Lymphangiogenesis is a novel prognostic parameter for several cancers that is preferentially quantified by immunohistochemistry of the lymphatic endothelium-specific hyaluronan receptor LYVE-1. Recently, the specificity of LYVE-1 was challenged by serendipitous observations of LYVE-1 expression in rare tissue macrophages. As expression of the hyaluronan receptor-like molecule stabilin-1 is shared by sinusoidal endothelium and macrophages, a thorough analysis of LYVE-1 expression was performed using macrophage-specific markers in vivo and in vitro. In murine tumour models and excisional wound healing, LYVE-1 expression occurred in a subset of CD11b(+), F4/80(+) tissue macrophages that preferentially co-expressed stabilin-1. Upon comparison of single- and double-labelling immunofluorescence, it became apparent that LYVE-1(+) macrophages mimic sprouting and collapsed lymphatic vessels. In vitro, LYVE-1 expression was induced in 25-40% of murine bone marrow-derived macrophages upon exposure to B16F1 melanoma-conditioned medium and IL-4/dexamethasone. By FACS analysis, 11.5% of bone marrow-derived macrophages were LYVE-1(+), stabilin-1(+) double-positive, while 9.9% were LYVE-1(+), stabilin-1(-) and 33.5% were LYVE-1(-), stabilin-1(+). Northern and western analyses confirmed expression of LYVE-1 mRNA and protein in bone marrow-derived macrophages. In the light of the current debate about true endothelial trans-differentiation versus endothelial mimicry of monocytes/macrophages, LYVE-1(+), stabilin-1(+) non-continuous endothelial-like macrophages will require further developmental and functional analyses. In conclusion, the findings imply that LYVE-1 staining must be supplemented by double labelling with macrophage markers in order to differentiate clearly between LYVE-1(+) lymphatics and LYVE-1(+) tumour-infiltrating macrophages. This improved approach will help to clarify the prognostic significance of lymphangiogenesis in malignant tumours.

SummaryGliotoxin is a secondary metabolite produced by several fungi including the opportunistic human pathogen Aspergillus fumigatus. As gliotoxin exerts immunosuppressive effects in vitro and in vivo, a role as a virulence determinant in invasive aspergillosis has been discussed for a long time but evidence has not been provided until now. Here, by the use of different selection marker genes A. fumigatus knock-out strains were generated that are deficient for the nonribosomal peptide synthetase GliP, the putative key enzyme of the gliotoxin biosynthesis. Deletion of the gliP gene resulted in loss of gliotoxin production, as analysed by high performance liquid chromatography and tandem mass spectrometry. No differences in morphology or growth kinetics between wild-type and gliP-deletion strains were observed. In vitro, the culture supernatant of the gliP-deficient strains showed a reduced cytotoxic effect on both macrophage-like cells and T cell lines. In a low-dose murine infection model of invasive aspergillosis, gliotoxin was detected in the lung and absent when mice were infected with the gliP deletion strain. However, gliP deletion strains showed no difference in virulence compared with the corresponding wildtype strains. Taken together, the non-ribosomal peptide synthetase GliP is essential for gliotoxin production in A. fumigatus. Gliotoxin is not required for pathogenicity of the fungus in immunocompromised mice, despite the fact that a reduced cytotoxicity of the culture supernatant of gliP deletion strains was demonstrated.

We used a novel approach for the direct ex vivo identification and characterization of T cell epitopes based on the screening of peptide spot libraries with freshly isolated splenocytes in a sensitive enzyme-linked immunospot (ELISPOT) assay. This technique was applied for the analysis of splenocytes from Listeria monocytogenes-infected BALB/c and C57BL/6 mice. The screening of peptide spot libraries covering the whole listeriolysin O and p60 of L. monocytogenes confirmed all known CD4 and CD8 T cell epitopes of these proteins and additionally revealed six new H-2d and six new H-2b-restricted T cell epitopes. New epitopes were categorized into CD4 and CD8 T cell epitopes by ex vivo ELISPOT analysis with separated T cell populations. The quantitative analysis of cells reactive with these CD4 and CD8 T cell epitopes revealed the existence of dominant and subdominant CD4 and CD8 T cell populations during L. monocytogenes infection. As a consequence of these data we suggest that ELISPOT-based screening of peptide spot libraries could be a general approach for the rapid identification and characterization of pathogen-specific T cell populations during various infectious diseases.