Elevated levels of mannan-binding leptin (MBL) and eosinophilia in patients of bronchial asthma with allergic rhinitis and allergic bronchopulmonary aspergillosis associate with a novel intronic polymorphism in MBL

Authors

Errata

This article is corrected by:

  1. Errata: Corrigendum Volume 144, Issue 3, 552, Article first published online: 22 May 2006

Dr Taruna Madan, Molecular Biochemistry and Diagnostics, Institute of Genomics and Integrative Biology, Mall Road, Delhi-110007, India.
E-mail: taruna@igib.res.in

Summary

Mannan-binding lectin (MBL), an important component of innate immunity, binds to a range of foreign antigens and initiates the lectin complement pathway. Earlier studies have reported high plasma MBL levels in allergic patients in comparison to healthy controls. In view of varied plasma MBL levels being determined by genetic polymorphisms in its collagen region, we investigated the association of single nucleotide polymorphisms (SNPs) in the collagen region of human MBL with respiratory allergic diseases. The study groups comprised patients of bronchial asthma with allergic rhinitis (n = 49) and allergic bronchopulmonary aspergillosis (APBA) (n = 11) and unrelated age-matched healthy controls of Indian origin (n = 84). A novel intronic SNP, G1011A of MBL, showed a significant association with both the patient groups in comparison to the controls (P < 0·01). Patients homozygous for the 1011A allele showed significantly higher plasma MBL levels and activity than those homozygous for the 1011G allele (P < 0·05). The 1011A allele also showed a significant correlation with high peripheral blood eosinophilia (P < 0·05) and low forced expiratory volume in 1 s (FEV1) (P < 0·05) of the patients. We conclude that the 1011A allele of MBL may contribute to elevated plasma MBL levels and activity and to increased severity of the disease markers in patients of bronchial asthma with allergic rhinitis and ABPA.

Introduction

The incidence of respiratory allergic diseases such as bronchial asthma, allergic rhinitis and allergic bronchopulmonary aspergillosis is rising globally, including in developing countries such as India [1]. The International Study of Asthma and Allergies in Childhood (ISAAC) questionnaire-based survey and the European Community Respiratory Health Survey (ECRHS) indicate that the prevalence of asthma in India ranges from 3·5% to 17%, depending on the methodology and definition of asthma used [1]. As high as 89% and 94% of asthmatic adolescents and asthmatic adults, respectively, show the simultaneous presence of allergic rhinitis in the lower airways [2]. About 7·6% of patients of persistent asthma show an aggressive pulmonary allergic response to the allergens/antigens of Aspergillus fumigatus (Afu), a fungal pathogen, and are categorized as allergic bronchopulmonary aspergillosis (ABPA) patients [3,4]. Bronchial asthma with allergic rhinitis and ABPA are complex genetic disorders with a heterogeneous phenotype, attributed largely to the interactions among many genes and between these genes and the environment. Genome screens and association studies have led to the identification of several polymorphisms in different candidate genes to be associated with asthma and related phenotypes [5]. Similarly, polymorphisms in several candidate genes such as cystic fibrosis trans-membrane regulatory proteins (CFTR) and human leucocyte antigen antigen-D related (HLA-DR) have been correlated with susceptibility to ABPA [6–9]. The identification of a comprehensive set of predisposing candidate genes may clarify some of the aetiopathogenic aspects of allergic diseases and improve their prophylaxis and therapy.

Recent insights into the complex mechanisms of asthma and related diseases suggest that the innate immunity genes may be relevant candidates for polymorphism screening [10]. Eder et al. [11] suggested that genetic variation in Toll-like receptor-2 (TLR2) is a major determinant of the susceptibility to asthma and allergies. Novel single nucleotide polymorphisms (SNPs) in the gene for lung surfactant protein-A-2, a C-type lectin with an important role in host defence against Afu, were observed to be associated with ABPA patients in the Indian population [12].

Mannan-binding lectin (MBL), another C-type lectin, constitutes an important effector molecule of innate immunity in the serum [13]. It binds allergens/antigens, activates the lectin complement pathway and modulates the levels of proinflammatory cytokines [14–17]. Varied levels of MBL in the serum have been associated with protection and/or susceptibility to several infectious and non-infectious diseases [13]. Earlier studies have reported elevated plasma MBL levels and complement activity in allergic patients [18,19]. These high plasma MBL levels could be a result of the disease process and/or they may be determined genetically. In our earlier studies, we observed that in a murine model of Afu hypersensitivity (4-week sensitization with Afu allergens) with immune responses characteristic of human ABPA there was an increase in the plasma mouse MBL-A levels [18]. Because genetic polymorphisms in the collagen region of the MBL gene (MBL) are known to contribute to varied plasma MBL levels and activity [13], in the present study we screened SNPs in exon 1 and intron 1 (encoding the collagen region) of MBL in patients of bronchial asthma with allergic rhinitis and ABPA.

