Download PDF
Synlett 2015; 26(15): 2121-2126
DOI: 10.1055/s-0035-1560052
letter
© Georg Thieme Verlag Stuttgart · New York

Zinc-Mediated Allylation Followed by Lactonization of Dialkyl 2-(3-Oxo-1,3-diarylpropyl)malonates: Construction of δ-Lactones with Multiple Stereocenters

Chennakesava Reddy
Indian Institute of Science Education and Research (IISER) Mohali, Knowledge City, Sector 81, SAS Nagar, Mohali, Manauli P.O., Punjab, 140306, India   Email: sababu@iisermohali.ac.in
,
Srinivasarao Arulananda Babu*
Indian Institute of Science Education and Research (IISER) Mohali, Knowledge City, Sector 81, SAS Nagar, Mohali, Manauli P.O., Punjab, 140306, India   Email: sababu@iisermohali.ac.in
› Author Affiliations
Further Information

Publication History

Received: 29 May 2015

Accepted after revision: 02 July 2015

Publication Date:
20 August 2015 (online)

 
 PDF Download Permissions and Reprints All articles of this category


Abstract

A variety of polysubstituted δ-lactones containing three or four stereocenters were prepared from various dialkyl 2-(3-oxo-1,3-diarylpropyl)malonates by a Barbier-type zinc-mediated allylation or cyclohexenylation of the keto group, followed by intramolecular lactonization/transesterification. The stereochemistry of the major isomers was confirmed by X-ray crystal structure analysis of representative compounds.


#

Key words

allylation - diastereoselectivity - lactones - stereoselective synthesis - zinc

δ-Lactones are present as core units in numerous natural products, and they are considered as very important organic molecules in the areas of medicinal and organic chemistry and flavor components.[1] [2] Numerous δ-lactone derivatives have been found to exhibit a wide range of biological activities, and δ-lactones are also important synthetic building blocks for the synthesis of various natural products and biologically active synthetic molecules.[1–4] For example, ophiodilactone A (I) and ophiodilactone B (II) (Scheme [1]) have been found to exhibit cytotoxic activity against P388 murine leukemia cells.[3a] [b] Aryl-substituted enol δ-lactones V and VI (Scheme [1]) have been shown to inhibit human neutrophil elastase and trypsin-like prote­ases.[3e]

Zoom Image
Scheme 1

Numerous methods have been reported for synthesizing multifunctionalized δ-lactones.[1] [2] [3] [4] [5] [6] Steglich’s group showed the asymmetric synthesis of polysubstituted δ-lactone calopin (III).[3c] Kobayashi’s group reported a concise stereoselective synthetic route for the synthesis of polysubstituted δ-lactone prelactone C (IV).[3d] In general,� stereoselective construction of polysubstituted δ-lactones is a relatively difficult task.[1] [2] [3] [4] [5] [6] We have accomplished a zinc-mediated Barbier-type allylation[7] of ketones, followed by lactoni­zation of the resulting metal alkoxide with one of the ester functional groups of the substrates 1 (Scheme [1]). In this letter, we report our preliminary investigations on the synthesis of various polysubstituted δ-lactones by using a zinc-mediated Barbier-type allylation[7] and subsequent lactoni­zation of dialkyl 2-(3-oxo-1,3-diarylpropyl)malonates 1.

At the outset, we performed optimization studies on the synthesis of δ-lactones by Barbier-type allylation/lactonization sequences from substrates 1a and 1b (Table [1]). Ini­tially, we performed the Barbier reaction on substrate 1a with allyl bromide in the presence of indium powder in anhydrous tetrahydrofuran; the allylation and subsequent transesterification promoted lactonization to afford the δ-lactone 3a in 86% yield. Allylation followed by lactonization of substrate 1a generated three stereocenters in δ-lactone 3a and, therefore, four diastereomers were expected to be formed in this reaction. However, we observed the formation of only two diastereomers of 3a in a diastereomeric ratio of 75:25 (Table [1], entry 1). Allylation of the substrate 1a with allyl bromide in the presence of indium powder in anhydrous N,N-dimethylformamide gave product 3a in 78% yield with moderate diastereoselectivity (dr = 65:35; entry 2).

