Email updates

Keep up to date with the latest news and content from Beilstein Journal of Organic Chemistry and BioMed Central.

Open Access Highly Accessed Full Research Paper

The oxanorbornene approach to 3-hydroxy, 3,4-dihydroxy and 3,4,5-trihydroxy derivatives of 2-aminocyclohexanecarboxylic acid

Ishmael B Masesane12*, Andrei S Batsanov1, Judith AK Howard1, Raju Mondal1 and Patrick G Steel1

Author Affiliations

1 Department of Chemistry, University of Durham, South Road, Durham, DH1 3LE, UK

2 Department of Chemistry, University of Botswana, P/bag 00704, Gaborone, Botswana

For all author emails, please log on.

Beilstein Journal of Organic Chemistry 2006, 2:9  doi:10.1186/1860-5397-2-9


The electronic version of this article is the complete one and can be found online at: http://bjoc.beilstein-journals.org/content/2/1/9


Received:18 April 2006
Accepted:4 May 2006
Published:4 May 2006

© 2006 Masesane et al; licensee Beilstein-Institut.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

The nitro oxanorbornene adduct derived from the Diels-Alder reaction of ethyl (E)-3-nitroacrylate and furan provides a versatile template for the stereoselective synthesis of hydroxylated derivatives of 2-aminocyclohexanecarboxylic acid (ACHC).

Graphical abstract

Introduction

In recent years there has been a surge of interest in cyclic β-amino acids and this has been accompanied by the proliferation of procedures for their synthesis.[1] Much of the interest in these compounds emanate from the ability of oligomers or short polymers of cyclic β-amino acids to adopt well-defined secondary structures analogous to those of natural peptides. [2-9] Such oligomers therefore have a particular appeal for extending understanding of protein structure and stabilization. However, much of this work has focused on simple cyclic β-amino acids and analogous studies of substituted derivatives are less well developed. One reason for this is that simple stereocontrolled routes to such derivatives have not been defined. In this respect, we have been exploring the use of oxanorbornene adducts derived from the Diels-Alder reaction of ethyl (E)-3-nitroacrylate and furan as versatile intermediates in the synthesis of a range of novel hydroxylated derivatives of 2-aminocyclohexanecarboxylic acid (ACHC). In previous preliminary communications we have described the application of the endo nitro adduct to the synthesis of various polyhydroxylated analogues.[10,11] In this paper, we provide the full details of this work and additionally describe the stereoselective synthesis of further novel hydroxylated derivatives of ACHC using the exo nitro oxanorbornene adduct as the template. The basis of our approach to these poly hydroxylated cyclohexane β-amino acids was the recognition that the bicyclic oxanorbornene cycloadduct, derived from the Diels-Alder reaction between furan and ethyl nitroacrylate, represented a versatile template that could be elaborated to the desired targets via a variety of complementary oxidative processes coupled with a base promoted ring fragmentation, Scheme 1.

Scheme 1

The furan-nitroacylate approach to poly hydroxylated ACHC derivatives

Results and Discussion

Oxanorbornene Synthesis

Efficient access to ethyl (E)-3-nitroacrylate 1 was achieved through a modification of the McMurry procedure involving the reaction of ethyl acrylate with N2O4 and I2 followed by careful elimination of HI with Hunig's base in ether.[12] With this sequence of reactions, 1 could be generated in 10–20 g batches, Scheme 1. Whilst this dienophile reacts with considerable regiocontrol (dominated by the nitro group) the stereoselectivity (endo:exo ratio) observed is frequently minimal. [13-15] Subjection of nitroacrylate 1 to cycloaddition reaction with furan in CHCl3 at room temperature gave a 2:1 mixture of cycloadducts favouring the endo nitro isomer 2a, Scheme 2. Increased endo selectivity could be achieved by carrying out the reaction at -20°C for 5 days, giving a 4:1 mixture of the two isomers in 90% yield. Attempts to further enhance this ratio through the use of a range of Lewis acids or alternative solvents were not successful. Interestingly, Just and co-workers have reported the preference for the exo-nitro adduct when the reaction of methyl 3-nitroacrylate and furan was carried out without a solvent.[16] However, in our hands, this reaction afforded a 3:1 mixture in favour of the endo-nitro adduct.

