Engineering Vascularized Skin Tissue in a 3D format supported by Recombinant Spider Silk

(1)

IN

DEGREE PROJECT MEDICAL ENGINEERING, SECOND CYCLE, 30 CREDITS

STOCKHOLM SWEDEN 2020,

Engineering Vascularized Skin

Tissue in a 3D format supported by Recombinant Spider Silk

SAVVINI GKOUMA

KTH ROYAL INSTITUTE OF TECHNOLOGY

SCHOOL OF ENGINEERING SCIENCES IN CHEMISTRY, BIOTECHNOLOGY AND HEALTH

(2)

(3)

Engineering vascularized skin tissue in a 3D format

supported by recombinant spider silk

SAVVINI GKOUMA

Master in Medical Engineering Date: September 23, 2020

Supervisor: Christos Panagiotis Tasiopoulos Examiner: Matilda Larsson

School of Engineering Sciences in Chemistry, Biotechnology and Health

Swedish title:

Vävnadskonstruktion av vaskulariserad hud med hjälp av rekombinant spindelsilke i 3D format

(4)

(5)

iii

Abstract

Skin is an organ with a complex structure which plays a crucial role in the body’s defence against external threats and in maintaining major homeostatic functions. The need for in vitro models that mimic the in vivo milieu is there- fore high and relevant with various applications including, among others, penetration, absorption, and toxicity studies. In this context, the choice of the biomaterial that will provide a 3D scaffold to the cultured cells is defining the model’s success. The FN-4RepCT silk is here suggested as a potent biomaterial for skin tissue engineering applications.

This recombinantly produced spider silk protein (FN-4RepCT), which can self-assemble into fibrils, creates a robust and elastic matrice with high bioactivity, due to its functionalization with the fibronectin derived RGD-containing peptide. Hence it overcomes the drawbacks of other available biomaterials either synthetic or based on animal derived proteins. Additionally, the FN- 4RepCT silk protein can be cast in various 3D formats, two of which are utilized within this project.

We herein present a bilayered skin tissue equivalent supported by the FN- 4RepCT silk. This is constructed by the combination of a foam format, integrated with dermal fibroblasts and endothelial cells, and a membrane format supporting epidermal keratinocytes. As a result, a vascularized dermal layer that contains ECM components (Collagen I, Collagen III, and Elastin) is constructed and attached to an epidermal layer of differentiated keratinocytes.

The protocol presented in this project offers a successful method of evenly in- tegrating cells in the FN-4RepCT silk scaffold, while preserving their ability to resume some of their major in vivo functions like proliferation, ECM se- cretion, construction of vascular networks, and differentiation. The obtained results were evaluated with immunofluorescence stainings of various markers of interest and further analysed, when necessary, with image processing tools.

The results that ensued from the herein presented protocol strongly suggest that the FN-4RepCT silk is a promising biomaterial for skin tissue engineering applications.

(6)

Chapter 1 Aim of the project

This project aims to investigate the potential of FN-4RepCT silk protein as a biomaterial for constructing a bilayered skin equivalent (BSE). The FN- 4RepCT protein was herein used to construct macrosized scaffolds, which were seeded with fibroblasts, endothelial and keratinocyte cells. The skin equivalents were evaluated in terms of cell viability and proliferation, dermal vascularization, ECM secretion, and expression of epidermal markers indicating keratinocyte differentiation.

For this purpose, the dermal part was constructed with a custom developed foaming method that ensures cell survival and integration in the FN-4RepCT fibrillar network, thereby resulting in scaffolds with an even cell distribution.

After 14 days of cultivation, and cell viability monitoring, the epidermal part was constructed as a keratinocyte monolayer over a membrane made of FN- 4RepCT silk. The BSE was constructed by the combination of the dermal and epidermal parts followed by another 14 days of culture at the Air-Liquid interface (ALi), in order to promote keratinocyte differentiation. After a total of 30-32 days, the BSEs were fixed and cryosectioned. Finally, immunofluorescence (IF) stainings were performed on the cryosections to detect the markers of interest. The resulting microscopy images were further processed and evaluated using Matlab .

1

(10)

Chapter 2 Methods

2.1 Cell Culture

Primary Human Dermal Fibroblasts of neonatal origin (HDFn) (Gibco; Thermo Fisher Scientific, Waltham, MA, USA) were cultured in DMEM/F-12 (Dul- becco’s Modified Eagle Medium F-12 Nutrient Mixture (Ham) with L-Glutamine and 15mM HEPES) (Gibco; Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 5% Fetal Bovine Serum (FBS) and 1% Penicillin and Strep- tomycin.

Primary Human Dermal Microvascular Endothelial Cells (HDMEC) (Promo- Cell, Heidelberg, Germany) were cultured in Endothelial Cell Growth Medium MV2, supplemented with fetal calf serum (5%), epidermal growth factor (recombinant human, 5 ng/ml), basic fibroblast growth factor (recombinant human, 10 ng/ml), insulin-like growth factor (Long R3 IGF, 20 ng/ml), vascular endothelial growth factor 165 (recombinant human, 0.5 ng/ml), ascorbic acid (1 µg/ml), and hydrocortisone (0.2 µg/ml), all included in the supplement mix, and 1% Antibiotic-Antimycotic.

Primary Normal Human Epidermal Keratinocytes (NHEK) (Lonza, Basel, Switzerland) were cultured in KBM-Gold^TM, Keratinocyte Cell Basal Medium, supplemented with hydrocortisone (0.50 ml), transferrin (0.50 ml), epinephrine (0.25 ml), GA-1000 (0.50 ml), BPE (2 ml), hEGF (0.50 ml), and insulin (0.50 ml) included in the KBM-GOLD^TMBulletKit^TM(Lonza, Basel, Switzerland).

2

(11)

CHAPTER 2. METHODS 3

HDFn and HDMEC were used between passages 5 and 6, and NHEK were used at passage 2. All cells were cultured in T75 culture flasks (Sarstedt, Nüm- brecht, Germany), in standard incubator conditions (37^oC, 5%CO2 and 95%

humidity) until reaching 80% confluency. At this point, they were enzymati- cally harvested, after 8 minutes of incubation with 1.5 ml of TrypLE (Thermo Fisher Scientific, Waltham, MA, USA), and further used in 3D culture, as described below.

2.2 Construction of the dermal part

The dermal part was constructed by seeding fibroblasts and endothelial cells in scaffolds made by the FN-silk protein in a foam format.

On Day 0 of the experiment, custom made PTFE molds were used to cover the bottom and walls of the wells of a 12-well plate (vwr, Radnor, PA, USA). The HDFn were harvested and FN-silk (3 mg/ml in PBS) was thawed just before use. A cell suspension of 270.000 cells (HDFn) in 1 ml (Table 2.1) was mixed with 2 ml of FN-silk inside a 35 ml beaker. The foam was formed by whipping the solution for 6-8 seconds with a domestic, battery operated, whisk. The freshly formed foam, enough for three constructs, was immediately transferred into the wells of the 12-well plate with a lab spoon. If more constructs were needed, the process was repeated. The newly formed foams were incubated without cell culture medium for 60 minutes in order to become stable, and then 2 ml of DMEM/F-12 was added dropwise into each well.

On Day 2 of the experiment, the culture medium was removed and the foam scaffolds were gently pressed against the bottom of a petri dish with a 1 ml syringe plunger, in order to remove air bubbles that were trapped inside the constructs. Further, each foam was carefully transferred into a 12-well plate transwell (Corning, NY, USA), using a lab spoon (Figure: 2.1.A). 700 µl of MV2 were added inside the transwell and 1300 µl inside the well. (Table 2.1) On Day 3 of the experiment, the HDMEC cells were harvested and prepared for seeding. A stock containing 10⁶cells/100 µl in MV2 medium was created.

