CA2216559A1 - G2 checkpoint inhibitors and assay - Google Patents

Abstract

Compounds are provided that inhibit the G2 checkpoint in the cell cycle and are useful for sensitizing G1 deficient cancer cells to killing by DNA damaging agents.
Also provided is an assay for G2 checkpoint inhibitors which includes applying a DNA damaging agent to p53 deficient cells and determining whether such cells do not arrest at G2 and proceed to mitosis after application of a sample to be tested. The assay may employ mitotic cell specific monoclonal antibodies. Compounds of this invention have the general formula:

wherein R5 is selected from the group comprising:

Description

CA 02216~9 1997-09-2~

Field of Invention This invention relates to substances which inhibit arrest of the cell cycle during the G2 phase and which are useful for sensitizing cancer cells to the effect of DNA
damaging agents. This invention also relates to an assay for such substances.

Background Normal cells respond to DNA damage in one of two ways, depending upon their type and the degree of damage. The cells may activate an apoptotic pathway leading to suicide of the cell and its removal or, survival of a damaged cell may be promoted by activating checkpoints that temporarily halt the normal cycle of growth and division to allow time for DNA repair. The checkpoints operate during the G1 phase of the cycle so that DNA is repaired before it is replicated in the S phase; and, during the G2 phase so that DNA is repaired before chromosomes are segregated in the mitosis phase (M).
Few G2 checkpoint inhibitors are known. Two groups have been found: the purine analogues (caffeine, pentoxifylline, 2-aminopurine, 6-dimethylaminopurine), and staurosporine with its derivative UCN-01 (7-hydroxystaurosporine). In addition, the protein phosphatase inhibitors (okadaic acid and fostriecin) can act as G2 checkpoint inhibitors but they also induce premature mitosis in the absence of DNA damage and have tumour-promoting activity.

CA 02216~9 1997-09-2~

Experiments employing cells deficient in the tumour suppressor protein p53 have demonstrated the value of the two groups of G2 checkpoint inhibitors described above, for selectively sensitizing cancer cells. Pentoxifylline has been shown to enhance cisplatin induced killing of p53~ MCF-7 cells 30-fold and radiation induced killing of p53~ A549 human lung adenocarcinoma cells 5-fold. Caffeine has been shown to enhance radiation induced killing of p53~ mouse embryonic fibroblasts 3-fold and radiation induced killing of p53~ A549 human lung adenocarcinoma cells 5-fold. UCN-01 has been shown to enhance cisplatin induced killing of p53~
MCF-7 cells 25-fold. UCN-01 is active in vitro as a G2 checkpoint inhibitor in the submicromolar range.
Over 50~ of human cancers exhibit a loss of function of the protein p53. Cells with mutated p53 are unable to activate the G1 checkpoint in response to DNA damage.
However, the G2 checkpoint (although usually weaker than in normal cells) still provides an opportunity to repair the DNA damage before cell division. Inhibition of the G2 checkpoint alone would have little effect on normal or cancer cells. However, G2 checkpoint inhibitors used in combination with a DNA damaging agent would increase the killing of cancer cells which cannot activate the G1 checkpoint.
The known G2 checkpoint inhibitors have been found serendipitously. An assay suitable for high throughput screening of compounds for G2 checkpoint inhibition activity would be desirable.

CA 02216~9 1997-09-2 Summary of The Invention In one aspect, this invention provides G2 checkpoint inhibitor compounds having the formula of Compound I as defined herein, including pharmaceutically acceptable salts thereof.
In another aspect, this invention provides the use of compounds having the formula of Compound I to inhibit the G2 checkpoint, including; release of cells that are arrested at the G2 checkpoint thereby permitting the cells to proceed to mitosis; and, to prevent G2 checkpoint arrest in cells (for example, in response to DNA damage). This invention also provides the use of compounds of the formula Compound I to sensitize cancer cells to the effects of DNA
damaging agents and the use of such compounds in the formulation of agents, including medicaments, for potentiating the effect of DNA damaging agents on cancer cells. This invention also provides the use of compounds having the formula of Compound I to increase the killing of cancer cells by DNA damaging agents.
This invention provides a method of increasing the killing of cancer cells having G1 checkpoint deficiency as a result of DNA damage, comprising the steps of administering a DNA damaging agent to said cancer cells thereby damaging DNA of said cells, and administering a G2 checkpoint inhibitor compound to said cells, wherein the G2 inhibitor compound is a compound or a pharmaceutically acceptable salt thereof having the formula of Compound I.

