- Email: [email protected]
The role of cell death and cell proliferation in the promotion of rat liver tumours by tamoxifen
CancerLetters 106(1996) 163-l 69
ELSEVIER
The role of cell death and cell proliferation in the promotion of rat liver tumours by tamoxifen P. Carthew*, B.M. Nolan, R.E. Edwards, L.L. Smith MRC
Toxicology
Unit, Hodgkin
Building,
University
ofleicestec
PO Box 138, Luncuster
Road, Leicesfer
I,El
9Hh’, UK
Received19April 1996;accepted2 May 1996
Abstract
Administration of tamoxifen to rats results in liver tumours with a latency time that is dependent on the strain of rat used. Wistar and Lewis rats develop liver tumours more rapidly than Fischer rats. Significant increases in the number of apoptotic hepatocytes were found in the Wistar and Lewis strains of rats after they were fed tamoxifen for up to 6 months, but not in Fischer rats. By 6 months of exposure to tamoxifen there were liver tumours in the Wistar and Lewis rats, but not the Fischers. Sustained elevations of the PCNA labelling index were found in the livers of tamoxifen-tre%ed Wistar and Lewis rats, over the first 6 months of tamoxifen treatment, but not Fischers. It is proposed that sustained cell death by apptosis may play a role in the mechanism of promotion of tamoxifen-induced liver tumours, by causing liver hyperpfasia. To support this concept it has been shown that cyloheximide, which causes apoptosis but not necrosis in the rat liver, c.auses DNA synthesis and cell division in hepatocytes. Keywords: Tamoxifen; Apoptosis; Proliferation; Liver; Tumour; Rat
1. Introduction
The hepatocarcinogenicity of tamoxifen to the rat has been demonstrated in numerous studies [l-6]. Because of the current prophylactic trial of tamoxifen in disease-free women, who have a high risk of developing breast cancer, the relevance of the rat liver tumours to women taking this drug is particularly important. To assess the possible risk factors to women, the mechanism of liver tumour development in rats given tamoxifen has received particular attention. There is no doubt that tamoxifen damages liver DNA,
resulting in adduct formation
* Correspondingauthor
[7-91, and this
has been quantitated in three strains of rats [lo]. Tamoxifen has also been shown to bind covalently to rat liver DNA [ 1 I], although the exact structures of all the DNA adducts have not been determined. Proposed tamoxifen DNA adducts include the bridge epoxide [12], and an adduct formed from a hydroxyethyl tamoxifen [ 131. Having established that tamoxifen is what could be described as a cumulative initiator of DNA damage, the well known role of tamoxifen as a promotor [14] has been examined. This demonstrated that tamoxifen, ar high doses, increased the labelling index in the liver as assessed by PCNA expression or BrdU incorporation into liver DNA in all three strains of rats examined for the first 3 months of exposure [lo]. Thereafter,
0304-3835/96/$12.00 0 1996Elsevier ScienceIrelandLtd. All tights reserved PII
SO304-3X35(96)04310-8
only two of
164
P. Carthew
et al. I Cancer
the three strains of rats maintained the increased labelling index, and these were the two strains that developed liver tumours by 6 months of exposure to tamoxifen. It seems reasonable to propose that the sustained hyperplasia of the liver in Wistar and Lewis strains of rat promotes the formation of liver tumours at 6 months, especially as the Fischer rats had no liver tumours at 6 months, and the initial increase in labelling index in these rats at 3 months was significantly depressed at 6 months [lo]. This pattern of promotion through increased cell proliferation led to the Wistar and Lewis rats all developing liver carcinomas by 11 months, while the Fischer rats took much longer to develop liver carcinomas (20 months) [lo]. The increase in cell proliferation in the liver could be due to an oestrogenic effect of tamoxifen in the liver; however, there is also the possibility that tamoxifen can cause toxicity in the liver resulting in cell death. If this were the case then compensatory hyperplasia might also be expected. Any increase in cell proliferation would augment the oestrogenic stimulation of the liver, increasing the promoting effect that cell proliferation could have on the initiated hepatocytes. In turn this would decrease the latency time to tumour in strains of rats where both effects were operating. To test this hypothesis we have examined the livers of the three strains of rats for which proliferation data and tumour incidence at the corresponding time points has been documented [lo] for evidence of cell death related to tamoxifen treatment. We have then correlated the presence of cell death (apoptosis) to hepatocellular proliferation, as measured by the PCNA labelling index and the interim sacrifice liver tumour incidence, as well as the overall latency time to tumour at the end of the bioassay. To underpin the hypothesis that cell death by apoptosis can, under some circumstances, lead to compensatory cell proliferation we have also examined the ability of cycloheximide to cause DNA synthesis and cell division in the rat liver. Cycloheximide is the only compound that has been shown so far to cause cell death by apoptosis, without necrosis, in experimental studies on the rat liver [ 151. If cycloheximide can cause hyperplasia of the rat liver at a dose known to cause significant apoptosis this would be consistent with the hypothesis that cell death by apoptosis could contribute to a promoting effect of tamoxifen in the rat liver.
