UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies in steroids and alkaloids Vlattas, Isidoros 1966

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1966_A1 V5.pdf [ 8.17MB ]
Metadata
JSON: 831-1.0062216.json
JSON-LD: 831-1.0062216-ld.json
RDF/XML (Pretty): 831-1.0062216-rdf.xml
RDF/JSON: 831-1.0062216-rdf.json
Turtle: 831-1.0062216-turtle.txt
N-Triples: 831-1.0062216-rdf-ntriples.txt
Original Record: 831-1.0062216-source.json
Full Text
831-1.0062216-fulltext.txt
Citation
831-1.0062216.ris

Full Text

STUDIES IN STEROIDS AND  ALKALOIDS  by  ISIDOROS Dipl.  VLATTAS  Chem. The U n i v e r s i t y  M.Sc., The U n i v e r s i t y  A T H E S I S SUBMITTED  o f Athens  - Greece,  o f B r i t i s h Columbia,  I N P A R T I A L FULFILMENT OF  1963.  THE  REQUIREMENTS FOR THE DEGREE OF DOCTOR OF in  PHILOSOPHY  t h e Department of Chemistry  We a c c e p t t h i s required  thesis  as c o n f o r m i n g t o t h e  standard  THE UNIVERSITY'OF B R I T I S H February,  1966  COLUMBIA  1959  In the  presenting  requirements  British  mission  for  I agree  that the Library  for extensive copying  representatives.,  cation of  this  for  Department  Ojd^e.-^i^i,  The U n i v e r s i t y o f B r i t i s h V a n c o u v e r 8, C a n a d a ti?  -  J  s h a l l make  of this  thesis  for  -  /  dlyrf  Columbia  it  that  of freely per-  scholarly or  that copying or  f i n a n c i a l gain shall  permission.  Date  the U n i v e r s i t y  I f u r t h e r agree  is understood  w i t h o u t my w r i t t e n  of  fulfilment of  by t h e Head o f my D e p a r t m e n t  It  thesis  in p a r t i a l  degree at  r e f e r e n c e and s t u d y .  p u r p o s e s may be g r a n t e d his  thesis  f o r an a d v a n c e d  Columbia,  available  this  not  by publi-  be a l l o w e d  The U n i v e r s i t y  of B r i t i s h  Columbia  FACULTY OF GRADUATE STUDIES  PROGRAMME OF THE FINAL ORAL EXAMINATION FOR THE  DEGREE OF  DOCTOR OF PHILOSOPHY  of  ISIDOR VLATTAS  Dipl.Chem., The U n i v e r s i t y M.Sc,  The U n i v e r s i t y  of Athens, 1959  of B r i t i s h Columbia, 1963  FRIDAY, FEBRUARY 25, 1966 at 3:30 P.M. IN ROOM 261,  CHEMISTRY BUILDING  COMMITTEE IN CHARGE Chairman:  D. H. C h i t t y  C. T. Beer W. R. C u l l e n J . P. Kutney External  Examiner:  C. A. McDowell T. Money R. E. P i n c o c k C. D j e r a s s i  Department of Chemistry Stanford University Stanford, C a l i f o r n i a Research S u p e r v i s o r :  J . P. Kutney  STUDIES IN STEROIDS AND  ALKALOIDS  ABSTRACT In p a r t I of t h i s t h e s i s are d e s c r i b e d our s t u d i e s i n the area of aza s t e r o i d s . These i n v e s t i g a t i o n s i n v o l v e chemical and s p e c t r o s c o p i c s t u d i e s of these compounds . L i t h i u m aluminum h y d r i d e r e d u c t i o n of 3P -hydroxyl l - a z a - 5 « ,22p - s p i r o s t - 8 ( 9 ) - e n - 1 2 - o n e (7.2) p r o v i d e s the enamine, ( 7 3 ) , which upon subsequent c o n v e r s i o n to i t s iminium s a l t , ( 7 5 ) , and b o r o h y d r i d e r e d u c t i o n y i e l d s 1 1 aza-5<x , 8 f , 9 ^ , 2 2 p - s p i r o s t a n - 3 p - o l ( 7 6 ) . T h i s r e a c t i o n f u r n i s h e s a convenient sequence f o r r e d u c t i o n of the 8,9-double bond i n 11-aza s t e r o i d d e r i v a t i v e s . Degradat i o n of the sapogenin s i d e c h a i n then a l l o w s e n t r y i n t o 11-aza pregnane d e r i v a t i v e s . The s y n t h e t i c sequence prov i d e s the f i r s t examples of 11-aza s t e r o i d analogues i n which r i n g C i s six-membered and completely s a t u r a t e d . A d e t a i l e d d i s c u s s i o n of the mass s p e c t r a of 611-aza s t e r o i d d e r i v a t i v e s i s p r e s e n t e d .  and  In p a r t I I of t h i s t h e s i s i s d e s c r i b e d our work which r e l a t e s to a s y n t h e t i c approach t o the Iboga and Aspidosperma a l k a l o i d s . The f i r s t s e c t i o n i n v o l v e s the s y n t h e s i s of 2-carbomethoxy-3-[« -hydroxy- p - ( 3 -  carbomethoxy-N-piperidyl)-ethyl3-indole (78). and 2>-[_f~ (3-carbomethoxy-N-piperdyl)-ethylj- indole-2-acetic methyl e s t e r ( 9 3 ) .  acid  The Hoesch r e a c t i o n was used f o r the s y n t h e s i s of 2-carbomethoxy-3-chloroacetylindole (75) from'2-carbomethoxy-indole ( 7 4 ) and c h l o r o a c e t o n i t r i l e . Treatment of 75 w i t h 3-carbomethoxy p i p e r i d i n e ( 7 6 ) y i e l d e d 2carbomethoxy-3-(3-carbomethoxy-N-piperidyl)-acetyl indole (77). The l a t t e r was reduced w i t h sodium b o r o h y d r i d e or by c a t a l y t i c h y d r o g e n a t i o n w i t h Raney n i c k e l t o 7 8 . Prolonged h y d r o g e n a t i o n of 77 or 78 w i t h Raney n i c k e l c a t a l y s t , p r o v i d e d 2-carbomethoxy-3-[5*-hydroxy- p - ( 3 carbomethoxy-N-piperidy1)ethylj -4,5,6,7-tetrahydroi n d o l e ( 7 9 ) . S i m i l a r l y 2-carbomethoxy-indole ( 7 4 ) was reduced t o 2 - c a r b o m e t h o x y - 4 , 5 , 6 , 7 - t e t r a h y d r o - i n d o l e ( 8 0 ) by h y d r o g e n a t i o n w i t h p l a t i n u m oxide c a t a l y s t . The Hoesch r e a c t i o n was a l s o used f o r the s y n t h e s i s of 3 - c h l o r o a c e t y l i n d o l e - 2 - a c e t i c a c i d methyl e s t e r ( 8 9 ) from i n d o l e - 2 - a c e t i c a c i d methyl e s t e r ( 8 8 ) and c h l o r o acetonitrile. Treatment of 89 w i t h 3-carbomethoxy  p i p e r d i n e (76) p r o v i d e d 3-(3-carbomethoxy-N-piperidyl)a c e t y l i n d o l e - 2 - a c e t i c a c i d methyl e s t e r (92) .. The l a t t e r substance was reduced w i t h d i b o r a n e t o 93. The second s e c t i o n p r o v i d e s t h e s y n t h e s i s of 1,2,3, 5 , 6 , l l , l l b ( £ ) - h e p t a h y d r o - 2 5 - (3-chloroprophyl)-2I - e t h y l s ' o x o - i n d o l o ( 2 , 3-g-) i n d o l i z i n e (118). The fundamental r e a c t i o n i n v o l v e d c o n d e n s a t i o n of tryptamine w i t h e i t h e r e t h y l o< - k e t o - y - ( )f - b e n z y l o x y p r o p y l ) - y - e t h y l - g l u t a r a t e (70b) or e t h y l - - ( y - b e n z y l oxypropyl )-oi, - e t h y l - s u c c i n a t e (70a). When g l u t a r a t e 70b was condensed w i t h t r y p t a m i n e the amides 110 and 111 were o b t a i n e d . On t h e other hand the s u c c i n a t e 70a r e a c t e d w i t h t r y p t a m i n e t o a f f o r d t h e d e s i r e d N-j~£ - ( 3 - i n d o l y l ) - e t h y l j - ( ^ - b e n z y l o x y p r o p y l ) <X - e t h y l - s u c c i n i m i d e (112). Treatment of t h e l a t t e r substance w i t h boron t r i b r o m i n d e y i e l d e d N-[? - ( 3 - i n d o l y l ) e t . h y l J - o < - (3-hydroxypropyl)-o< - e t h y l - s u c c i n i m i d e (115), which was subsequently c o n v e r t e d t o N-[p> - ( 3 - i n d o l y l ) e t h y l j - Q\ - ( 3 - c h l o r o p r o p y l ) - o( - e t h y l - s u c c i n i m i d e (116) with thionyl chloride. C y c l i z a t i o n of t h e l a t t e r subs t a n c e w i t h phosphorus p e n t o x i d e a f f o r d e d 2,3,5,6,11pentahydro-2^-(3-chloropropyl)-2^ - ethyl-3-oxo-indolo (2,3-g) i n d o l i z i n e (117), which on h y d r o g e n a t i o n w i t h p l a t i n u m oxide y i e l d e d 118. The g l u t a r a t e 70b and t h e s u c c i n a t e 70a i n v o l v e d i n t h e above syntheses were o b t a i n e d v i a a s e r i e s of e s t a b l i s h e d r e a c t i o n s , s t a r t i n g from b e n z y l c h l o r o p r o p y l e t h e r (101). GRADUATE STUDIES Fields  of Study:  Chemistry  Topics i n Physical  Chemistry  Seminar i n Chemistry T o p i c s i n I n o r g a n i c Chemistry T o p i c s i n Organic  Chemistry  A. Bree J . A. R. Coope J . P. Kutney N. B a r t l e t t W. R. C u l l e n D. E. McGreer T. E. P i n c o c k J . P. Kutney  P h y s i c a l Organic Chemistry R. Stewart R. E. P i n c o c k Organic R e a c t i o n Mechanisms S t r u c t u r e of Newer N a t u r a l Products s A. I . S c o t t Recent S y n t h e t i c Methods i n G. G. S. Dutton A. Rosenthal Organic Chemistry  PUBLICATIONS  J.P. Kutney, R.A. Johnson and I . V l a t t a s , Can.J.Chem., 41, 613 (1963). ; S y n t h e s i s of 6-Aza S t e r o i d s . A Novel C l a s s of Aza Androstane Analogues. J.P. Kutney, I . V l a t t a s and G.V. Rao, Can.J.Chem, 41, 958 (1963). S y n t h e s i s of 6-Aza S t e r o i d s . A Novel C l a s s of S t e r o i d a l Sapogenin Analogues. J.P. Kutney, I . V l a t t a s , S t e r o i d s , 4, 595 (1964). S y n t h e s i s of 11-Aza S t e r o i d s . A Novel C l a s s of Pregnane Analogues.  ABSTRACT  In part I of this thesis are described our studies in t h e a r e a o f aza steroids.  These im, es ligations involve chemical and spectroscopic r  studies of these compounds. Lithium aluminum hydride reduction of 33-hydroxy-ll-aza-5a 2265  spirost-8(9)-en-12-one (72) provides the enamine, (73),  which upon sub-  sequent conversion to its iminium salt, (75), and borohydride reduction yields l l - a z a - 5 a , 8 £ , 9 a , 2 2 6 - s p i r o s t a n - 3 B - o l  (76)-  This reaction furnishes  a convenient sequence for reduction of the 8,9-double bond in Il-aza steroid derivatives.  Degradation of the sapogenin side chain then allows  entry into 11-aza pregnane derivatives.  The synthetic sequence provides  the first examples of Il-aza steroid analogues in which ring C is sixmembered and completely saturated. A detailed discussion of the mass spectra of 6- and ll-aza steroid derivatives is presented. In part II of this thesis is described our work which relates to a synthetic approach to the Iboga and Aspidosperma alkaloids.  The first  section involves the synthesis of 2-carbomethoxy-3-[a-hydroxy-3-(3-carbomethoxy-N-piperidyl)-ethyl]-indole (78) and 3-[B-(3-carbomethoxy-N-piperdyl)-ethyl]-indole-2-acetic  acid methyl ester (93).  The Hoesch reaction was used for the synthesis of 2~carbomethoxy-3chloroacetylindole (75) from 2-carbomethoxy-indole (74) and chloroacetonitrile.  Treatment of 75 with 3-carbomethoxy piperidine (76) yielded  2-carbomethoxy— 3 — (3-carbomethoxy-N-piperidyl)-acetylindole ( 7 7 ) , latter compound was reduced with sodium borohydride or by catalytic  The  hydrogenation with Raney nickel--to 78.  Prolonged hydrogenation of 77 or  78 with Raney nickel catalyst provided  2-carbomethoxy-3-[a-hydroxy-6-(3-  carbomethoxy-N-piperidyl)-ethyl]-4,5,6,7-tetrahydro-indole (79). 2-carbomethoxy-indole  Similarly  (74) was reduced to 2-carbomethoxy-4,5,6,7-tetra-  hydro-indole (80) by hydrogenation with platinum oxide catalyst. The Hoesch reaction was also used for the synthesis of 3-chloroacetylindole-2-acetic acid methyl ester (89) from indole-2-acetic acid methyl ester (88) and chloroacetonitrile.  Treatment of 89 with 2-carbo-  methoxy pipe r i dine (76) provided 3-(3-carbomethoxy-N-piperidyl)-acetylindole-2-acetic acid methyl ester (92).  The latter substance was reduced  with diborane to 93. The second section provides the synthesis of 1,2 ,3,5,6,11 •,lib(£)heptahydro-2C-^chlorppropyl)-2£-ethyl-3-oxo-indolo(2,3-g)indolizine (118). The fundamental reaction involved condensation of tryptamine with either ethyl a-ketb-Y-(V-benzyloxyprppyl)-Y-ethyl-glutarate (70b) or ethyla-(Y-benzyloxypropyl)-a-ethyl-syccinate  (70a).  When glutarate 70b was  condensed with tryptamine the amides 110 and 111 were obtained... On the other hand, the succinate 70a reacted with tryptamine to afford the desired N-[g- (3-indolyl)-ethyl-a- (y-benzyloxypropyl)-a-ethyl-succinimide (112).  Treatment of the latter substance with boron tribromide yielded  N-[g-(3~indolyl)-ethyl]-a-(3-hydroxypropyl)-a-ethyl-succinimide (115), which was sybsequently converted to N-[8-(3-indolyl)-ethyl]-a-(3chloropropyl)-a-ethyl-succinimide (116) with thionyl chloride.  Cycli-  zation of the latter substance with phosphorus pentoxide afforded 2,3,5, 6,ll-pentahydro-2C-(3-chloropropyl)-2£-ethyl-3-oxo-indolo(2,3-g) indolizine (117), which on hydrogenation with platinum oxide yielded 118.  iv  The glutarate 70b and the succinate 70a involved in the above, syntheses were obtained via a series of .established reactions,, starting from benzyl y-chloropropyl ether (101).  V  TABLE OF CONTENTS Page ABSTRACT  ii v  TABLE OF CONTENTS LIST OF FIGURES  vii  ACKNOWLEDGEMENTS  xii  PART  I:  Studies in Aza Steroids  SECTION A: The synthesis of the 11-aza steroids Introduction  1  Discussion  26  Conclusion  40  Experimental  41  SECTION B: Mass spectra of the 6-aza and 11-aza steroids Introduction  48  Discussion  50  A.  The mass spectra of the 6-aza-5£- steroids  50  B.  The mass spectra of the 6-aza-5£-7-one steroids  61  C.  The mass spectra of the A -6-aza-7-one steroids  78  D.  The mass spectra of the 11-aza steroids  88  l+  Experimental  103  REFERENCES  104  vi  PART  II: Studies in the Alkaloid Field  Introduction  108  Discussion  1 2 8  Section a.  1 2 8  Section B.  1 4 S  Experimental REFERENCES  2 0 7  vii L I S T OF FIGURES  Figure  P  A  R  T  1  Page  1  2  2  3  3  5  4  ,  6  5  8  6  8  7  9  8  10  9  12  10  13  11  14  12  15  13  16  14  16  15  17  16  18  17  19  18  20  19  21  20  22  21  23  22  24  23  . . .  25  24  28  26  31  viii LIST  OF FIGURES (Continued)  Figure  Page  27  32  28  r  ..  34  29  36  30  38  31  .....  51  32  52  33  53  34  57  35  58  36  59  37  .  61  38  63  39  64  40  72  41  ....  73  42  74  43  78  44  79  45  80  46  82  47  83  48  ....  .  84  49  88  50  ..  51  ..  52  89 . ..  90 91  ix LIST OF FIGURES (Continued) Figures  Page  53  92  54  ...  93 PART  1  II ..  109  2  Ill  3  112  4  113  5  114  6  116  7  ...  ....  117  8  119  9  121  10  123  11  .  12 13  124 125  .  126  14  127  15  129  16  129  17  131  18  131  19  135  20  136  21  138  21a  141  X  LIST OF FIGURES (Continued) Figures  Page  22  144  23  .  145  24  146  25  146  26  146  27  147  28  147  29  148  30  149  31  151  32  151  33  152  34  152  35  154  36  154  37  156  38  156  39  157  40  160  41  162  42  162  43  16,5  44  166  45  167  46  170  xi LIST OF FIGURES (Continued) Figures  Page  47  170  48  172  49  175  xii  ACKNOWLEDGEMENTS  I wish to express my sincere appreciation to Dr. James P. Kutney for suggesting these interesting topics and also for his guidance and encouragement during the course of my research. Financial aid from the National Research Council of Canada, Smith Kline and French Laboratories, Life Insurance Medical Research Fund, B.C. Heart Foundation and the National Cancer Institute of Canada is very gratefully acknowledged. I am also grateful for having received a National Research Council of Canada studentship-during my study.  PART I  Studies in Aza Steroids  - 1 SECTION A The synthesis of the 11-aza steroids Introduction In recent years there have been numerous investigations concerned with the effect of substituents attached to the normal steroid skeleton on the biological properties of these important substances.  These investigations  have brought forth the realization that very dramatic alterations in these properties are indeed encountered when such substituents as halogen, particularly fluorine, hydroxyl and methyl are placed at rather specific positions 1-3 in the molecule.  These steroidal derivatives s t i l l possess the basic.  steroid skeleton so that the nature of-the molecule is not altered to a very significant extent.  Our own interest in this area was directed toward  the,introduction of a hetero atom in the steroid skeleton.  In particular,  the substitution of a nitrogen atom may provide aza steroids which exhibit new types of biological activity.  Dorfman et a l . had reported some 4 interesting findings in this regard. A more detailed discussion by  Doorenbos indicated that aza steroids indeed provide a rich source of new drugs against certain diseases.^ For example* the acetate derivatives of 17B-hydroxy-4-aza-androst-5-en- > 3-one,(l) and 176-hydrbxy-17a-methyl-4-aza-androst-5-en-3-one  (2) (Figure 1)  exhibited androgenic activity equivalent to one seventh and one.fifth that of testosterone by a chick comb injection test. The latter compound and 17g-hydroxy-4-aza-19-norandrost-5-en-3-one (3) when injected on the chick's comb at a dose of 2 mg, inhibited the action of testosterone administered 4 by subcutaneous injection.  The compounds, N-methyl-4-aza-38-methyl-5ci-  cholestane (4) and N N-dimethyl-4-aza-3B-benzyl-5a-cholestane iodide (5) >  Figure  1.  Some Biologically Active Aza Steroids  have.been found to block the reduction of desmosterol to cholesterol during the biosynthesis of cholesterol.-.  Acetyl-coenzyme A  Acetoacetyl-coenzyme A  3-Hydroxy-3-methylglutaryl-coenzyme A 4-  Mevalonic acid + Squalene  Lanosterol + Zymosterol 4-  Desmosterol + Cholesterol Figure 2.  Scheme Showing the , Important Intermediates in the Biosynthesis of Cholesterol.  Counsell and co-workers have also discovered that a group of diaza derivatives of cholesterol apparently block the conversion of 3-hydroxy-3methylglutaryl-coenzyme A into mevalonic acid.  One of the most effective  _ 4 -  compounds in this connection is 20,25-diazacholesterol  (6).  It is suggested  that these substances which block the biosynthesis of cholesterol may be of clinical value for the treatment of the disease atherosclerosis, which is associated with abnormally high serum cholesterol  levels.^'^  Some aza steroids may also exert a marked increase in cholesterol biosynthesis.  The most active of these are 3ot-N-ethanolamino-cholestane  and N-phenyl-4-aza-5-cholestan-3-one (8).  (7)  These derivatives are useful in  deducing sclerotic lesions in laboratory.animals without resorting to high levels of cholesterol in the diet. Besides these properties aza steroids have been reported to possess anabolic, anti-bacterial, anti-fungal, hypotensive, coronary artery dilating, CNS stimulant, CNS depressant, neuromuscular blocking, anti-inflammatory, and androgenic activities.^ The introduction of a nitrogen atom into the steroid nucleus has attracted the interests of chemists for some time.  As a result of numerous  investigations particularly during recent years, nitrogen atoms have been introduced into virtually every position on the steroid nucleus as well as in the side-chain. An excellent literature survey on the various aza steroid syntheses is 7 given by Djerassi 8.  whereas the side-chain introduction is given in reference  A review of the recent literature reveals that a considerable number of  investigations have been carried out in the field of aza^steroids.  It would  be pertinent to discuss this work briefly at this point of the thesis. The Beckmann rearrangement has been utilized frequently in these syntheses and recently the oximes of a 1-keto steroid (9) and the corresponding 1-keto-A^nor derivative (12) were converted to the 1-aza-A-homo (10) or  - 5 the 1-aza analogue (13)  possessing the conventional steroid skeleton.  The  1,10-seco-l-cyano compounds 11 and 14 also obtained via,abnormal rearrange9 ment.  (Figure 3).  Figure  3.  Synthesis.of 1-Aza and 1-Aza-A-homo Steroids.  The oxime of 5a-cholestan-2-one (15)  on Beckmann rearrangement provided  a mixture of 2- and 3-aza-A-homo lactams (16)  and (17)  which could be sepa-  rated and reduced with lithium aluminum hydride to the corresponding 2-  and  9 3-aza-A-homo-5a-cholestanes 18 and 19 respectively.  On the other hand, the  Beckmann rearrangement of the oxime of A-nor-5a-cholestan-2-one (20) an inseparable mixture of 2-aza-5a-cholestan-3^one (21) choles.tan-2-one (22).  10  provided  and 3^aza^5a-  The authors were able to separate the two products  as their dichloro derivatives (23)  and (24)  their reactivity with collidine (Figure 4).  and distinguish between them in The results concerning Beckmann  - 6 rearrangement of 2-keto steroids were in agreement with previous investigations on the 2-oximino-A-nor-53-cholanate s e r i e s . ^ Similarly the Beckmann rearrangement of the  5a-cholestan-3^one oxime  (25) provided an inseparable mixture of the 3- and 4-aza-A-homo derivatives 12 (26) and (27).  Figure  4.  On the other hand, Beckmann rearrangement of the oxime of a  Synthesis of Some 2- and 3-Aza Steroids.  '- 7 A^-S-keto steroid (28) provided only the unsaturated lactam (29) which after catalytic hydrogenation and lithium aluminum hydride reduction gave the 13 14 3- aza-A-homo steroid analogue.(30).  '  In this particular case i t was  found that the yield, of the lactam (29) was independent of the relative amounts of the syn and.anti forms of the oxime present in the mixture at least under the particular conditions used for the Beckmann rearrangement 14 (SOCI2  in dioxane).  These results indicated that the reaction product was  not necessarily related configurationally to the starting oxime as may be expected from the mechamism of the Beckmann.rearrangement (Figure 5). The synthesis of a 3-aza equilenin derivative (33) has been recently accomplished via an intramolecular interconversion of  2-aminoequilenin-3,4-  quinone (31) with peracetic acid and subsequent decarboxylation of the acid (32) with copper powder;^ (Figure 6). The high physiological activity associated with some 4-aza steroid derivatives has stimulated an extensive research towards the synthesis of these compounds;  Doorenbos and his coworkers have reported recently a  number of 3- and 4- substituted 4-aza steroids.  In this work intermediate  keto acid (34) was cyclized with the appropriate amines, to provide 16 17 18 4- hydroxy (35), 4-amino (36) and 4-alkylamino (37) steroid derivatives as well as other heterocyclic steroids which include nitrogen in the steroid 19 skeleton (38). On the other hand, 3-alkyl- or 3-aryl-4-aza steroids (41) have been obtained from the reaction of the saturated lactam (39) and an 20 21 appropriate Grignard reagent  '  followed by subsequent hydrogenation of  the resultant enamine (40) (Figure 7). 22 23 The synthesis of 5-aza-A-nor-B-homo and 5-aza-A-nor • steroids (45) and (49) has been acecomplished via a Beckmann rearrangement of the keto ester oximes (42) and (46), cyclization of the resulting lactams (43)  Figure  6.  Synthesis of 3-Aza-4-hydroxy Equilenin.  -  Figure  7.  9  -  Synthesis of Some 3 - and 4 - Substituted 4-Aza Steroids.  -  10  -  and (47) with phosphorus oxychloride and subsequent reduction of the imides (44) and (48) (Figure 8).  (45). 0 BiXl/Pyt. ^  A  I70-I80°C  V  CH 0-<-=O tf>  46  47  Figure  8.  Synthesis of Some 5-Aza Steroids.  A successful sequence leading to the synthesis of 6-aza steroids has 24 25 been developed in our laboratory several years ago sequence as well as a similar one independently  '  (Figure 9).  This  conceived by Jacobs and  Brownfield provided the f i r s t general synthesis of ring B aza steroids. Other workers have also succeeded in the synthesis of 6-aza steroids by 27 28 using a modified Curtius reaction ' . The elimination of the C-3 oxygen function during the course of the above sequences was a disadvantage for the synthesis of steroid derivatives in which such a function was to be 29 retained. Knof  was able to overcome this difficulty by reduction of the  intermediate isocyanante  (50) in the Curtius reaction sequence.  To support  these results Knof suggested the existence of an equilibrium between the isocyanate 50 and the lactone 51.  The latter intermediate on catalytic  hydrogenation or lithium aluminum hydride reduction leads to 52 and 53 respectively.  (Figure 10).  The total synthesis of 6-aza steroids in the estrogenic series has been independently  reported recently by three groups of workers.  Burckhalter  30 and Wattanabe were able to obtain the ketone (54) from which the synthesis, 31 involving several modifications of Johnson's  classic equilenin synthesis  led to dl-6-aza equilenin (55) (Figure 11). Another attractive approach to 6-aza equilenin and other 6-aza estrone 32 33 derivatives has been independently  reported  '  (Figure 12).  The synthesis of 6-aza-B-homo steroids (57) has been recently reported  1  (figure 13). Another sequence involving the Beckmann rearrangement of a 7-keto 34 oxime (58) provided the 7a-aza-B-homo steroids (59) (Figure 14). 15). a Beckmann rearrangement of the 5B-B-nor-6-ketoxime (60) (Figure The synthesis of the true 7-aza steroid system (61) was accomplished by 35  - 12 -  Figure  9. Preparation of 6-Aza Steroids Using Ring Cleavage and Enol-lactam Ring Closure.  - 13 -  Figure  10.  Synthesis Curtius  o f 6-Aza S t e r o i d s v i a a M o d i f i e d Reaction.  -  14  -  .55  Figure  11.  Total Synthesis  of dl-6-Aza  Equilenin.  - 15  ' H  Figure  12.  Total Synthesis of dl-6-Aza Equilenin and Other 6-Aza Estrone Derivatives.  Figure  14.  Synthesis  o f 7a-Aza-B-homo S t e r o i d s .  - 17 -  Figure  15.  S y n t h e s i s o f 7-Aza S t e r o i d s .  The t o t a l s y n t h e s i s o f aza s t e r o i d s c o n t a i n i n g a n i t r o g e n atom a t a 36 b r i d g e head p o s i t i o n has been r e c e n t l y r e p o r t e d . the s y n t h e s i s o f 8-aza e s t r o n e i n F i g u r e 16.  Clarkson  has a c c o m p l i s h e d  (62) and h i s s y n t h e t i c sequence i s summarized  - 18 -  Figure  16.  Synthesis of 8-Aza Estrone.  37 Meltzer et al  reported the formation of six diastereoisomers due  to the development of Cg, C13, and C ^  asymmetric centers in their  synthetic approach to 8-aza estrone (63) (Figure 17). 38 Meyers et al  have developed a synthetic approach to both 8- and  9-aza steroidal analogues (64) and (65) (Figure 18).  Figure  17.  Synthesis of 8-Aza Estrone Derivatives.  The synthesis of 13-aza-18-nor equilenin methyl ether (66) has been recently reported independently by two groups of.workers. 39 The synthetic sequence by Kessar et al is given in Figure 19. 40 Birch and Subba Rao  have developed a synthetic approach to  13-aza-18-nor as well as 13-aza-18-nor-D-homo equilenin methyl ether (66) and (67) (Figure 20). 41 Rakjit and Gut  synthesized 17-aza pregnane derivatives (69) using 42 the.lactam (68) as starting material. (Figure 21).  Figure  18.  Synthesis of 8- and 9-Aza Steroids.  o  19.  S y n t h e s i s o f 13-Aza-18-nor E q u i l e n i n Methyl E t h e r .  - 22 -  67  n =2  68  n =3  Figure  20. Synthesis of Some 13-Aza Equilenin Derivatives.  With appropriate modifications of the above sequence and using the 43 lactam 70 as starting material, 17-aza progesterone (71) was obtained (Figure 21). Some diaza cholesterol derivatives and particularly 20,25-diazacholesterol (6) was found to be extremely potent inhibitor of cholesterol biosynthesis in laboratory animals.  In an effort to obtain further insight  as to the mode of action of this compound a series of cholesterol 44 isosteres having only one nitrogen atom in the side chain was synthesized as shown in Figure 22.  Figure  21..  Synthesis of Some 17-Aza Pregnane  Derivati  - 24 -  X =0  R = -NHCH CHCH  3  23-aza cholesterol  X = 1  R = -NHCH(CH )  3 2  24-aza cholesterol  X = 2  R = -N(CH )  25-aza cholesterol  2  Figure  3  22.  2  Synthesis of Some Side Chain Aza Cholesterol Derivatives.  - 25 The above discussion has summarized the recent work in other laboratories on the synthesis of various aza steroids. :  I would now like to  return to our investigations in this area. I have already mentioned above that we had succeeded,in developing a synthetic sequence for the introduction of a nitrogen function into ring B of the steroid skeleton.  This work provided some of the f i r s t  examples of 6-aza steroids in which the normal steroid skeleton was retained. We subsequently extended out studies to the synthesis of ring C-aza steroids and succeeded in preparing the f i r s t 11-aza derivative in the steroidal 45 sapogenm series.  The synthetic sequence is illustrated in Figure 2 3 .  This part of the thesis describes the extension of this work to , . 46,47 11-aza pregnane derivatives. '  Figure *  2 3 . Synthesis of 11-Aza Steroids via the Seco-keto Acid Method.  -  26  -  Discussion The synthesis of 11-aza pregnane derivatives from the synthetic intermediate bearing the spiroketal side chain ( 7 2 ) required the investigations of two distinct reactions: (a) reduction of the 8 , 9 double bond to provide the saturated system in ring C and (b) degradation of the spiroketal side chain.  Since the latter phase of this problem was expected to follow the  well-known degradation of steroidal sapogenins to the pregnane series, i n i t i a l consideration was given to part (a). 25  From previous results in our laboratory in the area of 6-aza steroids 26  and those of other workers  i t was apparent that the hydride reduction of  an enol lactam system could lead to either an enamine or imine function. We decided to investigate this possibility in the case of the 11-aza intermediate ( 7 2 ) .  