Complement Receptor 1 Gene Polymorphisms in Sarcoidosis

Complement Receptor 1 Gene Polymorphisms in Sarcoidosis Michele Zorzetto, Cristina Bombieri, Ilaria Ferrarotti, Sara Medaglia, Carlo Agostini, Carmine Tinelli, Giovanni Malerba, Natalia Carrabino, Anna Beretta, Lucio Casali, Ernesto Pozzi, Pier Franco Pignatti, Gianpietro Semenzato, Maria Clara Cuccia, and Maurizio Luisetti Laboratorio di Biochimica e Genetica, Clinica di Malattie dell’Apparato Respiratorio, and Servizio di Biometria ed Epidemiologia Clinica IRCCS Policlinico San Matteo, Pavia; Dipartimento di Genetica e Microbiologia, Università di Pavia, Pavia; Immunologia Clinica, Dipartimento di Medicina Clinica e Sperimentale, Università di Padova, Padova; Cattedra di Malattie dell’Apparato Respiratorio, Università di Perugia; and Sezione di Biologia e Genetica, Dipartimento Materno Infantile e di Biologia-Genetica, Università di Verona, Verona, Italy Sarcoidosis is likely to result from exposure of genetically susceptible hosts to environmental agents. Erythrocyte (E) complement receptor 1 (CR1) is a membrane protein mediating the transport of immune complexes (ICs) to phagocytes, and at least three polymorphisms on the CR1 gene are related to erythrocyte surface density of CR1 molecules, in turn related to the rate of IC clearance from circulation. We hypothesized that sarcoidosis could be associated with increased frequency of the CR1 gene alleles coding for reduced CR1/E ratio. We studied 91 sarcoid patients and two control groups: 94 healthy volunteers and 71 patients with chronic obstructive pulmonary disease (COPD). Three polymorphic sites of CR1 gene, His1208Arg, intron 27 HindIII/RFLP, and Pro1827Arg, were analyzed. The three polymorphisms were in linkage disequilibrium. The GG genotype for the Pro1827Arg (C5507G) polymorphism was significantly associated with sarcoidosis in comparison to both control groups (odds ratio [OR] 3.13; 95% confidence interval [CI] 1.49–6.69 versus healthy control subjects, and OR 2.82, 95% CI 1.27–6.39 versus COPD control subjects). The same genotype was particularly associated to disease in females (OR 7.05; 95% CI 3.10–16.61 versus healthy control subjects). These findings agree with speculations on the role of CR1 gene as a possible susceptibility factor.

Sarcoidosis is a multisystem disorder characterized by T cell and mononuclear phagocyte infiltration of affected organs, granuloma formation, and distortion of the normal microarchitecture (1, 2). Although the cause of sarcoidosis remains unknown, there is evidence suggesting that the disorder results from exposure of genetically susceptible hosts to specific environmental agents (3, 4). Racial variation in the incidence and familial clustering are two robust lines of evidence strongly supporting the role of genetic susceptibility to sarcoidosis (2), and much effort has been spent in the last few years in searching for susceptibility genes (5, 6). Several studies, including a recent one on linkage to microsatellites in 122 siblings from 55 German nuclear families (7), have focused on the Major Histocompatibility Complex (MHC) genes as strong candidates for inherited

(Received in original form December 21, 2001 and in revised form March 11, 2002) Address correspondence to: Dr. Maurizio Luisetti, Laboratorio di Biochimica e Genetica, Clinica di Malattie dell’Apparato Respiratorio, IRCCS Policlinico San Matteo, Via Taramelli 5, 27100 Pavia, Italy. E-mail: [email protected] Abbreviations: base pairs, bp; cystic fibrosis transmembrane (conductance) regulator, CFTR; chronic obstructive pulmonary disease, COPD; complement receptor 1, CR1; erythrocyte, E; immune complexes, ICs; long homologous repeat, LHR; major histocompatibility complex, MHC; nucleotide, nt; polymerase chain reaction, PCR; restriction fragment length polymorphism, RFLP; short consensus repeat, SCR. Am. J. Respir. Cell Mol. Biol. Vol. 27, pp. 17–23, 2002 Internet address: www.atsjournals.org

