Fluorescent probes for investigation of isoprenoid configuration and size discrimination by bactoprenol-utilizing enzymes

Article abstract of DOI:10.1016/j.bmc.2013.06.007

Undecaprenyl Pyrophosphate Synthase (UPPS) is an enzyme critical to the production of complex polysaccharides in bacteria, as it produces the crucial bactoprenol scaffold on which these materials are assembled. Methods to characterize the systems associated with polysaccharide production are non-trivial, in part due to the lack of chemical tools to investigate their assembly. In this report, we develop a new fluorescent tool using UPPS to incorporate a powerful fluorescent anthranilamide moiety into bactoprenol. The activity of this analogue in polysaccharide biosynthesis is then tested with the initiating hexose-1-phosphate transferases involved in Capsular Polysaccharide A biosynthesis in the symbiont Bacteroides fragilis and the asparagine-linked glycosylation system of the pathogenic Campylobacter jejuni. In addition, it is shown that the UPPS used to make this probe is not specific for E-configured isoprenoid substrates and that elongation by UPPS is required for activity with the downstream enzymes.

Full text of DOI:10.1016/j.bmc.2013.06.007

Bioorganic & Medicinal Chemistry 21 (2013) 5428–5435
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Fluorescent probes for investigation of isoprenoid configuration
and size discrimination by bactoprenol-utilizing enzymes
⇑
Anahita Z. Mostafavi, Donovan K. Lujan, Katelyn M. Erickson, Christina D. Martinez, Jerry M. Troutman
University of North Carolina at Charlotte, Department of Chemistry, 9201 University City Blvd., Charlotte, NC 28223-0001, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Undecaprenyl Pyrophosphate Synthase (UPPS) is an enzyme critical to the production of complex poly-
saccharides in bacteria, as it produces the crucial bactoprenol scaffold on which these materials are
assembled. Methods to characterize the systems associated with polysaccharide production are non-triv-
ial, in part due to the lack of chemical tools to investigate their assembly. In this report, we develop a new
fluorescent tool using UPPS to incorporate a powerful fluorescent anthranilamide moiety into bactopre-
nol. The activity of this analogue in polysaccharide biosynthesis is then tested with the initiating hexose-
1-phosphate transferases involved in Capsular Polysaccharide A biosynthesis in the symbiont Bacteroides
fragilis and the asparagine-linked glycosylation system of the pathogenic Campylobacter jejuni. In addi-
tion, it is shown that the UPPS used to make this probe is not specific for E-configured isoprenoid sub-
strates and that elongation by UPPS is required for activity with the downstream enzymes.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 24 April 2013
Revised 28 May 2013
Accepted 5 June 2013
Available online 15 June 2013
Keywords:
UPPS
Bactoprenol
Undecaprenol
Isoprenoid
cis-Prenyltransferase
1. Introduction
fluorescent appendages, could allow for even more sensitive detec-
tion of products.
Complex polysaccharides are critical to interactions between
humans and bacterial pathogens as well as symbiotic microorgan-
isms.1–5 The wide array of roles that these molecules can play
makes the development of new tools to understand their biosyn-
thetic pathways crucial. Bactoprenol is an isoprenoid produced in
bacterial systems for the assembly of complex carbohydrates,
including teichoic acids, peptidoglycan, lipopolysaccharides and
polysaccharide capsules.6–12 The 55-carbon Bactoprenol (B-OH) is
assembled as bactoprenyl diphosphate (BPP) from the 15-carbon
precursor farnesyl diphosphate (FPP) and eight 5-carbon isopente-
nyl diphosphates (IPP), as shown in Scheme 1. This series of con-
densation reactions is catalyzed by Undecaprenyl Pyrophosphate
Synthase (UPPS).13
Recently, a bactoprenyl probe was developed as a new tool to
investigate complex bacterial oligosaccharide biosynthesis.14 This
probe was produced using UPPS from the symbiotic microbe
Bacteroides fragilis as an enzymatic tool to incorporate a 4-nitroani-
line moiety into BPP from the FPP analogue 4-nitroaniline geranyl
diphosphate (4NA-G-PP, Fig. 1). However, the analogue was difficult
to see in solution or on thin layer chromatography (TLC) plates. In
addition, while the extinction coefficient of a 4-nitroaniline is high
(9500 MÀ1 cmÀ1 at k395 nm), other types of functionalities, such as
In order to design more powerful probes to study complex oli-
gosaccharide biosynthetic pathways, it is imperative to know the
limitations on what UPPS can and cannot accept as a substrate,
and whether UPPS-catalyzed elongation of isoprenoid diphos-
phates is even necessary for downstream enzyme recognition.
The X-ray crystal structure of UPPS from Escherichia coli has been
solved as the apoenzyme and as the substrate or product analogue
bound complex.15–18 The FPP substrate is coordinated through its
diphosphate moiety to essential Asn and Arg residues and is posi-
tioned in close proximity to IPP.19 As the isoprenoid chain elon-
gates, the farnesyl end is thought to travel through
a
hydrophobic tunnel until it reaches a length-determining helical
floor. Once the product reaches this floor, it is released from the
enzyme.
Synthetic analogues of FPP that are to be used as probes of
downstream biosynthetic pathways must be able both to interact
with UPPS and to travel through the hydrophobic tunnel below
the enzyme active site. Other types of analogues have been used
to study bacterial glycosylation with success.12,20 However, the
more these substrate alternatives resemble bactoprenol , the more
likely they are to be accepted as a general probe for any bactopre-
nol-dependent bacterial glycosylation pathway. A clear distinction
between analogue ability to interact with UPPS and ability to be
elongated has been described with two analogues, MANT-O-GPP
and Coumarin-GPP, shown in Figure 1.21,22 The small appendage
in MANT-O-GPP replacing the terminal (
cepted as a substrate by UPPS, and elongated to a near full length
x) isoprene of FPP was ac-
⇑
Corresponding author. Tel.: +1 704 687 5180; fax: +1 704 687 0960.
