Proc. Nati. Acad. Sci. USA Vol. 77, No. 4, pp. 2219-2223, April 1980 Medical Sciences

Macromolecular l-adrenergic antagonists discriminating between receptor and antibody (polymeric drugs/antihormone antibodies/,-adrenergic receptors/steric strain/dextrans) JOSEF PITHA*, JORDAN ZJAWIONY*t, ROBERT J. LEFKOWITZt, AND MARC G. CARONt *National Institute on Aging, National Institutes of Health, Gerontology Research Center-Baltimore City Hospitals, Baltimore, Maryland 21224, tInstitute of Organic Chemistry, Technical University (Politechnika), Zwirki 36, 90-924 Lodz 40, Poland; and tHoward Hughes Medical Institute Laboratories, Department of Medicine and Biochemistry, Duke University Medical Center, Durham, North Carolina 27710 Communicated by Curt P. Richter, December 7,1979 ABSTRACT The P-adrenergic antagonist, alprenolol, was Dihydroalprenolol-SA-dextran. A solution of sodium hy- attached in an irreversible manner to macromolecular dextran droxide (2 g) in water (10 ml) was added to the solution of via side arms that differed in length. The ability of these mac- dextran romolecules to bind to the #-adrenergic receptor of frog eryth- (10 g) in water (100 ml). Epichlorohydrin (5.5 g) was rocytes and to catecholamine-binding antibies raised against added and the mixture was stirred at 60'C for 2 hr. After partially purified receptors was studied. Compared to the parent neutralization with saturated aqueous monosodium phosphate, dug the potency of binding of macromolecular alprenolol to 1 M sodium thiosulfate (60 ml) was added and the mixture was the receptor decreased about 1/10, 1/600, and 1/8000 when the stirred overnight at room temperature. After dialysis and length ofthe arm separating alprenolol from the dextran moiety clarification the supernatant was freeze-dried. The resulting was 13, 8, and 4 atoms, respectively. In contrast, the binding potencies of the parent drug and of all its macromolecular de- material (9.96 g) had 1.54% S-i.e., 1 mg contained 0.24 Aumol rivatives for the antibody were within the same order of mag- of thiosulfate residue. Sodium bicarbonate (5 ml, 0.1 M) was nitude. Thus, conversion of a drug to a macromolecular form added to the solution of thiosulfate dextran derivative (1 g, 0.24 may not only sustain its binding activity but may also leal to mmol of thiosulfate residue) in water (10 ml), followed by the a higher selectivity. The macromolecular derivatives described addition of dithiothreitol (462 mg, 3 mmol) in aqueous EDTA here may be suitable probes for investigation of the location and (4 ml, 1 mM). The mixture was stirred for 40 min at room of the molecular properties of the binding sites for P-adrenergic temperature and then dialyzed for 2 hr against nitrogen-purged drugs. water, and the pH of the solution was adjusted to 5 with acetic Conversion of a drug into macromolecular form affects its acid. Alprenolol hydrochloride (257 mg, 0.9 mmol) was then distribution and fate in the organism (1-4). Hydrophilic mac- added and the solution was stirred at 950C for 2 hr. During this romolecules are barred by the lipid bilayer from penetrating time a solution of potassium persulfate (243 mg, 0.9 mmol) in into cells and remain in the extracellular fluid until endocytosed water (1 ml) was added. After cooling, sodium borohydride (35 or excreted. For drugs that act on target tissues by binding to mg, 0.9 mmol) was added and the solution was stirred over- specific receptors located on the cell surface, the exclusion from night. After dialysis and clarification, freeze-drying gave 0.65 the cell interior and longer circulation time may be of distinct g of the material which had maxima in the ultraviolet spectrum pharmacological advantage. In this work we have studied the identical to alprenolol (271 