Abstract
The aim of the present study was to investigate the prevalence of C4 and C2 deficiencies and to characterize genomic alterations in C4 genes in a large cohort of 125 unselected patients with SLE. We determined the protein concentration and functional activity of C2 and C4, as well as the C4 phenotype. C4 genotyping included Taq 1 restricted fragment lengh polymorphism (RFLP) analysis and polymerase chain reaction using sequence-specific primers (SSP-PCR). Type I C2 deficiency was diagnosed by PCR. Overall, 79·2% of the patients exhibited abnormalities of the C4 genes including deletion, non-expression, gene conversion and duplication. Among C4-deficient patients (n = 66, 52·8% prevalence), 41·0% of the patients exhibited a C4A deficiency and 59·0% a C4B deficiency. Half of the C4 deficiencies were due to a gene deletion. There was a strong association between C4A and C4B gene deletion and the presence of the DRB1*03 allele. Among the silent C4A genes, only two cases were related to a 2-bp insertion in exon 29 of the C4A gene. A gene conversion was demonstrated in eight patients (6·4%). One patient had a homozygous C4A deficiency. Three (2·4%) patients presented with a heterozygous type I C2 deficiency and none with homozygous deficiency. Our results argue against a specific role for C4A gene deficiency in determining disease susceptibility among patients with SLE that are C4-deficient.
Keywords: complement, genetic deficiencies, systemic lupus erythematosus
Introduction
A number of genes have been identified as being associated with increased susceptibility to SLE in both human and mouse models of spontaneous SLE-like disease [1]. Two MHC-linked susceptibility factors for SLE patients of Caucasian background have been characterized, carried by the haplotypes B7-DR2 and B8-DR3 [1,2]. SLE is also associated with genetic deficiencies in complement components, i.e. C1q, and deficiencies in the C2 and C4 classical pathway proteins [3, 4]. The two isotypes of C4, C4A and C4B, are encoded by genes located in the vicinity of the genes encoding for C2, factor B and steroid 21-hydroxylase A and B (CYP21A and CYP21B) on chromosome 6. More than 75% of patients with complete C4 deficiency exhibit SLE [3]. Among patients with C4 deficiencies, patients with C4A null alleles (C4AQ0) have been predominantly reported as being susceptible to SLE, irrespective of ethnicity [2, 5]. A 28-kb deletion, involving the entire C4A gene and the adjacent CYP21A pseudogene, is frequently responsible for the deficiency [6, 7]; other null alleles are related to non-expressed genes [8]. There are only a few reports on an increased frequency of C4BQ0 alleles in SLE [9, 10]. C4B deficiencies result from a deletion in the C4B gene, or from a gene conversion event [11]. Of individuals with complete C2 deficiency, 30–50% exhibit SLE [4, 12].
The present study extensively characterized the abnormalities of the classical complement pathway components C2 and C4 at the protein and genomic levels in a large cohort of patients with SLE.
Patients and methods
Patients
One hundred and twenty-five unselected patients fulfilling the ACR criteria for the diagnosis of SLE were included in the study. Women represented 89% of the population. The median age of the patients was 36 years (17–73 years). The study group included 79 patients of Caucasian origin (63%), 19 patients from North Africa, 18 African patients and nine patients of Asian origin.
Plasma collected in EDTA 0·04 m was stored in aliquots at −80°C until use. Genomic DNA was extracted from cell pellets by the proteinase K/phenol method.
Antigenic and functional assessment of complement proteins
CH50, C2 and C4 haemolytic activities were measured according to standard procedures [13]. The results were expressed as the percentage of those obtained with a reference plasma pool of 100 healthy donors mainly comprising Caucasians. Plasma concentrations of C3, C4 and factor B antigens were measured by nephelemetry (Beckmann, Gagny, France). Normal concentrations established in our laboratory are 85 ± 25 mg/dl, 22 ± 10 mg/dl, 28 ± 12 mg/dl for C3, C4 and factor B, respectively. Complement consumption was diagnosed on the basis of concomitant decreases in CH50 and haemolytic C2 activity to below 70% of normal and a decrease in plasma concentrations of C3 antigen below 60 mg/dl. At the time of the study, 41 (32·8%) patients exhibited classical pathway-mediated complement consumption; 84 (67·2%) patients showed no evidence of complement activation.
Allotyping of C4
C4 allotypes were determined by high voltage agarose electrophoresis of plasma treated with neuraminidase type VIII (Sigma, St Louis, MO; 10 mm/ml) alone or in combination with carboxypeptidase B type I (Sigma), followed by immunofixation with goat anti-human C4 antiserum (INCSTAR, Stillwater, MN), as previously described [13]. C4 bands were visualized by staining with a brilliant blue R solution. C4 allotyping was analysed in 109 of 125 patients by considering the number of C4 bands and the relative intensities of the bands corresponding to the C4A- and C4B-specific alleles. The 16 remaining patients exhibited severe and persistent complement consumption, so only C4 cleavage products were found in plasma and C4 allotyping could not be analysed.
