Bivalent engagement of endothelial surface antigens is critical to prolonged surface targeting and protein delivery in vivo.

Targeted drug delivery to the endothelium has the potential to generate localized therapeutic effects at the blood-tissue interface. For some therapeutic cargoes, it is essential to maintain contact with the bloodstream to exert protective effects. The pharmacokinetics (PK) of endothelial surface-targeted affinity ligands and biotherapeutic cargo remain a largely unexplored area, despite obvious translational implications for this strategy. To bridge this gap, we site-specifically radiolabeled mono- (scFv) and bivalent (mAb) affinity ligands specific for the endothelial cell adhesion molecules, PECAM-1 (CD31) and ICAM-1 (CD54). Radiotracing revealed similar lung biodistribution at 30 minutes post-injection (79.3% ± 4.2% vs 80.4% ± 10.6% ID/g for αICAM and 58.9% ± 3.6% ID/g vs. 47.7% ± 5.8% ID/g for αPECAM mAb vs. scFv), but marked differences in organ residence time, with antibodies demonstrating an order of magnitude greater area under the lung concentration vs. time curve (AUCinf 1698 ± 352 vs. 53.3 ± 7.9 ID/g*hrs for αICAM and 1023 ± 507 vs. 114 ± 37 ID/g*hrs for αPECAM mAb vs scFv). A physiologically based pharmacokinetic model, fit to and validated using these data, indicated contributions from both superior binding characteristics and prolonged circulation time supporting multiple binding-detachment cycles. We tested the ability of each affinity ligand to deliver a prototypical surface cargo, thrombomodulin (TM), using one-to-one protein conjugates. Bivalent mAb-TM was superior to monovalent scFv-TM in both pulmonary targeting and lung residence time (AUCinf 141 ± 3.2 vs 12.4 ± 4.2 ID/g*hrs for ICAM and 188 ± 90 vs 34.7 ± 19.9 ID/g*hrs for PECAM), despite having similar blood PK, indicating that binding strength is more important parameter than the kinetics of binding. To maximize bivalent target engagement, we synthesized an oriented, end-to-end anti-ICAM mAb-TM conjugate and found that this therapeutic had the best lung residence time (AUCinf 253 ± 18 ID/g*hrs) of all TM modalities. These observations have implications not only for the delivery of TM, but also potentially all therapeutics targeted to the endothelial surface.

The transcription factor ERG regulates a low shear stress-induced anti-thrombotic pathway in the microvasculature.

Endothelial cells actively maintain an anti-thrombotic environment; loss of this protective function may lead to thrombosis and systemic coagulopathy. The transcription factor ERG is essential to maintain endothelial homeostasis. Here, we show that inducible endothelial ERG deletion (ErgiEC-KO) in mice is associated with spontaneous thrombosis, hemorrhages and systemic coagulopathy. We find that ERG drives transcription of the anticoagulant thrombomodulin (TM), as shown by reporter assays and chromatin immunoprecipitation. TM expression is regulated by shear stress (SS) via Krüppel-like factor 2 (KLF2). In vitro, ERG regulates TM expression under low SS conditions, by facilitating KLF2 binding to the TM promoter. However, ERG is dispensable for TM expression in high SS conditions. In ErgiEC-KO mice, TM expression is decreased in liver and lung microvasculature exposed to low SS but not in blood vessels exposed to high SS. Our study identifies an endogenous, vascular bed-specific anticoagulant pathway in microvasculature exposed to low SS.

Vascular endothelial effects of collaborative binding to platelet/endothelial cell adhesion molecule-1 (PECAM-1).

Targeting drugs to endothelial cells has shown the ability to improve outcomes in animal models of inflammatory, ischemic and thrombotic diseases. Previous studies have revealed that certain pairs of ligands (antibodies and antibody fragments) specific for adjacent, but distinct, epitopes on PECAM-1 enhance each other's binding, a phenomenon dubbed Collaborative Enhancement of Paired Affinity Ligands, or CEPAL. This discovery has been leveraged to enable simultaneous delivery of multiple therapeutics to the vascular endothelium. Given the known role of PECAM-1 in promoting endothelial quiescence and cell junction integrity, we sought here to determine if CEPAL might induce unintended vascular effects. Using a combination of in vitro and in vivo techniques and employing human and mouse endothelial cells under physiologic and pathologic conditions, we found only modest or non-significant effects in response to antibodies to PECAM-1, whether given solo or in pairs. In contrast, these methods detected significant elevation of endothelial permeability, pro-inflammatory vascular activation, and systemic cytokine release following antibody binding to the related endothelial junction protein, VE-Cadherin. These studies support the notion that PECAM-1-targeted CEPAL provides relatively well-tolerated endothelial drug delivery. Additionally, the analysis herein creates a template to evaluate potential toxicities of vascular-targeted nanoparticles and protein therapeutics.

Protection of cultured endothelial cells from hydrogen peroxide-induced injury by antibody-conjugated catalase.

The cytoprotective features of catalase-antibody conjugate prepared by covalent conjugation of catalase to rabbit antibody against mouse IgG is described. The bifunctional cross-linking agent m-maleimidobenzoic acid N-hydroxysuccinimide ester (MBS) was used for conjugation. Functionally active conjugate binds specifically to the plastic-adsorbed mouse IgG and to the surface of live human endothelial cells treated with mouse antiserum against human endothelial cells. Up to 4 units of catalase activity can bind to 1 cm2 of the endothelial monolayer. The targeted catalase protects endothelial cells from cytotoxic action of hydrogen peroxide: the minimal cytotoxic concentration of H2O2 for protected cells is 80-times higher than for intact cells. This effect is attributed partly to local reduction of H2O2 concentration in the cell microenvironment. Targeted catalase was estimated to reduce H2O2 concentration 8-fold near the cell surface with respect to average total concentration.

Avidin-induced lysis of biotinylated erythrocytes by homologous complement via the alternative pathway depends on avidin's ability of multipoint binding with biotinylated membrane.

It was reported that avidin and streptavidin induce lysis of prebiotinylated red blood cells via the alternative pathway of both homologous and heterologous complement. Both of these proteins have four biotin-binding sites, providing a polyvalent interaction with biotinylated components of the erythrocyte membrane. We have compared the effects of mono- and multipoint avidin attachment on the sensitivity of biotinylated erythrocytes to lysis by the complement system. In the presence of anti-avidin antibody, avidin-bearing biotinylated erythrocytes were rapidly lysed by heterologous serum. This lysis was independent from the mode of avidin attachment, implying that complement activation by the classical pathway triggered by interaction between C1 and avidin-bound antibody on the erythrocyte surface is independent from the avidin's ability of polyvalent (multipoint) binding with biotinylated membrane components. In the absence of anti-avidin antibody, biotinylated erythrocytes bearing polyvalently attached avidin were lysed by homologous complement better than cells bearing avidin, which possesses reduced ability for multipoint binding with biotinylated erythrocyte. Two independent approaches to reduce avidin's ability of multipoint binding were used: decrease in surface density of biotin on the erythrocyte membrane and blockage of biotin-binding sites of avidin. Both methods result in reduced lysis of avidin-bearing erythrocytes as compared with erythrocytes bearing an equal amount of polyvalent-bound avidin. Thus the activation of homologous complement via the alternative pathway depends on avidin's ability to 'cross-link' to the biotinylated components of the erythrocyte membrane.

Avidin attachment to biotinylated human neutrophils induces generation of superoxide anion.

The influence of biotinylation and subsequent attachment of avidin on generation of superoxide anion by human neutrophils was studied. Biotinylation of human neutrophils with succinimide ester of biotin does not reduce superoxide generation in response to activation with phorbol myristate acetate (PMA) and formyl peptide (FMLP). Addition of avidin to biotinylated, but not native, leukocytes induces generation of superoxide anion. The kinetics and level of superoxide generation by biotinylated neutrophils in response to addition of avidin were quite similar to those in response to activation with FMLP. The avidin sugar moiety and charge were not involved in superoxide generation, since streptavidin was also active. Both avidin- and PMA-induced superoxide generation were independent of the extracellular calcium, while FMLP-induced superoxide generation was dependent on the presence of calcium in solution. Therefore, interaction of avidin with biotinylated components of the neutrophil membrane alters functional activity of this cell and might induce 'activation-like' reaction of leukocytes.

Cytotoxicity of glucose oxidase conjugated with antibodies to target cells: killing efficiency depends on the conjugate internalization.