Reduction of 7 2 with lithium aluminum hydride in refluxing I  tetrahydrofuran provided a new reaction product and the spectral properties of this material quickly established that the lactam had been reduced.  The  strong lactam absorption present in the infrared spectrum of 7 2 had disappeared and instead, a weak absorption at 6 . 1 3 y characteristic of the enamine chromophore was evident. The expected shift to 6 . 0 2 y was also observed in the infrared spectrum of the hydrochloride salt of ( 7 3 ) in good agreement with published work on 48 a,B-unsaturated amines;  Furthermore, the characteristic absorption in the  ultraviolet spectrum of the enol lactam ( 7 2 ) ( X max 2 5 5 my) was also absent in the ultraviolet spectrum of the reduction product but now a new absorption at 2 3 6 my was noted.  This data was again in good agreement with the enamine  chromophore which has been extensively investigated in the ultraviolet 49  region by Leonard and co-workers.  This evidence excludes the imine  chromophore since the >C=N~ grouping would not exhibit these spectral  properties and we assign structure (73) to this reduction product.  In spite  of numerous attempts to obtain this enamine crystalline, we were unsuccessful and consequently characterized this compound as the acetate, (74). It should be noted at this point that Engel and Rakhit*^ have reported a similar reduction and also postualte an enamine grouping although no spectral data is presented for the product obtained directly from the reduction.  They also encountered difficulties in obtaining a crystalline  product and report spectral data on the crystalline acetate derivative of the enamine. 8 9  Having obtained the A ' -11-aza hecogenin derivative, (73) we then \  considered the reduction of the 8,9 double bond.  It is well known from  the normal steroid series that this double bond is particularly d i f f i c u l t to hydrogenate and the necessary conditions would almost certainly destroy the spiroketal system so that this method could not be seriously considered. It was felt that one way to eliminate this difficulty would be to convert the enamine, (73) into the iminium system, (75), and then subsequently reduce the latter by means of a hydride reagent.  The conversion,  >C=C-N< '—• >C-C=N<, is well known from the work of Leonard on unsaturated amines ^''^ and strong support for the success of this reaction was already evident from the observed shift in the infrared spectrum of the salt derivative of 73.  Indeed when the enamine, (73) was f i r s t treated with  anhydrous hydrogen chloride and the resulting crude product was subjected to reduction with sodium borohydride, we obtained after chromatographic purification, the desired reduction product, (76) m.p. 211-212°C, as one of the crystalline substances from this reaction.  Three other crystalline  products were actually obtained from this reaction and the pertinent data is discussed below although further work is necessary before complete  Figure  24.  - 29 structural assignments can be made in these instances.  The reduction product  (76), exhibited the expected spectral properties in agreement with the assigned structure.  The compound no longer showed any absorption in the  ultraviolet spectrum and the enamine absorption in the infrared region was also absent.  The characteristic spiroketal bands  52 53 ' were s t i l l present  in the fingerprint region of the infrared spectrum and therefore indicated that, as expected, the spiroketal system was s t i l l intact. Nuclear magnetic resonance (NMR)  spectroscopy also played an important  role in providing confirmatory evidence for a l l the structures in this investigation.  This was possible since in relation to another problem a  detailed analysis of the NMR  spectra of a large number of known steroidal 54 55  sapogenins had been carried out  '  and pertinent regions of the  spectra could be now assigned with certainty.  NMR  It is necessary at this point  to discuss briefly some of the relevant features of the previous work as they apply to the present study. Dr. Kutney had previously shown that in the instance of saturated sapogenins, the low field region of the NMR  spectra indicated only two sets  of signals. One set was observed as a broad multiplet which showed two main broad signals in the region 200-210 c.p.s., which was attributed to C26 protons of the spiroketal system and the other set at lower field (250-300 c.p.s.) was due to the lone proton at Cig and the C3 proton in the case pf C3-acetylated sapogenins.  With this information on hand one could  readily analyse the low field region of the NMR  spectrum of 76,  As expected the Cjg proton appeared as a broad signal in the region, 245-280 c.p.s.  The C26 and C3 proton signals were also clearly evident in  the region 190-230 c.p.s. but in addition to these signals, new sets of lines appeared in the region i30-185 c.p.s. and these corresponded in area  - 30 to three additional protons. This region is normally completely devoid of any signals as shown by the spectra of numerous steroidal sapogenins and i t was immediately obvious that these absorptions were due to protons on carbon atoms attached to the basic nitrogen atom (Cg and C 1 2 ) • This latter fact was confirmed when the NMR spectra of the crystalline monoacetate (77) and diacetate (78) were analyzed (See Figures 26 and 27).  One would expect that  in the former only the C3 proton would be shifted downfield and indeed this was the case.  The broad absorption signal in the 245-300 c.p.s. region  integrated for two protons while the proton absorption in the 130-190 c.p.s. region remained essentially unchanged.  In the case of the diacetate (78)  i t was now expected that the Cg and C ^ protons would be shifted downfield as well.  There was a general downfield shift of the signals in the 130-185  c.p.s. region and most important, this region now constituted an area due to only two protons.  A new one-proton signal now appeared as a broad doublet  at lower field (220-245 c.p.s.) and was attributed to the Cg proton.  The  presence of the 0-acetyl group in 77 was confirmed by a singlet at 117 c.p.s. Similarly the 0-acetyl and N-acetyl groupings in 78 were confirmed by the singlets at 118 and 128 c.p.s.  The proximity of the N-acetyl function to  the Cig angular methyl group could also be recognized by analysis of the high field region in the NMR spectra of 76 and 78.  In the former, the two  angular methyl groups were barely separated and were observed as two sharp lines at 53 and 55 c.p.s. respectively, whereas in the latter a larger separation was noted (47 and 58 c.p.s.) and, as expected, the C19 signal now occurred at lower field. reduction had been successful.  It was now clear that the sodium borohydride  Figure  26.  Figure  27.  - 33 Having assigned structure 76 to this reduction product, I would like to mention some data which has been obtained for the other three crystalline products (arbitrarily assigned as A, B and C  isolated from this reaction.  Mass spectrometry which became available to us after this project was essentially completed, has played a major role in providing structural evidence for these compounds. Although mass spectra are discussed in detail in the subsequent section of this thesis a brief mention of this data is necessary at this time. First of a l l , the IR spectra of a l l three compounds indicated the characteristic absorption bands (980, 923, 900 and 865 cm ketal side chain.  for the spiro-  In confirmation of this result was the presence of a  significant signal at m/e 139 in the mass spectra of these compounds - a feature characteristic of steroid sapogenins.  Furthermore the IR spectra  also indicated the presence of -NH and-QH groups in these compounds. The mass spectrum of A indicated that this substance was an isomer of 76 (at CQ and/or Cg] since i t possessed a molecular ion peak at m/e 417 and a series of signals which were common to both 76 and A. The mass spectra of B and C indicated molecular ion peaks at m/e 434 and also strong M-18 signal at m/e 416. This evidence suggested that an additional hydroxyl function was present in these compounds. Although an insufficient amount of B was available for NMR studies, the NMR spectrum of C indicated the characteristic signals of the CJS and C s protons at 263 2  and 204 c.p.s. respectively.  It was noted that signals due to five or six  protons were also present in the region 120-190 c.p.s.  These spectral  properties allow us to make a tentative assignment of structure 83 to the compounds B and C with the difference between them being one of stereochemistry at the asymmetric centers Cg and/or Cg. It is clear that further  - 34 work is necessary before any more definite conclusions can be reached.  Figure  The  28.  conversion of the iminium intermediate  75  to structure 83 could arise by  a simple hydrolysis of this intermediate to the amino-ketone 82  and subse-  quent reduction as shown in Figure 28. The stereochemistry of the newly created asymmetric centers (Cs and Cg) in (76) deserves some comment. Although i t is clear that the evidence presented here does not rigorously establish these asymmetric centers, consideration of the conformational expressions for the intermediates involved does allow tentative assignments. The i n i t i t a l process involving the conversion of the enamine, (73), to the iminium intermediate, (75)» generates an asymmetric carbon atom CQ and this should be considered presently. Conformational structures for both possibilities (8a and 83) are shown in  84  and  85  respectively, since i t  is felt that approach of the hydrogen atom is probable from either side of the molecule.  It is immediately obvious that in the 8a isomer, (84), ring B  adopts the boat conformation and serious interactions exist between the "flagpole" hydrogen atoms at C5 and Ce and also between the 76 hydrogen and  - 35 -  84  85  the Cjg angular methyl group.  On the other hand, in the 83 isomer, (85),  ring B is not in a boat conformation and there are not severe interactions of the type encountered in  84 . In fact the overall conformation of the  molecule closely approximates that of the normal trans-anti-trans backbone of the natural steroids. of  73  to  75  If one is justified to consider that the conversion  is a process which leads to equilibration, then the 83 isomer  (85), i s certainly the preferred structure.  Consideration of the next step,  namely the borohydride reduction of the iminium intermediate to the final product, (76), reveals that the approach of hydride from the 9a side in both 8a and 83 isomers is very much preferred.  In both  84  and  85  the two  angular methyl groups prevent effective approach from the 3 side of the molecule. The stereochemistry of sodium borohydride reductions of certain preformed iminium slats has been studied by Bohlmann.^  In this work i t was  shown that attack of hydride occurred from the least hindered side of the moleucle in i t s most stable conformer. On t h e basis of our considerations and the above investigation, we postulate that the most likely stereochemistry at CQ and Cg in the reduction product, (76), and in a l l subsequent substances reported below i s 83, 9a so  - 36 that the normal steroid stereochemistry persists.  Although this speculation  does not provide absolute proof for these stereochemical centers, i t is not possible to present any more rigorous data at this time since conclusive correlation to the conventional steroid system is not directly feasible. Now that the enamine group had been reduced" to a more stable system we then considered the well known degradation of the spiroketal side shain 57 to A -20-keto-pregnene derivatives. 16  Figure  (See Figure 29).  29.  When the reduction product, (76), was treated with acetic anhydride at 200° for ten hours, a brown oily product was obtained which, without further purification, was subjected to oxidation with chromium trioxide and the crude oily oxidation product was subjected to the action of .aqueous potassium hydroxide at room temperature. The final crude product was purified by chromatorgraphy on alumina to provide i n i t i a l l y the expected ll-aza-pren-16-en-20-one  derivativem (80).  The spectral properties of this crystalline material were in agreement with structure (80).  The infrared spectrum indicated three strong absorptions of  - 37 equal intensity at 5.78, 6.04 and 6.13u for the 0-acetyl,A -20-keto and 16  N-acetyl groupings.  The conjugated ketone chromophore was further confirmed  by the ultraviolet spectrum which showed an absorption at 325 my.  Finally  the NMR spectrum (Figure 30) was again very instructive and completely confirmed the structural assignment.  A broad,, one-proton signal centered at  403 c.p.s. indicated the presence of the C^s olefinic hydrogen atom and a very sharp line at 135 c.p.s. due to three protons was easily assigned to the C21 methyl group.  Indeed comparison of these signals with those  observed in the NMR spectrum of A -allopregnane-20-one which has the same 16  58 system in ring D, showed excellent agreement.  The effect of the N-acetyl  group on the two angular methyl protons was again clearly indicated since their signals are shifted to lower f i e l d (62 and 57 c.p.s.) relative to those observed in A -allopregnane-20-one (52 and 48 c.p.s.). 16  Finally the region  180-300 c.p.s, which is completely transparent in the spectrum of the allopregnane derivative indicates several sets of multiplets in the NMR spectrum of  80 . The total area under these signals corresponds to four  protons and apart from the lone proton at C3  which normally shows a broad  multiplet in the region,. 270-300 c.p.s., the remaining signals are obviously due to the C9 and C12 protons of this aza steroid derivative. A second solid product was obtained in the later chromatographic fractions of the reaction mixture from the sapogenin side chain degradation. The spectral properties suggested immediately that i t was merely the 36hydroxy-11-N-acetyl derivative, (79). The ultraviolet spectrum was identical with that observed for the diacetate, (80), but the infrared spectrum now indicated only two strong absorptions at 6.04 and 6.15 u for the conjugated ketone and N-acetyl groups respectively.  The structure  79 was conclusively  established when acetylation of this compound provided the triacetate, (80).  I 500 I  400 I  300 I  200  0 CPS I  100  \  solvent: C D C L CH, I C=0  0 II N-C-CH,  0 II O-C-CH,  3  C ,CH \ "7(8.03D 2  area=2H C +C H's 3  9  3  (7.75r) 135 120  area=2H C H's  (8.97T)62 ,57 (9.05 T)  l2  00  170  AcO  area = IH C, H 6  403 (3.27T)  8.0  int TMS  7.0  6.0  Figure 30.  282 266  5.0  225  4.0  204 . |89  3.0  2.0  1.0 p.p.m (8) 0  - 39 The final step in this sequence, namely the reduction of the 16,17double bond was readily accomplished by catalytic hydrogenation.  This  reduction product indicated no absorption in the ultraviolet region and the infrared spectrum revealed three strong absorption bands at 5.79,  5.88  and 6.08 u for the three carbonyl groups now present in the molecule.  The  NMR spectrum of this compound was also in agreement with the assigned structure, (81).  - 40  Conclusion  A synthesis of 11-aza steroids with the true steroid skeleton has been developed.  This represents the first synthesis of an 11-aza steroid of the  pregnane series.  The sequence employs hecogenin acetate as starting material  to provide the necessary intermediate 9,12-seco keto acid which cyclizes in the presence of ammonia to an enol lactam intermediate.  Lithium aluminum  hydride reduction of the enol lactam provides the corresponding enamine whose immonium salt can be reduced with sodium borohydride to an 11-aza steroidal sapogenin analogue.  Degradation of the sapogenin side chain  provides the 11-aza steroids possessing the acetyl side chain at C17. These intermediates may possibly provide a convenient route to the adrenocortical class as well.  - 41 Experimental A l l melting points were determined on a Kofler apparatus and are uncorrected.  The ultraviolet spectra were recorded in 95% ethyl alcohol on  a Cary 14 recording spectrophotometer and the rotations were taken in 1% chloroform solutions.  The infrared spectra were determined on a Perkin-  Elmer Model 21 spectrophotometer.  Analyses were performed by A. Bernhardt  and his associates, Mulheim (Ruhr), Germany. The NMR  spectra were taken in  deuteriochloroform. solutions on a Varian A60 instrument; the line positions or centers of multiplets are given in cycles per second (c.p.s.) scale with reference to tetramethylsilane as the internal standard.  The multiplicity,  and integrated area and type of protons are indicated in parentheses. Every molecular weight (M.W.) quoted was determined mass spectrometrically. 3B-Hydroxy-ll-aza-5a,22B-spirost-8(9)-ene  (73)  The enol lactam (72) (12.9 g) was dissolved in anhydrous tetrahydrofuran (1000 ml) and refluxed for 20 hours with lithium aluminum hydride (4 g) which was i n i t i a l l y placed in a Soxhlet apparatus and gradually brought into the vessel by the refluxing solvent.  The solvent was evapo-  rated iji_vacu£ and the residue decomposed cautiously by the addition of wet ethyl ether.  The mixture was then treated with water, the ether layer  separated and dried over anhydrous magnesium sulfate. solvent provided a semi-solid product. (10.6 g).  Removal of the  This material was treated  with ether (300 ml) and the white insoluble solid which remained undissolved was removed by f i l t r a t i o n (5.5 g).  Three recrystallizations of this material  from ether provided a pure sample, m.p. 168°); [ a l j j absorption.  22  -30°; infrared (KBr): 2.94  173-175° (block, preheated to about (broad), 6.lip; no ultraviolet  This material is very d i f f i c u l t to analyze and i t s structure  is s t i l l in question.  - 42 The ethereal f i l t r a t e was- then concentrated to yield (73) as an o i l (5 g) which resisted a l l attempts to crystallize; infrared (Nujol): 2.94 (broad), 6.13y  (1632 cm" , 1  weak); infrared of HCl salt (Nujol): 6.02y (1660  weak); ultraviolet: X 236 mu (alcohol); max  cm  X  . 272 my (alcohol solumax  tion containing a few drops of concentrated hydrochloric acid). The aqueous layer was extracted exhaustively with ether in a continuous, extraction apparatus (12 hrs). The ether extract was dried over anhydrous magnesium sulfate and then concentrated in vacuo, to yield a further 1.5 g of (73). 3B-Acetoxy-N-acetyl-ll-aza-5a,22  -spirost.-8(9) -ene (74)  The oily enamine, (73) (500 mg) was dissolved in pyridine (10 ml) and treated with acetic anhydride (20 ml).  The mixture was allowed to stand at  room temperature for 24 hours, after which time i t was treated cautiously with water and extracted with ether.  The ether extract was washed several  times with water, then with 5% aqueous sodium bicarbonate solution and again with water.  Filtration and removal of the solvent in vacuo provided an o i l  product (500 mg). activity III).  This material was chromatorgraphed on alumina (20 g,  Elution with petroleum ether-benzene  yielded an oily material (200 mg).  Final purification of this product was  accomplished by preparative thin-layer,chromotography chloroform-ethyl  (1:3) and benzene  ( s i l i c a gel G, with  acetate (1:1) as devoloping medium, R£ =0.65) and the  analytical sample (50 mg) of 74 was obtained as an amorphous solid: [a]  2 2 Q  + 56°; infrared (CHC1 ) 5.82, 6.15y. Found: C, 71.81; H,  0, 16.41; N, 2.55. N, 2.80  3  Calc. for C30H45O5N:  9.43;  C, 72,11; H, 9.08; 0, 16.01;  - 43 Sodium Borohydride Reduction of Enamine The enamine, (73),  (4.9 g) was dissolved in absolute methanol (500 ml)  and hydrogen chloride gas was then passed through the solution until i t was strongly acidic.  The solvent was removed in vacuo and the reddish residue  was taken up in absolute methanol (3000 ml).  The resulting solution was  treated with sodium borohydride (20 g) and the mixture then refluxed for five hours. ether.  The solvent was removed in vacuo and the residue dissolved in  The ether solution was first washed with water, than dried over  anhydrous magnesium sulfate and finally concentrated in vacuo to provide a white solid (4.6 g),  This material was dissolved in benzene and chromato-  graphed on alumina (18 g, activity III).  Elution with benzene chloroform  (9:1) provided a crystalline material (312 mg) which upon recrystallization 22 from ether-hexane yielded an analytical sample, m.p. 167-170°; [a]^ infrared (KBr): 2.96u; no ultraviolet absorption. 0, 12.26; N, 3.57; M.W. 417.  -60°;  Found: C, 73.97; H, 10.31;  Further work is required before a definite  structure can be assigned to this compound. The desired product (1.42 g) was eluted from the column with benzenechloroform (3:1).  Three recrystallizations from benzene provided an  analytical sample (1.3 g) of ll-aza-5ot,8£, a,223-spirostan-36-ol 211-212°; [ a ]  2 2 D  (76), m.p.  - 6 0 ° ; infrared (KBr): 2.84u (sharp, NH), 2.94y (broad, OH);  no ultraviolet absorption, NMR: 275-250 (broad multiplet, IH, Cig proton), 230-130 (complex pattern, 6H, C +C +C +C protons), 58, 55, 53, 49, 44, 3  9  12  25  (multiplet, 12H, C e+C 9+C i+C 7 methyl protons). 1  1  2  0, 11.40; N, 3.84; M.W. 417.  2  Calc. for C ^ H ^ O ^ :  Found: C, 74.56; H, 10.21; C, 74.77; H, 10.38;  0, 11.49; N, 3.35; M.W. 417. Elution with benzene-chloroform (3:2) provide another crystalline compound (410 mg).  Recrystallization from methylene chloride-hexane  -44 yielded an analytical sample, m.p.  179-183°; [a]  5.90, 6.09u; no ultraviolet absorption. N, 3.82; M.W.  434.  22  -100°; infrared (KBr):  D  Found: C, 69.61; H, 9.09; 0, 17.72;  Further work is required before a definite structure can  be deduced for this compound. Elution with benzene-chloroform  (1:3) and finally with chloroform  yielded the fourth crystalline compound (1.83 g).  Three recrystallizations  from methylene chloride-hexane provided the analytical sample, m.p. [a]  2 2  D  227-228°;  -109°; infrared (KBr): 2.84y (sharp, NH), 2.94u (broad, OH); no ultra-  violet absorption; NMR: (multiplet, 2H, C fc 2  275-250 (broad multiplet, IH, C i proton), 230-225 6  protons), 225-120 (complex pattern, approximately 5H)  60-37 (complex pattern, 11H, region of methyl protons). 70.40; H, 9.98, 9.76; 0, 15.82* 16.38; N, 3.29; 3.83; M.W.  Found: C, 70.58, 434.  Acetylation of 76 The alcohol, 76, (200 mg) was dissolved in pyridine (5 ml), treated with acetic anhydride (5 ml) and allowed to stand at room temperature for 24 hours.  The reaction mixture was cautiously treated with water and then  with ether.  The ether solution was washed several times with water, then  with 5% aqueous hydrochloric acid, water, and finally with 5% aqueous sodium carbonate solution.  After drying over••. anhydrous magnesium sulfate,  the solvent was removed in vacuo to provide an oily product (140 mg).  This  material was recrystallized twice from hexane to provide an analytical 22 sample of the diacetate, (78) m.p. 5.79, 6.07u; NMR  195-198°; [a]  -14°; infrared (KBr):  D  300-250 (broad multiplet, 2H, C +Ci protons), 250-135 3  (complex pattern, 5H, C +C +C 0 9  26  5  fl  12  protons),130 (.singlet, 3H, N-C-CH ), 117  (singlet, 3H, 0-C-CH ), 58, 55,,48, 44 (multiplet, 12H, 3  methyl protons).  Found: C, 71.59; H, 9.23; 0, 15.32; M.W.  3  C +Cio+C +C 18  501.  21  Calc.  27  for C oH 70 N: 3  4  C, 7 1 . 8 2 ; H, 9 . 4 4 ; 0 ,  5  45 1 5 . 9 5 ; M.W.  501.  The hydrochloric acid washings were made basic with 10% aqueous sodium hydroxide and the organic precipitate was.extracted with ether. The ether solution was washed with water, dried over anhydrous magnesium sulfate and concentrated in vacuo to provide a crystalline product (40 mg). Recrystallization of this compound from dichloromethane-hexane  provided an analytical 22  sample of the monoacetate, ( 7 7 ) , m.p. 2 1 0 . 2 1 3 ° ; 5.78  ; NMR: 3 0 0 - 2 5 0 (broad mulitplet, 2 H , C +C 3  [a] 16  - 5 9 ° ; infrared (KBr):  D  protons), 2 5 0 - 1 5 0 (complex . 0  pattern, 5 H , C +C 6 Ci2 protons), 117 (singlet, 3 H , 0 - C r C H ) , +  9  44  3  2  59, 55, 49,  (multiplet, 1 2 H , C + C + C 2 i + C 2 7 methyl protons). Found: C, 7 3 . 5 0 ; 18  H, 9 . 9 7 ; N, 3 . 6 6 .  19  Calc. for  C28H45O4N:  C, 7 3 . 1 6 ; H, 9 . 8 7 ; N, 3 . 0 5 .  Degradation of Sapogenin Side Chain 57  The procedure used here was essentially that of Wall  with slight  modifications. The alcohol, ( 7 6 ) , ( 4 . 2 g) was subjected to the action of acetic anhydride ( 6 0 ml) in a sealed tube at 200°C for 11 hours.  The reaction  mixture was treated with methanol to destroy the excess anhydride and the solution was concentrated in vacuo to yield a brownish oily residue. This residue was taken up in glacial acetic acid ( 1 0 0 ml) and the solution was cooled to 15°C.  To this cold solution, a solution of chromium trioxide  (2 g) in 80% acetic acid (16 ml) was added dropwise over a period of 20 (  minutes while the reaction temperature was kept at 15 C. 9  The mixture was  allowed to stand at 22°C for 15 hours, then treated W i t h water to allow the,organic material to precipitate,  This precipitate was extracted with  ether and the ether extract was washed successively with water, aqueous potassium carbonate and finally with water.  After drying over anhydrous  magnesium sulfate, the solvent was removed to provide a white amorphous  - 46 solid. This material was dissolved in t-butyl alcohol (100 ml) and a solution of potassium hydroxide (1 g potassium hydroxide in 2 ml of H 0 2  added.  was  This mixture was stirred vigorously for 2.5 hours at room temper-  ature. . The reaction mixture was treated with ether (500 ml) and the solvent, after drying over anhydrous magnesium sulfate, yielded an amorphous.product (2.7 g).  This material was dissolved in a small amount of benzene and  chromatographed on alumina (lib g, activity III). Elution with benzenechloroform (4:1) yielded 33,ll-diacetoxy-ll-aza-5a,8£;,9a-pregn-16-en-20-one (80) (850 mg), m.p.  185-188°; [ a ]  6.13y; ultraviolet: X  2 2 D  +110°; infrared (KBr): 5.78,  235 my (log e 3.99); NMR:  in 3.x  Ci  6  proton), 300-250 (complex pattern, 2H,  pattern, 2H, C  6.04,  404 (broad singlet, IH,  C3+C9  protons), 250-180 (complex  protons), 135 (singlet, 3H, C methyl protons), 120 0 0 (singlet, 3H, N-C-CH )^ 117 (singlet, 3H, O-C-CH3), 62 (singlet, 3H, C 12  21  3  methyl protons), 57 (singlet, 3H,  methyl protons).  H, 8.80; 0, 15.35; N, 3.79; M.W.  401.  H, 8.79; 0, 15.94; N, 3.49; M.W.  401.  ;  Calc. for  18  Found: C, 71.98;  02^350^:  C, 71.79;  Further elution with benzene-chloroform (2:3) provided a second product (1.1 g) which on two recrystallizations from ether yielded an analytical sample of the monoacetate, (79) m.p. 6.13y; ultraviolet: X  191-194°; infrared (KBr): 2.90,  234 my (log e 3.92); NMR:  6.04,  405 (broad singlet, IH,  - IHcLX  Cig proton), 300-250 (complex pattern, 2H, pattern, 2H, C  12  0  protons), 135 (singlet, 3H, C  (singlet, 3H, N-C-CH3) 60 (singlet, 3H, C 3H, C  19  methyl protons).  Calc. for C 2H 30 H: 2  3  C3+C9  3  18  21  protons), 250-170 (complex methyl protons), 119  methyl protons), 56 (singlet,  Found: C, 73.26; H, 9.52; N, 4.41; M.W.  C, 73.50; H, 9.25; N, 3.90; M.W.  359.  359.  - 47 Acetylation of (79) The alcohol, (79), (50 mg) was dissolved in pyridine (1 ml) and treated with acetic anhydride (1 ml).  After allowing to stand at room temperature  for 24 hours, the mixture was cautiously treated with water and.extracted with ether. The ethereal solution was washed successively with water, 5% ;  aqueous sodium carbonate and finally with water and then dried over anhydrous magnesium sulfate;  Removal of the solvent in vacuo provided a crystalline  product, (50 mg) which was shown to.be identical., in every respect with 80. Catalytic Reduction of (80) The diacetate, (80), (70 mg) was dissolved in 95% ethanol (10 ml) and hydrogenated at room temperature and atmospheric pressure with 10% palladium on,charcoal (30 mg).  The catalyst was filtered and the ethanol was removed  in vacuo to provide a crystalline product (70 mg).  Three recrystallizations  of this material from methylene chloride-hexane provided an analytical sample of 36,ll-diacetoxy-ll-aza-5a,8?,9a-pregnan-20-one, [a]  (81), m.p.  188-192°;.  +71°; infrared (KBr): 5.79, 5.88 and 6.08y; NMR: 130 (singlet, 3H, 0 0 C i methyl protons), 128 (singlet, 3H, N-ci-Cl^), 118 (singlet, 3H, 0-G-CH ), 2 2  n  2  3  58 (singlet, 3H, Ci$ methyl protons), 42 (singlet, 3H, Found:  C, 71.55; H, 9.16; 0,  15.23; N,  3.74; M.W.  C, 71.43; H, 9.24; 0, 15.86; N, 3.47; M.W.  403.  403.  