Irreversible strengthening of gelatin hydrogels by salt-directed polypeptide assembly and covalent crosslinking

(1)

1

Irreversible strengthening of gelatin hydrogels by salt-directed polypeptide assembly and covalent

crosslinking

Author: Patrick Shakari Supervisor: Ayan Samanta

Mars: 2019-13-03

(2)

2

Abstract

Corneal disease is one of the fastest growing issues in the world, and with an increasing population it has become one of the most demanding challenges in the developing world.

Therefore, new types of biomaterials and tissue engineering scaffolds are needed to be developed to combat these challenges. Gelatin-based hydrogels are one of the most commonly used materials in tissue engineering and regenerative medicine. Gelatin is cheap, biodegradable, biocompatible, and causes a low immune response. However, the disadvantage of gelatin-based hydrogels is that they suffer from poor mechanical strength. This thesis demonstrates the feasibility of preparing stronger and more ductile gelatin hydrogels using a two-step approach involving soaking of the gelatin physical gel (gelatin thermogel) in an ammonium sulfate solution to induce more polypeptide aggregates/assemblies in the hydrogel followed by covalent crosslinking of the polymer chains to permanently stabilize these aggregates in the hydrogel resulting in higher strength and modulus. This phenomenon is known as Hofmeister effect where protein solubility in water can be decreased in the presence of ammonium and sulfate ions. It has been observed that formation of such salt-directed polypeptide assemblies and the resulting increase in modulus of the hydrogel is not permanent when the hydrogels are stored in water. However, covalent crosslinking of the polypeptide chains post salt-directed assembly has made such aggregates inside the hydrogel permanent when the hydrogels are stored in water leading to the retention of the improved mechanical properties of the hydrogels during storage in water.

(3)

3

List of abbreviations

ECM = extracellular matrix HA = hyaluronic acid

RGD = Arg-Gly-Asp (amino acid repeat unit from fibronectin) LAP = lithium phenyl-2,4,6-trimethylbenzoylphosphinate

Gel = gelatin (unmodified) physical gel (thermogel, not crosslinked)) TNBS = 2,4,6-trinitrobenzenesulfonic acid

SH = DTT (Dithiotreitol) GNB = gelatin norbornene

GNB-SH = gelatin norbornene-dithiothreitol physical gel (thermogel, not crosslinked) GNB-SH-H = gelatin-norbornene-dithiothreitol (thermogel, not crosslinked) soaked in ammonium sulfate solution

GNB-SH-UV = gelatin-norbornene-dithiothreitol-UV irradiated (covalently crosslinked) GNB-SH-H-UV = gelatin-norbornene-dithiothreitol gel soaked in ammonium sulfate solution-UV irradiated (covalently crosslinked)

GelMa = methacrylamide modified gelatin G’ = Storage modulus (G’)

(5)

5

Aim

In this thesis, I aim to develop strong and ductile gelatin hydrogels by using the Hofmeister effect assisted polypeptide assemblies in a preformed gelatin physical gel and attempt to permanently stabilize such assemblies inside the hydrogel by covalent crosslinking leading to an irreversible strengthening of the hydrogels.. An important aspect was to prevent swelling of the resulting hydrogels in aqueous solution, since it can be problematic for implants that can swell depending on the application. The long-term goal is to be able to use these gels as either synthetic cornea or as a skin graft.

1.Introduction

1.0 Corneal blindness

Each year the world population continues to grow, mostly in the third world countries and with it the challenges in treating corneal blindness. According to a WHO report written by Rupert R A Bourne and colleagues, around 36 million people were blind in 2015, and this number will rise up to 115 million by the year of 2050 [1]. A vast majority of the blind people live in developing countries in East Asia, and most of them are women [1]. Poor living conditions and lack of socio-economic support pose a great challenge in treating corneal blindness. Simple vision impairment, if left untreated, can lead to corneal blindness. The deterioration can be due to different factors such as ocular diseases or chemical injuries, ocular surface disorders, keratoconus, viral, bacterial, or fungal infections, etc. Chemical injury, as one example, can vary in the degree of damage from mild irritation to the destruction of the cornea [2]. The effect that the chemicals have on the cornea is used to classify the chemical injury. There are two possible outcomes if an acid or base comes in contact with the cornea. If an acidic solution comes in contact with the cornea, the proteins in the cornea start to denature and precipitate. As an indication of this, the eye becomes red and starts to swell [2]. Similarly, alkalis can cause severe impairments by damaging the cell membrane of corneal epithelium and therefore, corneal stroma gets exposed to external infections and damage. [2]. The eye anatomy and different layers of cornea is shown in Figure 1. Hence, it is essential to create new ways of treating different types of ocular diseases and blindness. The field of tissue engineering and regenerative medicine is at the forefront of such new ideas. By using biomaterials, such as hydrogels, with incorporated stem cells, it is potentially possible to repair damaged tissues.

