Immunoglobulin Deficiencies and Susceptibility to Infection Among ...

P1: FLT pp891-joci-467317

JOCI.cls

June 19, 2003

14:48

C 2003) Journal of Clinical Immunology, Vol. 23, No. 4, July 2003 (!

Immunoglobulin Deficiencies and Susceptibility to Infection Among Homozygotes and Heterozygotes for C2 Deficiency CHESTER A. ALPER,1,2,7 JIANHUA XU,1,2 KATHERINE COSMOPOULOS,1 BRIAN DOLINSKI,1 ROSANNE STEIN,1 GABRIEL UKO,1,2 CHARLES E. LARSEN,1,3 DEVENDRA P. DUBEY,1 PETER DENSEN,4 5 ¨ LENNART TRUEDSSON,5 GUNNAR STURFELT,6 and ANDERS G. SJOHOLM

genes on the extended major histocompatibility complex (MHC) haplotype [HLA-B18, S042, DR2] (but probably not on type II C2-deficient haplotypes) similar to those previously identified on [HLA-B8, SC01, DR3] and [HLA-B18, F1C30, DR3].

Accepted: March 14, 2003

About 25% of C2-deficient homozygotes have increased susceptibility to severe bacterial infections. C2-deficient homozygotes had significantly lower serum levels of IgG2, IgG4, IgD, and Factor B, significantly higher levels of IgA and IgG3 and levels of IgG1 and IgM similar to controls. Type I (28 bp deletion in C2 exon 6 on the [HLA-B18, S042, DR2] haplotype or its fragments) and type II (non-type I) C2-deficient patients with increased susceptibility to bacterial infection had significantly lower mean levels of IgG4 ( p < 0.04) and IgA ( p < 0.01) than those without infections (who had a higher than normal mean IgA level) but similar mean levels of other immunoglobulins and Factor B. Of 13 C2-deficient homozygotes with infections, 85% had IgG4 deficiency, compared with 64% of 25 without infections. IgD deficiency was equally extraordinarily common among infection-prone (50%) and noninfection-prone (70%) homozygous type I C2-deficient patients. IgD deficiency was also common (35%) among 31 type I C2-deficient heterozygotes (with normal or type II haplotypes), but was not found in 5 type II C2-deficient heterozygotes or 1 homozygote. Thus, C2 deficiency itself is associated with many abnormalities in serum immunoglobulin levels, some of which, such as in IgG4 and IgA, may contribute to increased susceptibility to infection. In contrast, IgD deficiency appears not to contribute to increased infections and appears to be a dominant trait determined by a gene or

KEY WORDS: C2 Deficiency; immunodeficiency diseases; infection; MHC.

INTRODUCTION

Homozygous C2 deficiency is the most common inherited complete deficiency of a complement protein in European Caucasians (1, 2), with an estimated prevalence of 1 per 10,000. The structural gene for C2 lies within the major histocompatibility complexes (MHC), between the gene for HLA-B and that for complement Factor B (3). Most C2-deficient persons are homozygous for a 28 bp deletion in the sixth exon of the C2 gene (4) that results in a complete lack of C2 biosynthesis. This C2-deficiency gene is referred to as type I and is carried by the conserved extended MHC haplotype [HLA-B18, S042, DR2] (5–7) or its complotype-containing fragments. A much less common heterogeneous group of genes determining C2 deficiency is associated with a variety of MHC haplotypes and is referred to as type II (8). The first reported C2-deficient persons were healthy immunologists (1). As more cases were identified, it became evident that C2-deficient patients have an increased incidence of systemic lupus-like disease (9). In addition, as many as 25% of C2-deficient homozygotes have increased susceptibility to severe infection with bacteria such as S. pneumoniae, N. meningitidis, or H. influenzae (10, 11). Because the classical complement pathway is important in the generation of antibody-mediated enhanced phagocytosis and bactericidal activity, C2 deficiency itself has been implicated as the basis for the observed increased susceptibility to infection. Moreover, levels of Factor B of

1 The

Center for Blood Research, Harvard Medical School, Boston, Massachusetts. of Pediatrics, Harvard Medical School, Boston, Massachusetts. 3 Department of Medicine, Havard Medical School, Boston, Massachusetts. 4 Department of Internal Medicine, University of Iowa College of Medicine and VAMC, Iowa City, Iowa. 5 Institute of Laboratory Medicine, Section of Microbiology, Immunology and Glycobiology, University of Lund, Lund, Sweden. 6 Department of Rheumatology, University of Lund, Lund, Sweden. 7 To whom correspondence should be addressed at The Center for Blood Research, 800 Huntington Avenue, Boston, Massachusetts 02115; e-mail: [email protected]. 2 Department

