Factors influencing prenylation of an aromatic organolithium

Article abstract of DOI:10.3184/030823409X12491261540439

An apparently routine metal-halogen exchange (MHE) reaction gave variable low yields under typical conditions. ¹³C and ⁷Li NMR studies suggested a diversity of organolithiums were formed in the MHE reaction leading to a plethora of p

Full text of DOI:10.3184/030823409X12491261540439

468 RESEARCH PAPER
AUGUST, 468–472
JOURNAL OF CHEMICAL RESEARCH 2009
Factors influencing prenylation of an aromatic organolithium
Rosalyn Klein*, Suthananda N. Sunassee and Michael T. Davies-Coleman
Department of Chemistry, Rhodes University, Grahamstown 6140, South Africa
An apparently routine metal–halogen exchange (MHE) reaction gave variable low yields under typical conditions.
7
13C and Li NMR studies suggested a diversity of organolithiums were formed in the MHE reaction leading to a
plethora of products on subsequent electrophilic aromatic substitution with geranyl bromide. A reversal of reagent
addition simplified the 7Li NMR spectrum and the product distribution, in addition to increasing the isolated yield of
the desired product from a variable 5–40% to a consistent 65%.
Keywords: toluhydroquinones, metal–halogen exchange, 7Li NMR, organolithium
Ortho-polyprenylated toluquinones and toluhydroquinones,
e.g. 1 and 2, are an important class of bioactive secondary
metabolites regularly isolated from Southern African marine
invertebrates.1-3 The in vitro anti-proliferative activity of 2
against oesophageal cancer cells4 recently prompted us to
synthesise the related marine natural product analogues; 5-
methyl-2-[(2'E, 6'E)-3',7', 11'-trimethyl-2',6',10'-dodecatrienyl]-
2,5-cyclohexadiene-1,4-dione (3) and 5-methyl-2-[(2'E,6'E)-
3,7,11-trimethyl-2',6',10'-dodecatrienyl]-1,4-benzenediol (4).5
A number of strategies are available for the regioselective
synthesis of ortho-prenylated phenols including Claisen
rearrangements, directed ortho-metallation (DoM), metal-
mediated coupling and metal–halogen exchange (MHE).6,7
Of these, butyllithium-mediated MHE provided access to 5
(the methylated precursor of 3) from 6 and farnesyl bromide,
in relatively low and frustratingly variable isolated yield
(5–40%).5 Given the apparent vagaries of the prenylation of
bromodimethoxytoluene substrates using MHE methodology,
we present here our attempt to shed light on the factors that
may be influencing this reaction.
The n-BuLi-mediated MHE reaction between 6 and geranyl
bromide (9a), in the absence and presence of TMEDA,
7
was monitored by 13C and Li NMR spectroscopy. HPLC
analysis of the product mixture consistently suggested that
the same five components (Table 1 and Scheme 1) were
being formed in variable amounts in each experiment. It was
also noted that a reversal in the order in which the reactants
were added by Scheepers et al.5 reduced the number of side-
products and improved the yield of the desired prenylated
product. The possible rationale behind our observations is
discussed here.
Earlier studies of the structure of 2-lithioanisole indicated
that probable internal coordination between the oxygen of
the methoxy group and the lithium renders the anisole almost
unreactive to electrophiles.8 TMEDA and other strongly
coordinating solvent additives are therefore often used to
moderate aggregation and internal coordination as an aid to
enhance reactivity.9 This was borne out by the low ratios of
desired product 7 to unreacted geranyl bromide (4:5) obtained
in our small scale prenylation studies of 6 when TMEDA was
omitted (Table 1). Conversely, the addition of one equivalent
Table 1 Product ratios as determined from HPLC analysis of reaction mixtures (using a differential refractometer)a
Fraction
Compound
Without TMEDA
1 equiv. TMEDA
2 equiv. TMEDA
A
B
C
D
9a,b
7
7,10
6,11
5
1
6
1
6
1
1
4
0.6
0.6
1
0.8
aThere were also traces of unidentified products.