The indium-mediated allylation of the substrate 1a with allyl bromide in other solvents, such as ethanol, methanol, dimethyl sulfoxide, 1,4-dioxane, or 1,2-dichloroethane gave product 3a in low yield (5–28%) (Table [1], entries 3–7). The indium-mediated allylation of the substrate 1a in water or aqueous tetrahydrofuran did not afford 3a (entries 8 and 9). Allylation of substrate 1a with allyl bromide in the presence of zinc powder instead of indium powder in anhydrous tetrahydrofuran gave product 3a in high yield (98%) (entry 10) and with a comparable diastereoselectivity (dr 75:25) to that of the indium-mediated reaction (entry 1).

To corroborate the efficiency of the zinc-mediated allylation process, we treated diester 1b with allyl bromide in the presence of zinc powder in anhydrous tetrahydrofuran, and we obtained the corresponding product 3b in good yield (76%, dr 75:25) (Table [1], entry 11). We also carried out the zinc-mediated allylation of substrate 1a in N,N-dimethylformamide and in 1,4-dioxane, which gave the product 3a in 89 and 50% yield, respectively, but with low diastereoselectivities (entries 12 and 13). Allylation of the substrate 1b with allyl bromide in the presence of aluminum powder in anhydrous tetrahydrofuran gave product 3b in 40% yield (entry 14), but no product was obtained in the presence of tin powder (entry 15).

Next, the generality of this protocol was explored through the use of the substrates 1c–n (Scheme [2]). Zinc-mediated Barbier-type allylation followed by the lactonization of the substrates 1c–n gave the corresponding polysubstituted δ-lactones 3c–n, each containing three stereocenters, in 60–98% yield. In all these cases, the allylation and subsequent lactonization generated three stereocenters in the δ-lactone product 3c–n; correspondingly, each reaction was expected to provide four diastereomers. However, we observed the formation of only two of the diastereomers of 3c–n with a diastereomeric ratio of up to 80:20 (Scheme [2]). In the cases of substrates 1g, 1h, 1j, and 1k, which contained hetaryl groups, the zinc-mediated allylation reaction gave the corresponding products 3g, 3h, 3j, and 3k with relatively low diastereoselectivities.

We then performed a Barbier allylation/lactonization on substrates 1a and 1d by treatment with prenyl bromide in the presence of zinc powder and we successfully obtained the corresponding polysubstituted δ-lactones 3o (64%, dr 80:20) and 3p (56%, dr 75:25) with good diastereoselectivities (Scheme [3]). Allylation/lactonization of substrate 1b with 2,3-dibromoprop-1-ene in the presence of zinc powder in anhydrous tetrahydrofuran failed to give δ-lactone 3q. On the other hand, zinc-mediated allylation/lactonization of substrate 1d with 3-bromo-2-methylprop-1-ene in anhydrous tetrahydrofuran successfully gave the polysubstituted δ-lactone 3r in 91% yield with a diastereomeric ratio of 70:30. Next, we explored the synthesis of polysubstituted γ-lactones by the allylation/lactonization protocol. Accordingly, we carried out Barbier-type allylation/lactoni­zation of substrates 1o–q with allyl bromide in the presence of zinc powder and we successfully obtained the corresponding polysubstituted γ-lactones 3s (70%, dr 76:24), 3t (75%, dr 88:12), and 3u (58%, dr 80:20) with good diastereoselectivities (Scheme [3]). In these cases, we once again observed the formation of only two diastereomers of γ-lactones 3s–u.

Table 1 Optimization of the Reaction Conditions for the Construction of Polysubstituted δ-Lactones 3a and 3b a

Entry

R

Metal

Solvent (mL)

Time (h)

Product

Yield (%)

dr

 1

OMe

In

THF (1.5)

 7

3a

86

75:25

 2

OMe

In

DMF (1)

 7

3a

78

65:35

 3

OMe

In

EtOH (1)

 7

3a

25

50:50

 4

OMe

Zn or In

MeOH (1)

 8

3a

<5

–

 5

OMe

In

DMSO (1)

12

3a

28

47:53

 6

OMe

In

1,4-dioxane (1)

12

3a

15

50:50

 7

OMe

In

DCE (1)

30

3a

17

46:54

 8

OMe

In

H2O (5)

12

3a

 0

–

 9

OMe

In

THF–H2O (3, 1:1)

12

3a

 0

–

10

OMe

Zn

THF (1.5)

 7

3a

98

75:25

11

Cl

Zn

THF (1.5)

 2

3b

76

75:25

12

OMe

Zn

DMF (1.5)

 6.5

3a

89

56:44

13

OMe

Zn

1,4-dioxane (1)

 6.5

3a

50

62:38

14

Cl

Al

THF (1.5)

 3.5

3b

40

60:40

15

OMe

Sn

THF (1.5)

 7

3a

 0

–

a Reaction conditions: 1a/1b (0.25 mmol), allyl bromide (0.5 mmol), metal powder (0.37 mmol).