C4 genotyping by analysis of restriction fragment length polymorphism
Genomic DNA (10 μ g) was digested to completion with the restriction enzyme Taq I (New England Biolabs, Beverly, MA); Southern blotting was performed by standard techniques; blots were probed with a γ32P-radiolabelled 5′ end fragment of the C4 cDNA clone pAT-A [14]. The results were analysed as previously described [7].
Detection of non-expressed C4A genes
Polymerase chain reaction (PCR) amplification with sequence-specific primers was used to detect a 2-bp insertion in the exon 29 leading to C4A non-expression. C4A non-expression was defined by a profile of C4 allotyping characteristic of heterozygous C4A deficiency in the absence of the 6·4-kb DNA fragment specific for the 28-kb C4A deletion. Primers, 13 INS and A-down, and the procedure have previously been described [15]. For all PCR assays, negative and positive controls were performed with each set of samples that were tested. The presence of a 2-bp insertion was further demonstrated by DNA sequencing. A 986-bp fragment containing exon 29 was amplified by PCR using the primer A-down described above and the primer C4-ins selected according to the published C4 gene sequence (5′-CCTACTTGGGTACTGCGGAATC-3′). The amplification reaction was performed with 300 ng of genomic DNA, 50 pmol of each primer, 2 U Taq polymerase and 400 μm dNTPs using 55°C as the annealing temperature. Amplified DNA was then purified using the High pure PCR purification Kit' (Boehringer Mannheim, Meylan, France) according to the manufacturer's procedure. When C4-ins was used as the sequencing primer, sequencing was performed with the ‘Dye terminator cycle sequencing ready reaction, DNA sequencing kit’ (Abi Prism; Applied Biosystems, Perkin-Elmer, Courtaboeuf, France).
Gene conversion at the C4A and C4B locus
To exclude gene conversion in patients with C4A or C4B homozygous deficiency, as diagnosed by C4 allotyping, we amplified a 926-bp C4d fragment which encodes the antigenic determinant Chido 4 (Ch + 4) defining the C4B isotype. The oligonucleotide primers (L-3, L-4) and the PCR amplification conditions used have previously been described [11]. Following digestion with the enzyme NlaIV (New England Biolabs), one or two DNA fragments were observed in the case of Ch + 4 or Ch-4 sequences, respectively. In cases with C4A or C4B heterozygous deficiency without gene deletion, primers A-up and Ch5 or B-up and Ch-5 were used in order to detect C4B to C4A gene conversion and C4A to C4B gene conversion, respectively, using the experimental conditions previously described by Barba et al. [15].
PCR amplification of the C2 gene
PCR amplification was performed in order to detect the 28-bp deletion, characteristic of type I C2 deficiency [16].
HLA-DR genotyping
HLA-DR typing was performed using the HLA-DR oligo-detection kit (Biomérieux, Marcy-L'Etoile, France).
Statistical analysis
Descriptive analyses of the differences among the ethnic groups and the C4-deficient groups were performed using the χ2 distribution and the Fisher t-test analysis.
Results
C4 proteins and genes
The number of C4 alleles expressed was determined by assessing data derived from the antigenic and haemolytic levels of C4, combined with both phenotypic and genotypic analysis in 109 patients. The results are summarized in Table 1. Twenty-four patients expressed four C4 alleles. A partial C4 deficiency was found in 66 patients; 60 patients expressed three and six patients expressed two C4 alleles. Eleven patients expressed more than four C4 alleles. Results of C4 allotyping and genotyping were discrepant in eight patients. Finally, in the 16 patients in whom C4 phenotyping could not be performed, C4 genotyping was normal in 10 patients, three patients exhibited a C4 gene deletion and three additional patients a C4 gene duplication.
Table 1.
