Mono-, di-, tri-, and tetra-substituted fluorotyrosines: New probes for enzymes that use tyrosyl radicals in catalysis

Article abstract of DOI:10.1021/ja055926r

A set of N-acylated, carboxyamide fluorotyrosine (F_nY) analogues [Ac-S-FY-NH₂, Ac-3,5-F₂Y-NH₂, Ac-2,3-F ₂Y-NH₂, Ac-2,3,5-F₃Y-NH₂, Ac-2,3,6-F₃Y-NH₂ and Ac-2,3,5,6-F₄Y-NH ₂] have been synthesized from their corresponding amino acids to interrogate the detailed reaction mechanism(s) accessible to F _nY?s in small molecules and in proteins. These Ac-F _nY-NH₂ derivatives span a pK_a range from 5.6 to 8.4 and a reduction potential range of 320 mV in the pH region accessible to most proteins (6-9). DFT electronic-structure calculations capture the observed trends for both the reduction potentials and pK_aS. Dipeptides of the methyl ester of 4-benzoyl-L-phenylalanyl-F_nYs at pH 4 were examined with a nanosecond laser pulse and transient absorption spectroscopy to provide absorption spectra of F_nY?s. The EPR spectrum of each F _nY? has also been determined by UV photolysis of solutions at pH 11 and 77 K. The ability to vary systematically both pK_a and radical reduction potential, together with the facility to monitor radical formation with distinct absorption and EPR features, establishes that F_nYs will be useful in the study of biological charge-transport mechanisms involving tyrosine. To demonstrate the efficacy of the fluorotyrosine method in unraveling charge transport in complex biological systems, we report the global substitution of tyrosine by 3-fluorotyrosine (3-FY) in the R2 subunit of ribonucleotide reductase (RNR) and present the EPR spectrum along with its simulation of 3-FY122?. In the companion paper, we demonstrate the utility of F_nYs in providing insight into the mechanism of tyrosine oxidation in biological systems by incorporating them site-specifically at position 356 in the R2 subunit of Escherichia coli RNR.

Full text of DOI:10.1021/ja055926r

Published on Web 01/11/2006
Mono-, Di-, Tri-, and Tetra-Substituted Fluorotyrosines: New
Probes for Enzymes That Use Tyrosyl Radicals in Catalysis†
Mohammad R. Seyedsayamdost, Steven Y. Reece, Daniel G. Nocera,* and
JoAnne Stubbe*
Contribution from the Department of Chemistry, Massachusetts Institute of Technology,
77 Massachusetts AVenue, Cambridge, Massachusetts 02139-4307
Received August 28, 2005; E-mail: nocera@mit.edu; stubbe@mit.edu
Abstract: A set of N-acylated, carboxyamide fluorotyrosine (FnY) analogues [Ac-3-FY-NH2, Ac-3,5-
F2Y-NH2, Ac-2,3-F2Y-NH2, Ac-2,3,5-F3Y-NH2, Ac-2,3,6-F3Y-NH2 and Ac-2,3,5,6-F4Y-NH2] have
been synthesized from their corresponding amino acids to interrogate the detailed reaction mechanism(s)
accessible to FnY•s in small molecules and in proteins. These Ac-FnY-NH2 derivatives span a pKa range
from 5.6 to 8.4 and a reduction potential range of 320 mV in the pH region accessible to most proteins
(6-9). DFT electronic-structure calculations capture the observed trends for both the reduction potentials
and pKas. Dipeptides of the methyl ester of 4-benzoyl-L-phenylalanyl-FnYs at pH 4 were examined with a
nanosecond laser pulse and transient absorption spectroscopy to provide absorption spectra of FnY•s.
The EPR spectrum of each FnY• has also been determined by UV photolysis of solutions at pH 11 and 77
K. The ability to vary systematically both pKa and radical reduction potential, together with the facility to
monitor radical formation with distinct absorption and EPR features, establishes that FnYs will be useful in
the study of biological charge-transport mechanisms involving tyrosine. To demonstrate the efficacy of the
fluorotyrosine method in unraveling charge transport in complex biological systems, we report the global
substitution of tyrosine by 3-fluorotyrosine (3-FY) in the R2 subunit of ribonucleotide reductase (RNR) and
present the EPR spectrum along with its simulation of 3-FY122•. In the companion paper, we demonstrate
the utility of FnYs in providing insight into the mechanism of tyrosine oxidation in biological systems by
incorporating them site-specifically at position 356 in the R2 subunit of Escherichia coli RNR.
Introduction
enzymes, however, the occurrence of isolated ET is uncommon.
Many primary metabolic steps involving charge transport rely
Electron transfer (ET) has been the focus of most studies of
charge transport in biology. Proteins modified with redox-active
donors and acceptors,1-5 protein partners,6-10 and DNA
bioconjugates11-14 reveal that electrons transfer over long
distances according to the Marcus theory of ET,15 augmented
by considerations for electron tunneling.16,17 In functioning
on amino acid radicals in which ET is coupled to proton transfer
(PT).18-21 For such charge-transport reactions, the problem is
intrinsically more complicated because both the electron and
proton tunnel and these tunneling events can be coupled to one
another.22,23 As the electron moves, the pKa of the oxidized
cofactor will change; however, to predict kinetics, knowledge
of the driving forces for the ET and PT reactions is insufficient.
The reorganization energy will be affected by the charge
redistribution resulting from electron and proton motion.24-27
In addition, the electronic coupling will change parametrically
with the proton coordinate.
† Authors’ note: FnY refers to unprotected fluorotyrosines with free N-
and C- termini while Ac-FnY-NH2 refers to N-acylated, carboxyamide-
protected fluorotyrosines.
(1) Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: Weinheim,
Germany, 2001; Vol. 3, Part 1.
(2) Gray, H. B.; Winkler, J. R. Annu. ReV. Biochem. 1996, 65, 537.
(3) Gunner, M. R.; Robertson, D. E.; Dutton, P. L. J. Phys. Chem. 1986, 90,
3783.
(4) Farver, O.; Pecht, I. Biophys. Chem. 1994, 50, 203.
(5) Namslauer, A.; Braenden, M.; Brzezinski, P. Biochemistry 2002, 41, 10369.
(6) Tollin, G. In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-
VCH: Weinheim, Germany, 2001; Vol. 4, Part 1, p 202.
(7) Davidson, V. L. Acc. Chem. Res. 2000, 33, 87.
(8) McLendon, G.; Hake, R. Chem. ReV. 1992, 92, 481.
(9) Millett, F.; Durham, B. Biochemistry 2002, 41, 11315.
(10) Nocek, J. M.; Zhou, J. S.; Forest, S. D.; Priyadarshy, S.; Beratan, D. N.;
Onuchic, J. N.; Hoffman, B. M. Chem. ReV. 1996, 96, 2459.
(11) Lewis, F. D. Electron Transfer and Charge Transport in DNA. In Electron
Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: Weinheim, Germany,
2001; Vol. 3, Part 5, p 105.
(12) Boon, E. M.; Barton, J. K. Curr. Opin. Struct. Biol. 2002, 12, 320.
(13) Giese, Bernd. Curr. Opin. Chem. Biol. 2002, 6, 612.
(14) Jortner, J.; Bixon, M.; Langenbacher, T.; Michel-Beyerle, M. E. Proc. Natl.
Acad. Sci. U.S.A. 1998, 95, 12759.
(15) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265.
(16) Hopfield, J. J. Proc. Natl. Acad. Sci. U.S.A. 1974, 71, 3640.
Of the amino acid radicals participating in biological charge-
transfer processes, tyrosyl radical (Y•) is pre-eminent.20,21,28 This
(17) Levich, V. G. AdV. Electrochem. Electrochem. Eng. 1966, 4, 249.
(18) Stubbe, J.; Nocera, D. G.; Yee, C. S.; Chang, M. C. Y. Chem. ReV. 2003,
103, 2167.
(19) Chang, C. J.; Chang, M. C. Y.; Damrauer, N. H.; Nocera, D. G. Biochim.
Biophys. Acta 2004, 1655, 13.
(20) Stubbe, J.; van der Donk, W. A. Chem. ReV. 1998, 98, 705.
(21) Stubbe, J. Chem. Comm. 2003, 20, 2511.
(22) Cukier, R. I.; Nocera, D. G. Annu. ReV. Phys. Chem. 1998, 49, 337.
(23) Hammes-Schiffer, S. Acc. Chem. Res. 2001, 34, 273.
(24) Cukier, R. I. J. Phys. Chem. 1995, 99, 16101.
(25) Cukier, R. I. J. Phys. Chem. A 1999, 103, 5989.
(26) Soudackov, A. V.; Hammes-Schiffer, S. J. Chem. Phys. 1999, 111, 4672.
(27) Soudackov, A.; Hammes-Schiffer, S. J. Chem. Phys. 2000, 113, 2385.
(28) Pesavento, R. P.; van der Donk, W. A. AdV. Protein Chem. 2001, 58, 317.
9
10.1021/ja055926r CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 1569-1579
1569

A R T I C L E S
Seyedsayamdost et al.
Chart 1
one-electron, protein-based oxidant has been identified as both
a stable cofactor in deoxynucleotide biosynthesis29 and as a
transient intermediate with directed reactivity in photosynthetic
O2 evolution,30 prostaglandin biosynthesis,31 and disproportion-
ation of hydrogen peroxide.32 Despite extensive knowledge
about the structure of the Y•s in these systems and the ever
increasing number of Y•s identified in other proteins,33-35 the
precise mechanism of their formation and reactivity at a
functional biological level has, for the most part, remained
unknown. However, at a mechanistic level, it is well established
that oxidation of tyrosine under physiological conditions requires
the loss of both a proton and an electron, implicating proton-
coupled electron transfer (PCET) in the radical-forming mech-
anism.18,30,36 Therefore, the study of Y•s in the initiation and
propagation of charge in biology will necessarily demand that
new tools are needed to control both the electron and proton
for Y• formation within the enzyme. This is especially pertinent
to our studies of class I ribonucleotide reductase (RNR), where
reversible radical transfer is proposed to occur between a stable
Y122 radical (Y122•) on the R2 subunit and C439 on the R1
subunit.18,37 Radical migration over this unprecedented distance
has been suggested to involve a radical-hopping pathway
spanning both subunits with three other tyrosine residues, Y356
(R2), 731 (R1), and 730 (R1), leading into the R1 active site.
