Proteomic searches comparing two (R)-lacosamide affinity baits: An electrophilic arylisothiocyanate and a photoactivated arylazide group

Article abstract of DOI:10.1039/c000987c

We have advanced a novel strategy to search for lacosamide ((R)-1) targets in the brain proteome where protein binding is expected to be modest. Our approach used lacosamide agents containing "affinity bait" (AB) and "chemical reporter" (CR) units. The affinity bait moiety is designed to irreversibly react with the target, and the CR group permits protein detection and capture. In this study, we report the preparation and evaluation of (R)-N-(4-azido)benzyl 2-acetamido-3-(prop-2-ynyloxy)propionamide ((R)-3) and show that this compound exhibits potent anticonvulsant activities in the MES seizure model in rodents. We compared the utility of (R)-3 with its isostere, (R)-N-(4-isothiocyanato)benzyl 2-acetamido-3-(prop-2-ynyloxy)propionamide ((R)-2), in proteomic studies designed to identify potential (R)-1 targets. We showed that despite the two-fold improved anticonvulsant activity of (R)-3 compared with (R)-2, (R)-2 was superior in revealing potential binding targets in the mouse brain soluble proteome. The difference in these agents' utility has been attributed to the reactivity of the affinity baits (i.e., (R)-2: aryl isothiocyanate moiety; (R)-3: photoactivated aryl azide intermediates) in the irreversible protein modification step, and we conclude that this factor is a critical determinant of successful target detection where ligand (drug) binding is modest. The utility of (R)-2 and (R)-3 in in situ proteome studies is explored.

Full text of DOI:10.1039/c000987c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Proteomic searches comparing two (R)-lacosamide affinity baits: An
electrophilic arylisothiocyanate and a photoactivated arylazide group†
Ki Duk Park,a James P. Stables,b Rihe Liu*a,c and Harold Kohn*a,d
Received 18th January 2010, Accepted 23rd March 2010
First published as an Advance Article on the web 19th April 2010
DOI: 10.1039/c000987c
We have advanced a novel strategy to search for lacosamide ((R)-1) targets in the brain proteome where
protein binding is expected to be modest. Our approach used lacosamide agents containing “affinity
bait” (AB) and “chemical reporter” (CR) units. The affinity bait moiety is designed to irreversibly react
with the target, and the CR group permits protein detection and capture. In this study, we report the
preparation and evaluation of (R)-N-(4-azido)benzyl 2-acetamido-3-(prop-2-ynyloxy)propionamide
((R)-3) and show that this compound exhibits potent anticonvulsant activities in the MES seizure
model in rodents. We compared the utility of (R)-3 with its isostere, (R)-N-(4-isothiocyanato)benzyl
2-acetamido-3-(prop-2-ynyloxy)propionamide ((R)-2), in proteomic studies designed to identify
potential (R)-1 targets. We showed that despite the two-fold improved anticonvulsant activity of (R)-3
compared with (R)-2, (R)-2 was superior in revealing potential binding targets in the mouse brain
soluble proteome. The difference in these agents’ utility has been attributed to the reactivity of the
affinity baits (i.e., (R)-2: aryl isothiocyanate moiety; (R)-3: photoactivated aryl azide intermediates) in
the irreversible protein modification step, and we conclude that this factor is a critical determinant of
successful target detection where ligand (drug) binding is modest. The utility of (R)-2 and (R)-3 in in
situ proteome studies is explored.
Introduction
Lacosamide1 ((R)-1) is a first-in-class, antiepileptic drug (AED)
that has been approved in the United States and Europe for the
adjunctive treatment of partial-onset seizures in adults.2 While
recent reports indicate that (R)-1 may modulate both the slow
inactivation state of sodium channels3,4 and the CRMP2 signaling
pathway,5,6 the full mechanism of action of this AED remains
elusive. To gain additional information of (R)-1 function, we
advanced a strategy to identify drug targets where ligand binding is
expected to be modest.7 Our approach uses agents termed “affinity
bait” (AB) and “chemical reporter” (CR), where these AB and
CR units are incorporated within the drug framework (Scheme 1).
The AB group permits irreversible adduction of the modified drug
Scheme 1 “Affinity bait” and “chemical reporter” (AB&CR) strategy to
scan the brain proteome for (R)-lacosamide target(s).
to the receptor, and the CR unit allows protein detection and
capture upon subsequent attachment to a chemical probe (P). Key
to this strategy is that neither the AB nor the CR group impedes
drug binding to the cognate receptor(s).
aDivision of Medicinal Chemistry and Natural Products, UNC Eshelman
School of Pharmacy, University of North Carolina, Chapel Hill, North
Carolina 27599-7568, USA
bAnticonvulsant Screening Program, National Institute of Neurological
Disorders and Stroke, National Institutes of Health, 6001 Executive Blvd.,
Suite 2106, Rockville, MD 20892-9523, USA
cCarolina Center for Genome Sciences, University of North Carolina,
Chapel Hill, North Carolina 27599-7264, USA. E-mail: rliu@email.unc.edu;
Fax: +1 919-966-0204; Tel: +1 919-843-3635
dDepartment of Chemistry, University of North Carolina, Chapel Hill, North
Carolina 27599-3290, USA. E-mail: hkohn@email.unc.edu; Fax: +1 919-
966-0204; Tel: +1 919-843-8112
† Electronic supplementary information (ESI) available: 1H and 13C NMR
spectra of (R)-3 and (S)-3, Table S1 providing the relative intensities of
the CRMP2-containing band and other proteins bands versus an internal
reference protein, Fig. S1 (effect of TCEP on azide reduction in 3), Fig. S2
(treatment of GST-CRMP2 supplemented lysate with either (R)-3 or (S)-
3), Fig. S3 (treatment of GST-CRMP2 supplemented lysate with (R)-3
without and with excess (R)-1), Fig. S4 (effect of GSH on Cu(I)-mediated
cycloaddition reactions). See DOI: 10.1039/c000987c
Our initial report used the lacosamide AB&CR agent (R)-2
and (S)-2.7 We showed the preferential enantiospecific adduction
of CRMP2, a potential lacosamide target, by (R)-2 compared
with (S)-2. This stereochemical finding was in agreement with the
observation that the principal anticonvulsant activity of 1 resided
in the (R)-stereoisomer.1 The AB unit in 2 is the electrophilic
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2803–2813 | 2803
©

View Article Online
aryl isothiocyanate moiety.8,9 Isothiocyanates display an excellent
balance between stability and reactivity; they do not readily react
with hydroxylic solvents (permitting their dissolution in aqueous
solutions) but do react with nucleophiles (e.g., thiols, amines)
under appropriate conditions.10
amino acid residues within a binding pocket either as a nitrene
or as an electrophile and where protein adduction is expected to
occur on a shorter time scale than an isothiocyanate group. Third,
we used an alkyne as the CR group. We showed that replacing the
O-methoxy group in (R)-1 with an O-propargyl unit to give (R)-5
led to little loss of anticonvulsant activity (see Pharmacological
Activity).7
The aryl azide group has been extensively utilized as pho-
toaffinity agents in target adduction experiments.11,12 Significantly,
the aryl azide moiety is similar in size and shape to an aryl
isothiocyanate group.13 In this study, we report the synthesis and
whole animal pharmacological data for the lacosamide AB&CR
aryl azide agents, (R)-3 and (S)-3. We compare the utility of both
2 and 3 for their ability to identify (R)-1 targets in the mouse brain
soluble proteome and for their use in in situ proteomic studies. Our
findings show that while (R)-3 exhibited enhanced anticonvulsant
activity over (R)-2, (R)-2 was superior in selectively modifying
such potential targets of lacosamide as CRMP2. We document
that the reactivity of the AB unit is a critical determinant in
proteomic searches and further report the limitations of aryl azide
photoprobes in in situ investigations.
Finally, the bioorthogonal adduction of the CR alkyne group
in 3 with the rhodamine azide probe 619 under Cu(I)-mediated
cycloaddition conditions led to little background protein adduc-
tion by 6,7 thus making the alkyne unit an ideal CR moiety for
proteomic target studies.19
B. Synthesis of (R)-3 and (S)-3
Preparation of (R)-3 and (S)-3 followed the same general pro-
cedure developed for (R)-2 and (S)-2,7 except for the final
step. In brief, we started with the enantiomerically pure serine
methyl ester 7, N-tritylation followed by sequential mesylation
and treatment with a base gave the corresponding aziridine 10
(Scheme 2). Successive N-deprotection of the trityl group (TFA)
and acetylation (acetyl chloride, Et3N) gave N-acetyl aziridine 11
that was stereospecifically ring-opened20–22 with propargyl alcohol
and BF3·Et2O to afford 12. Hydrolysis of the serine methyl ester 12
followed by mixed anhydride coupling23 with 4-aminobenzylamine
(15) gave 16. Using the method of Moses and coworkers,24
treatment of 16 with tert-butyl nitrite and TMSN3 yielded 3.