susceptibility. Nevertheless, depending on the ethnic group studied, other genes have been reported to be associated with sarcoidosis, including ACE, cytokine/chemokine, vitamin D receptor, and cystic fibrosis transmembrane regulator (CFTR) (5, 6), suggesting that several genes may be acting together to result in sarcoidosis susceptibility. It is a long-standing notion that circulating immune complexes (ICs) are associated with sarcoidosis, especially in an acute form with erythema nodosum (8), and, although direct evidence is lacking, the hypothesis that ICs may play a role in the triggering or development of granuloma formation in sarcoidosis cannot be ruled out a priori (9). Under this viewpoint, it is possible to postulate that the polyclonal hypergammaglobulinemia observed in sarcoidosis may produce a combination of immunoglobulins and antigen(s) to form ICs (10) which, in turn, might contribute to granuloma formation (11). Complement receptor 1 (CR1; CD35; C3b/C4b receptor) is a membrane protein found in different cells, including erythrocytes, phagocytes, all B and some T cells, and dendritic cells (12), and is involved in different activities of the complement system (13). In particular, erythrocyte CR1 mediates the transport of ICs throughout the bloodstream to phagocytes in the liver and spleen, and it has been suggested that the rate of IC clearance from the circulation can be directly correlated to the number of CR1 molecules expressed on erythrocytes (CR1/E ratio) (14). Proteolytic cleavage may be responsible for loss of CR1 on the cell surface (15), but quantitative expression of CR1 on erythrocytes is also genetically determined (16). The CR1 structure is characterized by short consensus repeats (SCRs) of 65–70 amino acids with a conserved core of 11–14 amino acids. In the extracellular domain 7 SCRs constitute a long homologous repeat (LHR). Four LHRs (LHR-A, -B, -C, and -D) are followed by two additional SCRs, a single transmembrane domain, and a COOHterminal cytoplasmic domain of 43 amino acids. Three additional structural allotypes of CR1 have been found with three, five, or six LHRs, respectively (16). Several single nucleotide (nt) polymorphisms have been found on the CR1 gene, located on chromosome 1 (17). In this paper we focused on three polymorphisms: His1208Arg, intron 27 HindIII RFLP, and Pro1827Arg. His1208Arg results from an A→G substitution occurring in exon 22 at nt 3,650. The intron 27 HindIII RFLP gives rise to two alleles: H allele, associated with allotypes containing three, four, or five LHRs, and L allele, reported only in association with the allotype containing four LHRs (16). Pro1827Arg results from a C→G substitution occurring in exon 33 at nt 5,507 .

18

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 27 2002

The Pro1827Arg C allele and the His1208Arg A allele have been reported to be associated with the intron 27 HindIII RFLP H allele, whereas the Pro1827Arg G allele and the His1208Arg G allele have been associated with the intron 27 HindIII RFLP L allele in Caucasians (16, 17). The alleles are correlated respectively with either high (H) or low (L) CR1/E ratio (16, 18). Erythrocytes from individuals homozygous for the H allele may show as much as a 10-fold higher CR1/E ratio than those from individuals homozygous for the L allele; heterozygous individuals have intermediate CR1 expression (17–19). Because impaired clearance of ICs could result in increased tissue damage in a variety of inflammatory and infectious conditions (16), we hypothesized that sarcoidosis, an immune-mediated condition with a possible infectious trigger and occurring in some cases with circulating ICs, could be characterized by increased frequency of CR1 gene alleles associated with low CR1/E ratio. In this paper we provide evidence for an association between sarcoidosis and polymorphisms of the CR1 gene, thus suggesting that, at least in a subset of sarcoid patients, a reduced clearance of antigen(s) or ICs may occur. We speculate whether such finding would further support the concept that sarcoidosis is triggered by micro-organisms. In addition, this is the first report of a sex difference in sarcoidosis.

X-ray stage I to II, and from stage II to stage III], outcome [good spontaneous or prednisone-induced resolution of chest X-ray abnormalities; poor persistence or worsening of chest X-ray abnormalities], and relapses of the disease) were determined in patients with an available follow-up 1 yr: this subset included 84 out of 91 patients, in which the mean follow-up length was 91 55 mo (range 12–240). As controls, we studied two groups of ethnically and geographically matched individuals. The first control group consisted of 94 healthy volunteers (42 males and 52 females, mean age 42 15 yr) recruited from the clinical staff. Medical examination and a questionnaire, blood and urine chemistry, chest X-ray, and pulmonary function tests, performed during periodic medical surveys, excluded any disease. The second control group consisted of patients with a homogeneous lung disorder not related to interstitial lung disease. To this aim, we enrolled 71 unrelated patients with COPD (53 males and 18 females, mean age 65 13 yr), a disorder in which ICs are not likely to play a pathogenetic role (21). The diagnosis of COPD was made according to ATS standards (22) and subjects were enrolled based on irreversible airflow obstruction (FEV1 70%, 12% reversible). Sixty patients were smokers (42 former smokers and 19 current smokers), and 10 had never smoked. Their mean FEV 1 (% predicted) was 45.5 27.8 and the FEV1/FVC was 63.2 17.8. All individuals gave their consent before entering the study, which was approved by the Ethical Committees of the Institutions involved.

