<<

(CANCER RESEARCH 51, 6704-6707. December 15, 1991] Advances in Brief W-^S-DiacetoxypentyOdoxorubicin: A Novel Producing DNA Interstrand Cross-Linking and Rapid Endonucleolytic Cleavage in Human Cells1

Leonard A. Zwelling,2 Elizabeth Altschuler, Abdallah Cherif, and David Farquhar

Department of Medical Oncology, Division of Medicine, The University of Texas M. D. Anderson Cancer Center, Houston, Texas 77030

Abstract Materials and Methods

The cytotoxic and DNA-damaging effects of a novel alkylating anthra- The design and synthesis of A-(5,5-DAP)DOX (Fig. 1) has been cycline, yV-(5,5-diacetoxypentyl), were quantified in HL-60 described previously (2). The cellular pharmacology, biochemistry, and human leukemia cells and in an intercalator-resistant daughter line, III- molecular biology of HL-60 and its -resistant daughter line, HL-60/AMSA, have been described in detail (3, 4). The cells were 60/AMSA. The new drug was cytotoxic to both lines at doses as low as 50 nM for 1 h. /V-(5,5-Diacetoxypentyl)doxorubicin produced DNA inter- grown and radiolabeled for alkaline elution as described previously (3). strand cross-linking in both lines. The cross-linking appeared to increase Soft agar colony formation assays and alkaline elution assays using proteinase were performed and quantified as described in the past (3). in both lines following drug treatment, but the increase was greater in the resistant line. This appeared to be due to an underestimation of cross- To quantify endonucleolytic cleavage, cells with their DNA labeled as for alkaline elution were exposed to various concentrations of N- linking, particularly in sensitive HL-60, secondary to time-dependent (5,5-DAP)DOX for 1 h. The cells were harvested by centrifugation and DNA fragmentation that followed drug removal. This time-dependent washed in drug-free medium and then lysed during a 20-min incubation DNA fragmentation was probably endonucleolytic cleavage (a feature of with 10 mM Tris, I mivi EDTA, and 0.2% Triton X-100 (pH 7.5) on apoptosis) as characteristic nucleosomal ladders were produced by V- ice. This lysate was centrifuged at 13,000 x g for 10 min. The super (5,5-diacetoxypentyl)doxorubicin treatment in a cotemporal time-depend natant contained fragment DNA and the pellet contained intact chro- ent fashion. This novel anthracycline is the tirsi of a family of alkylating matin (5). This was confirmed by comparing the percentage of DNA in designed to be water soluble, easy to formulate, and the fragmented fraction with the percentage of DNA in the lysis fraction capable of producing DNA interstrand cross-linking. Because this last of filter elution assays performed with aliquots of the same cells (Table characteristic has previously been associated with doxorubicin analogues 1). of great potency and low toxicity, these newer, more readily formulated To demonstrate that the DNA in the lysis fraction truly represented drugs may have great clinical utility. endonucleolytic cleavage [i.e., apoptosis (6, 7)], we took aliquots of the identical cells used for quantification of endonucleolytic cleavage and lysed them in the same lysis buffer as above. The supernatant fraction Introduction (fragmented DNA) was then brought to 0.5 M NaCl (with 5 M NaCI) and to 50% isopropyl alcohol (with 100% isopropyl alcohol). Samples 7V-(5,5-DAP)DOX3 (Fig. 1) is