Association of Glutathione-S-Transferase-P1 (GST-P1 ...

6 downloads 71 Views 95KB Size Report
This study was supported by Emory/Egleston Children's Research, NIH R03 HD37674. Division of Neonatal-Perinatal Medicine (M.H.M., T.W.G., L.A.S.B.), ...
Original Article . . . . .

. . . . . . . . .

Association of Glutathione-S-Transferase-P1 (GST-P1) Polymorphisms with Bronchopulmonary Dysplasia Martha H. Manar Milton R. Brown Theresa W. Gauthier Lou Ann S. Brown, PhD

OBJECTIVE: Reactive oxygen species (ROS) contribute to oxidative lung injury. The glutathione-S-transferases (GST) family and microsomal epoxide hydrolase (mEPHx) enzymes detoxify ROS, and genetic polymorphisms alter this detoxification. We hypothesized that polymorphisms encoding for less efficient enzymes were associated with bronchopulmonary dysphasia (BPD).

STUDY DESIGN: We determined allelic distribution of these polymorphisms in a pilot study of 35 BPD cases and 98 controls. w2 and regression analysis were performed. RESULTS: GST P1 val105ile distribution differed between the groups, with the more efficient val/val allele predominately in controls (pr0.05). When controlling for race and sex, BPD cases were less likely to be homozygotes for the val/val isoform (OR 0.21, CI: 0.045–0.95, p ¼ 0.04) and more likely to possess the less efficient ile isoform (OR 4.5, CI: 1.0–20.7, p ¼ 0.05). CONCLUSIONS: This pilot study suggests that BPD is associated with the presence of the GST-P1 105ile allele. Future prospective studies are warranted. Journal of Perinatology (2004) 24, 30–35. doi:10.1038/sj.jp.7211020

INTRODUCTION Despite advances in neonatal care, bronchopulmonary dysplasia (BPD), or chronic lung disease, remains a significant problem for

Division of Neonatal-Perinatal Medicine (M.H.M., T.W.G., L.A.S.B.), Department of Pediatrics, Emory University School of Medicine, Atlanta, GA, USA and Division of Endocrinology (M.R.B.), Department of Pediatrics, Emory University School of Medicine, Atlanta, GA, USA. This study was supported by Emory/Egleston Children’s Research, NIH R03 HD37674. Address correspondence and reprint requests to Lou Ann S. Brown, PhD, Department of Pediatrics, Emory University, Atlanta, GA 30322, USA

premature infants, particularly those of extremely low birth weight. Infants with BPD have an increased potential for morbidity and mortality in the first 2 years of life, with significant respiratory and functional impairment throughout childhood.1 The incidence of BPD (defined by oxygen requirement at 36 weeks postmenstrual age) was recently reported to be 30% for infants with birth weights less than 1,000 grams.2 With improved survival in the extremely low birth weight population, increasing numbers of infants are at risk for developing BPD. There are few reports of the role of genetic factors in the development of BPD.3 The results of one study of low birth weight twins suggested a genetic susceptibility to BPD.4 BPD pathogenesis is influenced by a number of factors, including oxidative stress. Although reactive oxygen species (ROS) are an integral part of cell-signaling pathways and gene induction, oxidant stress and resultant ROS are key components of inflammation and tissue injury.5–8 Normal host antioxidant defenses include a number of cytoprotective enzymes that serve to reduce xenobiotics, ROS, and coordinate the inflammatory cascade. Glutathione-S-transferase enzymes (GST), particularly the pi, mu, and theta subclasses, are responsible for reducing DNA and lipid oxidation products in the lung, and modulating the synthesis of eicosanoids and other inflammatory mediators.9 Similarly, microsomal epoxide hydrolase (mEPHx) metabolizes toxic epoxide intermediates.10 Individuals vary in their ability to withstand oxidative stress, perhaps as a result of differences in the activities of these enzymes,9 which are part of a complex system of cell defenses against toxins and oxidation. Recent evidence suggests that genetic polymorphisms of these cytoprotective enzymes influence the development of many diverse diseases.10–14 The genes for the GST pi (GST-P1) subclass, as well as mEPHx, have genetic variations called polymorphisms (single base pair changes) within the general population that alter the amino-acid sequence of the enzyme and its enzymatic activity.5–8,15,16 The GST-P1 polymorphism at codon 105 encodes either a val or ile (val105ile). This variance resides within the enzymatic cleft and influences the efficiency of the enzyme activity. In vitro, the 105val isoform has increased efficiency for reducing oxidative toxins when compared to the 105ile isoform.15 This suggests that subjects with the 105ile isoform would have decreased ROS clearance and increased tissue injury. Indeed, there is increased frequency of the 105ile isoform in populations with decreased lung function growth, increased the risk or severity of Journal of Perinatology 2004; 24:30–35 r 2004 Nature Publishing Group All rights reserved. 0743-8346/04 $25

