Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Introduction

Hereditary tyrosinemia type 1 (HT1) (OMIM 276700) is an inherited metabolic disease, mainly of childhood. This pathological condition was referred to as hereditary tyrosinemia type 1 in the mid-1960s (reviewed in Mitchell et al. 2001; Russo et al. 2001), and it was later shown to result from a deficiency in fumarylacetoacetate hydrolase (FAH), the last enzyme of the tyrosine catabolic pathway (Lindblad et al. 1977; Fällström et al. 1979; Berger et al. 1981; Kvittingen et al. 1981; Tanguay et al. 1990).

HT1 is an autosomal recessive disease characterized by severe liver dysfunction, impaired coagulation, neurological crises, renal tubular dysfunctions and a high risk of hepatocellular carcinoma (HCC). Three main clinical forms of HT1 have been described: the acute form, which presents itself in the first months of life and is associated with acute liver failure; the subacute form (second half of the first year) that manifests a similar but less severe clinical picture presenting usually with hepatomegaly or hypophosphatemic rickets (due to tubular dysfunction); and the chronic form which appears after the first year of age and shows a slower progression (Tanguay et al. 1990; van Spronsen et al. 1994; Bergman et al. 1998; Russo et al. 2001). Patients affected with HT1 generally show failure to thrive and hepatic damage including hepatomegaly, cirrhosis, hepatic failure and HCC. Complications associated with liver damage include jaundice, ascites and bleeding. HT1 also disrupts kidney function causing multiple tubular dysfunctions, Fanconi-like syndrome and glomerulosclerosis. In 1992, the introduction of NTBC (2-(2-nitro-trifluoromethylbenzoyl) 1,3-cyclohexanedione, also known as nitisinone) (Lindstedt et al. 1992) has proven to be highly effective in preventing the progression of liver damage, neurological crises and kidney damages (Larochelle et al. 2012; Bartlett et al. 2014). NTBC in combination with a low-tyrosine diet represents the only treatment available for this disease. However, one of the most severe complications occurring in HT1 patients remains the development of HCC (Mitchell et al. 2001). Indeed, although regular administration of NTBC in HT1 patients, combined with a protein-restricted diet, prevents liver and kidney dysfunction, recent reports have documented the presence of HCC even under therapy (de Laet et al. 2013). Effectiveness of this treatment depends on how early the disease is recognized and treated; thus, recent retrospective studies highly recommend the implementation of newborn screening in more areas (Zytkovicz et al. 2013; Dehghani et al. 2013; De Laet et al. 2013; Mayorandan et al. 2014). For example, Mayorandan and collaborators in their retrospective study point out the necessity of neonatal programmes borne by the government or health insurance companies to allow early diagnosis and access to adequate treatment. Indeed they report that patients, who were diagnosed after the neonatal period and consequently received NTBC treatment later, had a 2–12-fold higher risk (depending on age at start of therapy) of developing hepatocellular carcinoma compared to patients treated as neonates.

Detection of succinylacetone (SA) in urine, blood and amniotic fluid is the most reliable biochemical diagnostic for HT1. Assay of FAH enzyme activity in skin fibroblasts is possible but not readily available. Advent of molecular genetic testing has greatly improved the diagnostic power for this disease. Mutation analysis is not essential for clinical management but is useful for prenatal diagnosis and reproductive counselling. In fact targeted mutation analysis for diseased alleles and sequence analysis of the entire fah coding region can detect mutations in more than 95% of affected individuals (Sniderman King et al. 2011). The database of the GTR (Genetic Testing Registry: https://www.ncbi.nlm.nih.gov/gtr/conditions/C0268490) reports 56 clinical tests for diagnosis and monitoring of this condition. Carrier testing for at-risk relatives and prenatal diagnosis for pregnancies at increased risk are possible if both disease-causing alleles in a family are known.

