A modular synthesis of teraryl-based α-helix mimetics, part 2: Synthesis of 5-pyridine boronic acid pinacol ester building blocks with amino acid side chains in 3-position

Article abstract of DOI:10.1002/chem.201203006

One of the most common protein-protein interactions (PPI) is the interaction of the α-helix of one protein with the surface of the second one. Terphenylic scaffolds are bioinspired motifs in the inhibition of PPIs and have been identified as suitable α-helix mimetics. One of the challenging aspects of this strategy is the poor solubility of terphenyls under physiological conditions. In the literature pyrrolopyrimidine-, pyrimidine- or pyridazine-based mimetics have been reported to show improved solubility. We present a new convergent strategy for the synthesis of linear pyridine-type teraryls based on a phenylic core unit. A general approach for the synthesis of 3,5-disubstituted pyridine-based boronic acid pinacol esters with amino acid side chains in the 3-position (representing Phe, Leu, Ile, Lys, Asp, Asn) is presented and exploits the functional group tolerance of the Knochel-Grignard reagents. The building blocks have been used in a convergent in situ two-step synthesis of teraryl α-helix mimetics. Tune in: The chemical orthogonality of Knochel's Grignard chemistry enables the synthesis of 3-substituted 5-pyridine-boronic esters with amino acid side chains, which can be used for convenient assembly of teraryl-based α-helix peptide mimetics by Suzuki coupling (see scheme; dppf=1,1′-bis(diphenylphosphino)ferrocene, DME= dimethoxyethane, Tf=trifluoromethanesulfonyl). Copyright

Full text of DOI:10.1002/chem.201203006

DOI: 10.1002/chem.201203006
A Modular Synthesis of Teraryl-Based a-Helix Mimetics, Part 2: Synthesis of
5-Pyridine Boronic Acid Pinacol Ester Building Blocks with Amino Acid
Side Chains in 3-Position
Martin Peters, Melanie Trobe, and Rolf Breinbauer*[a]
Dedicated to Professor Bernhard Krꢀutler
Abstract: One of the most common
protein–protein interactions (PPI) is
the interaction of the a-helix of one
protein with the surface of the second
one. Terphenylic scaffolds are bioin-
imidine-, pyrimidine- or pyridazine-
based mimetics have been reported to
show improved solubility. We present a
new convergent strategy for the synthe-
sis of linear pyridine-type teraryls
based on a phenylic core unit. A gener-
al approach for the synthesis of 3,5-di-
AHCTUNGTRENNGsUN ubstituted pyridine-based boronic acid
pinacol esters with amino acid side
chains in the 3-position (representing
Phe, Leu, Ile, Lys, Asp, Asn) is present-
ed and exploits the functional group
tolerance of the Knochel–Grignard re-
agents. The building blocks have been
used in a convergent in situ two-step
synthesis of teraryl a-helix mimetics.
ACHTUNGTRENNUNGspired motifs in the inhibition of PPIs
and have been identified as suitable a-
helix mimetics. One of the challenging
aspects of this strategy is the poor solu-
bility of terphenyls under physiological
conditions. In the literature pyrrolopyr-
Keywords: cross-coupling
· Kno-
chel–Grignard · peptidomimetics ·
protein–protein interactions · pyri-
dine-based 1,4-teraryls
Introduction
pyrrolopyrimidine-,[10] piperidino-,[11] or triazolo-based[12]
scaffolds. In addition, pyridine-based teraryls are known,
but their synthesis has turned out to be inflexible[13] and the
linear approach makes this work very time consuming.[14]
Here, we describe efficient synthetic access to amino acid
surrogate pyridine boronic acid building blocks 2, which to-
gether with our accompanying paper on convergent assem-
bly strategy, presents a universal and flexible approach for
the synthesis of pyridine-based heteroteraryl-based a-helix
mimetics 1 (Figure 1b).[15]
Protein–protein interactions (PPIs) are recognized as one of
the main factors in controlling protein function in living
cells. The number of different PPIs in human cells is esti-
mated to be approximately 65000,[1,2] and a-helices play a
prominent role in the interaction sites.[3] Typically, PPI do-
mains comprise about 35–150 amino acids.[4] For the study
and pharmaceutical intervention of PPIs tool compounds
are needed that allow the control of a particular interaction
of a specific target protein.[5] Hamilton and co-workers have
presented a quite general approach of mimicking a-helices
by suitable positioning of amino acid side chains around a
terarylic scaffold (Figure 1a).[6] They could impressively
demonstrate that for several examples selective PPI inhibi-
tors (terphenyl-based a-helical mimetics) with an affinity in
the nanomolar range can be developed using this design ap-
proach.[7]
As many of these terphenyls show poor solubility in aque-
ous solvents, more polar heteroaryl-based helical mimetics
have been developed, such as pyrimidine-,[8] pyridazine-,[9]
[a] Dr. M. Peters, M. Trobe, Prof. Dr. R. Breinbauer
Institute of Organic Chemistry, Graz University of Technology
Stremayrgasse 9, 8010 Graz (Austria)
Fax : (+43)316-873-32402
E-mail: breinbauer@tugraz.at
Figure 1. a) Schematic depiction of Hamilton’s terphenylic scaffold for
mimicking an a-helix. b) Retrosynthesis of our teraryls 1 as a convergent
synthesis starting from a para-iodo-triflate core unit 3.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201203006.
