WO2000021506A2

WO2000021506A2 - Pharmaceutical composition of glutathione modulators with antimony and/or arsenic for cancer therapy

Info

Publication number: WO2000021506A2
Application number: PCT/CA1999/000949
Authority: WO
Inventors: Wilson Miller; Gerald Batist; Kelly Davison
Original assignee: Mcgill University
Priority date: 1998-10-13
Filing date: 1999-10-12
Publication date: 2000-04-20
Also published as: WO2000021506A3; AU6074699A

Abstract

The present invention relates to combination of glutathione modulators with antimony and/or arsenic for cancer therapy. More particularly, the present invention relates to a pharmaceutical composition for the treatment of cancer, which comprises a therapeutical amount of a glutathione (GSH) modulator with antimony and/or arsenic in association with a pharmaceutically acceptable carrier.

Description

PHARMACEUTICAL COMPOSITION OF GLUTATHIONE MODULATORS WITH ANTIMONY AND/OR ARSENIC FOR CANCER THERAPY

BACKGROUND OF THE INVENTION (a) Field of the Invention

The invention relates to a pharmaceutical composition for cancer therapy based on the combination of glutathione modulators with antimony and/or arsenic, method of treatment and use thereof. (b) Description of Prior Art

The effective treatment of human cancer currently presents the single greatest obstacle in Western medicine. As such, the search for novel therapies must remain a priority as our population continues to expand and age. A form of arsenic (As), arsenic trioxide (As₂0₃) has recently been proven to provide an effective treatment for the rare myeloid leukemia, acute promyelocytic leukemia (APL) .

Acute promyelocytic leukemia (APL) represents about 10% of acute myeloid leukemia (AML) cases. APL is characterized by a specific differentiation block of the myeloid progenitor cells at the promyelocytic stage. At the molecular level, m the vast majority of cases (>95%), APL blasts harbor the balanced t(15;17) chromosomal translocation which fuses the PML gene located on chromosome 15 to the retinoic acid receptor α (RARα) gene on chromosome 17 (Kakizuka, A et al.

(1991) Cell , 66:663-674). The resulting PML-RARα fusion protein retains most of the functional domains of the parental PML and RARα proteins. The key role of the chimera in the differentiation block has initially been demonstrated in m vi tro (Kakizuka, A et al. (1991) Cell , 66:663-674) models and its leukemogenic potential has more recently been confirmed m transgenic mice. A unique feature of APL blasts is their ability to undergo terminal differentiation after retmoic acid (RA) treatment m vi tro as well as m vivo . Indeed, oral administration of all -transRA, the natural ligand for RARα, induces complete remission n t(15;17) APL patients, making APL the first example of differentiation therapy in the treatment of advanced cancer (Warrell, R.P. Jr et al. (1991) N. Engl . J. Med . , 324:1385). Several clinical studies have conclusively shown that the combination of RA with chemotherapy has improved the survival of patients with APL. Despite the broad success of RA therapy m APL, a significant percentage of patients relapse after initial remission and subsequently develop resistance to RA treatment. The clinical outcome of these patients is poor, as an effective substitute for RA is not yet available.

Arsenic and antimony are metals belonging to group Va of the periodic table of elements. Considered non-essential trace elements, these metals share several chemical and toxicological properties and occur naturally n both trivalent and pentavalent states. The trivalent state of both arsenic and antimony generally demonstrate more toxic effects than their pentavalent counterparts (Stemmer, K.L. ( 1976 j Pharmaco . Ther . A . , 1:157-160). Despite the cellular and organic damage associated with chronic exposure, however, both arsenic and antimony have found medical applications (Stemmer, K.L. (1976) Pharma co . Ther . A . , 1:157-160) . Antimony has, since the turn of the century, been used as a parasiticide and is still used for the treatment of schistosomiasis and leishmaniasis . More recently, arsenic has been shown to induce clinical remission, with limited side effects, both m patients having retmoid-sensitive and -resistant acute promyelocytic leukemia (APL) . One study reported a 90% success rate in the induction of complete remission m relapsed patients treated with arsenic alone for a period of 28-41 days (Shen, Z.X. et al . (1997) Blood, 89:3354-3360). In vi tro studies in our laboratory have demonstrated that AS₂O₃ induces apoptosis in both RA- resistant and sensitive APL cells m culture (Shao, W. et al. (1997) J. Na t . Cancer Ins . , 90:124-133). Arsenic has since been shown to induce apoptosis in other cancer cell types as well, although to a lesser degree than the more sensitive APL cells. It would be highly desirable to be provided with a novel approach for the treatment of cancer, including acute myeloid leukemias (AML) , such as acute promyelocytic leukemia (APL) .

SUMMARY OF THE INVENTION

One aim of the present invention is to provide a novel approach for the treatment of cancer, including acute myeloid leukemia (AML) and acute promyelocytic leukemia (APL) . One aim of the present invention is to provide a combination of glutathione modulators with antimony and/or arsenic for cancer therapy.