The cytotoxic action of glucose oxidase conjugated with antibodies against the target cells has been examined in a culture of human endothelial cells. Internalizable (anti-endothelial, MoAb E25) and non-internalizable (anti-fibronectin, MoAb FN) monoclonal antibodies were employed as vectors. Anti-endothelial monoclonal antibody E78 (whether it can be internalized by endothelial cells is unclear) and polyclonal mouse antiserum to the human endothelium were also used. The conjugates were prepared by oxidation of the enzyme carbohydrate moiety with periodate. Free conjugates display similar enzyme activity in glucose solution. In contrast to glucose oxidase, conjugated with no-immune IgG, antibody-conjugated glucose oxidase binds specifically to target cells. The efficiency of targeting was different for various conjugates. Targeting via the anti-fibronectin antibody and anti-endothelial antiserum provided maximal quantitative binding of glucose oxidase to endothelial cells, while the conjugates with MoAb E25 and MoAb E78 monoclonal antibodies provided less effective binding. In the presence of glucose, targeted glucose oxidase generated H2O2. Hydrogen peroxide is relatively stable in buffer, but rapidly decays in the culture medium supplemented with 20% human serum. Though the quantitative binding of MoAb E25-conjugated glucose oxidase was minimal comparing to other conjugates, targeting via MoAb E25 produced the maximal cytotoxic effect as well as targeting via polyclonal antiserum. The killing efficiencies of MoAb FN-conjugated and MoAb E78-conjugated glucose oxidase were about 30-fold lower. The high efficiency of the MoAb E25-conjugated enzyme may be due to its internalization by target cells. Internalization can lead to unaccessibility of generated H2O2 for extracellular scavengers and pH optimization for glucose oxidase activity, which provides valuable advantages for the cytotoxicity of the conjugate. Thus, cytotoxicity of antibody-conjugated glucose oxidase depends not only on the efficiency of specific binding to the target cell, but also on the fate of cell-bound conjugate. Cytotoxicity is extremely effective in case of 'internalizable' conjugate and drastically less effective in case of 'non-internalizable' conjugate.

Immunotargeting of erythrocyte-bound streptokinase provides local lysis of a fibrin clot.

The creation of an anticollagen antibody-erythrocyte-streptokinase complex has been described. Immobilization of both proteins on erythrocyte membrane has been performed using an avidin-biotin interaction. Modification of streptokinase with (6-biotinylamido)hexanoic acid N-hydroxysuccinimide ester at the concentration of 1.1 mM (20% modification of protein amino groups) provides effective (up to 90%) attachment of streptokinase to an avidin-carrying erythrocyte surface. The loss of streptokinase activity due to modification under these conditions is not significant. The maximal attachment of streptokinase was equal to about 50 ng per 10(6) erythrocytes, i.e., about 5 X 10(5) molecules of streptokinase per erythrocyte. The presence of streptokinase in the incubation mixture inhibited the attachment of antibodies by about 50%. Nevertheless, co-immobilization of anticollagen antibody (1.0 X 10(5) molecules per cell) and streptokinase (2.8 X 10(5) molecules per cell) on the erythrocyte surface provided firm and specific binding of such erythrocytes to a collagen-coated surface (1.6 X 10(6) bound cells per 1 cm2 on a collagen-coated surface against 0.006 X 10(6) bound cells on a bovine serum albumin-coated surface). Targeting of such erythrocytes led to local lysis of a fibrin clot in the target zone. The properties described offer in principle the possibility of the application of this or a similar system of fibrinolytic agent targeting for the preventive therapy of rethrombosis during surgical manipulations on vessels.

Target-sensitive immunoerythrocytes: interaction of biotinylated red blood cells with immobilized avidin induces their lysis by complement.

Red blood cells (RBC) coated with antibody (immunoerythrocytes) may be useful for drug targeting. Previously we have developed a methodology for avidin (streptavidin)-mediated attachment of biotinylated antibodies (b-Ab) to biotinylated RBC (B-RBC). We have observed that binding of avidin to B-RBC in suspension leads to their complement-mediated lysis by autologous serum. In the present work we have studied the interaction of B-RBC, which are not complement susceptible, with immobilized avidin and their consequent susceptibility to lysis by complement. B-RBC adhered tightly to avidin-coated surfaces and were rendered susceptible to lysis by autologous serum. A long biotin ester provided more effective binding of the B-RBC to immobilized avidin and greater lysis by complement, than a short biotin ester. Based on these results, we have hypothesized that targeting of serum-stable drug-loaded B-RBC attained by step-wise administration of b-Ab and streptavidin may provide target-sensitive lysis of B-RBC. To confirm this hypothesis, we have studied b-Ab and streptavidin mediated targeting of B-RBC to immobilized antigen. Step-wise addition of biotinylated antibody, avidin or streptavidin and b-RBC caused specific binding of B-RBC to immobilized antigen and their subsequent lysis by autologous serum. Therefore, our results obtained in an in vitro model demonstrate that B-RBC might be used for targeting and local release of drug.

