1 (Nk2 homeobox 1; blue). Note αSMA staining in muscle progenitors in the dorsal mesenchyme, Sox9 staining in cartilage progenitors in the ventrolateral mesenchyme, and Nkx2.1 staining in epithelial cells of the developing trachea (T). (D–F) In situ hybridization analysis of the trachea and lungs removed en bloc from E14.5 wild-type mouse embryos to identify and localize expression of mRNA encoding Shh (Sonic Hedgehog; D), Hspg2 (E), and Ror2 (receptor tyrosine kinase–like orphan receptor 2; F) in developing trachea. (E, H, and I) Note localization of Shh in epithelial cells of the developing trachea (T) and esophagus (E) (E); Hspg2 in mesenchymal cells surrounding the trachea and esophagus, especially in the dorsal tracheal mesenchyme (H); and Ror2 in the mesenchyme surrounding the tracheal epithelium (I). Tissues were also immunostained for Wnt9 (Wnt family member 9; blue). C = cartilage; Hspg2 = heparin sulfate proteoglycan 2; Sox9 = SRY (sex-determining region Y)-box 9.
[More] [Minimize]
Figure 5. Evaluation of human complete tracheal ring deformity candidate genes involved in Wnt signaling by determining the effects of Wls gene deletion on expression of their homologs in the developing mouse trachea. (A and B) Photograph of trachea and lungs removed en bloc from Embryonic Day 15.5 (E15.5) wild-type or Wls-conditional knockout (Wls-cKO) mouse embryos immunostained to identify αSMA (α-smooth muscle actin; green) and Sox9 (red) in the developing trachea (T) and lungs. (C and D) Representative photomicrographs of the trachea and surrounding mesenchyme from E13.5 wild-type or Wls-cKO mouse embryos sectioned perpendicular to the rostrocaudal axis and immunostained to identify αSMA (green), Sox9 (red), and Nkx2.1 (blue) in the developing trachea. Note the increased αSMA+ muscle progenitors located ectopically in the dorsolateral tracheal mesenchyme, reduced numbers of Sox9+ cartilaginous progenitors in the ventrolateral tracheal mesenchyme, and Nks2.1+ tracheal epithelial cells in Wls-cKO embryos. (E–H) Representative photomicrographs of the trachea and surrounding mesenchyme from E15.5 wild-type or Wls-cKO mouse embryos stained with Alcian blue to demonstrate the absence of tracheal cartilage in Wls-cKO (F and H) compared with the normal appearance of tracheal cartilage in wild-type mice (E and G). (I–Y) In situ hybridization analysis of the trachea and lungs removed en bloc from E14.5 wild-type or Wls-cKO mouse embryos to identify and localize expression of mRNA encoding Col2a1 (I and J), Wls (K and L), Wnt5a (M and N), Lrrc7 (O and P), Ror2 (Q and R), Myh11 (S and T), Axin2 (U and V), and Lef1 (X and Y) in developing trachea. In tracheal mesenchymal cells of Wls-cKO mouse embryos, note in lateral tracheal mesenchymal cells (T) the absence of Col2a1 (J), Wls (L), and Wnt5a (N) expression; reduced expression of Ror2 (R), Axin2 (V), and Lef1 (Y); increased expression of Lrrc7 (P) and Lef1 (Y); and ectopic expression of Myh11 (T). Axin2 = axis inhibition protein 2; Col2a1 = collagen type II, α1; E = esophagus; Lef1 = lymphoid enhancer binding factor 1; Lrrc7 = leucine-rich repeat containing 7: Myh11 = myosin heavy chain 11; Nkx2.1 = Nk2 homeobox 1; Ror2 = receptor tyrosine kinase–like orphan receptor 2; Sox9 = SRY (sex-determining region Y)-box 9; Wls = Wntless ligand secretion mediator; Wnt5a = Wnt family member 5a.
