Page 169 - IJB-10-4

International Journal of Bioprinting Printing collagen type IV membrane

were incubated in 5% bovine serum albumin (BSA) in PBS solution exhibited shear-thinning behavior (Figure 1B).
for 30 min at RT. This was followed by incubation with At minimum shear, its viscosity was measured to be 0.8
primary antibodies at 4 C overnight (Table S1). The cells ± 0.3 Pa·s; at maximum shear, its viscosity decreased
o
were then washed again three times in 1× PBS, followed significantly to 0.02 ± 0.0008 Pa·s (p < 0.05). These
by incubation in corresponding secondary antibodies results demonstrated that the neutralized Col-IV
(Table S1) for 2 h at RT in dark conditions. After incubation, solution with riboflavin can be a photo-crosslinkable
the cells were washed with PBS. Coverslips were gently and printable biomaterial.
added to the dishes, and the cells were examined using The printability test demonstrated that at a
confocal microscopy.
temperature of 22°C, the Pr value was closest to 1 with an
2.11. Mock Descemet membrane endothelial average of 0.97, indicating a good Pr. Significantly lower
keratoplasty surgical test Pr values were observed when printing at 7 and 10°C (p
The bioengineered Col-IV corneal endothelial membranes < 0.05). A printing speed of 150 mm/min resulted in the
were tested in a mock Descemet membrane endothelial best Pr value while increasing the printing speed above
keratoplasty (DMEK) surgery. The surgical test was this led to lower Pr values. A flow rate at 0.8 mm/min
conducted using artificial eyes and pig eyes (Wilmeat generated the Pr value closest to 1 with an average of
Cut Meats Pty Ltd, Australia). This test followed the basic 0.93, while lower and higher flow rates both led to lower
steps of a normal DMEK surgery. The membranes were Pr values. The optimal extruding printing conditions of
firstly trephined using 8.5 mm trephine and stained with the photo-crosslinkable Col-IV with Pr close to 1 were:
vision blue dye to enable visualization in the procedure. a printing temperature of 22°C, a printing speed of 150
The membrane was then aspirated into a Stryker injector mm/min, and a flow rate of 0.8 mm/min (Figure 1C). A
and injected into an artificial eye and two pig eyes, in four-layered Col-IV scaffold was also printed using these
which the membrane was required to unfurl. The test was printing parameters (Figure S4)
repeated three times. After successfully being transferred Notably, only extrusion-based bioprinting was used
into the pig eyes, the membranes were also removed in this study. Another two common printing methods
for imaging. Microscopic images of Col-IV membranes are inkjet-based and photopolymerized-based printing,
before and after transplantation were taken and compared which could also apply to this Col-IV ink. The inkjet-
using a phase contrast light microscope. based printer has the advantage of precise positioning of
22
2.12. Statistical analysis droplets to form a pattern. Previously, Col-I solution at 30
All statistical tests and relevant graphs in this study were mg/ml has been used in an electrohydrodynamic printer
23
conducted and generated using Prism 9.0 (GraphPad, (a special type of inkjet printer), but development was
USA). Student t-tests were used to compare two groups/ performed using acetic acid. Our Col-IV ink is developed
values obtained in previous tests. One-way ANOVA was in neutral pH with a lower concentration indicating its
performed to compare more than two groups. A statistically compatibility with inkjet printers. As the bioink is light-
significant difference was defined as p ≤ 0.05. activated, it is possible that photopolymerization-based
printing could also be a viable option. Alginate-based
3. Results and discussion bioink, which had G’ increased to 550 Pa in 300 s under
452 nm light, was used in the stereolithographic printer
3.1. Rheological properties and printability of LittleRP (LittleRP, USA) to print cubes or lattices. The
24
collagen type IV ink Col-IV ink developed in this study had a higher G’ at 915
The Col-IV powder can be successfully constructed Pa with 120 s of UV light activation. Therefore, Col-IV
using the Col-I neutralization method, and riboflavin is likely to be compatible with photopolymerization-
15
can be included as a photo-initiator. The rheology based printers and not restricted to only extrusion-based
data examined the photo-crosslinking profile of Col- printers.
IV, and the results indicated that 12 mg/mL Col-IV
can be crosslinked using UV light (Figure 1A). The 3.2. Physical examination of collagen
crosslinking started when the UV light was turned on type IV membranes
(Figure 1A, first arrow), where the storage modulus (G’) The 3D confocal results, which examined the topography
was approximately 9 Pa. G’ subsequently increased to of the Col-IV membranes generated using either printed
914.7 ± 151.3 Pa when G’ reached the plateau (Figure 1A, or molded methods, displayed a relatively smooth surface
dashed line). The average time for the Col-IV solution indicated by the consistent green color (Figure 2A). This
to be fully crosslinked was approximately 162 ± 21.4 s (< test also demonstrated that these membranes had an
3 min). Viscosity analysis demonstrated that the Col-IV average thickness of 120.9 ± 26 µm (for molded) and

Volume 10 Issue 4 (2024) 161 doi: 10.36922/ijb.3258