Squalene Cyclization Citation Schaefer, Phillip Cuthbert (1969) Squalene Cyclization. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/GV74-NT60. https://resolver.caltech.edu/CaltechETD:etd-10032002-124237 Abstract It has been observed that the proton, which is located at C-14 in squalene or C-13 in the "presterol cation," migrates less than 10% of the time (with a confidence level of 74%) to C-20 in lanosterol, and therefore 90% of the time (with a confidence level of 74%) to C-17 in lanosterol. If enzymatically controlled reactions are stereospecific, then at C-13/C17/C-20 in the "presterol cation," there are two sequential 1, 2-migrations during the backbone rearrangement to form lanosterol. A theoretical interpretation of this observation, which reflects on the stereochemistry of the C-18/C-19 double bond of squalene just prior to cyclization, is presented. A theoretical discussion of the minimal requirements for binding squalene-2, 3-oxide to an enzyme in order to obtain complete stereochemical control over the product is presented. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Chemistry) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Richards, John H. Thesis Committee: Unknown, Unknown Defense Date: 9 September 1968 Record Number: CaltechETD:etd-10032002-124237 Persistent URL: https://resolver.caltech.edu/CaltechETD:etd-10032002-124237 DOI: 10.7907/GV74-NT60 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 3881 Collection: CaltechTHESIS Deposited By: Imported from ETD-db Deposited On: 04 Oct 2002 Last Modified: 06 May 2024 21:20 Thesis Files Preview PDF (Schaefer_pc_1969.pdf) - Final Version See Usage Policy. 3MB Repository Staff Only: item control page

SQUALENE CYCLIZATION Thesis by Phillip C: Schaefer In Partial Fulfillment of the Requirements For the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1969 (Submitted September 9, 1968) To Mary Lou iii ACKNOWLEDGEMENTS I would like to thank Dr. N. Davidson and Dr. R. Bergman for their assistance with the completion of this thesis. I would also like to thank Dr. M. Raftery, particularly for his advice during the writing of this thesis. Thanks are also due to my adviser, Dr. J. H. Richards. I would like to thank the graduate students and postdoctoral fellows with whom I have worked. I would especially like to thank Dr. J. T. McFarland and Mr. F. W. Dahiquist for their constant encouragement and intellectual and scientific stimulation. For their financial assistance for these four years, I would like to thank the National Institutes of Health for a Predoctoral Fellowship and Traineeship, the National Science Foundation for a Summer Fellowship, and the California Institute of Technology for a teaching assistantship. Finally, I would like to thank Eileen McKoy for her assistance in the preparation of this manuscript. iv ABSTRACT It has been observed that the proton, which is located at C-14 in squalene or C-13 in the "presterol cation, " migrates less than 10% of the time (with a confidence level of 74%) to C-20 in lanosterol, and therefore 90% of the time (with a confidence level of 74%) to C-17 in lanosterol. If enzymatically controled reactions are stereospecific, then at C-13/C-17/C-20 in the "presterol cation," there are two sequential 1, 2-migrations during the backbone rearrangement to form. lanosterol. A theoretical interpretation of this cbservation, which reflects on the stereochemistry of the C-18/c-19 double bond of squalene just prior to cyclization, is presented. A theoretical discussion of the minimal requirements for bind- ing squalene-2, 3-oxide to an enzyme in order to obtain complete stereochemical control over the product is presented. Ii. VI. D. TABLE OF CONTENTS INTRODUCTION DISCUSSION A. Proton migrations B. Chain folding . RESULTS EXPERIMENTAL A. Chemical degradation of lanosterol B. Chemical synthesis of all trans- squalene-11,14-5H, LO C. Enzyme and biological preparations Materials and methods EH. Data _ REFERENCES . PROPOSITIONS 16 23 31 31 37 39 42 44 48 22 Squalene Cyclization I, INTRODUCTION The complete structure of lanosterol was determined by Ruzicka and co-workers in 1952 (1). Shortly thereafter, Woodward and Bloch (2), and later Dauben (3), advanced a new proposal for the folding of squalene, which accounts for the labelling pattern of cholesterol produced from acetate. This new suggestion replaced an earlier one by Robinson (Figure 1, scheme a), which was not con- sistent with the structure of lanosterol, the newly proposed. pre- cursor of cholesterol. Ruzicka and co-workers (4,5) have also advanced the "biogenetic isoprene rule" and have set forth a set of logically obtained rules for the cyclization of all trans-squalene. Included in this proposalis the second mechanism for the formation of lanosterol from squalene. | - Observations from degradative studies on squalene (6) and cholesterol (7, 8,9,10) have proved consistent with the Woodward- Bloch proposal (Figure 1, scheme b). In the cyclization mechanism set forth by Ruzicka (4,5), the full steroid nucleus (all four rings) is formed synchronously and the resulting cation (see Figure 2) is internally stabilized by C-17 via sigma bond participation (Figure 2, step 1f).. The result of the par- ticipation of C-17 is that the stereochemistry at C-19 is inverted. Subsequently, via a series of uninterrupted Wagner-Meerwein migrations, lanosterol is produced. Squalene Squalene 1 @) HO (a) Cholesterol (b) Cholesterol Figure 1. Proposed labelling patterns in cholesterol derived from carboxyl (+) labelled acetate via (a) Robinson's cyclization mechanism and (b) Woodward and Bloch's cyclization mechanism. 4 Ruzicka's Cyclization Mechanism ~HO > a) H-188 migrates to 198 b) H-14a migrates to 18a HO : a) CH,-158 migrates to 188 ~) | b) CH,-10a migrates to 15a c) H-118 is eliminated Lanosterol Figure 2 5 Van Tamelen's Cyclization Mechanism Squalene-2, 3-oxide ‘partial cyclization +H* a) H-14qa migrates to 15a 2 b) cyclization to spiro cation (charge at C-19} ec) H-188 migrates to 198 d) H-15a migrates to 18a 6 C-10/C-14 bond migrates ta aN C-10/C-15 position ™~ a) CH,-158 migrates to 148 b) CH,-10a migrates to lba c) H-11£ is eliminated Figure 3 Lanosterol _ In step 1d (see Figure 2) of the cyclization, the cyclization is directed so that a secondary cation is produced at C- 14, while an al- ternate cyclization could have led to a tertiary calion al C-15, which might be thought to be more stable and therefore favored. In order to avoid this possibly less favorable secondary cation, van ‘Tamelen, who has recently discovered an intermediate between squalene and lanosterol, i.e., squalene-2, 3-oxide (11), has proposed a third cyclization mechanism (Figure 3) (12). This mechanism di- verges trom that of Ruzicka at step id, where it proposes the forma- tion of a tertiary cation intermediate, which, after rearrangement to another tertiary cation (Figure 3, step 2a) cyclizes to a spiro cation. Thereaiter, two proton migrations and expansion of the third ring bring this mechanism to a point where it is again identical with the one proposed by Ruzicka. It should be noted that although van Tamelen's proposal alle- viates the potential problem of forming a secondary cation instead of a 3° cation, it is no longer a nonstop sequence. Specifically, in step 2a, the C—H bond at C-14 is orthogonal to the developing positive p orbital at C-15. Therefore, the migration of H-14 must be preceded by rotation of the C-14/C-15 bond or addition of a nucleophile at C-15, followed by rotation of the C-14/C-15 bond and displacement of the | nucleophile during step 2a. Additionally, step 3 suffers froma similar difficulty. In the previous step (2d), a proton migrated from the a position of C-15, so that if the methyl group attached to C-15 is to remain in the required 8 position, the migration of the C-10/C-14 (7) This cyclase requires no cofactors and no oxygen (18). (8) All trans-squalene is the biologically active isomer (19). The above observations place certain limits on the mechanism of converting squalene to lanosterol. (1) Except for the exchangeable hydroxyl position, if any proton from water is added to either squalene or 2, 