OH
O
O
O
2
1
OH
OH
O
O
4
3
OH
O
OMe
OMe
X
6
X = Br
5
8 X = Li
OMe
OMe
* Correspondent. E-mail: r.klein@ru.ac.za
PAPER: 09/0623

JOURNAL OF CHEMICAL RESEARCH 2009 469
9a Z = Br
9b Z = Bu
Z
OMe
OMe
Br
OMe
Isolated Yield:
42%
H3C
7
H3C
OMe
(1)TMEDA, ether
10
(2) n-BuLi, ether
Candidates for
OMe
10
unidentified coupled product
(3) geranyl bromide
10a
H3C
OMe
H
OMe
OMe
11 R = H H3C
OMe
10b
debrominated
starting material
H3C
OMe
Scheme 1 Prenylation of 6 (following method of Scheepers et al.).5
of TMEDA to the ether solution of 6 and n-BuLi resulted in
an increase in the proportion of 7 obtained with the proportion
of geranyl bromide dramatically reduced. We observed
that there was no added advantage gained by increasing the
amount of TMEDA from one to two equivalents (see Table 1),
which suggests that no more than one TMEDA is involved
in the rate-determining step.10 Frustratingly, the favourable
ratio of 7 suggested from HPLC analysis of the product
mixture obtained from the addition of TMEDA could not be
extrapolated to give a comparable isolated yield of 7 when the
reaction was scaled up. A closer investigation of the nuances
of this MHE reaction was thus required.
In order to gain a better understanding of the reaction and its
mechanism, we first attempted to identify the major by-product
(10) which appeared to be an alternative coupling product.
1
Mass spectrometric (m/z 288) and H NMR data (Fig. 2)
suggested that the unknown component 10 of the intractable
mixture obtained from semi-preparative HPLC (fraction C
from Table 1) is an isomer (10a or 10b) of the major product 7
(Scheme 1). The structures of the starting materials (6 and 9a)
and the by-products (9b and 11) in the reaction product
mixture were similarly determined from NMR data.
A combination of carbon and lithium NMR spectroscopy has
been shown to provide an insight into the mechanistic details
of lithium-mediated reactions.11 There was clear evidence from
an analysis of the 13C and 7Li NMR spectra and the structures
proposed for the products formed in this reaction that the MHE
had taken place. The signal corresponding to the brominated
carbon (dC 108) disappeared and two broad signals (dC 122
(a)
(b)
C
A
B
(c)
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
ppm
7.5
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
ppm
Fig. 1 7Li spectra (155.5 MHz, dLi 0–5.5, LiBr) of (a) n-BuLi
added to 6 (b) n-BuLi and TMEDA added to 6 and (c) 6 added
to n-BuLi and TMEDA.
Fig. 2 1H NMR (400 MHz) spectrum of fraction C obtained from
HPLC. Signals indicated by A, B and C refer to the aromatic,
benzylic methylene, and methyl signals respectively.
PAPER: 09/0623

470 JOURNAL OF CHEMICAL RESEARCH 2009
and 108) consistent with lithiated carbon were observed. The
chemical shift of the lithiated aromatic carbon signals normally
appear in the region 110–200 ppm, e.g. phenyllithium (dimer,
dC 188)12 and ortho-lithio-anisole (dC 115).13 The quadrupolar
broadening of the lithiated carbon resonances observed in
the 13C NMR spectra of the mélange of mixed aggregates is
typical of a system unsymmetrical around 7Li.14,15 In our case,
we also propose that competitive coordination between Li and
the ether, TMEDA and aromatic methoxy groups contributed
to quadrupolar broadening.16 Additional broadening of the
NMR signals could be attributed to solvent exchange processes
around 7Li.13 The signal broadening observed in the 13C NMR
spectrum was corroborated by similar broadening of the signals
in the 7Li NMR spectra (Fig. 1b).
