Synthesis of alkynyl-substituted camphor derivatives and their use in the preparation of paclitaxel-related compounds

Article abstract of DOI:10.3762/bjoc.13.122

Compounds containing two alkyne groups in close vicinity at the rigid skeleton of camphorsulfonamide show unique reactivities when treated with electrophiles or catalytic amounts of platinum(II). The formed product structures depend not only on the reagents used but also on the substituents attached to the triple bonds. Cycloisomerisations with perfect atom economy lead to polycyclic heterocycles that resemble to some extent the AB ring system of paclitaxel. Herein, we present practical synthetic methods for the selective synthesis of precursor dialkynes bearing different substituents (alkyl, aryl) at the triple bonds, based on ketals or an imine as protecting groups. We show for isomeric dialkynes that the reaction cascade induced by Pt(II) includes ring annulation, sulphur reduction, and ring enlargement. One isomeric dialkyne additionally allows for the isolation of a pentacyclic compound lacking the ring enlargement step, which we have proposed as a potential intermediate in the catalytic cycle.

Full text of DOI:10.3762/bjoc.13.122

Synthesis of alkynyl-substituted camphor derivatives and
their use in the preparation of paclitaxel-related compounds
M. Fernanda N. N. Carvalho1, Rudolf Herrmann2 and Gabriele Wagner*3,§
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2017, 13, 1230–1238.
1CQE, Instituto Superior Técnico, Universidade de Lisboa, Av.
Rovisco Pais, P-1049-001 Lisbon, Portugal, 2Institute of Physics,
University of Augsburg, Universitätsstr. 1, D-86135 Augsburg,
Germany and 3Department of Natural Sciences, University of
Chester, Thornton Science Park, Pool Lane, Ince, Chester, CH2 4NU,
United Kingdom
doi:10.3762/bjoc.13.122
Received: 07 March 2017
Accepted: 26 May 2017
Published: 26 June 2017
Associate Editor: I. R. Baxendale
Email:
Gabriele Wagner* - g.wagner@chester.ac.uk
© 2017 Carvalho et al.; licensee Beilstein-Institut.
License and terms: see end of document.
* Corresponding author
§ Tel.: +44 (0) 1244 512427
Keywords:
alkynes; camphor derivatives; catalysis; cycloisomerisation; platinum
Abstract
Compounds containing two alkyne groups in close vicinity at the rigid skeleton of camphorsulfonamide show unique reactivities
when treated with electrophiles or catalytic amounts of platinum(II). The formed product structures depend not only on the reagents
used but also on the substituents attached to the triple bonds. Cycloisomerisations with perfect atom economy lead to polycyclic
heterocycles that resemble to some extent the AB ring system of paclitaxel. Herein, we present practical synthetic methods for the
selective synthesis of precursor dialkynes bearing different substituents (alkyl, aryl) at the triple bonds, based on ketals or an imine
as protecting groups. We show for isomeric dialkynes that the reaction cascade induced by Pt(II) includes ring annulation, sulphur
reduction, and ring enlargement. One isomeric dialkyne additionally allows for the isolation of a pentacyclic compound lacking the
ring enlargement step, which we have proposed as a potential intermediate in the catalytic cycle.
Introduction
Enantiomerically pure raw materials, available in a sustainable bicyclic skeleton can easily be modified and adapted to the
manner from the natural “chiral pool” [1], offer a convenient purpose at hand, e.g., natural product synthesis [5]. The
entrance for the chemical synthesis of other chiral compounds, Wagner–Meerwein and Nametkin-type rearrangements are the
e.g., rare natural products and their analogues [2], or chiral cata- most common reaction patterns [6] and the addition of organo-
lysts [3,4]. A prominent example for such a “chiral pool” metallic reagents to the camphor carbonyl group allows for
starting material is camphor. Both its substituents and its selective introduction of additional substituents and functional
1230

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
groups (e.g., as in [7]). Camphor was the source of chirality in
Holton’s taxol synthesis [8] and other approaches to the taxane
group of compounds [9-11].
From cheap camphor-10-sulfonic acid (1), a cyclic sulfonimide
2 can easily be obtained which is readily converted into useful
auxiliaries [12], or oxidized to the oxoimide 3 (Scheme 1) [13-
15]. This versatile intermediate can be reduced to provide a
chelating ligand for chiral catalysis [16], or oxidised to
oxaziridines used as efficient chiral oxidising reagents [13,17-
19].