E-mail addresses: Jerry.Troutman@uncc.edu, jtroutm3@uncc.edu (J.M. Troutman).
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.06.007

A. Z. Mostafavi et al. / Bioorg. Med. Chem. 21 (2013) 5428–5435
5429
O-
O-
O
O
P
P
O
O-
O
O
P
O
P
O
O
O-
O-
O-
Farnesyl Diphosphate
FPP
+
UPPS
O
P
O
P
8
O
O
O-
O-
O-
Isopentenyl Diphosphate
IPP
Bactoprenyl Diphosphate
BPP
Scheme 1. Reaction catalyzed by UPPS probe.
properties of the substituents that interact unfavorably with the
enzyme active site or tunnel.
O
P
O
P
H
N
O
O
O-
O- O-
A previously established method to analyze complex oligosac-
charides linked to bactoprenyl diphosphates utilized a strongly
fluorescent anthranilamide moiety that was incorporated into the
free oligosaccharide following hydrolysis from the isoprenoid
diphosphate.8 After the oligosaccharide was cleaved and labeled,
it was then purified by normal phase HPLC and identified by MAL-
DI-MS.8 We were interested in whether the anthranilamide could
instead be linked directly to BPP to sensitively detect these linked
sugars. Such a procedure would bypass the labeling and cleavage
step and allow for further analysis of subsequent reactions through
retention of the diphosphate linkage. This work describes a power-
ful new probe to study oligosaccharide assembly that incorporates
this fluorescent tag into bactoprenol.
O2N
4NA-G-PP
H
N
O
P
O
P
O
O
O
O
O
O-
O-
O- O-
O
MANT-O-G-PP
Coumarin-G-PP
O
P
O
P
O
O
O-
O-
CF3
2. Results
2.1. Probe synthesis
H2N
O
O
P
O
P
H
Analogues of FPP can be used to incorporate new functionality
into bactoprenol using UPPS. However, it is not clear how wide a
range of functionality is acceptable to the enzyme. To probe the
N
O
O
O-
O- O-
dependence of a-isoprene E/Z-configuration on substrate recogni-
2AB-G-PP
tion, 2-aminobenzamide geranyl diphosphate (2AB-G-PP, Fig. 1)
and 2-aminobenzamide neryl diphosphate (2AB-N-PP, Fig. 1) were
designed to incorporate the easily detectable 2-aminobenzamide
moiety replacing the terminal isoprene of FPP. The analogues were
synthesized via a similar route (Scheme 2) from two allylic ace-
tates, neryl acetate and geranyl acetate.23–25 The allylic acetates
were oxidized to aldehydes at the terminal methyl positions, then
reductively aminated with 2-aminobenzamide. The resulting fluo-
rescent acetate was then saponified, halogenated and coupled to
the diphosphate, and purified by reverse phase (RP)-HPLC.
H2N
O
O
P
O
P
NH
O
O
O-
O- O-
2AB-N-PP
Figure 1. Analogues of FPP used to probe UPPS.
2.2. UPPS utilization of 2AB-G-PP
bactoprenol. However, the two-ring coumarin analogue was a poor
substrate for UPPS and acted more like a potent, competitive
inhibitor, possibly because its two-ring system was too bulky to
fit through the hydrophobic tunnel. However, the structural limits
of substrate activity for UPPS have not been systematically
explored, so it is difficult to determine whether the difference
in activity of these analogues was due to steric bulk or other
In order to test the activity of the 2-aminobenzamide analogues
with UPPS, assays were initially performed with 2AB-G-PP, which
was expected to be the preferred substrate due to its similarity
to native FPP. Reaction mixtures were prepared containing UPPS,
IPP and 2AB-G-PP, and long-chain isoprenoid products were phase
separated from water-soluble 2AB-G-PP through extraction into

5430
A. Z. Mostafavi et al. / Bioorg. Med. Chem. 21 (2013) 5428–5435
Scheme 2. Synthesis of 2AB-N-PP and 2AB-G-PP. Reagents and conditions: (a) t-BuOOH, salicylic acid, SeO2, Et2O; (b) 2-aminobenzamide, NaBH(OAc)3, DCE; (c) K2CO3,
MeOH, H2O; (d) PBr3, MeCN; (e) tris(tetra-N-butyl ammonium) diphosphate, MeCN.
chloroform. We found that an easily identifiable organic soluble
fluorescent product was extracted into the organic layer when
UPPS was present, and not when UPPS was left out of the reaction
mixture (Fig. 2a). In addition, the aqueous layer was significantly
more fluorescent when UPPS was left out of the reaction
(Fig. 2b). TLC analysis of the reaction mixtures (Fig. 2c) indicated
that a product was formed with increased hydrophobicity only
when UPPS was present. Reactions could be easily monitored
through this method by directly spotting the reaction mixture on
TLC plates, drying the plates at 130 °C, then resolving the plate in
a 73/53/11 chloroform/methanol/sodium borate buffer. Detection
limits were on the order of 10 pmol through visualization over a
365 nm UV lamp.
2.3.
a-Isoprene configuration is not critical for UPPS
recognition
Figure 3. (a) HPLC analysis of 2AB-G-PP reactions; (b) HPLC analysis of 2AB-N-PP.
In order to determine the effect of the a-isoprene configuration
on substrate recognition by UPPS, a RP-HPLC assay was utilized
with 2AB-G-PP and 2AB-N-PP. Identical reaction mixtures were
prepared containing each substrate and UPPS, which were
quenched with n-propanol after 1 hour then separated by RP-HPLC
(Fig. 3a). The starting material flowed rapidly through the column
with very little interaction, while three hydrophobic products re-
sulted in major peaks detected by fluorescence (ex. 350 nm em.
450 nm) at 5.6, 9.2 and 16.5 min. Interestingly, we observed iden-
tical activity with 2AB-N-PP, where levels of product were nearly
the same as with the 2AB-G-PP (Fig. 3b). These results suggested
that there was little effect on UPPS turnover with a change in the
E/Z configuration of the analogues. In addition, since retention
times of the products were identical, it is likely that the number
of isoprene units incorporated into each were identical.
