Design, synthesis and pharmacological characterization of analogs of 2-aminoethyl diphenylborinate (2-APB), a known store-operated calcium channel blocker, for inhibition of TRPV6-mediated calcium transport

Article abstract of DOI:10.1016/j.bmc.2013.03.037

2-Aminoethyl diphenylborinate (2-APB) is a known modulator of the IP ₃ receptor, the calcium ATPase SERCA, the calcium release-activated calcium channel Orai and TRP channels. More recently, it was shown that 2-APB is an efficient inhibitor of the epithelial calcium channel TRPV6 which is overexpressed in prostate cancer. We have conducted a structure-activity relationship study of 2-APB congeners to understand their inhibitory mode of action on TRPV6. Whereas modifying the aminoethyl moiety did not significantly change TRPV6 inhibition, substitution of the phenyl rings of 2-APB did. Our data show that the diaryl borinate moiety is required for biological activity and that the substitution pattern of the aryl rings can influence TRPV6 versus SOCE inhibition. We have also discovered that 2-APB is hydrolyzed and transesterified within minutes in solution.

Full text of DOI:10.1016/j.bmc.2013.03.037

Bioorganic & Medicinal Chemistry 21 (2013) 3202–3213
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Design, synthesis and pharmacological characterization of analogs
of 2-aminoethyl diphenylborinate (2-APB), a known store-operated
calcium channel blocker, for inhibition of TRPV6-mediated calcium
transport
Alexandre Hofer a, , Gergely Kovacs b,c, , Anna Zappatini a, Michele Leuenberger a,c, Matthias A. Hediger b,c
,
Martin Lochner a,c,
⇑
a Department of Chemistry and Biochemistry, University of Bern, Freiestrasse 3, 3012 Bern, Switzerland
b Institute of Biochemistry and Molecular Medicine, University of Bern, Bühlstrasse 28, 3012 Bern, Switzerland
c Swiss National Centre of Competence in Research, NCCR TransCure, University of Bern, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
2-Aminoethyl diphenylborinate (2-APB) is a known modulator of the IP3 receptor, the calcium ATPase
SERCA, the calcium release-activated calcium channel Orai and TRP channels. More recently, it was
shown that 2-APB is an efficient inhibitor of the epithelial calcium channel TRPV6 which is overexpressed
in prostate cancer. We have conducted a structure–activity relationship study of 2-APB congeners to
understand their inhibitory mode of action on TRPV6. Whereas modifying the aminoethyl moiety did
not significantly change TRPV6 inhibition, substitution of the phenyl rings of 2-APB did. Our data show
that the diaryl borinate moiety is required for biological activity and that the substitution pattern of
the aryl rings can influence TRPV6 versus SOCE inhibition. We have also discovered that 2-APB is hydro-
lyzed and transesterified within minutes in solution.
Received 11 December 2012
Revised 7 March 2013
Accepted 14 March 2013
Available online 1 April 2013
Keywords:
Calcium entry
TRPV6
Store-operated calcium channels
2-APB
Ó 2013 Elsevier Ltd. All rights reserved.
Inhibitors
1. Introduction
vitamin D-regulated Ca2+ uptake in the small intestine. Apart from
its expression in the duodenum, TRPV6 is highly expressed in pla-
The divalent Ca2+ ion is a second messenger involved in key
intracellular signaling processes that mediate a multitude of bio-
logical functions. Among many other things, intracellular Ca2+ con-
centrations also influence the cell cycle and thus proliferation.
Therefore, it is apparent that malfunctions in the regulation of
intracellular Ca2+ levels are involved in the formation and prolifer-
ation of certain types of cancer.
TRPV6 is a member of the vanilloid (V) subfamily of the large
superfamily of transient receptor potential (TRP) ion channels. It
is a constitutively active, highly selective Ca2+ channel (PCa/PNa
130) and responsible for the main mechanism of 1,25-dihydroxy
centa, pancreas, prostate and the salivary gland.1
Structural information on TRP channels is very scarce but it is
believed that TRPV6 assembles in the plasma membrane to gener-
ate a homotetramer that forms a cation-permeable pore. A TRPV6-
protomer is predicted to have both large intracellular N- and C-ter-
minal domains which are connected via six transmembrane heli-
ces.2 A hydrophobic stretch between helix 5 and 6 presumably
forms a re-entrant loop which is involved in channel pore forma-
tion and determination of cation selectivity. The N-terminal do-
main contains three ankyrin repeats that allow the connection of
TRPV6 to the cytoskeleton.3
Increased mRNA and protein levels of TRPV6 were found in
prostate cancer, as well as in other cancers of epithelial origin, such
as breast, thyroid, colon and ovarian cancers.4–6 Elevated TRPV6
expression in the prostate cancer cell line LNCaP promoted prolif-
eration at high rate and increased the resistance to apoptosis.7
Knockdown of TRPV6 by means of siRNA in the breast cancer cell
line T47D was shown to result in decreased viability of the cells.8
Thus, these studies suggest a potential role of TRPV6 both as cancer
marker and as a therapeutic target. Indeed, in a very recent study,
Abbreviations: 2-APB, 2-aminoethyl diphenylborinate; CRAC, Ca2+-release acti-
vated current; IP3, inositol 1,4,5-triphosphate; SAR, Structure–activity relationship;
SERCA, sarco/endoplasmatic reticulum Ca2+-ATPase; SOCs, store-operated calcium
channels; SOCE, store-operated calcium entry; STIM1, stromal interaction molecule
1; TRPV6, vanilloid transient receptor channel type 6.
⇑
Corresponding author. Tel.: +41 31 631 3311; fax: +41 31 631 4272.
E-mail address: martin.lochner@dcb.unibe.ch (M. Lochner).
These two authors contributed equally to the work.
0968-0896/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.03.037

A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
3203
Very recently, it was reported that 1 efficiently inhibits heterolo-
R2
gously expressed human TPRV6 channels with an IC50 value of
R3
22 l
M.16 Interestingly, the same study showed that 2-APB did
N
B
X
not inhibit the highly homologous TRPV5 channel. 2-APB also
inhibits TRPC1, TRPC4 and TRPC5 channels as well as TRPM7 (Mag-
NuM) currents.17–19 The calcium pump of the endoplasmic reticu-
lum (SERCA) is also blocked by 2-APB at very high
concentrations.20
NH2
B
O
R1
series 1: R1; R2 = R3 = H
1
, 2-APB
series 2: R1 = 2-aminoethoxy; R2 = R3
series 3: R1 = 2-aminoethoxy; R2 = H; R3
Numerous previous studies have used 2-APB and several syn-
thetic and commercially available analogs to study their influence
on SOCE mechanisms in different cell lines. For instance, Dobryd-
neva and Blackmore have established more than a decade ago that
2-APB and its analog diphenylborinic anhydride ((Ph2B)2O) are able
to prevent thapsigargin-activated Ca2+ and Sr2+ entry in human
platelets.21 An SAR study by the same group later showed that 2-
APB analogs with various amino alcohol side chains retained inhi-
bition activity of thrombin-induced Ca2+ influx in human plate-
lets.22 Mikoshiba and co-workers have synthesized a library of
bis-2-APB analogs and found some compounds that blocked both
store-operated and IP3R-induced Ca2+ influx in DT40 cells much
more potently than 2-APB.23 Similar bis-2-APB analogs have later
been used by the same group to study their influence on CRAC
channel subtypes24 and on SOCE in Jurka T cells.25 A more recent
SAR study with 2-APB and commercially available analogs has
emphasized the role of the R2B-O functional group in the ability
to potentiate and inhibit SOCE in different leukocyte cells lines.26
To date, no systematic SAR study of 2-APB analogs as modula-
tors of the TRPV6 channel was reported. Given the initial observa-
tion that 2-APB inhibits TRPV6, combined with fact that blocking
TRPV6 affects proliferation of certain cancer cells, the aim of our
work was to develop synthetic 2-APB analogs which are more po-
tent and more specific inhibitors of TRPV6. Apart from varying the
amino alcohol moiety, our goal was to synthesize novel analogs
with different substituents either on both or one of the two phenyl
rings of 2-APB (Fig. 1). Herein, we report the design and synthesis
of a small library of 2-APB congeners which were tested on human
TRPV6 and store-operated calcium channels (SOCs) in HEK293
cells.
Figure 1. Chemical structure of parent compound 2-APB (1) and schematic
overview of synthesized 2-APB analogs library.
compounds derived from known inhibitors of Ca2+ entry mecha-
nisms (econazole, miconazole and TH-1177) were tested both on
TRPV6 and the highly homologous TRPV5, and at least one com-
pound was shown to be a potent growth inhibitor of prostate can-
cer LNCaP (IC50 = 0.44 lM) and breast cancer T47D (IC50 = 39 lM)
cells.9 Even though the affinity of the reported compounds for
TRPV6 was in the high micromolar range, this study confirmed that
TRPV6 inhibitors can affect proliferation of cancer cells.
2-APB (1, Fig. 1) was initially characterized as a membrane-per-
meable modulator of IP3 receptors, which are localized at intracel-
lular calcium stores, but later it was found to directly influence
store-operated calcium entry (SOCE) mechanisms.10,11 The effect
of 2-APB on SOCE activity is nevertheless quite complex. In general,
2-APB potentiates ICRAC currents at low concentrations and inhibi-
tion is observed at higher concentrations.12 The best characterized
SOCE current is the so called ICRAC (Ca2+-release activated current)
in which Orai1, Orai2 and/or Orai3 are the pore-forming subunits.
2-APB was found to block Orai1 and to a lesser extent Orai2 activ-
ity whereas Orai3 currents were directly activated.13,14 The cellular
distribution of STIM1, which functions as an ER Ca2+-sensor and is
a key player in the activation of the SOCE, is also affected by 2-
APB.14 Further investigations revealed that it also interacts with
various TRP channels. For instance, a recent report showed that
2-APB enhances TRPV3 and TRPV4 activity and the researchers
were able to identify two residues in the N- and C-terminal
cytoplasmic domains which are required for 2-APB sensitivity.15
R
R'
Ph
B
a
Ph
N
Ph2B
BPh2
O
X
X = O, S
Ph
2a-i
NH
Ph
B
O
Ph
B
O
Ph
B
O
H
N
H
N
Ph
B
O
Ph
B
NH2
Ph
Ph
Ph
NMe2
Ph
N
O
2a
2b
2c
2d
2e
Ph
B
O
Ph
B
O
Ph
B
O
Ph
B
O
N
Ph
Ph
Ph
NH2
NH2
Ph
OH
N
O
2h
O
2i
2f
2g
H
N
Ph
B
O
Ph
B
S
Ph
Ph
Ph
O
Ph
NH2
B
NH2
Ph
NH2
OH
Ph
4
3
O
2k
2j
Scheme 1. Synthesis of aminoethanol variations of 2-APB (series 1). Reagents and conditions: (a) For 2a–g: amino alcohol, MeCN, 0 °C, 1–2 h, 46–99%; for 2h–j: glycine,
-serine or
-tryptophan, MeCN, 50 °C, 1–3 h, 75%, 44% and 32%, respectively; for 2k: cysteamine, MeCN, 0 °C, 5 d at ꢀ18 °C, 32%.
L
L