The culture medium was removed from both the transwells and the wells, and the HDMEC cells were seeded on top of each foam in two 10 µl drops. 100.000 cells were seeded with the first drop. After 15 minutes of incubation, another 100.000 cells were seeded, and the constructs were incubated for an additional

(12)

4 CHAPTER 2. METHODS

45 minutes. Once the incubation time was completed, any residual cell suspension inside the transwells was removed and MV2 culture medium was added (Table 2.1).

The single layered constructs (SLCs) were then cultured for 14 days, with the culture medium being changed every second day until Day 10 and everyday further on. On Day 14, they were either fixed in 4% PFA, or used to construct the BSEs.

Day Cell type

added Cells per

construct Culture medium Volume

in well Volume in transwell

0 HDFn 90.000 DMEM/F-12 2 ml -

2 - - MV2 2 ml -

3 HDMEC 200.000 MV2 1.3 ml 0.7 ml 14 NHEK 50.000 KBM-Gold^TM

MV2

- 1.5 ml

0.6 ml -

16-18 - -

KBM-Gold^TM(70%) +MV2(30%) +CaCl2(1.2 mM)

2.8 ml ALi

30-32 Construct fixation & harvest

Table 2.1: Summary of key time points for the construction of BSEs.

2.3 Construction of the BSE

The BSE is a combination of the previously described dermal part with an epidermal layer consisting of NHEK supported by a FN-4RepCT silk membrane.

For the construction of the epidermal layer, silk membranes were prepared as described in Gustafsson et al. (2020). The “air side” of each membrane was seeded with 150.000 NHEK (Table 2.1) and cultured in submerged conditions (i.e. 700 µl of KBM-Gold^TMinto the plate well and 300 µl into the insert) for 3 days. ^∗

On Day 14, the insert carrying the epidermal layer was carefully placed over the SLCs, 600 µl of KBM-Gold ^TM was added inside the membrane insert,

∗The FN-silk membranes were prepared by Christos Tasiopoulos.

(13)

and 1500 µl of MV2 inside the plate well. Depending on the NHEK level of confluency, the BSEs were kept in submerged conditions for 2-4 days (Figure 2.1.B). On Days 16-18, the BSEs were lifted to the ALi in order to promote keratinocyte differentiation. For this purpose, they were transferred into a 6- well plate, whereby 2.8 ml (70% KBM-Gold^TM - 30% MV2, supplemented with 1.2 mM CaCl2to promote the differentiation process^[1]) of fresh medium were added inside the plate well, and the epidermal layer was left exposed to the air (Table 2.1). Custom made holders were utilized to keep the constructs elevated as depicted in Figure 2.1.C. After 14 days of culture at the ALi, the constructs were fixed in 4% PFA and prepared for cryosectioning.

Figure 2.1: Schematically displaying the steps for constructing the BSE. A: The SLC inside the transwell after the air bubble removal. B: The BSE in submerged conditions (i.e. growth medium inside the well and inside the insert). The silk membrane is adhered underneath the insert sealing its outer walls. C: The BSE at the ALi (i.e. growth medium only inside the well and in contact with the bottom of the insert).

2.4 Cell viability assay

Cell viability was monitored, at 7 time points during the culture, using the alamarBlue^TM viability assay. The alamarBlue^TM Cell Viability Reagent (In- vitrogen, Carlsbad, CA, USA) was diluted (1:10) in the respective prewarmed growth medium depending on the exact culture time point (Table: 2.1). The culture medium was removed from the constructs and a total of 1000 µl of fresh solution was added per well. 500 µl was added inside the transwell and the remaining 500 µl was added in the 12-well plate well. 1000 µl of the solution was added in an empty well to serve as blank. After 2 hours of incubation (37^oC, 5%CO2, and 95% humidity), aliquots of 100 µl (n=4) were transferred to a 96-well plate (Greiner Bio-One GmbH, Kremsmünster Austria). Fresh growth medium was added to the culture plate after the removal of any resid- uals of diluted alamarBlue^TM.

(14)

The fluorescence intensity of the samples was measured in CLARIOstar plate reader (BMG LABTECH, Ortenberg, Germany) at room temperature with ex- citation at 544 nm and emission at 595 nm. The blank was deducted from each construct’s fluorescence intensity measurement and replicates’ mean value and standard deviation were calculated before plotting against time.

2.5 Cryosectioning

The constructs were fixed in 4% PFA for 10 minutes, washed in PBS twice, and embedded in OCT embedding matrix for frozen sections (CellPath, New- town, UK). They were allowed to freeze overnight at -20^oC and stored until cryosectioning.

On the day of cryosectioning, the constructs were cut in half along their diameter, resulting in two full-thickness pieces, and mounted together using OCT. The constructs were sectioned in 12 µm thick sections using a cryo- stat (Cryostar NX70, Thermo Fisher Scientific, Waltham, MA, USA). Sec- tions were obtained on Thermo Scientific^® SuperFrost Plus^® (Fisher Scien- tific, Hampton, NH, USA) adhesion slides and stored at room temperature until further use.

2.6 Immunofluorescence

Indirect immunofluorescence (IF) was performed on the cryosections in order to confirm the presence of the antigens of interest in the dermal and epidermal parts. The Sudan Black B (SB) (Sigma-Aldrich, St. Louis, MS, USA) reagent dissolved (0.3% w/v) in 70% ethanol was used to decrease background noise which is generated by the silk’s autofluorescence.^[2],[3]

Sections were hydrated for 10 minutes in PBS, incubated with SB for 15 minutes, and then rinsed with PBS until the SB was no longer visible. The cells were permeabilized for 10 minutes with 0.2% Triton X-100 in PBS and then, the sections were washed with 0.2% TWEEN in PBS two times, for 5 minutes each. The tissue sections were blocked for 1 hour using normal Goat Serum (GS) in PBS with 0.2% TWEEN. After blocking, the primary antibody of interest (Table 2.2) was added on each section, diluted in 1% GS in PBS with 0.2% TWEEN, and allowed to incubate overnight at +4^oC inside a box that

(15)

provides humidity.

The next day, the slides were placed inside a slide staining jar and washed, while being agitated, with 0.2% TWEEN in PBS. Further on, the secondary antibody (Table 2.3) diluted in 1% GS in PBS with 0.2% TWEEN was added.

It was allowed to incubate for 2 hours, at room temperature, inside a box that provides darkness and humidity. Once the incubation was completed, the sections were incubated in SB for 10 minutes before nuclei staining with DAPI diluted (1:800) in PBS for 10 minutes. Finally, the sections were washed twice for 5 minutes each time in 0.2% TWEEN in PBS, with agitation, before being mounted in #1.5 thickness cover slips (DURAN^®) using DAKO fluorescence mounting medium. The immunostained sections were stored in darkness at +4^oC until visualization using an inverted fluorescence microscope (Nikon Eclipse Ti) and the NIS-Elements BR software.

Primary Antibodies Species Dilution Manufacturer

CD31 mouse anti-human 1:200 Acris GmbH

Von Willebrand factor (vwf) mouse anti-human 1:200 Invitrogen Collagen type I (Col I) rabbit anti-human 1:500 Abcam Collagen type III (Col III) rabbit anti-human 1:500 Sigma Aldrich

Elastin guinea pig anti-human 1:200 Abcam

Keratin 5 (K5) rabbit anti-human 1:1000 BioLegend Keratin 10 (K10) rabbit anti-human 1:500 BioLegend Involucrin (INV) mouse anti-human 1:200 Sigma Aldrich

Table 2.2: Summary of the primary antibodies used for the IF stainings.

