| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
1 Cancer Biomarkers and Prevention Group, University of Leicester, Leicester, United Kingdom;
2 Department of Obstetrics and Gynaecology and
3 MRC Molecular Endocrinology Group, University of Leicester, Leicester, United Kingdom;
4 Jack Birch Unit for Environmental Carcinogenesis, Department of Biology, University of York, York, United Kingdom;
5 York District Hospital, York, United Kingdom;
6 Lawrence Livermore National Laboratory, University of California, Livermore, California; and
7 Syngenta Ltd., Cheshire, United Kingdom
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
18 h prior to hysterectomy or breast surgery. Nonmalignant uterine tissue was separated into myometrium and endometrium. DNA and protein were isolated and bound radiolabel determined by the sensitive technique of accelerator mass spectrometry. Levels of irreversible DNA binding of tamoxifen in the endometrium of treated patients were 237 ± 77 adducts/1012 nucleotides (mean ± SE, n = 10). In myometrial tissues, a similar extent of DNA binding was detected (492 ± 112 adducts/1012 nucleotides). Binding of tamoxifen to endometrial and myometrial proteins was 10 ± 3 and 20 ± 4 fmol/mg, respectively. In breast tissue, sufficient DNA could not be extracted but protein binding was an order of magnitude higher than that seen with endometrial proteins (358 ± 81 fmol/mg). These results demonstrate that after oral administration, tamoxifen forms adducts in human uterine DNA but at low numbers relative to those previously reported in women after long-term tamoxifen treatment where levels, when detected, ranged from 15,000 to 130,000 adducts/1012 nucleotides. Our findings support the hypothesis that the low level of DNA adducts in human uterus is unlikely to be involved with endometrial cancer development. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tamoxifen is carcinogenic in rat liver and long-term administration leads to a dose-dependent increase in hepatic tumors (5 , 6) . The mechanism by which tamoxifen causes cancer in rat liver is now well established as involving metabolism of tamoxifen to reactive intermediates, which bind irreversibly to hepatic DNA, forming DNA adducts. In rat liver, tamoxifen DNA adducts are formed in a dose-dependent manner and accumulate with time (6) . DNA adducts are an early initiating event in the carcinogenic process because if they are not removed by repair mechanisms they can cause mutations by the incorporation of an incorrect base during DNA replication. Mutations present in critical genes such as tumor suppressors or proto-oncogenes may ultimately result in tumors. However, the presence of DNA adducts does not always lead to tumor formation, as illustrated by the fact that tamoxifen forms hepatic DNA adducts in mice, but this species does not develop tumors with continued tamoxifen dosing (7) . This is probably because tamoxifen adducts are formed at lower levels in mice than rats, and they do not accumulate with time. It is not yet clear whether the development of endometrial tumors in women are associated with a genotoxic or epigenetic mechanism.
There is conflicting evidence for the presence of tamoxifen-DNA adducts in human endometrium (8) . Using the 32P-postlabeling assay, adducts in endometrial DNA have been reported as either absent or formed at low levels, close to the limits of detection achievable with this assay (9, 10, 11, 12) . Therefore, to address the key question of tamoxifen-DNA adduct formation in humans, we have used the alternative, more sensitive technique of AMS.3 AMS is used for measuring concentrations of rare long-lived radioisotopes such as 14C. It is generally accepted that AMS is the most sensitive technique available for detecting DNA adducts with a limit of detection of 1–10 adducts/1012 nucleotides, depending on the specific activity (13) . In addition, detection is specific for 14C-radiolabeled chemicals, unlike 32P-postlabeling, which detects adduct formation derived from a range of known and unknown genotoxins. In this study, women due to undergo uterine or breast surgery were given a single therapeutic dose of [14C]tamoxifen citrate (20 mg). DNA from uterine tissues was assayed for covalently bound radiolabel using AMS. Protein binding was also measured as a surrogate for DNA binding in breast tissue and compared with that in the uterus.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Patients.
