From DEPARTMENT OF BIOSCIENCES AT NOVUM
Karolinska Institutet, Stockholm, Sweden
SYNTHESIS OF INDOLE AND
PYRROLO OR IMIDAZOLO
All previously published papers have been reproduced with permission from the
Published and printed by Karolinska University Press
Box 200, SE-171 77 Stockholm, Sweden
© Stanley Rehn, 2004
The focus of this thesis is on the synthesis of oxindole- and indole-derivatives
incorporating pyrrolidins, pyrroles or imidazoles moieties.
Pyrrolidino-2-spiro-3’-oxindole derivatives have been prepared in high yielding
three-component reactions between isatin, α-amino acid derivatives, and suitable
dipolarophiles. Condensation between isatin and an α-amino acid yielded a cyclic
intermediate, an oxazolidinone, which decarboxylate to give a 1,3-dipolar species, an
azomethine ylide, which have been reacted with several dipolarophiles such as N-
benzylmaleimide and methyl acrylate. Both N-substituted and N-unsubstituted α-
amino acids have been used as the amine component.
3-Methyleneoxindole acetic acid ethyl ester was reacted with p-
toluenesulfonylmethyl isocyanide (TosMIC) under basic conditions which gave (in a
high yield) a colourless product. Two possible structures could be deduced from the
analytical data, a pyrroloquinolone and an isomeric ß-carboline. To clarify which one
of the alternatives that was actually formed from the TosMIC reaction both the ß-
carboline and the pyrroloquinolone were synthesised. The ß-carboline was obtained
when 3-ethoxycarbonylmethyl-1H-indole-2-carboxylic acid ethyl ester was treated
with a tosylimine. An alternative synthesis of the pyrroloquinolone was performed
via a reduction of a 2,3,4-trisubstituted pyrrole obtained in turn by treatment of a
vinyl sulfone with ethyl isocyanoacetate under basic conditions. This molecule (the
pyrroloquinolone), obtained in a low yield by a multistep procedure, proved to be
identical with the product obtained easily via the TosMIC route.
The reaction between 3-aminocrotonates and 3-acetonylideneoxindole in refluxing
toluene resulted in 2-pyrrolo-3’-yloxindoles in high yields, (around 90 %). At room
temperature the 2-pyrrolo-3’-yloxindoles exist as a mixture of keto-enol tautomers.
Treatment with POCl3 yielded the corresponding 2-chloro-3-pyrrolyl indole, which
gave a pyrrolo annulated indolopyrane upon basic hydrolysis of the ester function of
the methyl ester.
3-Imidazolylindoles were synthesised in good yields from the corresponding
benzylimine and TosMIC. Treatment of cyclohexanone benzylimine with α-
chloroacrylonitrile yielded, after expulsion of HCN by refluxing in ethanol, 1-benzyl-
4,5,6,7-tetrahydroindole. Formylation and benzylimine formation followed by
treatment with TosMIC furnished the desired 2-imidazolyltetrahydroindole.
Keywords: isatin, three-component reaction, α-amino acid, azomethine ylide,
pyrrolidino-3-spiro-3’-oxindole derivatives, 3-methyleneoxindole derivatives,
pyrroloquinolone, TosMIC, β-carboline, tosylimine, pyrrole, keto-enol tautomerism,
indolopyran-2-one, imidazole, benzylimine, tetrahydroindole.
List of publications
The thesis is based on the following papers, referred to in the text by the Roman
I. The Three-Component Reaction between Isatin, α-Amino Acids, and
Rehn, S.; Bergman, J.; Stensland, B. Eur. J. Org. Chem, 2004, 413-418.
II. Synthesis of 4-oxo-4,5-dihydro-3H-pyrrolo[2,3-c]quinoline-1-carboxylic acid
ethyl ester and its isomer 1-oxo-2,9-dihydro-1H-β-carboline-4-carboxylic acid
Bergman, J.; Rehn, S. Tetrahedron, 2002, 45, 9179-9185.
III. The reaction between 3-aminocrotonates and oxindole 3-ylidene derivatives:
synthesis of highly substituted pyrroles.
Rehn, S.; Bergman, J. Tetrahedron, accepted.
IV. Synthetic studies towards the alkaloid granulatimide: synthesis of 3-
imidazolylindole and 2-imidazolyltetrahydroindole.
Rehn, S.; Bergman, J. Manuscript
List of papers………………………………………………………………………….v
1 Introduction to isatin chemistry .............................................................................1
1.1 Synthesis of isatin ..............................................................................................1
1.2 Fundamental reactivity of isatins.......................................................................2
1.2.1 Aromatic substitution.....................................................................................2
1.2.2 N-Alkylation and N-acylation .......................................................................3
1.2.3 Carbonyl reactions .........................................................................................3
2.1 Naturally occuring 3-spiro-oxindoles................................................................9
2.2 Ninhydrin and the Strecker degradation............................................................9
2.3 Azomethine ylides ...........................................................................................10
2.3.1 1,2-prototropic shift .....................................................................................10
2.3.2 Decarboxylative condensation ....................................................................11
2.3.3 Three-component reactions (paper I) ..........................................................12
3 Reactions on 3-methyleneoxindole derivatives ...................................................15
3.1 3-Methyleneoxindole acetic acid ethyl ester...................................................15
3.2 Addition of TosMIC to 3-methyleneoxindole derivatives (paper II) .............16
3.2.1 Mechanistic aspects .....................................................................................17
3.3 3-(Pyrrol-4-yl)-oxindole (paper III) ................................................................22
3.3.2 3-Aminocrotonates and 3-methyleneoxindole acetic acid ethyl ester........23
3.3.3 3-Aminocrotonates and 3-acetonylideneoxindole ......................................24
3.3.4 Chlorination of pyrrolo-oxindoles with POCl3 ...........................................26
4 Synthetic studies towards the alkaloid granulatimide (paper IV) ...................28
4.1 Introduction to imidazolyl indoles...................................................................28
4.2 3-(Imidazolyl)-indoles .....................................................................................29
4.3 2-(Imidazolyl)-tetrahydroindoles ....................................................................29
4.3.1 Published procedures to granulatimide .......................................................29
4.3.2 Retrosynthesis of granulatimide..................................................................30
4.3.3 Synthesis of 2-imidazolyltetrahydroindole .................................................31
6 Appendix: supplementary material .....................................................................34
6.1 Experimental part to section 4 .........................................................................34
1 Introduction to isatin chemistry
Isatin 1 (indole-2,3-dione)1 has been known since 1841 when Erdmann and
Laurent prepared it by oxidation of indigo 2 by nitric and chromic acids. Although
known as a synthetic molecule for almost 140 years, isatin was later found in nature,
for instance in the fruits of the cannon ball tree, Couroupita quianensis Aubl.2 In
man, isatin has been found to function as an endogenous monoamine oxidase
5 2 N
6 N1 N
7 H H
Figure 1. Isatin 1 and indigo 2.
1.1 Synthesis of isatin
The importance of indigo as a possible synthetic dyestuff whithin the textile
industry led to intense research in the area of indigo chemistry. As an offspring to the
efforts in indigo research, the chemistry of isatin was explored, and several synthetic
pathways to isatin were developed. The oldest and the most important method of
synthesising isatin is the Sandmeyer methodology that starts from an aniline 3, which
reacts with chloral hydrate and hydroxylamine hydrochloride in water containing
sodium sulfate to form an isonitrosoacetanilide 4. The isolated isonitrosoanilide 4 is
then treated with concentrated sulfuric acid to yield the isatin 5.
R R R O
NH2 c N
3 4 5
Scheme 1. The Sandmeyer synthesis. a) Cl3CCH(OH)2, H2NOH·HCl, Na2SO4. b) H2SO4. c) H2O.
Second to Sandmeyer’s procedure of isatin synthesis is the method of Stollé
whereby the aniline (usually as its hydrochloride) is reacted with oxalyl chloride to
form an intermediate, chlorooxalyl anilide, which in turn can be cyclised in the
presence of a Lewis acid such as aluminium chloride to the isatin. A recent (but rarely
used) aproach (Scheme 2) was described by Gassman et al.4 in the late seventies.
From a suitable substituted aniline 3, a 3-methylthio-2-oxindole 8 was synthesised
and was subsequently 3-chlorinated with NCS and finally hydrolysed to the isatin 5.
The reaction proceeds via an azasulfonium salt followed by an abstraction of a
(a) Sumpter, W. C. Chem. Rev. 1944, 34, 393-434. (b) Popp, F. D. Adv. Heterocycl. Chem. 1975, 18,
1-58. (c) da Silva, J. F. M., Garden, S. J., Pinto, A. C. J. Braz. Chem. Soc. 2001, 12, 273-324.
Bergman, J., Lindström, J-O., Tilstam, U. Tetrahedron 1985, 41, 2879-2881.
Hamaue, N. Yakugaku Zasshi 2000, 120, 352-362.
Gassman, P. G.; Cue Jr, B. W.; Luh, T-Y. J. Org. Chem. 1977, 42, 1344-1348.
proton, to give a sulfur ylide 7, which underwent a Sommelet-Hauser type
rearrangement, and after ring closure yields the 3-methylthio-2-oxindole 8.5
R 6 X
SMe (1) NCS O
R O R O
Scheme 2. The Gassman synthesis.
1.2 Fundamental reactivity of isatins
Isatin will mainly react at three different sites, namely aromatic substitution at C-5,
N-alkylation, and carbonyl reactions at C-3. If the system carry electron-withdrawing
groups in the benzene ring or at the nitrogen attack at C-2 can also occur.1c
Figure 2. Reactivity of isatin.
1.2.1 Aromatic substitution
Nitration of isatin yields 5-nitroisatin where the reaction proceeds smoothly at
0 ûC6 but the temperature needs to be controlled precisely7 or else the nitration will
give rise to several nitrated products8. Nevertheless, 5,7-dinitroisatin can be
synthesised by merely heating 3,3,5,7-tetranitrooxindole, which in turn can be
obtained from the nitration of oxindole.9 When bromine was added to a solution of
(a) Gassman, P. G.; van Bergen, T. J. J. Am. Chem. Soc. 1974, 96, 5508-5512. (b) Gassman, P. G.;
Gruetzmacher, G.; van Bergen, T. J. J. Am. Chem. Soc. 1974, 96, 5512-5517.
Calvery, H. O.; Noller, C. R.; Adams, R. J. Am. Chem. Soc. 1925, 47, 3058-3060.
Daisley, R. W.; Shah, V. K. J. Pharm. Sci. 1984, 73, 407-408.
