GENETIC TESTINGVolume 11, Number 4, 2007© Mary Ann Liebert, Inc.DOI: 10.1089/gte.2007.0044

Buccal Cell DNA Extraction: Yield, Purity, and Cost: A Comparison of Two Methods

YOLANDE B. SAAB,1 WISSAM KABBARA,1 CHRISTIANE CHBIB,1 and PAUL RICHARD GARD2

ABSTRACT

Simple and cost-effective methods are needed to extract DNA in order to use it in large-scale studies. Bloodis an excellent DNA source; however, it is costly and invasive thus an alternative is needed. Several kits andchemical protocols using buccal cells have been proposed for DNA extraction. The objective of the study is toevaluate buccal NaOH chemical protocol and Nucleospin® Tissue Kit (BD Biosciences, Macery-Nagel, Ger-many) for DNA extraction. DNA swab samples were collected from 300 voluntary participants. DNA yieldsand purity were measured by NaOH and Nucleospin® Tissue Kit techniques; the cost and time consumptionfor DNA extraction per sample were assessed as well. Results have shown that DNA amount and purity ex-tracted by NaOH procedure was compared to that of the kit (p � 0.164; p � 0.249, respectively). NaOH methodwas considered cheaper and less time consuming (0.06 versus 3.80 USD, and 1.33 versus 3.59 minutes persample, p � 0.001). Buccal cell derived DNA extracted by NaOH protocol can be considered a feasible sub-stitute for more expensive and time-consuming kits.

413

INTRODUCTION

GENETIC RESEARCH has recently seen a marked increase asinterest in understanding the genetic basis of diseases and

drug regimens increases. Almost all of these studies requireDNA isolations. Blood samples, in particular white blood cells,are an excellent source to obtain large amounts of genomicDNA. However, because of the invasiveness and cost of ob-taining and testing blood samples, alternative methods areneeded particularly in epidemiologic studies (Feigelson et al.,2001; Garcia-Closas et al., 2001; Harty et al., 2000a, 2000b;Heath et al., 2001; Le Marchand et al., 1998). Buccal cells pro-vide a feasible and noninvasive approach to supply DNA forgenetic testing (Walker et al., 1999; Freeman et al., 1997; Lenchet al., 1988; Hayney et al., 1995; Lum and Le Marchand, 1998).There are two types of methods for collecting buccal cells: dryand wet procedures. Dry procedures use cytobrushes, buccalswabs, or other methods to scrape the oral mucosa. Wet pro-cedures include swishing liquids in the mouth and spitting theminto a collecting vessel (Feigelson et al., 2000). The swishmethod yields a higher amount and longer fragments of DNA(Feigelson et al., 2000; Heath et al., 2001), however, it requires

more steps, is more cumbersome, and is associated with an in-creased cost. Dry methods are simpler and more feasible forlarge-scale mailings; additionally, they have a lighter weightand are more cost effective. Recent studies have shown simi-lar successful DNA amplification and polymerase chain reac-tion (PCR) applications from both procedures (Walker et al.,1999; Freeman et al., 1997; Richards et al., 1993; Meulenbeltet al., 1995). Protocols to extract DNA by dry procedures in-clude the use of chemicals such as phenol-chloroform, NaOH,or already prepared kits containing all the components and amanual for DNA extraction steps. Studies comparing the yieldand purity of the obtained DNA among different methods ofextraction are needed. Using NaOH is less expensive than mar-keted kits. However, the significance of the difference in DNAyield, purity and quality between NaOH and marketed kitsneeds to be determined. The primary objective of this study wasto compare buccal cell DNA yield and purity extracted by twodifferent techniques, namely, NaOH protocol or NucleospinTissue® prepared kit (BD Biosciences, Macery-Nagel, Ger-many) using a dry DNA collection method (cotton swabs).Quantity, feasibility, and time consumption were assessed, aswell.

1Pharmacy School, Lebanese American University, Byblos, Lebanon.2Brighton University, Brighton, UK.

MATERIALS AND METHODS

Study population

For this study, 300 healthy volunteers were recruited fromthe general community (age range, 18–65 years) from Febru-ary until June 2006. Samples were divided into two groups: 150for DNA extraction via the chemical method group (NaOHgroup), and the other 150 samples were processed using the Nu-cleospin Tissue® kit.

