Structure-activity relationship of semicarbazone EGA furnishes photoaffinity inhibitors of anthrax toxin cellular entry

Article abstract of DOI:10.1021/ml400486k

EGA, 1, prevents the entry of multiple viruses and bacterial toxins into mammalian cells by inhibiting vesicular trafficking. The cellular target of 1 is unknown, and a structure-activity relationship study was conducted in order to develop a strategy for target identification. A compound with midnanomolar potency was identified (2), and three photoaffinity labels were synthesized (3-5). For this series, the expected photochemistry of the phenyl azide moiety is a more important factor than the IC₅₀ of the photoprobe in obtaining a successful photolabeling event. While 3 was the most effective reversible inhibitor of the series, it provided no protection to cells against anthrax lethal toxin (LT) following UV irradiation. Conversely, 5, which possessed weak bioactivity in the standard assay, conferred robust irreversible protection vs LT to cells upon UV photolysis.

Full text of DOI:10.1021/ml400486k

Letter
pubs.acs.org/acsmedchemlett
Structure−Activity Relationship of Semicarbazone EGA Furnishes
Photoaffinity Inhibitors of Anthrax Toxin Cellular Entry
Michael E. Jung,†,‡,* Brian T. Chamberlain,†,‡ Chi-Lee C. Ho,§ Eugene J. Gillespie,§
and Kenneth A. Bradley†,§
‡
§
†California NanoSystems Institute, Department of Chemistry and Biochemistry, Department of Microbiology, Immunology and
Molecular Genetics, University of California, Los Angeles, Los Angeles, California 90095, United States
S
* Supporting Information
ABSTRACT: EGA, 1, prevents the entry of multiple viruses and
bacterial toxins into mammalian cells by inhibiting vesicular
trafficking. The cellular target of 1 is unknown, and a structure−
activity relationship study was conducted in order to develop a
strategy for target identification. A compound with midnanomo-
lar potency was identified (2), and three photoaffinity labels were
synthesized (3−5). For this series, the expected photochemistry
of the phenyl azide moiety is a more important factor than the
IC50 of the photoprobe in obtaining a successful photolabeling
event. While 3 was the most effective reversible inhibitor of the
series, it provided no protection to cells against anthrax lethal toxin (LT) following UV irradiation. Conversely, 5, which
possessed weak bioactivity in the standard assay, conferred robust irreversible protection vs LT to cells upon UV photolysis.
KEYWORDS: Photoaffinity labeling, semicarbazone, aryl azide, anthrax lethal toxin, endosomal trafficking
common mechanism by which viruses and bacterial toxins
Aenter host cells involves trafficking from an early to a late
endosome followed by release into the cytoplasm as triggered
by endosomal acidification.1−3 Inhibition of this process
represents a host-targeted strategy for the development of
broad-spectrum antiviral and antibacterial drugs. Recently,
EGA, N1-(4-bromobenzylidene)-N4-(2,6-dimethylphenyl)-
semicarbazone, 1, was identified from a high-throughput
phenotypic screen to protect cells from anthrax lethal toxin
(LT).4 Further investigations showed that 1 protected cells
from multiple viruses and bacterial toxins that rely on pH-
dependent endosomal trafficking to enter cells. Although the
details of the biochemical mechanism have been studied, the
cellular target(s) remain unknown. We have conducted a
structure−activity relationship (SAR) study with the goal of
improving the potency of 1 and developing a strategy for
photoaffinity labeling and identification of the involved
proteins.
A series of derivatives of 1 were synthesized and tested,
revealing precise requirements for activity in a tight and
relatively flat structure−activity landscape. The 2,6-dimethyl
substitution in the N4-ring as well as an unaltered semi-
carbazone core were shown to be optimal for activity while
certain modifications to the N1-phenyl ring were tolerated.
From the SAR, a compound more potent than the original hit
was generated, namely the N1-2-fluoro-4-bromophenyl ana-
logue (IC50 = 0.4 μM), 2. In addition, three photoaffinity
probes, 3−5, two of which possess low micromolar activity,
were designed and synthesized. Our key finding is that the
expected photochemistry of the phenyl azide moiety is a more
important factor than the IC50 of the photoprobe in achieving
successful photolabeling. While 3 was the most potent inhibitor
of intoxication by LT in the standard assay, it did not provide
protection to the cells upon UV photolysis. Conversely 5,
which possesses two fluorine atoms ortho to the azido function,
was a poor inhibitor in the standard assay yet provided cells
with the most effective irreversible protection against LT upon
UV irradiation. These results represent significant progress
toward our goal of identifying the cellular target of this series of
compounds and highlight useful considerations for photo-
affinity labeling studies in general.
