Journal of Autoimmunity (1992) $149-160
C4 Polymorphism in Multiplex Families with Insulin Dependent Diabetes in the Tunisian Population: Standard C4 Typing Methods and RFLP Analysis
F. Jenhani,*t R. Bardi,* Y. Go@* *Laboratory
of Immunology,
K. Ayed* and M. Jeddit
Fact&e’ de me’decine de Tunis and j’Facultk de pharmacie de Monastir,
Tunisie
(Received 9 August 1991 and accepted 8 November 1991) The polymorphism of C4A and C4B genes was investigated in Tunisian patients with insulin dependent diabetes (IDDM) and compared to family members (sibs) and to healthy controls. Multiplex families were analysed. A significant increase in C4AQO (26.86% vs 6.90%) and C4BQO (40.29% vs 8.28%) phenotypes was noted in IDDM patients compared with controls. Using RPLP analysis, we confirmed the high frequency of C4 null alleles. We also observed that most of these alleles were genes deleted in IDDM patients (72.23% OS 20% for CA4QO and 74.07% vs 16.70% for C4BQO). A significant decrease in the C4B long (14.92% vs 67.12%) form of the gene was also demonstrated by RPLP analysis compared with controls. Two haplotypes were frequently associated with IDDM patients in whom the C4A and C4B were deleted genes.
Introduction C4 is one of the proteins of the classical pathway of complement. The C4 variants in
humans exist as two isotypes named C4A and C4B [ 11. These isotypes are coded by tandem loci situated in the HLA complex in chromosome 6 [2, 31 at a distance of 10 kb apart, and localized at 30 kb from the Bf and C2 genes. The C4 complement factor is highly polymorphic [4] with more than 30 different allotypic variants described [5]. In addition, several authors have reported the importance of these two loci of the C4 gene in autoimmune diseases, including in disseminated lupus erythematosus [6-91, type I juvenile insulin-diabetes mellitus [lO-121, and in juvenile Correspondence to: Dr Faouzi Jenhani, Boite Postale 365, 1004 El Menzeh, Tunis, Tunisia. 149 0896-8411/92/020149+
12 $03.00/O
0 1992 Academic Press Limited
150 F. Jenhaniet al. dermatomyositis [ 131. Type I insulin-dependent diabetes mellitus is associated with a high frequency of two haplotypes in linkage disequilibrium with DR3 and DR4, both carrying a C4 null allele: C4AQO and C4BQO. The standard method used to study C4 protein variants is electrophoresis of neuraminidase treated serum or plasma, followed by specific immunofixation [4]. The RFLPs of the HLA class III genes, and particulary those of C4, can be useful in this approach by providing multiple variants that disclose differences between genes whose products are indistinguishable by the standard method of Alper. The present studies were designed to investigate the polymorphism of the C4 gene in IDDM patients and in their sibs in the Tunisian population.
Materials and methods IDDM
patients and control subjects
Twenty-five unrelated multiplex families with more than two diabetic children per family were included in this study. These families were recruited after consultation at the Endocrinology Units of Tunis hospitals. This group of multiplex families was composed of 25 families with two diabetic children, three families with three diabetic subjects, the two families with four diabetic members. Ten patients with IDDM from simplex families were also recruited. Thus 67 children (39 females and 28 males, ranging in age from 1 to 21 years, average age 14 + 1.5 years, average duration of diabetes 3 + 1.5 years) were studied. One hundred and twenty six non-diabetic siblings (67 females and 59 males, ranging in age from 1 to 32 years, average age 36 f 2.5 years) were studied. All families were Tunisian. All patients were documented as having suffered from diabetes since before age 12, having ketosis or ketoacidosis, and to have required insulin since the time of diagnosis.
Control subjects Normal subjects were 73 healthy Tunisian volunteers.
C4 antigen typings Complement C4 typings were performed by high-voltage agarose electrophoresis and subsequent immunofixation according to Alper et al. (1972) and to Awedeh and Alper (1980) [ 14, 41. Specific C4 antiserum was obtained from Atlantic Autoantibodies Co. (ME, USA). Heterozygosity for C4AQO and C4BQO alleles was determined by comparing the intensities of stained C4A and C4B bands [15] as described elsewhere.