Secondary Antibodies Species Dilution Manufacturer AlexaFLuor 488 goat anti-mouse 1:500 Invitrogen AlexaFLuor 488 goat anti-rabbit 1:500 Invitrogen AlexaFLuor 488 goat anti-guinea pig 1:500 Invitrogen

Table 2.3: Summary of the secondary antibodies used for the IF stainings.

2.7 Image Analysis

The diameter of the lumen-like structures and the length of the sprout-like structures detected in the IF images of the dermal layers was measured using Matlab R2019a.

(16)

The lumens were detected in the IF images as green rings and the Flood Fill tool (Image Segmenter app) was used to manually create a mask of the ellipsoids that are defined by the rings’ inner diameter. The images containing sprout-like structures, were first filtered to remove the background (using im- fill(), bwareaopen(), and imbothat() functions) and then converted to the L*a*b colorspace in order to exploit the presence of the green and blue color layers.

The green layer was further separated according to the brightness values. This way, the highly bright, positively detected signal was isolated from the less bright signal originating from the silk’s autofluorescence. Finally, a mask of the CD31-positive areas (i.e. bright green signal) was created.

The regionprops() function was used to explore the properties of the regions that emerged from the masks described above. For the lumen structures, the ellipsoids were identified, and the major and minor axis values were returned.

In the case of the sprout constructs, the centroids of the masked regions were first determined, the smoothing spline model was used for the curve fitting, and the sprout length was calculated as the total arc length of the fitted line.

(17)

Chapter 3 Results

The skin equivalents, constructed with this experimental setup, are stable macrosized structures with fibroblasts and endothelial cells integrated in the FN- 4RepCT scaffold. The seeding density and methods were chosen in order to facilitate cell survival, homogeneous cell distribution in the 3D construct and adequate cell density throughout the culture (Figure 3.1). Additionally, the BSEs contain an epidermal part, consisting of multiple keratinocyte layers, which is strongly fused with the dermal one (Figure 3.2)

Figure 3.1: IF and brightfield images of the SLC. A: At Day 14, the endothelial cells (vWF in bright green), are homogeneously distributed inside the construct. B: Day 28 after two weeks at the ALi. The cells have populated the entire scaffold area and their density has increased.

C: Brightfield image of the SLC. Cell nuclei in blue (DAPI) in A & B. Weak autofluorescence originating from the silk scaffold is visible in the green and blue spectrum. Scale bars: 500 µm.

9

(18)

10 CHAPTER 3. RESULTS

Figure 3.2: IF and brightfield images of the BSE (Day 30). A: Brightfield image of the epidermal part. B: IF image of a BSE displaying the multiple NHEK layers of the epidermal part. C: Close brightfield image displaying the fusion of the dermal and epidermal parts of a BSE. The dashed lines mark the dermal-epidermal interface. Cell nuclei in blue (DAPI) in B.

Scale bars: 500 µm (A), 100 µm (B & C).

3.1 Construct morphology

Directly after the whipping process, the foams are fragile and tend to adhere to the culture plate walls. This adhesion, which could be damaging to the constructs, was prevented by the use of the PTFE molds. After 1 hour of incubation, the foams appeared as less compact constructs covering the surfaces of the culture well and containing multiple air cavities (Figure 3.3.A). Their shape started being formed and their stability was not affected by the addition of the cell culture medium. The following day (Day 1), the resulted foam constructs were not attached to the side walls of the wells and their respective size was smaller, indicating that the foams became denser. Multiple air bubbles were also visible inside and under the constructs (Figure 3.3.B). On

(19)

CHAPTER 3. RESULTS 11

Day 2, they were stable enough to withstand small mechanical forces, therefore the air bubbles were removed, in order to provide stability to the scaffolds and allow a more accurate representation of the in vivo conditions. The con- structs without the air cavities were flatter, more compact, and could remain submerged in the culture medium. In contrast, foam constructs were shown floating in cell growth medium on Days 0 and 1 (Figure 3.3.C). The cavities that initially appeared were minimized as the culture progressed, resulting in elastic, homogeneous constructs, which had assumed the circular shape of the transwell on Day 14 (Figure 3.3.D). Past this time point, the morphological changes of the SLCs were less noticeable, their elasticity did not decrease, and the two week culture at the ALi did not have any negative effect on their morphology (Figure 3.3.E). On the other hand, the two-week culture of the BSEs at the ALi, resulted in the constructs loosing some of their elasticity and becoming considerably more compact and stable, which is consistent with the role of the epidermis in the in vivo conditions (Figure 3.3.F).

Figure 3.3: Photographs of constructs at key time points of culture. A: Day 0, SLC after 1 hour of incubation. B: Day 1, SLC. C: Day 2, SLC after the removal of air bubbles. D: Day 14, SLC. E: Day 28, SLC after two weeks of culture at the ALi. F: Day 35, BSE after two weeks at ALi. An example of a foam in a well of a 12-well plate (A & B), inside a 12-well plate insert (D & E), and in a well of 24-well plate (C & F). Scale bars: 10 mm.

(20)

3.2 Cell viability assay

The alamarBlue^TMassay is a simple and fast method to monitor cell viability, without affecting the culture. In the presence of metabolically active cells, the reagent, diluted in the culture medium, becomes highly fluorescent.

Figure 3.4 captures the increase in the cells’ metabolic activity throughout the complete culture period. The metabolic activity is steeply increasing until Day 10, indicating that the cells survive the seeding process, proliferate, and are highly viable. Following this time point, the metabolic activity stabilizes and remains approximately at the same levels until the end of culture, indicating that the cells have successfully covered the available space. This provides an indication that the cells survive throughout the culture, remain metabolically active, and are not in distress because of the culture conditions (i.e. ALi cultivation).

Figure 3.4: Cell metabolic activity in BSEs and SLCs. BSEs (n=5) were cultured in sub- merged conditions from Day 0-13 and at the ALi from Day 14-28. SLCs were either cultured in submerged conditions until Day 28 (n=2) or at the ALi (n=2) from Day 14-28. The dashed lines represent culture at the ALi.

(21)

3.3 Dermal layer: ECM components

Positively detected markers for ECM production in the dermal part indicate

Figure 3.5: IF images of SLCs on Day 28, after 2 weeks at the ALi, stained for Collagen type I (A), Elastin (B), and Collagen type III (C) (green). D-F: The respective controls wherein the primary antibody was omitted. Cell nuclei in blue (DAPI). Scale bars: 250 µm.

(22)

that 14 days of submerged culture are adequate for the HDFn to produce major dermal ECM components (i.e. Collagen I, Elastin, and Collagen III); with elastin being highly abundant (Appendix Figure B.1). After two additional

Figure 3.6: IF images of BSEs on Day 32 (2 weeks at ALi) stained for Collagen type I (A), Elastin (B), and Collagen type III (C) (green). D-F: The respective controls wherein the primary antibody was omitted. Cell nuclei in blue (DAPI). Scale bars: 250 µm.

(23)

weeks of culture at the ALi, the SLCs appear to be more densely populated by cells and all the ECM markers give a brighter signal, suggesting increased quantities (Figure 3.5).

The same markers are also detected at the dermal part of the BSEs after 30-32 days of culture. Elastin appears to have been secreted at the same extent as in the SLC after two weeks at the ALi. Collagen I and III are also detected but in lesser amount (Figure 3.6).

3.4 Dermal layer: Vascularization

By Day 14, the endothelial cells have proliferated and migrated from the upper part of the foam, where they were seeded, to the entire scaffold area (Figure 3.1.A). At this time point, they have already formed sprout-like structures,

Figure 3.7: IF images of sprout-like vascular structures stained for CD31 (green). A-B:BSEs, on Day 30. C: SLC, on Day 28. Cell nuclei in blue (DAPI). Scale bars: 50 µm

(24)

which can be identified by the elongated nuclei and the directionality of positively detected transmembrane markers (i.e. CD31 and vWF) in one continuous line (Appendix Figure B.2).

