RuO4-catalyzed oxidative polycyclization of the Cs-symmetric isoprenoid polyene digeranyl. An unexpected stereochemical outcome

Article abstract of DOI:10.1016/j.tet.2008.09.052

The RuO₄-catalyzed oxidative polycyclization of digeranyl, a C_s-symmetric tetraene possessing a repetitive 1,5-diene structural motif, has been studied. The required substrate has been synthesized by Ti(III)-mediated tail-to-tail homocoupling of geranyl bromide. The process afforded two hitherto unknown isomeric tris-tetrahydrofuran products possessing unexpected all-threo cis-trans-cis and cis-trans-trans relative configuration. The new stereochemical outcome is explained based on previously formulated chelation or steric control models on the basis of structural differences between digeranyl and previously studied isoprenoid polyenes farnesyl acetate, geranylgeranyl acetate and squalene.

Full text of DOI:10.1016/j.tet.2008.09.052

Tetrahedron 64 (2008) 11185–11192
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
RuO4-catalyzed oxidative polycyclization of the Cs-symmetric isoprenoid polyene
digeranyl. An unexpected stereochemical outcome
,
b
b
Vincenzo Piccialli a , Nicola Borbone , Giorgia Oliviero
*
a
`
Dipartimento di Chimica Organica e Biochimica, Universita degli Studi di Napoli ‘Federico II’, Via Cynthia 4, 80126 Napoli, Italy
Dipartimento di Chimica delle Sostanze Naturali, Universita degli Studi di Napoli ‘Federico II’, Via D. Montesano 49, 80131 Napoli, Italy
b
`
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 15 July 2008
Received in revised form 4 September 2008
Accepted 18 September 2008
Available online 26 September 2008
The RuO4-catalyzed oxidative polycyclization of digeranyl, a Cs-symmetric tetraene possessing a re-
petitive 1,5-diene structural motif, has been studied. The required substrate has been synthesized by
Ti(III)-mediated tail-to-tail homocoupling of geranyl bromide. The process afforded two hitherto un-
known isomeric tris-tetrahydrofuran products possessing unexpected all-threo cis–trans–cis and cis–
trans–trans relative configuration. The new stereochemical outcome is explained based on previously
formulated chelation or steric control models on the basis of structural differences between digeranyl
and previously studied isoprenoid polyenes farnesyl acetate, geranylgeranyl acetate and squalene.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
RuO4
Oxidative polycyclization
Digeranyl
Cs-Symmetric tetraene
tris-THF diols
1. Introduction
polyene chain possesses a cis configuration, such as in (E,Z)-farnesyl
acetate,1b the process stops at the mono-THF level.
The ruthenium tetroxide mediated oxidative polycyclization
(OP) of polyenes characterized by a repetitive 1,5-diene structural
motif is a stereoselective process discovered a few years ago in our
laboratories (Scheme 1).1 It allows access to adjacently linked poly-
tetrahydrofuran products in a single step employing catalytic
amounts of ruthenium tetroxide in the presence of NaIO4 as co-
oxidant. Among the cases studied so far, the polycyclization of
squalene appears to be a remarkable process since a penta-THF
product (1) comprised of 10 chiral centres is obtained with a 50%
yield as a single diastereoisomer (Scheme 1), through what appears
to be a complex multistep cascade sequence.
Previous studies from our group allow a number of conclusions
to be drawn. (a) Poly-THF products derived from the oxidation of
all-trans-polyenes, all display a threo inter-THF stereorelationship
indicating a syn addition of oxygen pairs to all the involved double
bonds; (b) the first-formed THF along the poly-THF chain always
possesses a cis configuration, in agreement with the stereochemical
control operating in the strictly related oxidative mono-cyclization
of 1,5-dienes catalyzed by the same oxide;2 (c) the second and the
third THF rings possess cis and trans configurations, respectively, in
all the cases studied; (d) when the third double bond along the
Due to the still limited number of cases examined, many of
which concern the oxidation of all-trans, or predominantly trans,
isoprenoid polyenes, the factors governing the stereochemistry of
the process are not completely known and general stereochemical
rules, if any, cannot yet be formulated. In order to further amplify
our knowledge about this transformation we have now studied the
oxidative polycyclization of digeranyl (Scheme 2, 2), a Cs-symmetric
isoprenoid tetraene, synthesized by titanocene chloride-catalyzed
allylic homocoupling of geranyl bromide, according to a reported
procedure.3 In particular, we wanted to test the effect of the head-
to-tail fusion of the two central isoprene units in 2, never explored
in previous instances. As will be detailed below, we have found that
this structural feature profoundly affects the stereochemistry of the
process.
2. Results and discussion
Isolation of pure samples of 23,4 required careful chromatogra-
phy on silica gel impregnated with silver nitrate to separate the
required tetraene from the isomeric
obtained as the main side-product (20%) of the process (Scheme 2).
The
g0 coupling product 4 was also isolated in minute amount
a–
g0 tetraene (isogeranyl,5 3)
g–
(2%), and fully characterized for the first time. Compounds 2 and 3
are naturally occurring terpenes found in the commercially avail-
able bergamot oil.6
* Corresponding author. Tel.: þ39 081 674111; fax: þ39 081 674393.
E-mail address: vinpicci@unina.it (V. Piccialli).
0040-4020/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2008.09.052

11186
V. Piccialli et al. / Tetrahedron 64 (2008) 11185–11192
OAc
[from (E,E)- FA]
O
O
OH
H
H
HO
OAc
all-threo cis/cis
RuO4(cat.)/NaIO4
n
n=1 (E,E)-Farnesyl acetate [(E,E)]-FA
n=2 Geranylgeranyl acetate (GGA)
OAc
(from GGA)
O
O
O
OH
H
H
H
HO
all-threo cis/cis/trans
RuO4(cat.)/NaIO4
O
O
O
O
O
OH
H
H
H
H
H
H
HO
Squalene
all-threo cis/cis/trans/trans/trans
1
Scheme 1. Representative examples of the oxidative polycyclization of isoprenoid polyenes catalyzed by ruthenium tetroxide.
