Experimental and clinical allogeneic heart transplant rejection: correlations between histology and immune reactivity detected by cytokine messenger RNA.

BACKGROUND: Cytokines produced by host cells infiltrating allogeneic transplants are critical determinants of graft rejection but information on cytokine production during graft rejection remains limited. No reported study on cytokine profiles has compared experimental allograft rejection induced by withdrawal of cyclosporine with clinical transplant rejection that occurs in the presence of therapeutic levels of cyclosporine. METHODS: Functional activities of allograft-infiltrating host cells in sequential endomyocardial biopsies obtained before, during, and after acute heart transplant rejection were determined with the use of the reverse transcriptase-polymerase chain reaction to detect cytokine messenger RNA. These results were correlated with histologic findings in both an experimental canine model of heart transplant and rejection and in clinical human heart transplant recipients. RESULTS: When experimental rejection was induced by withdrawal of immunosuppression, rejection was characterized by the presence of mRNA encoding CD4, CD8, interleukin-2 (but not interleukin-4), interleukin-2 receptor, and tumor necrosis factor-beta. These findings are consistent with a classic T-helper, T-cytotoxic cell-mediated response. However, the cytokine profile of human, clinical heart transplant rejection occurring in the presence of therapeutic levels of immunosuppression differed strikingly. In clinical rejection in human beings, histologic evidence of rejection was not associated with detectable interleukin-2 or interleukin-2 receptor mRNA. CONCLUSIONS: Human, clinical heart rejection can occur in the absence of locally produced interleukin-2; the degree of immunosuppression achieved with cyclosporine A may explain the different results obtained in the canine withdrawal model versus human clinical allograft rejection.

Inhibition of atherosclerosis by dietary pectin in microswine with sustained hypercholesterolemia.

Sustained hypercholesterolemia is a known risk factor for development of atherosclerosis. In animal studies, grapefruit pectin fed concurrently with a high-lipid diet inhibits hypercholesterolemia and atherogenesis. The purpose of the present study was to determine if grapefruit pectin affects cholesterol levels and atherogenesis of animals with established hypercholesterolemia. Microswine were fed an atherogenic diet to establish hypercholesterolemia. Plasma cholesterol levels rose rapidly and for 360 days were sustained at levels 6- to 12-fold the normal level. Then, half the microswine, selected at random, were fed a diet in which 3% grapefruit pectin was substituted for cellulose, and the remaining animals received the original diet. Animals were killed 270 days later, and the extent of atherosclerosis was determined. In animals with established hypercholesterolemia, pectin did not lower their cholesterol levels. However, pectin reduced the extent of atherosclerosis in both the aorta and coronary arteries. The mean surface area covered by atherosclerosis in the aorta was 13.6% in the group that did not receive pectin compared with 5.3% in the group that did receive pectin. The mean coronary artery narrowing was 45% without pectin and 24% with pectin. We conclude that pectin may have a direct beneficial effect on atherosclerosis by a mechanism independent of cholesterol levels.

Modification of atrioventricular conduction using a combined laser-electrode catheter.

Ablation of the AV junction is an accepted technique for the management of selected supraventricular tachyarrhythmias. Radiofrequency ablation appears to be safe and effective for AV junction ablation in most patients, but the need for firm tissue contact may make it less effective for ventricular tachycardia and certain ectopic/atrial tachycardias. Laser energy can also be delivered through a catheter, and thus it may be an attractive alternative energy source for ablation. A new laser-electrode catheter was developed for modification of conduction through the AV node as a model for ablation of an arrhythmia substrate. A window for delivery of continuous-wave Nd:YAG laser energy was placed between the two electrodes of a bipolar electrode catheter. In vitro studies using a matrix of power versus time were performed to determine the energy that would create lesions of the appropriate size in vivo. Using this information, advanced AV block was successfully created in 16 of 17 dogs (94%) with the laser-electrode catheter. Advanced AV block was successfully created in all four dogs in the chronic study, and it persisted for 1-24 weeks of follow-up until sacrifice of the animals. Histologic examination demonstrated discrete thermal damage at the AV junction with no instances of septal perforation in the acute studies or progressive necrosis in chronically maintained dogs. Advanced AV block may be produced consistently and safely in dogs using a combined laser-electrode catheter.