Figure 3.8: IF images of lumen-like vascular structures. A & B: SLCs, on Day 28, stained for CD31 (green). C: BSE, on Day 32, stained for vWF (green). Cell nuclei in blue (DAPI).

Scale bars: 20 µm.

Figure 3.7 displays the sprout-like structures detected on Day 28 in a SLC (C) and on Day 30 at the dermal part of a BSE (A & B). The number of the sprouts, detected in the constructs, is higher compared to Day 14, and their length was estimated and found to be 200 µm ± 96 µm (mean ± std) (Appendix Table B.1). Additionally, lumen-like structures were also detected (Figure 3.8). They are brightly green, ring shaped formations connecting 2-4

Figure 3.9: Figures displaying the dimensions of a representative selection of capillary-like structures. Left: The diameters of the lumen-like structures detected in BSEs and SLCs (Fig- ure 3.8 and Appendix Figure B.3). Right: The lengths of the sprout-like structures detected in SLCs and BSEs (Figure 3.7 and Appendix Figure B.4).

(25)

nuclei, found in both the SLCs (A & B) and the BSEs (C). Their inner diameters were measured to be 22 µm, 12 µm ± 9 µm, 7 µm ([major axis, minor axis] (mean ± std)) (Figure 3.9 and Appendix Table B.2).

3.5 Epidermal layer: Keratinocytes’ differen- tiation

By Day 30, the epidermal layer consisted of multiple layers of proliferating and differentiating keratinocytes that have originated from the keratinocyte monolayer.

Keratin 5, an early differentiation marker of proliferating keratinocytes, was detected in the entire thickness of the epidermal part (Figure 3.10.A). In contrast, Keratin 10 expression was detected only at the upper areas of the epidermal layer and did not co-localize with neither of the other two markers (Figure 3.10.B). Involucrin, an intermediate differentiation marker of the suprabasal keratinocytes, was detected in middle areas of the epidermal layer (Figure 3.10.C).

Figure 3.10: IF images of the epidermal layer, in consecutive sections of BSEs on Day 30, stained for the keratinocytes’ differentiation markers. A: Keratin 5 (green). B: Keratin 10 (green). C: Involucrin (green). D-F: Respective controls for A-C wherein the primary anti- body was omitted. The dashed lines mark the limits of the epidermal layer. Cell nuclei in blue (DAPI). Scale bars: 100 µm.

(26)

Chapter 4 Discussion

This project explores the use of the FN-4RepCT silk protein as a matrice for 3D culture of human primary skin cells (i.e. dermal fibroblasts, dermal endothelial cells, and epidermal keratinocytes) and investigates its potential in constructing a tissue engineered BSE. For this purpose the silk protein was used in two formats: as a macrosized 3D foam for the dermal layer and as a membrane for the epidermal.

The developed protocol aims in achieving an optimal integration of the cells inside the dermal layer in terms of cell adhesion, survival, migration, and even distribution. The intrinsic properties of the FN-4RepCT silk facilitate the engineering of an elastic, yet robust matrix, which promotes cell adhesion, proliferation, and migration. The whipping process successfully integrates the fibroblasts inside the scaffold and is shown to be gentle enough for maintained viability and function. They are able to adhere to the biomaterial, within 1 hour, proliferate, and evenly populate the construct, within the course of 4 weeks.

Moreover, as confirmed by the IF images, the fibroblasts are able to resume functions they have in their in vivo environment and similarly to other in vitro models^[4],[5], secrete Collagen types I & III, and elastin. These dermal ECM components offer structural support to the cells and contribute to signaling with regards to major functions.^[6],[7] All the ECM components appear to be distributed in the entire thickness of the SLCs, as opposed to the BSEs where Col I is mainly detected at the lower areas of the dermal part. One interpreta-

18

(27)

CHAPTER 4. DISCUSSION 19

tion of this could be that the paracrine signaling between the keratinocytes and the fibroblasts affects the Col I secretion, resulting in denser and larger Col I fibers in the lower dermal areas (i.e. reticular dermis) of the BSEs similarly to the in vivo tissue.^[6],[7]However, further investigation of this claim is required.

The BSEs are constructed by combining the dermal and epidermal parts similarly to the tissue anatomy. The silk membrane, much like the basal lamina, acts as a permeable barrier separating the two layers, while promoting effi- cient attachment of the keratinocytes and enabling the fibroblast-keratinocyte crosstalk. Thus, this method provides a clear dermal-epidermal separation and facilitates the keratinocytes’ differentiation.

A first indication of the epidermal differentiation is the final thickness of the epidermal layer, which is in the range of 100 µm, compared to the initial membrane thickness, which is less than 1 µm^[8]. This is indicative of the keratinocytes making progress in the process of upward stratification. Fur- thermore, the IF stainings demonstrate that the keratinocytes’ differentiation corresponds to that of the skin tissue and other in vitro epidermal models^[4],[5]. The lack of K5 marker and the detection of K10 in the upper areas indi- cates that these cells have interrupted their mitotic cycle and have commit- ted to differentiation.^[9],[10] This is also supported by the localization of INV, a broader, intermediate, suprabasal marker detected in cells before the onset of envelop cross-linking.^[9],[10] Indeed, in the constructed BSEs, INV is not detected in the first, closest to the dermal-epidermal interface, layer of keratinocytes, corresponding to the basal layer, rather further up and just below the K10-positive area. Another indication suggesting the keratinocytes have progressed through the stages of differentiation is the detected nuclei in the upper areas of the epidermal layer. They are flattened and decreased in the K10-positive area, a finding consistent with the keratinocytes gradually los- ing their nuclei and organelles in the granular layer, while their keratin is not affected^[6],[7]. They are also not detectable at the outermost epidermal area.

Even though there are commercially available BSEs^[11],[12]with differentiated epidermal layers, vascularization remains a major challenge for tissue engineered skin models. This protocol provides a method for constructing a vascularized BSE containing only primary dermal cells, without the need for xenografting as in previously developed models.^[13][14]Exploiting the fast adhesion and migration properties of cells to the FN-4RepCT silk[15],[16][17]

, the HDMEC were seeded on top of the SLCs in dense cell suspensions and evenly

(28)

20 CHAPTER 4. DISCUSSION

distributed in the entire dermal layer. After 11 Days, as confirmed by the IF images, they have already migrated in clusters and formed sprout-like vascular structures. Following two additional weeks at the ALi, multiple sprout- and lumen-like vascular formations, consistent with other available models[17],[14],[18], were detected in both the SLCs and the BSEs indicating that the the long term culture at the ALi does not impede vascularization. Although further opti- mizations may be needed, such a prevascularized BSE containing lumen formations, in the range of dermal papillary arterioles^[19], could be potentially utilized in various applications where tissue vascularization is important.

Results presented herein hold great promise to tissue engineer a physiologically relevant skin model supported by the FN-silk, however, further investigation of some aspects is required. Firstly, it is yet to be explored whether the localization of the ECM components in the dermal part can be consistent with that of the in vivo tissue and more specifically, if a papillary-reticular dermis distinction can be achieved. The physiologic epidermal differentiation, although achieved to a big extent, also has to be further explored. The fact that the uppermost epidermal area is not positive for any of the tested keratinocyte markers suggests that more markers should be tested, in order to conclude about the stage of the keratinocyte differentiation. Further, the differences observed in the epidermal differentiation between cryosections coming from areas close to the constructs’ center (Figure 3.10) and those that are closer to the edges (Appendix Figure B.8) could be interpreted as uneven coverage of the silk membrane during cell seeding. As a result, keratinocytes may not differentiate simultaneously and homogeneously throughout the entire surface area. This could explain the formations of undefined shapes that were detected in the upper epidermal areas of cryosections obtained further away from the central point where keratinocytes were initially seeded (Appendix Figure B.8).

