Allergic lung inflammation is mediated by soluble tumor necrosis factor (TNF) and attenuated by dominant-negative TNF biologics

Tumor Necrosis Factor (TNF) is a pleiotropic cytokine consisting of soluble and transmembrane forms, with distinct roles in inflammation and immunity. TNF is an important factor in allergic airway inflammation. However, the disparate functions of soluble (sol) and transmembrane (tm) TNF in lung pathology are not well understood. Our aim was to assess the activities of solTNF and tmTNF in murine models of allergic airway disease, and to evaluate the efficacy of solTNF-selective inhibition. We used ovalbumin sensitization and challenge of TNF knockout, tmTNF knockin, and wild-type C57BL/6 mice to distinguish differences in airway inflammation and hyperreactivity mediated by solTNF and tmTNF. Functions of solTNF and tmTNF in hyperresponsive, wild-type Balb/c mice were assessed by comparing dominant-negative anti-TNF biologics, which antagonize solTNF yet spare tmTNF, to etanercept, a nonselective inhibitor of both TNF forms. Responses in transgenic C57BL/6 mice demonstrated that solTNF, and not tmTNF, is necessary to drive airway inflammation. In Balb/c mice, dominant-negative TNF biologics administered during immunization decreased the recruitment of eosinophils and lymphocytes into the bronchoalveolar space and lung parenchyma, reduced specific serum IgE, goblet-cell hyperplasia, and eosinophilic inflammation, and suppressed methacholine-induced airway hyperreactivity. Concentrations of IL-5, CCL5/RANTES, CCL11/eotaxin, and CCL17/TARC were also reduced in bronchoalveolar lavage. Dominant-negative TNFs reduced lung eosinophilia, even when given only during antigen challenge. The selective inhibition of soluble TNF suppresses inflammation, hyperreactivity, and remodeling in transgenic and wild-type murine models of allergic airway disease, and may offer safety advantages in therapies that preserve the immunoprotective functions of transmembrane TNF.

Soluble TNF, but not membrane TNF, is critical in LPS-induced hepatitis

BACKGROUND & AIMS

: Bacillus Calmette-Guérin (BCG) infection causes hepatic injury following granuloma formation and secretion of cytokines which renders mice highly sensitive to endotoxin-mediated hepatotoxicity. Tumor necrosis factor (TNF) is required for granuloma formation and is one of the most important cytokines in liver injury. TNF inhibitors are effective therapies for inflammatory diseases. However, clinical use of non-selective TNF inhibitors is associated with an increased risk of infections. This work investigates the differential roles of soluble TNF (solTNF) and membrane TNF (memTNF) in BCG infection, BCG/LPS- and D-GALN/LPS-induced liver injury.

METHODS

We have used both genetic and pharmacologic approaches and analyzed liver injury, TLR4, cytokine and iNOS activation induced by BCG, BCG/LPS and D-GALN/LPS.

RESULTS

BCG infection-induced liver injury is seen in wild-type mice but not in TNF(-/-), memTNF knock-in (KI), and sTNFR1-Fc transgenic mice. Severity of BCG-induced liver injury is correlated with BCG-granuloma number and hepatic expression of TLR4 and iNOS. In addition, protection from liver damage caused by BCG/LPS or D-GALN/LPS administration was observed in TNF(-/-), memTNF KI and sTNFR1-Fc transgenic mice. To extend the genetic findings, we then evaluated whether selective pharmacological inhibition of solTNF by dominant-negative (DN)-TNF neutralization and non-selective inhibition of solTNF and memTNF by anti-TNF antibodies and etanercept (TNFR2-IgG1) can protect the mice from liver injury. Both selective and non-selective inhibition of solTNF protected mice from BCG/LPS and D-GALN/LPS-induced liver damage.

CONCLUSIONS

These data suggest that memTNF is not mediating liver injury and that selective inhibition of solTNF sparing memTNF may represent a new therapeutic strategy to treat immune-mediated inflammatory liver diseases.

Dominant-negative tumor necrosis factor protects from Mycobacterium bovis Bacillus Calmette Guérin (BCG) and endotoxin-induced liver injury without compromising host immunity to BCG and Mycobacterium tuberculosis

BACKGROUND

Tumor necrosis factor (TNF) is associated with the development of inflammatory pathologies. Antibodies and soluble TNF (solTNF) receptors that neutralize excessive TNF are effective therapies for inflammatory and autoimmune diseases. However, clinical use of TNF inhibitors is associated with an increased risk of infections.