3204
A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
Ar
B
O
reaction mixture and could be collected by filtration in high pur-
Br
ity.27 Piperazinol derivative 2e was found to precipitate as the salt
of diphenylborinate while all other compounds were isolated as
single components. Of note are also the fluorescent properties of
2g (kmax abs 400 nm, kmax em 497 nm in CHCl3).28 For the prepara-
tion of amino acid adducts 2h–j the reaction mixture needed heat-
ing in order for the transformation to take place. The clearly
present triplet at 5.22 ppm (DMSO-d6) of the hydroxyl group in
the 1H NMR spectrum of 2i confirmed that an acyloxyborane was
formed rather than a borinic acid ester. Thiols and thioethers are
often used to stabilize the electron-deficient boron atom (e.g., as
in the BH3ꢁSMe2 complex) and therefore we synthesized sulfur
analog 2k, which was obtained by reacting diphenylborinic anhy-
dride with cysteamine. For the purpose of comparison we also syn-
thesized borinic acid 3 and carbon analog 4. The former was
obtained by acidic hydrolysis of 1 (1 M aq HCl, Me2CO/MeOH
1:1, rt) whereas the latter compound was synthesized in three
steps starting from diphenylmethanol.29,30
Ar
Ar
a
b
Ar
NH2
B
R
OH
5a-9a
5b-9b
5a,b:
7a,b:
6a,b:
CF3
MeO
8a,b:
F
9a,b:
OMe
Scheme 2. Synthesis of diarylborinates and corresponding 2-aminoethyl esters
(series 2). Reagents and conditions: (a) n-BuLi or s-BuLi, Et2O, ꢀ78 °C; (i-PrO)3B,
Et2O, ꢀ78 °C to rt; 1 M HCl, rt, 5–40%; (b) 2-aminoethanol, EtOH, rt or 2-
aminoethanol, EtOH/H2O, rt or 50 °C, 19–64%.
We then turned our attention to the synthesis of 2-APB congen-
ers with substituted phenyl rings (Scheme 2). To access these com-
pounds with identical substitution on both phenyl rings, various
aryl bromides were lithiated in a first step and then added to triiso-
propyl borate.31 The resulting borinic esters were hydrolyzed and
the crude borinic acids purified by column chromatography. Final-
ly, the diarylborinates were esterified with 2-aminoethanol32 to
yield final compounds 5b–9b. In general, we found that the 2-ami-
noethyl esters are crystalline, more stable and also more soluble
than the corresponding borinic acids.33
a
Br
b
Br
B
OH
Br
Br
10
11
Scheme 3. Synthesis of restricted 2-APB analog. Reagents and conditions: (a)
n-BuLi, THF, ꢀ78 °C, 76%; (b) s-BuLi, Et2O, ꢀ78 °C; (i-PrO)3B, Et2O, ꢀ78 °C to rt; 1 M
HCl, rt, 3%.
We were also interested to study the biological activity of a con-
formationally restricted analog of 1. To this end, we synthesized 11
where both phenyl rings were connected via a short linker
(Scheme 3). Treatment of 2-bromobenzyl bromide with n-BuLi
yielded bisbromide 1034 which was in turn lithiated, converted
to the borinic ester and hydrolyzed to borinic acid 11. We could
not obtain the corresponding 2-aminoethyl ester in crystalline
form and satisfactory purity. In addition, the purification of 11
proved very difficult, leading to a low isolated yield.
2. Results and discussion
2.1. Synthesis of 2-APB analogs
The first series of 2-APB analogs comprised borinic acid esters,
which featured various amino alcohols, and adducts of diphenylbo-
rinic acid and some amino acids (Scheme 1). These compounds
were synthesized by adapting a literature method,22 which en-
tailed the reaction of diphenylborinic anhydride with the amino
alcohol. The crystalline products mostly precipitated from the
For the preparation of a series of aryl(phenyl)borinic acids and
their corresponding 2-aminoethyl esters two synthetic strategies
were utilized (Scheme 4). Arylboronic acid esters were reacted
a
c
b
O
O
O
Ar
Ph
HO
HO
B
Ar
B
B
Ph
B
Ph
OH
O
13a-23a
12
d
O
B
Ph
Ph
B
O
O
Ar
NH2
24
13b-23b
13a,b:
17a,b:
21a,b:
14a,b:
15a,b:
16a,b:
F
n-Bu
S
18a,b:
19a,b:
20a,b:
Cl
N
Br
22a,b:
23a,b:
I
CF3
Scheme 4. Synthesis of aryl(phenyl)borinic acids and their corresponding 2-aminoethyl esters (series 3). Reagents and conditions: (a) PhLi, Et2O, ꢀ78 °C; 1 M HCl, rt, 16–39%;
(b) 2,2-dimethylpropane-1,3-diol, THF, rt, 98%; (c) ArLi, Et2O, ꢀ78 °C; 1 M HCl, rt; (d) 2-aminoethanol, EtOH/H2O, rt, 15–54% (for 2 steps).

A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
3205
with phenyllithium which yielded the borinic acids after hydroly-
sis. Alternatively, phenylboronic acid neopentylglycol ester (12)
was first synthesized from phenyl boronic acid and then treated
with aryllithium reagents, which in turn were freshly prepared
from the corresponding aryl bromides or iodides by lithium halo-
gen exchange. Furthermore, we found that, instead of 12, commer-
cially available phenylboronic acid pinacol ester (24) can also be
used effectively in this reaction. As above, borinic acids 13a–23a
were transformed into the corresponding 2-aminoethyl esters.
We were unable to isolate esters 20b and 23b, since their at-
tempted purification by RP-HPLC yielded the hydrolyzed borinic
acids 20a and 23a, respectively. Analogs 21a and 22a were synthe-
sized by mono-lithiation of 1,4-dibromo- and 1,4-diiodobenzene,
respectively, followed by treatment with 24 and hydrolysis. Inter-
estingly, when we attempted the bis-borination of 1,4-diiodoben-
zene by reacting the latter with two equivalents of n-BuLi, and of
24 subsequently, we only isolated mono-borinated C–C coupling
product 16a in fairly good yield.
temperature (Table S2). The yield of 1 ranged from 40 to 100% after
equilibration for 48 h. The transesterification of sulfur analog 2k in
particular was very fast and quantitative, presumable owing to the
longer and thus weaker boron-sulfur bond.
From these experiments it appears that 1 and its analogs can
undergo rapid dynamic changes in solution and potentially pro-
duce several species which are in equilibrium with each other. 2-
APB (1) has been used in numerous biological studies and, in our
mind, this liability has been largely ignored.
3. Pharmacological characterization of 2-APB analogs
With the synthetic 2-APB analogs in hand we proceeded to as-
sess their biological activity as inhibitors of human TRPV6 channel
and SOCE (Table 1). The IC50 values for human TRPV6 were deter-
mined on a high-throughput fluorescence microplate reader FLIPR
Tetra by measuring Ca2+ influx into stably hTRPV6-transfected
HEK293 cells utilizing a Ca2+-sensitive fluorescent dye (Fig. 2). It
has recently been shown that hTRPV6 expressed in HEK293 cells
is also permeable to Cd2+ which can be detected at much lower
concentration by the same fluorescent dye producing a more
2.2. Structure of 2-APB analogs
Several pieces of evidence and literature precedence35 indicate
that 1 and its analogs adopt a cyclic structure forming an intramo-
lecular N–B bond in the solid state as well as in solution. For in-
stance, the crystal structures of 2a, 2b and 2k (Figs. S1–3) clearly
show the interaction of the nitrogen with the central boron atom.
Of note is the significantly longer boron–sulfur bond in 2k (1.97 Å)
compared to the boron-oxygen bonds in 2a and 2b (1.47 Å and
1.48 Å, respectively). Furthermore, the 1H NMR signal for the N–
H proton(s) of the borinic esters (2b–d, 2h–k, 5b–9b, 13b–19b,
21b and 22b) is low field shifted to about 6–7 ppm (in DMSO-d6)
which is most probably due to donation of electron density from
the nitrogen to the electron deficient boron and thus deshielding
of the N–H protons. Notably, 2i displayed two distinct signals for
its amine protons which were separated by 0.9 ppm, further
emphasizing coordination of the nitrogen via its lone pair to the
boron.
Table 1
Inhibition of Ca2+ and Cd2+ influx via hTRPV6 and SOCs by 2-APB analogsa
Compd
IC50
(
l
M) hTRPV6
IC50
(
l
M) SOCE
Cd2+ influx
Ca2+ influx
Cd2+ influx
Ca2+ influx
2-APB (1)
2a
2b
2c
2d
2e
2f
2g
2h
2i
2j
2k
3
4
5a
5b
6a
6b
7a
7b
8a
8b
9b
11
13b
14b
15b
16b
17b
18b
19b
20a
20.7
63.6
18.8
18.5
16.2
15.0
34.9
NIb
24.5
19.8
28.0
30.6
28.1
24.3
41.5
NI
22.9
8.9
8.9
8.0
7.9
6.0
10.7
NI
11.9
5.1
10.2
10.1
8.2
2.2
10.2
NI
12.9
36.6
NI
30.6
20.2
NI
38.6
40.5
6.9
51.8
9.8
106.0
23.0
1.6
13.2
11.6
7.6
2.1
26.6
9.0
12.4
33.3
9.3
10.0
8.6
11.0
9.8
14.7
44.6
8.7
184.2
25.9
7.4
11.8
10.5
9.5
3.2
17.1
5.9
25.8
36.4
9.8
127.1
19.1
159.1
18.6
14.4
>500
>500
10.8
16.3
37.1
17.0
>100
>500
18.8
31.1
18.9
14.3
72.3
NI
93.5
31.5
>500
29.0
17.1
>500
412.3
11.2
13.7
61.6
14.2
89.8
>500
33.6
67.9
12.9
12.1
51.3
NI
2.3. Stability of 2-APB and its analogs in solution
While handling the compounds we have noticed that, for in-
stance, attempted purification of the 2-aminoethyl borinic esters
either by column chromatography or preparative RP-HPLC only
furnished the hydrolyzed product, namely the corresponding bori-
nic acid. We have thus carried out some stability studies with 1
and our analogs. Towards this end, 1 was dissolved in CHCl3, ex-
tracted once with H2O and the residue remaining in the organic
layer analyzed by 1H NMR. We observed substantial hydrolysis
(80%) of 1, that is, the main component of the residue was diphe-
nylborinic acid 3 mixed with 1. This is consistent with a recent
study which reported a small association constant (Ka) of 4.0 Mꢀ1
of 3 with 2-aminoethanol in pH 7 buffer.36 Interestingly, amino
acid complexes such as 2h and 2i remained unchanged during this
extraction. The remarkable stability of amino acid complexes of
diphenylborinic acid has also been demonstrated under mass spec-
trometry and HPLC conditions.37 Conversely, when borinic acid 3
was treated with 2-aminoethanol in DMSO-d6, formation of 1
could be observed by 1H NMR (Fig. S4) and 11B NMR (Fig. S5) with-
in minutes. Similar NMR studies in D2O were inconclusive partly
due to the very scarce solubility of 3 in D2O and also because most
of the diagnostic 1H NMR signals for the different species are pro-
duced by exchangeable protons. We have also investigated the sta-
bility of some of our borinic acid esters (2b, 2d, 2f, 2g, 2i and 2k)
under transesterification conditions. In the event, 2 equiv of 2-
amioethanol were added to the solution of borinic acid ester in
DMSO-d6 and in all cases we observed rapid formation of 1 at room
9.9
2.7
2.4
7.7
3.5
13.4
NI
NI
12.3
9.7
13.2
5.4
11.3
19.2
NI
10.0
12.0
9.8
30.9
21.8
3.4
25.2
6.4
4.9
7.2
NA
21b
22b
23a
5.2
2.8
12.5
17.7
NI
NAc
11.4
NI
(Ph2B)2O
B(OH)3
MeB(OH)2
PhB(OH)2
B(OPh)3
Ph2CHOH
Ph2CHO(CH2)2NMe2
13.4
NI
NI
NI
NI
NI
>500
NI
>500
NI
NI
NI
NI
NI
NI
NI
NI
NI
>500
>500
>500
>500
a
b
c
IC50 values are the average of six independent experiments.
NI, no inhibition.
NA, not attempted. Highest dose tested for each compound was 500 lM.

