Ectopic lymphoid tissue formation in the lungs of mice infected with Chlamydia pneumoniae is associated with epithelial macrophage inflammatory protein-2/CXCL2 expression.

Infection with Chlamydia pneumoniae (Cp) accounts for around 10% of community acquired bacterial pneumonia and has been associated with other chronic inflammatory conditions. We describe a C57/Bl6 murine model of Cp lung infection characterized by a dose-dependent, resolving neutrophilia followed by lymphocytic infiltration of the lungs. By 21 days post-infection, mice exhibit a T helper type 1 (Th1) polarized serum antibody response with local mucosal antibody secretion and organization of ectopic lymphoid tissue which persisted in the absence of detectable Cp DNA. Macrophage inflammatory protein (MIP)-2/CXCL2, which recruits neutrophils and lymphocytes and is associated with ectopic lymphoid tissue formation, was secreted in the lungs post-infection. In vitro, lung epithelial cells up-regulated MIP-2/CXCL2 in response to both rough lipopolysaccharide (reLPS) and Cp infection. We conclude that Cp infection can have long-term inflammatory effects on tissue that persist after clearance of active infection.

Production and characterization of two monoclonal antibodies to bovine tumour necrosis factor alpha (TNF-alpha) and their cross-reactivity with ovine TNF-alpha.

Tumour necrosis factor alpha (TNF-alpha) is an innate pro-inflammatory cytokine involved in protection against intracellular pathogens. Existing methods for measuring TNF-alpha production and function in ruminants are limited to ELISA and many rely on polyclonal antisera. With a view to developing improved detection methods for bovine (bov) TNF-alpha, monoclonal antibodies (mAb) were produced by immunising mice with a plasmid encoding bov TNF-alpha. Two of the resulting mAb, termed CC327 and CC328, were used to develop a sandwich ELISA capable of detecting both native and recombinant bov TNF-alpha. This ELISA did not detect recombinant ovine (ov) TNF-alpha. A luminometric method was applied to the ELISA to improve sensitivity for detection of native bov TNF-alpha in culture supernatants derived from bovine monocyte-derived dendritic cells (DC) infected with Mycobacterium bovis. Both CC327 and CC328 detected intracytoplasmic expression of TNF-alpha in mitogen-activated bovine T lymphocytes. However, only CC328 detected intracytoplasmic ovine TNF-alpha in transfected cells, explaining the failure of the sandwich ELISA to detect recombinant ov TNF-alpha. These mAbs have generated the capability to study the role of TNF-alpha in host immune protection and disease pathogenesis in ruminants.

Ovine chlamydial abortion: characterization of the inflammatory immune response in placental tissues.

Ovine chlamydial abortion is a serious cause of fetal mortality in several sheep-rearing countries. The causal agent, Chlamydophila abortus (Chlamydia psittaci), does not generally induce clinical signs in the ewe other than abortion; this is associated with macroscopically visible damage in the placenta, which may be inflamed and thickened. To investigate the nature of the placental inflammation, seven pregnant sheep were inoculated subcutaneously at 70 days' gestation with C. abortus (strain S 26/3). A further five pregnant sheep received control inoculum by the same route at the same stage of pregnancy. Three of the infected ewes produced stillborn lambs and four produced live lambs. Lesions characteristic of chlamydial infection were present in all placentas except for two from one ewe that gave birth to twins. Histopathological examination of placental tissues from aborted fetuses showed a mixed inflammatory cell infiltrate with vasculitis and thrombosis in the mesenchyme of the intercotyledonary membranes. Cells expressing the macrophage-associated molecule CD 14 were found to be numerous, as were cells expressing major histocompatibility complex class II (MHC II) molecules. Many cells expressing messenger RNA (mRNA) encoding for tumour necrosis factor-alpha (TNF-alpha) were demonstrated, but few cells expressing interferon gamma mRNA and none expressing interleukin-4 mRNA were detected. The fetal immune response included small numbers of CD4+ and CD8+ cells, gamma delta T cells and B cells. It is concluded that abortion is the result of several factors, including destruction of tissue by C. abortus, vascular thrombosis, and an inflammatory response by the fetus. Production of TNF-alpha by fetal macrophages expressing MHC II molecules may be of considerable significance in the pathogenesis of abortion.

Immune regulation during pregnancy and host-pathogen interactions in infectious abortion.