Prior to these studies taking place, protocols were independently reviewed by the Human Research Ethics Committee at York District Hospital or Leicester Health Authority, the Department of Health Committee on the Administration of Radioactive Substances to Persons–the Medicines (administration of radioactive substances) Regulation of 1978, United Kingdom, and the Institutional Review Board at the Lawrence Livermore National Laboratory. All patients who volunteered to take part were given an information leaflet describing the background and purpose of the study followed by a discussion with a member of the medical team. None of the patients had previously been treated with tamoxifen. Informed consent was obtained from all patients before studies commenced. Patients were also asked to fill in a questionnaire to obtain relevant personal information (such as age, height, and concurrent medication). Patient information can be found in Table 1
.
|
18 h before surgery. Urine was collected predose and up to 24 h after treatment. At surgery, nonmalignant breast or uterine tissue was removed. Samples were frozen and stored at -80°C. Uterine tissue was separated into myometrium and endometrium before use. Control tissue samples were also taken from 3 patients who did not receive [14C]tamoxifen.
Study 2: Leicester Royal Infirmary.
Ten female volunteers were recruited who were undergoing surgery for hysterectomy. Each subject was given 20 mg of tamoxifen citrate containing 1.85 MBq [14C]tamoxifen 18 h before surgery. This was a 5-fold higher dose of radioactivity than that used in the first study, although the dose of tamoxifen was the same and was necessary because the 0.37 MBq dose was found to be too low to enable unambiguous detection of radiolabel bound to uterine DNA. By using a higher radioactive dose, a greater proportion of any DNA adducts formed will be 14C-labeled, which increases the likelihood that they will be detectable by AMS. Venous blood (10 ml) was taken before the administration of the drug and at the time of surgery. Tissue samples were taken as described above, including from 5 women who had not received tamoxifen, which was used as controls.
Determination of Radioactivity in Urine, Blood, and Whole Tissues.
Radioactivity in pooled 24 h urine (1 ml aliquot) and plasma from whole blood (10 ml) were measured by liquid scintillation counting. Tissues (100 mg) were solubilized in Soluene (1 ml) overnight at 37°C and radioactivity measured by liquid scintillation counting or AMS. In some cases, both methods were used and values were averaged.
Determination of Covalent Binding of [14C]Tamoxifen to DNA and Protein.
Tissues were finely minced and digested in lysis buffer [800 mM guanidine-HCl, 30 mM EDTA, 30 mM Tris-HCl, 5% Tween 20, 0.5% Triton X-100 (pH 8.0)], containing proteinase K (1 mg/ml) for 18 h at 37°C. Undigested tissue was removed by centrifugation, and the supernatant was treated for 1 h at room temperature with RNase A, 0.19 mg/ml lysis buffer (14)
. DNA was extracted using Qiagen column chromatography (Qiagen Ltd., Crawley, United Kingdom) according to the manufacturers instructions. DNA purity was determined by the A260 nm/A280 nm ratio. Only DNA with a ratio of 1.6–1.9 was used. Using this DNA extraction method when tissue lysates are spiked with [14C]tamoxifen (226 fmol/mg tissue) at a concentration equivalent to tissue levels detected in this study, all 14C-radiolabel is efficiently extracted as demonstrated by there being no increase in 14C-radiolabel over that present in control DNA (data not shown). Furthermore, we have previously shown using this extraction method, and after DNA digestion and HPLC separation, the remaining 14C label coelutes on thin-layer chromatography plates with the tamoxifen-DNA adducts (15)
. Although it was not possible to carry out analogous studies with other known metabolites of tamoxifen because they are not available in radiolabeled form, they are all more polar than tamoxifen and therefore would be expected to be removed by Qiagen column chromatography more or as efficiently as tamoxifen itself. DNA samples (50–500 µg) were supplemented with a precise amount of carrier (2 mg of tributyrin) to provide sufficient carbon for optimal sample preparation and AMS analysis as described below. For protein isolation, tissues (50–100 mg) were homogenized in 20% (v/v) DMSO in methanol. The resulting protein pellets were exhaustively extracted with organic solvents as described previously (16)
. Proteins (2 mg) were dissolved in 0.1 M potassium hydroxide and submitted for analysis by AMS.
AMS Sample Preparation and Analysis.