3-oxidosqualene, it must be subsequently eliminated before the presently recognized stable products, 2, 3-oxidosqualene and lanosterol, are formed. Thus, the following intermediates (I, II) would be allowed (12). Figure 2 (2) Although all trans-squalene is the starting form of the Cx chain, a cis-trans isomerization, as by addition of a nucleophile (12), possibly at the cation stage, followed by a rotation and re-elimination of the nucleophile, can not be ruled out. (3) Addition ofa nucleophile, e.g., a part of the enzyme or water, to any double bond except C-22/C-23 is con- .Sistent with the above data. This work will show that most probably the sequence of the hydride migrations in the "presterol" cation is from C-13 to C-17 and C-17 to C-20 and that on theoretical grounds, the trans geometry in the pre C-17/C-20 double bond is required by this sequence of migrations. I. DISCUSSION A. Proton migrations It is known that when squalene produced biosynthetically from mevalonic-4R-4-°H acid, is enzymatically cyclized to lanosterol and cholesterol, one tritium is eliminated, one tritium is found in sterols at the 17-a@ position, one at either C-20 or C-22, one in the fragment C-23 to C-27, one at position 3-a, and one most probably at position d-a ( 20 ). 3. H 4H “A °H ® 3. H —\3a./ HO \?H Squalene _ Presterol Cation I A a Cholesterol Lanosterol Figure 5 10 In figure 5 , the "presumed" actual locations of all the tritium labels are shown. As can be seen, a label appears at position 17-a@ in both the "presterol cation" (I ) and subsequent products. Therefore, it has not been determined whether that tritium, which is at position 17-a in cholesterol, is the same one that was at position 17-a in the "“presterol cation." Although it has been shown that natural squalene has the all trans form (19), isomerization just prior to cyclization, as by a nucleophilic attack and elimination, perhaps during the cyclization, has not been eliminated. However, since the stereochemistry of the product lanosterol is known, the stereochemistry of squalene, just prior to cyclization, can be inferred if the migration sequence at C-13/C-17/C-20 is known. For example consider the outcome of the following migration sequence on all trans-squalene. If the migration sequence is from C-13 to C-20 (i.e., 1,3-migration), itis a reasonable assumption that the stereochemistry at C-17 would remain unchanged. There- fore the proton at C-17 is in the @ position in the "presterol cation," as it is found in lanosterol. If the methyl migrations are of the Wagner-Meerwein type, whereby the stereochemistry at each carbon atom is inverted during each migration, then the proton at C-13 must be in the a position, since C-18 methyl is ultimately in the 6-position. Furthermore, if it is assumed that the positive charge generated at il C-20 is stabilized from the side distal to the C-13/C-17 bond, then migration of the proton at C-13 to C-20 with inversion at C-20 re- sults in a product which is the C-20 epimer of lanosterol. all trans- squalene LL Lanosterol - 7 -20 epimer’ CHs 3. Cc p £ , Lanosterol R HL} Cc migration Figure 6 If the same C-13 to C-20 migration sequence is considered for squalene which has one cis double bond at C-18/C-19, the result is that the product has the same stereochemistry as lanostcrol. There- fore, if a 1, 3-migration pattern were observed in the rearrangement of the "presterol cation," it could be inferred that, just prior to cyclization, the C-16/C-19 bond in squalene was cis. 18, 19-cis-squalene Presterol Cation Hi ! OH, wR 1, 3- Lanosterol <—— : oo migration Figure 7 On the other hand, if two sequential 1, 2-migrations are con- sidered (i.e., proton from C-17 to C-20 and proton from C-13 to C-17), the outcomes are reversed. Now, the stereochemistry at C-17 must be inverted and therefore the proton at C-17 must be in the B 13 all trans-squalene Lanosterol +— ————_ migration 18,19-cis-squalene —» —> | Lanosterol- . — C-20 epimer migration Figure 8 . 14 position in this presterol cation. Tn the case of all trans-squalene, if the same assumptions about the stabilization of the positive charge at C-20 are invoked, migration of the proton from C-17 to C-20 and con- tinued Wagner-Meerwein migrations yield lanosterol as the product. In contrast, such sequential 1, 2-migrations on the "presterol cation" formed from 18, 19-cis-squalene results in a structure identical to lanosterol, except for the sterochemistry at C~20. It should be explained that the stabilization of the positive charge at C-20 is not unusual or new ( 21). Such stabilization is required in the rules for cyclization of all trans-squalene, which were set forth by Ruzicka (4,5). _Inhis case the stabilization was specific- ally internal, i.e., by carbon-carbon sigma bond participation. How- ever, this more general treatment includes internal stabilization as well as possible stabilization by the enzyme involved in the cyclization. The results of an experimental test of this migration sequence will be given in section III. This discussion has reflected mainly on the stereochemistry of the D ring of triterpenes. Some of the stereachemistry of this carbocyclic system is certainly controlled by the enzyme responsible for squalene cyclization. However, it seems likely that the enzyme need not "hold rigidly" the whole squalene molecule in order to com- pletely control the stereochemistry of the whole carbocycle which is formed. 15 \ aplxo-g ‘Z-auayenbg-ST)-6T ‘8T UOTIVABIUL Z [Ore}soueyT =8 H Figure 9 9prxo~¢ ‘z-auaTenog-sued} [O1a}souPT 16 B. Chain folding In the mechanisms for the cyclization of squalene proposed by Ruzicka (4) and van Tamelen (12), the basis of their arguments is analogy to known chemistry. Thus, one of the requirements set forth by Ruzicka (5) is that sequentially migrating groups be trans to each other and coplanar. The mechanism of van Tamelen (12) was proposed to avoid invoking the formation of a secondary cation in preference to a tertiary cation. It is certainly necessary to consider known chemistry when proposing a biosynthetic mechanism, but it would also seem to be use- ful to consider the biological implications of such proposals. One of the functions of squalene cyclase is to "hold the squal- ene chain in the conformation necessary for forming the final prod- uct.'' Possibly it was considered that this meant that all parts of the chain which eventually became part of the carbocyclic structure were held in the "proper orientation." | For the tetracyclic products of squalene, the minimal binding | requirements, a biologically desirable consideration, to exert full stereochemical control of the products have been considered. The results of this consideration will be offered here. | Initially three assumptions must be made: (1) that sequen- ially migrating groups must be trans to each other and coplanar [one of Ruzicka's rules (5 )], (2) that there is a significant energy re- lease by forming a five- or six-membered carbocyclic ring when a double bond is broken (2 carbon-carbon single bonds, 80 kcal each— 17 1 carbon-carbon double bond, 142 kcal; net gain 18 kcal), and (3) that the migrations along the backbone are energetically favorable, e.g., for strain relief (22). Since the epoxide ring in squalene-2, 3-oxide is opened during cyclization, it seems reasonable to assume that this is one of the likely sites for binding to an enzyme. If then the incipient A ring is held ina "‘chair"™ conformation (as much a chair as can be made from a cyclo- hexadiene type structure), then the 19-methyl group in a resulting tetracyclic compound would be in the 8 position, as it is always found. Therefore it will be assumed that the A ring is held in this "chair" conformation. There are two possibilities for the B ring, either incipient chair or incipient boat. First consider the incipient chair structure. Here the C-8 methyl of the cation would be held in the 8 position and the C-9 proton in the @ position, so that if there is to be any backbone migration, these two substituents are in the required trans coplanar relation. At this point, complete steric control over the rest of the cyclization is imposed, for there is only one conformation of ring C which places the C-14 methyl group of the cation trans and coplanar to the C-8 methyl group of the cation. That conformation is incipient chair, which restricts the C-14 methyl group to the @ position and the C-13 proton to the 8 position. 