We documented the enantiopurity of (R)-3 and (S)-3 using the
chiral NMR resolving agent (R)-(-)-mandelic acid and 1H NMR
spectroscopy.25
Results
A. Choice of lacosamide AB&CR agents
Several factors led to our selection of lacosamide AB&CR agent
3, which contained an aryl azide group as the AB moiety and
an alkyne moiety as the CR group. First, the azide moiety in 3
is similar in size and shape to the isothiocyanate group in 2.13
We previously showed that (R)-2 exhibited appreciable seizure
protection in the maximal electroshock (MES) seizure test14 in
rodents (see Pharmacological Activity).7 The MES test is a well-
established animal model for identifying drug candidates that have
potential human efficacy to treat generalized and partial seizures
and that are secondarily generalized. Thus, we did not expect
the replacement of the isothiocyanate moiety by the azide group
to impede binding to the (R)-1 receptor(s). Indeed, we recently
reported that N-(4-azido)benzyl 3-methoxypropionamide ((R)-4)
exhibited superb activity in the MES test (see Pharmacological
Activity).15 Second, aryl azides are among the most common
photoaffinity groups wherein activation occurs under 254–360 nm
UV light.11,12 The photochemistry of the aryl azides is complex
and leads to the generation of a short-lived singlet nitrene that
can undergo bond insertion, conversion to the ground state
triplet nitrene, and rearrangement to a strained dehydroazepine
(a ketenimine) that is rapidly trapped by nucleophiles.16–18 The
lifetime of the dehydroazepine is ~1 ms, in the absence of
nucleophiles.17 Thus, photoactivated aryl azides can react with
C. Pharmacological activity
In Table 1, we list the anticonvulsant activities determined at
the National Institute of Neurological Disorders and Stroke’s
(NINDS) Anticonvulsant Screening Program (ASP) for AB&CR
2 and 3, AB 4, and CR 5, using the procedures described by
Stables and Kupferberg.26 Where possible, we provide the activities
for these compounds in the MES,14 6 Hz psychomotor,27 and
neurological toxicity28 models, and compare them with AEDs (R)-
1,1 phenytoin, phenobarbital, and valproate.29 We observed that
(R)-3 displayed excellent seizure protection in the MES test in
both mice (intraperitoneally, ip) (MES ED50 = 20 mg kg-1) and
2804 | Org. Biomol. Chem., 2010, 8, 2803–2813
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 2 Synthesis of lacosamide AB&CR 3.
rats (orally, po) (MES ED50 = 12 mg kg-1). When we compared 3
with 1, we found that (R)-3 exhibited lower anticonvulsant activity
than (R)-1 (mice (ip), MES ED50 = 4.5 mg kg-1; rats (po), MES
ED50 = 3.9 mg kg-1),1 but that for both 11 and 3, anticonvulsant
activity principally resided in the (R)-stereoisomer. Similarly,
kindled seizure test,30 the ED50 values for (R)-3, (R)-4,15 and (R)-
5 were 42, 6.0, and 14 mg kg-1, respectively (Table 2). Finally,
the metabolic stability of (R)-3, (R)-4 and (R)-5 were evaluated
at the NINDS ASP against a panel of human cytochrome P450
enzymes.31 Using a 500 mM concentration of (R)-3, no significant
effects were observed for the cytochrome isozymes CYP 2A6,
2B6, 2C9, 2C19, and 2E1. For CYP 1A2 (Ki = 100 mM), 2D6
(Ki = 254 mM), and 3A4 (Ki = 131 mM), inhibition was observed.
We saw less cytochrome isozyme inhibition with (R)-4 and (R)-5.
Using a 500 mM (R)-4, no significant effects were observed for the
cytochrome isozymes CYP 1A2, 2A6, 2B6, 2C9, 2C19, 2D6 and
2E1. For CYP 3A4, a Ki of ~340 mM was determined. Similarly,
when (R)-5 was evaluated, no effects at 500 mM were found when
we used CYP 1A2, 2A6, 2C9, 2C19, 2D6 and 2E1. For CYP 3A4,
weak inhibition (Ki = 280 mM) was observed. We concluded that
(R)-4 (mice (ip), MES ED50 = 8.4 mg kg-1; rat (po), MES ED50
=
3.9 mg kg-1)15 and (R)-5 (mice (ip), MES ED50 = 16 mg kg-1;
rat (po), MES ED50 = 7.9 mg kg-1)7 exhibited exceptional seizure
protection while their (S)-counterparts, (S)-4 and (S)-5, did not.
When we compared (R)-3 with (R)-2 in mice (ip) ((R)-2: MES
ED50 = 45 mg kg-1), we found that (R)-3 was about two-fold
more potent. The superb activities of (R)-3, (R)-4, and (R)-5 war-
ranted further pharmacological testing. In the psychomotor 6 Hz
(32 mA) seizure test,27 (R)-3 and (R)-5 displayed ED50 values of 27
and 29 mg kg-1,15 respectively. In the sensitive rat (ip) hippocampal
Table 1 Pharmacological activities of select AB&CR, AB, and CR agents in the maximal electroshock and 6 Hz seizure modelsa
Mice (ip)b
Rat (po)c
Compd.
No.
MES, ED50/
Tox, TD50/
mg kg-1e
6 Hz, ED50/
MES, ED50/
Tox, TD50/
mg kg-1g
R
X
Stereo mg kg-1d
PIf mg kg-1 (32 mA) mg kg-1d
PIf
2h
3
CH2C CH
NCS (R)
(S)
N3 (R)
(S)
N3 (R)
(S)
45 [1] (41–48)
>300
20 [0.25] (18–22)
110 [0.25] (76–150) 2.4 53 [0.5] (42–68) >30 [0.5]
>30 [0.5]
NDi
>500
≡
>300
82 [0.25] (63–108)
NDi
NDi
≡
CH2C CH
4.1 27 [0.25] (19–37) 12 [1] (9.6–16)
>41
>64
>63
>130
160 [0.25] (110–230) > 300 [0.25]
NDi
> 30 [1]
> 30 [1]
4j
5h
1k
CH3
8.4 [0.25] (5.7–12)
>100, <300 [0.5]
16 [0.25] (13–19)
>300 [0.5]
4.5 [0.5] (3.7–5.5)
>100, <300
9.5 [2] (8.1–10)
22 [1] (15–23)
270 [0.25] (250–340) 430 [0.25] (370–450) 1.6
46 [0.25] (41–52)
>300 [0.5]
59 [0.25] (55–66)
>300 [0.5]
27 [0.25] (38–47)
>300
5.5 NDi
3.9 [0.5] (2.5–6.2) >250 [0.5]
>30 [0.5] >30 [0.5]
3.7 29 [0.5] (21–40) 7.9 [0.5] (4.7–11) > 500 [0.5]
>100 [0.5] >30 [0.5] >30 [0.5]
6.0 10 3.9 [0.5] (2.9–5.5) >500 [0.5]
>100 [0.5]
≡
CH2C CH
H
(R)
(S)
(R)
(S)
CH3
H
>30
>30
m
Phenytoinl
66 [2] (53–72)
69 [0.5] (63–73)
6.9
3.2
30 [4] (22–39)
9.1 [5] (7.6–12)
490 [0.5] (350–730) 280 [0.5] (190–350) 0.6
>100
6.7
Phenobarbitall
Valproatel
61 [0.5] (44–96)
a The compounds were tested through the auspices of the NINDS ASP. b The compounds were administered intraperitoneally. Numbers in parentheses
are 95% confidence intervals. The dose effect data was obtained at the “time of peak effect” (indicated in hours in the square brackets). c The compounds
were administered orally. d MES = maximal electroshock seizure test. e Tox = neurologic toxicity determined from rotorod test. f PI = protective index
(TD50/ED50). g Tox = behavioral toxicity. h Ref. 7. i ND = not determined. j Ref. 15. k Ref. 1. l Ref. 29. m No ataxia observed up to 3000 mg kg-1.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2803–2813 | 2805
©

View Article Online
Table 2 Advanced neurological test data for (R)-3, (R)-4, and (R)-5a
Preliminary hippocampal kindling test (rats)
Seizure Score
Pre-drug
After discharge duration/s
Hippocampal kindled rats (ip)
ED50 (interval)/mg kg-1
Compd. No.
