Inhibiting Effects of a Cyclic Peptide CNGRC on Proliferation and Migration of Tumor Cells In Vitro

The cyclic peptide Cys-Asn-Gly-Arg-Cys (CNGRC) has previously been demonstrated as a tumor vasculature-homing peptide, which can specifically bind to CD13/aminopeptidase N in vivo. However, the effect of the peptide (CNGRC) binding to tumor cells in vitro has not yet been reported. In this study, CNGRC and an irrelevant linear control peptide (SVSVG) were employed to investigate the specific binding properties and other cellular influences in vitro. Immunofluorescence revealed that the peptide CNGRC demonstrated high specificities to the cells MDA-MB-435S, A549, MDA-MB-231, SK-OV-3 and EA.hy926, respectively. The cell viability assay indicated that CNGRC inhibited the proliferation of tumor cells at 24, 48 and 72 h. Furthermore, the peptide efficiently inhibited the migration of tumor cells, but promoted the migration of the human umbilical vein cell line. These results demonstrate that the synthetic peptide CNGRC can bind to the tumor cells without aminopeptidase N (CD13) expressed on the membranes. Therefore, it is supposed that the mechanism of the peptide binding to tumor cells in vitro may be different from that in vivo.     

Structure-activity relationships of linear and cyclic peptides containing the NGR tumor-homing motif

Cyclic and linear peptides containing the Asn-Gly-Arg (NGR) motif have proven useful for delivering various anti-tumor compounds and viral particles to tumor vessels. We have investigated the role of cyclic constraints on the structure and tumor-homing properties of NGR peptides using tumor necrosis factor-alpha (TNF) derivatives containing disulfide-bridged (CNGRC-TNF) and linear (GNGRG-TNF) NGR domains. Experiments carried out in animal models showed that both GNGRG and CNGRC can target TNF to tumors. However, the anti-tumor activity of CNGRC-TNF was >10-fold higher than that of GNGRG-TNF. Molecular dynamic simulation of cyclic CNGRC showed the presence of a bend geometry involving residues Gly(3)-Arg(4). Molecular dynamic simulation of the same peptide without disulfide constraints showed that the most populated and thermodynamically favored configuration is characterized by the presence of a beta-turn involving residues Gly(3)-Arg(4) and hydrogen bonding interactions between the backbone atoms of Asn(2) and Cys(5). These results suggest that the NGR motif has a strong propensity to form beta-turn in linear peptides and may explain the finding that GNGRG peptide can target TNF to tumors, albeit to a lower extent than CNGRC. The disulfide bridge constraint is critical for stabilizing the bent conformation and for increasing the tumor targeting efficiency.     

Combination cytokine therapy inhibits tumor growth by generation of tumor-specific T-cell responses in a murine melanoma model

Various cytokines, including interferon α (IFNα), tumor necrosis factor α (TNFα), and granulocyte–macrophage colony-stimulating factor (GM-CSF), have been used as adjuvant therapy for advanced-stage melanoma with some success but with marked toxicity, which appears to be related to higher doses. We investigated the efficacy of IFNα, GM-CSF, and TNFα in various combinations to induce antitumor and immune responses in a B16F10 murine melanoma model. These studies showed that GM-CSF, IFNα, and TNFα, when injected together intratumorally, mediated significant inhibition of tumor growth. Tumor regression correlated with local tumor necrosis and significant infiltration of T cells. In addition, this injected intralesional cytokine cocktail also induced lymphadenopathy, with an increase in both CD4⁺ and CD8⁺ T cells in the draining lymph nodes. Furthermore, tumor-specific CD8⁺ T cells were identified from draining lymph nodes. These investigations identify the combined effects of IFNα, GM-CSF, and TNFα in induction of the adaptive immune response and generation of antigen-specific T-cell reactivity. These results support potential clinical trials of the low-dose cytokine combination as adjuvant therapy for melanoma.     

重组人白细胞介素－6与肿瘤坏死因子突变体融合蛋白的构建及表达

针对肿瘤坏死因子（ＴＮＦ）在肿瘤治疗剂量下产生的严重毒副作用及一些肿瘤细胞上白细胞介素－６（ＩＬ－６）受体明显增高的事实,根据ＴＮＦ结构与功能研究的最新信息,利用ＰＣＲ技术,对人ＴＮＦα基因进行了改造,并将其与人ＩＬ－６成熟肽编码区ｃＤＮＡ通过人工接头进行融合。融合蛋白在大肠杆菌中表达后,Ｗｅｓｔｅｒｎｂｌｏｔ分析表明,分子量约为３７ｋＤ;活性检测结果证实,该融合蛋白兼具有ＴＮＦ抗肿瘤活性和结合ＩＬ－６受体的能力,在高表达ＩＬ－６受体的人骨髓瘤细胞上测得的细胞毒活性较同样位点突变的ＴＮＦ高约３倍。     

Activation of tumor-infiltrating macrophages by a synthetic lipid A analog (ONO-4007) and its implication in antitumor effects

ONO-4007 is a novel synthetic analog of lipid A subunit and has been shown to exert antitumor activities on various experimental tumors with less toxicity than lipopolysaccharide. It remains unclear, however, what biological activities of this compound are relevant to its antitumor effects. We therefore investigated the activation of macrophages by ONO-4007 in vitro and in vivo and its implication in antitumor effects, using mouse MM46 mammary tumor as an experimental model. Intravenous injection of ONO-4007 produced significant therapeutic effects on this solid tumor. ONO-4007 could stimulate glycogen-elicited peritoneal macrophages in vitro, not only to produce tumor necrosis factor (TNF), but also to exert cytocidal activities against MM46 cells in vitro. Substantial TNF production was induced in tumor tissue by i. v. injection of ONO-4007, and its successive administration to tumor-bearing mice gave tumor-infiltrating macrophages a prominent in vitro tumoricidal activity and primed them for in vitro TNF secretion. These results suggest that activation of tumor-infiltrating macrophages to a direct tumoricidal state as well as to TNF secretion in tumor tissues may be at least some of the antitumor effects of this novel lipid A analog.     

Intratumoral coinjection of two adenoviruses, one encoding the chemokine IFN-gamma-inducible protein-10 and another encoding IL-12, results in marked antitumoral synergy

We have constructed a recombinant defective adenovirus that expresses functional murine IFN-gamma-inducible protein-10 (IP-10) chemokine (AdCMVIP-10). Injection of AdCMVIP-10 into s.c. tumor nodules derived from the CT26 murine colorectal adenocarcinoma cell line displayed some antitumor activity but it was not curative in most cases. Previous studies have shown that injection of similar s. c. CT26 tumor nodules with adenovirus-encoding IL-12 (AdCMVIL-12) induces tumor regression in nearly 70% of cases in association with generation of antitumor CTL activity. AdCMVIP-10 synergizes with the antitumor effect of suboptimal doses of AdCMVIL-12, reaching 100% of tumor eradication not only against injected, but also against distant noninjected tumor nodules. Colocalization of both adenoviruses at the same tumor nodule was required for the local and distant therapeutic effects. Importantly, intratumoral gene transfer with IL-12 and IP-10 generated a powerful tumor-specific CTL response in a synergistic fashion, while both CD4 and CD8 T cells appeared in the infiltrate of regressing tumors. Moreover, the antitumor activity of IP-10 plus IL-12 combined gene therapy was greatly diminished by simultaneous in vivo depletion of CD4+ and CD8+ T cells but was largely unaffected by single depletion of each T cell subset. An important role for NK cells was also suggested by asialo GM1 depletion experiments. From a clinical point of view, the effects of IP-10 permit one to lower the required gene transfer level of IL-12, thus preventing dose-dependent IL-12-mediated toxicity while improving the therapeutic efficacy of the elicited antitumor response.     

Differences in the biological activity of TNF alpha and TNF beta correlate with their different abilities for binding to the target cells.