DNA isolation and C4 gene RFLP analysis DNA was isolated and digested with the restriction enzymes Taq I and BamHI according to the manufacturer’s recommendations (Boerhinger Mannheim) and used
Table
151
in multiplex families with IDDM
C4 polymorphism
1. Frequencies of C4A and C4B phenotypes in IDDM
patients and in control
subjects IDDM patients (T = 67) C4A and C4B
phenotypes I-LOCUS C4AQO C4Al C4A2 C4A3 C4A6 C4AX II-LOCUS C4BQO C4Bl C4B2 C4BX
Controls (T = 73)
n
0, ,‘”
n
0 ,*
RR
x2
P
36 2 10 84 2 0
26.86 1.49 7.46 62.68 1.49 0
10 6 16 98 8 8
6.90 4.10 10.95 67.12 5.5 5.5
4.65 0.45 0.68 0.82 0.37 0.11
9.28 0.83 0.48 11.80 1.31 3.82
C4 Polymorphism in Multiplex Families with Insulin Dependent Diabetes in the Tunisian Population: Standard C4 Typing Methods and RFLP Analysis
F. Jenhani,*t R. Bardi,* Y. Go@* *Laboratory
of Immunology,
K. Ayed* and M. Jeddit
Fact&e’ de me’decine de Tunis and j’Facultk de pharmacie de Monastir,
Tunisie
(Received 9 August 1991 and accepted 8 November 1991) The polymorphism of C4A and C4B genes was investigated in Tunisian patients with insulin dependent diabetes (IDDM) and compared to family members (sibs) and to healthy controls. Multiplex families were analysed. A significant increase in C4AQO (26.86% vs 6.90%) and C4BQO (40.29% vs 8.28%) phenotypes was noted in IDDM patients compared with controls. Using RPLP analysis, we confirmed the high frequency of C4 null alleles. We also observed that most of these alleles were genes deleted in IDDM patients (72.23% OS 20% for CA4QO and 74.07% vs 16.70% for C4BQO). A significant decrease in the C4B long (14.92% vs 67.12%) form of the gene was also demonstrated by RPLP analysis compared with controls. Two haplotypes were frequently associated with IDDM patients in whom the C4A and C4B were deleted genes.
Introduction C4 is one of the proteins of the classical pathway of complement. The C4 variants in
humans exist as two isotypes named C4A and C4B [ 11. These isotypes are coded by tandem loci situated in the HLA complex in chromosome 6 [2, 31 at a distance of 10 kb apart, and localized at 30 kb from the Bf and C2 genes. The C4 complement factor is highly polymorphic [4] with more than 30 different allotypic variants described [5]. In addition, several authors have reported the importance of these two loci of the C4 gene in autoimmune diseases, including in disseminated lupus erythematosus [6-91, type I juvenile insulin-diabetes mellitus [lO-121, and in juvenile Correspondence to: Dr Faouzi Jenhani, Boite Postale 365, 1004 El Menzeh, Tunis, Tunisia. 149 0896-8411/92/020149+
12 $03.00/O
0 1992 Academic Press Limited
150 F. Jenhaniet al. dermatomyositis [ 131. Type I insulin-dependent diabetes mellitus is associated with a high frequency of two haplotypes in linkage disequilibrium with DR3 and DR4, both carrying a C4 null allele: C4AQO and C4BQO. The standard method used to study C4 protein variants is electrophoresis of neuraminidase treated serum or plasma, followed by specific immunofixation [4]. The RFLPs of the HLA class III genes, and particulary those of C4, can be useful in this approach by providing multiple variants that disclose differences between genes whose products are indistinguishable by the standard method of Alper. The present studies were designed to investigate the polymorphism of the C4 gene in IDDM patients and in their sibs in the Tunisian population.
Materials and methods IDDM
patients and control subjects
Twenty-five unrelated multiplex families with more than two diabetic children per family were included in this study. These families were recruited after consultation at the Endocrinology Units of Tunis hospitals. This group of multiplex families was composed of 25 families with two diabetic children, three families with three diabetic subjects, the two families with four diabetic members. Ten patients with IDDM from simplex families were also recruited. Thus 67 children (39 females and 28 males, ranging in age from 1 to 21 years, average age 14 + 1.5 years, average duration of diabetes 3 + 1.5 years) were studied. One hundred and twenty six non-diabetic siblings (67 females and 59 males, ranging in age from 1 to 32 years, average age 36 f 2.5 years) were studied. All families were Tunisian. All patients were documented as having suffered from diabetes since before age 12, having ketosis or ketoacidosis, and to have required insulin since the time of diagnosis.
Control subjects Normal subjects were 73 healthy Tunisian volunteers.
C4 antigen typings Complement C4 typings were performed by high-voltage agarose electrophoresis and subsequent immunofixation according to Alper et al. (1972) and to Awedeh and Alper (1980) [ 14, 41. Specific C4 antiserum was obtained from Atlantic Autoantibodies Co. (ME, USA). Heterozygosity for C4AQO and C4BQO alleles was determined by comparing the intensities of stained C4A and C4B bands [15] as described elsewhere.
DNA isolation and C4 gene RFLP analysis DNA was isolated and digested with the restriction enzymes Taq I and BamHI according to the manufacturer’s recommendations (Boerhinger Mannheim) and used
Table
151
in multiplex families with IDDM
C4 polymorphism
1. Frequencies of C4A and C4B phenotypes in IDDM
patients and in control
subjects IDDM patients (T = 67) C4A and C4B
phenotypes I-LOCUS C4AQO C4Al C4A2 C4A3 C4A6 C4AX II-LOCUS C4BQO C4Bl C4B2 C4BX
Controls (T = 73)
n
0, ,‘”
n
0 ,*
RR
x2
P
36 2 10 84 2 0
26.86 1.49 7.46 62.68 1.49 0
10 6 16 98 8 8
6.90 4.10 10.95 67.12 5.5 5.5
4.65 0.45 0.68 0.82 0.37 0.11
9.28 0.83 0.48 11.80 1.31 3.82
