For finding the potential alpha-biodiversity of botanical plant biodiversity of soils as well as agronomical weed infestination, would it be a good idea to count/test in the soil living seeds with genetic markers for plant identification methods?
Botanical plant monitoring is time consuming, time limiteded in the year and delivers different results depending when plants grow and blossoms are built. Different years and spots show also aweful different results.
I often asked me this matter of possibilities testing natural seed bank of soils with gentietcs methods. Or exists already research and literature in this issue linking genetic identification, biodiversity and weed/neobioata forecast?
Could such methods used also for detection of soil pests and invasions of introduced species as neophyts and neobiota?
I am not a genetic marker expert, but the proposed methods looks sparkling or not?
J HUMER
I think, it depends on the weeds that you are screening for. Checking soil for Ambrosia artemisiifolia seeds using genetic barcodes might be useful, e.g. in commercial soil transport.
On the other hand, I'm afraid about the practicability. You need to screen soil probes over a large area, so you would need to have many samples. Furthermore, you would need to take quite an amount of soil to actually get the seeds you are looking for. I don't have experience with obtaining DNA from, let's say, 100 g of soil, but I there might be some technial issue. If one can relyably obtain DNA from such amount of probe, then okay. Then, it's just a question of money whether it's economical to do such a screening instead of monitoring an area to look for plants or to basically ignore eventually harmful neophyte seeds in the gigantic! amount of soil/sand/gravel being shipped for gardening and construction etc.
You may consider to apply some treatment to the sampled soil probe, that stimulates germination of the soil seed bank. I mean a certain combination of moisture, temperature and light to be tested out, e.g. in a glass house. So you may get results a considerable time before the growth season. Most of the seedlings certainly can be identified with conventional methods in a very early stage. However, for some of the seedlings DNA barcoding can make sense for early identification. Estimations of costs can be found in the literature even as low as 2.5 USD per barcode - doesn't sound too expensive.
As it seems, for all problematic weeds there already exist reference barcodes in the barcode library (http://www.boldsystems.org/ ). However, for rare segetal plants, which may be interesting when the focus is on biodiversity, reference data are more sparse or even missing. For example, a barcode is still missing for Asperula arvensis, a species formerly rather widespread but now extinct in Austria, or even for the still rather widespread but rare Adonis flammea.
Herzliche Grüße
Franz Starlinger
In principle this approach sounds feasible to me. There is a nice kit for DNA extraction from dried, ground and homogenized soil samples (http://www.mpbio.com/product.php?pid=116560200) which I used for estimating root biomass of different maize lines a while ago. If the soil seed bank density is not terribly low you should be able to extract sufficient amounts of seed DNA for amplifying a barcoding sequence by PCR. If you have these amplicons NG sequenced with sufficient coverage you should be able to assign different reads to different species. This should be a reasonable estimate for the soil seed bank species composition. Using a correction factor for taking seed size into account you should be able to roughly estimate the frequencies with which species occur in the seed bank as well.
However this isn't as quantitatively reliable as a conventional germination study I guess. And of course the amount of root biomass left in the soil samples needs to be minimized, otherwise it will mess up the entire setup. Moreover, random factors affecting PCR amplification may result in false negative results, especially for lowly abundant species. This could be troublesome if universal barcoding primers like those for trnL are used. In that case probably DNA extraction and also barcode sequence amplification would need to be performed several times for the same homogenized soil sample to find representative average values.
Best regards,
Marcel
- Badges
- Science topic
- Similar topics
- Agriculture
- Soil Science
- Soil