This problem can potentially be resolved by changing the keratinocytes’ seeding method. Lastly, the signal detected in the upper epidermal region when staining for the K5 marker is not consistent with proliferating keratinocytes, due to the complete absence of nuclei, and could potentially be attributed to non specific binding of the antibody.

The macrosized FN-4RepCT foams presented herein combine the traits of previously developed FN-silk matrices,^[16],[17]in terms of cell adhesion and proliferation, as well as ECM secretion and vascularization. Thus, there is grow- ing evidence that the silk protein in foam and membrane formats is suitable for the developing of skin tissue models. Moreover, the vascularized BSEs,

(29)

CHAPTER 4. DISCUSSION 21

constructed by combining the foam and membrane matrices, are consistent with other bilayered skin models.[11],[12],[13],[14]

It is therefore shown that the FN-4RepCT silk protein is a potent biomaterial for skin tissue engineering.

The herein presented BSEs are not limited from drawbacks of animal derived matrices and are mechanically robust and highly bioactive. Furthermore, the availability of the recombinantly produced FN-silk in high amounts and the fast whipping technique that does not require complex lab equipment make their construction simple and repeatable. The macrosized BSE supported by the FN-4RepCT silk protein is therefore proposed as potent human skin in vitro model, with potential in applications like vascular supply perfusion sys- tems and cancer models.

(30)

Chapter 5 Conclusion

This project investigates the engineering of a physiologically relevant skin tissue model using a macrosized foam format supported by FN-4RepCT silk protein. The skin equivalents are constructed as a combination of the dermal and epidermal parts. The dermal layer, integrated with dermal fibroblasts and endothelial cells, contains ECM components (i.e. Col I, Col III, and Elastin) and vascular formations. The epidermal part has multiple layers of epidermal keratinocytes at different stages of the differentiation process. The dermal- epidermal interface is facilitated by a thin, silk membrane which enables attachment and nutrients diffusion. The herein constructed BSEs show promise to be used as in vitro models for skin research.

22

(31)

Acknowledgements

This thesis project was conducted in the Hedhammar Lab, at the Division of Protein Technology, KTH and supervised by Christos Tasiopoulos. I owe thanks to My Hedhammar for the opportunity to become a member of the Silk Group, work with the FN-4RepCT, and for all the means that were available to me. I would like to thank my supervisor Christos Tasiopoulos for supporting me throughout every single step of this project and for always being eager to share knowledge and answer my questions. Special thanks to Mona Widhe for always being willing to discuss the new outcomes, help overcome the difficul- ties that arose, and for offering valuable input regarding the project’s progress.

I would also like to thank all the members of the Silk Group for the warm welcome to the team and the friendly environment they created.

Finally, special thanks to the Spiber Technologies production team for provid- ing the FN-4RepCT protein.

23

(32)

24 CHAPTER 5. CONCLUSION

Abbreviations

BSE: Bilayered Skin Equivalent Col I: Collagen type I

Col III: Collagen type III ECM: Extracellular Matrix

FN-4RepCT: Functionalized silk protein with four repetitive units followed by a C-terminal

FN-silk: Abbreviation for FN-4RepCT ftSE: Full Thickness Skin Equivalent HA: Hyaluronic Acid

HDFn: Human Dermal Fibroblasts, neonatal

HDMEC: Human Dermal Microvascular Endothelial Cells HUVEC: Human Umbilical Vein Endothelial Cells

INV: Involucrin K5: Keratin 5 K10: Keratin 10

NHEK: Normal Human Epidermal Keratinocytes SLC: Single Layered Construct

PFA: Paraformaldehyde UV: Ultraviolet

(33)

Bibliography

[1] Daniel D Bikle, Zhongjian Xie, and Chia-Ling Tu. “Calcium regulation of keratinocyte differentiation”. In: Expert review of endocrinology &

metabolism 7.4 (2012), pp. 461–472.

[2] Lin Qi et al. “Pre-culture Sudan Black B treatment suppresses autoflu- orescence signals emitted from polymer tissue scaffolds”. In: Sci Rep 7.1 (2017).

[3] Puay Yong Neo et al. “Enhancing analysis of cells and proteins by fluorescence imaging on silk-based biomaterials: modulating the autofluo- rescence of silk”. In: Tissue engineering. Part C, Methods 21.2 (2015), pp. 218–228.

[4] Lydia Costello et al. “Engineering a Multilayered Skin Equivalent: The Importance of Endogenous Extracellular Matrix Maturation to Provide Robustness and Reproducibility”. In: Skin Tissue Engineering: Methods and Protocols. Ed. by Sophie Böttcher-Haberzeth and Thomas Bieder- mann. New York, NY: Springer New York, 2019, pp. 107–122. isbn:

978-1-4939-9473-1. doi: 10 . 1007 / 978 - 1 - 4939 - 9473 - 1 _ 9.

url: https : / / doi . org / 10 . 1007 / 978 - 1 - 4939 - 9473 - 1_9.

[5] Mathilde Roger et al. “Bioengineering the microanatomy of human skin”.

In: Journal of Anatomy 234.4 (2019), pp. 438–455. doi: 10.1111/

joa.12942. eprint: https://onlinelibrary.wiley.com/

doi/pdf/10.1111/joa.12942. url: https://onlinelibrary.

wiley.com/doi/abs/10.1111/joa.12942.

[6] Justine Fenner and Richard A.F. Clark. “Anatomy, Physiology, Histol- ogy, and Immunohistochemistry of Human Skin”. In: Skin Tissue En- gineering and Regenerative Medicine. Elsevier Inc., Jan. 2016, pp. 1–

25

(34)

26 BIBLIOGRAPHY

17. isbn: 9780128016541. doi: 10.1016/B978-0-12-801654- 1.00001-2.

[7] A. J. Paul Kolarsick, Maria Ann Kolarsick, and Carolyn Goodwin. “Anatomy and Physiology of the Skin”. In: Journal of the Dermatology Nurses’

Association 3 (2011), pp. 203–213. doi: 10.1097/JDN.0b013e3182274a98.

[8] Linnea Gustafsson et al. “Recombinant Spider Silk Forms Tough and Elastic Nanomembranes that are Protein-Permeable and Support Cell Attachment and Growth”. In: Advanced Functional Materials (2020), p. 2002982. doi: 10.1002/adfm.202002982.

[9] Patricia Rousselle, Edgar Gentilhomme, and Yves Neveux. “Markers of Epidermal Proliferation and Differentiation”. In: Apr. 2017, pp. 407–

415. isbn: 978-3-319-32381-7. doi: 10.1007/978-3-319-32383- 1_37.

[10] Watt F. M. “Involucrin and other markers of keratinocyte terminal dif- ferentiation”. In: The Journal of investigative dermatology 81.1 Suppl (1983), 100s–3s.

[11] Reconstructed Tissues. url: https : / / www . phenion . com / products / reconstructed - tissues % 7B % 5C # %7DTab - 754224%7B%5C_%7D2 (visited on 03/01/2020).

[12] EpiDermFT in vitro 3D Tissue |. url: https : / / www . mattek . com/products/epidermft/ (visited on 03/01/2020).

[13] Florian Groeber et al. “A first vascularized skin equivalent as an alter- native to animal experimentation”. In: Altex 33.4 (Nov. 2016), pp. 415–

422. issn: 18688551. doi: 10.14573/altex.1604041.

[14] Annie F. Black et al. “In vitro reconstruction of a human capillary-like network in a tissue-engineered skin equivalent”. In: The FASEB Journal 12.13 (Oct. 1998), pp. 1331–1340. issn: 0892-6638. doi: 10.1096/

fasebj.12.13.1331.