methyl protons). Calc. for  C21+H37O4N:  - 48 SECTION B Mass spectra of 6-aza and 11-aza steroids Introduction Mass spectromatry:has become a particularly effective physical method in the structural elucidation of natural products and in recent years a tremendous effort has been put forth on investigations with this technique. A number of excellent books on this subject have appeared from various laboratories but the: one most pertinent to this discussion has been published 59 by a Stanford University group.  This book deals with the application of  mass spectrometry tc structure elucidation of natural products with special reference to steroids-  'In this work i t is emphasized that the general  fragmentation pattern, of the steroid skeleton, upon electron impact, is highly susceptible to.the directing influence of substituents.  The ability  of the substituent to stabilize a positive charge in the molecular ion as well as in the fragment ions, determine the degree of i t s influence on the fragmentation process.  For example, nitrogen substituents as well as the  ethylene ketal and aromatic functions stabilize the positive charge so effectively that they often have the ability to direct the fragmentation in a specific manner.  In this way this influence often overcomes fission  reactions promoted by other more common steroid substitutents. The marked effect of the nitrogen atom in directing the fragmentation process Was well demonstrated from the numerous investigations in the alkaloid field.  Also of relevance to this discussion was the examination  of the mass spectra of a number of steroidal alkaloids and dimethyl amino steroid derivatives, in which the nitrogen function is connected in different ways to the steroid skeleton.  -  49  -  The successful synthesis of 6-aza and 11-aza steroids in our laboratory has furnished us with compounds in which the nitrogen atom i s incorporated into the steroid skeleton.  We considered i t of considerable interest to  examine the mass spectra of these compounds since such an investigation may not only provide some interesting fragmentation processes but will also allow us to extend this technique to other structural problems in aza steroids.  - 50 Discussion . For the sake of clarity this discussion will consider the mass spectra 6-aza and 11-aza steroids in separate sections. It must be indicated at the outset that throughout this entire discussion, I present "mechanistic rationalizations" in an attempt to provide some insight into the possible modes of fragmentation and explain the appearance of significant peaks in the mass spectra.  It must be emphasized that these postulates are by no  means established but are mainly put forth as being reasonable on the basis of the extensive investigations of Djerassi and other researchers in this field. A.  The mass spectra of 6-aza-5£;-steroids For studies in the 6-aza series, the mass spectra of 6-aza-5C-choles-  tane (86) and 176-hydroxy-6-aza-5£-androstane (87) were f i r s t examined (Figure 31). The 6-aza cholestane derivative (86) indicates significant peaks at m/e 538 (M-15), 344 (M-29), 330 (M-43) 316 (M-57), 302 (M-71), 164 and 124. Similarly the corresponding fragments in the androstane compound.(87) are also present (m/e 262, 248, 234, 220, 206 164 and 124. The M-15 peak is due to the loss of a methyl group, most probably the one at C19. The most .abundant M-57  fragment can be derived by homolytic cleavage  of the C 4 - C 5 linkage to provide the primary radical II. The a l l i l i c C^-C^o bond now cleaves to yield the radical III which in turn can lose a hydrogen atom to provide the fully conjugated ion IV (M-57).  -  51  -  Figure 3 1 .  On the other hand, homolysis of the C 5 - C 1 0 linkage provides the tertiary radical V which can be the common precursor for the fragment ions occurring at m/e 206  (M-71)  344  and  248  (M-29),  in the spectra of 8 6 and  330 87  and 2 3 4  (M-43)  respectively.  as well as  302  and  Loss of CH CH . (or 3  2  CH =CH plus one hydrogen) provides the M - 2 9 fragment ion V I and i t s 2  2  formation occurs only to a small extent.  Transfer of a hydrogen atom  from Cg to C10 provides an equally well stabilized tertiary radical V I I which can give rise to the primary radical the Cx-Cio bond.  The C 3 - . C 4 .  to the a l l y l i c radical I X .  VIII  by homolytic cleavage of  linkage is now activated and suffers homolysis The latter species loses one hydrogen atom to  provide the fully conjugated M - 4 3 ion X . Furthermore a hydrogen transfer from Cg to C^Q in V can generate the tertiary radical X I . Homolysis of the C -Cio bond will provide the secondary 9  Relative Intensity Q  no a  -t>  cr>  o  o  co  o  • < — i — i — i — i — i — i — i — i —  O O  1  O O  p-124  "fl  c CD  O  164  ro  12  o  O  3 \  CO  no cn O  _ f ^ - 2 8 8 (M-85)  o o  | r - 302 (M-71 )  3l6(M-57) 330(M-43) >- 344 ( M-29) O  E=  358 (M-15) 373 M'  o  Relative Intensity l\)  T  Q 1  1  o 1  CT)  1  o 1  _  00  1  o  1-  i  O O i  o o  TJ HOq  C  i-i CD  I— 124  cn  o  164  3 o o  W- 206(M-7I)  Y ==^220 l  (M-57)  234 (M-43) IN) i - 2 4 8 ( M - 2 9 )  CJilO _ 259(M-l8) "= 262(M-I5) r  277 M IS  +  - 54 -  -  55 -  radical XII which could then cleave at C 3 - C 4 to afford the a l l y l i c radical XIII.  The latter intermediate would Ipse one hydrogen atom at C7 to provide  the stable substituted dihydro pyridinium ion XIV (M-71). Another fragmentation process which is expected in this system, can be initiated by the cleavage of the C 7 - C 8 linkage to generate the secondary radical XV.  XVIII  XVII  In fact i t i s felt that the fragment occurring at m/3 124 i s possibly due to this type of fragmentation.  Homolytic cleavage of the C 9 - C 1 0 bond  in XV would provide the tertiary radical XVI, which loses the C5 hydrogen atom to afford the conjugated immonium ion XVII (irf/3 124). Another common signal (m/e 164) in the mass spectra of 86 and 87 i s possibly derived by the homolytic cleavage of the Ce-C^ and C 1 2 - C 1 3 bonds to the fragment XVIII. The spectrum of the cholestane analogue (86) also indicates a signal at m/e 288 which can be attributed to the loss of-CgH^. from the molecular ion and we believe that this represents cleavage of the C 2 0 - C 2 2 bond in the side chain.  - 56 The N-benzyl-6-aza steroids indicate a very similar fragmentation  , i  pattern but of course with the additional presence of a number of signals due to the presence of the benzyl group.  For example the mass spectra of  the N-benzyl-6-aza-5C-cholestane (88), 178-hydroxy-N-benzyl-6-aza-5?androstane (89) and 206-hydroxy-N-benzyl-6-aza-5£-pregnane  (90) possess a  common peak (m/e 91) due to the fragment C7Hy . Also the significant M-77 +  and M-91 fragments which occur at m/e 387, 289, 318 and 373, 275, 304 in the three spectra respectively are due to the loss of a phenyl (CgHs*) or benzyl (CgHsCI^-) moiety from the molecular ion. It is to be noted that these peaks are absent  in the spectra of the parent 6-aza compound. Again,  the most abundant peak (besides the molecular ion peak) in the spectra of these compounds is the one corresponding to the M-57 fragment. The fragments XVII and XVIII which were expected to appear at m/e 214 (124 + 90) and 254 (164 + 90) are no longer present in the spectra of these compounds.  This i s not too surprising when one considers the low abundance  • of these fragments in the spectra of the parent 6-aza steroids and the competing strong fragmentation which may be expected across the N-benzyl bond in the N-benzyl-6-aza homologues. The spectrum of the pregnane analogue (90) also indicates a signal at m/e 350 which can be attributed to the loss of CH3CH2C" from the molecular ion presumably due to the cleavage of the C 1 7 - C 2 0 bond in.the side chain. It is pertinent at this point to discuss briefly some recently published work^ on the mass spectra of cyclic amines such as piperidine. Due to the rigidity of the tetracyclic system of the 6-aza steroids, many of the mechanistic interpretations for the principal ions observed in the' mass spectrum of piperidine cannot be applied directly to our series but some comparisons are certainly of interest.  Three fragments which are  Relative Intensity o  ro o  -i> o  o"> o  - i — I I  oo o  i—i  i  •91  o o  3  H CD  O  ro O O  ro cn O ro CD  CM  o o  cn O  -r • — 373(M-9I) -379(M-85) • 387(M-77) " 3 9 3 (M-71)  O O  •407(M-57) —  42KM-43)  ?-435(M-29)  •4>  cn O  ==—449{M-I5) 4 6 4 M'  - LS -  O o i  Relative Intensity ro  CD  4>  00  pi — i — o o o i — i — i —  O  o o  91  o o Ti OQ  c I-i  CD  Ul  o ro O O  ro  05  ro ui  o 275(M-9I)  o o  2 9 5 (M-71)  289(M-77) 3 0 9 (M-57)  -323(M-43) 3 3 7 (M-29)  OJ Ul  o  _348(M-|8) L  35I(M-I5)  366 M  8S  -  +  Relative Intensity ro o  oo  CD  o  o o  o  o  O O 120  149  o  ro  o o  ro  3  i\)  CD  O  •258  OJ O O  304(M-9I) 3l8(M-77) •324 (M-71)  ~ .=  L  x 6  -  3  Lx 1.5  338(M-57) OJ cn O  r r  x  5  3  L x 2.1  350(M-45) 352(M-43)  366(M-29) 380(M-I5) x 1.8  O O  L  - 6S -  -x7,395,M x2.l  +  - 60 observed in the spectrum of piperidine at m/e 84 (XIX) (M-1), 56 (XX) and 30 (XXI) are of some relevance since they arise via fragmentation processes which are possible in the above compounds.  XXII  XVII  In the spectra of 6-aza steroids the fragments corresponding to XIX and XX can be represented by XXII and XVII.  The fragment XXIT (M-1) i s  considerably less abundant than the corresponding one (XIX) in the of piperidine.  spectrum  As i t was mentioned in the above discussion the signal  m/e 124 in the spectra of compounds 85 and 86 was attributed to fragment XVII.  Obviously, no conclusive indication can be drawn for the formation  of the fragment XXI (m/e 30) from the mass spectra of the aza steroids since this region is too complicated.  - 61 B. The mass spectra of 6-aza-5g-7-one steroids The presence of a carbonyl group at the C7 position of the 6-aza steroid skeleton has a dramatic effect on the fragmentation pattern of these compounds (Figure 37).  The mass spectra of these compounds reveal several  significant differences when compared to those discussed in the.preceding section.  These are: a) a very strong molecular ion peak as well as intense  M-l and M-2 peaks; b) considerable variation in the fragmentation process and c) the significant fragments retain the A and B rings of the molecule  Figure  37.  and arise from cleavage of bonds in rings C and D.  The fragmentation of  ring A occurs to a minor extent in this series. Evidence in support of the retention of ring A,and B in these fragments.is provided from several separate sets of results: a) the fact that the occurrence of these fragments is independent of the nature of the side chain at C i ; b) the fragmentation of the enol lactams (to be discussed 7  - 62 later) which possess an additional double bond at C^-Cs indicates, as expected, that these fragments now occur at m/e values which are lower by two units and c) the fragmentation of the N-benzyl-6-aza-7-one derivatives provides fragment ions at m/e values which are higher by ninety units.. I will attempt to provide possible mechanisms for the formation.of these fragments which occur at m/e 166, 179, 192, 206, 220 etc. in the spectra which were obtained. The formation of the m/e  166 fragment (XXIII) could be visualized as  arising through the homolysis of the C Q - C ^ and Cg-Cn  as shown in the  66 67 following scheme. However, in view of some very recent work ' the suggested *  hydrogen transfer may not occur because of the large interatomic distance between the hydrogen at C-15 and the oxygen atom at C-7.  Perhaps this frag-  ment arises via homolytic cleavage of the Cg-Cm bond.  rl  i x  x  r  n  The fragment XXVII (m/e 179) and XXIX (m/e 192) may be formed from the ion radical XXV which arises by a hydrogen transfer from C  9  to C in 8  the enol form XXIV of the molecular ion and subsequent homolysis of the  Relative Intensity  o  r*.  i  ro o i.  i  -i> o i  i  a> o i  O o  oo o  i—i—i—i  O O  o 166 179 192  ro o o  206  220  219 232 234 247  3  260 —261  292(M-95)  OJ  o o  302(M-85)  OJ s s s cn O  = = —  344(M-43) 359(M-28) 372 (M-15) MV387  ,x!0 ^x2.9  O O  - £9 -  Relative Intensity ro  i  O  1  CD  -4>  O  1  1  00  O  1  1  1  Q  1  o o  0Q  c H (6  CJ1  o  166 •179  3 CD  192 I96CM-95) — 206  O O E E E E =  —220 232  no  CJl O  o o  247  256(M-33) 26KM-28) 271 (M-18) 274(M-I5)  fr9  r  - 65 -  XXVII  XXIX  XXVIII  Cg-Cm bond to provide the radical XXVI. of the Cj j - C  12  The latter species by cleavage  bond provides the fragment XXVII. On the other hand, homolysis  of the C 2~C 3 bond provides the radical XXVIII which,loses one hydrogen 1  1  to give the conjugated immonium ion XXIX. A possible rationalization for the formation of the,other fragment utilizes a cleavage of the C^-Cm  linkage to generate the species XXX  which can be the intermediate for a number of fragment ions.  This inter-  mediate can lose the tertiary hydrogen at C 1 7 to provide the ion XXXI and further homolysis of the a l l y l i c C - C 15  15  and Cn-Ci2  linkages gives the  immonium ion XXXII (m/e 206). If the intermediate XXX cleaves at the Cn-C^  position i t will  provide the radical XXXIII which may proceed to the ion XXXIV (m/e 220) :  by the indicated pathway.  -  66  -  XXXIII  1  XXXIV  Another fragment of the same mass as XXXIV can be formed from XXX through an analogous sequence of intermediates.  Homolysis of the Ci  bond in XXX provides the radical XXXV which by hydrogen transfer from Cg to Cig generates the tertiary a l l y l i c radical XXXVI.  Cleavage of  the Cj1-C12 bond in the latter species provides a conjugated immpnium ion at m/e  220.  If hydrogen transfer from C sequent fission of the Ci^-Cn  15  to C  13  takes place in XXX, and sub-  bond occurs, the species XXXVII i s formed  - 67 -  and this intermediate can be the precursor for a,number of fragments as shown in the scheme below.  Homolysis of the C i 2 ~ C i 3 bond provides the  a l l y l i c radical XXXVIII (m/e 233), which can lose one hydrogen atom to afford the conjugated ion XXXIX (m/e 232).  If cleavage of the C - C 12  bond takes place with simultaneous hydrogen atom transfer from the substitutent R (R= -CeHi7 or - OH) to C , 12  obtained (m/e  13  C  17  the immonium ion XL is  234).  The formation of the fragments at m/e 232, 233, 234 can be also visualized through another fragmentation process.  Fission of the  C^-C^  bond in XXIV provide the intermediate radical XLI which upon hydrogen atom transfer from the C  17  quent cleavage of the (m/e 233).  substituent R(R= -CgH^ C13-C17  or-OH) to C  15  and subse-  bond generates the a l l y l i c radical XLII  This latter species could lose one hydrogen atom to provide  69  the fragment ion XXXIX.  If the cleavage of the C 1 3 - C 1 7 bond takes place  with simultaneous hydrogen transfer from C^g to C13 in XLI the fragment ion XLIII arises (m/e 324). The abundant fragment XLIV (m/e 274) can be formed by homolytic cleavage of the C27 -R(R=  XXX  CQHU  ,-OH) bond in the intermediate XXX.  .  XLIV  In an analogous fashion rationalizations in the formation of,the significant fragments occurring at m/e 246 (XLV), 247 (XLVI), 248 (XLVII) and 260 (XLVIII) are presented in the scheme below.  XLV  -  70 -  XLVII  XLVIII  There are rather intense peaks due to M-28, M-29 and M-43 fragment ions in the spectra of these compounds and these can be attributed to the loss of CO, CHO and CO plus *CH3 from the molecular ion respectively. The mass spectra of 91 and 92 possess relatively strong signals at m/e 292 and 196 respectively. These signals differ from the corresponding molecular ions by 95 mass units and therefore, their formation can be considered as being derived by loss of the elements of ring A. The formation of this M-95 fragment can be visualized by cleavage  - 71 of the Cg-C-xo bond in the molecular ion XXIV  to provide the tertiary-  radical XLIX, subsequent fission of the C5-N bond and a hydrogen transfer from C\ to N via a four-membered cyclic transition state.  The above mechanistic interpretation for the formation of the M-95 fragment is in great agreement with the main fragmentation of aliphatic amines as described in a recent publication.^ Other characteristic peaks in the spectra of these compounds are the M-15  for the loss of -CH  M-33  for the loss of H 0 and CH  3  from the molecular ion, M-18  2  significant signal at M-85  3  for the loss of water,  in the androstane analogue, (92) and a  for the loss of CeH  13  from the side chain in the  cholestane derivative, (91). It is now of considerable interest to compare the mass spectral results obtained in the N-benzyl compounds 93, 94 and 95 with those discussed above.  The abundant fragments which were mentioned above were  postulated as resulting from cleavage of the bonds in ring C and D.  Such  a suggestion would receive support i f the significant fragments in the  Relative  ro o  Intensity  an  4>  o ~l  o  1  oo o1  I  o oI  o o  Ti H*  O  c to o  ro o o . =  228  ro cn  256  o  —269 -282  CD  O O  -296 - =—310 -322 '324 = - 3 6 6  OJ  342  cn O 364 -372 r  -l> O O  cn O  386(M-9l)  .x5.3 ~x3.4  -392(M-85)  •400(M-77)_x2.6 >-xl.06  s=—-434  (M-43)  .m—449  (M-28) -462(M-I5) -xl.5 ^-xl3 L  - ZL -  x4.8  477M+  Relative Intensity ro O  O  4>  o  O  o o  GO  o 91  o o  CD  o  ro  o o  ?  cn  256  CO  _282 286 OJ  o o  304(M-77)  Ifi  m  OJ cn O  290(M-9I)  — 296  322 —324 336 351  —364 = 366(M-I5) x4.3 381 M"  O  o  Relative Intensity  _ 91 xl.6  149  o  H  CD  ro  o o  l= \  228  ro cn  o  CD  o o -332 •334  •315  OJ  cn O  r360(M-9l) 361 (M-90) -364(M-87) 374(M-77) 1  O O  -408(M-43) >-423(M-28) =  436(M-I5)  4>  M f - 4 5 1 XI2.I •x3.8 +  cn  O  PL -  - 75 N-benzyl series were now occurring at m/e values which were higher by 90 mass units.  This situation would of course prevail only i f fragmentation in  rings C and D occurred prior to any significant loss of the benzyl group. Indeed inspection of the mass spectra, as for example in the case of compound 93, reveals that this is the case - the peaks now occur at m/e  256  (166 + 90), 269 (179 + 90), 282 (192 + 90) etc. Although the abundance of these fragments is not very high i t must be recognized that in general, the intensity of a l l peaks relative to the molecular ion signal is considerably lower in these spectra than in those of compounds 91 and 92. As in the case of the N-benzyl compounds 88, 89 and 90, the mass spectra of the N-benzyl-6-aza-7-one analogues 93, 94 and 95 indicate the presence of fragments at m/e 91, m/e 386, 290, 360 (M-91) and m/e 400, 304, 374 (M-77) respectively which are due to the presence of the benzyl group. Again the peaks due to the fragments M-18  and M-33  are present in  the spectrum of 94 whereas the spectrum of 93 indicates the signal due to the loss of -CgH^  from the side chain, while that of 95 reveals a M-87 H fragment due to the loss of the side chain (CH -C-0AC). 3  I would like now to discuss briefly at this point the work published 62 Djerassi et al  on the mass spectra of lactams such as 2-piperidone. These  results have some relevance to our work in the mass spectra of 6-aza-5£-7one steroids, since they provide evidence for the fragmentation patterns on simple lactam systems.  However, only the arguments which are pertinent here  will be presented. In the mass spectrum of 2-piperidone a peak at m/e 71 (LII or LIII, M-28)  is noted and is believed to arise from the loss of ethylene rather  than carbon monoxide from the molecular ion L or LI. This conclusion is based on the unobserved shift of the peak at m/e 71 to m/e 73 in the  - 76 3,3-dideuterio compound.  LII  LIII  LV  If this type of fragmentation occurred in compounds 91 and 92, the fragments LIV or LV which correspond to LII or LIII would appear at m/e  139.  However, the spectra of these aza steroids do not possess a significant signal at this point (m/e 139). signals due to M-28  On the other hand, the relatively intense  fragments (which are also observed in the spectra of  the N-benzyl analogues 93, 94 and 95) are probably due to the loss of carbon monoxide or ethylene which could be expelled from another part of the molecule.  Further work with deuterium-labelled compounds would be  necessary before any distinction could be made between the two alternatives. For the mechanistic interpretation of the formation of the fragment at m/e  70 (M-29) three mechanisms were postulated and are shown below.  These workers showed with deuterium labelling that mechanism 1 predominates in the formation of this fragment.  In the case of the 6-aza-7-one steroids,  mechanism 1 cannot provide a fragment analogous to LVI and therefore no information can be obtained from the spectra of these compounds as far as  - 77 -  Mechanism 1  Mechanism 2  Mechanism 3 "  H LVIII  mechanism 1 is concerned. On the other hand, mechanism 2 could provide the fragment LIX of mass 138.  The mass spectra of compounds 6 and 7 possess no signal at m/e 138  H  LIX  LX  and this is in good agreement with Djerassi's results which indicated fragment LVII contributes only 2% to the composition of the fragment occurring at m/e 70. Finally, the mass spectra of compounds 91 and 92 (and the N-benzyl compounds 93, 94 and 95) indicate signals due to M-29 fragments.  These  results provide some support, for mechanism 3, which could generate the fragment LX. It is interesting to indicate here that the intensity ratio of the signals due to the M-28 and M-29 fragments is reverse to the one observed in the spectrum of 2-piperidone.  - 78 C.  The mass spectra of A -6-aza-7-one steroids lf  The previous discussion has provided rather lengthy explanations for a variety of fragments in which rings A and B are retained.  Consequently, i t  is immediately obvious that i f such species are significant in the fragmentation of the enol lactams, they must occur at m/e values which are lower by 2 mass units.  Indeed this situation is evident since the mass spectra  of 6-aza-4-cholestene-7-one (96) and  176-hydroxy-6-aza-4-androstene-7-one  (97) indicate a number of common signals at m/e 164, 177, 190, 204, 218. 230, 232. 244, 245, 246, 258 and 272 due to the A* analogues of the fragments 1  discussed in the case of the 6-aza-7-one compounds (91) and (92). The relatively abundant M-28, M-29 and M-43 fragments are again probably due to the loss of  Figure  CO, CHO and CO plus CH  3  43.  respectively.  Relative Intensity ro -i> CD co  o  —  i  —  i  —  o i  —  o o  i  —  i  —  o i  —  108  i  o —  — O i  —  i  o  109  cn O  164 177 190  no O O  204 218 230  CD  ro cn O  >246  -232  244  272 OJ  300(M-85)  O O  OJ  cn O  342 (M-43) 356 (M-29) 357(M-28) 370 (M-15) 385M+ I x2l.5  =  r  \x6 O O - 6L -  Relative Intensity ro  CD  -4>  a  a  — i — i — i — r  o  00  o  o o  I—I—I—I—I  oh o  o  164 177  ro  CD  o  190 204 218 .230  ro cn O  232 246 (M-43) 256CM-33) 261 (M-28) 274 (M-15) 289M  + x 5  .  x2.  o o - 08 -  - 81 The abundant M-95  fragment present in the spectra of the 6-aza-7-one  steroids is no longer present in the spectra of the A * compounds, (96) and 1  (97).  This is to be expected since the postulated mechanistic interpretation  for the formation of this fragment in the case of 6-aza-7-one steroids, cannot be applied in the case of the A  4  analogues.  In the case of the N-benzyl-A *-6-aza-7-one l  steroids the expected  signals at m/e 244 (164 + 90), 267 (177 + 90), 280 (190 + 90), etc. due to the fragmentation in ring C and D are very weak. The most intense peak, besides the molecular ion signal, is now due to the, M-28  fragment and is  probably due to the loss of CO from the molecular ion.  Similarly the M-29  and M-43  fragment, are relatively abundant in the spectra of these compounds  and they are again probably due to the loss of CHO and CO plus CH3 respectively. On the other hand, the N-benzyl-A^-6-aza steroids indicate three very characteristic signals at m/e occurring at m/e ,199.  198, 199 and 200 with the most intense peak  The formation of the corresponding fragments LXI  (m/e 198), LXII, (m/e .199) and LXIII (m/e 200) can be visualized via the following scheme. If such, a fragmentation was taking place in this series of compounds, one would expect that the spectra of the A^-6-aza steroids would possess signals at m/e  108 (198-90), 109, (199-90) and 110 (200-90).  Inspection of  this region in the spectrum of 97 reveals clearly that this is actually the case. The relatively intense signals at m/e  171 and 184 are also very  characteristic in the spectra of the N-benzyl-A -6-aza-7-one l+  series.  - 38 -  Relative  Intensity  - £8 -  Relative Intensity ro  o  i—i—i—r  O  o  T — i — i — or 91  O O  149  cn O  OP  c CD  o o  00  0)  ===  171  00  284  ro  o o  199  ro cn O 279  a> OJ  O  o  OJ  cn O  359(M-90) 362(M-87)  4>  O O  -406 ( M - 4 3 ) -42KM-28) xl.7 •434 (M-15) .M*449 r ' 2 x  cn O  X3.8  -  fr8  -  8  - 85 -  It has been mentioned in the previous discussion that the presence of the N-benzyl group in the 6-aza and 6-aza-7-one series is associated with the formation of fragments such as  (m/e 91), M-77 and M-91 due  to the loss of CeHs- and C-jHy from the corresponding molecular ions respectively.  Therefore, one would expect the fromatipn of the above  fragments in the N-benzyl-A -6-aza-7-one series as well. 1+  C7H7  +  Although the  fragment does appear as indicated by a very strong signal at m/e 91  in the spectra of the latter series, signals due to fragment ions corresponding to M-77 and M-91 are no longer present in these spectra. In order to provide an explanation of the absence of the above signals is necessary to comment on the fragmentation responsible for the formation of the fragment ions represented b y . C 7 H 7 , M-91 and M-77. +  It is well  known that a benzyl bond is easily cleaved to provide the benzyl ion  - 86 -  --•  LXIV in which the positive charge is stabilized by resonance.  Meyerson  has shown that this fragment is present rather as a tropilium ion (LXV) (m/e 91).  LXIV  LXV  On the other hand, the ability of a nitrogen atom ot stabilize a positive charge similarly can favor a fragmentation across the  N-CH<J> or 2  NCH-<j) bonds to provide the M-91  and M-77  CgH5CH - and CgHs* respectively.  The fact that these fragments are not  2  2  fragments and the stable radicals  M-77  present implies that the above mechanism is either not operative to any significant extent in these compounds, or else the species so generated are immediately fragmented further before they can be recorded in the mass  - 87 spectrometer.  We feel that a more preferable explanation lies in an  alternate mechanism (already postulated previously for the formation of the fragment LXI, LXII and LXIII) in which we suggest that a fission of the N-C=0 bond occurs in the i n i t i a l stages preferentially to the C-CH2<J> or NCH-<j> cleavage respectively. 2  The pregnane derivatives 95 and 100 indicate three very characteristic signals at m/e  149, 279 and 261 (M-90 in 95), 359 (M-90 in 100), which are  probably due to the nature of the side chain in these compounds. The mass spectra of A -6-aza-7-bne steroids, as in the case of the i+  6-aza-7-one series, provide an analogous comparison of results with respect to the formation of the signals at m/e 71 and 72 in the mass spectrum of 2-piperidone.  The intensity of the signals corresponding to the  M-28  ragment, in the mass spectra at the A -6-aza-7-one steroids (and particularly 1+  in the case of their N-benzyl analogues) is relatively higher than the one observed in the spectra of the 6-aza-7-one series.  We feel that, the loss  of ethylene via a retro-Diels Alder reaction in the former compounds may contribute significantly to the intensity of these signals.  It is also interesting to emphasize at this point the similarity of the mechanistic interpretation psotulated above for the formation of the intense signals at m/e  198 (LXI), 199 (LXII) arid 200 (LXIII) in the spectra  of the N-benzyl derivatives. 98, 99 and 100 (m/e 108, 109, and 110 in the spectrum of 96) and the one postulated for the formation of the most  - 88 intense signal at m/e 30 in the spectrum of 2-piperidone.  ^> H D.  +  m/e 30  The mass spectra of the 11-aza steroids For the studies of the mass spectra of the 11-aza series we f i r s t  examine the spectra of the compounds 79, 81 and 80 (Figure  49). The com-  pounds 79 and 80 indicate a series of common significant peaks at m/e 176, 190, 202, 203, 204, 206, 218, 216, 230, 232, 245, 249 and 262. On the other  Figure  49.  hand, compound 81 indicates an analogous series of signals which differ from those of 79 and 80 by only two mass units (m/e 178, 192, 204, 205, 206, 208, 218, 220, 232, 234, 247, 251 and 264).  