METHODS

A novel dominant-negative (DN) strategy of selective TNF neutralization, consisting of blocking solTNF while sparing transmembrane TNF (tmTNF), was tested in mouse models of mycobacterial infection and acute liver inflammation. XENP1595, a DN-TNF biologic, was compared with etanercept, a TNF receptor 2 (TNFR2)-IgG1 Fc fusion protein that inhibits murine solTNF and tmTNF.

RESULTS

XENP1595 protected mice from acute liver inflammation induced by endotoxin challenge in Mycobacterium bovis bacillus Calmette-Guérin (BCG)-infected mice, but, in contrast to etanercept, it did not compromise host immunity to acute M. bovis BCG and Mycobacterium tuberculosis infections in terms of bacterial burden, granuloma formation, and innate immune responses.

CONCLUSIONS

A selective inhibitor of solTNF efficiently protected mice from acute liver inflammation yet maintained immunity to mycobacterial infections. In contrast, nonselective inhibition of solTNF and tmTNF suppressed immunity to M. bovis BCG and M. tuberculosis. Therefore, selective inhibition of solTNF by DN-TNF biologics may represent a new therapeutic strategy for the treatment of inflammatory diseases without compromising host immunity.

Inhibition of B cell receptor-mediated activation of primary human B cells by coengagement of CD19 and FcgammaRIIb with Fc-engineered antibodies

The humoral immune response requires antigen-specific B cell activation and subsequent terminal differentiation into plasma cells. Engagement of B cell antigen receptor (BCR) on mature B cells activates an intracellular signaling cascade, including calcium mobilization, which leads to cell proliferation and differentiation. Coengagement by immune complex of BCR with the inhibitory Fc receptor FcgammaRIIb, the only IgG receptor expressed on B cells, inhibits B cell activation signals through a negative feedback loop. We now describe antibodies that mimic the inhibitory effects of immune complex by high-affinity coengagement of FcgammaRIIb and the BCR coreceptor complex on human B cells. We engineered the Fc domain of an anti-CD19 antibody to generate variants with up to approximately 430-fold greater affinity to FcgammaRIIb. Relative to native IgG1, the FcgammaRIIb binding-enhanced (IIbE) variants strongly inhibited BCR-induced calcium mobilization and viability in primary human B cells. Inhibitory effects involved phosphorylation of SH2-containing inositol polyphosphate 5-phosphatase (SHIP), which is known to be involved in FcgammaRIIb-induced negative feedback of B cell activation by immune complex. Coengagement of BCR and FcgammaRIIb by IIbE variants also overcame the anti-apoptotic effects of BCR activation. The use of a single antibody to suppress B cell functions by coengagement of BCR and FcgammaRIIb may represent a novel approach in the treatment of B cell-mediated autoimmune diseases.

Dominant-negative inhibitors of soluble TNF attenuate experimental arthritis without suppressing innate immunity to infection

TNF is a pleiotropic cytokine required for normal development and function of the immune system; however, TNF overexpression also induces inflammation and is associated with autoimmune diseases. TNF exists as both a soluble and a transmembrane protein. Genetic studies in mice have suggested that inflammation in disease models involves soluble TNF (solTNF) and that maintenance of innate immune function involves transmembrane TNF (tmTNF). These findings imply that selective pharmacologic inhibition of solTNF may be anti-inflammatory and yet preserve innate immunity to infection. To address this hypothesis, we now describe dominant-negative inhibitors of TNF (DN-TNFs) as a new class of biologics that selectively inhibits solTNF. DN-TNFs blocked solTNF activity in human and mouse cells, a human blood cytokine release assay, and two mouse arthritis models. In contrast, DN-TNFs neither inhibited the activity of human or mouse tmTNF nor suppressed innate immunity to Listeria infection in mice. These results establish DN-TNFs as the first selective inhibitors of solTNF, demonstrate that inflammation in mouse arthritis models is primarily driven by solTNF, and suggest that the maintenance of tmTNF activity may improve the therapeutic index of future anti-inflammatory agents.

Soluble TNF mediates the transition from pulmonary inflammation to fibrosis

Background

Fibrosis, the replacement of functional tissue with excessive fibrous tissue, can occur in all the main tissues and organ systems, resulting in various pathological disorders. Idiopathic Pulmonary Fibrosis is a prototype fibrotic disease involving abnormal wound healing in response to multiple sites of ongoing alveolar epithelial injury.

