Archives

  • 2018-07
  • 2018-10
  • 2018-11
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • Molecular dynamics studies have proved

    2022-01-14

    Molecular dynamics studies have proved an effective means of characterising transition state pathways between the inward- and outward-facing states, and providing the relative propensity for each state. For example, long-timescale simulations of the membrane-embedded LeuT dimer, totalling ∼21 μs, have been used to construct the free energy landscapes associated with the inward-facing to outward-facing transition. The extremely long timescales used in this study also highlighted the conformational atm kinase inhibitor and presence of low-energy “intermediate” states that may represent transition pathways from the outward-facing to inward-facing conformations [29]. Another MD study examining a membrane-embedded SERT dimer homology model showed that the type of ligand bound to the S1 site changes the propensity of SERT to transition towards an inward-facing or outward-facing state. The triplicate 60 ns simulations examined the effect of the substrate serotonin, the competitive inhibitor cocaine, and the non-competitive inhibitor noribogaine bound at the S1 site, and found that the competitive inhibitor cocaine promoted the outward-facing conformation, while serotonin and noribogaine induced transitions to the inward-facing state [32]. Similar findings have been shown for multiplex 100 ns simulations of a DAT homology model, performed with a range of inhibitors bound to the S1 site [33]. Of particular note is the conformational diversity sampled across the simulations from all three independent studies. One hallmark of MD simulations is the thermal noise inherent in the system from collective bond vibrations—molecular collisions reflective of a biological environment. While kinking and potential hinging of TM1a is observed in these simulations, collectively they show a complex ensemble of conformational changes associated with the transport cycle, which cannot be clearly delineated into any simplified model based on clear-cut morphing between crystallographic states.
    Membranes and lipids in SLC6 structure and function The role of lipids in SLC6 transporter dynamics, modulation and even structure determination is relatively unexplored. Membrane proteins are typically solubilised in detergent while being prepared for crystallographic structural determination, which removes the protein from its native bilayer environment. To date, only one SLC6 transporter, LeuT, has been crystallised in a lipid phase [34,35]. Including or removing lipids from the environment has non-trivial effects on protein structure: recent comparative studies have shown root-mean-square deviations (RMSDs) of up to 5 Å atm kinase inhibitor between structures of membrane proteins derived using crystallography and nuclear magnetic resonance (NMR) attributed to environmental differences [36]. In particular, NMR ensembles tend to include more kinks and a more diffuse packing arrangement in the TM helices for a given membrane protein than its corresponding crystal structure [36]. Likewise, gas phase mass spectrometry techniques show increased stability of membrane proteins in the presence of native lipids compared to their detergent solubilised counterparts [37]. These findings are supported by MD simulations of LeuT in DDM micelles, designed to mimic crystallographic conditions, which found that DDM binds to the proposed S2 substrate binding site, altering the transport stoichiometry of LeuT [38]. A follow-up computational investigation highlighted the effect of reconstitution environment: the retention of a phospholipid annulus around LeuT within the DDM micelle prevents the binding of DDM to the S2 site [12]. Therefore, if the membrane protein has been removed from its native lipid environment during crystallisation, any assessments of the impact of bilayer composition on its structural conformation, and how it influences protein function, should be interpreted with caution. Despite a large body of research into membrane proteins, it is challenging to accurately predict their specific lipid binding sites. For instance, one recent study examined 100 high resolution membrane protein crystal structures containing bound membrane lipids, and did not identify any common sequence motifs that could be used to predict lipid binding sites [39]. However, recent crystal structures of eukaryotic SLC6 transporters show densities for bound cholesterol, suggesting this lipid plays a structural role in transporter function. In DAT, cholesterol was resolved bound to the intracellular halves of TM1 and TM5 [7], while in SERT, cholesterol is bound to the extracellular half of TM12 [8]. In other structures, cholesterol [40,41] and the cholesterol-related compound cholesterol hemisuccinate [40] are bound to the intracellular halves of TM2, TM7 and TM11 of DAT [40,41]. While crystal structures of eukaryotic SLC6 transporters show distinct cholesterol-binding sites, as shown in yellow in Fig. 1B, only three [34] of the 53 prokaryotic structures, of LeuT, show a lipid or lipid-related compound bound to the transporter. Additionally, electron densities for a range of other molecules, including sugars and detergents, decorate the surface of SLC6 transporter crystal structures, as shown in orange and maroon in Fig. 1B. It is unclear whether these molecules, especially the lipids, are only required to crystallise the protein, or if they perform functional roles.