Therefore, i t becomes obvious  - 68 -  Relative Intensity ro  o  4>  CD  00  o  o  o  o o  _r  149  ai o  —164 178  Ti  c  OQ H  ro O  CD  192  xl.9  p204  o  L  •206 208 220  .232  234  ro cn O  xl.9  247 251  3 OJ o o  300 -318 r  228(M-75)x7.5 Rcl.5 343(M-60) r -X4.6 Lxl.2 x 7  OJ  8  r  cn O  360 (M-43)  388(M-I5)  Oh  403M+  r ' x 4  5  *-xl.4  o  - 06 -  Relative  Intensity  O CD ro 00 O O O o o i — i — i — i — i — i — i — i — i — i — i 149  O  162 _  ro o O  |76 190 203 -218 232  l\3 cn O  OJ  O O  -245  249  298 316  226(M-75)x2.2 24KM-60) 242(M-59)  OJ  cn O  x3.l x2  -358 (M-43)  386(M-I5) O O  401  - 16 -  xl.8 M"  - Z6 -  Relative Intensity  139  o c CD  tn  ro  o o  -231  ro cn O •273 L276' 3  290  OJ  o o  -303 —314 -318 •332 3 4 5 r xlO  OJ Oi  L 3  O  O O  X  —398  -426(M-75)  cn O  .441 ( M - 6 0 ) "L.442(M-59) - 4 5 8 (M-43)  -486(M-I5)  cn O O  •501 M  -  H  £ 6  -  - 94 from the above comparison that ring A or part of ring A is not present in the above fragments since the substituents at C3 do not alter the mass of these fragments.  On the other hand, these species must retain ring D since  the positions ,at which they occur are dependent on the functionality in this ring.  The mass spectra of compounds 76 and 78 are in further support of  this conclusion. If we accept the above postulate one would expect to observe that the corresponding fragment ions in the spectra of the latter compounds must shift by the appropriate values.  In compound 78 the peaks  should occur at m/e values which are higher by 98 units due to the. presence of the sapogenin side chain.  Indeed the spectrum of this compound does  possess signals at m/e 276, 290, 302 (weak), 303, 304 (weak), 316 (weak), 318 (weak), 330 (Weak), 332 and 345.  On the other hand, the spectrum of  76 which bears no N-acetyl function (42 mass units) indicates a series of signals.at m/e 220, 234, 248, 260, 261 (weak), 262, 264 (weak), 274, 276, :  290 and 303.  These values differ as expected by 56 mass units (98-42 = 56)  from those of compound 81 and.by 42 mass units fromthose of compound 78. The latter observation also indicates clearly the presence of the nitrogen function in these fragment ions.  It is therefore, reasonable to conclude  that the fragmentation responsible for these signals occurs via homolysis of bonds involving rings B and/or C. The m/e values recorded in parentheses in the following discussion are taken from the spectrum of compound 81, unless otherwise stated. A reasonable postulate for the fragmentation responsible for the generation of the most of the above fragments can be visualized as shown in the scheme below.  Homolytic cleavage of the C 9 - C 1 0 bond in the ion  radical LXVII which may fragment in two different ways.  For instance*  cleavage of the C 5 - C 5 bond in the latter generates the primary radical  - 95 -  - 96 LXVIII which by loss of a hydrogen atom i s converted to the fragment LXIX (m/e 247).  The latter species by loss of a methyl radical provides the less  abundant fragment LXX (m/e 232). On the other hand, homolysis of the C 5 - C 7 bond in LXVII generates the primary radical LXXI which by loss of a hydrogen atom is converted to the highly abundant conjugated ion LXXII (m/e 234).  This fragment can generate  another conjugated fragment ion LXXIII (m/e 218) by loss of a methyl group and a hydrogen atom as shown, below. Furthermore homolysis of the activated C 7 - C 8 bond in LXVII provides the a l l y l i c radical LXXIV from which loss of a,methyl group with simultaneous hydrogen transfer from  to C13 generates the fragment LXXV  (m/e 204) . A simple loss of the Cj^ hydrogen atom from LXXIV would yield the fragment LXXVI (m/e 220).  This fragment ions LXXVII (m/e 205) and  LXXVIII (m/e 206) can be derived from LXXVI as shown in the scheme. A further series of fragments can be derived from the intermediate LXVII by homolysis of the activated Cg-Cn, bond followed by hydrogen transfer from C12 to Cg to provide the secondary radical LXXIX.  Further  fragmentation of the latter at the Cg-Cg linkage generates the ion LXXX (m/e 208).  The fragment LXXXI (m/e 192) can be obtained from LXXX by  cleavage of the C 1 3 - C 1 8 bond and subsequent loss of the  hydrogen atom.  Another abundant pair of fragments LXXXI11 (m/e 208) and LXXXIV (m/e 178) are probably derived from a common ion radical LXXXII which is formed as shown.  Loss of a hydrogen atom from C17 in LXXXH provides the con-  jugated ion LXXXIII (m/e 178) while loss of the C i methyl group generates 2  the fragment LXXXIV (m/e 164). in the case of C^-C^  It is interesting to indicate here that  dehydro analogues 79 and 80 the fragment LXXXV  (m/e 162) arising via the above postulated fragmentation is much more  -  LXXXV  97  -  LXXXIII  - 98 abundant than the fragment LXXXIV (m/e 164) in the dihydro series. situation is possibly due to the fact that the  C16-C17  This  double bond in the  former compounds i s able to provide an aromatic btcyclic system. It is of further interest to emphasize at this point that the intensity of the signals due to the fragment ions discussed above, with the exception of those at m/e 345 and 303 is considerably lower in the spectrum of compound 78. Similarly an analogous difference in the intensity of ;  signals (with the exception of the one at m/e 290) i s observed in the mass spectrum of compound 76 and those of 79, 81 and 80.. It is not possible at this time to provide any conclusive explanation for the high intensity of the signals at m/e 345, 303 in the spectrum of compound 78 and at m/e 290 in the one of 76 but i t is suggested they are perhaps due to the fragments LXXXVI (as LXIX in 81), LXXXVII (as LXXVII in 81) and (LXXXVIII (as LXXII in 81) respectively.  0  ftcN  LXXXVI  LXXXVII  LXXXVIII  The important signal at m/e 139 (LXXXIX) common in compounds 76 and 78 is due to the fragmentation of the spiroketal side chain and i s a characeteristic feature in the mass spectra of sapogenins.  59  +  LXXXIX  The signals at m/e 274, 288, 303 and 345 in the spectrum of compound 76 can be derived via a fragmentation process of the spiroketal side chain u i .. 59 as shown below.  - 100 -  - 101 The abundant fragment M-60 (XC) and M-75 (XC-CH -) in the spectra of 3  compounds 81, 80 and 78 are very reminiscent of the spectrum of cholestan58 30-ol acetate  and are due to the loss of  group from the molecular ion.  CH C00H 3  and  CH3COOH  plus a methyl  In the case of compound 81 the corresponding  M-75 signals occur  at m/e 341 and 326 respectively.  The metastable, peak at  m/e 3l3 (which is also present in the spectrum of compound 80 at m/e 312) strongly supports the relationship between those two fragments.  The sub-  sequent retro-Diels-Alder reaction which would normally provide the fragment XCI is apparently not occurring in our system since there is no signal corresponding to this fragment in compounds 81, 80 and 78. On the other hand, the compounds 79, 81 and 80 and 78 indicate signals at m/e 316, 360, 358 and 458 (M-43) respectively, which are due to the loss of the CH3CO. group of the N-acetyl function from the molecular ion. The M-33 fragment (m/e 326) in the spectrum of compound 79 i s accordingly due to the loss of water plus a methyl group.  The signal at m/e 341  (M-18) is also relatively strong in the spectrum of this compound and i t is due to the loss of water. In connection with the mass spectra of the 6-aza steroids i t i s pertinent to discuss briefly at this point the work reported recently  64  -102 the mass spectra of 6-azar-equilenin (101) and 6-aza-14(3)-isoequilenin (102). The synthesis of these compounds is briefly described in the introduction 32 part of section A in this thesis. The fragmentation of these compounds is closely related with the one of equilenin,and 14(8)-isoequilenin which has been described in detail by Djerassi et a l . ^ In other words, the presence of hetero atom in these compounds does not provide a significant difference in the fragmentation pattern since this atom is part of a stable aromatic system.  The spectra  of compounds 101 and 102 indicate common signals at m/e 281 (M ) 266 (M-15) +  225 and 210.  Furthermore; the C/D trans isomer (101) indicates significant  signals at m/e 238 and 212.  The formation of the latter signals in the  C/D trans isomer is attributed to, stereochemically dependent processes. The examination of the mass spectra of 14,15-d2~azaequilenin (103) provided evidence for the plausible structures of these ions and for possible fragmentation pathways leading to their fprjnation as shown below.  - 103 Experimental The mass spectra of the compounds 78, 76 and 80 were obtained with a MS 9 mass spectrometer using a direct insertion probe.  The samples were  introduced directly into the mass spectrometer (MS 9) ion source and evaporated into the ionization region from the end of the sample probe situated only a few millimeters away. The ionizing emergy was maintained at 70 e.v. The rest of the mass spectra were obtained with an Atlas CH4 mass spectrometer using the direct insertion technique. maintained at 70 e.v.  The ionizing energy was  - 104 REFERENCES 1. J . Fried and A. Borman. 2.  V.A.'Drill and B. Riegel.  Vitamins and Hormones, 1 J , 303 (1958). Recent progress i n Hormone Research, 1 J ,  29 (1958) 3.  R. Hirshman, G.A. Bailey, R. Walke and J.M. Chemerda.  J. Am. Chem.  Soc., 81, 2832 (1959). ..  — , .  4. R.I. Dorfman, M. Uskokovic and M. Gut.  137th Meeting of the Am. Chem.  S o c , P. 19N, April (1960); R. Pappo. 5.  147th ibid. , P.17M,  April (1964). N. Doorenbos. 6th Drug Trade News Survey, Sejjjijtember (1962).  6. W.L. Holmes. "Drugs Affecting Lipid Synthesis", Smith Kline and French Lab., Philadelphia, Penn. (1964). 7.  C. Djerassi.  Steroid Reactions, Holden-Day, Inc., San Francisco (1963).  8. M. Alaudin and M. Martin-Smith.  J. Pharm. Pharmacol., 14, 325, 469  (1962). 9. CW. Shoppee, R.E. Lack and S.K. Roy. J . Chem. Soc, 3767 (1963). 10. 11.  N.J. Doorenbos and R.E. Havranek. J. Org. Chem. 3 J , 2474 (1965). S. Hara. Yakugaku Zasshi,,78, 1027 (1958); Chem. Abst., 53, 3273c (1959).  12.  CW. Shoppee and J, Sly. J . Chem. Soc., 3458 (1958).  13.  CW. Shoppee, R.E. Lack and B.C. Newman. J. Chem. Soc., 3388 (1964).  14.  R.H. Mazur.  15.  L. R. Morgan.  J. Org. Chem., 28, 248 (1963). Chem.and Ind., 293 (1963).  16. N.J.Doorenbos and Mu Tsu Wu. 17. N.J. Doorenbos and Mu Tsu Wu.  Ohem. and Ind., 648 (1965). . J. Heterocyclic Chem., 2, 212 (1965)  18. N.J. Doorenbos and R.E. Havranek. J. Phar. Sci. 54, 1163 (1965). 19. N.J. Doorenbos and Mu Tsu Wu. 20.  Chem. and Ind., 650 (1965).  N.J. Doorenbos and K.A. Kefridge.  J . Heterocyclic Chem. , 2, 126 (1965)  - 105 21.  V.C. Patel.  Diss. Abst., 24, 986 (1963).  22.  W.J. Rodewald and J. Wicha.  Bull. Acad.Polon. Sci. Ser. Sci. Chim.,  n, 437 (1963). 23.  W.J. Rodewald and J. Wicha. Bull. Acad.Polon. Sci. Ser. Sci. Chem., 12, 95 (1964).  24.  J.P. Kutney and R.A. Johnson.  Chem. and Ind., 1713 (1961).  25.  J.P. Kutney, R.A. Johnson and I. Vlattas.  Can. J . Chem. , 41, 613  (1963) . 26.  T.L. Jacobs and R.B. Brownfield. J. Am. Chem. Soc. , 82^ 4033 (1960).  27.  H. Lettre and L. Knof.  Chem. Ber., 93, 2860 (1960).  28.  A.D. Cross, F.A. Kind  and A. Bowers.  Chem. Abst. , 61, 16131 g  (1964) . 29.  L. Knof.  30.  J. H. Burekhalter and H. Watanabe. Chem. Eng. News. Feb. 4, (1963) P.40. W.S. Jonson, J.N. Petersen and CD. Gutsche. J. Am. Chem. S o c , 69,  31.  Annaleh, 670, 88 (1963).  2942 (1947). 32.  W.S. Speckamp, H. De Koning, U.K. Pandit and H.O. Huisman.  Tetra-  hedron, 21, 2517 (1965); Tetrahedron Letters, 1275 (1964). 33.  H. Smith, G.H. Douglas and CR. Walk.  Experientia, 20_, 418 (1964).  34.  CW. Shoppee, M.I. Akhtar and R.E. Lack.  35.  Y. Morisawa, Y. Kishida and K. Tanabe.  J . Chem. S o c , 3392 (1964). Chem. Pharm. Bull. Japan,  U , 686 (1963). 36.  R. Clarkson. J. Chem. Soc , 4900 (1965) •  37.  R.J. Meltzer, D.M. Lustgarten, R.J. Stanaback and R.E. Brown. Tetrahedron Letters, 1581 (1963).  38.  A.I. Meyers, G.G. Munoz, W. Sobotko and K. Baburao. Tetrahedron Letters, 255 (1965).  39.  S.V. Kessar, M. Singh and A. Kumar. Tetrahedron Letters, 3245 (1965)  40.  A.J. Birch and G.S.R. Subba Rao.  J. Chem. Soc., 3007 (1965).  - 106 41.  S. Rakhit and M. Gut. J. Org. Chem., 29, 859 (1964).  42.  R. Anliker, M. Muller, J. Wohlfahrt  and H. Heusser.  Helv. Chim. Acta.,  58, 1404 (1955). 43.  S. Rakhit and M. Gut. Steroids, 4, 291 (1964).  44.  R.E. Counsell, P.D. Klimstra, L.N. Nysted and R.E. Ranney. J . Medicin. Chem., 8, 45 (1965).  45.  J.P. Kutney, I Vlattas and G.V. Rao, Can. J. Chem., 4J, 958 (1963).  46.  J.P. Kutney and I. Vlattas.  47.  Part of this work was presented at the Symposium on Heterocyclic  Steroids, 5, 595 (1964).  Steroids, Abstracts, 147th Meeting of the Am. Chem. S o c , P. 18M, April fl964). 48.  N.J. Leonard and V.W. Gash, J. Am. Chem. S o c , Z6, 2781 (1954).  49.  N.J. Leonard and D.M. Locke, J. Am. Chem. Soc.,77, 437 (1955).  50.  Ch.R. Engel, and S. Rakhit. Can. J. Chem., 4_0_, 2153 (1962).  51.  N.J. Leonard, A.S. Hay, R.W. Fulmer, V.M. Gash. J . Am. Chem. Soc , 1_J, 439 (1955) .  52.  CR. Eddy, M.E. Wall and M.K. Scott. Anal. Chem., 2£, 266 (1953).  53.  R.N. Jones, E. Katzenellenbogen and K. Dobriner. J . Am. Chem. S o c , 75, 158 (1953).  54.  J.P. Kutney.  55.  J.P. Kutney and W. Cretney, unpublished results.  56.  F. Bohlmann, E. Winterfeldt, G. Boroschewshi, R. Mayer-Mader and B. Gatscheff. Ber., 9J, 1792 (1963). M.E. Wall, H.E. Kenney, and S. Rothman, J. Am. Chem. S o c , 77,5665  57.  Steroids, 2 255 (1963).  (1955) and references cited therein. 58.  NMR Sepctra Catalog, Varian Associates (1962).  59.  H. Budzikiewicz, C. Djerassi and D.H. Williams. "Structure Elucidation of Natural Products by Mass Spectroscopy," Vol. II, Holden-Day, Inc., San Francisco, California, 1964.  60.  A.M. Duffield, H. Budzikiewicz, D.H. Williams and C. Djerassi. J. Am.  - 107 Chem Soc., 87, 810 (1965). 61.  C. Djerassi and C. Fenselau.  J. Am. Chem. S o c , 8_2, 5752  62.  A.M. Duffield, H. Budzikiewicz and C. Djerassi.  (1965).  J. Am. Chem. Soc.,  8J, 5536 (1964). 63.  S. Meyerson.  Applied Spectroscopy, 9, 120 (1955).  64.  U.K. Pandit,  W.N. Speckamp and H.O. Huisman. Tetrahedron, 2J, 1767 (1965).  65.  C. Djerassi, J.M. Wilson, H. Budzikiewicz and J.W. Chamberlin, J. Am. Chem. Soc. , 8j,4549 (1962).  66.  C. Djerassi and L. Toke*s. J. Am. Chem. S o c , 88, 536 (1966).  67.  C. Djerassi, G. Von Mutzenbecher, J. Fajkos, D. H. Williams and H. Budzikiewicz. J. AM. Chem. S o c , 87, 817 1965.  PART  II  Studies in the Alkaloid Field  -. 108-  Introduction Of a large number of nitrogen bases occurring in Nature, a considerable portion contain the indole nucleus.  A wide category of  these compounds is the well known and very important family of so called "Indole alkaloids", which have so.far been isolated from upwards of twenty five genera  of plants and trees.  They include many  important and widely used alkaloids, such as the ergot bases, valuable as oxytocic drugs in childbirth; strychinine, valuable as a general tonic and also employed as a vermin k i l l e r ; yohimbine used in veterinary medicine as an aphrodisiac; the extracts of Rauwolfia serpentina Benth used in India for several purposes, chiefly as a sedative, etc. The pharmacological properties of a l l these plant extracts have stimulated chemical investigations into the structures of the alkaloidal constituents, and so far the structures of approximately three hundred indole alkaloids have been completely elucidated.''" Besides the isolation-of alkaloids and the chemical elucidation of their structure a considerable effort has been contributed to the total or partial synthesis of these substances.  In many of  these instances investigators have even tried to follow synthetic schemes in the laboratory which could possibly bear some relationship to the pathways utilized by Nature.  Although a good deal of  work s t i l l remains to be carried out in the area of alkaloid biosynthesis, a considerable amount of information is already available from experiments with radioactive tracers.  2  It has long been suspected that tryptophan ( 1 ) is the important amino acid which serves as the building unit for the 3  indole alkaloids.  Recent tracer experiments have provided  con-  - 109 formation of this postulate.  It is generally felt that after decar-  boxylation, is converted into tryptamine (2) which just like phenethylamine, is capable of undergoing a whole series of condensation reactions. (See Figure 1).  8 Figure  2.  1.  Biogenetic Derivation of Simple Indole Alkaloids.  For example, serotonin(5-hydroxytryptamine) (4), one of the more important naturally occurring indolylalkylamines is produced by hydroxylation of tryptophan (1) to 5-hydroxytryptophan (3) and subsequent lation.  decarboxy-  Tryptamine i t s e l f "cannot be hydroxylated. Psilocybine (5),  a phosphoric ester derivative of N,N-dimethyl-4-hydroxytryptamine, is  - 110 the active constituent of the of the Mexican hallucinogenic fungi of the genus "Psilocybe".  It is interesting to note that i t is one of the few  4-hydroxy indole analogues actually isolated in Nature. There are various hypotheses advanced for the biosynthesis of the "non-tryptophan" portion of the more complex indole alkaloids. We will concern ourselves here with only those postulates which are relevant to this thesis. 4  Wenkert  in 1962, elaborated an elegant scheme involving prephenic  acid (9) incorporation into the.non-tryptophan portion of several classes of indole alkaloids.  (Figure 2).  Rearrangement of prephenic acid (9) by a 1,2- shift of the pyruvate residue, with retention of configuration, followed by hydration affords a unit (10) readily discernible in yohimbine (12).  Condensation of the  above with a formaldehyde equivalent and retro-aldolisation then yields a "seco-prephenate-formaldehyde" (SPF) group (11) that can condense with tryptamine to eventually yield alkaloids typified by corynantheine (13) ajmalicine (14) sarpagine (15) and ajmaline (16). According to Wenkert the,role of the SPF unit (11) in the biosynthesis of corynantheine (13) and related alkaloids may be visualized as shown in Figure 3.  Formation of a SPF-tryptamine complex (18) is  followed by a Mannich-type condensation at the o-position of the indole system to give (19) which can then undergo appropriate modifications to yield the,various series known in Nature. The prephenic acid hypothesis also provides a comprehensive scheme for the biosynthesis of Strychnos and Iboga alkaloids and has the additional merit of encompassing the Aspidosperma series. The strychinine group (22) evolves from the tryptamine-SPF complex  - Ill -  Figure  2.  - 113 (20) as shown in Figure 4.  23;;  Figure  4.  22  Incorporation of the SPF Unit in the Strychnos Type Alkaloids.  Attack of the formyl acetate residue in the intermediate at the ot-position of the indole portion provides the iminium ion (21), which bears an obvious resemblance to the known alkaloid stemmadenine (23). A transannular cyclisation then provides the strychinine precursor (22). The relationship between tryptamine and the SPF unit in alkaloids  -  Figure 5.  114 -  Biogenetic Incorporation of the SPF Unit in the Aspidpsperma and Iboga Type Alkaloids.  - 115 of the Aspidosperma and Iboga type is not readily discernible.  It is  evident that these alkaloids arise from rearranged SPF units. The crucial rearrangement can be seen as proceeding via a retro-Michael reaction intermediate (24) (See Figure 5) and involves an activated hydrogen atom on a carbon atom a to either the iminium system or the acetyl group, with resultant cleavage of the SPF unit.  The cleavage  product (24) could be modified by unexceptional reactions to give either an Aspidosperma (25) or an Iboga (28) precursor.  These com-  pounds could then undergo a parallel series of reactions: Michael additions to the a3-unsaturated acid systems would afford the ninemembered ring compounds (26) and (29), which could then, by transannular cyclisations, yield the Aspidosperma and Iboga skeleta (27) and (30) respectively. There is as yet no direct proof of Wenkert's hypothesis and various researchers,have been.critical of its value in alkaloid biosynthesis.  The hypothesis does, however, stimulate further investi-  gations into some of the reactions proposed and recently some interesting work has been done with respect to the following equilibrium which is implied in the biosynthetic scheme.  37  6.  Correlation of the Configuration of Akuammicine with that of Condylocarpine.  - 117 A recent study^ of this equilibrium was involved in the interconversion of an alkaloid of the akuammicine type and an alkaloid of the aspidospermatine series.  The interconversion was carried out to verify  the absolute configuration of the alkaloid condylocarpine  (32) which had  been assigned on the.basis of a comparison of i t s optical rotation with that of akuammicine (31) whose configuration had been previously determined.  The results of this work are summarized in Figure 6. The equi-  librium between condyfoline (34) tubifoline (33) and the 20-epimer of condyfoline (36) could be accomplished by heating condyfoline under vacuo or under basic conditions, but not under acidic conditions. The transformation of condyfoline to tubifoline is postulated to occur as shown in Figure 7.  Figure 7.  - 118 It is felt that the c r i t i c a l step A +  B may involve abstraction -  of a proton by base as shown below.  \  N  /  The formation of the 20-epimer (36) of condyfoline undoubtedly involves an imine-enamine tautomerization of A prior to the transannular cyclization step (Figure 7). The biogentic significance of this work is that i t predicts that the antipodal relationship between the alkaloids of the akuammacine and those of the aspidospermatine type, merely arises from, appropriate oxidative cyclizations of a common precursor.  In support of this con-  tention both condyfoline and tubifoline give the same indole (35) on treatment with potassium borohydride in methanol. Some other recent work which has relevance to Wenkert's ionic 6— J  intermediate appears from a series of investigations by Kutney et. a l . This research provided the.first laboratory realization of the transannular cyclization process, and established that such a reaction may be of considerable u t i l i t y in the synthesis of complex indole alkaloids. It is appropriate at this point to discuss this work briefly since i t has a direct bearing on the reasons for the syntheses which were carried out and which are described in this thesis. In order to evaluate the feasibility of the transannular  cycli-  zation reaction, investigations in our laboratory, were initiated and dihydrocleavamine (38), readily available from other reasearch work on alkaloids from Vinca rosea linn, was utilized as the starting material. It was realized at that time that any successful results with this  \  - 119 compound could be extended to the natural Aspidosperma series, since the ring system present in dihydrocleavamine is identical with that already known in the Aspidosperma alkaloid quebrachamine (41). Indeed i t was 7  possible to convert the former to an Aspidosperma skeleton shown in Figure 8.  Figure  8. Synthesis of Aspidospermine Skeleton from Dihydrocleavamine.  (40) as  - 120 9 In subsequent work another successful series of reactions was carried out in which carbomethoxydihydrocleavamine (42) reacted as above with mercuric acetate to provide an iminium intermediate.  Cycli-  zation of the latter gave a vincadifformine-type skeleton (46) as the major product (Figure 9).  The, latter compound by reduction with zinc  in sulfuric acid gave (47) which, in turn, on treatment with hydrochloric acid and then reduction with lithium aluminum hydride, gave an Aspidosperma-type skeleton (48).  The latter product differed from  the previous compound . (40) only in stereochemistry. In the most recent paper^ i t was demonstrated that a,further possibility exists for the transannular cyclization process. mediate reaction product with the ,>N=Cs<  The inter-  grouping leads to the Iboga  skeleton and in this manner, coronaridine and dihydrocatharanthine were synthesized (See Figure 9).  The absolute configuration of the Iboga  alkaloids as proposed previously, by various workers is incorrect. The 12 correct stereo formula is given in structure 44.  Since a l l of these  transannular cyclizations made possible the interconversion and interrelation of the important and widespread groups, of Vinca, Aspidosperma, and Iboga alkaloids,, considerable attention was then directed toward the preparation of dihydrocleavamine,  carbomethdxy dihydrocleavamine  and quebrachamine since these "key" compounds provide the route to the total synthesis of these various classes of alkaloids. Before proceeding to our own work i t i s appropriate to review briefly at this point some very recent work which presents successful syntheses of some Aspidosperma and Iboga alkaloids. The f i r s t successful synthesis of dl-aspidospermine and dl-que13 brachamine by Stork  is shown in Figure 10. The stereochemistry of the  - 121 -  Figure  9.  Synthesis of Vincadifformine Skeleton from Carbomethoxydihydro cleavamine.  - 122 various bicyclic and t r i c y c l i c intermediates which are described in Stork's work is left open at this point . This stereochemical ambiguity is not significant since the indolenine (51) is formed under conditions which would lead to equilibration at the two centers C-12 and C-19 via a reverse Mannich reaction as shown below.  The most stable relative arrangement of the three asymmetric centers of (51) would thus be expected to result, regardless of the stereochemistry of the intermediates or the detailed course of the indolenine cyclization process.  There are good conformational arguments that this most stable  arrangement should coincide with that of dehydroaspidospermine. 14 Ban et al  have also investigated several possible routes for the  synthesis of dl-aspidospermine. intermediate  54  One of their approaches reached the  (See Figure 11) which has the same planar structure as  the corresponding one of Stork's synthesis, Physicochemical properties of this intermediate are quite different from those reported by Stork and this suggested that these compounds are diastereoisomeric. Ban carries through his synthetic sequence with intermediate  54  and indeed  succeeds in achieving a synthesis of dl-aspidospermine.  Figure  1 0 . Synthesis of dl-Aspidospermine  - 124 -  Figure  11.  - 125 -  Figure  12.  Synthesis of (±)-Eburnamine and (±)-3-Methyl Aspidospermidine.  ( 3 ^  H  R<L-C0CH  3  j W o l f f - K ishn«--r H«J.  The stereochemical assignments of the common intermediates in the Stork and Ban sequences were obtained from the NMR (50).  The conformations  55  and  Stork's lactam respectively. The synthesized from either equilibrium 51a  55  or  56  spectra of the lactam  were postulated to Ban's and  fact that dl-aspidospermine is 56  supports the existence of the  51b, postulated above and the stereospecific re-  duction of the most stable stereoisomer of the indolenine (51) with lithium aluminum hydride. An elegant synthetic path leading to Hunteria and Aspidosperma IS  alkaloids was recently reported by Barton and Harley-Mason.  Figure 12  indicates their synthetic sequence leading to (±)-eburnamine (59) and (*)-3-methyl aspidospermidine  (60).  A total synthesis of the Iboga skeleton has been also reported recently.  Buchi et a l ^ have achieved a total synthesis of (±)-ibogamine 1  (61) and (±)-epiibogamine (62) and their work is summarized in Figure 13. 17 Huffman et al  have also reported the synthesis of desethyl  ibogamine (63) and this sequence is summarized in Figure 14.  Figure 14.  Synthesis of Desethyl Ibogamine,  - 128 Discussion As already mentioned in the introduction, the transannular cyclization of the appropriate nine-membered ring intermediate provided an attractive synthetic entry into a wide variety of indole and dihydroindole.alkaloids. Consequently, i t was now desirable to develop a synthetic pathway to these "key" intermediates  for the proposed total syntheses of these natural  products. We have investigated in a preliminary manner, two synthetic approaches to the nine-membered ring system.  For a matter of convenience, the dis-  cussion of this part of the thesis will be divided into two sections, A and B, corresponding to.the two synthetic approaches, whose projected plans are outlined in Figures 15 and 16 respectively. Section A. In this synthetic sequence the i n i t i a l step requires the formation of an appropriate diester (66a or.b) which through subsequent reduction of the keto function and cyclization could generate the desired ninemembered ring intermediates  67 and 68.  