Diabetes Mellitus, Transient Neonatal, 1
A number sign (#) is used with this entry because transient neonatal diabetes mellitus-1 (TNDM1; '6q diabetes') is caused by overexpression of the paternal allele of the imprinted locus at chromosome 6q24, which contains only 2 expressed genes, PLAGL1 (603044) and HYMAI (606546).
Some TNMD1 cases are caused by homozygous or compound heterozygous mutations in the ZFP57 gene (612192) at chromosome 6p22, which affects the methylation status of the imprinted locus at 6q24.
DescriptionNeonatal diabetes mellitus (NDM), defined as insulin-requiring hyperglycemia within the first month of life, is a rare entity, with an estimated incidence of 1 in 400,000 neonates (Shield, 2000). In about half of the neonates, diabetes is transient and resolves at a median age of 3 months, whereas the rest have a permanent form of diabetes (606176). In a significant number of patients with transient neonatal diabetes mellitus, type II diabetes appears later in life (Arthur et al., 1997).
The major cause of transient neonatal diabetes (TND) is aberrant expression of imprinted genes at chromosome 6q24, associated in 20% of cases with DNA hypomethylation at the TND differentially methylated region (DMR), which lies within the imprinted promoter of the PLAGL1 gene (603044; Mackay et al., 2005). Over 50% of individuals with TND and hypomethylation at 6q24 also show mosaic DNA hypomethylation at other imprinted loci throughout the genome and a range of additional clinical features.
Genetic Heterogeneity of Transient Neonatal Diabetes
TNDM2 (610374) is caused by mutation in the ABCC8 gene (600509) on chromosome 11p15.1. TNDM3 (610582) is caused by mutation in the KCNJ11 gene (600937), also located on 11p15.1.
Clinical FeaturesTemple et al. (1996) reviewed the manifestations of transient neonatal diabetes mellitus and the evidence for an imprinted, paternally expressed gene on chromosome 6q22-q23. They reported that TNDM occurs with a frequency of approximately 1 in 500,000 births. Patients are born with intrauterine growth retardation and present within the first 6 weeks of life with severe failure to thrive, hyperglycemia, and dehydration. Temple et al. (1996) noted that there is evidence for failure of insulin (176730) production in response to glucose feeding and that insulin therapy is usually required. The condition usually resolves within the first 6 months of life. However, there is a predisposition toward type 2 (insulin resistant) diabetes (see 601283) later in life.
Christian et al. (1999) reported 2 patients who presented at birth with neonatal diabetes mellitus (NDM): one with paternal uniparental disomy for chromosome 6 and one with normal, biparental inheritance. The first child presented with low birth weight, macroglossia, hypertelorism, and clubfoot in addition to NDM. In this patient hyperglycemia was transient, and insulin treatment was discontinued at 4 months of age. The second child also presented with low birth weight but was normal in appearance, and insulin dependence continued after 5 years. Genetic analysis with polymorphic DNA markers for chromosome 6 indicated the presence of paternal uniparental disomy (UPD6) in the first case and normal, biparental inheritance in the second case. Christian et al. (1999) found reports of 8 previous cases of UPD6 of which 6 showed NDM. Three cases with paternal UPD6 also included additional anomalies, such as macroglossia, not usually associated with NDM. Christian et al. (1999) suggested, therefore, that the simultaneous finding of NDM and macroglossia should be a strong indicator for genetic testing. The genetic finding of paternal UPD6 allows prediction of a transient, rather than a permanent, form of diabetes mellitus and no increased recurrence risk of transient NDM in subsequent pregnancies.
Marquis et al. (2000) described 2 patients who suffered from transient neonatal diabetes mellitus (TNDM) due to paternal isodisomy of chromosome 6. One patient, 5 years old at the time of report, had severe intrauterine growth retardation, but recovered normal growth parameters. The other patient, 12 years old at the time of report, had a normal birth weight but showed impaired postnatal growth; in addition to TNDM, this patient presented with cardiac and thyroid abnormalities.
Mackay et al. (2006) reported 2 unrelated TNDM patients who had loss of maternal methylation both at 6q24 and at the centromeric DMR on 11p15.5 (KCNQ1OT1; 604115), which is involved in imprinting abnormalities in Beckwith-Wiedemann syndrome. Both patients presented with intrauterine growth retardation and TNDM without features of overgrowth. However, both had moderate macroglossia and abdominal wall defects, features occasionally found in both BWS and TNDM.