Methodology/Principal Findings

To decipher the role of TNF and TNF-mediated inflammation in the development of fibrosis, we have utilized the bleomycin-induced animal model of Pulmonary Fibrosis and a series of genetically modified mice lacking components of TNF signaling. Transmembrane TNF expression is shown to be sufficient to elicit an inflammatory response, but inadequate for the transition to the fibrotic phase of the disease. Soluble TNF expression is shown to be crucial for lymphocyte recruitment, a prerequisite for TGF-b1 expression and the development of fibrotic lesions. Moreover, through a series of bone marrow transfers, the necessary TNF expression is shown to originate from the non-hematopoietic compartment further localized in apoptosing epithelial cells.

Conclusions

These results suggest a primary detrimental role of soluble TNF in the pathologic cascade, separating it from the beneficial role of transmembrane TNF, and indicate the importance of assessing the efficacy of soluble TNF antagonists in the treatment of Idiopathic Pulmonary Fibrosis.

Blocking soluble tumor necrosis factor signaling with dominant-negative tumor necrosis factor inhibitor attenuates loss of dopaminergic neurons in models of Parkinson's disease

The mechanisms that trigger or contribute to loss of dopaminergic (DA) neurons in Parkinson's disease (PD) remain unclear and controversial. Elevated levels of tumor necrosis factor (TNF) in CSF and postmortem brains of PD patients and animal models of PD implicate this proinflammatory cytokine in the pathophysiology of the disease; but a role for TNF in mediating loss of DA neurons in PD has not been clearly demonstrated. Here, we report that neutralization of soluble TNF (solTNF) in vivo with the engineered dominant-negative TNF compound XENP345 (a PEGylated version of the TNF variant A145R/I97T) reduced by 50% the retrograde nigral degeneration induced by a striatal injection of the oxidative neurotoxin 6-hydroxydopamine (6-OHDA). XENP345 was neuroprotective only when infused into the nigra, not the striatum. XENP345/6-OHDA rats displayed attenuated amphetamine-induced rotational behavior, indicating preservation of striatal dopamine levels. Similar protective effects were observed with chronic in vivo coinfusion of XENP345 with bacterial lipopolysaccharide (LPS) into the substantia nigra, confirming a role for solTNF-dependent neuroinflammation in nigral degeneration. In embryonic rat midbrain neuron/glia cell cultures exposed to LPS, even delayed administration of XENP345 prevented selective degeneration of DA neurons despite sustained microglia activation and secretion of solTNF. XENP345 also attenuated 6-OHDA-induced DA neuron toxicity in vitro. Collectively, our data demonstrate a role for TNF in vitro and in vivo in two models of PD, and raise the possibility that delaying the progressive degeneration of the nigrostriatal pathway in humans is therapeutically feasible with agents capable of blocking solTNF in early stages of PD.

Creating the next generation of protein therapeutics through rational drug design

Biopharmaceuticals, or protein drugs, have proven to be safe and effective therapies in many disease indications. However, the first generation of biopharmaceuticals has largely been limited to replicating the functions of native human proteins, or inhibiting these functions through the use of monoclonal antibodies. Recent advances in the design of biopharmaceuticals include computational approaches to manipulate protein structure, improved screening processes to synthesize and assay libraries of novel proteins, and new methods to modify proteins post-translationally and during production. Protein drug structure and function can now be optimized in the same way that small molecules are optimized via medicinal chemistry. This review addresses recent developments in the field of protein 'medicinal biology', and provides examples of how these tools are being applied to create the next generation of biopharmaceuticals possessing optimized drug properties and novel functions.

Inactivation of TNF signaling by rationally designed dominant-negative TNF variants

Tumor necrosis factor (TNF) is a key regulator of inflammatory responses and has been implicated in many pathological conditions. We used structure-based design to engineer variant TNF proteins that rapidly form heterotrimers with native TNF to give complexes that neither bind to nor stimulate signaling through TNF receptors. Thus, TNF is inactivated by sequestration. Dominant-negative TNFs represent a possible approach to anti-inflammatory biotherapeutics, and experiments in animal models show that the strategy can attenuate TNF-mediated pathology. Similar rational design could be used to engineer inhibitors of additional TNF superfamily cytokines as well as other multimeric ligands.

