May 1, 2007 (Vol. 27, No. 9)
Acoustic Droplet Ejection Improves Screening by Delivering Accuracy at Small Volumes
The advantages of miniaturization are obvious. Conserving precious enzymes and substrates, reducing reagent cost, and increasing throughput continue to drive researchers toward lower reaction volumes.
However, dealing with nanoliter-range samples without proper liquid-handling equipment has consequences. Nonspecific binding of proteins to pins or nozzles during transfer alters concentration. Insufficient cleaning after any contact liquid transfer can cause cross contamination between samples. Shear forces from air-displacement liquid handling can injure cells. Regardless of the liquid-transfer technique used, poor transfer precision and low accuracy lead to inconsistent results.
To solve these bottleneck issues, non-contact acoustic droplet ejection (ADE) moves nanoliter and picoliter volumes of liquid with sound. ADE has been widely applied to DMSO-based solutions and has been demonstrated to improve screening results in drug discovery. The Labcyte® (www.labcyte.com) Echo® 555 liquid-handling system uses acoustic energy to transfer samples. Without touching the samples, the Echo system ejects 2.5-nL droplets from the sample surface upward onto inverted targets such as assay microplates or slides (Figure 1).
Fifty nanoliters of a wide range of aqueous solutions commonly used in biochemical applications were transferred to test the performance of the Echo system. The precision and accuracy were measured by fluorescence as described by Harris and Mutz in a 2006 JALA article. TRIS buffer and PBS were transferred with less than 2% error (average volume transferred was 49.8 nL for 10-mM TRIS and 49.3 nL for PBS; n=384 for each solution) and less than 2% CV. Cell culture medium DMEM was transferred with 2.78% error (average volume 51.4 nL) and 2.27% CV. Glycerol—a viscous solution that improves spot morphology in protein arrays and enhances low-temperature storage of enzymes and bacterial cell cultures—was tested at 50% concentration. We measured the error to be 1.98% and the precision to be 1.54% CV. Potassium phosphate, frequently used in biochemical assays, was transferred with 1.51% CV at 50 nL. The results indicate that the Echo liquid handler can transfer various solutions without losing precision or accuracy.
TRIS is used in many nucleic acid applications such as PCR and DNA sequencing. We used TRIS as a test solution to check the linearity of acoustic liquid transfers. Four different volumes (5, 10, 25, and 50 nL) of 10-mM TRIS were transferred with an Echo 555 liquid handler. Measured volume was linear with expected volume with a correlation of 0.9999. As shown in the Table, the precision of the transfer of 10-mM TRIS was <2.5% at all measured transfer volumes (n=384 for each volume). Deviation from expected was <2% in all cases. The transfer performance of the Echo is independent to solution volume transferred, as we did not observe decreasing precision or accuracy with different transfer volumes.
Volume transferred with the Echo system is also independent of buffer concentration. We transferred 50 nL of four different concentrations (10, 50, 100, and 200 mM) of TRIS and compared the results. Measured volumes from all four TRIS concentrations showed extremely low deviations from the expected 50 nL. This suggests that there is no dependency of transfer volume accuracy on the concentration of buffer. The transfer precision, as shown in the Table, stayed below 2% CV with 200-mM TRIS. This indicates that the precision of the Echo system is independent of buffer concentration.
In the quest to reduce volumes and increase assay density, there has been a growing tendency to eliminate the wells of a multiwell plate and go directly to arrays. We tested the Echo system to determine whether it could be used to set up low-density arrays.
An array was made by ejecting one 2.5-nL droplet of 50% glycerol in water containing Dylight 649 NHS ester onto a coated slide and then re-ejecting after moving the destination slide 800 µm. The resultant spots, as shown in Figure 2, showed no obvious distortion to the droplet shapes or to the drop alignment. The CV of the spot diameter was 1.78%. Spot spacing was 800 µm with an average 38-µm deviation. Arrays made of 2.5-nL droplets of TE buffer, PBS, PCR master mix, or DMEM also showed good grid alignment with no more than 50-µm deviation in spot spacing.
Because of its precise drop placements, ADE is a reliable technique for spot-on-spot arrays. Cy3 and Cy5 reactive dyes were diluted 1:1000 in neat dimethyl sulfoxide (Cy3-DMSO and Cy5-DMSO); DMSO is a common solvent for drug discovery applications. On an Echo system, we spotted arrays with Cy3-DMSO and Cy5-DMSO at 1,000-µm center-to-center row and column spacing (Figure 3). A 4 x 16 array was spotted with Cy5-DMSO on the left and a 4 x 16 array was spotted with Cy3-DMSO on the right, overlapping in the center eight columns (yellow spots.) Each single spot contains a 2.5-nL droplet, and each overlapping spot contains two 2.5-nL droplets.
ADE transfers a wide variety of aqueous solutions such as PBS, PCR master mix, DMEM, and TRIS buffer with high precision and accuracy. ADE can be used to transfer viscous solutions, such as those containing glycerol, which can be difficult to transfer with pipettes, nozzles, and pin tools. ADE reduces cross contamination to zero since nothing contacts the solutions. Transfer performance on the Echo 555 liquid handler was independent of transfer volume and buffer concentration.
With its excellent precision and accuracy in nanoliter liquid handling, ADE is a proper choice for miniaturization in aqueous applications, such as enzymatic assays, protein arrays, PCR, live cell transfers, and DNA sequencing.
Jean Shieh is applications specialist, Siobhan Pickett is director of marketing, and Joseph Olechno is vp of marketing at Labcyte. Phone: (408) 747-2000. Website: www.labcyte.com.
E-mail: [email protected].