Access to high value sp³-rich frameworks using photocatalyzed [2 + 2]-cycloadditions of γ-alkylidene–γ-lactams

Abstract

By harnessing an energy transfer process, new photocatalyzed [2 + 2]-cycloadditions occurring between γ-alkylidene–γ-lactams and unsaturated substrates have been developed. The reaction mode is particularly powerful because it leads to the formation of different high value sp³-rich frameworks and further diversity can be introduced through cascade sequences wherein strain releasing opening of the cyclobutane intermediates gives access to complex polycyclic alkaloid frameworks.

---

Achieving greater structural novelty¹ and a higher sp³ carbon count² have been identified as key ways to improve both the quality of pharmaceutical hits and the frequency with which they can be found. Thus, these two features have become essential prerequisites when developing new screening sets. As a result, three-dimensionally rich spirocyclic compounds³ have attracted particular attention. Interest that has been bolstered by the fact that the rigidity of these highly saturated frameworks endows the compounds with further benefits; such as, the lowering of the entropy penalty incurred when docking at a receptor site^3a and the potential to reduce receptor promiscuity.^1,2b

During our work on functionalizing γ-lactams,⁴ we noticed that lactams of type 1 (1a or 1b, Scheme 1A) had an inclination to isomerize substantially (from E to Z) when irradiated with blue light in the presence of [Ru(bpy)₃]Cl₂ (1a/b → 2a/b). This behavior was also seen with Ir(ppy)₃, albeit with lower final Z/E ratios; whereas, the direct irradiation of 1 without a catalyst led to only 8.5% isomerization. Lactam 1a′ proved to be inert to the isomerization conditions. Interestingly, this observation is in keeping with the known isomerization of the structurally similar heam-metabolite, bilirubine, which forms the basis of light therapy for neonatal jaundice.⁵ Thus, we began to develop a strategy predicated on the hypothesis that this isomerization involved an energy transfer reaction which might be harnessed to give a series of novel photocatalytic [2 + 2]-cycloadditions.

Scheme 1 (A) Observed E–Z isomerization of unsaturated lactams of type 1. (B) Photocatalytic [2 + 2]-cycloaddition reaction modes.

The initial target groups would be architecturally complex sp³-rich cyclobutanes formed by intra- or intermolecular reactions. Cyclobutanes not only feature in many biologically active compounds,⁶ but are also useful intermediates due to their potential to undergo ring strain releasing reactions.⁷ There are other methods to make cyclobutanes,⁸ but these are mostly eclipsed by the widespread use of [2 + 2]-photocycloadditions.⁹ Within this class the historic variant involves the olefin absorbing UV light directly.⁹ Milder and newer methods using visible light and a photocatalyst operate either via an energy transfer (EnT)¹⁰ or single electron transfer (SET) mechanisms.¹¹ The photocatalytic variants require some form of extended conjugation because it affords the substrates with lower triplet state energies and lower redox potentials to facilitate the generation of stabilized open shell intermediates (resonance stabilization, Scheme 1B). This feature has, therefore, often limited reaction scope to substrates bearing aromatic substituents. Photocatalyzed isomerization of conjugated alkenes from E to Z is also an energy transfer process sometimes used to access cis alkenes from their more readily synthesized trans analogs.¹² It was for this reason that we believed the previously observed isomerization might indicate that γ-alkylidene–γ-lactams could be uniquely suitable non-aromatic substrates for photocatalyzed [2 + 2]-cycloadditions to form new cyclobutanes via an energy transfer mechanism. Notably, a number of the N-substituted cyclobutane products would share key structural features with natural alkaloids and pharmaceuticals.^3a,13

