THE TOTAL SYNTHESES OF
{i} mam, amwmewme

Tim“ {ta-o {'56 Day?“ 0% M. 5.
MCEEQMQ SHE-1’13 MEWSETY
George Peter Nines
£967

 

\ V . --. 4,. .0_.' I! «lwnw‘F-‘d‘

 

LIBRARY

3 “w: Michigan §tatc
‘ Umversxty

 

ABSTRACT

THE TOTAL SYNTHESIS OF
(19 l-DEAZA-l-THIA-RESERPINE

By

George Peter Nilles

Reserpine, isolated from Rauwolfia Serpentina Benth., has wide-
spread use in modern pharmacology as an anti-hypertensive. The goal
of this investigation was to carry out the initial synthesis of a
compound in which the basic ring skeleton of reserpine was modified

while retaining all other functional groups intact (viz. l-deaza-l-

thia reserpine (A) ).

OCH3

OCH3
OCHS

 

The original total synthesis of reserpine (13,14) called for the

condensation of three molecular fragments (B, C, D).

George P. Nilles

CHSO , OH CL OCHs

C) I
OCHS OCH3
OCH3

(C) (D)

Molecules (C) and (D) have been previously described by this synthesis.
Thus attention was directed toward the synthesis of the sulfur analog of

(B); the thianapthyl derivative (E) of the indole molecule, having sulfur

in place of the indole nitrogen!

I
O S NHZ

CH3
(2)

Titus (10) in these laboratories had already synthesized the amide
from which the amine (E) could be obtained by simple reduction. He was
unable to effect such a reduction with any of the common reducing
agents.

Application of a new reductive process (19) in the present work,

utilizing diborane lead to the formation of (E) in good yield.

George P. Nilles

Molecule (C) was synthesized by a procedure which involved
some original work and incorporating the findings of other inves-
tigators (12, l3, 14). In the final stage of the synthesis of (A)
the sequence of the reactions necessary to effect condensation of
the molecules (C, D, E) was changed from those reported to take
advantage of favorable steric factors.

The 3,4,5-trimethoxybenzoyl function was attached to molecule (C)
before molecule (E) was added to obtain (A). The effect of this pro-
ceedure was to isolate exclusively the desired "thiareserpine" mol-
ecule having the proper stereéchemistry at the C-3 carbon thus ob-
viating the necessity to separate the thiareserpine from its C-3
epimer.

Advantage was taken of other newer developments since the advent
of the original reserpine synthesis to shorten the number of steps
required and to improve the overall yields of intermediates.

These included the conversion of the precursor of (C) i.e.
compound (1") by ozonolysis, to obtain directly the 18,8 ester function
eliminating the need for a laborious acid catalyzed epimerization of

the C-3 hydrogen.

COOCH3
o3 CHO

OCHQCHS ' OCH3 OCH? “3
OCH3 (C,

 

 

'0!

(F)

This and other minor changes in reaction conditions improved the
yields in some Specific steps in the initial synthetic sequence of
reactions to obtain (A). As a result the desired compound was synthe-

sized in an overall yield of 0.21 from its simplest starting materials.

THE TOTAL SYNTHESIS OF
('3'.) 1-DEAZA.- l-THIA-RESERPINE

By

George Peter Nilles

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirments
for the degree of

MASTER OF SCIENCE

Department of Chemistry

1967

6;?”(7p75/

{'95 £07

ACKNOWLEDGEMENT

The author wishes to express his sincere appreciation to
Professor Robert D. Schuetz for his guidance and encouragement
throughout the course of this investigation.

Special thanks are also extended to those individuals of
the Department of Chemistry who made available the necessary

instruments used in the experimental portion of this work.

ii

TABLE OF CONTENTS

Page
ACKNOWLEDGEMENTS 0 o o o o o o o o o o o O 0 O O o o o .00 ' ii
LIST OF FIGURES e e o o 0 e o e o o o o e e e o o e e e 0 V
INTRODUCTION AND HISTORICAL. . . . . . . . . . . . . . . .
DISCUSSION 0 Q Q 0 O O O O O O O O O O O O O O O O O O O O 9
PART I Preparation of 6-Methoxy-3-(2-Aminoethy1) Benzo-
thiophene. O O O O C O O O O C O O O O O O O O 9
PART II The Bicyclic Series . . . . . . . . . . . . . 11
PART III The Tetracyclic and Pentacyclic Series . . . 21
EXPERIMENTAL O O O O O O O O O O O O O O O O O O O O O O O 31
Preparations. O I O O O O O O 0 O O O O O O O O O O O 32
Meta-Methoxy Benzene Thiol. . . . . . . . . . . . . 32
Ethy1 (3-Ket0'4'Ch10rO) BUtanoate o o o e e o e o o 33
Ethyl 4-(Meta-Methoxy Phenyl Mercapto)-3-Keto
Butanoate O O O O O O O O O O O O O O O O O O O O O 33
Ethyl 6-(and -4-)-Methoxythianapthyl-3- Acetate. . 34
6-(and-4-)-Methoxy Thianapthyl 3-Acetamide. . . . . 3S
6-Methoxy-3-(2-Aminoethyl) Benzothiophene . . . . . 36
The BiCYCIic series 0 O O O O O O I O O O O C O O O O 37
cis-5,8-Diketo-l,4,5,8,9d,loflgHexahydronapthalene
1P.Carboxy11c AC1d. O O O O O O O O O O O O O O O O 37
cis-Sfl-Hydroxy-B-Keto-l,4,5,8,9d,lO‘-Hexahydro-
nathalene-Le-Carboxylic Acid. . . . . . . . . . . . 38
8-Keto-l,4,5,8,91,lOd-Hexahydronapthalene-y‘-
Carboxylic Acid 1,5-Lactone . . . . . . . . . . . . 39
59-Hydroxy-l,4,5,8,9d,lOd-Hexahydro-yo-Napthoic
ACid 1’8-MCtoneo a O O O O O O O O O O O O O O O I O 40
zfi-Bromo-l,2,3,4,5,8-cis-9x,lOu-Octahydro-6-ene-y6-
(3,5-Epoxy)-Napthoic Acid 1,8-Lactone . . . . . . . 41
Zd-Methoxy-l,2,3,4,5,8:cis-9d,10e-6-ene-Octahydro-
yfl-(3,S-Epoxy)-Napthoic Acid 1,8-Lactone. . . . . . 42
24-Methoxy-6d-Bromo-bB-Hydroxy-cis-Perhydro
(3,5-Epoxy)-leapthoic Acid 1,8-Lactone . . . . . . 43

iii

TABLE OF CONTENTS (CONTINUED)
Page

zfl-Methoxy-6dPBromo-7-Keto-cis-Perhydro-
(3,5-Epoxy)-y$Napthoic Acid 1,8 Lactone . . . . . . 44

Za-Methoxy-3fl-Hydroxy-5-ene-7-Keto-l,2,3,4,7,8-cis-
9‘,IW’octahYdrO'P'NapthOIC ACId o e s e t e e o s o 45

2—-Methoxy-33-Hydroxy-5-ene-7-Keto-l,2,3,4,7,8-cis-
9d,lei-Octahydro-ya-Napthoic Acid Methyl Ester. . . 46

3,4,5-Trimethoxy Benzoyl Chloride . . . . . . . . . 47

Zd-Methoxy-36-(3', 4', 5'—Trimethoxybenzyloxy)-5-ene-
7-Keto-l,2,3,4,7,8-cis-9d,lOugOctahydro-Lfi-Napthoic
Acid Methyl Ester . . . . . . . . . . . . . . . . . 47

The Tetracyclic and Pentacyclic Series. . . . . . . . 49

Ct) LB-Methoxycarbonyl-2fi-Methoxy-30-(3', 4', 5'-
Trimethoxybenzyloxy)-§6~Formyl-§§-Cyclohexyl Acetic
ACid O O O O O O O O O O O O O O O O O O O O O O O 49

(t) l-Deaza-l-Thia-Z,3-seco-3-Oxo-Reserpine . . . . 50
(t) l-Deaza-l-Thia-3,4-Dehydro-Reserpine Perchlorate 52
(t) l-Deaza-l-Thia-Reserpine. . . . . . . . . . . . S3
6-Methoxy-3-(2-Piperidinoethy1) Benzothiophene. . . 55

Methyl 3,4,5-Trimethoxy Benzoate and Methyl Cyclo-
hexane Carboxylate. . . . . . . . . . . . . . . . . 55

BIBLIOGRAPHY 0 O O O O O 0 O O O O O O O O 0 O O O O O O O 66

iv

Figure
l.

10.

11.

LIST OF FIGURES

The Infrared Spectrum of 1-Deaza-l-Thia-Reserpine

and a 0.100 M Mixture of Methyl Cyclohexane
Carboxylate, Methyl 3,4,5 Trimethoxy Benzoate and
6-Methoxy-3-2(Piperidinoethyl) BenzothiOphene in
Chloroform. . . . . . . . . . . . . . . . . . . .
The Ultraviolet Spectrum of 1-Deaza-l-Thia-Reserpine
and a 0.100 X 10 M Mixture of Methyl Cyclohexane
Carboxylate and 6-Methoxy-3- 2(Piperidinoethy1)Benzo-
thiophene in 95% Ethanol. . . . . . . . . . .
The Infrared Spectrum of ZuéMethoxy-3flh(3', 4' ,5'-Tri-
methoxybenzyloxy)- -5-ene—7-Keto-l, 2, 3, 4, 7, 8-cis-9d-
10N-Octahydro-lB—Napthoic Acid Methyl Ester in
Chloroform. . . . . . . . . . . . . . . . . . . .
The Infrared Spectrum of (:t) lfibMethoxycarbonyl-ZX-
Methoxy-3G=(3', 4' ,5'—Trimethoxybenzyloxy)- 58-Formyl-
Ge-Cyclohexyl Acetic Acid in Chloroform. . . . . .
The Infrared Spectrum of 6-Methoxy-3- 2(Aminoethyl)
BenzothiOphene between salts. . . . . . . . . . . .
The Infrared Spectrum of (i) l-Deaza-l-Thia-2,3-seco-
3—0xo-Reserpine in Chloroform. . . . . . . . . . . .
The Infrared Spectrum of (t) l-Deaza-l-Thia-3,4-
Dehydro-Reserpine Perchlorate in Chloroform. . . . .
The Infrared Spectrum of Ct) l-Deaza-l-Thia-Reserpine
in Chloroform. . . . . . . . . . . . . . . . . . . .
The Infrared Spectrum of l-Deaza-l-Thia-Reserpine in
KI. . . . . . . . . . . . . . . . . . . . . . . .
The Ultraviolet Spectrum of Qt) l-Deaza-l-Thia-Reser-
pine in 95% Ethanol. . . . . . . . . . . . . . .
The Ultraviolet Spectrum of (i) 1-DeazaJ1—Thia~3; 4a
Dehydro-Reserpine Perchlorate . . . . . . . . . . . .

Page

29

3O

57

58
59
6O
61
62
63

64

.
“-\

.65

INTRODUCTION AND HISTORICAL

Totally unaware of its chemical constitution, reserpine was
initially brought to the attention of western civilzation by Leonhart
Rauwolf some 400 years ago (1). Extracted from the root of the Indian

shrub Rauwolfia Serpentina Benth. ( named in honor of this sixteenth

 

century explorer ) it has stimulated the imagination of countless
men of chemistry and medicine.