2450
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2450 – 2456

FULL PAPER
Results and Discussion
We have already presented a convergent strategy for the
synthesis of linear terphenyls, using a benzene core unit fea-
turing two leaving groups of differentiated reactivity in the
Pd-catalyzed cross-coupling reaction.[15] Although this
design is in principle suitable for several a-helix mimetics
featuring highly polar side chains, the intrinsically poor solu-
bility of the benzene-based terphenyl core structure imposes
limitations for the use of this core structure as a scaffold for
hydrophobic side chains under physiological conditions. To
circumvent this problem, the design of more soluble teraryls,
containing pyridine moieties, is established.
For our design, we envisioned that pyridines as the top
and the bottom rings would increase the polarity and solu-
bility of our teraryl scaffold. We would prefer to have the
polar nitrogen atom at the site opposite to the amino acid
side chain, as this represents the water exposed surface of
the teraryl, and potentially has an advantageous effect on
the entropic cost of binding.[16] As a consequence of this
design, we required a universal approach to 3,5-disubstituted
pyridine boronic esters 2 featuring the amino acid side
chain as substituent in the 3-position (Scheme 1). We specu-
Scheme 2. Overview of the attempts of synthesizing 3-(hydroxymethyl)-5-
BPin-pyridine derivatives 6a,b starting from 3,5-dibromopyridine (4a).
Scheme 3. “Turbo-Grignard” approach for the derivatization of electron
poor N-heterocyclic compounds, such as pyridines.[20a] a) CuI (10 mol%),
NaI (4.0 equiv), N,Nꢁ-dimethylethylenediamine 1,4-dioxane (10 mol%),
1208C, 86% (gram scale).
Scheme 1. Original concept of synthesizing 5-pyridine boronic acid pina-
col esters with amino acid side chains in 3-position.
chemistry[21] should not only tolerate nitriles, OMe or CF3
groups, but also esters or amides.[20a,22]
À
À
When using dibromopyridine 4a, the first metal–halide
exchange worked quantitatively and dependent on the elec-
trophile the corresponding benzyl alcohol 5a,b or the BPin
derivative 7a could be isolated in good to excellent yields
(Scheme 2, routes b and c). However, the bromo derivatives
5a,b or 7a turned out to be problematic substrates for the
second metal–halide exchange under the same reaction con-
ditions due to increased byproduct formation and dramati-
cally prolonged reaction times.
We speculated that with the more reactive 3,5-diiodopyri-
dine (4b) we could address the problem of the second
metal–halide reaction (Scheme 3). After several failed ef-
forts to synthesize 4b by the sequential tBuLi mediated dili-
thiation and diiodination of 3,5-dibromopyridine (4a), we
turned our attention to Buchwaldꢁs variant for Finkelstein-
like iodination of haloarenes.[23] Buchwald and co-workers
had demonstrated the monoiodination of aryls and heteroar-
yls. Gratifyingly, this reaction worked also for the diiodina-
tion of our substrate. The halide exchange was catalyzed by
CuI in the presence of N,Nꢁ-dimethylethylenediamine and
furnished the desired product 4b in 86% on a multigram
scale (Scheme 3).
lated that the pinacol ester function could be first intro-
duced by electrophilic quenching of a suitable organolithium
derivative of pyridines 4a,b utilizing electrophiles, such as
PinBOiPr (11), which could then be followed by the intro-
duction of the amino acid side chain to finally afford the de-
sired boronic acid ester building blocks 2.
In our initial attempt we followed a protocol by Zhichkin
and co-workers and first used a modified tBuLi-mediated
method for the borylation of 3,5-disubstitued bromopyridine
4a, followed by an in situ second electrophilic quenching
with an aldehyde (Scheme 2, route a). However, we were
not able to isolate the desired 3-(hydroxymethyl)-5-BPin-
pyridine derivatives 6 with this in situ method.[17]
After several unsuccessful attempts to explore the dilithia-
tion of suitable pyridine derivatives, we were pleased to see
that in our hands the “turbo-Grignard” approach introduced
by Knochel[18] and other research groups[19] appears to be
the key to the successful substitution of dihalopyridine de-
rivatives 4a,b.[20] This method has the advantage that the
halides can successively be substituted with the combination
of magnetization/electrophilic quench on electron poor het-
erocycles (Scheme 3). Evidence from the literature suggest-
ed that the attractive orthogonality of this “turbo-Grignard”
Much to our delight, the twofold metal–halide exchange
reaction occurred with the more reactive 3,5-diiodopyridine
Chem. Eur. J. 2013, 19, 2450 – 2456
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2451

R. Breinbauer et al.
Table 1. First iodine–magnesium exchange followed by electrophilic
quenching with the carbonyl derivatives 8a–f to synthesize 3-(hydroxy-
methyl)-5-iodopyridine derivatives 9a–f.
Entry
Carbonyl
electrophile
Grignard
product
Yield
[%][a]
1
2
3
4
5
6
99
95
99[b]
68
83
84
Scheme 4. Strategy for the synthesis of 3,5-substituted pyridine-based
boronic acid pinacol esters 2.