representative of a new class were stored at -20°C overnight. The following day the samples were of extremely potent doxorubicin analogues designed and syn centrifuged at 13,000 x g for 10 min and the precipitated pellet was thesized by Farquhar and Cherif (1, 2).4 The molecule contains resuspended in 89 HIMTris-borate buffer (pH 8) (3). This was followed a 5,5-bisacetoxypentyl substitutent in the 3'-amino position of by adding one-sixth volume of 15 m\i EDTA, 2% SDS, and 50% glycerol loading buffer and heating the mixture to 65°Cfor 10 min. the sugar. This substituent was designed to undergo hydrolysis to the corresponding aldehyde in the pres The samples were loaded on a 1.5% agarose gel and run at 20 V ence of carboxylate esterases, enzymes that are ubiquitous in overnight in the presence of 0.5 Mg/ml of ethidium bromide to allow the direct visualization of DNA in the gel. tissue. It was anticipated that the liberated aldehyde would react with a nucleophilic group proximate to the site of drug inter calation or other form of DNA binding and thus form a covalent Results adduci. Hopefully, this approach would overcome cellular re The cytotoxic effects of 1-h treatments of HL-60 and HL- sistance to antitumGr anthracyclines. 60/AMSA cells with /V-(5,5-DAP)DOX were quantified using We now report on the activity of one of the first members of a soft agar colony formation assay (Fig. 2, top). At N-(5,5- this new class of compounds in a drug-sensitive/drug-resistant DAP)DOX concentrations in excess of 50 nM, survival in both cell pair in which the resistant line contains an intercalator- lines was markedly reduced, with the cytotoxicity being slightly resistant form of topoisomerase II (3). greater in effect on HL-60. The drug produced approximately equal frequencies of DNA interstrand cross-linking in both cell Received 8/30/91; accepted 10/29/91. lines immediately following a 1-h treatment (Fig. 2, bottom). The costs of publication of this article were defrayed in part by the payment However, /V-(5,5-DAP)DOX did not produce protein-associ of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ated DNA cleavage (data not shown). Together these observa 1This study was supported by USPHS Research Grant CA40090 (L. A. Z.). tions indicated that a cell line that contains an intercalator- American Cancer Society Grant CH-324D (L. A. Z.). a grant from the Physicians Referral Service of the M. D. Anderson Cancer Center (L. A. Z.), and a grant resistant form of topoisomerase II and that resists the actions from the University of Texas, M. D. Anderson Cancer Foundation (D. F.). of a number of topoisomerase II-reactive DNA intercalators 2To whom correspondence should be addressed, at Box 52, 1515 Holcombe including anthracyclines (3, 8) is not resistant to this new agent. Blvd., Houston, TX 77030. 3The abbreviations used are: A'-(5,5-DAP)DOX, iV-(5,5-diacetoxypen- Additionally, the DNA effect produced by this new drug is tyl)doxorubicin hydrochloride; MRA-CN, 3'-deamino-3'-(3-cyano-4-morphoIi- distinct from that produced by doxorubicin and other topoisom nyl)do\orubicin: SDS, sodium dodecyl sulfate. 4 A. Cherif and D. Farquhar. yV(5,5-Diacetoxypentyl)doxorubicin: a new in erase II-reactive anthracyclines (3, 9). tensely potent doxorubicin analogue, submitted for publication. Previously, we and others had demonstrated that another 6704