Materials and Methods

Analysis of Polymorphic Sites of CR1 Gene

Study Design

The three polymorphisms of CR1 gene were studied by restriction analysis. Regions encompassing each polymorphism were amplified from genomic DNA, according to Cornillet and coworkers (23), and to Xiang and colleagues (17). Primers, polymerase chain reaction (PCR) conditions, and restriction enzymes (all from New England Bio Labs, Beverly, MA) are summarized in Table 2. Amplimers were derived from published sequences (Gene Bank accession number: Y00816). Briefly, amplimers obtained were 682 bp for exon 22, 1,600 bp for intron 27, and 305 bp for exon 33. Amplimers of exon 22 were digested with RsaI and two fragments of 520 and 162 bp for A allele and three fragments of 458, 162, and 62 bp for G allele were obtained. Amplimers of intron 27 were digested with HindIII and one fragment of 1,600 bp for H allele, and two fragments of 1,100 and 500 bp for L allele were obtained. Amplimers of exon 33 were digested with MnlI and six fragments of 111, 80, 37, 33, 33, and 11 bp for C allele and five fragments 111, 80, 70, 33, and 11 bp for G allele were obtained. Amplification was performed using a Gene Amp PCR system 2,400 (Perkin-Elmer, Norwalk, CT). Conditions: 94C for 60 s, then 35 cycles of 94C for 15 s, annealing (63C for His1208Arg, 60C for

The investigation was designed as a case-control association study with a candidate gene, a powerful approach for finding genetic determinants of a complex disorder such as sarcoidosis (20).

Patients A total of 256 subjects, all Caucasians of Italian descent, were investigated. They were divided into three groups: patients with sarcoidosis (test group) and two control groups (healthy control subjects and patients with chronic obstructive pulmonary disease [COPD]). The sarcoidosis group included 91 consecutive, unrelated subjects from two Clinical Centers (Pavia, 51 patients, and Padua, 40 patients), both located in Northern Italy. Eighty-five of these patients had one or more positive biopsies. In six patients, who presented with Löfgren’s syndrome, biopsy confirmation was not obtained (3). Mean FEV1 was 92.1 10% predicted and FVC was 91.5 9.7. According to a common protocol, patients were examined for several features of the disease; their characteristics are reported in Table 1. In particular, clinical course features (stage progression [patients showing a progression from

TABLE 1

Characteristics of sarcoidosis patients No.

91

Age*

Sex

Biopsy

Onset†

35 10

Female 56 Male 35

Yes 85 No 6

Sympt. 55 Asympt. 36

Stage

0 I II III

5 24 44 18

Stage progression‡

Outcome§

Spread¶

Relapses

Yes 13 No 70

Good 28 Poor 56

Yes 31 No 60

Yes 27 No 55

* Mean age (yr) SD at the onset. † Sympt. symptoms; Asympt. routine chest X-ray. ‡ Patients with (Yes) or without (No) progression from one roentgenologic stage to the next during the follow-up (available in 83/91). § Good spontaneous or prednisone-induced resolution; Poor persistance or worsening of X-ray abnormalities (available in 84/91). ¶ Extrapulmonary spread: eyes, liver, skin, joints, parotid, spleen, peripheral lymphnodes. Available in 82/91.

Zorzetto, Bombieri, Ferrarotti, et al.: CR1 Gene in Sarcoidosis

19

TABLE 2

PCR and restriction analysis conditions used for the CR1 gene polymorphism detection Polymorphism

Primers

His1208Arg A3650G Exon 22

5-TTCACATTGCATAGCCAGAGC-3 5-CCAGAGGTTAATCTCCCTGGA-3

Intron 27

5-CCTTCAATGGAATGGTGCA-3 5-GGTCTGAACGGAATGTTCCC-3

Pro1827 Arg C5507G Exon 33

5-AAGCGCACAGTCACAGGTCAC-3 5-GGAGGTAGTTCTGTCTCTGAC-3

intron 27 HindIII RFLP, and 59C for Pro1827Arg) for 15 s, and 72C for 60 s. His1208Arg and intron 27 HindIII RFLP restriction fragments were resolved in 2% agarose gel. Pro1827Arg restriction fragments were resolved in non denaturing 16% polyacrylamide gel. Examples of genetic analysis of the three polymorphic sites of CR1 gene are shown in Figure 1. Pro1827Arg was analyzed also by direct sequencing, using a BigDye terminator cycle sequencing kit (Perkin-Elmer) (Figure 2). Sequencing was performed as a check of the restriction method in all samples showing the CG or GG genotype electrophoretic pattern and in 20 random samples showing the CC genotype electrophoretic pattern. A 100% concordance between gel-resolved PCR products and direct sequencing was found.