Moreover, the mechanism of formation of the stable Y122•,
which involves a putative di-iron, FeIII/FeIV intermediate as the
oxidant, also remains unknown.38-42
In this paper we report the synthesis and physical character-
ization of the N-acetyl and C-amide protected fluorotyrosines,
Ac-FnY-NH2s, shown in Chart 1. We also develop the
methodologies for determining the ∆G°ET and ∆G°PT for the
half of the PCET reaction involving fluorotyrosine oxidation
by measuring the reduction potentials for the radicals and the
pKa of the phenolic protons for Ac-FnY-NH2s, the latter of
which is shown to vary by up to 0.4 pKa units when compared
to FnYs. Furthermore, we report the spectroscopic properties
of the FnY•s, both EPR spectra and UV/vis absorbance spectra,
using transient laser-absorption spectroscopy. To demonstrate
the efficacy of the fluorotyrosine method in unraveling charge
transport in biological systems of complexity, we report the
global substitution of tyrosine by 3-fluorotyrosine (3-FY) in R2
and present the EPR spectrum of the 3-FY122•, along with its
simulation. In the companion paper,44 we demonstrate the utility
of FnYs in providing insight into the mechanism of tyrosine
oxidation in biological systems by incorporating them site-
specifically at position 356 in the R2 subunit of Escherichia
coli RNR and correlating RNR activity to reduction potentials
of the Ac-FnY-NH2s presented here.
Substitution of tyrosine with other redox-active amino acids
can provide insight into the mechanism of charge transport in
enzymes. However, only the natural amino acids tryptophan and
cysteine have reduction potentials close to that of tyrosine, and
they are poor structural substitutes. For this reason, we have
turned our attention to unnatural amino acids, specifically
fluorotyrosines, because they offer minimal structural perturba-
tions compared to tyrosine. They are isosteric with tyrosine,
and the van der Waals radius of fluorine is only 0.15 Å larger
than that of hydrogen, thus representing the most conservative
substitution barring isotopic replacement. Fluorination at various
points on the phenol ring affects a large range of pKa values
for the unprotected amino acid (FnYs).43 A similar variation in
the tyrosyl radical-reduction potentials upon fluorine substitution
would provide a systematic variation in ∆G°ET and ∆G°PT, the
factors that control the thermodynamics of PCET reactions.
Experimental Section
Materials and Methods. All fluorinated phenols, sodium pyruvate,
pyridoxal-5′-phosphate (PLP), N-acetyl-L-tryosinamide, 3-fluoro-L-
tyrosine, trifluoroacetic acid (TFA), thionyl chloride, acetic anhydride,
triethylamine, gaseous ammonia, 1-(3-(dimethylamino)-propyl)-3-eth-
ylcarbodiimide hydrochloride, and N-methylmorpholine (NMM) were
purchased from Sigma-Aldrich. L-Tyrosine methyl ester hydrochloride
(Y-OMe‚HCl) and 4-benzoyl-N-[(1,1-dimethylethoxy) carbonyl]-L-
phenylalanine (Boc-BPA-OH) were from Advanced ChemTech.
1-Hydroxybenzotriazole (HOBt) was obtained from NovaBiochem. All
chemicals were used as received.
Synthesis of FnYs. The plasmid, pTZTPL, encoding the enzyme
tyrosine phenol lyase (TPL), was a generous gift of Prof. Robert Phillips.
TPL purifications were carried out as described.45 FnYs were synthesized
by stirring a 1 L solution of 30 mM ammonium acetate, pH 8.0, 10
mM of the appropriate fluorophenol, 60 mM sodium pyruvate, 5 mM
â-mercaptoethanol, 40 µM PLP, and 30 units of TPL at room
temperature for 3-4 days in the dark. Synthesis of 2,3,5,6-tetrafluo-
rotyrosine required 300 units of TPL and a 3-4 week incubation period,
with an additional 20 units TPL and 1 mM tetrafluorophenol added
every week. It was important to keep the phenol concentration at 10
mM or lower, as higher concentrations resulted in enzyme denaturation.
After the incubation period, the mixture was acidified to pH ) 3.0
with concentrated HCl. The precipitated protein was removed by
filtering the mixture through a 2-3 cm thick Celite pad. The filtrate
was extracted once with 0.5 volumes of ethyl acetate. The aqueous
layer was loaded onto a 200 mL (4 cm × 17 cm) AG50W-X8 (50-
100 mesh) cation-exchange resin in the protonated state. After loading,
(29) Jordan, A.; Reichard, P. Annu. ReV. Biochem. 1998, 67, 71.
(30) Tommos, C.; Babcock, G. T. Biochim. Biophys. Acta 2000, 1458, 199.
(31) Karthein, R.; Nastainczyk, W.; Ruf, H. H. Eur. J. Biochem. 1987, 166,
173.
(32) Chouchane, S.; Girotto, S.; Yu, S.; Magliozzo, R. S. J. Biol. Chem. 2002,
277, 42633.
(33) Su, C.; Sahlin, M.; Oliw, E. H. J. Biol. Chem. 1998, 273, 20744.
(34) Ivancich, A.; Dorlet, P.; Goodin, D. B.; Un, S. J. Am. Chem. Soc. 2001,
123, 5050.
(35) Aubert, C.; Mathis, P.; Eker, A. P. M.; Brettel, K. Proc. Natl. Acad. Sci.
U.S.A. 1999, 96, 5423.
(36) Sjo¨din, M.; Styring, S.; Wolpher, H.; Xu, Y.; Sun, L.; Hammarstro¨m, L. J.
Am. Chem. Soc. 2005, 127, 3855.
(37) Stubbe, J.; Riggs-Gelasco, P. Trends Biochem. Sci. 1998, 23, 438.
(38) Bollinger, J. M., Jr.; Edmonson, D. E.; Huynh, B. H.; Filley, J.; Norton, J.
R.; Stubbe, J. Science 1991, 253, 292.
(39) Bollinger, J. M., Jr.; Tong, W. H.; Ravi, N.; Huynh, B. H.; Edmonson, D.
E.; Stubbe, J. J. Am. Chem. Soc. 1994, 116, 8015.
(40) Bollinger, J. M., Jr.; Tong, W. H.; Ravi, N.; Huynh, B. H.; Edmondson,
D. E.; Stubbe, J. J. Am. Chem. Soc. 1994, 116, 8024.
(41) Baldwin, J.; Krebs, C.; Ley, B. A.; Edmondson, D. E.; Huynh, B. H.;
Bollinger, J. M., Jr. J. Am. Chem. Soc. 2000 122, 12195.
(42) Krebs, C.; Chen, S.; Baldwin, J.; Ley, B. A.; Patel, U.; Edmondson, D. E.;
Huynh, B. H.; Bollinger, J. M., Jr. J. Am Chem. Soc. 2000 122, 12207.
(43) Kim, K.; Cole, P. A. J. Am. Chem. Soc. 1998, 120, 6851.
(44) Seyedsayamdost, M. R.; Yee, C. S.; Reece, S. Y.; Nocera D. G.; Stubbe,
J. J. Am. Chem. Soc. See accompanying paper.
(45) Chen, H.; Gollnick, P.; Phillips, R. S. Eur. J. Biochem. 1995, 229, 540.
9
1570 J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006

New Probes for Enzymes That Use Tyrosyl Radicals
A R T I C L E S
the column was washed with 10 volumes of water. The FnY analogue
was then eluted with a 10% ammonium hydroxide solution. Fractions
that gave positive ninhydrin tests (purple stain after heating) were
pooled, concentrated in vacuo, lyophilized to dryness, and stored at 4
°C. Yields ranged from 50 to 80% on the basis of the phenol as the
limiting reagent. The ninhydrin solution consisted of 0.19% w/v
ninhydrin, 95% v/v η-butanol, 0.5% v/v acetic acid, and 4.5% v/v
water.
3,5-F2Y: 1H NMR (300 MHz, D2O) δ ) 3.0 (dd, 1H, Câ-H1, 8.2
Hz, 14.7 Hz), 3.18 (dd, 1H, Câ-H2, 5.1 Hz, 14.6 Hz), 3.93 (dd, 1H,
CR-H, 5 Hz, 8 Hz), 6.88 (m, 2H, aromatic C-H). 19F NMR (300 MHz,
D2O) δ ) 26.9 (d, 2F, 8.3 Hz).
2,3-F2Y: 1H NMR (300 MHz, D2O) δ ) 2.74 (dd, 1H, Câ-H1, 7.7
Hz, 13.8 Hz), 2.92 (dd, 1H, Câ-H2, 5.8 Hz, 13.8 Hz), 3.43 (dd, 1H,
CR-H, 5.8 Hz, 7.7 Hz), 6.46 (m, 1H, aromatic C-H), 6.71 (m, 1H,
aromatic C-H). 19F NMR (300 MHz, D2O) δ ) -3.50 (dd, 1F, 8.3
Hz, 19 Hz), 16.8 (dd, 1F, 8.3 Hz, 19.8 Hz).
2,3,5-F3Y: 1H NMR (300 MHz, D2O) δ ) 2.73 (dd, 1H, Câ-H1,
7.5 Hz, 13.9 Hz), 2.89 (dd, 1H, Câ-H2, 5.9 Hz, 13.9 Hz), 3.43 (dd,
1H, CR-H, 5.8 Hz, 7.4 Hz), 6.66 (m, 1H, aromatic C-H). 19F NMR
(300 MHz, D2O) δ ) 0.28 (dd, 1F, 16.6 Hz, 19.4 Hz), 12.55 (m, 1F)
19.7 (m, 1F).