[More] [Minimize]CTRD and tracheomalacia can be considered to be extremes in a spectrum of phenotypic anomalies of tracheal cartilage and muscle development. To test the hypothesis that CTRD and tracheomalacia may arise as heterogeneous developmental responses to disruption of the same genes, we evaluated the effects of ablation of mouse homologs of human gene variants that we identified in patients with CTRD. Conditional deletion of Wls in the developing embryo (under temporospatial control synonymous with Shh expression) resulted in the absence of tracheal cartilage development and formation of ectopic, poorly organized smooth muscle in the mesenchyme encompassing the developing tracheal tube (Figures 5A–5J; compare Wls-cKO [right panels] with wild type [left panels]) (see also References 14, 24). Canonical (Wnt/β-catenin) signaling was reduced in Wls-cKO mice, as shown by reduced mRNA for Axin2 (Figure 5V) and Lef1 (Figure 5Y), which are direct targets of Wnt/β-catenin signaling (14, 34). Wls-cKO embryos had reduced expression of Wnt5a (Figure 5N), a ligand that binds to Ror2, as well as Ror2 itself (Figure 5R), a receptor expressed on tracheal cartilage and muscle progenitors. Finally, noncanonical signaling mediated by Jnk (Janus kinase) was downregulated in Wls-cKO mice, as indicated by reduced phosphor-JNK staining in tracheal cells in the developing embryo (Figures E1G and E1H).
As an additional approach, we evaluated the effects of disrupting Wnt/β-catenin signaling in two different reporter mice (TopGal and Axin2-LacZ) that express β-galactosidase from promoters activated by Wnt/β-catenin signaling (see online supplement for description) (24). Conditional deletion of Wls in the developing respiratory tract (under temporospatial control synonymous with Shh expression in respiratory epithelial cells) resulted in reduced reporter expression (indicating reduced Wnt/β-catenin signaling) in developing tracheal and lung tissues (Figures E1A–E1F). Conditional deletion of Wls also reduced expression of genes (Sox9, Col2a1) required for cartilage formation (Figures 5B and 5D) and increased expression of a gene (Myh11) involved in muscle formation (Figure 5T) (see also Gene Expression Omnibus accession no. GSE97445 and Reference 25).
Comparative mouse studies also revealed that mouse homologs of human variants in genes which participate in Wnt/β-catenin signaling were also downregulated in tracheal tissues of Wls-cKO embryos, including WLS (Figure 5L) and LRRC7 (Figure 5P) (Data File E2). Together, the genetic studies in human children with CTRD and genetically modified mice support a role for Wnt signaling in the differentiation of tracheal cartilage and muscle.
The present study demonstrates that CTRD is associated with absence of the trachealis muscle, widening of the tracheal lamina propria, increased numbers of submucosal glands, and tracheal stenosis. Trio exome sequencing analysis identified gene variants in the patients with CTRD, including a de novo mutation in SHH as well as several compound heterozygous and rare hemizygous variants in genes involved in Hedgehog and Wnt signaling that were predicted to disrupt the structure and function of the encoded protein and cause disease. Comparative analysis of mice with ablation of murine homologs of human genes in which CTRD-associated variants were found supports a role for disruption of Hedgehog and/or Wnt signaling in the pathogenesis of CTRD.
Absence of trachealis muscle in CTRD was a novel finding of this study. Whether this anomaly is caused by a secondary effect of abnormal cartilage development or results from primary failure of the tracheal mesenchyme to differentiate into muscle is unknown. The abnormality could occur if progenitor cells that normally form the trachealis muscle were either not specified or failed to proliferate owing to disruption of required morphogenic signals. Alternatively, it could occur as a secondary effect of an associated malformation such as arterial sling (2, 3), which was present in four of five patients with CTRD in our study. In the latter case, extrinsic compression by the sling could prevent radial expansion of the tracheal tube, thereby altering tracheal cartilage and muscle development.
