Base excision repair capacity in chronic renal failure patients undergoing hemodialysis treatment

Base excision repair capacity in chronic renal failure patientsundergoing hemodialysis treatment

Elitsa Stoyanova1, Susana Pastor1,2, Elisabet Coll3, Amaya Azqueta4,†, Andrew R. Collins4,† and Ricard Marcos1,2*1Grup de Mutagènesi, Departament de Genètica i de Microbiologia, Edifici Cn, Universitat Autònoma de Barcelona, Barcelona, Spain2CIBER Epidemiología y Salud Pública, ISCIII, Madrid, Spain3Fundació Puigvert, Barcelona, Spain4Department of Nutrition, Food Science and Toxicology, University of Navarra, Spain

The aim of this study was to determine if the differences observed in the levels of DNA damage in a group of patients suffering from chronicrenal failure are due to differences in the repair capability. DNA damage was initially measured with the comet assay in 106 hemodialysispatients. A selected group of 21 patients representing high (ten patients) and low (11 patients) levels of DNA damage were obtained fordetermination of base excision repair capacity. This was measured in an in vitro assay where protein extracts from lymphocytes wereincubated with a substrate of DNA containing 8-oxoguanine, and the rate of incision was measured with the comet assay. Patients with highlevels of genomic damage showed, as an average, significantly lower repair capacity (12·73 ± 1·84) in comparison with patients with lowlevels of genomic damage (18·13 ± 1·13). Nevertheless, the correlation coefficient between repair ability and levels of genomic damagewas found to be only close to the significance value (r:�0·423, p: 0·056). Although DNA damage was clearly related to time on hemodialysis,base excision repair capacity was not. This is one of the few studies providing information on the repair capacity of chronic renal failure patientsundergoing hemodialysis. As a summary, our results would indicate that DNA damage levels are in part associated to the repair capacity of thepatients, and this repair capacity is not associated with the duration of hemodialysis treatment. Copyright © 2013 John Wiley & Sons, Ltd.

key words—base excision repair; genetic damage; comet assay; chronic renal failure; hemodialysis

INTRODUCTION

Chronic kidney disease is characterized by the progressivedeterioration of renal function and is defined as a glomerularfiltration rate lower than 60mlmin�1, for at least 3months.Generally, the early stages (stages 1 and 2) of the diseaseare asymptomatic, in stages 3–5, kidney functions beginto decrease below 60mlmin�1 and eventually in stage 5,complete renal failure is reached, and a replacement therapyis required.1 Chronic renal failure (CRF) has become apathophysiological state with worldwide distribution andwith an increased incidence during the last decades, becauseof the increase in the prevalence of pathologies that causethis disease (high blood pressure and diabetes mellitus),and with increasing longevity of the population.2,3

It should be mentioned that CRF patients are considereda group of high health risk because of the relationshipof kidney disease with other chronic diseases.4 In this

population, there is a high risk of cardiovascular complica-tions,5 cancer incidence,6 premature ageing,7 immunitychanges,8 skin disorders,9 malnutrition10 and hypertension,11

among others.On the other hand, CRF patients have also been

proposed as a group under conditions of inflammation,oxidative stress and carbonyl stress.12 Patients treated withhemodialysis (HD) are thought to have an increasedgenetic instability, as indicated by the incidence of struc-tural chromosome abnormalities and elevated levels ofsister chromatid exchanges.13–15

Many studies have shown that the different stages of renalfailure are associated with increased levels of oxidativestress,16,17 generated from both reduced antioxidant capacityand increased reactive oxygen species production andrelated to the accumulation of uremic toxins18 or to deficien-cies in DNA repair ability.19

Chronic renal failure patients have been postulated as agroup with reduced repair capacity in comparison withcontrols,20 but conflicting results on this topic have beenreported. Thus, when DNA repair ability was measured, inresponse to in vitro ultraviolet irradiation, it was found tobe reduced in lymphocytes from predialysis patients, butthis ability was restored after HD.21 On the other hand, inpatients submitted to HD, the repair system was found to