Interaction of avidin-carrying red blood cells with nucleated cells.

In vivo application of red blood cells (RBC) modified with avidin-biotin complex has been suggested recently for various purposes. However, avidin attachment to RBC alters their biocompatibility. Thus, it has been described that avidin-carrying biotinylated RBC were lysed by the complement. In the present work interaction between avidin-carrying RBC and nucleated cells has been examined. It was found that attachment of avidin, but not streptavidin, to RBC led to binding of avidin-carrying RBC to nucleated cells. Adhesiveness of nucleated cells for avidin-carrying RBC varied for different types of nucleated cells. The strongest adhesion was observed with human fibroblasts and rat Kupffer cells, while rat liver endothelial cells were practically non-adhesive for avidin-carrying RBC of corresponding species. In contrast with avidin (streptavidin)-induced lysis by the complement, avidin-induced adhesion was independent of temperature, the presence of divalent ions and mode of avidin attachment. Polyanions (dextran sulphate and heparin) efficiently inhibited the adhesion presumably due to interaction with the membrane-bound avidin. Polyanions to a much lesser extent inhibited lysis of avidin-carrying RBC, which might be a result of their interaction with the complement components. Polycations also blocked adhesion of avidin-carrying RBC to nucleated cells, presumably due to interaction with negatively charged cell-surface components. Therefore, attachment of avidin to RBC alters their biocompatibility, due to both high positive charge of avidin and the cross-linking of biotinylated membrane proteins.

Targeting of antioxidant and anti-thrombotic drugs to endothelial cell adhesion molecules.

The endothelium represents an important therapeutic target for containment of oxidative stress, thrombosis and inflammation involved in a plethora of acute and chronic conditions including cardiovascular and pulmonary diseases and diabetes. However, rapid blood clearance and lack of affinity to the endothelium compromise delivery to target and restrict medical utility of antioxidant enzymes (e.g., catalase) and fibrinolytics. The use of "stealth" PEG-liposomes prolongs circulation, whereas conjugation with antibodies to endothelial determinants permits targeting. Constitutive endothelial cell adhesion molecules (CAM, such as ICAM-1 and PECAM-1, which are stably expressed and functionally involved in oxidative stress and thrombosis) are candidate determinants for targeting of antioxidants and fibrinolytics. CAM antibodies and compounds conjugated with anti-CAM bind to endothelial cells and accumulate in vascularized organs (preferentially, lungs). Pathological stimuli enhance ICAM-1 expression in endothelial cells and facilitate targeting, whereas PECAM-1 expression and targeting are stable. Endothelial cells internalize 100-300 nm diameter conjugates possessing multiple copies of anti-CAM, but not monomolecular antibodies or micron conjugates. This permits size-controlled sub-cellular targeting of antioxidants into the endothelial interior and fibrinolytics to the endothelial surface. Targeting catalase to PECAM-1 or ICAM-1 protects endothelial cells against injury by oxidants in culture and alleviates vascular oxidative stress in lungs in animals. Anti-CAM/catalase conjugates are active for a few hours prior to lysosomal degradation, which can be delayed by auxiliary drugs. Conjugation of fibrinolytics to monovalent anti-ICAM permits targeting and prolonged retention on the endothelial surface. Therefore, CAM targeting of antioxidants and fibrinolytics might help to contain oxidative and thrombotic stresses, with benefits of blocking CAM. Avenues for improvement and translation of this concept into the clinical domain are discussed.

Mouse model of microembolic stroke and reperfusion.

BACKGROUND AND PURPOSE: To test the role of fibrinolysis in stroke, we used a mouse model in which preformed 2.5- to 3-microm-diameter fibrin microemboli are injected into the cerebral circulation. The microemboli lodge in the downstream precapillary vasculature and are susceptible to fibrinolysis. METHODS: We injected various doses of microemboli into the internal carotid artery in mice and characterized their distribution, effects on cerebral blood flow, neurological deficit, infarct area, and spontaneous dissolution. By comparing wild-type and tissue plasminogen activator (tPA) knockout (tPA-/-) mice, we analyzed the role of endogenous tPA in acute thrombotic stroke. RESULTS: Microemboli cause dose-dependent brain injury. Although moderate doses of microemboli are followed by spontaneous reperfusion, they result in reproducible injury. Gene knockout of tPA markedly delays dissolution of cerebral emboli and restoration of blood flow and aggravates ischemic thrombotic infarction in the brain. CONCLUSIONS: We describe a microembolic model of stroke, in which degree of injury can be controlled by the dose of microemboli injected. Unlike vessel occlusion models, this model can be modulated to allow spontaneous fibrinolysis. Application to tPA-/- mice supports a key role of endogenous tPA in restoring cerebral blood flow and limiting infarct size after thrombosis.

