Monogenic and Syndromic Diabetes due to Endoplasmic Reticulum Stress

Abstract

The endoplasmic reticulum (ER) lies at the crossroads of protein folding, calcium storage, lipid metabolism, and the regulation of autophagy and apoptosis. Accordingly, dysregulation of ER homeostasis leads to β-cell dysfunction in type 1 and type 2 diabetes that ultimately culminates in cell death. The ER is therefore an emerging target for understanding the mechanisms of diabetes mellitus that captures the complex etiologies of this multifactorial class of metabolic disorders. Our strategy for developing ER-targeted diagnostics and therapeutics is to focus on monogenic forms of diabetes related to ER dysregulation in an effort to understand the exact contribution of ER stress to β-cell death. In this manner, we can develop personalized genetic medicine for ER stress-related diabetic disorders, such as Wolfram syndrome. In this article, we describe the phenotypes and molecular pathogenesis of ER stress-related monogenic forms of diabetes.

Introduction

Endoplasmic Reticulum as an Emerging Target for Diagnosis and Treatment of Diabetes Mellitus

Diabetes mellitus is a spectrum of disorders, ranging from absolute insulin deficiency to relative insulin deficiency, with heterogenous etiology. It has been now established that dysregulation of endoplasmic reticulum (ER) homeostasis underlies β cell dysfunction and death in type 1 and type 2 diabetes^1–6. Despite the underlying importance of ER dysfunction in diabetes, there are currently no diabetes treatments targeting the ER. Recent clinical and genetic studies have identified several monogenic forms of syndromic diabetes caused by mutations in genes involved in key components of the ER. Although these are rare diseases, their monogenic etiology may give us a clue to identify novel diagnostic and therapeutic targets for diabetes. Here we summarize the symptoms and molecular pathogenesis of monogenic and syndromic diabetes due to ER stress.

Endoplasmic Reticulum Stress and the Unfolded Protein Response (UPR)

The endoplasmic reticulum (ER) is a cellular organelle responsible for protein folding, calcium storage, and lipid metabolism. This organelle also integrates numerous other molecular pathways and contributes to reduction-oxidation regulation, autophagy, and cell death⁷. Dysregulation of ER homeostasis, especially the imbalance between protein-folding capacity and protein-folding load, causes ER stress⁸. An adaptive response to ER stress is the Unfolded Protein Response (UPR), which decreases the synthesis of new proteins while increasing the functional capacity of the ER⁹. When ER stress is too great, cells undergo apoptosis in order to prevent release of potentially harmful misfolded proteins².

The UPR is controlled by three master regulators: protein kinase R-like ER kinase (PERK), activating transcription factor 6 (ATF6), and inositol requiring enzyme 1 (IRE1). PERK is an ER transmembrane protein kinase whose N-terminal domain detects the accumulation of unfolded proteins within the lumen of the ER. In the presence of ER stress, PERK oligomerizes, promoting its own activation through trans-autophosphorylation, thereby allowing activated PERK to phosphorylate eukaryotic translational initiation factor 2 (eIF2α). This results in the general attenuation of mRNA translation, thereby reducing the translational burden of the ER¹⁰. ATF6 is a transcription factor localized to the ER membrane. In the setting of ER stress, ATF6 translocates from the ER to the Golgi where it undergoes proteolytic cleavage and activation (Figure 1B). Cleaved ATF6 translocates to the nucleus, leading to upregulation of various molecules involved in protein folding and protein degradation, including the ER chaperone immunoglobulin heavy chain-binding protein (BiP), which enhances ER folding capacity¹¹.

Figure 1. Molecular Mechanisms of ER Stress in Syndromic forms of Diabetes.

