Catalytic Enantioselective Desymmetrization of meso-Cyclopropane-Fused Cyclohexene-1,4-diones by a Formal C(sp²)–H Alkylation

Abstract

A bicyclo[4.1.0]heptane framework consisting of cis-fused cyclopropane and cyclohexane rings is found in several bioactive compounds. Given the symmetry of this core, catalytic desymmetrization can be considered as the most straightforward strategy for its enantioselective synthesis. Known desymmetrization reactions of meso-bicyclo[4.1.0]heptane derivatives proceed with opening of the cyclopropane ring. We now report the first ring-retentive desymmetrization of bicyclo[4.1.0]heptane derivatives, namely meso-cyclopropane-fused cyclohexene-1,4-diones, through a formal C(sp²)-H alkylation using a nitroalkane as the alkylating agent. This reaction is catalyzed by a dihydroquinine-derived bifunctional tertiary aminosquaramide and generates the products with up to 97:3 er. An application of this reaction is demonstrated by the first catalytic enantioselective synthesis of the natural product (–)-car-3-ene-2,5-dione.

Key words

organocatalysis - desymmetrization - cyclopropanes - bicyclo­heptanes - fused ring systems

Significant ring strain not only makes small carbocycles structurally rigid but also creates challenges for their construction.[1] Cyclopropane, the smallest carbocycle, has unique structural and electronic properties, and serves as a versatile building block in organic synthesis.[2] Bicyclo[4.1.0]heptane framework 1, consisting of fused cyclopropane and cyclohexane rings, is found in some natural products such as the carenes 2 and 3, bioactive compounds having anti-cancer properties (e.g, 4),[3] and aristolane-type sesquiterpenoids 5 and 6, which have been shown to be potent against neuropsychiatric disorders (Figure [1]).[4] In 1990, Mitsuhashi and co-workers isolated (±)-asarinol A (7), (±)-car-3-ene-2,5-dione (8), and (±)-asarinol B (9) from the roots of Asiasari radix, which is used in Chinese medicine as an anodyne or antitussive.[5]

Figure 1 Selected examples of natural products and bioactive compounds containing the bicyclo[4.1.0]heptane scaffold 1

Among the various methods known for accessing enantioenriched carbocyclic frameworks, desymmetrization of prochiral or meso compounds represents a powerful strategy.[6] Because of the symmetry of the meso-bicyclo[4.1.0]heptane core 1, catalytic enantioselective desymmetrization becomes a viable strategy for accessing these targets. Despite advancements in this area in the past few decades, desymmetrization of meso-bicyclo[4.1.0]heptane derivatives has only been achieved through the ring opening of the cyclopropane ring,[7] leading to mono- or disubstituted cyclohexene or cyclohexane derivatives. In 1994, Troxler and Scheffold reported a cob(I)alamin-catalyzed desymmetrization of a spiro-activated meso-bicyclo[4.1.0]heptane to deliver monosubstituted cyclohexene derivatives (Scheme [1]A, eq 1).[8] Secondary-amine-catalyzed ring openings of meso-cyclopropanecarbaldehydes through nucleophilic attack followed by capture of the resulting enamine with an electrophile were reported independently by the groups of Gilmour, Werz, and Vicario (Scheme [1]A, eq 2).[9] In 2005, Müller and Riegert reported the ring opening of spiro-activated cyclopropanes with thiophenols as the nucleophile using a superstoichiometric amount of cinchonidine as a chiral base (Scheme [1]A, eq 3).[10] However, ring-retentive desymmetrization of meso-bicyclo[4.1.0]heptane derivatives has never been realized.

Scheme 1 Catalytic enantioselective desymmetrization of meso-bicyclo[4.1.0]heptane derivatives

An ideal desymmetrization results in a loss of symmetry by virtue of the unmasking of stereocenter(s) already present in a symmetrical molecule without generating any additional stereocenter and without modifying the existing functionality. The biggest advantage of such enantioselective desymmetrization reactions lies in their ability to control the stereochemistry remote from the reaction site.[11] Following on our interest in organocatalytic enantioselective formal C(sp²)–H alkylative desymmetrization of prochiral 2,2-disubstituted cyclopentene-1,3-diones,[11h] meso-norbornenoquinones,[11f] and, as recently disclosed, trans-cyclobutane-fused cyclohexene-1,4-diones,[11b] we now report the application of the same strategy to meso-cyclopropane-fused cyclohexene-1,4-diones (Scheme [1]B). To the best of our knowledge, this is the first example of a ring-retentive enantioselective desymmetrization of any meso-bicyclo[4.1.0]heptane derivative.

