Pnp

6-Methylpurine (MeP)

Direct toxin/cellular necrosis

Present

viral thymidine kinases and a poor substrate for mammalian thymidine kinases, concentrations can be achieved that are lethal to cells expressing HSV-tk but are nontoxic to normal mammalian cells (10,11,15).

Culver et al. demonstrated the first in vivo application of suicide gene therapy for cancer using retroviral-mediated HSV-tk gene transfer via fibroblast packaging cells injected into brain tumors in mice (16,17). Since this initial study, HSV-tk suicide gene therapy has been applied and investigated in multiple tumor types, including thoracic, head and neck, and ovarian cancer (18-20). Critical to the tumor response described in these early studies was the observation that not all the tumor cells must express the HSV-tk gene for a complete or extensive tumoricidal effect. The term bystander effect has been attributed to the regression of noninfected surrounding tumor cells after HSV-tk delivery and GCV administration. There are both direct and indirect mechanisms driving the bystander effect that are discussed in more detail later in the chapter. Briefly, the direct effect is an exchange of toxic metabolites between HSV-tk infected tumor cells and neighboring noninfected cells (21,22). The indirect effect stems from antitumor immune responses add to the versatility of this suicide gene therapy strategy (23,24).

A potential disadvantage of the HSV-tk strategy is that it requires S-phase cell cycle activity and thus targets only dividing cells. At any given time, not all malignant cells within a tumor are cycling and different tumor types display different rates of cell doubling both in vitro and in vivo. Conditions must therefore be worked out that provide adequate HSV-tk gene expression and administration of GCV over long enough time to account for variability in cell cycling within a designated tumor. Although the bystander effect helps augment the antitumor effects, there is likely a threshold of HSV-tk transfer and expression and persistence of the enzyme that must be achieved to generate or sustain a significant therapeutic effect.

Although a potential disadvantage, the requirement for cell division may also be an advantage for both safety and tumor targeting for the HSV-tk strategy. Normal mammalian cells in general divide at much slower rates than tumor cells within an established malignancy. This differential in S-phase activity allows for design of HSV-tk delivery and expression that would preferentially kill tumor cells while minimizing or eliminating direct toxicity to surrounding or systemic normal tissues that may incorporate the gene and express the enzyme.

B. Cytosine Deaminase

The enzyme cytosine deaminase (CD) is found in a variety of fungi and bacteria, but is not found in mammalian cells. In these microbes, CD is activated during nutritional stress and normally catalyzes the deamination of cytosine to produce uracil. CD is also capable of converting the nontoxic prodrug 5-fluorocytosine (5-FC) into the metabolite 5-fluorouracil (5-FU) (24-26). In a second metabolic conversion process, intra-cellular enyzmes present in both microbial and mammalian cells then act on 5-FU to produce 5-fluorouridine 5'-triphos-phate and 5-fluoro-2'-deoxyuridine 5'-monophosphate. These toxic phosporylated metabolites disrupt both RNA and DNA synthesis resulting in direct cellular cytotoxicity. Because of the natural specificity of CD to fungi and relative lack of toxicity of 5-FC in human tissues at routine dosing, 5-FC has been developed as an antifungal agent. However, because 5-FU is also converted to toxic metabolites in mammalian cells, this compound has been developed and widely applied as a chemotherapeutic agent for cancer.

The original application of CD for cancer therapy involved implanting a capsule containing the enzyme into an established rat tumor whereupon the enzyme diffused into the tumor cells. The animals were subsequently treated with 5-FC, which resulted in an antitumor response (27). Upon the cloning of the gene and building on these early animal investigations, CD suicide gene therapy became a feasible strategy (24). Aside from the direct cytotoxicity with this system, a significant bystander effect has also been identified with this system. The CD system is different from the HSV-tk system and may have a bystander advantage because of the initial production of 5-FU. Because the initial conversion to a ''toxic metabolite'' (5-FC to 5-FU) does not involve phosporylation, the 5-FU produced within the cytoplasm may readily cross the cell membrane and enter surrounding tumor cells. Also, the phosphorylated toxic metabolites of 5-FU that disrupt DNA and RNA synthesis may also be transferred to surrounding cells via intercellular communication or similar mechanisms as reported for the HSV-tk system. In effect, the multiple conversion steps provide 2 possible means of bystander activity that theoretically provide an enhanced overall antitumor and bystander effect as compared with the HSV-tk system. The full efficiency of the CD strategy and its bystander effect, however, has not been elucidated and further preclini-cal investigations are needed to discern any significant bystander advantages.

