UBC Faculty Research and Publications

Oxidative modification of albumin in the parenchymal lung tissue of current smokers with chronic obstructive… Hackett, Tillie L; Scarci, Marco; Zheng, Lu; Tan, Wan; Treasure, Tom; Warner, Jane A Dec 22, 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12931_2010_Article_1023.pdf [ 458.29kB ]
JSON: 52383-1.0223451.json
JSON-LD: 52383-1.0223451-ld.json
RDF/XML (Pretty): 52383-1.0223451-rdf.xml
RDF/JSON: 52383-1.0223451-rdf.json
Turtle: 52383-1.0223451-turtle.txt
N-Triples: 52383-1.0223451-rdf-ntriples.txt
Original Record: 52383-1.0223451-source.json
Full Text

Full Text

RESEARCH Open AccessOxidative modification of albumin in theparenchymal lung tissue of current smokers withchronic obstructive pulmonary diseaseTillie L Hackett1,2*, Marco Scarci3, Lu Zheng2, Wan Tan2, Tom Treasure3, Jane A Warner1AbstractBackground: There is accumulating evidence that oxidative stress plays an important role in the pathophysiologyof chronic obstructive pulmonary disease (COPD). One current hypothesis is that the increased oxidant burden inthese patients is not adequately counterbalanced by the lung antioxidant systems.Objective: To determine the levels of oxidised human serum albumin (HSA) in COPD lung explants and the effectof oxidation on HSA degradation using an ex vivo lung explant model.Methods: Parenchymal lung tissue was obtained from 38 patients (15F/23M) undergoing lung resection andstratified by smoking history and disease using the GOLD guidelines and the lower limit of normal for FEV1/FVCratio. Lung tissue was homogenised and analysed by ELISA for total levels of HSA and carbonylated HSA. Todetermine oxidised HSA degradation lung tissue explants were incubated with either 200 μg/ml HSA or oxidisedHSA and supernatants collected at 1, 2, 4, 6, and 24 h and analysed for HSA using ELISA and immunoblot.Results: When stratified by disease, lung tissue from GOLD II (median = 38.2 μg/ml) and GOLD I (median = 48.4μg/ml) patients had lower levels of HSA compared to patients with normal lung function (median = 71.9 μg/ml, P< 0.05). In addition the number of carbonyl residues, which is a measure of oxidation was elevated in GOLD I andII tissue compared to individuals with normal lung function (P < 0.05). When analysing smoking status currentsmokers had lower levels of HSA (median = 43.3 μg/ml, P < 0.05) compared to ex smokers (median = 71.9 μg/ml)and non-smokers (median = 71.2 μg/ml) and significantly greater number of carbonyl residues per HSA molecule(P < 0.05). When incubated with either HSA or oxidised HSA lung tissue explants rapidly degraded the oxidisedHSA but not unmodified HSA (P < 0.05).Conclusion: We report on a reliable methodology for measuring levels of oxidised HSA in human lung tissue andcell culture supernatant. We propose that differences in the levels of oxidised HSA within lung tissue from COPDpatients and current smokers provides further evidence for an oxidant/antioxidant imbalance and has importantbiological implications for the disease.BackgroundThere is accumulating evidence that oxidative stressplays an important role in the pathophysiology ofchronic obstructive pulmonary disease (COPD) (1). Inparticular, studies have demonstrated elevated oxidativestress is associated with both severity of disease and epi-sodes of exacerbation (2). The elevated oxidative stressin these patients is thought to result both directly frominhaled oxidants in cigarette smoke or pollution andindirectly due to the release of reactive oxygen species(ROS) generated by various inflammatory, immune andepithelial cells (3). One current hypothesis is that theincreased oxidant burden in these patients is not ade-quately counterbalanced by the lung antioxidant sys-tems, leading to enhanced pro-inflammatory geneexpression and protein release, inactivation of antipro-teinases, and as a consequence oxidative tissue injury.The antioxidants present in serum, airway mucosa,alveolar lining fluid and cells include mucin, superoxide* Correspondence: Tillie.Hackett@hli.ubc.ca1School of Medicine, University of Southampton, Southampton, UKFull list of author information is available at the end of the articleHackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180© 2010 Hackett et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative CommonsAttribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction inany medium, provided the original work is properly cited.dismutase, glutathione, uric acid, ascorbic acid, andalbumin. Human serum albumin (HSA) is a single non-glycosylated polypeptide containing 35 cysteine residuesall involved in the formation of stabilising disulphidebonds except 34cysteine. In plasma, this free thiol groupis quantitatively the most important scavenger of oxi-dants (4-6), and is thus an important antioxidant withinthe body(7).The formation of carbonyl groups on amino acid resi-dues as a result of free radical-initiated reactions is welldocumented as a marker of protein degradation andturnover (8, 9). In fact the oxidative modification of pro-teins and lipids has been implicated in the etiology of anumber of diseases including atherogenesis and diabetes(10, 11). In particular oxidised HSA is a reliable markerof oxidative stress in patients with chronic renal