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1991 American Association for Cancer Research. DNA DAMAGE BY /V-(5,5-DAP)DOX

A45,5-DAP)DOX but rather immediately following . When SDS was included in the assay, large amounts of etopo- side-induced DNA fragmentation were detected. In the absence of SDS, DNA fragmentation following etoposide was com parable with that detected in untreated cells. This is undoubt edly due to etoposide-induced, topoisomerase II-mediated DNA CHjO cleavage, which depends upon strong dénaturantslike SDS for its production (14). Interestingly, 3 h after etoposide treatment, DNA fragmentation was detected by all the methods. Moreover, in separate experiments, we have shown that this later fragmen tation (but not that fragmentation revealed by SDS immediately . HCI after etoposide treatment) is associated with the production of a typical nucleosomal ladder in the gel assay (data not shown).

Discussion A variety of anthracycline analogues can produce interstrand Table 1 Comparison of methods by which to quantify endonucleolytic cleavage of cross-links. The best known of these analogues is MRA-CN DNA from drug-treated HL-60 cells (10-13). Because formulation and solubility problems may have % of cellular DNA in lysis solution with delayed the progression of this agent to the clinic, alternative following technique used anthracycline analogues that retain the ability of MRA-CN to following drug removal + cross-link DNA have been designed. /V-(5,5-DAP)DOX is one Proteinase*233871899495911723Filter-Triton'102861879212871220Pellet-Triton''123062899012841217 (min)]A^S.S-DAP) of these. It behaves similarly to MRA-CN in the cell culture nivi)060120180240EtoposideDOX (500 systems used in this work. In particular, /V-(5,5 DAP)DOX produced DNA fragmentation like MRA-CN (12, 13). The new data presented here suggest that this fragmentation is secondary to endonucleolytic cleavage [i.e., apoptosis, (6, 7) [Fig. 3]. This cleavage explains the excess DNA detected in the lysis fraction MM)0240Untreated0240Filter-SOS"203666899694901624Filter-SDS(100 of elution assays and the apparent decline in cross-linking following drug removal. The association between the cleavage and the decreased cross-

°Cells were deposited on a polyvinyl chloride filter and lysed with 5 ml of an SDS-containing lysis solution used in filter elution (2% SDS, 0.025 M disodium EDTA, pH 10). Values are percentages of the total DNA that eluted from the filter. * As in Footnote a with the addition of 0.5 n%/m\ of poroteinase. ' As in Footnote a except the lysis solution was the 0.2% Triton X-IOO lysis solution described in "Materials and Methods." d Quantification of endonculeolytic cleavage using Triton X-100 lysis and no filter as described in "Materials and Methods."

anthracycline, MRA-CN, produced interstrand cross-linking 100 (10-13). The DNA of cells treated for l h with that drug and incubated in drug-free medium at 37°Cfragmented within 3-6 hours. This same effect was produced by /V-(5,5-DAP)DOX (Fig. 3). Following treatment, cross-linking actually increased to a greater extent in HL-60/AMSA than in HL-60. This is probably because the increased amount of DNA fragmentation in HL-60 cells produced an underestimation of the cross-linking (12). This fragmentation is not secondary to generalized cellular necrosis but rather seems to be consistent with endonucleolytic cleavage (a feature of apoptosis; are Refs. 6 and 7) as evidenced by the DNA nucleosomal ladder patterns seen in the gels in 250 500 Fig. 3. Production of endonucleolytic cleavage is not an unique N-(5,5-DIACETOXYPENTYL)DOXORUBICIN-HCL or specific effect of /V-(5,5-DAP)DOX (see below). However, CONCENTRATION(nM) Fig. 2. Top, ability of various concentrations of A'-(5,5-DAP)DOX to reduce the onset of this effect was particularly rapid following drug the colony-forming ability of HL-60 and HL-60/AMSA human leukemia cells. treatment and occurred at drug concentrations as low as 50 nM All treatments were for I h at 37"C. Points, means of at least three independent (Fig. 3), a rather low drug concentration. determinations. Bars, SD. If fewer than three determinations were made, individ ual points are shown. Colony-forming efficiencies were 0.65 ±0.49 for HL-60 The data in Table 1 confirm the mechanistic connection and 0.98 ±0.4 for HL-60/AMSA (N = 5). Bottom, DNA interstrand cross- between the endonucleolytic cleavage visualized in Fig. 3 and linking frequency produced by l-h treatments of HL-60 and HL-60/AMSA cells with various concentrations of /V-(5,5-DAP)DOX. DNA interstrand cross-linking the fragmented DNA quantified in the lysis fraction of elution was quantified using alkaline elution with proteinase (see "Materials and Meth assays. However, the results from the various methods of frag ods" and Refs. 3, 8, 9. and 12) and expressed as rad equivalents (9). Each point mentation quantitation differed not following treatment with is the mean of at least four independent determinations. 6705

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1991 American Association for Cancer Research. DNA DAMAGE BY ,V-(5.5-DAP)DOX

