WO2000061183A2

WO2000061183A2 - Verotoxin treatment of lymphomas

Info

Publication number: WO2000061183A2
Application number: PCT/CA2000/000371
Authority: WO
Inventors: Clifford A. Lingwood; Gerald Arbus
Original assignee: HSC Research and Development LP
Current assignee: HSC Research and Development LP
Priority date: 1999-04-09
Filing date: 2000-04-07
Publication date: 2000-10-19
Anticipated expiration: 2001-10-09
Also published as: WO2000061183A3; AU3798500A

Abstract

Methods for treating lymphoma by the administration of an agent that binds Gb3 are disclosed. In one embodiment, the lymphoma may be, for example, PTLD.

Description

VEROTOXIN TREATMENT OF LYMPHOMAS

Related Application

This application is related to U.S. Patent Application Serial No. 08/563,394 filed on November 28, 1995, which is a continuation-in-part of U.S. Patent Application Serial No. 08/386,957, filed February 10, 1995, now abandoned, both of which are hereby incorporated by reference.

Field of the Invention

This invention relates to verotoxin pharmaceutical compositions and to methods of treating mammalian neoplasia, particularly, lymphomas, brain, ovarian and skin cancers, therewith.

Background to the Invention

Bacteriocins are bacterial proteins produced to prevent the growth of competing microorganisms in a particular biological niche. A preparation of bacteriocin from a particular strain of E. coli (HSCi Q) has long been shown to have anti-neoplastic activity against a variety of human tumour cell lines in vitro (1,2). This preparation, previously referred to as PPB (partially purified bacteriocin (2)) or ACP (anti-cancer proteins (2)) was also effective in a murine tumour model of preventing metastases to the lung (2). Verotoxins, also known as SHIGA-like toxins, comprise a family known as Verotoxin 1 , Verotoxin 2, Verotoxin 2c and Verotoxin 2e of subunit toxins elaborated by some strains of E. coli (3). These toxins are involved in the etiology of the hemolytic uremic syndrome (3,4) and haemorrhagic colitis (5). Cell cytotoxicity is mediated via the binding of the B subunit of the holotoxin to the receptor glycolipid, globotriaosylceramide, in sensitive cells (6). The verotoxin family of E coli elaborated toxins bind to the globo series glycolipid globotriaosylceramide and require terminal gal α-1-4 gal residue for binding. In addition, VT2e, the pig edema disease toxin, recognizes globotetraosylceramide (Gb_ι) containing an additional β 1-3 linked galNac residue. These glycolipids are the functional receptors for these toxins since incorporation of the glycolipid into receptor negative cells renders the recipient cells sensitive to cytotoxicity. The toxins inhibit protein synthesis via the A subunit. The A subunit is an N-glycanase which removes a specific adenine base in the 28S RNA of the 60S RNA ribosomal subunit. However, the specific cytotoxicity and specific activity is a function of the B subunit. In an in vitro translation system, the verotoxin A subunit is the most potent inhibitor of protein synthesis yet described, being effective at a concentration of about 8 pM. In the rabbit model of verocytotoxemia, pathology and toxin targeting is restricted to tissues which contain the glycolipid receptor and these comprise endothelial cells of a subset of the blood vasculature. Verotoxins have been strongly implicated as the etiological agents for hemolytic uremic syndrome and haemorrhagic colitis, microangiopathies of the glomerular or gastrointestinal capillaries respectively. Human umbilical vein endothelial cells (HUVEC) are sensitive to verotoxin but this sensitivity is variable according to cell line. Human adult renal endothelial cells are exquisitely sensitive to verotoxin in vitro and express a correspondingly high level of Gb3- However, HUS is primarily a disease of children under three and the elderly, following gastrointestinal VTEC infection. It has been shown that receptors for verotoxin are present in the glomeruli of infants under this age but are not expressed in the glomeruli of normal adults. HUNEC can be sensitized to the effect of verotoxin by pretreatment by tumour necrosis factor which results in a specific elevation of Gb3 synthesis (7,8). Human renal endothelial cells on the other hand, although they express high levels of Gb3 in culture, cannot be stimulated to increase Gbβ synthesis (8). It has been suggested that the transition from renal tissue to primary endothelial cell culture in vitro results in the maximum stimulation of Gb3 synthesis from a zero background (9). We therefore suspect that HUS in the elderly is the result of verotoxemia and a concomitant stimulation of renal endothelial cell Gb3 synthesis by some other factor, eg. LPS stimulation of serum α TΝF. Thus under these conditions, the majority of individuals (excepting the very young) would not be liable to VT induced renal pathology following systemic verotoxemia. It has also been shown that the verotoxin targets a sub-population of human B cells in vitro (10). These Gb3 containing B cells are found within the germinal centers of lymph nodes (11). It has been proposed that Gb3 may be involved in a germinal center homing by CDI9 positive B cells (12) and that Gb3 may be involved in the mechanisms of antigen presentation (13). Elevated levels of Gb3 have been associated with several other human tumors

(14-16), but ovarian tumors have not been previously investigated. Gb3 is the pk blood group antigen (17). Tissue surveys using anti-pk antisera have shown that human ovaries do not express this glycolipid (18, 19). Sensitivity to VTI cytotoxicity in vitro has been shown to be a function of cell growth, the stationary phase cells being refractile to cytotoxicity (20). The sequence homology between the receptor binding B subunit and the human α2-interferon receptor and the B cell marker CD 19 suggests that expression of Gb3 is involved in the mechanism of α2-interferon and CD 19 signal transduction (12). On surface ligation, Gb3 has been shown to undergo a retrograde intracellular transport via the rough endoplasmic reticulum to the nuclear membrane (21).

The astrocytoma is the most common primary human brain tumour. The majority of astrocytomas are malignant neoplasms which infiltrate diffusely into regions of normal brain. Despite the advent of promising adjuvant therapies and drugs which have impacted positively on patient survival in other tumor types in recent times, no such promising therapy has yet been found for the patient with a malignant astrocytoma. The median survival for patients with glioblastoma multiforme, the most malignant form of astrocytoma, is approximately 12 months and accordingly, it is imperative that new therapeutic treatments for malignant astrocytomas be found.

VTs consist of a 30kDa enzymatic A subunit which is capable of inhibiting protein synthesis. The A subunit is noncovalently associated with a pentameric 7kDa B subunit array which binds to Gb3. In addition to the cytotoxic effects of VTs on a wide range of cells by the A subunit inhibition of protein synthesis, recent evidence suggests that VTI, and the receptor binding B subunit alone, also induce morphological changes and DNA fragmentation characteristic of apoptosis in Gb3-positive cells (22, 23).

Reference List

The present specification refers to the following publications, each of which is incorporated herein by reference:

1. Farkas-Himsley, H. and R. Cheung. Bacterial Proteinaceous Products (bacteriocins as cytotoxic agents ofneoplasia). Cancer Res. 36:3561-3567,

(1976).

2. Hill, R.P. and H. Farkas-Himsley. Further studies of the action of a partially purified bacteriocin against a murine fibrosarcoma. Cancer Res. 51:1359-1365 (1991). 3. Karmali, M.A. Infection by Verocytotoxin-producing Escherichia coli. Clin.

Microbiol. Rev. 2: 15-38 (1989). 4. Karmali, M.A., M. Petric, C. Lim, P.C. Fleming, G.S. Arbus and H. Lior, 1985.

The association between hemolytic uremic syndrome and infection by

Verotoxin-producing Escherichia coli, J. Infect. Dis. 151:775. 5. Riley, L.W., R.S. Remis, S.D. Helgerson, H.B. McGee, J.G. Wells, B.R. Davis,

R.J. Hebert, E.S. Olcott, L.M. Johnson, N.T. Hargrett, P.A. Blake and M.C. Cohen. Haemorrhagic colitis associated with a rare Escherichia cold serotype. N. Engl. J. Med. 308:681 (1983). 6. Lingwood, CA.,. R. Bell, Y.A. Hannun and A.M. Jr., Advances in Lipid Research Academic Press. 25:189-211 (1993). 7. van de Kar, Ν.C.A.J., L.A.H. Monnens, M. Karmali and V.W.M. van

Hinsbergh. Tumour necrosis factor and interleukin-1 induce expression of the verotoxin receptor globotriaosyl ceramide on human endothelial cells. Implications for the pathogenesis of the Hemolytic Uremic Syndrome. Blood. 80:2755, (1992). 8. Obrig T., C. Louise, C. Lingwood, B. Boyd, L. Barley-Maloney and T. Daniel.

Endothelial heterogeneity in Shiga toxin receptors and responses. J. Biol. Chem. 268:15484-15488 (1993). 9. Lingwood, CA. Verotoxin-binding in human renal sections. Nephron. 66:21-28

(1994). 10. Cohen, A., V. Madrid-Marina, Z. Estrov, M. Freedman, CA. Lingwood and

H.M. Dosch Expression of glycolipid receptors to Shiga-like toxin on human B lymphocytes: a mechanism for the failure of long-lived antibody response to dysenteric disease. Int. Immunol. 2:1-8 (1990).

11. Gregory, CD., T. Turz, C.F. Edwards, C Tetaud, M. Talbot, B. Caillou, A.B. Rickenson and M. Lipinski. 1987. Identification of a subset of normal B cells with a Burkitt's Lymphoma (BL)-like phenotype. J. Immunol. 139:313-318 (1987).

12. Maloney, M.D. and CA. Lingwood, CD 19 has a potential CD77 (globotriaosyl ceramide) binding site with sequence similarity to verotoxin B-subunits: Implications of molecular mimicry for B cell adhesion and enterohemorrhagic

E. cold pathogenesis. J. Exp. Med. 180: 191-201, (1994).

13. Maloney, M. and C. Lingwood. Interaction of verotoxins with glycosphingolipids. TIGG. 5:23-31 (1993).

14. Li, S.C, S.K. Kundu, R. Degasperi and Y.T. Li. Accumulation of globotriaosylceramide in a case of leiomyosarcoma. Biochem. J. 240:925-927

(1986).

15. Mannori G., O. Cecconi, G. Mugnai and S. Ruggieri. Role of glycolipids in the metastatic process: Characteristics neutral glycolipids in clones with different metastatic potentials isolated from a murine fibrosarcoma cell line. Int. J. Cancer. 45:984-988 (1990).

16. Ohyama, C, Y. Fukushi, M. Satoh, S. Saitoh, S. Orikasa, E. Nudelman, M. Straud and S.I. Hakomori. Changes in glycolipid expression in human testicular

-4- tumours. Int. J. Cancer_. 45:1040-1044, (1990). 17. Naiki, M. and D.M. Marcus. Human erythrocyte P and P^* blood group antigens: Identification as glycosphingolipids. Biochem. Biophys. Res. Comm. 60:1105-1111, (1974). 18. Pallesen, G. and J. Zeuthen. Distribution of the Burkitt's-lymphoma-associated antigen (BLA) in normal human tissue and malignant lymphoma as defined by immunohistological staining with monoclonal antibody 38:13. J. Cancer Res. Clin. Oncol 113 :78-86 (1987).

19. Kasai, K., J. Galton, P. Terasaki, A. Wakisaka, M. Kawahara, T. Root and S.I. Hakomori. Tissue distribution of the Pk antigen as determined by a monoclonal antibody. J. Immunogenet. 12:213 (1985).