TNF alpha and TNF beta were compared regarding their binding to different types of target cells, cytotoxic/cytostatic activity against murine and human tumor cell lines as well as human capillary endothelial cells, their ability to induce differentiation in myeloid leukemia cell lines, and induction of hemorrhagic tumor necrosis and tumor regression as well as lethal toxicity in tumor-bearing mice. The results show considerable quantitative differences in the biological activity between TNF alpha and TNF beta depending on the type of target cell which has been used. TNF beta was 3 fold more cytotoxic than TNF alpha against murine L929 fibroblasts and 3-5 times more active concerning the induction of hemorrhagic tumor necrosis, complete tumor regression and more toxic in tumor-bearing mice. In contrast to this, TNF beta was markedly less cytotoxic against human capillary endothelial cells and the human mammary carcinoma cell line MCF7 and much less cytostatic against the human myeloid leukemia cell lines HL60 and U937. The lesser antiproliferative effect of TNF beta correlated with a lower ability for induction of differentiation in these cell lines. Competitive radioligand binding assays showed that TNF beta was about 4 fold more effective than TNF alpha in competing with 125I-labeled TNF alpha for the binding to murine L929 fibroblasts. But it was 15-20 times less effective in binding to the human MCF7 cells and the human myeloid leukemia cell lines HL60 and U937. This revealed that, at least for these targets, the differences in the biological activity between TNF alpha and TNF beta are due to different abilities for binding to the target cells. Possible mechanisms for these different binding abilities are discussed.     

Murine tumor cells transduced with the gene for tumor necrosis factor-alpha. Evidence for paracrine immune effects of tumor necrosis factor against tumors

Studies of the anti-tumor activity of TNF-alpha in vivo have been hampered by the need to administer systemically toxic doses of the cytokine to obtain a curative response. To facilitate studies of the effect of high local concentrations of TNF-alpha on tumor growth and host immunity, a newly induced murine sarcoma was transduced with the gene for human TNF-alpha and the biologic characteristics of these cells were examined. We identified high and low TNF-producing tumor clones which exhibited stable TNF secretion over time. Significant amounts of membrane associated TNF were found in a high-TNF producing clone as well. No difference in the in vitro growth rates between TNF-producing and nonproducing cell lines was observed. In contrast, in vivo studies demonstrate that although unmodified parental tumor cells grew progressively when implanted s.c. in animals, tumor cells transduced with the TNF gene were found to regress in a significant number of animals after an initial phase of growth. This effect correlated with the amount of TNF produced and could be blocked with a specific anti-TNF antibody. Regressions of TNF-producing cells occurred in the absence of any demonstrable toxicity in the animals bearing these tumors. TNF-producing tumor cells could function in a paracrine fashion by inhibiting the growth of unmodified, parental tumor cells implanted at the same site. The ability of tumor cells to regress was abrogated by in vivo depletion of CD4+ or CD8+ T cell subsets and animals that had experienced regression of TNF-producing tumors rejected subsequent challenges of parental tumor. Our studies thus show that tumor cells elaborating high local concentrations of TNF regress in the absence of toxicity in the host and that this process requires the existence of intact host immunity. Studies of the lymphocytes infiltrating the gene modified tumors and attempts to use TNF gene modified tumor infiltrating lymphocytes to deliver high local concentrations of TNF to the tumor site without inducing systemic toxicity are underway.     

Synergistic action of human recombinant tumor necrosis factor with endotoxins or nontoxic poly A:U against solid Meth A tumors in mice

