Unevolved De Novo Proteins Have Innate Tendencies to Bind Transition Metals

Life as we know it would not exist without the ability of protein sequences to bind metal ions. Transition metals, in particular, play essential roles in a wide range of structural and catalytic functions. The ubiquitous occurrence of metalloproteins in all organisms leads one to ask whether metal binding is an evolved trait that occurred only rarely in ancestral sequences, or alternatively, whether it is an innate property of amino acid sequences, occurring frequently in unevolved sequence space. To address this question, we studied 52 proteins from a combinatorial library of novel sequences designed to fold into 4-helix bundles. Although these sequences were neither designed nor evolved to bind metals, the majority of them have innate tendencies to bind the transition metals copper, cobalt, and zinc with high nanomolar to low-micromolar affinity.

. Amino acid Sequences of 52 well expressed Naïve Metal Binders (NMB) proteins. The controls S-842, S-824-HC, and S-824-HZ are also included. All proteins are 102 amino acids long. The four designed α-helices have combinatorial sequences of polar and nonpolar residues, while regions at the ends of the helices are constant. Sequences are presented in the following format: For S824, M is the first residue, G is the 51 st , G is the 52 nd and R is the 102 nd . Conserved regions are underlined, deliberate mutations are colored red. The combinatorial assembly of the library is imperfect so some of the tested proteins are slightly different than designed -these errors are indicated in blue.  -* 20 nM is near the Limit of Detection for ITC. This is the best fit of a two-site model, but should be viewed as an estimate rather than a precise determination.

Apparent Dissociation Constant:
The dissociation constant ( ) for one site binding one metal describes the following system at equilibrium With these, the of a single site is defined as the ratio of dissociation rate to association rate Which can be written in terms of bound receptor ([ ] ) and total receptor ([ ] ) Which can be rewritten as Equation (5) is what is used to fit the curve for a single binding site. For multiple independent binding sites with somewhat overlapping affinity, as we find in our de novo proteins, the observed binding is a linear combination of the contributing events. We denote the apparent affinity across n similar binding sites , , and define it as the average of all component dissociation constants , .
Because all receptors are on the same protein of concentration [ ] and total metal is in equilibrium across the system, the only additional variable is the metal bound to each receptor [ ] , . This can be rewritten to mirror equation (4 [ ] , is the bound concentration measured in the experiment, and is the total bound across all sites. Thus, the left-hand side of equation (8) is the "Bound Equivalents" y-axis in all equilibrium dialysis plots. Equation (8) is what is used to fit the , curve for all de novo proteins. Figure S1: Binding curves for all proteins characterized by equilibrium dialysis with Co 2+ , Cu 2+ , and Zn 2+ .  Bound Equivalents (mol / mol)

HisZero -Cu
Above 5µM free zinc, the nonspecific binding is apparent: Accounting for nonspecific binding gives the following curve: HisZero -Zn 2+ -nonspecific adjusted Figure S2: Binding curves for all proteins characterized by ITC with Co 2+ , Cu 2+ , and Zn 2+ . Raw data is presented at the top while the fit curve is below. Tabluated below each figure is the protein and metal concentrations used, and the determined affinity and thermodynamic terms, or the raw enthalpy of dilution for nonbinding events. Noise in S-824 binding copper is due to metal-mediated precipitation.

Protein: 20µM S-824
Metal: 500µM Co(II) N 1 : 0.94 ± 0.03 K D,1 : 20 ± 7 nM * ∆H 1 : -7.36 ± 0.3 kJ/mol -T∆S 1 : -36.9 kJ/mol ∆G 1 : -44.3 kJ/mol N 2 : 2.11 ± 0.06 K D,2 : 0.278 ± 0.01 µM ∆H 2 : -2.25 ± 0.27 kJ/mol -T∆S 2 : -35.2 kJ/mol ∆G 2 : -37.3 kJ/mol Amino Acid Abundance-Affinity Correlation: Figure S3: Lack of correlation between the abundance of metal binding residues (His, Asp, or Glu) and the observed binding to cobalt immobilized on a bead. The percentage of metal bound (i.e. remaining on the bead after stringent washes) is on the x axis, and the number of potential metal-binding residues is on the y axis.  Figure S4: Lack of correlation between the abundance of metal binding residues (His, Asp, or Glu) and the observed binding to copper immobilized on a bead. The percentage of metal bound (i.e. remaining on the bead after stringent washes) is on the x axis, and the number of potential metal-binding residues is on the y axis. Bound Fraction Bound Fraction Figure S5: Lack of correlation between the abundance of metal binding residues (His, Asp, or Glu) and the observed binding to zinc immobilized on a bead. The percentage of metal bound (i.e. remaining on the bead after stringent washes) is on the x axis, and the number of potential metal-binding residues is on the y axis.