Kathryn Miller-Jensen

Common effector processing mediates cell-specific responses to stimuli

The fundamental components of many signalling pathways are common to all cells. However, stimulating or perturbing the intracellular network often causes distinct phenotypes that are specific to a given cell type. This ‘cell specificity’ presents a challenge in understanding how intracellular networks regulate cell behaviour and an obstacle to developing drugs that treat signalling dysfunctions. Here we apply a systems-modelling approach to investigate how cell-specific signalling events are integrated through effector proteins to cause cell-specific outcomes. We focus on the synergy between tumour necrosis factor and an adenoviral vector as a therapeutically relevant stimulus that induces cell-specific responses. By constructing models that estimate how kinase-signalling events are processed into phenotypes through effector substrates, we find that accurate predictions of cell specificity are possible when different cell types share a common ‘effector-processing’ mechanism. Partial-least-squares regression models based on common effector processing accurately predict cell-specific apoptosis, chemokine release, gene induction, and drug sensitivity across divergent epithelial cell lines. We conclude that cell specificity originates from the differential activation of kinases and other upstream transducers, which together enable different cell types to use common effectors to generate diverse outcomes. The common processing of network signals by downstream effectors points towards an important cell biological principle, which can be applied to the understanding of cell-specific responses to targeted drug therapies.

Highly multiplexed profiling of single-cell effector functions reveals deep functional heterogeneity in response to pathogenic ligands

Significance We demonstrated codetection of 42 immune effector proteins in single cells, representing the highest multiplexing recorded to date for a single-cell secretion assay. Using this platform to profile differentiated macrophages stimulated with lipopolysaccharide reveals previously unobserved deep functional heterogeneity and varying levels of pathogenic activation, which is conserved throughout the cell activation process and prevails as it is extended to other Toll-like receptor (TLR) ligands and to primary human macrophages. The results indicate that the phenotypically similar cell population could still exhibit a large degree of intrinsic heterogeneity at the cell function level. This technology enables full-spectrum dissection of immune functional states in response to pathogenic stimulation and allows for more comprehensive and accurate monitoring of cellular immunity. Despite recent advances in single-cell genomic, transcriptional, and mass-cytometric profiling, it remains a challenge to collect highly multiplexed measurements of secreted proteins from single cells for comprehensive analysis of functional states. Herein, we combine spatial and spectral encoding with polydimethylsiloxane (PDMS) microchambers for codetection of 42 immune effector proteins secreted from single cells, representing the highest multiplexing recorded to date for a single-cell secretion assay. Using this platform to profile differentiated macrophages stimulated with lipopolysaccharide (LPS), the ligand of Toll-like receptor 4 (TLR4), reveals previously unobserved deep functional heterogeneity and varying levels of pathogenic activation. Uniquely protein profiling on the same single cells before and after LPS stimulation identified a role for macrophage inhibitory factor (MIF) to potentiate the activation of LPS-induced cytokine production. Advanced clustering analysis identified functional subsets including quiescent, polyfunctional fully activated, partially activated populations with different cytokine profiles. This population architecture is conserved throughout the cell activation process and prevails as it is extended to other TLR ligands and to primary macrophages derived from a healthy donor. This work demonstrates that the phenotypically similar cell population still exhibits a large degree of intrinsic heterogeneity at the functional and cell behavior level. This technology enables full-spectrum dissection of immune functional states in response to pathogenic or environmental stimulation, and opens opportunities to quantify deep functional heterogeneity for more comprehensive and accurate immune monitoring.

