- Email: [email protected]
Ascorbic acid (vitamin C) improves the antineoplastic activity of doxorubicin, cisplatin, and paclitaxel in human breast carcinoma cells in vitro
CAft#CE.R Cancer Letters
103 (1996)
185 I89
Ascorbic acid (vitamin C) improves the antineoplastic activity of doxorubicin, cispiatin, and paclitaxel in human breast carcinoma cells in vitro Christian M. Kurbachera-*, Uwe Wagnera, Bernd Kolsterb, Peter E. Andreottic, Dieter Krebsa, Howard W. Brucknerd
Kece~ved 8 February
1996; accepted 5 March
1996
IJtilising a microplate ATP bioluminescence assay, two human breast carcinoma cell lines, MCF-7 and MDA-MB-23 1. were testedagainst doxorubicin (DOX), cisplatin (DDP), and paclitaxel (TX) alone and in combination with ascorbic acid (Vit C). In both ceil lines, Vit C exhibited cytotoxic activity at high concentrations (i.e. 102-103~M). Both cell lines also were resistant to DOX. MCF-7 was found to be DDP-resistant. MDA-MB-231 was moderately sensitive to DDP Both cell lines were strongly sensitive to TX. Vit C both at non-cytotoxic (1 ,uM) and moderately cytotoxic concentrations (lO*/~M) improved the cytotoxicity of DOX, DDP. and TX significantly. Combination effects between Vit C and DDP or TX were partly synergistic and partly additive or subadditive whereas a consistent synergism was found between Vit C and DOX. The mechanisms by which Vit C potentiates the cytostatics studied are yet unclear and should he evaluated further.
Kr\~~~orci,c: Accorbic acid (vitamin C): ATP bioluminescence assay; Breast cancer: Cisplatin; Doxorubicin; Paclitaxel
1n Introduction ascorbic acid (vitamin C; Vit C) and a number of other substances which are not known as classical anticancer agents have been found to exhibit some ’ Corresponding author. Department of Gynecology and Ohstctrich, Unitcrsity of Cologne Medical Center. Kerpener Strasse 34. D-5093 1 Cologne. t;\x: 149 21: 478492’9
Germany.
Tel.:
:1.304-3X35/9h/4 17.00 0 1996 Eisevier !‘I!. SOI!)4-~X3S(‘)h)04211 7
+49
221 4784940/5196;
Science Ireland
degree of antineoplastic activity which seems to be associated with the generation of oxyradicals [ l-61. Correspondingly, these free radicals appear as an important determinant in the anticancer activity 01 several established cytostatics such as doxorubicin (DOX) [ 7 1. Overexpression of detoxifying enzymes like superoxide dismutase, catalase, or glutathione peroxidase may thus contribute to drug resistance protecting cells from oxidative stress [8]. Even though Vit C alone clinically failed to produce any
Ltd. All rights reserved
184
CM. Kurbuclwr ef nl./CuncerLetfers
significant antitumoral effect [9], a body of experimental work suggests that it might act as a chemomodulator potentiating the cytotoxic activity of different cytostatics [ 1l-l 31. Experimentally, MCF-7 breast cancer (BrCA) cells have been shown to exhibit some in vitro sensitivity to high concentrations of vitamin C which apparently was augmented by vitamin Ks and reversed by catalase suggesting a relationship with oxyradical formation [3]. Regarding the well recognized activity of DOX in clinical BrCA [14], it seems to be of particular interest that oxygen radical production may also be involved in DOX cytotoxicity in MCF-7 cells [IS]. However, little is known about possible interactions between cytotoxic agents and Vit C in this neoplasia. The present study using two human BrCA cell lines therefore was initiated in order to investigate the combination effects between ascorbic acid and DOX, cisplatin (DDP), and paclitaxel (TX), three agents which recently are considered among the most active for the treatment of clinical BrCA [ 141. 2. Material and methods 2.1. Breast carcinoma cell lines Estrogen receptor-positive MCF-7 cells were obtained from the tumor cell bank, Deutsches Krebsforschungszentrum (Heidelberg, Germany). Estrogen receptor-negative MDA-MB-231 cells were a kind gift from H. Arps (Institute of Pathology, Stadtisches Klinikum, Fulda, Germany). Cell lines were maintained in cell culture medium (CCM), RPM1 1640 supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 10 pug/ml insulin, 100 U/ml penicillin and 100 mg/ml streptomycin (all media and reagents from Gibco/BRL, Paisley, UK) at 37°C 5% CO* and 100% humidity. Cultures were periodically controlled for mycoplasma infection using 33528 dye (Hoechst, Frankfurt, Germany) and passaged once or twice a week at a I:2 ratio. 2.2. Drugs For this study, commercial formulations of DOX (Adriblastin@, Farmitalia Carlo Erba, Milan, Italy), DDP (Platinex@, Bristol-Myers Squibb, Munich, Germany), TX (Taxol@, Bristol-Myers Squibb, Mu-
IO.? (1996) 183-189
nich, Germany) and ascorbic acid (Vitamin CInjectopas@, Pascoe, Giessen, Germany) were used. Test concentrations were freshly prepared for each experiment. Single agents were tested at 1:lO dilutions with 10-4-10flM (DOX), 10e3-lo2 ,uM (DDP, TX) and 10-2-103 PM (Vit C). Clinically relevant cytostatic concentrations (i.e. l-100% peak plasma concentration) were represented by 1O-2-1 PM (DOX) and lo-‘--10pM (DDP, TX), respectively [16]. All antineoplastics were assayed at the above mentioned concentrations combined with fixed concentrations of Vit C (1 PM and lOOpM), as well. Additionally, Vit C at increasing concentrations was tested with fixed concentrations of DOX (lo-’ PM, IOpM), DDP (O.lpM, 102pM) and TX (O.lpM, lo2 PM). All drug concentrations and combinations were tested in triplicate. 2.3. Chemosensitivity assay In vitro chemosensitivity was determined by a microplate ATP tumor bioluminescence assay (ATPTCA) utilizing a standardized kit technique (TCAloo@, DCS, Hamburg, Germany) which was slightly modified in order to test permanent cell lines as previously described in detail [17]. In brief, drugs or drug combinations and cells were pipetted into 96-well microplates (Falcon; Becton & Dickinson, Heidelberg, Germany) with lo4 cells per well. Two controls, one with CCM (no inhibition; MO) and the other with maximum ATP inhibitor (maximum inhibition; MI) were added to six wells of each culture plate. Cultures then were incubated for 5 days at 37°C in a 95% air, 5% CO, atmosphere with 100% humidity. Each experiment was performed twice. 2.4. Assay evaluation At the end of incubation time, cellular ATP was extracted and stabilized by adding 50 ~1 tumor cell extraction reagent (TCER) to each well of the culture plates. Consecutively, ATP was measured in a LB 953-luminometer (Berthold, Wildbad, Germany) after automatically pipetting 55 pi of luciferinluciferase (Lu-Lu) into 50~1 of each cell lysate. Luminescence was measured as relative light units (RLU = photons/lo). The individual survival fraction
for each concentration of drug/drug (SF.,-,,,) was determined as:
combination
four categories of combined drug effects [ 171: (1) Supra-additivity (synergism): experimental survival < expected survival (2) Additivity: experimental survival = expected survival (3) Sub-additivity: experimental survival > expected survival (4) Antagonism: experimental survival > experimental survival for the best single agent.
