name : abdul aziz a. bin dukhyil department : …...keywords: forensic science, dna reference...

Name : Abdul Aziz A. Bin Dukhyil

Department : Medical Laboratory Sciences

Publications:

Indexing Number of

Publications

ISI 2

Scopus 0

Others 0

TOTAL 2

TECHNICAL NOTE

CRIMINALISTICS

Mohammed H. Albujja,1 M.Sc.; Abdul Aziz Bin Dukhyil,2 Ph.D.; Abdul Rauf Chaudhary,1 M.Phil.;Ahmed Ch. Kassab,1 Ph.D.; Ahmed M. Refaat,1 Ph.D.; Saranya Ramesh Babu,1 M.Sc.;Mohammad K. Okla,3 Ph.D.; and Sachil Kumar,1 Ph.D.

Evaluation of Skin Surface as an AlternativeSource of Reference DNA Samples: A PilotStudy

ABSTRACT: An acceptable area for collecting DNA reference sample is a part of the forensic DNA analysis development. The aim of thisstudy was to evaluate skin surface cells (SSC) as an alternate source of reference DNA sample. From each volunteer (n = 10), six samples fromskin surface areas (forearm and fingertips) and two traditional samples (blood and buccal cells) were collected. Genomic DNA was extractedand quantified then genotyped using standard techniques. The highest DNA concentration of SSC samples was collected using the tape/forearmmethod of collection (2.1 ng/lL). Cotton swabs moistened with ethanol yielded higher quantities of DNA than swabs moistened with salicylicacid, and it gave the highest percentage of full STR profiles (97%). This study supports the use of SSC as a noninvasive sampling techniqueand as a extremely useful source of DNA reference samples among certain cultures where the use of buccal swabs can be considered sociallyunacceptable.

KEYWORDS: forensic science, DNA reference sample, skin surface cells, fingertip, forearm, swabbing, tape lift

Forensic DNA profiling, or DNA typing, is a powerful inves-tigative tool because with the exception of monozygotic twins,no two individuals share the same DNA. The likelihood of twopeople having the same DNA profile is often over one in aquadrillion, a number many times higher than the total humanpopulation.The DNA evidence collected from a crime scene can be

linked to a suspect or can eliminate a suspect from being associ-ated with a crime. If no suspect has been identified, a DNA pro-file from crime scene evidence can be entered into a national orregional DNA database to identify a possible suspect. The analy-sis is based on comparison of the results of biological evidencerecovered from the crime scene with reference samples (usuallyblood or oral swabs) obtained from suspect(s) (1). Because theprocess of analysis and interpretation of results depends on thecomparison between reference samples (known) and questionsamples (unknown), there is a critical need for a rapid and sim-ple collection method of DNA reference samples and streamlinedprocessing for downstream applications (2). The type andmethod of DNA reference sample collection is very important inthe progression of the science of DNA analysis. The analysis of

a DNA reference sample is not only valuable in the forensicidentification of individuals but also plays a fundamental role inlarge-scale epidemiological studies (3–12) as well as in the con-struction of human DNA databases (13–15).Biological material (blood or buccal cells) for forensic genetic

analyses of reference samples may be collected on FTA cards,which use mainly blood as a source of biological sample and arenot comfortable for donors, especially children and elderly peo-ple; this also applies to a saliva samples. Infants and childrenpose an additional challenge for the collection of DNA referencesamples (11). On the basis of quantity and quality, blood is thepreferred source for collecting DNA reference samples. How-ever, difficulty exists in obtaining blood samples to establish aDNA databank (4, 12). In addition, buccal swabs are unsuitablein some cultures and have some disadvantages, such as their vul-nerability to degradation when swabs have not properly driedbefore packaging and due to not containing an antibacterial solu-tion to eliminate bacterial species, which exist in the buccal cav-ity (12). In some special cases, such as bone marrowtransplantation, we cannot rely on buccal or blood samples asreference samples because they may give a mixed profile. Thereis also a possibility of maternal cell contamination of buccalsmear samples taken from nursing neonates (16,17).These issues prompted us to research different or alternative

sources of DNA reference samples. As an alternative source toblood or buccal swabs, or both, there have been many trials todiscover a noninvasive method to collect DNA reference sam-ples from various specimens of the human body, such as nails(18,19), hair (10), urine (10), and saliva (by mouthwash) (8).These trials attempted to overcome the disadvantages of

1College of Forensic Sciences, Naif Arab University for Security Sciences,P.O. Box: 6830, Riyadh 11452, Saudi Arabia.

2College of Applied Medical Sciences, Majmaah University, P.O. Box:1816, Majmaah 11952, Saudi Arabia.

3College of Science, King Saud University, P.O. Box: 2455, Riyadh11451, Saudi Arabia.

Received 28 Sept. 2016; and in revised form 11 Jan. 2017; accepted 16Jan. 2017.

1© 2017 American Academy of Forensic Sciences

J Forensic Sci, 2017doi: 10.1111/1556-4029.13468

Available online at: onlinelibrary.wiley.com

traditional samples and benefit from the advantages of new sam-ples, but the main reason was to insure participation of the lar-gest proportion of subjects or volunteers in the study. Someattempts have been made to collect reference samples from skinsurface cells (SSC) (13–15). These studies justify this because itis easy, simple, and in an accessible area. Such a method couldbe extremely useful among certain cultures (e.g., in the MiddleEast), where the use of buccal swabs can be considered invasiveor is otherwise socially unacceptable (14).New methods for detecting, preserving, extracting, and quanti-

fying DNA are continually being developed and are aidingrecovery of DNA from biological material found at crimescenes. The value of DNA evidence in detecting crime wasreviewed (20). We also utilized modern short tandem repeats(STR)-genotyping technologies which obtain DNA profiles fromvery small amount of nucleic acid such as a single fingerprint(21) or even from one cell (22).The purpose of this article was to evaluate the skin surface as

an alternative source of reference DNA samples. Therefore, weexamined two methods (swab and tape) using two moisteningsolutions (70% ethanol and 10% salicylic acid). Our selection of70% ethanol was based on the study of Franki et al. (23). Etha-nol is well suited for quick storage because it does not requireany time for drying and is naturally antibacterial. However, sali-cylic acid has a keratolyic effect, which helps to collect moreSSC (24,25).

Materials and Methods

Samples were collected from 10 volunteers (6 males and 4females, 21 to 59 years old) after reading the instructions (col-lection method) and signing the informed consent form. They

neither had any hereditary or skin disease nor had they under-gone blood transfusion. We prepared a SSC collection kit to col-lect the samples from volunteers to insure no contaminationoccurred during sample collection and storage (Fig. 1).We collected eight samples from each volunteer, six from skin

surface areas (forearm and fingertips), and two traditional samples(blood and buccal cells). All skin surface samples were collectedby the volunteers themselves to reduce contamination in front ofthe investigator. Three samples were collected from the forearmand three samples from the fingertips. These areas were based onprevious studies (13–15, 26) and were chosen due to the easewith which samples may be taken from them. Lifts were takenfrom the fingertips and the forearms. The procedure for collectingskin surface samples was by gently placing the 25-mm2 finger-print tape (Transparent Transoflips, Cat. No. SAW1, ♠ACEFingerprint equipment laboratories, Inc.) on the skin and remov-ing it (Fig. 1). This process was repeated to cover the completesampling area. Each tape was placed on the area approximatelyfive times until it became saturated; this is indicated by a changein color from transparent to opaque or when the tape did not stickto the skin anymore. We applied tape on one area, and the swabtechnique was performed on the other area.Swabbing an area requires a moistened swab to traverse the

whole target area multiple times with some pressure and rotationof the swab so that the full surface area of the swab can con-tribute to the collection. Swabs were moistened with 70% etha-nol or 10% salicylic acid. Swabs were rolled over the surface ofthe skin in horizontal lines at an angle of about 10–15° whilebeing rotated around their longitudinal axis. After sample collec-tion, the tapes were placed in a sterile envelope and the swabsin EZ-DRY swab boxes from SIRCHIE� Company (SIRCHIE,Youngsville, NC).

FIG. 1––Consumable required for the collection of samples. 1—Transparent Transoflips, 25 mm2 Fingerprint Tape manufactured by ACE Fingerprint Equip-ment Laboratories, Inc., 2—Sterile disposable lancet, 3—Sterile cotton swab, 4—The envelop of the tapes, 5—Medical plaster, 6—Alcohol pads, 7—Powderedgloves, 8—EZDRY swab boxes from SIRCHIE company.

2 JOURNAL OF FORENSIC SCIENCES

Buccal swab and blood were collected from all the volunteersas a reference sample for DNA profile comparison. All sampleswere stored at room temperature (20 � 2°C) for a maximum of8 days before DNA extraction was performed.The process of collection included giving each volunteer sam-

ple a number and a letter, the letter tagged to the traditional sam-ples and the number tagged to the skin surface samples. Bothsamples were processed separately to insure the efficacy ofmatching between skin samples and traditional samples. Thesample collection was performed at random times and days. Noprevious hand-cleaning treatment was applied.