2.4. UPPS can incorporate eight isoprene units into 2AB-G-PP
In order to confirm the identity of the material formed in the
2AB-G-PP reactions, large scale (>1 lmol) reaction products were
isolated by RP-HPLC (Supplementary Fig. 1) and characterized by
ESI-MS. The longest HPLC retention time products were the 2-ami-
nobenzamide bactoprenyl diphosphates (2AB-B-PP, Fig. 4a–c),
which included seven (cis-7) or eight (cis-8) isoprene additions.
These results were surprising, as previously, we had observed a
maximum of seven isoprene additions with 4NA-G-PP.14 Another
peak was observed in the reaction at 5.6 min and is presumed to
be 2AB-B-PP cis-6 however enough material was not collected to
confirm this.
2.5. 1H NMR characterization of 2AB-B-PP cis-7
The 2AB-B-PP cis-7 purified as described above was next charac-
terized by 1H NMR (Fig. 5). The chemical shifts of the 2AB-G-PP
were identical to chemical shifts in the bactoprenyl diphosphate
with additional protons associated with the newly incorporated
isoprene units. Integration of the spectrum confirmed nine total
olefins as would be expected for the cis-7 isoprenoid. The integral
for the methylene and methyl protons were expected to be 36 and
27, respectively. The observed number of protons associated with
the methylene and methyl groups were 34 and 29, which were
very close to the expected value. The slight difference in the num-
ber of protons is likely due to error propagated from integration
Figure 2. Reactions of 2AB analogues. (a) organic extracts of reactions without
(ÀUPPS) or with (+UPPS), (b) aqueous extracts of reactions in a, (c) TLC analysis of
2AB-G-PP reaction with or without UPPS. 2AB-B-PP Rf = 0.5, 2AB-G-PP Rf = 0.13 in
73/53/11 CHCl3/MeOH/NaBorate buffer.

A. Z. Mostafavi et al. / Bioorg. Med. Chem. 21 (2013) 5428–5435
5431
the transfer of N-acetyl-4-amino-galactosamine-1-phosphate
(AADGal-1-P) from uridine diphosphate (UDP)-linked AADGal to
bactoprenyl monophosphate (Scheme 3). HPLC analysis indicated
that as low as 20 fmols of 2AB-B-PP cis-8 could be detected, which
was more than three orders of magnitude better than the nitroan-
iline analogue. This improved detection sensitivity prompted us to
test the analogue with WcfS. The 2AB-B-PP cis-7 was first con-
verted to the monophosphate (2AB-B-P) with an alkaline phospha-
tase.14 Reactions were then prepared with a WcfS containing E. coli
membrane fraction and UDP-AADGal, biosynthesized as previously
reported.26 Products of these reactions were analyzed by RP-HPLC
(Fig. 6a). We found that the retention was decreased with the
incorporation of AADGal-phosphate. All of the limiting 2AB-B-P ap-
peared to be consumed, in the reaction. Similar results were ob-
served for the cis-8 compound (data not shown). The isolated
WcfS product had the expected molecular weight of 2AB-B-PP-
AADGal based on ESI-MS (Fig. 6b). These results suggest that the
new 2AB probe could be utilized to provide highly sensitive detec-
tion for this initiating hexose-1-phosphate transferase, and will
provide a remarkable probe for studying downstream glycosyla-
tion reactions.
2.7. Short chain neryl analogue is not a substrate for WcfS
Our analysis of the UPPS reaction suggests that smaller isopre-
noids than the cis-8 and cis-7 can be made, and we have previously
shown that a cis-6 4-nitroaniline bactoprenyl phosphate is a sub-
strate for WcfS.26 We were interested in whether a short chain,
two isoprene, chemically synthesized neryl phosphate could be
utilized for these studies to allow us to bypass the use of UPPS in
synthesizing our probes. A 2AB-neryl phosphate was synthesized
through methods similar to the diphosphates. 2AB-N-P was mixed
with WcfS and UDP-AADGal and analyzed by RP-HPLC. No product
was observed using HPLC gradient conditions similar to those used
to purify 2AB-N-P (data not shown). These results suggested that
the longer chain isoprenoids were critical to recognition by the ini-
tiating hexose-1-phosphate transferase, and extension by UPPS is
crucial to the development of these tools.
Figure 4. Characterization of 2AB reaction products. (a) Structure of primary 2AB-
B-PP products; (b) MS analysis of cis-7; (c) MS analysis of cis-8. ⁄unidentified
contaminating peak at ꢀ1477 amu.
2.8. Assay development by thin layer chromatography
While HPLC is clearly the method of choice for detailed quanti-
tative reaction analysis and isolation of products, TLC can often be
a quicker method for screening reactions to identify the function of
an unknown protein, or for screening reaction conditions for
known proteins. Using the Campylobacter jejuni enzyme, PglC,
which catalyzes transfer of 4-N-Acetyl Bacillosamine (AcBac)-
phosphate to a bactoprenyl phosphate (Scheme 3), we tested the
ability to screen assay conditions by TLC. Duplicate reactions were
Figure 5. 1H NMR of 2AB-B-PP cis-7 on a 500 MHz spectrophotometer in d6 DMSO.
Integral values are given above each peak.
prepared with 96 lM 2AB-B-P cis-8 and either 160 lM or 80 lM
UDP-AcBac. The sub-stoichiometric UDP-linked sugar was ex-
pected to give less product than when the sugar was in excess.
normalization. Coupled with the mass spectrometry data, these
results confirmed the formation of the long chain isoprenoid with
2AB-G-PP as the UPPS substrate. Our previous analogue, 4NA-G-PP,
had been elongated using a tetra-N-butylammonium-linked IPP,
which resulted in a relatively complex 1H NMR spectrum due to
the counterion butyl group protons. Importantly, in this +NH4 form
IPP was used which greatly reduced the complexity of the spec-
trum and allowed for facile integration and analysis.