R
X
Drug
Pre-drug
Drug
≡
(R)-3
CH2C CH
CH3
N3
N3
H
4–5
4
4
5
4–5
1
0
0
0
3
37–63
29–65
32–50
33–53
42–72
76
0
35
0
42 (27–63)
6.0 (3–9)
14 (9–23)
14 (9.1–18)
34 (21–45)
20 (14–28)
210 (150–280)
(R)-4
≡
(R)-5
CH2C CH
(R)-1b,c
CH3
H
Phenytoinc
Phenobarbitalc
Valproatec
51
a The compounds were tested through the auspices of the NINDS ASP. b Ref. 50 c NINDS ASP internal control data.
the whole animal pharmacology for (R)-3 mirrored that of (R)-1
and that (S)-3 served as a valued control compound.
2, but the extent of CRMP2 adduction by (R)-2 compared with
(S)-2 was only 1.5–1.8 times higher.7 We attributed the modest
enantioselectivity differences for the in vitro CRMP2 adduction
to several factors.7 Among these were that the in vivo MES ED50
values are a composite of numerous pharmacological factors (e.g.,
metabolism, transport, efflux) not present in the in vitro experiment
and that different (R)-1 targets may have different stereochemical
preferences for the 2 enantiomers. The third protein selection
parameter is assessed by determining if target modification by
the AB&CR agent is competitively blocked by including excess
(R)-1 in the reaction solution.
In this study, we asked whether photoactivated AB&CR 3
would provide a protein adduction profile similar to that observed
for AB&CR 2, and whether the enhanced reactivity of the
photoactivated species generated from 3 compared with 2 would
markedly affect the parameters used to identify potential (R)-1
targets. We asked these questions because reagents that are more
reactive are generally less selective.32 To test whether our protein
selection criteria are affected by the reactivity of the AB moiety, we
compared the CRMP2-containing band adduction that appears
~62 kDa (arrow, lower band)7 with other proteins in the mouse
brain soluble lysate. Two laboratories have reported that this
protein is selectively adducted in the lysate by (R)-1 analogues.5–7
We expect that (R)-1 will preferentially bind to several proteins
in the brain but not all would be important for drug function.
Accordingly, the physiological significance of the (R)-1–CRMP2
interaction has not been determined. Thus, if our notions are
correct, AB&CR 3 may show decreased CRMP2 adduction levels
compared with non-specific and abundant proteins, a decrease in
preferential (R)-3 versus (S)-3 CRMP2 adduction, and diminished
competitive inhibition for (R)-3 adduction of target protein(s) by
excess (R)-1 when compared with (R)-2.
D. Investigation of mouse brain soluble proteome adduction with
3 and comparison with 2
AB&CR agents 2 and 3 are structurally similar in size and shape.13
Isothiocyanate 2 acts as an electrophilic, affinity-trapping agent,
while in azide 3 photoactivation is required and protein modifi-
cation occurs either by a nitrene insertion pathway or through
electrophilic trapping of the rearranged azepine ketenimine.16–18
Reports indicate that both nitrene insertion and ketenimine
adduction are more rapid17 than isothiocyanate addition reactions,
which is consistent with the notion that photoactivated azide is a
more reactive species than the corresponding isothiocyanate. Thus,
the mechanism of protein adduction, the inherent reactivities of
the AB unit, and the times required for adduction for AB&CRs 2
and 3 are all different.
We expected that the differences in isothiocyanate 2 and
azide 3 would affect three diagnostic parameters that are useful
in distinguishing selective from non-selective protein adduction
(Scheme 1). These are the preferential adduction of (R)-1-specific
binding targets versus nonspecific and abundant proteins; the
enantioselectivity of (R)-AB&CR versus (S)-AB&CR protein
adduction; and the affect of excessive (R)-1 on the (R)-AB&CR
protein adduction process. The first parameter is visualized in
the in-gel fluorescence experiments by determining the relative
intensities of the fluorescently-labeled bands and comparing
them to the intensities of protein bands in the corresponding
Coomassie blue-stained gels. The second parameter is determined
by comparing the in-gel fluorescent bands for the (R)-AB&CR
with the (S)-AB&CR reactions. A hallmark of (R)-1 function
is that the principal anticonvulsant activity resided in the D-
enantiomer ((R)-stereoisomer) compared with the L-isomer ((S)-
stereoisomer).1 For 1 in the MES test in mice (ip), this difference
exceeded 22-fold, for (R)-2 it was >6.6-fold, and for (R)-3, it was 8-
fold (Table 1). Indeed, using the mouse brain soluble proteome, we
reported that (R)-2 modified CRMP2 to a greater extent than (S)-
In Fig. 1A, we provide the in-gel fluorescence results for mouse
brain soluble proteome adduction using (R)-2, (S)-2, (R)-3, and
(S)-3. The lysate samples were passed through a Nap-10 column
prior to use. For compound 3, we activated the photoprobe
using 320 nm light (2 min, 4 ◦C), while protein adduction by 2
was promoted by incubation at 22 ◦C (30 min). The modified
2806 | Org. Biomol. Chem., 2010, 8, 2803–2813
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
and coworkers have shown that protein denaturation leads to a
reduction in selective AB protein modification.37 These collective
findings are consistent with the preferential adduction of CRMP2
by (R)-2.
Using (R)-3 and (S)-3, in place of (R)-2 and (S)-2, we observed
lower levels of CRMP2 modification compared with other ad-
ducted proteins in the lysate, and found no measurable differences
in the (R)-3 compared with the (S)-3 CRMP2 adduction levels
(Fig. 1A, lanes 5, 6). Little differences in the in-gel fluorescence
gel patterns were observed for (R)-3 and (S)-3 in the denatured
lysate reactions (Fig. 1A, lanes 7, 8), except for diminished levels
of the labeled protein bands observed at ~30 and 32 kDa. Since the
extent of CRMP2 modification by (R)-3 was small, we added to
the mouse brain soluble lysate GST-CRMP27 (2.4 mM) and then
either (R)-3 or (S)-3 (Fig. S2 in the ESI†). Again, we did not see
any preferential modification of GST-CRMP2 by (R)-3 compared
with (S)-3. In contrast, we observed higher levels of GST-CRMP2
adduction by (R)-2 compared with (S)-2 in the GST-CRMP2
supplemented lysate, as we previously reported.7 We attributed the
differences between the 2 and the 3 adduction profiles in Fig. 1 to
the enhanced reactivity of the photoprobe intermediates generated
from 3, compared with 2. Thus, our results are in agreement with
the general finding that more reactive agents are less selective than
less reactive agents.32
Next, we investigated whether potential (R)-1 targets could be
detected by conducting a competition experiment using excess
(R)-1. We reasoned that if a target is selectively adducted by a
lacosamide AB&CR agent then co-administration of (R)-1 could
reduce the extent of the adduction observed for the AB&CR agent
in the in-gel fluorescence gel. Accordingly, we added either 700
or 3500 equivalents of (R)-1 to the (R)-2 and (R)-3 adduction
experiments that used the mouse brain soluble proteome. We
observed a dose-dependent reduction of the ~62 kDa band (arrow,
corresponding to CRMP2) labeled by (R)-2 as the concentration
of (R)-1 was increased from 0 to 700 to 3500 equiv. (Fig. 2A,
lanes 1–3). After normalization of the ~62 kDa band versus the
internal ~40 kDa reference protein band (asterisk), we found a
33% intensity decrease of the ~62 kDa band with 700 equiv.
of (R)-1, and a 61% decrease with 3500 equiv. (Table S1 in the
ESI†). By comparison, we observed little changes in the level of
modification for other protein bands (i.e., ~34 kDa, ~37 kDa)
when increasing amounts of (R)-1 were added to the reaction.
When (S)-1 was used in place of (R)-1, the extent of competition
dropped significantly (Fig. 2A, lanes 5–7). We estimated only a 4%
decrease in the ~62 kDa band intensity with 700 equiv. of (S)-1
and a 19% decrease with 3500 equiv. of (S)-1, and again other
proteins bands were largely unaffected (Table S1 in the ESI†).
The competition experiments were repeated using (R)-3. We
detected no significant intensity loss of the weak band at ~62 kDa
in the (R)-3 reactions, as the amount of either (R)-1 (Fig. 2B, lanes
1–3) or (S)-1 (Fig. 2B, lanes 5–7) in the reaction was increased
from 0 to 3500 equiv. (Table S1 in the ESI†). A similar finding was
observed for the GST-CRMP2 (3.6 mM) supplemented lysate upon
addition of (R)-1 (700, 3500 equiv.) (Fig. S3 in the ESI†). Thus,
the (R)-1 competition experiments with (R)-2 and (R)-3 provided
different signatures for the selective CRMP2 adduction process.