Mazhilis, L. I.; Terent’ev, P. B.; Boltin, V.A. Chem. Heterocycl. Compd. (Engl. Transl.) 1989, 25,
Bergman, J.; Bergman, S.; Brimert, T. Tetrahedron 1999, 55, 10447-10466.
isatin in ethanol dibromination occurred to yield 5,7-dibromoisatin in good yield.10
Nevertheless mono bromination at C-5 has been achieved by treatment of isatin with
N-bromosuccinimide, while 5-chloroisatin was obtained using N-chlorosuccinimide.11
Recently, mono iodination of isatin yielding 5-iodoisatin was achieved by using an
aqueous potassium dichloroiodate (KICl2) as the iodinating agent.12
1.2.2 N-Alkylation and N-acylation
Commonly, N-alkylation of isatin proceeds via the sodium salt of isatin, which is
reacted with appropriate alkyl halide or alkyl sulfonate. Methylation can,
nevertheless, be achieved with other reagents, for example potassium tert-butoxide
and dimethyl oxalate (60% yield)13. However, normal alkylation conditions such as
dimethyl sulfate in ethanolic potassium hydroxide (80% yield)14 or benzylation with
sodium hydride and benzyl bromide (99% yield)15 give better yields.
N-Acetylation was performed by heating isatin in acetic anhydride for a couple of
hours,16 although a more recent procedure (sodium acetate and isatin were heated
shortly in acetic anhydride) has been published.17 Protecting amines as N-carbamates
is a commonly adopted strategy, which has also been applied to isatin. N-Boc isatin18
and N-CBz isatin19 have been synthesized in 89% yield and 85% yield, respectively.
1.2.3 Carbonyl reactions
All ketones, as well as, the C-3 carbonyl of isatin are susceptible towards
nucleophiles. Ketalisation serves this perfectly, as a good example of nucleophilic
attack on the carbonyl functionality. Thus employing ethylene glycol20, 1,2-
ethanedithiol21 or 2-mercapto ethanol20a on isatin yields different spiro ketals of
oxindole. Though the example above considers heteroatom dinucleophiles, carbon
nucleophiles do also react at C-3. Grignard reagents also attack at C-3 and yield the 3-
hydroxy-3-substituted oxindoles, which readily can be reduced to 3-substituted
When isatins are exposed to a weak base and a reagent with an active methylene,
3-substituted-3-hydroxy oxindoles are formed. The tertiary alcohol can easily be
dehydrated under acidic conditions to yield 3-methyleneoxindole derivatives. One
Lindwall, H. G.; Bandes, J.; Weinberg, L. J. Am. Chem. Soc. 1931, 53, 317-318.
Buu-Hoi, N. P. Rec. Trav. Chim. 1954, 73, 197-202.
Garden, S. J.; Torres, J. C.; de Souza Melo, S. C.; Lima, A. S.; Pinto, A. C.; Lima, E. L. S.
Tetrahedron Lett. 2001, 42, 2089–2092.
Bergman. J.; Norrby, P-O.; Sand, P. Tetrahedron 1990, 46, 6113-6124.
Harley-Mason, J.; Ingleby, R. F. J. J. Chem. Soc. 1958, 3639-3642.
Overman, L. E.; Peterson, E. A. Tetrahedron 2003, 59, 6905-6619.
Suida, W. Chem. Ber. 1878, 11, 584-590.
Somogyi, L. Bull. Chem. Soc. Jpn. 2001, 74, 873-881.
Wille, G.; Steglich, W. Synthesis 2001, 759-762.
Yamagishi, M.; Yamada, Y.; Ozaki, K.; Tani, J.; Suzuki, M. Chem. Pharm. Bull. 1991, 39, 626-629.
(a) Rajopadhye, M.; Popp, F. D. J. Med. Chem. 1988, 31, 1001-1005. (b) Cliffe, I. A.; Lien, E. L.;
Mansell, H. L.; Steiner, K. E.; Todd, R. S.; White, A. C.; Black, R. M. J. Med. Chem. 1992, 35,
(a ) Baker, J. T.; Duke, C. C. Aust. J. Chem. 1972, 25, 2467-2475. (b) Wenkert. E.; Bringi, N. V.;
Choulett, H. E. Acta Chem. Scand. 1982, B36, 348-350.
Bergman, J. Acta Chem. Scand. 1971, B25, 1277-1280.
example of this reaction is described in Scheme 3. Condensation between isatin 1 and
acetone 9 yields initially 3-acetonyl-3-hydroxy oxindole 10, which can eventually be
dehydrated to 3-acetonylideneoxindole 11.23 Utilising this Knoevenagel condensation
strategy other products between isatin and ketones were obtained and these
condensation products (e.g. 3-hydroxy-3-nitromethyloxindole)24b were further
manipulated to yield different oxindole derivatives.24 One other example is the
preparation of the α, β unsaturated ketone 3-phenacylideneoxindole from isatin and
acetophenone via 3-hydroxy-3-phenacyloxindole.25
O + O O
N N N
H H H
1 9 10 11
Scheme 3. Condensation between isatin and acetone.
The preparation of 3-methyleneoxindole acetic acid ethyl ester 12 was first
described in 1953 in a three step procedure involving a condensation between
oxindole 13 and diethyl oxalate to yield the stable enol 14 (Scheme 4).26 Catalytic
hydrogenation gave oxindole 15, which were followed by dehydration to yield the
α, β unsaturated ethyl ester 12. However α, β unsaturated esters are more readily
available via a Wittig type of reaction involving a phosphorus ylide. Thus isatin was
heated together with ethoxycarbonylmethylenetriphenylphosphorane in glacial acetic
acid and after a couple of hours the product, 3-methyleneoxindole acetic acid ethyl
ester, was obtained in 69 % yield.27 A couple of years later Franke published a
synthesis wherein he used a modified Wittig reagent (triethyl phosphonoacetate 16)
together with isatin 1. This readily performed Horner-Wadsworth-Emmons reaction
gave 3-methyleneoxindole acetic acid ethyl ester 12 in 65 % yield.28
(a) Braude, F.; Lindwall, H. G. J. Am. Chem. Soc. 1933, 55, 327-327. (b) Garden. S. J.; da Silva, R.
B.; Pinto, A. C. Tetrahedron, 2002, 58, 8399-8412.
(a) Pietra, S.; Tacconi, G. Farmaco 1958, 13, 893-910. (b) Tacconi, G.; Pietra, S. Farmaco 1963, 18,
409-423. (c) Tacconi, G. Gazz. Chim. Ital. 1968, 98, 344-357.
Lindwall, H. G.; Maclennan, J. S. J. Am. Chem. Soc. 1932, 54, 4739-4744.
Julian, P. L.; Printy, H. C.; Ketcham, R.; Doone, R. J. Am. Chem. Soc. 1953, 75, 5305-5309.
Brandman, H. A. J. Heterocyclic Chem., 1973, 10, 383-384.
Franke A. Liebigs Ann. Chem. 1978, 717-725.
O O O
N N N
H H H
13 14 15
O + EtO2C P OEt O
N OEt N
1 16 12
Scheme 4. Synthesis of 3-methyleneoxindole acetic acid ethyl ester. a) (CO2Et)2, NaOEt, EtOH. b) H2,
10 % Pd-C, EtOH. c) H2SO4 (cat), AcOH. d) NaOEt, DMF.
Reacting isatin with hydroxylamine or hydrazine derivatives give rise to the
expected condensed products. However when utilising amines a variety of products
can be obtained.29 Treatment of isatin with ammonia as shown in Scheme 5, gave rise
to a mixture of products, there among isamic acid 19 and isamide 20. The structures
of these two products were elucidated in 1976.30 Some years earlier a short
comunication31 presented the structure of isamic acid 19 and a year later the crystal
structure of the p-bromophenacyl isamate32 was published. It is believed that the 3-
imino isatin attacks a second molecule of isatin to yield a dimeric structure 17
whereupon the second isatin moiety is cleaved and further, via lactamization, gives
rise to a spiro oxazolidinone oxindole 18. Opening and re-closure of the spiro
oxindole forms the isamic acid 19. Reaction with a second equivalent of ammonia
yields the isamide 20.
Bergman, J.; Stålhandske, C.; Vallberg, H. Acta Chem. Scand. 1997, 51, 753-759.
Cornforth, J. W. J. Chem. Soc. Perkin Trans. 1 1976, 2004-2009.
Field, G. F. Chem. Commun. 1969, 886.
Blount, J. F. Chem. Commun. 1970, 432.
O NH2 N
18 19 : R=H
20 : R=NH2
Scheme 5. R = H, Isamic acid. R = NH2, isamide.
A more straightforward reaction is the condensation between isatin and aromatic
amines that usually yield the anils, as outlined in Scheme 6, where isatin 1 is
condensed with p-anisidine to yield the anil 21. These anils can be used in synthesis
of other oxindole containing systems. A recent example is the synthesis of 3-spiro β-
lactam oxindole 23 via the 3,3-disubstituted oxindole 22.33 Compound 22 was
obtained when anil 21 was treated with ketene silyl acetal of ethyl acetate in the
presence of a Lewis acid. A diverse pool of primary amines such as n-butylamine will
also give rise to imines together with isatin, and this particular imine was one part in a
recent Staudinger ketene-imine cycloaddition yielding a 3-spiro β-lactam oxindole.34
Scheme 6. a) p-anisidine, EtOH. b) CH2=COEt(OTMS), BF3·OEt2, DCM.
Nishikawa, T.; Kajii, S.; Isobe, M. Chem. Lett. 2004, 33, 440-441.
Lin, X.; Weinreb, S. M. Tetrahedron Lett. 2001, 42, 2631-3633.
Introduction of a second nucleophilic group besides the amino functionallity,
offers the possibility to form several products. Depicted in Scheme 7 is the reaction
between an isatin 24 and 3-amino propanol. Depending on the reaction conditions,
three different products can be obtained: the ring annulated product 25, the 3-spiro
oxindole 26 or the condensation product 27.35 Compounds like o-phenylenediamine
can also give rise to three different products depending on the reaction conditions,
thus for example 2-aminobenzylamine (the higher homologue of o-phenylene-
diamine) yielded only the spiro compound 28 when stirred at room temperature in
methanol as shown in Scheme 7.36
O O N
X NH b X a X
O O O
N N N
26 24 25
Scheme 7. a) X = F. 2 eq. H2N(CH2)3OH, EtOH, cat AcOH, 40-45 ºC, 3h (40%). b) X = F. 2 eq.
H2N(CH2)3OH, EtOH, cat AcOH, reflux temperature, 6h (70%). c) X = F. 2 eq. H2N(CH2)3OH, EtOH,
cat AcOH, reflux temperature, 25 h (70%). d) X = H. 1 eq. 2-aminobenzylamine, methanol, rt., 48h
Dandia, A.; Sati, M.; Sanan, S.; Joshi, R. Org. Prep. Proced. Int. 2003, 35, 433-438.
Bergman, J.; Engqvist, R.; Stålhandske, C.; Wallberg, H. Tetrahedron 2003, 59, 1033-1048.
A good example of the reactivity of isatin can be found in the synthesis37 of the
mitosane core (-) 29 (Scheme 8), which forms the core of the bifunctional DNA
alkylating mitomycin C.38
OMe H2N OMe
N NH Me N NH
(-) 29 mitomycin C
Figure 3. Mitosane core (-) 29 and mitomycin C.
