Chimeric papillomavirus virus-like particles induce a murine self-antigen-specific protective and therapeutic antitumor immune response

The use of chimeric virus-like particles represents a new strategy for delivering tumor antigens to the immune system for the initiation of antitumor immune responses. Immunization of DBA/2 mice with the P1A peptide derived from the P815 tumor-associated antigen P1A induced specific T-cell tolerance, resulting in progression of a regressor P815 cell line in all animals. However, immunization with a human papillomavirus type 16 L1 virus-like particle containing the P1A peptide in the absence of adjuvant induced a protective immune response in mice against a lethal tumor challenge with a progressor P815 tumor cell line. Additionally, we demonstrated that these chimeric virus-like particles could be used therapeutically to suppress the growth of established tumors, resulting in a significant survival advantage for chimeric virus-like particle-treated mice compared with untreated control mice. Chimeric virus-like particles can thus be used as a universal delivery vehicle for both tolerizing and antigenic peptides to induce a strong protective and therapeutic antigen-specific antitumor immune response.     

Targeting an anti-viral CD8+ T cell response to a growing tumor facilitates its rejection

 We demonstrate in a murine model that targeting an anti-viral T cell response to a growing tumor facilitates priming of a tumor-associated antigen (TAA)-specific, rejecting T cell response. Murine P815 mastocytoma cells grow aggressively in a syngeneic host. Transfected P815/S cells (expressing the hepatitis B surface antigen, HBsAg) also grow as subcutaneous tumors, but occasional ‘spontaneous’ rejections after transient growth are observed. Growth of P815/S tumors (but not of P815 tumors) is efficiently suppressed by a CD8⁺ cytotoxic T lymphocyte (CTL)-dependent immune mechanism in mice primed to HBsAg by DNA–immunization. In hosts immunized against HBsAg by DNA vaccination, HBsAg-specific CTL are generated. This specific CTL reactivity was targeted to s.c.-growing P815 tumors by intra tumor injections of either HBsAg-encoding plasmid DNA or viable P815/S cells; this treatment led to tumor rejection in 70–80% of the tumor-bearing animals. All rejecting animals showed a CD8⁺ CTL-dependent resistance to subsequent challenges by native, non-transfected P815 tumors. Targeting an established anti-viral (‘strong’) CTL response to a growing tumor hence is an efficient strategy to facilitate priming of a rejecting CTL response against (‘weak’) TAA in this system. Received: 18 December 1996 / Accepted: 6 February 1997     

Therapeutic effectiveness of the immunity elicited by P815 tumor cells engineered to express the B7-2 costimulatory molecule

 It is well accepted that inoculation of B7-1-transfected tumor cells into normal mice leads to tumor rejection and subsequent resistance to challenge. However, the effectiveness of B7-2-transfected tumor cells in eliciting protective antitumor immunity is less clear. Here we show that B7-2-transfected P815 tumor cells (B7-2⁺) are as effective as B7-1-transfected P815 tumor cells (B7-1⁺) in eliciting protective immunity in normal DBA/2 mice. In addition, B7-2⁺ cells were found to be at least as effective as B7-1⁺ cells in retarding tumor progression when admixed with parental P815 tumor cells prior to inoculation into normal mice. Moreover, the B7-2⁺ cells and the B7-1⁺ cells were equivalent in their ability to retard tumor growth when administered peritumorally into mice bearing established (approx. 3 mm in diameter) parental P815 tumors. Finally, P815 tumor cells infected with a recombinant replication-defective adenovirus encoding the murine B7-2 gene were effective in retarding the growth of established parental P815 tumors. Thus, B7-1 and B7-2 are comparable in terms of their ability to stimulate the generation of tumor-eradicating immunity in normal mice as well as in mice bearing established parental tumors. Moreover, adenovirus vectors can be used to generate B7-2-expressing tumor cells effective in the immunotherapy of established parental tumors. Received: 10 January 1996 / Accepted: 23 February 1996     

The antimetastatic function of concomitant antitumor immunity. I. Host Ly-1+2+ effector T cells prevent the enumeration of metastatic tumor cells in a biological assay