[a] Isolated yields. [b] Crude product; used in the next step without fur-
ther purification.
(4b) now as desired. First, we tried a modified reaction se-
quence of a twofold Grignard reaction as described by Kno-
chel and co-workers, in which we introduced the diverse ele-
ments of the side chains after having first formed the basic
pyridine boronic acid ester core (Scheme 4).[20a,24] The diio-
dopyridine 4b was borylated, to furnish the 5-BPin deriva-
tive 7b. After a second metal–halide exchange, the corre-
sponding 3-(hydroxymethyl)-5-BPin-pyridine derivative 6b
could be isolated (14%, starting from diiodopyridine 4b),
but because of the unexpected instability of compound 6b
we failed to deoxygenate the benzylic alcohol 6b to the cor-
responding desired final product 2b. Although this se-
quence, if successful, would have represented the most flexi-
ble access to our desired 3-substituted 5-pyridine boronic
acid building blocks 2, we speculated that changing the
order of the sequence in the electrophile addition might
lead us to the desired products.
Scheme 4 summarizes both the attempted as well as the
ultimately successful synthetic route for the assembly of 3,5-
disubstituted pyridine-based boronic acid pinacol esters 2.
The most critical step of the synthesis was the introduction
of the amino acid side chain during the first Grignard for-
mation.
As a test case for our strategy, we attempted to synthesize
the “Phe” building block 2a. For the introduction of the
side chain, we experimented with benzyl bromide as electro-
phile. Despite considerable optimization efforts, we did not
obtain decent coupling yields using this type of electrophile,
but struggled with alkylation of the pyridine nitrogen in-
stead. Fortunately, we found that highly reactive aldehyde
electrophiles (e.g., benzaldehyde, 8a; Table 1, entry 1), but
also comparably less reactive ketones (e.g., butan-2-one, 8c;
Table 1, entry 3) are excellent reactants for the electrophilic
quenching of the pyridine Grignard reagent, delivering
products 9a–f in good to excellent yields. Indeed the Kno-
chel conditions also tolerated functional groups such as ni-
trile and ester in the electrophile (Table 1, entries 5 and 6).
For most natural amino side chains (except Thr and Ser),
it is necessary to remove the superfluent hydroxyl group to
achieve the native alkyl form. A possible strategy is the
acid-catalyzed elimination of H2O leading to the unsaturat-
ed form, which easily occurs for tertiary alcohols like 9c
(Table 1, entry 3).[25] An alternative, and more general route,
is the reduction of the benzylic 3-(hydroxymethyl)-5-iodo-
pyridine derivatives 9 into the corresponding deoxygenated
form. However, none of the common reduction methods we
tried (H2, PtO2, MeOH; H2, Pd/C, MeOH; H2, Pd(OH)2/C,
MeOH; Zn, AcOH) led to the desired reduction product, as
any successful deoxygenation was afflicted with a concomi-
tant deiodination process as the main side reaction. In order
to overcome this problem we decided to convert the hydrox-
yl group first into chlorine, which should be amenable to de-
halogenation by a less aggressive reducing agent. In the
event, the pyridine carbinols 9 were smoothly transformed
into the corresponding 3-(chloromethyl)-5-iodopyridine de-
rivatives 10 with thionyl chloride (SOCl2) under neat condi-
2452
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2450 – 2456

Modular Synthesis of Teraryl-Based a-Helix Mimetics, Part 2
FULL PAPER
Table 2. Overall yields for the synthesis of building blocks 2 with representative resi-
dues (or a latent form; entries 5 and 6) for all groups of amino acids (nonpolar/hydro-
phobic, basic, acidic and polar/neutral or even non-natural).
tions (Scheme 4). Gratifyingly, the second iodine–
magnesium exchange on substrates 10 proceeds
without any side reaction from the potential inter-
nal electrophile represented by the a-aryl chloride.
Therefore, the electrophilic quench with PinBOiPr
(11) produced the borylated 3-(chloromethyl)-5-
BPin-pyridine derivatives 12. As these compounds
proved to be sensitive to heat and were also not
stable on silica gel, we used the intermediates 12
without any further purification in the next step, in
which they were easily dehalogenated with Zn/
AcOH. Importantly, this rather mild reduction
method did not cause any deborylation of the de-
sired products 2 (Scheme 4).