18 Ring A (chair) Rings A and B (chair-chair) CH CH CH CH, 3 CH, 3 3 . 9 8 CH, y R CH, Y 0) 0 H H H R = Cals; | R= C,H, Squalene-2, 3-oxide Squalene- 2, 3-oxide Chair-Chair-Chair-Boat (chair-chair-chair -boat) Cation (C-C-C-B) Figure 10 19 Assume that there will be 1, 2-migrations at C-13/C-17/C-20. Then there is only one conformation for the D ring, which is the incip- ient boat conformation. Thus the C-17 proton of the cation is placed in the @ position. A number of tetracyclic triterpenes have structures which indicate they could have arisen from sucha cation (23). Indeed, the variety is so great that the possibility of a full positive charge being located at C-20 seems highly likely. ‘If both the structures of dammaradienol, dammarenediol I and dammarenediol I (Figure 11) and the fact that these compounds are obtained from the same source ( 22) are considered, it would seem possible that they all arose from the same cation (Figures 10, 11, C-C-C-B) via elimination or hydroxyl capture. The other possibility for ring B is that it is held in an incip- ient boat. In this case, the C-8 methyl would be held in the a posi- tion and the C-9 proton in the 8 position. This requirement forces. ring C to be in the incipient boat form, where the C-14 methyl assumes the 8 position. Again assume that there will be 1, 2-migrations at C-13/ C-17/C-20. Since the C-13 proton is held in the a position, the C-17 proton must be held in the 8 position, which forces ring D into the incipient boat form. | Unfortunately, there are few triterpenes of this structure. Lanosterol and parkeol are the only good examples (Figure 12). 20 Euphorbol Dammarenediol IT (20 &,) ‘Figure li ai Rings A and B (chair-boat) CH, H f CH, Squalene-2, 3-oxide Lanosterol Parkeol Figure 12 2a As will be remembered from the prior discussion on migra- tions, the stereochemistry of C-17 and C-20 are equally dependent on the "folding of the molecule" and the migration sequence at C-13/ C-17/C-20. Whether the migration sequence is 1,2 or 1,3, the folding can be changed to accommodate either mechanism. Since cyclization is thermodynamically favored by 18 kcal per ring formed, and backbone rearrangement is probably stabiliz- ing (e.g., strain relief), then once the A and B rings are formed, the | greatest energy gain can only be realized in a single folding arrange- ment. Therefore, the minimal requirements for binding squalene- 2, 8-oxide to an enzyme need include only holding rings A and B in the proper stereochemical relation to obtain a fully cyclized product (or products) containing only one stereochemical structure. 23 Ili, RESULTS All trans-squalene-11, 14-°H, was synthesized by a previously reported method (24, 25 ) (Figure 13). Some tritium also exchanged into the methyl groups attached to C-10 and C-15, but this was not detrimental to the experiment. In Experiment A (see p. 44) this specifically labelled squalene was ‘ combined with squalene labelled with “4c at positions 1, 5, 9,16, 20 and 24, which had been prepared enzymatically from mevalonic-2-"*C acid. This doubly labelled squalene was converted to sterols with an enzymatically active homogenate of rat liver.. The sterols were separated by thin layer chromatography and lanosterol was extracted. This material was recrystallized, along with added lanosterol, until the *H/“*C ratio remained constant and the *H/“C ratio of the mate- rial in the supernatant from the crystallization was identical to the . ratio of the lanosterol from which it was filtered. This material was converted to lanosteryl benzoate, which was purified by thin layer chromatography. The lanosteryl benzoate was oxidized with chromium trioxide and the resulting acid was methylated with diazomethane. This material was purified by crystallization from ethanol and by thin layer chromatography. At this point only a small number of “C counts remained, so a number about equal to those already present was added as compound HI | (this material was obtained from a previous experiment). Compound III was reduced with zinc and acetic acid to give compound IV. . 24 “H-bT ‘7, -oueTenbg-suedy [[® vamormy “HW -PT ‘TT ~ouerenbs 8oga~ * baw H,Ond-3 ng-IOM S¢q°HO cs : . Pd “HO H,ONg a’oet A od + pdx H; _ A - H.Oug~-4- *$IHO-H, *baHO Hg 25 @COCl CrO, CHN, Py - AcOH - Lanosterol Zn/AcOH Ac,O ¢MgBr AcOH | NBS, hp | N, N-dimethylaniline CrOs AcOH Figure 14 26 Table 1 °H/!C 4 1o Squalene 22.6 +0.7 Lanosterol 16.841.3; 19.6 + 2.3 - Compound II 24.04 2.2 | Compound III + **¢c 12.0+2.2 Compound IV | 14,24 0.8 These data can be interpreted as showing that the amount of tritium at C-11, and therefore at.C-14, in this sample of squalene-11, 14-°H, is between 25.8 + 6.5% and 13.3% + 10. 6% of the total amount of tritium in the sample. 22.6 40.7 5 3 Ls ; -9.8t1.5 22.6 + 0.7 -19.6+2.3 see = 13.3, gx 13.3 = 10.6 3.0+42.4 , Consequently, in lanosterol produced from this squalene, of the total amount of tritium in the sample between 25. 8/(100-25.8) = 34.8% + 9.2% and 13. 3/(100-13. 3) = 15.4% + 13.6% should be at C-17 or C-20. | Much greater confidence is placed in the higher values (25.8% and 34.8%) since these result from counting a sample of lanosterol after it had been purified by thin layer chromatography and crystallized al from ethanol. The other values (13.3% and 15.4% were obtained after the second recrystallization of lanosterol. Since the *y/ “C ratio (16.8 + 4.6) for the material in the supernatant from the second re- crystallization is almost the same as that for the crystalline lanosterol (and it should be much lower if the 5H/"*C ratio of the crystalline pro- duct were lo increase) and the specific activities of the crystalline material and the material from the supernatant are also the same (see p. 44), there is good reason to believe that the material obtained after the first cryslallizalion was homogenous and that the *H/ “C ratio associated with this material was correct. Indeed, the large error in the *H/**C ratio from the twice crystallized material indicates that it could, in fact, be the same ratio as that from the singly crystallized material, which ratio was obtained with greater accuracy. Moreover, when squalene-11, 14-"H, was synthesized by the same method (25), the amount of deuterium at C-11 and C-14, as determined by pmr and mass spectroscopy, was approximately 83% of the total deuterium incorporated in the whole molecule. Therefore, it seems reasonablé to assume that a large amount of the tritium incor~ porated should have been at C-11 and C-14 (51.6%, if the higher values are used). | Additionally, it can be seen that the zinc and acetic acid reduc- tion has not affected the “H/C ratio. | In experiment B (p. 45) a much larger amount of all trans- squalene-11, 14-°H, was converted to sterols in the same fashion. Lanosterol was separated from the resulting mixture. It was con- 28 verted to its benzoate, which was oxidized with chromium trioxide. The oxidation product was methylated with diazomethane. The re- sulting methyl 25, 26, 27-trisnor-3-benzoyloxy-7, 11-diketo-A’ - 24- lanostenoate (I) was purified by thin layer chromatography and two recrystallizations from ethanol. To this material was added enough of the same compound, which was labelled with “*C at positions 1, 7,15, 22, and 31, to obtain a “H/C ratio of about 20. The material labelled with “C was obtain- ed by enzymatically converting d, 1-mevalonic- 2-"*c acid to sterols. The lanosterol obtained from this preparation was treated in exactly the same way as that in the above preparation of specifically tritiated lanosterol. After the *H/ “fC ratio was established, cold material of the same high purity was added to give a workable amount of compound, This now doubly labelled material was reduced with zinc in acetic acid to compound IV. Compound IV was treated with phenylmagnesium bromide two times, since under the reaction conditions employed (re- fluxing in ether, instead of