Subjects and methods

FAMILIES

The present report includes data from five families of Arabic origin, four from Saudi Arabia and one from Israel (family 89). DNA was available from a total of 32 subjects (12 affected, five of whom have situs inversus). The parents in all families are first cousins except in family 72 from Saudi Arabia. Diagnosis of affected subjects was made according to well established clinical criteria (table 1) supported by abnormal ciliary motility studies and electron microscopy (EM) of brushings or biopsy of nasal mucosa. An ultrastructural phenotype of absent outer dynein arms was shown in affected subjects where available.

MARKER TYPING

Genomic DNA was extracted from white cells by standard methods. DNA was amplified by PCR using fluorescently labelled primers. PCR was performed in 96 well microtitre plates (Hybaid). Each well contained 30-60 ng of genomic DNA, 1.5-2 mmol/l MgCl₂, 10 × reaction buffer (GIBCO), 200 mmol/l each of dATP, dTTP, dGTP, and dCTP, 50 ng of each primer, and 0.05 units of Taqpolymerase (GIBCO), in a total volume of 20 μl. Thirty cycles were performed in a thermocycler (Hybaid Omnigene^TM). Alleles were separated on a 6% polyacrylamide electrophoresis gel for three to four hours at a rate limiting voltage of 1000 V with a model 373XL DNA sequencer (Applied Biosystems). Analysis of the allele sizes was performed by GENESCAN^TM 672 (v 1.2) and Genotyper software (Applied Biosystems) using GENESCAN^TM -500 TAMRA as a size standard.

LINKAGE ANALYSIS

Linkage analysis was performed under the assumption of autosomal recessive inheritance with a penetrance of 0.9, disease allele frequency of 0.015, and a phenocopy rate of 0.001. Multipoint analysis was performed using the GENEHUNTER program.26 The marker loci from chromosome 19q that were included in this analysis were D19S572 - 4 cM - D19S418 - 3 cM - D19S605 - 5 cM - D19S404 - 2 cM -D19S214 - 1 cM - D19S218. Each marker was assumed to have five alleles of equal frequency. The relative positions of marker loci (fig 2) are based on recently published maps and chromosome 19 specific databases (Lawrence Livermore National Laboratory).27-29

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Pedigrees and chromosome 19q haplotypes. The genetic distance between D19S572 and D19S218 is approximately 15 cM (sex averaged). Regions of homozygosity by descent in affected subjects are highlighted in bold. Deduced marker data are shown in italics. Subjects with situs inversus are indicated by *.

Results

Over 300 microsatellite markers were typed and excess homozygosity was identified on chromosome 19q. Further analysis of high density mapping data using GENEHUNTER gave a maximum multipoint lod score (HLOD) of 4.4 at D19S404 and D19S218, allowing for heterogeneity, at α (proportion of linked families) = 0.7 on chromosome 19q13.3-qter. The HLOD remains over 4.0 from D19S572 to D19S214 with an α which varies only from 0.65 to 0.7. Haplotype analysis confirms linkage in families 66/67, 72, and 89 and non-linkage in families 65 and 69. Obligate recombinants at D19S572 and loss of homozygosity at D19S218 in family 89 define a critical region of about 15 cM. There is so far no evidence of linkage disequilibrium as no common haplotype or allele is present among affected subjects in the linked families. In family 65, the two affected sibs have inherited different parental chromosomes and in family 69 the affected subject shares common parental chromosomes with her unaffected sibs. These data therefore support non-linkage in these two families. The pedigrees and chromosome 19q haplotypes are shown in fig 2.

Discussion

This is the first report of the mapping of a locus for primary ciliary dyskinesia. Primary ciliary dyskinesia (PCD) is not an uncommon condition with an estimated incidence of about 1 in 20 000 live births. This is likely to be an underestimation because of the difficulties in making the diagnosis. Some of the clinical features such as otitis media, sinusitis, and recurrent pneumonia are common and are shared with other conditions including cystic fibrosis, immunodeficiency, and asthma. The presence of situs inversus obviously prompts consideration of the diagnosis and allows it to be made more confidently. Confirmation ideally requires nasal brushings/biopsy which is only available at specialised centres and is poorly tolerated by children. Disease severity also varies and some subjects may only manifest very minor or few symptoms.

