3
Fig. 4-14. Nonlinear regression analysis of the binding of [ H]dexamethasone to
glucocorticoid receptors in perfused mouse brain cytosol (derived from the
total binding data, BT, of figure 4-7 and displayed in the Scatchard coordinate
system).
Experimental conditions are indicated in the legend to figure 4-7, and the use of the
nonlinear regression program is described in Methods. The model to which the data were fit
consists of the sum of one saturable high-affinity (specific) component and one
nonsaturable (nonspecific) component as described by equation (4-6). The actual regression
3
equation does not contain F; it accepts B-j. and (total [ Hjdexamethasone concentration)
as pairs of input variables and then regresses B^ on the relatively error-free variable S^,
fitting the adjustable binding parameters K^, Bq, and C-j (the asymptotic "sink" of
nonspecific binding). Each input data point was weighted as 1/B-j-^, consistent with the
assumption that the coefficient of variation or percentage error in the measurement of B^
is constant over the range of B^.. For display the resulting complete 3-parameter
regression model was then plotted in the Scatchard coordinate system (curved line), and
each component was also plotted separately (2 straight lines). (Note that the x-axis is
plotted below y=0 in the figure.) Equilibrium dissociation constant K. = 1.1 nM, and
Q c. i
receptor concentration Bq = 2.4 nM (370 fmole/mg cytosol protein). = 2.4 x 10 M .
Agreement with the binding parameter estimates derived from the Scatchard plot of specific
binding fit by simple linear regression (figure 4-10) is excellent.