Characterization and mechanistic studies of DesII: A radical S-adenosyl-l-methionine enzyme involved in the biosynthesis of TDP-D-desosamine

Article abstract of DOI:10.1021/ja903354k

D-Desosamine (1) is a 3-(N,N-dimethylamino)-3,4,6-trideoxyhexose found in a number of macrolide antibiotics including methymycin (2), neomethymycin (3), pikromycin (4), and narbomycin (5) produced by Streptomyces venezuelae. It plays an essential role in

Full text of DOI:10.1021/ja903354k

Published on Web 09/11/2009
Characterization and Mechanistic Studies of DesII: A Radical
S-Adenosyl-L-methionine Enzyme Involved in the
Biosynthesis of TDP-D-Desosamine
Ping-Hui Szu,† Mark W. Ruszczycky, Sei-hyun Choi, Feng Yan,‡ and Hung-wen Liu*
DiVision of Medicinal Chemistry, College of Pharmacy, and Department of Chemistry and
Biochemistry, UniVersity of Texas at Austin, Austin, Texas 78712
Received April 25, 2009; E-mail: h.w.liu@mail.utexas.edu
Abstract: D-Desosamine (1) is a 3-(N,N-dimethylamino)-3,4,6-trideoxyhexose found in a number of macrolide
antibiotics including methymycin (2), neomethymycin (3), pikromycin (4), and narbomycin (5) produced by
Streptomyces venezuelae. It plays an essential role in conferring biological activities to its parent aglycones.
Previous genetic and biochemical studies of the biosynthesis of desosamine in S. venezuelae showed
that the conversion of TDP-4-amino-4,6-dideoxy-D-glucose (8) to TDP-3-keto-4,6-dideoxy-D-glucose (9) is
catalyzed by DesII, which is a member of the radical S-adenosyl-L-methionine (SAM) enzyme superfamily.
Here, we report the purification and reconstitution of His6-tagged DesII, characterization of its [4Fe-4S]
cluster using UV-vis and EPR spectroscopies, and the capability of flavodoxin, flavodoxin reductase, and
NADPH to reduce the [4Fe-4S]2+ cluster. Also included are a steady-state kinetic analysis of DesII-catalyzed
reaction and an investigation of the substrate flexibility of DesII. Studies of deuterium incorporation into
SAM using TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose as the substrate provides strong evidence for direct
hydrogen atom transfer to a 5′-deoxyadenosyl radical in the catalytic cycle. The fact that hydrogen atom
abstraction occurs at C-3 also sheds light on the mechanism of this intriguing deamination reaction.
D-Desosamine (1) is a 3-(N,N-dimethylamino)-3,4,6-trideoxy-
hexose found in a number of macrolide antibiotics including
methymycin (2), neomethymycin (3), narbomycin (4), and
pikromycin (5) produced by Streptomyces Venezuelae. Both
biochemical and structural studies have shown that desosamine
is an essential structural component crucial to the biological
activity of the parent aglycones (e.g., 12, 13).1 Desosamine is
biosynthetically derived from thymidine diphosphate (TDP)-D-
glucose (6), and a key step in its formation is the removal of
the C-4 hydroxyl group of the hexose ring. Early studies of
desosamine biosynthesis in S. Venezuelae revealed that the C-4
deoxygenation is unique among biological deoxygenation pro-
cesses2 and the reaction proceeds in two stages: the conversion
of TDP-4-keto-6-deoxy-D-glucose (7) to the corresponding
4-amino sugar intermediate (8), and the deamination of 8 to
afford TDP-3-keto-4,6-dideoxy-D-glucose as the final product
(9, Scheme 1). The former reaction is catalyzed by DesI, a
pyridoxal-5′-phosphate (PLP)-dependent aminotransferase,
whereas the latter reaction is catalyzed by DesII, a radical
S-adenosyl-L-methionine (SAM)-dependent enzyme.3 Subse-
quent C-3 aminotransfer (9 f 10) by DesV followed by N,N-
dimethylation (10 f 11) by DesVI complete the desosamine
biosynthesis.4 All enzymes in the pathway have been biochemi-
cally characterized, but the details of the catalytic properties of
DesII and the mechanism of its catalysis remain obscure.
Thus far, more than 2800 proteins have been identified,
mainly based on sequence alignment, as members of the radical
SAM enzyme superfamily.5 These enzymes catalyze a variety
of reactions, such as isomerization, protein radical formation,
sulfur insertion, ring formation, anaerobic oxidation, and unusual
methylations, and are involved in the biosynthesis of DNA
precursors, cofactors, vitamins, many types of secondary
metabolites, and various biodegradation pathways.6 Enzymes
of this superfamily possess a characteristic conserved sequence
motif, CXXXCXXC, which coordinates a [4Fe-4S] cluster.7
Four different oxidation states of the iron-sulfur cluster have
been observed. The intact iron-sulfur cluster exists mainly as
[4Fe-4S]2+ in the purified enzyme, but [4Fe-4S]3+ is found as
(3) (a) Szu, P.-h.; He, X.; Zhao, L.; Liu, H.-w. Angew. Chem., Int. Ed.
2005, 44, 6742–6746. (b) Zhao, L.; Borisova, S.; Yeung, S.-M.; Liu,
H.-w. J. Am. Chem. Soc. 2001, 123, 7909–7910.
† Current address: Department of Chemistry, Stanford University,
Stanford, CA 94305.
(4) Chen, H.; Yamase, H.; Murakami, K.; Chang, C.-w.; Zhao, L.; Zhao,
Z.; Liu, H.-w. Biochemistry 2002, 41, 9165–9183.
‡ Current address: Cellular and Molecular Pharmacology, University of
California, San Francisco, CA 94158.
(5) (a) Frey, P. A.; Hegeman, A. D.; Ruzicka, F. J. Crit. ReV. Biochem.
Mol. Biol. 2008, 43, 63–88. (b) Sofia, H. J.; Chen, G.; Hetzler, B. G.;
Reyes-Spindola, J. F.; Miller, N. E. Nucleic Acids Res. 2001, 29, 1097–
1106.
(1) Schlunzen, F.; Zarivach, R.; Harms, J.; Bashan, A.; Tocilj, A.; Albrecht,
R.; Yonath, A.; Franceschi, F. Nature 2001, 413, 814–821.
(2) (a) He, X., H.-w. Annu. ReV. Biochem. 2002, 71, 701–754. (b) He,
X. M.; Liu, H.-w. Curr. Opin. Chem. Biol. 2002, 6, 590–597. (c)
Thibodeaux, C. J.; Melancon, C. E., III; Liu, H.-w. Nature 2007, 446,
1008–1016. (d) Thibodeaux, C. J.; Melancon, C. E., III; Liu, H.-w.
Angew. Chem., Int. Ed. 2008, 47, 9814–9859.
(6) (a) Frey, P. A.; Magnusson, O. T. Chem. ReV. 2003, 103, 2129–2148.
(b) Cheek, J.; Broderick, J. B. J. Biol. Inorg. Chem. 2001, 6, 209–
226. (c) Fontecave, M.; Mulliez, E.; Ollagnier-de-Choudens, S. Curr.
Opin. Chem. Biol. 2001, 5, 506–511. (d) Jarrett, J. T. Curr. Opin.
Chem. Biol. 2003, 7, 174–182.
9
14030 J. AM. CHEM. SOC. 2009, 131, 14030–14042
10.1021/ja903354k CCC: $40.75  2009 American Chemical Society