2,3,6-F3Y: 1H NMR (300 MHz, D2O) δ ) 3.08 (dd, 1H, Câ-H1,
8.2 Hz, 15 Hz), 3.24 (dd, 1H, Câ-H2, 5.7 Hz, 9.2 Hz), 3.9 (dd, 1H,
CR-H, 5.7 Hz, 8.1 Hz), 6.42 (m, 1H, aromatic C-H). 19F NMR (300
MHz, D2O) δ ) -7 (m, 1F), 19.07 (d, 1F, 22.2 Hz), 38.13 (t, 1F, 11.1
Hz).
2,3,5,6-F4Y: 1H NMR (300 MHz, D2O) δ ) 3.12 (dd, 1H, Câ-H1,
8.3 Hz, 14.9 Hz), 3.27 (dd, 1H, Câ-H2, 5.4 Hz, 14.9 Hz), 3.9 (dd, 1H,
CR-H, 5.5 Hz, 8 Hz). 19F NMR (300 MHz, D2O) δ ) -5.14 (dd, 2F,
11.1 Hz, 27.8 Hz), 11.81 (dd, 2F, 11.1 Hz, 27.7 Hz).
H). 19F NMR (300 MHz, (CD3)2SO, 25 °C) δ ) -4.8 (s, 1F), 22.3 (d,
1F, 22.2 Hz), 40.9 (m, 1F).
2,3,5,6-F4Y-OMe‚HCl: 1H NMR (300 MHz, (CD3)2SO, 25 °C) δ
) 3.16 (m, 2H, Câ-H), 3.66 (s, 3H, -OCH3), 4.14 (m, 1H, CR-H),
8.75 (b.s., 3H, -NH3+), 11.56 (b.s., 1H, PhO-H). 19F NMR (300 MHz,
(CD3)2SO, 25 °C) δ ) -0.6 (d, 2F, 16.7 Hz), 16.2, (d, 1F, 22.2 Hz),
40.9 (m, 1F).
Synthesis of N-Acetyl-Fn-tyrosinamide (Ac-FnY-NH2). In a
typical synthesis, 460 µmol of FnY-OMe in 5.2 mL of MeCN were
stirred with 4 equiv of triethylamine and 3 equiv of acetic anhydride.
The reaction was monitored by TLC (20:1 CHCl3/MeOH). After 2 h,
the solvent was removed in vacuo. The product was redissolved in
MeOH followed by the removal of MeOH in vacuo (2×). The
N-acetylated, methyl ester derivative was converted to the amide by
stirring the compound at a concentration of 100 mM in NH3(g)-saturated
MeOH. The reaction was complete after 3 days. The solvent was
removed in vacuo, and the product was purified by silica gel
chromatography with 15:1 CH2Cl2/MeOH as the eluant. The product
was characterized by 1H and 19F NMR as well as LR-ESI-MS. Yields
of 80-95% were obtained for all analogues.
1
Ac-3-FY-NH2: H NMR (300 MHz, CD3OD) δ ) (s, 3H, Ac),
(dd, 1H, Câ-H1, 8.8 Hz, 14.0 Hz), (dd, 1H, Câ-H2, 5.8 Hz, 14.0 Hz),
(dd, 1H, CR-H, 5.8 Hz, 8.8 Hz), (m, 1H, aromatic C-H). 19F NMR
(300 MHz, CD3OD) δ ) (dd, 1F, 8.3 Hz, 19.4 Hz), (dd, 1F, 8.3 Hz,
19.4 Hz). LR-ESI-MS: m/z (-H) calcd 239.0, found 238.9.
Ac-3,5-F2Y-NH2: 1H NMR (300 MHz, CD3OD) δ ) 1.92 (s, 3H,
Ac), 2.75 (dd, 1H, Câ-H1, 9.2 Hz, 13.9 Hz), 3.03 (dd, 1H, Câ-H2, 5.5
Hz, 13.9 Hz), 4.55 (dd, 1H, CR-H, 5.7 Hz, 9.2 Hz), 6.83 (m, 2H,
aromatic C-H). 19F NMR (300 MHz, CD3OD) δ ) 25.2 (d, 2F, 8.3
Hz). LR-ESI-MS: m/z (-H) calcd 257.0, found 256.9.
Ac-2,3-F2Y-NH2: 1H NMR (300 MHz, CD3OD) δ ) 1.90 (s, 3H,
Ac), 2.85 (dd, 1H, Câ-H1, 8.8 Hz, 14.0 Hz), 3.12 (dd, 1H, Câ-H2, 5.8
Hz, 14.0 Hz), 4.61 (dd, 1H, CR-H, 5.8 Hz, 8.8 Hz), 6.83 (m, 1H,
aromatic C-H). 19F NMR (300 MHz, CD3OD) δ ) -3.12 (dd, 1F,
8.3 Hz, 19.4 Hz), 17.3 (dd, 1F, 8.3 Hz, 19.4 Hz). LR-ESI-MS: m/z
(-H) calcd 257.0, found 256.9.
Synthesis of Fn-Tyrosine Methyl Ester Hydrochloride (FnY-
OMe‚HCl). In a typical synthesis, the diammonium salt of the FnY
(0.6 mmol) was combined with 20 mL of methanol in a 50-mL round-
bottom flask. Thionyl chloride (400 µL, 5.5 mmol) was added dropwise,
and the solution was stirred for 3 days at room temperature. The solvent
was removed in vacuo, and the resulting white solid was dissolved in
a minimal amount of methanol. Ether was added dropwise to precipitate
NH4Cl, which was removed by filtration. Further dropwise addition of
ether induced crystallization of the desired compound. After cooling
to -20 °C, the white crystals were isolated by filtration and dried in
vacuo, providing product in 70-88% yield.
1
Ac-2,3,5-F3Y-NH2: H NMR (300 MHz, CD3OD) δ ) 1.93 (s,
3H, Ac), 2.78 (dd, 1H, Câ-H1, 8.4 Hz, 14.1 Hz), 3.03 (dd, 1H,
Câ-H2, 6 Hz, 14.3 Hz), 4.53 (dd, 1H, CR-H, 5.7 Hz, 8.2 Hz), 6.56
(m, 1H, aromatic C-H). 19F NMR (300 MHz, CD3OD) δ ) -0.95 (t,
1F, 19.4 Hz), 10.48 (m, 1F), 19.05 (dd, 1F, 11.1 Hz, 27.7 Hz). LR-
ESI-MS: m/z (-H) calcd 275, found 274.9.
Ac-2,3,6-F3Y-NH2: 1H -NMR (300 MHz, CD3OD) δ ) 1.94 (s,
3H, Ac), 2.93 (dd, 1H, Câ-H1, 8.8 Hz, 14 Hz), 3.17 (dd, 1H, Câ-H2,
5.5 Hz, 13.8 Hz), 4.65 (dd, 1H, CR-H, 5.7 Hz, 13.8 Hz), 6.51 (m, 1H,
aromatic C-H). 19F NMR (300 MHz, CD3OD) δ ) -10.4 (m, 1F)
17.4 (d, 1F, 21 Hz) 36.2 (t, 1F, 12 Hz). LR-ESI-MS: m/z (-H) calcd
275, found 274.9.
Ac-2,3,5,6-F4Y-NH2: 1H NMR (300 MHz, CD3OD) δ ) 1.92 (s,
3H, Ac), 2.96 (dd, 1H, Câ-H1, 8.4 Hz, 14.4 Hz), 3.2 (dd, 1H, Câ-H2,
6 Hz, 14.4 Hz), 4.64 (dd, 1H, CR-H, 6 Hz, 8.4 Hz). LR-ESI-MS: calcd
293, found 293.19F NMR (300 MHz, CD3OD) δ ) -5.0 (dd, 2F, 6
Hz, 22.5 Hz), 12.6 (dd, 2F, 6 Hz, 22.5 Hz). LR-ESI-MS: m/z (-H)
calcd 293, found 292.9.
3-FY-OMe‚HCl: 1H NMR (300 MHz, (CD3)2SO, 25 °C) δ ) 3.02
(m, 2H, Câ-H), 3.68 (s, 3H, -OCH3), 4.22 (m, 1H, CR-H), 6.80 (m,
1H, aromatic C-H), 6.90 (m, 1H, aromatic C-H), 7.02 (m, 1H,
aromatic C-H), 8.56 (b.s., 3H, -NH3+), 9.88 (b.s., 1H, PhO-H). 19
F
NMR (300 MHz, (CD3)2SO, 25 °C) δ ) 27.3 (t, 1F, 11.1 Hz).
1
3,5-F2Y-OMe‚HCl: H NMR (300 MHz, (CD3)2SO, 25 °C) δ )
3.05 (m, 2H, Câ-H), 3.70 (s, 3H, -OCH3), 4.27 (m, 1H, CR-H),
6.94 (d, 2H, aromatic C-H, 8.8 Hz), 8.63 (b.s., 3H, -NH3+), 10.15
(s, 1H, PhO-H). 19F NMR (300 MHz, (CD3)2SO, 25 °C) δ ) 32.1 (d,
2F,11.1 Hz).
1
2,3-F2Y-OMe‚HCl: H NMR (300 MHz, (CD3)2SO, 25 °C) δ )
3.09 (m, 2H, Câ-H), 3.65 (s, 3H, -OCH3), 4.14 (m, 1H, CR-H), 6.78
(m, 1H, aromatic C-H), 6.89 (m, 1H, aromatic C-H), 8.64 (b.s., 3H,
-NH3+), 10.51 (s, 1H, PhO-H). 19F NMR (300 MHz, (CD3)2SO, 25
°C) δ ) -0.5 (d, 1F, 22.2 Hz), 19.5 (d, 1F, 22.2 Hz).
2,3,5-F3Y-OMe‚HCl: 1H NMR (300 MHz, (CD3)2SO, 25 °C) δ )
3.12 (m, 2H, Câ-H), 3.69 (s, 3H, -OCH3), 4.22 (m, 1H, CR-H), 7.10
(m, 1H, aromatic C-H), 8.70 (b.s., 3H, -NH3+), 10.83 (b.s., 1H,
PhO-H). 19F NMR (300 MHz, (CD3)2SO, 25 °C) δ ) 5.5 (d, 1F, 16.7
Hz), 15.8, (s, 1F), 24.5 (s, 1F).