and 278 nm). The material had f3-adrenergic blocking drug alprenolol (Scheme I). This drug 0.51% S-i.e., 1 mg contained 0.16 Amol of S. From absorbancy binds to f3-adrenergic receptors located on the cell surface (5) data it was calculated that 1 mg of material contains 0.21 ,mol and can also bind with similar affinity to catecholamine-binding of alprenolol; by calculation the material should contain 0.29% antibodies that have been raised against partially purified re- N, but 0.73% N was actually found. ceptor preparations (6). Alprenolol was covalently bound to Dihydroalprenolol-MA-dextran. Aqueous sodium hydroxide dextran-and, by varying the length of the linkage arm between (38 ml, 40%) and 2-aminoethyl hydrogen sulfate (40 g) were the drug and soluble carrier (Scheme I; LA, longarm; MA, added to a solution of dextran (5 g) in water (5 ml). The mixture medium arm; SA, short arm), we obtained macromolecules with was placed in a pressure bottle and heated overnight at 900C. different affinities for 13-adrenergic receptor and the cate- After cooling, the mixture was neutralized with diluted hy- cholamine-binding antibodies. The ability of a modified drug drochloric acid and dialyzed at first against aqueous NaCl (10%) to discriminate between related binding sites such as the f3- and then against water. Centrifugation and freeze-drying of adrenergic receptor and these antibodies may find applications the supernatant gave 4.69 g of the material that had 1.59% in analytical measurements or in eventual therapeutic man- N-i.e., 1 mg of material contained 1.1 limol of aminoethyl agement of certain diseases. group. Carbonate buffer (20 ml, 1 M, pH 9.7) was added to the so- MATERIALS AND METHODS, lution of 2-aminoethyldextran (2 g) in water (20 ml). The Materials. Dextran of average Mr 40,000 and racemic al- mixture was cooled to 40C, N-acetyl-DL-homocysteinethio- prenolol hydrochloride were used in all experiments. Unless lactone (1 g) was added, and stirring at 4VC was continued otherwise indicated dialyses were against water at 4VC and in overnight. Dialysis followed by clarification and freeze-drying an exhaustive manner. Clarifications were by centrifugation yielded 1.44 g of product that contained 0.96% S and 0.61% at 10,000 rpm for 10 min. N-i.e., 1 mg of material contained 0.3 Amol of N-acetylho- mocysteine residue and 0.13 ,mol of aminoethyl group. When The publication costs of this article were defrayed in part by page the same synthesis was performed at room temperature, 1.8 g charge payment. This article must therefore be hereby marked "ad- of the material was obtained, containing 1.26% S. The latter vertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact. Abbreviations: LA, long arm; MA, medium arm; SA, short arm. 2219 Downloaded by guest on September 28, 2021 2220 Medical Sciences: Pitha et al. Proc. Natl. Acad. Sci. USA 77 (1980)

CHI=CH -CH2 ,+ Alprenolol O-CH2-CHOH-CH2-NH-CH (RI I CH3

CH3-CH2-CH2 -- (RI

Dihydroalprenolol D e x t -O-CH2-CHOH-CH2-S-CH2-CH2 CH2- IRI r a Dihydroalprenolol-SA-dextran n CH3 D CO e x NHIH t -O-CH2-CH2-NH-CO-CH-CH2-CH2-S-DC12-CH2 CH2 IRI r a Dihydroalprenolol-MA-dextran D n e x t-O-CH2-CHOH-CH2-O-CH2-O12-CH2-CH2-0-CH2-CHOH-CH2-S-CH2-CH2-CH2--- (RI r a Dihydroalprenolol-LA-dextran n Scheme I

