Dihydrofolate Reductase Gene Amplification and Possible ...

THEJOURNAL. OF BIOLOGICAL CHEMISTRY Vol. 257, No.24, Issue of December 25, pp. 15079-15086, 1982 Printed in U.S.A.

Dihydrofolate Reductase Gene Amplification and Possible Rearrangement in Estrogen-responsive Methotrexate-resistant Human Breast Cancer Cells* (Received for publication, October 20, 1981)

Kenneth H. Cowan$&Merrill E. Goldsmith$, Richard M. Levinell, Susan C. AitkenJI, Edwin Douglassll, Neil ClendeninnS,Arthur W. Nienhuisl, and Marc E. Lippmanll From the $Clinical Pharmacology Branch and IlMedicine Branch, National Cancer Institute and YClinical Hematology Branch, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20205

Methotrexate-resistant (MTXR)human breast cancer cells have been obtained which are 1000-fold less sensitive to this drug than thewild type MCF-7 cells from which they were derived. The resistant cells contain approximately a 25-fold increase in the activity of the target enzyme dihydrofolate (DHF) reductase. Enzyme inhibition studies andmethotrexate affinity studies fail to reveal any difference in the DHF reductase present in the MTXR cells compared to wild type MCF-7 cells. Cytogenetic analysis demonstrates the presence of elongated marker chromosomes in the resistant cells which are not found in the parental cell line. Analysis of the DNA from MTXR and wild type MCF-7 cells using Southern blot hybridization indicates that the MTXR MCF-7 cells contain more copies of DHFreductase gene sequences than do wildtype MCF-7 cells. These experiments also suggest that the amplified DHF reductase gene sequences in MTXR cells may have undergone a uniform structuralrearrangement involving the 5’ flanking sequences during theprocess of amplification. MTXR MCF-7 cells respond to estradiol by increasing cell growth, and thelevel of DHFreductase in theMTXR cells is further induced following administration of estradiol. Radiolabeling studies demonstrate that estrogen stimulates the actual synthesis of DHF reductase in human breast cancercells.

Methotrexate, a potentinhibitor of dihydrofolate reductase (EC 1.5.1.3), is an effective antineoplastic agent with activity in a wide variety of human malignancies. Its clinical usefulness, however, is limited by the relative ease with which tumors develop resistance to this agent. Previous studies using methotrexate-resistant animal cell lines have described several mechanisms associated with resistance to this drug, including the following: 1) defective drug transport (1-4); 2) structurally altered dihydrofolate reductase with a reduced affiity for methotrexate (5-8); and 3) increased levels of dihydrofolate reductase (9-17). The lattermechanism appears to be a common mechanism by which animal cells become resistant to methotrexate following exposure to stepwise increases in drug concentration in vitro. Subsequent studies * This work has been presented in part at theAmerican Society of Clinical Investigation, April 26,1981, San Francisco, CA. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 0 To whom reprints should be addressed at Clinical Pharmacology Branch, Building 10, Room 6N116, National Cancer Institute, National Institutes of Health, Bethesda, MD 20205.

have shown that in two methotrexate-resistant animal cell lines, each with elevated levels of DHF’ reductase, the increased concentration of enzyme results from an increased rate of enzyme synthesis (18, 19). Work by Schimke and coworkers has shown that this increased enzyme synthesis is associated not only with a corresponding increase in the concentration of DHF reductase messenger RNA (20)but also in the selective amplification of the gene which codes for the enzyme (21). Both of these findings have been confirmed in other resistant animal cell lines (22, 23). Cytogenetic abnormalities associated with methotrexate-resistant hamster cells werefist reported by Biedler and Spengler(24). These workers noted elongated marker chromosomes with homogeneously staining regions in resistant cells. Other cytogenetic abnormalities found in methotrexate-resistant cells have included the appearance of double minute chromosomes which are small pairs of chromosomes lacking centromeres (25). Little is known regarding the ways in whichhuman tumors develop resistance to this drug. One study of patients with acute myelocytic leukemia suggested that the natural resistance of these patients to therapy with methotrexate is associated with the inherent ability of acute myelocytic leukemia cells to rapidly synthesize DHF reductase and accumulate intracellular methotrexate (26). However, few biochemical studies have been done in human neoplasms which have become resistant following treatment with methotrexate. This lack of understanding is due in large part to the technical difficulties inherent in the study of human tumor samples. In order to facilitate the study of the mechanisms of resistance to methotrexate in human cells, we developed methotrexateresistant human breast cancer cells by serial passage of a human breast cancer cell line (MCF-7) in stepwise increasing concentrations of methotrexate. Eventually methotrexate-resistant MCF-7 cells emerged which are over 1000-fold less sensitive to methotrexate than the wild type MCF-7 cells. As w l i beshown, these methotrexate-resistant human breast cancer cells contain increased concentrations of DHF reductase having the same apparent Ki and Kd for methotrexate as the enzyme present in the parental cell line.In addition, these cells have distinct elongated marker chromosomes and contain increased gene copies for DHF reductase relative to the wild type MCF-7 cells. However, in contrast to previous studies in murine cells containing amplified DHF reductase genes (27),restriction endonuclease analysis suggests that the amplified sequences in the MTXR MCF-7 cells have apparently undergone a structural alterationor rearrangement dur-

’

The abbreviations used are: DHF, dihydrofolate; MTXR, methotrexate-resistant; kb, kilobase pair.

15079

15080

Estrogen Response in DHF Reductase Gene-amplified MCF-7 Cells

ing the courseof amplification. These MTX' human breast cancer cells also retain the sensitivity to estrogens previously described in the wild type MCF-7 cells (28). While the level of DHF reductase in these MTX' cells is almost 25-fold higher than in the wild type MCF-7 cells, there is yet a further induction following the addition of estradiol.

