Bronchopulmonary dysplasia (BPD) is a dangerous complication of preterm birth caused by an arrest in airway morphogenesis. Inflammation may contribute to BPD pathogenesis; in support of this idea, LPS can inhibit airway branching during lung development. Blackwell et al. (p. 2740) have now identified fetal lung-resident macrophages as the cells responsible for LPS-mediated disruption of branching. Macrophages appeared in the lung mesenchyme at the early stages of lung formation and remained there throughout development, and LPS stimulation of these cells induced activation of NF-κB and production of IL-1β and TNF-α. In a fetal lung explant model, LPS failed to impair saccular airway branching in the absence of macrophages or NF-κB signaling, suggesting that NF-κB activation in macrophages was responsible for disrupting lung development. In fact, macrophage-specific overexpression of IκB kinase β was sufficient to impair airway branching, both in the explant model and in intact fetal mice. In addition, in vivo activation of macrophages inhibited the expression of many genes important for lung morphogenesis, resulting in lungs with BPD-like morphology and often causing death soon after delivery. Identification of the pathogenic role of activated macrophages in this model of BPD provides insight that may aid in the development of treatments for this dangerous condition.
Timing Is Everything
Interaction between a TCR and its cognate peptide-MHC complex is required for T cell activation; however, the temporal requirements for this interaction during an immune response to infection are ambiguous. Blair et al. (p. 2310) developed an in vivo system to specifically address this issue by manipulating Ag availability using two peptide-MHC–specific blocking Abs. In two models of viral infection, prolonged antigenic stimulation (from 3 to 7 days, depending on the cell type and model) was required for maximal expansion of both CD4+ and CD8+ naive T cells. Memory CD8+ T cell generation was dependent on the duration of Ag exposure, but the threshold necessary for optimal memory generation was less than that required for initial T cell expansion. Development of memory CD4+ T cells, interestingly, was somewhat adversely affected by extended antigenic exposure. Analysis of recall responses revealed that secondary responses of both CD4+ and CD8+ T cells required a shorter period of antigenic stimulation than did their primary responses. As with CD4+ memory cell generation, CD4+ T cell recall responses were impaired by prolonged exposure to Ag. Thus, the ideal duration of Ag exposure for the generation of specific T cell responses must be determined to optimize vaccines under development.
Digging DAG Kinases
Proper development of invariant NKT (iNKT) cells requires specific signaling cascades initiated at their semi-invariant Vα14 TCRs. To investigate the regulation of signaling during iNKT cell development, Shen et al. (p. 2122) analyzed the involvement of the negative regulators of diacylglycerol (DAG) signaling, the DAG kinases (DGKs). Mice singly deficient in either DGKα or ζ showed no major impairments in iNKT cell development; however, mice lacking both of these kinases (DGKαζDKO) had a dramatic reduction in iNKT cell numbers in all organs analyzed, compared with controls. The redundant yet crucial functions of DGKα and ζ revealed by this loss of iNKT cells involved cell intrinsic activity that did not affect TCR recombination. Instead, DGKαζDKO thymocytes had increased DAG-mediated activation of the Ras-Erk1/2 and PKCθ–IκB kinase (IKK)–NF-κB signaling pathways relative to wild-type cells. To address the involvement of these pathways in iNKT cell development, the authors analyzed mice with constitutively active Ras or IKKβ. Constitutive Ras activity led to a late block in iNKT cell maturation, whereas constitutive IKKβ activity caused a severe reduction in iNKT cells that correlated with upregulation of ICOS and enhanced cell death. DAG-mediated signaling and its regulation by DGKα and ζ are therefore critical to multiple stages of iNKT cell development.
Tristetraprolin (TTP) is a zinc finger protein that binds to AU-rich elements (AREs) to destabilize cytokine and chemokine mRNAs. Mice deficient in TTP develop systemic inflammation and polyarticular arthritis, indicating the importance of TTP-mediated regulation of inflammatory cytokine production. To better understand the involvement of TTP in inflammatory regulation, Kang et al. (p. 2696) screened activated human monocytes for TTP binding partners and identified CCL3 mRNA as one of its targets. TTP bound to AREs in CCL3 mRNA and destabilized it in vivo. Compared with singly deficient TTP−/− mice, those doubly deficient in TTP and CCL3 demonstrated significantly ameliorated arthritis, suggesting a role for CCL3 in disease pathogenesis. However, the CCL3−/− TTP−/−- mice demonstrated similar systemic inflammation as TTP−/− mice, indicated by cachexia, myeloid hyperplasia, and increased plasma levels of the TTP targets TNF-α and IL-1β. In an APOE−/− model of atherosclerosis, TTP deficiency augmented disease, but this exacerbation was prevented by deletion of CCL3. Taken together, these data indicate that posttranscriptional regulation of CCL3 expression by TTP prevents localized tissue inflammation without affecting systemic inflammatory responses. This study reveals an important functional role for TTP beyond its well-characterized involvement in TNF-α regulation.
