Cryopreserving whole blood for functional assays using viable lymphocytes in molecular epidemiology studies

Cancer Letters 166 (2001) 155±163

www.elsevier.com/locate/canlet

Cryopreserving whole blood for functional assays using viable lymphocytes in molecular epidemiology studies L. Cheng, L.E. Wang, M.R. Spitz, Q. Wei* Department of Epidemiology, The University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA Received 2 November 2000; received in revised form 4 January 2001; accepted 5 January 2001

Abstract There is an increasing need for viable lymphocytes in performing phenotypic assays for biomarker studies. Both fresh and cryopreserved lymphocytes have been used for cell culture-based functional assays. However, fresh lymphocytes do not allow assays to be done in batches and cryopreservation of isolated lymphocytes results in a considerable loss of viable cells. To investigate the feasibility of using cryopreserved whole blood as a source of viable lymphocytes in molecular epidemiology studies, two well-established biomarkers, the host-cell reactivation (HCR) and mutagen sensitivity assays, were used to compare the method of cryopreserving whole blood with the traditional methods. In 25 paired blood samples assayed for DNA repair capacity (DRC) by the HCR assay, the DRC values of frozen whole blood (mean ^ SD, 11.59 ^ 3.07) were similar to those of frozen isolated lymphocytes (11.08 ^ 3.50). The correlation between the paired DRC values was 0.77 (P , 0:001). In 31 paired blood samples assayed for the g-radiation-induced chromatid breaks by the mutagen sensitivity assay, there was no signi®cant difference between the baseline level of chromatid breaks in lymphocytes from frozen blood (0.05 ^ 0.03) and fresh blood (0.06 ^ 0.03). The blastogenic rate and mitotic index of the cells used for the two assays were compared between the different processing methods. The lymphocytes from frozen whole blood were more sensitive to g-radiation, with a higher mean level of chromatid breaks (0.68 ^ 0.21) than that in fresh blood (0.42 ^ 0.12, P , 0:01), and the correlation between the numbers of chromatid breaks in the paired samples was statistically signi®cant (r ˆ 0:61, P , 0:001). These data suggest that within the limits of the parameters investigated here, cryopreserved whole blood is a good source of viable lymphocytes for biomarker assays in molecular epidemiological studies. q 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Biomarkers; DNA repair; Genetic susceptibility; Molecular epidemiology; Mutagen sensitivity

1. Introduction The role of genetic susceptibility in cancer etiology is under intensive investigation in molecular epidemiology studies. The interaction between environmental exposures and host susceptibility factors has been elaborated through the measurement of biologic markers in human cells, tissues, and body ¯uids [1±3]. * Corresponding author. Tel.: 11-713-792-3020; fax: 11-713792-0807. E-mail address: [email protected] (Q. Wei).

These studies improve cancer risk assessment and further the understanding of mechanisms of human carcinogenesis. However, a serious limitation in biomarker assays is the method of processing and storing samples, i.e. the need for large quantities of fresh biologic specimens and isolation of lymphocytes, which is labor-intensive and costly [4±6]. Furthermore, the large volume of blood required not only discourages collaborations in multidisciplinary research, but also makes it more likely that subjects will refuse to participate.

0304-3835/01/$ - see front matter q 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0304-383 5(01)00400-1

156

L. Cheng et al. / Cancer Letters 166 (2001) 155±163

In addition, there is an increasing need for using viable lymphocytes in performing phenotypic assays for biomarker studies. Traditionally, viable lymphocytes from fresh blood have been required for cytogenetic assays, and cryopreserved lymphocytes have been required for functional assays requiring shortterm cell culture. Both methods have limitations in that fresh blood does not allow assays to be done in batches due to a time lag between sample procurement, and cryopreservation of isolated lymphocytes results in a considerable loss of viable cells. To explore the feasibility of using cryopreserved whole blood as a source of viable lymphocytes, to reduce the amount of blood needed for biomarker assays, and to increase the ef®ciency of laboratory work in molecular epidemiological studies, we investigated the use of a wholeblood cryopreservation method for storing viable lymphocytes. It is a simple, quick, and reliable method that requires less blood volume and allows samples to be assayed in batches (i.e. mixed cases and controls). We tested its feasibility in comparison with the traditional blood-sample processing methods used for two well-established biomarker assays, the mutagen sensitivity assay [4] and the host-cell reactivation (HCR) assay for measuring DNA repair capacity (DRC) [5].

