Efficient and flexible synthesis of new photoactivatable propofol analogs

Article abstract of DOI:10.1016/j.bmcl.2021.127927

Propofol is a widely used general anesthetic, which acts by binding to and modulating several neuronal ion channels. We describe the synthesis of photoactivatable propofol analogs functionalized with an alkyne handle for bioorthogonal chemistry. Such tools are useful for detecting and isolating photolabeled proteins. We designed expedient and flexible synthetic routes to three new diazirine-based crosslinkable propofol derivatives, two of which have alkyne handles. As a proof of principle, we show that these compounds activate heterologously expressed Transient Receptor Potential Ankyrin 1 (TRPA1), a key ion channel of the pain pathway, with a similar potency as propofol in fluorescence-based functional assays. This work demonstrates that installation of the crosslinkable and clickable group on a short nonpolar spacer at the para position of propofol does not affect TRPA1 activation, supporting the utility of these chemical tools in identifying and characterizing potentially druggable binding sites in propofol-interacting proteins.

Full text of DOI:10.1016/j.bmcl.2021.127927

Bioorg. Med. Chem. Lett. 39 (2021) 127927
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Efficient and flexible synthesis of new photoactivatable propofol analogs
,
,
,
,*
Kenneth A. Skinnera 1, Joseph S. Wzorek2 b, Daniel Kahnea b, Rachelle Gaudeta
a Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA 02138, United States
b Department of Chemistry and Chemical Biology, Harvard University, Cambridge, MA 02138, United States
A R T I C L E I N F O
A B S T R A C T
Keywords:
Propofol
Propofol is a widely used general anesthetic, which acts by binding to and modulating several neuronal ion
channels. We describe the synthesis of photoactivatable propofol analogs functionalized with an alkyne handle
for bioorthogonal chemistry. Such tools are useful for detecting and isolating photolabeled proteins. We designed
expedient and flexible synthetic routes to three new diazirine-based crosslinkable propofol derivatives, two of
which have alkyne handles. As a proof of principle, we show that these compounds activate heterologously
expressed Transient Receptor Potential Ankyrin 1 (TRPA1), a key ion channel of the pain pathway, with a similar
potency as propofol in fluorescence-based functional assays. This work demonstrates that installation of the
crosslinkable and clickable group on a short nonpolar spacer at the para position of propofol does not affect
TRPA1 activation, supporting the utility of these chemical tools in identifying and characterizing potentially
druggable binding sites in propofol-interacting proteins.
TRPA1
Diazirine
Click chemistry
Propofol (2,6-diisopropylphenol) is an intravenous general anes-
thetic used in millions of surgical procedures worldwide.1 In addition to
the sedative effects it exerts on the central nervous system, propofol also
elicits a “burning” sensation upon injection in many patients.2 The
γ-aminobutyric acid A (GABAA) receptor is the principal target leading
to the anesthetic properties of propofol,3 whereas Transient Receptor
several photoaffinity analogs of propofol exist,6,7,12,13 there is a dearth
of propofol probes that also contain a chemical handle to enable sub-
sequent isolation of photolabeled protein, limiting the discovery of
novel propofol-interacting proteins. Thus, it is important to expand the
library of crosslinkable analogs to provide more tools to understand how
propofol and similar molecules, like thymol, carvacrol and zingerone,
which are plant phenols with multiple biological activities, interact with
their physiological targets.
