Biomedical Engineering Reference
In-Depth Information
66 CHAPTER 4. BIOREACTORS
may not be advantageous for creating a similar structure to the native tissue. Variations exist in
direct perfusion designs, but one commonality is a tight fit between the scaffold and walls of the
media chamber. If the scaffold has space around it, less fluid is forced through the pores and uniform
mechanical effects are not achieved. Basically, the chamber becomes a perfusion system and not a
direct perfusion system. A major benefit of this bioreactor is the continuous influx of fresh media to
cells through the thickness of a scaffold. Additionally, perfusion removes the need for manual media
changes, decreasing labor and reducing the risk of contamination.
Modifications can be made to direct perfusion bioreactors to alter the growth environment
of the samples. For example, media that has run through the system can easily be mixed in various
proportions with fresh media. Recycling some of the culture media keeps beneficial proteins secreted
endogenously by the cells i.e., growth factors, matrix molecules) in the system. Another possible
modification to the system involves controlling gas concentrations in the fluid. By adjusting the
oxygen content of the media, researchers can vary the concentration exposed to the cells [
539
]. If
the tissue becomes denser, more oxygen and nutrients can be added to compensate for the increased
oxygen usage. Experiments with variable levels of oxygen tension can also be easily controlled in a
perfusion system.
Direct perfusion of tissue engineered constructs can affect cell proliferation and viability,
matrix secretion, and tissue uniformity. Direct perfusion bioreactors with limited levels of shear
(
<
0.01 Pa) can stimulate cell proliferation and increase the production of proteoglycans and colla-
gen [
540
,
541
]. Cell-seeded scaffolds cultured in a direct perfusion bioreactor running at 1
μ
m/sec
(flow rate of 7.6
μ
L/min) for four weeks showed an increase of 184% in glycosaminoglycans, 155%
in
3
H-proline incorporation, and 118% in DNA content [
538
]. These increases are promising al-
though secreted molecules associated with injury response were not measured. Another research
group looked at applying direct perfusion at a higher linear velocity of 10.9
μ
m/sec (flow rate of
50
μ
L/min) [
459
]. The resulting constructs were composed of 25% (dry weight) glycosaminogly-
cans and 15% (dry weight) type II collagen (the balance being non-degraded polymer). While the
collagen composition is still significantly lower than native cartilage levels (50-73%), the absence of
type I collagen indicates that direct perfusion might be a possible option for growing hyaline car-
tilage. Unfortunately, tissue growth in the scaffold was non-uniform, with more matrix deposition
observed for the scaffold side facing fluid flow. Increasing the flow rate could mediate this problem
since the energy of the fluid flow dissipates as it passes through the scaffold, stimulating cells less as
it progresses through the scaffold. However, increased shear is likely to affect the cells at the surface
negatively. Finding a balance between complete perfusion and low shear forces is necessary if direct
perfusion bioreactors are to be used for cartilage tissue engineering. Rotating bioreactors, discussed
later in this chapter, are a possible solution to this problem.
Significant problems exist with using direct perfusion bioreactors for cartilage engineering.
One is that cellular secretions are non-uniform through the thickness of the construct. Since fluid
flows from one side of a scaffold to the other, the front surface experiences greater mechanical stress
due to the oncoming flow. Conversely, the back surface only experiences the shear stress inside its