Sample collection

Instructions for the collection of DNA using cotton swabs(Nuova Aptaca™, Canelli, Italy) were given for each subjectenrolled in the study by the experimenters. The subjects ab-stained from smoking, drinking, and/or eating for 45 min be-fore sample collection. The subjects were asked to twirl a ster-ile cotton swab on each inner cheek for 15 sec. The swabs werereturned to the laboratory. DNA was extracted from the cottonswabs within 24 hr. All samples were processed at room tem-perature (24°C) in accordance with good laboratory practices.

DNA extraction

NaOH protocol. DNA was extracted according to a protocolusing NaOH in 150 samples. The swabs were separated from thesticks with scissors and they were transferred to 1.5 ml Epen-drofTM tubes (Ismaning, Germany). Three hundred microliters ofNaOH 50 mM were added to each tube. The tubes were closedand vortexed for 10 sec. They were placed in a thermomixer for5 min at 95°C. Then, the swabs were removed and discarded.Thirty microliters of 1 M Tris HCl, pH � 8.0 was added to eachtube. The tubes were centrifuged at 13,000 rpm for 2 min. Thesupernatant (DNA) in each tube was used for analysis.

Nucleospin protocol. Nucleospin Tissue® prepared kit wasused to extract genomic DNA from the other 150 samples fol-lowing the Nucleospin Tissue® kit manual instructions (Nucle-ospin® Tissue Kit manual, 2004). In brief, the dry swab was placed

in 2-ml microcentrifuge tubes. Phosphate buffered saline (PBS),400 ml, was added with 25 �l proteinase K solution to the swabs.This step aimed at prelysing the sample. The lysis step involvedthe addition of 400 �l of a prepared buffer to the previous solu-tion. Vigourous vortexing was needed and the samples were in-cubated at 70°C for 10 min. Four hundred microliters of 98% eth-anol was added to each sample and revortexed, thus adjustingDNA binding conditions. To bind DNA, 600 �l of the sampleswere centrifuged at 11,000g for 1 min. The next step was to washthe silica membrane with a wash buffer and centrifuge for 1 minat 11,000g. A second wash with another buffer was done and cen-trifuged again. To dry the silica membrane, the samples were cen-trifuged one more time (residual ethanol was removed during thisstep). To elute highly pure DNA, 100 �l of a prewarmed elutionbuffer (70°C) was used and the samples were incubated at roomtemperature for 1 min. Finally, the samples were centrifuged for1 min at 11,000g once more. Obtained DNA was collected.

Yield and purity determination

Quartz cuvettes were used to read the absorbance at a wave-length of 260 nm and 280 nm using spectrophotometry. DNAconcentration was calculated by multiplying the absorbance at260 nm by 50 (1 unit is �g/ml). A 260:A280 ratio was calcu-lated to check for DNA purity.

To validate DNA quality and quantity, angiotensin-convert-ing enzyme (ACE) insertion (I)/deletion (D) gene polymor-phisms were amplified from the extracted DNA by either NaOHor Nucleospin Tissue Kit® using the PCR as described by Saabet al. (2007). The amplicons were separated and sized by elec-trophoresis on a 2.5% agarose gel and visualized directly withethidium bromide staining. The quality of the DNA from sam-ples was assessed with the insertion allele manifested as a 490bp and the deletion allele visualized as a 190 bp band. Repre-sentative results are shown in Figure 1.

Statistical analysis

Data were tabulated using Microsoft EXCEL (Microsoft,Inc., Redmond, WA), and analyzed using SigmaStat version 3.

SAAB ET AL.414

FIG. 1. Illustration of polymerase chain reaction (PCR) products from amplification of the Insertion (I)/ Deletion (D) poly-morphisms of the angiotensin-converting enzyme (ACE) gene. Lane 1: 20-bp DNA ladder; lanes 2, 4, 6, 7, 8 represent the ho-mozygous DD genotype (190 bp); lanes 3, 9 represent the heterozygous ID genotype (490 bp and 190 bp); II homozygous geno-type is illustrated in lane 5 (490 bp).

Differences in the mean of DNA yield, purity, cost, and timeconsumption between the two methods of extraction were testedusing the t test. The p values were recorded accordingly withp � 0.05 considered significant.

