- Email: [email protected]
Maturity Onset Diabetes of the Young (MODY) in Tunisia: Low frequencies of GCK and HNF1A mutations
Gene 651 (2018) 44–48
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Research paper
Maturity Onset Diabetes of the Young (MODY) in Tunisia: Low frequencies of GCK and HNF1A mutations
T
⁎
S. Ben Khelifaa, , R. Martinezb, A. Dandanaa, I. Khochtalic, S. Ferchichia, L. Castañob a
Unit of Clinical and Molecular Biology/UR17ES29, Faculty of Pharmacy, Monastir, Tunisia Endocrinology and Diabetes Research Group, Hospital Universitario Cruces, BioCruces, CIBERER, CIBERDEM, UPV-EHU, Barakaldo, Basque Country, Spain c Endocrinology Unit, Fatouma Bourguiba hospital, Monastir, Tunisia b
A R T I C L E I N F O
A B S T R A C T
Keywords: Heridity Monogenic diabetes HNF1A GCK MODY Genetic screening
Maturity Onset Diabetes of the Young (MODY) is a monogenic form of diabetes characterized by autosomal dominant inheritance, an early clinical onset and a primary defect in β-cell function. Mutations in the GCK and HNF1A genes are the most common cause of MODY among Caucasians. The etiology of MODY in Tunisia stills a challenge for researchers. The aim of this study was to screen for mutations in GCK, HNF1A, HNF4A and INS genes in North African Tunisians subjects, in whom the clinical profile was very suggestive of MODY. A total of 23 unrelated patients, with clinical presentation of MODY were tested for mutations in GCK, HNF1A, HNF4A and INS genes, using Denaturing High Performance Liquid Chromatography (DHPLC), Multiplex Ligation-depend Probe Amplification (MLPA) and sequencing analysis. We identified the previously reported mutation c169C > T in one patient as well as a new mutation c-457C > T in two unrelated patients. No mutations were detected in the HNF1A and INS genes. Despite restrictive clinical criteria used for selecting patients in this study, the most common genes known for MODY do not explain the majority of cases in Tunisians. This suggests that there are others candidate or unidentified genes contributing to the etiology of MODY in Tunisians families.
1. Introduction Maturity Onset Diabetes of the Young (MODY) is a genetically heterogeneous form of monogenic diabetes, characterized by an early onset (usually before 25 years of age), autosomal mode of inheritance and a primary defect in pancreatic β-cell function (Owen and Hattersley, 2001; Fajans et al., 2001). According to the Online Mendelian Inheritance in Man database (MIM # 606391), pathologic mutations in thirteen genes have been described in various MODY subtypes (HNF4A, GCK, HNF1A, PDX1, TCF2, NEUROD1, KLF11, CEL, PAX4, INS, BLK, ABCC8 and KCNJ11) (http://omim.org/). However, most studies have shown that HNF1A-MODY and GCK-MODY are the most prevalent forms (Velho and Robert, 2002). Their relative frequencies vary considerably among ethnics group. While HNF1A-MODY mutations are the most common cause of suspected MODY patients in the United Kingdom (53%) (Shields et al., 2010) and Germany (62%) (Shober et al., 2009), GCK-MODY are the most common cause of suspected MODY subjects in Spain (80%) (Estalella et al., 2007), France
(56%) (Froguel et al., 1993), and Italy (41%) (Froguel et al., 1993). The other forms of MODY are less common among studied populations (Shields et al., 2010). Besides individual population characteristics, recruitment criteria account for part of the difference between these two most common types. Children from general population screened regardless of glyceamic status show higher rate of GCK-MODY, whereas MODY typing of adult diabetic patients yields a higher prevalence of HNF1A-MODY (Estalella et al., 2007; Giuffrida and Reis, 2005). Their remains a group of families with autosomal dominant segregation of non-autoimmunity diabetes mellitus for whom the genetic culprit is still to be identified and is known as MODY X. It accounts for approximately 15–20% of Caucasians families fitting MODY criteria (Kim, 2015; Mantovani et al., 2003). In Tunisia, a previous study showed the absence of mutations in known MODY genes in 11 patients (Amara et al., 2012). The aim of the present study was to investigate the contribution of the MODY genes HNF4A, GCK, HNF1A and INS in the etiology of 23 unrelated Tunisian families.
Abbreviations: Apo A, apolipoprotein A; Apo B, apolipoprotein B; BMI, body mass index; DHPLC, Denaturing High Performance Liquid Chromatography; GAD, Anti-Glutamic acid Decarboxylase; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; HLA-DRB1, HLA class II histocompatibility antigen, DRB1 beta chain; hs-CRP, high sensitive C reactive protein; IA2, Anti-tyrosine phosphatase; Lp(a), lipoprotein a; MLPA, Multiplex Ligation-depend Probe Amplification; MODY, Maturity Onset Diabetes of the Young; PCR, polymerase chain reaction; SPSS, statistical Package for Social Sciences; SSO, enzyme-Specific Oligonucleotid ⁎ Corresponding author. E-mail address: [email protected] (S. Ben Khelifa). https://doi.org/10.1016/j.gene.2018.01.081 Received 20 April 2017; Received in revised form 14 January 2018; Accepted 24 January 2018 Available online 03 February 2018 0378-1119/ © 2018 Published by Elsevier B.V.
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
2. Patients and methods
members were analyzed for the appropriate exon in order to perform the corresponding segregation analysis. The Human Genome Mutation Database (HGMD) (http://www. hgmd.cf.ac.uk), the genome browser Ensembl (www.ensembl.org) and the previously published MODY mutation detection articles (Ellard and Colclough, 2006; Osbak et al., 2009) were used to ascertain whether the variant was novel or not.
2.1. Patients We analyzed 23 extended families. They were recruited from the Unit of Endocrinology of Monastir Hospital as well as several clinical centers located in the Sahel region of Tunisia. The main criteria used for the selection of these unrelated families were autosomal dominant mode of inheritance of diabetes mellitus, presence of the disease in at least three consecutive generations, presence of at least one proband with early onset diabetes initially not requiring insulin treatment, and normal Body Mass Index (BMI). In addition, 50 unrelated, healthy subjects served for the control of the DNA sequences. For blood collect, we have obtained approval from the Hospital authority and the ethics committee (No: 13/2010) after consent of patients.
2.5. HLA-DRB1 typing HLA-DRB1 genotyping was performed by the Polymerase Chain Reaction enzyme-Specific Oligonucleotid method (PCR-SSO) combined with Luminex technology as described (Urrutia et al., 2017). 2.6. Statistics
2.2. Clinical data
Statistical analysis was performed using statistical Package for Social Sciences (SPSS for windows, version 19.0). Results were expressed as means ± standard deviation (M ± SD). For quantitative traits, student's t-test was used. P < 0.05 was considered statistically significant.
Details of patients such as age, duration of diabetes, BMI, complications and current treatment were recorded. Blood samples were drawn for measurement of fasting glucose, HbA1c, insulin, C-peptide, creatinine, uric acid, cystatin C, high sensitive C reactive protein (hsCRP), lipid profile (triglycerides, total cholesterol, HDL-cholesterol (HDL-C)), Apolipoprotein A and B (Apo A and Apo B), Lipoprotein a (Lp (a)), Anti-Glutamic acid Decarboxylase (GAD) and Anti-tyrosine phosphatase (IA2).
