Hartnup Disorder

A number sign (#) is used with this entry because Hartnup disorder (HND) is caused by homozygous or compound heterozygous mutation in the SLC6A19 gene (608893) on chromosome 5p15.

Clinical Features

First described by Baron et al. (1956), this disorder is characterized by a pellagra-like light-sensitive rash, cerebellar ataxia, emotional instability, and amino aciduria. Scriver et al. (1985) suggested the existence of 2 forms of Hartnup disease: in the classic form the defect is expressed in both intestine and kidney; in a variant form it is expressed only in kidney. In the United States, cases of the full-blown clinical disorder are not seen, probably because of super-adequate diet.

Mahon and Levy (1986) reported on the childbearing experience in unrelated women with what they called Hartnup disorder and defined as an inborn error of neutral amino acid transport. Two living, unaffected offspring, born after untreated and uneventful pregnancies, one from each woman, had had normal development. This led Mahon and Levy (1986) to conclude that, unlike PKU, Hartnup disorder has no ill effects on the fetus. Normal ratios of amino acid concentrations between maternal and umbilical veins suggested that placental transport of free amino acids, unlike renal transport, is not reduced.

Nozaki et al. (2001) studied 2 Japanese families with first-cousin parents. The proband in the first family died of cirrhosis of the liver at the age of 42 years. Hartnup disorder was diagnosed on the basis of consistent patterns of monoamino-monocarboxylic aciduria and defects in the intestinal absorption of monoamino-monocarboxylic acids as determined by oral loading. He periodically had psychologic symptoms, ataxia, and diplopia. Niacin administration resolved these clinical symptoms. The patient was reported by Mori et al. (1989) as having no skin lesions. An older brother died at the age of 32 years of a progressive neurologic disease of unexplored pathogenesis. He was reported to have had mental retardation, periodic gait disturbances, and a skin rash on exposure to sunlight. A third son, younger than the other 2, had no health problems and was average scholastically in school. At age 33 years, large amounts of indican were found in his urine and he underwent oral amino acid loading showing decreased tryptophan absorption from the gut and suggesting that he was a Hartnup disorder carrier. In the second family, the proband had an eczematous skin rash at the age of 3 weeks and chronic diarrhea at the age of 3 months. Oral loading tests demonstrated impaired absorption from the gut, while the absorption of proline was intact. Liver biopsy showed extensive fatty liver without infiltration of inflammatory cells or fibrosis.

Biochemical Features

The defect in Hartnup disorder involves the intestinal and renal transport of certain neutral alpha-amino acids (Scriver, 1965). Seakins and Ersser (1967) described a patient in whom the intestinal transport defect was partially evident only under loading conditions. Lysine transport was impaired, whereas histidine transport was not. Studying uptake of amino acids by biopsied intestinal mucosa cells in vitro, Shih et al. (1971) found marked reduction in transport of methionine and tryptophan. Minimal reduction in transport of lysine and glycine correlated with the modest increases of these amino acids in the urine. Stool indoles and urinary indican were elevated after oral tryptophan loading. Occurrence of both Hartnup disease and methylmalonic aciduria in 2 families was considered coincidental (Shih et al., 1984).

Schmidtke et al. (1992) provided a detailed study of an affected girl who died in status epilepticus at age 2 years. The girl had a severe encephalopathy with an unusual pattern of cerebral gray and white matter pathology. Neurochemically there was evidence for impaired myelin formation. The authors considered whether this was a coincidence of a separate encephalopathy of unidentified type or an extreme form of the usually mild encephalopathy seen in the Hartnup syndrome. The child also had bisalbuminemia (103600) which was inherited from the mother.

Population Genetics

Hartnup disease was found to have about the same frequency in Massachusetts as phenylketonuria, i.e., 1 in 14,219 births (Levy et al., 1972).

Inheritance

Pomeroy et al. (1968) reported the first cases of affected persons (1 male, 1 female) who had children. In Colombia, Lopez et al. (1969) described 2 affected brothers whose parents were double second cousins. Two other deceased brothers were probably affected.

