| Is protein apo? |
Proteins may acquire metals in vivo during expression or in vitro during isolation (e.g. from buffers, salts, IMAC resins) which may be removed by incubation of protein with a strong metal chelator (e.g. EDTA) before the final purification step. |
| Are metal ions and protein cysteinyl ligands fully reduced? |
In studies that involve redox-active metal ions (e.g. FeII or CuI) or sulfhydryl groups of protein cysteines that act as metal ligands (both sensitive to oxidation), it is essential to maintain reducing conditions during the metal-binding assay. Proteins may be purified with reductant(s) such as DTT (and related thiols) or TCEP but, these reductants are strong metal chelators [91,104] and thus the purified proteins should ideally be transferred into an anaerobic chamber and separated via buffer-exchange into a deoxygenated buffer unless the reductants are demonstrated to have no interference with the downstream metal-binding experiments. Metals (ideally prepared in anaerobic solutions) should also be completely reduced under the experimental conditions, with [metal]red quantified by reaction with excess ligand (e.g. Bca, Bcs for CuI [8]; Fz for FeII [105]). Ellman's reagent can be used to quantify reduced protein thiols under denaturing conditions [106,107]. |
| Are probe and metal concentrations correct? |
Apo-probes can be quantified via absorbance using known extinction coefficients or by titration of probe with a calibrated metal stock (the latter is strongly recommended). Metal concentrations can be determined by AAS, ICP-MS, or simply by titration of metal into a calibrated probe solution. |
| Are buffers metal-free? |
Many buffers and salts contain trace metals. Glassware used for preparation of solutions can also contain metal contamination. For steps following removal of chelators (see above) select high-grade buffers with certified low levels of trace metal contamination. Alternatively, chelex-resins can be used to remove metal ions from reaction buffers and the metal-free solutions can be eluted into (i) glassware that has been washed with nitric acid (4% (v/v), overnight); or (ii) directly into clean plasticware. |
| Are buffer components competing for metal-binding? |
While strong metal chelators must be eliminated from metal-binding experiments, many buffers (e.g. Tris, phosphate) and salts (e.g. Cl−) display weak metal affinities and may contribute to metal speciation if present at high (millimolar) concentrations (e.g. millimolar [Cl−] can compete with Fs or Fz for binding Cu(I), see ref. [33]). A control metal titration of a probe-only solution can establish the reliability of the probe response and the potential contributions of buffers and salts (see section ‘Non-competitive controls, medium pH and buffers’). Non-coordinating buffers are available [95] and are preferred, especially for weak metal-binding experiments. |
| Are purification tags interfering? |
Purification tags with comparable metal affinities to the protein sites interfere with downstream experiments and thus must be cleaved post-purification. However, non-competing purification tags can be retained (see Figure 2F and ref. [35]). |
| Is pH controlled? |
Metal affinities and spectroscopic properties of probe ligands can be highly pH-dependent. Metals, ligands and reductants can influence buffer pH. Reaction solutions should be carefully monitored to ensure that pH is unchanged throughout the experiment. Where the competing ligand affinities are reported as ‘absolute’ formation constants, their conditional affinities at the experimental pH must be used in calculations [11]. |
| Is competition at thermodynamic equilibrium? |
Equilibration time for metal exchange between ligand competitors must be determined on a case-by-case basis. Where metal exchange is fast, competition experiments can be carried out as continuous titrations; but some competitions required prolonged incubation to reach equilibrium (see section ‘Kinetic controls’ and Figure 6). |
| Are stable ternary complexes present? |
The formation of stable ternary complexes prevents meaningful quantification of metal affinities via equation 7a/b. Several control experiments to confirm or disprove their presence have been described (see section ‘Ternary complexes’ and ref. [84]). |
| Are equilibrium measurements reliable? |
Ensure that there are some competitive data points where the probe responses lie within the 20–80% range of the control response in the absence of competition (see sections ‘Determination of metal–protein affinities via competition experiments’ and ‘Design and optimisation of ligand or inter-metal competition experiments’). This may be confirmed by control experiments in the absence of competition or tested by simulations of binding curves modelling responses for 10-fold tighter and weaker protein affinities (see examples in Figure 2). |