Hydrogels can be used as various types of scaffolds and can promote regeneration by triggering endogenous cells [3].

(6)

6 Figure 1. The five layers that create the cornea, from the inner parts to the outer parts. Image reproduced from [25], with permission from [allboutvision.com 2019-03-6].

1.1 Biomaterials and tissue engineering

Biomaterials can be broadly defined as materials that do not provoke an immune rejection or response when in contact with living tissue [4][5]. Different types of materials can be used as a biomaterial, from softer polymers to hardened titanium plates [4]. The field of biomaterials can be divided into three different generations; the first generation of biomaterials are known as the bioinert materials and have low to no immune response when placed in the body. The second generation of biomaterials was considered to be bioactive, which implies it causes a biological response from the surrounding tissue [4][5][6] and causes cells from the host to migrate into the implant [6]. The third-generation is bioactive, bioresorbable and in some cases bioinert.

These materials were often porous three-dimensional scaffolds that promote cell adhesion and migration. Bioactive bioresorbable materials trigger host cells to migrate into the implanted materials, remodel and regenerate the injured tissue.

Due to more advanced materials and an increasing understanding of the body's biological and chemical function, tissue engineering started to evolve from biomaterials. Some of the principles for biomaterials and tissue engineering are still the same, but the applications and the methods have in some ways changed. The key factors for tissue engineering lay in the regenerative function of the scaffolds. The requirement on these materials is that they should possess abilities to maintain, restore and, in some cases, improve the damaged tissue. Another essential and beneficial aspect is if the host's enzymes can degrade the implanted material.

Therefore, a lot of research has been focused on materials that resemble the extracellular matrix (ECM) [6][7]. The most abundant components in the ECM are collagen, hyaluronic acid (HA)

(7)

7 and gelatin (Gel), but there are many more [7]. Because of their abundance in natural ECM, the immune response from ECM-derived biomaterials is usually low, which makes them biocompatible. These materials resemble the natural niche for cells, and the cells can therefore attach, migrate and proliferate [5][6][7].

Collagen is one of the most abundant components found in natural ECM. From its triple helical structure, it is able to create extended fibrillar networks and can provide mechanical supports to all tissues in the body. It gives the skin a layer of protection against external heat, microbial organisms, and external chemical factors [7][8]. Using collagen for medical application is highly beneficial. Researchers were able to create a synthetic cornea made from collagen type III combined with copolymer, such as methacryloyloxyethyl phosphorylcholine, and this type of material is suitable for patients displaying high-risk of rejection of a donor transplant [9].

But collagen is usually expensive and therefore, has a limitation for large scale application in countries with weaker economy. However, gelatin, which is denatured and hydrolyzed collagen, is much cheaper and has proven to be equally useful in tissue engineering and regenerative medicine [10][11].

1.2 Hydrogels

In the field of tissue engineering hydrogels are commonly employed as a scaffold. The hydrogels are three-dimensional network of polymers with a high-water absorbing capacity.

Some hydrogels have the ability to absorb water up to almost 99% of their dry weight [10]. The polymer chains in a hydrogel can entangle and form physical bonds or can be chemically crosslinked. The chemical covalent bonds between the polymers are often more stable and robust than the physical bonds formed due to intermolecular interactions between the polymer chains [10][11]. The physical bonds can vary in type from van der waals interaction, hydrogen bonds or ionic interactions. The physical hydrogels can also be due to just physical entanglements of the long polymer chains. [10][11].

1.3 Gelatin hydrogels

There are a lot of different natural biomacromolecules that are of great interest in tissue engineering. Gelatin is considered to be one of them, due to its low cost and low immunogenic response. Another favorable property with gelatin is that it contains the amino acid sequence Arg-Gly-Asp (RGD)[10][12]. In the extracellular matrix (ECM), the RGD sequences act as the integrin-binding sequences and therefore, provides focal cell-adhesion [10] [12]. Gelatin A and B are obtained from tissues by acidic and basic treatments, respectively. Gelatin is slightly less immunogenic than collagen and is easier to dissolve in water/organic solvents [13][14][15]. A gelatin solution behaves as a gel at 25 ºC but turns into a viscous solution above 35 °C. This sol-gel transition from higher to lower temperature can be explained by the formation of the polypeptide assemblies in the hydrogel.

However, hydrogels consist of only gelatin tend to swell in aqueous solutions which is unfavorable for corneal transplantation. To avoid swelling of the hydrogel, the gelatin chains often need to be covalently crosslinked [11]. However, chemical crosslinking of gelatin can be difficult and often requires harsh conditions. Hence, researchers have developed modified gelatins such as gelatin methacrylamide (GelMa)[16]. According to Jason W. Nichol and colleagues, GelMa has been exceptionally good for microfluidic devices and microtissues usage. These materials have been shown to promote strong cell migration and proliferation [16].