Our standard polycyclization conditions [RuO2$2H2O (20 mol %)
unequivocal evidence on the relative configuration of either tris-
THF 5 or 6. A new set of NMR data was, therefore, collected for these
compounds in pyridine-d5 where a good proton dispersion was
observed for all methyl and angular THF proton resonances in their
1H NMR spectra, allowing to definitively clarify all the doubtful 2D
NMR correlations.
In particular, the cis configuration of ring A in all three com-
pounds 5–7 was suggested by the presence of a strong NOESY
correlation between the angularly positioned H-3 and Me-18, as
usually observed in cis-THF rings belonging to poly-THF com-
pounds of the same type1,7 (Fig. 1 shows significant NOESY
correlations for 5). Ring B in 5 and 7 was characterized by a
trans-arrangement as primarily suggested by the lack of a NOE
correlation between the H-7 and H-10 protons1,7 in their NOESY
spectra. Further support came from the key NOESY correlation
between Me-1(or Me-17) and H-10. The latter correlation, together
with the stronger one between Me-18 and H-7 indicated that A and
B rings are almost orthogonally oriented one another possibly due
to a hydrogen bond between the C2-OH and the ring-A THF oxygen,
as models show.
as pre-catalyst, NaIO4 (6 equiv), EtOAc–CH3CN–H2O (3:3:1)]1a were
initially employed for the oxidation of 2. In particular, according to
our previous results,1a the presence of four olefinic bonds in 2 called
for six molar equivalents of periodate (assuming a 4 equiv amount
for a 1,5-diene subunit plus one more equiv.1a for every additional
double bond). While mass recovery was in line with that observed
in previous instances,1 the HPLC profile of the crude reaction
mixture exhibited a more complex pattern. Among the major
peaks, tris-THF compounds 5 and 6 (Scheme 3) were isolated in
a 14% overall yield (5/6 dr 3:2) along with bis-THF lactone 7 (15%)
whose presence suggested that the process stops in part to the
second ring-closing step.
Determination of the stereostructure of these compounds
proved not trivial and required both extensive 2D NMR experi-
ments and chemical work. A full set of 2D NMR experiments was
initially carried out in CDCl3 for all three compounds 5–7. However,
due to the high degree of superimposition of some 1H and 13C
resonances pertaining to the THF rings, that occurred even in the
two-dimensional experiments, the collected data did not provide
PBr3 (0.4 equiv.)
THF, 0 °C, 1.5 h
82%.
OH
Br
Geraniol
Cp2TiCl2 (0.2 equiv.)
Mn dust (8 equiv.),
70%
α
α'
Digeranyl (2)
16
15
2
14 15
α
12
1
2
4
3
5
10
14
13
18
8
10
γ'
6
12
11
16
4
6 7
8
γ
γ'
11
3
5
7
9
9
1
19
20
19
17
17
13
20
18
Isogeranyl (3, 20%)
4 (2%)
Minor side-products derived from the
α-γ' and γ-γ' coupling of geranyl bromide
Scheme 2. Synthesis of digeranyl and two of its isomers by titanocene chloride-mediated homocoupling of geranyl bromide.

V. Piccialli et al. / Tetrahedron 64 (2008) 11185–11192
11187
1
20
16
17
14
6
11
7 10
O
3
+
O
O
O
O
O
OH
H
H
H
OH
H
H
H
HO
HO
H
H
19
18
2
cis/trans/trans (5)
cis/trans/cis (6)
.
RuO2 2H O (20 mol%)
NaIO4 (62equiv.)
EtOAc-CH3CN-H2O (3:3:1)
O
+
O
O
O
(5/6 dr 3:2)
HO
H
H
H
cis/trans (7)
Scheme 3. Oxidative polycyclization of tetraene 2 with RuO2(cat.)/NaIO4.
bis-THF lactone 7, the C2-symmetric bis-lactone 10, previously
obtained by penta-THF 1,1c and unreacted 6 in a 2:1:3 ratio, as
judged by 1H NMR spectroscopic analysis and confirmed by HPLC
separation. This experiment definitively inter-related bis-THF lac-
tone 7 and tris-THF 6 confirming the trans configuration of the
central THF ring in the latter compound as well. The all-threo ar-
rangement in compounds 5–7 was assumed on the basis of all
stereochemical results previously obtained in this type of process1
and the closely related Re(VII)-mediated bis- and tris-oxidative
cyclization of bis-homoallylic alcohols.8
18
20
H3C
CH3
H
H
6
7
OH
14
C
A
H
B
CH3
O
3
O
16
O
17
10
11
H3C
H
CH3
1
CH3
19
OH
Figure 1. Spatial arrangement of 5 as derived from NOESY data (arrows).
The stereochemical outcome of the above process was un-
expected in view of the previously observed stereochemistry of the
OP of other isoprenoid polyenes (Scheme 1), all giving poly-THF
compounds as single stereoisomers (dr>95%), as well as consider-
ing the configuration of poly-THF products 5–7. In fact, contrary to
what happened in all the previously studied cases, where the first
two THF rings both displayed a cis configuration, all three com-
pounds 5–7 possess the second THF with an unprecedented trans
configuration. On the other hand, the third THF ring in 5 and 6 is
either trans or cis in contrast to poly-THF products obtained from
the OP of geranylgeranyl acetate and squalene (Scheme 1) where
the third THF was constantly trans.
Finally ring C in 5 was shown to possess a trans configuration as
well on the basis of the lack of NOE contact between angular Me-19
and H-14 and by the NOESY correlations observed between the
proton pairs H-14/H-12b and Me-19/H-12a. A NOESY correlation
between Me-19 and H-10 was also observed that settled the rela-
tive arrangement of rings B and C, as depicted in Figure 1. This
spatial arrangement also agrees with that obtained for 5 through
molecular mechanics.
The trans nature of the central THF ring in 5 was also corrobo-
rated by the good agreement of the chemical shift values of the
oxygen-carrying carbons (C-6, C-7, C-10, C-11) belonging to the
central region of this compound with those displayed by the cor-
responding carbons in the structurally related cis-threo–trans-
threo–trans-threo tris-THF 9 (Scheme 4), previously obtained by
bidirectional PCC-mediated oxidative degradation of penta-THF 1,1c
followed by LAH reduction of the resulting bis-lactone 8 and
acetylation.