The immunological mechanisms that govern the success of pregnancy in outbred mammals are complex. During placental formation the invasion of fetal cells into maternal tissue must be controlled to prevent damage to the mother. Equally, maternal recognition of pregnancy must be such that allorejection of the fetus does not occur. Despite the complexity of this phenomenon, it is clear that cytokines play a crucial role at the maternofetal interface and in the periphery to ensure that pregnancy proceeds successfully. Inflammatory cytokines such as tumour necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma) can exert detrimental effects in the placenta and tend to be present at low concentrations, whereas the regulatory cytokines interleukin (IL)-10 and tranforming growth factor-beta (TGF-beta) are beneficial and tend to predominate. This means that infection with pathogens that target the placenta and that elicit inflammatory responses may cause abortion by giving rise to a detrimental combination of cytokines that causes damage but does not control the disease. Infectious abortion is discussed in the context of the modulation of host immune responses during pregnancy, taking into account the different placental structures present in human beings, rodents and ruminants.

Cytokine release by ovine macrophages following infection with Chlamydia psittaci.

Chlamydia psittaci is an obligate intracellular pathogen that causes abortion in both sheep and humans. The disease in sheep (but not humans) is characterized by a long-term persistent phase that appears to be under the control of interferon-gamma. However, nothing is known about cytokine induction that precedes the persistent phase in sheep. Primary alveolar lavage cells recovered from normal adult sheep were used to study cytokine production in the first 72 h of infection with C. psittaci. These cells were phenotypically characteristic of macrophages, being adherent, phagocytic, CD14+ and staining positive for non-specific esterase. In vitro infection of the macrophages with C. psittaci resulted in the release of IL-1beta, IL-8 and granulocyte-macrophage colony-stimulating factor (GM-CSF) as measured by ovine-specific ELISAs. Heat-treated chlamydiae (1 h at 65 degrees C) did not induce the release of IL-1beta, but the release of IL-8 was similar to that induced by untreated organisms. The cells from different sheep varied most notably in their patterns of GM-CSF release in response to heat-treated and untreated organisms.

Protective immune responses to Theileria annulata of relevance to vaccine development.

A series of projects on Theileria annulata funded by the European Union (STD1/STD2/STD3) have provided convincing evidence that macrophage and natural killer (NK) cell-dependent immune mechanisms may directly control the proliferation of different stages of T. annulata in cattle. The evidence for this conclusion and the implications for vaccine development are discussed in the following paper.

Development of a sandwich ELISA for ovine granulocyte/macrophage colony-stimulating factor.

Recombinant ovine granulocyte/macrophage colony-stimulating factor (rOv GM-CSF) has been expressed in Chinese hamster ovary cells. A stable, cloned line of these cells has been established which secretes high levels (40 mu g ml(-1)) of rOv GM-CSF. Three murine monoclonal antibodies (mAbs) were produced which reacted with rOv GM-CSF on Western blots. These mAbs also neutralised the activity of both recombinant and native Ov GM-CSF in a bone marrow haemopoietic progenitor cell assay. Two of the mAbs, which recognise mutually exclusive epitopes, were selected for the development of a sandwich enzyme-linked immunosorbent assay (ELISA) to measure GM-CSF in biological samples of ovine origin.

Production and characterisation of ovine GM-CSF expressed in mammalian and bacterial cells.

A cDNA encoding ovine granulocyte-macrophage colony-stimulating factor (GM-CSF) was isolated and two forms of recombinant ovine GM-CSF were produced. A glycosylated form was produced in mammalian cells infected with a recombinant vaccinia virus encoding ovine GM-CSF. Recombinant ovine GM-CSF was also produced in Escherichia coli and purified by affinity chromatography. Both forms of the protein were detected by ovine GM-CSF-specific monoclonal antibodies, and exhibited activity on ovine bone marrow haemopoetic progenitor cells.

A double monoclonal antibody ELISA for detecting pestivirus antigen in the blood of viraemic cattle and sheep.

A panel of monoclonal antibodies (mAbs) has been produced to the p125/p80 non-structural polypeptide of border disease virus (BDV) and bovine virus diarrhoea virus (BVDV). This polypeptide appears to be highly conserved among BDV and BVDV isolates and consequently the mAbs directed against it have a broad cross-reactivity with pestivirus isolates. The epitope specificities of these mAbs were determined by competitive binding and four of the mAbs with mutually exclusive epitope specificities were selected for the development of a diagnostic ELISA. Two mAbs were used to capture virus antigen prepared from the blood of infected cattle and sheep, then two different mAbs used to detect the captured antigen. This double mAb ELISA was compared to existing ELISAs which rely on polyclonal antibodies (pAbs) for detecting captured antigen. The mAb detection ELISA was more sensitive than the pAb detection ELISAs for both cattle and sheep and resulted in higher optical densities for positive samples without an increase in background readings of negative controls.