Results of C4 allotyping and genotyping in SLE patients (n = 125)
a. Four alleles expressed (n = 24)
| C4 allotyping | C4 genotyping | n |
|---|---|---|
| A3A3B1B1 | 4 C4 genes | 5 |
| A3A3B1B2 | 4 C4 genes | 4 |
| A3A3B1B3 | 4 C4 genes | 2 |
| A4A3B1B1 | 4 C4 genes | 1 |
| A2A3B1B1 | 4 C4 genes | 1 |
| A6A3B1B1 | 4 C4 genes | 1 |
| A2A92B1B1 | 4 C4 genes | 1 |
| A4A3B1B2 | 4 C4 genes | 1 |
| A4A3B2B2 | 4 C4 genes | 1 |
| A2A3A3B1BQ0 | C4B→C4A gene conversion | 2 |
| A4A2A3B2BQ0 | C4B→C4A gene conversion | 1 |
| A3AQ0B1B1B2 | C4A→C4B gene conversion | 1 |
| A2AQ0B1B1B2 | C4A→C4B gene conversion | 1 |
| A3AQ0B1B1B1 | ·C4A gene deletion + C4B | 1 |
| gene duplication | ||
| ·Non-expressed C4A gene + | 1 | |
| C4B gene duplication |
| b. Three alleles expressed (n = 60) | ||
|---|---|---|
| C4 allotyping | C4 genotyping | n |
| A3AQ0B1B1 | ·C4A gene deletion | 12 |
| ·Non-expressed C4A gene | 6 | |
| ·C4B gene deletion + C4A | 2 | |
| →C4B gene conversion | ||
| A3AQ0B1B2 | ·2pb C4A gene insertion | 1 |
| ·Non-expressed C4A gene | 1 | |
| A4AQ0B1B2 | ·2pb C4A gene insertion | 1 |
| ·C4A gene deletion | 1 | |
| A4AQ0B2B5 | Non-expressed C4A gene | 1 |
| A4AQ0B2B2 | Non-expressed C4A gene | 1 |
| A12AQ0B1B1 | C4A gene deletion | 1 |
| A3A3B1BQ0 | ·C4B gene deletion | 16 |
| ·Non-expressed C4B gene | 8 | |
| A2A3B1BQ0 | C4B gene deletion | 1 |
| A4A3B1BQ0 | Non-expressed C4B gene | 1 |
| A3A3B2BQ0 | Non-expressed C4B gene | 2 |
| A3A3B3BQ0 | C4B gene deletion | 1 |
| A4A3B2BQ0 | ·Non-expressed C4B gene | 1 |
| ·C4B→C4A gene conversion | 1 | |
| ·C4B gene deletion | 2 | |
| c. Two alleles expressed (n = 6) | ||
|---|---|---|
| C4 allotyping | C4 genotyping | n |
| AQ0B1 | Homozygous gene deletion | 1 |
| A3BQ0 | ·Homozygous gene deletion | 2 |
| ·Homozygous non-expressed gene | 2 | |
| A3AQ0B1BQ0 | C4A gene deletion + C4B gene deletion | 1 |
| d. More than four alleles expressed (n = 11) | ||
|---|---|---|
| C4 allotyping | C4 genotyping | n |
| A3A3A3B1B1 | C4A gene duplication | 2 |
| A3A3A6B1B1 | C4A gene duplication | 1 |
| A3A3A12B1B1 | C4A gene duplication | 1 |
| A4A2A2B2B2 | C4A gene duplication | 1 |
| A3A3B1B1B1 | C4B gene duplication | 1 |
| A2A3B1B1B1 | C4B gene duplication | 1 |
| A3A2B1B1B2 | C4B gene duplication | 2 |
| A5A3A3B3B3B1 | C4A + C4B genes duplication | 1 |
| A3AQ0B1B1B3B3 | Non-expressed C4A gene + C4B | 1 |
| gene duplication | ||
| e. Discrepant results (n = 8) | ||
|---|---|---|
| C4 allotyping | C4 genotyping | Total |
| A3<B1 | 4 C4 genes | 1 |
| A3A3B1B1 | C4B gene duplication | 1 |
| A3>B1 | C4B gene duplication | 2 |
| A3<B1 | C4B gene deletion | 2 |
| A6<B1 | C4B gene deletion | 1 |
| A3AQ0B1B3 | C4B gene deletion | 1 |
| f. C4 genotyping without C4 allotyping (n = 16) | |
|---|---|
| C4 genotyping | Total |
| 4 C4 genes | 10 |
| C4A gene deletion | 1 |
| C4A gene deletion + C4B gene duplication | 1 |
| Homozygous C4B gene deletion | 1 |
| C4B gene duplication | 3 |
In order to determine the impact of the ethnic origin of the patients on the prevalence of C4 gene abnormalities, we analysed the number of C4-expressed alleles in the Caucasian population compared with the non-Caucasian patients (Table 2). The percentages of patients exhibiting a partial C4 deficiency or expressing four normal C4 alleles did not differ between the two groups. In contrast, the percentage of patients expressing more than four alleles was significantly higher in the non-Caucasian population.
Table 2.
Number of expressed C4 alleles in the entire study population and in Caucasian and non-Caucasian patients
| Number of C4 alleles expressed | Total n = 125 | Caucasians n = 79 | Non-Caucasians n = 46 | P* |
|---|---|---|---|---|
| ≤ 3 alleles† | 66 (52·8%) | 42 (53%) | 24 (52%) | > 0·05 |
| 4 alleles† | 24 (19·2%) | 18 (23%) | 6 (13%) | > 0·05 |
| > 4 alleles† | 11 (9%) | 2 (2%) | 9 (20%) | 0·001 |
| Not determined: | ||||
| No C4 allotyping‡ | 16 (13%) | 10 (13%) | 6 (13%) | > 0·05 |
| Discrepant results§ | 8 (6%) | 7 (9%) | 1 (2%) | > 0·05 |
The percentages relative to the defined population are indicated in parentheses.
According to t-test comparing results obtained in Caucasian versus non-Caucasian groups.
As determined by combined data of C4 allotyping and genotyping.
C4 allotyping not determined because of persistent complement consumption.
