跨膜型肿瘤坏死因子α稳定表达细胞株的构建

目的：在体内,相对分子质量为26×103的跨膜型肿瘤坏死因子α（mTNF-α）可被TNF转化酶（TACE）酶解成相对分子质量为17×103的游离型TNF-α（sTNF-α）。为评价新研发的TNF-α全人源单克隆抗体与mTNF-α的结合能力,以及该单抗的ADCC作用,须构建只能稳定表达mTNF-α而不受TACE影响的Sp2/0细胞株。方法：PCR扩增缺失了TACE酶切位点的人TNF-α野生型基因序列,构建pcDNA3.1（＋）-dTNF-α重组质粒并测序鉴定;重组质粒转染Sp2/0细胞,构建只能稳定表达mTNF-α的细胞株,用流式细胞仪检测其表达情况。结果：pcDNA3.1（＋）-dTNF-α重组质粒测序结果与设计缺失基因的碱基序列完全一致,重组质粒构建成功;转染的阳性细胞株经流式细胞仪检测有表达,其中Clone11、15、16的表达量最高。结论：跨膜型TNF-α稳定表达细胞株构建成功,可用于TNF-α全人源单克隆抗体体外药效的评价。     

Transmembrane TNF-dependent uptake of anti-TNF antibodies

TNF-α (TNF), a pro-inflammatory cytokine is synthesized as a 26 kDa protein, anchors in the plasma membrane as transmembrane TNF (TmTNF), and is subjected to proteolysis by the TNF-α converting enzyme (TACE) to release the 15 kDa form of soluble TNF (sTNF). TmTNF and sTNF interact with 2 distinct receptors, TNF-R1 (p55) and TNF-R2 (p75), to mediate the multiple biologic effects of TNF described to date. Several anti-TNF biologics that bind to both forms of TNF and block their interactions with the TNF receptors are now approved for the treatment of a variety of immune-mediated diseases. Several reports suggest that binding of anti-TNFs to TmTNF delivers an outside-to-inside ‘reverse’ signal that may also contribute to the efficacy of anti-TNFs. Some patients, however, develop anti-TNF drug antibody responses (ADA or immunogenicity). Here, we demonstrate biochemically that TmTNF is transiently expressed on the surface of lipopolysaccharide-stimulated primary human monocytes, macrophages, and monocyte-derived dendritic cells (DCs) and expression of TmTNF on the cell surface is enhanced following treatment of cells with TAPI-2, a TACE inhibitor. Importantly, binding of anti-TNFs to TmTNF on DCs results in rapid internalization of the anti-TNF/TmTNF complex first into early endosomes and then lysosomes. The internalized anti-TNF is processed and anti-TNF peptides can be eluted from the surface of DCs. Finally, tetanus toxin peptides fused to anti-TNFs are presented by DCs to initiate T cell recall proliferation response. Collectively, these observations may provide new insights into understanding the biology of TmTNF, mode of action of anti-TNFs, biology of ADA response to anti-TNFs, and may help with the design of the next generation of anti-TNFs.     

The role of Tumor Necrosis Factor-alpha (TNF-α) in the pathogenesis of systemic lupus erythematosus

The Tumor Necrosis Factor-alpha (TNF-α) is a pleiotropic cytokine that produces different stimuli in various physiological and pathological conditions. TNF-α contributes importantly to the development of T cells, B cells, and dendritic cells. However, TNF-α is also a potent inflammatory mediator and apoptosis inducer. The significance of the TNF-α involvement in the pathogenesis of systemic lupus erythematosus (SLE) remains controversial. From the genetic standpoint, a number of studies suggest that the TNF-α gene polymorphism is involved in the susceptibility of SLE. Moreover, there is a close association between the TNF-α gene expression and clinical manifestations. In addition, the increased serum level of TNF-α is observed in SLE patients and associated with disease activity and certain systemic manifestations. Treatment with anti-TNF agents is, however, controversial in SLE since induction of antinuclear antibodies, anti-dsDNA, anticardiolipin antibodies, and cases of drug-induced lupus have been observed in rheumatoid arthritis patients. In this context, this study reviewed the importance of TNF-α in the pathogenesis of SLE.     