For the photoactivated agent (R)-3, which generates reactive
intermediates, no apparent competition by (R)-1 was detected, but
it was seen in the less-reactive isothiocyanate agent (R)-2, albeit
Fig. 1 Proteome reactivity profiles of lacosamide AB&CR isothio-
cyanate (NCS) agent 2 and lacosamide AB&CR azide (N3) agent 3. A:
Probe-labeled proteins were detected by in-gel fluorescence scanning after
Cu(I)-mediated cycloaddition to a rhodamine-azide reporter probe 6. An
arrow marks selectively labeled protein with 2 identified as CRMP27 and
an asterisk marks an internal reference protein band. Proteins observed
using heat-denatured lysate (ꢀ) were considered non-specifically labeled.
B: CRMP2 protein was detected by western blot using anti-CRMP2.
C: Proteins in lysate visualized by Coomassie blue staining after in-gel
fluorescence scanning. (All images shown in grayscale).
lysates were treated with 6 using Cu(I)-mediated cycloaddition
conditions (CuSO4, TCEP, 60 min).33–35 We independently showed
that the TCEP concentrations (1 mM) used for cycloaddition
reactions rapidly reduced the azide group in 3 to the corresponding
4-amino adduct 16 (<30 min) (Fig. S1 in the ESI†),36 while the
isothiocyanate moiety in 2 was largely unaffected by this reaction
(data not shown). Thus, protein modification by 3 presumably
occurred within minutes at 4 ◦C, while modification by 2 may
have occurred over 90 min. We observed preferential modification
of the ~62 kDa band for (R)-2 compared with (S)-2 (Fig. 1A,
lanes 1, 2; arrow), as previously reported.7 Using a protein band
(~40 kDa, asterisk) as an internal standard, we observed that the
in-gel fluorescence for the ~62 kDa was ~82% of the ~40 kDa
band for the (R)-2 reaction and ~45% for the (S)-2 reaction after
normalization. The relative increase in the fluorescence intensity
has been attributed to the preferential adduction of CRMP2 by
(R)-2 versus (S)-2, and CRMP2 modification is supported by
western blot analysis with anti-CRMP2 (Fig. 1B). Significantly,
when the lysate sample was denatured (75 ◦C, 5 min, 0.5% Triton
X100) prior to use, diminished amounts of the ~62 kDa band
adduction, versus other protein bands (Fig. 1A, lanes 3, 4), were
detected, and no difference was observed in the levels of (R)-2
versus (S)-2 adduction, further confirming that CRMP2 adduction
was specific. For both (R)-2 and (S)-2, the ~62 kDa band was 28–
29% the level of the ~40 kDa band after normalization. Cravatt
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2803–2813 | 2807
©

View Article Online
293T cells. Accordingly, we inc◦ubated cultured human 293T cells
with either (R)-2 or (R)-3 (37 C in the dark, 2 h) and then the
(R)-3-treated cells were irradiated with 312 nm light (4 ◦C, 10 min).
The (R)-2-treated cells were lysed, centrifuged to remove cellular
debris, and the supernatant treated with 6 under Cu(I)-mediated
cycloaddition conditions. The (R)-3 cells were similarly treated,
except that the recovered supernatant was equally divided and then
one of the two aliquots was re-irradiated at 312 nm (4 ◦C, 5 min)
before both aliquots were treated with 6. We observed extensive
protein modification for (R)-2 by in-gel fluorescence (Fig. 3A, lane
1) but essentially none for (R)-3 (Fig. 3A, lanes 2, 3), even though
the corresponding Coomassie blue-stained gels for total proteins
were similar (Fig. 3B).
Fig. 3 Proteome reactivity profiles of lacosamide AB&CR (R)-2 and
(R)-3 using 293T cells. A: Probe-labeled proteins were detected by
in-gel fluorescence scanning after Cu(I)-mediated cycloaddition to a
rhodamine-azide reporter probe 6. B: Proteins in lysate visualized by
Coomassie blue staining after in-gel fluorescence scanning. (All images
shown in grayscale).
Fig. 2 Competition experiments using lacosamide AB&CR isothio-
cyanate (NCS) agent (R)-2 and lacosamide AB&CR azide (N3) agent (R)-3
with either excess (R)-1 or (S)-1. A: Proteins detected by in-gel fluorescence
scanning after adduction with (R)-2 in the presence of excess 1 (room
temperature, 20 min) followed by Cu(I)-mediated cycloaddition with 6. An
arrow marks selectively labeled protein with 2 identified as CRMP27 and an
asterisk marks an internal reference protein band. B: Proteins detected by
in-gel fluorescence scanning after adduction with (R)-3 in the presence of
excess 1 (312 nm, 2 min, 4 ◦C) followed by Cu(I)-mediated cycloaddition
with 6. An arrow marks selectively labeled protein with 2 identified as
CRMP27 and an asterisk marks an internal reference protein band.
We briefly explored the difference between (R)-2 and (R)-3
results using 293T cells. Aryl azides are known to be reduced
by thiols,38 and glutathione (GSH) is a major constituent in the
cell (low mM).39 In addition, photolysis of aryl azides generates in-
termediates that can react with intracellular thiols.17 Accordingly,
we added GSH (0.1 mM, 0.5 mM) to the Nap-10-treated mouse
brain soluble proteome used in Fig. 1 and determined the relative
levels of protein adduction for 10 mM (R)-2 and 10 mM (R)-3.
For (R)-2, we observed no significant effect of 0.1 and 0.5 mM
GSH on the protein adduction profiles (Fig. 4, lanes 5–7). For
(R)-3, however, when we included 0.1 mM GSH in the lysate
we found a dramatic reduction in the protein band intensities,
which were further diminished with 0.5 mM GSH (Fig. 4, lanes
1–3). We tested whether GSH reacted with aryl azides. Using 0.5–
10 mM GSH, we observed no appreciable reduction of (R)-4 (TLC
analysis, data not shown). Next, we determined whether GSH
affected the Cu(I)-mediated cycloaddition reaction with 6. Our
finding that (R)-2 provided appreciable levels of protein adduction
in 293T cells (Fig. 4, lane 1) suggested that this was not likely.
Nonetheless, we added GSH (0.1 M, 0.5 mM) to the mouse brain
soluble lysate reactions after AB&CR 3 adduction. There was
no noticeable intensity loss for the in-gel fluorescence bands upon
reaction with 6 (Fig. S4, lanes 1–3 in the ESI†). However, when the
GSH concentration was raised to 2.5 mM a significant intensity
reduction of the in-gel fluorescence bands was observed (Fig. S4,
at high levels of (R)-1 (700–3500 equiv.). The high levels of (R)-1
needed in these experiments could be attributed to the irreversible
adduction process that follows the initial binding event (Scheme 1).
We expect that covalent modification of the target protein by (R)-1
AB&CR drives the reaction to completion, thus minimizing the
effect of competing amounts of (R)-1 during the initial binding
event.
In proteomic search for drug targets, it is important to identify
the proteins of interest within the cell. The significant anticonvul-
sant activities of (R)-2 and (R)-3 in rodents (Table 1) demonstrated
that both compounds cross into the central nervous system (CNS)
upon administration (ip, po) and likely enter neuronal cells. We
asked whether the AB&CR agents (R)-2 and (R)-3 were suitable
for in situ proteome experiments using readily accessible human
2808 | Org. Biomol. Chem., 2010, 8, 2803–2813
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
We first showed that (R)-3 exhibited excellent seizure protection
in both mice (ip) and rats (po) and that similar to 1 and 2,1,7
anticonvulsant activity principally resided in the (R)-stereiosomer,
compared with the corresponding (S)-isomer (mice (ip): 8-fold).
(R)-3 displayed a ~2-fold increase in potency compared with (R)-
2 in the MES test (mice, ip). In vitro liver enzyme assays using
(R)-3 and the corresponding AB agent (R)-4 showed that the aryl
azide moiety was not readily metabolized. These collective findings
suggested that (R)-3 may serve as an excellent AB&CR agent in
proteomic searches for (R)-1 binding targets.
Using 3, we found low adduction levels for the ~62 kDa band
(Fig. 1A, lanes 5, 6) containing CRMP2 (Fig. 1B), as determined
by in-gel fluorescence gel, and that there was little difference
between the (R)-3 and the (S)-3 experiments. Correspondingly,
higher levels of adduction of the ~62 kDa band were observed
for 2, and (R)-2 preferentially modified this band compared with
(S)-2 (Fig. 1A, lanes 1, 2). Protein denaturation of the soluble
lysate prior to AB&CR modification led to diminished levels of
~62 kDa protein(s) adduction by (R)-2 and a loss in the preferential
selectivity of (R)-2 versus (S)-2 protein modification (Fig. 1, lanes
3, 4). In contrast, no significant changes were observed for the (R)-
3 and (S)-3 adduction processes with protein denaturation (Fig. 1,
lanes 7, 8).
