The investigation examined all six known three-dimensional structures of cerium-protein adducts resolved by X-ray crystallography and nuclear magnetic resonance spectroscopy, assessing coordination numbers, binding motifs, and structural consequences across diverse biological frameworks. The perspective reveals that trivalent cerium substitutes for divalent calcium ions, whereas tetravalent cerium replaces trivalent ferric iron, all without inducing global structural rearrangements in host protein folds. Instead of altering macromolecular conformations, cerium ions predominantly anchor to aspartate and glutamate side chains through monodentate and bidentate coordination geometries. These assemblies exhibit coordination numbers ranging from six to ten, stabilized by ionic interactions with hard Lewis bases. While these findings outline design principles for stabilizing cerium complexes, the author emphasizes that the structural database remains exceedingly narrow, underscoring an urgent need for expanded structural characterization to advance the rational design of therapeutic cerium biomaterials.
Characterized by an electronic configuration of 4f1 5d1 6s2 and a melting point of about 795 degrees Celsius, cerium displays distinctive chemical and physiological behaviors stemming from its accessible oxidation states. The element exists predominantly as the trivalent cation Ce3+, retaining a xenon core with a single 4f electron, and as the tetravalent cation Ce4+, generated through loss of that lone electron. This redox versatility provides cerium with catalytic activity resembling superoxide dismutases and catalases, decomposing reactive oxygen species. Unlike native metal ions in the human body, Ce(III) modulates metabolic processes through catalytic properties and strong reducing power. Understanding how cerium cations interact with biological ligands is crucial for harnessing their catalytic and physiological activities while avoiding unintended damage.
The biological behavior of trivalent cerium is shaped by its steric resemblance to divalent calcium, as Ce3+ in coordination number six exhibits an ionic radius of 1.01 angstroms, nearly identical to the 1.00 angstrom ionic radius of Ca2+. Because of this dimensional parity, trivalent cerium replaces calcium ions in calcium-binding proteins, interfering with calcium-dependent reactions in blood. Due to its ability to activate prothrombin and factor XIII, antithrombotic drugs based on Ce3+ complexes have been developed. Research demonstrates that Ce3+ binds calcium- and magnesium-dependent ATPase, low-voltage-activated T-type calcium channels, and A-type potassium channels. Beyond pharmacology, binding of cerium ions to amino acids has been tested in aqueous systems to improve corrosion inhibition for aluminum alloys.
Despite therapeutic promise, biomedical deployment of cerium molecules remains constrained by poor targeting selectivity, causing non-specific tissue distribution, accumulation, and inflammatory damage. Trivalent cerium bipyridyl and phenanthroline complexes exhibit cytotoxic activities in vitro against cell lines, and cerium(III) complexes have been tested as antimicrobials. Cerium oxide diminishes reactive oxygen species in chondrocytes and cartilage tissue, counteracting interleukin-1-beta apoptosis within cartilage matrix. Cerium oxide has also been used for drug delivery and treating inflammatory conditions like Alzheimer’s disease, rheumatoid arthritis, and osteoporosis. To overcome non-specific distribution, conjugation with peptides and proteins is utilized. For example, fish skin collagen peptide chelates Ce3+ into an adduct that improves metal stability and cellular uptake. Furthermore, cerium conjugation produces artificial metalloenzymes and nanozymes with superoxide dismutase and catalase activities, scavenging reactive oxygen species and modulating inflammation by inhibiting the nuclear factor kappa-light-chain-enhancer of activated B cells pathway.
The structural basis of cerium binding was first revealed in 1997 through the nuclear magnetic resonance structure of the dicerium(III) complex of calmodulin’s N-terminal domain, termed Ce2-TR1C, deposited under code 1AK8. Bentrop and colleagues showed that the overall fold of Ce2-TR1C is almost identical to native calcium-bound calmodulin, demonstrating that trivalent cerium binding does not alter overall protein folding. Calmodulin coordinates two cerium ions. The first Ce3+ ion is coordinated by carboxylates of Asp20 at 2.9 to 3.1 angstroms, Asp22 at 2.1 to 3.0 angstroms, Asp24 at 2.0 to 4.3 angstroms, Glu31 at 1.8 to 4.9 angstroms, and the Thr26 carbonyl oxygen at 2.2 to 2.8 angstroms, yielding coordination number five to seven. The second Ce3+ coordinates Asp56 at 2.1 to 3.0 angstroms, Asp58 at 2.0 to 3.0 angstroms, Asn60 at 2.4 to 2.9 angstroms, Asp64 at 2.7 to 5.0 angstroms, Glu67 at 2.0 to 3.0 angstroms, and Thr62 carbonyl at 2.2 to 3.0 angstroms, with coordination number five to six. While coordination resembles Ca2+, Asp64 specifically binds Ce3+ but not Ca2+, and intermetallic distance is 11.1 plus or minus 0.1 angstroms versus 11.0 to 12.9 angstroms in Ca2-calmodulin.
