Synthesis of the TACO scaffold as a new selectively deprotectable conformationally restricted triazacyclophane based scaffold

Article abstract of DOI:10.1021/ol5012218

The synthesis of a new triazacyclophane scaffold (TACO scaffold) containing three selectively deprotectable amines is described. The TACO scaffold is conformationally more constrained than our frequently used TAC scaffold, due to introduction of a substituent on the para position of the benzoic acid hinge, which prevents ring flipping and makes it more attractive than the TAC scaffold for preparation of artificial receptor molecules or for mimicking discontinuous epitopes toward protein mimics when more preorganization is required.

Full text of DOI:10.1021/ol5012218

Letter
pubs.acs.org/OrgLett
Synthesis of the TACO Scaffold as a New Selectively Deprotectable
Conformationally Restricted Triazacyclophane Based Scaffold
Arwin J. Brouwer,†,§ Helmus van de Langemheen,†,§ Adriano Ciaffoni,† Kitty E. Schilder,†
and Rob M. J. Liskamp*,†,‡
†Medicinal Chemistry and Chemical Biology, Utrecht Institute for Pharmaceutical Sciences, Faculty of Science, Utrecht University,
P.O. Box 80082, 3508 TB Utrecht, The Netherlands
‡School of Chemistry Joseph Black Building, University of Glasgow, University Avenue, Glasgow G12 8QQ, United Kingdom
S
* Supporting Information
ABSTRACT: The synthesis of a new triazacyclophane scaffold (TACO scaffold) containing three selectively deprotectable
amines is described. The TACO scaffold is conformationally more constrained than our frequently used TAC scaffold, due to
introduction of a substituent on the para position of the benzoic acid hinge, which prevents ring flipping and makes it more
attractive than the TAC scaffold for preparation of artificial receptor molecules or for mimicking discontinuous epitopes toward
protein mimics when more preorganization is required.
olecular scaffolds have been described in the literature to
which three or more identical ligands such as peptides
approaches we are now able to introduce different ligands,
especially (cyclic) peptides, onto our scaffolds and therefore we
can now increasingly focus our attention on the much more
difficult issue of attempting to orient the ligands with respect to
each other in space. In order to try to achieve this we have so
far (1) varied the rigidity of the scaffold8 and (2) attempted to
influence the conformational space of the cyclophane ring and
therefore the orientation of the ligands, by changing its ring
size. Here we describe the “TACO” scaffold, a new O-alkylated
TAC-scaffold, in which the O-alkyl substituent may function as
an important orientation director (Figure 1). The TAC scaffold
consists of a benzoic acid hinge connected to a symmetric
bridge containing three amine functional groups with semi-
orthogonal protecting groups. Naturally, despite the presence
of the benzene ring as part of these large (14-membered)
cyclophane macrocycles, they are very flexible. As a
consequence, ring flipping occurs easily in these macrocycles9
leading to different conformers. In order to reduce ring flipping,
the triazacyclophane ring was reduced by one CH2 moiety,
which led to the A(symmetric)TAC scaffold.10 The ATAC
scaffold was less flexible, but not surprisingly, ring flipping still
occurred. We envisioned that, by introduction of a substituent
on the para position of the aromatic ring, thus affording a TAC-
O-alkylated molecular scaffold (Figure 1), ring flipping might
M
can be attached, but only a few are suitable for the selective
attachment of three different ligands.1 The TAC (TriAzaCy-
clophane) scaffold, developed in our group, was one of the first
selectively deprotectabe scaffolds.2 This scaffold consists of a
benzoic acid hinge connected to a symmetric cyclophane bridge
containing three secondary amines, which carry three different
selectively removable protecting groups (Figure 1). We have
Figure 1. TAC scaffold vs TACO scaffold.
used the TAC scaffold for the development of synthetic
receptors,3,4 synthetic vaccines5 involving mimicry of dis-
continuous epitopes,7,8 and combinatorial libraries,3,4,6 for
example toward artificial enzymes.6 Toward the mimicry of a
discontinuous epitope of HIV-gp120, we compared a small
number of chemically diverse scaffolds, which underlined the
importance of the nature of scaffolding.8 Using different
Received: April 29, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ol5012218 | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 1. Synthesis of Different TACO Scaffolds
be further reduced. Both simple physical molecular models and
preliminary molecular modeling experiments (See Supporting
Information) show that an O-alkyl group pushes P1 and P3
upward, whereas P2 will stay down (Figure 1).
