CLAIX Projects 2019 High Performance Computing

Metal ion transporters have gained increased attention due to their implication in the neurodegenerative diseases’ onset. Among these the family of “solute carrier 11” (SLC11) or “natural resistance-associated macrophage proteins” (NRAMPs) are of intense interest since they function as secondary active transporters, regulating the influx of essential divalent transition-metal ions across cellular membranes [1-4]. DMT1 (divalent metal transporter 1), the best characterized member of this family, is widely expressed in different human tissues [5-6] and carries out coupled divalent transition-metal ions/H+ transport [7], exhibiting selectivity towards Mn2+, Fe2+, Co2+ and Cd2+. Interestingly, mutations in DMT1 cause diseases (anemia) in humans and rodents [8]. Two X-ray structures, one of Staphylococcus capitis DMT1 (ScaDMT) [9], capturing an inward-facing conformation (IFC) of the transporter (i.e. the access route opens towards the cytoplasm) and another one from the bacterium Eremococcus coleocola (EcoDMT) with DMT1 in apo form and an outward facing conformation (OFC), revealed that DMT1 [10,11] work via an alternating access mechanism during which the protein is opened either towards the cytoplasm (IFC) or the extracellular matrix (OFC). In the first structure [9], a Mn2+ ion was identified coordinated by the carbonyl oxygen of A223, and the side chains of D49, N52 and M226. This structure also showed that the mutations identified in human patients with anemia are scattered in different parts of DMT1. The transport mechanism of this prokaryotic DMT1 is similar to the human one and depends on pH. Hence, DMT1 might actually couple Mn2+ transport with the symport of H+, although also uncoupled transport mechanisms have been detected [9,11]. In this project, we have used multiscale simulations to unravel the transport mechanism of metal ions in DMT1. This knowledge may be of key importance for therapeutic regulation of metal ion dys-homeostasis with potential applications in neurodegeneration. Scientific work accomplished and results obtained This system is extremely interesting to study as it is one of the few metal ions transporters for which an atomic level structure has been resolved. Since metal ion dys-homeostasis is increasingly implicated in major human diseases such as cancer and, in particular, neurodegeneration, in this project we tackle different important mechanistic traits, whose understanding may lead to novel regulatory strategies associated to metal ions transport dysfunctions. To this end multiscale simulations (unbiased MD and enhanced-sampling QM/MM MD) have been employed to unravel the mechanism of transport of metal ions. Specifically, we have addressed specific points of the coupled H+/Mn2+ transport mechanism and the origin of the specificity towards certain metals. Project ID: jara0178 EMILIANO IPPOLITI Institute for Advanced Simulations and Institute for Neuroscience and Medicine (IAS-5/INM-9), FZ Jülich, Germany MARIA LETIZIA MERLINI PhD student, EPFL, Lausanne, Switzerland ALESSANDRA MAGISTRATO CNR-IOM and SISSA, Trieste, Italy URSULA RÖTHLISBERGER EPFL, Lausanne, Switzerland PAOLO CARLONI Institute for Advanced Simulations and Institute for Neuroscience and Medicine (IAS-5/INM-9), FZ Jülich, Germany DANIELE NARZI EPFL, Lausanne, Switzerland Life Sciences Life Sciences Figure 1: Mutations of DMT1 associated to human diseases’ onset. Basic Biological and Medical Research | DFG 201 16 | Scientific Reports | Basic Biological and Medical Research Sub-project A: The specific aims of this sub-project were: (i) understand the molecular mechanism of Mn2+ release towards the cytosol, starting from the IFC of ScaDMT; (ii) identify a stable binding site for Mn2+ in the OFC, starting from the X-ray structure of EcoDMT; (iii) unravel the pathway for H+ transport across the transporter, identifying and characterizing the role of the residues crucially involved in this transport. Sub-project B: This part of the project aims at (iv) elucidating the origin of metal ions selectivity of DMT1 towards specific divalent transition metal ions, and (v) finally, at understanding the impact of specific mutations associated to human disease on the structural dynamical and transport mechanism. Realization of the project To run the FF-based simulations of both wild type and mutated ScaDMT and EcoDMT we have used the GROMACS 2016.4 programs. The two systems have ~266,000 atoms, and scale very well up to 4 nodes (96 core) keeping a computational efficiency of 72.5%. Therefore, all the classical MD simulations in this project were performed with 96 cores. Simulations of the transport mechanism of Mn2+ and other metals (Zn2+) required so far unbiased and biased Metadynamics-based QM/MM MD simulations by using CPMD 4.1 code. For this kind of simulations, we used 6 nodes (144 cores) on CLAIX-2016 (pure MPI) as this represented a good compromise between scaling performances and computational cost. The used codes were not modified during the execution of this project. The granted computing time was employed completely.