However, some studies have shown that the chain-growth photopolymerization used for

(8)

8 production of the GelMa hydrogels leads to lower compatibility with cells[17][18]. Therefore, a different approach was implemented in this thesis - thiol-ene step-growth polymerization initiated by UV light as developed by Tanja Greene and Chien-Chi Lin [17][18] (Figure 3).

Greene and Lin demonstrated that this reaction has higher compatibility for cells [17][18].

1.4 The Hofmeister effect

In the 19^th century, Franz Hofmeister observed the that different ions have a different effect on protein solubility and denaturation in aqueous solutions. Later on, scientists have observed similar behaviors of ions on thermoresponsive polymers in aqueous solutions [19][20]. The mechanism by which the observed phenomenon can be explained is still heavily debated and is not completely understood. The Hofmeister series of anions is denoted as follows [19][20]:

CO32−

> SO42−

> S2O32−

> H2PO⁴⁻ > F− > CH3COO− > Cl− > Br− > NO3− > I− > ClO4− >

SCN−

The anions residing on the right and left of the chloride are called chaotropes and kosmotropes, respectively. Chaotropes increase the protein solubility and do not compete for the water molecules (poorly hydrated ion) from the protein hydration shell. This prevents proteins from self-assembly and often causes protein denaturation. On the contrary, kosmotropes compete for the water molecules from the protein hydration shell and promotes protein self-assembly and in extreme case can lead to protein aggregation resulting in precipitation of the protein [19][20].

By use of kosmotropic ions Qingyuan He and colleagues have developed strong and ductile unmodified gelatin hydrogels [19]. In their approach more polypeptide self-assemblies were created by ammonium and sulfate ions, two well know kosmotropes, by soaking gelatin physical gels into ammonium sulfate solution. This, in turn, resulted in an improvement of the mechanical properties of the gelatin gels [19]. However, we observed that such improvement in the mechanical properties is only temporary and the gels revert back to their original poor strength when stored in water. This is expected as the ions are washed away when the gels are stored in water resulting in the break of the newly formed polypeptide assemblies. In this thesis, the feasibility of permanently stabilizing these polypeptide assemblies by covalent crosslinking is investigated.

1.5 Chemical crosslinking

There are two principal ways to generate radicals for chain growth or step growth polymerization, namely chemical and photochemical radical generation. In this thesis, photochemical generation of radical was used to initiate the thiol-ene reaction. For this purpose, lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP) was used as a photoinitiator because of its higher water solubility compared to commonly used photoinitiators such as Irgacure 2959 (I2959) [21][22]. Both LAP and I2959 are categorized as type I photo-initiators. After absorbing a photon, type I photo-initiators undergo unimolecular bond cleavage and generate two separate radicals. Figure 2 describes the thiol-ene reaction between gelatine norbornene and DTT that was used to covalently crosslink the hydrogels.

(9)

9

Figure 2. Reaction between LAP, DTT and gelatine norbornene in detail. [Drawn in Chemdraw 17,1 ]

2. MATERIALS

Gelatin (Typ A, Procine Skin 300 Blom) - Sigma Aldrich, Sweden Gelatin norbornene synthesized in the host lab

Ammonium sulfate - Sigma Aldrich, Sweden

Lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP) - Sigma Aldrich, Sweden

Dimethyl sulfoxide (DMSO) - Sigma Aldrich, Sweden

Dithiothreitol (DTT) - Sigma Aldrich, Sweden Deionized/degassed water

(10)

10 Carbic anhydride - Sigma Aldrich, Sweden

2,4,6-Trinitrobenzenesulfonic acid (TNBS) - Sigma Aldrich, Sweden Sodium bicarbonate - Sigma Aldrich, Sweden

HCl - Sigma Aldrich, Sweden NaOH - Sigma Aldrich, Sweden

3.METHODS

3.0 Synthesis of gelatin norbornene

10 g of porcine gelatin was dissolved in 1 L of deionized water in a round bottom flask by heating at 65 ºC. The resulting clear solution was cooled down to room temperature and the pH of the solution was adjusted from 6 to 10 by the dropwise addition of 2 M NaOH. Carbic anhydride (3.12 g, 5 eqv. w.r.t amines of gelatin considering 38 amines per gelatin molecule) dissolved in 30 ml of DMSO was added dropwise while maintaining the pH at 10 by intermittent addition of 2M NaOH. The reaction mixture was left overnight for stirring at room temperature.

The resulting reaction mixture was dialyzed (regenerative cellulose membrane with MWCO 12-14 kDa) against pH 10 water for 5-7 days followed by free drying to obtain norbornene- gelatin as a fluffy solid which was subjected to TNBS assay for determining the degree of modification and NMR characterization. The modification reaction has been described in Figure 3.

Figure 3. Synthesis of norbornene-gelatin.