Entry
Product
Amino acid side chain
Overall
yield [%][a]
1
2
“Phe”, nonpolar/hydrophobic
“Leu”, nonpolar/hydrophobic
76
45
3
“Ile”, nonpolar/hydrophobic
24[b]
In order to establish the scope and limitations of
this synthetic approach leading to 3,5-disubstituted
pyridine boronic esters 2, we selected target struc-
tures, which should be representative not only for
the reactivity of electrophiles (first iodine–magnesi-
um exchange), but also for the various classes of
amino acid side chains. We could confirm that non-
polar/hydrophobic (2a–c, Table 2, entries 1–3),
basic (2e, Table 2, entry 5), acidic (2 f, Table 2,
entry 6), polar/neutral (2g, Table 2, entry 7) and
even non-natural (2d, Table 2, entry 4) side chains
are accessible by this approach with 24–73% over-
all yields with the four-step reaction sequence. For
synthetic reasons we decided to synthesize the
4
5
6
7
non-natural/nonpolar/hydrophobic
“Lys”, basic
27
52
“Asp”, acidic
55
“Asn”, polar/neutral
37[c]
[a] Isolated yields over four steps starting from 3,5-diiodopyridine (4b). [b] The elimi-
nation product of compound 9c (Table 1, entry 3) was reduced with 10% Pd/C, after
borylation.[25] [c] Compound 2g was prepared from compound 2 f by aminolysis with
catalytic amounts of KCN in a 7m ammonia/MeOH mixture at 508C.[25]
building blocks for the “Lys” and “Asp” surrogates in a
masked form, as these are more stable in storage and more
advantageous in the subsequent Suzuki coupling assembly
and purification step. The masked “Lys” surrogates (2e,
Table 2, entry 5) can easily be hydrogenated to the corre-
sponding native structure at the teraryl stage, by using
Raney nickel in an ammonia/MeOH mixture, as we demon-
strated later (vide infra).
show that the combination of pyridine boronic acid ester
building blocks 2 and the convergent assembly by electroni-
cally differentiated functional core units 3 provides an effi-
Table 3. Synthesized pyridine-based teraryls 1a–c using the convergent
two-step synthesis without purification approach.
In order to highlight our convergent teraryl synthesis
strategy, we prepared representative pyridine-based teraryls
1 with a phenylic core unit 3. According to the established
procedure, it is possible to obtain teraryls 1 in a convergent,
in situ two-step synthesis.[15] The selective differentiation of
the two leaving groups is based on the different reactivity of
the leaving groups, the steric accessibility, and strength of
the applied bases during the Suzuki coupling with PdCl2-
ACHTUNGTRENNUNG
Entry
1
R1[a]
R2[a]
R3[a]
Teraryl
Yield
[%][b]
bled in 47% (Table 3, entry 1) and the Naph-Ile-Phe (1b)
mimetic in 66% overall yields (entry 2). For the synthesis of
the Leu-Val-Lys mimetic the second Suzuki coupling was
performed with boronic ester 2e to deliver compound 1c in
46% yield (Table 3, entry 3). The lysine side chain in its
latent form of a nitrile in compound 1c can be easily re-
duced to the primary amine under reductive conditions by
using Raney nickel in an H-CubeTM flow reactor to deliver
the desired Leu-Val-Lys mimetic 1d in 86% yield
(Scheme 5). Already these three examples convincingly
1a
47
2
3
1b
1c
66
46[c,d]
[a] Residues for the corresponding amino acids that are mimicked by
these residues. [b] Isolated yields. [c] Lys side chain (R3) in its latent form
of a nitrile. [d] The primary nitrile was reduced to the corresponding
amine with Raney nickel in ammonia/MeOH solution.[25]
Chem. Eur. J. 2013, 19, 2450 – 2456
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2453

R. Breinbauer et al.
and vacuum dried (1208C/6 h) lithium chloride (9.93 g, 0.23 mmol,
1.0 equiv, based on the formed Grignard). After being stirred for 12 h at
room temperature, the iPrMgCl·LiCl solution (13) was titrated again by
using the following procedure.[20]
Titration of isopropyl magnesium chloride lithium chloride solution
(iPrMgCl·LiCl) (13): A flame-dried amber glass Schlenk flask was charg-
ed with absolute, degassed toluene (200.0 mL) and absolute 2-butanol
(20.0 mL).[26] This stock solution was stored under argon over 3 ꢂ molec-
ular sieves (stable over months). The calculated concentration of this
stock solution (ccalcd =0.99m) was used as reference for the titration of
iPrMgCl·LiCl solution (13). N-Phenyl-4-phenylazoaniline (~2–5 mg) as
indicator and the Grignard solution were added dropwise, under inert
conditions, to the stock solution (2.0 mL). The equivalence point was in-
dicated by a sharp color change from yellow–orange to deep red. To
ensure a precise titration a triple determination was performed. The titra-
tion was carried out before every use of the Grignard solution.[27]
Scheme 5. Reduction of primary nitrile 1c to the corresponding amine
1d.
cient and general solution to the synthesis of polar teraryl-
based a-helix mimetics as potential inhibitors of PPIs.
3,5-Diiodopyridine (4b): In a flame-dried Schlenk flask (250 mL) 3,5-di-
bromopyridine (4a; 7.27 g, 30.7 mmol, 1.0 equiv), copper(I) iodide[28]
(585 mg, 3.1 mmol, 10 mol%) and sodium iodide (18.41 g, 0.13 mol,
4.0 equiv) were suspended in absolute, degassed 1,4-dioxane (50 mL).
After adding N,Nꢁ-dimethylethylenediamine (330 mL, 270 mg, 3.07 mmol,
10 mol%), the pale yellow suspension was stirred for approximately 20 h
at 1208C until complete conversion (Rf =0.68, cyclohexane/EtOAc, 9:1).