Patients and Methods

The present review is based on a current compilation of all HT1 alleles reported worldwide including those from patients identified in the Laboratory of Cellular and Developmental Genetics (LGCD), Université Laval, Quebec, Canada (Dr RM Tanguay), and the Department of Clinical Chemistry at Birmingham Children’s Hospital (BCH), Birmingham, UK (Dr G Gray), mostly between 2001 and 2013 (unpublished data). Screening of genetic databases (e.g. HGMD, NCBI, ENSEMBL) and HT1 literature has been made to classify the reported mutations and to identify the ethnic group of patients. The mutations reported so far and the patients’ origins are listed in Table 1.

Table 1 Compilation of hereditary tyrosinemia type 1 alleles worldwide
Full size table

Since there are inconsistencies in the literature of names of the mutations in this gene, we have used the Human Genome Variation Society’s nomenclature for the description of sequence variations (http://www.hgvs.org/mutnomen/recs.html) as the basis of nomenclature (den Dunnen and Antonarakis 2000) and used the fah cDNA sequence given as GenBank accession number BT007160.1 as our reference sequence. For splice defects we have also added the historical mutation nomenclature, since this is the most common way in which they are named worldwide.

Results and Discussion

Fah Gene Characteristics and Mutations

The first mutation reported in the fah gene was the c.47A>T (p.Asn16Ile) in a French Canadian patient and was shown to be causative of FAH deficiency (Phaneuf et al. 1992).

The human fah gene is located on chromosome 15q23-q25, spans 30–35 kb and consists of 14 exons. The cDNA has an open reading frame of 1,257 bp encoding 419 amino acids (Phaneuf et al. 1991; Labelle et al. 1993). Identification of this gene (Phaneuf et al. 1991) led to mutation screening of patients and characterization of a number of disease-causing alleles, some of which were present at relatively high frequencies in specific populations (St-Louis and Tanguay 1997).

Eighty-three disease-causing mutations are presently reported on Human Gene Mutation Database (HGMD® Professional 2014.2, accessed in August 2014). Recently, two new mutations were uncovered at LGCD, Quebec, and in BCH, Birmingham (unpublished data). The first was the c.726G>A (p.Trp242X) nonsense mutation, obtained by screening one English adult patient at BCH. The second, the c.775G>C (p.Val259Leu) a potential missense mutation, was observed in an American patient at the Quebec laboratory. This patient was heterozygous for the new c.775G>C (p.Val259Leu) allele and the already reported c.554-1G>T (IVS6-1G>T) (Grompe et al. 1994). Western blot analysis of his liver obtained after transplantation revealed the absence of FAH protein and no activity was detected by enzymatic assay (data no shown). RNA analysis suggested a defect in splicing affecting exon 9, and this was confirmed using minigene constructs transfected in HeLa cells (Dreumont and Tanguay, unpublished).

Reclassification of HT1 Mutations

After cross-checking of genetic databases and the literature on HT1 from the oldest publications to the present day, we updated the number of allelic variants with the two found by our group and others recently reported (Fig. 1 and Table 1). Next we decided to reclassify them in a unique list containing number of known alleles from patients and geographical distribution of the mutations most predominant for each country (Fig. 2, and Table 1). Indeed, frequency of reported alleles and origin of patients could be useful in helping clinicians to focus on mutations specific of certain regions, facilitating the targeted detection of diseased alleles.

Fig. 1
figure 1

Location of the 95 mutations identified on the fah gene. Among the known HT1 alleles causing mutations, 45 are missense mutations, 23 are splicing mutations, 13 are nonsense mutations, 10 are deletions and 4 are frameshift. Intronic mutations are illustrated at the bottom of the figure

Full size image
Fig. 2
figure 2

Geographical distribution of the most common HT1 alleles causing mutations worldwide. Pie chart representing distribution of ethnic groups in HT1 alleles. Where the patient provenance was not clear, the mutation is included in the continent of origin, i.e. undefined (un) on graphic. The top three mutations and the total number of alleles for each continent are reported. There are more than 894 HT1 alleles reported worldwide. The most frequent HT1 mutation encountered is the IVS12+5G>A splice mutation, which accounts for 33.7% of all HT1 alleles, followed by the IVS6-1G>T mutation (16.4%). The French Canadian population alone accounts for as much as a third of all HT1 alleles reported. Both mutations are the most reported globally

