Published ahead of print on October 27, 2005, doi:10.1164/rccm.200501-124OC
© 2006 American Thoracic Society doi: 10.1164/rccm.200501-124OC
HLA-C and Killer Cell Immunoglobulin-like Receptor Genes in Idiopathic BronchiectasisLung Immunology Group, Department of Biological Sciences and National Heart and Lung Institute, Faculty of Medicine, South Kensington Campus; Department of Occupational and Environmental Medicine, National Heart and Lung Institute, Brompton Campus; and Human Disease Immunogenetics Group, Department of Infectious Diseases, Hammersmith Campus, Imperial College; Host Defense Unit, Department of Respiratory Medicine, Royal Brompton and Harefield National Health Service Trust, London; Heart Science Centre, Harefield Hospital, Imperial College, Harefield; and Immunology Division, Department of Pathology, University of Cambridge, Cambridge, United Kingdom Correspondence and requests for reprints should be addressed to Rosemary Boyton, M.D., Lung Immunology Group, Department of Biological Sciences/National Heart and Lung Institute, Sir Alexander Fleming Building, South Kensington Campus, Faculty of Medicine, Imperial College, London SW7 2AZ, UK. E-mail: r.boyton{at}imperial.ac.uk
Rationale: In idiopathic bronchiectasis, lung inflammation and chronic bacterial infection lead to progressive lung damage. A possible role for natural killer (NK) cells is suggested by the observation that familial bronchiectasis occurs in a rare group of individuals with impaired HLA class I expression and consequent NK cell dysfunction. Objective: Because the HLA-C locus and killer cell immunoglobulin-like receptors (KIRs) are of key importance for NK cell recognition, we analyzed HLA-C/KIR combinations by genotyping patients with idiopathic bronchiectasis. Methods: Genomic DNA from 96 individuals with idiopathic bronchiectasis and 101 control subjects was analyzed by polymerase chain reaction with sequence-specific primers. High-resolution HLA-C genotyping was performed in addition to KIR analysis. Results: HLA-Cw*03 alleles and, in particular, HLA-C group 1 homozygosity are associated with the presence of bronchiectasis. Analysis of the relationship between HLA-C and KIR genes suggests a shift to activatory NK cell function. Conclusion: This is the first demonstration of genetic susceptibility in idiopathic bronchiectasis. The association with HLA-C group 1 homozygosity, and the interplay between HLA-C and KIR genes, argue for a role for NK cells in the progressive lung damage seen in this disease. This will require further investigation using functional studies.
Key Words: bronchiectasis • HLA-C • humans • immunity • killer cell immunoglobulin-like receptor Idiopathic bronchiectasis is a lung disease in which a dysregulated inflammatory response and recurrent bacterial infection result in progressive lung damage. Bronchiectasis in general can be defined as a common structural end point that can be reached by several pathological routes ranging from foreign body obstruction to postinfectious damage (Mycobacterium tuberculosis), genetic defects (cystic fibrosis), abnormal host defense (ciliary dyskinesia and hypogammaglobulinemia), and autoimmune disease (systemic lupus erythematosus, rheumatoid arthritis, and ulcerative colitis) (1). However, in idiopathic bronchiectasis, a clinically defined group of patients in whom there is bilateral, predominantly lower lobe disease associated with sinusitis, no underlying disease mechanism has been identified to date. In fact, the diagnosis can be made only after all known causes of bronchiectasis have been excluded.
Patients with bronchiectasis are commonly infected with bacterial pathogens such as Haemophilus influenzae, Streptococcus pneumoniae, and Pseudomonas and show persistent neutrophil trafficking (2, 3). It is thought that colonization of the lower respiratory tract by microorganisms causes a chronic inflammatory response characterized by neutrophil migration into the airways and secretion of tissue-damaging oxidants and enzymes such as neutrophil elastase and myeloperoxidase with impaired mucus clearance (4, 5). Although evidence clearly indicates a dysregulated inflammatory process, the specific underlying immune mechanism is unknown. Both innate and adaptive responses are implicated. There is evidence indicating the presence of interleukin 6 (IL-6), IL-8, tumor necrosis factor Natural killer (NK) cells rapidly accumulate in lung parenchyma during inflammation, recruiting other cell types including neutrophils and T cells (9–11). They play a critical role in early host protection against S. pneumoniae infection and other pathogens including M. tuberculosis (12, 13). Phenotypic differences in NK cells have been shown in mouse strains susceptible and resistant to chronic lung infection with Pseudomonas aeruginosa (14). CD1d knockout mice show reduced lung eradication of P. aeruginosa (15). Chemokine-mediated recruitment of NK cells is also a critical host defense mechanism in invasive aspergillosis (16). These are all infectious pathogens that are often seen in bronchiectasis. In addition, NK cells have been implicated by the development of familial bronchiectasis in individuals with mutations in TAP (transporter associated with antigen processing) genes (17). These individuals have impaired class I expression and consequent dysfunction of NK cells that may contribute to progressive lung damage (18, 19).
