Biomedical Engineering Reference
In-Depth Information
64 CHAPTER 4. BIOREACTORS
Spinner flasks and other impeller-based bioreactors are popular because they increase the
mass transfer rate to the cells. However, forming hyaline tissue via an impeller bioreactor is difficult.
Non-uniform mass transfer rates, nutrient and pH gradients, and shear gradients, which cause a non-
uniform mechanical stimulus over the sample, all contribute to inferior tissue formation compared to
other bioreactors [
354
]. Shear force at the surface of the impeller is ten times higher than anywhere
else within the bioreactor [
530
]. Because of this, samples closer to the impeller could experience
injurious levels of shear while samples further away might not be stimulated at all. If positioned
correctly, fibrous capsulation of the construct will be minimal, and the cells will still benefit from
enhanced nutrient transfer. However, low mixing rates or large distances from the shear source
could decrease mass transfer, creating a stagnant environment with increased pH caused by lessened
mixing. For successful use of a stirring bioreactor, a balance has to be obtained between the level of
shear force and the extent of nutrient and oxygen transfer in the media.
Some of the limitations mentioned above have been remedied by modifying the fluid shear
devices to lessen the magnitude of shear and create a more homogenous flow environment. The cone
viscometer, which consists of a small-angled cone that rotates in media above a flat surface seeded
with cells, can achieve a uniform shear distribution with values ranging from 10
−
3
to 10 Pa [
531
].
This type of device has been attractive to researchers because it can apply a laminar shear force at a
constant, controllable level. High-density chondrocyte monolayers exposed to a 1.6 Pa shear force
showed a 2-fold increase in GAG synthesis but also a 10- to 20-fold increase in prostaglandin
E
2
release and 9-fold increase in tissue inhibitor of metalloproteinase mRNA, both of which indicate
an inflammatory response to the fluid shear [
252
]. Additionally, interleukin-6 and nitric oxide levels,
which are reliable indicators of osteoarthritis, increased due to this type of mechanical stimulation,
and chondrocytic gene expression (aggrecan, type II collagen) decreased significantly [
532
,
533
].
These results show that fluid shear, at least when applied to chondrocytes not in a scaffold, might
not be beneficial to chondrogenesis.
Mechanically stirred bioreactors might not be optimal for growing hyaline cartilage, but they
do function well for attaching cells to fibrous mesh scaffolds. Spinner flasks are one of the most
efficient cell-seeding techniques when using pre-formed scaffolds [
525
]. Mixing provides for rapid,
high yield attachment and a more uniform distribution of cells throughout the scaffold, as well as
inducing better overall matrix production in the construct. Static seeding results in cells located
primarily in the lower half of the construct while dynamic seeding distributes cells more evenly
throughout the scaffold [
529
]. If the scaffold material is coated with protein (fibronectin, collagen,
etc.), then dynamic seeding can produce an even higher yield of attachment as well as increasing
migration of the cells into the scaffold [
423
]. Successful tissue engineering results rely heavily on
well-seeded scaffolds, and dynamic seeding provides a relatively simple approach for obtaining high
cell density constructs. However, some polymers, like hydrogels, do not need to be seeded in this
manner and can be distributed evenly at high densities without the use of a mechanically stirred
bioreactor.