Discrepant results between C4 allotyping and C4 genotyping.
Among the 66 patients with a partial C4 deficiency, as diagnosed on the basis of the combined data of C4 allotyping and genotyping, 27 (40·9%) patients exhibited a heterozygous C4A deficiency. The deficiency was due to a gene deletion in 15 (55·5%) cases. One patient exhibited a homozygous C4A deficiency due to gene deletion. The remaining cases of C4A deficiencies were caused by non-expression of the C4A gene (44·5%) (Table 3). In the latter patients, in whom we found no evidence of gene conversion, as well as in patients with severe complement consumption in whom C4 allotyping could not be performed, PCR amplification was used to detect a 2-bp insertion in exon 29 of the C4A gene to characterize further the mechanism of non-expression of the C4A gene. Two silent C4A genes were related to a 2-bp insertion in the C4A gene as detected by PCR. The results were further confirmed by sequencing the relevant portion of the gene (data not shown).
Table 3.
C4 abnormalities in SLE patients exhibiting C4 deficiency established by combined data of phenotypic and genotypic analysis
| n = 66 | C4A | C4B | |
|---|---|---|---|
| Deletion | Homozygous | 1 (1·5%) | 3 (4·6%) |
| Heterozygous | 15 (22·8%) | 21 (31·8%) | |
| Non-expression | Homozygous | 0 | 2 (3·0%) |
| Heterozygous | 11 (16·6%) | 13 (19·7%) | |
| Total | 27 (40·9%) | 39 (59·1%) |
Numbers in parentheses indicate the relative frequency of the mechanism responsible for C4 deficiencies.
Thirty-four patients (51·5% of patients with C4 deficiency) exhibited a heterozygous C4B deficiency. The deficiency was associated with a deletion in 21 (61·7%) individuals and with non-expression in the remaining 13 (38·3%) cases. Five patients presented with homozygous C4B deficiency and, in three of these, the deficiency was related to a homozygous deletion. One patient exhibited a combined C4A and C4B heterozygous deficiency. No evidence of gene conversion was found in our experimental conditions in patients presenting with C4B non-expression.
The relationship between the number of expressed C4 genes and C4 plasma levels was analysed in those patients selected on the basis of their lack of complement consumption (n = 84). Results indicated that C4 plasma levels were related to the number of expressed C4 alleles. Thus, C4 plasma levels were significantly higher in those patients with more than four alleles expressed (median 44 mg/dl) than in patients with four alleles expressed (median 23 mg/dl, P < 0·001). C4 plasma levels were also found to be higher in patients with four alleles expressed compared with patients with three alleles expressed (median 20 mg/dl, P = 0·03) (Fig. 1). C4 plasma levels tended to be lower in the group of patients expressing two alleles compared with the groups expressing three or four C4 alleles; however, the difference did not reach statistical significance, probably due to the low number of patients expressing two alleles (n = 4). Among the 41 patients who exhibited classical pathway-mediated complement consumption, no significant differences in C4 plasma levels were found between the patients exhibiting C4A or C4B deficiency compared with those without deficiency. In contrast, C4 plasma levels were significantly lower in patients with classical pathway-mediated complement consumption compared with patients with no evidence of complement consumption (median 9 mg/dl versus 21 mg/dl, P < 0·001).
Fig. 1.
Plasma levels of C4 in SLE patients according to (a) the number of C4-expressed alleles in patients with no evidence of complement consumption, and (b) the presence or absence of a C4 deficiency in patients exhibiting classical pathway-mediated complement consumption. C4 plasma levels were measured by nephelemetry. C4 null alleles were diagnosed on data obtained from both phenotypic and genotypic studies. Complement consumption was diagnosed on the basis of concomitant decreases in CH50 and haemolytic C2 to below 70% of normal and a decrease in plasma concentrations of C3 antigen to < 60 mg/dl (normal range 85 ± 25 mg/dl). Eighty-four patients showed no evidence of complement activation (a), 41 patients exhibited classical pathway-mediated complement consumption (b). Each square represents one patient. Medians are indicated for each group of patients.
More than four alleles were found to be expressed in 11 patients (Table 1d). A duplication of the C4A gene and of the C4B gene was found in five and four patients, respectively. One patient exhibited a gene duplication at both the C4A and C4B loci; one additional patient exhibited four C4B alleles. Ten patients displayed C4B gene duplication in association with another C4 gene abnormality, including C4A deletion and a silent C4A gene, so that the number of C4-expressed alleles did not exceed four. In three additional patients in whom allotyping was not available, genotyping revealed the presence of a C4B duplication.
Duplications of the C4A and C4B genes were predominantly found in patients who had originated from Africa (eight of 24), which may explain why African patients expressed a higher mean level of C4 proteins compared with non-African patients (24·5 ± 15·4 mg/dl (mean ± s.d.) versus 18·5 ± 10·5 mg/dl; P = 0·04, t-test).