The addition of geranyl bromide to the mixed aggregates
of 8 at 194 K resulted in the immediate appearance of a new
7
signal (dLi 2.7) in the Li NMR spectrum, accompanied by
the disappearance of at least one other smaller species and a
slight reduction in the major aggregate. As the temperature
was increased (at 20 minute intervals), the broad signal
at dLi 4.3 shifted upfield and then disappeared, while the
major aggregate (dLi 2.9) was completely consumed by the
time 222K was reached. It was not clear whether this was a
result of reaction of 8 with the geranyl bromide taking place
gradually over the intervening time period or as a result of the
temperature increase (Fig. 5).
During the course of this study we were still unable
to obtain 7 in reproducible and satisfactory yields. This
therefore prompted us to investigate the possible influence of
adventitious water despite our previous efforts to work under
strictly anhydrous conditions. In order that any traces of water
would react with excess butyllithium prior to the addition of
6 the order of reagents was reversed, i.e. n-BuLi was initially
combined with TMEDA in diethyl ether and cooled to –10°C,
after which a solution of 6 in ether was added dropwise.
Removal of any traces of water would consequently decrease
the loss of 8 due to protonation. The previous NMR studies
indicated that the MHE was rapid, so there would be very little
of 6 remaining in the presence of excess n-BuLi/TMEDA.
It was evident upon examination of the 7Li NMR spectrum,
acquired from the experiment in which the addition of
reagents had been reversed (Fig. 1c), that several of the lesser
aggregates had been eliminated, thus indicating that the effect
of the revised reagent addition method served to control the
number and nature of the organolithium species in solution
In the case of DoM in anisole, the reaction is perceived to
be a simple two step process: coordination of the butyllithium
dimer to the methoxy/electron-rich aromatic ring followed by
metal proton exchange.13, 17-19 The resulting complex includes
a mixed dimer of butyllithium and aryllithium.13, 20 Assuming
that the same process is followed for MHE in 6, then we
expected to see evidence of a mixed dimer product in the 13
C
7
and Li NMR spectra of the reaction mixture. Unfortunately,
despite several attempts, it was neither possible to conclusively
identify a mixed dimer, nor to establish the structure of any of
the aryllithium aggregates.
It is also significant that several signals were observed in both
the 13C NMR and 7Li NMR spectra, indicating the presence of
more than one aggregate (Fig. 1). In accordance with a general
medium effect, both the TMEDA (methyl and methylene 13C
resonances)andorganolithium(7Li)signalswereshifted(Fig.3).
7
The Li NMR spectrum of 8, acquired in the absence of
TMEDA, suggested three major aggregates (dLi 3.6, 3.5, 2.7,
Fig. 1a) and several minor aggregates. Interestingly, addition of
7
TMEDA simplified the Li NMR spectrum in which a single
new aggregate (dLi 2.9) was dominant (Fig. 1b).
257 K
In a dynamic NMR study of the TMEDA/organolithium
sample, the 7Li signals were observed to coalesce at or around
257 K suggesting that they exist in various dynamic equilibria
with one another21 (Fig. 4). Several possible structures have
been proposed for aggregates involving lithiated anisoles
with various combinations of TMEDA and ether22 and similar
opportunities for aggregation are envisaged for 8.
245 K
234 K
(a)
222 K
210 K
200 K
(b)
(c)
194 K
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
ppm
59 58 57 56 55 54 53 52 51 50 49 48 47 46 45 44 43 42
ppm
Fig. 4 Dynamic NMR study of 8 in TMEDA (155.5 MHz, dLi
0–5.5, LiBr) showing dynamic relationship between several
less populated aggregates and the sustained dominance of a
single species.
Fig. 3 Comparison of 13C NMR spectra (100 MHz, dC 40–80, CDCl3)
of (a) TMEDA, (b) 8 without TMEDA and (c) 8 with TMEDA.
PAPER: 09/0623

JOURNAL OF CHEMICAL RESEARCH 2009 471
was necessary were performed using glassware dried in a 150°C
oven overnight or dried at 600°C with a heatgun and flushed with
dry Ar to remove air and moisture. All reactions were performed
under an atmosphere of dry Ar. Immediately prior to their use Et2O
and THF were distilled from sodium metal/benzophenone ketyl.
