Heparin Functionalized Injectable Cryogel with Rapid Shape

and basic fibroblast growth factor. 13. (bFGF) have heparin binding domain that allow specific affinity with heparin functionalized. 14 biomaterials. ...
3 downloads 3 Views 3MB Size
Subscriber access provided by Kaohsiung Medical University

Heparin Functionalized Injectable Cryogel with Rapid Shape-Recovery Property for Neovascularization Inseon Kim, Seunghun S. Lee, Sunghoon Bae, Hoyon Lee, and Nathaniel S. Hwang Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.8b00331 • Publication Date (Web): 24 Apr 2018 Downloaded from http://pubs.acs.org on April 25, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

Heparin Functionalized Injectable Cryogel with

2

Rapid Shape-Recovery Property for

3

Neovascularization

4

Inseon Kim1, ‡, Seunghun S. Lee2, ‡, Sunghoon Bae1, Hoyon Lee1 and Nathaniel S. Hwang1, 2, 3*

5 6

1

7

National University, Seoul, 08826, Republic of Korea

8

2

9

Republic of Korea.

School of Chemical and Biological Engineering, the Institute of Chemical Processes, Seoul

Interdisciplinary Program in Bioengineering, Seoul National University, Seoul, 08826,

10

3

11

‡These authors contributed equally.

12

*Corresponding author

BioMAX/N-Bio Institute, Seoul National University, Seoul, 08826, Republic of Korea

13 14 15 16 17 18 19

ACS Paragon Plus Environment

1

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 40

1

ABSTRACT

2

Cryogel based scaffolds have high porosity with interconnected macropores that may provide

3

cell compatible microenvironment.

4

minimally invasive surgery due to its sponge-like properties, including rapid shape recovery and

5

injectability. Herein, we developed an injectable cryogel by conjugating heparin to gelatin as a

6

carrier for vascular endothelial growth factor (VEGF) and fibroblasts in hindlimb ischemic

7

disease. Our gelatin/heparin cryogel showed gelatin concentration-dependent mechanical

8

properties, swelling ratios, interconnected porosities, and elasticities. In addition, controlled

9

release of VEGF led to an effective angiogenic responses both in vitro and in vivo. Furthermore,

10

its sponge-like properties enabled cryogels to be applied as an injectable carrier system for in

11

vivo cells and growth factor delivery. Our heparin functionalized injectable cryogel facilitated

12

the angiogenic potential by facilitating neovascularization in hindlimb ischemia model.

In addition, cryogel based scaffolds can be utilized in

13 14 15

Keywords: heparin, cryogel, injectable, VEGF, neovascularization

16 17 18 19 20 21 22 23

ACS Paragon Plus Environment

2

Page 3 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

1. Introduction

2

Cryogel based scaffolds are characterized as hydrogels with high porosity and interconnected

3

macropores.1, 2 Due to its highly interconnected and macroporous structure, cryogels provide

4

cell compatible microenvironment that promotes cell proliferation and host cell infiltrating into

5

the cryogel upon implantation.3-8 In addition, cryogels have been investigated for drug or

6

therapeutic protein carriers for a site-specific release.9 However, unintended and uncontrolled

7

burst release of the therapeutic agents from cryogels have been considered as a major challenge

8

due to their high porosity.9, 10

9

For controlled release of growth factors from cryogels, several biomaterials moieties can

10

be incorporated in cryogels for sustained release of growth factors 11. For example, heparin is a

11

sulfated glycosaminoglycan of the extracellular matrix (ECM), and it has been widely used in

12

protein delivery due to its affinity to several growth factors. Vascular endothelial growth factor

13

(VEGF),12-14 bone morphogenic protein-2 (BMP-2),15-17 and basic fibroblast growth factor

14

(bFGF) have heparin binding domain that allow specific affinity with heparin functionalized

15

biomaterials.18-20 This characteristic capacity of heparin helps to minimize the undesirable burst

16

release of proteins from heparin functionalized cryogels, which improve therapeutic effect by

17

controlled release of the therapeutic proteins for a long period. In particular, heparin induced

18

controlled release of VEGF has been reported as a promising method to promote vascularization

19

in ischemic hind limb model.21-23

20

In recent years, injectable biomaterials are in high demand in regenerative medicine

21

owing to the need for minimally invasive surgery.24 However, injectable biomaterials based on