We began our studies with meso-bicyclo[4.1.0]hept-3-ene-2,5-dione (10a)[12] as a model substrate and nitromethane (11a) as the alkylating agent (Table [1]).[13] First, we screened a collection of achiral bases to identify a suitable terminal base that would cause a minimal background reaction. In this screening, Li₂CO₃ turned out to be optimal, as no product formation was observed when the reaction was carried out in the absence of any catalyst in toluene at 25 °C (Table [1], entry 1). However, a combination of 10 mol% of the quinine-derived tertiary aminothiourea[14] I and 1.5 equivalents of Li₂CO₃ in the presence of 5 Å MS as additive furnished the desired 3-methylbicyclo[4.1.0]hept-3-ene-2,5-dione (12aa) in only 28% NMR yield with 83:17 er after 96 hours (entry 2). On the other hand, Na₂CO₃ as the terminal base, under otherwise identical reaction conditions, resulted in complete consumption of 10a to deliver 12aa in 53% isolated yield, albeit with a notable erosion in enantioselectivity (entry 3). A significant improvement in yield and enantioselectivity was observed by using the corresponding squaramide II [15] as the catalyst (entry 4). Among the various bifunctional squaramide derivatives II–IV, the dihydroquinine-derived squaramide III proved to be the best in terms of enantioselectivity. A solvent screening at this point revealed CHCl₃ to be optimal (entries 7–11). To our delight, lowering the initial concentration to 0.1 M together with an increase in the catalyst loading to 20 mol% improved the reaction efficiency while maintaining the same level of enantioselectivity (entry 13).[13] When the reaction was scaled up under these conditions, 12aa was isolated in 88% yield with 92:8 er (entry 14).

Table 1 Optimization of the Catalyst and Reaction Conditions for the Enantioselective Formal C(sp²)–H Methylation of the meso-Bicyclocyclopropane 10a ^a

Entry	Catalyst	Solvent	Yieldb (%)	erc
1d	–	toluene	<5	–
2d	I	toluene	28	83:17
3	I	toluene	55 (53)	74:26
4	II	toluene	65 (61)	86:14
5	III	toluene	60	89:11
6	IV	toluene	39	82:18
7	III	THF	38	70:30
8	III	CH2Cl2	58	91:9
9	III	DCE	48	91:9
10	III	CHCl3	64	91:9
11	III	CCl4	56	87.5:12.5
12e	III	CHCl3	72	92:8
13f	III	CHCl3	95	91:9
14e–h	III	CHCl3	(88)	92:8
15e	V	CHCl3	60	91:9
16 e,g,i,j	VI	CHCl3	(56)	97:3

^a Unless stated otherwise, the reactions were performed on a 0.1 mmol scale.

^b Determined by ¹H NMR spectroscopy with 1,3,5-trimethoxybenzene as the internal standard. Isolated yields are given in parentheses.

^c Determined by chiral HPLC.

^d With Li₂CO₃ (1.5 equiv) as the terminal base.

^e Reaction with a 0.1 M initial concentration of 10a.

^f Reaction with 20 mol% of catalyst.

^g Reaction on a 0.3 mmol scale.

^h Reaction time: 144 h.

ⁱ Reaction with 11a (15 equiv) and Na₂CO₃ (2.5 equiv).

^j Reaction time 336 h.

To improve the enantioselectivity further, we screened two bulkier analogues of the dihydroquinine-derived squaramide catalyst, V and VI.[16] Although the use of the TMS-protected diarylmethanol-appended squaramide catalyst V did not produce any beneficial effect (entry 15), its unprotected counterpart VI gave 12aa with improved enantioselectivity, albeit at the expense of the reaction rate (entry 16). This unacceptable reaction rate deterred us from using catalyst VI further.

