Journal of Labelled Compounds and Radiopharmaceuticals
J Label Compd Radiopharm 2007; 50: 457–458.
Published online in Wiley InterScience
JLCR
(www.interscience.wiley.com). DOI: 10.1002/jlcr.1193
Short Research Article
14
13
y
The syntheses of [ C] and [ C ]pyromellitic acid
4
NICK SHIPLEY and KENNETH W. M. LAWRIE*
GlaxoSmithKline, Isotope Chemistry, Medicines Research Centre, Gunnels Wood Road, Stevenage SG1 2NY, UK
Received 17 August 2006; Revised 2 November 2006; Accepted 22 November 2006
Keywords: carbon-14; carbon-13; cyanation; oxidation
Introduction
target. We envisaged that a route common to both
isotopomers could be developed by cyanation of tetra-
1
3
14
In the drug development process appropriate salt
selection can be critical. For a recent project a
pyromellitic acid (1,2,4,5-benzene tetracarboxylic
acid), 1, salt proved optimal.
bromo or iodo benzene with either
CN or
CN.
However, under a variety of conditions (CuCN, KCN/
CuI, Zn(CN) /Pd(Ph P) ) no useful product was iso-
2
3
4
lated. An alternative related approach suggested itself
at this point. Oxidation of aryl methyl groups to
carboxylic acids is a well known and straightforward
1
O
O
O
O
process . We found that both 1,2,4,5-tetramethyl
benzene and 2,4,5,-trimethylbenzoic acid could be
oxidized by alkaline permanganate to pyromellitic acid
in good yield (70–80%). This discovery proved crucial.
Cyanation of 2,4,5-trimethyl bromobenzene with
HO
HO
OH
OH
(I)
1
4
Cu[ C]CN in NMP at 1758C followed by saponification
1
4
of the crude [ C]nitrile (NaOH, EtOH, reflux) gave
1
4
2
,4,5-trimethyl[carboxyl- C] benzoic acid in 84% yield
from copper cyanide. Oxidation by slow addition of
aliquots of 5% w/v KMnO in water to a solution of the
We required carbon-14 and stable labelled versions
of pyromellitic acid: the radio-isotopomer at 50–
4
radiolabelled acid in 1.5M KOH at 908C completed the
synthesis. The crude product was triturated with
ether/hexane and crystallized from hot water (60%
yield, radiochemical purity >98%, specific activity
53 mCi/mmol).
6
0 mCi/mmol for traditional ADME work and the stable
label as an internal standard in a LC/MS/MS assay. In
general an effective internal standard requires an
increase of at least 3 a.m.u. over the test substance
and a very high degree of isotopic incorporation
1
3
Our initial plan was to prepare [
4
C ]tetramethylben-
(
typically 50.1% unlabelled). We chose to fulfil these
zene by exhaustive metalation and methylation of
tetrabromobenzene and subsequent oxidation. We
anticipated that a large excess of MeI would be
criteria by mass labelling each of the carboxyl groups
with carbon-13.
1
3
required, but [ C]MeI is inexpensive and readily
available so this was not a major concern. Unfortu-
nately, we were unable to drive the reaction to
Results and discussion
1
3
1
,2,4,5-Tetracyanobenzene is readily saponified to
3
completion and isolated [methyl- C ]2,4,5-trimethyl-
pyromellitic acid. We therefore chose this as our initial
bromobenzene as the major product. However, we had
already demonstrated in the carbon-14 synthesis that
this could be readily converted to pyromellitic acid by a
cyanation, saponification and oxidation sequence.
*
Correspondence to: Kenneth W. M. Lawrie, GlaxoSmithKline, Isotope
Chemistry, Medicines Research Centre, Gunnels Wood Road, Stevenage
13
14
Simply substituting Cu[ C]CN for Cu[ C]CN gave
the desired product with a chemical purity of >98%
and no detectable unlabelled content by MS.
y
Proceedings of the Ninth International Symposium on the Synthesis
and Applications of Isotopically Labelled Compounds, Edinburgh,
1
6–20 July 2006.
Copyright # 2007 John Wiley & Sons, Ltd.
Shipley, Nick