Control of Stochastic Gene Expression by Host Factors at the HIV Promoter

The HIV promoter within the viral long terminal repeat (LTR) orchestrates many aspects of the viral life cycle, from the dynamics of viral gene expression and replication to the establishment of a latent state. In particular, after viral integration into the host genome, stochastic fluctuations in viral gene expression amplified by the Tat positive feedback loop can contribute to the formation of either a productive, transactivated state or an inactive state. In a significant fraction of cells harboring an integrated copy of the HIV-1 model provirus (LTR-GFP-IRES-Tat), this bimodal gene expression profile is dynamic, as cells spontaneously and continuously flip between active (Bright) and inactive (Off) expression modes. Furthermore, these switching dynamics may contribute to the establishment and maintenance of proviral latency, because after viral integration long delays in gene expression can occur before viral transactivation. The HIV-1 promoter contains cis-acting Sp1 and NF-κB elements that regulate gene expression via the recruitment of both activating and repressing complexes. We hypothesized that interplay in the recruitment of such positive and negative factors could modulate the stability of the Bright and Off modes and thereby alter the sensitivity of viral gene expression to stochastic fluctuations in the Tat feedback loop. Using model lentivirus variants with mutations introduced in the Sp1 and NF-κB elements, we employed flow cytometry, mRNA quantification, pharmacological perturbations, and chromatin immunoprecipitation to reveal significant functional differences in contributions of each site to viral gene regulation. Specifically, the Sp1 sites apparently stabilize both the Bright and the Off states, such that their mutation promotes noisy gene expression and reduction in the regulation of histone acetylation and deacetylation. Furthermore, the NF-κB sites exhibit distinct properties, with κB site I serving a stronger activating role than κB site II. Moreover, Sp1 site III plays a particularly important role in the recruitment of both p300 and RelA to the promoter. Finally, analysis of 362 clonal cell populations infected with the viral variants revealed that mutations in any of the Sp1 sites yield a 6-fold higher frequency of clonal bifurcation compared to that of the wild-type promoter. Thus, each Sp1 and NF-κB site differentially contributes to the regulation of viral gene expression, and Sp1 sites functionally “dampen” transcriptional noise and thereby modulate the frequency and maintenance of this model of viral latency. These results may have biomedical implications for the treatment of HIV latency.

Analysis of single-cell cytokine secretion reveals a role for paracrine signaling in coordinating macrophage responses to TLR4 stimulation.

Comparing isolated cells with cell populations uncovers the contributions of autocrine and paracrine signaling to the secretion of cytokines by macrophages. Coordinating macrophage responses with paracrine signals When macrophages are exposed to bacteria, they secrete various cytokines to instruct the immune system to fight the pathogen. The response of a macrophage population depends on the combined effects of the secreted cytokines on the cells that secreted them (autocrine signaling) and on neighboring cells (paracrine signaling). By comparing the cytokines secreted by individual macrophages exposed to the bacterial product lipopolysaccharide (LPS) with those secreted by LPS-stimulated populations of cells, Xue et al. distinguished between the contributions of autocrine and paracrine signaling to the overall cytokine secretion response of the cell population. Paracrine signaling by the inflammatory cytokine tumor necrosis factor–α (TNF-α) was the most substantial determinant of which cytokines were secreted by cells in the population; however, paracrine signaling was also required for secretion of the cytokine IL-10, which inhibits the effects of TNF-α. Macrophages not only produce multiple cytokines but also respond to multiple cytokines, which likely shapes the ultimate response of the population. To determine the role of paracrine signaling in shaping the profile of inflammatory cytokines secreted by macrophages in response to stimulation of Toll-like receptor 4 (TLR4) with lipopolysaccharide (LPS), we combined multiplexed, microwell-based measurements of cytokine secretion by single cells with analysis of cytokine secretion by cell populations. Loss of paracrine signaling as a result of cell isolation reduced the secretion by macrophage-like U937 cells and human monocyte-derived macrophages (MDMs) of a subset of LPS-stimulated cytokines, including interleukin-6 (IL-6) and IL-10. Graphical Gaussian modeling (GGM) of the single-cell data defined a regulatory network of paracrine signals, which was validated experimentally in the population through antibody-mediated neutralization of individual cytokines. Tumor necrosis factor–α (TNF-α) was the most influential cytokine in the GGM network. Paracrine signaling by TNF-α secreted from a small subpopulation of “high-secreting” cells was necessary, but not sufficient, for the secretion of large amounts of IL-6 and IL-10 by the cell population. Decreased relative IL-10 secretion by isolated MDMs was linked to increased TNF-α secretion, suggesting that inhibition of the inflammatory response also depends on paracrine signaling. Our results reveal a previously uncharacterized role for cell-to-cell communication within a population in coordinating a rapid innate immune response despite underlying cell-to-cell heterogeneity.

