244 Gray and Luin an ordinary agarose gel electrophoresis experiment. For duplex
DNA of this G + C content, the value of v0/[S] is close to 0.80 residues
released/terminus/min at an enzyme concentration of 1.0 U/mL (10).
Assuming a desired rate of 10 residues removed/terminus/min, an
enzyme concentration of 10/0.8 = 12.5 U/mL should be used indepen-
dently of the substrate concentration.
The older data for the S nuclease (6) give very different maximum
velocities/U/L of nuclease and an estimate of Km '~pp that is at least an
order of magnitude lower than those evidenced (although with consid-
erable error) from the more recent experiments (10). Because it turns out
that these data were measured under conditions where the total enzyme
concentration [E]tot actually was in excess of that of the substrate, the
usual assumption of Michaelis-Menten kinetics that [S] >> [E]tot was
not valid. The more recent data do not suffer from that limitation, allow
the dependence of v 0 on G + C content to be assessed (see Section 3.3.3.),
and appear to be quite accurate as long as [S] << Km app a condition that
will be met unless concentrations of hundreds of micrograms per
milliliter of fragments averaging several thousand base pairs in length
are used. Hence, use of the more recent data is recommended.
Recent studies on the duplex exonuclease activity of the F nuclease
(10) give results much more in agreement with the older data (6) and
strongly suggest that the values for gm app are generally more than an
order of magnitude lower than those for the S nuclease, which neces-
sitates the use of Eq. (2) without the assumption that [S] can be dropped
from the right-hand side. Here, values of Kin app and Vmax app were deter-
mined with good accuracy, because it was possible to access substrate
concentrations that were in the range of gm app.
These studies revealed that both gm app and Vm app decrease with
increasing length of the linear duplex substrate for the F nuclease.
Parallel studies could not be done for the S enzyme because of the error
in determination of gm app noted above. These data are consistent with
an interpretation of the mechanism in which nonspecific binding to the
duplex away from the ends is followed by a "search" process to form
a productive complex with nuclease bound at a terminus (10). The
values of Vmax app, but not those of Km app, seem to level off above a
length in the range of 2000-2500 bp. It appears that reasonable kinetic
parameters to use for the F nuclease, for molecules in this size range
or above, are 2.14 + 0.04 nmol/U/min for the maximum velocity/U of