4-Benzoyl-L-phenylalanyl-FnY Methyl Ester Trifluoroacetic Acid
(BPA-FnY-OMe‚CF3COOH). FnY-OMe (0.432 mmol, 1.0 equiv),
Boc-BPA-OH (160 mg, 0.432 mmol, 1.0 equiv), 1-(3-(dimethyl-
amino)propyl)-3-ethylcarbodiimide hydrochloride (91 mg, 0.475 mmol,
1.1 equiv), and 1-hydroxybenzotriazole (HOBt) (64 mg, 0.475 mg, 1.1
equiv) were combined in a 50-mL round-bottom flask with 20 mL of
methylene chloride. N-methylmorpholine (NMM) (175 µL, 1.73 mmol,
4 equiv) was added, and the solution was stirred overnight at room
temperature. The solution was diluted to 75 mL with methylene chloride
and washed with 2× 30 mL of 10% citric acid solution. The organic
layer was dried over MgSO4, and the solvent was removed in vacuo.
The resulting clear oil was dissolved in a few mLs of methylene chloride
2,3,6-F3Y-OMe‚HCl: 1H NMR (300 MHz, (CD3)2SO, 25 °C) δ )
3.07 (m, 2H, Câ-H), 3.64 (s, 3H, -OCH3), 4.07 (m, 1H, CR-H), 6.73
(m, 1H, aromatic C-H), 8.70 (b.s., 3H, -NH3+), 11.13 (s, 1H, PhO-
9
J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006 1571

A R T I C L E S
Seyedsayamdost et al.
and loaded onto a 1-mm thick silica gel Chromatotron plate. The product
was eluted with 3% methanol/methylene chloride and the solvent was
removed in vacuo. The resulting clear oil was dissolved in 1:1 TFA/
CH2Cl2 and stirred for 20 min at room temperature. The solvents were
then evaporated under a stream of N2(g), and the resulting clear oil was
dried under high vacuum. Trituration with CH2Cl2, followed by the
dropwise addition of ether, resulted in the formation of white crystals,
which were cooled to -20 °C and isolated by filtration (45-65% yield).
BPA-Y-OMe‚CF3COOH: 1H NMR (300 MHz, (CD3)2CO, 25
°C) δ ) 2.88-3.18 (m, 2H, Câ-H), 3.37 (m, 2H, Câ-H), 3.68 (s, 3H,
-OCH3), 4.62 (m, 1H, CR-H), 4.88 (m, 1H, CR-H), 6.78 (m, 2H,
phenol-H), 7.15 (m, 2H, phenol-H), 7.30 (d, 2H, C6H4OC6H5, 7.7 Hz),
7.55 (m, 2H, C6H4OC6H5), 7.67 (m, 3H, C6H4OC6H5), 7.76 (m, 2H,
C6H4OC6H5).
for the deprotonated phenolate form, was plotted vs pH and the data
fit to:
10(pH - pK )
a
A(pH) - A(pH ) 2)
A (pH ) 13) - A(pH ) 2)
) fY
)
(1)
-
1 + 10(pH - pK )
a
where A is absorbance at λmax(Ac-Y--NH2) and fY represents the
fraction of the amino acid in the phenolate form.
-
To determine the pKa of 2,3-F2Y by 19F NMR, the samples were
prepared in a similar fashion as described above for the pKa determina-
tions by UV spectroscopy. After preparation of the sample in an
Eppendorf tube, the pH was measured again, and the sample was
transferred to an NMR tube. Spectra were externally referenced to CFCl3
and 19F NMR recorded at different pH values. The change in the
chemical shift of each fluorine nucleus was plotted against pH and fit
to:
1
BPA-3-FY-OMe‚CF3COOH: H NMR (300 MHz, (CD3)2CO,
25 °C) δ ) 2.90-3.20 (m, 2H, Câ-H), 3.32-3.48 (m, 2H, Câ-H),
3.69 (s, 3H, -OCH3), 4.63 (m, 1H, CR-H), 4.96 (m, 1H, CR-H),
6.85-7.04 (m, 2H, phenol-H), 7.14 (m, 1H, phenol-H), 7.32 (m, 2H,
C6H4OC6H5), 7.50-7.82 (m, 7H, C6H4OC6H5)19F NMR (300 MHz,
(CD3)2CO, 25 °C) δ ) 27.17 (m, 1F), 89.76 (s, 3F).
10(pH - pK )
a
δ(pH) - δ(pH ) 5.3)
δ (pH ) 9.4) - δ(pH ) 5.3)
) fY
)
(2)
-
1 + 10(pH - pK )
a
BPA-3,5-F2Y-OMe‚CF3COOH: 1H NMR (300 MHz, (CD3)2CO,
25 °C) δ ) 2.89-3.25 (m, 2H, Câ-H), 3.38 (m, 2H, Câ-H), 3.70
(s, 3H, -OCH3), 4.65 (m, 1H, CR-H), 4.95 (m, 1H, CR-H), 7.03 (m,
2H, phenol-H), 7.33 (d, 2H, C6H4OC6H5, 8.3 Hz), 7.56 (m, 2H,
C6H4OC6H5), 7.62-7.85 (m, 5H, C6H4OC6H5). 19F NMR (300 MHz,
(CD3)2CO, 25 °C) δ ) 26.6 (s, 2F), 85.6 (s, 3F).
Differential Pulse Voltammetry (DPV). DPV was performed on
the Ac-FnY-NH2s using a CV-50W instrument from Bioanalytical
Systems as previously described.46 Briefly, measurements were per-
formed in a three-electrode glass cell with a glassy carbon working
electrode, a platinum wire counter-electrode, and a Ag/AgCl/3M NaCl
reference electrode. The glassy carbon electrode was polished with a
water/alumina slurry and sonicated for ∼1 min prior to each measure-
ment. Typically, the solution contained 150-200 µM of Ac-FnY-
NH2 in 10 mM potassium phosphate with 200 mM KCl as the
supporting electrolyte. The pH of each solution was adjusted with 1 M
HCl or 1 M KOH prior to addition of the analyte and spanned a range
of 1-13 at 0.3-0.5 pH unit intervals. After addition of the analyte to
the solution and immediately before each measurement, the pH was
measured with a pH meter, and the solution was degassed by Ar
sparging for 2 min. Measurements were made with the ferricyanide/
ferrocyanide couple as an internal standard (0.408 V47 vs NHE, pH 7).
DPV parameters were as follows: scan rate, 10 mV/s; sample width,
17 ms; pulse amplitude, 50 mV; pulse width, 50 ms; pulse period, 200
ms; quiet time, 4 s; sensitivity, 100 µA/V; temperature, 25 °C.
Generation of Oxidized FnYs and EPR Measurements. A
10-15 mM solution of each FnY was adjusted to pH 11 with NaOH.
The sample was transferred to an EPR tube, and the solution bubbled
with Ar for 15 min. The tube was then sealed under Ar pressure and
quick-frozen in liquid N2. The tube was immersed in liquid N2 in a
quartz finger dewar and irradiated for 5 min using unfiltered light from
a 1000 W high pressure Xe arc lamp. The photolyzed product was
never allowed to warm above 77 K, and the EPR spectra were sub-
sequently recorded on a Bruker ESP-300 X-band (9.4 GHz) spectrom-
eter equipped with an Oxford liquid He cryostat at 77 K. The EPR
parameters were as follows: power ) 0.2 mW; modulation frequency
) 100 kHz; modulation amplitude ) 1.5 G; time constant ) 5.12 ms;
and conversion time ) 20.48 ms. The resulting EPR spectra re-
mained unchanged after storage of the photolyzed product for one
month at 77 K in the dark. The EPR spectra of 3-FY-R2 and wild-
type (wt)-R2 (vide infra) were obtained at 25 K using the same
parameters listed above, except with a modulation amplitude ) 2 G.
The 3-FY-R2 spectrum was simulated using SIMPIP and QPOW
programs obtained from the Illinois EPR Research Center.
BPA-2,3-F2Y-OMe‚CF3COOH: 1H NMR (300 MHz, (CD3)2CO,
25 °C) δ ) 2.93-3.49 (m, 4H, Câ-H), 3.69 (s, 3H, -OCH3), 4.68
(m, 1H, CR-H), 4.95 (m, 1H, CR-H), 6.80 (m, 1H, phenol-H), 7.01
(m, 1H, phenol-H), 7.39 (d, 2H, C6H4OC6H5, 7.7 Hz), 7.56 (m, 2H,
C6H4OC6H5), 7.63-7.86 (m, 5H, C6H4OC6H5).19F NMR (300 MHz,
(CD3)2CO, 25 °C) δ ) -1.7 (m, 1F), 18.3 (m, 1F), 85.6 (s, 3F).
1
BPA-2,3,5-F3Y-OMe‚CF3COOH: H NMR (300 MHz, (CD3)2-
CO, 25 °C) δ ) 2.93-3.51 (m, 4H, Câ-H), 3.70 (s, 3H, -OCH3),
4.68 (m, 1H, CR-H, 5.04 (m, 1H, CR-H), 7.11 (m, 1H, phenol-H),
7.41 (d, 2H, C6H4OC6H5, 7.7 Hz), 7.55 (m, 2H, C6H4OC6H5), 7.62-
7.85 (m, 5H, C6H4OC6H5).19F NMR (300 MHz, (CD3)2CO, 25 °C) δ
) 7.93 (d, 1F, 22.2 Hz), 18.56 (m, 1F), 26.1 (s, 1F), 89.9 (s, 3F).
