FIELD OF THE INVENTION

The present invention relates generally to a method of screening for potential anticancer drugs using a cell differentiation assay.

BACKGROUND OF THE INVENTION

Cancer has often been depicted as a disease in which cell proliferation and differentiation become uncoupled and malignant cells proliferate uncontrollably. One proposed strategy of cancer chemotherapy derived from this viewpoint involves the induction of terminal differentiation of the malignant cell resulting in a cell with limited capacity for proliferation [1-3]. Such a strategy is attractive because the drug is targeted to a select cell population, but this approach has had limited application [1,4]. The main thrust of cancer chemotherapy has been directed at cytotoxic drugs that limit the proliferation of cells in general but are particularly effective with rapidly growing tumor cells. A drawback to this strategy is that slow growing cancer cells remain unaffected.

Accordingly, it would be desirable to provide a method of ascertaining whether compounds exhibit an efficacy for inducing terminal differentiation, and using such a method to determine whether known, or as yet unknown cancer drugs induce terminal differentiation in order to provide a predictive basis for searching for new potential anticancer drugs.

SUMMARY OF THE INVENTION

In one aspect the invention provides an in-vitro method of screening candidate anticancer compounds. The

screening method comprises the steps of, providing a plurality of cells from a hemopoetic cell line which are blockable in the pathway of terminal differentiation. The cells are treated in a solution comprising a cell differentiation inducing agent and a differentiation inhibiting agent for a predetermined period of time. The treated cells are then removed from contact with the inducing and inhibiting agents and the cells are exposed to the candidate compound for a predetermined period of time. The exposed cells are then subjected to a detection step for detecting the presence or absence of terminally differentiated cells among the cells exposed to the candidate compound.

In a more specific aspect of the invention the inducing agent is dimethylsulfoxide, the inhibiting agent is 3-aminobenzamide and commitment to terminal differentiation is detected by measuring the formation of hemoglobin detected by assaying said cells for benzidine positive results.

In another aspect of the invention, there is provided a pharmaceutical preparation for inhibiting the growth of cancer cells which comprises pharmaceutically acceptable G ₂ /mitosis blocking compounds. In this aspect of the invention the G ₂ /mitosis blocking drugs are from the class comprising Hoechst 33342, podophyllotoxin, puromycin, cycloheximide, amphotericin-B, griseofulvin and isoproterenol.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 illustrates the effect of anticancer drug concentration on commitment in FEL cells. Cells which were previously incubated in medium containing DMSO and 3AB were washed and incubated in medium containing (A) adriamycin (a-e concentration are 0.006, 0.025, 0.10, 0.20 and 0.40 μM respectively), (B) vincristine (a-e

concentrations are 1.1, 5.5, 11.0, 22.0 and 44.0 nM respectively), and (C) bleomycin (a-e concentrations are 0.002, 0.01, 0.05, 0.25 and 0.50 u/ml respectively). The cell viability is as follows: (A) adriamycin (a-e) 99, 98, 91, 67 and 61%, (B) vincristine (a-e) 96, 86, 80, 81 and 76%, and (C) bleomycin (a-e) 99, 98, 96, 80 and 65%. The experimental protocol is described in Materials and methods and in Table 1. The data represents the average of three experiments and the error bars equal one standard deviation. The controls which are not shown are similar to those in Table 1.

Figure 2 displays results of the study of the accumulation of cells block in G ₂ /mitosis following treatment by anticancer drugs. The basic experimental protocol is similar to that described herebelow in the Materials and methods section. The DMS0/3AB-treated cells were incubated in medium containing 0.1 μM adriamycin, 11 nM vincristine or 0.05 u/ml bleomycin and subsequently analyzed by flow cytometry using propidium iodide. The experiment has been repeated and representative data is shown.