After filtration, the reddish brown suspension was quenched with saturat-
ed NH4Cl solution (50 mL) and the deep blue aqueous phase was extract-
ed with DCM (4ꢃ70 mL). The combined yellow organic layers were
dried over Na2SO4, filtered and concentrated to dryness. The golden
yellow crude product (9.23 g, 91%) was recrystallized from EtOH
(235 mL; 8.72 g, 86%, pale golden shavings). As an alternative purifica-
tion method, small amounts could be sublimated at 110–1208C and
0.01 torr.[29] M.p. 166–1688C; sublim.=100–1108C, 0.01 torr; 1H NMR
Conclusion
The interaction of small molecules with biologically active
targets is the central focus of drug discovery. The field of in-
hibition of protein–protein interactions has become one of
the big challenges and this area of research is still in its nas-
cency. Our new modular approach for the assembly of more
water soluble teraryls, based on integrating pyridinic build-
ing blocks with our strategy of electronically differentiated
bifunctional core units 3, represents an important synthetic
advance for the convenient synthesis of libraries of such a-
helix mimetics. With only a set of 18 core building blocks 3
and 18 3,5-disubstituted pyridine boronic acid pinacol esters
2, any permutations of a-helix mimetics featuring all rele-
vant proteinogenic amino acids (excluding Pro and Gly)
could be prepared. We are currently working on the synthe-
sis of such a comprehensive set of building blocks, featuring
all 18 proteinogenic amino acids complemented by some
non-natural ones, like compound 2d.
(300 MHz, [D]CHCl3): d=8.75 (d, 4J
8.35 ppm (t, 4J(H,H)=1.8 Hz, 1H; H-4); 13C NMR (76 MHz, [D]CHCl3,
ACHTUNGTNER(NUNG H,H)=1.3 Hz, 2H; H-2, H-6),
AHCTUNGTRENNUNG
APT): d=154.3 (C-2, C-6), 151.7 (C-4), 94.0 ppm (Cq; C-3, C-5); GC-MS
(EI, 70 eV): m/z (%): 331 (100) [M+], 204 (46) [M+ÀI], 77 (17) [M+
À2I].
2-Isopropoxy-4,4,5,5-tetramethyl-1,3,2-dioxaborolane (PinBOiPr) (11): A
flame-dried and argon-flushed two-neck round-bottom flask (100 mL),
equipped with argon inlet, was charged with pinacol (6.92 g, 58.6 mmol,
1.0 equiv) and triisopropyl borate (13.50 mL, 11.00 g, 58.5 mmol,
1.0 equiv). After being stirred for 2 h at 688C, the formed 2-propanol was
removed under inert conditions, in vacuo, at room temperature
AHCTUNGTRENNUNG
a
AHCTUNGTRENNUNG
Experimental Section
AHCTUNGTRENNUNG
13C NMR (76 MHz, [D]CHCl3, APT): d=82.6 (Cq; CBPin), 67.5 (CH),
24.7, (CH3BPin), 24.5 ppm (CH3); GC-MS (EI, 70 eV): m/z (%): 186 (1)
[M+], 171 (100) [M+ÀCH3], 129 (33) [M+ÀC4H9].
General methods, additional information and further experimental proce-
dures are given in the Supporting Information.
Isopropyl magnesium chloride lithium chloride (iPrMgCl·LiCl) (13): A
flame-dried two-neck round-bottom flask (250 mL) with reflux condenser
(with argon inlet) charged with magnesium turnings (7.34 g, 0.30 mol,
1.2 equiv) was heated, in vacuo, with a heat gun for 5 min at maximum
power level. After being cooled to room temperature, absolute, degassed
THF (30 mL) was added and the suspension was sonicated for 10 min. 2-
Chloropropane (23.20 mL, 19.93 g, 0.25 mol, 1.0 equiv) was added via sy-
ringe under inert conditions. After activation by heating, the strong exo-
thermic reaction was kept between 55 and 608C by intensive ice-cooling.
If the reaction did not start by heating, a small crystal of iodine was
added, without stirring, to initiate the activation at a localized position.
After the Grignard formation had started, the suspension was diluted
with absolute, degassed THF (85 mL; overall 115 mL, ccalcd =2.2m). After
decay of the exothermic reaction, the mixture was heated for approxi-
mately 1 h at 808C and stirred, overnight, at room temperature. The
Grignard suspension was filtered under argon by using an inert filtration
funnel, and titration according to the procedure described below was per-
formed to determine the concentration of the isopropyl magnesium chlor-
ide. The pale gray/brown filtered Grignard solution was added to ground
Representative procedure for the first iodine–magnesium exchange for
synthesizing 3-(hydroxymethyl)-5-iodopyridine derivatives 9: A flame-
dried and argon-flushed two-neck round-bottom flask (100 mL), equip-
ped with argon inlet, was charged with 3,5-diiodopyridine (4b; 1.0 equiv)
and absolute, degassed THF was added until a clear solution was ob-
tained at room temperature (cꢀ0.2m). The colorless solution was cooled
to À788C and iPrMgCl·LiCl solution (13; 1.05 equiv) was added in one
portion. After being degassed, the pale yellow solution was stirred at
À788C until full conversion (~1.5–2 h). The iodine–magnesium exchange
was monitored by GC-MS. The GC samples were prepared by quenching
a small aliquot of the reaction mixture with saturated NH4Cl solution fol-
lowed by extraction with DCM. After quantitative conversion, the corre-
sponding electrophiles 8a–f were added and the reaction mixture was al-
lowed to warm to room temperature after 30 min at À788C, and stirred
until full conversion. The reaction mixture was quenched with saturated
NH4Cl solution, extracted with DCM (5ꢃ20 mL) and dried over Na2SO4.