Phorbol 12-myristate 13-acetate up-regulates the transcription of MUC2 intestinal mucin via Ras, ERK, and NF-kappa B

MUC2 is a secretory mucin normally expressed by goblet cells of the intestinal epithelium. It is overexpressed in mucinous type colorectal cancers but down-regulated in colorectal adenocarcinoma. Phorbol 12-myristate 13-acetate (PMA) treatment of colon cancer cell lines increases MUC2 expression, so we have undertaken a detailed analysis of the effects of PMA on the promoter activity of the 5'-flanking region of the MUC2 gene using stably and transiently transfected promoter reporter vectors. Protein kinase C inhibitors (bisindolylmaleimide, calphostin C) and inhibitors of mitogen-activated protein/extracellular signal regulated kinase kinase (MEK) (PD98059 and U0126) suppressed up-regulation of MUC2. Src tyrosine kinase inhibitor PP2, a protein kinase A inhibitor (KT5720), and a p38 inhibitor (SB 203580) did not affect transcription. Western blotting and reverse transcription-PCR analysis confirmed these results. In addition, co-transfections with mutants of Ras, Raf, and MEK showed that the induction of MUC2 promoter activity by PMA required these three signaling proteins. Our results demonstrate that PMA activates protein kinase C, stimulating MAP kinase through a Ras- and Raf-dependent mechanism. An important role for nuclear factor kappaB (NF-kappaB) was also demonstrated using the inhibitor caffeic acid phenethyl ester and electrophoretic mobility shift assays. Such identification of pathways involved in MUC2 up-regulation by PMA in the HM3 colon cancer cell line may serve as a model for the effects of cytokines and growth factors, which regulate MUC2 expression during the progression of colorectal cancer.

MUC17, a novel membrane-tethered mucin

Membrane mucins have several functions in epithelial cells including cytoprotection, extravasation during metastases, maintenance of luminal structure, and signal transduction. In this paper we describe a large membrane mucin expressed in the normal intestine. This novel mucin, designated MUC17, contains an extended, repetitive extracellular glycosylation domain and a carboxyl terminus with two EGF-like domains, a SEA module domain, a transmembrane domain, and a cytoplasmic domain with potential serine and tyrosine phosphorylation sites. RNA blot analysis and in situ hybridization indicates that MUC17 is expressed in select pancreatic and colon cancer cell lines and in intestinal absorptive cells. Radiation hybrid mapping localized MUC17 to chromosome 7q22 where it resides in close proximity with three other membrane mucin genes, MUC3A, MUC3B, and MUC12. Thus, these membrane mucins reside together in a gene cluster, but are expressed in different tissues and are likely to have different functions as well.

Non-specific antiviral activity of antisense molecules targeted to the E1 region of human papillomavirus

Antisense phosphorothioate oligonucleotides (ODN1 0x5 OMe) directed against the E1 start region of human papillomavirus 11 (HPV11) can inhibit papillomavirus induced growth of implanted human foreskin in a mouse xenograft model. Administration of a mismatch control oligonucleotide (ODN9 0x5 OMe), in which guanine was replaced with adenine in the same model, had no effect on papilloma induced growth. However, the apparent antiviral activity of ODN1 0x5 OMe was also shown in a lethal mouse cytomegalovirus (CMV) model, in which the oligonucleotides are not expected to have antisense activity. To understand the mechanisms of action of these oligonucleotides, a mismatch oligonucleotide (ODN61 0x5 OMe) was prepared which retained the CpG motifs of ODN1 0x5 OMe. This was tested in the mouse xenograft model and shown to have moderate inhibitory activity. As a definitive experiment, a comparison was made between the efficacy of the active oligonucleotide ODN1 0x5 OMe against two papilloma viruses HPV11 and HPV40. Both these viruses cause benign genital warts, but differ by four bases in their E1 sequence that was the target for ODN1 0x5 OMe. Papillomavirus induced growth in the mouse xenograft model was inhibited by ODN1 0x5 OMe in both cases, suggesting that oligonucleotide molecules have a non-specific antiviral activity that is not directly related to their antisense sequence.

