Abstract
Background Aspirin-exacerbated respiratory disease (AERD) is nonatopic asthma, and the role of heat shock protein (HSP) 70 in AERD remains unknown. We analyzed HSP70 gene polymorphisms in Japanese patients with AERD.
Methods The single-nucleotide polymorphisms in HSPA1B-179C>T and 1267A>G gene were examined in patients with AERD and those with aspirin-tolerant asthma (ATA). All patients were in a stable condition.
Results There were significant differences in total serum IgE levels, peripheral blood eosinophil count, and prevalence of atopy between AERD and ATA. The patients with AERD showed higher frequencies of the CT/TT genotype of the HSPA1B-179C>T than that of the CC genotype compared to ATA (P < 0.001). They showed higher frequencies of the GG genotype of the HSPA1B1267A>G than that of the GA/AA genotype compared to ATA (P < 0.001). These differences were irrespective of the sex for the genotypes analyzed. The frequency of HSPA1B-179C/1267A haplotype was significantly higher in AERD compared to ATA (P < 0.001; odds ratio, 3.154; 95% confidence interval, 1.916-5.193). Among the clinical and hematological characteristics investigated, AERD showed a significant variance in peripheral blood eosinophil count according to the association of the 2 HSP70 gene polymorphisms (P = 0.033), but not in ATA.
Conclusions Our findings first suggest that the association between HSPA1B-179C>T and 1267A>G gene sequence variations might be implicated in the development of AERD.
Heat shock protein (HSP), of which the HSP70 family is best understood, responds to a variety of stressful stimuli by augmentation of intracellular HSP gene expression1and subsequent inhibition of proinflammatory cellular functions.2Expression of HSP70 in leukocytes of patients with sepsis has been demonstrated, indicating that HSP70 may play a role in the outcome of the patients.3
There are 3 genes in the HSP70 family, HSPA1A, HSPA1B, and HSPA1L, located adjacent to each other in the class III region of the major histocompatibility complex (chromosome 6p21.3).4Nucleotide sequence analysis of the 2 intronless genes, HSPA1A and HSPA1B, encode an identical protein.5Both genes are expressed at high level in cells after heat shock, with HSPA1A also expressed constitutively at very low levels.4The HSPA1L gene is expressed at low levels both constitutively and after heat shock.6
A prospective cohort study of community-acquired pneumonia found that carriage of the AA homozygotes of the HSPA1B1267A>G gene was associated with a significantly greater risk of developing septic shock.7As HSPA1B1267A>G is a silent mutation, it is likely that another polymorphic site is responsible for the changes in biological function that explain the disease association. In fact, HSPA1B1267A>G and HSPA1B-179C>T were found to be in linkage disequilibrium.7Temple et al.8investigated the promoter region of HSPA1A and HSPA1B in healthy whites and Asians and demonstrated that HSPA1B-179C>T is in linkage disequilibrium with HSPA1B1267A>G and HSPA1B-179C>T affects HSP70 production, suggesting HSPA1B-179C>T as a key determinant of individual susceptibility to a variety of inflammatory diseases. The data were subanalyzed by race, and the same associations were observed in whites and Asians. They also suggested that the HSPA1B-179C>T:1276A>G haplotype is functional and may explain the association of the HSP70 gene with development of septic shock.9
Aspirin-exacerbated respiratory disease (AERD) is a complex clinical syndrome as recommended.10,11Here, we analyzed patients characterized by severe asthmatic attack after taking aspirin and/or nonsteroidal anti-inflammatory drugs. Genetic predisposition has been considered as a crucial determinant, and candidate genes have concentrated especially on cysteinyl leukotriene–related genes,12–15as the inhibitory action of aspirin and NSAIDs on cyclooxygenase activity may cause overproduction of cysteinyl leukotrienes. However, conflicting results have been reported,16,17and Higashi et al.18demonstrated that prostaglandin D2 was overproduced during aspirin-intolerant bronchoconstriction. So far, no worldly used assay for aspirin intolerance is available.19–22Alteration in the levels of cyclooxygenase and lipoxygenase products were published demonstrating elevated levels as well as no changes in blood from patients with AERD and those with aspirin-tolerant asthma (ATA),23indicating that the other genetic determinants remain to be identified. The recent investigations from our laboratory showed that the gene polymorphisms of such as Arg16Gly β2-adrenergic receptor, cytochrome P450 2C19, and IL-13 in patients with AERD were different from those in patients with ATA.24–27
Several studies have shown that HSP70 is overexpressed in the bronchi of asthmatic patients, correlating with intrapulmonary eosinophilia and the severity.28,29Autoprotective role of HSP70 expression in bronchial asthma has thus been suggested29; however, it also has been proposed that HSP70 expression might play a deleterious role in asthma.30On the other hand, Aron et al.31reported that no evidence was found for an independent role of HSP70 gene polymorphism in susceptibility to allergic asthma.