Section A outlines our investi-  gations directed at the synthesis of.the necessary indole units (64a and b), the alkylation reactions with the piperidine unit and finally reduction of the keto function in the product, (66).  In our i n i t i a l phases, 3-carbo-  methoxy piperidine (76), easily available from other work in our laboratory was used for study of various reaction conditions. lead to the desired intermediates  Although i t does not  (67) or (68) i t did enable us to develop  the necessary experimental conditions which could then be applied to reaction with appropriate piperidine moiety, (65).  - 129. -  Figure 16.  '  -  130 -  There are many well known reactions which provide a synthetic entry into the indole nucleous.  18- 2 2  Of the various alternatives we chose to  utilize the Reissert synthesis which is well suited for the preparation of indole derivatives possessing carboxyl or carbomethoxy groups at the 23-24 a-position. The known synthesis of 2-carbomethoxy-indole via the Reissert reaction is outlined in Figure 17.  24  In order to introduce a suitably reactive side chain at the 3-position 25 of the methyl ester  74 we considered the Hoesch synthesis  known in the indole chemistry.  which is well  This method Which proceeds via electro-  philic attack at the 0-position, is well suited for the preparation of 26 indole derivatives bearing side chains with ketonic functions. the ester  74  was treated with chloroacetbnitrile  Therefore  and hydrogen chloride  gas in dry chloroform as solvent in a sealed tube at 50° C.  Although the  more common Hoesch conditions employ abpslute ether as solvent, our compound  74  was only slightly soluble in this solvent.  The crude semi-  crystalline solid obtained from this reaction was purified by chromatography on alumina to provide a colourless crystalline material (m.p. 147.5148.5°) in 30% yield. The spectral properties of this substance were very instructive and served to establish the expected structure (75) for this compound. For example the NMR spectrum indicated the normal multiplet centered at 2.65 x for the aromatic protons of the indole nucleus, a two-proton singlet at ii  4.98 i (-C.CH C1) and a three-proton singlet at 5.98 x for the ester methyl 2  protons.  The• infrared spectrum of the Hoesch product showed a very sharp  band at 2.98y (>NH) and two carbonyl bands at 5.82y and 6.0y in complete agreement with the assigned structure.  Finally the ultraviolet absorption  - 131 -  Figure  17.  Figure  18.  - 132 at 220, 249 and 318 mu was distinctly different from that of the starting indole. The Hoesch product (75) was subsequently reacted with three mole equivalents of the piperidine derivative (76) to provide a viscous oily material.  This latter substance was purified by chromatography on neutral  alumina and crystallized from an ether-hexane mixture to provide a crystalline product, m.p. 121-125°C.  The NMR spectrum of the latter showed the  normal multiplet due to four aromatic protons at 2.6 T , two three-proton singlets at 6.08 and 6.4 T due to the ester methyl protons of the indole and piperidine moieties respectively, 0 6.2 T (-CCH_2N<) .  and a two-proton broad doublet at  The infrared spectrum with a sharp band at 3.1u (>NH)  and two carbonyl bands at 5.78u (strong) and 6.15u was in complete agreement with the presence of two carbomethoxy groups and the keto as expected in the desired reaction product.  function  Finally the ultraviolet  spectrum with maxima at 220, 248 and 314 mu indicated,a similarity to the indole  75 .  Indeed the,spectral data-was sufficiently instructive to  assign the structure 77 to this compound. The next obvious step in the sequence was to remove the carbonyl function from the two-carbon bridge in 77 and thereby obtain the intermediate necessary for the cyclization studies to generate the nine-membered ring.  Therefore this compound was treated with sodium borohydride in  methanol to provide an amorphous solid, which was purified by thin layer chromatography on alumina.  In spite of numerous attempts to obtain this  compound crystalline we were unsuccessful and therefore characterization of this product was done on the.amorphous solid.  We assigned the structure  78 to this compound on the basis of the following spectral data.  The NMR  spectrum showed the normal aromatic proton absorption at 2.65 x , two three-  - 133 0 II  proton singlets at 6.25 and 6.45 T (2 x -C-OCH ) and a one-proton broad OH r quartet centered at 4.4 T (-C-C). The carbonyl absorption due to the keto 3  ;  iL  group in the infrared spectrum of 77 had now disappeared and a broad absorption band 2.95p indicated the presence of a new hydroxyl group in this compound.  Furthermore in this compound the strong broad carbonyl  absorption band at 5.85u was in agreement with the presence of two carbomethoxy groups already evident from the NMR spectrum.  Finally the ultra-  violet spectrum of this compound had an absorption pattern (^  max  229 and  298 mu) very different from that of 77 but very similar to the one of 2-carbomethoxy-indole (74) (A \max J  v  218 and 294 mu). ' J  The removal of the alcoholic group and generation of the intermediate possessing a  -CH2CH2-  now necessary.  bridge between the indole and piperidine moiety was  We considered that dehydration of,the alcohol 78 could  provide the corresponding enamine (R-CH=CH-N<) which on reduction would yield the desired compound. The,success of the dehydration could be ascertained,by ultraviolet spectroscopy since the enamine chromophore will 27 exhibit a characteristic absorption.  However, in spite of numerous  attempts we were unable to obtain the desired enamine. We next considered catalytic hydrogenation of the keto function since i t was possible that the expected alcohol might suffer hydrogenolysis and thereby yield the desired compound.  This keto compound 77 on catalytic  hydrogenation with Raney Nickel,as catalyst smoothly absorbed one mole of hydrogen to provide an amorphous solid, which was shown to be identical,to 78 on the basis of NMR, IR and thin layer chromatography comparison. Prolonged hydrogenation of 77 or 78 under similar hydrogenation conditions lead to the same product, which also remained as an amorphous solid in .spite of numerous attempts at crystallization of this material.  - 134 The NMR spectrum of this compound showed n£ aromatic proton absorption, a broad one-jaroton peak at 1.35 T (H-N<) a one-proton multiplet centered at i  4.85 T (-CHCH-N) and two three-proton singlets at 6.29 and 6.48 x , indi2  cating the presence of two ester, methyl groups in this compound. The infrared spectrum,showed absorption bands at 2.9 (sharp -NH) and 3.0y. (broad -OH) and a doublet in the carbonyl region at 5.6 and 6.0u.  The  ultraviolet spectrum of this compound had a strong maximum at 290 my an absorption which is characteristic for a 2-carbomethoxy-3,4,5-trialkyl28 pyrrole system. In order to draw additional information about the structure of this reduction product we subjected 2-carbomethoxy-ihdole  (74) to catalytic  hydrogenation with platinum oxide in ethanol in the presence of a trace of concentrated hydrochloric, acid.  When one. mole of hydrogen was absorbed  (at a rather slow rate) the hydrogenation was interrupted.  The reaction  provided a mixture of starting material and a new hydrogenation product in an approximate ratio of 2:1.  Thin layer chromatography purification and  crystallization from aqueous methanol gave an analytical sample of this product (m.p. 156-157°C) . Elemental analysis was in agreement with the molecular formula  C10H13O2N  and spectroscopic data allowed us to assign  structure 80 to this compound.  The NMR spectrum showed the absence of  the normal aromatic proton mulitplet, the presence of a one-proton doublet at 3.52 xdue to the 6-proton of the pyrrole system, a three-proton singlet at 6.27 x (-OCH3) and two sets of four-proton multiplets centered at 7.5 and 8.25 x due to the protons of the cyclohexene ring.  The infrared  spectrum showed a sharp band at 3.02 y (>NH) and a strong carbonyl absorption at 6.0y. The ultraviolet spectrum showed a strong absorption at 287 my which is again characteristic of a 2-carbomethoxy-pyrrole  - 135 28 system. Considering  the s i m i l a r i t i e s  i n t h e NMR,  IR a n d UV s p e c t r a o f 80 a n d  t h e h y d r o g e n a t i o n p r o d u c t o f 77 o r 78 we a s s i g n e d latter  t h e s t r u c t u r e 79 t o t h e  compound. During the course o f these i n v e s t i g a t i o n s , t h e synthesis  o f 8-(2-carbo29-31  methoxy-3-indolyl)ethylchloride as shown i n F i g u r e  This  ( 8 1 ) was a c h i e v e d  19.  compound was t h e n s u c c e s s f u l l y c o u p l e d  65 t o p r o v i d e  i n our l a b o r a t o r y ,  With the p i p e r i d i n e d e r i v a t i v e  one o f t h e d e s i r e d s y n t h e t i c i n t e r m e d i a t e s  82.  Since  this  - 136 sequence proceeded without above was  turned  our a t t e n t i o n t o the s y n t h e s i s o f i n t e r m e d i a t e  a l s o d e s i r a b l e f o r o u r work,.  Indole-2-aeetic  the s y n t h e s i s o f the i n d o l e moiety was 32 sequence  the s y n t h e t i c approach as d i s c u s s e d  abandoned.  We now was  difficulty,  which i s shown i n F i g u r e  Figure  20.  20.  obtained  acid  66b which  (87) n e c e s s a r y  according  for  to a published  - 137 The  acid  (87), on treatment w i t h diazomethane p r o v i d e d  a c e t i c a c i d methyl e s t e r  indole-2-  (88) as a n i c e l y c r y s t a l l i n e m a t e r i a l .  The NMR  spectrum o f t h e e s t e r showed the normal a r o m a t i c p r o t o n a b s o r p t i o n  centered  (  at 2.8 T, a one-proton s i n g l e t at 3.84- x (B-proton), 0  a two-proton s i n g l e t  s i n g l e t at 6.41 x (-0-CH ).  at 6.37 T (-CH C-0Me) and a t h r e e - p r o t o n 2  3  The  i n f r a r e d spectrum showed a sharp peak a t 3.01u (>NH) and t h e expected carbonyl  band a t 5.85u.  The u l t r a v i o l e t  spectrum was very c h a r a c t e r i s t i c  o f a normal i n d o l e system with maxima at 218, 272 and 289 mu. The introduce  Hoesch s y n t h e s i s  was now a l s o used s u c c e s s f u l l y i n o r d e r t o  the n e c e s s a r y carbon c h a i n . a t  the 6 - p o s i t i o n  o f the i n d o l e r i n g .  When t h e e s t e r 88 was t r e a t e d w i t h c h l o r o a c e t o n i t r i l e and d r y hydrogen chloride, using  anhydrous, e t h e r  a c i d methyl e s t e r 70%  yield  as s o l v e n t  (89) was o b t a i n e d  3-chloroacetylindole-2-acetic  as a n i c e l y c r y s t a l l i n e m a t e r i a l i n  ( F i g u r e 21)  A comparison o f the s p e c t r a l p r o p e r t i e s  o f t h e Hoesch p r o d u c t 89  w i t h those mentioned above f o r the s t a r t i n g e s t e r immediately its  structure.  and  two c a r b o n y l  The i n f r a r e d spectrum showed a sharp band a t 3.0y (>NH)  p  0  absorptions  (  at 5.79 ( - C O C H 3 )  u l t r a v i o l e t , spectrum showed a very 244,  established  and 6.2y (-CHCH C1).  d i f f e r e n t absorption  267 and 309 my) from t h e normal, i n d o l e a b s o r p t i o n  2  pattern  (*  The m a x  214,  o f the e s t e r 88.  F i n a l l y t h e NMR spectrum showed t h e absence o f t h e one-proton s i n g l e t at x present  3.84 ring.  present 5.3  i n t h e spectrum o f 88 due t o t h e B-hydrogen on t h e i n d o l e  There were i n s t e a d two new two-proton s i n g l e t at 5.3 and 5.8 x now i n the spectrum o f 89.  We a t t r i b u t e d the two-proton s i n g l e t at  x t o the methylene protons o f the c h l o r o a c e t y l s i d e chain . ( - C C H 2 C I )  whereas the o t h e r new two-proton s i n g l e t a t 5. 8 x must be due t o t h e methylene protons o f the e s t e r s i d e c h a i n  (-CH2COOCH3) which have s h i f t e d  - 138 -  Figure  21.  - 139 down f i e l d upon i n t r o d u c t i o n o f the 8- s u b s t i t u e n t . This  assignment was confirmed when the NMR spectrum o f the correspond-  i n g 3-acetyl- d e r i v a t i v e  (90) was i n v e s t i g a t e d .  p r e p a r e d i n two d i f f e r e n t ways.  This  l a t t e r compound was  The f i r s t sequence u t i l i z e d  catalytic  h y d r o g e n a t i o n o f 89 with Raney n i c k e l as c a t a l y s t at room temperature and atmospheric pressure. conditions The  The r e a c t i o n prbceeded very r a p i d l y under these  and i t was i n t e r r u p t e d when t h e hydrogen uptake became v e r y slow.  product was o b t a i n e d from the r e a c t i o n mixture as a c r y s t a l l i n e  material.  I t was r e c r y s t a l l i z e d from a mixture o f e t h e r - d i c h l o r o m e t h a n e t o p r o v i d e an a n a l y t i c a l sample (m.p. 128-131°C). The compound  Hoesch r e a c t i o n was a l s o used f o r the s y n t h e s i s (90). However, i n t h i s case t h e r e a c t i o n between  a c i d methyl e s t e r the  (88) and a c e t o n i t r i l e p r o v i d e d  desired product.  (89) ( *  indole-2-acetic  a very s m a l l  yield of  The i n f r a r e d s p e c t r a o f the l a t t e r showed two c a r b o n y l  bands at 5.66 and 6.1u and the u l t r a v i o l e t o f compound  only  o f the a c e t y l  m a x  spectrum was i d e n t i c a l w i t h  215, 243, 272 and 303/.mu).  that  The NMR spectrum showed  the normal a r o m a t i c m u l t i p l e t c e n t e r e d at 2.8 T, one two-proton s i n g l e t at 5.71 T and a t h r e e - p r o t o n s i n g l e t at 6.33 (-0CH ). 3  The two-proton s i n g l e t  at 5.3 T p r e s e n t i n the spectrum o f 89 had now d i s a p p e a r e d and a new t h r e e 0 p r o t o n s i n g l e t was p r e s e n t at 7.4 T ( f - C ^ ) . We had now accomplished the s y n t h e s i s and  the next s t e p was t o c o n s i d e r  unit.  o f the d e s i r e d  the c o u p l i n g  indole  r e a c t i o n with t h e p i p e r i d i n e  Consequently the compound 80 was t r e a t e d with three mole  o f the p i p e r i d i n e d e r i v a t i v e  moiety  (76) and the r e s u l t a n t gummy b a s i c  equivalents reaction  product which was o b t a i n e d c o u l d be p u r i f i e d by t h i n l a y e r chromatography on  alumina.  In s p i t e o f numerous attempts at c r y s t a l l i z a t i o n , t h i s substance  remained as an amorphous s o l i d .  The most c h a r a c t e r i s t i c f e a t u r e s  i n the  - 140 NMR  spectrum  o f t h i s compound were, the aromatic p r o t o n m u l t i p l e t  x, two  at 2.7  -  two-proton s i n g l e t s x  0  at 5.67  T and two  three-proton  ii  (2 x - C O C H 3 ) .  s i n g l e t s at 6.3  and 6.46  singlet  t o the methylene protons  at 5.67  and 6.17  centered  I f we  a s s i g n the  two-proton  a t t a c h e d t o the i n d o l e moiety  (-CH2COOCH3) s i n c e t h e r e would be expected  to absorb  same p o s i t i o n  o f the compounds 89 and 90,  the> two  as the c o r r e s p o n d i n g protons  proton s i n g l e t  at 6.17  x  i s c l e a r l y due  at approximately  the  t o the methylene protons  a t t a c h e d t o the b a s i c n i t r o g e n atom o f the p i p e r i d i n e p o r t i o n o f the 0 molecule (-C-CHj-N-)., The chemical s h i f t o f the methylene protons from i n the c h l o r o compound 89 t o 6.17  then  5.3  x i n the amine (92) i s i n agreement w i t h 33  s e v e r a l examples known i n the l i t e r a t u r e , o f the o r d e r o f about 60  cycles/sec.  where t h i s  The  i n f r a r e d spectrum  c a r b o n y l a b s o r p t i o n at 5.77u due  t o the presence  groups and a weaker band at 6.1u  f o r the keto group.  spectrum 90 (A max  chemical  o f the two The  shift is  showed a s t r o n g carbomethoxy  ultraviolet  has an a b s o r p t i o n p a t t e r n s i m i l a r t o the one o f compounds 89 and 214,> 243,> .268,> and . 303 mp). On the b a s i s o f t h i s d a t a we a s s i & gned  the s t r u c t u r e 92 t o t h i s The next  compound  r e a c t i o n which was  (See F i g u r e 20). necessary  f o r our s t u d i e s i n v o l v e d the  r e d u c t i o n o f the ketone group and h o p e f u l l y , subsequent h y d r o g e n o l y s i s  of  the r e s u l t i n g a l c o h o l i c f u n c t i o n so as t o p r o v i d e complete removal o f the oxygen atom at t h i s p o s i t i o n .  An  analogous h y d r o g e n o l y s i s o f an oxygen -  34  f u n c t i o n v e r y s i m i l a r t o ours has been r e p o r t e d r e c e n t l y . The  compound 90 e a s i l y  these c a t a l y t i c h y d r o g e n a t i o n oxide i n methanol and v i s c o u s o i l y substance The NMR  spectrum  available studies.  from 89, was  which was  p u r i f i e d by t h i n  o f t h i s product was  21a)  chosen as a model f o r  Hydrogenation  concentrated h y d r o c h l o r i c acid  (Figure  o f 90 w i t h  platinum  (1%), afforded a  l a y e r chromatography.  v e r y i n s t r u c t i v e and  allowed us  to  - 141 -  O  Figure  21a.  assign structure 91 to this compound.  The aromatic proton multiplet was  centered at 2,85 x , but significantly the two-proton singlet due to the methylene group of the acetate spectrum of 90) to 6.28 x .  ; was shifted from 5.62 x  moiety  (NMR  This chemical shift is in agreement with  corresponding situation present in the NMR  spectrum of the indole-2-methyl  acetate (88) .• The three-proton singlet of the ester methyl protons appeared as expected at 6.31 x  but the three-proton singlet at 7.4 x  which was present in the NMR spectrum of 90 had now disappeared.  However,  a new signal characteristic of ah ethyl group was now present in the spectrum of the reduction product (a two-proton quartet centered at 7.38 x and a three-proton triplet centered at 8.8 x ) . this substance showed a sharp band at 2.9y absorption at 5.8y.  The infrared spectrum of  (>NH) and only one carbonyl  Furthermore, the ultraviolet spectrum showed the  normal indole absorption pattern (*  max  225, 276(sh), 285 and 293 my).  It was now clear that the reduction and hydrogenolysis had proceeded successfully and we turned our studies to the analogous sequence with the coupling product (92).  Unfortunately when the coupling product was  subjected to. the above hydrogenation conditions, no reduction was observed Under prolonged hydrogenation with a higher concentration of hydrochloric  -  142  -  acid, the coupling product was reduced to a compound which showed only one three-proton singlet (-OCH ) in the NMR spectrum and one carbonyl absorption 3  at 5 . 8 u in the IR spectrum. This indicates that under more drastic conditions the keto function as well as one carbomethoxy group, presumably the one which is part of the indole moiety were reduced.  In spite of many attempts,  we were unable to find optimum hydrogenation conditions for the reduction of the keto function preferentially to the carbomethoxy group.  Sodium  borohydride was the next reducing reagent to be considered in our study. The ester 90 was extremely reactive towards this reagent and even at 0 ° C the reduction product indicated no carbonyl absorption in the infrared spectrum.  Similar results were also observed with the compound 92 where  the keto function and one carbomethoxy group were simultaneously reduced with this  reagent.  It therefore became necessary for us to study the reduction of the keto function with a mild reducing reagent.  When compound 90 was treated  with diborane in dry tetrahydrofuran at 0 ° C , a mixture of two compounds was obtained, and these were separated by thin layer chromatography on alumina. The major and less polar component  of the mixture was found to be identical  with 91 by thin layer chromatography andIR spectra comparison. The minor and more polar compound showed no carbonyl absorption in the infrared spectrum which indicated that both carbonyl functions had been reduced in this particular product. It was now apparent that diboranei could effectively reduce the keto function preferentially to the carbomethoxy group.  Therefore we subse-  quently treated the coupling product (92) with diborane under the above reaction conditions and an amorphous solid product was obtained. The latter substance which now showed no carbonyl absorption at 6 „ 2 u in the  - 143 infrared spectrum did exhibit a new unexpected strong absorption band at 4.3 which was attributed to a boron-carbon stretching frequency.  This absorption  band indicated that this product was s t i l l in the form of a boron complex. In fact when this amorphous solid was treated with concentrated sulfuric acid in dry dioxane a mixture containing mainly two components was obtained. The infrared spectrum of the latter material now showed no absorption band at 4.3u.  The major and less polar component of the mixture was separated  as a gummy substance by thin layer chromatography on alumina.  The spectral  properties of this compound were in excellent agreement with the assigned structure 93.  The NMR  centered at 2.8 x  spectrum showed the normal aromatic proton multiplet  while two three-proton singlets at 6.26 and 6.3 x indi-  cated the presence of two carbomethoxy groups in this compound. There was only one two-proton singlet present in the spectrum at 6.2 x and this was obviously due to the methylene protons of the methyl acetate moiety.  This chemical shift was in agreement with that of the correspond-  ing protons in the NMR  spectra of indol-2-acetic acid methyl ester (80)  and 3-ethylindole-2-acetic acid methyl ester (91). The reduction product (93) was rather sensitive to air oxidation and was difficult to characterize in a satisfactory manner. Additional data was obtained from its hydrochloride salt which was also isolated as an amorphous solid.  The infrared spectrum of this derivative showed a carbonyl  absorption band at 5.79y and the ultraviolet spectrum with i t s typical indole absorption (X r  max  222,  275(sh),283 and 292 my) was in agreement with  the proposed structure (93). We now hope that the above outlined sequence which leads successfully to the model intermediate 93 would be applicable in the synthesis of the desired diesters 95 and 66b.  Dieckman cyclization of these compounds  - 144 could provide the necessary nine-membered ring intermediates of the Iboga (96) and Aspidosperma (67) type alkaloids (See Figure 22).  Figure  Unfortunately,  22.  the necessary piperidines 65 and 94, whose syntheses were being investigated 35 independently by other coworkers  in our laboratory, were not available to  us at this time and therefore we were forced to postpone our investigations in this direction. In the mean time,we decided to consider our approach from an entirely different synthetic aspect and the results from this latter investigation are outlined in detail in the next section of this thesis.  - 145 Section B An alternative approach to the synthesis of a medium-sized ring system utilizes a reaction in which an appropriate ring bond is cleaved to provide the desired nine-membered ring intermediate.  Before turning to the specific  sequence which was proposed for our purposes, i t is necessary to review briefly the literature which is relevant and which supports the feasibilty of the crucial "bond-cleavage" reaction which is illustrated below in the postulated synthetic pathway. The reductive cleavage of C-N bond in a quaternary ammonium salt dates back to the work of Emde. This reaction now well-known as the Emde degradation has been used quite extensively in the alkaloid field and examples from the tetrahydroquinoline and isoquinoline series are cited . 36, 37 below. K  Me.  Figure  23.  This type of reductive fission was subsequently accomplished with 38 39 sodium in liquid ammonia ' [Figure 24).  An extension of the latter  conditions to the indole alkaloids is shown in some degradation studies 40 of the Rauwalfia alkaloids (Figure 24).  - 146 -  Figure  24. 41  More  recently Wenkert et al  membered ring system  have reported the synthesis of a nine  (98) by means of lithium in liquid ammonia reduction  of the intermediate 97 (Figure 25).  Figure  25.  Additional evidence which supports the generality of such a reaction comes from some very recent work in our laboratory.  In these investigations,  Birch reduction conditions or the use of lithium aluminium hydride has met with success.  Figure  One example of this is given here (Figure 26).  42  26.  A closely related sequence in the indole series has been also reported recently. ^ 4  (Figure 27).  - 147 -  Li Al H<|  Figure  01^  27.  On the basis of the above results i t became immediately apparent that the formation of quebrachamine or i t s derivatives would result from a reductive cleavage of the intermediate  73.  a  R = H  b  R = C0 Et  Figure  2  73a and b.  R = H  28.  The following discussion in this section of the thesis presents our investigations which are concerned with the synthesis of these compounds. The proposed sequence for this synthesis is outlined in Figure 29.  Con-  densation of tryptamin with an appropriately substituted glutarate  (70b)  - 148 and succinate (70a) via a Pictet-Spengler or Bischler-Napieralski cyclization could provide the necessary indolo-indolizine derivatives 71a and 71b. The tosyl or chloro derivatives of the latter substance could then undergo intramolecular cyclization to afford the desired intermediates 73a and b.  Figure  29.  Before proceeding to the reactions outlined above i t is necessary to describe the syntheses of the ethyl a-ethyl-a-(y-benzyloxypropyl)-a-ketoglutarate (70b) and ethyl a-ethyl-a-(y-benzyloxypropyl)-succinate (70a) which are summarized in Figure 30. Trimethylene glycol (99) was treated with one mole equivalent of sodium metal and the resulting sodium salt was reacted with benzyl chloride to provide y-benzyloxypropanol  44  (100).  The latter compound (100) was  - 149 -  Na  H0(CH ) OH 2  <()CH 0(CH ) OH 2  3  2  3  <j>CH Cl  99  100  2  SOC1COOEt EtONa  (j>CH 0(CH ) CH 2  2  2  EtONa EtCHCC00Et)  EtI >lEtONa 2  2  3  COOEt I C CH CH 2  2  3  101  COOEt  102  <|>CH0 ( C H )  <j)CH 0(CH ) C 1  3  2  COOH I  KOH  <t>CH 0(CH ;) 3 C C H C H  3  2  2  2  3  COOH  COOEt  104  103  PCOOH -CH CH  COOR $CH 0(CH ) 2  2  •I  3  C CH CH 2  2  a)  ^~  CH N 2  2  <t>CH 0(CH ) 3 2  2  2  I H  b ) EtOH  3  105  106 a R = Me b R = Et COOCH 106 a  3  l  4> CNa 3  BrCH COOCH 2  <()CH 0CCH2) 3 C C H C H 2  2  3  3  CH  107  2  COOCH3  COOEt 1  4> CNa 3  106 b  ->  <f>CH 0(CH ) 3 C C H C H 2  2  BrCH C00Et  2  CH  2  1  3  70 a  2  COOEt  COOEt 106 b  > BrCH C0C00Et 2  (t)CH 0(CH ) 3 C C H C H ^ 2  2  2  COCOOEt  Figure  30.  3  70b  - 150 subsequently reacted with thionyl chloride i n the presence of N,N-dimethyl aniline to provide the benzyl-chloropropyl ether^ (101) in 83% y i e l d . Alkylation of diethyl malonate with the latter provided the ethyl 46 Y-benzyloxy malonate  (102) i n 77% y i e l d .  The l a t t e r compound was then  alkylated with, ethyl iodide to provide ethyl y-benzyloxypropylethyl malonate (103).  The l a t t e r compound was also obtained by alkylation of the diethyl •i  ethyl malonate with the ether 101. It may be appropriate at this point to discuss some of the spectral properties of the alkylation products.  The infrared spectrum of compound  103 showed a carbonyl absorption band at 5.81u.  The most characteristic  feature of the NMR spectrum of 103 (See Figure 32) was the new threeproton t r i p l e t at 9.2 x due to the methyl protons of the additional ethyl 0  group in this compound.  0  The one-proton t r i p l e t at 6.6 x (-C-Cfl-c'-) in the  NMR spectrum of 102 (Figure 31) was no longer present i n the spectrum of 103. The presence of the benzyloxy propyl function in the above mentioned compounds as well as i n the compounds in the following discussion was clearly depicted in the NMR spectra by a five-proton aromatic singlet at 2.65 x , a two proton singlet at 5.48 x due to the benzyl methylene and a two-proton t r i p l e t at 6.47 x due to the propyl methylene protons attached to the ether oxygen atom. Hydrolysis of 103 with potassium hydroxide provided Y-benzyloxypropylethyl malohic acid (104) as a nice crystalline material (m.p. 117120°C).  The infrared of this acid showed a doublet in the carbonyl region  at 5.74 (weak) and 5.89 (strong) y. The NMR spectrum (See Figure 33) was also in agreement with the structure of this compound.  Figure  32.  -  152  COOH <t>CH OCH CH CH C 2  2  2  CH CH  2  2  3  COOH  Figure  33'.  COOH l <j>CH 0CH CH CH C 2  2  2  2  H  Figure  34.  