Combining multiple tools outperforms individual methods in gene set enrichment analyses.

Motivation: Gene set enrichment (GSE) analysis allows researchers to efficiently extract biological insight from long lists of differentially expressed genes by interrogating them at a systems level. In recent years, there has been a proliferation of GSE analysis methods and hence it has become increasingly difficult for researchers to select an optimal GSE tool based on their particular dataset. Moreover, the majority of GSE analysis methods do not allow researchers to simultaneously compare gene set level results between multiple experimental conditions. Results: The ensemble of genes set enrichment analyses (EGSEA) is a method developed for RNA-sequencing data that combines results from twelve algorithms and calculates collective gene set scores to improve the biological relevance of the highest ranked gene sets. EGSEA's gene set database contains around 25 000 gene sets from sixteen collections. It has multiple visualization capabilities that allow researchers to view gene sets at various levels of granularity. EGSEA has been tested on simulated data and on a number of human and mouse datasets and, based on biologists' feedback, consistently outperforms the individual tools that have been combined. Our evaluation demonstrates the superiority of the ensemble approach for GSE analysis, and its utility to effectively and efficiently extrapolate biological functions and potential involvement in disease processes from lists of differentially regulated genes. Availability and Implementation: EGSEA is available as an R package at http://www.bioconductor.org/packages/EGSEA/ . The gene sets collections are available in the R package EGSEAdata from http://www.bioconductor.org/packages/EGSEAdata/ . Contacts: monther.alhamdoosh@csl.com.au mritchie@wehi.edu.au. Supplementary information: Supplementary data are available at Bioinformatics online.

Therapeutic Targeting of the G-CSF Receptor Reduces Neutrophil Trafficking and Joint Inflammation in Antibody-Mediated Inflammatory Arthritis.

G-CSF is a hemopoietic growth factor that has a role in steady state granulopoiesis, as well as in mature neutrophil activation and function. G-CSF- and G-CSF receptor-deficient mice are profoundly protected in several models of rheumatoid arthritis, and Ab blockade of G-CSF also protects against disease. To further investigate the actions of blocking G-CSF/G-CSF receptor signaling in inflammatory disease, and as a prelude to human studies of the same approach, we developed a neutralizing mAb to the murine G-CSF receptor, which potently antagonizes binding of murine G-CSF and thereby inhibits STAT3 phosphorylation and G-CSF receptor signaling. Anti-G-CSF receptor rapidly halted the progression of established disease in collagen Ab-induced arthritis in mice. Neutrophil accumulation in joints was inhibited, without rendering animals neutropenic, suggesting an effect of G-CSF receptor blockade on neutrophil homing to inflammatory sites. Consistent with this, neutrophils in the blood and arthritic joints of anti-G-CSF receptor-treated mice showed alterations in cell adhesion receptors, with reduced CXCR2 and increased CD62L expression. Furthermore, blocking neutrophil trafficking with anti-G-CSF receptor suppressed local production of proinflammatory cytokines (IL-1ß, IL-6) and chemokines (KC, MCP-1) known to drive tissue damage. Differential gene expression analysis of joint neutrophils showed a switch away from an inflammatory phenotype following anti-G-CSF receptor therapy in collagen Ab-induced arthritis. Importantly, G-CSF receptor blockade did not adversely affect viral clearance during influenza infection in mice. To our knowledge, we describe for the first time the effect of G-CSF receptor blockade in a therapeutic model of inflammatory joint disease and provide support for pursuing this therapeutic approach in treating neutrophil-associated inflammatory diseases.

Production of a human neutralizing monoclonal antibody and its crystal structure in complex with ectodomain 3 of the interleukin-13 receptor α1.

Gene deletion studies in mice have revealed critical roles for IL (interleukin)-4 and -13 in asthma development, with the latter controlling lung airways resistance and mucus secretion. We have now developed human neutralizing monoclonal antibodies against human IL-13Rα1 (IL-13 receptor α1) subunit that prevent activation of the receptor complex by both IL-4 and IL-13. We describe the crystal structures of the Fab fragment of antibody 10G5H6 alone and in complex with D3 (ectodomain 3) of IL-13Rα1. Although the structure showed significant domain swapping within a D3 dimer, we showed that Arg(230), Phe(233), Tyr(250), Gln(252) and Leu(293) in each D3 monomer and Ser(32), Asn(102) and Trp(103) in 10G5H6 Fab are the key interacting residues at the interface of the 10G5H6 Fab-D3 complex. One of the most striking contacts is the insertion of the ligand-contacting residue Leu(293) of D3 into a deep pocket on the surface of 10G5H6 Fab, and this appears to be a central determinant of the high binding affinity and neutralizing activity of the antibody.

