Re: [compost_tea] The soil metagenome: Methods review
You have to understand what is being measured, exactly, in order to know if
the tests used in molecualr biology measures what you want measured.
What is the question you want answered? Can these molecular methods do what
you want?
Let me review what the methods in Steve's e-mail are about. All of these
involve methods to assess nucleic acids, sequences of nucleic acids, and genetic
relationships between organisms.
rDNA sequencing
DNA extraction and PCR analysis
real-time PCR analysis
PCR (Polymerase Chain Reaction)
gel electrophoresis
gene expression
phylogenetic mapping
rDNA means ribosomal DNA. In eucaryotes, ribosomes are housed in the
mitochondria. You have mitochondrial nucleic acids in eucaryotes, which are more
highly conserved than the genomic DNA, as I understand.
Procaryotes do not have mitochondria, of course, so they don't have this
additional source of nucleic acid material. Our ability to understand genetic
relationships is therefore much more limited in bacteria and virus, since this
additional set of nucleic acids is not present in procaryotes.
DNA extraction is of course, just extracting DNA from any material.
Extracting DNA from mitochodria would be rDNA. Otherwise, as DNA is translated and
multiplied, it is RNA, since cells use different nucleic acids in producing the
messages inside cells than for making the material of the genome.
Extracting DNA from soil was a major problem for many years, and still messes
certain people up, as they haven't paid attention to the fact that clay
adsorbs nulceic acids quite well. ANY protein will be rapidly adsorbed on clay
surfaces. If the DNA you extract is sucked up by this material, the researcher
can, and it has happened, come to the comclusion that soil has low diversity.
Not the case - the methods were faulty.
Once you extract DNA, you have to multiply it. In order to work with DNA,
you need lots of it. But you have to get the sequences the same as were in the
original material. So, PCR is used to copy the sequences of DNA so where you
have one strand of DNA, you now have millions.
That's all that PCR is: a replication process. Polymerase is an enzyme that
copies nucleic acid sequences. But, if you extract DNA from all cells in the
soil, if you have one individual of one species, you will have only one DNA
sequence of that kind. And if you replicate each strand of the 600 million DNA
strands (the typical number of bacteria in a healthy soil, each bacterium
presumably having one genome) by a million times, you still only have one strand
of that DNA per 600 million strands. Will that strand be detected in the
analysis of the sequences?
Now, as we want to have our PCR product rapidly, we have moved to real-time
PCR, which has reduced the time for the whole multiplication process to take
place. But when we hurry things, guess what gets lost?
So, real-time diversity? Or actual diversity? When we to gel
electrophoresis of the products from PCR, are we likely to assess the full diversity of the
DNA strands? Not if we have hurried things along.
Going fast allows us to get an idea of the dominant species, not the full
diversity. Can there be a problem here, when interpreting this information? If
you gloss over the details, yes, conclusions can be made that are entirely
inappropriate. Care must be exercised in conclusions that are made.
Gene expression. Genetic material is not always expressed. For example, you
may have the genetic material which allows you to play the violin like a
virtuoso, but if you have never picked up a violin, that genetic material will
never be expressed. Yes, it's a bit more complicated than that, as practice is
involved, but I think you get what I mean. Genetic material is not always
used, not always expressed.
So, for example, when people want to use fatty acid expression to assess
bacterial species diversity, they often incorrectly interpret the data, because
bacteria do not always express the fatty acids of interest. Think of the errors
that are likely to ocurr.
But can assessing fatty acids gives us a good idea of diversity?
Show me data where full diversity has been assessed, and compare that to
fatty acid information. Good correlation, or not? Sometimes.... sometimese not.
So, when is the method going to be good, and when isn't it? I am not aware
of such analysis. If anyone is aware of such analysis, I'd love to get that
paper! But until then, use fatty acid analysis to make conclusions about
diversity with a solid grain of salt.
OK, back to the nucleic acid world.
Molecular biologists have started to use RNA as a method of assessing
organism diversity. EXCEPT, RNA is only produced when a gene is being expressed.
So, mRNA (messaenger RNA), rRNA (ribosomal RNA), whatever, are only produced
when DNA is being copied and translated.
So, can RNA ever be a method to assess diversity? No, most of the genes in
any genome are not being expressed at any time.
Can you use RNA probes to tell you if a bacterium is present, or not? What
if the RNA probe you are using is for an enzyme that is not being expressed by
that bacterium at this moment? You are in trouble.
So, for E. coli, for example, if E. coli is not active, it does not express
beta-galactosidase, does not make the RNA that is what the molecular probe
attaches to. Is this a problem? Yes, if you need to know if any E. coli is
present.
But maybe we only need to know about ACTIVE E. coli, in which case the RNA
probe would be fine.
Well, I could go on, but this is enough. Probably too much. I am always
happy to discuss this with folks, since there is more that I could learn about
these things.
But make it a discussion. No attacks on the people trying to understand or
share knowledge. And, as always, if I have made any errors of omission or made
a mistake in my information, a gentle correction is the most reasonable way
to start a discussion about any point.
Elaine R. Ingham
Soil Foodweb Inc., Corvallis, Oregon
Soil Foodweb Inc., Port Jefferson, New York
Soil Foodweb Institute, Lismore Australia
Soil Foodweb Institue Cambridge, New Zealand
Soil Foodweb Inc., Hilversum, The Netherlands
Laboratorios de Soil Foodweb, Culiacan, Mexico
Soil Foodweb Inc., Jerome, Idaho
Received on Wed Aug 18 2004 - 19:40:18 EDT
This archive was generated by hypermail 2.3.0
: Tue Feb 07 2012 - 14:15:27 EST