Although multiple gene variants relevant to tracheal development were identified and validated in patients with CTRD, the SHH mutation is of particular interest because ablation of the mouse homolog, Shh, abrogates tracheal cartilage formation (8, 27, 33). Shh is a secreted morphogen (35) that regulates tracheal patterning by establishing a gradient or “field” of Sox9 expression required for cartilage formation (8, 27). It is expressed during early embryogenesis in epithelial cells that induce adjacent peritracheal mesenchymal cells to differentiate into cartilage and muscle cells, and in the later stages of respiratory tract development, its expression is attenuated. Shh expression is readily detectable on E12–E14 and is diminished by E15.5, a time corresponding to about 8 weeks of gestation in humans. Although reactivation of SHH expression in the human respiratory tract in response to injury has been reported (28), expression was not detected in human fetal trachea at 20 weeks of gestation or in human adult primary airway epithelial cells, consistent with the pattern of transient expression during early embryogenesis in mice. Thus, lack of human tracheal specimens sufficiently early in embryogenesis (<8 wk of gestation) was a limitation of our study. Although less is known about a potential role for Shh in trachealis muscle formation, Shh is required for mesenchymal cell survival, proliferation, and differentiation into muscle cells in the developing ureter (36). Heterozygous variants in genes required for Hedgehog signaling were identified in patients with CTRD, including transcription factors GLI1 (glioma-associated oncogene homolog 1), GLI2, and GLI3. Ablation of the murine homologs, Gli2 and Gli3, causes tracheal stenosis (37). Thus, identification of variants in these genes in patients with CTRD further supports a role for disruption of Hedgehog signaling in the pathogenesis of CTRD.
Compound heterozygous and rare hemizygous variants were identified in human homologs of mouse genes required for tracheal cartilage development, including Hspg2, Wls, and Ror2. HSPG2 encodes a basement membrane–specific heparin sulfate proteoglycan (perlecan) located in the extracellular matrix that is required to form ligand gradients and attenuates ligand-mediated signaling in the developing embryo (38). In mice, perlecan is required for cartilage and vascular development (39–42) and interacts with Shh to localize and regulate its activity and signaling to chondrocytes (29, 43, 44). Thus, function-disrupting HSPG2 variants could be expected to disrupt Hedgehog signaling–dependent tracheal patterning by altering the SHH gradient or field (and/or binding of SHH to its receptor, Patched) required for normal induction of tracheal cartilage and muscle cell development during embryogenesis.
The variants identified in genes required for WNT signaling were also of interest, the ROR2 variant in particular. In mice, Ror2 is required to stimulate growth of the tracheal tube and conducting airways and for formation of tracheal cartilage (25, 30). We observed Ror2 expression in tracheal muscle cell progenitors, suggesting that it may be important in formation of the trachealis muscle, consistent with a recent report (32). Conditional gene ablation of Wls also causes absence of tracheal cartilage and ectopic muscle formation, a phenotype opposite to that of CTRD (14). Like Shh, Wnt signaling is active during early embryogenesis, is attenuated during later stages of tracheal development, and can be reactivated by injury. Wls is required for secretion of Wnt ligands, including Wnt5a, which binds Ror2. In mice, Wnt/β-catenin–dependent (canonical) and Wnt/β-catenin–independent (noncanonical) signaling are both important in respiratory tract development; however, delineating their precise roles is challenged by pathway overlap (14, 32). In humans, components of each pathway can be specifically detected by antiphosphoprotein antibodies; however, autopsy specimens from stillborn or premature infants were not sufficiently well preserved to permit detection. Similarly, human specimens were inadequate for detection of component-specific mRNA. Thus, this study limitation will require further experiments. Notum is a target and negative feedback modulator of Wnt/β-catenin signaling and may balance canonical and noncanonical (Wnt5a/Ror2) signaling during tracheal development. Notum gene ablation partially recapitulates the phenotype observed by gene ablation of Ror2 in that it causes tracheal stenosis; forms of ectopic, disorganized cartilage; and reduces amounts of disorganized trachealis muscle (25). These data support the concept that ROR2 mutations identified in patients with CTRD may drive pathogenesis. Ror2 mediates organization of the trachealis muscle in the developing trachea by binding Wnt5a, a ligand inducing Wnt/β-catenin–independent signaling, which is critical to mesenchymal cell polarization and differentiation (32). Thus, although ROR2 variants may contribute directly to the trachealis muscle defect, they may contribute to the pathogenesis of CTRD more broadly by disruption of Wnt signaling. Because Wnt5a also modulates Shh concentration and activity, interactions between Hedgehog and Wnt signaling pathways may be important in regulating mesenchymal cell differentiation during tracheal embryogenesis (45, 46).