•¿ »HL-60 O OHL-60/AMSA to

|0 |0 HL-60 I HL-60/AMSA

N (5.5 DAP) DOX 5005077820 50050222444322 010 500 50 5050 Concentration(nM)DNA (%iDNAin Supernatant 420 69 60 2758 (%)ISCin ElutKjnLysis 1918252 84 75788 32258

Frequency 14HL-60/AMSA014I 35500 74 60 120 180 (Rad-equivalents)HL-fiO TIME FOLLOWING A ONE HOUR TREATMENT WITH 500 nM N-(5.5-DlACETOXYPENTVL)DOXORUBICIN-HCL (min) Fig. 3. Left, effect of a 1-h treatment with 500 nM /V-(5,5-DAP)DOX on the integrity of the HL-60 and HL-60/AMSA cellular DNA in the time following drug treatment and removal. Top. DNA interstrand cross-linking; bottom. DNA in the lysis fraction prior to elution (see text). Right, effect of a 1-h treatment with different concentrations of A'-(5,5-DAP)DOX on DNA integrity as measured using agarose gel assays for endonucleolytic cleavage and alkaline elution (see text). The same cells were used for measurements of endonucleolytic cleavage by the gel assay, by quantitative ¡sopropylalcohol precipitation (DNA in supernatant), and by alkaline elulion (DNA in clulion lysis). DNA interstrand cross-linking (ISC) was quantified using alkaline elution with proteinase (4). Measurements »eremade immediately after the 1-h treatment (f0) or after a further 180-min incubation at 37"C in the absence of drug (t,m). linking is best appreciated in Fig. 3. Although cross-linking is position" might produce a good anticancer drug. In fact, Far- higher in resistant cells than in wild-type cells, there is an quhar et al. (23) previously reported the synthesis and biological accompanying lower amount of DNA detected in the lysis evaluation of a series of doxorubicin analogues bearing chemi fraction from resistant cells. Because the latter are intercalator cally reactive substituents at this position. A^S.S-DAPJDOX, resistant on the basis of a drug-resistant form of topoisomerase the new analogue described here, is a further extension of this II (3, 8), it is possible that topoisomerase II, a target for other approach. Studies are planned to define the size and composi intercalating anthracyclines (3, 9, 15), may be involved in the tion of the 3'-amino substituent for maximal cytotoxic activity manner in which /V-(5,5-DAP)DOX kills cells. On the other and to determine whether this new class of compounds is active hand, a large production of protein-associated DNA cleavage in vivo while retaining low cardiotoxicity. was not detected in cells treated with the new drug which suggests that its primary target is not topoisomerase II. Acknowledgments yV-(5,5-DAP)DOX also allowed us to demonstrate that the fragmented DNA we detect in the pH 10 elution lysis is prob We wish to thank Dr. Mike Story of the Department of Experimental ably that which constitutes the bulk of the DNA of the nucleo- Radiotherapy, M. D. Anderson Cancer Center, for teaching E. Altschu- somal ladder detected in gel assays of endonucleolytic cleavage, ler the method of assessing endonucleolytic cleavage and Dr. Raymond a feature of apoptosis (6, 7) (Table 1). These