Novel synthesis of [13C4,15N]1H-pyrrole-2, 3,5-tricarboxylic acid: An important biomarker for melatonin metabolism

Article abstract of DOI:10.1002/jlcr.1704

1H-pyrrole-2,3,5-tricarboxylic acid is a breakdown product of melatonin. A labeled version of this compound would serve as a key biomarker for drug candidates which track this substance to monitor their effectiveness (e.g. hyperpigmentation drugs). A Hantzsch synthesis using readily available starting materials was used to generate [¹³C₄,¹⁵N]1H- pyrrole-2,3,5-tricarboxylic acid in six steps (12% overall yield). Copyright

Full text of DOI:10.1002/jlcr.1704

Research Article
Received 15 May 2009,
Revised 21 August 2009,
Accepted 27 October 2009
Published online 4 January 2010 in Wiley Interscience
(
www.interscience.wiley.com) DOI: 10.1002/jlcr.1704
1
Novel synthesis of [ C , N]1H-pyrrole-2,3,
3
15
4
-tricarboxylic acid: an important biomarker
5
for melatonin metabolism
Ã
Marc B. Skaddan
1H-pyrrole-2,3,5-tricarboxylic acid is a breakdown product of melatonin. A labeled version of this compound would serve as
a key biomarker for drug candidates which track this substance to monitor their effectiveness (e.g. hyperpigmentation
1
3
15
drugs). A Hantzsch synthesis using readily available starting materials was used to generate [ C , N]1H-pyrrole-2,3,5-
4
tricarboxylic acid in six steps (12% overall yield).
1
Keywords: stable label; C; N; C-13; N-15; melatonin; pyrrole; biomarker
3
15
1
5
Introduction
4
acetoacetate, NH OH, and chloroacetaldehyde (Scheme 2).
The use of these labeled precursors satisfies the mass and
1H-pyrrole-2,3,5-tricarboxylic acid (1a) is a breakdown product regiochemical requirements of the desired compound. At this
of melatonin. To monitor the effectiveness of a new hyperpig-
mentation drug, a request was made for 200 mg of stable a carboxy electrophile to the 5-position. Subsequent oxidation
point, the electronics of the ring should direct substitution with
1
isotope labeled 1a (with a minimum mass increase of 3 AMU of the 2-methyl group, followed by hydrolysis, produces the final
and preferably 4 or 5 AMU). At least 3 Da mass difference had to
be present in the ring itself, since it was known that at least one
compound.
The synthesis was first performed with unlabeled material for
carboxylic acid residue was lost under the MS ionization optimization purposes. Using a slightly modified literature
2
procedure, NH OH was added to an aqueous mixture of ethyl
acetoacetate and chloroacetaldehyde, resulting in an exother-
conditions used to track the biomarker.
The discovery chemistry group had previously synthesized a
4
small amount of 1a via degradation of a commercially available mic reaction (Scheme 3, yields in parentheses refer to stable-
starting material, 5-hydroxyindole-2-carboxylic acid (Scheme 1). labeled product). Acidification and extractive workup provided
Yields were extremely poor and the final compound was difficult
to separate from reaction by-products by normal chromato-
pyrrole 2a.
Yields were modest due to a competing Feist–Benary reaction
graphic methods. Importantly, the synthesis was not clearly that produced an equimolar amount of furan 3 (Scheme 4).
amenable to the production of labeled 1a. Therefore, a new
route was sought to provide an analog with a mass increase of found that trituration of the crude oily extract with cold hexanes
AMU from commercially available stable isotope starting provided product that was pure by NMR.
While the furan can be removed by flash chromatography, it was
5
materials.