Full size image

Overall 95 mutations are now reported within the fah gene in this review (Fig. 1 and Table 1). All 95 HT1 alleles are divided in 45 missense mutations, 23 splice defects, 13 nonsense mutations, 10 deletions and 4 frameshift (Table 1). In addition the missense c.1021C>T (p.Arg341Trp) sequence variant is described as a pseudodeficiency variant since individuals homozygous for this mutation are healthy (Rootwelt et al. 1994b; Bergeron et al. 2001).

Predominance of Ethnic Groups in HT1 Distribution

Despite the fact that the worldwide incidence of HT1 is relatively low with one affected individual in approximately 100,000 healthy individuals (Hutchesson et al. 1996), specific populations stand out as they represent small clusters of diseased alleles (Fig. 2). The population that possesses the highest incidence of HT1 is the French Canadian population of the SLSJ region, in the province of Quebec (Canada) (De Braekeleer and Larochelle 1990; Poudrier et al. 1996). The prevalence of HT1 in the SLSJ region was as high as 1/1,042 births in 1971 but dropped to 1/1,846 births in 1986, most likely due to the implementation of a screening programme for HT1 in 1970 conducted by the Quebec Network of Genetic Medicine. The most predominant mutation in this region is the c.1062 + 5G>A (IVS12 + 5G>A) accounting ~90% of all the disease-causing alleles. Furthermore, even though the Quebec population accounts for only approximately 0.12% of the world population (estimated today to number of 7 billion), it represents ~33% of all HT1 alleles worldwide. Although these data may be biased by the fact that all newborns in Quebec are screened for HT1, it is clear that this region represents the highest incidence of HT1 and that the c.1062 + 5G>A mutation is predominant in this region.

A second cluster of HT1 is found in Scandinavia (Kvittingen et al. 1981). In the Finnish population of Pohjanmaa, 1 individual out of 5,000 is affected with HT1 (St-Louis et al. 1994), whereas the overall incidence of HT1 in Finland is 1:60,000 (Mustonen et al. 1997). In this region, one single mutation (c.786G > A, p.Trp262X) represents ~88% of all reported HT1 alleles (St-Louis et al. 1994). Indeed 40 of the 46 European c.786G>A alleles have been reported in this country.

New findings show a peculiar pattern of HT1 mutations also in Norway. In a recent report, 19 Norwegian HT1 patients were investigated in the Hospital of Oslo University and three new small deletions were found: c.615delT, (p.Phe205LeufsX2), c.744delG (p.Pro249HisfsX55) and c.835delC (p.Gln279ArgfsX25). The novel mutations lead to frameshift and premature termination codons. FAH protein structure is affected, and normal folding, function and stability of the protein cannot be expected (Bliksrud et al. 2012). The c.615delT, c.744delG and the c.835delC are found in 13.5%, 3.8% and 1.9% of the alleles, respectively. Around 65% of the Norwegian HT1 patients are heterozygous for different mutations. The relatively high incidence of HT1 in Norway (1 in 74,800 live births) has not been connected with a single founder effects or high incidence of parental consanguinity as in the previous areas (Bliksrud et al. 2012).

Another cluster occurs in an immigrant population from Pakistan living in the UK, predominantly in Birmingham (Hutchesson et al. 1998). Birmingham is a city in the West Midlands Region, which has a total population of approximately 5.3 million of which nearly 3% are of Pakistani origin. We have diagnosed 44 patients from the West Midlands with this disorder of which 30 (68%) were of Pakistani origin. This is over 22-fold higher than the frequency of people of Pakistani origin in this region. Mutation analysis revealed that five out of 12 index patients (42%) in this ethno-geographic group had the c.192G>T (p.Gln64His) mutation. This mutation was not detected in patients from any other close-by region suggesting a founder effect from the region of origin of this population. Indeed the frequency of this mutation in Pakistani from the UK was comparable to that of the common pan-ethnic c.1062 + 5G>A mutation.