A) Under normal conditions, immunoglobulin heavy chain-binding protein (BiP) is bound to PRK-like endoplasmic reticulum kinase (PERK) and ATF6. The eukaryotic initiation factor 2 (eIF2) complex serves as the rate limiting step in mRNA translation. p58^IPK negatively regulates PERK, while CReP negatively regulates the eIF2 complex. WFS1 works to inhibit ATF6 and assist in ER calcium regulation. CISD2 negatively regulates calpain 2 (CAPN2), an ER calcium sensitive protease, that when active can stimulate ER stress mediated cell death. B) Under ER stress conditions, BiP binds to misfolded proteins. Unbound PERK dimerizes and undergoes trans autophosphorylation and oligomerization. Subsequent phosphorylation of eIF2γ leads to attenuation of protein translation. Additionally, phosphorylation of eIF2γ leads to induction of ATF4 and CHOP, which serve as critical transcription factors in the ER stress response. IER3IP1 stimulates coat protein complex II (COPII) vesicles which transport unbound ATF6 to the Golgi. There ATF6 undergoes intramembrane proteolysis. The luminal side of the protein is retained in the ER, while the cytosolic fragment of ATF6 goes on to lead to upregulation of ER stress response genes. Gene mutated in syndromic forms of diabetes and ER stress are demonstrated in by orange stars with the disease name highlighted in orange. PNDM = Permanent neonatal diabetes mellitus; WRS = Wolcott-Rallison Syndrome; MEDS = Microcephaly, Epilepsy, and Diabetes Syndrome; Wolfram 1 = Wolfram syndrome 1, Wolfram 2 = Wolfram syndrome 2; ACPHD = Ataxia, Combined Cerebellar and Peripheral, with Hearing Loss, and Diabetes Mellitus; MSSGM2 = Microcephaly, short stature, and impaired glucose metabolism-2; MEHMO = Mental Retardation, Epileptic Seizures, Hypogonadism, and Hypogenitalism, Microcephaly, and Obesity. Pathways altered by ER stress are demonstrated by gray lines. Yellow stars indicate a phosphorylated state.

IRE1 is another transmembrane protein kinase localized to the ER. IRE1 also has two isoforms, IRE1α and IRE1β. IRE1 activation leads to the unconventional splicing of X-box binding protein 1 (XBP1) mRNA, producing spliced XBP1 (sXBP1) which encodes a transcription factor that upregulates genes encoding folding enzymes, molecular chaperones, and components of the ER-associated degradation (ERAD) machinery. Additionally, IRE1 decreases the synthetic demand on the ER by activating IRE1-dependent decay of ER-associated mRNAs¹². WFS1 and CISD2 are two other ER stress-modifying proteins localized to the ER. Although their exact functions are not entirely known, WFS1 is thought to negatively regulate ATF6 through ubiquitinization and proteasomal degradation¹³. CISD2 negatively regulates calpain 2, which when activated promotes ER stress-mediated cell death (Figure 1A)¹⁴.

ER Stress and Diabetes

The ER of insulin producing β-cells is highly developed, as it needs to continuously produce insulin throughout a given organism’s lifespan. Each β-cell is estimated to produce 1 million molecules of insulin each minute¹⁵. Insulin synthesis requires a complex cascade of biosynthetic events within the ER¹⁶. These events include cleavage of the signal sequence from pre-proinsulin to form pro-insulin, formation of two disulfide bonds, folding of pro-insulin into its native shape, trafficking of pro-insulin into the Golgi and secretory granules, and cleavage of C-peptide in order to form mature insulin¹⁷. The UPR is known to be involved in insulin production, and ER homeostasis plays a crucial role in β-cell health^{1, 2, 18}. As a result, ER stress is implicated in the pathogenesis of both type 1 and type 2 diabetes, as well as in several monogenic forms of diabetes, including Wolfram syndrome^19–21.

Genetic Syndromes of ER Stress & Diabetes

Several monogenic forms of ER stress disorders have now been described. These conditions have been critical for enhancing our understanding of the biology of β-cells and insulin secretion, and the roles of ER stress in human disease. The known conditions of ER stress and diabetes are summarized below and in Table 1.

Table 1.

Genetic Syndromes of Diabetes.