We began our investigation with the facile synthesis of the γ-alkylidene–γ-lactam 4a from furan 3a and allylamine using our previously developed photocatalytic protocol (Scheme 2).^4f Substrate 4a has a double bond on the amide side chain that could partner with the exocyclic double bond in a cross [2 + 2]-cyclization to yield the tightly packed sp³-rich polycycle 5a. Indeed, when 4a was treated with PC1 in CH₃CN (0.1 M) and irradiated with blue LEDs, the desired reaction occurred; however, the reactions did not reach completion even after 24 h (entries 1 and 2). In surveying photocatalysts, it was found that organic dyes PC2, PC3 and methylene blue (MB), all of which have only moderate triplet energies,¹⁴ were not competent catalysts for the reaction (entries 3–5). In contrast, PCs 4 and 5, which have high triplet energies,¹⁴ proved to be highly efficient converting 4a into 5a within 8 h in high yield (92 and 94% yield, respectively). The reaction did not proceed in the absence of the photocatalyst or light; although with the former, a small degree of double bond isomerisation did take place (E : Z, 2.4 : 1, entry 8). Extended conjugation in the substrates is a requirement as shown by the inert nature of 4a′ to the reaction conditions (entry 10). We expanded the set of substrates 4 and showed that they all reacted efficiently (85–95% yield, 1.6 : 1–2 : 1 dr, Scheme 3). Substrates that include groups sometimes sensitive to radicals and/or other photocatalyzed reactions they also successfully provided the desired products (5d and 5e). Interestingly, framework 5 contains the skeleton of the natural amino acid 2,4-methanoproline,¹⁵ which, along with its analogs, has been shown to exhibit a range of biological activities.¹⁶ It was also possible to relocate the partner double bond in these intramolecular [2 + 2]-photocycloadditions; for example, substrates of type 6 afforded polycyclic products 7 with excellent diastereoselectivity (7a–7c, 92–96% yield, Scheme 3).

Scheme 2 Optimization of the intramolecular [2 + 2]-cycloaddition of 4a. ^aDetermined by ¹H-NMR of the crude reaction mixture. ^bIsolated yield. ^cE/Z isomerization was observed (E/Z = 2.4 : 1).

Scheme 3 Photocatalytic intramolecular [2 + 2]-cycloaddition reactions of unsaturated lactams of type 4 and 6.

Next, we sought to move on to the more challenging intermolecular variant of the reaction in which the γ-alkylidene–γ-lactam would react with a second molecule containing an electron deficient double bond. The reaction would give us access to relatively rare 5,4-spirocycles of type 9 (Scheme 4). A variety of combinations were tested in which lactams 8 were combined with an excess (10 equiv.) of the unsaturated carbonyl compound under the previously optimized conditions (0.5% of PC5 in CH₃CN). All the reactions worked well, affording the products 9 in good to high yield (9a–9g, 60–82%) and with excellent regioselectivity. Despite the fact that every reaction furnished a mixture of diastereomers (1.6 : 1, by ¹H NMR), the products with a keto or aldehyde group were epimerized to a single stereoisomer upon chromatographic purification (9a–9c, 9e–9g). Only in the case of 9d (bearing an ester group less able to drive epimerization) did the reaction afford a 1 : 1 ratio of diastereomers which remained unchanged during purification. The intermolecular [2 + 2]-cycloaddition with the more electron-rich alkene styrene also works, but it is quite messy.

Scheme 4 Photocatalytic synthesis of spirocyclic compounds of type 9via intermolecular [2 + 2]-cycloaddition. ^aThe reaction occurred similarly using 0.5% of PC4.

An interesting result emerged when product 9g was treated with formic acid. The acidic conditions catalyzed a retro-Mannich reaction, presumably, driven by opening of the strained cyclobutane ring, which was followed by nucleophilic attack of the aromatic group on the resulting N-acyliminium cation, to furnish lactam 10a as a sole product (92% yield). Such N-containing aromatic polycycles of type 10 constitute the basic skeleton of many natural alkaloids (10a has been used as a precursor for the synthesis of erysotramidine, Scheme 5).^17a,b The retro-Mannich ring opening of 9g⁷ was intriguing because it opened up the possibility for developing cascade sequences that could diversify the type of products accessible to us through this methodology (8 → 10–12, Scheme 5). In practice, lactams 8, with internal nucleophilic groups at either R¹ or R², first underwent the [2 + 2]-photocycloaddition using an excess (10 equiv.) of acrolein or methyl vinyl ketone. The intermediates of type 9 were not isolated, but, instead, acid was added to attain a range of alternative products. More specifically, when R¹ bore an electron rich aromatic group, treatment with formic acid catalyzed further transformation into products of type 10 (overall 8 → 10a–10c in one pot with yields ranging from 60 to 69%). Similarly, when an appropriate nucleophile was appended to R², various products could be attained depending on the nature of that nucleophile. For example, spirocycles 11a and 11b were formed diastereoselectively in good yields in a single operation starting from lactam 8 (for 11a formic acid was used while for 11b PTSA was employed, Scheme 5). These compounds constitute the skeleton of marine Clavelina^17c and marineosins^17d alkaloids. Furthermore, chain homologation (a form of C–H activation) could be achieved for substrates with no internal nucleophile through a similar acid catalyzed cyclobutane ring opening reaction sequence. Thus, lactams of type 12 could be synthesized in one pot (8 → 12a–12c, yields 61–72%, Scheme 5). The retro-Mannich ring opening does not work with ester 9d, because the ester group is less electron withdrawing compared to aldehydes and ketones.