Zoom Image
Scheme 2 Synthesis of the δ-lactone scaffolds 3c–n. Reaction conditions: 1 (0.25 mmol), allylbromide (0.5 mmol), Zn powder (0.37 mmol), THF (1.5 mL). Reaction times: a 1.5 h. b 2 h. c 3 h. d 2.5 h. e 14 h. f 7 h.
Zoom Image
Scheme 3 Synthesis of γ- and δ-lactones 3o–u. Reaction temperatures: a 25 °C. b 32 °C. c –10 °C for 2 h followed by r.t. for 1 h.

Finally, we sought to extend the scope of this protocol to obtain cyclohexenylated δ-lactone scaffolds with multiple stereocenters (Scheme [4]). Accordingly, we performed the zinc-mediated Barbier-type reaction of substrate 1d with 3-bromocyclohexene and we obtained the polysubstituted δ-lactone 4a in 90% yield and high diastereoselectivity (dr 90:10) (Scheme [4]). Notably, we observed the formation of only two diastereomers of 4a. Encouraged by this result, we carried out the reactions of various dialkyl 2-(3-oxo-1,3-diarylpropyl)malonates 1 with 3-bromocyclohexene in the presence of zinc powder, and we obtained the corresponding polysubstituted δ-lactones 4b–h containing four stereocenters in 40–98% yield and moderate to high diastereoselectivities (dr ≤ 95:5); once more, in each case we observed the formation of only two diastereomers of the δ-lactone 4b–h.

Generally, in all the zinc-mediated ketone-group allylation/intramolecular lactonization reactions of substrates 1a–n and 1o–q, we observed the formation of only two corresponding diastereomers of each of 3a–r, 3s–u, and 4a–h. The structures and stereochemistries of the major diastereomers were established by means of single-crystal X-ray structure analyses of the major compounds 3b, 3c, 3t, and 4b (Figure [1]).[8] [9] On the basis of the X-ray analyses of the major compounds 3b, 3c, 3t, and 4b and the similarity in the NMR spectroscopic patterns of the respective series 3a–r, 3s–u, and 4a–h (see Supporting Information), we propose the structures of the major diastereomers shown in Schemes 2–4 and Table [1]. On the basis of the X-ray analyses (Figure [1]), the relative stereochemistry of the major isomer 3c at the C3(H), C4(H), and C6 stereocenters was assigned to be 3S*,4S*,6S*. Furthermore, with the help of 1H, 13C, DEPT-135, H,H-COSY, and HMQC NMR analyses of compounds 3c (major isomer) and 3c′ (minor isomer), the stereochemistry of the representative minor isomer 3c′ at the C3(H), C4(H), and C6 stereocenters was assigned to be 3S*,4S*,6R* (Figure [2], see also the Supporting Information). Accordingly, the assumption is that all other minor isomers are expected to have a stereochemistry similar to 3c′ (minor isomer). Furthermore, the assignment of the stereochemistry of 3c (major isomer) and 3c′ (minor isomer) indicates, plausibly, the diastereomers that are formed in the Barbier reaction process.[10a] However, because the lactones have an acidic α-hydrogen at the C3 stereocenter, we cannot ignore the possibility that epimerization at the C3 stereocenter during the reaction might contribute to the stereochemistry of the C3 stereocenter, a possibility regarding which we have no convincing evidence at this stage.[10b]

Zoom Image
Figure 1 X-ray structures of 3b, 3c, 3t, and 4b

In summary, we have described our preliminary investigations on the Barbier-type zinc-mediated allylation and cyclohexenylation of ketones with subsequent intramolecular lactonization. This protocol has led to the assembly of a variety of polysubstituted δ-lactones.

Zoom Image
Scheme 4 Synthesis of δ-lactones 4a–h. The starting materials for products 4a–h were 1d, 1e, 1f, 1g, 1l, 1m, 1a, and 1b, respectively. Reagents and conditions: 1 (0.25 mmol), 3-bromocyclohexene (0.5 mmol), Zn powder (0.37 mmol). Reaction times: a 2 h. b 1 h at –10 °C, then 1 h at r.t. c 3 h. d 2.5 h at –10 °C, then 1 h at r.t. e 2.5 h. f 14 h.
Zoom Image
Figure 2 Stereochemistry of the major and minor isomers

#

Acknowledgment

We thank IISER-Mohali for funding this research work and the NMR, HRMS, and X-ray facilities. C.R. thanks CSIR, New Delhi, for an SRF fellowship. We thank the reviewers for their valuable suggestions.

