Cancer Patent

Proaerolysin containing protease activation sequences and methods of use for treatment of prostate cancer

Cancer Patent Abstract



Disclosed herein are modified proaerolysin (PA) peptide. In some examples, the proteins include a prostate-specific protease cleavage site and can further include a prostate-tissue-specific binding domain which functionally replaces the native PA binding domain. In other examples, the proteins include a furin cleavage site and a prostate tissue-specific binding domain which functionally replaces the native PA binding domain. Methods of using such peptides to treat prostate cancer are also disclosed.

Cancer Patent Claims
We claim:

1. A purified peptide comprising a variant proaerolysin amino acid sequence, wherein the variation of the proaerolysin amino acid sequence consists of the replacement of the furin cleavage site corresponding to the sequence of amino acids at positions 427-432 of SEQ ID NO:2 by a prostate-specific protease cleavage site, wherein administration of 2.5-25 .mu.g of the peptide to a mouse having a prostate tumor xenograft reduces the volume of the tumor by 30% to 100%.

2. The peptide of claim 1, wherein the prostate-specific protease cleavage site comprises a prostate-specific antigen (PSA) cleavage site, a prostate specific membrane antigen (PSMA) cleavage site, or a human glandular kallikrein 2 (hK2) cleavage site.

3. The peptide of claim 2, wherein the PSA cleavage site comprises SEQ ID NO: 5, 8, 11, 14, 15, 16, 17, 18, 19, 20, or 21.

4. The peptide of claim 3, wherein the PSA cleavage site comprises SEQ ID NO: 5.

5. The purified peptide of claim 1, wherein the variant proaerolysin amino acid sequence is encoded by the nucleic acid sequence shown in SEQ ID NO: 3, 6, 9, or 12.

6. The purified peptide of claim 1, wherein the variant proaerolysin amino acid sequence comprises the amino acid sequence shown in SEQ ID NO: 4, 7, 10, or 13.

7. The purified peptide of claim 1, wherein the variant proaerolysin amino acid sequence consists of the amino acid sequence shown in SEQ ID NO: 4, 7, 10, or 13.

8. The purified peptide of claim 1, wherein the variant proaerolysin amino acid sequence consists of the amino acid sequence shown in SEQ ID NO: 4.

9. The peptide of claim 1, wherein the peptide is immobilized to a surface.

10. The peptide of claim 9, wherein the surface is a bead.

11. The peptide of claim 10, wherein the bead further comprises a prostate-specific ligand.

12. A purified peptide comprising a variant proaerolysin amino acid sequence, wherein the variations of the proaerolysin amino acid sequence consist of (i) the replacement of the furin cleavage site corresponding to the sequence of amino acids at positions 427-432 of SEQ ID NO:2 by a prostate-specific protease cleavage site and (ii) a modification of the amino acid sequence of the proaerolysin binding domain corresponding to the sequence of amino acids at positions 1-83 of SEQ ID NO: 2.

13. The peptide of claim 12, wherein the proaerolysin binding domain is modified by deletion of the amino acids at positions 1-83 of SEQ ID NO:2 or 4.

14. The peptide of claim 12, wherein the proaerolysin binding domain of SEQ ID NO:2 or 4 is modified by at least one amino acid replacement selected from the group consisting of W45A, 147E, M57A, Y61A, and K66Q.

15. The peptide of claim 12, wherein the proaerolysin binding domain is replaced with a prostate-tissue specific binding domain.

16. The peptide of claim 12, wherein a luteinizing hormone releasing hormone sequence is linked to an amino acid at position 215 or position 300 of SEQ ID NO:2 or 4, wherein the amino acid at position 215 or position 300 has been replaced by a cysteine.

17. The peptide of claim 15, wherein the prostate-tissue specific binding domain comprises a luteinizing hormone releasing hormone (LHRH) sequence.

18. The peptide of claim 17, wherein the luteinizing hormone releasing hormone sequence comprises SEQ ID NO: 22 or 23.

19. The peptide of claim 17, wherein the amino acid sequence of the peptide comprises SEQ ID NO:24 or 25.

20. The peptide of claim 13, wherein the peptide further comprises a luteinizing hormone releasing hormone sequence linked to the N-terminus of the variant proaerolysin.

21. The peptide of claim 15, wherein the prostate-tissue specific binding domain comprises an antibody that recognizes prostate-specific antigen, human glandular kallikrein 2, prostate-specific membrane antigen, or-luteinizing hormone releasing hormone.

22. The peptide of claim 21, wherein the antibody is linked to the N-terminus of the variant proaerolysin.

23. The peptide of claim 21, wherein the antibody is linked to the C-terminus of the variant proaerolysin.

24. A method for treating prostate cancer in a subject, comprising administration of the peptide of claim 1 to the subject.

25. The method of claim 24, wherein the peptide is administered intratumorally and/or intraprostatically.

26. The method of claim 24, wherein the peptide is administered intravenously, intramuscularly, subcutaneously, or orally.

27. The method of claim 24, wherein the subject has a localized prostate tumor.

28. The method of claim 24, wherein the subject has a metastatic prostate tumor.

29. The method of claim 24, further comprising administering GM-CSF to the subject.

30. The method of claim 24, further comprising administering irradiated prostate cancer cells to the subject.

31. The method of claim 24, wherein the prostate tumor cell volume is reduced by at least 50%.

32. The method of claim 24, wherein administration further results in a reduction of a metastatic prostate tumor.

33. The method of claim 24, wherein administration further results in treatment of a metastatic prostate tumor.

34. A method for treating prostate cancer in a subject, comprising contacting prostate cancer cells of the subject with the peptide of claim 1.

35. A method for treating prostate cancer in a subject, comprising administering the peptide of claim 12 to the subject.

36. A method for treating prostate cancer in a subject, comprising contacting prostate cancer cells of the subject with the peptide of claim 12.

Cancer Patent Description
FIELD

This application relates to novel variant proaerolysin (PA) proteins and methods of their use to treat localized and metastatic prostate cancer.

BACKGROUND

One of every three cancers diagnosed in American males is of prostatic origin, making prostate cancer the most commonly diagnosed malignancy in males in the United States (Berges et al. Clin. Cancer Res. 1:473-480, 1995). The incidence of prostate cancer in the U.S. has not been decreased by changes in lifestyle; in fact, the incidence rate of clinical prostate cancer has increased steadily since the 1930's (Pinski et al. Cancer Res. 61:6372-6, 2001). Prostate cancer incidence increases with age more rapidly than any other type of cancer; less than 1% of prostate cancers are diagnosed in men less than 50 years of age (Furuya et al. Cancer Res. 54:6167-75, 1994). Thus, as the life expectancy of the male population increases over time, the incidence of clinical prostate cancer will also increase (Furuya et al. Cancer Res. 54:6167-75, 1994).

Currently there is no treatment that significantly prolongs survival in men with metastatic prostate cancer (Khan and Denmeade. Prostate 45:80-83, 2000). Medical castration with oral estrogen (androgen ablation) was the first effective systemic therapy for cancer, and remains the most generally useful prostate cancer therapy. Although androgen ablation therapy has a substantial palliative benefit, it has little impact on overall survival (Berges et al. Clin. Cancer Res. 1:473-480, 1995). This therapy eventually fails because the metastatic prostate cancer within an individual patient is heterogeneously composed of androgen-dependent and androgen-independent cancer cells (Christensen et al. Bioorg. Medicinal Chem. 7:1273-80, 1999). Following androgen ablation, androgen-dependent cells within these tumors stop proliferating and activate a cellular suicide pathway termed programmed cell death (PCD) or apoptosis. Because of the elimination of this subgroup of androgen-dependent cells, the majority of men with metastatic prostate cancers have a beneficial response to androgen-deprivation therapy. However, all patients eventually relapse to a state unresponsive to further anti-androgen therapy, no matter how completely given, due to the presence of androgen-independent prostate cancer cells within the metastatic sites. Unfortunately, the disease is uniformly fatal at this point because currently there is no therapy that effectively eliminates androgen-independent prostate cancer cells (Khan and Denmeade. Prostate 45:80-83, 2000).

Several alternative approaches to the treatment of prostate cancer have been proposed. One has been to develop methods to aggressively screen for local disease while it is still in the prostate and thus potentially treatable by definitive local therapy. Localized cancers are often moderately differentiated and smaller in volume. During the last several decades, there have been improvements to the surgical and radiotherapeutic management of localized prostate cancer. These improvements have culminated over the last several years in the death rate of prostate cancer decreasing for the first time in fifty years.

However, while this advance increased the cure rate, there are still a large number of men who are not cured by local therapies and eventually die from metastatic disease. This clinical reality has led to the development of non-hormonal treatments for metastatic prostate cancer. Standard anti-proliferative chemotherapeutic agents have not been successful as treatment for prostate cancer. These types of agents may be ineffective against androgen-independent prostatic cancers because these cancers have a remarkably low rate of proliferation when compared to other tumor types and many normal tissues such as skin, gastrointestinal tract and bone marrow. For example, the growth fraction in 117 metastatic sites of prostate cancer obtained from 11 androgen ablation failing patients at "warm" autopsy was 7.1.+-.0.8%. (Pinski et al. Cancer Res. 61:6372-6, 2001). This low proliferative rate may explain the relative unresponsiveness of prostate cancer cells in humans to standard anti-proliferative chemotherapy, while highly proliferative androgen independent prostate cancer cell lines remain exquisitely sensitive to PCD induction in vitro.

Several strategies have been proposed for treatment of slowly-proliferating prostate cancers. One approach is to identify specific signaling pathways to which prostate cancer cells, during malignant transformation, acquire a unique dependence for survival. Once identified, small molecule or biological inhibitors of these pathways can be developed as therapeutics. An example of this approach is the use of small molecule or monoclonal antibody inhibitors of the Her2/neu or EGF receptor pathways. Another method is to inhibit a ubiquitous intracellular protein whose function is mandatory for survival of all cell types. This approach would overcome the problem of heterogeneity and "resistance" as all cancer cells within a tumor could be killed via this approach. However, the cytotoxicity would not be cell-type specific and administration of such a general toxin would be associated with significant systemic toxicity. Therefore, there is a need for a method for targeting cytotoxins directly to sites of prostate cancer.

Another strategy for treatment of slowly proliferating prostate cancers is to deliver. cytotoxins that kill cells not through induction of apoptosis following inhibition of critical signaling or metabolic pathways but rather through non-specific cytolysis via disruption of the plasma membrane. Many cytolytic toxins have been described (Lesieur et al. Mol. Membr. Biol. 14:45064, 1997). These cytolytic toxins are often of bacterial origin, and, in general, are beta-sheet proteins that oligomerize in the plasma membrane to produce well-characterized pores that, once formed, lead to rapid cytolytic cell death (Rossjohn et al. J. Struct. Biol. 121:92-100,1998). These toxins are also non-specific in their ability to kill cells, and therefore can not be administered as therapy without significant toxicity. Therefore, there is a need for agents to treat prostate cancer, which are predominantly cytotoxic to prostate cancer cells.

SUMMARY

Disclosed herein are variant proaerolysin (PA) molecules and methods of their use for treatment of localized and metastatic prostate cancers, such as slowly-proliferating prostate cancers.

In one example, a variant PA molecule includes a prostate-specific protease cleavage site, such as a prostate-specific antigen (PSA)-, prostate specific membrane antigen (PSMA)-, or human glandular kallikrein 2 (HK2)-specific cleavage site that functionally replaces the native PA furin cleavage site. In this way, administration of the disclosed variant PA molecules to a subject having prostate cancer results in activation of the disclosed variant PA molecules in the presence of the prostate-specific protease, and lysis of the cells, such as prostate cancer cells. In some examples, variant PA molecules also include a prostate tissue specific binding domain, to enhance targeting to cancer cells.