The reaction sequence started with the protection of the benzylic carbonyl of isatin
1 as a ketal to yield the spirodioxolane oxindole 30, which is a well-established
reaction . Secondly, protection of the nitrogen with (Boc)2O (DMAP, Et3N/DCM),
yielded compound 31. Saponification of the Boc-protected isatin ketal 31 gave the
acid 32, which was treated with allyl bromide to yield the diene 33.
a O b c
1 30 31
O O O O
O O O
N N NH
34 35 29
Scheme 8. a) ethylene glycol, p-TsOH, PhH, reflux; b) (Boc)2O, DMAP, Et3N, DCM, rt; c) NaOH,
THF/H2O, reflux; d) allyl bromide, NaH, DMF, 0 °C to rt.
Further transformations gave the diene 34, which was subjected to a ring closing
metathesis yielding the benzoazocin-5-one 35. Additional chemical transformations
led to the tricyclic mitosane core 29 via a transannular cyclization.
(a) Papaioannou, N.; Evans, C. A.; Blank, J. T.; Miller, S. J. Org. Lett. 2001, 3, 2879-2882. (b)
Papaioannou, N.; Blank, J. T.; Miller, S. J. J. Org. Chem. 2003, 68, 2728-2754.
Tomasz, M.; Palom, Y. Pharmacol. Therapeut. 1997, 76, 73-87
2.1 Naturally occuring 3-spiro-oxindoles
A number of naturally occurring pharmacologically active alkaloids have been
recognised to include the structural motif of pyrrolidino-3-spiro-3’-oxindoles.39
Examples (Figure 4) include the fairly simple (-)-horsfiline 3640 which has been
isolated from H. superba (a small tree indigenous to Malaysia) and the more
structurally complex spirotryprostatin B 3741 which exhibits cell cycle inhibition. In
addition to the naturally occurring pyrrolidino-3-spiro-3’-oxindoles synthetic
pyrrolidino- and piperidino-spiro-3’-oxindoles have been shown to exhibit local
anaesthetic properties42. Just recently it was also shown that 3,3-diaryloxindoles
could act as Ca2+-depleting translation initiation inhibitors.43
N HN N
Figure 4. (-)-Horsfiline 36 and spirotryprostatin B 37.
2.2 Ninhydrin and the Strecker degradation
Ninhydrin 38 is utilised for the chemical development of latent fingerprints, since
it works as a powerful indicator for α-amino acids.44 Fewer are aware of the
chemistry that is taking place when α-amino acids are detected by ninhydrin 38.45 The
chemistry of ninhydrin 38 and related compounds such as alloxan 39 (Figure 5) and
isatin can be dated back to 1862 when Strecker observed that alloxan 39 reacts with
alanine to give carbon dioxide and acetaldehyde.46 Hence the Strecker degradation
was discovered and was further investigated in 1948 by Schönberg et al.47
(a) Marti. C.; Carreira, E. M. Eur. J. Org. Chem. 2003, 2209-2219. (b) Joshi, K. C.; Jain, R.; Ghand,
P. Heterocycles 1985, 23, 957-996.
Jossang, A.; Jossang, P.; Hadi, H. A.; Sevenet, T.; Bodo, B. J. Org. Chem. 1991, 56, 6527-6530.
(a) Cui, C. B.; Kakeya, H.; Osada, H. J Antibiot. 1996, 49, 832-835. (b) Meyers, C.; Carreira, E. M.
Angew. Chem. Int. Ed. 2003, 42, 694-696, and references 2-6 therein.
Kornet, M. J.; Thio, A. P. J. Med. Chem. 1976, 19, 892-898.
Natarajan, A.; Fan, Y-H.; Chen, H.; Guo, Y.; Iysere, J.; Harbinski, F.; Christ, W. J.; Aktas, H.;
Halperin, J. A. J. Med. Chem. 2004, 47, 1882-1885.
Joullié, M. M.; Thompson, T. R.; Nemeroff, N. Tetrahedron 1991, 47, 8791-8830.
Petrovskaia, O.; Taylor, B. M.; Hauze, D. B.; Carroll, P.; Joullié, M. M. J. Org. Chem. 2001, 66,
Schönberg, A.; Moubasher, R. Chem. Rev. 1952, 50, 261-277.
(a) Schönberg, A.; Moubasher, R.; Mostafa, A. J. Chem. Soc. 1948, 176-182. (b) Schönberg, A.;
Moubasher, R. J. Chem. Soc. 1950, 1422.
O N O
Figure 5. Ninhydrin 38 and alloxan 39.
The mechanism of the Strecker degradation is exemplified in Scheme 9 whereby
alanine 40 is condensed with ninhydrin 38 to yield an azomethine carboxylic acid 41,
which will undergo decarboxylation. Hydrolysis of the imine 43 yields the amine 44
which is condensed with an additional ninhydrin molecule and finally, the end
product Ruhemann´s purple 45 containing the nitrogen atom from the alanine 40, is
formed. The remnant of the α-amino acid, the aldehyde, contains one carbon and one
nitrogen less as compared with to the initial α-amino acid.
O O O Me O
OH H2N -H2O -CO2
+ O N O
38 40 41
Me H H2O, H
O H2O O
43 Me CHO
O O O
H + 38
O O O
Scheme 9. Strecker degradation between ninhydrin and alanine.
2.3 Azomethine ylides
2.3.1 1,2-prototropic shift
When ninhydrin was treated with various α-amino acids together with a
dipolarophile (e.g. N-phenylmaleimide) formation of cycloadducts was observed.48
Thus, there exists a relationship between the Strecker degradation and the formation
of azomethine ylides. Schiff bases (imines from aldehydes) bearing an electron
withdrawing group at the α position will generate a stabilised azomethine ylide by a
prototropic shift when heated in toluene.49 These stabilised azomethine ylides can
Grigg, R.; Malone, J. F.; Mongkolaussavaratana, T.; Thianpatanagul, S. Tetrahedron 1989, 45,
(a) Amornraksa, K.; Grigg, R. Gunaratne, H. Q. N.; Kemp, J.; Sridharan, V. J. Chem. Soc. Perkin
Trans. 1 1987, 2285- 2296. (b) Tsuge. O.; Kanemasa, S.; Ohe, M.; Yorozu, K.; Takenaka, S.; Ueno,
K. Bull. Chem. Soc. Jpn. 1987, 60, 4067-4078.
then undergo cycloaddition reactions to yield pyrrolidines. Although there are several
routes to azomethine ylides such as the prototopic generation (discussed above) or
desilylation of N-silylmethyl amine reagents,50 I prefer to use the decarboxylative
condensation route due to: 1) The readily accessible pole of α-amino acids which are
cheap and available in large quantities. 2) The reaction can be performed by a simple
procedure and under neutral conditions. 3) The cycloaddition takes place with a
variety of dipolarophiles.
2.3.2 Decarboxylative condensation
In 1970 Rizzi reported evidence for a thermally generated nonstabilised
azomethine ylide51 intermediate from the decarboxylative condensation between
sarcosine and benzophenone.52 This way of generating the 1,3-dipolar azomethine
ylide is believed to proceed via initial formation of an oxazolidinone, which upon
heating will eliminate carbon dioxide. Tsuge et al.53 reacted N-trityl glycine 46 with
formaldehyde 47 to yield the precursor to the N-triphenylmethyl azomethine ylide 49,
i.e. the 3-(triphenylmethyl)-5-oxazolidinone 48 in almost quantatively yield. Heating
this trityloxazolidinon 48 resulted in expulsion of carbon dioxide and formation of an
azomethine ylide 49, which underwent a cycloaddition reaction with the N-(p-tolyl)-
maleimide 50 present to yield the bicyclic compound 51.
Trt N heat
N CO2H + HCHO(aq)
H O -CO2
46 47 48 O
N + N p-Tol Trt N N p-Tol
49 50 51
Scheme 10. Cycloaddition reaction between trityl-5-oxazolidinone 48 and N-(p-tolyl)-maleimide 50.
Since the creation of pyrrolidines is readily achieved via the azomethine ylide and
a dipolarophile, the three-component reaction between α-amino acids, isatin and
dipolarophile would be expected to lead to 3-spiro-(pyrrolidino)-oxindoles. In 1990
Grigg et al.54 showed that when isatin was condensed with benzylamine, a 1,2-
prototropy resulted in an azomethine ylide which was further reacted with methyl
acrylate in a cycloaddition reaction. A year later the decarboxylative condensation
(a) Padwa, A.; Chen, Y-Y. Tetrahedron Lett. 1983, 24, 3447-3450. (b) Vedejs, E.; West, F, G.
Chem. Rev. 1986, 86, 941-955.
Vedejs, E. In Nonstabilized azomethine ylides; Curran, D. P., Ed.; Advances in cycloaddition; Vol 1.
Jai; London, 1988; pp 33-51.
Rizzi, G. P. J. Org. Chem. 1970, 35, 2069-2072.
Tsuge, O.; Kanemasa, S.; Ohe, M.; Takenaka, S. Bull. Chem. Soc. Jpn. 1987, 60, 4079-4089.
Ardill, H.; Dorrity, M. J. R.; Grigg, R.; Leon-Ling, M–S.; Malone, J. F.; Sridharan, V.;
Thianpatanagul, S. Tetrahedron 1990, 46, 6433-6488.
between isatin and secondary α-amino acids to give the azomethine ylide, which in
turn was reacted with methyl acrylate, was published.55
2.3.3 Three-component reactions (paper I)
Since the major part of the decarboxylative condensation route to pyrrolidino-2-
spiro-3’-oxindole derivatives involves cyclic α-amino acids, a study involving acyclic
α-amino acids was initiated. This study included three acyclic N-substituted α-amino
acids, three N-unsubstituted α-amino acids and two cyclic α-amino acids. N-
Benzylmaleimide was chosen as the dipolarophile, both N-phenylmaleimide as well
as unsymmetrical dipolarophiles were also used.
Isatin 1 was reacted with the N-substituted acyclic α-amino acids sarcosine (N-
methyl glycine) 53a , N-benzylglycine 53b and N-methylalanine 53c in a
methanol/water medium at 90 ºC in the presence of N-benzylmaleimide 53a. The
tetracyclic spiro compounds 55 were obtained in good yields.
O R2 R3
1 a N R1
O + R N CO2H
+ R 3
N H O
1 53 54 55
Scheme 11. a) MeOH:H2O (3:1), 90 ºC.
Obviously the amide hydrogen present in isatin did not disturb the reaction since
the use of N-methylated isatin did not alter the course of the reaction.