Installation of the 5-carboxy group was accomplished by
heating 2a with trichloroacetyl chloride in dioxane, which
provided crude 4a quantitatively following workup. Ethanolysis
of the trichloromethyl ketone generated the 2,4,5-trisubstituted
pyrrole 5a in excellent yield after purification. Oxidation of the
Results and discussion
A thorough search of the literature revealed that no synthetic
routes to 1a had been published. A myriad of methods are
available for synthesizing substituted pyrroles, including the
Paal–Knorr, Hantzsch and Kenner syntheses, and 1,3-dipolar
5
challenge. Pyrroles are, in general, easily attacked by oxidizing
-methyl group to the carboxylic acid presents a special
1
reagents, frequently with complete decomposition. Moreover,
1
cycloadditions to alkynes. Since the labeled synthesis required
oxidation of alkyl substituents alpha to the nitrogen in pyrroles
are usually sluggish unless activated by an electron-withdrawing
group in the second alpha position. In the present case,
at least three stable labels in the ring itself, simple substitution
of readily accessed pyrrole starting materials was not feasible.
Moreover, many of the aforementioned methods rely on
precursors that are difficult to produce using known labeled
starting materials. The Hantzsch synthesis, however, was Pfizer Global Research and Development, Pfizer Inc., Eastern Point Road, Groton,
attractive due to the fact that it used inexpensive, readily CT 06340, USA
available labeled precursors.
*
8
Correspondence to: Marc B. Skaddan, Pfizer Inc., Eastern Point Road, Mail Stop
274-1342, Groton, CT 06340, USA.
Our envisioned route uses the Hantzsch reaction to construct
1
3
the 2,3-disubstituted pyrrole nucleus from [1,2,3,4- C ]ethyl E-mail: mskaddan@yahoo.com
4
J. Label Compd. Radiopharm 2010, 53 73–77
Copyright r 2009 John Wiley & Sons, Ltd.

M. B. Skaddan
carboethoxy moieties in the 2 and 4 positions set up pyrrole 5a the final compound 1a. Unfortunately, with only one aromatic
1
nicely for oxidation of the 5-methyl group. Direct conversion of proton, H NMR was unsupportive in this regard. Other NMR
1
3
15
1
a-methyl substituents to the acid/ester with SO
however, the literature substrates consist of tetrasubstituted experiments on 1a and other trisubstituted intermediates in
2 2
Cl are known; experiments, including C NMR, HMBC, and N- H IMPEACH
3
–6
13
pyrroles. The open 3-position in 5a would almost certainly be this sequence, were also inconclusive (for C experiments,
halogenated under these conditions. The better approach is to the resonances of the carbon signals did not closely match the
perform a stepwise oxidation of the alkyl moiety to the predicted resonances of either regioisomer). The definitive
aldehyde, then to the acid. Cerric ammonium nitrate (CAN) answer came from a C 2D INADEQUATE experiment (Figure 1),
oxidations of activated a-methyl substituents have been shown which unambiguously showed coupling of the pyrrole ring C-H
1
3
7
–9
to proceed in good yields;
reacting 5a with 6 equivalents of CAN in MeOH. This provided regioisomer would have showed only 1 other carbon coupling).
aldehyde 6a in modest yield, the main by-product being the With the successful implementation of the synthetic strategy on
therefore, we proceeded by carbon to two other pyrrole ring carbons (the 4-substituted
analogous 5-methoxymethyl pyrrole. Oxidation of the aldehyde unlabeled 1a, attention was turned to synthesizing the labeled
side chain using sodium chlorite failed to produce appreciable analog 1b. The synthesis proceeded without any noticeable
amounts of the desired acid. On the contrary, slow addition of deviations from the unlabeled synthesis, using the aforemen-
potassium permanganate to a dilute acetone solution of 6a, tioned labeled precursors (Scheme 3). Starting with 3.1 g
1
3
followed by heating at 551C, provided acid 7a in good yield. (23.0 mmol) of [1,2,3,4- C ]ethyl acetoacetate and 120 mmol
4
1
5
Basic hydrolysis furnished the desired triacid 1a quantitatively
extractive workup only). The entire sequence gave an overall 12%, which matched what was achieved with 1a. Mass spectro-
yield of 12%. metry clearly showed the molecular ion associated with the M15
4
NH OH, 560mg of 1b was produced in an overall yield of
(
Even though we predicted that reaction of 2a would occur species, with no M10 evident. The purity of the final compound
alpha to the nitrogen (vide supra), empirical evidence was (92% by UV-HPLC) was sufficient for the project team’s needs.