Methods

Study population

A case–control association study was conducted following approval of the Human Ethics Committee of Vallabhbhai Patel Chest Institute and Institute of Genomics and Integrative Biology. A 10 ml blood sample was collected from Indian individuals (n = 144), which included allergic patients (n = 60) in two categories: patients with bronchial asthma and allergic rhinitis (n = 49) and patients with ABPA (n = 11) enrolled at Vallabhbhai Patel Chest Institute and unrelated, age-matched healthy controls of similar ethnicity without any history of asthma or ABPA (n = 84). The age of the patients was in the range 12–42 years and that of the controls was in the range of 15–48 years. The asthma patients were included if they were aged 12 years and above, had never smoked and if asthma was confirmed by demonstrating a reversibility of at least 12% and an increase of 200 ml in forced expiratory volume in 1 s (FEV1) after inhalation of 200 µg of salbutamol from a metered-dose inhaler [20]. All patients with bronchial asthma had a positive skin-prick test against at least one of the tested local allergens, which included dust mite allergens, fungal allergens, dog and cat allergens and tree pollens. The skin test was considered positive if the orthogonal wheal diameter was at least 3 mm greater than the negative control and at least half the histamine control. A diagnosis of ABPA in the asthma patients was confirmed according to the criteria proposed by Rosenberg et al. [21]. The range of peripheral blood eosinophilia was 2–12% in patients with bronchial asthma and allergic rhinitis, 2–20% in ABPA patients and 0–5% in controls. The range of percentage-predicted FEV1 was 33–86% in the patients with bronchial asthma and allergic rhinitis, 24–86% in ABPA patients and 87–98% in controls. Informed written consent was obtained from all the participants.

Genomic DNA isolation

Genomic DNA was extracted from human peripheral blood mononuclear cells by the salting-out procedure [22].

Polymerase chain reaction (PCR)

Samples were screened for SNPs in exon 1 and intron 1 of MBL. A stretch of 417 base pairs (bp) from position 618–1034 of MBL (gene position as per accession no. AF08058), comprising the whole of exon 1 (256 bp) and 133 bp of 662 bp of intron 1, was amplified by PCR using primers and PCR conditions as reported previously [23].

Purification of PCR products and sequencing

Gel-purified PCR products were sequenced using gel terminator chemistry on an ABI Prism 3100 automated DNA sequencer (Applied Biosystems, Foster City, CA, USA). The sequencing data obtained were confirmed twice, by performing a complete repeat of the experimental procedure that included amplification of stock genomic DNA, fragment purification and sequencing of both DNA strands. Polymorphisms in the sequences obtained were identified using Basic Local Alignment Search Tool (blast) software located at the National Center for Biotechnology Information (NCBI) website, by comparing with the MBL sequence (accession no. AF080508).

Human plasma MBL levels and MBL-induced complement activity

To evaluate the functional effects of SNP G1011A, MBL levels and complement activity were estimated in some of the controls (n = 40) and some of the patients with bronchial asthma and allergic rhinitis (n = 49) and ABPA (n = 11). C4b deposited onto mannan bound by activated MBL is a measure of MBL-induced complement activity in the plasma. MBL levels and C4b deposition by activated MBL in the plasma samples were measured as described previously [18]. Purified human MBL was prepared as described previously [24]. The MBL functional activity assay was standardized against 1 µg/ml of MBL standard, added to plasma with known MBL content (500 ng/ml, wild-type/wild-type). The MBL standard was assigned arbitrarily the value of 1 unit of C4 deposition activity per microlitre (1 U C4/µl).

Statistical analysis

Fisher's exact test and Pearson's χ2 test were used to compare allelic frequencies between the patient groups and controls. Consistency of genotype frequencies with the Hardy–Weinberg equilibrium was tested using a χ2 test on a contingency table of observed versus predicted genotype frequencies. The values of χ2, odds ratio (OR) and P were calculated at 95% confidence interval (CI). Student's t-test was applied to analyse the difference between the genotype and peripheral blood eosinophil counts, and percentage-predicted FEV1 in asthma patients. A P value of less than 0·05 was considered to indicate statistical significance. Bonferroni's corrections were applied for multiple comparisons.

