Synthesis of 3,4-dihydroisoquinolines by a C(sp3)-H activation/electrocyclization strategy: Total synthesis of coralydine

Article abstract of DOI:10.1002/anie.200804444

(Chemical Equation Presented) Thanks to C-H activation: 3-Aryl-3,4-dihydroisoquinolines (2) are synthesized from bromobenzenes (1) by a sequence comprising a C(sp³)-H activation, a Curtius rearrangement, and a tandem electrocyclic ring-opening/

Full text of DOI:10.1002/anie.200804444

Angewandte
Chemie
DOI: 10.1002/anie.200804444
Synthetic Methods
3
ꢀ
Synthesis of 3,4-Dihydroisoquinolines by a C(sp ) H Activation/
Electrocyclization Strategy: Total Synthesis of Coralydine**
Manon Chaumontet, Riccardo Piccardi, and Olivier Baudoin*
In memory of Christian Marazano
Isoquinolines, in various oxidation states (for example,
isoquinolines, dihydro- and tetrahydroisoquinolines), are
found in numerous natural products[1] and bioactive mole-
cules, such as the tetrahydroprotoberberine alkaloid coraly-
dine[2] and the peripheral benzodiazepine receptor ligand
PK11195 (see below).[3]
Scheme 1. Overall strategy for the synthesis of dihydroisoquinolines.
Reagents and conditions: a) Pd(OAc)2 (10 mol%), P(tBu)3 (20 mol%),
K2CO3, DMF, 1408C; b) NaOH, MeOH/H2O, reflux; c) diphenylphos-
phoryl azide (DPPA), Et3N, toluene, reflux, then aq. HCl, 808C;
d) ArCHO (1 equiv), MgSO4, CH2Cl2, reflux; e) DMF, 1608C.
Dihydroisoquinolines (DHIQ) constitute synthetically
strategic molecules in this context, since they are precursors
to both isoquinolines and tetrahydroisoquinolines. Whereas a
range of new methods have been developed for the synthesis
of 1,2-DHIQ,[4] efforts have focused on the parent 3,4-DHIQ
to a much lesser extent. Classical syntheses of 3,4-DHIQ
involve Bischler-Napieralski-type reactions that rely on an
electrophilic aromatic substitution (SEAr) step.[5] However,
for the purpose of introducing broader structural diversity
onto this motif, the development of conceptually different
synthetic alternatives is of great interest. We envisioned the
construction of a variety of 3-aryl-3,4-DHIQ 5 from imino-
BCB 4 (BCB = benzocyclobutene) by thermal tandem elec-
trocyclic ring-opening/6p-electrocyclization via o-xylylene
intermediate A (Scheme 1).[6,7] In turn, imines 4 would arise
from amino-BCB 3, which should be accessible from BCB-
esters 2 by hydrolysis and Curtius rearrangement. BCB 2 can
be readily synthesized from bromobenzenes 1 by the palla-
ꢀ
dium-catalyzed C H activation of methyl groups that was
developed recently in our group.[8]
Amino-BCB 3a–d (Table 1) were synthesized from the
corresponding aryl bromides 1. The construction of the
ꢀ
cyclobutene ring by Pd-catalyzed C H activation/intramo-
ꢀ
lecular C C coupling was carried out in good yield as
described earlier (Scheme 1, step a).[8b] After ester hydrolysis
(Scheme 1, step b), the corresponding carboxylic acids under-
went a Curtius rearrangement in the presence of diphenyl-
phosphoryl azide (DPPA, Scheme 1, step c).[9] Amino-BCB
3e (Table 1, entry 12) was obtained from the corresponding
BCB-nitrile by hydrolysis to the primary amide and PhI-
(OCOCF3)2-mediated Hofmann rearrangement[10] (see the
Supporting Information). Amino-BCBs, such as 3, have
limited thermal stability when R2 = H, and therefore they
have very rarely been isolated and employed in synthesis.[11]
[*] Dr. R. Piccardi, Prof. Dr. O. Baudoin
Institut de Chimie et Biochimie Molꢀculaires et Supramolꢀculaires
CNRS UMR5246, Universitꢀ Claude Bernard Lyon 1,
43 bd du 11 Novembre 1918, 69622 Villeurbanne (France)
Fax: (+33)4-7243-2963
E-mail: olivier.baudoin@univ-lyon1.fr
Homepage: http://www.icbms.fr/user/main.asp?pag=348
In this case, the presence of a quaternary benzylic carbon
2
ꢀ
(R ¼ H), which is also necessary for the C H activation step
(a),[8b] stabilized the molecule, probably by raising the energy
barrier for the cyclobutene ring-opening, which allowed us to
isolate the free amines 3 and to engage them in the next step.
