PREVALENCE OF THE MISSENSE MUTATION ...

Diabetes Metab (Paris) 2002, 28, 39-44

PREVALENCE OF THE MISSENSE MUTATION GLY574SER IN THE HEPATOCYTE NUCLEAR FACTOR-1α IN AFRICANS WITH DIABETES C. COLLET (2), M. DUCORPS (1), H. MAYAUDON (1), O. DUPUY (1), F. CEPPA (1), P. BOUTIN (3), P. FROGUEL (2), B. BAUDUCEAU (1)

SUMMARY - Background: Clinical presentation and natural history of diabetes are somewhat different in Black Africans compared to Caucasians. This peculiar disease course could be at least partly related to a specific genetic profile that has not been studied in this population. Methods : Medical backgrounds, anthropometric and biologic parameters were obtained from 69 diabetic subjects in Dakar, Senegal, in 1998. Blood anti GAD and Islet Cell Antibodies were studied, using RIA and immunofluorescence assay. The HNF-1α gene was sequenced searching the Gly574Ser mutation, previously described in MODY 3. Results : Among these 69 diabetic patients, 11 (16%) were found to have the G574S mutation affecting the HNF-1α. These 11 patients carrying the mutation were compared respectively with the 58 non carriers. Mean age (57.5 yr. ± 11 vs 51.1 yr. ± 15) and duration of diabetes (11.9 vs 6.7 yr), were similar in the two groups. BMI was not different in patients with the mutation (26.3 vs 23.3, p = 0.06). Metabolic control (Glycosylated hemoglobin) was poor in the two groups (9.5% vs 9.2%). Chronic complications were equally found in the patients, but no mutation carrier had macroangiopathy. None of the anti GAD positive or ICA positive patients had the mutation. Conclusions : The HNF-1α Gly574Ser mutation was found in 16% of cases in a 69 diabetic patients group in Senegal. Diabetes was as severe as in non carriers of mutation. This mutation has been implicated in atypical diabetes of Afro-American children. The study confirms its prevalence in Africans with diabetes.

RE´SUME´ - Prévalence de la mutation Gly574Ser du HNF-1α dans une population de diabétiques africains. Contexte : L’histoire naturelle et la présentation clinique du diabète sont fréquemment atypiques chez le sujet noir. Ces particularités pourraient être partiellement liées à un profil génétique particulier, non encore étudié dans cette population. Méthodes : Les antécédents personnels et familiaux, les caractéristiques cliniques et biologiques de 69 patients diabétiques hospitalisés à Dakar, Sénégal, ont été étudiés. Pour chacun d’entre eux, les anticorps anti-GAD et Anti Cellules d’Ilots (ICA) ont été recherchés. Le gène du HNF-1α (MODY 3) a été séquencé à la recherche de la mutation Gly574Ser. Cette mutation a été impliquée dans une forme atypique de diabète chez des enfants noirs américains. Résultats : Parmi ces 69 patients, 11 (16 %) étaient porteurs de la mutation, et ils ont été comparés à ceux qui ne l’avaient pas. L’âge moyen (57,5 ± 11 ans vs 51,1 ± 15 ans) et la durée moyenne du diabète (11,9 ans vs 6,7 ans ) étaient similaires dans les deux groupes. L’Indice de Masse Corporelle n’était pas différent chez les patients ayant la mutation (26,3 vs 23,3, p = 0,06). L’équilibre glycémique était médiocre dans les deux groupes (HbA1c 9,5 % vs 9,2 %). L’incidence des complications microangiopathiques était similaire, mais aucun sujet porteur de la mutation n’avait de complication macroangiopathique, à facteurs de risque égaux. Chez les patients anti-GAD ou ICA positifs, aucun n’avait la mutation Gly574Ser. Conclusion : Cette étude confirme l’existence de la mutation G574S chez les sujets diabétiques noirs africains. Le diabète paraît aussi sévère que chez les non porteurs. Le rôle de la mutation dans l’expression du phénotype reste mal connu.

Key-words: diabetes mellitus, MODY, Africa, HNF 1α.

Mots-clés : Diabète sucré, MODY, Afrique, HNF 1α.