Differences in binding and effector functions between classes of TNF antagonists

There are currently two Food and Drug Administration-approved classes of biologic agents that target tumor necrosis factor-α (TNF-α): anti-TNF monoclonal antibodies (mAbs) (adalimumab and infliximab), and soluble TNF receptors (etanercept). This study examined the ability of the TNF antagonists to: (1) bind various polymorphic variants of cell surface-expressed Fc receptors (FcγRs) and the complement component C1q, and (2) mediate Ab-dependent cellular cytotoxicity (ADCC) and complement-mediated cytotoxicity (CDC) killing of cells expressing membrane-bound TNF (mTNF) in vitro. Both mAbs and the soluble TNF receptor demonstrated low-level binding to the activating receptors FcγRI, FcγRIIa, and FcγRIIIa, and the inhibitory receptor FcγRIIb, in the absence of exogenous TNF. However, upon addition of TNF, the mAbs, but not etanercept, showed significantly increased binding, in particular to the FcγRII and FcγRIII receptors. Infliximab and adalimumab induced ADCC much more potently than etanercept. In the presence of TNF, both mAbs bound C1q in in vitro assays, but etanercept did not bind C1q under any conditions. Infliximab and adalimumab also induced CDC in cells expressing mTNF more potently than etanercept. Differences in the ability to bind ligand and mediate cell death may account for the differences in efficacy and safety of TNF antagonists.     

Development and Optimization of Therapeutic Analogues of Anti-TNFα Antibody Infliximab

Previously, we have reported the crystal structures of Fab fragment of Infliximab in complex with TNFα. The structurally identified epitope on TNFα revealed the mechanism of TNFα inhibition by partially overlapping with the TNFα-receptor interface and the possibility to optimize the binding affinity. In this study, we launched a screen of a phage display library to isolate novel anti-TNFα antibodies based on the infliximab epitope. To develop novel anti-TNFα antibodies, structural analysis, the phage display antibody isolation, step by step antibody optimization, CDR residues random mutagenesis, and binding affinity characterization were performed. One of the novel antibodies generated on the backbone of infliximab, Inf3D6, has the superior binding affinity to TNFα, thus, demonstrating the potential for structure guided optimization for improvement of existing antibody-based therapeutics.     

Intact-cell-based surface plasmon resonance measurements for ligand affinity evaluation of a membrane receptor

Toward future applications to the discovery of drugs against membrane receptors on pathological cells, an intact-cell-based surface plasmon resonance (SPR) methodology has been developed. The injection of a suspension of epidermal carcinoma A431 cells (5×10(7)cells/ml), as an analyte, generated clear SPR responses to epidermal growth factor (EGF) immobilized on the sensor chip. Because the responses were competitively reduced by the free ligand EGF, added to the analyte cell suspension, they certainly reflect the specific interaction of the immobilized EGF with the extracellular region of its receptor, which is highly expressed on the surface of the A431 cells.     

TNF-α inhibitor reverse the effects of human umbilical cord-derived stem cells on experimental arthritis by increasing immunosuppression

To demonstrate the therapeutic potential for cartilage repair of mesenchymal stem cells derived from human umbilical cord (HUCSC), we studied the clinical and histopathological effects of intra-articular injection of HUCSCs in a collagen-induced arthritis (CIA) model. In our study, intra-articular injection of HUCSCs had no benefit in CIA mice; and accelerated the progression of arthritis in the presence of TNF-α. To determine the role of TNF-α, we injected the combination with HUCSCs and TNF inhibitor, showed reduced the disease signs in CIA mice. On exposure of TNF-α, stem cells significantly decreased the expression of CD90, HLA-G, and the levels of IL-10 in vitro and in vivo. Our data showed that TNF-α blocks the immunosuppressive effects of HUCSCs and that inhibition of TNF-α decreases cartilage destruction by suppressing the immunogenicity of HUCSCs. Injecting both a TNF inhibitor and HUCSCs may be a potential effective therapy for ameliorating the disease.     