297 C 2003 Plenum Publishing Corporation 0271-9142/03/0700-0297/0 !


JOCI.cls

June 19, 2003

14:48

298

the alternative complement pathway may be decreased in individual C2-deficient patients’ serum (12), so that defective antibody-mediated or nonantibody-mediated activation by that pathway may also contribute to infections. Subtle defects in antibody production have been reported in homozygotes for C2, C3, or C4 deficiency (13). Furthermore, patients with unspecified deficiencies of complement components (14) or of components of the classical pathway, including 14 C2-deficient homozygotes (15), had lower than normal IgG4 levels. This suggests that impaired complement function might interfere with immunoglobulin production and thereby perhaps contribute to increased susceptiblity to infection. We showed previously (16) that homozygotes for the conserved extended MHC haplotype [HLA-B8, SC01, DR3] had increased prevalences of IgA and IgG4 deficiencies compared with either heterozygotes or noncarriers, whereas both homozygotes and heterozygotes had increased prevalences of IgG3 and IgD deficiencies. We interpreted these findings as evidence for recessive genes for IgA and IgG4 deficiency and dominant genes for IgG3 and IgD deficiency on the [HLA-B8, SC01, DR3] haplotype. Moreover, although Sardinian (17) or Basque (18) homozygotes for the extended haplotype [HLA-B18, F1C30, DR3] did not have IgA or IgG subclass deficiency, homozygotes and heterozygotes among the Basques had IgD deficiency (18). As part of a search for genes for immunoglobulin deficiencies on other extended MHC haplotypes, we evaluated serum immunoglobulin levels in a series of C2-deficient patients, some of whom were ascertained because of increased susceptibility to bacterial infection. We hoped to better understand the part played by C2 deficiency per se, as well as putative MHC susceptibility genes, in immunoglobulin regulation and clinical abnormalities. MATERIALS AND METHODS

Subjects Serum was obtained from blood drawn from 28 heterozygous and 38 homozygous C2-deficient patients of whom 13 had known increased susceptibility to infection (two or more episodes of bacterial pneumonia, meningitis or septicemia, or the like). None of the 28 heterozygous C2-deficient/normal subjects had such a history. There were nine homozygous (B1–B9, two with infections) and 21 heterozygous C2-deficient Boston (B10– B30) and 21 homozygous C2-deficient Lund (L1–21) patients (of whom five had increased susceptibility to infection). An additional eight homozygous (of whom six had increased susceptibility to bacterial infection) and

ALPER ET AL.

seven heterozygous C2-deficient patients were ascertained in Iowa City (IC). There were 11 sib pairs among the C2deficient homozygotes. In five of these, neither sib had increased susceptibility to severe infections, in two pairs both sibs had infections and in four, only one had infections. All subjects gave informed consent and protocols were approved by the relevant institutional review boards. C2 levels were determined by both immunochemical (19) and hemolytic (20) assays. For direct comparisons of the distributions of immunoglobulin and Factor B levels in patients with those in control subjects, we used 53 parents of patients who underwent bone marrow transplantation at the Dana-Farber Cancer Institute, Boston, MA (DFCI). Serum Immunoglobulin Concentrations Serum levels of IgG, IgA, IgM, and Factor B were determined by immunonephelometry (21) using goat polyclonal antisera (Diasorin, Stillwater, MN). IgG2, IgG3, IgG4, and IgD serum levels were determined by ELISA (22) using specific polyclonal goat antisera (Diasorin) or mouse monoclonal antibodies (Pierce, Rockford, IL) or kits (The Binding Site, San Diego, CA). IgG1 levels were derived from IgG levels by subtracting the sum of the concentrations of IgG2, IgG3, and IgG4. We used previously published (16) immunoglobulin class and subclass levels determined by radial immunodiffusion to define immunoglobulin deficiencies as less than the geometric means −2 SD (IgD < 0.14, IgG2 < 21.9, IgG3 < 20.1, IgG4 < 3.8 mg/dL). For the mean −3 SD, the values were IgD < 0.05, IgG3 < 11.8, and IgG4 < 1.6 mg/dL. IgA deficiency was taken as 0.2 in all cases) in C2-deficient compared with normal subjects. In contrast, IgA and IgG3 mean values were highly significantly elevated ( p = 0.0003 and p < 0.0001) in the C2-deficient subjects. Moreover, IgG2, IgG4, IgD, and Factor B geometric mean levels were all highly significantly decreased in C2-deficient patients compared with controls ( p < 0.002, p < 0.0001, p < 0.0001, and p < 0.0001). Of 38 homozygotes for C2 deficiency (Fig. 1), there was 1 IgG3-deficient subject (3%, compared with 4.8% of controls), 5 IgG2-deficient subjects (14%, compared with 2.6% in controls), but no IgA-deficient persons. Two of the IgG2-deficient subjects were sibs, the others were unrelated. Moreover, 27 C2-deficient homozygotes (71%) were IgG4 deficient (of whom 23 of 33, or 70%, were type I/I homozygotes) and 16 (42%) were IgD deficient, compared with rates of 3.2% and 6.6%, respectively, in our control population. Among 20 type I C2-deficient homozygotes, 12, or 60%, were IgD deficient. No subject was deficient in IgG, IgG1, IgM, or Factor B.