Discrepant results between allotyping and genotyping of C4 were found in 16 of the 125 patients. Such a discrepancy was explained in 8/16 patients by a gene conversion from C4A to C4B (n = 4) and from C4B to C4A (n = 4).
HLA-DR genotype
In the Caucasian SLE population (analysed for HLA genotype), HLA class II alleles that were most frequently observed were DRB1*15 (17·0%), DRB1*13 (13·9%), DRB1*03 (15·1%) and DRB1*04 (15·1%). These frequencies did not differ from those observed in a French control population that consisted mainly of Caucasians, analysed in the same laboratory (Table 4). Unexpectedly, however, we found the frequency of HLA-DRB1*01 to be significantly decreased in the study population (0·049 versus 0·173, P (χ2) < 0·0001, compared with controls).
Table 4.
Results of HLA-DR B1* typing
| HLA-DR B1* | 01 | 03 | 04 | 07 | 08 | 09 | 10 | 11 | 12 | 13 | 14 | 15 | 16 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Caucasian SLE patients | 6·3 | 15·2 | 15·2 | 15·8 | 2·5 | 1·2 | 1·9 | 6·3 | 0 | 13·9 | 2·5 | 17 | 1·9 |
| (%, n = 158 alleles) | |||||||||||||
| Control group | 17·2 | 15·2 | 10·7 | 14·0 | 2·2 | 1·0 | 1·0 | 11·7 | 0·7 | 11·0 | 2·2 | 11·5 | 1·2 |
| (%, n = 400 alleles) | |||||||||||||
There was a strong association between C4A gene deletion and the presence of the DRB1*03 allele. Thus, 83% of the patients with C4A gene deletion carried this allele (P < 0·001). Forty-seven percent of patients with C4B gene deletion also exhibited the DRB1*03 allele (P < 0·001). No such association was noticed for non-expressed C4A and C4B genes.
C2 gene
A heterozygous deletion was found to be present in three patients (2·4%). All three patients exhibited the C4A*4B*2 variant, commonly associated with the C2Q0 allele. Two patients presented with a combined heterozygous C2 and C4B deficiency. One patient exhibited a combined heterozygous C2 and C4A deficiency.
Discussion
In the present study, we investigated in detail the classical pathway components C2 and C4, at the protein and genomic levels in a large cohort of patients with SLE. Overall, 79·2% of the patients exhibited abnormalities in the C4 genes including deletion, non-expression, gene conversion and duplication. Among C4-deficient patients in the study population, we found no increased prevalence of C4A deficiency compared with C4B deficiency, arguing against a specific role for C4A gene deficiency as a predisposing factor in SLE.
We have studied a large group of 125 unselected patients with SLE, irrespective of disease activity. The classical complement pathway was investigated in detail at the protein level, by measuring expression levels and the functional activity of the proteins and by determining C4 allotypes. In addition, defects were characterized at the genomic level. C4 deficiencies represented 71·7% of the C4 gene abnormalities. As previously reported [17], we found a relationship between plasma levels of C4 and the number of C4 expressed alleles. Plasma levels of haemolytic (data not shown) and immunochemical C4 were significantly decreased in C4-sufficient and C4-deficient patients with classical pathway activation compared with patients with no evidence of complement consumption, emphasizing that C4 plasma levels represent a primary indicator of complement consumption.
Previous reports have suggested that, among the C4 deficiencies, heterozygous and homozygous C4A deficiencies are more prevalent in patients with SLE [2, 5, 18, 19]. C4A deficiencies were reported to be heterozygous in 50–80% of SLE patients and homozygous in 10–22% of Caucasian SLE patients [19–22]. It was further suggested that the C4A, as opposed to the C4B, isotype preferentially transacylates onto amino group nucleophiles of immune complexes and so the C4A gene product would be more efficient at inhibiting immune precipitation and preventing immune complex clearance [3]. We found no difference in the relative prevalence of C4A and C4B deficiencies among patients with SLE, however. A deletion in the C4A gene was demonstrated in 15·2% of the study population. Moreover, only one case of homozygous C4A gene deficiency was observed, in striking contrast to the figures of 10–20% for homozygous C4A deficiencies previously reported [3]. These differences are unlikely to be related to the ethnic origin of the patients studied. C4A gene deletions were absent in a series of C4A-deficient patients from Japan [23] and none of our Asian patients presented with a C4A gene deletion (data not shown). Goldstein & Sengar found a frequency of C4A gene deletion of 12% in a population of French Canadian patients with SLE compared with 31% in non-French Canadian patients [24]. Reveille et al. also failed to demonstrate an association between SLE and C4A null alleles in three different ethnic groups [25].