Our research into the mechanism of action of these compounds has yielded a novel approach that may allow antimony (Sb) and arsenic (As) to be employed against other acute myeloid leukemias and potentially against cancers of other sorts.

In accordance with the present invention, there is provided a pharmaceutical composition for the treatment of cancer, which comprises a therapeutical amount of a glutathione (GSH) modulator with antimony and/or arsenic in association with a pharmaceutically acceptable carrier.

In accordance with the present invention, the glutathione (GSH) modulator may be a GSH-depletmg agent, including, without limitation, buthionme sulfoximme (BSO) .

In accordance with a preferred embodiment of the pharmaceutical composition of the present invention, BSO concentration is of about 4 to about 10 mM/kg, antimony concentration is of about 0 to about 20 mg/kg and arsenic concentration is about 0 to about 20 mg/kg, with the proviso that both antimony and arsenic concentrations are not together 0. In accordance with the present invention, there is provided a method for the treatment of cancer, which comprises administering to a patient in need thereof a therapeutical amount of a glutathione (GSH) modulator with antimony and/or arsenic. In accordance with the present invention, there is provided the use of a therapeutical amount of a glutathione (GSH) modulator with antimony and/or arsenic for the treatment of cancer.

For the purpose of the present invention the following terms are defined below.

The term "cancer" is intended to mean cancers of any sorts, including, without limitation, acute myeloid leukemia (AML) , acute promyelocytic leukemia (APL) , breast cancer, prostate cancer, lung cancer, colorectal cancer, pancreatic cancer, ovarian cancer, cervical cancer, endometπal cancer, lymphoma, other adenocarcmomas, squamous cell carcinomas, renal cell carcinoma and melanomas.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 illustrates the synergistic inhibition of myeloid leukemic cell growth by arsenic and BSO;

Fig. 2 illustrates the synergistic inhibition of myeloid leukemic cells by antimony and BSO cotreatement ; Fig. 3 illustrates the synergistic inhibition of breast cancer cell growth by arsenic and BSO;

Fig. 4 illustrates growth response of APL and other malignant cell lines to arsenic and antimony; Fig. 5 illustrates induced expression of PML-

RARα m the AML (non-APL) cell line, U937, did not sensitize the cells to As₂0₃;

Fig. 6 illustrates GSH depletion sensitizes APL and other malignant cell types to arsenic and antimony through the synergistic induction of apoptosis;

Fig. 7 illustrates an arsenic-resistant NB4 subclone is cross resistant to antimony, yet can be sensitized to arsenic upon glutathione depletion;

Fig. 8 illustrates baseline glutathione levels cannot predict tolerance to arsenic or antimony;

Fig. 9 illustrates dose-dependent CSH depletion by BSO directly correlates with cell death m response to arsenic;

Fig. 10 illustrates heavy metals activate AP-1, but glutathione depletion represses this metal-induced activation.

DETAILED DESCRIPTION OF THE INVENTION

Acute promyelocytic leukemia (APL) is characterized by a specific t(15;17) chromosomal translocation that fuses the genes encoding PML and the retmoic acid receptor α (RARα) . The PML-RARα protein induces a block m the differentiation of the myeloid progenitor cells, which can be released by retmoic acid (RA) m vi tro and in vivo . The RA-mduced differentiation of APL blasts is paralleled by the degradation of the fusion protein and the relocation of wild-type PML from aberrant nuclear structures to its normal localization m nuclear bodies. Recently, arsenic trioxide (As₂0₃) treatment was proposed as an alternative therapy m APL, as it can induce complete remission in both RA-sensitive and RA-resistant APL patients. Intrigumgly, As₂0₃ was also shown to induce degradation of the PML-RARα chimera and to reorganize PML nuclear bodies.

We have new evidence that a similar compound of the related element, antimony (Sb), may also be effective against APL. We have recently discovered that Sb also induces apoptosis in cultured APL, AML, and cervical carcinoma cells (HeLa) , and have considerable evidence suggesting that its mechanism of action is similar to that of arsenic.

Based on prior preliminary results, we examined whether biochemical modulation might potentiate the action of arsenic and antimony m other malignant cell lines. A recent abstract suggests that the particular sensitivity of NB4 cells to arsenic may be a result of the low level of reduced glutathione (GSH) present in these cells. Glutathione is a ubiquitously expressed tripeptide that serves as the largest source of non- protem thiol m mammalian cells. In keeping with its role as the cell's most important antioxidant, GSH is a substrate of glutathione peroxidases, which destroy organic and hydrogen peroxides. It also maintains the cell's reducing milieu, and is implicated in a variety of thiol-disulfide interconversions . A further major role for GSH is in the detoxification of endogenous and exogenous toxic compounds, including drugs and many environmental mutagens and carcinogens. In fact, cells resistant to a variety of chemotherapeutic drugs, particularly alkylating agents, often have an increased GSH content or disregulated GSH utilization pathways. There is also some evidence that GSH can bind free arsenic, forming a transient As(GS)₃ complex and thereby preventing arsenic from attacking its mtracellular target. Antimony, on the other hand, is known to be conjugated to glutathione prior to its excretion from cells.

