If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Laboratory of Bioinformatics and Systems Biology, Centre of New Technologies, University of Warsaw, Warsaw 02-089, PolandDepartment of Physical Chemistry, Faculty of Pharmacy, Medical University of Warsaw, Warsaw 02-097, Poland
Corresponding authors. Addresses: Department of Gastroenterology and Hepatology, Medical Center for Postgraduate Education, Warsaw 01-813, Poland. Tel.: +48 225462575; fax: +48 225463191, or Department of Gastroenterology, Hepatology and Feeding Disorders, Children’s Memorial Health Institute, Warsaw 04-730, Poland.
Corresponding authors. Addresses: Department of Gastroenterology and Hepatology, Medical Center for Postgraduate Education, Warsaw 01-813, Poland. Tel.: +48 225462575; fax: +48 225463191, or Department of Gastroenterology, Hepatology and Feeding Disorders, Children’s Memorial Health Institute, Warsaw 04-730, Poland.
Department of Gastroenterology and Hepatology, Medical Center for Postgraduate Education, Warsaw 01-813, PolandDepartment of Genetics, Cancer Center-Institute, Warsaw 02-781, Poland
Individuals with macro-AST demonstrate fluctuating, not persistent, elevation of AST levels in blood.
•
The causative gene of macro-AST has not yet been identified.
•
A genetic variant associated with macro-AST was found in theGOT1 gene.
•
Testing for this variant could aid diagnosis of macro-AST.
Background & Aims
Macro-aspartate aminotransferase (macro-AST) manifests as a persistent elevation of AST levels, because of association of the protein with immunoglobulins in the circulation. Macro-AST is a rare, benign condition without a previously confirmed genetic basis.
Methods
Whole exome sequencing (WES)-based screening was performed on 32 participants with suspected familial macro-AST, while validation of variants was performed on an extended cohort of 92 probands and 1,644 healthy controls using Taqman genotyping.
Results
A missense variant (p.Gln208Glu, rs374966349) in glutamate oxaloacetate transaminase 1 (GOT1) was found, as a putative causal variant predisposing to familial macro-AST. The GOT1 p.Gln208Glu mutation was detected in 50 (54.3%) of 92 probands from 20 of 29 (69%) families, while its prevalence in healthy controls was only 0.18%. In silico analysis demonstrated that the amino acid at this position is not conserved among different species and that, functionally, a negatively charged glutamate on the GOT1 surface could strongly anchor serum immunoglobulins.
Conclusions
Our data highlight that testing for the p.Gln208Glu genetic variant may be useful in diagnosis of macro-AST.
Lay summary
Higher than normal levels of aspartate aminotransferase (AST) in the bloodstream may be a sign of a health problem. Individuals with macro-AST have elevated blood AST levels, without ongoing disease and often undergo unnecessary medical tests before the diagnosis of macro-AST is established. We found a genetic variant in the GOT1 gene associated with macro-AST. Genetic testing for this variant may aid diagnosis of macro-AST.
Macroenzymes, serum high-molecular-weight compounds that are formed by polymerization or association with other serum constituents (primarily immunoglobulins
A typical macroenzyme is the macroenzyme form of aspartate aminotransferase (macro-AST), the presence of which is generally characterized by increased serum AST activity.
AST is highly conserved and expressed in many cell types, including hepatocytes, erythrocytes, and skeletal muscle cells. There are two similar isoenzymes of AST that localize to the cytoplasm and mitochondria and are encoded by GOT1 and GOT2 on chromosomes 10q24 and 16q12, respectively. Serum AST activity is largely of cytosolic origin,
and, while elevated serum AST activity is considered a basic biochemical marker of hepatic, cardiac, muscle, endocrine, and metabolic disorders, macro-AST is generally a benign condition. However, the diagnosis of macro-AST is usually delayed because of confusion about the reason for elevated AST activity, and many unnecessary medical tests and procedures are frequently performed, including invasive liver biopsies, before the final diagnosis is established.
