The seamless integration of laboratory experiments and detailed computational modeling provides an exciting route to uncovering many new insights into complex biological processes. In particular, the development of agent-based modeling using supercomputers has provided new opportunities for highly detailed, validated simulations that provide the researcher with greater understanding of these processes and new directions for investigation. This chapter examines some of the principles behind the powerful computational framework FLAME and its application in a number of different areas with a more detailed look at a particular signaling example involving the NF-κB cascade.

Dynamic single-molecule force spectroscopy was performed to monitor the unbinding of fibronectin with the proteoglycans syndecan-4 (SDC4) and decorin and to compare this with the unbinding characteristics of α5β1-integrin. A single energy barrier was sufficient to describe the unbinding of both SDC4 and decorin from fibronectin, whereas two barriers were observed for the dissociation of α5β1-integrin from fibronectin. The outer (high-affinity) barriers in the interactions of fibronectin with α5β1-integrin and SDC4 are characterized by larger barrier heights and widths and slower dissociation rates than those of the inner (low-affinity) barriers in the interactions of fibronectin with α5β1-integrin and decorin. These results indicate that SDC4 and (ultimately) α5β1-integrin have the ability to withstand deformation in their interactions with fibronectin, whereas the decorin-fibronectin interaction is considerably more brittle.

The transcription factor NF-κB is a biological component that is central to the regulation of genes involved in the innate immune system. Dysregulation of the pathway is known to be involved in a large number of inflammatory diseases. Although considerable research has been performed since its discovery in 1986, we are still not in a position to control the signalling pathway, and thus limit the effects of NF-κB within promotion of inflammatory diseases. We have developed an agent-based model of the IL-1 stimulated NF-κB signalling pathway, which has been calibrated to wet-lab data at the single-cell level. Through rigorous software engineering, we believe our model provides an abstracted view of the underlying real-world system, and can be used in a predictive capacity through in silico experimentation. In this study, we have focused on the dynamics of the IKK complex and its activation of NF-κB. Our agent-based model suggests that the pathway is sensitive to: variations in the binding probability of IKK to the inhibited NF-κB-IκBα complex; and variations in the temporal rebinding delay of IKK.

Expression of the interleukin-1 receptor type I (IL-1RI) co-receptor Toll-like and interleukin-1 receptor regulator (TILRR) is significantly increased in blood monocytes following myocardial infarction and in the atherosclerotic plaque, whereas levels in healthy tissue are low. TILRR association with IL-1RI at these sites causes aberrant activation of inflammatory genes, which underlie progression of cardiovascular disease. The authors show that genetic deletion of TILRR or antibody blocking of TILRR function reduces development of atherosclerotic plaques. Lesions exhibit decreased levels of monocytes, with increases in collagen and smooth muscle cells, characteristic features of stable plaques. The results suggest that TILRR may constitute a rational target for site- and signal-specific inhibition of vascular disease.

Agent based modelling is a methodology for simulating a variety of systems across a broad spectrum of fields. However, due to the complexity of the systems it is often impossible or impractical to model them at a one to one scale. In this paper we use a simple reaction rate model implemented using the FLAME framework to test the impact of common methods for reducing model complexity such as reducing scale, increasing iteration duration and reducing message overheads. We demonstrate that such approaches can have significant impact on simulation runtime albeit with increasing risk of aberrant system behaviour and errors, as the complexity of the model is reduced.