1
BPA-2,3,6-F3Y-OMe‚CF3COOH: H NMR (300 MHz, (CD3)2-
CO, 25 °C) δ ) 3.00-3.26 (m, 2H, Câ-H), 3.44 (m, 2H, Câ-H),
3.67 (s, 3H, -OCH3), 4.66 (m, 1H, CR-H), 5.12 (m, 1H, CR-H), 6.76
(m, 1H, phenol-H), 7.41-7.85 (m, 9H, C6H4OC6H5).19F NMR (300
MHz, (CD3)2CO, 25 °C) δ ) -1.86 (m, 1F), 25.2 (d, 1F, 22.2 Hz),
44.1 (m, 1F), 90.0 (s, 3F).
BPA-2,3,5,6-F4Y-OMe‚CF3COOH: 1H NMR (300 MHz,
(CD3)2CO, 25 °C) δ ) 3.10-3.52 (m, 4H, Câ-H), 3.68 (s, 3H,
-OCH3), 4.65 (m, 1H, CR-H), 5.05 (m, 1H, CR-H), 7.46-7.85 (m,
9H, C6H4OC6H5). 19F NMR (300 MHz, (CD3)2CO, 25 °C) δ ) -2.42
(d, 2F, 22.2 Hz), 14.6 (d, 2F, 22.2 Hz), 85.6 (s, 3F).
1
Physical Measurements. H and 19F NMR spectra were recorded
on a Varian 300 MHz NMR at the MIT Department of Chemistry In-
strumentation Facility (DCIF). Aqueous samples for 1H NMR and 19
F
NMR were carried out in D2O with 3-(trimethylsilyl)-propionic acid-
d4 (TSP) as an internal standard and CFCl3 as an external standard,
respectively. Nonaqueous 1H and 19F NMR were externally referenced
to tetramethylsilane and CFCl3, respectively. Steady-state absorption
spectra were recorded on an Agilent 8453 Diode Array spectropho-
tometer.
Transient Absorption (TA) Spectroscopy. Pump light for TA
measurements was provided by an Infinity Nd:YAG laser (Coherent)
running at 20 Hz. The setup of the instrument was slightly modified
from that previously described.48 The third harmonic (355 nm, 200 mJ
per pulse) was used to pump a Type II XPO (Coherent). The 600-nm
pKa Determination of Ac-FnY-NH2 Analogues using UV
Spectroscopy and 19F NMR Spectroscopy. The pH of solutions
containing 10 mM potassium phosphate and 200 mM KCl were adjusted
using either KOH or HCl spanning a range of 2-13 at 0.3-0.5 pH
unit intervals. The Ac-FnY-NH2 was diluted into each solution, and
the pH was measured before and after acquisition of the UV spectrum.
The absorption at λmax(Y-), the wavelength of maximum absorption
(46) Tommos, C.; Skalicky, J. J.; Pilloud, D. L.; Wand, A. J.; Dutton, P. L.
Biochemistry 1999, 38, 9495.
(47) O’Reilly, J. E. Biochim. Biophys. Acta 1973, 292, 509.
9
1572 J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006

New Probes for Enzymes That Use Tyrosyl Radicals
A R T I C L E S
signal beam was isolated, and the frequency was doubled in a XPO-
UV frequency-doubling crystal (Coherent) to produce a slightly
divergent 300-nm beam with a ∼5-ns pulse width and a pulse energy
of ∼500 µJ. This beam was then passed through a slow-focusing lens
and through the sample at ∼160°, producing a beam spot of 2-mm
diameter on the sample. A 75 W Xe arc lamp (unpulsed, PTI) provided
the probe light, which was focused onto the sample, overlapped with
the pump beam, recollimated after the sample, and focused onto the
entrance slit of a Triax 320 spectrometer. The probe beam was dispersed
by a 300 × 500 blazed grating and collected with either an intensified
gated CCD camera (ICCD, Andor Technology, 1024 pixels × 256
pixels, 26 µm2) for TA spectra or a photomultiplier tube (PMT) for
TA kinetics at a single wavelength. PMT outputs were collected and
averaged with a 1 GHz oscilloscope (LeCroy 9384CM). To produce a
TA spectrum, a series of four spectra were taken: IF (pump on/probe
off), I (pump on/probe on), IB (pump off/probe off), and I0 (pump off/
probe on). Uniblitz shutters driven by delay generators (DG535,
Stanford Research Systems) and T132 Uniblitz shutter drivers were
used to block the pump and probe light in this sequence using a TTL
trigger from the Q-switch of the laser. Transient spectra were corrected
for fluorescence and background light using these spectra and the
equation: ∆OD ) log[(I0 - IB)/(I - IF)]. Spectra reported are the
average of 1250 of the four-spectra sequences. The instrument control
and data analysis were performed by software written in LabView.
Aqueous BPA-FnY-OMe samples were prepared at 200-500 µM
concentrations in deionized water buffered to pH 4 with 20 mM succinic
acid. All TA measurements were performed in a 2-mm quartz optical
cell. For BPA-3-FY-OMe, 266-nm pump light was used as described
previously,49 and the sample was flowed through a 2-mm cell to ensure
fresh sample for each laser shot.
Computational Methods. Density functional theory calculations
(DFT) were performed using the Amsterdam Density Functional
(ADF2002.02) program,50,51 a home-built Linux cluster comprising 60
Intel processors organized in groups of 12 running in parallel. The
generalized gradient approximation was used as implemented in ADF
by the Becke-88 functional for exchange52 and the Perdew-Wang-
91functional for correlation.53 A basis set of triple-ú Slater-type
functions augmented by a polarization set (TZP) was used for all atoms;
spin restriction was lifted for all Ac-FnY•-NH2 radicals. Energies
reported are gas-phase internal energies at 0 K, without zero-point
vibrational energy, of the geometry optimized structure. Each structure
was optimized from the same amino acid backbone configuration, which
was kept nearly constant throughout the calculation. Orbital energies,
Cartesian coordinates, and atom-numbering schemes for each of the
protonated, deprotonated, and radical Ac-FnY-NH2s are provided in
the Supporting Information.
cells were grown to A600 nm of 1.4-1.6 and harvested ∼4 h after induc-
tion. The yield was 5 g/L and 20-25 mg/g of cells, similar to yields
obtained for wt R2 culture growths. 3-FY-R2 was purified using
published procedures.54
Extent of Incorporation of 3-FY into R2. The purified 3-FY-R2
(200-2000 pmol) was lyophilized to dryness in a Speedvac and
hydrolyzed in 200 µL of 6 N HCl (Pierce Sequence Grade) at 150 °C
for 1 h or 110 °C for 22 h under N2, with norleucine (1000 pmol)
added as an internal standard. The samples were then dried in vacuo,
treated with 2% borane/diisopropyl ethylamide complex solution (40
µL) to reduce methionine oxidation, and allowed to air-dry for 2 h.
The samples were then dissolved in 0.3 M sodium acetate buffer (pH
5.5) containing 0.025% w/v potassium EDTA, mixed rigorously, and
centrifuged at 15 000g on a tabletop centrifuge for 5 min to remove
particulate matter. Aliquots (30 µL) were then loaded onto an Applied
Biosystem model 420A Derivatizer using phenyl isothiocyanate as the
derivatizing agent, followed by automatic injection onto an Applied
Biosystem 130A Separation System, which uses a Spheri-5 PTC 5
micron, 220 mm × 2.1 mm column from Perkin-Elmer Brownlee. A
series of pure amino acid standards, including 3-FY (500 pmol), were
also derivatized and analyzed to determine the retention time of each
amino acid and for quantitation purposes. The data were collected and
analyzed using a Perkin-Elmer Applied Biosystem 610A Data Analysis
program by the MIT Biopolymers Lab.
Results and Discussion
pKa Determination of Ac-FnY-NH2s by UV Spectroscopy
and 2,3-F2Y by 19F NMR. A slightly modified method from
that of Kim and Cole43 and Phillips and co-workers,55 using
TPL-mediated enzymatic synthesis, yields the free FnYs in good
yields and high purity. The amine and acid functionalities of
the free amino acids were protected by acetylation and amidation
to avoid complications arising from different protonation states
of these groups. Figure 1 presents the absorption spectra of
phenol and phenolate forms of the Ac-FnY-NH2 amino acids
presented in Chart 1. Substitution of the aromatic protons of
tyrosine with fluorine results in a blue-shift of the λmax associated
with the π f π* transition (top panel), which red-shifts upon
deprotonation (bottom panel). This observed bathochromic shift
with deprotonation of the phenol group allows for the measure-
ment of pKa by UV spectroscopy.
The typical absorbance changes for Ac-FnY-NH2s induced
by changes in pH are shown in Figure 2 for Ac-2,3-F2Y-
-
NH2. The pKa values were determined by plotting fY as a func-
Global Incorporation of 3-FY into R2. Competent BL21(DE3)
cells were transformed with pTB2, an R2 overexpression plasmid, and
grown on an LB agar plate containing 100 mg/L ampicillin. A single
colony from the plate was used to seed a 200-mL culture containing
100 mg/L ampicillin. The culture was grown overnight and then centri-
fuged at 10 000g for 10 min. The cell pellet was re-suspended in 50
mL of minimal medium containing no L-tyrosine and 100 mg/L ampi-
cillin (details of the medium are reported in Supporting Information,
Table S1). The doubling time of the cells was typically ∼40 min, and
tion of pH and fitting the data to eq 1 (see inset). The pKa values
listed in Table 1 for the protected Ac-FnY-NH2 amino acids
differ from those previously determined for the unprotected FnY
analogues.43 N-acetyl and C-amide protected derivatives allow
us to monitor acid/base equilibria resulting only from the
fluorophenol moiety; therefore, the pKa values obtained on these
blocked derivatives better approximate the pKa that these
residues would have in a solvent-exposed, simple polypeptide
chain. The pKa values of FnY may be independently determined
by 19F NMR. Figure 3 shows the pH dependence of the two
fluorine signals of 2,3-F2Y; as the pH is decreased, these signals
shift downfield. When the chemical shift is plotted as a function
of pH and fit to eq 2, pKas of 7.53 (using the lower field 19F
nucleus) and 7.59 (using the higher field 19F nucleus) are
obtained, in good agreement with that previously reported for
at A600
of 0.6, 1 g/L of 3-FY was added to the culture. IPTG (0.5
nm
mM) was added to the culture 15 min after the addition of 3-FY. The
(48) Loh, Z.-H.; Miller, S. E.; Chang, C. J.; Carpenter, S. D.; Nocera, D. G. J.
