Role of Free Catecholamine in Thiol-Ene Crosslinking for Hyaluronic Acid Hydrogels with High Loading Efficiency of Anticancer Drugs

Abstract

Background:

Hyaluronic acid (HA) is a widely used polysaccharide in biomedical field because of its excellent biocompatibility. Its chemical structure can be modified with various functional groups. Recently, dopamine has been tethered onto the polymeric backbone to ensure long-term stability and tissue adhesiveness of HA hydrogel. However, the radical scavenging effect of dopamine on typical photo-induced crosslinking for hydrogels has not been specifically studied.

Methods:

Photo-crosslinkable norbornene-modified HA (NorHA) was synthesized and crosslinked by dithiothreitol containing dopamine at different concentrations. During in situ ultraviolet light-triggered crosslinking, storage moduli were monitored using an oscillatory rheometer. Additionally, the amount of thiol utilized for HA crosslinking was investigated under the presence and absence of dopamine. Finally, doxorubicin was encapsulated in the hydrogels, and the drug loading efficiency and release kinetics were measured.

Results:

Adding dopamine into the NorHA pre-gel solution delayed the gelation time, yet the final storage modulus of the hydrogel remained constant. That is, dopamine might partially consume the energy required for thiol-ene reaction to generate semiquinone radicals. Furthermore, the residual thiols which were not involved in the crosslinking decreased when the hydrogel was formed at a high concentration of dopamine, indicating the formation of Michael adducts of semiquinone and thiols. Interestingly, the presence of dopamine in the hydrogel increased the loading efficiency of the hydrophobic drugs due to π-π stacking and hydrogen bonding between dopamine and drugs.

Conclusion:

The presence of free catecholamines in a photo-crosslinkable polymer can delay the gelation time but improve the drug loading efficiency.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13770-021-00388-3.

Introduction

Hydrogel is the most popular material in the field of biomedical engineering. It has applications in disciplines such as tissue engineering because it exhibits an elastic modulus similar to that of human tissue [1]. Particularly, photo-crosslinkable hydrogels are used for biomedical applications in tissue engineering or as 3D bio-printing cartridges and drug delivery carriers because the functionality of the gel can be easily controlled [2]. There are various photo-crosslinkable hydrogels such as gelatin methacrylate (GelMA), methacrylated hyaluronic acid (HAMA), and norbornene modified hyaluronic acid (NorHA) [3–6]. Among them, NorHA is a distinct material that requires a small crosslinker, such as thiol-containing molecules, for additional crosslinking on demand. Therefore, we can manipulate the gelation kinetics by controlling the reactivity and stoichiometric ratio of the crosslinker involved in the polymers.

Hyaluronic acid (HA) is widely used due to its excellent biocompatibility. Additionally, it is abundant in the extracellular matrix (ECM) of cartilage tissue of a human body [7]. It is composed of alternating units of disaccharide that have functional groups such as carboxylic acid and primary hydroxyl groups in glucuronic acid, and secondary hydroxyl groups and N-acetyl group followed by deamidation of N-acetyl glucosamine [8]. These groups can be modified to other functional groups for applications such as self-healing property [9, 10] or biomolecular delivery [11]. Recently, polyphenol (e.g., catechol or gallol) conjugated HA that exhibited tissue adhesiveness has been introduced [11–13]. In particular, catechol exhibits adhesiveness and reactive oxygen species (ROS)/radical scavenging effect [14–16]. For instance, recent study showed that free catechols inhibit polymerization of acrylate monomers due to their radical scavenging effect [17]. Dopamine is a catecholamine which is characterized by a catechol group. It has two hydroxyl groups that can form hydrogen bonding and the ring structure can form π-π stacking [18]. While dopamine can interact with diverse surroundings, there is not sufficient research about effect of free dopamine in polymeric solution compared to dopamine conjugated polymers [17, 19–22]. Therefore, we aimed to study the effect of free dopamine on polymeric condition, especially thiol-ene photo-crosslinkable polymers, NorHA.