*Correspondence to: Ricard Marcos, Grup de Mutagènesi, Departamentde Genètica i de Microbiologia, Edifici Cn, Universitat Autònoma deBarcelona, 08193 Bellaterra, Cerdanyola del Vallès, Barcelona, Spain.E-mail: ricard.marcos@uab.es†Current address: Department of Nutrition, Institute of Basic MedicalSciences, University of Oslo, Oslo, Norway

Received 4 March 2013Revised 22 May 2013Accepted 6 June 2013Copyright © 2013 John Wiley & Sons, Ltd.

cell biochemistry and functionCell Biochem Funct 2014; 32: 177–182.Published online 19 July 2013 in Wiley Online Library(wileyonlinelibrary.com) DOI: 10.1002/cbf.2989

be normal or slightly enhanced during the first year oftreatment,22 which agrees with the results reported byBagatini et al.23 in patients with type 2 diabetes mellitusundergoing HD; although type 2 diabetes mellitus by itselfhas been shown to reduce repair ability.24 These conflictingresults could be due to the different methodologies used inthe different studies, and the reported repair capacity couldjust indicate the presence of more or less genomicdamage; thus, DNA repair, measured by the incorporationof 3H-thymidine, increased significantly following HD.This occurred in association with increased DNA damageinduced by oxidative stress.25

As a contribution to this discussion on the relationshipbetween CRF, genomic damage and DNA repair ability,we first determined DNA damage using the comet assay(single cell gel electrophoresis)26 in lymphocytes froma large group of 106 HD patients.27 From this largegroup, two subgroups of HD patients were then selectedaccording to their levels of genomic damage: a group often with the highest levels and a group of 11 with thelowest basal levels of DNA damage. We used proteinextracts from these patients' lymphocytes, containingrepair enzymes, to determine their base excision repair(BER) capacity by the ability to incise at 8-oxoguanineresidues in DNA from control cells treated with aphotosensitising agent (Ro 19-8022) plus light, measuringthe induced breaks with the comet assay.28 The BERpathway is the major system responsible for the removalof oxidized bases and also for the repair of DNA singlestrand breaks generated spontaneously or induced byexogenous DNA damaging factors.29 The effect of thedialysis time on the repair ability, and consequently ongenetic damage, means in practice a progressive clinicaldeterioration (premature ageing, higher risk of cancer andcardiovascular disease, inflammation, malnutrition, etc…)in this group of patients after a long period of dialysis (whilethey are waiting for a kidney transplant). This would agreewith the results of Vamvakas et al.22 who found that al-though a tendency for DNA repair capacity to increase wasdetected at the start of the HD treatment, this was signifi-cantly reduced after a certain time. This would suggest a re-lationship between length of treatment and repair capacity.In our study, a possible relation was sought between the

levels of DNA damage in peripheral lymphocytes and theability of CRF patients to repair this damage.

MATERIALS AND METHODS

Subjects

In a previous study of our group, blood samples werecollected from 106 CRF patients undergoing HD treatment,and the levels of DNA damage in lymphocytes wereanalysed using the comet assay. From this group ofdeceased patients, those with renal transplantation andpatients with cancer incidence were excluded, as well asthose who moved to another city, and then, 11 patients withthe lowest and ten patients with the highest values of

oxidized purines were selected for the present study. Allthe 21 patients included in the measurement of BER activi-ties were in stage 5 of kidney failure, undergoing standardHD treatment three times per week. The underlying causesof renal failure in the studied patients were glomerulonephri-tis (54·5% in the low damage group and 20% in thehigh damage group), diabetes mellitus (18·2% in the lowdamage group and 20% in the high damage group),nephroangiosclerosis (18·2% in the low damage group and20% in the high damage group), renovascular causes (9·1%of the patients in the low damage group), polycystic kidney(10% of high damage group) and another 30% of unknowncauses for the patients in the high damage group.A second blood sample was taken from the patients to

prepare the protein extracts needed for the repair assay.Average levels of DNA damage between groups weresignificantly different. Characteristics of the selectedpatients are indicated in Table 1.All patients were recruited in the Puigvert Foundation,