Members of the toll-like and IL-1 receptor family (TIR) are central regulators of immune and inflammatory responses. Signal activation is induced through ligand binding and controlled by system specific co-receptors. The IL-1RI coreceptor TILRR is a splice variant of FREM1. TILRR association with the signaling receptor magnifies IL-1 induced activation of the canonical and non-canonical NF-kB network by enhancing signal amplification at the level of the receptor complex and potentiates recruitment of the MyD88 adapter and PI3 kinase. TILRRcontrolled MyD88 dependent activation of the canonical pathway is regulated in a Ras-dependent manner, reflected in alterations in cytoskeletal structure and cell adhesion. The changes induced provide a process for rapid control of NF-kB, involving sequestration and release of cytoskeletal bound IkBα through a mechanism controlled by TILRR signal amplification. In silico simulations using agent based modeling of the NFkB network predict cytoskeletal control of inhibitor levels to provide a mechanism for signal calibration and to enable activation-sensitive regulation of NFkB induced inflammatory responses. Our studies have identified two functional sites within the TILRR core protein, which selectively control inflammatory and anti-apoptotic responses. The mechanisms underlying distinct network amplification and the relevance of pathway-specific regulation of canonical and non-canonical NFkB activation will be discussed.

Computational modelling and simulation is increasingly being used to complement traditional wet-lab techniques when investigating the mechanistic behaviours of complex biological systems. In order to ensure computational models are fit for purpose, it is essential that the abstracted view of biology captured in the computational model, is clearly and unambiguously defined within a conceptual model of the biological domain (a domain model), that acts to accurately represent the biological system and to document the functional requirements for the resultant computational model. We present a domain model of the IL-1 stimulated NF-κB signalling pathway, which unambiguously defines the spatial, temporal and stochastic requirements for our future computational model. Through the development of this model, we observe that, in isolation, UML is not sufficient for the purpose of creating a domain model, and that a number of descriptive and multivariate statistical techniques provide complementary perspectives, in particular when modelling the heterogeneity of dynamics at the single-cell level. We believe this approach of using UML to define the structure and interactions within a complex system, along with statistics to define the stochastic and dynamic nature of complex systems, is crucial for ensuring that conceptual models of complex dynamical biosystems, which are developed using UML, are fit for purpose, and unambiguously define the functional requirements for the resultant computational model.

The Nuclear Factor-kappa B (NF-κB) signalling pathway is one of the key signalling pathways involved in the control and regulation of the immune system [3]. Activation of the NF-κB transcription factor is a tightly regulated event, with NF-κB normally sequestered in the cytosol of non-stimulated cells. Following activation of a cell membrane receptor and propagation of the signal via intracellular signalling to the IκB Kinase (IKK), phosphorylation-induced degradation of IκB inhibitors occurs to facilitate the release of NF-κB and its translocation to the nucleus. Dysregulation of the pathway is known to be involved in a large number of inflammatory diseases. Although considerable research has been performed since its discovery in 1986, we are still not in a position to control the signalling pathway, and thus limit the effects of NF-κB within promotion of inflammatory diseases. Through adherence to the CoSMoS framework, we are developing a computational model of the IL-1 stimulated NF-κB intracellular signalling pathway, to assist in promoting our understanding of the mechanistic behaviours within the signalling network, and therefore identify potential targets for therapeutic interventions. We have previously developed a separate domain model [4, 5] as advocated by the CoSMoS framework, which captures the essential processes and entities of the system under study using; in particular, the emergent behaviour, at an appropriate level of abstraction using a mixture of cartoon and UML diagrams, along with statistical techniques to define the temporal-spatial dynamics.

The transcription factor NF-κB (nuclear factor kappa B) is activated by Toll-like receptors and controlled by mechanotransduction and changes in the cytoskeleton. In this study we combine 3-D predictive protein modelling and in vitro experiments with in silico simulations to determine the role of the cytoskeleton in regulation of NF-κB. Simulations used a comprehensive agent-based model of the NF-κB pathway, which includes the type 1 IL-1 receptor (IL-1R1) complex and signalling intermediates, as well as cytoskeletal components. Agent based modelling relies on in silico reproductions of systems through the interactions of its components, and provides a reliable tool in investigations of biological processes, which require spatial considerations and involve complex formation and translocation of regulatory components. We show that our model faithfully reproduces the multiple steps comprising the NF-κB pathway, and provides a framework from which we can explore novel aspects of the system. The analysis, using 3-D predictive protein modelling and in vitro assays, demonstrated that the NF-κB inhibitor, IκBα is sequestered to the actin/spectrin complex within the cytoskeleton of the resting cell, and released during IL-1 stimulation, through a process controlled by the IL-1RI co-receptor TILRR (Toll-like and IL-1 receptor regulator). In silico simulations using the agent-based model predict that the cytoskeletal pool of IκBα is released to adjust signal amplification in relation to input levels. The results suggest that the process provides a mechanism for signal calibration and enables efficient, activation-sensitive regulation of NF-κB and inflammatory responses.