However, NMR studies could not provide information on the
configuration of the central THF ring in 6 due to its C2-symmetry,
though the presence of a single bis-THF lactone (7) in the reaction
mixture suggested that also in 6 the first two THF rings possess
a cis–trans configuration since the former can be formed along the
polycyclization route eventually leading to tris-THF 5 and 6 (see
later for an explanation of the mechanistic relationship among
compounds 5–7). To confirm this inference compound 6 was sub-
jected to PCC oxidation in the same conditions previously
employed for penta-THF 11c (Scheme 5) to cleanly give a mixture of
The above results can be accounted for by the mechanistic path
shown in Scheme 6 for tetraene 2, that also takes into account our
previously formulated hypothesis1d,1e and the closely related oxi-
dative polycyclization of polyenic bis-homoallylic alcohols carried
out with rhenium (VII)-oxo species.7,8 In particular, a cascade of
consecutive ring-closing steps begins with the attack of RuO4 at
the terminal double bond to give ruthenium (VI) diester 11. Clo-
sure of the first THF ring proceeds with cis stereocontrol, as usu-
ally observed in the oxidative mono-cyclizations of 1,5-dienes
9
mediated by all three related oxo species RuO4,2 OsO4 and
MnO-4.10 Thus, a cis-threo mono-THF intermediate species (13) is
obtained through a [3þ2] cycloaddition of an O–Ru]O portion of
the intact ruthenium bis-glycolate 11 across the second C–C
double bond, with the molecule adopting the chair-like confor-
mation 12 in the transition state. Then, the metal, is oxidized at an
1. PCC/AcOH,
celite, CH2Cl2
O
O
O
O
O
O
O
O
O
O
O
O
OH
H
H
H
H
H
H
H
H
H
H
HO
1
8
1. LiAlH4
2. Ac2O/py
Carbon Tris-THF 5 Tris-THF 9
AcO
[ ]3
OH
OAc
[ ]3
83.33
84.88
85.12
83.96
83.23
84.68
85.09
84.05
6
11
6
7
6 7
11
10
10
7
O
O
O
O
O
O
H
cis
H
trans
H
H
HO
OH
H
cis
H
trans
H
H
HO
10
11
trans
trans
5
9
Scheme 4. Comparison of structure and selected 13C NMR data for tris-THF 5 and the closely related tris-THF diol 9.

11188
V. Piccialli et al. / Tetrahedron 64 (2008) 11185–11192
PCC (5.5 equiv.),
AcOH (70 equiv.),
CH2Cl2, 21h, r.t
O
+
O
O
O
O
O
O
O
O
O
O
O
H
H
trans
H
HO
H
trans
H
(6/7/10, 3:2:1)
H
H
H
H
OH
HO
cis
6
10
7
+
Unreacted 6
Steps (Ref. 1c)
O
O
O
O
O
OH
H
H
H
H
H
H
HO
1
Scheme 5. Bidirectional PCC oxidative degradation of tris-THF 6 and penta-THF diol 1, derived from the OP of squalene.
2
O
O
O
Ru
H
O
O
O
O
2
O
Ru
cis
Ru
O
O
O
O
O
Ruthenium (VI)
diester (11)
12
[3+2]
cycloaddition
(First cyclization
cis-selective)
11
NaIO4
Ru oxidation
[3+2] cycloaddition
11
2
O
O
H
H
(Second cyclization
trans-selective)
O
O
O
cis
O
O
cis
ORuLn
HO
Ru
H
H
H
Ru
O
trans
13
O
O
15
14
Scheme 6. Proposed mechanism for the first two cyclization steps of the Ru-catalyzed oxidative polycyclization of tetraene 2.
‘active’ oxidation state (intermediate 14) by NaIO4, thereby
allowing for the second cyclization step to take place via another
[3þ2] cycloaddition reaction once more involving a O–Ru]O and
the successive double bond in the carbon chain, to give the bis
cyclized intermediate 15.
control is operative (Scheme 7, 16). However, this appears not to be
the case for 2. We note that polyene 2, from a structural point of
view, possesses a methyl group bonded to the C-11 vinyl carbon and
a hydrogen at C-10, while in the previously studied polyenes,
shown in Scheme 1, the head-to-tail fusion of the third isoprene
From a stereochemical point of view, we have previously
hypothesized1d,1e that the observed cis selectivity of the second
cyclization step of polyenes such as FA, GGA or squalene (Scheme 1)
can be due to an arrangement of the molecule where the first-
formed THF ring (A) is coordinated to the metal, so that a chelation
H
Me
B
6
H
O
A
O
OH
O
Me
Me
Pseudoequatorial
2
2
H
Ln
O
9
Ru
O
O
H
A
Ru
H
6
O
Me
O
O
11
B
11
H
Methyls close
in the space
cis
H
trans
Non-chelated structure (17)
Chelated structure (16)
Closure of a
trans-THF
Closure of a
cis-THF
(Steric control)
(Chelation
control)
2
11
11
2
O
O
O
O
HO
ORuLn
ORuLn
HO
H
H
H
H
H
H
cis
trans
cis
cis
Figure 2. Computer-generated 3D structure showing the TS (16) for the second cycli-
zation step of 2. An octahedral arrangement of ruthenium is assumed. The Van der
Waals radii of H-9 and protons of Me-1, Me-11 are evidenced by white dots clouds.
Scheme 7. Chelation control versus steric control in the second cyclization step of the
Ru-catalyzed OP of 2.