Barcelona between 2007 and 2010, and the blood samplesfor the repair assay were collected in 2010. Informedconsent was obtained, and blood samples were collectedunder protocols approved by the Ethics Committee ofthe Puigvert Foundation. Clinical data were recorded fromtheir medical history, including pathologies and data forCRP (C-reactive protein), ferritin, parathyroid hormone,cholesterol, albumin, vitamin B12 and among otherbiochemical biomarkers.

Comet assay for measuring DNA damage (strand breaksand oxidized bases)

The comet assay protocol used for the measurement of DNAstrand breaks and oxidized bases was performed as previ-ously described by Stoyanova et al.27

In vitro repair comet assay

The in vitroDNA repair comet assaywas performed as previouslydescribed.30 Preparation of the extract is as follows. From eachpatient, 10ml of peripheral blood was obtained, and lymphocyteswere isolated with the Lymphoprep sedimentation method.Lymphocytes were resuspended in 10ml of buffer A diluted threetimes [4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid](HEPES), 0·4M KCl, 1mM ethylenediaminetetraacetic acid(EDTA), 0·1mM dithiothreitol, 10% glycerol, adjusted to pH7·8 with KOH]. A centrifugation at 700 g during 5min at 4 °C was performed, and the supernatant was removed. For each106 cells, 20μl of buffer A were added. From this suspension,50μl aliquots were prepared in cryotubes and stored at�80 °C.Preparation of the substrate is as follows. Twenty

millilitre of blood were obtained from a healthy donor, andlymphocytes were isolated with Lymphoprep. Cells wereresuspended in PBS and divided into two tubes. One tubewas treated with 20μl of 1mM Ro 19-8022 photosensitiser(F. Hoffman-La Roche Ltd, Basel, Switzerland), and theother tube was used as a control. Both tubes were placedon ice and irradiated for 5min at 30 cm from a 500Whalogen lamp to induce 8-oxoguanine. After centrifugation,

178 e. stoyanova ET AL.

Copyright © 2013 John Wiley & Sons, Ltd. Cell Biochem Funct 2014; 32: 177–182.

the supernatant was removed, and cells were resuspended in10ml of PBS, prior to the determination of cell concentra-tion. Tubes were centrifuged under the same conditions,and cells were resuspended in freezing medium (Dulbecco'sModified Eagles Medium +10% of dimethyl sulfoxide) at adensity of 106 cells/ml, and 0·5ml aliquots were slowlyfrozen to �80 °C.

For each experiment, two substrate aliquots were used(treated and control). A total of 60μl of cell suspension(0·5 x 106 cells/ml) was mixed with 280 μl of 1% low melt-ing point agarose in PBS, and by using a multi-dispensingpipettor, 12 gels of 5μl were made on each microscopicslide. For each experiment, four slides were prepared—twocontaining Ro 19-8022-treated nucleoids and the other twowith untreated nucleoids. Slides were transferred to lysissolution (2·5M NaCl, 100mM Na2EDTA, 10mM Tris-HCland 1%TritonX100 adjusted to pH 10·4) for aminimum of 1 h.

Two aliquots of each patient's extract were thawed. Extractswere centrifuged to remove nuclei and cell debris. Then 50μl ofthe supernatant was taken and mixed with 200μl of enzymebuffer (40mM hydroxyethyl piperazineethanesulfonic, 0·1MKCl, 0·5mM EDTA, 0·2mgml-1 bovine serum albumin,pH 8·0). A negative control was prepared like extract but with-out cells. The positive control was formamidopyrimidineDNA glycosilase (FPG).

After lysis, slides were washed in enzyme buffer and thenplaced in a 12-gel slide incubation chamber (SevernBiotech). To each gel, 30μl of extract (or negative controlsolution or FPG) was added. For each sample, two gelsper slide were prepared. Slides were incubated for 30minat 37 °C. Part of the extract was stored at �20 °C and usedfor subsequent protein determination.