SF,+,, = (RI,IJ.,,,\, - RLUM,)/(RLU~e - KIN,& Data then were graphically transformed into individual survival curves for each cell line and drug/drug combination, respectively. In order to estimate the combination effect between Vit C and the expected dose-response cytostatics, theoretically curves were constructed by multiplying the SF of the drug at v;niable concentrations with the SF of the drug added at a fixed concentration according to the following equation:
3. Results 3. I. Single
agent
acrivitJ
Both cell lines were not strongly sensitive to DOX and DDP. A significant SF reduction occurred only at relatively high concentrations exceeding 0.1 /dM (0.058pg/ml) for DOX (Fig. l&C), and 1,uM ((I.3 pug/ml) or 1OpM (3 pglml) for DDP (Fig. 2h,C).
l~oth theoretical and experimental curves for the combinations then were compared in order to define C
Doxorubicin
Concentration
DM]
1‘1s. i Combination effects between DOX and Vit C in MCF-7 (A and 13) and in MDA-MB-231.cells (C and I)). (A and (C) show results !W IlOX :I! variable concentrations combined with fixed concentrations of Vit C. (B) and (I>) show the reverse experiment. Dashed lines ir.~ithsmall symbols represent the theoretically expected curves assuming that DOX and Vit C act additively. Since 1 PM of Vit C did not produce an? redaction of SF in MDA-MB-231. the theoretical curve for the combination is identical with the dose-response curve for DOX .&me.
C.M. Kurhuchrr er 01. /Cunceri~~e~.s
186 A
t DDP - DDP+ I$.4“11c tDDP + 1oopM”rtC
loo= .L ?
60-
,j E
60
E II!
.0.
20
loo-
; s c
#
,,,,,s
0 10’
10’
SO-
60
E
.
‘lo
C
;
... 9
103 (1996) 183%189
__l
40 20 #
,,,,:y -...- . . . .
10’ Cisplatin
10’
10e
10’
Concentration
[rtvl]
10’
10'
0 10’
10 2
10’ Cisplatin
. .. ...L_ 10 1
10’
Concentration
..--.-.- . . ....__..10'
10’
IO’
10’
10’
WM]
20. 07 ,--I..., ..,..,.I ,,,-,,I ,,,....I ‘---.-I . 100 IO’ IO’ 10’ 10’ 10‘ Vtatin
C Concentration
IO'
IO'
CM]
10’
102
10'
viinin
100
IO'
c Concentration
IO'
$M]
Fig. 2. Combination effects between DDP and Vit C in MCF-7 (A and B) and in MDA-MB-23kells (C and D). (A) and (C) show results for DDP at variable concentrations combined with fixed concentrations of Vit C. (B) and (D) show the reverse experiment. Dashed lines with small symbols represent the theoretically expected curves assuming that DDP and Vit C act additively. Since 1 FM of Vit C did not produce any reduction of SF in MDA-MB-231, the theoretical curve for the combination is identical with the dose-response curve for DDP alone.
In contrast, the Tx-sensitivity of both cell lines was markedly higher with all concentrations producing >50% SF reduction (Fig. 3A,C). Vit C produced a weak antineoplastic activity against both MCF-7 and MDA-MB-231 with >50% reduction of SF at 100,~M (17.6pg/ml) (Figs. l-3B, l-3D). In all experiments, 1 ,uM of Vit C failed to exhibit any cytotoxicity. The SF at 100pM in contrast was 51.3% for MCF-7 and 17% for MDA-MB-23 1, respectively. 3.2.
Combined
drug
effects
Results for the combination of DOX and Vit C are presented in Fig. IA-D. For both cell lines a strong synergistic action was apparent regardless if either Vit C or DOX was used at fixed dosages. Whereas both the non-toxic (1 PM) and the moderately cytotoxic concentration (IOOpM) were nearly equal to potentiate DOX in MCF-7 (Fig. lA), a clear dose-
dependence could be found in MDA-MB-231 (Fig 1C). In Fig. 2A-D, results for Vit C combined with DDP are illustrated. In MCF-7 cells, findings were generally comparable to those with DOX showing a marked synergism between DDP and 1 ,uM or 100,uM of Vit C without striking differences between the higher and the lower dosage (Fig. 2A). When Vit C at varying dosages was combined with fixed DDP concentrations, combination effects were at best additive, mainly related to the low activity of DDP at both 0.1 PM and 10 PM (Fig 2B). In MDAMB-23 1 cells, only 100 PM Vit C was able to potentiate DDP (Fig. 2C). Correspondingly, DDP at 10 PM showed synergistic effects with Vit C, whereas the combination with 0.1 ,uM even was found to be subadditive at lower Vit C concentrations (Fig. 2D). Results for Vit C combined with TX are shown in Fig. 3A-D. As mentioned above, both cell lines ex-
Paclitaxel
Concentration
[PM]
Fig. 3. Combination effects between TX and Vit C in MCF-7 (A and t3) and in MDA-MB23kells (C and D). (A) and (C) show results for TX at variable concentrations combined with fixed concentrationsof Vit C. (B) and (D) show the reverse experiment. Dashed lines with ~n1:~11 symbols reprcsent the theoretically expected curves assuming that TX and Vit C act additively. Since 1 ,uM of Vit C did not produce ;my reduction of SF in MDA-MH-231, the theoretical curve for the combination is identical with the dose-responsecurve for TX alone.
hibited
marked
sensitivity
to single
agent TX (Fig.
