IN THIS ISSUE

doi:10.4049/jimmunol.0990058

B Cells at the Heart of the Attack

Patients with acute coronary syndrome (ACS) suffer from myocardial ischemia that can ultimately lead to myocardial infarction. Myocardial ischemia is commonly caused when atherosclerotic plaques (AP) rupture. Immune cells, particularly T cells and macrophages, accumulate within AP and are believed to contribute to plaque development, but their role remains unclear. To investigate the possibility of an Ab-mediated role in AP development, Burioni et al. (p. 2537 ) examined the IgG1/κ-coding gene repertoire of B cells present within coronary plaque (CP) samples from ACS patients. B cells derived from CP samples and patient-matched PBLs showed differences in their distribution of heavy chain and light chain gene usage, indicating the presence of two distinct B cell populations. Evidence of somatic hypermutation, affinity maturation, Ig class switching, and BCR revision demonstrated that the CP B cells had undergone Ag-driven evolution. Furthermore, analysis of framework and CDR silent and nonsilent mutations revealed local Ag-driven selection of the CP-derived B cells that was distinct from the PBL population. Taken together, these data revealed the presence of a positively selected population of B cells in human CP, suggesting the possible pathogenic involvement of B cells in CP development.

Proteases Pummel S. pneumoniae

Infection by the Gram-positive bacteria Streptococcus pneumoniae (Spn) is largely controlled by the antimicrobial response of neutrophils. Neutrophils use various mechanisms to kill pathogens, including generation of neutrophil extracellular traps (NETs) and opsonin-dependent engulfment into phagolysosomes, where pathogens are exposed to both oxidative and nonoxidative mechanisms of killing. To investigate how neutrophils kill Spn, Standish and Weiser (p. 2602 ) used a complement-dependent opsonophagocytic assay. Killing of Spn required complement-mediated phagocytosis and opsonization but not NETs. Inhibition of the respiratory oxidative burst significantly decreased killing of another Gram-positive bacterium, Staphylococcus aureus, but not that of Spn. Instead, Spn was vulnerable to the contents of azurophilic granules, which contain the serine proteases elastase, cathepsin G, and proteinase 3. Inhibition of intracellular serine protease activity by general and specific inhibitors increased the survival of Spn in the phagolysosomes, whereas treatment with purified elastase and cathepsin G killed Spn. These data suggest that serine proteases released by azurophilic granules into the phagolysosomes are sufficient for neutrophil-mediated killing of Spn and demonstrate that neutrophils do not kill all Gram-positive pathogens in the same way.

iNKT Cells Are Diced without Dicer

The RNase III enzyme Dicer trims a subset of pre-microRNAs (miRNAs) in the cytoplasm to generate functional miRNAs. Fedeli et al. (p. 2506 ) used Dicer conditional knockout mice to ask whether miRNAs play a role in invariant NK T (iNKT) cell development. As compared with wild-type controls, Dicer deletion at the CD4⁺CD8⁺ double-positive (DP) stage caused a 10-fold reduction in thymic iNKT cells and their disappearance from the periphery. In the earlier CD4⁻CD8⁻ double-negative 3 (DN3) stage, Dicer deficiency caused a complete loss of αGalCer-CD1d tetramer⁺ cells from thymus, spleen, and liver. To study the role of Dicer in the iNKT developmental pathway, iNKT cells of 4-wk-old DP Dicer-deficient mice were assessed. Whereas iNKT cells at stage 0 (HSA^highCD69⁺) appeared normal, differentiation was arrested at stage 2 (CD44^highNK1.1⁻). Further analysis revealed that immature iNKT cells exhibited abnormal cell cycling and significant cell death. In a bone marrow reconstitution assay, wild-type thymocytes could not rescue impaired iNKT development, thus indicating an iNKT cell-autonomous defect. Finally, comparison of miRNAs obtained from iNKT and T cells revealed that thymic iNKT and T cells expressed distinct miRNA profiles. Thus, these data demonstrate that iNKT cell development, unlike “mainstream” T cell development, requires Dicer-dependent miRNAs.

Malnourished Dendritic Cells

Severely malnourished (SM) children are prone to infections that contribute significantly to their morbidity and mortality. Commonly, T cell function in SM children is impaired, but the underlying cause is not known. Because appropriate T cell responses are dependent on dendritic cell (DC) phenotype and activity, Hughes et al. (p. 2818 ) studied DC function in SM children. For this study, blood samples were collected from SM Zambian children upon hospital admission, discharge, and recovery after discharge. DC numbers were low at admission but had nearly doubled upon discharge following standard treatment, principally due to a rise in myeloid DC numbers. Significantly, a low DC count (<18 cells/μl) was the primary independent predictor of mortality. Low DC numbers and decreased DC activation status negatively correlated with median levels of C-reactive protein and endotoxin. In 54% of the study cohort, DC fit an anergic phenotype as defined by down-regulation of HLA-DR, elevated IL-10 secretion, absent or low IL-12 production, and an inability to support T cell proliferation. As seen with DC numbers, IL-12 production improved with nutritional recovery. Taken together, these data suggest that DC numbers and their maturation status are the basis for the impaired T cell responses found in SM children.

