Demonstration of Ac-Arg-Gly-Asp-Ser-NH2 as an Antiaggregatory Agent in the Dog by Intracoronary Administration

This study compared the anti-platelet effect of Ac-RGDS-NH2 which is a peptide fragment from fibrinogen to Ac-RGES-NH2 in which the aspartic acid (D) of Ac-RGDS-NH2 has been replaced by glutamic acid (E). When Ac-RGDS-NH2 was infused intracoronary at concentrations of 100–400 mM, acute platelet dependent thrombus formation in the dog coronary artery was inhibited. However, infusion of Ac-RGES-NH2 intracoronary at similar concentrations to Ac-RGDS-NH2 failed to inhibit platelet dependent thrombus formation in the dog. Ac-RGDS-NH2 and Ac-RGES-NH2 were also tested for their ability to inhibit collagen-induced platelet aggregation in vitro. Ac-RGDS-NH2 elicited concentration-dependent inhibition of collagen-induced aggregation with no effect of Ac-RGES-NHz otr collagen-induced platelet aggregation. Thus, Ac-RGDS-NH2 is an effective antiplatelet agent after intracoronary administration in the dog and also inhibits collagen-induced platelet aggregation in vitro. Ac-RGDS-NH2 is a specific inhibitor of platelet aggregation as replacement of the aspartic acid in Ac-RGDS-NH2 with glutamic acid results in complete loss of biological activity.

Metal-carbon fiber composite femoral stems in hip replacements: a randomized controlled parallel-group study with mean ten-year follow-up

BACKGROUND

Attempts to improve proximal load transfer and minimize stress shielding have included reducing the stiffness of femoral stems and using alternative stem materials, including carbon fiber composites. An uncemented implant (SR71) composed of a carbon-fiber-composite distal section and a porous-coated titanium-alloy proximal section, designed to improve proximal load transfer and provide good fixation, was clinically evaluated in a prospective randomized study.

METHODS

Sixty patients were enrolled and randomized to receive either the SR71 stem or an all-metal stem (Stability). All patients received a cemented all-polyethylene acetabular component and a 28-mm metal femoral head. All uncemented stems were implanted by the same surgeon. Patients were followed for up to ten years with repeated assessments of bone mineral density, radiographs, Harris hip scores (HHS), and visual analog scale (VAS) pain scores.

RESULTS

Ten years postoperatively, nineteen patients who had been treated with the SR71 stem and not lost to follow-up showed a significantly greater increase in proximal bone mineral density (Gruen zones 1 [p = 0.003] and 7 [p = 0.0007]) from baseline than did the twenty-two who had been treated with the Stability stem and not lost to follow-up. In contrast, the Stability group showed a significantly greater increase in distal bone mineral density (Gruen zones 2 [p = 0.0004], 3 [p = 0.0001], and 5 [p = 0.0035]) compared with the SR71 group. Radiographs demonstrated one case of progressive migration of an acetabular component used with an SR71 stem and one case of bone resorption in Gruen zones 7 and 14 in a patient treated with a Stability stem. There was no significant difference between the SR71 and Stability stems in terms of changes in the total HHS, HHS for pain, HHS for range of motion, or VAS pain scores ten years postoperatively relative to preoperative levels. There was one reported revision of an SR71 femoral stem at the ten-year review.

CONCLUSIONS

The investigational SR71 implant provided increased proximal bone density and reduced distal bone density. The implant showed promising results at the time of early follow-up, and the clinical outcomes were similar to those of an all-metal stem at the time of a ten-year follow-up.

The development of potent and selective bisarylmaleimide GSK3 inhibitors

Many 3-aryl-4-(1,2,3,4-tetrahydro[1,4]diazepino[6,7,1-hi]indol-7-yl)maleimides exhibit potent GSK3 inhibitory activity (<100 nM IC(50)), although few show significant selectivity (>100x) versus CDK2, CDK4, or PKCbetaII. However, combining 3-(imidazo[1,2-a]pyridin-3-yl), 3-(pyrazolo[1,5-a]pyridin-3-yl) or aza-analogs with a 4-(2-acyl-(1,2,3,4-tetrahydro[1,4]diazepino[6,7,1-hi]indol-7-yl)) group on the maleimide resulted in very potent inhibitors of GSK3 (</=5 nM) with >160 to >10,000-fold selectivity versus CDK2/4 and PKCbetaII. These compounds also inhibited tau phosphorylation in cells and were effective in lowering plasma glucose in a rat model of type 2 diabetes (ZDF rat).