CA 02216~9 1997-09-2~

In a further aspect, this invention provides an assay for G2 checkpoint inhibitors. This invention provides a method for determining whether a sample contains a G2 checkpoint inhibitor, comprising the steps of:

(a) applying a DNA damaging agent to p53 deficient cells which will arrest at G2 in response to the DNA damaging agent;

(b) applying a sample to be tested and a microtubule depolymerizing agent to said cells;

(c) culturing said cells for a time sufficient for the cells to arrest at G2 in response to the DNA
damaging agent; and (d) subsequent to step (c), lysing said cells to provide an extract, and (e) applying a mitotic cell specific antibody to said extract and measuring antibody binding of said antibody to epitopes in the extract.

In the assay method, the sample may be applied at about the same time as the DNA damaging agent, or applied after the cells are arrested at G2. The measurement of antibody binding in step (e) may be compared to a measurement of antibody binding with cells arrested at G2, CA 02216~9 1997-09-2~

or the measurement of antibody binding in step (e) may be compared to a measurement of antibody binding with cells in a method wherein in step (b), a known G2 checkpoint inhibitor is applied to the cells.
In the assay, the antibody may be a monoclonal antibody specific for phospho-epitopes. The monoclonal antibody may be TG-3 and in step (e) of the method, TG-3 not bound to epitopes in the extract may be removed and an antibody capable of binding to TG-3 is applied with binding of TG-3 to said epitopes being measured by detecting production of a product of a reaction catalyzed by an enzyme linked to the TG-3 binding antibody.
Throughout this specification, the term "G2" or G2 phase" means the phase of the cell cycle between the end of DNA synthesis and the beginning of mitosis. A cell in G2 has an interphase nucleus as determined by microscopy, and duplicated DNA (usually determined by flow cytometry).
Throughout this specification, the term "G2 checkpoint inhibitor" means a substance which is capable of releasing a cell from arrest in G2 phase brought about by DNA damage, preventing a cell from arrest in G2 phase in response to DNA damage, or both.
Throughout this specification, the term "DNA damaging agent" means any substance or treatment which induces DNA
damage in a cell, including W irradiation, gamma irradiation, X-rays, alkylating agents, antibiotics that induce DNA damage by binding to DNA, inhibitors of topoisomerase and any compound used in chemotherapy which CA 02216~9 1997-09-2~

acts by causing DNA damage. Examples of specific compounds are cisplatin, VM-26, and procarbazine.

Brief Description of the Drawinqs In Fig. 1, Figs lA-E show flow cytometry analysis at various times following DNA damage, with caffeine (Fig. lD) and UCN-01 (Fig. lE) as G2 checkpoint inhibitors.
Figs. 2 and 3 are graphs showing the mitotic index of irradiated cells treated with or without caffeine or UCN-01 plotted against time and drug concentration.
Fig. 4 is a graph plotting the linear relationship between mitotic index and the optical measurements from an ELISA assay as described herein.
Fig. 5 is a histogram showing the mitotic index of cells subjected to gamma irradiation (~) or VM-26 treated with nocodazole as a control (-), nocodazole plus debromohymenialdisine (H), and nocodazole plus caffeine (C).
Fig. 6 is a graph showing potentiation of cytotoxicity of cisplatin by UCN-01 and debromohymenialdisine.

Detailed Description The assay of this invention involves treating cancer cells with a DNA damaging agent under conditions whereby arrest of the cells at G2 will be induced. A sample to be tested for G2 checkpoint inhibition activity and an agent which arrests cells in mitosis are applied to the cells, following which it is determined whether the G2 checkpoint CA 02216~9 1997-09-2~

is overcome and the cells enter mitosis. Release of the cells from arrest at G2 or prevention of arrest at G2, such that the cells proceed to mitosis, is the indicator of G2 checkpoint inhibition activity.
The assay of this invention requires the use of a cancer cell culture in which the cells are incapable of arrest at the G1 checkpoint in response to DNA damage. The cells may be from any of the numerous cancer cell lines for which it is known that the cells are incapable of arrest at G1. Such cells include cells which are p53 deficient, including: any cell line with a natural mutation or deletion in the p53 gene which renders p53 inactive; any cell line expressing a viral oncogene which disrupts p53;
any cell line expressing dominant-negative mutant p53 (some mutant p53 proteins inhibit wild-type protein such as in the p53~ MCF-7 cells used in the examples herein); and any primary or established cell line derived from p53 "negative transgenic" animals. Specific examples of suitable cancer cells are the following: cancer cells which have incorporated the human papillomavirus type-16 E6 gene, cell lines which are mutant for p53 function including Burkitt's human lymphoma cell line CA46 and the human colon carcinoma ~ell line HT-29 (CA46 and HT-29 are available from the American Type Culture Collection). Cell lines with a genetic deficiency which disrupts the G1 checkpoint other than by disruption of p53, may also be employed in this assay.