We therefore decided to use catalyst III under the reaction conditions shown in Table [1], entry 14 to examine the generality of this formal C(sp²)-H alkylative desymmetrization reaction (Scheme [2]).[13] [17] We initially focused on elucidating the scope of the nitroalkane for the alkylation of bicyclo[4.1.0]hept-3-ene-2,5-dione (10a). Under standard reaction conditions, nitroethane (11b) afforded the expected ethylated product 12ab in 10% yield only. Changing the reaction medium to toluene with an increased initial concentration of 0.5 M, improved the yield of 12ab to 47% with 88:12 er. A benzyl group (12ac) or a homobenzyl group (12ad) could be smoothly introduced in moderate yield, but with reduced enantioselectivity. In the case of TBS-protected 2-nitroethanol 11e, the product 12ae was obtained in good yield with moderate enantioselectivity. A further erosion in enantioselectivity was observed when the reaction was carried out with unprotected 2-nitroethanol (11f).

Scheme 2 Scope of the enantioselective formal C(sp²)–H alkylation of meso-cyclopropane-fused cyclohexenediones. Reactions were carried out on a 0.3 mmol scale of 10 with either 10 equivalents (11a,b) or 5 equivalents (11c–f) of the nitroalkane and 150 mg 5 Å MS in a sealed vial. Isolated yields after chromatographic purification are reported. The er was determined by HPLC analysis with a chiral stationary phase. ^a With 10 mol% III in toluene (0.5 M). ^b With VI (10 mol%) as catalyst.

To elucidate the effect of substitution at the 7-position of the cyclopropane ring in 10, the 7,7-disubstituted analogues 10b (R = CO₂Et) and 10c (R = Me) were subjected to methylation with nitromethane (11a) as the alkylating agent. With 7,7-diester 10b as the substrate, the desired methylated product 12ba was obtained in moderate yield and er (Scheme [2]B, entry 1). The enantioselectivity was improved by using VI as the catalyst (entry 2). We were pleased to note that the 7,7-dimethylcyclopropane 10c smoothly underwent methylation. The resulting C(sp²)-H methylated derivative 12ca, identical to the natural product car-3-ene-2,5-dione,[5] was obtained in 61% yield with 94.5:5.5 er. This is the first catalytic enantioselective synthesis of car-3-ene-2,5-dione. The absolute stereochemistry of 12ca was confirmed to be (1S,6R) by comparing its specific rotation with that of (–)-car-3-ene-2,5-dione[18] prepared from optically pure (+)-3-carene[19] (Scheme [3]A). The absolute stereochemistry of the other products shown in Scheme [2] was assigned by analogy.

Scheme 3 (A) The first catalytic enantioselective synthesis of (–)-car-3-ene-2,5-dione and confirmation of its absolute stereochemistry. (B) Larger-scale reaction. (C) Product derivatization.

To test the scalability of this protocol, a formal C(sp²)-H methylation of 10a was performed on a 1.0 mmol scale, giving 12aa in the same level of yield and enantioselectivity as the smaller-scale reaction (Scheme [3]B). However, this reaction requires seven days for completion. Column chromatographic purification permitted the recovery of the catalyst, which could be reused without any visible change in its performance. Subjecting 12aa to allylic bromination conditions delivered a separable mixture of the mono- and dibrominated products 15 and 16, respectively, in 83% combined yield without much erosion in enantioselectivity (Scheme [3]C).

In conclusion, we have accomplished the first ring-retentive desymmetrization of bicyclo[4.1.0]heptane derivatives, namely meso-cyclopropane-fused cyclohexene-1,4-diones. With dihydroquinine-derived bifunctional tertiary aminosquaramides as catalysts, this enantioselective desymmetrization proceeds through a formal C(sp²)–H alkylation with a nitroalkane as the alkylating agent. As an application of this protocol, the first enantioselective synthesis of the natural product (–)-car-3-ene-2,5-dione was demonstrated. Although the high catalyst loading and long reaction time are among the drawbacks of this strategy, the catalyst can be recovered and reused.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

S.R. thanks the Ministry of Education, Government of India for his doctoral fellowship through the Prime Minister's Research Fellowship (PMRF) scheme.