20. Pudymaitis, A. and CA. Lingwood. Susceptibility to verotoxin as a function of the cell cycle. J. Cell Physiol. 150:632-639 (1992).

21. Sandvig, K., O. Garred, K. Prydz, J. Kozlov, S. Hansen and B. van Deurs. Retrograde transport of endocytosed Shiga toxin to the endoplasmic reticulum.

Nature. 358:510-512 (1992).

22. Mangeney, M., Lingwood, C.A., Caillou, B., Taga, S., Tursz, T. and Wiels, J. Apoptosis induced in Burkitt's lymphoma cells via Gb3/CD77, a glycolipid antigen. Cancer Res. 53: 5314-5319, 1993. 23. Sandvig, K. and van Deurs, B. Toxin-Induced Cell Lysis: Protection by 3- Methyladenine and Cycloheximide. Exp Cell Res. 200: 253-262, 1992. 24. Ramotar, K., Boyd, B., Tyrrell, G., Gariepy, I., Lingwood, CA. and Brunton, J. characterization of Shiga-like toxin 1 B subunit purified from overproducing clones of the SLT-1 B cistron. Biochem. J. 272: 805-811 . 1990. 25. Costello, R. and Delmaestro, R., Human cerebral endothelium; Isolation and characterization of cell derived from microvessels of non-neoplastic and malignant glial tissue. J. Neuro-oncol. 8:231-243, 1990.

26. Pintus, C, Ransom, J. and Evans, C. Endothelial cell growth supplement: a cell cloning factor that promotes the growth of monoclonal antibody producing hybridoma cells. J. Immunological Methods. 61: 195-200, 1983.

27. Rutka J.T., Kleppe-Hoifodt H., Emma D.A., Giblin J.R., Dougherty D.V., McCulloch J.R., DeArmond S.J. and Rosenblum M.L., Characterization of normal human brain cultures: Evidence for the outgrowth of leptomeningeal cells. Laboratory Investigation 55: 71-85, 1986.

Although anti-neoplastic effects of bacterial preparations have been known for over twenty years, the neoplastic effect of verotoxin per se has, to-date, remained unknown. As a result of extensive investigations, we have discovered that verotoxin, particularly Verotoxin 1, is an active component within the ACP and that purified Verotoxin 1 has potent anti-neoplasia effect in vitro and in vivo. Most surprisingly, we have found effective in vivo anti-cancer treatments of human beings commensurate with non-toxic administered dosages.

Summary of the Invention

It is an object of the present invention to provide a pharmaceutical composition for the treatment of mammalian neoplasia and, particularly, lymphomas, brain cancers, skin cancers and ovarian cancers.

It is a further object of the present invention to provide a method of treating mammalian neoplasia, particularly, lymphomas, skin cancers, brain cancers and ovarian cancers. Accordingly, in one aspect the invention provides a pharmaceutical composition for the treatment of mammalian neoplasia comprising a non-lethal anti-neoplasia effective amount of a verotoxin, preferably, verotoxin 1, or the pentameric B subunit of verotoxin and a suitable pharmaceutically acceptable diluent, adjuvant or carrier therefor. The invention preferably provides a pharmaceutical composition and method of treatment for mammalian lymphomas, skin cancers, brain cancers and ovarian cancers. In a further aspect, the invention provides a process for the manufacture of a pharmaceutical composition for the treatment of mammalian neoplasia, said process comprising admixing an agent, e.g., verotoxin, the pentameric B subunit of verotoxin 1, verotoxin 2, or verotoxin 2c, with a pharmaceutically acceptable carrier, adjuvant or diluent therefor.

The present invention provides selective, specific cancer treatments wherein the agent, e.g., verotoxin, the pentameric B subunit of verotoxin 1, verotoxin 2, or verotoxin 2c, selectively binds with Gb3 in Gb3-containing cells. This is in contrast to the use of broad spectrum anti-neoplastic agents such as most chemotherapeutic agents, in that non-Gb3 containing cells are not affected by the agent, e.g., verotoxin. The present invention thus provides a most beneficial, cell-selective, therapeutic treatment.

The treatment is of value against cutaneous T-cell lymphomas, particularly, Mycosis Fungoides, sezary syndrome, related cutaneous disease lymphomatoid papilosis, and post transplant lymphoprohferative disorder (PTLD), e.g. PTLD following a transplant, e.g., a renal, heart, liver, or lung transplant. For example, Mycosis fungoides lesions in humans have been cleared without any observed adverse systemic effects by the application of VTI (5ng in 2 ml. solution) by interdermal injection in patients.

In a further aspect, the invention provides a method of treating mammalian neoplasia comprising treating said mammal with a non-lethal anti-neoplasia effective amount of an agent, e.g. a verotoxin, e.g., preferably verotoxin 1, the pentameric B subunit of verotoxin 1, verotoxin 2, or verotoxin 2c. In an alternate embodiment, the agent may be Pag adhesin linked to a toxin (e.g.. ricin) or an antibody to Gb3 linked to a toxin (e.g., ricin).

The agent, e.g., verotoxin or its B subunit, may be administered to the patient by methods well-known in the art, namely, intravenously, intra-arterially, topically, subcutaneously, by ingestion, intra-muscular injection, inhalation, and the like, as is appropriately suitable to the disease. For treatment of a skin cancer, sub-cutaneous application is preferred.

In the practice of the present invention, Verotoxin I has been injected intramuscularly into a patient with advanced ovarian carcinoma. No adverse affects were monitored on lymphocyte or renal function and a serum tumour marker was found to continue to rise when the patient was treated with relatively high doses of Verotoxin 1. This tumour was refractory to all conventional cancer therapies. No effect was found on hemoglobin levels. The agent, e.g., verotoxin or its B subunit, is, typically, administered in a suitable vehicle in which the active agent, e.g., verotoxin or B subunit, ingredient is either dissolved or suspended in a liquid, such as serum to permit the verotoxin to be delivered for example, in one aspect from the bloodstream or in an alternative aspect sub-cutaneously to the neoplastic cells. Alternative, for example, solutions are, typically, alcohol solutions, dimethyl sulfoxide solutions, or aqueous solutions containing, for example, polyethylene glycol containing, for example, polyethylene glycol 400, Cremophor-EL or Cyclodextrin. Such vehicles are well-known in the art, and useful for the purpose of delivering a pharmaceutical to the site of action.

Several multi-drug resistant cell lines were found to be hypersensitive to Verotoxin 1. For example, multidrug resistant ovarian cancer cell lines SKVLB and SKOVLC were more sensitive to VT cytotoxicity than corresponding non-multidrug resistant ovarian cancer cell line SKOV3. Such an observation indicates the possible beneficial effect for patients bearing the SKVLB cell line cancer than those with the SKOV3 cell line under VT treatment. Further, our observed binding of VTI to the lumen of blood vessels which vascularize the tumour mass, in addition to the tumour cells per se, may result in an anti-angiogenic effect to augment the direct anti-neoplastic effect of verotoxin. A series of human Gb containing astrocytoma cell lines were tested for sensitivity to VT. Although all cells were sensitive, the sensitivity varied over a 5000- fold range despite approximately equivalent Gb3 levels. We have found that treatment of the least sensitive cell line with sodium butyrate initiated a 5000-fold increase in VT sensitivity concomitant with an alteration in intracellular VT targeting.

Thus, we have also found that the efficacy of verotoxin and its B subunit may be significantly enhanced by a prior treatment of the neoplastic cells with a sensitizer, such as sodium butyrate.

Brief Description of the Drawings

In order that the invention may be better understood preferred embodiments will now be described, by way of example only, with reference to the accompanying drawings wherein: Figure 1 shows the selective neutralization of ACP cytotoxicity by anti VTI and or anti VTI B suburπt but not by anti VT2 antibodies as determined by cell density measurement after 48 hours;

Figure 2 shows the viability of selected ovarian and breast tumour cell lines to verotoxin concentration; Figure 3 represents VTI contained within ACP preparation binding to Gbβ (and

Gb₂);

Figure 4 represents VT thin layer chromatography overlay of ovarian tumour and ovary glycolipids;

Figure 5 represents VT thin layer chromatography overlay of selected cell line glycolipids;

Figure 6 represents in three graphs ovarian cell line sensitivity to VTI, VT2 and VT2c;

Figure 7 represents glioblastoma multiforme cell line sensitivity to VTI, VT2 and VT2c; Figure 8 represents the distribution of labelled VTI B subunit (VTB-¹³¹I) administered IP (inter-peridinually) in a Gb3 tumour bearing nude mouse;

Figure 9 represents the results of a three-day treatment of several human astrocytoma cell lines with VTI ;

Figures 10A - 10G represents a graph of the anti-proliferative effects of VTI on human astrocytoma cells;

Figures 11A and 1 IB provide a comparison of SF-539 and XF-498 sensitivity to VTI holotoxin;

-8- Figures 12A and 12B represent the detection of the VT-Receptor glycolipid, Gb3 in human astrocytoma cell lines;

Figure 13 shows the sensitivity of two astrocytoma cell lines to VTI after sensitizing culture; and Figure 14 shows the sensitivity to the B subunit of verotoxin VTI of the two cell lines used in tests shown in Figure 13.

Figures 15 a and b depict the FITC- VTB staining of LPD liver, showing the membrane staining of single cells dispersed throughout the tissue. Figures 15 c and d show the FITC- VTB staining of the LPD adenoid. Figure 15 e depicts background staining using FITC VT B of a normal liver. Figure 15 f depicts a normal liver using double labels, anti-CD20 and FITC VTB, it shows coincident expression of the lymphoid antigen and VT receptor.

Figure 16 depicts a fixed liver section processed for EBV nucleotide sequences using in situ hybridization. EBV positive cells are stained. Figure 17 depicts a gel of glycolipid extracts of EBV positive B cell lines, MB and TH9. The left panel of the figure shows orcinol detection and the right panel shows the VTI overlay which was used to detect the presence of Gb3.

Figure 18 depicts a graph of VTI cytotoxicity of MB and TH9 EBV transformed B cells.

Detailed Description of the Invention

The present invention pertains to methods and agents capable of inducing the death of lymphoma cells. "Lymphoma" is an art recognized term and refers to a potentially malignant neoplasm of the lymph and reticuloendoleitiial tissues. The term "lymphoma cell" is art recognized and includes the cells of the neoplasm as well as cells which have been exposed to the agent which caused the neoplasm or which could potentially cause a neoplasm.

One example of a common cause of lymphoma is the Epstein-Barr virus (EBV) which is a double stranded DNA virus of the herpesvirus family. The virus is transmitted by saliva, infects nasopharygeal epitheal cells and B lymphocytes, and is ubiquitous in human populations worldwide. It infects human B cells by binding specifically to the type 2 complement receptor (CR2) followed by receptor mediated endocytosis. Two types of cellular infections can occur. In a lytic infection, viral DNA, RNA and protein synthesis begin, followed by assembly of viral particles and lysis of the host cell. Alternatively, a latent non-lytic infection can occur, in which the viral DNA is incorporated into the host genome indefinitely. Various virally encoded

-9- antigens are detectable in infected cells. Epstein-Barr nuclear antigens (EBNAs) include at least four different nuclear proteins that are expressed early in lytic infections and may also be expressed by some latently infected cells. EBNAs are well-characterized EBV antigens which have been shown to be targets for specific cytolytic T lymphocytes (CTLs). Other viral structural protein antigens are expressed within infected cells and on released viral particles during lytic infections, including viral capsid antigens (VCAs). Antibodies specific for VCAs are present in acutely infected, recovering and remotely infected individuals.

