View PDF VersionPrevious ArticleNext Article
 
DOI: 10.1039/D4CC03655G (Communication) Chem. Commun., 2024, 60, 9946-9949

Fast one-step acrylate functionalization of hyaluronic acid via Williamson ether synthesis†

Cameron Milne a, Rijian Song a, Runqi Zhu a, Melissa Johnson a, Chunyu Zhao a, Francesca Santoro Ferrer a, Sigen A b, Jing Lyu a and Wenxin Wang *ac
aCharles Institute of Dermatology, School of Medicine, University College Dublin, Dublin 4, Ireland. E-mail: wenxin.wang@ucd.ie
bSchool of Medicine, Anhui University of Science and Technology, Huainan, China
cResearch and Clinical Translation Center of Gene Medicine and Tissue Engineering, School of Public Health, Anhui University of Science and Technology, Huainan, China

Received 21st July 2024 , Accepted 8th August 2024

First published on 15th August 2024

Abstract

The synthetic route presented for acrylate-modified hyaluronic acid (HA-A-BEA) offers a simple and efficient process, reducing reaction time and purification steps while retaining biocompatibility. This study demonstrates the ability of HA-A-BEA to form tunable hydrogels via versatile techniques suitable for biomedical applications.

Modified naturally occurring biopolymers have emerged as a cornerstone in the field of tissue engineering hydrogel research.1 The ability to synergistically utilize their inherent biological function and tailorable mechanical properties makes them ideal candidates for improving patient outcome in many biomedical applications.2–4 Hyaluronic acid (HA) is a naturally occurring polysaccharide and a major constituent of the extracellular matrix in the human body.5 HA consists of alternating units of D-glucuronic acid and N-acetyl-D-glucosamine, which endow HA with unique properties. This structure allows HA to contribute to various functions, including providing structure, lubrication and a cellular signalling platform for activities such as proliferation and migration.6 This structure also allows for versatile modification due to the lability of the hydroxyl (1° and 2°) and carboxylic acid groups, providing an ideal platform for tailorable and functional hydrogel scaffold design in tissue engineering.7,8 Some of the most frequently employed modified HAs are vinyl group modified HAs, such as methacrylate-modified HA (HA-MA) and acrylate-modified HA (HA-A). Modifying HA with vinyl groups enables crosslinking capabilities, such as free radical polymerisation, giving rise to rapid crosslinking and improved mechanical properties (<1 min).9,10 HA-A differs from HA-MA in that the α,β unsaturated vinyl group is able to selectively react with thiol-modified polymers, such as thiol-modified HA (HA-SH) for hydrogel preparation. This thiol-Michael addition reaction provides a biorthogonal and chemo selective method for crosslinking which is highly efficient, can be carried out at physiological pH (pH 7.4) and is fast which allows for in situ hydrogel formation.11–13 The thiol-Michael addition reaction has been utilised for synthesizing many in situ crosslinking HA hydrogels for drug delivery and tissue engineering applications.14–16

Despite several methods developed for manufacturing HA-A being reported, widespread adoption has been limited due to long reaction times (up to 5 days), reagent compatibility issues, and multi-step reaction and purification processes.17–20 Wang et al. previously reported a facile synthesis of HA-A with glycidyl acrylate as an intermediate in a one-pot reaction, which greatly reduced the cost involved and the need for extended purification.21 However, a long reaction time (5 days) is still required to favour the ring-opening reaction through the epoxide group over the reversible transesterification. This work presents a novel synthetic route for HA-A (HA-A-BEA) via a Williamson ether synthesis reaction (Scheme 1). This method uses a single modification reagent, 2-bromoethyl acrylate (BEA), a common base catalyst (triethylamine (TEA)) and only 12 hours of reaction time, without the need for multiple laborious purification steps.


image file: d4cc03655g-s1.tif
Scheme 1 Synthetic route for HA-A-BEA and hydrogel formation through a free radical polymerisation reaction with photoinitiator lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP) (HA-A-VL) and through a thiol-Michael addition reaction with HA-SH (HA-A-CX). Created with Biorender.com.