General laboratory solvents were distilled before use. Reactions were
monitored by thin layer chromatography (DC-Plastikfolien Kieselgel
60 F254 plates). Plates were visualised under UV light and developed
by spraying with either 10% conc. H2SO4 in methanol or by exposure
to iodine. Kieselgel 60 (230–400 mesh) was used for initial flash
chromatographic separations. Semi-preparative HPLC was performed
using a Whatman's Magnum 9 Partisil 10 column (10 mm i.d., length
50 cm) and a Waters 410 Differential Refractometer. All reagents
were commercially available, except 1-bromo-2,5-dimethoxy-4-
methylbenzene (6) which was prepared as described below.
257 K
245 K
234 K
222 K
General procedure for the preparation of NMR samples
Method A: The organolithium samples were prepared by reaction of
6 with n-BuLi. The samples of 6 (28 mg, 0.12 mmol) were weighed
into oven-dried 5 mm NMR tubes fitted with a screw cap and a
CDCl3 insert and flushed with argon. After dilution with ether and
cooling in a dry ice/acetone bath, each sample was treated with
a stoichiometric amount of n-BuLi (~1.5 equiv. of 1.5 M solution)
and 0, 1 or 2 equivalents of TMEDA and mixed by shaking and
applying a vortex. The resulting organolithium samples were then
inserted into the NMR probe precooled to 198 K. After a short time
for temperature equilibration, the field was locked and shimmed. This
was followed by acquisition of a 13C (proton decoupled) spectrum
210 K
200 K
194 K, after addition of geranyl bromide
194 K, before addition of geranyl bromide
7
(referenced to CDCl3) and a Li NMR spectrum (referenced to an
external standard of 1M LiBr in D2O at 303 K). Additional lithium
spectra were subsequently obtained at intervals as the temperature
was increased. The temperature of the NMR probe was calibrated
using a 4% MeOH in d4-methanol sample.
Method B: The n-BuLi aliquots were transferred into oven-dried
5 mm NMR tubes fitted with a screw cap and a CDCl3 insert and
flushed with argon. After dilution with ether, each sample was treated
with a stoichiometric amount of TMEDA and mixed by shaking
and applying a vortex before cooling in a dry ice/acetone bath.
The solution of 6 (28 mg, 0.12 mmol) in ether (0.2 mL) was added
dropwise and the resulting solution mixed carefully using a vortex
mixer. The resulting organolithium samples were then inserted into
the NMR probe precooled to 198 K. After a short time for temperature
equilibration, the field was locked and shimmed. This was followed
by acquisition of a 13C (proton decoupled) spectrum (referenced to
CDCl3) and a 7Li NMR spectrum (referenced to an external standard
of 1M LiBr in D2O at 303 K). Further lithium spectra were then
acquired at intervals as the temperature was increased.
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
ppm
Fig. 5 Variable temperature NMR study of 8 in TMEDA
following the addition of geranyl bromide (155.5 MHz, dLi
0–5.5, LiBr) showing the effect of increased temperature on the
progress of the reaction.
(and thereby reduce the potential for side reactions) The
gratifying outcome was a significant increase in the isolated
yield (65%), accompanied by an elimination of the alternative
coupling product (10). Furthermore, this method yields the
reproducible results absent when using the earlier method.5
In conclusion, we have established firstly, that even in the
presence of TMEDA, MHE-mediated prenylation reactions
of bromodimethoxytoluene substrates give rise to multiple
Li aggregates in the reaction mixture, ultimately resulting in
a complex cohort of products. Secondly, the low recovery
of unreacted 6 and the absence of corresponding signals
(such as C–Br) in the 13C NMR suggest that MHE is almost
quantitative in the presence of TMEDA and the low yields
often encountered in this reaction cannot be attributed to
the initial MHE step but rather to the electrophilic addition
step. Thirdly, competition between geranyl bromide and
trace amounts of water, gives rise to protonation of 8 and
the formation of dimethoxytoluene (11) as a significant
side product. Finally, we have established that alteration of
the sequence in which the reactants are added has improved
the yields of this reaction from a variable 5–40% to a
consistent 65%.
