The PAX genes, including PAX1, are a highly conserved family of developmental control genes that encode transcription factors and have been shown to play a role in pattern formation during embryogenesis in vertebrates (summary by McGaughran et al., 2003).

See PAX7 (167410) for a discussion of paired box domain genes.

Deutsch et al. (1988) identified a 3.1 kb-Pax 1 transcript during murine embryonic development, whereas no transcripts were detected in adult mouse tissues. Detailed in situ hybridization analysis with frozen embryonic sections demonstrated Pax1 transcripts in the perichordal zone of the developing vertebral column.

Burri et al. (1989) isolated and sequenced 3 human genes that contained paired domains with strong homology to those in genes in Drosophila involved in programming early development. One of these genes, called HuP48, had a paired domain similar to that of the Drosophila gene P29 and nearly identical to that of the mouse Pax1 gene.

McGaughran et al. (2003) determined the complete sequence of the human PAX1 gene. The deduced 440-residue protein has a calculated molecular mass of 45.7 kD. PAX1 has an extra 70-amino acids 5-prime to the paired box domain compared to mouse Pax1. They noted that PAX1 shows greatest similarity to PAX9 (167416).

McGaughran et al. (2003) determined that the PAX1 gene contains at least 4 exons and spans approximately 10 kb. Adham et al. (2005) stated that the PAX1 gene contains 5 exons.

Schnittger et al. (1992) demonstrated that the human PAX1 locus is situated on chromosome 20p. The map position of PAX1 after fluorescence in situ hybridization FL-pter value of 0.34 +/- 0.04 corresponds to band p11.2. (FL-pter = fractional length-pter (Lichter et al., 1990).) The mean FL-pter value for the centromere of the same chromosome 20 identified by Q-banding was 0.46 +/- 0.04. By PCR analysis in somatic cell hybrids, Pilz et al. (1993) confirmed the assignment of the PAX1 gene to chromosome 20. By analysis of somatic cell hybrids and by fluorescence in situ hybridization, Stapleton et al. (1993) mapped PAX1 to chromosome 20p11.

The mouse Pax1 gene was mapped by linkage analysis to distal mouse chromosome 2 in close proximity to 'undulated,' between the beta-2-microglobulin and 'agouti' loci (Balling et al., 1988). This segment of mouse chromosome 2 exhibits homology to human chromosome 20. Segmental trisomy (Francke, 1977) and monosomy (Schnittger et al., 1989; Anad et al., 1990) for chromosome 20p have been associated with anomalies of intervertebral discs.

The mouse Pax-1 gene encodes a sequence-specific DNA-binding protein with transcriptional activating properties (Deutsch et al., 1988; Chalepakis et al., 1991). The expression pattern of Pax1 during mouse embryogenesis indicates that it may play an important role in the development of the vertebral column.

Smith and Tuan (1994) detected expression of PAX1 in the fetal human vertebral column. Seven- to 8-week-old fetuses showed staining in mesenchymal cells making up the body of the intervertebral disks. Adjacent chondrocytes were PAX1-negative. PAX1 expression was also not detected in developing limbs. Ten- to 12-week-old fetal specimens no longer showed PAX1 expression. Smith and Tuan (1994) suggested that PAX1 is important for the proper formation of the segmented vertebral column in humans.

Using RT-PCR, Pohl et al. (2013) observed expression of Pax1 in the P6 mouse cochlea, supporting a role for PAX1 in the hearing process.

By whole-exome sequencing in affected members of a large consanguineous Turkish family with otofaciocervical syndrome (OTFCS2; 615560), Pohl et al. (2013) identified homozygosity for a missense variant in the PAX1 gene (G166V; 167411.0001) that segregated with disease in the family and was not found in more than 13,000 alleles from the Exome Variant Server database. Functional analysis indicated a reduced DNA-binding affinity of the mutant protein.

By whole-exome sequencing, Paganini et al. (2017) identified a homozygous nonsense mutation in the PAX1 gene (C368X; 167411.0002) in 2 distantly related Moroccan infants with OTFCS2 who also had thymic aplasia and died of severe combined immunodeficiency (SCID). Sanger sequencing confirmed the findings. The consanguineous parents from each sibship were heterozygous for the mutation, as was one of the affected infant's sibs who presented with preauricular pits.

In 2 sibs, born to first-cousin parents from India, with OTFCS2, Patil et al. (2018) identified a homozygous frameshift mutation in the PAX1 gene (167411.0003). The mutation, which was found by whole-exome sequencing and confirmed by Sanger sequencing, was present in heterozygous state in the parents.

In a 6-year-old German boy with OTFCS2, who had no thymus shadow on chest x-ray and showed significant T-cell lymphopenia, Yamazaki et al. (2020) identified homozygosity for an in-frame deletion in the PAX1 gene (167411.0004). In 4 similarly affected children from 2 sibships of a consanguineous Saudi Arabian pedigree, the authors identified homozygosity for a PAX1 missense mutation (V147L; 167411.0005). Functional studies showed reduced transactivation activity with the mutants compared to wildtype protein. Gene set enrichment analysis showed reduced expression of various genes involved in development of the thymus and T-cell commitment in patient thymic epithelial progenitor cells (TEPs) versus controls, indicating that multiple mechanisms contribute to the thymic defects associated with PAX1 deficiency. In addition, the authors observed differential expression of genes included in gene sets involving neural crest, ear, cartilage, pharyngeal, and skeletal development during differentiation of patient TEPs compared to controls.

