A new strategy for the stereoselective synthesis of 2,2′-bipyrrolidines

Article abstract of DOI:10.1016/j.tetlet.2009.06.068

A new strategy for the stereoselective synthesis of the 2,2′-bipyrrolidine scaffold is presented using a metathesis reaction followed by asymmetric dihydroxylation for the introduction of the stereogenic elements. This straightforward high-yielding process is suitable for application to the synthesis of additional heterocycles.

Full text of DOI:10.1016/j.tetlet.2009.06.068

Tetrahedron Letters 50 (2009) 4899–4902
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
A new strategy for the stereoselective synthesis of 2,20-bipyrrolidines
*
Mary J. Gresser, Paul A. Keller , Steven M. Wales
School of Chemistry, University of Wollongong, Wollongong, New South Wales 2522, Australia
a r t i c l e i n f o
a b s t r a c t
A new strategy for the stereoselective synthesis of the 2,20-bipyrrolidine scaffold is presented using a
metathesis reaction followed by asymmetric dihydroxylation for the introduction of the stereogenic ele-
ments. This straightforward high-yielding process is suitable for application to the synthesis of additional
heterocycles.
Article history:
Received 16 April 2009
Revised 18 May 2009
Accepted 12 June 2009
Available online 16 June 2009
Ó 2009 Elsevier Ltd. All rights reserved.
Dedicated to Professor John B. Bremner on
the occasion of his 66th birthday
Keywords:
Chiral bisamine
2,20-Bipyrrolidine
Metathesis
Asymmetric dihydroxylation
Chiral bisamine compounds are commonly used as ligands in
stereoselective metal-catalysed reactions, including allylic addi-
tions, reductions and asymmetric dihydroxylations.1 Our interest
in such compounds stems from the concept of invoking chiral dis-
crimination in reactions based on helix sense as a design principle
driven by the observation that all chiral molecules are also helical.
Ligand types of interest to us include bisphosphines, bisarsines, bis-
amines and helical ligands that possess a mixture of heteroatoms
with a view that a range of metal-ligand catalysed reactions could
be investigated in the future. Our general target structures are
encompassed by 1 where the helix is defined by the two stereogenic
atoms which are flanked by the metal co-ordinating heteroatoms,
thus forming the backbone ‘arc’ of helicity. The helical groove depth
and degree of twist could be modulated by a range of substituents
‘R0’, including the presence of fused rings. Importantly, any single
synthetic strategy towards these structures should not only be
highly stereoselective, but also be flexible to allow the incorporation
of a range of heteroatoms, for example, X = N or P or As atoms in a
controlled fashion. Further, the same strategy should allow articula-
tion to produce different ring sizes, fused-ring structures and a range
of substituted ligands stereoselectively, and in an efficient manner.
The starting point for our investigations was the stereoselective
synthesis of the known 2,20-bipyrrolidine and this is the subject of
the work reported here. The most common syntheses of the 2,20-
bipyrrolidine parent structure involve dimerisation of monomers
followed by resolution,2,3 but given our requirement for a single
stereoselective strategy to produce a range of derivatives, the lack
of convergency of this approach was deemed impractical. Our at-
tempts to reproduce the published 15-step chiral synthesis from tar-
taric acid4 proved extremely unreliable and poor yielding rendering
it of little use for our purposes. Therefore, we devised a new strategy
for the stereoselective synthesis of the 2,20-bipyrrolidine scaffold that
utilises simple chemistry and starts from cheap and readily available
materials. The key steps involved the formation of a symmetrical al-
kene using Grubbs’ metathesis chemistry and Sharpless asymmetric
dihydroxylation for the introduction of chirality—subsequent func-
tional group manipulations use established, high yielding chemis-
try.4 This Letter reports the synthesis of the protected 2,20-
bipyrrolidine 10 using this new strategy and is outlined in Scheme 1.
Pent-4-en-1-ol was protected and then subjected to metathesis
using standard Grubbs’ I reaction conditions5 giving the dimer 2 in
an optimised 75% yield with an E:Z ratio of 5:1 as determined by
1H NMR analysis of the methylene peaks adjacent to the olefin
(Scheme 1).6 A small quantity (5%) of a by-product was also formed
arising from reaction with the styrene generated from the catalyst.