A second NMR solution structure, deposited under code 1N65, resolved the monocerium derivative of the P43M mutant of Calbindin D9k. Calbindin D9k is a dicalcium-binding four-helix bundle with two EF-hand motifs connected by a flexible linker, binding one metal ion per loop. Paramagnetic Ce3+ was introduced to obtain long-range pseudocontact shifts, combined with nuclear Overhauser effects from the native protein to evaluate the initial unfolding step. In this structure, Ce3+ coordination closely mimics Ca2+, maintaining coordination number six to seven without altering protein folding. The metal is coordinated by side chains of Asp54 at 2.5 to 3.5 angstroms, Asn56 at 2.5 to 3.5 angstroms, Asp58 at 2.3 to 5.5 angstroms, and Glu65 at 3.2 to 5.7 angstroms, alongside the backbone oxygen of Glu60 at 3.5 to 3.6 angstroms. Both carboxylate oxygens of Asp58 and Glu65 coordinate the metal, showing that multidentate carboxylate coordination readily satisfies cerium’s steric requirements.
Cerium also binds iron-binding sites of proteins, as demonstrated by the crystal structure of human lactoferrin resolved at 2.2 angstroms under code 1FCK. Lactoferrin resembles human transferrin, possessing an N-lobe and a C-lobe that each bind Fe3+. Following treatment of lactoferrin with cerium(III) nitrate and hydrogen peroxide, Ce3+ oxidizes to Ce4+, replacing Fe3+ within both lobes. Despite Ce4+ possessing an ionic radius of 0.87 angstroms in coordination number six compared to 0.55 or 0.64 angstroms for Fe3+, the overall protein retains the unperturbed closed conformation of iron-bound lactoferrin with only slight coordination sphere expansion.
A native catalytic function for cerium is captured in the 1.6 angstrom crystal structure of methanol dehydrogenase from Methylacidiphilum fumariolicum SolV, deposited under code 4MAE. Methanol dehydrogenase is an asymmetric homodimer requiring lanthanides as essential cofactors to oxidize methanol. In both monomers A and B, the catalytic Ce3+ ion exhibits coordination number ten. This active-site adaptation improves catalytic performance, producing affinity constants for primary alcohols an order of magnitude lower by enhancing the electrophilicity of cofactor C5 to facilitate electron removal.
The remaining structures demonstrate cerium binding in landiscernin and de novo artificial enzymes. Code 9C8Y presents the 1.8 angstrom structure of landiscernin from Methylorubrum extorquens. Apo-landiscernin forms a disulfide-stabilized three-helix bundle with two chains burying 270 square angstroms of interface area. With CeCl3, landiscernin crystallizes with a single chain in the asymmetric unit, forming a metal-centered dimeric assembly between symmetry mates where Ce3+ occupies the lanthanum site with coordination number nine. Coordination involves Glu70 at 2.6 and 2.8 angstroms, Glu73 at 2.5 angstroms, and Glu75 at 2.6 angstroms from each chain, plus a water molecule on a twofold axis. Additionally, code 9SZU reports the 2.1 angstrom structure of PLZ2.3, an artificial metalloenzyme containing a TIM-barrel domain and ferredoxin lid developed via directed evolution and artificial intelligence-guided redesign to catalyze visible-light diol carbon-carbon cleavage. Incorporating mutations W85N and V87G, PLZ2.3 coordinates Ce3+ in coordination number eight via Glu31, Glu147, Glu193, and Glu309.
Comparative analysis of these six structures reveals general chemical principles governing cerium-protein recognition while emphasizing current scientific limitations. Cerium coordination numbers in proteins are characteristically high, ranging from six to ten, consistent with small-molecule data reporting coordination numbers between six and twelve. The geometry of the coordination sphere depends on coordination number, with Asp and Glu carboxylate groups serving as predominant coordinating groups through monodentate and bidentate modes. Ionic interactions predominate, while carbonyl oxygens and water molecules complete the coordination sphere. Lactoferrin represents an exception because its binding site evolved for Fe3+, utilizing tyrosine and histidine side chains. While structural data show Ce3+ and Ce4+ replace Ca2+ and Fe3+ without altering overall protein folds, Merlino notes the dataset is exceedingly limited. Little is known regarding how cerium binding affects protein flexibility and stability. Expanding the structural database of cerium-protein complexes will be crucial for understanding molecular recognition, predicting binding sites, and advancing rational design of cerium-based therapeutics and enzymes.
Subject of Research: Structural biology of cerium-protein interactions and coordination chemistry
Article Title: Cerium binding to proteins: a structural analysis
Article References: Merlino, A. (2026). Cerium binding to proteins: a structural analysis. Frontiers in Chemical Biology, 5, Article 1981754. https://doi.org/10.3389/fchbi.2026.1981754
Image Credits: AI Generated
DOI: 10.3389/fchbi.2026.1981754
Keywords: cerium, protein structures, bioinorganic chemistry, protein data bank, calmodulin, lactoferrin, methanol dehydrogenase, coordination chemistry, binding, proteins, structural, analysis
