Pimp Your Plasmid Growth Medium

I often wonder why it is that molecular biology researchers stubbornly refuse to change 4o-year old methods that, while work, are not as good as newer, faster and cheaper methods out there.

I suppose rational scientists often have irrational superstitions.

One example of an old method that could be improved is the growth media used for plasmid preparation.

The majority of us, throughout our university careers, have used either SOC, LB or TB, for recombinant plasmid propagation, typically in E. coli. LB or Luria-Bertani broth has been in use for almost 60 years or thereabouts, while SOC has certainly been in use for 2 decades.

Pimp Your media and Save Some Money!

But by adding in a few more ingredients or being more economical on others (especially yeast extract and tryptone) you could get a higher plasmid yield, quicker and with less money.

To counter the naysayers, nobody wants to make very complex media with 15 ingredients requiring filter sterilisation, as this obviously defeats the object of economy of time and budget. Indeed, there are trade-offs between optimising for biomass, plasmid yield, quality, stability and cost with the difference between protein production and plasmid production, since plasmid production requires only cell growth, division, and plasmid stability.

The good news is that Michael Danquah and Gareth Forde from Monash University down-under have devised a stoichiometrically optimised medium for plasmid production. PDM, supposedly yields under the conditions they tested, twice the amount of plasmid in both volumetric and specific yields compared to TB, and LB is left in the dust. Better yet, because it uses less tryptone and yeast extract, the cost per mg of DNA is roughly one-quarter of the cost of LB.

The recipes for LB, TB, SOC and PDM are shown below. If you decide to break with tradition and give PDM a go, be sure to tell us how it goes.


Note: Autoclave glucose, KH2PO4 and Na2HPO4 separately – if you want to know why check out Doesn’t Play Well with Others – The Chemistry of the Autoclave.

Originally Published 28 April 2008, Updated and Republished 19 December 2014.


  1. Shyam Bhakta on June 18, 2017 at 9:52 am

    LB with 30 g/L yeast extract (instead of 5 g/L) exceeds TB’s plasmid yield. The glycerol and phosphate-buffering of TB don’t matter. Source: http://www.sciencedirect.com/science/article/pii/S0167701216303591

    TB was second-best to PDM(R) in cost effectiveness according to the Danquah paper. LB with 24 or 30 g/L yeast extract would thus be at least more cost-effective than TB due to fewer components, maybe even as much as PDMR.

  2. Uma on February 24, 2016 at 11:02 am

    In my experience, studier media recipes are the best for protein expression and also plasmid yields. I suggest try them and check for yourself.

    Not all media recopies give same yield of plasmid, they vary because of the titre of cells. In addition, be cautious of limitations with plasmid purification kits, they are optimized with LB as reference. Generally, use less media when changing from LB and try for plasmid extraction. As a rule, we use studier media at 2ml for plasmid extraction with comparable yields with 5 – 10 ml of LB..

    In found Macherey Nagel plasmid extraction kits give high yields probably because of high binding capacity.

  3. Good Citizen on November 27, 2015 at 7:03 pm

    Tried the optimised “R” version of the PDM media with C/N ratio 2.8 and the yield was half of LB! This is consistent with what has been reported by a subset of users above using the non-optimised C/N ratio version of PDM. I tried it with Top-10 cells, shaking at 30 degrees and also with 37 degrees with same results. Ill give it a go again with DH5 cells to check if its any good or just baloney! FYI all reagents were from Sigma.

    • Stefy on November 21, 2016 at 12:43 am

      Hey I know it’s been a year since this post, but did you figure out the lower yields? Why? and were you able to fix it?
      For my part I will report that I think the calculations for the C N ratio to accomplish the PDMR are to add 5.454g of NH4Cl (~1.4g N) for the 10g of glucose (~4g of carbon in there) = 2.8. I did calculate to more sig figs according to my balance.

      Liam, (or anyone) did you try the PDMR to better/ worse success than the PDM?