Comparative analyses of a small molecule/enzyme interaction by multiple users of Biacore technology

To gauge the experimental variability associated with Biacore analysis, 36 different investigators analyzed a small molecule/enzyme interaction under similar conditions. Acetazolamide (222 g/mol) binding to carbonic anhydrase II (CAII; 30000 Da) was chosen as a model system. Both reagents were stable and their interaction posed a challenge to measure because of the low molecular weight of the analyte and the fast association rate constant. Each investigator created three different density surfaces of CAII and analyzed an identical dilution series of acetazolamide (ranging from 4.1 to 1000 nM). The greatest variability in the results was observed during the enzyme immobilization step since each investigator provided their own surface activating reagents. Variability in the quality of the acetazolamide binding responses was likely a product of how well the investigators' instruments had been maintained. To determine the reaction kinetics, the responses from the different density surfaces were fit globally to a 1:1 interaction model that included a term for mass transport. The averaged association and dissociation rate constants were 3.1+/-1.6 x 10(6)M(-1)s(-1) and 6.7+/-2.5 x 10(-2)s(-1), respectively, which corresponded to an average equilibrium dissociation constant (K(D) of 2.6+/-1.4 x 10(-8)M. The results provide a benchmark of variability in interpreting binding constants from the biosensor and highlight keys areas that should be considered when analyzing small molecule interactions.

Movement loci of selected points on the femoral head for individual total hip arthroplasty patients using three-dimensional computer simulation

Wear of ultra-high-molecular-weight polyethylene (UHMWPE) and the subsequent adverse tissue reaction to the wear particles has been cited as the predominant problem affecting total joint arthroplasties. Wear of the UHMWPE acetabular cup in total hip arthroplasties (THAs) is influenced by the sliding distance and direction of individual points on the femoral head, which has as yet been ascertained only for normal subjects. This study seeks to determine the trajectory of specified points on the femoral head for individual THA patients and the distances traversed by these points. A computer program was designed to use gait data from THA patients to simulate the motion of these points. Gait analysis was performed on 19 THA patients at a period of 5 years after operation. The orientation of the acetabular cup was ascertained from radiographs to determine the position of the points on the head most involved in the wear process. The loci of the points differed widely in size, shape, and direction between subjects. The largest average distance traversed was 140% greater than the smallest average distance traversed. Shorter longitudinal paths would tend to cause less wear than larger, wider paths, which cross a greater number of adjacent paths.

Pulmonary delivery of detirelix by intratracheal instillation and aerosol inhalation in the briefly anesthetized dog

Pulmonary delivery of the decapeptide detirelix was studied in briefly anesthetized dogs and the pharmacokinetics were examined following intravenous administration, intratracheal instillation, and aerosol inhalation. Detirelix administrations to the lung gave plasma profiles that were extended over two days, and that differed markedly from those of similarly sized peptides. Absorption from the lung after instillation was slow (Tmax = 6.5 +/- 3.6 h) with a relative bioavailability of 29 +/- 10%. Administration of detirelix-containing aerosols resulted in similar plasma profiles as for administration by instillation. Compartmental and non-compartmental methods of pharmacokinetic analysis indicated no faster absorption from aerosols than from instilled solutions; an absorption rate limiting process may be an explanation. Plasma profiles were not affected by the use of detirelix liquid crystal favoring formulations or destabilizing formulations, and suggested that in situ liquid crystal formation was not an explanation for the slow absorption. No significant changes in pharmacokinetics or systemic uptake were observed during the five-month period of repeated pulmonary administrations. Histopathologic examination revealed the lungs to be essentially normal.

Pharmacokinetics of detirelix following intratracheal instillation and aerosol inhalation in the unanesthetized awake sheep

The unanesthetized awake sheep was employed as large animal model for the determination of bioavailability and pharmacokinetics following the pulmonary instillation of the decapeptide detirelix. After intratracheal administration of a 80 micrograms/kg dose, the average t1/2 of elimination was 9.8 +/- 1.3 hours (n = 5) which was similar to the elimination kinetics of a 30 micrograms/kg i.v. dose (7.2 +/- 2.9 hours). Mean residence time (MRT) was prolonged to 10.3 +/- 2.0 hours vs. 2.7 +/- 0.8 hours i.v., and mean absorption time (MAT) was calculated to be 7.5 +/- 1.8 hours. Maximum plasma levels (cmax) of 9.2 ng/ml were reached after 2 hours. The average bioavailability was 9.8 +/- 3.9% of the dose. The pharmacokinetic profile was found to be similar after aerosol administration. It was concluded that detirelix was absorbed systemically when administered by pulmonary instillation or aerosolization and that the unanesthetized awake sheep is a suitable model to study resulting drug profiles.