22

hydrogel with small pores inhibit cell infiltration into the hydrogel, leading to an ineffective

23

tissue regeneration.25 In contrast, spongy-like injectable cryogel with uniformly formed

ACS Paragon Plus Environment

3

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 40

1

macropores may promote effective cellular proliferation and host cell infiltration. In addition, the

2

use of injectable cryogel prevents tissue injury from remained crosslinking reagents in the

3

hydrogel because remained crosslinking reagents will be released from swelling and drying

4

process prior actual usage

5

injectable cryogel system, including dibenzaldehyde-terminated poly (ethylene glycol) (PEG-

6

DA) 27, methacrylated gelatin (Gel-MA) 28 and hyaluronic acid (HA).29

7

26

. Following this trend, numerous macromolecules were utilized in

In this study, we have developed an injectable heparin/gelatin based cryogel for the

8

treatment of in vivo hind-limb ischemia model.

Furthermore, cryogels were designed and

9

fabricated to have (i) optimal VEGF release profile for functional angiogenesis and (ⅱ) optimal

10

injectability with viable cells.

Finally, we applied this cryogel to in vivo hind-limb ischemic

11

mouse model and demonstrate the recovery of blood flow and vascularization through the

12

necrotic muscle tissue.

13 14 15 16 17 18 19 20 21 22 23

ACS Paragon Plus Environment

4

Page 5 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

2. Experimental Section

2

2.1 Preparation of gelatin-heparin cryogels

3

Type A gelatin (from porcine skin, Sigma-Aldrich, G2500) and heparin (Merck Millipore,

4

375095) were respectively dissolved in deionized water and mixed together at each final

5

concentration of 1, 1.5 % (w/v) gelatin and 0.1, 0.3, 0.5 % (w/v) heparin. 50mM of 1-Ethyl-3-(3-

6

dimethylaminopropyl)carbodiimide (EDC) (Thermo Scientific, PG82073) and 25mM of sulfo-

7

Hydroxysuccinimide (NHS) (Thermo Scientific, PG82071) dissolved in deionized water were

8

respectively added into the gelatin/heparin solution as crosslinking reagents. After mixing all the

9

compounds, 85µL of the mixed solution was immediately pulled on pre-cooled cryogel mold and

10

placed on -20℃ overnight for cryogelation. During cryogelation, ice crystals were formed and

11

crosslinking reaction between gelatin and heparin was processed in the entire regions except for

12

ice crystals. When the cryogelation finished, scaffolds were lyophilized overnight and stored at -

13

20℃. All the followed experiments were processed with 5mm diameter, 3mm height cylindrical

14

cryogels.

15 16

2.2 Characterization of gelatin-heparin cryogels

17

For scanning electron microscopic (SEM) images, each cryogels dehydrated at least for 24hrs

18

fixed onto the stubs using carbon tape and coated with platinum scatter. Microstructure and pores

19

of the cryogels were visualized at 10kV with a microscope (Hitachi S3500-N, Hitachi Science

20

Systems Ltd., Tokyo, Japan). The area of pores was quantified by ImageJ. For swelling property,

21

lyophilized cryogels were weighed on a scale and incubated in PBS at RT for 24hrs. When

22

cryogels were fully swelled, the weights were measured and the swelling ratio of each cryogels

23

was calculated by the followed equation: Swelling ratio = (Ws-Wi)/Wi , where Wi means the

ACS Paragon Plus Environment

5

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 40

1

initial weight of lyophilized cryogel and Ws means the weight of fully swelled cryogel. The pore

2

connectivity analysis was examined as previously described by Bencherif et al.

3

cryogels were dehydrated on kimwipe until the water was completely removed away. The

4

percentage of interconnected porosity was calculated by the following equation: %

5

interconnected porosity = (Ws-Wde)/Ws, where Wde means the weight of dehydrated cryogels on

6

kimwipe. For the mechanical property, compression test was carried out using universal-testing-

7

machine (Shimadzu EZ-SX, Japan) with fully swelled cryogels. Gels were compressed with the

8

loading rate of 1mm/min. The result was obtained by the stress-strain curve and Young's

9

modulus was calculated by the measurement of the slope linearly increased in the region of the

10

30

Swelled

stress-strain curve.

