From OpenWetWare
Jump to navigationJump to search


This is an (as yet untested) work-in-progress protocol with the goal of replacing silica columns for bacterial minipreps with a magnetic bead based protocol. This protocol is based on. [1]. It may be possible to simplify the cleared lysate generation to a protocol more similar to that of Zymo's direct-from-culture lysis. The major predicted advantage of this protocol over spin-column protocols is cost; beads are ~$380 for enough for 3,750 purifications, or ~10c per purification. Whether the yield and purity match that of a spin column is unclear.

  • Andy Lane 20:42, 23 October 2014 (EDT): Tested 10/23/14: As written, the wash buffers supplied here seem to wash off all DNA as predicted; perhaps this is due to some difference between the carboxylate beads used here and those used by the authors. However, 80% EtOH wash instead retains a lot of DNA with a Nanodrop-reported yield of ~8.2ug from 3ml culture. Digest looked great, but uncut DNA did not look good; it ran at >12kb molecular weight (plasmid length 5.7kb). Seems to be running far higher than expected for a nicked circle.


Carboxy-coated magnetic beads are Sera-mag SpeedBeads (Fisher #09-981-123). These come at 50mg/ml; protocol calls for 20mg/ml in 0.5M EDTA pH7.2:

  • Solution 1 (Uses 30µl/prep - make 10ml; store at RT or at 4˚C to preserve RNAse A if preferred):
    • 50mM Glucose
    • 25mM Tris pH 8.0
    • 10mM EDTA pH 8.0
    • 100µg/ml RNAse A
  • Solution 2 (Uses 60µl/prep)
    • 0.2N NaOH
    • 1% SDS
  • Solution 3 (Uses 45µl/prep)
    • 3M KOAc
  • W1 (Uses 1ml per prep)
    • 5M NaCl
  • W2 (Uses 500µl per prep)
    • 25mM Tris Acetate pH 7.8
    • 100mM KOAc
    • 10mM Mg2OAc
    • 1mM DTT
  • 2X Binding buffer (uses 100µl per prep)
    • 20% PEG 8,000
    • 2.5M NaCl


Preparing magnetic bead stock

  1. Place 400µl beads into 1.5ml tube; collect on magnet.
  2. Remove supernatant
  3. Resuspend beads in 1ml 0.5M EDTA pH 7.2.
  4. Place tube back on magnet. Remove supe.
  5. Resuspend beads in 1ml 0.5M EDTA pH 7.2 again.

Store at 4˚C.

Purifying DNA:

  1. Spin down 1 ml bacterial culture.
  2. Resuspend pellet in 30µl Solution 1.
  3. Add 60µl Solution 2. Mix by inversion, make snot, incubate for a minute or two. More than five minutes is reported to permanently damage DNA.
  4. Neutralize with 45µl Solution 3; mix by inversion.
  5. Spin at max speed in room-temp microfuge for 10 minutes. Transfer 100µl of the supernatant (cleared lysate) to a new tube; trash the pellet.
  6. Add 10µl of your prepared magnetic beads to the cleared lysate along with 100µl 2X binding buffer. Incubate at RT for 5 minutes.
  7. Using magnetic rack, wash the beads with 500µl W1 twice. During washes, beads can be left as a pellet on the side of the tube. No need to resuspend.
  8. Wash once with 500µl W2.
  9. Resuspend beads in 50µl ddH2O. Spin to the bottom of the tube in a minifuge (givining ~1 min for DNA to be released from beads), then place back on rack. Collect the water containing DNA.

Left to right:

1: Marker (1kb+, Invitrogen).

2-5: Four clones that putatively contain a 5.7kb plasmid. Washed using 80% EtOH x2, 500ul each.

6-9: As 2-5, but a parallel prep washed using wash buffers outlined above.

10-13: As 2-5, but cut with XmnI. Pattern matches theoretical cuts for first 3 clones. Clone #4 is presumably negative.

14-17: As 6-9, but cut with XmnI.


Please feel free to post comments, questions, or improvements to this protocol. Happy to have your input!

  1. Andy Lane 18:53, 22 August 2014 (EDT) There are some things about the buffers that might be worth experimenting with. Buffer W2 is a little unexpected; why so complex (Mg, DTT)? Could it be replaced with 70% or 80% EtOH or as in Agencourt Ampure XP protocols? This would necessitate significant drying time to remove EtOH, though.
  2. Why not also have the bead stock stored already in the 2X Binding buffer rather than in EDTA, and add everything at once?

It may also be worth looking at Elkin et al., 2001 [2]. In summary, this paper generates a cleared lysate by a boiling-freezing protocol in STET-E1 buffer containing 3M NaCl, 165mM Tris-HCl pH 8.0, 45mM EDTA pH 8.0, 16.5% Tween 20, 200mg/ml RNAse A, 1mg/ml lysozyme. (185ul culture added to 60ul STET-E1) This is a slow protocol for single tubes, but relatively hands-off so suitable for 96-well plates (37 ˚C for 15 min, 90 ˚C for 10 minutes, -80 ˚C for ≥30 minutes; spin 90 min at 3200 x g; retrieve 100µl cleared lysate). To make up a bead stock, they add 52µl 5mg/ml washed beads to 448µl 2.5M NaCl/20% PEG-8000 (there's what looks like a typo in the paper at this part, though, making the ratios unclear). 145µl of this bead suspension is used per 100µl cleared lysate. The only washes are 6x 200µl 70% ethanol. After drying, DNA is eluted in 50µl H2O over 5 minutes on a shaker. Their stated yield is ~4µg plasmid from 185µl of Terrific-broth + 1.6% glycerol-grown E. coli, which sounds like a ton. How much could you get from a standard 3ml culture?

  • Potential upsides to the Elkin protocol
    • The washing sounds easier/less likely to lose DNA, likely to result in less-salty DNA.
    • They seem to get a lot of DNA out.
    • They use ~1/3 the beads of the Hawkins protocol (145µl of a 1:100 dilution of Sera-Mag, rather than 10µl of a 1:2.5 dilution of Sera-Mag), but that's for a 185µl culture.
  • Potential downsides to the Elkin protocol:
    • The lysis method isn't optimal for easy microfuge tube scale lysis. Would probably be better to use an alkline lysis method.


Relevant papers and books

  1. Hawkins TL, O'Connor-Morin T, Roy A, and Santillan C. DNA purification and isolation using a solid-phase. Nucleic Acids Res. 1994 Oct 25;22(21):4543-4. DOI:10.1093/nar/22.21.4543 | PubMed ID:7971285 | HubMed [Hawkins-NAR-1994]
  2. Elkin CJ, Richardson PM, Fourcade HM, Hammon NM, Pollard MJ, Predki PF, Glavina T, and Hawkins TL. High-throughput plasmid purification for capillary sequencing. Genome Res. 2001 Jul;11(7):1269-74. DOI:10.1101/gr.167801 | PubMed ID:11435410 | HubMed [Elkin-GenomeRes-2001]

All Medline abstracts: PubMed | HubMed


  • Who has experience with this protocol?

or instead, discuss this protocol.