After a 1 h incubation period with PglC, 2 lL of the reaction
(ꢀ200 pmols of anthranilamide) was removed and separated by
TLC (Fig. 7). We found that we could easily distinguish between
the two reactions and that the fluorescence was similar for the
duplicate measurements. The size and intensity of each of the sep-
arated TLC spots was also quantified and subtracted from back-
ground to determine if this was a viable method for estimating
reaction turnover. We observed 20 2% turnover for the 160
lM
reaction and 7.7 0.5% turnover for the 80 M reaction. These val-
l
2.6. Initiating hexose-1-phosphate transferase, WcfS
ues were consistent with expected values and with the appearance
of the spots on the plate. Identical reactions were prepared and
analyzed by RP-HPLC to determine the accuracy of these measure-
Recently, our group has isolated the key initiating hexose-1-
phosphate transferase, WcfS, involved in capsule biosynthesis for
the mammalian symbiont B. fragilis.26 This enzyme catalyzes
ments (Supplementary Fig. 2). We found that the 160
lM reaction
turnover based on fluorescent peak integrals was 14%, while the

5432
A. Z. Mostafavi et al. / Bioorg. Med. Chem. 21 (2013) 5428–5435
Scheme 3. Reactions catalyzed by WcfS and PglC.
was straightforward. Overall, these TLC based assays should be
readily usable for the analysis of other complex oligosaccharide
biosynthetic systems. Reactions with the analogue were also per-
formed in which full turnover was observed with PglC (data not
shown). Importantly, since both substrate and product can be ob-
served on the TLC plates, each lane can be monitored for relative
turnover, and does not require perfect spotting of the plates. How-
ever, it is clear that an HPLC-based analysis will be much more reli-
able with respect to quantitative accurate turnover measurements.
3. Discussion
3.1. Alternative substrates and UPPS
Numerous analogues of FPP have been designed to probe the
substrate specificity of other prenyltransferases, particularly those
involved in the farnesylation of critical mammalian signaling pro-
teins.23,24,27–31 These analogues included a diverse array of aromatic
moieties replacing multiple isoprenes of FPP for the investigation of
enzymes associated with those pathways. Application of such sys-
tematic studies to the structure function relationships between
FPP and UPPS would greatly benefit our ability to both design
new probes and gain better insight into the mechanism of substrate
recognition by the protein. The relatively small sampling of ana-
logues that have been tested with UPPS thus far makes it difficult
to discern why the 2AB-G-PP analogue accepts eight isoprene addi-
tions while the 4NA-G-PP does not. However, it is clear that a fair
degree of freedom is available for functionalization of an aromatic
system replacing the terminal isoprene of FPP.
Figure 6. (a) HPLC analysis of 2AB-B-P cis-7 reactions with UDP-AADGal catalyzed
by WcfS; (b) ESI-MS analysis of WcfS reaction product. Expected molecular weight
1109.65 amu.
3.2. Conclusions
The fluorescent bactoprenyl diphosphate provides an intriguing
new tool for the analysis of complex oligosaccharide biosynthetic
systems. The enormous detection sensitivity with this molecule
makes the analysis of these pathways much more tractable as very
low turnover reactions can be observed readily even by TLC analy-
sis. In addition, the ability to readily isolate products from these
reactions for characterization by ESI-MS is also an important fea-
ture of these probes. Of critical interest to us is whether these types
of molecules or precursors can be used for cellular analysis of oli-
gosaccharide assembly systems in vivo. In addition, we are con-
ducting ongoing studies to probe how well 2AB-G-PP acts as a
substrate relative to FPP. Both of these areas have broad signifi-
cance with respect to utilizing these probes to investigate unchar-
acterized polysaccharide biosynthesis pathways.
Figure 7. TLC analysis of PglC reactions. Reactions containing 96
2AB-B-P cis-8 were spotted on TLC plates and resolved in 73/53/11 CHCl3/MeOH/
sodium borate buffer. (1–2) duplicate spots of 160 M UDP-AcBac reaction, (3–4)
M UDP-AcBac reactions. 2AB-B-P Rf = 0.58 2AB-B-PP-AcBac
lM HPLC purified
l
duplicate spots of 80
l
Rf = 0.51.
80 lM reaction turnover was 6%. These results suggest that the TLC
assay results were useful for approximate turnover not perfect
quantification. The TLC assays were very quick to perform and
visualization of the 2AB substrate and product over a UV lamp

A. Z. Mostafavi et al. / Bioorg. Med. Chem. 21 (2013) 5428–5435
5433
2-Amideaniline-bactoprenyl monophosphate cis-8 expected
4. Methods
4.1. General
mass (911.64 amu) ESI-MS (MÀ1) 911.73 amu.
4.6. TLC and extraction analysis of 2AB-G-PP reactions
All general reagents and solvents were purchased from Sigma–
Aldrich, VWR or Fisher unless stated otherwise. 2-Aminobenzam-
ide-containing species were quantified at 350 nm using an extinc-
tion coefficient of 4500 MÀ1 cmÀ1. All HPLC was performed on an
Agilent 1100 system equipped with a diode array and fluorescence
detector and Gilson fraction collector. TLC was performed on K6f
non-fluorescent plates (Whatman). Synthetic protocol TLC was
performed in 1:1 hexanes:ethyl acetate, while bactoprenol TLC
was performed in 73/53/11 CHCl3/MeOH/0.1 M sodium borate.
