It is definitely best to eliminate red blood cells (RBCs), aka erythrocytes, from your prep before proceeding with culture or cryopreservation steps. RBCs have no nucleus or mitochondria, yet they are fairly metabolically active and consume glucose to generate energy primarily through glycolysis. This not only consumes glucose from the media but produces waste in the forms of pyruvate and lactate, and will ultimately adversely impact the pH of the culture media. A nice overview of the metabolism and functions of RBCs can be found here: http://www.bloodjournal.org/content/106/13/4034?sso-checked=true. For some cell types, like chondrocytes, just coming in direct contact with RBCs will adversely affect both viability and cell-cell adhesion. In addition, RBCs will eventually lyse and release their intracellular contents into the culture media, which includes divalent cations, hemoglobin, and iron – all of which may influence the biology of the culture system in an uncontrolled manner. The RBC cell membranes (or fragments thereof) will also adhere to tissue culture-treated or extracellular matrix-coated surfaces, thus consuming space that would otherwise be available for the primary cells of interest. Lastly, for all of us who favor photomicrographs to assess and document cellular morphology, confluency, etc., it is preferable to have those pesky RBCs eliminated from the picture.
There are many means of removing the RBCs, including lysis before plating with a buffer comprising Ammonium, Calcium, and Potassium (ACK Buffer), which will selectively lyse RBCs while leaving healthy nucleated cells intact. These buffers can be home-made or purchased through most vendors of culture media and related supplies. ACK buffer was developed for human whole blood, to eliminate the RBCs from peripheral blood mononuclear cell (PBMC) preps prior to flow cytometry or other assays, and it may have undesirable effects on primary cell viability or other important attributes, so it should be tested empirically with each primary cell type. For example, a controlled study in 2016 found that in addition to eliminating the majority of RBCs from primary chondrocyte preparations, the ACK buffer also reduced / eliminated apoptotic chondrocytes from the prep, resulting in improved viability and self-assembly for a tissue engineering application.
One of the most common means of eliminating RBCs from primary cell preps is the use of density gradient centrifugation, with standard gradient medias such as Ficoll®, Percoll®, OptiPrep®, or similar. Separation of RBCs can be accomplished concomitant with eliminating debris, dead cells, and/or fractionating the cells comprising a primary preparation based on inherent cell type-specific differences in buoyant density. A benefit of this approach is the availability of ultrapure reagents, it is well tolerated by most cell types, and doesn’t involve the addition of reagents (i.e, beads, antibodies) that are carried forward into subsequent processing steps. The gradient conditions may need optimization for each cell type and selected density gradient medium. A few examples can be found here: http://iccb.med.harvard.edu/files/iccb/files/optiprep.pdf
There are also some newer approaches to consider, including depletion of RBCs using dextran-coated magnetic particles coupled to anti-Glycophorin A antibody complexes, or microbubbles that are engineered to target and bind RBCs in a population and ‘float’ them to the top of the prep where they can be removed. Examples here: