Immunohistochemical detection of sepsis-induced lung injury in human autopsy material

Abstract

This review addresses our present-day knowledge on the role of different cellular adhesion molecules, cytokines and glycoproteins for the detection of sepsis-induced injury in the microvasculature of the human lung using immunohistochemistry. Through the induction and modulation of endothelial cell adhesion molecules, such as E-selectin (CD 62E), the vascular endothelium controls leukocyte extravasation into tissue. E-Selectin, not expressed by unstimulated endothelium, is activated by cytokines and initiates neutrophil recruitment in sepsis-induced lung injury. Since E-selectin is strongly expressed in the pulmonary microvasculature in sepsis-associated fatalities, the immunohistochemical detection of an intense expression of E-selectin in lung tissue is a valuable diagnostic tool in the forensic postmortem elucidation of death due to sepsis. VLA-4 (CD49d/CD29) is strongly expressed on intravascular, interstitial and intra-alveolar leukocytes in sepsis-associated fatalities, whereas in non-septic fatalities an irregular weak immunoreactivity can be observed on interstitial leukocytes and no positive immunohistochemical expression can be observed on intravascular or intra-alveolar leukocytes. ICAM-1 (CD54) is strongly expressed on endothelial cells of the pulmonary microvasculature and on pulmonary macrophages and lymphocytes in sepsis-associated fatalities. In contrast, an infrequent weak immunohistochemical reaction for ICAM-1 is found on pulmonary endothelium and on perivascular leukocytes in non-septic fatalities. The up-regulation of both cellular adhesion molecules can be considered as an useful immunohistochemical postmortem marker of sepsis. Lactoferrin (LF) is an iron-binding glycoprotein located in specific (secondary) granules of leukocytes and plays a central role in the host response to infectious stimuli in providing both bacteriostatic and bactericidal protection. There is a statistically significant association between an enhanced expression of LF on pulmonary leukocytes in sepsis-related fatalities in contrast to non-septic controls. The immunohistochemical detection of an enhanced expression of LF can contribute to the postmortem discrimination between sepsis and non-septic fatalities. Application of carbohydrate-specific lectins (ConA, UEA, GSA I, GSA II, MPA, PNA, Jac, WGA, MAA, LPA, SNA) on deparaffinated lung tissue sections from sepsis-associated fatalities and control cases results to some extent in different staining patterns of alveolar epithelial cells and subepithelial seromucous glands of the bronchi. Apart from differences in binding sites for alpha-mannose, N-acetyl-neuraminic acid and alpha-(2-6)-galactose (as detected by different expression for ConA, MAA and SNA), the main finding is that no binding sites for alpha-N-acetyl-galactosamine (as investigated by MPA immunoreactivity) can be detected on alveolar epithelial cells and mucous parts of subepithelial seromucous glands in sepsis cases in contrast to the presence of such binding sites in controls. Since most intracellular pathogens persist in macrophages and epithelial cells during infection, it is likely that these pathogens contribute to a continual deprivation or consumption, respectively, of glycoproteins physiologically secreted by alveolar epithelial and glandular cells at different time points and stages of infection and may, among other mechanisms, by reducing pathogen clearance amplify the inflammatory response. Vascular endothelial growth factor (VEGF), an angiogenic and chemotactic peptide, is abundantly expressed in normal lung tissue, especially in alveolar and bronchial epithelium, glandular cells of the bronchi, and activated alveolar macrophages. Pulmonary VEGF immunostaining differs in sepsis when compared to healthy individuals. In the latter a preponderant strong VEGF immunoreaction can be found on alveolar epithelium (predominately type II pneumocytes), bronchial epithelium and glandular cells of the bronchi and bronchioli, and activated alveolar macrophages. In contrast, in sepsis no VEGF immunopositivity can beivity can be observed on bronchial epithelium or glandular cells of the bronchi and bronchioli, and no or relatively sparse VEGF immunoreactivity is found on alveolar epithelial cells. The precise mechanisms of the decreased pulmonary VEGF expression in septic patients under conditions of intensive care medicine are not clear at present. During the complex cascade of excessive pro-inflammatory and anti-inflammatory mediator release involved in the host's systemic inflammatory response in the development of sepsis-induced lung injury, VEGF expression may be suppressed in sepsis by a hitherto not identified agent or the interaction of different mediators of cellular inflammation. For the detection of sepsis-induced lung injury the aforementioned markers can be used sufficiently, e.g. to give immunohistochemical evidence of a previously undiagnosed sepsis and to confirm or rule out a presumed diagnosis of a sepsis-associated fatality. The employment of the presented immunohistochemical methods will be particularly helpful when macroscopical and routine histological autopsy findings in cases of suspected fatal sepsis are unspecific or unconvincing, respectively, and clinical data on the patient's previous history are not available. Referring to the forensic argumentation regarding causality on the subject of possibly fatal septic complications, e.g. in the sequel of diagnostic or therapeutic iatrogenic injection procedures or being relevant to pressure sore-associated fatalities, aetiopathogenetic conclusions can be optimized on the basis of the described micromorphological investigations.