Summary Antitumor effects of i.v. injected human recombinant tumor necrosis factor (rTNF) against solid Meth A tumors in mice appeared to be critically dependent on the dose and were limited by its toxicity. Extensive necrosis and complete cures were only induced by doses having untoward effects, such as diarrhea, hypothermia, ruffled fur, and lethargy. Murine tumor necrosis serum (TNS, 0.5 ml) had about the same antitumor potential and induced all side effects except diarrhea. More extensive necrosis and approximate doubling of the incidence of complete regression in the absence of gross side effects were observed upon administration of a low dose of rTNF combined with detoxified endotoxin, nontoxic poly A:U, or submicrogram doses of toxic endotoxin. The separate constituents had little antitumor effects, if any at all. Increasing the dose of toxic endotoxin resulted in a further potentiation of necrosis, overt toxicity, but no cures. Muramyl dipeptide and interferon / did not potentiate effects of rTNF. In vitro growth of Meth A cells was not inhibited by toxic endotoxin, rTNF or the combination, although TNS was highly inhibitory. Data show that therapeutic effects of rTNF and its synergy with endotoxin are not due to direct effects on the tumor cells and that the extent of prompt in vivo tumor necrosis does not predict the course of tumor growth. Therapeutic effects of both TNS and toxic endotoxin probably involve a synergy between low levels of TNF and other factors/effects induced by endotoxin. Detoxified endotoxin and poly A:U probably induce the latter effects and little or no TNF, so explaining the absence of side effects, their weak antitumor potential, and their powerful synergistic action with rTNF. A role for interferon / as an induced synergistic factor is not likely. Muramyl dipeptide and TNF might share properties needed for synergy with endotoxins. Present address: Department of Immunotoxicology, State University of Utrecht, Biltstraat 172, 3572 BP Utrecht, The Netherlands     

Characterization of Immunological Activity of a Low Toxicity Antitumor Lipopolysaccharide from Bordetella pertussis

Immunological properties of a low toxicity lipopolysaccharide (BP-LPS) extracted from Bordetella pertussis (Tohama strain) which was reported to have high antitumor activity against murine tumors were examined and compared with those of LPS extracted from other enterobacteria. The activation or stimulation of murine macrophages and lymphocytes by these LPS, including TNF induction, was found to be similar. However, BP-LPS was clearly less active in its stimulation of murine and human neutrophils as estimated by neutrophil-adherence assay and by their TNF production than E. coli LPS. Furthermore, BP-LPS also suppressed the activation of human neutrophils by Escherichia coli LPS. A comparative study with 7 LPS preparations indicated that their toxicity in terms of animal body weight loss correlated with their ability to induce human neutrophil adherence. The inability of BP-LPS to activate neutrophils may thus have some bearing on its low toxicity.     

Local immune checkpoint blockade therapy by an adenovirus encoding a novel PD-L1 inhibitory peptide inhibits the growth of colon carcinoma in immunocompetent mice

A novel peptide that interferes with the PD-1/PD-L1 immune checkpoint pathway, termed PD-L1 inhibitory peptide 3 (PD-L1ip3), was computationally designed, experimentally validated for its specific binding to PD-L1, and evaluated for its antitumor effects in cell culture and in a mouse colon carcinoma syngeneic murine model. In several cell culture studies, direct treatment with PD-L1ip3, but not a similar peptide with a scrambled sequence, substantially increased death of CT26 colon carcinoma cells when co-cultured with murine CD8⁺ T cells primed by CT26 cell antigens. In a syngeneic mouse tumor model, the growth of CT26 tumor cells transduced with the PD-L1ip3 gene by an adenovirus vector was significantly slower than that of un-transduced CT26 cells in immunocompetent mice. This tumor growth attenuation was further enhanced by the coadministration of the peptide form of PD-L1ip3 (10 mg/kg/day). The current study suggests that this peptide can stimulate host antitumor immunity via blockade of the PD-1/PD-L1 pathway, thereby increasing CD8⁺ T cell-induced death of colon carcinoma cells. The tumor site-specific inhibition of PD-L1 by an adenovirus carrying the PD-L1ip3 gene, together with direct peptide treatment, may be used as a local immune checkpoint blockade therapy to inhibit colon carcinoma growth.     

Intratumoral tumor necrosis factor induction in tumor-bearing mice by exogenous/endogenous tumor necrosis factor therapy as compared with systemic administration of various biologic response modifiers.

