Anja J. Taverne-Thiele

Differences in mucus and serum immunoglobulin of carp (Cyprinus carpio L.)

Electrophoretic analysis did not reveal clear differences between skin mucus and serum immunoglobulin (Ig) of carp. The majority of both Igs were tetrameric (+/- 760 kDa) and composed of 25 kDa light (L) chains and 70 kDa heavy (H) chains, but dimeric and monomeric forms were found as well. Monoclonal antibody (mAb) WCI 12 produced from serum Ig appeared to react with the H chain of both molecules. After immunisation of mice with purified mucus Ig, mAbs could be selected that were reactive with mucus Ig only. Two of these mAbs (WCI M1 and WCI M2) were immunoreactive with the H chain of mucus Ig and not or hardly immunoreactive with the H chain of serum Ig, indicating differences in the composition of the H chains of both molecules. Because WCI M2 appeared to recognize a carbohydrate determinant, differences seem to occur in the protein as well as carbohydrate composition of mucus and serum Ig. Flow cytometric results showed that both mAbs were reactive with the same subpopulation of WCI 12-positive B cells. Immunohistochemical reactions on cryosections also showed a limited reaction by these mAbs compared with WCI 12; only epithelium of skin and bile ducts and capillaries in the liver were strongly positive with these mAbs. The presence of mucus Ig at these locations is discussed. Our results indicate structural and functional differences between mucus and serum Ig, which may explain the mucosal immune responses reported for fish. Such a specific mucosal defense system can be very important for fish, living in a pathogen-rich environment.

The gut-associated lymphoid tissue (GALT) of carp (Cyprinus carpio L.): An immunocytochemical analysis

The GALT of carp was studied with monoclonal antibodies reacting with carp Ig or carp leukocytes, using (dual) immunofluorescence or immunogold staining on cryosections, cytocentrifuge slides, and cell suspensions of the intestine. The intestinal epithelium contained many Ig-negative lymphoid cells and, in the hindgut, also many large Ig-positive macrophages, which appeared to bind Ig. The lamina propria contained numerous Ig-positive lymphoid cells next to Ig-negative lymphoid cells and granulocytes. Leukocytes isolated from the intestine mainly consisted of Ig-negative lymphoid cells (> 90%). With the methods used, leukocytes were poorly released from the connective tissue. Nevertheless, two types of Ig-containing cells were found: a conventional plasma cell, frequently showing Ig at its surface, and a more common smaller lymphoid cell having a narrow rim of Ig-positive cytoplasm but hardly any Ig on its surface. Many of the Ig-positive lymphoid cells observed in the lamina propria may represent these small Ig-containing cells. Isolated Ig-positive macrophages were frequently associated with B- and T-like cells. Our data strongly suggests an immunological function for the gut of carp, especially for the antigen-transporting hindgut.

Ontogeny of the thymus in a teleost fish, Cyprinus carpio L.: developing thymocytes in the epithelial microenvironment.

A monoclonal antibody, WCL9, specific for membrane molecules of a thymocyte subpopulation was used to detect these cells in situ during the ontogeny of thymus. Cryo-sections revealed WCL9+ cells in the rudiment of the thymus (day 4 post fertilization); thereafter, the positive cells were observed exclusively in the cortex from the first appearance of thymic regionalization (week 4 post fertilization) until adult age. Whole-mount immunostaining of the thymus with WCL9 revealed the three-dimensional structure of the cortex by specific staining. The presence and distribution of apoptotic cells during thymus development was studied by in situ end-labelling of fragmented DNA. From week 4 post fertilization onwards, apoptotic cells were more frequently detected in the cortex than medulla, suggesting a continuous selection of thymocytes in the cortex. Ultrastructural studies confirmed the presence of numerous cortical apoptotic cells inside macrophages. Electron microscopy provided evidence for the existence of epithelial heterogeneity in the thymus. During the ontogeny, the differentiation of epithelial cells was followed from the first weeks until the juvenile age. Cell types were classified on the basis of their localization and cytological characteristics as: i) limiting epithelial cells located in subcapsular, perivascular and peritrabecular zones; ii) reticular epithelial cells situated in medullary and cortical zones; iii) nurse-like cells at the border between the cortex and medulla, iiii) Hassalls body-like structures localized in the medulla. This study could suggest the occurrence of a wide range of lympho-epithelial interactions throughout thymocytes differentiation.