DesII: A Radical SAM Enzyme
A R T I C L E S
Scheme 1
a minor component and readily converts to [3Fe-4S]1+ via air
oxidation. Upon treatment with dithionite in the presence of
SAM, the iron-sulfur cluster is reduced to [4Fe-4S]1+, which
is the catalytically active form of all radical SAM enzymes.8
Catalysis by this class of enzymes is always initiated by one
electron transfer from the [4Fe-4S]1+ cluster to SAM. This
induces the homolytic cleavage of the C5′-S bond of SAM to
generate methionine and a 5′-deoxyadenosyl radical (22, see
Schemes 5 and 6).6 The subsequent abstraction of a hydrogen
atom from the substrate by the reactive 5′-deoxyadenosyl radical
triggers the chemical transformations during turnover.6,9
Studies of this class of enzymes are difficult because
reconstitution of the iron-sulfur cluster is technically challeng-
ing and the reduced [4Fe-4S]1+ cluster is highly oxygen
sensitive. Thus far, only a handful of radical SAM enzymes
have been investigated. Better known examples include lysine
2,3-aminomutase,7f,10 pyruvate formate-lyase activase,7a,8a,11
anaerobic ribonucleotide reductase activase,7b,8c,12 biotin
synthase,7d,13 lipoyl synthase,14 spore photoproduct lyase,15 and
BtrN,16 a radical SAM dehydrogenase involved in the biosyn-
thesis of butirosin. More recently, 4-amino-5-hydroxymethyl-
2-methylpyrimidine phosphate (HMP-P) synthase (ThiC) in the
thiamine pyrimidine biosynthetic pathway has been character-
ized as a previously unrecognized radical SAM enzyme using
SAM as a cosubstrate and producing HMP and 5′-deoxyad-
enosine as the products.17
(12) (a) Eliasson, R.; Fontecave, M.; Jo¨rnvall, H.; Krook, M.; Pontis, E.;
Reichard, P. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 3314–3318. (b)
Mulliez, E.; Fontecave, M.; Gaillard, J.; Reichard, P. J. Biol. Chem.
1993, 268, 2296–2299. (c) Liu, A.; Graslund, A. J. Biol. Chem. 2000,
275, 12367–12373. (d) Tamarit, J.; Mulliez, E.; Meier, C.; Trautwein,
A.; Fontecave, M. J. Biol. Chem. 1999, 274, 31291–31296.
(13) (a) Ugulava, N. B.; Gibney, B. R.; Jarrett, J. T. Biochemistry 2001,
40, 8343–8351. (b) Ugulava, N. B.; Sacanell, C. J.; Jarrett, J. T.
Biochemistry 2001, 40, 8352–8358. (c) Taylor, A. M.; Farrar, C.;
Jarrett, J. T. Biochemistry 2008, 47, 9309–9317.
(7) (a) Ku¨lzer, R.; Pils, T.; Kappl, R.; Hu¨ttermann, J.; Knappe, J. J. Biol.
Chem. 1998, 273, 4897–4903. (b) Tamarit, J.; Gerez, C.; Meier, C.;
Mulliez, E.; Trautwein, A.; Fontecave, M. J. Biol. Chem. 2000, 275,
15669–15675. (c) Layer, G.; Moser, J.; Heinz, D. W.; Jahn, D.;
Schubert, W.-D. EMBO J. 2003, 22, 6214–6224. (d) Berkovitch, F.;
Nicolet, Y.; Wan, J. T.; Jarrett, J. T.; Drennan, C. L. Science 2004,
303, 76–79. (e) Ha¨nzelmann, P.; Schindelin, H. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 12870–12875. (f) Lepore, B. W.; Ruzicka, F. J.;
Frey, P. A.; Ringe, D. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 13819–
13824.
(14) (a) Cicchillo, R. M.; Iwig, D. F.; Jones, A. D.; Nesbitt, N. M.; Baleanu-
Gogonea, C.; Souder, M. G.; Tu, L.; Booker, S. J. Biochemistry 2004,
43, 6378–6386. (b) Cicchillo, R. M.; Lee, K.-H.; Baleanu-Gogonea,
C.; Nesbitt, N. M.; Krebs, C.; Booker, S. J. Biochemistry 2004, 43,
11770–11781. (c) Cicchillo, R. M.; Booker, S. J. J. Am. Chem. Soc.
2005, 127, 2860–2861. (d) Douglas, P.; Kriek, M.; Bryant, P.; Roach,
P. L. Angew. Chem. 2006, 118, 5321–5323.
(15) (a) Rebeil, R.; Sun, Y.; Chooback, L.; Pedraza-Reyes, M.; Kinsland,
C.; Begley, T. P.; Nicholson, W. L. J. Bacteriol. 1998, 180, 4879–
4885. (b) Rebeil, R.; Nicholson, W. L. Proc. Natl. Acad. Sci. U.S.A.
2001, 98, 9038–9043. (c) Cheek, J.; Broderick, J. B. J. Am. Chem.
Soc. 2002, 124, 2860–2861. (d) Buis, J. M.; Cheek, J.; Kalliri, E.;
Broderick, J. B. J. Biol. Chem. 2006, 281, 25994–26003. (e) Chandra,
T.; Silver, S. C.; Zilinskas, E.; Shepard, E. M.; Broderick, W. E.;
Broderick, J. B. J. Am. Chem. Soc. 2009, 131, 2420–2421.
(8) (a) Henshaw, T. F.; Cheek, J.; Broderick, J. B. J. Am. Chem. Soc.
2000, 122, 8331–8332. (b) Lieder, K.; Booker, S.; Ruzicka, F. J.;
Beinert, H.; Reed, G. H.; Frey, P. A. Biochemistry 1998, 37, 2578–
2585. (c) Padovani, D.; Thomas, F.; Trautwein, A. X.; Mulliez, E.;
Fontecave, M. Biochemistry 2001, 40, 6713–6719.
(9) (a) Frey, P. A.; Magnusson, O. T. Chem. ReV. 2003, 103, 2129–2148.
(b) Frey, P. A. Annu. ReV. Biochem. 2001, 70, 121–148.
(16) (a) Yokoyama, K.; Numakura, M.; Kudo, F.; Ohmori, D.; Eguchi, T.
J. Am. Chem. Soc. 2007, 129, 15147–15155. (b) Yokoyama, K.;
Ohmori, D.; Kudo, F.; Eguchi, T. Biochemistry 2008, 47, 8950–8960.
(17) (a) Chatterjee, A.; Li, Y.; Zhang, Y.; Grove, T. L.; Lee, M.; Krebs,
C.; Booker, S. J.; Begley, T. P.; Ealick, S. E. Nat. Chem. Biol. 2008,
4, 758–765. (b) Martinez-Gomez, N. C.; Downs, D. M. Biochemistry
2008, 47, 9054–9056. (c) Martinez-Gomez, N. C.; Poyner, R. R.;
Mansoorabadi, S. O.; Reed, G. H.; Downs, D. M. Biochemistry 2009,
48, 217–219.
(10) (a) Chen, D.; Walsby, C.; Hoffman, B. M.; Frey, P. A. J. Am. Chem.
Soc. 2003, 125, 11788–11789. (b) Lieder, K.; Booker, S.; Ruzicka,
F. J.; Beinert, H.; Reed, G. H.; Frey, P. A. Biochemistry 1998, 37,
2578–2585. (c) Wang, S. C.; Frey, P. A. Biochemistry 2007, 46,
12889–12895.
(11) Vey, J. L.; Yang, J.; Li, M.; Broderick, W. E.; Broderick, J. B.;
Drennan, C. L. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 15137.
9
J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009 14031

A R T I C L E S
Szu et al.
chia coli strains using published procedures.3b,21 TDP-D-glucose
and the products of DesII and DesV were synthesized by literature
procedures.21a,22
DesII, which also contains the CXXXCXXC iron-sulfur
cluster binding motif, has been identified as a member of the
radical SAM enzyme superfamily.5b Previous studies have
established the dependence on a [4Fe-4S] cluster and SAM for
DesII activity.3 In this contribution, we report a full account of
the characterization of the [4Fe-4S] cluster in DesII, a steady-
state kinetic analysis of the DesII-catalyzed reaction, an
investigation of the substrate flexibility of DesII, and the
demonstration of a biological reducing system, flavodoxin,
flavodoxin reductase, and NADPH, capable of reducing the
[4Fe-4S]2+ cluster. Moreover, studies of deuterium incorporation
into SAM using TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose
(18, see Scheme 4) provide insight into the mechanism of this
intriguing deamination reaction.
Purification of DesII from Escherichia coli BL21 Star
(DE3)-desII/pET24 Cells. An overnight culture of E. coli BL21
Star (DE3)-desII/pET24b grown at 37 °C in Luria-Bertani (LB)
medium containing kanamycin (50 µg/mL) was used, in a 200-
fold dilution, to inoculate 6 L of the same medium. When the OD600
reached 0.4-0.6, the incubation temperature was lowered to 18
°C and isopropyl-ꢀ-D-galactopyranoside (IPTG) was added to a final
concentration of 0.1 mM to induce gene expression. Fe(NH4)2(SO4)2
was also added to a final concentration of 0.2 mM to provide
additional iron for iron-sulfur cluster biosynthesis. After incubation
for an additional 18 h at 18 °C, cells were harvested by centrifuga-
tion (4000g, 10 min) at 4 °C, washed with 50 mM potassium
phosphate buffer (pH 7.5), collected again by centrifugation (4000g,
10 min), and stored at -80 °C. The typical yield was 6 g of wet
cells per liter of culture. Subsequent purification was carried out at
4 °C, and all buffers were degassed and saturated with nitrogen
before use.
Thawed cell pellets were resuspended in 120 mL of lysis buffer
(50 mM sodium phosphate, 10% glycerol, 300 mM NaCl, 10 mM
imidazole, pH 8.0), and the cells were disrupted by sonication. Cell
debris was removed by centrifugation (12000g, 30 min), and the
supernatant was mixed by slow agitation with 10 mL of packed
Ni-NTA resin (Qiagen, Valencia, CA) for 1 h at 4 °C. The slurry
was poured into a capped column and washed with 100 mL of
wash buffer (50 mM sodium phosphate, 10% glycerol, 300 mM
NaCl, 20 mM imidazole, pH 8.0). The brownish protein was eluted
with elution buffer (50 mM sodium phosphate, 10% glycerol, 300
mM NaCl, 250 mM imidazole, pH 8.0). The desired fractions, as
detected by SDS-PAGE were pooled, dialyzed three times against
1 L of 20 mM tris(hydroxymethyl)aminomethane (Tris) ·HCl buffer
(pH 7.5) containing 15% glycerol, and concentrated by ultrafiltration
(YM-10 membrane). The collected protein was further purified by
FPLC on a MonoQ HR (16/10) column using buffer A (20 mM
Tris ·HCl, pH 7.5) and buffer B (Buffer A plus 0.5 M NaCl). The
column was run with a linear gradient of 0 to 50% B in 25 min,
followed by 50 to 70% B in 2 min. Finally, the column was washed
with 100% B for 5 min. The flow rate was 3 mL/min, and the
detector was set at 280 nm. The DesII protein eluted at 55% B. It
was then dialyzed against 20 mM Tris ·HCl buffer (pH 7.5) and
concentrated by ultrafiltration (YM-10 membrane) prior to storage
at -80 °C.
Construction of the Expression Plasmids for Flavodoxin
(fld) and Flavodoxin Reductase (fpr). The fld gene from E. coli
was PCR amplified and cloned into the pET24b(+) vector
(Novagen, Madison, WI) at the NdeI and XhoI restriction sites. The
primers used were: 5′-TCGCCATATGAACAGCTGCGTAT-
TAATAA-3′ (forward, with the NdeI site underlined) and 5′-
GCAACTCGAGTTAATAAAAACGTGGACA-3′ (reverse, with
the XhoI site underlined). Similarly, the fpr gene from E. coli was
amplified using 5′-AAAACCATGGCTGATTGGGTAAC-3′ (con-
taining an NcoI restriction site), and 5′-AAGTCTCGAGCCAG-
TAATGCTCCGCTGT-3′ (containing an XhoI restriction site) as
the forward and reverse primers, respectively, and cloned into the
pET28b(+) vector (Novagen) at the NcoI and XhoI restriction sites.
The resulting plasmids were used to separately transform E. coli
BL21 Star (DE3) cells for gene expression. The general methods
and protocols for recombinant DNA manipulations followed those
described by Sambrook et al.23
Experimental Procedures
General. Protein concentrations were determined by the method
of Bradford18 using bovine serum albumin as the standard. The
relative molecular mass and purity of enzyme samples were
determined using sodium dodecyl sulfate polyacrylamide gel
electrophoresis (SDS-PAGE) as described by Laemmli.19 The native
molecular mass of DesII was determined by the gel filtration method
of Andrews.20 1H,13C, 31P NMR spectra were recorded using either
a Varian Unity Plus 300, 500, or 600 MHz spectrometer. Mass
spectra were recorded by the MS facility at the Department of
Chemistry and Biochemistry of the University of Texas at Austin.
DNA sequencing and amino-terminal sequencing of the purified
enzyme were performed by the Core Facilities of the Institute of
Cellular and Molecular Biology at the University of Texas at Austin.
Electron paramagnetic resonance (EPR) spectra were recorded at
liquid helium temperature on a Bruker EMX EPR spectrometer at
the University of Texas at Houston Medical School. Analytical thin
layer chromatography (TLC) was carried out on precoated TLC
glass plates, and the spots on TLC plates were visualized first under
UV light and then by heating plates previously stained with
solutions of vanillin/ethanol/H2SO4 (1:98:1) or phosphomolybdic
acid (7% in EtOH). Flash column chromatography was performed
on silica gel by elution with the specified solvents. CarboPac PA1
HPLC columns (4 mm × 250 mm and 9 mm × 250 mm) were
obtained from Dionex (Sunnyvale, CA). The fast protein liquid
chromatography (FPLC) Mono-Q H/R 16/10 column was purchased
from Pharmacia (Uppsala, Sweden). Kinetic data were analyzed
by nonlinear regression using Grafit5 (Erithacus Software, UK).
All anaerobic experiments were performed in a glovebox (Coy
Laboratory Products, Grass Lake, MI) under an atmosphere
consisting of 95% N2 and 5% H2 at less than 5 ppm O2. The catalyst
used to remove the oxygen was regenerated each month by heating
at 120 °C for 2 h.
Materials. All chemicals were analytical grade or the highest
quality commercially available and were used without further
purification unless otherwise noted. Enzymes used in the cloning
experiment were obtained from Invitrogen (Carlsbad, CA), Amer-
sham (Arlington Heights, IL), Gibco BRL (Grand Island, NY) or
Promega (Madison, WI). Culture medium ingredients were pur-
chased from Difco (Detroit, MI). DNA minipreps were performed
using the Wizard DNA purification kit from Promega. All oligo-
nucleotide primers for polymerase chain reaction (PCR) amplifica-
tion of the desired inserts were prepared by Gibco BRL. All
electrophoresis materials were purchased from Gibco BRL or Bio-
Rad (Hercules, CA). The plasmid pET28b containing the desI gene
and the plasmid pET24b containing the desII gene were previously
constructed.3 Enzymes DesI, RfbB (a DesIV equivalent from
Salmonella typhi, which catalyzes the conversion of 6 to 7), and
DesV were purified from the corresponding recombinant Escheri-
(21) (a) Takahashi, H.; Liu, Y.-n.; Liu, H.-w. J. Am. Chem. Soc. 2006,
128, 1432–1433. (b) Zhao, L. Ph.D. Thesis, University of Minnesota,
2000.
(22) Chang, C.-w.; Zhao, L.; Yamase, H.; Liu, H.-w. Angew. Chem., Int.
Ed. 2000, 39, 2160–2163.
(23) Sambrook, J.; Russell, D. W. Molecular Cloning: A Laboratory
Manual, 3rd ed.; Cold Spring Harbor Laboratory Press: Plainview,
NY, 2001.
(18) Bradford, M. M. Anal. Biochem. 1976, 72, 248–254.
(19) Laemmli, U. K. Nature 1970, 227, 680–685.
(20) Andrews, P. Biochem. J. 1964, 91, 222–233.
9
14032 J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009