Table 1. Products obtained by the three-component reaction between α-amino acids 53, isatin 1 and
Entry Compound R1 R2 R3 R4 Time (h) Yield (%)
1 55aa Me H -C(O)NBnC(O)- 18 92
2 N-Me 55aa Me H -C(O)NBnC(O)- 18 82
3 55ba Bn H -C(O)NBnC(O)- 18 77
4 55ca Me Me -C(O)NBnC(O)- 18 79
5 55da H CH(CH3)2 -C(O)NBnC(O)- 2 94
6 55ea H Me -C(O)NBnC(O)- 18 95
7 55fa H H -C(O)NBnC(O)- 2 39
8 55ga -CH2SCH2- -C(O)NBnC(O)- 2 95
9 55ha -CH2CH2CH2- -C(O)NPhC(O)- 0.5 87
10 55bc Bn H CO2Me H 18 38
11 55gc -CH2SCH2- CO2Me H 2 69
Good single crystals were obtained when adduct 55ba was recrystallized from
acetonitrile, and an X-ray structure (Figure 6) could be obtained from these crystals.
Coulter, T.; Grigg, R.; Malone, J. F.; Sridharan, V. Tetrahedron Lett. 1991, 32, 5417-5420.
The stereochemical outcome of the cycloaddition was apparent from the X-ray
structure. Also observable is the aromatic T-stacking involving the phenyl ring from
the former N-benzylmaleimide and H15 on carbon 4 in the oxindole ring (i.e., C15 in
Figure 6). This T-stacking results in a shift upfield (δ = 6.17 ppm for H15 in
compound 55ba) compared to compound 55bc (δ = 6.91 ppm) obtained from isatin 1,
N-benzylglycine 53b and methyl acrylate 54c. On the other hand when N-
phenylmaleimide 54b was incorporated into the structure, an upfield shift was also
noticed (δ = 6.73 ppm for compound 55hb) but due to the lacking methylene the
influence is smaller.
Figure 6. Molecular structure of 55ba showing the atom numbering scheme. An acetonitrile solvate
molecule is excluded.
As discussed earlier, azomethine ylides have been prepared by condensation
between isatin and an unsubstituted amine followed by a 1,2-prototropy.54
Nevertheless, the use of an N-unsubstituted α-amino acid accompanied by isatin with
the intention to prepare an azomethine ylide via the decarboxylative condensation
route has gained scarce attention.56 The tetracyclic pyrrolidino-3-spiro-3’-oxindole
derivatives 55da and 55ea were prepared from the decarboxylative condensation
between isatin 1 and the α-amino acids valine 53d and alanine 53e together with and
N-benzylmaleimide 54a. The yield of these reactions were within the same range of
yields as for the N-substituted α-amino acids. Even glycine 53f yielded a pyrrolidino-
3-spiro-3’-oxindole derivative 55fa albeit in a low yield (39 %). A competing
reaction path involving 1,2-prototropy may be the reason.
In 2001 Azizian et al.57 published a study, wherein the reaction between isatin 1,
proline 53h and N-arylmaleimides in refluxing ethanol yielded pentacyclic
pyrrolidino-3-spiro-3’-oxindoles derivatives 55. Utilising the conditions in Scheme
11 with isatin 1, proline 53h and N-phenylmaleimide 54b gave the pyrrolidino-3-
spiro-3’-oxindole derivative 55hb in 87% yield. Also thiazolidine-4-carboxylic acid
(a) Fokas, D.; Ryan .W. J.; Casebier, D. S.; Coffen, D. L. Tetrahedron Lett. 1998, 39, 2235-2238. (b)
El-Ahl, A–A. S. Heteroatom Chem. 2002, 13, 324-329.
Azizian, J.; Asadi, A.; Jadidi, K. Synth. Commun. 2001, 31, 2727-2733.
53g will give rise to pentacyclic oxindole derivatives of the type 55ga (Scheme 11)
when the decarboxylative condensation taken place between isatin 1 and the imino
acid 53g, followed by a reaction with N-benzylmaleimide 54a as the dipolarophile.
The decarboxylative condensation between thiazolidine-4-carboxylic acid 53g and
proline 53h together with isatin 1 to yield the azomethine ylide have been subjected
to profound studies, both synthetically and theoretically58, therefore there is good
evidence for the mechanistic course and the stereochemical outcome (Scheme 12).
H O S O
O + HN H O
N O O
1 53g 56 57
Bn H H S
O S N
Scheme 12. Decarboxylative condensation path to azomethine ylide.
Interestingly, the reactions involving proline 53h, thiazolidine-4-carboxylic acid
53g and valine 53d proceeded approximately 10 times faster, as compared with e.g.
sarcosine 53a. The reason for this difference is perhaps due to the fact that the ring or
the extra bulkiness pushes the equilibrium towards the spiro oxazolidinone 57, and
that these features also will stabilise the azomethine ylide 58. This geometric feature
will induce the reactions to proceed faster by promoting the formation of the spiro
oxazolidinone 57, which I think is the species that will eliminate carbon dioxide to
give the reactive azomethine ylide 58. There is, nevertheless, also a possibility that
compound 56 is decarboxylated to yield the azomethine ylide 58.
(a) Pardasani, R. T.; Pardasani, P.; Ghosh, R.; Sherry, D.; Mukherjee, T. Heteroatom Chem. 1999,
10, 381-384. (b) Pardasani, P.; Pardasani, R. T.; Sherry, D.; Chaturvedi, V. Synth. Commun. 2002,
32, 435-441. (c) Pardasani, R. T.; Pardasani, P.; Yadav, S. K.; Bharatam, P. V. J. Heterocyclic
Chem., 2003, 40, 557-563. (d) Pardasani, R. T.; Pardasani, P.; Chaturvedi. V.; Yadav. S. K.; Saxena,
A.; Sharma, I. Heteroatom Chem. 2003, 14, 36-41.
3 Reactions on 3-methyleneoxindole derivatives
3.1 3-Methyleneoxindole acetic acid ethyl ester
In the pursue of the synthesis of pyrrolidino-3-spiro-3’-oxindoles I wanted to
explore the use of the easily available 3-methyleneoxindole derivatives as starting
materials. Spiro-oxindoles have already been prepared from 3-methyleneoxindoles.
Franke treated the 3-methyleneoxindole acetic acid ethyl ester 12 with diazomethane
and isolated the 3-spiro-pyrazolyl-oxindole 59, which upon heating expelled nitrogen
to yield 3-spiro-cyclopropano-oxindole 60 (Figure 7).28 Diels-Alder reactions
between 3-methyleneoxindole derivatives and dienes yielded as expected 3-spiro-
N N EtO2C Y R2
O O O
N N N
H H H
59 60 61
Figure 7. Spiro adducts from 3-methyleneoxindole acetic acid ethyl ester.
Worth mentioning in this context is the first step in an exquisite asymmetric total
synthesis of spirotryprostatin B wherein Williams and co-workers treated 3-
methyleneoxindole acetic acid ethyl ester 12 with an azomethine ylide generated from
aldehyde 63 and morpholine derivative 62.60 A similar type of 1,3-dipolar
cycloaddition to the oxindole ethyl ester 12 was performed with the azomethine ylide
generated from N-benzylglycine and isatin via the decarboxylative condensation
O Me Ph
EtO2C MeO O
O Me N
62 O O
N mol. sieves HN
Scheme 13. The first 1,3-dipolar step in the synthesis towards spirotryprostatin B.
(a) Kato, T.; Yamanaka, H.; Ichikawa, H. Chem. Pharm. Bull. 1969, 17(b) Richards, C. G.;
Thurston, D. E. Tetrahedron 1983, 39, 1817-1821. (b) Wenkert, E.; Liu, S. Synthesis 1992, 323-
Sebahar. P. R.; Williams, R. M. J. Am. Chem. Soc. 2000, 122, 5666-5667.
Rehn, S.; Bergman, J.; Stensland, B. Eur. J. Org. Chem. 2004, 413-418.
Isocyanides have received a lot of attention, especially in the synthesis of various
nitrogen containing heterocycles.62 One of the isocyanide reagents, TosMIC (p-
tosylmethylisocyanide), was ”discovered” in 1972 by van Leusen et al.63 In the years
that followed he showed that TosMIC is a versatile reagent for preparations of 5-
membered nitrogen containing heterocycles such as oxazoles64, imidazoles65,
thiazoles66 and pyrroles67.
3.2 Addition of TosMIC to 3-methyleneoxindole derivatives (paper II)
A reaction whereby 3-methyleneoxindole acetic acid ethyl ester 12 was treated
with TosMIC 65 under basic conditions was performed. The idea was to synthesise a
3-spiro-oxindole derivative. From this reaction a high melting (over 300 °C) white
solid was obtained with the elemental composition C14H12N2O3 in 74% yield.
Analysis of the 1H NMR spectrum revealed apart from a 1,2-disubstituted benzene
ring and an ethoxycarbonyl group three signals (two NH and one aromatic CH). From
this information, together with the information obtained from the 13C NMR and IR,
two possible structures were suggested, namely the β-carboline 66 and the
O O O
a NH R Yield (%)
O NH or OMe 52
N N N O Ph 79
H H O H
12 C 66 67
Scheme 14. a) KOtBu, THF, reflux 0.5h.
When the same conditions were applied to the corresponding methyl ester, a
compound of the same kind was obtained although in lower yield (52%). Also 3-
phenacylideneoxindole25 yielded a product (79%) where a benzoyl group replaced the
Marcaccini, S.; Torroba, T. Org. Prep. Proced. Int. 1993, 25, 141-208.
van Leusen, A. M.; Boerma, G. J. M.; Helmholdt, R. B.; Siderius, H.; Strating, J. Tetrahedron Lett.
1972, 13, 2367-2368.
van Leusen, A. M.; Hoogenboom, B. E.; Siderius, H. Tetrahedron Lett. 1972, 13, 2369-2372.
van Leusen, A. M.; Wildeman, J.; Oldenziel, O. H. J. Org. Chem. 1977, 42, 1153-1159.
van Leusen, A. M.; Wildeman, J. Synthesis 1977, 501-503.
van Leusen, A. M.; Siderius, H. Hoogenboom, B. E.; van Leusen, D. Tetrahedron Lett. 1972, 13,
3.2.1 Mechanistic aspects
Two tentative rationalisations are suggested in Scheme 15, one leading to the β-
carboline 66 while the other ends up giving the pyrroloquinolone 67. The initial step
involves a Michael addition to the exocyclic double bond leading to the intermediate
68. Michael adducts from 3-methylenoxindole derivatives have been reported e.g. the
reaction between some 3-methylenoxindole derivatives and malononitrile.68 To form
the β-carboline 66 intermediate 69 would have to ring-close onto isocyanide carbon, a
hydrogen shift and after elimination of p-toluensulfinate yield the indolo annulated
oxazepine 70. Valence tautomerism ought to yield the epoxide 71 which thereafter
opens, and finally after tautomerisation gives the β-carboline 66. The valence
tautomerisation between 70 and 71 can be compared with the corresponding oxepin
equilibrium. On the other hand, if TosMIC works as it normally does, a spiro pyrrolo
oxindole 72 should to be formed via ring closure and loss of p-touluensulfinate and a
hydrogen shift. Cleavage of the oxindole moiety, promoted by the basic conditions,
results in compound 73, which undergoes recyclisation to the pyrroloquinolone 67 via
isocyanate 74. This type of rearrangement of isatins to quinolones has precedent in
the literature.69 It seems that the very spiro structure of 72 render it sensitive to
secondary reactions such as ring expansion to quinolones. Cleavage of 3-spiro
oxindole derivatives was also observed when isatin was treated with ammonia
Higashiyam, K.; Nagase, H.; Yamaguchi, R.; Kawai, K-I.; Otomasu, H. Chem. Pharm. Bull. 1985,
(a) Bennet, G. B; Mason, R. B.; Shapiro, M. J. J. Org. Chem. 1978, 43, 4383-4385. (b) Eistert, B.;
Selzer, H. Chem. Ber. 1963, 96, 1234-1255.