DNA Extraction

25-mm2 tapes were cut into pieces then placed in a 2-mLtube. The heads of swabs were also cut into very small pieces tofacilitate the extraction by letting the extraction solution reachall cells trapped in the cotton to increase the quantity of DNAobtained. Seven hundred microliters of digestion buffer (1 MTris pH 8, 0.5 M EDTA, 20% SDS) and 25 lL of 20 mg/mLproteinase k were added to each sample, which was incubated at56°C overnight. Extraction was carried out both on the tapepieces, swabs of skin samples, and traditional samples using aphenol chloroform isoamyl alcohol (25:24:1 v/v) extractionmethod according to the method developed by Comey et al. (27)with some modifications. The swab and pieces of tape wereremoved from the tube and put in a spin basket (Cat. Q3258)(Sigma-Aldrich, St. Louis, MO), and the extract was placed in anew tube. The basket and new tube were centrifuged for 5 minat 17,900 9 g, and the basket was then discarded. About700 lL of phenol/chloroform was added to the extract, vortexedbriefly, and centrifuged for 5 min at 20,800 9 g. The aqueouslayer was transferred to a new tube with 500 lL of phenol/chloroform, vortexed, and centrifuged for another 5 min at20,800 9 g; the aqueous layer was placed in a new tube. Fivehundred lL of water-saturated n-butanol was added, vortexed,and centrifuged for 5 min at 20,800 9 g. Centricon 100 was usedfor purification according to the manufacturer’s recommendation(Millipore, Billerica, MA) (28).

DNA Quantification

DNA concentration was determined using the Applied Biosys-tems Quantifiler� Human DNA Quantification kit (Life Tech-nologies Inc., Carlsbad, CA) on Applied Biosystems 7500Real-Time PCR system and analyzed by SDS software v1.2.3program (Life Technologies Inc., Carlsbad, CA).

DNA Amplification and Genotyping

STR amplification was performed using the AmpFlSTR Iden-tifiler plus PCR Amplification kit on a Gene Amp 2720 Thermalcycler (Life Technologies, Inc.), according to the manufacturer’sinstructions. Electrophoresis was performed on an AppliedBiosystems 3130XL Genetic Analyzer and analyzed using Gen-eMaper� ID-X v 1.1 software (Life Technologies. Inc. Carlsbad,CA). The threshold for calling peaks in the analysis method wasset at 50 relative fluorescent units (rfu).

Statistical Analysis

All the data were analyzed using the Statistical Package forSocial Sciences (SPSS) version 22.0. The data are presented in

mean � SD and percentages. The Friedman test was used todetect differences in treatments across multiple test attempts. TheMann–Whitney U-Test was used to examine the differencesbetween two independent groups on a continuous scale. TheWilcoxon signed-rank test (nonparametric statistical hypothesistest) was used for comparing two related samples, matched sam-ples, or repeated measurements on a single sample to assesswhether their populations’ mean ranks differed. A p value lessthan 5% was considered statistically significant (p < 0.05).

Results

Traditional Samples DNA Concentration

The quantity of DNA extracted from volunteer’s blood sam-ples ranged from (1.2 to 36.4 ng/lL), while buccal samples ran-ged from (24.6 to 103.03 ng/lL).

SSC Samples DNA Concentration

The results revealed that the highest concentration was col-lected using the tape/forearm method of collection (2.1 ng/lL),while the lowest was from the salicylic acid/fingertip collectionmethod (0.003 ng/lL) (Table 1). Figure 2 shows the distributionof SSC samples according to four levels of concentration. Forty-seven samples (78%) were above 100 ng/lL, 30 of the 47 wereabove 200 ng/lL, while 13 samples (22%) came under thatlevel, and 5 of the 13 were less than 50 ng/lL.Results showed that the combination of wooden handle cotton

swabs moistened with ethanol/finger and tape/finger yielded thehighest quantities of DNA with values of 0.591 � 0.45 and0.550 � 0.66 ng/lL, respectively. By comparison, the combina-tion of swabs moistened with salicylic acid/forearm and ethanol/forearm yielded the lowest quantities of DNA with values of0.285 � 0.45 and 0.299 � 0.20 ng/lL, respectively. Analysisusing the Friedman test for quantity of DNA collected from SSCrevealed a significant interaction effect between collection meth-ods and area of collection (p = 0.014). The detailed resultshowed that only in the finger area, there are significant differ-ences between swabs moistened with ethanol and swabs moist-ened with SA. This indicates that different combinations ofcollection method/area of collection yielded different results(Fig. 3).

TABLE 1––DNA concentration (ng/lL) of samples collected from volunteersaccording to complex of the area and method of collection.

Volunteers

Method of Sample Collection

Arm Finger

70%EthanolSwab

10%Salicylic

Acid Swab Tape

70%EthanolSwab

10%Salicylic

Acid Swab Tape

1 0.173 0.056 0.099 0.262 0.187 0.1452 0.401 0.095 1.200 1.030 0.989 0.8033 0.700 0.147 0.185 0.498 0.037 0.4504 0.230 0.053 0.054 0.270 0.153 0.3485 0.074 0.036 0.142 0.830 0.003 0.3756 0.514 0.600 0.573 0.917 0.006 0.1187 0.211 0.122 0.467 0.280 0.106 0.5088 0.395 1.470 2.130 1.480 1.600 1.6609 0.162 0.168 0.561 0.117 0.012 0.11110 0.131 0.100 0.090 0.230 0.104 0.374Mean 0.161 0.150 0.550 0.406 0.181 0.489

ALBUJJA ET AL. . SKIN SURFACE AS A SOURCE OF REFERENCE DNA SAMPLES 3

In addition to comparing each collection method/area of col-lection combination, this study also conducted analysis to assessareas of collection with the same collection method. Comparisonusing the Wilcoxon test revealed that the finger area producedsignificantly higher yields than the forearm area when usingethanol as a moistening solution (p = 0.047). Control swabs andtapes showed no quantification level of DNA when quantifiedusing real-time quantification technology. Analysis using theMann–Whitney U-test revealed that there are no significant gen-der-related differences.

STR Analysis of SSC Samples

From the STR analysis of 60 SSC Samples, 43 Samples(71.7%) could be amplified and genotyped successfully, 10 Sam-ples (16.7%) showed partial profiles (<16 STRs, but alwaysmore than 10 markers), and only 7 samples (11.7%) showed par-tial profiles (1–9 markers) (Table 2).

To directly compare collection methods, the percentage ofdetected alleles was determined for each sample. The allele per-centages from the combination of body region/collection meth-ods decreased in the following order: forearm/ethanol swab,finger/ethanol swab, finger/tape, forearm/tape, forearm/salicylicacid, and finger/salicylic acid (Fig. 4).

Discussion and Conclusion

The results of the Samples Collection Questionnaire (SCQ)showed that most volunteers (9 of 10) prefer to give SSCsamples compared to blood samples and buccal swabs. Theonly volunteer who stated blood samples as a preference wasa diabetic patient, possibly due to his familiarity with bloodpunctures. This finding is consistent with a study conductedby Shannon (14), who stated that people in the Middle Eastprefer to give reference DNA samples from the surface of theskin.

FIG. 2––Distribution of samples according to the level of DNA concentration.

FIG. 3––DNA concentrations of different combinations of collection method and area of collection.


Six volunteers chose the forearm area over the fingers toprovide their reference samples. This choice supports our prefer-ence of the forearm area for collecting reference DNA samples.The advantage of this area is that it is less susceptible to con-tamination with secondary DNA, because this area is usuallycovered with cloth, especially in conservative societies. We rec-ommend conducting a survey on this issue with a larger numberof subjects.Two types of collection methods were used: tapes and swabs

(two moistening solutions, salicylic acid and ethanol). They werethen compared to establish which method and/or moisteningsolution were most effective in recovering skin cells. SSC sam-ples obtained from the forearm and fingertips are consideredeasy to collect and do not pose any hindrance to the sample col-lector or donor.DNA was extracted using standard organic extraction protocol

with some modifications (27), which gave the maximum amountof DNA from skin cells, especially after using the Centricon fil-ter. Other DNA extraction methods may result in even betterresults and less side effects from organic extraction (29,30).Somewhat surprisingly, swabs moistened with 70% ethanol gavebetter results than others which were moistened with salicylic

acid. Likewise, the quantitative PCR experiments showed no sig-nificant difference between forearms and fingers in percentage ofloci discovered.Results clearly showed that the differences in DNA concentra-

tion were affected by three factors: place of collection, extractionmethod, and tool of collection. In this study, the extraction wasthe same in all samples, even in the blood and buccal referencesamples. Other factors are comparable with previous studies (14,31). Phipps and Petricevic (32) mentioned that it may be morecomplicated than previously reported (33) to categorize a personas being a good or a bad shedder of skin cells and that if a goodshedder exists, they may significantly rarer than some have esti-mated. Therefore, we did not describe the skin characteristics ofeach subject, but we excluded any volunteer who suffered fromany skin diseases, according to the method used in the study ofKamphausen et al. (34), which concluded the considerableimpact of active skin diseases such as atopic dermatitis or psori-asis on the amplifiable DNA left by skin contact with surfaces.The samples which were collected from fingertips by tapes

gave higher DNA concentration followed by samples that werecollected from fingertips by swabs moistened with 70% ethanol.This result shows that finger areas shed cells more easily than

TABLE 2––Number of alleles detected (Percentage*) from skin surface cells.