3.1 TNBS assay

The chemistry of TNBS (2,4,6-trinitrobenzenesulfonic acid) assay to detect primary amines has been depicted in Figure 4 [23]. The conjugated trinitro derivative can be quantitatively detected by measuring its absorbance at 346 nm where all starting materials have negligible or no absorbance.

(11)

11

Figure 4. The reaction between TNBS and with the unreacted amines on the gelatin molecule. [Drawn in Chemdraw 17,1 ]

TNBS assay was performed on unmodified and norbornene modified gelatin to determine the degree of functionalization using the following equation:

(𝐴𝑏𝑠_𝑔𝑒𝑙− 𝐴𝑏𝑠_{𝑔𝑒𝑙𝑁𝐵})

𝐴𝑏𝑠_𝑔𝑒𝑙 𝑥100%

Definitions: 𝐴𝑏𝑠_𝑔𝑒𝑙:Absorbance of gelatin. 𝐴𝑏𝑠_{𝑔𝑒𝑙𝑁𝐵}: Absorbance of gelatin norbornene 6 mg of sample was dissolved in 4 w% NaHCO3 solution. 1 ml of a 0.5% TNBS solution was added. The blank was prepared by mixing 1 ml of 4 w% NaHCO3 solution, 1 ml of 0.5% TNBS solution and 3 ml of 6M HCl. All solutions were incubated in water bath with mild shaking at 40 ˚C for 2h, the unreacted TNBS was extracted with diethyl ether. The organic phase was discarded. All samples were measured at 346 nm, in a Lambda 35 UV/Vis spectrophotometer from PerkinElmer.

3.2 NMR

In order to prove the presence of norbornene functional group in the modified gelatin, NMR spectra were recorded for both unmodified and modified gelatin. Spectra were recorded on a Jeol JNM-ECP Serie FT-NMR with a field strength of 9.4T, operating at 400 MHz. Samples were recorded overnight to obtain a good signal-to-noise ratio. About 5 mg of sample was dissolved in 500 µl of D2O.

3.3 Preparation of hydrogels

The schematic in Figure 5 shows the steps used in the hydrogel preparation. The preparation protocol builds on the methodology developed by Qingyuan He et al, with some added necessary steps [19]. The total solid content for the GNB-SH gels were kept constant at 10 w%

while maintaining the norbornene:SH ratio to 1:1. For preparing gels with unmodified gelatin, the gelatin concentration in the hydrogel was maintained as indicated in individual experiment in Figure 11. First calculated amount of degassed water was added to the required amount of powdered gelatin (modified or unmodified) and heated to 50 ˚C in a water bath to obtain a solution. To this solution was added the required amounts of DTT to obtain SH:norbornene ratio of 1:1 and LAP to obtain a final concentration of 0.5 w% in the hydrogel. The resulting solutions were cast into syringe molds and kept in fridge for cooling at 4 °C for one hour. After this step, the process varied depending on the gel type. Thus, some of the samples were transferred to petri-dishes with 20% (NH2)4SO4 and 0.5% LAP (to prevent diffusion of LAP

(12)

12 out from the gels). The samples were stored overnight (12h). On the next day, the ammonium sulfate solution was removed, and a selected number of samples were irradiated for 2x5 min to create the covalent crosslinks. Four different gel types were prepared, gelatin norbornene-DTT (not crosslinked) (GNB-SH), gelatin norbornene-DTT-Hofmeister treated (not crosslinked) (GNB-SH-H), gelatin norbornene-DTT-UV crosslinked- (GNB-SH-UV), gelatin norbornene- DTT-Hofmeister-UV crosslinked (GNB-SH-H-UV).

Figure 5. Step by step schematic over the hydrogel preparation. [Drawn in inkscape, paint 3-D, Chemdraw 17,1]

3.4 Swelling test for the hydrogels

A swelling analysis was performed to find the right concentration for LAP. The least swollen gelatine hydrogel would indicate the best suitable concentration of LAP. Hydrogels with four different concentrations of LAP were prepared, 0.4, 0.5, 0.6 and 0.7 w%. The swelling of the hydrogels at 25 ºC was determined by weighing the gels after blotting the surface water before swelling or after 15 min of swelling in water.

The following equation was used for the calculation:

S=[^(𝑊𝑡¹⁵⁾

𝑊𝑡0 ] 𝑥100%

𝑊𝑡0 stands for initial weight (in grams) at 0 min, and 𝑊𝑡15is the weight after 15 min swelling in deionised water.

3.5 Rheology

Gels were characterised for their mechanical properties using a Discovery Hybrid Rheometer 2 (DHR-2) (TA Instruments, Sollentuna, Sweden). All gels were cut to a size of 8mm in diameter and 1.5 mm in thickness. Measurements were performed with an 8 mm parallel plate stainless

(13)

13 steel geometry at 25 °C. The storage moduli values at 1 Hz is used for comparison. All gels were measured as prepared/treated followed by washing for various duration and measured again.

