Attempted [2]Catenane Synthesis via a Quasi[1]catenane by a Templated Backfolding Strategy

Article abstract of DOI:10.1002/ejoc.201701325

A templated backfolding concept to construct a [2]catenane was attempted via a quasi[1]catenane showing an inverted spiro geometry. The template is covalently connected to the ketal-connected semi-perpendicular-arranged linear precursors and spatially directs the sterically congested backfolding macrocyclizations that are required to give a quasi[1]catenane. So far, we are unable to hydrolyze the cyclic ketal to liberate the [2]catenane.

Full text of DOI:10.1002/ejoc.201701325

A Journal of
Accepted Article
Title: Attempted [2]catenane synthesis via a quasi[1]catenane by a
templated backfolding strategy
Authors: Jan Herman Van Maarseveen, Luuk Steemers, Martin
Wanner, Bart van Leeuwen, and Henk Hiemstra
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201701325
Link to VoR: http://dx.doi.org/10.1002/ejoc.201701325
Supported by

10.1002/ejoc.201701325
European Journal of Organic Chemistry
COMMUNICATION
Attempted [2]catenane synthesis via a quasi[1]catenane by a
templated backfolding strategy
Luuk Steemers,[a] Martin J. Wanner,[a] Bart R. C. van Leeuwen,[a] Henk Hiemstra[a] and Jan H. van
Maarseveen*[a]
Abstract: A templated backfolding concept to construct a [2]catenane
was attempted via a quasi[1]catenane showing an inverted spiro
geometry. The template is covalently connected to the ketal-
connected semi-perpendicular arranged linear precursors and
Recently, we reported a strategy using the perpendicular
arrangement of a tetrahedral carbon atom for the synthesis of
bicyclic molecule 1 showing an inverted sipro architecture, coined
as a quasi[1]catenane (see Figure 1).[2] Inspired by the landmark
paper by Schill we report here our efforts to combine the best of
both worlds to arrive at [2]catenanes via quasi[1]catenane 2 by
introducing scissile bonds at the connecting tetrahedral carbon
atom. Figure 1 illustrates a comparison of the design of the two
quasi[1]catenanes differing in replacing the permanent central
fluorene moiety reported before by a rigid cyclic ketal linkage
based on L-(+)-tartaric acid in the current communication.
spatially
directs
the
sterically
congested
backfolding
macrocyclizations that are required to give a quasi[1]catenane. So far,
we are unable to hydrolyze the cyclic ketal to liberate the [2]catenane.
Introduction
To disclose the natural lasso peptide series in the far future, we
are currently exploring covalent strategies towards mechanically
interlocked molecules that do not contain the supramolecular
motifs generally applied in the current synthetic approaches. Back
in 1967 Schill reported one of the first [2]catenane syntheses
using a covalent approach.[1] Crucial in their approach was the
use of a cyclic ketal as motif for preorganization of the ring
fragment, ensuring a perpendicular arrangement of the two linear
ring precursors before ring-closure (see scheme 1).
27
26
O
O
HN
NH
O
O
NH
N
HN
N
N
N
N
N
N
N
H
H
N
N
O
O
N
N
N
O
O
27
N
CO2Me
31
O
MeO2C
CO2Me
MeO2C
O
1
quasi[1]catenanes
2
Cl
HO
OH
Figure 1. The previously synthesized permanent quasi[1]catenane 1 and
analog 2 employing a scissile ketal.
OAc
O
O
O
NAc
NAc
NH2
AcO
AcO
This cyclic ketal ensures the desired perpendicular arrangement
of the ring and thread fragments and is acid labile, allowing
hydrolysis during the final acidolytic cleavage step. L-(+)-tartaric
acid was chosen as a building block, as it possesses two
carboxylic acid moieties and two hydroxyl functionalities, allowing
ketal formation. Moreover, tartaric acid chemistry is well
developed and starting materials are readily available. As shown
in Scheme 2, the design of the linear precursor 4 features terminal
alkyne and alkene moieties to allow closure of the macrocyclic
rings via respective Cu-catalyzed azide alkyne cycloaddition
(CuAAC) and ring-closing metathesis (RCM) reactions. Central in
our approach is the use of a template 5 that is temporarily
connected to the tartaric acid-containing ring-precursor fragment
4 via esterification/lactonization to the acid cleavable benzylic
Cl
Scheme 1. Schill’s approach towards a [2]catenane via ketal preorganization.