Coded DNA and protein samples were sent to the Lawrence Livermore National Laboratory for AMS analysis; however, it was not possible to analyze controls and treated samples blind because the level of 14C in control samples is subtracted from that contained in the treated to determine the extent of drug binding. Protein and DNA samples were converted to elemental carbon using standard protocols, by combustion to CO2, followed by reduction to filamentous graphite (17)
. The resulting graphite was then packed into individual sample holders and analyzed by AMS. AMS measures the concentration of a rare isotope, in this case 14C, relative to a stable isotope, either 13C or 12C. From this ratio, the increase in 14C above background for a particular sample is determined by subtracting contributions due to the natural abundance of 14C in the sample and that due to sample processing (17
, 18) . The excess 14C content, which is caused by the presence of radiolabeled tamoxifen or related derivatives, was then converted into an appropriate form (fmol tamoxifen equivalents/mg protein or DNA adducts/1012 nucleotides), taking into account the specific activity of tamoxifen. The average Fraction Modern value of DNA from tamoxifen-treated patients in study 2, after subtraction of the carrier contribution, was 2.10 ± 0.09 compared with the natural radiocarbon content measured in untreated control patients (1.45 ± 0.04 Fraction Modern). The limits of detection were calculated based on the sum of the mean fraction modern for each set of control tissue samples and two times the SD of the mean.
Statistical Analysis.
Results are given as the mean ± SE and were tested using ANOVA with Dunnett’s test for significance at the 5% level. Each graphite sample was analyzed up to seven times for radiocarbon content by AMS or until the measurement variation was within ±5%.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). This agrees well with the report that a single dose of 20 mg gives a peak plasma level of 20 ng/ml after 24 h (19)
. The presence of radioactivity in tissues demonstrates that tamoxifen reached the uterus and breast, with 227 ± 36 fmol tamoxifen equivalents/mg (combined results from studies 1 and 2; n = 16) in uterine tissue and 335 ± 68 fmol tamoxifen equivalents/mg (n = 5) in breast tissue (Table 2)
. Similar levels of drug were detected in the uterine samples from studies 1 and 2, which was as expected because the same dose of tamoxifen was administered. Urinary excretion of radiolabel over the first 24 h after dosing with [14C]tamoxifen in patients from study 1 was 7.2 ± 0.6% (n = 11) of the total dose administered. These results are consistent with earlier studies that showed the majority of [14C]tamoxifen given to women was excreted in the feces (20)
.
|
and Table 2
). The typical limit of detection for samples analyzed from this study was 130 adducts/1012 nucleotides, lower than that achieved in study 1, which is a reflection of the 5-fold higher dose of radioactivity administered to the second group of patients. The average level of myometrial adducts was 2-fold higher than numbers of endometrial adducts, but this was not significant (P = 0.088). With breast tissue, sufficient DNA could not be extracted for AMS analysis, this was due partly to the small amount of tissue available and also because breast tissue is very fatty, which hinders the extraction process.
|
The extent of covalent binding of tamoxifen to breast and endometrial proteins was determined where sufficient material could be obtained. Protein binding was 10 ± 3 (n = 7) and 20 ± 4 (n = 9) fmol tamoxifen equivalents/mg protein for endometrium and myometrium, respectively (Fig. 2
, Table 2
). This is 14-fold greater than binding to DNA extracted from the same tissue. Binding of tamoxifen to normal breast tissue protein (358 ± 81 fmol tamoxifen equivalents/mg; n = 4) was 24-fold greater than that detected in the uterus in study 2 (Table 2)
.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Results from study 2 show that after this single dose procedure, only extremely low levels of uterine endometrial DNA adducts are detected in the order of 240 adducts/1012 nucleotides. Furthermore, there is no significant difference in DNA adduct levels between the endometrial and myometrial compartments. Although DNA adducts were not detectable in uterine tissue of patients recruited to study 1, increasing the dose of radioisotope in study 2 made it possible to detect the low levels of adducts formed. There need not be a linear correlation between DNA adduct level and the incidence of cancer, and therefore, low levels of DNA adducts do not necessarily equate to a low level of tumors or tumor risk; in fact, the presence of adducts can in some cases equate to no risk. For example, in rats, it has been shown that there is a threshold value of tamoxifen-induced DNA adducts (1,800,000 adducts/1012 nucleotides) required for subsequent induction of liver tumors (25) .