CH CH 2  2  - 153 The decarboxylation of 104 was accomplished smoothly at 140-150°C, to provide a-(Y-benzyloxypropyl)butyric acid (105) as a colorless viscous o i l which was characterized without further purification. The acid 105 was esterified with diazomethane or ethanol to provide the esters 106a and b as colorless viscous oils.  Both compounds showed a  carbonyl absorption band at 5.19u in the infrared spectrum.  The NMR spectra  (See Figure 35, 36) showed a new one-proton multiplet at 7.8 T due to the tertiary a-protbn  (CH-C00R) or the butyrate moiety which was generated from  the decarboxylation reaction. We had now achieved the syntheses of the desired methyl and ethyl c-(Y-benzyloxypropyl)butyrates  106a and b. which on subsequent alkylation  could provide the succinates 107 and 70a as shown in Figure 30. 47 It is well known  that sodium triphehyl methane is a strong base  and capable of forming the enolate of an aliphatic ester.  This enolate  can then be alkylated with an alkyl halide. The enolates of 106a and b were indeed formed when these esters were treated with sodium triphehyl methane in dry ether as solvent.  Subsequent  reaction of these intermediates with the corresponding methyl and ethyl bromo acetates, provide the desired methyl and ethyl succinates 107 and (70a).  In both cases the reaction product was chromatographed on neutral  alumina, and both succinates were eluted with benzene.  Vacuum distillation  provided these compounds as colorless oils. Both compounds 107 and 70a showed a strong carbonyl absorption band at 5.8u in the infrared spectra.  The most characteristic features of the  NMR spectra of these esters (See Figures 37, 38) were the absence of the one-proton multiplet at 7.8 T present in the spectra of 106a and b and the presence of a two-proton singlet at 7.4 T due to the methylene protons of  COOCH I <t>CH 0CH2CH2CH C C H C H 3  2  Figure 35.  Figure 36.  2  2  3  - 155 the newly introduced acetate unit.  The spectrum of 107 shows two three-  proton singlets at 6.4 and 6.46"T due to the two non-equivalent carbomethoxy groups present in this compound.  Similarly the spectrum of 70a shows a  four-proton octet centered at 5.87 T and a six-proton sextet centered at 8.79 x due to. the two non-equivalent carboethoxy groups present in this compound. The ethyl keto glutarate (70b) was similarly prepared from the ester (106b) using ethyl bromo pyruvate as the alkylating agent.  This compound  was obtained from the reaction mixture as a colorless o i l , after chromatography from neutral alumina followed by vacuum d i s t i l l a t i o n . The infrared spectrum of this compound showed only one strong carbonyl absorption band 5.8u.  The NMR spectrum (See Figure 39) showed a four-proton  quartet at 5.88 T. .and a six-proton triplet at 6.6 T due to the two carboethoxy groups present in this compound.  The new methylene proton of the  pyruvate unit give rise to a symmetrical quartet centered at 7.09 x which indicated the non-equivalence of these particular protons. The striking difference in the behavior of the methylene protons of the acetate and pyruvate units of the compounds 70a and 70b in the NMR spectra must be discussed briefly at this point. The protons of a methylene group adjacent to a center of molecular asymmetry can become magnetically non-equivalent and display AB-type splitting pattern: in the NMR spectra. The existence of prefered conformations of the methylene group with respect to the asymmetric center has generally been considered necessary for magnetic non-equivalence.  Another  important factor contributing to magnetic non-equivalence is the intrinsic 48 asymmetry of the molecule.  - 156 COOCH  Figure  3  37.  COOEt i  c^C^OCl^Cl^CH^C C H C H 2  CH I  3  2  COOEt  i  Figure  38.  _  I  I  !  L  COOEt  J  2  Figure  i  i  i  i  i  3  4  S  6  7  39.  .  i  i  l  8  9  10  - 158 The three staggered conformations of 70a which are pertinent to this discussion are I, II and III.  If we assume that protons Ha and Hb are not  equivalent in each of the above conformations, one would expect that they will give rise to an AB spectrum.  However, in order to be consistent with  the NMR spectrum of 70a in which these protons give rise to a singlet we can consider the two alternative posibilities which follow: a) the methylene protons are "accidentally" equivalent in each of the three conformations and therefore they will give rise to an A spectrum or b) we may also 2  assume that the alkyl groups, (J>CH OCH CH CH - (R) and CH CH -(Et) have a 2  2  2  2  3  2  similar effect on the methylene protons or in other words, the compound 70a which possesses an asymmetric center behaves similarly to the system R'-CH -CR R" (where R" is a very different substituent from R). The 2  2  methylene protons  Ha arid Hb will then be equivalent in conformation I.  Similarly when the interconversion between II and III is rapid and these two conformations are equally populated, the methylene protons will experience the same average shielding and therefore, will become equivalent. On this basis the shielding of the methylene group will then be the arithmetic mean of the three environments and the spectrum will be of the A  2  type.  - 159  -  If we now examine the conformations of compound 70b whose NMR  spectrum  indicates that the methylene protons are non-equivalent then the above  conditions cannot prevail.  Clearly the introduction of the pyruvate unit  which bears an a-keto carbomethoxy system must play some role in creating the non-equivalence of the methylene protons.  Whether this situation is  derived from the intrinsic asymmetry of the molecule or a decrease in the rate of interconversion of the various conformations is not known at the present time. case.  Let us briefly consider these two possibilities, in this  If rapid, interconversion does indeed occur and an equal population  of the individual conformations is obtained then obviously, the nonequivalence of the protons Ha and Hb must arise from the intrinsic asymmetry of the molecule 70b.  On the other hand, the a-keto carbomethoxy :  system can also introduce additional steric and electronic repulsions in the molecule and thereby prevent a very rapid interconversion from one conformation to another so that unequal populations of these conformations are obtained.  Furthermore, the methylene protons will become non-equivalent  - 160 providing that the environments of the Ha and Hb are not identical in conformations  we consider for this molecule.  I f we accept as in the case  of 70a, that the substituents R and Et have similar effects on protons Ha and Hb then only conformation V and VI w i l l give rise to the non-equivalence of these protons.  Since the latter alternative implies a) an equal  effect of R and Et and b) restricted rotation and thereby prefered conformations for this molecule I feel that intrinsic asymmetry plays the major role in providing the non-equivalence of the methylene protons in compound 70b. The difference of the chemical shifts (6Ha-6Hb) of Ha and Hb in the NMR spectrum of 70b was calculated 13 cycles per second  (c.p.s.)  We were now ready to study the condensation of the ethyl keto glutarate 70b with tryptamine via a Pictet-Spengler Wenkert et a l  4 1  50  type of cyclization.  were able to obtain the compound 108 by refluxing trypt-  amine hydrochloride with ethyl a-keto glutarate in ethanol for 60 hours.  Figure  40.  - 161 When the ethyl keto glutarate 70b was reacted with tryptamine hydrochloride under Wenkert's conditions, a mixture of neutral compounds was obtained. The major component of this mixture was separated by preparative thin layer chromatography.  The ultraviolet spectrum of this component showed an  absorption pattern characteristic of an indole system and the infrared spectrum indicated the presence of an amide carbonyl.  However, the com^  plexity of the NMR spectrum indicated that this compound was s t i l l a mixture of at least two components.  In fact, we were able to separate  two compounds from this mixture by a subsequent and more careful thin layer chromatographic purification.  For the sake of convenience, the  less polar compound obtained was labelled as compound A whereas the more polar one was labelled as B. Compound A remained as gummy material in spite of numerous attempts to obtain i t in crystalline form.  On the other  hand, compound B was crystallized from ether-hexane (m.p. 90-96°). Both compounds showed an identical ultaviolet absorption pattern characteristic  of an indole system, (X 223, 275(sh) 284 and 292 mu). . \ max ^ '  The infrared spectra of both compounds showed a sharp band at 2.79u (>NH) and two carbonyl absorptions at 5.82 and 5.9u corresponding to the presence of ester and amide functions respectively.  The NMR spectra of both compounds  were almost,identical (See Figures 41 and 42). integrated for 9 protons..  The aromatic proton region  The one-proton doublet at 3.05 x normally due to  the a-position of the indole moiety indicated that no cyclization had occurred.  The two-proton singlet at 5.53 x was due to the presence of the  benzyloxy methylene protons while the two-proton quartet centered at 5.8 x and the three-proton triplet centered at '8.78 x in the spectrum of compound A (the corresponding values for compound B appeared at 5.9 and 8.87 x respectively) indicated the presence of one carboethoxy group which was in  - 162 -  41.  Figure  i  \  i  1  3  4 Figure  :  42.  i  i  1  1  5  &  i  « • ?  1  L 10  -  agreement with the infrared spectrum.  163  -  The methylene protons of the pyruvate  unit gave rise again to a quartet centered at 7.0 T in compound A and at 7.IT in compound B. From the above spectral properties of the compounds A and B along with the analytical results obtained for the crystalline compound B, we assigned the structures 110 and 111 to either A or B. Furthermore, in the previous discussion concerning the non equivalence of the methylene protons -CH C0C02Et in the pyruvate 70b, we came to the conclusion that the a-keto — 00 2  U "  carbomethoxy group (C-C-0 Et) was mainly responsible for the non-equivalence of these protons.  One therefore expects that the difference in the chemical  shifts of Ha arid Hb (6Ha-6Hb) in 110 will be very similar to the one in the 0 0 glutarate 70b, since the methylene substituent -C-c'-O Et is common in both compounds. On the other hand, the value of <5Ha-6Hb is expected to be different in 70b and 111 due to the different nature of the methylene 0 0 II II  substituent in 111 (C-C-NH-). The value  <5Ha-6Hb was calculated 13 and 14.5  c.p.s. in compoundsA and B respectively (<SHa-<5Hb = 13 c.p.s. in compound 70b). However, we tentatively assigned the structures 110 and 111 to A and B respectively, although further work will be necessary in order to establish unambiguously the structure of these compounds. The mixture of compounds A and B was also obtained when equivalent amounts of tryptamine and the keto glutarate 70b were heated in ethanol in a sealed tube at 120° or when refluxed in benzene for several hours. The above results indicated that the keto glutarate 70b failed to follow the normal Pictet-Spengler type of cyclization with tryptamine. 50  Although the reason for the failure of this reaction is unknown, i t is felt that steric hindrance created by the two substitutents adjacent to the pyruvate carbonyl group may be responsible.  Instead, reaction at  - 164 either end of the keto glutarate molecule with tryptamine provided the amides 110 and 111. We now turned our attention to the condensation of the succinate 70a with tryptamine.  We felt that the i n i t i a l formation of an imide (112)  could subsequently lead to the indole-indolizine intermediate (113) via 51 Bischler-Napieralski cyslization reaction.  In the following reactions  the ethyl succinate 70a was used, for i t was more easily available in our laboratory. Indeed we were able to obtain the imide 112 by refluxing a mixture of tryptamine and 70a in diethylerie glycol monoethyl ether as solvent for 44 hours.  The imide (112) was isolated as a gummy material from the reaction  mixture by chromatography on alumina.  Iri spite of numerous attempts this  product failed to crystallize. The spectral properties were in excellent agreement with the structure 112 for this compound.  The molecular ion peak (m/e-418) in the mass  spectrum supported the molecular formula C^gF^gQ^. The NMR spectrum (See Figure 44) showed a broad one-proton singlet at 1.0 T (>NH), a nine-proton aromatic multiplet centered at 2.7 f , and a one-proton doublet at 3.06 T due to the a-proton of the indole system.  The absorption due to nine aromatic  protons and the two-proton singlet at 5.58 x was in agreement with the presence pf the benzyloxy group in this compound.  Two two-proton triplets  centered at 6.25 and 7.0 T were due to the ethylene bridge o f the tryptamine' moiety while the two-proton triplet centered at 6.68 T was readily assigned to the methylene protons on the carbon,atom attached to the ether oxygen atom (-CH 0). 2  Finally the two-proton singlet at 7.6 T could be easily  attributed to the methylene of the succinimide moiety.  -  Figure  43.  165  -  Figure  44.  - 167 The infrared spectrum of this compound showed a sharp peak at 2.86y (>NH)  and two carbonyl absorption bands at 5.67  (weak) and 5.91  (strong) u,.  The latter absorption pattern is a very characteristic feature of a succinimdie system?^  Furthermore, this compound gave typical indole  ultraviolet absorption spectrum ( A  m a x  221, 275(sh), 283 and 291  my).  41 Although Wenkert et al  were unable to cyclize N-[g-(3-indolyl)ethyl]-  succinimide (120), Morrison and Cetenko have recently succeeded in accomplishing the cyclization of the N-[6-(3-indolyl)ethyl)glutarimide (121) to provide the indoloquinolizine system 122 by the use of phosphorus pentoxide 5£U in refluxing xylene. (Figure 45).  121  120 Figure  122  45.  When we subjected the imide 112 to the above cyclization conditions we obtained among other products a compound in approximately 5% yield, whose 52 spectral properties were in excellent agreement with those reported  for  the compound 122. The ultraviolet spectrum of the former compound showed absorption maxima at 223(sh), 229, 314 and 327 my which is in good agreement with the absorption pattern reported for 122 (A  220(sh) 232, 308 and 319 my).  In  the presence of a strong acid e.g. concentrated hydrochloric acid a new .IN HCl strong absorption peak appeared at 388 my (compound 122 exhibited * my) .  m a x  395  - 168 The infrared (in KBr) of the cyclization product showed two carbonyl •' CHC13 absorptions at 5.95 and 6.05u (compound 122*)^  6.0 and 6.1u)  and a sharp  band at 3.05u (>NH). The NMR spectrum, which was obtained from a small amount of this product indicated that the benzyl group was no longer present since the aromatic proton multiplet now integrated for approximately four protons and the two-proton singlet due to the benzylic methylene was no longer present in this spectrum.  On the other hand, approximately four .  protons were present in the olefinic region of the NMR spectrum. The mass spectrum of the above compound showed the molecular ion peak at m/e 192 and this corresponded to the molecular formula C H2oON . On 19  2  the basis of the above data we assigned the structure 114 to this compound. It was now obvious that we had succeeded in accomplishing the cyclization of the imide although the loss of the benzyloxy group and the small yield of the desired product made this particular reaction impractical for our synthesis. In order to avoid the undesirable side reaction which could also be responsible for the low yield of the product we decided to replace the benzyloxy group with a substituent which could be stable under the cyclization conditions. A halogen atom appeared to be the most obvious substituent on the basis of our synthetic approach. Consequently we modified our synthetic sequence to allow for this substitution.  The benzyl ether function present in 112 was cleaved smoothly 53  with boron tribromide or trichloride, alcohol (115).  , to provide the corresponding  The cleavage of an ether with boron trihalides involves  the formation of an intermediate oxonium ion complex from which, a carbonium ion intermediate is formed by rupture of C-0 bond.  In the benzyl ether  112 the benzyl carbonium ion will be resonance stabl.ized and therefore the  - 169 alcohol 115 was exclusively obtained. The infrared spectrum of 115 showed a sharp band at 2.87u (>NH) which overlapped with a broad band at.2.95 (-0H) 5.66  and also two carbonyl peaks at  (weak) and 6.1u characteristic of the presence of the succinimide  chromophore. The NMR spectrum (Figure 46) showed a multiplet of four aromatic protons centered at 2.7 x while the two-proton singlet of the benzyloxy methylene normally appearing at 5.48 x was no longer present.  Another,  indication that the ether cleavage to the expected alcohol 115 had occurred was provided by the shift of the methylene protons attached to the oxygen function (-CH 0) from 6.47 in the ether to 6.58 x in the alcohol.  Finally  2  a one-proton singlet at 6.75 x was.assigned to the hydroxyl proton on the basis of the disappearance of this NMR peak on addition of D2O to the sample.  :  The mass spectrum of this compound indicated a molecular ion peak at m/e 328 which was in agreement with the molecular formula Ci9H ^03N2. 2  As expected the ultraviolet spectrum s t i l l showed a typical indole absorption spectrum (*  m  222, 275(sh), 283 and 292 mu) and thereby eliminated  any change in this portion of the molecule. Reaction of the alcohol 115 with thionyl chloride and pyridine in an ether-benzene solvent system, provided a mixture of products from which the chloro compound 116 was obtained as a gummy material by chromatography from neutral alumina.  This compound remained as a gum in spite of the  numerous attempts to obtain i t in crystalline form. The mass spectrum showed a molecular ion peak (m/e 346) which was in excellent agreement with the empirical formula CigH 30 N C1. 2  2  2  - 170 -  Figure  Figure  46.  47.  - 171 The infrared spectrum of this substance showed a sharp band at 2. 87p (>NH) and the characteristic carbonyl bands [5.67 (weak) and 6.1 (strong)] chromophore for the succinimide. The "NMR spectrum (Figure 47) indicated very clearly that the replacement of the hydroxyl group by the chlorine function had occurred.  In p a r t i s  cular the absence of,the hydroxyl proton peak which was present at 6.75 T in the spectrum of the alcohol and Chemical shift of the methylene protons attached to the hydroxyl group from 6.58 to 6.68 x in the corresponding chloro.compound 116 should be noted.  In addition the ultraviolet spectrum  indicated again the typical indole absorption pattern (A  223, 275(sh),  283 and 292 mp). When the chlorocompound 116 was subjected to the same cyclization conditions as :in the case of compound 112, we obtained among other products a substance (in 18% yield), which indicated the desired ultraviolet and infrared spectral properties for the expected enol lacta.m 117.  In our  attempt t o find more,optimum cyclization conditions i t was realized that the.yield of this porduct was slightly improved when the reflux period was reducted from five to one.and a half hours.  The product 117 was  obtained from the reaction mixture as an amorphous,solid after chromatography on neutral alumina.  This compound could not be obtained crystal-  lined and i t was characterized as an amorphous powder. The infrared spectrum of this compound showed a sharp band at 3.04p (>NH) and a carbonyl at 5.93 and 6.04y. The ultraviolet spectrum showed absorption maxima at 223(sh), 229, 283(sh), 303(sh), 313 and 326 mp.  When a drop of concentrated hydrochloric  acid was added to the above solution a new peak appeared at 387 mp while the intensity of the peaks at 313 and 326 mp was reduced considerably.  Figure  48.  - 173 These spectral properties are again in good agreement with the ones reported for the compound 122. The NMR spectrum was also very instructive and in excellent agree;  ment with the structure 117 assigned to this compound (Figure 48). This spectrum showed a broad one-proton-singlet at 1.6 T (>NH), a four-proton aromatic multiplet centered at 2.7 x and a sharp one-proton singlet at 6.64 x due to the olefinic proton at  Furthermore two two-proton  triplets centered at 6.15 and 6.96 x due to the C5 and Cg methylene protons a two-proton triplet at 6.59 x due to the -CH C1 group, a six-proton multi2  plet centered at 8.2  x  (-CH CH CH C1 and 2  2  ?  -CH7CH3)  and a three-proton t r i -  plet at 9.3 x (-CH CH3) were clearly evident in the spectrum. 2  Finally a  one-proton doublet at 3.0 x (3-proton of the indole) and the two proton singlet at 7.6 x (methylene protons of the succinimide moiety) present in the spectra of 112, 115 and 116 were no longer present in the spectrum of the cyclization product. The final piece of evidence in support of the cyclization product was obtained from the mass spectrum of this compound which showed a molecular ion peak at m/e 328 corresponding to the empirical formula C H 0N C1. 19  21  2  Hydrogenation of 117 with Ptp  2  in acetic acid provided a substance  which on purification by chromatography from neutral alumina provided a crystalline material (m.p. 148-156°).  This material showed two spots on  thin layer chromatopiates (alumina) when the latter was developed several times with benzene as eluant. The mass spectrum of this mixture indicated a molecular ion peak (m/e 329) corresponding to the empirical formula CigH N Cl. 23  2  We were able  to separate small amounts of the two compounds by careful preparative thin layer chromatography.  Both of these substances showed molecular ion peaks  -  at m/e  174  -  329 in their mass spectra and i t was clear that they were isomeric. The infrared spectrum of the mixture indicated an amide absorption'  at 5.99y and the ultraviolet spectrum exhibited tion pattern (A  224, 275(sh), 283 and 290  a  typical indole absorp-  my).  On the basis of the above spectral properties i t was obvious that the hydrogenation  of 117 had provided the two expected stereo isomers at  the newly created asymmetric center  llh .  The reduction of the lactam carbonyl was accomplished by refluxing the above mixture.118 with lithium aluminum hydride in tetrahydrofuran for 24 hours.  The reduction product was again a mixture of two components.  We were able to separate these compounds in small amounts by careful preparative thin layer chromatography.  Both components of this mixture indi-  cated a molecular ,ion peak at m/e 282 in their mass spectra - a value which, corresponded to the molecular formula C19H26N2.  Other features of the  spectrum were also in support of the fact that not only was the lactam carbonyl successfully reduced but unfortunately, the halogen had been lost. Although further work will be required, there appears l i t t l e doubt that the structure' of 119 is the correct one for these reduction products.  It  would appear that the two compounds are merely isomeric compounds. It is therefore necessary at this point to develop milder conditions for the reduction of the lactam so that the chlorine group will remain unaffected.  A possible alternative which is being considered at present is  to obtain the thiolactam (120) by treating the lactam 118 with phosphorous pentasulfide and subsequent reduction of the latter with Raney riickel (Figure 49).  The procedure has been recently employed in a synthesis of 54  the alkaloid vincamine.  - 175 -  Figure  49.  Another approach to the reduction of this carbonyl function could be the reduction of the enol lactam 117 with lithium aluminum hydride since this compound is expected to reduce under milder conditions than the lactam 118.  It is hoped that the conditions would be sufficiently mild to  prevent hydrogenolysis of the C-Cl bond.  Subsequent reduction of the  iminium salt 122 with sodium borohydride could lead to the desired cyclic product 121.  The subsequent cyclization of the latter could provide the  necessary intermediate, 73a.  The above alternatives are now under  investigation in our laboratory.  - 176 In conclusion, I would like to say that the synthetic sequence leading to the intermediate  73a is nearly complete.  As was already mentioned  above on several occasions, the small amounts of materials which were available or the gummy nature of some of.the reaction products prevented complete characterization of some of the compounds.  In particular,  analytical data are lacking in some instances and will be obtained when a subsequent repetition of the above, investigations on.a larger scale will be carried out.  - 177 Experimental Melting points were determined on a Kofler block and are uncorrected. The ultraviolet (UV) spectra were recorded in 95% ethanol on Cary 11 recording spectrometer, and the infrared (IR) spectra were taken on Perkin-Elmer Model 21 spectrometer.  Nuclear magnetic resonance (NMR) spectra were  recorded at 60 magacycles/sec on a Varian A60 instrument; the line positions or centers of multiplets are given in the Tiers T scale with reference to tetramethylsilane as the internal standard; the multiplicity, and integrated area and type of protons are indicated in parentheses.  S i l i c a gel  G and Woelm alumina were used for thin layer chromatopiates; the type of absorbent and the solvent system for developing the thin layer chromatoplates are given in parentheses.  The alumina used for column- chromato-  graphy.was Shawinigan reagent grade deactivated with 3% of 10% aqueous acetic acid.  Analyses were performed by Dr. A. Bernhardt and his associ-  ates, Mulheinv (Ruhr), Germany and by the Microanalytical Laboratory, University of British Columbia.  Every molecular weight (MW) quoted was  determined mass spectrometrically with,an Atlas CH-4 mass spectrometer. 2-Carbomethoxy-3-chloroacetyl  indole (75)  Dry 2-carbomethoxy-indole  (74) (5.644 g, 0.0032 mole) was placed in  a thick-walled glass tube and treated with dry chloroform (200 ml), and dry chloroacetonitrile (50 ml, 0.756 mole). Dry hydrogen chloride gas was bubbled into the mixture, cooled to 0°C, over a period of 2 hours.  After this time the reaction was cooled  to -78° (dry ice-acetone) and hydrogen chloride passed in for another 10 minutes at which point a yellow solid had formed in the reaction mixture.  The tube was the sealed and the reaction mixture was kept at  50° for 11 hours.  After this period the tube was slowly cooled to -78°,  - 178 then opened carefully and allowed to come to room temperature.  The pre-  cipitate was filtered, washed with ice-cold chloroform (2 x 25 ml) and the combined filtrate and washings were kept aside as they contained mainly starting material.  The deep yellow amorphous solid was dried under vacuum;  and then treated with water (50 ml) to affect hydrolysis of the iminium salt.  The salt dissolved instantly, but on hydrolysis to the correspond-  ing ketone an insoluble solid compound formed.  The precipitate was filtere  and subsequently extracted several times with boiling chloroform.  The  combined chloroform, extracts were washed once with water and dried over anhydrous magnesium sulfate.  Removal of the solvent afforded a light  yellow semicyrstalline product (5.278 g).  Ah additional 0.396 g of this  product was obtained from treatment of the i n i t i a l aqueous f i l t r a t e with :  ammonia, gas and removal of the yellow solid which was precipitated during this treatment.  The crude product (5.674 g) was chromatographed from  alumina. ' Initial elution with benzene-chloroform  (1:1) gave starting material  (1.023 g), but further elution with the same solvent gave the desired compound, (75) (1.76 g, 31% yield), m.p. X  249 (4.11), 318 (4.10) my.  147.5-148.5°. . Ultraviolet:  Infrared (CHC1 ): 2.98 (>NH), 5.80 3  Q  1H3.X  (-C00CH)^ 6.0 3  (-CCH2-),  6.58  (aromatic) u .  (multiplet, 4H, aromatic), 4.98 0 CH OC-)„  NMR signals ((CD ) CO): 3  2  2.65-  (singlet, 2H, -CH C1), 5.98 (singlet, 3H, 2  Found: C, 57.10; H, 4.11; 0, 18.97; N, 5.73; Cl, 14.07. Calc.  3  for C H O NCl: C, 57.24; H, 4.01; 0, 19.07; N, 5.56; Cl, 14.11. 12  10  3  2-carbomethoxy-3-(3-carbomethoxy-N-piperidyl)-acetylindole Well powdered 2-carbomethoxy-3-chloroacetyl  (77)  indole (75) (300 mg,  0.0012 mole) was thoroughly mixed with 3-carbomethoxy piperidine (76)  (500  mg, 0 . 0 0 3 5 mole).  179  -  The mixture was warmed slightly from time to time  in a steam bath, until a clear viscous solution,was obtained.  To this  viscous solution, ethyl ether (30 ml). and water (30 ml) were added and the mixture was shaken until the viscous o i l was distributed between the aqueous and the ether layer.  