Letters
106 (1996)
2. Materials
163-169
and methods
2.1. Animals and treatments 2.1.1. Tamoxifen administration Female F344/Tox (Fischer) and Wistar (LAC-P) rats were bred on site. LEW Ola (Lewis) rats were from Harlan Olac, Oxford, UK. The genetic authenticity of the inbred F344 and LEW rats was confirmed by comparing three microsatellite markers (ratcyp2a3a, ratcyp4al and ratiid2g [ 10,161, using DNA samples from the experimental animals and from five other F344 and LEW colonies. Animals, 6 weeks old, were kept in isolators, and fed either powdered control diet or one containing 420 ppm tamoxifen. The daily dietary intake of tamoxifen was monitored by twice weekly weighing of the feed containers. The daily dietary intake varied from 57 to 74 mg/kg body wt per day, increasing with age. This level of exposure to tamoxifen has previously been shown to produce a high incidence of hepatocellular carcinomas in rats [lo]. At 3 and 6 months, groups of five tamoxifen-treated rats and five controls in each strain were sacrificed. The remaining ten tamoxifenexposed rats of each strain were fed tamoxifencontaining diet until the 50% mortality endpoint for each group was reached, at which time the remaining animals in the group (and their respective controls) were sacrificed. The 50% mortality endpoint for Wistar and Lewis rats was 11 months and for the Fischer animals was 20 months. 2.1.2. Cycloheximide administration Wistar-derived 8 week old male Porton rats, in groups of four, were dosed with 3 mg/kg body weight of cycloheximide (or saline control) by the intraperitoneal route. After the administration of cycloheximide (or saline to control rats) all animals were allowed free access to water containing 80 mg/ 100 ml of BrdU (Sigma). Four animals were sacrificed at 6, 24, 48 and 72 h after cyclohexamide administration, as well as four controls at each time point. 2.2. Tissue preparation Livers were removed and weighed. Sections of all major lobes of the liver and other organs were fixed
P. Curthew et al. I Cancer Letters 106 (1996) 163-169
165
in 10% formaldehyde in buffered saline for routine histological examination and in Carnoy’s fluid for examination for evidence of apoptosis [17], and immunostaining for BrdU incorporation. 2.3. Histopathological
8
examination
Sections (5 pm) were preparedfrom paraffin waxembeddedliver tissue of all rats sacrificed. Representative sections of the left lateral, median, posterior, and crudate lobes were included on slides prepared and seained with haematoxylin and eosin for examination for apoptosis (Carnoy’s-fixed) and tumours (formalin-fixed). The index of apoptosis was determined for each animal by examining 7000 nuclei per animal and expressing the index as apoptotic bodies per 1000 nuclei using the criterion outlined previously [ 171. Tumour areas of the rat livers were excluded from examination in estimating the index of apoptosis. The PCNA labelling index data have been published previously [IO] and are included for comparison. The turnouts of the liver were classified according to tbeir hepwcellular origin [ 181.
6
g B 1” * 4
3
2
2.4. Qclokeximide-treated
mt liver cell replicative DNA synthesis as assessed by BrdV incorporation
548
Table
1
Body weights (g) of rats fed tamoxifen (0.42 g/kg) diet or basal powdered diet for up to 20 months Rat strain
F344 F344 Wistar Wistar kWlS
Lewis
Treatment
Tamoxifen Control Tamoxifen Control Tamoxifen Control
Sacrifice
in powdered
3 months
6 months
50% endpoint
152k-3 196*3 1.50+7 245 + 7 157*4 214k5
16824 222 * 5 16525 249+4 156+4 216+3
166k8 326 i 8 15959 284k9 15Ok.5 264k7
589
mortality
of F344 rats on dietary
tamoxifen
High Wycombe, Bucks., UK) as describedpreviously [19]. Immreactive nuclei were visual&d using 3,3’-diaminohentidine/H202 substrate aztd the sections were li@&Iy counterstained with haematoxylin. Sections of &o&urn, processedat the same time, served as positive controls. At least4QOQnuclei were counted on each section to derive the labelling index per 1000nuclei examined. 3. Resatts 3.1. Bwfy wei#zt tamoxifen
time
577 -w
Fig. 1. Cumulative
Paraffin sections(5 ,um) from Carnoy’s-fixed liver were rehydrated. Sections from rats treated with BrdU were treated first with a monoclonal rat antiBrdU antibody (1:lOO dilution, a gift from Dr. M. Omerod, Institute for Cancer Research, Sutton, Surrey, UK) and then with an anti-rat peroxidase conjugated second antibody (1:100 dilution, Dako Ltd.,
563
changes and nrortalily
for rats fed
The
basal di ment period (determined at sac&ice), are Shown in Table 1. Tamoxifen produces a severe inhibition of the normal growth of rats, as has been noted previously [4,9]. The cumulative mortality of the three
166
P. Carthew
et al. I Cancer
Letters
106 (1996)
163-169
strains of rats after the 6 month sacrifice is shown in Figs. l-3. 3.2. Index of apoptosisfor the three rat strains at 3 and 6 monthsof tamoxifen treatment Livers from three strains of rats administereddietary tamoxifen (420 ppm) were examined at interim sacrifice times in the lifetime bioassay. There was no evidence of necrosisin any of the rat livers after 3 or 6 months of tamoxifen administration. Apoptosis was significantly increasedin the non-tumour areasof the livers of Wistar rats at 3 and 6 months (Tables 2 and 3) of chronic tamoxifen treatment, and in the Lewis animals6 months after treatment (Table 3), but not in the livers of F344 rats at either the 3 or 6 month timepoints. Where there was a sustainedincrease in the numbers of apoptotic bodies or cells undergoing
Fig. 3. Cumulative mortality of Lewis rats on dietary tamoxifen.
apoptosisin the liver, the highest incidence of liver tumours was found (Wistar rats, Table 2). The Lewis rats had a lower incidence of tumours after 6 months, but had an increase in apoptotic index only at 6 months. F344 rats showed no evidence of liver tumours, or an increase in apoptotic index at 3 or 6 months in the study. The proliferation inde,x in these animals was increasedat 3 months, but significantly decreased at 6 months. The development of liver carcinomas in the F344 rats occurred over a longer time period (20 months for the 50% endpoint to be reached in this group, compared to 11 months in both the Wistar and Lewis rats [ IO]). 3.3. Effect of cycloheximide-induced apoptosison liver hyperplasia Fig. 2. Cumulative mortality of Wistar rats on dietary tamoxifen.