Gene	Protein	Syndrome	Inheritance
INS	Insulin	Permanent Neonatal Diabetes Mellitus	Autosomal domiant
EIF2AK3	PERK	Walcott-Rallison Syndrome	Autosomal recessive
EIF2S	eIF2γ	MEHMO Syndrome	X-linked recessive
PPP1R15B	CReP	MSSGM2	Autosomal recessive
DNAJC3	p58IPK	ACPHD	Autosomal recessive
IER3IP1	IER3IP1	MEDS	Autosomal recessive
WFS1	Wolframin	Wolfram Syndrome 1	Autosomal recessive*
CISD2	CISD2	Wolfram Syndrome 2	Autosomal recessive

MEHMO = Mental Retardation, Epileptic Seizures, Hypogonadism, and Hypogenitalism, Microcephaly, and Obesity; MSSGM2 = Microcephaly, short stature, and impaired glucose metabolism-2; ACPHD = Ataxia, Combined Cerebellar and Peripheral, with Hearing Loss, and Diabetes Mellitus; MEDS = Microcephaly, Epilepsy, and Diabetes Syndrome.

^*The majority of Wolfram syndrome 1 cases are autosomal recessive, however a small subset of patients carry only one mutant WFS1 allele.

Wolfram Syndrome and WFS1-related disorder

Wolfram Syndrome Type 1

Wolfram syndrome is a rare genetic form of syndromic diabetes caused by pathogenic variants in the Wolfram syndrome 1 (WFS1) gene^{22, 23}. More than 200 WFS1 variants associated with Wolfram syndrome have been reported. Cardinal manifestations in Wolfram syndrome are insulin-dependent diabetes mellitus, optic nerve atrophy and neurodegeneration^{23, 24}. Various other symptoms are often, but not always seen in patients with Wolfram syndrome, including diabetes insipidus, hearing loss, trigeminal neuralgia-like headaches, dysphagia, neurogenic bladder, anxiety, depression, and other neurologic and psychiatric conditions^{21, 25}. Wolfram syndrome is also referred as DIDMOAD (Diabetes Insipidus Diabetes Mellitus, Optic nerve Atrophy, and Deafness). Diabetes mellitus is typically the first manifestation, usually diagnosed around age 6, and optic nerve atrophy follows around age 11. Diabetes insipidus, sensorineural hearing loss, neurogenic bladder, and obstructive sleep apnea may also develop in the next two decades of life. Dysphagia, ataxia and central sleep apnea associated with cerebellar and brain stem atrophy may develop in later stages of this disease^{21, 24–28}.

Based on our clinical and genetic findings, Wolfram syndrome is best characterized as a spectrum of disorders, ranging in clinical severity from mild to severe (Figure 2)^{20, 29} (https://wolframsyndrome.dom.wustl.edu/). Wolfram syndrome patients carrying recessive and missense variants tend to have milder manifestations. The WFS1 p.R558C missense variant, for example, is associated with mild syndromic manifestations, but has a high carrier frequency (around 3%) in the Ashkenazi Jewish population³⁰.

Figure 2. Wolfram syndrome and WFS1-Related Disorders.

Several related medical conditions are caused by pathogenic WFS1 variants, ranging in clinical severity from mild to severe.

WFS1-related disorders

In addition, some pathogenic variants in the WFS1 gene are associated with a distinct subset of patients who develop only one or a few symptoms seen in Wolfram syndrome. Certain dominant pathogenic variants of the WFS1 gene cause deafness or diabetes alone^31–33. Other dominant WFS1 variants are associated with deafness together with mild optic nerve atrophy³⁴. It has been reported that autosomal dominant congenital cataracts are also associated with dominant variants of WFS1³⁵. We have recently identified several dominant de novo WFS1 variants associated with a genetic syndrome of neonatal/infancy-onset diabetes, congenital sensorineural deafness, and congenital cataracts²⁹. These patients have WFS1-related disorders, not Wolfram syndrome (Figure 2).

Wolfram Syndrome Type 2

Pathogenic variants in the CDGSH iron sulfur domain protein 2 (CISD2) gene have been found in a small fraction of patients^{22, 24, 36} (WFS2). Several CISD2 variants associated with Wolfram syndrome have been reported. CISD2 encodes a transmembrane protein localized to the ER and outer membrane of the mitochondria. The exact function of CISD2 is still not clear, but it has been shown to play a role in the regulation of pro-apoptotic molecule, calpain 2, aging process, and autophagy^{37, 38}. Wolfram syndrome patients carrying pathogenic variants in the CISD2 gene develop cardinal features of Wolfram syndrome, including diabetes mellitus and optic nerve atrophy, but they tend to develop other symptoms that are not typically seen in patients carrying pathogenic WFS1 variants, such as upper gastrointestinal ulceration and bleeding.