✍

: B. Bauduceau, Hôpital d’Instruction des Armées Bégin, 69 avenue de Paris. 94160, Saint-Mandé, France. Received : March 6th, 2001 ; revised : November 15th, 2001

(1) Service d’Endocrinologie, Hôpital d’Instruction des Armées Bégin, 94160 Saint Mandé, France (2) Service de Diabétologie, Hôpital Bichat-Claude Bernard, 46 Bd Huchard, 75018 Paris, France (3) Institut Pasteur de Lille, 1 rue du Professeur Calmette, BP 245, 59019 Lille Cedex, France.

© 2016 Elsevier Masson SAS. Tous droits réservés. - Document téléchargé le 08/02/2016 par Centre hospitalier de Versailles - (532257)

40

C. COLLET et al.

he prevalence of diabetes mellitus in Black Africa is relatively low, and most patients have type 2 diabetes. Another form of diabetes mellitus has been described in black African-American children [1] and in black African adults, called “atypical diabetes mellitus”. Patients exhibits features of both type 1 and type 2 diabetes mellitus, lacking the biochemical and genetic markers of type 1 diabetes. Patients are younger and leaner than usually observed in caucasian type 2 diabetes. Most of the time, they are not insulin treated, but can present rapid and severe decompensations requiring insulin therapy for short periods [2]. Mutations of the hepatocyte nuclear factors (HNF) are implicated in Maturity Onset Diabetes of the Young (MODY) [3]. Heterozygous mutations of HNF-1α cause a common form of MODY called MODY 3 [4]. Missense mutation Gly574Ser was first described in a black African subject from Mali, but never found in 200 Caucasians or Asians. Froguel et al. showed that this missense mutation occurs at a higher frequency among African-American children with atypical diabetes mellitus than in normal African-American children [1]. In order to assess the prevalence of this mutation in african diabetic subjects, and its role in phenotype, we screened a population of Black African diabetic subjects for the Gly574Ser mutation, and compared phenotypes of mutation carriers to non carriers.

T

m SUBJECTS AND METHODS The study was held in Dakar, Senegal, during 9 months (April 1997-July 1998), in one of the biggest hospitals of this country. All diabetic patients attending the Internal Medicine Department during this period were included. The ethnic distribution of these patients was representative of ethnic distribution in Senegal, and no significant difference was observed in clinical presentation of diabetes and ethnicity [5]. A total of 69 patients with diabetes were included. Diabetes mellitus was defined according to the WHO 1985 criteria [6], and the new diabetes mellitus classification was used [7]. For each patient, medical background was investigated, looking for familial history of diabetes, personal history of malnutrition or obesity, alcohol and smoking habits. Clinical examination was also performed, with special attention to chronic complications of diabetes. Height, weight, systolic and diastolic blood pressure were measured. Each subject had an ECG, abdominal X-ray or ultrasound, looking for pancreatic calcifications. An eye fundus was performed for each subject. Diabetic retinopathy, when present, was classified in four stages, ranging from absent to proliferating retinopathy. Diabetic neuropathy was de-

Diabetes Metab

fined by one or more of the following items: lower limbs reflexes missing, lower limbs paresthesia, impotency and neuropathic foot ulcer. All patients had their urine collected for 24 hours, from which a sample was tested for albuminuria. Urinary Albumin Excretion Rate (UAER) was defined as (total albuminuria in 24 h Urine) reported on total volume of 24 h urine, expressed in mg/day. When UAER was between 30 mg/d and 300 mg/d, the patient was considered having beginning diabetic nephropathy (microalbuminuria). When UAER was over 300 mg/d or had creatininemia over 150 µmol/l, he was considered having overt diabetic nephropathy. All patients were asked for clinical manifestations of angina pectoris, history of coronaropathy or peripheral vascular disease (intermittent claudication, antecedent of lower limb amputation), and lower limbs pulses were systematically checked. Silent myocardial infarction was looked for on a systematic resting electrocardiograph (ECG). Two blood samples and one urine sample were collected for each patient, and frozen. At the end of the study, all samples were sent to the Biochemical laboratory of Begin Hospital in Paris. The following measurements were made: Fasting blood glucose, creatininaemia, glycosylated haemoglobin and fructosamin, C-peptide, anti Glucose Amine Decarboxylase antibodies (anti GAD antibodies), islet cell antibodies (ICA) and anti-insulin antibodies. DNA was extracted from blood white cells in order to sequence Hepatocyte Nuclear Factor-1α (HNF-1α) gene. The Gly574Ser mutation is the result of the transition of a Guanine to an Adenine, in position 574, thus transforming Glycin into Serin Amino Acid. The Gly574Ser mutation does not modify the slicing sites of the gene; therefore RFLP technique is not usable to detect it. HNF-1α exon 9 was amplified by PCR technique, the sequenced using Big Dye Terminator Cycle Sequencing (Perkin Elmer). Electrophoresis gels were analyzed and compared to the wild type allele with Sequence Analysis and Sequence Navigator software’s. A genotype for codon 574 is then obtained: G/G, G/A or A/A. Statistics were performed on Statview 5.0 (SaS Institute, Inc.), using descriptive statistics. Comparisons between the three groups were made using Fischer’s exact test, because of small size of samples. Means and standard deviations (SD) were compared with umpaired T-test. An ANOVA comparing mutation carriers with non carriers, clinical type 2 subjetcs was performed for comparisons of age, BMI, C-peptide, diabetes duration, age at diagnosis of diabetes and SBP: Systolic blood pressure/DBP: Diastolic blood pressure. C-peptide and other variables with non normal distribution were compared with non parametric test. A 95% confidence interval was used.