Examination of rat livers 6 h after intraperitoneal
P. Carthew
et al. I Cancer
Letters
106 (1996)
167
163-169
Table 2 Summary of the relationship between apoptosis, cell proliferation and tumour development in three strains of rats given dietary tamoxifen (420 ppm) for 3 months
Table 4 Summary of the increase in DNA synthesis and cell division in male Wistar rat liver after cycloheximide (CHX) tmatment
Rat strain
Time after CHX (h-’ )
F344 Wistar Lewis
Treatment
Tamoxifen Control Tamoxifen Control Tamoxifen Control
Apoptosis indexa
0.4*Oo.l 0.2 + 0.1 0.9 f 0.2* 0.2 20.1 0.3 *O.l 0.2 f 0.1
PCNA labelling indexb 10.2 + 1 2.9 TIZ 0.6 6.5 + 0.9* 1 + 0.2 8.3 + 0.8* 1.3 i 0.4
Hepatocellular adenoma incidence o/5 o/5 015 o/5 o/5 o/5
“The index of apoptosis is expressed as the mean number f SE of apoptotic cells/IO3 nuclei; a minimum of 7 X lo3 nuclei were examined. bThese data are taken from Ref. [IO] and am included for detailed comparison. *Significantly increased compared to the control value at the 5% level.
injection of cycloheximide revealed numerous apoptotic cells and apoptotic bodiesoften present in hepatocytes, ashas been shown previously [ lS].There was
Table 3 Summary of the relationship between apoptosis, cell proliferation and tumour development in three strains of rats given dietary tamoxifen (420 ppm)for 6 months Ral strain
F344 Wistar Lewis
Treatment
Tamoxifen Control Tamoxifen Control Tamoxifen Control
Apoptosis indexa
0.9 f 0. I 0.7kO.l 3 zt 0.65* 0.8 + 0.4 I .9 + 0.2* 0.8 k 0. I
PCNA labelling indexb 2.2 + 0.3** 4.3 + 0.6 8.5 + 1.5* 2.4 f 0.3 3.6 + 0.5 2.53 + 0.3
Hepatocellular adenoma incidence o/5 o/5 315 O/5 115 o/5
“The index of apo tosis is expressed as the mean number * SE of apoptotic cells/IO t: nuclet; a mnnmum of 7 X lo3 nuclei were examined. bThese data are taken from Ref. [IO] and are included for detailed comparison. *Significantly increased compared to the control value at the 5% level. **Significantly decreased compared to the control value at the 5% level.
24 48 72
Number of nuclei incorporating BrdU/103 nucleia
Number of cells undergoin mitosis/l d cellsb
CHX
Control
CHX
Control
3 * 0.7 34.5 + 8 99.2 + 8.5”
o.10*t3.1 17zk3.7 30 + 6.2
0 0.42 it 0.24 6.3 + 1.1*
0.36 f 0.2 0.35 2 0.2 0.3 f 0.3
a1.5 X IO3 examined per animal. Values are expressed as the mean i SEM. bl.5 X lo3 examined per animal. Values are expressed as the mean + SEM. *Statistically significant at the 5% level.
no evidence of liver cell necrosisat 6, 24, 48 or 72 h. DNA synthesis as demonstratedfollowing the incorporation of BrdU into nuclei, demonstratedimmunocytochemically, did not increase above the control values at 24 or 48 h after dosing. This was attributed to the inhibition of protein synthesisthat is the major experimental use of cycloheximide. After 72 h there was a statistically significant increase in the number of liver cell nuclei incorporating BrdU in the cycloheximide-treated rats (Table 4). This was accompanied by an increase in the number of hepatocytes undergoing cell division (Table 4) indicating that there was compensatory hyperplasia of the liver in responseto cycloheximide-induced apoptosis. 4. Discussion The concept that apoptosis may stimulate compensatory hyperplasia seemsan unlikely one, given that it is a well documented mechanismfor the deletion of cells during the involution of organs and in embryonic development, where the biological objective would be to decreasethe number of cells in the organ concerned [20,21]. However, the process by which apoptosis occurs, which was thought to be activation of the endonuclease[22], is now known not to be a universal pathway in all cell types undergoing apoptosis [23]. If the mechanismsby which apoptosisoccurs can have differing pathways, it does not seem unreasonableto propose that the outcome
168
P. Carthew
et al. I Cancer
of apoptosis could also have more than one effect. Under some circumstances apoptosis could give rise to a sustained hyperplasia in an organ the way that necrosis is thought to, in some experimental exposures at the maximum tolerated dose. This has long been proposed as a mechanism that promotes tumour formation in rodent bioassys [24]. We are proposing that apoptosis as a result of sustained toxicity may have the same effect as necrosis. The data derived from the present tamoxifen exposure of three rat strains fits this concept over the first 6 months of exposure, as is summarised in Fig. 4. The ultimate experimental test of the hypothesis would be to examine a variety of different compounds that induce only apoptosis, in the absence of necrosis, to determine if this is a common phenomenon. This is not possible, at present, because there is only one compound that has been shown to have this property, namely cycloheximide [15]. We have shown that cyclohexamide caused regenerative hyperplasia in the rat liver, even though it is not regarded as a mitogenie compound. This result is consistent with the hypothesis that we are proposing; namely, that toxininduced apoptosis could stimulate regeneration of the liver and that this in turn could play a part in the mechanism of tumour promotion in the rat by tamoxifen. The recent finding that tamoxifen induces TGF-P
WISTAR
LEWIS
I
sustained (3
cell death
cell death
& 6 months)
(6 months)
L
1
sustained (3
1
hyperplasia
FISCHER 1
no cell death
1
hyperplasia
no sustained
& 6 months)
(3 months)
hyperplasia
I
L
1
3/5 rats with liver tumors
l/5
rats with
no liver tumors
liver tumors
Fig. 4. Summary of the possible factors involved in the mechanism of liver tumour formation by tamoxifen in three strains of rats’over the first 6 months of treatment.