Originally the entire ground root was used in the treatment
of hypertension. In 1952 a team of Swiss researchers (2) succeeded
in isolating the physiologically active principle, reserpine (1).
Although other compounds similar in structure, both natural and
artificial, have been discovered, none have an activity greater than

reserpine, which is effective at a dosage of 0.5 mg./day (3).

OCH3

OCH3 OCH3
(I) OCHs

 

Investigations directed towards its structural elucidation were
initiated independently both here and abroad. Dorfman and his colab-

orators (4,5) showed the overall structure to be (I).

Infrared; analysis of reserpine indicated the presence of an N-H bond,
two carbonyls probably as esters, and an ortho disubstituted benzene.
Elemental analysis gave an empirical formula, C33H4009N2. Alkaline
hydrolysis yielded 3,4,5-trimethoxy benzoic acid (II) and reserpic acid.
The reserpic acid was reconvertable to reserpine on treatment with
3,4,5-trimethoxy benzoyl chloride. The ultraViOIEt: Spectrum of reserpic
acid was almost identical to a tetrahydro76-carboline system (III, R=H)

and suggested a relationship to the Harmala alkaloid skeleton (III, R=0CH3).

3 OCH3 H
(II) (III)
Reduction of reserpine with lithium aluminum hydride afforded 3,4,5-
trimethoxy benzyl alcohol and reserpic alcohol (C22H3OO4N2). This diol
contained only two methoxy groups. The ll-methoxy group was confirmed on

the basis of permanganate oxidation followed by treatment with diazo-

methane to obtain the dimethyl ester of 4-methoxy-N-oxaly1 anthranilic
acid (IV).

CH3
CH3O OCH3

(IV)

A yohimbane type nucleus was evident upon selenium dehydrogenation
of reserpic acid, which yielded structure (V) which was confirmed by
synthesis. Dorfman (4,5) then postulated structure (V1) for reserpic
acid, which had to contain a methoxy group, a non phenolic hydroxyl,

and a carboxyl function in then unknown positions.

N 0 On —CCHs
HO H CH3 OCH3 H ——OH
«PCOOH
(V) -__d

(VI)

 

Potassium hydroxide fusion of reserpic acid led to 3-hydroxy
isophthalic acid (VII). Carbon atom number 14 (C-l4) should have been
lost as carbon dioxide. This assumption was based on the prior report
of the analogous reaction of the previously established rauwolscine.
This fact placed two of the groups named above as in structure (VIII).
An alternate structure (IX) was given less probability on biogenetic
grounds. The C-l6 carboxyl being considered more probable on comparison

to the Strychnos alkaloids (6).

O
HO

 

(VII) (IX)

On heating the methyl ester tosylate of reserpic acid in collidine
Dorfman obtained a compound C23I'12804N2 whose infrared -‘: and ultra violet
spectra suggested the presence of an ether substitutedegpunsaturated
ester. On this basis they assigned reserpine structure (I).

This work only partially established the complete structure since
a task of equal dimension still remained, namely the determination of
absolute configuration of reserpine. "Stereochemical abandon” as phrased
by Woodward, predominates among the members of the yohimbane family.
Reserpine itself contains six aswametric carbons. The first clue to
their orientations came from the fact that the methyl ester tosylate
of reserpic acid (X) easily formed a quaternary salt in collidine, and

this was postulated to have structure (XI) (7).

@Z
-_—‘f'()(:}425

 

 

2
0’\
OCHs

 

(X) (x1)

Structure (XI) for the quaternary salt demands a cis fusion of rings

D and E. Huebner and Wenkert (8) found a quaternary salt readily formed
in pyridine at room temperature upon the detosylation of 3-isoreserpinol
(XII) which requires a cis axial placement of the C-16 carbon function,
and shows that the C-16 group and the C-20 hydrogen are trans. Since res-
erpic acid readily forms a lactone (9) this places the C-18 hydroxyl

cis axial to the C-16 carboxyl (XIII).

 

 

(XII)

That reserpine possesses the epi allo configuration for the C-3
hydrogen ( the less thermodynamically stable form ) is shown by this
facile lactone formation. In the allo case (XIV) the carboxyl and the
hydroxyl functions are to far apart to form a lactone while the epi

allo case has these groups cis 1,3 diaxial.(XIII).

 

(XIV) (XIII)

The allo case can not be reoriented to the epi allo structure without
serious C-2, C-l6 group interactions (10). By molecular rotation
studies (10) the C-18 carbon was related to D-glyceraldehyde to give

the absolute configuration as in structure (XIII).

Further evidence of theII-oriented C-3 hydrogen came from infra .
studies (11,12). It was found that ale-oriented C-3 hydrogen provides
only a single C-H absorption at 2780 cm-1, while the-koriented C-3 hydro-
gen shows an absorption at 2740 cm-1as well. Since reserpine shows only
the 2780 cm-labsorption, the C-3 hydrogen must hep-oriented.

The remaining asymmetricn carbon at C—l7 was shown to be a by a
closer look at Dorfmans' work (9). Since the C-18 carbon group 18/3,
the formation of the quaternary salt (XV) should occur easilyionly if the
methoxy group at C-l7 could give anchimeric assistance to the back side
displacement of the C-18 tosylate. Therefore the C-17 methoxy group must
beta-oriented, thereby giving the absolute configuration shown in
structure (XVI).

 

That structure (XVI) is correct for natural reserpine was shown by the

brilliant total synthesis by Woodward (13,14).

7

Although numerous modifications of the basic reserpine skeleton
have been reported (1,15) these have involved only peripheral changes.
Of these the most effective from a phramacological point of view have
been the substitution of various ester functions at C-18.

Accordingly work was initiated in these laboratories to modify
the reserpine skeleton internally, namely replacement of ring B
(pyrrole) by thiOphene.

'Huckel molecular orbital calculations of bond orders in indole (l6)
and benzothiOphene (17) indicate a greater resonance contribution to

structure (XVII-A) than (XVII-B) in both benzothiophene and indole.

G _
X x’ ‘x 9
e e

(XVII-A) (XVII-B)

Calculations from average value bond lengths reveal the thiOphene
ring to be only 0.3 A 2 larger than pyrrole, 3.9 A 2 for thiophene
vs. 3.6 A 2 for pyrrole. These two facts tend to substantiate the
postulate that substitution of ptrrole by thiophene should not adversely
affect the phramacologic response of the molecule if this part is
engaged at either an active or locking enzymic site. Further support
of this postulate stems from the already established procedure of
substitution of thiophene for pyrrole in some phramaceuticals, albeit
with some decrease in activity (30). Hapefully this thia analog of
reserpine will overcome many of the adverse side effects of the natural
compound. This desire along with the want to establish and extend some
fundamental chemistry in the field of sulfur heterocyclic compounds
motivated this undertaking.

Woodwards' synthesis of reserpine involved the condensation of
three large molecules (XVIII-A), (XVIII-B), (XVIII-C) with five of
the six asymmetric carbons properly oriented in (XVIII-B).

| NH
0 S 2

(XIX)

C

O

OH CL 0 OCH3

OCH3 CH30 OCH3

(XVIII-B) (XVIII-C)

 

Therefore, attention was directed to the synthesis of the previously
unknown 6-methoxy-3-(2-aminoethy1) benzothiophene (XIX).

Titus (18) in previous work in these laboratories had synthesized
6-methoxy-3-acetamidobenzothiophene. Unfortunately he was unable to
reduce the amide to the desired amine. However, a recent paper by Brown
(19) late in 1964 showed a facile reduction of all types of amides to
the corresponding amines in high yield by borane in tetrahydrofuran.
With this new reductive procedure available, the total synthesis of

l-deaza-l-thia—reserpine was undertaken.

DISCUSSION

PART I Preparation of 6-methoxy-3-(2-aminoethyl) benzothiophene

Since Woodward had already described a stereospecific synthesis of
the E ring of reserpine (l4) attention was first directed toward the
synthesis of the sulfur heterocyclic amine necessary to form rings A,B
and part of C in the "thiareserpine" molecule.

Placement of the 6-methoxy group (i.e. ll-methoxy in thiareserpine)
was the initial consideration. Titus' original proce dure (18) called
for starting with meta-methoxy phenyl mercaptan (XXII) which he prepared
in a six step procedure. from the sodium salt of metanilic acid. It
was decided the overall procedure: could be considerably shortened by
taking advantage of a procedure» described by Godt and Wann (20).

Accordingly, meta-anisidine (XX) was diazotized and treated with
potassium ethyl xanthate to give meta-methoxy phenyl xanthate (XXI)
which was isolated but not purified. The crude xanthate upon alkaline
hydrolysis gave the desired mercaptan (XXII) in a 31.61 overall yield.

1. H030 1. OH
2. KC3H5820
CH3O NH2 CH 30 CH 30 SH
(XX) (XXII)
(XXI)

Using the mercaptan (XXII) Titus' procedure: was followed as out-

lined below to obtain the desired 6-methoxy-3-acetamido benzothiophene (XXIII)
in 267. yield. Q
1. Mg/Etzo I

“ 2. H 20

+
CH O NH2 CL OC2H5 v-

Q

 

 

10

0 P205 0 O
CH 0 S OCZHS CHSO U OC2H5

NHAOH

O

I
CH3O s NH2

(XXIII)

Titus' attempt to reduce the amide to the amine with lithium
aluminum hydride under a wide variety of experimental conditions gave
only a trace of the desired amine (18). Browns' general procedure: (19)
for the reduction of amides to amines using "diborane" (actually
borane-tetrahydrofuranate) readily gave the required amine (XIX) in 61.4%
yield. The amine was found to rapidly absorb carbon dioxide from the
atmosphere, and consequently had to be stored under nitrogen until
needed. Herz (21) reported similar behavior for 3-(2-aminoethy1)
benzothiophene. The melting point of the picrate derivative of the
amine differed by only a few degrees from that of Herz' desmethoxy
compound. With the amine in hand, work was directed toward preparing
the sterically complicated E ring (XVIII-B).

Considerable work has been exerted toward finding new routes
capable of effecting the synthesis of this reserpine intermediate in
greater overall yield and/or fewer steps in the experimental proceedure
(15). A number of these were utilized in the present work and will be
described in detail at the appropriate junctures.

11

PART II The Bicyclic Series

 

The initial step involved the Diels-Alder condensation of quinone
with vinylacrylic acid to form cis-5,8-diketo-l,4,5,8,9,lO-hexahydro-
napthalene l-carboxylic acid (XXIV). The stereochemical consequence of

the reaction is as shown.

 

The stereOChemistry of the adduct may be rationalized by invoking
two principles. Maximum molecular overlap favors a transition state in
which the carboxyl group is oriented to ap position. The formation of
a cis «5 ring fusion may be explained on the basis of the Woodward-
Hoffman rules (22) which state that an endo product should be expected
for a concerted cycloaddition reaction.involving 2 *’4‘W electrons.

The resultant product now has all three metrics carbons with
the proper stereochemistry for ring E of the thiareserpine molecule.

