Introduction

    Many important biological processes involve electron transfer reactions.  In the Stankovich group. We use the electron transfer reaction to learn more about various proteins.  The electron transfer path is able to tell us about changes that occur in a protein when it binds to another protein/subunit or substrate.  We are interested in the substrate molecules and their interactions with enzyme to form products  We are interested in the synthesis of deoxyribonucleotides, which  are formed through the reduction of ribonucleotides with the help of the enzyme ribonucleotide reductase.  Ribonucleotide reductase is essential to normal cells.  If the mechanism of ribonucleotide reductase could be fully understood, the knowledge could be used to restrict DNA synthesis, and therefore restrict the growth of tumors or virus replications (Sjöberg 1983).  This is vitally important, because it could lead to information that could help find a cure for cancer.

Importance of  the Stankovich Research

    Ribonucleotide reductase plays an important role in DNA biosynthesis, which makes the enzyme a favorable "target" for antiviral and antitumor agents (Stubbe 1990).  Recently, it was found that women with breast cancer who used hormone receptors had a high concentration of ribonucleotide reductase.  Full characterization of ribonucleotide reductase could allow the cancerous enzymes to be inhibited, while allowing the "healthy" enzymes to work as normal.  This could result in the development of drugs which combat the many types of cancer that take millions of lives each year.
 

Background

    The enzyme ribonucleotide reductase (RNR) is involved in a crucial step in DNA biosynthesis, the reduction of ribonucleotides to deoxyribonucleotides.  This reaction substitutes a hydrogen for the 2' hydroxyl group of the ribose, with retention of configuration to form the deoxyribonucleotide.
    Ribonucleotide reductase is an ubiquitous protein found in all living organisms, that can be divided into four distinct classes.  Each class contains an essential organic radical, which is formed through an oxidation-reduction reaction at the metal center (Stubbe 1990).  Class I reductases, the mammalian form of RNR, are tetramers composed of two proteins: R1 and R2. The protein can be purified from heat-inducible overproducing strains of E.coli. (Sjöberg 1986).  The X-ray crystal structures of the R1 component (Uhlin 1994) and the R2 component have been resolved (Nordlund 1993).  Ribonucleotide diphosphate reductase from aerobic E. coli  is archetypal of these enzymes, which contain a diiron cluster and a tyrosyl radical (Stubbe 1989).  The sole member of Class II ribonucleotide reductase is ribonucleotide triphosphate reductase from Lactobacillus leichmanni. (Lammers 1983). This enzyme contains a dimanganese cluster with a tyrosyl radical.  Class III, again, has only one representative enzyme.  The enzyme isolated from Brevibacerium ammoniagenes, contains a cobalt center with an adenosyl radical.  A ribonucleotide enzyme which can be isolated from E. coli, grown anaerobically, represents Class IV (Ashley 1986).  It contains an iron?sulfur center with a glycyl radical.  The Stankovich group’s research focuses on the Class I reductase.
 

Mechanism of Ribonucleotide Reductase
    The proposed mechanism of substrate reduction (Figure 2) occurs at the substrate binding site of the enzyme (Lammers 1983).  A hydrogen is abstracted from the radical of the protein at the 3' carbon of the substrate.  The hydroxyl group at the 2' position gains a proton and is then taken off as water.  A 3' ketone then further reduces the substrate, to form a 3'-deoxyribonucleotide radical concomitant, while oxidizing the two cysteine thiols, which causes the formation of a disulfide bond.  From that point, the 3' carbon of the substrate gains a hydrogen again, which forms the product (deoxyribonucleotide), which then will dissociate from the active site.  Once this has occurred, the reduced active site must be regenerated, which means that the protein disulfide bond has to be reduced through the transfer of reducing equivalents through an electron transfer chain.  This will occur by NADPH donating electrons that will be transported via one of two flavoprotein reductase systems: thioredoxin/thioredoxin reductase or glutaredoxin/glutathione reductase + glutathione (Sjöberg 1983).
 

R1
    R1 catalyzes the reduction of substrate and contains the binding sites for purines, pyrimidines, and allosteric cofactors (Mao 1992a). The X-ray crystal structure reveals that R1 is composed of three domains: a helical N-terminal domain, a a/b barrel domain, and a abaab region (Uhlin 1994).  Cys225 and Cys462 lie on one side of the a/b barrel domain in the crystal structure of the reduced protein.  In the crystal structure of the oxidized protein, the thiols from the residues form a disulfide bridge, which led to the idea that the two residues supply the reducing equivalents for substrate reduction by formation of a redox pair, which was suggested beforehand via site-directed mutagenesis (Mao 1992b).  The active site is located in a buried orifice between the N-terminal and b-barrel regions. .  R1 is composed of two polypeptide chains (a2), each containing 761 residues, yielding a molecular weight of 171 kDa.  One hypothesis is that R1 is very unstable and quite flexible when not bound to R2.  This flexibility would allow it to easily bind to R2.  The R1 dimer would then become more stable and conformationally rigid when it is joined to the top of the heart-shaped R2 dimer.  Thus a compact complex would be formed, while the active site remains accessible to substrate.
 