IA,C). in MCF-7, both stable concentrations
of Vit C
failed to improve TX significantly (Fig. 3A). Accordingly, the reverse experiment provided at best additive effects between Vit C and TX at 0.1 PM and
IOpM (Fig. 3B). MDA-MB-231 cells, however, expressed a somewhat different reaction pattern to TX plus Vit c‘. The high single agent activity of TX was even potentiated by Vit C in a dose-dependent manner with synergism (1 PM Vit C) or at least additive effects (100pM Vit C) between both substances (Fig. 3C). A corresponding result could also he achieved in the reverse experiment (Fig. 3D). 4. Discussion For almost two decades, ascorbic acid was proposed for cancer treatment and prevention [ 18-211. Even though the intrinsic antineoplastic activity of
Vit C observed clinically was disappointing [9], a body of experimental data arose suggesting Vit C to improve the cytotoxicity of different antineoplastic agents [lo-131. This effect was further improved by addition of vitamin Ks [3,22-241. Vit C induced potentiation of cytostatic drug effects thus is likely to he related to generation of oxyradicals because both vitamins are known to interact with enhanced formation of semiquinone
and dehydroascorbate
as poten-
tially sources of free radical production [25]. Accordingly, the cytotoxic potentiation induced by Vit C or Vit C plus vitamin K3 was completely reversed by adding catalase [3,23]. Interestingly, alterations in oxygen free radical susceptibility might be partly responsible for DOX-resistance occurring in human MCF-7 BrCA cells 126,271. Other mechanisms such as alteration of intracellular drug accumulation, however, might also he important for the chemomodulating action of ascorbic acid, since it was found to im-
188
CM. Kurbacher et al. /Cancer Letters 103 (1996) 183-189
prove cytostatics like vinca alkaloids which are not known to act primarily on DNA [ 12,22-241. Until today, the chemomodulating properties of Vit C are only poorly understood and have not been elucidated profoundly, particularly not in common neoplasms like breast cancer. In the experiments described here, we found Vit C both at non-toxic and at cytotoxic concentrations to produce a consistently synergistic antineoplastic activity with DOX in human BrCA cells. As mentioned above, this effect might be related to the generation of free radicals which have been shown to have an impact on DOX-induced cell kill in MCF-7 cells [ !5,26,27]. Since DOX is a key substance for the treatment of clinical BrCA, the combination with Vit C seems to be of particular interest inasmuch as ascorbic acid and its derivatives have been found to reduce anthracycline induced cardiotoxicity which, as one of its most important adverse effects, can compromise the clinical use of DOX [28]. Our results with Vit C and DDP generally are confirmative of observations which have been made with other malignancies in vitro and in vivo suggesting a potentiating effect when both drugs are combined [ 11,13,23, 241. Like for anthracyclines, formation of oxyradicals appears to mediate cytotoxicity of platinum compounds [29]. However, our findings in MDA-MB231, in which DDP and Vit C produced synergistic cytotoxicity solely when the latter was added at cytotoxic concentrations, might indicate that mechanisms other than those mediated by oxyradicals are more important for DDP-activity compared to DOX, at least in some breast tumors. Nevertheless, DDP plus Vit C could be another clinically feasible combination, since ascorbic acid might be able to reduce DDP-induced injury of renal cells [ 301. To our knowledge, no report exists about the combined activity of paclitaxel and Vit C neither in BrCA nor in any other tumor entity. Even though TX was markedly more active than DOX or DDP in both cell lines studied, we could document a significant potentiation of TX activity induced by Vit C. Currently, TX is not known to produce a direct DNA damage. In contrast, TX exhibits its major activity by stabilizing microtubule polymers, preventing them against depolymerization and thus inducing a block in the G2/M phase of the cell cycle [31]. As for other natural products, overexpression of p-glycoprotein
@gp) is one of the major mechanisms contributing to TX-resistance [31, 321. Interestingly, Vit C has been found to potentiate vinca alkaioids which are also influenced by pgp presumably by increasing cellular drug accumulation [12]. It is not clear at present, however, whether this effect is pgp-related and whether the activity of TX is modulated by the same or a yet unknown mechanism. In conclusion, we were able to demonstrate that ascorbic acid is likely to potentiate three of the most active drugs for the treatment of BrCA. Combination effects mostly were synergistic or at least additive. The mechanism by which Vit C is able to improve the cytostatics studied is not known at present and should be elucidated in further investigations. Due to the low toxicity of Vit C even at very high concentrations, combinations of ascorbic acid with cisplatin, doxorubicin, or paclitaxel seem to be attractive for the future treatment of breast cancer. Acknowledgements This work was supported, in part, by DCS Innovative Diagnostik Systeme, Hamburg, Germany and by Pascoe, Giessen, Germany. We wish to thank H. Hiibner and R. Klasen for their skillful technicai assistance. References [II Park,C.H., Amare, M., Savin, M.A. and Hoogstraten,B. (1980)Growthsuppression of humanleukemiccells in vitro by L-ascorbicacid.CancerRes.,40, 1062-1065. PI Bram,S., Froussard,P., Guichard,M., Jasmin,C., Augery, Y., Sinoussi-Barre, F. andWray,W. (1981)Vitamin C preferentialtoxicity for malignantmelanomacells. Nature, 284, 629-631.