After incubation, slides were washed with alkalinesolution to stop completely the reaction and then were left

for 20min for denaturation. Electrophoresis was run at25V for 20min at 4 °C. After electrophoresis, a wash withPBS for 10min was followed by a 10-min wash in water.Slides were dried by dipping in 70% ethanol followed by15min in absolute ethanol.

Slides were stained in Tris-EDTA buffer with anappropriate concentration of SybrGold. To score the slides,the Comet IV (Perceptive Instruments) programme wasused, and 30 randomly selected cells were analysed pergel. Results are presented as mean of the median of the rep-licate gels. The mean value with untreated substrate wassubtracted from the mean value with Ro 19-8022-treatedsubstrate; this gives the net accumulation of breaks duringthe incubation period, which indicates repair rate.

The protein determination for the extracts used in thein vitro repair assay was measured by NanoDrop (NanodropTechnologies, Inc. Wilmington, DE, USA) to test whetherthe protein amount was equal in all samples.

Statistical analysis

For the analysis of the data, the parameter percentage of DNAin the tail was used because it is a parameter that provides aclear indication of the formation of comets giving a goodlinear relationship with the frequency of DNA damage.31

As the number of patients used for the analysis islower than 30, a Shapiro–Wilks test was performed toassure the normality of the data.32 Because the distribu-tion of the data was normal, parametric tests were usedfor the statistical analysis. Comparisons between thetwo groups were analysed with the Student's t-test forcontinuous variables and the Chi-square test for discretevariables. All continuous variables were presented asmean ± standard error. Statistical analysis was performed

Table 1. Some characteristics of the studied population

Low damage group High damage group All patients (106 CRF)

Age (years) 58·73 ± 5·60 55·60 ± 5·40a 61,19 ± 1,45Sex (male/female) 6/5 5/5 66/40Time on HD treatment (months) 31·36 ± 4·84 80·10 ± 7·97b 39,38 ± 4,53Previous kidney transplant (yes/no) 4/7 5/5 29/80Hypertension (yes/no) 9/2 10/0 93/13Diabetes mellitus (yes/no) 6/5 2/8 26/79Cancer (yes/no) 0/11 0/10 24/82Cardiovascular disease (yes/no) 5/6 7/3 64/42Dyslipidemia (yes/no) 8/3 6/4 62/44Smoking (yes/no) 5/6 7/3 60/45Folic acid supplement (yes/no) 1/10 6/4 45/61ACEI/ARB (yes/no) 7/4 6/4 46/60CRP (mg l-1) 7·75 ± 3·70 7·26 ± 2·34 14,00 ± 2,63Ferritin (μg l-1) 199·15 ± 64·87 341·57 ± 77·50b 299,66 ± 24,70PTH (ng l-1) 102·10 ± 30·10 346·51 ± 141·38b 211,71 ± 20,40Vitamins B and C supplement (yes/no) 2/9 1/9 35/71

Values are expressed as mean ± standard error; there were no significant differences between groups except where indicated.aComparison between low damage group and high damage group p= 0·69 (Student's t-test for independent samples).bComparison between low damage group and high damage group p< 0·001 (Student's t-test for independent samples).CRF, chronic renal failure; CRP, C-reactive protein; PTH, parathyroid hormone; ACEI (angiotensin converting enzyme inhibitors) / ARB (angiotensin II-receptor blockers), drugs for hypertension treatment; HD, hemodialysis.

179ber in chronic renal failure patients

Copyright © 2013 John Wiley & Sons, Ltd. Cell Biochem Funct 2014; 32: 177–182.

using the PC version 19.0 of the Statistic Package of SocialSciences (IBM SPSS, Chicago, IL, USA) software forWindows, and statistical significance was defined p< 0·05.

