HETEROCYCLES

Paper | Regular issue | Vol. 87, No. 3, 2013, pp. 551-558
Received, 29th August, 2012, Accepted, 29th January, 2013, Published online, 5th February, 2013.
DOI: 10.3987/COM-12-12574

■ Synthesis and Biological Activity of Novel Heterocyclic Ring Systems: Imidazo[4’,5’:3,4]pyrido[2,1-a]isoquinolines and Imidazo[4,5-f][3]benzazecines

Robert Otto, Christoph Enzensperger, and Jochen Lehmann*

Institute of Pharmacy, Friedrich-Schiller-University, Philosophenweg 14, 07743 Jena, Germany

Abstract

Derivatives of two novel heterocyclic ring systems were synthesized and their affinities for dopamine receptors were measured. The compounds were obtained by reacting histamine with 2-(2-bromoethyl)benzaldehyde including an atypical Pictet-Spengler condensation, which afforded basic and not the usual neutral or acidic conditions. The resulting imidazo[4',5':3,4]pyrido[2,1-a]-isoquinoline derivative 4 was Boc protected at the most basic imidazole nitrogen, the isoquinoline nitrogen then quaternized by using methyl iodide and the tetracyclic isoquinolinium salt was both deprotected and cleaved under Birch conditions in one step to give a tricyclic imidazo[4,5-f][3]benzazecine derivative (3) by opening two 6-membered heterocycles towards one 10-membered. Radioligand binding studies showed a significant affinity of the moderately constrained 3 but not of 4 for dopamine receptors. Similar to the analogous indolo-benzazecine LE300, a preference of 3 for the D1 receptor family was observed, but with some loss of affinity over all.

INTRODUCTION
Dopamine is a key neurotransmitter in the brain with various physiological functions, e.g. regulation of locomotion, cognition, emotion and event prediction, and dopamine receptor antagonists play a crucial role in the treatment of neuropsychiatric diseases. Azecine-type dopamine receptor antagonists, with 7-methyl-6,7,8,9,14,15-hexahydro-5H-indolo[3,2-f][3]benzazecine (1, LE 300) as a lead, represent a chemically new class of potential antipsycotics with high affinities for dopamine receptors and a rather unique selectivity profile, showing prevalence for the dopamine D1 subtype receptor family.1-3 Both, the annulated azecine ring itself, but also a moderately rigidized phenylethyl- or heteroarylethyl-amine partial structures are considered to be essential for the biological activity,4 so further variations of these scaffolds are interesting. The azecine 1 and its precursor quinolizine 2 have been obtained in our group by a Bischler-Napieralski cyclization but synthesis has been recently improved by performing a Pictet- Spengler condensation of tryptamine and 2-(2-bromoethyl)benzaldehyde (Scheme 1).5

Changing the biogenic amines from tryptamine to histamine should give the novel heterocycles 4 and 3.

RESULTS AND DISCUSSION
Compound 3 combines structural aspects of both histamine and phenylethylamine in one heterocycle and hence might have interesting biological activities. We refluxed the histamine free base with 2-(2-bromoethyl)benzaldehyde6 in dichloromethane for 24 h (Scheme 2) which was expected to form the imidazo[4',5':3,4]pyrido[2,1-a]isoquinoline 4 by Pictet-Spengler cyclization, since many analogous indole- or thiophene-ethylamines have reacted this way.7 But in this particular case of histamine, the reaction stopped at the intermediate Schiff base 5 which did not undergo any further spontaneous ring closure (Scheme 2).

Pictet-Spengler condensations generally undergo various steps and the final cyclization of the intermediate Schiff base occurs either spontaneously or can be promoted by heating under acidic conditions.7 With histamine and 2-(2-bromoethyl)benzaldehyde these conditions did not lead to the fully cyclized product. We tried various solvents and pH ranges from neutral to acidic. Even with very strong acids like TFA in dichloromethane -these conditions easily induce cyclizations with tryptamine5- no reaction of 5 towards 4 was observed. But surprisingly, we detected the desired cyclized isoquinoline 4 by GC/MS analysis after preparing a sample of 5 for GC/MS analysis by adding aqueous ammonia solution and extracting with ethyl acetate. So we found and reconfirmed after repeating and upscaling that in this particular case a Pictet-Spengler condensation was achieved not under acidic but rather under basic conditions. Actually, a base-promoted (excess of Et3N) Pictet-Spengler reaction for serotonin has been reported8 and other authors reacted histamine with carbonyl compounds in ethanolic KOH.9 So we applied these conditions from the beginning by reacting histamine and 2-(2-bromoethyl)benzaldehyde in ethanol with an excess of Et3N, or alternatively by using ethanolic KOH. In both cases only traces (~ 0.4%) or no 4 at all were detected (GC/MS). Since the preparation of the intermediate 5 under non-acidic conditions showed to be simple, we just dissolved 5 in water and added aqueous conc. ammonia. Cyclization took place within less than 5 min. and 4 could be extracted with ethyl acetate.