Scheme 2

Reagents: i. Furan (see box); ii. Zn, HCl., EtOH then Boc2O, Et3N, 2b 89%, 3b 77%; iii. KHMDS, THF, -50°C, 4 71%, 5 69%; iv. Ac2O, Py then H2, Pd-C, 6 75%, 7 75%.

The two isomeric cycloadducts were easily separated by column chromatography. Whilst initial attempts to reduce the nitro group were complicated by difficulties in the isolation of the resultant amine, an efficient one-pot conversion into the protected aminoesters 2b and 3b was achieved by reduction with Zn/HCl followed by addition of (Boc)2O and a large excess of iPr2NEt. At this stage, confirmation of the correct assignment of the relative stereochemistry for each isomer was realised by an X-ray crystallographic structure determination of the endo carbamate 2b (See additional materials for details).

3-Hydroxy-2-aminocyclohexanecarboxylic acids

With the oxanorbornene template available in quantities we then explored the based promoted fragmentation needed to generate the 3-hydroxy-2-aminocyclohexanecarboxylate core. Related base promoted transformations have been described in the literature and our initial experiments followed these precedents. [17-20] Using carbamate 2b, treatment with LDA or LHMDS appeared, by TLC, to give complete conversion extremely rapidly. However, on workup, considerable amounts of the starting ester were recovered. Similar observations have been reported in related systems.[21] Speculating that this was due to either a reversible process or a stabilised, highly coordinated enolate alternative procedures were then explored. Whilst attempts to enhance the fragmentation through the use of various rapid, mildly acidic, inverse quenches or various trapping agents (e.g. TMSCl, TMSOTf) were partially successful, albeit only on very small scale, the use of a less coordinating potassium counterion (KHMDS) allowed the isolation of the dihydroanthranilate esters 4 and 5 in 71% and 69% yield respectively together with variable amounts of ethyl 3-hydroxybenzoate. Simple reduction of the corresponding acetates with H2 over Pd-C proved to be highly diastereoselective affording the anti-anti and syn-syn 3-hydroxy-2-aminocyclohexanecarboxylate esters respectively. The stereochemistry was assigned by a combination of NMR techniques using J values supported by nOe correlations, Figure 1. For example, the amide derivatives of the anti-anti isomer 8 and 9 show distinctive large di-axial couplings for 1-H, 2- H and 3-H. Final confirmation of the stereochemistry was achieved by reduction and deprotection of a single enantiomer of carbamate 4, obtained via resolution of bicyclic ester 2a, to provide 2-amino-3-hydroxycyclohexane-1-carboxylate with identical spectroscopic and analytical data to that described in the literature ([α]D21 -32 (c = 1, H2O) lit: [α]D21 -35 (c = 1, H2O).[22] We attribute the stereochemistry of the reduction of 4 to the directing effect of the NBoc carbamate vide infra, whilst the reversal observed with 5 can be attributed to the severe steric crowding of the β-face by the three syn substituents.

thumbnailFigure 1. Selected NOESY Correlations for 69

3,4-Dihydroxy-2-aminocyclohexanecarboxylic acids

Having established an efficient entry to the oxanorbornenes and dienyl carbamates we then sort to exploit the alkene units to provide a convenient source of poly-hydroxylated cyclohexane β-aminoacids. Consequently, treatment of endo derived hydroxy diene with mCPBA in DCM afforded a separable 9:1 mixture of epoxides 10 and 11 favouring addition syn to the carbamate, Scheme 3. Highly selective epoxidations of cycloalkenyl amides have been described[23] and, consistent with a hypothesis that the mixture of 10 and 11 arise through competing H-bonding directing effects of the carbamate and hydroxyl groups, acetylation of the free hydroxyl group prior to oxidation enhanced the selectivity leading to exclusive formation of 12. In a similar fashion, and in agreement with the idea of two directing groups working in concert, epoxidation of the exo derived diene 5 afforded exclusive formation of epoxide 13. Speculating that disruption of any directing effects through the use of more polar solvents would lead to increased amounts of 12 a range of alternative solvents were screened including THF, DMF, MeOH, CF3CH2OH, MeCN. Of these, acetonitrile proving the most successful, albeit only providing a 2:1 mixture of isomers. Attempts to further enhance this reversal of selectivity using a variety of alternative procedures were not successful with most methods affording significant quantities of the corresponding protected anthranilic acid.