Potential Ankyrin
1 (TRPA1), a calcium-permeable ion channel
expressed in peripheral sensory neurons, has been suggested to be the
key mediator of propofol-induced pain.4 We recently used a photo-
crosslinkable propofol analog to demonstrate that this analog does
indeed bind in the TRPA1 site identified by mutagenesis.5 Similar stra-
tegies have identified propofol-binding sites in the GABAA receptor.6,7
Also, molecular modeling and mutagenesis of the binding pocket resi-
dues corroborate photolabeled sites.8,9 These studies illustrate the
power of photoaffinity labeling (PAL) to identify the binding-site loca-
tion in complex biological systems, particularly where multiple binding
sites may exist.5,9 In addition, PAL probes may elucidate new drug tar-
gets and off-target proteins, facilitating drug discovery efforts.10,11
PAL finds widespread utility in drug discovery as a method to
identify new macromolecular targets and as a complement to structure
determination techniques like cryo-EM and X-ray crystallography to
probe the location and structure of ligand binding sites. Although
One of the ideal traits of a photoaffinity probe is structural similarity
to the parent molecule. In addition, the choice of crosslinker is impor-
tant; diazirines have favorable specific activities due to their activation
at relatively long wavelengths, the small size of the photoreactive group,
and the short half-life of the reactive carbene intermediate.14 The rapid
quenching of carbenes by reaction with water molecules minimizes
unspecific labeling, but also causes low yields of photolabeling. To
facilitate the detection and isolation of photolabeled protein, clickable
handles such as alkynes provide a convenient post-crosslinking method
to incorporate an azide-containing biotin or fluorescent tag into the
photoprobe.15 To the best of our knowledge, the only report of a
bifunctional propofol analogue has only been demonstrated to retain
biological activity for the GABAA receptor,6 and has not been tested for
other targets such as TRPA1 activation.
* Corresponding author.
E-mail address: gaudet@mcb.harvard.edu (R. Gaudet).
1
Present address: Departments of Medicine and Pharmacology, University of California, San Diego, La Jolla 92161, United States.
Present address: Novartis Institutes for BioMedical Research, Cambridge, MA 02139, United States.
2
https://doi.org/10.1016/j.bmcl.2021.127927
Received 4 December 2020; Received in revised form 24 February 2021; Accepted 26 February 2021
Available online 9 March 2021
0960-894X/© 2021 Elsevier Ltd. All rights reserved.

K.A. Skinner et al.
Bioorganic & Medicinal Chemistry Letters 39 (2021) 127927
Scheme 1. Synthesis of XPRO. Reagents and conditions: (a) Methyl vinyl ketone, Amberlyst-15, toluene, 50 ◦C, N2, 6 h, 80%; (b) TBSCl, imidazole, DMF, rt, N2, 16 h,
79%; (c) NH3, MeOH, NH2OSO3H, ꢀ 40 ◦C, N2, 7 h; TEA, iodine, MeOH, 45%; (d) TBAF-acetic acid, THF, rt, N2, 16 h, 76%.
Scheme 2. Synthesis of o-XPRO-Click. Reagents and conditions: (a) Methyl vinyl ketone, Amberlyst-15, toluene, 40 ◦C, N2, 8 h, 44%; (b) NBS, DCM, rt, N2, 16 h,
54%; (c) PPh3, PdCl2(PPh3)2, CuI, ethynyltrimethylsilane, TEA, rt, N2, 16 h, 59%; (d) TBSCl, imidazole, DMF, rt, N2, 16 h, 82%; (e) NH3, MeOH, NH2OSO3H, ꢀ 40 ◦C,
N2, 7 h; TEA, iodine, MeOH, 32%; (f) TBAF-acetic acid, THF, rt, N2, 16 h, 53%.
Scheme 3. Synthesis of p-XPRO-Click. Reagents and conditions: (a) Hexamethylenetetramine, TFA, HCl, reflux, 9 h, 90%; (b) TBSCl, imidazole, DCM, rt, N2, 16 h,
63%; (c) Ethynylmagnesium bromide, THF, ꢀ 78 ◦C, N2, 16 h, 83%; (d) cat. Cp*RuCl(µ2-SMe)2RuCP*Cl, NH4BF4, acetone, reflux, N2, 5 h, 83%; (e) NH3, MeOH,
NH2OSO3H, ꢀ 40 ◦C, N2, 7 h; TEA, iodine, MeOH, 52%; (f) TBAF-acetic acid, THF, rt, N2, 16 h, 55%.