Phys. Chem. A 2002, 106, 11700.
(49) Reece, S. Y.; Stubbe J.; Nocera, D. G. Biochim. Biophys. Acta 2005, 1706,
232.
(50) te Velde, G.; Bickelhaupt, F. M.; van Gisbergen, S. J. A.; Fonseca Guerra,
C.; Baerends, E. J.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22,
931.
(51) Fonseca Guerra, C.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor.
Chem. Acc. 1998, 99, 391.
(52) Becke, A. D. Phys. ReV. A 1988, 38, 3098.
(53) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M.
R.; Singh, D. J.; Fiolhais, C. Phys. Rev. B 1992, 46, 6671.
(54) Ge, J.; Yu, G.; Ator, M. A.; Stubbe, J. Biochemistry 2003, 42, 10071.
(55) Phillips, R. S.; Ravichandran, K.; Von Tersch, R. L. Enzyme Microb.
Technol. 1989, 11, 80.
9
J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006 1573

A R T I C L E S
Seyedsayamdost et al.
Table 1. Physical Data for Tyrosine and Fluorotyrosine Derivatives
a
λ
max(Ac
−
FnY
−
NH2)
λ
max(Ac
/ nm (
−
FnY-
−
NH2)
λ
max(FnY
•
)
Ep(Y
•
/Y-
)
δ(O−H)
/ ppmf
1
1
b
c
d
derivative
/ nm (
ꢀ
/M-1 cm-
)
ꢀ
/M-1 cm-
)
/ nm
pKa
pKa
pKa
/ mVe
Y
3-FY
275 (1400)
272 (1530)
263 (560)
265 (910)
267 (570)
266 (1050)
∼257 (420)
293 (2420)
289 (2740)
275 (1830)
279 (2060)
276 (1460)
277 (2360)
∼273 (1660)
407
400
395
408
400
415
411
10
-
9.9
8.4
7.2
7.8
6.4
7.0
5.6
642
705
755
810
853
911
968
9.44
9.88
8.4
6.8
7.6
6.1
6.6
5.2
8.4
7.2
7.7
6.4
6.9
5.6
3,5-F2Y
2,3-F2Y
2,3,5-F3Y
2,3,6-F3Y
F4Y
10.15
10.51
10.83
11.13
11.56
a
b
c
d
BPA-FnY-OMe dipeptides. FnY pKa as reported in ref 43. Ac-FnY-NH2 pKa as determined by the UV-vis absorption titration method.
Ac-FnY-NH2 pKa as determined by the DPV method. DPV peak potential vs NHE for the Ac-FnY•-NH2/Ac-FnY- -NH2 couple. FnY-OMe H
NMR in (CD3)2SO.
e
f
1
Figure 2. Changes in the absorption profile of Ac-2,3-F2Y-NH2 at
aqueous solution pH of 5.1 (red s), 7.0, 7.7, 8.0, 8.3, and 9.1 (black s).
(Inset) Fraction of Ac-2,3-F2Y- -NH2 vs pH (O) and fit (s) as described
by eq 1.
the pKa of each Ac-FnY-NH2, a pH-independent peak potential
is observed, indicative of the Ac-FnY•-NH2/Ac-FnY- -NH2
couple, and these values are listed in Table 1. As the pH is
decreased below the pKa, the peak potential increases by 59
mV per pH unit corresponding to the Ac-FnY•-NH2/Ac-
FnY-NH2 redox couple in which tyrosine oxidation is ac-
companied by proton loss. The pH dependence of Ep is in
accordance with the Nernst equation,
Figure 1. UV spectra of phenol (A) and phenolate (B) forms of (black s)
Ep ) Ep(Ac-Y•-NH2/Ac-Y--NH2) +
Ac-Y-NH2, (light blue s) Ac-3-FY-NH2, (dark blue s) Ac-3,5-F2Y-
NH2, (red s) Ac-2,3-F2Y-NH2, (green s) Ac-2,3,6-F3Y-NH2,
2.3 RT/nF log (1 + 10-pH/10-pK ) (3)
a
(purple s) Ac-2,3,5-F3Y-NH2, and (orange s) Ac-F4Y-NH2.
the unprotected amino acid (pKa ) 7.6).43 Within a protein
environment, the determination of pKa values of FnYs by the
19F NMR method may be easier to implement since large UV
absorptions from intraprotein Y and W residues will obscure
the bathochromic shift of FnY residues incorporated into the
protein. Indeed, 19F NMR signals of nonnatural FnY amino acids
have been used in small proteins to probe changes in the
environment of specific residues upon domain movements.56
pH-Dependent Oxidation of Ac-FnY-NH2’s. Oxidation
of Y is an irreversible one-electron event,57 and thus, only peak
potentials, Ep, may be obtained, most accurately by DPV.46
Sample DPV traces for oxidation of each Ac-FnY--NH2 are
provided in Figure S2. Plots of Ep vs pH for the Ac-FnY-
NH2 amino acids are shown in Figure 4. At pH values exceeding
where Ep(Ac-Y•-NH2/Ac-Y- -NH2) is the peak potential
of the deprotonated tyrosinate and the pKa accounts for the
Ac-Y-NH2 f Ac-Y- -NH2 + H+ reaction. Square schemes
describing the thermodynamics for oxidation58 of tyrosine show
that it is energetically disadvantageous to oxidize YH and then
deprotonate (ET/PT) and likewise to deprotonate YH followed
by oxidation (PT/ET).36 Excepting extreme thermodynamic
conditions,36 the diagonal pathway of concerted electron-proton
(CEP) is preferred for tyrosine oxidation.59 It is likely operating
(56) Bonaccio, M.; Ghaderi, N.; Borchardt, D.; Dunn, M. F. Biochemistry 2005,
44, 7656.
(57) Harriman, A. J. Phys. Chem. 1987, 91, 6102.
(58) Mayer, J. M. Annu. ReV. Phys. Chem. 2004, 55, 363.
(59) Sjo¨din, M.; Styring, S.; Åkermark, B.; Sun, L.; Hammarstro¨m, L. J. Am.
Chem. Soc. 2000, 122, 3932.
9
1574 J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006

New Probes for Enzymes That Use Tyrosyl Radicals
A R T I C L E S
Figure 3. Dependence of 19F NMR chemical shifts on pH. 19F NMR spectra
of free 2,3-F2Y at various pH values (bottom) and fraction of 2,3-F2Y- as
a function pH fit to eq 2 (top inset). Blue data points represent the signal
of the higher -field 19F nucleus; red data points represent that of the lower-
field 19F nucleus.
Figure 5. (Top) Plot of EP(Ac-FnY•-NH2/Ac-FnY- -NH2) vs
E(Ac-FnY•-NH2) - E(Ac-FnY- -NH2), the total bonding energy dif-
ference between Ac-FnY•-NH2 and Ac-FnY- -NH2 as calculated by
DFT, with linear fit (s). (Bottom) Plot of -ln(Ka) vs E(Ac-FnY- -NH2)
-
E(Ac-FnY-NH2), the total bonding energy difference between
Ac-FnY- -NH2 and Ac-FnY-NH2 as calculated by DFT, with linear fit
(s). Color coded points correspond to data for (black b) Ac-Y-NH2,
(light blue b) Ac-3-FY-NH2, (dark blue b) Ac-3,5-F2Y-NH2, (red b)
Ac-2,3-F2Y-NH2, (purple b) Ac-2,3,5-F3Y-NH2, (green b) Ac-2,3,6-
F3Y-NH2, and (orange b) Ac-F4Y-NH2.
Figure 4. Reduction potential vs NHE as a function of pH for Ac-FnY-
NH2 derivatives of Chart 1: (black s) Ac-Y-NH2, (light blue s)
Ac-3-FY-NH2, (dark blue s) Ac-3,5-F2Y-NH2, (red s) Ac-2,3-F2Y-
NH2, (purple s) Ac-2,3,5-F3Y-NH2, (green s) Ac-2,3,6-F3Y-NH2, and
(orange s) Ac-F4Y-NH2.
is captured computationally. Geometry optimizations on
Ac-FnY-NH2, Ac-FnY•-NH2, and Ac-FnY- -NH2 were
performed while keeping the amino acid backbone in the same
conformation. Figure 5 (top) compares the difference in total
bonding energy between the Ac-FnY•-NH2 and Ac-FnY- -
NH2 redox states, as calculated by DFT to the experimentally
determined Eps (Ac-FnY•-NH2/Ac-FnY- -NH2). The linear
correlation between observed and calculated results validates
the quality of the computation. The electronic-structure calcula-
tion summarized in Figure 6 shows that the observed trend in
the reduction potential arises primarily from stabilization of the
HOMO and HOMO-1 orbitals of the phenolate with increasing
fluorination of the aromatic ring of tyrosine. This stabiliza-
tion of HOMO and HOMO-1 levels is most pronounced for
Ac-FnY- -NH2; these levels exhibit only a modest shift for
Ac-FnY•-NH2 (and Ac-FnY-NH2) (see Figures S3-S5). The
Kohn-Sham representations of the frontier molecular orbitals,
here, too, for the flourotyrosines, though more detailed inves-
tigations are needed to establish unequivocally the mechanism
of oxidation. A fit of the Ep vs pH curves in Figure 4 to eq 3
yields pKas for Ac-FnY-NH2s that are in good agreement with
those obtained using the UV titration method already described
(see Table 1). The pKa of the oxidized [FnY•]+ radicals cannot
be determined from eq 3 because water is oxidized at potentials
below that needed for the electrochemical generation of [FnY•]+
at low pHs.