Conformationally constrained peptides and semipeptides derived from RGD as potent inhibitors of the platelet fibrinogen receptor and platelet aggregation

Structure-activity studies have been pursued on cyclo-S,S-[Ac-Cys-(N alpha-Me)Arg-Gly-Asp-Pen]-NH2, 2 (SK&F 106760), a potent inhibitor of platelet aggregation, in an effort to improve potency and affinity for the GPIIb/IIIa receptor. Modifications on the N- and C-termini of 2 produced a series of peptides which indicate that the C-terminal carboxylate group may be a secondary receptor-binding element. Further modification by replacing the disulfide tether N alpha-acetylcysteine/penicillamineamide with the novel, inexpensive, achiral, constrained, and more lipophilic tether 2-mercaptobenzoyl/2-mercaptoaniline (Mba/Man) afforded the semipeptide cyclo-S,S-[Mba-(N alpha-Me)Arg-Gly-Asp-Man], 18 (SK&F 107260), which exhibited significant enhancement in both affinity and potency. To further investigate the effect of the phenyl ring at the C-terminus, peptides bearing the novel (2R,3S)- and (2R,3R)-beta-phenylcysteines were synthesized, which culminated in the cyclo-S,S-[Ac-Cys-(N alpha-Me)Arg-Gly-Asp-(2R,3S)-beta-phenylCys]-OH peptide, 22, which displayed substantial affinity and potency. We describe, herein, the development of both 18 and 22 and the additional structural modifications within the constrained cyclic disulfide ring to probe the stereochemical and steric requirements for receptor interaction.

Analysis of the interaction between human interleukin-5 and the soluble domain of its receptor using a surface plasmon resonance biosensor

A surface plasmon resonance (SPR) biosensor was used to study the interaction of human interleukin-5 (hIL5) with its receptor. IL5 is a major growth factor in the production and activation of eosinophils. The receptor for IL5 is composed of two subunits, alpha and beta. The alpha subunit provides the specificity for IL5 and consists of an extracellular soluble domain, a single transmembrane region and a cytoplasmic tail. We expressed the soluble domain of the human IL5 receptor alpha subunit (shIL5R alpha) and human IL5 (hIL5) in Drosophila. Both hIL5 and shIL5R alpha were immobilized separately through amine groups onto the carboxylated dextran layer of sensor chips of the BIAcore (Pharmacia) SPR biosensor after N-hydroxysuccinimide/carbodiimide activation of the chip surface. Interactions were measured for the complementary macromolecule, either shIL5R alpha or hIL5, in solution. Kinetics of binding of soluble analyte to immobilized ligand were measured and from this the association rate constant, dissociation rate constant and equilibrium dissociation constant (Kd) were derived. With immobilized shIL5R alpha and soluble hIL5, the measured Kd was 2 nM. A similar value was obtained by titration calorimetry. The Kd for Drosophila expressed receptor and IL5 is higher than the values reported for proteins expressed in different systems, likely due to differences in the methods of interaction analysis used or differences in protein glycosylation. Receptor-IL5 binding was relatively pH independent between pH 6.5 and 9.5. Outside this range, the dissociation rate increased with comparatively little increase in association rate.(ABSTRACT TRUNCATED AT 250 WORDS)

Synthesis of poly(hydroxypropylglutamine-prazosin carbamate) and release studies

Prazosin, an antihypertensive drug with postsynaptic alpha 1-adrenergic blocking activity, has been coupled to poly-N5-(3-hydroxypropyl-L-glutamine) (PHPG) via a carbamate linkage. PHPG was activated by p-nitrophenyl chloroformate and then reacted with prazosin to form p(HPG-prazosin carbamate) conjugate. Drug loading was 23.9% (w/w). Activated polymer and conjugates were characterized by infrared spectroscopy and differential scanning calorimetry. In vitro studies proceeded in pH 7.4 isotonic phosphate-buffered saline solution. Prazosin was released at a rate of 0.92 mg/day/100 mg conjugate from p(HPG-prazosin carbamate) particles. In vivo studies were performed with New Zealand White rabbits. P(HPG-prazosin carbamate) conjugate particles (100 mg) were suspended in 2 ml saline and injected subcutaneously into both flanks of rabbits. P(HPG-prazosin carbamate) conjugates, following an initial burst, demonstrated a nearly constant plasma prazosin concentration profile above 2 ng/ml, which was maintained for 10 days.

