GATA6 Mutations Cause a Broad Phenotypic Spectrum of Diabetes From Pancreatic Agenesis to Adult-Onset Diabetes Without Exocrine Insufficiency

Patients were selected from an international cohort of 795 case subjects from 71 countries with diabetes diagnosed before 6 months of age. Pancreatic agenesis, defined as diabetes needing insulin and pancreatic exocrine insufficiency requiring enzyme supplementation therapy, was diagnosed in 39 of 795 case subjects. We previously studied 27 of these 39 case subjects with pancreatic agenesis (1). A genetic cause was identified in 624 of the 795 case subjects; these include 15 patients with pancreatic agenesis who are heterozygous for a GATA6 mutation (1). Mutations in PTF1A and PDX1 were excluded in all the patients with pancreatic agenesis.

We sequenced GATA6 in the remaining 171 patients in whom no genetic etiology had been identified. These included 24 case subjects with pancreatic agenesis (NDM requiring insulin and pancreatic exocrine insufficiency requiring enzyme supplementation therapy). NDM without exocrine insufficiency was reported in 147 patients: 126 with PNDM and 21 with TNDM. Mutations in ABCC8, KCNJ11, INS, and chromosome 6q24 abnormalities had been excluded.

Clinical information was provided by the referring clinicians via a NDM request form (available at www.diabetesgenes.org) or from clinical notes. The study was conducted in accordance with the Declaration of Helsinki principles with informed parental consent given on behalf of children.

We screened the coding sequence and ∼50 bp of flanking sequence of GATA6 using Sanger sequencing as described previously (primers) (1). Sequencing reactions were run on an ABI3730 capillary machine (Applied Biosystems, Warrington, U.K.) and analyzed using Mutation Surveyor v3.98 (SoftGenetics, State College, PA) (GATA6 nucleotide reference NM_005257.3).

The bioinformatics tools SIFT, PolyPhen-2, and Align GVGD were used to predict the effect of novel variants on the GATA6 protein (protein reference NP_005248.2).

Microsatellite analysis of parent/proband trios using the PowerPlex kit (Promega, Southampton, U.K.) was used to confirm apparently de novo mutations. The parental origin of de novo mutations in pedigrees with inherited mutations was investigated when grandparental samples were available by analyzing microsatellites D18S1149, D18S1104, D18S1067, D18S1001, D18S869, and D18S1002.

GATA6 mutations were identified in 9 of 171 probands and five parents. Six of the probands have pancreatic agenesis. In total, we have now identified GATA6 mutations in 21 of 39 (54%) patients with pancreatic agenesis (6 case subjects in this study and 15 reported by Lango Allen et al. [1]). Three of 147 patients (2%) with NDM but not on exocrine supplementation were heterozygous for GATA6 mutations; 2 have permanent diabetes, and 1 has transient episodes of hyperglycemia.

The nine novel mutations include three missense changes (p.C447R, p.G469E, and p.R479G) and four splicing, one frameshift, and one nonsense mutation. In total, 23 different GATA6 mutations have been identified in patients with diabetes (p.R456C was present in two unrelated patients) (Fig. 1). All missense mutations affect highly conserved residues within the second zinc-finger domain, which is needed for DNA binding and protein-protein interaction. None of the mutations were present in the dbSNP137, 1000 Genomes (March 2012 release, based on 1,197 individuals), or Exome Variants Server (June 2012 release, based on 6,503 individuals) databases. They were predicted to affect protein function by at least two of the bioinformatics predictive tools used (Supplementary Table 1). Three mutations affect conserved splice sites, and in silico analysis suggested aberrant splicing for the intronic c.1303–10C>G, c.1516+4A>G, c.1429–8T>G, and c.1429–41_1441del mutations (see Supplementary Table 2).

Parental DNA samples were available for eight of nine previously unreported patients. Mutations were shown to have arisen de novo in three patients and been inherited from a parent in five case subjects (see Fig. 2). The analysis of grandparental samples in families ISPAD 207, ISPAD 208, ISPAD 209, and ISPAD 212 showed that the mutation had arisen de novo in the parent on the grandmaternal allele in three case subjects and on the grandpaternal allele in the remaining one. The level of the GATA6 mutation in one parent (ISPAD 208-2) was estimated at ∼20% in leukocyte DNA.

The median age at diagnosis of diabetes for the 24 probands with GATA6 mutations was 2 days (IQR 1–7 days) and median birth weight was 1,588 g (SDS = −3.22, <1st percentile). Pancreatic imaging had been undertaken in 18 of 21 pancreatic agenesis case subjects and showed complete absence (n = 8) or marked hypoplasia (n = 10) of the pancreas. Measurement of fecal elastase or fecal fat in 15 case subjects showed severe exocrine pancreatic insufficiency.

The two probands with PNDM but not requiring treatment with pancreatic enzyme supplementation were diagnosed at 1 day (c.1303–1G>T) or 1 week (p.C447R) of age, and both have been treated with insulin since diagnosis. Although they had never received enzyme supplements, fecal elastase measurement in ISPAD 208 showed exocrine insufficiency (<15 µg/g, normal range >200 µg/g).

The patient with the de novo p.R479G mutation (ISPAD 206) was diagnosed with diabetes at age 8 weeks (glucose 28 mmol/L) and treated with insulin for 1 week during an enteroviral meningitis infection. During the past 4 years, he has twice presented with hyperglycemia, during a rotavirus infection and after surgery to repair his congenital heart defect. Now 5 years of age, his fecal elastase level is low (61 µg/g, normal range >200 µg/g), but he is not receiving insulin or pancreatic enzyme supplementation treatment.