[15] Mona Widhe, Nancy Dekki Shalaly, and My Hedhammar. “A fibronectin mimetic motif improves integrin mediated cell biding to recombinant spider silk matrices”. In: Biomaterials 74 (Jan. 2016), pp. 256–266.

issn: 18785905. doi: 10.1016/j.biomaterials.2015.10.

013.

[16] Mona Widhe et al. “Recombinant spider silk as matrices for cell cul- ture”. In: Biomaterials 31.36 (Dec. 2010), pp. 9575–9585. issn: 01429612.

doi: 10.1016/j.biomaterials.2010.08.061.

(35)

BIBLIOGRAPHY 27

[17] Ulrika Johansson et al. “Assembly of functionalized silk together with cells to obtain proliferative 3D cultures integrated in a network of ECM- like microfibers”. In: Scientific Reports 9.1 (Dec. 2019), pp. 1–13. issn:

20452322. doi: 10.1038/s41598-019-42541-y.

[18] Yuan Liu et al. “In Vitro Construction of Scaffold-Free Bilayered Tissue- Engineered Skin Containing Capillary Networks”. In: BioMed Research International 2013 (), pp. 1–8.

[19] Irwin M.Braverman. “The Cutaneous Microcirculation”. In: Journal of Investigative Dermatology Symposium Proceedings 5.1 (Dec. 2000), pp. 3–9.

[20] Joey E. Lai-Cheong and John A. McGrath. Structure and function of skin, hair and nails. 2013. doi: 10.1016/j.mpmed.2013.04.

017.

[21] Sylvia S. Mader and Patrick. Galliart. “Understanding human anatomy and physiology”. In: 5th. McGraw-Hill Higher Education, 2005. Chap. 5, p. 444. isbn: 0072464372.

[22] C. M. Chuong et al. What is the ’true’ function of skin? Apr. 2002. doi:

10 . 1034 / j . 1600 - 0625 . 2002 . 00112 . x. url: http : / / doi.wiley.com/10.1034/j.1600-0625.2002.00112.x.

[23] Carla A. Brohem et al. “Artificial skin in perspective: concepts and ap- plications”. In: Pigment Cell & Melanoma Research 24.1 (Feb. 2011), pp. 35–50. issn: 17551471. doi: 10.1111/j.1755-148X.2010.

00786 . x. url: http : / / doi . wiley . com / 10 . 1111 / j . 1755-148X.2010.00786.x.

[24] Biana Godin and Elka Touitou. Transdermal skin delivery: Predictions for humans from in vivo, ex vivo and animal models. Sept. 2007. doi:

10.1016/j.addr.2007.07.004.

[25] Ban on animal testing | Internal Market, Industry, Entrepreneurship and SMEs. url: https://ec.europa.eu/growth/sectors/

cosmetics / animal - testing % 7B % 5C _ %7Den (visited on 03/02/2020).

[26] Atul Chaudhari et al. “Future Prospects for Scaffolding Methods and Biomaterials in Skin Tissue Engineering: A Review”. In: International Journal of Molecular Sciences 17.12 (Nov. 2016), p. 1974. issn: 1422- 0067. doi: 10.3390/ijms17121974. url: http://www.mdpi.

com/1422-0067/17/12/1974.

(36)

28 BIBLIOGRAPHY

[27] Florian Groeber et al. “Skin tissue engineering - In vivo and in vitro applications”. In: Advanced Drug Delivery Reviews 63.4 (Apr. 2011), pp. 352–366. issn: 0169409X. doi: 10.1016/j.addr.2011.01.

005.

[28] Francesco Urciuolo et al. “Bioengineered Skin Substitutes: The Role of Extracellular Matrix and Vascularization in the Healing of Deep Wounds”.

In: Journal of Clinical Medicine 8.12 (Dec. 2019), p. 2083. issn: 2077- 0383. doi: 10.3390/jcm8122083. url: https://www.mdpi.

com/2077-0383/8/12/2083.

[29] Ahmad Sukari Halim et al. “Biologic and synthetic skin substitutes: An overview”. In: Indian J Plast Surg Supplement 1 43 (2010), S23–S28.

doi: 10.4103/0970-0358.70712.

[30] BIOBRANE biosynthetic skin dressing Products from Smith & Nephew | Smith+Nephew - Corporate. url: https://www.smith-nephew.

com / key - products / advanced - wound - management / other-wound-care-products/biobrane/ (visited on 03/01/2020).

[31] Hyalomatrix Wound Device | Medline Industries, Inc. url: https : / / www . medline . com / product / Hyalomatrix - Wound - Device/Z05-PF96940 (visited on 03/01/2020).

[32] OLYMPUS TERUMO BIOMATERIALS CORP. : Artificial Bone Re- placement Material : TERUDERMIS : Product information. url: https:

//www.biomaterial.co.jp/en/products/terudermis/

product/ (visited on 03/01/2020).

[33] AlloDerm RTM - BioHorizons. url: https://www.biohorizons.

com/alloderm.aspx (visited on 03/01/2020).

[34] Gøril Eide Flaten et al. In vitro skin models as a tool in optimization of drug formulation. May 2015. doi: 10.1016/j.ejps.2015.02.

018.

[35] Vesselina Laubach et al. “Integration of Langerhans-like cells into a human skin equivalent”. In: Archives of Dermatological Research 303.2 (Mar. 2011), pp. 135–139. issn: 03403696. doi: 10.1007/s00403- 010-1092-x.

[36] Christine Duval et al. “Keratinocytes control the pheo/eumelanin ra- tio in cultured normal human melanocytes”. In: Pigment Cell Research 15.6 (Dec. 2002), pp. 440–446. issn: 08935785. doi: 10.1034/j.

1600 - 0749 . 2002 . 02055 . x. url: http : / / doi . wiley . com/10.1034/j.1600-0749.2002.02055.x.

(37)

BIBLIOGRAPHY 29

[37] Stephanie H. Mathes, Heinz Ruffner, and Ursula Graf-Hausner. The use of skin models in drug development. Apr. 2014. doi: 10.1016/j.

addr.2013.12.006.

[38] My Hedhammar et al. “Structural properties of recombinant nonrepet- itive and repetitive parts of major ampullate spidroin 1 from Eupros- thenops australis: Implications for fiber formation”. In: Biochemistry 47.11 (Mar. 2008), pp. 3407–3417. issn: 00062960. doi: 10.1021/

bi702432y.

(38)

Appendix A

Theoretical Background

A.1 The human skin

A.1.1 Introduction

Skin is the largest organ of the human body. It covers its external surface and is continuous with the mucous membranes, lining the digestive, respiratory and urogenital tracts. An adult human’s skin has a surface area of approximately 1.8m² and accounts for 15% of the total body weight, about 4kg. It is structurally divided in three main layers (Figure A.1): the outermost is the epidermis, followed by the dermis and the innermost hypodermis or subcutis.[6],[7],[20]

The skin, which is also called cutaneous membrane or the integument, is a self regenerating tissue that serves a wide range of purposes, contributing to the body’s protection and physiological function in multiple and often complex ways.[21],[6],[7]

It is not only a difficult to penetrate barrier that prevents harm from external physical, chemical and biological threats, but is also metabolically active and together with its accessory structures (i.e. hair, nails, and glands) assists in maintaining the body’s homeostasis.^[20],[22]

A.1.2 Skin functions

The skin’s epidermal layer is the first line of defense between the body and the external environment. It serves as a physical barrier against microbe threats,

30

(39)

APPENDIX A. THEORETICAL BACKGROUND 31

but also synthesizes antimicrobial peptides (i.e. defensins, cytokines, chemokines etc) that together with the sebaceous and sweat glands’ secretions help weaken or kill bacteria, fungi and viruses.[21],[20],[22]