On the basis of our observations and published reports, we propose a mechanism for the regulation of tracheal patterning during embryogenesis in which SHH and Wnt signaling pathways govern specification and development of cartilage and muscle in large airways of the respiratory tract (Figure 6). We further propose that gene mutations disrupting these pathways drive the pathogenesis of CTRD. In mice, ablation of genes required for Hedgehog and Wnt signaling (e.g., Shh, Wls, and Ror2) disrupts tracheal cartilage and muscle patterning, resulting in abnormal tracheal development. In children with CTRD, we identified mutations in genes required for Hedgehog and Wnt signaling (SHH, PERLECAN, WLS, and ROR2) that are human homologs of murine genes critical to tracheal development in mice, which supports a potential role for variants in these genes in the pathogenesis of CTRD. In conclusion, in the present study, we found that trio-based exome sequencing analysis combined with focused mutational analysis of mouse homologs of the human genes associated with CTRD-related variants was particularly useful.
Figure 6. Proposed model of the regulatory control of tracheal mesenchyme differentiation. (Left) Photomicrograph of the mouse tracheal tube and surrounding mesenchyme taken at Embryonic Day 13.5 (E13.5) immunostained to identify cells expressing Nkx2.1 (purple), Sox9 (green), and α-smooth muscle actin (light blue). (Right) Schematic diagram of tracheal epithelial (purple) and cartilaginous (green) smooth muscle (light blue) progenitors in the surrounding mesenchyme, as well as the various signaling ligands and receptors that control patterning and formation of cartilaginous rings and trachealis muscle in the developing trachea. Tracheal epithelial cells express Wls, Shh, Wnt, and Wnt7b, whereas mesenchymal cells express Ror2, Ptc, Smo, GliA, Lrrc2, and Hspg2. Wls mediates secretion of Wnt ligands that activate canonical signaling mediated by β-catenin and noncanonical signaling mediated by the Ror2 receptor. β-Catenin induces Notum, which may promote Ror2-mediated (noncanonical) signaling and attenuates canonical signaling by β-catenin. Thus, Notum may regulate the balance between canonical and noncanonical signaling. Shh interacts with receptors Ptc and Smo to mediate cartilage ring formation via GliA. Hspg2 in the extracellular mesenchymal matrix appears to help establish a spatial gradient or field of Shh that regulates tracheal patterning and formation of cartilage and muscle. Lrrc7 is expressed in the dorsal mesenchyme and is a target of Wnt signaling that contributes to trachealis muscle cells. Although the precise interactions among Shh, Hspg2, Ror2, Wls, and Lrrc7 are unclear, the appropriate balanced expression and activity of the proteins encoded by the genes in this signaling network are critical for the formation of cartilage and tracheal muscle.
[More] [Minimize]The authors thank the children with CTRD and their parents for their collaboration on this study, Evan Meyer and Mike Muntifering for assistance with confocal imaging, Chuck Crimmel for assistance with graphic design, Kaulini Burra for assistance with immunofluorescence and microscopy, and Kathryn Wikenheuser-Brokamp and Jaime Reuss for obtaining human tracheal pathology specimens.