3206
A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
Figure 2. Representative traces depicting the effect of compound 2b, as an example, on the activity of TRPV6 (panels A and B) and SOCE (panels C and D). Using FLIPR Tetra
the changes of the fluorescence intensity of Calcium-5 Ca2+ (Cd2+)-sensitive fluorescence dye were monitored in RFP-hTRPV6 expressing HEK293 cells (TRPV6 activity) and
native HEK293 cell (SOCE activity). After establishing a stable baseline (50s), cells were treated with compound at different concentrations for 5 (TRPV6) or 10 min (SOCE) in
nominally Ca2+-free Krebs buffer. In HEK293 cells, store-operated calcium channel was activated with 1
lM thapsigargin (TG) which was applied together with the
compound. The rates of the increase of the fluorescence intensity were measured in response to administration of extracellular Ca2+ or Cd2+ as the measure of the activity of
TRPV6 (A, B) or SOCE (C, D).
intense signal.38 We have therefore determined the IC50 values for
Cd2+ entry for all the compounds in order to verify that they indeed
act on TRPV6. In addition, it allowed us to dissect the ion entry
mechanisms via TRPV6 from other Ca2+-dependent processes in
the cell. The IC50 values for SOCE were measured using non-trans-
fected HEK293 cells which were treated with thapsigargin, simul-
taneously with compound application, in order to passively release
Ca2+ from intracellular stores, which in turn results in SOCE activa-
tion. It is worth noting that non-transfected HEK293 cells do not
express endogenous TRPV6 or other TRP channels in appreciable
amounts.16 In addition, for the purpose of comparison, we have
also tested a few commercially available boron containing com-
pounds and 2-APB related carbon analogs (Table 1).
The data show that the boron atom is essential for hTRPV6
inhibitory activity as the carbon analog 4, diphenylmethanol and
the carbon analog of 2a, diphenhydramine (2-(diphenylmeth-
oxy)-N,N-dimethylethanamine, Ph2CHO(CH2)2NMe2), were signifi-
cantly less active or inactive. Similar to our finding regarding
SOCE activity, diphenhydramine was previously shown to be a
much weaker inhibitor of SOCE, or even inactive, in human plate-
lets and BL-41 cells compared to 2-APB,21,26 confirming the impor-
tance of the borinate functional group. What is more, the boronic
and boric acid derivatives (RB(OH)2 and B(OR)3, respectively) did
not show inhibition of hTRPV6 or SOCE, indicating that the oxida-
tion state of the boron is important as well.
the latter which is prone to hydrolysis thus potentially forming
2 equiv of 3, were as potent as 1 on Ca2+/Cd2+ influx via hTRPV6.
In our assay, 3 and diphenylborinic anhydride were slightly better
inhibitors of SOCE than 2-APB. A similar effect for diphenylborinic
anhydride was observed on SOCE of leukocyte cell lines.26 Interest-
ingly, hydrolytically stable amino acid derivatives 2h and 2i were
not considerably more active than the parent compound. Since free
2-aminoethanol, an important component of phospholipids, is
present in cells in a significant amount (estimated concentration
5–50 l
M)39 we cannot exclude that 2a–k, 3 and diphenylborinic
anhydride rapidly transesterify to 1 in vivo and therefore have sim-
ilar activity. Recent electrophysiology studies have shown how-
ever, that 1 inhibits hTRPV6 from the outside.16 Thus, at present,
the structure of the biologically active form of 2-APB (1) is not
clear.
The substitution of the phenyl rings of 1 affected the activity of
the compounds more significantly. Rigidified analog 11 for in-
stance showed a diminished potency when compared to corre-
sponding borinic acid 3. Similarly, ortho-substitution of the
phenyl rings, especially with electron-donating groups (6a, 6b),
resulted in markedly less active hTRPV6 inhibitors. Conversely,
para-substitution was better tolerated (e.g., 14b vs 15b) and small,
electron-withdrawing groups in para-position in particular slightly
enhanced TRPV6 inhibition potency (7a, 7b, 8b, 20a, 21b, 22b and
23a). Electron-donating (9b) and bulky groups (17b) in the para-
position on the other hand resulted in less active compounds.
Whereas replacing one phenyl ring by a pyridine ring, such as in
18b, had a detrimental effect on the activity, the thiophene ring
analog 19b was more active than 1. 2-Methylthiophene is a known
isosteric replacement of a phenyl ring, albeit with different
The 2-aminoethoxy analogs 2a–k exhibited inhibition of
hTRPV6 except 2g and 2j. However, none of these analogs was sig-
nificantly more potent than 1. In light of the observed rapid hydro-
lysis in solution, this finding is perhaps not surprising. In addition,
diphenylborinic acid (3) and diphenylborinic anhydride ((Ph2B)2O),

A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
3207
stereoelectronic and physicochemical properties. Our best TRPV6
inhibitor 22b had a para-iodo substituent on one of the two rings
and was about four times more potent than 1.
fractometer using mirror optics monochromated Mo K
(k = 0.71073 Å) and Al filtered.
a
radiation
The data show that our analogs are not selective since they also
inhibited SOCE. Analog 19b was a noticeable exception which
showed an approximately 2.5-fold selectivity for TRPV6 over SOCE.
It appears that if steric bulk around the central boron atom or rigid-
ity is increased (e.g., as in 2j, 6a, 6b and 11) TRPV6 inhibitory activ-
ity is dramatically lost, whereas SOCE inhibitory activity is
retained. Consequently, our analogs 2j, 6a, 6b and 11 are highly
SOCE-selective.
5.1.1. 3,3-Dimethyl-2,2-diphenyl-1,3,2-oxazaborolidin-3-ium-2-
uide (2a)
Diphenylborinic anhydride (200 mg, 0.578 mmol) was sus-
pended in MeCN (7 mL) and cooled to 0 °C. 2-(Dimethylamino)eth-
anol (59 mg, 0.665 mmol) dissolved in MeCN (1 mL) was added
and the mixture stirred at 0 °C. The anhydride slowly dissolved
and the product started to precipitate after 5 min. After 1 h of stir-
ring at 0 °C the mixture was allowed to stand for 15 min. Filtration
of the mixture yielded a first fraction of product (26 mg). The fil-
trate was stored at ꢀ20 °C to allow further precipitation. Filtration
yielded another fraction of product (81 mg). The fractions were
dried at ambient atmosphere to obtain 2a as a white crystalline so-
lid (total yield: 107 mg, 0.423 mmol, 73%). Some crystals were of
sufficient quality for X-ray analysis. 1H NMR (DMSO-d6,
300 MHz) d 7.80–7.53 (m, 4H), 7.18 (t, J = 7.2 Hz, 4H), 7.08 (t,
J = 7.2 Hz, 2H), 4.11 (t, J = 6.8 Hz, 2H), 2.93 (t, J = 6.8 Hz, 2H), 2.52
(s, 6H); MS (ESI+) m/z 254.17 [M+H]+; HRMS (ESI+) m/z calcd for
4. Conclusion
In an effort to discover new inhibitors of TRPV6 we have synthe-
sized several novel 2-APB analogs and evaluated these compounds
for activity as TRPV6 inhibitors. We have discovered several ana-
logs that were more potent inhibitors of TRPV6 than 2-APB, how-
ever, they inhibited SOCE as well. The most beneficial effects
were achieved when only one of the phenyl rings of 1 was replaced
by a 2-methylthiophene (19b). Throughout the course of our inves-
tigation, we have demonstrated that 2-APB and its amino alcohol
variants are not kinetically stable compounds since they hydrolyze
and transesterify rapidly in solution.
C
16H21BNO 254.1716 [M+H]+, found 254.1703 [M+H]+.
5.1.2. (3aS)-1,1-Diphenylhexahydro-1H-pyrrolo[1,2-
c][1,3,2]oxazaborol-7-ium-1-uide (2b)
Following the procedure for 2a, diphenylborinic anhydride
(198 mg, 0.572 mmol) was reacted with (S)-(+)-2-(hydroxy-
methyl)-pyrrolidine (74 mg, 0.732 mmol) in MeCN at 0 °C. This
yielded 2b as a white crystalline solid (70 mg, 0.264 mmol, 46%).
Some crystals were of sufficient quality for X-ray analysis. 1H
NMR (DMSO-d6, 300 MHz) d 7.45 (m, 4H), 7.28–6.93 (m, 7H),
3.79 (dd, J = 8.7, 7.2 Hz, 1H), 3.74–3.61 (m, 1H), 3.57 (dd, J = 8.7,
3.3 Hz, 1H), 2.98–2.78 (m, 1H), 2.45–2.31 (m, 1H), 2.15–1.96 (m,
1H), 1.87–1.45 (m, 3H); 13C NMR (CDCl3, 75 MHz) d 131.6 (broad),
127.5, 126.2, 67.9, 62.9, 49.8, 31.0, 26.7; MS (ESI+) m/z 266.17
[M+H]+; HRMS (ESI+) m/z calcd for C17H21BNO 266.1711 [M+H]+,
found 266.1709 [M+H]+.
5. Experimental section
5.1. Chemistry
Commercial solvents and reagents were used without further
purification. Dry solvents were filtered over columns of dried alu-
mina under a positive pressure of argon. The glassware was dried
by heating while flushing with argon or nitrogen. The reactions
were usually run under argon or nitrogen atmosphere. For flash
column chromatography, Fluka analytical 60752 silica gel, pore
size 60 Å, particle size 40–63 lm, was used. For purification by pre-
parative reverse-phase HPLC, a Waters Prep LC system equipped
with a C18 column was used. Detection was done with a Waters
2489 UV/Visible Detector at 214 nm wavelength. Elution was done
with MilliQ water/HPLC-grade MeCN +0.1% trifluoroacetic acid. The
gradient program was 10 min isocratic H2O/MeCN, 82:18 v/v fol-
lowed by a 20 min gradient to H2O/ACN, 10:90 v/v, the flow rate
5.1.3. (3aR)-1,1-Diphenylhexahydro-1H-pyrrolo[1,2-
c][1,3,2]oxazaborol-7-ium-1-uide (2c)
Following the procedure for 2a, (R)-(ꢀ)-2-(hydroxymethyl)-
pyrrolidine (35 mg, 0.346 mmol) was added to diphenylborinic
anhydride (102 mg, 0.295 mmol) in MeCN at 0 °C. The product 2c
was obtained as a white crystalline solid (42 mg, 0.158 mmol,
54%). 1H NMR (DMSO-d6, 300 MHz) d 7.45 (m, 4H), 7.24–6.96 (m,
7H), 3.79 (dd, J = 8.7, 7.0 Hz, 1H), 3.74–3.60 (m, 1H), 3.57 (dd,
J = 8.7, 3.4 Hz, 1H), 2.93–2.77 (m, 1H), 2.47–2.31 (m, 1H), 2.16–
1.95 (m, 1H), 1.88–1.46 (m, 3H); 13C NMR (CDCl3, 75 MHz) d
131.9 (broad), 127.5, 126.1, 67.9, 62.9, 49.8, 31.0, 26.7; MS (ESI+)
m/z 266.17 [M+H]+; HRMS (ESI+) m/z calcd for C17H21BNO
266.1711 [M+H]+, found 266.1715 [M+H]+.
was 60 mL minꢀ1
.
1H NMR and 13C NMR spectra were measured
on
a
Bruker Avance spectrometer (1H: 300.18 MHz, 13C:
75.48 MHz). 11B NMR spectra were measured by the NMR service
team in the Department of Chemistry and Biochemistry, University
of Bern (Dr. J. Furrer) on a Bruker Avance II spectrometer (11B:
128.38 MHz). The analytical UHPLC chromatograms were recorded
with a Dionex UltiMate 3000 Rapid Separation LC system equipped
with a C8 or C18 reverse-phase column. Detection was done by UV
absorption measurement at 214 nm and 254 nm wavelength. The
eluent was composed of MilliQ water/HPLC-grade MeCN + 0.1% tri-
fluoroacetic acid. A gradient of H2O/MeCN, 64:36–10:90 v/v in a
6 min program was used, the flow rate was 1.2 mL minꢀ1. All the
MS and HRMS spectra were obtained from the Mass Spectrometry
Group of the Department of Chemistry and Biochemistry, Univer-
sity of Bern (PD Dr. S. Schürch). ESI MS was performed on a Micro-
mass Platform (quadrupole mass spectrometer), NSI MS spectra
were recorded on an Applied Biosystems/Sciex QSTAR Pulsar Sys-
tem (hybrid quadrupole time-of-flight mass spectrometer). The
X-ray structures were measured and solved by the Group of Chem-
ical Crystallography of the Department of Chemistry and Biochem-
istry, University of Bern (PD Dr. P. Macchi). The measurements
were made on a Oxford Diffraction SuperNova area-detector dif-
5.1.4. 2,2-Diphenyl-1,3,2-oxazaborinan-3-ium-2-uide (2d)
Following the procedure for 2a, 3-amino-1-propanol (48 mg,
0.639 mmol) was added to diphenylborinic anhydride (102 mg,
0.295 mmol) in MeCN at 0 °C. The product 2d was obtained as a
white crystalline solid (79 mg, 0.330 mmol, 61% from the 2 equiv
of borinic acid obtained from its anhydride). 1H NMR (DMSO-d6,
300 MHz) d 7.42 (d, J = 6.8 Hz, 4H), 7.12 (t, J = 7.2 Hz, 4H), 6.99 (t,
J = 7.2 Hz, 2H), 5.86 (s, 2H, broad), 3.67 (t, J = 4.9 Hz, 2H), 2.86 (s,
2H, broad), 1.60 (s, 2H, broad); 13C NMR (CDCl3, 75 MHz) d 131.6,
127.7, 126.2, 60.6, 40.9, 27.5; MS (ESI+) m/z 240.16 [M+H]+; HRMS
(ESI+) m/z calcd for C15H19BNO 240.1554 [M+H]+, found 240.1560
[M+H]+.

