Different cancer cell lines resistant to the same drug exhibit differences in folate pathway dynamics Amrisha Bhosle and Nagasuma Chandra * Department of Biochemistry Indian Institute of Science Bangalore, India * Email: nchandra@biochem.iisc.ernet.in Abstract—Use of methotrexate (MTX), a widely used anti- cancer drug which targets primarily dihydrofolate reductase (DHFR) in the folate pathway is being limited by the emergence of resistance. Despite a large number of studies, a quantitative understanding of target pathway dynamics in resistant cancers is majorly lacking. In this work, we integrated gene expression data from different MTX-resistant cancer cell lines into kinetic models to study dynamics of the folate metabolic pathway. Given that all cell lines are derived from human cancers, the pathway is essentially the same consisting of 11 reactions catalyzed by the same set of enzymes and 1 non-enzymatic reaction. Kinetic models emulating pathway dynamics in MTX-untreated, MTX-treated- sensitive and MTX-treated-resistant conditions were generated and model behaviour at steady state was analysed with respect to concentrations of six folate metabolites and fluxes through the 12 reactions. We observed differences in steady-state properties across these cell lines even in the absence of MTX inhibition. More interestingly, the response of sensitive and resistant variants of each cancer type was also seen to vary in simulations of MTX- inhibition. However, accumulation of dihydrofolate at steady state for all sensitive cell lines along and a decrease towards normal levels for their resistant counterparts remained a common feature in most cases. Metabolic control analysis performed to identify crucial flux controlling elements in the pathway indicated that the enzymes methenyltetrahydrofolate cyclohydrolase (MTCH) and phosphoribosylglycinamide formyltransferase (PGT) could be targeted in combination with DHFR in MTX-resistant cancers for improved therapy. Keywords—Methotrexate effect, Gene expression profile, Kinetic model, Metabolic control analysis I. I NTRODUCTION Drug resistance is a major problem in the chemothera- peutic management of cancer. Resistance to anti-cancer drugs decreases drug efficacy and facilitates disease progression. It can be broadly classified into two categories: acquired or intrinsic. Mutations in drug target and increased production of target protein are two of the common mechanisms of acquired drug resistance. Methotrexate (MTX; 4-amino- N10- methylpteroylglutamic acid), a notable antineoplastic agent is a potent competitive inhibitor of dihydrofolate reductase (DHFR); an enzyme in the folate pathway. MTX and its polyg- We thank the Department of Biotechnology (DBT) and the Mathematical Biology Programe supported by the Department of Science and Technology (DST), Government of India, for financial support. The use of facilities at Department of Biochemistry, Indian Institute of Science is also gratefully acknowledged. lutamated forms also inhibit other enzymes in the same path- way such as thymidylate synthase (TS) non-competitively and AICAR transformylase (AICART) and GAR transformylase (PGT) competitively [1]. Chemotherapy with MTX has been rendered ineffective due to DHFR over-expression in resistant cancers [2], [3] and binding site mutations that reduce MTX binding affinity [4], [5]. This necessitates the development of new drug targeting strategies that can be used to alleviate if not obliterate the effects of resistance. The folate pathway lies at the interface of protein and nucleic acid metabolism pathways and therefore is of enor- mous pharmacological importance. Three most essential folate metabolites include 5-methyl THF required for methionine synthesis, 5, 10-methylene THF that provides C1 units for synthesis of deoxythymidine monophosphate (dTMP) and 10- formyl THF that is utilized as a co-factor in de novo purine synthesis reactions. Inhibition of this pathway depletes the cell of dTMP required for DNA synthesis thereby halting cell proliferation. Studies reporting change in pathway dynamics on inhibition by small molecules have been carried out employing mass spectrometric analyses and mathematical modelling [6], [7], [8]. Dynamics of complex biochemical pathways can be studied through numeric integration of inter-connected enzymatic reac- tions represented as a mathematical model consisting of ordi- nary differential equations. Different parameters in the model such as rate constants, V max and K M affect system behavior to different extents which can be measured by performing sensitivity analysis on the model. Metabolic Control Analysis (MCA) [9]; a form of sensitivity analysis is a powerful technique that can be used to quantify the contribution of every parameter in a mathematical model that captures a biological phenomemon such as drug resistance. MCA offers a scaled sensitivity analysis through Control coefficients which measure the response of a particular system variable to infinitesimal changes in rate of an enzymatic reaction which can be altered by chemotherapeutic agents. Use of MCA in identification of drug targets has been discussed in several studies [10], [11], [12]. In this study we generated models for seven cancer cell lines by incorporating (1) gene expression data for the 11 en- zymes in the pathway and (2) changes in enzymatic parameters due to enzyme inhibition in the basic folate pathway kinetic model [8]. Model behavior at steady state was studied for all cancer cell lines under three conditions viz. MTX-untreated (MU), MTX-treated-sensitive (MTS) and MTX-treated-resistant 2014 IEEE International Conference on Bioinformatics and Biomedicine 978-1-4799-5669-2/14/$31.00 ©2014 IEEE 209