Instead of the universal eukaryotic primer NS31 and the AM-specific primer AM1, I am wondering if anyone has used the primers developed by Lee et al (2008) that amplify a much higher proportion of all known AMF sequences. If so, how has it affected your interpretation of results? Presumably the sequences they produce have a lower % similarity to those in the MaarjAM database. How do you modify your analysis in accordance with this?
http://onlinelibrary.wiley.com/doi/10.1111/j.1574-6941.2008.00531.x/full
Dear Tom,
the Virtual Taxon (VT) system of the MaarjAM database is defveloped on the SSU fragment between primers NS31 and AM1 (or AML2, ca 10 bo downstream). Use of AML1 primer would add another 200 bp to the beginning of the amplicon, but the rest of it can be matched agains NS31-AM1sequences in MaarjAM (perhaps the majority of accessions), or the full AML1-AML2 amplicon would find a match to a full SSU sequences (there are some). In short, you'd get match between the overlapping parts of query and reference, which does not change the sequence identity % of the aligned portion.
Our team uses currently NS31 and AML2 (not ideal combination in terms of Tm, but good in terms of catching the most variable part of SSU and broad coverage of Glomeromycota clades, plus suitable length for 454-sequencing). With these primers we do get sequences from all Glomeromycota families (have not seen Geosiphon...). Excess of non-specific amplification, mostly of host plant, occurs in our samples in the case of hosts and ecosystems that are not particularly abundant in Glomeromycota biomass, e.g. some crucifers and desert or arctic ecosystems.
I hope this helps
Maarja
Dear Tom,
I very often use AML1-AML2 primer pair and I confirm that it actually covers a higher portion of AMF than the NS31-AM1 primer pair. However, according to the starting matrix you still get some non-target amplification (non-AMF, you get even plant DNA sometimes). Anyway, if you use DNA extracted from roots (a relatively simple matrix) you don't normally get a lot of aspecific amplification. Regarding the lower percentage similarity with the MaarjAM database, I must say I've never seen similarities that were lower than 96% (and normally I get 98-99%, I think it is an already very well implemented database), but I haven't made a very large use of this database so far.
cheers
Andrea
Dear Tom,
the Virtual Taxon (VT) system of the MaarjAM database is defveloped on the SSU fragment between primers NS31 and AM1 (or AML2, ca 10 bo downstream). Use of AML1 primer would add another 200 bp to the beginning of the amplicon, but the rest of it can be matched agains NS31-AM1sequences in MaarjAM (perhaps the majority of accessions), or the full AML1-AML2 amplicon would find a match to a full SSU sequences (there are some). In short, you'd get match between the overlapping parts of query and reference, which does not change the sequence identity % of the aligned portion.
Our team uses currently NS31 and AML2 (not ideal combination in terms of Tm, but good in terms of catching the most variable part of SSU and broad coverage of Glomeromycota clades, plus suitable length for 454-sequencing). With these primers we do get sequences from all Glomeromycota families (have not seen Geosiphon...). Excess of non-specific amplification, mostly of host plant, occurs in our samples in the case of hosts and ecosystems that are not particularly abundant in Glomeromycota biomass, e.g. some crucifers and desert or arctic ecosystems.
I hope this helps
Maarja
Dear Tom,
We also quite frequently use Lee's primers (AML1-AML2). As the region amplified/sequenced completely cover the NS31-AM1 we had no problem with using MaarjAM. The longer sequence might influence blast scores - I do not know whether that is the problem you asked. When we analyze AML1-2 sequences together with NS31-AM1 sequences we just trim the longer AML1-AML2 sequences or put '?' to NS31-AM1s.
Cheers,
Gabor
Dear Tom since the publication of the article by Lee, in the lab we applied the AML1-AML2 primers for both direct and nested PCR amplifications on root and soil DNA samples. I personally found that on roots that are do not very mycorrhized as actually says also Maarja you can get a fairly high percentage of non-specific sequences (especially belonging to the host plant or other fungal contaminants different from Glomeromycota). The same couple of primers used on soil DNA works quite well because cover almost all Glomeromycota groups (but I've never actually amplified sequences belonging to Paraglomus, for which I personally check the presence, both in soil and root, with other primers designed on ITS region, see Renker et al., 2003). Unfortunately the percentage of non-specific sequences vary greatly depending on sample analyzed (from 100% of Glomeromycota sequences in some samples up to soils where the percentage drops to 10% with nonspecific sequences belonging to Viridiplantae, Ascomycota, Chytridiomycota, Zygomycota, Eimeriidae ..... more of everything!). On DNA extracted from roots of plants of agricultural interest (Corn, wheat ...) the percentage of Ascomycota sequences could sometimes reach 30%. Anyway I continue to use this pair of primers for 18S because it allows to use the database MaarjAM and covers both the region of the primers NS31-AM1 (AM2, AM3) that the other hypervariable region amplified by the primers AMV4.5NF/AMDGR (Sato et al.
2005).
bye,
erica
You can try with the primer pairs of NS31 and (AM1+AM2+AM3/equal mixure) it would cover maximum phylotypes of AMF.
- Badges
- Science method