Demonstration of Ac-Arg-Gly-Asp-Ser-NH2 as an antiaggregatory agent in the dog by intracoronary administration

This study compared the anti-platelet effect of Ac-RGDS-NH2 which is a peptide fragment from fibrinogen to Ac-RGES-NH2 in which the aspartic acid (D) of Ac-RGDS-NH2 has been replaced by glutamic acid (E). When Ac-RGDS-NH2 was infused intracoronary at concentrations of 100-400 mM, acute platelet-dependent thrombus formation in the dog coronary artery was inhibited. However, infusion of Ac-RGES-NH2 intracoronary at similar concentrations to Ac-RGDS-NH2 failed to inhibit platelet-dependent thrombus formation in the dog. Ac-RGDS-NH2 and Ac-RGES-NH2 were also tested for their ability to inhibit collagen-induced platelet aggregation in vitro. Ac-RGDS-NH2 elicited concentration-dependent inhibition of collagen-induced aggregation with no effect of Ac-RGES-NH2 on collagen-induced platelet aggregation. Thus, Ac-RGDS-NH2 is an effective antiplatelet agent after intracoronary administration in the dog and also inhibits collagen-induced platelet aggregation in vitro. Ac-RGDS-NH2 is a specific inhibitor of platelet aggregation as replacement of the aspartic acid in Ac-RGDS-NH2 with glutamic acid results in complete loss of biological activity.

Animal model for evaluating the convulsive liability of beta-lactam antibiotics

The beta-lactam antibiotics imipenem-cilastatin, BMY-26225, and cefazolin significantly lowered the convulsive threshold of pentylenetetrazole in mice. In addition, imipenem-cilastatin and cefazolin were found to inhibit 3H-labeled gamma-aminobutyric acid binding to synaptic membranes from rat brains. Our results suggest that the pentylenetetrazole convulsive model may be useful in evaluating the proconvulsive liabilities of new carbapenems and other beta-lactam antibiotics and that the mechanism of imipenem-cilastatin and cefazolin toxicity may involve interaction with gamma-aminobutyric acid receptors.

Coupling of naltrexone to biodegradable poly(alpha-amino acids)

The narcotic antagonist naltrexone (I) was modified at the 3 and 14 hydroxyl positions and covalently coupled to a biodegradable poly(alpha-amino acid) backbone through a labile bond. Selective acetylation of I with acetic anhydride gave naltrexone-3-acetate (II), which was subsequently succinoylated to naltrexone-3-acetate-14-hemisuccinate (III) with succinic anhydride. The polymeric backbone chosen for initial coupling experiments was poly-N5-(3-hydroxypropyl)-L-glutamine (PHPG). The side-chain hydroxyl functionality permitted covalent bonding of III through an ester linkage. Hydrolysis of covalently bound drug to give naltrexone or its derivatives (II and III) should be much slower than diffusion of drug through the polymer matrix. While hydrolysis of naltrexone from the polymer side chain is first order, release of drug from the matrix can be zero order due to the geometry of the device and the physical and chemical interactions between naltrexone and the polymer matrix. In vitro studies of PHPG-naltrexone conjugate in disk form did not show constant release because of the hydrophilic nature of the polymer backbone and the changing local chemical environment upon hydrolysis of drug-polymer linkages. The conjugated system was made more hydrophobic by coupling drug to copolymers of hydroxypropyl-L-glutamine (HPG) and L-leucine. Conjugates of III coupled with copoly(HPG-70/Leu-30) demonstrated a nearly constant, but slightly declining release rate of naltrexone and its derivatives for 28 days in vitro.