3. Results Anti- GADA and Anti IA2 were absent in all subjects. Three index cases (13.05%) were shown to carry mutations in GCK and HNF4A genes. No molecular defect was revealed by the MLPA method. A rate of 86.95% could not be genetically explained in our study. We identified a previously described mutation c.-169C > T in the HNF4A gene in one family. A novel mutation, c.-457C > T, was also identified in two unrelated subjects. Segregation of this mutation in the two families wasn't possible as members refused to adhere to our study.
2.3. Assays Fasting glucose, Urea, Creatinine, Uric acid, Triglycerides, Total Cholesterol and HDL-C were evaluated on Beckman auto-analyzer by using commercial kits from Randox (Randox Diagnostics, Antrim, UK). HbA1c, Apo A, Apo B, Lp (a), cystatin C and hs-CRP were determined by immunoturbidimetric method (Roche Diagnostics, Mannheim, Germany). Fasting insulin, C-peptide were determined by chemiluminiscent microparticule immunoassay (Abbott Diagnostics, Rungis, France). Anti- GADA and Anti IA2 were performed by radioimmunoassay technology for all MODY patients.
3.1. Patient with the HNF4A mutation c.-169C ≥ T By analysising the HNF4A gene, we identified the c.-169C > T mutation in a proband as well as her mother (Fig. 1a). This mutation is located in the promoter region P2 (exon 1d). The HNF4A-MODY patient was a 15 years-old nonobese (BMI = 19.47 Kg/m2) female, from consanguineous parents. She was at term with a birth weight of 3.9 Kg. The neurological and physical developments were normal. The patient was diagnosed on the time of the study. The blood glucose, the Hb1Ac and the C-peptide levels were respectively 8.89 mmol/L, 9.16% (12 mmol/L) and 0.53 ng/mL. Physical examination revealed no other underlying disease. The patient was treated with insulin. Her family history was strongly positive for diabetes. The patient's sister, her mother and her grandfather on her mother's side were diagnosed with diabetes (Fig. 1b). Her mother was diagnosed with diabetes after presenting polyuria-polydipsia and nocturia syndromes. She was treated with OHA than with insulin injection. Her father wasn't diabetic, but he presented hypertension at 41 yearold.
2.4. Mutation screening Genomic DNA was extracted from peripheral blood lymphocytes using phenol-chloroform method. All GCK (MIM #138079) exons, including intron/exon boundaries and the promoter, were screened by PCR-DHPLC. In order to generate heteroduplices, amplified fragment were denaturated at 95 °C for 5 min, then slowly cooling. Each fragment was analyzed using WAVE DNA Fragment Analysis system (Transgenomic, Omaha, NE). A lineal gradient of buffer A (containing (0.1 mol/L triethylammonium acetate–TEAA) and buffer B (0.1 mol/L TEAA, 25% acetonitrile) was used to elute DNA from the column. Primers, PCR and DHPLC conditions are available on request. The exons of those patients with aberrant wave profile were subjected to sequence analysis on an ABI 3130 capillary DNA Analyzer (Applied Biosystems). Exons, flanking introns and promoters of the HNF1A (MIM #142410), HNF4A (MIM #600281) and INS (MIM #176730) genes were amplified by PCR and then screened for mutations in probands by direct sequencing. Primers and PCR conditions are available upon request. Families without a mutation in the four analyzed genes detectable by the previously described methods were screened with by Multiplex ligation-dependent probe amplification (MLPA), following the manufacturer's instructions (MLPA; MRC-Holland, Amsterdam, Holland). Commercially oligonucleotides probes for GCK, HNF1A and HNF4A exons were used. When a mutation is identified in a proband, all the available family
3.2. Patient with GCK promoter variation c.–457C ≥ T The analysis of the GCK gene revealed a novel variation c.457C > T in two unrelated probands (Fig. 2a). This substitution of cytosine to tyrosine, located in the promoter region P1, was absent in 50 non-diabetic controls. The first patient was a nonobese (22.37 Kg/m2) female of 24 yearold. She was from unrelated parents. She was born at term with a birth weight of 3.150 Kg. Her physical and neurological development were normal. The patient was diagnosed at 19 year-old, during the first pregnancy when routine testing was performed. Blood glucose, HbA1c and C-peptide levels were respectively 7.62 mmol/L, 7.3% (9 mmol/L) and 0.87 ng/mL. Physical examination revealed no other underlying 45
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
Fig. 2. (a): GCK gene mutation analysis of the proband showing a heterozygous single nucleotide substitution in the promoter c.-457C > T. (b): The pedigree structure of the GCK-MODY family (patient 1) with the mutation c457C > T. The arrow indicates the proband, double lines indicate consanguineous matings, white squares and circles indicate healthy males and females respectively, black squares and circles indicate males and females with the c-457C > T respectively, grey squares and circles indicate respectively males and females with the GCK-MODY phenotype clinical but untested for the mutation.
Fig. 1. (a): HNF4A gene mutation analysis of the proband showing a heterozygous single nucleotide substitution in the promoter P2 (c-169C > T). (b): The pedigree structure of the HNF4A-MODY family with the mutation c.-169C > T. The arrow indicates the proband, double lines indicate consanguineous matings, white squares and circles indicate healthy males and females respectively, black squares and circles indicate males and females with c-169C > T respectively, grey squares and circles indicate respectively males and females with the HNF4A-MODY phenotype clinical but untested for the mutation.
disease. The patient was treated with OHA. Her family history was strongly positive for diabetes. The patient's brother, her mother and her grandparents on her mother's side were diagnosed with diabetes (Fig. 2b). The mother was diagnosed at 34 year-old during her third pregnancy. She was treated with OHA. The second patient was a nonobese (24.06 Kg/m2) male of 31 yearold. He was from unrelated parents. He was born at term with a birth weight of 3.470 Kg. His physical and neurological development were normal. The patient was diagnosed at 31 year-old, during this study. Blood glucose, HbA1c and C-peptide levels were respectively 7.92 mmol/L, 7.3% (9 mmol/L) and 1.13 ng/mL. Physical examination revealed no other underlying disease. The patient was treated by diet. Her family history was positive for diabetes. The patient's father, his aunt and his grandparents from his father's side were diagnosed with diabetes (Fig. 3). Clinical data of the three MODY patients are presented in Table 1.
Fig. 3. The pedigree structure of the GCK-MODY family (patient 2) with the c-457C > T mutation. The arrow indicates the proband, white squares and circles indicate healthy males and females respectively, black squares and circles indicate males and females with c457C > T respectively, grey squares and circles indicate respectively males and females with the GCK-MODY phenotype clinical but untested for the mutation.
3.3. Subjects with no identified mutations About 87% of Tunisian subjects, clinically characterized as MODY, were with negative results in genetic study for HNF4A, GCK, HNF1A and INS genes. This group of diabetic patients showed lean BMI (24.01 ± 2.25 Kg/m2). One of these subjects presented retinopathy and one suffered from hypertension. Compared with a group of Tunisian non-diabetic controls, this group of diabetics presented low Cpeptide (p = 0.03) and Apo A (p = 0.01) levels and higher triglycerides and Apo B levels (p = 0.03 for both). The HLA-DRB1 typing showed the presence of the variants DR3 or DR4 (2: DR3/DR4; 1: DR4/DR4; 1: DR3/X with X corresponds to any HLA-DRB1 allele different from DR3 and DR4) in four patients. Characteristics of subjects with no identified mutations are shown in
Table 2.