In another example, a variant PA molecule includes a furin cleavage site, and a prostate tissue specific binding domain which functionally replaces the native PA binding domain, to assist in targeting to prostate cancer cells.

Methods are disclosed for treatment of localized and metastatic prostate cancers using the disclosed variant PA molecules. In addition, methods are disclosed for stimulating a subject's immune system to enhance treatment of localized and metastatic prostate cancer.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a schematic of proaerolysin domains (not drawn to scale) and shows the result of activation by furin.

FIG. 2 is a bar graph showing the results of a hemolysis assay in which PSA-PA1 is preincubated with human plasma or human plasma spiked with enzymatically active PSA (10,000 ng/ml).

FIG. 3 is a graph comparing the in vitro toxicity of several proaerolysin variants which include a PSA cleavage site in place of the native furin site, to wild-type proaerolysin.

FIG. 4 is a bar graph comparing the specificity of PSA-PA1 in PSA-producing (LNCaP), and non-PSA producing (SN12C) tumors, in vivo.

FIGS. 5A-5M are schematic drawings (not to scale) showing how a proaerolysin sequence can be altered to generate several different variant PA molecules. The "*" symbol represents one or more point mutations, and/or one or more deletions which decrease PA binding domain function (i.e. the ability to concentrate in a cell membrane).

SEQUENCE LISTING

The nucleic and amino acid sequences listed in the accompanying sequence listing are shown using standard letter abbreviations for nucleotide bases, and three letter code for amino acids. Only one strand of each nucleic acid sequence is shown, but the complementary strand is understood as included by any reference to the displayed strand.

SEQ ID NOS: 1 and 2 show a wild-type proaerolysin cDNA and protein sequence, respectively.

SEQ ID NOS: 3 and 4 show the PSA-PA1 cDNA and protein sequence, respectively, wherein the furin site of proaerolysin has been replaced with a PSA cleavage site.

SEQ ID NOS: 5 and 15-21 are PSA cleavage sites found in human semenogelin I and II proteins.

SEQ ID NOS: 6 and 7 show the PSA-1K cDNA and protein sequence, respectively, wherein the furin site of proaerolysin has been replaced with a PSA cleavage site.

SEQ ID NO: 8, 11, and 14-21 are alternative PSA cleavage sites.

SEQ ID NOS: 9 and 10 show the PSA-PA2 cDNA and protein sequence, respectively, wherein the furin site of proaerolysin has been replaced with a PSA cleavage site.

SEQ ID NOS: 12 and 13 show the PSA-PA3 cDNA and protein sequence, respectively, wherein the furin site of proaerolysin has been replaced with a PSA cleavage site.

SEQ ID NO: 22 is a native luteinizing hormone releasing hormone (LHRH) protein sequence.

SEQ ID NO: 23 is a modified LHRH protein sequence.

SEQ ID NO: 24 is a protein sequence of a variant PA peptide, wherein the furin site of PA has been replaced with a PSA cleavage site, and wherein the native binding domain of PA is deleted and replaced with SEQ ID NO: 23.

SEQ ID NO: 25 is a protein sequence of a variant PA peptide, wherein the furin site of PA is retained, and the native binding domain of PA has been deleted and replaced with SEQ ID NO: 23.

DETAILED DESCRIPTION OF SEVERAL EMBODIMENTS

Abbreviations and Terms

The following explanations of terms and methods are provided to better describe the present disclosure and to guide those of ordinary skill in the art in the practice of the present disclosure. As used herein and in the appended claims, the singular forms "a" or "an" or "the" include plural references unless the context clearly dictates otherwise. For example, reference to "a variant PA molecule" includes a plurality of such molecules and reference to "the antibody" includes reference to one or more antibodies and equivalents thereof known to those skilled in the art, and so forth.

Unless explained otherwise, all technical and scientific terms used herein have the same meaning as commonly understood to one of ordinary skill in the art to which this disclosure belongs.

Aerolysin: A channel-forming toxin produced as an inactive protoxin called proaerolysin (PA) (wild-type PA is shown in SEQ ID NOS: 1 and 2). The PA protein contains many discrete functionalities that include a binding domain (approximately amino acids 1-83 of SEQ ID NO: 2), a toxin domain (approximately amino acids 84-426 of SEQ ID NO: 2), and a C-terminal inhibitory peptide domain (approximately amino acids 427-470 of SEQ ID NO: 2) that contains a protease activation site (amino acids 427-432 of SEQ ID NO: 2).

The binding domain recognizes and binds to glycophosphatidylinositol (GPI) membrane anchors, such as are found in Thy-1 on T lymphocytes, the PIGA gene product found in erythrocyte membranes and Prostate Stem Cell Antigen (PSCA). Most mammalian cells express GPI anchored proteins on their surfaces. The activation or proteolysis site within proaerolysin is a six amino acid sequence that is recognized as a proteolytic substrate by the furin family of proteases. PA is activated upon hydrolysis of a C-terminal inhibitory segment by furin (FIG. 1). Activated aerolysin binds to GPI-anchored proteins in the cell membrane and forms a heptamer that inserts into the membrane producing well-defined channels of .about.17 .ANG.. Channel formation leads to rapid cell death via necrosis. Wild-type aerolysin is toxic to mammalian cells, including erythrocytes, for example at 1 nanomolar or less.

Animal: Living multicellular vertebrate organisms, a category which includes, for example, mammals and birds.

Antibody: Immunoglobulin molecules and immunologically active portions of immunoglobulin molecules, i.e., molecules that contain an antigen binding site which specifically binds (immunoreacts with) an antigen.

A naturally occurring antibody (e.g., IgG) includes four polypeptide chains, two heavy (H) chains and two light (L) chains inter-connected by disulfide bonds. However, the antigen-binding function of an antibody can be performed by fragments of a naturally occurring antibody. Thus, these antigen-binding fragments are also intended to be designated by the term antibody. Examples of binding fragments encompassed within the term antibody include (i) an Fab fragment consisting of the VL, VH, CL and CH1 domains; (ii) an Fd fragment consisting of the VH and CH1 domains; (iii) an Fv fragment consisting of the VL and VH domains of a single arm of an antibody, (iv) a dAb fragment (Ward et al., Nature 341:544-6, 1989) which consists of a VH domain; (v) an isolated complimentarily determining region (CDR); and (vi) an F(ab')2 fragment, a bivalent fragment comprising two Fab fragments linked by a disulfide bridge at the hinge region. Furthermore, although the two domains of the Fv fragment are coded for by separate genes, a synthetic linker can be made that enables them to be made as a single protein chain (known as single chain Fv (scFv); Bird et al. Science 242:423-6, 1988; and Huston et al., Proc. Natl. Acad. Sci. 85:5879-83, 1988) by recombinant methods. Such single chain antibodies are also included. In one embodiment, an antibody includes camelized antibodies (for example see Tanha et al., J. Biol. Chem. 276:24774-80, 2001).

In one example, antibody fragments are capable of crosslinking their target antigen, e.g., bivalent fragments such as F(ab')2 fragments. Alternatively, an antibody fragment which does not itself crosslink its target antigen (e.g., a Fab fragment) can be used in conjunction with a secondary antibody which serves to crosslink the antibody fragment, thereby crosslinking the target antigen. Antibodies can be fragmented using conventional techniques and the fragments screened for utility in the same manner as described for whole antibodies. An antibody is further intended to include bispecific and chimeric molecules that specifically bind the target antigen.

"Specifically binds" refers to the ability of individual antibodies to specifically immunoreact with an antigen. The binding is a non-random binding reaction between an antibody molecule and an antigenic determinant of the T cell surface molecule. The desired binding specificity is typically determined from the reference point of the ability of the antibody to differentially bind the T cell surface molecule and an unrelated antigen, and therefore distinguish between two different antigens, particularly where the two antigens have unique epitopes. An antibody that specifically binds to a particular epitope is referred to as a "specific antibody".

Cancer: Malignant neoplasm that has undergone characteristic anaplasia with loss of differentiation, increase rate of growth, invasion of surrounding tissue, and is capable of metastasis.

cDNA (complementary DNA): A piece of DNA lacking internal, non-coding segments (introns) and regulatory sequences which determine transcription. cDNA can be synthesized in the laboratory by reverse transcription from messenger RNA extracted from cells.

Chemical synthesis: An artificial means by which one can make a protein or peptide. A synthetic protein or peptide is one made by such artificial means.

Chemotherapy: In cancer treatment, chemotherapy refers to the administration of one or a combination of compounds to kill or slow the reproduction of rapidly multiplying cells. Chemotherapuetic agents include those known by those skilled in the art, including, but not limited to: 5-fluorouracil (5-FU), azathioprine, cyclophosphamide, antimetabolites (such as Fludarabine), antineoplastics (such as Etoposide, Doxorubicin, methotrexate, and Vincristine), carboplatin, cis-platinum and the taxanes, such as taxol and taxotere. Such agents can be co-administered with the disclosed variant PA molecules to a subject. Alternatively or in addition, chemotherapeutic agents can be administered prior to and/or subsequent to administration of the disclosed variant PA molecules to a subject. In one example, chemotherapeutic agents are co-administered with hormonal and radiation therapy, along with the disclosed variant PA molecules, for treatment of a localized prostate carcinoma.

Conservative substitution: One or more amino acid substitutions (for example 2, 5 or 10 residues) for amino acid residues having similar biochemical properties. Typically, conservative substitutions have little to no impact on the activity of a resulting polypeptide. For example, ideally, a modified PA peptide including one or more conservative substitutions retains proaerolysin activity. A polypeptide can be produced to contain one or more conservative substitutions by manipulating the nucleotide sequence that encodes that polypeptide using, for example, standard procedures such as site-directed mutagenesis or PCR.

Substitutional variants are those in which at least one residue in the amino acid sequence has been removed and a different residue inserted in its place. Examples of amino acids which may be substituted for an original amino acid in a protein and which are regarded as conservative substitutions include: Ser for Ala; Lys for Arg; Gln or His for Asn; Glu for Asp; Ser for Cys; Asn for Gln; Asp for Glu; Pro for Gly; Asn or Gln for His; Leu or Val for Ile; Ile or Val for Leu; Arg or Gln for Lys; Leu or Ile for Met; Met, Leu or Tyr for Phe; Thr for Ser; Ser for Thr; Tyr for Trp; Trp or Phe for Tyr; and Ile or Leu for Val.

Permissive substitutions are non-conservative amino acid substitutions, but also do not significantly alter proaerolysin activity. An example is substitution of Cys for Ala at position 300 of SEQ ID NO: 2 or 4.

Further information about conservative substitutions can be found in, among other locations in, Ben-Bassat et al., (J. Bacteriol. 169:751-7, 1987), O'Regan et al., (Gene 77:237-51, 1989), Sahin-Toth et al., (Protein Sci. 3:240-7, 1994), Hochuli et al., (Bio/Technology 6:1321-5, 1988), WO 00/67796 (Curd et al.) and in standard textbooks of genetics and molecular biology.

In one example, such variants can be readily selected for additional testing by performing an assay (such as those described in Examples 2-5) to determine if the variant retains variant PA activity.

Comprises: A term that means "including." For example, "comprising A or B" means including A or B, or both A and B, unless clearly indicated otherwise.