The reaction of the oxoimide 3 with two equivalents of the lithi-
um salt of a terminal alkyne leads to compounds 4 where two
alkynyl substituents, a sulfonamide and a hydroxy group are
found in vicinal positions (Scheme 1) [20]. A hydroxy group
neighbouring an alkynyl substituent, under treatment with acids,
Scheme 2: Reactions of bis-alkynyl camphor derivative 4a with TiCl4
and with Br2, respectively.
normally leads to Rupe and Meyer–Schuster rearrangements, of the starting material are found in the product, and thus fulfil
forming unsaturated carbonyl compounds. This was indeed ob- an important requirement of “green chemistry” [31].
served in camphor-derived bicyclic alcohols containing a single
ethinyl group [21,22], occasionally accompanied by a However, a different Pt(II)-catalysed reaction cascade was ob-
Wagner–Meerwein rearrangement [23]. However, no such served for 4c, with adamantyl groups at the alkynes. Here, the
products were found with any of the diynes 4. Our first attempts annulation step is followed by a C–H bond-activation process,
to employ 4a as a ligand with Ti(IV) resulted, somewhat sur- to establish an additional bond to one of the adamantyl groups
prisingly, in the addition of HCl under simultaneous annulation (8 in Scheme 3b) [32]. The simplest diyne 4d with R = H, in
(three-carbon expansion [24,25]) of a carbocyclic five-mem- contrast, gave a ring enlargement from six to seven members
bered ring to the 2,3-position of the bicyclic camphor-derived together with a 1,2-oxygen shift instead (9 in Scheme 3c) [33].
moiety (Scheme 2a) [20]. Reactions of 4a with halogens (e.g., This clearly demonstrates that the substituents at the triple
bromine) or acids were even more puzzling. In addition to the bonds have a decisive influence on the outcome of the catalytic
annulation, an unprecedented formation of a ketone accompa- reaction. This was also confirmed by the reaction of 4e
nied by the reduction of sulphur took place, to give a cyclic (R = benzyl): in addition to the expected product 7 in analogy to
sulfinamide 6 (Scheme 2b) [26]. In this case, the mechanism of that of 4b, there was also a considerable amount of a reduced
the reaction proceeded through cationic intermediates as evi- species 10 lacking sulphur reduction and ring enlargement, with
denced by in situ NMR spectroscopy.
structural similarity to the simple product from the Ti(IV) reac-
tion with 4a. The reducing agent is Pt(II), which is oxidised to
Catalysis by Pt(II) can drive the reaction even further: besides Pt(III) during the reaction (Scheme 3d) [34].
annulation and sulphur reduction, one finds a cleavage of the
C–C bond between the atoms bearing the OH and NH groups Scheme 3 also depicts paclitaxel (taxol, 11), an important anti-
(ring enlargement). The result is an isomerisation of 4a and 4b cancer drug, as there are some similarities (shown in red in
to form tricyclic compounds 7 containing a nine-membered Scheme 3) but also differences to our compounds obtained by
carbocyclic ring (Scheme 3a) [27]. Isomerisations are the best Pt(II) catalysis from, e.g., 4a. The eye-catching dimethylmeth-
examples for a perfect “atom economy” [28-30] since all atoms ylene bridge over the largest carbocyclic ring is of course the
Scheme 1: Synthesis of 3-oxo-camphorsulfonylimine (3) [13,15] and its bis-alkynyl derivatives 4 from camphor-10-sulfonic acid (1).
1231

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
between our materials and taxol do, of course, not mean that
similar biological activity is necessarily involved, but it might
be worthwhile investigating.
Since the results of the Pt(II) catalysis depend on the nature of
the substituents at the alkyne groups, it would be of interest to
explore the course of the reaction when the two substituents are
different. In this article, we develop reasonable synthetic proce-
dures for the starting diynes, and present the first result of a cat-
alytic Pt(II) reaction of such mixed substituted compounds.
Results and Discussion
For the preparation of the diynes 4, a 2:1 ratio (or slightly larger
for complete reaction) of the lithium salt of a terminal alkyne
and of the oxoimide 3 is applied. The ratio should, however, not
be increased too much. For instance, with benzylacetylene, the
expected diyne 4e (Scheme 3d) is obtained, with only traces of
monosubstituted compounds. However, with a 3:1 ratio, the
main product is formed by reaction of only one equivalent of
lithium salt with the C=N double bond, leaving the carbonyl
group intact. In addition, the formed propargyl group is isomer-
ized to an allene moiety, obviously due to the excess of the
strongly basic lithium salt [35]. This unexpected monosubstitu-
tion remains unexplained and is not applicable as basis for a
general selective synthesis of monoalkynylcamphor derivatives.