3208
A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
5.1.5. 1,1-Diphenyl-2-oxa-5,8-diaza-1-boraspiro[4.5]decan-5-
ium-1-uide (2e)
5.1.10. (4S)-4-((1H-Indol-3-yl)methyl)-5-oxo-2,2-diphenyl-
1,3,2-oxazaborolidin-3-ium-2-uide (2j)
Following the procedure for 2a, 1-(2-hydroxyethyl)piperazine
(54 mg, 0.415 mmol) was added to diphenylborinic anhydride
(108 mg, 0.312 mmol) in MeCN at 0 °C. The product 2e precipitated
as a white powder as its diphenyl borinic acid salt and was dried
under high vacuum (96 mg, 0.202 mmol, 65%). 1H NMR (DMSO-
d6/D2O/CD3CN, 300 MHz) d 7.75–7.53 (m, 8H), 7.29 (m, 12H),
3.46 (t, J = 6.2 Hz, 2H), 2.66 (m, 4H, broad), 2.50 (m, 4H, superim-
posed by DMSO signal), 2.37 (t, J = 6.2 Hz, 2H) (2 exchangeable pro-
tons not visible); MS (ESI+) m/z 295.20 [M+H]+; HRMS (ESI+) m/z
calcd for C18H24BN2O 295.1982 [M+H]+, found 295.1978 [M+H]+.
Following the procedure for 2h, diphenylborinic anhydride
(200 mg, 0.578 mmol) was reacted with L-tryptophan (136 mg,
0.665 mmol) in MeCN at 50 °C for 3 h. The mixture was cooled to
0 °C and stored at ꢀ15 °C overnight. 2j formed as beige solid which
was collected by filtration and dried (69 mg, 0.187 mmol, 32%). 1
H
NMR (DMSO-d6, 300 MHz) d 10.94 (d, J = 1.3 Hz, 1H), 7.46–6.93 (m,
15H), 6.49 (dd, J = 11.0, 9.2 Hz, 1H), 3.78–3.68 (m, 1H), 3.24 (dd,
J = 15.1, 3.8 Hz, 1H), 3.03 (dd, J = 15.1, 10.1 Hz, 1H); 13C NMR (ace-
tone-d6, 75 MHz) d 174.0, 137.7, 132.1, 131.9, 127.9, 127.8, 127.0,
126.9, 125.4, 122.5, 119.9, 119.0, 112.3, 109.3, 57.0, 26.5; MS
(ESI+) m/z 369.18 [M+H]+, 391.16 [M+Na]+; HRMS (ESI+) m/z calcd
for C23H22BN2O2 369.1774 [M+H]+, found 369.1769 [M+H]+.
5.1.6. 1,1-Diphenyl-1,3-dihydro-[1,3,2]oxazaborolo[3,4-
a]pyridin-8-ium-1-uide (2f)
5.1.11. 2,2-Diphenyl-1,3,2-thiazaborolidin-3-ium-2-uide (2k)
Following the procedure for 2a, cysteamine (51 mg,
0.661 mmol) was added to diphenylborinic anhydride (200 mg,
0.578 mmol) in MeCN at 0 °C. The product was obtained as a crys-
talline solid (61 mg, 0.253 mmol, 44%). Some crystals were of suf-
ficient quality for X-ray analysis. 1H NMR (DMSO-d6, 300 MHz) d
7.40 (m, 4H), 7.12 (m, 4H), 7.03 (m, 2H), 6.62 (br s, 2H), 3.00 (p,
J = 6.0 Hz, 2H), 2.67 (t, J = 6.5 Hz, 2H); 13C NMR (acetone-d6,
75 MHz) d 135.7, 133.6, 128.5, 52.3, 38.2; MS (ESI+) m/z 242.12
[M+H]+, 264.19 [M+Na]+; HRMS (ESI+) m/z calcd for C14H17BNS
242.1169 [M+H]+, found 242.1178 [M+H]+.
Following the procedure for 2a, 2-pyridinemethanol (48 mg,
0.440 mmol) was added to diphenylborinic anhydride (108 mg,
0.312 mmol) in MeCN at 0 °C. The product was obtained as a white
crystalline solid (59 mg, 0.216 mmol, 69%). 1H NMR (DMSO-d6,
300 MHz) d 8.64 (d, J = 5.6 Hz, 1H), 8.26 (t, J = 7.8 Hz, 1H), 7.87
(d, J = 7.8 Hz, 1H), 7.72 (t, J = 6.5 Hz, 1H), 7.33 (d, J = 6.5 Hz, 4H),
7.25–7.08 (m, 6H), 5.18 (s, 2H); 13C NMR (CDCl3, 75 MHz) d
159.7, 141.3, 140.6, 132.7, 127.4, 126.5, 123.9, 120.1, 69.0; MS
(ESI+) m/z 274.14 [M+H]+; HRMS (ESI+) m/z calcd for C18H17BNO
274.1403 [M+H]+, found 274.1395 [M+H]+.
5.1.7. 2,2-Diphenyl-2H-[1,3,2]oxazaborolo[5,4,3-ij]quinolin-3-
ium-2-uide (2g)
5.1.12. Hydroxydiphenylborane (3)
2-APB (1, 180 mg, 0.80 mmol) was dissolved in 0.8 mL acetone/
MeOH, 1:1 v/v and 1 mL of 1 M aq. HCl was added. The mixture
was stirred at rt for 1 h. The mixture was then extracted with
Et2O. The combined organic layers were washed with water and
brine and dried over anhydrous Na2SO4. Removal of the solvent
in vacuo and drying of the residue under high vacuum yielded
the product as a beige-white solid (83 mg, 0.456 mmol, 57%). 1H
NMR (DMSO-d6, 300 MHz) d 9.99 (s, 1H), 7.77–7.66 (m, 4H),
7.55–7.38 (m, 6H); 13C NMR (CDCl3, 75 MHz) d 135.9, 131.4,
128.0; MS (ESIꢀ) m/z 181.08 [MꢀH]ꢀ; HRMS (ESIꢀ) m/z calcd for
Following the procedure for 2a, 8-hydroxyquinoline (97 mg,
0.668 mmol) was added to diphenylborinic anhydride (200 mg,
0.578 mmol) in MeCN at 0 °C. The reaction mixture rapidly turned
yellow and was stirred for 1 h. The product was obtained as a yel-
low powder (177 mg, 0.573 mmol, 99%). 1H NMR (DMSO-d6,
300 MHz) d 9.16 (d, J = 4.9 Hz, 1H), 8.81 (d, J = 8.1 Hz, 1H), 7.92
(dd, J = 8.1, 4.9 Hz, 1H), 7.73 (t, J = 8.0 Hz, 1H), 7.55–7.32 (m, 5H),
7.32–7.11 (m, 7H); 13C NMR (CDCl3, 75 MHz) d 158.8, 139.4,
138.8, 137.6, 132.9, 132.1, 128.5, 127.7, 127.1, 122.8, 112.3,
109.7; MS (ESI+) m/z 310.14 [M+H]+; HRMS (ESI+) m/z calcd for
C
12H10BO 180.0819 [MꢀH]ꢀ, found 180.0828 [MꢀH]ꢀ.
C
21H17BNO 310.1398 [M+H]+, found 310.1402 [M+H]+.
5.1.13. 2-(Benzhydryloxy)ethanamine (4)
5.1.8. 5-Oxo-2,2-diphenyl-1,3,2-oxazaborolidin-3-ium-2-uide
(2h)
In the first step, diphenylmethanol (5 g, 27 mmol), 2-bromoeth-
anol (3.35 g, 27 mmol) and p-TsOH (100 mg, 0.53 mmol) were dis-
solved in toluene (50 mL) and refluxed in a Dean–Stark apparatus
for 3 h (approx. 0.5 mL H2O formed). The solvent was removed un-
der reduced pressure. The residue was diluted with Et2O and
washed with aq. K2CO3, H2O and brine. The organic phase was
dried over Na2SO4, evaporated under reduced pressure and dried
in vacuo. This gave ((2-bromoethoxy)methylene)dibenzene
(7.72 g, 26.5 mmol, 98%) as a white crystalline solid.
In the second step, ((2-bromoethoxy)methylene)dibenzene
(798 mg, 2.71 mmol) from above, phthalimide (398 mg,
2.71 mmol), and K2CO3 (550 mg, 3.98 mmol) in DMF (13.5 mL)
were stirred at 120 °C overnight. After 16 h, the solid was removed
and the filtrate was concentrated. The residue was dissolved in
CH2Cl2, washed with 0.2 M aq NaOH and H2O, dried over Na2SO4
and concentrated. The crude product was purified by washing with
hexane/Et2O 9:1 (product soluble in pure Et2O) which yielded 2-(2-
(benzhydryloxy)ethyl)isoindoline-1,3-dione (422 mg, 1.18 mmol,
44%) as a white crystalline solid.
Diphenylborinic anhydride (200 mg, 0.578 mmol) was sus-
pended in 5 mL of MeCN and glycine (50 mg, 0.665 mmol) was
added. The mixture was stirred for 2.5 h at 50 °C. The solvent
was then removed in vacuo and a beige-brown sticky solid was ob-
tained, which was washed with cold H2O and hexane to obtain 2h
as a light beige powder (103 mg, 0.431 mmol, 75%). 1H NMR
(DMSO-d6, 300 MHz) d 7.38 (d, J = 6.7 Hz, 4H), 7.31–7.12 (m, 6H),
7.12–7.01 (m, 2H, broad), 3.45 (t, J = 6.4 Hz, 2H); 13C NMR (ace-
tone-d6, 75 MHz) d 172.4, 132.2, 128.1, 127.2, 44.1; MS (ESI+) m/
z
240.12 [M+H]+; HRMS (ESI+) m/z calcd for C14H15BNO2
240.1190 [M+H]+, found 240.1183 [M+H]+.
5.1.9. (4S)-4-(Hydroxymethyl)-5-oxo-2,2-diphenyl-1,3,2-
oxazaborolidin-3-ium-2-uide (2i)
Following the procedure for 2h, diphenylborinic anhydride
(200 mg, 0.578 mmol) was reacted with
L-serine (70 mg,
0.666 mmol) in MeCN at 50 °C. This gave 2i as a light beige powder
(68 mg, 0.253 mmol, 44%). 1H NMR (DMSO-d6, 300 MHz) d 7.58–
7.33 (m, 5H), 7.31–7.08 (m, 6H), 6.29 (dd, J = 10.8, 7.4 Hz, 1H),
5.22 (t, J = 5.0 Hz, 1H), 3.83–3.58 (m, 3H); 13C NMR (acetone-d6,
75 MHz) d 172.7, 132.4, 132.3, 128.1, 128.1, 127.3, 59.8, 58.1; MS
(ESI+) m/z 270.13 [M+H]+; HRMS (ESI+) m/z calcd for C15H17BNO3
270.1296 [M+H]+, found 270.1296 [M+H]+.
In the third step, 2-(2-(benzhydryloxy)ethyl)isoindoline-1,3-
dione (220 mg, 0.616 mmol) from above was suspended in EtOH
(2 mL) and hydrazine (274 mg, 3 mmol) was added. The mixture
was heated to reflux for 5 min (the powder dissolved and a white
solid appeared shortly after). After cooling the reaction mixture,
the solid phthalazine formed was removed and the filtrate

