Established in the mid 1970's, New England Biolabs, Inc. (NEB) is the industry leader in the discovery and production of enzymes for molecular biology applications and now offers the largest selection of recombinant and native enzymes for genomic research. NEB continues to expand its product offerings into areas related to PCR, gene expression, sample preparation for next generation sequencing, synthetic biology, glycobiology, epigenetics and RNA analysis. Additionally, NEB is focused on strengthening alliances that enable new technologies to reach key market sectors, including molecular diagnostics development. New England Biolabs is a privately held company, headquartered in Ipswich, MA, and has extensive worldwide distribution through a network of exclusive distributors, agents and seven subsidiaries located in Canada, China, France, Germany, Japan, Singapore and the UK. For more information about New England Biolabs visit neb.

The Basics: How Alkaline Lysis Works

Content brought to you by New England Biolabs

plasmid-prep.jpgAlkaline lysis was first described by Birnboim and Doly in 1979 (Nucleic Acids Res. 7, 1513-1523) and has, with a few modifications, been the preferred method for plasmid DNA extraction from bacteria ever since. The easiest way to describe how alkaline lysis works is to go through the procedure and explain each step, so here goes.

1. Cell Growth and Harvesting

The procedure starts with the growth of the bacterial cell culture harboring your plasmid. When sufficient growth has been achieved, the cells are pelleted by centrifugation to remove them from the growth medium.

2. Re-suspension

The pellet is then re-suspended in a solution (normally called solution I, or similar in the kits) containing Tris, EDTA, glucose and RNase A. Divalent cations (Mg2+, Ca2+) are essential for DNase activity and the integrity of the bacterial cell wall. EDTA chelates divalent cations in the solution preventing DNases from damaging the plasmid and also helps by destabilizing the cell wall. Glucose maintains the osmotic pressure so the cells don’t burst and RNase A is included to degrade cellular RNA when the cells are lysed.

3. Lysis

The lysis buffer (aka solution 2) contains sodium hydroxide (NaOH) and the detergent Sodium Dodecyl (lauryl) Sulfate (SDS). SDS is there to solubilize the cell membrane. NaOH helps to break down the cell wall, but more importantly it disrupts the hydrogen bonding between the DNA bases, converting the double-stranded DNA (dsDNA) in the cell, including the genomic DNA (gDNA) and your plasmid, to single stranded DNA (ssDNA). This process is called denaturation and is central part of the procedure, which is why it’s called alkaline lysis. SDS also denatures most of the proteins in the cells, which helps with the separation of the proteins from the plasmid later in the process.

It is important during this step to make sure that the re-suspension and lysis buffers are well mixed, although not too vigorously (see below). Check out my article on 5 tips on vector preparation for gene cloning for more information and tips. Also remember that SDS and NaOH are pretty nasty so it’s advisable to wear gloves and eye protection when performing alkaline lysis.

4. Neutralization

Addition of potassium acetate (solution 3) returns decreases the alkalinity of the mixture. Under these conditions the hydrogen bonding between the bases of the single stranded DNA can be re-established, so the ssDNA can re-nature to dsDNA. This is the selective part. While it is easy for the the small circular plasmid DNA to re-nature it is impossible to properly anneal those huge gDNA stretches. This is why it’s important to be gentle during the lysis step because vigorous mixing or vortexing will shear the gDNA producing shorter stretches that can re-anneal and contaminate your plasmid prep.

While the double-stranded plasmid can dissolve easily in solution, the single stranded genomic DNA, the SDS and the denatured cellular proteins stick together through hydrophobic interactions to form a white precipitate. The precipitate can easily be separated from the plasmid DNA solution by centrifugation.

5. Cleaning and concentration

Now your plasmid DNA has been separated from the majority of the cell debris but is in a solution containing lots of salt, EDTA, RNase and residual cellular proteins and debris, so it’s not much use for downstream applications. The next step is to clean up the solution and concentrate the plasmid DNA.