RESULTS

Total DNA yield

Means of total DNA yield measured at 260 nm for NaOHvs Nucleospin Tissue® kit are presented in Table 1. The aver-age amount of DNA ranged from 10.61 �g/ml to 19.98 �g/mlusing the NaOH method, and 12.12 �g/ml to 20.65 �g/ml us-ing the Nucleospin Tissue® kit. A greater total DNA yield wasobtained in the kit group; however, the difference was not sig-nificant (p � 0.164).

DNA purity

A260:A280 ratio was used to determine the purity of DNApresent in the samples. Means of DNA purity for NaOH vs Nu-cleospin Tissue® kit are shown in Table 1. The ratio rangedfrom 0.90 to 1.32 in the NaOH group and 1.04 to 1.68 in thekit group. DNA purity was adequate in both groups. The DNAobtained by the Nucleospin Tissue® kit had a nonsignificanthigher A260:A280 ratio (p � 0.249).

In more than 90% of the DNA samples extracted by NaOHwe were able to amplify ACE I/D gene fragments (137/150).Slightly more samples amplified in the DNA extracted by Nu-cleospin Tissue® kit (96%; 144/150).

Cost analysis

The cost of extracting DNA per sample from the cottonswabs using NaOH and Nucleospin Tissue® kit was calculated.The cost was based on actual charges for equipment and sup-plies used in the study for both groups; i.e., the price of the kitsordered from the supplier and the materials and chemicals usedin the NaOH method. The cotton-swabs cost was not includedin the study for the same number and type of swabs were usedin both groups. Cost of the personnel time for processing thesamples was not considered, as well. The average cost per sam-ple in USD was 0.06 and 3.08 for NaOH and Nucleospin Tis-sue® kit, respectively (p � 0.001).

Time consumption

The average time consumption per sample for NaOH vs Nu-cleospin Tissue® kit is presented in Table 1. The elapsed timewas recorded for 12 samples processed together of each group,then calculated per sample. The average time for the process-ing of 12 samples was 16 min for the NaOH group and 43 minfor the Nucleospin Tissue® group. Time consumption usingNaOH protocol is significantly less than the time needed to ex-tract DNA by the Nucleospin® Tissue kit (p � 0.001).

DISCUSSION

The role of pharmacogenetics is going to have an increasedemphasis especially in the development and prevention of can-cer and other diseases. Thus, cost-effective methods of DNA

collection and extraction need to be determined. The aim of ourstudy was to evaluate NaOH versus Nucleospin Tissue® kit forDNA extraction collected from buccal cells, and then to deter-mine a simple method for DNA extraction. The two methodswere compared and both the quantity and purity of DNA andthe feasibility and cost of the procedures were assessed.

Many studies have shown that the collection of DNA withcytobrushes using simple instructions is cost effective in large-scale studies and yields sufficient quantity and good quality ofDNA for genotyping (Garcia-Closas et al., 2001; Walker et al.,1999; King et al., 2002). A total of 15.8 �g/ml of DNA fromcytobrushes and 12 �g/ml from mouthwash were obtained (Kinget al. 2002). The collection of human genomic DNA from buc-cal cells for genetic studies using cytobrush, mouthwash, andtreated cards was compared (Mulot et al. 2005). The mean DNAyield was found to be 3.5, 4, and 2.6 �g/ml for cytobrushes,mouthwashes, and treated cards, respectively. The results con-firmed that the protocol based on cytobrushes should be prefer-entially used because of its ease of collection, good DNA qual-ity and sufficient DNA quantity. Another study determined thelong-term stability, quantity, and quality of genomic DNA col-lected by the mouthwash method in order to use it in pharma-cogenomic studies (Andrisin et al., 2002). A median of 12.6�g/ml and 13.6 �g/ml of DNA extracted by NaOH and phenol-chloroform, respectively, using cytobrushes was obtained in anepidemiologic study (Garcia-Closas et al., 2001). DNA yield ob-tained in our study (14.1 �g/ml for NaOH versus 16.6 �g/mlfor the kit) was higher than previously reported studies usingbuccal cells for DNA extraction. DNA extracted in this studyfrom buccal cells using cotton swabs provides adequate amountof DNA for further usage. Both methods (NaOH and NucleospinTissue® Kit) used supplied sufficient DNA yields.