A mouse survival assay was evaluated for its suitability to enumerate metastatic P815 tumor cells in the draining lymph node and spleen of a B6D2 F1 (H-2b X H-2d) host bearing a primary intradermal P815 tumor. The mouse survival assay is based on the linear relationship between the log10 number of P815 tumor cells (H-2d) injected i.p. into mice and their mean survival time. It was found that the assay is capable of quantifying as few as 10 tumor cells in lymph node and spleen, but only if cell suspensions of these organs are treated with anti-H-2b serum and complement, in order to selectively destroy H-2bd host cells. This was necessary because host cells from the lymph node and spleen of a tumor-bearing host possessed antitumor functions, in that they were capable of destroying the H-2d P815 tumor cells when admixed with the tumor cells and injected i.p. into 800-rad irradiated test recipients. The kinetics of acquisition and loss of host cells with antitumor function and the Ly phenotype of these host cells suggest that they are the same cells that give the tumor-bearing host the capacity to express concomitant immunity against a tumor implant.     

The antimetastatic function of concomitant antitumor immunity. II. Evidence that the generation of Ly-1+2+ effector T cells temporarily causes the destruction of already disseminated tumor cells

Progressive growth of the P815 mastocytoma in an immunocompetent host evokes the generation of an antitumor immune response that can be measured in terms of the production of cytolytic Ly-1+2+ T cells in the draining lymph node and spleen. This immunity, designated concomitant immunity, is present on day 6 of tumor growth, peaks on day 9, and decays progressively thereafter. It fails to develop in mice made T cell deficient by thymectomy and lethal whole-body gamma-radiation, and reconstituted with syngeneic bone marrow cells (TXB mice). Employment of a mouse survival assay, capable of enumerating metastatic P815 cells in cell suspensions, showed that the P815 tumor metastasizes to the draining lymph node and spleen at the same rate in normal and TXB mice for the first 6 days of growth of an intradermal P815 tumor. By day 6 of tumor growth there were approximately 10(3) P815 cells in the draining lymph node in both types of mice. However, during the generation of concomitant immunity between days 6 and 9, the number of metastatic P815 cells in the draining lymph nodes and spleens of normal tumor-bearing mice declined by nearly 90%. After day 12, however, the number of tumor cells in the nodes and spleens increased concordantly with the decay of concomitant immunity. These findings, together with the demonstration that T cell-deficient mice failed to restrain the number of metastatic P815 cells in the draining lymph node and spleen, suggest that concomitant immunity is an important defense mechanism against the development of systemic disease. Additional evidence consistent with this interpretation was provided by studies which showed that adoptive immunization with spleen cells from concomitant immune donors significantly prolonged the median survival time of TXB tumor-bearing mice by destroying a substantial proportion of P815 tumor cells already seeded in the draining lymph node. Adoptive immunization also delayed the appearance of metastatic tumor cells in the spleen.     

Monocyte chemoattractant protein-1 (MCP-1) gene transduction: an effective tumor vaccine strategy for non-intracranial tumors

Recently, there has been renewed interest in the concept of tumor vaccines using genetically engineered tumor cells expressing a variety of cytokines to increase their immunogenicity. Human MCP-1 (JE) is a potent chemoattractant and activator of monocytes and T lymphocytes and thus a good candidate gene for a tumor vaccine. We therefore evaluated the efficacy of vaccines consisting of irradiated tumor cells transduced with the murine MCP-1 gene in the syngeneic 9L gliosarcoma brain tumor model. 9L cell lines stably expressing murine MCP-1 (9L-JE) and control cell lines expressing neomycin 3 phosphotransferase (9L-Neo) were generated by infection with a Moloney murine leukemia retroviral vector. Fisher 344 rats were immunized with intradermal injections of 5×10⁵ or 2×10⁶ irradiated (5000 cGy) 9L-JE, 9L-Neo, and wild-type 9L (9L-WT) cells. Two weeks later immunized an non-immunized animals were challenged with varyious doses of intradermal (5×10⁶–5×10⁷) or intracerebral (2×10⁴–5×10⁵) 9L-WT cells. Intradermal tumors grew in all non-immunized animals. No tumors grew in animals immunized with irradiated 9L-JE or 9L-Neo cells and challenged with inocula of fewer than 5×10⁵ 9L-WT cells. With higher inocula up to 10⁷ cells, tumors appeared in all the animals. Tumors in animals immunized with 9L-JE were always smaller than tumors in the other groups. In addition, only the 9L-JE vaccine protected against tumor inocula of 5×10⁷ cells. Thus vaccination with MCP-1-expressing cells was able to protect animals against at least a 100-fold larger number of challenge tumor cells than vaccination with control cells. In contrast to studies with intradermal tumors, immunization with 9L-JE and 9L-Neo produced only minimal protection against intracerebral tumors. There was no significant difference between the 9L-JE and 9L-Neo vaccines in intracerebral challenge. This study suggests that tumor vaccines expressing cytokine genes such as MCP-1 can increase the antitumor response. However, the protective effect of these vaccines appears to be largely limited to intradermal tumors rather than intracerebral tumors.     