11

For enzymatic degradation test, each scaffold was incubated in 1 unit/mL or 20 unit/mL

12

of Type ℃ collagenase (Worthington, LS004176) solution at 37℃. Collagenase solution was

13

replaced with fresh solution every 2-3 days. The degree of gel maintenance was calculated by the

14

percentage of the weight of remained cryogel to initial weight of the fully swelled cryogel.

15 16

2.3 Rheology test

17

Rheological behavior test was carried out in Demo lab (Anton Paar Korea) using rheometer

18

(MCR 302, Measuring cell: P-PTD & H-PTD 200, Measuring System: PP 25, Anton-Paar,

19

Austria). All the cryogels used in experiments had 8mm diameter and 2mm thickness cylindrical

20

shape. Firstly, amplitude sweep (strain sweep) test was applied from 0.01 to 100% at the

21

controlled frequency of 1Hz to determine the rheological change related to the shear stain.

22

Secondly, frequency sweep measurement was performed from 0.01 to 100Hz at a constant strain

23

of 0.2% to investigate the modulus change related to the oscillatory frequency. Finally,

ACS Paragon Plus Environment

6

Page 7 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

continuous step strain was measured by alternating 0.2% and 25% for 100s at a constant

2

frequency of 1Hz to confirm the shape recovery behavior.

3 4

2.4 Heparin detection via alcian blue assay

5

Unreacted heparin during cryogelation was detected by alcian blue assay as followed under the

6

modified standard protocol.31 Firstly, after cryogels swelled in PBS for 24hrs, the PBS was

7

transferred to 1.5mL tube. 20µL of collected PBS was mixed with 0.05% (w/v) alcian blue 8GX

8

(Sigma-Aldrich, A5268) in DW. After 3hrs incubation at RT, the precipitate of heparin was

9

formed and collected by centrifuging at 10,000rpm for 10min as a pellet. The supernatant was

10

removed and the pellet was washed with 500µL of 99.9% ethanol under centrifugation. After

11

that, ethanol was completely aspirated and the pellet was dissolved in 500µL of 7.5% (w/v)

12

sodium dodecyl sulfate (SDS) (Bio-Rad, 1610302) solution for an hour at 60℃. The absorbance

13

of the dissolved solution was detected at 678nm and then, analyzed by the standard curve with 0-

14

0.25mg/µL of heparin solution.

15 16

2.5 Synthesis of RITC-heparin-conjugated cryogel

17

Rhodamine B isothiocyanate (RITC) conjugated heparin was synthesized by adding 1mg of

18

RITC (Sigma-Aldrich, R1755) dissolved in 0.1mL of Dimethyl sulfoxide (DMSO) (Sigma-

19

Aldrich, M81802) to 2% (w/v) heparin solution for 1hr at 60℃. After the reaction, RITC-heparin

20

was placed in 3.5K MW dialysis tube and dialyzed in distilled water for 3days to remove

21

unbound RITC. Dialysis water was freshly replaced every day. In the last day, RITC-heparin was

22

freeze-dried and used for the fabrication of cryogels as described in 2.1. The microstructure of

ACS Paragon Plus Environment

7

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 40

1

RITC-conjugated cryogel was visualized with a confocal microscope (Olympus FluoView

2

FV100).

3 4

2.6 Heparin mediated protein release kinetics

5

For the observation of heparin-mediated sustained release of VEGF, 30µL of 0.01% (w/v) FITC-

6

BSA (Sigma-Aldrich, A9771) or 100ng/mL human VEGF (Pepro tech, 100-20) was soaked to

7

each dehydrated cryogels. After 5min to bind proteins to heparin, 500µL of PBS was added to

8

the gels and the PBS was collected at each time point. The amount of released FITC-BSA was

9

measured by fluorescence spectroscopy at excitation wavelength 490nm, emission wavelength

10

525nm. The amount of released VEGF was measured by Human VEGF ELISA Kit

11

(KOMABIOTECH, K0331132).