The ether layer was separated, washed once  with water and dried over anhydrous magnesium sulfate.  The ether was  removed by d i s t i l l a t i o n under vacuum to provide a viscous o i l (519 mg). This material was chromatographed from alumina (15 g).  The desired  material (77) (270 mg) was eluted with benzene-ethyl ether ( 9 - 1 ) as an amorphous solid.  Crystallization of this material from  dichloromethane-  ethyl ether provided and analytical sample m.p. 1 2 1 - 1 2 5 ° . Yield 6 0 % . 0 0 Infrared (KBr): 3.1 (>NH), 5.78 (-CQCH ), 6.1 (CCH-N<) y. Ultraviolet 3  A  max  220  ( 4 . 4 1 2 ) , 248  (4.155),314  /  2  ( 4 . 0 7 4 ) my.  NMR  0  signals  (CDC1 ): 3  2.6  11  (multiplet, 4H, aromatic), 6 . 0 8 (singlet, 3H, - C 0 C H ) , 6 . 4 (singlet, 3H, p 0 — 3  -d0CH ), 3  0,  22.98;  N,  7.82.  6.2  (broad doublet, 2 H , -CCH2-N<)  N, 6 . 9 6 .  Calc. for C H 0 N 2 : 1 9  2 2  5  .  Found:  C, 6 3 . 4 9 ; H,  C, 6 3 . 6 8 ; H, 6 . 1 9 ; 0 ,  6.54,  22.32;  2-carbomethoxy-3-: [ct-hydroxy-g- (3-carbomethoxy-N-piperidyl)ethyl]-indole (78) The compound 77 (250 mg, 0 . 0 0 0 7 mole) was dissolved in absolute methanol (50 ml) and sodium borohydride (400 mg, 0.01 mole) was added to this solution.  The mixture was stirred for 4 hours at room temperature.  The methanol was removed by distillation under vacuum and the residue was shaken with a mixture of water (30 ml) and ethyl ether (30 ml) until the residue was distributed between the aqueous and the ether layer.  The ether  layer was washed once with water and dried over anhydrous magnesium sulfate. The ether was removed by distillation under vacuum to provide an amorphous white solid (200 mg), which showed only one spot on thin layer chromato-  - 180 graphy (alumina, benzene-chloroform 1:1). This material, in spite of the numerous attempts, resisted crystallization.  To prepare an analytical  sample, this material was further purified by preparative thin layer chromatography (alumina, benzene-chloroform 1;1). The compound was eluted from the alumina with dry methanol.  Yield 80%.  Infrared (KBr): 2.95  (-0H,>NH), 5.85- (2 x -C0CH ) y. Ultraviolet: 'X 3  my.  &  229 (4.39), 298 (4.272)  NMR signals (CDC1 ): 2.65 (multiplet, 4H, aromatic), 6.25 (singlet, 0 0 OH 3  3H, -C0tH ) 6.45 (singlet, 3H, -C0CH ), 4.4 (broad quartet, IH, -CH-CH -). 3  3  2  Found: C, 63.26; H, 6.86; 0, 22.99, N, 6.90. Calc. for C H 19  21+  0N: 5  2  C, 63.32; H, 6.71; 0, 22.2; N, 7.77. Raney nickel hydrogenation of 77 The compound 77 (100 mg) was dissolved in 95% ethanol (30 ml) and hydrogenated over Raney nicker catalyst at room temperature and atmospheric pressure, for 1 1/2 hours.  The catalyst was filtered and the ethanol was  removed with d i s t i l l a t i o n under vacuum.  Preparative thin layer, chromato-  graphy (alumina, benzene-chloroform 1:1) of the residue provided an amorphous white solid (80,mg)i  This material was found to be identical to  78 by thin layer chromatography, NMR and IR spectra comparison.  Yield 84%.  2-carbomethoxy-3- [ot-hydroxy-g- (3-carbomethoxy-N-piperidyl)-ethyl]-4,5,6,7tetrahydro-indole (79) a) Raney nickel hydrogenation of 77 The compound 77 (100 mg) was dissolved in 95% ethanol (30 ml) and hydrogenated over Raney nickel catalyst at room temperature and atmospheric pressure, for a period of 20 hours.  The catalyst was filtered and the  ethanol was removed by distillation under vacuum.  Preparative thin layer  chromatography (alumina benzene-chloroform 1:4) of the residue provided an  - 181 amorphous, white solid (65 mg) which, in spite of numerous attempts resisted crystallization. 0 u (C0CH  Yield 64%.  Infrared (CHCl ): 2.9 (>NH), 3(-0H), 5.6 3  0 n (COCH3  of the piperidine moiety), 6 of the indole moiety) y. Ultraviolet: X 290 (4.193) my. NMR signals (CDC1 ):4.85 (broad multiOH 0 3  m  Diet, C0CH ) 3  a  3  x  IH,fcH-CH -N<),6.29 (singlet, 3H, COCH3), 6.48 (singlet, 3H, 2  x. Found: C, 5?.20; H, 6.77; 0, 22.69.  Calc. for C H 0 N : 19  28  5  2  C, 62.62; H, 7.74; 0, 21.95. . b)  Raney nickel hydrogenation of 78 The compound 78 (100 mg) was dissolved in 95% ethanol (30 ml) and  hydrogenated over Raney nickel catalyst at room temperature and atmospheric pressure for a. period of 20 hours.  The catalyst was filtered and the-  ethanol was removed by distillation uder vacuum. chromatography  Preparative thin layer  (alumina, benzene-chloroform 1:4) of the residue provided  the major component as a white amorphous solid (60 mg). This compound was found to be identical to 79 by thin layer, chromatography  (alumina, benzene-  chloroform 1:4), TR and NMR spectra comparison. 2-carbomethoxy-4,5,6,7-tetrahydro-indole. (80) 2-carbomethoxy-indole(74)(175 mg, 0.001 mole) was dissolved in 1% of concentrated hydrochloric acid in methanol solution (30 ml) and hydrogenated over Adams catalyst at room temperature and atmospheric pressure. When 1 mole of hydrogen was absorbed (in a period of approximately 6 hours), the hydrogenation was interrupted.  The catalyst was filtered and approxi-  mately 25 ml of methanol was removed by vacuum distillation. mixture was taken in ethyl ether (50 ml).  The remained  The ether solution was washed  once with water and dried over anhydrous magnesium sulfate.  The ether was  removed by distillation under vacuum to provide a white semicrystalline  - 182 solid.  Preparative thin layer chromatography (silica gel, chloroform) of  this solid provided starting material (95-mg) and the product 80 (50 mg). The latter was recrystallized from aqueous methanol m.p. 156-157°. 0  (KBr): 3.02 (>NH), 6.0 (C0CH )u.  Ultraviolet: X  3  Infrared  287 (4.253) my.  NMR  signals (CDCI3): 3.52 (doublet, IH, B-proton of the pyrrole), 6.27 (singlet 11  3H,  COCH3),  7.5  (multiplet, 4H), 8.25 (multiplet,  H, 7.68; 0 , 18,06; N, 7.86.  Calc. for C H O N. 10  13  2  4H)  x.  Found: C, 66;55;  C, 67.02; H, 7.31;  0 , 17.85; N, 7.82. Indole-2-carboxylic acid dimethylamide (83) To a stirred suspension of indole-2-carboxylic acid (31 g, 0.19 mole) in absolute benzene (500 ml), kept under a slow stream of nitrogen, was added thionyl chloride (50 ml), over a period of 1 hour.  The mixture was  then warmed to 45-50° and kept at this temperature for 1 1/2 hours, whereupon the acid went completely into solution.  In order to remove excess  thionyl chloride, 1/3 of the resulting solution was removed by distillation under reduced pressure.  After dilution of the remainder with benzene.  (200 ml) the solution was slowly added dropwise, to a stirred solution of dimethylamine in benzene.  The resulting mixture was treated with cold  water (100 ml) in order to dissolve the dimethylamine hydrochloride, thoroughly stirred, and filtered.  The solid was dried and crystallized from  ethanol, producing the dimethylamide (30 g, 83%) m.p. 182-184°.  Lit.:  m.p.  180-182°. 2-Dime thy1amin ome thy1in dole (84) To a stirred suspension of lithium aluminum hydride (28 g. 0.74 mole) in dry tetrahydrofuran (500 ml),  a solution of indoler2-carboxylic  acid dimethyl amide (56 g, 0.3 mole) in tetrahydrofuran (600 ml) was added  - 183 slowly over a period of 2 hours.  After completion of the addition, the  reaction mixture was stirred under an atmosphere of dry nitrogen at 45-50° for 5 hours, cooled to - 1 0 ° , and treated cautiously with water to destroy excess hydride.  The inorganic salts were separated by filtration and  washed thoroughly: with fresh volumes of tetrahydrofuran.  The combined  washings and filtrate were evaporated under reduced pressure.  The residual  oil was treated with ice-cold dilute (2N) aqueous sodium hydroxide solution C150 ml), and extracted thrice with ether.  The combined ether  extracts were washed twice with water and dried over anhydrous sodium sulfate.' Removal of the ether with vacuum distillation produced a viscous oil which was distilled under vacuum, affording a clear liquid (48.5 g, 93%) which after some time in the cold, crystallized.  A small amount was  recrystallized from pet. ether (b.p. 60-80°) affording colorless blocks. B.p. 104-107°/0.1 mm (Lit. b.p. 118-120° ,/0.3 mm) m.p. 59-61° (Lit. 6 0 - 6 1 ° ) . 2-dimethylaminomethylindole methiodide (85) To a stirred solution of 2-dimethylaminbmethylindole (17.4 g, 0.1 mole) in dry ethyl ether (100 ml) a solution of methyl iodide (15.6 g, 0.11 mole) in dry ethyl ether (50 ml) was added dropwise.  The quaternary salt  came out of solution as a white crystalline solid. After complete addition of the methyl iodide, the reaction mixture was stirred over-night at room temperature.  The ether was decanded from the solid and the residue was  dried under vacuum affording a white crystalline solid (31 g).  This  material turned to a gummy o i l when exposed to the open atmosphere and therefore i t was kept under vacuum in a desicator.  The material was used  for the next reaction step without further purification, since repeated attempts to crystallize from various solvents failed.  - 184 Indole-2-acetonitrile (86) A solution of 2-dimethylaminomethylindoie methiodide (18 g, 0.057 mole) and potassium cyanide (12 g, 0.185 mole) in absolute methanol (500 ml) was refluxed, with stirring, under an atmosphere of nitrogen, for 20 hours. After this time, approximately 350 ml of methanol was distilled and the residue poured into ice-water. The resulting mixture was extracted thrice with ether, the combined ether extracts were washed with ice-cold water, and then dried over anhydrous sodium sulfate.  Removal of the ether produced  a dark brown solid (9 g) which was chromatographed over alumina (500 g). The desired compound was eluted from the column with benzene-ether (9:1) which on recrystallization from ether-pet. ether (b.p. 60-80°) afforded white plates (4.5 g 56%) m.p. 95-97.5° (Lit. m.p. 96-98°). Indole-2-acetic acid (87) A solution of indole-2-acetonitrile (400 mg, 2.56 m.mole) in 95% ethanol (5 ml) was added to a solution of potassium hydroxide (410 mg, 7.3 m.mole) i n water (2 ml).  The resulting mixture was refluxed with  stirring under an atmosphere of nitrogen for 45 hours.  The solution was  diluted with water (10 ml) and the ethanol removed under vacuum. The resulting aqueous solution was washed twice with ether, and then carefully acidified with concentrated hydrochloric acid.  The mixture was extracted  twice with ether, the ether extracts were combined, washed once with water, and dried over anhydrous sodium sulfate.  Evaporation of the ether afforded  a crude brown product (440 mg, 98% crude yield).  The infrared spectrum  was nearly identical to that of indole-2-acetic acid, reproduced by 32 Schindler.  - 185 Indole-2-acetic acid, methyl ester (88) To a stirred solution of crude indole-2-acetic acid (420 mg) in anhydrous ether (10 ml) kept in an ice-water bath, a solution of diazomethane in ether-ethanol was added until the yellow color of the diazomethane persisted.  The resulting solution was allowed to stand for 30  minutes and the solvent was evaporated in a stream of nitrogen, affording crude product (436 mg).  The crude product was passed through a short  column of alumina (10 g) using benzene-ether (1:1) as eluant. Evaporation gave pale yellow crystals (327 mg) which upon crystallization from etherpentane afforded  colorless blocks (296 mg) m.p.  74-75°.  Infrared (Nujol):  0  3.06 (>NH), 5.85 (-C0CH) y. 3  Ultraviolet: X  218, 272, 289 my.  NMR  signals (CDC1 ): 2.55-3.35 (multiplet, 4H, aromatic) 3.84 (singlet, IH, (  B-proton of the indole), 6.67 (singlet, 2H, -CH  2  -CH  2  C00CH ). 3  COOCH )6.41 (singlet, 3H,  Found: C, 69.40; H, 6.22; N, 7.65.  3  Calc. for C H 0 N : n  n  2  C, 69.82; H, 5.86; 0, 16.91; N, 7.40. 3-chloroacetylindole-2-acetic acid methyl ester (89) Indole-2-acetic acid methyl ester (88) (1 g, 0.0053 mole) was dissolved in dry ether (7 ml) and chloroacetonitrile ( 8 ml) was added to this solution.  The mixture was cooled to 0° and dry hydrogen chloride  was passed through the solution for approximately 3 hours.  At the end of  this time a heavy white precipitate was formed, and the reaction mixture allowed to stay at 0° for another 2 hours.  The precipitate was filtered  and washed twice with dry ether. This precipitate was treated with water (100 ml) to affect hydrolysis of the iminium salt.  The salt dissolved  instantly, but on hydrolysis to the corresponding ketone yielded an insoluble solid compound.  This solid was filtered and washed twice with water  and then extracted three times with chloroform. The combined chloroform  - 186 extracts were washed twice with water and dried over anhyrous magnesium sulfate.  Evaporation of chloroform under vacuum afforded a white semi-  crystalline material. This material was passed through a short alumina column (25 g) using chloroform as eluant.  Evaporation  afforded a nice crystalline material (880 mg, 62%).  of the chloroform A small amount of  this material was recrystallized from dichloromethane-ethyl ether to provide^an analytical^sample m.p. 142-144°C. 5.79  (-COCH3),  6.2 (-(icH Cl) y. 2  Infrared (KBr): 3.0 (>NH),  Ultraviolet: X )  214 (4.523), 244 m3.x  (4.147), 267 (3.9555), 309 (4.048) my. NMR signals ((CD ) C0): 2.15-3.15 0 3  2  ti  (multiplet, 4H, aromatic), 5.3 (singlet, 2H, -C-CH C1), 5.8 (singlet, 2H, 2  -CH COOCH ), 6.4 (singlet, 3H, -CH COOCH ). 2  3  2  0, 17.81; N, 5.42; CI, 13.24.  3  Found: C, 58.62; H, 4.75;  Calc. for C H 0 NC1: :3  12  3  C, 58.76; H, 4.55;  0, 18.07; N, 5.27; CI, 13.35. 3-acetylindole-2-acetic acid methyl ester (90) 3-chloroacetylindole-2-acetic acid methyl ester (89) (71 mg) was dissolved in 95% ethanol (30 ml) and hydfogenated over Raney Nickel catalyst at room temperature and atmospheric pressure.  After a period of  30 minutes the uptake of hydrogen became very slow and the reaction was interrupted.  The catalyst was removed by filtration and the ethanol was  distilled under vacuum.  The residue was taken with dichloromethane, and  the insoluble inorganic salts i n i t i a l l y present in the ethanolic solution were filtered.  The filtrate was concentrated down to provide a crystalline  material ,(60 mg, 96%).  This was recrystallized from ethyl ether-dichloro-  methane to provide an analytical sample m.p. 128-131°. 0 0 3.16  Infrared (KBr):  ( NH), 5.66 (-C0CH ), 6.1 (-C-CH) y. Ultraviolet: X 3  3  243 (4.114), 272 (3.882), 303 (3.954) my.  m a x  215 (4.442),  NMR signals (CDC1 ): 2.8 3  (multiplet, 4H, aromatic), 5.71 (singlet, 2H, -CH2COOCH ), 6.33 (singlet, 3  - 187 0 3H, -CH COOCH ), 7.4 (singlet, 3H, C-CH ). 2  N, 5.86.  3  3  Calc. for Ci H 0 N: 3  13  3  Found:  C, 67.24; H, 5.60;  C, 67.53; H, 5.63; N, 6.06.  3-(3-carbomethoxy-N-piperidyl)-acetylindole-2-acetic  acid methyl  ester (92) Well powdered 3-chloroacetylindole-2-acetic acid methyl ester (89) (380 mg, 0.0014 mole) was mixed with 3-carbomethoxy piperidine (76) (840 mg, 0.0059 mole).  The mixture was warmed slightly from time to time in a  steam bath, until a clear solution was obtained.  This reaction mixture  was allowed to stay over night at room temperature.  At the end of this  time water (50 ml) and ethyl ether (50 ml) were added and the mixture was shaken until the viscous o i l was distributed between the aqueous and the ether layer.  The ether layer was washed once with water and dried  over anhydrous magnesium sulfate. The ether was removed by vacuum distillation to afford a gummy material (672 mg).  This material was  passed through a short s i l i c a gel column (6 g) using chloroform as eluant. The chloroform was removed by vacuum distillation to afford a gummy material (500 mg, 93%).  This material, in spite of numerous attempts  resisted crystallization.  For analytical purposes part of this material  was further purified by preparative thin layer chromatography (alumina, chloroform) to afford an amorphous solid. Infrared (KBr): 3.12 (>NH), 0 0 5.77 (2 x - C O C H 3 ) , 6.1 (-CCHN ) y. Ultraviolet: A 214 (4.604), 243 2  (4.266), 268 (4.066), 303 (4.14) my. NMR signals (CDC1 ): 2.7 (multiplet, 0 4H, aromatic), 5.67 (singlet, 2H, -CH COOCH ), 6.17 (singlet, 2H, CCH N<), 0 — 0 3  2  6.3  3  ?  (singlet, 3H, -C0CH ), 6.46 (singlet, 3H, C0CH ). 3  3  H, 6.72; 0, 22.84; N, 6.92. Calc. for C H O N : 20  0, 21.48; N, 7.52.  2i+  5  2  Found: C, 64.21; C, 64.5; H, 6.5;.  - 188 3-ethylindble-2-acetic acid methyl ester (91) (a) •By catalytic hydrogenation of 90. A solution of 3-acetylindole-2-acetic acid methyl ester (90) (225 mg) in 1 % hydrochloric acid in methanol (15 ml) was hydrogenated over Adams catalyst at room temperature and atmospheric pressure. After a period of 30 minutes, when approximately 2 moles of hydrogen had been absorbed, the reaction was interrupted.  The catalyst was filtered and the methanol was  removed by vacuum d i s t i l l a t i o n .  The residue was taken with ethyl ether,  and the ether extract was washed twice with water and dried over anhydrous magnesium sulfate.  Removal of the ether afforded an oily material (170 mg)  Preparative thin layer chromatography ( s i l i c a gel, chloroform-ethylacetate 1 : 1 ) of this substance provided the hydrogenation product (91) (front running material) as a viscous o i l ( 1 0 0 mg) and starting material (50 mg). 0  Infrared ( l i q . film): 2.9 (>NH), 5.8 276(sh), 285, 293 my. 6.28  (singlet,  (quartet, (b)  2H,  2H,  NMR signals  -CH2COOCH3),  -CH2CH ), 3  (C0CH ) 3  (CDC1 ): 3  6.31  y.  Ultraviolet: A  2.85  (multiplet,  225,  m a x  4H,  aromatic)^  (singlet, 3H, -CH COOCH ) 7.38 2  8.8 (triplet, 3H -CH CH ) 2  3  3  y  T.  By diborane reduction of (90) A solution of 3-acetylindole-2-acetic acid methyl ester (90) (50 mg,  0.0002  mole) in dry tetrahydrofuran  (10  ml) was cooled at  0°  and diborane,  generated from boron trifluoride etherate ( 2 ml) in diglyme (8 ml) and sodium borohydride (500 mg) in diglyme (15 ml), was passed through the solution for a period of 1 1 / 2 hours. stayat 0 ° for another 30 minutes.  The reaction mixture was allowed to  The excess of diborane was destroyed  with water and the mixture was extracted with ether. The ether extract was washed with water and dried over anhydrous magnesium sulfate.  The  - 189 -  ether was removed by vacuum distillation to provide a clear viscous oily material, which turned yellow in contact with the air. Preparative thin layer chromatography (silica gel, chloroform-ethyl acetate 1:1) of this material provided the desired product 91.  (Front running compound, 20 mg,  42%). 3-[8-(3-carbomethoxy-N^piperidyl)-ethyl]-indple-2-acetic acid methyl ester (93) The coupling product 92 (200 mg, 0.00054 mole) was dissolved in dry tetrahydrofuran and the solution cooled to 0°.  Diborane, generated  from boron trifluoride etherate (3 ml) in diglyme (15 ml), and sodium borohydride  (1 g) in diglyme (15 ml), was passed through the cold solution  for a period of 3 hours.  Dilute (10%) acetic acid (10 ml) was then added  slowly to the reaction mixture.  The resulted solution was allowed to stay  at room temperature for tern minutes and subsequently extracted twice with ether.  The combined ether extracts were washed once with 5% potassium  carbonate solution, twice with water and dried over anhydrous magnesium sulfate. (160 mg).  Removal of the ether afforded an amorphous yellowish solid This solid was dissolved in dry dioxane (5 ml) and a mixture  of concentrated sulfuric acid (0.2 ml) in dioxane (10 ml) were added to the above solution. The mixture was shaken well and allowed to stay at room temperature for 15 minutes. 10% sodium bicarbonate, ethyl  The sulfuric acid was neutralized with  ether (50 ml) was then added to the dioxane  solution and the mixture was washed several times with water to remove the dioxane.  The ether solution was dried over anhydrous magnesium  sulfate and the ether was removed with vacuum distillation to provide a gummy material (150 mg).  Preparative thin layer chromatography  (alumina, chloroform-ethyl acetate 1:1) of this  - 190 material afforded the desired reduction product (93) (Front running material, 60 mg, 28%).  This compound, in spite of numerous attempts,  resisted crystallization. aromatic), 6.2  NMR signals ( C D C 1 ) : 2.8 (multiplet, 4H, 0 (singlet, 2H, -CH0COOCH3), 6.26 (singlet, 3H, -COCH ) 6.3 3  3  0  /  (singlet, 3H, -COCH3).  This compound was further characterized as an  amorphous hydrochloride, which was prepared by passing dry hydrogen chloride through an ether solution of the product 93 and subsequent removal of the ether under vacuum to dryness.  Infrared (KBr): 5.79  0 (2  x -COCH3) p.  my.  Found:  Ultraviolet: 2 2 2 (4.3), 275(sh) 283 (3.7), 2.92 (3.664) y  C, 60.90; H, 7.33; 0, 16.04;  N,  5.85.  Calc. for C2oH270itN Cl: 2  C, 60.83; H, 6.89; 0, 16.2; N , 7.09. y-Benzyloxy propanol ( 1 0 0 ) Sodium (25 g, 1.08 moles) was added in portions to a hot (115-120°) and vigorously stirred solution of trimethyleneglycol (240 g, freshly d i s t i l l e d , (b.p. 115-119°/9 mm) in dry xylene (100 ml). was not necessary until near the end of the reaction.  External heating Benzyl chloride  (150 g) was slowly added with stirring to the hot ( 1 2 0 ° ) solution of the sodium derivative.  The mixture was heated ( 1 2 0 ° ) and stirred for 1 hour,  after a l l the benzyl bromide had been added.  The mixture was cooled to  room temperature and the precipitate was removed by f i l t r a t i o n .  The  filtrate was d i s t i l l e d under reduce pressure through a distillation column.  After a fore-run of xylene and dibenzyloxypropane the y-benzyloxy  propanol (125 g, 69%, Lit. 73%), boiled at 109-110°/8 mm) mm).  (Lit. 145-150°/13  NMR signals (neat): 2.76 (singlet, 5H, aromatic), 5.59 (singlet, IH,  -OH), 5.65 (singlet, 2H, C^CH^O-), 6.35  (triplet, 2H, -CH^OH), 6.51  (triplet, 2H, (JJCH^O CH -), 8.21 (quintet, 2H, -0CH CH_2CH 0H) x. 2  2  2  - 191 Benzyl y-chloropropyl ether (101) Thionyl chloride (120 g) was added drop by drop to a mixture of y-benzyloxy propanol (100) (160 g) and N-dimethyl aniline (130 g). The temperature of the reaction mixture was kept below 60° during the addition of the thionyl chloride.  After a l l the thionyl chloride was added the  reaction mixture was allowed to stay at 60° for another 1/2 hour.  At the  end of this time the reaction mixture was poured into excess of dilute hydrochloric acid, and the heavy o i l which separated was extracted with chloroform.  The chloroform extract was washed with dilute hydrochloric  acid to remove the dimethylaniline, then with water and dried over anhydrous sodium sulfate.  The solvent was evaporated and the o i l was d i s t i l l e d .  Benzyl-y-chloropropyl ether was at once obtained (150 g, 81%, Lit. 83%) as a colorless o i l b.p. 95-100°/l mm (Lit. 129°/16 mm). 2.72  NMR signal (neat):  (singlet, 5H, aromatic), 5.72 (singlet, 2H, 4>CH 0-), 6.57 (triplet, 2H, 2  <J>CH OCH -), 6.65 (triplet, 2H, -CH^Cl), 8.2 (quintet, 2H, -CH CH2CH -) T. 2  2  2  Ethyl y-benzyloxypropyl malonate (102) A solution of sodium (68 g, 2.96 gram atoms) in absolute alcohol (1250 ml) was made in a five l i t e r three-necked flask.  The solution was  cooled and treated with ethyl malonate (675 g, 4.23 moles).  The suspension  was refluxedand stirred while benzyl-y-chloropropyl ether (101) (520 g, 2.82 moles) was added over a period of 3 hours, and then the stirring and refluxing were continued' for 21 hours.  Most of the alcohol was removed  by d i s t i l l a t i o n , the residue cooled and water added to dissolve the inorganic salt. separated.  The layers were acidified with glacial acetic acid and  The water layer was extracted thrice with ether.  Oil and  extracts were combined, washed once with water, twice with 10% sodium  - 192 bicarbonate, dried  over  once  with  anhydrous  mm ( L i t . 1 9 3 - 2 0 0 ° / 4 2.71  5H,  (singlet,  saturated  sodium c h l o r i d e  sodium sulfate. mm) w e i g h e d  580"g  5.59  aromatic),  The p r o d u c t  solution  distilling  (76%, L i t . 7 7 % ) .  2XOCH2CH3),  6.6  (two t r i p l e t s , 3 H ,  plet,  4H,  Ethyl  Y-benzyloxypropylethyl  malonate  A solution (200  m l ) w a s made  cooled  all  (6  the ethyl  was  The- m o s t  allowed  to  acid.  combined ether  extracts  solution was  The p r o d u c t Infrared 5H,  gram atoms)  were  solution,  added  once  and the remained  (102)  malonate  and then  allowed  slowly  water  once  with  with  added  with water,  anhydrous  aromatic),  5.6  5.81  with twice  f o r 10  2H, C H C H 0 - ) , 6  5  2  5.88  t o 60°. After hours  The  (84%).  2.72 4H,  ether  pressure.  123 g .  (quartet  and the  chloride  reduced  (neat):  with  10% o f  with  magnesium s u l f a t e .  NMR s i g n a l s  g,  the mixture  ether  sodium  o i l was d i s t i l l e d u n d e r  (-COCH^) y .  (singlet,  to cool  stirring.  w a s d i s t i l l e d a t 2 2 0 - 2 2 2 ° / ! - 5 mm a n d w e i g h e d 0  ( l i q . film>:  (134.5  and a c i d i f i e d  twice  concentrated  ethanol  T h e s o l u t i o n was  was r e m o v e d b y d i s t i l l a t i o n ,  washed  dried over  i n absolute  flask.  The m i x t u r e was e x t r a c t e d  and f i n a l l y  evaporated  (103)  t o room t e m p e r a t u r e ,  glacial, acetic  (multi-  3  t h e s t i r r i n g was c o n t i n u e d  of the ethanol  cool  sodium bicarbonate  2  to boil  was t h e n  i o d i d e was a d d e d ,  \  2  5  y-benzyloxypropyl  0.64 m o l e )  g,  (quartet,  0  6 H , 2 x - 0 C H C H ) T.  0.56  (13 g ,  ethyl  (neat):  iodide  The m i x t u r e was h e a t e d  iodide  60°.  with  NMR s i g n a l s  1  6  i n a one l i t e r t h r e e - n e c k e d  and t r e a t e d  0.43 m l ) . Ethyl  o f sodium  150-160°/1  C H C H 0 C H 2 - a n d -C-CH-C -), 8.2  (triplet,  A l k y l a t i o n o f 102 w i t h e t h y l  (a)  at  8.86  -OC^Cj^Cj^C),  at  2 H , CgHsCHoO-), 5.9  (singlet,  — 4H,  and f i n a l l y -  (singlet,  2X-OCH2CH3),  1 6.6 8.85  (triplet, (triplet,  2 H , C H CH pC(-|g), 8.2 6  5  2  6H, 2xOCH CH ), 2  3  9.2  (multiplet, (triplet,  6H,  -OCHzQl^CH^CCH^GHg) ,  3H, C-CH CH ) x. 2  3  Found:  - 193 C, 67.75;.;.H, 8.31; 0, 23.94.  Calc. for C H280 : 19  5  C, 67.83; H, 8.39;  0, 23.78. (b).  Alkylation of ethyl diethyl malonate with 101 A solution of sodium (11.2 g, 0.48 moles) in absolute ethyl alcohol  (205 ml) was made in a one l i t e r three-necked flask.  The solution was cooled  and treated with ethyl diethyl malonate (90 g, 0.48 mole). The mixture was heated to boil and benzyl y-chloropropyl ether (101) (88 g, 0.48 mole) was . added in a period of 1/2 hour with continuous stirring of the reaction mixture.  The refluxing and stirring were continued for 10 hours.  At the  end of this time, part of the ethanol was d i s t i l l e d (180 ml), the mixture was cooled to room temperature and water (300 ml) was added.  The mixture  was acidified with glacial acetic and extracted twice with ether.  The  combined ether extracts were washed twice with 10% of sodium bicarbonate, once with saturated sodium chloride and dried over anhydrous magnesium sulfate.  The ether was evaporated to afford 150 ml of a heavy o i l , which  was d i s t i l l e d under vacuum. The desired material (103) boiled at 150-170°/ .02 mm and weighed 90 g (56%). y-benzyloxypropylethyl malonic acid (104) To a stirred solution of potassium hydroxide (90 g) in water (140 ml), ethanol (35 ml) and ethyl y-benzyloxypropylethyl malonate (103) (112 g) were added.' The mixture was heated (60°) with stirring for 15 hours.  The  ethanol was evaporated i n a steam bath and the alkaline solution was washed twice with ether (the ether washings were discarded). Water (100 ml) was then added, the mixture was acidified to Congo red with concentrated hydrochloric acid and extracted three times with ether.  The combined  ether extracts were washed twice with water, once with saturated sodium  - 194 chloride and dried over anhydrous magnesium sulfate.  The ether was  evaporated and the viscous oily residue was crystallized from n-hexaneether to provide colorless blocks (91 g, 97%), m.p. (KBr): 5.74  (weak), 5.89  (strong)y. NMR  singlet, 2H, 2x-C00H), 2.65 C H CH 0-), 6.47 6  5  117-120°. Infrared  signals (CDC1 ): -0.5 3  (singlet, 5H, aromatic), 5.48  (broad  (singlet, 2H,  (triplet, 2H, C H CH 0CH2-) , 7.8-8.8 (multiplet, 6H,  2  6  -0CH CH2C^CCH2-), 9-12 2  5  2  (triplet, 3H, -0CH CH3) 2  . Found: C, 64.14;  H, 7.12; 0, 28.62. Calc. for C H o0 : C, 64.27; H, 7.19; 0, 28.54. 15  2-(y-benzyloxypropyl)-butanoic  2  5  acid (105) v  The diacid (104) (41.85 g) was heated at 140-150° for 5 hours (the evolution of carbon dioxide ceased at the end of this time).  