R2
    R2 is a heart shaped dimer (Figure ???) with a molecular weight of 87 kDa (Lynch 1989).  It is composed of two identical polypeptide chains (b2), each containing 375 residues, which interact at an angle to form the dimer (Haskin 1996).  R2 contains 1.2 equivalents (per dimer) of the catalytically essential tyrosyl radical  (oY122).  It also contains 3.5 equivalents of Fe(III), assembled as µ-oxo bridged dinuclear clusters that are necessary for the generation and stability of the tyrosyl radical.  The formation of the tyrosyl radical occurs due to the reaction of the diiron cluster(II) with oxygen (Equation 1).

Fe(II)-Fe(II) + Y-OH + e- + O2 Æ Fe(III)-O-Fe(III) + Y-O* + OH-            (1)

    R2 may exist in any of three stable redox states.  The first is the catalytically active R2, containing oY122 and [Fe(III)Fe(III)].  The second state is R2met, which is composed of [Fe(III)Fe(III)] and Y122.  The third state, R2red, is fully reduced [Fe(II)Fe(II)] and Y122 (Elgren 19265).  Because of the known stochiometry amount of radical/dimer, in wild type R2, 33% of the protein can be found in the met form, while the remaining 67% is in the active form.
 The reaction of RNR is under study.  Thus far, the redox chemistry of the dinuclear iron centers has proven to be an excellent probe into the enzyme mechanism.  Reduction of  the tyrosyl radical in the catalytically active R2 results in the formation of R2met (Fe(III)Fe(III), Y122), which has an absorbance at 325 and 370 nm (from the ?-? transitions of the diiron cluster), but no sharp peak at 410 nm, (due to the tyrosyl radical).  This reduction to R2red can be accomplished via two routes: through chemical or electrochemical reduction (Lynch 1989).  The second, our method of choice, uses methyl viologen as an electron mediator .  Direct electron transfer to the protein from the electrodes is not possible since the diiron center and tyrosyl radical are buried deep within the protein structure.  Therefore, methyl viologen is reduced at the electrode surface and then transported to the protein surface where it reduces the protein.  R2met can be returned to the native protein by full reduction followed by the addition of oxygen.
 
    Coulometric and chemical titrations of R2ox with methyl viologen have shown that during the addition of 2.5 reducing equivalents to the oxidized form of R2, there is no observed build-up of reduced methylviologen (MVred), indicating fast transfer of electrons from reduced methyl viologen to the protein electron acceptor.  At this point in the reaction, the tyrosyl radical is fully reduced and 33% of the clusters are reduced (Fe(II)Fe(II)), while 66% remain Fe(III)Fe(III).  Further addition of reducing equivalents results in the formation of MVred which is then slowly oxidized by the remaining diferric clusters in R2.  This suggests that the rate of electron transfer from MVred to R2 has slowed significantly.  Therefore, researchers have concluded that two different populations of iron exist in R2, the fast reducing diiron cluster ("F") and the slow reducing diiron cluster ("S").  It was also noted that exposing the reduced F cluster to oxygen did not generate a radical.  Moreover, upon the addition of oxygen, the amount of radical produced directly correlated to the number of reduced S clusters.  Thus, there appears to be two different roles for the iron centers: generation of the tyrosyl radical upon exposure to oxygen and donation of the extra electron (Miller 1999).
 
    It has been reported that there is a ratio of 3.3 iron clusters per 1 tyrosyl radical.  This negates the idea that there should be one tyrosyl radical per 2 iron clusters (i.e. two tyrosyl radicals per 4 iron clusters).  If this is the case, where is the extra electron coming from?   If the tyrosyl radical can not donate the electron, what will?  We propose that one of the dinuclear iron centers reacts with oxygen to form the tyrosyl radical and the other acts as the electron donor.
 

Future Research of the Stankovich Group
    Previous work in the Stankovich lab will provide important building blocks for much of the proposed work.  It has been observed that when the R1-R2 complex is formed, the redox potential is  -226 mV, which is a considerable negative shift of over 100 mV potential of the unbound R2.  This indicates that R1 plays a role in electron transfer occurring in R2.  This project is designed to probe the reactivity properties of R1 bound to R2.