2. Materials and methods 2.1. Collection of blood samples Study subjects were recruited between May 1998 and January 1999 from an ongoing lung cancer casecontrol study in the Department of Epidemiology at the University of Texas M.D. Anderson Cancer Center. Approximately 20 ml of whole blood was obtained from each subject, drawn into 10-ml heparinized Vacutainers (Becton Dickinson, Franklin Lakes, NJ), and processed within 24 h. Whole-blood samples obtained from 31 subjects were used for the mutagen sensitivity assay; from each sample, 2 ml of fresh whole blood was used for two fresh cultures and 2 ml was used for whole-blood freezing and stored at 2808C for an average of 10 days (range 3 days to 2 months). Samples obtained from another 25 subjects were used for the HCR assay: 10 ml of whole blood was used for isolating and freezing lymphocytes and 5 ml was used for whole-blood freezing and stored at

2808C for an average of 30 days (range 1 week to 3 months). 2.2. Lymphocyte freezing procedure Lymphocytes were isolated from whole blood using a Ficoll-Hypaque gradient method [7]. Puri®ed lymphocytes were resuspended at 10 £ 10 6 cells/ml in ice-cold freezing medium consisting of 10% dimethyl sulfoxide (DMSO; Fisher Scienti®c Co., Pittsburgh, PA), 40% RPMI 1640, and 50% fetal bovine serum (FBS). The cell suspension was transferred to cryovials (Nalge Company, Rochester, NY) that were then submerged into a pre-cooled (48C) isoproponol container. The container was immediately placed in a 2808C freezer for storage, which provided a freezing speed of approximately 18C/min. 2.3. Whole-blood freezing procedure When the blood samples in 10 ml green-top Vacutainers (Becton Dickinson) were received, ice-cold (48C) DMSO was added to each tube (10%, v/v) and the samples were mixed gently. The mixture was pipetted into ice-cold 2 or 5 ml cryovials that had been labeled with the sample identi®cation number and date and the cryovials were then submerged in pre-cooled (48C) isoproponol in a container that was immediately put in a 2808C freezer. After the sample had been equilibrated to 2808C (approximately overnight), the cryovials were transferred to a 9 £ 9-well box for storage at 2808C. 2.4. Sample-thawing procedures and cell culture The paired samples in the cryovials containing frozen puri®ed lymphocytes and whole blood were taken from the freezer and quickly thawed by submersion in a 378C water bath. When the last ice crystal in the cryovials was still visible, they were placed on ice. The thawed cells in 1 unit of freezing medium volume were washed with 5 units of volume of ice-cold wash medium (50% FBS, 10% glucose, and 40% RPMI 1640) and centrifuged at 1000 rev./min at 48C for 10 min. The supernatants were discarded, and the cell pellets were resuspended and cultured in T-25 ¯asks at a cell density of approximately 10 £ 10 6 cells per 10 ml RPMI 1640 medium containing 20% FBS, and 56 mg/ml of phytohemagglutinin (PHA; Murex Diagnos-

L. Cheng et al. / Cancer Letters 166 (2001) 155±163

tics, Norcross, GA) was used to stimulate T-lymphocytes growth. 2.5. The mutagen sensitivity assay Standard lymphocyte cultures for the mutagen sensitivity assay were established as described previously [8], and g-radiation was used as the test mutagen to induce chromosome breaks [9]. Because the fresh blood was used, the paired samples were assayed separately. For each sample, 1 ml of either fresh or thawed frozen whole blood was inoculated into a T-25 culture ¯ask containing 9 ml of culture medium and incubated at 378C. Two cultures were prepared for each subject. Within 72 h of incubation, vigorous cellular growth occurred. At 67 h, one of the cultures was treated with 1.5 Gy of incident g-radiation from a 137Cs source (Cesium irradiator Mark 1, model 30; J.L. Shepherd and Associates, Glendale, CA). During the last hour of incubation, the cultures were treated with Colcemid (Life Technologies, Inc., Rockville, MD) at a ®nal concentration of 0.06 mg/ml to induce mitotic arrest. The cells were then harvested, spread on slides, and stained with 4% Giemsa without banding. For each culture, 50 wellspread metaphases from each coded sample were carefully examined by one of us (L.E.W.), and the number of chromatid breaks was counted for each metaphase and expressed as the average number of breaks per cell (b/c). Chromatid gaps were not recorded. The mitotic index (MI) was calculated as the percentage of total lymphocytes arrested in metaphase after receiving Colcemid treatment. 2.6. HCR (DRC) assay The HCR assay was performed as described previously [10]. The HCR assay measures the activity of the chloramphenicol acetyltransferase (CAT) gene, a bacterial drug resistance gene, in cells that have been transfected with UV-treated plasmid. Plasmids damaged by an incident dose of 800 J/m 2 UV light (about 254 nm) and undamaged plasmids were transfected into the lymphocytes using the diethylaminoethyl-dextran (Pharmacia Biotech Inc., Piscataway, NJ) method [11]; the transfections were performed in duplicate for each UV dose. Because a single unrepaired DNA adduct can effectively block CAT transcription, any CAT activity will re¯ect the