product was freshly reduced with dithiothreitol and condensed starting pH was adjusted to 7, the condensation reaction lowered with alprenolol as described for dihydroalprenolol-SA-dextran. it to 6, but amounts of alprenolol incorporated were only half From 0.4 g of material (0.3 mmol of mercapto group), we ob- of the above. tained 0.25 g of the product that had alprenolol-type maxima Binding of [3HJDihydroalprenolol and Macromolecular in ultraviolet spectrum (271 and 278 nm) and had 0.87% N and Alprenolol to Receptors and Antibodies. [3H]Dihydroalpre- 0.64% S. From that data it was calculated that 1 mg of material nolol binding to the various preparations was assessed as de- contained 0.13 /Amol of alprenolol residue, 0.2 ,umol of total S, scribed (6-8) with the following modifications. Purified frog and 0.29 gmol of aminoethyl group. erythrocyte membranes (7) suspended in 25 mM Tris-HCI, pH Dihydroalprenolol-LA-dextran. A solution of sodium hy- 7.4/2 mM were incubated at 25°C for 30 min with in water Mg9l2 droxide (2 g) (10 ml) and 1,4-butanediol diglycidyl [3H]dihydroalprenolol in the presence and absence of various ether (10 ml) were added to a of (10 in water solution dextran g) concentrations of competing ligands. Bound radioactivity was (50 ml). The mixture was stirred for 90 min at room tempera- determined by rapid collection of the incubation mixtures on ture and neutralized with saturated aqueous monosodium glass fiber filters. Binding to the antiserum (no. 349, heat in- phosphate. Sodium thiosulfate (60 ml, 1 M) was then added and activated at 56°C for 30 min) was determined as described the mixture was stirred overnight at room temperature. The above because it was found that the antibodies could be retained resulting alkaline solution (pH 11.5) was again neutralized, on glass fiber filters. Binding of [3H]dihydroalprenolol to sol- dialyzed, clarified, and freeze-dried as above. The product (9.11 ubilized preparations of frog erythrocyte ,3-adrenergic receptor g) contained 1.53% S-i.e., 1 mg of material contained 0.24 in 1% digitonin/100 mM NaCl/10 mM Tris-HCI, pH 7.4 was ilmol of thiosulfate residue. The reduction of the product and assessed at condensation with alprenolol was performed as above. From 4°C by incubating [3H]dihydroalprenolol (5-10 nM) with and without competing of receptor- 1 g (0.24 mmol of thiosulfate residue) of starting material, 0.8 drugs. Separation g of the product was obtained, which had maxima in ultraviolet bound [3H]dihydroalprenolol from free material was performed spectrum (271 and 278 nm) characteristic of alprenolol and on Sephadex G-50. were contained 0.96% S. From this data, it was calculated that 1 mg Stock solutions of the alprenolol derivatives prepared of material contained 0.08 ,umol of alprenolol and 0.3 ,mol of in water on the basis of their alprenolol residue content, which total S and should contain 0.11% N; 0.44% was found. was determined spectrophotometrically. Solubility of these The effects of the amount of persulfate and pH on the con- materials decreases upon storage but can be fully restored by densation with alprenolol were investigated in detail. When the mild reduction (e.g., 5 mM solution of dithiothreitol). The de- molar ratio of persulfate to thiol groups was 3, 2, 1, and 0.5 the crease in solubility is probably due to the formation of disulfide product contained 0.084, 0.077, 0.070, and 0.056 jumol of al- crosslinks by autooxidation of these mercapto groups, which prenolol residue per mg, respectively. The pH in these mixtures were not substituted in the Posner reaction. No free drug was was originally 5 but dropped during the reaction to the final released from the macromolecular preparations on storage or values of 1.80, 1.80, 2.00, and 2.70, respectively. When the air oxidation. Downloaded by guest on September 28, 2021 Medical Sciences: Pitha et al. Proc. Natl. Acad. Sci. USA 77 (1980) 2221

4o 4.z 0 c 0 U u 0o 100 0 A 60 m c so 0 0 .0P) .5 60 F .5OL 0 C 01o * 'O .5a. 40 H _s 0 _ > 20- 20 fn