RESULTS

Table 1 shows some characteristic of the patients included inthe study. Both sets of patients were selected from the groupof 106 individuals undergoing HD treatment. Significantdifferences were observed in the time patients wereundergoing HD treatment—the high damage group wastreated for longer periods in comparison with the lowdamage group. The periods of HD treatment cover a widerange (from 14 to 72months for the low damage groupand from 50 to 126months for the high damage group).Figure 1 shows the DNA damage values, measured as

strand breaks, of the group of 106 CRF patients, indicatingthose who were selected for the study. The average valueof DNA damage was 28·68 ± 1·29 (minimum value 3·29,maximum value 53·17).Table 2 shows the levels of DNA strand breaks and

base oxidation in lymphocytes from the two groups,measured with the comet assay. The difference in levelsof strand breaks between the groups was highly signifi-cant (p< 0·001, Student's t-test). Regarding the levelsof oxidative damage, significant differences were ob-served for the values of FPG sensitive sites.Individual data corresponding to repair ability of selected

patients are shown in Figure 2. It is seen that patientsbelonging to the group with high levels of DNA damagepresent a lower DNA repair efficiency (12·73 ± 1·84) thanthose belonging to the group with low levels of DNAdamage (18·13 ± 1·13). These differences attain statistical

significance (p= 0·020, Student's t-test). To demonstrate theexisting association between the basal levels of DNA dam-age and the ability to repair DNA damage, wehave determined the Pearson's coefficient of correlation.Results indicate that this correlation is almost significant(r: �0·423, p: 0·056), which would indicate a certainrelationship, although modulated by other factors.In addition, to determine whether the protein content was

different in the both groups, acting as a possible modulatingfactor, this parameter has also been evaluated. Resultsindicate that the protein content of the samples from bothgroups did not show significant differences among them,being 12·39 ± 2·18mgml�1 for the high damage group and9·77 ± 1·04mgml-1 for the low damage group (p= 0·300,Student's t-test).We have also tried to answer the question whether there is a

relationship between the ability of the CRF patients to repairDNA base oxidation and the time they have undergone HDtreatment. When the levels of DNA damage were considered,the average time on HD treatment for the group with low DNAdamage was 31·36±4·84months, and for the group with highDNA damage was 80·10±7·97months (54·57±7·02monthsfor the total of 21 patients), which indicate a statistically signifi-cant difference (p< 0·001, Student's t-test). When looking forthe repair ability of HD patients treated for a period longer than40months, it is observed that their ability to repair DNA damage(13·27±1·75) is lower than in patients treated for less than40months (18·07±1·25), and these differences attain statisticalsignificance (p=0·041). Nevertheless, when the associationbetween repair ability and time in HD was evaluated, no signifi-cant relationship was observed (correlation coefficient: �0·224,p: 0·330), as reflected in Figure 3.

DISCUSSION

Base excision repair is the predominant pathway dealingwith a wide range of base modifications resulting fromoxidation, alkylation and deamination.33 CRF patientsundergoing HD are a population with increased genomicinstability and a high incidence of different types ofcancer.34,35 We have previously found that CRF patientspresent high levels of genetic damage, significantly higherthan those observed in controls.27 In the current study, wedemonstrate that an excess of genetic damage is associatedwith lower DNA repair activity, at least for damage repaired

high DNA damage group; , low DNA damage group

Figure 1. DNA breaks in peripheral lymphocytes for the population ofchronic renal failure patients. Patients are placed (x-axis) according to theirlevels of DNA breaks, without the use of FPG enzyme. Selected patients areindicated using ■ and ▲ symbols

Table 2. DNA damage measured in peripheral lymphocytes with thecomet assay

% DNA in tail Low damage group High damage group pa

DNA strand breaks 12·04 ± 1·41 37·75 ± 2·90 <0·001Net FPG-sensitive sites(oxidized purines)

3·23 ± 0·69 22·25 ± 4·86 0·003

Values are expressed as mean ± standard error.aComparison between low and high damaged groups (Student's t-test forindependent samples). p, statistical significance if lower than 0·05.