required to prove that we did, in fact, get 5- vs 4-substitution for
In conclusion, we have synthesized the M15 melatonin
1
biomarker [ C , N]1H-pyrrole-2,3,5-tricarboxylic acid in a six
3
15
4
step 12% overall yield from relatively inexpensive and readily
available stable label starting materials. While some of the
transformations have modest yields, the chemistry is very
reproducible. This compound will serve as a valuable tool to
track melatonin metabolism in drug therapies designed to treat
various ailments involving this important hormone.
Scheme 1. Degradation route to 1a.
13
15
Scheme 2. Retrosynthetic route for [
4
C , N]1H-pyrrole-2,3,5-tricarboxylic acid.
Scheme 3. Synthesis of 1a/b via Hantzsch reaction.
Scheme 4. Competing Feist–Benary reaction to produce furan 3.
www.jlcr.org
Copyright r 2009 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 73–77

M. B. Skaddan
Figure 1. INADEQUATE spectrum of 1a.
1
3
15
Experimental
[ C4; N]1H-Pyrrole-2,3,5-tricarboxylic acid (1b)
The procedure used to prepare 1a was followed using 0.669 g 7b
(2.57 mmol) and 21 ml of the KOH solution (55.8 mmol). After 1.5h
All unlabeled reagents and solvents were obtained from Acros
and Aldrich and used without further purification. The
1
3
at 511C, the reaction mixture was worked up as described for 1a
to provide 1b (560 mg, 107%) as a slightly yellow solid. H NMR
[
Isotopes (CLM-3297, 99%) and the [ N]NH OH was purchased
4
1,2,3,4- C ]ethyl acetoacetate was purchased from Cambridge
1
1
5
4
(
1
DMSO-d , 300MHz) d 7.00 (s, 1H), 11.31 (d, J
2.71 (bs, 1H). ESIMS (neg. ion): m/z 203 (M-H, 100), 159 (M-CO
Hꢁ13
C
= 98.2 Hz, 1H),
from Aldrich (cat] 485454, 6 N, 98%). Mass spectrometric data
were collected on a PerkinElmer Sciex API 150EX (electrospray).
Reaction progress was monitored by analytical thin-layer
chromatography (Analtech scored glass 10 cm ꢀ 20 cm hard
TLC plates). TLC plates were visualized using short wave UV light
6
2
H,
47). HPLC conditions (A= 5:95 CH CN:0.1% TFA in water; B = 70:30
CH CN:0.1% TFA in water): 0–10 min 100% A, 10–11 min 100% A
to 100% B, 11–16min 100% B, 16–17 min 100%B to 100% A,
17–20 min 100% A. HPLC purity = 92% (t
impurity at t = 14.2 min (8%).
3
3
1
13
R
= 9.40 min). Minor
(
254 nm) or potassium permanganate. H and C NMR spectra
were obtained on a Bruker 300, 400, or 500 MHz spectrometer.
Resonances are reported in parts per million relative to the
R
incomplete deuteration signal from the NMR solvent. HPLC Ethyl 2-methyl-1H-pyrrole-3-carboxylate (2a)
analyses were performed on an Agilent 1100 LC system, using a
Phenomenex Luna C-18(2) 4.6 ꢀ 250 mm column, UV = 254 nm,
flow = 1.0 ml/min.