Scheme 5 Functionalization of γ-alkylidene–γ-lactams based on photocatalytic [2 + 2]-cyclization followed by cyclobutane ring opening. ^aThe second step was performed using PTSA·H₂O (2 equiv.) in CH₂Cl₂.

Finally, we wanted to investigate the reaction of γ-alkylidene–γ-lactams with 2,3-dimethylbuta-1,3-diene. 1,3-Dienes have previously been utilized in photocycloadditions, producing either vinylcyclobutans or cyclohexenes depending on which mechanism is in operation (EnT or ET).¹⁸ When the optimized conditions were applied to conjugated lactams of type 8 in the presence of 2,3-dimethylbuta-1,3-diene (4 equiv., Scheme 6), spirocyclic compounds of type 13 were formed as the sole products (13a–13c, 70–74% yield, Scheme 6). Monitoring the reaction by ¹H NMR, we observed the formation of the vinylcyclobutane 9h (Scheme 6) during the early stages of the process. This intermediate subsequently disappeared and spirocyclic product 13 was formed. This observation implies two sequential steps; an initial [2 + 2]-cycloaddition yielding 9h with subsequent rearrangement to 13 to give overall a [4 + 2]-transformation.

Scheme 6 Photocatalytic cyclization of unsaturated γ-alkylidene–γ-lactams with 2,3-dimethylbuta-1,3-diene. ^aThe reactions occurred similarly using 0.5% of PC4.

Stern–Volmer quenching and voltammetry studies were undertaken (see, ESI‡) and the results were consistent with an energy transfer from the excited state of the photocatalyst to the γ-alkylidene–γ-lactam being the mechanistic mode operating in these [2 + 2] cycloadditions. More precisely, compound 8a quenches the excited state of PC5 at a significantly higher rate than methylvinyl ketone or 2,3-dimethylbuta-1,3-diene and the redox potentials of PC4 and PC5 are not appropriate for initiation of an electron transfer pathway with 4a (see ESI‡).

Overall, a series of mild and highly efficient methodologies to access a diverse range of unusual rigid sp³-rich spirocycles, complex alkaloid frameworks or chain homologated products have been developed. Synthesis of the latter two groups was achieved by incorporating ring strain relieving cyclobutane opening into highly effective one pot cascade reaction sequences. The methodologies all rely initially on a novel photocatalyzed (visible light + PC) [2 + 2]-cycloaddition between γ-alkylidene–γ-lactams and an unsaturated partner which occurs via an energy transfer mechanism.

This research has been co-financed by the European Regional Development Fund of the European Union and Greek national funds through the Operational Program Competitiveness, Entrepreneurship and Innovation, under the call RESEARCH – CREATE – INNOVATE (project code:T2EDK-02364). We thank the Mass Spectrometry Facility of the University of Crete for obtaining the HRMS data. We are grateful to Prof. Athanassios G. Coutsolelos and Dr Georgios Charalambidis for help with voltammetry and Stern–Volmer studies.