Vol. 28, n° 1, 2002

m RESULTS Over a 9 months period, 69 diabetic patients were included in the study. Among these 69 patients, 11 (16%) were found to have the Gly574Ser mutation of the HNF-1α gene, and were compared with the non carriers subjects. Out of these 11 patients with the mutation, 10 were heterozygous for the Ser allele and 1 was homozygous. The main characteristics and differences of the patients are given in the (Table I). Hypertension prevalence was higher in affected subjects (8 out of 11) than in non affected subjects (17 out of 58). The difference persisted after exclusion of clinically type 1 diabetes patients among those without mutation (Table I). Among patients with G574S mutation, 2 had a clinical phenotype compatible with type 1 diabetes (former IDDM) and 9 had a clinical phenotype of type 2 diabetes (former NIDDM) according to the 1997 diabetes classification [6]. BMI was not significantly different between mutation carriers and non carriers, especially when type 1 phenotypic subjects were observed apart (Table I). One can note that the mean BMI of type 2 diabetes patients as of mutation carriers is relatively low, and certainly lower than what could be observed in a Caucasian type 2 diabetic population. This is a common characteristic of diabetic subjects in Africa, where obesity is still a rare condition. Among mutation carriers, 4 patients were treated with insulin alone, whereas 3 received metformine alone, 2 metformine in combination with glibenclamide and 2 were on diet alone. Among patients without the mutation, 15 had a clinical phenotype of type 1 diabetes, 42 had clinical type 2 diabetes and one had a diabetes secondary to chronic pancreatitis. Treatment of the non affected patients was as follow: 15 with diet alone, 11 received glibenclamide alone, 6 metformine alone, 4 metformine and glibenclamide in combination, 21 insulin alone, and one with a combination of the three treatments. Microangiopathic complications were equally found in mutation carriers than in non carriers. The presence of microalbuminuria was well correlated with systolic and diastolic blood pressure in the two groups. Despite a longer duration of diabetes and a higher prevalence of hypertension, none of the G574S patients had macroangiopathy, whereas 12% of non carrier subjects had coronaropathy, 7% had lower limb arteriopathy and 9% had cerebral stroke. The difference is not significant, but the power of the test was low due to small number of patients. None of the mutation carriers had positive antibodies (GAD, ICA or IAA). Anti GAD antibodies were present in the serum of 6 patients out of the 60 tested (positive >1 UI/l/ml). In each case, clinical classification was type 1 diabetes. Anti-insulin antibodies (IAA) were positive in 3 patients, all having type 1