Origin of cells in the coronary intima during acute vascular rejection of the transplanted human heart.

Acute vascular rejection of the coronary arteries in human heart transplants is characterized by a lymphocytic infiltrate and a thickened intima that contains numerous highly vacuolated cells. The origin of the vacuolated cells has been controversial. In this immunocytochemical and electron microscopic study of four patients with acute vascular rejection, the predominant cells in the coronary artery intima were host-derived lymphocytes and highly vacuolated smooth muscle cells. Lymphocytic infiltrates were composed of T cells with variable numbers of B cells. Macrophages were infrequent. Smooth muscle cells were identified by their reactivity to muscle-specific actin and ultrastructural features of a peripherally displaced elongated nucleus associated with abundant myofibrils. In addition, the vacuolated cells did not react with endothelial factor 8-related antigen or Ulex europaeus agglutinin, as would be expected of endothelial cells. The cytoplasmic vacuoles present in the smooth muscle cells appear to be swollen endoplasmic reticulum containing watery fluid consistent with the hypothesis that they result from altered ion movement across the plasma membrane in response to cellular injury.

In vitro laser recanalization of chronically occluded prosthetic grafts.

Polytetrafluoroethylene (PTFE) and Dacron grafts were implanted in canine femoral and carotid arteries using PTFE and Prolene suture, respectively. Arteries containing occluded grafts were explanted and laser recanalization was attempted in vitro. Laser recanalization was successful in 78% of PTFE grafts compared to 30% of Dacron grafts. Recanalization was complete (residual stenosis less than 5%) in opened PTFE grafts, whereas residual stenosis averaged 60% in recanalized Dacron grafts. PTFE graft/PTFE suture anastomotic tensile strength was unchanged after recanalization, while Dacron graft/Prolene suture anastomotic tensile strength decreased significantly. In addition, anastomotic bursting pressure was significantly higher for lased PTFE grafts with PTFE sutures (300 mg Hg) compared to lased Dacron grafts with Prolene sutures (70 mm Hg). Chronically occluded PTFE grafts with PTFE suture can be safely and effectively opened by laser recanalization. In contrast, attempted laser recanalization of Dacron grafts sutured with Prolene suture is seldom successful, significantly weakens the graft artery anastomosis, and should be avoided.

Acute vascular rejection of the coronary arteries in human heart transplantation: pathology and correlations with immunosuppression and cytomegalovirus infection.

Prior studies of vascular rejection in transplanted human hearts have stressed the importance of accelerated coronary arteriosclerosis (chronic vascular rejection). We, however, have had four patients with sudden onset of acute heart failure within 90 days of transplantation who have died without significant myocardial interstitial rejection or the concentric intimal thickening with dense collagen that is typical of chronic vascular rejection. In contrast, the coronary arteries in our patients had a prominent lymphocytic infiltrate, a loosely organized intimal thickening composed of smooth muscle cells, and extensive endothelial injury. We believe that these changes define acute vascular rejection of the coronary artery. In 14 transplanted hearts obtained consecutively, at autopsy or at a second transplant procedure, graft failure was caused by acute coronary vascular rejection in six cases and by chronic coronary vascular rejection in one case. The remaining seven patients showed no evidence of vascular rejection and died primarily of sepsis. Cytomegalovirus (CMV) disease was present in 6 of 7 patients with vascular rejection, of which 43% were CMV-negative recipients of hearts from CMV-positive donors. The adoption of a triple-drug protocol, in which azathioprine was added to cyclosporine and prednisone, reduced the incidence of acute vascular rejection from 27% to 8%. We conclude that acute coronary vascular rejection may be initially seen as global cardiac ischemia in the absence of significant interstitial myocardial rejection. Further, acute vascular rejection should be pathologically distinguished from chronic vascular rejection, although both are probably stages in the natural history of immune-mediated vascular injury.

Annular subaortic aneurysm resulting in sudden death.

A subvalvular aneurysm in a 34-year-old black woman developed beneath the left coronary cusp of the aortic valve and extended in the epicardium between the aortic root and left atrium. She was asymptomatic until she expired suddenly from myocardial ischemia caused by compression of the circumflex coronary artery by the aneurysm. A review of the literature reveals 13 reports of subaortic and 58 submitral aneurysms but only two instances of myocardial infarction secondary to coronary artery compression by spontaneous subvalvular aneurysms in either the subaortic or submitral position.