Autocrine stimulation of osteoblast activity by Wnt5a in response to TNF-α in human mesenchymal stem cells

Although anti-tumor necrosis factor (TNF)-α treatments efficiently block inflammation in ankylosing spondylitis (AS), they are inefficient to prevent excessive bone formation. In AS, ossification seems more prone to develop in sites where inflammation has resolved following anti-TNF therapy, suggesting that TNF-α indirectly stimulates ossification. In this context, our objectives were to determine and compare the involvement of Wnt proteins, which are potent growth factors of bone formation, in the effects of TNF-α on osteoblast function. In human mesenchymal stem cells (MSCs), TNF-α significantly increased the levels of Wnt10b and Wnt5a. Associated with this effect, TNF-α stimulated tissue-non specific alkaline phosphatase (TNAP) and mineralization. This effect was mimicked by activation of the canonical β-catenin pathway with either anti-Dkk1 antibodies, lithium chloride (LiCl) or SB216763. TNF-α reduced, and activation of β-catenin had little effect on expression of osteocalcin, a late marker of osteoblast differentiation. Surprisingly, TNF-α failed to stabilize β-catenin and Dkk1 did not inhibit TNF-α effects. In fact, Dkk1 expression was also enhanced in response to TNF-α, perhaps explaining why canonical signaling by Wnt10b was not activated by TNF-α. However, we found that Wnt5a also stimulated TNAP in MSCs cultured in osteogenic conditions, and increased the levels of inflammatory markers such as COX-2. Interestingly, treatment with anti-Wnt5a antibodies reduced endogenous TNAP expression and activity. Collectively, these data suggest that increased levels of Dkk1 may blunt the autocrine effects of Wnt10b, but not that of Wnt5a, acting through non-canonical signaling. Thus, Wnt5a may be potentially involved in the effects of inflammation on bone formation.     

Genetically recessive mutant of human monocytic leukemia U937 resistant to tumor necrosis factor-α-induced apoptosis

Tumor necrosis factor-α (TNF-α) is a cytokine that induces apoptosis in various cell systems by binding to the TNF receptor (TNFR). To study TNF-α-induced apoptosis, we isolated and characterized a novel TNF-α-resistant variant, U937/TNF clone UA, from human monocytic leukemia U937 cells. The UA cells resist apoptosis induced by TNF-α and anti-Fas antibody but not by anticancer drugs, such as VP-16 and Ara-C. Somatic cell hybridization between U937 and UA showed that apoptosis resistance to TNF-α in UA was genetically recessive. The hybridization analysis also showed that UA and another recessive mutant clone, UC, belong to different complementation groups in TNF-α-induced apoptosis signaling. In UA cells, TNF-α-induced disruption of mitochondrial membrane potential and CPP32 activation were abrogated. Expression of TNFR, Fas, and Bcl-2 family proteins was not changed in UA cells. These results suggest that the apoptosis resistant UA cells could have a functional defect in apoptosis signaling from the TNFR to mitochondria and interleukin-1β converting enzyme (ICE) family protease activation. UA cells could be used to study signaling linkage between cell death-inducing receptor and mitochondria. J. Cell. Physiol. 174:179–185, 1998. © 1998 Wiley-Liss, Inc.     