• References and Notes

• 1a Wiberg KB. Angew. Chem. Int. Ed. Engl. 1986; 25: 312
• 1b Kozina MP, Mastryukov VS, Mil’vitskaya EM. Russ. Chem. Rev. 1982; 51: 765
• 1c von Baeyer A. Ber. Dtsch. Chem. Ges. 1885; 18: 2269
• 2a Ma S, Mandalapu D, Wang S, Zhang Q. Nat. Prod. Rep. 2022; 39: 926
• 2b Sansinenea E, Ortiz A. Eur. J. Org. Chem. 2022; e202200210
• 2c Apel C, Christmann M. Tetrahedron 2021; 82: 131760
• 2d Liu J, Liu R, Wei Y, Shi M. Trends Chem. 2019; 1: 779
• 2e Novakov IA, Babushkin AS, Yablokov AS, Nawrozkij MB, Vostrikova OV, Shejkin DS, Mkrtchyan AS, Balakin KV. Russ. Chem. Bull. 2018; 67: 395
• 2f Wu W, Lin Z, Jiang H. Org. Biomol. Chem. 2018; 16: 7315
• 2g Ebner C, Carreira EM. Chem. Rev. 2017; 117: 11651
• 2h Fan Y.-Y, Gao X.-H, Yue J.-M. Sci. China Chem. 2016; 59: 1126
• 2i Cavitt MA, Phun LH, France S. Chem. Soc. Rev. 2014; 43: 804
• 2j Chen DY.-K, Pouwer RH, Richard J.-A. Chem. Soc. Rev. 2012; 41: 4631
• 2k Tang P, Qin Y. Synthesis 2012; 44: 2969
• 2l Wessjohann LA, Brandt W, Thiemann T. Chem. Rev. 2003; 103: 1625
• 2m Gnad F, Reiser O. Chem. Rev. 2003; 103: 1603
• 2n Lukina MY. Russ. Chem. Rev. 1962; 31: 419
• 2o Perkin WH. J. Chem. Soc., Trans. 1885; 47: 801
• 2p Perkin WH. Ber. Dtsch. Chem. Ges. 1884; 17: 323
• 3 Buzzetti F, Salle E, Lombardi P. DE 3719913 1987
• 4 Chen Y.-P, Ying S.-S, Zheng H.-H, Liu Y.-T, Wang Z.-P, Zhang H, Deng X, Wu Y.-J, Gao X.-M, Li T.-X, Zhu Y, Xu Y.-T, Wu H.-H. Sci. Rep. 2017; 7: 15114
• 5 Hashimoto K, Katsuhara T, Itoh H, Ikeya Y, Okada M, Mitsuhashi H. Phytochemistry 1990; 29: 3571

For selected reviews, see:
• 6a Nájera C, Foubelo F, Sansano JM, Yus M. Tetrahedron 2022; 106–107: 132629
• 6b Xu Y, Zhai T.-Y, Xu Z, Ye L.-W. Trends Chem. 2022; 4: 191
• 6c Chegondi R, Patel SM, Maurya S, Donthoju A. Asian J. Org. Chem. 2021; 10: 1267
• 6d Das T. ChemistrySelect 2020; 5: 14484
• 6e Borissov A, Davies TQ, Ellis SR, Fleming TA, Richardson MS. W, Dixon DJ. Chem. Soc. Rev. 2016; 45: 5474
• 6f Zeng X.-P, Cao Z.-Y, Wang Y.-H, Zhou F, Zhou J. Chem. Rev. 2016; 116: 7330
• 6g Manna MS, Mukherjee S. Org. Biomol. Chem. 2015; 13: 18
• 6h Fernández-Pérez H, Etayo P, Lao JR, Núñez-Rico JL, Vidal-Ferran A. Chem. Commun. 2013; 49: 10666
• 7a Pirenne V, Muriel B, Waser J. Chem. Rev. 2021; 121: 227
• 7b Xia Y, Liu X, Feng X. Angew. Chem. Int. Ed. 2021; 60: 9192
• 8 Troxler T, Scheffold R. Helv. Chim. Acta 1994; 77: 1193
• 9a Díaz E, Reyes E, Uria U, Carrillo L, Tejero T, Merino P, Vicario JL. Chem. Eur. J. 2018; 24:  8764
• 9b Wallbaum J, Garve LK. B, Jones PG, Werz DB. Chem. Eur. J. 2016; 22:  18756
• 9c Sparr C, Gilmour R. Angew. Chem. Int. Ed. 2011; 50:  8391
• 10 Müller P, Riegert D. Tetrahedron 2005; 61: 4373

