96 square deepwell plates (primarily used for clone-selection with hundreds or thousands of cell lines): Standard conditions for CHO-cells: • 0.5 or 0.6 ml culture volumes • 50 mm shaking amplitude • 300 rpm • sandwich covers CR1296a, CR1396, or CR1396b • clamps CR1700 (enabling quickly taking out and putting back MTPs + covers), or • clamp CR1800 (versatile clamp, ideal if also shallow well plates are used) Further details: Already slightly larger culture volumes (e.g. 0.75 ml) have been reported to give less good results (showing the importance of mixing down to the bottom for CHO-cells, see photo on the right side), as does a shaking diameter of 25 mm instead of 50 mm (see also videos of shaken cultures). An advantage of using shakers with a shaking amplitude of 50 mm is that it offers a "one- size-fits-all" solution for all CHO cell lines. Some cell-lines also thrive well with a 25 mm shaking amplitude (at 425 rpm) or with a 3 mm shaking amplitude (run at 1200 rpm), but some other cell lines appear to be sensitive to the higher shear forces (and increased foaming) associated with these higher shaking frequencies. Another advantage is that shaking at 300 rpm/50mm is mechanically more stable than at 3 mm/1200 rpm. An issue with 3 mm/1200 rpm is often that this high frequency easily gives rise to vibrations and vertical movements. Especially so at the outer sides of the shaker platform (more than in the middle of the shaker platform). This in its turn may cause different results for cultures in plates incubated at the sides of the shaker, in comparison to the centre, which is a nuisance when interpretating the experimental results, and can give rise to false positives or false negatives. However, we need to add here that a number of research groups have reported good results with 1200 rpm/3 mm, especially for transient transfactions, but also in (stable) clone selection projects. Most customers use our sandwich cover CR1296a , second item) for CHO cells. This will result in evaporation losses (at 80% humidity and 35 o C) of approx 5 µl per well per day, so 70 µl in 14 days. Which is acceptable for most projects, though still quite significant in comparison to the 600 ul starting volume. It is of course possible to compensate for these evaporation losses with the feeding regime. By diluting the feed solution (and increasing the feed volume), one can fine-tune this approach to compensate for evaporation, and keep the volume almost constant in the course of the full 10-14 days. This approach - by the way - also makes it possible to decide for the regular sandwich covers (with bigger holes in the silicone layer), such as CR1296 (only for advanced users). The easiest clamp system to use is CR1700 : it is very easy and quick to take the plates + covers in and out, which is relevant if you need to feed the cultures regularly. However, such clamps can only be used with deepwell plates. To be more flexible, approx half of our customers use our universal clamps CR1800, which can also be used for our half square deepwell plates CR1496c (3 stacked on top of each other, using our sandwich covers CR1396b), and Corning 24-roundwell plates in combination with sandwich covers CR1524a (both options: see below). Approx half of our customers use (and re-use after washing) our (detoxified) polypropylene 96-deepwell plates CR1496 or CH1496a, some of them after having tried several off-the-shelf plates and finding no or less growth in these (due to toxic leachables). However, other groups report no toxicity problems with their cell lines when using off-the-shelf plates. This is all dependent on the specific cell line, media, batch of microplates, inoculant size and quality, etc. The plates we supply (CR1496, and CR1496a) have not only been "detoxified" by boiling in alkaline and acid solutions respectively, but also the top of the plates has been flattened and polished. This ensures that the silicone rubber of the sandwich covers hermetically closes off the wells, and "forces" all exchange of headspace air (by passive diffusion) through the hole in the centre. With "off-the-shelve" polypropylene plates, the top surface is often not fully flat (not only dependent on trademark, but also of specific batch), which may - or may not - result in an incomplete closure of some wells. If higher evaporation rates from specific wells (as e.g. from wells A12-H12 from Greiner plates) is observed in practice, this is generally caused by this issue. Note: as an alternative to such polypropylene 96-square deepwell plates, some groups have recently started to use our polystyrene half-deepwell plates CR1496c, at culture volumes of 200-300 µl. If these lower culture volumes are acceptable, there are several advantages associated to their use: i) lower g-forces are required for full mixing: 50 mm/225 rpm and 25 mm/300rpm are sufficient (so the same shaking conditions as for the 24-square deepwell plates), ii) the throughput on one shaker is three times higher (three plates + covers CR1396b can be stacked in one position of clamp CR1800), iii) the plates are made from pure (disposable) polystyrene, and are supplied sterile, and individually wrapped.

24-square deepwell plates (for clone selection with relatively small libraries, and for studying selected clones iin more detail) Standard conditions for CHO-cells: • 2.5-3.5 ml culture volumes • 50 mm shaking amplitude at 225 rpm, or • 25 mm shaking amplitude at 300 rpm • sandwich covers CR1224a, or CR1224c • clamps CR1700 (enabling quickly taking out and putting back MTPs + covers), or • clamp CR1800 (versatile clamp, ideal if also shallow well plates are used) Further details: The use of 24-square deepwell microplates for the cultivation of CHO-cells is much less critical than the use of 96-square deepwell plates as described above. Both 50 mm / 225 rpm, and 25 mm / 300 rpm give good results for culture volumes between 2.5 and 3.5 ml. Also shakers with an amplitude of 12 mm have been reported to give good results (if run at approx 500 rpm). Shakers with an amplitude of 3 mm - however - can not be used for 24 square deepwell plates: the culture fluid can not follow the movement of the shaker, and so called out-of-phase conditions will occur (see photo on the right) Because of the higher culture volume, the results are generally well reproducible, and of course also the amounts of cells and their products (such as antibodies) are larger. Therefore, this plate format is often used for more detailed studies on clones that were initially selected in the 96-well plate format. If the cultures are to be fed every few days, it is advisable to use maximally 2.5 ml culture volume. At larger culture volumes, the culture is too close to the cover (and may even get in direct touch with the cover), and cross-contamination is more likely to occur during the feeding procedure (during taking of the sandwich cover, and putting it back on again).

24 roundwell microplates (e.g. from Corning) Standard conditions for CHO-cells: • 0.75-1 ml culture volumes • 50 mm shaking amplitude at 225 rpm, or • 25 mm shaking amplitude at 300 rpm • sandwich covers CR1524a or CR1324a • clamp CR1800 (three plates + sandwich covers can be stacked, see photo below) Further details: This plate format has been used already for many years for CHO-cells, starting in the time that CHO-cell lines were less robust than that they are at present. In a collaborative project of the UCL in London and Medimmune in Cambridge (UK), the cultivation of CHO-cells was validated and optimized in these plates from Corning and our sandwich covers and clamps (Silk et al, 2010, Biotechnology Letters, 32:73–78). Because of the round well format, orbital shaking gives rise to a gentle mixing (very low shear stress), and so even the most vulnerable animal cell lines can be cultivated in these plates. Another advantage of these plates (in comparison to the square-deepwell microtiter plates as described above) is that the cultures are separated by two walls rather than one, which further reduces the chance of cross- contamination. Furthermore, the 24-roundwell plates from Corning are available in various qualities in terms of binding characteristics of the polystyrene for proteins (TC-treated, ultra-low attachment, untreated, etc.)