Noto,V.. Taper,H.S..Yi-Hua,J., Janssens, I., Bonte, J. and De Loecker. W. (1989) Effects of sodium ascorbate (vitamin C) and 2-methyl-1,4-naphthoquinone (vitamin KS) treatment on humantumorcell growthin vitro. I. Synergismof combined vitamin C and Kg action. Cancer, 63,901-906. [41 Begleiter, A (1983) Cytocidal action of the quinone group and its relationship to antitumor activity. Cancer Res., 43, 481484. 151Powis, G. (1989) Freeradical formationby antitumorqui-
[31
nones.FreeRad.Biol. Med.,6,63-101. [61 Lee, Y.S. and Wurster, R.D. (1995) Methylene blue induces PI
cytotoxicity in human brain tumor cells. Cancer Lett., 88, 141-145. Lawn, J.W. (1985) Molecular mechanisms of action of
141
i9j
! lOj
i I!
IZi
131
anticancer agents involvrng f’ree radical intermediates. Free I
14 1 Haskcll. (‘.M. (1995) Systemic treatment for metastatic breast cancer. In: Cancer Treatment, pp. 354-368. Editor: i’ M. linskell. W.B. Saunders, Philadelphia. PA. 151 Sinha. H.K., Katkl, A.G., Hatist, G. Cowan, K H. and My,‘{\. (‘ I:. ( 1987) Adriamycin-stimulated hydroxyl radical fOVJlJ;l!iOli in human breast tumor cells. Hiochem. PharmaLOi 36. 733.--796. Ih] Albcrlh, l1.S. (1080) Tabular summary of pharmacokinertc parameters relevant to in vitro drug assay In: Cloning of Hum;rrt Tumor Stem (‘ells. Editor: S.E. Salmon. Alan I< Lish, New York 171 Kurbacher. CM., Bruckner. l1.W.. Andrcotti, P.E.. Kurha?hcr. .I.,~.. Sass, G. antI Krchs, D. (I Y9.5) In vitro activity ol tilanoccnedichloride versus cisplatin in four ovarian carc1noma ccl1 lines evaluated hy a microtiter plate ATY hioluminccccnts assay. Anti-Cancer Drugs, 6, 6Y7- 704. 181 Cameron. I-. and Pauling, L. ( 1976) Suppicmentni ascorhate m the supportive trcatmcnt of cancer: prolongation of sur.+lvnl times in terminal human cancer. Proc. Nat]. Acad Sci. IJSA. 73.3685 -3689. I!){ Marquardt, 11.. Rufling, 1.. and We&burger, J.11. (IY77i Mutagenic activity of mtritc treated foods. hurnan stomach t‘ancer ma> he related to dietary factors. Science. 196. 1000 lOOi
1201 Khan, P.K. and Sinha, S.P. (1993) Antimutagenic efficacy of higher doses of vitamin C. Mutat. Res., 298, 157-I 61. 12 11 Jain, M., Miller, A.B. and To, T. (1994) Premorbid diet and the prognosis of women with breast cancer. J. Natl. Cancer Inst., 86. 1390.-1397. [22] Taper, H.S., De Gerlache, J., Lans. M. and Roberfroid, M. (1987) Son-toxic potentiation of cancer chemotherapy by combined C and KY vitamin pretreatment. Int. .I. Cancer. 40, 57.5-579. 1231 Taper, H.S. and Koberfroid, M. (1992) Ken-toxic sensitization of cancer chemotherapy by combined vitamin C and K3 pretreatment in a mouse tumor resistant to oncovin. Anticancer Res., 13, 1651-1654. [24] De Loeckcr, W., Janssens, J., Bonte, J. and Taper. H.S. (1993) Effects of sodium ascorhate (vitamin C) and 2. methyl- I ,4+aphthoquinone (vitamin K3) treatment on human cell growth in vitro. II. Synergism with combined chemotherapy action. Anticancer Res., 13, 103 -104. 1251 Pething, R., Gascoync, P.R.C.. ,%lcLaughlin, J.A. and SzentGyorgyi, A. ( 1983) Ascorbate-quinonc interactions: electrochemical, free radical and cytotoxic properties. Proc. Natl. Acad. Sci. USA, 80, 129. 132. 1261 Doroshow, J. (lY86) Prevention of doxorubicin-induced killing of MCF-7 human breast cancer cells by oxygen radical scavengers and iron chelating agents. Biochem. Biophys. Kes. Commun., 135, 330-335. [271 Mimnaugh, E.ti., l)usre, I... Atwell, J. and Myers, C.E. 11989) Dirfcrential oxygen radical susceptibility of adriamycin-sensitive and -resistant MCF-7 human hreast tumor cells. Cancer Rcs., 49, 8-1.5 1281 Shimpo, K., Nagatsu, T., Yamada, K., Sate, T.. Niimi, H., Shamoto, M.. Takeuchi, T.. Umeznwa, II. and Fujita, K. (1991) Ascorbic acid and adriamycin toxicity. Am J Clin. ICulr., 54, 12YXS-13OlS. 1291 Tonctti, M.. Giovine, M., Gasparini, A.. Bcnatti, U. and De FIoJ-:I, A. (1993) Enhanced formation of reactive species from cis-diammine-( I, I -cyclobutanedicarboxylato)-platinum(Il) (carboplatin) m the presence of oxygen free radicals. Hiochem. Pharmacol, 46. 1377-1383. (301 Gemha, M. and Fukuishi, N. (I99 I) Amelioration by ascorbic acid of cisplatin-induced injury in cultured renal epithclial cells. Contrih. Nephrol., 9.5, 138-142. 13 I ] Spencer, C.M. and Faulds, D. (I 9Y4) Paclitaxel. A review of its pharmacodynamic and pharmacokinctic properties and therapeutic potential in the treatment of cancer. Drugs, 4X. 794 x47 1321 Roy, S.?. and llorwitz, S.B. (IYXS) A phosphoglycoprotein associated with Taxol-resistance in J774.2 cells. Cancer Kes.. 4s. 3856 3863.