  4. JackBean on September 16, 2013 at 6:14 pm

    What if I need a low-salt version of media (for zeocin antibiotic), what should I avoid in the medium?

  5. Ronny Ewanek on February 11, 2011 at 3:50 pm

    Some questions about this:

    Why can’t you mix everything and autoclave? The only thing I read on here was that it “doesn’t autoclave well.” What does this mean?

    How does one make the stock solutions (10X) when it’s in gram amounts? Convert to molar?

    Does the pH need to be altered in order for the Na2HP04 to dissolve?

    Sorry I’m a bit lost on how to go about making this 🙁

    • SeanMc on June 2, 2011 at 6:44 pm

      I see this question came in a while back, so it may be moot now, but if you are in a quandry as to how to make this, why don’t you just order it from Teknova? They have catalog # P0820-06, and product should normally ship out within a couple of days or so. Hope that helps.

  6. Lilian Guibert on February 4, 2011 at 5:32 pm

    Hello! We tried PDM medium with 6 clones which had low plasmid yields (less than 50 ng/ul) when grown in LB medium, but unfortunately we obtained similar or lower yields 🙁 . We used Qiagen´s Plasmid purification kit

  7. poorpostdoc on January 28, 2011 at 10:36 pm

    Has anyone tried to modify this for protein expression?  I expect there to be some issues with the glucose concentration if you’re using an ara-driven protein expression vector.  What about using glycerol as the carbon source?

  8. Mad-Scientist on December 17, 2010 at 6:36 pm

    This is my common protocol for miniprep

    one colony in 5 ml of LB +antibiotic
    Growth overnight
    spin down
    Qiagen miniprep kit
    elution in water 50 uL

    I tried PDM and the result is decrease by two of the DNA cocnentration.
    Have you any idea about That

  9. Denis Chaix on December 17, 2010 at 4:52 am

    I have red all the good comment about this medium.
    I have tried this PDM medium for miniprep and I was really desapointed, my DNA concentration was divide by two;I made the mistake of mix all the ingredients together and after that sterilize the liter of media.
    Have you any ideas or protocols to fix this bad rates of DNA purification.
    What is the good way to use this media lease it’s really important for me.

  10. Catherine Kibirige on July 28, 2010 at 9:11 pm

    Hey Sean!! It’s Catherine from the Margolick lab!! Thought I recognized the dreads… Just sent a “friend request”. Haven’t tried the media I’m afraid…

  11. sean evans on July 27, 2010 at 12:03 pm

    hi there,

    anyone try this media for recombinant protein expression? if so, how did it fare compared with lb?


  12. Benjamín Erranz on May 12, 2010 at 4:00 am

    in the Michael Danquah and Gareth Forde`s paper there is a medium referred to as PDMR, better than PDM, that is the PDM with optimized C:N ratio through variations in the concentration of NH4Cl. In the paper don`t appear the recipe for PDMR, or the C:N ratio of the PDM to infer the addition of NH4Cl
    Someone have a hint about this?

  13. Liam on November 17, 2009 at 3:40 pm

    Hi guys

    Yes, sometimes something does not dissolve or precipitates out, and I am not sure which component it is. I suspect it is the phosphate salts, and so I generally cool everything to room temperature or lower before combining, and keep a stirrer in the bottle to make sure everything mixes properly. It seems to happen more often with agar plates, so I cool everything in a 55oC waterbath before combining, although the buffer and glucose is alot cooler than that by the time it is added.

    • ghorvath on June 15, 2012 at 5:32 pm

      Ammonium chloride, potassium phosphate and the Magnesium salt react together and form Ammonium Magnesium Phosphate which is slightly soluble in water. The concentration of the Ammonium Magnesium Phosphate is near its solubility point (0.01%) and could percipitate.

      The same could happen with carbonate dissolved in the water and magnesium salt.