180 e. stoyanova ET AL.

Copyright © 2013 John Wiley & Sons, Ltd. Cell Biochem Funct 2014; 32: 177–182.

by the BER pathway. This association reaches border-line statistical significance (p: 0·056). It is interestingto point out that the highest levels of DNA damage,which would indicate certain inefficiency in removingoxidative damage, are present in patients with thelowest ability to repair. The fact that in some patientsthis association is not observed would indicate thatother underlying factors, possibly linked to the pathol-ogy itself, would act as modulating factors.

Nowadays, there are several assays for the detection ofthe DNA repair capacity. Probably, the simplest way toquantify DNA repair is to monitor genetic damage elimi-nation over time by using the comet assay,36–38 but ithas disadvantages (discussed in Collins and Azqueta28).To overcome these disadvantages, we have used anotherapproach to evaluate the BER capacity of cellularextracts from individual donors, by incubating them with

nucleoids containing specific damage.26 Although the BERassay we use measures only the initial step of base removal/apurinic/apyrimidinic (AP) site cleavage, this is generallyregarded as the rate-limiting step.39

The results following such an approach suggest twodifferent remarks. The first is that there is a tendencyshowing that levels of DNA damage are inversely relatedto the repair ability, as indicated by the relationshipbetween both parameters. The second is that althoughthere is a lack of significant association between time inHD and ability to repair DNA damage when the data fromall subjects tested are analysed together, the group of HDpatients treated for less than 40months had significantlyhigher DNA repair ability in comparison with thosetreated for longer periods. This is in agreement with theresults of Vamvakas et al.22 who found that although atendency for repair capacity to increase was detected atthe start of the HD treatment, this was significantlyreduced after a certain time. This would suggest arelationship between length of treatment and repair capac-ity. Because age could be a variable underlying thisrelationship, we have included patients with similar agepreserving the proportion of men and women, to avoidthe effects of such variables. In addition, as cancerincidence can also decrease DNA repair ability,40 in ourstudy, we have excluded those with cancer. Thus, theobserved effects cannot be explained by variables suchas age and cancer incidence.

In conclusion, from our study in CRF patients, itseems that the levels of DNA strand breaks and oxi-dized bases are modulated in part by the cells' ownBER capacity. Because the observed association is notso strong, this would indicate that other factors may alsobe involved in controlling the levels of DNA damageshown by CRF patients. On the other hand, from ourstudy, it does not seem that time in HD treatment actsas a significant lasting stimulus for DNA repair ability.

Figure 2. Individual base excision repair rates in extracts of peripheral lymphocytes from the selected patients. Patients are placed in order of their base ex-cision repair capacity. From 1 to 10, patients with low repair rate (12·73 ± 1·84) and high levels of basal DNA damage; from 1* to 11*, patients with high repairrate (18·13 ± 1·13) and low levels of basal DNA damage

Time on HD treatment (months)

Figure 3. Correlation between time on hemodialysis treatment and DNArepair capacity in peripheral lymphocytes

181ber in chronic renal failure patients

Copyright © 2013 John Wiley & Sons, Ltd. Cell Biochem Funct 2014; 32: 177–182.

CONFLICT OF INTEREST

The authors declare that there are no conflicts of interest.

ACKNOWLEDGMENTS

First of all, we thank all the volunteers that have participatedin this study. E. Stoyanova was supported by a postgraduateand a travel fellowship from the Generalitat de Catalunya. A.Azqueta was supported by the Juan de la Cierva Programme,2009, from the Spanish Ministry of Educación y Ciencia. Wethank F. Hoffmann-LaRoche Ltd, Basel, Switzerland for thegift of Ro 19-8022. This investigation has been supported inpart by the Generalitat de Catalunya (CIRIT, 2009SGR-725).

REFERENCES

1. NKF-KDOQI. Clinical practice guidelines for chronic kidney disease:evaluation, classification, and stratification. Am J Kidney Dis 2002; 39(2 suppl 1): S1–S266.