Thus amino-BCB 3 was treated with one equivalent of an
aromatic aldehyde (Scheme 1, step d) and the corresponding
imines 4 underwent the thermal tandem electrocyclic ring-
opening/ring-closing process (Scheme 1, step e). It was antici-
pated that the presence of the R2 alkyl substituent would
favor inward rotation of the imine group to give the Z isomer
of the o-xylylene intermediate (Scheme 1, A) that was
Dr. M. Chaumontet
Institut de Chimie des Substances Naturelles, CNRS UPR2301
Avenue de la Terrasse, 91198 Gif-sur-Yvette (France)
[**] We thank ICSN, Institut de Recherche Servier, and CNRS (ATIP
grant) for financial support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200804444.
Angew. Chem. Int. Ed. 2009, 48, 179 –182
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
179

Communications
required for the subsequent 6p-electrocyclization.[12] Indeed,
the thermolysis of imine 4a (R1 = H, R2 = Me, Ar= Ph),
derived from amine 3a and benzaldehyde, in DMF at 1608C
produced 3,4-DHIQ 5aa in 73% yield (Table 1, entry 1). The
reaction in DMF was found to be superior to that in more
conventional apolar solvents. Using these conditions, a variety
of amino-BCB 3a–e and aromatic aldehydes were combined
to give the corresponding 3,4-DHIQ 5 in satisfactory overall
yields (Table 1, entries 2–11). In particular the method
seemed compatible with various substituents R1 (Table 1,
entries 6-9) and R2 (Table 1, entries 10–11) on amino-BCB 3,
and with a variety of aromatic and heteroaromatic (Table 1,
entries 4, 5, 9) aldehydes.[13]
Table 1: Synthesis of 3,4-dihydroisoquinolines 5 from amino-BCB 3.
Entry
BCB-amine[a]
ArCHO
Product
Yield
[%][b]
1
73
65
3a
3a
5aa
5ab
5ac
2
3
To demonstrate the synthetic utility of the synthesized 3,4-
DHIQ, compounds, 5aa and 5ad were converted to the
corresponding isoquinolines 6a and 6d by aromatization in
the presence of Pd/C, and to tetrahydroisoquinolines 7a and
7d by reduction with sodium borohydride (Scheme 2).
3a
52
4
5
6
3a
3a
68
60
60
5ad
5ae
Scheme 2. Aromatization and reduction of 3,4-DHIQ. Reagents and
conditions: a) 10% Pd/C, decaline, reflux; b) NaBH4, MeOH, 208C.
In the latter case, an approximate 7:3 ratio of diastereo-
isomers was detected, the major diastereoisomer having the
expected cis configuration.