The relationship between the induction of tumor necrosis factor (TNF) (as an indicator of inflammatory reaction) in tumor tissues and its antitumor effect was investigated in tumor-bearing mice by using nine biologic response modifiers (BRMs) and by exogenous/endogenous TNF therapy following a previously reported protocol. Close correlation between the induction of TNF-rich inflammation in tumor tissues and the antitumor effect of BRM were observed. The results of this study suggest that the conditions necessary for exerting antitumor effects of biologic response modifiers may be the induction of TNF (50 to 200 U/g) at the tumor lesions at an early stage after BRM administration and maintenance of the detectable amount of TNF (approximately 10 U/g) for more than 6 hours. Tumor necrosis factor should also be induced in the liver and spleen so that its activity can be maintained in the tumor lesions.     

Aminopeptidase N/CD13 targeting fluorescent probes: synthesis and application to tumor cell imaging

A family of fluorescein-peptide conjugates (CNP1-3) for aminopeptidase N (APN/CD13) targeting fluorescent probes were designed and synthesized. Among the three conjugates, CNP1 bearing tumor-homing cyclic peptide CNGRC, could selectively label APN/CD13 over-expressing on the surface of tumor cells of HT-1080, as identified by means of fluorescent microscopic cell imaging. CNP1 was shown to be a promising fluorescent probe applicable to tumor-targeting molecular imaging.     

Targeting IFN-alpha to B cell lymphoma by a tumor-specific antibody elicits potent antitumor activities

IFN-alpha, a cytokine crucial for the innate immune response, also demonstrates antitumor activity. However, use of IFN-alpha as an anticancer drug is hampered by its short half-life and toxicity. One approach to improving IFN-alpha's therapeutic index is to increase its half-life and tumor localization by fusing it to a tumor-specific Ab. In the present study, we constructed a fusion protein consisting of anti-HER2/neu-IgG3 and IFN-alpha (anti-HER2/neu-IgG3-IFN-alpha) and investigated its effect on a murine B cell lymphoma, 38C13, expressing human HER2/neu. Anti-HER2/neu-IgG3-IFN-alpha exhibited potent inhibition of 38C13/HER2 tumor growth in vivo. Administration of three daily 1-microg doses of anti-HER2/neu-IgG3-IFN-alpha beginning 1 day after tumor challenge resulted in 88% of the mice remaining tumor free. Remarkably, anti-HER2/neu-IgG3-IFN-alpha demonstrated potent activity against established 38C13/HER2 tumors, with complete tumor remission observed in 38% of the mice treated with three daily doses of 5 microg of the fusion protein (p = 0.0001). Ab-mediated targeting of IFN-alpha induced growth arrest and apoptosis of lymphoma cells contributing to the antitumor effect. The fusion protein also had a longer in vivo half-life than rIFN-alpha. These results suggest that IFN-alpha Ab fusion proteins may be effective in the treatment of B cell lymphoma.     

IL-12 enhances the antitumor actions of trastuzumab via NK cell IFN-γ production

The antitumor effects of therapeutic mAbs may depend on immune effector cells that express FcRs for IgG. IL-12 is a cytokine that stimulates IFN-γ production from NK cells and T cells. We hypothesized that coadministration of IL-12 with a murine anti-HER2/neu mAb (4D5) would enhance the FcR-dependent immune mechanisms that contribute to its antitumor activity. Thrice-weekly therapy with IL-12 (1 μg) and 4D5 (1 mg/kg) significantly suppressed the growth of a murine colon adenocarcinoma that was engineered to express human HER2 (CT-26(HER2/neu)) in BALB/c mice compared with the result of therapy with IL-12, 4D5, or PBS alone. Combination therapy was associated with increased circulating levels of IFN-γ, monokine induced by IFN-γ, and RANTES. Experiments with IFN-γ-deficient mice demonstrated that this cytokine was necessary for the observed antitumor effects of therapy with IL-12 plus 4D5. Immune cell depletion experiments showed that NK cells (but not CD4(+) or CD8(+) T cells) mediated the antitumor effects of this treatment combination. Therapy of HER2/neu-positive tumors with trastuzumab plus IL-12 induced tumor necrosis but did not affect tumor proliferation, apoptosis, vascularity, or lymphocyte infiltration. In vitro experiments with CT-26(HER2/neu) tumor cells revealed that IFN-γ induced an intracellular signal but did not inhibit cellular proliferation or induce apoptosis. Taken together, these data suggest that tumor regression in response to trastuzumab plus IL-12 is mediated through NK cell IFN-γ production and provide a rationale for the coadministration of NK cell-activating cytokines with therapeutic mAbs.     

