The Bystander Effect

Although present to variable degrees for each suicide gene therapy strategy, the bystander effect provides a major advantage for these systems. The advantage of a bystander effect is obvious when considering that present gene transfer approaches provide relatively poor transduction efficiencies of tumors in vivo. Given that only a fraction of any targeted tumor can be effectively transduced, a supplemental antitumor effect that does not require exposure or uptake of the therapeutic gene is critical to the success of suicide gene therapy. As mentioned previously, the bystander effect is a phenomenon whereby neighboring nontransduced tumor cells are killed in conjunction with direct killing of transduced cells (Fig. 1). Multiple theories have been investigated and proposed as an explanation for this effect.

Figure 1 The bystander effect. The HSV-tk gene is delivered to a solid tumor resulting in effective transduction and gene expression in only a percentage of the tumor (shaded cells). Subsequent systemic treatment with ganciclovir (GCV) causes DNA disruption, cell death, and production of toxic metabolites. The toxic metabolites are then passed to surrounding nontransduced cells via gap junctions resulting in cell death.

When the Deo gene is expressed in target cells, the enyzme purine nucleoside phosphorylase (PNP) is produced. The PNP enzyme is not found in human cells, thus giving this system a level of selectivity and safety with respect to application in human tumors. PNP converts the prodrug 6-methyl purine deoxyriboside (MePdR) into a very toxic metabolite, 6-methyl purine (MeP). MePp is directly toxic to tumor cells and has

Figure 1 The bystander effect. The HSV-tk gene is delivered to a solid tumor resulting in effective transduction and gene expression in only a percentage of the tumor (shaded cells). Subsequent systemic treatment with ganciclovir (GCV) causes DNA disruption, cell death, and production of toxic metabolites. The toxic metabolites are then passed to surrounding nontransduced cells via gap junctions resulting in cell death.

A. Metabolic Cooperation and Gap Junctions

The most investigated and widely accepted theory to date is that toxic prodrug metabolites are passed between tumor cells through gap junctions. Gap junctions are small hexameric structures (2 nm) in the cell membrane that form part of a communication network between cells. This transfer of toxic metabolites via gap junctional intercelluar communication is founded in the principle of ''metabolic cooperation,'' which was first described by Subak-Sharpe et al. in 1966 (40). Metabolic cooperation is the process whereby low molecular weight molecules (< 1000 Daltons) are passed between cells that are in contact. Subsequent to these early studies, it was established that cells with gap junctions were ionically coupled and participated in metabolic cooperation (41). Cells that lacked gap junctions did not participate in this event.

The importance of cell contact in a suicide gene therapy system was identified by Moolten and colleagues who demonstrated bystander killing of nontransduced cells after HSV-tk gene transfer and GCV administration in vitro (10). Although the GCV prodrug can readily and passively diffuse across a cell membrane, the toxic phosphorylated metabolite is not a permeable molecule. However, phosphorylated GCV is approximately 400 Daltons in size and is well within the size limit for metabolic cooperation to occur via gap junctions (42). Subsequent to these early studies with the HSV-tk suicide gene therapy system, Bi et al. introduced the concept that the bystander effect was a result of metabolic cooperation (22). It was demonstrated that labeled phosphorylated GCV was able to enter adjacent contacting tumor cells and resulted in cell death. Gene transfer and expression of a much larger molecule such as p-galactosidase, however, was not transferred to nontransduced contacting tumor cells.

Gap junctional communication is believed to be mediated by a family of proteins called connexins of which 12 genes have been cloned (43). Definitive proof that gap junctional intercelluar communication and connexin activity played a major role in the bystander effect came from the investigations of Mensil et al. (44). In their experiments, HeLa cells were chosen because they exhibit very little gap junctional communication and have no detectable expression of known connexin genes (45). Upon gene delivery of HSV-tk to cultured HeLa cells, only the actual HSV-tk transfected cells were killed despite different levels of cell density or contact. When HeLa cells were transfected with a gene encoding the gap junctional protein connexin 43, both HSV-tk positive and non-transfected surrounding cells in contact were killed. This effect was abrogated when HSV-tk positive and negative HeLa cells were cocultured without cell-cell contact.

B. Transfer of Toxic Metabolites via Apoptotic Vesicles or Direct Transmembrane Diffusion

Another reported mechanism of bystander effect activity is the release of apoptotic vesicles by dying tumor cells after suicide gene therapy. Freeman and colleagues noted that HSV-tk-positive tumor cells exhibited characteristics of apoptosis when dying in culture (21). Microscopic analyses revealed cell shrinkage and detachments as well as chromatin condensation and vesicle formation. Further ultrastructural evaluation using transmission electron microscopy identified features consistent with apoptosis. Apoptotic vesicles released form HSV-tk- and GCV-treated tumor cells could transfer the toxic phosphorylated GCV metabolite or even the HSV-tk gene itself. The mechanism of transfer would involve phagocytoses of these vesicles by surrounding viable tumor cells. This presumed transfer of apoptotic vesicles was demonstrated using a fluorescent tracking dye and fluorescence microscopy and flow cytometry. Nontransfected tumor cells were able to phagocytose the labeled apoptotic vesicles generated from dying HSV-tk transfected cells. Further studies on the importance of this finding in mediating the extent of bystander activity are required before this principle is substantiated and accepted.