Mechanisms of Wolfram Syndrome

Wolfram syndrome is well-recognized as a prototype ER disorder because both WFS1 and CISD2 proteins are localized to the ER and play essential roles in maintaining ER homeostasis^{21, 39}. WFS1 is involved in the regulation of the unfolded protein response (UPR) in response to ER stress at the ER membrane through its interaction with ATF6 protein¹³. WFS1 also controls ER and cellular calcium homeostasis, which is critical for protein folding in the ER and secretion of hormones and neurotransmitters, including insulin¹⁴. In addition, misfolding of pathogenic WFS1 proteins may further disrupt the functions of the ER. Thus, pathogenic WFS1 variants cause pathological ER stress in pancreatic β-cells, neuron, retinal ganglion cells and oligodendrocytes, resulting in the dysfunction and degeneration of affected tissues^{13, 14, 40–42}. ER stress also leads to mitochondrial dysfunction. Wolfram syndrome is thus a systemic disease caused by ER dysfunction and a prototype of ER disorder.

Permanent Neonatal Diabetes Mellitus

Neonatal diabetes mellitus is defined as hyperglycemia requiring insulin within the first 12 months of life. Neonatal diabetes is rare, with an estimated incidence of about 1 in 90,000 – 1 in 160,000 neonates⁴³. Approximately half of these neonates have a transient form of diabetes which resolves at a median age of 3 months. If untreated, the remaining patients continue to experience hyperglycemia and are considered to have permanent neonatal diabetes mellitus (PNDM) (OMIM: 606176)⁴⁴. Other proposed names for this disorder include monogenic diabetes of infancy (MDI)⁴⁵ or specific disease names based on gene mutation⁴⁶.

Many genes have been implicated in the etiology of PNDM. Among these genes are pivotal transcription factors responsible for pancreatic or β-cell development, such as PDX1⁴⁷, PTF1A⁴⁸, and GATA6⁴⁹ and genes critical for insulin secretion, such as GCK⁵⁰, KCNJ11^{51, 52}, and ABCC8⁵³. Additionally, the gene encoding human insulin, INS, has been implicated in both PNDM and an ER stress-related form of diabetes⁵⁴. We have recently reported that WFS1, a causative gene for Wolfram syndrome, is also associated with PNDM.

Infants with PNDM due to INS mutations typically present around 9–11 weeks of life. Oftentimes presenting symptoms include symptomatic hyperglycemia and diabetic ketoacidosis^{54, 55}. All patients require insulin replacement therapy. Diabetes autoantibodies, when measured, are undetectable⁵⁵. Some studies estimate that INS gene mutations occur in approximately 10% of PNDM cases. Some patients with INS mutations present at a later age without other associated symptoms/affected organ systems⁵⁶.

The mutant INS peptide is hypothesized to cause a defect in insulin folding and secretion, that leads to β-cell death and failure to maintain β-cell mass over the first few months of life⁴⁵. Patient registries have identified hotspots for mutations in the INS gene⁵⁵. Recent publications have also identified intronic variants leading to PNDM^{57, 58}. ER stress is thought to occur as mutant proinsulin becomes trapped and accumulates in the ER. This leads to uncontrolled β-cell ER stress and subsequent apoptosis (Figure 1B). This accounts for the severe nature of INS mutations⁵⁴. Studies of insulin gene mutations in mice have helped confirm these hypotheses. The Akita mouse, harbors mutations in the INS2 gene and was first described in 1997. Electron microscopy of the β-cells of these mice demonstrate a dramatic reduction in insulin secretory granules and a remarkably enlarged ER lumen. BiP, a well-known ER stress responsive molecular chaperone, is upregulated is the β-cells of Akita mice⁵⁹.

