Abstract-The mass spectra (20 eV electron energy) of a wide range of A5-3B-hydroxy C,, steroid TMS ethers have been examined with the aid of high-resolution mass measurements, together with deuterium and oxygen-18 labelling data. The validity of many previously proposed fragmentation modes has been confirmed. A number of ions regarded as diagnostic have been shown to be less specific than had been formerly supposed. Several novel fragmentations have been observed and investigated.
I N T R O D U C T I O NCHOLESTEROL is present in practically all living organisms, and is the primary source of mammalian steroidal hormones. It may, for example, be metabolised to progesterone via 20,22-dihydroxycholesterol and pregnenolone. Alternatively, complete oxidative removal of the side chain leads to dehydroepiandrosterone (I) and to the C,, hormonal steroids. The A5-3j3-hydroxy steroids are precursors of hormones with the A4-3-one structure.l Consequently, A5-3p-hydroxy C,,-steroids (or their conjugates) are found in quantity only if 3,%hydroxysteroid dehydrogenase is not actively converting them to A4-3-keto steroids. This situation is encountered, for example, in newborn mammals and in certain pathological conditions.2The technique of combined gas chromatography-mass spectrometry (g.c.-m.s.) has been used in the identification of a large number of A5-3j3-hydroxy C,,-steroids, as their trimethylsilyl (TMS) ether derivatives. The following are representative of many samples studied: urine3 to and f a e c e~~~~.~ of newborn and infant humans, meconium of newborn humans,7 human umbilical cord p l a~m a ,~ human amniotic fluid,1° human bile,,, human peripheral plasma,12 to l5 plasma and urine of an eightyear old boy with 3p-hydroxy steroid dehydrogenase deficiency,16 urine of a pregnant irus monkey17 and of a newborn chimpanzee,18 and urine and faeces of female germfree and 'conventional' rats treated with a 3~-hydroxy-A5-oxidoreductase inhibitor.lg G.c.-m.s. has also been used to show that A5-3/3-hydroxy steroids were not excreted by an anencephalic newborn infant.20Despite the clinical significance of androstenols,21 there has been no systematic survey of the mass spectral fragmentations of their derived TMS ethers. We have accordingly examined the low-resolution mass spectra of a number of these compounds, and with the aid of isotope-labelling experiments and high-resolution mass measurements, have elucidated some routes of fragmentation.
925