U.S. State approaches to cannabis licensing.

U.S. states have taken varied approaches to licensing cannabis businesses under federal prohibition, but up to now there is limited research on cross-state licensing approaches. This paper provides a systematic analysis of the current licensing strategies taken by all states that have passed medical cannabis laws (MCLs)/recreational cannabis laws (RCLs). We construct comprehensive data on cannabis business licenses offered in each state, as well as metrics for license categories, cost, and issuance volume. We then analyze patterns between these metrics, also considering how long ago states implemented MCLs/RCLs, qualitative licensing aspects, state ideology and voting preference, and state cannabis taxation data. We observe that states tend to license medical cannabis more restrictively than adult-use cannabis: i.e., by offering licenses in fewer categories, at higher cost, in lower issuance volume, and more often mandating vertical integration. Additionally, states that implemented MCLs/RCLs earlier tend to offer licenses in more categories, at lower cost, and in greater volumes. Further, though states that implemented MCLs recently lean conservative and Republican, we do not observe clear relationships between ideology or voting preference and licensing policy. In our supporting results, we observe that a greater share of states with complex licensing structures impose non-retail price cannabis taxes than states overall, and we discuss how states have changed their licensing policies over time.

Targeting the Human ßc Receptor Inhibits Contact Dermatitis in a Transgenic Mouse Model.

Allergic contact dermatitis (ACD) is a prevalent and poorly controlled inflammatory disease caused by skin infiltration of T cells and granulocytes. The beta common (ßc) cytokines GM-CSF, IL-3, and IL-5 are powerful regulators of granulocyte function that signal through their common receptor subunit ßc, a property that has made ßc an attractive target to simultaneously inhibit these cytokines. However, the species specificity of ßc has precluded testing of inhibitors of human ßc in mouse models. To overcome this problem, we developed a human ßc receptor transgenic mouse strain with a hematopoietic cell‒specific expression of human ßc instead of mouse ßc. Human ßc receptor transgenic cells responded to mouse GM-CSF and IL-5 but not to IL-3 in vitro and developed tissue pathology and cellular inflammation comparable with those in wild-type mice in a model of ACD. Similarly, Il3-/- mice developed ACD pathology comparable with that of wild-type mice. Importantly, the blocking anti-human ßc antibody CSL311 strongly suppressed ear pinna thickening and histopathological changes typical of ACD and reduced accumulation of neutrophils, mast cells, and eosinophils in the skin. These results show that GM-CSF and IL-5 but not IL-3 are major mediators of ACD and define the human ßc receptor transgenic mouse as a unique platform to test the inhibitors of ßc in vivo.

Type 2 immune polarization is associated with cardiopulmonary disease in preterm infants.

Postnatal maturation of the immune system is poorly understood, as is its impact on illnesses afflicting term or preterm infants, such as bronchopulmonary dysplasia (BPD) and BPD-associated pulmonary hypertension. These are both cardiopulmonary inflammatory diseases that cause substantial mortality and morbidity with high treatment costs. Here, we characterized blood samples collected from 51 preterm infants longitudinally at five time points, 20 healthy term infants at birth and age 3 to 16 weeks, and 5 healthy adults. We observed strong associations between type 2 immune polarization in circulating CD3+CD4+ T cells and cardiopulmonary illness, with odds ratios up to 24. Maternal magnesium sulfate therapy, delayed hepatitis B vaccination, and increasing fetal, but not maternal, chorioamnionitis severity were associated with attenuated type 2 polarization. Blocking type 2 mediators such as interleukin-4 (IL-4), IL-5, IL-13, or signal transducer and activator of transcription 6 (STAT6) in murine neonatal cardiopulmonary disease in vivo prevented changes in cell type composition, increases in IL-1ß and IL-13, and losses of pulmonary capillaries, but not gains in larger vessels. Thereby, type 2 blockade ameliorated lung inflammation, protected alveolar and vascular integrity, and confirmed the pathological impact of type 2 cytokines and STAT6. In-depth flow cytometry and single-cell transcriptomics of mouse lungs further revealed complex associations between immune polarization and cardiopulmonary disease. Thus, this work advances knowledge on developmental immunology and its impact on early life disease and identifies multiple therapeutic approaches that may relieve inflammation-driven suffering in the youngest patients.