Although photoactive propofol derivatives do exist, several are un-
stable,13 or have cumbersome synthetic routes,12 and only one has a
click handle.6 We hypothesized that linking the propofol core with a
short nonpolar aliphatic spacer to separate the diazirine moiety from the
aryl ring would not only preserve the minimal pharmacophore but also
facilitate synthetic expediency.7,13 We designed three probes that fulfill
these criteria, and we synthesized them using relatively short and effi-
cient synthetic sequences (Schemes 1–3), which could readily be applied
to other similar scaffolds. Following synthesis, we used functional assays
using calcium influx and binding to a fluorescent indicator as a measure
of probe activation of TRPA1 to determine their promise in future
crosslinking experiments.
We next sought to generate an analog with a clickable group to
enable potential labeling studies. We envisioned placing an alkyne at
two possible locations within XPRO—in one analog one of the two iso-
propyl groups could be substituted with an alkyne (o-XPRO-Click) and in
a second analog the alkyne could be appended proximal to the diazirine
itself (p-XPRO-Click). At the outset, we chose to synthesize and profile
both designs, since it was unclear which would have the best potency/
reactivity profile.
To synthesize o-XPRO-Click (12; Scheme 2) starting from 2-isopro-
pylphenol (6), Amberlyst-15 was again used to generate the alkyl ke-
tone 7. Subsequent bromination with NBS, Sonogashira coupling with
TMS-acetylene, and phenol protection yielded the bis-silyl-protected
intermediate 10. Functionalization of the ketone using previously
established conditions, followed by deprotection, generated o-XPRO-
Click. In the case of p-XPRO-Click (18; Scheme 3), Duff formylation
generated a benzaldehyde intermediate which, after phenol protection,
was subjected to a Grignard reaction with ethynylmagnesium bromide.
Subsequent propargylic alkylation19 followed by diazirination and
deprotection yielded p-XPRO-Click. Moreover, o-XPRO-Click and p-
XPRO-conjugation experiments with 3-Azido-7-hydroxycoumarin,
whose fluorescence increases upon a successful click reaction with an
alkyne,20 demonstrate their ability to react with azide partners (Sup-
plementary fig. 1).
We designed our initial crosslinkable probe, XPRO (5), bearing the
linker and diazirine moiety in the para position to minimize potential
intramolecular interactions of the reactive carbene intermediate while
maintaining overall hydrophobicity of the propofol starting point.
Indeed, a propofol analog with a trifluoromethyl diazirine directly at the
para position was reported to be a poor photolabel.7,13 Synthesis of
XPRO was achieved in an expedient 5-step sequence (Scheme 1). Initial
functionalization of the para-position was achieved through an
Amberlyst-15-mediated alkylation16 followed by protection of the
phenol hydroxyl as a TBS silyl ether. The pendant methyl ketone was
subsequently converted to the requisite diazirine via a two-step
sequence.17,18 Lastly, deprotection of the TBS ether with TBAF affor-
ded XPRO. Overall yield for this sequence was 22%.
As a first assessment of the biological activity of our three new
photocrosslinkable propofol analogs, we tested their ability to activate
2

K.A. Skinner et al.
Bioorganic & Medicinal Chemistry Letters 39 (2021) 127927
Fig. 1. XPRO, o-XRO-Click, and p-XPRO-Click and propofol elicit calcium influx in Sf21 insect cells heterologously expressing TRPA1 but not TRPV1. (A, B, C) Time
traces showing the net change in Fura-2 340/380 fluorescence ratio (ΔRatio) in TRPA1-expressing insect cells. Adding XPRO (A), o-XPRO-Click (B) or p-XPRO-Click
(C) at 60 s caused a rise in [Ca2+]i. Traces showing activation of the equivalent batches of cells with 60 µM propofol (grey traces) are included as positive controls in
each panel. Addition of 1 mM allyl isothiocyanate (AITC) at 180 s provided a positive control for mouse TRPA1 expression. Data represent mean ± SEM (n = 4–14).