0o jL L-I A U ~~10-9 i 0 8 10-7 god6 lo-, 10-4 10-9 i 0-8 1-0-7 10o6 i O-S 10-4 Alprenolol residue, M Alprenolol residue, M FIG. 1. Competition of (+) alprenolol and its macromolecular FIG. 3. Competition of (+) alprenolol and its macromolecular derivatives with [3H]dihydroalprenolol binding to the frog erythrocyte derivatives with [3H]dihydroalprenolol binding to a 200-,u1 aliquot membrane (3-adrenergic receptor. Binding assays were performed with of a 1:500 dilution of an antiserum raised against partially purified 3.6 nM [3H]dihydroalprenolol. One hundred percent binding repre- frog erythrocyte f3-adrenergic receptors. [3H]Dihydroalprenolol was sented 0.42 pmol of [3H]dihydroalprenolol bound per assay. The at 6.5 nM in the assays. Binding (100%) represented 0.68 pmol of points are means of duplicate determinations from one experiment, [3H]dihydroalprenolol bound per assay. Symbols are as in Fig. 1. which is representative of two or three such experiments. A, Alpre- nolol; *, dihydroalprenolol-LA-dextran; 0, dihydroalprenolol- the dextran derivative and alprenolol hydrochloride, dihy- MA-dextran; 0, dihydroalprenolol-SA-dextran. droalprenolol-SA-dextran was prepared by the Posner reaction. To prepare the carrier with the medium arm (MA), the dextran RESULTS was converted to 2-aminoethyldextran by a procedure im- proved over the described one (4). The resulting 2-aminoeth- The f-adrenergic antagonist alprenolol (formula given in yldextran was condensed with Scheme I) was attached to macromolecules by the Posner re- N-acetylhomocysteinethiolacton action (9) (resulting in the formation of thioethers). In the to form macromolecular mercaptan from which dihydroal- Posner reaction a mercaptan is added to a double bond of the prenolol-MA-dextran was prepared again by the Posner reac- allyl group of the drug by a persulfate-initiated free-radical tion. To prepare the carrier having a long arm (LA) the dextran chain reaction (10, 11), forming a thioether as given in Scheme was -condensed with 1,4-butanediol diglycidyl ether. The 1. The water-soluble polysaccharide, dextran, was used as the conditions used favored the reaction of only one epoxy group starting material for the preparation of macromolecular mer- of the reagent with dextran. The remaining glycidyl group was captans. The mercapto group was attached to the polysaccha- converted as above via reaction with sodium thiosulfate and ride by three arms of different length. To prepare the carrier subsequent reduction into a 2-hydroxy-3-mercaptopropyl having a short arm (SA), the dextran was condensed with epi- group. From the substituted dextran thus obtained, dihy- chlorohydrin to yield glycidyldextran. The epoxy group of droalprenolol-LA-dextran was again prepared by the Posner glycidyldextran was converted by sodium thiosulfate into the reaction. All of the above compounds were purified by extensive corresponding thiosulfate ester (Bunte salt), which was then dialysis; freeze-drying gave colorless products with ultraviolet reduced with dithiothreitol to 2-hydroxy-3-mercaptopropyl- spectra consistent with the structures in Scheme I. dextran. A similar sequence of reactions was previously used [3H]Dihydroalprenolol has been shown to bind to f3-adren- to prepare 2-hydroxy-3-mercaptopropyl-Sephadex (12). From ergic receptors located on the plasma membranes of frog erythrocytes (5). The extent of the binding of the macromo- B.5 lecular alprenolol derivatives to these receptors was assessed by C 0 measuring their ability to compete with [3H]dihydroalprenolol 0 1001 for binding to these sites. The results in Fig. 1 show that binding of dihydroalprenolol-LA-dextran to membrane preparations as as XO 80 9 is about 1/10th strong that of alprenolol itself; dihy- 800- 'N ~ ~ ~ ~ 0 "I.0 s 60 IF Table 1. Equilibrium dissociation constants Apparent dissociation constants,* nM O 40 IF B-Adrenergic receptor Catecholamine preparations binding 20 Compounds Particulate Soluble antiserum Alprenolol 2.5 2.5 3.1 Dihydroalprenolol-LA- _ ~10910 10-7 104 -S 1074 dextran 28.6 25.0 19.0 Alprenolol residue, M Dihydroalprenolol-MA- FIG. 2. Competition of (1) aprenolol and* its macromolecular dextran 1,400 1,200 7.8 derivatives with [3H]dihydroalprenolol binding to solubilized (1% digitonin) frog erythrocyte membrane (3-adrenergic receptor. Con- Dihydroalprenolol-SA- ditions were as described in the legend to Fig. 1 with [3H]dihydroal- dextran 20,000 17,000 7.0 prenolol present at 10 nM in the assays. One hundred percent binding * Calculated as described (8) by using 2 nM as apparent dissociation represented 0.16 pmol of [3H]dihydroalprenolol bound per assay. constant of dihydroalprenolol for receptor preparations and anti- Symbols are as in Fig. 1. serum. Downloaded by guest on September 28, 2021 2222 Medical Sciences: Pitha et al. Proc. Natl. Acad. Sci. USA 77 (1980) droalprenolol-MA-dextran is about 1/600th as strong, whereas containing any ester bonds in the arm were found to be inferior dihydroalprenolol-SA-dextran is 1/8000th as strong. Similar to those described above. results were obtained with membranes solubilized by digitonin (Fig. 2). Thus, the large differences in binding observed for the DISCUSSION macromolecular forms of alprenolol cannot be due to steric The binding to 3-adrenergic receptor and pharmacological constraints imposed by components of the membrane other activity of compounds derived from I-alkylamino-3-aryloxy- than the receptor and must be due to interaction with the re- propan-2-ol (e.g., practolol, propranolol, and alprenolol) is well ceptor itself. understood. The results indicate that the substituent in the ortho Solubilized f-adrenergic receptor can be purified by affinity position of these drugs does not interfere when the drug is chromatography on alprenolol affixed to Sepharose (13, 14). bound to the receptor, because the compounds having an ex- A frog erythrocyte 3-adrenergic receptor preparation purified tended substituent in that position bind to the receptor with 1000- to 2000-fold over the membrane preparations was ob- high affinity (8, 13, 14, 17, 18). The same also applies for some tained by chromatography on an alprenolol affinity gel and substituents on the amino group of the above compounds (19, subsequent specific elution with isoproterenol. This preparation 20). The results with macromolecular alprenolol derivatives was used to raise antisera in rabbits. The antisera obtained were show that the branched polysaccharide dextran may create a found to bind catecholamines and related drugs with affinity prohibitive steric strain for binding to the receptor even when and specificity resembling that of the f-adrenergic receptor it is attached to the drug in the ortho position. The potency of (6). We have proposed that these antibodies might have been binding of the drug residue to the f3-adrenergic receptor in- raised in response to an isoproterenol-receptor complex. Be- creases with an increase in the length of the arm connecting the cause of the similarity of these binding sites with the receptor, bulky polysaccharide with the drug, as is clearly seen in Fig. it was of interest to assess the interaction of the alprenolol de- 4. rivatives with these sites. The binding of the alprenolol deriv- Consistent with this conclusion was the atives to the antiserum was evaluated by their ability to compete moderately strong with [3H]dihydroalprenolol binding. Results are summarized binding to the receptor of the compound formed by the Posner in Fig. 3. In contrast to the results obtained with the receptor reaction from glutathione and alprenolol and the lack of preparations, there were only small differences among the binding to the receptor of the copolymer of alprenolol and derivatives. Dihydroalprenolol-SA-dextran