Even if the pathogen intrusion is not prevented, skin cells (i.e. Langerhans cells) initiate immune responses to help neutralize the threat.^[21],[20]Another type of skin cells, the melanocytes, provide protection against the damaging effects of the UV light.[21],[20],[6]

Apart from protection, the human body relies on the integument for maintaining homeostatic conditions, by preventing loss of fluids through the water- impermeable uppermost layer of the epidermis (stratum corneum). Thermoreg- ulation is another function of the integumentary system, achieved through va- sodilation and vasoconstriction of the vascular plexi of the dermal part and the secretions of the sweat glands.^[21],[20]

Skin also enables the body’s sensory system to receive information from the outer world through the nerve bundles that are found along the capillaries in the dermal part. They run in parallel with the skin surface and facilitate the perception of touch, pressure, pain, itching, and temperature; with the nails also participating in the sensory perception.[21],[6],[20]

Moreover, the cutaneous membrane has a crucial role in vitamin D synthesis and therefore contributes to bone formation and calcium metabolism.[21],[20],[22]

Finally, the subcutaneous fat, found in the deeper layer of the skin, serves as an energy reserve, as insulation, and as an additional protection layer of the inner organs against trauma.^[20]

A.1.3 Skin anatomy

A.1.3.1 Epidermis

The epidermis is the outermost and thinnest skin layer. It is an avascular, strat- ified, squamous epithelium composed of different cell types: keratinocytes, melanocytes, Langerhans cells, and Merkel cells. Keratinocytes constitute at least 80% of the total epidermal cells and the state of their differentiation di- vides the epidermis further into four main sublayers: the basal layer (stratum basale), the squamous layer (stratum spinosum), the granular layer (stratum granulosum), and the cornified or horny layer (stratum corneum).^[6],[7]A representation of epidermis is illustrated in Figure A.2

(40)

32 APPENDIX A. THEORETICAL BACKGROUND

Figure A.1: Illustration of a full-thickness human skin with the vascularized dermal part subdivided into the papillary and reticular layers.

Keratinocytes

Keratinocytes originate from stem cells found in the layer closest to the dermis;

the basal layer. Keratinocytes migrate through the several epidermal layers in a two-stage process called keratinization. The first stage, the synthetic one, begins when the stem cells divide. Half of the daughter cells start to move towards the epidermal surface and differentiate, while the rest of them remain in the basal layer. At this state the keratinocytes have a columnar shape with their longer axis perpendicular to the dermis.^[6],[7]

In the next layer, the stratum spinosum, the columnar keratinocytes change to polygonal and begin synthesizing an insoluble protein (keratin) that acts as intermediate filament. Keratin is organized concentrically around the nucleus and along with the desmosomes form a shear-resistant layer that protects the epidermis from physical stress. Keratin filaments have their closest-to- the-nucleus end free while the other one is bound to desmosomal plaques, therefore creating mechanical intracellular bonds between the epidermal cells.

Desmosomes appear in a spine-like motif around the cells, giving this layer its name.^[6],[7]

As keratinocytes migrate through the layers they maturate and ultimately un-

(41)

Figure A.2: Illustration of the four epidermal sublayers as formed by keratinocyte cells.

dergo apoptosis. The granular layer is the uppermost epidermal layer in which keratinocyte cells are alive and get programmed to terminally differentiate.

The stage of keratinization is initiated as a result of the enzymatic degrada- tion of the keratinocytes’ nuclei and organelles, however, leaving their keratin unaffected. In the granular layer, the keratinocytes retain abundant keratohya- line granules in their cytoplasm, containing proteins responsible for keratin filament aggregation. The changes they undergo, leave the keratinocytes with an increasingly flattened shape as they progress towards the epidermal surface. Finally, as the lamellar bodies produced by keratinocytes release their contents, the epidermal lipid barrier is constructed.^[6],[7]

The cornified layer is the most superficial epidermal layer and composed of dead cells lacking nuclei and organelles. These flattened and elongated cells are surrounded by a protein envelope, filled with keratin proteins, and called corneocytes. The corneocytes are held together by a lipid extracellular matrix and the remains of the desmosomes, forming a resilient layer that protects from mechanical forces, water loss and external threats. The keratinocytes’ cycle is completed with desquamation occurring when the corneocytes reach the outer part of the stratum corneum and the desmosomes holding them together degrade, resulting in them being shed off.^[6],[7]

Melanocytes

Melanocytes are found in the basal layer in a ratio of 1 in every 10 keratinocytes

(42)

in most body areas, however being more frequent in the face, shins and geni- talia. Melanocytes’ main purpose is to protect the keratinocytes and therefore the epidermis from the damaging effects of the UV light.

For this purpose, they produce elongated pigment-producing organelles, the melanosomes. Melanocytes are dendritic cells able to reach many keratinocytes over long distances both in the basal and the squamous layer and are able to transfer melanosomes to them with cytocrine secretion. The melanosomes are moved in the ends of their dendritic processes, which are then phagocytosed by the keratinocytes. Once the melanosomes are taken in, they surround the keratinocytes’ nucleus in a protective layer that prevents UV damage.^[6],[7]

Merkel Cells

Merkel cells contribute to the sensing properties of the skin. They are slowly adapting type 1 mechanoreceptors that are found in the basal layer of areas sensitive to touch (e.g. digits, lips etc). Once they are stimulated by the small deformations of the keratinocytes they are attached to, they chemically signal the adjoining afferent neuron, which triggers an action potential and transfers the sensory signal to the brain.^[6],[7]

Langerhans Cells

Langerhans cells constitute 2-8% of the epidermal cell population and contribute to the immunologic defense of the skin.^[7]They are dendritic cells found mainly in the granular and also in the basal layer. They recognize and process microbial antibodies in the epidermis and once they become antigen present- ing cells they migrate to the lymph nodes and activate the T-cells.^[6]

A.1.3.2 Dermis

The dermal layer of the skin is found underneath the epidermis. The dermis is thicker (ranging from 0.3mm on the eyelids to 3mm on the back^[6]), does not show the depth dependent differentiation the epidermis does, and is further divided into two, the papillary and the reticular sublayers. It makes up most of the skin’s volume and provides its elasticity, flexibility, and tensile strength. The dermis accommodates a rich neurovascular network, provid- ing nutrition and sensation, as well as appendages like sweat and sebaceous glands, and hair follicles. The dermis and epidermis interact with each other in

(43)

order to maintain their properties and repair damaged tissue. The dermis also supports structurally the epidermis, nourishes it, and oxygenates it through diffusion.^[6],[7]

Structure

The dermis is of mesenchymal origin and the dermal fibroblasts are the main cells responsible for its structure. They secrete the structural components of the dermal extracellular matrix (ECM), i.e. collagen, elastin, and extrafibrillary matrix proteins.^[6],[7]

About 70% of the dermis consists of collagen fibers. They are thin and loose (type I and III) in the papillary layer, which also contains elastic fibers, con- nective tissue cells, and anchoring fibrils (type VII). In the reticular dermis, collagen fibers are larger in diameter and more dense forming a net-like surface parallel to the skin that provides resistance to tensile forces. While collagen contributes to skin durability, elastic fibers, composed of protein filaments and the protein elastin, offer its elasticity. Finally, the extrafibrillary matrix is a gel, mainly composed of proteoglycans and hyaluronan, which fills the space between the fibers and allows molecules, cells, and fluids to move through it.

[6],[7]

Vasculature

The dermis is highly vascularized with the arterial branch supplying the tissue with fresh oxygen and nutrients, as well as immune cells in case of inflamma- tion, while the venous branch circulates deoxygenated blood back to the heart.

