Neil Duldulao, Bio-Rad Laboratories, 2000 Alfred Nobel Drive, Hercules, CA 94547 USA
E. coli bacteria have long been used as hosts for plasmid propagation. Several methods have been used for preparing the plasmid, including alkaline lysis of the bacteria and alcohol precipitation of the DNA (Birnboim and Doly 1979). Once DNA had been observed to bind to glass (silicon dioxide) in the presence of chaotropic agents such as sodium iodide and sodium perchlorate, plasmid purification using glass powders and diatomaceous earth began (Vogelstein and Gillespie 1979, Yang et al. 1979, Marko et al. 1982). Typically, plasmid purification kits use alkaline lysis followed by guanidinium salts to bind the plasmid DNA to glass membranes. However, genomics laboratories with automated fluorescence-based sequencing systems have created a demand for high-throughput plasmid purification in a 96-well format. Accordingly, high-throughput plasmid purification kits must take into account the following criteria: sample purity, mass yield, concentration, extent of handling, and cost per sample. The plasmid samples must also give useful sequencing data from an automated fluorescent DNA sequencer. A total of six 96-well plasmid purification kits from various manufacturers were assessed using these criteria (Table 1).
E. coli strain DH5α (Invitrogen) containing plasmid pCMV (Stratagene) was streaked out on an LB-ampicillin (100 g/ml) agar plate and incubated at 37C overnight. The following day, 20 ml LB-ampicillin medium in a 50 ml centrifuge tube was inoculated with one bacterial colony from the agar plate and incubated overnight at 37C with shaking at 250 rpm. LB-ampicillin medium (500 ml) supplemented with 2% glycerol was then inoculated with 500 l of the starter culture and incubated overnight at 37C with shaking at 250 rpm. Culture density was measured by taking an absorbance reading of a 20-fold dilution of the culture at λ = 600 nm.
Samples of a 4 OD culture were transferred to each well in columns 16 of six 96-well grow blocks. Similarly, samples of a 10 OD culture were transferred to each well in columns 712. The grow blocks were then centrifuged at 1,500 x g for 10 min, the medium decanted, and the grow blocks blotted on paper towels to remove residual medium. Plasmid pCMV was subsequently purified from the bacteria using a different manufacturers 96-well plasmid purification kit for each grow block and following each kits protocol. For all kits, the plasmid DNA was eluted or resuspended in 80 l of the appropriate elution buffer. Plasmid DNA concentrations and quality were measured using UV spectrophotometry. All samples from the 4 OD culture were run on agarose gels to confirm purity and concentration of the plasmid. For each manufacturers kit, 5 of the samples purified from the 4 OD culture were sequenced on an Applera ABI PRISM 3100 capillary array sequencer, and the resulting electropherograms were analyzed for ambiguous counts within the first 600 bases to assess relative comparative read lengths.
Plasmid Yields and Concentrations
With respect to overall plasmid DNA mass yield, the Aurum plasmid 96 kit was one of the 2 best performers, giving the highest yield of all kits with the 4 OD culture, and second only to Wizard SV 96 with the 10 OD culture (Figure 1A). Aurum plasmid 96 also rendered the most concentrated samples of all kits from the 4 OD culture, and slightly less concentrated samples than those of QIAprep 96 Turbo and Wizard SV 96 from the 10 OD culture. Complete concentration performances of all 6 kits are shown in Figure 1B.
Most average A260/A280 ratios for the 6 kits were between 1.75 and 1.85; exceptions included the ratio for the plasmid purified with Montage from the 10 OD culture, and the ratios for Concert(96) samples, which were both inordinately low. Aurum plasmid 96 had average A260/A280 ratios of 1.84 and 1.83 for 4 and 10 OD cultures, respectively. Table 2 lists the average A260/A280 ratios for all 6 kits and for both bacterial densities processed.
Agarose Gel Electrophoresis
When plasmid samples isolated from the 4 OD culture were run on an agarose gel, Aurum plasmid 96 samples were confirmed to be more concentrated than those purified with Concert(96), Perfectprep, or Montage (Figure 2). One lane containing a Concert(96) sample produced significant UV fluorescence in the well, indicating possible genomic DNA contamination. There was no indication of RNA contamination or degradation of plasmid in any sample run on a gel.
ABI PRISM 3100 DNA
Sequencing All kits produced DNA with sequencing reads of at least 600 bases without any ambiguities. Aurum plasmid 96 sequencing reads (Figure 3) had among the longest usable read lengths.
Aurum Plasmid 96 Kit
Giving the best plasmid yields and concentrations from the 4 OD culture of all kits, the Aurum plasmid 96 kit also had the handling advantages of the binary plate configuration, in which both lysate filtration and plasmid binding plates remain outside the vacuum manifold chamber. When lysate filtration was complete, the lysate filtration plate was simply detached and discarded, without having to disassemble the manifold. Moreover, no extra time was spent binding the plasmid to the plasmid binding plate; lysate filtration and DNA binding were accomplished simultaneously. Since lysate filtration took less than 5 min and washing lasted less than 1 min, the entire plasmid purification was completed within 40 min. Sequencing electropherograms were of excellent quality.