CA 02216~9 1997-09-2~

Once a cell line incapable of G1 arrest is chosen, conditions for arresting a majority of the cells at G2 in response to a DNA damaging agent are optimized by determining appropriate culture conditions, incubation time, and type and dosage of DNA damaging agent.
Preferably, at least 50~ of the cells in a culture will be arrested at G2 in response to the DNA damaging agent.
Maximizing the proportion of cells in the population which are arrested at G2 will reduce the background signal.
Once the conditions for G2 arrest in the cell culture are established, the assay may be carried out in one of two ways. First, the cells may be arrested at G2 in response to the DNA damaging agent and then treated with the sample to determine whether there is release from G2 arrest or, the cells may be treated with the sample prior to the time when the majority of the cells would be expected to be arrested at G2 and determine whether the cells are prevented from G2 arrest.
Release from G2 arrest or prevention of G2 arrest is detected by a quantitative determination of the cells which proceed to mitosis. The assay culture is treated with a agent which will arrest such cells in mitosis. Such agents include microtubule depolymerizing agents that arrest cells in metaphase, such as nocodazole. This will increase the number of mitotic cells in the sample which will be detected.
In this assay, determination of the cells which proceed to mitosis is carried out using any of the known CA 02216~9 1997-09-2~

immunological methods by employing antibodies which have specificity for mitotic cells. Monoclonal antibodies demonstrating such specificity are known and include MPM-2 which was raised against mitotic HeLa cells and recognizes phospho-epitopes that are highly conserved in mitotic proteins of all eukaryotic species. Other examples are the monoclonal antibodies recognizing phospho-epitopes in the paired helical filament proteins (PHF) found in brain tissue of patients suffering from Alzheimer's disease as described in: PCT International Application published July 4, 1996 under No. WO 96/20218; and, Vincent, I. et al.
(1996) "Mitotic Mechanisms in Alzheimer's Disease?" The Journal of Cell Biology, 132:413-425. The examples in this specification make use of the antibody TG-3 described in the latter two references, which may be obtained from Albert Einstein College of Medicine of Yeshiva University, Bronx, New York.
TG-3 has a high specificity for mitotic cells. When TG-3 is used in the ELISA assay described herein, a very good signal/noise ratio is achieved permitting the ELISA
assay to be easily monitored by optical absorbance.
Dying cells exhibit some characteristics which are similar to mitotic cells. TG-3 doe5 not rea~t with dying cells which prevents apoptotic cells from being counted in the assay of this invention. The efficiency of the assay is not affected by the presence of dying cells.
Immunological methods useful for determination of mitotic cells in this assay include any method for CA 02216~9 1997-09-2~

determining antibody-antigen binding, including:
immunocytochemistry (eg. immunofluorescence), flow cytometry, immunoblotting, and ELISA. Several immunological methods are described in detail in the examples herein as well as in Vincent I. et al. [supra] .
Other immunological procedures not described herein are well-known in the art and may be readily adapted for use in this assay. However, high throughput testing of samples may best be achieved by use of ELISA.
The assay of this invention has been used to screen crude extracts from approximately 1,000 marine organisms.
As described in more detail in the examples herein, extracts from a marine sponge tested positively in the assay and analysis of the extracts resulted in the identification of a family of compounds having G2 checkpoint inhibition activity. Two such inhibitor compounds obtained from the sponge: hymenialdisine and debromohymenialdisine, have been previously isolated from sponge tissue and are known to have some cytotoxic activity and activity as alpha-adrenoceptor blockers. These compounds are now shown to be potent inhibitors of the G2 checkpoint and may be used to selectively sensitize cancer cells to the effect of DNA damaging agents thereby potentiating cytotoxity of DNA damaging agents. The compounds of this invention may be administered in conjunction with DNA damaging agents to increase killing of cancer cells which are deficient in the G1 checkpoint.

The G2 checkpoint inhibitor compounds of this invention have the formula:
z Rs Y~U Compound I
X

wherein R5 is selected ~om the grou~ co~

N~( N~ N~=( o~NR2 ~NR2 ~NR2 A ~ B '~ C '~

in which U is selected ~om the group CO

W' D 5 1 I N--(CH2)n ~~
R
W
E 5 1I N--CH=CH--~~
Rl F 5 CH=N--CH= ~~

W

G S 1I N=CH--CH= lo CA 022l6~9 l997-09-2 and wherein:
X = NR4 (R4 = H, or alkyl including straight chain, branched and cyclic aliphatic, optionally substituted with alkyl, alkene, aklyne, hydroxyl, carbonyl, aryl, amine, halogen, nitro, cyano, thio or sulfate) O, or S;
Y = H, F, Cl, Br, or I;
Z = H, F, Cl, Br, or I;
W = O, or H2;
R1 and R2 independently = H, or alkyl (as defined for R4);
R3 = H, alkyl (as defined for R4), or acyl (including aroyl); and n = 2 O.
This invention includes the E and Z configurations at the R5 linkage when that linkage is a double bond. The Z configuration is preferred.
Preferably, the alkyl and acyl substituents of the compounds of this invention will be from one to ten carbon