Asymmetric dihydroxylation utilised Sharpless conditions7 with
R'
R'
* *
X
R
X
R
1
ADmix
a and a mixture of the inseparable E:Z isomers of 2 yielding
the diol 3 as a mixture of chiral and meso diastereomers in overall
86% yield and 86% ee as determined by chiral HPLC. Similarly, the
* Corresponding author. Tel.: +61 2 4221 4692; fax: +61 2 4221 4287.
E-mail address: keller@uow.edu.au (P.A. Keller).
ee values are determined based upon conversion of the trans-alkene.
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.06.068

4900
M. J. Gresser et al. / Tetrahedron Letters 50 (2009) 4899–4902
ii
iii
BnO
viii R1
x
2
R2
2
R1
R1 = OH
BnO
N
N
2
NHBoc
Boc Boc
2
i
10
R2 = OH
8 R1 = OH
(R,R)-
R1 = OBn
+
3
4
5
6
7
iv
v
ix
R2 = OMs
R2 = N3
R1 = OMs
9
BnO
Ph
5%
vi
vii
R2 = NH2
R2 = NHBoc
Scheme 1. Reagents and conditions: (i) (a) NaH, THF, 0 °C, 30 min, (b) BnBr, 0 °C to rt, 23 h, 99%; (ii) 5 mol % Grubbs’ 1, CH2Cl2,
D
, 5 h, 75% (E:Z, 5:1); (iii) ADmix a, methane
sulfonamide, 1:1 tBuOH/H2O, 24 h, 86%; (iv) MsCl, Et3N, CH2Cl2, 20 min, 76%; (v) NaN3, DMF, 80 °C, 2 h, 71%; (vi) H2(g), Pd/C, EtOH, 2 h, 100%; (vii) Boc2O, Et3N, Et2O, 8 h, 59%;
(viii) H2(g), Pd/C, EtOH, 1.5 h, 85%; (ix) MsCl, Et3N, CH2Cl2, 20 min, 86%; (x) NaH, DMF, 2 h, 62%.
use of ADmix b resulted in an 80% yield of the alternative diol 3 with an
ee of 81%. Attempts to separate the chiral and meso isomers using col-
umn chromatography were unsuccessful. The remainder of the syn-
thesis to the bipyrrolidine utilised reported chemistry.4 While these
initial investigations established the viability of our strategy, the ste-
reochemical outcome required improvement. The problem arose from
the mixture of meso and chiral diols 3 being generated as a result of the
inability to separate the cis and trans isomers of 1 from the metathesis
reaction—this was largely due to the non-polar nature of the alkene
making simple separation using silica gel problematic. Therefore, a
more polar protecting group was chosen with the idea that silica gel
separation might be feasible and the results of this sequence synthes-
ising both the (R,R)- and (S,S)-bipyrrolidine 10 are summarised in
Scheme 2.
The asymmetric dihydroxylation of the alkene 12 was under-
taken with ADmix to give the (S,S)-diol 13 in a yield of 69% with
an ee of 98%. The yield can be improved (90%) using longer reaction
a
times but at the expense of stereochemical integrity (92%). In an
12
alkene
i then ii
O
S
O
O
PMBO
OPMB
Protection of pent-4-en-1-ol with PMBBr followed by metathe-
sis gave the alkene 12 in an optimised 86% yield with an E:Z ratio of
3.5:1. Again, the isomers could not be separated using silica gel col-
umn chromatography. Attempts to optimise the formation of the
E-isomer included the use of Grubbs’ II catalyst which resulted in
a mixture of products that were mostly non-symmetrical alkenes.
The use of 10 mol % benzoquinone8 in combination with Grubbs’ II
catalyst resulted in a poorer alkene yield (60%) with an E:Z ratio of
4:1, and additional impurities that could not be separated by col-
umn chromatography.