A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
3209
concentrated. The residue was dissolved in CH2Cl2 and HCl/dioxane
(1 mL) was added. The mixture was evaporated and subsequently
co-evaporated with CH2Cl2 (3 ꢂ 5 mL). The white solid was washed
with Et2O and Et2O/CH2Cl2 and purified by flash chromatography
(SiO2; CH2Cl2/MeOH/aq NH3 95:5:0.25) to obtain 4 as a pale yellow
oil (21 mg, 0.092 mmol, 15%). 1H NMR (DMSO-d6, 300 MHz) d
7.43–7.27 (m, 8H), 7.27–7.19 (m, 2H), 5.45 (s, 1H), 3.39 (t,
J = 5.7 Hz, 2H), 2.78 (t, J = 5.7 Hz, 2H); MS (ESI+) m/z 228.14
[M+H]+; HRMS (ESI+) m/z calcd for C15H18NO 228.1388 [M+H]+,
found 228.1376 [M+H]+.
NMR (acetone-d6, 75 MHz) d 163.1, 135.3, 127.6, 120.9, 110.2,
63.5, 55.4, 42.1; MS (ESI+) m/z 286.16 [M+H]+; HRMS (ESI+) m/z
calcd for C16H21BNO3 286.1609 [M+H]+, found 286.1606 [M+H]+.
5.1.18. 2,2-Bis(4-(trifluoromethyl)phenyl)-1,3,2-oxazaborolidin-
3-ium-2-uide (7b)
Following the procedure for 5a, 4-bromobenzotrifluoride
(475
1.69 mL, 4.225 mmol) and reacted with triisopropyl borate
(350
L, 1.515 mmol) in Et2O at ꢀ78 °C. The crude (brownish slur-
ry) was dissolved in EtOH/H2O 3:1 v/v (2 mL) and 2-aminoethanol
(100 L, 1.654 mmol) was added. The mixture was stirred at rt for
lL, 3.392 mmol) was lithiated with n-BuLi (2.5 M in hexanes,
l
5.1.14. Hydroxydi-p-tolylborane (5a)
l
Bromotoluene (750 mg, 4.385 mmol) was dissolved in dry Et2O
(8 mL) and the solution was cooled to ꢀ78 °C. s-BuLi solution
(1.4 M in cyclohexane, 3.15 mL, 4.410 mmol) was added slowly.
The mixture was allowed to reach 0 °C over 2 h, and then cooled
to ꢀ78 °C again. It was then transferred to a solution of triisopropyl
1 h, then for 1 h at 50 °C. The mixture was then poured into 10 mL
of Et2O. The organic layer was washed with H2O and brine, dried
over Na2SO4 and the solvent removed to obtain a brownish yellow
oil with white precipitate. The mixture was filtered and the solid
washed with chloroform to obtain 7b as a yellow-white solid
(217 mg, 0.601 mmol, 40%). 1H NMR (DMSO-d6, 300 MHz) d 7.63
(d, J = 7.8 Hz, 4H), 7.50 (d, J = 7.8 Hz, 4H), 6.38 (s, 2H), 3.81 (t,
J = 6.2 Hz, 2H), 2.89 (p, J = 6.2 Hz, 2H); 13C NMR (acetone-d6,
borate (450
lL, 1.949 mmol) in dry Et2O (10 mL) cooled to ꢀ78 °C.
The mixture was stirred and allowed to gradually reach rt over-
night. 1 M aq HCl (10 mL) was then added to stop the reaction,
and the biphasic mixture was stirred for 15 min at rt. The organic
layer was collected and the aqueous layer extracted three times
with Et2O. All organic layers were combined, dried over Na2SO4
and concentrated in vacuo to obtain the crude product as yellow
oil. One aliquot of the crude (85 mg) was purified by flash chroma-
tography (SiO2; hexane/ethyl acetate, 9:1) to obtain 5a as a white
solid (66 mg, 0.314 mmol, 16%). A second aliquot of crude product
was used directly for the esterification with 2-aminoethanol (vide
infra). 1H NMR (DMSO-d6, 300 MHz) d 9.82 (s, 1H), 7.61 (d,
J = 7.8 Hz, 4H), 7.24 (d, J = 7.8 Hz, 4H), 2.36 (s, 6H); 13C NMR (CDCl3,
75 MHz) d 141.2, 135.8, 134.9, 128.8, 21.7. MS ionization was not
possible with the available instrumentation used.
75 MHz)
d 134.6, 133.0, 129.6 (d, J = 270.8 Hz), 128.4 (q,
J = 31.3 Hz), 124.2 (q, J = 3.4 Hz), 62.1, 54.3; MS (ESI+) m/z 362.11
[M+H]+; HRMS (ESI+) m/z calcd for C16H15BF6NO 362.1145
[M+H]+, found 362.1147 [M+H]+.
5.1.19. Hydroxybis(4-(trifluoromethyl)phenyl)borane (7a)
Borinic ester 7b was dissolved in acetone/MeOH 1:1 v/v (1 mL),
and 1 M aq HCl (1 mL) was added. The mixture was stirred at rt for
1 h, then poured into Et2O. The ether layer was washed with H2O
and brine and dried over Na2SO4. The solvent was then removed
in vacuo and the residue dried under high vacuum to obtain 7a
as a yellow-white powder (56 mg, 0.176 mmol, 64%). 1H NMR
(DMSO-d6, 300 MHz) d 7.64 (d, J = 7.8 Hz, 4H), 7.50 (d, J = 7.8 Hz,
4H), 4.99 (s, 1H); 13C NMR (CDCl3, 75 MHz) d 135.9, 134.8, 133.1
(d, J = 32.5 Hz), 124.8 (q, J = 3.9 Hz), 124.0 (d, J = 272.4 Hz); MS ion-
ization was not possible with the available instrumentation used.
5.1.15. 2,2-Di-p-tolyl-1,3,2-oxazaborolidin-3-ium-2-uide (5b)
Crude 5a (150 mg, ca. 0.714 mmol) was dissolved in EtOH
(4 mL) and 2-aminoethanol (60 lL, 0.994 mmol) was added. The
reaction mixture was stirred for 1 h at rt. The product precipitated
from the mixture after partial removal of the solvent in vacuo and
was washed with Et2O to yield 5b as a white crystalline powder
(94 mg, 0.371 mmol, 19% (for 2 steps)). 1H NMR (DMSO-d6,
300 MHz) d 7.26 (d, J = 7.7 Hz, 4H), 6.95 (d, J = 7.7 Hz, 4H), 5.94
(s, 2H, broad), 3.75 (t, J = 6.4 Hz, 2H), 2.81 (p, J = 6.4 Hz, 2H), 2.22
(s, 6H); 13C NMR (acetone-d6, 75 MHz) d 135.1, 134.3, 128.3,
61.8, 54.1, 21.3; MS (ESI+) m/z 254.17 [M+H]+; HRMS (ESI+) m/z
calcd for C16H21BNO 254.1711 [M+H]+, found 254.1713 [M+H]+.
5.1.20. 2,2-Bis(4-fluorophenyl)-1,3,2-oxazaborolidin-3-ium-2-
uide (8b)
Following the procedure for 5a, 1-bromo-4-fluorobenzene (1 g,
5.71 mmol) was lithiated with n-BuLi (2.5 M in hexanes, 2.74 mL,
6.84 mmol) and reacted with triisopropyl borate (467 mg,
2.48 mmol) in Et2O at ꢀ78 °C. The crude product was dissolved
in MeCN (14 mL), cooled to 0 °C and 2-aminoethanol (167 mg,
2.73 mmol) in MeCN (2 mL) was added. After 30 min a precipitate
formed. Stirring was continued for 1 h, the solid was filtered off,
washed with MeCN (5 mL) and dried in vacuo. This yielded 8b as
off-white crystalline solid (343 mg, 1.31 mmol, 53%). 1H NMR
(DMSO-d6, 300 MHz) d 7.45–7.26 (m, 4H), 7.01–6.87 (m, 4H),
6.03 (br s, 2H), 3.75 (t, J = 6.5 Hz, 2H), 2.82 (p, J = 6.3 Hz, 2H); 13C
NMR (CDCl3, 75 MHz) d 177.7, 162.2 (d, J = 243.0 Hz), 134.6 (d,
J = 6.9 Hz), 114.0 (d, J = 19.1 Hz), 61.1, 53.5; MS (ESI+) m/z 262.12
[M+H]+; HRMS (ESI+) m/z calcd for C14H15BF2NO 262.1215
[M+H]+, found 262.1209 [M+H]+.
5.1.16. Hydroxybis(2-methoxyphenyl)borane (6a)
Following the procedure for 5a, 2-bromoanisole (570
4.611 mmol) was lithiated with n-BuLi (2.5 M in hexanes,
1.88 mL, 4.7 mmol) and reacted with triisopropyl borate (430 L,
lL,
l
1.863 mmol) in Et2O at ꢀ78 °C. The crude product (yellow oil)
was purified by column chromatography (SiO2; hexane/ethyl ace-
tate, 19:1) to obtain product 6a as a white solid (129 mg,
0.533 mmol, 29%). 1H NMR (DMSO-d6, 300 MHz) d 9.40 (s, 1H),
7.37 (ddd, J = 8.3, 7.4, 1.8 Hz, 2H), 7.24 (dd, J = 7.2, 1.8 Hz, 2H),
7.02–6.86 (m, 4H), 3.70 (s, 6H); MS (ESI+) m/z 243.12 [M+H]+;
HRMS (ESI+) m/z calcd for C14H16BO3 243.1192 [M+H]+, found
243.1186 [M+H]+.
5.1.21. Bis(4-fluorophenyl)(hydroxy)borane (8a)
Borinic ester 8b (50 mg, 0.19 mmol) was hydrolyzed following
the procedure for 7a, which gave 8a (37 mg, 0.17 mmol, 90%). 1H
NMR (DMSO-d6, 300 MHz) d 9.85 (s, 1H), 7.79–7.65 (m, 4H),
7.29–7.17 (m, 4H); MS ionization was not possible with the avail-
able instrumentation used.
5.1.17. 2,2-Bis(2-methoxyphenyl)-1,3,2-oxazaborolidin-3-ium-
2-uide (6b)
Following the procedure for 5b, 6a (78 mg, 0.322 mmol) was
treated with 2-aminoethanol (25 lL, 0.414 mmol) which gave the
product 6b as a white crystalline powder (35 mg, 0.123 mmol,
38%). 1H NMR (DMSO-d6, 300 MHz) d 7.12 (dd, J = 7.1, 1.7 Hz,
2H), 7.09–7.01 (m, 2H), 6.81–6.67 (m, 4H), 5.99 (s, 2H, broad),
3.68 (s, 6H), 3.65 (t, J = 6.3 Hz, 2 H), 2.91 (p, J = 6.3 Hz, 2H); 13C
5.1.22. 2,2-Bis(4-methoxyphenyl)-1,3,2-oxazaborolidin-3-ium-
2-uide (9b)
Following the procedure for 5a, 4-bromoanisole (1 g,
5.35 mmol) was lithiated with n-BuLi (2.5 M in hexanes, 2.61 mL,