The reaction has, however, resulted in a racemic mixture of the two
adducts.*

Advantage was now taken of the cis fusion of the rings to bring
about selective reduction of the S-keto group, using sodium borohydride
in a slightly basic medium. As Woodward (14) pointed out, sodium boro-
hydride is not particularly selective in its reducing capacity. However,
from structure (XXV) it may be seen that interaction of the l-carboxyl
function with the 8-keto group can lead to steric repulsion of the boro-
hydride ion.

 

*
Resolution has been carried out at this stage (12) but in view of the
of the large quantities of material involved in the present work, this
operation would have been rather cumbersome.

12

 

 

(XXV)

As a result the reduction proceeded smoothly to give the S-hydroxy
compound (XXVI). The keto-carboxyl interaction also aids in bringing
about an even greater steric consequence. It serves to "freeze" into
position the normal cage like structure of the cis decalin system. As
a result attack by the borohydride ion must occur (for steric reasons)
only from the rear or "convex" side of the molecule, thereby obtaining
the «sconfiguration of the C-5 hydrogen. Hinderence to attack at the
concave side of the system was employed extensively throughout the en-
tire (XVIII-B) synthesis.

Compound (XXVI) may be regarded as a point of juncture for two
different routes directed toward obtaining an intermediate with the
proper substituent at C-2 (C-l7 in thiareserpine). This intermediate
(XXVII) may be synthesized by the following route (14).

 

(XXVI)

13

   

Al(10Pr)3
iOPr '

(XXVII)

A significant increase in yield was obtained (22.2% vs. 13.9% for
the former route) by employing a synthesis through the five member

lactone as follows:

 

(XXVI) (XXVIII) (XXIX) J

 

   

l4

Dehydration of the hydroxy ketone (XXVI) by heating a solution of it
in benzene under reflux with acetic anhydride produced the 1,5 lactone
in a 60% yield. It should be noted here that this fixes the geometry of
the decalin ring system in one of the two alternate conformations, i.e.
the C-1 carboxyl is oriented cis axial to the C-5 hydroxyl (XXXI) com-
pared to the alternate structure (XXXII) where the carboxyl is oriented

cis equitorial.

 

That the 1,5 lactone is thermodynamically less stable than the 1,8
lactone (XXIX) is readily shown by its facile formation from (XXVIII)

upon a Meerwein-Verley-Pondorff reduction of the 8-ketone.

A1(iOPr)3
iOPr

 

   

(XXVIII) (XXIX)

15

Thus, when the 1,5 lactone was heated in isopropanol containing a
slight excess of aluminum isopropoxide, the 1,8 lactone resulted in a
531 yield. Woodward claimed a 90% yield for this translactonization and
although this procemhumx was repeated several times in the present work
(even using isopropanol dried over sodium and freshly distilled, together
with freshly distilled aluminum isopropoxide) no improvement in yield
could be realized.

Having obtained compound (XXIX) its structure was ready for intro-
duction of the desired substituent at C-2. From structure (XXIXPA) it
may be easily seen that the C-5 hydroxyl is readily disposed to aid in
electrophilic attack upon a bromonium ion through theelectron rich £52
system, with concomitant formation of a 3,5 epoxide bridge. This leads
directly to structure (XXX) since again the attacking species ( in this

case the bromonium ion ) must occur from the convex side of the molecule.

 

   

(XXIX-A) ~ (XXX)

By treating compound (XXIX) with bromine in methylene chloride
following Woodwards' procedure. (14) the bromo-epoxide was indeed ob-
tained. However, considerable bromination by addition also occured.

Presumably the dibromo material would have structure (XXXIII).

OH

O . Br
Br

(XXXIII)

 

16

In spite of the lack of experimental detail the procedure; described
by Velluz (12) for the preparation of compound (XXX). was utilized. This
involved the bromination of (XXIX) with N-bromo succinimide in tert-
butanol. Compound (XXX) was obtained in an 87% yield ( compared with
Woodwards' proceedure which gave a 47% yield). No detectable amount of
the dibromo compound was produced. The increased yield of the product
via the N-bromo succinimide route may be attributed to two factors. One,
the increased polarity of the protic solvent, tert-butanol, leads more
readily to the formation of a bromonium ion (23) and secondly the total
bromine concentration is lower than in the case where elemental bromine
is utilized directly.

The introduction of the thiareserpine C-l7 etmethoxy group was read-
ily accomplished by treatment of the bromolactone (XXX) with sodium
methoxide in methanol at room temperature. That the entering methoxy
group has the same configuration as its bromo predecessor may be explained

on the basis of an 8N1 reaction mechanism.

 

   

OCH3

e OCH3 (XXVII-A) (XXVII)

Dreiding models of the intermediate carbonium ion (XXVII-A) indi-
cate the structure must remain in the same locked configuration despite
the formation of an sp2 C-2 carbon. Attack of the methoxy ion then
occurs from the sterically accessible convex side to yield the methoxy
lactone (XXVII).

l7

Werk was now advanced to place substituents in the upper ring
which would allow subsequent conversion to the aldehyde (XVIII-B).

Addition of hypobromous acid to the 456-7 system was accomplished
using N-bromo succinimide in dilute sulfuric acid. At first glance it
may be difficult to visualize why the prefered bromohydrin is 6-bromo-
7-.hydmxy rather than 7-bromo-6-vhydrbxy. The structure may be assigned
on the following basis. The initial attack is by the 156.74? electron

system on the bromonium ion to give structure (XXXIV).

B'r

     
 

 

(XXXIV) OCH3

Inspection of carefully constructed Dreiding models reveals just
slightly less steric interference at the C-7 carbon than the C-6 carbon
( which suffers from steric hinderence by the methine hydrogen at C-2 ).
Consequently attack by water proceeds to give the trans diaxial bromo-
hydrin.

The oxidation of the bromohydrin was carried out using either
Woodwards‘ (14) or Velluz' (12) procdure . Essentially both involve
the use of chromium trioxide in a weak acidic medium. In the present
work it was possible to obtain only a 33% yield of the bromo ketone
(XXXVI) using Woodwards procedure with an aqueous acetic acid solvent.
However, by using a mixture of acetic acid, water, and phosphoric
acid employing a repetitive addition of the oxidizing medium in accord
with Velluz' method, a 56% yield of the bromo ketone resulted.

18

 

c5043
(XXXVI)

A brilliantly fortuitous reductive process described by Woodward (14)
was now carried out. By using zinc dust in glacial acetic acid, three
objectives were met in one step, namely, a dehydrohalogenation to form
a AS-éuf unsaturated ketone, the opening of the 3,5 epoxy bridge to
yield the 3p hydroxy function, and opening of the 1,8 lactone to form
the y# oarboxylic acid (XXXVII). Mechanistically, Woodward portrays

this process as follows:

 

(XXXVII)

The sucess of this reaction depends on two factors. The period of
reduction was very short (90 sec. in the present work). Furthermore,
when acetic acid only slightly wet with water, was used, an over reduc-

tion of the system occurred.

19

However, by using a solvent of sufficiently low water content, prepared
by distilling a mixture of glacial acetic acid containing 22 acetic
anhydride, the reduction could be carried out in 78% yield. The product
was readily identified on the basis of its ultraviolet spectrum, a
maximum occuring at 227mm , (£8 10,100) compared to Woodwards' (14)
values of 228 avg (£310,000). When over reduction of the.£P-6 system
occured the xmax frequently appeared in the neighborhood of 4000-6000.

The acid was now esterified with a slight excess of ethereal
diazomethane to obtain the 1,5 methoxy carbonyl ( 16/6 in thiareserpine)
compound (XXXVIII) in a 98% yield.

 

 

(XXXVII) (XXXVIII)

At this juncture it was decided to introduce the 3,4,5-trimethoxy
benzoyl group (XVIII-C) for two reasons. It markedly increased the total
amount of material available, and it was felt that the introduction of
this bulky substituent at this stage (Woodward introduced it only in
the last stage of his work) would exert a favorable steric guidance
effect on the later orientation of the hydrogen at the thiareserpine
C-3 carbon. As evidence for support of this approach, Protiva (24,25)
found that in his synthesis of l-deaza-l-thiadeserpidine, the product
was a 50/50 mixture of the C-3 epimers, when this group was introduced

prior to the C/D ring closure.

20

Accordingly this ester (XXXIX) was prepared in a carefully dried
mixture of pyridine and benzene by treating the alcohol (XXXVIII) with
3,4,5-trimethoxy benzoyl chloride.

 

OCH3
OCH3

 

CHsO

(XXXVIII) (XXXIX)

With the preparation of this compound (accomplished in 7.62 overall
yield from XXIV) work could be directed towards its condensation with
the thianapthyl amine system (XIX).

21

PART III The Tetracyclic and Pentacyclic Series

 

It was now desired to modify the upper ring of (XXXIX) by

conversion of carbon 5 to a formyl group. This would allow conden-

sation with the thianapthylamine (XIX) to form an aldimine (XXXXIV-A).

CHO

(WAS + Wm
(II-130 2

(XIX)

 

To effect this conversion, essentially two routes were open.
Woodwards' procedure: (14) called for the formation of the 5«,6£-
dihydroxy-7-ketone (XXXX) from the lga-hydroxyblh5~ene-7-one via
osmium tetroxide and cleavage of the dial to the formyl acid (XXXXI)

by means of periodic acid.

 

(XXXX) (XXXXI)

However, Velluz (12) prepared the intermediate (XXXXI) directly
from the eneone by ozonolysis. In each case the C-6 carbon is lost
as carbon dioxide. Blaha and his coworkers (26) prepared the aldehyde
(XXXXII) from (XXXIX) by ozonolysis, although they did not purify it.

CHO

HOOC o

CHSO , OCH3

(XXXIX) l'—0—3+ 0 (SC H3

2. H20

OCH
CH3o 3

(XXXXII)

Since the ozonolysis procedure“ is one step shorter, more economically

feasible, and less time consuming, it was adopted in the present work.
However, Blahas procedure: gave very poor yields of the desired

aldehyde as shown by the poor overall yield of the lactam (XXXXVI).

Improved yields were obtained using a highly modified proceedure. Here

the ozonolysis was carried out in methylene chloride at -300 using

a solution of potassium iodide as an external indicator. Use of even

a slight excess of ozone was found to decrease markedly the overall

yield of the desired lactam (XXXXVI).

 

23

The aldehyde was characterized as its previously unknown 2,4-dinitro-

phenylhydrazone. Treatment of the aldehyde with ethereal diazomethane

gave the triester (XXXXIII).

CH OOC CHO

3

0
CH 0 OCH3

3 e
O i .
OCH3 OCH3
CH 50

(XXXXIII)

Condensation of the triester (XXXXIII) with the thianapthylamine
(XIX) would be expected to produce the aldimine (XXXXIV-B) which upon
reduction with sodium borohydride would yield the amine (XXXXV). The
amine nitrogen could then attack the acetate function to give the
lactam (XXXXVI) by internal ammonolysis. The entire scheme being
analogous to that described by other authors (12, l3, 14, 15, 24, 26)

for the formation of such a system.