Herein, we demonstrated how free state of dopamine effects both on thiol-ene photo-crosslinking reaction time of norbornene-modified polymer (Fig. 1A) and loading efficiency of hydrophobic anti-cancer drug, doxorubicin (Fig. 1B). Dopamine contained polymeric solution showed delayed gelation time due to radical scavenging effect of dopamine, which makes polymerized dopamine, while no significant difference in storage modulus of final hydrogels. We also showed that dopamine embedded polymeric solution can increase loading efficiency of hydrophobic drugs forming π-π stacking and hydrogen bonding with their ring structure and hydroxyl moieties on catechol. Therefore, our study would be a promising strategy for encapsulating drugs in hydrogels while maintaining mechanical property for biomedical application particularly in cancer therapy field.

Fig. 1.

A schematic illustration of the effect of dopamine on thiol-ene crosslinking and doxorubicin encapsulation. A Illustration of Dopa@NorHA hydrogel crosslinking procedure. i) Materials in pre-solution of Dopa@NorHA, ii) delayed gelation time by radical scavenging effect of dopamine, and iii) crosslinked NorHA hydrogel. B Improved solubility of doxorubicin due to hydrogen bonding and π-π interaction between Dopa and DOX

Materials and methods

Synthesis of norbornene modified hyaluronic acid (NorHA)

Hyaluronic acid was functionalized with norbornene via 4-(dimethylamino)pyridine (DMAP, Sigma-Aldrich, St. Louis, MO, USA), di-tert-butyl dicarbonate (Boc₂O, Sigma-Aldrich) esterification reaction following as previous reported protocol [6]. Sodium hyaluronic acid (60 kDa, Lifecore Biomedical, Chaska, MN, USA) was modified to hyaluronic acid tetrabutylammonium (HA-TBA). The HA-TBA was dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich) and reacted with 5-norbornene-2-carboxylic acid (endo- and exo- mixture, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan), Boc₂O for 18 h. Supplementary figure S1 shows ¹H NMR spectrum of NorHA.

Dopa@NorHA hydrogel fabrication

NorHA (4% (w/v)) was dissolved in phosphate buffer saline (1X PBS, pH 7.4) with 0.05 wt% 2-hydroxy-4'-(2-hydroxyethoxy)-2-methylpropiophenone (I2959, Sigma-Aldrich), with a varying concentration of dithiothreitol (DTT, Roche, Switzerland) or dopamine hydrochloride (Dopa, Sigma-Aldrich). The pre-solution of Dopa@NorHA was irradiated with ultraviolet light (Omnicure S1500A, 320–390 nm, Waldbronn, Germany) for a desired time.

Rheology of Dopa@NorHA hydrogel

The rheology measurements were performed on a Discovery Hybrid Rheometer 2 (TA Instruments, New Castle, DE, USA) equipped with a UV-LED accessory technology. A 20 mm-diameter parallel plate geometry was used for the measurements. For photo-crosslinking Dopa@NorHA, a UV light source (320–390 nm) with an irradiance of 5 mW cm⁻² and 10 mW cm⁻² was used to cure the pre-hydrogel solution through a 2 cm-thick UV transparent quartz bottom plate. While conducting the process of curing, a constant oscillatory frequency of 1 Hz and a strain of 1% was used. The storage modulus was measured through a frequency sweep between 0.01 and 10.0 Hz with an oscillatory strain of 1% after curing as indicated by a plateau in the storage modulus during UV exposure. The storage modulus at 1 Hz was considered.

Thiol consumption test of Dopa@NorHA hydrogel

Pre-hydrogel solution (100 µL) was exposed under UV light (320–390 nm) for 3 min. After irradiation, non-reacted DTT was collected overnight by dipping the hydrogels in phosphate buffer (pH 8.0). The free thiol was assayed with Ellman's reagent (Sigma-Aldrich) with a multi-mode microplate reader (BioTek Synergy HTX, Winooski, VT, USA) absorbance peak at 412 nm [23]. The free thiol concentration was calculated using the calibration curve of DTT concentration of 20, 30, 40, 50, 60, 70, 80, and 100 µg/mL.

Doxorubicin loading on Dopa@NorHA hydrogel

Doxorubicin hydrochloride (DOX, 280 µg) was mixed with 100 µL of pre-hydrogel of Dopa@NorHA solution. Pre-hydrogels were crosslinked by irradiation of UV light with an intensity of 10 mW cm⁻² for 10 min. The insoluble DOX was removed by a centrifuge that was operated at 14,000 rpm for 2 min. Finally, the remaining solution was crosslinked and used for the release test.