Exclusion Studies

Smith and Tuan (1994) did not identify any mutations in the paired box sequence of the PAX1 gene in 3 female patients with Klippel-Feil anomaly (118100).

McGaughran et al. (2003) identified sequence abnormalities in the PAX1 gene in 8 of 63 individuals with Klippel-Feil anomaly. Two of the sequence variants were detected in controls, and several other variants were present in unaffected family members. The authors suggested that changes in PAX1 may play a role in the pathogenesis of the disorder.

Hol et al. (1996) identified a sequence variant in the PAX1 gene in a patient with spina bifida (see 182940); however, the mutation was also present in the unaffected mother and maternal grandmother, suggesting that additional factors are required for the phenotype.

Giampietro et al. (2005) identified a heterozygous change in exon 4 of the PAX1 gene in 2 of 48 patients with congenital vertebral malformations. One patient had T9 hypoplasia, T12 hemivertebrae, absent T10 pedicle, incomplete fusion of T7 posterior elements, ventricular septal defect, and polydactyly. He had a heterozygous pro413-to-leu (P413L) substitution, which was found in 0.3% of controls. A second patient, with a T11 wedge vertebra, had a heterozygous pro410-to-leu (P410L) substitution in the PAX1 gene, which was found in 0.8% of controls. Each patient's asymptomatic mother was heterozygous for the respective allele, and neither change occurred in a conserved region of the protein. Giampietro et al. (2005) stated that the most likely interpretation was 'that these variants have no functional significance' and are most likely 'not causally related to vertebral defects observed in their patients.' No functional expression analysis was reported.

The autosomal recessive mutation 'undulated' (un) in the mouse exhibits vertebral anomalies along the entire rostrocaudal axis. It was first described by Wright (1947). Balling et al. (1988) determined that the phenotype is caused by a G-to-A transition in the Pax1 gene, resulting in a gly15-to-ser (G15S) substitution in a highly conserved region of the paired box domain. The substitution dramatically decreases the DNA-binding affinity of the 'un' Pax1 protein and alters its DNA-binding specificity (Chalepakis et al., 1991).

Helwig et al. (1995) reported that mice who are doubly heterozygous for the mutants 'undulated' and 'Patch' (see PGDFRA; 173490) have a phenotype reminiscent of an extreme form of spina bifida occulta in humans (see 182940). The unexpected phenotype in double-mutant and not single-mutant mice showed that novel congenital anomalies such as spina bifida can result from interaction between products of independently segregating loci. This is an example of digenic inheritance.

The Pax1 gene is required for the normal development of 3 skeletal elements in the mouse: the vertebral column, sternum, and scapula. Three natural Pax1 mouse mutants exhibit phenotypes of different severity in these 3 elements. The phenotypes are referred to as 'undulated' (un); 'undulated extensive' (un-ex); and 'undulated short-tail' (Un-s). The 'undulated extensive' allele (un-ex) carries a deletion that includes the last exon of the Pax1 gene; however mutant Pax1 mRNA can be detected by RNase protection assay. In the 'undulated short-tail' allele (Un-s), the complete Pax1 locus is deleted. The first 2 mutations are thought to be recessive because severe skeletal malformations are found only in homozygous animals. Nevertheless, mild skeletal abnormalities have occasionally been described in heterozygotes of both alleles. In contrast, the Un-s mutation is semidominant as heterozygotes exhibit clear skeletal abnormalities including very short and strongly kinked tail. Homozygous Un-s mice die perinatally displaying the most severe skeletal malformations among the 'undulated' alleles (Wilm et al., 1998). Wilm et al. (1998) inactivated the Pax1 gene by gene targeting and found considerable differences between the phenotypes of the Pax1 knockout and the heterozygous Un-s mutant mice. The result was interpreted as indicating the contribution of an additional gene or genes to the Un-s mutant phenotype. Although heterozygous Pax1-null mice appeared externally normal, Wilm et al. (1998) demonstrated that almost 90% exhibited moderate skeletal malformations in the sternum and parts of the vertebral column. In contrast, the scapula and tail were normal in all heterozygotes, showing recessiveness of the null mutation in these structures. Previous studies on the skeletal phenotypes of the 'undulated' and 'undulated extensive' mutations had shown weak abnormalities in the sternum and in parts of the vertebral column in heterozygotes. Thus, haploinsufficiency of Pax1 in mice can also result in phenotypic abnormalities.

Adham et al. (2005) found that the autosomal recessive spontaneous mouse mutant 'scoliosis' (sco) carries a new Pax1 allele, 'undulated intermediate' (un-i). The un-i allele contains 2.0-kb and 4.5-kb deletions corresponding to lack of the 5-prime flanking region and exons 1 to 4, respectively. Homozygous mice show lumbar scoliosis and a kinky tail, which is a milder phenotype compared to other undulating mutations. Heterozygous animals were indistinguishable from wildtype. In homozygous embryos, the skeleton ossified early, ossification centers of the vertebral bodies were fused with ossification centers of the pedicles, and neural arches and spinous processes were underdeveloped. In the scapula, the acromion was missing, and the deltoid tuberosity of the proximal humerus was shortened and thickened. The thymus was also underdeveloped.