Volunteers

Method of Sample Collection

Arm Finger

70% Ethanol Swab 10% Salicylic Acid Swab Tape 70% Ethanol Swab 10% Salicylic Acid Swab Tape

1 16 (100) 12 (75) 12 (75) 16 (100) 16 (100) 16 (100)2 16 (100) 16 (100) 16 (100) 16 (100) 16 (100) 16 (100)3 16 (100) 16 (100) 16 (100) 16 (100) 4 (28) 16 (100)4 16 (100) 12 (78) 16 (100) 16 (100) 14 (88) 16 (100)5 15 (97) 7 (47) 16 (100) 16 (100) 1 (9) 16 (100)6 16 (100) 16 (100) 16 (100) 16 (100) 1 (6) 16 (100)7 16 (100) 16 (100) 16 (100) 16 (100) 8 (50) 16 (100)8 16 (100) 16 (100) 16 (100) 16 (100) 16 (100) 16 (100)9 16 (100) 15 (97) 9 (56) 15 (94) 6 (38) 14 (91)10 16 (100) 11 (69) 13 (84) 16 (100) 16 (100) 16 (100)Mean 99.7 86.6 91.5 99.4 61.9 99.1SD 0.9 18.5 15.3 1.9 39.8 2.8

*Percentage calculated by counting the number of allele calls and dividing it by the total allele calls possible.

FIG. 4––Percentage of loci detected from skin surface cell (SSC) samples (%) according to method and area of collection.


the forearm area because this area is more moistened than theforearm area. This finding has been confirmed in studies con-ducted by Shannon and Li (14, 26).The concentration of DNA obtained by swabs was near to

concentrations obtained by tapes. There were significant differ-ences between DNA concentrations extracted from skin areaswith swabs moistened with ethanol and salicylic acid.There were significant differences between DNA concentra-

tions extracted from the fingertips and the forearm area withswabs moistened with ethanol. These differences favor the fore-arm as the optimum area for obtaining DNA samples. This mayhave happened due to restricting the influence of ethanol on thecollection of normal shedding skin cells. As mentioned earlier,these findings correspond with the previous two studies by Shan-non and Li (14, 26).As for the adhesive tape, results indicated that there are no

significant differences between the two areas of collection. Sta-tistically, there were no significant differences between tools ofcollection in the same area. This finding differs with Shannon’sstudy (14) but is consistent with de Bruin’s study (35).This study demonstrated that there were no statistically signifi-

cant differences between males and females in concentration ofDNA extracted from them, regardless of collection tools or placeof collection, except samples collected by swabs moistened withsalicylic acid, which yielded more DNA from males. This maybe due to the greater pressure generated by males than femaleswhich increased the effect of salicylic acid by increasing frictionon the skin.Kopka and Daly showed that no significant differences exist

between genders (13, 36). According to our knowledge, no stud-ies have used exfoliate substances on skin surfaces to collectDNA reference samples.It is worth mentioning that when the statistical tests show that

there are significant differences between two variables, these dif-ferences are large enough to decide that these differences areunlikely to occur by chance.This study holds more importance in situations where the sub-

jects providing a DNA reference sample have undergone a bonemarrow transplant (BMT) or frequent blood transfusion (37), aswell as in some touch DNA studies which may need a referencesample from the skin surface of subjects (38).All profiles generated from skin lifts and swabs were consis-

tent with their corresponding buccal swab and blood, indicatingthat all two-body regions and both recovery methods yielded anaccurate profile.Regarding the use of the DNA profiles obtained from skin

surfaces, the results demonstrate the feasibility of using thesekinds of samples as a DNA source to construct databases (13).Reference samples from skin surfaces should be collected fromthe forearm by swabs moistened with 70% ethanol. Tape mayserve as a tool to collect cells and evidence (39).The goal of this study was to determine whether skin surface

samples are a valuable type of DNA reference sample, and thisfinding proposes new methods of collecting DNA samplesother than traditional methods. This sampling method involvesthe use of a noninvasive sampling technique, the handling ofsamples without biological risk, and the storage of sampleswithout bacterial contamination, and the ease with which sam-ples are easy to transport. This method is also extremely usefulamong certain cultures (e.g., in the Middle East), where theuse of buccal swabs can be considered invasive or is otherwisesocially unacceptable. Tapes can be used as a less-invasivemethod for collecting biological evidence for forensic DNA

analysis. At present, there is a lack of automatic collection pro-cedures to acquire reference samples. This study opens a win-dow for more studies on SSC as a source of DNA referencesampling.However, there are some limitations in the study related to a

small sample size and pressure variations. Further detailed stud-ies addressing a larger sample size, and fixing the pressure vari-able (automatic collection of sample with machine), is requiredto reach more comprehensive conclusions.

Acknowledgments

Special thanks to the College of Forensic Sciences, Naif ArabUniversity for Security Sciences, Saudi Arabia, for providing thefacilities for this study.

References

1. Sharma AK. DNA profiling: social, legal, or biological parentage. IndianJ Hum Genet 2007;13(3):88–92.

2. Butler JM. Advanced topics in forensic DNA typing: methodology. SanDiego, CA: Elsevier/Academic Press, 2012.

3. Freeman B, Powell J, Ball D, Hill L, Craig I, Plomin R. DNA by mail:an inexpensive and noninvasive method for collecting DNA samplesfrom widely dispersed populations. Behav Genet 1997;27(3):251–7.

4. Hansen TV, Simonsen MK, Nielsen FC, Hundrup YA. Collection ofblood, saliva, and buccal cell samples in a pilot study on the Danishnurse cohort: comparison of the response rate and quality of genomicDNA. Cancer Epidemiol Biomarkers Prev 2007;16(10):2072–6.

5. Mulot C, St€ucker I, Clavel J, Beaune P, Loriot MA. Collection of humangenomic DNA from buccal cells for genetics studies: comparisonbetween cytobrush, mouthwash, and treated card. Biomed Res Int2005;2005(3):291–6.

6. Le Marchand L, Lum-Jones A, Saltzman B, Visaya V, Nomura AM,Kolonel LN. Feasibility of collecting buccal cell DNA by mail in acohort study. Cancer Epidemiol Biomarkers Prev 2001;10(6):701–3.

7. Freeman B, Smith N, Curtis C, Huckett L, Mill J, Craig IW. DNA frombuccal swabs recruited by mail: evaluation of storage effects on long-term stability and suitability for multiplex polymerase chain reactiongenotyping. Behav Genet 2003;33(1):67–72.

8. King IB, Abouta JS, Thornquist MD, Bigler J, Patterson RE, Kristal AR,et al. Buccal cell DNA yield, quality, and collection costs comparison ofmethods for large-scale studies. Cancer Epidemiol Biomarkers Prev2002;11(10):1130–3.

9. van Wieren-de Wijer DB, Maitland-van der Zee AH, de Boer A, BelitserSV, Kroon AA, de Leeuw PW, et al. Determinants of DNA yield andpurity collected with buccal cell samples. Eur J Epidemiol 2009;24(11):677–82.

10. Ghatak S, Muthukumaran RB, Nachimuthu SK. A simple method ofgenomic DNA extraction from human samples for PCR-RFLP analysis. JBiomol Tech 2013;24(4):224–31.

11. Zheng S, Ma X, Buffler PA, Smith MT, Wiencke JK. Whole genomeamplification increases the efficiency and validity of buccal cell genotyp-ing in pediatric populations. Cancer Epidemiol Biomarkers Prev 2001;10(6):697–700.

12. Harty LC, Garcia-Closas M, Rothman N, Reid YA, Tucker MA, HartgeP. Collection of buccal cell DNA using treated cards. Cancer EpidemiolBiomarkers Prev 2000;9(5):501–6.

13. Kopka J, Leder M, Jaureguiberry SM, Brem G, Boselli GO. New opti-mized DNA extraction protocol for fingerprints deposited on a specialself-adhesive security seal and other latent samples used for human iden-tification. J Forensic Sci 2011;56(5):1235–40.

14. Shannon P. The optimized use and recovery of DNA from tape lifts [the-sis]. East Lasting, MI: Michigan State University, 2011.

15. Zamir A, Oz C, Wolf E, Vinokurov A, Glattstein B. A possible sourceof reference DNA from archived treated adhesive lifters. J Forensic Sci2004;49(1):68–70.