NMR spectroscopy was performed on a 500 MHz or 300 MHz Jeol
for 1H NMR and 13C NMR, and a 300 MHz Jeol spectrometer for
31P NMR. ESI-MS analysis was performed in negative ion mode
on a Thermo Electron Finnigan LTQ ESI-MS with injection volumes
Reactions were prepared with 367
M UPPS, 5 mM KCl, 0.5 mM MgCl2, and 0.1% DDM 50 mM
HEPES (pH 7.4). Reactions and control without UPPS were incu-
bated for 2 h at 37 °C. A 2.5 L aliquot of sample and control was
lM 2AB-G-PP, 10 mM IPP,
4.3
l
l
then spotted onto a non-fluorescent TLC plate. The plate was dried
for 3 min in a 130 °C oven, then product was separated in 73/53/11
CHCl3/MeOH/0.1 M sodium borate over 50 mm. A 20
was extracted into 400 L of 2:1 CHCl3/methanol then washed
with 3 Â 100 L Pure Solvent Upper Phase (PSUP, 4% KCl in 1:1
methanol/water). Tubes containing the organic or aqueous extracts
lL aliquot
l
l
and TLC plates were visualized over
(365 nm).
a UV transilluminator
of 200 lL and concentrations ranging from 50 to 300 lM. UPPS,
PglC, WcfS, and substrates were prepared as described previ-
ously.14,26,32,33 Synthetic protocols generally followed those estab-
lished by Spielmann and co-workers including 2b previously
reported.24
4.7. WcfS reactions with 2AB-B-P and 2AB-N-P
The 2AB-B-PP-AADGal was prepared and 2AB-N-P was tested
by adding 23
lg total protein in a membrane fraction of WcfS to
a 200 L solution containing 0.1 nmol 2AB-B-P cis-7 or short chain
l
4.2. 2AB-G-PP and 2AB-N-PP analytical scale reactions
2AB-N-P, 19.6 nmol UDP-AADGal, 1% Triton X-100 hydrogenated,
10 mM MgCl2, 0.5% DMSO, and 50 mM Tris–HCl (pH 8.0). The
2AB-B-P reaction was incubated at room temperature for 30 min,
after which it was analyzed using the RP-HPLC method described
above. 2AB-N-P was analyzed by RP-HPLC using the same RP-C18
column, but loaded onto the column in 20% acetonitrile/80%
25 mM ammonium bicarbonate, then separated with a linear gra-
dient over 30 min to 100% acetonitrile. Two negative controls were
used, one in the absence of UDP-AADGal and the other without
WcfS protein.
Reactions were prepared containing 6.8
N-PP, 0.1% n-dodecyl-b- -maltoside, 50 mM HEPES (pH 7.4),
5 mM KCl, 0.5 mM MgCl2, 0.2 mM IPP, 344 nM UPPS in 250 L total
volume. After 1 h at 37 °C, reactions were quenched with the addi-
tion of 250 L of n-propanol then analyzed by RP-HPLC. Isocratic
analytical HPLC on bactoprenol-linked materials was performed
on a C18 column (Agilent Eclipse XDB-C18 5
lM 2AB-G-PP or 2AB-
D
l
l
lm 4.6 Â 150 mm)
in 50% n-propanol/50% ammonium bicarbonate (25 mM), with
fluorescence detection at ex. 350 nm, em. 450 nm.
2-Amideaniline-bactoprenyl diphosphate AADGal cis-7 Ex-
pected mass (1109.65 amu) ESI-MS (MÀ1) 1109.66 amu.
4.3. 2AB-B-PP synthesis
2AB-B-PP was prepared in 50 mM HEPES (pH 7.4), 0.5 mM
4.8. PglC TLC reactions
MgCl2, 5 mM KCl, and 0.1% DDM, with 5.7
lM UPPS, 20 lmol of
IPP and 3 mol of 2AB-G-PP in a total volume of 1.5 mL. The reac-
l
Reaction mixtures were prepared in duplicate containing 96
2AB-B-P cis-8, 160 M or 80 M UDP-AcBac (synthesized as previ-
ously described32), 2% DMSO, 0.9 mg/mL PglC, 50 mM TrisOAc (pH
lM
tion was incubated at 37 °C for 6 h. Products were purified using an
isocratic method 75% n-propanol/25% ammonium bicarbonate
(25 mM) on a C18 Varian 250 Â 21.2 mm column (flow rate
4.0 mL/min).
l
l
8.0), 101% Triton-X-100, 10 mM MgCl2 in a total volume of 25
Reactions were incubated at room temperature for 1 h then 2
lL.
l
L
was spotted on TLC plates and product was separated in 73/53/
11 CHCl3/MeOH/0.1 M sodium borate buffer. Plates were visual-
ized over a transilluminator. Peak quantification was performed
using Adobe Photoshop where plate images were converted to
black and white then spot area and gray value were measured. Per-
cent turnover was determined from the gray value and pixel area
of the product minus background over the total for substrate and
product. RP-HPLC was performed as described with the WcfS reac-
tions except separations were performed in 60% n-propoanl/40%
25 mM ammonium bicarbonate.
4.4. 2-Amideaniline-bactoprenyl diphosphate cis-7
characterization
1H NMR (500 MHz): d 8.29 (t, 1H, J = 5.9 Hz), 7.56 (d, 1H,
J = 8.2 Hz), 7.15 (t, 1H, J = 7.5 Hz), 6.53 (d, 1H, J = 8.5 Hz), 6.45 (t,
1H, J = 7.5 Hz), 5.29 (t, 1H, J = 6.5 Hz), 5.25 (t, 1H, J = 6.5 Hz), 5.05
(m, 7H), 4.16 (t, 2H, J = 6.5 Hz), 3.60 (d, 2H, J = 4.9 Hz), 2.04 (q,
2H, J = 7 Hz), 1.95 (am, 32H), 1.62 (s, 3H), 1.59 (as, 20H), 1.54 (s,
3H), 1.51 (s, 3H). Expected mass (923.55 amu) ESI-MS (MÀ1):
923.53 amu.
2-Amideaniline-bactoprenyl diphosphate cis-8 Expected mass
4.9. Synthetic protocols
(991.61 amu) ESI-MS (MÀ1) 991.68 amu.
