Unlocking the secrets of the pancreatic β cell: man and mouse provide the key

Abstract

Failure of the pancreas to secrete sufficient insulin results in type 2 diabetes, but the pathogenesis of pancreatic β cell dysfunction is still poorly understood. New insights into β cell failure come from defining the genes involved in rare genetic subtypes of diabetes and creating appropriate animal models. A new mouse model of transient neonatal diabetes mellitus emphasizes that both the number of β cells and their function are critical for insulin secretion and may be regulated by imprinted genes.

The regulated secretion of insulin by the pancreatic β cell maintains blood sugar concentrations within a narrow physiological range. In over 150 million people worldwide, however, pancreatic β cells fail to secrete adequate insulin, usually in the presence of increased insulin resistance, which results in type 2 diabetes (T2D). Understanding the pathways that result in β cell dysfunction at a physiological and molecular level is critical for improved understanding and treatment of T2D.

Learning from rare genetic subtypes of diabetes

How can we study the pancreatic β cell in humans when these cells are not readily accessible? Accidents of nature in which a single gene defect results in severe β cell dysfunction, causing diabetes, offer the chance of gaining new insights into this disease if the responsible gene can be defined. The best example of such research has been the use of positional cloning to demonstrate that heterozygous mutations of the genes encoding the hepatic transcription factors HNF-1α and HNF-4α cause early-onset diabetes (1, 2). Subsequent studies have allowed the unraveling of a previously unexpected transcription factor network that is crucial to the maintenance of normal β cell development and function (3) and also involved in the susceptibility to T2D (4, 5). Although genetics provided the initial breakthrough, subsequent careful animal and molecular biological studies were needed to elucidate the underlying mechanism.

Transient neonatal diabetes: a disorder of imprinting in humans

Now, studies of the molecular genetics of transient neonatal diabetes mellitus (TNDM) in humans and mice have been combined to give new insights into the development and physiology of the β cell. TNDM is a rare condition (affecting approximately 1 in 600,000 live births) that is characterized by a unique clinical course (6). Affected babies have low birth weight, and high blood-glucose values are detected in the first week of life — features of low pancreatic insulin secretion in utero and after delivery, respectively. Initially, insulin treatment is needed, but by 12 weeks, endogenous insulin secretion has usually improved sufficiently to allow its discontinuation. Patients remain in apparent remission for many years, but 2/3 of them will subsequently develop diabetes, usually in adolescence. Their diabetes at this stage, despite their age and lack of obesity, is similar to T2D, with a loss of first-phase insulin secretion (7).

The first major clue to the etiology of this disappearing and reappearing diabetes came from genetic analysis implicating abnormalities of an imprinted locus on chromosome 6 (8). Three interrelated genetic mechanisms have been found to cause most TNDM (reviewed in ref. 6): (a) inheriting 2 copies of the paternal chromosome 6 (paternal uniparental isodisomy of chromosome 6); (b) paternally inheriting a duplication of 6q24; or (c) a maternal methylation defect within this region. These data are consistent with TNDM resulting from biallelic expression rather than the normal paternal monoallelic expression that results from methylation and hence inactivation of the maternally inherited allele. There are 2 overlapping imprinted genes with maternal allele silencing in the TNDM locus: ZAC (Z finger protein that regulates apoptosis and cell cycle arrest) and HYMAI (hydatidiform mole–associated and –imprinted transcript) (9). Overexpression of one or both of these genes could be responsible for the TNDM phenotype.

New insights from the mouse model TNDM29

This genetic information has allowed an excellent rodent model of TNDM to be created. Ma and colleagues, in this issue of the JCI (10), describe a transgenic mouse (TNDM29) with overexpression of a P1-derived artificial chromosome (PAC) containing the complete ZAC and HYMAI human genes. In keeping with the observation that TNDM in humans only results from the paternal inheritance of the duplication of 6q24, offspring generated by paternal transmission of the overexpressing PAC, but not maternal transmission, were hyperglycemic as neonates. The glycemia changes in TNDM29 mice mirrors those in TNDM in humans, with remission and normal glucose tolerance in juvenile mice followed by relapse and glucose intolerance in adulthood (Table 1).

