Unresponsiveness of intrahepatic lymphocytes to bacterial superantigen: rapid development of suppressive Mac-1(high) cells in the mouse liver.

We previously found that a small dose (2 microg per mouse) of staphylococcal enterotoxin B (SEB) induced early emerging unresponsiveness in intrahepatic-lymphocyte populations (IHLs). The purpose of this study was to reveal the inducing role of accessory cells involved in IHLs in this phenomenon. IHLs prepared at 3 to 24 hours after SEB injection failed to proliferate in response not only to SEB but also to SEA, representing ligand-nonspecific unresponsiveness, whereas spleen cells (SPCs) and mesenteric lymph-node cells showed transient proliferation. Unresponsiveness in IHLs was related to a deficit of their accessory cell function as measured by coculture of irradiated IHLs and antigen-specific, type 1 T-helper (Th1) clone cells. High levels of nitrite were detected in the culture supernatant. Supplement of N(G)-monomethyl-L-arginine lowered nitrite levels and concurrently restored the proliferative response of Th1 cells, indicating the involvement of nitric oxide in suppression. Adherent cells prepared from IHLs well reproduced these results. As shown by flow cytometry, Mac-1(high) Ia(+) cells, which mainly included F4/80(+) cells (macrophages) and a minor population of CD11c(+) cells (dendritic cells), increased in proportion in IHLs but not in SPCs at 6 to 24 hours. Depletion of Mac-1(high) cells from IHLs with antibody-coated magnetic beads recovered the proliferative response. Depleted Mac-1(high) cells had a monocytoid appearance. In immunostained sections, Kupffer cells came to highly express both Mac-1 and Ia at 12 hours. These results indicate that Mac-1(high)Ia(+) adherent cells, largely Kupffer cells activated by SEB, nonspecifically suppress the proliferation of Th1 cells via nitric oxide production before manifestation of ligand-specific unresponsiveness.

Antigen-driven clonal accumulation of peritoneal gammadelta T cells in vivo.

How the clonality of gammadelta T cells changes in response to exogenous antigens is uncertain. Here we analyzed kinetics of Vgamma1.1 and Vgamma2 T cell clonality after intraperitoneal injection of purified protein derivatives (PPD) by the heterogeneity of the third complementarity determining region (CDR3) length in Vgamma1.1-Jgamma4-Cgamma4 and Vgamma2-Jgamma1-Cgamma1 junctions. The V-J junctions were analyzed in intrahepatic lymphocytes (IHL), spleen cells, and peritoneal exudate cells (PEC) by polyacrylamide gel electrophoresis. Gammadelta T cells expressing Vgamma1.1 and Vgamma2 genes were heterogeneous in normal mice. Accumulation of specific Vgamma1.1 T cell clones was transiently detected 7 days after the injection in PEC, but no accumulation was observed in IHL and spleen cells. The accumulated clones disappeared by 4 weeks. Transient accumulation of Vgamma2 T cell clones was also observed in PEC at the early phase after the injection. These results suggest that gammadelta T cells with specific TCR respond to PPD and temporary accumulate in the peritoneal cavity, but not in liver and spleen.

Cultured murine parenchymal liver cells induce differentiation of bone marrow cells to macrophage-like cells which present antigen to Th1 clones but inhibit their proliferation by nitric oxide and prostaglandins.