4.9.1. (6Z)-3,7-Dimethyl-8-acetoxyl-2,6-octadien-1-al (2a)
6.983 g (0.035 mol) of 1a was added to a stirring solution of
0.491 g (3.5 mmol) of salicylic acid, 0.665 g (3.5 mmol) of SeO2
and 14.82 g (0.164 mol) of tert-butylhydroperoxide (70%) in
35 mL of diethyl ether at 0 °C and left to stir for 48 h. The volume
of the mixture was then reduced to one half and was washed with
15 mL of 5% NaHCO3, saturated CuSO4, repeatedly washed with
saturated Na2S2O3, water, brine and then dried over MgSO4.
4.5. 2AB-B-P synthesis
The 2AB bactoprenyl monophosphate was prepared by dissolv-
ing 2AB-B-PP in 20% n-propanol in 25 mM ammonium bicarbonate.
Sample was then treated with 2 lL of goat anti-rabbit alkaline
phosphate (Pierce) and incubated for 2 h. Product was purified
by HPLC as described for 2AB-B-PP.

5434
A. Z. Mostafavi et al. / Bioorg. Med. Chem. 21 (2013) 5428–5435
Solvent was removed by rotary evaporation and the crude oil was
then purified by flash chromatography (5% EtOAc/hexanes) to
afford 3.53 g (47%) of 2a. TLC: (Rf 0.7). 1H NMR (300 MHz): d 9.28
(s, 1H), 6.37 (t, 1H), 5.32 (t, 1H), 4.49 (d, 2H), 2.39 (q, 2H), 2.25
(t, 2H), 1.91 (s, 3H) 1.69 (s, 3H), 1.58 (s, 3H). 13C NMR (76 MHz):
d 195.2, 171.0, 153.2, 140.8, 139.7, 120.5, 60.7, 30.4, 27.3, 25.8
(contaminant), 23.2, 21.0, 9.2.
4.9.6. (6Z)-8-(2-Amideaniline)-3,7-dimethyl-1-bromo-2,6-
octadiene (5a)
0.100 g (0.4013 mmol) of 4a in 5 mL of MeCN was added to
0.0436 g (0.161 mmol) of PBr3 and allowed to stir for 15 min at
0 °C. The reaction mixture was then concentrated via rotary evap-
oration and sent through a short pad of silica with a 50% ethyl ace-
tate/hexanes solution. Solvent was removed by rotary evaporation
to yield a crude mixture of 5a. TLC: (Rf 0.43).
4.9.2. (6Z)-8-(2-Amideaniline)-3,7-dimethyl-1-acetoxyl-2,6-
octadiene (3a)
4.9.7. (6E)-8-(2-Amideaniline)-3,7-dimethyl-1-bromo-2,6-
3.53 g (0.0168 mol) of aldehyde 2a was added to a stirring solu-
tion of 2.51 g (0.0184 mol) of anthranilamide and 1.41 g
(0.0235 mol) of acetic acid in 75 mL of dichloroethane in a flame
dried flask. After the solution became homogenous, 5.34 g
(0.0250 mol) of NaBH(OAc)3 was added and left to stir for 24 h at
room temperature. The reaction mixture was then dried via rotary
evaporation to one half the initial volume and quenched with
200 mL of 5% NaHCO3. The mixture was then extracted with
diethyl ether three times (100 mL), dried with MgSO4, and concen-
trated via rotary evaporation. The crude oil was then purified by
flash chromatography (15% EtOAc/hexanes) to yield 0.9086 g
(22%) of 3a. TLC: (Rf 0.32) 1H NMR (300 MHz): d 7.97 (t,1H), 7.38
(d, 1H), 7.26 (t, 1H), 6.62 (d, 1H), 6.54 (t, 1H), 5.78 (br s, 2H),
5.34 (m, 2H), 4.47 (d, 2H), 3.70 (d, 2H), 2.13 (m, 4H), 2.02 (s, 3H),
1.73 (s, 3H), 1.65 (s, 3H). 13C NMR (76 MHz): d 172.2, 171.2,
150.4, 142.6, 133.5, 132.2, 128.3, 124.6, 119.5, 114.3, 113.2,
112.3, 61.0, 50.5, 31.8, 26.4, 23.4, 21.3, 14.5. Expected mass
(329.19 amu). Found: MÀ1: 329.3 amu.
octadiene (5b)
TLC: (Rf 0.43).
4.9.8. (6Z)-8-(2-Amideaniline)-3,7-dimethyl-2,6-octadiene
diphosphate (6a)
The crude mixture of (5a) was immediately added to 1.10 g
(1.20 mmol) of dry tris(tetra-N-butylammonium) hydrogen
diphosphate and allowed to stir for 24 h at room temperature.
The crude mixture was then dissolved in 50 mL of 25 mM ammo-
nium bicarbonate buffer and passed through an ion exchange col-
umn (ammonium form). The column was eluted with 200 mL of
buffer and that mixture was frozen and lyophilized overnight.
The solid material was dissolved in minimal buffer and purified
via reverse phase HPLC to yield 0.009 g (5%). 1H NMR (300 MHz):
d 7.39 (d, 1H), 7.23 (m, 2H), 6.68 (d, 1H), 6.60 (t, 3H), 5.24 (m,
2H), 4.23 (t, 2H), 3.58 (s, 2H), 2.00 (m, 4H), 1.52 (s, 3H), 1.47 (s,
3H). 31P NMR (122 MHz): d À7.51 (d, 1P), À9.93 (d, 1P). Expected
mass (447.11 amu). ESI-MS (MÀ1) 447.25 amu.
4.9.3. (6E)-8-(2-Amideaniline)-3,7-dimethyl-1-acetoxyl-2,6-
octadiene (3b)
4.9.9. (6E)-8-(2-Amideaniline)-3,7-dimethyl-2,6-octadiene
diphosphate (6b)
Yield (44%). TLC: (Rf 0.37) 1H NMR (300 MHz): d 8.00 (t, 1H),
7.37 (d, 1H), 7.28 (t, 1H), 6.62 (d, 1H), 6.56 (t, 1H), 5.84 (br s,
2H), 5.35 (t, 1H), 5.29 (t, 1H), 4.55 (d, 2H), 3.69 (d, 2H), 2.16 (m,
2H), 2.06 (m, 5H), 1.67 (s, 3H), 1.65 (s, 3H). 13C NMR (76 MHz): d
172.4, 164.8, 150.4, 137.9, 133.8, 131.6, 128.4, 124.5, 121.6,
114.8, 114.2, 112.9, 112.1, 59.3, 50.1, 38.9, 25.5, 16.4, 15.0. Ex-
pected mass (329.19 amu). Found: MÀ1 329.2 amu.