Evolution of Recognition of Ligands from Gram-Positive Bacteria: Similarities and Differences in the TLR2-Mediated Response between Mammalian Vertebrates and Teleost Fish

We investigated the role of the TLR2 receptor in the recognition of ligands from Gram-positive bacteria in fish. Comparative sequence analysis showed a highly conserved Toll/IL-1 receptor domain. Although the leucine-rich repeat domain was less conserved, the position of the critical peptidoglycan (PGN)-binding residues in the leucine-rich repeat domain of carp TLR2 were conserved. Transfection of human embryonic kidney 293 cells with TLR2 corroborated the ability of carp TLR2 to bind the prototypical mammalian vertebrate TLR2 ligands lipoteichoic acid (LTA) and PGN from Staphylococcus aureus. The synthethic triacylated lipopeptide N-palmitoyl-S-(2,3-bis(palmitoyloxy)-(2RS)-propyl)-(R)-Cys-(S)-Ser-(S)-Lys4 trihydrochloride (Pam3CSK4) but not the diacylated lipopeptide macrophage-activating lipopeptide-2 (MALP-2) also activated TLR2 transfected human cells. We identified clear differences between the mammalian vertebrates and carp TLR2-mediated response. The use of the same ligands on carp macrophages indicated that fish cells require high concentrations of ligands from Gram-positive bacteria (LTA, PGN) for activation and signal transduction, react less strongly (Pam3CSK4) or do not react at all (MALP-2). Overexpression of TLR2 in carp macrophages confirmed TLR2 reactivity of the response to LTA and PGN, low-responsiveness to Pam3CSK4 and nonresponsiveness to MALP-2. A putative relation with the apparent absence of accessory proteins such as CD14 from the fish TLR2-containing receptor complex is discussed. Moreover, activation of carp macrophages by PGN resulted in increased TLR2 gene expression and enhanced TLR2 mRNA stability, MAPK-p38 phosphorylation and increased radical production. Finally, we could show that NADPH oxidase-derived radicals and MAPK-p38 activation cooperatively determine the level of PGN-induced TLR2 gene expression. We propose that the H2O2-MAPK-p38–dependent axis is crucial for regulation of TLR2 gene expression in fish macrophages.

Differential contribution of neutrophilic granulocytes and macrophages to nitrosative stress in a host-parasite animal model

Tyrosine nitration is a hallmark for nitrosative stress caused by the release of reactive oxygen and nitrogen species by activated macrophages and neutrophilic granulocytes at sites of inflammation and infection. In the first part of the study, we used an informative host-parasite animal model to describe the differential contribution of macrophages and neutrophilic granulocytes to in vivo tissue nitration. To this purpose common carp (Cyprinus carpio) were infected with the extracellular blood parasite Trypanoplasma borreli (Kinetoplastida). After infection, serum nitrite levels significantly increased concurrently to the upregulation of inducible nitric oxide synthase (iNOS) gene expression. Tyrosine nitration, as measured by immunohistochemistry using an anti-nitrotyrosine antibody, dramatically increased in tissues from parasite-infected fish, demonstrating that elevated NO production during T. borreli infection coincides with nitrosative stress in immunologically active tissues. The combined use of an anti-nitrotyrosine antibody with a panel of monoclonal antibodies specific for several carp leukocytes, revealed that fish neutrophilic granulocytes strongly contribute to in vivo tissue nitration most likely through both, a peroxynitrite- and an MPO-mediated mechanism. Conversely, fish macrophages, by restricting the presence of radicals and enzymes to their intraphagosomal compartment, contribute to a much lesser extent to in vivo tissue nitration. In the second part of the study, we examined the effects of nitrosative stress on the parasite itself. Peroxynitrite, but not NO donor substances, exerted strong cytotoxicity on the parasite in vitro. In vivo, however, nitration of T. borreli was limited if not absent despite the presence of parasites in highly nitrated tissue areas. Further, we investigated parasite susceptibility to the human anti-trypanosome drug Melarsoprol (Arsobal), which directly interferes with the parasite-specific trypanothione anti-oxidant system. Arsobal treatment strongly decreased T. borreli viability both, in vitro and in vivo. All together, our data suggest an evolutionary conservation in modern bony fish of the function of neutrophilic granulocytes and macrophages in the nitration process and support the common carp as a suitable animal model for investigations on nitrosative stress in host-parasite interactions. The potential of T. borreli to serve as an alternative tool for pharmacological studies on human anti-trypanosome drugs is discussed.

Immunological differences in intestine and rectum of Atlantic cod (Gadus morhua L.).

The defence system of the distal gut (hindgut and rectum) of Atlantic cod, (Gadus morhua L.) was studied using (immuno)histochemical, electron microscopical and real-time quantitative PCR techniques. The uptake and transport of macromolecules in the intestinal epithelium was also investigated. In this study we observed that cod has many and large goblet cells in its intestinal epithelium and that IgM(+) cells are present in the lamina propria and their number is considerably higher in the rectum than in the intestine. Myeloperoxidase staining revealed low numbers of granulocytes in and under the epithelium of the distal intestine, whereas high numbers were found clustered in the submucosa of the rectum. Electron microscopy not only confirmed these observations, but also revealed the presence of lymphoid cells and macrophages within the intestinal epithelium. Acid phosphatase staining demonstrated more positive macrophage-like cells in the rectum than in the distal intestine. Antigen uptake studies showed a diffused absorption of horse radish peroxidase (HRP) and LTB-GFP, whereas ferritin uptake could not be detected. Basal gene expression of cytokines (IL-1beta, IL-8 and IL-10) and immune relevant molecules (hepcidin and BPI/LPB) were compared in both the intestine and rectum and revealed approximately 2-9 times higher expression in the rectum, of which IL-1beta expression showed the most prominent difference. The present results clearly indicate that intestinal immunity is very prominent in the rectum of cod.

CHARACTERISATION OF IMMUNOGLOBULIN-BINDING LEUCOCYTES IN CARP (Cyprinus carpio L.)

This study demonstrates the immunoglobulin(Ig)-binding capacity of Ig-positive carp macrophages employing immunofluorescence and immunogold methods. These methods allow for the characterisation of the Ig-binding cells. After internalisation of fluorescent- or gold-labelled Ig (30 min at room temperature), most macrophages from the hindgut were able to bind added purified carp Ig, which could be demonstrated clearly with a second fluorescent or gold label. In pronephros, an important haemopoietic organ in fish, a limited number of monocyte-like cells also showed Ig binding. Pronephros macrophages and neutrophilic granulocytes appeared to be Ig-negative. The use of goat anti-mouse Ig gold particles bound by carp anti-goat antibodies revealed that, in addition to hindgut macrophages and pronephric monocyte-like cells, some lymphoid cells in both hindgut and pronephros cell suspensions were also able to bind Ig. The classic erythrocyte-antibody rosette assay resulted in a limited number of small rosettes in cell suspensions from both organs.

Expression of MhcCyca class i and class ii molecules in the early life history of the common carp (Cyprinus carpio L.)

In this study transcription of class I alpha chain (Cyca-UA), beta2-microglobulin (Cyca-B2m) and class II alpha (Cyca-DXA) and beta (Cyca-DAB) during the early stages of embryo development was investigated by semiquantitative PCR. No transcripts of the genes under investigation were detected in the unfertilized egg. The expression of the genes encoding for the class II molecules revealed to be synchronized starting at day 1, unlike those for the class I molecules. Transcription of Cyca-B2m was first detected at day 7, whereas Cyca-UA was already present on day 1. This discrepancy would suggest absence of class I molecules during early development. The transcription of the Mhc genes in lymphoid organs was well established on day 21, with the exception of the spleen. In later stages of ontogeny cell surface expression of class I molecules was studied using polyclonal antibodies to Cyca-UA and Cyca-B2m in conjunction with detection of surface Ig. In week 3-10 Cyca-B2m was found on a higher percentage of cells from pronephros, spleen and thymus compared to Cyca-UA, suggesting the use of an alternative class I alpha chain. In the thymus, unlike the other organs, this difference remained present in the adult stage. The most likely candidates are alpha chains encoded by non-classical class I genes.