Suppressor cells were developed from nylon wool nonadherent CD4(-)8(-)TCRbeta(-) bone marrow cells cocultured with parenchymal liver cells for 2.5 days. The major suppressor cell population consisted of nylon wool/plastic dish-adherent, phagocytic Mac-1(+) CD3(-)4(-)8(-) cells (Ad cells), with 34% of the Ad cells being F4/80(+). These Ad cells suppressed the antigen-specific proliferation of Th1 clones in an MHC-nonrestricted manner. They showed a dose-dependent increase in suppressive activity, with both NO and PGE(2) levels in the culture supernatant rising with Ad cell concentration. OVA-pulsed Ad cells (OVA-Ad cells) were found to stimulate IFN-gamma production, resulting in an elevation of the NO and PGE(2) levels in wells containing OVA-specific Th1 clones. No DNA synthesis by these clones was detected in the absence of N(G)-monomethyl-l-arginine and indomethacin, yet the proliferation of the clone was induced in the presence of these chemicals. As proliferation is inhibited by NO and PGE(2) the Ad cells give the impression that they have no antigen-presenting function. This function is MHC-class-II-restricted. If cells such as Ad cells did actually exist in the hepatic sinusoid, they could by their nature play a major role in inducing the early emerging unresponsiveness of T cells in the liver which we reported in a previous paper.

Quantitative analysis of cytokine gene expression in the liver.

The relative levels of cytokine gene expression in the liver were analysed, focusing on IL-2, IL-4, IL-5, IL-6 and IFN-gamma compared to those in the spleen and Peyer's patch by using the reverse transcriptase polymerase chain reaction (RT-PCR). The levels of expression of cytokines in the liver mononuclear cells (MNC). especially that of IL-6, were significantly higher than in other organs when mice were reared under specific pathogen-free (SPF) or conventional conditions. Both the spleen and Peyer's patch MNC expressed little of any of the cytokines, except for IL-4 in Peyer's patch MNC. The liver MNC produced significant amounts of IL-6 in the culture supernatant upon concanavalin A stimulation. These findings suggest that the liver is a potent IL-6-producing organ, which may relate to B cell differentiation, liver regeneration and the induction of acute phase proteins.

Role of the liver in T cell differentiation--generation of CD3-CD4+/CD8+TCRbeta- cells and CD3-4-8-TCRbeta+ cells from CD4-8-TCRbeta- athymic nude bone marrow cells by culture with parenchymal liver cells.

To investigate the influence of the liver on differentiation of hematopoietic stem cells/pro-T cells, TN-NWP-BMC (athymic nude bone marrow cells that were treated with anti-TCRbeta, anti-CD4, and anti-CD8 Abs plus complement and then passed through a nylon wool column) were cultured on parenchymal liver cells. After culture for 2.5 days, CD3-4-8-TCRbeta+ cells and CD3-CD4+/CD8+TCRbeta- cells were developed from TN-NWP-BMC. TCRVbeta8+ cells comprised 19.9% of CD3-4-8-TCRbeta+ cells, and Vbeta8 mRNA was detected in the CD3-4-8-TCRbeta+ cells by reverse transcriptase-polymerase chain reaction. The CD3-CD4+/CD8+TCRbeta- cells contained not only single-positive cells but also CD4+8+ double-positive cells. The CD8 protein consisted of 88.9% CD8alpha+beta-, 10.1% CD8alpha+beta+, and 1% CD8alpha-beta+ molecules. From these results and the finding of co-expressed antigens, CD3-4-8-TCRbeta+ cells and CD3-CD4+/CD8+TCRbeta- cells appear to be immature cells not committed to a certain cell lineage.

A unique response to staphylococcal enterotoxin B by intrahepatic lymphocytes and its relevance to the induction of tolerance in the liver.

It has been reported that the intrahepatic lymphocyte (IHL) population is somewhat differently constituted from lymphocytes in other lymphoid tissues. Staphylococcal enterotoxin B (SEB) is a superantigen which can induce T-cell tolerance in mice. The authors investigated the in vitro and in vivo responses of mouse IHL to SEB. An intravenous injection of SEB did not result in the augmentation of the proliferative response of IHL, while mesenteric and splenic lymphocytes (mLNC and SPC, respectively) had augmented responses. Interleukin-2 (IL-2) mRNA was clearly detected in mLNC and SPC by reverse transcriptase-polymerase chain reaction (RT-PCR) shortly after the administration of SEB, but it was scarcely expressed in IHL. The expression of CD25 (IL-2 receptor) was also augmented in mLNC and SPC in the early period, while it was not changed in IHL. These findings suggested that the time required for tolerance induction is different locally and that the loss of augmentation of IL-2 and IL-2 receptor production by IHL may be relevant to the rapid induction of T-cell tolerance in the liver.