DesII: A Radical SAM Enzyme
A R T I C L E S
Expression and Purification of E. coli Flavodoxin (FLD)
and Flavodoxin Reductase (FPR). Overnight cultures of E. coli
BL21(DE3)-fld/pET24b(+) and fpr/pET28b(+) separately grown
at 37 °C in LB medium containing kanamycin (50 µg/mL) were
each used, in a 100-fold dilution, to inoculate 6 L of the same
medium. These cultures were incubated at 37 °C until the OD600
reached 0.5-0.6. Protein expression was then induced by the
addition of IPTG to a final concentration of 0.1 mM. After
incubation at 37 °C for an additional 6 h, the cells were harvested
by centrifugation (4000g, 10 min) at 4 °C, washed with 50 mM
potassium phosphate buffer (pH 7.5), and stored at -80 °C. The
typical yield was 6 g of wet cells per liter of culture. Procedures
for the subsequent protein purification using Ni-NTA affinity
chromatography were identical to those used for DesII purification.
Iron Content Determination. DesII samples (1 mL each) for
iron analysis24 were mixed with 500 µL of reagent A (1:1 of 4.5%
KMnO4:1.2 N HCl) and incubated at 60 °C for 2 h. To these
samples, 100 µL of reagent B (8.8 g of ascorbic acid, 9.7 g of
ammonium acetate, 80 mg of ferrozine, 80 mg of neocuproine, and
ddH2O to 25 mL total volume) was added with immediate
vortexing. The absorbance at 562 nm was recorded after the samples
had incubated for 1 h at room temperature. The iron concentration
was determined by comparing the reading to a standard curve
prepared using Fe(NH4)2(SO4)2.
Sulfur Content Determination. The sulfur content of DesII was
determined following a literature procedure.25 All glassware used
in this experiment was acid-washed and all solutions were prepared
using metal-free distilled H2O. Approximately 75 µg of DesII (1.5
nmol, in duplicate) was diluted with water or 20 mM Tris ·HCl
buffer (pH 7.6) to a total volume of 200 µL. A freshly prepared
deoxygenated 1% zinc acetate solution (600 µL) was added
followed immediately by the addition of 30 µL of 12% NaOH.
The mixture was stirred until the color became homogeneous. After
the reaction was incubated at room temperature for 1 h, 150 µL of
DMPD solution (0.1% N,N-dimethyl-p-phenylenediamine mono-
hydrochloride in 5 N HCl) was added. The resulting solution was
gently stirred until only the top 2-mm layer retained undissolved
zinc hydroxide as indicated by a pink color. A 30 µL FeCl3 (23
mM FeCl3 in 1.2 N HCl) solution was added and mixed rapidly to
generate a colorless solution. The mixture was incubated for 30
min and the color of the solution changed to blue. This blue solution
was centrifuged to remove precipitated protein, and the absorbance
at 670 nm (ε ) 34,500 M-1 cm-1) was recorded. A blank (in the
absence of DesII) and a series of Na2S standards (in 0.03% NaOH
aqueous solution) were measured in parallel with the DesII sample.
The sulfur content was determined by comparison to the standard
curve derived from the Na2S standards.
5 mM dithiothreitol [DTT]) for 6 h at room temperature. A
deaerated solution of ethylenediaminetetraacetic acid (EDTA) (2
mM final concentration) was then added to the protein. The resulting
solution was incubated for 30 min. Protein was loaded onto a 50
mL Sephadex G-25 column equilibrated and eluted with 100 mM
Tris ·HCl buffer, pH 8. Green-gray fractions were collected and
concentrated to approximately 10 mg/mL by ultrafiltration through
a YM10 membrane (Amicon) (75-80% recovery). This preparation
was then analyzed for iron and sulfur content, and the iron-sulfur
cluster was characterized by UV-vis and EPR spectroscopy.
Step 3: Reduction of [4Fe-4S] Center. Reduction of the [4Fe-
4S] center of the reconstituted protein was monitored by UV-vis
spectroscopy. The reducing agent was sodium dithionite, which was
prepared freshly in 100 mM Tris ·HCl buffer, pH 8, prior to the
reduction procedure. The protein solution was reduced with a 6-fold
molar excess of sodium dithionite over 1 h. Reduction was
monitored by the decrease of the absorption at 420 nm.
EPR Spectroscopy. All samples were freshly prepared anaero-
bically inside a glovebox. Each EPR sample (500 µL) contained 2
mM dithionite, 1.5 mM SAM, and 200 µM reconstituted DesII in
100 mM Tris · HCl buffer (pH 8.0, including 1 mM DTT). The
solution was mixed, transferred to an EPR tube, and frozen by
immersing the tube in cold isopentane (approximately -140 °C)
within 30 s after mixing. The dithionite was omitted in the control
sample. These EPR samples were stored in liquid nitrogen until
analysis. EPR spectra were recorded at liquid helium temperature
on a Bruker EMX EPR spectrometer. A GFS600 transfer line and
an ITC503 temperature controller were used to maintain the
temperature. An Oxford ESR900 cryostat was used to accommodate
the sample. Data analysis was conducted using WinEPR provided
by Bruker.
DesII Activity Assay Using Dithionite as the Reducing
Agent. The DesII protein was reduced with a 6-fold molar excess
of sodium dithionite over 1 h. A typical activity assay contained
0.5 mM TDP-4-amino-4,6-dideoxy-D-glucose (8), 0.1 mM dithion-
ite-reduced DesII, and 0.1 mM SAM in 1 mL of 100 mM Tris ·HCl
buffer (pH 8, including 2 mM DTT). The reaction mixture was
incubated at 25 °C for 3 h, and stopped by filtration through a YM10
membrane (Amicon) to remove the enzyme. The reaction mixture
was analyzed by HPLC on a Dionex CarboPac PA1 column (4 ×
250 mm). A linear gradient from 20% to 35% 1 M ammonium
acetate, pH 7.0, versus ddH2O over 40 min gave satisfactory
separation between substrate 8 (with a retention time of 5.9 min)
and the product, TDP-3-keto-4,6-dideoxy-D-glucose (9) (with a
retention time of 28.2 min). The flow rate was 0.6 mL/min, and
the detector was set at 267 nm. The product peak was collected
and its identity was verified by high-resolution mass spectrometry
analysis.
Coupled DesII and DesV Assay. The assay was carried out in
1 mL of 100 mM Tris ·HCl buffer (pH 8, including 2 mM DTT)
containing 0.5 mM TDP-4-amino-4,6-dideoxy-D-glucose (8), 0.1
mM dithionite-reduced DesII, and 0.1 mM SAM. After incubation
at 25 °C for 3 h, 10 mM of L-glutamate, 0.8 mM of PLP, and 0.1
mM DesV enzyme were added to the assay mixture. The reaction
was incubated at 25 °C for another 30 min and then was stopped
by removing the enzymes by ultrafiltration through a YM10
membrane (Amicon). The reaction solution was analyzed by HPLC
as described above. The new product, TDP-3-amino-3,4,6-trideoxy-
D-glucose (10), which eluted at 3.7 min, was collected and subjected
to high-resolution mass spectrometry verification.
Reconstitution of the [4Fe-4S] Cluster of DesII in vitro. All
procedures including the apoprotein preparation, reconstitution,
reduction of the iron-sulfur cluster, and activity assay were
performed anaerobically inside a glovebox. All solutions were
deoxygenated prior to use by repeated freeze-vacuum-thaw cycles
under argon, and all solids were deoxygenated prior to use by
vacuum and argon cycles.
Step 1: Preparation of Apo-DesII. Protein-bound iron was
removed by incubating the as-isolated DesII in the presence of 100
mM EDTA and 2 mM sodium dithionite in 100 mM Tris · HCl
buffer, pH 8, for 1 h at room temperature. The solutions became
colorless after 30 min of reduction with sodium dithionite. The
resulting protein was loaded onto a Sephadex G-25 column (50
mL) pre-equilibrated and eluted with 100 mM Tris ·HCl buffer,
pH 8, and concentrated to approximately 8 mg/mL by ultrafiltration
through a YM10 membrane (Amicon) (80% recovery).
DesII Activity Assay Using Flavodoxin and Flavodoxin
Reductase as the Reducing System. A typical activity assay
contained 0.5 mM TDP-4-amino-4,6-dideoxy-D-glucose (8), 0.1 mM
reconstituted DesII, 0.1 mM SAM, 1 mM NADPH, 12 µM
flavodoxin (FLD), and 6 µM flavodoxin reductase (FPR) in 1 mL
of 100 mM Tris ·HCl buffer (pH 8) with 2 mM DTT. The reaction
mixture was incubated at 25 °C for 1 h and stopped by ultrafiltration
through a YM10 membrane (Amicon) to remove the enzymes. The
reaction mixture was analyzed by HPLC on a Dionex CarboPac
Step 2: Reconstitution of [4Fe-4S] Center. The above apo-
protein was incubated with a 10-fold molar excess of Na2S and
Fe(NH4)2(SO4)2 in 1 mL of 100 mM Tris ·HCl buffer (pH 8, with
(24) Fish, W. W. Methods Enzymol. 1988, 158, 357–364.
(25) Beinert, H. Anal. Biochem. 1983, 131, 373–378.
9
J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009 14033