Therefore, we sought to determine whether biochemical modulation of mtracellular GSH levels could potentiate the effects of arsenic and antimony cancer cells intrinsically more resistant to these compounds than APL cells. We discovered that the relatively resistant, non-APL myeloid leukemic cell line PLB-985 could be rendered sensitive to arsenite when co-treated with the GSH-depletmg agent buthionme sulfoximme (BSO) , a selective inhibitor of α- glutamylcyste e synthetase, the rate-limiting enzyme m the synthesis of glutathione. MATERIAL AND METHODS

Cell culture and treatment of cells

NB4 cells, NB4R4 and PLB-985 cells were cultured m RPMI medium (Gibco, BRL) supplemented with antibiotics, glutamme and 10% fetal calf serum. MCF-7 cells and S30 cells were grown m minimal essential medium alpha supplemented with antibodies and 5% fetal calf serum. The HeLa cell-line stably overexpressmg PML(F) were as previously described. The cells were grown at 37°C in 5% C0₂ m Dulbecco' s modified minimal essential medium (Gibco, BRL) , supplemented with antibiotics, glutamme, 10% fetal calf serum and with G418 (Geneticm, Gibco, BRL) (750μg/ml) . As₂0₃ (Sigma) as a ImM stock solution in PBS, PAS (Aldrich) and meglumme antimonate Glucantime, Rhδne-Poulenc) as a lOμM stock solutions in H₂0. Sb₂0₃ (Fluka) was dissolved a minimal volume of HC1 and a lOOμM stock solution prepared m lOOμM Hepes, pH 7.5. BSO (Sigma) was dissolved in H₂0 at a concentration of 100 mM. Cells excluding Tryptan blue were counted at the indicated time following treatment initiation with hemocytometer. Growth Assays

NB4 and PLB-985 cells were seeded at 2x105 cells/ml in 6-well plates and HeLa cells at 1x104 cells/well in 24-well plates. Cells were treated with various concentrations of As203 (Sigma, Oakville, Ontario, Canada) or Sb203 (Fluka, Ronkonkoma, NY, USA) for seven days. Viable cells were counted by trypan blue (Gibco) exclusion with a hemacytometer . NB4 and PLB-985 cells were maintained at a density lower than 1x106 cells/ml through dilution as required, and HeLa cell media +/- treatment was replaced on every third day. TUNEL assay

Genomic DNA strand breaks characteristic of apoptosis were labeled m si tu by terminal deoxynucleotidyl transferase (TdT) using an m si tu cell death detection kit (Boehrmger Mannheim, Laval, Quebec, Canada) according to the manufacturer's instructions. Briefly, cytospms containing 100,000 cells were fixed 4% paraformaldehyde and permeabilized n 0.1% Triton X-100 and 0.1% sodium citrate. The cells were then exposed to the TUNEL reaction mixture containing TdT enzyme and fluorescem- labeled nucleotides, washed, and photographed under a fluorescence microscope.

Induction and detection of PML-RARα expression in U937- PR9 cells

U937 cells stably transfected with a zinc- mducible PML-RARα expression vector were seeded at lxlO³ cells/well m 96-well plates. Cells were treated with 100 μM ZnS04 (Sigma) for 24 hours prior to the addition of As203 to induce expression of the fusion protein. Nuclear extracts were prepared by first washing lxlO⁷ cells twice with PBS and resuspendmg in

2 ml PTG buffer (5 mM NaH₂P0₄, ImM DTT, 10% glycerol, ImM PMSF, 10 μg/ml each aprotmm and leupeptm, pH 7.4) at 4°C. Cells were then homogenized using a Dounce and pestle (pestle B) and the extracts transferred to centrifuge tubes. Following centrifugation for 30 minutes at 800g at 4°C, the nuclear pellet was washed with 1 ml PTG buffer and recentπfuged as above. Nuclear pellets were suspended in ice cold TTGK buffer (10 mM Tπs-Cl, pH 8.5, 1 mM EDTA, ImM DTT, 10% glycerol, 0.8 M KCl, ImM PMSF, 10 μg/ml each aprotmm and leupeptm) at 100 μL/lxlO⁷ cells. The suspension was incubated at 4°C for 1 hour, and was resuspended at 15 minute intervals. Extracts were then centrifuged at 14,000 rpm m a microfuge at 4°C, and supernatants transferred to fresh tubes. To detect PML-RARD, 10 μL of nuclear extract (approximately 1x106 cells) was added to 10 μL sample buffer and run on a 10% SDS-polyacrylamide gel. Proteins were transferred to nitrocellulose membranes (Bio-Rad, Mississauga, Ontario, Canada) and blocked with 25% fetal bovine serum m PBS containing 0.5% Triton X-100 for 1 hour at room temperature. The membrane was then hybridized overnight at 4°C with an antibody directed against the RARα F domain (provided by Dr. P. Chambon) . Following 3xl5-mmute washes with PBS containing 0.5% Triton X-100, PML-RARα was visualized by hybridization with 1251-labelled protein A and subsequent autoradiography . Toxicity Assays