(D, E, F) Addition of 60 µM XPRO (D), o-XPRO-Click (E) or p-XPRO-Click (F) or propofol (grey traces) at 60 s did not cause a rise in [Ca2+]i in rat TRPV1-expressing
insect cells. Addition of 75 nM capsaicin, a TRPV1-specific agonist, at 180 s did elicit a rise in [Ca2+]i, confirming TRPV1 expression in all samples. Data represent
mean ± SEM for n = 3–12.
TRPA1, leveraging our previously developed cell-based calcium-influx
assay.5 For TRPA1 production, we chose baculovirus-infected insect
cells, a common heterologous expression system for eukaryotic integral
membrane proteins.21 Cells were loaded with Fura-2, a calcium-
sensitive ratiometric fluorescent dye. Ligand binding and activation
causes ion channels like TRPA1 to open, allowing calcium to flow into
cells and bind Fura-2, resulting in an absorption shift.22
assessments should be performed before using these new compounds in
vivo. Furthermore, our bifunctional probes, o-XPRO-Click and p-XPRO-
Click, will facilitate profiling of the propofol proteome under cellular
conditions, for example enabling tagging to a fluorophore to determine
the localization of propofol-interacting proteins. The synthetic strategies
described here are also applicable to alkyl phenols like propofol, such as
carvacrol and thymol, and molecules that contain an analogous linker in
the para position, such as zingerone, all of which include TRPA1 mod-
ulation among their biological activities.27 Thus, the combination of the
diazirine and alkyne functionality enables not only detection and
enrichment of photolabeled biological targets but also interrogation of
the propofol-interacting proteome and generation of novel ion channel
modulators through click chemistry reactions.
After measuring background fluorescence levels, addition of 60 µM
XPRO or propofol at 60 s elicited a rise in [Ca2+]i in TRPA1-expressing
cells, resulting in an increase in the 340/380 ratio of the Fura-2 dye
(Fig. 1A). Expression of TRPA1 in all samples was confirmed by acti-
vation at 180 s with 1 mM AITC, an electrophilic TRPA1 agonist.23,24
Interestingly, o-XPRO-Click was a weaker TRPA1 agonist than propofol
(Fig. 1B). We hypothesize that the potency difference was due to
replacement of a key structural feature of propofol, one of the 2-isopro-
pyl group, with a smaller and non-branched alkyne. Consistent with this
hypothesis, p-XPRO-Click, which preserves both 2-isopropyl sub-
stituents, was at least as efficient as propofol (Fig. 1C).
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
To rule out nonspecific calcium influx due to increased hydropho-
bicity of XPRO relative to propofol, we also tested activation of TRPV1, a
related channel found on TRPA1-expressing sensory nerves and crucial
for pain sensation.25 As expected,26 neither propofol nor XPRO activated
TRPV1 while capsaicin, which serves as a positive control for functional
TRPV1 expression, was a potent TRPV1 agonist (Fig. 1D). In addition,
neither o-XPRO-Click nor p-XPRO-Click elicited calcium influx in
TRPV1-expressing cells (Fig. 1E and 1F). Taken together, calcium-influx
assays demonstrate that XPRO, o-XPRO-Click, and p-XPRO-Click share
similar TRPA1 potency with propofol.
Acknowledgements
We thank members of the Gaudet and Kahne lab for insightful dis-
cussions and technical assistance. Financial support for this work was
provided by the American Heart Association (16GRNT27250119). K.A.
S. was supported by an NSF Graduate Research Fellowship (DGE-
1144152) and a Ford Foundation Fellowship.