The level of tamoxifen equivalents in plasma (22 ng/ml) was an order of magnitude lower than the steady-state values of tamoxifen and its metabolites measured during long-term therapy (26
, 27)
. [14C]Radiolabel measured in uterine tissues was 6-fold higher than in plasma, however, after long-term exposure to tamoxifen, steady-state plasma levels are 10–60-fold higher than steady-state tissue levels (26
, 27)
. After a single dose of tamoxifen, drug metabolism and disposition may differ quantitatively compared with long-term exposure. The question as to whether tamoxifen forms adducts in endometrial DNA in treated women has been controversial. Previously, using indirect methods such as 32P-postlabeling, some groups report no 32P-postlabeled DNA adducts (9)
, whereas others have found low levels, ranging from 15,000 to 130,000 adducts/1012 nucleotides (12)
. More recently, using HPLC coupled with electrospray ionization tandem mass spectrometry analysis, tamoxifen DNA adducts were detected in the livers of rats and monkeys given this drug but were not detectable in endometrial samples from two women who had received tamoxifen (20 mg) for >1 year, presumably because they were below the limit of detection of the assay, which is 500 adducts/1012 nucleotides (28
, 29)
. The current study has shown that DNA adduct formation is possible in human endometrium and myometrium after a single dose of tamoxifen. If it were possible to conduct comparable studies in breast cancer patients receiving tamoxifen therapeutically, higher levels of uterine DNA adducts might be detected.
In rats given high doses (30 mg/kg), tamoxifen is a genotoxic liver carcinogen. 32P-Postlabeled DNA adducts are present at levels in the order of 30,000,000/1012 nucleotides after 6 months exposure when neoplasia is first detected (6)
. This level of adducts is comparable with 27 other rat liver carcinogens where the level of DNA adducts that resulted in a 50% tumor incidence spanned from 500,000 to 21,000,000 adducts/1012 nucleotides (30)
. No 32P-postlabeled uterine DNA adducts (31)
or uterine tumors (5)
have been reported in adult rats given tamoxifen in long-term studies. However, when rats are administered a single dose of [14C]tamoxifen, comparable with the dose used to treat women in this study, DNA adducts measured by AMS are detected in both the liver (3000 ± 300 adducts/1012 nucleotides; Ref. 32
) and uterus (160 ± 80 adducts/1012 nucleotides; unpublished data). In rats, this level of uterine DNA adduct formation is similar to that detected in the uterus of women in the present study, and both are at least an order of magnitude lower than that found in the liver of rats after a single dose.
For risk assessment purposes, comparative metabolism studies carried out in vitro show there is good reason to suppose that women will be far less sensitive to tamoxifen than rats (33) . For women, the risk of endometrial cancer increases with longer duration of tamoxifen use (P < 0.001), with relative risks of 2.0 for 2–5 years and 6.9 for at least 5 years compared with nonusers (34) . One mechanistic hypothesis proposed to explain how these tumors develop is that they arise as a result of a genotoxic action, involving the formation of low levels of endometrial DNA adducts after tamoxifen treatment, as detected in this study. However, the fact that similar levels of DNA adducts were also detected in the myometrium, a tissue where tumors do not develop in women, highlights that tumor formation is a multistep process, and DNA adduct formation is only one of the initial steps. Furthermore, it has been shown that a threshold of tamoxifen DNA adducts exists before liver tumors develop in rats (25) . Therefore, a more probable explanation in women is that uterine tumors develop as a result of the unopposed estrogenic action of this drug on cells of the endometrium, leading to increased mitotic activity of endometrial cells, increased number of DNA replication errors, and somatic mutations resulting in malignant phenotype (35) . Although a number of other drugs of this class such as toremifene and raloxifene are currently being tested as breast cancer therapeutic agents, the development of two others, levormeloxifene and idoxifene, has been discontinued primarily because of endometrial concerns after trials reported a higher proportion of pelvic organ prolapse in treated versus untreated women (36) .
In conclusion, tamoxifen forms adducts in human uterine DNA at very low levels compared with that seen in livers of rats treated with a single comparable dose or rats treated long term that develop liver tumors. This level of DNA adduct formation is unlikely to be mechanistically related to the development of endometrial cancer in women treated with tamoxifen.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: Elizabeth A. Martin, AstraZeneca, Genetic Toxicology Department, Safety Assessment UK, Mereside, Alderley Park, Macclesfield, Cheshire SK10 4TG, United Kingdom. Phone: 44-1625-231279; Fax: 44-1625-230782; E-mail: Elizabeth.martin{at}astrazeneca.com
