Plasmid vs. Genomic DNA Extraction: The Difference

To isolate plasmid DNA, you crack your cells open and perform a miniprep, trying hard to avoid contaminating genomic DNA. For genomic DNA, you crack your cells open in a different way and try to isolate as much of the contents as possible.

So what’s the difference in the protocols?

In this article, we will look at plasmid and genomic DNA extraction, and the ways in which these techniques differ.

Genomic DNA Extraction:

1. Lysis: Just Crack Them Open

Genomic DNA (gDNA) extraction is the simpler procedure because strong lysis is the only step necessary to release gDNA into solution. For yeast, plants and bacteria, lysis involves enzymatically breaking the strong, rigid cell wall before mechanically disrupting the plasma membrane.

Cell walls are usually digestible with lysozyme, which hydrolyzes cell wall peptidoglycans, and the serine protease proteinase K. For certain gram-positive species, lysostaphin will further aid enzymatic digestion. You may need to use different enzymes for more exotic species with different cell wall compositions.

Mechanical cell wall disruption represents a more universal lysis method for gDNA extraction. Bead beating is popular, and you can easily do this on a vortex using 0.1 mm glass beads or 0.15 mm fine garnet beads. Special vortex adapters help with performing multiple extractions at the same time with equal efficiency. Bead beating is faster than enzymatic lysis and generally more thorough.

For tough filamentous fungi (e.g. Aspergillus and Fusarium spp.), cellular material is often snap-frozen in liquid nitrogen and milled in a pestle and mortar followed by rapid vortexing in solution with an appropriate lysis buffer.

2. … and purify

Following cell lysis which brings the gDNA into solution, the only thing left to do is to purify the sample. You can do this by using phenol-chloroform or spin filter membrane technology with added guanidine salts that promote binding to silica.

3. A few words of advice

Chromosomes will break during purification because they are simply too large to stay intact, and for most applications this is not an issue. For PCR and qPCR, the breakage may actually be advantageous because it aids DNA melting, resulting in more efficient amplification reactions. However, for applications using large DNA fragments, e.g. long read sequencing and long range sequencing, this may be a serious issue. If you do need to isolate ultra-long DNA fragments for these applications, you should consider another setup – for example the recently developed SageHLS Platform from Sage Science.

The E.coli chromosome is just over 4.5 MB in size, amounting to approximately .005 picograms per cell. A typical overnight culture from a single starting colony will contain approximately 1-2×109 cells/ml. Theoretically, that means that 1 ml of culture should yield about 5 µg of gDNA per 109 bacterial cells. Take this into account when calculating how much DNA you need for your chosen application.

Plasmid DNA Extraction

Plasmid DNA extraction is a bit trickier because plasmid DNA must be kept separate from gDNA. This separation is based on size, and good separation relies on using the right lysis method.

1. Alkaline Lysis

For plasmid DNA extraction, the lysis has to be a lot more subtle than simply chewing up the cell wall with enzymes or bashing it with glass beads. Birnboim and Doly invented the (virtually) universal method for plasmid DNA extraction via alkaline lysis in 1979.

The lysis buffer contains sodium hydroxide and SDS, which completely denature plasmid and gDNA (i.e. separating the DNA into single strands). It is critical that this step is performed quickly because excessive denaturing may result in irreversibly denatured plasmid.

Next the sample is neutralized in a potassium acetate solution to renature the plasmid, and this is key to the separation of plasmid and gDNA.

Because plasmids are small, they can easily reanneal forming dsDNA. Genomic DNA, however, is too long to reanneal fully, and instead it tends to tangle so that complimentary strands remain separated.

During centrifugation, gDNA (bound to protein) forms a pellet while plasmid DNA remains soluble. It is key at this step not to vortex or mix the sample vigorously because gDNA breaks easily, and broken gDNA may be small enough to reanneal and go into solution with the plasmid.

2. Purification

Plasmid DNA in the supernatant can then be ethanol precipitated or cleaned up using phenol-chloroform or a spin filter. If you are using spin filter technology, the neutralization buffer will contain guanidine salts so the lysate can bind the silica directly for further washing and elution. The resulting DNA is pure enough for most downstream molecular biology applications. If you need the plasmid for transfection, anion-exchange purification is a better choice to remove contaminating endotoxin. Endotoxin removal is also possible using faster silica-based purification setups.

This method is compatible with mammalian and other eukaryotic plasmids as well as other small extra-chromosomal DNA species. Bear in mind however that plasmid copy number is often much lower in mammalian cells and plant extranuclear organelles.

3. … and some words of advice

Plasmid DNA is typically 3-5 kb depending on insert size. The specific origin of replication present will influence the plasmid copy number per cell. A typical high copy plasmid such as pUC or pBluescript should yield 4-5 µg of DNA per ml of culture.