A/W

6

C

D

MATERIALS AND METHODS*

RESULTS

Restriction Endonuclease Analysis-The finding of both increased concentrations of D H F reductase and elongated marker chromosomes in MTXH human breast cancer cells suggested that there might also be an increasein the number of gene copies of DHF reductase in these cells. DNA was isolated from both wild type and MTX' MCF-7 cells and treated in parallel with EcoRI restrictionendonuclease. The DNA fragments were analyzed bySouthern blothybridization to a radiolabeled mouse DHF reductase probe containing the entire coding region. Fig. 6 depicts an autoradiograph from such an experiment in which 35 pg of wild type DNA (lane W ) was compared to an identical amount of MTX" DNA which was analyzed in parallel (lane A ) . The autoradiograph for the wild type DNA hybridization was developed after 7 days at -80 "C,while the MTX' DNA autoradiograph was developed after an overnight exposure. Two important conclusions result from this study. First, it should be noted that the intensity of hybridization of radiolabeledmouse DHF reductase probe to each EcoRI fragmentproduced following digestion of MTX' DNA is as great if not greater than the hybridization to the wild type DNA. Since the intensity of hybridization is proportional to the length of the time of exposure, these results indicate that thereindeed is an amplification of DHF reductase gene sequences in the MTX" cells of approximately 10-fold over thatwhich is presentin the wild type MCF-7 cells. Second, the results depicted in Fig. 6 (lanes W and A ) indicate that there are not only differences in the quantityof the DHF reductase gene sequences in the MTXH cells, but there arealso qualitative differences as well. EcoRI digestion of wild type DNA results in three fragmentsof approximately equal intensity at 2.1 kb, 4.2 kb, and 6.6 kb, and a faint band which can occasionally be seen at 19 kb. In contrast,digestion of MTX" DNA produces two bands which are identical in size with that observed in the wild type DNA (6.6 kb and 4.2 kb) and a new band a t 1.8 kb which is not seen in the EcoRIdigest of wild type DNA. In this figure, the 2.1 kb band seen in the wild type EcoRI digest is not visible in the MTX" DNA. In other experimentsa faint 2.1 kb band is sometimesobserved in the EcoRIdigest of MTX' DNA which is clearly separate from the dense band appearing a t 1.8 kb and which always appears much lighterin intensity than the1.8,4.2, and 6.6 kb bands. Thus, although some of the restriction endonuclease sites are retained during amplification of DHF reductase gene sequences in theMTX" cells, some new EcoRIsitesare generated during the course of amplification. The results of Southern transfer experiments are consistent over a wide range of DNA concentrations. Furthermore,when the SouthPortions of this paper (including "Materials and Methods," part of "Results," Figs. 1-5, and Table I) are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 81M-2583, cite the authors, and include a set of photocopies. Full size check or money orderfor$5.60per photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

W

A

B

C

D

9.56.64.3-

2.11.9-

FIG. 6. Southern blot analysis of wild type and M T X R cell DNA. Wild type DNA (lane W ) andMTX' DNA (lanes A to D ) were treated with EcoRI, separated by agarose gel electrophoresis, transferred to nitrocellulose paper,andhybridized with different radiolabeled probes containing varying amounts of the mouse DHF reductase coding region (see top for individual probes). In each case, identical amounts (35 pg) of cell DNA were analyzed. The autoradiograph depicted in lane W (wild type DNA) was developed following a 7-day exposure; lanes A-D were developed following an overnight exposure. Lanes W and A were probed with the entire coding sequence; lane B with a 5' end probe; lane C with a probe containing the 3' end of the coding region; and lane D with the middle coding region probe. The numbers at the left represent the sizes (kilobase pairs) of the Hind111 fragments of XDNA which were run in parallel. On top is a diagram of the mouse DHF reductase-cloned cDNA as reportedbyChang et al. (37) and the fragments used in these experiments.

e m blot analysis is performed following digestion with other restriction enzymes, similar conclusions are reached, i.e. that there are both marked quantitative significant and qualitative changes in the structureof the DHF reductase gene sequences in the MTX" cells. In order to identify the nature of the apparent rearrangement of the amplified DHF reductase sequencesin the MTX' cells, EcoRI-digested MTX' DNA was electrophoresed and transferredto nitrocellulose paper as describedabove. In separate experiments, the digested DNA wasthen hybridized to radiolabeled probes corresponding to different portions of the mouse DHF reductase coding sequences (Fig. 6, lanes B, C, and D).The DNA in lane B was hybridized to a 5' end probe; lane C was hybridized to a 3' end probe; and lane D was hybridized to a probe containing the middle portion of the mouse DHF reductase coding sequence (see top, Fig. 6). The 4.2 kb EcoRI fragment of the MTX' genomic digest contains sequences homologous to the 3' end of the coding region (lane 0 , whereas the middle region of the coding sequences hybridizes predominantly with the 6.6 kb EcoRI fragment and toa lesser extent to the1.8 kb EcoRI fragment (lane D).Following hybridization with a radiolabeled fragment containing the 5' end of the mouse coding sequences,

Estrogen Response in

DHF Reductase Gene-amplified

only the 1.8 kb EcoRI fragment of the digested MTXR DNA is observed (lane B).This is the only band which appears in the amplified DNA and not in the EcoRI digest of the wild type DNA. These data areconsistent with the hypothesis that the MTXRDNA contains amplified DHF reductase sequences which have undergone an apparently uniform rearrangement in the upstream DNA sequences flanking the 5‘ end of the gene. Whether the original DHF reductase gene still exists in the MTXRcells is not answered by these studies but, since a 2.1 kb band is sometimes faintly visible in the MTXR DNA, it is possible that some of the original sequences may remain unaltered in the MTXR cells. These findings will be discussed in further detail below. MTXRMCF-7 Cells Are EstrogenResponsive-The parental MCF-7cell line has been shown to have high affinity binding proteins for estrogens (as),progesterone (46), glucocorticoids (46, 47), androgens (48, 49), and thyroid hormone (50). Furthermore, these cells respond to estradiol by an increase in cell growth (28), induction of DNA synthesis (28), and an induction of progesterone receptor (51). In order to determine if the MTXR MCF-7 cells retain the sensitivity to estrogen observed in the wild type MCF-7 cells, cellgrowthwasexaminedin the presence of either 1 nM estradiol or 1 IJM tamoxifen. As can be seen in Fig. 7, the growth of MTXR MCF-7 cellsis markedly increased by estradiol and inhibited by the anti-estrogen tamoxifen. By 14 days there is almost a 300% stimulation of cell growth in the presence of estradiol and approximately an 80%inhibition in