[Inhibitory effects of portal venous injection of type II collagen in collagen-induced arthritis].

Collagen-induced arthritis (CIA) is useful animal model for human rheumatoid arthritis. We investigated the inhibitory effects of portal venous (p.v.) injection of type II collagen (CII) in CIA. The arthritis was suppressed by p.v. injection of CII before immunization for CIA induction. The p.v. route was more effective than intravenous or intragastric routes in the induction of tolerance in CIA. The dose of CII necessary for CIA suppression was 10 micrograms/20 g body weight in p.v. injection. Both anti-CII IgG and anti-CII IgG 2 a levels in serum were reduced in mice injected CII before induction of CIA. However, anti-CII IgG 1 levels did not differ between mice injected with CII and mice injected with buffer alone. Thus, the specific reduction in anti-CII IgG 2 a levels in mice treated by p.v. injection before immunization suggests that the suppression of CIA could be responsible for hypofunction of Th 1 cells. Reduction of anti-CII IgG and suppression of arthritis were observed when CII was injected through portal vein after immunization for CIA as well.

[Induction of passive cutaneous anaphylaxis by oral administration of antigen].

Passive cutaneous anaphylaxis (PCA) reaction was developed by Ovary et al. as an animal model of mainly human type I allergic inflammation reaction. This is the most sensitive reaction for the detection of cutaneously sensitizing antibodies and provides a very effective means by which to investigate immunological reactions concerning the mechanisms of development and inhibition of allergic reactions, levels and specificity of antibodies, and the structure of antigens. These cutaneous inflammations mimic atopic dermatitis. Food ingestion has been pointed out as one of the worsened factors of atopic dermatitis. However, the body is generally protected against the invasion of high molecular substances such as non-ingested food by several barriers including digestive enzymes that break down complex food molecules into simpler substances and gastrointestinal mucosae. Accordingly, oral ingestion of food antigen seems to be physiologically and immunologically in conflict with the occurrence of dermatitis. With a view to determining whether allergic dermatitis occurs after oral ingestion of food, the present study was carried out on animals, utilizing the PCA reaction. C 57 BL/6 Ncr j mice were immunized with immunogen derived from commercially obtained eggs. Wistar rats were used as a model of PCA reaction. The results of the present investigation are summarized as follows. 1) Blue spots of 10 mm diameter were observed as a PCA reaction 50 min or more after oral administration of antigen, and the blue spots reached maximum size (20-21 mm, 1.8-1.7 micrograms) after 90-120 min. The PCA reaction was induced 20 min or more after intravenous administration of antigen. When the maximum reaction was compared between the oral and the intravenous routes after 50 min (29.5 micrograms) and 90 min (1.8 micrograms) respectively, there was about a 16-fold difference in the capacity to induce inflammations. 2) The maximum PCA reaction values were 100 and 1,600 for the oral and intravenous routes, respectively, there being a 16-fold difference between the two routes. 3) The minimum antigen concentration required to induce the PCA reaction was 10 mg/ ml for the oral route and 0.01 mg/ml for the intravenous, there being a 1,000-fold difference between the two routes. 4) Reactions with anti-egg mixture antibody and main egg constituents were specific. The PCA inhibition test results confirmed that the undigested structural portion of antigen was associated with the induction of PCA. The present investigation demonstrated that there existed a mechanism by which type I allergy is induced via the gastro-intestinal tract. This fact indicates that part of the food ingested undergoes no change in its molecular structure when transferred to the blood, thus acting as a PCA inducing antigen. This phenomenon suggests that this animal model of human type I allergic dermatitis is a useful system that strongly suggests the association of food antigen with the development of allergic dermatitis.