refluxing in benzene) the phenyl ketone [ uv(methylene chloride)A,,, = 240m], which would be expected from the single addition of phenyl erignard, resulted. The resulting di- phenyl carbinol was dehydrated with acetic anhydride/acetic acid to | give a diphenylethylene compound, V, which was recrystallized twice from n-butanol. . | The carefully purified diphenylethylene compound, V, was brominated with N-bromosuccinimide and dehydrobrominated with N, N-dimethylaniline to give a diphenylbutadiene compound, VI. This material was also recrystallized twice from n-butanol. The activities of tritium in these two compounds (V and VI were determined. The data from ten sequential 100 minute counts were combined to obtain the figures given (see p. for complete data). Table 2 | “H(cpm/um) *“H(dpm/ym) + lo + lo Diphenylethylene compound (V) 6.61+ 0.16 20.5 + 0.6 Diphenylbutadiene compound (VI) 6.614 0.13 20.5 + 0.5 The “H/*C ratios are not reported due to the large counting errors for these quantities. Amount of tritium at C-20: 6.61 + 0.16 - 6.614 0.13 | : 13 _ 0.0 = 0.2 or 0.0% + 3.0 6.614 0.16 2 or 0.0% 7 Fraction of tritium at C-14 which migrated to C-20: 6.61 +0,16 - 6.614 0.13 6.61 x 0.348 x 100 = 0.0 + 8.6% Maximum uncertainty due to the uncertainty in the amount _ of tritium at C-14: 0.20. 0.092 —— X —— 100 = 2.3 6.61 (0.3487 ® Total uncertainty fraction of tritium at C-14: - od [ (8. 6%)" + (2.3%)7] ° = 8.9% 30 ; These data show that there could have been as much as 0.0% 8.9% of the tritium, located originally at C-14 in squalene, at C-20 in lanosterol. Thus, within the confidence levels given (i.e., lo) it can be said that 8.9% of the tritium from C-13 of the presterol cation could have migrated to C-20 of lanosterol. Therefore it ap- pears that 91.1% of the tritium from C-13 of the presterol cation ‘should have migrated from C-14 to C-17. _ It is not unreasonable to assume that an enzymatically control- led reaction will be stereospecific. If this is the case, it seems most probable that the tritium originally at C-14 in the presterol cation has. migrated to C-17 in lanosterol and therefore the C-18/C-19 double bond was trans just prior to cyclization. | An even more conclusive experiment could be conducted if enough doubly labelled (*H and “‘C) material could be obtained so that the diphenylbutadiene compound (VI) could be oxidized to the keto compound (vn, from which the proton at C-17 could be exchanged, thereby quantitating the amount of tritium at this position. . $i IV. EXPERIMENTAL A. Chemical degradation of lanosterol 1. Lanosteryl benzoate Lanosterol (10 g, 23 mmol, Aldrich Chem. Co., recrystallized twice from acetone, mp 138-139.5°) was dissolved in 150 ml of ben- zene. The volume was reduced by one-third by distillation. To this was added 25 ml dry pyridine (distilled from barium oxide) and ben- zoyl chloride (25 ml, 0.276 mol). The solution was stirred overnight, after which most of the solvents were removed on a rotary evaporator. Water was added to destroy the excess benzoyl chloride. The product was extracted with three 100 ml portions of ether. The ether was washed repeatedly with water, 10% hydrochloric acid and saturated sodium bicarbonate solution. The ether layer was dried with three 30 ml portions of saturated sodium chloride solution and over mag- nesium sulfate. The ether was removed ona rotary evaporator. The product was recrystallized from methylene chloride-methanol. Yield: 4.5 g (36%), mp 182-186°, second crop, 2.2 ¢ (18%), mp 180-185° [reported mp 191.5° (26)]. | 2. Methyl 25, 26, 27-trisnor-3-benzoyloxy-7, 11-diketo- A’ ~24~lanostenoate (II ) (27, 28, 29) Toa suspension of lanosteryl benzoate (4.5 g, 8.5 mmol) in 20 ml methylene chloride and 80 ml acetic acid was added over thirty ‘minutes a solution of chromium trioxide (10 g, 100 mmol) in 100 ml of 90% acetic acid. The reaction was heated at 55° for three hours. Excess chromium trioxide was destroyed with 10 ml methanol. The 32 majority of the solvents was removed on a rotary evaporator. The products were taken up in 100 ml ether and 300 ml water. The water layer was extracted three times with 100 ml ether. The ether layer was washed repeatedly with 100 ml portions of water. The products were extracted with 10% sodium hydroxide solution, The basic extract was neutralized with concentrated hydrochloric acid and the product was extracted with three 100 ml portions of ether. The ether . was dried with three 30 ml portions of saturated sodium chloride solu- Lion and dried over sudiuim sulfate. The ether was removed on a rota- ry evayorator. Yield: 1.5 ¢. <A slurry of the acid in dry ether was treated with an ethereal solution of diazomethane, generated by treat- ing N-nitroso-N-methylurea (ig, 10mmol) with 3 m1 40% potassium hydroxide solution in the cold and decanting the ethereal solution of diazomethane (30), After three hours, the excess diazomethane was destroyed with acetic acid. The ether was removed on a rotary evap- orator and the product was recrystallized from ethanol. Yield: 0.404 g (8.5 %), mp 187-201°. Two further recrystallizations from ethanol gave 248 mg, mp 192-194.5°, uv (acetic acid) €,,, = 8,300, ir (KBr pellet) C=O at 1675, 1710 and 1735 em™*, C—O—C at 1275 em™ and methyl ester at 1170 cm™*. Analysis: 74.27% C, 8.31% H (Spang), calculated 74.43% C, 8.42% H. 3. Methyl 25, 26, 27-trisnor-3-benzoyloxy-7, 11-diketo- 24-lanostanoate (IV ) (31) Methyl 25, 26, 27-trisnor-3-benzoyloxy-7, 11-diketo-24- 33 lanostenoate (II ) (100 mg, 0.178 mmol) was dissolved in 5 ml acetic acid. The solution was heated to reflux. To this was added zinc dust (100 mg) and seven minutes later an additional portion of zine dust (100 mg) was added. After fifteen minutes of reacting, the solution was filtered and the acetic acid removed ona rotary evaporator. The. product was extracted in three 10 ml portions of ether. The ether was washed with three 5 ml portions of saturated sodium bicarbonate solu- tion, dried with three 5 ml portions of saturated sodium chloride solu- tion and dried over magnesium sulfate, The ether was removed ona rotary evaporator. Yield: 96 mg (95%). Recrystallization from ethanol gave material which had mp 212-215° [reported 220-221° (27)], ir (KBr pellet) C=O at 1695(s), 1708, and 1740 cm™* and C—~O—C at 1275 cm™* and methyl ester at 1170 cm™’. 4, Diphenyl carbinol derivative of IV To a solution of methyl 25, 26, 27-trisnor-3-benzoyloxy-", 1i- diketo-24-lanostanoate (IV ) (295 mg, 0.52 mmol) in 15 ml dry benzene was added . 1 ml of an ether solution of phenyl magnesium bromide (3 mmol, 3 M, Arapahoe). This reaction was heated under reflux for five hours. The salts were decomposed with ice and 20 ml water. The product was extracted with three 20 ml portions of ether. The . organic phase was dried with three 10 ml portions of saturated sodium chloride solution and then over magnesium sulfate. The solvents were removed on a rotary evaporator. The products were chromatographed ona 40 g silica gel column. The fractions eluted with 5% ether in 34 benzene contained the desired product. The eluant from these frac- tions was removed on a rotary evaporator. After recrystallization from ethanol, there was obtained 140 mg (39%), mp 200-210°, ir (KBr. pellet) O—H at 3460 cem™*, C=O at 1695, 1700(s), and 1715(s) em™*, C-—O—C at 1275 cm™* and aromatic bands at 680, 695, 705, 710, and 755 cm™*, | An alternate preparation for the diphenyl carbinol derivative of IV, used in Experiment B (see p. ), is described here. To a solution of methyl 25, 26, 27-trisnor-3-benzoyloxy-7, 11- diketo-24-lanostanoate (IV) (179 mg, 0.32 mmol) in 8 ml dry benzene, from which about 12 ml benzene had been distilled, was added 20 ml dry ether and 0.3 ml phenylmagnesium bromide (3 mmol, 3M, Arapahoe). This reaction was heated under gentle reflux for one hour. The salts were decomposed with 5% sulfuric acid. The product was extracted in three 10 ml portions of ether. The organic phase was » washed with saturated sodium bicarbonate and saturated sodium chloride. The solvents were removed ina stream of nitrogen. _The product was applied to two thick layer chromatography plates (silica gel F,.,, Brinkmann, 20 cm x 20 