A R T I C L E S
Szu et al.
Scheme 2
anion-exchange column (4 × 250 mm). The flow rate was 1.0 mL/
min, and the detector was set at 267 nm. A linear gradient from
2.5% to 15% 1 M ammonium acetate, pH 7.0, versus ddH2O over
20 min, followed by a second linear gradient from 15% to 20% 1
M ammonium acetate, pH 7.0, versus ddH2O over 20 min was used
to separate the substrate 14 (with a retention time of 18.9 min) and
the resulting product (with a retention time of 27.8 min).
Characterization of the DesII and DesII/DesV Products
from TDP-D-quinovose (14). A 4.3 mL solution of 5.25 mM TDP-
D-quinovose (14) containing an excess of SAM (7.8 mM) and
sodium dithionite (10 mM) was incubated anaerobically in the 2
mM DTT, 100 mM Tris ·HCl buffer, pH 8.0, with 2 µM DesII at
room temperature. Following a 7 h incubation the reaction was
judged >95% complete by analytical HPLC as previously described.
At this point, the reaction was divided into two equal fractions.
The first fraction was filtered to remove protein, and the new
product was isolated using a Dionex CarboPac PA1 semiprep
column with a flow rate of 3 mL/min and a two-stage linear
gradient: 100% ddH2O to 75% 300 mM NH4HCO3 in 20 min
followed by 75 to 100% 300 mM NH4HCO3 in 20 min. The new
product was collected at 24.8 min, lyophilized to dryness, and
characterized by electrospray ionization (ESI) mass spectroscopy.
The second fraction was made 1 mM in PLP, 25 mM in
glutamate, and approximately 2 mg DesV was added. Following a
30 min incubation at room temperature, the reaction was judged
>95% complete by analytical HPLC. The mixture was then filtered
to remove protein, and the DesII/DesV product was isolated using
the Dionex CarboPac PA1 semiprep HPLC methodology described
above. The product eluted at 28 min and was confirmed by
reinjections on the analytical HPLC. The isolated material was
lyophilized to dryness and characterized by ESI mass spectrometry,
PA1 column (4 × 250 mm) with a two-stage linear gradient: first
from 2.5% to 10% of 1 M ammonium acetate, pH 7.0 versus ddH2O
over 20 min, followed by a second gradient from 10% to 30% 1 M
ammonium acetate, pH 7.0 versus ddH2O over 20 min (1 mL/min).
The retention times are 18.3 and 37.2 min for 8 and the product,
TDP-3-keto-4,6-dideoxy-D-glucose (9), respectively. The product
peak was collected, and its identity was verified by high-resolution
mass spectrometry analysis.
Determination of Kinetic Parameters for the DesII-Cata-
lyzed Reaction. The steady-state kinetic parameters of DesII-
catalyzed reaction were determined by the activity assay as
described above3a using 3 µM dithionite-reduced DesII, 0.1 mM
SAM, and varied amounts of TDP-4-amino-4,6-dideoxy-D-glucose
(8, 3 µM to 1 mM) in 100 mM Tris·HCl buffer (pH 8.0). The
reaction volume was 0.3 mL, but larger volumes (0.6 mL) were
used for the lower substrate concentrations to facilitate HPLC
analysis. The reaction was incubated at 25 °C, and aliquots were
taken at appropriate time points from each sample and analyzed
by HPLC. Since the initial substrate concentration in each sample
was known, the percent conversion as determined by HPLC could
be used to calculate the amount of product formed for each time
point. The amounts of product formed for a given substrate
concentration at different time points were plotted against time,
and the initial rate was then plotted versus substrate concentration.
The resulting data were fit to the Michaelis-Menten equation by
nonlinear regression using Grafit 5 to determine the kcat and KM
values.
Preparation of TDP-D-quinovose (14) and TDP-D-fucose
(15). TDP-D-quinovose (14) and TDP-D-fucose (15) were prepared
following a literature procedure26 (Scheme 2). First, TDP-4-keto-
6-deoxy-D-glucose (7) was prepared by incubating TDP-D-glucose
(6, 46.2 mg, 8.2 mM) and RfbB (28 µM) in 10 mL of 50 mM
potassium phosphate buffer, pH 7.5, at 37 °C for 1 h. The enzyme
was removed by filtration through an Amicon ultrafiltration unit
(YM10 membrane). To the filtrate was added NaBH4 (10 µM), and
the reaction was allowed to proceed at room temperature for 1 h.
The reaction mixture was desalted on a Sephadex G-10 column
equilibrated with water. After lyophilization, the reaction mixture
was analyzed by HPLC using a Dionex CarboPac PA1 column (4
mm × 250 mm). The flow rate was 1.0 mL/min, and the detector
was set at 267 nm. A linear gradient from 2.5% to 10% 1 M
ammonium acetate, pH 7.0, versus ddH2O over 20 min, followed
by a second linear gradient from 10% to 30% 1 M ammonium
acetate, pH 7.0, versus ddH2O over 20 min gave satisfactory
separation between TDP-D-quinovose (14) and TDP-D-fucose (15)
(the ratio of 14:15 ) 3:1). The retention times are 18.9 and 13.5
min for TDP-D-quinovose (14) and TDP-D-fucose (15), respectively.
Both compounds were collected, lyophilized to dryness, and
confirmed by NMR spectroscopy.
1
1D and 2D H NMR spectroscopy.
Activity Assay Using TDP-D-fucose (15) as the Substrate.
Assays were carried out in 1 mL of 100 mM Tris ·HCl buffer (pH
8, with 2 mM DTT) in the presence of 0.5 mM TDP-D-fucose (15),
0.1 mM dithionite-reduced DesII, and 0.1 mM SAM. The reaction
mixture was incubated at 25 °C for 24 h and was stopped by
removing the enzymes by ultrafiltration through a YM10 membrane.
Analysis of the reaction mixture by HPLC followed the same
protocol as described above for the DesII activity assay using TDP-
D-quinovose (14).
Activity Assay Using TDP-3-amino-3,6-dideoxy-D-glucose
(17) as the Substrate. TDP-3-amino-3,6-dideoxy-D-glucose (17),
which is the product of the TylB reaction,27 was prepared according
to a published procedure28 (Scheme 3). The assay mixture contained
0.5 mM of 17, 50 µM dithionite-reduced DesII, and 50 µM SAM
in 1 mL of 100 mM Tris ·HCl buffer (pH 8 with 2 mM DTT). The
reaction was incubated at 25 °C for 3 h and quenched by removing
the enzyme (YM10 membrane, Amicon). The reaction was
subjected to HPLC analysis as described above using a CarboPac
PA1 column and a two-stage linear gradient: first from 2.5% to
10% of 1 M ammonium acetate, pH 7.0, versus ddH2O over 20
min, followed by a second gradient from 10% to 30% 1 M
ammonium acetate, pH 7.0, versus ddH2O over 20 min (1 mL/
min). The retention times are 18.1 and 37.2 min for 17 and the
resulting product, respectively. The product was verified via HPLC
co-injection with the DesII products from both TDP-4-amino-4,6-
dideoxy-D-glucose (8) and TDP-D-quinovose (14).
Activity Assay Using TDP-D-quinovose (14) as the Sub-
strate. To determine whether TDP-D-quinovose (14) can be
recognized and processed by DesII, 0.5 mM of 14 was incubated
with 0.1 mM dithionite-reduced DesII, 0.1 mM SAM in 1 mL of
100 mM Tris·HCl buffer (pH 8) containing 2 mM DTT. The
reaction mixture was incubated at 25 °C for 3 h, and stopped by
ultrafiltration through a YM10 membrane to remove the enzymes.
The reaction mixture was analyzed by HPLC on a CarboPac PA1
Coupled DesII and DesV Assay Using TDP-3-amino-
3,6-dideoxy-D-glucose (17) as the Substrate. After incubation of
the DesII assay mixture (0.5 mM of 17, 0.1 mM dithionite-reduced
DesII, and 0.1 mM SAM in 1 mL of 100 mM Tris ·HCl buffer, pH
8, with 2 mM DTT) at 25 °C for 3 h, 10 mM of L-glutamate, 0.8
mM of PLP, and 0.1 mM DesV enzyme were added to the assay
(27) Chen, H.; Yeung, S.-M.; Que, N. L. S.; Muller, T.; R., S. R.; Liu,
H.-w. J. Am. Chem. Soc. 1999, 121, 7166–7167.
(28) Melancon, C. E.; Hong, L.; White, J. A.; Liu, Y.-n.; Liu, H.-w.
Biochemistry 2007, 46, 577–590.
(26) Elling, L.; Rupprath, C.; Guenther, N.; Roemer, U.; Verseck, S.;
Weingarten, P.; Draeger, G.; Kirschning, A.; Piepersberg, W. Chem-
BioChem 2005, 6, 1423–1430.
9
14034 J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009