DESCRIPTION OF THE INVENTION A. Materials and Methods

Friend erythroleukemia (FEL) cells were cultured at 37°C in Iscoves modified Dulbecco\'s (Iscoves) medium supplemented with 10% fetal bovine serum (FBS) (Blocknek, BDH, Canada), penicillin (50 units/ml), and streptomycin (50 μg/ml) in a humidified atmosphere containing 5% C0 ₂ . Commitment was assayed as described [5]. Briefly, logarithmically grown FEL cells (K clone) were treated with 1.5% dimethylsulfoxide (DMSO) and 8 mM 3-aminobenzamide (3AB) (Pfaltz and Bauer, Waterbury, CT) for 64 h, centrifuged, washed, and exposed to various concentrations of drugs for 20h. The cells were subsequently transferred

to fresh medium containing cytochalasin B (1.0 μg/ml; Aldrich Chemicals, Milwaukee, WI) for 24 h to allow for the expression of the differentiated phenotype in the absence of further cell division. The cells were finally stained for Bz+ (hemoglobin-producing, differentiated) cells using the benzidine staining procedure as described by Gopalakrishnan and French Anderson, usually 96 hours after the start of the experiment [6]. A minimum of 300 cells were counted for each sample. Cell viability was assayed using the trypan blue exclusion assay [7] at the end of the experiment. Mitotic indices were measured as described [8]. Cytofluorometry analysis was performed using nuclei stained with propidium iodide as described [9] . For soft agar colony analysis, the anticancer drugs were removed from culture, the cells were washed, plated on soft agar for 4-5 days [10], stained, and examined for Bz+ colonies.

B. Results and Discussion

The FEL cell is a virally transformed murine cell line that has been a useful model for studying the differentiation process. The FEL cell differentiates following exposure to a variety of agents, including some antitumor drugs [11-13]. It has been demonstrated that treatment of FEL cells in medium with the inducer DMSO and the poly (ADP-ribose) polymerase inhibitor, 3AB for 56 h blocked the differentiation pathway just prior to commitment [10]. When these cells were washed to remove DMSO and 3AB and incubated in medium containing DMSO, commitment was rapid, indicating that many of the preliminary steps in the differentiation pathway had already occurred during the initial incubation with DMSO and 3AB. On the other hand, if these cells were returned to medium containing no additions, the cells resumed normal growth, and few differentiated cells were evident [5], Surprisingly, when the DMS0/3AB-treated cells were

incubated with the mitotic inhibitors nocodazole or colcemid, the cells became committed to terminal differentiation [14]. However, further incubation in the absence of the mitotic inhibitors was required for the expression of the differentiated phenotype. Since the number of mitotic cells in medium containing cytochalasin B, see [5], were approximately equivalent to the number of cells that differentiated, we suggested that mitosis may be a requirement for commitment. Extension of these studies to the antimitotic, anticancer drugs vincristine and vinblastine, indicated that they were also effective in commitment (Table 1). The antineoplastic drug adriamycin, known to damage DNA and depress mitotic activity [15] was also effective. A further survey of clinically used anticancer drugs, indicated that they all induced commitment (Table 1). Since antineoplastic drugs are known to cause a cell cycle delay in G ₂ [16] or mitosis, [17] we tested G ₂ mitosis blocking drugs [18-23] and found that they were also active in commitment (Table 2). The results shown in Tables 1 and 2 summarize data obtained from a larger experiment involving a range of drug concentrations. The data shown represents a single drug concentration, in which the effect on differentiation is clearly evident and where viability is reasonably maintained. The drug concentrations were selected on the basis of previously reported studies, and where the data was available, included concentrations which were attained during cancer chemotherepy [24-26] . Examples of these more detailed studies are shown in Fig. 1, in which the relationship between increasing drug concentration (adriamycin, vincristine, or bleomycin) and the appearance of differentiated Bz+ cells. The in-vivo plasma concentrations of adriamycin (approx. 1-27 nM), vincristine (approx. 10-15 nM) and bleomycin (0.13-10 mu/ml) reported to be attained during chemotherapy [24-26] also