The Williamson ether synthesis is a reliable and versatile method for the synthesis of ethers, and is widely used in synthetic chemistry.22–24 Dimethylformamide (DMF) was used as a co-solvent to improve the solubility of BEA in the reaction mixture, a frequently reported technique in HA modification synthesis.25,26 The HA-A-BEA synthesis proceeds through the reaction of the primary alcohol on the HA structure and the α-alkyl-bromide on the BEA structure. TEA was used as a catalyst as it is a sterically hindered base and under the high pH conditions induced by TEA, the primary alcohol in HA is predominantly deprotonated, forming primary alkoxides.8 Since primary alkoxides are more nucleophilic than carboxylates, acrylate substitution on the primary alcohol is favoured.27 This selective reaction helps preserve the biological functions of native HA, as the ionic interactions involving HA's carboxylic acid groups are crucial for cellular signaling.28,29

Excess TEA was removed by dialysis in acidic (pH 3.5) H2O on the first day before further dialysis in purified H2O for 2 days to remove excess BEA. The use of 2-chloroethyl acrylate (CEA) was also investigated for this reaction but 1H-NMR analysis showed that the reaction had been unsuccessful, even after a reaction at a molar feed ratio of 50[thin space (1/6-em)]:[thin space (1/6-em)]1 (CEA[thin space (1/6-em)]:[thin space (1/6-em)]HA). It is believed that this is due to bromine being less electronegative than chlorine, and the halide to α carbon bond is therefore less stable in BEA. Conjugation of BEA to HA was confirmed by the presence of proton peaks in the vinylic region (5.9–6.6 ppm) in the 1H-NMR spectrum (Fig. S1, ESI†). The degree of acrylate substitution (A-SD) was calculated by the comparison between the integrated HA methyl group peak on the N-acetyl-D-glucosamine peak at 2.01 ppm and the vinyl proton peaks at 6.0, 6.3 and 6.5 ppm in the 1H-NMR spectrum (eqn (S1), ESI†). A range of A-SD (10–40%) was achieved by adjusting the BEA molar feed ratio from 17 to 40. Fig. S2 (ESI†) shows a clear proportional relationship between SD and the BEA feed ratio from a range of HA-A-BEA batches demonstrating the tunability of the A-SD%.

These findings underscore the efficacy and predictability of the reaction conditions, establishing a robust foundation for scaling up and applying HA-A-BEA in hydrogel-based biomedical and therapeutic contexts.

To showcase the biocompatibility of the HA-A-BEA biopolymer, alamarBlue assays and live/dead staining were conducted using Normal Human Dermal Fibroblasts (NHDFs) over 24 and 72 h. As seen in Fig. 1, there is no significant reduction in cell viability after exposure to HA-A-BEA. According to ISO 10993-5:2009, this indicates that there is no significant cytotoxicity for HA-A-BEA over both time periods, showing that this acrylate modification maintains high biocompatibility within the HA structure.


image file: d4cc03655g-f1.tif
Fig. 1 (A) Cytocompatibility of the HA-A-BEA biopolymer after 24 and 72 h co-culture with NHDF cells. Data are presented as average ± standard deviation (n = 3). (B) LIVE/DEAD staining images of NHDF cells co-cultured with HA-A-BEA (2.5 mg mL−1) for 72 h. Scale bar = 100 μm.

The crosslinking ability of HA-A-BEA and the mechanical properties of HA-A-BEA hydrogels were investigated via a rheological assessment. HA-A-BEA was crosslinked with visible light (VL, 405 nm) (HA-A-VL), with photoinitiator lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP) and with HA-SH (with a range of thiol degrees of substitution, SH-SD%) via a thiol-Michael addition reaction (HA-A-CX). The formulations used in the HA-A-CX and HA-A-VL hydrogels are shown in Table 1. The crosslinking behaviour and mechanical properties of these HA-A-BEA hydrogels were initially evaluated via a rheological time-sweep experiment. Fig. 2A shows the time sweep rheology study results of the HA-A-CX hydrogels. The time sweep curves show an accelerated increase in the storage modulus (G′) over time for the systems with increased concentration or with excess HA-SH (HA-A-CX2) or HA-A (HA-A-CX3) substitution degrees (SDs). HA-A-CX2 hydrogels crosslink more rapidly (G′ 202.7 kPa after 10 min, HA-A-CX2-2%) than HA-A-CX1 and HA-A-CX3 as there are more thiol groups available for crosslinking (1.8[thin space (1/6-em)]:[thin space (1/6-em)]1, SH[thin space (1/6-em)]:[thin space (1/6-em)]A molar ratio, Table 1), leading to more frequent encounters between reactive groups. The second fastest crosslinking hydrogel was HA-A-CX3 (G′ 116.2 kPa after 10 min, HA-A-CX3-2%) due to the higher acrylate molar content compared to HA-A-CX1 (G′ 61.4 kPa after 10 min, HA-A-CX1-2%). The G′ value of HA-A-CX hydrogels after 60 min of crosslinking increased from 0.5 kPa (HA-A-CX1-1%) to 2.3 kPa (HA-A-CX2-2%) by varying biopolymer SD and concentrations. HA-A-VL hydrogels were formed with HA-A-BEA concentrations of 1 and 2% (w/v) with 0.5% LAP (w/v). The hydrogels formed instantly after exposure to VL.