2 0 atoms and n = 0, 1, or 2 .
More preferably, n = 1, the alkyl substituents will be methyl, ethyl or benzyl with benzyl being optionally substituted with OH, halogen, nitro, carboxylic acid, carboxaldehyde, methyl, ethyl, methoxy, sulfonyl or cyano;
25 the acyl substituents will be from two to four carbon atoms or benzoyl; and Y and Z will be H, Cl or Br.
When X = NH, Y = H or Br, Z = H, R5 = A
(Z configuration), W = O, R1 = H, R2 = H, R3 = H, and n = 1 CA 02216~9 1997-09-2~

in Compound I, the compound is hymenialdisine (Y = Br) or, debromohymenialdisine (Y = H).
While the examples in this specification describe purification of compounds from a natural source, the compounds of this invention can be synthesized by routine modification of the known methodology for total synthesis of hymenialdisine and debromohymenialdisine such as that described in: H. Annoura and T. Tatsuoka (1995) Tetrahedron Letters, 36:413-416 (scheme 1 at page 414), which used pyrrole-2-carboxylic acid as starting material, with an amide intermediate produced by reacting the starting material with an H2N(CH2)nCOOMe amino ester. The X, Y, and Z variations of the compounds of this invention arise from using substituted pyrroles (X = NR), furans (X
= O) or thiophenes (X = S) as the starting material.
Variations in "n" arise from using different H2N(CH2)nCOOMe amino esters in the reaction with the starting material to produce the amide intermediate. The latter agent may have to be modified by using a protected version of the ester for some values of "n" in order to prevent polymerization of the amino ester. W variations arise from reduction of the amide intermediate using LAH followed by reoxidation and esterification of the side chain. R1, R2, and R3 variations arise from alkylation and acylation reactions on the starting material (for R2, R3) or, on the amide intermediate (for R1).
Compounds of this invention, including pharmaceutically acceptable salts thereof (eg. HCl salts) CA 022l6~9 l997-09-2 are sufficiently water-soluble that the compounds may be administered in vivo in aqueous form. A preferred mode of administration to an animal (including human) would be intervenous with the goal being to achieve a circulating concentration of the compound in the patient of about 10-5 -10-4 molar.

EXAMPLE I

Cell Culture: Human m~mm~ry carcinoma cells (MCF-7) expressing a dominant-negative p53 mutant (p53~ MCF-7) as described in Fan. s., et al. (1995) "Disruption of p53 Function Sensitizes Breast Cancer MCF-7 Cells to Cisplatin and Pentoxifylline", Cancer Research 55:1649-1654, were grown in monolayer cultures at 37~ in humidified 5~ C02 in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10~ fetal calf serum, 100 U/ml penicillin and streptomycin, 2 mM L-glutamine, 1 mM sodium pyruvate, 1~ Modified Eagle Medium (MEM), non-essential amino acids, and 1 ng/ml human epidermal growth factor (all from Gibco, and 1 ~/ml bovine insulin, 1 ~g/ml hydrocortisone, and 1 ng/ml ~-estradiol (from Sigma). To arrest cells in mitotic metaphase, cultures were treated with 50 ng/ml nocodazole (Sigma) from a 1000-fold stock in dimethylsulfoxide stored at -20~C.

Immunofluorescence Microscopy: Cells were grown on coverslips coated with poly-L-lysine (Sigma), centrifuged at 1000 g for 5 minutes, fixed with 3~ formaldehyde in CA 022l6~9 l997-09-2~

Tris-buffered saline (TBS: 10 mM Tris-HCl pH 7.4, 0.15 M NaCl) for 40 minutes at 4~C, followed by ice-cold 100~
methanol for 5 minutes and then rinsed twice with TBS. The coverslips were blocked in 3~ dried milk (Carnation brand) in TBS for 1-2 hours at 4~C and then incubated with TG-3 hybridoma culture supernatant ( 15 ~g/ml IgM) diluted 1/10 in 3~ milk in TBS for 1 hour at room temperature.
After 2 rinses in TBS, the coverslips were incubated with CY3-conjugated goat anti-mouse secondary antibody (Jackson Immunoresearch Laboratories, West Grove, Philadelphia:
Cat. #115-165-006) diluted 1/500 in 3~ milk in TBS for 1 hour at room temperature. They were then rinsed in TBS, stained with the DNA dye Hoechst 33258TM (Sigma), mounted on slides in 10~ TBS in glycerol containing 0. 2 M n-propyl gallate, and photographed on Kodak TMax 400TM film with a Zeiss AxiophotTM microscope.