Cloning of the amino-terminal and 5'-flanking region of the human MUC5AC mucin gene and transcriptional up-regulation by bacterial exoproducts

To obtain gene regulatory sequence for the mucin gene MUC5AC, we have isolated the MUC5AC amino terminus cDNA and 5'-flanking region. This was possible through the use of rapid amplification of cDNA ends-polymerase chain reaction (RACE-PCR) in which the 5' sequence of the human gastric mucin cDNA HGM-1 (1) was used to design the first MUC5AC-specific primer. Primers for subsequent rounds of RACE were designed from the 5'-ends of amplified RACE products. After five rounds of RACE-PCR, we could no longer generate upstream extensions of the cDNA and hypothesized that we had reached the 5'-end. Primer extension and RNase protection analysis confirmed this. Combined nucleotide sequence for the RACE-PCR products was 3.3 kb with an open reading frame encoding 1100 amino acids. A putative translation start site was found at nucleotide +48. This was followed by a 45 nucleotide putative signal sequence. This amino-terminal sequence contains no tandem repeats but is >60% similar to the amino-terminal nucleotide sequence of MUC2. The positions of cysteine residues in this MUC2-similar region are almost 100% conserved between the two genes. Northern analysis showed expression of cognate RNA in the stomach and airway but not muscle and esophagus. This pattern was the same as that obtained using previously reported 3'-MUC5AC sequences. We have cloned approximately 4 kb of genomic DNA upstream of the transcription start site and have sequenced 1366 nucleotides containing a TATA box, a CACCC box, and putative binding sites for NFkappaB and Sp 1. Within 4 kb of the transcription start site are elements mediating transcriptional up-regulation in response to bacterial exoproducts.

Differential human nucleotide excision repair of paired and mispaired cisplatin-DNA adducts

In order to understand the action of the chemotherapeutic drug cisplatin, it is necessary to determine why some types of cisplatin-DNA intrastrand crosslinks are repaired better than others. Using cell extracts and circular duplex DNA, we compared nucleotide excision repair of uniquely placed 1,2-GG, 1,2-AG, and 1,3-GTG cisplatin-crosslinks, and a 2-acetylaminofluorene lesion. The 1,3 crosslink and the acetylaminofluorene lesion were repaired by normal cell extracts approximately 15-20 fold better than the 1,2 crosslinks. No evidence was found for selective shielding of 1,2 cisplatin crosslinks from repair by cellular proteins. Fractionation of cell extracts to remove putative shielding proteins did not improve repair of the 1,2-GG crosslink, and cell extracts did not selectively inhibit access of UvrABC incision nuclease to 1,2-GG crosslinks. The poorer repair of 1,2 crosslinks in comparison to the 1,3 crosslink is more likely a consequence of different structural alterations of the DNA helix. In support of this, a 1,2-GG-cisplatin crosslink was much better repaired when it was opposite one or two non-complementary thymines. Extracts from cells defective in the hMutSalpha mismatch binding activity also showed preferential repair of the 1,3 crosslink over the 1,2 crosslink, and increased repair of the 1,2 adduct when opposite thymines, showing that hMutSalphais not involved in the differential NER of these substrates in vitro. Mismatched cisplatin adducts could arise by translesion DNA synthesis, and improved repair of such adducts could promote cisplatin-induced mutagenesis in some cases.

An XPG DNA repair defect causing mutagen hypersensitivity in mouse leukemia L1210 cells

One of the most widely used antitumor drugs is cis-diamminedichloroplatinum(II) (cisplatin), and mechanisms of cisplatin resistance have been investigated in numerous model systems. Many studies have used mouse leukemia L1210/0 as a reference wild-type cell line, and cisplatin-resistant subclones have been derived from it. Increased DNA excision repair capacity is thought to play a key role in the acquired cisplatin resistance, and this has influenced development of drugs for clinical trials. We report here that the L1210/0 line is in fact severely deficient in nucleotide excision repair of damaged DNA in vivo and in vitro. L1210/0 cell extracts could be complemented by extracts from repair-defective human xeroderma pigmentosum (XP) or rodent excision repair cross-complementing (ERCC) mutant cells, except for XPG/ERCC5 mutants. Purified XPG protein could restore repair proficiency to L1210/0 extracts. Expression of mouse XPG mRNA was similar in all L1210 lines studied, suggesting a point mutation or small alteration of XPG in L1210/0 cells. The DNA repair capacity of a cisplatin-resistant subline, L1210/DDP10, is similar to that of type culture collection L1210 cells and to those of other normal mammalian cell lines. Nucleotide excision repair of DNA is thus clearly important in the intrinsic cellular defense against cisplatin. However, in contrast to what is generally believed, enhancement of DNA repair above the normal level in these rodent cells does not appear to be a mechanism of acquired resistance to the drug.

