Discussion

The plasmid copy number assay provides important information on gene dosage，gene regulation，and protein expression.Quantitation of the plasmid copy number in recombinant protein production strains is also required by drug regulatory agencies.Current plasmid copy number assays described in the literature are mainly divided into nucleic acid hybridizationbased assays and agarose gel electrophoresis-based assays.

Nucleic acid hybridization-based assays include Southern blot hybridization［3，12，13］，and the Threshold sequence-specific assay［14］.In Southern blot hybridization the use of radioactively labeled probes［12］represents a safety problem.Some authors have employed nonradioactive labels，such as digoxigenin［3］.Generally，the Southern blot hybridization procedure includes digestion of plasmid DNA with a restriction enzyme，separation of plasmid fragment and genomic DNA by agarose gel electrophoresis，and blot DNA transfer to the membrane，followed by hybridization with labeled specific probes.The procedure is evidently laborious and time-consuming.Furthermore，it is a semiquantitative assay that has poor reproducibility.Some investigators have recently reported a plasmid copy number determination method using the Threshold sequence-specific assay［14］.It was reported as a sensitive，specific，and accurate assay，but the assay is complicated and also very costly.

The agarose gel electrophoresis-based assays include purification of either total bacterial（genomic and plasmid）DNA［4，15，16］or plasmid DNA only［7，17，18］，followed by agarose gel electrophoresis and quantitation of DNA by densitometry.If the genomic DNA was not included in the assay as an internal DNA recovery control，the authors had to mix the standard transformant cells into the experimental cell culture to monitor the recovery of plasmid DNA［17，18］.This makes the assay more complicated.In cases where both genomic and plasmid DNAs were taken into analysis，the authors have run the agarose gel electrophoresis of undiluted sample only，followed by scanning either of negatives of the gel photographs［4］or of ethidium bromide-stained gels directly［15］.Sometimes an additional 1∶10 dilution of the sample was analyzed to account for the difference in chromosomal and plasmid DNA concentrations［16］.However，our experience shows that the linear ranges for genomic and plasmid DNA peak areas lie within different dilution factors of the samples.Moreover，the linear range dilution factor for plasmid DNA varies depending on the particular plasmid form（linear，nicked circular，or supercoiled）and the plasmid copy number of the given sample.As a rule，in a particular sample dilution the genomic DNA band is more saturated than are the plasmid DNA bands.Therefore，the plasmid copy number is well overestimated if the genomic and plasmid DNAs are quantitated at the same dilution number.

To overcome the described problems we have developed an easy and simple plasmid copy number assay.This assay combines isolation of total bacterial DNA，separation ofgenomic and plasmid DNA by agarose gel electrophoresis，and calculation of the plasmid copy number in a linear range after scanning the negative of an ethidium bromide-stained gel.The validation results have demonstrated high precision，accuracy，and sensitivity of the assay.The unique feature of our assay，compared to similar published approaches，is the quantitation of plasmid and genomic DNAs in linear range of their respective peak areas.This is achieved by running a series of twofold dilutions of total DNA samples，and applying strict data reduction rules to select for the reliable peak area values in the linear range，which are subsequently used for calculation of the final plasmid copy number.(https://www.daowen.com)

Another advantage of this assay is simultaneous isolation，separation，and quantitation of genomic and plasmid DNA.The plasmid copy number is calculated based on the ratio and sizes of genomic DNA and plasmid DNA.Here，genomic DNA is used as an internal control.This avoids an error caused by efficiency in cell lysis and increases the assay accuracy.Since the assay is based on the ratio of genomic DNA and plasmid DNA，it also greatly increases the assay sensitivity.The high sensitivity of a single copy of plasmid per genome was achieved not because of the sensitivity of detection by ethidium bromide staining，but rather because more cells could be used in the assay.

In conclusion，we believe that our easy and accurate assay for quantitation of bacterial plasmid copy numbers can be successfully applied to any other organism，with respective modifications in total DNA isolation procedure and data reduction rules.