Annulated isoquinolines as precursors for the N-methylated azecines were usually quaternized at the quinolizine nitrogen and then the central C-N bond cleaved with sodium in liq. ammonia. To prevent any alkylation of 4 at the imidazole, a Boc group was introduced giving the tert-butyl imidazo[4',5':3,4]pyrido[2,1-a]isoquinoline-3(5H)-carboxylate 6. It is not obvious at which of the imidazole nitrogens the Boc-group is attached, but only one regioisomer was formed. Comparison with calculated spectra (ACD-Chem-Sketch, 1H-NMR Predictor) suggests that the BOC group is at the more exposed position 3. The chemical shift for the proton attached to carbon 12b was calculated with δ = 5.12 ppm for the 3-Boc isomere and δ = 4.72 ppm for the 1-Boc isomere and it was found at δ = 5.04 ppm. So there is some preference for the 3-Boc compound. This preference is confirmed by the 13C shift for imidazole carbon 3a which was calculated with δ = 139.63 ppm for 1-Boc and δ = 124.35 ppm for 3-Boc and was found to be δ = 123.9 ppm. The position of the Boc group at the intermediates is not relevant in view of the desired target compounds; nevertheless, not only steric considerations but also NMR data strongly suggest the 3-Boc intermediates. The Boc-protected isoquinoline 6 was quaternized with methyl iodide yielding the corresponding isoquinolinium iodide 7 (Scheme 3). When performing the finalizing cleavage of the quinolizine C-N-bond towards the azecine ring under Birch conditions the Boc group fortunately was also cleaved off although it is described to be stable against Na in liq. ammonia.10,11

BIOLOGICAL ACTIVITIES
The affinities of the target compounds 3 and 4 for stably cloned human dopamine receptors were evaluated by radioligand binding experiments using a protocol which we have described previously.12 The Ki values (low value means high affinity) are given in Table 1.

The azecine 3 is the more active compound compared to its precursor 4, which is in line to all previous data comparing annulated isoquinoline and azecine derivatives. Target compound 3 is less affine compared to LE300 but still offers submicromolecular affinities for D1 and D5.

EXPERIMENTAL
Melting points are uncorrected and were measured in open capillary tubes, using a Gallenkamp melting point apparatus. 1H- and 13C-NMR spectral data were obtained from a Bruker Advance 250 spectrometer (250 MHz) and Advance 400 spectrometer (400 MHz). TLC was performed on silica gel F254 plates (Merck). MS data were determined by GC/MS, using a Hewlett Packard GCD-Plus (G1800C) apparatus (HP-5MS column; J&W Scientific). High resolution mass spectrometry (HRMS) data were determined on a TSQ Quantum Access Mass Spectrometer (Therma Electron Corporation). Purities of the compounds were determined by elemental analysis, performed on a Hereaus Vario EL apparatus. All values for C, H, and N were found to be within ± 0.4. All compounds showed > 95% purity.

2-[2-(1H-Imidazol-5-yl)ethyl]-3,4-dihydroisoquinolinium bromide hydrobromide (5)
In a 250 mL round bottom flask 610 mg (5 mmol) of histamine free base (purchased from Sigma Aldrich) and 1.4 g (6 mmol) of 2-(2-bromoethyl)- benzaldehyde6 were dissolved in 130 mL CH2Cl2 and refluxed for 24 h. The mixture was cooled to room temperature and the precipitated light yellow solid was filtered off, dried under reduced pressure and recrystallized from isopropanol to give 5 as white crystals. (950 mg, 49%); mp 213 °C; 1H NMR (DMSO-d6, 250 MHz): δ (ppm) 3.24 (t, J = 8.0 Hz, 2H, Hb), 3.35 (t, J = 6.9 Hz, 2H, H4),

4.12 (t, J = 8.0 Hz, 2H, Ha), 4.31 (t, J = 6.9 Hz, 2H, H3), 7.53 (m, 2H, H5 and H6), 7.63 (s, 1H, H4’), 7.82 (m, 2H, H8 and H7), 9.12 (s, 1H, H2’), 9.36 (s, 1H, H1); 13C NMR (DMSO-d6): δ (ppm) 22.76, 24.92, 48.35, 58.78, 117.71, 125.07, 128.56, 128.73, 128.82, 134.01, 135.09, 137.23, 138.19, 167.38; Anal. Calcd for C14H17Br2N3: C, 43.4; H, 4.43; N, 10.9. Found: C, 43.1; H, 4.62; N, 10.8. HRMS calcd for C14H16N3: 226.13442. Found: 226.13426.