2. Stolzmann KL, Camponeschi JL, Remington PL. The increasing inci-dence of end-stage renal disease in Wisconsin from 1982-2003: ananalysis by age, race, and primary diagnosis. WMJ 2005; 104: 66–71.

3. Coresh J, Selvin E, Stevens LA, et al. Prevalence of chronic kidneydisease in the United States. JAMA 2007; 298: 2038–2047.

4. Levey AS, Atkins R, Coresh J, et al. Chronic kidney disease as aglobal public health problem: approaches and initiatives—a positionstatement from kidney disease improving global outcomes. KidneyInt 2007; 72: 247–259.

5. London GM. Cardiovascular disease in chronic renal failure:pathophysiologic aspects. Semin Dial 2008; 16: 1–10.

6. Stewart JH, Buccianti G, Agodoa L, et al. Cancers of the kidney andurinary tract in patients on dialysis for end-stage renal disease: analysisof data from the United States, Europe, and Australia and NewZealand. J Am Soc Nephrol 2003; 14: 197–207.

7. Cohen BD. Premature ageing. Mol Cell Biochem 2007; 298: 195–198.8. Betjes MG, Langerak AW, van der Spek A, de Wit EA, Litjens NH.

Premature aging of circulating T cells in patients with end-stage renaldisease. Kidney Int 2011; 80: 208–217.

9. Udayakumar P, Balasubramanian S, Ramalingam KS, Lakshmi C,Srinivas CR, Mathew AC. Cutaneous manifestations in patients withchronic renal failure on hemodialysis. Indian J Dermatol VenereolLeprol 2006; 72: 119–125.

10. Garg AX, Blake PG, Clark WF, Clase CM, Haynes RB, Moist LM.Association between renal insufficiency and malnutrition in olderadults: results from the NHANES III. Kidney Int 2001; 60: 1867–1874.

11. Tedla FM, Brar A, Browne R, Brown C. Hypertension in chronickidney disease: navigating the evidence. Int J Hypertens 2011:132405.

12. Li Z, Su BH, Mi XH, Liu XR, Fan JM. Inflammation, oxidative stressand carbonyl stress in uremic patients. Sichuan Da Xue Xue Bao Yi XueBan 2006; 37: 123–125.

13. Cengiz K, Block AM, Hossfeld DK, Anthone R, Anthone S, SandbergAA. Sister chromatid exchange and chromosome abnormalities inuremic patients. Cancer Genet Cytogenet 1988; 36: 55–67.

14. Liu CS, Ko LY, Lim PS, Kao SH, Wei YH. Biomarkers of DNAdamage in patients with end-stage renal disease: mitochondrial DNAmutation in hair follicles. Nephrol Dial Transplant 2001; 16: 561–565.

15. Sandoval SB, Pastor S, Stoyanova E, et al. Genomic instability in chronicrenal failure patients. Environ Mol Mutagen 2012; 53: 343–349.

16. Galle J. Oxidative stress in chronic renal failure. Nephrol Dial Trans-plant 2001; 16: 2135–2137.

17. Himmelfarb J, Hakim RM. Oxidative stress in uremia. Curr OpinNephrol Hypertens 2003; 12: 593–598.

18. Vaziri ND. Oxidative stress in uremia: nature, mechanisms, and poten-tial consequences. Semin Nephrol 2004; 24: 469–473.

19. Stopper H, Schupp N, Bahner U, Sebekova K, Klassen A, Heidland A.Genomic damage in end-stage renal failure: potential involvement ofadvanced glycation end products and carbonyl stress. Semin Nephrol2004; 24: 474–478.

20. Stopper H, Meysen T, Böckenförde A, Bahner U, Heidland A,Vamvakas S. Increased genomic damage in lymphocytes of patientsbefore and after long-term maintenance hemodialysis therapy. Am JKidney Dis 1999; 34: 433–437.

21. Malachi T, Zevin D, Gafter U, Chagnac A, Slor H, Levi J. DNA repairand recovery of RNA synthesis in uremic patients. Kidney Int 1993;44: 385–389.