Mouse complement receptor-related gene y/p65-neutralized tumor vaccine induces antitumor activity in vivo

Two mouse tumor cell lines, Meth A (BALB/c mouse-derived fibrosarcoma) and MM46 (C3H/He mouse-derived mammary tumor), were shown to express high levels of complement receptor-related gene y/p65 (Crry/p65), a membrane-bound complement-regulatory protein. Inhibiting the complement-regulatory activity of Crry/p65 with mAb 5D5 induced high levels of C3 deposition on in vivo tumor-derived Meth A and MM46 cells. To determine the effect of Crry/p65 blockade and increased C3 deposition on in vivo tumor growth, Meth A and MM46 cells were treated with 5D5 mAb and injected into BALB/c and C3H/He mice, respectively. Pretreating MM46 cells with 5D5 mAb significantly suppressed their tumorigenicity when injected s.c. Pretreatment with 5D5 mAb had a modest effect on Meth A s.c. tumor growth. Because complement is involved in the induction of an immune response, we investigated the effect of Crry/p65 blockade and increased C3 deposition on the immunogenicity of the tumor cells in a vaccination protocol. Vaccination of mice with irradiated Meth A cells pretreated with 5D5 mAb protected mice from subsequent challenge. In contrast, vaccination with irradiated Meth A cells without pretreatment was not protective. Survival was correlated with a high titer IgM response and specific CTL activity. These data demonstrate that the functional inhibition of Crry/p65 on tumor cells affects tumor growth and immunogenicity, and that the complement deposition resulting from this inhibition can act in concert with antitumor effector mechanisms to elicit potent antitumor immunity in vivo.     

Differential resistance to growth of a tumor expressing incompatible minor alloantigens reflects regulatory influences rather than differences in anti-minor-CTL-P frequencies

In previous studies it was found that BALB/c (H-2d) was more susceptible than (BALB/c X A)F1 (H-2d X H-2a) to a tumor bearing multiple mismatched minor histocompatibility antigens, the DBA/2 (H-2d) mastocytoma P815, and that this resistance was H-2 linked. In the present studies the immunologic basis of this effect was examined by comparing the cytotoxic-T-lymphocyte (CTL) responses of BALB/c with those of (BALB/c X A)F1. Despite the BALB/c X A)F1's 34-fold greater resistance to P815 in vivo, the numbers of effector cell precursors were found to be similar in the two hosts as shown by (a) similar anti-P815 CTL responses in vitro with T-cell growth factor, (b) similar secondary anti-DBA/2 MiHA responses after in vivo priming with irradiated P815, and (c) similar frequencies of anti-DBA/2 CTL precursors by limiting-dilution analysis. However, priming with proliferating P815 in vivo revealed a defect in the BALB/c animals: Spleen cells from such animals were unable to control the growth of contaminating P815 cells in vitro or to mount strong secondary CTL responses to DBA/2 antigens. The defective priming of BALB/c could be corrected when DBA/2 spleen cells were added to the P815 inoculum. This impaired priming by living tumor cells was not seen in (BALB/c X A)F1. It is concluded that the use of living P815 tumor cells revealed a defect in immunoregulation in BALB/c mice, which rendered them susceptible to tumor growth in spite of apparently adequate numbers of anti-minor-CTL precursors. How the additional H-2 products expressed in the (BALB/c X A)F1 might correct this defect is discussed.     