Epstein-Barr virus has profound effects on B lymphocyte growth characteristics in vitro. First, the virus is a potent, T-cell independent polyclonal activator of B cell proliferation. Second, EBV can immortalize normal human B cells so that they will proliferate in culture indefinitely. The resulting long term B lymphoblastoid cell lines are latently infected with the virus and may express EBNA proteins, but they do not have a malignant phenotype (Abbas, A.K. et al, Cellular and Molecular Immunology, (W.B. Saunders Company: Philadelphia, 1991) p.343). EBV positive cells may also express Gb3.

B cell lymphomas, e.g. Burkitt's lymphoma and PTLD, occur at a high frequency in T-cell immunodeficient individuals, including individuals with congenital immunodefϊciences, AIDS patients, and kidney, renal or heart allograft recipients receiving immunosuppressive drugs. These individuals have deficiencies in normal T cell function. EBV infection proceeds unchecked in these individuals and EBV-induced polyclonal proliferation of B cells increases the chances of errors made by recombinases or isotype switching enzymes, resulting in a relatively high frequency of genetic translocations to Ig loci, potentially resulting in deregulation of genes and subsequent abnormal expression. B cell lymphomas, e.g. Burkitt's lymphoma and PTLD, are thought to be sequelae of unchecked EBV infections.

The present invention provides a method for inducing cell death, inhibiting protein synthesis, inducing apoptosis, in lymphoma cells, e.g., lymphoma cells of B cell origin, e.g., lymphoma cells of B cell origin which are EBV positive, and which bear Gb₃ receptors. Cells which express Gb₃ receptors are defined herein as "Gb₃ receptor positive cells." Cells which are of B cell origin are defined herein as cells which are descendent from B cells. Cells which are EBV positive are defined herein as cells expressing EBNA. These lymphomas include Burkitt's lymphoma and PTLD (post-transplant lymphoproliferative disease) e.g., PTLD involving renal and/or liver transplantation. PTLD includes disorders afflicting subjects who have undergone organ transplant surgery, e.g., renal transplantation, liver transplantation, and may be characterized by infiltrating lymphoma cells. Another aspect of the invention pertains to treating a subject, e.g., a human, having a disorder characterized by (or associated with) infiltrating lymphoma cells, e.g., lymphoma cells of B cell origin, e.g., lymphoma cells of B cell origin which are EBV positive. In one embodiment, the human is infected with HIV. These methods include the steps of administering an effective amount of an agent, e.g., a toxin of this invention, which is capable of inducing the cell death of the infiltrating lymphoma cells, e.g., by inhibiting the infiltrating lymphoma cell's protein synthesis, or by inducing apoptosis in the infiltrating lymphoma cell, such that treatment occurs. Non-limiting examples of disorders or diseases characterized by infiltrating lymphoma cells include lymphoma, e.g., Burkitt's lymphoma or PTLD, e.g., PTLD associated with renal, liver, heart, or lung transplant.

The term "disorder" includes a condition of a living organism or one of its parts which impairs normal or regular functioning, e.g., a disease, e.g., PTLD or Burkitt's lymphoma. The terms "treating" or "treatment" includes a reduction or alleviation of at least one adverse effect or symptom of a disorder or disease, e.g., a disorder or disease characterized by or associated with angiogenesis, a disorder characterized by infiltrating lymphoma cells.

The methods of the current invention also include administering an effective amount of an agent which binds Gb3 and induces cell death of infiltrating lymphoma cells, such that treatment occurs. The agent binds via Gb3 to cells which are involved in causing the disorder, e.g., infiltrating lymphoma cells, e.g., lymphoma cells of B cell origin, e.g., lymphoma cells of B cell origin which are EBV positive, and induces cell death. The agent may induce cell death of the cells causing the disorder through a variety of mechanisms. For example, the agent may, for example, bind the cell via Gb3, and become incorporated into the cell thus inducing cell death (e.g., by inhibition of protein synthesis, by induction of apoptosis).

Preferred agents capable of inducing cell death in infiltrating lymphoma cells include, among others, a verotoxin (VT). VTs, also known as SHIGA-like toxins, comprise a family known as VTI, VT2, VT2c, and VT2e of subunit toxins elaborated by some strains of E. coli. Cell toxicity is mediated via the binding of the B subunit of the holotoxin to Gb3- VTs are described in U.S. Patent Application Serial Number 08/563,394, entitled "Verotoxin Pharmaceutical Compositions and Medical Treatments Therewith", filed November 28, 1995. The isolation and purification of VTs have been earlier described. VTI can be prepared genetically from the high expression recombinant E. coli pJB28 (J. Bacteriol. 166:375 and 169:4313) and purified by protein purification procedures (FEMS MicrobioL Lett. 41 :63). VT2 can be obtained from R82 (Infect. Immun. 56:1926-1933 (1988)) and purified by protein purification procedures (FEMS MicrobioL Lett. 48:379- 383 (1987)). VT2c can be obtained from clinical strain E32511 and purified by protein purification procedures (FEMS MicrobioL Lett. 51:211-216 (1988)). VTI B subunit can be prepared according to Ramatour, et al. Biochem. J. 272:805-811 (1990).

The VTs consist of a 30kDa enzymatic subunit which is capable of inhibiting protein synthesis. The A subunit is noncovalently associated with a pentameric 7kDa B subunit array which binds to Gb_β. In addition to the cytotoxic effects of VTs on a wide range of cells by the A subunit inhibition of protein synthesis, VTI, and the receptor binding subunit alone, also induce morphological changes and DNA fragmentation characteristic of apoptosis in Gb3-positive cells.

Cell binding of the VTI B subunit alone can induce apoptosis in B cells and Gb3 containing B cells are prone to apoptosis during B-cell differentiation. Sensitivity to VTI is a function of cell cycle and cells at Gl/S boundary are particularly sensitive while stationary phase cells are refractory. Once internalized by receptor mediated endocytosi:**, Gb3-bound VTI can follow a unique pathway of intracellular retrograde transport to the Golgi/ER and nuclear membrane. Gb3 binding is involved in α-interferon receptor function, and in CD 19 signal transduction in germinal center B cells to mediate homotypic adhesion and apoptosis. Additional examples of agents capable of inducing cell death in infiltrating lymphoma cells include, among others, PagG adhesin (Kihlberg, et al. J. Am. Chem. Soc. 111:6364-6368 (1989) and antibodies to Gb3 or CD77 which can be linked to a toxin capable of inhibiting angiogenesis. Antibodies can be polyclonal, or more preferably, monoclonal. An intact antibody, or a fragment thereof (e.g., Fab or F(ab')2) can be used. A variety of monoclonal antibodies to Gb3 or CD77 are discussed in Oosterwijk, et al. (199V Int. J. Cancer 48(6):848-854; Kasai, et al. (1985) J. Immunogenet. 12(4-5):213-220; and Pallensen, et al. (1987,) J. Cancer Res. Clin. Oncol. 113(l):78-86, the contents of which are incorporated by reference herein. Also, anti- Gb3 is commercially available (AN 1003566, Biodesign International, Kennebunkport, ME, USA). Toxins which can be linked to these antibodies include, among others, VTs, and other immunotoxins known in the art, e.g., ricin.

The agents capable of inducing cell death in infiltrating lymphoma cells may be administered to the subject by methods well-known in the art, namely, intravenously, intra-arterially, topically, subcutaneously, by ingestion, intra-muscular injection, inhalation, and the like, as is appropriately suitable to the disease. For treatment of a skin cancer, subcutaneous application is preferred.

-12- The VT or its B subunit is typically administered in a suitable vehicle in which the active VT or B subunit ingredient is either dissolved or suspended in a liquid, such as serum to permit the VT to be delivered, for example, in one aspect from the bloodstream or in the alternative aspect subcutaneously to the cells. Alternatively, for example, solutions are typically alcohol solutions, dimethyl sulfoxide solutions, or aqueous solutions containing, for example, polyethylene glycol containing, for example, polyethylene glycol 400, Cremophor-EL, or Cyclodextrin. Such vehicles are well- known in the art and useful for the purpose of delivering a pharmaceutical to the site of action. The invention further provides a method for monitoring a previously diagnosed subject with a disorder characterized by abnormal cell proliferation, e.g., infiltrating lymphoma cells, e.g., lymphoma, e.g., Burkitt's lymphoma, PTLD, e.g., PTLD associated with liver transplantation, PTLD associated with renal transplantation. The method involves contacting a biological sample, e.g., a tissue sample, from the subject with an agent capable of detecting Gb3, e.g., fluorescently labeled VTI, determining the amount of Gb3 expressed in the sample, comparing the amount of Gb3 expressed in the sample to a the amount of Gb3 expressed in a sample previously obtained from the same subject to determine the progression of the disease, e.g., measuring the increase or decrease in levels of Gb3 over time in a subject. The following invention is further illustrated by the following examples, which should not be construed as further limiting. The contents of all references, pending patent applications and published patents cited throughout this application are hereby expressly incorporated by reference.

Exemplification of the Invention

EXAMPLE 1 : Isolation and Purification of Nerotoxins

Experimental

The isolation and purification of verotoxins VTI, VT2 and VT2c have been earlier described.

Verotoxin 1 was prepared genetically from the high expression recombinant E. coli pJB28, J. Bacteriol 166:37S and 169:4313. The generally protein purification procedure described in FEMS MicrobioL Lett_. 41 :63, was followed.

Verotoxin 2 was obtained from R82, Infect. Immun. 56:1926-1933; (1988); and purified according to FEMS MicrobioL Lett. 48:379-383 (1987). Verotoxin 2c was obtained from a clinical strain E32511 and purified according to FEMS MicrobioL Lett. 51 :211-216 (1988).

VTI B subunit was prepared according to Ramotar (24). VTs were aliquoted in PBS and stored at 70°C. The appropriate dilution for the treatment of astrocytoma cell lines was prepared freshly in media and added to the cells.

Purification of VTI from JB28

Pellet Preparation may be conducted as follows:

1. Prepare 6 x 1L LB broth in 3 x 5L jugs (media) and autoclave. Add carbenicillin to give a 100 μg/ml final cone, when cool. 2. Seed at least 6 ml of penassay (tubes in cold room) + 100 μg/ml carbenicillin with JB28 and incubate O/N at 37°C

3. Seed jugs (1 ml seed/litre broth) next morning and incubate for 24 hours at 37°C at 200 rpm (vigorous shaking).

4. Spin down bugs at 9K for 15 min. at 4°C and scrape pellet into a freezer bag for future use. Freeze at -70°C

Preparation of Crude Toxin Extract:

1. Retrieve pellet and dump into beaker. Resuspend in 400 ml of PBS containing 0.1 mg/ml polymyxin B, 50 mg PMSF using a blender. Blend thoroughly then sonicate on ice for - 1 minute to disperse further.

2. Incubate in shaking incubator, 200 rpm, or with vigorous stirring at 37°C for 1 hour.

3. Spin down cells at 9K for 15 minutes.

4. Pour off supernatant and keep. Resuspend pellet in 400 ml PBS with 0.1 mg/ml polymyxin B and PMSF. Blend and sonicate as before.

