Aminomethylation/hydrogenolysis as an alternative to direct methylation of metalated isoquinolines - A novel total synthesis of the alkaloid 7-hydroxy-6-methoxy-1-methylisoquinoline

Article abstract of DOI:10.3762/bjoc.14.8

Highly-substituted isoquinolines are important scaffolds in syntheses of natural products and in drug development and hence, effective synthetic approaches are required. Here we present a novel method for the introduction of a methyl group at C1 of isoquinolines. This is exemplified by a new total synthesis of the alkaloid 7-hydroxy-6-methoxy-1-methylisoquinoline. Direct metalation of 7-benzyloxy-6-methoxyisoquinoline with Knochel-Hauser base, followed by cuprate-mediated methylation gives the target alkaloid directly, but separation from the educt is cumbersome. Quenching the metalated intermediate with Eschenmoser's reagent gives an easy to clean tertiary benzylamine, which, after quaternization with iodomethane, is easily converted into the desired 1-methylisoquinoline by hydrogenolysis of both the benzylamine and benzyl ether groups.

Full text of DOI:10.3762/bjoc.14.8

Aminomethylation/hydrogenolysis as an alternative to direct
methylation of metalated isoquinolines – a novel total
synthesis of the alkaloid 7-hydroxy-6-methoxy-
1-methylisoquinoline
Benedikt C. Melzer, Jan G. Felber and Franz Bracher*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2018, 14, 130–134.
Department of Pharmacy – Center for Drug Research,
Ludwig-Maximilians University of Munich, Butenandtstr. 5–13,
D-81377 Munich, Germany
doi:10.3762/bjoc.14.8
Received: 13 October 2017
Accepted: 27 December 2017
Published: 11 January 2018
Email:
Franz Bracher* - Franz.Bracher@cup.uni-muenchen.de
Associate Editor: B. Stoltz
* Corresponding author
© 2018 Melzer et al.; licensee Beilstein-Institut.
License and terms: see end of document.
Keywords:
alkaloid; aminomethylation; hydrogenolysis; isoquinoline; metalation;
methylation
Abstract
Highly-substituted isoquinolines are important scaffolds in syntheses of natural products and in drug development and hence, effec-
tive synthetic approaches are required. Here we present a novel method for the introduction of a methyl group at C1 of isoquino-
lines. This is exemplified by a new total synthesis of the alkaloid 7-hydroxy-6-methoxy-1-methylisoquinoline. Direct metalation of
7-benzyloxy-6-methoxyisoquinoline with Knochel–Hauser base, followed by cuprate-mediated methylation gives the target alka-
loid directly, but separation from the educt is cumbersome. Quenching the metalated intermediate with Eschenmoser’s reagent
gives an easy to clean tertiary benzylamine, which, after quaternization with iodomethane, is easily converted into the desired
1-methylisoquinoline by hydrogenolysis of both the benzylamine and benzyl ether groups.
Introduction
The isoquinoline ring system has considerable relevance in drug candidates containing this partial structure are in clinical devel-
development, since it is found in both complex [1] and “simple” opment [4]. Consequently, synthetic approaches enabling a free
alkaloids (variously substituted derivatives of the native variation of substituents on this heteroaromatic ring system are
isoquinoline) isolated from different plants [2], and antiviral required. Numerous methods have been published over the
and antimicrobial activities have been found for numerous decades for the construction of highly substituted isoquinolines
isoquinoline alkaloids [3]. Further, isoquinoline is regarded as a [4]. Alternatively, subsequent functionalization of isoquino-
“privileged scaffold” in drug design, and a large number of drug lines is feasible, e.g., via Pd-catalyzed C–H functionalization
130