differentiate FEL cells under in-vitro conditions. The inventors have also noted that at higher drug concentrations there is decreased cell viability. One feature that distinguishes these anticancer drugs from other inducers is that the concentrations of drug required for commitment is low, and in many cases similar to levels attained in chemotherapy [24-26] . Although many agents induce FEL cell differentiation, [11-13] (R. Dinnen, unpublished observation) an important consideration for a potential anticancer drug is that differentiation occurs at physiologically tolerable concentrations. The following agents were found to be ineffective in the commitment assay: butyryl cAMP (0.05-2.5 mM), EGTA (0.14-6.75 mM), acetic acid (0.1-2mM), ethanol (0.5-2.5%), penicillin/streptomycin (500 μg/ml; 500 μg/ml), antimycin A (0.02-0.9 μM), rifampin (0.01-1.5 μM), N-ethylmaleimide (4-20 μM) and AZT (zidovudine) 1-200 μM). The concentrations selected were in the range used, and usually in excess of other studies [12, 27, 28].

A few of the anticancer drugs were studied in more detail. Cytofluorometric analysis of FEL cells treated with vincristine, adriamycin, or bleomycin under the conditions described in Tables 1 and 2 indicated accumulation of G ₂ /mitosis-delayed cells (Fig. 2). These drugs were previously observed to cause G ₂ /mitosis delay in other cell lines [15, 17, 29] . After vincristine treatment, metaphase cells accumulated, as indicated by staining for mitotic indices (Table 3). Furthermore, when FEL cells treated with vincristine were subsequently plated in soft agar [10], 55/133 were one-celled BZ+ colonies and the remainder composed of colonies with 2-10 Bz+ cells (summation of two experiments). This indicated that terminal differentiation had occurred. There were 23 mixed colonies composed of a few Bz+ cells and a large number of Bz- cells.

The purely undifferentiated colonies were also composed of large numbers of cells. Similar results were obtained in experiments involving bleomycin, amphotericin-B and hydroxyurea (data not shown).

The possibility that terminal differentiation might be the primary basis for drug-induced cell killing is not obviously supported by the observation that most anticancer drugs including vincristine are weak inducers of FEL cell differentiation [11-13] (R. Dinnen, unpublished observation). However, hemoglobin-production is an indirect assay for differentiation, and the total number of differentiated cells might be quite different by a more direct assay. The inventors have observed that vincristine treatment of FEL cells (745A line) yielded approximately 5% Bz+ cells. However, randomly selected clones derived from 745A gave wide variation in the number of Bz- cells ( <1- 27%). Five of these clones that differentiated to 5% or greater were recloned, and the phenotype confirmed. The K clone that was derived from 745A and used in the commitment assays also showed low numbers ( <1%) of Bz+ cells following vincristine treatment. These results indicate that (1) the commitment and direct assays for differentiation are quantitatively different and (2) the widely varied responses of different clones suggests that under some circumstances differentiation may be a significant factor in cell killing and subtle differences in clonal properties may account for the different sensitivities of tumors to chemotherapy.

The studies described here suggest a linkage between anticancer drugs, a cell cycle block and differentiation. The inventors have shown that anticancer drugs induce a cell cycle block in G ₂ /mitosis under the test conditions employed in the studies disclosed herein and that G ₂ mitosis blocking drugs induce differentiation. The relationship between many anticancer drugs and a G ₂ /mitosis

cell cycle block may be attributed to accompanying DNA damage caused by the drug [30] . Previously, DNA damage and FEL cell differentiation were found to be related [31-33], and in studies with the anticancer drug temozolomide it was noted that DNA damage, G ₂ arrest and differentiation of K562 leukemia cells were linked [34] . However in in-vivo studies it is uncertain whether differentiation is linked to tumor cell death. Isolated clinical observations with leukemia had indicated that some anticancer drugs induce differentiation of the cancer cell [35,36], These uncertainties do not alter the main inferences derived from these studies. Specifically, the high correlation between anticancer drugs, a cell cycle block and differentiation as disclosed herein, forms the basis of the method of using a differentiation assay to search for and screen for new anticancer drugs forming the subject invention. Furthermore, G ₂ /mitosis-blocking agents should be included in a survey of potential new anticancer drugs.