103 (1996)
185 I89
Ascorbic acid (vitamin C) improves the antineoplastic activity of doxorubicin, cispiatin, and paclitaxel in human breast carcinoma cells in vitro Christian M. Kurbachera-*, Uwe Wagnera, Bernd Kolsterb, Peter E. Andreottic, Dieter Krebsa, Howard W. Brucknerd
Kece~ved 8 February
1996; accepted 5 March
1996
IJtilising a microplate ATP bioluminescence assay, two human breast carcinoma cell lines, MCF-7 and MDA-MB-23 1. were testedagainst doxorubicin (DOX), cisplatin (DDP), and paclitaxel (TX) alone and in combination with ascorbic acid (Vit C). In both ceil lines, Vit C exhibited cytotoxic activity at high concentrations (i.e. 102-103~M). Both cell lines also were resistant to DOX. MCF-7 was found to be DDP-resistant. MDA-MB-231 was moderately sensitive to DDP Both cell lines were strongly sensitive to TX. Vit C both at non-cytotoxic (1 ,uM) and moderately cytotoxic concentrations (lO*/~M) improved the cytotoxicity of DOX, DDP. and TX significantly. Combination effects between Vit C and DDP or TX were partly synergistic and partly additive or subadditive whereas a consistent synergism was found between Vit C and DOX. The mechanisms by which Vit C potentiates the cytostatics studied are yet unclear and should he evaluated further.
Kr\~~~orci,c: Accorbic acid (vitamin C): ATP bioluminescence assay; Breast cancer: Cisplatin; Doxorubicin; Paclitaxel
1n Introduction ascorbic acid (vitamin C; Vit C) and a number of other substances which are not known as classical anticancer agents have been found to exhibit some ’ Corresponding author. Department of Gynecology and Ohstctrich, Unitcrsity of Cologne Medical Center. Kerpener Strasse 34. D-5093 1 Cologne. t;\x: 149 21: 478492’9
Germany.
Tel.:
:1.304-3X35/9h/4 17.00 0 1996 Eisevier !‘I!. SOI!)4-~X3S(‘)h)04211 7
+49
221 4784940/5196;
Science Ireland
degree of antineoplastic activity which seems to be associated with the generation of oxyradicals [ l-61. Correspondingly, these free radicals appear as an important determinant in the anticancer activity 01 several established cytostatics such as doxorubicin (DOX) [ 7 1. Overexpression of detoxifying enzymes like superoxide dismutase, catalase, or glutathione peroxidase may thus contribute to drug resistance protecting cells from oxidative stress [8]. Even though Vit C alone clinically failed to produce any
Ltd. All rights reserved
184
CM. Kurbuclwr ef nl./CuncerLetfers
significant antitumoral effect [9], a body of experimental work suggests that it might act as a chemomodulator potentiating the cytotoxic activity of different cytostatics [ 1l-l 31. Experimentally, MCF-7 breast cancer (BrCA) cells have been shown to exhibit some in vitro sensitivity to high concentrations of vitamin C which apparently was augmented by vitamin Ks and reversed by catalase suggesting a relationship with oxyradical formation [3]. Regarding the well recognized activity of DOX in clinical BrCA [14], it seems to be of particular interest that oxygen radical production may also be involved in DOX cytotoxicity in MCF-7 cells [IS]. However, little is known about possible interactions between cytotoxic agents and Vit C in this neoplasia. The present study using two human BrCA cell lines therefore was initiated in order to investigate the combination effects between ascorbic acid and DOX, cisplatin (DDP), and paclitaxel (TX), three agents which recently are considered among the most active for the treatment of clinical BrCA [ 141. 2. Material and methods 2.1. Breast carcinoma cell lines Estrogen receptor-positive MCF-7 cells were obtained from the tumor cell bank, Deutsches Krebsforschungszentrum (Heidelberg, Germany). Estrogen receptor-negative MDA-MB-231 cells were a kind gift from H. Arps (Institute of Pathology, Stadtisches Klinikum, Fulda, Germany). Cell lines were maintained in cell culture medium (CCM), RPM1 1640 supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine, 10 pug/ml insulin, 100 U/ml penicillin and 100 mg/ml streptomycin (all media and reagents from Gibco/BRL, Paisley, UK) at 37°C 5% CO* and 100% humidity. Cultures were periodically controlled for mycoplasma infection using 33528 dye (Hoechst, Frankfurt, Germany) and passaged once or twice a week at a I:2 ratio. 2.2. Drugs For this study, commercial formulations of DOX (Adriblastin@, Farmitalia Carlo Erba, Milan, Italy), DDP (Platinex@, Bristol-Myers Squibb, Munich, Germany), TX (Taxol@, Bristol-Myers Squibb, Mu-
IO.? (1996) 183-189
nich, Germany) and ascorbic acid (Vitamin CInjectopas@, Pascoe, Giessen, Germany) were used. Test concentrations were freshly prepared for each experiment. Single agents were tested at 1:lO dilutions with 10-4-10flM (DOX), 10e3-lo2 ,uM (DDP, TX) and 10-2-103 PM (Vit C). Clinically relevant cytostatic concentrations (i.e. l-100% peak plasma concentration) were represented by 1O-2-1 PM (DOX) and lo-‘--10pM (DDP, TX), respectively [16]. All antineoplastics were assayed at the above mentioned concentrations combined with fixed concentrations of Vit C (1 PM and lOOpM), as well. Additionally, Vit C at increasing concentrations was tested with fixed concentrations of DOX (lo-’ PM, IOpM), DDP (O.lpM, 102pM) and TX (O.lpM, lo2 PM). All drug concentrations and combinations were tested in triplicate. 2.3. Chemosensitivity assay In vitro chemosensitivity was determined by a microplate ATP tumor bioluminescence assay (ATPTCA) utilizing a standardized kit technique (TCAloo@, DCS, Hamburg, Germany) which was slightly modified in order to test permanent cell lines as previously described in detail [17]. In brief, drugs or drug combinations and cells were pipetted into 96-well microplates (Falcon; Becton & Dickinson, Heidelberg, Germany) with lo4 cells per well. Two controls, one with CCM (no inhibition; MO) and the other with maximum ATP inhibitor (maximum inhibition; MI) were added to six wells of each culture plate. Cultures then were incubated for 5 days at 37°C in a 95% air, 5% CO, atmosphere with 100% humidity. Each experiment was performed twice. 2.4. Assay evaluation At the end of incubation time, cellular ATP was extracted and stabilized by adding 50 ~1 tumor cell extraction reagent (TCER) to each well of the culture plates. Consecutively, ATP was measured in a LB 953-luminometer (Berthold, Wildbad, Germany) after automatically pipetting 55 pi of luciferinluciferase (Lu-Lu) into 50~1 of each cell lysate. Luminescence was measured as relative light units (RLU = photons/lo). The individual survival fraction
for each concentration of drug/drug (SF.,-,,,) was determined as:
combination
four categories of combined drug effects [ 171: (1) Supra-additivity (synergism): experimental survival < expected survival (2) Additivity: experimental survival = expected survival (3) Sub-additivity: experimental survival > expected survival (4) Antagonism: experimental survival > experimental survival for the best single agent.