DesII: A Radical SAM Enzyme
A R T I C L E S
Scheme 3
Scheme 4
mixture, and the reaction was incubated at 25 °C for 30 min. The
reaction was stopped by removing the enzymes through a YM10
membrane (Amicon ultrafiltration). Subsequent HPLC analysis was
the same as described above.
deuterium atom from TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose
(18) into 5′-deoxyadenosine. The assay contained 100 µM 18, 700
µM SAM, and 50 µM dithionite-reduced DesII in 100 mM Tris ·HCl
buffer (pH 8.0, with 1 mM DTT). The final volume was 540 µL.
The reaction was initiated by the addition of SAM and reconstituted
DesII. After incubation at 25 °C for 3 h, the reaction was quenched
by ultrafiltration through a YM10 membrane (Amicon) to remove
the enzyme. The reaction mixture was analyzed by HPLC on an
analytical Microsorb-MV C18 column as described above. 5′-
Deoxyadenosine, which was eluted at 10.6 min, was collected,
concentrated, and subjected to analysis by ESI mass spectrometry.
Synthesis of TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose (18).
The overall strategy for making TDP-[3-2H]-4-amino-4,6-dideoxy-
D-glucose (18) is depicted in Scheme 4. The synthesis of [3-2H]-
R-D-glucose (21), starting from 1,2:5,6-di-O-isopropylidene-R-D-
glucofuranose (19), was performed according to previously reported
procedures.29 The conversion of 21 to TDP-[3-2H]-4-amino-4,6-
dideoxy-D-glucose (18) followed the same procedure used to prepare
TDP-4-amino-4,6-dideoxy-D-glucose (8) as described before.5 The
overall yield of the five enzymatic conversions was 15%.
Deuterium Incorporation into S-Adenosyl-L-methionine (SAM).
The incorporation of deuterium from TDP-[3-2H]-4-amino-4,6-
dideoxy-D-glucose (18) into SAM was determined following a
literature procedure14b,16a,30 with some modifications. Briefly, the
assay contained 500 µM 18, 100 µM SAM, and 100 µM dithionite-
reduced DesII in 1 mL of 100 mM Tris·HCl buffer (pH 8.0, with
1 mM DTT). The reaction was initiated by the addition of SAM
and reconstituted DesII after incubation of the other components
of the assay mixture at 25 °C for 5 min. After incubation at 25 °C
for 3 h, the reaction was quenched by ultrafiltration through a YM10
membrane (Amicon) to remove the enzyme. The reaction mixture
was analyzed by HPLC on an analytical Microsorb-MV C18 column
(4.5 × 250 mm, 5 µm, Varian) with the detector set at 260 nm.
The column was pre-equilibrated with 0.1% trifluoroacetic acid
(TFA) in ddH2O and subsequently eluted with a 20-min linear
gradient of 0 to 45% acetonitrile containing 0.1% TFA at a flow
rate of 1 mL/min. S-adenosylmethionine (SAM), which was eluted
at 5.1 min, was collected, concentrated, and analyzed by ESI mass
spectrometry.
Results
Purification and Characterization of DesII Protein. The
C-terminal His6-tagged DesII was purified aerobically to near
homogeneity using a Ni-NTA column followed by FPLC on a
MonoQ column. The desired fractions were selected based on
SDS-PAGE results (Figure 1a). A 6-L culture yielded 15 mg
of homogeneous DesII protein, which is slightly brown in color.
The N-terminal peptide sequencing analysis confirmed that the
first 10 amino acid residues of the purified DesII are identical
to those of the translated DesII sequence except for the initiating
methionine residue, which must have been removed during
protein synthesis in E. coli. A molecular mass of 48 kDa for
DesII as estimated by gel filtration chromatography suggests
that His-tag-DesII, which has a calculated molecular mass of
54,265 Da (including the His6 tag), is a monomer in solution.
It is not clear that the observed monomeric state is due to the
presence of the His-tag.
UV-vis Spectral Analysis of Purified DesII. The UV-vis
spectrum of the purified protein exhibits a broad shoulder
between 400 and 500 nm (Figure 1b), which is characteristic
for iron-sulfur cluster containing proteins, such as the anaerobic
E. coli ribonucleotide reductase,31 spore photoproduct lyase,15d
BtrN in the biosynthesis of butirosin,16a and ThiC in thiamine
Deuterium Incorporation into 5′-Deoxyadenosine. A similar
procedure was followed to determine the incorporation of the
(29) Pieper, P. A.; Guo, Z.; Liu, H.-w. J. Am. Chem. Soc. 1995, 117, 5158–
5159.
(30) (a) Escalettes, F.; Florentin, D.; Tse Sum Bui, B.; Lesage, D.; Marquet,
A. J. Am. Chem. Soc. 1999, 121, 3571–3578. (b) Frey, M.; Rothe,
M.; Wagner, A. F. V.; Knappe, J. J. Biol. Chem. 1994, 269, 12432–
12437. (c) Moss, M.; Frey, P. A. J. Biol. Chem. 1987, 262, 14859–
14862.
(31) Ollangier, S.; Mulliez, E.; Gaillard, J.; Eliasson, R.; Fontecave, M.;
Reichard, P. J. Biol. Chem. 1996, 271, 9410–9416.
9
J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009 14035

A R T I C L E S
Szu et al.
Figure 1. (a) SDS-PAGE of purified DesII isolated from E. coli BL21
Star (DE3): lane 1, the molecular mass standards: MBP-ꢀ-galactosidase
(175 kDa), MBP-paramyosin (83 kDa), glutamic dehydrogenase (62 kDa),
aldolase (47.5 kDa), triosephosphate isomerase (32.5 kDa), ꢀ-lactoglobulin
A (25 kDa), lysozyme (16.5 kDa), and aprotinin (6.5 kDa), lanes 2-4,
fractions of DesII purified using a Ni-NTA column followed by FPLC on
a MonoQ HR (16/10) column. (b) UV-vis absorption spectrum of
aerobically purified DesII in 50 mM Tris ·HCl buffer, pH 7.5. The protein
concentration is 0.35 mM.
Figure 2. UV-vis absorption spectra of apo-DesII (solid line), reconstituted
DesII before (dotted line) and after (dashed line) reduction with 1.14 mM
sodium dithionite. The protein concentration is 0.19 mM in 100 mM
Tris·HCl buffer, pH 8.0.
pyrimidine biosynthesis.17a The extinction coefficients deter-
mined are ε420 ) 4000 M-1 ·cm-1 for DesII and ε420 ) 3700
M-1 ·cm-1 for the anaerobic E. coli ribonucleotide reductase.
Since the [4Fe-4S] cluster is oxygen sensitive and is known to
degrade to [3Fe-4S]1+ and/or [2Fe-2S]2+ clusters upon exposure
to oxygen,32 the UV-vis spectrum of the as-isolated DesII is
not likely due to an intact [4Fe-4S]2+ cluster.
diamine monohydrochloride to give methylene blue, which
could be detected at 670 nm. It was determined that the as-
isolated DesII contained approximately 0.5 sulfur equivalents
per monomer of protein, but the fully reconstituted DesII
contained nearly 4 sulfur equivalents per monomer. These data
are consistent with reconstituted DesII containing a [4Fe-4S]
cluster as suggested by its UV-vis spectrum (Figure 2).
Characterization of the [3Fe-4S]1+ Cluster in the Aero-
bically Purified DesII by EPR Spectroscopy. The aerobically
purified DesII exhibits a strong EPR signal at 10 K with a g
value centered at 2.010 (Figure 3a). This EPR signal is
characteristic for a [3Fe-4S]1+ cluster (spin state (S) ) 1/2),
which has been observed in several aerobically purified radical
SAM enzymes. For example, the aerobically purified spore
photoproduct lyase exhibits a similar EPR signal arising from
a [3Fe-4S]1+ cluster with a g value centered at 2.02.15d Likewise,
the anaerobic ribonucleotide reductase activase exhibits an EPR
signal for a [3Fe-4S]1+ cluster with a g value centered at 2.017
upon exposure to air.34 The [3Fe-4S]1+ cluster is thought to
result from air oxidation of the [4Fe-4S]3+ cluster and renders
the parent enzymes catalytically inactive. However, the majority
of aerobically purified DesII appears to be in the apoprotein
form with less than 10-20% of the monomers containing
residual iron as part of a [3Fe-4S]1+ cluster. This is based on
the observation of approximately 0.6 iron equivalents and 0.5
labile sulfur equivalents per protein monomer. Interestingly, no
EPR signal was observed for the as-isolated DesII protein in
the presence of dithionite recorded under the same conditions.
This observation suggests that its iron-sulfur cluster is in the
EPR-silent [3Fe-4S]0 form.
Reconstitution and Reduction of the Iron Sulfur Center of
DesII. The purified His-tag-DesII was anaerobically converted
to the colorless apoprotein by incubating with EDTA in the
presence of 2 mM sodium dithionite. Subsequent reconstitution
of the iron-sulfur cluster was achieved by incubating the apo-
DesII anaerobically with an excess of ferrous iron and sodium
sulfide in the presence of DTT for 6 h at 25 °C. After
reconstitution, the color of DesII changed to green-gray. The
reconstituted DesII exhibits featureless absorption across the
entire UV-vis range with a broad absorption hump at 420 nm
(ε420 ) 9200 M-1 ·cm-1) (Figure 2), which is typical for a [4Fe-
4S]2+ cluster. Similar spectra have also been observed for other
[4Fe-4S]2+ containing enzymes.7a,32 The catalytically active
form, [4Fe-4S]1+, of the iron-sulfur cluster was generated by
anaerobic reduction of the reconstituted DesII with sodium
dithionite. The reduction was monitored by following the
decrease in absorbance at 420 nm, which is associated with the
reduction of a [4Fe-4S]2+ cluster to the [4Fe-4S]1+ state33
(Figure 2).
The Iron and Sulfur Content in DesII. The iron titration
results showed that the as-isolated His-tag-DesII contained
approximately 0.6 iron equivalents per monomer of protein,
which clearly indicated iron depletion during aerobic purifica-
tion. In contrast, the fully reconstituted His-tag-DesII contained
approximately 4 iron equivalents per monomer. The labile sulfur
content of DesII was determined by quantitating the released
sulfide after protein denaturation using guanidinium hydrochlo-
ride in the presence of dithiothreitol.25 The released inorganic
sulfide was first absorbed onto zinc as zinc sulfide, and then
freed by acid treatment to react with N,N-dimethyl-p-phenylene-
Characterization of the [4Fe-4S]1+ Cluster in the Anaero-
bically Reconstituted DesII by EPR Spectroscopy. The recon-
stituted DesII, reduced by dithionite in the presence of SAM,
exhibits a rhombic EPR signal with g values at 2.01, 1.96, and
1.87 (gav ) 1.95) (Figure 3b), which is similar to that of the
[4Fe-4S]1+ cluster in the anaerobic ribonucleotide reductase
activase (ARR activase) when reduced by dithionite in the
(32) (a) Ugulava, N. B.; Gibney, B. R.; Jarrett, J. T. Biochemistry 2000,
39, 5206–5214. (b) Ollagnier, S.; Meier, C.; Mulliez, E.; Gaillard, J.;
Schuenemann, V.; Trautwein, A.; Mattioli, T.; Lutz, M.; Fontecave,
M. J. Am. Chem. Soc. 1999, 121, 6344–6350.
presence of SAM with g values at 2.00, 1.92, and 1.86 (gav
)
(33) Duin, E. C.; Lafferty, M. E.; Crouse, B. R.; Allen, R. M.; Sanyal, I.;
Flint, D. H.; Johnson, M. K. Biochemistry 1997, 36, 11811–11820.
(34) Torrents, E.; Buist, G.; Liu, A.; Eliasson, R.; Kok, J.; Gibert, I.;
Graslund, A.; Reichard, P. J. Biol. Chem. 2000, 275, 2463–2471.
9
14036 J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009