There are several ways to do this including phenol/chloroform extraction followed by ethanol precipitation and affinity chromotography-based methods using a support that preferentially binds to the plasmid DNA under certain conditions of salt or pH, but releases it under other conditions. The most common methods are detailed in my article on 5 ways to clean up a DNA sample.


  1. Roosevelt Anyanwu on September 28, 2018 at 7:38 pm

    I could be wrong but Glucose here seems optional since the cells need to be subsequently lysed. However, in the Lysozyme method, we definitely need to prevent cell bursting.

  2. suman on May 3, 2017 at 2:04 pm

    can Someone suggest me about addition og TE- Rnase in plasmid isolation process. I have added water by mistake. What can i do now for getting DNA.

    • Sagar on January 10, 2019 at 9:35 am

      Your plasmid is still dissolves in Water so dont worry, your plasmid is there in water

  3. benish on January 31, 2017 at 9:18 pm

    plz tell me which method for DNA extraction from bacterial cells is mostly used n successful

    • aashish siddharth on June 16, 2017 at 7:37 pm

      we have a methods

  4. Vijay Kumar on December 4, 2015 at 7:13 am

    Whether NaOH is used in genomic DNA isolation? At which stage, NaOH is used in genomic DNA isolation from bacteria? please help me out.

    • AHMET on December 11, 2015 at 3:29 am

      actually , we look to generally materials

      EDTA destroys the cell structure by connecting Mg ++ ions required for protection of the cell structure and DNA degradation enzymes that inhibit the cell.

      * SDS lysis removes the lipid molecules and disrupt the structure of cell membranes interrupting.

      * Phenol-chloroform, porém and makes it move away from the sedimented DNA.

      * Pure ethanol provides the H2O molecules in the environment compete with the DNA collapse.

      * Ethanol 70%, Na + as it allows monovalent cations and the salts to dissolve away from the environment.

      • AHMET on December 12, 2015 at 3:32 pm

        an add ,
        aim of the NaOH , connected the flexible structure of the DNA . ( EFFECT OF pH )

  5. Nasir Ali on October 21, 2015 at 6:14 pm

    Hi Nick,
    I have a question ?? Can we use Alkaline extraction method for genomic DNA ?or this method is specific for plasmid DNA .


  6. Richard Cowart on July 16, 2015 at 7:51 pm

    I have used the alkaline lysis method for plasmid purification a number of times with good success. I was wondering about RNA carry over upon EtOH precipitation of the plasmids. I have been searching around and have not found any hard discussions of evidence on this issue. I notice that in the protocol Nick recommends adding RNase A prior to addition of the NaOH/SDS. But, I do not see how the RNase A would not be denatured at this step. It occurred to me that one could add RNase A + TE buffer + lysozyme + cells and they would lyse with the RNase A then degrading RNA. After an incubation period one could then add the NaOH/SDS and proceed with the protocol. But, back to the RNA question. Does anyone know if RNAs are carried over with the EtOH precipitation?

  7. Agnese Kocere on June 2, 2015 at 11:06 pm

    Hello! I was very confused by the following; Glucose maintains the osmotic pressure so the cells don’t burst

    Why try to avoid cell bursting if you want to lyse them anyway?

    Can someone clarify this, and maybe give a better explanation to why we add glucose?

    • Shishir Pandey on February 5, 2017 at 8:11 am

      Glucose is added because it acts as a ph regulator.

  8. EduHerNav on May 10, 2013 at 9:25 pm

    Great! thanks a lot for sharing.

  9. samkjm on February 19, 2013 at 2:41 pm

    Why use alkali and not acid to lyse cells???

  10. earl on September 20, 2012 at 11:10 am

    Awesome article! Thank you! 🙂

  11. JANI_MAULIK on May 27, 2012 at 8:01 am

    i don’t think glucose in slo. 1 is to prevent cell burst,even in water these cells won’t burst..

    i might be wrong..

  12. laurie on October 24, 2011 at 3:27 pm

    i’m working on a alkaline extraction for gram+ gram- and yeast, for set up after a reat time for diagnose urinary tract infection. i would like to do a purifiation on column after all for an eventual automatisation. Curently i try to do this with Qiagen( qiagen DNA mini kit)silica column and their buffer and their protocole.i explain myself, after i extract my DNA i put AL ( chaotropics salts) i heat the all for 10 minutes at 70 degrees and after that put the all on the silica column and wash it with many buffer ( AW1,AW2…). in my extraction protocole they are 70% of ethanol. when i make my real time and i compare my method with qiagen method( proteinase K ATL= detersif ) i have a Cp of 14 with Qiagen method and 18 with mine for a bacteria culture of 10^8 at first. without the Qiagen column i have a Cp of 14 with my procedure. if someone have an idear?

    thank you a lot guys

  13. bumunang on September 19, 2011 at 3:26 pm

    Thanks Nick God bless you

  14. Jode Plank on August 4, 2011 at 4:24 am

    Monisha & sameerbau,
    The pKa of glucose is around 12, so I suspect that the glucose was originally added to Buffer 1 to control the pH after the addition of Buffer 2, but I don’t know that for sure. Obviously Qiagen has shown us that the glucose in Buffer 1 isn’t strictly necessary.

    This should work fine on a hypothetical RNA plasmid. In fact, RNase is often added to Buffer 1 to prevent RNA from co-purifying with the plasmid.

    Lista S,
    I assume the ghost band you’re describing runs just a little faster than the bulk of the DNA (negatively supercoiled) and is resistant to restriction digestion. The occurrence of this form is a function of both the time the plasmid is in the denaturing conditions as well as the temperature of the solution during that time. Rather than lowering the pH of Buffer 2, I would incubate the resuspended pellets on ice for a minute or two before adding Buffer 2 to lower the temperature of the solution during this step a little bit. After the addition of Buffer 2 and mixing, you should incubate the solution at room temperature – if you do this incubation on ice, cell lysis may not occur efficiently.

  15. Lista S on August 1, 2011 at 5:49 pm

    Thanks for such an informative article! I was wondering, I keep getting a “ghost” band of denatured plasmid DNA, which I know is an artifact of alkaline lysis, but I can’t seem to find the best method to completely rid my minipreps of it. The band lessens when I leave the cells in the lysis solution for only one minute, but it won’t completely dissappear. I am considering pH-ing my lysis solution down to 12 before doing my next miniprep, any ideas on whether this will cure it?

  16. pooja on July 20, 2011 at 2:27 pm

    if suppose there is a dsRNA, then can this method would be applicable for extraction of this dsRNA plasmid??

  17. Shreya Roy on May 9, 2011 at 11:50 am

    Wonderful article!!! The Alkaline lysis method is crystal clear to me now, thanks to this article….
    Nick could you please write an article on how Genomic DNA isolation by the CTAB method works…. I somehow never seem to get results with that method , so understanding how it works might help me to figure out why i never am able to isolate genomic dna.

  18. sameer thukral on July 7, 2010 at 5:38 pm

    Monisha out there asked a lot of days ago why we need glucose in solution I?
    Any answers as yet?

    • kesmarkie_1111 on September 24, 2018 at 12:11 pm

      it’s required to maintain osmotic pressure

  19. sameer thukral on July 7, 2010 at 5:38 pm

    Monisha out there asked a lot of days ago why we need glucose in solution I?
    Any answers as yet?

  20. sameer thukral on July 7, 2010 at 5:38 pm

    Monisha out there asked a lot of days ago why we need glucose in solution I?
    Any answers as yet?

  21. sameer thukral on July 7, 2010 at 5:38 pm

    Monisha out there asked a lot of days ago why we need glucose in solution I?
    Any answers as yet?

  22. Pranay Dogra on June 7, 2010 at 6:27 pm

    Excellent article. The discussion is even better. I had all my doubts regarding Plasmid DNA extraction cleared today. I have to deliver a lecture a on genomic DNA extraction and these concepts will come in handy answering the questions of the students.