Induction by DNA immunization of a protective antitumor cytotoxic T lymphocyte response against a minimal-epitope-expressing tumor

 In order to examine the use of DNA immunization to block tumor growth, we have developed a model system in which a defined 9-amino-acid epitope from the nucleoprotein of influenza virus is used as a surrogate tumor-associated antigen. A mastocytoma cell line of DBA/2 origin (P815) was transfected with a plasmid encoding the minimal H-2K^d-restricted NP(147–155) cytotoxic T lymphocyte (CTL) epitope, pCMV/NPep, to generate the cell line designated P815-NPep. Mice primed and boosted once with a plasmid encoding the full-length NP gene, pCMV/NP, but not with the minigene pCMV/NPep, developed a strong NP(147–155)-specific CTL response within 2 weeks after the boost. When challenged with 10⁴ P815-NPep cells, pCMV/NP-immunized DBA/2 mice were protected from tumor challenge, whereas control mice immunized with the vector backbone rapidly developed lethal tumor. Importantly, the P815-NPep-immune mice were also protected from a subsequent challenge with the untransfected parental tumor P815. By depleting the NP-immune mice of either CD4⁺ or CD8⁺ T cells and then challenging with 10⁴ P815-NPep tumor cells, it was determined that the CD8-depleted mice rapidly developed tumors, whereas the CD4-depleted or non-treated mice were protected. These data clearly indicate that intramuscular (i.m.) plasmid DNA immunization can be used to mobilize an effective CD8⁺ CTL-mediated antitumor response. Received: 8 May 1997 / Accepted: 28 August 1997     

CD8+ T cells circumvent immune privilege in the eye and mediate intraocular tumor rejection by a TNF-alpha-dependent mechanism

Although intraocular tumors reside in an immune-privileged environment, T cells can circumvent immune privilege and mediate tumor rejection without inducing damage to normal ocular tissue. In this study, we used a well-characterized tumor, Ad5E1 (adenovirus type 5 early region 1), to analyze the role of CD8+ T cells in the pristine rejection of intraocular tumors. It has been previously documented that Ad5E1 tumor rejection can occur in the absence of CD8+ T cells. However, here we find that CD8+ T cells infiltrated intraocular Ad5E1 tumors in C57BL/6 mice. Surprisingly, CD8+ T cells from tumor-rejector mice could mediate intraocular tumor rejection following adoptive transfer to SCID mice. In determining the mechanisms behind CD8+ T cell-mediated tumor rejection, we discovered that antitumor CTL activity was neither observed nor necessary for rejection of the intraocular tumors. CD8+ T cells from rejector mice did not produce IFN-gamma in response to Ad5E1 tumor Ags or use FasL to mediate intraocular tumor rejection. Also, CD8+ T cells did not use perforin or TRAIL, as CD8+ T cells from perforin knockout (KO) and TRAIL KO mice conferred protection to SCID recipient mice following adoptive transfer. We discovered that CD8+ T cells used TNF-alpha to mediate tumor rejection, because Ad5E1 tumor cells were highly sensitive to TNF-alpha-induced apoptosis and CD8+ T cells from TNF-alpha KO mice did not protect SCID mice from progressive Ad5E1 tumor growth. The results indicate that CD8+ T cells circumvent immune privilege and mediate intraocular tumor rejection by a TNF-alpha-dependent manner while leaving the eye intact and vision preserved.     

Intracamerally induced concomitant immunity: mice harboring progressively growing intraocular tumors are immune to spontaneous metastases and secondary tumor challenge

Intracameral inoculation of allogeneic P815 mastocytoma cells (DBA/2) into BALB/c mice resulted in progressively growing intraocular tumors. Intraocular tumor cells disseminated rapidly to the spleen and cervical lymph nodes, yet extraocular nests of tumor cells never developed into fulminant tumors. Further experiments showed that tumor cells were continuously seeded from the primary intraocular tumor and were rapidly cleared from extraocular sites. Hosts harboring intraocular P815 mastocytomas rejected tumorigenic doses of P815 cells inoculated subcutaneously or even into the contralateral anterior chamber. This systemic tumor immunity was found to be radiosensitive and T cell dependent. Spleen cells from animals with progressively growing intraocular tumors protected recipient mice challenged with intracamerally inoculated tumor cells and thus suggests that a cell-mediated mechanism is the underlying basis for this form of tumor immunity. The data indicate that mice harboring progressively growing intraocular tumors develop a potent state of "concomitant immunity," that prevents the development of metastases, yet is ineffective in controlling the primary tumor.     