[The role of hepatic sinusoidal cells in homing of lymphocytes. Preferential hepatic trapping of PNA(+) lymphocyte and its modulation during liver regeneration].

It was demonstrated that the reaction of some cell species to peanut agglutinin (PNA) involved the maturity of the cells. We investigated the selective adhesion of circulating hemopoietic cells and lymphoid cells to hepatic sinusoidal cells. We experimented with thymocyte (TYC) models that could be easily isolated into PNA positive [PNA(+)] and PNA negative [PNA(-)] subsets, and transfused these subsets, labeled with 51Cr into the isogenic mice, then measured the radioactivity in the kidney, lung, spleen, liver, Peyer's patch (P.P.), and thymus, after 48 hrs. The radioactivity (% total dose recovered) in the liver and spleen represented the degree of cellular distribution and indicated that a significantly higher dose had accumulated in the liver and spleen than in other organs. Additionally, TYC-PNA(+), recognized to be immature cells, demonstrated high aggregation in the liver, while TYC-PNA(-), recognized to be mature cells, demonstrated high aggregation in the spleen. Similar results were observed in our subsequent experiment using wheat germ agglutinin (WGA), other indicator of maturity, and spleen cells (SPC). Next, we transfused TYC-PNA(+) to mice at various intervals after 70% hepatectomy. The degree of aggregation (radioactivity/weight) in the liver was reduced 1 day after hepatectomy, and that in the spleen increased proportionately. Whereas the aggregation degree of TYC-PNA(-) in the same experiment did not change significantly. We postulated that hepatic sinusoidal cells recognize and, sequentially, trap specific carbohydrates on lymphoid cells by a lectin, PNA or WGA, like receptor. Reviewing our previous report, adhesion to Kupffer cells induced SPC activation and proliferation, this recognition and trapping mechanism seems to play an important role of intrahepatic hematolymphoid system.

Changes in arrangement and in conformation of molecular components of peripheral T-cell antigen receptor complex after ligand binding: analyses by co-precipitation profiles.

Amounts of co-precipitating CD3 components by anti-T-cell receptor (TCR)V beta or anti-CD4/8 monoclonal antibodies were compared between non-stimulated and stimulated splenic T cells. The amounts of co-precipitating CD3 delta, epsilon and gamma chains with TCR alpha beta and with CD4/8 were not significantly changed after TCR ligation. The apparent amount of CD3 zeta chain co-precipitated with TCR alpha beta increased up to threefold, while the actual amount of co-precipitating CD3 zeta with TCR alpha beta and the total amount of specifically precipitated CD3 zeta are not changed after cross linking of TCR. The apparent amount of CD3 zeta chain co-precipitated with CD4/8 also increased. Unlike co-precipitation with TCR alpha beta, the actual amount of CD3 zeta co-precipitated with CD4/8 increased significantly. This observation suggests a conformational change as well as the relocation of CD3 zeta molecules within the TCR complex after the signal delivery. After TCR ligation, CD3 zeta chains relocate to the vicinity of either CD4 or CD8 molecules. In addition, when cross linking and binding signals are compared, CD3 chains undergo two distinct phases of conformational change. The responses, while the later conformational change caused by the cross linking of TCR does not induce but enhances the proliferative response.

The effects of cyclosporin A on the polykaryocyte formation induced by measles virus in a monkey kidney cell line.

The effects of cyclosporin A (CsA) on the polykaryocyte formation induced by measles virus (MV) in a monkey kidney cell line (BSC-1) were studied. CsA inhibited virus-induced polykaryocyte formation as well as the production of infectious MV. The development of polykaryocyte formation in the presence of the CsA varied with virus strains, while pretreatment of the cells with 5 microM CsA for 24 hr before the virus infection enhanced polykaryocyte formation. These data demonstrated that CsA not only inhibits but also enhances virus-induced polykaryocyte formation depending on the conditions of its use.