Deletion: The removal of a sequence of a nucleic acid, for example DNA, the regions on either side being joined together.

DNA: Deoxyribonucleic acid. DNA is a long chain polymer which comprises the genetic material of most living organisms (some viruses have genes comprising ribonucleic acid, RNA). The repeating units in DNA polymers are four different nucleotides, each of which comprises one of the four bases, adenine, guanine, cytosine and thymine bound to a deoxynbose sugar to which a phosphate group is attached. Triplets of nucleotides, referred to as codons, in DNA molecules code for amino acid in a polypeptide. The term codon is also used for the corresponding (and complementary) sequences of three nucleotides in the mRNA into which the DNA sequence is transcribed.

Enhance: To improve the quality, amount, or strength of something. In one embodiment, a therapy enhances the ability of a subject to reduce tumors, such as a prostate carcinoma, in the subject if the subject is more effective at fighting tumors. In another embodiment, a therapy enhances the ability of an agent to reduce tumors, such as a prostate carcinoma, in a subject if the agent is more effective at reducing tumors. Such enhancement can be measured using the methods disclosed herein, for example determining the decrease in tumor volume (see Example 5).

Functional Deletion: A mutation, partial or complete deletion, insertion, or other variation made to a gene sequence which renders that part of the gene sequence non-functional.

For example, functional deletion of a PA binding domain results in a decrease in the ability of PA to bind to and concentrate in the cell membrane. This functional deletion can be reversed by inserting another functional binding domain into proaerolysin, such as a prostate-specific binding domain, for example, an LHRH peptide.

Examples of methods that can be used to functionally delete a proaerolysin binding domain, include, but are not limited to: deletion of about amino acids 1-83 of SEQ ID NO: 2 or fragments thereof, such as about amino acids 45-66 of SEQ ID NO: 2, or inserting one or more of the following mutations into a variant proaerolysin sequence W45A, I147E, M57A,Y61A, K66Q (amino acid numbers refer to SEQ ID NO: 2) (for example, see Mackenzie et al. J. Biol. Chem. 274: 22604-22609, 1999).

In another example, functional deletion of a native PA furin cleavage site results in a decrease in the ability of PA to be cleaved and activated by furin, when compared to a wild-type PA molecule.

Immobilized: Bound to a surface, such as a solid surface. A solid surface can be polymeric, such as polystyrene or polypropylene. In one embodiment, the solid surface is in the form of a bead. In another embodiment, the surface includes a modified PA toxin, and in some examples further includes one or more prostate-specific binding ligands, such as LHRH peptide, PSMA antibody, and PSMA single chain antibody. Ideally, the modified PA toxin is liberated from the bead once the bead reaches the prostate cell target. Methods of immobilizing peptides on a solid surface can be found in WO 94/29436, and U.S. Pat. No. 5,858,358.

Examples of how the molecules can be attached to the bead include, but are not limited to: PA toxin-PSA cleavage site-bead-prostate binding ligand; or prostate binding ligand-bead-PSA site-PA toxin-PSA cleavage site-inhibitor.

Isolated: An "isolated" biological component (such as a nucleic acid molecule or protein) has been substantially separated or purified away from other biological components in the cell of the organism in which the component naturally occurs (i.e. other chromosomal and extrachromosomal DNA and RNA). Nucleic acids and proteins that have been "isolated" include nucleic acids and proteins purified by standard purification methods. The term also embraces nucleic acids and proteins prepared by recombinant expression in a host cell as well as chemically synthesized nucleic acids and proteins.

An isolated cell is one which has been substantially separated or purified away from other biological components of the organism in which the cell naturally occurs.

Malignant: Cells which have the properties of anaplasia invasion and metastasis.

Mammal: This term includes both human and non-human mammals. Similarly, the term "subject" includes both human and veterinary subjects. Examples of mammals include, but are not limited to: humans, pigs, cows, goats, cats, dogs, rabbits and mice.

Neoplasm: Abnormal growth of cells.

Normal Cell: Non-tumor cell, non-malignant, uninfected cell.

Oligonucleotide: A linear polynucleotide sequence of up to about 200 nucleotide bases in length, for example a polynucleotide (such as DNA or RNA) which is at least about 6 nucleotides, for example at least 15, 50, 100 or 200 nucleotides long.

Operably linked: A first nucleic acid sequence is operably linked with a second nucleic acid sequence when the first nucleic acid sequence is placed in a functional relationship with the second nucleic acid sequence. For instance, a promoter is operably linked to a coding sequence if the promoter affects the transcription or expression of the coding sequence. Generally, operably linked DNA sequences are contiguous and, where necessary to join two protein coding regions, in the same reading frame.

ORF (open reading frame): A series of nucleotide triplets (codons) coding for amino acids without any termination codons. These sequences are usually translatable into a peptide.

Polynucleotide: A linear nucleic acid sequence of any length. Therefore, a polynucleotide includes molecules which are at least 15, 50, 100,200, 400, 500, 1000, 1100, or 1200 (oligonucleotides) and also nucleotides as long as a full-length cDNA or chromosome.

Proaerolysin: The inactive protoxin of aerolysin. The cDNA and protein of a wild-type or native proaerolysin are shown in SEQ ID NOS: 1 and 2, respectively.

In one example, a variant or modified proaerolysin molecule includes a prostate-specific protease cleavage site, such as a PSA-specific cleavage site, which permits activation of the variant PA in the presence of a prostate-specific protease such as PSA, PMSA, or HK2 (for example, see FIGS. 5C-I). In one example, a prostate-specific protease cleavage site is inserted into the native furin cleavage site of PA, such that PA is activated in the presence of a prostate-specific protease, but not furin (for example see FIGS. 5D-I). Alternatively, the furin cleavage site can be functionally deleted using mutagenesis of the six amino acid sequence, and insertion of a prostate-specific protease cleavage sequence (for example, see FIG. 5C). In another example, a variant PA molecule further includes deletion or substitution of one or more, such as at least two, of the native PA amino acids. In yet another example a variant PA molecule further includes another molecule (such as an antibody or peptide) linked or added to (or within) the variant PA molecule. In another example, a variant PA molecule includes a prostate-tissue specific binding domain.

In another example, a variant PA molecule further includes a functionally deleted binding domain (about amino acids 1-83 of SEQ ID NO: 2). Functional deletions can be made using any method known in the art, such as deletions, insertions, mutations, or substitutions. Examples include, but are not limited to deleting the entire binding domain (or portions thereof, (for example, see FIGS. 5G-I), or introduction of point mutations (such as those described above, (for example, see FIGS. 5D-F), which result in a binding domain with decreased function. For example, a PA molecule which has a functionally deleted binding domain (and no binding sequence substituted therefor), will have a decreased ability to accumulate in a cell membrane, and therefore lyse cells at a slower rate than a wild-type PA sequence (for example, see FIG. 5D). Also disclosed are variant PA molecules in which the native binding domain is functionally deleted and replaced with a prostate-tissue specific binding domain as described below (for example, see FIGS. 5E-I).

In another example, a variant or modified PA molecule includes a furin cleavage site, and a functionally deleted binding domain which is replaced with a prostate-tissue specific binding domain (for example, see FIGS. 5J-M). Such variant PA molecules are targeted to prostate cells via the prostate-tissue specific binding domain, and activated in the presence of furin.

Particular non-limiting examples of variant PSA proteins are shown in SEQ ID NOS: 4, 7, 10, 13, 24, and 25.

Modified PA activity is the activity of an agent in which the lysis of cells is affected. Cells include, but are not limited to prostate-specific protease secreting cells, such as PSA-secreting cells, such as prostate cancer cells, such as slow-proliferating prostate cancer cells. Agents include, but are not limited to, modified PA proteins, nucleic acids, specific binding agents, including variants, mutants, polymorphisms, fusions, and fragments thereof, disclosed herein. In one example, modified PA activity is said to be enhanced when modified PA proteins or nucleic acids, when contacted with a PSA-secreting cell (such as a prostate cancer cell), promote lysis and death of the cell, for example by at least 10%, or for example by at least 25%, 50%, 100%, 200% or even 500%, when compared to lysis of a non-PSA producing cell. In other examples, modified PA activity is said to be enhanced when modified PA proteins and nucleic acids, when contacted with a tumor, decrease tumor cell volume, such as a prostate tumor, for example by at least 10% for example by at least 20%, 30%, 40%, 50%, 60%, 70%, 75%, 80%, 85%, 90%, 95%, 98%, or even 100% (complete elimination of the tumor).

Assays which can be used to determine if an agent has modified PA activity are described herein, for example in Examples 2-5 and 9. For example, a modified PA peptide can be assessed for its ability to specifically lyse PSA-producing cells (Examples 2 and 5), be stable in human plasma (Example 3), be an efficient substrate for the enzymatic activity of PSA (Example 9). Functional protein activity could be detected by the preferential lysis of PSA-producing cells versus non-PSA-producing cells, decreasing prostate tumor volume, having a decreased toxicity when compared to wild-type PA, and having an increased stability in blood when compared to wild-type PA.

Similar assays can be used to determine if any modified PA agent disclosed herein can decrease tumor volume (such as a prostate tumor) and specifically lyse PSA-producing cells. For example, the modified PA peptides shown in SEQ ID NOS: 4, 24, and 25, are expected to decrease prostate tumor volume. Any of these assays can be modified by using in vivo expression of a modified PA gene, and variants, fusions, and fragments thereof, instead of administration of purified proteins.

Promoter: An array of nucleic acid control sequences which direct transcription of a nucleic acid. A promoter includes necessary nucleic acid sequences near the start site of transcription, such as, in the case of a polymerase II type promoter, a TATA element. A promoter also optionally includes distal enhancer or repressor elements which can be located as much as several thousand base pairs from the start site of transcription.

Prostate-specific promoter: A promoter responsive to testosterone and other androgens, which therefore promotes gene expression in prostate cells. Examples include, but are not limited to the probasin promoter; the prostate specific antigen (PSA) promoter; the prostate specific membrane antigen (PSMA) promoter; and the human glandular kallikrein 2 (HK2) promoter.

Prostate-specific protease cleavage site: A sequence of amino acids which is recognized and specifically and efficiently hydrolyzed (cleaved) by a prostate-specific protease. Examples include, but are not limited to a PSA-specific cleavage site, a PSMA-specific cleavage site and an HK2-specific cleavage site.

A PSA-specific cleavage site is a sequence of amino acids which is recognized and specifically and efficiently hydrolyzed (cleaved) by prostate specific antigen (PSA). Such peptide sequences can be introduced into other molecules, such as PA, to produce prodrugs that are activated by PSA. Upon activation of the modified PA by PSA, PA is activated and can exert its cytotoxicity. Examples of PSA-specific cleavage sites, include, but are not limited to are those shown in SEQ ID NOS: 5, 8, 11, and 14-21, those disclosed in U.S. Pat. Nos. 5,866,679 to DeFeo-Jones et al., U.S. Pat. No. 20 5,948,750 to Garsky et al., U.S. Pat. No. 5,998,362 to Feng et al., 6,265,540 to Isaacs et al., U.S. Pat. No. 6,368,598 to D'Amico et al., and U.S. Pat. No. 6,391,305 to Feng et al., (all herein incorporated by reference).

Particular examples of PSMA-specific cleavage sites can be found in WO/0243773 to Isaacs and Denmeade (herein incorporated by reference). Particular examples of HK2-specific cleavage sites are disclosed in WO/0109165 to Denmeade et al. (herein incorporated by reference).