To isolate high yields of plasmid DNA, use cultures in late log phase or early stationary phase. Prepare cultures using fresh single colonies and fresh selection antibiotic at the right concentration for plasmid maintenance. It is important not to overgrow bacterial cultures as this may result in gDNA contamination in the plasmid extract.

Nowadays, there is a kit for everything, and the net contains vast information about extracting DNA from plasmids, genomes and everything in between!

Further Reading:

Maximizing DNA Yield From Whole Blood.

Birnboim HC, Doly J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7(6):1513-23.

Originally published in 2014, republished in 2017.


  1. Ahmad Mukhtar on February 20, 2017 at 4:28 pm

    When i compare the methodologies of purification of plasmid and bacterial genomic DNA preparation,who can help me to give me some of the reasons for the differences?

  2. annie magnum on September 21, 2016 at 11:53 am

    Great Article. commentary – I loved the points , Does someone know where my business would be able to get access to a sample SSA-632-BK. example to edit ?

  3. Lakshay Sethi on June 18, 2016 at 3:45 am

    The genomic DNA isolation you have explained. Don’t you think it’ll also have plasmid DNA as well. Please correct me if I’m wrong.

  4. Krismyth on June 8, 2016 at 3:48 pm

    Hi Suzanne, I find your information very helpful in understanding basics of DNA extraction. I have a question – While extracting genomic DNA in which step will we get rid of plasmid DNA?

  5. Nasir on October 21, 2015 at 1:33 pm

    It is very nice to read your blog.I have query can we use Alkaline extraction method for Genomic DNA extraction???

  6. Dana Davidovsky on July 13, 2015 at 8:08 am

    Hi all,
    I have a question – I’m transfecting mouse cells with a resistance-carrying plasmid and after selection of 5 days I’m harvesting cells and purifying gDNA using proteinase K and precipitation with isopropanol. My plasmid is not replicated but I want to make sure that I don’t have any residual plasmid in my gDNA prep. Do you know if this method of gDNA purification eliminates all episomal DNA? Maybe I can use DpnI?

  7. SohailRAZA on September 23, 2014 at 3:43 am

    Hello Dear all
    I am extracting the BAC Plasmid (150kb) using alkaline lysis method. When i check the OD of extracted plasmid, it is 20ug/ul. But when i perform restriction enzyme analysis using HIND111 cut gel show just one band while i expect 8-10bands after cut. Can you please help me and suggest me where i have problems in BAC extraction.

  8. amitavagts on July 2, 2011 at 6:17 am

    hi has anyone tried isolation of plasmid dna by alkaline lysis method from mammalian cells transfected with plasmid of interest

    • abo177 on October 30, 2012 at 10:57 pm

      Hi Dr Suzanne,
      I am extracting plasmid DNA from transconjugates of Mupirocin High level resistant staphylococcus aureus, i use CTAB lysis and i use coloform/isopropanol to extract the plasmid DNA.
      My problem is that i am getting smears recently for no reason, I am following all precautions as written in almost all protocols.
      I am getting all figures of smears, upper tailed, global smear, or lower tailed..I am doubting the presence of either genomic DNA, or nucleases at my final prep. so shall i clean up the final TE- plasmid solution by boiling or what do u suggest?


  9. m v on April 4, 2011 at 7:55 pm

    better yet does anyone have a solid recommendation for a metering pump that can be programable without an rs cable or any other interface…just a stand alone pump?

  10. pijaw on March 10, 2011 at 1:15 am

    May i ask something?i recently conducted extraction of plasmid DNA from E.coli and one of the steps was to not vortex to avoid shearing genomic DNA. my question is-how does sheared genomic DNA contaminate the sample?what is wrong with having a sample contaminated by genomic DNA?how does that affect transformation?

    thank you.

    • m v on April 4, 2011 at 7:53 pm

      i dont think the editor or contributor is responding to anymore questions…did u run a gel? it might interfer with the quantification of the target by smearing on a gel. gnomic dna is long and can interfer wit hthe effectiveness of the target…better yet broken gnomic, nicked or open circular dna can decrease your overall percentage of supercoiled or plasmid.

    • Nick on April 5, 2011 at 7:39 pm

      Hi Pijaw

      gDNA contamination should not affect your transformation since none of the genomic pieces should be transformable. gDNA WOULD be a problem if you wanted to use your plasmid for anything PCR-based (since the gDNA would be non-specifically amplified). It would also be a problem if you wanted cut a piece of your plasmid using restriction enzymes, since the enzyme would cut in the genome, creating sticky ends that would interfere with downstream applications.