This vascular network shows structural differences between the papillary and the reticular layers. The larger blood vessels are found in the deep plexus, located in the interface between the reticular dermis and subcutis. Originat- ing from these vessels, the capillaries of the superficial plexus branch up in the papillary-reticular dermis junction. Dermal blood vessels are surrounded with smooth muscle cells, important for controlling perfusion.^[6],[7]

A.1.3.3 The dermo/epidermal junction

Between the dermis and the epidermis lies a porous and semipermeable membrane, called the basal lamina. This basement membrane mainly consists of collagen (type IV) and laminin (111) produced by the basal keratinocytes. The

(44)

basement membrane, in the dermoepidermal junction, is attached to the col- lagens (type I and III) of the papillary dermis by anchoring fibrils and affects the interactions between the two skin layers. It regulates the growth, adhesion and movement of the keratinocytes and fibroblasts, allows the diffusion of nutrients, provides structural support for the epidermal layers, and facilitates the dermal-epidermal attachment.^[6],[7]

A.1.3.4 Hypodermis

The innermost layer of the skin, hypodermis, lays over the muscles, is mainly composed of adipocyte cells, and serves a variety of purposes. The subcutaneous tissue protects the inner organs from trauma, creates an insulation layer and energy reservoir, provides buoyancy, and is considered an endocrine organ. The groupings (lobules) of the adipocytes in the hypodermis are separated by fibrous septa composed of large blood vessels and collagen continuous with the one in the dermis. A vascular plexus runs through the hypodermis facili- tating its nutrition and oxygenation needs.^[6],[7]

A.2 Skin Tissue Models

A.2.1 Introduction

The numerous skin functions and its important role in maintaining homeosta- sis, make constructing skin tissue models that can recreate the in vivo condi- tions a research topic of high interest.

Traditional two dimensional (2D) models where the cells are cultured in flat plastic surfaces, although relatively simple to maintain, do not offer a realistic representation of the actual tissue environment. The cell monolayers have to adapt on the rigid surfaces they are cultured on and as a result do not have the same density, metabolic activity, ECM composition, and proliferation capacity they would have had in a three dimensional (3D) environment. They also express different surface receptors and often have differences in the cell migration, signaling, and drug responses due to the model’s inability to represent an in vivo-like microenvironment. Additionally, the layered structure and the complex cell-cell and cell-ECM interactions that are crucial for maintaining skin tissue’s homeostatic conditions cannot be recreated in 2D constructs.^[23]

Animal models for in vivo studies have also been an option for studying various

(45)

tissues including skin. A number of different animal models have been suggested with the most common option being, pigs and rodents (rats, mice and guinea pigs). Porcine skin, frequently obtained from abattoirs, shares a lot of histological and biochemical similarities to the human tissue and is commonly used for permeability studies, however it is not easy to obtain. Rodents, on the other hand, are easy to handle, small sized and have a higher availability compared to pigs. Rat skin’s structure is the most similar to human among rodents, and the use of hairless species does not have the disadvantage of too high hair follicle density. However, animal models come with a number of drawbacks.

Despite sharing similarities to human tissue, in many cases a poor correlation between the species is indicated.^[24]Except for handling, cost, and availability problems, ethical issues are also raised in using them. Additionally, animal testings for the development of cosmetic products has been prohibited in the European Union (EU) for over a decade^[25], pushing the industry towards turn- ing to alternatives for skin models.^[24]

Tissue engineering holds promise to develop 3D skin tissue models with var- ious complexities that manage to recreate, the in vivo conditions. A number of applications has been studied to engineer the skin tissue which thereby, further increases the need for the development of more realistic models. Testing new cosmetic and pharmaceutic products for their safety and efficacy prior to market release is a mandatory step in the development process. Engineered dermal and epidermal models are suitable for testing new compounds for skin irritation, edema, erythrema, and toxicity as well as evaluating percutaneous absorption. Drugs penetration in the deeper skin layers and their associated effects on the tissue’s metabolic activity are properties of interest that can be evaluated through skin models representing all skin layers. Disease models for skin tissue, including cancer, as well as disorders like psoriasis and vitiligo, can offer insight on the cell-cell and cell-ECM signaling, and facilitate the evaluation of drug efficacy. The effects of sunlight’s UV radiation can also be studied in 3D models that include melanocytes. Finally, skin grafts that promote skin reconstruction and revascularization are applicable in cases of damaged skin tissue due to burns, injuries, surgical wounds and skin disorders.^[23]

A.2.2 Tissue engineered human skin equivalents

A number of different approaches, regarding both the biomaterials used and methods employed, have been followed in order to construct scaffold matri- ces suitable for mimicking the skin tissue’s in vivo microenvironment. These

(46)

approaches include the use of natural biomaterials, synthetic ones or com- binations of both, formatted mainly in porous, fibrous, hydrogel, and acel- lular constructs. Each of the scaffold types comes with certain advantages as well as drawbacks, but the main properties of interest for all the matrices are: biocompatibility, adhesion and penetration, scaffold-cell bioactivity, proliferation, and differentiation, vascularization capacity, tensile strength and biodegrability.^[26] Commercial products from various biomaterial combina- tions are available, aiming for one (dermal) or two (dermal-epidermal) layered skin regeneration for burns and skin wounds, or proposed as 3D models of the skin tissue for in vitro research.

A.2.2.1 Challenges of bilayered tissue engineered skin models

In order to construct a 3D skin tissue model that aims to mimic the native skin microenvironment, the complicated dermal-epidermal crosstalk needs to be recreated, therefore both the dermis and epidermis need to be mod- eled. Fibroblasts and keratinocytes participate in a complex double-paracrine signaling mechanism in which keratinocytes’ cytokine secretion stimulates the fibroblasts to secrete signaling molecules, which in turn promote the keratinocytes’ proliferation and differentiation. Furthermore, both cell types participate in synthesizing the components of the basal membrane in the dermal- epidermal junction.^[23],[27]Vascularization of these skin equivalents is of high importance for both in vitro and in vivo applications. In terms of 3D models, the presence of a capillary network would provide the construct with access to nutrients and oxygen, thereby not depending on passive diffusion. In addition, the conditions of native tissue will be more realistically approached, as well as the existing scale-up limitations will be overcomed. The in vivo impact of a prevascularized skin graft would also be significant, increasing the skin regeneration rate, limiting the need for autografting, and preventing necrosis due to lack of neovascularization.^[27]

A.2.2.2 Biomaterials

The most commonly used biomaterials for skin tissue regeneration are natural proteins like, collagen, gelatin, silk, and fibrin, or polysaccharide-based, like chitosan, hyaluronic acid (HA), and alginate.

Collagen is a fibroblast-produced protein abundant in the skin tissue. It is widely used as a scaffold material, as it facilitates cell adhesion and prolif-

Engineering Vascularized Skin Tissue in a 3D format supported by Recombinant Spider Silk

Engineering Vascularized Skin

Tissue in a 3D format supported by Recombinant Spider Silk

SAVVINI GKOUMA

Engineering vascularized skin tissue in a 3D format

supported by recombinant spider silk

SAVVINI GKOUMA

Abstract

Contents

Chapter 1

Aim of the project

Chapter 2 Methods

2.1 Cell Culture

2.2 Construction of the dermal part

2.3 Construction of the BSE

2.4 Cell viability assay

2.5 Cryosectioning

2.6 Immunofluorescence

2.7 Image Analysis

Chapter 3 Results

3.1 Construct morphology

3.2 Cell viability assay

3.3 Dermal layer: ECM components

3.4 Dermal layer: Vascularization

3.5 Epidermal layer: Keratinocytes’ differen- tiation

Chapter 4 Discussion

Chapter 5 Conclusion

Acknowledgements

Bibliography

Appendix A

Theoretical Background

A.1 The human skin

A.1.1 Introduction

A.1.2 Skin functions

A.1.3 Skin anatomy

A.2 Skin Tissue Models

A.2.1 Introduction

A.2.2 Tissue engineered human skin equivalents