Journal of Clinical Immunology, Vol. 23, No. 4, 2003

Fig. 1. Immunoglobulin class, IgG subclass, and Factor B serum levels in homozygous C2-deficient subjects and controls. Geometric mean levels ± 1 SD are shown as are the significant p values for the comparison between them. Individual levels higher than the geometric mean + 3 SD are indicated by an asterisk and have been eliminated from the analysis. For IgG3, IgD, IgG2, IgG4, and Factor B, the cutoff for defining deficiency is shown by a dotted line. The cutoffs defining Ig deficiency were published previously (16). For IgD levels, the lower limit of detection is shown as a dashed line and samples with no detectable IgD are given the value “0.”

Serum Immunoglobulin and Factor B Levels in Homozygous C2-Deficient Patients with and Without Infection Figure 2 shows the distributions of serum levels of IgG, IgG2, IgG3, IgG4, IgA, IgM, IgD, and Factor B in C2-deficient homozygotes with increased susceptibility to infection compared with those without such susceptibility. The geometric mean levels of IgG, IgG1, IgG3, IgM, IgD, and Factor B in infection-prone compared with noninfection-prone C2-deficient patients did not differ ( p > 0.2). On the other hand, the geometric mean level of IgA was significantly lower (and the same as in


JOCI.cls

June 19, 2003

14:48

300

ALPER ET AL.

Serum Immunoglobulin Levels in Heterozygotes for Type I C2 Deficiency Table I gives serum immunoglobulin concentrations in 27 healthy type I C2 deficiency/C2 normal heterozygous individuals and four heterozygous type I/type II C2deficient subjects (who were, of course, homozygous C2 deficient). The frequency of IgG4 deficiency in 27 type I C2-deficient heterozygotes (excluding those patients also heterozygous for type II C2 deficiency) was 7%, a rate similar to the 3% observed in our previously published controls (16). The second MHC haplotype in one of the two IgG4-deficient patients was [HLA-B8, SC01, DR3] and in the other was [HLA-B44, DR8]. The frequency of IgD deficiency among all type I C2deficient heterozygotes (including the type I/II double heterozygotes) was 11 in 31, or 35%, compared with the 60% of type I/I homozygotes who had IgD deficiency, as noted earlier. The ratio of the prevalence of IgD deficiency in type I/I homozygotes to prevalence in type I heterozygotes was therefore 60/35 or 1.71. Thus, IgD deficiency, as well as IgG4 deficiency, was far more common among homozygous type I C2-deficient persons than in the control population and IgD deficiency (but not IgG4 deficiency) was also common in heterozygotes for type I C2 deficiency. Serum Immunoglobulin Levels in Types I/II and II/II C2 Deficiency Of 4 I/II, 1 II/II, and 1 II/nl subjects, none had IgD, IgA, or IgG3 deficiency, although all but the type II/II C2-deficient homozygote and the II/nl person were IgG4 deficient (Table II).

Fig. 2. Immunoglobulin class, IgG subclass, and Factor B serum levels in homozygous C2-deficient subjects with increased susceptibility to bacterial infection compared with those without such susceptibility. See legend of Fig. 1 for other features. The number of infected subjects vary between 12 (IgG, IgM, IgG2, and Factor B) and 13. Similarly, the number of noninfected subjects varied between 24 (IgG, IgM, and Factor B) and 25.

DISCUSSION

controls, p > 0.6) in infected compared with noninfected C2-deficient subjects ( p < 0.01). The geometric mean IgG4 level was significantly lower in subjects with infection ( p < 0.04) than in non-infection-prone C2-deficient patients. The mean IgG2 level was lower in infectionsusceptible than in C2-deficient subjects without infections, but the difference was not statistically significant ( p < 0.1). IgG4 deficiency occurred in 11 of 13 (85%) C2deficient homozygotes with frequent infections and in 16 of 25 (64%) in those without infections ( p not significant). IgD deficiency occurred in 5 of 13 (38%) homozygous C2deficient patients with infections compared with 11 of 25 (44%) of those without (no significant difference).

Significant abnormalities in the concentrations of most immunoglobulins were found in C2-deficient patients compared with normal controls. C2-deficient patients had significantly lower geometric mean levels of IgG2, IgG4, IgD, and Factor B and significantly elevated geometric mean levels of IgA and IgG3. No differences from normal were found in levels of total IgG, IgG1, and IgM. There was a remarkably high prevalence of deficiencies of IgG4 and IgD among homozygous C2-deficient patients. In contrast, heterozygotes had only a high prevalence of IgD deficiency. Bird and Lachmann (15), in their study of IgG subclass concentrations in 52 complement-deficient patients, including 14 with C2 deficiency, found a severely low geometric mean level of IgG4 in patients with C1q, C1r, C4, C2 (the classical pathway), or C3 deficiency, but



JOCI.cls

June 19, 2003

14:48

C2 DEFICIENCY AND INFECTION

301 Table I. Immunoglobulin Levels in Type I C2-Deficient Heterozygotes

Haplotype 1a

ID B10 B11 B12 B13 B14 B15 B16 B17 B18 B19 B20 B21 B22 B23 B24 B25 B26 B27 B28 B29 B30 IC3 IC4 IC6 IC7 IC8 IC13 B8c B9c IC15c IC16c