Introduction Inflammatory responses are key drivers in the pathogenesis of atherosclerosis. We have identified a co-receptor, TILRR, a prominent activator of NF-κB controlled responses, which we have demonstrated is highly expressed in atherosclerotic lesions. Using Apoe–/– (Apo-lipoprotein E knockout) mice we have recently shown that administration of a blocking peptide antibody, which selectively targets amplification of inflammatory responses through TILRR, significantly reduces progression of atherosclerosis and promotes plaque stability. This study examines the molecular mechanisms underlying the reduced NF-κB activation in the TILRR knockout mouse. Methods and results Microarray analysis of blood and spleen monocytes from TILRR knockout mice and littermate controls, revealed pronounced reductions in NF-κB controlled inflammatory genes relevant to development of cardiovascular disease including Cathepsin L1, CXCL1 and CXCL13. Western analysis demonstrated a 65% decrease in IκBα expression and a 60% reduction in inhibitor degradation in IL-1 stimulated monocytes from TILRR-/- mice compared to responses in littermate controls. Parallel in vitro studies were carried out to determine the underlying mechanisms and consequences of these changes on interaction of the inhibitor with NF-κB. Deletion of the ANK 2 region of IκBα resulted in loss of function while the impact of ANK deletions 3, 4, and 5 was not significantly different from the wild type protein. 3-D modelling of the ANK2 deleted protein revealed that the modified protein maintained its overall structure. The modified protein lost its ability to interact with NF-κB, identifying a role for this region in IkBa/NF-kB complex formation. Ongoing studies are evaluating alanine-scanning mutants of conserved residues within the ANK2 region to determine their impact on amplified NF-kB responses. Conclusion Our results are consistent with a role for TILRR in regulation of NF-κB controlled inflammatory gene activation in vivo, and suggest that functional alterations in IκBα regulation in part are responsible for the reductions in the inflammatory response.

A study by Cheng et al. in this issue of Science Signaling highlights the distinct single-cell signaling characteristics conferred by pathways mediated by the adaptor proteins MyD88 and TRIF in the TLR4-dependent activation of the transcription factor nuclear factor κB (NF-κB).

Background TILRR, an IL-1R1 co-receptor, is a strong amplifier of IL-1-induced inflammatory and anti-apoptotic signals controlled by NF-κB.1 Our recent work has demonstrated that TILRR is highly expressed at areas of inflammation such the atherosclerotic plaque, and that inhibiting TILRR function, using a blocking peptide antibody, reduces plaque development in a mouse model of atherosclerosis (Samokhin et al, unpublished). This study focuses on characterisation of the inflammatory phenotype of a TILRR knock-out mouse, which was created to further assess the role of TILRR in vascular disease. Methods and results Wild-type and TILRR null mice were injected intraperitoneally with lipopolysaccharide (LPS) or PBS control. Animals were sacrificed at 3 or 12 h and cells from blood and spleen analysed by FACS. Protein and RNA levels were determined by western blotting, qPCR and microarray. Our results demonstrate a 50% decrease in neutrophil levels in the TILRR KO mice compared to wild type, following LPS injection. In addition, FACS analysis revealed a reduction in inflammatory monocytes, with a corresponding increase in the non-inflammatory phenotype. Further, TILRR knockout caused on average a 50% reduction in IL-1 receptor levels, and resulted in a decrease in ligand-induced activation of NF-κB and in release of inflammatory mediators, such as IL-6 and IL-8. Ongoing microarray analysis will determine the impact of these changes on overall gene activation profiles in TILRR knockouts. Based on initial studies using TILRR siRNA, the absence of TILRR expression is expected have pronounced effects on multiple genes relevant to control of vascular inflammation. Conclusion The results are consistent with a central role for TILRR in regulating inflammatory processes in vivo . Our data show that TILRR deficient mice exhibit reduced responses to systemic inflammation, and show that these are induced at the level of receptor function and signal transduction. The findings agree with work carried out in parallel, focusing on the role of TILRR in cardiovascular injury and disease. Supported by the BHF and BBSRC. References Zhang X, Shephard F, Kim HB, Palmer IR, Mcharg S, Fowler GJ, O’Neill LA, Kiss-Toth E, Qwarnstrom EE. TILRR, a novel IL-1RI co-receptor, potentiates MyD88 recruitment to control Ras-dependent amplification of NF-kappaB. J Biol Chem 2010;285:7222–32 Zhang X, Montagut Pino G, Shephard F, Kiss-Toth E, Qwarnstrom EE. Distinct control of MYD88-dependent and AKT-regulated responses by the IL-1RI co-receptor, TILRR. J Biol Chem 2012;287:12348–12352