In summary, the structure-function studies highlight promising
photoactivatable propofol analogs to move forward in crosslinking ex-
periments. We envision our analogs and synthesis schemes will benefit
future photolabeling strategies in several ways. For example, XPRO and
p-XPRO-Click retain both isopropyl arms, unlike most previously pub-
lished photoactivatable propofol analogs except p-4-AziC5-Pro,13 and
both of our new analogs have a shorter linker to the diazirine than p-4-
AziC5-Pro and a simpler synthetic route. Of note, given their structural
similarities, we anticipate that XPRO and p-XPRO-Click will activate the
GABAA receptor both in vitro and in vivo similarly to p-4-AziC5-Pro,
which was tested in zebrafish tadpoles,13 although pharmacokinetic
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.bmcl.2021.127927.
References
1 Jalota L, Kalira V, George E, et al. Prevention of pain on injection of propofol:
systematic review and meta-analysis. BMJ. 2011;342(mar15 1). https://doi.org/
10.1136/bmj.d1110. d1110.
3

K.A. Skinner et al.
Bioorganic & Medicinal Chemistry Letters 39 (2021) 127927
2 Picard P, Tramer MR. Prevention of pain on injection with propofol: a quantitative
systematic review. Anesth Analg. 2000;90(4):963–969. https://doi.org/10.1097/
00000539-200004000-00035.
15 Li Z, Hao P, Li L, et al. Design and synthesis of minimalist terminal alkyne-containing
diazirine photo-crosslinkers and their incorporation into kinase inhibitors for cell-
and tissue-based proteome profiling. Angew Chem Int Ed Engl. 2013;52(33):
8551–8556. https://doi.org/10.1002/anie.201300683.
3 Jurd R, Arras M, Lambert S, et al. General anesthetic actions in vivo strongly
attenuated by a point mutation in the GABA(A) receptor beta3 subunit. FASEB J.
2003;17(2):250–252. https://doi.org/10.1096/fj.02-0611fje.
16 Bunce RA, Reeves HD. Amberlyst-15 catalyzed addition of phenols to α,
β-unsaturated ketones. Synth Commun. 1989;19(5-6):1109–1117.
4 Matta JA, Cornett PM, Miyares RL, Abe K, Sahibzada N, Ahern GP. General
anesthetics activate a nociceptive ion channel to enhance pain and inflammation.
Proc Natl Acad Sci U S A. 2008;105(25):8784–8789. https://doi.org/10.1073/
pnas.0711038105.
17 Church RFR, Kende AS, Weiss MJ, Diazirines I. Some observations on the scope of the
ammonia-hydroxylamine-o-sulfonic acid diaziridine synthesis. The preparation of
certain steroid diaziridines and diazirines. J Am Chem Soc. 1965;87(12):2665–2671.
https://doi.org/10.1021/ja01090a025.
5 Woll KA, Skinner KA, Gianti E, et al. Sites contributing to TRPA1 activation by the
anesthetic propofol identified by photoaffinity labeling. Biophys J. 2017;113(10):
2168–2172. https://doi.org/10.1016/j.bpj.2017.08.040.
18 Church RFR, Diazirines II WMJ. Synthesis and properties of small functionalized
diazirine molecules. Observations on the reaction of a diaziridine with the iodine-
iodide ion system. J Organic Chem. 1970;35(8):2465–2471. https://doi.org/10.1021/
jo00833a001.
6 Woll KA, Murlidaran S, Pinch BJ, et al. A novel bifunctional alkylphenol anesthetic
allows characterization of gamma-aminobutyric acid, type A (GABAA), receptor
subunit binding selectivity in synaptosomes. J Biol Chem. 2016;291(39):
20473–20486. https://doi.org/10.1074/jbc.M116.736975.
19 Nishibayashi Y, Wakiji I, Ishii Y, Uemura S, Hidai M. Ruthenium-catalyzed
propargylic alkylation of propargylic alcohols with ketones: straightforward
synthesis of gamma-keto acetylenes. J Am Chem Soc. 2001;123(14):3393–3394.
https://doi.org/10.1021/ja015670z.