In light of these data, investigation of NK cell regulation through HLA-C/killer cell immunoglobulin-like receptor (KIR) gene analysis in individuals with idiopathic bronchiectasis could be an important step in the elucidation of mechanisms underlying disease susceptibility. NK cells are involved in the surveillance of pathogens through the interaction of receptors expressed by NK cells with major histocompatibility complex (MHC) class I molecules of infected cells. Cells with abnormal MHC class I expression become targets of NK lytic activity resulting from loss of inhibition of NK cell activation (20, 21). NK cells interact with HLA class I ligands through receptors such as KIR. Particularly relevant to NK recognition by KIRs are polymorphic HLA-C molecules. Structurally and functionally there are two KIR groups: stimulatory and inhibitory KIRs (21, 22). It is not known whether stimulatory KIRs bind directly to HLA-C, but experimental evidence suggests that this might be the case. It has been proposed that the binding of HLA-C to activating KIR 2DS1 and KIR 2DS2 is weaker than to their inhibitory counterparts KIR 2DL1 and KIR 2DL2/3 (21, 23–26). KIR haplotypes, on chromosome 19q13.4, are highly polymorphic but can broadly be classified as A or B (21). Similarly, HLA-C alleles can be divided into group 1 or group 2 depending on whether there is an asparagine or lysine present at position 80 of the We analyzed HLA-C/KIR combinations in individuals with idiopathic bronchiectasis. We identified an increase in the prevalence of HLA-C group 1 homozygosity in idiopathic bronchiectasis, which may result in increased probability of mismatches between KIR and HLA-C ligand with the potential to alter NK cell regulation and function.
Subjects with Bronchiectasis and Control Subjects The study group consisted of 96 unrelated individuals (mean age, 55 ± 1.4 yr; 32% male) with idiopathic bronchiectasis, all of whom gave written, informed consent. Peripheral blood samples were collected from patients attending the Host Defense Unit at Royal Brompton Hospital (London, UK). Patients seen in this unit are clinically phenotyped according to a detailed protocol of investigations. A diagnosis of idiopathic bronchiectasis is made, where there is predominantly bilateral lower lobe bronchiectasis and chronic rhinosinusitis, on the basis of clinical examination, pulmonary function tests, and high-resolution computed tomography. All patients undergo chest and sinus radiography, high-resolution thin-section computed tomography scan, respiratory function tests, and blood investigations including levels of IgG, IgM, IgA, IgE, and IgG subclasses, and testing for rare immunodeficiencies in selected patients where clinically indicated, Aspergillus radioimmunosorbent test and precipitins, rheumatoid factor, anti-nuclear antibodies, and  1-antiproteinase. Sputum is sent for microscopy, culture, and sensitivities, smear, and culture for acid-fast bacilli. Skin tests are performed for Aspergillus, and a sweat test proceeding to cystic fibrosis genotyping if abnormal. Cystic fibrosis transmembrane regulator genotyping is conducted to detect the following mutations:  F508, G551D, G542X, 621+1G > T, R553X, 1717-G > A, W1282X, N1303K, R117H, R1162X, R334W, and 3849+10kbC > T. These account for about 83% of mutations in the white population in the United Kingdom. Nasal mucociliary clearance and exhaled nasal nitric oxide are measured, proceeding to full cilia studies where indicated. In selected patients, fiberoptic bronchoscopy, barium swallow, respiratory muscle function tests, semen analysis, and tests for associated conditions are also conducted if clinically indicated. Consequently, patients with known underlying causes of bronchiectasis such as cystic fibrosis, immunoglobulin deficiency, and primary ciliary dyskinesia were excluded before a diagnosis of idiopathic bronchiectasis was made. Control subjects consisted of 101 U.K. heart/lung transplant organ donors (mean age, 24.8 ± 1.4 yr; 73% male). The majority of control subjects died unexpectedly after a road traffic accident, head injury, or cerebrovascular event. The study was approved by the Royal Brompton, Harefield, and National Heart and Lung Institute Ethics Committee. The two groups were matched for ethnicity. We acknowledge that they are not fully matched for age and sex (due to the preponderance of females in the patient group). We know of no bias imposed by sex on HLA-C gene frequencies. Relative ages of sample groups can be a confounding factor for a disease in which the most severely affected individuals die at a young age. There is no evidence of such an effect in this disease, where onset is generally in the twenties and thirties.