Id

-

MCF-7 Cells

15081

24 Hours

+El

T

FIG. 8. Effect of estrogen on DHF reductase activity. MTXR MCF-7 cells were plated in 47-mm Petri dishes in improved minimal essential medium containing 5% charcoal-stripped calf serum. When cells were nearly confluent the medium was changed to improved minimal essential medium without serum. Estradiol (10 nM) was added to half of the cells. At 24 h and 40 h the cells were harvested, sonicated, and the cytosol assayed for DHF reductase using radiolabeled folic acid. The results represent the specific activity of DHF reductase present in MTXRcells at various times after treatment and expressed as a fraction of the specific activity of DHF reductase present in cells at 24-h incubation in serum-free medium without estradiol. Each value represents the mean of triplicate samples & 1 standard deviation.

the presence of tamoxifen. These results are similar to those obtained with wild type MCF-7 (28). Since DHF reductase is an important enzyme in the synthesis of DNA, we were interested in determining whether estrogen had any effect on the level of DHF reductase in the resistant cells. As can be seen inFig. 8, following 24 h of incubation with estradiol there is a 50%increase in the specific activity of DHF reductase as compared to the basal level of activity present in these resistant cells incubated without hormone. Approximately a 100% increase in enzyme specific activity occurs after 40 h of incubation with estrogen. Thus, the base-line concentration of DHF reductase in the MTXR MCF-7cells is not only vastly increased compared to the parental cell line but can be increased still further by incubating the cells in the presence of estrogen. A similar although somewhat smaller (30%)induction in DHF reductase activity by estradiol is found in the wild type MCF-7 cells (data not shown). In order to identify the steps at which estrogen may regulate the level of DHF reductase in human breast cancer cells, 10 20 parallel cultures of MTXR MCF-7 cells wereincubated in the M estradiol. [35S]Methionine presence or absence of 1 X DAYS was added to the cultures 36 to 42 h after the addition of FIG. 7. Effects of estrogen andtamoxifen on growth of M T X R MCF-7 cells. MTXRMCF-7 (20,000cells/well) were plated in 47-mm estradiol. At the end of that period the cells were harvested and the amount of newly synthesized DHF reductase was Petri dishes in improved minimal essential medium containing 5% charcoal-stripped calf serum. The next day the medium was changed quantitated by affinity chromatography as described under and included either 10 nM estradiol (A-- -A) or 1 ~ L Mtamoxifen “Materials and Methods.” (W.. ..W). The control group consisted of MTXR cells (M)As shown in Table 11, following 42 h of incubation with grown in the improved minimal essential medium containing 5% estradiol there is a 61%increase in the level of DHF reductase charcoal-stripped calf serum only. The medium was changed every 4 as measured by [3H]methotrexate binding assay. During this days and a t various times the cells were harvested in phosphateperiod, estrogen increases total protein synthesis by 18%and buffered saline + EDTA and counted in a Coulter counter. Each value represents the mean cell number of triplicate samples & 1 there is a 34%increase in total soluble protein. These data are standard deviation. consistent with the observation that estrogen increases the

15082

Estrogen Response in DHF Reductase Gene-amplified

TABLE I1 Effect of estradiol on DHF reductase in MTXRMCF-7 cells MTXR MCF-7 cells were incubated in improved minimal essential medium f 10 nM estradiol for 36 h. One set of plates was analyzedfor actual DHF reductase level using r3H]methotrexate binding assay as describedunder“Materials and Methods.”Parallelcultureswere incubated for 6 h with 100 pCi of [35S]methioninein methionine-free media beginning at 36 h. Incorporation of radiolabel into total cell protein and into DHF reductase was analyzed as described under “Materials and Methods.” Results are expressed as the median of

tridicate samDles

* 1 standard deviation. [3HJMTx

Control +10 r m estradiol

Increase

dpm/mgprotem/6 h x IO-’

dpm/mgprotetn/6 h x

2.29 & .I2 2.70 f .06 18%

1.39 2 5 2.71 f .06 96%

F$f%{ez

14.41 & 0.45 23.2 f 2.2 61%

*

rate of growth of MTXRMCF-7 cells.During this same period there is a 96% increase in the actual rate of DHF reductase synthesis as determined by [35S]methionine labeling and methotrexate-Sepharose affinity chromatography. Thus estrogen not only increases total protein synthesis and accumulation in these breast cancer cells, it also stimulates an even greater increase in the rateof synthesis and accumulation of DHF reductase. DISCUSSION