Table 1 HA-A-BEA hydrogel formulations
Name HA-A-BEA conc.a (%) A-SDb (%) HA-SH conc.a (%) SH-SD (%) A:SH molar ratio
a Concentrations provided are w/v%. b A-SD values were calculated via1H-NMR analysis.
HA-A-CX1-1% 1 18 1 18 1[thin space (1/6-em)]:[thin space (1/6-em)]1
HA-A-CX1-2% 2 18 2 18 1[thin space (1/6-em)]:[thin space (1/6-em)]1
HA-A-CX2-1% 1 18 1 33 1[thin space (1/6-em)]:[thin space (1/6-em)]1.8
HA-A-CX2-2% 2 18 2 33 1[thin space (1/6-em)]:[thin space (1/6-em)]1.8
HA-A-CX3-1% 1 27 1 18 1[thin space (1/6-em)]:[thin space (1/6-em)]0.7
HA-A-CX3-2% 2 27 2 18 1[thin space (1/6-em)]:[thin space (1/6-em)]0.7
HA-A-VL-1% 1 18 N/A N/A N/A
HA-A-VL-2% 2 18 N/A N/A N/A


image file: d4cc03655g-f2.tif
Fig. 2 (A) Time sweep G′/G′′ assessment of HA-A-CX hydrogels via time sweep rheology studies at 1 Hz and 1% strain. (B) Maximum G′ values of HA-A-BEA hydrogels obtained at 1 Hz and 1% strain. Data are presented as average ± standard deviation (n = 3). (C) Compression test for HA-A-BEA hydrogels compressed at 1% strain/s. (D) Frequency sweep (0–10 Hz) test for HA-A-BEA hydrogels at 1% strain. (E) Degradation profiles of HA-A-BEA hydrogels in PBS and 100 U mL−1 HYASE. Data are presented as average ± standard deviation (n = 3).

As the G′ value of the HA-A-CX hydrogels was still increasing after 60 min, the final hydrogel G′ values were taken after 24 hours of preparation (Fig. 2B). The final G′ values for HA-A-VL-1% and HA-A-VL-2% were higher than that of the equivalent concentration HA-A-CX hydrogels due to the denser crosslinking nature of photo crosslinked hydrogels. In HA-A-CX2 hydrogels, the excess thiol groups were able to crosslink through a disulfide bond leading to a higher crosslinking density and greater max G′ than those of the other HA-A-CX hydrogels. HA-A-CX3 hydrogels showed a higher max G′ value than HA-A-CX1 due to the increased acrylate molar content but lower than HA-A-CX2 due to the lack of excess thiol disulfide crosslinking. However, for the 1% hydrogels, there was little difference between the G′ values of HA-A-CX1, HA-A-CX2 and HA-A-CX3, and at this lower concentration, there were fewer crosslinks and the impact of crosslinking density may be mitigated. Overall, all examined HA-A-BEA hydrogel formulations demonstrated the capacity to finely adjust both G′ and crosslinking potential by manipulating the SD and concentration of their constituent materials. The above behaviours observed in HA-A-BEA hydrogels align with established trends in modified HA, demonstrating their viability as promising candidates for achieving future bioengineered hydrogel advancements.26,30,31

