The melting temperature of the poly(dA) . poly(dT) double helix is exquisit
ely sensitive to salt concentration, and the helix-to-coil transition is sh
arp. Modem calorimetric instrumentation allows this transition to be detect
ed and characterized with high precision at extremely low duplex concentrat
ions. We have taken advantage of these properties to show that this duplex
can be used as a sensitive probe to detect and to characterize the influenc
e of other solutes on solution properties. We demonstrate how the temperatu
re associated with poly(dA).poly(dT) melting can be used to define the chan
ge in bulk solution cation concentration imparted by the presence of other
duplex and triplex solutes, in both their native and denatured states. We u
se this information to critically evaluate features of counterion condensat
ion theory, as well as to illustrate "crosstalk" between different, non-con
tacting solute molecules. Specifically, we probe the melting of a synthetic
homopolymer, poly(dA).poly(dT), in the presence of excess genomic salmon s
perm DNA, or in the presence of one of two synthetic RNA polymers (the poly
(rA).poly(rU) duplex or the poly(rU).poly(rA).poly(rU) triplex). We find th
at these additions cause a shift in the melting temperature of poly(dA).pol
y(dT), which is proportional to the concentration of the added polymer and
dependent on its conformational state (B versus A, native versus denatured,
and triplex versus duplex). To a first approximation, the magnitude of the
observed t(m) shift does not depend significantly on whether the added pol
ymer is RNA or DNA, but it does depend on the number of strands making up t
he helix of the added polymer. We ascribe the observed changes in melting t
emperature of poly(dA) . poly(dT) to the increase in ionic strength of the
bulk solution brought about by the presence of the added nucleic acid and i
ts associated counterions. We refer to this communication between non-conta
cting biopolymers in solution as solvent-mediated crosstalk. By comparison
with a known standard curve of t(m) versus log[Na+] for poly(dA).poly(dT),
we estimate the magnitude of the apparent change in ionic strength resultin
g from the presence of the bulk nucleic acid, and we compare these results
with predictions from theory. We find that current theoretical consideratio
ns correctly predict the direction of the t(m) shift (the melting temperatu
re increases), while overestimating its magnitude. Specifically, we observe
an apparent increase in ionic strength equal to 5 % of the concentration o
f the added duplex DNA or RNA (in mol phosphate), and an additional apparen
t increase of about 9.5 % of the nucleic acid concentration (mol phosphate)
upon denaturation of the added DNA or RNA, yielding a total apparent incre
ase of 14.5 %. For the poly(rU) . poly(rA).poly(rU) triplex, the total appa
rent increase in ionic strength corresponds to about 13.6 % of the amount o
f added triplex (moles phosphate). The effect we observe is due to coupled
equilibria between the solute molecules mediated by modulations in cation c
oncentration induced by the presence and/or the transition of one of the so
lute molecules. We note that our results are general, so one can use a diff
erent solute probe sensitive to proton binding to characterize subtle chang
es in solution pH induced by the presence of another solute in solution. We
discuss some of the broader implications of these measurements/results in
terms of nucleic acid melting in multicomponent systems, in terms of probin
g counterion environments, and in terms of potential regulatory mechanisms.
(C) 2001 Academic Press.