In this report we describe the selection of methotrexateresistant human breast cancer cells whichare more than 1000fold less sensitive to thedrug than the parentalcell line. Drug resistance in these MTXR MCF-7cells is associated with increased levels of DHF reductase. Enzyme inhibition studies and methotrexate affinity studies fail to identify any major differences between the DHFreductase present in MTXRand wild type MCF-7 cells (see Figs. 1-4 and Table I in Miniprint). In addition, we have measured the transport of methotrexate in both cell lines and found no difference in drug uptake into either cell line (data not shown). Thus, resistance to methotrexate in these human breast cancer cells is associated primarily with the presence of increased concentrations of DHF reductase. While the MTXRMCF-7 cells are more than 1000-fold less sensitive to the drug than the parental cell line, these cells contain only a 25-fold increase in DHF reductase concentration. This lack of correlation between the magnitude of resistance and the increase in enzyme levelhas also been found in methotrexate-resistant mouse and hamster cells (10,11,13, 16,51).Several explanations may account for this observation. First, although we have found no differences in methotrexate transport into either wild type or MTXR MCF-7 cells, drug uptake into wild type (45) and MTXR MCF-7 cells is nonlinear. Thus, a one log increase in extracellular methotrexate concentration does not result in a similar increase in intracellular methotrexate concentration. Secondly, other mechanisms besides increased DHF reductase might contribute to the lack of sensitivity to methotrexate in these resistant cells, including changes in other enzyme activities (i.e.thymidylate synthetase or thymidine kinase) (52). Furthermore, methotrexate undergoes a complex series of conversions within MCF-7 cellsto poly-L-glutamyl derivatives (53,54). While the importance of these metabolites is not clear, it is conceivable that alteration in methotrexate metabolism including alterations in polyglutamate formation might contribute to resistance.

MCF-7 Cells

The cytogenetic abnormalities present in theseMTXR MCF-7cells(shown in Fig. 5) are similar to those noted previously in methotrexate-resistant animal cell lines (17,24). Elongated marker chromosomes have been previously shown to be associated with a drug-resistant phenotype which is relatively stable in animal cells in the absence of continuous selective pressure. This appears to be true in these resistant human cells as well, sincewe have detected essentially no loss in resistance over 7 months in the absence of selective pressure. In situ hybridization studies in methotrexate-resistant hamster and mouse cell lines containing elongated marker chromosomes have shown that theamplified DHF reductase genes are specifically localized,presumably as tandem repeats within the homogeneously staining region (55, 56). In murine L5178Y cells the homogeneously staining region wasidentifed as being part of the mouse number 2 chromosome (56). Unfortunately, numerous chromosomal translocations in the parental MCF-7 cells make it difficult to identify with certainty the origin of the marker chromosomes present in these human methotrexate-resistant cells. However, we have frequently been able to identify part of human chromosome number 7 and number 10 as part of the elongated marker chromosomes. Detailed cytogenetic studies on subclones of these MTXR MCF-7 cells and other methotrexate-resistant breast cancer cell lines which we have developed are currently in progress. It is anticipated that such studies might enable identification of the human chromosome(s) whichismost often involvedin the amplification of the DHF reductase gene. The increase in size of these marker chromosomes is remarkable. Even if the human DHF reductase gene is as large as the estimated size of the mouse DHF reductase gene (>35 kb), the size of the elongated marker chromosomes in the MTXR MCF-7 cells isstill several orders of magnitude larger than that needed to code for the increased DHF reductase gene copies. Similar conclusions have been made in MTXR animal cells (55-57). The significance of this additional DNA and whether it codes for any additional proteins remain unknown at present. Sodium dodecyl sulfate-gel electrophoresis demonstrates the presence of only one protein band which is in marked higher concentrations in the MTXRcells compared to the wild type MCF-7 cells and this band co-migrates with purified DHF reductase obtained from these cells. Using Southern blotting hybridization techniques, we have shown that these methotrexate-resistant human breast cancer cells contain amplified DHF reductase DNA sequences. Moreover, there apparently is a marked change in the pattern of restriction endonuclease sites present within the amplified DHF reductase genes or its flanking DNA sequences compared to the parental cell line. The Southern blot analysis depicted in Fig.6 indicates that thealteration is quite uniform and occurred early in the process of gene amplification. As noted above, previous studies in methotrexate-resistant mouse cells containing amplified DHF reductase genesfailed to detect any obvious difference in the restriction endonuclease sites present in the DHF reductase genes in parental and resistant cells (27). However, recent studies by Tyler-Smith and Alderson (57) have also noted DHF reductase gene rearrangements in amplified mouseEL 4 lymphoma cells selected for in uitro resistance to methotrexate. Furthermore, studies by Hiscott et al. on revertants of temperature-sensitive SV40 tsA-transformed mouse embryo cells have also suggested that structural alteration or rearrangement may accompany specific gene amplification (58). At the present time we are unable to identify the precise location of the apparent alteration in the amplifiedDNA