5. Incubate with vigorous shaking/stirring at 37°C for 1 hour.

6. Spin at 10K for 15 minutes and save supernatant.

7. The supernatants should be quite yellow and the bacterial pellet should become more fine and diffuse with each extraction step. 8. Filter the combined supernatants through Whatman filter paper than through a glass fibre filter to clarify. This step is optional, but will greatly speed he concentration step. 9. Filter the combined supernatants at 70 psi (max.) using an Amicon YM10 membrane (takes about 200 hours) to concentrate to < 50 ml.

Chromatography: Hydroxylapatite

1. Equilibrate hydroxylapatite column with 1 OmM K or Na phosphate (several column volumes). 2. Load sample and wash with equilibration buffer until absorbance of effluent is negligible. 3. Add 2 column volumes (150 ml) of lOOmM K phosphate (until yellow-coloured fractions emerge) and collect 3 ml fractions.

4. Wash column with 500mM K phosphate and re-equilibrate with 10 mM K phosphate. Add 0.05% sodium azide.

Chromatofocussing

5. Measure fractions (A280) and Pool peak fractions from HA.

6. Dialyse against 2L 0.025M imidazole-HCl pH 7.4 O/N. Also equilibrate the chromatofocussing column O/N with the same (300 ml).

7. Load sample and follow with 400 ml polybuffer-HCl pH 5.0 (50ml polybuffer 74 + 350ml dH₂0, a 1 :7 dilution, -pH to 5.0 with HC1). NOTE: make sure the sample is equilibrated to the temperature that the column will be run at (usually room temperature) prior to loading. If the column is to be run at 4° then buffers must be pH'd at 4°C and the column equilibrated at this temperature.

8. Collect 1 ml fractions and test them for A28O and pH. 9. Plot the A28O and pool peak fractions at about pH 6.8 for VT-1 (pool side peaks separately). 10. Clean column with 100 ml IM NaCl. If really dirty follow with 100 ml IM HC1, but quickly equilibrate column with imidazole. Store column in 20% ethanol in 25mnM imidazole.

Cibachron blue

11. Equilibrate cibachron blue with 10 mM Na phosphate buffer, pH 7.2 (100ml).

12. Load sample directly from CF and follow with 60ml of same buffer.

13. Elute with 0.5M NaCl in above buffer and collect fractions. 14. Test fractions for N280 and cytotoxicity and pool appropriate ones.

15. Clean column with 25ml each of 8M Urea in wash buffer and IM NaCl in wash buffer.

16. Reequilibrate column with 1 OmM phosphate containing 0.1 % azide.

17. Dialyse peak fractions against wash buffer with one change. 18. Lyophilize and resuspend in 1 ml dH2θ.

19. Do protein assay and run SDS-PAGE to check purity.

Solutions: HA column potassium phosphate buffer (0.5M stock)

17.42g K₂HPO₄ up to 300 ml with dH2θ

6.8g KH₂PO₄ pH 7.2 with KOH CF column imidazole buffer

0.851g/500 ml H₂O pH 7.4 with HC1 CB column sodium phosphate buffer (Wash buffer- WB) 0.71g/500ml Na₂HPO₄ pH 7.2 with HAc degas Elution buffer Cleaning Buffers

2.922g NaCl/100 ml WB 12.012g Urea/25ml WB

1.461g NaCl/25ml WB

Purification of VT2 from R82 Pellet Preparation:

1. Prepare 3 x 2L penassay broth (Antibiotic Meida 3, DIFCO; pH 7.0) in 3 x 5L jugs and autoclave at 121°C for 20 minutes. Allow broth to cool to room temperature before use.

2. Seed minimum 3 x 2ml of penassay broth containing 75 μg/ml carbenicillin (Disodium salt, SIGMA) with R82 and incubate overnight at 37°C, with shaking.

3. Add 50 μg/ml carbenicillin to each of the 5L jugs (from step 1). Seed each jug with 2 ml of seed (step 2) and incubate for 24 hours at 37°C with shaking of approximately 120 rpm. 4. Heat incubator to 45°C and incubate for 30 minutes.

5. Reduce temperature to 37°C and incubate for another 3 hrs.

6. Spin down culture solution at 9,000xg for 15-20 min at 4°C. Discard supernatant and store pellets at -20°C

Preparation of Crude Toxin Extract:

1. Resuspend pellets in 100 ml of PBS (phosphate buffered saline, OXOID; pH 7.3).

2. Add 0.3 mg/ml PMSF (phenylmethyl-sulfonyl fluoride, SIGMA) dissolved in 0.5 ml acetone to pellet solution. Let acetone evaporate. Sonicate on ice at highest output possible for 5 min or until an homogeneous solution is obtained.

3. Spin down cell at 9,000xg at 4°C for 20 min. Discard pellets.

4. Concentrate supernatants using ultrafiltration (Model 8400 standard infiltration cell, AMICON) with N₂ no higher than 70 psi and using a 10,000 MW cutoff membrane filter (YM10 membrane, AMICON). 5. Using 12-14,000 MW cutoff tubing (SPECTRAPOR) (now and in all dialysis steps), dialyse toxin solution against 4L of 10M potassium phosphate overnight, with stirring at 4°C.

Chromatography: Hydroxylapatite (HA)

1. Equilibrate hydroxylapatite column (BSA binding capacity: 32 mg/g, approximately 113 ml volume; CALBIOCHEM (BEHRING DIAGNOSTICS)) with 2 column volumes of 1 OmM potassium phosphate.

2. Load sample and follow with 1 column volume lOmM potassium phosphate.

3. Add 2 column volumes of 200mM potassium phosphate and collect 2 ml fractions. The fractions containing the toxin should be coloured differently from the other fractions.

4. Wash column with 1 column volume of 500mM potassium phosphate and reequilibrate with 1 column volume of lOmM potassium phosphate. Add azide to the top of the column for storage.

Chromatofocussing (CF)

5. Pool peak fractions from HA column either by color or by cytotoxicity test on Vero cells (10-fold dilutions).

6. Dialyse pooled fractions against 4L 0.025M Histidine-HCl pH 6.2 (SIGMA) overnight. Also equilibrate the chromatofocussing column (PBE (polybuffer exchanger) 94, 1.5 cm diameter, 57 ml volume; PHARMACIA) overnight with the same buffer (300 ml).

7. Loan sample and follow with 400 ml polybuffer-HCl pH 4.0 (50 ml polybuffer 74 (PHARMACIA) + 350 ml dH₂0 - pH to 4.0 with HC1).

8. Collect 2 ml fractions and test the pH of each fraction. Once the pH has dropped to 3.95, stop collecting fractions. Test the fractions using absorbance of 280 n or by cytotoxicity on Vero cells (10-fold dilutions).

9. Pool peak fractions, and return pH to 7.0 using IN NaOH.

10. Clean column with 200 ml IM NaCl. If dirty follow with 100ml IM HC1, but quickly equilibrate column with 0.025M imidazole, otherwise equilibrate with 24% EtOH-H2θ.

Cibachron blue (CB) 11. Equilibrate cibachron blue (2 cm diameter, 82 ml volume, PIERCE) with 100 ml of lOmM sodium phosphate buffer (wash buffer).

12. Load sample and follow with 60 ml of wash buffer.

13. Elute with 0.5M NaCl in wash buffer and collect 2 ml fractions. 14. Test fractions for absorbance at 280 nm using the elusion buffer as a blank and cytotoxicity on Vero cells and pool appropriate fractions.

15. Clean column with 25 ml each of 8M Urea in wash buffer and IM NaCl in wash buffer.

16. Reequilibrate column with 100 ml of wash buffer and add azide to the top of the column for storage.

17. Dialyse peak fractions against 4L 0.0 IM Tris-Cl (pH 7.0, SIGMA).

18. Lyophilize sample and resuspend in 1-2 ml dH2θ (OPTIONAL).

19. Do protein assay (BCA Protein assay reagent, PIERCE) and rune SDS-PAGE gel (Schagger, H. and von Jagow, G.; Analytical Biochem 166, 368-379 (1987): 10%) T table 2; first line table 3) to check purity.

Solutions: HA Column potassium phosphate buffer (0.5M stock) 17.42g K2HPO4 up to 300 ml with dH₂O

6.8g KH₂PO₄ pH 7.2 with KOH

CF column

Histidine buffer (0.025M) 2.0g/500 ml H O pH 6.2 with HC1

CB column

Sodium phosphate buffer (Wash buffer- WB) 0.71g/500ml Na₂HPO₄ pH 7.2 with HAc, degassed

Elution buffer (0.5M) Cleaning Buffers

2,922g NaCl/100 ml WB 12.01g Urea/25 ml WB

1.46 NaCl/25 ml WB

0.01 M Tris 4.84 g Trizma Base

4 L ddH2θ pH to 7.2 with HC1

-19- Purification of VT2c from E32511 Pellet Preparation:

1. Prepare 3 x 2L penassay broth (Antibiotic Media 3, DIFCO; pH 7.0) in 3 x 5L jugs and autoclave at 121 °C for 20 minutes. Allow broth to cool to room temperature before use.

2. Seed minimum 3 x 2 ml of penassay broth with E32511 and incubate overnight at 37°C

3. Add 0.2 μg/ml Mitomycin C (1 ml of 0.4 mg/ml) (add 5 ml of ddH₂O to the vial) to each of the 5L jugs (from step 1). Seed each jug with 2 ml of seed (step

2) and incubate for 6 hrs at 37°C with shaking of approximately 120 rpm. It is very important to stagger the incubation by about 45 min/flask because the toxin begins to deteriorate after 6 hour exposure to Mitomycin C.

4. Spin down culture solution at 9,000xg for 15-20 min at 4°C Discard supernatant and store pellets at -20°C

Preparation of Crude Toxin Extract:

1. Resuspend pellets in 150 ml of PBS (Phosphate buffered saline, OXOID); pH

7.3). 2. Add 0.3 mg/ml PMSF (phenylmethyl-sulfonyl fluoride, SIGMA) dissolved in

0.5 ml acetone to pellet solution. Let acetone evaporate. Sonicate on ice at highest output possible for 3 min or until an homogeneous solution is obtained.

3. Add 0.1 mg/ml polymyxin B sulphate (Aerosporin, BURROUGHS WELLCOME INC.; 500,000 units) to solution and incubate with gentle shaking at 37°C for 1 hr.

4. Spin down cells at 9,000xg at 4°C for 20 min (to remove all cells and cell debris from solution).

5. Decant supernatant and store at 4°C Resuspend pellet in 75 ml PBS and add 0.1 mg/ml polymyxin B. 6. Incubate with gentle shaking at 37°C for 1 hr.

7. Spin down cell at 9,000xg at 4°C for 20 min and pool supernatants (from step

5). Discard pellets.

The next few steps should preferably be done at 4°C: 8. Add crystalline ammonium sulphate very slowly, with stirring to pooled supernatants to 30%> saturation. 9. Let stir for 20 min and then remove precipitate by centrifugation (lOOOOg for 10 min).