After filtration, the solvent was removed, in vacuo, by using a rotary
evaporator and the yellow or orange oil was purified by flash column
2454
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2450 – 2456

Modular Synthesis of Teraryl-Based a-Helix Mimetics, Part 2
FULL PAPER
chromatography (short SiO2 column, 4ꢃ4 cm, eluents are described in
the Supporting Information).[20a]
2.0 equiv) and PdCl2ACTHNGUTER(NNUG dppf)·DCM (5 mol%). After being dried, in vacuo,
a solution of the previously prepared crude intermediate in absolute, de-
gassed 1,2-DME (5 mL) was added. After being degassed, the reaction
mixture was stirred at 808C, overnight. The quantitative conversion was
confirmed by TLC, and after filtration of the black suspension through a
pad of SiO2 (3ꢃ2 cm, eluent: 100 mL MeOH) the filtrate was concentrat-
ed to dryness under reduced pressure and was purified by flash column
chromatography.
Representative procedure for the formation of 3-(chloromethyl)-5-iodo-
pyridine derivatives 10 from the corresponding 3-(hydroxymethyl)-5-io-
dopyridine derivatives 9: A round-bottom flask (50 mL) with reflux con-
denser was charged with the corresponding 3-(hydroxymethyl)-5-iodopyr-
idine derivative 9 (1.0 equiv). After being cooled to À128C, freshly distil-
led SOCl2 was added, and the mixture was stirred at the indicated tem-
perature. In some cases, DCM was added to ensure complete dissolution
of the formed pyridinium salt. Chlorination was monitored by GC-MS.
The GC samples were prepared by quenching a small aliquot of the reac-
tion mixture with saturated Na2CO3 solution followed by extraction with
DCM. After quantitative conversion, the SOCl2 was distilled off and the
crude product was quenched with saturated Na2CO3 solution. The aque-
ous phase (pH~8) was extracted with DCM (4ꢃ20 mL) and the com-
bined organic layers were washed with brine (1ꢃ15 mL). The pale yellow
or reddish organic phase was dried over Na2SO4, and after filtration, the
solvent was removed under reduced pressure. The crude product was pu-
rified by flash column chromatography (short SiO2 column, 4ꢃ4 cm; vide
infra).
Acknowledgements
We thank Prof. Reza Ahmadian (University of Dꢄsseldorf) and Dr. Ra-
dovan Dvrosky (MPI for Molecular Physiology, Dortmund) for helpful
discussions and continuous collaboration in the field of protein–protein
interactions. This research was funded by grants of the Volkswagenstif-
tung, Hannover, and the PLACEBO (Platform for Chemical Biology)
project as part of the Austrian Genome Project GEN-AU funded by the
Forschungsfçrderungsgesellschaft (FFG) and Bundesministerium fꢄr Wis-
senschaft und Forschung (BMWF). We gratefully acknowledge Elisabeth
Fuchs, Stefan Velikogne and Lukas Omann for skillful assistance in the
laboratory.
Representative procedure for the second iodine–magnesium exchange
for synthesizing 3-(chloromethyl)-5-BPin-pyridine derivatives 12:
A
flame-dried and argon-flushed two-neck round-bottom flask (50 mL),
equipped with argon inlet, was charged with the corresponding 3-(chloro-
methyl)-5-iodopyridine derivative 10 (1.0 equiv) and absolute, degassed
THF (cꢀ0.2m) was added. The solution was cooled to À788C, and
iPrMgCl·LiCl (13; 1.1 equiv) was added in one portion. After being de-
gassed, the solution was stirred at À788C until full conversion was detect-
ed by GC-MS (~2 h). The GC samples were prepared by quenching a
small aliquot of the reaction mixture with saturated NH4Cl solution and
extracted with DCM. PinBOiPr (11; 1.15 equiv) was added and the reac-
tion mixture was allowed to warm to room temperature and stirred until
full conversion (~1 h). The reaction mixture was quenched with saturated
NH4Cl solution and extracted with DCM (4ꢃ25 mL). The combined or-
ganic layers were washed with brine (1ꢃ15 mL) and dried over Na2SO4.
After filtration, the solvent was removed, in vacuo, by using a rotary
evaporator and the crude product was used in the next step without fur-
ther purification due to instability on SiO2.[20a]
[1] M. P. H. Stumpf, T. Thorne, E. de Silva, R. Stewart, H. J. An, M.
Lappe, C. Wiuf, Proc. Natl. Acad. Sci. USA 2008, 105, 6959–6964.
[2] K. Wagner, J. Sꢄhnel: http://ppi.fli-leibniz.de/.
[3] Analyses of the PDB databank show that more than 60% of depos-
ited structures of protein–protein complexes feature an a-helix at
the interface: a) A. L. Jochim, P. S. Arora, ACS Chem. Biol. 2010, 5,
919–923; b) A. L. Jochim, P. S. Arora, Mol. BioSyst. 2009, 5, 924–
926.
[4] T. Pawson, P. Nash, Science 2003, 300, 445–452.