 

(XXXXIII)
’ I N831]: I :N H

‘m’ CH 0 S (o

3 0 CH
CH 30 CH30

(XXXXIV-B) (XXXXV)

 

24

8 r ~-——“ 8 0
CH30

(XXXXVI)

These expectations were born out in the experimental procedures
employed. In order to minimize the losses of the labile intermediates,
they were not isolated but rather all operations were carried out in
the original methylene chloride solvent used in the ozonolysis step.

Frequently the sodium borohydride reduction reduced not only the
aldimine but also the l8fester to the 18; alcohol as shown by the loss
of absorption at 1330 cm.1 (assigned to C-C bending of the carbonyl
carbon and the phenyl ring of the trimethoxy benzoyl function). In
support of this assumption, Blaha and his coworkers (26) also found
this hydrolysis to occur especially when the reaction mixture was
heated to reflux with the sodium borohydride.

Therefore, in the present work the reduction was carried out
at 0°. Even at this temperature the yellow orange color of the reaction
mixture (due to the C-N double bond) was rapidly discharged upon treat-
ment with the borohydride. The ammonolysis step proceeded rapidly
merely by allowing the reaction mixture to stand a few minutes at
room temperature.

It should be noted in passing that the formation of a l,2,3,4-
tetrahydro benzothienol2,3-é]pyridine system via a Pictet-Spengler
type ring closure, i.e. structure (XXXXVII) may be precluded on the

basis of two points of experimental evidence.

 

25

8 NO

fl

(XXXXVII)

CHO

The yellow orange color of the aldimine solution was stable at 0°,
whereas if the ring closure had occured, the color would have been
rapidly destroyed without the use of the reducing agent or it would
have not developed at all. Secondly, the elemental analysis supports
structure (XXXXVI) although structure (XXXXVII) can not be ruled out
purely on the basis of the percentage values found. Structure (XXXXVI)
C33H39010NS demands 0= 61.7673, H= 6.127.. Structure (XXXXVII) C33H37010NS
demands C: 61.95%, H=5.83‘Z.. The analysis found C: 61.51%, H= 6.07%.
However, considering the carbon/hydrogen ratio, compound (XXXXVI) has
a C/H ratio of 10.09l:l, whereas compound (XXXXVII) has a C/H ratio
of 10.626:1. The ratio was found to be 10.133:l.

With the obtaining of the lactam (XXXXVI) in an overall yield of
752 from the aldehyde (XXXIX) an intermediate was now at hand to allow
a condensation to form the complete ring skeleton of thiareserpine.

To effect this ring closure a modified Bischler—Napieralski reaction
was carried out.

The lactam (XXXXVI) was heated at 650 for 45 minutes with
freshly distilled phosphorous oxychloride to obtain l-deaza-l-thia-
3,4-dehydro-reserpine chloride (XXXXVIII-A).

 

  

Is 00 OCH3
OCH?)

0

oCH H3
X'-CI' (XXXXVIII-A) CH O

x"- C1647 (XXXXVIIl-B) 3

This rapid and facile ring closure is undoubtably aided by the presence
of the electron donating ll-methoxy substituent which increases the
nucleophilic character of the C-2 carbon. Thus, Protiva and his
coworkers (24) found the 11-desmethoxy compound formed in good yield
only after heating the corresponding lactam for two hours in phos-
phorous oxychloride in the presence of phosphorous pentoxide. This
amorphous immonium chloride was converted to the perchlorate salt
(XXXXVIIl-B) for purification and subsequent reduction. The bright
orange immonium perchlorate was found to have a pale orange fluorescence
in the solid phase, a very intense blue-green-white fLuorescenceLin a
wide variety of solvents, and a strong absorption at 366e7~(loges 3.99)
tailing off into the visible region. The fluorescent: Effect which further
supports structure (XXXXVIII-B) is undoubtably the consequence of the
highly conjugated charge delocalized aromatic system in which the
positive charge of the N43 nitrogen may be supported by the ll-methoxy
oxygen and the thiophene sulfur as shown in the contributing resonance

structures .

 

27

Advantage was taken of this fluorescenteffect in the following
reduction of the immonium salt to the desired l-deaza-l-thiavreserpine.

The immonium perchlorate was heated under nitrogen with zinc dust
in an acetone-water-tetrahydrofuran-perchloric acid system. The reac—
tion mixture was illuminated at intervals with a long wave ultra violet
light. The reduction was assumed complete when the fluorescencehad fal-
Ienibo only slightly perceptable levels.

On the neutralization of the acidic reaction mixture with ammonium
hydroxide,two solid products were found. They were readily separated by

extraction with boiling ethanol. The non soluble material (m.p. 340°)

 

showed a strong broad absorption at 3450 cm"1 (KI) indicating the pre-
sence of a hydroxyl function. Since the absorption at 1330 cm.1 remained,

this material was assumed to be the 169 Carboxylate. (XXXXIX).

 

(XXXXIX) OCH3

The ethanol soluble portion of the reactiOanroducts did not possess

this 3450 cm.1 absorption whereas the band at 1330 cm”1 remained,

indicating that both ester functions had survived the reduction process.
The reduction introduced a new asymmetric center at C-3, and this

material was shown to be homogeneous by thin layer chromatography

(Rf‘= 0.65). Consequently it remained to be determined whether the

C-3 hydrogen conformation was-t or/a .

28

Wenkert and Roychaudhuri (11) had determined that allo systems
possess an absorption at 2740 cm"1 in the infrared;, while epiallo
systems (i.e. those having s C—3/3 hydrogen, such as reserpine) do
not have this absorption. The validity of extending this point in
the present study has its basis in the work of Protiva and Jirkovsky
(24) which established that l-deaza-l-thia—isodeserpidine (i.e.
ll-desmethoxy isothiareserpine) does have an absorption at 2760 cm-1
i; the infrared . The infrared- spectrum of the ethanol soluble
reduction product does not possess this absorption,p.29 and as a con-

sequence must belong to the epiallo system. Therefore the compound

 

synthesized is the desired l-deaza-l-thia-reserpine (L).

O
OCH3 0CH3

OC H3

    

(L)

For comparison the infrared '. spectrum, p29 of the synthesized
thiareserpine is shown superimposed on the spectrum of an equimolar
mixture of 6-methoxy 3-2(piperidinoethy1) benzothiophene, methyl
cyclohexane carboxylate, and methyl 3,4,5-trimethoxy benzoate. The
ultraviolet; spectrum of the thiareserpine is compared in a similar

manner, p.30.

29

It

.890uouo~£o :a..ut|.:.o:onnoasu0ucom

AHanuooefimwuumamvnumnzxonuozuo van uuuoucom axonuoawueun.e.m flange: .oueahxonuoo maxexonoHomo
Hague: mo ousuxfiz z ooH.o s was ocqmuomoMcsanyiaumumoana mo asuuooam HmonwuwcH 058 .H .m«m

 

 

 

 

75 8w 8.3 82 83 83 82 82 8.2 83 82
. m ... u."
n. . _ . n
3...... .... ..“ ..... A
a. .. :3 .u a .u _
u. n .n "u . . u” .
.. . .. f r. u .
.MMM C u. . _
n r. 1. .1 "s“. __
J u «. g a.“ "m u u.
a L .. M.

if

 

a .f..-....--...-\
I12:

 

 

 

uoyssrmsusal quaoaaa

Extinction Coefficient

60000

50000

40000

20000

10000

30

 

 

 

 

 

 

 

 

 

 

 

200 250 300 350 400 450

Wavelength in Millimicrons

 

Pig. 2. The Ultraviolet . Spectrum of l-Deaza-l-Thia-Reserpine
and a 0.100 x 10'4 M Mixture of Methyl Cyclohexane
Carboxylate, Methyl 3,4,5-Trimethoxy Benzoate and
6-Methoxy-3-2(Piperidinoethyl) Benzothiophene ---------
in 9§1>Etfianul.

EXPERIMENTAL

All melting points are uncorrected and were determined on an
Electrothermal melting point apparatus. Infraxxll spectra were de-
termined with a Beckman model IRS-A prism spectrophotometer or a
Perkin-Elmer model 2373 grating infra red spectrophotometer, in
either chloroform solution or as potassium iodide pellets. Ultra
violet spectra were determined with a Beckman model DB, a Beckman
model DKZ-A instrument, or a Unicam SP800 recording spectrophotometer.
All ultraviolet- spectra were determined in 95% ethanol. Ozone was

generated employing a Wellsbach model T-23 ozonator.

31

 

32

Preparation of Meta-Methoxy Benzene Thiol

073803 M.W. 140 (:fi

30 SH

(XXII)

This compound was prepared by a slight modification of the pro-
cedureC described by Godt and Wann (20). With constant stirring, 113 g.
(0.810 mole) of meta-anisidine was slowly added to a mixture of 150 g.
of crushed ice and 150 ml. of conc. hydrochloric acid. The solution of
the amine hydrochloride was cooled in an ice bath and an ice cold sol—
ution of 59.4 g. (0.860 mole) of sodium nitrite in 140 ml. of water
was added at such a rate that the reaction temperature was maintained
below 5°. The diazonium solution was stored in an ice bath and added
in small portions with constant stirring during a three hour period
to a solution of 152 g. (0.947 mole) of potassium ethyl xanthate in
200 ml. of water heated to 40-450. The reaction solution was stirred
an additional half hour following the addition of diazonium solution.
The red oily layer of crude product was separated and the aqueous layer
was extracted with ether (2x200 ml.). The oil and ether extracts were
combined, washed with water (2X100 ml.) and dried over calcium chloride.
The ether was removed in a rotary evaporator at water pump pressure
to leave the meta-methoxy phenyl xanthate as a red brown liquid.

The xanthate was dissolved in 500 ml. of 952 ethanol and heated
to reflux with stirring. External heating was discontinued and the
solution was maintained at its reflux temperature by the careful ad-
dition of 177 g. (3.00 moles) of potassium hydroxide pellets through
the condenser. The stirred solution was refluxed ten hours, cooled to
room temperature, and then poured cautiously into a beaker containing
350 ml. of conc. hydrochloric acid and 500 g, of ice. The resulting
oily suspension was extracted with benzene (5X100 ml.). The combined
extracts were dried over magnesium sulfate and the solvent removed
in vacuo. The residue was vacuum distilled and the fraction boiling

from 7o-75° 2 '°r" was collected. The yield was 35.7 g. (0.255 mole,

31.61). Literature values (20) b.p. 74.5° 2 '°”", in 292 yield.