Table 1

Comparison of the clinical features and pathophysiology related to glucose regulation in human TNDM with the transgenic mouse model, TNDM29

The initial studies of the TNDM29 transgenic mice have already suggested possible underlying mechanisms for the recurring β cell failure. The most fascinating findings in the TNDM model are the marked changes in β cell number compared with wild-type mice and the relationship of these changes in β cell number to the varying glucose tolerance and insulin secretion (Table 1). In the pancreata of the early embryonic TNDM29 transgenic mice, there was a marked reduction in the number of β and other pancreatic endocrine cells. This effect was probably mediated, at least partially, by downregulation of critical pancreatic transcription factors Pdx1, Ngn3, and Pax6. In late gestation and early postnatal life, there was a rapid increase in pancreatic β cell mass in the TNDM29 mouse, achieved primarily by an increased number of β cells (either through increased proliferation or decreased apoptosis), which help at least in part to compensate for the low initial number of β cells. Despite this, in the early neonate, the total insulin content of the pancreas was still reduced, and the animal was hyperglycemic as a result of inadequate insulin secretion. The number of β cells continued to increase and by 2–3 months (juvenile) was approximately twice the number observed in wild-type mice, although the total insulin content of the pancreas was unchanged, which suggests that each β cell contains less insulin. The increased number of β cells enabled normal glucose tolerance. However, the compensatory increase in β cell mass was not maintained, and adult TNDM29 mice had a β cell mass similar to that in wild-type animals. The glucose tolerance of the adult animals deteriorated and was characterized by reduced early insulin secretion. A key result is that disordered imprinting, like mutations in transcription factors (4), has led to both altered development and altered function of β cells.

New directions

As with all good science, these studies have raised more questions than they have answered. Why is the insulin deficiency less severe in mouse than human? Does the phenotype result only from the overexpression of ZAC, a potent cell cycle regulator, or is increased expression of HYMAI — an apparently untranslated mRNA of unknown function — also needed? Does the rapid increase in β cell mass in late intrauterine and early postnatal life represent secondary compensatory mechanisms, or is it directly mediated by ZAC/HYMAI? Is it the failure of β cell function as an adult a consequence of rapid compensation in early life? If the latter scenario is true, there could be parallels with fetal exposure to hyperglycemia in utero resulting in glucose intolerance as an adult. Again, the combination of human genetics and a resultant mouse model offers the opportunity for discovery of many more of the secrets of the pancreatic β cell.

Acknowledgments

Andrew Hattersley is a Wellcome Trust research leave fellow.

Footnotes

See the related article beginning on page 339.

Nonstandard abbreviations used: hydatidiform mole–associated and –imprinted transcript (HYMAI); P1-derived artificial chromosome (PAC); transient neonatal diabetes mellitus (TNDM); type 2 diabetes (T2D); Z finger protein that regulates apoptosis and cell cycle arrest (ZAC).

Conflict of interest: The author has declared that no conflict of interest exists.

References

1. Yamagata, K, et al. Mutations in the hepatic nuclear factor 1 alpha gene in maturity-onset diabetes of the young (MODY3). Nature. 1996. 384:455-458.
   View this article via: PubMed CrossRef Google Scholar
2. Yamagata, K, et al. Mutations in the hepatocyte nuclear factor 4 alpha gene in maturity-onset diabetes of the young (MODY1). Nature. 1996. 384:458-460.
   View this article via: PubMed CrossRef Google Scholar
3. Servitja, JM, Ferrer, J. Transcriptional networks controlling pancreatic development and beta cell function (review). Diabetologia. 2004. 47:597-613.
   View this article via: PubMed CrossRef Google Scholar
4. Silander, K, et al. Genetic variation near the hepatocyte nuclear factor-4 alpha gene predicts susceptibility to type 2 diabetes. Diabetes. 2004. 53:1141-1149.
   View this article via: PubMed CrossRef Google Scholar
5. Love-Gregory, LD, et al. A common polymorphism in the upstream promoter region of the hepatocyte nuclear factor-4 alpha gene on chromosome 20q is associated with type 2 diabetes and appears to contribute to the evidence for linkage in an Ashkenazi Jewish population. Diabetes. 2004. 53:1134-1140.
   View this article via: PubMed CrossRef Google Scholar
6. Temple, IK, Shield, JP. Transient neonatal diabetes, a disorder of imprinting. J. Med. Genet. 2002. 39:872-875.
   View this article via: PubMed CrossRef Google Scholar
7. Shield, JP, et al. Variable evidence of pancreatic insufficiency in those children with transient neonatal diabetes in remission [abstract]. Pediatr. Res. 2001. 49:P2-619.
   View this article via: CrossRef Google Scholar
8. Temple, IK, et al. An imprinted gene(s) for diabetes? Nat. Genet. 1995. 9:110-112.
   View this article via: PubMed CrossRef Google Scholar
9. Kamiya, M, et al. The cell cycle control gene ZAC/PLAGL1 is imprinted--a strong candidate gene for transient neonatal diabetes. Hum. Mol. Genet. 2000. 9:453-460.
   View this article via: PubMed CrossRef Google Scholar
10. Ma, D, et al. Impaired glucose homeostasis in transgenic mice expressing the human transient neonatal diabetes mellitus locus, TNDM. J. Clin. Invest. 2004. 114:339-348. doi:10.1172/JCI200419876.
    View this article via: JCI PubMed Google Scholar