Skip to content

Better Plasmid Midipreps Part II: What Causes Low Yields?

Recently we received a question from Bitesize Bio reader Sonia after our article How to: Get Better Plasmid Midiprep Yields. She asked: “What could be the problem when one sample gives a good yield while the other plasmid gives poor a one, when both the samples were processed simultaneously, and in the same way.”

This is a good question because many things can cause differences in yields between plasmid preps.  Let’s resolve this mystery one point at a time and go over some reasons why you might get low yields when you prep plasmids.

1. Plasmid backbone

You prepped two plasmids simultaneously using the exact same protocol and had different yields. If the plasmids have the same backbone (that is, they are both pUC, or pBluescript, etc.) then the reason leans towards the insert playing a role. Some inserts can be problematic for bacteria. It might be that a protein is made that makes the bacteria sick (for example, DNase) or it could be that the insert is unstable (for example, repetitive sequences). To overcome the problem, try using a specialized competent cell line. For unstable inserts, try the STBL2 cells from Life Technologies and for growing clones with toxic proteins, try the T7 Express LysY/Iq Competent cells from NEB.

Another important point is how the insert size changes the copy number of the plasmid. Large inserts will reduce the number of copies of the plasmid.

2. Copy number

If the genes are cloned into different vectors then the issue could be that the plasmids are replicating at different rates. One may be high copy and one may be a medium or even low copy plasmid. Some examples of  low copy plasmids are ones using the backbone pBR322 and pACYC, which are older and not used often in cloning work today. Many vectors used for protein expression are medium copy. This is desireable because when producing proteins, sometimes if growth is too fast, it enhances the chance of the protein becoming insoluble or forming inclusion bodies.

3.  Culture issues and antibiotic

Nick went over the technique for growing bacteria to obtain a healthy culture in late log phase where you can get the most plasmid in our original midiprep how-to article.  This technique works well as does using a 1:1000 dilution of starter culture into a large scale culture (so 100 ul into 100 ml) if you want to grow the bacteria overnight for 12-16 hrs.  One important consideration is the antibiotic. The bacteria are going to break down the antibiotic while they are growing in the culture. If not enough antibiotic is added or if the stock is old and not at the correct strength, the antibiotic selection pressure may not last very long and you could end up having a culture that was antibiotic-less for most of the culture time. Plasmid yields will go down without the selective pressure to keep it.

As a reminder, the state of the culture is critical for high yields of plasmid. For maximal yields, the culture should be in late log or early stationary phase. If the culture overgrows, you will be harvesting more dead bacteria than live cells and this also leads to genomic DNA contamination in the prep. If the culture is undergrown, then of course, yields are lower than expected. You can mistakenly undergrow a culture by using old colonies from plates or starting direct from a frozen stock and not from a colony. The lag time for the bacteria to ramp up is much longer when you use either of these approaches.

Streak fresh colonies:

One more point that people forget when setting up their starter cultures or overnight cultures is the age of the plate you are using to pick your colony. If your plate is old, you may have picked a nice big colony but it will not be all living cells. And if there were satellite colonies sitting around the original colony where the antibiotic no longer exists, those will not have plasmid and will be introduced into your culture. So streak a fresh plate before starting to ensure the best result.

So let’s assume that Sonia’s plasmids were the same vector, same antibiotic, grown exactly the same, and each have different inserts that are not too large or unstable or toxic. What else can it be?

4. Processing steps:

If the culture conditions are not the problem, then we have to look at something with the downstream steps. Since we are talking about Midipreps, let’s discuss what can go wrong with the anion-exchange procedure next.

Nick covered alkaline lysis in great detail and typically these reagents in the kits are stable and fine. Solution 2 (the one containing NaOH and SDS) can break down over time with exposure to air, but in general, they work for lysing bacteria for the life of the kit.

The other area of plasmid preparation where DNA can be lost are the final steps after anion-exchange which is the final precipitation step in isopropanol and finding the DNA pellet.

Isopropanol quality:

Many labs have isopropanol in large containers that have been opened and closed over the course of a year. For the best result in the precipitation step, make sure the isopropanol used is not the old bottom-of-the-barrel stuff. Use some isopropanol from a new bottle or a smaller bottle that is not who-knows how many years old. This makes a huge difference in the size of the DNA pellet you obtain after centrifugation.

Don’t lose the pellet!

Isopropanol pellets are glassy and clear and difficult to see. The best practice is to mark the side of the tube where you expect the pellet to form after centrifugation in a fixed angle rotor so when you decant the isopropanol, you know where to look for it. Keep an eye on the spot and look for the glassy material. Sometimes this is difficult because many people use the oakridge plastic tubes which are opaque. If you have glass corex tubes, this is a nice alternative and they can be baked to make them pyrogen free.

Sometimes, if you have concerns about losing the pellet, it is good practice to pour the isopropanol supernatant into a 15 ml tube to save it, just in case the pellet slipped off the wall. But this does not normally happen as long as you do not let the sample sit for long after the centrifuge stops. Once it is done, be right there to decant the sample. The only times I have seen a pellet come loose from the wall is when I was late getting to the centrifuge and it sat still for a few minutes.

Whether you use Oakridge tubes or glass, just note where that pellet should be. Once you wash with 70% ethanol, the pellet becomes visible.  When you are ready to resuspend your pellet, you’ll know exactly where to find it because you marked the tube.

Caution! It is not always a pellet!

Sometimes with fixed angle rotors, the DNA may not always form a nice tight pellet at the side wall. It can sometimes smear down the side. For this reason, I always use my resuspension buffer to wash down the side of the wall above my pellet to make sure I solubilize every molecule of plasmid that may be present even though I can’t see it.

It is a shame to do all that work and then lose the DNA pellet right at the end! More tips on perfect DNA pellet recovery is found here too.

We had a nice discussion about DNA precipitation in a previous article and it was the consensus that the most important factor in obtaining high yields is centrifugation speed and time. Don’t cut the centrifuge time short unless you can turn up the speed.

I thought it would be good to mention that many plasmid kit manufacturers have recognized that the pelleting step is problematic for some users so have developed kits that desalt the DNA using “precipitators” or silica disc filters. These are a fast alternative to centrifugation. However, you still need good isopropanol for these to work so always use fresh.


Large scale plasmid DNA preps have a lot of steps where things can go wrong but in my experience, the problem is usually either the culture or the DNA precipitation.  To check if your culture is healthy, just take 1-2 mls out of your 100 ml flask and then do a quick miniprep on it to see how much plasmid/ml is there. That will give you a good idea of what you will get from the rest of the sample.

So remember to get great plasmid yields, do a little background first on your vector and insert to make sure there is no reason for the DNA itself to be a problem and then start with a fresh colony and a starter culture and fresh antibiotic. And at the end, fresh isopropanol will be key to a thorough precipitation of all the plasmid DNA.

[sponsor] [adrotate banner=”69″][/sponsor]


  1. Dr Amanda Welch on June 6, 2016 at 11:46 pm

    It looks like pMAL is ampicillin resistant. You can use chloramphenicol (another antibiotic) amplification. You can read a bit about it here (https://bitesizebio.com/22824/how-to-manipulate-plasmid-copy-number/) and there’s a link out to more resources in that article.

  2. Gimme vector on June 6, 2016 at 10:30 pm

    How do you go about getting appreciable yields from low-copy plasmids like the pMal series from NEB?

Leave a Comment

You must be logged in to post a comment.

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

Scroll To Top
Share via
Copy link