Tumor necrosis factor-α receptor type 1, not type 2, mediates its acute responses in the kidney

Acute administration of tumor necrosis factor-α (TNF-α) resulted in decreases in renal blood flow (RBF) and glomerular filtration rate (GFR) but induced diuretic and natriuretic responses in mice. To define the receptor subtypes involved in these renal responses, experiments were conducted to assess the responses to human recombinant TNF-α (0.3 ng·min(-1)·g body wt(-1) iv infusion for 75 min) in gene knockout (KO) mice for TNF-α receptor type 1 (TNFαR1 KO, n = 5) or type 2 (TNFαR2 KO, n = 6), and the results were compared with those obtained in corresponding wild-type [WT (C57BL/6), n = 6] mice. Basal levels of RBF (PAH clearance) and GFR (inulin clearance) were similar in TNFαR1 KO, but were lower in TNFαR2 KO, than WT mice. TNF-α infusion in WT mice decreased RBF and GFR but caused a natriuretic response, as reported previously. In TNFαR1 KO mice, TNF-α infusion failed to cause such vasoconstrictor or natriuretic responses; rather, there was an increase in RBF and a decrease in renal vascular resistance. Similar responses were also observed with infusion of murine recombinant TNF-α in TNFαR1 KO mice (n = 5). However, TNF-α infusion in TNFαR2 KO mice caused changes in renal parameters qualitatively similar to those observed in WT mice. Immunohistochemical analysis in kidney slices from WT mice demonstrated that while both receptor types were generally located in the renal vascular and tubular cells, only TNFαR1 was located in vascular smooth muscle cells. There was an increase in TNFαR1 immunoreactivity in TNFαR2 KO mice, and vice versa, compared with WT mice. Collectively, these functional and immunohistological findings in the present study demonstrate that the activation of TNFαR1, not TNFαR2, is mainly involved in mediating the acute renal vasoconstrictor and natriuretic actions of TNF-α.     

TNF receptor II fusion protein with tandemly repeated Fc domains

The extracellular domain of tumour necrosis factor (TNF) receptor II fused with the human IgG1 Fc region (TNFRII-Fc), as well as antibodies against TNF, has been used to treat rheumatoid arthritis. However, TNFRII-Fc is less effective than these antibodies in terms of antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC) against cells bearing TNF on the cell surface. We hypothesized that these activities could be increased by fusing TNFRII with tandemly repeated Fc (TNFRII-Fc-Fc). The affinities of TNFRII-Fc-Fc for soluble TNF-α and transmembrane TNF-α and the TNF-α cytotoxicity-inhibitory activity were as potent as those of TNFRII-Fc. TNFRII-Fc-Fc showed much higher binding avidity for Fcγ receptors than TNFRII-Fc and was more potent in terms of both ADCC and CDC against cells expressing transmembrane TNF-α. TNFRII-Fc-Fc of 80 kDa, as well as TNFRII-Fc-Fc of 200 kDa, was detected. TNFRII-Fc-Fc (80 kDa) was as potent as TNFRII-Fc in terms of both ADCC and CDC. These results suggest that Fc multimerization of receptor-Fc fusion proteins can augment effector functions such as ADCC and CDC, and thereby have the potential to provide a superior therapeutic effect. This may be the case not only for TNFRII-Fc but also for other receptor-Fc fusion proteins.     