is produced by formation of an ∼250 kDa complex of AST with serum immunoglobulins (IgA, IgG, or both), and the specific enzyme binding site is mostly on the Fab and F(ab1)2 fragment of the immunoglobulin molecule,
leading to altered renal clearance and excretion of the enzyme and, ultimately, to higher levels of activity. The mechanisms of the immune complex formation are unclear, but may be due to autoimmunity, with immunoglobulins targeting enzymes as antigens via molecular mimicry.
in 1978 in two healthy women with an unexplained persistent elevation of AST activity. Their serum AST isoenzymes contained a fraction that migrated between cytoplasmic and mitochondrial AST on electrophoresis, and an AST-IgG complex was confirmed in one of them.
), wherein diagnosis was performed using different methods, including electrophoresis, gel permeation chromatography, polyethylene glycol precipitation, and heat stability measurements.
Macro-AST is rare in the general population, with a prevalence of 0.014% and 9.09% among a general population of gastroenterological patients and those with isolated increased AST activity without liver abnormalities, respectively.
While in adult patients macro-AST is often accompanied by other diseases, including neoplasms and autoimmune diseases, isolated macro-AST is more frequently identified in children.
To date, no genetic tendency has been described in either adult or pediatric macro-AST, probably because of the rarity of the condition, particularly in children. Only a single case of suspected familial macro-AST has previously been reported in a mother and her 2-month-old baby with macro-AST, who were otherwise healthy.
Between 2005 and 2016, a clinical diagnosis of macro-AST was established in 69 out of 744 pediatric patients admitted to the Departments of Gastroenterology, Hepatology, Nutritional Disorders, or Pediatrics, Children’s Health Memorial Institute, Warsaw for diagnosis of an unexplained increase in AST activity levels. The patients affected by macro-AST exhibited asymptomatic, fluctuating (rather than persistent) increased AST activity observed over months or years. None of these children had suffered from any major medical problems, except one who was diagnosed with neuroblastoma during clinical follow-up. None were treated with hepatotoxic drugs or with drugs that could affect liver enzyme activities, none had symptoms of liver disease or cholelithasis, and neuromuscular status was normal. Except for elevated serum AST activity, the results of the following tests were normal: serum alanine aminotransferase, alkaline phosphatase, gamma glutamyltransferase, cholinesterase, creatine kinase and its isoenzymes, lactate dehydrogenase and its isoenzymes, aldolase, amylase, total bilirubin, conjugated bilirubin, and prothrombin time. Tests for viral (hepatitis A, B, and C viruses, cytomegalovirus, and Epstein-Barr virus), metabolic (α-1-antitrypsin, ceruloplasmin, and urinary copper before and after penicillamine load), and autoimmune (antinuclear, anti-smooth muscle, antimitochondrial, and anti-liver-kidney microsomal and anti-transglutaminase antibodies) conditions also revealed no abnormalities. Liver ultrasound was normal. Additionally, hemolytic, renal, and pancreatic causes of elevated AST activity were excluded.
Laboratory tests for diagnosis of macro-AST
Polyethylene glycol (PEG) precipitation tests. Equal volumes (100 μl) of a 24% solution of PEG 6000 and patients’ sera were mixed, equilibrated for 10 min. at 37 °C and centrifuged at 1,500g for 20 min at 22 °C. To calculate the PEG precipitable activity (PPA), AST activity was measured in the supernatant, and after correction for dilution was compared with samples obtained from unprecipitated serum.
AST isoenzyme electrophoresis. Electrophoresis on 1% agarose gel was carried out in a barbiturate buffer (pH 8.6) at 100 Volts, then gels were soaked in a staining reagent mixture (10 µM alpha-ketoglutarate, 200 µM L-cysteine sulfinate, 0.1 mg m-PMS, 0.8 mg MTT, 2 µM EDTA, 20 mg dextran and 100 µM imidazole buffer [pH 7.5]) for 20 min. at 37 °C, immersed in 10% acetic acid for 5 min, washed with water and dried in an oven at 65 °C. Purple bands indicated the presence of free-AST-molecules or AST complexes.