To our knowledge, no studies have evaluated its association with AERD. Therefore, taking all into account, we hypothesize that HSP70 gene polymorphism might be involved in the susceptibility of AERD, and we have expanded our studies in this paper. This is the first study analyzing HSPA1B-179C>T and 1267A>G gene polymorphisms in patients with AERD, and demonstrating its possible influence on the susceptibility to AERD.
MATERIALS AND METHODS
Subjects and Clinical Assessment
This study was performed with the approval of the Institutional Ethics Committee of Gunma Institute for Allergy and Asthma, Gunma Hospital for Allergic and Respiratory Diseases, and written informed consent was obtained from each individual before the study commenced. All subjects were nonsmoking Japanese and were recruited from the outpatient clinics of Gunma Hospital for Allergic and Respiratory Diseases, Yukawa Clinic of Internal Medicine, and Hiroshima Allergy and Respiratory Clinic in Japan.
Diagnosis of bronchial asthma was confirmed by using the Global Initiative for Asthma guidelines, and all patients were given a diagnosis by experienced pulmonologists. Airway reversibility was defined as a greater than 12% and greater than 200-mL increase in forced expiratory volume in 1 second (FEV1) from baseline after inhalation of short-acting β2-adrenergic bronchodilators. The diagnosis of AERD was made on the basis of either a positive result on lysine-aspirin challenge test11or apparent history of more than one self-reported episode of bronchial response to aspirin or NSAID ingestion. The provocation test could not be applied to the subjects who did not give written informed consent. Patients with AERD combined with aspirin-intolerant urticaria were excluded from this study. Aspirin-tolerant asthma was defined as bronchial asthma with no history of NSAID-induced asthma attack. No lysine-aspirin challenge test was performed in patients with ATA. The patients consisted of 102 patients with AERD and 300 patients with ATA as shown in Table 1. All patients were in a stable clinical condition and had not been treated with oral corticosteroids. A total of 100 nonsmoking subjects with no history of bronchial asthma or other respiratory symptoms were selected from healthy volunteers and comprised normal controls. Pulmonary function test and hematological test were not performed in the normal controls. The serum levels of total immunoglobulin E (IgE) and specific house dust mite IgE antibody were analyzed by the Pharmacia CAP system (Phadia, Uppsala, Sweden), and the patients with positive scores for the antibody were classified as patients with atopy. The sensitization was defined as a specific IgE level of 0.35 IU/mL or greater with a detection limit of 0.1 IU/mL as described.32,33The total eosinophil count was measured in peripheral blood using a flow cytometer (Coulter Maxm; Beckman-Coulter Inc, Fullerton, CA). The presence of rhinosinusitis and nasal polyps could not be determined because of the absence of otolaryngologists at the hospital who were specialized in using paranasal sinus x-ray and rhinoscopy.
HSPA1B-179C>T and HSPA1B1267A>G Polymorphism Genotyping
DNA in the specimens obtained by rubbing buccal mucosa by a cotton swab was extracted by using QIAamp 96 DNA blood kits (Qiagen, Hilden, Germany). The target DNA sequence of the HSPA1B-179C>T was amplified using a set of primers (forward: 5′-AAAGGCGGGTCTCCACGAC-3′, reverse: 5′-GTTCGCGCTCTGGAAAGCCTTG-3′) and that of 1267A>G was amplified using a set of primers (forward: 5′-ACAAGGCCCAGATTCACG-3′, reverse: 5′-GTTTTCCAGCTTGCATGTCC-3′). The target DNA sequence of the IL-17A -737C>T was amplified using a set of primers (forward: 5′-CCCCCATCATGTCTCCTCTCC-3′, reverse: 5′-GGGTTGATGCTCTTGTTCAG-3′). Allelic discrimination assay for single-nucleotide polymorphisms (SNPs) relating to the expressions of HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) was carried out by an SNP detective system as described.18–21All subjects and investigators remained unaware of the genotype until the final analysis.