Wolcott-Rallison Syndrome

The cardinal features of Wolcott-Rallison syndrome (WRS) (OMIM: 226980) include diabetes mellitus which occurs early in infancy and multiple epiphyseal dysplasia⁶⁰. Diabetes typically occurs early in infancy (within the first 6 months of life). It often presents acutely with severe diabetic ketoacidosis^61–63. WRS is quite rare, with fewer than 60 cases described in the literature. A large majority of affected patients are from middle eastern countries, where consanguineous marriages are common. WRS is considered to be the most frequent cause of PNDM in consanguineous families⁶³. As WRS is a monogenic form of diabetes, the typical autoantibodies found in children with type 1 diabetes are not present⁶⁴.

Another cardinal feature of WRS is hepatic dysfunction which manifests as elevated liver enzymes, hepatomegaly, and recurrent acute liver failure⁶⁵. The skeletal phenotype of WRS includes multiple epiphyseo-metaphyseal dysplasia which mainly affects the long bones, pelvis, and spine. Additionally, patients experience osteopenia and are prone to multiple and frequent fractures⁶⁶. Impaired renal function can occur in WRS, along with hepatic manifestations⁶⁴. Some patients have been described to have pancreatic atrophy or exocrine pancreatic dysfunction⁶⁷. Intellectual disability and developmental delays are common⁶⁵ with variable motor deficits, microcephaly, simplified gyral patterns, and epilepsy^{68, 69}. Hypothyroidism^{70, 71}, neutropenia, and recurrent infections are also found in patients with WRS⁶⁵. Skin abnormalities, dental discoloration, and dysmorphic facial features have also been reported as rarer manifestations of WRS^{66, 67, 72}. Case reports describe remarkable clinical variability between patients. Diabetes is diagnosed in the first 6 months, while bone dysplasia occurs within the first 2 years of life. Hepatic disease can occur at any age and may be the next manifestation after the onset of diabetes⁶².

WRS is inherited in an autosomal recessive pattern and is due to mutations in the gene encoding eukaryotic translation initiation factor 2α kinase 3 (EIF2AK3). Nine different pathogenic variants in EIF2AK3 have been reported in patients with WRS. EIF2AK3 is also known as pancreatic EIF2α kinase (PEK) and PKR-like endoplasmic reticulum kinase (PERK)⁷³. PERK is a transmembrane protein kinase located to the ER membrane and plays a critical role in the UPR. PERK together with IRE1 and ATF6 acts a stress sensor in the ER, which detects the accumulation of misfolded proteins and initiates the UPR. Upon activation, PERK phosphorylates the translation initiation factor eIF2α, which in turn reduces protein synthesis. PERK also activates expression of ATF4, and CHOP which are key transcription factors in the UPR (Figure 1B)^{19, 74}. The majority of EIF2AK3 mutations include nonsense/frameshift mutations resulting in truncated protein or missense mutations in the kinase domains of the protein, which leads to reduced phosphorylation of eIF2α⁶². It has been shown that eIF2α phosphorylation in response to ER stress plays a key role in the survival of ER stressed cells^75–77. The attenuation of protein synthesis and client protein translocation into the lumen of the ER by eIF2α phosphorylation protects cells from pathological ER stress⁷⁸. Thus, reduced eIF2α phosphorylation has a negative impact on secretory cell survival and is thought to be the underlying cause of WRS.

Mental Retardation, Epileptic Seizures, Hypogonadism, and Hypogenitalism, Microcephaly, and Obesity (MEHMO) Syndrome

Mental Retardation, Epileptic Seizures, Hypogonadism, and Hypogenitalism, Microcephaly, and Obesity (MEHMO) Syndrome (OMIM: 300148) is another rare syndrome related to ER stress. It was first described in 1998 in a large kindred featuring several affected male children presenting with mental retardation, epilepsy, hypogonadism, hypogenitalism, microcephaly, and obesity (MEHMO)⁷⁹. Affected children had a life expectancy less than 2 years. Due to its inheritance pattern it was thought to be an X-linked recessive condition⁸⁰. MEHMO is extremely rare, with only less than 30 patients described in the medical literature. Some cases describe a pattern of neonatal hypoglycemia, followed by infant onset diabetes⁸¹. Some patients with MEHMO also demonstrate hypopituitarism including growth hormone and thyroid stimulating hormone deficiencies⁸².

