Side-chain functionalisation of unsaturated polyesters from ring-opening polymerisation of macrolactones by thiol–ene click chemistry

Abstract

A novel method for the modification of polyglobalide, produced by enzymatic ring-opening polymerisation, with pendant side-chains viathiol–ene click chemistry is disclosed. This methodology may offer a chemically simplistic route to the (bio)functionalisation of polyesters.

---

Aliphatic polyesters have received significant recent attention owing to their use in a number of biomedical applications, with key research focussed particularly on their use as degradable drug delivery vehicles and in tissue engineering.¹ Notable advantageous material properties include excellent biodegradability, biocompatibility, and nontoxicity. In addition, the production of biocompatible degradation products renders the application of polyesters as biomaterials suitably appropriate. Aliphatic polyesters can be prepared by either a step-growth polycondensation process or the ring-opening polymerisation (ROP) of cyclic esters (lactones). The popularity of the latter methodology stems from the fact that it produces polyesters with a high degree of control, with respect to molecular weight, polydispersity and end-group fidelity, under relatively mild reaction conditions. A wide range of catalytic systems have been reported for the ROP of cyclic esters including organometallic,² organic,^3,4 and enzyme catalysts.^5,6

Despite the suitability of incorporating polyesters into biomaterials, their use in highly sophisticated biomedical applications is hindered by the difficulty of polyester side-chain functionalisation. Material modification to allow specific interactions within biological systems significantly enhances biomaterial performance, potentially offering the material a dynamic role in its application. However, as recently highlighted by Pounder and Dove, the introduction of functional groups throughout the polymer chain viaROP remains highly challenging.⁷ The chemical polymerisation process is non-selective and does not tolerate the presence of even the most simple functional groups, such as hydroxy or amines, without inevitably producing a cross-linked polymer. Despite this synthetic hindrance there are numerous examples of side chain functionalised linear polyesters produced from ROP which underlines the demand for these materials. However, most examples reported are produced via rigorous, often multi-step, synthetic procedures to derivatise lactone or lactide monomers.^7–9 An alternative synthetic approach is the derivatisation of suitable substituted aliphatic polyesters, in which a stable substituted lactone precursor is polymerised and the polymer subsequently modified selectively. Successful examples of this methodology include ‘click’ reactions such as the Michael addition reactions of thiols on polymers from γ-acrylic-ε-caprolactone¹⁰ and Huisgens cycloadditions on azide functionalised polymers derived from the nucleophilic substitution of halogen functionalised polycaprolactone.^11–13

In this paper we propose the chemically straightforward modification of aliphatic polyesters by utilising highly efficient and robust thiol–ene ‘click’ reactions directly on unsaturated polyesters obtained from ROP of unsaturated macrolactone globalide (Gl). This reaction may be employed for the post-polymerisation modification of unsaturated polyesters to allow the introduction of numerous different chemical groups to polymeric backbones, thereby potentially enabling the formation of biofunctional materials through considered materials engineering.^14,15 A few recent publications describe thiol–ene reactions on fatty acid monomers, however, to our knowledge this paper offers the primary example of the thiol–ene functionalisation of a fatty acid derived polymer.^16–18

We have previously reported the enzymatic synthesis of homopolymers from globalide, which is derived from hydroxy fatty acids and is proven to be non-toxic.¹⁹ Due to the low ring tension of macrolactones only enzymatic ROP permits fast polymerisation to form high molecular weight polymers.²⁰ The ROP of globalide (Scheme 1) was carried out in solution with Novozym 435 (Candida antarcticaLipase B (CALB) immobilised on macroporous resin) to a molecular weight of 16 000 g mol⁻¹ and a typically broad polydispersity of 2.5 due to transesterification reactions.

Scheme 1 Polymerisation of globalide and thiol–ene reaction of polyglobalide with 6-mercapto-1-hexanol (MH), butyl-3-mercapto propionate (BMP) and N-acetylcysteamine (nACA). Globalide is a mixture of two constitutional isomers with the double bond at the 11 or 12 position. For reasons of clarity only one isomer is shown.

The thiol–ene functionalisation of polyglobalide (PGl) was undertaken by a thermally initiated reaction catalysed by AIBN in the bulk above the melting point of PGl or with a minimum amount of solvent when required to enable complete reactant miscibility. Initially butyl-3-mercaptopropionate (BMP) was chosen as a suitable model compound for this polymeric functionalisation owing to its thiol functionality.

Generally, internal double bonds as present in PGl and some fatty acids are less reactive than terminal ones due to the reversibility of the C–S bond formation. We therefore paid specific attention to the quantification of the thiol–ene reaction on the PGl by NMR spectroscopy.

The ¹H NMR obtained for BMP functionalised PGl reveals characteristic peaks corresponding to the repeat units of PGl and signals from the BMP, which suggests a successful reaction. Significantly, the quantification of the thiol–ene coupling of BMP to PGl revealed a reduction of the double bond peak at around 5.4 ppm in the spectra of the final product, indicating that the thiol–ene coupling efficiency was greater than 75%.

