Perfluorocarbon functionalized hyaluronic acid derivatives as oxygenating systems for cell culture

Abstract

A set of new hyaluronic acid (HA) derivatives was obtained by binding fluorinated oxadiazole (OXA) moieties to an amino derivative of the polysaccharide (HA-EDA). The obtained HA-EDA-OXA biomaterials are potentially able to improve oxygenation into a scaffold for tissue engineering purposes. The oxygen solubility in aqueous dispersions of the obtained derivatives showed that polymers were able to improve oxygen uptake and maintenance in the medium. The HA-EDA-OXA was employed to form a hydrogel in situ by reaction with a vinyl sulphone derivative of inulin, under physiological conditions. The influence of the presence of OXA moieties on the mechanical properties of the obtained hydrogels as well as on the metabolic activity of incorporated primary fibroblasts was investigated. The produced HA-EDA-OXA biomaterials were able to promote cell growth under hypoxic conditions.

---

Introduction

Many projects to build cell encapsulating systems fail because of the insufficient oxygen supply and inadequate scavenging of waste gas, such as carbon dioxide. Indeed cells, especially those proliferating in the inner parts of constructs beneath 1 mm in depth, in the absence of tissue perfusion or any adequate solution start to experience metabolic suffering with consequent apoptosis and cell death.¹ For this reason, oxygen level is a crucial parameter for the tissue development both in vitro and in vivo, thus limiting the applicability of synthetic extracellular matrix in regenerative medicine.² Despite research efforts in these areas,³ oxygen transport remains one of the main limitations in maintaining cell viability and functionality.

Many attempts have been reported to overcome issues related to the lack of oxygen in hydrogel or scaffold systems.³ The widest field of interest is probably related to the development of oxygen-generating biomaterials. Among these systems, the most extensively studied ones are those employing peroxides (such as CaO₂,⁴ MgO₂,⁵ etc.) and enzymes coupled with H₂O₂ as oxygen generating systems.⁶ Nevertheless, both systems involve the production, without an accurate kinetic control, of reactive oxygenated species, whose levels could harm cells and tissues.⁷

Alternatively, perfluorocarbons (PFCs) can be used as oxygen carriers. PFCs are inert organic molecules in which fluorine atoms replace hydrogen ones. This peculiar characteristic allow them to dissolve large amounts of physiologically important gases, such as oxygen and carbon dioxide. Functional studies have shown that O₂ solubility of PFCs is approximately 50 times higher than that of culture medium. The PFCs have been known as blood substitutes because of their high oxygen solubilising ability.⁸ For example, oxygen solubility in perfluorotributylamine (PFTBA) is 35 mM, more than a fifteenfold increase with respect to its solubility in water, which is only 2.2 mM.⁹ Additionally, since PFCs have only a limited capacity to supply oxygen and do not become reoxygenated, they can increase oxygenation by enhancing dissolved oxygen effective diffusivity through the matrix and not by serving as an oxygen reservoir. Therefore, PFC emulsions have a linear relationship between oxygen partial pressure and oxygen concentration,¹⁰ and could rapidly increase oxygenation of tissues under hypoxic condition. Furthermore, PFCs are advantageous because of their commercial availability, chemical and biological inertness, and easy sterilization. In particular, fluorinated oxadiazoles are an emerging class of PFCs due to their properties as oxygen carriers.^11,12 In fact, despite 1,2,4-oxadiazoles derivatives are more frequently studied for their application in synthesis^13–16 and in medicinal chemistry,¹⁷ this heterocycle show great potential for applications in advanced materials¹⁸ due to its tendency to self-assembly when associated with perfluoroalkyl chains.^{11,12,18–20}

Polysaccharide-based hydrogels have attracted much attention as matrices for application in tissue engineering due to their excellent biocompatibility and biodegradability.²¹ Moreover, because of their high water affinity they are perfect candidates as biomimetic materials of extracellular matrix (ECM).²² Hyaluronic acid for example is a glycosaminoglycan (GAG) abundant in the ECM of animal tissues, implied in structural functionalities such as osmotic tissue tone maintenance and biological regulatory activity on cell differentiation and control of growth factors production and diffusion.^23,24 Several types of HA graft derivatives have been proposed with the aim to either improve the control over the biological and physicochemical performance of biomaterials resulting from appropriate processing, or to exploit the favourable biological properties of HA^25–27 through the recognition and activation of a series of receptors that can control cellular activity.^28,29 HA in particular has known properties that stimulate regeneration and differentiation in several tissues, and it is employed for wound healing and cartilage regeneration.³⁰ HA hydrogels, either as non functionalized biomaterials or in conjunction with other kind of ECM's cues or soluble factors, have been also studied as injectable systems to enhance regeneration of ischemic heart diseases, and showed to improve myocytes growth and differentiation in vitro, and to improve cardiac functions by reducing infarct size in vivo.^31,32 Moreover, oxygen generating biomaterials applied on wound areas or ischemic tissues have demonstrated to improve regeneration of tissues damages restoring their functionality.^33,34 Our approach is based on the tethering of a fluorinated oxadiazole moiety to the hyaluronic acid hydrophilic backbone, thus obtaining novel HA derivatives possessing both the innate HA biomimetic properties and the ability of PFCs to increase diffusivity of gases and nutrients. This strategy will allow to obtain new biomaterials able to favour oxygen exchange.