12 13

2.7 HUVEC migration and tube formation assay

14

Firstly, 1x104 Human Umbilical Vein Endothelial Cells (HUVEC) were seeded in gelatin coated

15

24-well plate. Then, HUVECs were primed in daily collected assay medium (EBM-2 with 10%

16

FBS containing VEGF released from cryogels) for a week. For wound migration assay, 100%

17

confluent HUVECs were scratched by 1000µL pipet tip, which made vertical empty space. Cells

18

were washed with PBS and incubated in the assay medium. After 2days, the cells were fixed in

19

4% PFA for 10min and stained with DAPI. The rate of migration healing was quantified by

20

measuring the gap distance compared to the beginning interval. For tube formation assay, 150µL

21

of BD matrigelTM matrix (BD Biosciences, 354234) in ice was transferred to a pre-cooled 48-

22

well plate and incubated for 30min at 37℃. 2.5x105 VEGF primed HUVECs were trypsinized,

ACS Paragon Plus Environment

8

Page 9 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

suspended in 125µL of assay medium and seeded onto the matrigel. After 8hrs incubation, cells

2

were visualized using a light microscope.

3 4

2.8 Injectability test

5

To confirm the injectability of the scaffolds, cryogels were loaded in 5ml syringe with 1mL of

6

PBS and injected through the 17G needle to 50mL tubes. The remained cryogels after injection

7

were weighed and the degree of weight loss was calculated by the ratio of lost weight to initial

8

weight of cryogels. For the degree of FITC-BSA retention, 0.01 % (w/v) FITC-BSA loaded

9

cryogels were utilized. Original cryogels and remained gels after injection was completely

10

digested in Papain (Worthington, LS003127) for 16hrs at 60℃. The digested solution was

11

analyzed using fluorescence spectroscopy at excitation wavelength 490nm, emission wavelength

12

525nm. The ratio was calculated by the same method of weight retention.

13 14

2.9 Preparation of cell-seeded scaffolds

15

Freeze-dried cryogels were swelled in PBS for 24hrs to remove excess EDC. Swelled cryogels

16

were washed 2 more times in PBS and sterilized by UV irradiation for 3hrs. Then 5x104 NIH-

17

3T3 fibroblast cells, mouse embryo fibroblasts, were suspended in 30µL of growth medium

18

(DMEM, 10% FBS, 1% L-glutamate, 1% Pen-strep). Cell solution was soaked to the sterile

19

dehydrated cryogel and incubated at 37℃ for 2-3hrs for cell attachment. After that, 300µL of

20

fresh medium was added to the cryogels to provide sufficient nutrients for cell growth.

21 22

2.10 PicoGreen and live/dead assay

ACS Paragon Plus Environment

9

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 40

1

For the confirmation of the cell number preserved in the scaffold after injection, injected cell-

2

seeded scaffolds were degraded by papain digestion as described in 2.8. Remained cells were

3

quantified with PicoGreen™ dsDNA Assay Kit (Invitrogen, P11496). This method was also

4

applied to original cell-seeded scaffolds as a control. For the viability test, cell-seeded scaffolds

5

before, after injection were stained for 30min in 0.5µL/mL of calcein-AM and 2µL/mL of

6

ethidium homodimer-1 (Eth-1) from LIVE/DEAD™ Viability/Cytotoxicity Kit, for mammalian

7

cells (Invitrogen, L3224). GFP positive live cells and RFP positive dead cells were visualized

8

with a confocal microscope and fluorescent cells were counted for the quantification of the cell

9

viability.

10 11

2.11 In vivo ischemic hindlimb mouse model

12

Balb/c nude mice (female, 6-weeks) was used for the animal experiments with the protocols

13

approved by the Seoul National University. All the cryogels utilized for in vivo experiments was

14

G1H0.3. In this study, 5 types of experimental groups were developed as follows: sham cryogels

15

(Control), an acellular cryogel (Gel), cryogel with 100ng/mL VEGF (VEGF), cryogel with 1x106

16

NIH-3T3 cells (NIH) and cryogels with 1x106 NIH-3T3 cells and 100ng/mL VEGF

17

(NIH_VEGF). The ischemic hind limb was induced on the left leg excising three vessels

18

followed by the reference.32 After the development of the animal model, cryogels were injected

19

into the surgical region through 17G syringe needle with sterile PBS. Blood perfusion was

20

evaluated using a laser Doppler perfusion imaging (LDPI) system (Moor instrument, Devon,

21

UK) right after surgery and every week for 28days to observe the recovery rate of the blood. The

22

rate of blood flow recovery was expressed as the LDPI ratio of ischemic to non-ischemic hind

23

limb blood perfusion. And a limb necrosis was scored as the following scale: entire recovery

ACS Paragon Plus Environment

10

Page 11 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

(score 6), minor necrosis or nail loss (score 5), partial toe amputation (score 4), total toe