Table 1. Yields of Alkylations of 3 with Different Alkyl
Halides (R−X), Affording 4a−4g (1 equiv of 3, 1.2 equiv of
alkyl halide, and K2CO3, DMF, 75 °C)
In principle substituents such as alkyl, amine, ether, and ester
can be introduced. However, an ether linkage was preferred
since it is very stable, neutral, and easily accessible. Instead of
introducing the alkoxy substituent at the para (4) position of
3,5-dimethylbenzoic acid, which is the starting material in the
TAC scaffold synthesis,10 it was decided that the methyl groups
would be introduced at both meta (3 and 5) positions of
methyl 4-hydroxybenzoate (1, Scheme 1). This was syntheti-
cally attractive, and the use of methyl 4-hydroxybenzoate as a
starting material was also substantially cheaper than using 3,5-
dimethylbenzoic acid.
Thus, methyl-4-hydroxybenzoate (1) was subjected to a
modified Duff reaction11 by treatment with hexamine in
refluxing TFA as a solvent, leading to bis-formylated benzoate
2 in 70% yield (Scheme 1). For the possibility of coupling an
alkyl group to the phenolic−OH at the last stage of the
synthesis, temporary protection was attempted using silyl
protecting groups (TBDMS and TMSE). Unfortunately, all
attempts were unsuccessful, pointing to the low nucleophilicity
of the phenolic−OH, likely due to delocalization of the
negative charge of the phenolate on both aldehyde function-
alities in 2. To increase its nucleophilicity, it was decided that
both aldehydes of 2 would be reduced using NaBH4 in
methanol, affording trihydroxy benzoate 3 in high yield (87%).
Selective protection of a phenolic−OH in the presence of
benzylic alcohols is impossible,12,13 therefore alkylation of the
phenolic−OH was decided for this stage of the synthesis, since
the resulting ether moiety will be inert to subsequent synthetic
transformations. Standard phenol alkylation conditions using an
alkyl halide in acetone or DMF with K2CO3 as a base14,15 were
optimized leading to treatment with a slight excess of the alkyl
bromide (1.2 equiv) in DMF at 75 °C (Table 1). Good to high
yields (66% to 100%) were obtained with several alkylating
agents. Not surprisingly, the methyl ether (4a) was prepared in
high yield too. Alkylation with functionalized alkyl bromides
Boc-2-bromoethylamine and propargyl bromide also proceeded
smoothly leading to ethers 4f and 4g, which now contained an
additional handle for convenient functionalization (CuAAC,
a
b
Temperature: 40 °C. Temperature: 50 °C.
thiol−ene, conjugation, etc.) after completion of a TACO-
molecular construct.
In attempts to equip the benzyl alcohols with mesylate
leaving groups, bis-benzylic alcohols 4a−4g were reacted with
methanesulfonyl chloride and triethylamine as a base. Not
unexpectedly, no bis-mesylates were found, but instead bis-
benzylic chlorides 5a−5g were isolated16 which were
sufficiently pure for direct use in the macrocyclization reaction
leading to 7a−7i (Scheme 1). For evaluation of the scope of
this reaction, the more easily accessible triamine containing
three oNBS-groups (6a, P1 = P2 = oNBS) was used.3 The
macrocyclization reaction was performed using 1 equiv of both
the bis-benzylic chloride and triamine 6a in DMF, in the
presence of potassium iodide and cesium carbonate as a base.
The macrocyclization reaction using the bis-benzylic chlorides
5a−5g proceeded equally well as when using a bis-benzylic
bromide,10 and TACO scaffolds 7a−7g, including those
containing groups suitable for further functionalization (7f
and 7g), were obtained in moderate yields (42−67%, over two
steps) (Table 2), after column chromatography. Even TACO
B
dx.doi.org/10.1021/ol5012218 | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
scaffold 7e with the sterically demanding 3,5-di-tert-butylbenzyl
group was afforded in a moderate yield of 51%.
Thus, TACO-scaffold 8b was coupled to Tentagel resin
containing a Rink amide linker, using BOP as a coupling
reagent and DiPEA as a base, affording resin 9 (Scheme 2).
Removal of the oNBS-groups by 2-mercaptoethanol and DBU
in DMF was followed by subsequent coupling of Fmoc-
Ser(tBu)-OH and Fmoc-Lys(Boc)-OH using BOP and DiPEA.
Removal of the Fmoc-groups by piperidine and then acetylation
afforded TACO-tris-dipeptide 10. Cu(I)-assisted azide−alkyne
cycloaddition (CuAAC)17,18 was carried out after treatment of
resin 10 with TFA/TIS/H2O to afford 11, which was purified
by preparative HPLC. Reaction with azidoPhe-Leu-OMe (12),
using CuSO4 and sodium ascorbate in a DMF/water mixture
under microwave irradiation conditions, afforded TACO
scaffold 13 now containing four peptides in a good yield of
66%, after preparative HPLC.
Table 2. Yields of TACO Scaffolds 7a−7i after the
Macrocyclization Reaction with Protected Triamines 6a
(entries 1−7) and 6b (entries 8−9) (1.0 equiv of 5a−5g and
6a or 6b, 4.0 equiv of Cs2CO3, 0.1 equiv of KI, DMF, rt)
To determine whether an additional substituent at the para
position of the benzoic acid hinge is capable of inhibition of the
ring flipping process, TACO scaffold 7a was compared with a
TAC scaffold containing three oNBS-groups (Figure 2). Ring
1
flipping can be easily monitored using variable temperature H
NMR experiments (Figure 3).9,10
a
CH3CN, reflux, 1 h.
After these successful macrocyclizations, triamine 6b,10
containing now three different protecting groups (P1 = Aloc,
P2 = TFA), was used in this alkylation reaction with bis-
benzylic chlorides 5a and 5f, and TACO scaffolds 7h and 7i
were indeed obtained in reasonable yields (41% and 49%,
respectively). Finally, saponification of the methyl ester in
scaffolds 7g and 7i, and simultaneous cleavage of the TFA-
group (in 7i), followed by introduction of an Fmoc-group (for
preparing 8a) afforded TACO scaffolds 8a and 8b in high yields
(96% and 95%, respectively), after column chromatography.
To investigate whether an additional substituent indeed can
be introduced on the alkyne group of a TACO scaffold already
containing three peptides, TACO-scaffold 10 was synthesized.
Figure 2. Superimposed ring flipping isomers of the TACO and TAC
scaffolds (P1−3 = oNBS).3
At room temperature, the benzylic CH2 groups in the
scaffolds were visible as a singlet for the TAC scaffold and an
AB system for the TACO scaffold. This at least indicated that
the TAC scaffold can ring flip easily at rt, but that the TACO
scaffold is conformationally more restricted because the
benzylic protons are magnetically nonequivalent. The singlet
of the TAC scaffold starts to broaden significantly after cooling
to −80 °C, indicating slower ring flipping. The TACO scaffold
was measured at a high temperature (110 °C), in order to
stimulate ring flipping. Surprisingly, there was hardly any
Scheme 2. Attachment of Three Dipeptides to the Triazacyclophane Ring and an Additional Peptide Connected to the Alkyne
Moiety Using CuAAC Chemistry
C
dx.doi.org/10.1021/ol5012218 | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
REFERENCES
■
(1) (a) Van der Plas, S. E.; Gea, A.; Figaroli, S.; De Clercq, P. J.;
Madder, A. Eur. J. Org. Chem. 2008, 1582−1588. (b) Van der Plas, S.
E.; Van Hoeck, E.; Lynen, F.; Sandra, P.; Madder, A. Eur. J. Org. Chem.
2009, 1796−1805. (c) Verzele, D.; Figaroli, S.; Madder, A. Molecules
2011, 16, 10168−10186. (d) Franke, R.; Hirsch, T.; Eichler, J. Journal
of Receptors and Signal Transduction 2006, 26, 453−460. (e) Franke,
R.; Hirsch, T.; Overwin, H.; Eichler, J. Angew. Chem., Int. Ed. 2007, 46,
1253−1255. (f) Sudarman, E.; Bollati-Fogolín, M.; Hafner, M.; Muller,
̈
Figure 3. Variable temperature 1H NMR signals for the diastereotopic
benzylic CH2 protons from TACO scaffold 7a and the TAC scaffold
(Figure 2, P1−3 = oNBS).
W.; Scheller, J.; Rose-John, S.; Eichler, J. Chem. Biol. Drug Des. 2008,
71, 495−500. (g) Virta, P.; Karskela, M.; Lonnberg, H. J. Org. Chem.
̈
2006, 71, 1989−1999.
(2) Opatz, T.; Liskamp, R. M. J. Org. Lett. 2001, 3, 3499−3502.
(3) Monnee, M. C. F.; Brouwer, A. J.; Liskamp, R. M. J. QSAR Comb.
Sci. 2004, 23, 546−559.
1
change in the H NMR spectrum. The coupling constants for
(4) Monnee, M. C. F.; Brouwer, A. J.; Verbeek, L. M.; van
Wageningen, A. M. A.; Liskamp, R. M. J. Bioorg. Med. Chem. Lett.
2001, 11, 1521−1525.