Eukaryotic translation initiation factor 2 subunit 3 (EIF2S3 or eIF2γ) has been shown to be the causative gene in MEHMO syndrome⁸³. Several pathogenic variants in the EIF2S3 gene have been reported to be associated with MEHMO syndrome. EIF2S3 is strongly expressed in the developing endocrine pancreas as well as the hypothalamus and pituitary glands⁸⁴. EIF2S3 functions downstream of EIF2AK3 (PERK) as another aspect of the eukaryotic translation initiation factor pathway⁸⁵. As a result, patients with MEHMO syndrome experience increased ER stress and cell death in affected tissues (Figure 1B).

Microcephaly, Short Stature, and Impaired Glucose Metabolism-2 (MSSGM2)

Microcephaly, short stature, and impaired glucose metabolism-2 (MSSGM2) (OMIM: 616817) is an extremely rare genetic syndrome. The few patients with this condition have been reported to have diabetes mellitus, short stature, microcephaly, and intellectual disability. Affected patients seem to develop diabetes in the 2nd to 3rd decade of life⁸⁶. Other symptoms include delayed puberty, growth retardation, sensorineural deafness, liver cirrhosis, kyphoscoliosis, pectus excavatum, oligodontia, dental hypoplasia, and sparse hair^{86, 87}.

Affected patients with MSSGM2 were found to harbor mutations in the PPP1R15B (Protein Phosphatase 1 Regulatory Subunit 15B) gene (also known as constitutive repressor of eIF2α phosphorylation [CReP]) (Figure 1A)⁸⁶. Since the initial cohort was described, 4 additional patients were identified to harbor pathogenic variants in PPP1R15B^{87, 88}. A few pathogenic variants in PPP1R15B have been reported in patients with MSSGM2. These patients were relatively young, so they have yet to demonstrate hyperglycemia. Of note MSSGM2 should not be confused with MSSGM1, which is caused by mutations in the TRM10A gene⁸⁹.

PPP1R15B (a.k.a. CReP) plays a role in eIF2α dephosphorylation⁹⁰. Therefore, loss of function of PPP1R15B leads to increased eIF2α phosphorylation, which in turn induces cell death through the proapoptotic BH3-Only Proteins DP5, PUMA, and Bim⁸⁶. Thus, β cell death mediated by DP5, PUMA, and Bim would be the underlying mechanism of diabetes in MSSGM2. Increased eIF2α phosphorylation leads to β cell death in MSSGM2, whereas reduced eIF2α phosphorylation causes β cell death in Wolcott-Rallison syndrome. These clinical and genetic findings strongly suggest that dysregulation of eIF2α is an emerging mechanism for β cell death in diabetes.

Ataxia, Combined Cerebellar and Peripheral, with Hearing Loss, and Diabetes Mellitus (ACPHD)

Combined cerebellar and peripheral ataxia with hearing loss and diabetes mellitus (ACPHD) (OMIM: 616192) is another extremely rare genetic syndrome. Neurologic features of ACPHD include combined cerebellar and afferent ataxia, mild upper motor neuron damage, demyelinating sensorimotor peripheral neuropathy, and sensorineural hearing loss. Diabetes typically develops in the second decade of life. Neurological symptoms can vary widely and include sensorineural hearing loss, abnormal gait, and generalized cerebral atrophy⁹¹. Some patients experienced primary hypothyroidism, short stature, and low body mass index⁹².

Affected patients with ACPHD were found to have mutations the DNAJC3 gene, which encodes p58^IPK, a co-chaperone for BiP, which works to inhibit EIF2AK3 (PERK) (Figure 1A). Therefore, mutations in DNAJC3 lead to increased ER stress and apoptosis in β-cells⁹³. A few pathogenic variants in DNAJC3, as well as deletions of DNAJC3 locus, have been reported in patients with ACPHD.