3.6 TGA

Thermogravimetric analysis (TGA) is an analytical measurement technique which allows the measurement of the change in weight during a temperature increase. The samples are heated often in a sealed chamber with different gases, in this experiment we used air. The temperature which the samples can be heated from is between the range of 25 °C to 1000 T °C [24]. The measurements were performed using a TGA Q500 (TA Instruments, Sollentuna, Sweden) with sample sizes of 10-25 mg. The temperature was increased at a rate of 10°C/min to a maximum of 250 °C. This experiment was performed to determine the water content of the hydrogels.

4.RESULTS & DISCUSSION 4.1 Gelatin norbornene synthesis

4.1.1 TNBS results

The results from the three different TNBS assay experiments show different degree of modifications on the gelatin. The degree of modification was found to be 35-68% (Table 1).

This variation in the degree of modification could be due to the variations in the gelatin stocks or due to an experimental error.

Table 1: UV absorbance data for TNBS assay. Calculation of degree of modification in percentage with respect to amines in gelatin.

Gelatin

Gelatin norbornene

Degree of modification %

Sample 1 0,471 0,151 68%

Sample 2 0,471 0,255 46%

Sample 3 0,471 0,305 35%

4.1.2 NMR

The ¹H-NMR spectra of unmodified gelatin and norbornene functionalized gelatin are shown in Figure 6. The vinylic proton peaks originating from the norbornene functional group appear in the region 5.99-6.30 δ-ppm. Hence, from the NMR data it can be concluded that the

functionalization of gelatin with carbic anhydride to obtain norbornene modified gelatin was successful. These results are in agreement with previously reported data on gelatin

functionalization with norbornene moieties [18].

(14)

14

Figure 6. NMR spectra of gelatin and gelatin norbornene indicating the peaks originating from the norbornene functional group.

4.2 Swelling measurement

Hydrogel swelling measurements indicate the efficacy of crosslinking in the hydrogel. To determine the optimum photo-initiator concentration, gels were prepared using various

concentrations of LAP photo-initiator, irradiated at 365 nm and subjected to swelling in water (Figure 7). Least swelling, thus indicating highest degree of covalent crosslinking between thiol and norbornene groups, was observed at a LAP concentration of 0.5%. Hence, all following experiments were conducted with 0.5% LAP concentration. These results are in agreement with radical step-growth polymerization where too few or too much generation of free radicals in the initiation step decreases the degree of crosslinking. This is due to either very low concentrations of thiyl radicals generated or due to radical-radical coupling occurring at higher radical concentrations, respectively.

Figure7. The collected data points from the swelling experiment to find the optimal PI concentration. Data presented as mean ± standard deviation, n = 4.

1,5 1,7 1,9 2,1 2,3 2,5 2,7 2,9

0,4 0,5 0,6 0,7

Increase in weight x times

PI Concentration in %

Swelling test

(15)

15

4.3 Hydrogels mechanical properties

The thiol to norbornene ratio was kept constant at 1:1 for all gel formulations to obtain maximum crosslinking. The hydrogels were subjected to oscillatory rheology to determine their modulus. The washing process was performed to wash out the side-products from the thiol-ene reaction and to observe the gels´ behavior when stored in water for a longer duration. The storage moduli (G´) values at 1 Hz, which is well within the linear viscoelastic region in a frequency sweep experiment, was taken for comparison (Figure 8-10). Decrease in moduli values upon storage of the hydrogels in water is indicative of swelling and loss of mechanical strength and stiffness. Complementary pictures showing the behavior of gels during washing can be found in the appendix (Figure 14-24).

Figure 8. Average G’ values obtained at 1Hz after 0, 15, 60 and 120 min washing of GNB-SH, GNB-SH-H, GNB-SH-UV and GNB-SH-H-UV gels in water. NW indicates non-washed samples. Data reported as mean ± SD, n=6.Figure 8

The effect of storing the hydrogels for up to 2 hours on their mechanical properties are shown in Figure 8. Comparisons were made between GNB-SH, GNB-SH-H, GNB-SH-H-UV, and GNB-SH-UV (see page 12 for description of the gels).

As expected, GNB-SH showed the lowest initial G´ value. GNB-SH-H showed a significant increase in the G´ value, which could be expected from earlier published work by Qingyan et al [19]. The moduli values of GNB-SH-UV were higher than GNB-SH but not higher than the GNB-SH-H before storing them in water. This is plausibly due to the physical crosslinking formed as a result of the polypeptide assemblies created by the Hofmeister effect. The shorter the distance to each chain the higher the possibility for entanglement [11]. However, light scattering experiments should be performed to further confirm this hypothesis. Due to lack of time and equipment it was out of the scoop of this thesis. The picture in Figure 14 in the appendix, shows the size difference between the gels. The ammonium sulfate treated gels were smaller in width and height than the non- treated gels which could be explained due to the difference in osmotic pressure between the gels which are prepared in water and ammonium