DNA isolation, exome sequencing, and genotyping
The study protocol was approved by the local Bioethical Committee, and all participants, or their parents, provided written informed consent. The study protocol conforms to the ethical guidelines of the 1975 Declaration of Helsinki. Genomic DNA was extracted from whole blood treated with EDTA as previously described.
Human exome sequencing libraries were constructed using AmpliSeq™ Exome technology (Thermo Fisher Scientific), and generated amplicons were sequenced using the Ion Proton™ platform (Thermo Fisher Scientific) as previously described.
Raw reads were processed using the Torrent Suite version 5.0 analysis pipeline and mapped to human genome assembly hg19 with TMAP Aligner (Thermo Fisher Scientific). Variant calls were made using Torrent Variant Caller version 5.0-13
with high stringency for base insertions/deletions (indels), and low stringency for single nucleotide polymorphisms (SNPs). If multiple alleles were reported in one position, they were separated using the vcflib script, vcfbreakmulti.
overall minor allele frequency [MAF] <1%), non-synonymous variants, the Fisher’s exact test for allele over-representation was conducted using WES data obtained from a cohort of 77 pediatric patients with Wilson’s disease (manuscript in preparation) as controls. Variants that reached genome-wide statistical significance were additionally checked using a variant-filter script.
Genotyping of the rs374966349 variant was performed with TaqMan SNP Genotyping Assays (Thermo) on an ABI 7900HT qPCR system (Thermo), as previously described.
Sequencing data (fragments of BAM files containing variant rs374966349) have been deposited in The European Nucleotide Archive under the accession number PRJEB20091.
Pediatric patients admitted with unexplained elevated levels of AST activity and no evidence of associated diseases, underwent PEG screening test and AST isoenzyme electrophoresis examination to confirm or exclude the presence of macro-AST. Although the PEG precipitation technique has been considered a gold standard for the diagnosis of macro-AST, PPA reference ranges are still not defined.
In our study, gel electrophoresis of the sera with PPA of <52% exhibited typical electrophoretic patterns of isoenzymes, composed of an anionic and a cationic band, which represent the cystolic or supernatant (cAST/sAST) and the mitochondrial (mAST) isozymes, respectively. In turn, in 69 patients whose PPA ranged between 59% and 78%, electrophoresis revealed a single band with an abnormal migration, either between the sAST and the mAST bands or near to the sAST position. As suggested by Moriyama et al., the abnormal bands may indicate the type, amount and isoenzyme specificity of bound immunoglobulin and may differentiate between IgG and IgA complexed-AST.
Once macro-AST was diagnosed, the diagnostic procedure for macro-AST was also performed in parents and siblings. In 37 families, at least one other family member was also macro-AST-positive (familial macro-AST), while all relatives from the other 32 families were macro-AST-negative (sporadic macro-AST). To investigate the genetic background of patients with suspected familial macro-AST, whole exome sequencing (WES)-based screening was performed on 32 participants, including 15 male children, eight female children, six male adults, and three female adults.
On average, 21,514,330 reads (median = 21,415,862) that mapped to human genome assembly hg19 were generated per sequencing run. An average of 74% of bases exhibited coverage of ⩾20×. In patient samples, 65,472 variants were discovered in coding regions: 1,423 indels and 64,244 SNPs. Each patient had a mean of 18,444 variants, of which 18,257 were SNPs and 177 were indels. Among these, four rare non-synonymous variants passed all filters and reached the genome-wide significance level (p <5×e−8) by Fisher’s exact test, including a rare (MAF in ExAC database as low as 0.02%) missense variant in GOT1, encoding p.Gln208Glu (rs374966349; two-tailed p = 1.73×e−11). Other variants were located in the genes MUC4 (two variants) and NBPF10; however, according to their Gene Damage Index Scores, both of these genes are highly tolerant to mutations.