Electronic-Structure Calculations of Ac-FnY-NH2s. The
Ac-FnY•-NH2/Ac-FnY- -NH2 redox couple was found to
generally increase with the number of fluorine substituents on
the phenol ring. Within the series of di- and trifluorotyrosines,
the couples follow a trend: Ac-2,3-F2Y-NH2 > Ac-3,5-F2Y-
NH2 and Ac-2,3,6-F3Y-NH2 > Ac-2,3,5-F3Y-NH2, which
9
J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006 1575

A R T I C L E S
Seyedsayamdost et al.
Figure 6. Energy-level diagram for frontier molecular orbitals of the
phenolates: (black s) Ac-Y- -NH2, (light blue s) Ac-3-FY- -NH2,
(dark blue s) Ac-3,5-F2Y- -NH2, (red s) Ac-2,3-F2Y- -NH2, (purple
s) Ac-2,3,5-F3Y- -NH2, (green s) Ac-2,3,6-F3Y- -NH2, and (orange
s) Ac-F4Y- -NH2. Selected Kohn-Sham representations of orbitals for
the tyrosine analogue are shown at the 95% probability level. Kohn-Sham
representation of the HOMO orbital for Ac-3,5-F2Y- -NH2 is presented
in the frame.
shown for tyrosinate in Figure 6, reveal that the HOMO orbital
carries significant π-electron density at the 3 and 5 positions
of the aromatic ring, whereas the HOMO-1 has π-electron
density localized on the orbital of the oxygen of the phenol, in
an in-plane antibonding interaction with the aromatic ring. The
interaction of the HOMO and HOMO-1 ring orbitals with the
pπ-orbitals of the fluorine atom substituents allows subtle
variations in the reduction potential among members of the
phenolate series to be understood. For instance, substitution of
fluorine at the 2,3 vs 3,5 ring positions results in a 65 mV
increase in the reduction potential of the radical. The HOMO
of Ac-3,5-F2Y--NH2, which is presented in the framed box
of Figure 6, exhibits an antibonding interaction between the
π-orbitals of the fluorine atoms and the 3,5-carbon pπ-orbitals
of the aromatic ring. By moving a fluorine from a 5 to a 2
position, this antibonding interaction is relaxed significantly,
resulting in stabilization of the HOMO. A similar effect is
observed for 2,3,5- and 2,3,6-trifluorotyrosines.
A parallel calculation was performed to assess the difference
between the observed ln(Ka) and the total bonding energy
difference of protonated and deprotonated Ac-FnY-NH2s. The
calculation predicts the general trend that increasing the number
of fluorine substituents decreases the pKa of the phenolic proton
(Figure 5, bottom), however a poorer correlation is observed
as compared to that for the reduction potentials. The anomaly
is most obvious for Ac-2,3-F2Y-NH2 and Ac-3,5-F2Y-NH2,
which are calculated to have similar pKas despite an observed
difference of 0.6 pKa units. We believe that such deviations
arise from the deficiency of gas-phase DFT calculations in
modeling specific hydrogen-bonding interactions that are im-
portant in determining the overall pKa of the fluorotyrosines.60-62
Spectroscopic Characterization of FnY•. Distinct EPR
signals are observed for oxidized Y and FnYs. The oxidized
species were generated by the method of Hulsebosch et al.,63
using UV photolysis at 77 K. The spectra obtained are shown
Figure 7. Normalized EPR spectra of oxidized: (black s) Y, (light blue
s) 3-FY, (dark blue s) 3,5-F2Y, (red s) 2,3-F2Y, (purple s) 2,3,5-F3Y,
(green s) 2,3,6-F3Y, and (orange s) F4Y at 77 K. Oxidized product was
formed by UV photolysis of pH 11 solutions of the FnYs at 77K.
in Figure 7; the signal was stable for at least 1 month at 77 K.
All FnY•s display broader signals than Y•; lower modulation
amplitudes and larger sweep widths did not uncover any
additional hyperfine interactions. Qualitatively, the spectra of
2,3,5-F3Y•, 2,3,6-F3Y•, and 2,3,5,6-F4Y• are similar, whereas
those of 2,3-F2Y• and 3,5-F2Y• are distinctly different.
Attempts to simulate the EPR spectra displayed in Figure 7
by several methods were unsuccessful. Both standard programs
(60) Baker, A. W.; Kaeding, W. W. J. Am. Chem. Soc. 1959, 81, 5904.
(61) West, R.; Powell, D. L.; Whatley, L. S.; Lee, M. K. T.; Schleyer, P. von
R. J. Am. Chem. Soc. 1962, 84, 3221.
(63) Hulsebosch, R. J.; van den Brink, J. S.; Nieuwenhuis, S. A. M.; Gast, P.;
(62) Carosati, E.; Sciabola, S.; Cruciani, G. J. Med. Chem. 2004, 47, 5114.
Raap, J.; Lugtenburg, J.; Hoff, A. J. J. Am. Chem. Soc. 1997, 119, 8685.
9
1576 J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006

New Probes for Enzymes That Use Tyrosyl Radicals
A R T I C L E S
400 ns and 1 µs show that the overall profile shape does
not change with time. The transient signals decayed concomi-
tantly with a rate constant of 5.6 × 106 s-1; single wave-
length kinetics data recorded at 547 nm are shown in the inset
of Figure 8. This unimolecular decay is consistent with charge
recombination between BPA• and Y•. The kinetics were
unaffected by the presence or absence of oxygen in solution. A
minor residual absorbance (<10%), which persists for tens of
microseconds, is likely due to a small fraction of ET product
3
formed in the bimolecular reaction between BPA-Y-OMe
and ground-state BPA-Y-OMe. This reaction is favored at
the high concentrations of dipeptide required for the TA
experiment.
Figure 9 shows the TA spectra obtained 100 ns after 300 nm
excitation of each of the BPA-FnY-OMe dipeptides. The
amount of BPA•-FnY•-OMe produced in each experiment
varies due to changes in pump-probe beam overlap and sample
concentration. Therefore, to allow for comparison of the FnY•
absorption maxima between experiments, the spectra in Figure
9 are normalized to the 547-nm peak of the BPA• ketyl radical.
The feature is invariant for each of the dipeptides, whereas the
absorption band of each FnY• species varies in its maximum
between 395 and 415 nm. The values of λmax for each of the
FnY•s are listed in Table 1. An interesting comparison can be
made between the FnY•s and Y• absorbance features. The 2,3-
F2Y•, 2,3,6-F3Y•, and F4Y• all have the familiar double-hump
absorbance feature of Y•, though they are slightly broader and
red-shifted. The spectrum of 3-FY• also contains the double-
hump feature, though its λmax is now shifted to the blue by 7
nm. Of particular note are the single peaks at 395 and 400 nm
for 3,5-F2Y• and 2,3,5-F3Y•, respectively, which are signifi-
cantly blue-shifted from the larger 407 nm absorbance peak of
Y•.
Incorporation of 3-FY into R2. To demonstrate the ability
of FnYs as spectroscopic probes of Y sites in biological systems,
we replaced all Ys in R2 with 3-FY using global incorporation
methods. The successful procedure required the growth of the
bacteria in minimal media from which Y was omitted and
replaced by 3-FY. Under these conditions, the doubling time
of the bacteria was 40 min. SDS-PAGE analysis of whole cells
after induction with IPTG suggests that the expression level of
3-FY-R2 is quite similar to that observed for wt R2 (20-25
mg/g of cell paste), which has a monomeric MW of ∼43 kDa
(Figure 10, lane 1). 3-FY-R2 was isolated using the standard
isolation procedure for wt R2. SDS-PAGE analysis of purified
3-FY-R2 (Figure 10, lane 4) shows that the purified protein is
∼90% homogeneous. Amino acid analysis indicates ∼90%
incorporation of 3-FY. EPR analysis of this material using wt
E. coli R2 as an integration standard69 revealed 0.6-0.8 FY•s/
dimer. The absorption spectrum of the 3-FY122• (Figure S1,
Supporting Information) shows that λmax blue-shifts to 402 nm
(wt Y122• has a λmax of 411 nm), similar to the results from
the TA experiments described above, suggesting that the protein
environment is not perturbing the spectrum.
Figure 8. Transient absorption spectra recorded at (red s) 100 ns, (black
s) 400 ns, and (green s) 1 µs following 300 nm, 5 ns excitation of a 500
µM solution of BPA-Y-OMe buffered to pH 4.0 with 20 mM succinic
acid. Inset: Single wavelength kinetics of the 547 nm absorption (O) and
the single-exponential decay fit (red s) to the data.
as well as DFT-based simulations were unsatisfactory, as has
been previously observed with 3-FY• (David Britt, personal
communication). These results suggest that frozen glasses of
FnY• contain a wide distribution of dihedral angles of the
â-methylene hydrogen atoms. In Y•, one of these hydrogen
atoms is responsible for the strongest hyperfine coupling to the
unpaired electron.63 Since the strength of this interaction varies
as cos2 θ, where θ is the dihedral angle between the plane of
the aromatic ring and the â-methylene hydrogen, a distribution
over these angles can broaden the spectra.64
The absorption spectra of the fluorotyrosyl radical analogues
in a dipeptide (BPA-FnY-OMe) were obtained using the flash
photolysis technique with benzophenone as the photooxidant
of tyrosine. The triplet excited state of benzophenone has been
shown to be a competent, bimolecular oxidant of tyrosine with
an excited-state reduction potential of 1.69 V vs NHE and a
bimolecular rate constant for the reaction with Y of 2.6 ( 0.2
× 109 M-1 s-1 65
.
Moreover, the absorption spectrum of the
benzophenone ketyl radical, produced upon tyrosine photooxi-
dation, does not significantly overlap with that of the fluoro-
tyrosyl radical analogue.
