In situ growth induction of the corneal stroma cells using uniaxially aligned composite fibrous scaffolds

Abstract

Uniaxially aligned composite fibrous scaffolds of gelatin and poly-L-lactic acid (PLLA) were fabricated using electrospinning and the scaffolds were implanted into the corneal stroma layers of New Zealand white rabbits (NZWRs) to observe the in situ growth induction of the stroma cells. The effects on cell growth were evaluated by both apparent observation and pathological analysis. It was demonstrated that the scaffolds had a good compatibility with the corneal tissues and were nontoxic by observing the changes of the structure and the physiological activity of the corneal tissues around the scaffolds using a slit lamp and in vivo confocal images. The in vivo confocal images of the scaffolds implanted into the eyes of NZWRs show the process of the cells’ ingrowth and the tissue regeneration, which indicated that the uniaxially aligned fibers could induce the polarized ingrowth of keratocytes which may provide the basis for clinical application to the in situ repair of corneal stromata.

---

1. Introduction

The cornea is a transparent tissue on the surface of eyeball, which plays important roles in both protecting the eyes and the formation of vision. So, the health of the cornea is a pre-condition for normal vision.^1,2 Corneal disease has already become the second most prevalent eye disease leading to blindness in the world.^3,4 The shortage of donors is a major problem in the treatment of corneal grafts.^5,6 Corneal tissue engineering provides a new approach for corneal repair.^7,8 However, the greatest challenge of corneal tissue engineering is how to realize the special physiological function of the cornea.⁹ The special physiological structure of the corneal stroma is the key to maintaining its function.^10,11 The human corneal stroma is about 500 μm thick, and its extracellular matrix consists of more than 200 thin layers of parallel and staggered collagen fibers with a uniform diameter of 30 nm.^12,13

Although gels and sponges with a disordered structure composed of collagen or other biological materials have realized the reconstruction of the corneal tissue in vitro,¹⁴ at present none of these tissue engineered corneas have the above-mentioned ordered structure in their reconstructed stromata.¹⁵ Research has shown that,^16,17 compared with a randomly oriented fibrous scaffold, the uniaxially aligned scaffold of gelatin and PLLA, which has the same parallel arrangement as the normal corneal cells, can induce the growth of matrix cells along the fibers superiorly. At the same time, this scaffold allows the corneal cells to grow into it and form a three-dimensional structure. Thus, it is speculated that this scaffold can provide physical support and induce the directional growth of corneal cells.

On the basis of the previous work,^16,17 the uniaxially aligned composite fibrous membrane of gelatin and PLLA was prepared and used to induce the in situ growth of autologous corneal stroma cells of NZWRs.

2. Experimental section

2.1. Materials

Gelatin (type A, from porcine skin, Sigma) and poly-L-lactic acid (PLLA, M_w = 300 000, from the Shandong Institute of Medical Appliances, People’s Republic of China) were used as the materials for electrospinning. 1,1,1,3,3,3-Hexafluoro-2-propanol (HFIP) and N,N-dimethylformamide (DMF, from Atoz Fine Chemicals) were used as the solvents for electrospinning. Ethanol (EtOH) and dimethyl sulfoxide (DMSO, also from Atoz Fine Chemicals) were used as received. The xylazine hydrocloride and tetracaine hydrochloride injections (from the animal experimental center of Norman Bethune College Of Medicine, Jilin University) were used as anaesthetics. A 3,3′-diaminobenzidine color developing reagent kit (DBA, Shanghai Sangon Biotech Company) was used for immunohistochemical staining. The animals used in the experiments were NZWRs purchased from the animal experimental center of Norman Bethune College Of Medicine, Jilin University. The sterile surgical suture MANI10-0 was imported from Japan. All the other chemicals and agents were analytically pure.