failureand individuals on hemodialysis therapy (12). In light ofthese findings the quantification of carbonyl residuesmay provide further evidence to support a role of oxida-tive stress in COPD pathology. There are several meth-odologies for the quantification of carbonyl residues; inthe majority of them 2,4-dinitrophenyl hydrazine isallowed to react with the protein carbonyls to form thecorresponding hydrazone, which can be analysed opti-cally by radioactive counting or immunohistochemistry.In this study we have adapted a previously publishedmethodology based on ELISA to analyse the levels ofcarbonylated HSA in human lung tissue from COPDpatients (13). In addition, we have investigated the effectof oxidation on HSA degradation within human lungtissue explants.MethodsPatient characteristics for human lung tissue experimentsParenchymal lung tissue from the normal margin sur-rounding the tumour site was obtained from 38 patients(15F/23M) undergoing resection for carcinoma at Guy’sHospital London. The study was approved by the StThomas’ Hospital Research Ethics committee, referencenumber EC01/047, and all volunteers gave their signedinformed consent. The Global Initiative for ChronicObstructive Pulmonary Disease (GOLD) guidelines wereused to stratify patients with COPD by disease severitybased on measurements of airflow limitation duringforced expiration (14, 15). Each stage is determined bythe volume of air that can be forcibly exhaled in onesecond (FEV1) and by the ratio of FEV1 to the forcedvital capacity (FVC); lower stages indicate less severedisease. Using the GOLD guidelines our patient cohortwas stratified into the following groups, GOLD I (FEV1/FVC < 70%, FEV1 ≥ 80% predicted), GOLD II (FEV1/FVC < 70%, 50% ≤ FEV1 < 80% predicted) and indivi-duals with normal lung function (FEV1/FVC > 70%,FEV1 ≥ 90% predicted). Table 1 shows the number ofpatients in each GOLD stage and their demographicswhich include age, gender, lung function and smokinghistory. The patient cohort was also reclassified usingthe prediction equations from the National Health andNutrition Examination Survey (NHANES) III (16) fromthe United States and the Health Survey for England(HSE) (17) to determine the lower limit of normal(LLN) for FEV1/FVC. This analysis was performed usingSPSS 14.0 for Windows (SPSS, Chicago, Illinois, USA),data are given in Table 2. For the purposes of this studyex-smokers were defined as that had given up smokingfor ≥3 years to ensure for smoking cessation. AllTable 1 Patient characteristics of subjects prior to theremoval of lung tissueClassification Normal LungFunctionFEV1/FVC > 70%FEV1 ≥ 90%predictedGOLD IFEV1/FVC ≤70%FEV1 ≥ 80%PredictedGOLD IIFEV1/FVC ≤70%50% ≤ FEV1 <80%PredictedNo. subjects 16 13 9Age 64.7 ± 14.1 68.2 ± 9.9 64.3 ± 12.3Gender 6F 7F 2F10M 6M 7MPre-bronchodilatorFEV1/FVC0.78 ± 0.08 0.62 ± 0.04 0.53 ± 0.1Smokingstatus6 current smokers 5 currentsmokers7 currentsmokers8 ex-smokers 5 ex-smokers 2 ex-smokers2 non-smokers 3 non-smokersTissue samples were taken from 38 patients. Patient details including age,gender, lung function given as the ratio of air that can be forcibly exhaled inone second (FEV1) to the forced vital capacity (FVC) pre-bronchodilator useand smoking status. Data given are the mean ± SD of each group.Table 2 Reclassification of subjects using lower limit ofnormal FEV1/FVC to define COPDClassification Normal LungFunctionGOLD I GOLD IINo. subjects 12 8 11Age 63.3 ± 4.7 71.4 ± 2.3 62.5 ± 10.6Gender 4F 4F 4F8M 4M 7MHeight (M) 1.71 ± 0.01 1.68 ± 0.03 1.74 ± 0.1Weight (Kg) 81.0 ± 4.4 67.29 ± 4.3 82.3 ± 9.7LLN FEV1predicted0.91 ± 0.1 0.87 ± 0.02 0.65 ± 0.3Smoking status 6 current smokers 4 currentsmokers7 currentsmokers5 ex-smokers 3 ex-smokers 2 ex-smokers2 non-smokers 2 non-smokersTissue samples were taken from 31 patients. Patient details including age,gender, height, weight, lung function given as lower limit of normal (LLN) ofair that can be forcibly exhaled in one second (FEV1) and smoking status.Data given are the mean ± SD of each group.Hackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 2 of 10demography data was available up to the date of surgeryand none of the subjects were treated with inhaled ororal corticosteroids or bronchodilators.Preparation of human lung tissue for primary cell cultureLung tissue was finely chopped using dissection scissorsinto fragments during several washes with Tyrode’s buf-fer containing 0.1% sodium bicarbonate. 5-6 explants(total weight approx. 30 mg) were incubated in a 24 wellplate with RPMI-1650 medium containing 1% penicillin,1% streptomycin and, 1% gentamycin at 37°C in 5% car-bon dioxide/air for 16 hours (18). Tissue was then eitherincubated with 200 μg/ml HSA or oxidised HSA andlung tissue and supernatant were harvested at 1, 2, 4, 6,and 24 hour time points, weighed and stored at -80°C.Human Serum Albumin ELISAFor measuring total levels of HSA in samples we devel-oped a specific ELISA assay. Briefly, a 96 well plate wasincubated with 14 ng/ml of rabbit HSA antibody incoating buffer at 4°C for 6 hours. Following incubation,the plate was washed and incubated overnight withPBS-Tween containing 5% milk. The following day theplate was washed again and a HSA standard curve(1.5-1000 μg/ml) and samples were added and incubatedat 