Geoffrey Grove, Ed Alderman
Applied Science and Technology, Caliper Life Sciences, Inc., Hopkinton, MA 01748
Monica Adams, Jason Brennan
Development Integration, Amersham BioSciences, 800 Centennial Avenue, Piscataway, NJ 08855
The TempliPhi DNA Sequencing Template Amplification Kit is designed to eliminate the requirement for extended bacterial growth prior to sequencing. The automation of this kit on the Sciclone ALH 3000 produced 3.0 μg of DNA, which is comparable to the manual technique. Using the DYEnamic ET Terminator Sequencing Kit, Phred20 scores averaging 603 bp across four 96-well plates were observed, with a maximum readlength of 810 bp. IF/THEN programming functions provided by the control software are used to create scalable methods capable of drawing enzyme reagents from various source types. An overview of the basic workflow and software is described.
Amplification of source DNA is a basic building block technique in molecular biology research. The TempliPhi DNA Sequencing Template Amplification Kit employs bacteriophage j29 DNA polymerase applied in an isothermal rolling circle amplification (RCA)1 to produce large amounts of product DNA (1-3 μg)2 and eliminate the need for extended bacterial growth prior to sequencing. Amplification times may vary from 4-18 hours, with increased yields achieved with overnight amplification.
The ability to quickly set up overnight amplification of DNA templates ensures a ready supply of starting materials to feed a variety of research areas. The reaction uses picogram quantities of starting material, and for our work, we used pUC19 control template. The principle concerns for automation included transfers of 0.5 μL of starting template, minimizing cross-contamination by using disposable tips, and temperature control f or TempliPhi Premix, which contains the active enzyme. These conditions were the primary concerns driving the decisions for selection of the automation presented in this note.
In addition, we addressed throughput scalability with a method that allows the user to select whether the Premix source will be a microcentrifuge tube or a column reservoir. Using IF/THEN type method control, at the beginning of the method a user can specify the Premix source container. For low throughput situations, the user may choose to use only a single microcentrifuge tube, while for higher throughput applications the user can specify a column reservoir with a higher capacity. Even higher capacity reservoirs would be simple to include in the method, however, we choose to do all our work in a single temperature controlled reservoir which has both tube and column options available. The enzyme in the Premix is the most costly component of the TempliPhi kit, and temperature control of the Premix is necessary to maintain enzyme activity.
Timing comparison results are presented for 96-well plate format for both tube and reservoir sources of Premix.
Materials and Methods
Liquid Handler (Sciclone ALH 3000 with software version 3.1.14, firmware version 1.14 [CaliperLS]). The Sciclone was equipped with the following:
- Low Volume Head (P/N 103453/0)
- 96 Mandrel Array (P/N 103969/1)
- Press Station (P/N 104826/0)
- Z-8 Pipettor (P/N 102425/C)
- Temperature-Controlled Locator (P/N 103412)
- Temperature-Controlled Locator Block (P/N 103481), includes 3 column reservoirs and holds 12 microcentrifuge tubes.
Thermocycler (MJ Research PTC-200)
Capillary Sequencer (Amersham BioSciences MegaBACE 1000)
96-well PCR microplates (MJ Research Hard-Shell Thin Wall P/N HSP-9601)
Tempo-100 Pipet Tips (P/N 66670)
Tempo 25 μL disposable tips (P/N 104035)
pUC19 DNA control template (United States Biochemicals P/N 76632)
TempliPhi DNA Sequencing Template Amplification Kit (Amersham BioSciences P/N 25-6400-01)
DYEnamic ET Terminator Sequencing Kit (Amersham BioSciences P/N US81095)
PicoGreen dsDNA Quantitation Kit (Molecular Probes P/N P-7589)
Deck Layout: For the TempliPhi Kit, the deck layout is shown in Figure 1. Position A1 is a tipbox, A2 is tip disposal, B2 is the Denature buffer reservoir, C2 is the TempliPhi reaction plate, D2 is the pUC19 DNA source plate, A4 is the temperature-controlled locator containing the Premix in either tube or column reservoir format, and C5 is the Press Station (used for attaching 96 disposable tips to the 96 Mandrel Array). For the DYEnamic Kit, the same layout is used, except positions A2 and B2 are left empty, while the temperature-controlled reservoir at A4 is used to hold the Premix solution and the product plate from the TempliPhi reaction is now used as the source plate in position D2. The Premix contains enzyme, dNTP's, and ddNTP's and is supplied in the DYEnamic Kit. Prior to the run, this was manually mixed with 5 μM primer (M13 Forward) and water.