MUTATIONS IN AFRICANS WITH DIABETES

41

diabetes. Islet Cell Antibodies (ICA) were positive in 9 patients (17%). Two patients, all classified as type 1 diabetes, were found to have the 3 different antibodies. Two patients with type 1 diabetes had two positive antibodies, just as had one patient with type 2 diabetes. Three type 1 diabetes subjects had only 1 positive antibody, as had two type 2 diabetic patients and one with chronic pancreatitis. m DISCUSSION World wide prevalence of diabetes mellitus is actually exploding [8, 9]. There are no updated studies on prevalence of type 2 diabetes in Africa, where it is considered to be rare, around 1% of general population [11-15]. Prevalence of diabetes mellitus is inversely high in black population of the Diaspora, reaching 9% of general population in the AfroCaribbean’s and 11% in African-Americans [16-18]. It is commonly admitted that Afro-Americans and Africans from western Africa share common ancestry. We can therefore think that the prevalence of diabetes in these two populations should meet in the future, when socioeconomic differences will lessen. The G574S mutation was first described in one adult diabetic subject from Mali. Froguel et al. [1] showed that it was very frequently found in black African-American children with atypical type 2 diabetes. The prevalence of this mutation was much higher in black AfroAmerican diabetic children and type 2 diabetic AfroAmerican adults than in the control population: 66.7% of African-American diabetic children tested had the Gly574Ser genotype, compared to 7% of non diabetic African-American children. Among African-American adults, 7.1% of diabetic subjects had the mutation compared to none of the non diabetic subjects [1]. There is no disponible evaluation of the prevalence of this mutation in non diabetic population of western Africa. In our study, no control group was available, due to the local conditions of survey. But we confirmed the high prevalence of the G574S mutation in 16% of the diabetic black subjects tested in this study. And according to the hypothesis of a common ancestry, one can think that what was found in diabetic and non diabetic Afro-Americans, could be true for west Africans. The Gly574Ser mutation is a missense mutation of the exon 9 of the Hepatic Nuclear Factor-1α (HNF1α) on chromosome 12q. It is part of the mutations described in diabetes caused by genetic defects of β-pancreatic cells, formerly called MODY. The mutation affecting HNF-1α causes MODY 3, which account for 1/3 of MODY Diabetes in Europe [3, 4]. Several mutations of the HNF-1α have been described to date, in Caucasian and Asian populations, and data are missing in black population [19-21]. Some of these publications have shown that mutations of HNF-1α can be associated with late onset type 2


42

C. COLLET et al.

Diabetes Metab

TABLE I. Principal characteristics of subjects with and without Gly574Ser genotype.[p= comparison between mutation (+) subjects and mutation(–) subjects. p* = comparison between mutation (+) subjects and mutation (–) subjects clinically type 2. p** = comparison between mutation (+) subjects and mutation (–) subjects clinically type 1].

G574S Mutation Absent Clinical classification type 2

G574S Mutation Absent Clinical classification type 1

p= p* = p** =

Item studied

G574S Mutation Present

G574S Mutation Absent All subjects

Number of subjects

11 (16 %)

58 (84 %)

42

15

–

57.5 ± 11.6

51.1 ± 15.1

57.4 ± 10.1

33.8 ± 13.4

p = NS p* = 0.98 NS p** < 0.001

Male= 5 (45 %)

Male= 27 (47 %)

20

6

p = NS

Female = 6 (55 %)

Female = 3 (53 %)

22

9

p*, p** = NS

7/3 (66 %)

25/31 (43 %)

17/25

6/8

p = NS p*, p** = NS

26.35 ± 5.06

23.33 ± 4.79

24.6 ± 4.5

19.4 ± 3.3

p =0.06 p* = 0.33 NS p** = 0.0003

11.9

6.7

5.8

9.3

p = 0.03 p* = 0.03 p** = 0.45

45

44.5

51.6

24.5

p = NS p* = 0.08 p** = 0.0001

SBP/DBP (mmHg)

155 ± 25/ 99 ± 18

138 ± 27/ 83 ± 16

135 ± 23/ 82 ± 15

146 ± 36/ 83 ± 19

p = 0.05/0.004 p* = 0.08/0.01 p** = 0.49/0.04

HbA1c (%, N < 6 %)

9.5 ± 2.4

9.2 ± 2.7

1.97 ± 1.96 (n = 10)

0.89 ± 0.71 (n = 29)

Diabetic Retinopathy

12 %

27 %

p = NS

Diabetic neuropathy

27 %

24 %

p = NS

7 (60 %)

23 (39 %)

p = NS

– 30 mg/d < UAER < 300 mg/d

2 (18 %)

10 (17 %)

– UAER > 300 mg/d

4 (36 %)

9 (15 %)

1

4

9/2

8/50

p = NS

– angina pectoris/myocardial infarction

0

12 %

p =NS

– Lower limbs Arteriopathy/Amputation

0

7%

p = NS

0

9%

p = NS

0/0/0

5/9/3

Age (years)

Sex

Familial history of diabetes (Yes/No) BMI (kg/m2)

Duration of diabetes

Age at diagnosis (years)