Beilstein J. Org. Chem. 2018, 14, 130–134.
[5] or regioselective direct ring metalation (at C1) [6-8]. be converted into the corresponding 1-methyl product. Since the
General aspects of the direct methylation of electron-deficient 7-hydroxy group of target alkaloid 1 is not compatible with the
N-heterocycles have been reviewed [9].
metalation reagent, the corresponding benzyl ether 3 was
selected as central building block.
In a project aimed at the synthesis of tri- and tetracylic alka-
loids containing the isoquinoline scaffold, we were interested in Results and Discussion
isoquinoline building blocks which bear a methyl group at C1, In our previous work we prepared isoquinoline 3 in a three-step
since this group should, due to its intrinsic C–H acidity, open procedure starting from commercially available O-benzyliso-
opportunities for further functionalization. Since most isoquino- vanillin (2) in a modified Pomeranz–Fritsch reaction published
line alkaloids bear hydroxy and/or methoxy substituents at the by Reimann and Renz [14]. In the present investigations, we
positions 6 and 7, we regarded 7-hydroxy-6-methoxy-1- even optimized this procedure, ending up with a protocol that
methylisoquinoline (1) as well as analogues containing a pro- does not afford purification of any of the intermediates, and
tected 7-hydroxy group as highly attractive building blocks for gives isoquinoline 3 in a straightforward operation in 65% iso-
our project. A literature search revealed that compound 1 is in lated yield. For this purpose, the starting aldehyde 2 was sub-
fact a natural product. It was isolated from the trunk bark of the jected to a reductive amination with aminoacetaldehyde
Taiwanese tree Hernandia nymphaeifolia (Hernandiaceae) in dimethyl acetal and NaBH4, followed by N-tosylation and
1996 [10]. No data on the biological activities of this alkaloid hydrochloric acid-mediated cyclization under concomitant
have been reported so far. Further, this literature search showed N-detosylation and aromatization. Direct ring metalation of 3
that two total syntheses of isoquinoline 1 had been published with TMPMgCl∙LiCl was performed as described by us previ-
even before its identification as a natural product (Figure 1). In ously [7]. First attempts for a direct methylation at C1 with
1963, Franck and Blaschke [11] obtained 1 by dehydrogenation iodomethane failed completely. Only upon addition of catalytic
of its 1,2,3,4-tetrahydro analogue (which itself had to be pre- amounts of CuCN∙2LiCl [15] significant methylation took
pared in several steps) with MnO2 in poor yield (24% crude place. The crude reaction mixture contained almost equimolar
product; yield of the final crystallization step for purification amounts of the desired 1-methylisoquinoline 4 and starting ma-
not given). In 1965, Bruderer and Brossi [12] reported on an ap- terial 3. Due to the very similar polarities of 3 and 4, chromato-
proach whose central step was a Pictet–Gams cyclization of graphic separation was very tedious, and only 34% of methyl
N-acetylated β,3-dimethoxy-4-benzyloxyphenethylamine, compound 4 was isolated, accompanied by about 30% of
which gave the 7-O-benzyl derivative of alkaloid 1 in 27% starting material 3 and mixed fractions. Debenzylation of 4 by
yield. The required precursor is available from commercially catalytic hydrogenation in methanol solution under palladium
available precursor 2 via a nitrostyrene intermediate in three catalysis gave the alkaloid 1 in almost quantitative yield
steps in 45% overall yield [13]. In conclusion, no efficient ap- (Scheme 1).
proach to this “simple” alkaloid 1 has yet been published.
So this new metalation/methylation protocol offered a new ap-
This prompted us to work out a novel synthetic access to alka- proach to alkaloid 1 with an overall yield (22% from commer-
loid 1. Our new strategy was fueled by our previous findings in cially available precursor 2) much higher than in the above
the course of new approaches to benzylisoquinoline, oxoapor- mentioned previously published methods [11,12]. However, the
phine [7], and oxoisoaporphine alkaloids [8], where we could unsatisfactory separation of 1-methylisoquinoline intermediate
demonstrate the power of regioselective direct ring metalations 4 from its unmethylated precursor 3 rendered this protocol unat-
of isoquinolines at C1 with sterically hindered amide bases like tractive. This is a problem that is common to numerous methyl-
TMPMgCl∙LiCl (Knochel–Hauser base) [6]. For our present ation protocols, including methylation of metalated arenes [16],
purpose, an appropriate 1-metalated isoquinoline species was to methylation of arenes bearing directing groups with methyl-
Figure 1: Previously published total syntheses of alkaloid 1 [11,12].
131