(5) Hijnen, M.; van Zoelen, D. J.; Chamorro, C.; Gageldonk, P.;
Mooi, F. R.; Berbers, G.; Liskamp, R. M. J. Vaccine 2007, 25, 6807−
6817.
the AB systems of the benzylic CH2 groups were identical at
both temperatures, showing that even at high temperatures
likely no ring flipping occurred in a TACO scaffold even
containing a substituent as small as a methoxy group at the 4-
position of the benzoate hinge (7a). These results indicate that
the TACO scaffold may be promising for use as a more
preorganized scaffold for the construction of protein mimics.
In conclusion, we have described an efficient and high
yielding method for the multigram scale synthesis of a new
(selectively deprotectable) and conformationally more re-
stricted TACO scaffold. Although this scaffold is very similar
to the TAC scaffold frequently used by us, the ether substituent
at the 4-position of the benzene ring has a great influence on
the flexibility of the cyclophane ring. VT NMR measurements
showed that indeed the TACO scaffold is more preorganized
than the TAC scaffold. In addition, a suitably functionalized
ether substituent at this position provided an anchoring
position for another substituent, so that four different ones
can be introduced lining the cyclophane ring, while there is still
the possibility of a fifth substituent to be incorporated onto the
carboxyl moiety of the benzoate. We believe we have provided
a very valuable addition to our repertoire of scaffolds, which can
be applied to the synthesis of highly functionalized protein
mimics or other densely functionalized (bio)molecular
constructs. The present investigation includes their application
in the assembly of synthetic vaccines and other challenging
biomimics.
(6) Albada, H. B.; Liskamp, R. M. J. J. Comb. Chem. 2008, 10, 814−
824.
(7) (a) Chamorro, C.; Kruijtzer, J. A. W.; Farsaraki, M.; Balzarini, J.;
Liskamp, R. M. J. Chem. Commun. 2009, 821−823. (b) Mulder, G. E.;
Kruijtzer, J. A. W.; Liskamp, R. M. J. Chem. Commun. 2012, 48,
10007−10009.
(8) Mulder, G. E.; Quarles van Ufford, H. C.; van Ameijde, J.;
Brouwer, A. J.; Kruijtzer, J. A. W.; Liskamp, R. M. J. Org. Biomol. Chem.
2013, 11, 2676−2684.
(9) (a) Piątek, P.; Kalisiak, J.; Jurczak, J. Tetrahedron Lett. 2004, 45,
3309−3311. (b) Kalisiak, J.; Kalisiak, E.; Jurczak, J. Tetrahedron 2006,
62, 5905−5914. (c) Bogdan, A. R.; Jerome, S. V.; Houk, K. N.; James,
K. J. Am. Chem. Soc. 2012, 134, 2127−2138.
(10) Brouwer, A. J.; van de Langemheen, H.; Liskamp, R. M. J.
Tetrahedron 2014, 70, 4002−4007.
(11) Routaslo, T.; Helaja, J.; Kavakka, J.; Koskinen, A. M. P. Eur. J.
Org. Chem. 2008, 18, 3190.
(12) Amir, R. J.; Danieli, E.; Shabat, D. Chem.Eur. J. 2007, 13,
812−821.
(13) Veldhuyzen, W. F.; Pande, P.; Rokita, S. E. J. Am. Chem. Soc.
2003, 125, 14005−14013.
(14) Masci, B.; Pasquale, S.; Thuer
4838.
́
y, P. Org. Lett. 2008, 10, 4835−
(15) Tsubaki, K.; Otsubo, T.; Morimoto, T.; Maruoka, H.; Furukawa,
M.; Momose, Y.; Shang, M.; Fuji, K. J. Org. Chem. 2002, 67, 8151−
8156.
(16) van Oijen, A. H.; Huck, N. P. M.; Kruijtzer, J. A. W.; Erkelens,
C.; van Boom, J. H.; Liskamp, R. M. J. J. Org. Chem. 1994, 59, 2399−
2408.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details and characterization of the products. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(17) Tornoe, C. W.; Christensen, C.; Meldal, M. J. Org. Chem. 2002,
67, 3057−3069.
(18) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B.
Angew. Chem., Int. Ed. 2002, 41, 2596−2599.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: r.m.j.liskamp@uu.nl, robert.liskamp@glasgow.ac.uk.
Author Contributions
§These authors contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We would like to thank Dr. Johann Jastrzebski and Dr. Johan
Kemmink, for the help with the variable temperature NMR
measurements, and Dr. Ir. D. T. S. Rijkers and Dr. Remco
Merkx, for the kind gift of azidopeptide 12.
D
dx.doi.org/10.1021/ol5012218 | Org. Lett. XXXX, XXX, XXX−XXX