Cells were seeded at lxlO³ cells/well/0.1 ml m 96-well plates. 24 hours after seeding, buthionme sulfoximme (Sigma) was added at a final concentration of either 25 μM or 50 μM to half of the wells from each plate, and media was added to all other wells. Cells were cultured overnight prior to the addition of heavy metal to allow for mtracellular GSH depletion. As₂0₃ or Sb₂0₃ was then added at concentrations ranging from 5 nM to 100 μM. Following a 96-hour treatment, cells were pelleted if necessary, fixed with 50% TCA, washed, and stained with sulforhodamme B (Sigma) . Protein stammg was then measured spectrophotometrically at 630 nm. GSH assay

Intracellular reduced glutathione (GSH) levels were assessed enzymatically with glutathione reductase. 10 Million cells were harvested, washed and lysed in 0.9 ml of 100 mM Tπs-Cl by freeze-thaw cycling. Cell lysates were centrifuged and the supernatants transferred to eppendorf tubes containing 0.1 ml of 30% S-sulfosalicyclic acid. Following 15 minutes on ice, cell extracts were centrifuged for 2 minutes at 12,000 rpm and the supernatants transferred once again to new tubes. GSH levels per 100 μl sample were then measured spectrophotometrically by conversation of 5,5'-dιthιo- bis (2-nιtrobenzoιc acid) to its colored product upon reduction by GSH-dependent glutathione reductase. AP-1 Transactivation Assay

2 x 10⁶ HeLa cells stably expressing an AP-1 reporter gene construct, TRE6-TATA-CAT (Komg A et al., Blood 1997; 90 (2) :562-70) , were seeded in 10 cm plates and allowed to attach overnight. Cells were then serum starved for 24 hours, and then treated with As₂0₃ (Sigma) or Sb₂0₃ (Fluka) +/- BSO (Sigma) for 16 hours. Cell extracts were prepared by 3 freeze-thaw cycles, and CAT activity assessed as previously described (Rosenauer A et al . , Blood 1996; 88 (7 ): 2671-82) . Briefly, cell extracts standardized for protein content (Bio-Rad) were incubated with cold chloramphenicol (ICN, Costa Mesa, CA, USA) and tritiated acetyl-CoA (Mandel, Montreal, Quebec, Canada) at 37°C for 2 hours. The generation of acetylated chloramphenicol was then measured by scintillation counting following its extraction from free acetly-CoA by the organic diffusion method (Rosenauer A et al., Blood 1996; 88 (7) :2671-82) .

RESULTS

Following a 24 hour treatment, PLB-985 cells treated with both 1 μM arsenic trioxide and 100 μM BSO had a viability of only 22.9%, while that of cells treated with As₂0₃ or BSO alone was 96.9% and 97.1%, respectively (Fig. 1). After 48 hours, the viable cell count in the co-treatment group was only 2.22% that of the untreated control. Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) analysis confirmed that this loss of cell viability was due to the induction of apoptosis. Furthermore, a GSH assay revealed that 100 μM BSO alone or in combination with arsenic was sufficient to reduce glutathione levels from 18.6 nmol/mg protein in control cell extracts to undetectable levels after a 24 hour treatment (Table 1) .

Table 1 Effect of GSH depletion upon arsenic-induced apoptosis

Effect of arsenic treatment upon mtracellular GSH levels

A similar result was obtained with antimony and BSO co-treatment, but with a slightly delayed response (Fig. 2) . Here, a 72-hour treatment yielded 8% of cells viable in the co-treatment group, while antimony- or BSO-treated cells had 86.9% and 80.8% viability, respectively (Fig. 2). Recently, this result has also been extended to the more common deadly malignancy, breast cancer (Fig. 3) . Here we discovered that both estrogen receptor-positive and negative breast cancer cell lines, some of which display considerable resistance to arsenic, undergo apoptosis when depleted of mtracellular GSH at the time of 1 μM arsenic trioxide treatment (Fig. 3). For example, while MCF-7 cells cultured the presence of arsenic were 78.1% viable after 48 hours, relative to an untreated control group, co-treated cells had a viability of 5.74% (Fig. 3) . In vivo studies aimed at confirming the results of these cell culture experiments in tumor- bearing animals are currently m progress.

