Page 172 - IJB-10-4

International Journal of Bioprinting Printing collagen type IV membrane

3.3. Comparison of adhesion and wound healing of 3–5 days of culturing in 5% hPL, with a 100% detachment
human endothelial cells between collagen type IV rate observed among the 26 Col-IV membranes tested.
and type I Approximately 11.5% of membranes detached on day 3,
The adhesion test results displayed no significant differences 46.2% detached on day 4, and 42.3% detaching on day 5.
between the fabrication methods used to produce the Col- Notably, the Col-IV membranes cultured in 5% fetal calf
IV membrane (Figure 3A). This is likely because both serum (FCS) media remained adhered to the coverslips
membranes have a similar topography to Descemet’s (data not disclosed). A key difference between hPL and
membrane. Col-IV membranes exhibited significantly FCS is the presence of fibrinogen. Fibrinogen could lead to
higher cell attachment than Col-I, chondroitin/laminin, coagulation of cell culture if heparin is not added and could
and the control (tissue culture vessels without coating). affect the cell and matrix protein interaction. 30,31 Although
Both collagen types demonstrated at least sevenfold higher no gelling of media was observed, this may be a reason for
cell attachment than chondroitin/laminin-coated wells or at the self-attachment. Further investigation is warranted to
least 22-fold higher than controls (p < 0.01). No significant determine the exact mechanism of the self-attachment.
difference was found between chondroitin/laminin- The culturing system of the Col-IV membrane with
coated wells and controls. This finding is consistent with hPL addition enabled the self-detachment of confluent
a published study, which reported that Col-IV is essential cell membranes, eliminating the need for manual removal
for human corneal endothelial cell attachment compared and minimizing potential mechanical damage to the cells.
to Matrigel, laminin, or fibronection. Unlike the previous Other studies have used thermo-responsive materials
27
study, which conducted the test by coating Col-IV onto to detach cell sheets from culturing vessels without
an atelocollagen sheet (Col-I sheet), we fabricated a requiring mechanical handling. These materials include
27
pure Col-IV sheet without Col-I. Therefore, our results N-isopropylacrylamide-co-glycidylmethacrylate and
32
confirmed that Col-IV is superior to other substrates, elastin-like polypeptides functionalized with arginine-
including Col-I, for endothelial cell attachment. glycine-aspartic acid (RGD), demonstrating the benefits
33
The scratch assay results demonstrated that Col-IV can of easy handling. However, these materials were not tested
facilitate endothelial cell wound healing (Figure 3B). Full with primary human endothelial cells; instead, they were
recovery was observed on Col-IV membranes by day 3, examined with rabbit corneal endothelial cells and human
whereas only partial wound recovery was observed on Col-I adipose stem cell-derived corneal endothelial-like cells,
membranes and controls. The role of extracellular Col-IV both of which inherently exhibit higher proliferation rates.
in wound healing remains unknown. However, studies Therefore, it remains unclear whether these materials have
have been conducted to evaluate the effect of Descemet’s the same cell compatibility as Col-IV. Nevertheless, our
membrane, with Col-IV being the key matrix protein, on findings present an alternative that requires no additional
corneal endothelium repair. In vitro and in vivo studies on mechanical or chemical means for membrane removal and
rabbit corneal endothelial cells have reported that intact has been tested on primary human endothelial cells.
Descemet’s membrane supported corneal endothelial cell The average B4G12 cell density on the Col-IV
regeneration compared to stripped Descemet’s membrane, membrane was found to be 3710 ± 791.3 cells/mm 2
and transplanting decellularized Descemet’s membrane on day 5 (Figure 4). Conversely, the average density
28
facilitated recovery of the endothelium. In contrast, the of P-HCECs was 1193 ± 120.2 cells/mm on day 10
2
transplantation of decellularized Descemet’s membrane (Figure 4). This difference was likely due to the initially
to treat human corneal endothelial diseases has yielded low seeding number and a slower rate of proliferation for
mixed results, warranting further studies on the biological P-HCECs compared to B4G12. In addition, the B4G12
role of Descemet’s membrane and Col-IV in regulating the cell line could be stratified in a crowded area, resulting
29
cellular activities of corneal endothelial cells.
in high cell counts. The average light transmittance of
3.4. Characterization of bioengineered the generated Col-IV membrane seeded with B4G12 and
corneal endothelium — morphology and P-HCECs cells was > 90% (Figure 2B), which exceeded the
26
immunostaining results reported 80% of the native cornea.
Both B4G12 and P-HCECs could reach confluence on The immunostaining results of B4G12 cells revealed
Col-IV membranes (Figure 4A and B). Additionally, strong expressions of ZO-1 and Na /K -ATPase, the two
+
+
when the seeded Col-IV membrane (with either B4G12 typical corneal endothelial cell markers, as well as laminin
or P-HCECs) was cultured in 5% hPL, it started to self- and Ki67 in all culturing conditions (Figure 5). Since no
detach from the coverslip it was printed or molded on laminin was added to the Col-I and Col-IV membranes, the
(Figure 4C). This self-detachment typically occurred within detection of laminin expression indicated that B4G12 cells

Volume 10 Issue 4 (2024) 164 doi: 10.36922/ijb.3258