The photosensitized oxidation of tyrosine by benzophenone
was accomplished by intramolecular ET within the dipeptide
pair. Excitation of a solution of BPA-Y-OMe at pH 4.0 with
a 300-nm nanosecond laser pulse produces a species with the
transient absorption spectrum shown in Figure 8 at 100 ns. The
strong, sharp peak with λmax ) 334 nm and broad peak at λmax
) 547 nm with a shoulder to the blue are the spectral signatures
of the reduced benzophenone ketyl radical66 (ꢀ550 nm(BPK•) )
3300 ( 700 M-1 cm-1 in cyclohexane),67 whereas the absorp-
tion bands at λmax ) 390 and 407 nm are characteristic of Y•
(ꢀ410 nm(Y•) ) 2750 ( 200 M-1 cm-1).68 The Y• peak is slightly
blue-shifted from that of free Y•, owing to their overlap with
the tailing absorption of the 334 nm band of the ketyl radi-
cal of BPA (BPA•). The subsequent gray traces recorded at
The EPR spectrum of wt R2 from E. coli has previously been
characterized at 9, 35, 140, 195, 245, and 280 GHz and has
anisotropic g values of gx ) 2.00912-2.00868, gy ) 2.00457-
2.0043, and gz ) 2.00225-2.00208.28 The X-band EPR
spectrum of wt R2 is dominated by the well-resolved, nearly
isotropic coupling to one of its â-Hs (Ax ) 61.2 MHz, Ay ) Az
(64) Warncke, K.; Babcock, G. T.; McCracken, J. J. Phys. Chem. 1996, 100,
4654.
(65) Canonica, S.; Hellrung, B.; Wirz, J. J. Phys. Chem. A 2000, 104, 1226.
(66) Kajii, Y.; Itabashi, H.; Shibuya, K.; Obi, K. J. Phys. Chem. 1992, 96, 7244.
(67) Johnston, J. L.; Lougnot, D. J.; Wintgens, V.; Scaiano, J. C. J. Am. Chem.
Soc. 1988, 110, 518.
(68) Feitelson, J.; Hayon, E. J. Phys. Chem. 1973, 77, 10.
9
J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006 1577

A R T I C L E S
Seyedsayamdost et al.
Figure 10. SDS gel analysis of 3-FY-R2 with lanes: (1) overexpression
of 3-FY-R2, (2) high-range molecular weight marker, (3) low-range
molecular weight marker, and (4) purified 3-FY-R2.
Figure 11. (Top) EPR spectrum (solid line) and simulation (dotted line)
of 3-FY-R2 at 25 K. The values used for the simulation are listed in Table
2. (Bottom) EPR spectrum of wt R2 under the same conditions.
Table 2. A-Tensors of Nuclei with Significant Hyperfine Couplings
in 3-FY•-R2
nucleus
Ax (MHz)
Ay (MHz)
Az (MHz)
φa
P3-F
Pâ-H1
P5-H
-15 (6)
48 (6)
-25 (1)
3
-3 (4)
54 (4)
-8 (1)
9
180 (2)
50 (2)
-17 (1)
6
0
0
25
65
Figure 9. The transient absorption spectrum of (black s) BPA-Y-OMe,
(light blue s) BPA-3-FY-OMe, (dark blue s) BPA-3,5-F2Y-OMe, (red
s) BPA-2,3-F2Y-OMe, (purple s) BPA-2,3,5-F3Y-OMe, (green s)
BPA-2,3,6-F3Y-OMe, and (orange s) BPA-F4Y-OMe normalized to
the peak at 547 nm obtained 100 ns after excitation of ∼500 µM solutions
of each dipeptide buffered to pH 4.0 with 20 mM succinic acid.
P2,6-H
a
φ angles describe rotation about the axis normal to the plane of the
phenol ring between the g tensor and hyperfine tensors.
) 53.7 MHz).70 The second â-H coupling is much smaller and
is usually buried in the EPR line width.71 This study presents
the first observation of the magnetic properties of a 3-FY• in a
protein. The X-band EPR spectrum of this radical, along with
the simulation, is shown in Figure 11. Table 2 presents the
A-tensors of the nuclei with significant hyperfine couplings to
the unpaired electron. The splitting due to coupling to the â-H
is similar to that observed for the Y122• in wt R2. However, a
strongly split doublet of doublets, demarcated by brackets in
Figure 11, is apparent at the extremes of the spectrum due to
coupling to the fluorine nucleus. Together, the results above
show that 3-FY can be incorporated at high efficiency into R2,
yielding 3-FY-R2 with high radical content and activity.
Further, the EPR spectrum shows that the 3-FY122• has well-
9
1578 J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006

New Probes for Enzymes That Use Tyrosyl Radicals
A R T I C L E S
resolved hyperfine interactions that can be simulated, supporting
our contention that the FnY• EPR spectra of Figure 7 are broad
due to a distribution in the dihedral angles of â-methylene
hydrogens.
RNR activity is preserved with global substitution of the 14
Y residues of R2. The activity of the 3-FY-R2/R1 complex
was measured by a coupled spectrophotometric assay using TR/
TRR/NADPH as the source of reductant.69 The specific activity
obtained is ∼65%, that of wt R2, suggesting that substitution
of the 14 Y residues in R2 with 3-FY minimally perturbs the
tertiary structure of 3-FY-R2.
The FnYs will be useful for a variety of mechanistic studies
in biology, in which tyrosine is integral to radical initiation and
transport. We demonstrate one example of this utility in the
companion paper in which Y356 of R2 is site-specifically
replaced with the FnYs presented in this study. A study of the
pH-dependent activity of each of these new semisynthetic
enzymes cast with the backdrop of the Ac-FnY-NH2’s pKa
and Ac-FnY•-NH2’s reduction potential data, reported herein,
allows for mechanistic details of the PCET reactivity at this
site to be unveiled.
Acknowledgment. We thank Prof. Robert Phillips for the
generous gift of the plasmid, pTZTPL, used for encoding the
enzyme tyrosine phenol lyase (TPL), Dave Bray of the MIT
DCIF for help with Figure 3, Julien Bachmann for assis-
tance with DFT calculations, R. David Britt for discussions
regarding fitting of the FnY• EPR spectra, Stoyan K. Smoukov
and Brian M. Hoffman for fitting the EPR spectrum of
3-FY122•, Hiep-Hoa T. Nguyen for preparation of 3-FY-R2,
Cyril S. Yee for initial work with Ac-2,3-F2Y-NH2, John H.
Robblee for helpful discussions, and the National Institutes of
Health for support of this work GM47274 (D.G.N.) and
GM29595 (J.S.).
Conclusions
The mechanistic complexity arising from the coupling of
electron and proton transfers in charge-transport processes
involving amino acid radicals requires the development of new
tools and methods for their study. We show here that FnYs will
be useful in the study of biological charge-transport mechanisms
involving tyrosine for the following reasons. (1) They provide
a range of pKas for the phenolic proton and Eps for the Y•/Y
couple, which allows for tuning of the ∆G°PT and ∆G°ET for
the PCET reaction. UV-vis and DPV analysis illustrate that
the Ac-FnY-NH2’s offer a range of almost 5 units in pKa and
320 mV in redox potential within the pH range of 6-9, where
proteins can be studied. (2) The oxidized forms of FnY•s possess
unique spectral signals allowing a distinction between Y• and
FnY• in the same system. The time-resolved absorption spectra
of BPA-FnY-OMe dipeptides show that the λmax for FnY•
absorption can be shifted by ∼10 nm relative to Y•. (3) In
addition to glutathione transferase72,73 and other proteins,74-79
3-FY can be incorporated into RNR with relatively minimal
structural perturbation, as demonstrated by the preserved activity
of the R2 subunit of class I E. coli RNR, in which 14 Y residues
were globally replaced by 3-FY.
Supporting Information Available: Details of the E. coli
growth medium for 3-FY-R2 preparation; the UV-vis spec-
trum of 3-FY-R2; sample different pulse voltammograms of
Ac-Y- -NH2 and Ac-FnY- -NH2s; Cartesian coordinates
and atom-numbering scheme used for DFT calculations; MO
energy-level diagrams of the tyrosine and fluorotyrosine deriva-
tives. This material is available free of charge via the Internet
at http://pubs.acs.org.
JA055926R
(74) Hull, W. E.; Sykes, B. D. Biochemistry 1976, 15, 1535.
(75) Kimber, B. J.; Griffths, D. V.; Birdsall, B.; King, R. W.; Scudder, P.;
Freeney, J.; Roberts, G. C. K.; Burgen, A. S. V. Biochemistry 1977, 16,
3492.
(69) Ge, J.; Perlstein, D. L.; Nguyen, H.-H.; Bar, G.; Griffin, R. G.; Stubbe, J.
Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 10067.
(70) Bender, C. J.; Sahlin, M.; Babcock, G. T.; Barry, B. A.; Chandrashekar,
T. K.; Salowe, S. P.; Stubbe, J.; Lindstro¨m, B.; Petersson, L.; Ehrenberg,
A.; Sjo¨berg, B.-M. J. Am. Chem. Soc. 1989, 111, 8076.
(76) Sixl, F.; King, R. W.; Bracken, M.; Feeney, J. Biochem. J. 1990, 266,
545.
(71) Bar, G.; Bennati, M.; Nguyen, H. H.; Ge, J.; Stubbe, J.; Griffin, R. G. J.
Am. Chem. Soc. 2001, 123, 3569.
(77) Eccleston, J. F.; Molloy, D. P.; Hinds, M. G.; King, R. W.; Feeney, J. Eur.
J. Biochem. 1993, 278, 1041.
(72) Parsons, J. F.; Armstrong, R. N. J. Am. Chem. Soc. 1996, 118, 2295.
(73) Xiao, G.; Parsons, J. F.; Armstrong, R. N.; Gilliland, G. L. J. Am. Chem.
Soc. 1997, 119, 9325.
(78) Lui, S. M.; Cowan, J. A. J. Am. Chem. Soc. 1994, 116, 4483.
(79) Pal, P. P.; Bae, J. H.; Azim, M. K.; Hess, P.; Friedrich, R.; Huber, R.;
Moroder, L.; Budisa, N. Biochemistry 2005, 44, 3663.
9
J. AM. CHEM. SOC. VOL. 128, NO. 5, 2006 1579