3 The abbreviations used are: AMS, accelerator mass spectrometry; HPLC, high-performance liquid chromatography. 
Received 7/18/03. Revised 9/16/03. Accepted 9/22/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Hydroxytamoxifen is a substrate of hydroxysteroid (alcohol) sulfotransferase, resulting in tamoxifen DNA adducts. Cancer Res., 58: 647-653, 1998.This article has been cited by other articles:
|
|
 |
|
  K. Brown Is tamoxifen a genotoxic carcinogen in women? Mutagenesis, September 1, 2009; 24(5): 391 - 404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Schild-Hay, T. A. Leil, R. L. Divi, O. A. Olivero, A. Weston, and M. C. Poirier Tamoxifen Induces Expression of Immune Response-Related Genes in Cultured Normal Human Mammary Epithelial Cells Cancer Res., February 1, 2009; 69(3): 1150 - 1155. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Liapis, K. I.E. McLuckie, P. D. Lewis, P. B. Farmer, and K. Brown Mutagenicity of tamoxifen DNA adducts in human endometrial cells and in silico prediction of p53 mutation hotspots Nucleic Acids Res., October 1, 2008; 36(18): 5933 - 5945. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Brown, E. M. Tompkins, D. J. Boocock, E. A. Martin, P. B. Farmer, K. W. Turteltaub, E. Ubick, D. Hemingway, E. Horner-Glister, and I. N.H. White Tamoxifen Forms DNA Adducts in Human Colon after Administration of a Single [14C]-Labeled Therapeutic Dose Cancer Res., July 15, 2007; 67(14): 6995 - 7002. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. M. Arlt, E. Frei, and H. H. Schmeiser ECNIS-sponsored workshop on biomarkers of exposure and cancer risk: DNA damage and DNA adduct detection and 6th GUM-32P-postlabelling workshop, German Cancer Research Center, Heidelberg, Germany, 29-30 September 2006 Mutagenesis, January 1, 2007; 22(1): 83 - 88. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Y. Kim, N. Suzuki, Y. R. S. Laxmi, and S. Shibutani INEFFICIENT REPAIR OF TAMOXIFEN-DNA ADDUCTS IN RATS AND MICE Drug Metab. Dispos., February 1, 2006; 34(2): 311 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Y. Kim, Y. R. S. Laxmi, N. Suzuki, K. Ogura, T. Watabe, M. W. Duffel, and S. Shibutani FORMATION OF TAMOXIFEN-DNA ADDUCTS VIA O-SULFONATION, NOT O-ACETYLATION, OF {alpha}-HYDROXYTAMOXIFEN IN RAT AND HUMAN LIVERS Drug Metab. Dispos., November 1, 2005; 33(11): 1673 - 1678. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. H. Phillips, A. Hewer, M. R. Osborne, K. J. Cole, C. Churchill, and V. M. Arlt Organ specificity of DNA adduct formation by tamoxifen and {alpha}-hydroxytamoxifen in the rat: implications for understanding the mechanism(s) of tamoxifen carcinogenicity and for human risk assessment Mutagenesis, July 1, 2005; 20(4): 297 - 303. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. J. Schild, D. H. Phillips, M. R. Osborne, A. Hewer, F. A. Beland, M. I. Churchwell, K. Brown, M. Gaskell, E. Wright, and M. C. Poirier Hepatic DNA adduct dosimetry in rats fed tamoxifen: a comparison of methods Mutagenesis, March 1, 2005; 20(2): 115 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. A. Beland, M. I. Churchwell, D. R. Doerge, D. R. Parkin, D. Malejka-Giganti, A. Hewer, D. H. Phillips, P. L. Carmichael, G. Gamboa da Costa, and M. M. Marques Electrospray Ionization-Tandem Mass Spectrometry and 32P-Postlabeling Analyses of Tamoxifen-DNA Adducts in Humans J Natl Cancer Inst, July 21, 2004; 96(14): 1099 - 1104. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) and myometrial (
) DNA 18 h after the administration of 1.85 MBq of 14C-radiolabeled tamoxifen. After tissue digestion using proteinase K and RNase A, DNA was extracted using Qiagen column chromatography according to the manufacturers instructions. DNA samples were converted to elemental carbon by combustion to CO2, followed by reduction to filamentous graphite and analyzed by AMS to determine the concentration of 14C relative to 13C. From this measurement, the increase in 14C content over naturally occurring background levels (i.e., because of bound 14C-radiolabel) could be calculated. For endometrium, where tissue was limited, results were based on a single DNA extraction for patients 4, 5, 6, 7, 9, and 10; two extractions for patients 1, 2, and 8; and four extractions for patient 3. For myometrial tissue, between three and six extractions were performed for each patient.