The affected subjects in our families have been investigated appropriately (table 1). Investigations to exclude cystic fibrosis and immunodeficiency have been performed in most patients. The presence of bronchiectasis has been documented by x ray or CT scan. All families except 69 have at least one affected subject with situs inversus. The affection status can be firmly established on the available evidence from the clinical features and laboratory investigations. The parameters of our linkage analysis are appropriate, assuming a highly penetrant recessive model with a phenocopy rate of 0.001. The disease allele frequency of 0.015 was chosen to correlate with the estimated incidence.

Our data provide conclusive evidence for a PCD locus on chromosome 19q in a subset of these families. Genetic heterogeneity is expected in PCD in view of its wide range of ultrastructural phenotype. Grouping of families according to ultrastructural phenotype or population of origin is therefore a sensible strategy in a disease with expected genetic heterogeneity. All five families are Arabic in origin and affected subjects have a similar ultrastructural phenotype of absent outer dynein arms. Our linkage data support the presence of genetic heterogeneity of PCD even within these families of common population origin and ultrastructural phenotype. Work is in progress to examine for evidence of linkage in this region in other families/population groups, especially those with absent outer dynein arms.

The gene will be identified using a positional candidate cloning approach. The present 15 cM critical region is defined by recombinations at D19S572 and loss of homozygosity at D19S218 in family 89. This critical region will be further refined by investigation of meioses with flanking recombinants or identification of linkage disequilibrium. The extent of linkage disequilibrium depends on the existence of an ancestral disease mutation, the age of the mutation, and the degree of inbreeding. At present no evidence of linkage disequilibrium has been detected. High density marker typing in the linked families and the rest of the family resource with absent outer dynein arms at this locus is in progress.

Existing knowledge of the biology of cilia and of LRA determination provides a relatively firm foundation for the evaluation of transcripts as potential candidates in this region. Obvious candidates will include genes encoding structural components of cilia (for example, dyneins, tubulins, and microtubule associated proteins). Chromosome 19q13.3-qter is a gene rich region and contains a large cluster of zinc finger containing (ZNF) genes. This is of potential interest as mutations inZIC3, a zinc finger transcription factor, have been identified in subjects with X linked and sporadic laterality defects.22

The association of situs inversus with PCD (Kartagener syndrome) remains unexplained. These patients have an increased risk of cardiac malformations which confers significant morbidity. The cloning oflrd, an axonemal dynein heavy chain gene, in the inversus viscerum(iv) mouse support an important role for dyneins in the LRA determination pathway.23 More recently, work on nodal cilia in mice has provided strong evidence that cilia play a primary role in initiating this process. Nodal cilia are monocilia which are located in the primitive node in the normal embryo. They lack the two central microtubules and therefore adopt the 9 + 0 ultrastructural configuration. They have been shown to be motile and to create a directed flow of extraembryonic fluid which could produce a gradient of putative morphogen along the left-right axis in the node. Mice with targeted disruption of the kinesin family memberkif3b gene, which encodes a component of a microtubule plus end directed motor involved in intraciliary transportation, exhibit absence of nodal monocilia and randomisation of left-right asymmetry.25 It is therefore plausible that any defect producing abnormal cilia including the nodal cilia will result in PCD with randomisation of left-right asymmetry.

Determination of the molecular basis of PCD will provide information relevant both to clinical practice and basic biomedical science. New non-invasive approaches to diagnosis and carrier detection will become available and ultimately new methods of treatment should emerge. Gene therapy, for example, is a feasible long term option in a recessive disorder dominated by respiratory tract pathology. Moreover, it is possible that allelic variation at these loci causes a broader phenotype than presently recognised, including male subfertility or chronic secretory otitis media. The identification of the protein products encoded by PCD genes should provide new insights into the molecular mechanisms involved in the assembly and function of cilia and contribute to understanding of the pathway which determines the left-right axis in vertebrates.