JOURNAL OF MASS SPECTROMETRY
J. Mass Spectrom. 33, 1032 (1998)
JMS Letters
Dear Sir,
concentrations ranging between 40.6 and 373 nmol l~1. Using
the method under the conditions of a reference method,6 we
were able to satisfy the proposed analytical speciÐcations, i.e.
a maximum total analytical error of 3%, a maximum total
imprecision (n \ 6) of 2% and a maximum bias of 0.9%. From
these results, we recommend the use of TMCS in MeOH for
the methylation of thyroxine prior to perÑuoroacylation and
quantitative IDÈGC/MS analysis.
On the use of trimethylchlorosilane in methanol for methylation
of thyroxine prior to perÑuoroacylation and isotope
dilution–gas chromatography/mass spectrometry
For gas chromatographic analysis, thyroxine is most
commonly derivatized in a two-step reaction to the N,O-bis
(
perÑuoroacyl) methyl ester.1h4 Usually, the methyl ester is
formed from reaction with methanolic HCl, which is gener-
ated from introducing gaseous HCl in dry MeOH or from
hydrolysis of acetylchloride dropped into MeOH. However,
in our hands, this methylation reagent posed problems, in
particular because we applied it for the quantitative analysis
of thyroxine in serum by isotope dilutionÈgas
chromatography/mass spectrometry (IDÈGC/MS). We
obtained irreproducible isotope ratios, presumably owing to
the uncontrollable concentration of HCl in the reagent (e.g.
dependent on the moisture content of the MeOH and on the
conditions of the in situ hydrolysis of acetyl chloride) and/or
the formation of byproducts. Sometimes, the reagent even
caused cleavage of the phenyl ether bond. This was evidenced
by the presence of unlabeled thyroxine after derivatization of
Acknowledgements
This work was supported by the Belgian Government through
Contract No. 12050690 and the Fund for Medical ScientiÐc
Research through Contract No. 3.30007.91.
Y ours,
V. I. DE BRABANDERE, D. ST O CKL,¤
L. M. THIENPONT¤ and A. P. DE LEENHEER*
Laboratoria voor Medische Biochemie en Klinische Analyse,
Faculteit van de Farmaceutische Wetenschappen,
Universiteit Gent,
pure [13C ]thyroxine in the presence of diiodotyrosine used
6
as carrier (during derivatization the 13C -labeled tyrosine
6
moiety was thus exchanged for unlabeled diiodotyrosine).
Harelbekestraat 72,
Because of these problems, we developed a methylation pro-
cedure with diazomethane followed by perÑuoroacylation.5
Although the alternative methylation procedure allowed us to
quantify thyroxine in serum reliably with IDÈGC/MS, it was
not fully satisfactory. The reasons for this were that methyl-
ation with diazomethane resulted in a di†erent reaction
product, i.e. the O-methyl-N-perÑuoroacyl methyl ester deriv-
ative and lower GC/MS sensitivity (signal-to-noise ratio) and
also diazomethane is very hazardous and toxic. Because of
these problems, we continued to search for another methyl-
ation procedure.
B-9000 Ghent,
Belgium
References
1. D. B. Ramsden, in Mass Spectrometry, edited by A. Lawson, p.
6
23. Walter de Gruyter, Berlin (1989).
. B. Moller, O. Falk and I. Bjorkhem, Clin. Chem. 29, 2106
1983).
. D. B. Ramsden and M. J. Farmer, Biomed. Mass Spectrom. 11,
21 (1984).
. L. Siekmann, Biomed. Environ. Mass Spectrom. 14, 683
1987).
5. L. M. Thienpont, V. I. De Brabandere, D. Sto
2
3
4
(
4
Here we report the methylation of thyroxine with a reagent
consisting of trimethylchlorosilane (TMCS) in methanol
(
ckl and A. P. De
(
MachereyÈNagel, Du
ren, Germany). Although this reagent is
Leenheer, Biol. Mass Spectrom. 23, 475 (1994).
6. L. M. Thienpont, C. Franzini, J. Kratochvila, J. Middle, C. Ricos,
known as a methylation agent, to the best of our knowledge it
has never been described for the derivatization of thyroxine.
The optimized methylation conditions consist of reaction in
L. Siekmann and D. Sto ckl, Eur. J. Clin. Chem. Clin. Biochem.
3, 949 (1995).
3
100 ll of methanol and 30 ll of TMCS at 70 ¡C for 1 h. The
advantages of the proposed methylation procedure are multi-
ple: it is based on the same methylation reagent (methanolic
HCl) as used before,1h4 and thus also results in the N,O-
bis(perÑuoroacyl) methyl ester; however, the in situ pro-
duction of HCl from TMCS in MeOH is easy and user-
friendly to perform and apparently happens in a more
controlled way, since reproducible isotope ratios are obtained;
compared with methylation using diazomethane, it results in
better GC/MS sensitivity.
* Correspondence to: A. P. De Leenheer, Laboratoria voor
Medische Biochemie en Klinische Analyse, Faculteit van de Farm-
aceutische Wetenschappen, Universiteit Gent, Hardbekstraat 72,
B-9000 Gent, Belgium.
Contract/grant sponsor: Belgian Government; Contract/grant
number: 12050690.
Contract/grant sponsor: Fund for Medical ScientiÐc Research;
Contract/grant number: 3.30007.91.
¤
Present address: Laboratorium voor Analytische Chemie, Facul-
We have used TMCS in methanol in our IDÈGC/MS
method5 for thyroxine quantiÐcation in 31 human sera with
teit van de Farmaceutische Wetenschappen, Universiteit Gent, Hare-
lbekestraat 72, B-9000 Gent, Belgium.
CCC 1076È5174/98/101032È01 $17.50
Received 20 July 1998
Accepted 22 July 1998
(
1998 John Wiley & Sons, Ltd.
De Brabandere