157

ability of the transfected cells to remove UV damage from the plasmids. Therefore, this assay provides a quantitative measurement of the DRC of the host cells. The CAT gene expression (or activity) was measured 40 h after transfection as described previously [10]. Brie¯y, the tubes containing the cell culture with transfections were then centrifuged; the cell pellets were collected and washed twice with 1.5 ml Tris-buffered saline (TBS) and resuspended in 31.5 ml of 0.25 M Tris TBS in a 1.5 ml Eppendorf tube. The cells were lysed by three 10 min cycles of freezing and thawing in a dry ice/ethanol bath and a 378C water bath. Cell extracts were then assayed for CAT expression or activities. The expression of the repaired CAT gene was measured by a scintillation counter for the formation of [ 3H]monoacetylated and diacetylated chloramphenicols through the reaction between chloramphenicol and [ 3H]acetyl coenzyme A catalyzed by CAT protein in the cell extract. DRC is de®ned as the ratio of the CAT activity of cells transfected with UV-treated plasmids to that (baseline) of cells transfected with untreated plasmids, i.e. DRC ˆ …CATUV800 =CATUV0 † £ 100%. The CAT activity of cells transfected with undamaged plasmids provides an experimental internal control, because it is derived under the same experimental conditions as the CATUV0 [10] and from the same number of cells from the same individual. In this assay, the frozen isolated lymphocytes and whole blood from each paired sample were thawed at the same time in batches. The blastogenic rate (BR) and actual radioactivity counts (baseline expression level) of undamaged plasmids were also recorded for comparison of the differences in the effect of cryopreservation on frozen isolated lymphocytes and frozen whole blood. 2.7. BR measurement The total number of viable lymphocytes was counted using a microscope and 0.4% Trypan Blue exclusion (Sigma, St. Louis, MO) [12] at 0 and 72 h of cell culture. The BR was calculated as the percentage of lymphocytes that responded to mitogen (PHA) stimulation. 2.8. Statistical analysis The ranking of the measurements of the paired

158

L. Cheng et al. / Cancer Letters 166 (2001) 155±163

samples from the two assays was evaluated using the Wilcoxon rank sum test, and differences in the means were compared using the paired t-test. The Pearson correlation coef®cient was computed for the paired samples for both chromatid b/c and DRC values. Analysis of the data was performed with the PC version (6.0) of Statistical Analysis System (SAS) software (SAS Institute Inc., Cary, NC). 3. Results We ®rst compared the BR in response to mitogen stimulation to identify differences in the effect of

cryopreservation on frozen isolated lymphocytes versus frozen whole blood. The BR of the lymphocytes recovered from frozen whole blood was lower than that of either frozen isolated lymphocytes in the CAT assay or fresh culture in the mutagen sensitivity assay (Tables 1 and 2), but these differences were not statistically signi®cant (P ˆ 0:243 and P ˆ 0:510, respectively). These results suggest that the lymphocytes recovered from frozen whole blood may have a similar response to mitogen stimulation. We next examined the differences in the baseline expression level as measured by the actual radioactivity counts from the cell extracts containing CAT enzymes from undamaged plasmids. Baseline CAT

Table 1 Comparison of the DRC of paired samples using two different cryopreservation methods Rank (DRC) a

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

F F F F F F M F F M F F F M M M M M M M F M M F F

Mean ^ SD P value d a b c d

Sex

Age (years)

Frozen cells a

Frozen blood b

BR (%) c

Baseline (cpm)

DRC (%)

BR (%) c

Baseline (cpm)

DRC (%)