In a review, Yorifuji et al. (2018) stated that the most important clinical feature of chromosome 6q24-related diabetes mellitus is a small-for-gestational-age birthweight, which reflects the lack of insulin in utero. They noted that many patients relapse after puberty. Diabetes is then permanent and is characterized by diminished insulin secretion without obesity and the absence of autoantibodies, which mimics maturity-onset diabetes of the young (MODY; see 125850).
DiagnosisBisulfite sequencing of the differentially methylated region (DMR) of 6q24 facilitated development of a diagnostic test for TNDM based on ratiometric methylation-specific PCR. Mackay et al. (2005) applied this method to 45 cases of TNDM, including 12 with paternal UPD6, 11 with an isolated methylation mutation at 6q24, 16 with a duplication of 6q24, and 6 of unknown etiology, together with 29 normal controls. All were correctly assigned.
MappingMackay et al. (2008) studied a transient neonatal diabetes cohort which contained 13 probands with hypomethylation of multiple imprinted loci from 12 families. Six families were consanguineous, including 1 with 2 affected sibs. Genomewide SNP genotyping revealed a single region of homozygosity within 6p22.2-6p21.1 shared between 5 consanguineous pedigrees and the proband of the sixth, who was homozygous by descent.
Mapping to Chromosome 6q
Temple et al. (1996) reported a family with TNDM that showed linkage to D6S310, a marker in the 6q22-q23 region.
Gardner et al. (1999) refined the critical region in transient NDM to a region of 6q24 defined by markers D6S1699 and D6S1010, within an interval of approximately 5.4 Mb. By further sequencing, Gardner et al. (2000) refined the region to a 300 to 400 kb region which contains several CpG islands. At one island, they noted differential DNA methylation between patients with paternal UPD of chromosome 6 and normal controls. In addition, 2 patients with TNDM, in whom neither paternal UPD of chromosome 6 nor duplication of 6q24 had been found, showed a DNA methylation pattern identical to that of patients with paternal UPD of chromosome 6. Control individuals showed a hemizygous methylation pattern. The authors concluded that TNDM can be associated with a methylation change and that there is a novel methylation imprint on chromosome 6 associated with TNDM.
PathogenesisPaternal Isodisomy or Duplication
Temple et al. (1996) reported a family with a duplication in the 6q22-6q23 region associated with TNDM.
The association between TNDM and either paternal isodisomy or duplication of 6q22-q23 raised the possibility of an imprinted gene in this location. Arthur et al. (1997) reported diabetes that developed in a baby girl immediately after birth and resolved after 7 weeks of insulin treatment. Because of her relatively coarse facial features and a protruding tongue, cytogenetic analysis was performed, showing an inverted duplication of 6q: invdup(6)(q22q23). The duplicated segment was located between DNA loci D6S308 and D6S1684. Arthur et al. (1997) concluded that the patient findings supported the assumption that an imprinted gene exists on 6q22-q23. The duplication in the patient reported by Arthur et al. (1997) was of paternal origin and spans the same region as that in a family reported by Temple et al. (1996).
Gardner et al. (1998) analyzed samples from their cohort of patients with transient neonatal diabetes mellitus by uniparental disomy of chromosome 6 using polymorphic microsatellite repeat analysis. They reported the fifth case of paternal uniparental disomy of chromosome 6 associated with classic transient neonatal diabetes mellitus and estimated that uniparental disomy of chromosome 6 accounts for approximately one-fifth of cases of this syndrome.
Since either duplication of a portion of chromosome 6q or uniparental disomy have been associated with transient neonatal diabetes mellitus, overexpression of an imprinted gene in this disorder is suggested. Prior to the report by Das et al. (2000), all patients with transient neonatal diabetes mellitus and uniparental disomy had had complete paternal isodisomy. Das et al. (2000) described a patient with neonatal diabetes, macroglossia, and craniofacial abnormalities who had partial paternal uniparental disomy of chromosome 6 involving the distal portion of 6q (6q24-qter). This observation demonstrated that mitotic recombination of chromosome 6 can also give rise to uniparental disomy and neonatal diabetes, a situation similar to that observed in Beckwith-Wiedemann syndrome (BWS; 130650), another imprinting disorder.