| 1. | Antón-Pacheco JL, Cano I, García A, Martínez A, Cuadros J, Berchi FJ. Patterns of management of congenital tracheal stenosis. J Pediatr Surg 2003;38:1452–1458. |
| 2. | Herrera P, Caldarone C, Forte V, Campisi P, Holtby H, Chait P, et al. The current state of congenital tracheal stenosis. Pediatr Surg Int 2007;23:1033–1044. |
| 3. | Dodge-Khatami A, Tsang V, Roebuck D, Elliott M. Management of congenital tracheal stenosis: a multidisciplinary approach. Images Paediatr Cardiol 2000;2:29–39. |
| 4. | Hofferberth SC, Watters K, Rahbar R, Fynn-Thompson F. Management of congenital tracheal stenosis. Pediatrics 2015;136:e660–e669. |
| 5. | Antón-Pacheco JL, Morante R. Operative or non-operative treatment of congenital tracheal stenosis: is there something new? J Thorac Dis 2017;9:4878–4880. |
| 6. | Fraga JC, Gabra HO, Calkoen EE, McLaren CA, Roebuck DJ, Elliott MJ. Long segment congenital tracheal stenosis in twins successfully treated by slide tracheoplasty. J Pediatr Surg 2009;44:640–643. |
| 7. | Elluru RG, Thompson F, Reece A. Fibroblast growth factor 18 gives growth and directional cues to airway cartilage. Laryngoscope 2009;119:1153–1165. |
| 8. | Park J, Zhang JJ, Moro A, Kushida M, Wegner M, Kim PC. Regulation of Sox9 by Sonic Hedgehog (Shh) is essential for patterning and formation of tracheal cartilage. Dev Dyn 2010;239:514–526. |
| 9. | Turcatel G, Rubin N, Menke DB, Martin G, Shi W, Warburton D. Lung mesenchymal expression of Sox9 plays a critical role in tracheal development. BMC Biol 2013;11:117. |
| 10. | Akiyama H, Lyons JP, Mori-Akiyama Y, Yang X, Zhang R, Zhang Z, et al. Interactions between Sox9 and β-catenin control chondrocyte differentiation. Genes Dev 2004;18:1072–1087. |
| 11. | Hatakeyama Y, Tuan RS, Shum L. Distinct functions of BMP4 and GDF5 in the regulation of chondrogenesis. J Cell Biochem 2004;91:1204–1217. |
| 12. | Goldring MB, Tsuchimochi K, Ijiri K. The control of chondrogenesis. J Cell Biochem 2006;97:33–44. |
| 13. | Morrisey EE, Hogan BL. Preparing for the first breath: genetic and cellular mechanisms in lung development. Dev Cell 2010;18:8–23. |
| 14. | Snowball J, Ambalavanan M, Whitsett J, Sinner D. Endodermal Wnt signaling is required for tracheal cartilage formation. Dev Biol 2015;405:56–70. |
| 15. | Marcus CL, Smith RJ, Mankarious LA, Arens R, Mitchell GS, Elluru RG, et al. Developmental aspects of the upper airway: report from an NHLBI Workshop, March 5-6, 2009. Proc Am Thorac Soc 2009;6:513–520. |
| 16. | Chen JC, Holinger LD. Congenital tracheal anomalies: pathology study using serial macrosections and review of the literature. Pediatr Pathol 1994;14:513–537. |
| 17. | Faust RA, Stroh B, Rimell F. The near complete tracheal ring deformity. Int J Pediatr Otorhinolaryngol 1998;45:171–176. |
| 18. | Hines EA, Jones MK, Verheyden JM, Harvey JF, Sun X. Establishment of smooth muscle and cartilage juxtaposition in the developing mouse upper airways. Proc Natl Acad Sci USA 2013;110:19444–19449. |
| 19. | Ng SB, Buckingham KJ, Lee C, Bigham AW, Tabor HK, Dent KM, et al. Exome sequencing identifies the cause of a Mendelian disorder. Nat Genet 2010;42:30–35. |