66 60 51 54 66 73 78 49 67 60 49 60 55 61 67 47 67 73 56 61 56 76 65 62 61

36.0 49.2 36.0 184.0 50.6 59.0 89.0 50.0 36.9 32.0 74.2 44.2 102.0 37.9 23.0 47.8 29.8 16.0 85.0 30.7 24.0 18.9 22.4 56.0 24.1

20614 33451 39384 16064 11446 15693 14230 9367 12302 20998 13297 41159 16824 14384 7244 16514 47633 9574 28950 37435 8108 14865 10632 11673 43058

6.9 7.0 7.1 7.8 8.0 8.2 8.2 8.9 9.3 9.6 10.1 10.2 10.4 10.9 11.2 11.3 11.5 11.8 12.7 13.2 13.9 14.1 15.8 17.7 21.3

130.0 27.7 46.0 60.0 33.3 53.0 86.0 26.8 22.3 33.7 45.5 51.0 18.5 25.0 13.2 34.6 50.0 12.6 53.0 18.0 14.5 33.9 63.3 22.0 22.6

33071 25460 42611 45188 24028 27788 46247 22562 55797 25030 33557 24499 16308 32992 28407 48941 26405 11083 37408 12843 22647 33806 16931 96015 25768

7.8 7.1 9.9 12.7 8.5 8.3 11.9 10.9 11.1 9.7 9.5 10.0 11.7 11.7 9.2 11.2 10.7 13.4 9.4 12.1 14.0 17.2 16.0 19.8 15.2

61 ^ 8

50.3 ^ 35.9 Ref.

20595 ^ 12389 Ref.

11.1 ^ 3.5 Ref.

39.8 ^ 26.2 0.243

32615 ^ 17362 0.007

11.6 ^ 3.1 0.595

Rank by DRC of frozen isolated lymphocytes from fresh blood. Lymphocytes from frozen whole blood. BR in response to mitogen stimulation. Ref., reference group; P values were derived from a two-sided Student's t-test.

F F F F F M M M F M F F F F F F M F F F F M F M M M M M F M M

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

c

b

a

54 69 69 55 49 60 63 68 58 78 37 62 67 74 56 64 61 49 60 34 46 56 50 66 76 61 74 63 55 47 47

Age (years)

71.6 ^ 42.6 Ref.

44.4 25.8 140.0 128.0 200.0 47.0 54.3 64.0 141.0 74.5 28.8 200.0 100.0 100.0 30.5 41.1 90.0 50.0 150.0 150.0 40.0 62.8 47.1 100.0 58.8 63.8 100.0 54.9 131.0 42.0 42.0 7.5 ^ 1.6 Ref.

7.0 8.4 8.6 8.4 8.4 7.8 5.2 7.0 8.6 7.0 7.8 6.8 8.6 8.2 8.4 5.2 7.6 4.8 7.5 9.0 9.8 5.8 8.4 11.2 6.2 7.8 5.2 7.0 8.4 9.6 7.8 0.05 ^ 0.03 Ref.

0.10 0.08 0.08 0.02 0.00 0.08 0.04 0.06 0.04 0.04 0.00 0.06 0.04 0.04 0.06 0.06 0.04 0.04 0.14 0.04 0.12 0.02 0.02 0.02 0.02 0.02 0.08 0.10 0.06 0.08 0.06 0.42 ^ 0.12 Ref.

0.26 0.30 0.30 0.30 0.32 0.32 0.34 0.34 0.36 0.36 0.38 0.38 0.38 0.38 0.38 0.38 0.40 0.42 0.42 0.42 0.42 0.44 0.46 0.46 0.48 0.50 0.50 0.52 0.58 0.76 0.80

MS b/c (%)

65.0 ^ 35.4 0.510

100.0 30.0 88.9 61.2 57.8 53.0 91.6 81.0 30.0 38.1 31.2 34.0 62.5 140.0 70.8 41.3 47.9 29.7 150.0 33.3 132.0 28.6 80.9 70.2 152.0 82.6 116.0 83.3 31.2 20.8 20.8

BR (%)

Baseline b/c (%)

BR (%)

MI (%)

Frozen blood b

Fresh blood a

Rank by MS (mutagen sensitivity) of lymphocytes cultured from fresh whole-blood samples. Lymphocytes cultured from frozen whole-blood samples. Ref., reference group; P values were derived from a two-sided Student's t-test.