To a vigorously stirred solution of ethyl acetoacetate (2.60g,
0.0mmol) in water (10 ml) was added chloroacetaldehyde
2.80 ml of a 50 wt% in water solution, 22.0mmol), followed
2
(
immediately by ammonium hydroxide (9.70 ml of a 28wt% water
solution, 144 mmol). The reaction exothermed and was stirred
1
H-Pyrrole-2,3,5-tricarboxylic acid (1a)
2
Following a slightly modified literature procedure, compound overnight, providing an oily, globular precipitate. Analysis of the
a (949 mg, 3.72 mmol) was dissolved in 2 M KOH (30 ml of a reaction by TLC (40% EtOAc/hexanes) showed the presence of
0:10 water:EtOH solution). The reaction was followed by MS, product (R = 0.41) and a single, more polar impurity (equal
which showed steady but sluggish hydrolysis. After 24 h, the intensity under UV light, R = 0.26). The heterogeneous solution
7
9
f
f
reaction was heated for 1.5 h at 511C and this resulted in was extracted with 20 ml diethyl ether. The aqueous phase was
complete conversion to the triacid. The reaction was cooled in extracted with another 10ml of ether, and the organics were
an ice bath and acidified with 6 M HCl. A precipitate formed combined. The combined organics were washed sequentially with
from the yellow solution. The solid was filtered, and the filter 10 ml 10% NaOH, 10ml water, and 10 ml 5% HCl. Following the
cake was washed with cold water, diethyl ether, hexanes, and HCl wash, the TLC analysis showed complete conversion of the
EtOAc. The yellow solid was dried on a high vacuum line to 0.26 R
impurity to a less polar impurity (R
, 300 MHz) d 7.03 (s, compound 3 from the competing Feist–Benary reaction. The
6
= 0.62), presumably
f
f
1
provide 1 (802 mg, 108%). H NMR (DMSO-d
1
3
1
1
1
H), 11.72 (s, 1H). C NMR (DMSO-d , 100 MHz) d 117.6, 120.2, organic layer was dried with sodium sulfate and evaporated to an
6
23.8, 131.4, 161.1, 161.8, 165.7. ESIMS (neg. ion): m/z 198 (M-H, oily residue with crystalline solids in it. Ten milliliters of hexanes
00), 154 (M-CO
2
H, 50).
was added, and the mixture was stirred for 30 min in an ice bath
J. Label Compd. Radiopharm 2010, 53 73–77
Copyright r 2009 John Wiley & Sons, Ltd.
www.jlcr.org

M. B. Skaddan
to triturate the product. The cold mixture was filtered and the filter showed complete conversion of the starting material. The
cake washed with a few milliliters of cold hexanes. This provided a reaction was cooled to room temperature and then filtered. The
slightly off-white, pinkish powder and an oily filtrate. The powder
K
2
CO
was dried in vacuo for 2 h to provide 2 (1.02 g, 33%). H NMR evaporated in vacuo to afford a brown solid. The solid was
CDCl , 300 MHz): d 1.37 (t, J =7.3 Hz, 3H), 2.56 (s, 3H), 4.30 purified by flash chromatography (SiO , 20% EtOAc/hexanes) to
q, J= 7.3 Hz, 2H), 6.59 (s, 1H), 8.40 (bs, 1H). ESIMS: m/z 176 provide 5a as a slightly yellow fluffy solid (2.64 g, 96%). H NMR
3
cake was washed with 3 ml EtOH, and the filtrate was
1
(
3
2
1
(
1
1
(M1H1Na , 100), 154 (M1H , 79), 126 (27). Isolation of 3 from (DMSO-d , 500 MHz, 5 mm Cryoprobe) d 1.24 (t, J = 7.2 Hz, 3H),
6
the filtrate by flash chromatography (SiO , 20–30% EtOAc/hexanes) 1.27 (t, J = 7.2 Hz, 3H), 2.42 (s, 3H), 4.14 (q, J = 7.2 Hz, 2H), 4.21 (q,
2
provided a pungent-smelling, volatile oil (1.17 g, with some EtOAc J = 7.2 Hz, 2H), 6.99 (s, 1H), 12.33 (s, 1H). ESIMS: m/z 248 (M1H1
1
1
1
still present). H NMR (CDCl , 300MHz): d 1.37 (t, J=7.3Hz, 3H), 2.59 Na , 94), 226 (M1H , 100), 198 (21).
3
(
s, 3H), 4.31 (q, J= 7.3 Hz, 2H), 6.66 (s, 1H), 7.24 (s, 1H).