Statistical Analysis Clinical data are presented as mean SD. Hardy-Weinberg equilibrium was assessed by goodness-of-fit 2 test for biallelic markers. Frequencies were compared with 2 test, and differences considered statistically significant when the P value was 0.05. Linkage disequilibrium and haplotype frequencies were estimated with the ASSOCIATE software program (Rockefeller University, New York, NY) (25). Statistical analysis of CR1 genotype frequencies and clinical features of sarcoidosis was calculated by 2 contingency table analysis with appropriate degrees of freedom.

Size (bp)

Annealing temperature (°C)

Restriction enzyme

Fragment size (bp)

682

63

RsaI

A: 520, 162 G: 458, 162, 62

1,600

60

HindIII

H: 1,600 L: 1,100, 500

305

59

MnlI

C: 111, 80, 37, 33, 33, 11 G: 111, 80, 70, 33, 11

Results Table 3 displays the genotypes of the three CR1 polymorphic sites examined. Linkage disequilibrium among the three polymorphisms taken two by two was found (D22,i27 0.9879; D22,330.9773; Di27,33 0.9642). We were able to confirm the linkage disequilibrium, already reported in Caucasians (16) for the intron 27 HindIII RFLP and the His1208Arg and Pro1827Arg polymorphisms, in our series of Caucasian subjects. The less common His1208Arg G allele and Pro1827Arg G allele are therefore believed to be linked to a low (L) CR1/E ratio (17). Table 4 reports the deduced haplotype frequencies in the three groups of individuals investigated. Frequencies between sarcoidosis and both control groups were significantly different (2 7.12, P 0.0284 versus healthy control subjects; 2 7.10, P 0.0287 versus COPD control subjects). The frequency of the Pro1827Arg polymorphism in the three groups of subjects is described in Table 5. Genotypes of controls were in accordance with the Hardy-Weinberg equilibrium. In marked contrast, those of the sarcoidosis patients were out of Hardy-Weinberg (P 0.007), and this Figure 1. Electrophoretic patterns of the three CR1 gene polymorphic sites examined. (A) PCR product of the exon 22 His1208Arg polymorphism after digestion with RsaI. Samples 1 and 2 are homozygous for the A allele, sample 3 is AG heterozygous, sample 4 is homozygous for the G allele. (B) Amplimer of the intron 27 polymorphism digested with HindIII. Sample 1 is HL heterozygous, sample 2 is homozygous for the L allele, and sample 3 is homozygous for the H allele. Marker used in both A and B is Hyper Ladder III (Bioline). (C) Amplimer of exon 33 Pro1827Arg polymorphism after digestion with MnI. Sample 1 is homozygous for the C allele, sample 2 is homozygous for the G allele, sample 3 is CG heterozygous. Marker 20 bp (Sigma, St. Louis, MO).

20


cal characteristics, assumed to be representative of the heterogeneity of the disease, i.e., age of onset, roentgenologic stage, progression, outcome, extrapulmonary spread, and relapses (data not shown). We could not associate any CR1 genotype, and the GG genotype in particular, to any clinical characteristics of the disease considered.

Discussion

Figure 2. Sequence of CR1 exon 33 Pro1827Arg polymorphism. (A) Sample homozygous for C allele. (B) Sample homozygous for G allele. (C) CG heterozygous sample. Arrows show the position of the mutation.

feature was even stronger when the female sarcoid subset was considered (P 0.0005). We found that the GG genotype was overrepresented in the sarcoidosis group (0.18 versus 0.06 in healthy control subjects, and 0.07 in COPD control subjects; 2 5.531, OR 3.13 95%, CI 1.49–6.69, P 0.016 and 2 3.927, OR 2.82, 95% CI 1.27–6.39, P 0.038, respectively). This finding is strengthened by the significant difference (P 0.0114) between sarcoidosis and control subjects when the GG LL GG multiple genotype (Table 3) was examined within a multiple genotype comparison (performed in 243 out of 256 individulas). After stratifying by sex, the difference was restricted to female sarcoidosis subjects (0.22) with respect to healthy control subjects (0.04) (2 7.591, OR 7.05, 95% CI 3.10–16.61, P 0.005). The presence of two less common alleles seems necessary for the association, in accordance with a possible recessive gene effect. The Pro1827Arg GG genotype distribution did not differ among male subjects from any group (Table 5). Next, we examined whether or not there was any relationship between the Pro1827Arg polymorphism and clini-