Mean ^ SD P value c

Sex

Rank (MS) a

Table 2 Comparison of g-ray-induced mutagen sensitivity and cytogenetic characteristics of two methods

5.3 ^ 1.8 0.001

7.2 5.2 8.4 4.2 8.6 4.2 5.8 3.0 7.8 2.8 7.6 5.8 7.6 6.4 5.4 4.0 5.6 5.2 5.6 4.4 9.2 4.8 8.2 3.8 3.8 4.2 3.4 4.6 5.2 3.4 5.0

MI (%)

0.06 ^ 0.03 0.194

0.08 0.02 0.08 0.04 0.06 0.06 0.04 0.06 0.00 0.08 0.04 0.14 0.04 0.04 0.04 0.04 0.04 0.02 0.14 0.08 0.08 0.06 0.04 0.04 0.02 0.06 0.08 0.06 0.04 0.14 0.12

Baseline b/c (%)

0.68 ^ 0.21 0.001

0.42 0.58 0.42 0.60 0.50 0.70 0.76 0.84 0.68 0.92 0.44 0.74 0.78 0.96 0.72 0.76 0.52 0.54 0.56 0.72 0.58 0.62 0.50 0.72 0.46 0.62 0.74 0.56 0.76 1.32 1.18

MS b/c (%)

L. Cheng et al. / Cancer Letters 166 (2001) 155±163 159

160

L. Cheng et al. / Cancer Letters 166 (2001) 155±163

gene expression in cultured lymphocytes recovered from frozen whole blood (mean ^ SD, 32 615 ^ 17 362 cpm) was signi®cantly higher (P ˆ 0:007) than that in frozen isolated lymphocytes (20 595 ^ 12 389 cpm), suggesting that the conditions or functions of the lymphocytes recovered from frozen whole blood were as good as those of the cryopreserved lymphocytes. Whereas previously at least 10 ml of whole blood was required to perform the HCR assay [13], the lymphocytes cultured from only 5 ml of frozen whole blood were suf®cient to successfully perform the assay in this study (Table 1). The mean DRC values in the frozen isolated lymphocytes (11.1 1 3.5) and in the lymphocytes recovered from frozen whole blood (11.6 1 3.1) were nearly identical (P . 0:10). Using the Wilcoxon rank sum test, we found no signi®cant difference between the results of the two methods (P . 0:10). More importantly, the DRC values for the two methods showed a strong correlation with a Pearson's correlation coef®cient of 0.77 (P , 0:001) (Fig. 1). These results demonstrated that recovering lymphocytes from frozen whole blood generated HCR assay results comparable with those of cryopreserving isolated lymphocytes. Therefore, using frozen whole blood would reduce both laborious processing at the time of receiving the samples, particularly on weekends, and require only half the blood sample for the assay.

To assess whether lymphocytes cultured from frozen whole blood could be used in the mutagen sensitivity assay, we established lymphocyte cultures from frozen whole-blood samples. As shown in Table 2, there were no signi®cant differences in the baseline level of chromatid breaks between fresh (0.05 ^ 0.03) and frozen (0.06 ^ 0.03) paired samples (P ˆ 0:194). Although there was no difference in response to PHA stimulation, the lymphocyte cultures from frozen whole blood had a decreased MI (5.3 ^ 1.8%) compared with that of fresh cultures (7.5 ^ 1.6%) (P ˆ 0:001), but this difference may be due to a physically delayed response to mitogen stimulation in the lymphocytes after freezing. In evaluating induced chromatid-break levels, we found that the lymphocytes recovered from frozen whole blood exhibited greater sensitivity to g-radiation (0.68 ^ 0.21) than did those from fresh blood (0.42 ^ 0.12) (P , 0:001). The Wilcoxon rank sum test showed a signi®cant difference between the two methods of obtaining lymphocytes (P , 0:01), suggesting that freezing and storage could cause the cells to be more sensitive to mutagenesis. However, the b/c values for the paired fresh and frozen samples (Fig. 2) had a Pearson correlation coef®cient of 0.61 (P , 0:01). The lymphocyte cultures from 1 ml frozen whole-blood samples provided a suf®cient

Fig. 1. Correlation of DRC values of isolated frozen lymphocytes and lymphocytes recovered from whole blood in 25 paired samples.

Fig. 2. Correlation between the b/c values for fresh and frozen whole blood in 31 paired samples in the mutagen sensitivity assay.