Keeping Eosinophils in Check
Eosinophils reside in the intestinal lamina propria under steady-state conditions, where their release of cytotoxic granules is tightly regulated through mechanisms that are not yet clear. In this issue, Verjan Garcia et al. (p. 2268) found that intestinal eosinophils constitutively expressed a high level of the inhibitory receptor signal regulatory protein α (SIRPα)/CD172a and, to a lesser degree, the degranulation-associated molecule CD63. In wild-type eosinophils, crosslinking of SIRPα/CD172a inhibited the release of the cytotoxic protein eosinophil peroxidase (EPO). However, in mice lacking the cytoplasmic domain of SIRPα/CD172a (SIRPαCyto−/− mice), receptor crosslinking did not affect EPO release. Instead, increased eosinophil degranulation, accompanied by increased CD63 expression, occurred independent of stimulation in the SIRPαCyto−/− mice. Eosinophils from these mice also spontaneously underwent apoptosis both in vitro and in vivo. These effects were mirrored in mice lacking CD47, which can serve as a receptor for SIRPα/CD172a. During a Th2-type inflammatory response, eosinophils in SIRPαCyto−/− mice demonstrated increased degranulation and decreased survival compared with cells in wild-type mice. These data indicate that SIRPα/CD172a signaling prolongs the survival of eosinophils in the intestinal lamina propria while keeping their cytotoxic activities in check.
An ideal vaccine should elicit CD8+ T cells with a variety of specificities; however, immunodominance of one or more T cell specificities often emerges and limits the repertoire of T cells primed by vaccines. To investigate the basis for vaccine-induced immunodominance, Reiser et al. (p. 2172) analyzed CD8+ T cell responses to the pp65 Ag of human CMV in HLA-A*0201-transgenic mice. Of the eight epitopes (e1–8) examined, high CD8+ T cell responses to e6, and lower responses to e3 and e8, were elicited by DNA-based vaccination against pp65. The e6-specific response suppressed T cell responses to the other epitopes, indicating the immunodominance of this epitope. Interestingly, epitope-specific DNA- or peptide-based vaccines primed CD8+ T cell responses specific for e3 and e8, but not e6. Unlike responses to the other two epitopes, the development of CD8+ T cell responses to e6 required help from CD4+ T cells. This help could be provided by CD4+ T cells specific for one or more epitopes overlapping with e6, or by more distantly related CD4+ T cells. In addition, the position of the e6 sequence within the pp65 Ag was unimportant in the establishment of the immunodominant CD8+ T cell response. The design of T cell-stimulating vaccines will benefit from this insight into the vaccine-elicited development of immunodominant and subdominant T cell responses.
V(D)J recombination of the Tcra/Tcrd locus occurs in a developmentally regulated fashion such that Tcrd genes form in double-negative (DN) thymocytes, whereas Tcra genes are assembled in double-positive thymocytes. The enhancer Eδ is active during Tcrd recombination, but its activities are less well understood than those of Eα, which acts over long distances to regulate Tcra rearrangement. To clarify the role of Eδ in the selection of Vδ gene segments, Hao and Krangel (p. 2484) compared histone modifications, germline transcription, and recombination on wild-type versus Eδ-deficient alleles. In adult mouse thymocytes, Eδ acted locally, only influencing the chromatin structure and germline transcription of regions within ∼10 kb. In fetal thymocytes, however, Eδ acted over a longer distance to control histone acetylation and germline transcription of the Vδ segment TRDV4. Recombination involving TRDV4 occurred in fetal but not adult thymocytes and required the presence of Eδ. This restriction of TRDV4 rearrangement to fetal thymocytes was found to be caused by a shift in the balance of histone acetyltransferase and deacetylase activity as mice reached adulthood. Thus, more than 10 years after Eδ was first shown to regulate Tcrd recombination, a clearer understanding of how it modulates Tcrd chromatin structure and specific steps of γδ TCR generation is beginning to emerge.
DM and DQ2 Stay in Touch
HLA-DM (DM) facilitates the loading of antigenic peptide onto MHC II and modulates the repertoire of peptides selected for presentation through peptide editing. The HLA-DQ2 (DQ2) allele is strongly associated with celiac disease and is known to resist much of the activity of DM. To understand the structural basis for the inefficient interaction between DQ2 and DM, Hou et al. (p. 2442) introduced amino acid substitutions into the DQ2 α-chain in the region predicted to contact DM. All three substitutions analyzed decreased the stability of the peptide-MHC II complex and increased DM-mediated peptide exchange. However, the α+53G substitution specifically enhanced DQ2 susceptibility to DM activity and increased cell surface expression of DQ2. This apparently improved interaction between DM and DQ2 molecules bearing the α+53G substitution suppressed the presentation of α−II gliadin 62–70, an immunodominant epitope in patients with celiac disease. These data suggested that DM normally selects against this epitope, but in the presence of wild-type DQ2, an impaired interaction with DM leads to the epitope's immunodominance. By improving the current understanding of the DM:DQ2 interaction, this study may provide insight into the basis for DQ2-associated susceptibility to disease.
Summaries written by Jennifer Hartt Meyers, Ph.D.