The system proposed offers the opportunity to introduce a range of functionalities into the polyester backbone via the coupling of various thiol terminated linkers that also possess other chemical groups which offer potential for further chemical modification. 6-Mercapto-1-hexanol (MH) was selected as a suitable candidate for the introduction of a primary alcohol to the polymervia the methodology proposed. Upon formation, this alcohol-functionalised polyester may be further utilised for the introduction of carboxylic acid-terminated groups, such as amino acids, into the polyester via esterification as a method of biofunctionalisation. The ¹H NMRs of MH, PGl and the product formed are given in Fig. 1. Analogous to PGl functionalisation with BMP, a reduction in the peak corresponding to the polyester double bond was observed. By comparing the integration of the peaks that correspond to the double bond in PGl at around 5.4 ppm (a and a′) with the peaks representative of the methylene group adjacent to hydroxyl terminal group (b), it was determined that the fraction of thiol–ene coupling was in excess of 95% which highlights the high efficiency of the thiol–ene reaction undertaken. Moreover, all characteristic peaks of both MH and PGl are present in the product spectrum. The peak of the methylene group adjacent to the thiol group at 2.6 ppm (A) shifted slightly upfield to 2.45 ppm after the thiol–ene reaction. Most importantly, the signal of the methylene group adjacent to the hydroxyl group of 6-mercapto-1-hexanol was still detected at 3.6 ppm as a triplet (B) confirming the hydroxy functionalisation of the polyester.

Fig. 1 ¹H NMR spectra of 6-mercapto-1-hexanol (MH), polyglobalide (PGl) and the thiol–ene coupling product from MH and PGl. Spectra recorded in CDCl₃ using a 400 MHz spectrometer. Globalide is a mixture of two constitutional isomers with the double bond at the 11 or 12 position.

Amine functionality may be introduced into the polyester via the thiol–ene coupling of N-acetylcysteamine (nACA). Subsequent deacetylation would reveal free amine groups, affording a biofunctional polyester to which direct protein modification is readily achievable. Analysis of the formed acetylated thiol–ene product by NMR (Fig. 2) enabled the comparison of integrals corresponding to the double bond of PGl (5.4 ppm (a and a′)) with those representative of the amine proton (b). Optimal coupling conditions were found to give a reaction yield of ≥95% further highlighting the efficiency of the thiol–ene reaction.

Fig. 2 ¹H NMR spectra of N-acetylcysteamine (nACA) conjugated polyglobalide (PGl–nACA). The spectrum was recorded in CDCl₃ using a 400 MHz spectrometer.

Variation in the reaction conditions employed for the coupling of the three thiol-terminated side chains described to PGl had a marked effect on the coupling yield. Poor reactant miscibility dictated the necessary addition of THF to enhance thiol–ene reactions and consequent product formation for some reactions. However, excessive solvent addition reduces reactant concentrations to the detriment of the coupling yield (ESI, Table S1‡). The optimal reaction conditions found for the thiol–ene coupling of MH and nACA to PGl are given in Table 1. An excess of the thiol-bearing molecule was required for effective coupling due to the low reactivity of the PGl double bond. Further investigation into the coupling of nACA to PGl revealed that the initiator concentration also had a significant bearing on the reaction yield (ESI, Table S2‡). A 95% yield was recorded for the coupling of 6.6 mmol nACA to 0.01 mmol PGl (0.6 mmol double bonds) using 50 mg AIBN in contrast to a 60% yield when 10 mg AIBN was used.

Table 1 The optimal reaction conditions found for the thiol–ene coupling of three selected side-chains to PGl (M_n = 16 000 g mol⁻¹)

	PGl–BMP	PGl–MH	PGl–nACA
a The concentration of PGl double bonds was estimated from the molecular weight of the polymer as determined by GPC with polystyrene calibration. Due to the inherent error of this calculation only a relative comparison of the results can be made.
Double bond concentration^a/mmol	0.6	4.0	6.6
Initiator concentration/mg	10	10	50
Reaction temperature/°C	80	80	80
Solvent	None	1 mL THF	1 mL THF
M n/g mol^−1	15 900	26 000	23 500
M w/Mn	3.17	2.22	1.82
Coupling yield (%)	75	≥95	≥95

---

DSC analysis of the obtained thiol–ene functionalised polymers was performed to ascertain their thermal properties. PGl has a semi-crystalline structure with a melting point of 48 °C. As the vast majority of double bonds present on the polyester backbone partake in the thiol–ene reaction, the crystalline structure of the polymer is disrupted resulting in an amorphous material as shown clearly in the DSC thermograms presented in Fig. 3 (PGl, PGl–MH and PGl–nACA).

Fig. 3 DSC thermograms (second heating cycle) of polyglobalide (PGl) and polyglobalide after thiol–ene coupling independently with 6-mercapto-1-hexanol (PGl–MH) and N-acetylcysteamine (PGl–nACA).