L. Cheng et al. / Cancer Letters 166 (2001) 155±163

number of cells for the required cytogenetic reading of the metaphases (50 per culture) [8], and the metaphases were satisfactory for evaluating chromatid breaks in the mutagen sensitivity assay. Furthermore, the correlation of measurements between cultures from fresh and frozen blood samples suggested that it was feasible to use frozen whole blood for the mutagen sensitivity assay. 4. Discussion In this study, we successfully cultured the lymphocytes recovered from frozen whole blood and used them for the mutagen sensitivity and HCR assays. We also demonstrated that the results from these two assays using viable lymphocytes obtained with the cryopreservation method were similar to those using traditional methods [4,13]. Our results are also consistent with a recent report in which frozen lymphocytes stored for up to more than a year did not have signi®cant changes in their cellular functions [14]. Therefore, the use of frozen whole blood provides a simple, quick, and economical way of storing viable lymphocytes for measurement of phenotypic biomarkers in future molecular epidemiology studies. It is conceivable that the blood itself is the best medium for storing viable lymphocytes. The advantages of this approach over the traditional ones are several. First, a smaller (reduced to a half) blood sample is needed for a functional assay that requires viable lymphocytes, which is particularly important when blood is the only biologic specimen available for biomarker studies. Second, less blood will be used in one assay, which leaves more blood available for other complementary assays. Third, the storage of viable lymphocytes in frozen whole blood allows functional assays to be performed in batches with an adequate control for assay variation due to samples being assayed on different days. The HCR assay is a well-established assay that has been used to measure DRC in several laboratories [5,10,13]. DRC varies among individuals and reduced DRC is considered to be a genetic predisposition to cancer [15±17]. Hence, a population-based assay using cellular DNA repair as a biomarker of genetic susceptibility to cancer is a vital tool for identifying

161

individuals at high risk of cancer. However, this assay requires a relatively large volume of blood (15±20 ml) because of the need to isolate lymphocytes, which limits its use in molecular epidemiology studies. The whole-blood cryopreservation method described here helps reduce the amount of blood (to only 5 ml) needed for the assay. In addition, the time-consuming steps of isolating lymphocytes at the time of receipt of blood samples are also reduced from several hours to one, which is particularly practical when samples come on weekends. This represents a great increase in laboratory ef®ciency in terms of time and costs. The mutagen sensitivity assay is an in vitro assay developed by Hsu [18] that quanti®es bleomycininduced breakage in cultured lymphocytes. To perform this assay, fresh whole-blood samples are brie¯y cultured with a mitogen to stimulate lymphocyte growth. After in vitro exposure to the test mutagens, the cells are arrested in metaphases in which chromosome aberrations can be evaluated. Combined cytogenetic and epidemiologic data have shown that mutagen sensitivity is a signi®cant risk factor in cancers of the lung, upper aerodigestive tract, and glioma [19±21]. In a large Nordic cohort study, increased levels of chromosomal breakage were also shown to be a predictor of cancer risk [22], which appeared to be independent of age, sex, and environmental exposure [23]. Thus, this relatively simple and inexpensive cytogenetic assay provides a biomarker for genetic susceptibility to cancer. However, the need for fresh blood cultures for this assay has limited its use. The daily processing and harvesting of the cells as samples arrive in the laboratory not only increase the possibility of assay variation but also make it dif®cult to schedule laboratory work due to the need to harvest cells after 72 h of stimulation, particularly for the weekends. The use of frozen whole blood would eliminate most of these limitations. More importantly, this cryopreservation approach allows us to perform the mutagen sensitivity assay in batches with an equal number of cases and controls, which should greatly improve the reproducibility of the assay. By using frozen whole-blood samples, we found a signi®cant increase in the number of g-radiationinduced chromatid breaks in the lymphocytes compared with that using fresh whole-blood samples. One possible explanation is that the freezing proce-

