After 30 years of having performed hundreds of CFAs using a large number of different mammalian tumor cell lines, the CFAs don't work any longer. We used to seed 300-500 cells per well into a 6-well plate and always had >80% plating efficiency. Now the cloning efficiency is 0% at these cell numbers. I am embarrassed that we can't figure out what's wrong.

The key problem is this: after seeding, the cells don't attach (and therefore don’t proliferate and form colonies). In comparison, when the cells are seeded at much higher cell densities, they attach perfectly well at >95% efficiency. We are working with standard monolayer tumor cell lines: MDA-MB-213; MCF7; B16; T98G; etc. The problem is the same with all cell lines we have tested (and used successfully in the past). For the past year, we have been unable to perform any CFAs with any of these cell lines.

There is an apparent threshold: when we seed any cell number from 100 - 5,000 cells per well (6-well), plating efficiency is 5,000 cells per well, cell attachment (plating efficiency) shoots up to >90% (at 24 hours after seeding, the cells are flattened and firmly attached to the surface of the culture plate, and they start proliferating). But of course, with >5,000 cells per well, you cannot get quantifiable colonies; after a week, you have a dense monolayer and CFAs are impossible.

Over the past year, we tried different plate formats (6-, 12-, 24-wells; 10-cm plates) from different vendors; used more FBS or conditioned medium; pre-coated with Matrigel; used trypsin-free cell collection before seeding; purchased DMEM from different vendors; used more/less medium in the wells; tested for mycoplasma. But we always get the same outcome with each cell line: once cell density during seeding is below about 500 cells per square centimeter, plating efficiency drops to near-zero; above that number, plating efficiency shoots up to close to 100%. Regrettably, under these conditions, CFAs are not possible. Any suggestions what else we could try?