3210
A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
6.52 mmol) and reacted with triisopropyl borate (0.44 g,
oil. Drying under high vacuum yielded crude phenylboronic acid
as a beige solid (1.41 mg, ca. 11.56 mmol, 90%). A portion of this
crude product (400 mg, ca. 2.558 mmol) and 2,2-dimethyl-1,3-pro-
panediol (376 mg, 3.609 mmol) were dissolved in THF (9 mL) and
the mixture was stirred at rt for 3.5 h. The solvent was then re-
moved in vacuo and the obtained yellow viscous liquid was dis-
solved in CH2Cl2. The solution was washed with H2O, dried over
Na2SO4 and the solvent was removed in vacuo to obtain the title
compound as a yellow-white solid (613 mg, 3.225 mmol, 98%).
1H NMR (DMSO-d6, 300 MHz) d 7.71 (d, J = 8.0 Hz, 2H), 7.56–7.28
(m, 3H), 3.77 (s, 4H), 0.97 (s, 6H).
2.33 mmol) in Et2O at ꢀ78 °C. This gave crude 9a as a colorless
oil. To this crude, 2-aminoethanol (160 ll, 2.56 mmol) in EtOH
was added. The mixture was left to stirring at rt for 2 h and then
stored at ꢀ20 °C. No precipitation was obtained within 11 days.
Different solvent systems and crystallization methods were tried,
and eventually the residue was dissolved in a mixture of EtOAc/
CH2Cl2 (5 mL) and stored at ꢀ20 °C. The precipitate which had
formed after 3 days was collected by filtration to yield 9b as a
white solid (20 mg, 0.07 mmol, 1%). 1H NMR (DMSO-d6,
300 MHz) d 7.25 (d, J = 8.5 Hz, 4H), 6.71 (d, J = 8.5 Hz, 4H), 5.88
(br s, 2H), 3.73 (t, J = 6.4 Hz, 2H), 3.67 (s, 6H), 2.80 (p, J = 6.2 Hz,
2H); MS (ESI+) m/z 286.16 [M+H]+, 308.14 [M+Na]+; HRMS (ESI+)
m/z calcd for C16H21BNO3 286.1609 [M+H]+, found 286.1607
[M+H]+.
5.1.26. 2-(4-Fluorophenyl)-2-phenyl-1,3,2-oxazaborolidin-3-
ium-2-uide (13b)
4-Fluorophenylboronic acid neopentylglycol ester (187 mg,
0.9 mmol) was dissolved in dry Et2O (8 mL) and the solution cooled
to ꢀ78 °C. PhLi (1.8 M in n-Bu2O, 0.5 mL, 0.9 mmol) was then
added slowly to the solution and the mixture was stirred at
ꢀ78 °C. The reaction was quenched after 30 min at ꢀ78 °C with
5 mL 1 M aq HCl, and the biphasic mixture was stirred for
15 min at rt. The organic layer was then collected and the aqueous
layer extracted with Et2O. All organic layers were combined, dried
over Na2SO4 and concentrated in vacuo to obtain a yellow-orange
viscous oil. This residue was dissolved in EtOH/H2O 3:1 (4 mL) and
5.1.23. 1,2-Bis(2-bromophenyl)ethane (10)
2-Bromobenzyl bromide (3.70 g, 14.80 mmol) was dissolved in
dry THF (20 mL). The solution was then cooled to ꢀ78 °C and n-
BuLi, (2.5 M in hexanes, 2.9 mL, 7.250 mmol) was added dropwise.
After 45 min of stirring at ꢀ78 °C, the reaction was quenched with
H2O (10 mL). The biphasic mixture was allowed to reach rt. The
layers were separated and the water phase extracted with Et2O.
The combined organic layers were washed with satd NH4Cl, dried
over Na2SO4 and concentrated under reduced pressure to obtain
a white solid. Purification by flash column chromatography (SiO2,
petrol ether) yielded 10 as a white crystalline solid (1.87 g,
2-aminoethanol (60 lL, 1.000 mmol) was added. The reaction mix-
ture was stirred at 50 °C for 1 h and then poured into Et2O. The or-
ganic layer was collected, and the aqueous layer extracted with
Et2O. All organic layers were combined, dried over Na2SO4 and
the solvent was removed in vacuo to obtain a yellow oil. This crude
product was dissolved in chloroform and the solution stored at
5.50 mmol, 76%). 1H NMR (DMSO-d6, 300 MHz)
d 7.61 (d,
J = 7.9 Hz, 2H), 7.36–7.31 (m, 4H), 7.24–7.13 (m, 2H), 2.99 (s, 4H).
5.1.24. 10,11-Dihydro-5H-dibenzo[b,f]borepin-5-ol (11)
ꢀ20 °C. The product crystallized as
a white solid (96 mg,
Following the procedure for 5a, bisbromide 10 (809 mg,
2.379 mmol) was lithiated with s-BuLi (1.5 M in cyclohexanes,
3.7 mL, 5.550 mmol) and reacted with triisopropyl borate
0.395 mmol, 44%). 1H NMR (DMSO-d6, 300 MHz) d 7.51–7.31 (m,
4H), 7.15 (t, J = 7.2 Hz, 2H), 7.05 (t, J = 7.2 Hz, 1H), 6.94 (dd,
J = 9.8, 8.6 Hz, 2H), 6.09 (br s, 2H), 3.78 (t, J = 6.5 Hz, 2H), 2.84 (p,
(540
l
L, 2.340 mmol) in Et2O (10 mL) at ꢀ78 °C. The crude (orange
J = 6.5 Hz, 2H); 13C NMR (acetone-d6, 75 MHz)
d 162.7 (d,
viscous oil) was purified by flash column chromatography (SiO2,
J = 240.9 Hz), 135.9 (d, J = 6.9 Hz), 134.4 (d, J = 6.8 Hz), 134.1,
132.6, 127.6, 126.5, 114.0 (d, J = 19.0 Hz), 61.8, 54.1; MS (ESI+)
m/z 244.13 [M+H]+; HRMS (ESI+) m/z calcd for C14H16BFNO
244.1303 [M+H]+, found 244.1305 [M+H]+.
hexane/ethyl acetate, 9:1) to obtain impure product. The solid
was dissolved in EtOH (4 mL) and 2-aminoethanol (100 lL,
1.654 mmol) was added. The mixture was stirred at rt for 2 h,
and a beige precipitate formed which was the impure 2-amino-
ethyl ester of 11. This precipitate was dissolved in acetone/MeOH
(0.8 mL) and 1 M aq HCl (1 mL) was added. The mixture was stirred
for 2 h at rt and then poured into Et2O (10 mL). The organic layer
was washed with H2O and the aqueous layer then extracted with
Et2O. All combined organic layers were dried over Na2SO4 and
the solvent was removed to obtain impure 11 as an orange oil.
Purification by flash column chromatography (SiO2, hexane/ethyl
acetate, 95:5 and hexane/ethyl acetate, 9:1) and preparative TLC
(SiO2, hexane/ethyl acetate, 9:1) yielded 11 as a beige sticky solid
(17 mg, 0.082 mmol, 3%). 1H NMR (DMSO-d6, 300 MHz) d 9.84 (s,
1H), 7.94 (dd, J = 7.4, 1.4 Hz, 2H), 7.35 (td, J = 7.4, 1.4 Hz, 2H),
7.27–7.16 (m, 4H), 3.04 (s, 4H); 13C NMR (CDCl3, 75 MHz) d
150.8, 136.0, 133.4, 131.7, 129.0, 125.7, 37.1; MS ionization was
not possible with the available instrumentation used.
5.1.27. 2-Phenyl-2-(o-tolyl)-1,3,2-oxazaborolidin-3-ium-2-uide
(14b)
2-Iodotoluene (128 lL, 1.000 mmol) was dissolved in dry Et2O
(8 mL) and the solution cooled to ꢀ78 °C. n-BuLi (2.5 M in hexanes,
0.5 mL, 1.250 mmol) was added slowly to the solution and the mix-
ture was stirred at ꢀ78 °C for 45 min. Then, freshly prepared 12
(190 mg, 1 mmol) in Et2O (10 mL) was added at ꢀ78 °C and the
mixture was allowed to reach ꢀ20 °C over 5 h. 1 M aq HCl (5 mL)
was then added to quench the reaction. The biphasic mixture
was stirred for 45 min at rt. Then, the organic layer was collected
and the aqueous layer extracted with Et2O. All organic layers were
combined and concentrated in vacuo to obtain the crude product
as an orange sticky solid. This solid was dissolved in EtOH/H2O
3:1 (2 mL) and 2-aminoethanol (66 lL, 1.100 mmol) was added.
The reaction mixture was stirred at 50 °C for 1.5 h. It was then
cooled to rt and extracted with Et2O. All organic layers were com-
bined, dried over Na2SO4 and the solvent was removed in vacuo to
obtain a yellow sticky solid. The product was crystallized from
MeCN and washed with hexane and H2O to obtain a light yellow
crystalline solid (65 mg, 0.272 mmol, 27%). 1H NMR (DMSO-d6,
300 MHz) d 7.51 (d, J = 6.5 Hz, 1H), 7.29 (d, J = 6.8 Hz, 2H), 7.20–
6.83 (m, 6H), 6.18 (br s, 2H), 3.78 (br s, 1H), 3.69 (br s, 1H), 2.87
(br s, 2H), 2.15 (s, 3H); 13C NMR (acetone-d6, 75 MHz) d 143.5,
134.5, 134.2, 133.0, 130.7, 127.6, 126.7, 126.4, 124.8, 62.0, 53.7,
22.9; MS (ESI+) m/z 240.16 [M+H]+; HRMS (ESI+) m/z calcd for
5.1.25. 5,5-Dimethyl-2-phenyl-1,3,2-dioxaborinane (12)
Triisopropyl borate (2.95 mL, 12.80 mmol) was dissolved in dry
Et2O (30 mL) and the solution was cooled to ꢀ78 °C. PhLi (1.8 M in
n-Bu2O, 7.1 mL, 12.80 mmol) was added over 15 min, and the mix-
ture was stirred for 1 h at ꢀ78 °C. The mixture by then almost to-
tally solidified and the reaction was allowed to reach rt and was
quenched by adding 1 M aq HCl (15 mL). The biphasic mixture
was stirred at rt for 30 min. The ether layer was then collected
and the aqueous layer extracted with Et2O, basified with sat. NaH-
CO3 and extracted with Et2O. All organic layers were combined,
dried over Na2SO4 and the solvent removed to obtain a yellow
C
15H19BNO 240.1554 [M+H]+, found 240.1553 [M+H]+.

A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
3211
5.1.28. 2-Phenyl-2-(p-tolyl)-1,3,2-oxazaborolidin-3-ium-2-uide
(15b)
Following the procedure for 14b, 4-bromotoluene (1 g,
5.85 mmol) was lithiated with n-BuLi (2.5 M in hexanes, 3.51 mL,
8.77 mmol) and then reacted with phenylboronic acid pinacol ester
(24, 1.19 g, 5.85 mmol). The crude borinic acid 15a (colorless oil)
out further purification. 1H NMR for this intermediate (DMSO-d6,
300 MHz) d 7.78 (m, 2H), 7.66 (m, 4H), 7.47 (m, 2H), 7.38 (m,
1H), 3.78 (s, 4H), 0.97 (s, 6H).
Following the procedure for 13b, boronic acid neopentylglycol
ester from above (1.103 g, 3.82 mmol) was reacted with PhLi
(1.8 M in n-Bu2O, 2.2 mL) in Et2O/THF at ꢀ78 °C. The crude borinic
acid (yellow oil) was purified by washing it with hexane to obtain
17a as a white solid (750 mg). This was used in the next reaction
without further purification. 1H NMR for 17a (DMSO-d6,
300 MHz) d 9.98 (s, 1H), 7.85–7.68 (m, 6H), 7.53–7.35 (m, 8H).
Borinic acid 17a (750 mg, 1.50 mmol) was esterified with 2-
was esterified with 2-aminoethanol (380 lL, 6.44 mmol) in abso-
lute EtOH to give the title compound as white crystals (0.35 g,
1.5 mmol, 25%). 1H NMR (DMSO-d6, 300 MHz) d 7.37 (m, 2H),
7.26 (d, J = 7.8 Hz, 2H), 7.11 (m, 2H), 7.07–6.97 (m, 1H), 6.94 (d,
J = 7.5 Hz, 2H), 5.99 (br s, 2H), 3.74 (t, J = 6.5 Hz, 2H), 2.81 (p,
J = 6.3 Hz, 2H), 2.20 (s, 3H); 13C NMR (acetone-d6, 75 MHz) d
135.2, 134.7, 134.1, 132.7, 128.3, 127.4, 126.3, 125.9, 61.6, 53.9,
21.3; MS (ESI+) m/z 240.16 [M+H]+, 262.14 [M+Na]+; HRMS (ESI+)
aminoethanol (181 lL, 3 mmol) in absolute EtOH yielding the title
compound as white crystals (399 mg, 1.32 mmol, 88%). 1H NMR
(DMSO-d6, 300 MHz) d 7.58 (m, 2H), 7.51–7.38 (m, 8H), 7.29 (m,
1H), 7.14 (m, 2H), 7.04 (m, 1H), 6.11 (br s, 2H), 3.79 (t, J = 6.4 Hz,
2H), 2.94–2.77 (m, 2H); 13C NMR (acetone-d6, 75 MHz) d 142.6,
138.8, 134.8, 134.1, 133.3, 132.6, 129.5, 127.5, 127.4, 126.4,
126.0, 61.8, 54.0; MS (ESI+) m/z 302.17 [M+H]+, 324.15 [M+Na]+,
363.22 [M+2-aminoethanol+H]+; HRMS (ESI+) m/z calcd for
m/z calcd for
C
15H19BNO 240.1554 [M+H]+, found 240.1556
[M+H]+.
5.1.29. 2-(4-Butylphenyl)-2-phenyl-1,3,2-oxazaborolidin-3-
ium-2-uide (16b)
A solution of 1,4-diiodobenzene (567 mg, 1.72 mmol) in anhy-
drous THF (20 mL) was cooled to ꢀ78 °C. n-BuLi (2.5 M in hexanes,
1.44 mL, 3.6 mmol) was added drop-wise, keeping the temperature
below ꢀ70 °C. The white milky mixture was left to stirring for
1.5 h. An argon-bubbled solution of phenylboronic acid pinacol es-
ter (24, 735 mg, 3.43 mmol) in diethyl ether (10 mL) was added
slowly keeping the temperature below ꢀ70 °C. The cooling was
continued for 1 h and then the mixture was left to slowly warm
up to room temperature overnight. The reaction was quenched
using 1 M aq HCl (20 mL). The biphasic mixture was stirred for
15 min and then poured into Et2O (20 mL). The phases were sepa-
rated and the aqueous phase was extracted three times using Et2O
(40 mL each). The combined organic phases were washed with
brine (20 mL), dried over Na2SO4 and the solvent was removed in
vacuo giving a colorless oil. The crude was purified by silica gel
flash column chromatography using hexane/EtOAc (20:1 v/v) to af-
ford the borinic acid 16a as a yellowish oil (380 mg, 1.34 mmol,
78%). 1H NMR for 16a (DMSO-d6, 300 MHz) d 9.89 (s, 1H), 7.73–
7.65 (m, 2H), 7.62 (d, J = 7.9 Hz, 2H), 7.51–7.32 (m, 3H) 7.24 (d,
J = 7.9 Hz, 2H) 2.69–2.57 (m, 2H), 1.66–1.52 (m, 2H), 1.38–1.30
(m, 2H), 0.90 (t, J = 7.3 Hz, 3H).
C
20H21BNO 302.1711 [M+H]+, found 302.1708 [M+H]+.
5.1.31. 2-Phenyl-2-(pyridin-3-yl)-1,3,2-oxazaborolidin-3-ium-2-
uide (18b)
Following the procedure for 14b, 3-Bromopyridine (0.61 mL,
6.33 mmol) was lithiated with n-BuLi (2.5 M in hexanes, 2.6 mL,
6.50 mmol) and reacted with 24 (1.29 g, 6.33 mmol) in Et2O at
ꢀ78 °C. The crude borinic acid 18a (beige-white solid) was esteri-
fied with 2-aminoethanol (500 lL, 8.284 mmol) in EtOH to give
18b as a beige-white powder (210 mg, 0.93 mmol, 15%). 1H NMR
(DMSO-d6, 300 MHz) d 8.53 (s, 1H), 8.25 (d, J = 3.1 Hz, 1H), 7.67
(d, J = 7.3 Hz, 1H), 7.38 (d, J = 6.9 Hz, 2H), 7.26–6.96 (m, 4H), 6.24
(br s, 1H), 6.18 (br s, 1H), 3.78 (t, J = 6.4 Hz, 2H), 2.85 (m, 2H);
13C NMR (acetone-d6, 75 MHz) d 155.1, 147.6, 141.9, 134.1, 132.6,
127.8, 126.7, 123.3, 62.0, 54.2; MS (ESI+) m/z 227.14 [M+H]+;
HRMS (ESI+) m/z calcd for C13H16BN2O 227.1350 [M+H]+, found
227.1354 [M+H]+.
5.1.32. 2-(5-Methylthiophen-2-yl)-2-phenyl-1,3,2-
oxazaborolidin-3-ium-2-uide (19b)
Following the procedure for 13b, 5-methylthiophene-2-boronic
acid pinacol ester (1.07 mL, 4.46 mmol) was reacted with PhLi
(1.8 M in n-Bu2O, 2.5 mL, 4.50 mmol) in Et2O at ꢀ78 °C. The crude
borinic acid 19a (yellow sticky solid) was esterified with 2-amino-
Borinic acid 16a (380 mg, 1.60 mmol) from above was dissolved
in absolute EtOH (3 mL) and 2-aminoethanol added (215 lL,
3.56 mmol). The mixture was left to stirring at room temperature
for 2 h and then stored in the freezer. Precipitation was attempted
using different methods, of which only the most successful one is
described here: The solvent was removed in vacuo. The resulting
yellowish oil was mixed with some water and Et2O. The resulting bi-
phasic mixture was left to standing overnight. After 15 h a white
precipitate had formed, which was isolated by filtration and washed
with water giving the title compound as white powder (207 mg,
0.74 mmol, 46%). 1H NMR (DMSO-d6, 300 MHz) d 7.38 (m, 2H),
7.28 (m, 2H), 7.12 (m, 2H), 7.02 (m, 1H), 6.94 (m, 2H), 6.00 (br s,
2H), 3.75 (t, J = 6.5 Hz, 2H), 2.81 (p, J = 6.3 Hz, 2H), 2.49–2.39 (m,
2H), 1.64–1.37 (m, 2H), 0.87 (t, J = 7.3 Hz, 3H); 13C NMR (CD3OD,
75 MHz) d 141.1, 133.0, 132.9, 128.2, 128.0, 126.7, 63.9, 42.6, 36.5,
35.1, 23.3, 14.3; MS (ESI+) m/z 282.20 [M+H]+, 304.18 [M+Na]+,
343.25 [M+2-aminoethanol+H]+; HRMS (ESI+) m/z calcd for
ethanol (60 lL, 1.00 mmol) in EtOH to yield 19b as a white solid
(204 mg, 0.832 mmol, 19%) crystallized from hexane/CHCl3. 1H
NMR (DMSO-d6, 300 MHz) d 7.43 (d, J = 6.9 Hz, 2H), 7.29–6.96
(m, 3H), 6.66 (s, 1H), 6.61 (s, 1H), 6.08 (br s, 2H), 3.96–3.64 (m,
2H), 3.07–2.70 (m, 2H), 2.37 (s, 3H); 13C NMR (acetone-d6,
75 MHz) d 140.6, 134.0, 132.6, 130.9, 127.6, 126.7, 126.2, 62.0,
53.9, 15.3; MS (ESI+) m/z 268.09 [M+Na]+; HRMS (ESI+) m/z calcd
for C13H16BNNaOS 268.0938 [M+Na]+, found 268.0929 [M+Na]+.
5.1.33. (4-Chlorophenyl)(hydroxy)(phenyl)borane (20a)
4-Chlorophenylboronic acid (400 mg, 2.558 mmol) and 2,2-di-
methyl-1,3-propanediol (293 mg, 2.814 mmol) were dissolved in
THF (7 mL) and the mixture was stirred at rt for 2.5 h. The solvent
was then removed in vacuo and the obtained beige solid was dis-
solved in CH2Cl2. The solution was washed with H2O, dried over
Na2SO4 and the solvent was removed in vacuo to obtain para-chlo-
rophenyl boronic acid neopentylglycol ester as a white solid
(499 mg, 2.223 mmol, 87%). 1H NMR for this intermediate
(DMSO-d6, 300 MHz) d 7.71 (d, J = 8.3 Hz, 2H), 7.43 (d, J = 8.3 Hz,
2H), 3.77 (s, 4H), 0.97 (s, 6H).
C
18H25BNO 282.2024 [M+H]+, found 282.2019 [M+H]+.
5.1.30. 2-([1,10-Biphenyl]-4-yl)-2-phenyl-1,3,2-oxazaborolidin-
3-ium-2-uide (17b)
A solution of biphenyl-4-boronic acid (756 mg, 3.82 mmol) and
2,2-dimethyl-1,3-propanediol (437 mg, 4.20 mmol) in THF was
stirred at room temperature for 2 h. Removal of the solvent yielded
the crude boronic acid neopentylglycol ester as a white solid
(1.103 g, quantitative), which was used in the next reaction with-
Following the procedure for 13b, freshly prepared para-chloro-
phenyl boronic acid neopentylglycol ester from above (202 mg,
0.90 mmol) was reacted with PhLi (1.8 M in n-Bu2O, 0.5 mL,

3212
A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
0.90 mmol) in Et2O at ꢀ78 °C. The crude borinic acid 20a (beige
sticky solid) was esterified with 2-aminoethanol (60 L,
300 MHz) d 8.77 (s, 1H), 7.82 (d, J = 7.9 Hz, 2H), 7.70 (d,
l
J = 7.9 Hz, 2H), 7.64 (dd, J = 7.4, 1.9 Hz, 2H), 7.42–7.31 (m, 3H);
1.00 mmol) in EtOH/H2O 3:1. Attempts to crystallize the product
20b failed. Instead, purification by preparative RP-HPLC hydro-
lyzed the ester, and 20a was obtained as a beige solid (55 mg,
0.254 mmol, 28%). 1H NMR (DMSO-d6, 300 MHz) d 9.78 (s, 1H),
7.83–7.59 (m, 4H), 7.57–7.31 (m, 5H); 13C NMR (acetone-d6,
75 MHz) d 137.2, 135.7, 134.4, 131.7, 128.3, 128.1; MS (ESIꢀ) m/z
215.04 [MꢀH]ꢀ; HRMS (ESIꢀ) m/z calcd for C12H9BClO 215.0440
[MꢀH]ꢀ, found 215.0442 [MꢀH]ꢀ.
MS (ESIꢀ) m/z 249.07 [MꢀH]ꢀ; HRMS (ESIꢀ) m/z calcd for
C
13H9BF3O 249.0704 [MꢀH]ꢀ, found 249.0704 [MꢀH]ꢀ.
5.2. Pharmacology
Cell culture reagents as well as Lipofectamine 2000 were ob-
tained from Invitrogen Life Technologies Europe B.V. The
pTagRFP-C1 vector was purchased from Evrogen Joint Stock Com-
pany. Calcium-5 was bought from Molecular Devices LLC. 2-APB,
diphenhydramine hydrochloride was obtained from Tocris Biosci-
ence. All other, non-selfsynthesized chemicals were purchased
from Sigma or Aldrich.
5.1.34. 2-(4-Bromophenyl)-2-phenyl-1,3,2-oxazaborolidin-3-
ium-2-uide (21b)
Following the procedure for 14b, 1,4-dibromobenzene (578 mg,
2.45 mmol) was mono-lithiated with n-BuLi (2.5 M in hexanes,
1 mL, 2.50 mmol) and reacted with 24 (500 mg, 2.45 mmol) in
dry THF at ꢀ78 °C. The crude borinic acid was purified using silica
gel flash column chromatography (EtOAc/ hexane 1:5 v/v) to afford
21a as a yellowish oil (347 mg, 1.33 mmol, 55%). 1H NMR for 21a
(DMSO-d6, 300 MHz) d 9.71 (s, 1H), 7.69–7.63 (m, 2H), 7.60 (s,
4H), 7.47–7.34 (m, 3H).
5.2.1. Cell culture and establishment of RFP-C1-hTRPV6
expressing HEK293 cell clones
HEK293 cells (ATTC) were cultured at 37 °C in DMEM cell cul-
ture media supplemented with 10% FBS, 10 mM HEPES, 1 mM so-
dium pyruvate, and 1% penicillin/streptomycin. Cells were
subcultivated at 90% confluency.
Borinic acid 21a (374 mg, 1.33 mmol) was esterified with 2-
aminoethanol (161 lL, 2.66 mmol) in EtOH and the ester 21b
To establish pTagRFP-C1-hTRPV6 expressing HEK 293 cell
clones, cells plated the day before on poly-D-lysine coated 35 mm
was obtained as a white powder (236 mg, 0.78 mmol, 59%) after
precipitation from a biphasic Et2O/H2O mixture. 1H NMR (DMSO-
d6, 300 MHz) d 7.41–7.25 (m, 6H), 7.13 (m, 2H), 7.07–7.00 (m,
1H), 6.11 (br s, 2H), 3.75 (t, J = 6.5 Hz, 2H), 2.92–2.76 (m, 2H);
13C NMR (CD3OD, 75 MHz) d 150.1, 135.0, 132.8, 131.0, 128.2,
126.9, 120.9, 64.0, 42.6; MS (ESI+) m/z 304.05, 306.05 [M+H]+,
326.03, 328.03 [M+Na]+, 365.10, 367.10 [M+2-aminoethanol+H]+;
HRMS (ESI+) m/z calcd for C14H16BBrNO 304.0503 [M+H]+, found
304.0508 [M+H]+.
dish were transfected with 2 g pTagRFP-C1-TRPV6 using 5 lL
l
Lipofectamine 2000 per well as described in the manufacturer’s
protocol. Transfection medium was changed to antibiotic-free
medium after 4 h. On the following day the medium was then
changed with selection antibiotic (G418) containing medium. From
then on, the cells were kept in this selection medium. After a mas-
sive cell death of the non-transfected cells, surviving cells were
trypsinized and replated in a 96-well plate at such a dilution that
1 cell/well density was obtained. After several days, colonies of
cells displaying red fluorescence were selected as hTRPV6-express-
ing positive clones using fluorescence microscopy.
5.1.35. 2-(4-Iodophenyl)-2-phenyl-1,3,2-oxazaborolidin-3-ium-
2-uide (22b)
Following the procedure for 14b, 1,4-diiodobenzene (808 mg,
2.45 mmol) was mono-lithiated with n-BuLi (2.5 M in hexanes,
980 lL, 2.45 mmol) and then reacted with 24 (500 mg, 2.45 mmol)
5.2.2. Fluorescence ion measurement experiments using
FLIPRTetra
Cells were trypsinized and plated at 40,000 cells/well density in
in THF/Et2O at ꢀ78 °C. The crude borinic acid was purified by silica
gel flash column chromatography (EtOAc/hexane 1:5 v/v) to obtain
22a as a yellow oil (549 mg, 1.78 mmol, 73%). 1H NMR for 22a
(DMSO-d6, 300 MHz) d 9.72 (s, 1H), 7.78 (m, 2H), 7.72–7.55 (m,
2H), 7.55–7.22 (m, 5H).
100
lL volume onto 96-well black plates coated with 100
lg/mL
poly-D-lysine 36 h before the experiments. HEK293 cells were used
for testing the effects of compounds on store-operated calcium
channel activity; whereas hTRPV6 activity was measured using
hTRPV6-expressing HEK293 cell clones. 36 h later the medium
Borinic acid 22a (549 mg, 1.78 mmol) was esterified with 2-
was replaced with 100
(modified Krebs buffer containing 117 mM NaCl, 4.8 mM KCl,
1 mM MgCl2, 5 mM -glucose, 10 mM HEPES, and calcium-5 fluo-
lL of nominally calcium-free loading buffer
aminoethanol (215 lL, 3.56 mmol) in MeCN which gave the title
compound as a white powder (465 mg, 1.33 mmol, 75%) after pre-
cipitation from a biphasic Et2O/H2O mixture. 1H NMR (DMSO-d6,
300 MHz) d 7.48 (m, 2H), 7.36 (m, 2H), 7.20 (m, 2H), 7.13 (m,
2H), 7.03 (m, 1H), 6.10 (br s, 2H), 3.74 (t, J = 6.5 Hz, 2H), 2.92–
2.67 (m, 2H); 13C NMR (CD3OD, 75 MHz) d 137.1, 135.3, 132.8,
128.1, 126.9, 92.3, 64.0, 42.6, 25.2; MS (ESI+) m/z 352.04 [M+H]+,
374.02 [M+Na]+, 413.09 [M+2-aminoethanol+H]+; HRMS (ESI+)
D
rescence dye). Cells were then incubated in the loading buffer at
37 °C for 1 h. Fluorescence Ca2+ and Cd2+ measurements were car-
ried out using FLIPTetra high-throughput, fluorescence microplate
reader. Cells were excited using a 470–495 nm LED module, and
the emitted fluorescence signal was filtered with a 515–575 nm
emission filter. After establishment of a stable baseline in nomi-
nally calcium-free Krebs buffer, cells were incubated with the com-
pounds under calcium-free conditions for 5 min followed by the
m/z calcd for
C
14H16BINO 352.0364 [M+H]+, found 352.0362
[M+H]+.
administration of either 2 mM CaCl2 or 50 lM of CdCl2. The activity
of hTPRV6 was measured by the initial rate of the increase in fluo-
5.1.36. Hydroxy(phenyl)(4-(trifluoromethyl)phenyl)borane
(23a)
rescence intensity.
Following the procedure for 14b, 1-Bromo-4-(trifluoro-
For store-operated calcium channel activity, measurements
were performed in HEK293 cells. The procedure was the same as
mentioned above for hTRPV6 except that compounds were admin-
methyl)benzene (140 lL, 1.000 mmol) was lithiated with n-BuLi
(2.5 M in hexanes, 0.5 mL, 1.250 mmol) and subsequently reacted
with freshly prepared 12 (190 mg, 1 mmol) in Et2O at ꢀ78 °C.
The crude borinic acid 23a (orange sticky solid) was esterified with
istered together with 1
passively.
lM thapsigargin, in order to activate SOCs
2-aminoethanol (66 lL, 1.100 mmol) in EtOH. Attempts to crystal-
Experiments were done with 6 repeats per group. Relative IC50
values were calculated with Screenworks 3.1.1.4 (Molecular De-
vices LLC, USA).
lize the ester 23b failed. Instead, purification by preparative RP-
HPLC hydrolyzed the ester, and the title compound was obtained
as a beige solid (41 mg, 0.164 mmol, 16%). 1H NMR (DMSO-d6,

A. Hofer et al. / Bioorg. Med. Chem. 21 (2013) 3202–3213
12. Prakriya, M.; Lewis, R. S. J. Physiol. 2001, 536, 3.
3213
Acknowledgments
13. Lis, A.; Peinelt, C.; Beck, A.; Parvez, S.; Monteilh-Zoller, M.; Fleig, A.; Penner, R.
Curr. Biol. 2007, 17, 794.
We thank PD Dr. Piero Macchi for solving the crystal structures.
The authors would also like to thank Maria Feher, Michelle Frey,
Geraldine Lechot, and Hedna Scowell for their excellent cell culture
assistance. This work was supported by the Swiss National Science
Foundation and the University of Bern. This is a publication from
the NCCR TransCure at the University of Bern.
14. DeHaven, W. I.; Smyth, J. T.; Boyles, R. R.; Bird, G. S.; Putney, J. W., Jr. J. Biol.
Chem. 2008, 283, 19265.
15. Hu, H.; Grandl, J.; Bandell, M.; Petrus, M.; Patapoutian, A. Proc. Natl. Acad. Sci.
U.S.A. 2009, 106, 1626.
16. Kovacs, G.; Montalbetti, N.; Simonin, A.; Danko, T.; Balazs, B.; Zsembery, A.;
Hediger, M. A. Cell Calcium 2012, 52, 468.
17. Clapham, D. E. Cell 2007, 131, 1047.
18. Rychkov, G.; Barritt, G. J. Handb. Exp. Pharmacol. 2007, 23.
19. Chokshi, R.; Fruasaha, P.; Kozak, J. A. Channels 2012, 6, 362.
20. Bilmen, J. G.; Wootton, L. L.; Godfrey, R. E.; Smart, O. S.; Michelangeli, F. Eur. J.
Biochem. 2002, 269, 3678.
Supplementary data
21. Dobrydneva, Y.; Blackmore, P. F. Mol. Pharmacol. 2001, 60, 541.
22. Dobrydneva, Y.; Abelt, C. J.; Dovel, B.; Thadigiri, C. M.; Williams, R. L.;
Blackmore, P. F. Mol. Pharmacol. 2006, 69, 247.
23. Zhou, H.; Iwasaki, H.; Nakamura, T.; Nakamura, K.; Maruyama, T.; Hamano, S.;
Ozaki, S.; Mizutani, A.; Mikoshiba, K. Biochem. Biophys. Res. Commun. 2007, 352,
277.
24. Goto, J.-I.; Suzuki, A. Z.; Ozaki, S.; Matsumoto, N.; Nakamura, T.; Ebisui, E.;
Fleig, A.; Penner, R.; Mikoshiba, K. Cell Calcium 2010, 47, 1.
25. Suzuki, A. Z.; Ozaki, S.; Goto, J.-I.; Mikoshiba, K. Bioorg. Med. Chem. Lett. 2010,
20, 1395.
Supplementary data (HPLC purity assessment for all final com-
pounds, crystal structures of 2a, 2b and 2k, NMR stability studies
of 2-APB and analogs, and NMR spectra for all final compounds)
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.bmc.2013.03.037. These data include
MOL files and InChiKeys of the most important compounds de-
scribed in this article.
26. Dellis, O.; Mercier, P.; Chomienne, C. BMC Pharmacol. 2011, 11, 1.
27. It is worth noting that reaction of diphenylborinic anhydride with amino
alcohols choline and 4-(2-hydroxyethyl)morpholine, and alcohols 1-propanol
and ethan-1,2-diol, did not yield any product, even under forcing conditions.
28. Douglass, J. J. Org. Chem. 1961, 26, 1312.
29. Vischer, H. F.; Hulshof, J. W.; Hulscher, S.; Fratantoni, S. A.; Verheij, M. H. P.;
Victorina, J.; Smit, M. J.; de Esch, I. J. P.; Leurs, R. Bioorg. Med. Chem. 2010, 18,
675.
References and notes
1. Peng, J. B.; Brown, E. M.; Hediger, M. A. J. Physiol. 2003, 551, 729.
2. Peng, J. B.; Chen, X. Z.; Berger, U. V.; Vassilev, P. M.; Tsukaguchi, H.; Brown, E.
M.; Hediger, M. A. J. Biol. Chem. 1999, 274, 22739.
3. den Dekker, E.; Hoenderop, J. G.; Nilius, B.; Bindels, R. J. Cell Calcium 2003, 33,
497.
4. Peng, J. B.; Zhuang, L.; Berger, U. V.; Adam, R. M.; Williams, B. J.; Brown, E. M.;
Hediger, M. A.; Freeman, M. R. Biochem. Biophys. Res. Commun. 2001, 282, 729.
5. Wissenbach, U.; Niemeyer, B. A.; Fixemer, T.; Schneidewind, A.; Trost, C.;
Cavalie, A.; Reus, K.; Meese, E.; Bonkhoff, H.; Flockerzi, V. J. Biol. Chem. 2001,
276, 19461.
6. Zhuang, L.; Peng, J.-B.; Tou, L.; Takanaga, H.; Adam, R. M.; Hediger, M. A.;
Freeman, M. R. Lab. Invest. 2002, 82, 1755.
7. Lehen’Kyi, V.; Flouriakis, M.; Skryma, R.; Prevarskaya, N. Oncogene 2007, 26,
7380.
8. Bolanz, K. A.; Hediger, M. A.; Landowski, C. P. Mol. Cancer Ther. 2008, 7, 271.
9. Landowski, C. P.; Bolanz, K. A.; Suzuki, Y.; Hediger, M. A. Pharm. Res. 2011, 28,
322.
10. Maruyama, T.; Kanaji, T.; Nakade, S.; Kanno, T.; Mikoshiba, K. J. Biochem. 1997,
122, 498.
30. Caldirola, P. M.; van der Goot, H.; Timmerman, H. Eur. J. Med. Chem. 1992, 27,
571.
31. Brown, H. C.; Srebnik, M.; Cole, T. E. Organometallics 1986, 5, 2300.
32. Mori, Y.; Kobayashi, J.; Manabe, K.; Kobayashi, S. Tetrahedron 2002, 58, 8263.
33. Letsinger, R. L.; Skoog, I. J. Am. Chem. Soc. 1955, 77, 2491.
34. Wyatt, P.; Hudson, A.; Charmant, J.; Orpen, A. G.; Phetmung, H. Org. Biomol.
Chem. 2006, 4, 2218.
35. Rettig, S. J.; Trotter, J. Can. J. Chem. 1976, 54, 3130.
36. Chudzinksi, M. G.; Chi, Y.; Taylor, M. S. Aust. J. Chem. 2011, 64, 1466.
37. Flückiger, R.; Henson, E.; Hess, G. M.; Gallop, P. M. Biomed. Mass Spectrom.
1984, 11, 611.
38. Kovacs, G.; Danko, T.; Bergeron, M. J.; Balazs, B.; Suzuki, Y.; Zsembery, A.;
Hediger, M. A. Cell Calcium 2011, 49, 43.
39. Shiao, Y. J.; Vance, J. E. Biochem. J. 1995, 310, 673.
11. Bootman, M. D.; Collins, T. J.; Mackenzie, L.; Roderick, H. L.; Berridge, M. J.;
Pepiatt, C. M. FASEB J. 2002, 16, 1145.