Yield 14%. 1H NMR (300 MHz) d 7.41 (d, 1H, J = 7.5 Hz), 7.24 (t,
1H, J = 7.2 Hz), 6.68 (d, 1H, J = 8.7 Hz), 6.63 (t, 1H, J = 7.8 Hz), 5.18
(m, 2H, J = 7.5 Hz), 4.25 (t, 2H, J = 6.3 Hz), 3.58 (s, 2H), 2.00 (t, 2H,
J = 7.2 Hz), 1.89 (q, 2H, J = 6.6 Hz) 1.49 (s, 3H), 1.46 (s, 3H). 13C
NMR (76 MHz): 183.1, 148.6, 142.6, 133.5, 132.0, 128.8, 125.8,
119.6, 116.8, 115.9, 113.8, 62.8, 50.0, 38.3, 25.1, 15.5, 13.5. 31P
NMR (122 MHz) d À7.5 (d, 1P), À10.5 (d, 1P). Expected mass
(447.11 amu). ESI-MS (MÀ1) 447.22 amu.
4.9.4. (6Z)-8-(2-Amideaniline)-3,7-dimethyl -2,6-octadien-1-ol
(4a)
4.9.10. (6Z)-8-(2-Amideaniline)-3,7-dimethyl-2,6-octadiene
monophosphate
0.500 g (2.016 mmol) of 3a was dissolved in 21 mL of methanol
and introduced into a flask containing 0.834 g (6.04 mmol) of
K2CO3 dissolved in 1.5 mL of water. The reaction was refluxed at
70 °C for 24 h. The solution was then dissolved in water and ex-
tracted four times (50 mL) with chloroform, washed with brine
and dried with MgSO4. The crude solution was then concentrated
via rotary evaporation and purified by flash chromatography
(20% EtOAc/hexanes) to yield 0.200 g (40%) of 4a. TLC: (Rf 0.1) 1H
NMR (300 MHz) d 7.94 (t,1H), 7.39 (d,1H), 7.24 (t, 1H), 6.61
(d,1H), 6.51 (t,1H), 6.37 (br s, 2H), 5.35 (m,2H), 4.07 (d, 2H), 3.64
(d,2H), 2.96 (br s, 1H), 2.09 (m,4H), 1.68(s, 3H), 1.64 (s, 3H). 13C
NMR (76 MHz) d 172.8, 150.6, 138.19, 133.3, 132.1, 128.7, 125.3,
124.3, 114.5, 113.3, 112.1, 58.9, 50.3, 31.6, 25.7, 23.3, 15.2 Ex-
pected mass (287.18 amu). Found: MÀ1 287.3 amu.
A crude mixture of 5a was immediately added to 0.634 g
(0.690 mmol) of dry di (tetra-N-butyl ammonium) hydrogen
monophosphate and allowed to stir at room temperature for
24 h. The crude mixture was then dissolved in 50 mL of 25 mM
ammonium bicarbonate buffer and passed through an ion ex-
change column (ammonium form). The column was eluted with
200 mL of buffer and that mixture was frozen and lyophilized over-
night. The solid material was dissolved in minimal buffer and puri-
fied via reverse phase HPLC to afford 3.0 mg (2.3%) of
monophosphate product. 1H NMR (300 MHz): d 7.41 (d, 1H), 7.27
(t, 1H), 6.70 (d, 1H), 6.63 (t, 1H), 5.24 (m, 2H), 4.13 (t, 2H), 3.61
(s, 2H), 2.01 (m, 4H), 1.55 (s, 3H), 1.50 (s, 3H). 31P NMR
(122 MHz): d 2.17 (s). Expected mass (367.14 amu). ESI-MS (MÀ1
)
367.33 amu.
4.9.5. (6E)-8-(2-Amideaniline)-3,7-dimethyl-2,6-octadien-1-ol
(4b)
Yield (60%). TLC: (Rf 0.16) 1H NMR (300 MHz): d 7.86 (br t, 1H),
7.31(d, 1H), 7.22 (t, 1H), 6.58 (d, 1H), 6.48 (t, 1H), 5.77 (br s, 2H),
5.32 (m, 2H), 4.06 (d, 2H), 3.60 (d, 2H), 2.11 (q, 2H), 1.99(t, 2H),
1.60 (s, 3H), 1.57 (s, 3H). 13C NMR (76 MHz): d 172.2, 164.8,
150.4, 137.9, 133.8, 131.6, 128.4, 124.5, 121.3, 114.8, 112.9,
112.1, 59.2, 50.1, 38.9, 25.5, 16.4, 15.0. Expected mass
(287.18 amu). Found: MÀ1 287.3 amu.
Acknowledgments
We thank, Dr. Barbara Imperiali for supplying PglC expressing
cells, Dr. Jian Liu for allowing us to use his MS system and Michael
Morrison and Dr. Richard Jew for critical reading of this manuscript
and helpful suggestions. This work was supported by National
Institutes of Health AREA Grant R15GM100402 (J.M.T)

A. Z. Mostafavi et al. / Bioorg. Med. Chem. 21 (2013) 5428–5435
5435
14. Lujan, D. K.; Stanziale, J. A.; Mostafavi, A. Z.; Sharma, S.; Troutman, J. M.
Carbohydr. Res. 2012, 359, 44.