15083

human breast sequences of the MTXR MCF-7 cells. Hybridization using (63). N-(phosphonacety1)-L-aspartate-resistant different portions of mouse DHF reductase coding sequence cancer cells contain increased levels of aspartate transcarbahas shown that the alteration or rearrangement has appar- mylase (63)and also contain amplified DNAsequences coding ently occurred in the 5’ end of the DHFreductase gene or its for this enzyme.4However, in these drug-resistant human immediate upstream flanking sequences. However, an alter- cells, Southern blot analysis fails to detect anydifferences in native explanation for the difference in the wild type and the pattern of restriction endonuclease sites in the amplified MTXR genomic blotting patterns is possible. Recently, two DNA sequences compared to the parental cell line. Thus, it highly conserved intronless pseudogenes for DHF reductase appears that gene amplification may in fact be a common have been isolated from human recombinant DNA librarie~.~mechanism for development of drug resistance in human It has also been suggested that atleast in hamster cells there tumor cells, as has been demonstrated previously in animal may be at least two DHF reductase genes coding for enzymes cell lines.It is also apparent from these studies that structural with different molecular weights (23).The presence of at least alteration need not invariably accompany specific gene amtwo human pseudogenes in addition to one or more normal plification in human cells. Finally, we have shown that the MTXR MCF-7 cellsretain human DHF reductase genes, all of which hybridize to the DHF reductase probe, make it difficult to interpret genomic the hormonal sensitivity previously noted in the wild type blots. It is likely that not all the DHF reductase genes are MCF-7 cells (28)and both parental and MTXR cells respond to estradiol with increases in cell growth and DNA synthesis. amplified in our MTXR cells. An alternative explanation for the observation of a 2.1 kb In addition, we found that there is an 1.5- to 2-fold induction EcoRI fragment in the wild type DNA hybridizations and a of DHF reductase activity in MTXRMCF-7 cells in response 1.8 kb fragment in the MTXR DNA is possible. Perhaps the to estradiol. Thus, even though MTXR MCF-7 cellscontain a wild type 2.1 kb Eco fragment represents a portion of a DHF 25-fold increase in DHF reductase concentration relative to reductase gene or pseudogene which is not amplified in the that present in wild type MCF-7 cells, these resistant cells resistant cells. Hybridization to this fragment in the MTXR respond to estrogen by a further induction in DHF reductase DNA would be faint or undetectable compared to the ampli- levels. This induction of DHF reductase has been noted using fied sequences depending on the relative frequencies of the both a direct assay of enzyme activity and a radiolabeled two DHF reductase genesin the MTXR DNA. It is also methotrexate binding assay and indicates that there is an possible that the MTXR 1.8 kb fragment does not indicate a actual increase in the enzyme concentration and not simply gene rearrangement in the amplified DNAsequences but may an increase in activity secondary to a change in substrate instead represent the true5‘ end of the normal DHF reductase concentration, pH, or salt concentration, all factors known to genewhich is undetected by hybridization with wild type influence DHF reductase activity. The radiolabeling studies shown in Table I1 demonstrate DNA. If exon 1 of the human DHF reductase gene contains only a small region of homology with the mouse gene probe, that estrogen increases total protein synthesis in these drugthen these sequences may only bedetected when they become resistant breast cancer cells. It is furthermore apparent from amplified in the MTXK cells. In fact, the sequence of the these studies that estrogen has an even greater stimulatory recently isolated human DHF reductase cDNA is approxi- effect on the rate of synthesis of DHF reductase. We cannot mately 80% homologous to the first 72 nucleotides of the exclude the possibility that estrogen may also have an effect mouse codingsequences but thetwo sequences diverge signif- onenzyme stability whichmay contribute tothe overall icantly just 12 nucleotides upstream from the initiation site increase in DHF reductase levels observed followingincuba(59). We have attempted to identify the 5’ end of the DHF tion of these cells with estrogen. The increased rate of DHF reductase gene in the wild type DNA but, despite the use of reductase synthesis which followsthe administration of estrohigh concentrations of DNA (up to 60 pg) and prolonged gen suggests that thehormone or ahormone-inducible factor exposure times, these experiments have failed to determine may be acting at the level of the amplified DHF reductase with certainty this region of the wild type DNA. The use of a genes. Of course, it is possible that estrogen may increase human cDNA containing a longer region of homology of the DHF reductase synthesis through other mechanisms such as immediate 5’ flanking sequences should help to clarify the increased DHF reductase mRNA stability or increased effk question regarding rearrangement of the amplified genes. Con- ciency of translation. Mariani et al. (64) have recently demsidering the complexity of the DHF reductase gene locus, onstrated that DHF reductase synthesis increases during S including the presence of more than one DHF reductase gene phase. Since estrogen results in an increased rate of growth of and pseudogenes, definitive analysis of the amplifiedgene the MTXR MCF-7 cells, it is possible that the effect of the sequences in the MTXR cells must await the isolation and hormone on DHF reductase synthesis does not represent a structural characterization ofcloned DHF reductase gene direct hormone receptor-mediated event at the level of the sequences from both wild type and MTXR cells. DHF reductase gene but instead reflects a series of events Gene amplification as a mechanism of drug resistance is not which occur secondarily during the estrogen stimulation of unique to methotrexate, but hasalso been shown to occur in cell growth. animal cells resistant to N-(phosphonacety1)-L-aspartate(60The estrogen stimulation of DHF reductase synthesis sughamster cells gests that theamplified DHF reductase gene sequences pres62). N-(phosphonacety1)-L-aspartate-resistant contain increased levels of a multifunctional protein (CAD ent in the MTXR MCF-7 cells retain certain regulatory seprotein) which includes the enzyme activity inhibited by N - quences. The fact that amplified genes do retain regulatory (phosphonacety1)-L-aspartate (aspartate transcarbamylase) controls has been previously noted in studies of methotrexate(61).These resistant hamster cells have been shown to contain resistant animal cells, in which DHF reductase activity was increased copies of messenger RNA coding for the CAD further induced following addition of serum (65,66)and during protein and amplified DNA sequences complementary to this viral infection (67). In contrast, recent studies by Mayo and mRNA (62). We have also isolated N-(phosphonacetyl)-L- Palmiter have demonstrated that amplified metallothionein I aspartate-resistant human breast cancer cells in a fashion genes in cadmium-resistant mouse sarcoma cells may lose the similar to that described for isolation of MTXR MCF-7 cells ‘IC.H. Cowan, M. E. Goldsmith, R. M. Levine, S. C. Aitken, E.

Chen, J., Shimada, T., Moulton, A. D., Harrison, M., and Nienhuis, A. W. (1982)Proc. Natl. Acad. Sei.U. S. A , , in press.

Douglas,N. Clendeninn, A. W. Nienhuis, and M. E. Lippman, unpublished observations.