3.4. Polymorphisms The analysis of the HNF1A gene revealed the presence of the variant p.Ser487Asn (rs2464196) in 9 patients. Four of them presented also the p.Ile27Leu variant (rs1169288). Patients carrying the p.Ile27Leu and p.Ser487Asn were aged 21.19 ± 2.84 years at diagnosis. There BMI was 24.06 ± 2.15 Kg/m2. These patients presented the polydipsiapolyurea symptoms at diagnosis. HbA1c levels was on average 9.58 ± 1.79% (12 ± 0.3 mmol/L). These patients were treated by 46
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
it likely affects the cooperative action of transcription factors (HNF1α and β) at the HNF4A P2 promoter and to decrease it's basal activity in pancreatic islets (Wirsing et al., 2010). The examination of the Slovakian as well as our Tunisian family showed an early age of onset and a progressive impaired glucose stimulation insulin release. The c.169C > T is likely to be a relatively severe mutation, which might be revealed mostly in MODY patients diagnosed in infancy period. In this study, we report also the c.-457C > T variation located in the β-cell promoter of GCK gene. Previously, the activity of β-cell GCK promoter in INS-1 cells has been studied by preparing gene constructs containing different gene promoter fragment lengths ranging from −263 bp to −1031 bp with respect to ATG start site. The luciferase activity was detectable with all GCK promotor constructs. Increased expression was seen with fragment −430 bp and decreased expression was seen with fragment −618 bp. This suggests the presence of repressor elements within the −430 bp to −618 bp. Thus the c.457C > T could be a possible etiological point mutation for the GCKMODY, which might affect a transcriptional repressor. This mutation cosegrated with diabetes in the two probands, with a clinical phenotype similar to that caused by the only GCK promoter mutation -71G > C as well as that caused by a mutation in the coding region of the gene. This can be explained by the compensation provided by the wild type allele (Gasperikova et al., 2009). In fact, GCK-MODY is characterized by mild and stable hyperglycemia (5.5–8.0 mmol/L) that shows little deterioration with age and an HbA1c level below 8% (10 mmol/L). Patients are generally asymptomatic and do not develop diabetes-related microvascular complications, hyperglycemia is commonly discovered during routine screening such as during pregnancy (Gardner and Tai, 2012; Colom and Corcoy, 2010). Subjects with no identified mutations accounted for most of the clinical MODY cases in this study. The majority of these mutation-negative patients were lean and provided with insulin secretory capacity, suggesting that the insulin resistance syndrome didn't form a major part of these subjects' phenotype and the probability that these patients could have T2D was minor. We noted also the presence of the p.Ile27Leu and p.Ser487Asn polymorphisms respectively in 7 and 5 patients. These two variants have been reported to be associated with reduced glucose-stimulated insulin secretion/insulin resistance and compromised beta-cell function/reduced transcription activity (Awa et al., 2011). Thus, they could be part of the causes of diabetes onset in these patients. Additionally, four of the subjects with no identified mutations presented DR3 or DR4 variants. This genetic component could also be involved in the onset of diabetes in these patients. Nevertheless, the precise cause of MODY in the Tunisian population remains unknown. MODY is characterized by a variety of genetically and clinically heterogeneous forms of monogenic diabetes. However, all of these forms have common features that can distinguish them from other types of diabetes (autosomal dominant inheritance, early age at onset, betacell dysfunction) (Gardner and Tai, 2012; Stride and Hattersley, 2002). Although the prevalence of specific mutations in MODY genes differs widely among ethnics groups, mutations in the GCK and HNF1A genes, are the most frequent cause of MODY in the majority of populations studied. They account approximately for 50% of cases (Giuffrida and Reis, 2005). Despite the clinical profile that was very suggestive of MODY, about 87% of patients were not genetically explained in this study. The analysis of GCK, HNF1A, HNF4A and INS showed similar results in Asians and Brazilian populations. In fact, the prevalence of these subjects accounts for 80–90% of clinical MODY cases in Asians and approximately 75% of cases in Brazilians (Furuzawa et al., 2008; Hwang et al., 2006). These findings suggested that the genetic background in Caucasians is distinguishable from that of Asians, American Latinians and North-African Tunisians. The previous Tunisian study of Amara et al. showed the absence of mutations in known MODY genes in a cohort of 11 patients (Amara et al., 2012). It's unclear why mutational frequency is low in the
Table 1 Clinical data of MODY Tunisian patients.
Age (years) BMI (Kg/m2) Fasting glyceamia (nmol/L) HbA1c (%)/(mmol/ L) Insulin (pmol/L) C-peptide (ng/mL) Triglycerides (mmol/L) Total Cholesterol (mmol/L) HDL-Cholesterol (mmol/L) Apo A (mg/L) Apo B (mg/L) Lp (a) (mg/L) Creatinin (mol/L) Complications Current treatment
Patient with c.169C > T mutation
Patient 1 with c.457C > T mutation
Patient 2 with c.457C > T mutation
15 19.47 8.89
24 22.37 7.62
31 24.06 7.92
9.16/12
7.3/9
7.6/9.5
62 0.53 0.6
37 0.87 1.14
92 1.13 0.81
3.6
4.1
3.7
0.77
1.03
1.12
1.03 0.98 0.31 75 None Insulin
0.97 1.23 0.51 86 None OHA
1.2 0.91 0.39 96 None Diet
Table 2 Clinical characteristics of Tunisian subjects with no identified mutation.
Number (male/female) Age (years) Age at diagnosis (years) Duration of diabetes (years) BMI (Kg/m2) Fasting glyceamia (nmol/L) HbA1c (%) Insulin (pmol/L) C-peptide (ng/mL) Triglycerides (mmol/L) Total Cholesterol (mmol/L) HDL-Cholesterol (mmol/L) Apo A (g/L) Apo B (g/L) Lp (a) (g/L) Creatinine (mol/L) Cystatin C Microvascular complications (%) Microvascular complications (%) Treatment (Diet/Inslin/OHA)
Subjects with no identified mutation
Non-diabetic controls
p-Value
6/14 33.25 ± 10.38 21.55 ± 3.1 12.2 ± 9.57
10/15 24.15 ± 4.06 – –
– – – –
24.01 ± 2.25 9.96 ± 2.38
23.78 ± 1.45 4.71 ± 0.52
0.00 0.012
9.57 ± 1.87 40.35 ± 6.72 0.53 ± 0.17 1.15 ± 0.42 4.16 ± 0.8
4.86 48.4 1.35 0.86 4.44
0.56 7.1 0.31 0.18 0.63
0.000 0.000 0.401 0.03 0.535
1.11 ± 0.34
0.99 ± 0.28
0.646
1.02 ± 0.2 0.66 ± 0.29 0.23 ± 0.15 66.65 ± 16.78 0.68 ± 0.3 5
1.51 ± 0.4 1.01 ± 0.16 0.2 ± 0.1 73.24 ± 11.45 0.82 ± 0.29 –
0.01 0.03 0.057 0.799 0.65 –
5
–
–
15/35/50
–
–
± ± ± ± ±
OHA (46.15%) and insulin (53.85%). 4. Discussion Mutational analysis of GCK, HNF4A, HNF1A and INS genes was performed in 23 Tunisians families, presenting the clinical phenotype of MODY. After molecular analysis, we identified the previously described mutation c.-169C > T in one family (Wirsing et al., 2010). We identified also one novel variation c.-457C > T in two unrelated probands. No mutation was found in the HNF1A as well as in the INS genes. The heterozygous HNF4A c.-169C > T mutation found in one patient has been reported earlier to cosegregate with diabetes in 6 Slovakian families. The functional study of this mutation revealed that 47
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
Tunisian population, but this could be in part explained by the implication of other single genes as recruiting criteria in this study were restrictive. So, further investigations of other candidate or susceptible genes not yet identified that are involved in transcription, development of islet cells and glucose sensing in islet cells, can provide clue for the pathogenesis of MODY in Tunisia. Subsequently, we need to carry out the next generation sequencing and to widen the study sample. Besides, future researches must focus on specific biomarkers that can discriminate MODY from type 1 and type 2 diabetes in the Tunisian population, since standard clinical criteria alone (described for the Caucasian population) are not enough discriminant.