Induction of antigen-specific immune responses against malignant brain tumors by intramuscular injection of sindbis DNA encoding gp100 and IL-18

We constructed pSin-SV40-HDV-SV40pA, an improved Sindbis DNA expression vector, and evaluated the potential of this vector system for brain tumor therapy. We investigated whether immunizing mice with xenogeneic DNA encoding human gp100 and mouse IL-18 would enhance the antitumor responses. To study the immune mechanisms involved in tumor regression, we examined tumor growth in B16-gp100-implanted brain tumor models using T-cell subset-depleted and IFN-gamma-neutralized mice. Hugp100/mIL-18 vaccination was also investigated for its antitumor effects against the wild-type murine B16 tumor, which expresses the murine gp100 molecule. Genetic immunization using plasmid pSin 9001 DNA codelivery of human gp100 and mouse IL-18 resulted in enhanced protective and therapeutic effects on the malignant brain tumors. The antitumor and protective effects were mediated by both CD4(+)/CD8(+) T cells and IFN-gamma. Vaccination with hugp100/mIL-18 conferred a significant survival merit to wild-type B16 tumor-harboring mice. Immunogene therapy with the improved Sindbis virus vector expressing xenogeneic gp100 and syngeneic IL-18 may be an excellent approach for developing a new treatment protocol. Thus, the Sindbis DNA system may represent a novel approach for the treatment of malignant brain tumors.     

Combination therapy with a synthetic peptide of C-reactive protein and interleukin 2: augmented survival and eradication of pulmonary metastases

A synthetic peptide (RS-83277) derived from the structure of human C-reactive protein (CRP) was previously shown to have antitumor activity in three different murine tumor models when administered in multilamellar vesicles (MLV). The therapeutic effects were comparable to those seen with MLV-encapsulated native CRP. The present study evaluated the therapeutic and immunomodulatory effects of administering CRP peptide RS-83277 MLV simultaneously with low-dose recombinant interleukin-2 (IL-2) to C57B1/6 mice bearing established pulmonary metastases of fibrosarcoma T241. Results demonstrated that the capacity of RS-83277 MLV to inhibit tumor metastases and prolong survival was significantly augmented by combination with 10 000 U/day IL-2 i.p. Treated animals showed no evidence of toxicity. By immunohistochemistry, increased Thy 1.2⁺ cells were detectable in lungs of RS-83277 MLV/IL-2-treated animals compared to those receiving RS-83277 MLV alone. Circulating tumor necrosis factor (TNF) and interferon (IFN) were not detectable in animals receiving RS-83277 MLV alone, but TNF was significantly elevated in animals receiving IL-2. In the presence of combination therapy, however, circulating TNF was not detectable. Results suggest that the combination of synthetic CRP peptide RS-83277 MLV and low-dose IL-2 offers a therapeutic advantage over either agent alone.This work was supported in part by grant CA49950 from the National Cancer Institute (B.P.B.), and grant 43618 from the National American Cancer Society (S.D.D.)     