DesII: A Radical SAM Enzyme
A R T I C L E S
with TDP-4-amino-4,6-dideoxy-D-glucose (8) in the presence
of SAM, a product was formed. The retention time of 28.2 min
for this product (9) under the HPLC conditions is identical to
that of the chemoenzymatically synthesized standard. The high-
resolution fast atom bombardment (FAB) MS data (calcd for
C16H24N2O14P2 [M-H]- 529.0625, found 529.0620) is consistent
with the composition of the assigned product (9). To gain
additional evidence for the formation of 9 in the DesII reaction,
the above reaction mixture was incubated with DesV, PLP and
L-glutamate to convert 9 to TDP-3-amino-3,4,6-trideoxy-D-
glucose (10). A new HPLC peak with a retention time of 3.7
min was observed. This new product coeluted with a standard
of 10 (Figure 4a). The identity of 10 was further validated by
high-resolution FAB-MS analysis (calcd for C16H27N3O13P2
[M-H]- 530.0941, found 530.0949). These results unequivocally
established that the product of DesII reaction is TDP-3-keto-
4,6-dideoxy-D-glucose (9).
Determination of the Kinetic Parameters for DesII-
Catalyzed Reaction. The discontinuous HPLC assay was used
to determine the steady state kinetic parameters for His-tag-
DesII-catalyzed conversion of 8 to 9. The reconstituted enzyme
(3 µM) was activated by dithionite reduction of its [4Fe-4S]2+
center to [4Fe-4S]+. The plot of V0 (the initial velocity) versus
[S] (3 µM to 1 mM) (Figure 5) was fit to the Michaelis-Menten
equation by nonlinear regression to yield a kcat of 1.0 ( 0.1
min-1 and a Km of 50 ( 2 µM for 8. These parameters
correspond to the TDP-4-amino-4,6-dideoxy-D-glucose substrate
in the presence of a constant concentration of 0.1 mM SAM.
Purification and Characterization of E. coli Flavodoxin
(FLD) and Flavodoxin Reductase (FPR). The C-terminal His6-
tagged flavodoxin (FLD) and flavodoxin reductase (FPR) were
purified to near homogeneity using Ni-NTA affinity chroma-
tography. A 6-L culture yielded 90-100 mg of the desired
proteins. Judging by SDS-PAGE, the molecular masses of FLD
and FPR were estimated to be 22 kDa and 29 kDa, respectively,
which correlates well with the corresponding calculated mo-
lecular mass of 22,765 Da and 29,765 Da (including the His6
tag). The UV-vis spectrum of the purified FPR exhibits peaks
at 360 and 460 nm as well as a shoulder near 500 nm, revealing
the presence of a flavin mononucleotide (FMN) cofactor. The
observed UV-vis spectrum of FPR, which exhibits absorption
maxima at 400 and 465 nm, and a shoulder at 480 nm, is
characteristic for a flavin adenine dinucleotide (FAD) in its
active-site.
Reduction of DesII Mediated by Flavodoxin (FLD) and
Flavodoxin Reductase (FPR). Similar to the results of using
dithionite-reduced DesII in the assay, anaerobic incubation of
substrate (8), the reconstituted DesII, SAM, NADPH, fla-
vodoxin, and flavodoxin reductase also led to the production
of TDP-3-keto-4,6-dideoxy-D-glucose (9). In this experiment,
the [4Fe-4S]2+ of DesII is reduced to [4Fe-4S]1+ by stepwise
electron transfer from NADPH through flavodoxin and fla-
vodoxin reductase. Accordingly, a system consisting of fla-
vodoxin/flavodoxin reductase-NADPH or its equivalent from
the cellular pool may serve as the in ViVo reducing system for
the DesII-catalyzed reaction.
Figure 3. (a) EPR spectrum of as-purified DesII containing a [3Fe-4S]1+
cluster with a g value centered at 2.010. The protein concentration is 0.2
mM in 100 mM Tris ·HCl buffer, pH 8.0. The spectrum was recoded at 10
K with a modulation amplitude of 10 G, a microwave power of 1 mW, and
a microwave frequency of 9.60 GHz. (b) EPR spectrum of reconstituted
DesII containing a [4Fe-4S]1+ cluster with g values at 2.01, 1.96, and 1.87
(gav ) 1.95). The protein concentration is 0.2 mM in 100 mM Tris ·HCl
buffer, pH 8.0. The spectrum was recoded at 10 K with a modulation
amplitude of 10 G, a microwave power of 10 mW, and a microwave
frequency of 9.60 GHz.
1.93).12c Unlike ARR activase whose EPR signal is more axial
than rhombic in the absence of SAM, DesII is EPR silent in
the absence of SAM. It is likely that the [4Fe-4S]2+/1+ cluster
of DesII is unstable without SAM binding to its open iron site
and can easily lose one iron to become a [3Fe-4S]1+ cluster.
The one-electron reduced form of the latter cluster, [3Fe-4S]0,
is EPR-silent. Another possibility is that the [4Fe-4S]2+ cluster
of DesII is more difficult to reduce in the absence of SAM.
Precedence for such a phenomenon has been set by lysine 2,3-
aminomutase, where SAM binding elevates the reduction
potential of its [4Fe-4S]2+ cluster by 50-86 mV.35 No EPR
signal was observed in the reconstituted DesII protein in the
absence of dithionite recorded under the same conditions,
suggesting the iron-sulfur cluster is in its EPR-silent [4Fe-
4S]2+ state.
Characterization of TDP-D-quinovose (14) and TDP-D-
fucose (15). These two compounds were synthesized from TDP-
4-keto-6-deoxy-D-glucose (7) by reducing the C-4 keto group
using NaBH4 (Scheme 2). The yields of 14 and 15 were 40%
and 13%, respectively. The spectral data for TDP-D-quinovose
Activity Assay of the DesII-Catalyzed Reaction. When the
fully reconstituted and dithionite reduced DesII was incubated
1
(14): H NMR (300 MHz, D2O) δ 0.80 (3H, d, J ) 5.7 Hz,
(35) Hinckley, G. T.; Frey, P. A. Biochemistry 2006, 45, 3219–3225.
5-Me), 1.52 (3H, s, 5′′-Me), 1.98-2.08 (2H, m, 2′-Hs), 2.84
9
J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009 14037