cm x 2 mm), which were developed in 10% ether in benzene. The product (rf ~ 0.18) was scraped from the plates and eluted with acetone. Yield 144 mg (76%), uv(methylene chloride )A_ _ = 242. This was the phenylketone which ak resulted from the single addition of phenylmagnesium bromide to IV. Of this material, 100 mg (0.17 mmol) was dissolved in 15 ml dry ether. To this was added 0.3 ml phenylmagnesium bromide. The reaction was heated under gentle reflux for one hour. The salts were decomposed with 5% sulfuric acid. The product was extracted with three 15 ml portions of ether. The ether was dried with three 10 ml portions of saturated sodium chloride. The ether was evaporated ina stream of nitrogen. The resulting product was used successfully to prepare the diphenyl ethylene derivative of IV. 5. Diphenyl ethylene derivative of IV A solution of the diphenyl carbinol derivative of IV (140 mg, 0.204 mmol) in 11 ml of acetic acid-acetic anhydride (10:1) was heated under reflux for one hour. Five ml water was added and the solution was allowed to stand for 2 hours. The product was extracted with three 10 ml portions of ether. The ether layer was washed with three 10 ml portions of saturated sodium bicarbonate solution, dried with three 10 ml portions of saturated sodium chloride solution and dried over magnesium sulfate. The ether was removed on a rotary evapora- tor. The product was recrystallized twice from n-butanol. Yield: 60 mg (44%), mp .243-248° [reported 286-287° (27)|, ir (KBr pellet) C=O at 1695 and 1715 em™!, C~O—C at 1272 cm™! and aromatic bands at 695, 700, 710, 760, 770, and 800 cm™, uv (methylene chloride) maximum at 252 mu jreported uv (chloroform) maximum at 255 mu (27)]. This material was homogeneous by thin layer chromatography. 6. Diphenyl butadiene derivative of IV To a solution of the diphenyl ethylene derivative of IV (30 mg, 0.045 mmol) in 5 ml carbon tetrachloride was added N-bromosuccini- 36 mide (9 mg, 0.050 mmol, MCB). The solution was heated and irradiated with a 150 watt light bulb for thirty minutes. The reaction was cooled and the succinimide was filtered from the solution. Carbon tetrachloride was removed on a rotary evaporator. Two ml N, N-dimethylaniline was addcd to the flask and the reaction was heated at 70° for ten minutes. Water (10 ml) was added _ to the reaction and the products were extracted with three 10 ml por- . tions of ether. The ether layer was washed with three 10 ml portions of 10% hydrochloric acid, dried with three 5 m1 portions of saturated sodium chloride solution and dried over magnesium sulfate. The ether was evaporated on a rotary evaporator. The product was recrystal- lized from n-butanol. Yield 10 mg (33%), mp 253-258° [reported mp 236- 237° (27)], uv (chloroform) maximum 310 my [reported uv (chloro- form) maximum 310 mu (27)]. This product was homogeneous by thin layer chromatography. 7. 22,23, 24, 25, 26, 27-Hexanor-3-benzoyloxy-lanosta- 7,11, 20-trione (VIL) To a solution of the diphenyl butadiene derivative of IV (10 mg, 0.015 mmol) in one-half ml methylene chloride and 3 ml acetic acid was added a solution of chromium trioxide (20 mg, 0.20 mmol) in one ml 90% acetic acid. The reaction was heated at 55° for 2hours. The excess chromium trioxide was destroyed with methanol and the solvents were removed on a rotary evaporator. Water (10 m1) was added to the reaction. The product was extracted with three 37 15 ml portions of ether. The ether was washed with two 10 ml portions of water, two 10 ml portions of saturated sodium bicarbonate solution, dried with three 10 ml portions of saturated sodium chloride solution and dried over magnesium sulfate. The solvent was removed on a ro- tary evaporator. The residue was applied to a thin layer chromato- graphy plate, which was then developed with 5% ether in benzene. A band with an rf of 0.1 contained 2.2 mg (0.0045 mmol, 30% yield) of the desired product. This material had a uv (methylene chloride) maximum at 230 my and a very small uv maximum between 280 and 290 my [reported uv (chloroform) maximum at 295 mu (27)]. The ir (KBr pellet) of this material showed C=O at 1705 cm” (very broad), C—O-—C at 1272 cm™ and an aromatic band at 710 cm™’. B. Chemical synthesis of all trans-squalene-11, 14-°H, 1. Tetramethylene-1, 4-bistriphenylphosphonium dibromide (32) To a 250 ml. flask, equipped with reflux condenser and dry- ing tube, was added 90 ml acetonitrile (MCB reagent), triphenyl- phosphine (43.2 g, 165 mmol, MCB, mp 79-80°) and 1, 4-dibromo- butane [17.3 g, 80 mmol, MCB redistilled, bp 63-65° (5 mm)]. The reaction was heated under reflux for two days, after which the solids were dissolved in hot chloroform and precipitated with ether. The product was dried under vacuum at 180° over phosphorous pentoxide for two days. The recovery was 57.7 g, mp 298-300° (reported mp 296~-298° (32)] . ) 38 2. Tertiary-butyl aleohol-1-°H To dry t-butyl alcohol (about 15 ml, distilled frora calcium hydride) was added tritiated water (about 500 me in about 0.1 ml, Nuclear Chicago). After two hours, calcium hydride (2 g, 47.7 mmol) was added to the solution. The solution was heated under reflux for two and one-half hours, after which the t-butyl alcohol-1-"H was distilled into a 100 ml three necked flask. Yield: about 12 ml. 8. Squalene-11, 14~*H, (24, 25) | To the dry tritiated t-butyl alcohol was added potassium (320 mg, 8.2 mmol). After six hours the potassium had reacted completely and tetramethylene-1, 4-bistriphenylphosphonium dibromide (2.8 g, 3.8 mmol) was added to the solution. The flask was flushed with ni- trogen.and then sealed. After sixteen hours, trans-geranyl acetone (1.61 g, 8.4 mmol, 99+ % trans by vpe analysis) was added to the reac- tion. After five hours, a small portion of the reaction medium was removed and diluted to determine the specific activity of t-butyl alcohol, sp. activity = 9,90 mc/mmol, The reaction was quenched by addition of water (15 ml). The aqueous phase was extracted with three 15 ml portions of ligroin (30-60°). The ligroin phase was washed with three 10 ml portions of water, dried with three 10 ml portions of saturated sodium chloride solution and dried over sodium Sulfate. Yield: 48.6 mc. _ The product was applied to a chromatography column of Woelm activity II neutral alumina (60 g). The first four fractions, 39 ligroin (30-60°), contained squalene. These fractions were combined and concentrated. Yield: about 1.5 g. This material was divided into four parts, each of which was applied to a thin layer chromatography plate (10 cm by 20 cm by 2 mm, Brinkmann, Silica gel F,,,). The plates were developed in benzene-ligroin (1:1). The squalene bands (rf 0.7) were scraped from the plates and eluted with ether, acetone, and methanol. The solvents were evaporated in a stream of nitrogen. Yield: about 600 mg. | To the flask containing the squalene was added 100 ml meth- anol, saturated with thiourea, and thiourea (1 g, powdered). After the solution had been stirred in a cold room (4°) for four days, the solid thiourea clathrate was filtered from the solution and it was washed with three 5 ml portions of cold methanol (33). The thiourea clathrate was decomposed with water (75 ml) and the squalene was extracted with three 30 ml portions of ligroin (30-60°). The ligroin layer was dried with three 20 ml portions of saturated sodium chloride solution and over sodium sulfate. Yield: 4 me. C. Enzyme and biological preparations 1. Preparation of rat liver homogenate (16) Livers were excised from male Sprague-Dawley rats (130-165 g), shortly after they had been sacrificed by a blow on the 40 head. * The livers were stored on ice until they were taken to a cold room (4°C), where they were homogenized, usually within thirty min- utes of the time when the rat was sacrificed. The livers were forced through a stainless steel frit, which contained about 50 holes of 1.5 | mm diameter (34}. This material was further homogenized in a tef- lon and glass tissue homogenizer (0. 