G3139 and other CpG-containing immunostimulatory phosphorothioate oligodeoxynucleotides are potent suppressors of the growth of human tumor xenografts in nude mice

Several phosphorothioate antisense oligodeoxynucleotides (ODN) are developed to target factors potentially involved in tumor growth and apoptosis suppression. Among them, the 18-mer G3139 (Oblimersen), which targets Bcl-2, is currently being tested in phase II and phase III clinical trials for various tumors in combination with chemotherapy. On the other hand, ODNs containing CpG dinucleotides (CpG-ODN) within specific-sequence contexts (CpG motifs) have been shown to activate rodent or primate immune cells via toll-like receptor 9 (TLR9) and have demonstrated remarkable T cell-dependent antitumor efficacy in a series of murine tumor models. However, immune cell activation by CpG-ODN is largely diminished upon C-5 methylation at CpG cytosine. As G3139 contains CpG motifs, we questioned whether the antitumor effects seen in human tumor xenografts might be abrogated by cytosine C-5 methylation of G3139, which retained the ability of G3139 to suppress Bcl-2 expression in tissue culture, or by similar derivatization of other phosphorothioate ODNs developed for the immune activation of rodent or human cells. The in vivo antitumor efficacy of the immunostimulatory H1826 and H2006 ODNs was compared with that of G3139. Bcl-2 suppression achieved by G3139 purportedly sensitizes tumor cells toward cytotoxic agents, and some of the experiments employed combinations of ODN with such drugs as cisplatin or etoposide. H1826, H2006, and G3139 all produced similar, striking, growth inhibitory effects on either H69 SCLC, A2780 ovarian carcinoma, or A549 lung adenocarcinoma human tumor xenografts at doses of 0.3 mg/kg and 1 mg/kg (H1826, H2006) or 12 mg/kg (G3139) per day. In contrast, the H2006-mC (1 mg/kg) or G3139-mC (12 mg/kg) derivatives demonstrated no significant antitumor effects. The combination of G3139 (12 mg/kg) with cisplatin produced some additive antitumor efficacy, which was not seen in combinations of G3139-mC (12 mg/kg) or H1826 (1 mg/kg) with cisplatin. G3139, at a dose of 12 mg/kg, alone induced extensive enlargement of the spleen. Immunostimulation was evaluated in vitro by flow cytometric measurements of the CD80 and CD86 activation markers found on CD19+ murine splenocytes. The CpG-ODN producing strong antitumor effects in vivo also induced these activation markers in vitro, in contrast to the in vivo inactive G3139-mC. Our data indicate a significant contribution of the immunostimulatory properties of CpG-ODN (including G3139) to the antitumor effects observed in nude mouse xenograft models. This is in contrast to previous data presented by other authors indicating that the activity of G3139 in human tumor xenografts was Bcl-2 specific. Furthermore, as nude mice are devoid of T cells, a T cell-mediated immune response apparently is not required for the potent antitumor responses observed here; innate immune responses are sufficient.     

TNF-mediated toxicity after massive induction of specific CD8+ T cells following immunization of mice with a tumor-specific peptide

We immunized mice with antigenic peptide P815E, which is presented by H-2K(d) and recognized by tumor-specific CTL raised against P815 tumor cells. This peptide is encoded by the ubiquitously expressed gene MsrA and carries a mutated residue conferring tumor specificity. Unexpectedly, we observed a severe toxicity occurring in the early hours after the third injection, resulting in the death of most mice within 24 h. The toxic syndrome was reminiscent of TNF-induced shock, and the sera of ill mice contained high levels of TNF. Toxicity was prevented by injection of neutralizing anti-TNF Abs, confirming the involvement of TNF. Depletion of CD8+ T cells could also prevent toxicity, and ex vivo experiments confirmed that CD8+ lymphocytes were the major cellular source of TNF in immunized mice. Tetramer analysis of the lymphocytes of immunized mice indicated a massive expansion of P815E-specific T cells, up to >60% of circulating CD8+ lymphocytes. A similar toxicity was observed after massive expansion of specific CD8+ T cells following immunization with another P815 peptide, which is encoded by gene P1A and was injected in a form covalently linked to an immunostimulatory peptide derived from IL-1. We conclude that the toxicity is caused by specific CD8+ lymphocytes, which are extensively amplified by peptide immunization in a QS21-based adjuvant and produce toxic levels of TNF upon further stimulation with the peptide. Our results suggest that immunotherapy trials involving new peptides should be pursued with caution and should include a careful monitoring of the T cell response.