Results

SNP screening in exon 1 and intron 1 of MBL

Five SNPs were observed in exon 1 and intron 1 of MBL in 144 individuals of Indian origin (Table 1). As well as the three reported structural polymorphisms in exon 1 of MBL (C868T, G875A and G884A), we identified two novel polymorphisms, A816G, in codon 34 of exon 1 and G1011A in intron 1 of MBL (submitted to NCBI GenBank under accession nos ss12535136 and ss46532912, respectively). Within all study groups, the genotype distributions of the observed SNPs were consistent with Hardy–Weinberg equilibrium. Table 1 describes the genotype and allele frequencies of these SNPs in patients and controls.

Table 1.  Distribution of genotype and allele frequencies in exon 1 and intron 1 of mannan-binding lectin (MBL) in patient groups and controls.
SNPa Location AA position AA changeStudy groupsGenotype ne (%frequency)Allele n (%frequency)χ2 (2 d.f.f, 1 d.f.)Pg value (2 d.f., 1 d.f.)
  1. ABPA: allergic bronchopulmonary aspergillosis patients; BAAR: bronchial asthma with allergic rhinitis patients; single nucleotide polymorphisms: SNPs. aNucleotide position of MBL single nucleotide polymorphisms (SNPs) is according to NCBI accession no. AF080508; bnovel SNPs identified in the study;cbronchial asthma with allergic rhinitis patients;dallergic bronchopulmonary patients;eno. of individuals;fdegrees of freedom;gP-values have been calculated using the actual number of individuals in each group by Fisher's exact test ( d.f. = 1) or by online χ2 test using a 2 × 3 contingency table ( d.f. = 2);*Significant P-values (P < 0·05) of the patient groups versus controls after Bonferroni's corrections

A816Gb AAAGGGAG  
 Exon 1BAARc (49)29 (59)20 (40)0 (0) 78 (80)20 (20)0·33, 0·40P < 1, P < 1
 Codon 34ABPAd (11) 9 (81) 2 (18)0 (0) 20 (9) 2 (91)1·04, 0·95P < 1, P < 1
 Ala to AlaControls (84)55 (65)29 (35)0 (0)139 (83)29 (17)  
C868T CCCTTTCT  
 Exon 1BAAR (49)48 (98) 1 (2)0 (0) 97 (99) 1 (1)0·01, 0·01P < 1, P < 1
 Codon 52ABPA (11)11 (100) 0 (0)0 (0) 22 (100) 0 (0)0·23, 0·27P < 1, P < 1
 Arg to CysControls (84)82 (98) 2 (2)0 (0)166 (99) 2 (1)  
G875A GGGAAAGA  
 Exon 1BAAR (49)43 (87) 6 (13)0 (0) 92 (94) 6 (6)1·14, 1·59P < 1, P < 1
 Codon 54ABPA (11) 8 (73) 3 (27)0 (0) 19 (86) 3 (14)0·17, 0·16P < 1, P < 1
 Gly to AspControls (84)66 (79)18 (21)0 (0)150 (89)18 (11)  
G884A GGGAAAGA  
 Exon 1BAAR (49)39 (80)10 (2)0 (0) 88 (90)10 (10)0·19, 0·26P < 1, P < 1
 Codon 57ABPA (11)10 (90) 1 (1)0 (0) 21 (95) 1 (5)0·37, 0·38P < 1, P < 1
 Gly to GluControls (84)70 (83)14 (17)0 (0)154 (92)14 (8)  
G1011Ab GGGAAAGA  
 Intron 1BAAR (49)23 (46)21 (42)5 (1) 67 (68)31 (32)26·8, 29·29P < 0·003*, P < 0·002*
ABPA (11) 5 (45) 4 (36)2 (18) 14 (64) 8 (36)20·6, 19·2P < 0·003*, P < 0·002*
Controls (84)73 (87)11 (13)0 (0)157 (93)11 (7)  

Association of the 1011A allele with allergic patients

A comparison of the allele frequencies of G1011A SNP between the two patient groups and controls revealed that the A allele of the novel intronic G1011A SNP was significantly more frequent in patients with bronchial asthma and allergic rhinitis (χ2 = 29·29, P = 0·00, OR = 6·60, 3·13 < OR < 13·91) and ABPA (χ2 = 19·21, P = 0·00, OR = 8·15, 2·8194 < OR < 23·593) than controls (Table 1). A comparison of the genotype frequencies revealed that there was a significant increase of the GA genotype in the patient groups compared to controls (P < 0·001). The AA genotype was present only in the patients and completely absent in the controls.