[5] For an authoritative review of small molecule PPI inhibitors, see: T.
Berg, Angew. Chem. 2003, 115, 2566–2586; Angew. Chem. Int. Ed.
2003, 42, 2462–2481.
[6] a) A. Kazi, J. Sun, K. Doi, S.-S. Sung, Y. Takahashi, H. Yin, J. M.
Rodriguez, J. Becerril, N. Berndt, A. D. Hamilton, H.-G. Wang,
S. M. Sebti, J. Biol. Chem. 2011, 286, 9382–9392; b) B. P. Orner, J. T.
Ernst, A. D. Hamilton, J. Am. Chem. Soc. 2001, 123, 5382–5383;
c) M. A. Kotharꢅ, J. Ohkanda, J. W. Lockman, Y. Qian, M. A. Blas-
kovich, S. M. Sebti, A. D. Hamilton, Tetrahedron 2000, 56, 9833–
9841; d) C. G. Cummings, A. D. Hamilton, Curr. Opin. Chem. Biol.
2010, 14, 341–346; e) J. M. Davis, L. K. Tsou, A. D. Hamilton,
Chem. Soc. Rev. 2007, 36, 326–334.
[7] a) I. Saraogi, A. D. Hamilton, Biochem. Soc. Trans. 2008, 36, 1414–
1417; b) L. Chen, H. Yin, B. Farooqi, S. Sebti, A. D. Hamilton, J.
Chen, Mol. Cancer Ther. 2005, 4, 1019–1025; c) H. Yin, G.-i. Lee,
H. S. Park, G. A. Payne, J. M. Rodriguez, S. M. Sebti, A. D. Hamil-
ton, Angew. Chem. 2005, 117, 2764–2767; Angew. Chem. Int. Ed.
2005, 44, 2704–2707.
[8] M. L. McLaughlin, L. Anderson, M. Zhou, University of South Flor-
ida, USA, H. Lee Moffitt Cancer Center and Research Institute,
Inc., WO2010083501A2, 2010.
[9] a) L. Moisan, T. J. Dale, N. Gombosuren, S. M. Biros, E. Mann, J.-L.
Hou, F. P. Crisostomo, J. Rebek, Jr., Heterocycles 2007, 73, 661–671;
b) A. Volonterio, L. Moisan, J. Rebek, Jr., Org. Lett. 2007, 9, 3733–
3736.
Representative procedure for the dechlorination of 3-(chloromethyl)-5-
(BPin)-pyridine derivatives 12 to the corresponding 3,5-disubstituted pyr-
idine-based boronic acid pinacol esters 2: In
a round-bottom flask
(250 mL) the corresponding 3-(chloromethyl)-5-(BPin)pyridine derivative
12 (1.0 equiv) was dissolved in approximately 1.0m DCM/glacial acetic
acid (21 equiv) solution. After addition of zinc dust (<10 mm, equivalents
are denoted in the Supporting Information), the green suspension was
stirred until full conversion was detected by GC-MS analysis. The GC
samples were prepared by quenching a small aliquot of the reaction mix-
ture with saturated Na2CO3 solution followed by extraction with DCM
and filtration through
a plug of cotton. The reaction mixture was
quenched with saturated Na2CO3 solution, extracted with DCM (4ꢃ
20 mL) and washed with brine (1ꢃ20 mL). The combined organic layers
were dried over Na2SO4 and filtered. The crude product was concentrat-
ed to dryness and the pale yellow or orange crude oily product was puri-
fied by Kugelrohr distillation (temperature and pressure are mentioned).
Representative procedure for the synthesis of linear pyridine-type terar-
yls 1a–c by consecutive double Suzuki coupling: A flame-dried two-neck
round-bottom flask with argon inlet was charged with the corresponding
3,5-disubstituted pyridine-based boronic acid pinacol ester 2 (1.0 equiv),
caesium fluoride (CsF; 2.0 equiv) and PdCl2ACTHNUTRGNEUNG(dppf)·DCM (5 mol%).
After being dried, in vacuo, a solution of the corresponding core unit 3
(1.0 equiv) in absolute, degassed 1,2-dimethoxyethane (DME, 4 mL) was
added. After being degassed, the reaction mixture was stirred at 808C
until full conversion was detected by TLC (~7 h). The brown suspension
was filtered through a pad of SiO2 (3ꢃ2 cm, eluent: 100 mL EtOAc) and
the filtrate was concentrated to dryness by using a rotary evaporator. A
second flame-dried two-neck round-bottom flask with argon inlet was
charged with the second corresponding 3,5-disubstituted pyridine-based
boronic acid pinacol ester 2 (1.05 equiv), caesium carbonate (Cs2CO3;
[10] J. H. Lee, Q. Zhang, S. Jo, S. C. Chai, M. Oh, W. Im, H. Lu, H.-S.
Lim, J. Am. Chem. Soc. 2011, 133, 676–679.
[11] P. Maity, B. Kçnig, Org. Lett. 2008, 10, 1473–1476.
[12] I. Ehlers, P. Maity, J. Aube, B. Kçnig, Eur. J. Org. Chem. 2011,
2474–2490.
[13] J. M. Davis, A. Truong, A. D. Hamilton, Org. Lett. 2005, 7, 5405–
5408.