All Sciclone ALH 3000 movements were performed at 100% travel speed except when entering and exiting liquids. Liquid entry/exit was typically done at an approach speed of 20% and a retraction speed of 5%.
Pipette Tips: Automation Certified Tempo-100 tips were used for all Z- 8 pipetting. For Mandrel Array pipetting, Tempo 25 μL tips were used (P/N 104035).
Plates: For the TempliPhi reaction, the reaction plate and DNA source plates, positions C2 and D2 respectively, were PCR plates. To avoid air bubbles being trapped in the bottom of the "V" shape of PCR wells, when dispensing the 10 μL of Denature buffer, we dispensed 5 μL of the volume very close to the bottom of the well, and then moved up 1 mm and dispensed the remaining volume. The pUC19 source plate was manually pre-loaded with 20 μL of DNA template (2 μg/mL) per well, so the final per well concentration is 1 ng of DNA.
Reservoir: The type of reservoir for holding Denature buffer simply needs to be sufficient to hold the required volume, and be made of a chemically compatible material.
Pipetting technique: The liquid classes used for this work are device dependent. For the Low Volume Head, aspiration and dispense speeds of 0.5 μL/sec were used with a 2 μL pre-sample air gap. For the Z-8, eight independent channel pipetting device, the liquid class settings included no pre or post air-gaps, aspiration and dispense speeds of 2000, Breakoff and Start Speeds of 500, and a backlash setting of 0 as shown in Figure 2. Aspiration and dispense speeds of 3 μL/sec, with a pre-sample air gap of 5 μL, were used for the mixing step in the DYEnamic kit.
The mean DNA template yield per-well was >3.0 μg and the %CV was <4.2 as shown in Figure 3. The smallest volume transfer was 0.5 μL. Results from the sequencing of the TempliPhi reaction are shown in Figure 4.
The results of this feasibility study show that liquid handling capabilities of the Sciclone ALH 3000 are sufficient to produce results that are consistent with the manual results expected for the TempliPhi and DYEnamic kits, which yield ~3.2 g of DNA per well, and readlengths of >600 bp. The use of the same Sciclone ALH 3000 configuration for both of these kits was done with furthe r automation in mind. It is possible that the two kits could be fully automated to run sequentially. However, storage and transport solutions would need to be considered for this system integration. This is another aspect of "scalability", in that the semi-automated system presented here could be integrated into a fully automated system. The MJ Research Hard-ShellThin Wall plates were chosen for their rigidity before and after thermocycling. Plates that maintain constant physical dimensions are crucial for automation.
Travel entry and exit speeds of 20% and 5% were chosen based on observations where rapid entry and exit can cause non-uniform droplet breakoff. Higher entry speeds would likely have little affect, however, exit speeds can have a dramatic affect on precision and accuracy, as evidenced by visual observation of droplet residue left on the outside of pipetting devices.
Tips are ejected immediately prior to loading in our protocols as a matter of programming style. This style takes into account the case where a user has paused or stopped a method with tips left on.
The precision and accuracy of the Sciclone ALH 3000 was sufficient to produce yields comparable to the manual results. Since the well-to-well %CV was <4.2 and the smallest volume transfer was 0.5 μL, it is reasonable to infer that the Sciclone ALH 3000 performed within its operating specifications for the types of liquids used in these kits, that the PCR reaction provides a leveling effect, or both.
Scalable Automation of the TempliPhi DNA Sequencing Template Amplification and DYEnamic ET Terminator Sequencing Kits on the Sciclone ALH 3000, produce data equivalent or superior to historical manual results3. Since all of this work was performed on an "out-of-the-box" system, it is likely that optimization of the system and liquid classes could produce improvements in both yield and throughput.
The use of IF/THEN method control and dialog boxes by the method developer provides the end-user with an environment where all instructions for running the method can be contained within the runtime environment, eliminating the requirement for a paper protocol. Although the improved reliability of automated pipetting is the primary benefit for automating these kits, by using a common layout between each of the kits, greater walk-a-way times are created, with less user intervention required. Furthermore, with the addition of a plate transport and storage device, a fully automated solution could easily be built.
1 Lizardi, P. et. al. 1998. Nat. Genet. 19 pp. 225-232.
2 Nelson J.R. et. al. 2002. Genomics. 80 (6) pp. 691-698.
3 Nelson J.R. et. al. 2002. Biotechniques. June Suppl. pp. 44-47.