Induction of delayed-type hypersensitivity and antitumor immunity by systemic BCG

Both delayed-type hypersensitivity (DTH) and antitumor resistance induced in mice by intravenous (i.v.) and local injection of highly immunogenic irradiated Meth A cells were potentiated by prior systemic BCG infection. DTH and antitumor immunity were not elicited by i.v. injection of poorly immunogenic irradiated mastocytoma cells, P 815 (MA), but were induced by the local injection of these cells when animals were systemically infected with BCG. The level of the potentiated response corresponded with the dose of immunogen up to an optimum, beyond which additional immunogen was suppressive. At all dose levels the subcutaneous (s.c.) route of immunogen inoculation was more effective than the i.v. route. Significant DTH was first detected 7 days after the local administration of immunogen and was correlated with antitumor immunity. Systemically administered BCG grew mainly in the liver and spleen until the development of maximal tuberculin sensitivity when the number of organisms decreased. However, the small number of mycobacteria that reached the peripheral lymph nodes remained constant after maximal tuberculin sensitivity but failed to augment the cell proliferation that occurred in these lymph nodes as a result of the local inoculation of irradiated tumor cells. Autoradiographs of such nodes revealed proliferation in the thymus-dependent areas whereas nodes from mice immunized with immunogen alone manifested B- as well as T-cell activity. Local immunization in both BCG-infected and uninfected hosts was also associated with a proliferative response in the red pulp of the spleen but the BCG-infected hosts differed conspicuously by virtue of the presence of tubercles and depletion of lymphoid cells from the periarteriolar sheath. Immunity generated by the local administration of immunogen in systemically infected mice was tumor specific and could be adoptively transferred with spleen cells.     

Augmentation of in vivo cytotoxic T lymphocyte activity and reduction of tumor growth by large multivalent immunogen.

Class I alloantigen incorporated into cell-size supported membranes provides an effective stimulus for in vitro stimulation of CTL responses. When alloantigen-bearing cell-size (5 microns) microspheres, termed large multivalent immunogen (LMI), were administered in vivo, no primary cytotoxic response to the Ag could be detected. However, coadministration of LMI and allogeneic tumor stimulator cells resulted in substantial augmentation of the resulting CTL response, compared with that obtained from mice that received just stimulator cells. Responses were augmented only when the same alloantigen was present on the LMI and on the stimulator cells, and the effector cells remained specific for the cognate alloantigen-bearing targets. The physical form of the alloantigen was critical for augmentation; alloantigen in liposomes had no effect on response levels. Tumor cell Ag in the form of purified plasma membrane vesicles can also be incorporated onto the surface of cell-size microspheres. As with allogeneic responses, tumor Ag on LMI specifically augmented the in vivo CTL activity generated in response to irradiated tumor cells in syngeneic mice. Administration of Ag-bearing LMI to mice inoculated i.p. with live P815, EL4, or RDM4 tumor cells resulted in a significant reduction in growth of the tumors in their syngeneic hosts. Similarly, LMI treatment significantly reduced growth of P815 as a solid s.c. tumor. LMI-mediated growth reduction occurred only when plasma membrane Ag from the cognate tumor was used to prepare the LMI, and Ag in the form of free plasma membrane vesicles was not effective. Although Ag has been used to manipulate in vivo humoral and Th responses, this has proven to be much more difficult for CTL responses. The ability of Ag-bearing LMI to affect significantly the in vivo levels of cytolytic response and to reduce syngeneic tumor growth has potential for application to tumor immunotherapy and, possibly, treatment of other diseases in which CTL can provide a protective effect.     

Generation of cytotoxic lymphocytes to syngeneic tumor by using co-stimulator (Interleukin 2)

Cytotoxic lymphocytes (CL) highly active against the syngeneic mastocytoma, P815, were generated from spleen cells of DBA mice cultured with co-stimulator (Interleukin 2) and P815. More CL activity was generated from spleen cells of P815 tumor-bearing mice than from spleen cells of normal mice. Thymus cells from tumor-bearing mice, however, did not produce increased CL activity. Most of the CL were Thy 1 and Ly 1 positive. The optimal culture conditions and kinetics were similar to those for the generation of allogeneic cytotoxic T lymphocytes. The cytotoxic activity against syngeneic P815 was similar in magnitude to the response of DBA spleen cells to allogeneic tumor lines and to the response of allogeneic CBA spleen cells to P815. Although CL generated from tumor-bearing mice did not lyse normal DBA cells, they did lyse, to a much lesser degree, a number of tumor cell lines other than the sensitizing P815. This nonspecific lysis was not H2 restricted nor was it restricted to tumors of lymphoid origin. Generation of nonspecific cytolytic activity was antigen independent, occurring in the presence of co-stimulator alone.     