C4A deficiencies result from either gene deletion or non-expression caused by either a point mutation, responsible for the occurrence of a termination codon, or a gene conversion to the C4B isotype [8]. C4B deficiencies result from a deletion in the C4B gene or from a gene conversion event in which two C4 genes, both with C4A sequences, occur on the same chromosome [11]. In our study population, C4A null alleles were caused by a gene deletion in 55·9% of the cases, and by a silent gene in 44·1%. Only two of these cases were related to a 2-bp insertion in exon 29 of the C4A gene. The frequency of the 2-bp insertion (2/15) was lower than that previously reported [8, 26]. Thirteen patients presented with a silent C4A gene that could not be associated with a specific genetic defect. Pseudogenes may result from a point mutation or a defect at the translational level. With regard to the very low prevalence of homozygous C4A deficiency in our study population, we could not exclude the possibility of homozygous C4A non-expression not related to a 2-bp insertion in the small group of patients in whom C4 allotyping could not be performed.
Half of the C4B null alleles that we observed were due to a gene deletion and half to non-expression. Five patients presented with a homozygous C4B deficiency which was due to a homozygous gene deletion in three patients and to homozygous non-expression in the two remaining cases. It has been suggested that homozygous C4B deficiency may result from either deletion or gene conversion [11]. We found no evidence, however, in our experimental conditions for a C4B to C4A gene conversion in the cases of C4B non-expression. Gene conversion was responsible for the presence of a C4 null allele (C4AQ0 or C4BQ0) in 6·4% of patients. The functional properties of these chimeric alleles remain to be characterized. The prevalence of gene duplication that we observed was similar to that previously reported in patients with SLE [27].
Our study did not confirm the high prevalence of type I C2 deficiency in patients with SLE, as previously reported [28, 29]. Only 2·4% of the patients in our cohort exhibited a partial C2 gene deficiency, a prevalence which is similar to that described in the normal Caucasian population. All patients with heterozygous C2 deficiency also exhibited a heterozygous C4A or C4B gene deletion. We have previously reported on such combined heterozygous C2–C4 deficiencies and their occurrence in patients with SLE [12]. The observed frequency of combined deficiency in the present cohort of SLE patients was higher than the expected frequency of the combined C4–C2 deficiency (approximately 0·001 in a normal Caucasian population). None of the patients displayed a profile of complement proteins that suggested a type II C2 deficiency.
Although a primary association between the alleles of HLA-DR locus and SLE has been previously reported [1], HLA class II genotyping failed to demonstrate an increased prevalence of the DRB1*03 allele in this study. A possible explanation for this finding is the lower prevalence of C4 gene deletion in our study population compared with previous reports [5, 25, 30].
SLE is a multifactorial and a multigenic disease. The involvement of heterozygous C4 and C2 deficiencies as a predisposing factor for the occurrence of SLE remains in dispute. Homozygous deficiencies of the early proteins of the classical complement pathway C1, C2 and C4 have been suggested to be associated with an increased risk of SLE because of their role in inhibiting immune precipitation [3]. A specific role for C1q in the clearance of apoptotic cells has recently been emphasized [31] and studies in C1q ‘knock-out’ mice have indicated that C1q deficiency may lead to SLE-like disease in animals of certain genetic backgrounds [32]. Other genetic factors relating to the capacity of the host to process circulating immune complexes and apoptotic cells have been identified, such as specific Fcγ receptor type II and type III gene polymorphisms [33, 34]. Genetic studies based on microsatellite analysis have provided evidence for a genetic linkage between the Fcγ receptor locus and the susceptibility to SLE in African Americans [35]. Genome-wide linkage studies in murine models [36] and in multiplex SLE families [37] have allowed the identification of new clusters of susceptibility genes. Thus, SLE is clearly a complex genetic trait with contributions from both MHC and non-MHC genes. Among the genes that are located within the MHC, results from the present study argue against a specific role for C4A gene deficiency in determining disease susceptibility among C4-deficient patients with SLE.
Acknowledgments
This work was supported by the Délégation à la Recherche Clinique/Assistance Publique-Hôpitaux de Paris (AP-HP) (project no. AOB94039) and the Institut National de la Santé et de la Recherche Médicale (INSERM). We gratefully acknowledge Eric Oksenhendler, Christian Jacquot, Jean-François Delfraissy, Cécile Goujard, Catherine Picard-Dahan, Michel Vayssairat, Luc Mouthon for including patients in the study.
