N-carbamylglutamate enhancement of ureagenesis leads to discovery of a novel deleterious mutation in a newly defined enhancer of the NAGS gene and to effective therapy

Abstract

N-acetylglutamate synthase (NAGS) catalyzes the conversion of glutamate and acetyl-CoA to NAG, the essential allosteric activator of carbamyl phosphate synthetase I, the first urea cycle enzyme in mammals. A 17-year-old female with recurrent hyperammonemia attacks, the cause of which remained undiagnosed for 8 years in spite of multiple molecular and biochemical investigations, showed markedly enhanced ureagenesis (measured by isotope incorporation) in response to N-carbamylglutamate (NCG). This led to sequencing of the regulatory regions of the NAGS gene and identification of a deleterious single-base substitution in the upstream enhancer. The homozygous mutation (c.-3064C>A), affecting a highly conserved nucleotide within the hepatic nuclear factor 1 (HNF-1) binding site, was not found in single nucleotide polymorphism databases and in a screen of 1,086 alleles from a diverse population. Functional assays demonstrated that this mutation decreases transcription and binding of HNF-1 to the NAGS gene, while a consensus HNF-1 binding sequence enhances binding to HNF-1 and increases transcription. Oral daily NCG therapy restored ureagenesis in this patient, normalizing her biochemical markers, and allowing discontinuation of alternate pathway therapy and normalization of her diet with no recurrence of hyperammonemia. Inc.

Figure 1

Experimental time line. Identical studies were performed before and immediately following a 3-day course of NCG at a dose of 2.2 g/m²/d. Plasma samples for ammonia, urea, amino acids, and [¹³C]urea were obtained at 0, 15, 30, 45, 60, 90, 120, 180, and 240 minutes prior (time 0) and following the ingestion of isotopic acetate.

Figure 2

Consensus HNF-1 binding site. The human NAGS HNF-1 binding site and matrices for HNF-1 binding sites from the TRANSFAC database. The -3063C>A mutation is indicated. Standard nucleotide codes are used.

Figure 3

Isotopic enrichment of [¹³C]urea. Increase over time in isotopic enrichment of [¹³C]urea (A) and plasma concentration of [¹³C]urea (B) in the patient, who was administered 27.5 mg/kg of [1-¹³C]sodium acetate at time 0 of each study day.

Figure 4

DNA sequence of the HNF-1 binding site within the enhancer of the NAGS gene. The patient is homozygous for a single base substitution at position -3063. The patient’s parents are heterozygous for this allele at the same position. Asterisk (*) indicates the location of the mutation.

Figure 5

Transcription factor binding sites in the NAGS enhancer. Transcription factors HNF-1 and NF-Y bind within the human NAGS enhancer.

Figure 6

Alignment of the -3063C>A mutation with the HNF-1 binding site in mammalian species. Reference mammalian DNA sequence of the HNF-1 binding site was aligned with the patient’s HNF-1 binding site. The -3063C>A mutation (box) is indicated.

Figure 7

Effects of the -3063C>A mutation and consensus HNF-1 binding site on gene expression. Luciferase expression is significantly different between wild-type, -3063C>A, and consensus enhancers when expressed in conjunction with the human wild-type NAGS promoter (A) or the minimal TATA-box promoter (B). All results are an average of three independent experiments that were each carried out in triplicate. Calculated error was from three trials and reported as ±SEM. One asterisk (*) indicates p<0.05 and two asterisks (**) indicate p<0.005.

Figure 8

Effects of the -3063C>A mutation and consensus HNF-1 binding site on “pull-down” of HNF-1 protein. Amounts of HFN-1 protein pulled down were significantly different between enhancers containing the wild-type, -3063C>A mutated, or consensus HNF-1 binding site (A). Immunoblots (B) show HNF-1 and NF-Y protein following overnight incubation with whole nuclear extract. Values are normalized means from three replicate experiments. Immunoblots are representative of the results obtained by three replicate experiments. Calculated error was from three trials and reported as ±SEM. One asterisk (*) indicates p<0.05 and two asterisks (**) indicate p<0.005.