10. Add crystalline ammonium sulphate very slowly, with stirring to pooled supernatants to 70% saturation.

11. Let stir for 20 min and then centrifuge at 1 OOOOg for 10 min. 12. Resuspend pellet from step 11 in 15 ml of 0.01 M Potassium phosphate buffer. 13. Using 12-14,000 MW cutoff tubing (SPECTRAPOR) (now and in all dialysis steps), dialyse toxin solution against 4L of lOmM potassium phosphate overnight, with stirring at 4°C

Chromatography:

Hydroxylapatite (HA)

1. Equilibrate hydroxylapatite column (BSA binding capacity: 32 mg/g, approximately 113 ml volume; CALBIOCHEM (BEHRING DIAGNOSTICS)) with 2 column volumes of lOmM potassium phosphate. 2. Load sample and follow with 1 column volume lOmM potassium phosphate.

3. Add 2 column volumes of 100mM-200mM potassium phosphate and collect 2 ml fractions. The fractions containing the toxin should be coloured differently from the other fractions.

4. Wash column with 1 column volume of 500mM potassium phosphate and reequilibrate with 1 column volume of lOmM K phosphate. Add azide to the top of the column for storage.

Chromatofocussing (CF)

6. Dialyse pooled fractions against 4L 0.025M imidazole-HCl pH 7.4 (SIGMA) overnight. Also equilibrate the chromatofocussing column (PBE (polybuffer exchanger) 94, 1.5 cm diameter, 57 ml volume; PHARMACIA) overnight with the same buffer (300 ml). 7. Load sample and follow with 200 ml polybuffer-HCl pH 5.0 (25 ml polybuffer

74 (PHARMACIA) + 175 ml dH₂O - pH to 5.0 with HC1). 8. Collect 2 ml fractions and test the pH of each fraction. Once the pH has dropped to 5.95, stop collecting fractions. Test the fractions for cytotoxicity on Vero cells (10-fold dilutions). 9. Pool peak fractions.

10. Clean column with 200 ml IM NaCl. If really dirty follow with 100 ml lm HC1, but quickly equilibrate column with 0.025M imidazole. Cibachron blue (CB)

11. Equilibrate cibachron blue (2 cm diameter, 82 ml volume, PIERCE) with 100 ml of lOmM sodium phosphate buffer (wash buffer). 12. Load sample and follow with 60 ml of wash buffer.

13. Elute with 0.5M NaCl in wash buffer and collect 2 ml fractions.

14. Test fractions for absorbance at 280 nm using the elusion buffer as a blank and cytotoxicity on Vero cells and pool appropriate fractions.

17. Dialyse peak fractions against 4L 0.01M Tris-CL (pH 7.0, SIGMA).

18. Lyophilize sample and resuspend in 1-2 ml dH2θ (OPTIONAL). 19. Do protein assay (BCA Protein assay reagent, PIERCE) and run SDS-PAGE gel (Schagger, H. and von Jagow, G.; Analytical Biochem 166, 368-379 (1987): 10%> T table 2; first line table 3) to check purity.

Solutions: HA column potassium phosphate buffer (0 5M stock)

17.42g K2HPO4 up to 300 ml with dH O

6.8g KH2PO4 pH 7.2 with KOH

CF column imidazole buffer (0.025M)

0.851g/500 ml H₂O pH 7.4 with HC1 CB column sodium phosphate buffer (Wash buffer- WB) 0.71g/500 ml Na₂HPO4 pH 7.2 with HAc degas

Elution buffer Cleaning buffers 2.922 g NaCl/lOOml WB 12.012g Urea/25ml WB

1.461 g NaCl/25ml WB

-22- 4.84 g Trizma Base 4 L ddH2θ pH to 7.2 with HCl

Affinity purification of verotoxins

500μg globotriaosyl ceramide in 1 ml chloroform was mixed and dried with lg of dried celite. The chloroform was evaporated and the celite suspended in PBS and poured in a column. Crude polymyxin extract 20 ml (25 mg protein) the toxin producing E. coli was applied to the column and incubated at room temp for 15 mins. The column was washed with PBS and purified verotoxin eluted with 10 ml IM Tris pH 9.6. The eluate was neutralized and dialysed. This method is applicable for purification of all verotoxins. [Boulanger, J., Huesca M., Arab, S and Lingwood, CA. "Universal method for the facile production of glycolipid/lipid matrices for the affinity purification of binding ligands" Anal Biochem 217: 1-6 (1994)]

EXAMPLE 2: Cell Based Assays of Verotoxin Activity

Preparation of verotoxin 1 doses VTI was purified from the E. coli strain as previously described which overexpresses the cloned toxin genes. The purified toxin was free of endotoxin contamination. The protein concentration of this batch of verotoxin was determined and the toxin aliquoted and stored at -70°C.

To prepare VTI doses for patients, VTI was diluted into injection grade sterile saline containing 0.2%> v/v of the patient's own serum. 210 μl of sterile patient serum was added to 10 ml of sterile injection saline and 93.9 ml of purified VTI (6.7 g/ml) added to give a final toxin concentration of 62.5 ng/ml or 12.5 ng per 0.2 ml. dose. The final toxin preparation was sterile-filtered using a 0.2 mm syringe filter and dispensed in 2 ml aliquots into 10 ml vials. One working vial may be stored at 4°C and the remaining vials frozen until needed.

FITC labeling of VTI : FITC was added directly to VTI (in a 1 :1, w/w ratio) in 0.5M Na2CO3/NaHCO3 conjugated buffer pH 9.5 and the mixture gently rotated for 1.2 hours at room temperature. Free FITC was removed by a Centricon filter.

Fluorescent Staining of Sections: Samples of surgically removed ovarian tumours were embedded in OCT compound, flash frozen in liquid nitrogen, and stored at -70°C until

-23- use. Five μm sections of frozen sample were thawed, allowed to dry and stained with FITC-labelled VTI in PBS (0.5 mg/ml) containing 0.1% BSA for 1 h at room temperature. Sections were extensively washed with PBS and mounted with mounting medium containing DABCO. Sections were observed under a Polyvar fluorescent microscope.

Fluorescent Staining of Cells: Cells growing on coverslips were washed once with PBS, fixed for 2 min at room temperature with 2% formalin rinsed with PBS twice and incubated with FITC-VT1 for lh at room temperature. The cells were washed 5 times with PBS, mounted with DABCO and observed under a Polyvar fluorescent microscope.

Quantification of VTI antitumour activity: SKOV3 (drug sensitive human ovarian cell line), SKOVLC (SKOV3, resistant to Vincristine, and SKOVLB (SKOV3, resistant to Vinblastine) were each grown in α - MEM supplemented with 10%_> fetal calf-serum and tested for their sensitivity to VTs. Equal numbers of cells (approximately 1000 per/ml of media) were adde to the wells of Linbro 98 well plate. 10-fold dilution of VTs were tested in triplicate and incubated for 48h at 37°C in a humidified atmosphere containing 5% CO₂. Cells were then fixed with 2% Formalin, stained with Crystal Violet, and read with ELISA plate reader. To quantify the anticancer activity of VT 1 , SKOV3 , SKOVLC, and SKOVLB

(human ovarian cell line) were incubated with 10-fold dilution of VTI for 48h. SKOVLC & SKOVLB (drug resistant cell lines) are more sensitive to VTI antitumour activity than SKOV3.

Preparation of ¹³¹I-VT1B

This material may be made by the following procedure. 1. Dissolve 20 mg of iodogen in 2.0 ml of chloroform (10 mg/ml). Make a 1:10 dilution by adding 0.25 ml of the 10 mg/ml solution to 2.25 ml chloroform (1 mg/ml). 2. Dispense 20 μl of this dilute solution into a clean, dry sterilized glass tub. Add

500 μl of chloroform and evaporate to dryness under N₂.

3. Add 1.5 mg. in 0.66 ml of VTI B subunit to the test tube.

4. Add 5 MCi of 1311 sodium iodide in 100 μl. Allow labelling to proceed for 10 mins. 5. Wash a PD-10 column with 25 ml of Sodium Chloride Injection USP.

6. Dilute ¹³ ll-VTIB to 2.5 ml total volume with 1% HSA in Sodium Chloride

Injection USP. Load onto PD-10 column. Elute column with 3.5 ml 1% HSA

-24- in saline. 7. Measure I³ l activity of eluant and column to determine LE. Draw up pooled fractions into a syringe with spinal needle attached. Detach spinal needle and attach Millex GV filter. 8. Filter into a sterile 10 ml multidose vial. Note volume filtered and assay vial for

131l in dose calibrator. Calculate concentration. 9. Draw up 0.1 ml of I³ l-VT1B and dispense 0.05 ml into each of two 5 ml sterile multidose vials (one for sterility test and one for pyrogen test). Vials already contain 2 ml saline (=1:50 dilution). 10. Determine RCP by PC (Whatman No. 1 ) in 85%* MeOH and by size exclusion

HPLC 11. Conduct sterility and pyrogen tests.

Astrocytoma Cell Lines. Endothelial Cells and Culture Conditions:

Six permanent human malignant astrocytoma cell lines (SF-126, SF-188, SF- 539, U 87-MG, U 251-MG, and XF-498) were selected for study. SF-126, SF-188, and SF-539 were kindly provided by Dr. Mark Rosenblum, Henry Ford Hospital. U 87-MG and U 251-MG were kindly provided by Dr. Jan Ponten, University of Uppsala, Sweden; and XF-498 was a gift of Dolores Dougherty, University of California San Francisco. Astrocytoma cells were cultured in alpha-MEM, nonessential amino acids, glutamine, gentamycin, and 10% heat-inactivated fetal bovine serum. The cultures were incubated at 37°C and equilibrated in 5%> CO₂ and air. Cells were harvested with 0.25%> trypsin (Gibco, Santa Clara, CA) in Ca^++~ and Mg^++~ free Hank's balanced salt solution and were subcultured weekly.

Human capillary endothelial cells were isolated after the method of Costello (25) and were derived from samples of normal human brain taken from patients undergoing neurosurgical procedures for epilepsy, trauma, and resection of arteriovenous malformations. The capillary cells were grown as described above in media supplemented with 15 μg/ml endothelial growth factor (Sigma, St. Louis) (26). The endothelial origin of the cells in culture was established by immunocytochemical analysis using anti-human factor- VHI-related antigen antisera (Dako, Santa Barbara, CA) as described previously (27).

Approximately 1-5 x 10^ cells were added to 24-well plates and incubated in MEM in 5%> CO₂ at 37°C After 24 hours, the growth medium was replaced with medium containing various concentrations of the holotoxin VTI (0, 0.1, 5, 50, and 100 ng/ml). The treated astrocytoma cell lines and endothelial cells were trypsinized and counted at intervals throughout the growth curve. Cell viability was assessed by trypan blue dye exclusion. Cell counts were plotted again time for the various concentrations of VTI and B subunit. For each time point analyzed, the wells were set-up in triplicate.

For selected cell lines, the B subunit of VTI, VT2, and VT2c was added alone to the astrocytoma cells at same concentrations listed above. In these experiments, a single dose of VTI, VT2, and VT2c was added to confluent astrocytoma cells in microtiter wells. Cell survival at 72 hours was monitored by staining with 0.1%) crystal violet, and measuring the optical density at 590 nm using a Dynatek microtiter plate reader.

