The origins of binding specificity of a lanthanide ion binding peptide

Lanthanides (Ln) are a series of 15 elements that exist as trivalent cations (Ln³⁺) in aqueous solution and differ in ionic radius by less than 0.26 Å. Their unique magnetic and optical properties underpin applications in materials such as batteries, phosphors, and magnets, motivating efficient recovery and separation methods. Conventional chelators (e.g., EDTA, NTA) generally show gradually increasing affinity as Ln³⁺ ionic radius decreases, reflecting increasing acidity, but lack selectivity for individual lanthanides. A 17-residue lanthanide binding tag (LBT), engineered from calmodulin, exhibits a distinct, biphasic affinity trend across the Ln³⁺ series, with pronounced specificity (e.g., >60-fold difference for certain pairs in prior variants). The research question addressed here is: what molecular mechanism underlies the unusual, biphasic binding specificity of the LBT variant LBT3 toward different Ln³⁺ ions? The study aims to integrate thermodynamics, solution structures, and molecular dynamics to reveal how ion size, hydration, and peptide coordination govern specificity.

Prior work shows small chelators (EDTA, NTA) bind Ln³⁺ with affinities that increase monotonically as ionic radius decreases, due to increasing acidity, but they do not discriminate well among individual lanthanides. LBT peptides, developed through combinatorial screening from calmodulin motifs, provide high-affinity binding and have been widely used as protein tags for NMR and biophysical studies. Notably, the LBT variant LBTv displays a biphasic affinity profile across the Ln series (with Eu³⁺ as a branch point), indicating a mechanism distinct from simple acidity trends. Structural studies have shown LBTs coordinate Ln³⁺ via side-chain carboxylates and carbonyl oxygens, forming compact complexes, but the origin of specificity across the series had remained unresolved.

• Isothermal titration calorimetry (ITC): Measured thermodynamics (ΔG, ΔH, ΔS) for LBT3 binding to a series of Ln³⁺ at pH 6.0 (50 mM MES, 100 mM NaCl) and 10 °C. Typical setup: 15 µM LBT3 in cell, 300 µM Ln(NO₃)₃ in syringe; 20 injections with appropriate controls for dilution.
• NMR spectroscopy: 1D ¹H and multidimensional TOCSY, NOESY, HSQC were recorded in 30 mM MES-d₁₃ (H₂O/D₂O 90:10, pH 6.0). Diamagnetic ions (La³⁺, Lu³⁺) enabled detailed comparisons; Sm³⁺ used for paramagnetic probing. CLEANEX-PM experiments assessed solvent accessibility/proton exchange for selected amide protons. Titrations and variable temperature studies characterized exchange regimes.
• Size-exclusion chromatography: Assessed oligomeric state of free LBT3 and LBT3-Ln complexes at pH 6.0; all species behaved as monomers.
• Structure calculations: CYANA 2.1 used with NOESY-derived restraints to model free LBT3 and complexes. For 6-coordinate models (6LBT3-Ln), coordination to D3 (Oγ), N5 (Oδ1), D7 (Oγ), W9 (carbonyl O), E11 (Oε1/Oε2), and E14 (Oε1/Oε2) was enforced with distance bounds (2.1–3.0 Å) based on an LBTv-Tb crystal structure. Additional 5-coordinate models (5LBT3-Ln) omitted N5-Oδ1 coordination.
• Molecular dynamics (MD): GROMACS with CHARMM36m for peptide, TIP3P water, and lanthanide parameters from Migliorati et al. Systems (~2589 waters) were neutralized (net -2 with Na⁺ counterions). Equilibration followed by 100 ns production runs at multiple temperatures (285–345 K). Analyses included distances for coordinating atoms, radial distribution functions (RDFs) for water around ions and selected protons, and monitoring coordination patterns including water infiltration. Accelerated weight histogram (AWH) free-energy calculations at 300 K (200 ns) mapped the landscape for N5-Oδ1 association/dissociation with weak backbone restraints to isolate local effects.
• Data deposition: ¹H/¹³C chemical shifts (BMRB 36356, 36357) and structures (PDB 7CCN, 7CCO).