The molecular weights and polydispersities of PGl, PGl–BMP, PGl–MH and PGl–nACA were determined by GPC. Analysis of the GPC data reveals a significant increase in the molecular weight and polydispersity of both PG–MH (26 000 g mol⁻¹) and PGl–nACA (23 500 g mol⁻¹) relative to unmodified PGl (16 000 g mol⁻¹) (Table 1). This increase in molecular weight corresponds well with the ¹H NMR data which suggested that in both instances a highly efficient thiol–ene coupling reaction occurred. A less significant change in M_n was observed for PGl–BMP relative to unfunctionalised PG, however, the modified polymer possessed a greater polydispersity in comparison to PG which made accurate molecular weight assignment difficult but suggests that PG functionalisation has occurred. It also has to be noted that upon thiol–ene grafting a branched polymer is produced, which makes comparison of GPC results with the linear precursor complex. Nevertheless, the shift in M_n, particularly in the formation of PGl–MH and PGl–nACA, validates this methodology for the functionalisation of PGl with three different thiol terminated linkers, offering a novel and chemically simplistic technique for polyester functionalisation.

The work detailed in this paper offers a simple route for the synthesis of thiol–ene functionalised polyesters. Enzymatic ROP of polyglobalide offers a highly controlled mechanism to create linear polyesters possessing alkene functionality. Click chemistry may be utilised to exploit this functionality via highly efficient thiol–ene coupling reactions to precisely introduce thiol terminated linkers pendant to the polyester backbone. The introduction of primary alcohol terminated 6-mercapto-1-hexanol and N-acetylcysteamine was demonstrated and both potentially enable further polymeric modification, via esterification or amidation of the deacetylated amine respectively, to introduce functional amino acid groups, amongst others, that possess a wide range of chemical and physical properties. This novel polymerisation therefore readily enables the highly efficient and chemically simplistic biofunctionalisation of a polyester compound to yield significant products of potential biomedical importance.

Acknowledgements

This work was funded by the Science Foundation Ireland (SFI) Principle Investigator Award 07/IN1/B1792. AH is a SFI Stokes Senior Lecturer (07/SK/B1241).

Notes and references

1. A. C. Albertsson and I. K. Varma, Biomacromolecules, 2003, 4, 1466–1486.
2. O. Dechy-Cabaret, B. Martin-Vaca and D. Bourissou, Chem. Rev., 2004, 104, 6147–6176.
3. A. P. Dove, R. C. Pratt, B. G. G. Lohmeijer, D. A. Culkin, E. C. Hagberg, G. W. Nyce, R. M. Waymouth and J. L. Hedrick, Polymer, 2006, 47, 4018–4025.
4. M. K. Kiesewetter, E. J. Shin, J. L. Hedrick and R. M. Waymouth, Macromolecules, 2010, 43, 2093–2107.
5. S. Kobayashi and A. Makino, Chem. Rev., 2009, 109, 5288–5353.
6. A. Heise and A. R. A. Palmans, Adv. Polym. Sci., 2010, 237, 79–113.
7. R. J. Pounder and A. P. Dove, Polym. Chem., 2010, 1, 260–271.
8. C. K. Williams, Chem. Soc. Rev., 2007, 36, 1573–1580.
9. C. Jerome and P. Lecomte, Adv. Drug Delivery Rev., 2008, 60, 1056–1076.
10. J. Rieger, K. van Butsele, P. Lecomte, C. Detrembleur, R. Jerome and C. Jerome, Chem. Commun., 2005, 274–276.
11. B. Parrish, R. Breitenkamp and T. J. Emrick, J. Am. Chem. Soc., 2005, 127, 7404–7410.
12. R. Riva, S. Schmeits, F. S. Stofelbach, C. Jerome, R. Jerome and P. Lecompte, Chem. Commun., 2005, 5334–5336.
13. R. Riva, S. Schmeits, C. Jerome, R. Jerome and P. Lecompte, Macromolecules, 2007, 40, 796–803.
14. A. Dondoni, Angew. Chem., Int. Ed., 2008, 47, 8995–8997.
15. A. B. Lowe, Polym. Chem., 2010, 1, 17–36.
16. O. Türünç and M. A. R. Meier, Macromol. Rapid Commun., 2010, 31, 1822–1866.
17. J. Samuelsson, M. Jonsson, T. Brinck and M. Johansson, J. Polym. Sci., Part A: Polym. Chem., 2004, 42, 6346–6352.
18. G. B. Bantchev, J. A. Kenar, G. Biresaw and M. G. Han, J. Agric. Food Chem., 2009, 57, 1282–1290.
19. I. van der Meulen, M. de Geus, H. Antheunis, R. Deumens, E. A. J. Joosten, C. E. Koning and A. Heise, Biomacromolecules, 2008, 9, 3404–3410.
20. M. de Geus, I. van der Meulen, B. Goderis, K. Van Hecke, M. Dorschu, H. van der Werff, C. E. Koning and A. Heise, Polym. Chem., 2010, 1, 525–533.

---

Footnotes

† This paper is part of a Polymer Chemistry issue highlighting the work of emerging investigators in the polymer chemistry field. Guest editors: Rachel O'Reilly and Andrew Dove.

‡ Electronic supplementary information (ESI) available: Experimental procedures and comparative product yield data. See DOI: 10.1039/c0py00294a

---