15084


responsiveness to glucocorticoid regulation while maintaining the response t o h e a v y metals observed in drug-sensitive ce& (68). Thus, amplified genes need n o th a v e the identical response to control signals as that observed inthe parental cells.

tFir

Acknowledgments"We thank J. Whang-Peng and E. Lee for assistance in the cytogenetic analysisand L. Ulsh for expert technical assistance. We thank B. Chabner for helpful discussions and for critically reviewing this manuscript. We also gratefully acknowledge the assistance of R. A. Rodbell and K. Moore in preparation of this manuscript. REFERENCES

1. Flintoff, W. F., Davidson, S. V., and Siminovitch, L. (1976) Somatic Cell Genet. 2, 245-261 2. Fischer, G. A. (1962) Biochem. Pharmacol. 11, 1233-1234 3. Sirotnak, F. M., Kurita, S., and Hutchison, D. J. (1968) Cancer Res. 28,75-80 4. Harrap, K. R., Hill, B. T., Furness, M. E., and Hart, L. I. (1971) Ann. N. Y.Acad. Sei. 186,312-324 5. Gupta, R. S., Flintoff, W. F., and Siminovitch, L. (1977) Can. J. Biochem. 55,445-452 6. Flintoff, W. F., and Essani, K. (1980) Biochemistry 19,4321- 4327 7. Jackson, R. C., and Niethammer, D. (1977) Eur. J. Cancer 13, 567-575 8. Haber, D. A., Beverley, S. M., Kiely, M. L., and Schimke, R. T. (1981) J. Bwl. Chem. 256,9501-9510 9. Albrecht, A. M., Biedler, J . L., and Hutchison, D. J. (1972) Cancer Res. 32, 1539-1546 10. Nakamura, H., and Littlefield, J . W. (1972) J. Biol. Chem. 247, 179-187 11. Littlefield, J. W. (1969) Proc. Natl. Acad. Sci.U. S. A . 62, 88-95 12. Fischer, G. A. (1961) Biochem. Pharmacol. 7, 75-77 13. Hakala, M. T., Zakrzewski, S. F., and Nichol, C. A. (1961) J. Biol. Chem. 236,952-958 14. Harding, N. G. L., Martelli, M. F., and Huennekens, F. M. (1970) Arch. Biochem. Biophys. 137,295-296 15. Biedler, J. L., Albrecht, A. M., Hutchison, D. J., and Spengler, B. A. (1972) Cancer Res. 32, 153-161 16. Biedler, J. L., Albrecht, A. M., and Spengler, B. A. (1978) Eur. J. Cancer 14,41-49 17. Bostock, C. J., Clark, E. M., Harding, N. G. L., Mounts, P. M., Tyler-Smith, C., van Heyningen, V., and Walker, P. M. B. (1979) Chromosoma 74, 153-172 18. Alt, F. W., Kellems, R. E., and Schimke, R. T. (1976) J. Biol. Chem. 251, 3063-3074 19. Hanggi, U. J., and Littlefield, J. W. (1976) J. Biol. Chem. 251, 3075-3080 20. Kellems, R. E., Alt, F. W., and Schimke, R. T. (1976) J. Biol. Chem. 251,6987-6993 21. Alt, F. W., Kellems, R. E., Bertino, J. R., and Schimke, R. T. (1978) J. Biol. Chem. 253, 1357-1370 22. Chang, S . E., and Littlefield, J . W. (1976) Cell 7, 391-396 23. Melera, P. W., Lewis, J . A,, Biedler, J . L., and Hession, C. (1980) J. Biol. Chem. 255,7024-7028 24. Biedler, J. L., and Spengler, B. A. (1976) Science (Wash. D. C . ) 191, 185-187 25. Kaufman, R. J., Brown, P. C., and Schimke, R. T. (1979) Proc. Natl. Acad. Sei. I/. S. A. 76,5669-5673 26. Bertino, J . R., Sawicki, W. L., Cashmore, A. R., Cadman, E. C., and Skeel, R. T. (1977) Cancer Treat. Rep. 61,667-673 27. Nunberg, J. H., Kaufman, R. J., Chang, A. C. Y., Cohen, S. N., and Schimke, R. T . (1980) Cell 19,355-364 28 Lippman, M., Bolan, G., and Huff, K. (1976) Cancer Treat. Rep. 60, 1421-1429 29. Soule, iI. D., Vazquez, J., Long, A., Albert, S., and Brennan, M. (1973) J. Natl. Cancer Inst. 51, 1409-1416 30. Richter, A., Sanford, K.K., and Evans, V. J . (1972) J. Natl. Cancer Znst. 49, 1705-1712 31. Rothenberg, S. P. (1966) Anal. Biochem. 16, 176-179 32. Hangg, U. J., and Littlefield, J. W. (1974) J. Biol. Chem. 249,

1399-1397 33. Lowry, 0.H., Rosebrough, N. J., F m , A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 34. Myers, C. E., Lippman, M. E., Eliot, H. M., and Chabner, B. A. (1975) Proc. Natl. Acad. Sei.U. S. A . 72,3683-3686 35. Oliverio, V. T. (1961) Anal. Chem. 33, 263-265 36. Zaharko, D. S., Bruckner, H., and Oliverio, V. T. (1969) Science (Wash.D. C . ) 166,887-888 37. Chang, A. c, y,,Nunberg, J. H,,Kaufman, R. J,, H, A,,

Schimke, R. T., and Cohen, S. N. (1978) Nature (Lond.)275, 617-624 38. Kretschmer. P. J.. Kaufman. R. E.. Coon. H. C.. Chen. M.-J. Y.. Geist, C. E., and NienhuG, A. W . (1980) J. Biol. Chem.~255; 3204-3211 39. Clewell, D. B. (1972) J. Bucteriol. 110, 667-676 40. Benz, E. J., Jr., Kretschmer, P. J., Geist, C. E., Kantor, J. A,, Turner, P. H., and Nienhuis, A. W. (1978) J. Biol. Chem. 254, 6880-6888