In this review article, we discuss the current state of computational modelling of the nuclear factor-kappa B (NF-ΚB) signalling pathway. NF-ΚB is a transcription factor, which is ubiquitous within cells and controls a number of immune responses, including inflammation and apoptosis. The NF-ΚB signalling pathway is tightly regulated, commencing with activation at the cell membrane, signal transduction through various components within the cytoplasm, translocation of NF-ΚB into the nucleus and, finally, the transcription of various genes relating to the innate and adaptive immune responses. There have been a number of computational (mathematical) models developed of the signalling pathway over the past decade. This review describes how these approaches have helped advance our understanding of NF-ΚB control.

Introduction Interleukin-1 (IL-1) and its signalling receptor (IL-1RI) are well known as central regulators of vascular repair and injury. Specifically, activation of the NF-kB pathway and its control of inflammatory and anti-apoptotic signals have pronounced effects on development of atherosclerosis. TILRR is a recently identified co-receptor to IL-1RI, which amplifies NF-kB activities and alters gene activation profiles induced by IL-1.1 TILRR is highly expressed at sites of vascular inflammation where its recruitment to the IL-1 receptor complex enhances inflammatory signals, reduces apoptosis and triggers cytoskeletal collapse during IL-1 stimulation. Mutants of TILRR (R425 and D448) have been produced, and demonstrated to selectively reduce enhanced inflammatory or anti-apoptotic signals through the IL-1 receptor complex.2 Methods and Results We have developed an agent-based model of the IL-1R signalling pathway capable of qualitatively reproducing in vitro data of IkBa degradation and NF-kB nuclear translocation. The model describes accurately the impact of varying TILRR expression, the effects of varying the affinity for adaptor proteins at the level of the receptor complex, and the specific regulation of NF-kB controlled responses induced by wild type TILRR and the R425 and D448 mutants. Further investigation with this model identifies a novel regulation of the IL-1 signalling response through the binding and release of IkBa by the cytoskeleton, which has previously been shown to sequester 2/3 of cellular IkBa.3 Here we demonstrate the relevance of the cytoskeletal pool of IkBa in NF-kB control using the computational model in combination with in vitro experiments. Conclusions Using agent-based modelling we predict that both cytoskeletal sequestering of IkBa and its release upon IL-1 stimulation impact the kinetics and levels of the cellular response. The effects are expected to play a significant part in the altered gene activation profiles induced by IL-1RI activation at sites of inflammation during development of atherosclerosis. Acknowledgements The research is supported by BBSRC grant BB/J009687/1 to EEQ and MWH. References Zhang X, Shephard F, Kim HB, Palmer IR, McHarg S, Fowler GJ, O’Neill LA, Kiss-Toth E, and Qwarnstrom EE. TILRR, a novel IL-1RI co-receptor, potentiates MYD88-recruitment to control ras-dependent amplification of NF-kappB. J Biol Chem 2010;285(10):7222–7232 Zhang X, Montago-Pino G, Shephard F, Kiss-Toth E, and Qwarnstrom EE. Distinct control of MyD88 adapter-dependent and Akt kinase regulated responses by the IL-1RI co-receptor TILRR . J Biol Chem2012 ;287:12348–12352; doi:10.1074/jbc.C111.321711 Pogson M, Holcombe M, Smallwood R, Yang L, and Qwarnstrom EE. Introducing spatial information into predictive NF-kB modelling- an agent based approach. PLoS ONE 2008;3(6):e2367