V. Piccialli et al. / Tetrahedron 64 (2008) 11185–11192
11189
Methyls close
in the space
and C-11, when passing from GGA and squalene to 2, plays a crucial
role in determining the balance of trans/cis configuration of the
newly formed THF ring. In particular, referring to a cis-selective
step, arrangement 18 would be involved where the C-10 grouping is
pseudoaxially disposed. The situation for the OP of GGA and
squalene, where a methyl at C-10 is present, is analogous to that
seen for the second cyclization of 2 (Scheme 7) though, in this case,
the Me-15 (analogous to Me-11 in 16) does not pose steric problems
due to the trans configuration of the B-THF. Overall the approach of
the C–C double bond to the O–Ru]O portion appears to be
unfavourable in the TS for steric reasons and the alternative ar-
rangement 19, where the B-ring is not coordinated to the metal and
the C-10 grouping is pseudoequatorial, is favoured for these prod-
ucts, leading to a trans-THF through steric control. As far as 2 is
concerned, the presence of a hydrogen at C-10, in place of a methyl
group, renders arrangement 18, and the resulting closure of the cis-
THF, less unfavourable (Fig. 4 compares the computer-generated 3D
structures of the TS for the cis- and trans-selective third cyclization
steps for 2 and GGA/squalene). On the other hand, conformation 19
appears now to be less favourable than for the OP of GGA and
squalene due to the presence of Me-11 that is in close proximity
with the terminal Me at C-15 as well as with ruthenium (in Fig. 5
the vicinity in the space of Me-11 and Me-15 is shown).
The above reasoning can also explain the unprecedented for-
mation of bis-THF lactone 7. Overall, the third cyclization is prob-
lematic and the alternative hydrolysis of the ruthenium ester
intermediates (18/19) can favourably compete with the third cy-
clization step to give the bis-THF species 20 (Scheme 8). Then, the
independent oxidation of the remaining double bond, according to
the known reactivity of RuO4 (Scheme 8),11 would follow. In par-
ticular, the following steps can be envisaged to occur involving the
terminal olefin: (i) dihydroxylation; (ii) oxidative scission of the so-
formed diol system; (iii) hemiacetal formation involving the C11-OH
and (iv) the successive oxidation of the latter to a lactone. On the
other hand, formation of 7 from tris-THFs 5/6, successive to their
formation, appears less probable since no trace of the isomeric
trans–trans bis-THF lactone, in principle derivable from 5 or 6
through oxidative degradation of the trans- or cis-THF terminus,
respectively, has been noticed by an accurate HPLC separation of
the reaction mixture.
A-THF
A-THF
H (Me)
Pseudoaxial
Ln
Ru
O
B
10
O
H
O
Me
O
Me (H)
O
H
Me
11
B
O
Ru
Me (Chain)
Me (H)
15
O
11
C
O
C
10
(Me) H
13
cis
15
(Chain) Me
H
trans
H
Pseudoequatorial
19
18
Figure 3. Arrangements that help to explain a cis-selective (18) third ring-closing step
for tetraene 2 and a trans-selective (19) third ring-closing step for 2 and GGA/squalene.
For 2: C10-H/C11-Me/C15-Me; for squalene and GGA: C10-Me/C11-H/C15-chain.
unit places a methyl at C-10 and a hydrogen at C-11 in these
compounds. This different substitution pattern in 2 appears to be
responsible for the different stereochemical outcome of the second
cyclization step in this polyene that is now trans-selective. In par-
ticular, when referring to the transition state for the second cycli-
zation step, the C-11 Me group is in close proximity to Me-1
(Scheme 7, 16). This, in addition to the 1,3-diaxyal interaction be-
tween the C-6 grouping and the Hax-9, hampers the approach of the
vinyl unit to the O–Ru]O portion. Figure 2 shows a computer-
generated 3D structure of TS 16 where the vicinity in the space of
M-11 and Me-1 can be best appreciated. Therefore, the alternative
arrangement 17 (Scheme 7), where the C(2)O–Ru bond is broken
and the A-THF ring is not coordinated to the metal, and pseudoe-
quatorially disposed, appears to be now the favoured arrangement,
that leads to a trans-THF through steric control. According to this
model, we presume that the second cyclization step for the OP of 2
is under steric control.
Finally, we have to explain the stereochemistry of the third ring-
closing step. This cyclization step, different from the OP of GGA and
squalene, that typically give a trans-THF, is less stereoselective be-
cause either a cis-THF or a trans-THF is formed, though the com-
pound embodying the trans C-THF (5) is still slightly predominant.
The trans configuration of the second THF ring pushes the A-ring
away from the metal preventing its coordination (Fig. 3). However,
just as seen for the second cyclization step, B-THF ring can, in
principle, coordinate to ruthenium, in which case the cyclization
step would be cis-selective through a chelation control, or not co-
ordinate to it, leading to a trans-THF through steric control. Once
again, we believe that the exchange of the Me/H pair between C-10
Attempts to improve the yields of the tris-THF products were
carried out varying both the co-oxidant (6 to 8 equiv) and the
Figure 4. Computer-generated 3D structure showing the TS (18) for the cis-selective third cyclization step of 2 (left). On the right is shown the effect of a methyl at C-10 as in GGA
and squalene. An octahedral arrangement of ruthenium is assumed. Van der Waals radii of relevant protons are evidenced by white dots clouds.

11190
V. Piccialli et al. / Tetrahedron 64 (2008) 11185–11192
a similar way, Go˘ksel and Stark recently used a related mono-THF-g-
lactone to synthesize cis-solamin, a mono-THF acetogenin,14 while
Sinha et al. employed bis-THF lactones strictly similar to 7 to assemble
the bis-THF diol core of a complete library of bis-THF Annonaceous
acetogenins.8j
3. Conclusion
In conclusion, the oxidative polycyclization of tetraene 2 allowed
access to poly-THF products with unprecedented configuration. The
new stereochemical outcome here reported has shown for the first
time that the type of alkyl substitution of the polyene can strongly
affect, and even reverse, the stereochemistry of the RuO4-mediated
OP process of isoprenoid polyenes thus opening new opportunities
for the use of this transformation in the synthesis of stereochemi-
cally defined poly-THF compounds by suitably tuning the structure
of the starting polyene. The explanation given for the observed
stereocontrol of each ring-closing step in terms of chelation or steric
control is consistent with, and further supports, the previously for-
mulated hypotheses for this type of process. The hitherto unknown
all-threo tris-THF materials obtained are potentially able of cation
transport. Experiments to verify this will be carried out.
Figure 5. Computer-generated 3D structure showing the TS for the third trans-
selective cyclization step of 2. An octahedral arrangement of ruthenium is assumed.
Van der Waals radii of relevant protons are evidenced by white dots clouds.
4. Experimental
4.1. General methods
pre-catalyst (RuO2 10% to 30%) amounts, but no substantial increase
of the overall yield (lactone 7þtris-THFs 5/6) was observed.
Carrying out the process under dilute conditions was also in-
effective. On the other hand, when the process was conducted at
room temperature the two tris-THFs were isolated in a ca. 1:1 ratio.
All-erythro tris-THF diol compounds possessing the same consti-
tution as compounds 5 and 6 have previously been synthesized via
epoxide chemistry12 and have been shownto possesstransport ability
for physiologically important cations such as Naþ and Kþ. Poly-THF
All reagents and anhydrous solvents were purchased (Aldrich)
at the highest commercial quality and used without further puri-
fication. Where necessary, flame-dried and argon-charged glass-
ware was used. Oxygen free THF was obtained by bubbling Argon
into the solvent for 30 min. Reactions were monitored by thin-layer
chromatography carried out on precoated silica gel plates (Merck
60, F254, 0.25 mm thick). Merck silica gel (Kieselgel 40, particle size
0.063–0.200 mm) was used for column chromatography. Na2SO4
was used as drying agent in all extractive work-ups. HPLC separa-
tions were carried out on a Varian 2510 apparatus equipped with
a Waters R403 dual cell differential refractometer using phenom-
compounds possessing a terminal g-lactone moiety such as com-
pound 7 have synthetic value since this functionality can be further
elaborated. Recently, these substances have been elected as key in-
termediates in the synthesis of the poly-THF core of Annonaceous
acetogenins, a groupof plant-derived active metabolites. Forexample,
enex 250ꢀ10 mm and 250ꢀ4.6 mm (both 5
m) columns.
All NMR experiments were performed on a Varian Unity INOVA-
500 spectrometer equipped with a 5 mm inverse detection z-gra-
dient probe and Varian 300 spectrometer. The 1H, 13C and 2D NMR
spectra were measured at 25 ꢁC using CDCl3 and pyridine-d5 as
solvents. Proton chemical shifts were referenced to the residual
a mono-THF-g-lactone, synthesized through related permanganate
chemistry, has been used by Brown et al.13 as an intermediate in the
synthesis of membranacin, an adjacent bis-THF acetogenin. In
Bis-cyclization
and hydrolysis
O
O
HO
H
H
H
OH
2
20
RuO4
(Dihydroxylation)
RuO4
Oxidative scission
of the 1,2-diol
O
O
O
O
O
OH
HO
HO
OH
H
H
OH
OH
H
H
H
H
RuO4
1.Lactol
formation
2.Oxidation
to lactone
O
O
O
O
HO
H
H
trans
H
cis
7
Scheme 8. Mechanistic route to bis-THF lactone 7.

V. Piccialli et al. / Tetrahedron 64 (2008) 11185–11192
11191
CHCl3 (7.26 ppm) or pyridine signals (8.71, 7.56, 7.19 ppm); 13C NMR
4.2.3. Compound 4
chemical shifts were referenced to the solvent (CDCl3, 77.0 ppm;
pyridine-d5, 149.9, 135.5, 123.5 ppm). Abbreviations for signal
coupling are as follows: s, singlet; d, doublet; t, triplet; q, quartet;
m, multiplet, br, broad. J values are given in hertz. One-dimensional
1H and 13C NMR spectra were acquired under standard conditions.
The pulse programs of the gCOSY, gHSQC, TOCSY, gHMBC, NOESY
and HSQC-TOCSYexperiments were taken from the Varian software
library. The gHSQC experiments were optimized for a one-bond
heteronuclear coupling constant of 145 Hz. The gHMBC experi-
ments were optimized for long-range coupling constants of 8 Hz.
Two-dimensional TOCSY experiments were acquired with a mixing
time of 80 ms while NOESY spectra were acquired with mixing
times of 80, 200 and 350 ms.
IR spectra were collected on a Jasco FTIR-430 spectrometer. ESI
mass spectrometric analyses were recorded on a Waters Micromass
ZQ mass spectrometer equipped with an Electrospray source used
in the positive mode. HRESIMS spectra were recorded on a Bruker
APEX II FT-ICR mass spectrometer using electron spray ionization
(ESI) technique in positive mode. HREIMS spectra were recorded on
a Fisons VG Prospect mass spectrometer.
Oil. 1H NMR (300 MHz, CDCl3):
d
5.76 (2H, dd, J¼17.6, 11.0, H-17
and H-19), 5.10 (2H, dd, J¼11.0, 1.6, Ha-18 and Ha-20), 5.08 (2H, br t,
overlapped with the Ha-18/Ha-20 protons, J¼7.5, H-3 and H-10),
4.88 (2H, dd, J¼17.6, 1.6, Hb-18 and Hb-20), 1.67, 1.56 (6H each, s,
vinyl methyls), 0.93 (6H, s, CH3-14 and CH3-15). 13C NMR (75 MHz,
CDCl3):
d 16.3, 17.6, 23.37, 25.7, 34.7, 45.3, 113.8, 125.4, 130.9, 144.1.
HREIMS calcd for C20H34 274.2661, found 274.2675 (Mþ).
4.3. Oxidative cyclization of 2
To a solution of digeranyl (2, 156 mg, 0.569 mmol) in the bi-
phasic mixture EtOAc/CH3CN/H2O (3:3:1) (50 mL) were added in
sequence NaIO4 (6 equiv, 732 mg, 3.41 mmol) and RuO2$2H2O
(20 mol %, 15.2 mg, 0.11 mmol) under vigorous stirring at 0 ꢁC. After
20 min a saturated Na2S2O3$5H2O solution (1 mL) was added and,
after a further 10 min stirring, the mixture was extracted with
EtOAc (3ꢀ10 mL). The combined organic phase was dried and
evaporated to give 152 mg of an oily product. HPLC separation
(250ꢀ10 mm column, hexane–EtOAc, 55:45, flow 2.5 mL/min,
30 mg/injection) afforded tris-THF diol
6
(10.1 mg, 5%,
tR¼19.5 min), tris-THF diol 5 (18.2 mg, 9%, tR¼22.0 min) and bis-
4.2. Synthesis of tetraene 2
THF lactone 7 (26.6 mg, 15%, tR¼32.5 min) as oils.