Most Frequent HT1 Alleles Around the World

Although Quebec, Finland, Norway and Pakistani in the UK stand out as populations with the higher frequency of HT1, reports highlight a specific tendency in mutational distribution among ethnic groups. The c.1062 + 5G>A (IVS12 + 5G>A) mutation is found frequently in patients from a wide range of ethnic groups over a large geographical distribution. Given the high frequency and wide spread of this mutation, it is likely to be a very old mutation and it was originally reported in a French Canadian patient and in two patients of Iranian origin (Grompe et al. 1994). Although this is the most frequent HT1 mutation encountered worldwide (302/894 HT1 alleles), the c.554-1G>T (IVS6-1G>T) splice mutation is also frequently observed (147/894 HT1 alleles), showing a high prevalence in the Mediterranean region and in southern Europe. In a recent cross-sectional retrospective study on 168 HT1 patients originating from Europe, Turkey and Israel, mutational analysis performed in 58/168 patients revealed the predominance of the IVS12 + 5G>A (11 patients) and IVS6-1G>T (13 patients) mutations in these ethnic groups (Mayorandan et al. 2014).

The mutation that ranks third in prevalence in Europe is the c.786G>A (p.Trp262X) nonsense mutation. This ranking is due to its predominance in the Finnish population. A number of others mutations have also been associated with specific ethnic or geographic groups, as described below (Table 1, Fig. 2). The c.1062 + 5G>A (IVS12 + 5G>A), the c.607-6T>G (IVS7-6T/G) and the c.554-1G>T (IVS6-1G>T) splicing mutations and the c.786G>A (p.Trp262X) nonsense mutation all together represent 60% of mutant alleles in the general US population (Sniderman King et al. 2011). Surprisingly, only one HT1 allele was reported until now in Mexico and this allele carried the c.1062 + 5G>A mutation most prevalent in Quebec (Table 1). 16 new cases have recently been described in Brazil (Neto et al. 2014), with only two alleles reported at this time, and these harboured the c.554-1G>T (IVS6-1G>T) mutation, most prevalent in the Mediterranean area (Table 1).

Arranz et al. in their work based on a panel of 29 patients mostly from southern Europe demonstrated a high homogeneity of the mutational spectrum in this region (Arranz et al. 2002). In a retrospective study on European HT1-affected individuals (Couce et al. 2011), mutational analysis on 34 Spanish patients reported nine different mutations in this population, documenting c.554-1G>T (IVS6-1G>T) as the most prevalent, in accordance with the previous literature (Arranz et al. 2002). Molecular genetics analysis of the fah gene in 11 Czech patients with HT1, diagnosed in the Medical Faculty of Charles University in Prague between 1982 and 2006, revealed three mutations not previously described: the c.579C>A nonsense mutation (p.Cys193X) and the c.680G>T (p.Gly227Val) and c.1210G>A (p.Gly404Ser) missense mutations (Vondrackova et al. 2010).

The Middle East is interesting in the sense that even though patients harbour the common c.1062 + 5G>A (IVS12 + 5G>A) and c.554-1G > T (IVS6-1G>T) mutations, many of the other mutations reported are typical to this region. One such example is the already described c.192G>T (p.Gln64His) mutation, which is thus far found only in people originating from Pakistan, the Middle East and North West India. This mutation accounts for over one third of all HT1 alleles in these populations (Rootwelt et al. 1994a; Rootwelt et al. 1996). Another mutation that is often detected in patients from the Middle East is the c.709C>T (p.Arg237X) mutation (Imtiaz et al. 2011). In Turkey, the c.698A>T (p.Asp233Val) mutation, which has not been reported elsewhere, accounts for 20% of the reported alleles (Rootwelt et al. 1994a; Rootwelt et al. 1996; Dursun et al. 2011). Moreover, other different mutations, although not at high frequency, are peculiar for this population (Table 1).

