Concise synthesis of 2-N-hydroxy-3,4-dihydroisoquinol-2-one: A bacterial siderophore and human 5-lipooxygenase inhibitor

Article abstract of DOI:10.1016/j.tetlet.2018.03.017

A short synthesis of 2-N-hydroxy-3,4-dihydroisoquinol-2-one has been devised, which is based on the Castagnoli-Cushman reaction and 3 times more high-yielding compared to an earlier reported sequence.

Full text of DOI:10.1016/j.tetlet.2018.03.017

Accepted Manuscript
Concise Synthesis of 2-N-Hydroxy-3,4-dihydroisoquinol-2-one: a Bacterial
Siderophore and Human 5-Lipooxygenase Inhibitor
Anna Tutov, Olga Bakulina, Dmitry Dar'in, Mikhail Krasavin
PII:
S0040-4039(18)30314-9
https://doi.org/10.1016/j.tetlet.2018.03.017
TETL 49786
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
1 February 2018
2 March 2018
6 March 2018
Please cite this article as: Tutov, A., Bakulina, O., Dar'in, D., Krasavin, M., Concise Synthesis of 2-N-Hydroxy-3,4-
dihydroisoquinol-2-one: a Bacterial Siderophore and Human 5-Lipooxygenase Inhibitor, Tetrahedron Letters
(2018), doi: https://doi.org/10.1016/j.tetlet.2018.03.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Concise Synthesis of 2-N-Hydroxy-3,4-dihydroisoquinol-2-one: a Bacterial
Siderophore and Human 5-Lipooxygenase Inhibitor
Anna Tutov,† Olga Bakulina, Dmitry Dar’in and Mikhail Krasavin*
Saint Petersburg State University, Saint Petersburg 199034 Russian Federation
* Corresponding author.
Phone: +7 931 361 7872, fax: +7 812 428 6939, e-mail: mkrasavin@spbu.ru
URL: https://krasavin-group.org
†On research training leave from Leibniz Universität Hannover, Germany.
Abstract: A short synthesis of 2-N-hydroxy-3,4-dihydroisoquinol-2-one has been devised,
which is based on the Castagnoli-Cushman reaction and 3 times more high-yielding compared to
an earlier reported sequence.
Keywords: Castagnoli-Cushman reaction; formaldoxime; decarboxylation; O-debenzylation.
Microbial metabolites excreted by bacteria to scavenge for iron in iron-scarce environments are
designated so-called bacterial siderophores. A typical siderophore molecule comprises of one to
several iron-chelating motifs (such as catechols2 or hydroxamic acids3) and, when, loaded with
an iron cargo, is transported back into the bacteria where it releases its load for the microbe’s
nutritional needs. The respective bacterial transporters often have a high degree of tolerance
which has been exploited in the recent years as a ‘Trojan Horse’ strategy for antibiotic delivery.4
2-N-Hydroxy-3,4-dihydroisoqionol-2-one (1) is a very simple siderophore isolated over 20 years
ago from Streptomyces spec. 2002-104, with confirmed binding affinity to ferric ions (2).5 Three
years earlier, it emerged, as a fully synthetic compound, in a Bristol-Myers Squibb program
directed at developing anti-inflammatory 5-lipooxygenage inhibitors (5-LO).6 Considering that
5-LO contains iron at its active site, it was surprising that the effect of 1 in a whole-cell assay

was quite modest (Fig. 1). Still, a linear four-step synthesis with 13.1% overall yield was
reported.6a
Figure 1. Structure of 1, its 5-LO inhibitory activity and its suggested hexadendate complex 2
with Fe 3+ (only one possible isomer of 2 is shown).
Recently, we developed a unified approach towards cyclic N-hydroxytetrahydroquinolonic acids
3 (related to 1) via the Castagnoli-Cushman reaction (CCR) of homophthalic anhydride (HPA)
and unprotected oximes 4 (Scheme 1).7
Scheme 1. Synthesis of N-hydroxytetrahydroquinolonic acids 3.
The unprotected version of formaldoxime 4a did not undergo the CCR of HPA and needed to be
protected as O-benzyl ether 5. The latter underwent a CCR with HPA and, without further
purification, the resulting adduct 6 was taken through to the decarboxylation step. At this stage, a
42% yield of pure compound 7 was obtained chromatographically along with 5% of an impurity
(for discussion, vide infra). Compound 7 thus obtained was cleanly hydrogenated over 10% Pd
on charcoal to afford target compound 1 in 38% overall yield (Scheme 2).
Scheme 2. Synthesis of 1 developed in this work.
The persistent impurity separated from 7 was identified as 8. Its formation (observed previously
by us8 and others9) can be justified as shown in Scheme 3. Since O-benzylformaldoxime (5) had