Clearly, modulation of glutathione levels provides a powerful means by which to sensitize tumor cells to the action of heavy metals. Because treatment with effective doses of arsenic alone may prove to be too toxic for the majority of malignancies outside of APL, this sensitization may allow for an increase m the anti-cancer activity of this heavy metal without an associated increase in toxicity. This is particularly plausible when one considers that many tumors have a seemingly increased need for glutathione, and thus when depleted of it may be placed at a disadvantage over normal cells. Furthermore, the novel combination of antimony and BSO may prove to be an equally effective treatment against APL and other malignancies, with potentially less toxicity than arsenic.

APL cells are uniquely sensitive to As₂0₃ and Sb₂0₃- induced apoptosis Our lab and others have previously documented the induction of apoptosis in the APL cell line, NB4, upon treatment with arsenic. In this report, we investigated the induction of apoptosis by As₂0₃ and Sb₂0₃ in several other hematologic and solid tumor cell lines. As Fig. 4 illustrates, NB4 cells were more sensitive to both arsenic and antimony than any other cell line examined, including the non-APL myeloid leukemic PLB-985 cell line, and solid tumor-derived HeLa cell line. Cells were treated with the indicated concentration of arsenic trioxide (As₂0₃) (Fig. 4A) or antimony trioxide (Sb₂0₃) (Fig. 4B) for one week. Viable cells were assessed by counting trypan blue- excluding cells. Each data point represents the average of triplicate or quadruplicate samples, and standard error bars are shown. As shown, 2 μM AS₂O₃ and Sb₂0₃ led to 100% cell death m NB4 cells, while both PLB-985 and HeLa cell numbers continued to increase after 7 days of treatment with 2-3 μM As₂0₃ or 5μM Sb₂0₃.

Induced expression of PML-RARα does not sensitize non- APL cells to heavy metals

In spite of the increased sensitivity of APL cells to treatment with arsenic and antimony, we hypothesized that PML-RARα expression is not a major determinant of cellular sensitivity to these agents. To test this hypothesis, we determined the sensitivity of the non- APL, myeloid leukemic U937 cell line, to As₂0₃ m the presence and absence of exogenously expressed PML-RARα. U937-PR9 cells stably expressing the fusion orotem under the control of a zmc-mducible promoter and mock-transfected U937-SN4 cells were seeded at lxl0^J cells/well in 96-well plates, and PML-RARα expression was induced by treatment with 100 μM ZnS0₄ for 24 hours. Treatment with ZnS0₄ did not cause additional toxicity itself as the IC₅₀ value for arsenic in control, mock-transfected SN4 cells was not affected by its presence. Induction of PML-RARα expression in PR9 cells was verified through western blotting, and the level of expression of this protein was found to be comparable to that NB4 cells (Fig. 5A) . 7-PR9 and -SN4 cells were treated for 24 hours with 100 μM ZnS0₄ - -15 _c -

to induce the expression of PML-RARα. Western blotting with an antibody directed against RARα was performed to detect PML-RARα induction and to compare its level of expression relative to that of NB4 cells (A) . Cells were then treated with As₂0₃, at concentrations ranging from 5 nM to 100 μM, for four days. The absence of a leftward shift in the toxicity curve for zinc-treated

PR9 cells (Fig. 5B) demonstrates that expression of the fusion protein does not change the sensitivity of this non-APL cell line to arsenic. U937-PR9 cells were then treated for four days with As₂0₃, and survival was assessed in cells expressing PML-RARα or not ( ■ and ♦, respectively) as a measure of protein concentration per well with the sulforhodamme B assay (B) .

Glutathione depletion sensitizes tolerant and sensitive cells to arsenic- and antimony-induced apoptosis

We therefore investigated an alternative hypothesis suggested by reports that a lower GSH content APL and other cells might mediate their sensitivity to arsenic and antimony. We tested the response of a variety of cancer cell types to As₂0₃ and Sb₂0₃ m combination with BSO, an agent that depletes mtracellular GSH. We found that all cell lines examined could be sensitized to AS₂O₃ and Sb₂θ₃ by glutathione depletion, regardless of their inherent tolerance or sensitivity to metal alone. As Fig. 6A illustrates, pretreatment with 25 or 50 μM BSO sensitized cells to a four-day treatment with arsenic or antimony. (A) A variety of malignant cell types were treated overnight m the presence or absence of 25 or 50 DM BSO followed by a four day treatment with a range of concentrations of As₂0 (A, left column) or Sb₂0₃ (B, right column) . Cell survival was assessed by the SRB assay and expressed as a fraction of untreated controls. The sensitization achieved upon GSH depletion with BSO ranged from 4.3 to 75-fold for As₂0₃ and 20 to 87.5-fold for Sb₂θ₃ in the four cell lines examined. These included, in addition to NB4, an ER+ breast cancer cell line, MCF-7, an MCF-7-derived adriamycin-resistant line, NIH-ADR, and the cervical carcinoma cell line HeLa.