Fig. 4 Glutathione (GSH) effect on in vitro labeling with lacosamide
AB&CR (R)-2 and (R)-3 in the mouse brain soluble proteome.
lane 4†). We concluded that the GSH levels in the 293T cells did
not appreciably affect the Cu(I)-mediated cycloaddition reaction
and that this step did not account for the near total absence of
protein bands in the (R)-3 experiments. Rather, our collective
findings suggested that intracellular nucleophiles inactivated the
photoactivated 3 aryl azide intermediates, thereby preventing them
from reacting with 293T cellular proteins.
When the (R)-2 experiments were repeated in the presence
of excess (R)-1, we observed a significant loss in the in-gel
fluorescence intensity of the ~62 kDa band (Fig. 2A, lanes 1–3).
Diminished competition was observed when (S)-1 was substituted
for (R)-1 (Fig. 2A, lanes 5–7). These findings are consistent with
the notion that (R)-1 binds to the same binding pocket in CRMP2
as (R)-2 does and that (R)-1 can block protein modification by
(R)-2. When the competition experiment was conducted using
(R)-3, no intensity loss of the ~62 kDa band by excess (R)-1
was observed (Fig. 2B, lanes 1–3). Similar results were obtained
when GST-CRMP2 was added to the lysate (Fig. S3, ESI†). Thus,
the less-reactive AB&CR agent (R)-2 was more sensitive to (R)-1
competition than its more reactive counterpart (R)-3.
Together, these results provided valuable information on the
AB properties needed to identify protein binding targets for (R)-
1, a ligand for which protein binding is expected to be modest.
We found that the less reactive aryl isothiocyanate AB moiety
exhibited better enantioselectivity for protein modification and
proved to be more sensitive in competition experiments with excess
(R)-1 than the more reactive aryl azide AB group.
We compared the potential use of AB&CR agents 2 and 3 for
in situ proteomic search experiments. Using human 293T cells,
we observed significant labeling with (R)-2 but not with (R)-
3 (Fig. 3A, lanes 1–3). Whole animal pharmacological studies
demonstrated that both compounds exhibited significant seizure
protection (Table 1), and therefore, we have assumed both agents
can enter 293T cells. We briefly explored why (R)-3 was ineffective
in labeling proteins under the irradiation conditions we used.
The mouse brain soluble proteome shown in Fig. 1 and 2 was
passed through a Nap-10 column prior to use. This size exclusion
column preferentially removes small molecules from the lysate.
One such molecule is likely to be GSH. Thiols are known to react
with aryl azides,37 and they are also expected to react with the
aryl azide intermediates generated by photolysis.17 We showed
that progressively adding GSH (0.1–0.5 mM) to the Nap-10-
treated lysate steadily decreased the amounts of protein adduction
Discussion
We have advanced a general approach to investigating the
proteome for ligands (drugs) whose binding is expected to be
modest (Scheme 1).7 Key to the implementation of this approach
is selecting an AB unit that irreversibly modifies the target site.
The AB group should not perturb drug binding, be sufficiently
stable to permit dissolution and protein binding, yet sufficiently
reactive to irreversibly modify the target. While the importance of
these factors for selective protein adduction is recognized, little is
known how AB reactivity can affect the protein selection process.
Aryl isothiocyanates8,9 and aryl azides11,12 are established affinity
baits used in receptor tagging and identification studies. While
structurally similar,13 their methods of activation, pathways of
protein adduction, and inherent reactivity may impact their use
in proteomic searches. As an electrophilic affinity bait, the isoth-
iocyanate group displays an excellent balance between stability
and reactivity, thus allowing dissolution in aqueous solutions but
permitting reaction with nucleophilic amino acids (e.g., cysteine,
lysine).10 Accordingly, isothiocyanates have proven effective in
selectively modifying target proteins within brain membrane
fractions8,9 and have been used in in vivo protein adduction studies
after intracerebroventricular (icv) administration.40 Similarly, aryl
azides have been extensively used in protein labeling studies.11,12
Indeed, an aryl azide analog of (S)-levetiracetam was employed to
identify the synaptic vesicle protein SV2A as the binding site for
this AED.41,42
Using the AB&CR strategy, we showed that the lacosamide
AB&CR agent (R)-2, containing the electrophilic isothiocyanate
group, preferentially modified CRMP2, a protein previously
reported to bind (R)-1.5–7 In this study, we explored the utility
of 3, an isostere of 2, to identify lacosamide binding targets in
the mouse brain soluble proteome fraction, and we compared 3
with 2.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2803–2813 | 2809
©

View Article Online
observed upon (R)-3 photoactivation (Fig. 4, lanes 1–3). Control
experiments indicated that GSH did not noticeably reduce the
azide group in the (R)-3. Similar results for other aryl azides
and GSH have been reported.38a In addition, moderate levels of
GSH did not markedly affect the Cu(I)-cycloaddition reaction
with 6 under the employed conditions (Fig. S4, ESI†). These
findings indicated that (R)-3 photoactivated intermediates are
likely destroyed by intracellular constituents, thus making this
AB&CR a poor choice for identifying drug binding targets in
intact cells.
AB&CR agent, along with the agent’s pharmacological properties
and structural size, are important determinants for successful
identification of potential ligand binding partners in the proteome.
Finally, we showed that the utility of the aryl azide affinity bait
for in situ experiments can be affected by the ease with which
the photoactivated aryl azide intermediate(s) is quenched by
intracellular thiols.
Experimental section
The findings for our (R)-3 proteomic labeling studies need to be
placed in the context of the anticipated (R)-3-protein interaction.
Lacosamide is a low molecular weight drug (MW = 250) whose
protein(s) binding is anticipated to be modest. Using (R)-3, we ob-
served no preferential protein labeling in the mouse brain soluble
lysate. We have attributed this result to the modest binding strength
of the (R)-3–protein binding interaction and to the reactivity of
the (R)-3 photoprobe intermediates. Correspondingly, we expect
that aryl azide-containing ligands43 that strongly bind to proteins
will lead to selective adduction, provided that nucleophiles in the
lysate or in the cell do not destroy the photoactivated species.
We anticipate that lysate preparations that have been passed
through size-exclusion columns, extensively washed, or dialyzed
will provide higher yields of photoactivated aryl azide protein
adducts than those that have not been pretreated. It should be
noted that the selective adduction of CRMP2 by (R)-2 in the
mouse brain soluble lysate does not, by itself, document that this
protein is a physiological target for (R)-1. We suspect that several
proteins can bind (R)-144 and that target validation will require
detailed functional and pharmacological studies.
General methods
Melting points were determined in open capillary tubes using a
Thomas-Hoover melting point apparatus and are uncorrected.
Infrared spectra (IR) were run on an ATI Mattson Genesis FT-
IR spectrometer. Absorption values are expressed in wavenum-
bers (cm-1). Optical rotations were obtained on a Jasco P-1030
polarimeter at the sodium D line (589 nm) using a 1 dm path
length cell and are given in units of deg cm3 g-1 dm-1. NMR
spectra were obtained at 300 MHz (1H) and 75 MHz (13C)
using TMS as an internal standard. Chemical shifts (d) are
reported in parts per million (ppm) from tetramethylsilane. Low-
resolution mass spectra were obtained with a BioToF-II-Bruker
Daltonics spectrometer by Drs Matt Crowe and S. Habibi at
the University of North Carolina Department of Chemistry.
The high-resolution mass spectrum was performed on a Bruker
Apex-Q 12 Telsa FTICR spectrometer by Drs Matt Crowe and
S. Habibi. Microanalyses were performed by Atlantic Microlab,
Inc. (Norcross, GA). Reactions were monitored by analytical
thin-layer chromatography (TLC) plates (Aldrich, Cat # Z12272-
6) and analyzed with 254 nm light. The reaction mixtures
were purified by flash column chromatography using silica gel
(Dynamic Adsorbents Inc., Cat #02826-25). All chemicals and
solvents were reagent grade and used as obtained from commercial
sources without further purification. THF was distilled from
blue sodium benzophenone ketyl. Yields reported are for purified
products and were not optimized. All compounds were checked
by TLC, 1H and 13C NMR, MS, and elemental analyses. The TLC,
NMR and the analytical data confirmed the purity of the products
was ≥95%.