18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18,S042,14 18,S042,14 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, S042, 2 18, 2 18, 2 18, 2 18, 2 18, 2 18, 2 18, S042, 14 18, S042, 14 18, S042, 2 18, S042, 2

Haplotype 2

IgA (mg/dL)

IgD (mg/dL)

IgG3 (mg/dL)

IgG4 (mg/dL)

310 237 331 380 510 242 231 207 183 311 176 276 350 456 370 339 348 215 283 278 1104 737 521 123 184 1115 969 92 247 217 204

0.04b

57 70 46 102 77 323 119 59 92 92 96 57 217 54 123 145 195 50 169 167 131 368 181 263 255 386 343 35 35 134 134

9.9 73.9 41.2 20.1 27.4 54.4 46.9 40.8 2.9 93.1 46.1 7.5 11.5 17.7 6.4 14.4 11.0 80.1 75.8 17.0 20.0 52.9 36.7 14.0 0.0 15.5 17.9 0.4 0.5 0.3 0.5

18, F1C30, 4 35, FC33, 10 51, SC31, X 51, SC30, 9 18, SC31, 5 58, FC31, 8 58, FC31, 8 62, SC33, 4 8, SC01, 3 14, SC21, 1 35, FC3.20, 3 37, SC02, 13 60, SC02, 13 X, SC30, 1 18, SC31, 5 7, SC31, 2 27, SC31, 2 7, SC31, 2 8, SC01, 3 58, FC31, 6 8, SC01, 3 7, 2 5, 2 7, 2 44, 8 5, 5 18, 8 35, F030, 4 35, F030, 4 5, S031, 4 5, S031, 4

0.03 0.28 0.04 0.57 2.02 0.82 0.08 0.00 0.10 0.15 0.39 0.16 0.00 0.00 1.60 1.15 0.78 0.00 0.10 4.84 0.51 0.49 0.57 5.33 0.00 1.21 2.16 3.47 8.39 1.11

haplotypes given as HLA-B, complotype, HLA-DR alleles. Underlined haplotypes carry C2∗ Q0 (C2 deficiency gene). b Italics: ≤normal geometric mean −2 SD; Bold italics: ≤normal geometric mean −3 SD. c Heterozygous type I, but homozygous C2 deficient (I/II).

a MHC

normal levels in patients with C5-9 deficiency, properdin deficiency or hereditary angioedema. Moreover, 50% of their C2-deficient patients were IgG4 deficient by our definition. Thus, our present findings confirm theirs with respect to IgG4 and C2 deficiency in general. Because the low IgG4 levels were found in patients with a variety of

classical complement component deficiencies, many not encoded on chromosome 6p (as are deficiencies of C4 or C2), Bird and Lachmann concluded that classical component deficiency per se led to IgG4 deficiency. These authors did not study IgD, IgA, IgM, or Factor B levels and did not provide HLA haplotypes or clinical data for

Table II. Immunoglobulin Levels in Persons Heterozygous or Homozygous for Type II C2 Deficiency ID

Type

Haplotype 1a

Homozygous C2 deficient B8 I/II 18, S042, 14 B9 I/II 18, S042, 14 IC12 II/II 5, S031, 4 IC15 I/II 18, S042, 2 IC16 I/II 18, S042, 2 Heterozygous C2 deficient IC14 II/nl 5, 4

Haplotype 2

IgA (mg/dL)

IgD (mg/dL)

IgG3 (mg/dL)

IgG4 (mg/dL)

35, F030, 4 35, F030, 4 35, S031, 1 5, S031, 4 5, S031, 4

92 247 1091 217 204

2.16 3.47 0.62 8.39 1.11

35 35 337 134 134

0.4b 0.5 6.9 0.3 0.5

18, 8

933

5.70

275

21.7

a MHC haplotypes given as HLA-B, complotype, HLA-DR alleles. Underlined haplotypes carry C2∗ Q0 (C2 deficiency gene). b Italics: ≤normal geometric mean −2 SD; Bold italics: ≤normal geometric mean −3 SD.