HLA-C Analysis
KIR Gene Analysis
Statistical Analysis
Individuals with Idiopathic Bronchiectasis Show an Increased Frequency of HLA-Cw*03, whereas the Frequency of HLA-Cw*06 Is Reduced Ninety-six unrelated patients with idiopathic bronchiectasis and 101 control subjects were studied. The frequencies of HLA-C alleles in individuals with bronchiectasis compared with control subjects are shown in Table 1. HLA-Cw*03 was more common in subjects with bronchiectasis compared with control subjects. HLA-Cw*03 was identified in 19.8% of patients, but in only 9.9% of the control subjects (OR, 2.25; 95% CI, 1.25–4.02; p < 0.006). In contrast, HLA-Cw*06 was identified in 4.2% of patients and 14.4% of control subjects (OR, 0.26; 95% CI, 0.12–0.58; p < 0.0005; Figure 1A). The Cw*03 allele was associated with a 2.25-fold increased risk of bronchiectasis and the Cw*06 allele with a 0.26-fold reduced risk.
Increased Frequency of HLA-C Group 1 Alleles in Individuals with Idiopathic Bronchiectasis Dimorphisms in the HLA-C 1 domain that are characterized by Ser-77/Asn-80 and Asn-77/Lys-80 define serologically distinct HLA-C groups termed HLA-C group 1 and HLA-C group 2, respectively (20, 27). HLA-C alleles are defined in terms of group 1 and 2 (Table 2). HLA-C group 1 was more common in subjects with bronchiectasis compared with control subjects. HLA-C group 1 was present in 66% of patients with bronchiectasis compared with 51% in control subjects (OR, 1.88; 95% CI, 1.25–2.82; p < 0.002; Figure 1B). Furthermore, a trend toward significance is seen when HLA-Cw*03 and HLA-Cw*06 alleles are not included in the analysis, excluding the possibility that these alleles are singularly driving the group 1/group 2 effect (Figure 1C).
HLA-C Group 1 Homozygosity in Idiopathic Bronchiectasis Approximately half the patients with bronchiectasis were homozygous for HLA-C group 1 (Figure 2A) compared with one-quarter of the control subjects (OR, 2.56; 95% CI, 1.41–4.64; p < 0.002). HLA-C groups 1 and 2 act as ligands for KIR 2Ds and it is through this interaction that HLA-C molecules protect healthy cells from NK cell–mediated cytolysis. KIR 2DL1 and 2DS1 interact with group 2, whereas KIR 2DL2, 2DL3, and 2DS2 interact with group 1 (Table 2). The specificity for the HLA-C type is defined by a single amino acid substitution at KIR 2D position 44 (21, 27).
Relationship between HLA-C and KIR Haplotype in Bronchiectasis The overall KIR gene frequency for 2DL2, 2DL3, 2DS1, and 2DS2 was not significantly different in patients with idiopathic bronchiectasis compared with control subjects (Table 3).
Individuals only expressing HLA-C group 1 on targets will overall have fewer NK cells under inhibitory receptor control due to the missing inhibitory receptor ligand. This would allow stimulatory KIRs greater impact on NK cell function, enhancing activation (29). We examined the impact of HLA-C group 1 homozygosity in the presence of 2DS1 and/or 2DS2 and found that genotypes with stimulatory KIR expressing only HLA-C group 1 were significantly overrepresented in idiopathic bronchiectasis (33% in patients and 16% in control subjects; OR, 2.733; 95% CI, 1.31–5.70; p < 0.006; Figure 2B). We then explored the relationship between the presence and absence of the activating KIR 2DS1 and 2DS2 in individuals expressing both group 1 and group 2 HLA-C. We determined the combinatorial frequencies of HLA-C group 1 and 2 heterozygotes with 2DS1 and 2DS2. By the criteria defined in the model described by Nelson and colleagues (29), the presence of HLA-C group 1/2 heterozygosity along with KIR 2DS1 and 2DS2 is expected to lie at the most "inhibitory" end of the NK cell activation gradient. This genotype was significantly underrepresented in individuals with bronchiectasis (Figure 2C, upper pair of bars; 19% in control subjects compared with 5% in patients; OR, 0.25; 95% CI, 0.09–0.70; p < 0.005). About one-quarter of the patients with idiopathic bronchiectasis were homozygous for the AA*01 KIR haplotype (defined in this study as the presence of 2DL3 and the absence of 2DL2, 2DS1, and 2DS2; Figure 3A). The relationship between HLA-C group 1 homozygosity and KIR group AA*01 homozygosity in individuals with idiopathic bronchiectasis and control subjects is shown (Figure 3). Thus it is the increased prevalence of HLA-C group 1 homozygosity in idiopathic bronchiectasis that allows a greater proportion of potential mismatches between KIRs and HLA-C ligands present, which may result in altered NK cell regulation.
