Rudy Willebrords1 and Bart Moyson2
1Johnson & Johnson Pharmaceutical Research & Development, a division of Janssen Pharmaceutica N.V., Laboratory of Virology, Beerse, Belgium
2Caliper Life Sciences Benelux NV, Teralfene, Belgium
The purpose of this paper is to report the performance obtained by dispensing 1μL organic sample (DMSO) containing tartrazine into 384 well plates containing water. Precision and accuracy results are shown for this work.
All testing was performed using a Sciclone Advanced Liquid Handler (ALH) with a 384 Fixed-Cannula Array (Caliper Life Sciences). A Bulk Reagent Dispenser (Caliper Life Sciences) facilitated the continuous delivery of water across the columns of a 384 Well, Flat Bottom, Untreated, Polystyrene Assay plates (COSTAR Catalogue # 9837). Absorbance was measured at two wavelengths using a SPECTRAmax (Molecular Devices Corporation) 2000/384 reader (690 nm is subtracted from the 450 nm).
Tartrazine (Sigma) solution are prepared in dimethyl sulfoxide (DMSO) at 3.0 g/L and dispensed at 1μL.
Qualification of the Bulk Reagent Dispenser
Qualification of the bulk reagent dispenser is done to confirm the accuracy of the volume dispensed. This bulk dispenser is qualified by dispensing volumes of water into a plate, which is then weighed on a balance, and the accuracy calculated. For our work, the Bulk Reagent Dispenser on the ALH (Caliper Life Sciences) was used.
Qualification of the Fixed-Cannula Array
Qualification of the 384 fixed array is done to confirm the accuracy of the volume dispensed. When testing the ALH for 1μL volume dispensing, 20μL water is added to the plates before or after (with chase) the tartrazine dispense. For our work, the 384 fixed-cannula array on the ALH (Caliper Life Sciences) was used.
A gravimetric dispense test is done to determine the accuracy of the bulk (Table 1) and 384 fixed -cannula array (Table 2) dispense operations.
Fixed Cannula Precision Determination (1μL dispensing)
a. Water (20μL) was dispensed into each column of a 384-well plate with the Bulk Dispenser in a separate routine. Starting with a 3μL air gap, Tartrazine (5μL) was aspirated in a larger volume than needed (10μL). While still in the reservoir a 5μL volume was pushed out once. This leaves 5μL in the cannulas and, more importantly, residual on the outside of the cannulas. To control this the cannulas were touched off using a low volume dispense (0.2μL) on the surface of an upside down flat bottom plate followed by a small vertical tip touch. Finally the Tartrazine was consecutive dispensed (1μL) into each quadrant of a 384 well plate with the low volume head.
b. Starting with a 3μL air gap, a chase water volume of 20μL was aspirated followed by a 1μL air gap. Then tartrazine (1μL) was aspirated and finally the whole content of the cannulas was dispensed into a quadrant of a 384 well plate with the low volume head.
Absorbance was measured and the data exported to ExcelTM for CV determination. The results of the dispensed data for fixed cannula dispensing are shown below. The overall CV's that compromise every well of a 384 well plate are shown for each procedure tested. The average %CV per procedure is calculated in each table.
In this study, we demonstrate the high level of precision and accuracy that can be expected of the Sciclone ALH Bulk Reagent Dispenser and the Low Volume Head with the 384 Fixed-Cannula Array. We achieved accuracy levels of 0.03 % for a volume of 50μL of water with the Bulk Dispenser. The dispense precision for the fixed cannulas were 0.52%, 0.22% and 3.07% for volumes of 50μL, 20μL and 1μL respectively. And accuracy levels of less than 1% for volumes of 50μL and 20μL with the low volume head and a 384 fixed cannula array.
The experimental data shows that excellent pipetting performance can be obtained from the Sciclone ALH (low volume head) equipped with the 384 Fixed-Cannula Array. The results surpass the published factory specifications ( CV <= 8%) at 1μL dispense volumes, likely as a result of good pipetting technique. We believe this % CV can even be reduced by using degassed water. We hadn't the opportunity to do so at the test facility.
We have found that residual on the outside of the cannulas becomes the most influential liquid when determining CV's in low volume dispensing. Optimized touch offs in combination with time delays at tip touches and very slow draw backs out of liquids increases remarkably the performance of the pipettor.