Scheme 3

Reagents: i. mCPBA, CH2Cl2 11 : 12 (9 : 1) 77%, 15 82%; ii. Ac2O, Py then mCPBA, CH2Cl2 (13 only) 65%; iii. mCPBA, MeCN, NaHCO3 11 : 12 (1 : 2) 95%; iv. Ac2O, pyridine; v. H2, Pd/C, 14 88% from 12, 15 84% from 11, 16 97%.

Having introduced the epoxide function to either face of the diene unit we then sought conditions for regioselective cleavage. In this context, palladium mediated reactions of vinyl epoxides are known to proceed with allylic cleavage.[24] Subsequent addition of hydrogen to the intermediate π-allyl palladium complex would provide the desired 4-hydroxy-ACHC derivative. In agreement with this analysis, reaction of the epoxides with hydrogen in the presence of Pd-C afforded the corresponding 3,4-dihydroxy-2-aminocyclohexancarboxylate derivates 14–16 in excellent yield and diastereoselectivity. In each case the reduction afforded the product with a cis relationship between the 3-hydroxy group and C-1 carboxylate as ascertained by NOESY experiments, figure 2. Whilst reactions of allylic electrophiles normally occur with anti attack by the Pd catalyst, the result using 12 suggests possible involvement of the carbamate. In this case the exception occurs with 13 which can be attributed to the severe steric crowding inhibiting such a co-ordination mode.

thumbnailFigure 2. Selected NOESY Correlations for 1416 (R = Boc)

3,4,5-Trihydroxy-2-aminocyclohexanecarboxylic acids

The final group of targets we wished to generate was the trihydroxy analogs. Initial attempts addressed the generation of the syn 4,5 set. Consequently, OsO4 mediated dihydroxylation and subsequent peracylation of dienes 4 and 5 proved to be facio-specific and afforded cyclohexenyl derivatives 18 and 20 in 75 % yield over the two steps, Scheme 4. Reduction, as previously, afforded the trihydroxy β-aminoacid derivatives in excellent yields.

Scheme 4

Reagents: i. OsO4, Me3NO.H2O, acetone 18 78% from 4 ii. TBSCl, imidazole, CH2Cl2, 17 66% from 2b; iii. LiHMDS, THF, -50-25 °C, 48%; iv. nBu4NF, THF; v. Ac2O, pyridine; vi. H2, Pd-C, 21 67% from 18, 22 51% from 19, 23 74% from 5.

At this stage it is instructive to address the interesting stereochemical outcome of this dihydroxylation reaction. Concurring with our results, Donohoe has reported that cyclic homoallylic carbamates give high levels of syn selectivity in the OsO4 mediated dihydroxylation reactions.[25,26] On the other hand, Kishi has established that the OsO4 mediated oxidation of cyclic allylic alcohols led preferentially to syn-anti triols.[27,28] On the basis of the results from our system where the two processes acted against each other, one can conclude that the homoallylic carbamate ability of stereocontrolling the OsO4 mediated reaction overrides that of the allylic hydroxyl group. Moreover, such directing effects are sufficiently strong that attempts using TMEDA/OsO4 combinations known to afford syn dihydroxylation of cyclic allylic alcohols were not successful in reversing the selectivity of the reaction with 4.

In order to introduce the syn diol unit on the opposite face to the carbamate in 4 we returned to the oxanorborne cycloadduct in which electrophile additions are known to occur from the exo face. As predicted, smooth selective formation of the desired exo diol 17 was achieved. Following double protection as a bis TBS ether, initial attempts to achieve the desired fragmentation using KHMDS failed. Fortunately, use of LiHMDS was successful albeit accompanied by the unexpected loss of Boc group. Similar strategies using either the free diol or the corresponding acetonide were not successful leading to extensive decomposition. Following TBAF promoted desilylation, peracylation and reduction afforded the trans β-amino acid derivative 22 as established by NMR experiments.