    Thanks. You are a life saver Nick 🙂

  23. Watanyoo Sopipong on May 12, 2010 at 2:02 pm

    Thank you ;D

  24. Jode on March 26, 2010 at 6:01 am


    While phenol and chloroform have low dielectric constants, they are almost completely immiscible in water. Therefore, the dielectric constant of the aqueous phase (where the DNA is) is minimally effected during an extraction. Since small alcohols are miscible in water, they can drop the dielectric constant and precipitate DNA.

  25. Jode on March 26, 2010 at 5:52 am

    Sam M,

    Strictly speaking, E. coli aren’t soluble, so they scatter light. Once you add the lysis solution, the cells pop and all the components of E. coli are solublized (surrounded by water molecules), either by the water directly, or with the detergent’s help, and the solution clears.

  26. Jode on March 26, 2010 at 5:37 am


    I’m a biochemist, which is kinda-like a chemist. Outside of that most of what I’ve learned came about after having something go horribly wrong, and then trying to figure out why. I’ve also been pretty lucky and frequently have had people around me that could help fill in the gaps in my knowledge. Hopefully, this web site can help others who don’t have that go-to person down the hall.

  27. fridah gakii on March 25, 2010 at 3:10 pm

    thank you so much.a
    question that is not discussed is:explain the reasons why the experiment of dna extraction can not be terminated after the lysis step?

  28. yoo on March 24, 2010 at 6:42 pm

    Excuse me, Why doesn’t DNA precipitated during phenol/chloroform extraction?
    Both phenol/chloroform and EtOH has low Dielectric constant?

  29. Sam M on February 17, 2010 at 12:27 pm

    When buffer II is added (the NaOH and SDS) the suspension goes clear; why does this happen? Is it something to do with the breaking of the hydrongen bonds in the DNA bases?

    • Vanshika Rai on August 30, 2017 at 6:22 am

      Solution becomes clear due to cell lysis

  30. fenil on February 2, 2010 at 5:34 pm


  31. Leighton on January 27, 2010 at 9:30 am

    This is really helpful! Thanks!!

  32. Ranga on January 17, 2010 at 7:32 pm

    Jode.. Thanks for that fantastic explanation. I’m just wondering how you guys can come up with amazing explanations like this. Are you a chemist? I get so many such questions, and I do come up with answers. But what I lack is the confidence.. I’m not sure how good my reasons are. I search the internet, and if I don’t find an answer, that’s it. That question is lost in obscurity.

  33. Rishabh jain on December 14, 2009 at 4:01 pm

    hello nick,
    I had a Practical exam today,My teacher asked me that “what would happen if we increase the incubation period after adding 3M potassium acetate in plasmid dna isolation by alkaline lysis method”?

    As we know that potassium acetate will help in renaturation of plasmid dna,but chromosomal dna will not renature.

    will the chromosomal dna will also renature after long incubation???

  34. Rishabh jain on December 14, 2009 at 3:54 pm

    hello nick,
    you have cleared my year long doubts about isolation of the DNA.
    thanks for doing such a wonderful job.

  35. Nick on November 16, 2009 at 4:27 am

    Jode, thanks for that fantastic explanation.

  36. Jode on November 14, 2009 at 8:47 pm


    Everything Nick described in Step 4 is true, but there is something else going on here as well. SDS is actually a salt, and when it dissolves in an aqueous solution the sodium disassociates from the dodecyl sulfate. The detergent, a long hydrophobic chain with a negatively charged sulfate group at one end, then goes on to find other hydrophobic molecules to associate with, namely the interiors of membranes and proteins, disrupting the structures of both. Solution Three contains free potassium ions, however, and potassium dodecyl sulfate is an insoluble salt (in the presence of excess K+). Once Solution Three is added to the dodecyl sulphate-denatured proteins and membranes, the potassium binds to the sulfate groups and neutralizes the charge, and the resulting KDS precipitates with the proteins and membranes in tow.

    (I was surprised to read that Vlokar’s lab uses sodium acetate in Solution Three – in this situation I would expect that all precipitation of protein is occurring via a pH mediated mechanism and that the resulting clarified supernatant would still contain some dodecyl sulfate and more protein than if potassium acetate were used in Solution Three.)