The critical role of the colony-stimulating factor-1 receptor in the differentiation of myeloblastic leukemia cells.

How diverse stimuli control hemopoietic lineage development is unknown. An early event during induction of macrophage differentiation in the myeloblastic leukemia M1 cell line by different stimuli, such as leukemia inhibitory factor (LIF) and interleukin-6 (IL-6), is expression of the colony-stimulating factor-1 receptor (CSF-1R). We report that expression of active CSF-1R in M1 cells accelerated their subsequent terminal differentiation into macrophages in response to LIF and IL-6 when compared with cells lacking the CSF-1R or expressing the receptor with compromised kinase activity; however, there was no requirement for signaling through the CSF-1R, for example, via endogenous CSF-1, during the actual LIF-induced and IL-6-induced differentiation stage. Differences were noted in the signaling pathways downstream of the LIF receptor depending on the presence of the CSF-1R. Both LIF and IL-6 gave an additive response with CSF-1, consistent with LIF and IL-6 acting via a different signaling pathway (signal transducer and activator of transcription 3 dependent) than CSF-1 (extracellular signal-regulated kinase dependent). Based at least on this cell model, we propose that terminal macrophage differentiation involves a critical priming or deterministic phase in which signaling by the CSF-1R prepares a precursor population for subsequent rapid terminal macrophage differentiation by diverse stimuli. We also propose that expression and activation of the CSF-1R explain much prior literature on macrophage lineage commitment in M1 leukemic cells and may be important in controlling the progression of certain myeloid leukemias.

Machine learning applied to whole-blood RNA-sequencing data uncovers distinct subsets of patients with systemic lupus erythematosus.

OBJECTIVES: Systemic lupus erythematosus (SLE) is a heterogeneous autoimmune disease that is difficult to treat. There is currently no optimal stratification of patients with SLE, and thus, responses to available treatments are unpredictable. Here, we developed a new stratification scheme for patients with SLE, based on the computational analysis of patients' whole-blood transcriptomes. METHODS: We applied machine learning approaches to RNA-sequencing (RNA-seq) data sets to stratify patients with SLE into four distinct clusters based on their gene expression profiles. A meta-analysis on three recently published whole-blood RNA-seq data sets was carried out, and an additional similar data set of 30 patients with SLE and 29 healthy donors was incorporated in this study; a total of 161 patients with SLE and 57 healthy donors were analysed. RESULTS: Examination of SLE clusters, as opposed to unstratified SLE patients, revealed underappreciated differences in the pattern of expression of disease-related genes relative to clinical presentation. Moreover, gene signatures correlated with flare activity were successfully identified. CONCLUSION: Given that SLE disease heterogeneity is a key challenge hindering the design of optimal clinical trials and the adequate management of patients, our approach opens a new possible avenue addressing this limitation via a greater understanding of SLE heterogeneity in humans. Stratification of patients based on gene expression signatures may be a valuable strategy allowing the identification of separate molecular mechanisms underpinning disease in SLE. Further, this approach may have a use in understanding the variability in responsiveness to therapeutics, thereby improving the design of clinical trials and advancing personalised therapy.

A potential association between IL-3 and type I and III interferons in systemic lupus erythematosus.