16. Babovic-Vuksanovic D, Michels VV, Law ME, Bailey R, Wyatt WA,Lindor NM, et al. Guidelines for buccal smear collection in breast-fedinfants. Am J Med Genet 1999;84(4):357–60.

17. Babovic-Vuksanovic D, Michels VV, Law ME, Lindor NM, Jalal SM.Maternal cell contamination of buccal smear samples in nursing neo-nates. Clin Genet 1998;53(2):114–8.


18. Hogervorst JG, Godschalk RW, van den Brandt PA, Weijenberg MP,Verhage BA, Jonkers L, et al. DNA from nails for genetic analyses inlarge-scale epidemiologic studies. Cancer Epidemiol Biomarkers Prev2014;23(12):2703–12.

19. Klassen TL, von R€uden EL, Drabek J, Noebels JL, Goldman AM. Compar-ative analytical utility of DNA derived from alternative human specimensfor molecular autopsy and diagnostics. J Mol Diagn 2012;14(5):451–7.

20. Bond JW. Value of DNA evidence in detecting crime. J Forensic Sci2007;52(1):128–36.

21. Van Oorschot RA, Jones MK. DNA fingerprints from fingerprints. Nat-ure 1997;387(6635):767.

22. Findlay I, Taylor A, Quirke P, Frazier R, Urquhart A. DNA fingerprint-ing from single cells. Nature 1997;389(6651):555–6.

23. Franke N, Augustin C, P€uschel K. Optimization of DNA-extraction andtyping from contact stains. Forensic Sci Int: Genet Suppl Series 2008;1(1):423–5.

24. Huber C, Christophers E. “Keratolytic” effect of salicylic acid. Arch Der-matol Res 1977;257(3):293–7.

25. Imayama S, Ueda S, Isoda M. Histologic changes in the skin of hairlessmice following peeling with salicylic acid. Arch Dermatol 2000;136(11):1390–5.

26. Li RC, Harris HA. Using hydrophilic adhesive tape for collection of evi-dence for forensic DNA analysis. J Forensic Sci 2003;48(6):1318–21.

27. Comey CT, Koons BW, Presley KW, Smerick JB, Sobieralski CA, Stan-ley DM, et al. DNA extraction strategies for amplified fragment poly-morphism analysis. J Forensic Sci 1994;39(5):1254–69.

28. http://kirschner.med.harvard.edu/files/protocols/Millipore_Centricons.pdf.29. Agency for Toxic Substances and Disease Registry. Toxicological profile

for chloroform. Atlanta, GA: U.S. Department of Health and Human Ser-vices, Public Health Service, 1997.

30. Agency for Toxic Substances and Disease Registry. Toxicological profilefor phenol. Atlanta, GA: U.S. Department of Health and Human Ser-vices, Public Health Service, 2008.

31. Alessandrini F, Cecati M, Pesaresi M, Turchi C, Carle F, Tagliabracci A.Fingerprints as evidence for a genetic profile: morphological study on

fingerprints and analysis of exogenous and individual factors affectingDNA typing. J Forensic Sci 2003;48(3):586–92.

32. Phipps M, Petricevic S. The tendency of individuals to transfer DNA tohandled items. Forensic Sci Int 2007;168(2–3):162–8.

33. Lowe A, Murray C, Whitaker J, Tully G, Gill P. The propensity of indi-viduals to deposit DNA and secondary transfer of low level DNA fromindividuals to inert surfaces. Forensic Sci Int 2002;129(1):25–34.

34. Kamphausen T, Schadendorf D, Von Wurmb-Schwark N, Bajanowski T,Poetsch M. Good shedder or bad shedder – the influence of skin diseaseson forensic DNA analysis from epithelial abrasions. Int J Legal Med2012;126:179–83.

35. de Bruin KG, Verheij SM, Veenhoven M, Sijen T. Comparison of stub-bing and the double swab method for collecting offender epithelial mate-rial from a victim’s skin. Forensic Sci Int Genet 2012;6(2):219–23.

36. Daly DJ, Murphy C, McDermott SD. The transfer of touch DNA fromhands to glass, fabric and wood. Forensic Sci Int Genet 2012;6(1):41–6.

37. Pope S, Chapman H, Lambert J. The effect of bone marrow transplantson DNA profiles; a case example. Sci Justice 2006;46(4):231–7.

38. Lempan A, Riproumsup K, Panvisavas N, Kusamran T. DNA recoveryfrom forensic clothing samples by tape-lift. In: Faculty of GraduateStudies. Proceedings of the 8th national grand research conference; 2007Sept 7-8; Bangkok, Thailand: Mahidol University, 2007; http://forensic.sc.mahidol.ac.th/proceeding/49_Aree.pdf.

39. Zamir A, Glattstein B, Springer E. Fingerprints and DNA: STR typingof DNA extracted from adhesive tape after processing for fingerprints. JForensic Sci 2000;45(3):687–8.

Additional information and reprint requests:Mohammed H. Albujja, M.Sc.College of Forensic SciencesNaif Arab University for Security SciencesP.O. Box: 6830 Riyadh 11452Saudi ArabiaE-mail: [email protected]


http://forensic.sc.mahidol.ac.th/proceeding/49_Aree.pdf

http://forensic.sc.mahidol.ac.th/proceeding/49_Aree.pdf

1926

Ischemic stroke patients are at high risk of having new cardio-vascular events. The cumulative risk of vascular recurrence after

a first stroke reaches up to 48% at 10 years.1 Acetylsalicylic acid (aspirin) and clopidogrel are the most widely used antiplatelet agents in secondary prevention of stroke. Aspirin has been associ-ated with a relative risk reduction of 25% in secondary cardiovas-cular events and 13% to 15% in secondary prevention of stroke.2 A high platelet reactivity phenotype can be present in ≤60% of aspirin-treated patients and has been associated with an increased risk of recurrent ischemic events, and although it is a substantially heritable phenotype, very few genetic determinant have been identified. Furthermore, the causes of aspirin failure goes beyond high platelet reactivity and are still largely unknown.3

Only polymorphisms in PEAR1 have been confirmed as genetic risk factors of aspirin failure.4 The rs12041331 polymorphism is associated with platelet aggregation, increased platelet reactivity,5 and vascular recurrence in aspirin-treated patients,4 although these polymorphisms do not explain the whole variability observed in aspirin resis-tance or in vascular responsiveness. Other genomic regula-tions, such as epigenetics modifications could be associated with the risk of vascular recurrence in patients treated with aspirin.

We aimed to analyze the whole epigenome of stroke patients treated with aspirin, to find altered methylation sites associated with vascular recurrence.

Background and Purpose—Despite great efforts by pharmacogenetic studies, the causes of aspirin failure to prevent the recurrence of ischemic events remain unclear. Our aim was to study whether epigenetics could be associated with the risk of vascular recurrence in aspirin-treated stroke patients.

Methods—We performed an epigenetic joint analysis study in 327 patients treated with aspirin. In the discovery stage, we performed a nested case–control study in 38 matched ischemic stroke patients in whom 450 000 methylation sites were analyzed. Nineteen patients presented vascular recurrence after stroke, and 19 matched patients did not present vascular recurrence during the first year of follow-up. In a second stage, 289 new patients were analyzed by EpiTYPER.

Results—The following 3 differentially methylated candidate CpG sites, were identified in the discovery stage and analyzed in the second stage: cg26039762 (P=9.69×10−06, RAF1), cg04985020 (P=3.47×10−03, PPM1A), and cg08419850 (P=3.47×10−03, KCNQ1). Joint analysis identified an epigenome-wide association for cg04985020 (PPM1A; P=1.78×10−07), with vascular recurrence in patients treated with aspirin.

Conclusions—The pattern of differential methylation in PPM1A is associated with vascular recurrence in aspirin-treated stroke patients. (Stroke. 2016;47:1926-1929. DOI: 10.1161/STROKEAHA.116.013340.)