4C for 2 hours. Following incubation, the plate waswashed and a rabbit anti-HSA antibody conjugated toHRP was added at a concentration of 130 ng/ml for 2hours before a final wash. The plate was developed withthe HRP substrate system (TMB), the reaction stoppedwith 1 M H2SO4 and optical density read at 450 nm.The limit of detection for this protocol was 0.3 ng/ml.Oxidation and derivatisation of the HSA and humantissueA stock solution of 30 mg/ml of HSA was oxidised withequal volumes of 9% hydrogen peroxide and incubatedat room temperature for 30 mins. 100 μl of the oxidisedHSA was then derivatised with 100 μl of 10 mM DNPHin trifluroacetic acid and 100 μl of H2O. Samples werethen incubated at room temperature for 45 mins, withvortexing every 10-15 mins. Derivatised protein wasthen precipitated on ice with 10% trichloroacetic acidfor 30 mins. Following which the sample was centri-fuged at 15,000 g for 5 mins and the supernatantremoved. The pellet was then washed 3 times with 100μl of ethanol/ethyl acetate (1:1) and then allowed to dry.Finally the pellet was broken up with sonication and re-suspended in 0.5 mls of 6 M guanidine hydrochloride in0.5 M potassium phosphate (pH 2.5). The A375 was thenmeasured and the carbonyl content of the oxidised HSAstandard was then determined using ε375 22,000M-1 cm-1(8). For baseline human tissue all samples were deriva-tised using the method described above.Carbonylated human serum albumin ELISATo measure total levels of oxidised human serum albu-min we adapted a previously published method used tomeasure total carbonylated protein (13). Briefly, a 96well plate was incubated with 10 ng/ml of mouse anti-HSA antibody in coating buffer at 4°C for 6 hours. Fol-lowing incubation, the plate was washed and incubatedovernight with 0.1% PBS-Tween containing 5% soyamilk. Following the overnight block, plates were washedand a derivatised HSA standard curve (0.04 - 45.4 μg/ml)and derivatised samples added and incubated at 4°C for2 hours. Following the incubation with samples, the platewas washed and incubated with 1:5000 rabbit anti-dinitrophenyl (DNP) antibody, which had a specific anti-body concentration of 1.0 - 1.7 μg/μl, for 2 hours at 4°C.Finally after washing, the plate was coated with 60 ng/mlof anti-rabbit HRP conjugate for 2 hours at 4°C. Theplate was developed with TMB, the reaction stoppedwith 1 M H2SO4 and optical density read at 450 nm. Thelimit of detection for this was 0.02 ng/ml.ImmunoblotSamples were separated by electrophoresis on 10% SDS-polyacrylamide electrophoresis gels. The proteins weretransferred to a nitrocellulose membrane (Bio-Rad) andblocked overnight with 20% milk. Blots were incubatedwith 1:1000 peroxidase conjugated anti-human albuminantibody (DAKO, Denmark) or 1:1000 anti-DNP anti-body (Sigma, UK). Sites of antibody binding were visua-lised by Super signal west (Pierce, UK).Bicinchonic acid (BCA) assayTotal protein levels of lung homogenates were measuredusing a commercially available BCA assay from BioRadusing a Human Serum Albumin (HSA) standard curve.Limit of detection for HSA was 4 μg/ml.Lactate dehydrogenase assayLDH levels were measured in lung supernatant using acommercially available assay and LDH standard (0.9 -2000 pg/ml) from Roche (Indianapolis IN, USA). Tostandardize for the maximum concentration of LDHpresent tissue was homogenised on ice using a sonicatorset at amplitude of 2 microns; for 12 cycles of 10 sec-onds sonication followed by 20 seconds rest. Followingsonication samples were centrifuged at 15,000 g for 15minutes at 4°C, and supernatant removed for storage.The limit of detection of the assay was 0.5 pg/ml.Statistical analysisStatistical analyses of results were carried out using Stat-view software™. The non-parametric Kruskal Wallis testwas used to analyse all of the data except for the paireddata where Non-parametric Wilcoxon Signed RankHackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 3 of 10analysis was carried out. P < 0.05 was considered assignificant.Multivariate linear regressions for COPD and non-COPD were performed to test for associations withHSA and carbonylated HSA. Confounding factorsincluded for analyses of age, gender, COPD defined as(FEV1/FVC < 70%; FEV1 ≤ 80% predicted) and smokingstatus using Statistica software™. COPD by smokinginteractions were tested in the study by adding a multi-plicative term to the regression models.ResultsRelationship between baseline levels of human serumalbumin and GOLD I & IIParenchymal lung tissue from 38 individuals categorisedas GOLD I (mild), II (moderate) or patients with no evi-dence of airway obstruction, was homogenised and thelevels of HSA analysed using ELISA. As Figure 1 indi-cates, the level of HSA was decreased in lung tissuefrom GOLD II (median = 38.2 μg/ml, IQR = 15.5-48.9,P < 0.05) and GOLD I patients (median = 48.4 μg/ml,IQR = 36.6-93.4, P < 0.05) compared to individuals withnormal lung function (median = 71.9 μg/ml, IQR =52.2-87.6).Relationship between GOLD I & II and levels ofcarbonylated HSAThe tissue homogenates shown in Figure 1 were alsoderivatised and the level of carbonyl residues per HSAmolecule measured by ELISA. The numbers of carbonylresidues together with the values for total HSA shownin Figure 1 were used to calculate the number of carbo-nyl residues per HSA molecule. As shown if Figure 2lung tissue from patients with normal lung function hadvery little carbonylated HSA (median = 0.40 carbonylresidues/HSA molecule, IQR = 0.2-0.7, P < 0.05). How-ever we found the number of carbonyl residues permolecule of HSA was elevated in lung tissue fromGOLD I patients (median of 2.3 carbonyl residues/HSAmolecule, IQR = 1.9-2.5, P < 0.05) and was furtherelevated to a median of 5.0 carbonyl residues/HSAmolecule in lung tissue from GOLD II patients (IQR =4.0-7.6, P < 0.05).HSA µg/mg of tissueGOL D     1050100150200GOL D     2    Normal lung functionGOL D S tatusP  < 0. 