C-Peptide (ng/ml)

Diabetic Nephropathy

– creatininemia > 150 µmol/l Tabagism: Yes/No

p = NS 1.03 ± 0.72 (n=34)

0.45 ± 0.45 (n = 12)

p = 0.01 p* = 0.02 p** = 0.01

Macroangiopathy:

– Stroke Antibodies: anti-GAD/ICA/IAA

1/4/1


4/5/2

p = NS

Vol. 28, n° 1, 2002

diabetes, as well as apparent type 1 diabetes [22, 23]. The clinical and genetic heterogeneity of MODY 3 is big. It appears now that HNF-1α mutations (linked to chromosome 12q) can show an incomplete penetrance for the disease status. Phenotypic analysis of members from four large Finnish MODY 3 kindred revealed that diabetic members and normoglycemic members both had a severe impairment of insulin secretion when they had the MODY 3 gene [24]. Stimulated insulin secretion was not studied in our diabetic subjects, but basal C-peptide was mesured for most of them and showed normal value (0.66-2.5 ng/ml). Classically, MODY 3 usually begins after puberty, it has an autosomal dominant inheritance and frequently needs insulin to achieve metabolic control. The severity of this type of diabetes is estimated between MODY 2 (caused by glucokinase mutation) and common type 2 diabetes. The prevalence of microangiopathic complications is similar as the one observed in type 2 diabetes, with a trend towards a higher frequency of proteinuria [3]. This has not been studied in black population. A surprising finding is the lack of macroangiopathy in the mutated group despite a longer duration of the disease, a higher mean age and a greater prevalence of hypertension. The G574S polymorphism could therefore possibly confer a relative protection against macroangiopathy. The mechanisms involved in this protection are unknown. Hepatic nuclear factors are expressed in β pancreatic cells, but also in other tissues. Another study is available on the prevalence of macroangiopathy in MODY 3 diabetes [10], among a finnish population (n = 57 patients), and reports a smaller prevalence of macroangiopathy compared to type 2 diabetes patients (16% versus 33%, p < 0.02). The prevalence of microangiopathy in this study is similar in the two groups of diabetic patients. The mean age of our diabetic population was relatively high, and the mean age at diagnosis of diabetes among mutation carriers was not significantly lower than in type 2 non carriers (45 years versus 51), despite a tendency, which doesn’t plead on MODY3‘s diagnosis favor. We see two objections to that point of view. First of all, it is only recently that the prevalence of diabetes is rising in western Africa, due to increased urbanization and occidental way of living. Diabetes still remains a disease of wealthy “old” people, and it is not surprising to have an older population that in occidentalised countries. Secondly, it is possible that this particular mutation has an incomplete penetrance and only causes overt diabetes in older subjects. Further studies will help precise the exact role of this frequent mutation in diabetes mellitus in black subjects.

MUTATIONS IN AFRICANS WITH DIABETES

43

m CONCLUSION Prevalence of diabetes mellitus is actually exploding worldwide and especially in developing countries. Clinical presentation of diabetes is often atypical in tropical Africa. We showed that the prevalence of the G574S mutation of the HNF-1α is high in a diabetic black population. The mutation seems to be associated with a different phenotype, similar to MODY 3 phenotype described in Caucasians and a very low prevalence of macroangiopathy. Unfortunately, no study on the prevalence of the G574S mutation among non diabetic subjects of the same origin is available to date, as well as no functional study of the mutated HNF-1α has been reported to our knowledge. These two points really need to be lightened before conclusions on the role of the G574S mutation in diabetes of African subjects can be made.