Prior to launching the structure−activity relationship study,
we confirmed that the intact semicarbazone structure of 1 was
the entity responsible for the inhibition of membrane
trafficking. In a previous study, semicarbazones derived from
aldehyde-based Cathepsin K inhibitors were shown to have
poor CN bond stability, which, along with other evidence,
led researchers to conclude that the semicarbazones were
functioning as prodrugs delivering a bioactive aldehyde.5 On
the other hand, structurally distinct peptide-semicarbazones
were shown to be stable in acidic media, requiring reflux to
induce decomposition.6 In our case, incubating RAW 264.7
macrophages with the semicarbazide (6) and the benzaldehyde
component of 1 (both individually and in combination) prior
to addition of LT did not prevent toxin-induced cell death,
Received: November 27, 2013
Accepted: December 26, 2013
Published: January 21, 2014
© 2014 American Chemical Society
363
dx.doi.org/10.1021/ml400486k | ACS Med. Chem. Lett. 2014, 5, 363−367

ACS Medicinal Chemistry Letters
Letter
which indicates that the complete semicarbazone structure is
required for bioactivity (Supporting Information, Figure 1).
The structure−activity relationship study began by evaluating
the importance of the N4-2,6-dimethylphenyl moiety for the
bioactivity of 1. Several N-substituted semicarbazides (6, 7a−c)
were synthesized by reaction of hydrazine with an intermediate
phenyl carbamate,7 or via direct addition of hydrazine with tert-
butyl isocyanate (8). Condensation of these semicarbazides
with 4-bromobenzaldehyde proceeded well in all examples, and
the desired products (1, 9a−c, and 10) were obtained pure by
recrystallization (Scheme 1). The stereochemistry of the
semicarbazone double bond is assigned as E based on analogy
with known benzaldehyde semicarbazone structures.8−12
Table 2. Inhibition of LT-Induced Cell Death by 1, 2, 12−
13ab, 15, and 28
Scheme 1. Synthesis of 1, 9a−c, and 10
Reagents and conditions: (a) ClCO2Ph, Et3N, DCM, rt; (b) NH2NH2·
H2O, DCM, rt; (c) 4-BrC6H4CHO, EtOH, HOAc, reflux. 6 = 2,6-
dimethylphenyl semicarbazide, 7a = 2,4-dimethylphenyl semicarba-
zide, 7b = 2,4,6-trimethylphenyl semicarbazide, 7c = 2,6-diethylphenyl
semicarbazide, 8 = tert-butyl.
Cell viability data demonstrated the critical importance of the
2,6-dimethylphenyl moiety for the inhibition of membrane
trafficking [For the less active analogues, precise IC50 values
were not determined due to poor solubility above 25 μM.
These compounds are characterized by an activity limit (i.e.,
>12.5 μM, >25 μM) or as “not protective” if no bioactivity was
observed up to 25 μM] (Table 1). Thus, compounds lacking
Table 1. Inhibition of LT-Induced Cell Death by 1, 9a−c, 10,
and 11
entry
R
IC50 (μM)
1
2,6-(Me)2C6H3
2,4-(Me)2C6H3
2,4,6-(Me)3C6H2
2,6-(Et)2C6H3
t-Bu
1.4 0.4
NP
9a
9b
9c
10
11*
12.6 2.2
>12.5
NP
Ph
NP
*
N1-(4-Cl-benzylidene)-N4-phenylsemicarbazone 11 was obtained
commercially and tested previously.4 Although compound 11 has a
4-Cl substituent rather than a 4-Br one, that difference should not
cause a large change in potency (compare activity of 13a to that of 1,
Table 2). NP = not protective.
this motif (9a, 10, and 114) were not protective, and a 2,6-
diethylphenyl moiety (9c) displayed significantly diminished
bioactivity (IC50 >12.5 μM). Additionally, the N4-2,4,6-
trimethylphenyl analogue 9b was approximately 10-fold less
active than 1. Although we did not systematically investigate the
activity of analogues with modifications at all the positions of
the N4-phenyl ring, these results demonstrate the limited
potential for modifications in this position of 1.