Although we have previously reported the induction of apoptosis by As₂0₃ in NB4 cells, we were interested in determining whether the synergism between BSO and arsenic or antimony was mediated by a synergistic induction of programmed cell death. Evidence for the induction of apoptosis was obtained through Annexin V staining of arsenic treated NB4 cells, +/- BSO. Here we found that a 24-hour treament with 1 μM As₂0₃ and 100 μM BSO significantly increased the number of apoptotic Annexin V-staining cells, as compared to either the arsenic or BSO-treated groups alone (Fig. 6B) . (B) NB4 cells were pre-treated for 24 hours with 100 μM BSO, followed by a 48 hour treatment with the indicated concentration of arsenic. Apoptosis was assessed by Annexin V-FITC staining, and apoptotic cells quantitated through flow cytometry. We also investigated the induction of apoptosis in the relatively arsenic-resistant AML cell line, PLB-985. Fig. 6C shows a clear,^' time-dependent sensitization to 0.3 or 0.5 μM As₂0₃ upon GSH depletion. (C) PLB-985 cells were co-treated with 25 μM BSO (•) and 0.3 ( -*■ ) or 0.5 μM As₂03 (♦) . Viable cells were counted by trypan blue exclusion with a hemacytometer over a period of five days. TUNEL analysis, shown in Fig. 6D, was performed to confirm that arsenic and BSO synergize in the induction of apoptosis in these cells, as shown for the more sensitive NB4 cell line. (D) Cytospins of 2 x 10⁵ PLB-985 cells were prepared from cells treated for 48 hours with either 25 μM BSO, 1 μM As₂0₃, or both. Apoptosis was assessed by terminal nucleotidyl transferase-mediated dUTP nick end labelling (TUNEL) and visualized by fluorescence microscopy. Here a 48- hour treatment with 1 μM AS₂O₃ caused only a slight increase in the number of apoptotic cells over the untreated control. Co-treatment with 25 μM BSO, however, sensitized the cells such that a 1 μM treatment induced a response similar to that induced by 10 μM AS₂O₃ alone, corresponding to a roughly 10-fold sensitization by BSO.

Arsenic-resistant NB4 cells can be re-sensitized to heavy metals through GSH depletion

Just as we have previously used retmoid resistant subclones of the APL cell line, NB4, to investigate mechanisms of response to RA, we developed arsenic-resistant subclones to address the question of how heavy metals exert their effects. These cells were generated by constant culture in the presence of As₂0₃ at concentrations that were slowly increased over time. Once a population of cells were obtained that could be sustained m 1 μM As₂0 , single clones were selected by plating methylcellulose . The two clones that were selected for further expansion are nov routinely cultured m the presence of 2 μM As₂0₃. As Fig. 7 illustrates, AsR3 cells are approximately 10-fold less sensitive to AS₂O₃, and they also show significant cross-resistance to Sb₂0₃. (Fig. 7A) Arsenic-resistant AsR3 and their sensitive parental NB4 cells were seeded at 1,000 cells/well 98-well plates (♦ and ■, respectively) and treated with a range of concentrations of arsenic (left panel) or antimony

(right panel) for a period of four days. Cell survival was assessed by measuring protein content through the SRB assay, and expressed as a fraction of control groups. (Fig. 7B) Sensitization by GSH depletion was demonstrated by plating cells as above, then treating for four days with arsenic following a 24 hour pre- treatment with 100 μM BSO. Despite this resistance, these cells could also be rendered sensitive to AS₂O₃ (Fig. 7) and Sb₂0₃ by mtracellular GSH depletion with a non-toxic dose of BSO.

Baseline glutathione levels alone cannot predict tolerance or sensitivity to metals, although upregulation of GSH may be a mechanism for resistance in As₂0₃-resιstant NB4 cells

To determine whether low GSH content could explain the unique sensitivity of APL cells to heavy metals, we assessed baseline GSH levels in several cell types and correlated these with their IC₅₀ values for a four-day treatment with As₂0₃ or Sb₂θ₃. While we found no correlation between baseline GSH and inherent sensitivity or tolerance to metals among cells of different types (Fig. 8A) , we did find that glutathione levels were elevated in the arsenic resistant cell lines, AsR2 and AsR3 (Fig. 8B) . (Fig. 8A) Intracellular reduced glutathione levels were assessed in a variety of cancer cell lines havmg different IC₅₀ values for As₂0j and Sb₂U₃. GSH levels were measured indirectly by reduction of by GSH-dependent glutathione reductase, and expressed as nmol/mg protein. (Fig. 8B) GSH levels were also assessed in parental, sensitive NB4 cells and compared to those in two subclones of arsenic-resistant cells. GSH values from a representative experiment are shown, and are the mean of triplicate measurements.