JOCI.cls

June 19, 2003

14:48

302

their patients. Although there was a higher geometric mean level of IgG3 and a lower mean level of IgG2 in their early component-deficient subjects compared with controls, the differences were not statistically significant. Their failure to find significant differences (as we did) may have been due to a heterogeneous patient population or their smaller number of C2-deficient subjects or both. Because of the increased susceptibility to infection of some C2-deficient homozygotes (10, 11), it is possible that one or the other of these immunoglobulin deficiencies contributes to this enhanced susceptibility. Low concentrations of IgG2 (32, 33) and of IgG4 (34, 35) have been reported to be associated with increased susceptibility to infection. There have also been individual case reports of C2-deficient infection-prone individuals who also had immunoglobulin deficiencies (36–38). There were no increased infections or immunoglobulin abnormalities (except for one subject with IgD deficiency) in nine C2-deficient heterozygotes and one normal person from families with at least one infection-prone C2-deficient homozygote. In our present study, there was a significantly higher frequency of IgG4 deficiency and a significantly lower geometric mean serum level of IgG4 ( p < 0.04) among infection-prone C2-deficient patients compared to the C2-deficient patients without increased susceptibility to infection. The mean serum level of IgG2 was somewhat lower and the frequency of IgG2 deficiency among infection-prone patients was somewhat higher than that in infection-free C2-deficient patients, but the differences were not significant ( p < 0.1). Thus, our results suggest that IgG4 deficiency, and possibly also IgG2 deficiency, may contribute to increased susceptibility to infection in C2-deficient patients. Although C2-deficient patients with increased susceptibility to infection had a lower geometric mean level of IgA than C2-deficient patients without infections, it was the same as that in normals. In other words, the uninfected subjects accounted for the higher level of IgA in C2-deficient patients overall. Perhaps the higher IgA levels in subjects without infections were protective against infection, whereas the lower levels of IgG2 and particularly of IgG4 contributed to susceptibility to infection. There were no differences in the concentrations of IgG1, IgG3, IgM, IgD, or Factor B in C2-deficient patients with or without increased susceptibility to infection, despite the remarkably high incidence of IgD deficiency, the higher mean level of IgG3 and the significantly lower mean level of Factor B in C2-deficient subjects in general. IgD deficiency is the most common of the immunoglobulin deficiencies among normal Caucasian individuals, with a frequency of 6–8% (39). In our population of previously studied normal subjects, 6.6% were IgD deficient

ALPER ET AL.

as defined by −2 SD and 6.1% by −3 SD (16, 18). The corresponding frequencies for IgG4 deficiency were 3.2 and 2.7%. IgD is primarily a surface membrane protein on B cells. The roles of both serum and membrane-bound IgD are uncertain. It is clear that the classical complement pathway has a major role in B-cell development and differentiation (40– 43). Perhaps IgG4 and IgD deficiencies in C2 deficiency reflect a specific requirement, either directly or indirectly, for switching and/or differentiation of B cells and their plasma cell progeny secreting these immunoglobulins. This argument was advanced by Bird and Lachmann (15). Individuals homozygous for deficiencies of C2, C4 or C3 have defective antibody production to the bacteriophage ϕ × 174 (13). Furthermore, C2 seems to be required for the normal development of antibody responses to antigens that activate the mannose-binding lectin pathway (44) and might promote antibody responses to antigens that activate the alternative pathway. Mice rendered homozygous deficient by gene knockout for the complement C3 receptor CD21/CD35 have a failure in normal lymph node germinal center development as well as in B-cell differentiation (42, 43). Fairly little is known about immune responses to polysaccharides in patients with deficiencies of classical pathway components. Recently, bactericidal antibody responses by C2-deficient patients to immunization with N. meningitidis and H. influenzae type b vaccines were described, emphasizing the protective potential of such antibodies in the absence of a functional classical complement pathway (44). Low concentrations of anticapsular antibody to S. pneumoniae have been found in patients with C3 deficiency (45), but not C2 deficiency (45, 46). Development of protective antibody levels may be delayed in classical pathway deficiencies, but this remains to be shown. Our present finding of high IgA levels in C2-deficient patients without infection might possibly relate to the presence of opsonic IgA antibodies to capsular polysaccharides (47). An alternative mechanism by which C2 deficiency could lead to immunoglobulin deficiency is via distinct MHC immunoglobulin deficiency genes closely linked to and in linkage disequilibrium with C2 deficiency genes. This kind of mechanism appears to occur in IgA deficiency, in which four different conserved, extended haplotypes, including [HLA-B8, SC01, DR3], are increased in frequency among patients (48). In prospective studies of persons homozygous, heterozygous, or noncarrying for [HLA-B8, SC01, DR3] (17), we found IgA deficiency in 13% of homozygotes, but almost none of the heterozygotes or noncarriers. Similarly, IgG4 deficiency was found in 30% of homozygotes, but rarely in heterozygotes. This suggested that IgA and IgG4 deficiencies are