Prostate tissue-specific binding domain: A molecule, such as a peptide ligand, toxin, or antibody, which has a higher specificity for prostate cells than for other cell types. In one example, a prostate tissue specific binding domain has a lower K.sub.D in prostate tissue or cells than in other cell types, (i.e. binds selectively to prostate tissues as compared to other normal tissues of the subject), for example at least a 10-fold lower K.sub.D, such as an at least 20-, 50-, 75-, 100- or even 200-fold lower K.sub.D. Such sequences can be used to target an agent, such as a variant PA molecule, to the prostate. Examples include, but are not limited to: antibodies which recognize proteins that are relatively prostate-specific such as PSA, PSMA, hK2, prostasin, and hepsin; ligands which have prostate-selective receptors such as natural and synthetic luteinizing hormone releasing hormone (LHRH); and endothelin (binding to cognate endothelin receptor).

Purified: The term "purified" does not require absolute purity; rather, it is intended as a relative term. Thus, for example, a substantially purified protein or nucleic acid preparation (such as the modified PA toxins disclosed herein) is one in which the protein or nucleic acid referred to is more pure than the protein in its natural environment within a cell or within a production reaction chamber (as appropriate). For example, a preparation of a modified PA protein is purified if the protein represents at least 50%, for example at least 70%, of the total protein content of the preparation. Methods for purification of proteins and nucleic acids are well known in the art. Examples of methods that can be used to purify a protein, such as a modified PA, include, but are not limited to the methods disclosed in Sambrook et al. (Molecular Cloning: A Laboratory Manual, Cold Spring Harbor, N.Y., 1989, Ch. 17).

Recombinant: A recombinant nucleic acid is one that has a sequence that is not naturally occurring or has a sequence that is made by an artificial combination of two otherwise separated segments of sequence. This artificial combination is often accomplished by chemical synthesis or, more commonly, by the artificial manipulation of isolated segments of nucleic acids, e.g., by genetic engineering techniques. A recombinant protein is one that results from expressing a recombinant nucleic acid encoding the protein.

Sample: Biological samples containing genomic DNA, cDNA, RNA, or protein obtained from the cells of a subject, such as those present in peripheral blood, urine, saliva, semen, tissue biopsy, surgical specimen, fine needle aspriates, amniocentesis samples and autopsy material. In one example, a sample includes prostate cancer cells obtained from a subject.

Sequence identity/similarity: The identity/similarity between two or more nucleic acid sequences, or two or more amino acid sequences, is expressed in terms of the identity or similarity between the sequences. Sequence identity can be measured in terms of percentage identity; the higher the percentage, the more identical the sequences are. Sequence similarity can be measured in terms of percentage similarity (which takes into account conservative amino acid substitutions); the higher the percentage, the more similar the sequences are.

Methods of alignment of sequences for comparison are well known in the art. Various programs and alignment algorithms are described in: Smith & Waterman, Adv. Appl. Math. 2:482, 1981; Needleman & Wunsch, J. Mol. Biol. 48:443, 1970; Pearson & Lipman, Proc. Natl. Acad. Sci. USA 85:2444, 1988; Higgins & Sharp, Gene, 73:23744, 1988; Higgins & Sharp, CABIOS 5:151-3, 1989; Corpet etal., Nuc. Acids Res. 16:10881-90, 1988; Huang etal. Computer Appls. in the Biosciences 8, 155-65, 1992; and Pearson et al., Meth. Mol. Bio. 24:307-31, 1994. Altschul et al., J. Mol. Biol. 215:403-10, 1990, presents a detailed consideration of sequence alignment methods and homology calculations.

The NCBI Basic Local Alignment Search Tool (BLAST) (Altschul et al., J. Mol. Biol. 215:403-10, 1990) is available from several sources, including the National Center for Biological Information (NCBI, National Library of Medicine, Building 38A, Room 8N805, Bethesda, Md. 20894) and on the Internet, for use in connection with the sequence analysis programs blastp, blastn, blastx, tblastn and tblastx. Additional information can be found at the NCBI web site.

For comparisons of amino acid sequences of greater than about 30 amino acids, the Blast 2 sequences function is employed using the default BLOSUM62 matrix set to default parameters, (gap existence cost of 11, and a per residue gap cost of 1). When aligning short peptides (fewer than around amino acids), the alignment should be performed using the Blast 2 sequences function, employing the PAM30 matrix set to default parameters (open gap 9, extension gap 1 penalties). Proteins with even greater similarity to the reference sequence will show increasing percentage identities when assessed by this method, such as at least 70%, 75%, 80%, 85%, 90%, 95%, or even 99% sequence identity. When less than the entire sequence is being compared for sequence identity, homologs will typically possess at least 75% sequence identity over short windows of 10-20 amino acids, and can possess sequence identities of at least 85%, 90%, 95% or 98% depending on their identity to the reference sequence. Methods for determining sequence identity over such short windows are described at the NCBI web site.

Protein homologs are typically characterized by possession of at least 70%, such as at least 75%, 80%, 85%, 90%, 95% or even 98% sequence identity, counted over the full-length alignment with the amino acid sequence using the NCBI Basic Blast 2.0, gapped blastp with databases such as the nr or swissprot database. Queries searched with the blastn program are filtered with DUST (Hancock and Armstrong, 1994, Comput. Appl. Biosci. 10:67-70). Other programs use SEG.

One of skill in the art will appreciate that these sequence identity ranges are provided for guidance only; it is possible that strongly significant homologs could be obtained that fall outside the ranges provided. Provided herein are the peptide homologs described above, as well as nucleic acid molecules that encode such homologs.

Nucleic acid sequences that do not show a high degree of identity may nevertheless encode identical or similar (conserved) amino acid sequences, due to the degeneracy of the genetic code. Changes in a nucleic acid sequence can be made using this degeneracy to produce multiple nucleic acid molecules that all encode substantially the same protein. Such homologous peptides can, for example, possess at least 75%, 80%, 90%, 95%, 98%, or 99% sequence identity determined by this method. When less than the entire sequence is being compared for sequence identity, homologs can, for example, possess at least 75%, 85% 90%, 95%, 98% or 99% sequence identity over short windows of 10-20 amino acids. Methods for determining sequence identity over such short windows can be found at the NCBI web site. One of skill in the art will appreciate that these sequence identity ranges are provided for guidance only; it is possible that significant homologs or other variants can be obtained that fall outside the ranges provided.

Subject: Living multicellular vertebrate organisms, a category which includes, both human and veterinary subjects that require an increase in the desired biological effect. Examples include, but are not limited to: humans, apes, dogs, cats, mice, rats, rabbits, horses, pigs, and cows.

Therapeutically Effective Amount: An amount sufficient to achieve a desired biological effect, for example an amount that is effective to decrease the size (i.e. volume), side effects and/or metastasis of prostate cancer. In one example, it is an amount sufficient to decrease the symptoms or effects of a prostate carcinoma, such as the size of the tumor. In particular examples, it is an amount effective to decrease the size of a prostate tumor and/or prostate metastasis by at least 30%, 40%, 50%, 70%, 80%, 90%, 95%, 99% or even 100% (complete elimination of the tumor).

In particular examples, it is an amount of a modified PA molecule effective to decrease a prostate tumor and/or an amount of prostate cancer cells lysed by a modified PA, such as in a subject to whom it is administered, for example a subject having one or more prostate carcinomas. In other examples, it is an amount of a modified PA molecule and/or an amount of prostate cancer cells lysed by such a modified PA molecule, effective to decrease the metastasis of a prostate carcinoma.

In one embodiment, the therapeutically effective amount also includes a quantity of modified PA and/or an amount of prostate cancer cells lysed by a modified PA sufficient to achieve a desired effect in a subject being treated. For instance, these can be an amount necessary to improve signs and/or symptoms a disease such as cancer, for example prostate cancer.

An effective amount of modified PA and/or prostate cancer cells lysed by such a modified PA molecule can be administered in a single dose, or in several doses, for example daily, during a course of treatment. However, the effective amount of will be dependent on the subject being treated, the severity and type of the condition being treated, and the manner of administration. For example, a therapeutically effective amount of modified PA can vary from about 1-10 mg per 70 kg body weight, for example about 2.8 mg, if administered iv and about 10-100 mg per 70 kg body weight, for example about 28 mg, if administered intraprostatically or intratumorally. In addition, a therapeutically effective amount of prostate cancer cells lysed by PA (variant or wild-type) can vary from about 10.sup.6 to 10.sup.8 cells.

Therapeutically effective dose: In one example, a dose of modified PA sufficient to decrease tumor cell volume, such as a prostate carcinoma, in a subject to whom it is administered, resulting in a regression of a pathological condition, or which is capable of relieving signs or symptoms caused by the condition. In a particular example, it is a dose of modified PA sufficient to decrease metastasis of a prostate cancer.

In yet another example, it is a dose of cell lysate resulting from contact of cells with a modified PA sufficient to decrease tumor cell volume, such as a prostate carcinoma, in a subject to whom it is administered, resulting in a regression of a pathological condition, or which is capable of relieving signs or symptoms caused by the condition. In a particular example, it is a dose of cell lysate resulting from contact of cells with a modified or wild-type PA sufficient to decrease metastasis of a prostate cancer.

Tumor: A neoplasm. Includes solid and hematological (or liquid) tumors.

Examples of hematological tumors include, but are not limited to: leukemias, including acute leukemias (such as acute lymphocytic leukemia, acute myelocytic leukemia, acute myelogenous leukemia and myeloblastic, promyelocytic, myelomonocytic, monocytic and erythroleukemia), chronic leukemias (such as chronic myelocytic (granulocytic) leukemia, chronic myelogenous leukemia, and chronic lymphocytic leukemia), polycythemia vera, lymphoma, Hodgkin's disease, non-Hodgkin's lymphoma (including low-, intermediate-, and high-grade), multiple myeloma, Waldenstrdm's macroglobulinemia, heavy chain disease, myelodysplastic syndrome, mantle cell lymphoma and myelodysplasia.

Examples of solid tumors, such as sarcomas and carcinomas, include, but are not limited to: fibrosarcoma, myxosarcoma, liposarcoma, chondrosarcoma, osteogenic sarcoma, and other sarcomas, synovioma, mesothelioma, Ewing's tumor, leiomyosarcoma, rhabdomyosarcoma, colon carcinoma, lymphoid malignancy, pancreatic cancer, breast cancer, lung cancers, ovarian cancer, prostate cancer, hepatocellular carcinoma, squamous cell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinomas, medullary carcinoma, bronchogenic carcinoma, renal cell carcinoma, hepatoma, bile duct carcinoma, choriocarcinoma, Wilms' tumor, cervical cancer, testicular tumor, bladder carcinoma, and CNS tumors (such as a glioma, astrocytoma, medulloblastoma, craniopharyogioma, ependymoma, pinealoma, hemangioblastoma, acoustic neuroma, oligodendroglioma, menangioma, melanoma, neuroblastoma and retinoblastoma).

Transformed: A transformed cell is a cell into which has been introduced a nucleic acid molecule by molecular biology techniques. As used herein, the term transformation encompasses all techniques by which a nucleic acid molecule might be introduced into such a cell, including transfection with viral vectors, transformation with plasmid vectors, and introduction of naked DNA by electroporation, lipofection, and particle gun acceleration.

Transgenic Cell: Transformed cells which contain foreign, non-native DNA.

Transgenic mammal: Transformed mammals which contain foreign, non-native DNA. In one embodiment, the non-native DNA is a modified PA which includes a PSA cleavage site, such as SEQ ID NO: 3, or a nucleic acid sequence which encodes for a protein shown in SEQ ID NOS: 24 or 25.