The  product, a yellowish viscous o i l , was used for the subsequent reactions without further purification. (broad, -OH absorption) -COOH), 2.71  Infrared ( l i q . film): 5.87  . NMR signals (CDC1 ): -0.6  (triplet, 2H, C H CH 0CH^-) 7.65 5  5  2  6H, -OCHzC^CH^CCH^CHg) , 9.11 Methyl  (singlet, IH,  3  (singlet 5H, aromatic), 5.55  (- C0H),2.8-3.7  (singlet, 2H, C^CH^O-), 6.56  (multiplet, IH, CHC00H), 8.45  (triplet, 3H,  (multiplet,  CHCH CH ) T. 2  3  2-(y-benzyloxypropyl)-butanoate(106a) The viscous oily acid (105) was dissolved in dry ethyl ether  (50 ml) and the mixture was cooled to 0°.  A solution of diazomethane in  ether-ethanol was added to the above solution until the yellow color of the diazomethane persisted. The ether was evaporated in a stream of nitrogen and the resulted o i l was d i s t i l l e d under reduced pressure to afford a colorless o i l (28 g, 75%) b.p. 180-186°/1.2 mm. (liq. film): 5.19 aromatic), 5.67  (-C0CH ) V'. 3  NMR signals (neat) : 2.75  (singlet, 2H, C^CH^O-) , 6.67  Infrared (singlet, 5H,  (triplet, 2H, CeHsCH^CH^-)  - 195 6.5  (singlet, 3H, -OCH ), 7.8 3  6H, -0CH CH CH2(|cH2CH3), 9.2 H, 8.48;  (triplet, 3H, C-CH CH3) T.  2  2  0, 19.76.  (multiplet, IH, CHCOOCH3), 8.5 2  Calc. for C H 0 : 15  22  Found:  C, 71.97; H, 8.86;  3  (multiplet, C, 71.76;  0, 19.17.  Ethyl 2- (y-benzyloxypropyl) -but;anoate(106b) To a solution of 2-(y-benzyloxpropyl)-butyric acid (105)  (14.9  g)  in absolute ethanol (250 ml), concentrated sulfuric acid (2 ml) was added, and the mixture was  refluxed for 6 hours.  Part of the ethanol was  distilled  (150 ml) and the rest of the reaction mixture was poured into ice-water. The mixture was extracted twice with ether and the combined ether extracts were washed once with water, once with 5% sodium bicarbonate solution, once again with water and dried over anhydrous magnesium,sulfate. was evaporated and the resulted o i l was  The ether  d i s t i l l e d under reduced pressure  to afford the ethyl ester (106b) (12.8  g, 76%) as colorless o i l . b.p. 0  159°/.25 mm.  (C0C H ) y.  2.73 2H,  Infrared ( l i q film): 5.8  (singlet, 5H, aromatic) , 5.6 -OCH2CH3),  6.61  2  NMR  5  signals  (triplet, 2H, C Hc;CH 0CH -) 7.75 R  7  ?  (multiplet,  (multiplet, 6H, OCH CH CH CCH CHci) , 8.85  -OCH CH ), 9.15  (triplet, 3H, C-CH CH ) T.  2  5  3  0, 17.85.  ?  2  Calc. for C H 16  21+  ?  3  2  ?  3  IH,  (triplet,  Found: C, 73.16; H,  0 : C, 72.69; H, 9.15;  (quartet,  ?  CHC00C H ), 8.4 2  (neat):  (singlet, 2H, Cf;H CH 0-), 5.95 q  156-  3H,  8.99;  0, 18.16.  Preparation of sodium triphenyl methane Sodium (3 g) covered with dry xylene was flame.  carefully melted with free  To this melted sodium, kept under a dry nitrogen atmosphere,  mercury (200 g) was  slowly added.  The remained xylene was  the amalgam, allowed to cool to room temperature.  decanded and  To the cold amalgam,  ethyl ether (50 ml) and triphenyl methyl chloride (11 g) were added and the mixture was shaken vigorously for 6 hours.  At the end of this time  - 196 another  80 ml of dry ether were added, the mixture was shaken well and  allowed to rest for 5 to 6 hours.  The dark-red ether solution was then  transferred into the reaction flask with caution, under a dry nitrogen atmosphere.  The concentration of this solution was approximately 0.2 mole/  lit. Methyl a-(y-benzyloxypropyl)-ot-ethyl-succinate  (107)  To an ether solution of sodium triphenyl methane (0.03 mole) the ester 106a (7.5 g, 0.03 mole) was added and the mixture was stirred at room temperature for 1/2 hour.  At the end of this time, methyl bromoacetate  (5 g, 0.03 mole) was slowly added and the mixture was stirred for another 15 minutes.  The reaction mixture Was then treated with water (50 ml) and  the layers were separated.  The ether layer was washed once with water and  dried over anhydrous magnesium sulfate. a yellowish viscous o i l .  Removal of the ether afforded  To this oily residue, benzene (10 ml) was added  and the mixture allowed to rest until most of the triphenyl methane had crystallized.  The triphenyl methane was filtered and the filtrate was  chromatographed from alumina (200 g).  The triphenyl methane which remained  in the filtrate was eluted with benzene-pt. ether (30-60°) (1:1) and i t was followed by the unreacted ester 106a (2.5 g).  Finally the desired product  (107) was eluted'-with benzene as yellowish viscous o i l and i t was further purified by vacuum distillation (bath temperature 200-220°/.1mm, 3.6 g, 37%). Infrared (liq. film): 5.8 (2xCOCH ) u. 3  5H, aromatic), 5.56  6  3  C-CH CH ) T. 2  3  signals (CDC1 ): 2.74 3  (singlet, 2H, C H CH_20-), 6.4  (singlet, 3H, -0CH ), 6.6 -CH^COOCH^ 8.32  NMR  5  (singlet, 3H, -0CH ) 6.46 3  (triplet, 2H, C H CH 0CH2-) 6  5  2  ;  (multiplet, 6H, OCHpCHpCH^CCHpCH^ 9.2  Found: C, 67.27; H, 7.74; 0, 24.99.  (singlet,  7.4 (singlet, 2H, (triplet, 3H,  Calc. for C H 0 : 18  26  5  - 197 C,  6 7 . 0 8 ; H, 8.07;  Ethyl  a-(y-benzyloxypropyl)-ot-ethyl-succinate To  ( 5 g , 0.019  m o l e ) was a d d e d and  1 1/2 h o u r s a t room t e m p e r a t u r e .  m o l e ) was t h e n s l o w l y a d d e d and 10 m i n u t e s . ml)  and  the  w a t e r and  the mixture  allowed  was  chromatographed from alumina  remained i n the  (70a)  (singlet,  f o r another  washed t w i c e  o i l , benzene  most o f the  with  the  (10 ml) was  t r i p h e n y l methane  and  the  filtrate  The t r i p h e n y l m e t h a n e w h i c h  e l u t e d with benzene-pet. ether e s t e r 106b  g, 20%).  Infrared (liq. film):  3  5H, a r o m a t i c ) ,  (2.3  (bath  g).  (30-60°)  The  temperature  5.8 (-COC H ) u . 2  200-  NMR  5  5.57 ( s i n g l e t , 2H,  ( o c t e t , 4H, 2X-0CH CHCJ) , 6.6 ( t r i p l e t , 7  2H, C H C H 0 C H - ) 6  5  2  2  2 H , - C H C 0 0 C H ) , 8.4 ( m u l t i p l e t , 6 H , OCH CH;,CH CCH CH ), 7  2  s  ?  ( s e x t e t , 6 H , 2 x - O C H C H ) , 9.2 ( t r i p l e t , 2  6 8 . 4 7 ; H, 8.55;  0, 22.83.  0.019  was e l u t e d w i t h b e n z e n e a s a y e l l o w i s h v i s c o u s o i l  ( C D C 1 ) 2.75 ( s i n g l e t ,  CeHsCH^O-^ 5.87  C,  g).  g,  Removal o f t h e  i t was f u r t h e r p u r i f i e d b y vacuum d i s t i l l a t i o n  signals:  8.79  l a y e r was  To t h i s  i t was f o l l o w e d b y t h e u n r e a c t e d  220°/.1 mm 1.93  7.4  (200  (3.2  t r e a t e d w i t h w a t e r (50  The t r i p h e n y l m e t h a n e was f i l t e r e d  f i l t r a t e was f i r s t  desired material and  oil.  torest until  crystallized.  and  The e t h e r  a yellowish viscous  was s t i r r e d  was f u r t h e r s t i r r e d  d r i e d o v e r anhydrous magnesium s u l f a t e .  had  (1:1)  the mixture  was s u b s e q u e n t l y  l a y e r s were s e p a r a t e d .  (0.019 m o l e ) t h e  E t h y l bromoacetate  the mixture  The r e a c t i o n m i x t u r e  ether provided a d d e d and  (70 a)  an e t h e r s o l u t i o n o f s o d i u m t r i p h e n y l methane  e t h y l e s t e r 106b for  0, 2 4 . 8 1 .  0,  3  22.99.  Calc. f o r C oH 2  ?  3H, C - C H C H ) 2  3 0  05:  3  .  C, 6 8 . 5 4 ; H,  ?  3  Found: 8.63;  J  - 198 Ethyl  -  g-keto-y-(Y-benzyloxypropyl)-Y-ethyl-glutarate  (70b)  To an e t h e r s o l u t i o n o f sodium t r i p h e n y l methane (0.018 mole), e s t e r 106b 1 1/2 was  g, 0.018  mole) was  hours at room temperature.  added and the mixture E t h y l bromopyruvate  then s l o w l y added and the mixture  minutes. and  (4.9  The  r e a c t i o n mixture  was  the l a y e r s were s e p a r a t e d .  was  stirred for g, 0.018  f o r another  d i l u t e d w i t h benzene (10 ml)  f i l t r a t e was  alumina (200  t r i p h e n y l methane which remained i n the  f i r s t e l u t e d w i t h benzene-pet. e t h e r  by the u n r e a c t e d  e s t e r 106b.  The  bejizene as a y e l l o w i s h v i s c o u s o i l and i t was distillation f i l m ) : 5.8  a r o m a t i c ) , 5.57 6.6  signals:  ( t r i p l e t , 2H,  6  C H CH 0CH -)., 7.09 fi  s  ?  7.88;  ?  0,  25.52.  and  i t was  was  e l u t e d with  g, 23%).  (CDC1 ) : 2.73 3  ( q u a r t e t , 4H,  5  ( q u a r t e t , 2H,  Calc. for C i H  followed  2  Infrared ( l i q . (singlet,  2x-0CH?CHQ,  -CH C00C?HQ, ?  3 0  5H,  8.35  O : C, 66.64; H,  7.99;  6  25.37.  Condensation o f the keto g l u t a r a t e (70b) A mixture tryptamine was  1.4  ( s i n g l e t , 2 H , C H C H 2 0 - ) , 5.88  Found: C, 66.80; H, 0,  NMR  (70b)  filtrate  f u r t h e r p u r i f i e d by vacuum  (bath temperature 210-230°/ 1 mm, (-CCOEt, -COOEt) \i.  The  chromatographed from  (30-60°) (1:1)  d e s i r e d product  and  crystallized.  f i l t e r e d and the  was  ml)  Removal o f the e t h e r a f f o r d e d  t r i p h e n y l methane was The  10  washed once w i t h water  allowed t o r e s t u n t i l most o f the t r i p h e n y l methane had  g).  mole)  t r e a t e d with water (50  e t h e r l a y e r was  and d r i e d over anhydrous magnesium s u l f a t e . a y e l l o w i s h v i s c o u s o i l , which was  (3.63  further stirred  subsequently  The  was  the  w i t h tryptamine  o f the keto g l u t a r a t e (70b)  hydrochloride  (70 mg,  (240 mg,  0.00026 mole)  hydrochloride  0.006 mole) and  i n absolute ethanol  r e f l u x e d f o r 60 hours under a n i t r o g e n atmosphere.  The  ethanol  (2 was  ml)  - 199 evaporated under vacuum and the residue,was treated with an ether-water mixture until i t was distributed between the twq layers.  The' ether layer  was washed with water, dried over anhydrous magnesium sulfate.  Removal of  the ether provided an oily material (250 mg) which showed three spots on thin layer chromatography ( s i l i c a gel, chloroform). chromatographed from alumina (20 g).  This mixture was  The unreacted keto glutarate (70b)  (100 mg) was eluted with benzene and i t was followed by an oily substance (20 mg) which indicated no ultraviolet absorption corresponding to an indole chromophore. benzene-ether (9:1).  Finally, a gummy material (40 mg) was eluted with This latter material showed two spots on thin layer  chromatography (alumina, chloroform^ethyl acetate 9:1) when the plate was developed three times in this solvent system. (sharp, > NH), 5.92 p.  Infrared  (CHCI3):  Ultraviolet: 233, 275(sh), 284, 292 mp.  2.79 Further  information about this material are given in the following experiment. Condensation of the keto glutarate 70b with tryptamine Tryptamien hydrochloride (400 mg, 0.0013 mole) was dissolved in water. The solution was basified with sodium bicarbonate solution and the free base was extracted with ether.  The ether extract was dried over anhydrous  magnesium sulfate and the ether was removed under vacuum. The triptamine residue was dissolved in absolute ethanol (20 ml) and keto glutarate (70b) (500. mg, 0.0032 mole) was added to this solution.  The mixture was heated  (120°) in a sealed tube for 20 hours.  The solution was cooled to room  temperature, the ethanol was removed  under vacuum and the residue was  taken with ether.  The ether solution was washed once with 5% hydrochloric  acid solution,, once with water and dried over anhydrous magnesium sulfate. Removal of the ether afforded a gummy material (650 mg).  This material  - 200 showed two spots on thin layer chromatography (alumina, chloroform-ethyl acetate 9:1) when the plate was developed three times in this solvent system.  Thin layer chromatography, IR and UV spectra indicated that the  above mixture was identical with the one obtained from the reaction of ketoglutarate (70b) with tryptamine hydrochloride as i t is described in the previous experiment.  Preparative thin layer chromatography (alumina,  chloroform-ethyl acetate 9:1) of the reaction mixture (450 mg) - the plate was developed 5 times in the above solvent system - afforded the less polar material (A) (196 mg) and the more polar one (B) (164 mg) as gummy substances.  The compound B was crystallized from n-hexane-ether to pro-  vide an analytical sample, m.p. 90-96°.  Compound A, in spite of numerous  attempts, resisted crystallization.. Compound A. X  Infrared (CHC1 ): 2.79 (sharp,>NH) 5.81, 5.93 u. 3  223, 275(sh), 284 and 292 mu.  Ultraviolet:  NMR signals (CDC1 ): 1.7 (singlet, IH, 3  >NH) , 2.7 (multiplet, 9H, aromatic) , 3.05 (doublet, IH, ot-proton of the indole), 5.53 (singlet, 2H, C H CH 0-), 5.8 (quartet, 2H, -OCHpCHQ , 6.5 fi  (multiplet, 6H, -CH CH N< 7  s  ?  and CRH^CH^OCH?-), 7.0 (quartet, 2H, -CH COCO-),  ?  ?  8.26 (multiplet, 6H, -0CH CH CH CCH CH3), 8.78 (triplet, 3H, -OCH CH ) , ?  ?  ?  ?  2  3  9.18 (triplet, 3H, C-CH CH ) x. 2  Compound B. X  3  Infrared (CHC13) : 2.78 (sharp, >NH)^5.8, 9.2 p.  223, 275(sh), 284 and 292 my.  Ultraviolet:  NMR signals (CDC1 ): 1.7 (singlet, IH, 3  IT13.X  >NH), 2.75 (multiplet, 9H, aromatic), 3.02 (broad singlet, IH, B-proton of the indole), 5.6' (singlet, 2H, C H CH_20-), 5.9 (quartet, 2H, -OCH^CHQ , 6  5  6.6 (multiplet, 6H, -CH?CH-N< and C H CH 0CH?-), 7.1 (quartet, 2H, ?  fi  fi  ?  -CH C0C0-), 8.35 (multiplet, 6H, 0CH CH CH CCH CH,3), ?  7  ?  ?  -0CH CH ), 9.15 (triplet, 3H, C-CH CH ) x. 2  3  0, 16.98; N, 6.02.  2  3  7  8.8 (triplet, 3H,  Found: C, 69.75; H, 7.32;  Calc. from C 9H 0 N2: C, 70.11; H, 7.37; 0, 16.24; 2  36  5  - 201 N,  5.69. The structures 110 and 111 has been tentatively assigned to the  compounds A and B respectively as i t is described in the foregoing discussion of this thesis. N- [8- (3-indolyl).-ethyl] -a- (y-benzyloxypropyl) - a-ethyl-succinimide  (112)  Tryptamine hydrochloride (10 g, 0.033 mole) was dissolved in water, the aqueous solution was basified with 20% sodium bicarbonate solution and the tryptamine base was extraceted with ether.  The ether extract was  washed once with water, dried over anhydrous magnesium sulfate and the ether was removed under vacuum.  To the crystalline residue, a solution  of ethyl succinate (70a) (4.8 g, 0.013 mole) in greshly d i s t i l l e d diethylene glycol mpnoethyl ether (60 ml) was added and 20 ml of solvent was subsequently d i s t i l l e d in order to azeotrope traces of water from the mixture.  Tryptamine hydrochloride , (300 mg) was then added and the  mixture was refluxed for 40 hours in a nitrogen atmosphere.  The reaction  mixture was allowed to cool to room temperature and diluted with ethyl ether (200 ml).  The ether solution was washed 5 times with water, twice  with 10% glacial acetic acid solution once with 5% sodium bicarbonate solution, once again with water and dried over anhydrous magnesium sulfate. The ether was removed and the brown residue was chromatographed from alumina (500 g). The desired product (112) was eluted with benzenechloroform; (9: 1) as a gummy material' (3.9 g, 69%).  This compound, in  spite of numerous' attempts, resisted crystallization.  Infrared(CHCl3):  2.86  m  (>NH), 5.67 weak, 5.91  283, 291 my. • NMR  2H, C H CH20-), 6.25 5  Ultraviolet: X  221, 275(sh)  signals (CDC1 ):T.0 (singlet, 1H,>NH), 2.7  9H, aromatic), 3.06 6  (strong) y. 3  (doublet, IH, g-proton of the indole)., 5.58  (multiplet, (singlet,  (triplet, 2H, -CH?CH?N ), 7.0 (triplet, 2H,-CH?CH N<), ?  - 202 6.68  P  (triplet, 2H, C H CH OCH2-), 7.6 (singlet, 2H, C-CHp-C-), 8.5 6  5  2  (multiplet, 6H, -OCH CH CH CCH CHQ, 9.27 (triplet, 3H, C-CH CH ) x; MW, 7  418.  ?  7  7  2  3  Calc. for C H O N 2 : MW, 4 l 8 . 25  30  3  Bischler-Napieraski cyclization of the imide (112) The benzyl-imide (112) (230 mg) was dissolved in dry xylene (35 ml) in a 50 ml three-necked round bottom flask.  From above solution, xylene  (10 ml) was removed by distillation in order to azeotrope traces of water. To the stirred refluxing solution, phosphorous pentoxide (approximately 1.5 g) was added in three portions (500 mg) over a period of 45 minutes under a dry nitrogen atmosphere. continued for another 2 hours.  The refluxing and stirring was further  The reaction mixture was allowed to cool  to room temperature, the xylene was decanded and the brown precipitate was washed twice with ether.  (The xylene and the ether washings were  d i s c a r d e d ) T h e brown precipitate was treated with ice-water (50 ml), the mixture was made strongly alkaline with concentrated potassium hydrohydroxide and extracted 5 times with ether.  The combined ether extracts  were washed once with water and dried over anhydrous magnesium sulfate. Removal of the ether afforded an amorphous yellowish solid (130 mg), which was subsequently chromatographed from alumina.  Elution with benzene  afforded a yellowish amorphous solid (20 mg) which was further purified by preparative thin layer chromatography (alumina, benzene-chloroform  1:1).  The recovered white amorphous solid (10 mg) indicated the following spectral properties.  Infrared CHCl ): 3  223(sh), 229, 314, 327 mp.  3.95, 6.05 (weak) p. Ultraviolet: A Founds MW, 292 . Calc. for C H H : MW. 292 . 19  20  2  N-[B-(3-indolyl)-ethyl]-a-(3-hydroxypropyl)-q-ethyl-succinimide (115) To an ice-cold solution of the imide (112) (3.95 g) in dry dichloro-  - 203 methane (25 ml), boron tribromide (3 ml) was added and the mixture was allowed to stay at 0° for 10 minutes.  At the end of this time ethyl ether  (100 ml) was added and the resulted mixture was subsequently treated with ice-cold water.  The layers were separated and the ether layer was washed  once with water and dried over anhydrous magnesium sulfate. Removal of the solvent afforded a yellowish gummy residue (3.7 g) which was chromatographed from alumina (100 g). The desired alcohol (115) was eluted with benzene-chloroform (1:4) as a colorless gummy material (2 g, 65%). This material, in spite of numerous attempts, resisted crystallization.  Infrared  (CHC1 ): 2.87 (>NH), 2.95 (-OH), 5.66 (weak), 6.1 y. . Ultraviolet: X  '  222  3  IHclX  275(sh), 283, 292, my. NMR signals (CDC1 ): 1.47 (singlet, IH, >NH), 2.7 3  (multiplet, 4H, aromatic), 3.06 (doublet, IH, g-proton of the indole), 6.17 (triplet, 2H,. -CH CH2N<), 6.95 (triplet, 2H, -CHCH?N<) , 6.58 (triplet, 2H, 2  H O C H 2 - ) , 6.75  ?  (singlet, IH, H0CH -) 7..61 (singlet, 2H, C-CH^-), 8.5 2  /  (multiplet, 6H, -OCH CH CH CCH CHO, 9.28 (triplet, 3H, C-CH CH ) . 2  ?  7  9  Found: MW, 328. Calc, for C H 19  2  21t  3  0 N : MW, 328. 3  2  N-[8-(3-indolyl)-ethyl-q-(3-chloropropyl)-a-ethyl-succinimide (116) To an ice-cold solution of the alcohol 115 (2 g) in dry ether (50 ml), a mixture of thionyl; chloride (10 ml), dry pyridine (10 ml) and dry ether (25 ml) was slowly added with stirring.  The reaction mixture was subse-  quently stirred for 15 hours at room temperature.  At the end of this time .  benzene (25 ml) was added and.the mixture was refluxed for 3 hours. The reaction mixture was subsequently cooled to 0° and the thionyl chloride was destroyed with ice-cold water.  The organic layer was washed twice  with water and dried over anhydrous magnesium sulfate. Removal of the solvent afforded,a gummy residue (1,8 g) which was subsequently chromato-  - 204 graphed from alumina with  The d e s i r e d  chloride  b e n z e n e as a c o l o r l e s s gummy s u b s t a n c e  crystallization. chloroform. A  (100 g).  (CHCl3):  eluted  which r e s i s t e d  ( 2 9 0 mg) was e l u t e d  2.87 (>NH), 5.67 ( w e a k ) , 6.1 y.  2 2 3 , 275 ( s h ) , 2 8 3 , 292 my. NMR  v  (0.87 g, 4 1 % ) ,  Unreacted s t a r t i n g m a t e r i a l  Infrared  ( 1 1 6 ) was  (CDCI3):  signal  with  Ultraviolet:  1.85 ( s i n g l e t , I H ,  in 3.x  >NH),  2.7 ( m u l t i p l e t , 4 H , a r o m a t i c ) ,  indole),  (triplet,  6.15  (triplet,  2 H , -CH?C1),  Calc.  for C  1 9  H  2 3  IH, g-proton o f the 2 H , -CH2CH N<)  ?  7.59 ( s i n g l e t , (triplet,  0 N C1: 2  (doublet,  2 H , - C H ? C H N < ) , 6.93 ( t r i p l e t , 0  9.25  -OCH2CH2CH2CCH2CH3),  3.0  2  MW,  2  2 H , ( C - C H ^ C ) , 8.5  3H, C - C H £ H ) T. 2  3  (multiplet,6 H ,  F o u n d : MW,  346.  346.5.  2,3,5,6,ll-pentahydro-2fr(3-chloropropyl)-2g-ethyl-3-oxo-indolo(2,3-g)i n d o l i z i n e (117) The  chloro-imide  .116 ( 1 2 0 mg) was d i s s o l v e d i n d r y x y l e n e  a three-necked round bottom f l a s k . was  removed b y d i s t i l l a t i o n  i n order  From t h e a b o v e s o l u t i o n , x y l e n e t o azeotrope traces  s t i r r e d , b o i l i n g s o l u t i o n , phosphorous pentoxide portions  ( 5 0 0 mg) o v e r a p e r i o d  o f 2 0 minutes  atmosphere.  The m i x t u r e was f u r t h e r s t i r r e d  subsequently  cooled  was  washed t w i c e  discarded).  t o 0°.  with  The x y l e n e  dry ether  (thexylene  The  r e s u l t e d e m u l s i o n was e x t r a c t e d extracts.were  magnesium s u l f a t e .  washed once w i t h  ( 1 . 5 g) was a d d e d i n t h r e e  and under a d r y n i t r o g e n  and t h e e t h e r  concentrated  w a t e r and d r i e d o v e r afforded  were  (30 m l ) a n d  potassium  ether.  solid  washings  i c e - c o l d water  5 times with  Removal o f t h e e t h e r  To t h e  and r e f l u x e d f o r l h o u r , and  The b r o w n s o l i d was t r e a t e d w i t h  m i x t u r e was made' s t r o n g l y b a s i c w i t h  ( 1 0 ml)  o f water.  was d e c a n d e d a n d t h e b r o w n  the  ether  (35 ml) i n  hydroxide.  The c o m b i n e d anhydrous,  a brownish  amorphous  solid. The  a b o v e p r o c e d u r e was r e p e a t e d 5 t i m e s ,  using  each time  1 2 0 mg o f  - 205 the chloro-imide 116. The products of the above five reactions were combined (450 mg) and chromatographed  from alumina (40 g). The desired  product (117) was eluted with benzene as a yellowish amorphous solid (104. mg, 18.5%), which, in spite of numerous attempts, resisted crystallization. Infrared: (KBr) : 3.04 (>NH), 5.93, 6.04 y. Ultraviolet:. X 229 (4.653), 303(sh)  m a x  223(sh),  (4.375), 313 (4.481), 327 (4.38) my: X ^ ° ' H  230(sh), 303(sh), 314, 327, 387 my.  HC1)  220,  NMR signals ( C D C I 3 ) : 1.6 (singlet,  IH, >NH), 2.7 (multiplet, 4H, aromatic), 4.64 (singlet, IH, Cj olefinic proton), 6.15 (triplet, 2H, C methylene), 6.96 (triplet, 2H, C 5  6  methylene), 6.59 (triplet, 2H, -^201)^8.2 (multiplet, 6H, 0CH CH CH CCH?CH3), 9.3 (triplet, 3H, C-CHyCHQ T. Found: C, 68.88; ?  ?  ?  H, 6.58; N, 8.3; C l , 10.11; MW, 328. Calc. for C H iON Cl: 19  2  1  C, 69.4;  H, 6.43; 0, 4.86; N, 8.52; C l , 10.87; MW, 328.5. I, 2,3,5,6,11,11b(g)-heptahydro-2g-(3-chloropropyl)-2g-ethyl-3-oxo-indolo(2,3-g)indolizine (118) A solution of enol lactan 117 (60 mg) in glacial acetic acid was hydrogenated over Adams catalyst at room temperature and atmospheric pressure until the hydrogen uptake ceased.  The catalyst was filtered,  the f i l t r a t e was diluted with chloroform and the mixture was carefully treated with 20% potassium bicarbonate solution.  The chloroform layer  was washed once with' water and dried over anhydrous magnesium sulfate. Removal of the chloroform provided an amorphous white solid which was subsequently chromatographed  from alumina (80 mg). Elution with benzene  provided a crystalline compound (40 mg), which was recrystallized from nrhexane-dichloromethane, m.p. 148-156°C:.  Thin layer chromatogrpahy  (alumina, benzene 1:1) of this material indicated two distinct spots  - 206 -  0 A  when the plate was developed 5 times. Ultraviolet: X  Infrared (KBr): 5.99 (>N-C-) y.  224, 275(sh), 283, 290 my.  Found:. C, 67.94;.H, 6.71;.  N, 8.04; CI, 9.69; MW, 329. Calc. for C H 30N C1: 19  0, 4.83;  2  2  C, 68.96;.H, 7.00;  • N, 8.46; CI, 10.72; MW,330.5.  The two components in the above mixture were effectively separated by preparative thin layer chromatography (alumina, benzene) when the plate was developed 5 times.  Both components indicated very similar IR ( A  CHCl , 5.95 ) and UV (X 3  m a x  .224, 275(sh), 283, 290 my) spectra and MW, 329.  Lithium aluminium hydride reduction of 118 A mixture of the two isomers of.the amide 118 (20 mg) was treated with lithium aluminum hydride in dry tetrahydrofuran with refluxing for 24 hours.  Theexcess hydride was destroyed with water and the inorganic  salts were filtered.  The f i l t r a t e was diluted with.ether, the mixture was  washed 3 times with water and dried over anhydrous magnesium sulfate. Removal of the ether provided an amorphous solid' (15 mg) which showed two spots on thin layer chromatography (alumina, benzene) when the plate was developed 5 times. r  Ultraviolet: X 225, 275(sh), 283, 290 my. max ^ 1  The two components were- separated by preparative thin layer chromatography (alumina, benzene) when the plate was developed 5 times. 5  components indicated MW, 282. Calc. for C H N 19  16  2  MW, 282.  Both  -  207  -  REFERENCES 1.  M. Hesse.  "Indolalkaloide", Springer-Verlag, Berlin-GSttingen-Heidelberg, (1964).  2.  K. Mothes and H.R. Schutte.  3.  R. Robinson.  Angew. Chem. (Int. Ed.),  2,  441  (1963).  "The Structural Relations of the Natural Products",  Clarenton Press, Oxford J. Aiii. Chem. S o c ,  (1955).  4.  E. Wenkert.  5.  D. Schumann and H. Schmid.  6.  J.P. Kutney, J. Trotter, T. Tabata, A. Kerigan and N. Camerman. Chem. and Ind.,  648  84, 98  (1962).  Helv. Chim. Acta,  4 J , 1996  (1963).  (1963).  7.  J.P. Kutney and E. Piers.  J. Am. Chem. S o c ,  8.  A. Camerman, N. Camerman, J.P. Kutney, E. Piers and J. Trotter. Tetrahedron Letters, U,  9. 10.  637  86, 953  (1964).  (1965).  J.P. Kutney, R.T. Brown and E, Piers. J. Am. Chem. Soc., 8 6 , J.P. Kutney, R.T. Brown and E. Piers.  2286  J. Am. Chem. Soc ,  8 J , 2287  447  (1964).  (1964) . 11.  J.P. Kutney, R.T. Brown and E. Piers.  Lloydia, V.  12.  J.P. Kutney, R.T. Brown and E. Piers.  Can. J. Chem., in press.  13.  S. Stork and J.E. Dolfini.  14.  Y. Ban, V. Sato, I. Inoue, M. Nagai, T. Oishi, M. Terashima, 0. Youemitsy and Y. Kanaoka. Tetrahedron Letters, 2261  J. Am. Chem. Soc. ,  8J,  27,  4,  (1964).  2872.(1963).  (1965) . 15.  J.E.D. Barton and J. Harley-Mason.  16.  G. Buchi, D.L. Coffen, Karoly Kocsis, P.E. Sonnet and F.E. Ziegler. J. Am. Chem. S o c , § 7 , 2073 ( 1 9 6 5 ) . J.W. Huffman, C.B.S. Rao, T. Kamiya. J. Am. Chem. S o c , 8 J , 2288  17.  Chem. Comm.,  298  (1965).  (1965).  18.  CW. Sumpter and F.M.. Miller.  "The Chemistry of Heterocyclic  Compounds", Intern. Publishers, Inc., N.Y. and London, Vol. VIII pp.  3-23  (1954).  - 208 19.  C.F.H. Allen and CV. Wilson.  J. Am. Chem Soc., 65, 611 (1943).  20.  F.T. Tyson.  21.  H.J. Stephen.  22.  A. Bischler and P. Fireman.  23.  A. Reissert.  24.  P.P. Shorygin and K.S. Polyakova.  J. Am. Chem. Soc., 6_3, 2024 (1941). J. Chem. Soc., 1874 (1925). Chem. Ber., 26, 1336 (1893).  Chem. Ber., §0, 1030 (1897). Sintezy Dushistyky Veshchestv,  Sbornik Statri, 130 (1939); Chem. Abst., 36 , 3802 (1942). 25;  R. Adams. Organic Reactions, Vol. V, pp. 387-411  26.  R. Seka.  27.  U. Renner, D.A. Prins, A.L. Burlingame and K. Biemann, Helv. Chim. Acta, 4§, 2186 (1963).  28.  A.J. Scott.  Chem. Ber., §6.-, 2058 (1923).  "Interpretation of the Ultraviolet Spectra of Natural Products",  Vol. 7, p.166.  29.  N. Abdurahman.  Unpublished results.  30.  R.R. Phillips.  Organic Reactions, Vol. X, p.143.  31.  Van Order and Lind Nail.  32.  W. Schindler. Helv. Chim, Acta, 4J, 2156 (1957).  33.  G.V.D. Tiers. "N.M.R. Summary", Central-Research Department, St. Paul 6,  Chem. Revs., 3Q, 69 (1942).  Minnesota. 34.  K.T. Potts and D.R. Lidjegren. J. Org. Chem., 28, 3066 (1963).  35.  B. McKague and C.E. Gletsos. Unpublished results.  36.  H. Emde. Annalen, 391, 88 (1912).  37.  H, Emde and H. Kull. Arch.  38.  D.B. Clayson.  39.  G. Childs and E.J. Forbes. . J. Chem. Soc., 2024 (1959).  40.  E. Leete.  Pharm., 274, 173 (1936).  J . Chem. Soc., 2016 (1949).  Chem. and Ind., 692 (1960);  J. Am. Chem. Soc, 8J, 6338  (1960) . 41.  E. Wenkert, S. Garratt and K.G. Dave.  42.  E. Piers and T. Inaba.  Can. J. Chem., | 2 , 489 (1964).  Unpublished results.  - 209 43.  L.J. Dolby and D.L. Booth.  J. Org. Chem.,  30,1550 (1965).  44.  L.I. Smith and J.A. Spring. J. Am. Chem. S o c , 65,  45.  G.M. Bennett and A.L. Hock.  46.  L.C Cherry and J.R. Prening.  J. Am. Chem. Soc., 67, 2213 (1945).  47.  B.E. Hudson and CR. Hauser.  J. Am. Chem. Soc., 6_3_, 3156 (1941) .  48.  CM. Whitesides, D. Haltz and J.D.. Roberts.  1 2 7 1  (1943).  J. Chem. Soc., 128, 472 (1927). -  J. Am. Chem. S o c , §6;,  2628 (1964). 49.  L.M. Jackman.  "Applications of NMR Spectroscopy jn Organic Chemistry",  Pergamon Press. (1959), p.89. 50.  W.M. Whaley and T.R. Govindachari.  Organic Reactions, Vol. VI, p.151.  51.  W.M. Whaley and T.R. Govindachari.  Organic Reactions, Vol. VI, p.74.  52.  G.C Morrison, W. Cetenko and J. Shavel.  J. Org. Chem., 29, 2771  (1964). 53.  R.D. Youssefyeh  54. M.E. Kuehne. 55.  and Y. Mazur.  Chem. and Ind., 609 (1963).  J. Am. Chem. S o c , 86, 2946 (1964).  The Sadtler Spec-Finder, Research Laboratories, Fifteen/Seventeen Vine Street, Philadelphia 2, Pennsylvania, Spectra No, 19268-19277.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0062216/manifest

Comment

Related Items