| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Functional Genomics Unit, Institute of Genomics and Integrative Biology (Mr. Ahsan and Dr. Qadar Pasha), Delhi; Department of Medicine, Sonam Norboo Memorial Hospital (Dr. Ghulam) Leh, Ladakh, J&K, India; Ladakh Heart Foundation and Ladakh Institute of Prevention, (Dr. Norboo) Leh, Ladakh, Jammu and Kashmir, India; and Department of Biochemistry (Dr. Baig) Hamdard University, New Delhi, India.
Correspondence to: M. A. Qadar Pasha, PhD, Functional Genomics Unit, Institute of Genomics and Integrative Biology, Mall Rd, Delhi-110 007, India; e-mail: qpasha{at}igib.res.in
Abstract
Study objectives: High-altitude pulmonary edema (HAPE), which develops on exertion under hypoxic conditions, aggravates due to endothelial dysfunction. Repeat events of the disorder suggests of genetic susceptibility. Endothelial nitric oxide synthase gene (NOS3), a regulator of vasodilation, has emerged as a strong candidate marker. In the present study, we investigated G894T, 27-base-pair 4b/4a (variable number of tandem repeat), –922A/G, and –786T/C polymorphisms of NOS3, individually or in combination, for an association with HAPE.
Design: A cross-sectional case control study.
Settings: Blood samples of HAPE-resistant lowlanders (HAPE-r) were obtained at sea level, and blood samples of patients with HAPE (HAPE-p) were obtained at Sonam Norboo Memorial Hospital, Leh, at 3,500 m.
Participants: The study groups consisted of 60 HAPE-r inducted two to three times to altitudes > 3,600 m; and 72 HAPE-p, who had HAPE on their first visit to high altitude.
Results: Nitrogen oxides (NOx) at 77.9 ± 28.6 µmol/L were significantly elevated in HAPE-r as compared to 42.39 ± 12.93 µmol/L in HAPE-p (p < 0.0001). Genotype distribution of G894T and 4b/4a polymorphisms was significantly different in the two groups (p = 0.001 and 0.009, respectively). Haplotype analysis revealed –922A/G and –786T/C polymorphisms in complete linkage disequilibrium. The wild-type haplotypes G-b (G894T, 4b/4a), G-A (G894T, –922A/G), and G-b-A (G894T, 4b/4a, –922A/G) were significantly overrepresented in HAPE-r (p < 0.0001, p = 0.03, and p = 0.02, respectively). The heterozygote genotype combination GTba as compared to wild-type combination GGbb was significantly higher in HAPE-p (
2 = 18.62, p = 0.00009; odds ratio, 7.20; 95% confidence interval, 2.82 to 18.38). The combination of four heterozygotes GTbaAGTC was overrepresented in HAPE-p (p = 0.04), whereas the wild-type genotype combination GGbbAATT was overrepresented in HAPE-r (p = 0.002). Furthermore, the GGbb combination correlated with significantly elevated NOx as compared to remaining combinations as a whole in both HAPE-r and HAPE-p (p = 0.01 and 0.004, respectively).
Conclusions: Reduced NOx and combination of heterozygotes associate with the susceptibility to HAPE. The study impels another step toward application of NOx as a diagnostic marker for HAPE. The NOS3 GTba and GTbaAGTC genotype combinations may find application as genetic markers for predicting the risk for HAPE.
