Adenine phosphoribosyltransferase (APRT) deficiency: identification of a novel nonsense mutation

Adenine phosphoribosyltransferase deficiency (APRTD) is a rare autosomal recessive metabolic disorder due to a mutation of the APRT gene [1]. APRT is a purine-metabolism enzyme that catalyzes the formation of 5′-adenosine monophosphate (5′-AMP) and pyrophosphate (PP) from adenine and 5-phosphoribosyl-1-pyrophosphate [2, 3]. In patients with complete APRT deficiency, adenine is oxidized by xanthine oxidase (XO) to the highly insoluble and nephrotoxic derivative 2,8-dihydroxyadenine (2,8-DHA) [4], leading to urolithiasis and renal failure caused by intratubular crystalline precipitation [5, 6]. The APRT gene, located on chromosome 16q24 [7], is approximately 2.6 kb long, contains five exons and four introns, and encodes a protein of 180 amino acid residues [8]. The human enzyme, present in a variety of cell types including erythrocyte [9], is a homodimer composed of two identical subunit species with a molecular weight of about 19.481 Da [10]. Currently, there are two isoforms produced by alternative splicing: the isoform 1 (P07741-1) and the isoform 2 (P07741-2); the isoform 1 has been considered as the ‘canonical’ one.

In the pathologic allelic variants, more than 40 mutations have been identified in the coding region of APRT gene in over 300 affected individuals from more than 25 countries, including at least 200 individuals from Japan. Approximately 10% of mutant alleles in affected white individuals and 5% in affected Japanese haven’t been yet identified. APRT gene alterations include missense, frameshift, and nonsense mutations and small deletions/insertions ranging in size from 1 to 8 base pairs. The estimated heterozygosity in different populations ranges from 0.4 to 1.2% [11], suggesting that the prevalence of a homozygous state is at least 1:50,000 to 1:100,000.

Mutant alleles responsible for the disease have been classified as APRT*Q0 for type I and APRT*J for type II APRTD. Type I APRT deficiency (complete deficiency in vivo or in vitro) has been found in patients from many different countries [12‐15]. Type II deficiency (complete enzyme deficiency in vivo but partial deficiency in cell extracts) has been found mainly in Japan [16‐18]. However, this distinction is only of historical interest, because APRT enzyme activity in intact cells has been shown to be approximately 1% in both types [19].

The most common mutations in affected European individuals are: (i) T insertion at the intron 4 splice donor site (IVS4 + 2insT) which leads to the deletion of exon 4 from the mRNA because of aberrant splicing. This mutation has been found in individuals from many European countries as well as in an affected individual from the US, (ii) A-to-T transversion in exon 3 (g.194A > T, p.Asp65Val), described in affected individuals from Iceland, Britain, and Spain. The three most common mutations in affected Japanese individuals, in order of decreasing frequency, are: (i) T-to-C missense mutation in exon 5 (g.442 T > C), (ii) G-to-A nonsense mutation in exon 3 (g.329G > A) and (iii) a four-base pair (CCGA) duplication in exon 3 that leads to a frameshift after codon 186 [20, 21].

APRT deficiency is a disease caused by mutations in the APRT gene. There are more than 40 types of mutations in APRT gene, but currently only five allelic variants are responsible for the complete inactivation of the APRT protein both in vitro and in vivo.

This disease is characterized by the presence of 2,8-DHA crystals in urine. These crystals are radiolucent and are often considered as uric acid stones or, in renal biopsies, as oxalate crystals, causing an erroneous diagnosis. Ultraviolet or infrared spectrophotometry is required for their correct identification.

Accurate diagnosis of APRT deficiency is crucial because early treatment with allopurinol or low-purine diet effectively prevents the stone formation and may improve patients’ renal function. Nowadays, several diagnostic tools are available for the identification of the APRT deficiency such as 2,8-DHA crystal microscopic detection, APRT enzyme activity assays and molecular analysis of the APRT gene. In the present study, we analyzed an Italian patient with a clinical history of recurrent nephrolithiasis and chronic renal failure due to obstructive nephropathy and recurrent urinary tract infections. The patient received a renal transplant in 2010 but, unfortunately, he died 10 months after the surgical procedure. The definitive diagnosis of APRT deficiency was made by microscopic detection of 2,8-DHA crystals on the biopsy of the failing transplanted kidney and by documenting decreased APRT activity in red blood cells. Genetic analysis revealed a combination of a previously described homozygous state for rs8191483 SNP and a novel nonsense mutation p.Gln147X. The identification of two allelic variants in homozygous state suggest the consanguinity in this family. The carriers are asymptomatic and they are usually identified during family screening.

The normal human APRT is a protein of 180 amino acids, composed of 9 β-strands and 6 α-helices, which can be divided into the “core” (residues 33–169), the “hood” (residues 5–34), and the “flexible loop” (residues 95–113) domains. On the basis of the APRT structure, which has been widely described [21], it is evident that Ala131 and Leu159 are essential residues for the specific recognition of adenine among different purines through hydrophobic interactions. Moreover, the importance of the Leu159 residue is also confirmed by the presence in the same position of a lysine residue in the PRTases that binds hypoxanthine, xanthine, or guanine. This difference between residues located at the amino acid 159 position explains the specificity of type I PRTases for their respective purines [25].

The novel nonsense mutation reported in this study caused the introduction of a stop codon leading to the loss of 33 amino acids, corresponding to the C-terminal domain of the APRT protein. This portion represents the “core” of the APRT protein. In particular, the fundamental Leu159 residue, responsible for the correct activity of the APRT enzyme, is located in the S8 strand. Therefore, the complete inactivation of the APRT protein reported in our patient is due to the impaired binding of the specific substrate to the active site, since the g.2098C > T mutation deeply affects the enzyme structural integrity.

In conclusion, the novel nonsense mutation p.Gln147X might be the main cause for the therapeutic failure observed in our patient.

Moreover, the evidence of this novel “loss-of-function” mutation might be extremely useful as a new genetic diagnostic marker for the early identification of the APRT deficiency.

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor of this journal.