Applications of mass spectrometry in the steroid field.

Among the various physical methods enjoying wide use by organic chemists, mass spectrometry is the most recent and probably also the most spectacular addition. It is not surprising, therefore, that it has also found rapid acceptance in the steroid field. Indeed, many of the recent generalisations about mass spectral fragmentation mechanisms of organic molecules were first established by model studies in the steroid series. The present lecture deals with the following aspects by employing androstanes, sterols and steroidal sapogenins as suitable illustrations: (a) Use of isotopically labelled steroids for elucidation of mass spectral fragmentation mechanisms. (b) Use of molecular ion peak determinations in the sterol field. (c) Use of specific and characteristic mass spectral fragmentation processes for structure determination of unknown steroids. INTRODUCTION Mass spectrometry is the most recent physical method which has found wide acceptance in the steroid field. Whereas ten years ago it was difficult to find any papers in the steroid literature citing the use of mass spectrometry, it is equally difficult in 1969 to encounter articles dealing with steroid chemistry in which mass spectrometry is not at least used for analytical purposes and frequently for partial or even complete identification purposes. The literature up to 1964 dealing with the role of mass spectrometry in the steroid field has been completely reviewed in a book from our laboratory1 but since then a veritable flood of articles has appeared in which the use of mass spectrometry in the steroid area has been emphasized. Unfortunately, a summary or review of this research since 1964 is sorely lacking. The scope of mass spectrometric applications among steroids can be well illustrated by simply selecting a few pertinent papers from the last two volumes (1968) of Steroids. In them, mass spectrometry was employed to determine the sterol composition of human faeces2, of algae3' ', of human metabolites of the oral contraceptive norgestrel,5 and to study the typical fragmentation patterns of such diverse steroid types as 3,6,20-trioxygenated pregnanes6, unsaturated (z14 and 4J5) ketals7, and various adrenal cortical steroids8. Another illustration of the actual and potential importance of mass spectrometry in the clinical area is afforded by the very substantial Part 191 in the Stanford series 'Mass Spectrometry in Structural and Stereocoemical Problems'. 205 CARL DJERASSI number of articles9 dealing with the combined use of gas liquid chromatography and mass spectrometry of trimethylsilyl derivatives of steroids. Even this very cursory coverage of the contents of one single journal for one year yields a potpourri which defies adequate coverage within the confines of one lecture. Therefore, I have limited myself largely to a description of some recent research in our laboratory which illustrates the methodology and basis on which ultimately most applications of mass spectrometry in organic chemistry in general, and in the steroid field in particular, and must rest. DETERMINATION OF EMPIRICAL FORMULA The best known and most easily comprehensible use of mass spectrometry pertains to the information extractable from the molecular ion region of a mass spectrum. Thus, high resolution mass spectral measurements will yield the precise empirical formula, which in the sterol field can never be obtained unambiguously by microanalysis (e.g. Calcd. for C27H460: C, 8387; H, 1199. Calcd. for C28H480: C,8393; H, 12.08). Even more striking is the fact that such information can be deduced with equal facility from an impure sample and that mass spectrometry is frequently the tool par excellence for the analysis of mixtures on a sub-milligram scale. A typical example is shown in Figure 1, which contains the high mass range of the spectrum ChoLesterol 100 386 (C27 H460)