A mean A260:A280 ratio was 1.8 and 1.9 for the brush andmouthwash, respectively, in a large-scale study (King et al.,2002). PCR success rate was 100% for all samples (de Vries etal., 1996; Walker et al., 1999; Zheng et al., 2001). DNA pu-rity from another study resulted in ratios of 1.6, 1.7, and 1.1 forcytobrush, mouthwash, and treated cards, respectively (Mulotet al., 2005). DNA ratios of 1.9 and 1.3 for phenol-chloroformand NaOH, respectively, for cytobrushes were obtained in anepidemiologic study (Garcia-Closas et al., 2001). Samples col-lected by mouthwash had ratios of 1.9 for both phenol-chloro-form and PureGene Kit (Gentra Systems, Minneapolis, MN).In this study, a mean ratio of 1.08 and 1.245 were obtained us-ing NaOH and Nucleospin Kit®, respectively. Comparing DNA

BUCCAL CELL DNA EXTRACTION 415

TABLE 1. NaOH VS. NUCLEOSPIN TISSUE® KIT DNA: YIELD (DNA CONCENTRATION), PURITY

(A260�A280 RATIO), AND TIME CONSUMPTION

Extractionmethod n Mean SD Mean SD Mean SD

NaOH 150 14.18 1.68 1.08 0.07 1.33 0.51Kit 150 16.67 1.69 1.24 0.10 3.59 0.81p 0.164 0.249 � 0.001

n, sample size; SD, standard deviation.

Timeconsumptionper sample(minutes)

A260�A280ratio

DNA conc(�g/ml)

purity with the NaOH method and the Nucleospin Tissue® Kitrevealed a slight insignificant difference. Although the samplesare not 100% pure (ratio � 1.8), this will not limit the DNAusage for further applications; PCR amplification was success-ful for assessed samples by either method.

In terms of cost, the relevant studies compared differentmethods of DNA extraction. In one study, the estimated costper person for buccal brush DNA was $8.5 and $18 for themouthwash (King et al., 2002). All of the samples were ex-tracted using QIAamp Mini Kits (Quiagen, Hilden, Germany).The final cost in another study was per sample: $4.66, $7.17,and $2.81 for cytobrush, mouthwash, and treated cards, re-spectively (Mulot et al., 2005). The cost per sample of ex-tracting DNA by using NaOH ($0.06 per sample) is more than50 times less expensive than the Nucleospin Tissue® Kit ($3.80per sample) and much less expensive than in reported studies.Given the comparable yield and purity between the two meth-ods, the NaOH protocol for DNA extraction provides a cost-ef-fective approach.

As for collection time, 1.7 min was needed for DNA ex-traction per sample step for either cytobrush or mouthwash ina human genomic DNA collection study (Mulot et al., 2005).Personnel time for DNA extraction using the Nucleospin Tis-sue Kit (3.59 min per sample) is more than double the timeneeded for extraction following the NaOH protocol (1.33 minper sample). This also favors NaOH as being a time saving al-ternative to Nucleospin Tissue® Kit.

CONCLUSION

In summary, DNA yield and purity are comparable whetherDNA was extracted by NaOH protocol or Nucleospin Tissue®

Kit. NaOH protocol is less expensive and less time consuming,therefore, it provides a more feasible and cost-effective alter-native for DNA extraction. We recommend using NaOH pro-tocol for DNA extraction in large-scale studies. Further studiesare needed to compare DNA extracted by different chemicalmethods versus other marketed kits.

ACKNOWLEDGMENT

We would like to thank LAU School of Pharmacy for the fi-nancial help.

REFERENCES

Andrisin TE, Humma LM, Johnson JA (2002) Collection of genomicDNA by the noninvasive mouthwash method for use in pharmaco-genetic studies. Pharmacotherapy 22:954–960.

de Vries HG, Collee JM, van Veldhuizen MH, Achterhof L, SmitSibinga CT, Scheffer H, Buys CH, ten Kate LP (1996) Validation ofthe determination of F508 mutations of the cystic fibrosis gene inover 11 000 mouthwashes. Hum Genet 97:334–336.

Feigelson HS, Rodriguez C, Robertson AS, Jacobs EJ, Calle EE, ReidYA, Thun MJ (2001) Determinants of DNA yield and quality frombuccal cell samples collected with mouthwash. Cancer EpidemiolBiomarkers Prev 10:1005–1008.

Freeman B, Powell J, Ball D, Hill L, Craig I, Plomin R (1997) DNA by

mail: An inexpensive and noninvasive method for collecting DNAsamples from widely dispersed populations. Behav Genet 27:251–257.