006-0.009 inch clearance, Arthur H. Thomas Co. ) in two volumes of pH 7.4 potassium phosphate buffer (0.03 M in nicotinamide, 0.004 M in magnesium chloride) per weight of fresh liver. Complete homogenization required three to five strokes. The homogenate was centrifuged at 12, 000 gat 2°. The supernatant was filtered through cheesecloth. Yield: about 15 ml enzyme preparation, S,,, for 10 g of liver. © 2. Preparation of labelled squalene (35) Forty ml of S,, liver homogenate was placed in a 250 ml arlenmeyer flask. Nitrogen was passed through the homogenate until the oxygen content was less than 5% of that in air saturated water, as determined with a YSI Oxygen Monitor. Tor the duration of the reac= tion time, a slow stream of nitrogen was passed through the solution, To the enzyme preparation was added d, 1-mevalonic- 2-"*C acid (2 to ak ; It was observed that homogenate prepared from older rats (400-500 g¢) were more active for synthesis of squalene, but not for synthesis of sterols. 41 3 mg, 11. 5 to 17 uc, Tracerlab) as its N, N’ -dibenzylethylene diamine salt. Each of the following cofactors was added at time = 0, 1, and 2 hours, adenosine triphosphate (59 mg, 0.1 mmol), nicotine adenine dinucleotide, reduced (7 mg, 0.01 mmol), nicotine adenine dinucleo- tide phosphate (4 mg, 0.005 mmol), and glucose-6-phosphate (2 mg, 0.0067 mmol), and at time = 0, sodium fluoride (8.4 mg, 0.5 mmol), _ After three hours incubation on a shaker in a 37° room, the reaction was stopped by the addition of 80 ml 15% potassium hydroxide in 50% ethanol-water. The mixture was hydrolyzed for three hours at 70°. After the hydrolysate was cooled to room temperature, it was extracted with three 50 ml portions of ligroin (30-60°). The ligroin layer was washed with two 25 ml portions of water, dried with three 25 ml por- © _ tions of saturated sodium chloride solution and dried over sodium sulf- ate. The ligroin was removed on a rotary evaporator. Yield: 3.4 uc {about 50% for the dform). The total product was chromatographed on a 2g column of Woelm activity II neutral alumina. The first fraction (ligroin, 10 m1) contained squalene (1.3 uc) and the second fraction (acetone-ether 1:1, 10 m1) contained sterols (2.1 uc). 3. Preparation of sterols from labelled squalene (20) A sample of biosynthetic squalene-"*C (derived from d, 1- mevalonic~2-'"C acid) (25,000 dpm, about 1.2 mum or about 0.5 pg) was added to a 25 ml test tube. The solvent was evaporated in a stream of nitrogen. To this was added lanosterol (100 ug), 1, 2-propanediol - (20 yl) and acetone (50 1). Rat liver homogenate (5 ml) was added to 42, the sample. Each of the following cofactors was added at time = 0, 1, and 2 hours, nicotine adenine dinucleotide phosphate (2,2 mg, 0.0029 mmol) and glucose-6-phasphate (4.5 mg, 0.015 mmol). The sample was placed on a shaker in a 37° room for three hours, after which time it was quenched by the addition of 10 ml 15% potassium hydroxide in 50% ethanol-water, The hydrolysis was carried out at 70° for two hours. After the hydrolysate had been cooled to room temperature, it was extracted with three 5 ml portions of ligroin (30-60°). The ligroin layer was dried with three 5 ml portions of saturated sodium chloride solution and over sodium sulfate. The products were separated by thin layer chromatography on analytical plates (5 cm by 20 cm by 0. 25 mm, Brinkmann, silica gel F,,,), developed in 20% ether in benzene. Yield: lanosterol (6,000 dpm, 24%), cholesterol (10, 000 dpm, 40%). D, Materials and methods 1. Scintillation counting Radioactivity in samples, which contained **C, °H, or both, was determined by scintillation counting in a Packard Tri-Carb Scintil- lation Counter, model 3324, for which the channel settings and gain had been optimized. The efficiency of counting was monitored by using the automatic external standardization unit of this instrument. | Two different scintillation solutions were employed. For analytical work, when samples were checked for purity by thin layer chromatography, samples, which were still deposited on silica. gel, 43 were placed in Bray's solution, a dioxane based solution (36). For all other work, the samples were counted in a toluene based solution, which was prepared by diluting Packard 25X Concentrated Liquid Scintillator (40 ml, Packard Instrument Co.) with reagent toluene to one liter [final solution contained 2, 5-diphenyloxazole (PPO, 5g/1) and 1, 4-bis- (4-methyl-5-phenyloxazoyl)-benzené (dimethyl POPOBP 0.1¢/1)]. To determine “H/‘C ratios, the data obtained from the scintil- lation counter were analyzed on an IBM 7090/7094 computer. The data were analyzed to determine the absolute number of disintigrations per minute (dpm) for each isotope and the ratio of these numbers. Further- more, the statistical error in each number was determined and the error in the *H/““C ratio was determined by calculation of the root mean square of the errors for additive operations and of the percent error for multiplicative operations. 2. Chemicals Adenousine~0d’ -lriphusphale, from equine muscle—-Sigma Chemical Co. | Glucose-8=phosphate disodium salt--Calbiochem. Nicotine-Adenine dinucleotide phosphate monosodium-- Calbiochem. Nicotine-Adenine dinucleotide reduced disodium--Calbiochem., d, 1- Mevalonic-2-"“C acid, dibenzy ethylene diamine salt-- Tracerlab. d, 1-Mevalonic-5-°H acid--Tracerlab. Lanosterol--Aldrich Chemical Co. 44 E. Data Data for Experiment A Wt.(mg) “H(dpm/mg) “H/C Squalene-11, 14-°H, + squalene~1, 5,9, 16, 20, 26-"*c, 22.640. Lanosterol-°H, ““C crystallized once 12645 16.841. Supernatant from above crystal- lization 370415 16.821. Lanosterol- “Hy, ““C crystallized twice , 97+ 4 19.622. Supernatant from above crystallization 89 +5 16.824. Lanosteryl benzoate | (60 mg lanosteryl benzoate added) 39.7 22.042. Compound III 33.4 24.042. Compound III + II - “*c (67 mg unlabelled compound II | _ added) 12.022. Compound IV 100 14.2+0. 45 Data for Experiment B Theat *3(dpm) Wt.(mg) (dpm) *H(dpm mg Squalene-11, 14-"H, Sterols . Lanosterol Lanosteryl benzoate Lanosteryl benzoate + — unlabelled lanosteryl benzoate Compound III © Compound III, crystallized Compound III, recrystallized Compound III + unlabel- -led Compound III Compound IV Compound Vv Compound VI 1.8x10 2.6 x 10° 2.4x 10° 1.3x10 3x10 oo ee 1.1x 10° 8.1x 10° 8.1x 10° 7.0.x 10° 500, 100 5x 10° 107 1200 \ 507 384 150 366 80 13928 71 «115#4 201 40.341.38 22.74+0.7 187 =38.7+0.6 21.940.3 16.7 30.640.9 20.540.6 3.5 30.640.7 20.54+0.5 46 Counting data (100 minute counts) used to determine the specific activities of compounds V and VI. Count Number Blank Compound V Compound VI 1 1125 _ 1814 2093 2 1122 1883 (1953) 3 1115 1886 2027 4 1142 —«-: 1898 2049 5 1145 1871 2023 6 1096 1889 2094 7 1127 1894 2007 8 1089 1906 2068 9 1134 1817 | 2028 10 1128 1851 — 2022 11 — - 2108 11223 18709, 20519 Average 1122 1871 2052 Sum of deviation? 2057 9741 11813 Average deviation 296 974 1181 Standard deviation - 18.1 32.9 34.4 Note: Count # 2 for Compound VI was not used since it was 2.88 standards from the mean. 47 Specific activities of compounds V and VI. Gross Net sample . Counts Counts Counts | Weight mg Compound V 18709+137 74862173 0.76 mg 98502233 Compound VI © 205192143 92964178 0.94 mg 9889+ 196 Counts CPM . DPM um im Lm Compound V 66064156 6.61+0.16 20.52+0.6 Compound VI 66124131 6.6140.13 20.52+0.5 Note: The efficiency of tritium counting was 32. 28% + 0.46%. Although, theoretically, there were a few “C counts, there were so few (less than 0.2 cpm in the tritium channel, where the “Cc efficiency was 17%) that they were neglected for these calculations. Since the counting efficiency in the two samples, as measured by the automatic external standardization unit of the Packard Scintillation Counter was the same, it is sufficient to compare the CPM/um, where the error is somewhat less, since the error in determining the tritium counting efficiency does not appear in these figures. .An error of 0, 5% in the sample weight was allowed, which is greater than the 2.5 ug error (5 mg range) specizied by the manufacturer of the balance, Cahn. 10. 11, 12. 