For selected examples of such remote stereocontrol, see:
• 11a Lou Y, Wei J, Li M, Zhu Y. J. Am. Chem. Soc. 2022; 144: 123
• 11b Ray S, Mondal S, Mukherjee S. Angew. Chem. Int. Ed. 2022; 61: e202201584
• 11c Genov GR, Douthwaite JL, Lahdenperä AS. K, Gibson DC, Phipps RJ. Science 2020; 367: 1246
• 11d Shi H, Herron AN, Shao Y, Shao Q, Yu J.-Q. Nature 2018; 558: 581
• 11e Su B, Zhou T.-G, Xu P.-L, Shi Z.-J, Hartwig JF. Angew. Chem. Int. Ed. 2017; 56: 7205
• 11f Sarkar R, Mukherjee S. Org. Lett. 2016; 18: 6160
• 11g Manna MS, Sarkar R, Mukherjee S. Chem. Eur. J. 2016; 22: 14912
• 11h Manna MS, Mukherjee S. J. Am. Chem. Soc. 2015; 137: 130
• 11i Zhou F, Tan C, Tang J, Zhang Y.-Y, Gao W.-M, Wu H.-H, Yu Y.-H, Zhou J. J. Am. Chem. Soc. 2013; 135: 10994
• 11j Lewis CA, Gustafson JL, Chiu A, Balsells J, Pollard D, Murry J, Reamer RA, Hansen KB, Miller SJ. J. Am. Chem. Soc. 2008; 130: 16358
• 11k Cefalo DR, Kiely AF, Wuchrer M, Jamieson JY, Schrock RR, Hoveyda AH. J. Am. Chem. Soc. 2001; 123: 3139
• 12 Mal D, Ray S. Eur. J. Org. Chem. 2008; 2008: 3014
• 13 For details, see the Supporting Information

For pioneering work on bifunctional thiourea catalysis, see:
• 14a Okino T, Hoashi Y, Takemoto Y. J. Am. Chem. Soc. 2003; 125: 12672. For seminal works on cinchonaalkaloid-based bifunctional thiourea catalysts, see
• 14b Li B.-J, Jiang L, Liu M, Chen Y.-C, Ding L.-S, Wu Y. Synlett 2005; 603
• 14c McCooey SH, Connon SJ. Angew. Chem. Int. Ed. 2005; 44: 6367
• 14d Vakulya B, Varga S, Csámpai A, Soós T. Org. Lett. 2005; 7: 1967
• 14e Ye J, Dixon DJ, Hynes PS. Chem. Commun. 2005; 4481
• 15 For seminal work on bifunctional squaramide catalysts, see: Malerich JP, Hagihara K, Rawal VH. J. Am. Chem. Soc. 2008; 130: 14416
• 16 Odriozola A, Oiarbide M, Palomo C. Chem. Eur. J. 2017; 23: 12758