In vitro release test

DOX containing Dopa@NorHA hydrogel was moved to a 24-transwell with 1X PBS buffer (pH 7.4, 1 mL) solution and it was incubated at 37 °C. The entire volume of the sample solution was replaced with a fresh buffer at pre-determined time intervals (0.5, 1.5, 4, 12, 18, and 24 h). Subsequently, the amount of DOX released in the sample solutions was analyzed by an ultraviolet–visible (UV–vis) spectrophotometer (Agilent 8453, Santa Clara, CA, USA) using a calibration curve at various concentrations (14.3, 16.7, 20, 25, 33.3, and 50 µg/mL) obtained by the UV–Vis spectrometer at a wavelength of 480 nm.

Statistical analysis

The data was presented as the mean ± standard error of the mean (SEM) values, and n is the number of samples. The significance differences between groups were analyzed by one-way analysis of variance (ANOVA) with GraphPad Prism version 7 (GraphPad Software Inc., La Jolla, CA, USA). A p value less than or equal to 0.05 was considered statistically significant.

Results

Rheology of Dopa@NorHA

The change of rheology of Dopa@NorHA while UV crosslinking was measured through a photorheometer (Fig. 2). The gelation time was delayed with an increase in the [Dopa/Nor] mol ratio. The tendency of delayed time followed the [Dopa/Nor] mol ratio independent of the intensity of the light (Fig. 2C, F). Additionally, the storage modulus after gelation was approximately 10⁴ Pa which remained constant for all [Dopa / Nor] mol ratios (Fig. 2B, E). Furthermore, we conducted the rheology of Dopa@NorHA hydrogel with various DTT ratio to Nor. The difference in the gelation time was insignificant between various DTT concentrations, while storage modulus was slightly increased at the ratio of 0.5 (3925.5 Pa) to 1 (5109.3 Pa) since more crosslinked NorHA strands have formed (Supplementary Fig. S2).

Fig. 2.

Rheology of Dopa@NorHA hydrogel. A In-situ photorheometry. B Storage modulus of Dopa@NorHA at various concentrations of dopamine at a UV intensity of 5 mW cm⁻². C Delayed gelation time at a stoichiometric ratio of Dopa to Nor under UV irradiation (5 mW cm⁻²). D In-situ photorheometry. E Storage modulus of Dopa@NorHA at various concentrations of dopamine at a UV intensity of 10 mW cm⁻². F Delayed time according to [Dopa/Nor] ratio at a UV intensity of 10 mW cm⁻² (n = 3, Error bars on gelation were omitted for clarity)

Thiol consumption of dopamine

Variation on Dopa ratio

The amount of thiol consumption changed with an increase in the [Dopa/Nor] ratio (Fig. 3A). In the middle of variables, the ratio of 0.5, which is indicated by a red arrow, showed approximately 15% decreased thiol consumption than that at other concentrations.

Fig. 3.

Thiol consumption test of non-reacted components of Dopa@NorHA hydrogels via Ellman’s assay. A Thiol consumption of Dopa@NorHA hydrogel at various [Dopa/Nor] mol ratio. Red arrow indicates the value of the ratio at 0.5. B Thiol consumption of Dopa@NorHA hydrogel at various [DTT/2*Nor] molar ratio while [Dopa/Nor] molar ratio was fixed with 1. Red arrow indicates the value of the ratio at 1.5 (n = 3)

Variation on DTT ratio

To investigate the effect of DTT on gelation at a fixed ratio of Dopa embedded in the NorHA solution, we varied the mol ratio of DTT to Nor as [DTT/2*Nor]—which is based on the fact that two Nor can react with one DTT containing two thiol groups (Fig. 3B). Particularly, the [Dopa/2*Nor] mol ratio of 1.5 exhibited a thiol consumption of approximately 85%, which has a gap between with theoretical consumption of crosslinker.

Improved DOX loading efficiency

The loading efficiency of DOX was easily observed since undissolved DOX was precipitated. Additionally, DOX exhibits a red color. Therefore, the separated supernatant (pre-hydrogel solution) color which was red indicated a higher amount of DOX. Hence, the loading efficiency increased with an increase in the [Dopa/Nor] ratio of Dopa@NorHA solution (Fig. 4A).