Garcia-Closas M, Egan KM, Abruzzo J, Newcomb PA, Titus-ErnstoffL, Franklin T, Bender PK, Beck JC, Le Marchand L, Lum A, Ala-vanja M, Hayes RB, Rutter J, Buetow K, Brinton LA, Rothman N(2001) Collection of genomic DNA from adults in epidemiologicalstudies by buccal cytobrush and mouthwash. Cancer Epidemiol Bio-markers Prev 10:687–696.

Harty LC, Garcia-Closas M, Rothman N, Reid YA, Tucker MA, andHartge P (2000a) Collection of buccal cell DNA using treated cards.Cancer Epidemiol Biomarkers Prev 9:501–506.

Harty LC, Shields PG, Winn DM, Caporaso NE, Hayes RB (2000b)Self-collection of oral epithelial cell DNA under instruction from epi-demiological interviewers. Am J Epidemiol 151:199–205.

Hayney MS, Dimanlig P, Lipsky JJ, Poland GA (1995) Utility of a“swish and spit” technique for the collection of buccal cells for TAPhalotype determination. Mayo Clinic Proc 70:951–954.

Heath EM, Morken NW, Campbell KA, Tkach D, Boyd EA, Strom DA(2001) Use of buccal cells collected in mouthwash as a source ofDNA for clinical testing. Arch Pathol Lab Med 125:127–133.

King IB, Satia-Abouta J, Thornquist MD, Bigler J, Patterson RE, KristalAR, Shattuck AL, Potter JD, White E (2002) Buccal cell DNA yield,quality, and collection costs: Comparison of methods for large-scalestudies. Cancer Epidemiol Biomarkers Prev 11:1130–1133.

Le Marchand L, A Lum-Jones, B Saltzman, V Visaya, AM Nomura, andLN Kolonel. 2001. Feasibility of collecting buccal cell DNA by mailin a cohort study. Cancer Epidemiol Biomarkers Prev 10:701–703.

Lench N, Starnier P, Williamson R (1988) Simple non invasive methodto obtain DNA for gene analysis. Lancet 2:1356–1358.

Lum A, Le Marchand L (1998) A simple mouthwash method for ob-taining genomic DNA in molecular epidemiological studies. CancerEpidemiol Biomark Prev 7:719–724.

Meulenbelt I, Droog S, Trommelen GJ, Boomsma DI, Slagboom PE(1995) High yield noninvasive human genomic DNA isolationmethod for genetic studies in geographically dispersed families andpopulations. Am J Hum Genet 57:1252–1254.

Mulot C, Stucker I, Clavel J, Beaune P, Loriot MA (2005) Collectionof human genomic DNA from buccal cells for genetics studies: Com-parison between cytobrush, mouthwash, and treated card. J BiomedBiotech 3:291–296.

Nucleospin® Tissue kit [manual]. Heidelberg, Germany: BD Bio-sciences, 2004.

Richards B, Skoletsky J, Shuber AP, Balfour R, Stern RC, Dorkin HL,Parad RB, Witt D, Klinger KW (1993) Multiplex PCR amplificationfrom CFTR gene using DNA prepared from buccal brushes/swabs.Hum Mol Genet 2:159–163.

Saab YB, Gard PR, Yeoman MS, Mfarrej BG, El-Moalem HE, IngramMJ (2007) Renin-angiotensin- system gene polymorphisms and de-pression. Prog Neuropsychopharmacol Biol Psychiatry 31:1113–1118.

Walker AH, Najarian D, White DL, Jaffe JF, Kanetsky PA, RebbeckTR (1999) Collection of genomic DNA by buccal swabs for poly-merase chain reaction based biomarker assays. Environ Health Per-spect 107:517–520.

Zheng S, Ma X, Buffler PA, Smith MT, Wiencke JK (2001) Whole ge-nome amplification increases the efficiency and validity of buccalcell genotyping in pediatric populations. Cancer Epidemiol Bio-markers Prev 10:697–700.

Address reprint requests to:Yolande B. Saab, Pharm.D., Ph.D.

School of PharmacyP.O. Box: 36 F 19

Lebanese American UniversityByblos, Lebanon

E-mail: ysaab@lau.edu.lb

SAAB ET AL.416