Alkynes can also be cleaved from the bis-alkynyl compounds 4
by reaction with CuCl, but again, there are selectivity issues that
prevent a general application [36]. We therefore set out to
explore whether some selectivity is observed when lithium salts
of terminal alkynes are reacted with the oxoimide 3 in a 1:1
ratio (Table 1). As alkyne precursors, phenylacetylene,
1-heptyne, 1-ethynyladamantane, and 1-ethynyl-1-methoxycy-
clohexane were used. In all cases, mixtures of 12 and 13,
together with starting oxoimide 3 and the bis-alkynyl product 4
were obtained. The very slow addition of the acetylide and
dilute oxoimide solutions slightly improved the yields of the
mono-adducts 12 and 13, but the formation of the bis-adducts
could not be fully suppressed. With increasing bulkiness of the
Scheme 3: Reactions of bis-alkynylcamphor derivatives 4a–e with cat-
alytic amounts of PtCl2(PhCN)2.
most striking similarity, although this ring is in our compounds alkyne substituent (R = adamantyl, methoxycyclohexyl) the
one-carbon unit smaller (9 vs 10 members in taxol) and lacks reaction tends to become more selective towards mono-addi-
the oxygen substituent bearing the amino acid side group. In tion, but less selective with respect to the alkylation site, and the
both compounds, we find a keto group in the largest ring. 2-alkynyl and 3-alkynyl products are formed in similar
Instead of the six-membered ring annulation in taxol, there are amounts. Alkynes with small substituents (e.g., heptyne or
the substituents of the original triple bonds in our compounds, phenylacetylene) preferentially attack at the C=N bond of the
precisely in the same positions. In addition, where taxol has the sulfonylimine, suggesting that the carbon atom of the C=N is
bridgehead hydroxy and its neighbouring benzoate groups, we more electrophilic than that of the C=O bond. However, with
find the heterocyclic reduced isothiazole ring in the product of bulkier alkynes (e.g., 1-adamantylacetylene or 1-methoxy-1-
the Pt(II) catalysis. Other camphor derivatives prepared as ethynylcyclohexane), an attack at the carbonyl group becomes
entrance to taxoid compounds have a carbocyclic ring at this more pronounced. Presumably, the C=O carbon atom in 3-oxo-
place in the precursor [10] and a nine-membered ring with a camphorsulfonylimine is sterically more accessible, whereas the
keto group after oxidative bond cleavage [11]. The similarities sulfonylimine C=N carbon is more electrophilic. Steric and
1232

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
electronic properties thus counteract and the overall selectivity doublets spaced about 0.2 ppm apart in 13. Compounds 12
of the reaction depends on a balance between both factors.
display a singlet near 5 ppm for the SO2NH proton, whereas
spectra of 13 show a singlet for the OH group at about 3.2 ppm.
Similarly, the 13C NMR signals of the methyl groups nearly co-
incide in 13 but are approximately 2 ppm apart in 12. Com-
pounds 12 show a signal for the C=O near 206 ppm and one for
the sulfonamide carbon at about 65 ppm. In compounds 13,
there is a signal for the C=N around 194 ppm and one for the
tertiary alcohol carbon near 73 ppm. All other signals of the
camphor framework and the alkynyl substituent come at rela-
tively similar values and do not allow distinguishing between
12 and 13 reliably.
Table 1: Selectivity in the reaction of oxoimide 3 with alkynyllithium
compounds.
In view of the above results, it was clear that a more selective
method was needed that generally allows for mono-alkynyl-
ation and for differentiation between the carbonyl group and the
sulfonylimine. One way to achieve this could be in the chemo-
selective reduction of the C=N group of oxoimide 3, addition of
the alkynyl moiety to the carbonyl group, and re-oxidation of
the sultam to the sulfonimide (see Scheme 4). There are a num-
ber of reductions of oxoimide 3 reported in the literature, using
a variety of reducing agents. All of them lead either to com-
plete or to unselective reduction of the C=N and C=O bond.
Thus, LiAlH4 reduces the C=O and C=N group of 3, and 3-exo-
hydroxycamphorsultam is obtained in good yields [16]. Under
Meerwein–Ponndorf–Verley conditions (Al(OiPr)3/iPrOH), a
Entry
R
12:13
Yield 12 + 13
1
2
3
4
phenyl
80:20
90:10
70:30
50:50
42%
35%
88%
67%
n-pentyl
1-adamantyl
1-methoxycyclohexyl
Whilst chromatographic separation of the mixture of the mixture of the 3-hydroxyimine, 3-oxocamphorsultam and
monoadducts 12 + 13 from the starting material 3 and the bis- 3-exo-hydroxycamphorsultam is produced. Prolonged reaction
adduct 4 is straightforward, the isolation of pure 12 and 13 over several weeks eventually leads to 3-exo-hydroxycamphor-
requires careful control of the chromatography conditions sultam as the sole product.