Letters
106 (1996)
163-169
in mice [25] could explain the mechanism by which tamoxifen can induce apoptosis in some strains of rats. TGF-/3 has been shown experimentally to induce apoptosis in the rat liver and in isolated rat hepatocytes [23], and if this were the case in Wistar and Lewis rats, in particular, then the finding of apoptosis in the livers of these two strains of rats given tamoxifen in the present study, would not be unexpected. References [II
Rattel, B., Loser, R., Dahme, E.G., Liehn, H.D. and Seibel, K. (1987) Comparative toxicology of droloxifene (3-OH tamoxifen) and tamoxifen. Hepatocellular carcinomas induced by tamoxifen. Biennial Int. Breast Cancer Res. Conf., F18, Miami, FL. M.V. and VI Williams, G.M., latropoulos, M.J., Djordjevic, Kaltenberg, O.P. (1993) The triphenylethylene drug tamoxifen is a strong liver carcinogen in the rat. Carcinogenesis, 14, 315-317. S., Goonetilleke, R., Nunn, G., Topham, J. and [31 &eaves, Orton, T. (1993) Two-year carcinogenicity study of tamoxifen in Alderley Park Wistar-derived rats. Cancer Res., 53,3919-3924. P., Hirsimaki, Y., Nieminen, L. and Payne, B.J. [41 Hirsimaki, (1993) Tamoxifen induces hepatocellular carcinoma in rat liver: a l-year study with two antiestrogens. Arch. Toxicol., 67.49-54. P., Martin, E.A., White, I.N.H., Edwards, R.E., [51 Carthew, Dorman, B.M. and Smith, L.L. (1995) Tamoxifen induces short-term cumulative DNA damage and liver tumours, with and without promotion by phenobarbital. Cancer Res., 55, 544-547. PI Dahme, E. and Rattel, B. (1994) Unlike tamoxifen, droloxifene produces no hepatic tumors in the rat. Oncologie, 17 (Sl), 6-16. of covalent DNA [71 Han, X. and Liehr, J.C. (1992) Induction adducts in rodents by tamoxifen. Cancer Res., 52, 13601363. 181 White, I.N.H., De Matteis, F., Davies, A., Smith, L.L., Crofton-Sleigh, C., Venitt, S., Hewer, A. and Phillips, D.H. (1992) Genotoxic potential of tamoxifen and analogues in female Fischer F344/n rats, DBAR and C57/6 mice and in human MCL-5 cells. Carcinogenesis, 13, 2197-2203. M.J, Jordon, K., Radi. L., Kaltenr91 Hard, G.C., latropoulos, berg, O.P., Imondi, A.R. and Williams, G.M. (1993) Major differences in the hepatocarcinogenicity and DNA adduct forming ability between toremifene and tamoxifen in female Crl:CD(BR) rats. Cancer Res., 53.4534-4541. [lOI Carthew, P., Rich. K.J., Martin, E.A., DeMatteis, F., Lim, C.K., Manson, M.M., Festing, M.F.W., Gant, T.W., White, I.N.H. and Smith, L.L. (1994) DNA damage as assessed by 32P-postlabelling in three rat strains exposed to dietary tamoxifen: the relationship between cell proliferation and liver tumour formation. Carcinogenesis, 16, 1299-1304.
P. Curthew [II]
1121
[I 31
[ 141
[15]
[I61
[17]
[ 181
et al. / Cuncer
Martin, E.A., Turtletaub, K.W., Heydon, R., Davies, A.M., White, I.N.H. and Smith, L.L. (1995). Characterisation of tamoxifen-induced DNA adducts formed in rat liver. Toxicologist, 15, 152 (Abstract). Phillips, D.H., Hewer, A., White, I.N.H. and Farmer, P.B. (I 994) Co-chromatography of a tamoxifen epoxide deoxyguanylic acid adduct with a major DNA adduct formed in the livers of tamoxifen treated rats. Carcinogenesis, 15, 793795. Phillips, D.H., Carmichael, P.L., Hewer, A., Cole, K.J. and Poon. C.K. (1994) a-Hydroxytamoxifen, a metabolite of tamoxifen with exceptionally high DNA binding activity in rat hepatocytes. Cancer Res., 54,55 18-5522. Dragan. Y.P., Xu, Y. and Pitot, H.C. (1991) Tumor promotion as a target for estrogen/antiestrogen effects in rat hepatocarcinogenesis. Prev. Med., 20, 15-26. Ledda-Columbano, G.M., Coni, P., Faa, G., Manenti, G. and Columbano, A. (1992) Rapid induction of apoptosis in rat liver by cycloheximide. Am. J. Pathol., 140,545-549. Kunieda, T., Kobayashi, E., Tachibana, M., Ikadai, H. and Imamichl, T. (1992) Microsatellite foci of the rat (Rattus nomegicus). Mammalian Genome, 3, 564-567. Bursch. W., Paffe, S., Putz, B., Barthel, G. and SchulteHerman, R. (1990) Determination of the length of the histological stages of apoptosis in normal liver and in altered hepatic foci of rats. Carcinogenesis, I I, 847-853. Maronpot, R.R., Montgomery, C.A., Boorman, G.A. and McConnell, E.E. (1986) National Toxicology Program nomenclature for hepatoproliferative lesions of rats. Toxicol. Palhol., 14, 263-273.
Letters
106 (1996)
163-169
I69
1191 Green, J.A., Edwards, R.E. and Manson. M.M (1992) lmmunohistochemical detection bromodeoxyuridine-labelled nuclei for in vivo cell kinetic studies. In: Methods in Molecular Biology. Vol. 10, pp. 131-135. Editor M.M. Manson. Humana Press, Totowa, NJ. [20] Kerr. J F-R., Wyllie, A.H. and Currie. A.R. i 1972) .4poptosis: a basic biological phenomenon with wide-ranging im. plications m tissue kinetics. Br. J. Cancer, 26, 219-257 1211 Walker, N.I. and Globe, G.C. (1987) Ceil death anll cell proliferation during atrophy of the rat parotid gland induced by duct obstruction. J. Pathol., 153, 333-344 [22] Arends, M.J. and Wyllie. A.H. (1991) Apoptosis: rnechanisms and roles in parholagy. In: International Review of Experimental Pathology, 30: Molecular Cell Pathology, pp. 223-254. Editors: G.W. Richter and K. St>lez Academic Press, New York. [23] Oberhammer, F., Fritsch. G , Pavelka. M., Froschl, G., Tiefenbacher, R., Purchio, T. and Schulte-Herman. R. (1992) Induction of apoptosis in cultured hepatocytes and in the regressing liver by transforming growth f;ztor-beta 1 occurs without activation of an endonuclcast: Toxicol l,etl., 64.701-704. 1241 Carr. C.J. ancl Kolbye, A.C. (19Y I) A crilicluc of rhc use of maximum tolerated dose in bioassays to a;scss cancer risk from chemicals. Regul. Toxicol. Pharmacol , 14. 78-8’7 [2S] Grainger, D.J., Witchell, CM. and Metcalfe, J.C. rlY95) Tamoxifen elevates transforming growth factor-b and suppresses diet-induced formation of lipid It:zmnc in mouse aorta. Nature Med., 1067-l 071.