VT Receptor Analysis of Human Astrocytoma Cells:

Cultured human astrocytoma cells were homogenized in a minimum volume of PBS and extracted with 20 volumes of a 2:1 by volume chloroform:methanol solution. The extract was partitioned against water and the lower phase partitioned again against theoretical upper phase. The lower phase was dried completely and dissolved in a known volume of 2: 1 chloroform:methanol. The presence of Gb3 was detected by TLC overlay binding with VTI . Astrocytoma lower phase and standard Gb3 from human kidney each were separated by TLC [(chloroform:methanol:water •= 65:25:4 (v/v/v)]. The TLC plates were dried and blocked with 1%> gelatin in water at 37°C overnight. Then they were washed three times with 50 mm TBS (Tris Buffer Salin) for 5 min and incubated with 0.1 μg/ml VTI for 1 hour. After further washing with TBS, the plates were incubated with a mouse monoclonal PHI and anti -VTI antibody (2 μg/ml), followed, after washing, by peroxidase-conjugated goat anti-mouse antibody or peroxidase conjugated goat anti-rabbit antibody as appropriate. Finally, the plates were washed with TBS, and VTI binding was visualized with 4-chloro-l naphthol peroxidase substrate. A similar plate was prepared and stained with orcinol carbohydrate spray for comparison.

Nuclear staining with propidium iodide:

SF-539 cells grown on the cover slips overnight were incubated at 37°C with VTI B-subunit (50 μg/ml) for 1.5 hrs or 10 hrs and fixed (with 1%> paraformaldehyde for 3 minutes), permeabilized with 0.1 %> Triton X in 100 mm PBS for 5 min, and stained with 5 μg/ml propidium iodide (Sigma). After extensive wash with 50 mm PBS, the fixed cells were mounted with DABCO (1 ,4-Diazabicyclo-Octane, sigma), and nuclear staining observed under incident uv illumination.

Flow Cytometry:

Apoptosis of astrocytoma cells, incubated with 10 ng/ml of VTI for 24-36 hrs in the presence of 10%> bovine fetal serum was analyzed on an Epics Profile Analyzer (Coulter Electronics, Pathology, University of Toronto). After treatment, cells were trypsinized and the 200Xg centrifuged cell pellet was suspended in 1ml of hypotonic fluorochrome solution of 50 μg/ml propidium iodide (Sigma) and stained for 30 min at 4 C. To remove RNA prior to staining, cells were treated with 100 μl of 200 μg/ml DNase-free RNase A at 37 C for 30 min. Cell cycle distribution was determined using manual gating. Flow cytometric quantitation of apoptotic cells within the propidium iodide-stained population was performed. Debris and dead cells were excluded on the basis of their forward and side light-scattering properties. Astrocytoma cells grown simultaneously in the absence of VTI served as controls.

Ultrastructural Analysis of VT-treated Astrocytoma Cells: Cells were cultivated on a transferable 9 mm cyclopore membrane (0.45 μm pore size, Falcon) to form a confluent monolayer and were incubated at 37°C with VTI (10 ng/ml). Cells were fixed at room temperature by addition of 1.6% glutaraldehyde to the well and then incubated in 0.066 M Sorensen buffer (pH 7.4) containing 1.5% glutaraldehyde for 1 h at 4°C After 2 h of washing with 0.1 M phosphate buffer, cells were post- fixed in 2% osmium tetroxide in the same buffer. After dehydration in graded ethanols and propylene oxide, Epon embedding and uranyl-lead staining were performed. Thin sections were examined in a Philips EM 400 electron microscope and ultrastructural features of apoptosis was analysed.

Fig. 1 relates to the neutralization of ACP cytotoxicity by anti-VT. KHT cell monolayers were incubated with 35 ng/ml ACP from E.coli HSCi Q, or 10 pg/ml VTI, VT2 or VT2c in the rresence of monoclonal anti-VTl(PHl), monoclonal anti VT2 or polyclonal rabbit anti VTI B subunit. The cells were incubated for 72 hours at 37°C and viable adherent cells were detected by fixation and staining with crystal violet. Cytotoxity of VTI and ACP was completely neutralized in the presence of anti VTI or anti VT1B subunit (anti-VT2 serum had no effect).

From measurement of the cytotoxic assay of ACP on vero cells (cells from Africa green monkey kidney that are very sensitive to verotoxin), relative to a pure VTI standard, it was estimated that the ACP preparation contained 0.05%> VTI. This concentration of purified VTI was as effective as ACP in inhibiting the growth of various tumour cell lines in vitro (Fig. 2). Thus, VTI mimics the anti-neoplastic effect of ACP in vitro. VTI was tested for the ability to inhibit the metastases of KHT fibrosarcoma cells in the mouse model as had been previously reported for ACP. The equivalent dose of VTI was as effective as ACP, reducing the number of lung metastases to background levels, following a primary subcutaneous tumour inoculum (Table 1).

Table 1. Response of KHT cells, growing as lung modules, to treatment with VTI or ACP.

Mice were treated with VT-1 or ACP(l-p) 1 day after cell injection (1000 KHT cells/mouse i-v).

Lung nodules counted (@ 20 days after cell injection.

* Mean change in gp wi-max during 10 days (Expl 1) or 4 days (Expt 2) after VT-1 or ACP injection. Max wt loss @ days 7-8. ** Death occurred @ days 2-3 after ACP injection *** Deaths occurred @ days 7-8

Purified VTI was found to mimic the anti-metastatic effect of ACP on the growth of this tumour from a primary subcutaneous site. Lung metastasis was completely inhibited. Moreover, prior immunization of mice with the purified B-subunit of verotoxin completely prevented any protective effect of ACP when the animals were subsequently treated with the tumour and ACP (Table 2). Table 2. Response of KHT lung nodules, growing to immunized mice, to treatment with VTI or ACP.

Mice were treated with VT- 1 or ACP(i-p) 1 day after cell injection (1000 KHT cells/mouse).

Lung nodules counted (@ 20 days after cell injection (i-v).

*Immunization was 2 injections of VT-1 B (lOug/mouse +/- Freund's Adjuvant (FA) given (i-p) 4 weeks and 2 weeks before cell injection.

** Mean change in gp wt - max during 13 days. Maximum weight loss @ day 7-8.

ACP was tested for glycolipid binding by TLC overlay using monoclonal anti-

VT1 or anti-VT2c. Anti-VTl shows extensive binding of a component within the ACP preparation to globotriaosylceramide and galabiosyl ceramide (Fig. 3). This binding specificity is identical to that reported for purified VT1(8). No binding component reactive with anti-VT2 was detected. In Fig. 3 anti VT antibodies were used to detect binding to the immobilized glycolipids. Arrows indicate position of standard (from the

-30- top) galabiosyl ceramide, globotriaosyl ceramide and globotetraosyl ceramide. Panel 1- detection using anti VTI, panel 2-detection using anti VT2c.

VTI demonstrated in vitro activity against a variety of ovarian carcinoma cell lines. A large number of primary human ovarian tumour biopsies were screened for the expression of Gb3 via TLC overlay using purified VTI . It was found that Gb3 was barely detectable in normal ovary tissue, whereas in all cases a significant increase in expression of Gb3 was observed in the ovarian carcinoma. Similarly, elevated levels of Gb3 were found in acites tumour and in tumours that had metastized to the omentum, (Fig. 4) which defines lane 1, ovarian omentum metastasis; lane 2: tumour biopsy; lane 3, tumour biopsy; lanes 3-6, normal ovary; lane 7, human kidney Gb3 standard.

Surprisingly, we have found that multi-drug resistant variants of ovarian tumour cell lines were considerably more sensitive to VTI cytotoxicity than the drug sensitive parental cell line (Figs 2, 5 and 6). Similar effects had been observed for ACP. Fig 2 shows human ovarian tumour cell lines sensitive to ACP tested for VT sensitivity. Human ovarian and breast tumour derived cell lines were tested for VTI sensitivity wherein ovarian I, 2, 3, 4 and 5 are denoted π, +, x, N and o respectively, and breast- SKBR3 , 468 ♦, 453 >, 231 ^A . The cell lines 1-ovarian, 453 and SKBR3, previously shown to be resistant to ACP, were also resistant to up to 20 ng/ml VTI.

The 1, 2, 3 and 4 cells were from ovarian cancer patients; the 453 cells were from a breast cancer patient; 231 and SKBR3 are breast adenocarcinoma cell lines, and 5, SKOV3 and SKOVLB are adenomacarcinous ovarian cancer cell lines. The lines 1, 453 and SKBR3, resistant to ACP, were co-resistant to VTI. Fig. 5 shows VT sensitive and resistant cell lines tested for the presence of Gb3 by VT binding in tic overlay. Glycolipid from an equal number of cells were extracted and separated by tic prior to toxin binding. In Fig. 5, lane 1:SKBR3, lane 2:468, lane 3:231, lane 4:453, lane 5 Gb3 standard, lane 6:SKOV3, lane 7:SKOVLB. Cell lines SKBR3, 468, 231 and 453 are derived from breast tumours. Only 231 is sensitive to VTI. SKOVLB is a multiple drug resistant ovarian tumour cell line derived from SKOV3.

Ovarian tumour cells were highly sensitive to VT (Fig. 3) and contained elevated levels of the VT receptor, Gb3 (Fig. 4). Breast cancer cells were for the most part, toxin resistant (Fig. 3) and receptor negative (Fig. 5). Low levels of Gb3 were detected in normal ovarian tissue but these were markedly elevated for the ovarian tumour tissue samples.

The specific elevation of Gb3 in ovarian tumours as opposed to normal ovary tissue provides the feasibility of using the toxin in the management of this malignancy. Ovarian tumours are often refractory to chemotherapy and prognosis is poor. Indeed, preliminary phase 1 clinical trials using a ACP injected directly into skin malignancies (Mycosis fungoides) have proven successful without adverse systemic effects.

With reference now to Fig. 6, human derived ovarian tumour cell lines were tested for VTI, VT2, and VT2c sensitivity. The cells were grown to confluence in 48- well plates, then incubated for 48 hrs. in the presence of increasing doses of VTs.

SKOVLB, the multiple drug resistant variant of SKOV3 ovarian line, showed the most sensitivity to VT's with SKOVLC being the next most sensitive to the VT's.

We have found that both drug resistant cells are approximately 500 to 1000 times more sensitive to verotoxin cytotoxicity than the parental SKOV3 cell line. Fig. 7 shows the effect after 48 hrs. of treatment of the brain tumour SF-539 cell line derived from a recurrent, right temporoparictal glioblastoma multiform with VTI, VT2, and VT2c. This cell line, as others, was highly sensitive to VT's.

Fig. 8 provides the results from imaging a nude mouse with 131I-VT1B (CPM distribution in different organs). VTlB-l^ l cpm distribution in nude mouse with implanted ovarian tumour showed that a considerable amount of radiolabled VTI B had been concentrated ir the ovarian tumour. Only a trace amount of VTI B was located in the brain where the potential VTI side effect was considered. Since the lung in human adult is not the site of concern for VTI toxicity this does not present a problem for treatment of human adult with ovarian tumour. In addition the CPM in kidney includes the excreted radiolabelled VTI B subunit. Accordingly, based on this test, imaging with labelled VTI B subunit can be a very useful method for screening the susceptible patient to VTI cytotoxiciry.

Fig. 9 shows the sensitivity of a variety of human astrocytoma cell lines to VTI. All these cells contain Gb3 but show variable sensitivity to VTI induced cytotoxicity. This suggests that certain astrocytomas will be more susceptible to verotoxin than other astrocytomas. This is important since astrocytomas are very refractory to treatment at the present time and cell sensitivity in vitro to concentrations as low as 5ng per/ml is rare.