Temple and Shield (2002) reviewed TNDM as a disorder of imprinting. They noted that 3 genetic mechanisms had been shown to result in TNDM: paternal uniparental isodisomy of chromosome 6, paternally inherited duplication of 6q24, and a methylation defect at a CpG island overlapping exon 1 of ZAC/HYMAI.
Hypomethylation of Imprinted Loci
Kant et al. (2005) reported female monochorionic, triamniotic, monozygous triplets, 2 of whom had TNDM. Methylation-specific PCR showed that the 2 affected children had isolated loss of maternal methylation within the TNDM DMR; the third unaffected child had normal PCR results. The discordant phenotype in 2 of 3 triplets suggested that the imprinting error may have preceded twinning in this case.
Arima et al. (2001) showed that the differentially methylated CpG island that partially overlaps Zac1 (PLAGL1; 603044) and Hymai (606546) at the mouse locus syntenic for 6q24 is a likely imprinting control region (ICR) for the 120- to 200-kb domain. The region is unmethylated in sperm but probably methylated in oocytes, a difference that persists between parental alleles throughout pre- and postimplantation development. Within this ICR, there is a region that exhibits a high degree of homology between mouse and human and acts as a strong transcriptional repressor when methylated. In 5 of 6 TNDM patients studied with a normal karyotype, loss of methylation at 8 CpG sites within the region was demonstrated. ZAC/PLAGL1 is a transcriptional regulator of the type 1 receptor for pituitary adenylate cyclase-activating polypeptide (102981), potent known insulin secretagogue and an important mediator of autocrine control of insulin secretion in the pancreatic islet. The authors proposed that the ICR adjacent to ZAC may regulate expression of imprinted genes within the domain, and that epigenetic or genetic mutations of this region probably result in TNDM by affecting expression of ZAC in the pancreas and/or the pituitary.
Mackay et al. (2006) performed DNA methylation analysis on a cohort of 12 patients with TNDM and total loss of maternal methylation on 6q24. Six of these patients showed a spectrum of methylation loss that was mosaic with respect to the extent of the methylation loss, the tissues affected, and the genetic loci involved. These patients had higher birth weight and were more phenotypically diverse than other TNDM patients, presumably reflecting the influence of dysregulation of multiple imprinted genes. Mackay et al. (2006) proposed the existence of a maternal hypomethylation syndrome and suggested that any patient with methylation loss at 1 maternally-methylated locus might also manifest methylation loss at other loci, potentially complicating or even confounding the clinical presentation.
Molecular GeneticsThe mosaic DNA hypomethylation of multiple imprinted loci throughout the genome in over 50% of individuals with transient neonatal diabetes suggested disruption of imprinting not in the germline but during very early embryonic development. Using genomewide SNP genotyping, Mackay et al. (2008) identified a single region of homozygosity within chromosome 6p22.2-6p21.1 common to 5 consanguineous pedigrees and a proband who was homozygous by descent. On the basis of selection under the twin criteria of involvement in DNA binding or transcription regulation and expression in mouse oocyte or early zygote, and because of its expression in undifferentiated stem cell lineages and downregulation upon stem cell differentiation, Mackay et al. (2008) chose ZFP57 (612192) as the best candidate among the genes in the critical interval on 6p22.2-6p21.1. They identified ZFP57 mutations in 7 families with hypomethylation of multiple imprinted loci. Missense, nonsense, and frameshift mutations were identified. In addition to showing the cardinal epigenetic feature of TND, hypomethylation of the TND differentially methylated region (DMR), individuals with ZFP57 mutations were all hypomethylated at the PEG3 (601483) and GRB10 (601523) differentially methylated regions (DMRs). Hypomethylation of the KCNQ1OT1 (604115) and PEG1 (601029) DMRs was found in individuals with or without mutations in ZFP57. The diabetic presentation was typical for 6q TND in the 8 infant probands.
Mackay et al. (2008) found that 6 of 9 individuals with TNDM and ZFP57 mutations had additional clinical features atypical of TNDM associated with hypomethylation at 6q24. Developmental delay was noted in 6 cases, including 2 with severe developmental delay and hypoplasia of the corpus callosum; however, ZFP57 mutation was also compatible with normal intelligence and development. Three probands were reported to have cardiac defects. These anomalies are less prevalent among individuals with hypomethylation of multiple imprinted loci but without mutations in ZFP57.