13. W 48 REFERENCES . Voser., M. V. Mijovic, H. Heuser, O. Jeger, and L. Ruzicka, Helv. Chim. Acta, 35, 2414 (1952). . B. Woodward and K. Bloch, J. Am, Chem. Soc., 75, 2023 (1953). . G. Dauben, S. Abraham, S. Hotta, I. L. Charkoff, H. L. Bradlow, and A. H. Soloway, J. Am. Chem. Soc., 15, 3038 (1953). L. Ruzicka, Experientia, 9, 357 (1953). . Eschenmoser, L. Ruzicka, O. Jeger, and D. Arigoni, Helv. Chim. Acta, 38, 1890 (1955). . W. Cornforth and G. Popjak, Biochem. J., 58, 403 (1954), . W. Cornforth, G. D. Hunter, and G. Popjak, Biochem. J., 54, 590 (1953). . W. Cornforth and I. Y. Gore, Biochem. J., 65, 94 (1957). K. Bloch, Helv. Chim, Acta, 36, 1611 (1953). Ww . G, Dauben and K. H. Takemura, J, Am. Chem. Soc., 75, 6302 (1953). E. E. van Tamelen, J. D. Willett, R. B. Clayton, and K. E. Lord, J. Am, Chem. Soc., 88, 4752 (1966). ._ E, van Tamelen, J. D. Willett, and R. B. Clayton, J. Am. Chem. Soc., 89, 3371 (1967). . 8. Goodman, J. Biol. Chem., 236, 2429 (1961). 14. 15, 16. 17. 18. 19. 20. al. 22. 23. 24. 49 J. D. Willett, K. B. Sharpless, K. E. Lord, E. E. van Tamelen, and R. B. Clayton, J. Biol, Chem., 242, 4182 (1967). R. K. Maudgal, T. T. Tchen, and K. Bloch, J. Am. Chem. Soc., 80, 2589 (1958). | J. W. Cornforth, R. H. Cornforth, A. Pelter, M. G. Horning, and G. Popjék, Tetrahedron, 5, 311 (1959). D. Arigoni, Experientia, 14, 153 (1958). P. D. G. Dean, P. R. Oritz de Montellano, K. Bloch, and E. J. Corey, J. Biol. Chem., 242, 3014 (1967). N. Nicolaides and F. J. Laves, J. Am. Chem. Soc., 76, " 2596 (1954). J. W. Cornforth, R. H. Cornforth, C. Donninger, G. Popjak, Y. Shimizu, S. Ichii, E. Forchielli, and E. Caspi, J. Am. Chem. Soc., 87, 3224 (1965). J. H. Richards and J. B. Hendrikson, "The Biosynthesis of Steroids, Terpenes, and Acetogenins,'" W. A. Benajmin, Inc., New York, 1964, pp 240-278. H. W. Whitlock, Jr., and M. C. Smith, Tetrahedron Lett., 1968, 821. | G. Ourisson, P. Crabbé, andO. R. Rodig, "'Tetracyclic Triterpenes, ' Holden-Day, Inc., San Francisco, 1964, pp. 1384-180. R. E. Wolff and L. Pichat, Compt. rend., 246, 1868 (1958). 25. 26. 27. 28. 29. 30. 31. 32. 33. 34, 35. 36. 50 Phillip C. Schaefer, ''Candidacy Examination, " California Institute of Technology, Pasadena, California, 1966. I. N. Heilbron, A. H. Cook, H. M. Bunbury, and D. H. Hey (editors), "Dictionary of Organic Compounds," 4th edition, Oxford University Press, New York, 1965, p. 2000. J. F. McGhie, M. K. Pradhan, J. F. Cavalla, and S. A. Knight, Chem. and Ind., 1951, 1165. W. Voser, O. Jeger, and L. Ruzicka, Helv. Chim. Acta, 35, 497 (1952). | | W. Voser, O. Jeger, and L, Ruzicka, Helv. Chim, Acta, 30, 503 (1952). A. H. Blatt (editor), "Organic Syntheses," Coll. Vol. 2, John Wiley & Sons, Inc., New York, 1943, p. 165. C. Dorée, J. F. McGhie, and F. Kurzer, J. Chem. Soc., 1948, 988. L. Horner, H. Hoffmann, W. Klink, H. Ertel, and V. G. Toscano, Chom. Ber., 85, 581 (1062). OQ. Isler, R. Riiegg, L. Chopard-dit-Jean, H. Wagner, and K, Bernhard, Helv. Chim. Acta, 39, 897 (1956). _G. Popjdk, Private Communication, 1967. D. S. Goodman and G. Popjdk, J. Lipid Res., 1, 286 (1960). G. A. Bray, Anal. Biochem. , 1, 279 (1960). ol VI. PROPOSITIONS Abstracts of Propositions I. Itis proposed that 1, 2-(a-ketotetramethylene)ferrocene, which can be resolved into d and 1 forms, be investigated as a photochemi- cal sensitizer which might induce residual optical activity in a race- mate by causing the d and 1 forms to undergo photochemical trans- formations at different rates. Ii. An epoxide is proposed as an intermediate in the enzymatic oxidation of 8-carotene to retinal. II. It is proposed that the mechanism of the opening of the 9, 19- cyclopropane ring of cycloartenol be investigated by observing the fate of a tritium label, which would be placed at C-8 by an enzymatic synthesis. Iv. fit is proposed that the mechanism of squalene cyclization be investigated by preparation of 20, 21-dehydrosqualene and its enzy- matic conversion to tetracyclic compounds. V. Steroids, which are prepared from specifically deuterated squalenes, are proposed as substances for the further investigation of the mass spectra of steroids. 52 PROPOSITION I In 1965 Hammond and Cole (1) showed that an optically active sensitizer (I) could induce residual optical activity in a racemate (Ii), by causing the d and 1 isomers to undergo photochemical transforma- tions at different rates. Since that time, there has been no report of ok CH,CHNHOCCH, i ¢ p H I d,1-trans-1, 2-diphenylcyclopropane (IL) any other sensitizers which have been used to the same end success- fully. It is proposed that a large number of compounds derivable from 1, 2-(a-ketotetramethylene)-ferrocene (II), which is resolvable into d and 1 forms (2), be investigated as photochemical sensitizers. This O- WO Fe Oo | Fe (+) - Ill . . (-) - I compound offers an unusual variety of optically active derivatives, obtainable without risk of stereochemical inversion. Ferrocene has been shown lo sensitize the cis-trans isomeriza~- tion of piperylenes and there is some evidence that this isomerization a3 proceeds via a complex between the sensitizer and substrate (3,4, 5). Furthermore, there is crystallographic evidence that, in the complex between tetracyanoethylene (TCNE) and ferrocene, TCNE is planar to and situated above one of the five-membered rings of ferrocene (6). Therefore, it is possible that other complexes would form in this fashion (rather than directly to the d-orbitals of the iron atom), there- by facilitating close interaction between the assymetric center of Ii and a substrate (e.g., I). Although complexes may form to both rings of ferrocene, to the extent that one forms to the substituted ring, some induction of opti- cal activity via photochemical sensitization may be observed. Addi- tionally, such a substrate-sensitizer complex has been postulated in the sensitized stereochemical racemization of IV by naphthalene (7). O A 8 “CH, R IV R= CH,, Br, OH Even if the parent compound (III) proved to be an inefficient sensitizer, it would be possible to prepare other derivatives fe. Bes V(2)], which might be more suitable for such investigations. Or” Fe OC) Vv a4 Although it might seem that for many photochemical reactions there is insufficient energy available from the first triplet of ferro- cene (about 43 kcal/mol) (3), there has been a report of an emission from higher excited states of ferrocene (5,10). More importantly, ferrocene sensitizes reactions (3) which require about 60 kcal/mol (3). Therefore, it seems possible that the proposed compounds (II, V and related compounds) might serve as seusilizers, which could be used to further investigate the interactions between sensitizers and substrates in photochemical reactions. 10. 95 References G, 8, Hammond and R, S, Cole, J, Am, Chem, Soc., 87, 3256 (1965), K. Schlégi, Fortschr. Chem. Forsch., 6, 479 (1966) and references therein. J. J, Dannenberg and J. H. Richards, J, Am. Chem. Soc., 87, 1626 (1965). | J. J. Dannenberg, Ph.D. Thesis, California Institute of Technology, 1967. J. P. Guillory, C. F. Cook and D. R. Scott, J. Am. Chem. Soc., 89, 6776 (1967). E. Adman, M. Rosenblum, S. Sullivan, and T. N. Margulis, J. Am. Chem, Soc., 89, 4540 (1967). 2958 (1968). R. S. Cooke and G. S. Hammond, J. Am. Chem. Soc., 20, K. Mislow and A. J. Gordon, J. Am. Chem. Soc., 85, 3521 (1963). P. J. Wagner, J. Am. Chem. Soc., 89, 2820 (1967). D. R. Becker and R. S. Scott, J, Chem. Phys., 35, 516 (1961). 36 PROPOSITION II DeWitt 8. Goodman has published a series of articles about 6-carotene (I), retinal, retinoic acid, vitamin A (II), the biosyntheses of these compounds and their biological distributions (1-7). A pro- posed mechanism for the biological conversion of 8-carotene to retinal has been supported in this work (4). R__7....0 aN H L H R i Figure 1. Goodman's Mechanism (4) Purification of the enzyme responsible for this oxidation was not possible (6). It is proposed that an intermediate epoxide (III) exists and that more than one enzyme is required for the conversion of B-carotene to retinal. Further, il is pruposed that 15, 15’ -oxide-g-carotene be synthesized by the method outlined (8) in Figure 2, and that this 57 YQ AQ ANS QR OER oO Ol compound be tested for its biological activity. CHO + [(CH,),N],P ——> I Retinal Figure 2 D D ON D8 References | . S, Huang and D, 8. Goodman, J. Biol. Chem., 240, 2839 (1965). . 8, Goodman, H. S. Huang, and T. Shiratori, J. Lipid Res., 6, 390 (1965). . §. Goodman and H. S. Huang, Science, 149 879 (1965). ere, . 