30

www.nature.com/jp

GST-P1 with Bronchopulmonary Dysplasia

respiratory infections in school-aged children, increased airway reactivity, asthma, and COPD.9,12,13,17,18 Alternatively, the polymorphisms can result in a null allele with no formation of the enzyme product. Such polymorphisms exist for the mu and theta subclasses of GST (GST-M1 and GST-T1, respectively).19 Therefore, individuals that are ‘‘double-null’’ homozygotes have no GST-M1 or T1 enzymes. Not surprisingly, these polymorphisms are associated with increased frequencies in several pathologies including decreased childhood lung function,17 increased childhood respiratory infections,18 and childhood bronchial asthma.20 For mEPHx, the 113his isoform has been called the ‘‘slow’’ allele because its catalysis of bioreduction reactions is less efficient than that of the 113tyr isoform. Studies have documented lower frequencies of the mEPHx 113tyr isoform in populations with emphysema and COPD and correspondingly higher frequencies of the 113his isoform.8,14 We designed a case–control pilot study to investigate the genotype frequencies of the GST family and mEPHx polymorphisms in BPD patients. We hypothesized that the alleles encoding the less efficient isoforms (GST-P1 115ile and mEPHx 113his) and the null alleles of GST-M1 and GST-T1 would have increased frequency in the BPD population when compared to race- and sex-matched controls. The purpose of this study was to provide preliminary data to support further studies investigating a role for polymorphisms in antioxidant defenses as contributors to the development of BPD.

MATERIALS AND METHODS Sample Collection Eligible BPD cases (infants with radiographic changes and oxygen requirement at 36 weeks postconceptional age) were recruited from the NICU and the developmental clinic of Grady Hospital in Atlanta, GA. To control for differences in polymorphic distributions due to gender and ethnic groups,19,21 eligible controls were term or near-term infants without respiratory distress and matched to BPD cases by sex and race. No sibling pairs were enrolled. The study was reviewed and approved by the Human Investigations Committee at Emory University School of Medicine and informed consent obtained. Genotyping A buccal swab was obtained from each infant, air-dried and stored at room temperature. Using an BuccalAmp DNA Extraction Kit (Epicentre; Madison, WI), DNA (2 to 5 ng/ml) was isolated from each sample according to the directions of the manufacturer. DNA was successfully isolated from 100% of cases and 93% of controls. DNA was genotyped using the polymerase chain reaction (PCR). GST-P1 and MEPHx were classified by restriction fragment length polymorphisms (RFLP). Each 5 ml PCR reaction contained 2 to 5 ng DNA, thermophilic buffer (10 mM Tris-HCl and 50 mM KCl in Journal of Perinatology 2004; 24:30–35

Manar et al.

0.1% Triton-X 100), 3.0 mM MgCl, 200 mM of each dNTP, 0.6 mM primers, and 0.125 U Taq polymerase. Polymorphism of GST-P1 The GST-P1 polymorphism was detected as described by Harries et al.22 The DNA was denatured at 941C for 3 min, underwent 30 cycles of denaturation (941C for 30 seconds), annealing (551C for 30 seconds), and extension (721C for 30 seconds), and a final extension step at 721C for 2 minutes. BsmA1 was used to digest the PCR products at 551C for 1 hour. This resulted in cleavage of the 105val sequence into 91 and 85 bp fragments but left the 105ile sequence intact (176 bp). The digest products were then separated on 8% acrylamide/8% glycerol gels and stained with silver nitrate. RFLP products were classified as Ile-Ile, Ile-Val, or Val-Val. Multiplex PCR for GST-M1, T1, and b-globulin This assay was performed using three primer pairs for GST-M1, GST-T1, and b-globulin.19 The PCR consisted of 941C for 2 minutes, 35 cycles of 941C for 10 seconds, 581C for 20 seconds, 721C for 45 seconds, and then a final step of 721C for 5 minutes.12 The products were separated on 8% acrylamide/8% glycerol gels as above. Products were classified as either homozygous null or heterozygous/homozygous M1 (215 bp) or T1 (480 bp). b-globulin (268 bp) served as an internal control. The absence of a M1 or T1 PCR product in the presence of a b-globulin product was scored as a null allele. Polymorphism of MEPHx The mEPHx tyr113his polymorphism assay was detected as described.11 The PCR cycling program was 941C for 2 minutes, then 35 cycles of 941C for 30 seconds, 461C for 30 seconds, and 721C for 20 seconds, and a final extension at 721C for 5 minutes. The PCR products were digested with EcoRV at 371C for 1 hour. The 113his product remained uncleaved (162 bp). The 113tyr PCR product was cleaved into 140 and 22 bp fragments. The digest products were separated on 10% acrylamide/8% glycerol gels as above. RFLP products were classified as Tyr-Tyr, Tyr-His, or HisHis. Statistics Statistical analysis was performed using SPSS for Windows. Twotailed w2 analysis were used with Yates continuity correction when appropriate. Multivariate regression analysis was also performed controlling for race and sex.

RESULTS Demographics In this pilot study, 98 controls and 35 cases were enrolled (Table 1). The race was divided into three groups: African American, Caucasian and Other (including Native African, Middle Eastern, Indian, and Hispanic). The majority of the patients evaluated were 31

Manar et al.

GST-P1 with Bronchopulmonary Dysplasia

Case

p-value

83/98 (B85%) 5/98 (B5%) 10/98 (B10%)

23/35 (B66%) 5/35 (B14%) 7/35 (B20%)

p=0.05

51/98 (B52%) 47/98 (B48%) 39.62 (±1.89)

22/35 (B63%) 13/35 (B37%) 26.62 (±2.87)

p=0.36

of African-American descent, reflective of our patient population in the Grady Healthcare System. The racial distribution between the two groups reached statistical significance (p ¼ 0.05). The male/ female ratio in the BPD cases was B2:1. There was no difference in the sex distribution between cases and control. Per study design, the estimated gestational age of cases was significantly lower than controls (p