The ability of a hydrogel to withstand mechanical stresses is an important factor when designing a hydrogel system.32 Compressive stress limits were tested for both the HA-A-VL and HA-A-CX hydrogels to test the hydrogels' compressive strength (Fig. 2C). All compressive stress limit results for HA-A-BEA hydrogels showed an expected trend of an increased stress limit but a decreased strain limit with an increase of the HA-A-BEA concentration from 1 to 2% (w/v). Among 2% hydrogels, HA-A-CX3 hydrogels were able to withstand a higher compressive strain than VL, CX1 and CX2 hydrogels. This is due to the larger pore density resulting from the excess acrylate groups compared to thiol groups, which allows the hydrogel to deform with greater flexibility. HA-A-CX2 contains excess thiol groups, which lead to an increase in its crosslinking density at both 1% and 2% concentrations, enhancing the hydrogel's ability to withstand higher stress loads. However, the higher crosslinking density also makes the hydrogel more brittle, as indicated by their lower critical strain values than those of HA-A-CX1 and HA-A-CX3 at equivalent concentrations. As expected, the HA-A-VL hydrogel exhibited the lowest critical strain values among hydrogels of the same concentrations due to it densely crosslinked structure.

To further explore the viscoelastic behaviour of the hydrogels, frequency sweep experiments were conducted (Fig. 2D). A high G′ compared to the loss modulus (G′′) indicates strong intermolecular interactions within the gel structure. This suggests the hydrogel's ability to resist intermolecular slippage and that the hydrogel has a robust three-dimensional network formed by these interactions. Moreover, as shown in Fig. 2D, the G′ value displayed a nearly frequency-independent trend between 0.1 and 10 Hz and all samples exhibited consistent behaviour characteristic of elastic gels, with G′ consistently greater than G′′.

The swelling and degradation profiles of HA-A-BEA hydrogels in the presence of PBS and hyaluronidase (HYASE, 100 U mL−1) as a function of time are presented in Fig. 2E. HA-A-VL-2% and HA-A-CX1-2% were chosen as representatives of the photo crosslinked and chemically crosslinked systems, respectively. In PBS, the HA-A-CX1 hydrogel initially swelled and then subsequently degraded more quickly than the HA-A-VL hydrogel due to a larger pore size. The HA-A-VL hydrogel, being more tightly crosslinked, exhibited limited swelling and less mass loss over time. Once stable, both hydrogels lost minimal mass over 31 days. When exposed to HYASE, HA-A-CX1 degraded more quickly, as expected, due to its less dense network structure that makes the HA chains more accessible to enzymatic degradation. The increased rate of degradation in the presence of HYASE suggests that the HA backbone structure is well preserved after HA-A-BEA synthesis.

In summary, this work showcased a novel and facile synthetic route, through a Williamson ether synthesis, for acrylate-modified HA (HA-A-BEA) with only one reagent, BEA, that reduces costs and reaction time and eliminates the need for extensive purification. This reaction represents a notable improvement over previous approaches, enhancing the feasibility of large-scale production and offering versatility in acrylate substitution percentages (10–40%) by adjusting the BEA feed ratio. Based on HA-A-BEA, the formation of tunable and stable hydrogels via biorthogonal Michael addition reactions and visible light photo-crosslinking is reported. Rheological assessments confirm the robustness of HA-A-BEA hydrogels, and the biocompatibility of HA-A-BEA was affirmed by high in vitro cell viability, showcasing their high potential for advanced tissue engineering applications. Overall, our findings underscore the significance of this synthesis route in advancing the development of tailored HA-based hydrogel systems for biomedical applications and beyond.

The ToC figure was created by BioRender.com.