Key Words: genotype combination • high-altitude hypoxia • high-altitude pulmonary edema • nitric oxide • NOS3 polymorphisms
A significant number of people visit the Himalayas and other high mountains for activities such as sight-seeing, mountaineering, and trekking. In modern times, duty-bound induction (ie, the recruitment of soldiers for duty at high altitudes) has drawn increased attention to high altitude (HA) studies. Most of the sojourners travel to mountains without knowing about adverse health effects such as mountain disorders. HA pulmonary edema (HAPE), which normally occurs at altitudes 
2,500 m, accounts for most of the deaths due to altitude sickness.1 Hypoxia and physical exertion are the main predisposing factors toward the development of HAPE that occurs in unacclimated sojourners.23 However, the precise etiology remains unknown, and there is no reliable diagnostic method for predicting a sojourner who might become ill. Those who have had HAPE run a significant risk of recurrence, and this is one feature that points to the involvement of genetic components in the disorder.4
Advances in understanding the biochemical pathways that alter the physiologic response to hypoxia have led to greater insights into the pathogenesis of HAPE. Polymorphisms within the genes of these pathways may explain the interindividual differences in response to hypoxia5678 and perhaps indicate susceptibility to the disorder. HAPE is characterized by unequal hypoxic pulmonary vasoconstriction and regional overperfusion of capillaries in areas of least arterial vasoconstriction leading to increased capillary pressure and leakage.910 Endothelial dysfunction is an important clinical condition that determines the excessive pulmonary vasoconstriction in HAPE.211 Nitric oxide (NO), also called endothelium-dependent relaxing factor, is the major vasodilator in the capillaries of the endothelium,12 and increased nitrogen oxides (NOx) are involved in the mechanisms of offsetting hypoxia.7 Functional variants in the endothelial NO synthase gene (NOS3) may have the potential to alter the expression of the enzyme,13 thereby accounting for the difference in the NOx. The available literature141516 suggests a positive association of these genetic variants with the susceptibility of various cardiovascular disorders (CVD) that are also demonstrated to be involved with the pathophysiology of NO production; however, their involvement in HAPE has remained ambiguous.171819 Other candidate genes were also tested for an association with HAPE; however, the functionality of not even a single genetic variation could be ascertained.202122 The susceptibility of HAPE could be because of single or several such variants within a gene or even involving with several genes.
In the present case-control study, we investigated G894T, 27-base-pair 4b/4a variable number of tandem repeat, –922A/G, and –786T/C polymorphisms of the NOS3 for an association with HAPE. Endogenous NOx were estimated as a biochemical marker to find a correlation with the individual genotype, haplotypes, and genotype combinations of these polymorphisms.
Materials and Methods
Subjects
The study groups consisted of 60 HAPE-resistant lowlanders (HAPE-r) and 72 patients with HAPE (HAPE-p). The blood samples of HAPE-r were obtained at sea level, whereas those of HAPE-p were obtained at Sonam Norboo Memorial Hospital, Leh, at 3,500 m. The groups were age matched (30 to 40 years) and were unrelated men of the same ethnicity. HAPE-r were healthy individuals who had been recruited for duty two to three times at altitudes > 3,500 m and had performed routine strenuous physical activities without having the disorder; in contrast, the other group had HAPE on their very first visit. We diagnosed HAPE on the basis of standard criteria,423 which included assessment of onset of typical symptoms at HA, including cough and dyspnea at rest, absence of infection, presence of pulmonary rales and cyanosis, disappearance of symptoms and signs within 3 days of the start of treatment with supplemental oxygen, and bed rest. Chest radiographic infiltrates consistent with pulmonary edema confirmed the disorder (refer to supplementary file). After recovery, HAPE-p were examined to exclude any previous cardiopulmonary diseases. Written informed consent of the subjects was obtained before the investigation. The Human Ethics Committee of our institute approved the investigation.
Genotype Analysis
DNA was isolated from peripheral blood leukocytes using a modification of the salting-out procedure.24 Polymerase chain reactions (PCRs) for the four polymorphisms were performed using standard protocols252627; 1.5% agarose gel electrophoresis confirmed the size of the PCR products. The G894T, –922A/G, and –786T/C polymorphisms were screened by restriction fragment length polymorphism using 8 U of Ban II, Bsl I, and Ngo M IV restriction endonucleases, respectively, per 30.0 µL of reaction mixture. The difference in the alleles was resolved by 12% polyacrylamide gel electrophoresis. The 4b/4a variable number of tandem repeat polymorphism, a 27-base-pair difference, was resolved by 3% agarose gel electrophoresis. The PCR products were randomly sequenced to validate the restriction fragment length polymorphism results to avoid any mistyping. For this purpose, the PCR products were isolated by columns (GFX columns; Amersham Biosciences; Piscataway, NJ) and were subjected to cycle sequencing using a dye terminator cycle sequencing kit (377 DNA Sequencer; Perkin-Elmer, Applied Biosystems; Foster City, CA).
Biochemical Marker Estimation
Plasma NOx were estimated by an enzymatic Griess method28 on a high-throughput spectrophotometer (SpectraMax 190; Molecular Devices; Sunnyvale, CA). NO, being a short-lived molecule, rapidly gets converted into nitrite and nitrate. In a two-step method, the nitrate is first converted to nitrite by nitrate reductase (Sigma; St. Louis MO). The total nitrite is then estimated to reflect NOx. Estimations were performed in duplicate, and the results were confirmed by repetition in those samples in which the coefficient of variation was > 5%. Inference was drawn from internal standards of nitrite.