Correlation between renal membrane binding and nephrotoxicity of aminoglycosides

The kinetics of aminoglycoside binding to renal brush border and basolateral membrane vesicles from rat renal cortex were studied by using [3H]amikacin. [3H]amikacin binding to renal membranes was found to be a rapid, saturable process with a fourfold greater affinity for basolateral membranes than for brush border membranes (Kd basolateral = 607 microM; Kd brush border = 2,535 microM). Renal membranes prepared from immature rats (2 to 3 weeks old) exhibited a significantly lower affinity compared with membranes from adults (Kd basolateral = 2,262 microM; Kd brush border = 6,216 microM). Additionally, the inhibitory behavior of several aminoglycosides versus [3H]amikacin binding to brush border membranes revealed the following rank order of potency: neomycin greater than tobramycin approximately gentamicin approximately netilmicin greater than amikacin approximately neamine greater than streptomycin. The relative insensitivity of immature rats to aminoglycoside-induced nephrotoxicity in vivo and the comparative nephrotoxicity of the various aminoglycosides suggest that renal membrane-binding affinity is closely correlated to the nephrotoxic potential of these antibiotics.

Inhibition of renal membrane binding and nephrotoxicity of aminoglycosides

The initial event in the renal tubular reabsorption of nephrotoxic aminoglycosides involves binding to brush border membranes. This primary event was measured in renal brush border membrane vesicles prepared from rat renal cortex utilizing [3H]gentamicin. In order to gain structure-activity information regarding this interaction the effect of substances having chemical similarities to aminoglycosides (sugars, polyamines and amino acids) on gentamicin binding to brush border membranes was determined. Polyamino acids were found to possess the greatest inhibitory potency. In addition to polymers of cationic amino acids (lysine, ornithine, arginine and histidine), polymers of neutral (asparagine) and acidic (aspartic and glutamic acid) amino acids also exhibited inhibition of the membrane binding of gentamicin. Inasmuch as inhibition of renal membrane binding has the potential to decrease aminoglycoside nephrotoxicity, several polyamino acids that inhibited membrane binding were tested in vivo for potential protective activity vs. gentamicin- and amikacin-induced nephrotoxicity. Polyasparagine90 and polyaspartic acid100 inhibited gentamicin and amikacin nephrotoxicity completely when coadministered to rats with the aminoglycosides. Polylysine20 provided complete and partial inhibition of gentamicin and amikacin nephrotoxicity, respectively. Whereas in vivo distribution studies revealed that cortical levels of [3H]amikacin were elevated slightly by the coadministration of polyaspartic acid, brush border and basolateral membranes contained significantly lower levels of the aminoglycoside (46 and 41% inhibition, respectively). These results question the role of charge per se in the binding of aminoglycosides to renal membranes and further confirm the importance of membrane binding in the pathogenesis of aminoglycoside nephrotoxicity.

Stereospecific opiate-binding sites occur in coated vesicles

We prepared clathrin-coated vesicles from bovine forebrain utilizing sucrose or deuterium oxide-Ficoll density gradient centrifugation followed by permeation chromatography. Homogeneity was monitored by electron microscopy (EM) and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). EM revealed that the predominant (up to 98% of the total) organelles were coated vesicles and empty hexagonal baskets. Diameters of the coated vesicles ranged from 37 to 120 nm with a mean of 65.2 +/- 2.2. Upon SDS-PAGE of the coated vesicle fraction, the most prominent band appeared at 180,000 daltons. There were also three additional bands at 100,000, 50,000 and 35,000 daltons, giving the overall pattern characteristic of coated vesicles. Both 0.5 nM tritiated naltrexone and etorphine displayed specific binding to coated vesicles. Naltrexone binding in coated vesicles from gradient fractions was increased 2.5-fold over the original 100,000 X g pellet. An additional 4-fold enrichment in specific binding was observed after permeation chromatography which was concomitant with an increase in the volume density of coated vesicles in electron micrographs. Naltrexone binding was stereospecific and etorphine binding was inhibited by 100 mM NaCl (40%). Both naltrexone and etorphine binding were inhibited by 50 microM guanyl-5'-yl imidodiphosphate (40 to 50%). In summary, purified bovine brain-coated vesicles contained high affinity stereospecific opiate alkaloid-binding sites with characteristic opioid binding properties.