Direct laser and laser-thermal irradiation of normal canine coronary arteries: implications for laser delivery methods.

Clinical use of laser recanalization has been mostly limited by arterial perforation. Two modifications of the optical fiber were used to decrease the perforation rate: (1) a guidewire to align the fiber in a coaxial position with the vascular lumen and (2) encapsulation of the optical fiber tip with a metal cap. Fourteen dogs were studied. Argon laser radiation was delivered through optical fibers, 9 with a 1.7-mm metal encapsulated fiber (Laserprobe-PLR,¿ Trimedyne, Santa Ana, CA) and 5 with a 2-mm metal encapsulated fiber with a window at the tip (Spectraprobe-PLR). Three dogs served as control, and no guidewire was used. In 11 dogs, laser irradiation was done either in advancing the probe or on pullback over a guidewire. Energy used ranged from 25 to 50 J. In vitro, this raised the probe temperature in blood to a maximum of 500 degrees C. Arterial perforation occurred in 2 of 4 lased arteries without guidewire (p less than 0.05). There was no difference in perforation rated comparing the Laserprobe (1 out of 17 arteries) and Spectraprobe (1 out of 8 arteries) (p greater than 0.05). At high laser energy (50 J), there was an increased incidence of thrombus formation, which appeared to be associated with adherence of the metal cap to the arterial wall (3 out of 6 vs. 1 out of 19, p less than 0.05).(ABSTRACT TRUNCATED AT 250 WORDS)

Transvascular argon laser ablation of atrioventricular conduction in dogs: feasibility and morphological results.

The purpose of this study was to ablate atrioventricular (AV) conduction in dogs with an argon laser using a transvascular approach. Six dogs were anesthetized and underwent a right thoracotomy and atriotomy. A bipolar #7 French lumen catheter containing a 400 u silica fiber was used to map the region of the AV junction. When a stable His deflection was obtained, the silica fiber was extruded from the end of the catheter and argon laser radiation was delivered for up to 20 s at 3.0-4.5 watts. In five of six dogs, complete heart block was successfully created after 3 to 15 lasings. A continuous His bundle electrogram in one dog showed gradual prolongation of the HV interval before the development of complete infrahisian block. The mean R-R intervals in the five dogs with complete heart block increased from 435 +/- 56 ms to 1216 +/- 197 ms (P less than 0.001), and the QRS duration increased from 78 +/- 26 ms to 91 +/- 19 ms (P = ns). Gross inspection showed multiple 1-2 mm2 craters in the atria just above the tricuspid valve ring. Complications included one instance each of aortic root perforation and creation of a small ventricular septal defect. These lesions were so small that there was no hemodynamic compromise in either animal in which they occurred and the lesions themselves were only detected on careful postmortem examination. Histology revealed two patterns of injury involved in conduction ablation. One was direct vaporization of tissue resulting in a wedge shaped incision and the other was formation of a hematoma and thermal necrosis at the lasing site with little evidence of tissue vaporization. Although catheter modifications are necessary to avoid perforation, this study demonstrates that ablation of conduction can be successfully accomplished using an argon laser.

Host endocrine responses during tumor growth.

Transplantation of the EL-4 lymphoma to syngeneic recipients caused significant endocrine changes which occurred very early as well as late after transplantation. Among the hormonal changes induced were a biphasic increase in the level of serum corticosterone, a biphasic decrease in serum insulin levels, an early decrease in prolactin and a terminal severe deficiency in thyroxine. The mechanism by which corticosterone levels are increased immediately following tumor transplantation appears to involve post-thymic T cells. In addition, the corticosterone response after tumor transplantation seems to be restricted to syngeneic recipients and does not seem to occur with allogeneic tumor transplantation. Further, the phenomenon may require an immunogenic tumor since the relatively nonimmunogenic mammary tumor virus (MTV) induced adenocarcinoma did not increase corticosterone in syngeneic C3H/He mice. Such data are consistent with the proposition that recognition of tumor antigen by mature T cells occurs within hours of tumor transplantation. This recognition appears to be MHC restricted. Whereas mitogen stimulation of T cells produces a glucocorticoid increasing factor designated GIF (Besedovsky et al., 1985b), it is reasonable to suggest that GIF is produced in vivo as part of the T cell response to tumor antigen. GIF in turn stimulates hypophyseal release of ACTH with a subsequent release of corticosterone from the adrenal gland. The biological relevance of this physiological increase in serum levels of corticosterone was examined with respect to the anti-inflammatory phenomenon often observed after tumor transplantation. First, a concordance was noted following tumor transplantation between elevated corticosterone levels and anti-inflammation. Similarly, transplantation of the MTV induced mammary adenocarcinoma which failed to increase serum levels of corticosterone did not exhibit anti-inflammation. Consistent with the concept that corticosterone levels increase following T cell recognition of tumor antigens, it is known that anti-inflammation does not occur with weakly immunogenic tumors but does follow transplantation of moderately immunogenic tumors (Normann, 1985b; Normann et al., 1985a). Second, adrenalectomy prevented the corticosterone response to tumor transplantation and eliminated tumor associated anti-inflammation. Additional studies are necessary to determine if the increase in serum levels of corticosterone alters other parameters of the host response to tumors. Anti-inflammation was shown to occur following tumor transplantation via a corticosterone dependent pathway.(ABSTRACT TRUNCATED AT 400 WORDS)

Isolation of a low molecular weight inhibitor of lymphocyte proliferation from tumorous ascites.

Intraperitoneal growth of P-815 mastocytoma cells in syngeneic DBA/2 mice produces ascites fluid which strongly inhibits mitogen-stimulated lymphocyte proliferation. The less than 10,000 m.w. fraction from gel filtration chromatography of tumorous ascites on Sephadex G-150 showed no inhibition of proliferation when eluted under physiologic conditions but was inhibitory when eluted with a high ionic strength, acidic buffer. The organic phase of a chloroform/methanol extract of the low m.w. fraction contained all the inhibitory activity. Purification of the inhibitor to relative homogeneity was achieved by reverse phase, HPLC with a gradient of acetonitrile in dilute acetate buffer. Inhibitory activity eluted between 30 and 35% acetonitrile. The active fraction contained less than 30 pg/ml PGE by RIA which was insufficient to inhibit proliferation and may actually have been stimulatory. Inhibition comparable to that produced by the ascites fraction required greater than 300 pg/ml of PGE. This low m.w. (less than 10,000), lipid-like inhibitor of lymphocyte proliferation is acid stable, not sensitive to proteolytic enzymes, soluble in both aqueous and organic solvents and occurs normally bound to a higher m.w. carrier molecule.

Distribution and survival of passively transferred lymphoblasts in normal and tumor-bearing mice.

Although most circulating lymphocytes are long-lived small lymphocytes, it is the recently-divided medium and large-sized lymphoblasts that most clearly migrate to sites of inflammation, and they do so without regard to immunologic commitment. The tumor-bearing state frequently suppresses macrophage accumulation at inflammatory sites but it is not known whether lymphocyte traffic is similarly affected. We addressed this issue using passively transferred radiolabelled lymphoblasts. Lymphoblasts readily entered s.c. fibrosarcomas transplanted to C57BL/6 mice. Large and small tumors acquired the same number of lymphoblasts per gram of tumor, indicating the absence of an anti-inflammatory reaction. However, the intra-tumoral survival of lymphoblasts was shorter in large than in small tumors. Shortened survival was noted also in spleen and whole body of animals bearing tumors. Shortened survival rather than impaired emigration may contribute to the increase in tumor:lymphocyte ratio observed with large tumors. The adverse effect of tumor bearing on lymphoblast survival may be limited to tumors which acquire a high percentage of passively transferred lymphocytes since intraperitoneal P-815 mastocytomas in DBA/2 mice acquired few labelled lymphoblasts and did not alter lymphoblast survival. These results are important for host resistance to tumor growth and therapeutically for the use of lymphocyte cytotoxic cells induced to proliferate in vitro.

The argon contact laser scalpel: a study of its effect on atherosclerosis.

A new argon laser scalpel (ALS) that delivers radiation to tissue by direct contact was used to perform endarterectomies on atherosclerotic rabbit aortas in vivo. The resultant effects were compared to those induced by CO2 laser (CO2) and conventional surgical endarterectomy (END) to determine whether this instrument might be useful in the treatment of occlusive cerebrovascular disease. Light microscopy of the treated aortic segments revealed significantly more atheroma removed and less damage to the underlying media in the ALS segments compared to the CO2 segments. Electron microscopy showed that the ALS surface and distal intima-media interface were smoother and more even than those of the CO2 or END groups. Prostacyclin synthesis, as measured by 6-keto-prostaglandin F1a levels, was significantly reduced in the ALS compared to the END and control segments. These results indicate that the ALS is superior to CO2 in performing open laser endarterectomies, but such treatment places the atherosclerotic blood vessel at greater risk for thrombotic complications during the early postoperative period that does surgical endarterectomy. It is conceivable that a contact laser may be useful in the smooth welding of the distal edge of an atheroma (i.e., during carotid endarterectomy) and for the transcatheter ablation of surgically inaccessible obstructions of the cerebral circulation.

Tumor cytokinetics in the presence of normal, alloimmune, or Bacillus Calmette-Guérin-activated host cells simultaneously assayed in vivo and in vitro.

Rates of tumor cell loss, replication, and growth were determined simultaneously for P-815 mastocytoma cells placed in culture or transplanted as a peritoneal ascites tumor. Cytokinetic parameters were concordant in vitro to in vivo for P-815 cells growing in the presence of syngeneic DBA/2 resident or proteose peptone-elicited macrophages and under allogeneic C57BL/6 nonimmune conditions. Under alloimmune conditions, measured parameters differed in vitro from in vivo but conclusions were consistent in that alloimmune host cells were cytolytic and cytostatic and caused tumor regression. In contrast, syngeneic Bacillus Calmette-Guérin-activated macrophages were cytolytic and cytostatic in vitro but not in vivo despite equivalent or greater effector to target ratios, presence or absence of endotoxin in the tumor inoculi, or changes in the injection schedule of Bacillus Calmette-Guérin. Similarly, Bacillus Calmette-Guérin-activated macrophages were cytolytic in vitro but not in vivo when admixed with tumor cells prior to injection into the leg. This study is the first simultaneously conducted cytokinetic analysis of a common pool of labeled tumor cells growing in vitro and in vivo using randomly selected mice as donors of host effector cells or as recipients of tumor transplantation. It demonstrates that activated macrophages which are cytolytic and cytostatic in vitro for P-815 cells may not function analogously in vivo in controlling tumor growth.

Differences in Fc receptor expression between small and large guinea pig monocytes.

Guinea pig monocytes have been separated into subpopulations differing in size, morphology, and cytochemistry. We report now that these peripheral blood monocyte subpopulations differ also in native Fc receptor expression. All freshly isolated small monocytes formed Fc rosettes with sensitized sheep erythrocytes, whereas Fc receptor expression was not detectable on large monocytes. However, Fc receptor expression developed rapidly in culture such that by 48 hours all monocytes were positive. This phenomenon was inhibited by 2-deoxy-D-glucose and cycloheximide, suggesting that Fc receptor expression depended upon glycolysis and protein synthesis. Large monocytes participated in antibody-dependent lysis of sheep erythrocyte targets. That large monocytes lacking Fc receptors were cytolytic in this assay was concordant with expression of Fc receptors during the 16-hour test period. We believe that the large number of lymphocytes present in the Fc receptor-positive small monocyte fraction interfered with their measurement of antibody-dependent cytotoxicity. We conclude that Fc receptor expression on guinea pig monocytes is heterogeneous within the peripheral circulation. These differences quickly diminish in vitro with the expression of Fc receptors by all monocytes. Additionally, antibody-dependent lysis of sensitized erythrocytes by guinea pig mononuclear cells is due to monocytes and not lymphocytes or Kurloff cells.

Population kinetic study of guinea pig monocytes and their subsets during acute inflammation.

Guinea pig peripheral blood monocytes exist in four subsets which differ in cytochemistry, function, rates of production, and circulatory time during steady state. These subsets are designated small, intermediate, large, and very large vacuolated monocytes. To address the question as to whether all subsets participate equally in the acute inflammatory response, population kinetics were performed on monocyte subsets isolated by counterflow centrifugation elutriation following a single intravenous pulse of tritiated thymidine given concurrently with an intraperitoneal injection of phytohemagglutinin as phylogistic agent. Inflammation reduced the cell cycle times of the precursors for all monocyte subsets, increasing their production. However, inflammation increased the number of precursors only for large monocytes. In addition, a reserve of exclusively large monocytes existed which appeared in the circulation within 3 hr of inflammation induction. The subsequent loss of large monocytes from the circulation exceeded their production in contrast to all other monocyte fractions. Over 92% of all monocytes entering the acute inflammatory site were large monocytes despite the fact that they constituted only 58% of monocytes under steady state conditions. Small, intermediate, and very large vacuolated monocytes were only minor participants in the acute inflammatory response. These results indicate heterogeneity in the monocyte subset response to acute inflammation.

