To eliminate bottlenecks in experimental precedures and enjoy the higher processibility of ABI3100, it is necessary to accelerate the clean-up step of sequence reaction products. It is also better for us to employ cheaper method. I have tested the following protocol to clean-up the reaction products using Tomy Degital Biology's model EX-125 centrifuge.

The model EX-125 centrifuge can hold four 96-well plates at a time and spin them at 2810 x G, 4C.

To each reaction well:
Add 3uL of NaOAc-EDTA solution (0.5M Na-acetate, 83mM EDTA, pH8).
Add 50uL of ice-cold ethanol.
Seal the plate and invert it once to mix. Make sure that the sequence reaction + ethanol solution comes off from the plate bottom*¹. Inverting too many times will cause leakage.
Spin the plate at 2810 x G, 4C for 80 min (fast acceleration and weak break). The rfc of this condition is equivalent to 15000 x G, 15 min*².
When it finishes, invert the plate onto Kim-Towel to drain (not completely drained).
Place an 8cm x 12cm sheet of Whatman's #17 chromatogram paper (or Advantec's #50), and set the plate with a plate-base onto the paper inversely. Spin it briefly at 50 x G to drain off completely*³.



Add 50uL of ice-chilled 70% ethanol*⁴. Mixing is unnecessary.
Spin the plate at 2810 x G, 4C for 15 min (fast acceleration and weak break).
When it finishes, invert the plate, drain, and spin inversely again as above.

Seal the plate and keep dark and cold until electrophoresis.

After the introduction of the centrifuge for plates, we encountered another inconveniency. At first, we used it for both ethanol precipitation and flush-down of reaction solution on plates. Ethanol precipitation step, however, occupies the machine for 80 min at the longest, and a waiting line appeared with concerns about inactivation of enzymes, etc. We had got to purchase another small table-top centrifuge for plates. Hitachi Koki's himac CT2Y may be the smallest and cheapest for flushing-down plates*⁵.

*¹ Complete mixing is important. Otherwise loss of signals may occur. Vortexing is not enough.
*² 1h may be enough.
*³ Once I made a mistake to spin it at the top speed. Nevertheless it was fine.
*⁴ Because ice-chilled alcohol absorbs humidness in the air, I make a little more conc solution (c.a. 75%).
*⁵ Recently, Kubota released a cheaper machine (PlateSpin). It spins a swing bucket rotor so that deep welled plates or else can be set.