Kinetic evidence for differential agonist and antagonist binding to bovine hippocampal synaptic membrane opioid receptors

To examine the kinetics of opioid receptor binding, the agonists [D-Ala2-D-Leu5]enkephalin (DADL) and [D-Ala2-MePhe4-Gly-ol5]enkephalin (DAGO) and the antagonists diprenorphine and naltrexone were used with bovine hippocampal synaptic plasma membranes. By computer modeling of equilibrium binding displacement curves utilizing the LIGAND program, we found opioid peptides bind with high affinity to single populations of synaptic plasma membranes receptors, whereas opiate alkaloids bind to multiple sites. Initial kinetic experiments revealed that agonist rates of association were radioligand concentration-independent. Pseudo first-order rate constants for DADL, DAGO, diprenorphine, and naltrexone association were estimated to be 5.63 X 10(5), 5.08 X 10(5), 4.60 X 10(6), and 2.3 X 10(6) mol-1 X s-1, respectively. After preincubation of 0.2-1 nM radioligand for variable time intervals, dissociation was initiated by addition of 1 microM unlabeled ligand. If saturation binding was achieved before dissociation was initiated, then nearly monophasic dissociation of DADL, DAGO, and diprenorphine and a biphasic off-rate for naltrexone were observed. When association times were reduced to pre-equilibrium intervals, the kinetics of dissociation of agonists became biphasic and association time-dependent, but that for antagonists did not change significantly. Comparisons by both graphical methods and computerized nonlinear regression analyses of rate constants revealed that the fraction of the rapid component of agonist dissociation decreases and that of the slow component is elevated with increasing receptor occupancy. In the presence of 100 mM NaCl, DADL dissociation became association time-independent. These data are consistent with the idea that the Na+ effect is brought about by a change of receptor to an antagonist-like conformation. On the basis of both association and dissociation kinetic data, opioid agonists appear to interact in a multistep process in which a rapid, reversible association is followed by the formation of a more tightly bound complex.

Differential inhibition of activated tyrosine hydroxylase

Under conditions of cyclic AMP-dependent protein phosphorylation, tyrosine hydroxylase (EC 1.14.16.2; TH) is activated. Kinetic analysis reveals that, upon activation the affinity of the enzyme cofactor tetrahydrobiopterin, Vmax, as well as the Ki of its putative feedback inhibitor dopamine, are increased. Catecholic inhibitors of rat striatal TH have been assessed for the structural requirements that impart differential sensitivity to activated and control enzyme. By varying cofactor and inhibitor concentrations, Ki's were generated from Dixon plots. Structural analogs of dopamine in which the amino group was fixed in a cis conformation, i.e., 6,7-dihydroxytetrahydroisoquinolines, exhibit the same Ki for activated and nonactivated TH. However, 2-amino-6,7-dihydroxytetralin (ADTN), in which the nitrogen is extended in a fixed trans conformation of the beta-rotamer, exhibited a fourfold increase in Ki upon assaying tyrosine hydroxylase under phosphorylation conditions. By systematically increasing the hydrophobicity of the substituent at C-1 of 1-carboxy-6,7-dihydroxytetrahydroisoquinolines the inhibitory potency was enhanced, suggesting the presence of a hydrophobic region near the catecholic binding site. If the hydrophobic group was rigid as in the catechol estrogens, 2-hydroxy-estradiol and 2-hydroxyestrone, the Ki was relatively low (2 X 10(-5) M) despite the absence of an amino group. Upon activation the Ki increased fourfold. These studies provide insight into the topography of the catecholic binding site on TH and to attendant changes occurring upon activation. The results suggest that the catechol binding site includes both amino group-interacting and hydrophobic regions which are influenced by enzyme activation.

Ontogeny of benzomorphan-selective (kappa) sites: a computerized analysis

In an investigation of the postnatal development of kappa opiate receptors, the affinity and capacity of 0.5 nM [3H]-ethylketocyclazocine (EKC) binding in crude rat brain homogenates was measured by displacement with unlabeled EKC, morphine, or D-ala2-D-leu5-enkephalin (DADL). Displacement curves were analyzed using a weighted, non-linear regression, curve fitting computer program. At all stages of development, [3H]-EKC binding fit a two site model significantly better than a one site model. Affinities of EKC, morphine, or DADL for the high affinity [3H]-EKC binding site did not change during the postnatal period. The density of the high affinity [3H]-EKC binding site increased linearly with age, whereas the levels of the low affinity site rose more rapidly during the second week.