Synthesis of 1-bromo-2,5-dimethoxy-4-methylbenzene (6)23
N-bromosuccinimide (4.40 g, 24.8 mmol) was added to a solution
of 1,4-dimethoxy-2-methylbenzene (3.15 g, 20.7 mmol) in MeCN
(200 mL) and stirred overnight at room temperature. The solvent was
then removed under reduced pressure and the residue taken up in
CH2Cl2 (50 mL), washed with sodium sulfite solution (50 mL) and
water (50 mL) dried over anhydrous MgSO4 and concentrated under
reduced pressure to give a white solid (8.56 g). Flash chromatography
of the crude product in 9:1 hexane: EtOAc yielded pure 6 (3.94 g,
17.0 mmol, 82%) as white plates from hexane; m.p 82–86°C, lit.24
81–82°C; IR nmax 2945, 2833, 1505, 1217, 1037, 866 cm-1; dH
(600 MHz, CDCl3) 6.98 (1H, s, H-3), 6.73 (3H, s, H-6), 3.83 (3H, s,
OMe), 3.77 (3H, s, OMe), 2.17 (3H, s, H-7). dC (150 MHz, CDCl3)
152.2, 149.7, 126.8, 115.4, 115.2, 108.0, 57.0, 56.1, 16.3. Calcd for
C9H11O279Br [M+] 229.9942. Found 229.9943.
Synthesis of 1-[(2E)-3,7-dimethylocta-2',6'-dienyl]-2,5-dimethoxy-4-
methyl-benzene (7)
TMEDA (1.01 g, 8.65 mmol, 2.0 equiv.) followed by 1.6 M n-BuLi
(5.4 mL, 8.65 mmol) was added to dry Et2O (1 mL) at –10°C and
allowed to stir for 15 min before a solution of 6 (1.0 g, 4.33 mmol,
1.0 equiv.) in dry Et2O (5 mL) was added. The resulting mixture was
stirred at –10°C for 15 min before geranyl bromide (1.88 g, 8.65 mmol,
2.0 equiv.) was added dropwise and the mixture was allowed to stir
overnight at room temperature. The reaction was thereafter quenched
with sat. NH4Cl (10 mL) and the aqueous solution extracted with Et2O
(3 ¥ 5 mL). The combined ethereal extracts were washed with water
and sat. brine and dried over anhydrous MgSO4. Concentration in
vacuo afforded the crude mixture (1.97 g) as a brown oil. Purification
using NP HPLC (96:4 hexane: EtOAc) yielded 7 (810 mg, 2.8 mmol,
65%) as a pale yellow oil. IR nmax 2915, 2849, 1506, 1400, 1211, 857;
Experimental
Melting points were determined using a Reichert hot stage microscope
and are uncorrected. IR spectra were recorded on a Perkin Elmer
Spectrum 2000 FT-IR spectrometer with the compounds as films
(neat) on NaCl discs. NMR spectra were acquired on Bruker 400
MHz Avance and 600 MHz Avance II spectrometers using standard
pulse sequences. Chemical shifts are reported in ppm, referenced to
residual solvent resonances (CDCl3 dH 7.25, dC 77.0), and coupling
constants are reported in Hz. HRFABMS data were obtained on
a JEOL SX102 spectrometer. Reactions where exclusion of water
PAPER: 09/0623

472 JOURNAL OF CHEMICAL RESEARCH 2009
5
B.A. Scheepers, R. Klein and M.T. Davies-Coleman, Tetrahedron Lett.,
2006, 47, 8243.
S.I. Odejinmi and D.I. Wiemer, Tetrahedron Lett., 2005, 46, 3871.