Data availability

Data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Notes and references

  1. S. Van Vlierberghe, P. Dubruel and E. Schacht, Biomacromolecules, 2011, 12, 1387–1408 CrossRef CAS PubMed .
  2. Y. S. Zhang and A. Khademhosseini, Science, 2017, 356, eaaf3627 CrossRef PubMed .
  3. A. La Gatta, E. Bedini, M. Aschettino, R. Finamore and C. Schiraldi, Polymers, 2022, 14, 2402 CrossRef CAS PubMed .
  4. D. K. Patel, E. Jung, S. Priya, S.-Y. Won and S. S. Han, Carbohydr. Polym., 2024, 323, 121408 CrossRef CAS PubMed .
  5. A. Fallacara, E. Baldini, S. Manfredini and S. Vertuani, Polymers, 2018, 10, 701 CrossRef PubMed .
  6. J. A. Burdick and G. D. Prestwich, Adv. Mater., 2011, 23, H41–H56 CrossRef CAS PubMed .
  7. S. Trombino, C. Servidio, F. Curcio and R. Cassano, Pharmaceutics, 2019, 11, 407 CrossRef CAS PubMed .
  8. C. E. Schanté, G. Zuber, C. Herlin and T. F. Vandamme, Carbohydr. Polym., 2011, 85, 469–489 CrossRef .
  9. C. Milne, R. Song, M. Johnson, C. Zhao, F. Santoro Ferrer, J. Lyu and W. Wang, Biomacromolecules, 2024, 25, 2645–2655 CrossRef CAS PubMed .
  10. M. Zeng, M. Johnson, J. Creagh-Flynn, Q. Xu, H. Tai and W. Wang, Biomacromolecules, 2020, 21, 2229–2235 CrossRef PubMed .
  11. T. M. FitzSimons, E. V. Anslyn and A. M. Rosales, ACS Polym. Au, 2022, 2, 129–136 CrossRef CAS PubMed .
  12. M. J. Kade, D. J. Burke and C. J. Hawker, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 743–750 CrossRef CAS .
  13. D. Berne, V. Ladmiral, E. Leclerc and S. Caillol, Polymers, 2022, 14, 4457 CrossRef CAS PubMed .
  14. J. L. Vanderhooft, B. K. Mann and G. D. Prestwich, Biomacromolecules, 2007, 8, 2883–2889 CrossRef CAS PubMed .
  15. M. A. Serban and G. D. Prestwich, Biomacromolecules, 2007, 8, 2821–2828 CrossRef CAS PubMed .
  16. S. K. Hahn, E. J. Oh, H. Miyamoto and T. Shimobouji, Int. J. Pharm., 2006, 322, 44–51 CrossRef CAS PubMed .
  17. Y. Lei, S. Gojgini, J. Lam and T. Segura, Biomaterials, 2011, 32, 39–47 CrossRef CAS PubMed .
  18. J. Becher, S. Möller and M. Schnabelrauch, Carbohydr. Polym., 2013, 93, 438–441 CrossRef CAS PubMed .
  19. S. Khetan, J. S. Katz and J. A. Burdick, Soft Matter, 2009, 5, 1601–1606 RSC .
  20. S. Khetan and J. A. Burdick, Biomaterials, 2010, 31, 8228–8234 CrossRef CAS PubMed .
  21. S. A. Q. Xu, P. McMichael, Y. Gao, X. Li, X. Wang, U. Greiser, D. Zhou and W. Wang, Chem. Commun., 2018, 54, 1081–1084 RSC .
  22. K. Kokado, A. Nagai and Y. Chujo, Macromolecules, 2010, 43, 6463–6468 CrossRef CAS .
  23. C. Lin and I. Gitsov, Macromolecules, 2010, 43, 10017–10030 CrossRef CAS .
  24. H. Peng, M. Lei, L. Yang, J. Sun, L. Yang and Q. Zhao, ACS Omega, 2020, 5, 21468–21475 CrossRef CAS PubMed .
  25. E. Hachet, H. Van Den Berghe, E. Bayma, M. R. Block and R. Auzély-Velty, Biomacromolecules, 2012, 13, 1818–1827 CrossRef CAS PubMed .
  26. S. A. Bencherif, A. Srinivasan, F. Horkay, J. O. Hollinger, K. Matyjaszewski and N. R. Washburn, Biomaterials, 2008, 29, 1739–1749 CrossRef CAS PubMed .
  27. V. Hintze, M. Schnabelrauch and S. Rother, Front. Chem., 2022, 10, 830671 CrossRef CAS PubMed .
  28. X. Zhang, P. Sun, L. Huangshan, B.-H. Hu and P. B. Messersmith, Chem. Commun., 2015, 51, 9662–9665 RSC .
  29. V. B. Lokeshwar and M. G. Selzer, J. Biol. Chem., 2000, 275, 27641–27649 CrossRef CAS PubMed .
  30. C. C. L. Schuurmans, M. Mihajlovic, C. Hiemstra, K. Ito, W. E. Hennink and T. Vermonden, Biomaterials, 2021, 268, 120602 CrossRef CAS PubMed .
  31. M. Pérez-Lloret and A. Erxleben, ACS Omega, 2024, 9, 25914–25921 CrossRef PubMed .
  32. X. Lin, X. Zhao, C. Xu, L. Wang and Y. Xia, J. Polym. Sci., 2022, 60, 2525–2542 CrossRef CAS .

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4cc03655g
This journal is © The Royal Society of Chemistry 2024