    If you would like to test what I mentioned above, though, simply mix a SDS solution with a KCl (or KCH3COO) solution in an eppendorff tube. You can increase the solubility of the KDS by heating it, and decrease the solubility by incubating on ice. This is why an incubation on ice and a 4°C centrifugation is common in plasmid prep protocols, particularly for protocols utilizing DEAE resins downstream, where residual dodecyl sulfate will interfere with DNA binding to the resin.

    You are right that the acetate that hitched a ride with the potassium into solution three won’t neutralize the hydroxide from solution two. However, solution three has been pH’d to ~5.5 with acetic acid. Since the concentration of hydroxide is relatively low (~0.1M), the addition of a high concentration of a weak acid is able to neutralize it without damaging the DNA, as would happen if the NaOH were neutralized by a strong acid like HCl. (This would create local environments of very low pH during mixing, which would cause nicks and breaks in the DNA.)

    The addition of all these salts does nothing (in a practical sense) to the solubility of the plasmid DNA because we are still in a solution with a high dielectric constant. During ethanol or isopropanol DNA precipitation, the cation (usually Na+) is only able to neutralize the negative charge on DNA and precipitate it once the dielectric constant of the solution is lowered by the addition of the alcohol. See Nick’s article on DNA precipitation (https://bitesizebio.com/2007/12/04/the-basics-how-ethanol-precipitation-of-dna-and-rna-works/) for more clarification.

  37. Ranga on November 11, 2009 at 9:00 pm

    I have a question: how does potassium acetate help here in plasmid extraction? In DNA precipitation, we add sodium acetate, the function of which would be to (1) neutralize the phosphate group making DNA less soluble (2) disrupt the hydration layer/hydrogen bonding (not sure which) by CH3COO-, again decreasing the solubility of DNA. Considering this fact, how does potassium acetate increase the solubility of plasmid? And on a separate note, forgive me if I’m dumb: how does CH3COOK *decrease* the alkalinity of the solution?

  38. Vlokar on September 23, 2009 at 3:19 pm

    Hi Monisha. It’s not necessary providing you have a good buffer. However I can see it’s purpose in preventing cells bursting whilst you are resuspending the cell pellet. If cells burst then, you may induce some mechanical shearing of the plasmid DNA with the pipette. Another factor might be controling the timing of the de-naturing of the DNA. Knowing the first exposure the DNA receives to any de-naturing compound is when you add that NaOH/SDS. Peace of mind I guess.

  39. monisha on September 10, 2009 at 2:15 pm

    If ultimately what we want is to get out the cell contents,why use glucose to prevent cells from bursting?

  40. Vlokar on August 29, 2009 at 10:18 am

    Hi Richa. The NaOH is around pH 12. This will de-protonate (remove + charge from) the Hydrogen bonds between the nucleic acid bases in the core of the double stranded DNA… allowing the strands to disassociate into single strands. Should this not occur and a “flocculant” of chromosomal DNA and proteins not clump together you will take up ds chromosomal DNA and proteins with your plasmid and it will be contaminated for whatever your doing with it next. Glacial acetic acid is just pure acetic acid. Our lab uses sodium acetate for nuetraliztion rather than potassium, (probably cheaper lol). Maybe your protocol also adds a salt or is relying on free ions in the solution. I encourage you to read the article by Birnboim and Doly that Nick Oswald cited in his main article for further clarification.

  41. richa on August 28, 2009 at 5:54 pm

    thnx 4 sharing this useful information. could u please tell what is the pH of lysis solution and why it is required to be alkaline? what if the solution is not that much alkaline ? what is the function of glacial acetic acid in neutralization solution? 52

  42. Vlokar on August 27, 2009 at 1:28 pm

    I hope this imparts a little extra insight. Sloppy technique can incorporate relative amounts of the flocculent precipitate which may include compromised gDNA. As well as mechanical breaks residual endonucleases could degrade ss or ds gDNA into small fragments. Prolonged exposure to ethanol can dehydrate the small segments of DNA causing denaturing of double strands. Small segments of ss DNA with repeat sequences can form hairpin structures which could be incorporated into the plasmid DNA extract. (Nucleic Acids Research. 15, 8739-8754). Please feel free to correct me should I have misunderstood.