2.2. Preparation of aligned gelatin/PLLA fibrous scaffold

A certain amount of gelatin and PLLA (G70P30, mass ratio of gelatin to PLLA was 70 : 30) was successively dissolved in the mixed solvent of HFIP and DMF (volume ratio of HFIP to DMF was 4 : 1) to prepare a 10% (w/v) polymer solution. The polymer solution was placed into a 5 mL syringe with a needle of a 0.4 mm inner diameter and was pumped using a syringe pump at a rate of 30 μL min⁻¹. The electrospinning voltage was 15 kV, the tip-to-collector distance was 18 cm and the ambient humidity was controlled at 60–70%. The copper wire drum (Fig. 1) was used as a receiver for the aligned composite fibrous scaffold. Then, the aligned composite fibrous scaffold was placed into a vacuum drying oven to remove residuary solvent.

Fig. 1 The electrospinning receiver for the aligned nanofibers – a rotating copper wire drum.

2.3. Analytical methods

2.3.1. Morphology of aligned composite fibrous scaffold. The morphology of the electrospun aligned gelatin/PLLA fibrous membrane was characterized using scanning electron microscopy (SEM, Shimadzu SSX-550) at an accelerating voltage of 3 kV. The average fiber diameter was measured with the Photoshop software.

2.3.2. Damage model and implantation experiment. The fibrous membranes were cut into circular discs of 6 mm in diameter and 0.5 mm in thickness and disinfected with 75% (volume fraction) ethanol for 30 min. Then, the cut scaffolds were washed with sterile Hank’s solution for 3 times, immersed in the serum-free medium DMEM and then placed overnight into a refrigerator at 4 °C. These operations were performed on a super clean bench. The disinfected scaffolds were washed with a sterile phosphate buffer solution to wipe off the medium before implantation.

18 healthy NZWRs were generally anesthetized with xylazine hydrochloride (applied amount: 0.4 mg kg⁻¹) via intramuscular injection, and then locally anesthetized by 1% tetracaine hydrochloride injection. A slight mark was drilled on the cornea by a trephine with a diameter of 7.5 mm and an incision, which was about 2 cm long, was opened beside the drilled mark. After that, a matrix vesicle of 6.5 mm in diameter was made in the middle of the corneal stroma using an iris separator, and on the vesicle, an artificial minimally invasive wound was created. Then, the disinfected scaffolds were implanted into the vesicle, and spread evenly. Finally, the incision was stitched up with no. 10 sterile surgical suture, and the operation was completed. All animal handling procedures conformed to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

2.3.3. Clinical characterization. The 18 rabbits were divided into 6 groups and the samples for clinical characterization were taken from the rabbits after the operation for 0 days, 2, 6, 12, 24 and 32 weeks respectively. The clinical phenomenon was observed by a slit lamp, anterior segment optical coherence tomography (OCT) and intravital confocal microscopy through focusing (CMTF).

After the operation, the experimental eyes were observed with a slit lamp (HPG100, OLYMPUS, Japan) after 0 days, 2, 6, 12, 24 and 32 weeks to record the corneal transparency and the growth of new blood vessels.

Anterior segment OCT (made in Germany, Carl Zeiss company) was used to estimate the degradation of the fibrous scaffolds and the ingrowth rate of the cells and to measure the thickness of the cornea. The distance between the instrument and the experimental eye was adjusted until a vertical white stripe appeared in the center of the cornea, then real-time scanning was carried out. Lastly, the images were analyzed to measure the corneal curvature and thickness.

The phenotype and density of the cells were inspected using intravital CMTF (made in Japan, NIDEK company, fourth generation). The magnification was set to 1000×, the scanning range was 340 × 255 μm, and the scanning thickness was about 700 μm. 350 images were measured in every scan. The surface of the experimental cornea was perpendicular and close to the objective lens, and cells of different layers were scanned and recorded as images. Then, the NAVIS software was used for image analysis to count the cell density of the epithelium, anterior stroma, posterior stroma and endothelium.

2.3.4. Pathology characterization. Samples were taken at 2, 6, 12 and 32 weeks after the operation and quickly immersed in 4% paraformaldehyde solvent to fix the cells, then the samples were stained using conventional HE and immunohistochemical methods. In the immunohistochemical staining of vimentin in the corneal stroma cytoplasm, the first antibody (Ab) used was rabbit mono-clonal anti-vimentin, and the second Ab was goat anti-rabbit IgG-HRP. In the immunohistochemical staining of vimentin in the corneal epithelial cytoplasm, the first Ab used was the rabbit monoclonal anti-keratin K3/K76 antibody, and the second Ab was goat anti-rabbit IgG-HRP. The concentrations of the first and second Abs were 1 : 200 and 1 : 400, respectively. Then, the samples were observed and analyzed using an optical microscope and the Scopephoto software.

