Cfrench:IdentifyingBacteria

Identification of Bacteria
Our current standard method of identifying bacteria is to amplify the 16S rRNA gene by PCR using 'universal' bacterial primers, then sequence the PCR product. The primers we use are from a fairly old paper (Weisburg, W.G., Barns, S.M., Pelletier, D.A., and Lanes, D.J. 1991. 16S ribosomal DNA amplification for phylogenetic study. Journal of Bacteriology 173, 697-703) and I am not sure how universal they really are, but certainly we have used them with no trouble for various proteobacteria and high-GC Gram positive bacteria with, as far as I recall, 100% success to date. The primer sequences are:


 * fD1: agagtttgatcctggctcag
 * rD1: aaggaggtgatccagcc

The protocol is as follows:

1. Obtain a colony of your bacterium of interest on a plate.

2. Suspend the colony in sterile DI water (about 0.1 ml).

3. Use as template in a PCR reaction with Taq polymerase (or optionally a more expensive polymerase like Pfu or Kod, but for routine purposes Taq should be fine). A standard reaction mixture with Promega GoTaq would be:


 * 31 microlitres water
 * 10 microlitres 5 x reaction buffer
 * 3 microlitres 25 mM MgCl2 (not required unless using 'Flexi' buffer)
 * 2 microlitres primer mixture fD1 + rD1 (10 pmol/microlitre each)
 * 1 microlitre dNTP mixture (10 mM each)
 * 2 microlitres cell suspension
 * 1 microlitre Taq polymerase
 * TOTAL 50 microlitres

4. Perform PCR, annealing at 58 degrees C and extending for 90 seconds.

5. Check the PCR product on a gel. There should be a single band around 1.5 kb.

6. If all is well, purify the PCR product from the PCR mixture (not from the gel, unless absolutely necessary). Use 20 microlitres of glass beads and elute to 40 microlitres of EB (see 'Purifying DNA from a PCR reaction', link to be added here).

7. Submit the PCR product for sequencing. Prepare 2 reaction mixtures, each with about 2 microlitres of PCR product and 3 microlitres of water (depending on how strong your PCR product was on the gel) and 1 microlitre of primer, either fD1 or rD1, at 5 pmol/microlitre.

8. Wait for the sequence to come back. If you are lucky, you may get overlap and can edit the two sequences into a single near-complete 16S rRNA sequence. Otherwise, just use both sequences separately. Submit as BLAST to Entrez. The top results will indicate the organism in the database which is most closely related to your bacterium.

Update: 31 Oct 2008
Here is a modified procedure for identifying multiple isolates at once.

1. Prepare cell suspensions of all of the organisms from subculture plates as follows: pipette 10 microlitres of sterile water onto a colony on the plate, suspend cells in the droplet by pipetting up and down, and transfer the liquid to a sample tube.

2. Prepare PCR master mix for n reactions as follows (recipe is for Promega GoTaq using the non-flexi-buffer which contains magnesium):


 * water 13.4 x (n+1) microlitres
 * 5 x buffer 4.0 x (n+1) microlitres
 * primer mix (10 pmol/microlitre each fD1 and rD1) 0.8 x (n+1) microlitres
 * dNTP mix (10 mM each) 0.4 x (n+1) microlitres
 * GoTaq polymerase 0.4 x (n+1) microlitres

Mix well by pipetting up and down. You will note that this prepares enough solution for one more reaction than you will actually do. In our experience, if you prepare exactly the right amount, you will always be a little bit short by the time you get to the last reaction. Maybe our pipettes just need recalibrating.

3. Transfer 19 microlitres of reaction mix to each PCR tube and add 1 microlitre of cell suspension.

4. Perform PCR and proceed as described above.

Back to Main Page