Microcephaly, Epilepsy, and Diabetes Syndrome (MEDS)

Microcephaly with a simplified gyral pattern, epilepsy, and infant onset diabetes are the cardinal features of microcephaly, epilepsy, and diabetes syndrome (MEDS) (OMIM: 614231). This disease was first described in 2011, in a small cohort of patients symptomatically resembling WRS⁹⁴. All known patients with MEDS develop severe, refractory insulin dependent diabetes. They have congenital microcephaly (occipitofrontal circumference < 2–3 SD below the mean), with progressive head growth deceleration⁹⁵. Furthermore, each case has severe motor deficits and diminished vision/hearing. Epilepsy develops within the first 2 months of life with burst suppression and simplified gyral pattern on MRI. Agenesis/hypoplasia of the corpus callosum and hypoplastic cerebellar vermis are frequent findings. Some children with MEDS harbor skeletal defects, however this has not been specifically documented in all cases. Unfortunately, most infants succumb to respiratory infections and die at an early age⁹⁶.

MEDS is exceedingly rare, with only 8 cases documented in the medical literature⁹⁶. Similar to WRS, MEDS is an autosomal recessive condition. It is caused by mutations in IER3IP1⁹⁴. IER3IP1 encodes a small, 10 kDA, polypeptide protein. The overwhelming majority of cases are caused by homozygous p.Leu78Pro mutation^{94, 95, 97}. This might be due to a founder effect as most cases are reported in the Mediterranean area⁹⁶. IER3IP1 peptide localizes to the ER, via its C-terminal transmembrane domain. This suggests it is involved in COP-II vesicle budding (Figure 1A)⁹⁸. Similar to WRS, it is hypothesized that loss-of-function mutations in IER3IP1 reduce the ability of the cell to respond to ER stress. As a result, the brain and pancreas are particularly prone to ER stress-mediated cell death⁹⁶. This theory has been corroborated by the severe degree of apoptosis found in post-mortem brain samples⁹⁴.

Clinical next-generation sequencing for personalized genetic medicine

Clinical and genetic heterogeneity, as well as variable expressivity, each poses a challenge for making accurate diagnosis and designing effective therapies in patients with monogenic and syndromic forms of diabetes. Genetic testing is the best tool for physicians to diagnose genetic forms of diabetes, offer genetic education and counseling, and improve their disease management. The American Diabetes Association recommends genetic testing for all children diagnosed with diabetes in the first 6 months of life and patients diagnosed in early adulthood who have diabetes not characteristic of type 1 or type 2 diabetes that occurs in successive generations⁹⁹. In addition, insulin deficient, antibody negative diabetes in childhood and young adults should also be investigated, particularly if accompanied by neurologic dysfunction or other disorders. Genetic testing based on next-generation sequencing (NGS) technology, including exome and genome sequencing, has the ability to do this by identifying DNA variants that highly correlate with each patient’s clinical signs and symptoms. Routine use of NGS-based genetic testing will not only facilitate patient counseling by clinicians, medical geneticists, and genetic counselors, but will also serve as a first step towards designing personalized treatments for patients with monogenic and syndromic diabetes. Such genetic testing is available to patients with atypical diabetes at the Washington University Medical Center in St. Louis (https://gps.wustl.edu/patient-care/diabetes-and-er-stress/), as well as in other academic and commercial labs (https://www.ncbi.nlm.nih.gov/gtr/).

Conclusion

Despite their rarity, genetic syndromes of ER stress represent an important group of diseases that continue to enhance our understanding of the biological underpinnings of ER stress and diabetes. It is important for clinicians to be familiar with these syndromes so the appropriate testing can be ordered. A rapid diagnosis may help guide medical management, shed light on risk of future affected pregnancies, and potentially guide the patient towards relevant clinical trials. The advent and proliferation of DNA sequencing technology has made this more practical than ever before. Personalized genetic medicine will improve the quality of life and prognosis of patients with atypical diabetes.

Highlights.

• The endoplasmic reticulum (ER) is an emerging target for understanding the mechanisms of diabetes mellitus.

• Monogenic and syndromic forms of diabetes related to ER dysregulation is an important model for understanding the exact contribution of ER stress to β-cell death

• Genetic testing would be the best tool for physicians to diagnose genetic forms of diabetes, offer genetic education and counseling, and improve their disease management.

• Wolfram syndrome is a prototype of ER stress-associated monogenic and syndromic diabetes.