The rs374966349 variant was further genotyped in extended cohorts of probands and healthy controls. Among 1,644 controls, only three individuals were carriers of the heterozygous GOT1 mutation, indicating a prevalence of 0.18%. The mutated genotype was detected in 38 and 6 macro-AST–positive and macro-AST–negative probands, respectively, recruited from families with familial macro-AST (Fig. 1). Out of 12 patients who were macro-AST–positive, but only had macro-AST–negative relatives (sporadic macro-AST), three children were carriers of the heterozygous mutations; among their family members one mutation carrier was found in each of the three families who were otherwise macro-AST–negative (Fig. 1, family #3, #7 and #20). The only homozygous GOT1 mutation was found in a father of two sons; all three of them were affected by macro-AST (Fig. 1, family #16). Altogether, the GOT1 mutation was present in 50 probands from 20 (69%) families studied; 41 carriers of the mutation were macro-AST–positive and nine were macro-AST–negative (Fig. 1). While this mutation has less than 100% penetrance, and in nine (31%) families with familial macro-AST no mutation was detected, we cannot exclude the possibility that familial macro-AST has polygenic inheritance, in which the same clinical phenotype arises from different impact loci and/or from additive interactions between them.
Fig. 1Prevalence of the glutamate oxaloacetate transaminase 1 (GOT1) p.Gln208Glu mutation in families with at least two occurrences of macro-aspartate aminotransferase.
The serum AST activity levels are summarized in Table 1. As expected, elevated AST activity was determined in macro-AST–positive patients; normal or slightly over the upper limit levels were found in the macro-AST–negative carriers of GOT1 mutations; and normal levels were detected in macro-AST–negative relatives exhibiting the wild-type GOT1 genotype. Interestingly, in subjects diagnosed with macro-AST who were the mutation carriers, serum AST levels were significantly higher compared to non-carriers (p value = 0.043; Welch Two Sample t test) (Fig. 2).
Table 1Relation between the GOT1 genotype and serum AST activity levels in patients with a clinically diagnosed familial macro-AST and their unaffected relatives.
GOT1 genotype/clinical diagnosis of macro-AST
19 families with mutated genotype
9 families with wild-type genotype
Mutated/Macro-AST positive
Mutated/Macro-AST negative
Wild type/Macro-AST negative
Wild type/Macro-AST positive
Wild type/Macro-AST negative
Proband number
41
9
13
18
11
AST; min–max
55–228
26–54
13–40
45–173
13–35
AST; median
91
37
31
70
17
AST, aspartate aminotransferase; normal range, 10–40 U/L.
Fig. 2The serum aspartate aminotransferase (AST) activity levels in subjects diagnosed with macro-AST who were either carriers or non-carries of glutamate oxaloacetate transaminase 1 p.Gln208Glu mutation. The statistical difference was assessed with Welch Two Sample t test.
To characterize the potential influence of the identified missense mutation on protein function at the structural level, we mapped the mutation onto the three-dimensional structure of the human GOT1 homodimer (pdb|3ii0) (Fig. 3). This analysis demonstrated that GOT1 p.Gln208Glu is located away from the protein’s active site, on the surface of a helix exposed to the solution, and on the opposite side from the site that mediates dimerization. Further bioinformatic analysis revealed that the mutated amino acid at this position is not conserved in different species. Overall, the results indicate that this mutation, which generates a surface exposed, negatively charged glutamate, may have a causative function in macro-AST by inducing a stronger association between AST and serum components, likely immunoglobulins. Although this rare variant is incompletely penetrant and probably not the only factor causing familial macro-AST, testing for this genetic variant could be useful in diagnosis of macro-AST, at least in some families.
