IN THIS ISSUE

The infectious agent of severe acute respiratory syndrome (SARS) in humans is a novel coronavirus (SARS-CoV) that also infects a variety of animals including mice. Glass et al. (p. 4030 ) noted that virus accumulated at peak levels in the epithelial lining of bronchi and terminal bronchioles in lungs of C57BL/6 (B6) mice infected intranasally 3 days earlier with a clinical isolate of SARS-CoV. Infections were resolved by day 9 postinfection (p.i.), and the animals had no clinical signs of disease except slower weight gain than mock-infected controls. Kinetics of virus replication and clearance were nearly identical in wild-type B6 mice, in B6 mice lacking NK cells or in CD1^−/− B6 animals lacking NK-T cells. SARS-CoV gene and viral RNA expression were detected in lungs of infected RAG1^−/− mice through day 6 p.i., although no infectious virus was recovered. Epithelial damage similar to that seen in lungs from infected wild-type mice was noted in lungs of all except the RAG1^−/− mice. Increased expression of mRNA and protein for a number of inflammatory chemokines and their receptors was detected in lungs from virus-infected wild-type, NK cell-deficient, and RAG1^−/− B6 mice. SARS-CoV RNA also was detected by RT-PCR in brain, heart, liver, and spleen, but not in kidney, of B6 animals through day 9 p.i. without obvious histological changes; live virus was isolated from brain on days 9–15 p.i. and was found predominantly in the hippocampus. The data suggest that the mouse model of SARS-CoV, in which viral clearance does not require NK cells, NK-T cells, or T and B cells, might mimic a subclinical human infection.

Members of the human carcinoembryonic Ag (CEA) Ig-superfamily (CEACAM) interact with each other to modulate innate and adaptive immune responses. The Mandelboim group showed that CEACAM1 homophilic interactions inhibit NK cell lysis of melanoma cells. Others have shown that CEACAM6 replaces CEACAM1 on some colorectal tumors. In a follow-up to their melanoma study, Markel et al. (p. 3732 ) demonstrated binding of a fusion protein, CEACAM1-Ig, to CEACAM1 expressed from a vector transfected into human cells; the fusion protein did not bind to cells expressing CEACAM6. Murine cells expressing a CEACAM1 protein containing mouse sequences released IL-2 when incubated with irradiated human cells expressing CEACAM1, but not when incubated with cells expressing CEACAM6. Replacement of amino acid R43 and amino acid Q44 in the N terminus of CEACAM1 with amino acid S43 and amino acid L44 found in CEACAM6, respectively, abolished homophilic binding with CEACAM1-Ig. The reciprocal substitution in CEACAM6 conferred strong binding of CEACAM1-Ig and induced release of IL-2 from mouse cells expressing the mouse CEACAM1 protein. Human CEACAM1-positive NK clones were unable to kill human cells expressing CEACAM1 or expressing CEACAM6 with both the amino acid R43 and amino acid Q44 substitutions. The data show that amino acid R43 and amino acid Q44 in the N terminus of CEACAM1 expressed on tumor cells protects them against NK cell killing.

IDO, an enzyme of tryptophan catabolism expressed in IFN-γ-activated tolerogenic dendritic cells (DCs), induces immunosuppression. The only ligand known to affect IDO expression in DCs is CTLA-4. Fallarino et al. (p. 3748 ) found that all types of mouse DCs, particularly plasmacytoid DCs, bound soluble CD200-Ig, a fusion protein of Ig and a widely distributed cell surface glycoprotein, CD200. CD200-Ig did not modify the suppressive activity of untreated CD8⁺ DCs for CD8⁻ DCs loaded with a nonapeptide that acted as a mimotope for autoimmune diabetes in mice. However, CD200-Ig treatment of CD8⁻ DCs or plasmacytoid DCs caused suppression of T cell-mediated footpad swelling induced by the peptide-loaded CD8⁻ DCs; suppression by the plasmacytoid DCs, but not by the CD8⁻ DCs, was lost with the addition of an IDO inhibitor. IDO protein expression in CD11c⁺B220⁺ DCs exposed to CD200-Ig alone was enhanced by coexposure to CpG. Coexposure to CD200-Ig and CpG induced IFN-α production, whereas CD200-Ig alone had no effect. Purified plasmacytoid DCs from IFN-αβR^−/− mice did not produce IDO protein in response to treatment with the combination of CD200-Ig and CpG and did not convert tryptophan to kynurenine. The authors have identified CD200-CD200R interactions on a variety of DC subtypes and show that those interactions result in an IDO-dependent tolerance mechanism mediated by type I IFNs.

Transcriptional activation of kinase signaling pathway members regulates inflammatory gene expression in LPS-stimulated monocytes. Histone deacetylases (HDACs) are thought to influence transcriptional activity by modifying the acetylation of several DNA-associated proteins. Mahlknecht et al. (p. 3979 ) identified recombinant human HDAC3 interaction with MAPK11 on high-density protein expression filter membranes and with in vitro-translated MAPK11; HDAC3 coimmunoprecipitated with MAPK11 from lysates of cultured cells and primary macrophages. Cotransfection of HDAC3 and MAPK11 cDNA constructs into mammalian cells resulted in activated expression of a reporter construct in a two-hybrid assay, but truncation of the HDAC3 N terminus eliminated the activity. Transient transfection of promonocytic human cells with the HDAC3-expressing vector inhibited LPS-induced phosphorylation of MAPK11 and of activating transcription factor 2 (ATF-2) and reduced the amount of endogenous MAPK11/HDAC3 interacting complexes. Other transient transfection experiments showed that HDAC3 over-expression inhibited MAPK11-mediated ATF-2 activation; small double-stranded inhibitory HDAC3 RNA reversed transcriptional repression of ATF-2 by HDAC3 and increased NF-κB activation. Transient transfection of cells with an HDAC3 expressing vector eliminated both a LPS-induced increase in ATF-2 association with TNF promoter sequences and TNF promoter activity and decreased TNF mRNA levels. The data suggest a model in which HDAC3 controls expression of the proinflammatory cytokine TNF by repressing MAPK11/ATF-2 signaling.