可溶性肿瘤坏死因子受体II-脂联素球部融合蛋白的真核表达及生物学活性检测

本研究设计和构建了一种人肿瘤坏死因子受体II胞外区与人脂联素球部的融合基因sTNFRII-gAD,且相应的融合蛋白在哺乳动物细胞BHK-21S的无血清培养体系中实现了表达,并对该融合蛋白进行了初步鉴定。首先,用RT-PCR方法从人的外周血淋巴细胞总RNA中扩增人肿瘤坏死因子II型受体胞外区基因片段,与脂联素球部基因片段融合,克隆至pAAV2neo表达载体中,构建成pAAV2neo-sTNFRII-gAD。随后,用pAAV2neo-sTNFRII-gAD转染BHK-21S细胞获得G418抗性细胞BHK-21S/pAAV2neo-sTNFRII-gAD;然后,将原来含有血清的培养液换成无血清的化学成分限定的培养液,细胞从贴壁培养方式转换成悬浮培养方式;最后,收集BHK-21S/pAAV2neo-sTNFRII-gAD无血清悬浮培养24h后的培养上清,进行sTNFRII-gAD融合蛋白的鉴定分析。酶切鉴定和测序结果显示,所构建的pAAV2neo-sTNFRII-gAD质粒结构正确,sTNFRII-gAD序列与预期一致;分别用抗人肿瘤坏死因子受体II和抗人脂联素球部的单克隆抗体检测pAAV2neo-sTNFRII-gAD瞬时转染的BHK-21S细胞,免疫荧光呈现阳性;免疫印迹分析在pAAV2neo-sTNFRII-gAD稳定转染的BHK-21S细胞上清中检测到sTNFRII-gAD融合蛋白的表达,并以单体、三聚体和三聚体以上的多聚体形式存在。活性测定结果表明,sTNFRII-gAD融合蛋白具有显著抑制TNFα杀伤L929细胞的活性。因此,本研究为下一步大量制备sTNFRII-gAD融合蛋白用于体内外功能研究提供了良好基础。     

Inhibition of Varicella-Zoster Virus Glycoprotein Expression by Peripheral Blood Mononuclear Cells

The effect of peripheral blood mononuclear cells (PBMC) on expression of varicella-zoster virus (VZV) glycoproteins (Gps) was analyzed by flow cytometry. PBMC from VZV seropositive and seronegative donors and supernatant of PBMC co-cultured with VZV-infected human embryonic fibroblasts reduced VZV Gp expression. Neutralization of supernatant fluid with mixture of anti-interferons (IFN)-α, -β, -γ, and tumor necrosis factor (TNF)-α partially reduced inhibitory activity of supernatant on VZV Gp expression. Deletion of natural killer (NK) cells and adherent cells from PBMC reduced inhibitory activity of PBMC on VZV Gp expression. These results suggest that IFN-α, -β, -γ, TNF-α and other soluble factors released from NK cells and monocytes by co-cultivation with VZV-infected fibroblasts inhibit VZV Gp expression.     

Differential risk of tuberculosis reactivation among anti-TNF therapies is due to drug binding kinetics and permeability

Increased rates of tuberculosis (TB) reactivation have been reported in humans treated with TNF-α (TNF)-neutralizing drugs, and higher rates are observed with anti-TNF Abs (e.g., infliximab) as compared with TNF receptor fusion protein (etanercept). Mechanisms driving differential reactivation rates and differences in drug action are not known. We use a computational model of a TB granuloma formation that includes TNF/TNF receptor dynamics to elucidate these mechanisms. Our analyses yield three important insights. First, drug binding to membrane-bound TNF critically impairs granuloma function. Second, a higher risk of reactivation induced from Ab-type treatments is primarily due to differences in TNF/drug binding kinetics and permeability. Apoptotic and cytolytic activities of Abs and pharmacokinetic fluctuations in blood concentration of drug are not essential to inducing TB reactivation. Third, we predict specific host factors that, if augmented, would improve granuloma function during anti-TNF therapy. Our findings have implications for the development of safer anti-TNF drugs to treat inflammatory diseases.     

Tumor necrosis factor alpha-induced inflammation is increased but apoptosis is inhibited by common food additive carrageenan