While the screening method of the present invention has been described as including the step of exposing a hemopoetic cell line to a medium comprising an inducer and an inhibitor, it will be appreciated that in the broadest aspect of the invention, this step could be dispensed with and the hemopoetic cell line exposed only to the candidate compound being tested for a predetermined period of time followed by searching for postcommitment events. In this embodiment of the screening method, the number of committed cells will only be a few percent and greater detection sensitivity would be required. In the preferred embodiment previously described, the step of exposing the cells to the inducing and inhibiting agents results in a higher percentage of committed cells.

Those skilled in the art will appreciate that while the screening method disclosed herein has been described using cells from the FEL cell line, other cell

lines may be employed. A required functional feature of the cell line being employed is that the cells are capable of being blocked in the pathway of terminal differentiation by inhibiting agents prior to the cells passing the "commitment point". The cell line HL-60 may also be useful in this regard.

Similarly, the invention has been described using DMSO as the inducing agent. However, other known inducing agents include hexamethylene bis acetamide as an example from the class of planar polar compounds, butyric acid as an example from the class of fatty acids and hypoxanthine from the class of purine and purine derivatives.

In summary, part of the invention disclosed herein relates to an in-vitro method of screening compounds to determine their efficacy for inducing terminal differentiation in FEL cells. Such compounds may form the basis of anticancer drugs if they induce terminal differentiation in pharmaceutically suitable dosages. In its broadest form, the assay comprises treating cells from a hemopoetic cell line which are blockable in terminal differentiation when exposed to medium comprising a differentiation inducer and a differentiation inhibitor for a predetermined period of time. The cells are then removed from the solution and exposed to candidate compounds for a predetermined period of time. Whether or not the candidate compound induced terminal differentiation is determined by detecting any one of several postcommitment events, including for example the formation of hemoglobin, differentiated markers, cessation of DNA replication or cell death just to mention a few.

The method of screening may also include the step of transferring the cells after treatment with the inducer and inhibitor to a solution containing cytochalasin B for a predetermined period of time subsequent to exposure to the candidate compounds. The exposure to cytochalasin B

prevents cell division of all species and therefore provides a means of estimating the number of committed (differentiated) cells without complications introduced by cell replication.

The results and studies summarized herein also indicate that compounds which cause a cell cycle delay in G ₂ or mitosis exhibit an efficacy in inducing terminal differentiation.

Therefore, while the method of screening compounds which forms the subject invention has been described with reference to specific examples of the cell line in which differentiation is being induced, the inducing agents, inhibiting agents and means for detecting postcommitment events, it will be readily apparent to those skilled in the art that there exist numerous variations of these agents and steps in the screening method disclosed herein, and that using any of these variations will not result in a departure from the invention disclosed herein.

The basic experimental protocol is described in the Materials and Methods section. The 3AB/DMS0-treated cells were incubated in the presence of various anticancer drugs for 20h. When DMSO was substituted for the above drugs following the 3AB/DMS0 treatment, there was an average of 37 ± 8.0% benzidine-positive (differentiated) cells. In another control in which cells were treated with 1.5% DMSO for 108 h, there were 95 ± 5% benzidine-positive cells. The results (only one drug concentration is shown) are the means of three separate experiments corrected for the controls (not containing the anticancer or G ₂ /mitosis- blocking drug) (mean = 3.8 ± 1.7%). This value changes by altering factors such as 3AB concentrations, serum, cell titre, and length of initial incubation. In cases in which drugs were dissolved in DMSO, the final concentration of DMSO did not exceed 0.16%, except taxol (0.15%). Corresponding controls contained equivalent amounts of DMSO. Errors are equal to one standard deviation from the mean.