Inflammation and tissue damage in systemic lupus erythematosus (SLE) patients are induced by abnormally activated, necrotic T cells. NO mediates CD3/CD28 costimulation-induced mitochondrial hyperpolarization (MHP), production of reactive oxygen intermediates (ROI), and Ca²⁺ fluxing in normal T cells. However, the role of NO in SLE has not been determined. Nagy et al. (p. 3676 ) found that cytosolic and mitochondrial Ca²⁺ concentrations were higher in unstimulated and CD3/CD28-activated T cells from lupus patients compared with normal controls. NO production in lupus T cells increased to 24 h poststimulation but began to decline in normal T cells at 4 h. A population of T cells with MHP, NO production, and increased Ca²⁺ levels seen at 4 h in activated normal T cells was absent in activated lupus T cells. However, only lupus T cells had an early (2 min) and sustained surge in cytosolic Ca²⁺ concentration. Mitochondria in lupus PBLs were increased in numbers per cell and in size compared with mitochondria in normal PBLs. Although lupus T cells produced more NO than normal T cells, lupus monocytes produced twice as much NO as normal monocytes. Mitochondrial transmembrane potential and cytosolic Ca²⁺ concentrations were increased in normal monocyte-depleted PBLs incubated with lupus monocytes or with NO alone. Preincubation of normal T cells with NO also induced MHP, increased mitochondrial mass, and rapid Ca²⁺ fluxing following activation. The authors propose that intercellular NO signaling between monocytes and T cells results in persistent MHP and altered Ca²⁺ fluxing in SLE.

The fate of most immature lymphoid progenitors is determined by signals they receive through contact with thymic epithelial cells. Exposure to Notch ligands directs the cells along the T cell pathway of development, whereas absence of Notch1 signaling directs them along the B cell pathway. Höflinger et al. (p. 3935 ) found that pro-B cells from mice deficient for the Pax5 B-lineage commitment factor, cultured with bone marrow-derived stromal cells ectopically expressing the Notch ligand Delta-like1 (OP9-DL1), expressed a high level of Notch1R mRNA compared with normal stromal cells. Single c-Kit⁺B220⁺Pax5^−/− pro-B cells cultured with OP9-DL1 cells in the presence of IL-7 and Flt3 ligand (Flt3L) developed into double-positive T cells (59%) which coexpressed TCRβ on the surface (55%); TCRγδ⁺ cells also were generated (11%). No T cell differentiation was seen in cocultures with control OP9 cells. T cell differentiation to the CD44⁻CD25⁺ stage was seen for Kit^lowB220⁺ pro-B cells isolated from Pax5^−/− bone marrow and cultured with IL-7, Flt3L, and OP9-DL1 cells; similarly treated Notch1^−/−Pax5^−/− pro-B cells failed to develop into T lineage cells. Pax5^−/− bone marrow pro-B cells cocultured with OP9-DL1 cells, IL-7 and Flt3L had increased Thy1.2, c-Kit, and CD25 expression and TCRγ, δ, and β gene rearrangements; B cell-specific gene expression was reduced. Expression of several early T cell-specific genes was induced in Pax5^−/− pro-B cells, but not in Notch1^−/−Pax5^−/− pro-B cells, cultured with OP9-DL1 cells. The authors conclude that in vitro T cell development of the Pax5^−/− pro-B cells is dependent on Notch1R signaling.

Dendritic cell (DC) activation of NK cells is a critical step in host defense. NK cells express IL-15R, and IL-15 is known to support NK cell survival and proliferation. Yet a role for IL-15 in DC/NK interactions has not been established. Koka et al. (p. 3594 ) detected elevated expression of IL-12 and IL-15Rα, a subunit of the IL-15R complex, in LPS-stimulated bone marrow-derived DCs (BMDCs) from wild-type mice. However, there were no differences between the ability of wild-type and IL-15Rα^−/− BMDCs to differentiate into myeloid DCs and release IL-12 and IL-15. Survival of splenic NK cells required coculturing with BMDCs (expressing or lacking IL-15Rα). Activation of NK cells to produce IFN-γ was induced by LPS-stimulated BMDCs expressing IL-15Rα, but not by LPS-stimulated IL-15Rα^−/− or IL-15^−/− BMDCs; anti-IL-15Rα, anti-IL-12, and anti-IL-2Rβ, but not anti-IL-2Rα, Abs inhibited the IFN-γ release. IL-15Rα expression on BMDCs also was required to prime NK cell cytolytic activity; anti-IL-15Rα or anti-IL-2Rβ, but not anti-IL-12, Abs blocked the priming activity. The ability of IL-15Rα-expressing BMDCs to stimulate IFN-γ production from NK cells lacking IL-15Rα was prevented by anti-IL-15Rα Ab. NK cell activation was increased in the presence of IL-15 bound to IL-15Rα attached to culture plates and soluble IL-12. The data provide evidence that murine NK cells are activated by trans presentation of IL-15 by IL-15Rα expressed on the surface of DCs.

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