  14. comp3v on November 2, 2009 at 5:12 am

    Can anybody comment about purity of the plasmid? I mean, all plasmid preparation protocols are adjusted for defined portion of bacterial mass, and if this media increase amount of bacterial cells – one will get excess of material => incomplete lysis etc…

  15. BioKyle on October 30, 2009 at 7:17 pm

    I made a 10x solution of KH2PO4 + Na2HPO4, 10x Glucose, and a 1.2x of the rest. I autoclaved them separately, but when I mixed together something salted out and won’t dissolve, even upon boiling. I’ll repeat to once more to make sure I didn’t screw up the weighing, but this should work right?

  16. Gokay on September 14, 2009 at 4:58 pm

    Simple idea behind PDM is that the chemicals in the media “react” in the cell to form complex biostructures. They calculated the optimal concentration of each ingredient that would give you a high biomass and lots of plasmid. When you are isolating plasmid, you want your cells to divide like crazy without worrying about spending their energy on transcription or translation, obviously aside from the required amount to sustain cell growth. Hence, PDM would probably suck for protein isolation, although LB or TB can be used for this purpose because of the fact that they are not optimized for constant cell division and low protein synthesis.

    About the autoclaving: Glucose will not brown unless heated with amino acids (called a Maillard reaction) as Liam said. Phosphates should be autoclaved separately as usual.

    With this medium, I have successfully isolated 60-70 mg pure high-copy plasmid from 1L cultures (no kits involved).

  17. Liam on July 13, 2009 at 8:06 pm

    Nice, I am glad it worked

  18. CH on July 13, 2009 at 6:02 pm

    I tried this medium and got a whopping 6.5x increase in plasmid DNA production compared to LB! Highly recommended!!

  19. Liam on February 19, 2009 at 1:46 pm

    Sorry, that’s meant to read “without”

  20. Liam on February 19, 2009 at 1:34 pm

    Sure you can filter glucose, but as long as it is autoclaved with yeast extract or another amino acid source, there will be no caramel, I’ve done it plenty times and never had a hassle.

  21. matt on February 19, 2009 at 12:30 pm

    One point : Glucose + autoclave = Caramel!

    You should always filter sterilise glucose.

  22. CK on June 24, 2008 at 4:08 pm

    Looks promising based on the feedback here. The lazier option: inoculating your LB overnight 3 hours earlier than usual? 😉

  23. Paulo on May 30, 2008 at 5:39 pm

    Normal growth time for LB is about 15-17.5 hours. In the paper presented by Danquah and Forde, it displays 12 hour growth as approximately the best growth time when using PDM. Is this what others have used?

  24. Liam on May 7, 2008 at 11:16 am

    Thanks Sudheer, you pimp that media.


  25. Sudheer Reddy on May 7, 2008 at 9:22 am

    Hi to all,

    I tried with the above mentioned PDM media to grow my plasmid. Trust me, this media will really work very efficiently for the maximum production of cell density.

    Good work!!!!!!!!!!!!!!1 Keep it up

  26. Liam on May 5, 2008 at 6:31 am


    As I understand it, it allows greater cell densities to be reached, and the buffering and salts allow for greater plasmid stability, the glucose allows for quicker cell growth. Apart from that, I don’t think there is much more,

  27. Liam on May 5, 2008 at 6:21 am

    Hi Paulo

    The medium is for 1L of medium.

  28. Paulo on May 1, 2008 at 4:08 pm

    How much of the PDM podwer should be added to 1 liter of diH2O?

  29. CK on April 30, 2008 at 10:08 am

    Not sure what the idea is here. Is the medium facilitating more rapid E. coli growth, resulting in greater cell density and therefore higher plasmid yield?

  30. Liam on April 30, 2008 at 6:50 am


    You sterilise everything by autoclaving, but the glucose in a separate bottle, and the phosphate buffers in a separate bottle. When combining them all together, they don’t autoclave so well, but separately they’re fine.

    Good luck

  31. Ivonne on April 29, 2008 at 3:32 pm

    Glucose, KH2PO4 and Na2HPO4 are sterilized by autoclave.
    Have I to autoclave the others ingredients, or by filtration?

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.