EtO2C EtO2C EtO2C
O O O
N N N
H H H
12 68 69
EtO2C N N
N H H
O 72 70
EtO2C EtO2C 1 2 N
N 9 3
+ H+ 8 N O
N 7 71
C N O
O 6 H5
5 4 2
7 N O
Scheme 15. a) TosMIC, KOtBu, THF.
In spite of the efforts to analyse the spectroscopic data available no enlightenment
to the true structure of the TosMIC adduct (66 or 67) could be achieved. Two-
dimensional NMR studies did not exclude or point towards any of the two candidate
stuctures. The task was then clear, alternative synthesis of both the β-carboline 66 and
the pyrroloquinolone 67 had to be performed.
There are four different carbolines 75, α-, β-, γ- and δ-carbolines,which differs in
the position of the nitrogen in the annulated pyridine ring. β-Carbolines are the most
abundant of the carbolines.
N NH Cl NH
N α N N
H H Cl O
75 76 77
Figure 8. Carbolines, tetrahydro β-carboline and baurine C.
Nature has been endowed with many different β-carbolines due to their
accessibility from the amino acid tryptophan, which serves as the starting material in
the biosynthetic pathway of β-carbolines. Both the non-aromatic tetrahydro-β-
carboline 76 and the fully aromatic β-carboline have been isolated from various
sources. Tetrahydro β-carbolines 76 have for instance been identified in several
commercial sausages such as salami and Spanish chorizo.70 Baurine C 77, a chlorine
containing β-carboline which has been isolated from the terrestrial blue-green alga
Dichothrix baueriana, shows activity against herpes simplex virus type 2.71 Since β-
carbolines are well-documented biologically active compounds, they are interesting
goals in synthetic chemistry.72 The most common way to prepare β-carbolines is the
Pictet–Spengler type of reaction whereby tryptamines and aldehydes are condensed to
give the tetrahydro-β-carbolines followed by oxidation to the fully aromatised β-
carbolines.73 Nevertheless, this strategy cannot be applied to the synthesis of β-
carboline 66, hence a new route had to be developed. Retrosynthesis of compound 66,
shown in Scheme 16, begun with breaking the amide bond which ought to be formed
easily by introducing an ammonia source to enol 78. Formylation to the 2,3-
disubstituted indole 79, available by Fischer indolisation according to Robinson et
al.74, should yield the enol 78.
EtO2C EtO2C OH EtO2C
NH OEt OEt
N O N O N O
H H H
66 78 79
Scheme 16. Retrosynthesis of β-carboline 66.
The 3-ethoxycarbonylmethyl-indole-2-carboxylic acid ethyl ester 79 was treated
with NaH and ethyl formate to yield the formylated species 78. Ring closure was
attempted with ammonium acetate in refluxing DMF but the annulated pyridone I
sought after could not be isolated. Instead lactonization to the indolopyrone 80
occurred. Maybe an attack by the ammonia was prevented by an initial deprotonation
of the enol and therefore no lactamisation occurred.
Herraiz, T.; Papavergou, E. J. Agric. Food Chem. 2004, 52, 2652-2658.
Larsen, L. K.; Moore, R. E.; Patterson, G. M. L. J. Nat. Prod., 1994, 57, 419-421.
Donova, A. K.; Statkova-Abeghe, S. S.; Venkov, A. P.; Ivanov, I. I. Synth.. Commun. 2004, 34,
2813-2821, and references cited therein.
Love, B. E. Org. Prep. Proced. Int. 1996, 28, 1-64, and references cited therein.
Robinson, J. R.; Good, N. E. Can. J. Chem. 1975, 35, 1578-1581.
EtO2C EtO2C OH
NMe2 c a
Me 79 H
81 (56%) 78 (87%)
N O O N O
82 (42%) 66 (20%) 80 (27%)
Scheme 17. a) NaH, ethyl formate, Et2O. b) NH4OAc, p-TosOH(cat), DMF, reflux. c) DMFDMA,
DMF, heat. d) NaH, Et2O.
To circumvent the possible acidity problem with the enol, a dimethylvinylamino
group was introduced by treating the indole 79 with DMFDMA. However, in addition
to the introduction of the dimethylvinylamino group this reagent also N-methylated,
perhaps not unexpectedly,75 the indole nitrogen. Together with the known N-
methylated dimethylvinyl indole 8176 the N-methylated diethyl ester 97 was obtained
in 23% yield. Now, treatment with an ammonia source yielded the mono N-
methylated β-carboline 82 in 42%. Finally a strategy involving the tosylimine 8377
was adopted whereby the tosylimine 83 served as a synthetic equivalent to a C-N
fragment. In this fashion the previously unknown β-carboline 66 was prepared in a
modest yield (20%). The NMR data of the TosMIC adduct and the β-carboline 66 did
not match. Consequently, the TosMIC adduct was not the β-carboline 66.
A survey of the literature revealed very few reports leading to pyrrolo[2,3-
c]quinolones 84, even though structures of this type had been shown to inhibit
proliferation of tumour cells.78 German researchers have synthesised pyrrolo[2,3-
c]quinolone 86 by an intramolecular amide bond formation via reduction of a 3-(2-
nitropheny)-5-methyl-2-ethoxycarbonyl pyrrole 85 (Figure 9).79 The chemistry of the
parent compound of 84, the non-oxidised pyrrolo[2,3-c]quinoline 87, has been
reviewed and does also occur as a moiety in certain alkaloids80.
(a) Stanovinik, B.; Tišler, M.; Hribar, A.; Barlin, G. B.; Brown, D, J. Aust. J. Chem. 1981, 34, 1729-
1738. (b) Middleton, R. W.; Monney. H.; Parrik, J. Synthesis 1991, 740-743.
Gray, N. M.; Dappen, M. S.; Cheng, B. K.; Cordi, A. A.; Biesterfeldt, J. P. J. Med. Chem.1991, 34,
Anglada, L.; Marquez, M.; Sacristan, A.; Ortiz, J. A. Eur. J. Med. Chem. Chim. Ther. 1988, 23, 97-
Thal, C.; Boye, O.; Guenard, D.; Potier, P. WO 9 002 733, 1990.
Görlitzer, K.; Fabian, J.; Frohberg, P.; Drutkowski, G. Pharmazie, 2002, 57, 243-247.
Khan, M. A.;.da Rocha, J. F. Heterocycles 1978, 9, 1617-1629.
NH NH NH NH
N O NO2 N O N
84 85 86 87
Figure 9. Pyrrolo[2,3-c]quinoline derivatives.
The retrosynthesis starts with breakage of the amide bond that then will lead to the
trisubstituted pyrrole 88. This pyrrole, which ought to be acylated in the 2-position,
should be available from the cinnamic acid derivative 89 and TosMIC
O2N + TosMIC
N O R N NO2
67 88 89
Scheme 18. Retrosyntheis of pyrroloquinolone 67.
The cinnamic acid derivative 8981 could indeed be reacted with TosMIC to yield
the adduct 4-(o-nitrophenyl)-pyrrole-3-carboxylic acid ethyl ester 88 (R=H, Scheme
18) which in turn was subjected to several acylating agents such as trichloroacetyl
chloride, oxalyl chloride and TFAA. None of these reagents were successful in their
acylation task, most likely due to the two electron withdrawing groups present in the
Changing the starting material to the α, β-unsaturated sulfone 90 obtained from a
condensation between o-nitrobenzaldehyde and tosylacetonitrile. Treating the α, β-
unsaturated sulfone 90 with ethyl isocyanoacetate yielded the 2,3,4-trisubstituted
pyrrole 91 in 72% yield in a modified Barton-Zard reaction.82 Ring closure to the
relatively unusual pyrroloquinolone 92 (65% yield) was achieved by reduction of the
nitro functionality and the formation of an intramolecular amide bond. Once the
skeleton was correctly assembled, transformation to the desired carbethoxy group was
achieved by hydrolysis in a sulfuric acid/ethanol medium. The hydrolysis yielded a
4:1 mixture of the acid 93 and the ethyl ester 67. The proton NMR spectrum of the
pyrroloquinolone ethyl ester 67 obtained via the lengthy route (11% total yield) in
Scheme 19 matched perfectly with the product obtained from the high yielding (74%)
reaction wherein TosMIC was added to 3-methyleneoxindole acetic acid ethyl ester.
Sinisterra, J. V.; Mouloungui, Z.; Dekmas, M.; Gaser, A. Synthesis 1985, 1097-1100.
(a) Barton, D. H.; Kervagoret, J.; Zard, S. Z. Tetrahedron 1990, 46, 7587-7598. (b) Fumoto, Y. Uno,
H.; Murashima, T. Ono, N. Heterocycles, 2001, 54, 705-720.
NC Tos NC NC
a b c
NO2 NO2 N O
90 (71%) 91 (72%) 92 (65%)
N O N O
Scheme 19. a) ethyl isocyanoacetate, DBU, THF, 0 °C. b) Fe, AcOH, reflux. c) EtOH/ conc. H2SO4
3.3 3-(Pyrrol-4-yl)-oxindole (paper III)
Enaminones (e.g. 3-aminocrotonates) are versatile reagents that have been used in
the synthesis of a multitude of heterocycles such as pyridines, pyrroles, imidazoles
and pyrimidines.83 There are two electrophilic centers namely C-1 and C-4 (cf. α, β-
unsaturated esters) and two nucleophilic centers C-2 and the amino functionallity (cf.
enamines). Thus enaminones are suitable for reactions with polydentate reagents, a
reaction that usually will give heterocycles. One example is the reaction between
methyl propiolate 94 and the enaminone 3-amino-cyclohex-2-enone 95 gives the
tetrahydroquinolone derivative 96.84 Another well-known example is the Nenitzescu
indole synthesis wherein an enaminone is reacted with a quinone to yield an indole
H O O
CO2Me H2N O N
94 95 96
Scheme 20. Reaction of an enaminone to give tetrahydroquinolone derivative 96.