33

O 0

Preparation of Ethyl (3-keto-4-chloro) Butanoate

C6H9C103 M. W. 164 CL OC2H5

The procedure: employed in obtaining this material is essentially
that of Titus (18). A 2 liter, 3 neck flask fitted with dropping funnel,
overhead stirrer, and condenser topped with a drying tube was charged

with 24.0 g. (1.00 mole) of magnesium chips, and l. g. of mercuric

"11

chloride. By means of the dropping funnel, 50 ml. of a solution of

245 g. (2.00 moles) of ethyl chloroacetate in 200 ml. of anhydrous
ether was added to the flask. The flask was gently heated by immersion
in a water bath until the reaction had been initiated as evidenced by
the formation of a milky suspension. The water bath was removed and

the remainder of the ester solution was added at a rate sufficient to
maintain the reaction mixture under reflux. When the magnesium had

been nearly consumed, external heat was applied and the reaction
mixture allowed to reflux four hours. Finally it was hydrolyzed by
pouring it onto crushed ice with stirring. The precipitate of magnesium
salts was dissolved by slowly adding 4 N sulfuric acid until the
aqueous solution reached a pH of 5 (Hydrion B). The ethereal layer

was separated and the aqueous layer was extracted with ether (3x200 ml.).
The ethereal solutions were combined,dried over magnesium sulfate,

and filtered. The ether was removed by flash distillation. The oily
residue was vacuum distilled and the material boiling 80-900 2 torr.
was collected. The yield was 78.3 g. (0.476 mole, 47.61) 111)24 1.4460.

Literature values (27) loo-103° '2 t°rr°, nD17 1.4545.

Preparation of Ethyl 4-(meta-methoxy phenyl mercapto)-3-Keto Butanoate
C H O S M.W. 268

13 16 4 C) C)

C H 30 OC2H5

The experimental procedure» described by Titus (18) was slightly

modified to prepare this ester.

34

A 500 m1. 3 neck flask equipped with an overhead stirrer, dropping
funnel, and a thermometer extending below 180 ml. of pyridine in the
flask was supported in an ice bath. With constant stirring, 35.8 g.
(0.256 mole) of meta-methoxy benzene thiol was added in one portion.
This was followed by the dropwise addition of 41.9 g. (0.256 mole) of
ethyl (3-keto-4-chloro) butanoate at such a rate that the reaction
temperature was maintained between 25-300. The ice bath was removed
and the solution was heated to 70-800 for ten minutes on a steam
bath after which it was allowed to cool to room temperature. The pH
of the solution was adjusted to 5 (Hydrion B) by the slow addition
of 6 N hydrochloric acid, approxiametly 400 ml. being required. The
resulting oil which separated was removed and the aqueous layer was
extracted with ether (2X50 ml.). The ether extract was dried over
magnesium sulfate, filtered, and the ether was removed in a.rotary
evaporator to yield a yellow odoriferous residue. Since Titus re-
ported decomposition upon attempting to distill this material, it
was used without further purification in the suceeding ring closure

step. The yield of the crude material was 62.4 g. (0.233 mole 91.01).

Preparation of Ethyl [6-(and -4-) Methoxythianapthyl-3J) Acetate

C13H14O3S M.W. 250

 

Titus' experimental procedure: (18) was slightly modified for
the preparation of this material. A solution of 30.6 g. (0.114 mole)
of ethyl 4-(meta-methoxy phenyl mercapto)-3-keto butanoate in 200 ml.
of chlorobenzene was treated with 50 m1. of 852 ortho phosphoric
acid and 100 g. of phosphorous pentoxide. After heating the reaction
mixture at its reflux temperature for three hours, the chlorobenzene
was decanted and replaced with 200 ml. of dry benzene, and the reaction

mixture was again heated under reflux for three hours.

35

The benzene was decanted and combined with the previous chlorobenzene
solution. The aromatic solution was washed succusively with 10%

sodium bicarbonate solution (50 ml.) and water (2X50 ml.). The solvents
were removed in a rotary evaporator at 800 20 torr.. The reddish orange
oil was used without further purification to prepare the corresponding

amide. The crude yield was 25.6 g. (0.102 mole, 90.5%).

Preparation of 6-(and-4-)-Methoxy Thianapthyl 3-Acetamide
C H N0 8 M.W. 221

11 11 2 (OCH3)

I
o NH2

CH3
(XXIII)

Titus' laboratory procedure (18) was used with some modification to
obtain this amide. A 4 liter resin kettle equipped with an air driven
overhead stirrer was charged with 400 ml. of cone. ammonium hydroxide.
To this, 17.0 3. (0.0692 mole) of ethyl 6-(and-4-) methoxythianapthyl-
3-) acetate was added. The mixture was stirred for seven days at room
temperature. The resulting crude gummy amide was dissolved in 100 ml.
of boiling ethanol. Upon cooling, the amide crystallized. It was col-
lected by vacuum filtration and recrystallized from 70 m1. of hot
ethanol to yield 4.82 g. (0.0218 mole, 31.72) of pale yellow plates
m.p. 192-1930. The 4-methoxy isomer was recovered from the mother
liquor of the 6-methoxy compound as follows. The alcohol filtrate

was reduced to near dryness by evaporation in vacuo. The residue was
dissolved in a minimum amount of chloroform and placed on an alumina
column (Matheson Activated Alumina, 80-200 mesh, dried at 200° for

18 hours) measuring 3X45 cm. The column was eluted with chloroform
which was collected in 50 ml. fractions. Fractions 10 through 13
inclusive contained the 4-methoxy thianapthyl-3- acetamide. The melting
point for the 4-isomer was 199-2000, Literature values (18) for the

4 isomer 200-200.5°, for the 6 isomer l92.8-193.3°.

36

Preparation of 6-Methoxy-3-(2-Aminoethyl) Benzothiophene

CllHIBNOS M.W. 207

CHSO NH2

(XIX)

A solution of 40 ml. of borane tetrahydrofuranate (Metal Hydrides
Inc., Beverly, Mass.) was cooled to 00 under nitrogen. With constant
stirring, 1.10 g. (5.00 mmoles.) of 6-methoxy-3-thianapthy1 acetamide
was added in one portion. The tetrahydrofuran insoluable amide rapidly
went into solution with the evolution of gas. The reaction solution
was allowed to warm to room temperature and was then heated under
reflux for eight hours, after which it was set aside at room temp-
erature for sixteen hours. The solution was carefully acidified with
20 ml. of 6 I hydrochloric acid. The tetrahydrofuran was removed in a
rotary evaporator at water pump pressure under nitrogen. The aqueous
residue was cooled in an ice bath. The solution was adjusted to a pH
of 10 (Hydrion B) with 5 M sodium hydroxide and extracted with ether
(5X30 ml.). The ether extracts were combined, dried over magnesium
sulfate, and flash distilled. The oily yellow liquid remaining was
vacuum distilled to yield 0.636 g. (3.08 mmoles., 61.4%) of the de-
sired amine boiling from 13o-140° 0'3 t°"', “n25 1.5964.

The rapid reaction of the amine with atmospheric carbon dioxide made
it necessary to store it under nitrogen. A picrate derivative of the
thianapthylamine was prepared in benzene in the usual manner and
recrystallized three times from ethanol for analysis, melting point
177-178°.

Analysis; Calculated for C17HI6N4808:C, 46.78; H, 3.69; N, 12.84;

S, 7.35; O, 29.34. Foundzc, 46.60; H, 4.26; N, 12.77; S, 7.31;
0, (by difference) 29.06.

THE BICYCLIC SERIES

Preparation of cis-5,8-Diketo-l,4,5,8,9«,1047Hexahydronaptha1ene-
Ll-Carboxylic Acid
011111004 M.W. 206

 

I Vinylacrylic Acid

A sufficient quantity of this material was prepared by a slight
modification of Woodwards' procedure: (14) and combining such prep-
arations. A typical synthesis follows. With vigorous stirring, 240 g.
(2.73 moles) of malonic acid was dissolved in 520 ml. of anhydrous
pyridine. After precooling the solution to 50 by immersion in an
ice bath, 160 g. (3.50 moles) of acrolein (Matheson practical grade,
stabilized with hydroquinone) was added dropwise during a half hour
while maintaining the reaction temperature below 12°. Following the
addition of the acrolein, the reaction mixture was stirred at 0-50
for three hours, then heated to 35-400 in a water bath with stirring
for an additional five hours. The reaction mixture was then quickly
poured into 520 ml. Of 50% aqueous sulfuric acid (1:1 conc.) pre-
viously cooled to -10°. The temperature of the acid solution was main-
tained below 200 by immersing it in an isopropanol—dry ice bath.
The acid (litmus) reaction mixture, which now contained considerable
precipitate, was extracted six times with ether (3x500 m1., 3x300 ml.).
The ethereal extracts were combined and ca. 2000 m1. of ether was
removed by distillation. The residue in the distillation flask was

stoppered and set aside overnight at -78°.

37

“h

38

II The Quinone Adduct

A 3 liter, 3 neck flask equipped with condenser, overhead
stirrer, thermometer, and Dean-Stark trap was charged with 140 g.
(1.30 moles) of quinone (Eastman Yellow Label) and 600 m1. of ben-
zene. The vinylacrylic acid solution, prepared as described above
was added in one portion, and the reaction mixture was heated at its
reflux temperature with vigorous stirring. The solvent was slowly fl
removed by distillation through the Dean-Stark trap until a still I
head temperature of 600 was reached, after which the reaction mix-
ture was heated under reflux for two hours. During this time the
adduct product separated as a gray precipitate.

The solvent was then distilled until the reaction mixture had
been reduced to half of its initial volume. The reaction temperature was
rdduced to 500 and the precipitated adduct was collected by vacuum
filtration. The mother liquor was returned to the reaction vessel
and again concentrated to half of its volume by distillation and then
filtered hot. The combined product material was washed with two
liters of pentane containing 100 m1. of acetone. The crude adduct
was dissolved in 800 m1. of acetone and 1000 ml. of pentane was
added. The material was allowed to crystallize slowly overnight in
the ice box and was then recovered by vacuum filtration. The yield
was 92.4 g. (0.402 mole, 31.01 overall) of the adduct in the form
of pale yellow plates. The melting point of 215-2250 was identical
to that of Woodwards' as was the infraed Spectrum; (as a KI pellet) (14).

Preparation of cis-5p-Hydroxy-8-Keto-1,4,5,8,9q,10%-Hexahydro-
napthalene-yp-Carboxylic Acid

C11H1204 M.W. 208 ()ri

 
 

O

This material was prepared as described by Woodward (14) using

(m!)

a slightly different product isolation procedure”

39

A 4 liter resin kettle immersed in an ice bath and equipped with
overhead stirrer was charged with 700 ml. of water and 107 g.
(0.522 mole) of the diketo adduct (XXIV). A thin layer of ethyl
acetate was maintained over the vigorously stirred slurry to con-
trol foaming while a solution of 43.8 g. (0.522 mole) of sodium
bicarbonate in 700 m1. of water was slowly added. When the adduct
had dissolved, 12.5 g. (0.330 mole) of sodium borohydride dissolved
in 40 m1. of water was added in small portions during a period of
15 minutes. The reaction mixture was stirred an additional 15 min-
utes, then acidified (congo red) with 202 (4:1 conc.) of aqueous
sulfuric acid. The acidic solution was allowed to stir an additional
30 minutes at 0°. The product was separated by vacuum filtration
and air dried to yield 87.0 g. (0.418 mole, 79.81) of material
melting at 175-1780. Reported melting point (14) 179-180°. This
hydroxy ketone was identical to that reported by Woodward by

infra ned‘ spectrum in a KI pellet.