  43. Vlokar on August 25, 2009 at 11:07 am

    So class moved onto alkaline lysis. One question has been posed regarding how the precipitated gDNA could damage potentially extracted plasmid DNA…. Over excessive mixing resulting in shearing of gDNA which will then contaminate the plasmid extract is a definite… (thankyou). Apparently there is another means, (I guess there are several)by which the chromosomal DNA can contaminate the plasmid DNA. The only other thing I can think of is incomplete precipitation of proteins in the TENS buffer (NaOH/SDS) hence leading to digestion of gDNA by free DNAses resulting in the reannealing of some small damaged excised gDNA that could move into solution with the plasmid. However I think I’m only expounding my own ignorance here… can anyone elucidate on this please?

  44. atiqah on August 20, 2009 at 5:11 pm

    why were the bacteria grown overnight in medium containing selective antibiotic?

  45. Vaughan on August 13, 2009 at 11:59 am

    This was a great source for info! Tremendous insight into the method and really helped with compiling my Practical report. Thanx Alot! 🙂

  46. jessica on August 5, 2009 at 12:36 pm

    What does the white flocculant consist of?

    • Nick on August 5, 2009 at 1:23 pm

      Hi Jessica

      It is the membranes, denatured proteins etc. In theory it should be everything except the nucleic acids.


  47. Piet on April 20, 2009 at 8:50 pm

    Dear Nick Oswald,
    thank you for this very useful information, you can’t imagine how much time one looses to find excact information. I have a question: is there a difference in lysis buffers for gram positive and negative bacteria? (in the article from Birnboim and Doly, who described the first alkaline extraction, they use lyzosyme. Is the function of lyzosome taken over by NaOH by means of denaturation of proteins?) thank you in advance

    • Nick on August 5, 2009 at 1:07 pm

      Hi Peit

      Sorry for taking so long to reply! The difference between gram +ve and -ve bacteria is in their cell walls and while the role of lysozyme is to break down the cell wall, it works on gram -ve bacteria, but only some gram +ve bacteria. In the alkaline lysis procedure desrcribed, I think that the NaOH and the lysozyme are redundant — they will both break down the cell wall so if you remove the lysozyme from B&D’s protocol, it still works.

  48. arul on July 28, 2008 at 2:04 am

    what is the exact pH at which cell wall lysis occurs and the plasmids are released?

  49. ritu on July 15, 2008 at 9:27 am

    does any concentration of NaOH react with CO2 in air…because in lab v generally prepare 10N NaOH and store it for months..is that recommended??

  50. Nick on April 5, 2008 at 2:38 pm

    maan – the the pH of solution 2 goes up, the lysis will be inefficient.

  51. CK on April 4, 2008 at 4:51 pm

    Thanks for writing these articles Nick. This kind of info. is scarce thus much appreciated. Can stop hassling the senior postdoc with these kinds of questions now!

  52. maan on March 27, 2008 at 2:53 am

    what happens when the pH of the solution is lowered when the solution 2 prepared was not fresh?

  53. Nick on March 21, 2008 at 5:17 am

    Hi Tina

    I am reliably informed that the reason solution 2 has to be made fresh is that the NaOH reacts with CO2 in the air, thus lowering the pH.

  54. Tina on March 20, 2008 at 5:05 pm

    the protocols always say that Solution 2 must be freshly prepared but they do state the reason why. Do you any reason why Solution 2 must be freshly prepared?

  55. SANKET on March 19, 2008 at 9:52 am

    Hi sir,this is amazing source for alkaline lysis method,thanks for making such a wonderful source to help students.Best up luck.

  56. Rikesh Patel on January 22, 2008 at 10:58 pm

    This is amazing. Thanks so much for sharing this information with us. It gave me a perfect insight into each step and gave me a starting point in my research for University coursework.

  57. Rach on November 7, 2007 at 5:42 pm

    He he, in our lab this one is known as “The Snot Extraction” 😉

Leave a Comment

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

Share via
Copy link