The ketal motif was cleaved in a late stage using aqueous HBr,
liberating the ketone and a catechol moiety. To the best of our
knowledge, this approach is the only successful synthesis of a
mechanically interlocked product employing a cyclic ketal as motif
for pre-organization, despite its relative simplicity and facile
accessibility.
tethers.
This
forces
both
the
subsequent
CuAAC
macrocyclizations as well as the final RCM macrocyclization in a
backfolding fashion. Cleavage of the lactones via
transesterification and protolytic cleavage of the benzylic linkages
provides quasi[1]catenane 2. Final hydrolysis of the ketal in 2
liberates the mechanically locked [2]catenane 3.
[a]
Van ‘t Hoff Institute for Molecular Sciences
University of Amsterdam
Science Park 904, 1098XH Amsterdam, The Netherlands
Homepage: http://hims.uva.nl/soc
E-mail: j.h.vanmaarseveen@uva.nl
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701325
European Journal of Organic Chemistry
COMMUNICATION
However, under the various reaction conditions tested, only
hydrolysis of the acetal was observed (forming the starting ketone
7 again) with no traces of the desired tartaric acid ketal 10. It is
thought that the close proximity of the methyl ester moieties to the
acetal prevents the desired reaction to take place, probably by an
unwanted but favored 5-exo-trig attack of the ester carbonyls at
the intermediate oxonium intermediate. Therefore, we chose to
mask the ester moieties on the 4-oxopimelic acid side during
ketalization. It was decided to use an alkene as a carboxylate
synthon, eventually transforming it through sequential oxidation
via the bis-aldehyde to the bis-carboxylic acid. This synthetic
detour started with a Grignard reaction of 4-bromo-1-butene 11
with ethyl formate to give the symmetric secondary alcohol in
quantitative yield (see scheme 4.[4] Subsequent oxidation with
pyridinium chlorochromate (PCC) gave ketone 12 in 91% yield
over the two steps, after column chromatography.[5]
O
O
N
HN
NH
HO
OH
O
N
H
N
H
N
O
N
O
N
N
N
CO2Me
MeO2C
[2]catenane 3
CuAAC
RCM
O
O
N
H
N
O
O
O
H
N
N
✂
O
scissile bond
bicyclic ring-
precursor 4
1) Mg, THF
HC(OMe)3,
2) HCO2Et
MeO
OH
HO
OMe
pTsOH, MeOH
Br
CuAAC
O
12 (91%, 2 steps)
11
3) PCC, SiO2
DCM
N3
O
esterification
HO
OH
C6F5O
HO2C
CO2H
O
N3
OC6F5
template 5
O
O
L-(+)-tartaric acid 6
9, pTsOH, DCM
MeO2C
CO2Me
MeO OMe
Scheme 2. Outline of the covalent template-directed backfolding strategy.
13
14 (71%, 2 steps)
OHC
CHO
HO2C
CO2H
1) O3
NaClO2
NaH2PO4
Results and Discussion
O
O
O
O
2) PPh3
DCM
t-BuOH/H2O
MeO2C
CO2Me
MeO2C
CO2Me
The synthesis commenced with the build up of the tartaric acid-
based acyclic ring precursor. It was decided to install the pivotal
cyclic ketal moiety as early on as possible in the synthesis. The
most logical reaction to form the desired skeleton was via
ketalization of dimethyl tartrate 9 with a diester of 4-oxopimelic
acid 7 as this would directly form the desired core 10 with the
correct oxidation states. Although orthogonally protected esters
would be most desirable, it was decided to first test the ketal
formation with dimethyl tartrate 9 and diethyl 4-oxopimelate 7
(both commercially available). To activate the ketone moiety, it
was first reacted with excess trimethyl orthoformate in methanol,
forming acetal 8 (see Scheme 3).[ 3 ] Note that under these
conditions the ethyl esters were transformed to methyl esters.