Introduction The toll-like and IL-1 receptor regulator (TILRR) is a co-receptor that binds the type I IL-1 signalling receptor (IL-1RI), and enhances signal amplification at the level of the receptor complex.1–3 IL-1 is a pro-inflammatory cytokine, central to development of vascular diseases. Initial experiments demonstrated high levels of TILRR expression in atherosclerotic plaques in human vessels. In this study we evaluate the role of TILRR in development of atherosclerosis, and in vascular remodelling following carotid artery ligation using a TILRR knockout mice. Methods and results Initial histological analysis of major organs from TILRR-/- mice demonstrated the same morphology as in tissues from wild type mice. To assess the role of TILRR on vascular repair TILRR-/- mice were used in a model of carotid artery ligation, and neointima formation and smooth muscle cell proliferation were assessed up to 4 weeks. Immunostaining of serial cross sections demonstrated high expression levels of TILRR in the area of ligation. TILRR knockout caused a decrease in neointima size by 48%, and about a 50% reduction in intima/media ratio after 4 weeks of ligation. Further, the number of smooth muscle cells in the neointima and medial area corresponded to 56% of level in wild type mice. To evaluate the role of TILRR in atherosclerosis, ApoE-/- mice we kept on a high fat diet for 8 weeks and injected with a peptide antibody directed against the TILRR functional site (i.v. 130ng/kg; twice a week). Immunohistochemical analysis of lesions from ApoE-/- mice revealed extensive expression of TILRR in atherosclerotic plaques. TILRR-antibody treatment resulted in reduction of plaque size in the aortic root by 28% and in the brachiocephalic artery by 40%. In addition, blocking TILRR function resulted in a two-fold increase in collagen content and a four-fold increase in SMC content of the lesions, characteristic of more stable plaques. TILRR-antibody treatment also reduced the number of elastin fibre breaks in the media and the number of buried fibrous caps within the plaques by about 60%. FACS analysis of blood demonstrated a reduction in inflammatory (Gr1-positive) monocytes by about 30%, known to be the major source of macrophages in plaques, and revealed a corresponding increase in the non-inflammatory phenotype. Conclusion The data demonstrate a role for TILRR control in in vivo models of vascular disease, with effects on blood vessel response to injury and on development of atherosclerosis, and are consistent with TILRR control of IL-1RI function and with reduced inflammatory responses in the TILRR KO mice. Supported by the British Heart Foundation. References Zhang X, et al. J Biol Chem 2010;285(10):7222–7232 Zhang X, et al. J Biol Chem 2012;287:12348–12352 Hudson RC, et al. MDPI Biology 2012;1:484–494