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. J.-L. Reymond, Acc. Chem. Res., 2015, 48, 722.
2. (a) F. Lovering, J. Bikker and C. Humblet, J. Med. Chem., 2009, 52, 6752; (b) P. A. Clemons, N. E. Bodycombe, H. A. Carrinski, J. A. Wilson, A. F. Shamji, B. K. Wagner, A. N. Koehler and S. L. Schreiber, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 18787.
3. (a) E. M. Carreira and T. C. Fessard, Chem. Rev., 2014, 114, 8257; (b) Y. Zheng, C. M. Tice and S. B. Singh, Bioorg. Med. Chem. Lett., 2014, 24, 3673; (c) K. Hiesinger, D. Dar’in, E. Proschak and M. Krasavin, J. Med. Chem., 2021, 64, 150.
4. (a) D. Kalaitzakis, A. Bosveli, K. Sfakianaki, T. Montagnon and G. Vassilikogiannakis, Angew. Chem., Int. Ed., 2021, 60, 4335; (b) S. Hoxha, D. Kalaitzakis, A. Bosveli, T. Montagnon and G. Vassilikogiannakis, Org. Lett., 2021, 23, 5354; (c) T. Montagnon, D. Kalaitzakis, M. Sofiadis and G. Vassilikogiannakis, Org. Biomol. Chem., 2020, 18, 180; (d) G. I. Ioannou, T. Montagnon, D. Kalaitzakis, S. A. Pergantis and G. Vassilikogiannakis, ChemPhotoChem, 2018, 2, 860; (e) D. Kalaitzakis, M. Triantafyllakis, G. I. Ioannou and G. Vassilikogiannakis, Angew. Chem., Int. Ed., 2017, 56, 4020; (f) D. Kalaitzakis, A. Kouridaki, D. Noutsias, T. Montagnon and G. Vassilikogiannakis, Angew. Chem., Int. Ed., 2015, 54, 6283.
5. (a) R. Bonnett, J. E. Davies and M. B. Hursthouse, Nature, 1976, 262, 327; (b) K. Mreihil, A. F. McDonagh, B. Nakstad and T. W. R. Hansen, Pediatr. Res., 2010, 67, 656.
6. (a) J. Li, K. Gao, M. Bian and H. Ding, Org. Chem. Front., 2020, 7, 136; (b) M. Wang and P. Lu, Org. Chem. Front., 2018, 5, 254.
7. J. D. Winkler, C. M. Bowen and F. Liotta, Chem. Rev., 1995, 95, 2003.
8. (a) M. R. Fructos and A. Prieto, Tetrahedron, 2016, 72, 355; (b) Y. Xu, M. L. Conner and M. K. Brown, Angew. Chem., Int. Ed., 2015, 54, 11918.
9. (a) N. Hoffmann, Chem. Rev., 2008, 108, 1052; (b) M. D. Kärkäs, J. A. Porco and C. R. J. Stephenson, Chem. Rev., 2016, 116, 9683; (c) S. Poplata, A. Tröster, Y.-Q. Zou and T. Bach, Chem. Rev., 2016, 116, 9748; (d) B. Cox, K. I. Booker-Milburn, L. D. Elliott, M. Robertson-Ralph and V. Zdorichenko, ACS Med. Chem. Lett., 2019, 10, 1512.
10. For selected reviews see: (a) F. Strieth-Kalthoff, M. J. James, M. Teders, L. Pitzer and F. Glorius, Chem. Soc. Rev., 2018, 47, 7190; (b) Q.-Q. Zhou, Y.-Q. Zou, L.-Q. Lu and W.-J. Xiao, Angew. Chem., Int. Ed., 2019, 58, 1586 ; For selected non-enantioselective examples see: ; (c) Z. Lu and T. P. Yoon, Angew. Chem., Int. Ed., 2012, 51, 10329; (d) Y.-Q. Zou, S.-W. Duan, X.-G. Meng, X.-Q. Hua, S. Gao, J.-R. Chen and W.-J. Xiao, Tetrahedron, 2012, 68, 6914; (e) A. E. Hurtley, Z. Lu and T. P. Yoon, Angew. Chem., Int. Ed., 2014, 53, 8991; (f) V. Mojr, E. Svobodová, K. Straková, T. Neveselý, J. Chudoba, H. Dvořáková and R. Cibulka, Chem. Commun., 2015, 51, 12036; (g) J. Zhao, J. L. Brosmer, Q. Tang, Z. Yang, K. N. Houk, P. L. Diaconescu and O. Kwon, J. Am. Chem. Soc., 2017, 139, 9807; (h) M. J. James, J. L. Schwarz, F. Strieth-Kalthoff, B. Wibbeling and F. Glorius, J. Am. Chem. Soc., 2018, 140, 8624; (i) R. F. Higgins, S. M. Fatur, N. H. Damrauer, E. M. Ferreira, A. K. Rappé and M. P. Shores, ACS Catal., 2018, 8, 9216; (j) F. M. Hörmann, T. S. Chung, E. Rodriguez, M. Jakob and T. Bach, Angew. Chem., Int. Ed., 2018, 57, 827; (k) T. Neveselý, C. G. Daniliuc and R. Gilmour, Org. Lett., 2019, 21, 9724; (l) L. D. Elliott, S. Kayal, M. W. George and K. Booker-Milburn, J. Am. Chem. Soc., 2020, 142, 14947.