Statistical Analysis
Data are presented as means ± SD or numbers (%) of observations, unless stated otherwise. Differences in the mean value of the phenotypic characteristics within the groups were compared using either analysis of variance test or t test, and qualitative data were compared by the χ2 test. Allele frequencies were estimated by gene counting method. Significant departures of genotype frequency from the Hardy-Weinberg equilibrium at each SNP were tested by the χ2 analysis. Differences in minor allele frequencies of HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) in the patients with AERD and the control subjects were compared with those of the patients with ATA by the χ2 test. Logistic regression analysis was used to estimate odds ratio (OR) and 95% confidence interval (CI). Each gene polymorphism related to the asthma phenotype was examined by multivariable logistic regression models with adjustment for covariates, namely, with the asthma phenotype as dependent variable and independent variables including age (continuous value), sex (men, 0; and women, 1), 2 alternative genotype models that were combined heterozygous CT genotype group and homozygous TT genotype in HSPA1B-179C>T (rs6457452), and homozygous GG genotype and combined heterozygous GA genotype and homozygous TT genotype group in HSPA1B1267A>G (rs1061581). In addition, subgroup analyses with sex of the multivariable logistic regression analysis were performed in each gene, and the interactions between sex and the genotype were tested using the Wald statistic. Differences in the clinical characteristics according to the association of the HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) gene polymorphisms in the patients were compared by the F-test, and qualitative data were compared by the χ2 test. Linkage disequilibrium (LD) evaluated by D′ coefficient was calculated to evaluate LD by the Haploview version 4.0 software program. Haplotype frequency analysis with multivariate adjustment for age and sex was determined using the H-Plus haplotype version 2.5 software program.34,35Statistical analyses were undertaken using SPSS for Windows version 17 (SPSS Inc, Chicago, IL). Our study has to be considered exploratory in nature. Therefore, no correction for multiple testing was carried out. P < 0.05 was considered to be significant.
RESULTS
Patients’ Characteristics
All patients were in a stable clinical condition, and there was no significant difference between the patients with AERD patients and those with ATA with regard to age, sex, and FEV1 (percent predicted). The levels of total serum IgE in the patients with AERD were lower than those in the patients with ATA (P < 0.001). The patients with AERD had a higher peripheral blood total eosinophil count compared to the patients with ATA (P < 0.001). A significant difference was found in the prevalence of atopy between the patients with AERD and those with ATA (P < 0.001) (Table 1).
Genotype of HSPA1B-179C>T and HSPA1B1267A>G (rs6457452 and rs1061581)
The genotype distribution fulfills the Hardy-Weinberg equilibrium in each group (Table 2). The frequencies of the T allele of the HSPA1B-179C>T (rs6457452) genotype was the same in the patients with ATA and the control subjects (frequency of allele [q] = 0.030), whereas the frequency in the patients with AERD was increased (q = 0.181). The frequency of the T allele in the patients with AERD was higher than that in the patients with ATA (P < 0.001). On the other hand, the frequencies of the G allele of the HSPA1B1267A>G (rs1061581) genotype was similar in the patients with AERD and the control subjects (q = 0.706 and q = 0.650, respectively), whereas the frequency in the patients with ATA was decreased (q = 0.535). The frequency of the G allele in the patients with AERD was higher than that in the patients with ATA (P < 0.001).
Table 3 (A) presents the results of multivariable logistic regression analysis of the HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) genotype controlling age and sex in the patients with AERD compared with those in the patients with ATA. The patients with AERD demonstrated higher frequencies of combined heterozygous CT and homozygous TT genotype group of the HSPA1B-179C>T (rs6457452) than those of homozygous CC genotype when compared to the patients with ATA (P < 0.001). Comparing the patients with AERD patients and those with ATA, the OR of the combined CT and TT genotype group compared to the CC genotype was 7.527 (95% CI, 3.933–14.407). The patients with AERD demonstrated higher frequencies of homozygous GG genotype of the HSPA1B1267A>G (rs1061581) than those of combined GA and AA genotype group when compared to the patients with ATA (P < 0.001). Comparing the patients with AERD and those with ATA, the OR of homozygous GG genotype compared with the combined GA and AA genotype group was 3.126 (95% CI, 1.953–5.001).
Subgroup analyses with sex showed the positive association between asthma phenotype and the HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) genotype both in men and women as shown in Table 3 (B). The patients with AERD demonstrated higher frequencies of the combined CT and TT genotype group of the HSPA1B-179C>T (rs6457452) than those of the CC genotype compared with the patients with ATA (P = 0.001; OR, 8.500; 95% CI, 2.297–31.45 in men, and P < 0.001; OR, 7.382; 95% CI, 3.459–15.754 in women). The interaction between sex and the genotype of the HSPA1B-179C>T (rs6457452) was not significant (P = 0.910). The patients with AERD demonstrated higher frequencies of the homozygous GG genotype of the HSPA1B1267A>G (rs1061581) than those of the combined GA and AA genotype group compared with the patients with ATA (P = 0.022; OR, 2.758; 95% CI, 1.161–6.550 in men, and P < 0.001; OR, 3.308; 95% CI, 1.887–5.796 in women). The interaction between sex and the genotype of the HSPA1B1267A>G (rs1061581) was not significant (P = 0.915).
Haplotype of HSPA1B-179C>T and HSPA1B1267A>G (rs6457452 and rs1061581)
The 2 SNPs in the HSP70, HSPA1B-179C>T and HSP1267A>G (rs6457452 and rs1061581), have been shown to be in the LD with each other in the normal controls, and there was no significant difference between the white and Asian groups.8,9Based on the LD coefficient (D′ = 1.000) between the 2 genotype SNPs (rs6457452 and rs1061581) in the normal controls in the present study, we inferred the haplotype frequencies. The prevalence of haplotype [C-A] was significantly higher in the patients with AERD than in the patients with ATA (P < 0.001; OR, 3.154; 95% CI, 1.916–5.193; Table 4).
Hematological Characteristics According to the Association of HSPA1B-179C>T and HSPA1B1267A>G (rs6457452 and rs1061581) Gene Polymorphisms
Among the hematological characteristics investigated earlier, their association with HSPA1B-179C>T and HSP1267A>G (rs6457452 and rs1061581) gene polymorphisms was analyzed by the F-test, and a significant variance in peripheral blood total eosinophil count was shown to be present in the patients with AERD (P = 0.033) but not in the patients with ATA (Table 5). However, there was no significant variance in total IgE levels and the prevalence of atopy both in the patients with AERD and those with ATA.
DISCUSSION
Genetic predisposition has been considered as a crucial determinant in patients with AERD, and candidate genes have concentrated on leukotriene-related genes.12–15However, conflicting results have been reported.16,17Aron et al.31suggested that HSP70 overexpression in asthma was independent on HSP gene polymorphisms, and Smith et al.36reported that HSPA1A and HSPA1B do not share common patterns of polymorphisms. In this study, we analyzed HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) gene polymorphisms in patients with AERD.
In this study, the patients with AERD and those with ATA revealed comparable distribution of age and sex. Then, we assessed the clinical and hematological characteristics of the subjects. All patients were in a stable clinical condition, and there was no difference between the 2 groups in FEV1. There were significant differences between the patients with AERD and those with ATA with regard to the levels of total serum IgE, peripheral blood total eosinophil count, and the prevalence of atopy.
Next, we investigated the frequencies of the HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) genotype in the patients with AERD and those with ATA. This is the first study that demonstrated that both in male and female patients with AERD, the frequencies of the combined CT and TT genotype group of the HSPA1B-179C>T (rs6457452) was higher than those of the CC genotype compared to the patients with ATA. In addition, we showed that both in male and female patients with AERD, the frequencies of the GG genotype of the HSPA1B1267A>G (rs1061581) was higher than those of the combined GA and AA genotype group compared to the patients with ATA.
HSPA1B1267A>G is a silent polymorphism,37and it is likely that another polymorphic site is involved in the biological function that explains the disease association. In fact, it has been reported that HSPA1B-179C>T is in LD with HSPA1B1267A>G and the A allele of HSPA1B1267 is in linkage with the C allele of the HSPA1B179, which is associated with lower levels of gene expression,8suggesting that HSPA1B gene polymorphism is one of key determinants of individual susceptibility to a variety of infectious and inflammatory diseases. A prospective cohort study of community-acquired pneumonia found that carriage of the AA homozygotes of HSPA1B1267A>G gene was associated with a significantly greater risk of developing septic shock.7
The 2 SNPs in the HSP70, HSPA1B-179C>T and HSP1267A>G, have been shown to be in the LD with each other in the normal controls,8,9which corresponds to the result in this study. Therefore, we investigated the frequencies of the HSPA1B-179C>T and 1267A>G (rs6457452 and rs1061581) haplotype and demonstrated that the prevalence of haplotype [C-A] was significantly higher in the patients with AERD than in those with ATA.