As a second illustration of this strategy , we synthesized
the fused heterocycle 8, which is of potential medicinal
interest (Scheme 3).[14] To this end, the novel BCB-ester 2 f,
incorporating a p-methoxybenzyl (PMB)-protected terminal
alcohol was first synthesized in three steps from bromide 9 by
3b
3b
5ba
7
8
59
61
60
61
61
5bb
5ca
ꢀ
sequential alkylations and palladium-catalyzed C H activa-
tion. As reported for other substrates,[8b] the C H activation
ꢀ
step occurred, completely regioselectively, on the methyl
group without affecting the other alkyl chain (Scheme 3,
step c, 73% yield). After ester hydrolysis (Scheme 3, step d),
we first attempted the direct conversion of the corresponding
carboxylic acid to amine 3 f by Curtius rearrangement and
HCl-mediated hydrolysis of the isocyanate intermediate, as
before (Scheme 1), but the concomitant cleavage of the PMB
protecting group could not be avoided. As the unprotected
alcohol proved unsatisfactory in the subsequent steps, we
opted for a two-step procedure (Scheme 3, steps e,f). Thus
the Curtius isocyanate intermediate was quenched with allyl
alcohol, to give the corresponding allylcarbamate, which was
converted to amino-BCB 3 f under palladium catalysis.[15]
Imine formation from 3 f and benzaldehyde and subsequent
thermolysis furnished 3,4-DHIQ 5 f in 63% yield (Scheme 3,
step g). The NaBH4-mediated reduction of 5 f gave tetrahy-
droisoquinoline cis-7 f as the isolated major diastereoisomer
in 74% yield (Scheme 3, step h). Finally, removal of the PMB
group using cerium(IV) ammonium nitrate (CAN), followed
3c
3c
9
5cb
5d
10
11
3d
3e
5e
[a] See Supporting Information for detailed preparative procedures.
[b] Yield of the isolated product, calculated from 3.
180
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 179 –182

Angewandte
Chemie
Scheme 3. Synthesis of fused heterocycle 8. Reagents and conditions:
a) NaHMDS (HMDS=hexamethyldisilazide), PMBO(CH2)4I (PMB=p-
methoxybenzyl), THF, 208C, 76%; b) NaHMDS, MeI, THF, 74%;
c) Pd(OAc)2 (10 mol%), P(tBu)3 (20 mol%), K2CO3, DMF, 1408C,
73%; d) NaOH, MeOH/H2O, reflux, 85%; e) DPPA, Et3N, toluene,
reflux, then allyl alcohol, 808C, 83%; f) [Pd(PPh3)4] (2 mol%), PPh3
(9 mol%), 2-ethylhexanoic acid, CH2Cl2, 208C, 68%; g) benzaldehyde,
MgSO4, CH2Cl2, reflux, then DMF, 1608C, 63%; h) NaBH4, MeOH,
208C (d.r.=7:1), 74%; i) [Ce(NH4)2(NO3)6] (CAN), MeCN/H2O, 208C,
78%; j) HBF4, then PPh3, diisopropyl azodicarboxylate (DIAD), THF,
reflux, 67%.
Scheme 4. Synthesis of (ꢁ)-coralydine. Reagents and conditions:
a) NaOH, MeOH/H2O, reflux, 90%; b) DPPA, Et3N, toluene, reflux,
then aq. HCl, 808C, 69%; c) 10, CH2Cl2, 208C, then DMF, 1608C,
52%; d) NaBH4, MeOH, 208C (d.r.=6:1), 68%; e) nBu4NF, THF,
208C, 86%; f) HBF4, then PPh3, DIAD, THF, reflux, 63%.
isoquinoline-containing molecules, including the tetrahydro-
protoberberine alkaloid coralydine.
Experimental Section
[16]
General procedure for the synthesis of 3,4-DHIQ 5: The aldehyde
(0.2–0.6 mmol) and MgSO4 (5 equivalents) were added at 208C to a
solution of amino-BCB 3 (1 equivalent) in CH2Cl2 (0.1–0.2m concen-
tration) under argon. The mixture was heated at reflux until
completion of the reaction (monitored by 1H NMR spectroscopy).