REFERENCES
- 1.Vyse TJ, Kotzin BL. Genetic susceptibility to systemic lupus erythematosus. Annu Rev Immunol. 1998;16:261–92. doi: 10.1146/annurev.immunol.16.1.261. [DOI] [PubMed] [Google Scholar]
- 2.Hartung K, Baur MPC, Fricke M, et al. Major histocompatibility complex haplotypes and complement C4 alleles in systemic lupus erythematosus. Results of a multicenter study. J Clin Invest. 1992;90:1346–51. doi: 10.1172/JCI116000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hauptmann G, Tappeiner G, Schifferli J. Inherited deficiency of the fourth component of human complement. In: Rosen F, Seligmann M, editors. Switzerland: Chur, Harwood Academic; 1993. [PubMed] [Google Scholar]
- 4.Walport M. Complement deficiency and disease. Br J Rheumatol. 1993;32:269–73. doi: 10.1093/rheumatology/32.4.269. [DOI] [PubMed] [Google Scholar]
- 5.Petri M, Watson R, Winkelstein J, McLean R. Clinical expression of systemic lupus erythematosus in patients with C4A deficiency. Medicine. 1993;72:236–44. doi: 10.1097/00005792-199307000-00003. [DOI] [PubMed] [Google Scholar]
- 6.Carroll M, Palsdottir A, Belt K, Porter R. Deletion of complement C4 and steroid 21-hydroxylase genes in the HLA class III region. EMBO J. 1985;4:2547–52. doi: 10.1002/j.1460-2075.1985.tb03969.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Schneider P, Carroll M, Alper C, Rittner C, Whitehead A, Yunis E, Colten H. Polymorphism of the human complement C4 and steroid 21-hydroxylase genes. Restriction fragment length polymorphisms revealing structural deletions, homoduplications and size variants. J Clin Invest. 1986;78:650–7. doi: 10.1172/JCI112623. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Barba G, Rittner C, Schneider P. Genetic basis of human complement C4A deficiency. Detection of a point mutation leading to nonexpression. J Clin Invest. 1993;91:1681–6. doi: 10.1172/JCI116377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Reveille J, Moulds J, Arnett F. Major histocompatibility complex class II and C4 alleles in Mexican Americans with systemic lupus erythematosus. Tissue Antigens. 1995;45:91–97. doi: 10.1111/j.1399-0039.1995.tb02423.x. [DOI] [PubMed] [Google Scholar]
- 10.Naves M, Hajeer A, Teh L, et al. Complement C4B null allele status confers risk for systemic lupus erythematosus in a Spanish population. Eur J Immunogenet. 1998;25:317–20. doi: 10.1046/j.1365-2370.1998.00110.x. [DOI] [PubMed] [Google Scholar]
- 11.Braun L, Schneider P, Giles C, Bertrams J, Rittner C. Null alleles of human complement C4. Evidence for pseudogenes at the C4A locus and for gene conversion at the C4B locus. J Exp Med. 1990;171:129–40. doi: 10.1084/jem.171.1.129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hartmann D, Frémeaux-Bacchi V, Weiss L, Meyer A, Blouin J, Hauptmann G, Kazatchkine M, Uring-Lambert B. Combined heterozygous deficiency of the classical complement pathway proteins C2 and C4. J Clin Immunol. 1997;17:176–84. doi: 10.1023/a:1027334716982. [DOI] [PubMed] [Google Scholar]
- 13.Kazatchkine M, Haupmann G, Nydegger U. Techniques du complement. Paris: L. E. INSERM; 1985. [Google Scholar]
- 14.Belt K, Yu C, Carroll M, Porter R. The structural basis of the multiple forms of human complement component C4. Cell. 1984;36:907–14. doi: 10.1016/0092-8674(84)90040-0. [DOI] [PubMed] [Google Scholar]
- 15.Barba G, Braun-Heimer L, Rittner C, Schneider P. A new PCR-based typing of the Rodgers and Chido antigenic determinants of the fourth component of human complement. Eur J Immunogenet. 1994;21:325–39. doi: 10.1111/j.1744-313x.1994.tb00203.x. [DOI] [PubMed] [Google Scholar]
- 16.Johnson C, Densen P, Hurford RKJC, Wetsel R. Type I human complement C2 deficiency. A 28-base pair gene deletion causes skipping of exon 6 during RNA splicing. J Biol Chem. 1992;267:9347–53. [PubMed] [Google Scholar]
- 17.Hammond A, Ollier W, Walport M. Effects of C4 null alleles and homoduplication on quantitative expression of C4A and C4B. Clin Exp Immunol. 1992;88:163–8. doi: 10.1111/j.1365-2249.1992.tb03057.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Reveille J, Barger B, Hodge T. HLA-DR2-DRB1 allele frequencies in DR-positive black Americans with and without systemic lupus erythematosus. Tissue Antigens. 1991;38:178–80. doi: 10.1111/j.1399-0039.1991.tb01892.x. [DOI] [PubMed] [Google Scholar]