Tumor necrosis factor (TNF)-α, a homotrimeric, pleiotropic cytokine, is secreted in response to inflammatory stimuli in diseases such as rheumatoid arthritis and inflammatory bowel disease. TNF-α mediates both apoptosis and inflammation, stimulating an inflammatory cascade through the non-canonical pathway of NF-κB activation, leading to increased nuclear RelB and p52. In contrast, the common food additive carrageenan (CGN) stimulates inflammation through both the canonical and non-canonical pathways of NF-κB activation and utilizes the adaptor molecule BCL10 (B-cell leukemia/lymphoma 10). In a series of experiments, colonic epithelial cells and mouse embryonic fibroblasts were treated with TNF-α and carrageenan in order to simulate the possible effects of exposure to dietary CGN in the setting of a TNF-α-mediated inflammatory disease process. A marked increase in secretion of IL-8 occurred, attributable to synergistic effects on phosphorylated NF-κB-inducing kinase (NIK) in the non-canonical pathway. TNF-α induced the ubiquitination of TRAF2 (TNF receptor-associated factor 2), which interacts with NIK, and CGN induced phosphorylation of BCL10, leading to increased NIK phosphorylation. These results suggest that TNF-α and CGN in combination act to increase NIK phosphorylation, thereby increasing activation of the non-canonical pathway of NF-κB activation. In contrast, the apoptotic effects of TNF-α, including activation of caspase-8 and PARP-1 (poly(ADP-ribose) polymerase 1) fragmentation, were markedly reduced in the presence of CGN, and CGN caused reduced expression of Fas. These findings demonstrate that exposure to CGN drives TNF-α-stimulated cells toward inflammation rather than toward apoptotic cell death and suggest that CGN exposure may compromise the effectiveness of anti-TNF-α therapy.     

The combined effects of anti-TNFα antibody and IL-1 receptor antagonist in human rheumatoid arthritis synovial membrane

TNFα and IL-1 are the pivotal cytokines involved in rheumatoid arthritis (RA). More recently, the biological therapy targeting TNFα or IL-1 has been impressively effective for many RA patients, however, it remains insufficient in some patients. In the present study, we examined the combined effects of two agents against TNFα and IL-1 in human RA synovial membrane. Synovial explants (an ex vivo model) and synovial fibroblasts (an in vitro model) were prepared from 11 RA patients, and then anti-TNFα antibody (Anti-TNFα) and IL-1 receptor antagonist (IL-1Ra), either alone or in combination, were added to the synovial explants and fibroblasts. IL-6 and MMP-3 production were measured after incubation. As a result, their production significantly decreased by the combination of agents compared with the control group in both the synovial explants and fibroblasts. The efficacy of this combination was also observed for IL-6 production compared with each agent alone in the synovial explants, and for IL-6 and MMP-3 production compared with each agent alone in the synovial fibroblasts. Therefore, the combination of Anti-TNFα and IL-1Ra appears more beneficial in synovial membrane, particularly when compared with a single agent alone.     

粘着斑激酶在TNF-α作用下影响SMMC-7721细胞PKB的表达水平

 探讨在肿瘤坏死因子α(TNF-α)作用下 ,粘着斑激酶 (focal adhesion kinase,FAK)对蛋白激酶 B(PKB)蛋白水平的影响 .利用构建、转染 FAK反义质粒来特异性降低 SMMC- 772 1细胞的FAK含量 ,及用 Western杂交的方法来检测 PKB的蛋白含量 .文献报道 TNF- α能够激活磷脂酰肌醇 3-激酶 (PI3K)而使 PKB发生磷酸化 .但是至于 TNF-α对 PKB蛋白水平的影响目前并无报道 .研究发现 ,当用 wortmannin特异性抑制 PI3K活性后可以显著降低 PKB的蛋白含量 .提示PI3K对维持 PKB的基础蛋白水平是必需的 .但是 TNF- α本身对 PKB的蛋白水平无明显影响 .而当用不同浓度的 TNF- α和 wortmannin处理 SMMC- 772 1细胞时 ,发现 PKB的蛋白含量随着TNF-α浓度的增加而降低 .提示 TNF-α可能除了通过 PI3K外 ,还可能通过另一条途径来下调PKB的表达 .而当用 FAK反义质粒转染 SMMC- 772 1细胞后 (FAK下降了 60 % ) ,发现在当用不同浓度的 TNF- α处理的情况下 ,FAK反义质粒转染株 AS- 772 1细胞的 PKB含量降低为对照的70 % ;而在用 TNF-α和 wortmannin处理的情况下 ,下降为对照的 40 %～ 60 % .TNF-α能够通过PI3K及另一未知途径来影响 PKB的蛋白水平 .而 FAK在 TNF- α作用下能够不通过 PI3K来影响PKB的蛋白水平 .     