JOCI.cls

June 19, 2003

14:48


recessive MHC-determined traits. IgG4 deficiency would be an apparently recessively determined MHC trait either whether determined by a gene or genes on the [HLA-B18, S042, DR2] haplotype or as a result of C2 deficiency itself (since only completely C2-deficient patients have an increased prevalence of IgG4 deficiency). The fact that one of the two type I C2-deficient heterozygotes with IgG4 deficiency carried [HLA-B8, SC01, DR3] suggests that there is a recessive gene for IgG4 deficiency on the [HLA-B18, S042, DR2] haplotype. It is conceivable, but not likely, that homozygous C4A deficiency contributed to IgG4 deficiency in the [HLA-B8, SC01, DR3] homozygotes. In the present study, we found that the homozygous C2-deficient patients who were type I/type II double heterozygotes also had a high (100%) frequency of IgG4 deficiency. All of these IgG4-deficient subjects had normally expressed C4A genes on both haplotypes. As for genes in linkage disequilibrium with C2 null genes, there were three different MHC haplotypes carrying type II C2 deficiency associated with IgG4 deficiency in these patients. If the MHC were solely responsible for IgG4 deficiency in these subjects, one would thus have to assume that at least two of the three different type II C2 deficiency MHC haplotypes carried specific susceptibility genes for IgG4 deficiency. This is unlikely since type II C2 deficency is rare and IgG4 deficiency is common among patients with non-MHC-encoded deficiencies for C1q, C1r/s, and C3. The lack of IgG4 deficiency in the single type II/II homozygous C2-deficient patient could well be due to chance since not all homozygous C2deficient subjects have IgG4 deficiency. The presence of IgG4 deficiency in two type I C2-deficient/C2 normal heterozygotes could reflect the presence of IgG4 deficiency susceptibility genes on both [HLA-B18, S042, DR2] and [HLA-B8, SC01, DR3], as well as on the random MHC haplotype HLA-B44, DR8 in the second patient. The frequency of IgG4 deficiency in C2-deficient subjects in our present study (71%) was considerably higher than the 30% of our previous study (16) of [HLA-B8, SC01, DR3] homozygotes, which again suggests a different mechanism (or, conceivably, different penetrance rates of different alleles). It is conceivable that IgG4 deficiency results from an additive interplay between MHC genes and C2 deficiency itself but the latter clearly contributes in a major way. The considerations with respect to IgD deficiency are different. The putative susceptibility genes for IgD deficiency on the [HLA-B8, SC01, DR3] (16) and [HLA-B18, F1C30, DR3] haplotypes (18) act dominantly with an apparent penetrance of 37% in homozygotes and around 20% in heterozygotes for each of these haplotypes. The apparent penetrance for IgD deficiency in relation to the pres-


303

ence of the [HLA-B18, S042, DR2] haplotype is 60% in homozygotes and 35% in heterozygotes. Apparent penetrance for IgD deficiency is thus considerably higher in both homozygotes and heterozygotes for type I C2 deficiency than for [HLA-B8, SC01, DR3] or [HLA-B18, F1C30, DR3]. Nevertheless, as was observed in the two earlier studies, IgD deficiency had a higher frequency in homozygotes than in heterozygotes. The higher frequency of IgD deficiency in type I C2-deficient homozygotes could reflect the combined effects of a dominantly expressed MHC susceptibility gene on [HLA-B18, S042, DR2] as well as of C2 deficiency per se. It is doubtful that the 35% prevalence of IgD deficiency among type I heterozygotes is contributed to by half-normal C2 levels, since heterozygotes are presumed to have adequate C2 levels for facilitation of immune functions. The differences in frequencies of IgD deficiency between homozygotes and heterozygotes for [HLA-B18, S042, DR2] and those among homozygotes and heterozygotes for other extended haplotypes may reflect differences in penetrance of the IgD deficiency susceptibility genes rather than C2 deficiency itself. Assuming that penetrance reflects a stochastic process affecting a dominant MHC susceptibility gene (31), the prevalence of IgD deficiency in type I C2-deficient heterozygotes of 35% predicts a prevalence in homozygotes of 58% and a ratio of IgD deficiency in homozygotes to heterozygotes of 1.65. These values are remarkably close to those observed (60% and 1.71, respectively), strengthening the argument for a distinct MHC susceptibility gene. The occurrence among the heterozygotes for [HLA-B18, S042, DR2] of IgD deficiency may have been artificially high because of the presence in three heterozygotes of the [HLA-B8, SC01, DR3] haplotype (two of whom had a low IgD level) and of (HLA-B18, F1C30), DR4, (a fragment of [HLA-B18, F1C30, DR3], also known to be associated with IgD deficiency). Since IgD deficiency had the same frequency in C2-deficient homozygotes with or without infections, it is unlikely that it contributed to this susceptibility.

ACKNOWLEDGMENTS

This work was supported by grants nos. HL-29583 (CAA) and AI-14157 (CAA) from the National Heart, Lung, and Blood Institute and the Allergy and Infectious Diseases Institute of the National Institutes of Health, by a Merit Review Award from the Department of Veterans Affairs (PD), and by grants from the Swedish Medical Council, the Swedish National Health Association against Rheumatism, and the King Gustaf V 80th Birthday Fund (AGS, GS, and LT).


