I am working with plant samples (leafs and berries) and I have a problem with sequences generated by pyrosequencing when I perform PCR amplification of the V6 region of the 16S rRNA gene. The results obtained show that 60-90% of reads are plant sequences, mostly from the 16S of chloroplastidial DNA. Has someone had this problem too and solved it or has any idea of how to remove or reduce the amount of chloroplastidial DNA? Or are there any other ideas of how to reduce the percentage of plant sequences?
Hi Diogo, we have same issue. We sequenced using Illumina and got approx same range of chloroplast and also about 3% of mitochondrial amplicons. I think the best way is to filter out these sequences, either in mothur or Qiime and continue with analyses. I do not know about any method which could reduce the plant sequences after PCR since the chloroplast and mitochondria probably originate from bacteria, so their 16S rDNA genes amplify quite nice with universal bacterial primers.
Another alternative, not for this dataset but for future data, would be to try to eliminate plant material by just swabbing the plant surface, or washing it and using that wash/rinse. I am not sure if that would be applicable, but that would increase the bacteria to plant DNA ratio.
But I think switching to a different primer pair, as suggested by Naupaka, would be a good choice as well.
Good luck!
Dear Diogo, I agree with Elisabeth Bik. In order to avoid non-target sequences from getting amplified, the metagenome extraction method must be critically optimized. Another way is to choose optimum primer pairs that amplify the target domain by in silico approach using available tools and databases (such as RDP, GreenGene, Silva). This NAR paper will definitely help you out to solve your problem at the stage of PCR amplification http://nar.oxfordjournals.org/content/41/1/e1.long.
It is normal to have some chloroplastidial 16S. They are usually assigned as cyanobacteria-like. However, the percentage you obtained is so big. To avoid such problem, you have to carefully select the couple of primers to use. I would suggest to read all the papers from the field of the epiphyte and leaf associated communities and to test in silico their primers.
For the next time you can also consider PNA-clamping to avoid the chloroplastidial 16S.
Hi Diogo, you can try and do subtractive hybridization of a sort, with a batch of the plant DNA prior to the 16S PCR. Isolate your “pure” plant DNA from a “sterile” compartment of the plant and label with biotin. Also isolate the DNA that you care about, the one with the bacterial component in it. Hybridize you sample of interest to an excess of the “pure” labeled batch, remove hybrid molecules with streptavidin magnetic beads – the leftover should now contain mostly bacteria, fungi, etc., PCR 16S and NG Sequence. I do not expect that process will remove all, but it is a fairly simple procedure and may be worth trying.
Hi Diogo. I am not sure if you have already overcome on the problem of plastid contamination. But in case someone still has the same problem, it's been recently published a paper in which the authors have designed several modifications of the 799F primer that, combined with primer 1392R, reduced remarkably plastid contamination.
http://www.sciencedirect.com/science/article/pii/S0167701213002510
Hi Diogo,
I hope that you have found a solution in the meantime. Otherwise, I might have an idea even though being someone from the commercial side.
Like in other eukarytotes, also the genome of plants is methylated; the most prominent motif methylated is the dinucleiotide CpG (only motif in vertebrates) but also CHH and CHG. In contrast to this, chloroplast DNA does not seem to be methylated (see http://www.plantmethods.com/content/9/1/16). In general, prokaryotes do not feature CpG methylation.
Our LOOXSTER Enrichment Kit enables the enrichment of prokaryotic DNÀ as non-methylated CpG motifs are targeted by this kit. Even though not evaluated yet, I would assume the plant genomic DNA to be depleted significantly when using LOOXSTER. However, as chloroplast DNA doesn't seem to be methylated, this DNA would be also enriched.
I hope this info is of some help to you.
Best,
Ingo
- Badges
- Science method