The basic experimental protocol is outlined in Materials and Methods section. Control data applies as outlined in Table 1.

TABLE 3. MITOTIC INDICES OF FEL CELLS TREATED WITH VINCRISTINE FOR 20 H. UNDER CONDITIONS WHICH GAVE DIFFERENTIATION-COMMITMENT

Mitotic cells

Expt 2

1.6

15

15

19 27

The basic protocol is outlined in the Materials and Methods section.

REFERENCES

1. Degos L. (1990) Differentiating agents in the treatment of leukemia. Leukemia Res 14, 717.

2. Sachs L. (1986) The development and reversal of malignancy. Cancer Rev 2, 48.

3. Hozumi M. (1983) Fundamentals of chemotherapy of myeloid leukemia by induction of leukemia cell differentiation. In Advances in Cancer Research (Klein G. & Weinhouse S. Eds), Vol. 38, pp. 121-169. Academic Press. New York.

4. Marks P. A. & Rif ind R. A. (1990) Role of differentiation induction in tumor suppression. In Tumour Suppressor Genes (Klein G. Ed. ) pp. 201-216. Marcel Dekker, Inc., New York.

5. Ebisuzaki K. , Casley W. L. Griffiths A. & Wheaton L. (1991) Temporal mapping of the differentiation pathway of the murine erythroleukemia cell. Cancer Res 51, 1668.

6. Gopalakrishnan T. V. & French-Anderson W. (1979) Mouse erythroleukemia cells. Meth. Enzym. 58, 506.

7. Patterson M. K. Jr (1979) Measurement of growth and viability of cells in culture. Meth. Enzym. 58, 141.

8. Trent J. M. & Thompson F. H. (1987) Methods for chromosome banding of human and experimental tumors in vitro. Meth. Enzym. 151, 267.

SUBSTITUTESHEET

9. Zucker R. M. , Elstein K. H. Easterling R. E. & Massaro E. J. (1989) Flow cytometric comparison of the effects of triakyltins on the murine erythroleukemia cell. Toxicology 58, 107.

10. Brae T. & Ebisuzaki K. (1987) PolyADP-ribosylation and Friend erythroleukemic-cell differentiation: action of poly (ADP-ribose) polymerase inhibitors. Differentiation 34, 139.

11. Marks P. A. & Rifkind R. A. (1978) Erythroleukemic differentiation. Ann Rev. Biochem. 47, 419.

12. Ebert P. S. Wars I. & Buell D. N. (1976) Erythroid differentiation in cultured Friend leukemia cells treated with metabolic inhibitors. Cancer Res. 36, 1809.

13. Sugano H., Furusawa M. , Kawaguchi T. & Ikawa Y. (1973) Enhancement of erythrocytic maturation of Friend virus-induced leukemia cells in vitro. In Unifying Concepts of Leukemia Bibl. Haematology (Dutcher R. M. & Chieco-Bianchi L., Eds), Vol. 39. pp. 943-954. Karger, Basel.

14. Dinnen R. & Ebisuzaki K. (1991) Mitosis may be an obligatory route to terminal differentiation in the Friend erythroleukemia cell. Exp. Cell Res. 191, 149.

15. Kitaura K., Imai R. , Ishihara Y., Yanai H. & Takahira H. (1972) Mode of action of adriamycin on HeLa S-3 Cells in vitro. J. Antibiot 25, 509.

16. Drewinko B. (1980) Cellular pharmacology. In Cancer and Chemotherapy (Crooke S. . & Prestayko A. W. Eds),

^S UBSTITUTESHEET

Vol. 1, pp. 95-121. Academic Press, New York.

17. Dustin P. (1984) Microtubules, Second Edition. Springer, Berlin.

18. Lepoint A., De Paermentier F., Bassleer R. & Desaive Cl. (1975) Dosages cytophotometriques des acides desoxyribonucleques et des proteines totales dans les cellules tumorales d\'Ehrlich traitees par la daunomycine ou 1\'amphotericine B. Ann. Histochim. 20, 101.