Results and discussion

Fluorinated oxadiazole 1 was chosen for the perfluorocarbon-like functionalization of the hyaluronic acid backbone. The amino functionalized HA derivative HA-EDA was β chosen to perform the coupling reaction with the oxadiazole 1 through SN_Ar reaction at the fluorinated aromatic group, at the favoured ortho position (Scheme 1).^13,35

Scheme 1 Derivatization reaction of HA derivatives with 1.

The presence of primary amino groups as side chain in HA-EDA derivative is fundamental to obtain an efficient functionalization with oxadiazole 1. In fact, reactions performed with HA in similar reaction condition did not result in any detectable functionalization. In details, three reaction conditions were set to produce a series of derivatives having different physical and chemical characteristics (Table 1).

Table 1 Reaction conditions and size exclusion chromatography data of HA-EDA^a and HA-EDA-OXA^a polymers

Sample	Temp.	DDOXA^b (mol%)	Unreact. NH2^c (mol%)	Mw (kDa)	PDI
a %molar ratio between EDA groups and HA repeating units is equal to 50%.b %molar ratio between OXA moieties and amino groups.c %molar ratio between free amino groups and HA repeating units.
HA-EDA	—	—	50	380	1.87
HA-EDA-OXA a	40 °C	21.6	39.2	369	1.81
HA-EDA-OXA b	60 °C	36.0	32.0	354	2.15
HA-EDA-OXA c	60 °C (MW)	98.2	0.9	252	2.05

---

In order to determine the derivatization degree, we have performed an UV analysis, adopting a procedure elsewhere reported for other conjugates.^12,20 UV absorption spectrum of derivatives in DMSO–H₂O (1 : 1), in the 200–400 nm range, shows an absorption maximum at 328 nm, while HA-EDA does not show absorption in the same range. The obtained spectra are similar to that reported for the starting oxadiazole 1 (OXA) which shows one maximum at 328 nm (ε = 24 500 ± 10 M⁻¹ cm⁻¹). However, considering the structural differences between oxadiazole 1 and the tethered oxadiazole moiety in the polymer, we decided to use oxadiazole 2 as a better model of the oxadiazole chromophore in the HA-EDA-OXA derivative. Therefore, compound 2 was synthesized through SN_Ar of oxadiazole 1 with n-butylamine (Scheme 2),^13,14,35 and used as a reference compound to build the calibration curve for the spectrophotometric determination of the polymer's oxadiazolic derivatization degree (DD_OXA) (Table 1).

Scheme 2 Synthesis of reference compound 2.

Indeed, due to a hypochromic effect of the amino substitution on the fluorinated aromatic ring, oxadiazole 2 presented a maximum at the same wavelength (328 nm) but with a much lower molar extinction coefficient value (ε = 1440 ± 5 M⁻¹ cm⁻¹) with respect to the starting oxadiazole 1. The content of polymer-linked oxadiazole units was calculated by UV analysis, comparing E^1%₃₂₈ of each polymer sample with that of oxadiazole 2. As reported in Table 1, varying the reaction temperature from 40° to 60 °C resulted in about a 50% increase of the DD. When the reaction was performed in a microwave reactor with a power intensity equal to 25 W for 1 h the resulting HA-EDA-OXA c derivative showed an almost complete derivatization of amino groups.

SEC measurements showed negligible effects of the reaction conditions on the stability of the HA backbone for derivatives a and b. In fact, no significant reduction of M_w has been registered, also taking into account the probable reduction on hydrodynamic size caused by the hydrophobicity of the linked oxadiazoles. A significant degradation of HA-EDA-OXA c derivative was instead registered, with almost a 30% of molecular weight loss, probably due to the effect of microwaves on the polysaccharidic backbone.³⁶ In order to obtain data about HA-EDA-OXA graft derivatives behaviour as potential carriers for O₂, gas release kinetics were performed at different temperatures (25 and 37 °C) and concentrations (1 mg ml⁻¹ and 30 mg ml⁻¹). Due to the slightly lower DD% obtained for HA-EDA-OXA a, with respect to HA-EDA-OXA b, the former was not included in the oxygen solubility measurements. Oxygen solubility was determined after saturation of the medium and monitored, as a function of time, at atmospheric pressure. Accordingly to previously reported in vivo experiments,³⁷ the desaturation curves were approximated, by a single exponential function (eqn (1)).