Having prepared the 4,5-cis diol, we turned our attention to the synthesis of the corresponding trans diols. These could be prepared by simple hydrolytic cleavage of epoxides 10–13. In line with this plan, treatment of 12 with aqueous perchloric acid led to a single trans diol, albeit accompanied by the loss of the Boc protecting group. Characterisation of the peracetylated derivative 27 suggested that nucleophilic attack had occurred at the allylic position. Support for this was confirmed by ring opening with ZnCl2 to afford the crystalline chlorohydrin 28 which was characterised by x-ray diffraction. Reduction of the double bond then afforded the expected all anti trihydroxy ACHC isomer 24 as confirmed by 1D and 2D NMR experiments. The observation of 10.0 Hz HaHb coupling constants in the 1H NMR spectrum of 24 for the ring protons (H-1, H-2, H-3, H-4 and H-5) is consistent with anti diaxial relationships. Such assignments are supported by nOe interactions between H-1 and H-3, H-1 and H-5 and also H-2 and H-4, Figure 3.

thumbnailFigure 3. Selected NOESY correlations for 24 26

Similar treatment of epoxides 11 and 13 then afforded the corresponding anti 4,5 diols. Subsequent reduction of the double bond gave the anti-syn-anti-anti and anti-syn-syn-anti isomers 25 and 26 as the only detectable products, Scheme 5. In each case it is noteworthy that the reduction proceeded with delivery of hydrogen to the same face as that occupied by the carbamate and opposite to the 3-acetoxy group. As previously, the stereochemical relationships were initially assigned on the basis of 2D NMR experiments, particularly NOESY. Subsequently, the structure of 26 was confirmed by X-ray crystallographic analysis.

Scheme 5

Reagents: i. HClO4, H2O/acetone; v. Ac2O, pyridine; iii. H2, Pd/C, 24 71%, 25 67%, 26 71%

Attempt to generate the remaining epoxide isomer were confounded by an inability to overcome the directing influence of the carbamate and deliver the epoxide to the opposite face of diene 5. Similarly, whilst the alternative regiochemistry in the epoxide ring opening product could be achieved exploiting neighbouring group participation by the NBoc group to afford the cyclic carbamate 29, all attempts to induce fragmentation of this polycyclic compound have thus far proved unsuccessful.

Conclusion

In conclusion, we have amplified the utility of the Diels-Alder adducts of ethyl (E)-nitroacrylate and furan by devising stereoselective routes to hydroxylated derivatives of 2-aminocyclohexanecarboxylic acid. The principal feature of this approach is the reliance on simple reactions exploiting substrate-controlled selectivity. In this respect, the ability of the carbamate group as a directing group is particularly noteworthy. The use of these hydroxylated cyclohexyl β-amino acids in the synthesis of short peptides is currently ongoing in our laboratory and will be described in due course.

Description of Additional Materials

Experimental procedures and spectroscopic data are provided for all new compounds including details of the X-ray diffraction studies for compounds 2b, 26 and 28. (see 1)

Additional File 1. Additional file 1 (doc) contains experimental details

Format: DOC Size: 446KB Download file

This file can be viewed with: Microsoft Word ViewerOpen Data

Acknowledgements

We thank the University of Botswana (IBM) for financial support, Dr A. M. Kenwright for assistance with NMR experiments, Dr M. Jones and the EPSRC National Mass Spectrometry Service, Swansea for mass spectra and Dr Darren Orton for some preliminary experiments.