(a) Elassar, A-Z. A.; El-Khair, A. A. Tetrahedron 2003, 59, 8463-8480. (b) Stanovnik, B.; Svete, J.
Chem. Rev. 2004, 104, 2433-2480. (c) Negri, G.; Kascheres, C.; Kascheres, A. J. J. Heterocycl.
Chem. 2004, 41, 461-491.
(a) Sluyter, M. A. T.; Pandit, U. K.; Speckamp, W. N.; Huisman, H. O. Tetrahedron Lett. 1966, 7,
87-90. (b) Ruda, M.; Bergman, J.; Koehler, K.; Ye, L. Heterocycl. Commun. 2003, 9, 571-574.
Allen, G. R. Org. Reactions 1973, 20, 337-454.
As seen previously 3-methylene oxindole derivatives are attacked by nucleophiles
on the exocyclic double bond. That is precisely what Tacconi et al.86 reported in 1976
when N-methyl-3-phenacylideneoxindole was treated with 1-pyrrolidinocyclo-
pentene. The reaction proceeded via an initial attack on the exocyclic double bond by
the enamine that gave an intermediary zwitterionic compound 97, which later yielded
a new enamine 98. Hydrolysis of enamine 98 and treatment with aniline yielded the
Ph H O
O H H Ph
Figure 10. Enamine adducts based on 3-phenacylideneoxindole.
A similar reaction between 1-pyrrolidinocyclohexene and N-methyl-3-
phenacylideneoxindole gave an adduct in the same way as described above. This
adduct was carefully hydrolysed and depending on the conditions during the work-up,
two different diastereoisomers were isolated, one of them, compound 100, is shown
in Figure 10. The structure of compound 100 was supported by extensive 1H NMR
studies and finally confirmed by an X-ray diffraction analysis.87 Since enaminones
share the reactivity of enamines, a study on the reactions of enaminones with 3-
methylenoxindole derivatives was also initiated.
3.3.2 3-Aminocrotonates and 3-methyleneoxindole acetic acid ethyl ester
The two enaminones 101, methyl- and ethyl- 3-aminocrotonate, were refluxed in
ethanol together with 3-methyleneoxindole acetic acid ethyl ester 12. This reaction
yielded a colourless solid 103, which was isolated after partial evaporation of the
solvent. The formation of 103 proceeds via an initial zwitterionic adduct 102 as
Tacconi, G.; Invernizzi, A., G.; Desimoni, G. J. Chem. Soc. Perkin Trans. 1 1976, 1872-1879.
López-Alvarado, P.; García-Granda, S.; Álvarez-Rúa, C.; Avendaño, C. Eur. J. Org. Chem. 2002,
O HN Me
H2N Me NH2
EtO2C EtO2C R Yield (%)
O CO2R a: Et 62
b: Me 45
Scheme 21. Addition of 3-aminocrotonates to 3-methyleneoxindole acetic acid ethyl ester.
The 1H NMR spectrum of 103a showed the presence of two isomers, the E, Z pair
of the double bond. Recrystallisation from ethyl acetate yielded only one isomer,
although isomerisation took place in solution. On the contrary, the methyl ester 103b
was collected as one single isomer. To elucidate the stereochemistry of the double
bond in the methyl ester 103b, NOE difference spectra were recorded. The methyl
signal at 1.94 ppm (the vinylic methyl group) was irradiated and it resulted in an
NOE interaction at 3.96 ppm. This signal corresponds to the α-proton of the ethyl
ester. On the other hand, when the signal corresponding to the methoxy group (3.22
ppm) was irradiated no NOE interaction could be detected at 1.94 ppm. This piece of
information corroborated the double bond to be in Z configuration, as depicted in
3.3.3 3-Aminocrotonates and 3-acetonylideneoxindole
A literature survey regarding 3-aminocrotonates and α, β-unsaturated carbonyl
compounds ended up in the classical Hantzsch pyrrole synthesis dating back to
1890.88 The reaction involves an in situ formation of a 3-aminocrotonate ester, which
is alkylated by a halo–ketone or –aldehyde and finally cyclised to the pyrrole. A
couple of years ago a one-pot synthesis of pyrrole derivatives was published wherein
the pyrrole 105 was prepared from 3-aminocrotonates such as 101b and
dibenzoylethylene 104.89 Interestingly, this reaction was performed without solvent
and gave the pyrrole in almost quantatively yield.
MeO2C O MeO2C
+ Ph O
Me NH2 -H2O Me Ph
101b 104 105
Scheme 22. One-pot synthesis of pyrroles in the solid-state.
(a) Hantzsch, A. Ber. 1890, 23, 1474-1476. (b) Roomi, M., W.; MacDonald, S. F. Can. J. Chem.
1970, 48, 1689-1697.
Kaupp, G.; Schmeyers, J.; Kuse, A.; Atfeh, A. Angew. Chem. Int. Ed. 1999, 38, 2896-2899.
The 3-acetonylideneoxindole 11 was refluxed in toluene together with three
different 3-aminocrotonates, namely ethyl 3-aminocrotonate 101a, methyl 3-
aminocrotonate 101b and methyl 3-(methylamino)crotonate 101c. Solid products
were obtained in good yields from all these reactions.
a H CO2R
O + O
N HN Me N -H2O
11 101 106
N Me Me
Me Me R1 R2 Yield 107':107''
H 1 1
CO2R CO2R a: Et H 91% 3:2
b: Me H 89% 5:3
N c: Me Me 88% 2:1
Scheme 23. Synthesis of pyrrolo oxindole. a) Toluene, reflux.
As seen in Scheme 23 the products obtained were the cyclised pyrrolo-oxindoles
107. Why do the adducts 106 ring close and not the adduct 103 obtained in the
reaction described in Scheme 21? The explanation can be found in the different
oxidation level in the α, β-unsaturated ester and the α, β-unsaturated ketone. The
adduct 103 incorporates an ester but adduct 106 contains a more reactive carbonyl
(ketone) which then cyclises more easily. Synthesis of 3-pyrrol-3’-yloxindoles have
been achieved, in very low yields (3-5%), by reacting 3-diazooxindoles with pyrroles
in the presence of a rhodium(II) acetate catalyst, even though the main product from
this reaction were 3-pyrrol-3’yloxindoles obtained in good yields (60-70%).90 The
reaction between 3-aminocrotonate esters and 3-acetonylideneoxindole give a pyrrole
moiety with much less expensive reagents and with higer yields (around 90%).
In contrast to the non-cyclised adducts 103 the pyrrolo-oxindoles 107 occure as
keto-enol tautomers with a slight preference towards the keto form 107’. This keto-
enol tautomerism could be observed in the NMR spectra as a doubling of the signals,
i.e. the characteristic pattern of an ethyl group was doubled. When the NMR tube was
heated to 125 °C both 13C spectra and 1H spectra only showed one set of signals. A
plausible explanation for the shift towards the keto form may be found in the
increased steric hindrance in the planar enol form.
As good crystals were obtained by recrystallisation of 107c an X-ray structure
analysis was conducted. In the solid state the only form present was the racemic keto
form. Hence in solution three different species were present, the enol 107c’’ and the
(a) Muthusamy, S.; Gunanathan, C. Synlett 2002, 1783-1786. (b) Muthusamy, S.; Gunanathan, C.;
Nethaji, M. J. Org. Chem. 2004, 69, 5631-5637.
two enantiomers 107c’(S) and 107c’(R). In the solid state the molecules are joined as
dimers by hydrogen bonding, the indolic hydrogen is paired with the amide oxygen.
Figure 11. The racemic compound 107c’. The atom-labelling scheme is shown.
Incorporating an α, β-unsaturated ketone, 3-phenacylideneoxindole 108 was also
reacted with methyl 3-aminocrotonate 101a to yield the pyrrolo-oxindole 109.
Ph N Me
O Ph MeO2C
O O OH
N (92%) N N
H H H
108 109' 109''
Scheme 24. a) methyl 3-amino crotonate 101a, toluene, reflux, 3h.
Somewhat surprisingly the keto-enol equilibrium was more shifted towards the
keto form (90%) compared to the pyrrolo oxindoles obtained from 3-acetonylidene-
oxindole. The predominance of the keto form can be explained in terms of π-stacking
between the oxindole nucleus and the phenyl ring. When the pyrrolo oxindole 109 is
toggled between the keto 109’and the enol form 109’’ C-3 is switching between sp3
and sp2 hybridisation. As an sp2 carbon exhibit a planar geometry, this fact will force
the indole and the pyrrole ring to be in the same plane. This configuration diminishes
the possibility of π-stacking between the phenyl ring and the oxindole. Consequently
when C-3 is sp3 hybridised it adopts a tetrahedral geometry and permit the phenyl
ring to bend over the oxindole to form π-stacking between the phenyl ring and the
3.3.4 Chlorination of pyrrolo-oxindoles with POCl3
To achieve a permanent transformation of the oxindole to an indolic moiety,
treatment with phosphorus oxychloride (POCl3) ought to trap the enol form as a 2-
chloroindole derivative. Thus the pyrrolo-oxindole 107c was heated in neat POCl3
(and quenched by pouring it onto an ice/water mixture and made alkaline with a 45%
KOH(aq) solution). Two products were obtained, the minor one was the expected 2-
chloroindole derivative 110 and the major product was identified as the uncommon
pyrrolo annulated indolo-2-pyranone 111. The cause of the formation of the pyrrolo
annunlated indolo-2-pyranone derivative 111 was attributed to the addition of the
very strong alkaline KOH(aq) solution, which probably hydrolysed the methyl ester.
Ring closure was then effected by the attack of the carboxylate anion on the 2-
Me Me Me
Me N Me
N N Me
a b O
O Cl O
N (69%) N (quant.) N
H H H
107c 110 111
Scheme 25. a) POCl3, Et3N, MeCN reflux over night. b) 45% KOH(aq), MeOH, reflux 3h.
In order to gain more information about the formation of the pyrrolo annulated
indolo-2-pyranone 111 the chlorination reaction was performed in a two-steps, first
the POCl3 chlorination, and then a separate ester hydrolysis. Changing the conditions
of the chlorination step to POCl3 together with triethylamine in refluxing acetonitrile
gave the 2-chloro-3-pyrrolo-indole derivative 110 in 69% yield. Hydrolysis of the
methyl ester was performed and yielded the pyrrolo annulated indolo-2-pyranone 111
in quantitative yield.
4 Synthetic studies towards the alkaloid granulatimide (paper
4.1 Introduction to imidazolyl indoles
Among the 20 α amino acids that are usually found in proteins three incorporate
nitrogen heterocycles. One of them is the saturated nitrogen heterocycle proline
(pyrrolidine-2-carboxylic acid). The other two, histidine and tryptophane, have
aromatic nitrogen heterocycles in the side chain. Tryptophan contains an indole and
histidine is an imidazole derivative. Noteworthy, both of these α amino acids are
essential amino acids for vertebrates, which means they cannot be biosynthesised
from other sources but have to be ingested via the food.