3. Results and discussion

3.1. Morphology of aligned composite fibrous scaffold

An aligned gelatin/PLLA (G70P30) fibrous membrane with 10% w/v concentration was made and its morphology was characterized using SEM, as shown in Fig. 2. As can be seen in Fig. 2, the fibers were straight and smooth, and arranged approximately in a single orientation. The diameters were mainly distributed between 750–1000 nm, as shown in Fig. 3.

Fig. 2 SEM image of aligned fibers of G70P30 and the magnification of one segment (insert).

Fig. 3 Diameter distribution of aligned fibers of G70P30.

3.2. Apparent evaluation of biocompatibility

The biocompatibility of the scaffold was evaluated by observing the scaffold implanted in the eye of NZWRs. The results in Fig. 4, obtained with a slit lamp, show that the fibrous scaffold implanted into the corneal stroma layers presented a semi-transparent gel state at the first hour (0 days). Two weeks later, it showed an observable improvement in transparency. However, the transparency declined slightly at the 6th week because of shrinkage. At the 12th week, the scaffold shrank apparently and became more transparent. At the 24th week, the scaffold became hard to be distinguished from the normal cornea, and the whole cornea was almost completely transparent at the 32nd week. This showed the uniaxially aligned fibrous scaffold has a good compatibility with the cornea. According to the literature,^18,19 stylolite could cause the neogenesis of new vessels. As can be seen in Fig. 4, a small amount of new blood vessels was generated around the seams after the 2nd week, but gradually reduced in later observations until they completely disappeared after 32 weeks. Therefore, it could be suggested that the scaffold could not cause the generation of new vessels, which was favorable for the repair of the rabbit corneal stroma.

Fig. 4 Restorative process of the corneal transparency of NZWRs during a 32 weeks post-operative slit-lamp examination.

The in vivo confocal images of the allopelagic central cornea of NZWRs for 32 weeks post-operative in Fig. 5 show that nerve fibers (marked by arrows) were growing in stroma layers at different depths, which indicated that the scaffolds have some functions to the cornea regeneration. All the adjacent keratocytes have a normal phenotype, namely an even rank of cells and a short rod-like cell nucleus. The wing-like epithelial and basal cell layers were clearly visible. And the endothelial cells with a uniform size and regular hexagonal shape were arranged very closely. Table 1 shows the results of the statistical analysis of the corneal cell densities of different central cornea layers. It shows that there were no significant differences in cell density among the different central corneal layers. These results indicated that the implanted aligned fibrous scaffolds had no toxic effect on the surrounding tissues.

Fig. 5 Typical in vivo confocal images of the central cornea of NZWRs for 32 weeks postoperative. (A) Basal epithelial cornea, the innermost epithelial layer; (inset of (A)) the outermost epithelial layer; (B) anterior stroma; (C and D) middle depths of stroma; (E) posterior stroma; (F) endothelial cornea which is a single layer in the posterior cornea. The arrows mark the nerve fibers.

Table 1 Corneal cell densities of different central cornea layers

Cell density (cells per mm^2)	Epithelium	Anterior stroma	Posterior stroma	Endothelium
Control	4827 ± 156	897 ± 143	465 ± 158	2502 ± 191
32 weeks	4736 ± 145	927 ± 197	486 ± 137	2473 ± 177

---

The results of the HE staining are shown in Fig. 6. As can be seen, all the epithelial cells for the different time periods were arranged very closely. The thickness of the epithelium layer had increased a little 2 weeks after the operation (Fig. 6B), and then gradually normalized (Fig. 6C and D). The endothelium was complete with a normal thickness (Fig. 6E–H). The keratocytes increased in the surrounding tissues between the 2nd and 12th week, and then returned to normal until the 32nd week. The corneal epithelial cells were stained using immunohistochemistry, and the chromogenic results after DAB developing are shown in Fig. 7. Consistent with the results of the HE staining (Fig. 6), the thickness of the epithelium increased after the 2nd week. This could be explained by the fact that the epithelial layer was damaged to some extent during the operation, which might lead to a certain degree of mild edema. The thickness was back to normal at the 12th week due to the self-healing process. It’s also shown in Fig. 7 that the endothelium was formed by a single layer of endothelial cells, and there was no obvious change in cell size and arrangement during the different periods. These suggested that the implantation of the fibrous scaffolds caused neither negative effects on the endothelium, nor lesions on the surrounding tissues, and this also proved that the scaffolds had a good histocompatibility.

