Alexey Goltsov

Systems biology reveals new strategies for personalizing cancer medicine and confirms the role of PTEN in resistance to trastuzumab

Resistance to targeted cancer therapies such as trastuzumab is a frequent clinical problem not solely because of insufficient expression of HER2 receptor but also because of the overriding activation states of cell signaling pathways. Systems biology approaches lend themselves to rapid in silico testing of factors, which may confer resistance to targeted therapies. Inthis study, we aimed to develop a new kinetic model that could be interrogated to predict resistance to receptor tyrosine kinase (RTK) inhibitor therapies and directly test predictions in vitro and in clinical samples. The new mathematical model included RTK inhibitor antibody binding, HER2/HER3 dimerization and inhibition, AKT/mitogen-activated protein kinase cross-talk, and the regulatory properties of PTEN. The model was parameterized using quantitative phosphoprotein expression data from cancer cell lines using reverse-phase protein microarrays. Quantitative PTEN protein expression was found to be the key determinant of resistance to anti-HER2 therapy in silico, which was predictive of unseen experiments in vitro using the PTEN inhibitor bp(V). When measured in cancer cell lines, PTEN expression predicts sensitivity to anti-HER2 therapy; furthermore, this quantitative measurement is more predictive of response (relative risk, 3.0; 95% confidence interval, 1.6-5.5; P < 0.0001) than other pathway components taken in isolation and when tested by multivariate analysis in a cohort of 122 breast cancers treated with trastuzumab. For the first time, a systems biology approach has successfully been used to stratify patients for personalized therapy in cancer and is further compelling evidence that PTEN, appropriately measured in the clinical setting, refines clinical decision making in patients treated with anti-HER2 therapies.

Infrared laser pulse triggers increased singlet oxygen production in tumour cells

Photodynamic therapy (PDT) is a technique developed to treat the ever-increasing global incidence of cancer. This technique utilises singlet oxygen (1O2) generation via a laser excited photosensitiser (PS) to kill cancer cells. However, prolonged sensitivity to intensive light (6–8 weeks for lung cancer), relatively low tissue penetration by activating light (630 nm up to 4 mm), and the cost of PS administration can limit progressive PDT applications. The development of quantum-dot laser diodes emitting in the highest absorption region (1268 nm) of triplet oxygen (3O2) presents the possibility of inducing apoptosis in tumour cells through direct 3O2 → 1O2 transition. Here we demonstrate that a single laser pulse triggers dose-dependent 1O2 generation in both normal keratinocytes and tumour cells and show that tumour cells yield the highest 1O2 far beyond the initial laser pulse exposure. Our modelling and experimental results support the development of direct infrared (IR) laser-induced tumour treatment as a promising approach in tumour PDT.

Model-based global sensitivity analysis as applied to identification of anti-cancer drug targets and biomarkers of drug resistance in the ErbB2/3 network

High levels of variability in cancer-related cellular signalling networks and a lack of parameter identifiability in large-scale network models hamper translation of the results of modelling studies into the process of anti-cancer drug development. Recently global sensitivity analysis (GSA) has been recognised as a useful technique, capable of addressing the uncertainty of the model parameters and generating valid predictions on parametric sensitivities. Here we propose a novel implementation of model-based GSA specially designed to explore how multi-parametric network perturbations affect signal propagation through cancer-related networks. We use area-under-the-curve for time course of changes in phosphorylation of proteins as a characteristic for sensitivity analysis and rank network parameters with regard to their impact on the level of key cancer-related outputs, separating strong inhibitory from stimulatory effects. This allows interpretation of the results in terms which can incorporate the effects of potential anti-cancer drugs on targets and the associated biological markers of cancer. To illustrate the method we applied it to an ErbB signalling network model and explored the sensitivity profile of its key model readout, phosphorylated Akt, in the absence and presence of the ErbB2 inhibitor pertuzumab. The method successfully identified the parameters associated with elevation or suppression of Akt phosphorylation in the ErbB2/3 network. From analysis and comparison of the sensitivity profiles of pAkt in the absence and presence of targeted drugs we derived predictions of drug targets, cancer-related biomarkers and generated hypotheses for combinatorial therapy. Several key predictions have been confirmed in experiments using human ovarian carcinoma cell lines. We also compared GSA-derived predictions with the results of local sensitivity analysis and discuss the applicability of both methods. We propose that the developed GSA procedure can serve as a refining tool in combinatorial anti-cancer drug discovery.