[O₂] = [O₂]_∞ + [O₂]_load exp(−t/k) (1)

where: [O₂]_∞ = oxygen solubility at t_∞; [O₂]_load = [O₂]₀ − [O₂]_∞ with [O₂]₀ = oxygen solubility at t₀; k = clearance constant (min).

The obtained data are reported in Table 2 and in Fig. 1 and 2.

Table 2 Parameters of eqn (1) determined for desaturation curves as a function of time of oxygen saturated aqueous solutions of polymers

Polymer	C = 1 mg ml^−1		C = 30 mg ml^−1
	T = 25 °C	T = 37 °C	T = 25 °C	T = 37 °C
HA-EDA	[O2]0 = 33.49 ± 0.03 ppm	[O2]0 = 23.09 ± 0.02 ppm	[O2]0 = 20.93 ± 0.02 ppm	[O2]0 = 20.57 ± 0.05 ppm
	[O2]∞ = 7.70 ± 0.01 ppm	[O2]∞ = 6.40 ± 0.01 ppm	[O2]∞ = 7.12 ± 0.01 ppm	[O2]∞ = 5.48 ± 0.02 ppm
	k = 71.2 ± 0.1 min	k = 62.7 ± 0.1 min	k = 277.3 ± 0.7 min	k = 232 ± 1 min
HA-EDA-OXA b	[O2]0 = 35.77 ± 0.02 ppm	[O2]0 = 24.04 ± 0.02 ppm	[O2]0 = 33.21 ± 0.02 ppm	[O2]0 = 31.59 ± 0.04 ppm
	[O2]∞ = 7.13 ± 0.01 ppm	[O2]∞ = 6.18 ± 0.01 ppm	[O2]∞ = 7.07 ± 0.01 ppm	[O2]∞ = 6.98 ± 0.02 ppm
	k = 275.6 ± 0.1 min	k = 70.9 ± 0.1 min	k = 327.8 ± 0.3 min	k = 351 ± 1 min
HA-EDA-OXA c	[O2]0 = 36.24 ± 0.03 ppm	[O2]0 = 26.83 ± 0.01 ppm	ND	ND
	[O2]∞ = 7.21 ± 0.02 ppm	[O2]∞ = 6.18 ± 0.01 ppm
	k = 54.6 ± 0.1 min	k = 34.6 ± 0.1 min

---

Fig. 1 Oxygen release curves from aqueous solutions containing polymers at 1 mg ml⁻¹ at 25 °C (left) and 37 °C (right).

Fig. 2 Oxygen release curves from aqueous solutions containing polymers at 30 mg ml⁻¹ at 25 °C (left) and 37 °C (right).

Data from Table 2 and Fig. 1 show that, at polymer concentrations of 1 mg ml⁻¹, the fluoro functionalization only slightly increases the dissolved oxygen content [O₂]₀ at saturation. In terms of oxygen release, a significant increase of clearance constant value was observed only for HA-EDA-OXA b at 25 °C compared to HA-EDA. Notably, for HA-EDA-OXA c a faster O₂ release was observed with respect to the unfluorinated derivative. At 37 °C, differences between HA-EDA and HA-EDA-OXA are less marked and both derivatives behave in a similar way, although a faster O₂ release was still observed for HA-EDA-OXA c (Fig. 1). Beside lowering the intrinsic solubility of gases, the increase of temperature interferes with the interaction with fluorinated moieties at polymer concentration of 1 mg ml⁻¹. Unfortunately, measurements of polymer dispersions with higher concentration (30 mg ml⁻¹) of HA-EDA-OXA c were unreliable due to the high tendency to aggregation and precipitation. Nevertheless, measurements performed on the other samples reveal that the presence of fluorinated moieties on HA-EDA-OXA b, strongly affects the ability to dissolve higher oxygen content than HA-EDA (Table 1 and Fig. 2). This is particularly evident at 37 °C were the fluorinated medium also slow down O₂ release, showing a considerably higher k value (Table 1). These data are particularly interesting considering that, at higher concentration, the fluorinated phase is able to overcome the effect of temperature on lowering the intrinsic solubility of gases.