Renewing Stressed HSC

Thioredoxin interacting protein (TXNIP) is a stress-responsive protein that appears to link oxidative stress and metabolism. In this issue, Jeong et al. (p. 2495 ) asked whether TXNIP plays a role in maintaining the self-renewal capacity of hematopoietic stem cells (HSCs). In bone marrow (BM), TXNIP expression levels were highest in the most immature long-term HSCs (LT-HSCs) and decreased with maturation. HSCs from young Txnip^−/− and wild-type (WT) mice exhibited similar hematopoietic dynamics, but aged Txnip^−/− mice had decreased LT-HSC numbers. Furthermore, although WT and Txnip^−/− cells reconstituted lethally irradiated WT recipient mice equally well, the frequency of Txnip^−/− LT-HSCs significantly decreased with aging. To test the role of TXNIP under conditions of stress, Txnip^−/− mice were treated with 5′-fluorouracil (5-FU), which depleted actively cycling cells. After 5-FU treatment, Txnip^−/− mice took greater time to regenerate HSCs and exhibited decreased BM cellularity and increased numbers of immature cells in secondary hematopoietic organs, as compared with WT mice. The loss of HSC renewal capacity under stress in Txnip^−/− mice was linked to overactive and sustained Wnt signaling. Finally, the role of TXNIP was not cell autonomous, as it also modulated CXCL12- and osteopontin-mediated interactions between HSCs and their BM niche. Thus, these data suggest that during aging and stress, TXNIP is necessary for maintaining HSC self-renewal.

An Apolarized Response

Developed over 75 years ago, the tuberculosis vaccine Bacillus Calmette-Guérin (BCG) remains the tuberculosis vaccine of choice. However, a new vaccine formulation is widely sought after, as the protective rate from BCG vaccination is extremely variable and the complexity of the BCG adjuvant can lead to adverse side effects. To identify the components responsible for its potent activity, Swetman Andersen et al. (p. 2294 ) separated the BCG lipid extract into polar and apolar lipid fractions. Two mouse strains with different Th1/Th2 biases were vaccinated using the mycobacterial vaccine Ag Ag85B-ESAT-6, both directly with total polar or apolar fractions and coformulated with cationic liposomes. Subsequently, PBMCs were isolated, restimulated with various concentrations of vaccine Ag, and assessed for IFN-γ production. The most prominent Th1 response occurred with the apolar fraction coformulated with liposomes. Furthermore, this formulation was most successful at reducing bacteria in the lungs of vaccinated mice challenged with virulent Mycobacterium tuberculosis. Significantly, monomycoloyl glycerol (MMG) was identified as the key active component and a synthetic analog, C32 MMG, gave rise to a comparable Th1 response. These data identify MMG and its synthetic analog C32 MMG as potent mycobacteria-derived adjuvants.

Fewer Foxp3⁺DP in the Coop

Natural CD4⁺Foxp3⁺ regulatory T cells (Treg) develop in the thymus and recent studies report that up to ∼30% of the total Foxp3⁺ thymocyte population are CD4⁺CD8⁺ double positive (DP). To study this population, Lee and Hsieh (p. 2261 ) used routine FACS conditions to isolate Foxp3⁺DP cells. Analysis, however, was hampered by low sort purities due to contaminating Foxp3⁻DP and Foxp3⁺CD4 single positive (SP) cells. Further scrutiny revealed that many apparent Foxp3⁺DP events were doublets formed by the joining of Foxp3⁻DP and Foxp3⁺CD4SP cells. Following experimental optimization, Foxp3⁺DP cells comprised only ∼1% of Foxp3⁺ thymocytes, suggesting that contaminating doublets may have compromised previously reported findings. Analyses of singlet Foxp3⁺DP cells showed that Foxp3⁺DP cells expressed traditional Treg markers and underwent positive selection before transitioning to the Foxp3⁺CD4SP state. To determine whether Foxp3⁺DP cells were a significant population in neonatal thymi, the development of Treg from bone marrow precursors in neonatal bone marrow chimeras was analyzed. As in the adult thymus, neonatal Foxp3⁺DP cells were only a minor subset of Foxp3⁺ thymocytes. Thus, these data indicate that the frequency of Foxp3⁺DP cells is lower than previously believed.

Short Telomeres Snag STAT5a

Telomeres are chromatin caps that protect the ends of chromosomes but shorten with age as telomerase activity diminishes. To investigate a possible connection between telomere length and age-associated defects in macrophages (Mφ), Sebastián et al. (p. 2356 ) studied Mφ derived from the bone marrow of young (6-wk-old) and old (19- to 24-wk-old) mice. Measurement of relative telomere lengths revealed that Mφ from old mice had shorter telomeres than Mφ from young mice. In vitro M-CSF and GM-CSF stimulation revealed that, whereas Mφ of both young and old mice proliferated in response to M-CSF, only Mφ from young mice responded to GM-CSF. Analysis of the GM-CSF signaling pathway revealed that STAT5a phosphorylation was reduced in Mφ of old wild-type (WT) mice and third generation telomerase knockout (G3 Terc^−/−) mice, which also have shortened telomeres. Compared with Mφ from young mice, Mφ from both old WT and G3 Terc^−/− mice exhibited enhanced oxidative stress and decreased STAT5a oxidation, which appeared to cause decreased phosphorylation of STAT5a. In contrast, generation 1 (G1) Terc^−/− mice, which lack telomerase but have normal length telomeres, exhibited no STAT5a defect. Thus, the impaired STAT5a phosphorylation in Mφ of old WT and G3 Terc^−/− mice appears to correlate not with decreased telomerase activity but with the resultant shortening of telomere length.

Summaries written by Meredith G. Safford, Ph.D.