Effects of chronic arterial hypertension on constitutive and induced intercellular adhesion molecule-1 expression in vivo.

Recent reports indicate that bacterial endotoxin (lipopolysaccharide) and cytokines elicit a more profound increase in the surface expression of intercellular adhesion molecule-1 (ICAM-1) in cultured endothelial cells derived from spontaneously hypertensive (SHR) versus normotensive Wistar-Kyoto rats (WKY). Our objective in this study was to characterize and compare in vivo ICAM-1 expression in SHR and WKY under basal conditions and after 5 hours of endothelial cell activation with either lipopolysaccharide (5 mg/kg i.p.) or tumor necrosis factor-alpha (TNF-alpha; 1, 5, and 10 micrograms/kg i.p.). ICAM-1 expression was quantified in different tissues by the double-radiolabeled monoclonal antibody technique. When constitutive (baseline) ICAM-1 expression was corrected for endothelial cell surface area, significantly higher values were noted in SHR than WKY but only in splanchnic organs. Lipopolysaccharide and TNF-alpha elicited significant increases in ICAM-1 expression in all tissues of both WKY and SHR. However, the magnitude of the lipopolysaccharide-induced ICAM-1 upregulation in heart, stomach, skeletal muscle, and brain was significantly lower in SHR than WKY. A similar blunted ICAM-1 upregulation was noted in the stomach of SHR after administration of 5 micrograms/kg TNF-alpha. The differences in induced ICAM-1 expression between SHR and WKY do not appear to be due to differences in endothelial cell surface area or plasma glucocorticoid levels. These results suggest that chronic arterial hypertension results in altered ICAM-1 expression on the endothelium, which may contribute to the abnormal inflammatory responses associated with this disease.

Streptavidin-induced lysis of homologous biotinylated erythrocytes. Evidence against the key role of the avidin charge in complement activation via the alternative pathway.

It is shown that non-covalent attachment of streptavidin, as well as of avidin, to biotinylated human erythrocytes induces homologous hemolysis by complement. Rabbit antiserum against human C3 is found to inhibit the lysis specifically as compared with non-immune rabbit serum. Efficiency of lysis inhibition is greater for avidin- and streptavidin-induced lysis of biotinylated human erythrocytes than for antibody-sensitized sheep erythrocytes. In contrast to positively charged avidin (pI 11), streptavidin is a neutral protein. Hence, hemolysis of streptavidin-carrying erythrocytes is inconsistent with the suggestion on the crucial role of avidin charge in lysis. Membrane alterations (cross-linking and clusterization of biotinylated components) induced by avidin (streptavidin) seem to be a more plausible explanation for the lysis.

Avidin attachment to biotinylated amino groups of the erythrocyte membrane eliminates homologous restriction of both classical and alternative pathways of the complement.

Lysis of avidin-coated biotinylated sheep red blood cells (RBC) via the classical pathway of homologous (sheep) and heterologous (guinea pig) complement has been studied. The minimal surface density of avidin inducing antibody-dependent lysis via the classical pathway is smaller than that inducing antibody-independent lysis via the alternative pathway. Heterologous lysis via the classical pathway does not depend on the mode of avidin attachment: both biotinylation of membrane amino groups and insertion of biotinyl-lipid into the membrane provide the same lysis of avidin-coated RBCs by guinea pig serum in the presence of anti-avidin antibody. Avidin-free sheep RBC sensitized with hemolytic anti-RBC antibody were lysed by guinea pig, but not by sheep serum, confirming high efficiency of homologous restriction of the complement. However, avidin-coated RBCs were lysed by homologous serum in the presence of anti-avidin antibody at low surface density of avidin attached. The elimination of the homologous restriction depends on the mode of avidin attachment: biotinylation of membrane amino groups provides antibody-mediated lysis via the classical pathway of homologous complement, while insertion of biotinyl-lipid does not provide lysis.

Targeting of enzyme immobilized on erythrocyte membrane to collagen-coated surface.