Dose-dependent GSH depletion by BSO correlates with response of NB4 cells to As₂0₃ and Sb₂0₃ In order to determine whether sensitization to heavy metals by BSO results from GSH depletion, we correlated the dose-dependent suppression of GSH levels in NB4 cells with the growth inhibitory effect by BSO plus As₂0₃. Fig. 9A shows a linear relationship between BSO concentration and depletion of mtracellular GSH following a three day treatment with doses ranging from 0.1 to 10 μM. (Fig. 9A) NB4 cells were treated for three days with sublethal doses of BSO ranging from 0.1 to 10 μM, after which glutathione content was assessed. NB4 cells were also co-treated with BSO doses m this range and 1 μM As₂0₃. This dose-dependent depletion of GSH directly correlated with the degree of sensitization conferred by BSO, as shown in Fig. 9B, strongly suggesting that glutathione depletion sensitizes NB4 cells to arsenic and antimony. Cell survival was measured by counting trypan-blue excluding cells, and the values shown represent triplicate measurements (Fig. 9B) .

As₂0₃ and Sb₂0₃ cause a dose-dependent increase in AP-1 activity in HeLa cells, and GSH depletion represses this metal-induced activation

Based on previous reports that arsenic can induce phosphorylation of JNK, an activator of the AP-1 transcription factor, we investigated the ability of

As₂0₃ and Sb₂θ₃ to induce AP-1 activity. As Fig. 10 illustrates, 16 hour treatment with arsenic or antimony induced a strong dose-dependent activation of AP-1 activity, as measured by chloramphenicol acetyl transferase (CAT) activity in HeLa cells stably expressing a (TRE) ₆-TATA-CAT reporter gene construct. Treatment with 1, 5, or 15 μM As₂0 induced a 1.12-, 3.26-, and 22.6-fold induction, respectively. HeLa cells stably expressing the AP-1 reporter gene, TRE₆- TATA-CAT, were serum-starved overnight and treated for 16 hours with the indicated dose of arsenic or antimony m the presence or absence of 100 μM BSO. CAT activity was measured and normalizeα for protein content following cell lysis. Treatment with the same concentrations of Sb₂0₃ triggered 1.11-, 2.78-, and 37.7-fold inductions. Because GSH depletion has independently been shown to activate JNK and AP-1, we investigated the ability of coupled glutathione depletion and heavy metal treatment to synergistically induce AP-1 transactivation . Interestingly, we found that pretreatment with 100DM BSO repressed AP-1 activation by both arsenic and antimony. This repression was complete in the case of antimony, while GSH depletion repressed arsenic-mduced AP-1 activity to an intermediate level . Discussion

AS₂O₃ has been reported to be an effective therapy against a specific hematologic malignancy, acute promyelocytic leukemia (APL) , which is known to be uniquely responsive m vi tro and m vivo to RA. Recent m vi tro results suggest that Sb₂θ₃ may also have activity against APL cells. To date, however, arsenic and antimony have been less effective m vi tro against other cancer cells. Because the PML-RARα fusion protein is the characteristic feature of APL, and since both RA and these metals induce its degradation, we examined whether the exogenous expression of this protein could modulate sensitivity to arsenic or antimony. By inducing the expression of PML-RARα from a zmc-mducible promoter in stably transfected U937 cells, we found that the expression of the fusion protein alone did not measurably increase sensitivity to arsenic. This is consistent with previous reports by our lab and others that a RA-resistant NB4 subclone, NB4.306, which lacks detectable PML-RARlJ expression, is equally sensitive to arsenic as its parental cell line. These data suggest that the unique sensitivity of APL cells may result from their biochemical or cellular background, not merely from the presence of the fusion protein. This leads to our hypothesis that modulation of the biochemical environment of the cell might increase the efficacy of arsenic and antimony against other, non-APL malignancies.

An alternative explanation for the unique sensitivity of APL cells is their reportedly low GSH content, as compared to that of cell lines of other hematological malignancies. GSH confers protection against xenobiotics and several chemotherapeutic agents, including alkylatmg agents and cisplatin, and plays a major role m maintaining mtracellular redox potential. Also, all cancer cells tested, regardless of GSH content or their initial response to AS₂O₃ or Sb₂0₃, could be sensitized by mtracellular GSH depletion.

Thus, steady state GSH content appears to be only one factor dictating cellular resistance or sensitivity to heavy metals. The ability to synthesize GSH may be an additional factor in determining cellular tolerance to heavy metals, suggesting an important role for the enzymes m the biosynthetic pathway of GSH. Also, the expression level of other proteins, including GSH utilizing enzymes (including glutathione S- transferases, GSTs) and those involved m metal ion chelating (metallothione ) , free radical scavenging, and peroxide metabolism (GSH peroxidase, catalase) may play important roles. Because arsenic is exported from the cell in an ATP-dependent manner by the multidrug- resistance associated protein (MRP) , high expression of this membrane pump could also cause metal resistance. Each of these proteins has been shown to be mducible by arsenic, or overexpressed some arsenic-resistant cell systems, and their relative abundance may exert a combined effect on overall As₂Oj or Sb0₃ tolerance. Expression of pGp, on the other hand, may not play a role in arsenic or antimony resistance, as we find that an adriamycin-resistant cell line known to overexpress this membrane pump, NIH ADR, is not cross-resistant to these metals (Fig. 6A) . Similarly, a cisplatm- resistant squamous cell carcinoma cell line displayed no increased resistance to arsenic or antimony, and was sensitized by BSO to the same degree as its cisplatin- sensitive parental cell line. These results do suggest that this combined therapy may be effective against tumors resistant to other chemotherapeutic agents.