OBJECTIVES: Plasmacytoid dendritic cells (pDCs), through the production of type 1 interferons (IFNs) and other cytokines, are major contributors to systemic lupus erythematosus (SLE) pathogenesis. IL-3 promotes pDC survival, but its role in SLE is not well characterised. This study investigated serum IL-3 and IFN levels, and a whole blood 'IL-3 gene signature', in human SLE. METHODS: Serum cytokine levels were measured by ELISA in n = 42 SLE patients, and n = 44 healthy donors. IL-3-regulated genes were determined by RNASeq of healthy donor whole blood cells (WBCs) stimulated in vitro with IL-3 for 6 or 24 h. Whole blood cell RNASeq analysis was undertaken in a separate cohort of n = 31 SLE patients, and n = 28 healthy donors. RESULTS: Serum IL-3 levels correlated with IFNα (r = 0.612, 95% CI 0.455-0.733, P < 0.001) and type III IFN (r = 0.585, 95% CI 0.406-0.720, P < 0.0001). IL-3 stimulation of WBC in vitro altered 794 genes (-1 ≥ logFC ≥ 1, FDR < 0.05), of which 35 overlapped with genes differentially expressed between SLE and healthy donors. These 35 genes were expressed in 27/31 SLE donors, revealing the presence of an 'IL-3 gene signature'. There was strong correlation between the IL-3 signature and an IFN signature, as determined by hierarchical clustering of the 500 most variable genes in SLE donors (r = 0.939, 95% CI 0.898-0.964, P < 0.0001). CONCLUSION: A dual IL-3/IFN gene signature is a feature of SLE. An association between IL-3 and IFN raises the possibility that dual blockade of IL-3 and IFN may be especially useful for SLE patients with this dual cytokine gene signature.

A dual role for the N-terminal domain of the IL-3 receptor in cell signalling.

The interleukin-3 (IL-3) receptor is a cell-surface heterodimer that links the haemopoietic, vascular and immune systems and is overexpressed in acute and chronic myeloid leukaemia progenitor cells. It belongs to the type I cytokine receptor family in which the α-subunits consist of two fibronectin III-like domains that bind cytokine, and a third, evolutionarily unrelated and topologically conserved, N-terminal domain (NTD) with unknown function. Here we show by crystallography that, while the NTD of IL3Rα is highly mobile in the presence of IL-3, it becomes surprisingly rigid in the presence of IL-3 K116W. Mutagenesis, biochemical and functional studies show that the NTD of IL3Rα regulates IL-3 binding and signalling and reveal an unexpected role in preventing spontaneous receptor dimerisation. Our work identifies a dual role for the NTD in this cytokine receptor family, protecting against inappropriate signalling and dynamically regulating cytokine receptor binding and function.

Role of the ß Common (ßc) Family of Cytokines in Health and Disease.

The ß common ([ßc]/CD131) family of cytokines comprises granulocyte macrophage colony-stimulating factor (GM-CSF), interleukin (IL)-3, and IL-5, all of which use ßc as their key signaling receptor subunit. This is a prototypic signaling subunit-sharing cytokine family that has unveiled many biological paradigms and structural principles applicable to the IL-2, IL-4, and IL-6 receptor families, all of which also share one or more signaling subunits. Originally identified for their functions in the hematopoietic system, the ßc cytokines are now known to be truly pleiotropic, impacting on multiple cell types, organs, and biological systems, and thereby controlling the balance between health and disease. This review will focus on the emerging biological roles for the ßc cytokines, our progress toward understanding the mechanisms of receptor assembly and signaling, and the application of this knowledge to develop exciting new therapeutic approaches against human disease.

Targeted therapeutics in SLE: emerging strategies to modulate the interferon pathway.

Systemic lupus erythematosus (SLE) is a prototypic autoimmune disease characterized by impaired immune tolerance, resulting in the generation of pathogenic autoantibodies and immune complexes. Although autoreactive B lymphocytes have been the first targets for biologic therapies in SLE, the importance of the innate immune system, and in particular, pathways involved in interferon (IFN) signaling, has emerged. There are now data supporting a central role for a plasmacytoid dendritic cell-derived type I IFN pathway in SLE, with a number of biologic therapeutics and small-molecule inhibitors undergoing clinical trials. Monoclonal antibodies targeting IFN-α have completed phase II clinical trials, and an antibody against the type I IFN receptor is entering a phase III trial. However, other IFNs, such as IFN gamma, and the more recently discovered type III IFNs, are also emerging as targets in SLE; and blockade of upstream components of the IFN signaling pathway may enable inhibition of more than one IFN subtype. In this review, we discuss the current understanding of IFNs in SLE, focusing on emerging therapies.

A cytotoxic anti-IL-3Rα antibody targets key cells and cytokines implicated in systemic lupus erythematosus.