Interpreting interest in interferon-α

The aim of the present study was to investigate the expression of Fas in periarticular tenocytes of patients with osteoarthritis (OA) and to study their susceptibility to Fas ligand-mediated apoptosis. Tendon samples were obtained from the quadriceps femoris muscle of patients with knee OA and used for histological evaluation, for immunohistochemical detection of Fas, and to establish tenocyte cultures. The expression of Fas mRNA was determined by quantitative PCR. Levels of soluble Fas and soluble tumour necrosis factor (TNF) receptor I were measured using ELISA. Apoptosis was induced with recombinant human Fas ligand and measured by a histone fragmentation assay and flow cytometry. The effects of TNF-α were studied by stimulation with TNF-α alone or 24 hours before the induction of apoptosis. Tendon samples from non-OA patients were used as controls. Histological evaluation revealed degenerative changes in the tendons of all OA patients but not in the controls. Fas was detected by immunohistochemistry in all specimens, but quantitative PCR revealed significantly higher levels of Fas mRNA in OA tenocytes. In contrast, lower levels of soluble Fas were found in OA tenocytes by ELISA. OA tenocytes were significantly more susceptible to Fas ligand induced apoptosis than were control cells. TNF-α reduced the Fas ligand induced apoptosis in OA tenocytes but had no effects on control tenocytes. These data suggest that knee OA is associated with higher susceptibility of periarticular tenocytes to Fas ligand induced apoptosis because of higher expression of Fas but lower levels of apoptosis-inhibiting soluble Fas. These changes may contribute to decreased cellularity in degenerative tendons and promote their rupturing. The antiapoptotic effects of TNF-α in OA tenocytes most likely reflect regenerative attempts and must be taken into account when anti-TNF strategies are considered for OA.     

Role of tumour necrosis factor in the tumour-necrotizing activity of agents with diverging toxicity

Summary We investigated the ability of various tumournecrotizing agents with diverging toxicity to induce tumour necrosis factor (TNF) and cytostatic activity inPropionibacterium-acnes-primed Swiss and tumour-bearing BALB/c mice, and the capacity of anti-TNF antibodies to inhibit induction of tumour necrosis by the agents. Lipid A and especially its combination with muramyl dipeptide induced high TNF levels in Swiss mice, as measured in the serum. Lower levels were induced by detoxified lipid A and the nontoxic dsRNA, polyadenylic polyuridylic acid, either alone or combined with muramyl dipeptide. The toxic agents also appeared the strongest inducers of mediators with cytostatic activity against cultured endothelial cells and MethA tumour cells. Anti-TNF antibodies partially reduced the cytostatic activity of the sera against MethA cells. Tumour-bearing BALB/c mice produced only low levels of TNF and cytostatic factors in response to all agents. Recombinant mouse TNF hardly reduced the DNA synthesis of MethA cells, unless normal mouse serum was added. Serum fromP.-acnes-treated Swiss mice and tumour-bearing BALB/c mice, that were inhibitory on their own, failed to potentiate the action of TNF. Serum from Swiss mice treated with toxic, but not detoxified, lipid A caused extensive tumour necrosis upon injection into MethA-bearing BALB/c mice. This activity was completely abolished by pre-incubation of the serum with anti-TNF. The tumour-necrotizing activity of the agents could be partially reduced by prior injection of these antibodies. Results show that the capacity of the agents to induce TNF and cytostatic activity is not related to their antitumour potential. Although TNF is likely to be a crucial mediator of the tumour-necrotizing action of the toxic as well as the nontoxic agents, it is probably not the sole mediator. Data also indicate that induction of tumour necrosis does not require induction of high and, thus toxic, TNF levels in the serum.