41. Wigler, M., Pellicer, A., Silverstein, S., and Axel, R. (1978) Cell 14, 725-731 42. Burton, K. (1956) Biochem. J. 62, 315-323 43. Southern, E. M. (1975) J. Mol. Biol. 98, 503-517 44. Lawn, R. M., Fritsch, E. F., Parker, R.C., Blake, G., and Maniatis, T . (1978) Cell 15, 1157-1174 45. Schilsky, R. L., Bailey, B. D., and Chabner, B. A. (1981) Biochem. Pharmacol. 30, 1537-1542 46. Horwitz, K. B., Costlow, M. E., and McGuire, W.L. (1975) Steroids 26, 785-795 47. Lippman, M., Bolan, G., and Huff, K. (1976) Cancer Res. 36, 4610-4618 48. Lippman, M., Bolan, G., and Huff, K. (1976) Cancer Res. 36, 4602-4609 49. Burke, R. E., Zava, D. T., and McGuire, W. L. (1977) Clin. Res. 25,461a 50. Horwitz, K. B., and McGuire, W. L. (1977) Clin. Res. 25, 295a 51. Horwitz, K. B., and McGuire, W. L. (1978) J. Biol. Chem. 253, 2223-2228 52. White, J. C., and Goldman, I. D. (1981) J. Biol. Chem. 256,57225727 53. Schilsky, R. L., Bailey, B. D., and Chabner, B. A. (1980) Proc. Natl. Acad. Sei. U. S. A . 77, 2919-2922 54. Jolivet, J., and Schilsky, R. L. (1981) Biochem. Pharmacol. 30, 1387-1390 55. Nunberg, J. H., Kaufman, R. J., Schimke, R. T., Urlaub, G., and Chasin, L. A. (1978) Proc. Natl. Acad. Sei. U. S. A . 75, 55535556 56. Dolnick, B. J., Berenson, R. J., Bertino, J . B., Kaufman, R. J., Nunberg, J. H., and Schimke, R. T. (1979) J . Cell Biol. 83,394402 57. Tyler-Smith, C., and Alderson, T. (1981) J. Mol. Biol. 153, 203218 58. Hiscott, J., Murphy, D., and Defendi, V. (1980) Cell 22,535-543 59. Morandi, C., Masters, J . N., Mottes, M., and Attardi, G. (1982) J. Mol. Biol. 156,583-607 60. Kempe, T. D., Swyryd, E. A,, Bruist, M., and Stark, G . R. (1976) Cell 9,541-550 61. Wahl, G . M., Padgett, R. A., and Stark, G. R. (1979) J. Biol. Chem. 254,8679-8689 62. Coleman, P. F., Suttle, D. P., and Stark, G. R. (1977) J. Biol. Chem. 252,6379-6385 63. Cowan, K., and Lippman, M. (1980) Proc. Am. Assoc. Cancer Res. 2 1 , 4 1 64. Mariani, B. D., Slate, D. L., and R. T. Schimke (1981) Proc. Natl. Acad. Sci. U. S. A . 78,4985-4989 65. Wiedemann, L. M., and Johnson, L. F. (1979) Proc. Natl. Acad. Sci. U. S. A . 76,2818-2822 66. Johnson, L. F., Fuhrman, C. L., and Wiedemann, L. M. (1978) J. Cell Phys. 97, 397-406 67. Kellems, R. E., Morhenn, V.B., Pfendt, E. A., Alt, F. W., and Schimke, R. T. (1979) J. Biol. Chem. 254, 309-318 68. Mayo, K. E., and Palmiter, R.D. (1981) J. Biol. Chem. 256,26212624

Estrogen Response

in DHF Reductase Gene-amplified

MCF-7 Cells

15085

Supplementary Materlal to: DIHYDROFOLATE REDUC'IASE GENE AMPLIFICATION AND POSSIBLE REARRANGEMENT I N E S T R O G E Y - R E S P O N S I V EM E T H O T R E X A T E - R E S I S T A N TH U M M BREAST CANCER CELLS

Kenneth H . C o w a n . Merrill E . Galdsmrth. Rlchard H. Levlne, S u s a n c . Artken, E i i w l n D o u g l a s s . Ne11 Clendenlnn. b t h u r W. Ylenhuis. and Marc E . Llppman

.

_.

plier. D N l i fragments were separated by electrophoresis ~n 1% agdrose slab gels a s described previously ( 4 0 ) . DNA fragments contalning DHFR sequences were ldentlfled by Southern transfer 1 4 3 ) of the DNA from agarose g e l s to n l t r ~ c e i l u l o ~filters e u s l n g a procedure descrcbed by Lawn et ai ( 4 4 1 , except that the transfer buffer was 6 x standard saline c l t r d t e ( 1 1 S S C equals 0.15 M NaCl and . 0 1 5 M S o d l u m citrate, pH 7 . 0 ) . Nlck translation was performed using reagents and cOndlt1ons supplied by Bekhesda Research Laboratories. The DNA probe used was the P e t 1-691 1 1 fragment of the plasmid pDHFR 11 which contains the 5' end and a l l Of the coding sequences of the c D N l ( 3 7 1 . Previous thermal denaturation studies have suggested a high d e g r e e Of homology between mouse and human DHFR gene sequencee 1 2 7 ) . I n fact. recomblnani hacterlophage Contalnlng human DHFR g e n e sequences have been rsalated a n d the coding ~ e q u e n u e s a r e 8 5 t o 90% hOmOloqOUS to mouse DHFR codrng sequences IJ. Chen and A . D a v l s . personal communicationl. The radlolabeled D N l i was preclpltdted I n ethanol and purlfled by chromatography over Sephadex G l O O prior to use in hybrldiratlon experiments. Follovrng hybridlratian the fkIters w e r e wa-hed four LLmeS I n 100 m l of 0.1 r S S C c o n t a i n i n g 0.18 Si>S and ".I% sodrum pyrophosphate at 55'C. RUtOradlOqraDhS were ~ r e u a r e d usinq Kodak XR5 fllm with Dupont Llghtrng p l u s Lntenslfi;r ;creens a t -SO'C. DHFR

Synthesis Studies

nrxn

MCF-7 c e l l s were qrown ~n IWEM conraining 5 % charcoal-stripped calf serum for two pas4ages and t h e n plated ~n rr~plicate i n 35 m m Llrnbra dishes at a cell density of 400,000 cells per well. T W O days later the medla were changed to serum-free IMEM. Half of the cells were xncubated wlth 1 X IO-'M estradiol. Following a n 18-hour incubation the media I n each well Were replaced w i t h two m l of methlonlne-free media Contalnlnq 100 U C l of C35S1methlonine with a speciflc Of 1175 C i / m o l ( N e w England Nuclear. B o s t o n . M A ) . After 6 hours the cells were washed with PBS, harvested by scraprnq. centrlfuged a t 1000 x g for 5 minutes, and the cell pellets frozen and stored at -20'C.

