For optimum results with automated fluorescent sequencing, plasmid template of sufficient quality and quantity must be supplied. The starting template DNA is the single most important determinant of the quality of the final sequencing data.

The vast majority of our customers have obtained excellent results using the Qiagen* plasmid DNA kits (QiaPrep, Qiagen-tip 20, Qiagen-tip 100, Qiaprep spin, etc.) or the Qiawell plasmid kits. The QiaPrep and Qiawells are the most reliable way of isolating high quality plasmid DNA suitable for automated fluorescent sequencing, and we strongly recommend their use for template DNA purification.

Although the QiaPrep and Qiawell plasmid kits are very reliable, there are some common mistakes which compromise the final quality of sequence data:

• The isopropanol-precipitated DNA is not washed with 70% ethanol to remove excess salt. Wash the DNA pellet at least once as directed with 70% ethanol. Residual salt in the final template will interfere with the activity of Taq polymerase resulting in sequence data which extends less than 200 bases from the primer and exhibits a low signal to noise ratio.
• The template DNA is not dried completely before final resuspension in H2O or TE. To remove residual ethanol, dry the DNA for 5 min. in a properly operating speedvac. If air-drying is preferred, make sure that the DNA is dry (no fluid in the tube, the DNA pellet doesn't look wet). When air drying, a brief 15 min incubation of the open tube at 65 oC is often sufficient to completely dry the DNA. Residual ethanol is very detrimental to Taq cycle sequencing resulting in data with a drastically reduced signal.
• The directions for cell growth are not followed resulting in overloading the Qiagen resin. Usually, a poor yield of plasmid DNA results, presumably due to competition with RNA fragments for binding to the Qiagen resin. Use the recommended quantity of LB broth (don't use Terrificbroth) for cell growth.

In addition to the Qiagen procedure, there are two other recommended methods for consistently preparing high quality template DNA:

• Ultracentrifugation in CsCl density gradients often yields excellent template DNA. Following centrifugation, one must carefully remove residual CsCl from the DNA either by dialysis and/or ethanol precipitation (Cs inhibits Taq polymerase).
• The Wizard (formerly Magic) DNA purification system gives good DNA if extra steps are added to the standard procedure. Adding an extra ethanol precipitation at the end greatly improves the reliability of this method for preparing high quality template DNA. Overall the Wizard method appears to be somewhat less consistent than the Qiagen procedure, but in the hands of an experienced user, roughly 90% of Wizard templates generate good sequence data.

Besides the method of template preparation, the quantity of template is an important determinant of the accuracy and reliability of the final sequence data.

• Too little template results in reactions with little or no signal and poor or no basecalling.
• Too much DNA produces reactions which terminate prematurely, often with less than 250 bases of reliable sequence data.

The optimum amount of plasmid or cosmid DNA for a dRhodamine Taq FS Dye-terminator cycle sequencing reaction is 300-2000ng (see standard reaction guidelines); good data is generated when 0.3 to 2 µg of template is used per reaction. Therefore, the accurate quantitation of template DNA is an important step in the overall sequencing process.

• We recommend that when possible the DNA concentration is determined by spectrophotometry (recall that the absorbance @ 260 nm of a 50 µg/ml solution of ds DNA is 1). When this is not possible, run at least two different amounts of each template on an agarose gel adjacent to a dilution series of a known amount of molecular weight standard (50-500 ng). The template amount should be chosen to fall between the largest and smallest amounts of the standard DNA and within the linear response range for photography of an ethidium bromide-stained agarose gel. Accurate quantitation of high quality template will lead to highly accurate and reliable DNA sequence data: 550 to 750 bases with a 98-99% accuracy.

One final factor which can influence the quality of sequence data is the E. coli host strain used to propagate the plasmid.

• Most commonly used strains, such as HB101, DH1, DH5a, and XL1Blue consistently produce high quality plasmid DNA; some like JM101 demonstrate some variability with respect to template quality. Furthermore, attention to good microbiological practices--using antibiotic selection when propagating plasmids, and streaking out colonies from transformation plates to obtain "pure" single bacterial clones, is beneficial in obtaining good plasmid DNA yields and a homogenous population of plasmid DNA molecules (only one plasmid in the final DNA preparation).

* This does not represent an endorsement of Qiagen products. There are numerous methods and plasmid preparation kits on the market which can give template of excellent quality.