• Thermodynamics: All bindings were endothermic. ΔG became more favorable from La³⁺ (−6.9 kcal/mol) to Tb³⁺ (−9.1 kcal/mol), remained roughly constant through Yb³⁺, then was slightly less favorable for Lu³⁺. ΔH and ΔS both decreased from La³⁺ to Sm³⁺, increased from Sm³⁺ to Yb³⁺, and decreased again for Lu³⁺, indicating complex, size-dependent contributions beyond simple acidity.
• NMR exchange and spectra: LBT3-Lu exhibited sharp peaks and slow exchange upon titration, while LBT3-La showed broadened peaks and intermediate exchange. CLEANEX-PM detected solvent exchange for N5-HN and N6-HN in La (at 10 °C and 25 °C) and Sm (at 25 °C), but not in Lu, indicating greater water accessibility for larger ions. Chemical shift changes (e.g., N5-HN ~0.5 ppm; G8-HN ~0.7 ppm between La and Lu complexes) support differing local environments.
• Solution structures: CYANA-derived 6LBT3-La and 6LBT3-Lu complexes were highly similar and compact, coordinating Ln³⁺ with six sites (D3, N5, D7, W9O, E11, E14). Backbone rmsd between La and Lu complexes was ~0.354 Å and similar to an LBTv-Tb crystal structure (rmsd ~0.412 Å).
• MD evidence for N5-Oδ1 lability: In 6LBT3-La simulations, N5-Oδ1 frequently dissociated (distance increasing >0.4 nm), while it mostly remained bound in 6LBT3-Lu (except late at 345 K). Free-energy profiles showed the associated state is 7.1 kcal/mol more stable than dissociated for Lu, but only 4.6 kcal/mol for La, indicating easier N5-Oδ1 dissociation in La complexes.
• Water coordination drives instability for larger ions: For La, direct water coordination to Ln³⁺ was frequently observed, with transitions among coordination patterns: (i) six LBT3 ligands only; (ii) six ligands plus one water; (iii) five ligands (excluding N5-Oδ1) plus one water (often H-bonded to N5-Oδ1); and (iv) five ligands plus two waters. Water infiltration next to N5 promotes and stabilizes N5 dissociation, leading to more flexible, less stable complexes. For Lu, only forms analogous to (i), (iii), and a unique (v) five-ligand state without water were seen; direct water coordination was rare and only at highest temperature.
• Carboxylate binding mode depends on ion size: D3 and D7 carboxylates tended towards bidentate chelation for La, but monodentate for Lu (with the second oxygen engaging water). Systematic simulations across Ln radii showed D3-O and D7-O distance trends consistent with shrinking ionic radii, bidentate-to-monodentate transitions, and increased lability of certain oxygens for smaller ions.
• Three binding categories across the series: Category I (La³⁺–Sm³⁺) features frequent water coordination and N5 dissociation, yielding lower affinity. Category II (Sm³⁺–Yb³⁺) shows LBT3 fully covering the ion, minimal water access, and higher, relatively constant affinity (with a modest increase Sm→Tb due to rising acidity, then compensation by steric factors Tb→Yb). Category III (Lu³⁺) is fully covered with the smallest ion; although fewer waters need to be displaced (lower ΔH), fewer waters are released (lower ΔS), resulting in slightly reduced affinity vs Yb.
• Quantitative differences: Relative affinities under current conditions match prior LBTv trends; e.g., LBT3 shows >40-fold difference between La³⁺ and Lu³⁺ here, and LBTv displayed >60-fold differences (e.g., La vs Tb, KD 3500 nM vs 57 nM) in prior work.

The study demonstrates that LBT3 specificity for different Ln³⁺ arises from how ion size governs water accessibility and coordination within the complex. Larger ions (La–Sm) leave space for water to infiltrate and directly coordinate the ion, which in turn displaces the weakest ligand (N5-Oδ1), destabilizing the complex and lowering affinity. For intermediate to smaller ions (Sm–Yb), the peptide fully encapsulates the metal, excluding water, stabilizing all coordination sites, and enhancing affinity, with modest variations arising from competing effects of increased acidity (favors tighter binding) and steric crowding (can weaken certain interactions). For the smallest ion (Lu), complete encapsulation persists, but thermodynamic signatures reflect fewer water molecules being displaced upon binding, reducing the entropic driving force and slightly decreasing affinity relative to Yb. These mechanistic insights reconcile the biphasic affinity profile with structural and dynamic observations, highlighting solvent-mediated effects in peptide–metal recognition.

This work elucidates the molecular origin of the biphasic binding specificity of the lanthanide-binding peptide LBT3 across the Ln³⁺ series. The key determinant is water’s ability to directly coordinate the ion within the complex: permissible for larger ions, promoting N5-Oδ1 dissociation and low affinity, and excluded for smaller ions, yielding stable, high-affinity complexes, with Lu³⁺ showing an additional entropic limitation due to fewer displaced waters. The integrated ITC, NMR, structure calculation, and MD simulations provide a coherent picture of how sub-angstrom ion size differences translate into distinct thermodynamic and structural regimes. These findings inform the rational design of peptides and ligands that harness or modulate internal water coordination to achieve selective recognition among closely similar metal ions. Future work could explore sequence modifications that tune cavity size and hydrogen-bond networks, extend to other metal series, and exploit solvent coupling for enhanced selectivity in separation technologies.

• NMR structural analysis of paramagnetic Ln³⁺ complexes is limited; detailed solution structures were derived primarily for diamagnetic La³⁺ and Lu³⁺, with Sm³⁺ used sparingly due to paramagnetism.
• MD simulations, while extensive (100 ns per condition), may not capture all rare events; some N5-Oδ1 dissociations were only observed under select temperatures and timescales. Backbone restraints were used in AWH free-energy calculations to isolate local effects, which may limit conformational sampling.
• ITC conditions (pH 6.0, 10 °C) differ from some prior studies, leading to systematic differences in absolute affinities, though relative trends are consistent.
• Force-field and ion parameterization choices (e.g., TIP3P water, specific Ln³⁺ models) carry known approximations that could affect hydration and coordination energetics.