Genotype Combinations
A combination analysis of the genotypes of two, three, and four polymorphisms was performed. The combinations were studied for an association with the disorder and for a correlation with NOx so as to establish functionality in the disorder.
Biostatistical Analysis
Statistical software (Statistical Package for the Social Sciences for Windows, release 10; SPSS; Chicago, IL; and Epi Info 6; Centers for Disease Control and Prevention; Atlanta, GA) was used for statistical analysis. A goodness-of-fit test was used for testing the Hardy-Weinberg (HW) equilibrium, and a 2 test compared the genotype and allele frequencies between the two groups. The odds ratio (OR) and 95% confidence interval (CI) were used as a measure of strength for the association between different genotypic combinations with the disease. Logistic regression analysis was used for relative risk (RR) determination for the alleles under consideration. Haplotype frequencies of the four polymorphisms were estimated and compared by 2 test and permutation analysis (SNP Alyze, version 3.1; Dynacom; Mobara-shi, Japan). To determine the extent of association of the G894T, 4b/4a, –922A/G, and-786T/C, the Lewontin coefficient (D’) and squared correlation coefficient (r2) for pairwise linkage disequilibrium (LD) were calculated (Graphical Overview of Linkage Disequilibrium; Center for Statistical Genetics; Ann Arbor, MI; and SNP Alyze). The biochemical variable was analyzed for normal distribution using a Kolmogrov-Smirnov test. Variables that did not follow normal distribution were considered as log-normal values and were analyzed using Kruskal-Wallis and Mann-Whitney U tests. NOx were expressed as mean ± SD. Differences between the groups were analyzed by one-way analysis of variance (ANOVA) and unpaired t-test with two-tailed values. Where appropriate, p values for pairwise differences were corrected for multiple comparisons using a Bonferroni correction test. A p value < 0.05 was considered statistically significant.
Results
Genotype Distribution and Allele Frequencies
The genotype distribution of each of the four polymorphisms was in HW equilibrium in HAPE-r. In case of HAPE-p, –922A/G and-786T/C polymorphisms were in HW equilibrium, whereas G894T and 4b/4a polymorphisms were not, because of the significant differences between observed and expected likelihood of genotypes (2 = 7.78, p = 0.02 and 2 = 7.62, p = 0.02) [refer to supplementary file]. The G894T and 4b/4a genotype distribution was significantly different between the two groups (p = 0.001 and 0.009, respectively) [Table 1
]. The wild-type genotypes 894GG and 4bb were significantly overrepresented in HAPE-r (42% and 26%, respectively) compared to HAPE-p. In case of G894T polymorphism, the 894G allele was significantly overrepresented in HAPE-r compared to HAPE-p (p = 0.03). The distribution of the genotypes and alleles for –922A/G and –786T/C polymorphisms did not reach significance between the cases and control subjects (Table 1), although the wild-type genotypes –922AA and –786TT were 15% greater in HAPE-r than HAPE-p.
|
|
|
|
2 = 15.67, p = 0.0004), whereas the frequency of the wild-type combination was decreased in HAPE-p and was dominated by the heterozygote combination (2 = 18.62, p = 0.00009) [Fig 2
]. With respect to genotype combinations of four polymorphisms, 10 different combinations were observed (Table 4
). Here again, the wild-type genotype combination GGbbAATT of the four polymorphisms was significantly overrepresented (37% vs 15%) in HAPE-r (2 = 12.58, p = 0.002), whereas the heterozygous GTbaAGTC combination was overrepresented (17% vs 5%) in HAPE-p (2 = 7.35, p = 0.04). It is clearly visible that the wild-type alleles are being more frequently replaced by mutant alleles in HAPE-p. Further, it can be seen from Figure 3
that the homozygous wild-type GGbb combination correlated with significantly elevated NOx as compared to remaining combinations as a whole in both HAPE-r and HAPE-p (p = 0.01 and 0.004, respectively). Furthermore, a comparison between GGbb and GTba specifically revealed noticeable differences in NOx. NOx obtained for HAPE-r were 92.09 ± 22.26 µmol/L for the GGbb combination and 78.34 ± 20.87 µmol/L for the GTba combination. In HAPE-p, the levels were 55.40 ± 9.91 µmol/L for the GGbb combination and 46.20 ± 10.12 µmol/L for the GTba combination. It was evident that the heterozygote combination GTba as compared to the wild-type homozygous combination GGbb correlated with decreased NOx in HAPE-r and HAPE-p; the levels, however, did not reach significance, which could be attributed to fewer number of GTba individuals (Fig 3). The correlation of NOx with the four genotype combinations was in confirmation to two genotype combinations. The NOx differed significantly for the genotype combinations GGbbAATT (p = 0.01), GTbbAATT (p < 0.0001), GTbaAGTC (p < 0.0001), and GGbaAATT (p = 0.02) between HAPE-r and HAPE-p (Table 4). It is obvious from these findings that the presence of any of the four mutant alleles is associated with reduced NOx in both groups.