REFERENCES 1 Boutin P, Gresh L, Cisse A et al. Missense mutation Gly574Ser in the transcription factor HNF-1α is a marker of atypical diabetes mellitus in African-American children. Diabetologia, 1999, 42, 380-381. 2 Ducorps M, Ndong W, Jupkwo B, Belmejdoub G, Poirier JM, Mayaudon H, Bauduceau B. Epidemiological aspects of diabetes in Cameroon: What is the role of tropical diabetes? Diabetes Metab, 1997, 23, 61-67. 3 Froguel P, Vaxillaire M, Velho G. Genetic and metabolic heterogeneity of maturity onset diabetes of the young. Diabetes Reviews, 1997, 5, 123-130. 4 Vaxillaire M, Boccio V, Philippi A et al. A gene for maturity onset diabetes of the young (MODY) maps to chromosome 12q. Nature Genetics, 1995, 9, 418-423. 5 Collet-Burgel C, Cellier C, Mbaye PS. Aspects cliniques du diabète sucré en Afrique sub-saharienne: étude prospective à Dakar, Sénégal. Revue française d’Endocrinologie Clinique, 1998, 39, 603-608. 6 World Health Organization: Diabetes Mellitus: Report of a WHO Study Group. Geneva, World Health Org (Tech. Rep. Ser. no. 727) 1985. 7 The expert committee on the diagnosis and classification of diabetes mellitus. Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care, 1997, 20, 1183-1197. 8 Amos AF, McCarty DJ, Zimmet P. The rising global burden of diabetes and its complications: Estimates and projections to the year 2010. Diabetic Medicine, 1997, 14, S7-S85. 9 King H, Rewers M. Diabetes in adults is now a Third World problem. Bulletin of the World Health Organization, 1991, 69, 643-648. 10 Isomaa B, Henricsson M, Lehto M et al. Chronic diabetic complications in patients with MODY3 diabetes. Diabetologia, 1998, 41, 467-473. 11 Owoaje EE, Rotimi CN, Kaufman JS, Tracy J, Cooper RS Prevalence of adult diabetes in Ibadan, Nigeria. East Afr Med, 1997, 74, 299-302. 12 Mbanya JC, Ngonang J, Salah JN. Prevalence of NIDDM and impaired glucose tolerance in a rural and an urban population in Cameroon. Diabetologia, 1997, 40, 824-829. 13 Ducorps M, Baleynaud S, Mayaudon H, Castagne C, Bauduceau B. A prevalence survey of diabetes in Mauritania. Diabetes Care, 1996, 19, 761-763. 14 Mac Larty DG, Swai ABM, Kitange HM et al. Prevalence of diabetes and impaired glucose tolerance in rural Tanzania. Lancet, 1989 , 871-874. 15 Swai ABM, Lutale J, McLarty DG. Diabetes in tropical Africa: a prospective study, 1981-7. BMJ, 1990, 300, 1103-1110.


44

C. COLLET et al.

16 Rotimi CN, Cooper RS, Okosun IS et al. Prevalence of diabetes and impaired glucose tolerance in Nigerians, Jamaicans and US blacks. Ethn Dis Spring-Summer, 1999, 9, 190-200. 17 Osei K. Metabolic consequences of the West African diaspora: lessons from the thrifty gene. J Lab Clin Med, 1999, 133, 98-111. 18 Cooper RS, Rotimi CN, Kaufman JS et al. Prevalence of NIDDM among populations of the African Diaspora. Diabetes Care, 1997, 20, 343-348. 19 Yamada S, Tomura H, Nishigori H et al. Identification of mutations in the hepatocyte nuclear factor-1α gene in Japanese subjects with earlyonset NIDDM and functional analysis of the mutant proteins. Diabetes, 1999, 48, 645-648. 20 Elbein SC, Teng K, Yount P, Scroggin E. Linkage and molecular scanning analyses of MODY 3/hepatocyte Nuclear factor 1α gene in typical familial type 2 diabetes: evidence for novel mutations in exons 8 and 10. J Clin Endocrinol Metab, 1998, 83, 2059-2065.

Diabetes Metab

21 Kaisaki P, Menzel S, Lindner T et al. Mutations in the hepatocyte nuclear factor 1α gene in MODY and Early-onset NIDDM. Diabetes, 1997, 46, 528-535. 22 Moller AM, Dalgaard LT, Pociot F, Nerup J, Hansen T, Pedersen O. Mutations in the hepatocyte nuclear factor 1α gene in caucasian families originally classified as having type 1 diabetes. Diabetologia, 1998, 41, 1528-1531. 23 Awata T, Kurihara S, Inoue K et al. A novel missense mutation in the homeodomain of the hepatocyte nuclear factor 1α/maturity onset diabetes of the young 3 in a Japanese earlyonset type 2 diabetic patient and time- course of glucose-stimulated insulin secretion. Diabetes Care, 1998, 21, 1569-1570. 24 Lehto M, Tuomi T, Mahtani MM et al. Characterization of the MODY 3 Phenotype; Early onset diabetes caused by an insulin secretion defect. J Clin Invest, 1997, 99, 582-591.

© Masson, Paris, 2002