same changes were Meyn for critical reading of the manuscript. We also thank Diane produced by a topoisomerase II-reactive drug that does not Rivera for editing the manuscript. produce interstrand cross-links (etoposide) (Table 1). Thus, apoptosis can be triggered by cytotoxic agents with different References initial effects on cellular DNA (6, 7, 16-21). Note, however, 1. Farquhar. D., and Cherif. A. Potently tutnoricida! anthracycline analogues. that the DNA fragmentation following ¿V-(5,5-DAP)DOXwas U. S. Patent Application 07/608.149, 1990. clearly not topoisomerase II-mediated, because such DNA 2. Cherif, A., Zwelling, L. A., Altschuler. E.. and Farquhar. D. Doxorubicin cleavage (e.g., etoposide-induced, protein-associated DNA analogues bearing latent alkylating groups. Proc. Am. Assoc. Cancer Res., 32:405, 1991. cleavage) required SDS to detect (Table 1), while DNA frag 3. Zwelling, L. A., Hinds, M., Chan, D., Mayes, J.. Sie, K. L., Parker, E., mentation following /V-(5,5-DAP)DOX did not. Silberman, L.. Radcliffe, A., Beran. M., and Blick, M. Characterization of an amsacrinc-rcsistant line of human leukemia cells. Evidence for a drug- The chemical nature of the interstrand cross-link produced resistant form of topoisomerase II. J. Biol. Chem.. 264: 16411-16420, 1989. by /V-(5,5-DAP)DOX is not known. However, the facts that 4. Beran, M. and Andersson, B. S. Development and characterization of a both this agent and MRA-CN are modified in the 3'-amino human tnyelogenous leukemia cell line resistant to 4'-(9-acridinylamino)-3- methanesulfon-m-anisidide. Cancer Res.. 47: 1897-1904, 1987. position of doxorubicin (Fig. 1) and that both agents cross-link 5. Sellins, K. S., and Cohen, J. J. Gene induction by -»-irradiationleads to DNA DNA suggest that this site may be uniquely positioned to fragmentation in lymphocytes. J. Immunol, 139: 3199-3206, 1987. 6. Kaufmann, S. Induction of endonucleolytic DNA cleavage in human acute produce cross-linking once the drug has intercalated. In this myelogenous leukemia cells by etoposide. , and other cytotoxic regard, Gao et al. (22) have demonstrated that in the presence anticancer drugs: a cautionary note. Cancer Res., 49: 5870-5878, 1989. 7. Cotter, T. G., Lennon, S. V., Glynn, J. G.. and Martin, S. J. via of formaldehyde a analogue modified with a sugar apoptosis and its relationship to growth, development and differentiation of at the 4'-oxygen position of daunosamine formed a covalent both tumor and normal cells. Anticancer Res., 10: 1153-1160, 1990. adduct with a DNA hexamer through a 3'-amino méthylène 8. Zwelling, L. A.. Mayes, J.. Hinds. M.. Chan, D.. Altschuler, E.. Carroll, B., bridge. These authors suggested that an anthracycline with "a Parker, E., Deisseroth, K.. Radcliffe, A., Seligman, M., Li, L., and Farquhar, D. Cross-resistance of an amsacrine-resistant human leukemia line to topoi reactive (e.g., alkylating) functional group at the 3'-nitrogen somerase II reactive DNA intercalating agents. Evidence for two topoisom- 6706