2

amputation (score 3), partial/total foot amputation (score 2), or partial/total limb amputation

3

(score 1).33

4 5

2.12 OCT embedding

6

Muscle tissues or ex vivo cryogels collected after in vivo study were embedded in OCT (Sakura

7

Fineteck) to determine the degree of angiogenesis via histology and immunostaining. Firstly,

8

tissues or gels were fixed in 4% PFA at 4℃ overnight. Then, samples washed twice with PBS

9

and stored in 20 % (w/v) sucrose solution overnight to maintain the structures. In addition,

10

sucrose solution was replaced to OCT/ sucrose=1:1 solution and placed at 4℃ overnight. Finally,

11

the cross-sectioned area of the samples was located in the bottom of a cryomold and filled with

12

OCT. After removing bubbles, the cryomold was frozen in the liquid nitrogen. Constructed

13

cryoblocks were stored at -80℃. For the histology or immunostaining, samples were

14

cryosectioned at 7µm thickness.

15 16

2.13 Histology & immunostaining

17

Cryosectioned samples were washed 3 times with PBS containing 1% (w/v) BSA (BSA/PBS) to

18

remove OCT. For H&E staining, samples were stained with hematoxylin (Sigma-Aldrich,

19

HHS16) for 2min and then, washed in the running water for 5 min. If hematoxylin was

20

overstained, washed with 1% HCl in 70% EtOH for the 30s. Finally, samples were stained with

21

eosin (Sigma-Aldrich, HT110332) for 1min and washed with 95% EtOH 4 times for 1min in

22

series. Stained samples were mounted with Canada balsam (Sigma-Aldrich, C1795) in xylene.

23

For immunofluorescence staining, samples were permeabilized in the blocking solution based on

ACS Paragon Plus Environment

11

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 40

1

PBS containing 1% (w/v) BSA, 10% normal goat serum and 0.1% Triton X-100 (Sigma-Aldrich,

2

X100) for 45min at RT and then, incubated in primary Rat Anti-Mouse CD31 (BD Biosciences,

3

550274) diluted to 1:40 in the blocking solution overnight at 4℃. The samples were washed 3

4

times with BSA/PBS. 1:500 diluted 594-conjugated secondary antibody in BSA/PBS was

5

applied to the samples for 2hrs at RT in the dark. The samples were washed 3 times and stained

6

with 1:200 diluted DAPI for 10min. Fluorescence was visualized by EVOS FL Imaging System

7

AMF4300 (Life Technologies).

8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

ACS Paragon Plus Environment

12

Page 13 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1 2 3 4 5 6

3. Results and discussion

7

3.1

8

Gelatin has sufficient primary amine residues (e.g. arginine, histidine, and lysine) which can be

9

conjugated with carboxyl groups on heparin polysaccharide chain via EDC/NHS reaction,

10

resulting in the formation of amide cryogels by a covalent conjugation.34 We performed the

11

gelatin/heparin cyrogel via mixing the solutions and EDC/NHS crosslinking reactions at -20 ℃.

Fabrication of gelatin/heparin cryogel

12

In the process of cryogelation, macroporous structures within cryogels were formed by

13

ice crystals that were formed. Furthermore, completely frozen cryogels were then lyophilized

14

prior to the cell seeding and cryogel characterizations (Figure. 1A). We confirmed the EDC/NHS

15

reaction during sub-zero degree condition (i.e., formation of amide groups between gelatin and

16

heparin), by FT-IR analysis (Figure. S1). In FT-IR analysis, a single peak of amide A at 3300cm-

17

1

18

1550cm-1 attributed to the bending frequency of N-H conjugated with C-O was observed with

19

amide I peak around 1630cm-1 which also indicated stretching vibrations of the C-O in amide

20

groups. All these peaks indicated successful conjugation of gelatin and heparin forming amide

21

groups via EDC/NHS crosslinking reaction during cryogelation process.

presented the stretching frequency of N-H in amide groups. A strong amide II peak around

22

ACS Paragon Plus Environment

13

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 40

1 2 3 4 5 6 7 8

Figure 1. Fabrication and experimental strategy of gelatin/heparin cryogel. (A) Schematic illustration of gelatin/heparin cryogel developing procedure. Gelatin was crosslinked with heparin under EDC/NHS mediated cryogelation process. (B) Photographs of gelatin/heparin cryogels fabricated with the combination of 1, 1.5% (w/v) gelatin (G1, G1.5) and 0.1, 0.3, 0.5% (w/v) heparin (H0.1, H0.3, H0.5). Lyophilized gelatin/heparin cryogels were fully swelled in PBS for 24hrs (right) and dehydrated on the kimwipes (left).