(SiO2, CHCl3/diethyl ether gradient 0 to 10%), and has been
used on a small scale for analytical purposes only. The ratios
12:13 given in Table 1 were determined by integration of the
signals of the methyl groups in the 1H NMR. The ratios in the
crude product and after isolation of the mixture of 12 and 13 by
chromatography were found consistent.
In the IR spectra, compounds 12 show a typical NH stretching
vibration at approximately 3217 cm−1, and a C=O stretch at
1764 cm−1. Compounds 13, in contrast, can be recognised by
their OH stretching vibration at 3444 cm−1 and the C=N stretch
at 1653 cm−1. Both compounds display a C≡C stretch at very
similar wavenumbers around 2227 cm−1, and asymmetric and
symmetric SO2 stretching vibrations. These are found at about
1314 and 1143 cm−1 in compounds 12, and at slightly higher
Scheme 4: Attempted selective synthesis of 3-alkynyl derivatives via
sulfonylimine reduction of oxoimide 3.
wavenumbers of approx. 1330 and 1160 cm−1 in 13. In the
1H NMR spectra, the signals of the geminal methyl groups
come with a chemical shift difference of about 0.2 ppm in 12,
but almost coincide to form one signal in compounds 13. The We now found that the sulfonylimine of 3 could be reduced
opposite holds true for the diastereotopic protons of the selectively to the sultam 14 in the presence of Zn/HOAc, as
CH2SO2 moiety, which nearly coincide in 12, but come as two shown in Scheme 4. The product shows an IR stretching vibra-
1233

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
tion at 1760 cm−1 and a 13C NMR signal at 209.7 ppm for the
unreacted C=O group. The presence of the sultam can be
deduced from the NH stretching vibration at 3180 cm−1, a 13C
signal of the sulfonamide carbon at 65.0 ppm and a 1H NMR
signal for the NH at 5.57 ppm. These data are very similar to
those of compound 12 where C=O and sulfonamide coexist. In
contrast to 12, the 1H NMR signal of the NH in 14 comes as a
doublet, due to coupling with the adjacent CH proton at
3.58 ppm. The latter proton also appears as a doublet. The intro-
duction of an alkynyl substituent into the 3-position does not
pose any problem and can be achieved by reaction with
PhC≡CLi under standard conditions, to produce 15 in high
yield. Two equivalents of the acetylide are required because the
first one deprotonates the nitrogen of the sultam before the
second equivalent undergoes the desired nucleophilic addition
to the carbonyl group. All attempts to re-oxidise the sultam 15
to the sulfonimide 13, using Cl2/pyridine [37] or N-tert-
butylphenylsulfinimidoyl chloride/DBU [38] under literature
conditions, were unsuccessful due to the sensitivity of the
alkyne to oxidising conditions.
Scheme 5: Selective synthesis of 2-alkynyl derivatives by protection of
the 3-oxo group as an acetal.
Our next strategy was to modify the carbonyl group in the 3-po-
sition and introduce the alkyne at the sulfonamide side, as
shown in Scheme 5 and Scheme 6. The carbonyl group in 3 can
be selectively protected as an acetal by reaction with orthofor-
mates at room temperature in the presence of an acid. Thus, 3,3-
dimethoxycamphersulfonylimine 16 and 3,3-diethoxycampher-
sulfonylimine 16’ were prepared [18,39]. The subsequent reac-
tion of these acetals with one equivalent of lithium phenyl-
acetylide or 1-heptynyllithium under conditions described
above for the synthesis of 12 and 13 introduced the alkyne into
the 2-position of the camphor skeleton, to provide the sultams
17. The removal of the acetal-protecting group occurred under
comparatively mild conditions by stirring a mixture of 17 with
acetone and conc. HCl. Acetone as solvent was best as it allows
for trans-acetalisation to acetone dimethylacetal and 12 rather
Scheme 6: Selective synthesis of 2-alkynyl derivatives by protection of
the 3-oxo group as an imine.