|
|
|
The large occurrence of HAPE in our subcontinent because of global attraction to the Himalayas, duty-bound induction, and the absence of the knowledge about the predisposing mechanisms emphasize the need to unravel the molecular linings of the disorder as the threat looms large on unacclimated visitors. The thin balance maintained between vasoconstriction and vasodilation under the extreme conditions of hypoxia and cold at HA is nucleus to any impending threat of escalation of these pathologic conditions. Endothelial dysfunction and extra vasoconstriction, the characteristics of HA disorders such as HAPE, are also the clinical features of CVD.11 Therefore, the role of few of the candidates, in the present case NOS3, which are similar in HA and CVDs, is anticipated. Among our findings, the first striking note was the distortion of HW equilibrium in HAPE-p for G894T and 4b/4a polymorphisms, which did not happen in case of HAPE-r. This suggests the possibility of these two loci being the susceptibility loci or the genetic markers in close proximity to the diseased locus/loci. Since the number of samples in HAPE-p is more than HAPE-r, the significance of this finding cannot be ruled out and in fact compelled us to further look for the functionality of these two polymorphisms. We observed significant overrepresentation of the wild-type homozygotes 894GG and 4bb of G894T, 4b/4a polymorphisms in HAPE-r compared to HAPE-p. The overrepresentation of 894T and 4a alleles in HAPE-p may cause a reduction in NO synthesis by NOS3 (a structure-function relationship), thereby contributing to the endothelial dysfunction and pulmonary hypertension. The G894T polymorphism situates itself in exon 7 of the NOS3 and changes an amino-acid Glu at 298th position to Asp, whereas 4b/4a polymorphism being localized in intron 4 does not make any structural change; however, studies1529 have revealed its functionality in posttranscriptional modifications. It has been observed that presence of the 894T allele reduces the expression of NOS3 through increased proteolysis of the enzyme that may contribute to decrement in the NOx,30 which in a given biological environment, such as hypoxia, may initiate and/or aggravate the pathophysiology of a disease state. It is encouraging to note that already reported positive association of the mutant alleles, 894T and 4a, with CVD15 and the overrepresentation of these alleles in HAPE-p support our hypothesis of the involvement of common factors in the pathogenesis of these disorders. The –922G and –786C alleles of the remaining two polymorphisms were marginally greater in HAPE-p than HAPE-r. In earlier studies,1416 these alleles have been observed to be positively associated with the susceptibility of CVD. Although the mutant alleles –922G and –786C were marginally overrepresented in HAPE-p, their being in the promoter region and in complete LD may provide them with additive effects in regulating the expression of NOS3.
Apart from the individual causative effect of these polymorphisms, the other important findings of the present investigation have been the haplotype and genotype combination analyses of these four polymorphisms, which to our knowledge have been performed for the first time in HAPE-p. It was observed that the haplotypes between the two wild-type alleles G-b, G-A, and three wild-type alleles G-b-A were significantly overrepresented in HAPE-r, whereas the mutant allele haplotypes T-a, T-G, a-G, and T-a-G were more prevalent in HAPE-p. The haplotypes between two and three polymorphisms revealed the greater probability of cross-talk between mutant alleles in HAPE-p. The genotype combination analysis further added extra weight to our presumption of their role in the disease. In this context, it would be desirous to refer to our previous findings,7 in which we have reported the significant association of wild-type genotype combination GGbb with HA adaptation (66% and 69% in HA groups vs 44% in lowlanders). In the present study, the GGbb combination as compared to remaining combinations was significantly underrepresented in HAPE-p (16%) than HAPE-r (40%). It is evident that natural selection of GGbb has been dominant during the process of HA adaptation in order to eliminate combinations harboring mutant alleles to restrict the incidences of mountain disorders, in the present case HAPE. This is exactly what we observed by analyzing the two- and four-genotype combinations of G894T, 4b/4a –922A/G, and –786T/C polymorphisms, the significant overrepresentation of the heterozygotes GTba and GTbaAGTC, and underrepresentation of GGbb and GGbbAATT in HAPE-p.