Subsequent acid catalyzed trans-acetalization with dimethyl
tartrate was conducted next.
15
16 95% (2 steps)
Scheme 4. Synthetic scheme of the L-(+)-tartaric acid based central core.
Next, the acid catalyzed ketal formation of ketone 12 with dimethyl
tartrate was tested. Refluxing with p-TsOH in toluene with a Dean-
Stark trap gave no conversion and also refluxing with BF3 Et2O in
.
dry CH2Cl2 failed to give conversion. Therefore, the same mode
of activation was chosen as for the attempted synthesis of 10.[6,7,8]
As a result, ketone 12 was first transformed into dimethyl acetal
13 with trimethyl orthoformate in dry methanol. Subsequent p-
TsOH catalyzed trans-acetalization with two equivalents of
dimethyl tartrate 9 in refluxing dichloromethane gave a clean
conversion to cyclic ketal 14, which was isolated in 71% yield over
the two steps. Subsequent oxidation of the terminal alkenes
towards the aldehydes giving 15 was optimized next. Initially, the
Lemieux-Johnson reaction,[9] i.e. oxidation with OsO4 and excess
NaIO4, was used, but yields of dialdehyde 15 were not
satisfactory. Fortunately, ozonolysis of the terminal alkenes,
followed by reduction with PPh3 gave a clean conversion to
dialdehyde 15. Reduction of the ozonide intermediate with
dimethyl sulfide, thiourea or NMMO gave lower yields and/or more
byproducts. Because isolated yields of the sensitive dialdehyde
after column chromatography were mediocre at best, it was
decided to use the crude dialdehyde still contaminated with PPh3
and PPh3O directly in the subsequent Pinnick oxidation step.[10,11]
Treatment with NaH2PO4 and NaClO2 using 2-methyl-2-butene as
HC(OMe)3
EtO2C
CO2Et
MeO2C
CO2Me
MeO OMe
pTsOH,
MeOH
7
O
8
HO
OH
MeO2C
CO2Me
9
MeO2C
CO2Me
O
O
pTsOH or PPTS
toluene, MeCN or neat
MeO2C
CO2Me
10
Scheme 3. Attempted synthesis of central L-(+)-tartaric acid-based core.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701325
European Journal of Organic Chemistry
COMMUNICATION
the HOCl scavenger yielded diacid 16 in 95% yield over the two
steps. Purification of the diacid was achieved by extracting it into
the water layer with NaHCO3 and washing the water layer with
EtOAc to remove the PPh3, PPh3O and other apolar organic
residues. Subsequent careful acidification of the water layer
allowed extraction of the product into the organic phase, giving
fairly pure diacid 16.
aldehyde 21 by the Hass-Bender oxidation using 2-nitropropane
in 75% yield.[14] Further oxidation to the esters with V2O5 and
hydrogen peroxide in acidic ethanol yielded the terephthalic ester
in 69% yield.[15] Because some mono-carboxylic acid was still
present due to hydrolysis this yield may be improved. Fortunately,
the subsequent radical bromination went cleanly, giving
dibromide 22 in 66% yield after recrystallization. The azides were
introduced cleanly and almost quantitatively, giving the di-azide
as a colorless solid. Saponification with LiOH gave di-acid 23 in
quantitative yield. Treatment of di-acid 23 with pentafluorophenol,
DIPEA and HBTU as coupling reagent gave the activated
template 5 as a colorless solid in a moderate 52% yield.
Next, coupling of diacid 16 with 5-amino-1-pentyne[ 12 ] was
optimized.