11. For selected examples see: (a) N. L. Bauld, Tetrahedron, 1989, 45, 5307; (b) M. A. Miranda and H. García, Chem. Rev., 1994, 94, 1063; (c) T. P. Yoon, ACS Catal., 2013, 3, 895; (d) M. Riener and D. A. Nicewicz, Chem. Sci., 2013, 4, 2625.
12. For relative reviews see: (a) Y. Yokoyama, Chem. Rev., 2000, 100, 1717; (b) J. B. Metternich and R. Gilmour, Synlett, 2016, 2541; (c) J. J. Molloy, T. Morack and R. Gilmour, Angew. Chem., Int. Ed., 2019, 58, 13654.
13. (a) V. M. Dembitsky, J. Nat. Med., 2008, 62, 1; (b) V. M. Dembitsky, Phytomedicine, 2014, 21, 1559; (c) A. Sergeiko, V. V. Poroikov, L. O. Hanuš and V. M. Dembitsky, Open J. Med. Chem., 2008, 2, 26.
14. For the triplet energy of the current applied photosensitizers see: (a) For PC1: D. P. Rillema, G. Allen, T. J. Meyer and D. Conrad, Inorg. Chem., 1983, 22, 1617; (b) For PC2, PC3 and MB: T. Shen, Z.-G. Zhao, Q. Yu and H.-J. Xu, J. Photochem. Photobiol., A, 1989, 47, 203; (c) For PC4: J. D. Slinker, A. A. Gorodetsky, M. S. Lowry, J. Wang, S. Parker, R. Rohl, S. Bernhard and G. G. Malliaras, J. Am. Chem. Soc., 2004, 126, 2763; (d) For PC5: A. Singh, K. Teegardin, M. Kelly, K. S. Prasad, S. Krishnan and J. D. Weaver, J. Organomet. Chem., 2015, 776, 51.
15. (a) E. A. Bell, M. Y. Qureshi, R. J. Pryce, D. H. Janzen, P. Lemke and J. Clardy, J. Am. Chem. Soc., 1980, 102, 1409; (b) C. V. Stevens, G. Smagghe, T. Rammeloo and N. De Kimpe, J. Agric. Food Chem., 2005, 53, 1945.
16. B. Cox, J. Duffy, V. Zhdorichenko, C. Bellanger, J. Hurcum, B. Laleu, K. I. Booker-Milburn, L. D. Elliot, M. Robertson-Ralph, C. J. Swain, S. J. Bishop, I. Hallyburton and M. Anderson, ACS Med. Chem. Lett., 2020, 11, 2497.
17. (a) M. E. Amer, M. Shamma and A. J. Freyer, J. Nat. Prod., 1991, 54, 329; (b) A. J. Blake, C. Gill, D. A. Greenhalgh, N. S. Simpkins and F. Zhang, Synthesis, 2005, 3287; (c) S. M. Weinreb, Chem. Rev., 2006, 106, 2531; (d) C. Boonlarppradab, C. A. Kauffman, P. R. Jensen and W. Fenical, Org. Lett., 2008, 10, 5505.
18. Energy transfer based cyclizations with 1,3-dienes afford vinylcyclobutanes. For examples see: (a) Ref. 10c, e, j; (b) F. Tang, L. Tang, Z. Guan and Y.-H. He, Tetrahedron, 2018, 74, 6694; (c) T. R. Blum, Z. D. Miller, D. M. Bates, I. A. Guzei and T. P. Yoon, Science, 2016, 354, 1391; (d) Recently (ref. 10i), using isoprene as the 1,3-diene an overall-[4 + 2] reaction was observed.

---

Footnotes

† Dedicated to Prof. Gerasimos J. Karabatsos for his 90th birthday.

‡ Electronic supplementary information (ESI) available: [DETAILS]. See DOI: https://doi.org/10.1039/d2cc03009h

§ These authors have contributed equally to this work.

---