Key Words: aspirin ◼ methylation ◼ phenotype ◼ platelet aggregation ◼ stroke

PPM1A Methylation Is Associated With Vascular Recurrence in Aspirin-Treated Patients

Cristina Gallego-Fabrega, MSc; Caty Carrera, MD, MSc; Jean-Luc Reny, MD, PhD; Pierre Fontana, MD, PhD; Agnieszka Slowik, MD, PhD; Joanna Pera, MD, PhD; Alessandro Pezzini, MD; Gemma Serrano-Heras, PhD; Tomás Segura, MD, PhD;

Abdul-Aziz A. Bin Dukhyil, PhD; Joan Martí-Fàbregas, MD, PhD; Elena Muiño, MD; Natalia Cullell, MSc; Joan Montaner, MD, PhD; Jerzy Krupinski, MD, PhD;

Israel Fernandez-Cadenas, PhD

Received March 8, 2016; final revision received May 3, 2016; accepted May 17, 2016.From the Stroke Pharmacogenomics and Genetics, Fundació Docència i Recerca MutuaTerrassa, Hospital Mútua de Terrassa, Terrassa, Spain (C.G.-F.,

E.M., N.C., I.F.-C.); School of Medicine, University of Barcelona, Barcelona, Spain (C.G.-F.); Neurovascular Research Laboratory, Vall d’Hebron Institute of Research, Universitat Autonoma de Barcelona, Barcelona, Spain (C.C., J.M.); Division of Internal Medicine and Rehabilitation (J.-L.R.) and Division of Angiology and Haemostasis (P.F.), Geneva University Hospitals, Switzerland; Geneva Platelet Group, Faculty of Medicine, Geneva, Switzerland (J.-L.R., P.F.); Department of Neurology, Jagiellonian University Medical College, Krakow, Poland (A.S., J.P.); Dipartimento di Scienze Cliniche e Sperimentali, Clinica Neurologica, Università degli Studi di Brescia, Brescia, Italy (A.P.); Neurology Department, Albacete Hospital, Albacete, Spain (G.S.-H., T.S.); College of Applied Medical Sciences, Majmaah University, Saudi Arabia (A.-A.A.B.D.); Department of Neurology, Hospital de la Santa Creu i Sant Pau, IIB-Sant Pau, Barcelona, Spain (J.M.-F.); Neurology Service, Hospital Universitari Mútua Terrassa, Terrasa, Spain (J.K.); and School of Healthcare Science, Manchester Metropolitan University, Manchester, UK (J.K.).

The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA. 116.013340/-/DC1.

Correspondence to Israel Fernández-Cadenas, PhD, Fundació Docència i Recerca MutuaTerrassa, C/Sant Antoni 19, 08221 Terrassa, Spain. E-mail [email protected]

© 2016 American Heart Association, Inc.

Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.116.013340

by guest on January 14, 2017http://stroke.ahajournals.org/

Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from


Dow

nloaded from

mailto:[email protected]

http://stroke.ahajournals.org/
















Gallego-Fabrega et al Epigenetic Regulation of PPM1A 1927


Study PatientsThirty-eight subjects from a cohort of 1900 ischemic stroke patients recruited prospectively at Vall d’Hebron Hospital (Spain), who started aspirin treatment after a first ischemic stroke, were ana-lyzed. Nineteen participants presented a new vascular event (isch-emic stroke, myocardial infarction, peripheral vascular disease, or cardiovascular death) during the first year of follow-up and were matched one-to-one (age, sex, TOAST [Trial of Org 10172 in Acute Stroke Treatment]) with 19 participants without a vascular event. Second-stage analysis was performed in 289 prospective enrolled participants from 4 independent cohorts with ischemic stroke or ischemic atherothrombotic disease; 29 presented a vascular recur-rence within 1 year of follow-up (Table I in the online-only Data Supplement).

Discovery: HumanMethylation450 AssayDNA was extracted from blood samples using standard methods. Samples were obtained during the first 24 hours after stroke onset, before initial aspirin administration.

Genome-wide DNA methylation (DNAm) was assessed using the HumanMethylation450 (Illumina) assay. All samples (n=38)

were processed in a single working batch. Quality Control and dif-ferentially methylated CpG (DMC) sites analysis were performed as described elsewhere,6 using the R computing environment (3.1.3 ver-sion; Table II in the online-only Data Supplement).

Rs120411331 polymorphism (PEAR1) was checked to discover whether it could be a confounding factor in the EWAS (Epigenome-Wide Association Study) analysis.

CpGs SelectionOne DMC with epigenome-wide significance7 (RAF1; P<10−06) was selected for further replication. Candidate DMCs for second-stage analysis were defined as: top 100 significant DMCs from the uni-variate analysis, which also appear in the top 100 DMCs after multi-variable analysis. Multivariable analyses were adjusted for principal components and DNAm potential covariates (age, sex, and current smoking). From a total of 36 candidate CpGs, those mapped on genes previously associated with cardiovascular disease were considered for further analysis. A PubMed search (http://www.ncbi.nlm.nih.gov/pubmed) was undertaken using the following key words: aspirin, atherosclerosis, cardiovascular, stroke, and vascular. Finally, from 4 candidate DMCs, only 2 CpGs (PPM1A and KCNQ1) were located in regions suitable for EpiTYPER analysis (Figure IV in the online-only Data Supplement).

Figure. A, Manhattan plot for the EWAS (Epigenome-Wide Association Study) of aspirin-treated stroke patients. x axis: chromosome position; y axis: −log10 of P values. B, Hierarchical cluster analysis of most significant differentially methylated CpG associated (Table) with vascular recurrence. CA indicates recurrence patients; and CO, nonrecurrence patients.


Dow

nloaded from


1928 Stroke July 2016

Second-Stage EpiTYPER AssayAn averaged 400-bp region was sequenced using EpiTYPER (Sequenom) around each CpG selected from the discovery stage in 289 new patients.

Statistical AnalysisDMC sites were calculated using the Mann–Whitney U test and multivariable generalized linear analysis. Sample size calculation indicated that 19 subjects per condition were needed to achieve a 10−05 significance level and 80% statistical power. A joint analysis strategy using the DerSimonian–Laid test was performed to increase the statistical power in the 2-stage study as previously suggested.8

Complete methodology is available in the Materials and Methods section in the online-only Data Supplement.

ResultsDiscoveryDNAm levels were obtained for 485 577 CpGs sites (Figure [A]). After data preprocessing and quality control analysis one DMC was associated with vascular recurrence at epigenomewide level (P<10−06), covariant adjustment analysis was also performed (Table; Table IV in the online-only Data Supplement). Table IV in the online-only Data Supplement shows the 36 candidate CpG sites which were best ranked in both univariate and covari-ate analyses, including cg04985020 (PPM1A) and cg08419850 (KCNQ1). Figure (B) shows a hierarchical cluster analysis, which was able to distinguish vascular recurrence patients from nonvascular patients, with only 2 samples misclassified.

Second-Stage AnalysisSecond-stage analysis was performed for the epigenome-wide significant CpG, cg26039762 (RAF1; P=9.69×10−06),

and 2 additional candidate CpG sites, cg04985020 (PPM1A; P=3.46×10−04) and cg08419850 (KCNQ1; P=3.46×10−04). All 3 CpGs map genes involved in atherosclerosis and vascular process. The cg04985020 (PPM1A) was associated with vas-cular recurrence (P=0.027). The same cg04985020 methyla-tion pattern was observed in each of the 4 validation-stage cohorts and when nonstroke and cardioembolic patients were excluded from the analyses (Materials and Methods section in the online-only Data Supplement). Additionally, the fol-lowing 3 CpG sites surrounding cg26039762 (RAF1) were de novo associated with vascular recurrence: RAR1_10 (P=0.0015), RAF1_6 (P=0.0043), and RAF1_3 (P=0.01; Table VI in the online-only Data Supplement). However, cg26039762 (RAF1) and cg08419850 (KCNQ1) were not associated in this second stage.

Joint analysis of discovery and replication stage revealed an epigenome-wide statistically significant meta P value8 for PPM1A (meta P=1.78×10−07) but not for RAF1. PPM1A was independently associated with vascular recurrence when ana-lyzed possible DNAm confounding factors (age, sex, and cur-rent smoking; P=5.74×10−03) and also PEAR1 polymorphism (rs12041331; P=9.82×10−03). Additionally, PPM1A methyla-tion levels were not influenced by cell-type proportions.

DiscussionPatients with vascular recurrence presented higher meth-ylation levels of cg04985020 CpG (PPM1A) compared with nonrecurrent patients. Despite cg04985020 not being sig-nificant on the epigenome-wide level, in the discovery stage (P=3.46×10−04), the joint analysis revealed a meta P value of 1.78×10−07 that was significant on the epigenome-wide level.7 These results suggest an association between DNAm levels of PPM1A and vascular recurrence during on-aspirin treatment, regardless of the disease (cardiovascular or stroke). The asso-ciation was not confounded by common DNAm confounding factors or PEAR1 polymorphism. The cg26039762 (RAF1) was not replicated, but 3 surrounding CpGs were associ-ated de novo with vascular recurrence. Patients with vascu-lar recurrence presented lower methylation levels of these 3 surrounding CpGs compared with nonvascular recurrence patients, similar to cg26039762 (RAF1) in the discovery stage. Additionally, hierarchical cluster analysis showed a potential role of DMCs as a model to classify patients within highest or lowest risk of vascular recurrence.

Protein phosphatase magnesium dependent 1A (PPM1A) is involved in the regulation of transforming growth factor (TGF)-β1 signaling and plasminogen activator inhibitor-1 transcription.9 Plasminogen activator inhibitor-1 levels are associated with increased risk of cardiovascular events, par-ticularly in the context of elevated tissue TGF-β1.10 Raf1 proto-oncogene, serine/threonine kinase (RAF1), has been described in many vascular diseases, such as vascular dementia, angio-genesis processes, or cardiovascular events.11 RAF1 has also been described as taking part in signal transduction in the TGF-β/Smad pathway.12 Furthermore, aspirin administration has been associated with decreasing TGF-β1 serum levels in hypercholesterolemic rats.13 We hypothesize that PPM1A- and RAF1-altered methylation could be associated with vascular

Table. DMC Sites Associated With Vascular Recurrence (P<10−05)

DMCs Associated With Vascular Recurrence. Univariate Analysis.