Synthesis of tumour necrosis factor-alpha and interferons by mononuclear cells from Theileria annulata-infected cattle.

Bovine macrophage-derived tumour necrosis factor-alpha/cachectin (TNF-alpha) was synthesized when peripheral blood mononuclear cells (PBMC) and purified adherent PBMC from naive and Theileria annulata-infected cattle were incubated in vitro with concanavalin A (Con-A) or bovine recombinant interferon gamma (Bo rIFN-gamma). TNF-alpha production was also induced when adherent PBMC were cultured with T. annulata macroschizont-infected cells. In contrast, non-adherent PBMC from sublethally infected cattle produced interferon (IFN) when incubated with Hu rIL-2, Con-A, phytohaemagglutinin (PHA) or T. annulata macroschizont-infected cells growing as cell lines in vitro. Whilst PBMC from lethally infected cattle spontaneously produced IFN-gamma during advanced stages of infection, the sera of such animals contained type 1 IFN (alpha/beta). IFN was also produced by T. annulata macroschizont-infected cell lines maintained in vitro. This work suggests that cytokines serve as crucial links between proliferating Theileira-infected cells and the characteristic clinical symptoms of tropical theileriosis.

Response of efferent lymph and popliteal lymph node to epidermal infection of sheep with orf virus.

Functional and phenotypic changes in the cell populations were monitored in the popliteal efferent lymph of sheep following experimental epidermal infection with orf virus. In another group of sheep, cells from the popliteal lymph node draining the site of infection were similarly monitored and compared with the cells from contralateral popliteal and mesenteric lymph nodes. All sheep showed serological evidence of previous exposure to orf virus. Following infection, anti-orf antibody titres rose and efferent lymphocyte and blast cell output increased. Interferon-like activity was detected in efferent lymph early after orf virus but not mock infection. Lymphocytes from the draining popliteal lymph node showed antigen-specific lymphoproliferation on Days 3-7 while cells in the efferent lymph demonstrated proliferative activity on Days 4-6. The requirement for exogenous antigen-presenting cells in the culture of efferent lymphocytes varied between individual sheep. The culture supernatant from proliferating lymph node cells contained interferon-like activity but no anti-orf antibodies, the reverse of that from cultured efferent lymphocytes, perhaps indicating a different reactive T cell population. During the course of the experiment there was an increase in the percentage of efferent lymphocytes expressing MHC Class II antigens and surface immunoglobulins, the latter being recorded as a double peak. The short-term nature of the local T cell response may in part explain the incompleteness of immunity to orf virus in sheep.

Detection of border disease virus in sheep efferent lymphocytes by immunocytochemical and in situ hybridisation techniques.

The prefemoral efferent lymphatics of four sheep persistently infected with a non cytopathic (NCP) isolate of border disease virus (BDV) were cannulated. Recovered lymphocytes were examined for the presence of virus by an immunocytochemical technique employing a pool of monoclonal antibodies which recognise the 120K non-structural polypeptide of NCP BDV. The results revealed that 9.5% of the lymphocytes carried virus antigen. Lymphocytes from two of the sheep were studied by in situ hybridisation using a viral antisense RNA probe complementary to the region of the BDV genome coding for the 120K polypeptide. This showed that 70-80% of the cells were infected, confirming the greater sensitivity of the in situ hybridisation technique.

Cytopathic and non-cytopathic biotypes of border disease virus induce polypeptides of different molecular weight with common antigenic determinants.

Ten monoclonal antibodies have been raised against lysates of cells infected with cytopathic border disease virus (BDV). These antibodies all recognize non-cytopathic BDV and react with a number of different strains of bovine viral diarrhoea virus (BVDV). Studies with radiolabelled cell lysates show that all the antibodies precipitate two polypeptides of apparent Mr 80,000 and 130,000 from cells infected with cytopathic virus and a single polypeptide of apparent Mr 120,000 from cells infected with non-cytopathic virus. Two of the monoclonal antibodies react on immunoblots and show the same pattern of reactivity indicating that these three polypeptides are antigenically related.