Among the hematological characteristics investigated, a significant elevation in peripheral blood total eosinophil count was present in the patients with AERD according to the association of HSPA1B-179C>T and HSP1267A>G (rs6457452 and rs1061581) gene polymorphisms, but not in the patients with ATA.
Aspirin-exacerbated respiratory disease is known to be associated with higher peripheral blood eosinophil count than ATA,38,39which corresponds to the results in this study. This study demonstrated that the prevalence of haplotype [C-A] was significantly higher in patients with AERD than in patients with ATA. Whereas studies investigating the levels of translated protein still need to be performed, the A allele of HSPA1B1267 has been shown to be in linkage with the C allele of the HSPA1B179, which is associated with lower levels of HSP70 gene expression.8Therefore, a lower production of intracellular HSP70 may have a minimal effect on inhibiting proinflammatory cellular functions potentially involved in AERD. The present study provides evidence that a significant elevated peripheral blood total eosinophil count according to the association of the 2 SNPs in the patients with AERD, but not in the patients with ATA, suggesting that the association with HSPA1B-179C>T and 1267A>G gene polymorphisms may be involved in the susceptibility to AERD.
The present study has certain limitations as reported.24–27The diagnosis of AERD was made based on either a positive result on lysine-aspirin challenge test13or an apparent history of more than one self-reported episode of bronchial response to aspirin or NSAID ingestion as recommended by the European Network for Drug Allergy.40The suspected drugs were not applied to the subjects in this study because of patients’ refusal and our safety policy of inpatient provocations only.41Ideally, everyone in the study should have been challenged to rule in/out bronchial hypersensitivity to aspirin/NSAIDs. However, a good number of the patients with AERD were recruited from the outpatient clinics of the experienced pulmonologists in Japan. Finally, this work did not describe the mechanisms linking the HSP70 gene genotype and a significant elevated peripheral blood total eosinophil count according to the association of the 2 SNPs in the patients with AERD but not in the patients with ATA. However, we can propose several possibilities.
Although studies in cultured cells have demonstrated that NSAIDs can potentiate heat-induced HSP70 expession,42the use of NSAIDs has been recommended to be carefully monitored in cancer patients undergoing hyperthermic treatment.43,44Mortaz et al.45demonstrated that NSAIDs induced HSP70 from bone marrow–derived mast cells, which was closely paralleled with inhibition of tumor necrosis factor (TNF) production. They also demonstrated that aspirin-induced release of HSP70 from mast cells results in cell activation through Toll-like receptor pathway.46Interestingly, Kee et al.9found that individuals with the haplotype containing the sepsis-associated genotype, HSPA1B-179*C:HSPA1B1267*A, have decreased expression of HSP70 in mononuclear cells and increased production of TNF. Tumor necrosis factor is a well-known proinflammatory cytokine released from inflammatory cells including mast cells47,48and is increased in asthmatic airways.49,50As the HSP genes lie in the major histocompatibility complex class III region,4it is possible that linkage of HSPA1B-179C>T with other polymorphisms in this region, and the adjacent TNF genes may account for some of the functional associations. On the other hand, many papers, including that by Higashi et al.,18have shown that mast cells may be involved in the pathogenesis of AERD,51–53suggesting a role of mast cells in AERD through aspects of HSP70.54Interestingly, Zander et al.55reported that protein microassay analysis of nasal polyps from aspirin-sensitive patients with chronic rhinosinusitis showed a greater expression of HSP70 than that from aspirin-tolerant patients.
In conclusion, we were the first to analyze the HSP70 gene polymorphisms in Japanese patients with AERD and to show that the association between HSPA1B-179C>T and HSPA1B1267A>G gene sequence variations might be implicated in the development of AERD. Our data suggest that HSP70 gene polymorphism profiles might be a useful diagnostic tool in assessment of the susceptibility of subgroup of patients with AERD with severe asthma without further comorbidities. The findings of this study are based on small size in a Japanese population, and to conduct further validation studies in independent populations are required to provide reassurance about the findings in this study.
ACKNOWLEDGMENTS
The authors thank the assistance in the accumulation of samples by Jiroh Okusawa, MD; Masayuki Hashimoto, MD; Junichiro Morioka, MD; and Hiroaki Inamura, MD, PhD, at the Gunma Institute for Allergy and Asthma, Gunma Hospital for Allergic and Respiratory Diseases, Ohra-machi, Gunma, Japan.