Preparation of 8-Keto-l,4,5,8,9a,IDs-Hexahydronapthalene-hp-
Carboxylic acid 1,5-Lactone
01,31003 M.W. 190

 

(XXVIII)

This compound was prepared following the procedure. of Woodward (14).
A vigorously stirred mixture of 20.0 3. (0.0960 mole) of the hydroxy
ketone (XXVI), 20 m1. of acetic anhydride, 4 g. of anhydrous sodium
acetate, and 400 m1. of anhydrous benzene was heated at its reflux
temperature for one hour. Heating was discontinued and the reaction
mixture was cooled by immersion in an ice bath. Approxiametly 200 m1.

of ethyl acetate along with 50 ml. of ice cold water was added.

40

The organic layer was separated and extracted with 50 ml. of 10%
sodium bicarbonate solution. The aqueous phases were combined and
washed with ethyl acetate (3X30 ml.). The ethyl acetate solutions
were combined, washed with saturated sodium chloride solution (2x25 m1.)
and dried over 10 g. of anhydrous sodium sulfate. The filtrate was
concentrated to 20 ml., 10 m1. of ether was added, and the mixture
was set aside for several hours in the cold. The crystalline lactone
which separated was recovered by filtration, washed with ether and
air dried to obtain a pale yellow solid weighing 11.0 3. (0.0579 mole,
60.4%) which melted at 121°. The melting point was not reported by
Woodward but the infra-eds. spectrum in K1 was identical to that
reported (14).

Preparation of 5f-Hydroxy-l,4;5,8,9d,lou-Hexahydro-yB-Napthoic acid
1,8-Lactate

011111203 M.W. 192

 

(XXIX)

The material was prepared by a slight modification of Woodwards'
method (14). To 30.7 g. (0.182 mole) of the 1,5 lactone (XXVIII)
suspended in 400 m1. of anhydrous isopropanol, 37.0 g. (0.181 mole)
of aluminum isopropoxide (Eastman Yellow Label) was added. The reaction
mixture was stirred and heated to remove the isopropanol by distillation
which was checked for its acetone content with aqueous 2,4-dinitro
phenylhydrazine hydrochloride solution (28). From time to time fresh
isopropanol was added to maintain the reaction level approxiametly
constant. After two hours the distillate failed to give a positive

acetone reaction.

41

The reaction solution was heated at its reflux temperature for an
additional 45 minutes. The solvent was removed in a rotary evaporator.
The residue was dissolved in 400 ml. of ethyl acetate and cooled to
0°. It was added to 600 m1. of a solution of 500 g. of potassium
sodium tartrate and 37 g. of sodium bicarbonate in water. The milky
aqueous phase was separated and extracted with ethyl acetate (3X100 m1. ).
The ethyl acetate solutions were combined and washed with saturated
sodium bicarbonate solution (3X50 ml.). Finally the solution was dried
with 50 g. of anhydrous sodium sulfate and filtered. The solvent was
removed in a rotary evaporator, leaving a viscous yellow colored
residual oil which solidified when set aside. It was triturated with
ether to give a nearly pure white solid lactone, melting point
116-1180, reported (14) 120-122°. The yield was 16.4 3. (0.0863 mole,
53.41). The infrared spectrum (in chloroform) was identical to that
reported by Woodward(l4).

Preparation of 24-8romo-l,2,3,4,5,8-cis-9d,10d-0ctahydro-6-ene-yfi‘
(3,5-epoxy)-Napthoic Acid 1,8-Lactone

C11H1103Br M.W. 271

 

B's

(XXX)

The preparation described by Velluz (12) was modified slightly for

the preparation of this compound. An 11.5 g. (0.600 mole) quantity

of the previously prepared 1,8 lactone (XXIX) was dissolved by

heating it in 120 m1. of tert-butanol, on the steam bath. The steam
bath was removed and the hot solution was allowed to cool to room
temperature, in one portion 11.1 3. (0.0623 mole) of N-bromo succinimide

was added to the solution.

4 ti

 

42

The reaction solution was stirred for ten minutes during which most of
the N-bromo succinimide had dissolved. The mixture was then evaporated
to dryness in vacuo. The residue was suspended in 100 ml. of water

and filtered. The solid bromo compound remaining was dried to obtain
14.2 3. (0.0522 mole, 87.2%) of pale tan colored crystalline material
melting at 120-122°. Reported (14) 120-124°. This material was iden-
tical by infrared spectrum (in chloroform) to that reported by
Woodward for the bromo lactone (XXX) (14).

Preparation of 2*PMethoxy-l,2,3,4,5,8-cis-9d,10d-6-ene-Octahydro-yfl-
(3,5-epoxy)-Napthoic Acid 1,8-Lactone

0121-11404 M.W. 222

 

OCH 3

(XXVII)

This material was prepared as described by Woodward (14). A 5.59 3.
(0.0206 mole) quantity of the bromo lactone (XXX) was dissolved in
80 ml. of anhydrous methanol, previously dried over magnesium and
freshly distilled. A 50 ml. volume of sodium methoxide solution,pre-
pared by dissolving 0.49 g. (0.021 mole) of sodium in 50 ml. of
anhydrous methanol, was added in one portion. The reaction mixture was
set aside at room temperature for two hours and thirty minutes. The
basic solution was neutralized with glacial acetic acid and concen-
trated in vacuo to ca. 10 ml.. The reaction mixture was diluted with
50 m1. of methylene chloride and 20 m1. of ice cold water. The aqueous
layer was separated and the organic layer was washed sucessively with
15 m1. of 10% aqueous sodium bicarbonate, 10 m1. of saturated sodium

chloride, and dried over 3 g. of anhydrous sodium sulfate.

43

The salt was removed by filtration and the filtrate was concentrated
to approxiametely 5 ml. in vacuo. The methoxy lactone (XXVII)
crystallized from the solution when it was set aside at room temper-
ature to yield 4.48 3. (0.0202 mole, 98.02). The melting point was
98-1000. The reported melting point was 100-1020 (14). The infra-
redspectrumuwas identical to that reported by Woodward as a KI
pellet (14).

Preparation of 2K-Methoxy-6A-Bromo-ZG-Hydroxy-cis-Perhydro-(3,5-epoxy)-
beapthoic Acid 1,8-Lactone Ear
C H 0 Br M.W. 318 3

12 15 5 +1()

  

00H3
(XXXIV)

Woodwards' procedure. (14) was used with slight modification
for the preparation of this bromohydrin. A 4.00 3. (0.0180 mole)
quantity of the methoxy lactone (XXVII) was dissolved by heating it
do 80° in 30 ml. of water containing 8 m1. of l N sulfuric acid.
After cooling to room temperature, the acidic solution was reheated
to 60-70° by immersion in an oil bath and 3.21 g. (0.0180 mole) of
N-bromo succinimide (freshly recrystallized from glacial acetic acid)
was added with constant stirring during a perion of ten minutes. The
reaction temperature was then increased to 80-900 and held there for
90 minutes. Heating was discontinued and the reaction was allowed to
cool to room temperature. Approxiametely l.g. of sodium bisulfite
was added'to the aqueous reaction mixture and it was continuously
extracted with chloroform for four hours. The chloroform was re-

moved in vacuo.

44

The residual yellow colored oil (identical to Woodwards' bromohydrin

as determined by infra red spectrum in chloroform) was utilized in the r
next step of the synthetic sequence without further purification.

The yield of the crude bromohydrin was 7.32 3. (0.0234 mole, 130%).
Woodward by a similar proceedure reports (14) a 1291 yield of this
compound. Infrared: analysis showed the crude bromohydrin was con-

taminated with succinimide.

Preparation of Qd-Methoxy-6u-Bromo-7 -Keto-cis-Perhydro-(3,5-epoxy)-
l-Napthoic Acid 1,8-Lactone

012H13053r M.W. 316 Sr

 

00H3

(XXXVI)

The method reported by Velluz (12) was used without modification.
A suspension of 15.5 3. (0.0488 mole) of the previuosly prepared
bromohydrin (XXXIV) in 200 ml. of l,2-dichloroethane was treated
with 35 m1. of glacial acetic acid. An oxidizing medium consisting
of 14 g. of chromium trioxide, 14 m1. of water, 12 m1. of ortho-
phophoric acid and 25 m1. of acetic acid was prepared. A 30 ml.
quantity of this oxidizing medium was added to the bromohydrin
contained in a flask immersed in an ice bath at such a rate that
the reaction temperature was held below 300 during the addition of
the oxidizing agent. With occasional swirling of the reaction flask
10 ml. of fresh oxidizing agent was added in ten minutes. The re-
mainder of the oxidizing agent was added after another 30 minutes.
The reaction flask was then removed from the ice bath, and stored
in the cold until frozen. After thawing, the reaction mixture was
filtered. The collected material was washed on the filter with
acetic acid followed by ether.

45

The nearly pure white crystalline mass of the bromo ketone, weighing
8.74 g. (0.0296 mole, 56.52) had a melting point of 172-173°. It

was reported as 166-1670 (14). This bromo ketone was shown to be
identical to the material prepared by Woodward by infraradfl spectrum
in KI.

Preparation of Zq-Methoxy-3B-Hydroxy-5-ene-7-Keto-1,2,3,4,7,8-cis-
9d-lOd-0ctahydro-yeNapthoic Acid

C12H1506 M.W. 240

 

(XXXVII)

This compound was prepared exactly as described by Woodward (14).
A 1 liter, 3 neck flask equipped with an overhead stirrer, a widebore
funnel, and a thermometer extending to the solvent, was charged with
1.87 g. (5.92 mmoles) of the bromo ketone (XXXVI) and 500 m1. of
freshly distilled glacial acetic acid. The ketone was dissolved by
heating the solvent to 70°. External heating was discontinued and
the acetic acid solution chilled to 180 by immersion in an ice bath.
The ice bath was removed and 7.5 g. of zinc dust (200 mesh) previously
cooled to 0°, was added rapidly in one portion. The acid solution was
stirred very vigorously for 90 seconds and immediately filtered
through a medium porosity sintered glass funnel. The filtrate was
concentrated to dryness by distillation at water pump pressure,
heat being supplied by a water bath heated to 80°. The residue was
dissolved: in 25 ml. of water and made basic (Hydrion B) with solid
sodium bicarbonate. The basic solution was extracted continuously

with ether for three hours.

'1

46

The ether extract was discarded, and the aqueous phase was acidified
(congo red) with conc. hydrochloric acid, saturated with solid sodium
chloride, and re-extracted with ether for 36 hours. Concentration of
the extract in vacuo produced a white crystalline solid which was
triturated with a small amount of acetone to yield the acid (XXXVII)
as a pure white solid weighing 1.12 g. (4.67 mmoles, 79.02). This
material had a melting point of 204-2060, while the reported melting
point was 204-2060 (14). This material was shown to be identical

to that prepared by Woodward by infrared spectrum in XI. The ultra-
violet spectrum showed akmax 227 ty‘,€'10,100, the reportedlmax
at 228wr5@10,000 (14).