Heterogeneity

Genetic heterogeneity probably exists because cases have been described in which only the urinary characteristics of Hartnup disease were present, and there was no evidence of an intestinal transport defect (Srikantia et al., 1964).

Scriver et al. (1987) suggested that Hartnup disorder is multifactorial. They compared developmental outcomes and medical history of 21 Hartnup subjects, identified through newborn screening, with those of 19 control sibs. They found 2 tissue-specific forms of the Hartnup transport phenotype: renal and intestinal involvement in 15 families and renal involvement alone in 1 family. They concluded that whereas deficient activity of the Hartnup transport system was monogenic, the associated plasma amino acid value is polygenic. In general, they found no significant difference between means of the summed plasma values for amino acids affected by the Hartnup gene in the 2 groups. The 2 Hartnup subjects with clinical manifestations (impaired somatic growth and IQ in one, impaired growth and a 'pellagrin' episode in the other) had, however, the lowest summed plasma amino acid values in the Hartnup group. Furthermore, they found that the corresponding values for these patients' sibs were the low outliers in the control group. They concluded that there is a polygenic determination of amino acid values and that superimposed expression of the Hartnup gene increases liability for the disease.

Mapping

By homozygosity mapping, Nozaki et al. (2001) assigned the Hartnup disease locus to chromosome 5p15.

Seow et al. (2004) confirmed the assignment and narrowed the cytogenetic location to 5p15.33.

Molecular Genetics

Using homozygosity mapping and fine mapping in the consanguineous English family in which Hartnup disorder was originally discovered, Kleta et al. (2004) confirmed previous results showing linkage to 5p15. Two members of the SLC6 family of transporters mapped to the mouse chromosomal region that is homologous to human 5p15: SLC6A18 (610300) and SLC6A19. Both show abundant expression in mouse kidney, as assessed by real-time RT-PCR. Immunohistochemistry confirmed expression of mouse B(0)AT1 at the brush border of small intestine and kidney proximal tubule cells. As a primary candidate for the gene causing Hartnup disorder, the human homolog, SLC6A19, was sequenced in 6 families, with identification of 5 mutations in 4 families. In the family in which Hartnup disorder was first described, a homozygous splice site mutation, IVS8+2T-G (608893.0001), segregated with the disease phenotype.

In Australia, Seow et al. (2004) likewise identified the SLC6A19 gene as the site of mutations causing Hartnup disorder. They identified 6 mutations that cosegregated with the disease in the predicted recessive manner, with most affected individuals being compound heterozygotes. A common mutation, 517G-A (608893.0003), showed a population frequency of 0.007; the second most frequent mutant allele, 718C-T (608893.0004), had an estimated frequency of 0.001. An analogy was drawn to cystic fibrosis (219700) in relation to the distribution of the disorder: the 517G-A mutation is relatively frequent, and homozygosity was estimated to occur at a rate of approximately 1 in 20,000. This rate is consistent with the frequency of Hartnup disorder in European populations.

Animal Model

Symula et al. (1997) mapped hyperphenylalaninemia 2 (hph2), a recessive mutation in the mouse that causes deficient amino acid transport similar to Hartnup disease. The hph2 mouse locus was mapped in 3 separate crosses to identify candidate genes and a region of homology in the human genome where they proposed that the human disorder may map. The gene maps to mouse chromosome 7 close to a marker in the fibroblast growth factor-3 gene (164950) which in the human is located on 11q13. The mouse mutant was isolated after N-ethyl-N-nitrosourea (ENU) mutagenesis on the basis of delayed plasma clearance of an injected load of phenylalanine. Symula et al. (1997) found that animals homozygous for the mutation excrete elevated concentrations of many of the neutral amino acids in urine, while plasma concentrations of these amino acids are normal. In contrast, mutant homozygotes excrete normal levels of glucose and phosphorus. Symula et al. (1997) presented experiments indicating that the mouse disorder is a model for heart disease: the urine amino acid profiles were similar; in both species, there was a deficiency in brush-border amino acid transport; and both displayed a niacin-reversible syndrome influenced by diet and genetic background.