Supplementary data
15. Teng, K. H.; Liang, P. H. Bioorg. Chem. 2012, 43, 51.
Supplementary data (RP-HPLC of large scale HPLC preparation
of 2AB-B-PP, PglC reactions with 2AB-B-P cis-8, ESI-MS of 2AB-B-
P cis-8, full chemical spectra of BPP precursors) associated with this
article can be found, in the online version, at http://dx.doi.org/
10.1016/j.bmc.2013.06.007.
16. Kuo, C. J.; Guo, R. T.; Lu, I. L.; Liu, H. G.; Wu, S. Y.; Ko, T. P.; Wang, A. H. J.; Liang,
P. H. J. Biomed. Biotechnol. 2008. doi:10.1155/2008/841312.
17. Ko, T. P.; Chen, Y. K.; Robinson, H.; Tsai, P. C.; Gao, Y. G.; Chen, A. P. C.; Wang, A.
H. J.; Liang, P. H. J. Biol. Chem. 2001, 276, 47474.
18. Guo, R. T.; Ko, T. P.; Chen, A. P. C.; Kuo, C. J.; Wang, A. H. J.; Liang, P. H. J. Biol.
Chem. 2005, 280, 20762.
19. Pan, J. J.; Yang, L. W.; Liang, P. H. Biochemistry 2000, 39, 13856.
20. Perlstein, D. L.; Zhang, Y.; Wang, T. S.; Kahne, D. E.; Walker, S. J. Am. Chem. Soc.
2007, 129, 12674.
References and notes
21. Teng, K. H.; Chen, A. P. C.; Kuo, C. J.; Li, Y. C.; Liu, H. G.; Chen, C. T.; Liang, P. H.
Anal. Biochem. 2011, 417, 136.
22. Chen, A. P. C.; Chen, Y. H.; Liu, H. P.; Li, Y. C.; Chen, C. T.; Liang, P. H. J. Am. Chem.
Soc. 2002, 124, 15217.
1. Surana, N. K.; Kasper, D. L. Immunol. Rev. 2012, 245, 13.
2. Ochoa-Reparaz, J.; Mielcarz, D. W.; Wang, Y.; Begum-Haque, S.; Dasgupta, S.;
Kasper, D. L.; Kasper, L. H. Mucosal Immunol. 2010, 3, 487.
3. Mazmanian, S. K.; Liu, C. H.; Tzianabos, A. O.; Kasper, D. L. Cell 2005, 122, 107.
4. Fong, J. C. N.; Syed, K. A.; Klose, K. E.; Yildiz, F. H. Microbiology-Sgm 2010, 156,
2757.
5. Linkous, D. A.; Oliver, J. D. FEMS Microbiol. Lett. 1999, 174, 207.
6. Woodward, R.; Yi, W.; Li, L.; Zhao, G. H.; Eguchi, H.; Sridhar, P. R.; Guo, H. J.;
Song, J. K.; Motari, E.; Cai, L.; Kelleher, P.; Liu, X. W.; Han, W. Q.; Zhang, W. P.;
Ding, Y.; Li, M.; Wang, P. G. Nat. Chem. Biol. 2010, 6, 418.
7. Whitfield, C. Annu. Rev. Biochem. 2006, 75, 39.
23. Chehade, K. A. H.; Andres, D. A.; Spielmann, H. P. Biochemistry 2000, 39, 1560.
24. Chehade, K. A. H.; Kiegiel, K.; Isaacs, R. J.; Pickett, J. S.; Bowers, K. E.; Fierke, C.
A.; Andres, D. A.; Spielmann, H. P. J. Am. Chem. Soc. 2002, 124, 8206.
25. Davisson, V. J.; Woodside, A. B.; Neal, T. R.; Stremler, K. E.; Muehlbacher, M.;
Poulter, C. D. J. Org. Chem. 1986, 51, 4768.
26. Mostafavi, A. Z.; Troutman, J. M. Biochemistry 1939, 2013, 52.
27. Placzek, A. T.; Hougland, J. L.; Gibbs, R. A. Org. Lett. 2012, 14, 4038.
28. Rawat, D. S.; Krzysiak, A. J.; Gibbs, R. A. J. Org. Chem. 2008, 73, 1881.
29. Kim, M. K.; Kleckley, T. S.; Wiemer, A. J.; Holstein, S. A.; Hohl, R. J.; Wiemer, D. F.
J. Org. Chem. 2004, 69, 8186.
8. Glover, K. J.; Weerapana, E.; Imperiali, B. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,
14255.
9. Swoboda, J. G.; Campbell, J.; Meredith, T. C.; Walker, S. ChemBioChem 2010, 11,
35.
10. Perlstein, D. L.; Wang, T. S. A.; Doud, E. H.; Kahne, D.; Walker, S. J. Am. Chem. Soc.
2010, 132, 48.
11. Greenfield, L. K.; Whitfield, C. Carbohydr. Res. 2012, 356, 12.
12. Han, W. Q.; Wu, B. L.; Li, L.; Zhao, G. H.; Woodward, R.; Pettit, N.; Cai, L.; Thon,
V.; Wang, P. G. J. Biol. Chem. 2012, 287, 5357.
30. Henry, O.; Lopez-Gallego, F.; Agger, S. A.; Schmidt-Dannert, C.; Sen, S.; Shintani,
D.; Cornish, K.; Distefano, M. D. Bioorg. Med. Chem. 2009, 17, 4797.
31. Dursina, B.; Reents, R.; Delon, C.; Wu, Y. W.; Kulharia, M.; Thutewohl, M.;
Veligodsky, A.; Kalinin, A.; Evstifeev, V.; Ciobanu, D.; Szedlacsek, S. E.;
Waldmann, H.; Goody, R. S.; Alexandrov, K. J. Am. Chem. Soc. 2006, 128, 2822.
32. Olivier, N. B.; Chen, M. M.; Behr, J. R.; Imperiali, B. Biochemistry 2006, 45, 13659.
33. Glover, K. J.; Weerapana, E.; Chen, M. M.; Imperiali, B. Biochemistry 2006, 45,
5343.
13. Teng, K. H.; Liang, P. H. Mol. Membr. Biol. 2012, 29, 267.