Preparation for Cell Sorting: Cultured cells were detached using trypsin, and fixed in 3~ formaldehyde in TBS for 40 minutes at 4~C, followed by 70~ ethanol for 5 minutes at 4~C. The cells were blocked in 1~ bovine serum albumin (Sigma) in TBS (B-TBS) for 1 hour at 4~C and then incubated overnight at 4~C with TG-~ hybridoma culture ~upernatant diluted 1/10 in B-TBS. After two rinses in TBS, they were incubated for 1 hour at 4~C with fluorescein-isothiocyanate (FITC)-conjugated goat anti-mouse secondary antibody (Pierce, Rockford, Illinois; Cat. #31560) diluted 1/100 in B-TBS, rinsed twice in TBS and then resuspended in 5 ~g/ml CA 022l6~9 l997-09-2 PI in TBS. The cells were subjected to flow cytometry for sorting.

Preparation for Cell Cycle Analysis: To determine the percentage of cells in each of the G1, S, G2, and M phases of the cell cycle, cultured cells were incubated for 20 minutes with 30 ~M debromodeoxyuridine (BrdUrd). The cells were detached using trypsin, fixed in 70~ ethanol for 30 minutes at 4~C, treated with 2 N HC1 in 0.5~ Triton 0 X-lOOTM at room temperature for 30 minutes, followed by 0.1 M Na2B4O7, pH 8.5 for 5 minutes. After 1 hour at 4~C in 0.5~
TweenTM 20 and 1~ bovine serum albumin in TBS (TB-TBS), they were incubated at room temperature for 30 minutes with monoclonal anti-BrdUrd antibody (Becton Dickinson, San Jose, California; Cat #347580) diluted 1/3 in TB-TBS, to which was then added TG-3 hybridoma culture supernatant diluted 1/10 in TB-TBS and the cells were left overnight at 4~C. After two rinses in TBS, the cells were incubated in FITC-conjugated goat anti-mouse secondary antibody diluted 1/25 in TB-TBS for 2 hours at 4~C. Cells were then rinsed twice in TBS, resuspended in 5 ~g/ml PI in TBS and subjected to flow cytometry.

Flow Cytometry: A Coulter Epics Elite ESPTM flow cytometer (Coulter Corp., Miami, Florida) equipped with an EnterpriseTM ion laser (Coherent, Santa Clara, California) with an output of 400 mW at 488 nm was used. Forward and side scatter were measured simultaneously and used to gate CA 02216~9 1997-09-2~

out debris. Time of flight was used to gate for single cells. A minimum of 20,000 gated cells was collected.
FITC fluorescence was collected using a 525 nm bandpass filter, and propidium iodide fluorescence was collected using a 610 nm longpass filter. All data was saved in listmode for further gating and analysis. TG-3 positive cells were sorted according to fluorescence levels determined empirically. For sorting, a minimum of 50,000 cells was collected and 5,000 were reanalysed by flow cytometry to determine the purity of the sort.

Mitotic Index: Cycling cells stained both with the monoclonal antibody TG-3 using indirect immunofluorescence and with the DNA dye Hoechst 33258TM were examined to determine their mitotic stage. In interphase cells, TG-3 staining is extremely weak and is restricted to small speckles within the nucleus. Cells in mitosis show a dramatic increase in staining. Prophase cells, in which the DNA has condensed into chromosomes but the nuclear lamina has not yet broken down, show intense staining throughout the nucleus. The intense staining is maintained in metaphase cells but is also spread throughout the cytoplasm. Staining is particularly strong around the chromosomes. Cytoplasmic staining is reduced slightly in anaphase cells and further reduced in telophase cells. In telophase cells, discrete speckles of fluorescence reappear both within the nucleus and extranuclearly, but the latter disappear after cytokinesis. The relative intensity of CA 022l6~9 l997-09-2 TG-3 immunofluorescence in interphase and mitotic cells can be seen by comparing these cells (for example, in the same photograph) and is at its most extreme in metaphase.
When cycling cells stained with TG-3 antibody and propidium iodide are subjected to flow cytometry, a histogram of the fluorescence emitted from propidium iodide shows the standard distribution of cells consisting of a peak of G1 cells with diploid DNA, a diffuse plateau of S phase cells with increasing fluorescence, and a peak of cells in G2+M in which the mean fluorescence intensity is double that of the G1 peak. A histogram of FITC
fluorescence associated with TG-3 immunoreactivity shows a major peak of cells with low fluorescence and a minor peak of cells with about 50-fold higher fluorescence. Dual parameter analysis of FITC and propidium iodide shows that the subpopulation with intense FITC fluorescence is within the G2+M phase. Such TG-3 positive cells represented 2.2 + 0.3~ of the total cell population, comparable to the value of 3.16 + 0.9~ mitotic cells obtained independently from the same samples by fluorescence microscopy using chromosome condensation as the marker for mitosis. The TG-3 positive subpopulation can be seen most clearly in nocodazole-treated samples. Cells previously treated with nocodazole for 4 hours before antibody staining when subjected to flow cytometry gave 17.3~ _ 0.9~ TG-3 positive cells which agrees well with the value of 19.7 + 2.9~
mitotic cells obtained by microscopy. Similar results are obtained with cells arrested in mitosis with other CA 02216~9 1997-09-2~