O
O
1) H2N
PyBOP, DiPEA,
DCM (94%)
N
H
N
O
O
H
With all building blocks in hand the scene is set for the final
assembly of the quasi[1]catenane 2. Macrocyclization of tetra-
amide 4 with template 5 was performed using optimized
transesterification conditions[2] by stirring the two components
with 10 equiv of Cs2CO3 and 4Å molecular sieves in acetonitrile at
high dilution (2 mM), giving macrocycle bis-ester 24 in 67% yield
(Scheme 8). As observed in our previous work on the synthesis
of the structurally similar quasi[1]catenane 1,[2] the 1H-NMR
spectrum of the macrocycle 24 shows a high complexity (see
supporting information). Besides rotamers emerging from the
tertiary amides the complexity of the NMR spectrum of 24 might
be further increased due to the presence of two diastereomers
caused by hindered rotation around the endocyclic terephthalic
core connecting ester single bonds introducing a center of planar
chirality, as depicted in the cartoon below (Scheme 7). The 27-
membered ring probably does not allow free rotation of the ester
bonds as was the case in the similar precursor towards the
synthesis of quasi[1]catenane 1. In contrast to the perfect flat
central five-membered ring within the fluorene core as in 1, the
1,3-dioxolane ring as in 24 is slighlty puckered thus further
16
2) NaOH, THF/H2O/
MeOH (69%)
HO2C
CO2H
17
+
NH
DCC, HOBt,
DMAP, DCM
4
18
(61%)
H3CO
OH
Scheme 5. Build up of the ring-precursor fragment.
Standard coupling conditions between the amine and diacid 16
with DCC and HOBt gave low yields. However, treatment with 2.5
equivalents of PyBOP gave full conversion, giving the diamide in
94% yield, which was pure enough for the next step. The two
methyl esters were smoothly saponified with aqueous NaOH,
giving diacid 17 in 69% yield. Impurities were removed by washing
the basic water layer with ethyl acetate. The product was
extracted after careful acidification of the water layer. Next, diacid
17 was coupled to amine 18[2] with DCC and HOBt as coupling
reagents, giving ring-precursor tetra-amide 4 in 61% yield.
With ring precursor 4 in hand, the di-azide template 5 had to be
synthesized.
1
breaking the symmetry also contributing to the H-NMR spectra
complexity.
Me
Me Br
NO2
Me
CHO
(CH2O)n, HBr,
HOAc (54%)
O
O
O
O
NaH, DMF
(75%)
OHC
19
20
21
Me
Br Me
Me
Br
1) V2O5, H2O2,
HCl, EtOH (69%)
1) NaN3, DMF (97%)
CO2Et
24
single bond rotation
Scheme 7. Possible diastereomer formation due to hindered rotation at the
terephthalic single bond ester connections.
2) LiOH, H2O, MeOH/
THF (quant.)
EtO2C
2) NBS, DCM, hν
(66%)
N3
22
Br
N3
O
Subsequent CuAAC reaction with catalytic Cu(CH3CN)4BF4 and
TBTA as ligand in dichloromethane gave bis-triazole cage
molecule 25 uneventfully in 80% yield.[16] Subsequent ring closure
via olefin metathesis using Grubbs 2nd generation catalyst was
unexpectedly difficult. Despite various experiments, the
macrocyclic olefin 26 was obtained in 25% yield only. Moreover,
as observed in our previously reported quasi[1]catenane
synthesis, trace amounts of the CH2-truncated product were also
observed.[2] Cleavage of the lactone esters was achieved by
treatment with excess K2CO3 in methanol, giving the diester 27 in
82% yield. These conditions suppress unwanted saponification
due to trace amounts of water in the reaction mixture as observed
occasionally when employing NaOCH3 in ‘anhydrous’ methanol.
CO2H
OC6F5
C6F5OH, HBTU,
C6F5O
HO2C
N3
DiPEA, DCM
(52%)
O
N3
23
template 5
Scheme 6. Scaffold synthesis.
The synthesis started with bromomethylation of p-xylene 19 to
give dibromide 20 in a moderate but acceptable 54% yield (see
scheme 6).[13] Bis-benzylic bromide 20 was transformed into di-
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701325
European Journal of Organic Chemistry
COMMUNICATION
Next, the alkene E/Z-mixture was removed by catalytic
hydrogenation to give 27-H2. The final steps of the attempted
synthesis of [2]catenane 3 included acidolytic cleavage of the
benzyl groups and acidic hydrolysis of the ketal moiety.
