Figure 4 - Characteristics of the 14% eSF hydrogel microfluidic
platform. A) Images showing the unique mechanical properties of the
developed platform after stretching (tension), torsion, and bending at
days 1 and 3. B) Perfusion of blue ink through the serpentine channel
and formation of a soluble ink gradient due to the hydrophilic nature of
the eSF hydrogel. Scale bar: 7.3 mm. C) Representative microscopy
snapshots showing the flow of microstructures along the serpentine
channel.
Next, we aimed at building a very simplistic colorectal tumor model, as
shown in Figure 5 A. For this, we encapsulated colorectal cancer cells
within the bulk of the eSF hydrogel and HCoMECS were seeded in the walls
of the serpentine microchannel. We first investigated the ability of
cells to adhere and spread in the microfluidics channel. Hydrogels
obtained from Bombyx mori are not utterly absent of binding
sequences. Silk proteins contain a main heavy chain, considered a
hydrophobic protein with a co-block design
(32). Hydrophilic segments are involved
in the self-assembling process, causing differences in water amount and
the mechanical properties of the final hydrogel. This can elucidate the
mechanisms by which a certain level of cell attachment and migration
happens in Bombyx mori derived hydrogels
(14). Moreover, sequences that are
RGD-like are present in the N-terminal segment of heavy-chain, VTTDSDGNE
and NINDFDED (32). Indeed, previous works
have shown the culture of endothelial cells in eSF hydrogels with good
cell adhesion and growth (11).
Figure 5 B focuses on the culture and adhesion of HCoMECS on the eSF
microchannel. Cells showed normal morphology and elongation 1 day after
the seeding. After 3 days, HCoMECS reached a confluent layer
characteristic of native vasculature
(33). Finally, due to the hydrogel’s high
permeability, DAPI/Phalloidin injection also stained HCT-116 located in
the bulk of the hydrogels.
We next assessed the viability of HCT-116 cells encapsulated within the
14% eSF hydrogel to determine if the cells survived the fabrication
procedure. The live/dead assay performed at a random location of the
device on day 1 and day 3 showed that cells were viable and
proliferating (Figure 5 C-D). This experiment also showed the cells were
isotropically distributed across the eSF hydrogel.
To further confirm cell viability, we quantified the levels of ATP both
in 2D static and in the eSF microfluidic chip for up to 7 days. A
calibration curve was used to define its relationship with
bioluminescence in 2D. We performed a direct correlation using the same
number of cells to correlate the results in our eSF platform. The signal
obtained when testing the hydrogels shows that the cells remained highly
viable (Figure 5 E). In 2D, there is a statistical difference in every
day within the same timepoint due to cell proliferation in cell flasks
without constrains. In 3D, there is only a statistical difference when
comparing day 3 to day 1. Being encapsulated in a rather stiff hydrogel,
cells proliferate at lower rates. Natural 3D hydrogels confer different
characteristics to cells when compared to 2D, such as high cell
viability but controlled proliferation and differentiation
(34). Finally, it is worth noticing the
controls, which can provide valuable information about the homeostasis
of the cells in the system. This means that the chosen design allows for
cell viability: feeding the system through the inlet via the serpentine
is enough to provide nutrients/oxygen diffusion. A significant
statistical difference was found in metabolic activity on days 1 and 3,
with increased activity on day 3 (Figure 5 E). It showed the ability of
the eSF hydrogel microfluidic platform to support a stable and viable
culture of HCT-116 cells under perfusion.
Finally, we evaluated the effect of dynamically injecting 5 µM GEM
(through the serpentine channel) on HCT-116 cells for up to 7 days to
validate the model (35). GEM is now being
tested as treatment of patients with advanced CRC
(36). Alamar blue (cell viability) assay
shows no statistical differences in the presence or absence of GEM on
day 1, most probably due to the time it takes for the drug to diffuse
across the hydrogel (Figure 5 F). Moreover, it was observed a slight
decrease in metabolic activity for GEM between day 1 and day 3. The
anti-neoplastic effect was only observed after 7 days, as corroborated
by other studies that show this “prolonged” effect
(37). Overall, the obtained results (AB
and ATP) demonstrated that cells are viable when encapsulated in a
highly concentrated mesh of eSF.
In brief, we obtained an eSF microfluidic platform that can be used for
many biological applications, such as organ-on-chip and point-of-care
diagnostics. As an alternative to PDMS, it can be used in the scale up
of soft microfluidics devices for commercial or industrial purposes
since it is cost-effective, and the stiffness can be tailored to the
model of choice.