Association of killer cell immunoglobulin-like receptors with scleroderma




Scleroderma is an autoimmune disorder of unknown etiology. A genetic contribution has been demonstrated, and genes influencing activation of the immune system have been potentially identified as candidate genes in this process. The repertoire of killer cell immunoglobulin-like receptors (KIRs) that are involved in the activation of T cells and natural killer cells is highly variable. Recently, an association of KIR2DS2 with vasculitis in patients with rheumatoid arthritis has been reported. Because scleroderma is characterized by an involvement of the vascular system, we sought to determine whether KIR2DS2 is associated with scleroderma.


We typed 9 KIR genes in 102 patients with scleroderma and in 100 blood donors, using polymerase chain reaction on genomic DNA.


Twelve patients with scleroderma, compared with only 2 blood donors, had KIR phenotypes characterized by the presence of the activating KIR2DS2 and the absence of the corresponding inactivating KIR2DL2 (P = 0.005).


The genetic combination of KIR2DS2+ and KIR2DL2− is associated with scleroderma.

Scleroderma is an autoimmune disease characterized by tissue fibrosis, inflammation, and vascular injury and by the production of autoantibodies against nuclear antigens. Twin and family studies have revealed a genetic contribution to the etiology of the disease (1). Chronic viral infections may induce the disease (2). In addition, oligoclonal expansion of T cells has been observed in skin lesions, suggesting that T cells are involved in the pathogenesis of scleroderma (3, 4). Therefore, genes regulating T cell activation may be genetic risk factors for scleroderma.

Activation of T cells is mediated by the T cell receptor (TCR). This activation may be enhanced by the coreceptors CD4 and CD8 and modulated by killer cell immunoglobulin-like receptors (KIRs) (5). KIRs are members of the immunoglobulin superfamily and are expressed on natural killer (NK) cells and subsets of T cells. Inhibitory KIR molecules bind to target cell major histocompatibility complex (MHC) class I molecules and prevent the attack of NK cells on normal cells (6, 7). MHC class I molecules, the ligands for inhibitory KIR molecules, are extremely polymorphic. Inhibitory KIRs bind distinct subsets of MHC class I molecules. Thus, KIR3DL1 binds to the Bw4 subset of HLA–B allotypes. KIR2DL1 binds to HLA–C, and this is characterized by a Lys80 residue (HLA–Cw4 and related alleles), whereas KIR2DL2 and KIR2DL3 recognize HLA–C with an Asn80 residue (HLA–Cw3 and related alleles) (8). Varying expression of KIRs on NK and T cell clones may allow subsets of these cells to recognize different virus-infected or malignant cells.

Since KIR molecules bind to the complex of MHC class I–peptides, they can interact on T cells via binding of the TCR to the same structure and can modulate activation via the TCR (9, 10). Activating KIRs in addition to inactivating molecules have been described. Although the extracellular sequences of the activating KIRs and the corresponding inactivating KIRs differ only slightly, the ligands of activating KIRs have not yet been clearly defined. Only a weak interaction of KIR2DS1 and Lys80 HLA–C molecules has been reported so far (8).

The KIR genes localized on chromosome 19q13.4 are highly diverse, since there are at least 13 different loci that are not universally expressed (11, 12). Recently, the presence of KIR2DS2 has been found to be associated with vasculitis in patients with rheumatoid arthritis (13). In addition, the presence of KIR2DS2 and KIR2DL2 in the absence of ligands for KIR2DL2 is associated with psoriatic arthritis (14). Therefore, in the present study, we examined whether there is also an association of KIR2DS2 with scleroderma.



After written consent was secured, DNA of 102 patients with scleroderma was obtained from a stock in Hannover, Germany that is being collected for a multicenter study on the genetics of connective tissue diseases. The patients were recruited from dermatology outpatient clinics in Cologne, and 22 of the patients were recruited from self-help groups in Hannover. All patients were German Caucasians and fulfilled the American College of Rheumatology (formerly, the American Rheumatism Association) criteria for the classification of scleroderma (15). As a control, DNA from 100 German Caucasian blood donors was obtained.

KIR typing.

KIRs were typed in scleroderma patients and controls using a polymerase chain reaction (PCR) method described earlier (16). Genomic DNA was used as the template. The KIRs examined and the respective pairs of primers are summarized in Table 1. For the PCR, 50–100 ng DNA, 50 mM KCl, 10 mM Tris HCl, pH 8.3, 2.5 mM MgCl2, 400 μM dNTPs, 0.4 μM primer, and 1 unit of Taq polymerase (PE Applied Biosystems, Foster City, CA) were mixed in a total volume of 50 μl. A PCR was performed with 30 cycles (each for 1 minute at 94°C, 30 seconds at 57°C, and 1 minute at 72°C). The PCR products were analyzed on a 1.2% agarose gel after electrophoresis.

Table 1. Primers used for definition of killer cell immunoglobulin-like receptor (KIR) genes
KIR5′ primer3′ primer

Statistical evaluation.

Scleroderma patients and controls were segregated into those with and those without each of the 9 KIRs examined, and were also separated into those with and those without the presence of an activating KIR without the simultaneous presence of the extracellularly most-related inactivating KIR (KIR2DS1+/KIR2DL1−, KIR2DS2+/KIR2DL2−, or KIR2DS3+/KIR2DL3−). It is important to note that other inactivating KIRs may be structurally related to activating KIRs; for example, KIR2DL3 is very similar to KIR2DS2.

Associations of KIRs or the combinations of KIRs with scleroderma were calculated using Fisher's exact test. Initially, we set out to look for an association of scleroderma with only KIR2DS2, KIR2DL2, and the combination of the two. Using Bonferroni's correction for multiple testing for those 3 associations, a probability of association of <0.0169 was regarded as statistically significant. Subsequently, we examined the correlation between scleroderma and the presence of all 9 KIRs and 3 KIR combinations (12 variables), and a probability of association of <0.0042 was regarded as statistically significant.


We analyzed DNA samples from 102 scleroderma patients and 100 blood donors, to look for an association of scleroderma with KIR2DS2 and the corresponding inhibiting KIR2DL2. KIR2DS2 was detected in 59 of 102 scleroderma patients and in 53 of 100 blood donors (P not significant [NS]), and KIR2DL2 was found in 47 of 102 scleroderma patients and in 51 of 100 blood donors (P NS) (Table 2). However, the combination KIR2DS2+/KIR2DL2− was present in 12 of 102 scleroderma patients but in only 2 of 100 blood donors (P = 0.005) (Table 3). When clinical, laboratory, and demographic data on the scleroderma patients with and those without the combination KIR2DS2+/KIR2DL2− were compared, no significant differences were found with regard to the presence of esophageal, pulmonary, and renal involvement, calcinosis, CENP-B and Scl70 autoantibodies, mean age (63 years for both groups), and disease duration (7 years and 6 years, respectively) (data not shown).

Table 2. Frequency of 9 KIR genes in scleroderma patients and control blood donors*
KIRScleroderma patients, % (n = 102)Controls, % (n = 100)P
  • *

    The weak association of KIR2DS3 and scleroderma could be disregarded after Bonferroni's correction for the multiple parameters examined. NS = not significant (see Table 1 for other definitions).

Table 3. Frequency of combinations of activating KIRs and the corresponding inhibitory KIRs in scleroderma patients (n = 102) and control blood donors (n = 100)*
KIR combinationsScleroderma patients, %Controls, %P
  • *

    See Tables 1 and 2 for definitions.


We then examined the association of scleroderma with KIR2DS3 and KIR2DL3, which are extracellularly more than 90% identical to KIR2DS2 and KIR2DL2, respectively. KIR2DS3 was present in 34 of 102 scleroderma patients and in 23 of 100 blood donors (P = 0.033). KIR2DL3 was present in 94 of 102 scleroderma patients and in 91 of 100 blood donors (P NS) (Table 2).

In order to exclude the possibility that the associations of scleroderma with KIR2DS3 and with the combination KIR2DS2+/KIR2DL2− were not indirectly caused by an association with other KIRs in linkage disequilibrium, we typed 5 further KIRs in patients and controls to compare the distribution of the combinations of KIRs, that is, the “KIR phenotypes” that were originally described by Uhrberg et al (16) and subsequently by Crum et al (12). None of the KIR phenotypes was significantly associated with scleroderma (Table 4). Moreover, the further combinations of activating KIRs and corresponding antagonistic KIRs 2DS1+/2DL1− (6 of 102 scleroderma patients versus 5 of 100 controls), 2DS3+/2DL3− (3 of 102 scleroderma patients versus 5 of 100 controls), and 3DS1+/3DL1− (8 of 102 scleroderma patients versus 3 of 100 controls) were not associated with scleroderma (Table 3).

Table 4. Frequency of the 10 most frequent KIR phenotypes in scleroderma patients (n = 102) and control blood donors (n = 100)*
KIR phenotype2DL12DL22DL32DS12DS22DS32DS43DS13DL1Scleroderma patients, %Controls, %
  • *

    Phenotypes defined according to Crum et al (12). No significant differences were noted. See Table 1 for definitions.

7 and 8+++++++++9.88
Others         27.523


Vascular damage is one of the hallmarks of scleroderma, but its pathogenesis is unclear. Family studies have demonstrated a contribution of genetic factors to the pathogenesis of scleroderma. The candidate genes in scleroderma may be involved in T cell activation, since T cells are oligoclonally expanded in skin lesions and appear to be involved in the pathogenesis of scleroderma (4). The genes that regulate the activation of T cells include the KIR genes, which modulate activation via the TCR and, in addition, regulate the cytotoxicity of NK cells.

Both activating KIRs and inhibitory KIRs have been characterized (6, 7). MHC binding of the inactivating receptors induces phosphorylation of immunoregulatory tyrosine-based motifs (ITIM) in their cytoplasmic domain (17). The phosphorylated ITIMs recruit the cytosolic protein tyrosine phosphatases SHP-1 and SHP-2 and, subsequently, substrates critical for cellular activation are dephosphorylated. In contrast, activating KIRs lack the ITIM and possess a charged residue in their transmembrane domains that mediates interaction with the DAP12 signal transduction chain (18). DAP12 contains an immunoreceptor tyrosine-based activation motif in its cytoplasmic domain. Cross-linking of KIR–DAP12 complexes results in cellular activation and binding of ZAP-70 and Syk protein tyrosine kinases, similar to the activation pathway of T cell and B cell antigen receptors.

Of the KIR genes described so far, KIR2DS2 has gained particular interest, since it has been shown to be associated with vasculitis in rheumatoid arthritis patients (13). A second study detected an association of psoriatic arthritis with the presence of KIR2DS2 and KIR2DL2, in the absence of HLA–C molecules containing an asparagine in position 80 (ligands of KIR2DL2) (14). Interaction of KIR2DL2, which is expressed on NK cells and subsets of activated CD4+ and CD8+ T cells, with HLA–C molecules can block T cell activation and target cell lysis of NK cells. The ligand of KIR2DS2, which is extracellularly almost identical to both KIR2DL2 and KIR2DL3, is still unknown. Using tetramers of HLA–Cw3, binding to KIR2DL2, but not to KIR2DS2, could be demonstrated (19). On the other hand, functional analysis using soluble HLA revealed signaling via activating KIRs (20). It is, therefore, likely that activating KIRs bind to the same targets as those of inactivating KIRs, but with a lower affinity. In contrast, activation of NK cells with monoclonal antibodies against 2DS2, 2DL2, and 2DL3 revealed that low concentrations of antibody stimulate preferentially the activating receptor, whereas high antibody concentrations stimulate both receptor types (21). Thus, inactivating receptors appear to have a higher affinity toward the antigen, but also a higher activation threshold.

In this respect, KIR2DS2 and KIR2DL2/KIR2DL3 would be antagonistic molecules involved in fine tuning of the threshold for NK and T cell activation. The physiologic relevance of KIR molecules appears to be in its surveillance of tumors and defense against virus. For example, the KIR repertoire has been shown to influence the prognosis of human immunodeficiency virus infection (22). It is conceivable that the KIR repertoire also influences the immune response against a putative viral infection initiating scleroderma.

In our study, not KIR2DS2 itself, but the combination of KIR2DS2+/KIR2DL2− was associated with scleroderma. It is not clear whether this will have any functional impact, since KIR2DL3, which seems to bind to similar MHC molecules as that of KIR2DL2, was present in all but 1 of the patients with the combination KIR2DS2+/KIR2DL2−. It remains to be shown how the presence of KIR2DS2 without KIR2DL2 is generated. A recent description of KIR haplotypes did not include that combination (23). Since the other KIR molecules were not uniformly distributed in the 11 scleroderma patients with the unusual KIR combination, it is likely that several rare KIR haplotypes are associated with scleroderma.

With regard to the findings regarding the involvement of KIR2DS2 and KIR2DL2 in the pathogenesis of autoimmune disorders, 2 examples have been clearly defined so far: 1) association of KIR2DS2 itself with vascular disorders in rheumatoid arthritis patients, and 2) association of KIR2DS2 and the absence of ligands of KIR2DL2 in psoriatic arthritis. Herein, we have described a third way in which KIR2DS2 in the absence of KIR2DL2 is involved in disease pathogenesis in its association with scleroderma. In addition, in our study there was an association of scleroderma with KIR2DS3, which is structurally and functionally related to KIR2DS2; however, this weak association with KIR2DS3 could be disregarded after Bonferroni's correction for multiple parameters tested.

The likelihood cannot be ruled out that, rather than KIR2DS2 and KIR2DS3, other genes in linkage disequilibrium are really associated with scleroderma. It is, however, unlikely that such a gene may be one of the KIRs themselves, since the remaining KIRs or combinations of KIRs, the so-called “KIR phenotypes” and polymorphisms of ILT-2 (a receptor adjacent to the KIR locus) (data not shown), were not associated with scleroderma in our study.

The prevalence of the combination KIR2DS2+/KIR2DL2− in the control blood donors in our study is identical to the prevalence observed in control subjects in a recent study conducted in Australia (11). In a recent report, the first KIR haplotypes were described (23). None of the haplotypes included the combination KIR2DS2+/KIR2DL2−. The prevalence of KIR2DS2+/KIR2DL2−, however, differs from that in another study performed in an Irish population (12). Therefore, further studies of scleroderma patients from different ethnic backgrounds are necessary. Nevertheless, so far, the presence of KIR2DS2 and its ligands seems to be associated with a number of autoimmune disorders, including scleroderma, and thus may provide a target for therapeutic interventions.