- 19.Fielder A, Walport M, Batchelor J, Rynes R, Black C, Dodi I, Hughes G. Family study of the major histocompatibility complex in patients with systemic lupus erythematosus: importance of null alleles of C4A and C4B in determining disease susceptibility. Br Med J. 1983;286:425–8. doi: 10.1136/bmj.286.6363.425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Reveille J, Arnett F, Wilson R, Bias W, McLean R. Null alleles of the fourth component of complement and HLA haplotypes in familial systemic lupus erythematosus. Immunogenetics. 1985;21:299–311. doi: 10.1007/BF00430796. [DOI] [PubMed] [Google Scholar]
- 21.Kemp M, Atkinson J, Skanes V, Levine R, Chaplin D. Deletion of C4A genes in patients with systemic lupus erythematosus. Arthritis Rheum. 1987;30:1015–22. doi: 10.1002/art.1780300908. [DOI] [PubMed] [Google Scholar]
- 22.Davies E, Steers G, Ollier W, Grennan D, Cooper R, Hay E, Hillarby M. Relative contributions of HLA-DQA and complement C4A loci in determining susceptibility to systemic lupus erythematosus. Br J Rheumatol. 1995;34:221–5. doi: 10.1093/rheumatology/34.3.221. [DOI] [PubMed] [Google Scholar]
- 23.Yamada H, Watanabe A, Mimori A, et al. Lack of gene deletion for complement C4A deficiency in Japanese patients with systemic lupus erythematosus. J Rheumatol. 1990;17:1054–7. [PubMed] [Google Scholar]
- 24.Goldstein R, Sengar D. Comparative studies of the major histocompatibility complex in French Canadian and non French Canadian Caucasians with systemic lupus erythematosus. Arthritis Rheum. 1993;36:1121–7. doi: 10.1002/art.1780360813. [DOI] [PubMed] [Google Scholar]
- 25.Reveille J, Moulds J, Ahn C, Friedman A, Baethge B, Roseman J, Straaton K, Alarcon G. Systemic lupus erythematosus in three ethnic groups. I. The effects of HLA Class II, C4, and CR1 alleles, socioeconomic factors, and ethnicity at disease onset. Arthritis Rheum. 1998;41:1161–72. doi: 10.1002/1529-0131(199807)41:7<1161::AID-ART4>3.0.CO;2-K. [DOI] [PubMed] [Google Scholar]
- 26.Sullivan K, Kim N, Goldman D, Petri M. C4A deficiency due to a 2bp insertion is increased in patients with systemic lupus erythematosus. J Rheumatol. 1999;26:2144–7. [PubMed] [Google Scholar]
- 27.Fan Q, Uring-Lambert B, Weill B, Menkes C, Delpech M. Complement component C4 deficiencies and gene alterations in patients with systemic lupus erythematosus. Eur J Immunogenet. 1993;20:11–21. doi: 10.1111/j.1744-313x.1993.tb00091.x. [DOI] [PubMed] [Google Scholar]
- 28.Truedsson L, Sturfelt G, Nived O. Prevalence of the type I complement C2 deficiency gene in Swedish systemic lupus erythematosus patients. Lupus. 1993;2:325–7. doi: 10.1177/096120339300200509. [DOI] [PubMed] [Google Scholar]
- 29.Sullivan K, Petri M, Schmeckpeper B, McLean R, Winkelstein J. Prevalence of a mutation causing C2 deficiency in systemic lupus erythematosus. J Rheumatol. 1994;21:1128–33. [PubMed] [Google Scholar]
- 30.Hartung K, Fontana A, Klar M, et al. Associations of class I, II and III MHC-gene products with systemic lupus erythematosus: results of a central European multicenter study. Rheumatol Int. 1989;9:13–18. doi: 10.1007/BF00270284. [DOI] [PubMed] [Google Scholar]
- 31.Walport M, Davies K, Botto M. C1q and systemic lupus erythematosus. Immunobiol. 1998;199:265–85. doi: 10.1016/S0171-2985(98)80032-6. [DOI] [PubMed] [Google Scholar]
- 32.Botto M, Dell'agnola C, Bygrave A, et al. Homozygous C1q deficiency causes glomerulonephritis associated with multiple apoptotic bodies. Nat Genet. 1998;19:56–59. doi: 10.1038/ng0598-56. [DOI] [PubMed] [Google Scholar]
- 33.Manger K, Repp R, Spriewald B, et al. Fcγ receptor IIa polymorphism in Caucasian patients with systemic lupus erythematosus: association with clinical symptoms. Arthritis Rheum. 1998;41:1181–9. doi: 10.1002/1529-0131(199807)41:7<1181::AID-ART6>3.0.CO;2-C. [DOI] [PubMed] [Google Scholar]
- 34.Koene H, Kleijer M, Swaag A, et al. The FCγRIIIA-158F allele is a risk factor for systemic lupus erythematosus. Arthritis Rheum. 1998;41:1813–8. doi: 10.1002/1529-0131(199810)41:10<1813::AID-ART13>3.0.CO;2-6. [DOI] [PubMed] [Google Scholar]
- 35.Moser K, Naes B, Salmon J, et al. Genome scan of human systemic lupus erythematosus: evidence for linkage on chromosome 1q in African–American pedigrees. Proc Natl Acad Sci USA. 1998;95:1469–74. doi: 10.1073/pnas.95.25.14869. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Hogarth M, Slingsby J, Allen P, et al. Multiple lupus susceptibility loci map to chromosome 1 in BXSB mice. J Immunol. 1998;161:2753–61. [PubMed] [Google Scholar]
- 37.Gaffner P, Kearns G, Shark K, et al. A genome-wide search for susceptibility genes in human systemic lupus erythematosus sib-pair families. Proc Natl Acad Sci USA. 1998;95:14875–9. doi: 10.1073/pnas.95.25.14875. [DOI] [PMC free article] [PubMed] [Google Scholar]