The lack of natural CD within mammalian cells and safety of 5-FC dosing makes this enzyme-prodrug combination a natural choice for human cancer investigation. CD is presently the second most common suicide gene under preclinical and clinical investigation; however, it has certain limitations that may affect successful application of this strategy. The toxicity of 5-FU is not S-phase cell-cycle specific, but it does depend on cell proliferation for its effect. As with the HSV-tk system, the CD strategy preferentially affects tumor cells that have a higher rate of proliferation. Although this provides some inherent safety and tumor targeting specificity, heterogenous tumor cell populations with variable levels of cellular proliferation may reduce the overall therapeutic effect. To achieve substantial tumor cell toxicity, high doses of 5-FU at the cellular level are generally required. The need for high cellular levels of 5-FU requires both efficient CD gene transfer and expression followed by adequate systemic dosing of 5-FC. The need for higher doses of 5-FC prodrug introduces the potential for systemic intestinal toxicity. The reason for this potential intestinal side effect is that residing enterobacteria present in normal human gastrointestinal tissues produce CD. These normal intestinal bacteria thus are able to convert 5-FC to 5-FU and produce local cellular injury. Another major limiting factor is that 5-FU is only active for a maximum of 10 min. Thus, prolonged systemic treatment with 5-FC after CD gene transfer is required to maintain the production of intracellular 5-FU. Possibly the most limiting factor is the complexity of the metabolic conversion pathway itself. A multistep complex pathway provides many opportunities for tumor cells to acquire resistance to the actual therapy (28).

C. Varicella Zoster Virus Thymidine Kinase

The varicella zoster virus is also capable of expressing a unique thymidine kinase (VZV-tk) whose substrate specificity is distinct fromboth mammalian cellular kinases and the HSV-tk enyzme. Upon gene transfer of VSV-tk into recipient cells, cytotoxicity is induced by administration of the prodrug 9-(b-D-arabinofuranosyl)-6-methoxy-9H-purine, also known as araM (29). VZV-tk enyzme initially phosphorylates araM that is further metabolized by natural cellular enzymes (AMP deaminase, AMP kinase, nucleoside diphosphate kinase, and ad-enylosucinate synthestase lysase) into adenine arabinonucleo-side triphosphate (araATP). ara-ATP is highly toxic and therefore only small quantities of ara-M in the range of 1 to 100 ^m are required to directly kill cells that contain the VZV-tk enzyme (29). As with GCV in the HSV-tk system, araM is an excellent substrate for VZV-tk but not mammalian nucle-oside kinases. Normal mammalian cells are able to withstand over 1500 ^m of ara-M exposure. The overall sensitivity of a transduced cell to the ara-M prodrug is directly proportional to level of the VZ-tk activity.

This system is still relatively new for preclinical investigation, and the presence of a bystander effect remains to be proven. The disadvantages of VZV-tk suicide gene therapy are similar to those of HSV-tk, and further experimentation is required to define the efficacy and safety rationale for selecting between these two strategies.

D. Escherichia coli Nitroreductase

The NTR enzyme activates the relatively nontoxic prodrug dintrophenylaziridine CB1954 through a reduction process that generates a 4-hydroxylamine metabolite. This intermediate molecule further reacts with intracellular thioesters such as acetyl-CoA to produce a highly cytotoxic alkylating agent that is 10,000 times more toxic than the original prodrug (30).

There are a few proposed mechanisms by which CB1954 mediates its cellular toxicity after reduction by nitroreductase (NTR). The most commonly reported mechanism is through cross-linking DNA strands causing disruption of synthesis and DNA breaks that are directly cytotoxic to both dividing and nondividing cells (31). Some investigators have reported increased apoptosis after delivery of NTR and CB1954 to targeted cells (32) which was presumed to be a result of the DNA alterations incurred. It appears that CD1954 acts more rapidly than other prodrugs such as GCV in the HSV-tk system, with reports of cytolytic activity as early as 4 h after prodrug administration (31,33). One explanation for this com paratively rapid response is that the NTR strategy does not require cells to be in the S phase of growth.

This lack of specificity for dividing cells may prove advantageous in achieving maximal antitumor efficacy; however, it raises a significant safety concern regarding normal surrounding tissues. A critical issue in the future application of this system will be the development of tumor-specific targeting so that normal somatic cells will not be exposed to the cytotoxic effects of the enzyme and prodrug. Another potential drawback to the CB1954/NTR system is that a bystander effect has not been identified to date (32). These 2 limitations may preclude the clinical application and benefit of this system until increased efficiency of gene transfer and tumor-specific targeting is achieved.