Production of macrophage colony-stimulating factor by murine liver in vivo.

In previous reports, the authors demonstrated that M-CSF was produced by primary-cultured non-parenchymal (NPLC) and parenchymal (PLC) liver cells. In order to clarify the biological role of M-CSF produced by the liver, macrophage colony-stimulating factor (M-CSF)-producing cells in vivo were investigated using reverse transcriptase polymerase chain reaction (RT-PCR), dot blot analysis, in situ hybridization and immunohistochemistry. M-CSF mRNA was constantly identified by RT-PCR in the liver, NPLC and PLC, before and after partial hepatectomy. Dot blot analysis showed that fluctuations of M-CSF mRNA level after partial hepatectomy were not statistically significant. In situ hybridization revealed that M-CSF mRNA was expressed mainly in NPLC and vascular endothelial cells (VEC). In addition, a small number of PLC also expressed M-CSF mRNA. Neither the distribution nor the frequency of M-CSF mRNA positive cells in regenerative livers differed significantly from normal livers. M-CSF immunoreactivity was present in NPLC and VEC at all the times before and after partial hepatectomy, while PLC exhibited M-CSF immunostaining 0.5 days after partial hepatectomy. As normal liver expressed M-CSF mRNA to the same degree as regenerative liver, hepatic M-CSF mRNA production in vivo may be related to the physiological function of the liver. However, transient expression of M-CSF protein in PLC at an early stage after partial hepatectomy may be associated with liver regeneration.

Production of B cell differentiation factors by mouse parenchymal liver cells.

Previously we reported that most antibody secreting cells secreted IgA in the liver. Here we assessed the possibility that parenchymal liver cells (PLC) produced factors, transforming growth factor (TGF)-beta and IL-5, which participate in the differentiation of B cells to IgA-secreting cells. We showed that TGF-beta activity was present in the culture supernatant of PLC, and IL-5 activity was in the lysate of PLC. Moreover, it was confirmed that IL-5 protein produced by PLC was mainly localized in the cell membrane by histochemical staining. The findings that both TGF-beta and IL-5 were produced by PLC should provide useful information concerning the fact that IgA-secreting cells were dominant in the liver.

Interleukin 7 receptor-deficient mice lack gammadelta T cells.

The interleukin 7 receptor (IL-7R) plays a crucial role in early B- and T-cell development. It consists of a unique a chain and a common gamma chain [IL-2 receptor gamma chain (IL-2Rgamma)]. Gene inactivation of IL-7, IL-7R, and IL-2Rgamma resulted in severe impairment of B and T lymphopoiesis in mice. In addition, IL-2Rgamma-deficient mice lack gammadelta T cells in the skin and have the impaired development of natural killer (NK) cells and intraepithelial lymphocytes. To explore the role of IL-7/IL-7R system in gammadelta T- and NK-cell development, we have generated and analyzed IL-7R-deficient mice. gammadelta T cells were absent from skin, gut, liver, and spleen in the deficient mice. In contrast, alphabeta T and B cells were detected in reduced, but certain, numbers, and NK cells developed normally. The gammadelta T-cell development in fetal and adult thymus was also completely blocked. These results clearly demonstrate that the signal from IL-7R is indispensable for gammadelta T-cell development in both thymic and extrathymic pathways. On the contrary, it is suggested that NK-cell development requires cytokine(s) other than IL-7.

Inactivation of human immunodeficiency virus (HIV)-1 envelope-specific CD8+ cytotoxic T lymphocytes by free antigenic peptide: a self-veto mechanism?