Five patients had inherited their GATA6 mutation from a parent (Fig. 2). The parent with a mosaic mutation is not diabetic at 43 years of age, but the four heterozygotes were diagnosed with diabetes at 12–46 years of age and treated with insulin (three case subjects) or metformin. The parent diagnosed at 12 years of age tested negative for GAD autoantibodies and is receiving a nonreplacement dose of insulin (0.35 units/kg/day). Fecal elastase levels were measured in three parents and showed subclinical exocrine insufficiency (40 μg/g) in one of them (ISPAD 212-02).

All the patients with NDM and three of five parents with a GATA6 mutation also show extrapancreatic features (see Table 1 and Fig. 2). Congenital heart defects are most common, present in 21 of 24 probands and three parents (83%), and required surgical correction in 19 case subjects.

Additional extrapancreatic features in the probands include congenital hypothyroidism (n = 6), hepatobiliary malformations (in particular gallbladder agenesis and biliary atresia, n = 5), and gut abnormalities (intestinal malrotation and hernias, n = 6). Eleven patients also have a significant neurocognitive deficit. Three parents had cardiac malformations, but no other features were reported.

This study of a large cohort of patients with NDM confirms that GATA6 mutations are the major cause of pancreatic agenesis (1). Familial studies showed that heterozygous GATA6 mutations can also cause adolescent/adult-onset diabetes. Four parents were diagnosed with diabetes at 12–46 years of age and treated with insulin (n = 3) or metformin, and none had received pancreatic enzyme supplements. Fecal elastase measurement in two diabetic parents showed subclinical exocrine insufficiency in one and a normal level in the other. Our data therefore extend the diabetic phenotypic spectrum of GATA6 mutations from pancreatic agenesis to adult-onset diabetes without exocrine insufficiency.

This study describes 29 subjects (24 probands and 5 parents) with GATA6 mutations from an international cohort of 795 subjects with NDM Within this large cohort, GATA6 mutations account for at least 3% of NDM cases. Of 24 probands with GATA6 mutations, 21 had pancreatic agenesis, and GATA6 mutations were present in 21 of 39 (54%) patients with pancreatic agenesis. Mutations were also identified in three patients with NDM who were not receiving enzyme supplementation; fecal elastase was measured in two case subjects and showed subclinical exocrine insufficiency.

The mutation in the four diabetic parents appeared heterozygous in leukocyte DNA samples, consistent with a spontaneous mutation in a parental germ cell rather than a postzygotic mosaic mutation in the developing embryo. One parent was shown to be a somatic mosaic (suggesting a postzygotic event) and was not diabetic at 43 years of age, although she had surgery to correct a patent ductus arteriosus as a child. Her risk of developing diabetes will depend on the mutational load in the pancreas.

These data confirm previous evidence of the essential function of GATA6 in the development of the human pancreas. In mouse models, GATA6 expression was significantly increased during endocrine differentiation (22), suggesting that it might also be involved in the last stages of β-cell development. The variation in the pancreatic phenotype is also seen with heterozygous mutations in HNF1B, another β-cell transcription factor associated with early pancreatic development (23).

Consistent with the initial discovery of GATA6 mutations as a rare cause of isolated congenital heart defects (2–4), cardiac malformations were the most common extrapancreatic feature (83% of case subjects with a GATA6 mutation) in our cohort. Studies in mouse models have confirmed the role of GATA6 as one of the master regulators of heart formation during embryo development (24). Our study extends the phenotypic spectrum of GATA6 mutations from isolated congenital heart defects to pancreatic agenesis with cardiac malformations and adult-onset diabetes with/without heart defects.

In addition to congenital heart defects, a wide range of extrapancreatic malformations were observed in patients with GATA6 mutations. The most common features were hypothyroidism, hepatobiliary malformations (mostly gallbladder agenesis and biliary atresia), and gut abnormalities (mostly congenital hernias). All these defects affect organs of endodermal origin, suggesting a defect in early embryonic differentiation. Gata6^−/− mice die early during embryo development (embryonic day 6.5–7.5) and show a specific defect in endodermal differentiation, including severely downregulated expression of endodermal markers such as Gata4, Hnf4a, Hnf3b, and Afp (25). Our data suggest a role for GATA6 in the development of endodermal-derived organs in humans too.

This study shows that GATA6 mutations cause a variable diabetic phenotype, ranging from pancreatic agenesis to adult-onset diabetes. Most, but not all, patients have exocrine insufficiency and other extrapancreatic features. Further work is needed to establish the prevalence of GATA6 mutations as a cause of monogenic diabetes presenting outside the neonatal period.

The research leading to these results received funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement number FP7-PEOPLE-ITN-2008 (Marie Curie Initial Training Networks, Biology of Liver and Pancreatic Development and Disease) and grant agreement number 223211 (Collaborative European Effort to Develop Diabetes Diagnostics, CEED3), Diabetes UK (ref. 11/0004193), and the Wellcome Trust. S.E. and A.T.H. are employed as core members of staff within the National Institute for Health Research (NIHR)–funded Peninsula Clinical Research Facility. A.T.H. is an NIHR senior investigator.

No potential conflicts of interest relevant to this article were reported.

E.D.F. researched the molecular genetics data, contributed to discussion, and wrote the manuscript. C.S.-S. and M.H.S. researched the clinical data, contributed to discussion, and reviewed the manuscript. S.E.F. researched the molecular genetics data, contributed to discussion, and reviewed the manuscript. The International NDM Consortium authors identified the patients, researched the clinical data, and reviewed the manuscript. A.T.H. directed the research, contributed to discussion, and reviewed the manuscript. S.E. directed the research, contributed to discussion, and wrote the manuscript. A.T.H. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

The authors thank the families who participated in this study.