162

L. Cheng et al. / Cancer Letters 166 (2001) 155±163

dure may have increased the sensitivity of the lymphocytes by destroying all erythrocytes [24] and causing hemolysis during thawing. The consequent release of hemoglobin and its hazardous heme group into the cell cultures could then catalyze auto or photo oxidation reactions, potentially leading to free radical formation [25,26]. However, this effect was minimized by the cell-washing after thawing and should have been the same for every sample to be tested. It has been reported that the correlation coef®cient was only 0.72 for the same slides read by the same individual [27]. Therefore, the reading variation in b/c values rather than cryopreservation may have a major contribution to the relatively low correlation between fresh and frozen samples observed in this study. Although the cryopreservation of wholeblood samples will not reduce the overall reading variation in the mutagen sensitivity assay, it should reduce the variation associated with the assays performed at a different time. In summary, we have successfully grown lymphocytes from frozen whole blood and subjected them to the mutagen sensitivity and HCR assays. Conventional approaches using fresh blood and isolated frozen lymphocytes, respectively, for these two assays have limited their use in molecular epidemiologic studies, because of the need for a large amount of blood, the time-consuming processing of blood samples, particularly for weekends when the blood samples arrive, and the relatively high cost, particularly on a large scale. Therefore, the data presented here demonstrate the feasibility and ef®ciency of using frozen whole-blood samples for biomarker (functional) assays in future molecular epidemiologic studies. Nevertheless, the application of this approach of using frozen whole blood represents a major change in the study protocol, which should be further validated with samples stored for longer periods of time. Acknowledgements We thank Ms Xinyan Lu and Ms Ping Xiong for their technical support, Mr Don Norwood for his scienti®c editing, and Ms Joanne Sider and Ms Joyce Brown for manuscript preparation. This study was supported in part by National Institutes of Health

grants R03 CA 78425 (to L.C.), R01 CA 55769 (to M.S.), R01 CA 74851 and R29 70334 (to Q.W.) and by the Faculty Achievement Award in Cancer Prevention (to M.S.).

References [1] F.P. Perera, I.B. Weinstein, Molecular epidemiology and carcinogen-DNA adduct detection: new approaches to studies of human cancer causation, J. Chron. Dis. 35 (1982) 581± 600. [2] F.P. Perera, M.C. Poirier, S.H. Yuspa, J. Nakayama, A. Jaretzki, M.M. Curren, A pilot project in molecular cancer epidemiology: determination of benzo[a]pyrene-DNA adducts in animal and human tissues by immuno-assay, Carcinogenesis 3 (1982) 1405±1410. [3] P.A. Schulte, F.P. Perera, Molecular Epidemiology: Principles and Practices, Academic Press, New York, 1993. [4] M.R. Spitz, T.C. Hsu, X. Wu, J.J. Fueger, C.I. Amos, J.A. Roth, Mutagen sensitivity as a biologic marker of lung cancer risk in African Americans, Cancer Epidemiol. Biomarkers Prev. 4 (1995) 99±103. [5] Q. Wei, L. Cheng, W.K. Hong, M.R. Spitz, Reduced DNA repair capacity in lung cancer patients, Cancer Res. 56 (1996) 4103±4107. [6] S.J. Duthie, A. Hawdon, DNA instability (strand breakage, uracil misincorporation, and defective repair) is increased by folic acid depletion in human lymphocytes in vitro, Fed. Am. Soc. Exp. Biol. J. 12 (1998) 1491±1497. [7] A. Boyum, Separation of leukocytes from blood and bone marrow with special reference to factors which in¯uence and modify sedimentation properties of hematopoietic cells, Scand. J. Clin. Lab. Invest. 21 (1968) 1±109. [8] T.C. Hsu, E.J. Shillitoe, L.M. Cherry, Q. Lin, S.P. Schantz, C. Furlong, Cytogenetic characterization of 20 human lymphoblastoid lines derived from human individuals differing in bleomycin sensitivity, In Vitro Cell. Dev. Biol. 26 (1990) 80±84. [9] M.L. Bondy, A.P. Kyritsis, J. Gu, M. de Andrade, J. Cunningham, V.A. Levin, J.M. Bruner, Q. Wei, Mutagen sensitivity and risk of gliomas: a case-control analysis, Cancer Res. 56 (1996) 1484±1486. [10] W.F. Athas, M. Hedayati, G. Matanoski, E.R. Farmer, L. Grossman, Development and ®eld-test validation of an assay for DNA repair in circulating human lymphocytes, Cancer Res. 51 (1991) 5786±5793. [11] J.H. McCuchan, J.S. Pagano, Enhancement of the infectivity of simian virus 40 deoxyribonucleic acid with diethylaminoethyl-dextran, J. Natl. Cancer Inst. 41 (1968) 351±356. [12] A.N. Moshell, R.E. Tarone, S.A. New®eld, A.D. Andrews, J.H. Robbins, A simple and rapid method for evaluating the survival of xeroderma pigmentosum lymphoblastoid cell lines after irradiation with ultraviolet, In Vitro 17 (1981) 299±307. [13] Q. Wei, M.A. Hedayati, G.M. Matanoski, DNA repair and