In summary, data show how generally the introduction of fluorinated moieties could increase the oxygen solubility, but the functionalization with the fluorinated oxadiazole is more effective on increasing k value with respect to the unfluorinated polymer; the observed behaviour is not correlated to the derivatization degree of obtained derivatives and is strongly dependent from temperature and concentration. Differences of k values, suggest that at higher concentration the likely formation of polymeric aggregates driven by the hydrophobic entanglement of fluorinated chains results in the constitution of a reservoir for oxygen, thus explaining the high k values observed for HA-EDA-OXA b.

Moreover, fluorinated HA-EDA-OXA derivatives showed higher k values than those found for other perfluorochemicals^11,37,38 thus suggesting their potential use as water soluble polymeric carriers when long oxygenation times are required.

To assay the appropriateness of the new derivatives as constituent of biomaterials for potential biomedical applications, an in situ forming hydrogel scaffold suitable for cell encapsulation strategies has been designed. Pendant amino groups, still available on the derivatives, can readily react with vinyl groups (vinyl sulfones, acrylates, methacrylates, etc.) in physiological media. Although HA-EDA-OXA c showed more pronounced capabilities to retain oxygen, it was hardly dispersible at concentrations higher than 1% w/v; moreover, the almost complete functionalization of amino groups in this derivative would obstacle hydrogel formation by Michael-type reaction and the decrease in molecular weight observed would negatively affect mechanical properties of obtained hydrogels. Therefore, only HA-EDA and HA-EDA-OXA b derivatives were chosen for the in situ formation of hydrogel by reaction with the vinyl sulfone derivative of inulin (INUDV). Inulin was chosen considering its known biocompatibility and biological inertness.³⁹ Chemically crosslinked hydrogels were obtained by mixing polymer solutions of INUDV and amino derivatives of hyaluronic acid (HA-EDA or HA-EDA-OXA b) in DPBS at pH 7.4, both at 3% w/v concentration; this value was chosen since preliminary studies showed that for lower concentrations formed hydrogels were unstable. In particular, polymeric solutions of HA-EDA or HA-EDA-OXA b and INUDV were mixed according to weight percentages equal to 80 : 20 and 70 : 30, respectively and characterized in terms of time required for gel formation and mechanical properties. For all HA-derivative/INUDV characterized samples, molar ratios between reacting amino groups and vinyl sulfone have been calculated and reported in Table 3 together with the gelation time evaluated with the tube inversion test and the values of elastic modulus E.

Table 3 Physicochemical characterization of obtained hydrogels: amino groups/vinyl sulphones ratio, gelation time, and elastic modulus (E) were reported as function of starting solution composition

Sample	NH2/vinyl sulphone groups ratio	Gel formation time (min)	E (kPa)
HA-EDA/INUDV 80 : 20	2.6	21.3 ± 0.4	10.3 ± 0.4
HA-EDA/INUDV 70 : 30	1.5	24.5 ± 0.3	9.2 ± 0.6
HA-EDA-OXA/INUDV 80 : 20	1.7	27.2 ± 1.4	7.9 ± 1.1
HA-EDA-OXA/INUDV 70 : 30	1.0	30.1 ± 1.2	2.4 ± 0.8

---

Both gelation time and elastic modulus were directly related to the amino groups/vinyl sulfone molar ratios due to a more efficient crosslinking in the presence of an molar excess of NH₂ moieties.

By comparing hydrogels HA-EDA/INUDV 70/30 and HA-EDA-OXA b/INUDV 80/20 one can notice that to a slight increase of the NH₂/vinyl ratio from 1.5 to 1.7 corresponds a slight increase of gelation time and decrease of the elastic modulus. Additionally, the significantly lower value of E observed for HA-EDA-OXA b/INUDV 70/30, could not be justified on the basis of a lower NH₂/vinyl sulphone ratio. These data clearly suggest a weakening effect of the oxadiazole moieties on the final hydrogel mechanical resistance.

Fibroblasts were encapsulated within selected gel forming systems with the aim of confirming the ability of introduced OXA moieties to favour oxygen exchanges. First viability of fibroblasts encapsulated inside the hydrogel performed on the weight ratio 70/30 after 12 h of incubation revealed an acceptable cytocompatibility of the crosslinking reaction (75 ± 12 and 81 ± 19% for HA-EDA/INUDV and HA-EDA-OXA/INUDV respectively). In the normoxic culture condition the metabolic activity of cells within each construct was assessed after 7 days using an MTS assay, and reported in Fig. 3.

Fig. 3 Viability of rat fibroblasts encapsulated in either HA-EDA/INUDV or HA-EDA-OXA/INUDV gel forming solutions after 7 days of incubation time. The experiment was performed incubating in normoxic condition (−O₂) or in normoxic condition with the supply of oxygen (+O₂).