05P  < 0. 05Figure 1 Relationship between GOLD I and II patients andbaseline levels of HSA. Human lung tissue from 38 individualsclassified using the GOLD guidelines was homogenised andadjusted for total protein. HSA levels were measured in lunghomogenates using ELISA. The median is marked as a solid bar andexpressed as μg/ml. Data was analysed using the non-parametricKruskal Wallis test, P < 0.05 was considered to be statisticallysignificant.0123456   Normal  lung function    GO LD     1GO LD     2GO L D S tatusCarbonyl residues/ HSA moleculeP  < 0. 05P  < 0. 05Figure 2 Relationship between GOLD I and II patients andbaseline levels of carbonylated HSA. Human lung tissue from 38individuals classified using the GOLD guidelines was homogenised,derivatised and the number of carbonyl residues measured usingELISA. The median is marked as a solid bar and expressed ascarbonyl residues/HSA molecule. Data was analysed using the non-parametric Kruskal Wallis test, P < 0.05 was considered to bestatistically significant.Hackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 4 of 10Re-classification of subjects using LLN for FEV1/FVC todefine COPDThe GOLD guidelines define airway obstruction as afixed FEV1/FVC ratio of 0.70 which has been demon-strated to misdiagnose airway obstruction becauseFEV1/FVC varies with age, height and gender. Thus were-classified the subjects in our study using the spirome-try reference prediction equations from the NHANESIII (16) and NSE (17) studies to confirm that the sub-jects defined with COPD by the GOLD guidelines didhave a spirometry FEV1/FVC lower that the lower limitof normal FEV1/FVC (Table 2). From the 38 patients inthis study, data on age, height and weight was onlyavailable for 31 of the subjects. All of the patients classi-fied with COPD using the GOLD guidelines were alsofound to have obstructive lung disease using LLN FEV1/FVC. Using the LLN re-classified subjects we foundindividuals defined by the GOLD guidelines as GOLD IIhad significantly decreased levels of HSA compared toindividuals with normal lung function and GOLD Ipatients (P = 0.0128, Figure 3A). We also observed thatthe number of carbonyl residues/HSA molecule wasincreased in individuals defined with COPD using theLLN for FEV1/FVC and GOLD guidelines stratification(Figure 3B).Relationship between baseline levels of human serumalbumin and smoking statusHaving observed an inverse relationship between GOLDI and II patients and levels of HSA we turned ourattention to the other clinical parameters collected inthe study. When analysing smoking histories the dataindicated that current smokers had lower levels of HSA(median = 43.3 μg/ml, IQR = 23.8-62.0, P < 0.05) com-pared to ex smokers (median = 71.9 μg/ml, IQR = 38.8-122.7) and non-smokers (median = 71.2 μg/ml, IQR =44.9-80.3.7), as shown in Figure 4. We analyzed bothCOPD and smoking for an association with the levels ofHSA in the study cohort. The data in Table 3 suggestedan association with COPD and HSA levels (P = 0.001),and a significant interaction of COPD with smoking(P < 0.001).Relationship between smoking status and levels ofcarbonylated HSASince smoking status influenced baseline levels of HSAwe next investigated whether levels of carbonylatedHSA were also affected. We found no differencebetween the number of carbonylated HSA molecules inex-smokers (median = 1.9 carbonyl residues/HSA mole-cule, IQR = 0.3-2.2) and the non-smokers (median =1.51 carbonyl residues/HSA molecule, IQR = 0.6-2.2,Figure 5). This was in contrast to lung tissue from cur-rent smokers which exhibited a significantly greaternumber of carbonyl residues per HSA molecule (median= 3.60 carbonyl residues/HSA molecule, IQR = 0.7-4.9,P < 0.05).We analyzed both COPD and smoking for an associa-tion with the levels of carbonylated HSA in the studycohort. The data in Table 3 suggested there was anGOLD status defined using LLN GOLD Status defined using LLN Normal  LungFunction Normal  LungFunctionGOLD I GOLD IGOLD II GOLD II051015P < 0.0001P < 0.002Carbonyl residies/HSAmolecule050100150200P = 0.0128HSA µg/mg of tissueFigure 3 Reclassification of subjects using LLN FEV1/FVC to define COPD. Subjects from Figure 1 and 2 were re-classified using the lowerlimit of normal (LLN) for FEV1/FVC using prediction equations from the NHANES III and NSE studies to confirm COPD and then categorized bythe GOLD guidelines. Data was analysed using the non-parametric Kruskal Wallis test, P < 0.05 was considered to be statistically significant.Hackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 5 of 10association with COPD and smoking with carbonylatedHSA levels (P = 0.001), and a