Genotype/Phenotype CorrelationsIn a study of 97 patients with diabetes diagnosed in the first 6 months of life whose diabetes remitted before the age of 5 years, 64 of whom had previously been reported, Flanagan et al. (2007) found that 70 of 97 (72%) had an abnormality at the 6q24 locus: 27 (39%) had paternal uniparental isodisomy, 28 (40%) had a duplication of the paternal allele, and 15 (22%) had a methylation defect. Of the remaining 28 TNDM patients without 6q24 abnormalities, 13 were found to have a mutation in the ABCC8 gene (600509) and 12 in the KCNJ11 gene (600937); in 3 patients no mutation was found. There were no significant differences in clinical characteristics between patients with ABCC8 and KCNJ11 mutations, whereas patients with a 6q24 abnormality had a significantly lower birth weight and an earlier diagnosis and remission than patients with K(ATP) channel mutations (p less than 0.001 for all). Atypical phenotypes were observed in all 3 patients in whom the genetic etiology was not defined, including premature birth, epilepsy, and multiple episodes of relapse and remission.
Of 16 Japanese patients with TNDM, Suzuki et al. (2007) identified the 6q24 abnormality in 11 and a KCNJ11 mutation in 2; of 15 patients with PNDM, they identified a KCNJ11 mutation in 7 and an ABCC8 mutation in 2. Compared to patients with a KCNJ11 mutation, patients with the 6q24 abnormality had earlier onset of diabetes, a lower frequency of diabetic ketoacidosis at onset, and a higher proportion of patients with macroglossia at initial presentation.
Diatloff-Zito et al. (2007) reported a group of 13 sporadic transient neonatal diabetes cases, including 5 with birth defects (congenital abnormalities of heart, brain, and bone) and 8 without. Two of the patients had paternal uniparental disomy-6 (UPD6); of the remaining 11 cases, 2 had complete and 3 had partial loss of the maternal methylation signature upstream of the ZAC1-HYMAI imprinted genes in non-UPD cases. There was 1 case of hemizygous deletion among all 13 cases, in a patient with severe congenital malformations. Diatloff-Zito et al. (2007) raised the hypothesis that the deletion had an effect on regulatory elements critical for imprinting and tissue-specific gene expression in early development.
Animal ModelThe inheritance pattern of transient neonatal diabetes mellitus implicates overexpression of 1 or both genes within the TNDM locus: ZAC (603044), which encodes a proapoptotic zinc finger protein, and HYMAI (606546), which encodes an untranslated mRNA. To investigate the consequences for pancreatic function, Ma et al. (2004) developed a high-copy transgenic mouse line carrying the human TNDM locus. Neonates of this line displayed hyperglycemia, and older adults, glucose intolerance. Neonatal hyperglycemia occurred only on paternal transmission, analogous to paternal dependence of TNDM in humans. The embryonic pancreata of these mice showed reductions in expression of endocrine differentiation factors and numbers of insulin-staining structures. By contrast, beta-cell mass was normal or elevated at all postnatal stages, whereas pancreatic insulin content in neonates and peak serum insulin levels after glucose infusion in adults were reduced. Expression of human ZAC and HYMAI in these transgenic mice thus recapitulated key features of TNDM and implicated impaired development of the endocrine pancreas and beta-cell function in disease pathogenesis.
Other FeaturesHurst and McVean (1997) examined the conflict theory for the evolution of genomic imprinting. The theory proposes that imprinting is an intraindividual manifestation of classic parent-offspring conflict. The theory predicts that imprinted genes expressed from the paternally derived genome should be enhancers of pre- and postnatal growth, while those expressed from the maternally derived genome should be growth suppressors. Hurst and McVean (1997) examined this prediction by reviewing the literature on growth of human and mouse progeny that inherited both copies (or part thereof) of a particular chromosome from only 1 parent. They found that much of the data do not support the conflict theory hypothesis. They pointed to paternal uniparental disomy of chromosome 6 (UPD6), which is associated with severe growth retardation. The conflict theory would suggest that growth should be enhanced. In their Figure 1, they illustrated the patient reported by Ferguson and Milner (1970).