As clearly shown for both kind of experimental condition set (i.e. normoxic and oxygen supplementation) viability of fibroblasts on HA-EDA-OXA b/INUDV hydrogels was significantly higher than viability on HA-EDA/INUDV. Moreover, fibroblasts viability in HA-EDA/INUDV hydrogel was very poor in normoxic culture condition indicating a probable hypoxia, however partially resolved with the supplementary oxygenation. In HA-EDA-OXA b based constructs, cell viability was significantly higher than that observed in HA-EDA/INUDV media and was not affected by the supplementary oxygenation. This data, in addition to the oxygen exchange studies performed on the not crosslinked derivatives, suggest that HA-EDA-OXA b derivative renders the crosslinked scaffold very permeable to oxygen compared to the HA-EDA construct, allowing it to reach, already in normoxic culture conditions, the sufficient oxygenation levels to assure the maximum metabolic activity of fibroblasts.

The experiment performed under hypoxic conditions showed how cells incubated inside the OXA functionalized HA derivative were significantly more viable if compared to HA-EDA/INUDV construct. As shown in Fig. 4 viability of fibroblasts on HA-EDA-OXA is significantly higher than viability on HA-EDA already after 24 h (p < 0.05) and decreases significantly less after 72 (p = 0.002) and 168 h (p = 7.7 × 10⁻⁶).

Fig. 4 Viability of rat fibroblasts encapsulated in HA-EDA/INUDV or HA-EDA-OXA/INUDV gel forming solutions (weight ratio 70/30) and incubated under hypoxic condition.

Despite the highly hypoxic condition created inside the sealable pouch, oxygen loading, due to the culture medium repletion, seem more efficiently retained by the OXA functionalized construct. The biological in vitro tests considered together are in accordance to the oxygen solubility performances evidenced in the kinetic study, and demonstrated how HA-EDA-OXA based scaffolds can better exploit environmental oxygen if compared to the not fluorinated derivatives. Finally, SEM pictures of OXA containing hydrogels highlight adhering (blue arrows) encapsulated cells, which show affinity for tested materials (Fig. 5).

Fig. 5 SEM pictures of cell-encapsulating HA-EDA-OXA 80/20 (up) and 70/30 (down) after 7 days of culture. Arrows indicated fibroblasts attached on scaffold.

Experimental

All reagents were of analytical grade and unless otherwise stated were used as received. 2-(2,3,5,6-Tetrafluoro-4-(3-perfluoroheptyl-1,2,4-oxadiazol-5-yl)-phenylamino)acetic acid 1 (OXA) was prepared as previously reported.¹¹ Other experimental details are given on ESI.†

Synthetic procedures

Synthesis of vinyl sulfone derivative of inulin (INUDV). Vinyl sulfone moieties have been tethered to inulin backbone according to a previously published procedure.⁴⁰ Briefly, inulin has been solved in anhydrous DMF at 5% w/v. Then, divinyl sulfone (DV) and triethylamine (TEA) have been added both with a molar ratio equal to 5 respect to inulin repeating units and the reaction has been carried out at 60 °C for 24 h under argon. The pure product has been recovered after precipitation and several washings in diethyl ether and finally dried under vacuum. ¹H-NMR spectrum in D₂O showed peaks at δ 3.50–4.0 (5H, m: –CH₂–OH; –CH–CH₂–OH; –CH₂–CH₂–O–), 4.14 (1H, t: CH–OH), 4.25 (1H, d: CH–OH), 6.43 and 6.90 (3H, 2q: CH₂=CH–). The degree of derivatization (DD%) in DV was determined by comparing the peak integrals at δ 6.43 and 6.90 (3H, 2q: CH₂=CH–) relative to DV double-bond protons, with the peaks between δ 3.5–4.25 relative to inulin fructose unit protons (7H). The value of DD% in DV was 30 ± 1 mol%.

Synthesis ethylendiamino derivatives of hyaluronic acid (HA-EDA, HA-TBA-EDA). Low molecular weight HA (M_w 250 kDa, polydispersity 2.1) and ethylendiamino derivatives HA-EDA and HA-TBA-EDA have been obtained according to already published synthetic protocols.⁴¹ Briefly, 1% w/v HA-TBA in anhydrous DMSO has been reacted for 4 h at 40 °C with 4-NPBC, at 0.5 molar ratio respect to HA repeating units, in order to activate primary hydroxyl groups of N-acetyl glucosamine residues. Then, ethylenediamine has been added to the activated polymer solution and left to react for 3 h at 40 °C. The pure product has been recovered after precipitation, washings in acetone and freeze-drying. To obtain HA-EDA, TBA has been removed by adding aqueous NaCl saturated solution after reaction completion. The product has been recovered after washings with a mixture of acetone–water 8 : 2, with acetone alone and final freeze-drying. ¹H-NMR spectrum in D₂O showed principal peaks at δ 2.0 (–NH–CO–CH₃), δ 3.1 (–CO–NH–CH₂–CH₂–NH₂–), δ 3.3–3.8 (pyranosyl CH of HA). The derivatization degree in EDA portions linked to HA was calculated by ¹H-NMR comparing the peak at δ 3.1 attributable to the methylene groups of EDA portion with the peak at δ 2.0 attributable to acetamido group of HA and by TNBS colorimetric assay and resulted to be 50 ± 2 mol%.