The c.782C>T (p.Pro261Leu) missense mutation was found in 100% of Ashkenazi-Jewish examined in Israel (Elpeleg et al. 2002). Direct sequencing in 43 HT1-affected patients originating from Saudi Arabia, Egypt and Iran identified a total of 17 different homozygous mutations. Eleven of these (8 missense, 1 nonsense, 1 splice site and 1 deletion) had not been reported previously (Imtiaz et al. 2011).

Little information about the epidemiology and molecular defects in HT1 patients from East Asia is available at this time. Sakai and Kitagawa (1957) reported the first case of HT1 in a two-month-old Japanese patient, but genetic analysis was not possible at that time. The c.185T>G (p.Phe62Cys) represents the first and the only allele reported in Japan to date (Awata et al. 1994). Recent findings start to describe HT1 mutations in China. The missense mutation c.1124T>C (p.Leu375Pro) represents the first case of HT1 analysed by molecular genetics in this area (Cao et al. 2012). This mutation, affecting the secondary protein structure, decreases the stability of FAH enzyme and compromises the protein’s functions. Another report represents the first case of HT1 in a two-month-old Hong Kong Chinese patient (Mak et al. 2013). Genetic analysis of this patient showed two novel mutations, the c.1063-1G>A splicing mutation and the c.1035_1037del. Recently, clinical data on 3 HT1 Chinese patients showed five mutations in the FAH gene: c.455G>A (p.Trp152X), c.520C>T (p.Arg174X), c.974_976delCGAinsGC, c.1027G>A (p.Gly343Arg) and c.1100G>A (p.Trp367X) (Yang et al. 2012; Dou et al. 2013). The c.455G>A, c.974_976delCGAinsGC and c.1100G>A mutations have not been described elsewhere. Currently, few cases of HT1 have been reported in Korea. Mutational analysis of two female neonates admitted to hospital for further work-up of an abnormal newborn screening test revealed three novel mutations (one deletion, one missense and one splice defect) that have not been reported elsewhere (Park et al. 2009; Choi et al. 2014).

To our knowledge no mutations in HT1 have yet been documented in Central America or in the Oceania continent.

Conclusions

The advent of neonatal screening, prenatal diagnosis and carrier tests for genetic disorders has shown the importance of establishing the population frequencies and ethno-geographic spread of mutations for the evaluation of future screening strategies. To highlight the prevalence of HT1 mutations in a geographical context, we compiled all reported HT1 alleles worldwide, including those not yet reported in the common databases, and another two, discovered in the screening of HT1 patients in our laboratories over a period between 2001 and 2013 (summarized in Table 1 and Fig. 2). Obvious conclusions can be drawn when we examine the incidence of HT1 worldwide (Fig. 2 and Table 1).

According to the data gathered so far, a preferential screening for those mutations in regions in which they show a higher prevalence could provide some improvement in carrier diagnostic efficiency and may enable the establishment of family pedigrees for adequate counselling in some cases. Currently, screening is carried out in Quebec, the USA and Europe (Morrissey et al. 2011; Barnby 2014). In this case it is obviously important to know the pattern of mutations in the respective populations.

However, it is necessary to bear in mind that this compilation may be partly biased by the fact that: (1) very few cases of HT1 are overlooked in some countries as in the province of Quebec due to a screening programme for HT1 established early in 1970 and (2) not all cases of HT1 are described in the literature. Many of these probably occur in countries with no or limited access to service for diagnosis of genetic disease and as a result remain undiagnosed (De Laet et al. 2013). This could lead to some geographical bias reflected in the fact that the majority of the patients whose mutations have been described are residents of Europe, the Middle East or North America. (3) Whilst we have carefully attempted to ensure that patients are not counted twice because they appear in more than one publication, this may occur in a few cases.

In summary, this report allows a detailed identification of the mutations causing HT1 worldwide, with diagnostic and methodological consequences implementing the groundwork for future carrier and prenatal testing, premarital screening and pre-implantation genetic diagnosis.