Date of Thesis


Thesis Type

Masters Thesis (Bucknell Access Only)

Degree Type

Master of Science



First Advisor

William D. Kerber


Metal ions bound within proteins, also called metallocofactors, are commonly found in nature and metalloproteins must be able to bind the correct metal in order to accomplish their intended biological function. In some cases, active site coordination chemistry is responsible for binding the correct metal from the mixture that is available in a cell. Heterobimetallic proteins contain two different metal ions in close proximity that work together in the protein active site. One specific example is the bimetallic active site of subclass Ic RNR R2 (R2c) proteins which contains both a manganese and an iron ion in nearly identical binding sites. The way that nature assembles the specific cofactor is intriguing. Loading of Fe and Mn into RNR R2c proteins is selective as manganese is observed in site 1 and iron is observed in site 2. Manganese and iron are similar metals in that they are neighboring elements in the first row of transition metals; however, there is a slight preference for iron based on the Irving-Williams series for divalent transition metals. The selectivity for manganese in site 1 is thought to be the result of a Jahn-Teller distortion where site 1 is a regular octahedron and site 2 is a distorted octahedron.

The goal of this research was to investigate small molecule complexes that modeled the active site of R2c proteins. To begin, a symmetric model with two identical binding sites, 5-fluoro-2-hydroxy-1,3-xylene-α,α'-diamine N,N,N',N'-tetraacetic acid (F-HXTA), was synthesized and its ability to discriminate between Fe(II) and Mn(II) was investigated. The metal complexes in equilibrium were observed using 19F-NMR spectroscopy. The metal exchange is slow on the NMR timescale, but equilibrium mixtures are still possible. The symmetric ligand showed a slight preference for Fe(II) over Mn(II). Overall, the heterobimetallic cofactor is more stable than expected, compared to the homobimetallic Fe/Fe or Mn/Mn cofactors as the first displacement of Mn with Fe was favored over the second displacement.

Both binding sites in F-HXTA are distorted, which helps to reinforce the Irving-Williams preference for Fe(II) over Mn(II). To further investigate the specific binding of these two metal ions in the native metalloenzyme, a nonsymmetric ligand model, 4-fluoro-2-hydroxy-1,3-xylene-α,α'-diamine-N,N-diacetic-N',N'-dipropionic acid (F-HXAP), was designed. The nonsymmetric F-HXAP ligand proposed was a variation on the symmetric F-HXTA ligand that was modified to contain two different size chelating arms on either side of the phenol. The different chelating arms change the geometry at the metal center and should make one binding site closer to a regular octahedron allowing Mn(II) to compete with Fe(II) more effectively. The synthesis of the nonsymmetric aminomethylated phenol was attempted via two routes: Sequential Mannich alkylations of the phenol with different secondary amines (chelating arms) and alkylation of 2-hydroxybenzyl chlorides followed by Mannich alkylation. The sequential Mannich alkylations strategy was unsuccessful on a meta-substituted phenol while the benzyl chloride alkylation followed by Mannich alkylation was also unsuccessful; however, a protection of the 2-hydroxybenzyl alcohol was developed in addition to the reversal of the order of the second route (Mannich alkylation prior to benzyl chloride alkylation) to afford a ligand with two different chelating arms. Crude material of the desired nonsymmetric ligand has been prepared, but the compound has not yet been purified.