Chapter 7. Beer Bottle Minipreps

Joe Rokicki


Minipreps are the bread and butter of the genetic engineer. After constructing or procuring a DNA sequence of interest, one of the first next steps is often to insert that DNA into a larger circular piece of DNA called a plasmid. We do this because plasmids have the wonderful property of being autonomously amplified inside of bacteria. If you can get your DNA of interest into a plasmid and you can get that plasmid into a bacterium, all of the fancy, error-correcting, DNA copying machinery of the bacterium will go to work for you, automatically copying your DNA. You can sit back and relax and grow up a bacterial broth that will be rich in your DNA sequence. The only thing you have to worry about is breaking the cells open and purifying your amplified DNA back out, away from all the other genomic DNA, RNA, proteins, and miscellaneous cellular debris. This is where the miniprep comes in.

A miniprep is a column-based plasmid purification protocol where crude cell lysate is passed through a resin that specifically binds plasmid DNA and lets everything else pass through. The plasmid DNA is later released from the column in a highly pure and concentrated form. The miniprep’s speed, ease of use, and purity have made it the de facto standard over other plasmid purification protocols such as phenol chloroform extractions and cesium chloride gradients. Yet, despite being one of the most universally performed protocols in molecular biology, the miniprep is also often one of the least understood, even among scientists, for the simple reason that the vast majority of labs take advantage of commercially prepared kits that require zero understanding of what you are actually doing. A little investigation quickly reveals that the active ingredient of the mystery column resin is actually silica, the most abundant mineral on earth. Further, the reaction that causes silica to specifically and reversibly bind to DNA turns out to require nothing more than salt. While the identity of the salt affects the efficiency of the binding (I will go into more detail about this in the next section), at the end of the day, this isn’t an exotic reaction requiring rare chemicals or dangerous reactions. It is simply sticking DNA to sand with salt.

Richard Feynman once famously said eight words that, decades later, scientists at the Venter Institute went on to immortalize (mortalize?) by encoding into the genomic sequence of the first synthetic organism: "What I cannot create, I do not understand." In this article, I will explain the basic principles that allow the miniprep process to isolate plasmids and show you how to create (and therefore understand!) your own DIY miniprep spin columns from commonly available materials.


Commercial minipreps work on two principles. The first is alkaline lysis. In a highly basic solution with a little detergent, bacterial cells break open and their contents denature. DNA’s iconic double-stranded helix unwinds into separate single strands, and proteins unfold from the characteristic shapes they need to function. If the pH is just right, the plasmid DNA will behave differently from all the genomic DNA. Because it is in a highly twisted “supercoiled” state, it is able to better resist the DNA unwinding effects of a basic solution. As a result, when the alkaline lysate is later neutralized with acid, the denatured proteins and genomic DNA will all coagulate into an insoluble mass and precipitate with the detergent; but the plasmid DNA, which was never denatured in the first place, will stay separate and in solution.

The second principle minipreps leverage is silica adsorbtion of DNA at high concentrations of chaotropic salt. A chaotropic salt is a salt that is really good at denaturing proteins. It turns out that one of two things usually happens when you add a lot of salt to a protein solution. Some salts mess up the sphere of water molecules that surround a protein and cause the protein to aggregate with other proteins and precipitate. This is called “salting out.” Other salts denature proteins into super soluble unfolded amino acid chains. This is called “salting in.” Salts can be ordered according to their propensity to either “salt in” or “salt out.” The “salting in” side of the salt spectrum is called chaotropic. The “salting out” side is called kosmotropic.

For some reason, at high concentrations of chaotropic salt, nucleic acid sticks to silica. There are differing theories on why exactly this happens; and while the chaotropic nature of the salt seems to increase the efficiency of adsorption, even common kosmotropic salts like NaCl have been reported to cause adsorption of nucleic acid (, LA Biohackers).

I’m leaving out a few details here, but in general, this is how a commercial miniprep works. First, the cells are pelleted and resuspended in a neutral buffer. This is important so that the pH of the solution can be tightly controlled in the next steps. Second, we add the highly basic solution with SDS that will lyse the cells and irreversibly denature everything except for the plasmid DNA. Third, we add an acidic solution that will neutralize the basic solution we just added and cause the denatured genomic DNA and proteins to precipitate out. They are then removed by a high-speed centrifugation step. The acidic solution also contains tons and tons of chaotropic salt so when we run the salty, neutralized, clarified cell lysate through a silica column, the plasmid DNA is adsorbed. The column is then washed with an ethanol solution that will remove the excess salt but leave the plasmid DNA on the column. Finally, a small volume of water is added that is sufficient to break the adsorption interaction between the plasmid DNA and the silica and elute the DNA from the column. This small volume of pure and concentrated plasmid DNA is our final product.

DIY Minipreps

To make our own version of this system completely from scratch, we will need to make our own solutions and our own spin columns.

Miniprep Solutions

Several excellent recipes for DIY miniprep solutions are already available online. In the following experiments, I used the solutions P1, P2, N3, and PE described here with one small modification. I used acetic acid to bring the pH of solution N3 down to ~4.2, which was the pH I measured for the commercial solutions. Alternatively, in true DIY spirit, LA Biohackers have published an alternative protocol that uses only commonly available chemicals and materials.