ELSEVIER
The role of cell death and cell proliferation in the promotion of rat liver tumours by tamoxifen P. Carthew*, B.M. Nolan, R.E. Edwards, L.L. Smith MRC
Toxicology
Unit, Hodgkin
Building,
University
ofleicestec
PO Box 138, Luncuster
Road, Leicesfer
I,El
9Hh’, UK
Received19April 1996;accepted2 May 1996
Abstract
Administration of tamoxifen to rats results in liver tumours with a latency time that is dependent on the strain of rat used. Wistar and Lewis rats develop liver tumours more rapidly than Fischer rats. Significant increases in the number of apoptotic hepatocytes were found in the Wistar and Lewis strains of rats after they were fed tamoxifen for up to 6 months, but not in Fischer rats. By 6 months of exposure to tamoxifen there were liver tumours in the Wistar and Lewis rats, but not the Fischers. Sustained elevations of the PCNA labelling index were found in the livers of tamoxifen-tre%ed Wistar and Lewis rats, over the first 6 months of tamoxifen treatment, but not Fischers. It is proposed that sustained cell death by apptosis may play a role in the mechanism of promotion of tamoxifen-induced liver tumours, by causing liver hyperpfasia. To support this concept it has been shown that cyloheximide, which causes apoptosis but not necrosis in the rat liver, c.auses DNA synthesis and cell division in hepatocytes. Keywords: Tamoxifen; Apoptosis; Proliferation; Liver; Tumour; Rat
1. Introduction
The hepatocarcinogenicity of tamoxifen to the rat has been demonstrated in numerous studies [l-6]. Because of the current prophylactic trial of tamoxifen in disease-free women, who have a high risk of developing breast cancer, the relevance of the rat liver tumours to women taking this drug is particularly important. To assess the possible risk factors to women, the mechanism of liver tumour development in rats given tamoxifen has received particular attention. There is no doubt that tamoxifen damages liver DNA,
resulting in adduct formation
* Correspondingauthor
[7-91, and this
has been quantitated in three strains of rats [lo]. Tamoxifen has also been shown to bind covalently to rat liver DNA [ 1 I], although the exact structures of all the DNA adducts have not been determined. Proposed tamoxifen DNA adducts include the bridge epoxide [12], and an adduct formed from a hydroxyethyl tamoxifen [ 131. Having established that tamoxifen is what could be described as a cumulative initiator of DNA damage, the well known role of tamoxifen as a promotor [14] has been examined. This demonstrated that tamoxifen, ar high doses, increased the labelling index in the liver as assessed by PCNA expression or BrdU incorporation into liver DNA in all three strains of rats examined for the first 3 months of exposure [lo]. Thereafter,
0304-3835/96/$12.00 0 1996Elsevier ScienceIrelandLtd. All tights reserved PII
SO304-3X35(96)04310-8
only two of
164
P. Carthew
et al. I Cancer
the three strains of rats maintained the increased labelling index, and these were the two strains that developed liver tumours by 6 months of exposure to tamoxifen. It seems reasonable to propose that the sustained hyperplasia of the liver in Wistar and Lewis strains of rat promotes the formation of liver tumours at 6 months, especially as the Fischer rats had no liver tumours at 6 months, and the initial increase in labelling index in these rats at 3 months was significantly depressed at 6 months [lo]. This pattern of promotion through increased cell proliferation led to the Wistar and Lewis rats all developing liver carcinomas by 11 months, while the Fischer rats took much longer to develop liver carcinomas (20 months) [lo]. The increase in cell proliferation in the liver could be due to an oestrogenic effect of tamoxifen in the liver; however, there is also the possibility that tamoxifen can cause toxicity in the liver resulting in cell death. If this were the case then compensatory hyperplasia might also be expected. Any increase in cell proliferation would augment the oestrogenic stimulation of the liver, increasing the promoting effect that cell proliferation could have on the initiated hepatocytes. In turn this would decrease the latency time to tumour in strains of rats where both effects were operating. To test this hypothesis we have examined the livers of the three strains of rats for which proliferation data and tumour incidence at the corresponding time points has been documented [lo] for evidence of cell death related to tamoxifen treatment. We have then correlated the presence of cell death (apoptosis) to hepatocellular proliferation, as measured by the PCNA labelling index and the interim sacrifice liver tumour incidence, as well as the overall latency time to tumour at the end of the bioassay. To underpin the hypothesis that cell death by apoptosis can, under some circumstances, lead to compensatory cell proliferation we have also examined the ability of cycloheximide to cause DNA synthesis and cell division in the rat liver. Cycloheximide is the only compound that has been shown so far to cause cell death by apoptosis, without necrosis, in experimental studies on the rat liver [ 151. If cycloheximide can cause hyperplasia of the rat liver at a dose known to cause significant apoptosis this would be consistent with the hypothesis that cell death by apoptosis could contribute to a promoting effect of tamoxifen in the rat liver.