diabetes of the young. Hum. Mutat. 27, 854–869. Estalella, I., Rica, I., De Nanclares, G.P., Bilbao, J.R., Vasquez, J.A., San Pedro, J.I., et al., 2007. Mutations in GCK and HNF-1α explain the majority of cases with clinical diagnosis of MODY in Spain. Clin. Endocrinol. 67, 538–546. Fajans, S.S., Bell, G.I., Polonsky, K.S., 2001. Molecular mechanisms and clinical pathophysiology of maturity-onset diabetes of the young. N. Engl. J. Med. 345, 971–980. Froguel, P., Zouali, H., Vionnet, N., Velho, G., Vaxillaire, M., Sun, F., Lesage, S., Stoffel, M., Takeda, J., Passa, P., 1993. Familial hyperglycemia due to mutations in glucokinase. Definition of a subtype of diabetes mellitus. N. Engl. J. Med. 11 (328), 697–702. Furuzawa, G.K., Giuffrida, F.M.A., Oliveira, C.S.V., Chacr, A.R., Dib, S.A., Reis, A.F., 2008. Low prevalence of MODY 2 and MODY 3 mutations in Brazilian individuals with clinical MODY phenotype. Diabetes Res. Clin. Pract. 81, e12-e14. Gardner, D.S., Tai, E.S., 2012. Clinical features and treatment of maturity onset diabetes of the young (MODY). Diabetes Metab. Syndr. Obes. 5, 101–108. Gasperikova, D., Tribble, N.D., Stanik, J., Huckova, M., Misovicova, N., et al., 2009. Identification of a novel beta-cell glucokinase (GCK) promoter mutation (−71G > C) that modulates GCK gene expression through loss of allele-specific Sp1 binding causing mild fasting hyperglycemia in humans. Diabetes 58, 1929–1935. Genome browser Ensembl available from. http://www.ensembl.org. Giuffrida, F.M.A., Reis, A.F., 2005. Genetic and clinical characteristics of maturity-onset diabetes of the young. Diabetes Obes. Metab. 7, 318–326. Human Genome Mutation Database (HGMD) available from. http://www.hgmd.cf.ac.uk. Hwang, J.S., Shin, C.H., Yang, S.W., Jung, S.Y., Huh, N., 2006. Genetic and clinical characteristics of Korean maturity-onset diabetes of the young (MODY) patients. Diabetes Res. Clin. Pract. 74, 75–81. Kim, S.H., 2015. Maturity-onset diabetes of the young: what do clinicians need to know? Diabetes Metab. J. 39, 468–477. Mantovani, V., Salardi, S., Cerreta, V., Bastia, D., Cenci, M., Ragni, L., et al., 2003. Identification of eight novel glucokinase mutations in Italian children with maturityonset diabetes of the young. Hum. Mutat. 22, 338–344. Online Mendelian Inheritance in Man O. Johns Hopkins University, Baltimore, MD. MIM Number: 606391: 09/21/2010. World Wide Web (available from) http://omim.org/. Osbak, K.K., Colclough, K., Saint-Martin, C., Beer, N.L., Bellanné-Chantelot, C., Ellard, S., et al., 2009. Update on mutations in glucokinase (GCK), which cause maturity-onset diabetes of the young, permanent neonatal diabetes, and hyperinsulinemic hypoglycemia. Hum. Mutat. 30, 1512–1526. Owen, K., Hattersley, A.T., 2001. Maturity-onset diabetes of the young: from clinical description to molecular genetic characterization. Best Pract. Res. Clin. Endocrinol. Metab. 15, 309–323. Shields, B.M., Hicks, S., Shepherd, M.H., Colclough, K., Hattersley, A.T., Ellard, S., 2010. Maturity-onset diabetes of the young (MODY): how many cases are we missing. Diabetologia 53, 2504–2508. Shober, E., Grabert, R.M., Kapellen, A.T.T., Reinehr, T., Holl, R.W., 2009. Phenotypical of maturity-onset diabetes of the young (MODY diabetes) in comparision with type 2 diabetes mellutis (T2DM) in children and adolescents: experience from a large multicentre database. Diabet. Med. 26, 466–473. Stride, A., Hattersley, A.T., 2002. Differents genes, differents diabetes: lessons from maturity-onset diabetes of the young. Ann. Med. 34, 207–216. Urrutia, I., Martínez, R., López-Euba, T., Velayos, T., Martínez de LaPiscina, I., Bilbao, J.R., et al., 2017. Lower frequency of HLA-DRB1 type 1 diabetes risk alleles in pediatric patients with MODY. PLoS One 4 (12), e0169389. Velho, G., Robert, J.J., 2002. 3. Maturity-onset diabetes of the young (MODY): genetic and clinical characteristics. Horm. Res. 57 (Suppl. 1), 29–33. Wirsing, A., Johnstone, K.A., Harries, L.W., Ellard, S., Ryffel, G.U., Stanik, J., et al., 2010. Novel monogenic diabetes mutations in the P2 promoter of the HNF4A gene are associated with impaired function in vitro. Diabet. Med. 27, 631–635.