18 + diastereomer
iii then iv
13
diol
Scheme 3. Separation of stereoisomers via the corresponding sulfite. Reagents and
conditions: (i) ADmix
or ADmix b, methane sulfonamide, tBuOH/H2O (1:1), 17 h;
(ii) SOCl2, Et3N, CH2Cl2, 72 h; (iii) chromatographic separation, (R,R)-18 58%, (S,S)-18
62% (yields are based on the two step conversion from the trans alkene 12); (iv)
NaOMe, MeOH, 116 h, (R,R)-13 77%, (S,S)-13 69%.
a
R1
R1 =OH
11 R1 =OPMB
ii
i
PMBO
2
12
iii
iv
(S,S)-13 R2 = OH
(R,R)-13
v
(S,S)-
(R,R)-
(R,R)-
R2 = OMs
14
15
16
(R,R)-14
vi
PMBO
R2 = N3
R2 = NH2
PMBO
ix
2
R2
(S,S)-15
(S,S)-16
(S,S)-17
2
R2
vii
viii
(R,R)-17 R2 = NHBoc
ix
R1
R1
R1 = OH
R1 = OMs
(S,S)-8
(S,S)-9
(R,R)-8
(R,R)-9
2
2
x
NHBoc
NHBoc
xi
xi
N
N
N
N
Boc Boc
(S,S)-10
Boc Boc
(R,R)-10
Scheme 2. Reagents and conditions: (i) (a) NaH, THF, 0 °C. (b) PMBBr, 0 °C to rt, 4 h, 73%; (ii) 2.5 mol % Grubbs’ 1, CH2Cl2,
D, 20 h, 86%, E:Z, 3.5:1; (iii) ADmix a, methane
sulfonamide, tBuOH:H2O (1:1), 24 h (69%, 98% ee) or 20 h (90%, 92% ee); (iv) ADmix b, methane sulfonamide, tBuOH:H2O (1:1), 65 h (98%, 88% ee) or 24 h (93%, 70% ee); (v)
MsCl, Et3N, CH2Cl2; (S,S)-14 25 min, 99%; (R,R)-14 20 min, 95%; (vi) NaN3, DMF, 80 °C, (R,R)-15 23 h, 73%; (S,S)-15 18 h, 76%; (vii) H2(g), Pd/C, abs. EtOH, 2 h, (R,R)-16 6 h, 100%;
(S,S)-16 3.5 h, 100%; (viii) Boc2O, Et3N, Et2O, 35 °C, (R,R)-17 17 h, 40%; (S,S)-17 15 h, 75%; (ix) H2(g), Pd/C, abs. EtOH, (R,R)-8 1.5 h, 85%; (S,S)-8 5 h, 95%; (x) MsCl, Et3N, CH2Cl2,
(R,R)-9 20 min, 86%; (S,S)-9 10 min, 76%; (xi) NaH, DMF, (R,R)-10 2 h, 62%; (S,S)-10 3 h, 76%.

M. J. Gresser et al. / Tetrahedron Letters 50 (2009) 4899–4902
4901
H
i
BnO
v
13
(diol)
OH
O
BnO
2
BnO
18
ii
R
3
BnO
19 R=OMs
iii
iv
20
R=Br
21 R=PPh3Br
Scheme 4. Synthesis of alkene 3 using the Wittig chemistry. Reagents and conditions: (i) DMP, CH2Cl2, 86%; (ii) MsCl, Et3N, CH2Cl2, 30 min; (iii) LiBr, THF, 72 h, 82% over two
steps; (iv) PPh3, CH3CN,
, 164 h then KOH(aq), THF, 3.5 h, 47% over two steps; (v) LiBr, PhLi, THF, ꢁ78 °C to rt, 30 min, then 18, ꢁ78 °C to rt, 12 h, 30%, 97% E.
D
analogous fashion, the use of ADmix b yielded the (R,R) isomer in
98% yield with an ee of 88%. The chiral diols could not be separated
from the meso diol at this stage. Subsequent mesylation of both the
(S,S) and (R,R) isomers gave 14 (99% and 95% yields, respectively)
and allowed for the incorporation of the heteroatom in an SN2
reaction using sodium azide giving (R,R)-15 (73%) and (S,S)-15
(76%). Reduction of the azide derivatives using H2 gas and palla-
dium on carbon gave the amines (R,R)-16 (100%) and (S,S)-16
(100%) with no removal of the PMB protecting group evident.
