MATERIALS AND METHODS

Families and patients

Two unrelated Iranian families with isolated congenital anosmia (families 1 and 2), reported previously,⁴ were analysed in the present study. Family 1 consisted of three affected members in three generations, and family 2 contained nine patients in two generations. A total of 54 members of the two families agreed to participate in the present study with written informed consent. Of the 54 members, seven were diagnosed as having isolated congenital anosmia (fig 1). By clinical history, physical examination, and smelling testing by intravenous injection of combined vitamins (Alinamin™, Takeda Pharmaceutical Co Ltd, Japan), the disease was confirmed in each affected member. All affected people had a history of never being able to smell, and had never had hypogonadism signs, nasal polyposis, rhinoscleroma, or any underlying infections or other neurological disorders. Clinical manifestations of deceased persons (I-1 in family 1, and I-2, I-10, and II-13 in family 2) were presumed by their relatives’ recollection.

• Download figure
• Open in new tab
• Download powerpoint

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Pedigrees of families 1 and 2 with haplotypes at marker loci on chromosome 18p11.23-q12.2. Solid squares and circles show individuals with isolated congenital anosmia. Numbers in open boxes show a possible disease associated haplotype. Heavy short lines indicate definite recombination sites, and heavy double short lines depict recombination sites that could have occurred on either side of the corresponding markers. Haplotypes of I-1 in family 1, and I-2 and I-3 in family 2 are deduced from those in their offspring. Asterisks indicate alleles showing novel microsatellite mutations.

Genotyping, linkage analysis, and sequencing

Genomic DNA was extracted from whole blood of the 54 family members, and PCR was performed using standard methods. The amplified samples were genotyped at 400 microsatellite marker loci with an average distance of 10 cM in the genome (Linkage Mapping Set version 2 (LMS-MD10, Applied Biosystems, Foster City, CA) on a DNA sequencer (Model 377, Applied Biosystems), and marker alleles were assigned using software, Genescan™ and Genotyper™ (Applied Biosystems). When data assigned to a chromosomal region were obtained, additional markers around the critical region were used to confirm a linkage and narrow the region (fig 1).

Two point linkage analysis was performed under an assumption that isolated congenital anosmia in the two families is inherited in an autosomal dominant fashion with incomplete penetrance. Lod scores were calculated separately in each family, at the same penetrance, using the computer program MLINK of FASTLINK version 4.1P.^10–12 Allele frequencies at the 400 marker loci were set as 1/N (where N is the observed number of alleles), while those at seven loci (D18S452, D18S53, D18S66, D18S1108, D18S56, D18S1133, and D18S1128) around the critical region were calculated among 40 chromosomes from unrelated healthy Iranian individuals.

Eight genes or expressed sequence tags of interest; GNAL, VAPA, PTPRM, PTPN2, CABYR, RNMT, CDH2, and NOL4, were analysed for their sequences, to see whether affected members in the families had mutations in them. All their exons and exon-intron boundaries were amplified by PCR. PCR products were sequenced directly using BigDye™ Terminator ver. 3.0 Cycle Sequencing kit (Perkin-Elmer, Foster, CA, USA), and samples were run on an automated sequencer model 3100 (Applied Biosystems).

RESULTS

The genome-wide linkage analysis first ruled out the loci for Kallmann syndrome types 1 and 2 (Xp22.3 for KAL1 and 8p11.2-p11.1 for KAL2) from the candidacy for the anosmia locus in the two families. The other locus, KAL3, was also ruled out normally because of its X-linked inheritance. A high two point lod score of 3.03 was obtained at the D18S53 locus on chromosome 18. Extensive linkage analysis to define the critical region using 52 additional markers flanking D18S53 finally reached the maximum lod score of 5.14 (1.85 for family 1, and 3.29 for family 2) at D18S1108, when penetrance was set at 0.8 for both families, and the recombination fraction to be 0.00 (table 1). Haplotype analysis demonstrated that affected members in each family each had a common haplotype including alleles at 35 marker loci between D18S452 and D18S475 (fig 1). A recombination between D18S452 and D18S1163 or between D18S464 and D18S1153 was observed in a patient (IV-12) of family 1 (fig 1). However, as only two markers have been known between D18S452 and D18S1153, more precise identification of recombination sites was not possible in this patient. Two recombination events were observed in a patient (II-8) of family 2, one between D18S452 and D18S1163, and the other between D18S57 (or D18S1128) and D18S475. These findings indicated that the isolated congenital anosmia locus lies in a 45.9 cM segment between D18S452 and D18S475 at 18p11.23-q12.2. The sequence analysis of the eight genes or expressed sequence tags revealed no mutation in any affected members of the two families.

DISCUSSION

Isolated congenital anosmia in our two families is compatible with an autosomal dominant pattern of inheritance with incomplete penetrance. Individuals II-5 in family 1, III-28 and III-37 in family 2 are definite non-penetrants, who inherited the disease-associated haplotype but had a normal sense of smell, as determined both by past medical history and a smelling test. The isolated congenital anosmia gene was transmitted by one of them (II-5 in family 1) to his affected son (III-16), but the other two non-penetrants did not have any children. Five unaffected members (III-4 and III-14 in family 1, and III-16, III-22, and III-27, in family 2) inherited a part of the critical haplotype (fig 1). However, it remains unknown who are non-penetrants or normal individuals, because none had any affected children. Inheritance patterns in family 2 appeared unusual; especially those in its third generation in which none of a total of 25 children of patients were affected (fig 1). This seemingly strange observation may be explained by incomplete penetrance, or just by chance as suggested in our previous report,⁴ or the existence of a modifier gene.

We have assigned the isolated congenital anosmia locus in the two families to the 45.9 cM interval between D18S452 and D18S47 at 18p11.23-q12.2. However, this interval is the maximum extent estimated under an assumption that all the seven unaffected members carrying the critical haplotype were non-penetrants or carriers. The interval might be reduced to 31.2 cM, if the recombination in IV-12 of family 1 had occurred between D18S464 and D18S1153 (fig 1). Among several genes and expressed sequence tags that are expressed in the central nervous system and have been assigned within the isolated congenital anosmia region, the GNAL (or the G protein α subunit, G (olf)) gene (guanine nucleotide binding protein, α activating activity polypeptide, olfactory type) located between D18S53 and D18S71 is one of the most interesting candidate genes for isolated congenital anosmia, since mice homozygous for a null mutation in Gnal are anosmic, showing reduced electrophysiological response of the primary olfactory sensory neurones to odours.¹³ However, by our preliminary study of affected members of the two families, no mutation has been found in this gene, or in seven other genes, VAPA, PTPRM, PTPN2, CABYR, RNMT, CDH2, and NOL.4, all located within the critical region.

The sense of smell is one of the first developed special sense organs in animals¹⁴ and plays a role in feeding, differentiation of a parent from others, and in reproductive behaviour.¹⁵ Male mice do not show mating behaviours or sexual development when their olfactory bulbs are removed shortly after birth.^16,17 The role of the human olfactory system in reproduction and sexual development is supported by clinical features of Kallmann syndrome in which congenital anosmia is associated with hypogonadotropic hypogonadism. The existence of familial isolated congenital anosmia with a normal reproductive system, as has been shown in affected members of our families, may indicate the independent migration of the olfactory system from the LH-RH cells, and suggest that other factors may influence the relationship between the reproductive and olfactory systems. Isolation of the putative isolated congenital anosmia gene and its functional analysis will contribute to the better understanding of the development of the olfactory system.