Top Reads

Sepsis triggers a wide range of immune responses, and there is a critical need for rapid, functional diagnostics that would allow early detection of these pathogenic responses. In this Top Read, Mazer et al. (p. 23) describe a whole blood ELISpot assay that can provide a functional readout of immune responses in septic patients. They compared IFN-γ and TNF-α ELISpot assays of whole blood or PBMCs from critically ill septic or nonseptic patients, as well as healthy controls. Whole blood ELISpot data were generally consistent with PBMC data but also detected nonmonocyte, myeloid populations as a major TNF-α source. Samples from critically ill septic patients who later died showed evidence of innate and adaptive immune suppression, but another group of septic patients who had elevated cytokine production relative to controls showed evidence of protective or excessive immune responses. These findings indicate that whole blood ELISpot may be an easier and faster approach for characterization of immune responses in septic patients.

The novel coronavirus SARS-CoV-2 is the causative agent of COVID-19, and an intense effort is underway to define associated immune responses. In this Top Read, Law et al. (p. 37) characterized virus-specific T cell responses in 13 individuals who had recovered from SARS-CoV2 and compared these with seasonal influenza responses. The majority of SARS-CoV-2 convalescent individuals showed CD4⁺ T cell recall responses in vitro to either spike (S) or nucleocapsid (N) in samples collected 4–12 wk after symptom onset, and CD4⁺ T cell responses dominated over those of CD8⁺ T cells. Recall responses to either SARS-CoV-2 or influenza were associated with a central memory phenotype for CD4⁺ T cells. SARS-CoV-2 recall responses exhibited higher production of IL-2 and TNF, whereas influenza-specific responses were characterized by higher levels of IFN-γ production. Peripheral T follicular helper (pTfh) IL-2⁺CCR7⁺CXCR5⁺ cells were also detected in some convalescent individuals upon in vitro stimulation with SARS-CoV-2 Ags. The presence of pTfh correlated directly with serum neutralization levels and receptor binding domain–specific IgA, but overall frequencies were lower relative to influenza-specific responses. These data provide insight into virus-specific recall responses at early time points following SARS-CoV-2 infection.

High-dimensional cytometry, which includes mass cytometry and spectral flow cytometry, has revolutionized the analysis of patient samples at the single-cell resolution. A major limitation of these methods is the difficulty in comparing datasets collected from different sites or at different time points, because many datasets lack appropriate technical replicates required for batch correction. In this Top Read, Ogishi et al. (p. 206) now describe the integration of multibatch cytometry datasets (iMUBAC), which is a flexible, robust, and scalable computational framework for rational comparison of cell subsets across multiple batches of different high-dimensional cytometry datasets. In this issue, they show that iMUBAC works as a technical replicate–independent batch correction and unsupervised clustering framework that is flexible with respect to study design, is linearly scalable, and is robust in terms of immune cell identification for both abundant and rare subsets. iMUBAC was developed using healthy control samples and validated against publicly available and in-house datasets derived from patients with autoimmune disorders or melanoma, identifying disease-associated immunophenotypes across different batches. Thus, iMUBAC is a computational framework that is likely to support the analysis of high-dimensional cytometry data in a practical manner.

Invariant NKT (iNKT) cells are a unique cell subset with innate, TCR-independent responses, including glycolipid Ag recognition, as well as TCR-dependent activation. In this Top Read, Anderson et al. (p. 132) now define the differential signaling requirements for TCR-dependent and -independent responses by iNKT cells. They generated mice with T cell–specific conditional deficiencies in IL-12Rβ2 or MyD88 signaling and observed that neither pathway is required for iNKT cell development or peripheral localization. IL-12, IL-18, and TLRs are not required for TCR-dependent iNKT activation in the presence of a strong agonist. In contrast, IL-12 was necessary but not sufficient for TCR-independent iNKT activation mediated by murine CMV infection. Surprisingly, splenic iNKT cells required IL-18 for TCR-independent activation, but IL-18 signaling was not needed for hepatic iNKT cells. These findings better define the diverse signaling requirements used by iNKT cell subsets for TCR-dependent and -independent activation.

In humans, a lack of signal peptide peptidase–like 2a (SPPL2a) results in Mendelian susceptibility to mycobacterial disease (MSMD), which is attributed to the loss of conventional dendritic cells (cDC). In this Top Read, Gradtke et al. (p. 164) showed SPPL2a-deficient mice produced more proinflammatory cytokines in response to mycobacteria. Although SPPL2a^−/− mice had reduced numbers of cDC in lymphatic tissue, these numbers were restored to wild-type (WT) levels in mice lacking both SPPL2a and CD74, suggesting the loss is CD74 dependent. However, bone marrow–derived DC (BMDC) from SPPL2a^−/− mice displayed no impairment in DC differentiation. Stimulation of SPPL2a^−/− BMDC with both heat-killed and live Mycobacterium bovis resulted in decreased IL-10 and increased IL-1β expression compared to WT BMDC, which was again dependent on CD74. This unbalanced cytokine response was caused by enhanced TLR4 responsiveness and decreased Dectin-1 surface expression and resulted in a proinflammatory cytokine profile in response to mycobacteria. Together, these data show that SPPL2a likely contributes to the mechanism of MSMD.