Tm of DNA denaturation is surely DNA concentration dependent. If you
visualize the denaturation and reannealing process as a reversible
reaction, it is easy to see that a basic assumption is taken in determining
the Tm: steady state equilibrium. It is assumed that you do not take
away the denatured DNA during Tm determination. See what if you constantly
remove the denatured single stranded DNA ? Your double stranded DNA will
eventually fully denatured, regardless of the temperature (practical temp.
Then, you may think of hybridization. Higher single stranded DNA will
favour hybridization. This approximates to a second order kinetics.
But remember this is a kinetics problem. Rate of reaction have nothing to
do with equilibrium state. High temperature favours ssDNA state, and low
temperature favours dsDNA state. High temperature increase rate of molecular
bombardment, thus increase rate of re-annealing.
Going back to my most original question. I wish to determine the washing
condition of oligonucleotide hybridization. This is also a kinetics
problem. In membrane washing, the free ssDNA concentration is kept
practically at zero. Allowed you wash the membrane with sufficient time,
no matter what temperature you use (practical temp. :-) ), all the annealing
oligonucleotide probe will be washed away. Thus, the timing of washing is
important. Usually, the washing temperature is sufficiently low compared
with the Tm so that a double stranded state is strongly favoured. Extensive
washing can that be done, with no appreciate quantity of annealed DNA probe
being washed away. However, in oligonucleotide hybridization, this washing
step is tricky. In order to wash away the probe with single mismatch, a
washing condition closer to the dissociation state then ordinary washing
should be used. Thus, you can either extend the washing of the membrane,
or choose a higher temperature. A compromise between ligitimate binding
and single mismatch binding should be determined. This is an optimization
process. You wash away the mismatch binding, but wish to retain the
ligitimate binding. This process is governed by the thermodynamic laws of
energetics and kinetics. Thus the optimum state can theoretically be
predicted if 1) the binding energy 2) dissociation constant are known.
Both of these parameters can be determined experimentally. The
neighbourhood binding entropy is a measure of the first parameter. But for
the second parameter, I have no information.
Obviously, all the computer programs I encountered (including "primer"
"oligo" etc.) calculate the Tm only. The kinetics problem, though
important, is not considered. If, I am saying if, such program is available,
not only membrane washing can be optimized, the PCR process may be more
finely optimized with respect to time in each segment.