C. Hoarau and T.R.R. Pettus, Synlett, 2003, 127.
N.J.R. van Eikema Hommes and P. von Ragué Schleyer, Tetrahedron,
1994, 50, 5903.
dH (600 MHz, CDCl3) 6.67 (1H, s, H-6), 6.65 (1H, s, H-3), 5.30 (1H,
t, J = 7.3 Hz, H-2'), 5.10 (1H, t, J = 6.8 Hz, H-6'), 3.77 (3H, s, OMe-1),
3.76 (3H, s, OMe-4), 3.30 (2H, d, J = 7.3 Hz, H-1'), 2.18 (3H, s, H-7),
2.10 (2H, dd, J = 14.4, 6.8 Hz, H-5'), 2.04 (2H, m, H-4'), 1.70 (3H, s,
H-10'), 1.67 (3H, s, H-8'), 1.59 (3H, s, H-8'). dC (150 MHz, CDCl3)
151.6, 151.0, 136.1, 131.4, 128.0, 124.4, 124.3, 122.7, 114.0, 112.4,
56.3, 56.1, 39.8, 28.2, 26.8, 25.7, 17.7, 16.1. Calcd for C19H28O2
[M+] 288.2089. Found 288.2090
6
7
8
9
D.B. Collum, Acc. Chem. Res., 1992, 25, 448.
10 A. Galiano-Roth and D.B. Collum, J. Am. Chem. Soc., 1989, 111, 6772.
11 S.J. Zuend, A. Ramirez, E. Lobkovsky and D.B. Collum, J. Am. Chem.
Soc., 2006, 128, 5939.
12 H.J.Reich,W.S.Goldenberg,A.W.Sanders,K.L.JantziandC.C.Tzschucke,
J. Am. Chem. Soc., 2003, 125, 3509.
13 W. Bauer and P. von R. Schleyer, J. Am. Chem. Soc., 1989, 111, 7191.
14 A. Borman and D. Johnels, Magn. Reson. Chem., 2000, 38, 853.
15 T.D.W. Claridge, High resolution NMR techniques in organic chemistry,
Elsevier, Amsterdam, 1999, p. 41.
16 A. Ramirez and D.B. Collum, J. Am. Chem. Soc., 1999, 121, 11114.
17 S.T. Chadwick, R.A. Rennels, J.L. Rutherford and D.B. Collum, J. Am.
Chem. Soc., 2000, 122, 8640.
18 R.A. Rennels, A.J. Maliakal and D.B. Collum, J. Am. Chem. Soc., 1998,
120, 421-422.
We thank the South African National Research Foundation
and the Medical Research Council for financial support, and
DAAD for a PhD bursary (SNS).
Received 4 June 2009; accepted 29 July 2009
Paper 09/0623 doi: 10.3184/030823409X12491261540439
Published online: 14 August 2009
References
19 V. Snieckus, Chem. Rev., 1990, 90, 879.
1
K.L. McPhail, M.T. Davies-Coleman and J. Starmer, J. Nat. Prod., 2001,
64, 1183.
20 O. Mendoza, L.P. Cuffe, F.K. Rehmann and M. Tacke, J. Organomet.
Chem., 2005, 690, 1511.
2
3
G.J. Hooper and M.T. Davies-Coleman, Tetrahedron Lett., 1995, 36, 3265.
G. Koren-Goldshlager, P. Klein, Y. Benayahu, M. Schleyer and
Y. Kaschman, J. Nat. Prod., 1996, 59, 262.
C. Whibley, K. McPhail, R. Keyzers, M. Maritz, V. Leaner, M. Birrer,
M. Davies-Coleman and D. Hendricks, Mol. Cancer Ther., 2007, 6, 2535.
21 E.L. Eliel and S.H. Wilen, Stereochemistry of organic compounds, 1994,
pp. 502-507.
22 H. Gunther, J. Brazil Chem. Soc., 1999, 10, 241.
23 J.L. Bloomer and W. Zheng, Synth. Commun., 1998, 28, 2087.
24 M. Yoshida, Y. Shoji and K. Shishido, Org. Lett., 2009, 11, 1441.
4
PAPER: 09/0623