| 20. | Vissers LE, de Ligt J, Gilissen C, Janssen I, Steehouwer M, de Vries P, et al. A de novo paradigm for mental retardation. Nat Genet 2010;42:1109–1112. |
| 21. | Xu B, Roos JL, Dexheimer P, Boone B, Plummer B, Levy S, et al. Exome sequencing supports a de novo mutational paradigm for schizophrenia. Nat Genet 2011;43:864–868. |
| 22. | Gokey JJ, Snowball J, Sridharan A, Speth JP, Black KE, Hariri LP, et al. MEG3 is increased in idiopathic pulmonary fibrosis and regulates epithelial cell differentiation. JCI Insight 2018;3:122490. |
| 23. | Carpenter AC, Rao S, Wells JM, Campbell K, Lang RA. Generation of mice with a conditional null allele for Wntless. Genesis 2010;48:554–558. |
| 24. | Snowball J, Ambalavanan M, Sinner D. Studying Wnt signaling during patterning of conducting airways. J Vis Exp 2016;(116):53910. |
| 25. | Gerhardt B, Leesman L, Burra K, Snowball J, Rosenzweig R, Guzman N, et al. Notum attenuates Wnt/β-catenin signaling to promote tracheal cartilage patterning. Dev Biol 2018;436:14–27. |
| 26. | Wang F, Flanagan J, Su N, Wang LC, Bui S, Nielson A, et al. RNAscope: a novel in situ RNA analysis platform for formalin-fixed, paraffin-embedded tissues. J Mol Diagn 2012;14:22–29. |
| 27. | Miller LA, Wert SE, Clark JC, Xu Y, Perl AK, Whitsett JA. Role of Sonic hedgehog in patterning of tracheal-bronchial cartilage and the peripheral lung. Dev Dyn 2004;231:57–71. |
| 28. | Liu L, Kugler MC, Loomis CA, Samdani R, Zhao Z, Chen GJ, et al. Hedgehog signaling in neonatal and adult lung. Am J Respir Cell Mol Biol 2013;48:703–710. |
| 29. | Palma V, Carrasco H, Reinchisi G, Olivares G, Faunes F, Larraín J. SHh activity and localization is regulated by perlecan. Biol Res 2011;44:63–67. |
| 30. | Oishi I, Suzuki H, Onishi N, Takada R, Kani S, Ohkawara B, et al. The receptor tyrosine kinase Ror2 is involved in non-canonical Wnt5a/JNK signalling pathway. Genes Cells 2003;8:645–654. |
| 31. | Li C, Xiao J, Hormi K, Borok Z, Minoo P. Wnt5a participates in distal lung morphogenesis. Dev Biol 2002;248:68–81. |
| 32. | Kishimoto K, Tamura M, Nishita M, Minami Y, Yamaoka A, Abe T, et al. Synchronized mesenchymal cell polarization and differentiation shape the formation of the murine trachea and esophagus. Nat Commun 2018;9:2816. |
| 33. | Sala FG, Del Moral PM, Tiozzo C, Alam DA, Warburton D, Grikscheit T, et al. FGF10 controls the patterning of the tracheal cartilage rings via Shh. Development 2011;138:273–282. |
| 34. | Snowball J, Ambalavanan M, Cornett B, Lang R, Whitsett J, Sinner D. Mesenchymal Wnt signaling promotes formation of sternum and thoracic body wall. Dev Biol 2015;401:264–275. |
| 35. | Ingham PW, McMahon AP. Hedgehog signaling in animal development: paradigms and principles. Genes Dev 2001;15:3059–3087. |
| 36. | Bohnenpoll T, Weiss AC, Labuhn M, Lüdtke TH, Trowe MO, Kispert A. Retinoic acid signaling maintains epithelial and mesenchymal progenitors in the developing mouse ureter. Sci Rep 2017;7:14803. |
| 37. | Motoyama J, Liu J, Mo R, Ding Q, Post M, Hui CC. Essential function of Gli2 and Gli3 in the formation of lung, trachea and oesophagus. Nat Genet 1998;20:54–57. |
| 38. | Lin X. Functions of heparan sulfate proteoglycans in cell signaling during development. Development 2004;131:6009–6021. |