[14] G. T. Bourne, D. J. Kuster, G. R. Marshall, Chem. Eur. J. 2010, 16,
8439–8445.
Chem. Eur. J. 2013, 19, 2450 – 2456
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
2455

R. Breinbauer et al.
[15] See the accompanying paper: M. Peters, M. Trobe, H. Tan, R. Klei-
neweischede, R. Breinbauer, Chem. Eur. J. 2012; DOI: 10.1002/
chem.201203005.
[16] G. Klebe, Wirkstoffdesign: Entwurf und Wirkung von Arzneistoffen,
2 ed., Spektrum, Marburg, 2009.
[21] a) P. Knochel, M. A. Schade, S. Bernhardt, G. Manolikakes, A.
Metzger, F. M. Piller, C. J. Rohbogner, M. Mosrin, Beilstein J. Org.
Chem. 2011, 7, 1261–1277, No. 1147; b) B. Haag, M. Mosrin, H. Ila,
V. Malakhov, P. Knochel, Angew. Chem. 2011, 123, 9968–9999;
Angew. Chem. Int. Ed. 2011, 50, 9794–9824; c) P. Knochel, C.
Diene, C. R. Chim. 2011, 14, 842–850; d) L. Ackermann, A. Al-
thammer, Chem. Unserer Zeit 2009, 43, 74–83; e) H. Ila, O. Baron,
A. J. Wagner, P. Knochel, Chem. Commun. 2006, 583–593.
[22] a) A. Krasovskiy, A. Tishkov, V. del Amo, H. Mayr, P. Knochel,
Angew. Chem. 2006, 118, 5132–5136; Angew. Chem. Int. Ed. 2006,
45, 5010–5014; b) M. S. Maji, S. Murarka, A. Studer, Org. Lett.
2010, 12, 3878–3881; c) T. Brꢄckl, I. Thoma, A. J. Wagner, P. Kno-
chel, T. Carell, Eur. J. Org. Chem. 2010, 6517–6519.
[23] A. Klapars, S. L. Buchwald, J. Am. Chem. Soc. 2002, 124, 14844–
14845.
[24] Knochel and co-workers used allyl bromide in the present of
CuCN·2LiCl as electrophile for the second metal–halide exchange.
[25] See the Supporting Information.
[26] 2-Butanol, Sigma–Aldrich Cat. No.: 294810, anhydrous, 99.5%.
[27] According to the Sigma–Aldrich titration method, Titrimetric analy-
sis of organolithium compounds and Grignard reagents.
[28] Commercially available CuI was washed with THF by using a Soxh-
let extractor before use. After being dried in vacuo, CuI was stored
in a freezer under argon.
[29] J. Bartoszewski, Acta Pol. Pharm. 1954, 11, 189–194.
[30] R. Smoum, A. Rubinstein, M. Srebnik, Bioorg. Chem. 2003, 31,
464–474.
[17] Q. Jiang, M. Ryan, P. Zhichkin, J. Org. Chem. 2007, 72, 6618–6620.
[18] a) A. Krasovskiy, B. F. Straub, P. Knochel, Angew. Chem. 2006, 118,
165–169; Angew. Chem. Int. Ed. 2006, 45, 159–162; b) A. Krasov-
skiy, P. Knochel, Angew. Chem. 2004, 116, 3396–3399; Angew.
Chem. Int. Ed. 2004, 43, 3333–3336; c) P. Knochel, W. Dohle, N.
Gommermann, F. F. Kneisel, F. Kopp, T. Korn, I. Sapountzis, V. A.
Vu, Angew. Chem. 2003, 115, 4438–4456; Angew. Chem. Int. Ed.
2003, 42, 4302–4320.
[19] a) C. J. Rohbogner, G. C. Clososki, P. Knochel, Angew. Chem. 2008,
120, 1526–1530; Angew. Chem. Int. Ed. 2008, 47, 1503–1507;
b) D. R. Armstrong, P. Garcꢆa-ꢇlvarez, A. R. Kennedy, R. E.
Mulvey, J. A. Parkinson, Angew. Chem. 2010, 122, 3253–3256;
Angew. Chem. Int. Ed. 2010, 49, 3185–3188; c) P. Garcꢆa-ꢇlvarez,
D. V. Graham, E. Hevia, A. R. Kennedy, J. Klett, R. E. Mulvey,
C. T. OꢁHara, S. Weatherstone, Angew. Chem. 2008, 120, 8199–8201;
Angew. Chem. Int. Ed. 2008, 47, 8079–8081.
[20] a) O. Baron, P. Knochel, Angew. Chem. 2005, 117, 3193–3195;
Angew. Chem. Int. Ed. 2005, 44, 3133–3135. The Grignard reagent
isopropyl magnesium chloride lithium chloride (iPrMgCl·LiCl, 13)
can easily be synthesized, but in solution it is not very stable. We
have stored iPrMgCl·LiCl solution under argon at À288C. If the
concentration of the active Grignard had decreased by more than a
half, the reaction time of the halide–magnesium exchange was dra-
matically prolonged.
Received: August 23, 2012
Revised: November 13, 2012
Published online: December 23, 2012
2456
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 2450 – 2456