Population kinetic study on the origin of guinea pig monocyte heterogeneity.

Population kinetic studies were performed on guinea pig peripheral blood monocyte fractions isolated by counter-flow centrifugation elutriation following a single in vivo pulse of tritiated thymidine. Labeled large monocytes (volume 317 micron3; relative distribution 49%; circulating half-life 5.7 hr; and production rate 17,000 cells/ml blood/hr) accumulated in the circulation more rapidly, had a faster turnover time, and were produced in greater numbers than small monocytes (volume 283 micron3; relative distribution 34%; circulating half-life 10.8 hr; and production rate of 6100 cells/ml blood/hr). The kinetic data do not support a maturational sequence of small into large monocytes. Intermediate monocytes (volume 300 micron3; relative distribution 11%; circulating half-life 18.2 hr) and very large monocytes (volume 354 micron3; relative distribution 6%; circulating half-life 36.5 hr) had production rates, respectively, of only 1200 and 320 cells/ml blood/hr. Maxima in the labeling index curve for small and large monocytes suggested a generation time of 24 hr while grain count analysis revealed that these two cell fractions were derived from a precursor population with similar numbers of reductive divisions. Grain count analysis of intermediate and very large monocytes revealed that these cells differed from both small and large monocytes. Our data support the concept that monocyte subsets exist in guinea pig peripheral blood with different kinetics of production and survival.

The healing process in normal canine arteries and in atherosclerotic monkey arteries after transluminal laser irradiation.

To evaluate the healing response of both normal and atherosclerotic arteries to laser radiation, 7 mongrel dogs and 4 hypercholesterolemic atherosclerotic monkeys underwent catheterization with a right Judkins catheter-optical fiber system designed to maximize arterial wall injury. Argon laser radiation was then delivered to the abdominal aorta and iliofemoral arteries. In the 11 animals, a total of 917 sites were irradiated in 33 arteries. Angiography did not reveal dissection or aneurysm formation; occlusive thrombosis was found in 2 arteries. Perforation of the arterial wall was a frequent complication. In animals killed between 1 hour and 4 days, light and electron microscopy of lased sites showed craters filled with a coagulum of blood and cellular debris with only a few adherent platelets. Healing occurred with a minimal inflammatory response and involved both fibroblasts and smooth muscle cells. Reendothelialization was seen in all animals killed between 7 and 14 days after lasing and was complete by 30 to 60 days. Within this period, no accelerated atherosclerosis was seen at lased sites in the hypercholesterolemic monkeys. It is concluded that transluminal lasing of normal arteries in dogs and in atherosclerotic arteries of monkeys is followed by healing and reendothelialization within a few weeks.

Antileukocyte activity I. Systemic inhibition of cellular emigration following local inflammation.

Inflammation in progress at one site decreases edema formation at a second and separate inflammatory focus. This clinically important phenomenon is known as counter-irritation. Since its effect on leukocyte responses has not been defined, we investigated in rats the systemic anti-inflammatory effect of local irritant injection on cellular emigration, particularly monocytes. Macrophage accumulation at a subcutaneous inflammatory site was severely depressed by prior intraperitoneal irritant injection despite continued macrophage accumulation in the peritoneal cavity and normal circulating monocyte levels. The phenomenon also existed in the peritoneal cavity to subcutaneously administered irritants and involved PMNs as readily as macrophages. Anti-inflammation occurred only when the counter-irritant was injected before or simultaneously with the measured inflammatory response while the degree and duration of inhibition depended upon the nature and amount of counter-irritant injected. These studies demonstrate that local inflammation inhibits leukocyte reactivity. Transplantation of syngeneic tumor but not normal cells also produced a depression in macrophage inflammatory responses. This inhibition differed from counter-irritation by not affecting granulocytes and by being transient despite tumor persistence.