*
Compound was obtained commercially and tested previously.4 NP =
not protective.
Examination of the N1-position began with the synthesis and
testing of ten compounds prepared from a variety of
benzaldehydes (13a−j) (Table 2). The bioactivity supported
the hypothesis from preliminary SAR work (i.e., 12)4 that
substitution in the 4-position was critical for potency. Next,
nine compounds prepared from 4-substituted benzaldehydes
were examined in order to give more detailed characterization
364
dx.doi.org/10.1021/ml400486k | ACS Med. Chem. Lett. 2014, 5, 363−367

ACS Medicinal Chemistry Letters
Letter
of that critical position (13k−s). Although no compound from
this set was more potent than 1, substitution in this position
was found to be generally well tolerated, with 9 of the 13
compounds featuring a benzylidene motif substituted only in
the 4-position registering an IC50 between 1 and 4 μM. Five
analogues consisting of heterocyclic rings were also synthesized.
While 13t−13w did not offer protection to cells challenged
with LT, the 5-bromothiophen-2-ylmethylene example (13x)
possessed bioactivity similar to that of 1 (IC50 1.7 μM and 1.4
μM, respectively).
The thiosemicarbazones (19a−c) were synthesized from N-
2,6-dimethylphenyl thiosemicarbazide, 18, and the correspond-
ing benzaldehyde in refluxing ethanol and acetic acid13
(Scheme 3). Substituting the carbonyl of 1 with a thiocarbonyl,
Scheme 3. Synthesis and Bioactivity of 19a−c and 22
The fact that the N1-2,4-difluoro analogue (13f) was notably
more potent than the analogue from 4-fluorobenzaldehyde
(13e) (IC50 = 2.1 μM vs IC50 = 7.0 μM) led us to examine
whether including a fluorine in the 2-position of the N1-ring
would similarly augment the potency of 1. Indeed, compound 2
displayed a significant increase in activity (IC50 = 0.4 μM).
Encouraged by this result, we synthesized and tested
compounds 13y−13ab, which gave IC50 values that were all
larger than the original lead compound 1.
Reagents and conditions: (a) CS2, NaOH, DMF; (b) NH2NH2·H2O,
DMF, 60 °C; (c) 4-BrC6H4CHO, EtOH, HOAc, reflux; (d) MeI,
EtOH, reflux, Na2CO3; (e) NH2NH2·H2O, EtOH, HOAc, reflux; (f)
4-BrC6H4CHO, EtOH, HOAc, reflux.
The effects of modifications made to the semicarbazone core
are compiled in Scheme 2. The imine bond of 1 could be
Scheme 2. Synthesis and Bioactivity of 14, 16, and 17
19a, had little effect on the potency of the compound (IC50
=
1.5 μM) while exacerbating an already problematic solubility
profile. For example, compounds 19b and 19c were poorly
soluble at assay concentrations (cLogP: 19a, 5.8; 19b, 6.0; 19c,
6.1; 1, 5.2) and did not return meaningful dose−response
curves. A significantly more soluble CNH compound (22)
(cLogP: 3.9−4.0) was elaborated from the S-methylation of
(2,6-dimethylphenyl)thiourea,14 20, to give 21, followed by
displacement with hydrazine and condensation with 4-
bromobenzaldehyde to give the hydrazinecarboximidamide,
22. Unfortunately, the bioactivity of this compound was
diminished (IC50 = 13.8 μM).
From the structure−activity relationship data, it was evident
that the most effective photoaffinity probe would contain an
N4-2,6-dimethylphenyl unit, an unmodified semicarbazone
core, and a 4-substituted benzylidene motif at the N1-position.
As the course of phenyl azide photolysis is known to be
sensitive to the substituents of the phenyl azide,15,16 three
photoaffinity labels were designed. The first, 3, would be
predicted by the SAR to provide the highest potency in the in
vitro assay. The second, 4, would consist of a simple phenyl
azide commonly employed in the biochemical literature.17 The
third molecule, 5, would flank the azido group with fluorines, a
modification known to give a longer-lived singlet nitrene that
can more effectively yield genuine insertion products upon
photolysis.18,19
Compounds 3, 4, and 5 were synthesized from 6 and the
desired azido-containing benzaldehyde (4-azidobenzaldehyde
(25), 4-azido-2-fluorobenzaldehdye (26), and 4-azido-2,3,5,6-
tetrafluorobenzaldehyde (27)), in the usual manner. Com-
pound 25 could be obtained from the ethylene acetal of 4-
nitrobenzaldehyde via catalytic hydrogenation and diazotization
followed by reaction with sodium azide.20 In this study we
chose to perform an Ullmann-type coupling with 4-iodobenzyl
alcohol (23) and sodium azide21 followed by oxidation to the
aldehyde (Scheme 4). This protocol was convenient on the
milligram scale and provided access to the previously
undescribed 26 from 4-bromo-2-fluorobenzyl alcohol, 24.