5. Conclusion Unlike Caucasian population, GCK-MODY and HNF1A-MODY did not explain the majority of MODY cases in Tunisia, and the major gene of this pathology remains to be identified in our population. So further, studies must consider other single genes involved in β-cell function and other screening methods such us the Next generation sequencing. Competing interests The authors declare that they have no competing interests. Acknowledgements We thank patients and families, and collaborating physicians for participating in this study. We thank so much the staff of the Biochemistry Laboratory of Faculty of Pharmacy of Monastir-Tunisia, the Unit of Endocrinology of Fatouma Bourguiba Hospital of MonastirTunisia and the members of the Endocrinology and Diabetes Research Group of the Cruces hospital of Barakaldo- Spain, for their valuable contributions in this work. References Amara, A., Chadli-Chaieb, M., Ghezaiel, H., Philippe, J., Brahem, R., Dechaume, A., et al., 2012. Familial early-onset diabetes is not a typical MODY in several Tunisian patients. Tunis. Med. 90, 882–887. Awa, W.L., Thon, A., Raile, K., Grulich-Henn, J., Meissner, T., Schober, E., et al., 2011. Genetic and clinical characteristics of patients with HNF1A gene variations from the German–Austrian DPV database. Eur. J. Endocrinol. 164, 513–520. Colom, C., Corcoy, R., 2010. Maturity onset diabetes of the young and pregnancy. Best Pract. Res. Clin. Endocrinol. Metab. 24, 605–615. Ellard, S., Colclough, K., 2006. Mutations in the genes encoding the transcription factors hepatocyte nuclear factor 1 alpha (HNF1A) and 4 alpha (HNF4A) in maturity-onset
48
Contents lists available at ScienceDirect
Gene journal homepage: www.elsevier.com/locate/gene
Research paper
Maturity Onset Diabetes of the Young (MODY) in Tunisia: Low frequencies of GCK and HNF1A mutations
T
⁎
S. Ben Khelifaa, , R. Martinezb, A. Dandanaa, I. Khochtalic, S. Ferchichia, L. Castañob a
Unit of Clinical and Molecular Biology/UR17ES29, Faculty of Pharmacy, Monastir, Tunisia Endocrinology and Diabetes Research Group, Hospital Universitario Cruces, BioCruces, CIBERER, CIBERDEM, UPV-EHU, Barakaldo, Basque Country, Spain c Endocrinology Unit, Fatouma Bourguiba hospital, Monastir, Tunisia b
A R T I C L E I N F O
A B S T R A C T
Keywords: Heridity Monogenic diabetes HNF1A GCK MODY Genetic screening
Maturity Onset Diabetes of the Young (MODY) is a monogenic form of diabetes characterized by autosomal dominant inheritance, an early clinical onset and a primary defect in β-cell function. Mutations in the GCK and HNF1A genes are the most common cause of MODY among Caucasians. The etiology of MODY in Tunisia stills a challenge for researchers. The aim of this study was to screen for mutations in GCK, HNF1A, HNF4A and INS genes in North African Tunisians subjects, in whom the clinical profile was very suggestive of MODY. A total of 23 unrelated patients, with clinical presentation of MODY were tested for mutations in GCK, HNF1A, HNF4A and INS genes, using Denaturing High Performance Liquid Chromatography (DHPLC), Multiplex Ligation-depend Probe Amplification (MLPA) and sequencing analysis. We identified the previously reported mutation c169C > T in one patient as well as a new mutation c-457C > T in two unrelated patients. No mutations were detected in the HNF1A and INS genes. Despite restrictive clinical criteria used for selecting patients in this study, the most common genes known for MODY do not explain the majority of cases in Tunisians. This suggests that there are others candidate or unidentified genes contributing to the etiology of MODY in Tunisians families.
1. Introduction Maturity Onset Diabetes of the Young (MODY) is a genetically heterogeneous form of monogenic diabetes, characterized by an early onset (usually before 25 years of age), autosomal mode of inheritance and a primary defect in pancreatic β-cell function (Owen and Hattersley, 2001; Fajans et al., 2001). According to the Online Mendelian Inheritance in Man database (MIM # 606391), pathologic mutations in thirteen genes have been described in various MODY subtypes (HNF4A, GCK, HNF1A, PDX1, TCF2, NEUROD1, KLF11, CEL, PAX4, INS, BLK, ABCC8 and KCNJ11) (http://omim.org/). However, most studies have shown that HNF1A-MODY and GCK-MODY are the most prevalent forms (Velho and Robert, 2002). Their relative frequencies vary considerably among ethnics group. While HNF1A-MODY mutations are the most common cause of suspected MODY patients in the United Kingdom (53%) (Shields et al., 2010) and Germany (62%) (Shober et al., 2009), GCK-MODY are the most common cause of suspected MODY subjects in Spain (80%) (Estalella et al., 2007), France
(56%) (Froguel et al., 1993), and Italy (41%) (Froguel et al., 1993). The other forms of MODY are less common among studied populations (Shields et al., 2010). Besides individual population characteristics, recruitment criteria account for part of the difference between these two most common types. Children from general population screened regardless of glyceamic status show higher rate of GCK-MODY, whereas MODY typing of adult diabetic patients yields a higher prevalence of HNF1A-MODY (Estalella et al., 2007; Giuffrida and Reis, 2005). Their remains a group of families with autosomal dominant segregation of non-autoimmunity diabetes mellitus for whom the genetic culprit is still to be identified and is known as MODY X. It accounts for approximately 15–20% of Caucasians families fitting MODY criteria (Kim, 2015; Mantovani et al., 2003). In Tunisia, a previous study showed the absence of mutations in known MODY genes in 11 patients (Amara et al., 2012). The aim of the present study was to investigate the contribution of the MODY genes HNF4A, GCK, HNF1A and INS in the etiology of 23 unrelated Tunisian families.
Abbreviations: Apo A, apolipoprotein A; Apo B, apolipoprotein B; BMI, body mass index; DHPLC, Denaturing High Performance Liquid Chromatography; GAD, Anti-Glutamic acid Decarboxylase; HbA1c, glycated hemoglobin; HDL-C, high-density lipoprotein cholesterol; HLA-DRB1, HLA class II histocompatibility antigen, DRB1 beta chain; hs-CRP, high sensitive C reactive protein; IA2, Anti-tyrosine phosphatase; Lp(a), lipoprotein a; MLPA, Multiplex Ligation-depend Probe Amplification; MODY, Maturity Onset Diabetes of the Young; PCR, polymerase chain reaction; SPSS, statistical Package for Social Sciences; SSO, enzyme-Specific Oligonucleotid ⁎ Corresponding author. E-mail address: [email protected] (S. Ben Khelifa). https://doi.org/10.1016/j.gene.2018.01.081 Received 20 April 2017; Received in revised form 14 January 2018; Accepted 24 January 2018 Available online 03 February 2018 0378-1119/ © 2018 Published by Elsevier B.V.
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
2. Patients and methods
members were analyzed for the appropriate exon in order to perform the corresponding segregation analysis. The Human Genome Mutation Database (HGMD) (http://www. hgmd.cf.ac.uk), the genome browser Ensembl (www.ensembl.org) and the previously published MODY mutation detection articles (Ellard and Colclough, 2006; Osbak et al., 2009) were used to ascertain whether the variant was novel or not.
2.1. Patients We analyzed 23 extended families. They were recruited from the Unit of Endocrinology of Monastir Hospital as well as several clinical centers located in the Sahel region of Tunisia. The main criteria used for the selection of these unrelated families were autosomal dominant mode of inheritance of diabetes mellitus, presence of the disease in at least three consecutive generations, presence of at least one proband with early onset diabetes initially not requiring insulin treatment, and normal Body Mass Index (BMI). In addition, 50 unrelated, healthy subjects served for the control of the DNA sequences. For blood collect, we have obtained approval from the Hospital authority and the ethics committee (No: 13/2010) after consent of patients.
2.5. HLA-DRB1 typing HLA-DRB1 genotyping was performed by the Polymerase Chain Reaction enzyme-Specific Oligonucleotid method (PCR-SSO) combined with Luminex technology as described (Urrutia et al., 2017). 2.6. Statistics
2.2. Clinical data
Statistical analysis was performed using statistical Package for Social Sciences (SPSS for windows, version 19.0). Results were expressed as means ± standard deviation (M ± SD). For quantitative traits, student's t-test was used. P < 0.05 was considered statistically significant.
Details of patients such as age, duration of diabetes, BMI, complications and current treatment were recorded. Blood samples were drawn for measurement of fasting glucose, HbA1c, insulin, C-peptide, creatinine, uric acid, cystatin C, high sensitive C reactive protein (hsCRP), lipid profile (triglycerides, total cholesterol, HDL-cholesterol (HDL-C)), Apolipoprotein A and B (Apo A and Apo B), Lipoprotein a (Lp (a)), Anti-Glutamic acid Decarboxylase (GAD) and Anti-tyrosine phosphatase (IA2).