T w o mi of a cell suspenslon contalning 50,000 to 100,000 c e l l s of elther W . T . HCF-7 or MTXR MCF-7 were plated in triplrcate ~n 6-well plastic Llnbro d l s h e s 135 _n diameter) and incubated at 3 7 ° C under a humidified atmosphere cnntainlnq 5 % CO2. Twenty-four houra later the media was changed to I N E M cont a l n l n g 5% charcoal-strlpped calf serum 1 2 8 1 . 0 . 1 u f r n l of regular ~ n s u l i n ( E l i LlllY Co., Indianapolrs, I N 1 and varylng concentratlone Of mietharrexare. me charcoal treatment reduces the levels of nucleosides in the serum which are capable of rescuing cells from methotrexate toxicity and also reduces the l e v e l of Sterold hormones. Under these condltians the sensltivrty of MCF-7 cells to

methotrexate Lnhlbltlon and to e s t r o g e n 4 t i m u l a r i o n o f c e l l growth are markedly enhanced. The rnedlurn was changed on day 5 and the cella harvested on day 8 u s i n g 2 rnl of mlbecca's phosphate buffered saline (HEM Research, B e t h e s d a , Y D I contalnlng 0 . 0 2 % E D T A . The cells were diluted with Isoton I1 1 C o u l t e r ElectronLC Inc.1 and counted i n a particle counter ( C o u l r e rE l e c t r o n i c Inc.. Hlnleah, FL) Preparation of C e l l Eltracts Confluent c e l l monolayers verevashedvlth P B S I E D T A and the cells suspended L" a small volume of the same s o l u t i o n by scrap~ngwith a rubber policeman. Following centrlfugatlan at 600 x g for 10 mln, cells were resuspended ~n 1 m l o f 0 . 0 1 M T r l s . pH 7 . 5 . ,001 M E D T A , and sonicated Cvlce far 10 eec in a Brans o n SonLCator (Branson S o n i c Power Co., Danbury, CT) for 15 set. m e supernatant fluld was then used i n bOth dihydrofolate reductase actlvlty assays and r a c l i o l a b e l e d methotrexate b l n d l n g assays. Dlhydrofolate Reductase Rssay Dihydrofolate reductaee acilvlty l a 3 measured using C3Hlfolic a c i d *-cord'ng to a procedure described by Rothenberg 131) and modlfied by Nakamura and Llttlefleld ( 3 2 1 except that the incubatlon mixture was increased to 2 0 0 u 1 and the speclfxc actlvity of the [3illfolic acid was 10 m c i f m m o l . Blank v a l u e s o b i a l n e d without addltion of cell extract were subtracted from each sample. P r o t e l n was #measured by the method of Lowry uslnq bovine serum albunln as a standard ( 3 2 1 . M e t h o t r e x a t e lnhihltton rtudles were done Ysinga spectrophotometric assay for D H f R previously descrlbed (121. The assay m i x t u r e consisted of 0 . 2 "mol drhydrofolate. 0 . 2 ~ l m o lN A D P H . 0 . 1 m r n o l e potasslum phosphate, pH 6.8. 0 . 4 mmol K C I . and 1 " m o l of dithlothrertol l n a totdl volume of 1 ml. The reaction was r u n at 25'C with continuous measurement of the decrease l n absorbance at 340 nrn for 10 m l n u k e s in a double beam Beckman DU spectrophotometer. One u n l t of actlvlry 18 defined as the amount of enzyme which reduces 1 nmale of dlhydrof"1ate I" 1 minute a t 25.c.

Cells were resuspended i n 1.0 m l Of PBS and sonlcated far 20 seconda as described above. Following centrifugation at 1000 g x 15 minutes, a n aliquot of the supernatant bovine serum dlbumln was added to a flnal ConCentraLlon of 100 u q / r n l and the nurture w a s precipitated ~n 10% trlchloraceClc acld. An aliquot of the supernatant I300 "11 was mlred wlrh 600 1 ' 1 of 1.0 M potasslum phosphate buffer. pH 6.2, and 100 p l of i i n l a b e l e d rnethionlne ( 1 0 mg/ml) and passed through d me2hot=eiare-=eph~ro4eafflnlty column (0.5 x 1.0 C m ) 5 L l m e 4 . The methotrerate-sepharose was kindly provlded by D r . Bernard Kaufman. The column was then washed wlth 1 M potasslum phospllate buffer u n t i l the radioacrivlry eluted from the column was approrlmarely background. The [35Sllabele0 DHFR was then eluted with 10 m l of 1.0 W potasslum phosphate buffer conra~ning 1 mH methotrexate. m e eluate was collected ~n 2 m l fractions and allquots of each fraction were counted in a liquid s c i n t ~ l l a t l o n counter. RESULTS

The sensitivity of W.T. and MTXR WCF-7 cells to methotrexate 1s shown 1. While the growth of W . T . MCF-7 c e l l s i s markedly lnhiblted by methotrexate concentrations as low a3 4 x lo-%, t h e r e is essentlally no rnhlbition of growth of MTXR MCF-7 cells u n t l l the concentcation of methotrexate in the nedlum z e greater than 1 x IO-%. The MTX concehtrailons re ulred for 50% inhibitlon of cell growth i s Over 1000-fold greater ~n