CpG ID Gene P Value

cg26039762 RAF1 9.69×10−6

cg00094487 SPG21 1.20×10−5

cg26568880 RDBP; SKIV2L 1.47×10−5

cg20702204 CHFR 2.19×10−5

cg22352818 … 2.19×10−5

cg01112035 … 2.19×10−5

cg16966962 LMOD1 3.23×10−5

cg09747456 PANK1 3.90×10−5

cg07580707 XPC; LSM3 4.69×10−5

cg02533998 … 5.62×10−5

cg00932677 … 6.72×10−5

cg08859247 … 6.72×10−5

cg00608860 MTHFSD; FLJ30679 6.72×10−5

cg16933664 LOC100130987 8.02×10−5

cg08808677 … 8.02×10−5

cg26577201 CRYBA2 8.02×10−5

DMC indicates differentially methylated CpG.


Dow

nloaded from


Gallego-Fabrega et al Epigenetic Regulation of PPM1A 1929

recurrence because of the regulation of the TGF-β pathway; however, further studies are needed to confirm this hypothesis.

Pharmacoepigenomics is a growing field that could be used in the future to assist in personalized antiplatelet therapy or to find potential new drug targets. Our study suggests a remark-able role for epigenetics in the modulation of aspirin response.

LimitationsThe small sample size in the discovery stage impeded the achievement of more significant values; however, joint anal-ysis was useful to improve the results of our study. mRNA studies are needed to confirm the biological significance of our results.

Sources of FundingThis study was supported by Miguel Servet grant (CP12/03298, Instituto de Salud Carlos III and Fondo Europeo de Desarrollo Regional), Sheikh Abdullah Bin Abdul Mohsen Al Tuwaijri Chair for Applied Research in Stroke, Majmaah University, Saudi Arabia, and SEDMAN Project. J. Pera is supported by the Jagiellonian University Medical College by grant no. K/ZDS/003844. J. Martí-Fabregas is supported by Redes Temáticas de Investigación Cooperativa en Salud RD12-0014-0002.

DisclosuresP. Fontana discloses honoraria from Bayer. The other authors report no conflicts.

References 1. van Wijk I, Kappelle LJ, van Gijn J, Koudstaal PJ, Franke CL, Vermeulen

M, et al; LiLAC study group. Long-term survival and vascular event risk after transient ischaemic attack or minor ischaemic stroke: a cohort study. Lancet. 2005;365:2098–2104. doi: 10.1016/S0140-6736(05)66734-7.

2. Greer DM. Aspirin and antiplatelet agent resistance: implications for prevention of secondary stroke. CNS Drugs. 2010;24:1027–1040. doi: 10.2165/11539160-000000000-00000.

3. Pettersen AA, Arnesen H, Seljeflot I. A brief review on high on-aspirin residual platelet reactivity. Vascul Pharmacol. 2015;67-69:6–9. doi: 10.1016/j.vph.2015.03.018.

4. Lewis JP, Ryan K, O’Connell JR, Horenstein RB, Damcott CM, Gibson Q, et al. Genetic variation in PEAR1 is associated with platelet aggrega-tion and cardiovascular outcomes. Circ Cardiovasc Genet. 2013;6:184–192. doi: 10.1161/CIRCGENETICS.111.964627.

5. Würtz M, Nissen PH, Grove EL, Kristensen SD, Hvas AM. Genetic determinants of on-aspirin platelet reactivity: focus on the influ-ence of PEAR1. PLoS One. 2014;9:e111816. doi: 10.1371/journal.pone.0111816.

6. Gallego-Fabrega C, Carrera C, Reny JL, Fontana P, Slowik A, Pera J, et al. TRAF3 epigenetic regulation is associated with vascular recurrence in patients with ischemic stroke. Stroke. 2016;47:1180–1186. doi: 10.1161/STROKEAHA.115.012237.

7. Rakyan VK, Down TA, Balding DJ, Beck S. Epigenome-wide associa-tion studies for common human diseases. Nat Rev Genet. 2011;12:529–541. doi: 10.1038/nrg3000.

8. Skol AD, Scott LJ, Abecasis GR, Boehnke M. Joint analysis is more efficient than replication-based analysis for two-stage genome-wide association studies. Nat Genet. 2006;38:209–213. doi: 10.1038/ng1706.

9. Samarakoon R, Chitnis SS, Higgins SP, Higgins CE, Krepinsky JC, Higgins PJ. Redox-induced Src kinase and caveolin-1 signaling in TGF-β1-initiated SMAD2/3 activation and PAI-1 expression. PLoS One. 2011;6:e22896. doi: 10.1371/journal.pone.0022896.

10. Otsuka G, Agah R, Frutkin AD, Wight TN, Dichek DA. Transforming growth factor beta 1 induces neointima formation through plasminogen activator inhibitor-1-dependent pathways. Arterioscler Thromb Vasc Biol. 2006;26:737–743. doi: 10.1161/01.ATV.0000201087.23877.e1.

11. Liu Z, Liu Y, Li L, Xu Z, Bi B, Wang Y, et al. MiR-7-5p is frequently downregulated in glioblastoma microvasculature and inhibits vas-cular endothelial cell proliferation by targeting RAF1. Tumour Biol. 2014;35:10177–10184. doi: 10.1007/s13277-014-2318-x.

12. Watanabe-Takano H, Takano K, Hatano M, Tokuhisa T, Endo T. DA-Raf-mediated suppression of the Ras–ERK pathway is essential for TGF-β1-induced epithelial-mesenchymal transition in alveolar epithe-lial type 2 cells. PLoS One. 2015;10:e0127888. doi: 10.1371/journal.pone.0127888.

13. Mohamed AR, El-Hadidy WF, Mannaa HF. Assessment of the pro-phylactic role of aspirin and/or clopidogrel on experimentally induced acute myocardial infarction in hypercholesterolemic rats. Drugs R D. 2014;14:233–239. doi: 10.1007/s40268-014-0059-3.


Dow

nloaded from


and Israel Fernandez-CadenasDukhyil, Joan Martí-Fàbregas, Elena Muiño, Natalia Cullell, Joan Montaner, Jerzy KrupinskiJoanna Pera, Alessandro Pezzini, Gemma Serrano-Heras, Tomás Segura, Abdul-Aziz A. Bin Cristina Gallego-Fabrega, Caty Carrera, Jean-Luc Reny, Pierre Fontana, Agnieszka Slowik,

Methylation Is Associated With Vascular Recurrence in Aspirin-Treated PatientsPPM1A

Print ISSN: 0039-2499. Online ISSN: 1524-4628 Copyright © 2016 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Stroke doi: 10.1161/STROKEAHA.116.013340

2016;47:1926-1929; originally published online June 14, 2016;Stroke.

http://stroke.ahajournals.org/content/47/7/1926World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://stroke.ahajournals.org/content/suppl/2016/06/27/STROKEAHA.116.013340.DC1.htmlData Supplement (unedited) at:

http://stroke.ahajournals.org//subscriptions/

is online at: Stroke Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer process is available in the

Request Permissions in the middle column of the Web page under Services. Further information about thisOnce the online version of the published article for which permission is being requested is located, click

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Strokein Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:


Dow

nloaded from

http://stroke.ahajournals.org/content/47/7/1926

http://stroke.ahajournals.org/content/suppl/2016/06/27/STROKEAHA.116.013340.DC1.html

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://stroke.ahajournals.org//subscriptions/


Supplemental Material


Clinical protocol

From a cohort of 1,900 stroke patients from Vall d’Hebron University Hospital (Barcelona,

Spain), 38 subjects were selected. 19 ischemic stroke patients treated with aspirin with a

new vascular event (new ischemic stroke, myocardial infarction, peripheral vascular

disease and cardiovascular death) and 19 ischemic stroke patients treated with aspirin

without a new vascular event were matched one-to-one by age (±7years), sex, and TOAST1

classification (Table I).

Cases were defined as ischemic stroke patients under aspirin treatment which had vascular

recurrence in the first year of follow up with a good adherence to the treatment measured

by the Morisky-Green test2, controls were defined as ischemic stroke patients under aspirin

treatment with a good adherence to the treatment measured using the Morisky-Green test

and without vascular recurrence during the first year of follow up. Vascular recurrence was

described as new ischemic stroke, myocardial infarction, peripheral vascular disease or

cardiovascular death and was detected by phone calls every three months or direct clinical

reporting.

Second stage analysis was performed on 289 new samples from 4 international cohorts,

three ischemic stroke patients’ cohorts and one cardiovascular disease patients’ cohort.