Variants or fragments or fusion proteins: The production of modified PA protein can be accomplished in a variety of ways (for example see Examples 12 and 16). DNA sequences which encode for a modified PA protein or fuision protein, or a fragment or variant of a protein (for example a fragment or variant having 80%, 90% or 95% sequence identity to a modified PA) can be engineered to allow the protein to be expressed in eukaryotic cells or organisms, bacteria, insects, and/or plants. To obtain expression, the DNA sequence can be altered and operably linked to other regulatory sequences. The final product, which contains the regulatory sequences and the therapeutic modified PA protein, is referred to as a vector. This vector can be introduced into eukaryotic, bacteria, insect, and/or plant cells. Once inside the cell the vector allows the protein to be produced.

A fusion protein which includes a modified PA, (or variants, polymorphisms, mutants, or fragments thereof) linked to other amino acid sequences that do not inhibit the desired activity of the protein, for example the ability to lyse PSA-secreting cells. In one example, the other amino acid sequences are no more than 5, 6, 7, 8, 9, 10, 20, 30, or 50 amino acid residues in length.

One of ordinary skill in the art will appreciate that the DNA can be altered in numerous ways without affecting the biological activity of the encoded protein. For example, PCR can be used to produce variations in the DNA sequence which encodes a variant PA toxin. Such variants can be variants optimized for codon preference in a host cell used to express the protein, or other sequence changes that facilitate expression.

Vector: A nucleic acid molecule as introduced into a host cell, thereby producing a transformed host cell. A vector can include nucleic acid sequences that permit it to replicate in the host cell, such as an origin of replication. A vector can also include one or more selectable marker genes and other genetic elements known in the art

Additional definitions of terms commonly used in molecular genetics can be found in Benjamin Lewin, Genes V published by Oxford University Press, 1994 (ISBN 0-19-854287-9); Kendrew et al (eds.), The Encyclopedia of Molecular Biology, published by Blackwell Science Ltd., 1994 (ISBN 0-632-02182-9); and Robert A. Meyers (ed.), Molecular Biology and Biotechnology: a Comprehensive Desk Reference, published by VCH Publishers, Inc., 1995 (ISBN 1-56081-569-8).

Variant Proaerolysin Molecules

Bacterial toxins, such as aerolysin produced by Aeromonas hydrophilia and .alpha.-hemolysin produced by Staph aureus, are beta-sheet proteins that oligomerize in the plasma membrane to produce pores that lead to rapid cytolytic cell death (FIG. 1). Pore formation physically disrupts the cell membranes, and results in death of cells in all phases of the cell cycle, including non-proliferating cells (i.e. G0 arrested). However, wild-type aerolysin kills cells indiscriminately. Herein disclosed is an inactive protoxin form of aerolysin (a variant PA) that can be targeted to, and activated by, prostate cancer specific proteins. One advantage of the disclosed variant PA molecules for treatment of localized and metastatic prostate cancer, is that it combines a proliferation independent therapy with prostate-specific drug delivery, resulting in minimal side effects to patients. One skilled in the art will understand that other protoxins, such as Clostridium septicum alpha toxin, Bacillus thuringiensis delta-toxin, and human perforin, can be substituted for proaerolysin.

Disclosed herein are variant PA molecules, including both DNA and protein sequences, which include a prostate-specific protease cleavage sequence. Examples of prostate-specific protease cleavage sequences include, but are not limited to: PSA, PSMA, and HK2 cleavage sequences. The prostate-specific protease cleavage sequence functionally replaces the native furin cleavage site of PA (for example, see FIGS. 5B-I). This replacement results in a proaerolysin variant that only becomes cytolytically active in the presence of enzymatically active protease, such as PSA, PSMA, or hK2. PSA is a serine protease with the ability to recognize and hydrolyze specific peptide sequences. It is secreted by normal and malignant prostate cells in an enzymatically active form and becomes inactivated upon entering the circulation. Since neither blood nor normal tissue other than the prostate contains enzymatically active PSA, the proteolytic activity of PSA was used to activate protoxins at sites of prostate cancer. Any PSA, PSMA, or hK2 cleavage site can be used. Examples of PSA cleavage sites include, but are not limited to, those shown in SEQ ID NOS: 5, 8, 11, and 14-21. In a particular example, the PSA cleavage site includes SEQ ID NO: 5.

In some examples, the furin cleavage site of PA (amino acids 427-432 of SEQ ID NO: 2) is deleted and a prostate-specific protease cleavage site, such as a PSA cleavage site, is inserted (for example, see FIG. 5B). In other examples, the furin cleavage site of PA is mutated and a prostate-specific protease cleavage site, such as a PSA cleavage site, inserted with in, or added to the N- or C-terminus of the furin site (for example, see FIG. 5C).

Also disclosed are variant PA molecules in which the PA binding domain is functionally deleted (for example, see FIGS. 5D-M). Such variant PA molecules can contain a native furin cleavage site (for example, see FIGS. 5J-M), whereby targeting to prostate cells is achieved by functionally replacing the PA binding domain with a prostate-tissue specific binding domain. Alternatively, variant PA molecules contain a prostate-specific protease cleavage site (for example, see FIGS. 5D-I), whereby activation of the protoxin primarily occurs in cells that secrete a prostate-specific protease. The PA binding domain includes about amino acids 1-83 of SEQ ID NO: 2. The binding domain can be functionally deleted using any method known in the art, for example by deletion of all or some of the amino acids of the binding domain, such as deletion of amino acids 1-83 of SEQ ID NO: 2 or 4 (for example see FIG. 5G), or such as deletion of one or more amino acids shown as amino acids 45-66 of SEQ ID NO: 2 or 4 (for example see FIG. 5D where the * represents one or more deletions). In other examples, the binding domain is functionally deleted by introduction of one or more site-specific mutations into the variant PA sequence, such as W45A, 147E, M57A,Y61A, and K66Q of SEQ ID NO: 2 or 4 (for example see FIG. 5D, where the * represents one or more mutations).

Variant PA molecules which include a prostate-tissue specific binding domain which functionally substitutes for the native PA binding domain are disclosed (for example, see FIGS. 5E, 5F, 5H and 5I-M). The use of one or more prostate-tissue specific binding domains can increase targeting of the disclosed variant PA molecules to the prostate cells and its metastases. Several prostate-tissue specific binding domains are known. Examples include, but are not limited to a luteinizing hormone releasing hormone (LHRH) sequence, such as those shown in SEQ ID NOS: 22 and 23, and antibodies that recognize PSA and/or PSMA.

One or more prostate-tissue specific binding domains can be linked to one or more amino acids of the disclosed variant PA molecules, but ideally, do not interfere significantly with the ability of the variant PA to be activated by a prostate-specific protease such as PSA, and the ability to form pores in cell membranes. For example, prostate tissue specific binding domains can be linked or inserted at an N- and/or C-terminus of a variant PA (for example, see FIGS. 5E and 5H). In some examples, the native binding domain of PA is deleted (i.e. amino acids 1-83 of SEQ ID NO: 2 or 4), such that attachment or linking of a prostate tissue specific binding domain to the N-terminus results in attachment to amino acid 84 of SEQ ID NO: 2 or 4 (for example, see FIGS. 5H and 5L). In other examples, smaller deletions or point mutations are introduced into the native binding domain of PA, such that attachment or linking of a prostate tissue specific binding domain to the N-terminus results in attachment to amino acid 1 of SEQ ID NO: 2 or 4 (or whichever amino acid is N terminal following functional deletion of the native PA binding domain) (for example, see FIGS. 5E and 5I). In some examples, the N-terminal amino acid of PA is changed to a Cys or other amino acid to before attaching a prostate-tissue specific binding domain, to assist in linking the prostate-tissue specific binding domain to the variant PA protein.

Alternatively or in addition, one or more prostate tissue specific binding domains can be attached or linked to other amino acids of a variant PA molecule, such as amino acid 215 or 300 of SEQ ID NO: 2 or 4 (for example, see FIGS. 5F, 5I, 5K and 5M). In some examples, a Cys amino acid replaces the native amino acid at that position. For example, the following changes can be made to SEQ ID NO: 2 or 4: Tyr215Cys or Ala300Cys. In one example, where the prostate tissue specific binding domain is an antibody, crosslinking can be used to attach antibodies to a variant PA, for example by reacting amino groups on the antibody with cysteine located in the PA variant (such as amino acids Cys19, Cys75, Cys159, and/or Cys164 of SEQ ID NO: 2).

Also disclosed are particular variant PA molecules, such as those shown in SEQ ID NOS: 3, 4, 6, 7, 9, 10, 12, 13, 24 and 25.

In some examples the disclosed variant PA molecules are linked or immobilized to a surface, such as a bead. The bead can also include a prostate-specific ligand to enhance targeting to a prostate cell, such as a localized or metastasized prostate cancer cell.

Treatment of Prostate Cancer Using Modified Proaerolysin

The variant PA molecules disclosed and discussed above are specifically activated to potent cytotoxins within prostate cancer sites via the proteolytic activity of prostate-specific proteases such as PSA, PSMA, and hK2. Targeting in some examples is achieved by including one or more prostate-tissue specific binding domains, such as LHRH peptide which can bind to its cognate LHRH receptor expressed by prostate cancer cells, or PSMA or LHRH antibodies, which can bind to PSMA or LHRH expressed on the surface of prostate cancer cells. One skilled in the art will recognize that the use of a variant PA molecule which includes a furin cleavage site and an LHRH peptide or antibody, can be used to treat other cancers which express LHRH receptors, such as melanoma and cancers of the breast, ovary and lung, using the variant PA molecules and methods disclosed herein. Furthermore, one skilled in the art will recognize that the use of a variant PA molecule which includes a furin or PSMA cleavage site, and/or a PSMA antibody, can be used to treat other cancers in which PSMA is expressed (e.g. in the vasculature of the tumor), such as cancers of the breast, colon, kidney, bladder and brain, using the variant PA molecules and methods disclosed herein.

The disclosed variant PA molecules, such as nucleic acids and/or proteins, can be administered locally or systemically using any method known in the art, to subjects having localized or metastatic prostate cancer. In addition, the disclosed variant PA molecules can be administered to a subject for immunostimulatory therapy. Due to the specificity of binding and activation of the disclosed variant PA molecules, local and systemic administration should have minimal effect on a patient's normal tissues and ideally produce little to no side effects.

In one example, the disclosed variant PA molecules are injected into the prostate gland (intraprostatically) and/or into the prostate tumor (intratumorally) in a subject having prostate cancer, such as a localized tumor. Such localized injection and subsequent lysis of prostate cancer cells within the prostate gland can produce an immunostimulatory effect leading to a decrease or elimination of micrometastatic disease in treated subjects. In this way, systemic disease is treated or reduced through a minimally toxic, locally applied therapy.

In addition, or alternatively, the disclosed variant PA molecules can be administered systemically, for example intravenously, intramuscularly, subcutaneously, or orally, to a subject having prostate cancer, such as a metastatic prostate tumor. Systemic therapy can also have an immunostimulatory anti-tumor effect. The disclosed variant PA molecules which include a PSA-cleavage site are not hydrolyzed by serum proteases or enzymatically inactive PSA within the blood. Instead, the unhydrolyzed disclosed variant PA molecules are delivered via the blood to the extracellular fluid within metastatic cancer deposits where they can be hydrolyzed to the active therapeutic toxin by the enzymatically active PSA secreted by these prostate cancer cells. Once hydrolyzed, the liberated toxin enters PSA-producing and non-producing bystander cells in the immediate vicinity due to its high membrane penetrating ability and induces the cytolytic death of these cells.