Purified mouse mammary tumor and lymphoid cells in immune assays

Summary Tumor and lymphoid cell components from primary mammary adenocarcinomas of C3H/He mice were isolated simultaneously by velocity gradients. Viable tumor cells were obtained in sufficient numbers to test their in vivo and in vitro growth. Isolated tumor cells grew in 97% of inoculated syngeneic animals. In six assays with different tumors the effects of tumor-associated lymphoid cells (TAL) on in vivo tumor growth varied, enhancing in three and delaying in two experiments. Isolated tumor cells from animals with enhancing TAL grew faster in nonirradiated mice, whereas tumor cells from animals with inhibitory TAL grew better in irradiated animals. Isolated tumor cells also proliferated in cell culture, where they averaged 35% primary plating efficiency. Separated tumor cells were used in short-term ⁵¹Cr-release assays with TAL, tumor-bearer lymph node and spleen effectors. Cytotoxicity was detected in only five of 25 assays. In no case was there killing by lymphocyte populations from normal animals. In the present report we describe a technique for the isolation of viable tumor and lymphoid cells from murine adenocarcinomas that allows study of interactions between these populations from the original tumor-bearing host.Postdoctoral fellow of the Fogarty International Foundation. Present address: Department of Surgery, Harvard Medical School and Brigham & Women's Hospital, Boston, MA 02215, USA     

Inhibition of tumor rejection by gammadelta T cells and IL-10

Although many tumors express tumor-specific antigens, most fail to stimulate effective immune responses. Tumors generally lack co-stimulatory molecules, which can lead to tolerance of tumor-specific T cells and progressive tumor growth. Here, we demonstrate that the ovalbumin (OVA) transfected EL4 tumor, E.G7-OVA, grows progressively in syngeneic mice even though the tumor can be rejected if the mice are immunized with OVA in adjuvant. E.G7-OVA grew more rapidly in RAG-1 deficient than sufficient mice suggesting that normal mice make an abortive immune response to this tumor. Depletion of gammadelta T cells or IL-10 augmented the ability of B6 mice to reject E.G7-OVA. Spleen cells from normal, but not IL-10 knockout, mice reconstituted rapid tumor growth in gammadelta T cell-deficient mice. Thus, gammadelta T cells play an important role in preventing immune elimination of this tumor by a mechanism that directly or indirectly involves IL-10.     

Identification of a second major tumor-specific antigen recognized by CTLs on mouse mastocytoma P815

Murine mastocytoma P815 induces CTL responses against at least four distinct Ags (AB, C, D, and E). Recent studies have shown that the main component of the CTL response against the P815 tumor is targeted against Ags P815AB and P815E. The gene P1A has been well characterized. It encodes the P815AB Ag in the form of a nonameric peptide containing two epitopes, P815A and P815B, which are recognized by different CTLs. Here, we report the identification of the P815E Ag. Using a cDNA library derived from tumor P815, we identified the gene coding for P815E. We also characterized the antigenic peptide that anti-P815E CTLs recognize on the MHC class I molecule H-2Kd. The P815E Ag results from a mutation within an ubiquitously expressed gene encoding methionine sulfoxide reductase, an enzyme that is believed to be important in the protection of proteins against the by-products of aerobic metabolism. Surprisingly, immunizing mice i.p. with syngeneic tumor cells (L1210) that were constructed to express B7-1 and P815E did not induce resistance against live P815, even though a strong anti-P815E CTL response was observed with splenocytes from immunized animals.     

In vivo activity of tumor necrosis factor (TNF) mutants. Secretory but not membrane-bound TNF mediates the regression of retrovirally transduced murine tumor.