SF,+,, = (RI,IJ.,,,\, - RLUM,)/(RLU~e - KIN,& Data then were graphically transformed into individual survival curves for each cell line and drug/drug combination, respectively. In order to estimate the combination effect between Vit C and the expected dose-response cytostatics, theoretically curves were constructed by multiplying the SF of the drug at v;niable concentrations with the SF of the drug added at a fixed concentration according to the following equation:
3. Results 3. I. Single
agent
acrivitJ
Both cell lines were not strongly sensitive to DOX and DDP. A significant SF reduction occurred only at relatively high concentrations exceeding 0.1 /dM (0.058pg/ml) for DOX (Fig. l&C), and 1,uM ((I.3 pug/ml) or 1OpM (3 pglml) for DDP (Fig. 2h,C).
l~oth theoretical and experimental curves for the combinations then were compared in order to define C
Doxorubicin
Concentration
DM]
1‘1s. i Combination effects between DOX and Vit C in MCF-7 (A and 13) and in MDA-MB-231.cells (C and I)). (A and (C) show results !W IlOX :I! variable concentrations combined with fixed concentrations of Vit C. (B) and (I>) show the reverse experiment. Dashed lines ir.~ithsmall symbols represent the theoretically expected curves assuming that DOX and Vit C act additively. Since 1 PM of Vit C did not produce an? redaction of SF in MDA-MB-231. the theoretical curve for the combination is identical with the dose-response curve for DOX .&me.
C.M. Kurhuchrr er 01. /Cunceri~~e~.s
186 A
t DDP - DDP+ I$.4“11c tDDP + 1oopM”rtC
loo= .L ?
60-
,j E
60
E II!
.0.
20
loo-
; s c
#
,,,,,s
0 10’
10’
SO-
60
E
.
‘lo
C
;
... 9
103 (1996) 183%189
__l
40 20 #
,,,,:y -...- . . . .
10’ Cisplatin
10’
10e
10’
Concentration
[rtvl]
10’
10'
0 10’
10 2
10’ Cisplatin
. .. ...L_ 10 1
10’
Concentration
..--.-.- . . ....__..10'
10’
IO’
10’
10’
WM]
20. 07 ,--I..., ..,..,.I ,,,-,,I ,,,....I ‘---.-I . 100 IO’ IO’ 10’ 10’ 10‘ Vtatin
C Concentration
IO'
IO'
CM]
10’
102
10'
viinin
100
IO'
c Concentration
IO'
$M]
Fig. 2. Combination effects between DDP and Vit C in MCF-7 (A and B) and in MDA-MB-23kells (C and D). (A) and (C) show results for DDP at variable concentrations combined with fixed concentrations of Vit C. (B) and (D) show the reverse experiment. Dashed lines with small symbols represent the theoretically expected curves assuming that DDP and Vit C act additively. Since 1 FM of Vit C did not produce any reduction of SF in MDA-MB-231, the theoretical curve for the combination is identical with the dose-response curve for DDP alone.
In contrast, the Tx-sensitivity of both cell lines was markedly higher with all concentrations producing >50% SF reduction (Fig. 3A,C). Vit C produced a weak antineoplastic activity against both MCF-7 and MDA-MB-231 with >50% reduction of SF at 100,~M (17.6pg/ml) (Figs. l-3B, l-3D). In all experiments, 1 ,uM of Vit C failed to exhibit any cytotoxicity. The SF at 100pM in contrast was 51.3% for MCF-7 and 17% for MDA-MB-23 1, respectively. 3.2.
Combined
drug
effects
Results for the combination of DOX and Vit C are presented in Fig. IA-D. For both cell lines a strong synergistic action was apparent regardless if either Vit C or DOX was used at fixed dosages. Whereas both the non-toxic (1 PM) and the moderately cytotoxic concentration (IOOpM) were nearly equal to potentiate DOX in MCF-7 (Fig. lA), a clear dose-
dependence could be found in MDA-MB-231 (Fig 1C). In Fig. 2A-D, results for Vit C combined with DDP are illustrated. In MCF-7 cells, findings were generally comparable to those with DOX showing a marked synergism between DDP and 1 ,uM or 100,uM of Vit C without striking differences between the higher and the lower dosage (Fig. 2A). When Vit C at varying dosages was combined with fixed DDP concentrations, combination effects were at best additive, mainly related to the low activity of DDP at both 0.1 PM and 10 PM (Fig 2B). In MDAMB-23 1 cells, only 100 PM Vit C was able to potentiate DDP (Fig. 2C). Correspondingly, DDP at 10 PM showed synergistic effects with Vit C, whereas the combination with 0.1 ,uM even was found to be subadditive at lower Vit C concentrations (Fig. 2D). Results for Vit C combined with TX are shown in Fig. 3A-D. As mentioned above, both cell lines ex-
Paclitaxel
Concentration
[PM]
Fig. 3. Combination effects between TX and Vit C in MCF-7 (A and t3) and in MDA-MB23kells (C and D). (A) and (C) show results for TX at variable concentrations combined with fixed concentrationsof Vit C. (B) and (D) show the reverse experiment. Dashed lines with ~n1:~11 symbols reprcsent the theoretically expected curves assuming that TX and Vit C act additively. Since 1 ,uM of Vit C did not produce ;my reduction of SF in MDA-MH-231, the theoretical curve for the combination is identical with the dose-responsecurve for TX alone.
hibited
marked
sensitivity
to single
agent TX (Fig.