Free peptide has been found to inhibit cytotoxic T lymphocyte (CTL) activity, and veto cells bearing peptide-major histocompatibility complex (MHC) complexes have been found to inactivate CTL, but the two phenomena have not been connected. Here we show that a common mechanism may apply to both. CD8+ CTL lines or clones specific for a determinant of the human immunodeficiency virus (HIV) 1 IIIB envelope protein gp160, P18IIIB, are inhibited by as little as 10 min exposure to the minimal 10-mer peptide, I-10, within P18IIIB, free in solution, in contrast to peptide already bound to antigen-presenting cells (APC), which does not inhibit. Several lines of evidence suggest that the peptide must be processed and presented by H-2Dd on the CTL itself to the specific T cell receptor (TCR) to be inhibitory. The inhibition was not killing, in that CTL did not kill 51Cr-labeled sister CTL in the presence of free peptide, and in mixing experiments with CTL lines of different specificities restricted by the same MHC molecule, Dd, the presence of free peptide recognized by one CTL line did not inhibit the activity of the other CTL line that could present the peptide. Also, partial recovery of activity could be elicited by restimulation with cell-bound peptide, supporting the conclusion that neither fratricide nor suicide (apoptosis) was involved. The classic veto phenomenon was ruled out by failure of peptide-bearing CTL to inactivate others. Using pairs of CTL lines of differing specificity but similar MHC restriction, each pulsed with the peptide for which the other is specific, we showed that the minimal requirement is simultaneous engagement of the TCR and class I MHC molecules of the same cell. This could occur in single cells or pairs of cells presenting peptide to each other. Thus, mechanistically, the inhibition is analogous to veto, and might be called self-veto. As a clue to a possible mechanism, we found that free I-10 peptide induced apparent downregulation of expression of specific TCR as well as interleukin 2 receptor, CD69, lymphocyte function-associated antigen 1, and CD8. This self-veto effect also has implications for in vivo immunization and mechanisms of viral escape from CTL immunity.

Characteristic immunological tolerance in intrahepatic lymphocytes induced by bacterial superantigen SEB.

We have reported that the liver is closely associated with the hematolymphoid system and that the liver, as a lymphoid organ, may play an important role in the differentiation of lymphocytes and oral tolerance. To investigate the relevance of the liver to immunological tolerance, we compared the response pattern to Staphylococcus enterotoxin B (SEB) in intrahepatic lymphocytes (IHL) with those of spleen cells (SPC) and mesenteric lymph node cells (mLNC) during tolerance induction. Tolerance was induced by i.v. injections of SEB in all three organs. IHL had a distinct pattern from the other cells in the induction of unresponsiveness to SEB. While SPC and mLNC have a proliferation phase before the reduction of DNA synthesis, IHL become tolerant immediately without this phase. When IHL became tolerant in the early period after SEB injection, their nylon-wool column non-adherent (NW-NA) IHL were not recovered from unresponsiveness in vitro in spite of the addition of irradiated normal SPC as accessory cells. NW-NA IHL did not secrete IL-2 during this process, indicating that NW-NA IHL were in a state of functional anergy soon after the i.v. administration of SEB. Phenotypic analysis revealed that the characteristic changes of IHL during the induction of unresponsiveness might be associated with the kinetic modulation of CD4+ V beta 8+ T cells. These results suggest that the liver may hold a different mechanism for the induction of tolerance to SEB from that of SPC and mLNC.

Mouse parenchymal liver cells in culture secrete a growth inhibitor for myeloma cells.

METHODS: Growth inhibitory activity in the conditioned medium of mouse parenchymal liver cells was examined in three strains of myeloma cells. RESULTS: Two strains of myeloma cells were highly sensitive to a low concentration of mouse parenchymal liver cell derived growth inhibitor, whereas one strain was resistant to the same concentration. Interferon-alphabeta and transforming growth factor-beta activity were detected in mouse parenchymal liver cells, while interferon-gamma and tumor necrosis factor-alpha were not. The growth suppression exerted by mouse parenchymal liver cell derived growth inhibitor in the three myeloma strains was distinct from that exerted by transforming growth factor-beta, tumor necrosis factor-alpha, interferon-alphabeta and interferon-gamma. The mouse parenchymal liver cell derived growth inhibitor was eluted with a peak activity in the 18 kDa range and focused into pI values of 3.8-4.0, and it was lost when mouse parenchymal liver cells were treated with heat or trypsin. CONCLUSION: These results indicate that mouse parenchymal liver cell derived growth inhibitor differs from the well-characterized growth inhibitors, transforming growth factor-beta, tumor necrosis factor-alpha, interferon-alphabeta and interferon-gamma.