Letters
106 (1996)
2. Materials
163-169
and methods
2.1. Animals and treatments 2.1.1. Tamoxifen administration Female F344/Tox (Fischer) and Wistar (LAC-P) rats were bred on site. LEW Ola (Lewis) rats were from Harlan Olac, Oxford, UK. The genetic authenticity of the inbred F344 and LEW rats was confirmed by comparing three microsatellite markers (ratcyp2a3a, ratcyp4al and ratiid2g [ 10,161, using DNA samples from the experimental animals and from five other F344 and LEW colonies. Animals, 6 weeks old, were kept in isolators, and fed either powdered control diet or one containing 420 ppm tamoxifen. The daily dietary intake of tamoxifen was monitored by twice weekly weighing of the feed containers. The daily dietary intake varied from 57 to 74 mg/kg body wt per day, increasing with age. This level of exposure to tamoxifen has previously been shown to produce a high incidence of hepatocellular carcinomas in rats [lo]. At 3 and 6 months, groups of five tamoxifen-treated rats and five controls in each strain were sacrificed. The remaining ten tamoxifenexposed rats of each strain were fed tamoxifencontaining diet until the 50% mortality endpoint for each group was reached, at which time the remaining animals in the group (and their respective controls) were sacrificed. The 50% mortality endpoint for Wistar and Lewis rats was 11 months and for the Fischer animals was 20 months. 2.1.2. Cycloheximide administration Wistar-derived 8 week old male Porton rats, in groups of four, were dosed with 3 mg/kg body weight of cycloheximide (or saline control) by the intraperitoneal route. After the administration of cycloheximide (or saline to control rats) all animals were allowed free access to water containing 80 mg/ 100 ml of BrdU (Sigma). Four animals were sacrificed at 6, 24, 48 and 72 h after cyclohexamide administration, as well as four controls at each time point. 2.2. Tissue preparation Livers were removed and weighed. Sections of all major lobes of the liver and other organs were fixed
P. Curthew et al. I Cancer Letters 106 (1996) 163-169
165
in 10% formaldehyde in buffered saline for routine histological examination and in Carnoy’s fluid for examination for evidence of apoptosis [17], and immunostaining for BrdU incorporation. 2.3. Histopathological
8
examination
Sections (5 pm) were preparedfrom paraffin waxembeddedliver tissue of all rats sacrificed. Representative sections of the left lateral, median, posterior, and crudate lobes were included on slides prepared and seained with haematoxylin and eosin for examination for apoptosis (Carnoy’s-fixed) and tumours (formalin-fixed). The index of apoptosis was determined for each animal by examining 7000 nuclei per animal and expressing the index as apoptotic bodies per 1000 nuclei using the criterion outlined previously [ 171. Tumour areas of the rat livers were excluded from examination in estimating the index of apoptosis. The PCNA labelling index data have been published previously [IO] and are included for comparison. The turnouts of the liver were classified according to tbeir hepwcellular origin [ 181.
6
g B 1” * 4
3
2
2.4. Qclokeximide-treated
mt liver cell replicative DNA synthesis as assessed by BrdV incorporation
548
Table
1
Body weights (g) of rats fed tamoxifen (0.42 g/kg) diet or basal powdered diet for up to 20 months Rat strain
F344 F344 Wistar Wistar kWlS
Lewis
Treatment
Tamoxifen Control Tamoxifen Control Tamoxifen Control
Sacrifice
in powdered
3 months
6 months
50% endpoint
152k-3 196*3 1.50+7 245 + 7 157*4 214k5
16824 222 * 5 16525 249+4 156+4 216+3
166k8 326 i 8 15959 284k9 15Ok.5 264k7
589
mortality
of F344 rats on dietary
tamoxifen
High Wycombe, Bucks., UK) as describedpreviously [19]. Immreactive nuclei were visual&d using 3,3’-diaminohentidine/H202 substrate aztd the sections were li@&Iy counterstained with haematoxylin. Sections of &o&urn, processedat the same time, served as positive controls. At least4QOQnuclei were counted on each section to derive the labelling index per 1000nuclei examined. 3. Resatts 3.1. Bwfy wei#zt tamoxifen
time
577 -w
Fig. 1. Cumulative
Paraffin sections(5 ,um) from Carnoy’s-fixed liver were rehydrated. Sections from rats treated with BrdU were treated first with a monoclonal rat antiBrdU antibody (1:lOO dilution, a gift from Dr. M. Omerod, Institute for Cancer Research, Sutton, Surrey, UK) and then with an anti-rat peroxidase conjugated second antibody (1:100 dilution, Dako Ltd.,
563
changes and nrortalily
for rats fed
The
basal di ment period (determined at sac&ice), are Shown in Table 1. Tamoxifen produces a severe inhibition of the normal growth of rats, as has been noted previously [4,9]. The cumulative mortality of the three
166
P. Carthew
et al. I Cancer
Letters
106 (1996)
163-169
strains of rats after the 6 month sacrifice is shown in Figs. l-3. 3.2. Index of apoptosisfor the three rat strains at 3 and 6 monthsof tamoxifen treatment Livers from three strains of rats administereddietary tamoxifen (420 ppm) were examined at interim sacrifice times in the lifetime bioassay. There was no evidence of necrosisin any of the rat livers after 3 or 6 months of tamoxifen administration. Apoptosis was significantly increasedin the non-tumour areasof the livers of Wistar rats at 3 and 6 months (Tables 2 and 3) of chronic tamoxifen treatment, and in the Lewis animals6 months after treatment (Table 3), but not in the livers of F344 rats at either the 3 or 6 month timepoints. Where there was a sustainedincrease in the numbers of apoptotic bodies or cells undergoing
Fig. 3. Cumulative mortality of Lewis rats on dietary tamoxifen.
apoptosisin the liver, the highest incidence of liver tumours was found (Wistar rats, Table 2). The Lewis rats had a lower incidence of tumours after 6 months, but had an increase in apoptotic index only at 6 months. F344 rats showed no evidence of liver tumours, or an increase in apoptotic index at 3 or 6 months in the study. The proliferation inde,x in these animals was increasedat 3 months, but significantly decreased at 6 months. The development of liver carcinomas in the F344 rats occurred over a longer time period (20 months for the 50% endpoint to be reached in this group, compared to 11 months in both the Wistar and Lewis rats [ IO]). 3.3. Effect of cycloheximide-induced apoptosison liver hyperplasia Fig. 2. Cumulative mortality of Wistar rats on dietary tamoxifen.