The second stage cohort included 29 patients with a new vascular event and 261 patients

without a new vascular event during the first year of follow up (Table II). The Italian,

Spanish and Polish cohorts consist of consecutively recruited patients that started aspirin

treatment after the first ischemic stroke. Vascular recurrence information was available from

all patients. The Geneva cohort consists of consecutively recruited patients with

symptomatic documented ischemic atherothrombotic disease (coronary artery disease

[CAD], ischemic cerebrovascular disease and/or peripheral artery disease) treated with

aspirin and/or clopidogrel for < 5 years. Information about clinical ischemic events over an

ongoing 3-year follow-up was available from all patients. Only aspirin treated patients were

selected from this cohort.

The local ethical committee approved the study (PR(AG) 03/2007).

Discovery, HumanMethylation450 assay.

Genome-wide DNA methylation was assessed using the Infinium HumanMethylation450

BeadChip (Illumina Inc., San Diego Ca). All samples were processed in a single working

batch.

All pre-processing, correction and normalization steps were implemented using the R

statistical computing environment (3.1.3 version) with Bioconductor packages, (Table III).

Plots were also produced using R functions. Quality control metrics were examined to

determine the success of the bisulphite conversion and subsequent array hybridization.

Fluorescence intensities were imported from GenomeStudio, then probe filtering was

performed to remove probes that have failed to hybridize (detection p-value>0.05) and that

were not represented by a minimum of 3 beads on the array, as described elsewere3,4. CpG

sites containing documented single nucleotide polymorphisms (SNPs) were also excluded5.

Multidimensional scaling (MDS) plots were used to evaluate gender outliers based on

chromosome X data. MDS and PC were also used to check unknown population structures

within the sample. Then, probes mapping to sex chromosomes were removed. We also

checked the white cell count (neutrophils, lymphocytes and monocytes) as a possible

confounding factor. Finally, a subset quantile normalization was performed using a

background adjustment between-array normalization and a dye bias correction, following

previous recommendations4.

The methylation level of each cytosine was expressed as a Beta value (β-value), ranging

between 0 and 1, unmethylated to completely methylated respectively. Differentially

methylated CpG sites (DMCs) were analyzed using the non-parametric Mann-Whitney U

test for independent samples, p-values<10-06 were selected as statistically significant6 and

p-values<10-05 as having nominal association. Multivariable generalized linear analyses

adjusting for Principal Components and DNA methylation potential covariates (age, sex and

current smoking) were also used.

Second stage, EpiTYPER assay.

Quantitative DNA methylation analysis was performed using the MassARRAY EpiTYPER

(Sequenom, San Diego, CA, USA), on selected CpGs. An averaged 400 pb region was

sequenced around each selected CpGs, containing multiple CpG sites and aggregated

CpG sites. Target-specific primers were designed using the online Epidesigner software

(http://www.epidesigner.com), table V. The quantitative methylation data obtained for each

CpG site, or aggregates of multiple CpG sites, were analyzed with the EpiTYPER software

(Sequenom). Statistical analyses were performed using R (3.1.3 version). P-values<0.05

were considered as statistically significant, after the Mann-Whitney U test.

PEAR1 analysis.

OmniQuad Human 1M array (Illumina) was used to genotype the samples. Data was

analyzed following WTCC recommendation and the specific methodological guidelines from

the Broad Institute and Harvard University. Quality controls (QC) were performed before the

analyses: Missings (-0.3) and checksex QCs were performed for each individual. Only

individuals with Caucasian ethnicity were included. We also tested for population

stratifications by analyzing the individual IBS, performing MDS plots and adjusting for

principal components (PCs). Hardy-Weinberg (p-value>10-08), missingness (0.01), Mishap

(p-value>10-09), were analyzed for each SNP.

After the QC process, the genotypic results were analyzed by Plink, Haploview, STATA,

SNPtest and GTOOL software solutions.

Statistical analysis.

Sample size calculation was performed using the pwr package (version: 1.1-2) from

Bioconductor (www.bioconductor.org). Nineteen subjects per condition were needed in

order to achieve a 10-05 significance level and 80% statistical power, considering a Cohen's

effect size = 0.9.

Results

Epigenome-wide analysis.

DNA methylation levels were obtained for 485.577 CpG sites across the whole genome.

After pre-processing and QC analysis, 34.999 probes and three sample were removed from

further analysis: 1.209 CpGs with detection p-value>0.05, 1.765 CpGs with a beadcount

lower than 3 in more than 5% of the samples, 20.935 CpGs overlapping with SNPs and

11.090 CpGs located on sex chromosomes. One samples was removed because 1% of the

CpG sites had detection p-values>0.05 (Figure I.A) and two more samples were removed

because they showed sex discrepancies (Figure I.B).

A heatmap of the 16 differentially methylated CpGs is shown in the supplemental material,

Figure II.

Second stage analysis.

Higher methylation levels of cg04985020 (PPM1A) were observed in vascular recurrence

patients treated with aspirin in each of the four validation-stage cohorts. The same pattern

was observed after excluding the Geneva's non-stroke patients and the cardioembolic

stroke patients, although the association was not statistically significant due to a reduction

of the sample size (Figure III). Indicating that cg04985020 (PPM1A) association with

vascular recurrence may be independent of the cardiovascular disease.

Further analysis

Frequency distribution of the rs12041331 risk allele was equally distributed between

patients (6 vascular recurrent patients) and controls (2 non-vascular recurrence patients) p-

value=0.076.

The Laboratory of Stroke Pharmacogenomics and Genetics is part of the International

Stroke Genetics Consortium (ISGC, www.strokegenetics.com) and coordinates the Spanish

Stroke Genetics Consortium (Genestroke, www.genestroke.com). I. F-C. is supported by

the Miguel Servet programme (CP12/03298), Instituto de Salud Carlos III. We thank Dr.

Raid S Al Baradie for providing the necessary support.

References

1. Adams HP, Bendixen BH, Kappelle LJ, Biller J, Love BB, Gordon DL, et al. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Stroke. 1993;24:35–41.

2. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med. Care. 1986;24:67–74.

3. Touleimat N, Tost J. Complete pipeline for Infinium(®) Human Methylation 450K BeadChip data processing using subset quantile normalization for accurate DNA methylation estimation. Epigenomics. 2012;4:325–41.

4. Pidsley R, Y Wong CC, Volta M, Lunnon K, Mill J, Schalkwyk LC. A data-driven approach to preprocessing Illumina 450K methylation array data. BMC Genomics. 2013;14:293.

5. Price ME, Cotton AM, Lam LL, Farré P, Emberly E, Brown CJ, et al. Additional annotation enhances potential for biologically-relevant analysis of the Illumina Infinium HumanMethylation450 BeadChip array. Epigenetics Chromatin. 2013;6:4.

6. Rakyan VK, Down TA, Balding DJ, Beck S. Epigenome-wide association studies for common human diseases. Nat. Rev. Genet. 2011;12:529–41.

Tables. Table I: Descriptive characteristics of the study population. Discovery cohort n=38,

validaion cohort n=289.

Vascular Recurrence

Non-Vascular

Recurrence

Vascular Recurrence

Non-Vascular Recurrence

Discovery Stage Cohort (n= 38)

Second Stage Cohort (n= 289)

N 19 (50%) 19 (50%) 29 (10%) 260 (90%)

Age (years) 69.2±12.9 68.9±12.8 69±10 65±12

Male 17 (89.4%) 17 (89.4%) 17 (58.5%) 183 (69.5%)

Female 2 (10.5%) 2 (10.5%) 12 (41.3%) 77 (29.3%)

TOAST

Aterothrombotic 7 (21%) 7 (21%) 1 (3.4%) 12 (4.6%)

Lacunar 3 (15.8%) 3 (15.8%) 1 (3.4%) 10 (3.8%)

Cardioembolic - - 6 (20.6%) 20 (7.6%)

Undetermined 9 (47.3%) 9 (47.3%) 10 (34.4%) 38 (14.4%)

Other 1 (3.4%) 9 (3.4%) Dyslipidemia 11 (57.9%) 6 (31.6%) 6 (20.6%) 90 (34.2%)

Diabetes Mellitus 8 (42.1%) 3 (15.8%) 12 (41.3%) 65 (24.7%)

Hypertension 11 (57.9%) 11 (57.9%) 21 (72.2%) 154 (58.5%)

Current Smoker 4 (21%) 5 (31.6%) 8 (27.5%) 67 (25.5%)

Alcohol intake 7 (21%) 4 (21%)

Table II: Descriptive characteristics of each international cohort of the second stage

analysis.