quasi[1]catenane architecture of 2 by 1H-NMR is still severely
hampered. Heating up the NMR sample in deuterated DMSO to
120 °C resulted in a less complex spectrum in which the peaks of
the conformers coalesced. At that temperature in the 7-8 ppm
region the terephtalic, triazole and two different amide-N protons
gave four broad but discrete signals. Also noteworthy are the
ester methyl signals that now gave one peak. Disappointingly,
various attempts to hydrolyze the ketal in 2 failed to give any of
the desired [2]catenane, usually resulting in acid catalyzed
hydrolysis of the methyl esters of the terephthalate moiety only
(see supplementary information). We reason that the electron
withdrawing esters of the tartaric acid moiety thwart protonation
of the dioxolane oxygen. In addition, the severe steric hindrance
around the endocyclic ketal hampers hydrolysis. After a landmark
publication by the Sauvage group mentioning the catenand effect,
various other reports describe the inert environment inside the
core of interlocked molecules due to steric isolation.[17] Therefore,
the fact that 2 is completely reluctant to hydrolysis can be seen as
an indirect proof of the structure. Currently, work is in progress to
to synthesize the sterically less encumbered [2]rotaxanes using
the same strategy. In addition, a more electron-rich ketal will be
installed in combination with replacing the aliphatic chains for
peptidic chains allowing water molecules to enter the
quasi[1]catenane cavity to facilitate hydrolysis. Furthermore, by
replacing the ketal for a imidazolidin-4-one as the perpendicular
and cleavable thread/ring connecting moiety, hydrolysis will result
in a peptide thus coming closer to the far future lasso peptide
series.
O
O
Cs2CO3
4Å MS
N
N
H
O
O
H
4 + 5
CH3CN
67%
N
O
N
O
O
MeO
OMe
O
N3
O
N3
24
O
O
O
Cu(CH3CN)4BF4,
TBTA, DCM
(80%)
N
H
N
O
O
H
N
O
N
N
Grubbs II, DCM
(25%)
O
O
N
N
N
N
N
MeO
OMe
O
O
O
O
25
O
N
H
N
H
O
O
1) MeOH, K2CO3,
DCM, EtOAc
(82%)
2) H2, Pd(C),
MeOH, EtOAc
(quant.)
N
N
N
N
O
Conclusions
O
N
N
N
N
MeO
O
OMe
Previously we have developed a template-directed covalent
strategy in which a linear chain is forced to backfold enabling
macrocyclization over another linear molecule. This appoach led
to a fascinating compound class coined as quasi[1]catenanes and
characterized as bicyclic compounds with an inverted spiro
architecture in which the rings are connected via a fluorene-
centered tetrahedral C-atom. In this communication we have
replaced the fluorene by a tartaric acid derived ketal having a
similar geometry. Hydrolysis of the ketal liberates the
mechanically locked [2]catenane skeleton. All steps towards the
ketal centered quasi[1]catenane, i.e. macrolactonization, CuAAC
macrocyclizations and RCM towards a cycloolefin, worked well.
Unfortunately, so far we are unable to hydrolyze the central ketal
moiety.
O
O
26
O
O
N
O
N
H
H
O
O
N
N
OH
OH
N
TFA, Et3SiH, DCM
(85%)
O
O
N
N
2
N
N
N
MeO
OMe
MeO2C
CO2Me
27-H2
Scheme 8. End game towards quasi[1]catenane 2.
Treatment with excess triethylsilane as a cation scavenger in TFA
gave quasi[1]catenane 2 in ca. 85% yield. HRMS of the product
showed the presence of the desired mass (m/z 1039, M+Na+),
however also trace amounts of m/z 1037 were observed,
corresponding to the M+Na+ signal of the non-hydrogenated
analogue of 2, meaning that the hydrogenation step had not
reached completion yet. Surprisingly, no hydrolysis of the ketal
was observed during the LCMS analysis. Most probably due to
the presence of multiple conformations and two diasteromers as
discussed above, the unambiguous assignment of the
Acknowledgements
The Netherlands Research Council (NWO) is acknowledged for
financial support (ECHO-grant 711.012.009) to J.H.v.M. We thank
E. Zuidinga and J.M. Ernsting for high-resolution mass
spectrometry and NMR analysis.
