Frederick N. Miller

In vivo targeting of inducible NO synthase with oligodeoxynucleotides protects rat kidney against ischemia.

Gene products of all three distinct nitric oxide synthases are present in the mammalian kidney. This mosaic topography of nitric oxide synthase (NOS) isoforms probably reflects distinct functional role played by each enzyme. While nitric oxide (NO) is cytotoxic to isolated renal tubules, inhibition of NO production in vivo invariably results in the aggravation of renal dysfunction in various models of acute renal failure. We reasoned that the existing ambiguity on the role of nitric oxide in acute renal failure is in part due to the lack of selective NOS inhibitors. Phosphorothioated derivatives of antisense oligodeoxynucleotides targeting a conserved sequence within the open reading frame of the cDNA encoding the inducible NOS (iNOS) were designed to produce a selective knock-down of this enzyme. In vivo use of these antisense constructs attenuated acute renal failure in rats subjected to renal ischemia. This effect was due, at least in part, to the rescue of tubular epithelium from lethal injury. Application of antisense constructs did not affect endothelial NOS, as evidenced by a spared NO release after the infusion of bradykinin during in vivo monitoring with an NO-selective microelectrode. In conclusion, the data provide direct evidence for the cytotoxic effects of NO produced via iNOS in the course of ischemic acute renal failure, and offer a novel method to selectively prevent the induction of this enzyme.

Quantitation of vasodilator-induced macromolecular leakage by in vivo fluorescent microscopy

Abstract In vivo television microscopy of the rat cremaster muscle was used to investigate the phenomenon of drug-induced protein leakage from the vasculature. Concentration-response curves of the vasodilators, isoproterenol, histamine, and nitroprusside, were used to determine if vasodilation of this skeletal muscle vasculature would lead to or be associated with protein leakage. This objective required development of the methodology to quantitate protein leakage that occurs in any one vascular area of the microcirculation. Fluorescein isothiocyanate-tagged rat serum albumin was injected into the rats (ia) and an argon laser was used for excitation of the fluorochrome. The fluoroscent image was recorded on videotape for later analysis. Vascular diameters were measured off the television monitor. Protein leakage was quantitated by determining the change in interstitial fluorescent intensity that occurred during the experiment. These studies demonstrated that protein leakage started within 1 min of topical application of histamine and peak response occurred in 5–7 min. Nitroprusside and histamine produced protein leakage but isoproterenol just produced vasodilation. Furthermore, significant differences in pD2 values (−log ED 50 ), which were used as a measure of drug sensitivity, indicated that protein leakage and vasodilation were probably not related phenomena for histamine and nitroprusside. Thus, this method of quantitation of protein leakage in the microcirculation of the rat cremaster muscle, gave reproducible data that can be used to determine the time course of protein leakage and to compare concentration-dependent phenomena.

Mechanisms of α-thrombin, histamine, and bradykinin induced endothelial permeability

α‐Thrombin, bradykinin, and histamine are endogenous mediators that increase endothelial permeability. We examined the mechanism by which these three vasoactive mediators could alter permeability to albumin of human umbilical vein endothelial cells (HUVEC). HUVEC were grown to confluence on Transwell membranes and we monitored the flux of fluorescein isothiocyanate‐labeled human serum albumin across the membrane from the upper to lower chamber of the Transwell. Addition of α‐thrombin, bradykinin, or histamine increased the permeability coefficient of the HUVEC monolayer. At 30 min the permeability coefficient for α‐thrombin was 4.92 × 10−6 cm/sec while histamine was 4.47 × 10−6 cm/sec. Maximum changes in the permeability coefficient were about three‐fold control baseline values (1.59 × 10−6 cm/sec). There was also a temporal difference in the magnitude of the permeability coefficient. α‐Thrombin and bradykinin induced HUVEC permeability was increased for the first 90 min after which it returned to control levels. In contrast, histamine increased the permeability of the HUVEC monolayer throughout the 2 h experiment. To determine a possible intracellular mechanism of the altered permeability coefficients, HUVEC were labeled with FURA‐2 and intracellular calcium was monitored by digital fluorescence ratio imaging. Maximum intracellular calcium in HUVEC was increased by α‐thrombin (245 ± 20 nM) and histamine (210 ± 22 nM), but not by bradykinin (70 ± 7 nM) as compared to control (69 ± 10). Fluorescent photomicrographs of HUVEC stimulated with the three agonists indicated that α‐thrombin and histamine substantially altered HUVEC f‐actin arrangement, while bradykinin had no effect on HUVEC f‐actin distribution. These data support previous in vitro and in vivo studies demonstrating increased permeability by all three agonists. These data also show, for the first time, that histamine and α‐thrombin increased permeability by calcium‐dependent intracellular pathways, but bradykinin operates through a calcium‐independent mechanism.