JOCI.cls

June 19, 2003

14:48

304 REFERENCES 1. Klemperer MR, Woodworth HC, Rosen FS, Austen KF: Hereditary deficiency of the second component of complement (C& 2) in man. J Clin Invest 45:880–890, 1966 2. Schneider PM, Rittner C: Complement genetics. In Complement— A Practical Approach. A Dodds, RB Sim (eds). Oxford, UK, Oxford University Press, 1997, pp 165–198 3. Carroll MR, Campbell D, Bentley DR, Porter RR: A molecular map of the human major histocompatibility complex class III region linking complement genes C4, C2 and factor B. Nature (Lond) 307:237– 241, 1984 4. Johnson CA, Densen P, Hurford RK Jr, Colten HR, Wetsel RA: Type I human complement C2 deficiency. A 28-base pair gene deletion causes skipping of exon 6 during RNA splicing. J Biol Chem 267:9347–9353, 1992 5. Fu SM, Stern R, Kunkel HG, Dupont B, Hansen JA, Day NK, Good RA: Evidence for linkage between HL-A histocompatibility genes and those involved in the synthesis of the second component of complement. J Exp Med 140:1108–1111, 1975 6. Awdeh ZL, Raum DD, Glass D, Agnello V, Schur PH, Johnston RB Jr, Gelfand EW, Ballow M, Yunis E, Alper CA: Complementhuman histocompatibility antigens in C2 deficiency. J Clin Invest 67:581–583, 1981 7. Truedsson L, Alper CA, Awdeh ZL, Johansen P, Sjöholm AG, Sturfelt G: Characterization of type I complement C2 deficiency haplotypes. Strong conservation of the complotype/HLA-B-region and absence of disease association due to linked class II genes. J Immunol 151:5856–5863, 1993 8. Wetsel RA, Kulics J, Lokki M-L, Kiepiela P, Akama H, Johnson CAC, Densen P, Colten HR: Type II human complement C2 deficiency. Allele-specific amino acid substitutions (Ser189 → Phe; Gly444 → Arg) cause impaired C2 secretion. J Biol Chem 271:5824– 5831, 1996 9. Agnello V, deBracco MME, Kunkel HG: Hereditary C2 deficiency with some manifestations of systemic lupus erythematosus. J Immunol 108:837–840, 1972 10. Densen P: Complement deficiencies and meningococcal disease. Clin Exp Immunol 86(Suppl 1):57–62, 1991 11. DeWitt CC, Scher DP, Winkelstein J: Group B streptococcal disease in a child beyond early infancy with a deficiency of the second component of complement (C2). Pediatr Infect Dis J 18:77–78, 1999 12. Truedsson L, Gullstrand B, Jönsson T, Klint C: Serum concentration of C4 isotypes and factor B in Type I C2 deficiency suggest haplotype-dependent quantitative expression of MHC class III complement genes. Exp Clin Immunogenet 12:66–73, 1995 13. Ochs HD, Wedgwood RJ, Heller SR, Beatty PG: Complement, membrane glycoproteins, and complement receptors: their role in regulation of the immune response. Clin Immunol Immunopathol 40:94– 104, 1986 14. Wedgwood RJ, Ochs HD, Oxelius V-A: IgG subclass levels in the serum of patients with primary immunodeficiencies. Monogr Allergy 20:80–89, 1986 15. Bird P, Lachmann PJ: The regulation of IgG subclass production in man: low serum IgG4 in inherited deficiencies of the classical pathway of C3 activation. Eur J Immunol 18:1217–1222, 1988 16. Alper CA, Marcus-Bagley D, Awdeh Z, Kruskall MS, Eisenbarth GS, Brink SJ, Katz A, Stein R, Bing DH, Yunis EJ, Schur PH: Prospective analysis suggests susceptibility genes for deficiencies of IgA and several other immunoglobulins on the [HLA-B8, SC01, DR3] conserved extended haplotype. Tissue Antigens 56:207–216, 2000

ALPER ET AL.