After cooling, the reaction mixture was filtered under argon through
syringe filter, to remove undissolved material, and the solvent
removed under reduced pressure to give the crude imine 4. The
crude imine was dissolved in degassed anhydrous DMF (0.01m
solution based on total conversion) was heated to 1608C for 2.5–
3.5 h to form the corresponding 3,4-DHIQ 5. After completion of the
reaction (monitored by 1H NMR spectroscopy), the solvent was
removed under reduced pressure and the residue was purified by flash
chromatography on silica gel using ethyl acetate and heptanes as the
eluent.
by the Mitsunobu reaction in the presence of HBF4
furnished target compound 8 in 52% yield for the final two
steps.
As a final demonstration of this strategy, the convergent
synthesis of the tetrahydroprotoberberine alkaloid coralydine
was undertaken (Scheme 4). BCB-ester 2c was synthesized in
three steps from commercially available starting materials
[8b]
ꢀ
according to our C H activation method.
It was then
converted to amino-BCB 3c by hydrolysis and Curtius
rearrangement (as described in Scheme 1). Imine formation,
from amino-BCB 3c and the known aldehyde 10,[17] and
subsequent thermolysis provided 3,4-DHIQ 5cc in 52% yield
from 3c. This somewhat lower yield might be imputable to the
presence of an ortho substituent on the arylaldehyde. The
reduction of the imine gave a 7:1 mixture of diasteroisomers
(96% combined yield), from which the major cis diastereo-
isomer 11 was isolated (68% yield). Desilylation and HBF4-
mediated Mitsunobu reaction (as in Scheme 3) furnished
racemic coralydine (54% yield from 11), which had identical
physical properties to those reported for the natural prod-
uct.[2] This total synthesis gives an overall yield of 6.2% for a
nine-step linear sequence.
Received: September 9, 2008
Published online: November 26, 2008
ꢀ
Keywords: alkaloids · C H activation · electrocyclic reactions ·
heterocycles · natural products
.
[1] a) K. W. Bentley, Nat. Prod. Rep. 2006, 23, 444 – 463, and
references therein; b) L. Grycovꢀ, J. Dostꢀl, R. Marek, Phyto-
chemistry 2007, 68, 150 – 175.
[2] a) S.-T. Lu, T.-L. Su, T. Kametani, A. Ujiie, M. Ihara, K.
Fukumoto, Heterocycles 1975, 3, 459 – 464; b) H. Bruderer, J.
Metzger, A. Brossi, J. J. Daly, Helv. Chim. Acta 1976, 59, 2793 –
2807.
In conclusion, we have developed a novel strategy for the
synthesis of 3-aryl-3,4-dihydroisoquinolines via amino-ben-
zocyclobutenes. Key steps include a palladium-catalyzed
C(sp3)-H activation and a tandem electrocyclic ring-open-
ing/6p-electrocyclization. The synthetic utility of this
approach was demonstrated by the synthesis of various
Angew. Chem. Int. Ed. 2009, 48, 179 –182
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
181

Communications
[3] P. Casellas, S. Galiegue, A. S. Basile, Neurochem. Int. 2002, 40,
[8] a) O. Baudoin, A. Herrbach, F. Guꢁritte, Angew. Chem. 2003,
115, 5914 – 5918; Angew. Chem. Int. Ed. 2003, 42, 5736 – 5740;
b) M. Chaumontet, R. Piccardi, N. Audic, J. Hitce, J.-L. Peglion,
E. Clot, O. Baudoin, J. Am. Chem. Soc. 2008, 130, 15157 – 15166.
[9] T. Shioiri, K. Ninomiya, S. Yamada, J. Am. Chem. Soc. 1972, 94,
6203 – 6205.
[10] I. Erdelmeier, C. Tailhan-Lomont, J.-C. Yadan, J. Org. Chem.
2000, 65, 8152 – 8157.
[11] a) J. A. Skorcz, J. E. Robertson, J. Med. Chem. 1965, 8, 255 – 257;
b) W. A. Oppolzer, Synthesis 1978, 793 – 802.
[12] a) C. W. Jefford, G. Bernardinelli, Y. Wang, D. C. Spellmeyer, A.