Adenoviral vector saturates Akt pro-survival signaling and blocks insulin-mediated rescue of tumor-necrosis-factor-induced apoptosis

Recombinant adenoviruses are used extensively as delivery vectors in clinical gene therapy and in molecular biology, but little is known about how the viral carrier itself contributes to cellular responses. Here we show that infection with an E1/E3-deleted adenoviral vector (Adv) sensitizes human epithelial cells to tumor necrosis factor (TNF)-induced apoptosis. To explore the mechanism of Adv-mediated sensitization, we measured activity time courses for three protein kinases (MK2, IKK and JNK1) centrally involved in the TNF-receptor signaling network, as well as two kinases (Akt and ERK) activated by growth factors. Both the pro-apoptotic signal MK2 and the anti-apoptotic signal Akt were upregulated when Adv-infected cells were stimulated with TNF, and MK2 and Akt each contributed significantly to TNF-induced cell fate. Surprisingly, further activation of Akt in Adv-infected cells via insulin treatment did not significantly reduce apoptosis or MK2 activity. We show that the ineffectiveness of insulin-mediated anti-apoptotic signaling through Akt is due to saturation of Akt-effector substrate phosphorylation in Adv-infected cells. Normalizing Akt signaling relative to its Adv-induced baseline activity identified a global dose-response curve that relates Akt signaling to cellular survival. Thus, the background Akt activity induced by Adv limits the transmission of anti-apoptotic signals in response to further cytokine or growth-factor stimulation. The phenotypic and intracellular synergy between Adv and TNF may have implications for interpreting cellular responses in gene-therapy and laboratory applications.

Genetic selection for context-dependent stochastic phenotypes: Sp1 and TATA mutations increase phenotypic noise in HIV-1 gene expression.

The sequence of a promoter within a genome does not uniquely determine gene expression levels and their variability; rather, promoter sequence can additionally interact with its location in the genome, or genomic context, to shape eukaryotic gene expression. Retroviruses, such as human immunodeficiency virus-1 (HIV), integrate their genomes into those of their host and thereby provide a biomedically-relevant model system to quantitatively explore the relationship between promoter sequence, genomic context, and noise-driven variability on viral gene expression. Using an in vitro model of the HIV Tat-mediated positive-feedback loop, we previously demonstrated that fluctuations in viral Tat-transactivating protein levels generate integration-site-dependent, stochastically-driven phenotypes, in which infected cells randomly ‘switch’ between high and low expressing states in a manner that may be related to viral latency. Here we extended this model and designed a forward genetic screen to systematically identify genetic elements in the HIV LTR promoter that modulate the fraction of genomic integrations that specify ‘Switching’ phenotypes. Our screen identified mutations in core promoter regions, including Sp1 and TATA transcription factor binding sites, which increased the Switching fraction several fold. By integrating single-cell experiments with computational modeling, we further investigated the mechanism of Switching-fraction enhancement for a selected Sp1 mutation. Our experimental observations demonstrated that the Sp1 mutation both impaired Tat-transactivated expression and also altered basal expression in the absence of Tat. Computational analysis demonstrated that the observed change in basal expression could contribute significantly to the observed increase in viral integrations that specify a Switching phenotype, provided that the selected mutation affected Tat-mediated noise amplification differentially across genomic contexts. Our study thus demonstrates a methodology to identify and characterize promoter elements that affect the distribution of stochastic phenotypes over genomic contexts, and advances our understanding of how promoter mutations may control the frequency of latent HIV infection.

Chromatin accessibility at the HIV LTR promoter sets a threshold for NF-κB mediated viral gene expression

Higher order chromatin structure in eukaryotes can lead to differential gene expression in response to the same transcription factor; however, how transcription factor inputs integrate with quantitative features of the chromatin environment to regulate gene expression is not clear. In vitro models of HIV gene regulation, in which repressive mechanisms acting locally at an integration site keep proviruses transcriptionally silent until appropriately stimulated, provide a powerful system to study gene expression regulation in different chromatin environments. Here we quantified HIV expression as a function of activating transcription factor nuclear factor-κB RelA/p65 (RelA) levels and chromatin features at a panel of viral integration sites. Variable RelA overexpression demonstrated that the viral genomic location sets a threshold RelA level necessary to induce gene expression. However, once the induction threshold is reached, gene expression increases similarly for all integration sites. Furthermore, we found that higher induction thresholds are associated with repressive histone marks and a decreased sensitivity to nuclease digestion at the LTR promoter. Increasing chromatin accessibility via inhibition of histone deacetylation or DNA methylation lowered the induction threshold, demonstrating that chromatin accessibility sets the level of RelA required to activate gene expression. Finally, a functional relationship between gene expression, RelA level, and chromatin accessibility accurately predicted synergistic HIV activation in response to combinatorial pharmacological perturbations. Different genomic environments thus set a threshold for transcription factor activation of a key viral promoter, which may point toward biological principles that underlie selective gene expression and inform strategies for combinatorial therapies to combat latent HIV.