½
13C4; 15NꢂDiethyl 5-methyl-1H-pyrrole-2,4-dicarboxylate (5b)
[13C4; 15NꢂEthyl 2-methyl-1H-pyrrole-3-carboxylate (2b)
The procedure used to prepare 5a was followed, using 2.54 g
The procedure used to prepare 2a was followed, using 20 ml (ꢃ7.85 mmol) 4b, 15 ml EtOH, and 1.08 g (7.85 mmol) K CO
2
3
.
15
13
(
119.6 mmol) 6 N [ N]NH
acetoacetate, and 3.20 ml (25.3 mmol) chloroacetaldehyde. After 6 h, provide 5b as a slightly yellow, fluffy solid (1.70 g, 94%). H NMR
the reaction was worked up as described for 2a to afford 2b as (CDCl , 300 MHz) d 1.38 (m, 6H) 2.59 (d, JHꢁ13C = 129.6 Hz, 3H),
4 4
OH, 3.10 g (23.0 mmol) [1,2,3,4- C
]ethyl The reaction was worked up and purified as described for 5a to
1
3
1
a white powder (1.26 g, 35%). H NMR (CDCl
3
, 300MHz) d 1.37 4.33 (m, 4H), 7.26 (s, 1H), 9.23 (d, J
Hꢁ15N
= 97.0 Hz, 1H). ESIMS: m/z
1
1
(t, J=7.5Hz, 3H), 2.56 (d, JHꢁ =129Hz, 3H), 4.29 (q, J=7.0Hz, 2H), 253 (M1H1Na , 100), 231 (M1H , 85), 203 (15).
13 C
6.59 (m, 2H), 8.17 (d, J
1
Hꢁ15
=129Hz, 1H). ESIMS: m/z 181 (M1H1Na
N
1
100), 159 (M1H , 80), 131 (20).
,
Diethyl 5-formyl-1H-pyrrole-2,4-dicarboxylate (6a)
7
Following Bobal’s procedure, a solution of 5a (2.57 g,
1.4 mmol) in MeOH (50 ml) was cooled in an ice bath. To this
Ethyl 2-methyl-5-(2,2,2-trichloroacetyl)-1H-pyrrole-3-carboxylate
4a)
1
was added a solution of CAN (25.6 g, 46.7 mmol) in water (25 ml).
(
To a solution of 2a (1.89 g, 12.3 mmol) in dioxane (20 ml) and 2,6- The reaction turned a deep claret color and the ice bath was
lutidine (1.58 ml, 13.5 mmol) was added the trichloroacetyl removed. After 25 min, the reaction had changed to a yellow
chloride (1.51 ml, 13.5 mmol) at room temperature. The turbid color. After 1 h, analysis by TLC showed incomplete conversion
mixture was heated to 851C. After 1 h, a milky white precipitate of starting material to the aldehyde. After 1.5 h, an additional
formed, and TLC showed complete disappearance of the starting 12.5 g (22.8 mmol) of CAN was added to the reaction mixture
material. The reaction was cooled and half of the dioxane was and the reaction mixture turned claret. After another 70 min
removed in vacuo. The remaining mixture was partitioned (total reaction time = 2.5 h), the reaction was an orange color
between 50 ml EtOAc and 10 ml water. The EtOAc layer was and TLC showed complete disappearance of the starting
washed with 10 ml saturated NaCl. The combined aqueous material. The MeOH was evaporated in vacuo, during which
washes were back-extracted with EtOAc (20 ml110 ml). The time a solid formed. The aqueous suspension was extracted
combines organics were washed with 4 ml saturated NaCl, then with CH Cl (1 ꢀ 50 ml, 2 ꢀ 35 ml), and the combined extracts
2
2
2 4
2
dried with Na SO and evaporated in vacuo to provide 4a as an were dried with Na SO4 and evaporated to an orange solid.
orange solid (4.16 g, 4100%). This crude material was used for Purification by flash chromatography (SiO2, 20% EtOAc/
1
the next reaction without further purification. H NMR (CDCl
3
,
hexanes) provided 6a as a white fluffy solid (1.29 g, 47%)
00 MHz) d 1.40 (t, J = 7.0Hz, 3H), 2.65 (s, 3H), 4.34 (q, J = 7.0 Hz, containing a minor impurity by TLC (2.5% acetone/CH Cl ). This
3
2
1
2
2
1
3
H), 7.73 (s, 1H), 9.39 (bs, 1H). C NMR (CDCl , 400 MHz) d 13.81, was used in the next reaction without further purification.