There was no significant difference in the allele frequencies and genotype distributions of any of the three reported structural polymorphisms (C868T, G875A, G884A) in exon 1 of MBL between the patient groups and the controls. None of the individuals in the present study was homozygous for the three structural mutations of MBL. The novel A816G SNP, which involved the wobble base of codon 34, also did not show any significant association with the disease.

Association of MBL genotypes at SNP G1011A with MBL levels and activity

We studied the functional relevance of the intronic SNP G1011A by measuring the MBL levels and complement activity in some of the patients (n = 60) and controls (n = 40). The individuals homozygous for the A allele at position 1011 of MBL showed significantly higher mean MBL level and mean MBL-induced complement activity than the individuals homozygous for the G allele (P < 0·05, Fig. 1).

Figure 1.

Bar diagrams comparing (a) mannan-binding lectin (MBL) levels (µg/ml) and (b) MBL-induced complement activity (U C4/µl) in different study groups stratified according to the MBL genotypes at position 1011, with mean values indicated by lines and standard deviations by whiskers. *Significant P-values (P < 0·05) after Bonferroni's corrections. Individuals with the GG genotype were significantly different from those with GA and AA genotypes in their MBL levels and MBL-induced complement activity. The AA genotype was completely absent in the controls. As a whole, MBL levels and activity in the patient groups were significantly higher in the patient groups in comparison to controls (P < 0·05). BAAR: bronchial asthma with allergic rhinitis patients (n = 49). ABPA: allergic bronchopulmonary aspergillosis patients (n = 11).

Association of MBL genotypes at SNP G1011A with peripheral blood eosinophilia and percentage-predicted FEV1

Allergic patients carrying the A/A genotype at SNP G1011A had significantly higher peripheral blood eosinophilia than patients with the G/G genotype (P < 0·05). Also, the percentage-predicted FEV1 was significantly lower in patients with the A/A genotype than patients with the G/G genotype (P < 0·05) (Table 2).

Table 2.  Mean percentage of peripheral blood eosinophilia and percentage-predicted forced expiratory volume in 1 s (FEV1) of patients in patients stratified according to mannan-binding lectin (MBL) genotype at 1011.
PatientsncMean % eosinophild (s.d.e)Mean percentage FEV1g (s.d.)Pf-valuePf-value
  1. ABPA: allergic bronchopulmonary aspergillosis patients; BAAR: bronchial asthma with allergic rhinitis patients. aBronchial asthma with allergic rhinitis patients;ballergic bronchopulmonary aspergillosis patients;cno. of individuals;dpercentage if peripheral blood eosinophilia;estandard deviation;fP-values have been calculated using online Student's t-test;gpercentage-predicted forced expiratory volume in 1 s (FEV1); *significant P-values (P < 0·05) of GA and AA genotypes versus the GG genotype after Bonferroni's corrections.

BAARa
 GG21 4·66 (2·24) 82·97 (13·54) 
 GA18 7·66 (2·70)0·001*72·8 (14·4)0·04*
 AA 512·4 (3·57)0·02*61·2 (9·70)0·008*
ABPAb
 GG 5 4·4 (2·5) 68·8 (18) 
 GA 4 9·5 (3·31)0·0667·25 (14·86)1·6
 AA 217 (2·24)0·04*29 (7·07)0·006*

Discussion

In the present study, a significant association of the 1011A allele of a G1011A polymorphism in intron 1 of MBL with patients of bronchial asthma with allergic rhinitis and ABPA suggests a role for the 1011A allele in genetic predisposition to respiratory allergic diseases. A higher severity of the two clinical markers, namely, high peripheral blood eosinophil counts and low FEV1 in allergic patients homozygous for the 1011A (AA) than in patients with alternative genotypes (GG and GA), indicates further a probable contribution of this allele in the pathogenesis of asthma and allergy.

The presence of the 1011A allele was associated significantly with elevated plasma MBL levels and activity in allergic patients. We believe that this is the first report of the intronic SNP, G1011A in MBL. Being an intronic SNP it may be involved in regulating MBL expression, possibly by affecting the splicing mechanisms. One of the MBL haplotypes, HYP, resulting from three polymorphic sites (H/l, X/Y and P/Q) in the promoter region of MBL, associates with high plasma MBL levels [25]. Hence, it is likely that this intronic SNP may be in linkage disequilibrium with the promoter polymorphisms in MBL. The discovery of a novel SNP correlating with high MBL levels may provide a useful tool to further studies on the association of MBL levels and disease.