Conclusions
Our findings for 2 and 3 have provided insights into the utility and
limitations of the aryl isothiocyanate and aryl azide affinity baits,
two functional groups commonly employed in target identification
studies.8–12 In our studies, we have incorporated either an elec-
trophilic isothiocyanate ((R)-2) or a photoactivated ((R)-3) affinity
bait on the (R)-1 framework. The two affinity baits are structurally
similar in size and shape,13 but differ in their reactivity and where
photoactivation of the aryl azide group in (R)-3 leads to a more
reactive agent than the isothiocyanate group in (R)-2. (R)-1 is a low
molecular weight drug that is expected to bind to its receptors (not
necessary at the active site) in the brain with only modest affinity.
Accordingly, our investigation differs from activity-based protein
profiling (ABPP) methodologies,33b,c,37,46–48 that utilize mechanism-
based small molecules to identify active enzymes within the
proteome, and from studies that incorporate affinity baits on
ligands that bind tightly to their cognate receptors.49 In these
investigations, protein adduction is expected to be efficient. We
showed that despite the two-fold improved anticonvulsant activity
of (R)-3 compared with (R)-2, (R)-2 was superior in revealing
potential binding targets in the mouse brain soluble proteome.
Moreover, the incorporation of an isothiocyante group in 2
allowed us to document the enantioselective adduction of CRMP2
by (R)-2 compared with (S)-2 within the proteome, and that
(R)-2 adduction was competitively blocked by excess amount
of (R)-1. Thus, our findings demonstrated that in a case where
ligand binding is modest, the reactivity of the AB unit within the
(R)-N-(4-Azidobenzyl) 2-acetamido-3-(prop-2-
ynyloxy)propionamide ((R)-3)
◦
To a cooled (0 C) CH3CN solution (20 mL) of (R)-167
(1.04 g, 3.60 mmol) maintained under Ar, t-BuONO (1.28 mL,
10.80 mmol) followed by TMSN3 (1.14 mL, 8.64 mmol) were
slowly added. The resulting solution was allowed to stir at room
temperature (16 h). The solvent was removed in vacuo, and
the product was purified by column chromatography (SiO2; 1 : 9
acetone–EtOAc) to give 0.87 g (76%) of (R)-3 as a yellowish
◦
white solid: mp 135–136 C; [a]2D4 -15.7 (c 1.0, CHCl3); Rf 0.40
(1 : 9 acetone–EtOAc); IR (nujol mull) 3277, 2927, 2110, 1635,
1553, 1459 cm-1; 1H NMR (CDCl3) d 2.04 (s, CH3C(O)), 2.46 (t,
J = 2.5 Hz, CH2CCH), 3.64 (dd, J = 7.1, 9.3 Hz, CHH¢OCH2),
3.93 (dd, J = 4.2, 9.3 Hz, CHH¢OCH2), 4.15 (1/2HH¢q, J = 2.5,
16.0 Hz, OCHH¢C), 4.23 (1/2HH¢q, J = 2.5, 16.0 Hz, OCHH¢C),
4.37-4.51 (m, CH2Ar), 4.57-4.63 (m, CH), 6.52 (br d, J =
6.3 Hz, NHCH), 6.75-6.80 (br m, NHCH2), 6.96-7.01 (m, 2 ArH),
2810 | Org. Biomol. Chem., 2010, 8, 2803–2813
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
7.25-7.27 (m, 2 ArH), addition of excess (R)-(-)-mandelic acid to
a CDCl3 solution of (R)-3 gave only a single signal for the acetyl
methyl protons and the alkyne proton, addition of excess (R)-(-)-
mandelic acid to a CDCl3 solution of (R)-3 and (S)-3 (1 : 2 ratio)
gave two signals for the acetyl methyl protons (d 2.006 (R) and
2.021 (S) (Dppm = 0.015)), and two signals for the alkyne proton
(d 2.402 (S) and 2.448 (R) (Dppm = 0.046)) in a ~1 : 2 ratio; 13C
NMR (CDCl3) d 23.4 (CH3C(O)), 43.2 (CH2Ar), 52.7 (CH), 58.9
(CH2CCH), 69.2 (CH2OCH2), 75.6 (CH2CCH), 79.0 (CH2CCH),
119.5, 129.2, 134.8, 139.5 (C6H4), 169.9, 170.6 (2 C(O)); HRMS
(ESI) Calcd. for C15H18N5O3 (MH+) 316.1410. Found 316.1407.
Found: C, 57.30; H, 5.46; N, 22.10. C15H18N5O3 requires C, 57.13;
H, 5.43; N, 22.21%.
(pH 7.8)) were treated with either (R)-2 or (S)-2 (10 mM) without
or with competing ligands ((R)-1, (S)-1) at room temperature
(20 min). The modified lysates were sequentially treated with
rhodamine reporter tag 6 (50 mM), TCEP (1 mM), TBTA (100 mM)
and CuSO4 (1 mM). Samples were shaken and allowed to rotate
using Roto-shake (8 rpm, Scientific Industries Inc., Model No.
SI-1100, Bohemia, NY) at room temperature (1 h). Proteins were
separated by SDS-PAGE after addition of 4¥ SDS-PAGE loading
buffer and visualized by in-gel fluorescence using a typhoon 9400
scanner (GE Healthcare/Amersham Bioscience) with excitation
at 555 nm and detection at 580 nm.
AB&CR 3 labeling, cycloaddition reaction, and in-gel fluorescence
scanning
(S)-N-(4-Azidobenzyl) 2-acetamido-3-(prop-2-
ynyloxy)propionamide ((S)-3)
Utilizing the preceding procedure, lysate aliquots (50 mL of 2.0 mg
mL-1 protein in 50 mM HEPES buffer (pH 7.8)) were treated
with either (R)-3 or (S)-3 (10 mM) without or with competing
ligands ((R)-1, (S)-1) at 4 ◦C (10 min) and irradiated with 312 nm
light (8 W, Spectroline EB-280C, Spectronics Corp., New York,
USA) at 4 ◦C (2 min). The reaction mixtures were quenched
with TCEP (1 mM). The modified lysates were sequentially
treated with rhodamine reporter tag 6 (rhodamine-azide (Rho-
N3) (50 mM)), TBTA (100 mM) and CuSO4 (1 mM). Samples
were shaken and allowed to rotate using Roto-shake (8 rpm,
Scientific Industries Inc., Model No. SI-1100, Bohemia, NY) at
room temperature (1 h). Proteins were separated by SDS-PAGE
after addition of 4¥ SDS-PAGE loading buffer and visualized
by in-gel fluorescence using a typhoon 9400 scanner (Amersham
Bioscience) with excitation at 532 nm and detection at 580 nm.
Utilizing the preceding procedure, and using (S)-167 (1.10 g,
3.81 mmol), t-BuONO (1.36 mL, 11.42 mmol), TMSN3 (1.20 mL,
9.14 mmol) gave 0.89 g (74%) of (S)-3 as a yellowish white
◦
solid: mp 135-136.5 C; [a]2D4 +15.4 (c 1.0, CHCl3); Rf 0.40 (1 : 9
acetone–EtOAc); IR (nujol mull) 3279, 2924, 2111, 1638, 1551,
1
1458 cm-1; H NMR (CDCl3) d 2.03 (s, CH3C(O)), 2.46 (t, J =
2.4 Hz, CH2CCH), 3.65 (dd, J = 7.2, 9.1 Hz, CHH¢OCH2), 3.92
(dd, J = 4.4, 9.1 Hz, CHH¢OCH2), 4.15 (1/2HH¢q, J = 2.4,
15.9 Hz, OCHH¢C), 4.23 (1/2HH¢q, J = 2.4, 15.9 Hz, OCHH¢C),
4.36-4.50 (m, CH2Ar), 4.58-4.64 (m, CH), 6.48 (br d, J = 6.9 Hz,
NHCH), 6.80-6.84 (br m, NHCH2), 6.96-7.00 (m, 2 ArH), 7.24-
7.28 (m, 2 ArH), addition of excess (R)-(-)-mandelic acid to a
CDCl3 solution of (S)-3 gave only a single signal for the acetyl
methyl protons and the alkyne proton, addition of excess (R)-(-)-
mandelic acid to a CDCl3 solution of (R)-3 and (S)-3 (1 : 2 ratio)
gave two signals for the acetyl methyl protons (d 2.006 (R) and
2.021 (S) (Dppm = 0.015)), and two signals for the alkyne proton
(d 2.402 (S) and 2.448 (R) (Dppm = 0.046)) in a ~1 : 2 ratio; 13C
NMR (CDCl3) d 23.4 (CH3C(O)), 43.2 (CH2Ar), 52.7 (CH), 58.9
(CH2CCH), 69.3 (CH2OCH2), 75.6 (CH2CCH), 79.1 (CH2CCH),
119.5, 129.2, 134.8, 139.5 (C6H4), 169.9, 170.6 (2 C(O)); HRMS
(ESI) Calcd. for C15H18N5O3 (MH+) 316.1410. Found 316.1406.