Myeloid-targeted immunotherapies act in synergy to induce inflammation and antitumor immunity

Eliciting effective antitumor immune responses in patients who fail checkpoint inhibitor therapy is a critical challenge in cancer immunotherapy, and in such patients, tumor-associated myeloid cells and macrophages (TAMs) are promising therapeutic targets. We demonstrate in an autochthonous, poorly immunogenic mouse model of melanoma that combination therapy with an agonistic anti-CD40 mAb and CSF-1R inhibitor potently suppressed tumor growth. Microwell assays to measure multiplex protein secretion by single cells identified that untreated tumors have distinct TAM subpopulations secreting MMP9 or cosecreting CCL17/22, characteristic of an M2-like state. Combination therapy reduced the frequency of these subsets, while simultaneously inducing a separate polyfunctional inflammatory TAM subset cosecreting TNF-&agr;, IL-6, and IL-12. Tumor suppression by this combined therapy was partially dependent on T cells, and on TNF-&agr; and IFN-&ggr;. Together, this study demonstrates the potential for targeting TAMs to convert a “cold” into an “inflamed” tumor microenvironment capable of eliciting protective T cell responses.

Distinct promoter activation mechanisms modulate noise-driven HIV gene expression

Latent human immunodeficiency virus (HIV) infections occur when the virus occupies a transcriptionally silent but reversible state, presenting a major obstacle to cure. There is experimental evidence that random fluctuations in gene expression, when coupled to the strong positive feedback encoded by the HIV genetic circuit, act as a ‘molecular switch’ controlling cell fate, i.e., viral replication versus latency. Here, we implemented a stochastic computational modeling approach to explore how different promoter activation mechanisms in the presence of positive feedback would affect noise-driven activation from latency. We modeled the HIV promoter as existing in one, two, or three states that are representative of increasingly complex mechanisms of promoter repression underlying latency. We demonstrate that two-state and three-state models are associated with greater variability in noisy activation behaviors, and we find that Fano factor (defined as variance over mean) proves to be a useful noise metric to compare variability across model structures and parameter values. Finally, we show how three-state promoter models can be used to qualitatively describe complex reactivation phenotypes in response to therapeutic perturbations that we observe experimentally. Ultimately, our analysis suggests that multi-state models more accurately reflect observed heterogeneous reactivation and may be better suited to evaluate how noise affects viral clearance.

Generalized selection to overcome innate immunity selects for host breadth in an RNA virus.

Virus‐host coevolution has selected for generalized host defense against viruses, exemplified by interferon production/signaling and other innate immune function in eukaryotes such as humans. Although cell‐surface binding primarily limits virus infection success, generalized adaptation to counteract innate immunity across disparate hosts may contribute to RNA virus emergence potential. We examined this idea using vesicular stomatitis virus (VSV) populations previously evolved on strictly immune‐deficient (HeLa) cells, strictly immune competent (MDCK) cells, or on alternating deficient/competent cells. By measuring viral fitness in unselected human cancer cells of differing innate immunity, we confirmed that HeLa‐adapted populations were specialized for innate immune‐deficient hosts, whereas MDCK‐adapted populations were relatively more generalized for fitness on hosts of differing innate immune capacity and of different species origin. We also confirmed that HeLa‐evolved populations maintained fitness in immune‐deficient nonhuman primate cells. These results suggest that innate immunity is more prominent than host species in determining viral fitness at the host‐cell level. Finally, our prediction was inexact that selection on alternating deficient/competent hosts should produce innate viral generalists. Rather, fitness differences among alternating host‐evolved VSV populations indicated variable capacities to evade innate immunity. Our results suggest that the evolutionary history of innate immune selection can affect whether RNA viruses evolve greater host‐breadth.