To geraniol (2.15 g, 13.95 mmol) dissolved in anhydrous THF
4.3.1. Compound 5
(10 mL), phosphorous tribromide (0.4 equiv, 5.6 mmol, 530
mL) was
Oil. IR (neat) nmax 3420 (br) cmꢂ1
.
1H NMR (500 MHz, CDCl3,
added dropwise under stirring, at room temperature. When the
starting product disappeared (30 min, TLC monitoring, hexane–
EtOAc, 7:3), ice was added and stirring continued for further
a 10 min. The mixture was extracted with hexane (3ꢀ10 mL), dried
and taken to dryness in vacuo to give geranyl bromide (2.472 g,
82%) that was used in the next step without further purification.
Crude geranyl bromide (1.884 g, 8.65 mmol) was dissolved in an-
hydrous and strictly deoxygenated THF (20 mL), then Cp2TiCl
(431 mg, 0.2 equiv, 1.73 mmol) was added followed by portionwise
addition of Mn dust (3.80 g, 8 equiv, 69.2 mmol), under an argon
atmosphere. The red-wine solution turned to green within 2–
3 min. After 5 min, TLC analysis (hexane/EtOAc, 95:5) revealed the
disappearance of the starting allyl halide (Rf¼0.6) and formation of
a product at Rf¼0.8. The mixture was diluted with ether and 1 N HCl
was cautiously added (gas evolution) until effervescence ceased.
The mixture was extracted with t-BuOMe (3ꢀ10 mL), dried and
concentrated under reduced pressure to give an oily product
(1.215 g) that was filtered on a short pad of silica gel eluting with
hexane/t-BuOMe (4:1). 1H NMR analysis showed the crude reaction
attributions by 2D NMR): d 3.95–3.90 (1H, m, H-10), 3.90–3.86 (1H,
dd, J¼8.2, 4.0, H-3), 3.84–3.79 (1H, m, H-7), 3.73 (1H, dd, J¼9.2, 5.9,
H-14), 2.28 (1H, m, Ha-5), 1.60 (Hb-5), 2.18, 1.65 (H2-12), 2.10, 1.98
(H2-4), 1.97 (H2-9), 1.96 (H2-8), 1.85, 1.75 (H2-13), 1.24 (3H, s, Me-1),
1.18 (3H, s, Me-20), 1.12 (3H, s, Me-18), 1.11 (3H, s, Me-16), 1.10 (3H,
s, Me-19), 1.08 (3H, s, Me-17). 1H NMR (500 MHz, pyridine-d5, at-
tributions by 2D NMR):
d
4.04 (1H, dd, J¼8.4, 6.8, H-14), 4.02–3.98
(1H, m, H-10), 3.95 (1H, dd, J¼7.2, 6.2, H-3), 3.94–3.90 (1H, m, H-7),
2.25, 1.52 (H2-5), 2.23, 1.61 (H2-12), 2.10, 1.90 (H2-4), 2.00, 1.98 (H2-
13), 1.93, 1.86 (H2-9), 1.90, 1.88 (H2-8), 1.39 (3H, s, Me-1), 1.37 (3H, s,
Me-16), 1.33 (3H, s, Me-20), 1.25 (3H, s, Me-17), 1.13 (6H, s, Me-18,
Me-19). 13C NMR (125 MHz, CDCl3):
d 87.1 (CH-14), 86.1 (CH-3), 85.1
(CH-10), 84.9 (CH-7), 84.0 (C-11), 83.3 (C-6), 72.2 (C-2), 70.7 (C-15),
35.1 (CH2-5), 35.0 (CH2-12), 27.9 (CH3-1), 27.4 (CH3-20), 27.1 (CH2-
9), 26.8 (CH2-13), 25.8 (CH2-4), 25.2 (CH3-17), 24.6 (CH3-18), 23.9
(CH3-19), 23.7 (CH3-16). HRESIMS calcd for C20H36NaO5 379.2460,
found 379.2481 (MþNa)þ.
4.3.2. Compound 6
product to be composed by a mixture of
a
,a
0- and
a,
g0 coupling
Oil. IR (neat) nmax 3420 cmꢂ1
.
1H NMR (500 MHz, CDCl3, attri-
products in a 3.5:1 approximate ratio. The crude mixture (906 mg)
was further chromatographed on AgNO3 (20%)-impregnated silica
gel. Slow elution with hexane/EtOAc (98:2) gave pure 2 (581 mg,
butions by 2D NMR): d 3.93 (2H, m, H-7 and H-10), 3.84 (2H, dd,
J¼7.3, 5.8, H-3 and H-14), 2.20–2.13 (1H, m, Ha-5 and Ha-12), 1.62
(Hb-5 and Hb-12), 2.04, 1.92 (H2-4 and H2-13), 2.00, 1.91 (H2-8 and
H2-9), 1.24 (6H, s, Me-1 and Me-16), 1.13 (6H, s, Me-18 and Me-19),
1.07 (6H, s, Me-17 and Me-20). 1H NMR (500 MHz, pyridine-d5,
70%), the
amount of the
a,
g0 coupling isomer 3 (166 mg, 20%) as well as a minute
g,
g0 coupling isomer 4 (16.5 mg, 2%).
attributions by 2D NMR):
d
3.97 (2H, br t, J¼5.9, H-7 and H-10), 3.88
4.2.1. Compound 23,4
(2H, t, J¼7.0, H-3 and H-14), 2.19, 1.50 (H2-5 and H2-12), 2.06, 1.84
(H2-4 and H2-13), 1.83 (H2-8 and H2-9), 1.40 (6H, s, Me-1 and Me-
16), 1.12 (6H, s, Me-18 and Me-19), 1.21 (6H, s, Me-17 and Me-20).