Found: C, 56.99; H, 5.45; N, 22.10. C15H17N5O3 requires C, 57.13;
H, 5.43; N, 22.21%.
In situ proteome reactivity profiling
293T cells were grown to ~50% confluency in DMEM supple-
mented with 10% fetal bovine serum in T75 flask. Medium was
changed, and DMSO solution (200 mL, 5 mM) of either (R)-2
or (R)-3 was added directly to 10 mL medium in the flask. After
incubating (37 ◦C in the dark, 2 h), the (R)-3 treated cells were
irradiated with 312 nm light (4 ◦C, 10 min). Both (R)-2 and (R)-3
treated cells were washed with PBS (¥3) and harvested by scraping.
Soluble proteome extracts were prepared by needle sonication of
cell pellets in 50 mM HEPES buffer (pH 7.8) and centrifugation
(100 000 ¥ g for 1 h at 4 ◦C). The supernatant of (R)-2 treated cells
was treated with 6 under Cu(I)-mediated cycloaddition conditions.
The supernatant of (R)-3 treated cells was first equally divided, and
Preparation of mouse brain soluble proteomes
Mouse brains (male, 2 months, ICR mice [Rockland Im-
munochemicals, Gilbertsville, PA]) were Dounce-homogenized in
50 mM HEPES buffer (pH 7.4). The◦lysate was centrifuged at
slow speed (1200 ¥ g for 12 min at 4 C) to remove debris. The
supernatant was then centrifuged at high speed (100 000 ¥ g for
1 h at 4 ◦C). The resulting supernatant was collected and stored at
-80 ◦C until use. The total protein concentration was determined
by using the Bradford assay.
◦
then one of the two aliquots was re-irradiated at 312 nm (4 C,
5 min) before both aliquots were treated with 6. Proteins were
separated by SDS-PAGE after addition of 4¥ SDS-PAGE loading
buffer and visualized by in-gel fluorescence using a typhoon 9400
scanner (Amersham Bioscience) with excitation at 532 nm and
detection at 580 nm.
Pharmacology
AB&CR 2 labeling, cycloaddition reaction, and in-gel fluorescence
scanning
Compounds were screened under the auspices of the National
Institutes of Health’s Anticonvulsant Screening Program. Exper-
iments were performed in male rodents [albino Carworth Farms
No. 1 mice (intraperitoneal route, ip), albino Spague-Dawley
rats (oral route, po)]. Housing, handling, and feeding were in
accordance with recommendations contained in the ‘Guide for
Mouse brain lysate (1 mL, 50 mM HEPES buffer (pH 7.4)) was
passed through a Nap-10 column (GE Healthcare) to exchange
buffer to an aqueous 50 mM HEPES buffer (pH 7.8). Lysate
aliquots (50 mL of 2.0 mg mL-1 protein in 50 mM HEPES buffer
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2803–2813 | 2811
©

View Article Online
the Care and Use of Laboratory Animals’. Anticonvulsant activity
was established using the MES test,14 6 Hz,26 hippocampal kindled
seizure,29 and the scMet test,25 according to previously reported
methods.45
19 (a) A. E. Speers and B. F. Cravatt, Chem. Biol., 2004, 11, 535–546;
(b) K. D. Park, R. Liu and H. Kohn, Chem. Biol., 2009, 16, 763–772.
20 M. Bucciarelli, A. Forni, I. Moretti, F. Prati and G. Torre, Tetrahedron:
Asymmetry, 1995, 6, 2073–2080.
21 U. Larsson and R. Carlson, Acta Chem. Scand., 1994, 48, 511–516.
22 M. Belohradsky, L. Ridvan and J. Zavada, Collect. Czech. Chem.
Commun., 2003, 68, 1319–1328.
Acknowledgements
23 G. W. Anderson, J. E. Zimmerman and F. M. Callahan, J. Am. Chem.
Soc., 1967, 89, 5012–5017.
The authors thank the NINDS and the Anticonvulsant Screening
Program (ASP) at the National Institutes of Health with Drs Tracy
Chen and Jeffrey Jiang for kindly performing the pharmacological
studies via the ASP’s contract site at the University of Utah with
Drs H. Wolfe, H.S. White, and K. Wilcox. Funds for this study were
generously provided by grant R01NS054112 (H.K., R.L.) from the
National Institute of Neurological Disorders and Stroke, and by
the Korean Research Foundation Grant funded by the Korean
Government (MOEHRD) grant KRF-2006-352-C00042 (KDP).
The content is solely the responsibility of the authors and does not
necessarily represent the official views of the National Institute of
Neurological Disorders and Stroke or the National Institutes of
Health. Harold Kohn has a royalty-stake position in (R)-1.
24 K. Barral, A. D. Moorhouse and J. E. Moses, Org. Lett., 2007, 9,
1809–1811.
25 For comparable procedures for resolving stereoisomers, see the follow-
ing: (a) G. R. Weisman, in Asymmetric Synthesis-Analytical Methods,
J. D. Morrison, Academic Press, New York, 1983, vol. 1, pp. 153-171.;
(b) D. Parker and R. J. Taylor, Tetrahedron, 1987, 43, 5451–5456.
26 J. P. Stables and H. G. Kupferberg, in Molecular and Cellular Targets
for Antiepileptic Drugs, ed. G. Avanzini, P. Tanganelli and M. Avoli,
John Libbey, London, 1977, pp 191–198.
27 M. E. Barton, B. D. Klein, H. H. Wolff and H. S. White, Epilepsy Res.,
2001, 47, 217–227.
28 (a) N. W. Dunham and T.-S. Miya, J. Am. Pharm. Assoc., 1957, 46,
208–209; (b) H. S. White, J. H. Woodhead, K. S. Wilcox, J. P. Stables,
H. J. Kupferberg and H. H. Wolf, in Antiepileptic Drugs, ed. R. H. Levy,
R. H. Mattson, B. S. Meldrum and E. Perruca, Lippincott, Williams
and Wilkins, Philadelphia, PA, 5th edn, 2002, pp 36-48.
29 R. J. Porter, J. J. Cereghino, G. D. Gladding, B. J. Hessie, H. J.
Kupferberg, B. Scoville and B. G. White, Cleveland Clin. Q., 1984,
51, 293–305.
References
30 (a) K. Morimoto, M. Fahnestock and R. J. Racine, Prog. Neurobiol.,
2004, 73, 1–60; (b) E. W. Lothman and J. M. Williamson, Brain Res.,
1994, 649, 71–84; (c) E. W. Lothman, R. A. Salerno, J. B. Perlin and
D. L. Kaiser, Epilepsy Res., 1988, 2, 367–379; (d) E. W. Lothman, J. B.
Perlin and R. A. Salerno, Epilepsy Res., 1988, 2, 356–366.
31 (a) L. L. von Moltke, D. J. Greenblatt, S. X. Duan, J. Schmider,
L. Kudchadker, S. M. Fogelman, J. S. Harmatz and R. I. Shader,
Psychopharmacology, 1996, 128, 398–407; (b) R. Pearce, D. Greenway
and A. Parkinson, Arch. Biochem. Biophys., 1992, 298, 211–225;
(c) E. L. Code, C. L. Crespi, B. W. Penman, F. J. Gonzalez, T. K.
Chang and D. J. Waxman, Drug Metab. Dispos., 1997, 25, 985–993;
(d) T. Leemann, D. Transon and P. Dayer, Life Sci., 1993, 52, 29–34;
(e) T. Shimada, K. S. Misono and F. P. Guengerich, J. Biol. Chem.,
1986, 261, 909–921; (f) T. Kronbach, D. Mathys, J. Gut, T. Catin and
U. A. Meyer, Anal. Biochem., 1987, 162, 24–32; (g) R. Peter, R. Bocker,
P. H. Beaune, M. Iwasaki, P. P. Guengerich and C. S. Yang, Chem. Res.
Toxicol., 1990, 3, 566–573; (h) F. P. Guengerich, M. V. Martin, P. H.
Beaune, P. Kremers, T. Wolff and D. J. Waxman, J. Biol. Chem., 1986,
261, 5051–5060.
1 D. Choi, J. P. Stables and H. Kohn, J. Med. Chem., 1996, 39, 1907–1916.
2 E. Perucca, U. Yasothan, G. Clincke and P. Kirkpatrick, Nat. Rev. Drug
Discovery, 2008, 7, 973–974.