To analyze the mechanism by which arsenic/antimony and BSO synergize in the induction of apoptosis, we first examined the ability of these drugs to modulate the transcriptional activity of the AP-1 transcription factor. AP-1 activity is induced both by mitogens and cellular stresses including oxidative stress and arsenite. As expected, we observed that both metals induced a dose-dependent activation of AP-1 in HeLa cells. On the other hand, although GSH depletion promotes an oxidative environment and BSO treatment has been shown to activate the SAP kinase JNK, BSO treatment did not increase AP-1 activity HeLa cells. This latter observation contrasts with the report that, under conditions of GSH depletion, baseline and tert- butylhydroquinone-induced AP-1 binding activity increased in HepG2 cells. AP-1 activity can be induced by three different MAP/SAP kinase cascades, which can have opposite effects on cell proliferation and apoptosis. It is possible that AP-1 activation in response to heavy metals may reflect a protective response which, when inhibited by glutathione depletion, may leave the cell more vulnerable to their toxic effects. This argument is supported by the finding that GSH depletion repressed the induction of AP-1 activity by antimony to a greater extent than that by arsenic (Fig. 10), while consistently leading to a greater sensitization to antimony than that by arsenic (Fig. 6A) .

Induction of apoptosis of mouse epidermal JB6 cells by high doses of arsenic has recently been shown to proceed through induction of the JNK SAP kinase. On the other hand, the extra cellular signal-regulated kmases (ERKs) are also induced in a delayed fashion, and activation of ERKs was found to mediate arsenic- induced transformation of JB6 cells. It will be of interest to assess which signalling pathways are modulated by BSO treatment. Indeed, BSO may synergize with arsenic or antimony by inhibiting the anti- apoptotic ERK pathway, or by activating pro-apoptotic pathways that do not induce AP-1.

In order for a treatments therapeutic index to be improved, its efficacy must be increased more than its associated toxicity. Although several studies have shown that malignant cells often have significantly elevated glutathione levels, GSH is also an important component of normal cell function. If GSH is depleted and repleted from various normal and tumor tissues with different kinetics, however, it may be possible to establish a protocol that optimizes antitumor efficacy and therapeutic gam. In fact, m KHT sarcoma-oearmg mice, Siemann et al found that GSH levels m tumor cells were depleted more slowly than those m bone marrow cells, but also recovered more slowly than the normal marrow. This provided a window of time between the recovery of bone marrow cells and reestablishment of GSH levels in the tumor during which co-treatment with BSO and the alkylatmg agent, melphalan, enhanced the tumor cell death.

While the invention has been described in con- nection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations, uses, or adaptations of the invention following, m general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features hereinbefore set forth, and as follows in the scope of the appended claims.

Claims

WHAT IS CLAIMED IS:

1. A pharmaceutical composition for the treatment of cancer, which comprises a therapeutical amount of a glutathione (GSH) modulator with antimony and/or arsenic in association with a pharmaceutically acceptable carrier.

2. The pharmaceutical composition of claim 1, wherein said glutathione (GSH) modulator is GSH- depletmg agent.

3. The pharmaceutical composition of claim 2, wherein said GSH-depletmg agent is buthionme sulfoximine (BSO) .

4. The pharmaceutical composition of claim 3, wherein BSO concentration is of about 4 to about 10 mM/kg, antimony concentration is of about 0 to about 20 mg/kg and arsenic concentration is about 0 to about 20 mg/kg, with the proviso that both antimony and arsenic concentrations are not together 0.

5. The pharmaceutical composition of claim 1, wherein said cancer is selected from the group consisting of acute myeloid leukemia (AML) , acute promyelocytic leukemia (APL) , breast cancer, prostate cancer, lung cancer, colorectal cancer, pancreatic cancer, ovarian cancer, cervical cancer, endometπal cancer, lymphoma, other adenocarcmomas , squamous cell carcinomas, renal cell carcinoma and melanomas.

6. A method for the treatment of cancer, which comprises administering to a patient in need thereof a therapeutical amount of a glutathione (GSH) modulator with antimony and/or arsenic.

7. The use of a therapeutical amount of a glutathione (GSH) modulator with antimony and/or arsenic for the treatment of cancer.