| 39. | Arikawa-Hirasawa E, Watanabe H, Takami H, Hassell JR, Yamada Y. Perlecan is essential for cartilage and cephalic development. Nat Genet 1999;23:354–358. |
| 40. | French MM, Smith SE, Akanbi K, Sanford T, Hecht J, Farach-Carson MC, et al. Expression of the heparan sulfate proteoglycan, perlecan, during mouse embryogenesis and perlecan chondrogenic activity in vitro. J Cell Biol 1999;145:1103–1115. |
| 41. | Costell M, Gustafsson E, Aszódi A, Mörgelin M, Bloch W, Hunziker E, et al. Perlecan maintains the integrity of cartilage and some basement membranes. J Cell Biol 1999;147:1109–1122. |
| 42. | Costell M, Carmona R, Gustafsson E, González-Iriarte M, Fässler R, Muñoz-Chápuli R. Hyperplastic conotruncal endocardial cushions and transposition of great arteries in perlecan-null mice. Circ Res 2002;91:158–164. |
| 43. | Datta S, Pierce M, Datta MW. Perlecan signaling: helping hedgehog stimulate prostate cancer growth. Int J Biochem Cell Biol 2006;38:1855–1861. |
| 44. | Park Y, Rangel C, Reynolds MM, Caldwell MC, Johns M, Nayak M, et al. Drosophila perlecan modulates FGF and hedgehog signals to activate neural stem cell division. Dev Biol 2003;253:247–257. |
| 45. | Li C, Hu L, Xiao J, Chen H, Li JT, Bellusci S, et al. Wnt5a regulates Shh and Fgf10 signaling during lung development. Dev Biol 2005;287:86–97. |
| 46. | Loscertales M, Mikels AJ, Hu JK, Donahoe PK, Roberts DJ. Chick pulmonary Wnt5a directs airway and vascular tubulogenesis. Development 2008;135:1365–1376. |
| 47. | Tiozzo C, De Langhe S, Carraro G, Alam DA, Nagy A, Wigfall C, et al. Fibroblast growth factor 10 plays a causative role in the tracheal cartilage defects in a mouse model of Apert syndrome. Pediatr Res 2009;66:386–390. |
| 48. | Geng Y, Dong Y, Yu M, Zhang L, Yan X, Sun J, et al. Follistatin-like 1 (Fstl1) is a bone morphogenetic protein (BMP) 4 signaling antagonist in controlling mouse lung development. Proc Natl Acad Sci USA 2011;108:7058–7063. |
| 49. | Arora R, Metzger RJ, Papaioannou VE. Multiple roles and interactions of Tbx4 and Tbx5 in development of the respiratory system. PLoS Genet 2012;8:e1002866. |
| 50. | Que J, Luo X, Schwartz RJ, Hogan BL. Multiple roles for Sox2 in the developing and adult mouse trachea. Development 2009;136:1899–1907. |
| 51. | Boucherat O, Montaron S, Bérubé-Simard FA, Aubin J, Philippidou P, Wellik DM, et al. Partial functional redundancy between Hoxa5 and Hoxb5 paralog genes during lung morphogenesis. Am J Physiol Lung Cell Mol Physiol 2013;304:L817–L830. |
| 52. | Aubin J, Lemieux M, Tremblay M, Bérard J, Jeannotte L. Early postnatal lethality in Hoxa-5 mutant mice is attributable to respiratory tract defects. Dev Biol 1997;192:432–445. |
| 53. | Vermot J, Niederreither K, Garnier JM, Chambon P, Dollé P. Decreased embryonic retinoic acid synthesis results in a DiGeorge syndrome phenotype in newborn mice. Proc Natl Acad Sci USA 2003;100:1763–1768. |
| 54. | Li C, Li A, Li M, Xing Y, Chen H, Hu L, et al. Stabilized β-catenin in lung epithelial cells changes cell fate and leads to tracheal and bronchial polyposis. Dev Biol 2009;334:97–108. |
| 55. | Bell SM, Schreiner CM, Wert SE, Mucenski ML, Scott WJ, Whitsett JA. R-spondin 2 is required for normal laryngeal-tracheal, lung and limb morphogenesis. Development 2008;135:1049–1058. |