Figs. 10A - 10G show the anti -pro liferative effects of VTI on human astrocytoma cells. All astrocytoma cell lines showed at least some inhibition of growth following VTI treatment. The most sensitive cell line was SF-539 (Fig. 10A), and the least sensitive was SF-126 (Fig. 10F). Human cerebral capillary endothelial cells were largely resistant to the growth-inhibitory effects of VTI except at high doses (100 ng/ml) (Figs. 10G). U-251 MG and U-87 MG were sensitive to VTI (Figs. 10B and 10C), whereas XF 498 and SF-188 were somewhat less sensitive to VTI (Figs. 10D, 10E and IE) than were U-251 MG and U-87 MG. Figs. 11 A and 1 IB provide a comparison of SF-539 and XF-498 sensitivity to VTI holotoxin (upper panel) and B-subunit (lower panel). Forty-eight hrs following the treatment of SF-539 and XF-498 cells in monolayer culture, the percent cell survival was calculated. VTI was cytotoxic to SF-539 astrocytoma cells at doses as low as 0.01 ng/ml (upper panel). XF-498 cells were resistant to VTI holotoxin. When the VTI B- subunit was employed, only SF-539 was sensitive to this toxin (lower panel).

Figs. 12 A and 12B represent the detection of the VT-Receptor glycolipid, Gb3 in human astrocytoma cell lines. Astrocytoma neutral glycolipids were prepared from 1 x 106 cells and separated by TLC. (A) Glycolipids were visualized by orcinol and bands representing Gb3 are seen in all astrocytoma cell lines. (B) The same blot was assayed by VTI overlay. In this study, VTI binds to Gb3 extracted from astrocytoma cells as shown (arrow). SF-539 astrocytoma cells showed maximal binding of Gb3/YT1. Lane 1, U87 MG; lane 2, U251 MG; lane 3, SF-126; lane 4, SF-188; lane 5, XF-498; lane 6, SF-539, lane 7, standard Gb₃ (0.3 ug/ml). Fig. 13 compares the sensitivity of two astrocytoma cell lines SF539 (sensitive),

XF498 (less sensitive) and XF 498, following three days of culture of XF498 in sodium butyrate. It is seen that the sensitivity of XF498 is increased to that or even more than that of the most sensitive cell line SF539. Fig. 14 shows the same effect for the B subunit of verotoxin 1.

Anti-Proliferative Effects of Verotoxin on Human Astrocytoma Cells:

Figs. 10A - 10G show that all astrocytoma cell lines studied were sensitive to VTI. The most sensitive cell line in terms of growth inhibition was SF-539 (Fig. 10A) and the least sensitive was SF-188 (Fig. 10E). When treated with other members of the VT family including VT2, and VT2c, SF539 was growth inhibited. VT-1 was the most potent species (Fig. 11). Interestingly, human cerebral endothelial cells were largely resistant to the growth inhibitory and cytotoxic effects of VT-1 (Fig. 10G). Only when doses as high as 100 ng/ml were used were endothelial cells inhibited.

A comparison between the sensitivity of SF 539 and XF498 for VTI and VTI B subunit is shown in Fig. 11 A and Fig. 11 B. XF498 cells were considerably less sensitive to the B subunit than to the VT-1 holotoxin. By comparison, SF 539 astrocytoma cells were significantly more sensitive to the B subunit alone than were XF 498 astrocytoma cells, since 50%> cell death was observed in the presence of 50 ng/ml.

VT-Receptor Analysis of Human Astrocytoma Cells:

The glycolipid profile of the 6 human astrocytoma cell lines analyzed for Gb3 content as detected with orcinol is shown in Fig. 12 A. All of the astrocytoma cell lines

-33- expressed significant levels of Gb3 and showed binding with VTI in the overlay assay used (Fig. 12B). SF-539 cells expressed the highest levels of Gb3 with maximal binding to VTI.

Flow Cytometric Analysis:

To determine the extent of astrocytoma cells death by apoptosis, cells were analysed by flow cytometry. SF-539 and XF-498 astrocytoma cells exposed to VTI (10 ng/ml) revealed the characteristic features of apoptosis. As a result of chromatin condensation and DNA cleavage, apoptotic cells show less propidium iodide fluorescence than viable cells and can be quantified as the "subdiploid" population or pre-Gl position in cell cycle (Fig. 13, arrow head). Presence of cells with fractional DNA content, typical of apoptosis was more marked in SF-539 than XF-498 cells. A cell cycle analysis of the non- apoptotic cell population revealed marked differences in the proportion of cells in the respective phases of the cell cycle. In VTI- sensitive SF- 539 cells, a pronounced loss of S phase cells from 33 to 15 and 10%) was seen whereas with the less VTI sensitive XF-498 cells, the loss of S phase cells observed was only 75 to 69 and 65%. Changes in the proportion of cells in G2-M phase were also seen (Fig. 13).

Propidium iodide stains

For detection of apoptotic morphology in cells treated with VTI or VTI B- subunit, permeabilized SF— 539 and XF-498 cells were stained with the DNA- intercalating agent propidium iodide and were analyzed by fluorescence microscopy. VTI treated cells displayed characteristic features of apoptosis, such as marked reduction in diameter, condensed chromatin. Nuclear segmentation and subnuclear bodies were prominent in cells treated with, VTI B-subunit for 1.5 or 10 hours.

Ultrastructural Analysis of VT-Treated Astrocytoma Cells: By electron microscopy, VTl-treated astrocytoma cells (SF-539, XF-498 demonstrated characteristic features of apoptosis such as, blebbing of the cytoplasmic membrane, fragmentation of heterochromatin, condensation of the nucleolar membrane, loss of cell junctions and microvilli, nuclear disintegration, and apoptotic bodies. The results herein show that VTI inhibits the growth of a series of human astrocytoma cell lines. All cell lines showed significant sensitivity to VTI, contained the Gb3 receptor for VT, and demonstrated ultrastructural features indicative of apoptosis following VT treatment. These results show that VTs provide the basis of new agents active against human astrocytoma cells. The results show that the most toxin sensitive astrocytoma cell line, SF-539, is also highly sensitive to B subunit induced apoptosis.

Definitive morphological evidence of apoptosis (nuclear shrinkage and choromatine condensation) were observed within 1.5 hrs of toxin or B subunit administration to astrocytoma cells. This is considerably more rapid than has previously been described for induced apoptosis by anticancer drugs. Accumulation of VTI -treated astrocytoma cells in pre-Gl position in cell cycle (Fig. 13) is strong evidence for apoptosis. Additional evidence in support of VTI causing apoptosis in sensitive astrocytoma cells include nuclear staining with propidium iodide and ultrastructural alterations indicative of apoptosis.

We have found that there was relative insensitivity of human cerebral capillary endothelial cells to VT.

EXAMPLE 3: Case Studies of Subjects with PTLD Characterized by Infiltrating Lymphoma cells

The following case studies illustrate the effectiveness of verotoxins to bind specifically to EBV infected cells from PTLD infected patients. The in vitro studies show that verotoxin is an effective agent for binding specifically to cells which upregulate Gb₃.

Case 1. A previously healthy 9 year old boy presented acute fulminant hepatitis on November 1990. Although the etiology was believed to be viral, all serology screening tests were negative including HAV, BHV, CMV and EBV. The diagnosis given was non A non B hepatitis with histologic findings on liver biopsy of syncytial giant cell hepatitis. Despite full supportive conservative treatment, including a trial of prostaglandin E, the patients' condition deteriorated, from acute hepatic failure, with impaired gluconeogenesis and abnormal coagulation functions, urea cycle dysfunction, and encephalopathy. One month after his presentation the patient received a reduced size liver transplant. Induction immuno suppression included methylprednisolone, cyclosporine and azathioprine.

The patient was stable until day 10 post-transplant when he developed an acute, grade III rejection, which was treated with a ten day course of OKT3. Concomitantly, the patient developed Staphylococcus aureus sepsis. Antibiotic treatment included vancomycin initially, followed by a combination of cloxacillin, ampicillin and gentamicin. Ganciclovir treatment was initiated because of an increase in the Anti- EBVCA titer.

Despite an initial response to drug therapy, liver function and the patient's general condition continued to deteriorate. Apart from the acute rejection process, a liver biopsy revealed changes suggestive of cholangitis secondary to biliary obstruction. A second 14 day course of OKT3 was administrated. The patient's course was complicated by a colon laceration and hemothorax from a percutaneous needle liver biopsy, causing a serious bleeding event requiring a number of surgical interventions for hemostasis and hematoma evacuation. A laceration of the left femoral artery following a dialysis line insertion was also surgically repaired. The patient's course was complicated by acute renal failure presumed to have been caused by both acute tubular necrosis related to sepsis with hypotension and drug toxicity. The patient became oligo-anuric leading to the institution of continuous arterio-venous hemofiltration followed by hemodialysis.

Although liver function continued to deteriorate necessitating a second liver transplant, there was no indication of a malignant process in all imaging studies and histopathologic investigations. A second liver transplant was attempted three months after the first one, during which the patient died from uncontrollable bleeding. A subsequent histopathologic examination of the first graft revealed a malignant lymphoma infiltrating the bile duct and the peripheral tissues with patchy parenchymal involvement.

Case 2. This boy presented jaundice in early infancy secondary to congenital biliary artresia. Two attempts at palliation with Kasai procedures were unsuccessful resulting in progressive cholestasis and cirrhosis.

At nine months of age the patient had an acute deterioration from a gastro- intestinal bleeding and peritonitis requiring admission to the intensive care unit. Later, November 1991, the patient received a reduced size liver transplant. Initial immunosuppression consisted of an 8 day course of OKT3, methylprednisolone, and azathioprine. Cyclosporine was started on the sixth day post-transplant. Because of the donor's positive CMV status, the patient received prophylactic therapy with acyclovir and CMV hyperimmune globulin. Prior to transplant the patient's serology screening tests were positive for both CMV and EBV.

In the first two months following transplantation, the patient had eight liver biopsies most of which revealed evidence of acute cellular rejection, of varying severity. These rejection events were treated with two additional courses of OKT3 for periods of 11 and 14 days respectively.

On the third biopsy, a focal area suggestive of early post-transplant lymphoproliferative disease (PTLD) was identified. EBV and CMV were not detected by PCR on this sample and on the following five biopsies there was no evidence of an evolving lymphoproliferative process. Three months after transplant, while the patient's liver functions were stable and the rejection seemed to be under control, he had a prolonged upper respiratory tract infection causing upper airway obstruction and evidence of enlarged adenoids on a lateral cervical x-ray. An adenoidectomy was performed with the histopathologic examination of the adenoids revealing a B-cell lymphoma. Both CMV and EBV DNA were identified by PCR. Azathioprene treatment was stopped and treatment with gancyclovir and alpha-interferon was started for a period of six months. Prednisone and cyclosporine were continued according to the standard liver transplant protocol. No evidence of tumor spread was found on an extensive imaging work-up.

The patient responded to therapy and a follow-up with sequential imaging studies as well as additional liver biopsies showed no further evidence of PTLD. Liver function remained stable apart from a few episodes of mild rejection which responded well to an increased dose of prednisone.