4.44, 60.29, 94.46, 115.89, 120.57, 123.29, 143.58, 164.01, 173.49.
1
CN, 48), 298 (M , 100), 270 (7).
3
1
H NMR (CDCl , 300 MHz) d 1.42 (m, 6H), 4.41 (m, 4H), 7.34
3
1
(s, 1H), 10.10 (bs, 1H), 10.35 (s, 1H). ESIMS: m/z 262 (M1H1Na ,
MS (APCI): m/z 339 (M1CH
3
1
00), 240 (M1H , 31).
1
[13
3
C4; 15N]Ethyl 2-methyl-5-(2,2,2-trichloroacetyl)-1H-pyrrole-
-carboxylate (4b)
[13
C4; 15N]Diethyl 5-formyl-1H-pyrrole-2,4-dicarboxylate (6b)
The procedure used to prepare 6a was followed, using 1.69 g
(7.34 mmol) 5b, 33ml MeOH, 16.5g (30.1mmol) CAN in 16ml
water, and 8.00 g (14.6 mmol) of additional CAN. The reaction was
worked up as described for 6a, then purified by flash chromato-
The procedure used to prepare 4a was followed, using 1.24 g
7.85 mmol) of 2b, 0.997 ml (8.64 mmol) of 2,6-lutidine,13 ml
(
dioxane, and 0.964 ml (8.64 mmol) of trichloroacetyl chloride.
After 1 h, TLC revealed the presence of starting material. An
additional 100 ml (0.896 mmol) of trichloroacetyl chloride was
added, and the solution was heated for an additional 30 min,
with no change in the progress of the reaction. The reaction was
worked up as described for 4a to provide 4b as a light brown
solid (2.54 g, 107%), which was used in the next step without
further purification.
graphy (SiO
2 2 2
, 1.5% acetone/CH Cl ) to give 6b as a white
solid (787 mg, 44%). H NMR (CDCl , 300MHz) d 1.42 (m, 6H),
15 N
.41 (m, 4H), 7.34 (s, 1H), 10.0 (d, JHꢁ = 101 Hz, 1H), 10.29
1
3
4
(
2
1
dd, JHꢁ = 192.0, 30.2 Hz, 1H). ESIMS: m/z 267 (M1H1Na , 100),
13 C
1
45 (M1H , 33).
3
,5-Bis(ethoxycarbonyl)-1H-pyrrole-2-carboxylic acid (7a)
Diethyl 5-methyl-1H-pyrrole-2,4-dicarboxylate (5a)
1
Using a slightly modified literature procedure, a solution of
0
To a solution of 4a (3.66 g, 12.2 mmol) in 23 ml EtOH was added KMnO (2.44 g, 15.4 mmol) in 125 ml water and 125 ml acetone
4
2 3
the K CO (952 mg, 6.89 mmol). The slurry was heated at 901C was added dropwise via addition funnel to a solution of 6a
for 2 h, during which the reaction mixture turned brown. TLC (1.23 g, 5.14 mmol) in acetone (150 ml) over the course of 3 h.
www.jlcr.org
Copyright r 2009 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2010, 53 73–77

M. B. Skaddan
The reaction was monitored by MS, and showed slow but steady 10.29 (d, JHꢁ = 98.7 Hz, 1H). ESIMS (neg. ion): m/z 259
13C
progress. After 6 h, MS showed a 58:42 ratio of compound 7a:6a. (M-H, 81), 214 (M-CO
The reaction mixture was heated at 551C for 1 h, then room
temperature overnight. The reaction was checked again by MS,
showing a 92:8 ratio of compound 7a:6a, and the reaction was
2
H, 100).