This is the first report of an association between CR1 gene polymorphism and sarcoidosis or, to our knowledge, to any disease. A low number of CR1 on erythrocytes has been reported in several disorders, including systemic lupus erythematosus (SLE), immune hemolytic anemias, AIDS, and lepromatous leprosy (26). Data seem to support the concept that, at least in AIDS (27), and in SLE (28), the defect was acquired and not inherited, because H and L alleles did not show altered frequencies. The number of CR1 on erythrocytes in sarcoidosis is unknown; one report has dealt with the soluble form of CR1 (sCR1), shed from activated neutrophils, in a number of lung diseases, including sarcoidosis, showing that sCR1 is increased in bronchoalveolar lavage fluids, but not in the sera (29). We paid particular attention to the Pro1827Arg polymorphism since this was available in the largest series of subjects and because the C→G substitution leading to the Pro→Arg amino acid change at position 1,827 of the protein sequence in the proximal extramembrane region of the L allotype, according to Herrera and colleagues (16), may create a cleavage site allowing its accelerated proteolysis. Our study shows that the G allele and GG genotype of the CR1 gene Pro1827Arg polymorphism are significantly more frequent in subjects with sarcoidosis, a diffuse interstitial lung disorder, than in sex-, age- and ethnicallymatched healthy controls (Table 5). We also had evidence that the polymorphism was more frequent in sarcoidosis than in COPD, an obstructive lung disorder, assumed as a “positive” control group, i.e., a group of subjects in whom the same organ was affected, but by a disorder whose biochemical bases are completely different from those of sarcoidosis. This observation is, however, limited by the fact that, although ethnically matched, the COPD series differed in terms of mean age and sex: in fact, by definition, COPD arises between the 6th and 7th decade of life, and mostly affects men. CR1 polymorphism frequency did not differ between healthy control subjects and subjects with COPD. Of interest was the finding that the GG genotype seemed to be particularly frequent in sarcoidosis patients of female sex (22%), whereas 12% of male patients were homozygous for the G allele (Table 5). It should be borne in mind that most studies suggest a slightly higher disease rate in women (3). The difference with female healthy controls and female COPD controls was striking (OR 7.056 and 10.29, respectively). However, comparison with the latter group was limited by the low number of female subjects, as mentioned above. An interesting point is that genotype data in sarcoidosis are out of the Hardy-Weinberg equilibrium, especially after stratification by sex. In the hypothesis that the gene may be involved in the disease, altered genotype frequency


21

TABLE 3

CR1 gene polymorphism multiple genotypes (%) Healthy control subjects

COPD control subjects

Sarcoidosis

His1208Arg HindIII RFLP Pro1827Arg (i27 H/L) (C5507G) (A3650G)

His1208Arg HindIII RFLP Pro1827Arg (A3650G) (i27 H/L) (C5507G)

His1208Arg HindIII RFLP Pro1827Arg (A3650G) (i27 H/L) (C5507G)

AA AG AG AG AG AG GG GG

HH HH HH HL HL HL HL LL

CC CC CG CG CC GG GG GG

Total

56 (61) 2 1 25 (27) 1 1 2 3 (3) — 91

AA AG AG AG AG GG GG GG

HH HH HH HL HL HL HL LL

CC CC CG CG CC CG GG GG

44 (64) 2 1 14 (20) 1 1 1 4 (5) — 68

AA AG AG AG AA AG

HH HH HH HL HL HL

CC CC CG CG CC GG

40 (47) 1 3 25 (29) 1 1

GG

LL

GG

13 (15) — 84

13 subjects were not typed for either His1208Arg or intron 27 HindIII RFLP polymorphisms due to insufficient DNA.

could be due to the increased frequency of the Pro1827Arg G allele in cases. Alternatively, a chance effect could be postulated, whereas we can reasonably rule out other possible explanations, such as genotyping errors, kinship among studied subjects, or population admixture. Beside the hypothesis that CR1 could be a susceptibility gene for sarcoidosis, we also investigated whether it could act as a gene modifying the pattern of the disease presentation. To this end we explored the relationships between the Pro1827Arg polymorphism and clinical variables. We did not find any statistically significant association. We therefore conclude that the CR1 gene does not seem to act as a modifier factor in sarcoidosis. However, this point deserves further investigation in larger series of subjects. On the basis of our findings, we may postulate a role for the Pro1827Arg G allele in the pathogenesis of sarcoidosis. There is general consensus that sarcoidosis can be triggered by unidentified antigen(s) (1–4). Based on features of sarcoid granulomata, it has been believed that poorly degradable antigen(s) or antigen–antibody complexes are most likely to be important in the formation of the sarcoid granulomata (9). As a consequence of the probable ubiq-

TABLE 4

Deduced haplotype frequencies of the CR1 gene polymorphisms Haplotype (e22–i27–e33)*

A A A A G G G G

H H L L H H L L

C G C G C G C G

Sarcoidosis†

0.6546 0.0060 0.0060 0.0000 0.0060 0.0179 0.0000 0.3094

(110) (1) (1) (0) (1) (3) (0) (52)

Healthy control subjects

0.7747 0.0055 0.0000 0.0000 0.0110 0.0165 0.0055 0.1868

(141) (1) (0) (0) (2) (3) (1) (34)


0.7794 0.0000 0.0000 0.0000 0.0219 0.0149 0.0076 0.1763

(106) (0) (0) (0) (3) (2) (1) (24)

* e22 His1208Arg; i27 HindIIIRFLP; e33 Pro1827Arg. † 2 7.12, P 0.0284 versus healthy controls; 2 7.10, P 0.0287 versus COPD controls; 2 10.15, P 0.0063 versus healthy COPD control subjects. Numbers in parentheses indicate number of haplotypes estimated from deduced frequencies. 13 subjects were not typed for either His1208Arg or intron 27 HindIII RFLP due to insufficient DNA.

uity of the antigen(s), the development of the disease should be linked to the host’s immune response to the initiating agent, and a missing or delayed eradication of the antigen(s) and/or the ICs may play a role. Therefore, a low CR1/E ratio may be associated with CR1 gene polymorphism, impaired clearance of ICs, and their deposition outside the reticuloendothelial system (16). In this framework, we postulate that the CR1 polymorphism might be a genetic determinant of the susceptibility to sarcoidosis, at least in a subset of patients (maybe females). We are aware that the major limitation of our report is the lack of a relationship between genetic issues and the corresponding phenotype, because the CR1/E ratio was not determined in our study. Nevertheless, even if we may consider our report as a sort of serendipitous finding, we believe that it might be of some interest from a genetic point of view, because it opens up interesting speculations about many things, from the gene to the role of the protein on pathogenetic processes in sarcoidosis, and it suggests future areas of research. As an example, an intriguing question is: can our finding help speculation about the nature of the triggering antigen(s) of sarcoidosis? It has been suggested that one or more of micro-organisms behaving in a noninfectious fashion may trigger the sarcoidosis granulomatous process (4). Micro-organisms activate host complement: in some cases, such as pyogenic bacteria, the complement opsonin C3b is a major defense against infection (13), and in other cases, such as human immunodeficiency virus (30) or mycobacteria (31), activated complement causes C3b to bind to their surface, and C3b receptors are used by the microbes to enter the cell. The intimate mechanism, if any, by which the CR1 polymorphism could be involved in an “alternative” relationship between microbe(s) and host remains to be elucidated. Previous studies have dealt with infection susceptibility genes and sarcoidosis. No associations have been found for mannose-binding lectin (a collectin binding to arrays of terminal mannose group of bacteria) gene polymorphisms (32), whereas a mutation in the natural resistance-associated macrophage protein (NRAMP) gene, coding for a protein associated with susceptibility to tuberculosis, has been found to play a protective role toward sarcoidosis in

22


TABLE 5

Frequencies of CR1gene Pro1827Arg (C5507G) polymorphism in control and sarcoidosis groups Healthy control subjects

Genotype CC CG GG Allele C G


Sarcoidosis

Total (94)

M (42)

F (52)

Total (71)

M (53)

F (18)

60 (64)* 28 (30) 6 (6)

25 (60) 13 (31) 4 (9)

35 (67) 15 (29) 2 (4)

49 (69) 17 (24) 5 (7)

38 (72) 10 (19) 5 (9)

63 (75) 21 (25)

85 (82) 19 (18)

115 (81) 27 (19)

86 (81) 20 (19)

148 (79) 40 (21)

Total (91)

M (36)

F (55)

11 (61) 7 (39) 0

46 (50) 29 (32) 16 (18)†

16 (44) 16 (44) 4 (12)

30 (54) 13 (24) 12 (22)‡

29 (80) 7 (20)

121 (66) 61 (34)§

48 (67) 24 (33)

73 (66) 37 (34)¶

* Numbers in parentheses indicate percentages. † 2 5.531, OR 3.13, 95% CI 1.49–6.69, P 0.016 versus healthy control subjects; 2 3.927, OR 2.82, 95% CI 1.27 to 6.39 P 0.038 versus COPD. ‡ 2 7.591, OR 7.05, 95% CI 3.10–16.61, P 0.005 versus healthy control subjects; 2 4.699, OR 10.29 (Haldane correction) P 0.024 versus COPD. § 2 6.98, OR 1.87, 95% CI 1.14–3.05, P 0.008 versus healthy control subjects; 2 8.479, OR 2.15, 95% CI 1.24 to 3.74 P 0.003 versus COPD. ¶ 2 6.534, OR 2.27, 95% CI 1.15–4.50, P 0.008 versus healthy control subjects.

African Americans (33). By contrast, a polymorphism in the vitamin D receptor gene, linked to susceptibility to tuberculosis and leprosy, has been reported to be significantly associated with sarcoidosis in the Japanese (34), and the slight, but significant increase in cystic fibrosis gene mutations in Italian patients with sarcoidosis (35) raises the hypothesis that CFTR might play a role, based on its function of microbial receptor. In conclusion, we have found evidence of a significant association between sarcoidosis in Italian subjects, especially females, and polymorphisms of the CR1 gene, which in turn has been previously associated in a number of settings to a low CR1/E ratio, thus suggesting that, at least in a subset of sarcoidosis patients, a reduced clearance of antigen(s) or ICs may occur. Interestingly, the association seems to be particularly restricted to female subjects, and, to our knowledge, this is the first report of a genetic sex difference in sarcoidosis. Although not linked to any particular pattern of disease presentation, we speculate about a possible role for CR1 as a susceptibility gene for sarcoidosis and, more interestingly, this finding, if confirmed by other, independent studies and in different ethnic groups, would further support the hypothesis that sarcoidosis is triggered by micro-organisms. Acknowledgments: The linguistic assistance of Dr. Rachel Stenner is gratefully acknowledged. Supported in part by IRCCS Policlinico San Matteo, Ricerca Corrente grant 1999 – 2001, and by Italian CNR Target Project Biotechnology.

References 1. Müller-Quernheim, J. 1998. Sarcoidosis: immunopathogenetic concepts and their clinical application. Eur. Respir. J. 12:716–738. 2. Newman, L. S., C. S. Rose, and L. A. Maier. 1997. Sarcoidosis. N. Engl. J. Med. 336:1224–1234. 3. Hunninghake, G. W., U. Costabel, M. Ando, R. Baughman, J. F. Cordier, R. du Bois, A. Eklund, M. Kitaichi, J. Lynch, G. Rizzato, C. Rose, O. Selroos, G. Semenzato, and O. P. Sharma. 1999. ATS/ERS/WASOG statement on sarcoidosis. Sarcoidosis Vasc. Diffuse Lung Dis. 16:149–173. 4. Grath, D. S., N. Goh, P. J. Foley, and R. M. du Bois. 2001. Sarcoidosis: genes and microbes–soil or seed? Sarcoidosis Vasc. Diffuse Lung Dis. 18:149–164. 5. Rybicki, B. A., M. J. Maliarik, M. Major, J. Popovich, Jr., and M. C. Iannuzzi. 1997. Genetics of sarcoidosis. Clin. Chest Med. 18:707–717. 6. Luisetti, M., A. Beretta, and L. Casali. 2000. Genetic aspects in sarcoidosis. Eur. Respir. J. 16:768–780. 7. Schürmann, M., P. A. Lympany, P. Reichel, B. Müller-Myhsok, K. Wurm, M. Schlaak, J. Müller-Quernheim, R. M. du Bois, and E. Schwinger. 2000. Familial sarcoidosis is linked to the Major Histocompatibility Complex region. Am. J. Respir. Crit. Care Med. 162:861–864.