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1991 American Association for Cancer Research. DNA DAMAGE BY A'-(5,5-DAP)DOX

erase II directed drug actions. Biochemistry, 30: 4048-4055, 1991 M. Topoisomerase inhibitors induce irreversible fragmentation of replicated 9. Zwelling, L. A., Michaels, S., Erickson, L. C., Ungerleider. R. S., Nichols, DNA in concanavalin A stimulated splenocytcs. Biochemistry, 27: 95-99, M.. and Kohn. K. W. Protein-associated DNA strand breaks in L1210 cells 1988. treated with the DNA intercalating agents 4'(9-acridinylamino) methanesul- 17. Barry, M. A., Behnke, C. A., and Eastman, A. Activation of programmed fon-m-anisidide and Adriamycin. Biochemistry, 20:6553-6563, 1981. cell death (apoptosis) by . other anticancer drugs, toxins, and hyper- 10. Westendorf, J., Groth, G.. Steinheider, G., and Marquardt. H. Formation of thermia. Biochem. Pharmacol., 40: 2353-2362, 1990. DNA adducts and induction of DNA-crosslinks and chromosomal aberra 18. Alnemri, E. S., and Litwack, G. Activation of internucleosomal DNA cleav tions by the new potent anthracycline antitumor antibiotics: morpholinodau- age in human CEM lymphocytes by glucocorticoid and novobiocin. J. Biol. nomycin, cyanomorpholinodaunomycin and cyanomorpholinoadriamycin. Chem., 265: 17323-17333, 1990 Cell Biol. Toxicol., /: 87-101, 1985. 19. Del Bino. G., and Dar/.ynkiewicz, Z. Camptothecin, , or 4'-(9- 11. Begleiter, A., and Johnston, J. B. DNA crosslinking by 3'-(3-cyano-4-mor- acridinylamino)-3-methanesulfon-m-anisidide, but not or dox pholinyl)-3'-deaminoadriamycin in HT-29 human colon carcinoma cells in orubicin, induces degradation of nuclear DNA in the of HL-60 cells. vitro. Biochem. Biophys. Res. Commun.. 131: 336-338, 1985. Cancer Res., 51: 1165-1169, 1991. 12. Lau, D. H. M., Ross, K. L., and Sikic, B. I. Paradoxical increase in DNA 20. Kyprianou, N., Alexander, R. B.. and Issacs, J. T. Activation of programmed cross-linking in a human ovarian carcinoma cell line resistant to cyanomor- cell death by recombinant human tumor necrosis factor plays topoisomerase pholino doxorubicin. Cancer Res., 50: 4056-4060. 1990. II-targeted drugs in L929 cells. J. Nati. Cancer Inst., 83: 346-350. 1981. 13. Wassermann. K., Zwelling, L. A., Mullins. T. D., Silberman, L. E., Anders- 21. Walker, P. R., Smith, C., Youdale, T., Leblanc, J., Whitfield, J. F.. and son, B. S., Bakic. M.. Acton, E. M., and Newman, R. A. Effects of 3'- Sikorska, M. Topoisomerase II-reactive chemotherapeutic drugs induce deamino-3'-(3-cyano-4-morpholinyl)doxorubicinand doxorubicin on the sur apoptosis in thymocytes. Cancer Res., 51: 1078-1085, 1991. vival, DNA integrity, and nucleolar morphology of human leukemia cells in 22. Gao, Y-G.. Liaw, Y-O, Li, Y-K., van der Marel, G. A., van Boom, J. H., vitro. Cancer Res.. 46: 4041-4046, 1986. and Wang. A. H-J. Facile formation of a crosslinked adduct between DNA 14. Liu, L. F. DNA topoisomerase poisons as antitumor drugs. Annu. Rev. and the duanorubicin derivative MAR70 mediated by formaldehyde: molec Biochem., SS: 351-375, 1989. ular structure of the MAR70-d(CGT"ACG) covalent adduct. Proc. Nail. 15. Tewey, K. M., Rowe, T. C.. Yang, L.. Halligan, B. D., and Liu, F. L. Acad. Sci. USA, SS: 4845-4849. 1991. Adriamycin-induced DNA damage mediated by mammalian DNA topoisom 23. Farquhar. D., Newman, R. A., Zuckerman, J. E.. and Andcrsson, B. S. erase II. Science (Washington DC), 226: 446-468, 1984. Doxorubicin analogues incorporating chemically reactive substituents. J. 16. Jaxel, C., Taudou, G., Portemer, C.. Mirambeau, G., Panijel, J., and Duguet. Mcd. Chem.. 34: 561-564. 1991.

6707

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1991 American Association for Cancer Research. N-(5,5-Diacetoxypentyl)doxorubicin: A Novel Anthracycline Producing DNA Interstrand Cross-Linking and Rapid Endonucleolytic Cleavage in Human Leukemia Cells

Leonard A. Zwelling, Elizabeth Altschuler, Abdallah Cherif, et al.

Cancer Res 1991;51:6704-6707.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/51/24/6704

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/51/24/6704. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1991 American Association for Cancer Research.