The effect of photodynamic therapy on the microcirculation

Photodynamic therapy (PDT) is a new form of cancer therapy involving tumor localization by photosensitizing drugs such as dihematoporphyrin ether (DHE). Light irradiation of drug-sensitized tissue results in photoactivation of DHE and tumor necrosis. The mechanism of action is incompletely understood but involves the generation of singlet oxygen which produces cytotoxic effects on tissues containing the compound. In addition, microcirculatory aberrations have been described during PDT. We have studied the acute effects of PDT on the microcirculation using in vivo television microscopy of the rat cremaster. This has enabled us to observe the effects of PDT on both paired and unpaired arterioles and venules using two different wavelengths of activating light (blue, 450-490 nm, and green, 530-560 nm). For both wavelengths of activating light, significant reduction in blood flow of all vessels was seen during PDT. This, in combination with the formation and embolization of platelet thrombi, resulted in stasis of blood flow in 80% of arterioles and 90% of venules. Observation for 2 hr after the completion of photoactivation revealed reperfusion in 20% of the arterioles but none of the venules. Blood flow was reduced by a combination of vasoconstriction and platelet thrombus formation. Reducing the total activating energy from 120J/cm2 to 24J/cm2 significantly reduced the response in venules and the incidence of stasis in both arterioles and venules. We conclude that the photoactivation of DHE results in significant vasoconstriction and thrombosis of normal vessels and that if these effects are seen at later times after DHE administration and in tumor neovasculature they may contribute to the mechanism by which PDT results in tumor necrosis.

Differential macromolecular leakage from the vasculature of tumors.

Tumor‐induced neovascularization is essential for invasion, metastases, and exponential growth of solid tumors. The authors studied the differences in macromolecular leakage from the neovasculature of a fast‐growing, early‐metastasizing tumor, the Walker 256 carcinosarcoma, and a slow‐growing, nonmetastasizing tumor, a rat chondrosarcoma. A 1‐mm3 piece of the Walker 256 carcinoma or the chondrosarcoma was implanted in the cremaster muscle of rats. Five days after surgery the cremaster muscle with the implanted tumor was placed in a special bath containing Krebs solution such that the circulation and nerves from the animal to the cremaster were intact. Fluorescein isothiocyanate‐labeled rat serum albumin (FITC‐RSA) was injected (intra‐arterially) into each rat to permit visualization of the vasculature by fluorescent microscopy. A closed‐circuit television system was used to quantitate macromolecular leakage as a change in interstitial fluorescent intensity. Data are given as a relative fluorescent intensity (mean ± standard error of the mean) in an area of the cremaster with tumor‐induced neovascularization.

A Comparison of the Effects of Photodynamic Therapy on Normal and Tumor Blood Vessels in the Rat Microcirculation

The effects of light activation of the tumor photosensitizer dihematoporphyrin ether (DHE) were studied in the microcirculation of the rat cremaster muscle. Arterioles and venules in an implanted chondrosarcoma were studied by in vivo television microscopy and were compared to normal vessels of the same size elsewhere in the preparation and in control preparations. Activation with green light (530-560 nm, 200 mW/cm2, 120 J/cm2) 48 h after intraperitoneal injection of DHE (10 mg/kg body wt) resulted in significant narrowing of diameters of red blood cell columns in tumor arterioles and venules. The response in normal and control arterioles and venules was not significantly different from that seen in the tumor vessels except that the control arterioles did not remain significantly constricted during the treatment period. Treatment resulted in stasis of blood flow in 90% of tumor and normal arterioles at the completion of light activation. In venules, stasis of blood flow was observed in 75% of tumor and 70% of normal vessels. Vasoconstriction was the primary response in arterioles, while thrombosis predominated in venules. Morphologic assessment of light-activated vessels in the cremaster preparation by transmission electron microscopy revealed platelet aggregation with damage to endothelial cells and smooth muscle cells. Perivascular effects observed included interstitial edema and damage to skeletal muscle cells. In the tumor-bearing preparation, no direct cytotoxic effect on the tumor cells was shown. The surrounding vessels exhibited similar vascular stasis, but the lining cells appeared minimally affected. Photoactivation of DHE results in significant changes in the microcirculation which lead to stasis of blood flow. In this model, the response was similar for the normal microvasculature and for the microcirculation of an implanted chondrosarcoma. These effects may account, in part, for the mechanism of action of photodynamic therapy.

Mechanisms of leukocyte-mediated tissue injury induced by interleukin-2

Abstract In this review we discuss the effects of interleukin-2 (IL-2) therapy on trafficking of leukocytes and platelets to normal organs. The use of animal models has allowed the elucidation of events leading to damage and/or dysfunction of normal tissues after IL-2 administration. These studies have shown that acute toxicity induced by IL-2 is mediated primarily by neutrophils. Chronic toxicity results from the adhesion and transmigration of activated lymphocytes into normal organs. Platelet-derived microvascular thrombosis also contributes to the vascular toxicity of IL-2. A better understanding of these mechanisms may lead to the development of interventions that will significantly improve the therapeutic efficacy of IL-2.