than hydrolysis to methanol and 12. Overall, yields are higher As an alternative to the introduction of an acetal, an imine was
throughout when the methyl acetal is used. To obtain the tested for its suitability as a protecting group for the carbonyl
camphor-derived hydroxysultam 18 bearing an alkyne substitu- moiety, as shown in Scheme 6. 3-Oxocamphorsulfonylimine 3
ent in 2-position rather than in 3-position, 12 was reduced with was converted into the imine 19 by reaction with 2-phenylethyl-
NaBH4 under standard conditions. Compound 18 was prepared amine in the presence of TiCl4 [40]. The reaction of 19 with one
for comparison of the analytical data with that of compound 15. equivalent of the lithium phenylacetylide under the conditions
Although the IR spectra look fairly similar, the two isomers can described above for the synthesis of 12 and 13 provided the
be distinguished from their NMR spectra, in particular the 1H sultam 20 selectively, with exclusive introduction of the alkyne
and 13C signals of the atoms in 2 and 3 position. Thus, the into the 2-position at the camphor skeleton [41]. When two
proton in the 3-position in 18 comes at higher chemical shift equivalents of the acetylide are used, the reaction still produces
than the one in the 2-position in 15 (4.31 ppm vs 3.67 ppm). 20 only, and no reaction at the 3-position in the imine was ob-
Likewise, the signals of carbons 2 and 3 are further downfield, served. The high selectivity of this reaction can be explained by
when hydrogen is attached to them. In addition, carbon 4 is the large difference between the electron-deficient sulfonyl-
affected and appears at higher chemical shift in compound 15, imine and the electron-rich imine. The sulfonylimine carbon
where the alkynyl substituent is nearby.
atom is significantly more electrophilic and thus more prone to
1234

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
attack by the acetylide. In the case of 3-oxocamphorsul- plete within 30 min at 80 °C, without impairing the selectivity.
fonylimine, this difference between the reactive functional The structure of 22a is analogous to the one observed previ-
groups (C=O vs C=N-SO2R) was much less pronounced, ously with the bis-phenylalkynyl compound [27] as a starting
leading to poor selectivity. The general 1H and 13C signal material. Cyclisation of the alkynes and a three-carbon ring
pattern of the camphor skeleton in compound 20 is fairly simi- enlargement lead in a single step to a rare bicyclic carbon
lar to the one found for compounds 12, indicating that the framework that bears some similarity to that of the anticancer
alkyne has indeed been introduced at the sulfonamide side. Also drug paclitaxel. Remarkably, the sulfonamide group was
the 13C signal at 175.8 ppm shows clearly the presence of the reduced to a sulfinamide in the course of the reaction, and one
imino group in 3-position. For a sulfonimide, a signal at higher of the former alkynyl carbons was oxidised to a ketone. The
ppm values (approx. 190 ppm) would have been expected. The presence of the sulfinamide can be deduced from the fragmenta-
removal of the imine-protecting group turned out to be some- tion pattern in the mass spectrum where the loss of SO can be
what difficult and could be performed only under relatively seen which is further supported by the IR spectrum. Only one
harsh acidic conditions using aqueous HCl under reflux to S=O vibration can be identified at 1098 cm−1, in a similar posi-
provide 12 in moderate yield. Overall, the protection of the car- tion as the S=O vibrations in DMSO (1050 cm−1) or tert-butyl-
bonyl group as an acetal appears more convenient than as an sulfinamide (1032 cm−1) [43]. The strong band at about
imine.
1330 cm−1, typical for the asymmetric S=O stretch in sulfon-
amides and sulfones, is absent in 22a. The IR spectrum also
Starting from the 2-alkynyl-3-oxo compound 12a, the mixed shows the presence of two carbonyl groups at 1680 and
bis-alkynyl compound 21a was prepared by reaction with 1611 cm−1, and these are further confirmed in the 13C NMR
1-heptynyllithium, and accordingly, 21b was obtained from spectrum (signals at 198.3 and 210.8 ppm). The NH of the sulfi-
reaction of 12b with PhC≡CLi (Scheme 7) [41,42]. Two equiv- namide seems to be strongly involved in hydrogen bonding with
alents of the alkynyllithium compound are necessary because the carbonyl group nearby, as evidenced from the NH stretching
the first one is required for the deprotonation of the relatively vibration in the IR at 3110 cm−1 and the 1H NMR signal at an
acidic proton at the sulfonamide nitrogen.
unusually high chemical shift of 11.97 ppm. The ring-expanded
carbon skeleton has been corroborated from 1H/13C NMR
Compounds 21a and 21b were then reacted with 5 mol % together with 2D experiments, which allowed for a complete
PtCl2(PhCN)2 at 60 °C in CHCl3 [42]. For 1H NMR monitor- and unequivocal assignment of all signals.
ing, the same conditions were applied, but CDCl3 was used as
solvent. Within 10 hours, compound 21a converts cleanly into Upon reaction with catalytic amounts of PtCl2(PhCN)2, com-
product 22a without the observation of intermediates or side pound 21b converts into two products, apparently in a parallel
products. Under microwave irradiation, the reaction is com- reaction, and these were separated by chromatography. One
product is the expected 22b, whose structure is analogous to
22a described above. The other one, 23, has undergone alkyne
cyclisation but the ring expansion has not yet taken place. The
sulfonamide reduction and formation of the ketone are just
about to occur, as the oxygen atom involved is on its way of
being transferred from the sulphur onto the carbon atom.