Supporting Information

    Supporting information for this article is available online at http://dx.doi.org.accesdistant.sorbonne-universite.fr/10.1055/s-0035-1560052.
  • Supporting Information

  • References and Notes

  • 8 Crystallographic data for compounds 3b, 3c, 3t, and 4b have been deposited with the accession numbers CCDC 1048540, 1048541, 1048542, and 1048543, respectively, and can be obtained free of charge from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44(1223)336033; E-mail: deposit@ccdc.cam.ac.uk; Web site: www.ccdc.cam.ac.uk/conts/retrieving.html.
  • 9 γ-Lactones 3s–u and δ-Lactones 3a–r and 4a–h; General Procedure Zn powder (0.37 mmol) and allylbromide or 3-bromocyclohexene (0.5 mmol) were added to a solution of the appropriate malonate substrate 1 (0.25 mmol) in anhyd THF (1.5 mL) under N2, and the mixture was stirred at 30–35 °C for the appropriate time (see Table 1 and Schemes 2–4). The reaction was then quenched by adding H2O (2 mL), and the mixture was allowed to stand. The mixture was transferred to a separatory funnel and extracted with EtOAc (3 × 8 mL), and the organic layers were combined, dried (Na2SO4), filtered, and concentrated under vacuum. The resulting crude product was purified by column chromatography (silica gel, EtOAc–hexane). Ethyl (3S*,4S*,6S*)-6-Allyl-4-(4-chlorophenyl)-2-oxo-6-phenyltetrahydro-2H-pyran-3-carboxylate (3b) Prepared by the general procedure from 1b, and purified by column chromatography [silica gel, EtOAc–hexanes (13: 87)] as a colorless solid (major isomer); yield: 97 mg (98%; dr 75:25); mp 87–89 °C. IR (KBr): 1747, 1726, 1494 and 1155 cm–1. 1H NMR (400 MHz, CDCl3): δ = 7.41–7.28 (m, 5 H), 7.32 (d, J = 8.4 Hz, 2 H), 7.13 (d, J = 8.4 Hz, 2 H), 5.66–5.56 (m, 1 H), 5.16–5.11 (m, 2 H), 4.16–4.10 (m, 2 H), 3.81 (td, J1  = 12.2 Hz, J2  = 4.7 Hz, 1 H), 3.54 (d, J = 12.2 Hz, 1 H), 2.92 (dd, J1  = 14.2 Hz, J2  = 6.7 Hz, 1 H), 2.84 (dd, J1  = 14.2 Hz, 1 H, J2  = 7.6 Hz), 2.67 (dd, J1  = 14.3 Hz, J2  = 4.7 Hz, 1 H), 2.29 (dd, J1  = 14.3 Hz, J2  = 12.2 Hz, 1 H), 1.14 (t, J = 7.2 Hz, 3 H). 13C NMR (100 MHz, CDCl3): δ = 167.9, 166.9, 143.2, 138.8, 133.6, 131.4, 131.4, 129.2, 128.7, 128.4, 128.0, 124.6, 120.0, 86.1, 61.8, 54.2, 47.0, 39.3, 38.3, 14.0. HRMS (ESI): m/z [M + Na]+ calcd for C23H23ClNaO4: 421.1183; found: 421.1174.
    • 10a This observation implies the stereoselection might occur at the Barbier reaction step through a plausible chelation effect involving the malonate moiety. This possibility is based on the observation that reaction in polar protic solvents such as EtOH gave no selectivity (Compare entries 1 and 3 in Table 1).
    • 10b There was no significant change in the diastereoselectivity with respect to the two diastereomers obtained from reactions performed for different times. Furthermore, the diastereoselectivity of the crude reaction mixture did not differ significantly from that of the pure mixture of diastereomers obtained after isolation by column chromatography on silica gel.