In idiopathic bronchiectasis chronic bacterial infection and inflammation result in progressive lung damage. Bilateral, predominantly lower lobe bronchiectasis is one of the clinical features of TAP deficiency syndrome (individuals with a mutation in the TAP gene), which can also encompass chronic sinusitis, necrotizing granulomatous skin lesions, and recurrent bacterial pneumonia (18, 19, 32). Mutations preventing expression of either or both TAP1 and TAP2 proteins cause reduced cell surface expression of HLA class I molecules and, as a consequence, expansions of NK and   T cells with increased cytolytic activity are seen (18, 19, 32–35). Although this rare group of patients represents a special case of bronchiectasis, the data indicate that bronchiectasis may be associated with dysregulated NK cell function. The hypothetical framework for our study was thus that idiopathic bronchiectasis, generally considered a disease of unknown etiology, may involve a genetic susceptibility to inappropriate or dysregulated NK cell surveillance of bacterial infection in the lung. The functional impact on NK cell killing of KIR/HLA-C interactions is often analyzed with respect to susceptibility to viruses or tumors, in which mechanisms subverting adaptive immunity through class I down-regulation are a common feature of pathogenesis, leading to a presumed importance of NK cell surveillance. However, bacterial genomes contain, in addition to the well-documented TLR-mediated capacity to up-regulate MHC, gene products capable of specific interference with MHC antigen presentation (36). Individual differences in NK cell interactions are dependent on combinations of variable KIR and HLA class I gene products. Because the two loci segregate independently, NK cells can express KIRs for which there is no known HLA ligand present. As different receptor–ligand interactions may result in altered NK cell–mediated immunity against pathogens, it is proposed that the relationship between these genes may be important in a disease such as bronchiectasis, where chronic bacterial infection and progressive lung damage may be the result of a dysregulated immune response to infectious pathogens and/or self-antigens. To gain insights into whether NK cell function may be more broadly significant in this disease, the key polymorphic ligand/receptor HLA-C and KIR 2D genes were analyzed. We found HLA-Cw*03 to be present more commonly in individuals with idiopathic bronchiectasis, and HLA-Cw*06 less so. The Cw*03 allele was associated with a 2.25-fold increased risk, and the Cw*06 allele with a 0.26-fold reduced risk, of bronchiectasis. More generally, HLA-C group 1 motifs were more common in bronchiectasis, whereas group 2 motifs were less so. HLA-C group 1 homozygosity was associated with markedly increased susceptibility to bronchiectasis. This is noteworthy because a homozygosity effect of this type is not easily reconciled with a simple immune response gene effect whereby a particular HLA class I allele is required for adaptive response to an epitope from a particular pathogen. It is highly reminiscent of several other NK cell disease models, whereby individuals (homozygotes) who are missing ligands for inhibitory receptors will have fewer NK cells under inhibitory control (28–30). An absence of inhibitory receptors KIR 2DL1 and 2DL2/3, in the presence of the respective homologous activating KIR 2DS1 and 2DS2 receptors, has been associated with the HLA-Cw*0602–associated inflammatory disease psoriatic arthritis (28–30). The bronchiectasis group contained a significantly increased number of individuals expressing only HLA-C group 1 with 2DS1 and/or 2DS2 stimulatory KIRs: this would be considered at the "top end" of the activation spectrum for NK cells. At present, such interpretations depend on a model encompassing many unknowns including the relative affinities of inhibitory and activating KIRs, competitive binding, influence of peptide binding and clonotypic expression of receptors. Nevertheless, the implication that will need to be pursued through functional experiments is that an element of the pathogenesis in this disease relates to excessive activation of NK cells. Antigenic peptides can be either permissive or prohibitive to KIR recognition of HLA class I (21, 25, 37–39). This has led to the proposal that a basal set of peptides presented by healthy cells contains permissive peptides, tuned just above the threshold that permits inhibition of NK cells. However, subtle changes in the peptide pool, resulting from, for example, infection, might trigger NK activation (21, 25). This may be particularly relevant for HLA-C molecules, as they are expressed at one-tenth to one-third of the level of HLA-A and HLA-B and are therefore closer to the threshold necessary for NK cell inhibition (39, 40). Using activating KIR tetramer–binding studies, formal dependence on the nature of HLA-C–presented peptides has been