The structure of 23 was established by two-dimensional
1H/13C NMR experiments, and by comparison with relatively
similar compounds obtained in the reaction of the bis(phenyl-
alkynyl)-derivative of 4a with triflic acid [26], or with K[ReO4]
[27]. The polycyclic carbon skeleton in these is identical, but
the latter compounds are protonated at the sulfonimide nitrogen
and form the triflate or perrhenate salts. The newly formed five-
membered ring in 23 is evident from the 13C NMR signals at
84.0, 90.0, 138.5, 136.0 and 125.5 ppm, and a 1H signal at
5.80 ppm. The enolate carbon is detected at 151.4 ppm in
13C NMR. Neither NMR nor IR show any evidence for carbon-
Scheme 7: Synthesis of the bis-alkynyl derivatives bearing different
alkyne substituents and their platinum-catalysed cycloisomerisation.
yl groups, but there is a medium intense C=C stretching vibra-
Compounds 21a and 22a: R1 = Ph, R2 = pentyl; compounds 21b, 22b
and 23: R1 = pentyl, R2 = Ph.
tion at 1653 cm−1 and two strong bands at 1324 cm−1 and
1059 cm−1 for the asymmetric and symmetric stretches of the
1235

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
O=S=N moiety. Compared to the SO2 moiety in sulfonamides turned out to be most efficient for the protection of carbonyl
(e.g., 21b with 1330 cm−1 and 1128 cm−1) the symmetric groups, leading to pure dialkynes with a well-defined substitu-
stretch is at unusually low wavenumbers.
tion pattern. The reactivity of a pair of isomers containing a
phenyl and a pentyl group attached to the triple bonds towards
The formation of products 22 and 23 can be explained from the cycloisomerisation induced by Pt(II) catalysis was studied. The
proposed reaction mechanism, shown in Scheme 8. The attack expected annulation–sulphur reduction–ring enlargement
of the catalyst at the OH and possibly at the alkyne in the imme- cascade leading to a product resembling paclitaxel to some
diate neighbourhood could be inferred from the fact that the 1H extent was found in both cases. However, one of the isomers
signal of the OH of 21 broadens significantly upon addition of yielded a second product lacking the ring-enlargement step and
PtCl2(PhCN)2. Concomitantly, a small amount of uncoordi- containing an additional sulphur–oxygen–carbon linkage. The
nated PhCN is observed in the 1H NMR spectrum. The coordi- platinum complex of this compound was postulated before as an
nated Pt centre, as a Lewis acid, promotes the release of the OH intermediate in the sulphur-reduction step of the cascade reac-
proton from A and its transfer onto one of the alkyne carbon tion, and the isolated product thus supports our mechanistic
atoms to form intermediate B. The resulting vinylic carbocation considerations. Platinum(II) catalysis applied to camphor-
undergoes an electrophilic attack at the neighbouring alkyne, derived dialkynes with two different substituents can thus not
which in turn reacts with the sulfonamide oxygen nearby. Both only give valuable insight in the mechanism of such cycloiso-
steps are strongly facilitated by the geometry changes when the merisations and help to clarify the role of the substituents, but
linear alkyne moieties convert into bent alkene ones, as the also yields a novel type of taxoid compounds of complex poly-
reactive centre is literally pushed into the functional group it cyclic structures with potential biological effects.
next reacts with. The intermediate D can be regarded as the
common precursor for the parallel formation of the observed
Supporting Information
products, 23 by dissociation from the Pt catalyst and proton
migration from the sulfonamide to the alcoholate, and 22 by a
Supporting Information File 1
cascade of electron movements as indicated by the arrows in D.
Experimental procedures and copies of 1H and 13C NMR
spectra of compounds 12a, 12b, 13a, 17a, 17b, 18, 19, 20,
21a, 21b, 22a, 22b and 23.
Conclusion
Several methods for the synthesis of camphor-derived dialkynes
having two different alkynyl substituents in close vicinity to
each other and to a sulfonamide group were developed. Ketals
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-13-122-S1.pdf]
Scheme 8: Proposed mechanism of the platinum-catalysed cycloisomerisation.
1236

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
17.Meladinis, V.; Herrmann, R.; Steigelmann, O.; Müller, G.