Certain toxic metabolites such as 5-FU and 6-methyl purine (MeP) do not require gap junctions or the phagocytoses of apoptotic vesicles in order to enter surrounding tumor cells. 5-FU and MeP are permeable to most cellular membranes and thus can freely diffuse throughout the tumor, resulting in a strong bystander effect that is not restricted by tumor cell-dependent numbers of gap junctions or phagocytotic capabilities.

C. Local Antitumor Immune Responses

Although metabolic cooperation via gap junctions has been consider the major mechanism of the bystander effect, there is a growing interest in the role of the immune system in this phenomenon (Fig. 2). A number of studies have reported the presence of an inflammatory infiltrate in dying or regressing tumors after both HSV-tk and CD gene therapy (46,47). Other investigations have described a lessened response to HSV-tk and GCV therapy for tumors grown in nude mice as compared with immunocompetent animals (23,48). The decreased effect in these athymic and therefore T cell-deficient mice suggests that a T cell-mediated immune response plays some role in tumor regression. In these studies, nude mice and sublethally irradiated mice failed to demonstrate subcutaneous tumor regression when the tumor cell population consisted of 50% HSV-tk-transduced cells. The same experiments in immuno-competent mice, however, did show tumor rejection with the 50% proportion of HSV-tk-transduced cells.

More specific immune and cytokine analyses were subsequently performed using an intraperitoneal tumor model (49). Upon treatment of established tumors with HSV-tk-trans-duced cells, the peritoneal exudate was analyzed for the presence of various cytokines. Expression of tumor necrosis factor-alpha (TNF-a), interleukin 1-alpha (IL-1a), and IL-6 was identified at 24 h after tumor treatment with HSV-tk-positive cells. After 48 h interferon gamma (INF-7) was identified and at 96 h, granulocyte-macrophage colony-stimulating factor (GM-CSF) was produced. There appeared to be a defined cascade of cytokine production that was specific to the HSV-

Iniii.il killinu 11 f~tk+ Dcliw MSV-lk niu. tu in or I ran sil u coil tmnor cdl>

\lllh:ims :]ru] [vll]ll:isll of 1'iminr Vinton*. Stimulus ]immmc t'clE In! illrj[n:ni

\lllh:ims :]ru] [vll]ll:isll of 1'iminr Vinton*. Stimulus ]immmc t'clE In! illrj[n:ni

Figure 2 Immune component to the bystander effect. As tumor cells are killed by the suicide gene therapy, there is a release of cellular debris and tumor antigens within the local tumor environment. The cell necrosis and antigen load stimulates recruitment and activation of tumor-specific immune cells. The immune cells attack the local-regional tumor and also initiate a systemic response that may target and kill metastatic disease or prevent recurrence.

Figure 2 Immune component to the bystander effect. As tumor cells are killed by the suicide gene therapy, there is a release of cellular debris and tumor antigens within the local tumor environment. The cell necrosis and antigen load stimulates recruitment and activation of tumor-specific immune cells. The immune cells attack the local-regional tumor and also initiate a systemic response that may target and kill metastatic disease or prevent recurrence.

tk-treated tumors. Immunohistochemical staining identified immune infiltrates consisting of macrophages and T lymphocytes that were predominantly in and surrounding regressing tumors subsequent to HSV-tk therapy. These representative studies as well as others have established a role for local immune responses in generating killing of both transduced and nontransduced tumor cells. The generation of an antitumor immune response against both transduced and nontransduced tumor cells greatly enhances the power of suicide gene therapy and its associated bystander effect.

D. Systemic Antitumor Immune Responses

There is increasing evidence that not only a direct local immune response but also a systemic immune response is generated from suicide gene therapy. The development of a systemic immunity has been demonstrated for both the HSV-tk and CD strategies. In animal studies with HSV-tk-expressing tumor cells, administration of GCV resulted in both direct tumor cell killing and an ''initial'' protection against a second challenge of wild-type tumor cells (23,50). This effect could not be sustained and eventually tumors grew at the site of second challenge. This initial antitumor immunity was not generated against a second local challenge of syngeneic but heterologous (antigen distinct) tumor cells. Using the same tumor strategy in T cell-deficient nude mice, the antitumor effects and initial systemic immunity against a second-site tumor challenge was greatly reduced. These findings again support the importance of a T cell-mediated tumor-specific immune response. This initial systemic antitumor immunity has been documented in multiple tumor cell types for the

HSV-tk strategy, although the extent and period of immune effect has been variable among studies.