Buda, K. N. Houk, J. Am. Chem. Soc. 1992, 114, 1157 – 1165;
b) N. Mariet, H. Pellissier, J.-L. Parrain, M. Santelli, Tetrahedron
2004, 60, 2829 – 2835.
[13] With aliphatic aldehydes, the isolated product was that depicted
below, rather than the desired DHIQ: It presumably arose from
outward rotation of the imine, followed by tandem [1,5]-
hydrogen shifts [Eq (1)].
475 – 486.
[4] Selected work: a) M. Ohtaka, H. Nakamura, Y. Yamamoto,
Tetrahedron Lett. 2004, 45, 7339 – 7341; b) N. Asao, S. Yudha, T.
Nogami, Y. Yamamoto, Angew. Chem. 2005, 117, 5662 – 5664;
Angew. Chem. Int. Ed. 2005, 44, 5526 – 5528; c) R. Yanada, S.
Obika, H. Kono, Y. Takemoto, Angew. Chem. 2006, 118, 3906 –
3909; Angew. Chem. Int. Ed. 2006, 45, 3822 – 3825; d) N. Asao, K.
Iso, S. Yudha, Org. Lett. 2006, 8, 4149 – 4151; e) S. Obika, H.
Kono, Y. Yasui, R. Yanada, Y. Takemoto, J. Org. Chem. 2007, 72,
4462 – 4468; f) K. Gao, J. Wu, J. Org. Chem. 2007, 72, 8611 –
8613; g) S. Su, J. A. Porco, Jr., Org. Lett. 2007, 9, 4983 – 4986.
[5] M. Chrzanowska, M. D. Rozwadowska, Chem. Rev. 2004, 104,
3341 – 3370.
[6] A. K. Sadana, R. K. Saini, W. E. Billups, Chem. Rev. 2003, 103,
1539 – 1602.
[7] For examples of electrocyclizations from BCB, see ref. [2a] and:
a) T. Kametani, K. Ogasawara, T. Takahashi, J. Chem. Soc.
Chem. Commun. 1972, 675 – 676; b) T. Kametani, K.
Ogasawara, T. Takahashi, Tetrahedron 1973, 29, 73 – 76;
c) T. Kametani, C. Ohtsuka, H. Nemoto, K. Fukumoto,
Chem. Pharm. Bull. 1976, 24, 2525 – 2531; d) K. Shish-
ido, M. Ito, S. Shimada, K. Fukumoto, T. Kametani,
Chem. Lett. 1984, 1943 – 1946; e) K. Shishido, E.
Shitara, K. Fukumoto, T. Kametani, J. Am. Chem.
Soc. 1985, 107, 5810 – 5812; f) K. Shishido, K. Hiroya,
K. Fukumoto, T. Kametani, Heterocycles 1989, 28, 39 –
41; g) K. Shishido, A. Yamashita, K. Hiroya, K. Fukumoto, T.
Kametani, Tetrahedron Lett. 1989, 30, 111 – 112; for more recent
reports: h) K. Ohmori, K. Mori, Y. Ishikawa, H. Tsuruta, S.
Kuwahara, N. Harada, K. Suzuki, Angew. Chem. 2004, 116,
3229 – 3233; Angew. Chem. Int. Ed. 2004, 43, 3167 – 3171; i) Y.
Matsuya, N. Ohsawa, H. Nemoto, J. Am. Chem. Soc. 2006, 128,
13072 – 13073.
[14] L. C. Gobbi, T. Luebbers, P. Mattei, R. Narquizian, P. C. Wyss
(Hoffmann-La Roche AG), WO03055881, 2003.
[15] P. D. Jeffrey, S. W. McCombie, J. Org. Chem. 1982, 47, 587 – 590.
[16] P. G. Sammes, S. Smith, J. Chem. Soc. Chem. Commun. 1983,
682 – 684.
[17] N. Sotomayor, E. Domꢂnguez, E. Lete, J. Org. Chem. 1996, 61,
4062 – 4072.
182
www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 179 –182