Synthesis of fluorinated derivatives of hyaluronic acid (HA-EDA-OXA). Derivatization of HA-TBA-EDA with OXA to obtain HA-EDA-OXA derivatives was carried out in anhydrous DMSO with two different reaction temperature. A solution of OXA in anhydrous DMSO and TEA has been added to previously dispersed HA-TBA-EDA in DMSO, the molar ratios between OXA and EDA and between TEA and OXA was set equal to 2.3 and 1 respectively. The reaction was carried out for 24 h at 40 or 60 °C under continuous stirring. After 24 h, the reaction was stopped adding aqueous NaCl saturated solution, then precipitated with an excess of acetone (to remove TBA). Samples were washed with acetone–water 8 : 2 and then acetone alone. Obtained products were dried under vacuum, dissolved in water and finally freeze-dried.

Enhanced microwave synthesis was used as a tool to increase OXA derivatization degree. In this case, the reaction was carried out in a microwave reactor at 40 W for 1 h, keeping the reaction mixture temperature at 60 °C via external cooling.

The amount of oxadiazole linked was determined by means of UV spectroscopy by comparing the absorbance at 328 nm of each polymer in DMSO–H₂O (1 : 1) solution with a calibration curve of the reference compound 2. The molar amount of unreacted amino groups was finally confirmed by ¹H-NMR analysis and TNBS colorimetric assay.

Biological procedures

Rat dermal fibroblasts. Rat fibroblasts were isolated as described:⁴² dermis was accurately shaved, cut into 1 cm² pieces and immersed in sterile DPBS with 1% v/v of penicillin–streptomycin solution. Specimens were treated with an aqueous cold ethanol solution (70% v/v) for 2 minutes, washed several times with sterile DPBS and reduced in small pieces with a scalpel. So obtained pieces were incubated with a Dispase II solution (2.5 U ml⁻¹) for 1 hour then the epidermis was separated from the dermis with the use of forceps.

Dermis pieces were kept in the bottom of T-75 culture flask for 1 h prior to add DMEM supplemented with 10% (v/v) of FBS, 1% (v/v) of penicillin–streptomycin solution, 1% (v/v) of glutamine solution and 0.1% v/v amphotericin B solution. Specimens were cultured for 2 weeks by changing the culture medium every 2–3 days until fibroblast migrate from the dermis to the culture flask. The so obtained cells were cultured from passage 1 to 7 in standard fibroblast medium with the above mentioned composition. Cells were used within passage 7.

Cell encapsulation and cytotoxicity. Rat fibroblasts (passage 6) were suspended in 200 μl of 3% w/v gel forming mixture of either HA-EDA/INUDV or HA-EDA-OXA/INUDV at a concentration of 1 × 10⁶ cells ml⁻¹. After 1 h cell medium was added to crosslinked hydrogels and all constructs were cultured in 1 ml of complete DMEM at 37 °C. In particular, two different culture environments were chosen, normoxic condition (5% CO₂ in humidified incubators) or reduced oxygen culture using the BD GasPak EZ Gas Generating System sachet. The latter system allow to generate a hypoxic condition inside a resealable pouch (atmosphere with approximately 1% oxygen after 24 h of incubation). In the experiment performed incubating in normoxic condition 2 groups of constructs were set: the first one received only cell medium exchange every day without further supply; the cell/hydrogel constructs belonging to the second group, in addition to daily fresh medium exchange, were bubbled everyday with pure O₂ for 4 minutes in order to saturate the systems. After 7 days the viability of encapsulated cells was measured by MTS assay on HA-EDA-OXA/INUDV and HA-EDA/INUDV weight ratios 80/20 and 70/30 respectively.

In the experiment performed in the anaerobic pouch the medium was changed every 2 days, while the gas generating system was replaced after 3 days. In this latter experiment viability of cells inside the hydrogel constructs was assayed via MTS after 24, 72 and 168 hours on HA-EDA-OXA/INU-DV and HA-EDA/INUDV weight ratio 70/30.