22. Vamvakas S, Bahner U, Becker P, Steinle A, Götz R, Heidland A.Impairment of DNA repair in the course of long-term hemodialysisand under cyclosporine immunosuppression after renal transplantation.Transplant Proc 1996; 28: 3468–3473.

23. Bagatini PB, Palazzo RP, Rodrigues MT, Costa CH, Maluf SW.Induction and removal of DNA damage in blood leukocytes of patientswith type 2 diabetes mellitus undergoing hemodialysis. Mutat Res2008; 657: 111–115.

24. Blasiak J, Arabski M, Krupa R, et al. DNA damage and repair in type 2diabetes mellitus. Mutat Res 2004; 554: 297–304.

25. Herman M, Ori Y, Chagnac A, et al. Spontaneous DNA repairincreases during hemodialysis. Nephron Clin Pract 2008;108: 188–193.

26. Collins AR. The comet assay for DNA damage and repair: principles,applications, and limitations. Mol Biotechnol 2004; 26: 249–261.

27. Stoyanova E, Sandoval SB, Zúñiga LA, et al. Oxidative DNAdamage in chronic renal failure patients. Nephrol Dia. Transplant2010; 25: 879–885.

28. Collins AR, Azqueta A. DNA repair as a biomarker in humanbiomonitoring studies; further applications of the comet assay. MutatRes 2012; 736: 122–129.

29. Sharma RA, Dianov GL. Targeting base excision repair to improvecancer therapies. Mol Aspects Med 2007; 28: 345–374.

30. Collins AR, Dusinská M, Horváthová E, Munro E, Savio M, Stĕtina R.Inter-individual differences in repair of DNA base oxidation, measuredin vitro with the comet assay. Mutagenesis 2001; 16: 297–301.

31. Cederberg H, Henriksson J, Binderup ML. DNA damage detected bythe alkaline comet assay in the liver of mice after oral administrationof tetrachloroethylene. Mutagenesis 2010; 25: 133–138.

32. Shapiro SS, Wilk MB. An analysis of variance test for normality(complete samples). Biometrika 1965; 52: 591–611.

33. Baute J, Depicker A. Base excision repair and its role in maintaininggenome stability. Crit Rev Biochem Mol Biol 2008; 43: 239–276.

34. Maisonneuve P, Agodoa L, Gellert R, et al. Cancer in patients ondialysis for end-stage renal disease: an international collaborativestudy. Lancet 1999; 354: 93–99.

35. Rao SN. Cancer screening in end-stage renal disease. Saudi J KidneyDis Transp 2009; 20: 737–740.

36. Jałoszyński P, Kujawski M, Czub-Swierczek M, Markowska J, SzyfterK. Bleomycin-induced DNA damage and its removal in lymphocytesof breast cancer patients studied by comet assay. Mutat Res 1997;385: 223–233.

37. Schmezer P, Rajaee-Behbahani N, Risch A, et al. Rapid screeningassay for mutagen sensitivity and DNA repair capacity in humanperipheral blood lymphocytes. Mutagenesis 2001; 16: 25–30.

38. Speit G, Rothfuss A. The comet assay: a sensitive genotoxicity testfor the detection of DNA damage and repair. Methods Mol Biol2012; 920: 79–90.

39. Hill J, Hazra T, Izumi T, Mitra S. Stimulation of human8-oxoguanine-DNA glycosylase by AP-endonuclease: potentialcoordination of the initial steps in base excision repair. Nucleic AcidsRes 2001; 29: 430–438.

40. Rusin P, Olszewski J, Morawiec-Bajda A, et al. Role of impaired DNArepair in genotoxic susceptibility of patients with head and neckcancer. Cell Biol Toxicol 2009; 25: 489–497.

182 e. stoyanova ET AL.

Copyright © 2013 John Wiley & Sons, Ltd. Cell Biochem Funct 2014; 32: 177–182.