3. Results Anti- GADA and Anti IA2 were absent in all subjects. Three index cases (13.05%) were shown to carry mutations in GCK and HNF4A genes. No molecular defect was revealed by the MLPA method. A rate of 86.95% could not be genetically explained in our study. We identified a previously described mutation c.-169C > T in the HNF4A gene in one family. A novel mutation, c.-457C > T, was also identified in two unrelated subjects. Segregation of this mutation in the two families wasn't possible as members refused to adhere to our study.
2.3. Assays Fasting glucose, Urea, Creatinine, Uric acid, Triglycerides, Total Cholesterol and HDL-C were evaluated on Beckman auto-analyzer by using commercial kits from Randox (Randox Diagnostics, Antrim, UK). HbA1c, Apo A, Apo B, Lp (a), cystatin C and hs-CRP were determined by immunoturbidimetric method (Roche Diagnostics, Mannheim, Germany). Fasting insulin, C-peptide were determined by chemiluminiscent microparticule immunoassay (Abbott Diagnostics, Rungis, France). Anti- GADA and Anti IA2 were performed by radioimmunoassay technology for all MODY patients.
3.1. Patient with the HNF4A mutation c.-169C ≥ T By analysising the HNF4A gene, we identified the c.-169C > T mutation in a proband as well as her mother (Fig. 1a). This mutation is located in the promoter region P2 (exon 1d). The HNF4A-MODY patient was a 15 years-old nonobese (BMI = 19.47 Kg/m2) female, from consanguineous parents. She was at term with a birth weight of 3.9 Kg. The neurological and physical developments were normal. The patient was diagnosed on the time of the study. The blood glucose, the Hb1Ac and the C-peptide levels were respectively 8.89 mmol/L, 9.16% (12 mmol/L) and 0.53 ng/mL. Physical examination revealed no other underlying disease. The patient was treated with insulin. Her family history was strongly positive for diabetes. The patient's sister, her mother and her grandfather on her mother's side were diagnosed with diabetes (Fig. 1b). Her mother was diagnosed with diabetes after presenting polyuria-polydipsia and nocturia syndromes. She was treated with OHA than with insulin injection. Her father wasn't diabetic, but he presented hypertension at 41 yearold.
2.4. Mutation screening Genomic DNA was extracted from peripheral blood lymphocytes using phenol-chloroform method. All GCK (MIM #138079) exons, including intron/exon boundaries and the promoter, were screened by PCR-DHPLC. In order to generate heteroduplices, amplified fragment were denaturated at 95 °C for 5 min, then slowly cooling. Each fragment was analyzed using WAVE DNA Fragment Analysis system (Transgenomic, Omaha, NE). A lineal gradient of buffer A (containing (0.1 mol/L triethylammonium acetate–TEAA) and buffer B (0.1 mol/L TEAA, 25% acetonitrile) was used to elute DNA from the column. Primers, PCR and DHPLC conditions are available on request. The exons of those patients with aberrant wave profile were subjected to sequence analysis on an ABI 3130 capillary DNA Analyzer (Applied Biosystems). Exons, flanking introns and promoters of the HNF1A (MIM #142410), HNF4A (MIM #600281) and INS (MIM #176730) genes were amplified by PCR and then screened for mutations in probands by direct sequencing. Primers and PCR conditions are available upon request. Families without a mutation in the four analyzed genes detectable by the previously described methods were screened with by Multiplex ligation-dependent probe amplification (MLPA), following the manufacturer's instructions (MLPA; MRC-Holland, Amsterdam, Holland). Commercially oligonucleotides probes for GCK, HNF1A and HNF4A exons were used. When a mutation is identified in a proband, all the available family
3.2. Patient with GCK promoter variation c.–457C ≥ T The analysis of the GCK gene revealed a novel variation c.457C > T in two unrelated probands (Fig. 2a). This substitution of cytosine to tyrosine, located in the promoter region P1, was absent in 50 non-diabetic controls. The first patient was a nonobese (22.37 Kg/m2) female of 24 yearold. She was from unrelated parents. She was born at term with a birth weight of 3.150 Kg. Her physical and neurological development were normal. The patient was diagnosed at 19 year-old, during the first pregnancy when routine testing was performed. Blood glucose, HbA1c and C-peptide levels were respectively 7.62 mmol/L, 7.3% (9 mmol/L) and 0.87 ng/mL. Physical examination revealed no other underlying 45
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
Fig. 2. (a): GCK gene mutation analysis of the proband showing a heterozygous single nucleotide substitution in the promoter c.-457C > T. (b): The pedigree structure of the GCK-MODY family (patient 1) with the mutation c457C > T. The arrow indicates the proband, double lines indicate consanguineous matings, white squares and circles indicate healthy males and females respectively, black squares and circles indicate males and females with the c-457C > T respectively, grey squares and circles indicate respectively males and females with the GCK-MODY phenotype clinical but untested for the mutation.
Fig. 1. (a): HNF4A gene mutation analysis of the proband showing a heterozygous single nucleotide substitution in the promoter P2 (c-169C > T). (b): The pedigree structure of the HNF4A-MODY family with the mutation c.-169C > T. The arrow indicates the proband, double lines indicate consanguineous matings, white squares and circles indicate healthy males and females respectively, black squares and circles indicate males and females with c-169C > T respectively, grey squares and circles indicate respectively males and females with the HNF4A-MODY phenotype clinical but untested for the mutation.
disease. The patient was treated with OHA. Her family history was strongly positive for diabetes. The patient's brother, her mother and her grandparents on her mother's side were diagnosed with diabetes (Fig. 2b). The mother was diagnosed at 34 year-old during her third pregnancy. She was treated with OHA. The second patient was a nonobese (24.06 Kg/m2) male of 31 yearold. He was from unrelated parents. He was born at term with a birth weight of 3.470 Kg. His physical and neurological development were normal. The patient was diagnosed at 31 year-old, during this study. Blood glucose, HbA1c and C-peptide levels were respectively 7.92 mmol/L, 7.3% (9 mmol/L) and 1.13 ng/mL. Physical examination revealed no other underlying disease. The patient was treated by diet. Her family history was positive for diabetes. The patient's father, his aunt and his grandparents from his father's side were diagnosed with diabetes (Fig. 3). Clinical data of the three MODY patients are presented in Table 1.
Fig. 3. The pedigree structure of the GCK-MODY family (patient 2) with the c-457C > T mutation. The arrow indicates the proband, white squares and circles indicate healthy males and females respectively, black squares and circles indicate males and females with c457C > T respectively, grey squares and circles indicate respectively males and females with the GCK-MODY phenotype clinical but untested for the mutation.
3.3. Subjects with no identified mutations About 87% of Tunisian subjects, clinically characterized as MODY, were with negative results in genetic study for HNF4A, GCK, HNF1A and INS genes. This group of diabetic patients showed lean BMI (24.01 ± 2.25 Kg/m2). One of these subjects presented retinopathy and one suffered from hypertension. Compared with a group of Tunisian non-diabetic controls, this group of diabetics presented low Cpeptide (p = 0.03) and Apo A (p = 0.01) levels and higher triglycerides and Apo B levels (p = 0.03 for both). The HLA-DRB1 typing showed the presence of the variants DR3 or DR4 (2: DR3/DR4; 1: DR4/DR4; 1: DR3/X with X corresponds to any HLA-DRB1 allele different from DR3 and DR4) in four patients. Characteristics of subjects with no identified mutations are shown in
Table 2.