Histidine incorporates an imidazole side chine that possesses several properties
that have key features in proteins. Having a pKa value near 7, imidazole is well suited
to play a crucial role in enzymes where proton transport is required. Decarboxylated
histidine, histamine, is intimately connected with inflammatory and allergic reactions.
As being important not only in the biosynthesis and functionality of proteins,
tryptophan and histidine are present independently or in combination in several
alkaloids. Some examples are granulatimide 112,91 isogranulatimide 113,91
didemnimide A 11492 and grossularine 11593.
N O H
O O N
N N N
O N NH
H H N
Figure 12. Alkaloids incorporating imidazolyl indoles.
Granulatimide 112 and isogranulatimide 113 have been isolated from the ascidian
Didemnum granulatum and crude methanol extracts from the ascidian showed
activity during screening for G2 cell cycle checkpoint inhibitors.91 The synthesis
(Scheme 27) of the granulatimides 112 and 113 showed that isogranulatimide 113
Berlinck, R. G. S.; Britton, R.; Piers, E.; Lim, L.; Roberge, M.; da Rocha, R. M.; Andersen, R. J. J.
Org. Chem., 1998, 63, 9850-9856.
Vervoort, H. C.; Richard-Gross, S. E.; Fenical, W. J. Org.Chem. 1997, 62, 1486-1490.
(a) Moquin-Pattey, C.; Guyot, M. Tetrahedron 1989, 45, 3445-3450. (b) Choshi, T.; Yamada, S.;
Sugino, E.; Kuwada, T.; Hibino, S. J. Org. Chem. 1995, 60, 5899-5904.
was the natural occurring G2 checkpoint inhibitor but also granulatimide 112
exhibited this pharmacological property.91
There are two main ways of synthesising 3-(imidazolyl)-indoles. One starts with
an imidazole and then the indole moiety is built from there and one from an existing
indole derivative. As TosMIC has been used successfully to synthesise imidazoles,
the choice of starting from an indole derivative was easy since 3-formyl indole is a
commercial compound and an excellent staring material for an imidazole synthesis.
Thus conversion of 3-formyl indole into the benzyl imine 116 proceeded as expected,
likewise did the addition of TosMIC 65a to the imine 116, which gave a 3-
(imidazolyl)-indole derivative 117a in 79% yield.
Tos Et3N R
+ N C
116 65 117
a: H (79%)
b: Ph (66%)
c: p-methoxphenyl (34%)
Scheme 26. Synthesis of 3-imidazolyl indoles 117.
Aryl substituted TosMIC derivatives (65b and 65c) yielded tri-substituted
imidazole derivatives (117b and 117c). When p-methoxyphenyl substituted TosMIC
65c was used a lower yield of the imidazole 117c was obtained. The destabilising
effects that the p-methoxyphenyl group exerts on the intermediate anion created can
rationalise this decrease of the yield. Sisko et al.94 have published a procedure
wherein the imine is formed in situ and thereafter reacted with an aryl substituted
TosMIC reagent to yield 1,4,5-trisubstituted imidazoles. By adopting and modify this
sequence by first reacting 3-formylindole with benzylamine and then adding TosMIC
and triethylamine yielded 3-(3-benzyl-5-phenyl-3H-imidazol-4-yl)-1H-indole 117b in
86% yield. This sequence was also tried with plain TosMIC but this only resulted in a
plethora of non-separable compounds. These results are in accordance with what has
already been reported.94
4.3.1 Published procedures to granulatimide
The first published syntheses of granulatimide 112 and isogranulatimide 113
appeared together with the isolation, and followed a biomimetic pathway via
didemnimide A 114.91 The indole part was derived from indole-3-acetamide 118,
which was condensed with the imidazole derivative 119 to yield, after reduction and
removal of the methoxymethyl protecting group, the alkaloid didemnimide A 114. A
Sisko, J.; Kassick, A. J.; Mellinger, M.; Filan, J. J.; Allen, A.; Olsen, M. A. J. Org. Chem. 2000, 65,
photolysis of didemnimide A 114 resulted in formation of granulatimide 112 in 91%
yield and isogranulatimide 113 in 8% yield.
N O N O
+ NH 10% Pd-C, NH
MeO2C O MeCN
N N N N
O O O N O
N Br Br Br
OMe 123 N
+ EtMgBr H
Scheme 27. Published syntheses of granulatimide.
The second synthesis of granulatimide 112, developed by Japanese researchers,
also utilised a photoreaction to obtain the alkaloid.95 The 2-stannylindole 120 was
prepared in situ from the 2-lithiated 1-methoxyindole and then utilised in Stille
coupling reaction with 4-iodo-imidazole 121 to yield the key intermediate 2-
imidazolyl indole derivative 122. The dibromomaleimide 123 was attached to the 3-
position of indole via the magnesiumbromide salt of compound 122 to yield the 2,3-
disubstituted indole 124. To create the bond between the imidazole and the maleimide
ring compound 124 was irradiated with UV-light as outlined above. Granulatimide
112 was then finally obtained after deprotection of the imidazole part.
4.3.2 Retrosynthesis of granulatimide
As described earlier in the synthesis of (3-benzyl-3H-imidazol-4-yl)-1H-indole
117a the construction of an imidazole directly bonded to an indole was easily effected
by reacting a benzylimine with the TosMIC reagent. This TosMIC strategy was
chosen to build the 2-imidazolyl indole 122 required as the key intermediate for the
synthesis of granulatimide 112 according the retrosynthesis in Scheme 28. An
Yoshida, T.; Nishiyachi, M.; Nakashima, N.; Murase, M.; Kotani, E. Chem. Pharm. Bull. 2002, 50,
addition of TosMIC to a Schiff base like 126 ought to yield an 2-imidazolylindole
122. The imine 126 should be obtainable from a 2-formylindole 127. The second
important compound is the 2-formyl indole 127, which unfortunately is not as
accessible as 3-formylindole. Hence a new route to 2-formylindoles had to be
O N O
NH N + O N O TosMIC
N N N N
112 122 125
Scheme 28. Retrosynthesis of granulatimide 112.
4.3.3 Synthesis of 2-imidazolyltetrahydroindole
By using the tetrahydroindole it is possible to introduce the formyl group by the
Vilsmeier reagent hence pyrroles do react with electrophiles at the 2-position. This is
contrary to indoles where electrophilic reagents end up in the 3-position. A survey of
the literature regarding the synthesis of tetrahydroindoles revealed a promising
procedure published by a Danish group. Thus Madsen et al96 have prepared
tetrahydroindoles by reacting α-chloroacrylonitrile with methyl- and cyclohexan-
cyclohexanone imine. In a similar approach, cyclohexanone benzylimine 128 was
reacted with α-chloroacrylonitrile to yield the ring-closed compound 129.
N NH N N
Bn Bn Bn
128 129 c 130: R=H (90%)
131: R=CHO (61%)
d 132: R=CHNBn
Scheme 29. a) α-chloroacrylonitrile, Et3N, MeCN, 0 °C. b)EtOH, reflux. c) POCl3, DMF, 0 °C to rt. d)
BnNH2, THF, rt.
To eliminate HCN Madsen et al.96 performed a pyrolysis. This procedure could be
further simplified when it was discovered that mere heating of 129 in ethanol caused
elimination of HCN. Purification of tetrahydroindole 130 can be carried out by
chromatography but it is not imperative. Now, with the previously known97
tetrahydroindole 130 in hand formylation with the Vilsmeier reagent (DMF/POCl3)
Madsen, J. Ø.; Meldal, M.; Mortensen, S.; Olsson, B. Acta Chem. Scand. 1981, B35, 77-81.
Trost, B. M.; Keinan, E. J. Org. Chem. 1980, 45, 2741-2746.
proceeded smoothly to yield 2-formyltetrahydroindole derivative 131. The synthesis
of the Schiff base 132 was executed in almost quantative yield in a THF solution.
During the work with the imine 132 we noticed that it was very sensitive towards
acids. In an attempt to purify small amounts of the imine 132 by chromatography on
silica gel it only resulted in hydrolysis of the imine back to the aldehyde 131.
Formation of the 2-imidazolyltetrahydroindole 133 was observed when the
conditions for imidazole synthesis described in Scheme 26 were applied. Better yields
was obtained when the imine 132 as refluxed in DCM together with TosMIC and
Et3N (Scheme 30). Using these conditions the 2-imidazolyltetrahydroindole 133 was
obtained in 72% yield.
N Bn a N
N (72%) N N
Scheme 30. Et3N, TosMIC, DCM, reflux, 24h.
To conclude, the synthesis of a potential key intermediate, 2-
imidazolyletetrahydroindole 133, in a synthetic route towards the alkaloid
granulatimide 112 have been presented. The construction of the imidazole moiety was
effected with TosMIC chemistry. Moreover 3-imidazolylindoles were synthesised
utilising the same TosMIC strategy.
Min handledare professor Jan Bergman som en gång antog mig som doktorand. Du
har lärt mig kemisk ”know-how” och visat upp ett brinnande intresse för den
organiska kemins underbara värld.
Docent Alf Claesson för ditt engagemang som extern handledare och tack
AstraZeneca R&D, Södertälje för det finansiella stödet.
Robert Engqvist för delandet av dragskåpsrad och labutrymme samt diskussioner
rörandes musik, sport och där i mellan kemi.
Dr. Tomasz Janosik som alltid har en bestämd åsikt om det mesta, speciellt inom
den gastronomiska sfären. Speciellt tack till din ”rödfärgning” av avhandlingen.
Dr. Per Wiklund och Dr. Niklas Wahlström som har kamperat i samma lab och
bidragit till kemidiskussioner och övrigt stämningshöjande som t.ex. bränder…
Dr. Birgitta Stensland för dina ”bilder” av några av mina föreningar.
Malin Björk för att ha stått ut med mig som skrivbordsgranne, nu får du extra
utrymme för dina papper.
Ann-Louise Johnson för ditt slit med att hålla köket rent och snyggt. RESPEKT.
Solveig, din närvaro har en uppiggande effekt på alla tröga doktorander.
Kollegorna i skrivrummet. De nuvarande Jealux, Sassa och Birgitta men även de
som har passerat revy i skrivrummet.
Medlemmar i JB gruppen; Johnny, Jeff, Hamid, Ivan men även alla andra som har
passerat revy under mina år i gruppen.
Vänner i Ödeshög och i Södertälje.
Ett innerligt tack till mina föräldrar och syskon utan stödet från er skulle inte
denna avhandling finnas.
Åsa tack för att du sa ja!
6 Appendix: supplementary material
6.1 Experimental part to section 4
General Remarks: Solvents (PA grade) were commercial and used without
further purification. NMR spectra were recorded in DMSO-d6 solutions at 25 °C,
unless otherwise stated, on a Bruker DPX 300 spectrometer, operating at 300 MHz
for 1H and 75 MHz for 13C. δ Values are reported in ppm and J values in Herz. IR
spectra were recorded on a Thermo Nicolet Avatar 330 FT-IR instrument using
single-reflection ATR. Melting points were determined with a Büchi melting point B-
545 apparatus and are uncorrected.
Triethylamine (0.18 g, 1.73 mmol) was added to a solution of benzyl-[1-(1H-
indole-3-yl)methylidene]-amine 116 (0.27 g, 1.15 mmol) and TosMIC (0.27 g, 1.38
mmol) in methanol (6 ml) and THF (3 ml). After 4 h of reflux the reaction was
quenched with 80 ml ice/water and acidified with acetic acid. Extraction with EtOAc
(4x20 ml) and washing of the combined organic phases with sat. aq. NaHCO3 and
brine yielded after drying with MgSO4 and evaporation of the solvent 0.39 g of a
brownish gum. Purification by dry flash cromatography (hexane-EtOAc) yielded 3-
(3-benzyl-3H-imidazol-4-yl)-1H-indole 117a (0.25 g, 79 %). Recrystallization from
acetonitrile afforded analytically pure 3-(3-benzyl-3H-imidazol-4-yl)-1H-indole
117a. Mp:139-140 ûC; IR (neat) 1450, 1226, 818, 756, 712 cm-1; δH = 5.29 (2H, s),
6.96-6.98 (2H, m), 7.02-7.08 (1H, m), 7.10-7.30 (6H, m), 7.41 (1H, d, J=8.0), 7.53
(1H, d, J=7.8), 7.81 (1H, s), 11.32 (1H ,s); δC = 48.5 (t), 103.8 (s), 111.8 (d), 119.1
(d), 120.1 (d), 122.2 (d), 124.2 (d), 126.6 (d), 127.1 (s), 127.3 (s), 127.7 (d), 128.7
(d), 136.3 (s), 137.0 (s), 137.6 (d)
Sequencial procedure. 3-Formylindole (1.84 g, 13 mmol) and benzylamine (4.35
g , 41 mmol) were stirred together with MgSO4 (6.5 g) in THF (70 ml). After 24 h of
stirring α-tosylbenzyl isocyanide 65b (2.76 g, 10 mmol) and Et3N (1.31 g, 13 mmol)
were added and the reaction was stirred for 3 days at rt. Removal of all solid material
(MgSO4) and evaporation of the solvent gave a crude yellow solid. Trituration with
hot DCM (30 ml) gave 2.36 g (66%) of the trisubstituted imidazole 117b as a pale
One pot procedure. Benzylamine (2.23 g, 21 mmol) and Et3N (0.79 g, 8 mmol)
were added to a stirred suspension of 3-formylindole (1.14 g, 8 mmol), α-tosylbenzyl
isocyanide 65b (1.93 g, 7 mmol) and MgSO4(5 g) in THF (50 ml). After two days of
stirring at ambient temperature the reaction mixture was filtered and added to water
(300 ml) and acidified with AcOH. The trisubstituted imidazole 117b was collected
as a yellow solid. Mp 232-233 ûC; IR (neat 1456, 1444, 939, 779, 731, 717, 696, 653
cm-1; δH = 5.01 (2H, s), 6.89-6.95 (3H, m), 7.01-7.25 (8H, m), 7.31 (1H, d, J = 2.5),
7.45-7.52 (3H, m), 7.96 (1H, s), 11.46 (1H, s); δC = 47.5 (t), 103,4 (s), 112.0 (d),
118.7 (d), 119.5 (d), 121.6 (d), 121.9 (s), 125.3 (d), 125.7 (d), 126.4 (d), 126.6 (d),
126.9 (s), 127.3 (d), 127.9 (d), 128.4 (d), 135.3 (s), 136.1 (s), 137.85 (s), 137.89 (d),
Triethylamine (1.11 g, 11 mmol) was added to a solution of benzylimine 116 (1.28
g, 5 mmol) and p-methoxyphenylTosMIC 65c (1.65 g, 5 mmol) in a medium consist
of MeOH (5 ml) and THF (20 ml). After 24 h of stirring at rt the mixture was added
to water (150 ml) and acidified with AcOH, and extracted with EtOAc (4x30 ml). The
organic phase (combined) was washed with sat. aq. NaHCO3 (30 ml) and brine (30
ml), dried (MgSO4) and evaporated. Recrystallization from ethanol yielded 0.72 g (34
%) of the trisubstituted imidazole 117c. Mp 223-224 ûC; IR (neat): 1443, 1247, 1176,
836, 756, 725, 692, 661, 610 cm-1; δH = 3.64 (3H, s), 4.99 (2H, s), 6.69 (2H, d, J=8.9),
6.90-6.94 (3H, m), 7.02 (1H, d, J=7.8), 7.09-7.14 (1H, m), 7.17-7.24 (3H, m), 7.40-
7.46 (3H, m), 7.91 (1H, s), 11.44 (1H, s); δC = 47.5 (t), 54.9 (q), 103.6 (s), 111.9 (d),
113.4 (d), 118.8 (d), 119.5 (d), 120.7 (s), 121.6 (d), 126.4 (d), 126.6 (d), 126.7 (d),
127.0 (s), 127.3 (d), 128.1 (s), 128.4 (d), 136.1 (s), 137.6 (d), 137.97(s), 138.05 (s),
During a period of 15 min α-chloroacrylonitrile (21.1g, 0.24 mol) was added to
stirred solution of cyclohexylidenebenzylamine 128 (45.1g, 0.24 mol) and
triethylamine (40ml, 0.29 mol) in acetonitrile (200 ml), while the tempertature was
kept between 25 °C and 30 °C. After 1h of stirring at rt the reaction mixture was
cooled to 5 °C and Et3N·HCl was filtered off and the solvent was evaporated. The
residue was treated with ether (100 ml) and the solution was aged over night at 5 °C
whereupon more Et3N·HCl was removed and the ether was evaporated to yield crude
129 as a yellow oil. This product was refluxed in ethanol (150 ml) for 2 days and then
evaporated to yield 46.64 g (90%) of crude N-benzyl-4,5,6,7-tetrahydroindole 130 as
a brownish oil. Purifiction could be achieved by chromatography. The NMR data
were in accordance with published data.97 δH (CDCl3) = 1.77-1.84 (4H, m), 2.45-2.49
(2H, m), 2.58-2.61 (2H, m), 5.00 (2H, s), 6.03 (1H, d, J=2.7), 6.61 (1H, d, J=2.7),
7.07-7.39 (5H, m); δC (CDCl3) = 21.9 (t), 23.40 (t), 23.45 (t), 23.7 (t), 50.1 (t), 106.6
(d), 117.8 (s), 119.5 (d), 126.7 (d), 127.3 (d), 128.1 (s), 128.8 (d), 138.7 (s)
POCl3 (1.87 g, 12 mmol) was added to a stirred solution of the terahydroindole
130 (1.29 g, 6 mmol) in DMF (5 ml) at 0 °C. After 5 minutes at this temperature the
reaction mixture was allowed to reach rt and then stirred for 16 h and then quenched
by adding it to a suspension of ice/water (100 ml) and made alkaline by adding 2M
KOH(aq). When all ice melted the mixture was extracted trice with EtOAc (50 ml) and
the organic phase (combined) was washed with water (4x50 ml) and brine (50 ml)
finally dried (MgSO4) to yield a brown oily crude product (1.32 g). Purification by
dry flash cromatography yielded 0.89 g (61%) of the 2-formylpyrrole derivative as a
yellow oil. IR (neat): 2931, 2853, 1650, 729, 695 cm-1; δH (CDCl3) = 1.70-1.85 (4H,
m), 2.48-2.59 (4H, m), 5.58 (2H, s), 6.76 (1H, s), 7.03-7.05 (2H, m), 7.23-7.33 (3H,
m), 9.45 (1H, s); δC (CDCl3) = 22.1 (t), 22.6 (t), 22.9 (t), 23.2 (t), 48.1 (t), 120.8 (s),
123.5 (d), 126.5 (d), 127.2 (d), 128.7 (d), 130.7 (s), 137.9 (s), 140.7 (s), 178.5 (d)
Benzylamine (0.73 g, 7 mmol) was added to the 2-formyl pyrrole 131 (1.63 g, 0.73
mmol) in THF (20 ml) and stirred at room temperature for 16 h. The solvent was then
evaporated to yield the crude benzyl imine 132 (2.17 g, 96 % yield). This imine was
then used without further purification. IR (neat): 2927, 2828, 1632, 726, 694 cm-1; δH
(CDCl3) = 1.74-1.84 (4H, m), 2.49-2.60 (4H, m), 4.65 (2H, s), 5.76 (2H, s), 6.42 (1H,
s), 7.00-7.39 (11H, m) 8.19 (1H ,s); δC (CDCl3) = 22.1 (t), 23.0 (t), 23.1 (t), 23.5 (t),
48.1 (t), 65.2 (t), 116.0 (d), 118.8 (s), 126.41 (d), 126.43 (d), 126.7 (d), 127.5 (d),
128.2 (d), 128.4 (s), 128.5 (d), 135.3 (s), 139.3 (s), 140.7 (s), 153.5 (d)
Triethylamine (0.96 g, 9.5 mmol) was added to a solution of benzylimine 132
(1.56 g, 4.7 mmol) and TosMIC (0.93 g, 4.7 mmol) in DCM (30 ml). The mixture
was then heated at reflux for one day and then evaporated and purified by dry flash
chromatography (hexane/ethyl acetate) to yield 2-imidazolyltetrahydroindole 133
1.27 g (72 % yield) as a yellow oil which solidified upon standing. Mp:97-103 ûC; IR
(neat): 2924, 2850, 719, 694, 657 cm-1; δH (CDCl3) = 1.77-1.85 (4H, m), 2.42-2.46
(2H, m), 2.55-2.60 (2H, m), 4.75 (2H, s), 4.83 (2H ,s), 6.01 (1H, s), 6.81-6.83 (2H,
m), 6.92-6.95 (3H, m), 7.22-7.29 (6H, m), 7.50 (1H, s); δH (CDCl3) = 22.2 (t), 23.0
(t), 23.2 (t), 23.5 (t), 46.7 (t), 48.2 (t) 110.4 (d), 117.7 (s), 118.4 (s), 124.8 (s), 125.8
(d), 126.9 (d), 127.2 (d), 127.7 (d), 128.45 (d), 128.53 (d), 129.6 (d), 130.0 (s), 136.6
(s), 137.7 (d), 138.6 (s).
AcOH acetic acid
(Boc)2O di-tert-butyl dicarbonate
DMFDMA dimethylformamide dimethyl acetal
DNA deoxyribonucleic acid
NMR nuclear magnetic resonance
NOE nuclear Overhauser effect
RCM ring closing metathesis
rt room temperature
TFAA trifluoroacetic anhydride
p-TsOH para-toluensulfonic acid