Preparation of 2fi-Methoxy-Qfl-Hydroxy-5-ene-7-Keto-1,2,3,4,7,8-cis-
91-10d-Octahydro-l-Napthoic Acid Methyl Ester

C13HI7O6 M.W. 254

 

The proceduree given by Woodward (14) was slightly modified for the
preparation of this ester. A 60 ml. volume of hot anhydrous dioxane was
used to dissolve 1.20 g. (5.00 mmoles) of the previously prepared
acid (XXXVII). The dioxane solution was cooled to 130 by immersion
in an ice bath and 55 m1. of an ethereal solution of diazomethane
(0.136 mmoles./m1., 7.47 mmoles. total) was added to the acid solution
in one portion. The ice bath was removed hnd the reaction mixture was
allowed to warm to room temperature. The solvent was immediately
removed in a rotary evaporator, holding the bath temperature at 40°.
The oily residue was triturated with 5 ml. of ether containing a
few drops of acetone, whereupon the ester crystallized. The crystalline
ester was collected by filtration and washed with ether. The ester
was recrystallized from a small volume of ether/acetone 0190/10) to

yield the colorless ester weighing 1.25 g. (4.92 mmoles., 98.5%).

47

This material melted at 130-1310 compared to the reported melting
point of l34-l36°(14). The ester had an identical infrared. spectrum
compared to that reported by Woodward (14).

Preparation of 3,4,5-Trimethoxy Benzoyl Chloride

c a 0 c1 M.W. 230 ()
(3143;3’ lllii' (:L
CHSO

10 11 4
00143

(XVIII-C)

A mixture of 21.2 g. (0.100 mole) of 3,4,5-trimethoxy benzoic
acid and 20.8 g. (0.100 mole) of phosphorous pentachloride was meta-
thesized and set aside until the vigorous evolution of hydrogen chloride
had ceased. The reaction mixture was heated on a steam bath for 1 hour.
The phosphorous oxychloride was removed at water pump pressure. The
resulting violet colored acid chloride was recrystallized from 300 m1.
of 60-900 petroleum ether/pentane/benzene (8/1/1) to give 21.5 g.
(0.0934 mole, 93.4%) of colorless needles of the title compound,
melting at 81-820. Literature value (29) 81-820.

Preparation of Zfl-Methoxy-3fl-(3',4',5'-Trimethoxybenzy1oxy)-5-ene-7-
Keto-1,2,3,4,7,8-cis-9fl-10£-0ctahydro-yB-Napthoic Acid
Methyl Ester
C23H28°9 C)
M.W. 448

    

48

The method of Weichet, Pelz, and Blaha (26) was employed with
minor differences to obtain this ester. A solution of 0.500 g.
(1.98 mmoles.) of the previously prepared ester (XXXVIII) was
formed in 15 ml. of anhydrous pyridine (dried over barium oxide and
distilled from barium oxide). The pyridine solution was cooled to 50
by immersion in an ice bath and added to a chilled solution of
0.685 g, (2.96 mmoles.) of 3,4,5-trimethoxy benzoyl chloride in
12 ml. of benzene (previously dried over sodium wire and distilled).
The solution was allowed to warm to room temperature and was then set
aside undisturbed for 20 hours. The reaction mixture was hydrolyzed
with 8 ml. of water and stirred for a half hour. A 20 g. quantity of
ice was added and the pH of the reaction mixture was adjusted to 5
(Hydroin B) with cone. hydrochloric acid. The acidic solution was
extracted with chloroform (3X40 ml.). The chloroform solution of the
diester was washed with saturated sodium bicarbonate solution (2X20 ml.)
and saturated sodium chloride (2X10 ml.). The combined chloroform extracts
were taken to dryness in a rotary evaporator at reduced pressure. The
residue was suspended in 10 ml. of methanol and filtered. The methanol
filtrate was treated with 5 ml. of ether and set aside in the cold for
a day to allow crystallization of the product. This was collected to
yield 0.701 g. (1.56 mmoles., 79.02) of pale tan colored crystalline
material melting at 110-1140. The reported melting point was 107-1150
(26). The ultravioleta spectrum showed absorbtion maxima at 268 my;
€‘10,900; 217 mpqe'38,300. The infraredl spectrum is shown on page 57.

THE TETRACYCLIC AND PENTACYCLIC SERIES

Preparation of Qt) hD-Methoxycarbonyl-ZI-Methoxy-36-(3',4',5'-Tri-
methoxybenzyloxy)-53-Formy1-69~Cyclohexy1 Acetic Acid

c22H28011 M.W. 468
COOH
.
356...: *1‘
OCH3
(XXXXII)

A 224 mg. (0.500 mmole.) quantity of the diester (XXXIX) was
dissolved in 10 ml. of anhydrous methylene chloride contained in

the reaction vessel attached to the apparatus shown.

C®Q Tygon Coupl ings\

 

 

OzoneI _ $2 2 Exit Gases
, _ ._ W

1

24/40 3

 

 

 

 

 

 

 

52 otassium

Iodide Solution

I!
a u\\ Indicating in Water

 

 

 

 

Drierite"
-50-+50o Thermometer

~‘\\“Ciass "c" Frit
3 ’ CHZCl2 Solution of the Diester
\

Dewar Flask

 

 

 

 

 

 

Dry Ice-Isopropanol
49 '

50

The isopropanol bath temperature was lowered to -300 by adding
small pieces of dry ice to the bath. Ozone («:11 V/V concentration
of O3 in 02;
chloride solution, the exit gases being passed into 30 ml. of a 5%

flow rate ~a0.002 ft3/min.) was passed into the methylene

aqueous potassium iodide solution. After five min. and twenty sec.
the iodide solution changed rapidly from colorless to yellow brown
and the flow of 0 -0 gas was terminated. This corresponds, on the

3 2
basis of gas flow time, to an uptake of ~'470 mmoles. of 0 The

solution of ozonide at -300 was purged of ozone by passingBa stream of
dry nitrogen through it for ten min.. The solution was then heated,
under reflux, for 45 min. in an atmosphere of nitrogen with 2 m1. of
water containing 0.01 g. of hydroquinone. After cooling the solution
to room temperature, the aqueous layer was separated, washed with
methylene chloride (2X5 m1.) and the combined methylene chloride
solutions were dried by filtering them through a bed of anhydrous
sodium sulfate. The aldehyde solution was treated with a slight excess
of 2,4-dinitro phenylhydrazine in 20 m1. of isopropanol. After
standing several hours at room temperature the solid material was
collected by vacuum filtration and recrystallized three times from
ethanol to obtain brilliant red colored needles melting at 128-1310.
Analysis: Calculated for C H O N ' C, 51.88; H, 4.97; NO, 43.18

28 32 14 4'
Found: C, 51.60; H, 5.10; N0, (by difference) 43.30.

Preparation of fit) l-Deaza-l-Thia-Z,3-seco-3-0xo Reserpine

CH30 DUO N

  

51

The aldehyde described above and obtained from 224 mg. (0.500
mmoles.) of the diester (XXXXIX) in the original methylene chloride.
solution was immediately cooled to 00 by immersion in an ice bath
and treated with a slight excess of ethereal diazomethane. The ice
bath was removed and the reaction mixture was set aside for 10 min.
at room temperature, and then concentrated to % its initial volume
in an atmosphere of nitrogen in a rotary evaporator. The concentrated
solution of the triester was chilled to 00 in an ice bath and a solution
of 104 mg. (0.503 mmole.) of 6-methoxy-3-(2-aminoethy1) benzothiophene
(XIX) in 1.4 ml. of anhydrous benzene was added in one portion. The
reaction solution which took on a yellow orange coloration was set
aside for ten min. and 20 ml. of anhydrous methanol was added. After
recooling the reaction mixture to 0°, a solution of 19.0 mg. (0.500
mmole.) of sodium borohydride in 2 m1. of anhydrous methanol was added
to the mixture, during a period of five minutes. Acetic acid (2 drops)
was added to the mixture and all solvents were removed, first in an
atmosphere of nitrogen in a rotary evaporator and finally with an oil
pump at 0.01 torr.. The dry residue was triturated with ether, col-
lected by vacuum filtration and washed well with ether. In order to
re-esterify and alcohol which may have formed during the reduction
with sodium borohydride, the throughly dry lactam was dissolved in
15 ml. of anhydrous pyridine and treated with 115 mg. (0.500 mmole.)
of solid 3,4,5-trimethoxy benzoyl chloride. After being set aside at
room temperature in an atmosphere of nitrogen for four days, the
pyridine was removed in a rotary evaporator under nitrogen. The
residue was dissolved in 10 ml. of chloroform and the chloroform
solution was extracted with 5 ml. of 51 aqueous sodium bicarbonate.
After drying over anhydrous sodium sulfate, the chloroform solution
was heated with a small amount of Norit—A and filtered. Ether was added
to precipitate the lactam which was collected by vacuum filtration

and dried over anhydrous calcium sulfate in a vacuum desiccator.

52

The yield of pale tan powder product was 240 mg. (0.374 mmole., 74.82)
The melting point was 145-1480 dec. (sealed capillary). For analysis
the material was purified by repitative precipitation from ethyl
acetate by adding ether.
Analysis Calculated for C33H39010NS: C, 61.76; H, 6.12; N, 2.18
S, 5.00: Found: C, 61.51; H, 6.07; N, 3.00; 8,4.67.

Preparation of fit) l-Deaza-l-Thia-3,4-Dehydro—Reserpine Perchlorate

C33H38013NSC1 M.W. 723

 

00H3

3
(XXXXVIII~B) OCH3 OC'---|30C‘n|3

A 100 mg. quantity of the lactam (XXXXVI) was dissolved in 2 m1.
of freshly distilled phosphorous oxychloride. The reaction mixture
was heated in an atmosphere of nitrogen by immersion in an oil bath
previously heated to 65°. After 45 min. the phosphorous oxychloride
was distilled at water aspirator pressure. Finally the reaction mix-
ture was evaporated to dryness at 0.01 torr. using an oil pump. The
oil bath was removed and the glassy residue in the reaction vessel
was dissolved in 4 ml. of acetone. The acetone solution of the flor-
escent immonium chloride was treated with 3.5 ml. of 0.1 N perchloric
acid. The semi solid which formed was not removed but rather the
entire suspension was concentrated in.a rotary evaporator under ni-
trogen to remove the acetone. The aqueous suspension of the immonium
perchlorate was extracted (3X5 ml.) with chloroform. The combined
chloroform extracts were dried over anhydrous sodium sulfate and

taken to dryness in a rotary evaporator under nitrogen.

53

The residue was triturated with ether, collected by vacuum filtration
and washed with ether to obtain 101 mg. (0.140 mmole., 89.82) of the
immonium perchlorate melting at 178-1860 dec. (sealed capillary).

For analysis the crude product was recrystallized from ethanol/acetone
5/1 to obtain an analytical sample in the form of fine orange colored
needles melting at 203-2050 dec. (sealed capillary).

33H38013NSC1: C, 54.80; H, 5.29; N, 1.94;

S, 4.43; C10, 33.52; Found: C, 54.28; H, 5.53; N, 2.07; S, 4.64:

010, (by difference) 33.58.

Analysis Calculated for C

Preparation of Qt) l-Deaza-l-Thia-Reserpine

C33H3909NS M.W. 626

 

(L) OCH3

A solution of 111 mg. (0.153 mmole.) of the immonium perchlorate
(XXXXVIII-B) in 5 m1. of acetone was placed in a 25 m1. 3 neck flask
fitted with nitrogen inlet, reflux condenser and mercury trap. To the
acetone solution of the perchlorate, 1.5 m1. of 0.7 N perchloric acid
was added. Sufficient tetrahydrofuran was added to the mixture to ef-
fect a clear solution. The reaction vessel was purged with nitrogen
and placed in an oil bath previously heated to 70°. When the reaction
mixture commenced refluxing, it was stirred vigorously by means of a

magnetic stirrer and 0.15 g. of zinc dust was added to the mixture.