Catalytic Enantioselective Alkylative Desymmetrization of meso-Cyclopropane-Fused Cyclohexene-1,4-Diones; General ProcedureA glass vial was charged with freshly activated 5 Å MS (150 mg), catalyst III (0.06 mmol, 0.2 equiv), Na₂CO₃ (0.45 mmol, 1.5 equiv), and the appropriate cyclopropane-fused enedione 10 (0.30 mmol, 1.0 equiv) under a positive argon pressure. Distilled CHCl₃ (3.0 mL) was added and the resulting suspension was stirred at 25 °C for 10 min. Nitroalkane 11 (11b or 11c: 3.0 mmol; 11c–f: 1.5 mmol) was then added, and the resulting mixture was stirred at 25 °C until complete conversion of 10 (TLC). The mixture was then diluted with CH₂Cl₂ (2 mL) and filtered through Celite, which was washed with additional CH₂Cl₂ (3 × 5 mL). The combined organic phase was concentrated under reduced pressure and the residue was purified by flash column chromatography (silica gel).(1R,6S)-3-Methylbicyclo[4.1.0]hept-3-ene-2,5-dione (12aa)Prepared according to the general procedure, and purified by flash column chromatography (silica gel, 20–25% EtOAc–PE) to give a thick yellow oil; yield: 36 mg (0.264 mmol, 88%); [α]_D ²² –83.7 (c 1.0, CHCl₃) for an enantiomerically enriched sample with 92:8 er. HPLC [Daicel Chiralpak IG, hexane–EtOH (60:40), 1.0 mL/min, 20 °C, λ = 254 nm]: t _minor = 14.7 min, t _major = 17.5 min. FTIR (thin film): 3340 , 3062 , 2924 , 1726 , 1674 , 1620 , 1440 , 1348 , 1282 cm^–1. ¹H NMR (400 MHz, CDCl₃): δ = 6.29 (s, 1 H), 2.55–2.46 (m, 2 H), 1.95 (s, 3 H), 1.67 (dt, J ₁ = 8.7, J ₂ = 4.9 Hz, 1 H), 1.58 (q, J = 5.1 Hz, 1 H). ¹³C NMR (100 MHz, CDCl₃): δ = 195.7, 194.8, 146.4, 133.7, 27.8, 27.0, 20.1, 16.5. HRMS (APCI): m/z [M + H]⁺ calcd for C₈H₉O₂: 137.0603; found: 137.0605. (1R,6S)-3-Benzylbicyclo[4.1.0]hept-3-ene-2,5-dione (12ac)Prepared according to the general procedure and purified by flash column chromatography (silica gel, 12–15% EtOAc–PE) as a yellow oil; yield: 33 mg (0.155 mmol, 52%); [α]_D ²² –43.3 (c 2.0, CHCl₃) for an enantiomerically enriched sample with 79:21 er.HPLC [Daicel Chiralpak IG, hexane–EtOH (60:40), 1.0 mL/min, 20 °C, λ = 254 nm]: t _minor = 10.9 min, t _major = 27.5 min. FTIR (thin film): 3061 , 3028 , 2924 , 1675 , 1614 , 1495 , 1346 , 1279 , 1041 , 700 cm^–1. ¹H NMR (400 MHz, CDCl₃): δ = 7.26–7.22 (m, 2 H), 7.20–7.18 (m, 1 H), 7.07 (d, J = 7.1 Hz, 2 H), 6.00 (br m, 1 H), 3.57 (dd, J ₁ = 29.6, J ₂ = 16.2 Hz, 2 H), 2.50–2.40 (m, 2 H), 1.61–1.55 (m, 1 H), 1.47 (q, J = 5.0 Hz, 1 H). ¹³C NMR (100 MHz, CDCl₃): δ = 195.1, 194.9, 149.2, 136.5, 133.3, 129.3, 129.0, 127.1, 35.8, 27.9, 27.2, 19.9. HRMS (ESI+): m/z [M + H]⁺ calcd for C₁₄H₁₃O₂: 213.0916; found: 213.0918.(–)-Car-3-ene-2,5-dione (12ca)Prepared according to the general procedure and purified by flash column chromatography (silica gel, 15–17% EtOAc–PE) as a yellow crystalline solid; yield: 31 mg (0.183 mmol, 61%); mp 77–79 °C; [α]_D ²² –13.9 (c 0.25, CHCl₃) for an enantiomerically enriched sample with 94.5:5.5 er.HPLC [Daicel Chiralpak IG, hexane–EtOH (60:40), 1.0 mL/min, 20 °C, λ = 254 nm]: t _minor = 9.4 min, t _major = 17.3 min. FTIR (thin film): 3034 , 2962 , 1662 , 1447 , 1370 , 1289 , 1125 cm^–1. ¹H NMR (400 MHz, CDCl₃): δ = 6.50–6.49 (br m, 1 H), 2.35–2.30 (m, 2 H), 1.97 (d, J = 1.3 Hz, 3 H), 1.32 (s, 3 H), 1.31 (s, 3 H). ¹³C NMR (100 MHz, CDCl₃): δ = 195.1, 194.5, 150.1, 137.8, 40.0, 39.1, 33.7, 29.2, 16.3, 15.6. HRMS (ESI+): m/z [M + H]⁺ calcd for C₁₀H₁₃O₂: 165.0916; found: 165.0918.
• 18 Galin FZ, Kashina YA, Zainullin RA, Kukovinets OS, Khalilov LM, Tolstikov GA. Russ. Chem. Bull. 1998; 47: 185
• 19 Gomes SQ, Salles AG. Jr. Synth. Commun. 2019; 49: 3389

Publication History

Received: 24 June 2022

Accepted after revision: 28 July 2022

Accepted Manuscript online:
28 July 2022

Article published online:
27 September 2022