A R T I C L E S
Szu et al.
Figure 4. (a) HPLC traces demonstrating DesII activity using dithionite as the reducing agent. (b) HPLC traces demonstrating DesII substrate specificity
using TDP-D-quinovose (14) and TDP-D-fucose (15) in the incubation. (c) HPLC traces demonstrating DesII substrate specificity using TylB product (17)
in the incubation. See Experimental Procedures for details.
(1H, dd, J3,4 ) J4,5 ) 9.9 Hz, 4-H), 3.22-3.31 (1H, m, 2-H),
3.28 (1H, dd, J2,3 ) 9.3, J3,4 ) 9.9 Hz, 3-H), 3.53 (1H, m, 5-H),
3.75-3.88 (3H, m, 4′-H, 5′-Hs), 4.18-4.22 (1H, m, 3′-H), 5.15
(1H, dd, J1,p ) 6.8, J1,2 ) 3.1 Hz, 1-H), 5.94 (1H, t, J ) 6.7
Hz, 1′-H), 7.29 (1H, s, 6′′-H); 31P NMR (121 MHz, D2O) δ
-11.79 (d, J ) 18.3 Hz), -10.34 (d, J ) 18.3 Hz). The spectral
data for TDP-D-fucose (15): 1H NMR (300 MHz, D2O) δ 1.21
(3H, d, J ) 6.6 Hz, 5-Me), 1.93 (3H, s, 5”-Me), 2.36 (2H, m,
2′-Hs), 3.74 (1H, dt, J ) 10.5, 3.2 Hz, 2-H), 3.81 (1H, J ) 3.2
Hz, 4-H), 3.91 (1H, dd, J ) 10.5, 3.2 Hz, 3-H), 4.17 (3H, m,
4′-H and 5′-Hs), 4.28 (1H, J ) 6.6 Hz, 5-H), 4.62 (1H, m, 3′-
H), 5.56 (1H, dd, J ) 6.8, 3.6 Hz, 1-H), 6.34 (1H, t, J ) 7.0
Hz, 1′-H), 7.74 (1H, s, 6′′-H); 31P NMR (121 MHz, D2O) δ
-11.75 (d, J ) 18.3 Hz), -10.39 (d, J ) 18.3 Hz). These
spectral data are in agreement with those reported in the
literature.26
Testing the Competence of TDP-D-quinovose (14), TDP-D-
fucose (15) and TDP-3-amino-3,6-dideoxy-D-glucose (17) as
Substrates for DesII. Compounds 14, 15, and 17 were used to
assess the ability of DesII to accept different substrate analogues
(HPLC traces in b and c of Figure 4). Anaerobic incubation of
TDP-D-quinovose (14) with SAM and the reconstituted and
reduced DesII led to the formation of TDP-3-keto-6-deoxy-D-
glucose (16). Assignment is based on ESI MS. In the negative
ion mode, peaks identified at 544.93 and 567.13 m/z are
consistent with the (M - H+)- and (MNa+ - 2H+)- ions of
9
14038 J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009

DesII: A Radical SAM Enzyme
A R T I C L E S
distinct from 9 upon corresponding HPLC co-injections. When
TDP-D-fucose (15) was tested, no turnover was observed (Figure
4b).
Synthesis of TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose (18).
Compound 18 was prepared from 1,2:5,6-di-O-isopropylidene-
ꢀ-D-glucofuranose (19) as shown in Scheme 4. The deuterium
was incorporated into 1,2:5,6-di-O-isopropylidene-ꢀ-D-ribo-
hexofuranos-3-ulose (20) by the reduction of the C3-keto group
using NaBD4. Subsequent inversion of the C-3 configuration
followed by acid hydrolysis gave [3-2H]-glucose (21). The con-
version of 21 to 18 was carried out enzymatically, where
hexokinase, phosphoglucomutase, RfbA, RfbB, and DesI were
employed to afford TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose
(18).3a Spectral data of 18: H NMR (500 MHz, D2O) δ 0.80
1
(3H, d, J ) 5.7 Hz, 5-Me), 1.52 (3H, s, 5′-Me), 1.98-2.03 (2H,
m, 2′-Hs), 2.84 (1H, d, J ) 9.9 Hz, 4-H), 3.28 (1H, d, J ) 3.6
Hz, 2-H), 3.53 (1H, m, 5-H), 3.75-3.88 (3H, m, 4′-H, 5′-Hs),
4.18-4.22 (1H, m, 3′-H), 5.48 (1H, dd, J ) 6.8, 3.1 Hz, 1-H),
5.94 (1H, t, J ) 6.7 Hz, 1′-H), 7.29 (1H, s, 6′-H); 13C NMR
(125 MHz, D2O) δ 12.8, 17.6, 39.7, 55.9, 66.6 (d, J ) 6.0 Hz),
70.1 (t, J ) 20 Hz), 70.6 (d, J ) 8.4 Hz), 72.1, 73.3, 86.2, 86.4
(d, J ) 9.2 Hz), 95.6 (d, J ) 6.1 Hz), 112.8, 138.5, 153.0,
167.9; 31P NMR (202 MHz, D2O) δ -13.5 (d, J ) 20.7 Hz),
-11.8 (d, J ) 20.7 Hz); ESI-HRMS calcd for C16H25DN3O14P2
[M - H]- 547.0958, found 547.0951.
Deuterium Incorporation into S-Adenosyl-L-methionine
(SAM) from TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose (18). In
a similar manner, the reaction mixture obtained after anaerobic
incubation of the deuterium-labeled substrate (18), was analyzed
by HPLC on an analytical C18 column. SAM, which was eluted
at 5.1 min, was collected, concentrated, and subjected to ESI-
MS analysis (ESI-HRMS calcd for C15H212H2N6O5S [M + H]+
401.1571, found 401.1562). A control using the nonlabeled
substrate was also carried out in parallel (ESI-MS analysis of
the isolated SAM: ESI-HRMS calcd for C15H23N6O5S [M +
H]+ 399.1450, found 399.1452). The increase of two mass units
of SAM derived from the labeled sample clearly indicated the
incorporation of two deuterium atoms.
Deuterium Incorporation into 5′-Deoxyadenosine from
TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose (18). After anaerobic
incubation of the deuterium-labeled substrate (18), the reaction
mixture was analyzed by HPLC on an analytical C18 column.
The 5′-deoxyadenosine, which was eluted at 10.6 min, was
collected, concentrated, and subjected to ESI-MS analysis (ESI-
HRMS calcd for C10H132HN5O3 [M + H]+ 253.1152, found
253.1134). A control using the nonlabeled substrate was carried
out in parallel (ESI-MS analysis of the isolated 5′-deoxyad-
enosine: ESI-HRMS calcd for C10H14N5O3 [M + H]+ 252.1132,
found 252.1091). The increase of one mass unit of 5′-
deoxyadenosine derived from the labeled sample revealed the
incorporation of one deuterium atom.
Figure 5. Plot of Vo versus [S] (8) determined by an HPLC assay from
which the steady state kinetic constants for the DesII-catalyzed reaction
were determined. See Experimental Procedures for details.
16, which has a neutral monoisotopic mass of 546. In the
positive ion mode, peaks at 568.87, 590.80, and 612.87 m/z are
consistent with the MNa+, MNa2+, and MNa3 ions of 16. In
+
both spectra, no peaks were observed corresponding to ions of
either 9, neutral mass 530, or its hydrate, neutral mass 548.
Consistent with this assignment, the DesII product from 14 is
indistinguishable by analytical HPLC from the Tyl1a product
(16, see Scheme 3) where co-injection yielded an inseparable
composite peak but was distinctly separable on co-injection with
9. Unfortunately, product 16 itself is relatively unstable, and
the 1H NMR spectra obtained from the isolated material
contained a number of contaminating species, making peak
assignments difficult.
Subsequent reaction of DesV with the DesII product (16) from
TDP-D-quinovose yielded TDP-3-amino-3,6-dideoxy-D-glucose
(17), consistent with the known ability of DesV to replace TylB
in ViVo.21b Positive ion ESI MS of the isolated material
demonstrated peaks at 547.93, 569.80, and 591.67 m/z consistent
with the MH+, MNa+, and MNa2 ions of 17, which has a
+
neutral monoisotopic mass of 547. No peaks corresponding to
ions of 10 where observed (neutral mass 531). In the negative
ion ESI MS a signal at 546.13 is present, consistent with the
(M - H+)- anion of 17, while no peaks corresponding to ions
of 10 were detected.
In contrast to the DesII product from 14, the coupled DesII/
DesV product (17) is significantly more stable such that a much
1
cleaner H NMR spectrum was obtained. This spectrum is
1
consistent with the assignment of structure 17: H NMR (600
MHz, D2O) δ 1.18 (3H, d, J ) 6.3 Hz, 5-Me), 1.82 (3H, s,
5′′-Me), 2.26 (2H, m, 2′-Hs), 3.25 (1H, dd, J4,5 ) J4,3 ) 9.6
Hz, 4-H), 3.33 (1H, dd, J3,4 ) J3,2 ) 10.5 Hz, 3-H), 3.68 (1H,
ddd, J2,1 ) J2,P ) 3.6 Hz, J2,3 ) 10.2 Hz, 2-H), 3.91 (1H, dq,
J5,4 ) 9.6 Hz, J5,6 ) 6.6 Hz, 5-H), 4.07 (3H, m, 4′-H and 5′-
Hs), 4.51 (1H, m, 3′-H), 5.48 (1H, dd, J1,2 ) 3.6 Hz, J1,P ) 7.2
Hz, 1-H), 6.23 (1H, dd, J1′,2′R ) J1′,2′S ) 6.6 Hz, 1′-H), 7.62
(1H, s, 6′′-H). Cross-coupling assignments were confirmed by
1H COSY NMR. On analytical HPLC, co-injection of the DesII/
DesV product from 14 is indistinguishable from the TylB
product (17) but is separable from 10.
Discussion
At the time when the function of DesII from S. Venezuelae
was established in 2005,3a DesII was one of the first radical
SAM enzymes involved in secondary metabolite biosynthesis
to be characterized at the enzymatic level. Previous biochemical
studies had shown that DesII together with DesI carries out the
C-4 deoxygenation of TDP-4-keto-6-deoxy-D-glucose (7) to give
TDP-3-keto-4,6-dideoxy-D-glucose (9) as the product. The
function of DesII is to convert the TDP-4-amino-4,6-dideoxy-
D-glucose (8) intermediate generated by DesI to the final product
9.
Interestingly, TDP-3-amino-3,6-dideoxy-D-glucose (17) is
itself a substrate for DesII (Figure 4c). The product was
identified as TDP-3-keto-6-deoxy-D-glucose (16) as it is indis-
tinguishable from the DesII product from 14, although clearly
9
J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009 14039