Oil. 1H NMR (300 MHz, CDCl3):
d 5.19–5.05 (4H, m, olefinic
protons), 2.13–1.92 (12H, m), 1.68 (6H, s), 1.60 (12H, s). 13C NMR
(75 MHz, CDCl3):
d
15.9, 17.6, 25.6, 26.7, 28.1, 39.7, 124.2, 124.3, 131.1,
13C NMR (125 MHz, CDCl3):
d 86.1 (CH-3, CH-14), 85.5 (CH-7, CH-
134.9. HREIMS calcd for C20H34 274.2661, found 274.2650 (Mþ).
10), 83.8 (C-6, C-11), 71.8 (C-2, C-15), 35.4 (CH2-5, CH2-12), 28.3
(CH2-8, CH2-9), 28.1 (CH3-1, CH3-16), 26.3 (CH2-4, CH2-13), 25.3
(CH3-17, CH3-20), 23.6 (CH3-18, CH3-19). HRESIMS calcd for
C20H36NaO5 379.2460, found 379.2475 (MþNa)þ.
4.2.2. Compound 35
Oil. 1H NMR (300 MHz, CDCl3):
d
5.74 (1H, dd, J¼17.7, 11.0, H-19),
5.08 (3H two apparent br t, overlapping, both J¼6.0, H-3, H-7, H-
12), 4.98 (1H, dd, J¼11.0, 1.7, Ha-20), 4.90 (1H, dd, J¼17.7, 1.7, Hb-20),
2.12–1.92 (8H, m), 1.67 (6H, br s, 2ꢀvinyl methyls), 1.60 (6H, br s,
2ꢀvinyl methyls), 1.58 (3H, s, vinyl methyl), 1.35–1.25 (2H, m), 0.95
(3H, s, CH3-17). HREIMS calcd for C20H34 274.2661, found 274.2681
(Mþ).
4.3.3. Compound 7
Oil. IR (neat) nmax 3422 (br), 1766 cmꢂ1 1H NMR (500 MHz,
.
CDCl3, attributions by 2D NMR):
d
3.96 (1H, t, J¼7.7, H-10), 3.87 (1H,
dd, J¼8.0, 4.9, H-3), 3.84 (1H, dd, J¼9.7, 5.2, H-7), 2.82–2.68 (1H, m,
H2-13), 2.50 (1H, dd, J¼10.8, 3.3, H2-13), 2.44, 1.96 (2H, H2-12), 2.15,

11192
V. Piccialli et al. / Tetrahedron 64 (2008) 11185–11192
Caserta, T.; Gomez-Paloma, L.; Piccialli, V. Tetrahedron Lett. 2003, 44, 5499–
5503; (c) Caserta, T.; Piccialli, V.; Gomez-Paloma, L.; Bifulco, G. Tetrahedron
2005, 61, 927–939; (d) Piccialli, V.; Caserta, T.; Caruso, L.; Gomez-Paloma, L.;
Bifulco, G. Tetrahedron 2006, 62, 10989–11007; (e) Piccialli, V. Synthesis 2007, 17,
2585–2607.
2. (a) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. J. Org. Chem. 1981,
46, 3936–3938; (b) Piccialli, V.; Cavallo, N. Tetrahedron Lett. 2001, 42, 4695–
4699; (c) Albarella, L.; Musumeci, D.; Sica, D. Eur. J. Org. Chem. 2001, 5, 997–
1003; (d) Roth, S.; Go¨hler, S.; Cheng, H.; Stark, C. B. W. Eur. J. Org. Chem. 2005, 19,
4109–4118; (e) Go¨hler, S.; Cheng, H.; Stark, C. B. W. Org. Biomol. Chem. 2007, 5,
1605–1614; (f) Go¨hler, S.; Roth, S.; Cheng, H.; Go¨ksel, H.; Rupp, A.; Haustedt,
L. O.; Stark, C. B. W. Synthesis 2007, 17, 2751–2754.
1.63 (H2-5), 2.06, 1.96 (H2-4), 2.04 (H2-9), 1.97 (H2-8), 1.33 (3H, s,
Me-17),1.24 (3H, s, Me-1),1.10 (3H, s, Me-16),1.09 (3H, s, Me-15). 1H
NMR (500 MHz, pyridine-d5, attributions by 2D NMR):
d 3.92 (1H,
overlapping m, H-10), 3.91 (1H, overlapping m, H-3), 3.83–3.78 (1H,
m, H-7), 2.81 (1H, dt, J¼17.3, 9.4, H2-13), 2.47 (1H, ddd, J¼17.3, 10.6,
3.6, H2-13), 2.32 (1H, ddd, J¼12.4, 10.6, 3.6, H2-12), 2.06, 1.47 (H2-5),
2.03, 1.85 (H2-4), 1.83, 1.80 (H2-9), 1.81, 1.77 (H2-8), 1.78 (H2-12), 1.16
(3H, s, Me-17), 1.36 (3H, s, Me-1), 1.06 (3H, s, Me-16), 1.26 (3H, s,
Me-15). 13C NMR (125 MHz, CDCl3):
d 178.0 (C-14), 86.2 (C-11), 85.9
(CH-7), 85.8 (CH-3), 85.3 (CH-10), 83.1 (C-6), 72.2 (C-2), 35.3 (CH2-
5), 32.7 (CH2-12), 29.8 (CH2-13), 28.0 (CH3-1), 27.6 (CH2-8), 26.8
(CH2-9), 25.9 (CH2-4), 24.9 (CH3-15), 24.4 (CH3-16), 23.4 (CH3-17).
HRESIMS calcd for C17H28NaO5 335.1835, found 335.1839 (MþNa)þ.
3. (a) Barrero, A. F.; Herrador, M. M.; Quilez del Moral, J.-F.; Arteaga, P.; Arteaga, J.
F.; Piedra, M.; Sanchez, E. M. Org. Lett. 2005, 7, 2301–2304 and references
therein; (b) Barrero, A. F.; Quilez del Moral, J.-F.; Sanchez, E. M.; Arteaga, J. F.