An additional method for systemically treating prostate cancer in a subject is also disclosed. In this method, prostate cancer cells are removed from the subject having prostate cancer, such as a metastatic prostate tumor. Alternatively or in addition, established prostate cancer cell lines can be used. Examples of prostate cancer cell lines that can be used include, but are not limited to: PSA-producing cells such as LNCaP (such as ATTC Nos. CRL-1740 and CRL-10995) and CWR22R (ATCC No. CRL-2505 and Nagabhushan et al., Cancer Res. 56(13):3042-6, 1996), or PSA non-producing cells such as PC-3 (ATCC No. CRL-1435) and DU 145 (ATCC No. HTB-81). The removed cells or cell lines are incubated or contacted with the disclosed variant PA molecules. This incubation results in lysis of the cells by the variant PA molecules, and production of a cell lysate which is administered to the subject. In one example, the method further includes administration of immunostimulatory factors, lysates from prostate cancer cells engineered to produce immunostimulatory factors, and/or irradiated prostate cancer cells (including prostate cancer cells engineered to produce immunostimulatory factors). Examples of immunostimulatory factors include, but are not limited to: granulocyte macrophage colony stimulatory factor (GM-CSF); members of the interleukin family of proteins such as but not limited to interleukin-2 and interleukin-6, granulocyte colony stimulatory factor (G-CSF); and members of interferon family such as interferon alpha, beta or gamma. Administration of such materials to a subject can be simultaneous with the cell lysate (co-administration), before administration of the cell lysate, and/or subsequent to administration of the cell lysate.

In one example, such administration enhances the ability of a subject to decrease the volume of a prostate tumor and/or a metastatic tumor. For example, the disclosed methods can reduce prostate tumor cell volume and/or a metastatic tumor cell volume, such as by at least 10%, for example by at least 20% or more. In addition, the disclosed methods can result in a decrease in the symptoms associated with a prostate tumor and/or a metastatic prostate tumor.

The disclosed variant PA molecules can be administered as a single modality therapy or used in combination with other therapies, such as radiation therapy and/or androgen ablative therapies (such as LHRH receptor agonists/antagonists, antiandrogens, estrogens, adrenal steroid synthesis inhibitors ketoconazole and aminoglutethimide). In addition, administration of the disclosed variant PA molecules can be alone, or in combination with a pharmaceutically acceptable carrier, and/or in combination with other therapeutic compounds, such as those that reduce the production of antibodies to the administered variant PA proteins (for example Rituximab and steroids) and other anti-tumor agents.

Disclosure of certain specific examples is not meant to exclude other embodiments. In addition, any treatments described herein are not necessarily exclusive of other treatment, but can be combined with other bioactive agents or treatment modalities.

EXAMPLE 1

Generation of PSA-Activated Proaerolysin Toxins

This example describes methods used to produce the variant proaerolysin toxins shown in Table 1, which are activated by PSA. One skilled in the art will understand that similar methods can be used to produce other variant proaerolysin proteins which are activated by PSA or any other prostate-specific protease. Such proteins can be produced by substituting the furin sequence of proaerolysin with a prostate-specific protease cleavage site, such as a PSA-specific cleavage sequence (see Example 9).

TABLE-US-00001 TABLE 1 PSA-specific proaerolysin variants Comparison to wt Proaerolysin Mutant ADSKVRRARSVDGAGQGLRLEIPLD (SEQ ID NO) Change(s) made (SEQ ID NO) (aa 424 448 of SEQ ID NO: 2) PSA-PA1 KVRRAR (aa 427 432 of SEQ ID NO: ADSHSSKLQSVDGAGQGLRLEIPLD (3 & 4) 2) changed to HSSKLQ (5) (aa 424 448 of SEQ ID NO: 4) PSA-1K KVRRARSV (aa 427 434 of SEQ ID ADSHSSKLQSADGAGQGRLEIPLD (6 & 7) NO: 2) changed to HSSKLQSA (8) (aa 424 448 of SEQ ID NO: 7) PSA-PA2 KVRRAR (aa 427 432 of SEQ ID NO: ADSQFYSSNSVDGAGQGLRLEIPLD (9 & 10) 2) changed to QFYSSN (11) (aa 424 448 of SEQ ID NO: 10) PSA-PA3 KVRRAR (aa 427 432 of SEQ ID NO: ADSGISSFQSSVDGAGQGLRLEIPLD (12 & 13) 2) changed to GISSFQS (14) (aa 424 448 of SEQ ID NO: 13)

The variant or modified proaerolysins (PA) shown in Table 1, include a proaerolysin sequence (wild-type PA shown in SEQ ID NOS: 1 and 2) in which the six amino acid furin protease recognition site (amino acids 427-432 of SEQ ID NO: 2) was replaced with a PSA substrate. For example, the variant proaerolysin (PA) toxin termed PSA-PA1 (SEQ ID NOS: 3 and 4), includes a PA sequence in which the furin cleavage site was replaced by the PSA substrate HSSKLQ (SEQ ID NO: 5).

Recombinant PCR was used to substitute the furin site of aerolysin (amino acids 427-432 of SEQ ID NO: 2) with a PSA-specific cleavage site (SEQ ID NO: 5, 8, 11 or 14) using methods previously described (Vallette et al., Nucl. Acids Res. 17:723-33, 1988). Briefly, recombinant PCR was performed in a final volume of 50 .mu.l which contained 0.2 mM deoxynucleoside triphosphate (dNTPs), 0.5 .mu.M forward and reverse primers, 0.1 .mu.g template DNA and 2.5 units cloned pfu polymerase in pfu Reaction Buffer [20 mM Tris-HCl (pH 8.8), 10 mM KCl, 10 mM (NH.sub.4).sub.2SO.sub.4, 2 mM MgSO.sub.4, 0.1% Triton X-100, and 0.1 mg/ml BSA].

Screening transformed cells for the proaerolysin insert was performed by PCR using Taq polyrnerase. A cocktail was prepared in PCR reaction buffer [50 mM KCl, 1.5 mM MgCl.sub.2, and 10 mM Tris-HCl (pH 9.0)] containing 0.2 mM dNTPs, 0.5 .mu.M forward and reverse primers and 5 units of Taq polymerase. Ten .mu.l samples of this cocktail were aliquoted into 0.2 ml tubes and transformed cells were added using sterile toothpicks.

The final PCR products were digested using appropriate restriction enzymes, then ligated into the cloning vector pTZ18u (BioRad) for amplification. Briefly, restriction digests were performed at 37.degree. C. for 90 minutes in Pharmacia One-Phor-All buffer [10 mM Tris-acetate (pH 7.5), 10 mM Mg-acetate, and 50 mM K-acetate] containing about one unit of restriction enzyme for every .mu.g of DNA. The resulting insert, and pTZ18u vector DNA, were mixed together in a ratio of approximately 5:1 and heated at 45.degree. C. for 15 minutes. Subsequently, the samples were diluted in One-Phor All buffer and ATP added to a final concentration of 1 mM for cohesive-end ligations or 0.5 mM for blunt-end ligations. Then, 11 units of T4 DNA ligase were added to each sample and the samples mixed gently. Ligations were carried out at 13.degree. C. for 4 hours (cohesive-end ligations) or 16 hours (blunt-end ligations).

DNA sequencing was performed to ensure the correct substitutions were made. The insert was subsequently isolated from the cloning vector and subcloned into the broad-host-range plasmid pMMB66HE (Furste et al., Gene 48:119-131, 1986) for expression in E. coli. E. coli DH5.alpha. cells were made competent using the CaCl.sub.2 wash method described previously (Cohen et al. Proc. Nat. Acad. Sci. USA 69:2110-4, 1972). Cells in log-phase (OD.sub.600=0.4-0.7) were harvested by centrifugation and washed in 1/4 volume of cold 100 mM MgCl.sub.2. The cells were pelleted again, and resuspended in two volumes of cold 100 mM CaCl.sub.2. The cells were then incubated on ice for approximately 45 minutes. The cells were then centrifuged and resuspended in 1/10 volume of 100 mM CaCl.sub.2. Incubation continued for an additional 45 minutes before the addition of glycerol to a final concentration of 15%. Competent cells were stored at -70.degree. C. until use.

Transformation of recombinant plasmids into competent E. coli cells was performed according to the method of Inoue et al. (Gene 96: 23-8, 1990). Competent cells (200 .mu.l aliquots) were incubated with 0.5-10 ng of DNA for one hour on ice. The cells were then subjected to heat shock at 42.degree. C. for 4 minutes. The cells were quickly transferred back onto ice for 5 minutes. Subsequently, 500 .mu.l of LB media was added to each sample and the cells incubated for 1 hour at 37.degree. C. with mild agitation. Aliquots (150 .mu.l) were plated onto LB agar containing 50 .mu.g/ml ampicillin. These plates were incubated ON at 37.degree. C.

Recombinant pMMB66HE clones were transferred into Aeromonas salmonicida strain CB3 (see Buckley, Biochem. Cell. Biol. 68:221-4, 1990) by conjugation using the filter-mating technique of Harayama et al. (Mol. Gen. Genet. 180:47-56, 1980). Use of this protease-deficient strain of A. salmonicida resulted in production of proaerolysin variants that were not contaminated by activated aerolysin and resulted in production of large quantities of protein. The final proaerolysin protein was purified by hydroxyapatite chromatography and ion exchange chromatography as previously described (Buckley, Biochem. Cell. Biol. 68:221-4, 1990). This method resulted in preparations of the proaerolysin proteins identical from batch to batch.

EXAMPLE 2

PSA-PAL Specifically Lyses PSA-Producing Cells in Vitro

This example describes methods used to determine the specificity of the proaerolysin variants described in Example 1. Such methods can be used to test the specificity of any proaerolysin variant that includes a PSA-specific cleavage site.

PSA-PA1 toxin was tested against PSA-producing LNCaP cells (American Type Culture Collection, Manassas, Va.) and non-PSA-producing TSU cells (Dr. T. Itzumi, Teikyo University, Japan). Cells were incubated in the presence of 10.sup.-12 M to 10.sup.-6 M toxin for 24 hours. Subsequently, cells were counted and scored for percent viable cells based on ability to exclude Trypan Blue. Concentration required to kill 50% of cells (IC.sub.50) was determined for the toxin against both LNCaP and TSU lines.

The LD.sub.50 for PSA-PA1 against PSA-producing cells was 10.sup.-10 M. In contrast, against non-PSA producing TSU cells the LD.sub.50 was .about.5.times.10.sup.-8 M. This result demonstrates that the PSA-PA1 toxin is specifically activated by PSA as evidenced by a 500-fold difference in toxicity against PSA-producing versus non-PSA producing human cancer cell lines.

EXAMPLE 3

PSA-PA1 is Not Activated in Blood Containing PSA

The disclosed prostate-specific protease-activated variant proaerolysin peptides, such as those described in Example 1, can be injected intraprostatically (or intratumorally) as local therapy for prostate cancer. The toxin can also be injected intravenously (or intramuscularly) as systemic therapy for metastatic prostate cancer. Those variant PA molecules which include a PSA site should not be activated in blood, because PSA is enzymatically inactivated in the blood due to the presence of a large molar excess of serum protease inhibitors such as .alpha..sub.1-antichymotrypsin and .alpha..sub.2-macroglobulin.

To test for non-specific activation of PSA-PA1 by other serum proteases and PSA in human serum, a sensitive hemolysis assay was performed as follows. Red blood cells (RBCs, 2% v/v) were added to plasma or buffer containing PSA-PA1 toxin.+-.PSA. The extent of hemolysis was assayed by measuring release of hemoglobin into the supernatant. Addition of 0.1% Triton results in 100% hemolysis within a few seconds and was used as the positive control. Amount of hydrolysis was expressed as a ratio of sample absorbance at 540 nm to absorbance of Triton treated sample. Pre-incubation of the PSA-PA1 toxin (10.sup.-8 M) with PSA in aqueous buffer alone for 1 hour prior to adding RBCs resulted in .about.45% hemolysis (FIG. 2).

To determine whether PSA-PA1 becomes activated in human plasma, PSA-PA1 toxin (10.sup.-8 M) was incubated in 50% human plasma for 1 hour. In a related experiment, excess PSA (10,000 ng/ml) was first added to the human plasma and allowed to incubate for several hours. The PSA-PA1 containing plasma.+-.PSA was then incubated with human RBCs (2% v/v). The addition of PSA-PA1 to human plasma, or human plasma spiked with high concentration of PSA, resulted in no appreciable hemolysis (i.e. <1% of Triton control) (FIG. 2). These results demonstrate that PSA-PA1 can be administered systemically without any significant activation in the blood, even if the blood contains measurable PSA.

EXAMPLE 4

In Vitro and In Vivo Toxicity Proaerolysin Variants

This example describes methods used to determine the in vitro and in vivo toxicity of the disclosed modified proaerolysin proteins. Such methods can be used to measure the toxicity of any prostate-specific protease-cleavable proaerolysin variant protein.

To determine in vitro toxicity, a cell viability assay was performed as follows. E14 mouse T-cell lymphoma cells (ATCC TIB-39) were cultured at 10.sup.+5 cells per well in MTS/PMS Cell Titer 96 (Promega). Proaerolysin variants at 1.times.10.sup.-13 M-1.times.10.sup.-7 M was added as shown in FIG. 3, and incubated with the cells for 4 hours at 37.degree. C. Cell viability was subsequently determined by reading the plate on a plate reader, as directed by the manufacturer of the MTS/PMS kit. As shown in FIG. 3, the proaerolysin variants are less toxic than wild-type proaerolysin, with an LC.sub.50 of 4.times.10.sup.-9(PSA-PA1), 1.times.10.sup.-9(PSA-1K), and 1.times.10.sup.-7(PSA-PA2), in contrast to an LC.sub.50 of 1.5.times.10.sup.-10 for wild-type.

To determine in vivo toxicity, proaerolysin variants were administered to mice intravenously. Wild-type proaerolysin (SEQ ID NO: 2) was highly toxic to mice; a dose of 1 .mu.g caused death within one hour and the LD.sub.100 at 24 hours (i.e. the dose that kills 100% of animals within 24 hours) following a single IV injection was 0.1 .mu.g. In contrast, the LD.sub.100 of PSA-PA1 (SEQ ID NO: 4) at 24 hours post injection was 25-fold higher (i.e. 2.5 .mu.g total dose).

EXAMPLE 5

PSA-PA1 Reduces PSA-Secreting Tumor Cell Volume

On the basis of the toxicity data described in Example 4, a series of LNCaP bearing mice (human LNCaP prostate cancer xenografs which produce PSA) and a series of SN12C bearing mice (control mice which have a human renal carcinoma xenograft which does not produce PSA) were administered a single 100 .mu.l intratumoral injection of 0.25-25 .mu.g PSA-PA1 (0.1-10 times the LD.sub.100 dose). At 48 hours post injection, tumors were harvested, fixed and stained with H&E, and for Ki-67 (proliferative index) and Tunel (apoptotic index). The percent of tumor within sample was determined by calculating the ratio of viable tumor to total tumor area following image analysis of thin tumor sections.

As shown in FIG. 4, administration of 2.5-25 .mu.g of PSA-PA1, reduced tumor cell volume by 85-99%, while <20% reduction was observed at 0.25 .mu.g of PSA-PA1. In contrast, when mice bearing non-PSA producing SN12C human renal carcinoma xenografts (cells provided by Dr. Isaiah Fidler Anderson Cancer Center, Houston Tex.), were injected intratumorally with PSA-PA1, no significant reduction (i.e. <25%) in percent of viable tumor was observed over the same dose range (FIG. 4).

In control tumors (SN12C mice), the proliferative index was .about.40% 48 hours post administration of intratumoral PSA-PA1. In contrast, in the responding PSA-PA1 treated tumors the percent of Ki-67 positive cells was <1% at 48 hours. In addition, in the control tumors the apoptotic index was <1% 48 hours post administration of intratumoral PSA-PA1, while in the PSA-PA1 treated tumors all cells were positive by 48 hours (i.e. apoptotic index>99%).

These results demonstrate that the PSA-PA1 toxin efficiently and rapidly kills PSA-producing cells and this cytotoxicity in vivo is specific and dependent on the presence of PSA present in the extracellular fluid of tumors.

EXAMPLE 6

Targeting the Protoxin Via Prostate-Specific Binding Domains

This example describes methods to generate and use variant prostate specific protease-activated proaerolysin toxins in which the wild-type (or native) PA binding domain (approximately amino acids 1-83 of SEQ ID NO: 2) is functionally replaced with a prostate-tissue specific binding domain, such as the LHRH peptide. The native binding domain of PA can be functionally deleted by any method known in the art, for example by deletion of about amino acids 1-83 of SEQ ID NO: 2 (or fragments thereof, such as 45-66), or by insertion of mutations which decrease the ability of PA to concentrate in cell membranes.

The N-terminal GPI-anchor protein binding domain of PA (approximately amino acids 1-83 of SEQ ID NO: 2) can be functionally deleted (for example by deletion of amino acids 1-83 or by insertion of mutations which render the domain non-functional) without significantly effecting pore-formation. A binding domain is needed, however, to concentrate the toxin in the cell membrane. Mutant proteins that lack the binding domain are cytolytic, but only when cells are exposed to several orders of magnitude higher concentrations of toxin. Most in vivo toxicity of wild type and variant PSA-activated PA described above in Example 4 is due to non-specific binding to GPI-anchor proteins expressed by most mammalian cells. To generate a more specific, and less systemically toxic protoxin, the non-specific GPI-anchor protein binding domain of proaerolysin can be functionally deleted and replaced with a prostate tissue-specific binding domain.

Antibodies

An example of a binding moiety that can be used to target a modified proaerolysin protein with high specificity to prostate tissue is a fusion protein which includes a single chain antibody to a prostate-specific membrane protein, and antibodies that recognize PSMA or LHRH fused to the toxin domains. Ideally, attachment of the antibody to a modified PA molecule will not significantly interfere with the ability of the molecule to bind to and concentrate in a cell membrane, resulting in cell death. Antibodies can be attached to the N- or C-terminus of a modified PA using gene fusion methods well known in the art (for example see Debinski and Pastan, Clin. Cancer Res. 1:1015-22, 1995). For example, the antibody could replace the native small lobe of proaerolysin (amino acids 1-83 of SEQ ID NO: 2), or the antibody could be added to a PA molecule having mutations in the native binding domain. Such a modified proaerolysin can also include a PSA activation sequence to increase specificity. In one example, the antibody is a single chain antibody to PSMA fused to the toxin domain of PA.

Alternatively, antibodies or FAB fragments can be attached to a modified PA by covalent crosslinking (for example see Woo et al., Arch. Pharm. Res. 22(5):459-63, 1999 and Debinski and Pastan, Clin. Cancer Res. 1(9):1015-22, 1995). Crosslinking can be non-specific, for example by using a homobifunctional-lysine-reactive crosslinking agent, or it can be specific, for example by using a crosslinking agent that reacts with amino groups on the antibody and with cysteine located in the proaerolysin variant (such as amino acids Cys19, Cys75, Cys159, and/or Cys164 of SEQ ID NO: 2).

Ligands

Other binding moieties that can be used are small peptide ligands that bind to their cognate receptor expressed on the membrane of prostate cancer cells. An example includes, but is not limited to, natural and synthetic luteinizing hormone releasing hormone (LHRH) agonist peptides (for example see Genbank Accession No. CAA25526 and SEQ ID NOS: 22 and 23), which bind with high affinity to LHRH receptors, and peptides that can bind selectively to PMSA. LHRH receptors are expressed by a high percentage of human prostate cancers, but not by hematopoeitic cells. This differential expression provides binding specificity.

It is known that certain residues of LHRH, such as the Gly at the 6th position (Gly6), can be substituted without compromising receptor binding affinity (Janaky et al., Proc. Natl. Acad. Sci. USA 89:972-6, 1992; Nechushtan et al., J. Biol. Chem., 272:11597-603, 1997). Therefore, a variant PA toxin (in which the native binding domain is functionally deleted) can be produced which is covalently coupled to purified LHRH D-Lys6 (at the epsilon amine of this lysine).

LHRH D-Lys6 (SEQ ID NO: 23) can be attached to various portions of a modified proaerolysin protein having a functionally deleted binding domain. Ideally however, such placement will not significantly interfere with the ability of the toxin to insert into the membrane to form a pore. For example, the epsilon amine of the D-Lys6 analog can be coupled to the amino terminus of the modified proaerolysin via a dicarboxylic acid linker. Cleavage by furin or a prostate-specific protease (depending on which cleavage site is present in the proaerolysin peptide) will result in release of the C-terminal inhibitory portion while the toxin remains bound to the LHRH receptor.

Alternatively or in addition, the epsilon amine of the D-Lys6 analog of LHRH can be coupled directly to the C-terminal carboxyl of the modified proaerolysin lacking a functional wild type binding domain, by the addition of a Cys to the C-terminus of the modified PA, then crosslinking this Cys to the epsilon amine of the D-Lys6 analog of LHRH. This coupling will produce a derivative PA protein in which the LHRH peptide is attached to the C-terminal inhibitory domain of PA. Cleavage by furin or by a prostate specific protease, such as PSA, will liberate the toxin and leave the inhibitory fragment bound to the LHRH receptor. Therefore, pore formation should not be inhibited by tight binding to the receptor. In addition, recombinant fusion proteins can be produced in which modified LHRH peptides are fused to both the N- and C-terminus of the modified PA toxin lacking a functional native binding domain.

In other examples, a cysteine residue is introduced into the 6th position of the LHRH peptide and the peptide attached to the modified PA toxins via a disulfide bridge, for example at amino acids 215 and/or 300 of SEQ ID NO: 2, wherein amino acids 215 and/or 300 has been mutated to a cysteine. In another example, a recombinant protein is produced in which LHRH peptide is fused to the amino terminus of the modified PA toxin.

The resulting modified proaerolysin proteins which include a prostate tissue specific binding domain functionally substituted for the native PA binding domain, are tested in vitro and in vivo for binding specificity and toxin activation, using the methods described in Examples 1-5.

To demonstrate that prostate-tissue specific binding domains, such as LHRH or PSMA, can be functionally substituted for the native PA binding domain the following experiments can be performed. The methods described below describe the use of a molecule in which the native PA binding domain has been deleted, and LHRH linked to the resulting proaerolysin. However, similar methods can be used to test any variant PA molecule, such as molecules in which other prostate-tissue specific binding domains are used, and where the PA binding domain is mutated (for example by insertion of one or more of the following mutations: W45A, I47E, M57A, Y61A, K66Q, and W324A) instead of deleted.

LHRH-proaerolysin proteins containing the native PA furin activation sequence are produced. The specificity of binding of these toxins to LHRH receptor positive (LNCaP) and negative (TSU) cells is compared using methods described in Example 2. Both cell lines activate the wild type, furin-activation-site-containing PA. Therefore, while each line may activate the LHRH-proaerolysin proteins, the ideal peptide is one that is toxic at low concentrations to LHRH receptor positive cells. Using these methods, regions of proaerolysin to which the LHRH peptide can be attached without interfering with channel formation by the toxin are identified.

To demonstrate that LHRH-proaerolysin proteins both bind to LHRH receptor and are activated by PSA, toxins that contain a PSA-activation site instead of the furin site are generated using the methods described in Example 1. The activation of these toxins by LHRH positive, PSA-producing LNCaP cells is compared to activation by LHRH receptor negative, PSA non-producing TSU cells, using the methods described in Example 2.

LHRH-PSA-activated proaerolysin toxins are tested in vivo to demonstrate that introduction of the LHRH binding moiety results in decreased non-specific toxicity and increased targeting ability. As described above, the LD.sub.100 of these toxins following a single intravenous injection is determined and compared to the corresponding non-LHRH containing, PSA-activated toxin, using the methods described in Example 4. Subsequently, the toxin is administered at varying doses, such as 0.1 .mu.g to 1 mg, intravenously to LNCaP bearing animals to demonstrate that an enhanced anti-tumor effect is observed at higher, less toxic doses of toxin.

EXAMPLE 7

Determination of Proaerolysin Antigenicity

This example provides methods to determine if the modified proaerolysin peptides disclosed herein are antigenic. In addition, methods to reduce potential antigenicity are disclosed.

As described in the Examples above, intratumoral injection of PSA-PA1 (SEQ ID NO: 4) demonstrates the usefulness of the toxin as therapy for localized prostate cancer via intraprostatic injection. However, such a therapy can also be administered by other routes, such as intravenous (iv), intramuscularly, orally, etc., as a systemic therapy for metastatic prostate cancer. However, systemic administration of the variant PA peptides disclosed herein may result in the development of a neutralizing antibody response that would limit repeat dosing.

The kinetics and magnitude of the antibody response to any of the PA variants disclosed herein can be determined as follows. For example, the antigenic response to PSA-PA1 (SEQ ID NO: 4) can be determined in immunocompetent mice, to develop a dosing regimen that can be used in a immunocompetent human. Immunocompetent mice (C57-BL6) are administered iv doses of PSA-PA1 (SEQ ID NO: 5) both daily.times.5 and weekly.times.3 at a dose range from 0.1 .mu.g to 5 .mu.g. Mice are sacrificed at varying intervals (e.g. following single dose, following multiple doses) and serum obtained. An ELISA-based assay can be used to detect presence of anti-proaerolysin antibodies. In this assay, a defined quantity of proaerolysin is fixed to the polystyrene surface in 96-well plates. Following adequate blocking with bovine serum albumin (BSA), serum from mice exposed to proaerolysin is added to the wells at varying dilutions. After a defined incubation time, wells are washed, and alkaline phosphatase linked goat-anti-mouse secondary antibody is added, followed by substrate. The amount of antibody present is determined by measuring absorbance in a spectrophotometer, which permits determination of the time course and magnitude of the antibody response by varying schedules and doses of iv PSA-PA1.

To decrease antigenicity of proaerolysin, the native binding domain can be functionally deleted and replaced, for example with LHRH, as described in Example 6. The antigenicity of such peptides can be determined following exposure to varying schedules of LHRH-proaerolysin proteins which lack portions of the native binding domain using the methods described above. Another method that can be used to allow continued treatment with prostate-specific protease activated toxins is to use alternative lytic toxins with non-overlapping antigenicity (see Example 10). One example is to use a modified, structurally related bacterial toxin such as Clostridium septicum alpha toxin that can also be activated by a prostate-specific protease, such as PSA, but would not be recognized or neutralized by antibodies that recognize PA (see Example 10). Another example is to use a pore-forming toxin produced by human tissues, such as human perforin produced by cytolytic human T cells. Modified perforins in which the wild type activation sequence is replaced by peptides that are substrates for prostate-specific proteases, such as PSA, can be administered and not produce an antibody response because the proteins are of human origin.

In addition, methods are provided to determine whether the antibody produced can neutralize the cytolytic properties of aerolysin. A disclosed modified PA (such as SEQ ID NOS: 4, 24 and 25) is incubated with PSA to activate the toxin. Activated toxin is incubated with antibody containing serum at varying doses. Washed RBC's are added to determine degree of hemolysis versus control (non-serum exposed toxin) as described above in Example 2.

In addition, PSA-producing tumor bearing animals can be inoculated twice with PSA-activated proaerolysin toxins, then rechallenged with a lethal dose of toxin (see Example 4) to determine if antibody neutralizes toxicity in vivo. Subsequently, the anti-tumor response following intratumoral injection of vaccinated animals is assessed, to determine if antibodies neutralize toxin when injected intratumorally.

EXAMPLE 8

Induction of a Systemic Immunostimulatory Response

This example provides methods that can be used to demonstrate that proaerolysin-mediated cell lysis produces a systemic immunostimulatory effect. Such a systematic immunostimulatory effect resulting from an intraprostatic administration of the toxins disclosed herein would provide both local therapy for prostate cancer within the prostate gland, while simultaneously induce a systemic antitumor effect against occult micrometastatic disease. Alternatively or in addition, subjects can be vaccinated with modified proaerolysin-lysed prostate cancer cells in the presence or absence of cytokines, such as GMCSF, to treat recurrent or initial metastatic disease.

Administration of Prostate Tumor Cells Lysed with Modified Proaerolysin

To demonstrate that modified proaerolysin treated cells stimulate a systemic immune response in a subject, the following methods can be used. Briefly, subjects (such as immunocompetent mice or a human subject having prostate cancer) are administered prostate tumor cells (such as TC2 mouse prostate tumor cells, prostate cancer cells obtained from the subject having prostate cancer, and/or human prostate cancer cell lines for administration to patients) which have been lysed with one or more modified proaerolysin molecules disclosed herein. To determine whether a systemic immune response occurs, mice that have been administered lysed prostate cancer cells can be rechallenged with the same cells and growth of these inoculated tumor cells measured. For example, C57-BL6 mice are subcutaneously injected with proaerolysin lysed TC2 cells. To accomplish this, 10.sup.7 TC2 cells are lysed by incubation with proaerolysin for 1 hour at 37.degree. C. in sterile phosphate buffered saline (PBS). Animals are administered two weekly injections of this cell lysate to stimulate an immune response to the TC2 cells. Animals are subsequently rechallenged with a subcutaneous injection of TC2 cancer cells one week after the second lysate inoculation. TC2 is a mouse prostate cancer cell line derived from cancerous tissue isolated from a TRAMP mouse (Foster et al., Cancer Res. 57:3325-30, 1997). Control subjects will receive a similar number of cells that have been lysed by freezing and thawing, or that have been treated with radiation to induce apoptosis. Another control group will receive only the modified proaerolysin peptide. Tumor growth is compared between the groups using the methods described in Example 5.

For human subjects, prostate cancer cells can be obtained from patient at time of prostatectomy or biopsy or human prostate cancer cell lines can be used. Examples of human prostate cancer cell lines include PSA-producing LNCaP (such as ATTC Nos. CRL-1740 and CRL-10995) or CWR22R (ATCC No. CRL-2505 and Nagabhushan et al., Cancer Res. 56(13):3042-6, 1996) cells, or PSA non-producing PC-3 (ATCC No. CRL-1435) and DU 145 (ATCC No. HTB-81). Approximately 10.sup.7-10.sup.8 cells from either source (patient or cell lines) are incubated with 1 .mu.g of PA (wild-type or variant) for 1 hour at 37.degree. C. in sterile PBS. The resulting lysate is mixed thoroughly using vigorous pipetting, and suspension is administered to a subject via subcutaneous injection of .about.0.5 ml. Patients can be treated weekly for 3 doses. Patients are closely monitored with weekly physical exams. Immune response to subcutaneous PA can be monitored by assaying presence of antibodies to PSA, PSMA, and hK2, and assaying for B and T cell response using methodology previously described (Simons et al. Cancer Res. 59:5160-8, 1999).

Alternatively, or in addition, prostate cancer cells from a patient or prostate cancer cell lines engineered to express immunostimulatory proteins (such as GM-CSF) are incubated with PA (wild-type or variant) and used to produce the lysate (for example see Simons et al. Cancer Res. 59:5160-8, 1999). PA-lysed prostate cancer cells can be co-administered with irradiated prostate cancer cells producing immunostimulatory molecules. Irradiation induces cells to undergo apoptosis. The combination of cells killed by cytolysis and cells killed by induction of apoptosis is expected to produce superior immunostimulatory effects (Sauter et al. J. Exp. Med. 191:423-33, 2000). In one example, a subject is co-administered PA-lysed prostate cancer cells mixed with immunostimulatory protein. In another example, PA-lysed prostate cancer cells are administered subcutaneously, and an immunostimulatory protein, such as GM-CSF, interleukins, interferons, G-CSF (Dranoff et al. Proc. Nati. Acad. Sci. USA 90:3539-43, 1993) is systemically administered.

Intraprostatic Administration of Modified Proaerolysin

Another method that can be used to stimulate a systemic immune response in a subject having prostate cancer is to administer the modified proaerolysin toxins disclosed herein directly into the prostate gland of the subject. For example, modified proaerolysin toxins can be administered directly into the prostate gland (or the tumor itself) of a human subject having a prostate tumor, or into transgenic ProHA mice followed by adoptive transfer of hemagglutinin (HA) primed T cells. It is expected that cytolysis following intraprostatic injection of modified proaerolysin toxins will provoke immune response to HA in the prostate gland of ProHA mice, or to prostate tumor antigens in humans. ProHA transgenic mice express the influenza protein HA under the control of the prostate-restricted promoter probasin. ProHA mice express HA only in the prostate gland. In this system, specific CD4 and CD8 T cells from separate donor mice are primed by exposure to HA and then adoptively transferred into the HA transgenic animal.

Subjects are administered a sublethal intraprostatic dose of modified proaerolysin (for instance determined using the methods described in Example 4). Because ProHA mice do not produce PSA or an enzymatically equivalent homolog, wild type aerolysin is used. However in humans, one or more of the disclosed modified proaerolysin toxins are administered. Twenty-four hours after the intraprostatic injection, some mice are sacrificed and prostates removed and analyzed for degree of necrosis. A second group of injected mice will receive HA specific CD4 and CD8 T cells, which have been harvested from TCR transgenic donors inoculated with 10.sup.9 pfu of recombinant hemagglutinin expressing vaccinia virus as previously described (Staveley-O'Carroll et al., Proc. Natl. Acad. Sci. USA 95:1178-83, 1998). These T cells are Thy 1.1 + and are transferred into the ProHA mice which are Thy1.2+. Four days after adoptive transfer into the proaerolysin treated ProHA mice, recipient mice are sacrificed, and the spleen, axillary (irrelevant) and prostate draining nodes dissected, dissociated and washed.

To assess activation of specific T cells, the following assays can be performed: One million isolated cells are stained using a Phycoerythrin labeled anti Thy1.1 antibody and cychrome labeled antiCD4 or CD8 antibody (Pharmingen, San Diego, Calif.). Isolated cells are analyzed using FacsScan (Becton Dickinson). The index of expansion of HA specific T cells is calculated by divi