Introduction The toll-like and IL-1 receptor regulator (TILRR) is a recently described co-receptor that binds the type I IL-1 signalling receptor (IL-1RI), and enhances signal amplification at the level of the receptor complex. IL-1 is a pro-inflammatory cytokine with a central role in responses linked to progression of atherosclerosis. The current project assesses the impact of TILRR on vascular pathology and remodelling. Immunocytochemistry demonstrated a cell specific increase in TILRR expression in the atherosclerotic plaque and a transient up-regulation in response to vessel injury. The data are consistent with a role for TILRR in IL-1-mediated signals during atherosclerosis and vascular repair. Methods and Results Atherosclerotic plaques from human and mice (ApoE−/−, LDLR−/−) were stained with an anti-TILRR antibody and the monocyte marker Mac-3, using two-colour fluorescence. The data show a high level of TILRR expression in areas containing vascular lesions, with low levels in undamaged tissue. Specifically, the experiments demonstrate a correlation of TILRR expression (DyLight 594) within the plaque, with staining for Mac-3 (DyLight 488), in agreement with data from FACS analysis showing high levels of TILRR in the macrophage/monocyte population in atherosclerotic mice. The role of TILRR in development of atherosclerosis was assessed by selective blocking of TILRR function in ApoE −/− mice on a high fat diet. Results showed a significant reduction in plaque development following injection of an antibody targeting TILRR/IL-1RI interaction. These observations are consistent with initial results from screening human samples by qPCR, which demonstrate increased TILRR expression in patients with vascular disease compared to levels in healthy controls. A set of experiments, which used carotid ligation to assess the role of TILRR in vascular remodelling revealed a transient increase in TILRR expression at early times after ligation (2 wks) with only weak staining at later stages (4 wks). Ongoing studies use TILRR KO to further analyse the role of TILRR in vascular repair and disease. Conclusion The data demonstrate an increase in TILRR expression in macrophage/monocyte cell types and are consistent with a role for TILRR in the inflammatory response underlying development of atherosclerosis. Further, the transient increase in expression following carotid ligation suggests a role for TILRR in early stages of vascular repair.

The TLRs and IL-1 receptors have evolved to coordinate the innate immune response following pathogen invasion. Receptors and signalling intermediates of these systems are generally characterised by a high level of evolutionary conservation. The recently described IL-1R1 co-receptor TILRR is a transcriptional variant of the FREM1 gene. Here we investigate whether innate co-receptor differences between teleosts and mammals extend to the expression of the TILRR isoform of FREM1. Bioinformatic and phylogenetic approaches were used to analyse the genome sequences of FREM1 from eukaryotic organisms including 37 tetrapods and five teleost fish. The TILRR consensus peptide sequence was present in the FREM1 gene of the tetrapods, but not in fish orthologs of FREM1, and neither FREM1 nor TILRR were present in invertebrates. The TILRR gene appears to have arisen via incorporation of adjacent non-coding DNA with a contiguous exonic sequence after the teleost divergence. Comparing co-receptors in other systems, points to their origin during the same stages of evolution. Our results show that modern teleost fish do not possess the IL-1RI co-receptor TILRR, but that this is maintained in tetrapods as early as amphibians. Further, they are consistent with data showing that co-receptors are recent additions to these regulatory systems and suggest this may underlie differences in innate immune responses between mammals and fish.

Toll-like and IL-1 receptors control inflammatory responses. TILRR (Toll-like IL-1 Receptor Regulator), is an IL-1 co-receptor which associates with the type I IL-1 receptor (IL-1RI) to amplify activation of NF-kappaB and inflammatory responses. Earlier studies have demonstrated increased levels of TILRR in the atherosclerotic plaque. Further, that injection of a polyclonal anti-TILRR antibody, which blocks TILRR/IL-1RI association and reduces inflammatory responses, causes a 25% decrease in plaque formation in ApoE-/- mice on a high fat diet. Alanine scanning mutagenesis identified two sites within the TILRR core protein, which allow distinct control of Il-1 activities. The R425A substitution blocks enhanced cell survival, but functions as wild-type in relation to inflammatory responses. In contrast, a D448A substitution reduces MyD88-dependent inflammatory responses, but has no impact on cell survival [1]. Current studies use peptides designed to block these distinct functional sites to further analyse consequences of selective inhibition on downstream events. Results show a successive reduction in inflammatory responses by the peptide designed to block D448 dependent interactions, with no effect of the peptide targeting anti-apoptotic signals or a non-specific control. Ongoing studies are testing the effect of the peptides on IL-1-induced cell survival. 1. Zhang, X., et al., JBC, 2012, doi:10.1074/jbc.C111.321711