In vitro and in vivo studies of PEO-grafted blood-contacting cardiovascular prostheses

The initial step of thrombus formation on blood-contacting biomaterials is known to be adsorption of blood proteins followed by platelet adhesion. Poly(ethylene oxide) (PEO) has been frequently used to modify biomaterial surfaces to minimize or prevent protein adsorption and cell adhesion. PEO was grafted onto a number of biomaterials in our laboratory. Nitinol stents and glass tubes were grafted with PEO by priming the metal surface with trichlorovinylsilane (TCVS) followed by adsorption of Pluronic and γ-irradiation. Nitinol stents were also coated with Carbothane® for PEO grafting. Chemically inert polymeric biomaterials, such as Carbothane, polyethylene, silicone rubber, and expanded polytetrafluoroethylene (e-PTFE), were first adsorbed with PEO-polybutadiene-PEO (PEO-PB-PEO) triblock copolymers and then exposed to γ-irradiation for covalent grafting. For PEO grafting to Dacron® (polyethylene terephthalate), the surface was sequentially treated with PEO-PB-PEO and Pluronics® followed by γ-irradiation. In vitro studies showed substantial reduction in fibrinogen adsorption and platelet adhesion to the PEO-grafted surfaces compared with control surfaces. Fibrinogen adsorption was reduced by 70-95% by PEO grafting on all surfaces, except for e-PTFE. The platelet adhesion corresponded to the fibrinogen adsorption. When the PEO-grafted surfaces were tested ex vivo/in vivo, however, the expected beneficial effect of PEO grafting was inconsistent. The beneficial effect of the PEO grafting was most pronounced on the PEO-grafted nitinol stents. Thrombus formation was reduced by more than 85% by PEO grafting on metallic stents. Only moderate improvement (i.e. 35% decrease in platelet deposition) was observed with PEO-grafted tubes of polyethylene, silicone rubber, and glass. For PEO-grafted heart valves made of Dacron, however, no effect of PEO grafting was observed at all. It appears that the extent of thrombus formation on PEO-grafted biomaterials was directly related to the extent of tissue damage during implantation surgery. Platelets can be activated and form aggregates in the bulk blood, and the formed platelet aggregates may be able to deposit on the PEO monolayer overcoming its repulsive property. Our studies indicate that the testing of in vitro platelet adhesion should include adhesion of large platelet aggregates, in addition to adhesion of individual platelets. Furthermore, the surface modification methods should be improved over the current monolayer grafting concept so that the repulsive force by the grafted PEO layers is large enough to prevent adhesion of platelet aggregates formed in the bulk blood before arriving at the biomaterial surface.     

Inorganic coatings for cardiovascular stents: In vitro and in vivo studies

The in-vitro and in-vivo biocompatibility of two oxides (TiO and ZrO) and diamond-like carbon (D) coated stents has been assessed and compared with uncoated stainless steel (St) stents. In vitro studies demonstrated that both fibrinogen adsorption and platelet adhesion were significantly higher on D coating compared to those on oxide coatings and uncoated stainless steel. In addition TiO and ZrO coatings showed evidence of a minor inflammatory response and more complete endothelialization of the aorta than that seen around D coated and uncoated St stents. The resulting neointimal growth in the aorta with TiO, ZrO, and D coated and uncoated St stents, measured 8 weeks after stenting (the ratio of the neointima in the stented artery to the non-stented artery) was 1.03 + 0.28, 0.85 + 0.36, 1.78 + 1.26, and 1.15 + 0.56, accordingly. From the data obtained it could be concluded that the increased neointima measured around D-coated stents, may be due to both, the inferior haemocompatibility of the diamond-like carbon coating and mechanical instability of D coating observed in an in vivo environment.     

In vitro and in vivo evaluations of dacron velour and knit prostheses.

Dacron velour and knit prostheses were compared with respect to long-term patency when used as canine aortic implants. These grafts were rated, according to the average numerical value of six equally weighted criteria, in decreasing order of performance: velour I, velour II, circular knit III, velour IV, circular knit V, warp-knit VI, warp-knit VII and VIII, warp-knit IX and warp-knit X. In general, compacting and crimping methods using halogenated hydrocarbons produced the least successful grafts. Grafts of identical brand but compacted and crimped by different methods exhibited different patencies. No true healing of the grafts was observed. Layered deposits of inner fibrous capsule were formed at rates and thicknesses characteristic for each type of graft employed. A functioning Dacron femoral-popliteal bypass removed from a human after 9 months exhibited inner and outer fibrous capsules similar to those from canine implants.     