F and S Cluster of R2
    Research will first focus on studying the electrochemical potential difference between the S and F cluster.  The redox chemistry of the dinuclear iron cluster will be continued to be used as a probe of the enzyme mechanism.  First, it must be determined if there is a potential difference between the S and F cluster.  The S cluster will be investigated by measuring its potential as the protein is quickly reduced, adding 2.5 reducing equivalents (the number of reducing equivalents associated with the fast phase) to the protein.  Several redox potential measurements will be taken during the fast reduction.  The slow phase will be probed by fully reducing R2 and allowing it to reoxidize.  As it oxidizes, the potential will be measured until it reaches the fast phase (measuring from 4.7 reducing equivalents to 2.5 equivalents).  Additionally, the effect of R1 binding on the potentials of the F and S clusters, as well as the kinetics of their formation will be studied, using stop flow measurements.

R1 bound to R2
    Another project will look at ribonucleotide reductase (R1 bound to R2).  Ribonucleotide reductase is quite puzzling, due to the necessity of both R1 and R2 to form the catalytically active protein.   The tyrosyl radical abstracts a hydrogen, then reduction takes place by the thiols.  The tyrosyl radical is located in R2, while the thiols and substrate binding sites are found in R1.  Many questions can be posed.  How do R1 and R2 interact to form the catalytically active protein and the active site?  Does a conformational change have to occur for binding to take place?  Answers to these questions can be found using electrochemistry.
 
    Our lab has looked at R2 alone to observe the kinetics of reduction and redox potential involved.  Fortunately, R2ox is spectrally active, with the tyrosyl radical having an absorbance at 410 nm and the diiron center having absorbances at both 325 and 370 nm.  It is interesting to note that the redox potential of the diiron center shifts to -226 mV when bound to R1.  Therefore, there must be some sort of conformational change occurring near the diiron center when R2 is bound to R1.  Surprisingly, the spectral properties (EPR and Mössbauer) of the diiron center did not change when R1 was bound to R2.  This indicates that electrochemical methods are the best way to understand what is happening when R1 binds to R2.  Therefore, experiments should be performed on R1 bound to R2 to observe the interactions and the conformational change involved.
    What causes the redox potential to change when R1 is bound to R2?  Our hypothesis is that rearrangement by the hydrogen bonds along the electron pathway, upon binding of R1, could possibly cause the redox potential of the iron centers of R2 to change.  This hypothesis will be explored by mutating residues along the electron transfer pathway (Tyr730, Tyr731, Tyr356, Trp48, Asp237, and His118). Residues  along the R1/R2 interface (Glu350, Tyr356, Ser343, and Val353) will also be examined via site-directed mutagenesis.  All mutants have been made by Britt-Marie Sjöberg’s group, and she has generously given us the clones.  The redox properties of the mutants will be characterized, and the effect of R1 bound to R2 on the redox potential will be explored.
 

Site-directed mutagenesis has been performed on R2 alone to try to elucidate the electron pathway in R2.
Y356
     Climent changed the R2 tyrosine 356 to an alanine (Climent 1992). Mutating with alanine basically deletes the ligand (changing from a hydroxyl group to a methyl group).  This mutation results in the mutant being inactive due to the electron transfer path being blocked and the electron not being able to move along its usual path, but still forms the normal amount of the catalytically essential tyrosyl radical.
Trp48
    Tryptophan 48 has also been converted to tyrosine, phenyl alanine, and glutamine.  Only the W48Y formed the catalytically essential tyrosyl radical, but yet again, the enzyme was inactive (Parkin 1998).  The hydroxyl group was able to donate an electron to make the catalytically essential radical.
Asp237
    Aspartic acid 237 was changed by an alanine (deleting the ligand to the iron), asparagine (adding an amine group), and glutamic acid (adding a methyl in the R group).  With the glutamic acid mutant, the tyrosyl radical is formed and low enzymatic activity is seen (~10% of wild type) (Ekberg 1997). This change could be due to the aspartate only being one methyl group shorter than the glutamate.  It has been reported that the longer glutamate could possibly still form the necessary hydrogen bonds and still allow electron transfer (Persson 1997).
 Equation 7 enables us to relate the shift in redox potential of our redox active species (R2) to the binding constants of R1 ligands with the oxidized and reduced protein.

Substrate and Activation Binding
    Another question to be answered involves substrate binding and activation binding.  How does substrate and allosteric effector binding affect the interactions between R1 and R2.  Binding will increase the rate of reaction between R1 and R2, but will it also affect the redox potential?  Measuring the effect of substrate and allosteric effector binding on the potential of the R1-R2 complex will enable this question to be answered.  The allosteric effectors that will be investigated are ATP and dATP.
 

Back to:  Stankovich Group Home Page

works cited