IN THIS ISSUE

Communication between cells of the immune system occurs through interaction of cell surface molecules. A novel mechanism for intercellular communication is presented in the paper by Önfelt et al. (p. 1511 ). The authors used single photon-excitation resonance scanning confocal microscopy to detect nanotubes connecting a variety of cells in culture. GFP-labeled GPI was detected in nanotubes between a transfectant expressing the GFP protein and NK cells, as well as on the surface of the cell connected to the transfectant. Three-way nanotubes extending >80 μm were seen connecting murine macrophages. Detection of bulges along nanotubes, stained vesicles within them, and stained material on their surfaces suggested transportation of material between the cells through the structures and along their surfaces. Nanotube formation between cells separated after contact was captured by time-lapse micrography. The average length of nanotubes was 30 μm, with some nanotubes growing to over 140 μm. Duration of connections was several minutes, with one connection lasting longer than 15 min. The authors present their observations as evidence of a new form of intercellular communication among immune cells. As yet, these structures have not been detected in vivo.

Immunoglobulin controls influenza virus infection of the respiratory tract. However, there is controversy regarding the relative contributions of IgG and secretory IgA. Renegar et al. (p. 1978 ) injected mice i.v. with equivalent virus-neutralizing doses of anti-influenza virus IgG₁ and anti-polymeric IgA (pIgA) mAbs 4 h before virus challenge. pIgA was 10 times more efficient than IgG₁ in reducing viral shedding in the nasopharynx by 98%. The resultant serum IgA levels were the same as those of normal influenza convalescent mice 6 wk after virus infection; the IgG₁ serum levels were seven times higher. Transudation of the injected anti-influenza virus mAbs from plasma to nasal secretions occurred at lower serum titers of pIgA than of IgG₁. Mice injected 4 h before virus infection with anti-influenza IgG₁, either in amounts equivalent to those found in influenza convalescent mice or at the higher maximal virus-neutralizing dose, had pathologic changes in nasal or tracheal epithelium similar to those of untreated infected controls. However, mice injected with anti-influenza pIgA at the maximal virus-neutralizing dose had no viral pathology in their nasal mucosa or trachea. Intrapulmonary bronchioles were protected from influenza virus damage by both doses of the anti-influenza IgG₁, but influenza virus-specific pIgA had no effect. The authors conclude that while both Ab classes play an important role in controlling influenza virus infection, secretory IgA is more important than IgG in protection of the upper respiratory tract and IgG is more important than IgA in the protection of the lungs.

Adhesion of leukocytes to endothelium can result in inflammation in such conditions as atherosclerosis and inflammatory bowel disease. One approach to treating pathological inflammation would be to decrease expression of endothelial cell adhesion molecules. Dagia et al. (p. 2041 ) found that phenyl methimazole, or compound 10 (C10), a derivative of an agent already used clinically to treat several autoimmune diseases, reduced VCAM-1 mRNA and protein expression that had been induced in human aortic endothelial cells (HAEC) by treatment with TNF-α for 4 or 24 h. C10 also reduced E-selectin expression induced by 24-h exposure to TNF-α. No effect of C10 on ICAM-1 expression was seen at either time point. Using an in vitro flow chamber that mimicked in vivo conditions, the authors showed that monocyte cell adhesion to HAEC treated with TNF-α for 4 or 24 h was blocked by a combination of anti-E-selectin mAb and C10. Only C10 alone blocked cell adhesion to HAEC treated for 24 h with TNF-α. The ability of C10 to reduce TNF-α-induced activity of a luciferase reporter plasmid containing the VCAM-1 promoter was mapped to −85 to +12 bp by promoter deletion mutants. EMSA of nuclear extracts from HEAC treated with TNF-α and C10 demonstrated decreased binding of IRF-1, but not NF-κB, to the VCAM-1 promoter; C10 also reduced TNF-α-induced IRF-1 mRNA and protein expression in HAEC. The data indicate that C10-mediated reduction of monocyte adhesion to endothelial cells is the result of blocking IRF-1 transcription, thereby preventing IRF-1 activation of the VCAM-1 promoter.

The size of the mouse postnatal thymus remains fairly constant even though thymic progenitors enter and mature thymocytes exit. Factors controlling recruitment of thymic progenitors are unknown. Prockop and Petrie (p. 1604 ) studied the ability of wild-type bone marrow to proliferate and differentiate in several nonirradiated immunodeficient mouse strains. Thymuses of mice lacking the α-chain of the IL-7R (IL-7Rα^−/−) were fully reconstituted with wild-type bone marrow and attained normal sizes. Thymuses of mice deficient in MLR-2 (RAG-2^−/−) were refractory to reconstitution, even with large numbers of wild-type bone marrow, and remained small. Moreover, transplantation of IL-7Rα^−/− mice with RAG-2^−/− bone marrow rendered the IL-7Rα^−/− mice refractory to reconstitution with wild-type marrow. IL-7Rα^−/− and other strains of mice responsive to reconstitution had low numbers of CD4⁻8⁻ (DN) thymocytes compared with mice refractory to reconstitution. Further analysis showed that a lack of CD25⁺44^low (DN3) cells in the thymus most closely correlated with the ability of wild-type bone marrow cells to expand within it. Low numbers of cells at the DN2 (CD25⁺44⁺) stage had limited correlation with the ability of wild-type bone marrow cells to expand, and there was no correlation with low numbers of cells at the DN1 (CD25⁻44⁺) stage. The IL-7Rα^−/− thymus had a normal architecture and low numbers of DN2 and DN3 cells, whereas the RAG-2^−/− thymus was poorly compartmentalized and contained primarily DN3 cells. The authors propose that the number of cells at the DN3 stage and, to a lesser extent, the DN2 stage, regulate the influx of thymic progenitors that compete for specific stromal niches.