Statistical analysis

All results are reported as mean ± standard deviation and, when applicable, statistical analysis for significance was performed by means of Student's t-test, using Microsoft Excel statistical function for t-tests, assuming unequal variance and two tailed distribution; values of p < 0.05 were considered statistically significant.

Conclusions

The amino derivative of HA (HA-EDA) has been reacted with a fluorinated oxadiazole to obtain a biomaterial able to increase the solubility of biologically relevant gases. Our synthetic approach allowed us to obtain a set of derivatives with different OXA derivatization degrees based on reaction conditions. In particular, HA-EDA-OXA b resulted a good candidate to be employed for the development of a hydrogel for tissue engineering purposes. Besides having a good degree of derivatization in OXA moieties and showing good capacity to enhance oxygen solubility, HA-EDA-OXA b maintains the 2/3 of the amino residues available for the Michael-type addition with vinyl sulphone groups tethered to inulin. Since this reaction occurred at physiological conditions it was possible to develop a new in situ forming hydrogel system. In vitro biological tests evidenced that fluorinated OXA moieties within the hydrogel structure promote cell growth and metabolic activity, avoiding the cells to experience hypoxia, in comparison to analogous unfluorinated systems.

Acknowledgements

Financial support through University of Palermo is gratefully acknowledged.

Notes and references

1. J. Folkman and M. Hochberg, J. Exp. Med., 1973, 138, 745.
2. A. Khademhosseini, R. Langer, J. Borenstein and J. P. Vacanti, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 2480.
3. G. Camci-Unal, N. Alemdar, N. Annabia and A. Khademhosseini, Polym. Int., 2013, 62, 843.
4. S. H. Oh, C. L. Ward, A. Atala, J. J. Yoo and B. S. Harrison, Biomaterials, 2009, 30, 757.
5. B. S. Harrison, D. Eberli, S. J. Lee, A. Atala and J. J. Yoo, Biomaterials, 2007, 28, 4628.
6. S. I. Abdi, S. M. Ng and J. O. Lim, Int. J. Pharm., 2011, 409, 203.
7. Z. Li, X. Guo and J. Guan, Biomaterials, 2012, 33, 5914.
8. M. P. Krafft, Adv. Drug Delivery Rev., 2001, 47, 209.
9. K. C. Lowe, M. R. Davey and J. B. Power, Trends Biotechnol., 1998, 16, 272.
10. J. G. Riess, Artif. Cells, Blood Substitutes, Biotechnol., 2006, 34, 567.
11. A. Palumbo Piccionello, A. Guarcello, A. Calabrese, I. Pibiri, A. Pace and S. Buscemi, Org. Biomol. Chem., 2012, 10, 3044.
12. D. Mandracchia, A. Palumbo Piccionello, G. Pitarresi, A. Pace, S. Buscemi and G. Giammona, Macromol. Biosci., 2007, 7, 836; G. Pitarresi, A. Palumbo Piccionello, R. Calabrese, A. Pace, S. Buscemi and G. Giammona, J. Fluorine Chem., 2008, 129, 1096.
13. S. Buscemi, A. Pace, A. Palumbo Piccionello, I. Pibiri and N. Vivona, Heterocycles, 2004, 63, 1619.
14. A. Palumbo Piccionello, A. Pace, I. Pibiri, S. Buscemi and N. Vivona, Tetrahedron, 2006, 62, 8792.
15. A. Palumbo Piccionello, A. Pace, S. Buscemi, N. Vivona and G. Giorgi, Tetrahedron Lett., 2009, 50, 1472.
16. A. Terenzi, G. Barone, A. Palumbo Piccionello, G. Giorgi, A. Guarcello and A. Pace, Inorg. Chim. Acta, 2011, 373, 62.