IA,C). in MCF-7, both stable concentrations
of Vit C
failed to improve TX significantly (Fig. 3A). Accordingly, the reverse experiment provided at best additive effects between Vit C and TX at 0.1 PM and
IOpM (Fig. 3B). MDA-MB-231 cells, however, expressed a somewhat different reaction pattern to TX plus Vit c‘. The high single agent activity of TX was even potentiated by Vit C in a dose-dependent manner with synergism (1 PM Vit C) or at least additive effects (100pM Vit C) between both substances (Fig. 3C). A corresponding result could also he achieved in the reverse experiment (Fig. 3D). 4. Discussion For almost two decades, ascorbic acid was proposed for cancer treatment and prevention [ 18-211. Even though the intrinsic antineoplastic activity of
Vit C observed clinically was disappointing [9], a body of experimental data arose suggesting Vit C to improve the cytotoxicity of different antineoplastic agents [lo-131. This effect was further improved by addition of vitamin Ks [3,22-241. Vit C induced potentiation of cytostatic drug effects thus is likely to he related to generation of oxyradicals because both vitamins are known to interact with enhanced formation of semiquinone
and dehydroascorbate
as poten-
tially sources of free radical production [25]. Accordingly, the cytotoxic potentiation induced by Vit C or Vit C plus vitamin K3 was completely reversed by adding catalase [3,23]. Interestingly, alterations in oxygen free radical susceptibility might be partly responsible for DOX-resistance occurring in human MCF-7 BrCA cells 126,271. Other mechanisms such as alteration of intracellular drug accumulation, however, might also he important for the chemomodulating action of ascorbic acid, since it was found to im-
188
CM. Kurbacher et al. /Cancer Letters 103 (1996) 183-189
prove cytostatics like vinca alkaloids which are not known to act primarily on DNA [ 12,22-241. Until today, the chemomodulating properties of Vit C are only poorly understood and have not been elucidated profoundly, particularly not in common neoplasms like breast cancer. In the experiments described here, we found Vit C both at non-toxic and at cytotoxic concentrations to produce a consistently synergistic antineoplastic activity with DOX in human BrCA cells. As mentioned above, this effect might be related to the generation of free radicals which have been shown to have an impact on DOX-induced cell kill in MCF-7 cells [ !5,26,27]. Since DOX is a key substance for the treatment of clinical BrCA, the combination with Vit C seems to be of particular interest inasmuch as ascorbic acid and its derivatives have been found to reduce anthracycline induced cardiotoxicity which, as one of its most important adverse effects, can compromise the clinical use of DOX [28]. Our results with Vit C and DDP generally are confirmative of observations which have been made with other malignancies in vitro and in vivo suggesting a potentiating effect when both drugs are combined [ 11,13,23, 241. Like for anthracyclines, formation of oxyradicals appears to mediate cytotoxicity of platinum compounds [29]. However, our findings in MDA-MB231, in which DDP and Vit C produced synergistic cytotoxicity solely when the latter was added at cytotoxic concentrations, might indicate that mechanisms other than those mediated by oxyradicals are more important for DDP-activity compared to DOX, at least in some breast tumors. Nevertheless, DDP plus Vit C could be another clinically feasible combination, since ascorbic acid might be able to reduce DDP-induced injury of renal cells [ 301. To our knowledge, no report exists about the combined activity of paclitaxel and Vit C neither in BrCA nor in any other tumor entity. Even though TX was markedly more active than DOX or DDP in both cell lines studied, we could document a significant potentiation of TX activity induced by Vit C. Currently, TX is not known to produce a direct DNA damage. In contrast, TX exhibits its major activity by stabilizing microtubule polymers, preventing them against depolymerization and thus inducing a block in the G2/M phase of the cell cycle [31]. As for other natural products, overexpression of p-glycoprotein
@gp) is one of the major mechanisms contributing to TX-resistance [31, 321. Interestingly, Vit C has been found to potentiate vinca alkaioids which are also influenced by pgp presumably by increasing cellular drug accumulation [12]. It is not clear at present, however, whether this effect is pgp-related and whether the activity of TX is modulated by the same or a yet unknown mechanism. In conclusion, we were able to demonstrate that ascorbic acid is likely to potentiate three of the most active drugs for the treatment of BrCA. Combination effects mostly were synergistic or at least additive. The mechanism by which Vit C is able to improve the cytostatics studied is not known at present and should be elucidated in further investigations. Due to the low toxicity of Vit C even at very high concentrations, combinations of ascorbic acid with cisplatin, doxorubicin, or paclitaxel seem to be attractive for the future treatment of breast cancer. Acknowledgements This work was supported, in part, by DCS Innovative Diagnostik Systeme, Hamburg, Germany and by Pascoe, Giessen, Germany. We wish to thank H. Hiibner and R. Klasen for their skillful technicai assistance. References [II Park,C.H., Amare, M., Savin, M.A. and Hoogstraten,B. (1980)Growthsuppression of humanleukemiccells in vitro by L-ascorbicacid.CancerRes.,40, 1062-1065. PI Bram,S., Froussard,P., Guichard,M., Jasmin,C., Augery, Y., Sinoussi-Barre, F. andWray,W. (1981)Vitamin C preferentialtoxicity for malignantmelanomacells. Nature, 284, 629-631.
