Perturbation biology nominates upstream-downstream drug combinations in RAF inhibitor resistant melanoma cells.

Resistance to targeted cancer therapies is an important clinical problem. The discovery of anti-resistance drug combinations is challenging as resistance can arise by diverse escape mechanisms. To address this challenge, we improved and applied the experimental-computational perturbation biology method. Using statistical inference, we build network models from high-throughput measurements of molecular and phenotypic responses to combinatorial targeted perturbations. The models are computationally executed to predict the effects of thousands of untested perturbations. In RAF-inhibitor resistant melanoma cells, we measured 143 proteomic/phenotypic entities under 89 perturbation conditions and predicted c-Myc as an effective therapeutic co-target with BRAF or MEK. Experiments using the BET bromodomain inhibitor JQ1 affecting the level of c-Myc protein and protein kinase inhibitors targeting the ERK pathway confirmed the prediction. In conclusion, we propose an anti-cancer strategy of co-targeting a specific upstream alteration and a general downstream point of vulnerability to prevent or overcome resistance to targeted drugs.

Spatial normalization of reverse phase protein array data.

Reverse phase protein arrays (RPPA) are an efficient, high-throughput, cost-effective method for the quantification of specific proteins in complex biological samples. The quality of RPPA data may be affected by various sources of error. One of these, spatial variation, is caused by uneven exposure of different parts of an RPPA slide to the reagents used in protein detection. We present a method for the determination and correction of systematic spatial variation in RPPA slides using positive control spots printed on each slide. The method uses a simple bi-linear interpolation technique to obtain a surface representing the spatial variation occurring across the dimensions of a slide. This surface is used to calculate correction factors that can normalize the relative protein concentrations of the samples on each slide. The adoption of the method results in increased agreement between technical and biological replicates of various tumor and cell-line derived samples. Further, in data from a study of the melanoma cell-line SKMEL-133, several slides that had previously been rejected because they had a coefficient of variation (CV) greater than 15%, are rescued by reduction of CV below this threshold in each case. The method is implemented in the R statistical programing language. It is compatible with MicroVigene and SuperCurve, packages commonly used in RPPA data analysis. The method is made available, along with suggestions for implementation, at http://bitbucket.org/rppa_preprocess/rppa_preprocess/src.

Perturbation biology: inferring signaling networks in cellular systems.

We present a powerful experimental-computational technology for inferring network models that predict the response of cells to perturbations, and that may be useful in the design of combinatorial therapy against cancer. The experiments are systematic series of perturbations of cancer cell lines by targeted drugs, singly or in combination. The response to perturbation is quantified in terms of relative changes in the measured levels of proteins, phospho-proteins and cellular phenotypes such as viability. Computational network models are derived de novo, i.e., without prior knowledge of signaling pathways, and are based on simple non-linear differential equations. The prohibitively large solution space of all possible network models is explored efficiently using a probabilistic algorithm, Belief Propagation (BP), which is three orders of magnitude faster than standard Monte Carlo methods. Explicit executable models are derived for a set of perturbation experiments in SKMEL-133 melanoma cell lines, which are resistant to the therapeutically important inhibitor of RAF kinase. The resulting network models reproduce and extend known pathway biology. They empower potential discoveries of new molecular interactions and predict efficacious novel drug perturbations, such as the inhibition of PLK1, which is verified experimentally. This technology is suitable for application to larger systems in diverse areas of molecular biology.

Drug synergy screen and network modeling in dedifferentiated liposarcoma identifies CDK4 and IGF1R as synergistic drug targets.

Dedifferentiated liposarcoma (DDLS) is a rare but aggressive cancer with high recurrence and low response rates to targeted therapies. Increasing treatment efficacy may require combinations of targeted agents that counteract the effects of multiple abnormalities. To identify a possible multicomponent therapy, we performed a combinatorial drug screen in a DDLS-derived cell line and identified cyclin-dependent kinase 4 (CDK4) and insulin-like growth factor 1 receptor (IGF1R) as synergistic drug targets. We measured the phosphorylation of multiple proteins and cell viability in response to systematic drug combinations and derived computational models of the signaling network. These models predict that the observed synergy in reducing cell viability with CDK4 and IGF1R inhibitors depends on the activity of the AKT pathway. Experiments confirmed that combined inhibition of CDK4 and IGF1R cooperatively suppresses the activation of proteins within the AKT pathway. Consistent with these findings, synergistic reductions in cell viability were also found when combining CDK4 inhibition with inhibition of either AKT or epidermal growth factor receptor (EGFR), another receptor similar to IGF1R that activates AKT. Thus, network models derived from context-specific proteomic measurements of systematically perturbed cancer cells may reveal cancer-specific signaling mechanisms and aid in the design of effective combination therapies.

Modeling local and global intracellular calcium responses mediated by diffusely distributed inositol 1,4,5-trisphosphate receptors.

Considerable insight into intracellular Ca2+ responses has been obtained through the development of whole cell models that are based on molecular mechanisms, e.g., single channel kinetics of the inositol 1,4,5-trisphosphate (IP3) receptor Ca2+ channel. However, a limitation of most whole cell models to date is the assumption that IP3 receptor Ca2+ channels (IP3Rs) are globally coupled by a "continuously stirred" bulk cytosolic [Ca2+], when in fact open IP3Rs experience elevated "domain" Ca2+ concentrations. Here we present a 2N+2-compartment whole cell model of local and global Ca2+ responses mediated by N=100,000 diffusely distributed IP3Rs, each represented by a four-state Markov chain. Two of these compartments correspond to bulk cytosolic and luminal Ca2+ concentrations, and the remaining 2N compartments represent time-dependent cytosolic and luminal Ca2+ domains associated with each IP3R. Using this Monte Carlo model as a starting point, we present an alternative formulation that solves a system of advection-reaction equations for the probability density of cytosolic and luminal domain [Ca2+] jointly distributed with IP3R state. When these equations are coupled to ordinary differential equations for the bulk cytosolic and luminal [Ca2+], a realistic but minimal model of whole cell Ca2+ dynamics is produced that accounts for the influence of local Ca2+ signaling on channel gating and global Ca2+ responses. The probability density approach is benchmarked and validated by comparison to Monte Carlo simulations, and the two methods are shown to agree when the number of Ca2+ channels is large (i.e., physiologically realistic). Using the probability density approach, we show that the time scale of Ca2+ domain formation and collapse (both cytosolic and luminal) may influence global Ca2+ oscillations, and we derive two reduced models of global Ca2+ dynamics that account for the influence of local Ca2+ signaling on global Ca2+ dynamics when there is a separation of time scales between the stochastic gating of IP3Rs and the dynamics of domain Ca2+.