QIAprep 96 Turbo Miniprep Kit
The QIAprep 96 Turbo purification plates had one of the fastest rates of liquid passage through the plates of all kits tested, less than 1 min each for the filtration and 2 wash steps. User handling involved with this kit, however, was much more extensive than with Aurum plasmid 96, because of the QIAprep kits extra steps (e.g., washes) and the need to place the DNA binding plate inside the vacuum manifold. Repositioning of the DNA binding plate required that the manifold be disassembled and then reassembled, increasing handling. As a result, QIAprep required about 50 min to purify 96 plasmid DNA samples, compared to 40 min for Aurum. QIAprep samples were of sufficient concentration and purity to give good sequencing results, but purifying these samples was costly: QIAprep 96 Turbo cost $1.90 per prep, compared to $1.56 per prep for Aurum*. QIAprep delivered quality sequencing electropherograms, but Aurum plasmid 96 provided the same sequencing quality in less time, with fewer handling steps, and at a lower cost.
Wizard SV 96 Plasmid DNA Purification System
One of the best performers in plasmid yield from the 10 OD culture, the Wizard SV 96 kit did not perform as well as Aurum plasmid 96 with respect to average yield and concentration for the 4 OD culture, the density of bacteria obtainable in 96-well grow blocks. As with Aurum plasmid 96, Wizard SV 96 filtration and DNA binding were accomplished simultaneously, saving time. However, the required extra steps (e.g., second neutralization and manifold disassembly/reassembly) and longer incubations increased Wizard SV 96 prep time to 52 min compared to 40 min for Aurum. The Wizard SV 96 cost was $1.93 per prep. Aurum was faster and much less expensive than Wizard SV 96 while still giving high-quality plasmid DNA sequencing results.
Perfectprep Plasmid 96 Vac, Direct Bind Purification System
Aurum plasmid 96 gave higher average yields and concentrations than did Perfectprep for both 4 and 10 OD cultures, giving almost twice as much plasmid yield at more than twice the concentration. Because the Perfectprep kit had difficulty filtering the 10 OD samples (i.e., wells clogged), plasmid yields from these culture samples were lower than anticipated. Although Perfectprep gave reasonable sequencing electropherograms, prep time was longer than for Aurum (50 min vs. 40 min). Additionally Perfectprep kit handling was greater, since the plasmid binding plate needed to be placed inside the vacuum manifold.
Montage Plasmid Miniprep96 Kit
Among the other plasmid purification kits tested, Montage average yields from both 4 and 10 OD cultures were most similar to those of Aurum plasmid 96. However, Montage average concentrations were only about 60% of Aurum averages. Furthermore, the Montage protocol took about 80 min to complete, twice as long as the Aurum prep time. The protracted sample prep time for Montage was primarily due to an extremely slow plasmid binding step; wells containing lysates from 4 OD culture samples took 16 min to pass through the plasmid binding plate, as opposed to the 7 min indicated in the manual. Washing the plasmid binding plate required extra time as well. Plasmid recovery consisted of a manual transfer of resuspended DNA into a 96-well microplate via micropipet, as required in the protocol. The extent of handling in this step was consequently greater than in vacuum elutions.
The Concert(96) kit offered the user the opportunity to perform the lysis without centrifuging the bacteria beforehand (the direct load method), saving time. Any time saved, however, was lost by the lengthy centrifugations required to filter the plasmid samples, extending the plasmid preparation time to 70 min. Although plasmid sequencing quality was good, the lysate filtration and washes were unsatisfactory. Visible precipitates were observed in approximately 30 of 96 resuspended plasmid samples. The aberrant fluorescence observed on the agarose gel (Figure 2A) was probably caused by such precipitates. The low average A260/A280 ratios obtained with Concert(96) (ratios of 1.44 and 1.23) gave further evidence of sample impurity.
Obtaining high-quality DNA sequencing results is typically the primary criterion for choosing a certain 96-well plasmid purification kit. When all kits seem to satisfy this requirement, other considerations come into play, such as plasmid purity and yield, usage requirements (handling and time), and cost. Sufficiently concentrated template is also an important requirement for some applications such as cycle sequencing. No other 96-well plasmid purification kit performed as quickly as Aurum plasmid 96, and Aurum was one of the 2 best performing kits in plasmid yield and concentration. In addition, Aurum requires a minimum of handling with its innovative binary plate configuration. No other kit produced better sequencing data than did Aurum plasmid 96. And with a cost of $1.56 per prep, Aurum plasmid 96 represents a cost-effective choice for high-throughput plasmid purification.
ABI PRISM 3100 DNA sequencing was performed by Davis Sequencing, LLC, Davis, CA, USA.
Birnboim HC and Doly J, A rapid alkaline extraction procedure for screening recombinant plasmid DNA, Nucleic Acids Res 7, 15131523 (1979)
Marko MA et al., A procedure for the large-scale isolation of highly purified plasmid DNA using alkaline extraction and binding to glass powder, Anal Biochem 121, 382387 (1982)
Vogelstein B and Gillespie D, Preparative and analytical purification of DNA from agarose, Proc Natl Acad Sci USA 76, 615619 (1979)
Yang RCA et al., Elution of DNA from agarose gels after electrophoresis, Methods Enzymol 68, 176182 (1979)
* Costs based on 2001 US list prices.
back to top