In vitro and in vivo studies of heparinized-collageno-elastic tubes

Heparin was covalently coupled to collageno-elastic grafts (CET) derived from lamb carotid arteries, by using the crosslinking agent 1-ethyl-3 (3-dimethyl-aminopropyl) carbodiimide (EDC). The collagenous grafts were pretreated with various aminating agents in order to enhance the number of available binding sites on the collagen surface. By varying the EDC/heparin weight ratio, the pH of the immobilization media, and the pretreatment agent, a global search pattern maximized heparin loading at 3.90 +/- 0.36 USP heparin/cm2 collagenous graft surface when the EDC/heparin ratio was 2:1 at a pH of 1.5 with 1 M hydroxylamine sulfate as the pretreatment agent. Heparinized CETs were superior to nonheparinized CETs by exhibiting both enhanced antiplatelet activity in using an in vitro differential recirculation reactor with chromium-51 tagged platelets and enhanced patency when interposed in canine carotid arteries. Both antiplatelet activity and patency duration for heparinized CETs were independent of heparin loading.     

In vivo and in vitro studies of thymulin in marginally zinc-deficient mice

Thymulin (or serum thymic factor, FTS-Zn), a well-defined thymic hormone previously shown to be a nonapeptide binding the metal zinc, was studied in mice subjected to a long-term marginally Zn-deficient diet. In spite of the absence of thymic atrophy, we observed a significant decrease in the serum levels of thymulin as early as two months after the onset of treatment. However, these levels could be consistently restored after in vitro addition of ZnCl₂. The analysis of thymuses from Zn-deficient mice showed that, despite the apparently normal network of epithelial cells, there was a progressive increase in the number of thymulin-containing cells (assessed by immunofluorescence with anti-thymulin monoclonal antibodies) that was already significant after two months of treatment. These results are in keeping with those of previous investigators, showing a specific, altered, thymic endocrine function following Zn deprivation. Nonetheless, our results strongly suggest that the nonactive Zn-deprived peptide is secreted under these experimental conditions. Furthermore, the fact that the augmented numbers of thymulin-containing cells were observed in the thymuses following a decrease in the peripheral thymulin (biologically active) brings further evidence for the existence of a feedback mechanism for the secretion of this hormone.     

In vitro and in vivo studies of macrophage functions in amebiasis.

Experimental intrahepatic inoculation of the gerbil with Entamoeba histolytica trophozoites was used as a model of liver amebiasis to study the cellular immune response elicited by the parasite. It was shown that abscess-derived macrophages (5 to 20 days old) were deficient in their capacity to develop a respiratory burst, to secrete and express membrane-bound interleukin-1-like activity, and to kill E. histolytica trophozoites as well as to respond to lymphokines in vitro. However, macrophages isolated from the spleen and peritoneal cavities from the same infected animals were not significantly down regulated in these functions. Splenocytes from infected gerbils were shown to develop a strong responsiveness to amebic antigen, whereas their response to concanavalin A was suppressed. Crude E. histolytica extracts or conditioned medium down regulated murine BALB/c macrophage accessory and effector cell functions in vitro in a manner similar to abscess-derived macrophages, whereas crude extracts of the nonvirulent E. histolytica-like Laredo strain did not. Our results indicate that intrinsic or secreted products or both from E. histolytica are actively regulating macrophage functions at the abscess site and can possibly mediate other immunoregulatory mechanisms at distant targets.     

In vivo andin vitro comparative studies of animal articular surfaces

The contours in the articular surfaces of rabbit and canine knee joints were measuredin vivo andin vitro using the scanning electron microscope (SEM), replication, and light-microscopy techniques. Using the light microscope, patterns of highlights of the order 10 μm were observed when the living joint-surfaces were exposed. Replicas were made of these surface contours using Xantopren, a light-bodied dental silicone impression-material. More than 50% of the replicas were discarded because of defects. In the SEM, a crazing phenomenom was sometimes evident on the replicated surfaces which may have been attributable to distortion of the silicone medium, possiblyin vacuo. The interpretable replicas from the living joint-surfaces contained 7 to 20 μm diameter humps that varied from 0.5 to 2.5 μm high. These corresponded to the patterns of highlights seen with the light microscope at surgery. Examination of the replicas takenin vitro and direct examination of the cartilage in the SEM revealed comparable data. Theliving articular-cartilage contours, therefore, appear to correspond to those previously identifiedin vitro. We emphasize that although these values give some indication of the anticipatedin vivo contours, the actual contours in a load-carrying situation may differ.     