Compound 27 was synthesized according to a known
procedure via an SNAr reaction of sodium azide with
pentafluorobenzaldehyde.18
Reagents and conditions: (a) BH3·THF, 50 °C; (b) MeI, K2CO3,
DMF, rt; (c) BrCH2CCH, K2CO3, DMF, rt; (d) MeNHNH2·H2O,
EtOH, reflux; (e) Ph 2,6-Me2phenyl carbamate, Et3N, DCM, rt.
effectively reduced to the disubstituted hydrazine derivative
(14) using an excess of borane in THF with heating. The
compound was not protective at the concentrations tested. The
1-(4-bromophenyl)ethylidene analogue (15) was synthesized
by the condensation of the semicarbazide 6 with 4′-
bromoacetophenone. This change caused a minor reduction
in potency, with 15 showing an IC50 of 2.5 μM (Table 2). The
semicarbazone core of 1 could be methylated with high
regioselectivity by treatment with iodomethane and K2CO3 in
DMF to give 16. Synthesis of 16 via addition of methyl
hydrazine to 4-bromobenzaldehyde followed by reaction with
phenyl (2,6-dimethylphenyl)carbamate gave a product with
1
identical H NMR and 13C NMR spectra and thus confirmed
the regiochemistry of methylation at the N2-position. The
bioactivity of 16 was not significantly affected compared to that
of 1 (IC50 = 2.0 μM and 1.4 μM, respectively). Incorporating a
propargyl group into this position by alkylation with propargyl
bromide generated 17, which had somewhat impaired
bioactivity (IC50 > 12.5 μM).
365
dx.doi.org/10.1021/ml400486k | ACS Med. Chem. Lett. 2014, 5, 363−367

ACS Medicinal Chemistry Letters
Letter
in the N4-phenyl ring, it is not apparent how such a “clickable”
linker could be included into a bioactive photoaffinity version
of 1. Nevertheless, the convenience of this approach led us to
synthesize compound 28, an N2-propargyl analogue of 4 (Table
2). This compound did not offer any protection to cells at any
concentration tested. Accordingly, radiolabeled versions of 5
are being considered as a means of identifying the bound
proteins.
Scheme 4. Synthesis and Bioactivity of Photoaffinity Probes
3 and 4
In conclusion, we have conducted a structure−activity
relationship study that has examined how modifications in
the structure of 1 affect the cellular entry of LT. The data
indicate a tight and relatively flat SAR, with many of the original
structural features of 1 being preferred for bioactivity. Inclusion
of a fluorine in the 2-position of the N1-benzylidene portion
appears to be a general strategy for increasing the potency of
these arylsemicarbazones and, in the case of 2, improved
potency to midnanomolar levels. The photoaffinity probes 3, 4,
and 5 were synthesized from azido-containing benzaldehydes.
Photolysis of 4 and 5 in the presence of RAW 264.7 cells
conferred irreversible resistance to LT and implies that covalent
bonds with the cellular target were generated. We are in the
process of designing and preparing radiolabeled versions of
these compounds to facilitate target identification.
Reagents and conditions: (a) NaN3, CuI, MeHNCH2CH2NHMe, Na
ascorbate, DMSO/H2O; (b) PCC, DCM, rt; (c) 6, EtOH, HOAc,
reflux.