Beilstein J. Org. Chem. 2018, 14, 130–134.
Scheme 1: Total synthesis of alkaloid 1 via direct ring metalation and methylation.
magnesium bromide [17], and direct methylation using radical [21]. Obviously, and in contrast to earlier reports on related
reactions [9,18,19].
naphthol Mannich bases [20], the benzylamine moiety of 5 is
resistant to hydrogenolysis, whereas the benzyl ether is readily
Hence, we searched for a more viable method for the introduc- removed. This order of reactivity is known from previous work
tion of the methyl group. A prerequisite was that the organome- [22]. Since it is further known that quaternary benzylammo-
tallic intermediate should be trapped by an electrophile which nium compounds undergo hydrogenolysis easier than the corre-
could later be transformed into a methyl group, but the trapping sponding tertiary benzylamines [22], we converted amine 5 into
product should be easily separable from the starting material 3. methoiodide 7 by treatment with iodomethane. This salt was ob-
For this purpose we selected Eschenmoser’s salt (N,N-dimethyl- tained in pure form in 78% yield by simply collecting the pre-
methyleniminium iodide) as an electrophile. Trapping of the cipitate, and there was no indication of an undesired methyla-
metalated species was expected to give the N,N-dimethylamino- tion of the ring nitrogen of the isoquinoline. High-pressure
methyl derivative 5 in reasonable yield. Later on, hydrogeno- hydrogenolysis of 7 gave only minor amounts (about 7%) of the
lytic cleavage of the generated benzylamine-type group at C1 desired alkaloid 1, but 1-methyl compound 4 with intact
should give the desired 1-methyl moiety. Related reductive O-benzyl residue was isolated in 54% yield. The surprising
cleavage reactions have been published earlier by Möhrle [20] stability of the O-benzyl residue in this experiment might be
for phenolic Mannich bases in the course of the total synthesis due to the iodide counterion, which is known to be a poison for
of the naphthalene-derived natural product plumbagin.
palladium catalysts. Cleavage of the highly reactive benzylam-
monium residue still takes place, but O-debenzylation is pre-
In fact, trapping 1-metalated isoquinoline 3 with Eschenmoser’s dominantly suppressed by this catalyst poison. Finally,
salt gave the aminomethyl derivative 5 in 37% yield. Chromato- poisoning of the catalyst was prevented by simply passing a
graphic separation from starting material 3 (recovered yield: solution of the methoiodide 7 through a chloride-loaded ion
32%) was unproblematic.
exchanger prior to catalytic hydrogenation. In this manner alka-
loid 1 was obtained in 94% yield (18% overall yield from com-
Surprising results were obtained in our hydrogenolysis experi- mercially available precursor 2, Scheme 2).
ments with 5, which were aimed at simultaneous O-debenzyla-
tion at the 7-position and conversion of the N,N-dimethylamino- Conclusion
methyl group at C1 into a methyl group. Hydrogenation in pres- In conclusion, we have worked out two novel protocols for the
ence of palladium as catalyst at 1 bar in the presence or absence introduction of a methyl group at C1 of isoquinolines. These
of small amounts of sulfuric acid gave the phenolic product 6 in were applied to the total synthesis of the alkaloid 7-hydroxy-6-
high yield with unchanged N,N-dimethylaminomethyl group. methoxy-1-methylisoquinoline (1), but should also be of value
The same result was obtained at high pressure (40 bar) and for the synthesis of other 1-methylisoquinolines. We could
upon addition of formic acid for accelerating hydrogenolysis demonstrate that the aminomethylation of metalated arenes with
132