3 A. C. Errington, T. Stoehr, C. Heers and G. Lees, Mol. Pharmacol.,
2008, 73, 157–169.
4 P. L. Sheets, C. Heers, T. Stoehr and T. R. Cummins, J. Pharmacol.
Exp. Ther., 2008, 326, 89–99.
5 J. M. Freitag, B. Beyreuther, C. Heers and T. Stoehr, Epilepsia, 2007,
48, 320. Suppl 8, 72-73.
6 B. K. Beyreuther, J. Freitag, C. Heers, N. Krebsfaenger, U. Scharfe-
necker and T. Stoehr, CNS Drug Rev., 2007, 13, 21–42.
7 K. D. Park, P. Morieux, C. Salome, S. W. Cotten, O. Reamtong, C.
Eyers, S. J. Gaskell, J. P. Stables, R. Liu and H. Kohn, J. Med. Chem.,
2009, 52, 6897–6911.
8 B. R. de Costa, R. B. Rothman, V. Bykov, A. E. Jacobson and K. C.
Rice, J. Med. Chem., 1989, 32, 281–283.
9 B. R. de Costa, A. H. Lewin, K. C. Rice, P. Skolnick and J. A.
Schoenheimer, J. Med. Chem., 1991, 34, 1531–1538.
10 (a) A. E. Takemori and P. S. Portoghese, Annu. Rev. Pharmacol.
Toxicol., 1985, 25, 193–223; (b) J. G. Robertson, Biochemistry, 2005,
44, 5561–5571; (c) C. Drahl, B. F. Cravatt and E. J. Sorensen, Angew.
Chem., Int. Ed., 2005, 44, 5788–5809.
32 (a) D. L. Boger, A. Santillan Jr., M. Searcey, S. R. Brunette, S. E.
Wolkenberg, M. P. Hedrick and Q. Jin, J. Org. Chem., 2000, 65, 4101–
4111; (b) K. S. MacMillan and D. L. Boger, J. Med. Chem., 2009, 52,
5771–5780.
33 (a) E. Saxon and C. R. Bertozzi, Science, 2000, 287, 2007–2010; (b) A. E.
Speers and B. F. Cravatt, ChemBioChem, 2004, 5, 41–47; (c) H. Ovaa,
P. F. van Swieten, B. M. Kessler, M. A. Leeuwenburgh, E. Fiebiger,
A. M. C. H. Van Den Nieuwendijk, P. J. Galardy, G. A. Van Der
Marel, H. L. Ploegh and H. S. Overkleeft, Angew. Chem., Int. Ed.,
2003, 42, 3626–3629; (d) Y. Kho, S. C. Kim, C. Jiang, D. Barma, S. W.
Kwon, J. Cheng, J. Jaunbergs, C. Weinbaum, F. Tamanoi, J. Falck and
Y. Z h a o , Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 12479–12484.
34 V. V. Rostovtsev, L. G. Green, V. V. Fokin and K. B. Sharpless, Angew.
Chem., Int. Ed., 2002, 41, 2596–2599.
11 (a) H. Bayley and J. V. Staros, in Azides and Nitrenes: Reactivity
and Utility, Academic Press, Orlando, 1984, Ch. 9, pp. 433-490.;
(b) G. Bucher, in CRC Handbook of Organic Photochemistry and
Photobiology, ed. W. Horspool and F. Lenci, CRC Press, Boca Raton,
2nd edn, 2003, Ch. 44, pp 1-31.
12 (a) G. H. Hur, J. L. Meier, J. Baskin, J. A. Codelli, C. R. Bertozzi, M. A.
Marahiel and M. D. Burkart, Chem. Biol., 2009, 16, 372–381; (b) J.
Gubbens, E. Ruijter, L. E. V. de Fays, J. M. A. Damen, B. de Kruijff,
M. Slijper, D. T. S. Rijkers, R. M. J. Liskamp and A. I. P. M. de Kroon,
Chem. Biol., 2009, 16, 3–14.
35 C. W. Tornøe, C. Christensen and M. Meldal, J. Org. Chem., 2002, 67,
13 (a) M. F. Budyka, Russ. Chem. Rev., 2008, 77, 709–723; (b) P. Mani, H.
Umamaheswari, B. D. Joshua and N. Sundaraganesan, THEOCHEM,
2008, 863, 44–49.
3057–3064.
36 For a similar study, see: A.-M. Faucher and C. Grand-Maitre, Synth.
Commun., 2003, 33, 3503–3511.
37 (a) K. T. Barglow and B. F. Cravatt, Chem. Biol., 2004, 11, 1523–1531;
(b) A. E. Speers, G. C. Adam and B. F. Cravatt, J. Am. Chem. Soc.,
2003, 125, 4686–4687.
38 (a) J. V. Staros, H. Bayley, D. N. Standring and J. R. Knowles, Biochem.
Biophys. Res. Commun., 1978, 80, 568–572; (b) A. L. Handlon and N. J.
Oppenheimer, Pharm. Res., 1988, 5, 297–299.
39 R. Paroni, E. De Vecchi, G. Cighetti, C. Arcelloni, I. Fermo, A. Grossi
and P. Bonini, Clin. Chem., 1995, 41, 448–454.
14 R. H. Levy, R. Mattson and B. Meldrum, Antiepileptic Drugs, Raven
Press, New York, 4th edn, 1995, ch. 6, pp. 99-110.
15 C. Salome, E. Salome-Grosjean, K. D. Park, P. Morieux, R. Swendi-
man, E. DeMarco, J. P. Stables and H. Kohn, J. Med. Chem., 2010, 53,
1288–1305.
16 S. Mandel, J. Liu, C. M. Hadad and M. S. Platz, J. Phys. Chem. A,
2005, 109, 2816–2821.
17 C. J. Shields, D. R. Chrisope, G. B. Schuster, A. J. Dixon, M. Poliakoff
and J. J. Turner, J. Am. Chem. Soc., 1987, 109, 4723–4726.
18 O. L. Chapman and J. P. Le Roux, J. Am. Chem. Soc., 1978, 100, 282–
285.
40 B. D. De Costa, R. B. Rothman, V. Bykov, L. Band, A. Pert, A. E.
Jacobson and K. C. Rice, J. Med. Chem., 1990, 33, 1171–1176.
2812 | Org. Biomol. Chem., 2010, 8, 2803–2813
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
41 B. A. Lynch, N. Lamberg, K. Nocka, P. Kensel-Hammes, S. M.
Bajjalieh, A. Matagne and B. Fuks, Proc. Natl. Acad. Sci. U. S. A.,
2004, 101, 9861–9866.
42 B. Fuks, M. Gillard, P. Michel, B. Lynch, P. Vertongen, P. Leprince, H.
Klitgaard and P. Chatelain, Eur. J. Pharmacol., 2003, 478, 11–19.
43 For a recent discussion on the potential use of aryl azides in in situ
studies, see: H. Baruah, S. Puthenveetil, Y.-A. Choi, S. Shah and A. Y.
Ting, Angew. Chem., Int. Ed., 2008, 47, 7018–7021.
46 M. J. Evans and B. F. Cravatt, Chem. Rev., 2006, 106, 3279–3301.
47 (a) A. M. Sadaghiani, S. H. L. Verhelst and M. Bogyo, Curr. Opin.
Chem. Biol., 2007, 11, 20–28; (b) S. H. L. Verhelst, M. Fonovic and M.
Bogyo, Angew. Chem., Int. Ed., 2007, 46, 1284–1286.
48 (a) K. A. Kalesh, L. P. Tan, K. Lu, L. Gao, J. Wang and S. Q. Yao,
Chem. Commun., 2010, 46, 589–591; (b) P.-Y. Yang, K. Liu, M. H. Ngai,
M. J. Lear, M. R. Wenk and S. Q. Yao, J. Am. Chem. Soc., 2010, 132,
656–666.
44 C. Temperini, A. Innocenti, A. Scozzafava, S. Parkkila and C. T.
Supuran, J. Med. Chem., 2010, 53, 850–854.
49 A. L. MacKinnon, J. L. Garrison, R. S. Hegde and J. Taunton, J. Am.
Chem. Soc., 2007, 129, 14560–14561.
45 A. LeTiran, J. P. Stables and H. Kohn, J. Med. Chem., 2002, 45, 4762–
4773.
50 T. Stoehr, H. J. Kupferberg, J. P. Stables, D. Choi, R. H. Harris, H.
Kohn, N. Walton and H. S. White, Epil. Res., 2007, 74, 147–154.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2803–2813 | 2813
©