and dihydroal- acrylamide (unpublished data). It is important to note that there prenolol-MA-dextran are about one half as potent as alprenolol is a structural similarity between the linkage arms of dihy- itself, and dihydroalprenolol-LA-dextran, about one-sixth as droalprenolol-SA-dextran and dihydroalprenolol-LA-dextran potent. It may be noted that free alprenolol and dihydroal- on one hand and between dihydroalprenolol-MA-dextran and prenolol have identical affinity at the receptor and, further- dihydroalprenololglutathione on the other. This indicates that, more, alprenolol itself has a comparable affinity for the re- although some specific interactions between the arms and re- ceptors and the antibody. The equilibrium dissociation constants ceptor may exist, these are not the predominant ones. (Kds) of alprenolol and the dextran-alprenolol derivatives for The binding site for alprenolol derivatives on the receptor receptor and antibody binding sites are summarized in Table may be considered to be buried or otherwise in a sterically constrained position. In contrast, the drug binding site of the The above differences in binding of macromolecular forms antibody must be in a very accessible position because the of alprenolol to the receptor and antibody convincingly show binding affinity of the drug was not decreased even when the that the method used for affixing the drug to polysaccharide polysaccharide was attached by a short arm to the drug. leads to a very stable linkage-i.e., there is no measurable These findings are in agreement with other observations. The leakage of small molecular weight drug derivatives from the successful application of affinity chromatography depends on macromolecular drug. This chemical stability was achieved by strong and specific binding of the immobilized ligand with the choice of preparative methods rather than by subsequent pu- protein to be purified. In the purification of many receptors the rifications. Preparative methods, using as starting materials necessity of an extended link between the ligand and matrix either partly oxidized dextran (15) or dextran activated by has been noted (21, 22), suggesting that receptors may have cyanogen bromide (16), had to be abandoned; also materials rather constrained binding sites. The architecture of various antibodies is known and their binding sites are generally in 1o-91 exposed positions (23) with the result that the steric constraints Alprenolol in these molecules are rather low and can be detected only with ...... _=_....= 1 0-8 bulky antigens (24). Differences in steric requirements for binding to antibodies and to receptors may have potential applications to the man- .; 10-7h 0 Dihydroalprenolol-LA- agement of clinical situations in which an antihormone antibody dextran interferes with hormone action. An example would be anti- .0 insulin antibodies causing insulin resistance in insulin-treated 0 io'F- anC. diabetics (25). In such a case, a suitably substituted hormone Dihydroalprenolol-MA-dextran or drug might be used to selectively and specifically neutralize m0 1 0-h circulating antibodies without affecting the physiologically LO important receptor binding sites. were 'o-4h ' Dihydroalprenolol-SA-dextran Macromolecular adrenergic agonists synthesized pre- viously and were shown to have biological potency (26, 27). The presently described compounds are the first macromolecules

0 5 10 15 20 containing antagonist moieties and they have the advantage Atoms separating alprenolol from dextran moiety of being stable and of high affinity for the f-adrenergic re- FIG. 4. Dependence of binding of alprenolol derivatives to ceptors. A number of potential applications for such macro- membranes of frog erythrocytes on number of atoms in the arm sep- molecules may be envisaged. Adrenergic antagonists are widely arating the drug from polysaccharide. used to treat hypertension; unfortunately, they have concom- Downloaded by guest on September 28, 2021 Medical Sciences: Pitha et al. Proc. Natl. Acad. Sci. USA 77 (1980) 2223