GNB- SH- NW

GNB- SH-

H- NW

GNB- SH- UV- NW

GNB- SH-

H- UV- NW

GNB- SH- 15m

GNB- SH-

H- 15m

GNB- SH- UV- 15m

GNB- SH-

H- UV- 15m

GNB- SH-

1h GNB-

SH- H-1h

GNB- SH- UV- 1h

GNB- SH-

H- UV-

1h GNB-

SH- 2h

GNB- SH-- H 2h

GNB- SH- UV- 2h

GNB- SH-

H- UV-

2h Av(G') 2745, 29068 9829, 50119 2296, 145711465942700 3414, 171801613237317 3637, 227731676434673

0,0 10000,0 20000,0 30000,0 40000,0 50000,0 60000,0 70000,0

Storage Modulus (G') Pa

Values take at 1 Hz

0 min-2h Washing

(16)

16 sulfate solution. The slight yellow color of some of the gels is likely due to the photodecomposition products of LAP. However, more experiments are required to confirm this assumption.

The results obtained after washing the gels from 15 min and up to 2h (Figure 8) reveals that the trend of the G´ values is different for different gels. The ammoniums sulfate treated gels (GNB-SH-H and GNB-SH-H-UV) have in general a much higher G´ value when compared to the non-treated gels. However, the loss in the G´ values of the hydrogels during storage in water is significantly higher for non-crosslinked gels (GNB-SH-H) compared to UV-crosslinked gels (GNB-SH-H-UV). This phenomenon indicates that Hofmeister effect assisted polypeptide self- assemblies in the hydrogels are reversible and such gels cannot be stored in water without compromising on their mechanical properties. However, when the assemblies are permanently stabilized by covalent UV-crosslinking, the resulting gels do not lose their mechanical strength when stored in water. The slight decrease followed by increase in G´ for all gels except GNB- SH-H-UV during storage can be due the combination of three different competing effects. (1) Increase in modulus due to polymer chain stretching during swelling; (2) decrease in modulus due to the size increase of gels during swelling (dilution of polymer chains in gels); (3) reformation of gelatin polypeptide self-assemblies under thermodynamic equilibrium (appendix Figure 16-19). However, more experiments will be needed to confirm this hypothesis [11]. Hence, from Figure 8 and appendix Figure 17-18, it can be concluded that through covalent crosslinking it is possible to permanently stabilize the Hofmeister effect assisted polypeptide self-assemblies in gelatin gels and the resulting gels retain their mechanical strengths when stored in water for up to 2 hours.

Figure 9. Average values are taken at 1Hz. 18h washing and 36 washing for GNB-SH, GNB-SH-H, GNB-SH-UV and GNB-SH-H- UV. The standard deviation was calculated by measurement of 6 different gels on each bar.

The behavior of the gels during 18-36 hours of storage in water is decribed in Figure 9. The G' for both GNB-SH-H and GNB-SH-UV decreased by almost 5 kPa during storage in water for 18 hours. The GNB-SH-H gel swelled to almost double its size. However, polypeptide assemblies started to break druing the 18-36h of washning and did not reassemble again.The GNB-SH gels started to degrade at a faster rate. These gels had almost lost all their integrity and shape. The last gel could be measured at 54h for the GNB-SH. The only gel with a gradual

GNB-SH- 18h

GNB-SH-H- 18h

GNB-SH- UV-18h

GNB-SH-H- UV-18h

GNB-SH- 36h

GNB-SH-H- 36h

GNB-SH- UV-36h

GNB-SH-H- UV-36h AV(G') 3828,6 17472,4 11738,5 40071,9 1647,5 15334,9 9867,1 30496,3

0,0 10000,0 20000,0 30000,0 40000,0 50000,0 60000,0

Storage Modulus (G´) Pa

18h-36h washing

(17)

17 decreasing slope in G´ is GNB-SH-H-UV during the 18-36h of storage. However, after almost 36 h of storing in H2O, the G’ was at high levels. All gels became transparent after this period of time.

Figure 10. Average values are taken at 1Hz. 54-90h washing for GNB-SH, GNB-SH-H, GNB-SH-UV and GNB-SH-H-UV. The standard deviation was calculated by measurement of 6 different gels on each bar.

After 54 h of washing in the aqueous medium, all of the gels, except GNB-SH-H-UV, had lost their integrity and shape. Figure 25 in the appendix shows the width and hight of the four different disks. After the 54h measurement, the GNB-SH gels fell apart, thus explaining the low G' value. The same type of behavior was shown for GNB-UV. The longer the gels were washed the worse the integrity of the shape of the gel were observed. This can be seen in Figure 26 in the appendix. GNB-SH-H gels showed some signs of degradation, such as cracks in the gel.

The only gel that kept its structural integrity was the GNB-SH-H-UV, for which almost no visible changes could be noted.