3.4. Polymorphisms The analysis of the HNF1A gene revealed the presence of the variant p.Ser487Asn (rs2464196) in 9 patients. Four of them presented also the p.Ile27Leu variant (rs1169288). Patients carrying the p.Ile27Leu and p.Ser487Asn were aged 21.19 ± 2.84 years at diagnosis. There BMI was 24.06 ± 2.15 Kg/m2. These patients presented the polydipsiapolyurea symptoms at diagnosis. HbA1c levels was on average 9.58 ± 1.79% (12 ± 0.3 mmol/L). These patients were treated by 46
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
it likely affects the cooperative action of transcription factors (HNF1α and β) at the HNF4A P2 promoter and to decrease it's basal activity in pancreatic islets (Wirsing et al., 2010). The examination of the Slovakian as well as our Tunisian family showed an early age of onset and a progressive impaired glucose stimulation insulin release. The c.169C > T is likely to be a relatively severe mutation, which might be revealed mostly in MODY patients diagnosed in infancy period. In this study, we report also the c.-457C > T variation located in the β-cell promoter of GCK gene. Previously, the activity of β-cell GCK promoter in INS-1 cells has been studied by preparing gene constructs containing different gene promoter fragment lengths ranging from −263 bp to −1031 bp with respect to ATG start site. The luciferase activity was detectable with all GCK promotor constructs. Increased expression was seen with fragment −430 bp and decreased expression was seen with fragment −618 bp. This suggests the presence of repressor elements within the −430 bp to −618 bp. Thus the c.457C > T could be a possible etiological point mutation for the GCKMODY, which might affect a transcriptional repressor. This mutation cosegrated with diabetes in the two probands, with a clinical phenotype similar to that caused by the only GCK promoter mutation -71G > C as well as that caused by a mutation in the coding region of the gene. This can be explained by the compensation provided by the wild type allele (Gasperikova et al., 2009). In fact, GCK-MODY is characterized by mild and stable hyperglycemia (5.5–8.0 mmol/L) that shows little deterioration with age and an HbA1c level below 8% (10 mmol/L). Patients are generally asymptomatic and do not develop diabetes-related microvascular complications, hyperglycemia is commonly discovered during routine screening such as during pregnancy (Gardner and Tai, 2012; Colom and Corcoy, 2010). Subjects with no identified mutations accounted for most of the clinical MODY cases in this study. The majority of these mutation-negative patients were lean and provided with insulin secretory capacity, suggesting that the insulin resistance syndrome didn't form a major part of these subjects' phenotype and the probability that these patients could have T2D was minor. We noted also the presence of the p.Ile27Leu and p.Ser487Asn polymorphisms respectively in 7 and 5 patients. These two variants have been reported to be associated with reduced glucose-stimulated insulin secretion/insulin resistance and compromised beta-cell function/reduced transcription activity (Awa et al., 2011). Thus, they could be part of the causes of diabetes onset in these patients. Additionally, four of the subjects with no identified mutations presented DR3 or DR4 variants. This genetic component could also be involved in the onset of diabetes in these patients. Nevertheless, the precise cause of MODY in the Tunisian population remains unknown. MODY is characterized by a variety of genetically and clinically heterogeneous forms of monogenic diabetes. However, all of these forms have common features that can distinguish them from other types of diabetes (autosomal dominant inheritance, early age at onset, betacell dysfunction) (Gardner and Tai, 2012; Stride and Hattersley, 2002). Although the prevalence of specific mutations in MODY genes differs widely among ethnics groups, mutations in the GCK and HNF1A genes, are the most frequent cause of MODY in the majority of populations studied. They account approximately for 50% of cases (Giuffrida and Reis, 2005). Despite the clinical profile that was very suggestive of MODY, about 87% of patients were not genetically explained in this study. The analysis of GCK, HNF1A, HNF4A and INS showed similar results in Asians and Brazilian populations. In fact, the prevalence of these subjects accounts for 80–90% of clinical MODY cases in Asians and approximately 75% of cases in Brazilians (Furuzawa et al., 2008; Hwang et al., 2006). These findings suggested that the genetic background in Caucasians is distinguishable from that of Asians, American Latinians and North-African Tunisians. The previous Tunisian study of Amara et al. showed the absence of mutations in known MODY genes in a cohort of 11 patients (Amara et al., 2012). It's unclear why mutational frequency is low in the
Table 1 Clinical data of MODY Tunisian patients.
Age (years) BMI (Kg/m2) Fasting glyceamia (nmol/L) HbA1c (%)/(mmol/ L) Insulin (pmol/L) C-peptide (ng/mL) Triglycerides (mmol/L) Total Cholesterol (mmol/L) HDL-Cholesterol (mmol/L) Apo A (mg/L) Apo B (mg/L) Lp (a) (mg/L) Creatinin (mol/L) Complications Current treatment
Patient with c.169C > T mutation
Patient 1 with c.457C > T mutation
Patient 2 with c.457C > T mutation
15 19.47 8.89
24 22.37 7.62
31 24.06 7.92
9.16/12
7.3/9
7.6/9.5
62 0.53 0.6
37 0.87 1.14
92 1.13 0.81
3.6
4.1
3.7
0.77
1.03
1.12
1.03 0.98 0.31 75 None Insulin
0.97 1.23 0.51 86 None OHA
1.2 0.91 0.39 96 None Diet
Table 2 Clinical characteristics of Tunisian subjects with no identified mutation.
Number (male/female) Age (years) Age at diagnosis (years) Duration of diabetes (years) BMI (Kg/m2) Fasting glyceamia (nmol/L) HbA1c (%) Insulin (pmol/L) C-peptide (ng/mL) Triglycerides (mmol/L) Total Cholesterol (mmol/L) HDL-Cholesterol (mmol/L) Apo A (g/L) Apo B (g/L) Lp (a) (g/L) Creatinine (mol/L) Cystatin C Microvascular complications (%) Microvascular complications (%) Treatment (Diet/Inslin/OHA)
Subjects with no identified mutation
Non-diabetic controls
p-Value
6/14 33.25 ± 10.38 21.55 ± 3.1 12.2 ± 9.57
10/15 24.15 ± 4.06 – –
– – – –
24.01 ± 2.25 9.96 ± 2.38
23.78 ± 1.45 4.71 ± 0.52
0.00 0.012
9.57 ± 1.87 40.35 ± 6.72 0.53 ± 0.17 1.15 ± 0.42 4.16 ± 0.8
4.86 48.4 1.35 0.86 4.44
0.56 7.1 0.31 0.18 0.63
0.000 0.000 0.401 0.03 0.535
1.11 ± 0.34
0.99 ± 0.28
0.646
1.02 ± 0.2 0.66 ± 0.29 0.23 ± 0.15 66.65 ± 16.78 0.68 ± 0.3 5
1.51 ± 0.4 1.01 ± 0.16 0.2 ± 0.1 73.24 ± 11.45 0.82 ± 0.29 –
0.01 0.03 0.057 0.799 0.65 –
5
–
–
15/35/50
–
–
± ± ± ± ±
OHA (46.15%) and insulin (53.85%). 4. Discussion Mutational analysis of GCK, HNF4A, HNF1A and INS genes was performed in 23 Tunisians families, presenting the clinical phenotype of MODY. After molecular analysis, we identified the previously described mutation c.-169C > T in one family (Wirsing et al., 2010). We identified also one novel variation c.-457C > T in two unrelated probands. No mutation was found in the HNF1A as well as in the INS genes. The heterozygous HNF4A c.-169C > T mutation found in one patient has been reported earlier to cosegregate with diabetes in 6 Slovakian families. The functional study of this mutation revealed that 47
Gene 651 (2018) 44–48
S. Ben Khelifa et al.
Tunisian population, but this could be in part explained by the implication of other single genes as recruiting criteria in this study were restrictive. So, further investigations of other candidate or susceptible genes not yet identified that are involved in transcription, development of islet cells and glucose sensing in islet cells, can provide clue for the pathogenesis of MODY in Tunisia. Subsequently, we need to carry out the next generation sequencing and to widen the study sample. Besides, future researches must focus on specific biomarkers that can discriminate MODY from type 1 and type 2 diabetes in the Tunisian population, since standard clinical criteria alone (described for the Caucasian population) are not enough discriminant.