Geneva's Cohort (178)

Polish's Cohort (59)

Italy's Cohort (26)

Spanish's Cohort

(27)

Rec Non-Rec Rec Non-

Rec Rec Non-Rec Rec Non-

Rec

N 10 (5,6%)

168 (94,4%)

13 (21,9%)

46 (77,7%)

2 (7,7%)

24 (92,4%)

4 (14,8%)

23 (85,1%)

Age (years) 69±9 64±11 70±10 66±12 76±10 66±13 66±11 60±12

Male 8 (80%)

127 (74.9%)

7 (53.8%)

24 (52.2%)

1 (50%)

16 (66.6%)

1 (25%)

17 (73.8%)

Female 2 (20%)

41 (24.2%)

6 (46.1%)

22 (47.7%)

1 (50%)

8 (33.3%)

3 (75%)

6 (26%)

TOAST Aterothrombotic -

32 (19%) -

2 (4.3%)

1 (50%)

7 (29.1%) -

3 (13%)

Lacunar - - 1 (7.7%)

5 (10,9%) - 3

(12.5%) - 2 (8.7%)

Cardioembolic - - 5

(38.5%) 19

(41,23%) - - 1

(25%) 1

(4.3%)

Undetermined - - 7 (53.8%)

19 (41,2%) - 2

(8.3%) 3

(75%) 17

(73.8%)

Other - - - 1 (2.27%)

1 (50%)

8 (33.3%)

Dyslipidemia

5 (50%)

88 (51.9%) - - - -

1 (25%)

2 (8.7%)

Diabetes Mellitus

1 (10%)

39 (23%)

8 (61.5%)

20 (43.4%)

2 (100%)

3 (12.5%)

1 (25%)

3 (13%)

Hypertension

7 (70%)

101 (59.66%)

9 (69.2%)

35 (75.9%)

2 (100%)

11 (45.8%)

3 (75%)

7 (30.4%)

Current Smoker 4 (40%)

42 (24.8%)

3 (23.1%)

12 (26%) - 8

(33.3%) 1

(25%) 5

(21.7%)

Table III: Bioconductor packages for the processing and analysis of array-based DNA

methylation data.

DNA methylation processing/analysis step Bioconductor packages Commands

Methylation data loading MethyLumi methylumiR()

Quality control sample/probe wateRmelon, minfi pfilter()

mdsPlot()

Normalization and background correction wateRmelon dasen()

Table IV: DMC sites associated with vascular recurrence (p-value<10-05). List of the 36

candidate DMC sites which were best ranked in both, raw and covariate analysis. Six

CpGs, highlighted in bold letters, were initially selected for further analysis, even though

finally only RAF1, PPM1A and KCQ1 CpG sites were suitable for MassARRAY EpiTYPER

analysis.

DMCs associated with vascular recurrence. Univariate analysis.

CpG ID Chr Position Gene p-value Covar. p-value

cg26039762 3 13177788 RAF1 9,69E-06 0,01815912 cg00094487 15 65255928 SPG21 1,20E-05 0,00493098

cg26568880 6 31926569 RDBP; SKIV2L 1,47E-05 0,00253664

cg20702204 12 133430077 CHFR 2,19E-05 0,00274526

cg22352818 2 97193289

2,19E-05 0,00345245

cg01112035 2 105200688

2,19E-05 0,08986818

cg16966962 1 201915989 LMOD1 3,23E-05 0,00815474

cg09747456 10 91405300 PANK1 3,90E-05 0,0031413

cg07580707 3 14220061 XPC; LSM3 4,69E-05 0,00633959

cg02533998 12 101540053

5,62E-05 0,29100396

cg00932677 4 187776068

6,72E-05 0,00305652

cg08859247 17 17286854

6,72E-05 0,00848049

cg00608860 16 86588622 MTHFSD; FLJ30679 6,72E-05 0,08698691

cg16933664 11 67085326 LOC100130987 8,02E-05 0,00328628

cg08808677 2 6911141

8,02E-05 0,00630309

cg26577201 2 219857793 CRYBA2 8,02E-05 0,00745067

36 candidate DMCs best ranked in both, univariate and covariate analysis

CpG ID Chr Position Gene p-value Covar. p-value

cg00094487 15 65255928 SPG21 1,20E-05 0,00493098

cg26568880 6 31926569 RDBP 1,47E-05 0,00253664

cg20702204 12 133430077 CHFR 2,19E-05 0,00274526

cg22352818 2 97193289

2,19E-05 0,00345245

cg09747456 10 91405300 PANK1 3,90E-05 0,0031413

cg07580707 3 14220061 XPC;LSM3 4,69E-05 0,00633959

cg00932677 4 187776068

6,72E-05 0,00305652

cg08808677 2 6911141

8,02E-05 0,00630309

cg16933664 11 67085326 LOC100130987 8,02E-05 0,00328628

cg26577201 2 219857793 CRYBA2 8,02E-05 0,00745067

cg05529249 2 24272480 FKBP1B 0,00011304 0,00542232

cg08121755 12 52545978

0,00015767 0,00743803

cg25379116 1 160254873 PEX19 0,00015767 0,00417774

cg16413445 16 75659884

0,00018548 0,00501087

cg23130010 6 29855462 HLA-H 0,00018548 0,0055877

cg02076826 15 61477299 RORA 0,00021767 0,00547324

cg08015776 21 27543229 APP 0,00021767 0,00358557 cg05500125 15 101390023

0,0002548 0,00476859

cg03968943 16 1121050 LOC146336 0,00029758 0,00522032

cg06651180 7 101559888 CUX1 0,00029758 0,00689386

cg16618752 1 230080330

0,00029758 0,00580725

cg17965019 6 27858545 HIST1H3J 0,00029758 0,00517746

cg23289021 13 69459913

0,00029758 0,00688949

cg00814733 12 4382051 CCND2 0,00034671 0,00564052

cg04985020 14 60711502 PPM1A 0,00034671 0,00574663 cg08419850 11 2569801 KCNQ1 0,00034671 0,00481029 cg24311382 6 132272408 CTGF 0,00034671 0,00382345 cg00822187 8 26291560

0,00040304 0,00666538

cg02587673 7 592662 PRKAR1B 0,00040304 0,00679647

cg07979106 13 98086433 RAP2A 0,00040304 0,00747707

cg25066224 17 19617395 SLC47A2 0,00040304 0,00557317

cg04646186 15 27215757 GABRG3 0,00046747 0,0058128

cg06006403 20 305636 SOX12 0,00046747 0,00453558

cg09044174 20 60308212 CDH4 0,00046747 0,00486864

cg10799705 19 4358875 MPND 0,00046747 0,00737485

cg21546286 11 48923668 0,00046747 0,00702784

Table V: Sequences of primers used in the MassARRAY EpiTYPER second stage

analysis.

Gene Primer* Size Sequence Product Size

Nº of CpG's Coverage

PPM1A LP 25 GTAGGTTAGGGTGTAGGGGTATGAT

484 8 8 RP 25 CTTACAACCCAAAAACAAATTCAAC

KCNQ1 LP 25 TGTTGTTTTTTTTGGATTTTGTTTT

499 17 12 RP 25 CATTTATACACAACCTAAACACCCC

RAF1 LP 25 TATGTTTTGGTTATGGGAGGTTTTA

335 10 10 RP 25 ACCCCAATTTAAAAAAATAATTCCC

* LP, Left Primer; RP, Right Primer.

Table VI: Differentially methylated CpG sites identified de novo in the EpiTYPER second

stage analysis.

CpG ID Chr Gene p-value

PPM1A_4 14 PPM1A 0.02733

RAF1_3 3 RAF1 0.01049

RAF1_6 3 RAF1 0.00432

RAF1_10 3 RAF1 0.00153

Figures. Figure I: Quality control analysis. (a) Density plot showing the methylation levels

distribution in each sample. The same bimodal distribution was observed in all samples

except for one sample (removed from further analysis). X-axis indicates methylation β-

values and Y-axis frequency, each line represents one sample. (b) Multidimensional

Scaling plot showing two clusters, male samples and female samples. Two samples

(indicated by arrow) were group in the wrong cluster (removed from further analysis). X-axis

indicates the first dimension of variance and Y-axis indicates the second dimension of

variance.

Figure II: Clustering heatmap of the 16 diferentially methylated CpG sites identified by 450k

analysis. Each column represents a sample and each horizontal line represents the

methylation levels of a given CpG across samples. Methylation levels are expressed as 0-1

β-values (green and red, unmethylated and completely methylated, respectively).

Figure III: PPM1A methylation levels on stroke and cardiovascular patients, only stroke

patients, and no-cardioembolic stroke patients from validation cohorts. All three analysis

show higher methylation levels of vascular recurrent patients compared with non-vascular

recurrent patients, regardless of the pathology (stoke or cardiovascular disease).

Figure IV: Diagram of candidate CpG sites selected for second stage study.

Top 100

Differentially Methylated CpG sites

Top 100 Differentially Methylated CpG sites

including covariates

EWAS 475.500 CpG sites

analyzed

36 coincident CpG sites

2 CpG sites suitable for

EpiTYPER analysis

4 CpG sites previously associated with

atherosclerotic process

name : abdul aziz a. bin dukhyil department : …...keywords: forensic science, dna reference...

Documents