We have previously demonstrated that murine tumor cells transduced with a retrovirus containing the cDNA encoding wild-type human TNF regress in vivo when injected into immunocompetent mice; this regression is T cell mediated. To determine whether membrane-associated or secreted TNF was responsible for tumor regression, we transduced a cloned murine fibrosarcoma 205 F4 with retroviruses encoding modified human TNF genes. The cloned tumor lines of one retroviral transduction expressed only membrane bound 26-kDa TNF. This TNF could not be cleaved or secreted, but was present on the cell surface. A second retrovirus caused the expression of only secretory 17-kDa TNF, as the transmembrane domain of the cDNA was deleted. The TNF produced by tumor cells transduced with either retroviral vector was functional in vitro as direct lysis of the TNF-sensitive target L929 by transduced tumor cells was demonstrated. The TNF present on 26-kDa expressing tumors was membrane bound as supernatants from cultured 17-kDa TNF expressing tumor cells but not 26-kDa TNF expressing tumors mediated the lysis of L929 cells. Both tumors were injected s.c. into syngeneic mice and tumor growth was measured serially. In repeated experiments, 26-kDa TNF expressing tumors grew progressively in all mice. In contrast, 17-kDa TNF expressing tumors grew for 10 days and then regressed with all animals free of tumor at 28 days. Tumor regression was abrogated by in vivo injection of an anti-TNF antibody. Similar results were obtained in a second tumor model, 203 E4. Thus regression of TNF transduced tumors in vivo requires secretion of TNF, as membrane-bound TNF is insufficient to elicit the host response.     

Host B7-1 and B7-2 costimulatory molecules contribute to the eradication of B7-1-transfected P815 tumor cells via a CD8+ T cell-dependent mechanism

B7-1 (CD80)-transfected P815 tumor cells were previously shown to elicit tumor-eradicating immunity that leads to the regression of B7-1+ P815 tumors after transient growth in normal syngeneic (DBA/2) mice. Here, we show that not only the B7-1 molecule but also the B7-2 (CD86) molecule contributed to the eradication of B7-1+ P815 tumors. The B7-1 molecule that contributed to the eradication of B7-1+ P815 tumors was expressed not only on the tumor cells but also on host APCs, including MAC-1+ cells. The B7-2 molecule that contributed to the eradication of B7-1+ P815 tumors was expressed only on host APCs, such as B220+ cells, and not on the tumor cells. In spite of the fact that B7-expressing host APCs contributed to the eradication of B7-1+ P815 tumors, only CD8+ T cells without help from CD4+ T cells were important for tumor eradication. Taken together, these findings indicate that in addition to the ability of B7-1-transfected tumor cells to stimulate CD8+ T cell-mediated tumor-eradicating immunity directly, such tumor cells can also stimulate CD8+ T cell-mediated tumor-eradicating immunity indirectly as a result of cross-priming through B7-expressing host APCs.     

Antitumor and antimetastatic activity of IL-23

The structure and T cell stimulatory effects of the recently discovered cytokine IL-23 are similar to, but distinct from, those of IL-12. Although the antitumor activities of IL-12 are well characterized, the effect of IL-23 on tumor growth is not known. In this study, murine CT26 colon adenocarcinoma and B16F1 melanoma cells were engineered using retroviral vectors to release single-chain IL-23 (scIL-23) to evaluate its antitumor activity. In BALB/c mice, scIL-23-transduced CT26 cells grew progressively until day 26 to an average size of 521 +/- 333 mm(3), then the tumors started to regress in most animals, resulting in a final 70% rate of complete tumor rejection. scIL-23 transduction also significantly suppressed lung metastases of CT26 and B16F1 tumor cells. In addition, mice that rejected scIL-23-transduced tumors developed a memory response against subsequent wild-type tumor challenge. Compared with scIL-12-expressing CT26 cells, scIL-23-transduced tumors lacked the early response, but achieved comparable antitumor and antimetastatic activity. These results demonstrated that IL-23, like IL-12, provided effective protection against malignant diseases, but it probably acted by different antitumor mechanisms. As a first step in identifying these antitumor mechanisms, tumor challenge studies were performed in immunocompromised hosts and in animals selectively depleted of various lymphocyte populations. The results showed that CD8(+) T cells, but not CD4(+) T cells or NK cells, were crucial for the antitumor activity of IL-23.