Boc protection of the amines proceeded under standard conditions
giving (R,R)-17 (40%) and (S,S)-17 (75%). At this stage, separation of
the meso from the chiral molecule was possible by recrystallisation
from CH2Cl2/hexanes. Removal of the PMB protecting groups using
H2 gas and palladium on carbon proceeded smoothly [(R,R)-8 (85%)
and (S,S)-8 (95%)] and subsequent mesylation gave the penulti-
mate derivatives 9 in 86% (R,R) and 76% (S,S) yields. The ring-clos-
ing reaction proceeded by generation of the anion using NaH and
produced the target N-Boc-protected bipyrrolidines (R,R)-10 and
(S,S)-10 in 62% and 76% yields, respectively,9 with excellent stere-
reaction time and temperature using toluene at reflux favoured
the formation of the desired salt albeit in low yield (entry 2). The
reaction was also slow in a polar solvent (entry 3), however, an ex-
tended reaction time afforded the phosphonium salt 21 in good
yield with a small amount of the persistent by-product. In this
case, the benzyl salt was removed by selective hydrolysis with KO-
H(aq) in THF, providing the pure phosphonium salt in an overall
yield of 47% after hydrolysis.
Due to the difficulties encountered with the benzyl salt by-
product, we also investigated the reaction with iodine. Therefore,
iodination of 4-benzyloxy-1-butanol was carried out under stan-
dard conditions with PPh3, I2 and imidazole in 94% yield. Although
subsequent reaction with PPh3 was still slow (entry 4), prolonged
heating in toluene at 80 °C gave the phosphonium iodide salt in
74% yield with only a trace amount of the benzyl by-product de-
tected by NMR (entry 5).
The trans selective modification of the Wittig reaction12 using
PhLi as the base and the bromide salt 21 furnished the alkene 3
in high stereoselectivity (97% E), albeit in a modest yield of 30%
(Scheme 4). A secondary alcohol was also isolated as a by-prod-
uct (25%) arising from attack of PhLi on the unreacted aldehyde.
The same yield was also obtained with the iodide salt, although
the stereoselectivity was slightly lower (94% E). The poor yields
were a result of incomplete ylide generation presumably due to
the competing acidity of the benzylic protons of the protecting
group. The AD reaction was carried out using the trans alkene
(97% E) to afford the (S,S)-diol in 77% yield (ee 87%) and the
oselectivity [(S,S)-10 ½a D23
ꢀ
ꢁ43.7 (c 0.198, CHCl3), cf. lit.4 ½a D23
ꢁ40.6
ꢀ
(c 0.36, CHCl3)].
In an attempt to improve the stereochemical efficiency of the
sequence, we attempted to remove the meso isomer by converting
the product from the dihydroxylation of the alkene 12 into the cor-
responding sulfite, followed by chromatographic separation10 and
hydrolysis back to the diol (Scheme 3). This three-step process pro-
vided the (S,S)-diol 13 in 43% overall yield with an ee of 96% and
the (R,R)-diol 13 in 45% yield with an ee of 68%.
(R,R)-diol in 73% yield (ee 97%) using ADmix
a and ADmix b,
Although the diastereomeric separation of the chiral sulfites al-
lows for the successful completion of the synthesis with good ste-
reoselectivity, the sequence would clearly be more efficient if none
of the cis alkene was produced in the first instance. Thus, we inves-
tigated the generation of alkene 12 using modified Wittig chemis-
try. Protected 4-benzyloxy-1-butanol was converted into the
required aldehyde 18 using DMP in 86% yield (Scheme 4). The
use of PCC for the oxidation gave lower yields and side products
that were difficult to separate. The same alcohol was also con-
verted into the bromide 20 in 82% yield via mesylation and bro-
mination with LiBr. Synthesis of the phosphonium bromide salt
21 under the reported conditions11 of xylenes at reflux (Table 1,
entry 1) was sluggish and benzyltriphenylphosphonium bromide
was formed unexpectedly as the major product. Decreasing the
respectively.