8. Goodman, H. 8. Huang, and T. Shiratori, J. Biol. Chem., 241, 1929 (1966). . S. Goodman, R. Blomstrand, B. Werner, H. S. Huang, and T. Shiratori, J. Clin. Invest., 45, 1615 (1966). . S, Goodman, H. S. Huang, M. Kanai, and T. Shiratori, 5. Biol. Chem., 242, 3543 (1967). . H. Fidge, T. Shiratori, J. Ganguly, and D. S. Goodman, J. Lipid Res., 9, 103 (1968). . Mark, in E. J. Corey (ed.), "Organic Synthesis, '' Volume 46, John Wiley and Sons, Inc., New York, 1966, pp 31-33. 09 PROPOSITION III Cycloartenol (I) serves as the common intermediate between trans-squalene-2, 3-oxide and plant sterols (1-4). However, there is a major structural distinction between cycloartenol and most phyto- sterols, that being the presence of a 9, 19-cyclopropane ring. 3 H It is known that cycloartenol, biosynthesized from mevalonic-~ 4-R-4-5H, 2-*C acid, retains six tritium lahels (5). One of these labels is at C-8 (5). It is proposed that the mechanism of the opening of the 9, 19- cyclopropane ring be investigated by observing the fate of the tritium at C-8, This can be readily done by studying *H/"*C ratios in other phytosterols (II-V), which can be obtained from the same sources. as cycloartenol (6). H 24-Methylenecycloartenol (II) 24-Methylenecycloeucalenol (IIT) 60 H 24-Methylenelophenol (IV) 24-Ethylidenelophenol (V) In the lophenols, it might be expected that the tritium from C-8 would be at C-9 (via 1, 2-migration), at C-19 (via intra- or inter- molecular transfer) or to be absent from the product (via elimination or abstraction). If the label remains in the product, oxidation with N-bromo- succinimide to the 7, 9(11)-diene would show whether the label were at C-9. This is not an unexpected possibility, since acid treatment of cycloartenol opens the cyclopropyl ring, generating a cationic center at C-9. Presence of the label at C-19 requires a highly strained con- formation for proton transfer , and therefore is unexpected. Absence of this label might indicate that a new intermediate (VI) exists. HO VI 61 References P. Benveniste, L. Hirth, and G. Ourisson, Phytochemistry, 5, 31 (1966). | P. Benveniste, L. Hirth, and G. Ourisson, Phytochemistry, 5, 45 (1966). P. Benveniste and R. A. Massy-Westropp, Tetrahedron Lett., 1967, 3553. | H. H. Rees, L. J. Goad, and T. W. Goodwin, Tetrahedron Lett,, 1968, 723. H. H. Rees, L. J. Goad, and T. W. Goodwin, Biochem. J., 107, 417 (1968). M. von Ardenne, G. Osske, K. Schreiber, K. Steinfelder, and R. Tiimmiler, Kulturpflanze, 13, 101 (1965). 62 PROPOSITION IV The most recently proposed mechanism for the cyclization of squalene-2, 3-oxide (I) is the one of van Tamelen (1), in which a spiro intermediate cation (II) is supported instead of Ruzicka's cation (II) Ruzicka It is proposed that 20, 21-dehydro~squalene (IV) be synthesized and converted enzymatically in a stepwise fashion to its terminal epoxide and its cyclized product (V or VI). As C-20 becomes a full cation, it is quite probable that the charge would become delocalized over the available 7-system (C-22 to C-25) of the substituted butadiene. Quenching of this cation at C-25 by water would lead to a neutral product (V or VI). If VI were the observed product, then van Tamelen's proposal would be strongly indicated. However, if V were the observed 63 product, it would seem that Ruzicka's original proposal was correct. In the proposed synthesis, reaction 1 has been reported pre- viously (3). 5-Bromo-2-pentanone is readily available (4). Com- pounds which contain ketone functions and bromine can be converted to ketals without affecting the bromine (5), and ketals are not affected by Wittig reactions (6). Although step three of the proposed synthesis of IV will gener- ate cis and trans isomers, it is possible that the cis isomer will cycl- ize to a cyclohexadiene, making it readily separable from the trans isomer. However, the inclusion of some cis isomer would not be a serious drawback, since it is expected that either bond isomer in the final product would be usable (7). CH2OH 64 ys 2 a Al(O-4-Bu)3 Sy Br (HOGHg)o O3P ¥ 5 IV Figure 1 oh AAO @) tots p-BuLi > 9p OOS 3 -O 90 Laman “ny ) 0: 65 aerobic incubation with IV + squalene-2, 3-oxide > IV-2, 3-oxide , rat liver homogenate anaerobic incubation with IV-2, 3-oxide —n o rat liver homogeaate or HO ’ Figure 1 (continued) ao ee WD bw Q Bono E 66 References . E. van Tamelen, J. D. Willett, and R. B. Clayton, J. Am. Chem. Soc., 89, 3371 (1967). Ruzicka, Experientia, 10, 357 (1953). A. Braude and C. J. Timmons, J. Chem. Soc., 1950, 2007. . dager and R. Keyner, Arch. Pharm., 293, 896 (1960). Djerassi, ed., "Steroid Reactions, ' Holden-Day Inc., San Francisco, 1963, p. 8. . Djerassi, ed., "Steroid Reactions, '' Holden-Day Inc., San Francisco, 1963, p. 2. | . J. Corey andS. K. Gross, J. Am. Chem. Soc., 89, 4561 (1967). 67 PROPOSITION V The mass spectra of steroids has been a topic of interest in the past few years. Thus far the mass spectra of steroids have been used as fingerprints for these compounds. The major problem in analyzing the mass spectrum ofa steroid is the lack of specific knowledge con- cerning fragment formation during ionization. The bulk of information about fragmentation has been obtained by qualitative comparison of mass spectra of known compounds of similar structure (1, 2,3, 4). Keto-steroids have been reasonably well investigated by sub- stituting deuterium at exchangeable positions in the molecule by Djerassi and co-workers and others (5, 6, 7, 8,9,10,11,12,13). Anal- ysis of the spectra of these deuterated keto-steroids has allowed Djerassi to make considerable inroads into the problem of fragment formation. However, the important field of steroidal alcohols has been virtually untouched, except for comparison of spectra of similar compounds. Many interesting problems can be solved by using prop- erly labelled steroids, but due to lack of suitable functionality, deut- eration is not easy and requires much more chemistry than is neces- sary with steroidal ketones. Since the discovery of the intermediacy of 2, 3-oxido-squalene in the conversion of squalene to lanosterol (14), it is much easier to control the products obtained; for example, by using an anaerobic incubation, only lanosterol will be obtained, even with a crude enzyme preparation. 68 It is proposed that numerous deuterated squalenes and their 2, 3-oxides be prepared and be converted enzymatically to lanosterol. The mass spectra of such specifically deuterated steroids would per- mit many problems associated with fragment formation to be solved. Many positions in the steroid nucleus are available for deuteration, via synthetic squalene (Figure 1). ?) Na0D/D,0 ’ 0) PCDCOBr SN —tHop App ? 3 1B ? 5 . p cD. g | . coy CH.CO-E t [)NGOE?/Et0D y 2) KOD/D.O CDoBr $,P=CDCD.CD,C D=PQ, {Bu00 , NBS/+Bu0H, f KC Os/eoHl, 10. 11. 12. 70 References H. Budzikiewicz, C. Djerassi, and D. H. Williams, "Structure Elucidation of Natural Products by Mass Spectroscopy," Holden-Day, Inc., San Francisco, 1964, Vol. Tl, pp. 25-110. K. Biemann, "Mass Spectroscopy, Organic Chemical Applica- tions, '' McGraw-Hill, New York, 1962, pp. 334-48. 8S. 8. Friedland, G. H. Lane, Jr., R. T. Longman, K. E. Train, and M, J. O'Neal, Jr., Anal. Chem., 31, 169 (1959). L. Peterson, Anal, Chem,, 34, 1781 (1962). D. H. Williams, J. M. Wilson, H. Budzikiewicz, and C, Djerassi, J. Am. Chem. Soc., 85, 2091 (1963). R. H. Shapiro, D. H. Williams, H. Budzikiewicz, and C. Djerassi, J. Am, Chem. Soc., 86, 2837 (1964). J.E. Gurst and C. Djerassi, J. Am. Chem. Soc., 86, 5542 (1964). H. Budzikiewicz and C. Djerassi, J. Am. Chem. Soc., 84 1430 (1962). Z. Pelah, D. H. Williams, H. Budzikiewicz, and C. Djerassi, J. Am. Chem. Soc., 87, 574 (1965). - H. Audier, A. Diara, M. deJ. Durazo, M. Fétizon, P. Foy, and > W. Vetter, Bull. Chim. Soc. Fr., 1963, 2827. H. Audier, M. Fétizon,and W. Vetter, Bull. Chim. Soc. Fr., 1964, 415, J. L. Courtney and J, S. Shannon, Tetrahedron Letters, 1963, 13. 71 13. J. S. Shannon, C. G. Macdonald, and J. L. Courtney, Tetrahedron Letters, 1963, 173. 14. E, E. van Tamelen, J. D. Willet, R. B. Clayton, and kK. E. Lord, J, Am, Chem. Soc., 88, 4752 (1966).