9

In this study, we fabricated 6 different types of cryogels with varying amount of gelatin

10

and heparin concentration as depicted Table 1. As shown in Figure. 1B, cryogels based on 1%

11

gelatin (G1 cryogels) were more transparent than cryogels based on 1.5% gelatin (G1.5 cryogels)

12

when completely swollen in PBS for 24hrs (right). In addition, G1 cryogels had a skinnier shape

13

when mechanically dehydrated on kimwipe (left) showing a more sponge-like property.

14

Interestingly, cryogels synthesized without heparin displayed a crumbled shape and different

15

property compared to the heparin incorporated cryogels (data was not shown). Although Type A

16

gelatin can be polymerized without heparin, because gelatin has 0.077 ratio of carboxyl groups to

17

amine groups, the crosslinking efficiency is less than 26% in lower EDC, NHS concentration

35,

ACS Paragon Plus Environment

14

Page 15 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

36

2

reaction for gelation of our injectable cryogel system.

3

. As a result, heparin is a key material as it enables to build up carboxyl acids for crosslinking

Table 1. Weight/volume (w/v) percent composition of each gelatin/heparin cryogel.

Sample name Gelatin % (w/v) Heparin % (w/v)

G1 cryogels G1H0.1 G1H0.3 1 1 0.1

0.3

G1H0.5 1

G1.5 cryogels G1.5H0.1 G1.5H0.3 1.5 1.5

G1.5H0.5 1.5

0.5

0.1

0.5

0.3

4 5

3.2

Characterization of gelatin/heparin cryogels

6

To analyze the microarchitecture of gelatin/heparin cryogels, cross-sectional areas of

7

lyophilized cryogels were visualized by scanning electron microscopy (SEM). All cryogels

8

showed monodisperse structures with interconnected pores in a whole area, indicating uniformly

9

constructed network by conjugating heparin to gelatin in the slow crosslinking reaction at a low

10

temperature (Figure. 2A). In addition, the pore area was calculated from SEM images of cross

11

sections of each crygels via ImageJ program following the method from previous paper. 7 There

12

was a trend of smaller average pore area as the centration of gelatin and heparin increased due to

13

the higher degree of crosslinking between gelatin and heparin; however, the difference between

14

each samples were not significant due to the limited varying amount of gelatin and heparin

15

utilized in our study (Figure. S2). As shown in the figure S2, all cryogels had over 2000 um2 of

16

average pores. This is probably due to the similar crygelation conditioned applied for all the

17

samples. For example, even though each cryogel sample had varying amount of gelatin and

18

heparin, they were all crosslinked under the same freezing condition. This condition may have

19

contributed to the similar degree of ice crystal and macropore structure formation.

ACS Paragon Plus Environment

15

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 40

1

Cryogels are interconnected supermacroporous gels prepared at sub-zero temperatures

2

having applications in various research fields. The process of cryogelation is ideally thought to

3

take place via following steps: phase separation with ice-crystal formation, cross-linking and

4

polymerization followed by thawing of ice-crystals to form an interconnected porous cryogel

5

network. Thus, factors such as the concentration of reactants and nonfrozen liquid microphase

6

are considered to be important for cryogelation.37, 38 Among them, the condition of sub-zero

7

temperature is crucial for cryogels since the reaction rate affects the porosity. Lozinsky et al.

8

reported that decrease in temperature lowers the mobility of the solutes which decelerates the

9

reaction by Arrhenius principle.37 Therefore, since the cryogelation proceeds much faster than

10

the usual reaction at elevated temperature, the EDC/NHS chemistry mediated crosslinking of

11

heparin to gelatin was conducted in sub-zero temperature in order to synthesize the injectable

12

cryogel with optimized porosity.