Fig. 6 Corneal histocompatibility of the aligned scaffold with HE staining. (A–D) Epithelium and anterior stroma, (E–H) posterior stroma and endothelium. (A and E) Pre-operative; (B and F) 2 weeks; (C and G) 12 weeks; (F and H) 32 weeks. Magnification: 100×.

Fig. 7 Corneal histocompatibility of scaffold with immunohistochemical staining. (A–D) are the changes of the epithelium and (E–H) are the changes of the endothelium. (A and E) Pre-operative; (B and F) 2 weeks; (C and G) 12 weeks; (F and H) 32 weeks. Magnification: 400×.

3.3. Apparent evaluation of the regeneration process

Anterior segment OCT was used to observe the physiological structure of the anterior segment and to measure the thickness and curvature of the cornea for examining the regeneration process of the stroma coloboma. The results are shown in Fig. 8. The difference in density between the fibrous scaffold and normal corneal tissue led to a significant difference in gray level between the scaffolds and surrounding tissues in the OCT images. The scaffolds were highly reflective, as shown in Fig. 8A. The scaffolds were still distinct at the 6th week. The difference in gray level was blurring after the 12th week and disappeared completely by the 32nd week. It was speculated that the scaffolds gradually degraded in the observation period and the proliferation of the keratocytes around the scaffolds was activated according to the previous pathology characterization. Meanwhile, the scaffolds could gradually induce the cell growth into the scaffolds to form the extracellular matrix which resulted in the increasing approximation of the whole density of the scaffolds to that of the surrounding normal tissues.

Fig. 8 (A–F) Dewarped cross-sectional images of the central cornea of NZWRs for 32 weeks post-operative using optical coherence tomography (OCT) overlaid with the detected corneal boundaries; (G) central corneal curvature; (H) thickness. Data are shown in mean ± SD. Arrows mark the scaffolds.

In addition, Fig. 8G shows that there was no significant change in the corneal curvature 6 weeks after the operation. Fig. 8H shows that the corneal thickness returned to normal gradually within 12 weeks after the operation. It indicated that the implanted scaffolds had no negative effect on the corneal structure which was consistent with the result of the apparent evaluation.

The corneal images of the implant position were obtained using laser scanning confocal microscopy (LSCM), as shown in Fig. 9. The images show the migration of the stroma cells toward the fibrous scaffolds. As can be seen, the scaffolds were highly reflective and in close contact to the surrounding tissues at the initial postoperative stage (Fig. 9A). The cells gradually grew into the scaffolds during the subsequent observation (Fig. 9B–F). The scaffolds became non-reflective by the 32nd week. The cells of the central cornea were clearly visible with a rod-like nucleus, having the same phenotype as the normal corneal stroma. This changing process was consistent with the results shown in Fig. 8, namely, the embedded aligned fibrous scaffolds could induce the ingrowth of the surrounding keratocytes and keep the normal phenotype of the keratocytes. A lamellar structure was finally formed with an increasing amount of keratocytes.

Fig. 9 The typical in vivo confocal images of the central cornea of NZWRs at 32 weeks post-operative show the process of cell ingrowth and tissue regeneration. (A) 0 days, post-operative; (B) 2 weeks; (C) 6 weeks; (D) 12 weeks; (E) 24 weeks; (F) 32 weeks. The “d” values indicate the depths of confocal imaging.