  • References and Notes

  • 8 Crystallographic data for compounds 3b, 3c, 3t, and 4b have been deposited with the accession numbers CCDC 1048540, 1048541, 1048542, and 1048543, respectively, and can be obtained free of charge from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44(1223)336033; E-mail: deposit@ccdc.cam.ac.uk; Web site: www.ccdc.cam.ac.uk/conts/retrieving.html.
  • 9 γ-Lactones 3s–u and δ-Lactones 3a–r and 4a–h; General Procedure Zn powder (0.37 mmol) and allylbromide or 3-bromocyclohexene (0.5 mmol) were added to a solution of the appropriate malonate substrate 1 (0.25 mmol) in anhyd THF (1.5 mL) under N2, and the mixture was stirred at 30–35 °C for the appropriate time (see Table 1 and Schemes 2–4). The reaction was then quenched by adding H2O (2 mL), and the mixture was allowed to stand. The mixture was transferred to a separatory funnel and extracted with EtOAc (3 × 8 mL), and the organic layers were combined, dried (Na2SO4), filtered, and concentrated under vacuum. The resulting crude product was purified by column chromatography (silica gel, EtOAc–hexane). Ethyl (3S*,4S*,6S*)-6-Allyl-4-(4-chlorophenyl)-2-oxo-6-phenyltetrahydro-2H-pyran-3-carboxylate (3b) Prepared by the general procedure from 1b, and purified by column chromatography [silica gel, EtOAc–hexanes (13: 87)] as a colorless solid (major isomer); yield: 97 mg (98%; dr 75:25); mp 87–89 °C. IR (KBr): 1747, 1726, 1494 and 1155 cm–1. 1H NMR (400 MHz, CDCl3): δ = 7.41–7.28 (m, 5 H), 7.32 (d, J = 8.4 Hz, 2 H), 7.13 (d, J = 8.4 Hz, 2 H), 5.66–5.56 (m, 1 H), 5.16–5.11 (m, 2 H), 4.16–4.10 (m, 2 H), 3.81 (td, J1  = 12.2 Hz, J2  = 4.7 Hz, 1 H), 3.54 (d, J = 12.2 Hz, 1 H), 2.92 (dd, J1  = 14.2 Hz, J2  = 6.7 Hz, 1 H), 2.84 (dd, J1  = 14.2 Hz, 1 H, J2  = 7.6 Hz), 2.67 (dd, J1  = 14.3 Hz, J2  = 4.7 Hz, 1 H), 2.29 (dd, J1  = 14.3 Hz, J2  = 12.2 Hz, 1 H), 1.14 (t, J = 7.2 Hz, 3 H). 13C NMR (100 MHz, CDCl3): δ = 167.9, 166.9, 143.2, 138.8, 133.6, 131.4, 131.4, 129.2, 128.7, 128.4, 128.0, 124.6, 120.0, 86.1, 61.8, 54.2, 47.0, 39.3, 38.3, 14.0. HRMS (ESI): m/z [M + Na]+ calcd for C23H23ClNaO4: 421.1183; found: 421.1174.
    • 10a This observation implies the stereoselection might occur at the Barbier reaction step through a plausible chelation effect involving the malonate moiety. This possibility is based on the observation that reaction in polar protic solvents such as EtOH gave no selectivity (Compare entries 1 and 3 in Table 1).
    • 10b There was no significant change in the diastereoselectivity with respect to the two diastereomers obtained from reactions performed for different times. Furthermore, the diastereoselectivity of the crude reaction mixture did not differ significantly from that of the pure mixture of diastereomers obtained after isolation by column chromatography on silica gel.

   Permissions and Reprints All articles of this category
Zoom Image
Scheme 1
Zoom Image
Scheme 2 Synthesis of the δ-lactone scaffolds 3c–n. Reaction conditions: 1 (0.25 mmol), allylbromide (0.5 mmol), Zn powder (0.37 mmol), THF (1.5 mL). Reaction times: a 1.5 h. b 2 h. c 3 h. d 2.5 h. e 14 h. f 7 h.
Zoom Image
Scheme 3 Synthesis of γ- and δ-lactones 3o–u. Reaction temperatures: a 25 °C. b 32 °C. c –10 °C for 2 h followed by r.t. for 1 h.
Zoom Image
Figure 1 X-ray structures of 3b, 3c, 3t, and 4b
Zoom Image
Scheme 4 Synthesis of δ-lactones 4a–h. The starting materials for products 4a–h were 1d, 1e, 1f, 1g, 1l, 1m, 1a, and 1b, respectively. Reagents and conditions: 1 (0.25 mmol), 3-bromocyclohexene (0.5 mmol), Zn powder (0.37 mmol). Reaction times: a 2 h. b 1 h at –10 °C, then 1 h at r.t. c 3 h. d 2.5 h at –10 °C, then 1 h at r.t. e 2.5 h. f 14 h.
Zoom Image
Figure 2 Stereochemistry of the major and minor isomers