To date, the major target of biologic therapeutics in systemic lupus erythematosus (SLE) has been the B cell, which produces pathogenic autoantibodies. Recently, targeting type I IFN, which is elaborated by plasmacytoid dendritic cells (pDCs) in response to endosomal TLR7 and TLR9 stimulation by SLE immune complexes, has shown promising results. pDCs express high levels of the IL-3Rα chain (CD123), suggesting an alternative potential targeting strategy. We have developed an anti-CD123 monoclonal antibody, CSL362, and show here that it affects key cell types and cytokines that contribute to SLE. CSL362 potently depletes pDCs via antibody-dependent cell-mediated cytotoxicity, markedly reducing TLR7, TLR9, and SLE serum-induced IFN-α production and IFN-α-upregulated gene expression. The antibody also inhibits TLR7- and TLR9-induced plasmablast expansion by reducing IFN-α and IL-6 production. These effects are more pronounced than with IFN-α blockade alone, possibly because pDC depletion reduces production of other IFN subtypes, such as type III, as well as non-IFN proinflammatory cytokines, such as IL-6. In addition, CSL362 depletes basophils and inhibits IL-3 signaling. These effects were confirmed in cells derived from a heterogeneous population of SLE donors, various IFN-dependent autoimmune diseases, and healthy controls. We also demonstrate in vivo activity of CSL362 following its s.c. administration to cynomolgus monkeys. This spectrum of effects provides a preclinical rationale for the therapeutic evaluation of CSL362 in SLE.

CSL311, a novel, potent, therapeutic monoclonal antibody for the treatment of diseases mediated by the common ß chain of the IL-3, GM-CSF and IL-5 receptors.

The ß common-signaling cytokines interleukin (IL)-3, granulocyte-macrophage colony stimulating factor (GM-CSF) and IL-5 stimulate pro-inflammatory activities of haematopoietic cells via a receptor complex incorporating cytokine-specific α and shared ß common (ßc, CD131) receptor. Evidence from animal models and recent clinical trials demonstrate that these cytokines are critical mediators of the pathogenesis of inflammatory airway disease such as asthma. However, no therapeutic agents, other than steroids, that specifically and effectively target inflammation mediated by all 3 of these cytokines exist. We employed phage display technology to identify and optimize a novel, human monoclonal antibody (CSL311) that binds to a unique epitope that is specific to the cytokine-binding site of the human ßc receptor. The binding epitope of CSL311 on the ßc receptor was defined by X-ray crystallography and site-directed mutagenesis. CSL311 has picomolar binding affinity for the human ßc receptor, and at therapeutic concentrations is a highly potent antagonist of the combined activities of IL-3, GM-CSF and IL-5 on primary eosinophil survival in vitro. Importantly, CSL311 inhibited the survival of inflammatory cells present in induced sputum from human allergic asthmatic subjects undergoing allergen bronchoprovocation. Due to its high potency and ability to simultaneously suppress the activity of all 3 ß common cytokines, CSL311 may provide a new strategy for the treatment of chronic inflammatory diseases where the human ßc receptor is central to pathogenesis. The coordinates for the ßc/CSL311 Fab complex structure have been deposited with the RCSB Protein Data Bank (PDB 5DWU).

cAMP inhibits CSF-1-stimulated tyrosine phosphorylation but augments CSF-1R-mediated macrophage differentiation and ERK activation.

Macrophage colony stimulating factor (M-CSF) or CSF-1 controls the development of the macrophage lineage through its receptor tyrosine kinase, c-Fms. cAMP has been shown to influence proliferation and differentiation in many cell types, including macrophages. In addition, modulation of cellular ERK activity often occurs when cAMP levels are raised. We have shown previously that agents that increase cellular cAMP inhibited CSF-1-dependent proliferation in murine bone marrow-derived macrophages (BMM) which was associated with an enhanced extracellular signal-regulated kinase (ERK) activity. We report here that increasing cAMP levels, by addition of either 8-bromo cAMP (8BrcAMP) or prostaglandin E(1) (PGE1), can induce macrophage differentiation in M1 myeloid cells engineered to express the CSF-1 receptor (M1/WT cells) and can potentiate CSF-1-induced differentiation in the same cells. The enhanced CSF-1-dependent differentiation induced by raising cAMP levels correlated with enhanced ERK activity. Thus, elevated cAMP can promote either CSF-1-induced differentiation or inhibit CSF-1-induced proliferation depending on the cellular context. The mitogen-activated protein kinase/extracellular signal-related protein kinase kinase (MEK) inhibitor, PD98059, inhibited both the cAMP- and the CSF-1R-dependent macrophage differentiation of M1/WT cells suggesting that ERK activity might be important for differentiation in the M1/WT cells. Surprisingly, addition of 8BrcAMP or PGE1 to either CSF-1-treated M1/WT or BMM cells suppressed the CSF-1R-dependent tyrosine phosphorylation of cellular substrates, including that of the CSF-1R itself. It appears that there are at least two CSF-1-dependent pathway(s), one MEK/ERK dependent pathway and another controlling the bulk of the tyrosine phosphorylation, and that cAMP can modulate signalling through both of these pathways.