itant side effects on the central nervous system, especially in 11. Griesbaum, K. (1970) Angew. Chem. Int. Ed. Engl. 9, 273- the elderly (28). The present compounds, while retaining their 287. binding to the adrenergic receptor, may be expected, due to 12. Axen, R., Drevin, H. & Carlsson, J. (1975) Acta Chem. Scand. their lipophobic character, to be unable to cross the blood-brain Ser. B: 29,471-474. barrier (29) and thus may have diminished side effects. The 13. Caron, M. G., Srinivasan, Y., Pitha, J., Kociolek, K. & Lefkowitz, inability of the presently described compounds to penetrate into R. J. (1979) J. Biol. Chem. 254,2923-2927. cells also makes them useful as 14. Vauguelin, G., Geynet, P., Hanoune, J. & Strosberg, A. D. (1977) potentially selective markers for Proc. Nat!. Acad. Sci. USA 74,3710-3714. adrenergic receptors. Finally, the macromolecular character 15. Pitha, J. (1978) Eur. J. Biochem. 82, 285-292. and the presence of mercapto groups in these compounds makes 16. Kagedal, L. & Akerstrom, S. (1971) Acta Chem. Scand. 25, it possible to tag them with large amounts of fluorescent, ra- 1855-1859. dioactive, or electron-dense markers. 17. Rzeszotarski, W. J., Gibson, R. E., Eckelman, W. C. & Reba, R. C. (1979) J. Med. Chem. 22,735-737. M.G.C. and R.J.L. thank Dr. Ablad of Haessle Pharmaceutical for 18. Bartsch, W. W., Dietmann, K., Leinert, H. & Sponer, G. (1977) his generous gift of (±) alprenolol hydrochloride. J.P. appreciates the Arzneim. Forsch. 27(I), 1022-1026. comments of Drs. E. Dax, C. Heckman, and W. Rzeszotarski on the 19. Atlas, D., Yaffe, D. & Skutelsky, E. (1978) FEBS Lett. 95, data, and we thank Mrs. D. Lamartin for help with the manuscript. 173-176. 20. Meier, K. E. & Ruoho, A. E. (1977) J. Solid Phase Biochem. 2, 1. Ringsdorf, H. (1975) J. Polym. Sci. Polym. Symp. 51, 135- 105-109. 153. 21. Cuatrecasas, P. & Anfinsen, C. B. (1971) Annu. Rev. Biochem. 2. Muck, K. F., Christ, 0. & Kellner, H. M. (1977) Makromol. 41,259-278. Chem. 178, 2785-2797. 22. Jakoby, W. B. & Wilchek, M., eds. (1974) Methods Enzymol. 3. Noronha-Blob, L., Vengris, V. E., Pitha, P. M. & Pitha, J. (1977) 34. J. Med. Chem. 20,356-59. 23. Nisonoff, A., Hopper, J. E. & Spring, S. B. (1975) The Antibody 4. Chu, B. C. F. & Whiteley, J. M. (1977) Mol. Pharmacol. 13, Molecule (Academic, New York). 80-88. 24. Streetkerk, D. G., Marijula, B. N. & Glaudemans, C. P. J. (1979) 5. Lefkowitz, R. J., Limbird, L. E., Mukherjee, C. & Caron M. G. J. Immunol. 122,537-541. (1976) Biochim. Biophys. Acta 457,1-39. 25. Bar, R. S., Harrison, L. C., Muggeo, M., Gorden, P., Kahn, C. R. 6. Caron, M. G., Srinivasan, Y., Snyderman, R. & Lefkowitz, R. J. & Roth, J. (1979) Adv. Intern. Med. 24,23-52. (1979) Proc. Natl. Acad. Sci. USA 76,2263-2267. 26. Verlander, M. S., Venter, J. C., Goodman, M., Kaplan, N. 0. & 7. Caron, M. G. & Lefkowitz, R. J. (1976) J. Biol. Chem. 251, Saks, B. (1976) Proc. Natl. Acad. Sci. USA 73, 1009-1013. 2374-2384. 27. Melmon, K. L., Weinstein, Y., Bourne, H. R., Poon, T., Shearer, 8. Mukherjee, C., Caron, M. G., Mullikin, D. & Lefkowitz, R. J. G. & Castagnoli, N. (1976) Mol. Pharmacol. 12,701-710. (1976) Mol. Pharmacol. 12, 16-31. 28. Hammond, J. J. & Kirkendall, W. M. (1979) Geriatrics 34, 9. Posner, T. (1905) Ber. Dtsch. Chem. Ges. 38, 646. 27-36. 10. Kharash, M. S., Read, A. T. & Mayo, F. R. (1938) Chem. Ind. 57, 29. Ohno, K., Pettigrew, K. D. & Rapoport, S. (1978) Am. J. Physiol. 752. 235(3), H299-H307. Downloaded by guest on September 28, 2021