19. Wilson L., Bamburg J. R. , Mizel S.B., Grisham L. M. & Creswell K. M. (1974) Interaction of drugs with microtubule proteins. Fed. Proc. Fed. Am. Soc. Exp Biol. 33, 158.

20. Weber K. , Wehland J. & Herzog W. (1976) Griseofulvin interacts with microtubules both in vivo and in vitro. J. Mol. Biol. 102, 817.

21. Radley J. M. & Hodgson G. S. (1971) Effect of isoprenaline on cells in different phases of the mitotic cycle. Exp. Cell Res. 69, 148.

22. Schiff P. B. & Horwitz S. B. (1980) Taxol stabilizes microtubules in mouse fibroblast cells. Proc. natn. Acad. Sci. U.S.A. 77, 1561.

23. Steuer B., Breuer B. & Alonso A. (1990) Differentiation of EC cells in vitro by the fluorescent dye Hoechst 33342. Exp Cell Res. 186, 149.

24. Chabner B. A. & Collins J. M. (1990) Cancer chemotherapy: Principles and Practice. J. B.

Lippincott Company, Philadephia.

25. Chabner B. (1982) Pharmacologic Principles of Cancer Treatment. W. B. Saunders Company, Philadelphia.

26. Speth P. A., Linssen P. C. M., Holdrinet R.S.G. & Haanen C. (1987) Plasma and cellular Adriamycin concentrations in patients with myeloma treated with ninety-six-hour continuous infusion. Clin. Pharmacol. Ther. 41, 661.

27. Pincus S. H. & Wehrly K. (1990) AZT demonstrates anti- HIV-1 activity in persistently infected cell lines: Implications for combination chemotherapy and immunotherapy. J. Infect. Dis. 162, 1233.

28. Bittles A. H., Baum H. & Monks N. J. (1988) Differential growth inhibition of human diploid fibroblasts by 2-deoxyglucose and antimycin-A with ageing in vitro. Gerontology 34, 236.

29. Barlogie B., Drewinko B., Schumann J. & Freireich E.J. (1976) Pulse cytophotometric analysis of cell cycle perturbation with bleomycin in vitro. Cancer Res. 36, 1182.

30. Tobey R. A. (1975) Different drugs arrest cells at a number of distinct stages in G ₂ Nature 254, 245.

31. Nomura S. & Oishi M. (1983) Indirect induction of erythroid differentiation in mouse Friend cells: Evidence for two intracellular reactions involved in the differentiation. Proc. natn. Acad. Sci. U.S.A. 80, 210.

SUBSTITUTESHEET

32. Terada M., Mudel U., Fibach E., Rifkind R. A. & Marks P.A. (1978) Changes in DNA associated with induction of erythroid differentiation by dimethyl sulfoxide in murine erythroleukemia cells. Cancer Res. 38, 835.

33. Scher W. & Friend C. (1978) Breakage of DNA and alterations in folded genomes by inducers of differentiation in Friend erythroleukemia cells. Cancer Res. 38, 841.

34. Zucchetti M. , Catapano C. V., Filippeschi S., Erba E. & D\'lncalci M. (1989) Temozolomide induced differentiation of K562 leukemia cells is not mediated by gene hypomethylation. Biochem. Pharmacol. 38, 2069.

35. Fearon E. R., Burke P.J., Schiffer C.A., Zehnbauer B. A. & Volgelstein B. (1986) Differentiation of leukemia cells to polymorphonuclear leukocytes in patients with acute nonlymphocytic leukemia. New Engl. J. Med. 315, 15.

36. Beran M. , Hittelman W. N., Andersson B. S. & McCredie K. B. (1986) Induction of differentiation in human myeloid leukemia cells with cytosine arabinoside. Leukemia Res. 10, 1033.

SUBSTITUTE SHEET