8. Daniele, R. P., L. J. McMillan, J. H. Dauber, and M. D. Rossman. 1978. Immune complexes in sarcoidosis: a correlation with activity and duration of disease. Chest 74:261–264. 9. Thomas, P. D., and G. W. Hunninghake. 1987. Current concepts of the pathogenesis of sarcoidosis. Am. Rev. Respir. Dis. 135:747–760. 10. Hunninghake, G. W., and R. G. Crystal. 1981. Mechanisms of hypergammaglobulinemia in pulmonary sarcoidosis. Site of increased antibody production and role of T lymphocytes. J. Clin. Invest. 67:86–92. 11. Spector, W. G., and N. Heesom. 1969. The production of granulomata by antigen-antibody complexes. J. Pathol. 98:31–39. 12. Fearon, D. T., and W. W. Wong. 1983. Complement ligand-receptor interactions that mediate biological responses. Annu. Rev. Immunol. 1:243–271. 13. Walport, M. J. 2001. Complement: first of two parts. N. Engl. J. Med. 344: 1058–1066. 14. Oudin, S., M. Tonye Libyh, D. Goossens, X. Dervillez, F. Philibert, B. Reveil, F. Bougy, T. Tabary, P. Rouger, D. Klatzmann, and J. H. Cohen. 2000. A soluble recombinant multimeric anti-Rh(D) single-chain Fv/CR1 molecule restores the immune complex binding ability of CR1-deficient erythrocytes. J. Immunol. 164:1505–1513. 15. Ripoche, J., and R. B. Sim. 1986. Loss of complement receptor 1 (CR1) on ageing of erythrocytes: studies of proteolytic release of the receptor. Biochem. J. 235:815–821. 16. Herrera, A. H., L. Xiang, S. G. Martin, J. Lewis, and J. G. Wilson. 1998. Analysis of complement receptor type 1 (CR1) expression on erythrocytes and of CR1 allelic markers in Caucasian and African American populations. Clin. Immunol. Immunopathol. 87:176–183. 17. Xiang, L., J. R. Rundles, D. R. Hamilton, and J. G. Wilson. 1999. Quantitative alleles of CR1: coding sequence analysis and comparison of haplotypes in two ethnic group. J. Immunol. 163:4939–4945. 18. Wilson, J. G., E. E. Murphy, W. W. Wong, L. B. Klickstein, J. H. Weis, and D. T. Fearon. 1986. Identification of a restriction fragment length polymorphism by a CR1 cDNA that correlates with the number of CR1 on erythrocytes. J. Exp. Med. 164:50–59. 19. Moulds, J. M., M. Brai, J. Cohen, A. Cortelazzo, M. Cuccia, M. Lin, S. Sadallah, J. Schifferli, V. Bala Subramian, L. Truedsson, G. W. Wu, F. Zhang, and J. P. Atkinson. 1998. Reference typing report for complement receptor 1 (CR1). Exp. Clin. Immunogenet. 15:291–294. 20. Silverman, E. K., and L. J. Palmer. 2000. Case-control association studies for the genetics of complex respiratory disorder. Am. J. Respir. Cell Mol. Biol. 22:645–648. 21. Stockley, R. A. 1996. Cellular mechanisms in the pathogenesis of COPD. Eur Respir Rev 6:264–269. 22. American Thoracic Society. 1995. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 152(Suppl.):S78–S121. 23. Cornillet, P., F. Philbert, M. D. Kazatchkine, and J. H. M. Cohen. 1991. Genomic determination of the CR1 (CD35) density polymorphism on erythrocytes using polymerase chain reaction amplification and HindIII restriction enzyme digestion. J. Immunol. Methods 136:193–197. 24. Vik, D. P., and W. W. Wong. 1993. Structure of the gene for the F allele of complement receptor type 1 and sequence of the coding region unique to the S allele. J. Immunol. 151:6214–6224. 25. Ott, J. 1999. Analysis of Human Genetic Linkage, 3rd ed. John Hopkins University Press, Baltimore, MD. 26. Lach-Trifilieff, E., J. Marfurt, S. Schwarz, S. Sadallah, and J. Schifferli. 1999. Complement receptor 1 (CD35) on human reticulocytes: normal expression in systemic lupus erythematosus amd HIV-infected patients. J. Immunol. 162:7549–7554. 27. Cohen, J. H., C. Geffriaud, V. Caudwell, and M. D. Kazatchkine. 1989. Ge-


28.

29. 30. 31. 32.

netic analysis of the CR1 (the C3b complement receptor, CD35) expression on erythrocytes of HIV-infected individuals. AIDS 3:397–399. Kumar, A., A. Kumar, S. Shina, P. S. Khandekar, K. Banerjee, and L. M. Srivastava. 1995. HindIII genomic polymorphism of the C3b receptor (CR1) in patients with SLE: low erythrocyte CR1 expression is an acquired phenomenon. Immunol. Cell Biol. 73:457–462. Hamacher, J., S. Sadallah, J. A. Schiefferli, J. Villard, and L. P. Nicod. 1998. Soluble complement receptor type 1 (CD35) in bronchoalveolar lavage of inflammatory lung disease. Eur. Respir. J. 11:112–119. Speth, C., L. Kacani, and M. P. Dierich. 1997. Complement receptors in HIV infection. Immunol. Rev. 159:49–67. Schorey, J. S., M. C. Carroll, and E. J. Brown. 1997. A macrophage invasion mechanism of pathogenic mycobacteria. Science 277:1091–1093. Foley, P. J., C. G. Mullighan, D. S. McGrath, P. Pantalidis, S. Marshall,

23

P. A. Lympany, K. I. Welsh, and R. M. du Bois. 2000. Mannose-binding lectin promoter and structural gene variants in sarcoidosis. Eur. J. Clin. Invest. 30:549–552. 33. Maliarik, M. J., K. Mei Chen, R. G. Sheffer, B. A. Rybicki, M. L. Major, J. Popovich, Jr, and M. C. Iannuzzi. The natural resistance-associated macrophage protein gene in African Americans with sarcoidosis. Am. J. Respir. Cell Mol. Biol. 22:672–675. 34. Niimi, T., H. Tomita, S. Sato, H. Kawaguchi, K. Akita, H. Maeda, Y. Sugiura, and R. Ueda. 1999. Vitamin D receptor gene polymorphism in patients with sarcoidosis. Am. J. Respir. Crit. Care Med. 160:1107–1109. 35. Bombieri, C., M. Luisetti, F. Belpinati, E. Zuliani, A. Beretta, J. Baccheschi, L. Casali, and P. F. Pignatti. 2000. Increased frequency of CFTR gene mutation in sarcoidosis: a case/control association study. Eur. J. Hum. Genet. 8:717–720.