The Menkes protein (MNK) and Wilson protein (WND) are transmembrane, CPX-type Cu-ATPases with six metal binding sites (MBSs) in the N-terminal region containing the motif GMXCXXC. In cells cultured in low copper concentration MNK and WND localize to the transGolgi network but in high copper relocalize either to the plasma membrane (MNK) or a vesicular compartment (WND). In this paper we investigate the role of the MBSs in Cu-transport and trafficking. The copper transport activity of MBS mutants of MNK was determined by their ability to complement a strain of Saccharomyces cerevisiae deficient in CCC2 (delta ccc2), the yeast MNK/WND homologue. Mutants (CXXC to SXXS) of MBS1, MBS6, and MBSs1-3 were able to complement delta ccc2 while mutants of MBS4-6, MBS5-6 and all six MBS inactivated the protein. Each of the inactive mutants also failed to display Cu-induced trafficking suggesting a correlation between trafficking and transport activity. A similar correlation was found with mutants of MNK in which various MBSs were deleted, but two constructs with deletion of MBS5-6 were unable to traffic despite retaining 25% of copper transport activity. Chimeras in which the N-terminal MBSs of MNK were replaced with the corresponding MBSs of WND were used to investigate the region of the molecules that is responsible for the difference in Cu-trafficking of MNK and WND. The chimera which included the complete WND N-terminus localized to a vesicular compartment, similar to WND in elevated copper. Deletions of various MBSs of the WND N-terminus in the chimera indicate that a targeting signal in the region of MBS6 directs either WND/MNK or WND to a vesicular compartment of the cell.

Context. Separators, which are in many ways the three-dimensional equivalent to two-dimensional nulls, are important sites for magnetic reconnection. Magnetic reconnection occurs in strong current layers which have very short length scales. Aims. The aim of this work is to explore the nature of current layers around separators. A separator is a special field line which lies along the intersection of two separatrix surfaces and forms the boundary between four topologically distinct flux domains. In particular, here the current layer about a separator that joins two 3D nulls and lies along the intersection of their separatrix surfaces is investigated. Methods. A magnetic configuration containing a single separator embedded in a uniform plasma with a uniform electric current parallel to the separator is considered. This initial magnetic setup, which is not in equilibrium, relaxes in a non-resistive manner to form an equilibrium. The relaxation is achieved using the 3D MHD code, Lare3d, with resistivity set to zero. A series of experiments with varying initial current are run to investigate the characteristics of the resulting current layers present in the final (quasi-)equilibrium states. Results. In each experiment, the separator collapses and a current layer forms along it. The dimensions and strength of the current layer increase with initial current. It is found that separator current layers formed from current parallel to the separator are twisted. Also the collapse of the separator is a process that evolves like an infinite-time singularity where the length, width and peak current in the layer grow slowly whilst the depth of the current layer decreases.

Magnetic separators, which lie on the boundary between four topologically distinct flux domains, are prime locations in three‐dimensional magnetic fields for reconnection, especially in the magnetosphere between the planetary and interplanetary magnetic fields and also in the solar atmosphere. Little is known about the details of separator reconnection, and so the aim of this paper, which is the first of two, is to study the properties of magnetic reconnection at a single separator. Three‐dimensional, resistive magnetohydrodynamic numerical experiments are run to study separator reconnection starting from a magnetohydrostatic equilibrium which contains a twisted current layer along a single separator linking a pair of opposite‐polarity null points. The resulting reconnection occurs in two phases. The first is short involving rapid reconnection in which the current at the separator is reduced by a factor of around 2.3. Most (75%) of the magnetic energy is converted during this phase, via Ohmic dissipation, directly into internal energy, with just 0.1% going into kinetic energy. During this phase the reconnection occurs along most of the separator away from its ends (the nulls) but in an asymmetric manner which changes both spatially and temporally over time. The second phase is much longer and involves slow impulsive bursty reconnection. Again, Ohmic heating dominates over viscous damping. Here the reconnection occurs in small localized bursts at random anywhere along the separator.

Aims. The aim of this work is to investigate and characterise particle behaviour in a 3D magnetohydrodynamic (MHD) model of a reconnecting magnetic separator. Methods. We use a relativistic guiding-centre test-particle code to investigate electron and proton acceleration in snapshots from 3D MHD separator reconnection experiments, and compare the results with findings from an analytical separator reconnection model studied in a previous investigation. Results. The behaviour and acceleration of large distributions of particles are examined in detail for both analytical and numerical separator reconnection models. Accelerated particle orbit trajectories are shown to follow the separator before leaving the system along the separatrix surface of one of the nulls (determined by particle species) in the system of both models. A sufficiently localised electric field about the separator causes the orbits to appear to follow the spine bounding the separatrix surface field lines instead. We analyse and discuss the locations and spread of accelerated particle orbit final positions, which are seen to change over time in the numerical separator reconnection model. We deduce a simple relationship between the final energy range of particle orbits and the model dimensions, and discuss its implications for observed magnetic separators in the solar corona.