A R T I C L E S
Szu et al.
the reaction facilitated by the reduced [4Fe-4S]1+ cluster as
found in other radical SAM-dependent enzymes.6 The chemical
conversion is triggered by hydrogen atom abstraction from 8
by 22. Abstraction likely occurs at C-3 to give 23 as an
intermediate. The mechanism for the subsequent transformation
is less obvious, but may parallel the reaction catalyzed by the
coenzyme B12-dependent ethanolamine ammonia-lyase, which
converts ethanolamine to ammonia and acetaldehyde.41 As
shown in Scheme 6 path A, the key step may be a radical-
induced 1,2-amino shift (23 f 24/25 f 26), to form an aminol
radical 26. Reclaiming a hydrogen atom from 5′-deoxyadenosine
(27) would then result in the formation of 28 and regeneration
of 22, which could recombine with methionine to reform the
[4Fe-4S]1+-SAM complex.9,10a Elimination of an ammonium
ion from 28 would afford the final product 9. The reaction may
also proceed by deprotonation of the 3-hydroxyl group of 23
to yield a ketyl radical anion 29, whose resonance form 30
facilitates the ꢀ-elimination of the 4-amino group (Scheme 6,
path B). The reaction catalyzed by (R)-2-hydroxyacyl-CoA
dehydratase42 in the fermentation of ꢀ-amino acids by anaerobic
bacteria provides such a precedent.
Scheme 5
To investigate the proposed mechanisms, we first considered
hydrogen atom abstraction by the 5′-deoxyadenosyl radical (22).
When TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose (18) was
used as substrate, deuterium incorporation into SAM and 5′-
deoxyadenosine were observed. This provides strong evidence
for cleavage of the C-H bond at C-3 (8 f 23) during turnover.
The mechanistic implications of these findings are at least 2-fold.
First, they provide compelling evidence supporting the inter-
mediacy of a sugar C-3 radical (23) in DesII catalysis. Second,
they support the hypothesis that SAM serves catalytically as a
cofactor that is regenerated with each turnover of the TDP-
sugar substrate. As depicted in Scheme 6, the outcome is
consistent with the formation of [5′-2H]-5′-deoxyadenosine ([5′-
Although chemical reductants such as dithionite and 5-dea-
zaflavin are commonly used to reduce the [4Fe-4S]2+ cluster in
enzymes, flavodoxin, flavodoxin reductase, and NADPH have
been demonstrated to be an effective reducing system for several
radical SAM enzymes.9 The reducing equivalents are transferred
from NADPH to the [4Fe-4S]2+ cluster by relay through
flavodoxin and flavodoxin reductase (Scheme 5). Not surpris-
ingly, similar results were also noted for DesII catalysis. The
in ViVo effectiveness of this reducing system for DesII activity
has already been demonstrated by studies of TDP-D-desosamine
biosynthesis in a 6-deoxyerythromycin D-producing recombi-
nant E. coli strain, where coexpression of flavodoxin and
flavodoxin reductase from E. coli increased DesI/DesII activ-
ity.36
A kcat of 1.0 ( 0.1 min-1 and a Km of 50 ( 2 µM with respect
to the sugar substrate 8 were determined for the DesII-catalyzed
reaction with SAM constant at 0.1 mM. Kinetic parameters are
also known for lysine 2,3-aminomutase,37 pyruvate formate-
lyase activase,38 anaerobic ribonucleotide reductase activase,39
biotin synthase,40 lipoyl synthase,14a and BtrN,16a with the kcat
values ranging from 27 s-1 (lysine 2,3-aminomutase) to 0.175
min-1 (lipoyl synthase), and the Km values ranging from 1.4
µM (pyruvate formate lyase-activase) to 8.0 mM (lysine 2,3-
aminomutase). Clearly, radical SAM enzymes demonstrate a
large range of turnover rates, and the reaction catalyzed by DesII
appears to be more sluggish than many others. However, it
cannot be ruled out that the relatively low rate of DesII turnover
may be due to the presence of the C-terminal His6 tag used in
this study.
2
2H]-27) as a result of H-atom abstraction from 18 by the 5′-
deoxyadenosyl radical (22). Owing to the primary deuterium
isotope effect, a H-atom rather than the 2H-atom from the methyl
group of [5′-2H]-27 will be preferentially reclaimed by the
aminol radical intermediate (26) in path A or the enol radical
intermediate (31/32) in path B to yield a [5′-2H]-5′-deoxyad-
enosyl radical ([5′-2H]-22), from which the [4Fe-4S]1+-[5′-
2H]SAM complex reforms to complete the catalytic cycle. A
second-round of turnover with another molecule of deuterated
substrate 18 would then lead to [5′-2H2]SAM as was detected
by mass spectrometry. The precedence is set by the reactions
of lysine 2,3-aminomutase9 and spore photoproduct lyase,15 in
which SAM functions catalytically as a temporary oxidant and
is regenerated at the end of each catalytic cycle.
However, these results do not rule out stoichiometric turnover
of SAM suggested by the observation of the formation of [5′-
2H]-5′-deoxyadenosine when 18 was used in the reaction. Under
this scenario, the consumption of one molecule of SAM should
be accompanied by the production of one molecule of [5′-2H]-
5′-deoxyadenosine after each catalytic cycle. The formation of
Two possible mechanisms have been proposed for the DesII
reaction.3a As depicted in Scheme 6, generation of a 5′-
deoxyadenosyl radical (22) is expected to be the first part of
(41) (a) Banerjee, R. Chemistry and Biochemistry of B12; John Wiley &
Sons, Inc.: New York, 1999. (b) LoBrutto, R.; Bandarian, V.;
Magnusson, O. T.; Chen, X.; Schramm, V. L.; Reed, G. H. Biochem-
istry 2001, 40, 9–14. (c) Sandala, G. M.; Smith, D. M.; Radom, L.
J. Am. Chem. Soc. 2005, 127, 8856–8864.
(36) Lee, H. Y.; Khosla, C. PLoS Biology 2007, 5, 243–250.
(37) Wu, W.; Lieder, K. W.; Reed, G. H.; Frey, P. A. Biochemistry 1995,
34, 10532–10537.
(38) Wagner, A. F.; Demand, J.; Schilling, G.; Pils, T.; Knappe, J. Biochem.
Biophys. Res. Commun. 1999, 254, 306–310.
(39) Eliasson, R.; Pontis, E.; Sun, X.; Reichard, P. J. Biol. Chem. 1994,
269, 26052–26057.
(42) (a) Buckel, W. FEBS Lett. 1996, 389, 20–24. (b) Buckel, W. Appl.
Microbiol. Biotechnol. 2001, 57, 263–273. (c) Buckel, W.; Golding,
B. T. FEMS Microbiol. ReV. 1998, 22, 523–541. (d) Kim, J.; Hetzel,
M.; Boiangiu, C. D.; Buckel, W. FEMS Microbiol. ReV. 2004, 28,
455–468.
(40) Picciocchi, A.; Douce, R.; Alban, C. Plant Physiol. 2001, 127, 1224–
1233.
9
14040 J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009

DesII: A Radical SAM Enzyme
A R T I C L E S
Scheme 6
Scheme 7
[5′-2H2]SAM in this case may be ascribed to the reversible
formation of the substrate radical (23) via direct H-atom
interchange with 5′-deoxyadenosine (27). As depicted in Scheme
7, H-atom return from [5′-2H]-27 to the substrate radical (23)
could eventually lead to a Michaelis complex with [5′-2H2]SAM,
which will then be released. Evidence for such reversible H-atom
interchange between 5′-deoxyadenosine and a substrate radical
has been gathered for the radical SAM dehydrogenase BtrN.16
Overall, a catalytic role for SAM in DesII-catalyzed reaction is
attractive and consistent with the observation of [5′-2H2]SAM
in the reaction with TDP-[3-2H]-4-amino-4,6-dideoxy-D-glucose.
However, the possibility that SAM may not be used in a catalytic
fashion is implicated by the observation of monodeuterated 5′-
deoxyadenosine produced in the DesII reaction and the near
unit concentration ratios of substrates to enzyme employed in
the present studies. Efforts are currently underway in our
laboratory to resolve these mechanistic ambiguities.
In exploring the substrate specificity of DesII, it was
discovered that DesII catalyzes oxidation of TDP-D-quinovose
(14) and TDP-3-amino-3,6-dideoxy-D-glucose (17) to form in
9
J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009 14041

A R T I C L E S
Szu et al.
both cases TDP-3-keto-6-deoxy-D-glucose (16). The latter
substrate is presumably first dehydrogenated to a Schiff base
that subsequently hydrolyzes to give the ketone. An analogous
reaction is reported for BtrN, which catalyzes the oxidation of
the C-3 hydroxyl group of 2-deoxy-scyllo-inosamine to yield
the corresponding C-3 ketone.16 In contrast, TDP-D-fucose (15),
which contains an axial hydroxyl group at C-4, is not a substrate
for DesII. Therefore, the stereochemical orientation of the
substituent at the C-4 position appears to be important for
substrate recognition. These results demonstrate that DesII
tolerates variation in substituents at both the C-3 and C-4
positions; however, while an amino group at C-4 leads to a redox
neutral deamination reaction, the presence of a hydroxyl group
at C-4 leads to dehydrogenation at C-3.
superfamily. Although the details of its catalytic mechanism
remain elusive, participation of DesII with the transaminase DesI
provides another unique method by which C-O cleavage can
proceed in biological systems. Mechanistic studies of this
intriguing enzyme are in progress.
Acknowledgment. This work was supported in part by the
National Institutes of Health Grants (GM35906, 54346). We thank
Professor Ah-Lim Tsai at the University of Texas at Houston
Medical School for his assistance in acquiring the EPR spectra.
Construction of the expression plasmids for flavodoxin and fla-
vodoxin reductase was assisted by Dr. Allen Yu. Dr. Xiaotao Pu
and Dr. Steven O. Mansoorabadi are acknowledged for their
valuable suggestions on the synthesis of the deuterium-labeled
substrate and EPR studies, respectively. The Tyl1a and TylB
enzymes were generously provided by Dr. Charles E. Melanc¸on.
We also thank Dr. Steven Mansoorabadi for his critical comments
on the manuscript.
In conclusion, DesII has been fully established as a radical
SAM-dependent deaminase involved in C-4 deoxygenation in
TDP-D-desosamine biosynthesis. The fact that DesII is respon-
sible for a radical-mediated deamination reaction further il-
lustrates the catalytic versatility of the radical SAM enzyme
JA903354K
9
14042 J. AM. CHEM. SOC. VOL. 131, NO. 39, 2009