Fig. 4.

Encapsulation and in vitro release test of doxorubicin loaded Dopa@NorHA hydrogel. A Effect of dopamine concentration on doxorubicin loading efficiency. (Scale bar = 3 mm). B In vitro release profile of doxorubicin from the hydrogel containing various concentrations of dopamine. (n = 3)

Release profile of DOX-loaded Dopa@NorHA hydrogel

In accordance with the result obtained for DOX loading efficiency, a greater amount of DOX was released from the hydrogel when the [Dopa/Nor] mol ratio was 5 (Fig. 4B). The mean value of cumulative DOX release at [Dopa/Nor] mol ratio of 5 (199.6 µg) was approximately sixfold higher than that of the ratio of 0 (32.5 µg) at 24 h. The free dopamine was also existed in the releasates (Supplementary Fig. S3).

Discussion

If dopamine was involved in the crosslinking reaction, it would have formed a bond with the thiol group of dithiothreitol or alkene group of norbornene and change the modulus of hydrogels. However, a significant difference in the storage modulus was not observed for various ratios of dopamine to norbornene. This indicated that dopamine only affected the gelation time of thiol-ene reaction, delaying the time to reach the critical radical concentration to complete gelation. We hypothesized that dopamine can partially absorb the light energy and form semi-quinone radicals which are crosslinked as di-catechol or generate indole form (e.g., 5,6-Dihydroxyindole (DHI)) (Fig. 1A) [24]. In addition, the formation of DHI can trigger generation of polydopamine nanoparticles with small diameters within the final Dopa@NorHA hydrogels, further enhancing therapeutic efficacy in cancer treatment due to its cellular internalization by high affinity to metal ion present in lysosome of cancer cells [25].

In experiments of thiol consumption test varying [Dopa/Nor] mol ratio, we observed that the difference in storage modulus was insignificant over ~ 80% thiol consumption because statistical tests of modulus did not exhibit a significant value. At a ratio of 0.5, which means that the mol ratio of Dopa and DTT was same, showed a decrease in the thiol consumption. The low thiol consumption indicates that crosslinkers are not fully involved in gelation, decreasing crosslinking density. We conclude that if dopamine was absent in the Dopa@NorHA solution, where the ratio was 0, hindering effect of Dopa was not observed and 100% of thiols reacted with Nor. However, the presence of Dopa in the solution can minutely hinder the interaction between thiols and alkenes of Nor as a radical scavenger. At the critical point, (here at 0.5 indicated with a red arrow in Fig. 3A), radicals of Dopa and DTT can interact with another reversibly to ensure that DTT can revert to its original free thiol condition resulting in low thiol consumption. Furthermore, at a fixed ratio of Dopa/Nor, an increment of crosslinker rather maintains ~ 85% of thiol consumption (red arrow, Fig. 3B). From the result, we conjectured that high concentration of Dopa generates DTT-Dopa conjugates rather than acting as a radical scavenger. So, the thiol consumption at a [DTT/2*Nor] mol ratio of 1.5 or 1.25 is not significantly different.

While doxorubicin is the most popular anti-cancer drug which exhibits water soluble property with hydrochloride, its solubility decreased significantly after dissociation with hydrochloride due to an increase in the ring structure. The absence of dopamine in NorHA solution shows considerable precipitation of doxorubicin, whereas precipitation was not observed for a NorHA solution with high concentration of dopamine. As illustrated in the overall scheme (Fig. 1B), we hypothesized that the interaction between ring structure and hydroxyl groups of dopamine and doxorubicin forms π-π stacking and hydrogen bonding. Additionally, dopamine itself has a stacking effect that led to dopamine wrapping doxorubicin which resulted in an increase in the solubility of doxorubicin and enhancement of cellular uptake for anti-cancer effects [26–28].

In conclusion, mixing free polyphenols in thiol-ene photo-crosslinkable biomedical polymers can increase the loading efficiency of a hydrophobic drug and it does not significantly affect the modulus of the hydrogel.

Supplementary Information

Below is the link to the electronic supplementary material.

Declarations

Conflict of interest

The authors have declared that they have no conflict of interest.

Ethical Statement

There are no animal experiments carried out for this article.