significant interaction ofCOPD with smoking (P = 0.007).Degradation of HSA in human lung tissueAs we observed a reduction in the total levels of HSA inlung tissue from COPD patients and smokers (Figure 1,3 and 4), but an increase in the number of carbonylresidues per molecule of HSA (Figure 2, 3 and 4), thisindicated that oxidation may be effecting HSA turnover. Thus we investigated whether exogenously addedoxidised HSA compared to unmodified HSA, isdegraded in human lung tissue. To evaluate HSA degra-dation, human lung tissue explants from 12 individuals(6 ex, 5 current and 1 non-smoker, 5F/7 M, averageFEV1/FVC = 0.64, average age = 68.1) were culturedwith either 200 μg/ml HSA or oxidised HSA for 1, 2, 4,6 and 24 hours and supernatants analysed using a HSAELISA. As shown in Figure 6 when the tissue was incu-bated with non-oxidized HSA, the levels of HSA in thesupernatant remained relatively constant over the 24hour duration. In contrast, when tissue was incubated050100150200   Non- smoker   E x- smokerC urrent smokerS mok ing  s tatusP  < 0. 05P  < 0. 05HSA µg/mg of tissueFigure 4 Relationship between smoking status and baselinelevels of HSA. Human lung tissue from current smokers (n = 18),ex smokers (n = 15) and non-smokers (n = 5) was homogenisedand adjusted for total protein. HSA levels were measured in lunghomogenates using ELISA. The median is marked as a solid bar andexpressed as μg/ml. Data was analysed using the non-parametricKruskal Wallis test, P < 0.05 was considered to be statisticallysignificant.Table 3 Analysis of COPD and smoking interactions onHSA and carbonylated HSAHSATerm Β SE P valueCOPD -0.6090 0.0029 0.001Smoking status -0.0651 0.0580 0.037COPD × smoking -0.3716 0.0170 <0.001Carbonylated HSA molecules/HSA moleculeTerm Β SE P valueCOPD -0.579 0.0053 0.001Smoking status -0.861 0.0035 0.001COPD × smoking -0.553 0.0033 0.007Values are means ± plusorminus SD for non-continuous data unless otherwisestatedHSA, human serum albumin; COPD, ratio of air forcibly exhaled in one second(FEV1) to the forced vital capacity (FVC) pre-bronchodilator use (FEV1/FVC <70%) and FEV1 ≤ 80% predicted; Smoking, current smoking history.02.557.510   Non- smoker   E x- smokerC urrent smokerS mok in g s tatusCarbonyl residues/HSA molecule P  < 0.05P  < 0.05Figure 5 Relationship between levels of carbonylated HSA andsmoking status. Human lung tissue from current smokers (n = 18),ex smokers (n = 15) and non-smokers (n = 5) was homogenised.Samples were derivatised and the number of carbonyl residuesmeasured using ELISA. The median is marked as a solid bar andexpressed as carbonyl residues/HSA molecule. Data was analysedusing the non-parametric Kruskal Wallis test, P < 0.05 wasconsidered to be statistically significant.Hackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 6 of 10with oxidised HSA we observed a dramatic decrease inthe detectable levels of HSA after 4 hours. Indeed, after24 hours the levels of oxidised HSA had decreased to105.7 μg/ml compared to 213.5 μg/ml for unmodifiedHSA, P < 0.05. The representative blot in Figure 7 forHSA in lung explant supernatants demonstrates thesame pattern of rapid (A) oxidized HSA turnover over24 hours compared to (B) unmodified HSA.DiscussionIn the present study, we investigated the oxidation anddegradation of HSA, an abundant sacrificial anti-oxi-dant, in explants of human lung tissue obtained frompatients with and without COPD. We found parenchy-mal tissue from COPD patients who were current smo-kers contained lower levels of total HSA, but hadproportionally greater levels of carbonylated HSA, com-pared to patients with normal lung function. Lung tissuefrom current smokers was also found to contain lowerlevels of HSA which was highly carbonylated comparedto lung tissue from ex smokers and non-smokers. Cigar-ette smoking has been associated for many years withdecreased levels of the anti-oxidants such as ascorbateand vitamin C (19-21). In addition, recent studies haveshown decreased levels of ascorbic acid and Vitamin Ein COPD patients during exacerbations compared tostable periods (22). However, this is the first study toprovide evidence of reduced levels of the anti-oxidantHSA within parenchymal tissue from current smokerswith COPD.Serum albumin is one of the major antioxidants in therespiratory tract lining fluid, which also includes mucin,superoxide dismutase, glutathione, uric acid and ascor-bic acid. The pathogenesis of COPD is thought toinvolve an increased oxidant burden both directly as aresult of smoking and indirectly by the release of ROSwhich may not be adequately counterbalanced by thepulmonary antioxidant systems, resulting in net oxida-tive stress. Decreased levels of HSA in current smokerswith COPD could therefore contribute to the excessiveaccumulation of oxidants which would lead to enhancedexpression of pro-inflammatory mediators, inactivationof anti-proteinases and ultimately oxidative tissue injury.It is unlikely that current smokers with COPD aregenetically predisposed to produce lower levels of HSA.Although single nucleotide polymorphisms in the genehave been documented, those that affect synthesis of theprotein are extremely rare (23, 24). Alternatively it ispossible that HSA like many genes emerging from theliterature could be