We have developed a quantitative computerized subtraction technique to demonstrate in rat brain the regional distribution of #1 sites, a common very-high-affinity binding site for both morphine and the enkephalins. Autoradiography provides an effective means to accurately map opiate-binding sites within the rat brain and correlate their distribution with brain regions known to mediate opiate actions (1-6). In addition, autoradiographic studies have greatly strengthened the suggestion that there are distinct ,u and 8 receptors (7-9), by demonstrating their unique regional distributions within the brain (10,11). Recently, we proposed that morphine and the enkephalins label three, not two, subpopulations of sites (12,13). In addition to the enkephalinpreferring 8 sites and the classical morphine-selective (p.2) sites demonstrated in the guinea pig ileum (7,14), there is a common site (p.l) to which both morphine and the enkephalins bind with very high affinity. This p.l site, corresponding to the high-affinity site identified in 1975 (15), differs from the pU2 and 8 sites developmentally, phylogenetically, biochemically, and pharmacologically, playing a major role in a number of important opiate actions, including supraspinal analgesia, but not in others, such as respiratory depression and most signs of physical dependence (12)(13)(14)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34).Homogenate binding studies reveal a heterogeneous distribution of Al sites in rat brain (31, 35), but they lack the necessary detailed resolution. Autoradiography, on the other hand, faced a number of difficulties, including the lack of a ligand that could selectively label pul sites. England Nuclear) in the absence or presence of morphine sulfate (5 nM). Nonspecific binding was determined from sections incubated with the radioligand and levallorphan (1 ,uM). A section from each group was stained with hematoxylin and eosin for use in identifying anatomical regions. Slides were washed twice for 5 min each in Tris buffer (0°C), dipped once in distilled water, dried with a stream of cold dry air, placed in an x-ray film cassette along with sections containing known concentrations of tritium (dpm/mg of tissue, dry weight), exposed to LKB Ultrofilm at 4°C with desiccant for 6 months, and developed. Standards were prepared as previously described (39).The autoradiographs of the tissue sections and standards were digitized by using a Vidion-based EYECOM II image processing system linked to a PDP11/23 computer. Multiple standards on each film varied by less than 5%. The correlation of optical densities of the standards to their activity levels (dpm/mg of tissue) were fit to a polynomial equation and used to calculate the concentrations of radioligand in the tissue sections (fmol/mg of tissue). Selected regions of the experimental sections were identified on adjacent sections stained with hematoxylin and eosin. Results represent averages of values determined in two to four areas of two sections of a rat b...