Compensatory effects in the PI3K/PTEN/AKT signaling network following receptor tyrosine kinase inhibition

Overcoming de novo and acquired resistance to anticancer drugs that target signaling networks is a formidable challenge for drug design and effective cancer therapy. Understanding the mechanisms by which this resistance arises may offer a route to addressing the insensitivity of signaling networks to drug intervention and restore the efficacy of anticancer therapy. Extending our recent work identifying PTEN as a key regulator of Herceptin sensitivity, we present an integrated theoretical and experimental approach to study the compensatory mechanisms within the PI3K/PTEN/AKT signaling network that afford resistance to receptor tyrosine kinase (RTK) inhibition by anti-HER2 monoclonal antibodies. In a computational model representing the dynamics of the signaling network, we define a single control parameter that encapsulates the balance of activities of the enzymes involved in the PI3K/PTEN/AKT cycle. By varying this control parameter we are able to demonstrate both distinct dynamic regimes of behavior of the signaling network and the transitions between those regimes. We demonstrate resistance, sensitivity, and suppression of RTK signals by the signaling network. Through model analysis we link the sensitivity-to-resistance transition to specific compensatory mechanisms within the signaling network. We study this transition in detail theoretically by variation of activities of PTEN, PI3K, AKT enzymes, and use the results to inform experiments that perturb the signaling network using combinatorial inhibition of RTK, PTEN, and PI3K enzymes in human ovarian carcinoma cell lines. We find good alignment between theoretical predictions and experimental results. We discuss the application of the results to the challenges of hypersensitivity of the signaling network to RTK signals, suppression of drug resistance, and efficacy of drug combinations in anticancer therapy.

Engineering simulations for cancer systems biology

Computer simulation can be used to inform in vivo and in vitro experimentation, enabling rapid, low-cost hypothesis generation and directing experimental design in order to test those hypotheses. In this way, in silico models become a scientific instrument for investigation, and so should be developed to high standards, be carefully calibrated and their findings presented in such that they may be reproduced. Here, we outline a framework that supports developing simulations as scientific instruments, and we select cancer systems biology as an exemplar domain, with a particular focus on cellular signalling models. We consider the challenges of lack of data, incomplete knowledge and modelling in the context of a rapidly changing knowledge base. Our framework comprises a process to clearly separate scientific and engineering concerns in model and simulation development, and an argumentation approach to documenting models for rigorous way of recording assumptions and knowledge gaps. We propose interactive, dynamic visualisation tools to enable the biological community to interact with cellular signalling models directly for experimental design. There is a mismatch in scale between these cellular models and tissue structures that are affected by tumours, and bridging this gap requires substantial computational resource. We present concurrent programming as a technology to link scales without losing important details through model simplification. We discuss the value of combining this technology, interactive visualisation, argumentation and model separation to support development of multi-scale models that represent biologically plausible cells arranged in biologically plausible structures that model cell behaviour, interactions and response to therapeutic interventions.

Feedforward and feedback regulation of the MAPK and PI3K oscillatory circuit in breast cancer

Although the theoretical possibility of oscillations in MAPK signalling has long been described, experimental validation has proven more elusive. In this study we observed oscillations in MAPK and PI3K signalling in breast cancer cells in response to epidermal growth factor receptor-family stimulation. Using systems level analysis with a kinetic model, we demonstrate that receptor amplification, loss of transcriptional feedback, or pathway crosstalk, are responsible for oscillations in MAPK and PI3K signalling. Transcriptional profiling reveals architectural motifs likely to be responsible for feedback control of oscillations. Overexpression of the HER2 oncogene and inhibition of transcriptional feedback increase the amplitude of oscillations and provide experimental validation of the computational findings.

Features of the reversible sensitivity-resistance transition in PI3K/PTEN/AKT signalling network after HER2 inhibition

Systems biology approaches that combine experimental data and theoretical modelling to understand cellular signalling network dynamics offer a useful platform to investigate the mechanisms of resistance to drug interventions and to identify combination drug treatments. Extending our work on modelling the PI3K/PTEN/AKT signalling network (SN), we analyse the sensitivity of the SN output signal, phospho-AKT, to inhibition of HER2 receptor. We model typical aberrations in this SN identified in cancer development and drug resistance: loss of PTEN activity, PI3K and AKT mutations, HER2 overexpression, and overproduction of GSK3β and CK2 kinases controlling PTEN phosphorylation. We show that HER2 inhibition by the monoclonal antibody pertuzumab increases SN sensitivity, both to external signals and to changes in kinetic parameters of the proteins and their expression levels induced by mutations in the SN. This increase in sensitivity arises from the transition of SN functioning from saturation to non-saturation mode in response to HER2 inhibition. PTEN loss or PIK3CA mutation causes resistance to anti-HER2 inhibitor and leads to the restoration of saturation mode in SN functioning with a consequent decrease in SN sensitivity. We suggest that a drug-induced increase in SN sensitivity to internal perturbations, and specifically mutations, causes SN fragility. In particular, the SN is vulnerable to mutations that compensate for drug action and this may result in a sensitivity-to-resistance transition. The combination of HER2 and PI3K inhibition does not sensitise the SN to internal perturbations (mutations) in the PI3K/PTEN/AKT pathway: this combination treatment provides both synergetic inhibition and may prevent the SN from acquired mutations causing drug resistance. Through combination inhibition treatments, we studied the impact of upstream and downstream interventions to suppress resistance to the HER2 inhibitor in the SN with PTEN loss. Comparison of experimental results of PI3K inhibition in the PTEN upstream pathway with PDK1 inhibition in the PTEN downstream pathway shows that upstream inhibition abrogates resistance to pertuzumab more effectively than downstream inhibition. This difference in inhibition effect arises from the compensatory mechanism of an activation loop induced in the downstream pathway by PTEN loss. We highlight that drug target identification for combination anti-cancer therapy needs to account for the mutation effects on the upstream and downstream pathways.