17. See as selected examples: A. Terenzi, G. Barone, A. Palumbo Piccionello, G. Giorgi, A. Guarcello, P. Portanova, G. Calvaruso, S. Buscemi, N. Vivona and A. Pace, Dalton Trans., 2010, 39, 9140; A. Palumbo Piccionello, R. Musumeci, C. Cocuzza, C. G. Fortuna, A. Guarcello, P. Pierro and A. Pace, Eur. J. Med. Chem., 2012, 50, 441; C. G. Fortuna, C. Bonaccorso, A. Bulbarelli, G. Caltabiano, L. Rizzi, L. Goracci, G. Musumarra, A. Pace, A. Palumbo Piccionello, A. Guarcello, P. Pierro, C. E. A. Cocuzza and R. Musumeci, Eur. J. Med. Chem., 2013, 65, 533; L. Lentini, R. Melfi, A. Di Leonardo, A. Spinello, G. Barone, A. Pace, A. Palumbo Piccionello and I. Pibiri, Mol. Pharmaceutics, 2014, 11, 653.
18. I. Pibiri, A. Pace, S. Buscemi, V. Causin, F. Rastrelli and G. Saielli, Phys. Chem. Chem. Phys., 2012, 14, 14306.
19. F. Lo Celso, I. Pibiri, A. Triolo, R. Triolo, A. Pace, S. Buscemi and N. Vivona, J. Mater. Chem., 2007, 17, 1201.
20. A. Palumbo Piccionello, G. Pitarresi, A. Pace, D. Triolo, P. Picone, S. Buscemi and G. Giammona, J. Drug Targeting, 2012, 20, 433.
21. J. Kuang, K. Y. Yuk and K. M. Huh, Carbohydr. Polym., 2011, 83, 284.
22. J. Kopecek, Biomaterials, 2007, 28, 5185.
23. G. Kogan, L. Soltés, R. Stern and P. Gemeiner, Biotechnol. Lett., 2007, 29, 17.
24. R. Stern, A. A. Asari and K. N. Sugahara, Eur. J. Cell Biol., 2006, 85, 699.
25. C. Eenschooten, F. Guillaumie, G. Kontogeorgis, E. Stenby and K. Schwach-Abdellaoui, Carbohydr. Polym., 2010, 79, 597.
26. E. J. Oha, K. Park, J. S. Choi, C. K. Joo and S. K. Hahn, Biomaterials, 2009, 30, 6026.
27. F. S. Palumbo, G. Pitarresi, C. Fiorica, S. Rigogliuso, G. Ghersi and G. Giammona, Mater. Sci. Eng., C, 2013, 33, 2541.
28. J. Entwistle, C. L. Hall and E. A. Turley, J. Cell. Biochem., 1996, 61, 569.
29. B. P. Toole, Hyaluronan-cell interactions in morphogenesis, Portland Press, London, 1998.
30. G. Kogan, L. Soltes, R. Stern and P. Gemeiner, Biotechnol. Lett., 2007, 29, 17.
31. A. Chopra, V. Lin, A. McCollough, S. Atzet, G. D. Prestwich, A. S. Wechsler, M. E. Murray, S. A. Oake, J. Y. Kresh and P. A. Janmey, J. Biomech., 2012, 45, 824.
32. S. J. Yoon, Y. H. Fang, C. H. Lim, B. S. Kim, H. S. Son, Y. Park and K. Sun, J. Biomed. Mater. Res., Part B, 2009, 91, 163.
33. C. Ward, B. T. Corona, J. J. Yoo, B. S. Harrison and G. J. Christ, PLoS One, 2013, 8, e72485.
34. A. Wijekoon, N. Fountas-Davis and N. D. Leipzig, Acta Biomater., 2013, 9, 5653.
35. S. Buscemi, A. Pace, A. Palumbo Piccionello and N. Vivona, J. Fluorine Chem., 2006, 127, 1601.
36. M. M. Mecwan, G. E. Rapalo, S. R. Mishra, W. O. Haggard and J. D. Bumgardner, J. Biomed. Mater. Res., Part A, 2011, 97, 66.
37. B. A. Berkowitz, C. A. Wilson and D. L. Hatchell, Invest. Ophthalmol. Visual Sci., 1991, 32, 2382.
38. M. P. Krafft, J. Polym. Sci., Part A: Polym. Chem., 2006, 44, 4251.
39. B. D. Mather, K. Viswanathan, K. M. Miller and T. E. Long, Prog. Polym. Sci., 2006, 31, 487.
40. G. Pitarresi, G. Tripodo, R. Calabrese, E. F. Craparo, M. Licciardi and G. Giammona, Macromol. Biosci., 2008, 8, 891; G. Pitarresi, G. Tripodo, D. Triolo, C. Fiorica and G. Giammona, J. Drug Delivery Sci. Technol., 2009, 19, 413.
41. F. S. Palumbo, G. Pitarresi, A. Albanese, F. Calascibetta and G. Giammona, Acta Biomater., 2010, 6, 195; F. S. Palumbo, G. Pitarresi, C. Fiorica, P. Matricardi, A. Albanese and G. Giammona, Soft Matter, 2012, 8, 4918.
42. C. Fiorica, G. Pitarresi, F. S. Palumbo, M. Di Stefano, F. Calascibetta and G. Giammona, Int. J. Pharm., 2013, 30, 457.

---

Footnote

† Electronic supplementary information (ESI) available: Details of materials, methods and other experimental procedures. See DOI: 10.1039/c4ra01502a

---