Keywords: catenanes • templated synthesis • synthetic
methodology • spiro compounds •
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701325
European Journal of Organic Chemistry
COMMUNICATION
[1]
[2]
G. Schill, Chem. Ber. 1967, 100, 2021-2037.
L. Steemers, M. J. Wanner, M. Lutz, H. Hiemstra, J. H. Van
Maarseveen, Nat. Commun. 2017, 8, 15392.
M. Sander, E. V. Dehmlow, Eur. J. Org. Chem. 2001, 2, 399-
404.
G. Lemiere, V. Gandon, K. Cariou, A. Hours, T. Fukuyama, A.-
L. Dhimane, L. Fensterbank, M. J. Malacria, Am. Chem. Soc.
2009, 131, 2993-3006.
[11] B. S. Bal, W. E. Childers, H. W. Pinnick, Tetrahedron, 1981,
37, 2091-2096.
[12] T. B. Yu, J. Z. Bai, Z. Guan, Angew. Chem. Int. Ed. 2009, 48,
1097-1101.
[13] H. L. Zhang, G. Y. Zhu, G. Wang, M. Meng, C. Y. Liu, H. Lei,
Inorg. Chem. 2015, 54, 11314-11322.
[14] H. B. Hass, M. L. Bender, J. Am. Chem. Soc. 1949, 71, 1767-
1769.
[3]
[4]
[5]
[6]
J.-C. Legeay, W. Lewis, R. A. Stockman, Chem. Comm. 2009,
16, 2207-2209.
[15] R. Gopinath, B. K. Patel, Org. Lett. 2000, 2, 577-579.
[16] T. R. Chan, R. Hilgraf, K. B. Sharpless, V. V. Fokin, Org. Lett.
2004, 6, 2853-2855.
[17] a) A.-M. Albrecht-Gary, Z. Saad, C. O. Dietrich-Buchecker, J.-
P. Sauvage, J. Am. Chem. Soc. 1985, 107, 3205-3209. b) A.
H. Parham, B. Windisch, F. Vögtle, Eur. J. Org. Chem. 1999,
1233-1238. c) J. Winn, A. Pinczewska, S. M. Goldup, J. Am.
Chem. Soc. 2013, 135, 13318-13321. d) L. Steemers, A. W.
Ehlers, M. J. Wanner, H. Hiemstra, J. H. van Maarseveen,
Org. Lett. 2017, 19, 2342-2345.
J. Weng, Y.-B. Li, R.-B. Wang, F.-Q. Li, C. Liu, A. S. C. Chan,
G. Lu, J. Org. Chem. 2010, 75, 3125-3128.
D. Crich, X. J. Hao, Org. Chem. 1999, 64, 4016-4024.
N. J. Lajkiewicz, S. P. Roche, B. Gerard, J. A. Porco Jr, J. Am.
Chem. Soc. 2012, 134, 13108-13113.
[7]
[8]
[9]
R. Pappo, D. S. Allen, R. U. Lemieux, W. S. Johnson, J. Org.
Chem. 1956, 21, 478-479.
[10] B. O. Lindgren, T. Nilsson, S. Husebye, O. Mikalsen, K.
Leander, C.-G. Swahn, Acta Chemica Scandinavica, 1973,
27, 888-890.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201701325
European Journal of Organic Chemistry
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
COMMUNICATION
Catenanes
Luuk Steemers, Martin J. Wanner, Bart
R. C. van Leeuwen, Henk Hiemstra and
Jan H. van Maarseveen*
Page No. – Page No.
Attempted [2]catenane synthesis via
a quasi[1]catenane by a templated
backfolding strategy
A tartaric acid derived ketal-centered quasi[1]catenane is prepared by a template-
directed covalent strategy forcing backfolding macrocyclizations. All key steps
towards the quasi[1]catenane, i.e. macrolactonization, CuAAC macrocyclizations
and RCM towards a cycloolefin, worked well. Unfortunately, all attempts to
hydrolyze the ketal liberating the [2]catenane architecture failed.
Molecular knotting
O
H2O, H+
steps
O
O
HO OH
O
O
quasi[1]catenane
[2]catenane
This article is protected by copyright. All rights reserved.