epigenetically regulated.In an attempt to elucidate other possible mechanismsthat could underpin the reduced expression of this anti-oxidant, we examined whether COPD and smokingaffected the levels of oxidised HSA, and as a result itsdegradation. Our data demonstrate that the number ofcarbonyl residues per HSA molecule is increased inTime (h ou rs )HSA (µg/mg of tissue)0501001502002500 12 24**Oxidized HSANon-oxidized HSAFigure 6 Degradation of HSA and oxidised HSA in human lungtissue. Human lung tissue (n = 12) was incubated with 200 μg/mlHSA (open circles) or 200 μg/ml oxidised HSA (filled circles) for 1, 2,4, 6, and 24 hours. Samples were analysed for the levels of HSAusing ELISA. Values given are the mean ± SEM and are expressed asμg/ml. The data was statistically analysed using the Wilcoxon-Signedrank test, * indicates a P value < 0.05.Time (h)            1         2        4        6       24HSA std 65 KDa200 µg/ml HSATime (h)            1         2        4        6        24HSA std 65 KDa200 µg/ml oxidized HSAABFigure 7 Western blot analysis of HSA and oxidised HSAdegradation in human lung tissue. Human lung tissue (n = 12)was cultured with 200 μg/ml HSA or oxidised HSA and incubatedfor 1, 2, 4, 6, or 24 hours. Supernatants were separated on a 12%SDS-polyacrylamide gel and analysed for HSA expression usingimmunoblot. The supernatants cultured with HSA are depicted inFigure 7a and the supernatants cultured with oxidised HSA areshown in figure 7b. The blot depicted is a typical example of themolecular profile of HSA observed for all individuals in the study.Hackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 7 of 10COPD patients. However within the study we were notable to obtain lung tissue from GOLD III and IV stageCOPD patients to determine if the expression of HSAdecreases with disease severity. However we could con-firm that the subjects classified with COPD had obstruc-tive lung function whether they were defined using theGOLD guidelines or the lower limit of normal for FEV1/FVC ratio using the prediction equation from theNHANES III (16) and NSE(17) studies. With both clas-sifications we consistently found that GOLD II patientshad decreased levels of HSA molecules which had agreater number of carbonylated residues. We alsoobserved that lung explants from current smokers hadelevated numbers of carbonyl residues per HSA mole-cule compared to those from ex and non-smokers. Theassociation of COPD and smoking with levels of carbo-nylated HSA and a COPD × smoking interaction withlevels of HSA indicates that the two cofactors arerequired to be present for the effects to manifest. Insupport of this, cigarette smoke has been shown tomodify human plasma proteins, producing carbonyl pro-teins with lost sulfhydryl groups (25, 26). In the clinicalsetting it has been shown that the content of oxidisedproteins recovered in BAL is greater in smokers com-pared with non-smoking control subjects (27). Moreimportantly Rahman et al reported that plasma anti-oxi-dant activity is decreased acutely in cigarette smokers,following acute exacerbations in COPD patients (28). Inaddition oxidised HSA has previously been reported inBAL from COPD patients (29). As the parenchymallung explants could not be inflated for histology, it wasnot possible to determine the localisation of HSA, whichis a limitation of our study. The carbonylated HSA mea-sured with the lung tissue could therefore be present inthe intravascular space, extracellular fluid or intracellu-lar environment. In the clinical setting it would thus beimportant to determine if the levels of carbonylatedHSA were derived primarily from the lung or the sys-temic circulation. Ultimately independent of the sourceof HSA, decreased levels of the protein, could contributeto the oxidative burden within the lungs of smokerswith COPD and potentially result in lung tissue damage.Of particular note is our observation that lung tissuefrom ex smokers, defined as having given up smokingfor at least 3 years, had the same mean concentration ofcarbonylated HSA as non-smokers. This may suggestthat smoking cessation could prevent the elevated oxida-tion and degradation of HSA at least in part, contribut-ing to the restoration of the oxidant/anti-oxidantbalance within the lung. It is well documented thatsmoking cessation in addition to other therapies such asinhaled steroids and bronchodilators can be effectivetreatments for COPD, decreasing the accelerated declinein lung function and disease progression. If as our datasuggests that the oxidant/anti-oxidant imbalance isresolved with smoking cessation it further supports therole of antioxidant disturbances in the progression ofCOPD. The data however can not indicate the timescale required for the resolution of smoking related oxi-dative stress within the lung.In this current study we found that the proportion ofcarbonylated HSA was greatest in smokers with COPD.As carbonylated proteins are degraded more rapidly wehypothesised that in these patients’ total levels of HSAare decreased due to rapid degradation of the carbony-lated protein. Using an in vitro lung tissue culture sys-tem we added exogenous oxidised HSA to model theeffects of oxidised HSA within the extracellular fluid ofthe lung. In support of this hypothesis our