Spin Columns

This is where you get to do some MacGyvering. With minimal effort and only commonly available materials, we can hack together a surprisingly professional miniprep spin column. This method is adapted from a protocol for eluting water from small pieces of filter paper, but I have found it works excellently for DIY minipreps as well. First, place a 0.65 mL tube inside of a 1.5 or 2 mL flat bottom freezer tube. Puncture the 0.65 mL tube twice with a small-gauge needle. I used a 5/8 inch, 25-gauge needle that was just long enough to puncture the 0.65 mL tube, but not long enough to reach the bottom of the 2 mL freezer tube, so I didn’t have to worry about sticking myself (see Figure 7-1). It is important to use a flat-bottom type tube for the collection tube as opposed to a standard 1.5 mL tapered tube because you want to have at least 500 uL of volume below the inner 0.65 mL tube to catch column flow through. The last step is to fill the inner 0.65 mL with a silica-containing material.

The needle is used to puncture the smaller 0.5 mL tube. The 0.5 mL tube is filled with a silica-containing compound. This is then placed inside a larger flat bottom 2.0 mL tube to form the DIY miniprep column.
Figure 7-1. The needle is used to puncture the smaller 0.5 mL tube. The 0.5 mL tube is filled with a silica-containing compound. This is then placed inside a larger flat bottom 2.0 mL tube to form the DIY miniprep column.

Silica Resin

To find a good resin to pack our column with, I experimented with several commonly available sources of silica. WARNING: Silica dust is dangerous to breathe. If silica is being crushed, do so in a hood, underwater, or while wearing a respirator. Don’t breathe dust containing silica:

Crushed “crystal” cat litter
This stuff is great. It’s tiny chunks of silica gel. It’s the same stuff in the desiccant packets they stick in with shoes and beef jerky to keep them dry. It looks kind of like broken safety glass. I placed two small chunks in a column and crushed it with a screwdriver. It crumbles easily into powder.
Uncrushed crystal cat litter
To make sure the crushing was necessary, I tried a column with two pieces of uncrushed silica gel.
Powdered 90 Shilling beer bottle
I thought it would be cool to make a "beer bottle miniprep," but, compared to working with silica gel, crushing glass bottles is a huge pain. I broke the bottle in a plastic bag with some water in it using a hammer, moved the slurry to a 50 mL conical tube, and continued to crush the glass to powder underwater using a screwdriver. After much grinding, I resuspended the slurry, allowed it to settle for an hour, decanted the supernatant, spun it down, and collected my hard-earned glass powder.
Celite 454
This is fancy diatomaceous earth that I had on hand. You can buy food-grade diatomaceous earth online for cheap.
Regular, fine beach sand.
Commercially available miniprep column
I used a Qiagen miniprep column.


  • A flask of 25 mL LB containing 100 ug/mL Amp was inoculated with E. coli containing high-copy plasmid A. A second flask of 25 mL LB containing 100 ug/mL Amp was inoculated with E. coli containing high-copy plasmid B. These two cultures were allowed to grow overnight then were pelleted and stored at –20°C until use.
  • The cultures were each resuspended in 2.5 mL of buffer P1 and combined for a total of 5 mL of saturated bacterial culture.
  • This suspension was divided into 18 aliquots of 250 uL in 1.5 mL tubes.
  • 250 uL of buffer P2 was added to each solution. The solutions were inverted five times.
  • 350 uL of buffer N3 was added to each solution. The solutions were inverted five times.
  • The suspensions were spun at full speed for 10 minutes on a microcentrifuge (>10,000 RPM).
  • The following procedure was performed three times on each of the six columns described above.
  • 500 uL of supernatant was added to the column. Column spun at full speed for 15 seconds. Flow-through discarded.
  • The remaining ~170 uL of supernatent was added. Column spun at full speed for 15 seconds. Flow-through discarded.
  • 500 uL of buffer PE was added to the column. Column spun at full speed for 15 seconds. Flow-through discarded.
  • 500 uL of PE was added to the column a second time. Column spun at full speed for 15 seconds. Flow-through discarded.
  • Column spun at full speed empty for one minute to remove excess wash buffer.
  • Column moved to fresh 1.5 mL tube. 50 uL of nuclease-free water added to column. Column spun for one minute at full speed.
  • Eluate (50 uL) reapplied to column and spun again for one minute at full speed.


The three gels here are technical replicates. Each gel contains six lanes where the only difference from lane to lane is the column used to perform the purification. The columns used in each lanes are: 1—crushed crystal cat litter, 2—uncrushed crystal cat litter, 3—crushed beer bottle glass, 4—diatomaceous earth, 5—sand, 6—commerically available miniprep column. The two bands visible in most of the lanes correspond to the two plasmid that were copurified as described in the methods. Uncrushed crystal cat litter and sand both failed to purify any DNA in any of the experiments. Crushed crystal cat litter, beer bottle glass, diatomaceous earth, and the commerical miniprep column purified DNA in all replicates, although the DNA purified with beer bottle glass ran slower, perhaps because of DNA sheering. In summary:

  • The crushed-crystal-cat-litter resin worked as well as a commercial miniprep, followed closely by diatomaceous earth.
  • Crushing the crystal cat litter is necessary; the column with uncrushed resin didn’t yield any DNA.
  • Powdered beer-bottle glass appears to sheer the plasmid DNA, resulting in bands that migrate slower than the plasmid DNA purified with other resins.


These columns are cheap and easy enough to be disposable. To save even more time and money, this recycling protocol is an effective way to destroy any contaminating DNA and reuse the same column many times. Have fun!