Thus, we have demonstrated an efficient stereoselective synthe-
sis of 2,20-bipyrrolidine utilising cheap starting materials and
straightforward chemistry. The key steps involved ensuring the
integrity of the trans alkene to enable asymmetric dihydroxylation
to produce chiral diols in good enantiomeric excess. Alternatively,
any remaining meso isomers can be removed by recrystallisation
partway through the sequence. Subsequent functional group
manipulations have been shown by us, and others, to maintain
the chirality through to the 2,20-bipyrrolidines. The advantages of
this new synthetic route are the extremely simple and highly
reproducible reactions and the versatility of design which allows
for the incorporation of different heteroatoms giving rise to a range
of possible heterocycles using the identical strategy. Importantly, a
range of substituted alkenes could be used in the initial stages and
thus, this strategy could be used to give access to numerous differ-
ent symmetrical systems, and by the early use of cross-metathesis
reactions, it is possible to produce non-symmetrical heterocyclic
systems. We are currently investigating the versatility of our ster-
eoselective strategy through the synthesis of novel heterocycles.
Table 1
Optimisation of the generation of the the Wittig salt 21; X = halogen
Entry
X
Solvent
Time (h) Temperature
Yielda (%)
Butyl phos. Benzyl phos.
halide
halide
1
2
3
4
5
Br Xylenes
Br Toluene
Br Acetonitrile 164
100
6
Reflux
Reflux
Reflux
Reflux
80 °C
36
21
71
44
74
46
5
20
—
Acknowledgements
I
I
Benzene
Toluene
21
158
Financial support from UoW small grants schemes is gratefully
acknowledged. M.J.G. and S.M.W. thank the Australian Research
Council for PhD scholarships.
—
Relative yields determined by 1H NMR analysis of the salt mixture.
a

4902
M. J. Gresser et al. / Tetrahedron Letters 50 (2009) 4899–4902
6. A reported synthesis of this olefin by metathesis using the Schrock catalyst
References and notes
gave an 85% yield with an E:Z ratio of 3.7:1, see: Miller, A. L., II; Bowden, N. B.
Chem. Commun. 2007, 2051–2053.
7. Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. Rev. 1994, 94,
2483–2547.
8. Hong, S. H.; Sanders, D. P.; Lee, C. W.; Grubbs, R. H. J. Am. Chem. Soc. 2005, 127,
17160–17161.
9. In the double cyclisation to produce the bipyrrolidine 10, there was
never any evidence for the formation of the alternate fused decahydro-
1,5-naphthyridine structure. This is consistent with the previous reports—
see Ref. 4.
1. Selected general reviews include: (a) Lucet, D.; Le Gall, T.; Mioskowski, C.
Angew. Chem., Int. Ed. 1998, 37, 2580–2627; (b) Bennani, Y. L.; Hanessian, S.
Chem. Rev. 1997, 97, 3161–3196; (c) Noe, M. C.; Letavic, M. A.; Snow, S. L. Org.
React. 2005, 66, 109–625.
2. Denmark, S. E.; Fu, J.; Lawler, M. J.; Lee, S.; Huntsman, E.; Grabowski, E. J. J. Org.
Synth. 2006, 83, 121–130.
3. (a) Li, S.; Dieter, R. K. J. Org. Chem. 2003, 68, 969–973; (b) Dieter, R. K.; Li, S.;
Chen, N. J. Org. Chem. 2004, 69, 2867–2870; (c) Oishi, T.; Hirama, M.; Sita, L. R.;
Masamune, S. Synthesis 1991, 789–792.
4. Kotsuki, H.; Hiroko, K.; Gohda, T.; Fukuhara, M.; Ochi, M.; Oishi, T.; Hirama, M.;
Shiro, M. Tetrahedron: Asymmetry 1995, 6, 2227–2236.
10. (a) Pandey, S. K.; Kumar, P. Tetrahedron Lett. 2006, 47, 4167–4169; (b)
Fernandes, R. A.; Kumar, P. Eur. J. Org. Chem. 2002, 2921–2923.
11. Bianchi, D. A.; Rua, F.; Kaufman, T. S. Tetrahedron Lett. 2004, 45,
411–415.
5. For a review of Grubbs’ metathesis reactions, see: (a) Schuster, M.; Blechert, S.
Angew. Chem., Int. Ed. 1997, 36, 2036–2056; (b) Bremner, J. B.; Coates, J. A.;
Coghlan, D. R.; David, D. M.; Keller, P. A.; Pyne, S. G. New J. Chem. 2002, 26,
1549–1552.
12. Wang, Q.; Deredas, D.; Huynh, C.; Schlosser, M. Chem. Eur. J. 2003, 9,
570–574.