17. Cucca F, Zhu Z-B, Khanna A, Cossu F, Congia M, Badiali M, Lampis R, Frau F, De Virgiliis S, Cao A, Arnone M, Piras P, Campbell RD, Cooper MD, Volanakis JE, Powis SH: Evaluation of IgA deficiency in Sardinians indicates a susceptibility gene is encoded within the HLA class III region. Clin Exp Immunol 111:76–80, 1998 18. Calvo B, Castaño L, Marcus-Bagley D, Fici DA, Awdeh Z, Alper CA: The [HLA-B18, F1C30, DR3] conserved extended haplotype carries a susceptibility gene for IgD deficiency. J Clin Immunol 20:216–220, 2000 19. Laurell C-B: Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Anal Biochem 15:45–52, 1966 20. Austen KF, Beer F: The measurement of second component of human complement by its interaction with EAC’1agp , 4gp cells. J Immunol 92:946–952, 1964 21. Ritchie RF, Alper CA, Graves J, Pearson N, Larson C: Automated quantitation of proteins in serum and other biologic fluids. Am J Clin Pathol 59:151–159, 1973 22. Engvall E, Perlmann P: Enzyme-linked immunosorbent assay, ELISA. III. Quantitation of specific antibodies by enzyme-labeled anti-immunoglobulin in antigen-coated tubes. J Immunol 109:129– 135, 1972 23. Raum D, Awdeh Z, Yunis EJ, Alper CA, Gabbay KH: Extended major histocompatibility complex haplotypes in type 1 diabetes mellitus. J Clin Invest 74:449–454, 1984 24. Marcus-Bagley D, Alper CA: Methods for allotyping complement proteins. In Manual of Clinical Laboratory Immunology, 4th edn. NR Rose, E Conway de Macario, JL Fahey, H Friedman, GM Penn (eds). Washington, DC, American Society for Microbiology, 1992, pp 124–141 25. Alper CA, Raum D, Karp S, Awdeh ZL, Yunis EJ: Serum complement ‘supergenes’ of the major histocompatibility complex in man (complotypes). Vox Sang 45:62–67, 1983 26. Zachary AA, Teresi GA: ASHI Laboratory Manual, 2nd Edn. New York, American Society for Histocompatibility and Immunogenetics, 1990, p 195 27. Ray JG Jr, Hare DB, Pedersen PD, Mulally DI: NIAID Manual of Tissue Typing Techniques 1976–1977. Washington, DC, Government Printing Office, (DHEW publication no. NIH 76–545), 1976 28. Tiercy J-M, Gorski J, Jeannet M, Mach B: Oligonucleotide typing analysis of the polymorphism of DRB1 and DRB3 genes within DRw52 haplotypes. In Immunobiology of HLA. B Dupont (ed). New York, Springer, 1989, pp 248–250 29. Zar JH: In Biostatistical Analysis, 2nd edn. New Jersey, PrenticeHall, 1984, p 238 30. Pagano M, Gauvreau K: In Principles of Biostatistics. Belmont, CA, Wadsworth, 1992, p 339 31. Alper CA, Awdeh Z: Incomplete penetrance of MHC susceptibility genes: Prospective analysis of polygenic MHC-determined traits. Tissue Antigens 56:199–206, 2000 32. Umetsu DT, Ambrosino DM, Quinti I, Siber GR, Geha RS: Recurrent sinopulmonary infection and impaired antibody response to bacterial capsular polysaccharide antigen in children with selective IgG-subclass deficiency. N Engl J Med 313:1247–1251, 1985 33. Quinti I, Paretti C, Testi R, Bonomo R, Aiuti F: IgG subclass deficiency in adults: A clinical and immunological study. Monogr Allergy 20:143–148, 1986 34. Heiner DC, Lee SI, Short JA: IgG4 subclass deficiency syndromes. Monogr Allergy 20:149–156, 1986 35. Moss RB, Carmack MA, Esrig S: Deficiency of IgG4 in children: association of isolated IgG4 deficiency with recurrent respiratory infection. J Pediatr 120:16–21, 1992



JOCI.cls

June 19, 2003

14:48


36. Seligmann M, Brouet J-C, Sasportes M: Hereditary C2 deficiency associated with common variable immunodeficiency. Ann Intern Med 91:216–217, 1979 37. Sjöholm AG, Hallberg T, Oxelius V-A, Hammarström L, Smith CIE, Lindgren F: C2 deficiency, moderately low IgG2 concentrations and lack of the G2m(23) allotype marker in a child with repeated bacterial infections. Acta Paediatr Scand 76:533–538, 1987 38. Zimmerli W, Schaffner A, Schneidegger C, Scherz R, Späth PJ: Humoral immune response to pneumococcal antigen 23-F in an asplenic patient with recurrent fulminant pneumococcemia. J Infect 22:59–69, 1991 39. Fraser PA, Schur PH: Hypoimmunoglobulin D: Frequency, family studies, and association with HLA. Clin Immunol Immunopathol 19:67–74, 1981 40. Pepys MB: Role of complement in the induction of immunological responses. Transpl Rev 32:93–120, 1976 41. Rajewsky K, Forster I, Cumano A: Evolutionary and somatic selection of the antibody repertoire in the mouse. Science 238:1088–1094, 1987 42. Carroll MC: The role of complement and complement receptors in induction and regulation of immunity. Annu Rev Immunol 16:545– 568, 1998


305

43. Carroll MC: The role of complement in B cell activation and tolerance. Adv Immunol 74:61–88, 2000 44. Selander B, Käyhty H, Wedege E, Holmström E, Truedsson L, Söderström C, Sjöholm AG: Vaccination responses to capsular polysaccharides of Neisseria meningitidis and Haemophilus influenzae type b in two C2-deficient sisters: alternative pathway-mediated bacterial killing, and evidence for a novel type of blocking IgG. J Clin Immunol 20:138–149, 2000 45. Hazlewood MA, Kumaratne DS, Webster ADB, Goodall M, Bird P, Daha M: An association between homozygous C3 deficiency and low levels of anti-pneumococcal capsular polysaccharide antibodies. Clin Exp Immunol 87:404–409, 1992 46. Selander B, Weintraub A, Holmström E, Sturfelt G, Truedsson L, M˚artensson U, Jensenius JC, Sjöholm AG: Low concentrations of immunoglobulin G antibodies to Salmonella serogroup C in C2 deficiency: Suggestion of a mannan-binding lectin pathway-dependent mechanism. Scand J Immunol 50:555–561, 1999 47. Janoff EN, Fasching C, Orenstein JM, Rubins JB, Opstad NL, Dalmasso AP: Killing of Streptococcus pneumoniae by capsular polysaccharide-specific polymeric IgA, complement, and phagocytes. J Clin Invest 104:1139–1147, 1999 48. Burrows PD, Cooper MD: IgA deficiency. Adv Immunol 65:245– 276, 1977