been distilled prior to use, participation from adventitious formaldehyde could be excluded. Most
likely, the process was triggered by the Mannich-type addition of HPA onto O-
benzylformaldoxime (5) to give 9. The latter can eliminate O-benzylhydroxylamine to produce
strong Michael acceptor 10 which can react with another HPA molecule and, after skeletal
reorganization within 11 (described in detail from a mechanistic point of view by Bogdanov and
colleagues9), gives tetracyclic by-product 12 (co-eluting with 6 since both are carboxylic acids).
It should be noted that 12 was not isolated and characterized in this work. Instead, joint
decarboxylation of 6 and 12 produced 7 and 8, respectively, which led to easier chromatographic
separation of 8 from 7.
Although being a natural siderophore on its own, 1 does not contain any functionality that would
make it employable in the design of siderophore-antibiotic ‘Trojan Horse’ constructs. Therefore,
we employed preparative HPLC to purify 6 from 8 and obtained compound 13 in 90% yield
after hydrogenolysis (Scheme 4). The latter essentially represents a functionalized natural
siderophore 1 suitable for conjugation to antibiotic drug molecules.
Scheme 3. Formation of by-product 8.
Scheme 4. Production of compound 13.

Supporting Information
Experimental procedures and characterization data (compounds 7, 1, 6, 9, 13); copies of 1H and
13C NMR spectra.
Acknowledgements
This research was supported by the Russian Scientific Fund (project grant 14-50-00069). NMR,
mass spectrometry were performed at the Research Centre for Magnetic Resonance, the Centre
for Chemical Analysis and Materials Research of Saint Petersburg State University Research
Park.
Rereferences and notes
1. Neilands, J. B. 1995, 270, 26723–26726.
2. Wyckoff, E. E.; Allred, B. E.; Raymond, K. N.; Payne, S. M. J. Bacteriol. 2015, 197, 2840-
2849.
3. Ejje, N.; Soe, C. Z.; Gu, J.; Codd, R. Metallomics 2013, 5, 1519-1528.
4. Ferreira, K.; Hu, H.-Y.; Fetz, V.; Prochnow, H.; Rais, B.; Müller, P. P.; Brönstrup, M.
Angew. Chem. Int. Ed. 2017, 56, 8772-8276.
5. Gräfe, U.; Ritzau, M.; Ihn, W.; Möllmann, Fleck, W. F.; Groth, J.; Reissbbrodt, R. J. Basic
Microbiol. 1994, 34, 351-355.
6. (a) Misra, R. N.; Botti, C. M.; Haslanger, M.F.; Engefrecht, J. R.; Mahoney, E. M.; Ciosek,
C. P. Bioorg. Med. Chem. Lett. 1991, 1, 295-298; (b) Misra, R. N. US Pat. 4,757,078;
Chem. Abstr. 1988, 109, 128847.
7. Bakulina, O.; Bannykh, A.; Dar’in, D.; Krasavin, M. Chem. Eur. J. 2017, 23, 17677-17673.
8. Guranova, N.; Dar’in, M.; Krasavin, M. Synthesis 2018, in press.
9. Bogdanov, M. G.; Mitrev, Y.; Tiritiris, I. Eur. J. Org. Chem. 2011, 377-384.

 Shorter, more high-yielding synthesis of an important N-hydroxylactam is described
 The synthesis is based on the Castagnoli-Cushman reaction
 Persistent impurity described in the literature was formed
 The impurity was separated after total decarboxylation of the product mixture
 Carboxylic acid version of 2-N-hydroxy-3,4-dihydroisoquinol-2-one was obtained

Concise Synthesis of 2-N-Hydroxy-3,4-dihydroisoquinol-2-one: a Bacterial
Siderophore and Human 5-Lipooxygenase Inhibitor
Anna Tutov, Olga Bakulina, Dmitry Dar’in and Mikhail Krasavin