diabetes of the young. Hum. Mutat. 27, 854–869. Estalella, I., Rica, I., De Nanclares, G.P., Bilbao, J.R., Vasquez, J.A., San Pedro, J.I., et al., 2007. Mutations in GCK and HNF-1α explain the majority of cases with clinical diagnosis of MODY in Spain. Clin. Endocrinol. 67, 538–546. Fajans, S.S., Bell, G.I., Polonsky, K.S., 2001. Molecular mechanisms and clinical pathophysiology of maturity-onset diabetes of the young. N. Engl. J. Med. 345, 971–980. Froguel, P., Zouali, H., Vionnet, N., Velho, G., Vaxillaire, M., Sun, F., Lesage, S., Stoffel, M., Takeda, J., Passa, P., 1993. Familial hyperglycemia due to mutations in glucokinase. Definition of a subtype of diabetes mellitus. N. Engl. J. Med. 11 (328), 697–702. Furuzawa, G.K., Giuffrida, F.M.A., Oliveira, C.S.V., Chacr, A.R., Dib, S.A., Reis, A.F., 2008. Low prevalence of MODY 2 and MODY 3 mutations in Brazilian individuals with clinical MODY phenotype. Diabetes Res. Clin. Pract. 81, e12-e14. Gardner, D.S., Tai, E.S., 2012. Clinical features and treatment of maturity onset diabetes of the young (MODY). Diabetes Metab. Syndr. Obes. 5, 101–108. Gasperikova, D., Tribble, N.D., Stanik, J., Huckova, M., Misovicova, N., et al., 2009. Identification of a novel beta-cell glucokinase (GCK) promoter mutation (−71G > C) that modulates GCK gene expression through loss of allele-specific Sp1 binding causing mild fasting hyperglycemia in humans. Diabetes 58, 1929–1935. Genome browser Ensembl available from. http://www.ensembl.org. Giuffrida, F.M.A., Reis, A.F., 2005. Genetic and clinical characteristics of maturity-onset diabetes of the young. Diabetes Obes. Metab. 7, 318–326. Human Genome Mutation Database (HGMD) available from. http://www.hgmd.cf.ac.uk. Hwang, J.S., Shin, C.H., Yang, S.W., Jung, S.Y., Huh, N., 2006. Genetic and clinical characteristics of Korean maturity-onset diabetes of the young (MODY) patients. Diabetes Res. Clin. Pract. 74, 75–81. Kim, S.H., 2015. Maturity-onset diabetes of the young: what do clinicians need to know? Diabetes Metab. J. 39, 468–477. Mantovani, V., Salardi, S., Cerreta, V., Bastia, D., Cenci, M., Ragni, L., et al., 2003. Identification of eight novel glucokinase mutations in Italian children with maturityonset diabetes of the young. Hum. Mutat. 22, 338–344. Online Mendelian Inheritance in Man O. Johns Hopkins University, Baltimore, MD. MIM Number: 606391: 09/21/2010. World Wide Web (available from) http://omim.org/. Osbak, K.K., Colclough, K., Saint-Martin, C., Beer, N.L., Bellanné-Chantelot, C., Ellard, S., et al., 2009. Update on mutations in glucokinase (GCK), which cause maturity-onset diabetes of the young, permanent neonatal diabetes, and hyperinsulinemic hypoglycemia. Hum. Mutat. 30, 1512–1526. Owen, K., Hattersley, A.T., 2001. Maturity-onset diabetes of the young: from clinical description to molecular genetic characterization. Best Pract. Res. Clin. Endocrinol. Metab. 15, 309–323. Shields, B.M., Hicks, S., Shepherd, M.H., Colclough, K., Hattersley, A.T., Ellard, S., 2010. Maturity-onset diabetes of the young (MODY): how many cases are we missing. Diabetologia 53, 2504–2508. Shober, E., Grabert, R.M., Kapellen, A.T.T., Reinehr, T., Holl, R.W., 2009. Phenotypical of maturity-onset diabetes of the young (MODY diabetes) in comparision with type 2 diabetes mellutis (T2DM) in children and adolescents: experience from a large multicentre database. Diabet. Med. 26, 466–473. Stride, A., Hattersley, A.T., 2002. Differents genes, differents diabetes: lessons from maturity-onset diabetes of the young. Ann. Med. 34, 207–216. Urrutia, I., Martínez, R., López-Euba, T., Velayos, T., Martínez de LaPiscina, I., Bilbao, J.R., et al., 2017. Lower frequency of HLA-DRB1 type 1 diabetes risk alleles in pediatric patients with MODY. PLoS One 4 (12), e0169389. Velho, G., Robert, J.J., 2002. 3. Maturity-onset diabetes of the young (MODY): genetic and clinical characteristics. Horm. Res. 57 (Suppl. 1), 29–33. Wirsing, A., Johnstone, K.A., Harries, L.W., Ellard, S., Ryffel, G.U., Stanik, J., et al., 2010. Novel monogenic diabetes mutations in the P2 promoter of the HNF4A gene are associated with impaired function in vitro. Diabet. Med. 27, 631–635.
5. Conclusion Unlike Caucasian population, GCK-MODY and HNF1A-MODY did not explain the majority of MODY cases in Tunisia, and the major gene of this pathology remains to be identified in our population. So further, studies must consider other single genes involved in β-cell function and other screening methods such us the Next generation sequencing. Competing interests The authors declare that they have no competing interests. Acknowledgements We thank patients and families, and collaborating physicians for participating in this study. We thank so much the staff of the Biochemistry Laboratory of Faculty of Pharmacy of Monastir-Tunisia, the Unit of Endocrinology of Fatouma Bourguiba Hospital of MonastirTunisia and the members of the Endocrinology and Diabetes Research Group of the Cruces hospital of Barakaldo- Spain, for their valuable contributions in this work. References Amara, A., Chadli-Chaieb, M., Ghezaiel, H., Philippe, J., Brahem, R., Dechaume, A., et al., 2012. Familial early-onset diabetes is not a typical MODY in several Tunisian patients. Tunis. Med. 90, 882–887. Awa, W.L., Thon, A., Raile, K., Grulich-Henn, J., Meissner, T., Schober, E., et al., 2011. Genetic and clinical characteristics of patients with HNF1A gene variations from the German–Austrian DPV database. Eur. J. Endocrinol. 164, 513–520. Colom, C., Corcoy, R., 2010. Maturity onset diabetes of the young and pregnancy. Best Pract. Res. Clin. Endocrinol. Metab. 24, 605–615. Ellard, S., Colclough, K., 2006. Mutations in the genes encoding the transcription factors hepatocyte nuclear factor 1 alpha (HNF1A) and 4 alpha (HNF4A) in maturity-onset
48