Eur. J. Org. Chem. 2006, 7, 1627–1641.
4. Hoshino, T.; Kumai, Y.; Kudo, Y.; Nakano, S.; Ohashi, S. Org. Biomol. Chem. 2004,
2, 2650–2657.
5. (a) Baldwin, J. E.; Kelly, D. P. J. Chem. Soc., Chem. Commun. 1968, 899–900; (b)
Momose, D.-I.; Iguchi, k.; Toshikatzu, S.; Yamada, Y. Chem Pharm. Bull. 1984, 32,
1840–1853.
4.4. Bis-lactone 5 and monolactones 3 and 4
6. Soucek, H.; Herout, V.; Sorm, F.; Ceskoslov, A. Collect. Czech. Chem. Commun.
1961, 26, 2551–2556.
7. Morimoto, Y.; Iwai, T. J. Am. Chem. Soc. 1998, 120, 1633–1634 and references
therein.
To a solution of 6 (1.5 mg, 0.00421 mmol) in CH2Cl2 (0.5 mL)
were added Celite (25 mg), PCC (5.0 mg, 0.0231 mmol,) and AcOH
(70 mmol, 20 mL) and the resulting heterogeneous mixture was
8. (a) Kennedy, R. M.; Tang, S. Tetrahedron Lett. 1992, 33, 3729–3732; (b) Tang, S.;
Kennedy, R. M. Tetrahedron Lett. 1992, 33, 5299–5302; (c) Tang, S.; Kennedy,
R. M. Tetrahedron Lett. 1992, 33, 5303–5306; (d) Boyce, R. S.; Kennedy, R. M.
Tetrahedron Lett. 1994, 35, 5133–5136; (e) McDonald, F. E.; Towne, T. B. J. Org.
Chem. 1995, 60, 5750–5751; (f) Towne, T. B.; McDonald, F. E. J. Am. Chem. Soc.
1997, 119, 6022–6028; (g) Keinan, E.; Sinha, S. C. Pure Appl. Chem. 2002, 74, 93–
105; (h) Morimoto, Y.; Kinoshiya, T.; Toshiyuki, T. Chirality 2002, 14, 578–586;
(i) Sinha, S. C.; Keinan, E.; Sinha, S. C. J. Am. Chem. Soc. 1998, 120, 9076–9077; (j)
Das, S.; Li, L.-S.; Abraham, S.; Chen, Z.; Sinha, S. C. J. Org. Chem. 2005, 70, 5922–
5931 and references therein.
stirred at room temperature for 21 h and then diluted with CH2Cl2
and filtered through a short silica gel pad. Elution with CHCl3/
MeOH 9:1 (10 mL) afforded a colourless oil. The crude was parti-
tioned between CHCl3 (5 mL) and a saturated aqueous NaHCO3
solution (5 mL). The organic layer was washed with water, dried
and concentrated in vacuo to give 1.2 mg of an oily material that
was separated by HPLC (250ꢀ4.6 mm column, hexane–EtOAc, 1:1)
to give pure unreacted 6 (0.5 mg, tR¼7.0 min), monolactone 7
(0.4 mg, 30% respect to reacted 6, tR¼11.0 min) and bis-lactone 10
(0.2 mg, 18% respect to reacted 6, tR¼12.4 min).
´
9. (a) de Champdore, M.; Lasalvia, M.; Piccialli, V. Tetrahedron Lett. 1998, 39, 9781–
9784; (b) Donohoe, T. J.; Winter, J. J. G.; Helliwell, M.; Stemp, G. Tetrahedron Lett.
2001, 42, 971–974; (c) Donohoe, T. J.; Butterworth, S. Angew. Chem., Int. Ed.
2003, 42, 948–951.
10. (a) Klein, E.; Rojahn, W. Tetrahedron 1965, 21, 2353–2358; (b) Baldwin, J. E.;
Crossley, M. J.; Lehtonen, E.-M. M. J. Chem. Soc., Chem. Commun. 1979, 918–919;
(c) Walba, D. M.; Wand, M. D.; Wilkes, M. C. J. Am. Chem. Soc. 1979, 101, 4396–
4397; (d) Walba, D. M.; Edwards, P. D. Tetrahedron Lett. 1980, 21, 3531–3534; (e)
Spino, C.; Weiler, L. Tetrahedron Lett. 1987, 28, 731–734; (f) Walba, D. M.;
Przybyla, C. A.; Walker, C. B. J. J. Am. Chem. Soc. 1990, 112, 5624–5625; (g) Ko-
cienski, P. J.; Brown, R. C. D.; Pommier, A.; Procter, M.; Schmidt, B. J. Chem. Soc.,
Perkin Trans. 1998, 9–39; (h) Brown, R. C. D.; Hughes, R. M.; Keily, J.; Kenney, A.
Chem. Commun. 2000, 1735–1736; (i) Brown, R. C. D.; Keily, J. F. Angew. Chem.,
Int. Ed 2001, 40, 4496–4498.
Acknowledgements
We are grateful to MURST, Italy (PRIN 2003), for financial sup-
port in favour of this investigation and to the ‘Centro di Metodo-
logie Chimico-Fisiche dell’Universita di Napoli ‘Federico II’ ’ and
‘Centro Interdipartimentale di Analisi Strumentale (CSIAS)’ for MS
and NMR facilities.
`
11. Plietker, B. Synthesis 2005, 15, 2453–2472.
12. Morimoto, Y.; Iway, T.; Yoshimura, T.; Takamasa, K. Bioorg. Med. Chem. Lett.
1998, 8, 2005–2010.
References and notes
13. Head, G. D.; Whittingham, W. G.; Brown, R. C. D. Synlett 2004, 1437–1439.
14. Go˘ksel, H.; Stark, C. B. W. Org. Lett. 2006, 8, 3433–3436.
1. (a) Bifulco, G.; Caserta, T.; Gomez-Paloma, L.; Piccialli, V. Tetrahedron Lett. 2002,
43, 9265–9269; corrigendum Tetrahedron Lett. 2003, 44, 3429; (b) Bifulco, G.;