Examination of rat livers 6 h after intraperitoneal
P. Carthew
et al. I Cancer
Letters
106 (1996)
167
163-169
Table 2 Summary of the relationship between apoptosis, cell proliferation and tumour development in three strains of rats given dietary tamoxifen (420 ppm) for 3 months
Table 4 Summary of the increase in DNA synthesis and cell division in male Wistar rat liver after cycloheximide (CHX) tmatment
Rat strain
Time after CHX (h-’ )
F344 Wistar Lewis
Treatment
Tamoxifen Control Tamoxifen Control Tamoxifen Control
Apoptosis indexa
0.4*Oo.l 0.2 + 0.1 0.9 f 0.2* 0.2 20.1 0.3 *O.l 0.2 f 0.1
PCNA labelling indexb 10.2 + 1 2.9 TIZ 0.6 6.5 + 0.9* 1 + 0.2 8.3 + 0.8* 1.3 i 0.4
Hepatocellular adenoma incidence o/5 o/5 015 o/5 o/5 o/5
“The index of apoptosis is expressed as the mean number f SE of apoptotic cells/IO3 nuclei; a minimum of 7 X lo3 nuclei were examined. bThese data are taken from Ref. [IO] and am included for detailed comparison. *Significantly increased compared to the control value at the 5% level.
injection of cycloheximide revealed numerous apoptotic cells and apoptotic bodiesoften present in hepatocytes, ashas been shown previously [ lS].There was
Table 3 Summary of the relationship between apoptosis, cell proliferation and tumour development in three strains of rats given dietary tamoxifen (420 ppm)for 6 months Ral strain
F344 Wistar Lewis
Treatment
Tamoxifen Control Tamoxifen Control Tamoxifen Control
Apoptosis indexa
0.9 f 0. I 0.7kO.l 3 zt 0.65* 0.8 + 0.4 I .9 + 0.2* 0.8 k 0. I
PCNA labelling indexb 2.2 + 0.3** 4.3 + 0.6 8.5 + 1.5* 2.4 f 0.3 3.6 + 0.5 2.53 + 0.3
Hepatocellular adenoma incidence o/5 o/5 315 O/5 115 o/5
“The index of apo tosis is expressed as the mean number * SE of apoptotic cells/IO t: nuclet; a mnnmum of 7 X lo3 nuclei were examined. bThese data are taken from Ref. [IO] and are included for detailed comparison. *Significantly increased compared to the control value at the 5% level. **Significantly decreased compared to the control value at the 5% level.
24 48 72
Number of nuclei incorporating BrdU/103 nucleia
Number of cells undergoin mitosis/l d cellsb
CHX
Control
CHX
Control
3 * 0.7 34.5 + 8 99.2 + 8.5”
o.10*t3.1 17zk3.7 30 + 6.2
0 0.42 it 0.24 6.3 + 1.1*
0.36 f 0.2 0.35 2 0.2 0.3 f 0.3
a1.5 X IO3 examined per animal. Values are expressed as the mean i SEM. bl.5 X lo3 examined per animal. Values are expressed as the mean + SEM. *Statistically significant at the 5% level.
no evidence of liver cell necrosisat 6, 24, 48 or 72 h. DNA synthesis as demonstratedfollowing the incorporation of BrdU into nuclei, demonstratedimmunocytochemically, did not increase above the control values at 24 or 48 h after dosing. This was attributed to the inhibition of protein synthesisthat is the major experimental use of cycloheximide. After 72 h there was a statistically significant increase in the number of liver cell nuclei incorporating BrdU in the cycloheximide-treated rats (Table 4). This was accompanied by an increase in the number of hepatocytes undergoing cell division (Table 4) indicating that there was compensatory hyperplasia of the liver in responseto cycloheximide-induced apoptosis. 4. Discussion The concept that apoptosis may stimulate compensatory hyperplasia seemsan unlikely one, given that it is a well documented mechanismfor the deletion of cells during the involution of organs and in embryonic development, where the biological objective would be to decreasethe number of cells in the organ concerned [20,21]. However, the process by which apoptosis occurs, which was thought to be activation of the endonuclease[22], is now known not to be a universal pathway in all cell types undergoing apoptosis [23]. If the mechanismsby which apoptosisoccurs can have differing pathways, it does not seem unreasonableto propose that the outcome
168
P. Carthew
et al. I Cancer
of apoptosis could also have more than one effect. Under some circumstances apoptosis could give rise to a sustained hyperplasia in an organ the way that necrosis is thought to, in some experimental exposures at the maximum tolerated dose. This has long been proposed as a mechanism that promotes tumour formation in rodent bioassys [24]. We are proposing that apoptosis as a result of sustained toxicity may have the same effect as necrosis. The data derived from the present tamoxifen exposure of three rat strains fits this concept over the first 6 months of exposure, as is summarised in Fig. 4. The ultimate experimental test of the hypothesis would be to examine a variety of different compounds that induce only apoptosis, in the absence of necrosis, to determine if this is a common phenomenon. This is not possible, at present, because there is only one compound that has been shown to have this property, namely cycloheximide [15]. We have shown that cyclohexamide caused regenerative hyperplasia in the rat liver, even though it is not regarded as a mitogenie compound. This result is consistent with the hypothesis that we are proposing; namely, that toxininduced apoptosis could stimulate regeneration of the liver and that this in turn could play a part in the mechanism of tumour promotion in the rat by tamoxifen. The recent finding that tamoxifen induces TGF-P
WISTAR
LEWIS
I
sustained (3
cell death
cell death
& 6 months)
(6 months)
L
1
sustained (3
1
hyperplasia
FISCHER 1
no cell death
1
hyperplasia
no sustained
& 6 months)
(3 months)
hyperplasia
I
L
1
3/5 rats with liver tumors
l/5
rats with
no liver tumors
liver tumors
Fig. 4. Summary of the possible factors involved in the mechanism of liver tumour formation by tamoxifen in three strains of rats’over the first 6 months of treatment.