It is suggested to use 'enzyme(s)-erythrocyte-antibody' complex for modulation of the microenvironment in definite compartments of blood circulation. A model system including peroxidase, human erythrocytes and anti-collagen antibodies was chosen to illustrate the principle. Peroxidase was conjugated to the erythrocyte surface via periodate-oxidized enzyme carbohydrate moiety; biotinylated antibodies were linked by avidin to the biotinylated erythrocytes. The properties of the immunocomplexes obtained have been investigated in an artificial system simulating an injured blood vessel wall. The advantages in using erythrocyte-mediated immunoenzyme complexes for enzyme (drug) targeting are discussed.

Red blood cell targeting to collagen-coated surfaces.

The interaction of human red blood cells carrying antihuman collagen antibody with collagen-coated surfaces was studied. Avidin was used as bifunctional crosslinking agent for the attachment of antibody to the red blood cell surface. Antibody-carrying red blood cells efficiently and specifically bound to collagen-coated surface covering a significant part of the surface. The components of normal blood had an insignificant effect on red blood cell binding. A model of drug targeting to the injured sites(s) of blood vessel wall is proposed.

Red blood cell targeting to smooth muscle cells.

Monoclonal antibody discriminating between endothelial and smooth muscle cells is suggested to be used as a vector for directed transport of drugs to injured (denuded) areas of blood vessel wall. An in vitro model system was used in the studies: vascular smooth muscle or endothelial cells grown on plastic surface were treated with specific mouse monoclonal antibody recognizing an antigen localized on the surface of smooth muscle rather than endothelial cells; then erythrocytes coated with secondary (rabbit antimouse) antibodies were added. The results were analyzed spectrophotometrically or with scanning electron microscopy. Under the experimental conditions, erythrocytes, possible 'containers' for carrying the drugs, were found to bind only to smooth muscle cells. The data show that antibody provides absolute discrimination between endothelial and smooth muscle cells and, thus, may be used as a vector for drug targeting.

Delivery of antioxidant enzyme proteins to the lung.

Protection of alveolar epithelial cells (alveolocytes) and vascular endothelial cells against pulmonary oxidative stress is an important problem. An inadequate delivery to the target cells limits the protective utility of the antioxidant enzymes, superoxide dismutase (SOD) and catalase. SOD and catalase modifications, such as coupling with polyethylene glycol and encapsulation in liposomes, prolong the life span of the active enzymes in vivo. The airway administration of SOD and catalase protects alveolocytes against hyperoxic oxidative stress. Although pulmonary endothelium is poorly accessible from the airways, it is accessible from circulation. However, antioxidant enzymes and their derivatives display poor targeting to pulmonary endothelium. To improve the targeting and provide intracellular delivery to endothelium, the enzymes can be conjugated with antibodies against endothelial antigens, such as angiotensin-converting enzyme and adhesion molecules [intercellular adhesion molecule-1 (ICAM-1) or platelet-endothelial cell adhesion molecule-1 (PECAM-1)]. These immunoconjugates accumulate in the pulmonary vasculature in intact animals, enter endothelium, and augment the antioxidant defenses. The immunoconjugates directed against ICAM-1 and PECAM-1 may also provide a secondary therapeutic benefit by blocking of sequestration and infiltration of leukocytes in the lungs. Further investigations are necessary to evaluate the therapeutic effectiveness of the vascular immunotargeting of antioxidant enzymes and solve technical problems associated with production of safe, clinically useful conjugates.

Local tissue injury induced by glucose oxidase conjugated with anti-collagen antibody.

The conjugation of glucose oxidase with anti-collagen antibody using periodate oxidation of the enzyme carbohydrate moiety is described. After conjugation, the antibody retained its antigen-binding capacity and the enzyme retained hydrogen peroxide-generating activity. Intradermal administration of the immune conjugate into rats induced local tissue injury at doses 10-100 micrograms. Pronounced damage (local hyperemia and edema) occurred 24 h after injection and necrosis developed 1 week later. The enzyme was tightly bound to the fibrillar components of the extracellular matrix and retained its activity in vivo for a prolonged period of time. In contrast, non-immune IgG-conjugated glucose oxidase was removed rapidly from the site of injection and did not induce tissue damage. Pure native anti-collagen antibody was retained at the site of injection for 8 days, but caused no tissue injury. These results suggest that active glucose oxidase conjugated with antibodies to tissue antigen can be accumulated and retained in the tissues. At the site of accumulation local 'proinflammatory' damage develops even in the absence of the halide-peroxidase system. Similar conjugates could be potential agents for local modulation of inflammation.