A novel 110 kDa form of myosin XVIIIA (MysPDZ) is tyrosine-phosphorylated after colony-stimulating factor-1 receptor signalling.

Macrophage colony-stimulating factor (M-CSF or CSF-1) controls the development of macrophage lineage cells via activation of its tyrosine kinase receptor, c-Fms. After adding CSF-1 to M1 myeloid cells expressing CSF-1R (CSF-1 receptor), tyrosine phosphorylation of many cellular proteins occurs, which might be linked to subsequent macrophage differentiation. The biological significance and characterization of such proteins were explored by a dual strategy comprising two-dimensional SDS/PAGE analysis of cell lysates of CSF-1-treated M1 cells expressing the wild-type or a mutated receptor, together with an enrichment strategy involving a tyrosine-phosphorylated receptor construct. In the present study, we report the identification by MS of a novel, low-abundance, 110 kDa form of myosin XVIIIA (MysPDZ, myosin containing PDZ domain), which appears to be preferentially tyrosine-phosphorylated after CSF-1R activation when compared with other known isoforms. Receptor mutation studies indicate that CSF-1R-dependent tyrosine phosphorylation of p110myosin XVIIIA requires Tyr-559 in the cytoplasmic domain of the receptor and is therefore Src-family kinase-dependent. Gelsolin, Erp61 protein disulphide-isomerase and possibly non-muscle myosin IIA were also tyrosine-phosphorylated under similar conditions. Similar to the more abundant p190 isoform, p110 myosin XVIIIA lacks a PDZ domain and, in addition, it may lack motor activity. The phosphorylation of p110 myosin XVIIIA by CSF-1 may alter its cellular localization or target its association with other proteins.

Identification of potent antagonist antibodies against mouse IL-13Rα1 using novel bioassays.

Interleukin-13 (IL-13) is a cytokine implicated in airway diseases such as asthma and idiopathic pulmonary fibrosis. IL-13 signals through a heterodimeric receptor complex consisting of IL-13Rα1 and IL-4Rα, known as the type II IL-4R. IL-4 also signals through this receptor and as such many of the biological effects of IL-13 and IL-4 are similar. Here we describe the development of two sensitive bioassays to determine the potency of antagonists of the mouse type II IL-4R. Both IL-13 and IL-4 dose-dependently induce CCL17 production from J774 mouse monocytic cells and CCL11 production from NIH3T3 mouse fibroblasts in the presence of TNFα. The assays were optimized to minimize TNFα concentration, cell number and incubation time whilst retaining a suitable signal-to-background ratio. Anti-cytokine antibodies or recombinant soluble receptors completely neutralized IL-13 or IL-4 activity in these bioassays. The J774 assay was used to screen a panel of anti-mIL-13Rα1 antibodies for neutralizing activity against this receptor. We report the identification of the first monoclonal antibodies that bind mouse IL-13Rα1 and neutralize both IL-13-induced and IL-4-induced cellular function. These antibodies should prove useful for determining the effects of neutralizing IL-13Rα1 in mouse models of disease. In addition, these bioassays may be used for measuring the bioactivity of mouse IL-13 and IL-4 and for the discovery of additional antagonists of the mouse IL-13Rα1/IL-4Rα complex.

Unexpected mechanisms of action for a cytokine receptor-blocking antibody.

CSL362 is a humanized interleukin-3 (IL-3)-neutralizing monoclonal antibody with enhanced effector function that binds the α subunit of the IL-3 receptor (IL3Rα). The crystal structure of an IL3Rα:CSL362 complex shows that IL3Rα adopts "open" and "closed" conformations. CSL362 blocks IL-3 function through both IL3Rα conformations but via distinct and unexpected mechanisms.