For confirmation of the reversibility of the Hofmeister effect assisted polypeptide assemblies in gelatin hydrogels and concomitant increase in the gel modulus, few different hydrogels were prepared using unmodified gelatin and evaluated; (1) a gelatin physical gel with 10 w% gelatin concentration (G10 Control), (2) G10 Control soaked overnight in 20% ammonium sulfate solution to further promote the Hofmeister effect assisted polypeptide assembly in the gel (G10A20), (3) G10A20 washed for 15 min in water (AFW15m_G10A20), (4) a gelatin physical gel with 15 w% gelatin concentration soaked overnight in 20% ammonium sulfate solution to further promote the Hofmeister effect assisted polypeptide assembly in the gel (G15A20), (5) G15A20 washed for 15 min in water (AFW15m_G15A20). The moduli values are described in Figure 11. The results, shown in Figure 11, clearly reveal that the kosmotropic ion directed polypeptide assemblies in the gelatin physical gels are completely reversible and the concomitant improvement in the mechanical properties are only temporary. Such strong gels cannot be stored in water solution and therefore, poses a great limitation for use in biomedical or clinical applications.

GNB- SH-54h

GNB- SH-H- 54h

GNB- SH-UV-

54h

GNB- SH-H- UV- 54h

GNB- SH-72h

GNB- SH-H- 72h

GNB- SH-UV-

72h

GNB- SH-H- UV- 72h

GNB- SH-90h

GNB- SH-H- 90h

GNB- SH-UV-

90h

GNB- SH-H- UV- 90h AV(G') 917,5 16157, 5710,9 38453, 0,0 16376, 5091,1 34313, 0,0 16315, 4386,6 41955,

0,0 10000,0 20000,0 30000,0 40000,0 50000,0 60000,0 70000,0

Storage Moldulus (G') Pa

Values take at 1 Hz

54h-90h washing

(18)

18

Figure 11 Average G’ values obtained at 1Hz. 0 min before washing and 15 min after washing for 10% and 15 Gelatin gels soaked in 20% ammonium sulfate. AFW denotes after washing. For description of the gel abbreviations, see p12.

4.4 Thermogravimetric analysis (TGA)

The TGA was measured to see the water content of four of the GNB-SH-H-UV disks. However, due to the instrument unavailability, the measurement was not performed until two weeks after the gels were measured by rheology. The gels which were supposed to be measured was already divided into smaller pieces. During this period water was changed every day. The results are shown in Figure (12-15). Average water content for these hydrogels were found to be 9-10%.

However, as mention earlier, due to several experimental and instrument related errors the results could be affected and therefore, should be repeated.

Figure 12 TGA diagram which shows the water content for disk 1 of GNB-SH-H-UV G10 Control G10A20 AFW15m_G10A

20 G15A20 AFW15m_G15A

20

G' 10913 55235,7 6018,85 61301,4 10913

0 10000 20000 30000 40000 50000 60000 70000

Storage Modulus (G´) Pa

Values taken at 1Hz

0-15 min washing

(19)

19

Figure 13. TGA diagram which shows the water content for disk 2 of GNB-SH-H-UV.

(20)

20

5. Conclusion

By soaking gelatin and gelatin norbornene hydrogels in an ammonium sulfate solution, we have created strong and ductile gels. In this thesis we have shown that it is possible to reproduce the work of Qingyan et al [19]. However, upon storing in an aqueous medium, the gels revert back to their initial properties. By introducing permeant covalent crosslinking in GNB gels, the gels were locked in the state with aggregated polypeptide chains and therefore, these polypeptide assemblies cannot be broken upon storing the gel in water. The bonds created by covalent crosslinking hindered the gels from swelling and the GNB gels showed a remarkable ability to retain their modulus even after 90h of storage in water.

(21)

21

6.REFERENCES

1. Bourne RRA, Flaxman SR, Braithwaite T, Cicinelli M V., Das A, Jonas JB, et al.

Magnitude, temporal trends, and projections of the global prevalence of blindness and distance and near vision impairment: a systematic review and meta-analysis. Lancet Glob Heal. 2017;5(9):e888–97.

2. Eslani M, Baradaran-Rafii A, Movahedan A, Djalilian AR. The Ocular Surface Chemical Burns. J Ophthalmol. 2014;2014:1–9.

3. Popa M, Lazar V, Iordache C, Beresteanu ŞVG, Ficai A, Grigore R, et al. Soft tissue engineering and microbial infections. Nanobiomaterials Soft Tissue Eng. 2016;1–29.

4. Temenoff,J.S, Mikos A.G, Biomaterials: The Intersactio of biology and materials science: Pages 2-150, publicist: 2009

5. Raghavendra GM, Varaprasad K, Jayaramudu T. 2.1 overview. 2015;21–44.

6. B. Yue, “Biology of the Extracellular Matrix: An Overview,” J. Glaucoma, vol. 23, pp. S20– S23, 2014.

7. Tang X, Thankappan SK, Lee P, Fard SE, Harmon MD, Tran K, et al. Polymeric Biomaterials in Tissue Engineering and Regenerative Medicine. 2014;351–71.

