(0%, 30%, 50%, 70%, 80%, 90%, 100%) by submerging samples

(2
5 min) at each progressive concentration.

5. Air-dry, sputter-coat with 10 nm gold, and image bioreactor

surfaces using scanning electron microscopy (SEM) (Fig. 3f).

We have found that 3D morphology differs greatly with

cell type.

3.8

EV-Associated

RNA

EV-associated RNA can be isolated by resuspending large or small

EV pellets in 50–100 μL PBS and adding TRIzol, or by using other

RNA purification reagents [9].

4

Notes

1. There is a variant of the CELLine flask available for suspension

cells which requires passaging to maintain an appropriate cell

number. This protocol has been optimized for adherent

cells only.

2. Media choices will depend on the type of cells used. We recom-

mend using commercially available defined “advanced media”

formulations that contain serum replacements when possible to

support long-term, high density growth with minimal serum

usage, but some cells may perform better in standard media.

Media formulations should be tested in conventional cultures

using prior to inoculation.

3. The use of CDM-HD or other serum replacement is optional,

and cells should be tested prior to bioreactor inoculation to

determine if it affects their viability. We have found that

CDM-HD offers great benefits with several different cell lines

in maintaining cell viability and reducing serum usage.

4. Initially seeding adherent cells with 10% serum then weaning

them off serum gradually seems to help them initially adhere to

the fibrous polyethylene terephthalate (PET) growth surface.

5. When handling the bioreactor, it is important to do everything

gently as the 10 kDa membrane is only 8 μm thick and is the

most frequent cause of the bioreactor’s failure.

6. Media adaptation will be highly cell-dependent, with some cells

able to be adapted to their final media prior to inoculation or

adapted more quickly once inoculated. We recommend slow

adaptation in most cases.

7. It is common to retrieve less than 15 mL from the cell chamber

due to osmotic gradients, do not apply excessive suction with a

serological pipette if less than 15 mL is collected because there

is a risk of rupturing the membrane. Repeatedly retrieving

more than 20 mL from the cell chamber is a sign that the

190

Anastasiia Artuyants et al.