3.4. Pathology evaluation of the regeneration process

It was proven by the apparent evaluation that the uniaxially aligned fibrous scaffolds could induce the ingrowth of surrounding keratocytes and keep their normal phenotype. In this section, the ingrowth process and the formation process of collagen type I fibers were further confirmed using tissue section staining. The results of the HE staining are shown in Fig. 10. Fig. 10A and F are the pre-operative contrast images. In comparison, the scaffolds were clearly discernible at the 2nd week post-operative, and had begun to heal over with the surrounding tissues between the 6th and 12th week. Then, the scaffolds completely disappeared and entirely normal tissues were formed by the 32nd week. The zoomed-in images in Fig. 10F–N revealed the ingrowth process more clearly.

Fig. 10 Restorative process of the corneal stroma of NZWRs using HE staining. (A and F) Post-operative, (B and G) 2 weeks, (C and H) 6 weeks, (D and M) 12 weeks, (E and N) 32 weeks. Arrows mark scaffolds.

The immunohistochemical staining results are shown in Fig. 11. It can be seen in Fig. 11 that the structure of the collagen fibers normalized gradually after the operation (Fig. 11B), and the keratocytes of the new tissues could be shown by staining of a specific protein—vimentin (Fig. 11C). This structure of the regenerated corneal stroma, composed of in parallel aligned collagen fibers and orderly arranged keratocytes, played an important role in keeping the corneal transparency.^20,21

Fig. 11 Restorative process of the corneal stroma of NZWRs using immunohistochemical staining of (A) both collagen type I and vimentin, (B) collagen type I, and (C) vimentin. Magnification: (A) 200×, (B) 400×, (C) 200×.

3.5. Mechanism of in situ induction and repair

According to the above experimental results, the repair mechanism of the in situ induction of the uniaxially aligned composite fibrous scaffolds into the rabbit corneal stroma can be speculated as follows.

In the first period, the scaffolds provided the necessary conditions for cell living, activation and proliferation of the keratocytes from the surrounding tissues. The activated keratocytes were the cell source for in situ regeneration. In this way, the immunological rejection of foreign cells was avoided. Besides, the good histocompatibility of the scaffolds made the scaffolds closely connect with the surrounding tissues, which would stimulate the ingrowth of activated keratocytes.

In the medium term, the uniaxially aligned arrangement of fibers in the scaffolds induced the keratocytes to arrange in parallel and kept their polarized growth until the repair was completed. And this character was essential for maintaining the corneal transparency.^22–24 Because of the proper porous structure of the fibrous scaffolds, they enabled the ingrowth of adequate cells to form a three-dimensional structure.^25–27

During the medium to the last period, the corneal keratocytes grown into the scaffolds formed a normal cellular phenotype with physiological function. On the other hand, the uniaxially aligned scaffolds disappeared because of degradation, which would ensure that the collagen fibers arranged easily without physical obstacles.

4. Conclusion

The uniaxially aligned composite scaffolds showed good compatibility with the corneal tissues, which provided the basic security assurance for the successful repair of the corneal stroma. The apparent and pathology evaluation results of the repair process showed that the implanted scaffolds could induce the ingrowth of corneal keratocytes and keep their normal physiological function, which provided the possibility of promoting the formation of new tissue. The composite scaffolds could provide the three-dimensional space for cell growth and induce autologous cells to form a normal cellular phenotype with a physiological structure.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 21104023) and the Key Program of the Science and Technology Department of Jilin Province, PR China (no. 20140204053GX).