Beilstein J. Org. Chem. 2018, 14, 130–134.
Scheme 2: Total synthesis of alkaloid 1 via the aminomethyl intermediate 5 and selectivity’s found in debenzylation experiments.
2. Menachery, M. D.; Lavanier, G. L.; Wetherly, M. L.; Guinaudeau, H.;
Shamma, M. J. Nat. Prod. 1986, 49, 745–778.
doi:10.1021/np50047a001
Eschenmoser’s salt followed by hydrogenolytic cleavage is a
highly attractive alternative to direct ring methylations. Further,
new insights into selectivity of O- and N-debenzylation reac-
tions should be useful for future natural product and drug syn-
theses.
3. Qing, Z.-X.; Yang, P.; Tang, Q.; Cheng, P.; Liu, X.-B.; Zheng, Y.-j.;
Liu, Y.-S.; Zeng, J.-G. Curr. Org. Chem. 2017, 21, 1920–1934.
doi:10.2174/1385272821666170207114214
4. Roesch, E. S. Isoquinolines. In Privileged Scaffolds in Medicinal
Chemistry: Design, Synthesis, Evaluation; Bräse, S., Ed.; Drug
Discovery, Vol. 50; Royal Society of Chemistry, 2016.
doi:10.1039/9781782622246-00147
Experimental
For experimental procedures and copies of 1H and 13C NMR
spectra of all compounds see Supporting Information File 1.
5. Tiwari, V. K.; Pawar, G. G.; Jena, H. S.; Kapur, M. Chem. Commun.
2014, 50, 7322–7325. doi:10.1039/c4cc03165b
6. Kraskovskiy, A.; Kraskovskaya, V.; Knochel, P. Angew. Chem., Int. Ed.
2006, 45, 2958–2961. doi:10.1002/anie.200504024
7. Melzer, B.; Bracher, F. Org. Biomol. Chem. 2015, 13, 7664–7672.
doi:10.1039/C5OB00926J
Supporting Information
Supporting Information File 1
Experimental part and NMR spectra.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-14-8-S1.pdf]
8. Melzer, B. C.; Bracher, F. Beilstein J. Org. Chem. 2017, 13,
1564–1571. doi:10.3762/bjoc.13.156
9. Kim, J.; Cho, S. H. Synlett 2016, 27, 2525–2529.
doi:10.1055/s-0036-1588073
10.Chen, J.-J.; Tsai, I.-L.; Ishikawa, T.; Wang, C.-J.; Chen, I.-S.
Phytochemistry 1996, 42, 1479–1484.
Acknowledgements
doi:10.1016/0031-9422(96)00123-9
The authors thank Prof. Dr. Eberhard Reimann for helpful
11.Franck, B.; Blaschke, G. Justus Liebigs Ann. Chem. 1963, 668,
145–164. doi:10.1002/jlac.19636680117
discussions and Susanne Gerndt for support in benchwork.
12.Bruderer, H.; Brossi, A. Helv. Chim. Acta 1965, 48, 1945–1956.
doi:10.1002/hlca.19650480814
ORCID® iDs
13.Merz, K. W.; Fink, J. Arch. Pharm. 1956, 289, 347–358.
doi:10.1002/ardp.19562890705
Jan G. Felber - https://orcid.org/0000-0002-5010-9624
Franz Bracher - https://orcid.org/0000-0003-0009-8629
14.Reimann, E.; Renz, H. Arch. Pharm. 1993, 326, 253–258.
doi:10.1002/ardp.19933260502
References
1. Phillipson, J. D.; Roberts, M. F.; Zenk, M. H. The Chemistry and
Biology of Isoquinoline Alkaloids; Springer-Verlag: Berlin, Heidelberg,
1985. doi:10.1007/978-3-642-70128-3
15.Erdik, E. Tetrahedron 1984, 40, 641–657.
doi:10.1016/S0040-4020(01)91092-0
133

Beilstein J. Org. Chem. 2018, 14, 130–134.
16.Deslongchamps, P.; Bélanger, A.; Berney, D. J. F.; Borschberg, H.-J.;
Brousseau, R.; Doutheau, A.; Durand, R.; Katayama, H.; Lapalme, R.;
Leturc, D. M.; Liao, C.-C.; MacLachlan, F. N.; Maffrand, J.-P.;
Marazza, F.; Martino, R.; Moreau, C.; Ruest, L.; Saint-Laurent, L.;
Saintonge, R.; Soucy, P. Can. J. Chem. 1990, 68, 115–126.
doi:10.1139/v90-021
17.Sato, T.; Yoshida, T.; Al Mamari, H. H.; Ilies, L.; Nakamura, E.
Org. Lett. 2017, 19, 5458–5461. doi:10.1021/acs.orglett.7b02778
18.Bracher, F.; Puzik, A. J. Heterocycl. Chem. 2004, 41, 173–176.
doi:10.1002/jhet.5570410206
19.Caporaso, R.; Manna, S.; Zinken, S.; Kochnev, A. R.;
Lukyanenko, E. R.; Kurkin, A. V.; Antonchick, A. P. Chem. Commun.
2016, 52, 12486–12489. doi:10.1039/C6CC07196A
20.Möhrle, H.; Folttmann, H. Arch. Pharm. 1988, 321, 259–261.
doi:10.1002/ardp.19883210504
21.Wardle, R. B.; Edwards, W. W. Hydrogenolysis of
2,4,6,8,10,12-hexabenzyl-2,4,6,8,10,12-hexaazatetracyclo[5.5.0.05,9.0
3,11]dodecane. U.S. Patent US5,739,325, April 14, 1998.
22.Birkofer, L. Ber. Dtsch. Chem. Ges. B 1942, 75, 429–441.
doi:10.1002/cber.19420750420
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.14.8
134