7 Yip GMS, Chen Z-W, Edge CJ, et al. A propofol binding site on mammalian GABAA
receptors identified by photolabeling. Nat Chem Biol. 2013;9(11):715–720. https://
doi.org/10.1038/nchembio.1340.
20 Sivakumar K, Xie F, Cash BM, Long S, Barnhill HN, Wang Q. A fluorogenic 1,3-
dipolar cycloaddition reaction of 3-azidocoumarins and acetylenes. Org Lett. 2004;6
(24):4603–4606. https://doi.org/10.1021/ol047955x10.1021/ol047955x.s001.
21 He Y, Wang K, Yan N. The recombinant expression systems for structure
determination of eukaryotic membrane proteins. Protein Cell. 2014;5(9):658–672.
https://doi.org/10.1007/s13238-014-0086-4.
8 Ton HT, Phan TX, Abramyan AM, Shi L, Ahern GP. Identification of a putative
binding site critical for general anesthetic activation of TRPA1. Proc Natl Acad Sci U S
A. 2017;114(14):3762–3767. https://doi.org/10.1073/pnas.1618144114.
9 Ton HT, Phan TX, Ahern GP. Inhibition of ligand-gated TRPA1 by general
anesthetics. Mol Pharmacol. 2020;98(3):185–191. https://doi.org/10.1124/
mol.119.118851.
22 Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with
greatly improved fluorescence properties. J Biol Chem. 1985;260(6):3440–3450.
23 Hinman A, Chuang HH, Bautista DM, Julius D. TRP channel activation by reversible
covalent modification. Proc Natl Acad Sci U S A. 2006;103(51):19564–19568.
https://doi.org/10.1073/pnas.0609598103.
10 Hill JR, Robertson AAB. Fishing for drug targets: a focus on diazirine photoaffinity
probe synthesis. J Med Chem. 2018;61(16):6945–6963. https://doi.org/10.1021/acs.
jmedchem.7b01561.
11 Prevet H, Collins I. Labelled chemical probes for demonstrating direct target
engagement in living systems. Future Med Chem. 2019;11(10):1195–1224. https://
doi.org/10.4155/fmc-2018-0370.
24 Macpherson LJ, Dubin AE, Evans MJ, et al. Noxious compounds activate TRPA1 ion
channels through covalent modification of cysteines. Nature. 2007;445(7127):
541–545. https://doi.org/10.1038/nature05544.
12 Hall MA, Xi J, Lor C, et al. m-Azipropofol (AziPm) a photoactive analogue of the
intravenous general anesthetic propofol. J Med Chem. 2010;53(15):5667–5675.
https://doi.org/10.1021/jm1004072.
25 Fernandes ES, Fernandes MA, Keeble JE. The functions of TRPA1 and TRPV1:
moving away from sensory nerves. Br J Pharmacol. 2012;166(2):510–521. https://
doi.org/10.1111/j.1476-5381.2012.01851.x.
13 Stewart DS, Savechenkov PY, Dostalova Z, et al. p-(4-Azipentyl)propofol: a potent
photoreactive general anesthetic derivative of propofol. J Med Chem. 2011;54(23):
8124–8135. https://doi.org/10.1021/jm200943f.
26 Nishimoto R, Kashio M, Tominaga M. Propofol-induced pain sensation involves
multiple mechanisms in sensory neurons. Pflugers Arch. 2015;467(9):2011–2020.
https://doi.org/10.1007/s00424-014-1620-1.
14 Dubinsky L, Krom BP, Meijler MM. Diazirine based photoaffinity labeling. Bioorg Med
Chem. 2012;20(2):554–570. https://doi.org/10.1016/j.bmc.2011.06.066.
27 Lee SP, Buber MT, Yang Q, et al. Thymol and related alkyl phenols activate the
hTRPA1 channel. Br J Pharmacol. 2008;153(8): 1739–49. 10.1038/bjp.2008.85.
4