| 56. | Caprioli A, Villasenor A, Wylie LA, Braitsch C, Marty-Santos L, Barry D, et al. Wnt4 is essential to normal mammalian lung development. Dev Biol 2015;406:222–234. |
| 57. | Mahlapuu M, Enerbäck S, Carlsson P. Haploinsufficiency of the forkhead gene Foxf1, a target for sonic hedgehog signaling, causes lung and foregut malformations. Development 2001;128:2397–2406. |
| 58. | Bonvin E, Le Rouzic P, Bernaudin JF, Cottart CH, Vandebrouck C, Crié A, et al. Congenital tracheal malformation in cystic fibrosis transmembrane conductance regulator-deficient mice. J Physiol 2008;586:3231–3243. |
| 59. | Gradus B, Alon I, Hornstein E. miRNAs control tracheal chondrocyte differentiation. Dev Biol 2011;360:58–65. |
| 60. | Rock JR, Futtner CR, Harfe BD. The transmembrane protein TMEM16A is required for normal development of the murine trachea. Dev Biol 2008;321:141–149. |
| 61. | Li M, Li C, Liu YH, Xing Y, Hu L, Borok Z, et al. Mesodermal deletion of transforming growth factor-β receptor II disrupts lung epithelial morphogenesis: cross-talk between TGF-β and Sonic hedgehog pathways. J Biol Chem 2008;283:36257–36264. |
| 62. | Whitsett JA, Clark JC, Picard L, Tichelaar JW, Wert SE, Itoh N, et al. Fibroblast growth factor 18 influences proximal programming during lung morphogenesis. J Biol Chem 2002;277:22743–22749. |
| 63. | Lin SS, Tzeng BH, Lee KR, Smith RJ, Campbell KP, Chen CC. Cav3.2 T-type calcium channel is required for the NFAT-dependent Sox9 expression in tracheal cartilage. Proc Natl Acad Sci USA 2014;111:E1990–E1998. |
| 64. | Boucherat O, Nadeau V, Bérubé-Simard FA, Charron J, Jeannotte L. Crucial requirement of ERK/MAPK signaling in respiratory tract development. Development 2014;141:3197–3211. |
| 65. | Han L, Xu J, Grigg E, Slack M, Chaturvedi P, Jiang R, et al. Osr1 functions downstream of Hedgehog pathway to regulate foregut development. Dev Biol 2017;427:72–83. |
Supported by NIH grants R03 HL133420 and R01 HL144774 (D.I.S.), U54 HL127672 (B.C.T.), and U01 HL122642 and U01 HL134745 (J.A.W.); by NIH grant GH006888 (J.B.H.); by the U.S. Department of Veterans Affairs (K.M.K. and J.B.H.); and by the Translational Pulmonary Science Center and Division of Pulmonary Biology, Cincinnati Children’s Hospital Medical Center.
Author Contributions: B.C.T. designed and led the clinical study. B.C. and D.Z. conducted study visits, collected clinical specimens, and prepared DNA. K.M.K. and J.B.H. organized the collection, management, and initial analysis of nucleotide sequence data. R.E.W., M.J.R., A.d.A., and R.G.E. provided patient care and/or human tracheal specimens. D.I.S. designed, led, and conducted the studies in mice and histological evaluation of human tracheas with assistance from L.L. D.I.S., J.A.W., and B.C.T. interpreted data and drafted the manuscript. All authors participated in writing the manuscript and approved the final version.
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org.
Originally Published in Press as DOI: 10.1164/rccm.201809-1626OC on June 19, 2019
Author disclosures are available with the text of this article at www.atsjournals.org.