Notes and references

1. S. Salehi, M. Fathi, S. H. Javanmard, T. Bahners, J. S. Gutmann, S. Ergün, K. P. Steuhl and T. A. Fuchsluger, Macromol. Mater. Eng., 2014, 299, 455.
2. H. J. Levis, I. Massie, M. A. Dziasko, A. Kaasi and J. T. Daniels, Biomaterials, 2013, 34, 8860.
3. S. Salehi, T. Bahners, J. S. Gutmann, S. L. Gao, E. Mäder and T. A. Fuchsluger, RSC Adv., 2014, 4, 16951.
4. B. J. Klenkler, D. Dwivedi, J. A. West-Mays and H. Sheardown, J. Biomed. Mater. Res., Part A, 2010, 93, 1043.
5. L. N. Guan, P. Tian, H. Y. Ge, X. L. Tang, H. Zhang, L. L. Du and P. Liu, J. Mol. Histol., 2013, 44, 609.
6. H. L. Luo, Y. B. Lu, T. T. Wu, M. Zhang, Y. J. Zhang and Y. Jin, Biomaterials, 2013, 34, 6748.
7. R. Sakai, B. John, M. Okamoto, J. V. Seppälä, J. Vaithilingam, H. Hussein and R. Goodridge, Macromol. Mater. Eng., 2013, 298, 45.
8. J. Y. Lai, RSC Adv., 2014, 4, 18871.
9. J. Wu, Y. Q. Du, M. M. Mann, J. L. Funderburgh and W. R. Wagner, Exp. Eye Res., 2014, 120, 71.
10. S. Koo, R. Muhammad, G. S. Peh, J. S. Mehta and E. K. Yim, Acta Biomater., 2014, 10, 1975.
11. N. A. Ghafar, K. H. Chua, W. Z. W. Ngah, J. C. Hamzah, F. Othman, R. A. Rahman and R. H. Idrus, Cell Tissue Banking, 2014, 15, 25.
12. M. Alaminos, M. D. C. Sánchez-Quevedo, J. I. Muñoz-Ávila, D. Serrano, S. Medialdea, I. Carreras and A. Campos, Invest. Ophthalmol. Visual Sci., 2006, 47, 3311.
13. N. E. Vrana, N. Builles, V. Justin, J. Bednarz, G. Pellegrini, B. Ferrari, O. Damour, D. J. S. Hulmes and V. Hasirci, Invest. Ophthalmol. Visual Sci., 2008, 49, 5325.
14. R. Lanza, R. Langer and J. P. Vacanti, Principles of tissue engineering, Academic Press, 2011.
15. D. Duan, B. J. Klenkler and H. Sheardown, Expert Rev. Med. Devices, 2006, 3, 59.
16. J. Yan, L. H. Qiang, Y. Gao, X. J. Cui, H. Y. Zhou, S. L. Zhong, Q. Wang and H. Y. Wang, J. Biomed. Mater. Res., Part A, 2012, 100, 527.
17. Y. Gao, J. Yan, X. J. Cui, H. Y. Wang and Q. Wang, Chem. Res. Chin. Univ., 2012, 28, 1022.
18. B. Ozcelik, K. D. Brown, A. Blencowe, M. Daniell, G. W. Stevens and G. G. Qiao, Acta Biomater., 2013, 9, 6594.
19. J. Zavala, G. R. L. Jaime, C. A. R. Barrientos and J. Valdez-Garcia, Eye, 2013, 27, 579.
20. N. Builles, H. Janin-Manificat, M. Malbouyres, V. Justin, M. R. Rovère, G. Pellegrini, J. Torbet, D. J. S. Hulmes, C. Burillon, O. Damour and F. Ruggiero, Biomaterials, 2010, 31, 8313.
21. T. J. Duncan, Y. Tanaka, D. Shi, A. Kubota, A. J. Quantock and K. Nishida, Biomaterials, 2010, 31, 8996.
22. S. Salehi, M. Fathi, S. H. Javanmard, T. Bahners, J. S. Gutmann, S. Ergün, K. P. Steuhl and T. A. Fuchsluger, Macromol. Mater. Eng., 2013, 299, 455.
23. K. Tonsomboon and M. L. Oyen, J. Mech. Behav. Biomed. Mater., 2013, 21, 185.
24. B. D. Lawrence, J. K. Marchant, M. A. Pindrus, F. G. Omenetto and D. L. Kaplan, Biomaterials, 2009, 30, 1299.
25. W. G. Liu, C. Deng, C. R. McLaughlin, P. Fagerholm, N. S. Lagali, B. Heyne, J. C. Scaiano, M. A. Watsky, Y. Kato, R. Munger, N. Shinozaki, F. F. Li and M. Griffith, Biomaterials, 2009, 30, 1551.
26. M. Rafat, F. Li, P. Fagerholm, N. S. Lagali, M. A. Watsky, R. Munger, T. Matsuura and M. Griffith, Biomaterials, 2008, 29, 3960.
27. C. R. McLaughlin, M. C. Acosta, C. Luna, W. G. Liu, C. Belmonte, M. Griffith and J. Gallar, Biomaterials, 2010, 31, 2770.

---