Increased activation of the MBL-mediated complement pathway has been associated previously with various pathological conditions. For example, a role of high MBL levels and complement activity has been suggested in rheumatic heart disease, chronic renal failure and diabetic nephropathy in type 1 diabetes [26–28]. Like other complement pathways, the MBL-mediated lectin complement pathway, after the activation of C4 to C4b, culminates in the production of opsonins such as C3b, membrane attack complex (MAC) and complement activation products such as C3a and C5a anaphylatoxins [29]. The interaction of these complement fragments with their corresponding receptors on basophils provides critical immunomodulatory signals to the adaptive immune system, resulting in the production of interleukin (IL)-4, IL-13 and the cysteinyl leukotrienes, leading to potent bronchoconstriction and development of allergic inflammation [28]. It has been shown that plasma C3a levels are elevated in asthmatic subjects [30]. ABPA patients also show elevated levels of complement factors that contribute to lung damage [31]. Increased MBL levels and activity in allergic patients attributed to the observed intronic polymorphism may contribute to increased complement activation via the lectin pathway, and hence to an increased severity of allergic markers such as an increase in blood eosinophil counts. Also, complement activation MBL may lead directly to the development of allergy by affecting the levels of proinflammatory cytokines [32]. A recent study on animal models of Afu-induced asthma by Hogaboam et al. [32] showed that in Afu-sensitized murine MBL-A-deficient mice whole lung T helper cell type 2 cytokine levels were decreased significantly, and whole lung interferon-gamma levels were increased significantly when compared with normal mice, indicating that the murine MBL-A contributes to the development and maintenance of airway hyperresponsiveness.

As well as the novel intronic SNP, the three exon 1 SNPs [G875A (B allele), G884A (C allele) and C868T (D allele)] at codons 54, 57 and 52, respectively, leading to low plasma MBL levels that have been reported in other populations including the Indian population, were also present in our study population [13,33–36]. The B allele frequency in the controls of the present study (0·11) is similar to that reported earlier for Indian control individuals (0·11) and comparable to that reported for Caucasians, Chinese and Eskimos (0·11–0·17) [33,34]. The frequency of the D allele in our study (0·01) is also comparable with the earlier frequency reported in Indian individuals (0·04) and to that of Caucasians (0·05) and Africans (0·05) [33–36]. Also, the frequency of the C allele in the present study (0·08) is comparable to that reported earlier in Indian individuals (0·03) and also to Caucasians (0·02), but lower than in Africans (0·23–0·29) [35]. The presence of polymorphisms in MBL contributing to high as well as low levels of MBL associated with varied immune functioning and affecting susceptibility to various diseases suggests that the MBL system represents a balanced polymorphism. For example, individuals with polymorphisms leading to low plasma MBL levels and complement activity may have an increased susceptibility to infectious diseases, but may be less likely to develop complement-mediated diseases such as allergic diseases.

Mannan-binding lectin has been documented as a weak acute-phase reactant and its plasma levels increase by up to threefold in response to environmental stimuli such as infection or surgery [13]. Earlier studies have reported increased plasma MBL levels and activity in allergic patients [18,19]. High MBL levels in allergic patients also associate with their increased blood eosinophils, which are systemic components of airway inflammation in asthma, and these high levels are sustained even in remission [18,19]. In view of these observations and that of the present study it may be proposed that, in response to an environmental insult, MBL levels may rise in all the individuals, but in those individuals who have elevated plasma MBL levels due inherently to the presence of gene polymorphisms, MBL levels may rise even further and contribute to the increased severity of allergic diseases.

To summarize, the present study infers that the 1011A allele of the intronic SNP G1011A of MBL may lead to high plasma MBL levels and complement activity which contribute plausibly to the increased severity of clinical markers in respiratory allergic diseases. The study emphasizes the need for exploring the precise role of MBL and its associated complement activity in asthma and allergy.

Acknowledgements

This work was supported by the Council for Scientific and Industrial Research (COR0011). We are grateful to the Functional Genomics Unit, Institute for Genomics and Integrative Biology for the sequencing of PCR products. We acknowledge Dr Puneet Khanna, Dr Chandramani and Dr Vikas Maurya, Vallabhbhai Patel Chest Institute, for their help in collection of patient samples. We acknowledge the efforts of Dr Soren Hansen, Department of Cell Biology, Duke University and Medical Center, Durham, for initiation of our collaboration with Professor Steffen Thiel. Ms Savneet Kaur is the recipient of a Senior Research Fellowship of the Council of Scientific and Industrial Research.

Ancillary