Acknowledgements
heated again for 1 h at 551C. The ratio of 7a:6a was now 97:3,
and the acetone was removed in vacuo. The purple solution was
treated with 10% sodium metabisulfite in 1 M HCl solution
We thank Stephen Grode from the Rapid ID Lab (Portage, MI) for
his work on the INADEQUATE experiments, as well as David J.
Russell and Gary Martin from the Structure Elucidation Group
(
95 ml). The purple solution turned clear and exothermed. It was
extracted with CH Cl (3 ꢀ 50 ml), and the organics were dried
2
2
15
1
(
Portage, MI) for their help in the HMBC and N- H IMPEACH
with Na SO and evaporated in vacuo to a slightly yellow solid
2
4
1
NMR experiments. We also thank the members of the Kalamazoo
Radiosynthesis Lab for their scientific input and support.
(
4
1
1.02 g, 78%). H NMR (CDCl
3
, 300 MHz) d 1.43 (m, 6H), 4.43 (m,
H), 7.37 (s, 1H), 10.43 (bs, 1H). ESIMS (neg. ion): m/z 254 (M-H,
00), 210 (M-CO H, 87).
2
[13C4; 15N]3,5-Bis(ethoxycarbonyl)-1H-pyrrole-2-carboxylic
acid (7b)
References
[
1] J. A. Joule, K. Mills, G. F. Smith, Pyrroles: reactions and synthesis,
Heterocyclic Chemistry 3rd ed., Chapman & Hall, New York, 1995,
pp. 229–258.
The procedure used to prepare 7a was followed, using 777 mg
(
3.18 mmol) 6b, 93 ml acetone, and 1.51 g (9.54 mmol) KMnO
4
in
77 ml water and 77 ml acetone. The reaction mixture was heated
[
2] A. Treibs, W. Ott, Justus Liebigs Ann. Chem. 1958, 615, 137.
to 551C for 3 h. An additional 251 mg (1.59 mmol) KMnO was
[3] U. Eisner, R. P. Linstead, E. A. Parkes, E. Stephen, J. Chem. Soc.
956, 1655.
4
1
added and the reaction mixture was stirred for 1 h. The heat was
removed and the reaction allowed to stir overnight at room
temperature. The solution was evaporated in vacuo (ꢃ130 ml
acetone removed), and the resulting aqueous solution was
poured into 60 ml 10% sodium metabisulfite in 1 M HCl (cooled
to 51C). The clear solution was worked up as described in 7a to
[4] B. Morgan, D. Dolphin, R. H. Jones, T. Jones, F. W. B. Einstein,
J. Org. Chem. 1987, 52, 4628.
[5] M. Scarsella, G. Sleiter, Gazzetta Chimica Italiana 1988, 118, 757.
[6] E. Vogel, M. Balci, K. Pramod, P. Koch, J. Lex, O. Ermer,
Angewandte Chemie 1987, 99, 909.
[
[
7] P. Bobal, D. A. Lightner, J. Heterocycl. Chem. 2001, 38, 1219.
8] A. J. Moreno-Vargas, J. G. Fernandez-Bolanos, J. Fuentes,
I. Robina, Tetrahedron: Asymmetry 2001, 12, 3257.
provide 7b as a white solid (679 mg, 82%), which was used
1
in the next step without further purification.
CDCl
H NMR
, 300 MHz) d 1.44 (m, 6H), 4.44 (m, 4H), 7.36 (s, 1H),
[9] T. Thyrann, D. A. Lightner, Tetrahedron Lett. 1996, 37, 315.
(
3
[10] P. Barker, P. Gendler, H. Rapoport, J. Org. Chem. 1978, 43, 4849.
J. Label Compd. Radiopharm 2010, 53 73–77
Copyright r 2009 John Wiley & Sons, Ltd.
www.jlcr.org