Fig. 3Location of the p.Gln208Glu mutation, identified in patients with macro-aspartate aminotransferase (AST), in the structure of the human glutamate oxaloacetate transaminase 1 homodimer. The amino acid mutated in macro-AST is shown in red. Monomers are colored blue and gray, while key residues from both active sites are indicated in yellow and magenta. (This figure appears in colour on the web.)
Enhanced utility of family-centered diagnostic exome sequencing with inheritance model-based analysis: results from 500 unselected families with undiagnosed genetic conditions.
WES is a cost-effective technology for detection of disease variants underlying Mendelian disorders, as well as for cataloguing common and rare disease-related genomic alterations;
however, its diagnostic yield depends on age of disease onset, the presence of a positive family history, and specific clinical phenotypes. The diagnostic success rate for WES-based identification of rare causative variants is only 15–30%.
Enhanced utility of family-centered diagnostic exome sequencing with inheritance model-based analysis: results from 500 unselected families with undiagnosed genetic conditions.
Identification of macro-AST causative variants from existing incidental findings and variants of unknown/uncertain significance was possible because, over 11 years, we have assembled the largest cohort of macro-AST patients published to date. Furthermore, we describe familial macro-AST for the first time. Selection of such a large number of patients was possible because staff in our Departments of Gastroenterology, Hepatology, Nutritional Disorders, and Pediatrics at the Children’s Health Memorial Institute are trained in diagnosis of macro-AST, and 744 pediatric patients with unexplained AST elevation have been directed to our diagnostic center from hospitals across Poland. Furthermore, when well-defined macro-AST was established in a pediatric patient, the diagnostic procedure was extended to other members of the immediate family. Finally, WES-based genetic screening was conducted in an academic laboratory to identify a non-synonymous change in the gene encoding cytoplasmic GOT1, which co-segregates convincingly with the macro-AST phenotype and has functional plausibility as the causative agent of this phenomenon.
Conclusions
The AST protein is conserved in all prokaryotic and eukaryotic organisms.
only one report has revealed a direct impact of a rare in-frame deletion of three nucleotides in GOT1, encoding asparagine at position 389 (p.Asn389del), and leading to decreased GOT1 serum enzyme activity.
Using WES and strict filtering criteria we uncovered, for the first time, a very rare amino acid substitution in GOT1, which co-segregated with familial macro-AST. Additionally, by performing in silico analysis we identified a potential relationship between AST structure and function, underlying the formation of high-molecular-weight serum macro-AST. Finally, screening for this alteration in cases of unexplained elevation of AST activity has the potential to prevent unnecessary and intrusive tests.
Financial support
This work was supported by: the National Science Centre [2013/11/B/NZ2/00130] and the Children’s Memorial Health Institute [S131/2013]. JO was supported by the National Science Centre [2011/02/A/NZ5/00339]. KG and ML were supported by the Foundation for Polish Science (TEAM).
Conflict of interest
Prof. Czlonkowska and Prof. Ostrowski report grants from The National Science Centre, during the conduct of the study.
Please refer to the accompanying ICMJE disclosure forms for further details.
Authors’ contributions
Conception and design of the study: J.O., P.S.; Patients recruitment and clinical data compilation: P.S., A.W., A.H., W.J., A.C.; DNA isolation, exome sequencing, Taqman genotyping: A.P., J.K., M.D.; Exome-Seq dataset analyses and interpretation: M.K., M.M.; In silico GOT1 analysis and interpretation: K.G., M.L.; Drafting of the manuscript: J.O., M.K., M.M.
Enhanced utility of family-centered diagnostic exome sequencing with inheritance model-based analysis: results from 500 unselected families with undiagnosed genetic conditions.
32146-3&id=fx1.jpg){kind=link}
32146-3&id=gr1.jpg){kind=link}
32146-3&id=gr2.jpg){kind=link}
32146-3&id=gr3.jpg){kind=link}