antitubulin agents, including demecolcine, hemiasterlin, and hemiasterlin A.
Cycling and nocodazole-treated preparations were subjected to cell sorting based on their TG-3 immunofluorescence. The sorted fractions where subjected to flow cytometry, giving 94.6~ and 96.8~ TG-3 positive cells. Samples of the sorted TG-3 positive fractions were prepared for microscopy and the percentage of mitotic cells determined to be 93.5% and 95.0~ respectively. The TG-3 positive fraction sorted from cycling cells comprised 13.8~
prophase, 73.1~ metaphase, 6.9~ anaphase, 0~ telophase, and 6.2~ interphase cells, while the fraction sorted from nocodazole-treated cells consisted almost exclusively of cells in metaphase, with 2~ prophase, 94~ metaphase and 4~
interphase. Microscopic ex~m~n~tion of 150 mitotic cells from a cycling population revealed 16~ prophase, 81 metaphase, and 2~ telophase.
Although the TG-3 antigen is present throughout interphase at a very low level as nuclear speckles, high levels of antigen are found only in mitotic cells. This dramatic difference can be used to distinguish mitotic cells from nonmitotic cells. 1~ mitotic cells may be detected with accuracy comparable to that of conventional microscopy.

The TG-3 antibody and cell cycle analysis: As the ability to determine the proportion of cells in each of the G1, S, G2 and M phases of the cell cycle would be useful, the CA 02216~9 1997-09-2 above-described flow cytometry protocol was used in which standard methods using antibodies to incorporated BrdUrd to identify S phase cells and propidium iodide labelling of total DNA to identify Gl and G2+M phase cells was combined 5 with TG-3 antibody binding to distinguish G2 and M phase cells. Cycling cells that had incorporated BrdUrd for 2 0 minutes were fixed and their DNA was denatured. The cells were then incubated with either mouse monoclonal TG-3 antibody, or mouse monoclonal antibody to BrdUrd, or both, followed by FITC-con~ugated goat anti-mouse secondary antibodies. Propidium iodide was used to label total DNA.
Dual parameter analysis of the fluorescence emitted by FITC
versus propidium iodide was carried out. The antigen recognised by TG-3 was able to withstand the acid 15 denaturation required for detection of S phase cells.
Cells stained with TG-3 alone show the same pattern as described above for cycling cells not exposed to BrdUrd.
The mitotic subpopulation of the G2+M phase showing intense FITC staining, indicating that 1~ of the cells are mitotic.
20 Antibodies to BrdUrd alone show a typical horseshoe-shaped pattern of FITC-staining cells spanning the DNA content range of diploid to tetraploid with 35~ in Gl, 50~ in S, and 15~ in G2+M. When both TG-3 antibody and antibody to BrdUrd are used, two discrete FITC-positive populations are 25 observed, one like that obtained with antibodies to BrdUrd alone and another that is more intense, like that obtained with TG-3 antibodies alone. The percentage of cells in the different phases of the cells cycle agrees with those CA 02216~9 1997-09-2~

observed with TG-3 antibody or antibody to BrdUrd alone:
29~ in G1, 54~ in S, 16~ in G2, and 1~ in M.

EXAMPLE II

Preparation of Sponge Extracts: Specimens (87 g, wet) of Stylissa flabelliformis (PNG94-51: S. flabelliformis Hentschel, 1912; Class Demospongiae, Order Halichondrida, Family Axinellidae; registered as ZMA POR, 11416) were thawed and extracted exhaustively with MeOH (250 ml x 5, each about one day). The MeOH extract was filtered and concentrated in vacuo to give a dark brown solid (~5 g).
An ageous solution of approximately 3 g of the solid was chromatographed on a SephadexTM LH-20 column using MeOH as the eluent, yielding nine fractions which were subjected to the G2 checkpoint inhibitor assay described below. Samples (10 ~g/ml) of three of the fractions gave positive results and those factions were further purified as described below. Each purification step was monitored using the G2 checkpoint inhibitor assay.