  11. m v on November 16, 2010 at 3:12 pm

    what exp. do you guys have correlating biomass to plasmid yield? My last results were off the charts…

  12. Aapjeet Aapjeet on October 17, 2010 at 11:44 pm

    very informative. Thanks

  13. Aapjeet Aapjeet on October 17, 2010 at 11:41 pm

    Very informative. Thanks

  14. Chris Grandlic on September 29, 2010 at 5:51 pm

    Hi Suzanne,

    I need to extract large plasmids (and separate them from genomic DNA for sequecing) and I am considering using a CsCl method. Do you have a method that you would recommend?


    • Suzanne on October 9, 2010 at 12:54 am

      Hi Chris,
      I have used the old fashioned cesium chloride preps before but I don’t think I would recommend this. How big is your plasmid? I would try the Qiagen Large Construct Kit if they still sell it. It will be easier and it works well. I used it for 100 kb PI clones and it was much better than pulling plasmids out of cesium gradients. The yields will be low- probably around 0.1-0.2 ug/ml of culture.
      Since the recipes for the anion-exchange kit are published, you could make your own P1, P2, and P3 and then process large volumes of culture, precipitate down the lysate, and then add to a Midi column to maximize the yield.


  15. Suzanne Kennedy on July 23, 2010 at 4:01 pm

    Hi Brian,
    When the culture reaches death phase, the cells will start lysing. This will lead to genomic DNA in the supernatant. If you can, image broken cell walls with DNA attached but hanging free. They are brought down with the bacterial pellet.
    However it is going to shear because it is long and also because of DNases, so you will have pieces of genomic DNA now mixed with the plasmid.
    Normally, the genomic DNA is discarded with the cell walls and protein after you precipitate everything in the third solution- but this requires that the DNA is intact. If you have broken bits of released genomic DNA floating around, it will be in your plasmid supernatant.

    So for plasmid preps, the best yield comes from end of log/early stationary. If you overgrow- either because you inoculated too heavy, grew too long, or used a rich media, you’ll push the culture into death phase too soon and loose your plasmid and have a lot more genomic DNA.


    • stefano.erreni on January 8, 2013 at 1:28 pm

      Hi Suzanne,
      does the fact that death phase is not good for plasmid extraction because of shearing of genomic DNA is true also for a cell culture pelletted and conserved at -20°C for some days?

  16. Brian Cady on July 23, 2010 at 2:07 pm

    “It is important not to overgrow the culture or it may result in genomic DNA contamination in the plasmid prep.”

    Why is this? How does the stage of growth affect isolation of plasmid from gneomic DNA?


  17. Alumi on April 8, 2010 at 5:46 am

    I need to extract genomic DNA from E.coli, but mostly I get a lot of plasmid DNA, too. I followed the protocols from “Current protocols in molecular biology” (unit 2.4), but I couldn’t separate plasmid DNA from genome well. Which step I should notice to prevent plasmid when I extract genomic DNA? Thank you^^

    • Suzanne on April 8, 2010 at 2:32 pm

      Hi Alumin,
      Is this in the absence of antibiotic? You should leave antibiotic out so that the plasmid has not pressure to be maintained.
      When you isolate genomic DNA, you isolate all the DNA in the cell. The best way to get rid of the plasmid is to isolate pure colonies that do not have it. This can be done by streaking colonies on antibiotic-less plates or growing in medium without it.
      If the E.coli strain doesn’t matter and you just want cells without plasmid, you could try buying some competent cells (like XLIblue, DH5Alpha) and then plate them. They will have no plasmid.


  18. dsbreak on February 25, 2010 at 12:49 am

    I noticed that some protocols (Epicentre, MO BIO) include a heating step at 65 degrees during the disruption of yeast cell walls. Any insight/references into how this works?

    • Suzanne on February 27, 2010 at 7:03 am

      The heating step serves to help weaken the cell walls before beating using beads. Heating denatures proteins and disrupts bonding, making it easier to break the cells with shearing forces.

  19. Suzanne on April 17, 2009 at 5:33 am

    Hi Kurt,
    That is a good point to mention. Plasmids (BAC clones) up to 300 kb can be separated from genomic DNA successfully as long as you are very gentle with the mixing so you do not break the clone. I have isolated PI clones that were 150-200 kb on anion-exchange resins and with cesium chloride with good results.
    Only anion-exchange kits (such as Qiagen Maxi kits) can be used for plasmids greater than 50 kb.
    The main issue I had with large constructs is the low copy number and the need to scale up the culture to get a decent yield. Large constructs exist as <10 copies per cell so expect to recover around 0.1-0.2 ug of DNA per ml of culture (so 50 ug / 500 ml).

  20. Kurt on April 17, 2009 at 4:48 am

    The problem comes when your dealing with big plasmids up to 100 kb, that need to be separated from the chromosomal DNA.

Leave a Comment