3,4,5,7,8,12b-Hexahydroimidazo[4',5':3,4]pyrido[2,1-a]isoquinoline (4)
In a 100 mL round bottom flask 810 mg 5 (2.1 mmol) were dissolved in 20 mL water and conc. ammonia solution was added to adjust the pH at 10. After 5 min, the mixture was extracted with EtOAc (3 x 15 mL). The combined organic phases were dried over Na2SO4, evaporated and dried under reduced pressure to give 4 as a yellow amorphous solid. (320 mg, 67%) For analytical purpose a dihydrochloric salt was formed by dissolving 4 in isopropanol and adding conc. HCl dropwise until the pH reached 5. The formed salt

was collected and recrystallized from acetone/water. mp 275 °C; 1H NMR (DMSO-d6, 400 MHz): δ (ppm) 3.04 (m, 2H, H8), 3.19 (m, 2H, H4), 3.51 (m, 2H, H5), 3.71 (m, 2H, H7), 6.04 (s, 1H, H12b), 7.32-7.36 (m, 3H, H9-11), 7.50 (d, J = 7.1 Hz, 1H, H12), 9.06 (s, 1H, H2); 13C NMR (DMSO-d6): δ (ppm) 17.77, 23.68, 46.26, 46.80, 54.40, 123.90, 124.58, 127.65, 128.21, 128.95, 129.09, 129.50, 131.41, 135.81; Anal. Calcd for C14H17Cl2N3 + 1/3 H2O: C, 55.3; H, 5.85; N, 13.8. Found: C, 55.1; H, 5.58; N, 13.6. HRMS Calcd for C14H16N3: 226.13442. Found: 226.13519.

References

1. M. Decker, K. J. Schleifer, M. Nieger, and J. Lehmann, Eur. J. Med. Chem., 2004, 39, 481. CrossRef
2. B. Hoefgen, M. Decker, P. Mohr, A. M. Schramm, S. A. Rostom, H. El-Subbagh, P. M. Schweikert, D. R. Rudolf, M. U. Kassack, and J. Lehmann, J. Med. Chem., 2006, 49, 760. CrossRef
3. T. Witt, F. J. Hock, and J. Lehmann, J. Med. Chem., 2000, 43, 2079. CrossRef
4. C. D. Duarte, E. J. Barreiro, and C. A. Fraga, Mini Rev. Med. Chem., 2007, 7, 1108. CrossRef
5. D. Robaa, R. Kretschmer, O. Siol, S. E. Abulazm, E. Elkhawass, J. Lehmann, and C. Enzensperger, Arch. Pharm. (Weinheim), 2011, 344, 28.. CrossRef
6. P. C. B. Page, G. A. Rassias, D. Barros, D. Bethell, and M. B. Schilling, J. Chem. Soc. Perkin Trans. 1, 2000, 3325. CrossRef
7. E. D. Cox and J. M. Cook, Chem. Rev., 1995, 95, 1797. CrossRef
8. K. Yamada, Y. Namerikawa, T. Haruyama, Y. Miwa, R. Yanada, and M. Ishikura, Eur. J. Org. Chem., 2009, 5752. CrossRef
9. N. N. Smolyar, M. G. Abramyants, T. I. Zavyazkina, D. I. Matveeva, Y. S. Borodkin, and I. A. Voloskii, Russ. J. Org. Chem., 2009, 45, 1219. CrossRef
10. Organic Chemistry Portal, http://www.organic-chemistry.org/protectivegroups/amino/boc-amino.htm
.
11. T. W. Green, and P. G. M. Wuts, Protective Groups in Organic Synthesis, Wiley-Interscience, New York, 1999, p. 518 to 525 and 736 to 739.
12. C. Enzensperger, F. K. Muller, B. Schmalwasser, P. Wiecha, H. Traber, and J. Lehmann, J. Med. Chem., 2007, 50, 4528. CrossRef