Repair by human cell extracts of single (6-4) and cyclobutane thymine-thymine photoproducts in DNA

One cis-syn cyclobutane thymine dimer or one (6-4) thymine-thymine photoproduct was built into an identical sequence of a closed-circular M13 duplex DNA, and nucleotide excision repair synthesis carried out by human cell extracts in the area containing each lesion was determined. Extracts from normal cells repaired the (6-4) photoproduct with a patch size of approximately 20-30 nucleotides, but repair was at least 10-fold lower at the cyclobutane dimer. The (6-4) lesion was repaired with comparable efficiency to a single acetylamino-fluorene-guanine adduct in a similar location. Extract from nucleotide excision repair-deficient xeroderma pigmentosum group A cells could not remove any of these adducts but could complete repair of the lesions after incision with Escherichia coli UvrABC proteins. This direct comparison of repair of two UV photoproducts, in an in vitro system where chromatin assembly and transcription are absent, suggests that the more rapid repair of the (6-4) lesion observed in the mammalian cell genome overall is due in part to a significant difference in the ability of the repair complex to locate and incise these lesions in DNA.

Co-correction of the ERCC1, ERCC4 and xeroderma pigmentosum group F DNA repair defects in vitro

The mammalian ERCC1-encoded polypeptide is required for nucleotide excision repair of damaged DNA and is homologous to Saccharomyces cerevisiae RAD10, which functions in repair and mitotic intrachromosomal recombination. Rodent cells representing repair complementation group 1 have nonfunctional ERCC1. We report that repair of UV-irradiated DNA can be reconstituted by combining rodent group 1 cell extracts with correcting protein from HeLa cells. Background repair was minimized by employing fractionated rodent cell extracts supplemented with human replication proteins RPA and PCNA. Group 1-correcting activity has a native molecular mass of 100 kDa and contains the 33 kDa ERCC1 polypeptide, as well as complementing activities for extracts from rodent group 4 and xeroderma pigmentosum group F (XP-F) cells. Extracts of group 1, group 4 or XP-F cells do not complement one another in vitro, although they complement extracts from other groups. The amount of ERCC1 detectable by immunoblotting is reduced in group 1, group 4 and XP-F extracts. Recombinant ERCC1 from Escherichia coli only weakly corrected the group 1 defect. The data suggest that ERCC1 is part of a functional protein complex with group 4 and XP-F correcting activities. The latter two may be equivalent to one another and analogous to S. cerevisiae RAD1.

Electron microscopy of DNA excision repair patches produced by human cell extracts

To characterize the process by which the mammalian nucleotide excision repair complex interacts with DNA to recognize and repair lesions, we have investigated the size and distribution of repair patches induced by human cell extracts in ultraviolet light-irradiated plasmid DNA. Repair synthesis was carried out in a buffer substituting biotinylated dUTP for dTTP, to allow repair patches to be detected by electron microscopy after streptavidin/colloidal gold labelling. Individual repair events on circular plasmids that had undergone repair synthesis in cell extracts were scored as gold particles bound specifically to irradiated molecules. Samples of over 2000 irradiated and unirradiated plasmids were counted. Repair synthesis at ultraviolet light photoproducts typically replaced about 30 nucleotides, since 69% of patches contained only one particle of 10 nm gold and 24% of patches contained two gold particles (each covering approx. 29 nucleotides). In addition, the ordering of repair events among damaged plasmids closely fitted a Poisson distribution, indicating that repair of lesions is achieved via a non-processive, random diffusion mechanism. This suggests that the repair complex is not intrinsically processive.

An intrastrand d(GpG) platinum crosslink in duplex M13 DNA is refractory to repair by human cell extracts

We have examined the ability of human cell extracts to repair the most frequent DNA adduct caused by the cancer chemotherapeutic agent cis-diamminedichloroplatinum(II). A circular DNA duplex with an intrastrand d(GpG) crosslink positioned at a specific site was synthesized. Human cell extracts were unable to induce repair synthesis in a 29-base-pair region encompassing the adduct or in adjacent regions. The same extracts could repair a single defined 2-acetylaminofluorene lesion in a similar location. When molecules containing the platinum adduct were cleaved by Escherichia coli UvrABC enzyme, human cell extracts could perform repair synthesis at the damaged site, suggesting that human enzymes fail to make incisions near the d(GpG) crosslink but can complete repair once incisions are made. This result indicates that most repair synthesis in DNA damaged with multiple cis-diamminedichloroplatinum(II) adducts takes place at lesions other than the predominant d(GpG) crosslink. These data support the idea that the clinical effectiveness of cis-diamminedichloroplatinum(II) may be explained by the inefficient repair of the major DNA adduct caused by this drug.