Equally interesting and important are our findings of NOx. The significantly decreased endogenous NOx in HAPE-p as compared to HAPE-r strengthens the presence of endothelial dysfunction in the disorder. The available literature731 suggests the benefits of elevated NO levels for the survival at HA under the systemic hypoxic state of an individual. Inhalation therapy of NO to the HAPE patients improves the localized pulmonary hypoxic state of the disorder.3233 Although NOx were measured after the return of HAPE-r to sea level, if the levels were measured at HA, they would have either remained same or elevated but certainly higher than HAPE-p. The available literature3435 suggests decreased exhaled NO levels in individuals susceptible to HAPE and unchanged levels in HAPE-resistant individuals. We presume the NOx levels in HAPE-r need to be higher as a compensatory mechanism against hypoxia, as have been observed in adapted HA natives.731 In this context, recently we inducted ourselves as a part of a research scheme and accentuated the elevation of NOx on acclimatization (data not shown). Of further note, the correlation of NOx with the genotypes, in the present study, strengthened our belief of a strong genetic basis of the disorder. The significantly elevated NOx in subjects with wild-type GGbb combination and the correlation of reduced NOx with heterozygote combination GTba of G894T and 4b/4a polymorphisms indeed confirmed that the mutant alleles are involved in the pathogenesis of HAPE. Moreover, similar to a two heterozygote combination GTba, the combination of four heterozygotes GTbaAGTC was also found to correlate with decreased NOx in comparison to GGbbAATT, the wild-type homozygote combination. It seems that although the combination of four heterozygotes, GTbaAGTC, associates with the disorder, the two polymorphism genotype combination GTba has a more pronounced impact on the endogenous NO production.
In conclusion, the results indicated the involvement of reduced NOx in HAPE, which sets the stage for the application of NOx as a diagnostic marker for the disease. The overrepresentation of heterozygote combinations GTba and GTbaAGTC in patients associates with the susceptibility of the disease. The genotype findings become stronger due the correlation of genotype combinations of heterozygotes with decreased NOx. The absence of mutant homozygotes of the four polymorphisms in both the groups could be beneficial for HA performance because it is possible that presence of four mutant alleles in an individual could be highly risky, when inclusion of a single mutant allele (heterozygote) is so effective. Bringing some new insights in the genetic predisposition to HAPE, the findings of the present investigation may find application in tourist industry, mountaineering, and even official recruitment. The trends of association of allelic distribution and the correlation with NOx that did not reach significance necessitate larger sample set; nevertheless, since HAPE samples are difficult to arrange, the present study has its own importance.
Acknowledgements
The authors thank Ven. Thubten Choegyal, Chairman, Ladakh Heart Foundation, Leh, and Mr. Mohd. Iqbal, Sonam Norboo Memorial Hospital, Leh, especially in the collection of blood samples during our visits. We highly appreciate the constant encouragement of the Director, Institute of Genomics and Integrative Biology, in our endeavors.
Footnotes
Abbreviations: ANOVA = analysis of variance; CI = confidence interval; CVD = cardiovascular disorders; D' = Lewontin coefficient; HA = high altitude; HAPE = high-altitude pulmonary edema; HAPE-p = patients with HAPE; HAPE-r = HAPE-resistant lowlanders; HW = Hardy-Weinberg; LD = linkage disequilibrium; NO = nitric oxide; NOS3 = endothelial nitric oxide synthase gene; NOx = nitrogen oxides; OR = odds ratio; PCR = polymerase chain reaction; RR = relative risk; SNP = single-nucleotide polymorphism
The Council of Scientific and Industrial Research supported this work.
The authors have no conflicts of interest to disclose.
Received for publication January 2, 2006. Accepted for publication April 18, 2006.
References
C mutation in the 5'-flanking region of the endothelial nitric oxide synthase gene is associated with coronary spasm. Circulation 1999;99,2864-2870C endothelial nitric oxide synthase genotype is a novel risk factor for coronary artery disease in Caucasian patients of the GENICA study. J Am Coll Cardiol 2003;41,930-937
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