As predicted, 3 and 4 were inhibitors of LT membrane
trafficking, with 4 having an IC50 of 2.8 μM and 3 giving an IC50
of 2.2 μM, while 5 was only weakly protective against LT,
beginning to show a biological effect at 25 μM. The reversibility
of inhibition by 3−5 was confirmed by incubating cells with
inhibitor-containing media and then exchanging it for fresh
media prior to intoxication with LT. As expected, the
arylsemicarbazones could be “washed-out” and there was no
cell viability in these experiments upon exposure to LT (Figure
1). Irradiating the cell cultures incubated with 4 and 5 with 5
ASSOCIATED CONTENT
■
S
* Supporting Information
Synthetic procedures and 1H NMR, 19F NMR (when
applicable), and HRMS characterization data for all compounds
tested in bioassay. For select compounds, additional character-
ization includes 13C NMR (2, 3, 4, 13x, 14, 15, 16, 17, 19a,
28), FTIR (2−5, 13x, 28), and melting point (2, 13x). NMR
spectra are provided for select compounds. Materials and
methods for cellular intoxication assays and photolabeling
experiments are included. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: jung@chem.ucla.edu.
Notes
The authors declare no competing financial interest.
REFERENCES
Figure 1. Photoaffinity labeling of RAW264.7 cells with aryl azide
photoprobes. LT = Anthrax lethal toxin, UV = 5 min ultraviolet
(UVB) irradiation, “washout” indicates replacing cell media with fresh
media lacking 1, 3, 4, or 5 prior to intoxication with LT. Compounds 4
and 5 provided statistically significant protection (* = p < 0.05, ** = p
< 0.01) following photoirradiation vs washout.
■
(1) Di Simone, C.; Zandonatti, M. A.; Buchmeier, M. J. Acidic pH
Triggers LCMV Membrane Fusion Activity and Conformational
Change in the Glycoprotein Spike. Virology 1994, 198, 455−465.
(2) Knodler, L. A.; Celli, J.; Finlay, B. B. Pathogenic trickery:
deception of host cell processes. Nat. Rev. Mol. Cell Biol. 2001, 2, 578−
588.
(3) White, J.; Matlin, K.; Helenius, A. Cell fusion by Semliki Forest,
influenza, and vesicular stomatitis viruses. J. Cell Biol. 1981, 89, 674−
679.
(4) Gillespie, E. J.; Ho, C.-L.; Balaji, K.; Li, Z.; Thomas, D.; Clemens,
D. L.; Deng, G.; Wang, Y.; Elsaesser, H.; Tamilsevam, B.; France, B.;
Chamberlain, B. T.; Blanke, S. R.; Cheng, G.; Brooks, D.; Jung, M. E.;
Manchester, M.; Zink, J.; Colicelli, J.; Damoiseaux, R.; Bradley, K. A.
Selective Inhibitor of Endosomal Trafficking Pathways Exploited by
Multiple Toxins and Viruses. Proc. Natl. Acad. Sci. U.S.A., early ed.;
2013 (Nov 4 2013); pp 1−9.
min of ultraviolet (UVB) light prior to media exchange
conferred significant protection on cells against LT. Compound
3 did not offer statistically significant protection after UVB
photolysis and media exchange. The irreversible protection
provided to cells upon the irradiation of 4 and 5 implies that
these compounds form covalent linkages with the target protein
upon exposure to UVB light.
A common strategy for identifying proteins bound by
photoaffinity probes is to incorporate an alkynyl moiety into
the photoaffinity probe that can be conjugated with an azido-
containing fluorescent dye following cell lysis.22 Given the low
activity of 17 and the limited potential for making modifications
(5) Adkison, K. K.; Barrett, D. G.; Deaton, D. N.; Gampe, R. T.;
Hassell, A. M.; Long, S. T.; McFadyen, R. B.; Miller, A. B.; Miller, L.
R.; Payne, J. A.; Shewchuk, L. M.; Wells-Knecht, K. J.; Willard, D. H.;
Wright, L. L. Semicarbazone-based inhibitors of cathepsin K, are they
366
dx.doi.org/10.1021/ml400486k | ACS Med. Chem. Lett. 2014, 5, 363−367

ACS Medicinal Chemistry Letters
Letter
prodrugs for aldehyde inhibitors? Bioorg. Med. Chem. Lett. 2006, 16,
978−983.
(6) Leban, J.; Blisse, M.; Krauss, B.; Rath, S.; Baumgartner, R.; Seifert,
M. H. J. Proteasome inhibition by peptide-seimcarbazones. Bioorg.
Med. Chem. 2008, 16, 4579−4588.
(7) Yogeeswari, P.; Sriram, D.; Thirumurugan, R.; Raghavendran, J.