Letters
106 (1996)
163-169
in mice [25] could explain the mechanism by which tamoxifen can induce apoptosis in some strains of rats. TGF-/3 has been shown experimentally to induce apoptosis in the rat liver and in isolated rat hepatocytes [23], and if this were the case in Wistar and Lewis rats, in particular, then the finding of apoptosis in the livers of these two strains of rats given tamoxifen in the present study, would not be unexpected. References [II
Rattel, B., Loser, R., Dahme, E.G., Liehn, H.D. and Seibel, K. (1987) Comparative toxicology of droloxifene (3-OH tamoxifen) and tamoxifen. Hepatocellular carcinomas induced by tamoxifen. Biennial Int. Breast Cancer Res. Conf., F18, Miami, FL. M.V. and VI Williams, G.M., latropoulos, M.J., Djordjevic, Kaltenberg, O.P. (1993) The triphenylethylene drug tamoxifen is a strong liver carcinogen in the rat. Carcinogenesis, 14, 315-317. S., Goonetilleke, R., Nunn, G., Topham, J. and [31 &eaves, Orton, T. (1993) Two-year carcinogenicity study of tamoxifen in Alderley Park Wistar-derived rats. Cancer Res., 53,3919-3924. P., Hirsimaki, Y., Nieminen, L. and Payne, B.J. [41 Hirsimaki, (1993) Tamoxifen induces hepatocellular carcinoma in rat liver: a l-year study with two antiestrogens. Arch. Toxicol., 67.49-54. P., Martin, E.A., White, I.N.H., Edwards, R.E., [51 Carthew, Dorman, B.M. and Smith, L.L. (1995) Tamoxifen induces short-term cumulative DNA damage and liver tumours, with and without promotion by phenobarbital. Cancer Res., 55, 544-547. PI Dahme, E. and Rattel, B. (1994) Unlike tamoxifen, droloxifene produces no hepatic tumors in the rat. Oncologie, 17 (Sl), 6-16. of covalent DNA [71 Han, X. and Liehr, J.C. (1992) Induction adducts in rodents by tamoxifen. Cancer Res., 52, 13601363. 181 White, I.N.H., De Matteis, F., Davies, A., Smith, L.L., Crofton-Sleigh, C., Venitt, S., Hewer, A. and Phillips, D.H. (1992) Genotoxic potential of tamoxifen and analogues in female Fischer F344/n rats, DBAR and C57/6 mice and in human MCL-5 cells. Carcinogenesis, 13, 2197-2203. M.J, Jordon, K., Radi. L., Kaltenr91 Hard, G.C., latropoulos, berg, O.P., Imondi, A.R. and Williams, G.M. (1993) Major differences in the hepatocarcinogenicity and DNA adduct forming ability between toremifene and tamoxifen in female Crl:CD(BR) rats. Cancer Res., 53.4534-4541. [lOI Carthew, P., Rich. K.J., Martin, E.A., DeMatteis, F., Lim, C.K., Manson, M.M., Festing, M.F.W., Gant, T.W., White, I.N.H. and Smith, L.L. (1994) DNA damage as assessed by 32P-postlabelling in three rat strains exposed to dietary tamoxifen: the relationship between cell proliferation and liver tumour formation. Carcinogenesis, 16, 1299-1304.
P. Curthew [II]
1121
[I 31
[ 141
[15]
[I61
[17]
[ 181
et al. / Cuncer
Martin, E.A., Turtletaub, K.W., Heydon, R., Davies, A.M., White, I.N.H. and Smith, L.L. (1995). Characterisation of tamoxifen-induced DNA adducts formed in rat liver. Toxicologist, 15, 152 (Abstract). Phillips, D.H., Hewer, A., White, I.N.H. and Farmer, P.B. (I 994) Co-chromatography of a tamoxifen epoxide deoxyguanylic acid adduct with a major DNA adduct formed in the livers of tamoxifen treated rats. Carcinogenesis, 15, 793795. Phillips, D.H., Carmichael, P.L., Hewer, A., Cole, K.J. and Poon. C.K. (1994) a-Hydroxytamoxifen, a metabolite of tamoxifen with exceptionally high DNA binding activity in rat hepatocytes. Cancer Res., 54,55 18-5522. Dragan. Y.P., Xu, Y. and Pitot, H.C. (1991) Tumor promotion as a target for estrogen/antiestrogen effects in rat hepatocarcinogenesis. Prev. Med., 20, 15-26. Ledda-Columbano, G.M., Coni, P., Faa, G., Manenti, G. and Columbano, A. (1992) Rapid induction of apoptosis in rat liver by cycloheximide. Am. J. Pathol., 140,545-549. Kunieda, T., Kobayashi, E., Tachibana, M., Ikadai, H. and Imamichl, T. (1992) Microsatellite foci of the rat (Rattus nomegicus). Mammalian Genome, 3, 564-567. Bursch. W., Paffe, S., Putz, B., Barthel, G. and SchulteHerman, R. (1990) Determination of the length of the histological stages of apoptosis in normal liver and in altered hepatic foci of rats. Carcinogenesis, I I, 847-853. Maronpot, R.R., Montgomery, C.A., Boorman, G.A. and McConnell, E.E. (1986) National Toxicology Program nomenclature for hepatoproliferative lesions of rats. Toxicol. Palhol., 14, 263-273.
Letters
106 (1996)
163-169
I69
1191 Green, J.A., Edwards, R.E. and Manson. M.M (1992) lmmunohistochemical detection bromodeoxyuridine-labelled nuclei for in vivo cell kinetic studies. In: Methods in Molecular Biology. Vol. 10, pp. 131-135. Editor M.M. Manson. Humana Press, Totowa, NJ. [20] Kerr. J F-R., Wyllie, A.H. and Currie. A.R. i 1972) .4poptosis: a basic biological phenomenon with wide-ranging im. plications m tissue kinetics. Br. J. Cancer, 26, 219-257 1211 Walker, N.I. and Globe, G.C. (1987) Ceil death anll cell proliferation during atrophy of the rat parotid gland induced by duct obstruction. J. Pathol., 153, 333-344 [22] Arends, M.J. and Wyllie. A.H. (1991) Apoptosis: rnechanisms and roles in parholagy. In: International Review of Experimental Pathology, 30: Molecular Cell Pathology, pp. 223-254. Editors: G.W. Richter and K. St>lez Academic Press, New York. [23] Oberhammer, F., Fritsch. G , Pavelka. M., Froschl, G., Tiefenbacher, R., Purchio, T. and Schulte-Herman. R. (1992) Induction of apoptosis in cultured hepatocytes and in the regressing liver by transforming growth f;ztor-beta 1 occurs without activation of an endonuclcast: Toxicol l,etl., 64.701-704. 1241 Carr. C.J. ancl Kolbye, A.C. (19Y I) A crilicluc of rhc use of maximum tolerated dose in bioassays to a;scss cancer risk from chemicals. Regul. Toxicol. Pharmacol , 14. 78-8’7 [2S] Grainger, D.J., Witchell, CM. and Metcalfe, J.C. rlY95) Tamoxifen elevates transforming growth factor-b and suppresses diet-induced formation of lipid It:zmnc in mouse aorta. Nature Med., 1067-l 071.