 

54

After 10 min., a second 0.15 g. portion of ziné dust was added to
the mixture followed by a third similar quantity after another 10
minute interval. The reaction was considered complete after stirring
the mixture an additional 10 min.. At this point the reaction mixture
showed only a very slight fluorescence.The oil bath was removed and
the reaction mixture was allowed to cool to room temperature, and the
zinc was removed by filtration. The filtrate was made basic (pH 9,
Hydrion B) with concentrated ammonium hydroxide and 10 m1. of chloro-
form was added to the basic filtrate. The layers were separated and
the aqueous layer was extracted (2X5 ml.) with chloroform. The com-
bined chloroform extracts were filtered through anhydrous sodium
sulfate. The solvents were removed in a rotary evaporator under ni-
trogen. The residue (71 mg.) was triturated with 10 ml. of boiling
ethanol and filtered while hot. The residue collected on the filter u
( 17 mg., m.p. 340o dec., sealed capillary) was recrystallized from
acetone/ether (1/5) and showed a broad absorption at 3450 cm-1(KI).
This material was assumed to be the 16,6 carboxylic acid, although no
further characterization, was carried out.

The ethanol filtrate obtained above was concentrated to 3 ml.
in a rotary evaporator under nitrogen and 15 m1. of ether was added
dropwise with constant stirring. The precipinabe(54 mg., 0.086 mmole.
562) of l-deaza—l-thia reserpine had a melting point of 182-1890 dec.
(sealed capillary). It was shown to homogeneous by thin layer chroma-
tography on Woelm activity 11 alumina eluting with chloroform-methanol-
benzene (10:3:1), Rf: 0.65. For analysis the product was recrystallized
three times from a minimum amount of ethanol/ether (9/1) to obtain
21 mg. of pure thiareserpine melting at 188-1910 (sealed capillary).
Analysis Calculated for C H 0 NS: C, 63.34; H, 6.28; N, 2.24; SO, 28.13

33 39 9
Found: C, 63.08; H, 6.75; N, 2.53; SO (by difference) 27.64.

55

The Synthetic Spectrum

The materials used in the synthetic spectrum, p. 29 were prepared in

the following manner.

Preparation of 6-Methoxy-3-(2-Piperidinoethy1) Benzothiophene

C16H210NS M.W. 275

(314:5C)

A 0.298 g. (1.00 mmole.) quantity of the hydrochloride salt of
this compound as prepared by Titus (18) was dissolved in 5 ml. of
water in a separatory funnel. To this 10 ml. of ether was added and
then 5 m1. of l N sodium hydroxide. The ether layer was separated and
the aqueous layer was extracted (2X5 ml.) with ether. The ethereal
solutions were combined, dried over anhydrous soudium sulfate and
filtered. The filtrate was evaporated first in a rotary evaporator
and finally with an oil pump at 0.01 torr. to obtain the pure amine

as a pale yellow colored oil.

Preparation of Methyl 3,4,5-Trimethoxy Benzoate and Methyl Cyclo-
hexane Carboxylate C)

9
CH30 0 OCH5 OCH3
CH30
OCH3

C11111405 PM" 216 C8H1402 M.W. 142

The individual acids (1.00 mmole. each in anhydrous ether) of the

corresponding esters were treated with a 10% excess of ethereal diazo-

methane.

56

The ether was initially removed in a rotary evaporator under nitro-
gen and finally with an oil pump at 0.01 torr. to obtain the methyl
cyclohexane carboxylate and the methyl 3,4,5-trimethoxy benzoate as
pure colorless materials.

These three reagents were combined in chloroform and diluted to
0.100 M. for the determination of the infraredi spectrum. The ultra-
violet spectrum was determined by diluting lO/ul. of the chloroform
solution to give a concentration of all species of 10"5 M in 95%
ethanol. These spectra are reported on pp 29-30 and shown superimposed

on the spectra of the thiareserpine.

57

.¢.m.m.Huouoxunuoconmnaxonaucenzxosumagma:.m ..q ..mV11Wuhxonuoztxm mo asuuoemm mousuwcH 039 .m .wfim

HI

20

com

ooo~

.Buowonofino o« woumm Hanuoz vwo< owonumm2u\mnouvhnmuooiaoa.smim«0nw.h

meme

coed

coca

coma

ooo~

Gena

coon

acmm

 

 

_

 

 

 

 

_

 

 

 

 

 

 

 

_

_

_

 

 

X

 

 

ON

as

00

cm

cod

uorssrmsusal :uaoaad

58

HI

20

com

oooH

ooNH

coed

Enomouo~no a“ cued oauoo< me030ao>otu©u~%EuomlnwlAmxo
iamnconaxonuoswueim .3 ..mvucnuhxosuozl1Nu~>sonumo>xonuozt3 Aflv no 8950on .. vouwumcu 05. J .wwm

coca

coma

ooo~

comm

ooom

comm

 

 

 

 

_

 

_

 

 

 

 

_

 

 

_

.llll om

 

IIHuooH

uorssrmsuea; queoaaa

59

o

005

mafimm newsman econmowsuouaem AfihnueocHE¢vNumuhxoauozuo mo asuuoomm mmoumumcH 05H .n .wflm

cow

oooH

coma

coma

coon coma ooom

comm ooom

 

 

 

 

_

 

I ,

 

ON

cod

 

 

 

uoIsstsusJL nuaolaa

60

H...

to

Epomouofino a“ ocwmuemomloxOTMTouoeum.Nisanfiuuteueoaua Auo mo Enuuuomm «voueumcH one .o .wwm
comm

cow

oooH

OONH

ocoa

coca

coma

ocoN

OOmN

coon

 

 

_

A

 

_

 

#

 

A

 

.flflflflfioo~

 

1101 88111181121}, 31.183135

61

Showouoano cw oueuoflnouom ecamuomomloummsoaueamtewnfinausueoata Aav mo Bauuoomm -poumumcm one .5 .wwm
00mm

HI.

:0

com

oooH

coma

coma

ooofi

coma

ooom

oomN

ooom

 

 

_

 

 

 

_

 

_

 

J.

 

 

ON

co

co

ow

OOH

00188111181191}, JUBOJBJ

62

20

00ml

oooH

oo~H.

Euomouoano :« ecflmuomflwsanauatensoaud mo Bauuoomm MvmumuwaH oak .m .wfim

coed

coed

coma

ooo~

OOnN

ooom

comm

 

 

_

_

 

 

 

 

 

ON

co

co

ow

ooH

111181191], 30331.8(]

uotss

63

:0 com coca coma

HM cw scamuoeoalsunauaususonua mo Enuuoemm .vmumuwcH onH .m .wam

coca

coed

coma

coo~

comm

oocm

comm

 

_ _

camamm .w1\oo¢

 

 

 

_

ofimasa.ws\oo~

 

_

_

cw

co

co

ow

 

 

00H

uoIssTmsusJL nuaozaa

Extinction Coefficient

60000

50000

40000

30000

20000

10000

64

 

 

 

j
200 250 300 350 400 450

 

 

 

Wavelength in Millimicrons

Fig. 10 The Ultraviolet; Spectrum of (t) l-Deaza—l-Thia-2,3-
seco-3-0xo Reserpine in 952 Ethanol

Extinction Coefficient

60000

50000

40000

30000

20000 I

10000

65

 

 

 

 

J I |

200 250 300 350 400 450

 

Wavelength in Millimicrons

Fig. 11. The Ultraviolet- Spectrum of (19 l-Deaza-l-Thia-3,4-
Dehydro Reserpine Perchlorate

BIBLIOGRAPHY

1. M. Gordon in ”Medicinal Chemistry," 2nd. ed., A. Burger, Ed.,
Interscience Publishers, New York, N. Y., 1960, pp 397-398.

2. J. M. Mueller, E. Schlittler, and H. J. Bein, Experientia,
‘8, 338 (1952).

J. F. Kerwin, C. D. Balaut, and G. E. Ullyot in "Medicinal Chemistry,"
ibid., p 563.

L. Dorfman et. a1., Helv. Chem. Acta, 31, 59 (1954).

. Dorfman et. a1., Experientia, 2, 368 (1953).

. B. Woodward, Nature, 192, 155 (1948).

A. Diassi et. al., J. Am. Chem..Soc., 11, 4687 (1955).

. F. Huebner and E. Wenkert, J. Am. Chem. Soc., 11, 4180 (1955).
. . E. Aldrich et. al., J. Am. Chem. Soc., 81, 2485 (1959).

10. op. cit., p 2481.

11. E. Wenkert and D. K. Roychaudhuri, J. Am.Chem. Soc., IQ, 6417 (1956).

12. L. Velluz et. a1., Bull. Chem. Francd, 673 (1958).

13. R. B. Woodward et. a1., J. Am.Chem. Soc., 18, 2023, 2657 (1956).

14. R. B. Woodward et. a1., Tet., g, 1 (1958).

15. E. Schlittler in "The Alkaloids," Vol. VIII, R. H. F. Manske, Ed.,
-Academic Press Inc., New York, N. Y., 1965, chap. 13.

16. K. Fukui et. a1., J. Chem. Phys., 22, 1433 (1954).

17. J. C. Patel, J. Sci. Ind. Research, (India), 168, 370 (1957).
Chem. Abs., 22, 17031 (1958).

w

0

\OQNO‘UIb
woywr‘

in

18. R. L. Titus, Doctoral Dissertation, Michigan State University, (1964).
19. H. C. Brown and P. Heine, J. Am. Chem. Soc., 86, 3566 (1964).
20. H. C. Godt and R. E. Wann, J. Org. Chem., 26, 4050 (1961).

21. W. Herz, J. Am. Chem. Soc., 13, 4999 (1950).

22. R. Hoffmann and R. B. Woodward, J. Am. Chem. Soc., 81 2047 (1965).

23. "Modern Synthetic Reactions," by H. 0. House, W. A. Benjamin Co.,
New York, N. Y., 1965, p 139.

24. I. Jirkovsky and M. Protiva, Coll. Czech. Chem. Commun.,
38, 2582 (1963).

66

25.

26.

27.
28.

29.

30.

67

M. Protiva and I. Jirkovsky, Czech. Pat. #108,565, in Chem. Abs.,
_6_9, 8079 (1964).

J. Weichet, K.Pelz and L. Blaha, Coll. Czech. Chem. Commun.,
29, 1529 (1961).

J. F. Hamel, Bull. Chem. Soc. France, 22, 390 (1921).

A. L. Wilds in "Organic Reactions," Vol. 2, John Wiley and Sons Inc.,
New York, N. Y., p 200.

"Dictionary of Organic Compounds,” Oxford University Press, New York,
N.Y., 1965.
V. B. Schatz in "Medicinal Chemistry" ibid. p. 80.

 

”11117), E11 M!!! Millfflfli'tilfiwfijlfilfl'“