G2 Checkpoint Inhibition Assay: p53~ MCF-7 cells as described in Example I are cultured as monolayers in DMEM
supplemented with 10~ fetal bovine serum, 2 Mm L-glutamine, 50 units/ml penicillin, 50 ~g/ml streptomycin, 1 Mm sodium pyruvate, MEM non-essential amino acids, 1 ~g/ml bovine insulin, 1 ~g/ml hydrocortisone, 1 n/ml human epidermal growth factor, and 1 ng/ml ~-estradiol at 37~C in CA 02216~9 1997-09-2~

humidified 5~ C02. The cells are seeded at 10,000 cells per well of 96-well polystyrene tissue culture plates (Falcon) in a volume of 200 ~l cell culture medium. The cells are allowed to grow for 24 hours and irradiated with 6.5 Gy using a 60Co source (Gammacell 200TM, Atomic Commission of Canada) delivering ~-rays at a dose rate of 1.4 Gy/min.
Immediately after irradiation, 100 ng/ml nocodazole is added (Sigma, from a 1000-fold stock in DMSO) together with samples to be tested at about 1-10 ~g/ml (from 1000-fold stocks in DMSO). The cells are incubated for 20 hours (G2 arrest preventer assay). Alternatively, nocodazole and extracts may be added 16 hours after irradiation and the cells incubated for 8 hours (G2 arrest releaser assay).
Caffeine at 2 mM (Sigma, from a 100 mM solution in PBS) may be used as a positive control. After treatment, the cell culture medium is removed completely and the 96-well tissue culture plates frozen at -70~C for 1-14 hours). The frozen cells are thawed by addition of 100 ~l of ice-cold lysis buffer (1 mM EGTA, Ph 7.4, 0.5 mM PMSF) and lysed by pipeting up-and-down 10 times. Cell lysates are transferred to 96-well PolySorpTM Elisa plates (Nunc) and dried completely by blowing warm air at about 37~C with a hair drier positioned about 3 feet above the plates.
Protein binding sites are blocked by adding 200 ~l per well of TBSM (10 mM Tris HCl pH 7.4, 150 mM NaCl, 0.1 mM PMSF,

3~ (w/v) dried non-fat milk (Carnation)) for 1 hour at room temperature. The blocking medium is removed and replaced with 100 ~l TSM containing 0.1-0.15 ug/ml TG-3 monoclonal CA 02216~9 1997-09-2 antibody and horseradish peroxidase-labelled goat anti-mouse IgM (1021-05, Southern Biotechnology Associates) at a dilution 1/500. After overnight incubation at 4~C, the antibody solution is removed and the wells rinsed 3 times with 200 ~1 rinse buffer (10 mM Tris Hcl Ph 7.4, 0.02~ TweenTM 20). 100 ~1 ABTS buffer (120 mM Na2HPO4, 100 mM citric acid, pH4.0) containing 0.5 mg/ml 2,2'-azino-bis (3-ethylbenzthiazoline-6-sulfonic acid) and 0.01~ hydrogen peroxide added for 1 hour at room temperature. The plates are read at 405 nm using a BioTekTM plate reader. Caffeine positive controls give absorbance readings of about 1.0, corresponding to about 60~ mitotic cells.

Purification and Characterization of Active Compounds: The sponge extract fractions exhibiting G2 checkpoint inhibitor activity (typical absorbance reading in assay = 0.5 - 1.0) were individually preabsorbed onto NP silica gel columns and subjected to silica-gel flash chromatography using stepwise gradient elution (100~ CH2Cl2 to 50~ CH2Cl2:MeOH
saturated with NH3 (g)). Further purification was achieved by repeated fractionation on reversed-phase HPLC using 80:20:0.05 H2O/MeOH/CF~COOH (TFA) as eluent. One pure product was debromoaxinohydantoin which was inactive as a G2 checkpoint inhibitor. Two pure products, recovered as TFA salts were converted to the hydrochloride salts by acidification with 3N HCL and concentrated i~ vacuo. The latter two products were active G2 checkpoint inhibitors (typical absorbance reading in assay = 1.0 - 1.5). The CA 02216~9 1997-09-2 products were identified as debromohymenialdisine and hymenialdisine and were present in the SephadexTM fractions as taurine salts. An inactive cleavage product, debromopyrrololactam was obtain in pure form from one of the active SephadexTM fractions by subjecting the fraction to a 5% K2CO3 solution, followed by concentration in vacuo and then reversed-phase HPLC using 80: 20 H2O/MeOH as eluent.