in vitro datademonstrated that oxidised HSA was degraded morerapidly than unmodified HSA in cultured human lungtissue explants, when analysed by ELISA and westernblot. Larger molecular proteins such as albumin are pri-marily cleared from the lung by paracellular mechan-isms, into the systemic circulation. However, as thesupernatant and tissue were analysed in our model itsuggests that carbonylated HSA could be degraded bythe parenchymal lung explants. In support of this find-ing, it has been demonstrated that both albumin andother high molecular weight proteins can be directlycleared by the epithelium through epithelial receptormediated endocytosis or pinocytosis, and these proteinsare catabolised through lysosomal degradation (30-32).Recent evidence suggests that oxidation of HSAdecreases its denaturation enthalpy, suggesting that oxi-dation of HSA renders it to be denatured more easily(33). The precise mechanisms involved in the metabolicturnover of HSA have not been fully elucidated. Theyare thought also to involve the uptake of damaged pro-teins by type A scavenger receptors found on macro-phages and the sinusoidal liver epithelial cells (34, 35).The tissue culture experiments were performed on par-enchymal tissue from donors with and without COPDand different smoking histories. Although no differenceswere observed between the responses of parenchymaltissue from different donors, the sample size was toosmall for statistical analysis, which is a limitation todetermine the effects of smoking and disease on HSAturnover.In summary, our study provides further evidence forthe role of oxidative stress in current smokers withCOPD and is the first study to evaluate the effect of oxi-dation on HSA degradation in human lung tissue. HSAis currently used clinically to maintain colloid osmoticpressure and is also viewed as an important antioxidantin patients with damaged vascular endothelium andpatients with acute lung injury (7, 36, 37). Our data sug-gests that it might also be important not only toHackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 8 of 10consider oxidised HSA as a marker of oxidative stress incurrent smokers with COPD, but also the potential ther-apeutic role of HSA in the homeostasis of the oxidant/anti-oxidant balance, where there is a large unmetclinical need.AcknowledgementsWe would like to thank the cardiothoracic team at Guy’s Hospital for theirinvaluable support in providing surgical specimens and continued support.TLH is a recipient of a Canadian Institute for Health Research/Canadian LungAssociation/GSK, IMPACT strategic training initiative and Michael SmithFoundation for Health Research fellowships.Author details1School of Medicine, University of Southampton, Southampton, UK. 2JamesHogg Research Centre, Heart + Lung Institute, University of British Columbia,Vancouver, Canada. 3Department of Thoracic Surgery, Guy’s Hospital, Greatmaze pond, London, UK.Authors’ contributionsTLH participated in the study design carried out the tissue culture studies,immunoassays, performed the statistical analysis and drafted the manuscript.MS, LZ, WT and TT participated in patient data collection, statistical analysisand manuscript revision. JAW conceived of the study, participated in itsdesign, analysis and manuscript preparation. All authors read and approvedthe final manuscript.Competing interestsThe authors declare that they have no competing interests.Received: 20 April 2010 Accepted: 22 December 2010Published: 22 December 2010References1. MacNee W: Pulmonary and systemic oxidant/antioxidant imbalance inchronic obstructive pulmonary disease. Proc Am Thorac Soc 2005,2(1):50-60.2. Bowler RP, Barnes PJ, Crapo JD: The role of oxidative stress in chronicobstructive pulmonary disease. Copd 2004, 1(2):255-77.3. Pryor WA, Stone K: Oxidants in cigarette smoke. Radicals, hydrogenperoxide, peroxynitrate, and peroxynitrite. Ann N Y Acad Sci 1993,686:12-27.4. Cha MK, Kim IH: Glutathione-linked thiol peroxidase activity of humanserum albumin: a possible antioxidant role of serum albumin in bloodplasma. Biochem Biophys Res Commun 1996, 222(2):619-25.5. Era S, Kuwata K, Imai H, Nakamura K, Hayashi T, Sogami M: Age-relatedchange in redox state of human serum albumin. Biochim Biophys Acta1995, 1247(1):12-6.6. Soriani M, Pietraforte D, Minetti M: Antioxidant potential of anaerobichuman plasma: role of serum albumin and thiols as scavengers ofcarbon radicals. Arch Biochem Biophys 1994, 312(1):180-8.7. Bourdon E, Loreau N, Blache D: Glucose and free radicals impair theantioxidant properties of serum albumin. Faseb J 1999, 13(2):233-44.8. Levine RL, Garland D, Oliver CN, Amici A, Climent I, Lenz AG, et al:Determination of carbonyl content in oxidatively modified proteins.Methods Enzymol 1990, 186:464-78.9. Davies KJ, Lin SW, Pacifici RE: Protein damage and degradation by oxygenradicals. IV. Degradation of denatured protein. J Biol Chem 1987,262(20):9914-20.10. Wright E, Scism-Bacon JL, Glass LC: Oxidative stress in type 2 diabetes:the role of fasting and postprandial glycaemia. Int J Clin Pract 2006,60(3):308-14.11. Matsuura E, Kobayashi K, Tabuchi M, Lopez LR: Oxidative modification oflow-density lipoprotein and immune regulation of atherosclerosis. ProgLipid Res 2006, 45(6):466-86.12. Himmelfarb J, McMonagle E: Albumin is the major plasma protein targetof oxidant stress in uremia. Kidney Int 2001, 60(1):358-63.13. Buss H, Chan TP, Sluis KB, Domigan NM, Winterbourn CC: Protein carbonylmeasurement by a sensitive ELISA method. Free Radic Biol Med 1997,23(3):361-6.14. GOLD: Global strategy for the diagnosis, management, and preventionof chronic obstructive pulmonary disease. GOLD guidelines 2006 2006[http://WWW.goldcopd.org], [cited [cited 2007].15. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS: Global strategy forthe diagnosis, management, and prevention of chronic obstructivepulmonary disease. NHLBI/WHO Global Initiative for Chronic ObstructiveLung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med 2001,163(5):1256-76.16. Swanney MP, Ruppel G, Enright PL, Pedersen OF, Crapo RO, Miller MR, et al:Using the lower limit of normal for the FEV1/FVC ratio reduces themisclassification of airway obstruction. Thorax 2008, 63(12):1046-51.17. Falaschetti E, Laiho J, Primatesta P, Purdon S: Prediction equations fornormal and low lung function from the Health Survey for England. EurRespir J 2004, 23(3):456-63.18. Schleimer RP, Schulman ES, MacGlashan DW, Peters SP, Hayes EC,Adams GK, et al: Effects of dexamethasone on mediator release fromhuman lung fragments and purified human lung mast cells. J Clin Invest1983, 71(6):1830-5.19. Duthie GG, Arthur JR, Beattie JA, Brown KM, Morrice PC, Robertson JD, et al:Cigarette smoking, antioxidants, lipid peroxidation, and coronary heartdisease. Ann N Y Acad Sci 1993, 686:120-9.20. Pelletier O: Vitamin C and cigarette smokers. Ann N Y Acad Sci 1975,258:156-68.21. Anderson R, Theron AJ, Ras GJ: Ascorbic acid neutralizes reactive oxidantsreleased by hyperactive phagocytes from cigarette smokers. Lung 1988,166(3):149-59.22. Tug T, Karatas F, Terzi SM: Antioxidant vitamins (A, C and E) andmalondialdehyde levels in acute exacerbation and stable periods ofpatients with chronic obstructive pulmonary disease. Clin Invest Med2004, 27(3):123-8.23. Murray JC, Demopulos CM, Lawn RM, Motulsky AG: Molecular genetics ofhuman serum albumin: restriction enzyme fragment length polymorphismsand analbuminemia. Proc Natl Acad Sci USA 1983, 80(19):5951-5.24. Koot BG, Houwen R, Pot DJ, Nauta J: Congenital analbuminaemia:biochemical and clinical implications. A case report and literaturereview. Eur J Pediatr 2004, 163(11):664-70.25. Reznick AZ, Cross CE, Hu ML, Suzuki YJ, Khwaja S, Safadi A, et al:Modification of plasma proteins by cigarette smoke as measured byprotein carbonyl formation. Biochem J 1992, 286(Pt 2):607-11.26. O’Neill CA, Halliwell B, van der Vliet A, Davis PA, Packer L, Tritschler H, et al:Aldehyde-induced protein modifications in human plasma: protectionby glutathione and dihydrolipoic acid. J Lab Clin Med 1994, 124(3):359-70.27. Lenz AG, Costabel U, Maier KL: Oxidized BAL fluid proteins in patientswith interstitial lung diseases. Eur Respir J 1996, 9(2):307-12.28. Rahman I, MacNee W: Oxidant/antioxidant imbalance in smokers andchronic obstructive pulmonary disease. Thorax 1996, 51(4):348-50.29. Foreman RC, Mercer PF, Kroegel C, Warner JA: Role of the eosinophil inprotein oxidation in asthma: possible effects on proteinase/antiproteinase balance. Int Arch Allergy Immunol 1999, 118(2-4):183-6.30. Hastings RH, Folkesson HG, Petersen V, Ciriales R, Matthay MA: Cellularuptake of albumin from lungs of anesthetized rabbits. Am J Physiol 1995,269(4 Pt 1):L453-62.31. Grune T, Davies KJ: Breakdown of oxidized proteins as a part ofsecondary antioxidant defenses in mammalian cells. Biofactors 1997,6(2):165-72.32. Das S, Horowitz S, Robbins CG, el-Sabban ME, Sahgal N, Davis JM:Intracellular uptake of recombinant superoxide dismutase afterintratracheal administration. Am J Physiol 1998, 274(5 Pt 1):L673-7.33. Anraku M, Yamasaki K, Maruyama T, Kragh-Hansen U, Otagiri M: Effect ofoxidative stress on the structure and function of human serum albumin.Pharm Res 2001, 18(5):632-9.34. Swart PJ, Beljaars L, Kuipers ME, Smit C, Nieuwenhuis P, Meijer DK: Homingof negatively charged albumins to the lymphatic system: generalimplications for drug targeting to peripheral tissues and viral reservoirs.Biochem Pharmacol 1999, 58(9):1425-35.35. Duryee MJ, Freeman TL, Willis MS, Hunter CD, Hamilton BC, Suzuki H, et al:Scavenger receptors on sinusoidal liver endothelial cells are involved inHackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 9 of 10the uptake of aldehyde-modified proteins. Mol Pharmacol 2005,68(5):1423-30.36. Lang JD, McArdle PJ, O’Reilly PJ, Matalon S: Oxidant-antioxidant balance inacute lung injury. Chest 2002, 122(6 Suppl):314S-20S.37. Quinlan GJ, Mumby S, Martin GS, Bernard GR, Gutteridge JM, Evans TW:Albumin influences total plasma antioxidant capacity favorably inpatients with acute lung injury. Crit Care Med 2004, 32(3):755-9.doi:10.1186/1465-9921-11-180Cite this article as: Hackett et al.: Oxidative modification of albumin inthe parenchymal lung tissue of current smokers with chronicobstructive pulmonary disease. Respiratory Research 2010 11:180.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitHackett et al. Respiratory Research 2010, 11:180http://respiratory-research.com/content/11/1/180Page 10 of 10


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items