The mouse mast cell protease 6 (mMCP-6) is activated by heparin and is stored intracellularly in a complex with it. Heparin also is involved in the assembly and stabilization of the enzymatically active mMCP-6 tetramer. Hallgren et al. (p. 1868 ) determined the amino acid residues within mMCP-6 that interact with heparin. They used site-directed mutagenesis to convert four of eight His residues that were conserved between mMCP-6 and its human homologue, β-tryptase, to Ala. The four His were determined by molecular modeling to be exposed on the surface of the tetramer. Wild-type and eight mutants (single, double, or triple) were molecularly cloned and expressed in a human kidney cell line and the proteins were purified. Tetramerization of the mutated proteins in the presence of heparin was diminished compared with that of the wild-type protein, with the greatest defect seen with two triple mutants. Although the protease triple mutants had greatly reduced enzymatic activation by heparin compared with wild-type mMCP-6, mutant tetramers, once formed, had enzymatic activities comparable to those of the wild-type tetramer. Tetramer affinity for heparin was lowered for most mutant proteins with the greatest reduction seen with the triple mutant. The data demonstrate the importance of specific surface-exposed His residues of mMCP-6 in heparin binding, tetramer formation and enzyme activation.

Continued involvement of CD4⁺ T cells in Ig class switching and in inducing CTL activity during immune responses to Ags becomes a challenge for HIV-positive individuals whose CD4⁺ T cells decrease over time. Yao et al. (p. 1951 ) looked at the ability of virus-like particles (VLPs) of different compositions to elicit humoral and cellular immune responses. CD4⁺ T cell-deficient mice were injected with chimeric VLPs, containing the hemagglutinin (HA) of influenza virus, a SIV protein and an HIV protein (HA/SHIV VLP). Those animals had higher levels of anti-HIV protein IgG₁ and IgG_2a Abs than CD4⁺ T cell-deficient mice injected with VLPs containing only the two retroviral proteins (SHIV VLP). Both groups had lower levels of viral specific IgGs than wild-type mice immunized with either VLP. Systemic immunization was more effective than mucosal immunization. Anti-HIV protein IgA was detected only in CD4⁺ T cell-deficient mice immunized mucosally with either VLP. Sera from CD4⁺ T cell-deficient mice immunized with HA/SHIV VLPs had 3-fold higher HIV-neutralizing activity in vitro compared with sera from those immunized with SHIV VLPs; splenocytes from CD4⁺ T cell-deficient mice immunized with HA/SHIV VLP had greater cytolytic activity than splenocytes from those immunized with SHIV VLP. Wild-type dendritic cells bound more HA/SHIV VLPs than SHIV VLPs, had an ∼10% increase in activation markers and produced higher levels of TNF-α and IFN-γ compared with wild-type dendritic cells that bound SHIV VLPs. The data indicate that HA has a possible adjuvant activity in enhancing both humoral and cellular anti-HIV responses to chimeric VLPs in mice lacking CD4⁺ T cells.

Interaction of the human anaphylatoxin C5a with its receptor, C5aR, initiates a series of proinflammatory responses that can result in septic shock, adult respiratory distress syndrome, rheumatoid arthritis and other immune complex-dependent diseases. Yet details of the intracellular response to C5aR activation have not been established. Melendez and Ibrahim (p. 1596 ) studied the involvement of sphingosine-1-phosphate (S1P), known to increase in response to activation of several plasma membrane receptors, in C5a-stimulated processes. C5a activation of its receptor on human macrophages induced S1P production, sphingosine kinase (SPHK1) activity and movement of SPHK1 from the cytoplasm to the membrane. Pretreatment of cells with an anti-sense SPHK1 oligonucleotide reduced SPHK1 mRNA levels by ∼80% and inhibited S1P formation; a scrambled control oligonucleotide had no effect. The anti-sense SPHK1 oligonucleotide also inhibited C5a-stimulated Ca²⁺ release, degranulation, chemotaxis, and release of TNF-α, IL-6, and IL-8, all processes stimulated by treatment of the macrophages with C5a. The data indicate that SPHK1, and S1P, mediate the responses triggered by C5a in human macrophages and offer SPHK1 as a potential novel target for therapeutic intervention in treatment of inflammatory diseases.

Summaries written by Dorothy L. Buchhagen, Ph.D.