EXAMPLE III

Correspondence between microscopic and ELISA determination of mitotic index: p53~ MCF-7 cells grown in 10 cm diameter cell culture dishes were treated with 0-200 nM nocodazole for 16 hours. The cells were harvested by trypsinization, washed once with PBS and a small sample processed for microscopic determination of the percent mitotic cells as follows: cells were swelled in hypotonic medium (75 mM
KCl), fixed with methanol:acetic acid (3:1), spotted onto microscope slides, stained with bisbenzimide and observed using a ZeissTM standard microscope (Guo et al. (1995). At least 300 cells were counted for each sample. The remaining cells were lysed in ice-cold lysis buffer (1 mM
EGTA, pH 7.4, 0.5 mM PMSF). The protein concentration of the cell lysates was determined using a BioRadTM protein assay and BSA as a standard, and the extract was kept at -70~C as aliquots. Volumes of cell extracts corresponding to 10,000 cells were processed for ELISA according to the method described in Example II. The standard curve CA 02216~9 1997-09-2 (Fig. 4) shows a linear response which correlates highly (r~0.99) with results obtained with the microscopy analysis.

Analy~is by flow cytometry and microscopy: p53~ MCF-7 cells were cultured and irradiated as described in Example II.
After irradiation, ~90~ of the cells arrest in G2/M within 16 hours as determined by flow cytometry, and the cells remain arrested for an additional 8 hours. Fig. 1 shows analysis by flow cytometry of the cells irradiated at 0 hours (Fig. lA) and left for 16 hours (Fig. lB). At 16 hours, cell samples were left untreated (Fig. lC), or were treated with 2 mM caffeine (Fig lD) or 100 nM UCN-01 (Fig. lE). After 8 hours, most of the treated cells had gone from G2/M to G1 but not the untreated cells.
Flow cytometry per se cannot distinguish between cells in G2 and M. For analysis by microscopy, cells were irradiated at 0 hours, nocodazole (see Example II) was added immediately and cell samples were harvested at different time for microscopic determination of the percentage of cells in mitosis showing condensed metaphase chromosomes (mitotic index). At either 0 or 16 hours, 2 mM
caffeine wa~ added to ~ome sample~ and the mitotic index determined at various times. The results shown in Fig. 2 show that no cells entered mitosis in the 24 hours following DNA damage (open circles), but cells treated with caffeine at the time of irradiation (checkpoint inhibitor assay; closed circles) and at 16 hours (checkpoint releaser CA 02216~9 1997-09-2 assay; closed rectangles) entered mitosis within 8 hours following treatment. In a parallel procedure shown in Fig. 3, cells were irradiated at 0 hours, and at 16 hours nocodazole and different concentration of caffeine (circles) and UCN-01 (rectangles) were added. The mitotic index was measured 8 hours later. Within a 16-24 hour post-damage period, the majority of cells can be made to enter mitosis by a G2 checkpoint inhibitor while a negligible number of cells do so in the absence of the inhibitor.

EXAMPLE IV

G2 Checkpoint Inhibition by Debromohymenialdisine: Using the assay procedure described in Example II, p53~ MCF-7 cells were exposed to 6.5 Gy of ~-irradiation; or, to 1 ~M
VM-26 (Bristol-Myers) for 100 minutes. After 16 hours, 100 ng/ml nocodazole was added to all samples. Controls received no further treatment. Other samples also received caffeine or debromohymenialdisine with the nocodazole.
Eig. 5 shows the results obtained 8 hours after treatment with no further addition (-); 20 ~g/ml debromohymenialdisine (H); and, 2 mM caffeine (C).

EXAMPLE V

Potentiation of the Cytotoxicity of a DNA-Damaging Agent by Debromohymenialdisine: p53-MCF-7 cells were plated at 1000 CA 02216~9 1997-09-2 cells per well of a 96-well plate and exposed to different concentrations of cisplatin for 2 hours. The cisplatin was washed away and fresh medium containing: (with reference to Fig. 6) 2 0 ~g/ml debromohymenialdisine (filled circles);
100 nM UCN-O1 (closed triangles); or, nothing (open circles). The cultures were grown for 5 days after which cell survival was measured optically. Average and S.D. of triplicate measurements are shown in Fig. 3.
Although various aspects of the present invention have been described in detail, it will be apparent that changes and modification of those aspects described herein will fall within the scope of the appended claims. All publications referred to herein are incorporated by reference.

Claims