V.; Sudhan, K.; Pavana, R. K.; Stables, J. Discovery of N-(2,6-
Dimethylphenyl)-Substituted Semicarbazones as Anticonvulsants:
Hybrid Pharmacophore-Based Design. J. Med. Chem. 2005, 48,
6202−6211.
(8) Sarojini, B. K.; Narayana, B.; Bindya, S.; Yathirajan, H. S.; Bolte,
M. 4-(Methylthio)benzaldehyde semicarbazone. Acta Crystallogr., Sect.
E: Struct. Rep. Online 2007, E63, o2946.
(9) Naik, D. V.; Palenik, G. J. The crystal and molecular structures of
acetone semicarbazone and benzaldehyde semicarbazone. Acta
Crystallogr., Sect. B 1974, 30B, 2396−2401.
(10) Basuli, F.; Peng, S.-M.; Bhattacharya, S. Chemical Control on
the Coordination Mode of Benzaldehyde Semicarbazone Ligands.
Synthesis, Structure, and Redox Properties of Ruthenium Complexes.
Inorg. Chem. 2001, 40, 1126−1133.
(11) Casas, J. S.; Garcia-Tasende, M. S.; Sordo, J. Main group metal
complexes of semicarbazones and thiosemicarbazones. A structural
review. Coord. Chem. Rev. 2000, 209, 197−261.
(12) Lin, L.; Fan, W.; Chen, S.; Ma, J.; Hu, W.; Lin, Y.; Zhang, H.;
Huang, R. Photochromism of (E)-4-phenyl-1-(pyridine-2-
ylmethylene)semicarbazide. New J. Chem. 2012, 36, 2562−2567.
(13) Tripathi, L.; Kumar, P.; Singh, R.; Stables, J. P. Design, synthesis
and anticonvulsant evaluation of novel N-(4-substituted phenyl)-2-[4-
(substituted) benzylidene]hydrazinecarbothio amides. Eur. J. Med.
Chem. 2012, 47, 153−166.
(14) Rasmussen, C. R.; Villani, F. J., Jr.; Weaner, L. E.; Reynolds, B.
E.; Hood, A. R.; Hecker, L. R.; Nortey, S. O.; Hanslin, A.; Costanzo,
M. J.; Powell, E. T.; Molinari, A. J. Improved Procedures for the
Preparation of Cycloalkyl-, Arylakyl-, and Arylthioureas. Synthesis
1988, 456−459.
(15) Gritsan, N. P.; Platz, M. S. Kinetics, Spectroscopy, and
Computational Chemistry of Arylnitrenes. Chem. Rev. 2006, 106,
3844−3867.
(16) Li, Y. Z.; Kirby, J. P.; George, M. W.; Poliakoff, M.; Schuster, G.
B. 1,2-Didehydroazepines from the Photolysis of Substituted Aryl
Azides: Analysis of Their Chemical and Physical Properties by Time-
Resolved Spectroscopic Methods. J. Am. Chem. Soc. 1988, 110, 8092−
8098.
(17) Kotzyba-Hibert, F.; Kapfer, I.; Goeldner, M. Recent Trends in
Photoaffinity Labeling. Angew. Chem., Int. Ed. Engl. 1995, 34, 1296−
1312.
(18) Keana, J. F. W.; Cai, S. X. New Reagents for Photoaffinity
Labeling: Synthesis and Photolysis of Functionalized Perfluorophenyl
Azides. J. Org. Chem. 1990, 55, 3640−3647.
(19) Schnapp, K. A.; Poe, R.; Leyva, E.; Soundararajan, N.; Platz, M.
S. Exploratory Photochemistry of Fluorinated Aryl Azides. Implica-
tions for the Design of Photoaffinity Labeling Reagents. Bioconjugate
Chem. 1993, 4, 172−177.
(20) Walton, R.; Lahti, P. M. An Efficient Simple Synthesis of 4-
Azidobenzaldehyde. Synth. Commun. 1998, 28, 1087−1092.
(21) Andersen, J.; Madsen, U.; Bjorkling, F.; Liang, X. Rapid
Synthesis of Aryl Azides from Aryl Halides under Mild Conditions.
Synlett 2005, 2209−2213.
(22) Lapinsky, D. J. Tandem photoaffinity labeling−bioorthogonal
conjugation in medicinal chemistry. Bioorg. Med. Chem. 2012, 20,
6237−6247.
367
dx.doi.org/10.1021/ml400486k | ACS Med. Chem. Lett. 2014, 5, 363−367