ACS Paragon Plus Environment

16

Page 17 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1 2 3 4 5 6 7 8

Figure 2. Characterization of gelatin-heparin cryogels with various combination of gelatin and heparin concentration. (A) Representative SEM images of microstructure of macroporous cryogels (n=3). (B) Swelling ratio (n=3), (C) Interconnected porosity (n=3) and (D) Young’s modulus (n=3) of cryogels. (E) An enzymatic degradation test in 1 unit/ml or 20 unit/ml collagenase solution (n=3). All groups were applied to 500µL of collagenase freshly replaced every 2-3 days. Error bars indicate SD. Next, we monitored swelling ratio and interconnected porosity of swollen gelatin/heparin

9

cryogels. First, for the swelling ratio, G1 cryogels showed higher swelling ratio than G1.5

10

cryogels except the H0.1 group (Figure. 2B). In addition, G1 cryogels showed higher

ACS Paragon Plus Environment

17

Biomacromolecules 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 40

1

interconnected porosity than G1.5 cryogels (Figure. 2C). Interestingly, as heparin concentration

2

increased, swelling ratio and interconnected porosity increased in G1 cryogels while both

3

properties decreased in G1.5 cryogels. However, there was no significant difference among the

4

cryogels in G1 cryogels. Besides, in the mechanical compression test, G1 cryogels exhibited a

5

lower Young's modulus than G1.5 cryogels did (Figure. 2D). This indicates that G1.5 cryogels

6

are mechanically stronger and able to endure higher strain against the structure deformation than

7

G1 cryogels are.

8

From these investigations, we found that various properties including swelling property,

9

porosity and mechanical property were determined by gelatin concentration, which means the

10

extent of crosslinking was determined by gelatin concentration.39 Due to the lower crosslinking

11

extent, G1 cryogels had a mesh structure with larger pores than G1.5 cryogels, resulting in higher

12

swelling ratio, interconnectivity, and lower modulus. In addition, an increase of the heparin

13

concentration in G1.5 cryogels causes much chemical conjugation of heparin to gelatin, leading

14

to lower swelling ratio, interconnected porosity, and higher Young's modulus. This phenomenon

15

can be explained from the higher extent of crosslinking in H0.5 cryogels which lead to the

16

thicker rigid polymer network that shows lower water uptake than empty pore space.40

17

Additionally, other several factors have contributed to the overall crosslinking density of the

18

cryogel. This chemistry not only contributed to the conjugation of heparin to gelatin, it also has

19

contributed to the intermolecular crosslinking of gelatin which would increase the Young’s

20

modulus. Therefore, we measured not only the physical crosslinking formed by gelatin, but also

21

covalent intermolecular crosslinking formed due to the EDC/NHS mediated intermolecular

22

crosslinking.

ACS Paragon Plus Environment

18

Page 19 of 40 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

1

To observe scaffold stability in enzyme solution, in vitro enzymatic degradation test was

2

performed by incubating cryogels in two different concentrations of collagenase solution. When

3

1 unit/mL collagenase was applied, the weight of G1 cryogels remained over 80% until 36 days

4

and in G1.5 cryogels, over 90% left due to a higher degree of crosslinking of G1.5 cryogels

5

(Figure. 2E). Similarly, in 20 unit/mL collagenase solution, G1 cryogels was slightly more

6

degraded than in 1unit/mL, but still, over 70% remained. This indicates that our gelatin/heparin

7

cryogels are able to maintain their structure with high stability in enzyme solution for a long

8

period against the enzymatic hydrolysis even in high enzyme concentration.

9 10

3.3

Rheological properties of gelatin/heparin cryogels

11

We also characterized our GHs by investigating the rheological properties to confirm the

12

dimensional stability of the inner network. In the rheological study, viscoelastic solids display

13

higher storage modulus (G’) than loss modulus (G’’), while viscoelastic liquids have higher G’’

14

than G’. Therefore, chemically cross-linked cryogels exhibited G’>G’’. In the frequency sweep,

15

G1.5 cryogels showed higher G’ than G1 cryogels, indicating that G1.5 cryogels had higher

16

degree of crosslinking with a stiffer property (Figure. 3A). In addition, as the frequency

17

increases, G’ and G’’ of both G1 and G1.5 cryogels increased as well which indicate the

18

relaxation process of cryogel. In the Figure. 3B, the strain sweep test shows that G1.5 cryogels

19

maintained the stable network against a deformation showing stable G’ and G’’ below 2% of

20

strain, while structural deformation was detected above 30% of strain which is resulted from the

21

crossover of G’ and G’’. In contrast, G1 cryogels preserved the balanced network in the wider

22

range of shear strain (