Acknowledgements
Z. Naturforsch., B: J. Chem. Sci. 1989, 44, 1453–1458.
doi:10.1515/znb-1989-1122
The authors are grateful to the Dr. Ing. Leonhard Lorenz Foun-
dation, the Fond der Chemischen Industrie and the Fundação
para a Ciência e Tecnologia, Portugal (Projects UID/QUI/
00100/2013, RECI/QEQ-QIN/0189/2012) for financial support.
The BAYER AG (Leverkusen/Germany) is acknowledged for a
generous gift of 1-ethynyladamantane, and Anton Paar for the
provision of a Monowave 400 microwave reactor. We are par-
ticularly grateful for the experimental assistance given by
Adrian Mahamoodally (M.Sc. thesis, University of Surrey/UK,
18.Verfürth, U.; Herrmann, R. J. Chem. Soc., Perkin Trans. 1 1990,
2919–2928. doi:10.1039/P19900002919
19.Čudić, M.; Herrmann, R. Magn. Reson. Chem. 1993, 31, 461–467.
doi:10.1002/mrc.1260310509
20.Wagner, G.; Heiß, C.; Verfürth, U.; Herrmann, R.
Z. Naturforsch., B: J. Chem. Sci. 1996, 51b, 1655–1662.
doi:10.1515/znb-1996-1121
21.Kagawa, M. Chem. Pharm. Bull. 1959, 7, 306–315.
doi:10.1248/cpb.7.306
2004), Joseph Yalley (M.Sc. thesis, University of Surrey/UK, 22.Kagawa, M. Chem. Pharm. Bull. 1961, 9, 391–403.
doi:10.1248/cpb.9.391
2007) and Sarah Belsey (Nuffield Science Bursary, University
23.Kagawa, M. Chem. Pharm. Bull. 1959, 7, 751–756.
of Surrey/UK, 2008). Special thanks also to GSK Stevenage for
doi:10.1248/cpb.7.751
NMR spectroscopy support.
24.Trost, B. M.; Vincent, J. E. J. Am. Chem. Soc. 1980, 102, 5680–5683.
doi:10.1021/ja00537a049
References
1. Blaser, H.-U. Chem. Rev. 1992, 92, 935–952.
25.Mi, B.; Maleczka, R. E., Jr. Tetrahedron Lett. 1999, 40, 2053–2056.
doi:10.1016/S0040-4039(99)00163-X
26.Wagner, G.; Herrmann, R.; Schier, A. J. Chem. Soc., Perkin Trans. 1
1997, 701–707. doi:10.1039/a605472b
doi:10.1021/cr00013a009
2. Crane, E. A.; Gademann, K. Angew. Chem., Int. Ed. 2016, 55,
3882–3902. doi:10.1002/anie.201505863
27.Carvalho, M. F. N. N.; Pombeiro, A. J. L.; Wagner, G.; Pedersen, B.;
Herrmann, R. Z. Naturforsch., B: J. Chem. Sci. 1999, 54b, 725–733.
doi:10.1515/znb-1999-0604
3. Trost, B. M. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 5348–5355.
doi:10.1073/pnas.0306715101
28.Trost, B. M. Science 1991, 254, 1471–1477.
4. Diéguez, M.; Claver, C.; Pàmies, O. Eur. J. Org. Chem. 2007,
4621–4634. doi:10.1002/ejoc.200700082
doi:10.1126/science.1962206
29.Trost, B. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 259–281.
doi:10.1002/anie.199502591
5. Money, T. Nat. Prod. Rep. 1985, 2, 253–289.
doi:10.1039/np9850200253
30.Trost, B. M. Acc. Chem. Res. 2002, 35, 695–705.
doi:10.1021/ar010068z
6. Shokova, E. A.; Kim, J. K.; Kovalev, V. V. Russ. J. Org. Chem. 2016,
52, 459–488. doi:10.1134/S1070428016040011
7. Chelucci, G. Chem. Soc. Rev. 2006, 35, 1230–1243.
doi:10.1039/b604793a
31.Li, C.-J.; Trost, B. M. Proc. Natl. Acad. Sci. U. S. A. 2008, 105,
13197–13202. doi:10.1073/pnas.0804348105
32.Wagner, G.; Pedersen, B.; Herrmann, R. In Electronic Conference on
Trends in Organometallic Chemistry (ECTOC-3), June 30–July 31,
1997; Rzepa, H. S.; Leach, C., Eds.; Royal Society of Chemistry.