Noto,V.. Taper,H.S..Yi-Hua,J., Janssens, I., Bonte, J. and De Loecker. W. (1989) Effects of sodium ascorbate (vitamin C) and 2-methyl-1,4-naphthoquinone (vitamin KS) treatment on humantumorcell growthin vitro. I. Synergismof combined vitamin C and Kg action. Cancer, 63,901-906. [41 Begleiter, A (1983) Cytocidal action of the quinone group and its relationship to antitumor activity. Cancer Res., 43, 481484. 151Powis, G. (1989) Freeradical formationby antitumorqui-
[31
nones.FreeRad.Biol. Med.,6,63-101. [61 Lee, Y.S. and Wurster, R.D. (1995) Methylene blue induces PI
cytotoxicity in human brain tumor cells. Cancer Lett., 88, 141-145. Lawn, J.W. (1985) Molecular mechanisms of action of
141
i9j
! lOj
i I!
IZi
131
anticancer agents involvrng f’ree radical intermediates. Free I
14 1 Haskcll. (‘.M. (1995) Systemic treatment for metastatic breast cancer. In: Cancer Treatment, pp. 354-368. Editor: i’ M. linskell. W.B. Saunders, Philadelphia. PA. 151 Sinha. H.K., Katkl, A.G., Hatist, G. Cowan, K H. and My,‘{\. (‘ I:. ( 1987) Adriamycin-stimulated hydroxyl radical fOVJlJ;l!iOli in human breast tumor cells. Hiochem. PharmaLOi 36. 733.--796. Ih] Albcrlh, l1.S. (1080) Tabular summary of pharmacokinertc parameters relevant to in vitro drug assay In: Cloning of Hum;rrt Tumor Stem (‘ells. Editor: S.E. Salmon. Alan I< Lish, New York 171 Kurbacher. CM., Bruckner. l1.W.. Andrcotti, P.E.. Kurha?hcr. .I.,~.. Sass, G. antI Krchs, D. (I Y9.5) In vitro activity ol tilanoccnedichloride versus cisplatin in four ovarian carc1noma ccl1 lines evaluated hy a microtiter plate ATY hioluminccccnts assay. Anti-Cancer Drugs, 6, 6Y7- 704. 181 Cameron. I-. and Pauling, L. ( 1976) Suppicmentni ascorhate m the supportive trcatmcnt of cancer: prolongation of sur.+lvnl times in terminal human cancer. Proc. Nat]. Acad Sci. IJSA. 73.3685 -3689. I!){ Marquardt, 11.. Rufling, 1.. and We&burger, J.11. (IY77i Mutagenic activity of mtritc treated foods. hurnan stomach t‘ancer ma> he related to dietary factors. Science. 196. 1000 lOOi
1201 Khan, P.K. and Sinha, S.P. (1993) Antimutagenic efficacy of higher doses of vitamin C. Mutat. Res., 298, 157-I 61. 12 11 Jain, M., Miller, A.B. and To, T. (1994) Premorbid diet and the prognosis of women with breast cancer. J. Natl. Cancer Inst., 86. 1390.-1397. [22] Taper, H.S., De Gerlache, J., Lans. M. and Roberfroid, M. (1987) Son-toxic potentiation of cancer chemotherapy by combined C and KY vitamin pretreatment. Int. .I. Cancer. 40, 57.5-579. 1231 Taper, H.S. and Koberfroid, M. (1992) Ken-toxic sensitization of cancer chemotherapy by combined vitamin C and K3 pretreatment in a mouse tumor resistant to oncovin. Anticancer Res., 13, 1651-1654. [24] De Loeckcr, W., Janssens, J., Bonte, J. and Taper. H.S. (1993) Effects of sodium ascorhate (vitamin C) and 2. methyl- I ,4+aphthoquinone (vitamin K3) treatment on human cell growth in vitro. II. Synergism with combined chemotherapy action. Anticancer Res., 13, 103 -104. 1251 Pething, R., Gascoync, P.R.C.. ,%lcLaughlin, J.A. and SzentGyorgyi, A. ( 1983) Ascorbate-quinonc interactions: electrochemical, free radical and cytotoxic properties. Proc. Natl. Acad. Sci. USA, 80, 129. 132. 1261 Doroshow, J. (lY86) Prevention of doxorubicin-induced killing of MCF-7 human breast cancer cells by oxygen radical scavengers and iron chelating agents. Biochem. Biophys. Kes. Commun., 135, 330-335. [271 Mimnaugh, E.ti., l)usre, I... Atwell, J. and Myers, C.E. 11989) Dirfcrential oxygen radical susceptibility of adriamycin-sensitive and -resistant MCF-7 human hreast tumor cells. Cancer Rcs., 49, 8-1.5 1281 Shimpo, K., Nagatsu, T., Yamada, K., Sate, T.. Niimi, H., Shamoto, M.. Takeuchi, T.. Umeznwa, II. and Fujita, K. (1991) Ascorbic acid and adriamycin toxicity. Am J Clin. ICulr., 54, 12YXS-13OlS. 1291 Tonctti, M.. Giovine, M., Gasparini, A.. Bcnatti, U. and De FIoJ-:I, A. (1993) Enhanced formation of reactive species from cis-diammine-( I, I -cyclobutanedicarboxylato)-platinum(Il) (carboplatin) m the presence of oxygen free radicals. Hiochem. Pharmacol, 46. 1377-1383. (301 Gemha, M. and Fukuishi, N. (I99 I) Amelioration by ascorbic acid of cisplatin-induced injury in cultured renal epithclial cells. Contrih. Nephrol., 9.5, 138-142. 13 I ] Spencer, C.M. and Faulds, D. (I 9Y4) Paclitaxel. A review of its pharmacodynamic and pharmacokinctic properties and therapeutic potential in the treatment of cancer. Drugs, 4X. 794 x47 1321 Roy, S.?. and llorwitz, S.B. (IYXS) A phosphoglycoprotein associated with Taxol-resistance in J774.2 cells. Cancer Kes.. 4s. 3856 3863.