In vitro and in vivo anti-coagulant activity and toxicological studies of marine sulfated glycosaminoglycans

The present study aimed to characterize and evaluate the in vitro and in vivo anticoagulant activity of sulfated glycosaminoglycans from the skins of smooth hound (SHSG) and grey triggerfish (GTSG). The analysis of SHSG and GTSG with acetate cellulose electrophoresis in Zn-acetate revealed the presence of hyaluronic acid (HA), chondroitin sulfate (CS) and dermatan sulfate (DS). Both glycosaminoglycans were evaluated for their in vitro anticoagulant activities using activated partial thromboplastin time (aPTT), thrombin time (TT) and prothrombine time (PT) tests. SHSG and GTSG and calciparin were tested as in vivo anticoagulants by subcutaneous (s.c) injection to adult female Wistar rats in a concentration of 75 mg/kg of body weight. The administration of SHSG, GTSG and calciparin to rats induced a significant decrease of platelet rates compared to the control. The aPTT assay of SHSG and GTSG was prolonged 1.3 and 1.23-fold respectively compared with the control. Toxicity studies were performed to investigate whether or not SHSG and GTSG can cause pathological changes in the liver, proteins and DNA. The concentration and catalytic activity of liver oxidative stress markers and enzymes, respectively, as well as the observed hepatic morphological changes indicated that calciparin induced hepatic toxicity and oxidative damage in the liver. The higher activity and lower toxicity of SHSG and GTSG recommended these compounds as a better drug candidate than calciparin.     

In vitro and in vivo studies on biodegradable CaMgZnSrYb high-entropy bulk metallic glass

In order to enhance the corrosion resistance of the Ca₆₅Mg₁₅Zn₂₀ bulk metallic glass, which has too fast a degradation rate for biomedical applications, we fabricated the Ca₂₀Mg₂₀Zn₂₀Sr₂₀Yb₂₀ high-entropy bulk metallic glass because of the unique properties of high-entropy alloys. Our results showed that the mechanical properties and corrosion behavior were enhanced. The in vitro tests showed that the Ca₂₀Mg₂₀Zn₂₀Sr₂₀Yb₂₀ high-entropy bulk metallic glass could stimulate the proliferation and differentiation of cultured osteoblasts. The in vivo animal tests showed that the Ca₂₀Mg₂₀Zn₂₀Sr₂₀Yb₂₀ high-entropy bulk metallic glass did not show any obvious degradation after 4 weeks of implantation, and they can promote osteogenesis and new bone formation after 2 weeks of implantation. The improved mechanical properties and corrosion behavior can be attributed to the different chemical composition as well as the formation of a unique high-entropy atomic structure with a maximum degree of disorder.     

In vitro and in vivo studies for modified ethyl cyanoacrylate regimens for sclerotherapy

Cyanoacrylates have known for their ability to polymerize rapidly in the presence of traces of weakly basic moieties such as water. The tissue adhesive, Histoacryl(R) (N-butyl 2-cyanoacrylate), has been reported to control bleeding through endoscopic sclerotherapy. But the commercially available Histoacryl(R) is expensive, and it has the problem like other cyanoacrylates that the glue tends to flow/run away from the point of application, which is inherent to the low viscosity, making precise application difficult. In this study, ethyl cyanoacrylate (ECA), the main constituent of "super glue," was employed instead of Histoacryl(R) due to its lower cost. The aim of the research is to modify the compositions of ECA regimen and evaluate its feasibility for sclerosant application through both in vitro flow circuit model and in vivo animal tests. It was noted that the difference in the relative hardening rate between the in vitro Hepes-Tyrodes buffer flowing model and the in vivo rat model for the ECA and Histoacryl(R) was related to the existence of the blood protein, such as albumin, in the physiological milieu. It was also noticed that the ECA setting rate was greatly increased either in Hepes-Tyrodes buffer or in blood (to a comparable rate as Histoacryl(R) in vivo) by adding a few doses of caffeine, which acts as a polymerization initiator. This would lead to far better injection precision during sclerotherapy. Furthermore, in vivo histological examination for the occluded lumen of the rat's inferior vena cava and a clinical piglet portal vein occlusion experiment have suggested this new sclerosant regimen, caffeine/ECA, is of great promise in endoscopic sclerotherapy.     

In vivo evaluation of modified mandrel-grown vascular prostheses.

The Omniflowtrade mark Vascular Prosthesis (OVP) has been manufactured and extensively tested in animal and human trials. It has mechanical and biological qualities superior to synthetic and biological conduits, particularly in low flow conditions. For further development into the smaller diameter coronary prostheses, the inner luminal surface is of paramount importance. In a previous study this inner surface was modified to produce a more uniformly thicker nonundulating surface. In this study the mandrels of these modified OVPs were treated with either collagen or heparin; the OVPs were evaluated for patency, tissue integration and wound healing, and endothelialization using a dog model comparable to that used to evaluate the unmodified OVP. In all instances, each of the modified prostheses were fully patent and had no signs of any deleterious effects caused by these modifications; no thrombus or aneurysms were visible. The tissue response was rapid with excellent new host collagen deposition within the vessel wall and minimal inflammatory and foreign body giant cells. Endothelialization was noted at the earliest explant time point in central regions of the prostheses, albeit that the histological picture at this time point appeared to reflect a complex atypical intimal layer.     