demonstrated (25). Furthermore, virally infected cells showed enhanced binding for both stimulatory and inhibitory KIRs (25). The precise, physiological ligands for the activating KIRs are poorly defined. It has been elucidated that binding of HLA-C to activating KIR 2DS1 and 2DS2 would be weaker than to their inhibitory counterparts KIR 2DL1 and 2DL2/3 (23–25). KIR tetramer–binding studies suggest that activating and inhibitory receptors recognize the same set of HLA class I molecules, differing in their binding affinities, such that the stimulatory KIR is not always sufficient to trigger an NK cell response to ligand. This allows fine control during cellular activation (25). There is support for the notion that non-HLA molecules (such as foreign or microbial antigens, aberrantly expressed normal cell surface proteins, or complexes of pathogen-derived peptides bound to MHC class I molecules) may behave as ligands for activating KIRs (28). Furthermore, the presence of an activating KIR (KIR 3DS1) gene along with specific HLA alleles encoding Bw4–80I has an epistatic protective effect on AIDS progression (41). On the basis of gene content, two distinct primary sets of haplotypes have been determined for KIR genes, termed A and B (22). Haplotype A has seven loci: 2DL1, 2DL3, 2DL4, 2DS4, 3DL1, 3DL2, and 3DL3. An important and probably functionally relevant difference between A and B haplotypes is the presence of stimulatory KIR genes. Haplotype B contains various combinations of 2DS1, 2DS2, 2DS3, 2DS5, 3DS1, and 2DS4, whereas haplotype A contains only a single stimulatory KIR gene, 2DS4. The 2DS4 gene has a null allele with a population frequency of about 84% (42). Therefore, most individuals who are homozygous for haplotype A may express no activating KIR (43). The frequency of the two major haplotype groups, A and B, is different across ethnic groups (44–51). For example, the A haplotype allele frequency for Japanese populations is as high as 75% and for Australian aborigines as low as 15% (52, 53). It is noteworthy that the prevalence of childhood bronchiectasis among Australian aborigines is extremely high at 14.7 per 1,000, a rate that is 40-fold greater than in nonindigenous Australian populations (0.35 per 1,000) (52). Furthermore, a chronic, HLA class I–associated lung disease similar to bronchiectasis, diffuse panbronchiolitis, is relatively common in Japanese populations (53). Thus, in two ethnic groups with increased probability of polarized HLA class I/KIR combinations, an unusually high incidence of bronchiectasis-like disease is described. About one-quarter of the patients in this study were homozygous for the KIR A haplotype (defined in this study as the presence of 2DL3 and the absence of 2DL2, 2DS1, and 2DS2). The increased prevalence of HLA-C group 1 allotype homozygosity in bronchiectasis allows a greater proportion of potential mismatches between KIRs and HLA-C ligands, which may impact NK cell regulation. The present study is the first to establish genetic susceptibility in idiopathic bronchiectasis: the association with HLA-C group 1 homozygosity, and the interplay between HLA-C/KIR genes, taken together with findings in TAP deficiency syndrome, are consistent with a role for NK cells in disease pathology. More specifically, the data suggest a role for excessive or inappropriate NK cell activation. Certainly, for this chronic progressive lung disease, for which there is no clear mechanistic hypothesis, the findings suggest that experiments on NK cell function may be informative. More generally, the findings reveal a new level of complexity to our understanding of the relationship between HLA class I and KIR genes and the impact that this has on disease susceptibility. In evolutionary terms it seems that, to benefit from the genetic diversity of this system that allows the generation of an efficient and effective immune response to environmental pathogens, there is a cost: mismatches can occur at an individual or population level, ending in disease through an inappropriately regulated immune response.
The authors thank Professors Malcolm Green, Anthony Newman-Taylor, Tim Evans, Peter Cole, and Maggie Dallman for continued support and advice.
Supported by grants from the Royal Brompton and Harefield NHS Trust Clinical Research Committee, the Welton Foundation, and the Medical Research Council, United Kingdom. R.J.B. is supported by a Medical Research Council Clinician Scientist Fellowship. Originally Published in Press as DOI: 10.1164/rccm.200501-124OC on October 27, 2005 Conflict of Interest Statement: None of the authors have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form January 26, 2005; accepted in final form October 26, 2005
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
2 or Fisher's exact test. p Values less than or equal to 0.01 were regarded as significant and those less than or equal to 0.001 were regarded as highly significant. 