8. Sumrita Bhat and Ashok Kuma, Biomaterials and bioengineering tomorrow’s healthcare. Biomatter 2013; 3: e24717

9. M. Mirazul Islam, Biomaterials-enabled cornea regeneration in patients at high risk for rejection of donor tissue transplantation, npj Regenerative Medicine (2018)

10. Lee J, Kim H. Emerging properties of hydrogels in tissue engineering. 2018;0–3.

11. Hellio D, Djabourov M. Influence of Weak and Covalent Bonds on Formation and Hydrolysis of Gelatin Networks Se. 2004;1662–6.

12. Bellis SL. NIH Public Access. 2012;32(18):4205–10.

13. Tang X, Thankappan SK, Lee P, Fard SE, Harmon MD, Tran K, et al. Polymeric Biomaterials in Tissue Engineering and Regenerative Medicine. 2014;351–71

14. Liu D, Nikoo M, Zhou P, Regenstein JM. Collagen and Gelatin. Annu. Rev. Food Sci.

Technol. 2015. 6:527–57

15. Singh S, Rao KVR, Venugopal K, Manikandan R. Alteration in Dissolution

Characteristics of Gelatin-Containing Formulations A Review of the Problem, Test Methods , and Solutions. 2002;(April).

16. Nichol JW, Koshy ST, Bae H, Hwang CM, Yamanlar S, Khademhosseini A.

Biomaterials Cell-laden microengineered gelatin methacrylate hydrogels. Biomaterials [Internet]. 2010;31(21):5536–44. Available from:

http://dx.doi.org/10.1016/j.biomaterials.2010.03.064 17. Lin C. Biomaterials Science. 2014;2(8).

18. Greene T, Lin CC. Modular Cross-Linking of Gelatin-Based Thiol-Norbornene Hydrogels for in Vitro 3D Culture of Hepatocellular Carcinoma Cells. ACS Biomater Sci Eng. 2015;1(12):1314–23.

19. He Q, Huang Y, Wang S. Hofmeister Effect-Assisted One Step Fabrication of Ductile and Strong Gelatin Hydrogels. 2018;1705069:1–10.

20. Jaspers M, Rowan AE, Kouwer PHJ. Tuning Hydrogel Mechanics Using the Hofmeister Effect. 2015;6503–10.

21. Thakur Raghu Raj Singh, Garry Laverty, Ryan Donnelly Hydrogels: Design, Synthesis and Application in Drug Delivery and Regenerative Medicine CRC Press Published February 28, 2018

(22)

22 22. Benjamin D. Fairbanks Photoinitiated polymerization of PEG-diacrylate with lithium phenyl-2,4,6- trimethylbenzoylphosphinate: polymerization rate and cytocompatibility, Biomaterials 30 (2009) 6702–6707

23. Fields R. TNBS-amino acid. 1969;454(1960):464–8.

24. Heal GR. Thermogravimetry and Derivative Thermogravimetry, Chapter 2 pages:10-, publicist:2002.

25. Copyright owned by AllAboutVision.com, 2019-03-21

APPENDIX

Figure 16. Size comperasing comperasing 0 min before washing and rheology testing for GNB-SH, GNB-SH-H, GNB-SH-UV and GNB-SH-H-UV.[Picture taken by my phone]

Figure 17. Size comperasing comperasing 15 min after washing and rheology testing for GNB-SH-UV and GNB- SH-H-UV.[Picture taken by my phone]

(23)

23

Figure 18. Size comperasing , 15 min after washing and rheology testing for GNB-SH-(no crosslinking), and 1h of washing for GNB-SH-UV and GNB-SH-H-UV.[Picture taken by my phone]

Figure 19. Size comperasing comperasing , 1h of washing and rheology testing for GNB-SH-(no crosslinking), and 2h of washing for GNB-SH-UV and GNB-SH-H-UV.[Picture taken by my phone]

Figure 20. Size comperasing , 15 min of washing and rheology testing for GNB-SH-H(no crosslinking), and 2h of washing for GNB-SH-(no crosslinking), GNB-SH-UV and GNB-SH-H-UV.[Picture taken by my phone].

(24)

24

Figure 21. Size comperasing , 2h of washing and rheology testing for GNB-SH-H(no crosslinking) GNB-SH-(no crosslinking), GNB-SH-UV and GNB-SH-H-UV.[Picture taken by my phone]

Figure 22. Size comperasing , 18h of washing and rheology testing for GNB-SH-H(no crosslinking) GNB-SH-(no crosslinking), GNB-SH-UV and GNB-SH-H-UV.[Picture taken by my phone].

(25)

25

(26)

26

Irreversible strengthening of gelatin hydrogels by salt-directed polypeptide assembly and covalent crosslinking