http://www.ch.ic.ac.uk/ectoc/ectoc-3/pub/042/
8. Holton, R. A.; Somoza, C.; Kim, H.-B.; Liang, F.; Biediger, R. J.;
Boatman, P. D.; Shindo, M.; Smith, C. C.; Kim, S.; Nadizadeh, H.;
Suzuki, Y.; Tao, C.; Vu, P.; Tang, S.; Zhang, P.; Murthi, K. K.;
Gentile, L. N.; Liu, J. H. ACS Symp. Ser. 1995, 583, 288–301.
doi:10.1021/bk-1995-0583.ch021
33.Lampropoulou, M.; Herrmann, R.; Wagner, G. Tetrahedron 2004, 60,
4635–4643. doi:10.1016/j.tet.2004.03.078
9. Paquette, L. A.; Zhao, M.; Montgomery, F.; Zeng, Q.; Wang, T. Z.;
Elmore, S.; Combrink, K.; Wang, H.-L.; Bailey, S.; Su, Z.
Pure Appl. Chem. 1998, 70, 1449–1457.
34.Carvalho, M. F. N. N.; Ferreira, A. S. D.; Herrmann, R.
Synth. Commun. 2013, 43, 2305–2313.
doi:10.1080/00397911.2012.702290
doi:10.1351/pac199870081449
35.Santos, A. M.; Carvalho, M. F. N. N.; Galvão, A.; Pombeiro, A. J. L.
Z. Naturforsch., B: J. Chem. Sci. 2002, 57b, 691–698.
36.Carvalho, M. F. N. N.; Fernandes, T. A.; Ferreira, A. S. D.; Alves, L. G.;
Herrmann, R. J. Organomet. Chem. 2008, 693, 2847–2855.
doi:10.1016/j.jorganchem.2008.06.004
10.Gaisina, I. N.; Tikhonov, O. V.; Selezneva, N. K.; Abutkov, A. V.;
Fatykhov, A. A.; Spirikhin, L. V.; Miftakhov, M. S. Russ. J. Org. Chem.
1999, 35, 995–999.
11.Gaisina, I. N.; Tikhonov, O. V.; Spirikhin, L. V.; Miftakhov, M. S.
Russ. J. Org. Chem. 1999, 35, 1240–1241.
37.Rassukana, Y. V.; Onys’ko, P. P.; Davydova, K. O.; Sinitsa, A. D.
Eur. J. Org. Chem. 2004, 3643–3649. doi:10.1002/ejoc.200400086
38.Matsuo, J.-i.; Kawana, A.; Fukuda, Y.; Mukaiyama, T. Chem. Lett.
2001, 30, 712–713. doi:10.1246/cl.2001.712
12.Oppolzer, W. Pure Appl. Chem. 1990, 62, 1241–1250.
doi:10.1351/pac199062071241
13.Glahsl, G.; Herrmann, R. J. Chem. Soc., Perkin Trans. 1 1988,
1753–1757. doi:10.1039/p19880001753
39.Chen, B.-C.; Murphy, C. K.; Kumar, A.; Reddy, R. T.; Clark, C.;
Zhou, P.; Lewis, B. M.; Gala, D.; Mergelsberg, I.; Scherer, D.;
Buckley, J.; DiBenedetto, D.; Davis, F. A. Org. Synth. 1996, 73,
159–173. doi:10.15227/orgsyn.073.0159
14.Towson, J. C.; Weismiller, M. C.; Lal, G. S.; Sheppard, A. C.;
Kumar, A.; Davis, F. A. Org. Synth. 1990, 69, 158.
doi:10.15227/orgsyn.069.0158
15.Belsey, S.; Danks, T. N.; Wagner, G. Synth. Commun. 2006, 36,
1019–1024. doi:10.1080/00397910500501557
16.Ketter, A.; Glahsl, G.; Herrmann, R. J. Chem. Res., Synop. 1990,
278–279.
40.Čudić, M. Campher-Derivate als chirale Reagentien in stereoselektiven
Reaktionen. Ph.D. Thesis, Technical University of Munich, Germany,
1996.
41.Mahamoodally, A. Synthesis of Taxol-like compounds from camphor
derivatives. M.Sc. Thesis, University of Surrey, U.K., 2004.
J. Chem. Res. M 1990, 2118–2156.
1237

Beilstein J. Org. Chem. 2017, 13, 1230–1238.
42.Yalley, J. Synthesis of Taxol-like derivatives from camphorsulfonic
acid. M.Sc. Thesis, University of Surrey, U.K., 2007.
43.SDBS Spectral Database for Organic Compounds, National Institute of
Advanced Industrial Science and Technology. http://sdbs.db.aist.go.jp
(accessed March 6, 2017).
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.13.122
1238