In vivo and in vitro studies of plaque type mutants of an RNA virus

Summary Six plaque types all belonging to the same antigenic group (E₅₄) of vesicular exanthema of swine virus (VESV) were studied for mutational changes in swine and in pig and canine cell cultures. The plaque types differed in virulence, the largest in size being the most virulent and the smallest the least virulent type. Neutralization tests revealed that only animals which showed frank disease developed specific viral antibodies. The inoculated type was generally mixed with other plaque types after recovery from the host. The contaminating new types were not the same for the different plaque type inoculations. Similar changes were observed in the two types of cell cultures. Since all inocula consisted of virus from single cell yields which were clonal in origin, it was concluded that mutational changes occurredin vivo andin vitro. Such mutations, together with the requisite selective forces, could lead to erroneous classification of a variant in regard to virulence.This work was sponsored by the Office of Naval Research under the terms of a contract with the Regents of the University of California. Reproduction in whole or in part is permitted for any purpose of the United States Government.     

In vivo and in vitro studies on possible pathogenic mechanisms of Actinomyces viscosus.

Actinomycotic infections are characterized by long-term inflammatory lesions containing large numbers of polymorphonuclear leukocytes (PMNs) and mononuclear cells. The pathogenic mechanisms involved in these lesions are not understood. Homogenates of Actinomyces viscosus (AVIS) induce an acute inflammatory response with a predominance of PMNs within 6 h after injection into the footpads of nonimmunized mice. These homogenates, when tested in vitro, contain potent chemotactic activity for human PMNs. In vitro chemotactic activity for human monocytes is weak but statistically significant (P less than 0.025). Doses of AVIS, which alone have little chemotactic activity, cause the generation of PMN chemotactic activity in fresh, but not complement-inactivated, serum. The injection of AVIS into the footpads of immunized mice induces an acute inflammatory response followed within 48 h by a mononuclear cell infiltrate, suggesting that factors affecting monocyte accumulation are generated by the immune host in response to challenge with the bacterial antigens. These findings indicate that the pathogenicity of the Actinomyces may result in part from (i) their direct chemotactic effect on PMNs, (ii) their cytotaxigenic effects on serum, and (iii) their ability to stimulate host immune cells to produce and release mediators of inflammation.     

Inhibition of tumor growth induced by histamine: In vivo and in vitro studies

An experimental mammary carcinoma was induced in rats by the i.p. administration of N-nitroso N-methylurea (NMU) in 3 doses of 50 mg/kg. In order to study the role of histamine (Hi) and its receptors in tumor growth,in vivo andin vitro treatments were carried out. Different groups of animals received a daily s.c. injection of Hi 0.1 mg/kg, starting with the first dose of NMU. Hi significantly reduced tumor incidence and the number of tumors developed per rat. Thein vitro studies, using the clonogenic agar technique, showed no difference in colony formation between control, 0.01 and 0.1 M Hi treatment, while 1 and 10 M Hi concentrations induced inhibition of cell growth up to 60%. This effect was abolished by H₁ antagonists. Conversely, the action of the H₂ antagonists was a significant inhibition of colony formation. We may then conclude that in these experimental tumors, histamine exerts a regulatory function on cell growth by acting directly on specific membrane receptors.     

In vitro andin vivo studies of radiographic contrast media-induced histamine release in pigs

Routine clinical use of radiographic contrast media (RCM) causes adverse reactions in some patients. To elucidate the mechanisms of these reactions bothin vitro andin vivo studies are necessary. In this study, RCM-induced histamine release from isolated mast cells was compared with thein vivo release of histamine and cardiovascular symptoms using a porcine model. The 2 non-ionic preparations examined (Solutrast^® and Ultravist^®) released little or no histamine from the 4 cell types tested (porcine pulmonary, cardiac, hepatic, and renal mast cells). The 4 ionic preparations (Angiographin^®, Hexabrix^®, Rayvist^®, and Telebrix^®) caused histamine release from most of the cell suspensions. In almost all cases, the cardiac mast cells were the most sensitive followed by the hepatic mast cells. All 4 RCM testedin vivo produced elevated plasma histamine levels in some animals. The highest incidence was observed using the ionic, high osmolal Rayvist^® (6 of 12 animals), followed by the non-ionic RCM with the lowest osmolality Ultravist^® (4 of 12 animals).In vivo, mechanisms in addition to direct histamine release may also be involved in RCM-induced adverse reactions, since low osmolal, non-ionic RCM can cause elevated plasma histamine levels withoutin vitro release. The susceptibility of cardiac mast cells to RCM-induced histamine release suggests that patients undergoing e.g. coronary angiography may be especially at risk for an adverse reaction.