Systems analysis of drug-induced receptor tyrosine kinase reprogramming following targeted mono- and combination anti-cancer therapy

The receptor tyrosine kinases (RTKs) are key drivers of cancer progression and targets for drug therapy. A major challenge in anti-RTK treatment is the dependence of drug effectiveness on co-expression of multiple RTKs which defines resistance to single drug therapy. Reprogramming of the RTK network leading to alteration in RTK co-expression in response to drug intervention is a dynamic mechanism of acquired resistance to single drug therapy in many cancers. One route to overcome this resistance is combination therapy. We describe the results of a joint in silico, in vitro, and in vivo investigations on the efficacy of trastuzumab, pertuzumab and their combination to target the HER2 receptors. Computational modelling revealed that these two drugs alone and in combination differentially suppressed RTK network activation depending on RTK co-expression. Analyses of mRNA expression in SKOV3 ovarian tumour xenograft showed up-regulation of HER3 following treatment. Considering this in a computational model revealed that HER3 up-regulation reprograms RTK kinetics from HER2 homodimerisation to HER3/HER2 heterodimerisation. The results showed synergy of the trastuzumab and pertuzumab combination treatment of the HER2 overexpressing tumour can be due to an independence of the combination effect on HER3/HER2 composition when it changes due to drug-induced RTK reprogramming.

Synergetics of the Membrane Self-Assembly: A Micelle-to-Vesicle Transition.

A model approach is developed to study intermediate steps and transientstructures in a course of the membrane self-assembly. The approach isbased on investigation of mixed lipid/protein-detergent systems capable ofthe temperature induced transformation from a solubilized micellar stateto closed membrane vesicles. We performed a theoretical analysis ofself-assembling molecular structures formed in binary mixtures ofdimyristoylphosphatidylcholine (DMPC) and sodium cholate (NaC). Thetheoretical model is based on the Helfrich theory of curvature elasticity,which relates geometrical shapes of the structures to their free energy inthe Ginzburg-Landau approximation. The driving force for the shapetransformation is spontaneous curvature of amphiphilic aggregates which isnonlinearly dependent on the lipid/detergent composition. An analysis ofthe free energy in the regular solution approximation shows that theformation of mixed structures of different shapes (discoidal micelles,rod-like micelles, multilayer membrane structures and vesicles) ispossible in a certain range of detergent/lipid ratios. A transition fromthe flat discoidal micelles to the rod-like cylindrical micelles isinduced by curvature instabilities resulting from acyl chain melting andinsertion of detergent molecules into the lipid phase. Nonideal mixing ofthe NaC and DMPC molecules results in formation of nonideal cylindricalaggregates with elliptical cross section. Further dissolution of NaCmolecules in DMPC may be accompanied with a change of their orientation inthe lipid phase and leads to temperature-induced curvature instabilitiesin the highly curved cylindrical geometry. As a result the rod-likemicelles fuse into less curved bilayer structures which transformeventually to the unilamellar and multilamellar membrane vesicles. Thetheoretical analysis performed shows that a sequence of shapetransformations in the DMPC/NaC mixed systems is determined by thesynergism of four major factors: detergent/lipid ratio, temperature (acylchain melting), DMPC and NaC mixing, and reorientation of NaC molecules inmixed aggregates.

Customizing the therapeutic response of signaling networks to promote antitumor responses by drug combinations

Drug resistance, de novo and acquired, pervades cellular signaling networks (SNs) from one signaling motif to another as a result of cancer progression and/or drug intervention. This resistance is one of the key determinants of efficacy in targeted anti-cancer drug therapy. Although poorly understood, drug resistance is already being addressed in combination therapy by selecting drug targets where SN sensitivity increases due to combination components or as a result of de novo or acquired mutations. Additionally, successive drug combinations have shown low resistance potential. To promote a rational, systematic development of combination therapies, it is necessary to establish the underlying mechanisms that drive the advantages of combination therapies, and design methods to determine drug targets for combination regimens. Based on a joint systems analysis of cellular SN response and its sensitivity to drug action and oncogenic mutations, we describe an in silico method to analyze the targets of drug combinations. Our method explores mechanisms of sensitizing the SN through a combination of two drugs targeting vertical signaling pathways. We propose a paradigm of SN response customization by one drug to both maximize the effect of another drug in combination and promote a robust therapeutic response against oncogenic mutations. The method was applied to customize the response of the ErbB/PI3K/PTEN/AKT pathway by combination of drugs targeting HER2 receptors and proteins in the down-stream pathway. The results of a computational experiment showed that the modification of the SN response from hyperbolic to smooth sigmoid response by manipulation of two drugs in combination leads to greater robustness in therapeutic response against oncogenic mutations determining cancer heterogeneity. The application of this method in drug combination co-development suggests a combined evaluation of inhibition effects together with the capability of drug combinations to suppress resistance mechanisms before they become clinically manifest.