Case 3. This girl was born with right sided reflux nephropathy and a left non- functioning dysplastic kidney which was surgically removed in infancy. In addition the child had multiple congenital malformations including: anal stenosis, spinal bifida, cicornuate uterus, Wolff-Parkinson- White syndrome, mild sensorineural hearing loss, fused left radius and ulna, hypoplastic femoral condyles and mild scoliosis; mental development was normal. In infancy she had a partial bowel resection due to volvulus and throughout childhood was treated with bronchodilators for moderate asthma.

The reflux nephropathy ultimately led to end stage renal disease. In November 1996, at the age of 16 years, the patient received a preemptive living related donor renal transplant.

Induction immunosuppression for her transplant consisted of prednisone, azathioprine and a 7-day course of OKT3. Cyclosporine was started on day 3 post- transplant. Prior to transplant both recipient and donor were CMV positive, but only the donor was EBV positive. The post-transplant course was complicated by one episode of rejection which occurred 3 months after transplantation, which was successfully treated with an increased dose of methylprednisolone that was tapered down over a period of two weeks, maintenance azathioprine was replaced by mycophenolate mofetil following this rejection episode. The patient had two episodes of pyelonephritis requiring intravenous antibiotic treatment, hypertension secondary to renal artery stenosis treated with balloon angioplasty and cholelithiasis induced cholecystitis which required cholecystectomy.

Sixteen months post-transplant she presented with an exacerbation of her asthma which was treated with inhaled bronchodilators and steroids. The patient then developed acute sinusitis complicated by bacterial tracheitis causing upper airway obstruction requiring incubation and mechanical ventilation. She had a septic shock event with hypotension causing acute renal failure. Bronchoalveolar lavage revealed heavy growth of staphylococcus aureus and a high load of EBV. The patient responded to treatment with a combination of cloxacillin gancyclovir and CMV hyperimmune globulin accompanied by discontinuation of MMF. The prednisone which was initially increased to 2 mg/kg/day for treatment of a possible asthmatic involvement in her respiratory deterioration, was scheduled for a slow gradual taper. Her kidney function recovered completely and a pituitary lesion which was seen on an MRI study and was unclear in origin was to be followed by further imaging studies.

The patient was discharged after more than 1 month hospitalization but unfortunately presented a week after her discharge with similar symptoms of sinusitis and trachiitis deteriorating to airway obstruction requiring ventilation. A biopsy from the nasal mucosa revealed infiltration of monoclonal B cells and was diffusely positive for EBV by in situ hybridization. At this point cyclosporine was stopped, prednisone continued to be gradually decreased to 10 mg/day and treatment with gancyclovir and CMV hyperimmune globulin was continued. Despite an initial response to treatment, the patient developed diffuse swelling of her parotid glands which were biopsied and revealed EBV related large cell lymphoma. Aggressive chemotherapy was ineffective and the patient died.

Materials and Methods FITC/VTI B Subunit staining of LPD liver section: Serial 5 μM cryosections of samples were thawed, dried, blocked with BSA and stained with FITC- VTI B in PBS (0.5 μg/ml) containing 0/1 % BSA for 1 hr at room temperature. Sections were extensively washed with PBS, mounted without media with antifading agent DABCO and observed under incident UV illumination.

Results

FITC- VTB staining of LPD livers from cases 1 and 2 (Figure 15 a and b, respectively), and the adenoid from case 3 (Figures 15 c and d) show the membrane staining of single cells dispersed throughout the tissue. Background staining for FITC VT B was seen for a normal liver section (case 1, Figure 15 e). Double labeling using anti-CD20 and FITC VTB showed coincident expression of the lymphoid antigen and VT receptor (Figure 15 f). Figure 16 shows a fixed liver section from case 1 which was processed for EBV nucleotide sequences using in situ hybridization techniques. The EBV positive cells are stained.

Discussion Malignancy, as a serious complication of organ transplantation, was recognized many years ago. In the late 1960s, it was reported that renal transplant recipients, compared to the general population, were one hundred times more likely to develop malignancies [Morrison, V.A. et al. Am. J. Med. (1994) 97:14-24]. Subsequently, it was stated that renal allograft recipients were forty times more likely than the general population to develop PTLD [Morrison et al., supra}. As a plethora of anti-rejection medication became available since the 1980's, including OKT3, cyclosporine and tacrolimus, [Swinnen, L.J. et al. New EngJMed (1990) 323:1723-1728; Newell, K.A. et al. Transplantation (1996) 62:370-375] there has been an increase in PTLD with lymphoproliferative disorders complicating the clinical course of 1 -20%> of organ transplant recipients [Randhawa, P.S. et al. New EngJMed (1992) 327:1710-17141; Montone, K.T. et al. Surgery (1996) 119:544-551].

The EBV, a known herpes virus, has been associated with several human cancers including nasopharyngeal carcinoma, BL, and a B cell lymphoma seen in immunodeficient hosts. All humans carry some EBV infected lymphocytes

[Moghaddam, A. et A. Science (1997) 276:2030-2033]. By giving immunosuppressive agents following organ transplantation, immunosurveillance is compromised and the EBV positive population is allowed to expand unchecked, increasing the likelihood of lymphoma development and PTLD. The main histologic types of lymphomas seen include polyclonal B cell proliferation and a malignant monoclonal lymphoma.

With the identification of a malignant lymphoma in a patient with PTLD, there is usually reduction or cessation of most or all, immunosuppression, and therapy with antiviral agents (acyclovir or gancyclovir), intravenous immunoglobulin therapy, and interferon-alpha therapy. At times, graft removal is necessary. More recently, therapy with anti-B cell antibodies (CD21 and CD24 monoclonal antibodies) has been suggested [Lazarovitis, A. I. et al. Clin Invest Med (1994) 17:621-625]. Despite all attempts at eradicating the malignancy, death from infection, tumor or loss of a necessary allograft is seen and can be as high as 80%> [Morrison, V.A., supra; Swinnen, L.J. et al. Blood (1995) 86:3333-3340]. In an attempt to improve the treatment of PTLD, verotoxin was administered.

Since BL cells are EBV induced and express Gb3 receptors, it was proposed that malignant monoclonal lymphoma seen following organ transplantation might also express upregulated Gb3 receptors since it is also an EBV induced lymphoma.

Approximately 20%> of tonsillar B cells express Gb3 receptors [Cohen, A. et al., Int. Immunol (1990) 2:1-8]; whereas T cells do not; i.e. less than 2%. The wild type

Daudi Burkitt B cell lymphoma cell line and EBV induced B lymphoblasts express Gb3 receptors and are highly sensitive to VT cytotoxicity in vitro [Cohen, A. et al. J. Biol. Chem. (1987) 262:17088-17099].

Frozen tissue of the three reported cases above of PTLD were analyzed. At present, only single core biopsies are usually taken and the tissue is rarely frozen, therefore, access to frozen tissue is not readily available.. All three of the cases showed cells that are EBV induced in that they had markers for EBV. Additionally these same cells showed an up-regulation of the Gb3 receptor. None of the other tissue surrounding these cells whether from liver, adenoids or parotid gland showed Gb3 on their surface.

EXAMPLE 4: Animal Model for PTLD

Animal systems have been previously used to study EBV transformed human B lymphocytes and the development PTLD in vivo. Immunodeficient mice, in particular strains with severe combined immunodeficiency (SCID) accept human lymphoid xenografts and allow their endogenous expansion over periods of several months. If these grafts are derived from EBV carriers, then eventually PTLD will develop as surveillant T cells are slowly lost from the graft. This provides an excellent model for the natural course of PTLD.

PTLD development requires several months in xenografted SCID mice and engraftment is successful in 705 of cases. To examine the behavior of these lymphomas in an accelerated fashion requires engraftment of already (in vitro) EBV transformed B lineage cells. This is a valid model for PTLD, as only the long phase for escape from EBV immunosurveillance is bypassed. It has been proposed to employ this accelerated model for the study of VTI in vivo effectiveness.

Two such EBV transformed B cell lines (MB and TH9) were generated and tested for expression of Gb3 and for sensitivity to VTI cytotoxicity in vitro. Figure 17 shows that both cell lines express significant levels of Gb3, which is recognized by VTI.

Significantly, the VT receptor species identified are, predominantly, the more slowly migrating Gb₃ isoforms which had been previously correlated in other cell lines with high sensitivity to VT in vitro and in vivo and with MDR expression. As can be seen from the cytotoxicity profile (Figure 18), both EBV positive B cell lines are highly susceptible to VTI (CD50~5pg/ml). Thus injections of either of these cell lines into SCID mice provides a good model for determining the efficacy of systemic VT administration.

EQUIVALENTS:

-41- While the invention has been described in detail and with reference to specific embodiments thereof, it will be apparent to one skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention.

Claims

CLAIMSWhat is claimed is:

1. A method for inducing lymphoma cell death, the method comprising administering an effective amount of an agent that binds Gb₃, such that lymphoma cell death occurs.

2. The method of claim 1 , wherein the lymphoma cell is a lymphoma cell of B cell origin.

3. The method of claim 2, wherein the lymphoma cell of B cell origin is EBV positive.

4. The method of claim 1 , wherein the agent is a verotoxin.

5. The method of claim 4, wherein the verotoxin is selected from the group consisting of verotoxin 1, verotoxin 1 B-subunit, verotoxin 2, and verotoxin 2c.

6. A method for treating a disorder characterized by infiltrating lymphoma cells, the method comprising administering to a subject in need thereof an effective amount of an agent capable of binding Gb₃ and inducing the death of the lymphoma cells such that the disorder is treated.

7. The method of claim 6, wherein the disorder is lymphoma.

8. The method of claim 7, wherein said lymphoma is a cutaneous T- cell lymphoma.

9. The method of claim 8, wherein said lymphoma is selected from the group consisting of Mycosis Fungoides, sezary syndrome, related cutaneous disease lymphomatoid papilosis, and PTLD.

10. The method of claim 9, wherein said lymphoma is PTLD.

11. The method of claim 10, wherein said PTLD is associated with renal, heart, lung or liver transplantation.

12. The method of claim 6, wherein the agent is a verotoxin.

13. The method of claim 12, wherein the verotoxin is selected from the group consisting of verotoxin 1, verotoxin 1 B-subunit, verotoxin 2, and verotoxin

2c.

14. The method of claim 6, wherein the agent is administered in a pharmaceutically acceptable carrier.

15. The method of claim 6, wherein the agent is selected from the group consisting of Pag adhesin linked to a toxin and an antibody to Gb₃ linked to a toxin.

16. The method of claim 15, wherein the toxin is ricin or a ricin derivative.

17. The method of claim 6, wherein the subject is human.

18. The method of claim 17, wherein said human is HIV positive.

19. A kit comprising an agent capable of binding Gbβ, such that lymphoma cell death occurs, a container, and instructions for administering said agent such that lymphoma cell death occurs.

20. The kit of claim 19 wherein said agent is a verotoxin.

21. A method of treating PTLD in a human comprising administering an effective amount of a verotoxin, such that PTLD is treated.

22. The method of claim 21 , wherein said human is a liver transplant recipient.

23. The method of claim 21, wherein said human is a renal transplant recipient.

24. The method of claim 21 , wherein said human is a heart transplant recipient.

25. The method of claim 21 , wherein said human is a lung transplant recipient.