References

  1. Liu M, Sibi MP:

    Tetrahedron. 2002, 58:7991-8035. Publisher Full Text OpenURL

  2. Fülöp F, Martinek TA, Tóth GK:

    Chem Soc Rev. 2006, 35:323-334. PubMed Abstract | Publisher Full Text OpenURL

  3. Gellman SH:

    Acc Chem Res. 1998, 31:173-180. Publisher Full Text OpenURL

  4. Appella DH, Christianson LA, Karle IL, Powell DR, Gellman SH:

    J Am Chem Soc. 1996, 118:13071-13072. Publisher Full Text OpenURL

  5. Cheng RP, Gellman SH, DeGrado WF:

    Chem Rev. 2001, 101:3219-3232. PubMed Abstract | Publisher Full Text OpenURL

  6. Raguse TL, Lai JR, Le Plae PR, Gellman SH:

    Org Lett. 2001, 3:3963-3966. PubMed Abstract | Publisher Full Text OpenURL

  7. Appella DH, Christianson LA, Klein DA, Powell RD, Huang X, Barchi JJJ, Gellman HS:

    Nature. 1997, 387:381-384. PubMed Abstract | Publisher Full Text OpenURL

  8. Gademann K, Seebach D:

    Helv Chim Acta. 2001, 84:2925-2937. Publisher Full Text OpenURL

  9. Guichard G, Abele S, Seebach D:

    Helv Chim Acta. 1998, 81:187. Publisher Full Text OpenURL

  10. Masesane IB, Steel PG:

    Synlett. 2003, 735-737. OpenURL

  11. Masesane IB, Steel PG:

    Tetrahedron Lett. 2004, 45:5007-5009. Publisher Full Text OpenURL

  12. McMurry JE, Musser JH:

    Org Synth. 1977, 56:65-68. OpenURL

  13. Danishefsky S, Prisbylla MP, Hiner S:

    J Am Chem Soc. 1978, 100:2918.

    For an exception to this observation see reference 15

    Publisher Full Text OpenURL

  14. Clarke C, Fleming I, Fortunak JMD, Gallagher PT, Honan MC, Mann A, Nubling CO, Raithby PR, Wolff JJ:

    Tetrahedron. 1988, 44:3931.

    For an exception to this observation see reference 15

    Publisher Full Text OpenURL

  15. Michael JP, Blom NF, Boeyens JCA:

    J Chem Soc Perkin Trans. 1984, 1:1739. Publisher Full Text OpenURL

  16. Just G, Martel A, Grozinger K, Ramjeesingh M:

    Can J Chem. 1975, 53:131-137. Publisher Full Text OpenURL

  17. Brion F:

    Tetrahedron Lett. 1982, 23:5299-5302. Publisher Full Text OpenURL

  18. Rajapaksa D, Keay BA, Rodrigo R:

    Can J Chem. 1984, 62:826-827. Publisher Full Text OpenURL

  19. Campbell MM, Sainsbury M, Searle PA:

    Synthesis. 1993, 179-193. Publisher Full Text OpenURL

  20. Evans DA, Barnes DM:

    Tetrahedron Lett. 1997, 38:57-58. Publisher Full Text OpenURL

  21. Couche E, Deschatrettes R, Poumellec K, Bortolussi M, Mandville G, Bloch R:

    Synlett. 1999, 87-89.

    See also reference 20

    OpenURL

  22. McCormick DRJ, Reichenthal J, Hirsch U, Sjolander NO:

    J Am Chem Soc. 1961, 83:4104-4105. Publisher Full Text OpenURL

  23. Barrett S, O'Brien P, Steffens HC, Towers TD, Voith M:

    Tetrahedron. 2000, 56:9633-9640. Publisher Full Text OpenURL

  24. Trost BM, Angle SR:

    J Am Chem Soc. 1985, 107:6123-6124. Publisher Full Text OpenURL

  25. Donohoe TJ, Mitchell L, Waring MJ, Helliwell M, Bell A, Newcombe NJ:

    Tetrahedron Lett. 2001, 42:8951-8954. Publisher Full Text OpenURL

  26. Donohoe TJ, Blades K, Moore PR, Waring MJ, Winter JJG, Helliwell M, Newcombe NJ, Stemp G:

    J Org Chem. 2002, 67:7946-7956. PubMed Abstract | Publisher Full Text OpenURL

  27. Cha JK, Christ WJ, Kishi Y:

    Tetrahedron Lett. 1983, 24:3943-3946. Publisher Full Text OpenURL

  28. Cha JK, Christ WJ, Kishi Y:

    Tetrahedron. 1984, 40:2247-2255. Publisher Full Text OpenURL

< previous full text next >