Pentoxifylline inhibits interleukin-2-induced toxicity in C57BL/6 mice but preserves antitumor efficacy.

Interleukin 2 (IL-2) mediates the regression of metastatic cancer but clinical use has been limited due to associated toxicities. Tumor necrosis factor (TNF) is an important mediator of IL-2 toxicity and may have a limited role in IL-2 antitumor efficacy. Because pentoxifylline (PTXF) inhibits TNF production, we hypothesized that PTXF would ameliorate IL-2 toxicity without compromising antitumor efficacy. Four groups of female C57BL/6 mice with pulmonary metastases from a 3-methylcholanthrene-induced fibrosarcoma (MCA-105) and four groups of nontumored mice were treated every 6 h for 4 d by intraperitoneal injections of either IL-2 alone, IL-2 and PTXF, PTXF alone, or equal volumes of saline. Upon completion of therapy, we found that PTXF suppressed many of the IL-2-induced effects including TNF production, lymphocytic infiltration of multiple organs, multiple organ edema, hepatic dysfunction, leukopenia, and thrombocytopenia. Tumor response was determined 21 d after cessation of therapy by quantitating the number and surface area of pulmonary metastases. PTXF preserved antitumor efficacy while reducing the morbidity and mortality caused by IL-2 treatment. These data strongly support the use of PTXF in extending the therapeutic index of IL-2 in the treatment of cancer.

Differentiation of light-dye effects in the microcirculation

Activation of photosensitive compounds has been used in the treatment of tumors and as a technique to study various microcirculatory phenomena. This technique may be accompanied by deleterious effects which may complicate interpretations of experimental results. However, the relevant physiological mechanisms that induce toxicity and the light doses needed to produce different toxic reactions have not been well defined. In the current study, the rat cremaster muscle preparation was used with in vivo fluorescent television microscopy and subsequently with electron and light microscopy to evaluate toxic reactions of light activation of fluorescein isothiocyanate. The most sensitive photoactive reactions were macromolecular leakage and platelet activation, occurring with 120 J/cm2 activation energy. Macromolecular leakage was at least partially restricted by perivenular and pericapillary pericytes and there was no morphological damage with this light dose. Since macromolecular leakage was significantly inhibited by pretreatment with diphenhydramine or Compound 48/80, it is in part due to the release of histamine from tissue mast cells. 720 J/cm2 reduced the red blood cell column in the venules by over 50% due to platelet thrombus formation, an effect that was accentuated by pretreatment with indomethacin. This suggests an inhibitory role of prostaglandins in platelet thrombus formation. In addition, 720 J/cm2 caused endothelial and smooth muscle cell swelling and ruptures, gap formation, and leukocyte and protein accumulation in the vessel walls.

THE ROLE OF CYTOKINES, ADHESION MOLECULES, AND CHEMOKINES IN INTERLEUKIN-2-INDUCED LYMPHOCYTIC INFILTRATION IN C57BL/6 MICE

IL-2 mediates the regression of certain malignancies, but clinical use is limited because of associated toxicities, including parenchymal lymphocytic infiltration with multiple organ failure. Secondarily induced cytokines are important mediators of IL-2 toxicity and IL-2-induced lymphocyte-endothelial adherence and trafficking. The recently discovered C-C chemokines, RANTES (regulated on activation, normal T expressed and secreted) and macrophage inflammatory protein-1alpha, have also been implicated in lymphocytic migration. We hypothesized that IL-2 alters cytokine, C-C chemokine, and adhesion molecule expression in association with parenchymal lymphocytic infiltration. C57BL/6 mice were injected with 3x10(5) IU of IL-2 or 0.1 ml of 5% dextrose intraperitoneally every 8 h for 6 d, then killed. IL-2 induced massive lymphocytic infiltration in the liver and lung and moderate infiltration in the kidney in association with organ edema and dysfunction. Immunostaining showed increased intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) expression in association with this organ-specific lymphocytic infiltration. Flow cytometry showed increased expression of the corresponding ligands (lymphocyte function-associated antigen-1 and very late antigen-4) on splenocytes. IL-2 increased TNF-alpha mRNA and protein expression in the liver. Organs infiltrated by lymphocytes had increased TNF-alpha mRNA, whereas RANTES mRNA was increased in all organs, regardless of lymphocytic infiltration. IL-2 toxicity involves organ-specific TNF-alpha and RANTES production with increased ICAM-1 and VCAM-1 expression as potential mechanisms facilitating lymphocytic infiltration and organ dysfunction.