Developmental heterogeneity of V gamma 1.1 T cells in the mouse liver.

Modifications at V-(D)-J junctions increase the diversity of T-cell receptors (TCR). It has been shown that the levels of N-nucleotide insertion at the V-(D)-J junction in TCR transcripts are different between fetal and adult stages. To clarify developmental stages and pathways of gamma delta T cells in the liver, we analysed the nucleotide sequence of V gamma 1.1-J gamma 4 junctions of intra-hepatic lymphocytes (IHL), spleen cells and developing thymocytes from normal and athymic nude mice. The level of N-insertion increased in thymocytes during ontogeny. The percentage of V gamma 1.1-J gamma 4 transcripts with N-insertion was 3% at day 16 of gestation, 42% at newborn, and 89% at 7 weeks. Transcripts from normal IHL showed intermediate levels of N-insertion between those of newborn and adult thymocytes. In contrast the percentage of N-insertion in nude IHL was 47%, and this value was comparable to that of newborn thymocytes. Among the transcripts of normal IHL, the sequences common with nude IHL showed a newborn level of N-insertion (38%), and the remaining sequences showed an adult level (89%). These results suggested the possibility that V gamma 1.1-expressing T cells in IHL might be a heterogeneous population consisting of the cells developed extrathymically as well as the cells developed intrathymically. The V gamma 1.1-J gamma 4 junctions from spleen cells showed less variability than those from IHL and adult thymocytes. It suggested that gamma delta T cells bearing specific V gamma 1.1 TCR develop and/or home in the spleen.

Changes in M-CSF-like activity during chicken embryonic development.

We have established a method for separation of chicken bone-marrow cells using Percoll density gradient centrifugation, and have developed a new method for determining chicken M-CSF-like activity employing a liquid culture. Using this method, we determined M-CSF-like activities in egg yolk, chorioallantoic fluid (CAF) and amniotic fluid (AmF), and studied the effects of M-CSF on development of chicken embryos. M-CSF-like activity in egg yolk was at a high level before the incubation of the egg; it began to decrease on the third day of incubation and rapidly decreased on the fourth day, and no significant activity was detected after the tenth day of incubation. M-CSF-like activity in CAF was very low, and it exhibited almost no change during development. No M-CSF-like activity was detected in AmF throughout the experimental period.

Patterns of lymphokine production by Semliki Forest virus-specific T-cell hybridomas stimulated with different antigen-presenting cells.

The development of infection seems to be influenced by the characteristics of antigen-presenting cells (APC) in the infection site. Thus, we compared the Semliki Forest virus (SFV)-antigen-presenting capacity of spleen cells, B-cell lymphomas, bone marrow-derived mast cells and nonparenchymal liver cells by measuring the production of lymphokines in SFV-specific T-cell hybridomas. Spleen cells were able to provide the signals needed to stimulate the production of IL-2, IL-4, IL-6 and IFN-gamma, while B lymphomas the signals leading to only IL-2 production. When bone marrow-derived mast cells were used as APC, SFV-specific T-cell hybridomas produced IL-2, IL-4 and IL-6 in the presence of soluble anti-CD3 antibody. However, no lymphokine production was detected when the SFV antigen was used instead of the antibody. Nonparenchymal liver cells containing liver endothelial cells and Kupffer cells have an APC function stimulating the production of IL-2 and IL-6. These findings confirmed that the T-cell hybridomas can be selectively stimulated by different APC to produce different lymphokines, and it would influence the development of the immune-mediated inflammatory response.