L. Cheng et al. / Cancer Letters 166 (2001) 155±163

[14]

[15] [16]

[17] [18] [19]

[20]

[21]

aging in basal cell carcinoma: a molecular epidemiology study, Proc. Natl. Acad. Sci. USA 90 (1993) 1614±1618. P. Schmezer, N. Rajaee-Behbahani, A. Risch, W. Ritten, K. Kayser, H. Dienemann, V. Schulz, P. Drings, H. Bartsch, Mutagen sensitivity and DNA repair capacity (DRC) as risk factors for non-small cell lung cancer, Proc. Am. Assoc. Cancer Res. 41 (2000) 322. P. Modrich, Mismatch repair, genetic stability, and cancer, Science 266 (1994) 1959±1960. R.B. Setlow, Variations in DNA repair among people, in: E. Castellani (Ed.), Epidemiology and Quantitation of Environmental Risk in Humans from Radiation and Other Agents, Plenum Press, New York, 1985, pp. 205±212. F. Oesch, W. Aulman, K.L. Platt, Individual differences in DNA repair capacities in man, Arch. Toxicol. Suppl. 10 (1987) 172±179. T.C. Hsu, Genetic instability in the human population: a working hypothesis, Hereditas 98 (1983) 1±9. Q. Wei, J. Gu, L. Cheng, M.L. Bondy, H. Jiang, W.K. Hong, M.R. Spitz, Benzo[a]pyrene diol epoxide-induced chromosomal aberrations and risk of lung cancer, Cancer Res. 56 (1996) 3975±3979. M.R. Spitz, J.J. Fueger, S. Halabi, S.P. Schantz, D. Sample, T.C. Hsu, Mutagen sensitivity in upper aerodigestive tract cancer: a case-control analysis, Cancer Epidemiol. Biomarkers Prev. 2 (1993) 329±333. M.L. Bondy, A.P. Kyritsis, J. Gu, M. Andrade, J. Cunningham, V.A. Levin, J.M. Bruner, Q. Wei, Mutagen sensitivity and risk of gliomas: a case-control analysis, Cancer Res. 56 (1996) 1484±1486.

163

[22] L. Hagmar, A. Brogger, I.L. Hansteen, S. Heim, B. Hogstedt, L. Knudsen, B. Lambert, K. Linnainmaa, F. Mitelman, I. Nordenson, C. Reuterwall, S. Salomaa, S. Skerfving, M. Sorsa, Cancer risk in humans predicted by increased levels of chromosomal aberrations in lymphocytes: Nordic study group on the health risk of chromosome damage, Cancer Res. 54 (11) (1994) 2919±2922. [23] S. Bonassi, L. Hagmar, U. Stromberg, A.H. Montagud, H. Tinnerberg, A. Forni, P. Heikkila, S. Wanders, P. Wilhardt, I.L. Hansteen, L.E. Knudsen, H. Norppa, Chromosomal aberrations in lymphocytes predict human cancer independently of exposure to carcinogens. European Study Group on Cytogenetic Biomarkers and Health, Cancer Res. 60 (6) (2000) 1619± 1625. [24] J.L. Wilmer, G.L. Erexson, A.D. Kligerman, Implications of elevated sister-chromatid exchange frequency in rat lymphocytes cultures in the absence of erythrocytes, Mutat. Res. 109 (1983) 231±248. [25] C.S. Foote, Photosensitized oxidation and singlet oxygen: consequences in biological systems, in: W.A. Pryor (Ed.), The Function of Red Blood Cells: Erythrocyte Pathobiology, Alan R. Liss, New York, 1976, pp. 153±172. [26] F.J. Hawco, P.J. O'Brien, Hydroxyl-radical formation during prostaglandin formation catalysed by prostaglandin cyclooxygenase, Biochem. Soc. Trans. 6 (1978) 1169±1171. [27] J. Lee, T. ZoltaÂn, T.C. Hsu, M.R. Spitz, W.K. Hong, A statistical analysis of the reliability and classi®cation error in application of the mutagen sensitivity assay, Cancer Epidemiol. Biomarkers Prev. 5 (1996) 191±197.