Although most of the work centering on immune responses and suicide gene therapy has centered on HSV-tk therapy, the immune component to the bystander effect has also been identified in the CD scheme. Both fibrosarcoma and adenocar-cinoma cells that were retrovirally transduced to express CD showed variable levels of resistance to second wild-type tumor challenges subsequent to systemic administration of the prodrug 5-FC (51). Second tumor challenges with antigen distinct tumor lines from the same murine background, however, displayed normal tumor growth, indicating a tumor specificity to the immune response. These findings parallel the results noted in the HSV-tk system.

In the previous investigations, the fibrosarcoma and adeno-carcinoma cell lines were generated via carcinogen induction. Subsequent studies have confirmed both a local and systemic protective immune responses after CD and 5-FC therapy in a spontaneously occurring mammary carcinoma (49). Antibody depletion against specific lymphocytes was performed in the animal tumor model system to further define specific components to these immune responses. Upon antibody depletion of CD8-positive T lymphocytes in vivo, a significantly decreased level of local tumor regression and even increased tumor growth was noted in CD-expressing tumors that were treated with 5-FC. Antibody depletion of granulocytes also appeared to limit the antitumor responses. Deletion of CD4 T lymphocytes had no significant effect on local tumor regression from the CD and 5-FC treatment. Interestingly, depletion of CD4-positive T lymphocytes dramatically reduced or in some cases eliminated the systemic immunity against a second local challenge of wild-type tumor. Depletion of CD8 T lymphocytes also appeared to limit the systemic antitumor bystander effects of a second challenge of wild-type tumor. Overall, these studies support the ability of suicide gene therapy to generate some level of systemic immunity. The systemic protection, however, was present at variable levels and only short-term systemic immunity was evaluated.

Although systemic immune responses can be generated against second tumor challenges after suicide gene therapy, could it be possible to induce regression of existing tumor metastases after treatment of the primary tumor? Original mu-rine studies in multiple tumor types evaluating responses against an established subcutaneous wild-type tumor at a second site were not encouraging. Tumor killing and regression identified in subcutaneous flank tumors that expressed CD or HSV-tk did not induce a significant effect on wild-type (nontransduced) tumors growing on the opposite flank (21,26,51). This apparent lack of effect on the presence of established tumors at a distant site suggested that the immune bystander effect was not going to be effective against gross metastases present at the time of primary tumor therapy.

Further studies involving orthotopic metastases, however, have defined some role of this immune bystander effect in the treatment of metastatic disease. Consalvo et al. demonstrated regression of lung metastases in 20% of mice whose primary CD-expressing subcutaneous tumors regressed after 5-FC ad

Iniii.il killinu 11 f~tk+ Dcliw MSV-lk niu. tu in or I ran sil u coil tmnor cdl>

ministration (47). Up to 90% regression of HSV-tk-negative liver metastases was also described after elimination of HSV-tk-expressing tumors growing intraperitoneal in a murine colon adenocarcinoma model (52). Immunohistochemical studies confirmed the presence of increased inflammatory infiltrates in these regressing metastases as compared with controls. These representative studies support the concept that a systemic antitumor immunity or bystander effect is capable of generating variable levels of tumor regression in distant metastases.

The issue has not been resolved as to why established second site wild-type tumors are not affected by suicide gene therapy directed at the primary tumor but metastatic lung and liver tumors show some regression. Excluding possibilities such as natural antigenicity of various tumor types and inherent differences among animal tumor models, there are a few hypotheses to explain this somewhat contradictory finding. The first hypothesis is that anatomic location of second site or metastatic disease is an important issue in generating effective immune-mediated regression after primary tumor therapy. The lung and liver have an increased number of cells from the reticuloendothelial system as compared with the subcutaneous tissues. The increased reticuloendothelial cells surrounding the metastases may facilitate or enhance antitumor immune activity generated from the initial primary tumor treatment. Also, this environment within the lung or liver may promote increased or more efficient antigen presentation. The second hypothesis may be related to gross tumor burden. Although in multiple animal models a second site established tumor has not regressed after primary tumor treatment with suicide genes, an initial protection against repeat tumor challenges has been well documented. The second hypothesis is based on this finding. It is possible that the systemic immune response generated from the initial tumor treatment is not strong enough to manage the gross tumor burden of an established tumor but can handle the smaller tumorigenic doses given for a second challenge. Further experiments to evaluate this consideration of tumor burden or rate of cell division within the metastatic lung and liver models may help to answer this question.

Was this article helpful?

0 0
How To Bolster Your Immune System

How To Bolster Your Immune System

All Natural Immune Boosters Proven To Fight Infection, Disease And More. Discover A Natural, Safe Effective Way To Boost Your Immune System Using Ingredients From Your Kitchen Cupboard. The only common sense, no holds barred guide to hit the market today no gimmicks, no pills, just old fashioned common sense remedies to cure colds, influenza, viral infections and more.

Get My Free Audio Book


Post a comment