E. Cytochrome P450 2B1

The hepatic enzyme cytochrome P450 B1 (CYP2B1) will convert the inert lipophilic prodrug cyclophosphamide (CPA) into an effective anticancer agent (34). CPA is initially converted into 4-hydroxy-cyclophosphamide (4-HCPA), which is naturally unstable and will spontaneously decompose into 2 toxic metabolites, acrolein and phosphoramide mustard (PM). Acrolein will promote covalent links in cellular proteins and PM induces DNA alyklation and results in DNA strand breaks during replication. The importance of acrolein in causing tumor cytotoxicity in vivo has yet to be proven; therefore, the major anticancer metabolite appears to be PM. The cytotoxic-ity of PM affects both dividing and nondividing cells and so may prove useful for tumors that have low levels of S-phase activity such as the glioblastoma brain tumor (35).

Tumor cells usually express low levels of CYP2B1 but the liver expresses higher levels. Systemic exposure to CPA at concentrations that would prove effective in killing tumor cells will result in high levels of toxic metabolites produced in normal liver tissue. These high levels of toxic metabolites not only injure normal liver tissue, but also are released into the circulation. Systemic toxicity has therefore limited the ability to use these prodrugs alone as cancer chemotherapeutic agents. Gene transfer of the CYP2B1 gene and the resulting upregulated expression of the enzyme, however, will allow for administration of smaller systemic levels of CPA prodrug to maintain tumor cytotoxicity while minimizing systemic toxicity.

One important advantage of this system is that the intermediate metabolite 4-HCPA is lipophilic and so can pass through cellular membranes (36). Diffusion of these metabolites throughout the nontransduced tumor cells results in a strong bystander effect. Despite this apparent augmentation of overall antitumor effects, the free diffusion of 4-HCPA and the fact that the CYP2B1 gene therapy strategy does not require cell division also carries the disadvantage of possible toxicity in surrounding normal tissue.

F. Carboxypeptidase G2

The bacterial enzyme carboxypeptidase G2, (CPG2) has no mammalian homolog and has been shown to activate the pro-

drug 4-[(2-chloroethyl)(2-mesyloxyehtyl0amino]bensoyl-L-glutamic acid (CMDA), which is a derivative of a benzoic acid mustard. The CPG2 enzyme removes the glutamic acid moiety from the CMDA prodrug and releases a toxic benzoic acid mustard (37), which requires no further enzymatic or decomposition process. The benzoic acid mustard is a strong alkylating agent and cross-links DNA, thus imparting toxicity to both dividing and nondividing transduced cells.

The single-step process of converting CMDA to the toxic mustard metabolite offers an advantage over other suicide gene therapy strategies that have intermediate metabolites and multistep conversion processes within a targeted tumor cell. If the cellular enzymes that are responsible for the second phase or multistep activation process become defective or deficient in the tumor cell, a significant resistance to the prodrug could develop (38). However, when the toxic metabolite is released directly from the initial step or prodrug cleavage, there is much less chance of developing resistance to the suicide gene therapy. Also, mustard alkylating agents such as benzoic acid mustard have the advantage that their cytotoxic-ity is dose related. This important factor further reduces the chances of resistance.

As mentioned with other suicide gene therapy strategies above, this broad killing is beneficial for tumors with significant number of cells in Go at the time of gene transfer and prodrug administration. However, the potential continues to exist for direct toxicity to surrounding normal tissues that are transduced with this suicide gene.

A substantial bystander effect has been documented with the CPG2 strategy in vitro when as few as 3.7% of the tumor cells were expressing CPG2 (39). This is one of the strongest reported bystander effects and serves as an advantage for this strategy.

The gtp gene encodes the enzyme xanthine guanine phophori-bosyltransferase known as XGPRT. This enzyme converts the xanthine analog, 6-thioxanthine (6-TX) into a weakly toxic purine analog that is subsequently phosphorylated into 6-thi-oxanthine monophosphate (6-XMP) by the same XGPRT enzyme. 6-XMP is then converted into the very toxic 6-thiogua-nine monophosphate (6-GMP). This suicide gene therapy has been studied in a retrovirus-mediated gene transfer strategy in sarcoma and glioma tumor cell line and animal model experiments (39a,39b). XGPRT may also prove valuable in ade-novirus-mediated gene transfer strategies.

H. E. coli Deo a strong bystander effect. The main reason for its strong bystander effect is its ability to freely diffuse through cell membranes and travel into surrounding tumor cells. Significant tumor necrosis has been demonstrated when as little as 1 in 100 to 1000 tumor cells have been effectively trasduced with viral vectors carrying the E. coli Deo gene. This strong suicide gene therapy effect and corresponding bystander effect has been studied in human colon, glioma, and melanoma cell lines in vitro and ovarian and glioma cell lines in vivo (39c,39d).

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment