Search

Little is known about the pharmacology or cell biology of human bombesin (Bn) receptors, because they are usually present at low levels and both subtypes are frequently present in the same tissues. Human gastrin-releasing peptide (GRP) receptors (huGRP-R) and human neuromedin B (NMB) receptors (huNMB-R) were stably transfected into BALB/3T3 fibroblasts. Both receptor types were glycosylated, with 35% of the huGRP-R and 38% of the huNMB-R representing carbohydrate residues. The extent of glycosylation of the transfected huGRP-R was the same as that seen in the human glioblastoma cell line U-118. Radiolabeled agonist ligands were rapidly internalized, whereas noninternalized ligand readily dissociated in a temperature-dependent fashion. The affinities of various agonists for binding to the huGRP-R were Bn (Ki = 1.4 +/- 0.2 nM) = 4 x GRP = 300 x NMB. In contrast, affinities for the huNMB-R were NMB (Ki = 8.1 +/- 5.2 nM) = 4 x Bn = 600 x GRP. [F5-D-Phe6,D-Ala11]Bn(6-13)methyl ester was the most potent huGRP-R antagonist, whereas D-Nal-Cys-Tyr-D-Trp-Lys-Val-Cys-Nal-NH2 was the most potent huNMB-R antagonist. Agonist binding to either receptor type caused activation of phospholipase C and increased cellular [3H]inositol phosphate levels. GRP was potent at increasing [3H]inositol phosphate generation in cells expressing the huGRP-R (EC50 = 13.6 +/- 1.3 nM), whereas NMB was similarly potent when acting upon cells expressing the huNMB-R (EC50 = 9.3 +/- 1.4 nM). However, neither receptor type, when stimulated with agonist, caused an increase in cAMP levels. These data show that stably transfected huGRP-R exhibit similar pharmacology for agonists and antagonists, are appropriately glycosylated, and function similarly with respect to their ability to alter biological activity, compared with natively expressed receptors. Minimal native huNMB-R data are available for comparison, but in general the huNMB-R is similar to the rat NMB receptor in its pharmacology and cell biology.

From structure-function studies it has been proposed that two subtypes of receptors may mediate galanin's actions in the gastrointestinal tract and other tissues and their effects can be either direct or neurally mediated. We have recently demonstrated that isolated gastric smooth muscle cells possess high-affinity galanin receptors, activation of which increases AMP and causes relaxation. Because this cell system contains no neural elements, contains a single class of galanin receptors and allows binding to be correlated with function, it is a good system to investigate peptide requirements for cell activation. Porcine galanin (p-Gal) and rat galanin (r-Gal) were equipotent to the NH2-terminal fragments r,p-Gal(1-10), r,p-Gal(1-15), r,p-Gal(1-20), p-Gal(2-29) and p-Gal(3-29) at inhibiting binding of 125I-galanin to gastric smooth muscle cells from guinea pig (Kd 5-8 nM). The midmolecule fragment p-Gal(9-25) and the long COOH-terminal fragment r-Gal(9-29) were equipotent and 25-fold less potent than p-Gal. Acetylation of r-Gal(9-29) increased potency to that of p-Gal. The short COOH-terminal fragment, p-Gal(21-29) and r-Gal(21-29), had very low affinity (Kd 3 microM). Each peptide alone (1 microM) caused no effect on cell length, but inhibited carbachol-induced contraction. The inhibition was 81% to 92% for r-Gal or p-Gal, r,p-Gal(1-10), r,p-Gal(1-15), r,p-Gal(1-20), p-Gal(2-29), p-Gal(3-29) and Ac-r-Gal(9-29); 47% for p-Gal(9-25) and r-Gal(9-29); and 16% to 17% for p-Gal(21-29).(ABSTRACT TRUNCATED AT 250 WORDS)

Stimulation of the gastrin-releasing peptide receptor (GRP-R) in Swiss 3T3 cells resembles that of a number of other recently described G protein-coupled receptors, insofar as both the phospholipase C and adenylyl cyclase signal transduction pathways are activated. GRP-R activation induces numerous alterations in both the cell and the receptor, but because two signal transduction pathways are activated it is difficult to determine the specific contributions of either pathway. We have found that BALB/3T3 fibroblasts transfected with the coding sequence for the GRP-R are pharmacologically indistinguishable from native receptor-expressing cells and activate phospholipase C in a manner similar to that of the native receptor but fail to increase cAMP in response to bombesin; thus, they may be useful cells to explore the role of activation of each pathway in altering cell and receptor function. Swiss 3T3 cells and GRP-R-transfected BALB/3T3 cells expressed identically glycosylated receptors that bound various agonists and antagonists similarly. G protein activation, as determined by evaluation of agonist-induced activation of phospholipase C and by analysis of the effect of guanosine-5'-(beta,gamma-imido)triphosphate on GRP-R binding affinity, was indistinguishable. Agonist stimulation of GRP-R caused similar receptor changes (internalization and down-regulation) and homologous desensitization in both cell types. Bombesin stimulation of Swiss 3T3 cells that had been preincubated with forskolin increased cAMP levels 9-fold, but no bombesin-specific increase in cAMP levels was detected in transfected cells, even though forskolin and cholera toxin increased cAMP levels in these cells. Quiescent Swiss 3T3 cells treated with bombesin rapidly increased c-fos mRNA levels and [3H]thymidine incorporation, whereas both effects were potentiated by forskolin. The specific protein kinase A inhibitor H-89 blocked increases in c-fos levels and [3H]thymidine incorporation induced by low concentrations of bombesin. GRP-R-transfected BALB/3T3 cells increased c-fos mRNA levels and [3H]thymidine incorporation with the addition of serum but not bombesin. These data suggest that bombesin-stimulated increases in cellular levels of cAMP appear not to be an important mediator of GRP-R internalization, down-regulation, or desensitization but do play an important role in bombesin-induced mitogenesis.

Inhibition both in vivo and in vitro of pepsinogen secretion by somatostatin (SS) and the histological demonstration that fundic D-cells contain long cytoplasmic processes extending to chief cells suggest a possible direct effect of SS on chief cell function. The aim of the present study was to determine whether SS interacts directly with receptors on isolated gastric chief cells and, if so, how SS alters cell function. Binding of 125I-[Tyr11]SS14 to chief cells was saturable, time and temperature dependent, and was inhibited by both SS14 (Ki 1.6 nM) and SS28 (Ki 5.2 nM). SMS-201-995 was 1,300-fold less potent than SS14. Calcium-mobilizing secretagogues reduced binding of 125I-[Tyr11]SS14 with efficacies of cholecystokinin octapeptide (CCK-8) > carbachol > gastrin. Adenosine 3',5'-cyclic monophosphate (cAMP)-activating secretagogues also inhibited binding with efficacies of secretin > vasoactive intestinal polypeptide (VIP). 12-O-tetradecanoylphorbol 13-acetate (TPA) or A-23187 also decreased binding. Analyses demonstrated that CCK-8 and TPA were decreasing the affinity of SS receptors for 125I-[Tyr11]SS14 without affecting their binding capacity. Both SS14 and SS28 at a maximally effective concentration inhibited cAMP production caused by VIP or secretin (20-30%) but did not alter cytosolic calcium ([Ca2+]i), inositol phosphates, or pepsinogen release. We conclude that chief cells possess SS receptors with a high affinity for both SS14 and SS28 but low affinity for SMS-201-995 and thus resemble the SSB receptors described in the rat cerebral cortex. Although occupation of these receptors by SS has no effect on pepsinogen release induced by secretagogues acting through either the calcium or the cAMP pathway, SS receptor occupation is regulated by agents activating phospholipase C, adenylate cyclase, protein kinase C, and [Ca2]i.

Galanin-like immunoactivity occurs in nerves and plexi in muscle layers throughout gastrointestinal tract including the stomach. Galanin can affect gastric emptying and contraction or relaxation of gastric muscle in different species. The aim of this study was to investigate the direct effect of galanin on dispersed gastric smooth muscle cells and to characterize any galanin receptors that mediated any effect. Dispersed gastric smooth muscle cells were prepared from guinea pig stomach by collagenase digestion. Porcine galanin (p-galanin; 1 microM) did not stimulate contraction when present alone; however, p-galanin (1 microM) inhibited carbachol-induced contraction with a half-maximal effect at 7 nM. p-Galanin (1 microM) increased cellular adenosine 3',5'-cyclic monophosphate (cAMP) content by 10 s and caused a maximal increase of 80% over basal. 125I-galanin (porcine) bound to dispersed cells in a time- and temperature-dependent manner. Binding was saturable, reversible, and specific. Binding of 125I-galanin was inhibited almost equally by porcine and rat galanin (Ki = 6-8 nM) but was not inhibited by the galanin-associated peptide [preprogalanin-(108-123)]. The fragment galanin-(1-16) was equally potent to rat galanin; however, the fragment galanin-(9-29) was 56-fold less potent (Ki = 370 nM). Computer analysis demonstrated there were two binding sites for p-galanin on gastric smooth muscle cells, a high-affinity site (Kd = 2.6 nM) with low capacity (Bmax = 175 fmol/mg protein) and a low-affinity site (Kd = 150 nM) with large capacity (Bmax = 3,611 fmol/mg protein).(ABSTRACT TRUNCATED AT 250 WORDS)

We used thapsigargin (TG), 2,5-di-tert-butyl-1,4-benzohydroquinone (BHQ) and cyclopiazonic acid (CPA), each of which inhibits microsomal Ca(2+)-ATPase, to evaluate the effects of this inhibition on cytoplasmic free calcium ([Ca2+]i) and secretagogue-stimulated enzyme secretion in rat pancreatic acini. Using single-cell microspectrofluorimetry of fura-2-loaded acini we found that all three agents caused a sustained increase in [Ca2+]i by mobilizing calcium from inositol-(1,4,5)-trisphosphate-sensitive intracellular calcium stores and by promoting influx of extracellular calcium. Concentrations of all three agents that increased [Ca2+]i potentiated the stimulation of enzyme secretion caused by secretagogues that activate adenylate cyclase but inhibited the stimulation of enzyme secretion caused by secretagogues that activate phospholipase C. With BHQ, potentiation of adenylate cyclase-mediated enzyme secretion occurred immediately whereas inhibition of phospholipase C-mediated enzyme secretion occurred only after several min of incubation. In addition, the effects of BHQ and CPA on both [Ca2+]i and secretagogue-stimulated enzyme secretion were reversed completely by washing whereas the actions of TG could not be reversed by washing. Concentrations of BHQ in excess of those that caused maximal changes in [Ca2+]i inhibited all modes of stimulated enzyme secretion by a mechanism that was apparently unrelated to changes in [Ca2+]i. Finally, in contrast to the findings with TG and BHQ, CPA inhibited bombesin-stimulated enzyme secretion over a range of concentrations that was at least 10-fold lower than the range of concentrations over which CPA potentiated VIP-stimulated enzyme secretion.

Galanin has numerous effects on gastrointestinal smooth muscle. However, because of the lack of specific inhibitors, it is not known which are physiological and which are pharmacological. This study investigates the ability of two chimeric galanin analogs, [# 1-galantide = (M-15) = [galanin (1-13)-substance P(5-11)] and #2-M-35[galanin(1-13)bradykinin (2-9)], which were recently reported to function as galanin-receptor antagonists in the CNS, to interact with galanin receptors on rat jejunal muscle strips or dispersed smooth muscle cells from guinea pig stomach. In both systems each chimeric analog had agonist activity and was as efficacious as galanin. Cross-desensitization experiments demonstrated that in the jejunal muscle strips, both chimeric analogs were causing muscle contraction by interacting with the galanin receptor. In dispersed smooth muscle cells, galanin, as well as each chimeric analog, caused muscle relaxation, whereas substance P and bradykinin both caused muscle contraction. Each chimeric analog was equipotent to galanin in inhibiting binding of 125I-galanin, and there was close agreement between their abilities to occupy the galanin receptor and cause relaxation. Each chimeric analog also activated adenylate cyclase and increased cAMP characteristic of relaxants. These studies demonstrate these chimeric analogs will not be useful for defining the physiological role of galanin in altering gastrointestinal motility, because they function as full galanin-receptor agonists instead of as galanin-receptor antagonists.

Pituitary adenylate cyclase-activating polypeptide (PACAP) is a 38-amino acid peptide of the secretin-vasoactive intestinal peptide (VIP) family. To investigate whether PACAP alters chief cell function, we prepared isolated chief cells (> 90% pure) from guinea pig stomach. PACAP-38, PACAP-27, VIP, and secretin all caused a threefold increase in pepsinogen release. The dose-response curves of PACAP-38, PACAP-27, and VIP were biphasic, whereas with secretin it was not. The first phase comprised 40% of maximal release, and each of the three peptides (PACAP-38, PACAP-27, and VIP) were equipotent (EC50 0.1-0.3 nM). For the second phase, comprising 60% of maximal release, the relative potencies were PACAP-38 > PACAP-27 = VIP. 125I-labeled secretin, 125I-VIP, and 125I-PACAP-27 all demonstrated saturable binding to chief cells. Binding of both 125I-PACAP-27 and 125I-VIP was inhibited completely and with similar potencies by PACAP-38, PACAP-27, and VIP. Secretin had a > 500-fold lower affinity than PACAP-38 for displacing both 125I-PACAP-27 and 125I-VIP. With 125I-secretin, secretin was the most potent, and was 197 times more potent than PACAP-38, which was 6-8 times more potent than both PACAP-27 and VIP. We conclude that both PACAP-38 and PACAP-27 stimulate pepsinogen secretion from dispersed chief cells. In contrast to a number of other tissues, no evidence for a high-affinity receptor that interacted only with PACAP was found. PACAP and VIP interact with equal high affinity with a common receptor and with low affinity with the secretin receptor.

On the basis of opioid-stimulated contraction of dispersed gastric smooth muscle cells it has been suggested that these cells possess opioid receptors of three subtypes: kappa (kappa), mu (mu), and delta (delta). We have used selective peptidase-resistant radioligands, agonists and antagonists, to examine receptor subtypes on dispersed gastric smooth muscle cells from guinea pigs prepared by collagenase digestion. The kappa-agonist U-50488H, the mu-agonist [D-Ala2,N-Me-Phe4,Gly5-ol]enkephalin (DAGO), and the delta-agonist [D-Pen2,Pen5]enkephalin (DPDPE) each caused muscle contraction. The concentrations required to caused half-maximal contraction were U50488H (6 pM) greater than DAGO (13 pM) greater than DPDPE (6 nM). The abilities of these agonists to inhibit binding of [3H]U-69593 (kappa-preferring) by 50% were U50488H (43 nM) greater than DAGO (43 microM) greater than DPDPE (200 microM). Their abilities to inhibit binding of [3H]naloxone (mu-preferring) by 50% were DAGO (0.2 microM) greater than U50488H (10 microM) greater than DPDPE (greater than 100 microM). No binding could be detected with the delta-selective ligand [3H]DPDPE. The kappa-preferring antagonist Mr2266 (10 nM) preferentially inhibited contraction stimulated by the kappa-agonist U50488H, and naltrexone (10 nM) (mu-selective antagonist) preferentially inhibited contraction stimulated by the mu-agonist DAGO. ICI 174864 (200 microM; delta-selective antagonist) had no effect on contraction stimulated by mu-, kappa-, or delta-agonists. Contraction stimulated by the delta-agonist DPDPE was inhibited by both kappa- and mu-receptor antagonists. Studies on the effect of the antagonists on binding of [3H]naloxone and [3H]U69593 also provided evidence for kappa- and mu-sites but nor for delta-sites.(ABSTRACT TRUNCATED AT 250 WORDS)

The effects of somatostatin-28, somatostatin-14, and a synthetic somatostatin octapeptide analogue, D-Phe-Cys-Tyr-D-Trp-Lys-Thr-Cys-Nal-NH2 (cyclo SS-8) were examined on contraction of dispersed gastric smooth muscle cells from guinea pigs. The somatostatins did not cause contraction of gastric smooth muscle cells, nor did they inhibit carbachol-stimulated contraction. However, they reversed vasoactive intestinal peptide (VIP)-induced inhibition (relaxation) of carbachol-stimulated contraction. Somatostatin-28 had a half-maximal effect (EC50) at 1.6 +/- 0.8 nM, cyclo SS-8 at 0.6 +/- 0.3 nM, but somatostatin-14 had no effect even when used in concentrations as high as 1 microM. Incubation of muscle cells with peptidase inhibitors phosphoramidon (1 microM) plus amastatin (10 microM) had no effect on the EC50 of somatostatin-28 or cyclo SS-8 but increased the potency of somatostatin-14 greater than 1,000-fold. When peptides were incubated with muscle cells and the products applied to high-performance liquid chromatography, cyclo SS-8 was not degraded, but somatostatin-14 was rapidly degraded when present alone, and the addition of peptidase inhibitors partially inhibited the degradation. Cyclo SS-8 had its maximal effect at 0.5-1 min and inhibited relaxation induced by VIP, isoproterenol, glucagon, or dibutyryl adenosine 3',5'-cyclic monophosphate (DBcAMP). Cyclo SS-8 partially inhibited the increase in VIP-stimulated cAMP. Preincubation with pertussis toxin blocked the inhibitory action of cyclo SS-8 on VIP or DBcAMP-induced relaxation. These results indicate that gastric smooth muscle cells rapidly degrade somatostatin-14 and suggest that muscle cell peptidases could have a major effect on the actions of somatostatin-14.(ABSTRACT TRUNCATED AT 250 WORDS)

We have previously described that [Tyr0]CGRP(28-37) acts as a receptor antagonist of rat CGRP in guinea pig pancreatic acini. We therefore examined other C-terminal peptides of CGRP for such activity. CGRP-acetyl(28-37) acetate did act as a rat CGRP antagonist. However, C-terminal CGRP peptides of 4 to 8 amino acid residues did not antagonize the actions of rat CGRP but stimulated amylase secretion. In pancreatic acini, a maximally effective concentration of rat CGRP (100 nM) caused a 2.1-fold increase in amylase secretion. When the C-terminal peptides of CGRP were tested in at 100 microM, CGRP(34-37) caused a 1.8-fold increase in amylase secretion, CGRP(33-37) a 2.8-fold increase. CGRP(32-37) a 9.2-fold increase, CGRP(31-37) a 4.1-fold increase, and CGRP(30-37) a 5.1-fold increase. Further studies with the most effective peptide, CGRP(32-37), demonstrated that it did not cause release of lactate dehydrogenase, and thus did not cause amylase release by cell damage. Unlike rat CGRP, CGRP(32-37) did not increase cellular cyclic AMP, but did stimulate outflux of 45Ca. CGRP(32-37)-stimulated amylase release was not inhibited by the substance P receptor antagonist, spantide, by the bombesin receptor antagonist, [D-Phe6]bombesin(6-13) propylamide, or by the muscarinic receptor antagonist, atropine, but was inhibited by the CCK receptor antagonist L364,718. C-terminal peptides of CGRP inhibited binding of 125I-BH-CCK-8, with the relative potencies of the peptides being the same as their relative potencies for stimulating amylase secretion.(ABSTRACT TRUNCATED AT 250 WORDS)

In guinea pig pancreatic acini rat calcitonin gene-related peptide (CGRP) increased amylase release 2-fold, salmon calcitonin had an efficacy of only 44% of that of CGRP and [Tyr0]CGRP(28-37) and human calcitonin had no actions. [Tyr0]CGRP(28-37), but not human calcitonin, antagonized the actions of CGRP in pancreatic acini with an IC50 of 3 microM. [Tyr0]CGRP(28-37) produced a parallel rightward shift in the dose-response curve for CGRP-stimulated amylase secretion. The inhibition was specific for CGRP and was reversible. Studies with 125I-CGRP demonstrated that CGRP, salmon calcitonin and [Tyr0]CGRP, but not human calcitonin, interacted with CGRP receptors on pancreatic acini. These results indicate that various CGRP-related peptides demonstrate different relationships between their abilities to occupy the CGRP receptor and to affect biologic activity, with CGRP itself being a full agonist, salmon calcitonin a partial agonist, [Tyr0]CGRP(28-37) a competitive antagonist, and human calcitonin having no actions.

Trimethylpsoralen (TMP) levels in the blood of vitiligo patients were determined through the use of a high-performance liquid chromatographic method. TMP was extracted from blood buffered at pH 9.0 with 95:5 (V/V) hexane-isopropanol mixture; evaporated to dryness, and reconstituted in 50 microliters of ethanol. A 10-microliters aliquot was injected into a Micropack MCH-10 column (Varian HPLC model #5000). The mobile phase consisted of a mixture of water and acetonitrile with a linear gradient. The retention time of TMP was 16.5 min. The calibration curve of the external standards was linear. Overall recovery of the internal stands was 75-92%, with the lower detection limit of TMP at 2 ng/ml. Peak blood levels as low as 140 ng/ml and as high as 800 ng/ml were obtained in vitiligo patients 1-2 hr following the oral administration of 30 mg of trioxsalen tablets (Paul B. Elder Co.). Blood TMP levels peaked consistently at 2 hr when patients were fasted for 8 hr prior to drug ingestion. These results are consistent with the clinical observation that maximum response due to phototherapy is obtained 1-2 hr after oral administration of the drug. Two hours after oral drug administration, TMP levels in the epidermis, dermis, and whole skin of the guinea pig (in ng per g tissue) were: epidermis, 226 +/- 15; dermis, 25 +/- 6; and whole skin 176 +/- 12. Also detected were TMP levels 244 +/- 17 ng/ml in aqueous humor and 63 +/- 6 mg/ml in vitreous humor. These results point to the fact that the eyes of patients must be protected from overexposure to sunlight after psoralen-ultraviolet treatment.

When adenylate cyclase activities in purified membranes from normal rat liver and from a series of rapid growing transplantable Morris hepatomas were examined at various temperatures, several unique features were observed. Two of the hepatomas yielded patterns similar to that of normal liver, even though glucagon did not activate either tumor adenylate cyclase but did activate the normal liver enzyme. The patterns of the third tumor line were completely different from normal. This clearly shows the heterogeneity in cancers of similar origin.

Adenylate kinases, purified from adult rat liver and skeletal muscle and from a fast growing hepatoma, were examined in the presence of a series of prostaglandins. Prostaglandins A1, A2, E1, E2 and F2alpha stimulated liver adenylate kinase 27% to 149%. All of these protaglandins stimulated the skeletal muscle enzyme from 47% to 82%. While prostaglandins A2 and E2 stimulated, prostaglandins A1 and F2alpha inhibited, and prostaglandin E1 was without effect on the hepatoma adenylate kinase.

The effects of prostaglandins on adenylate cyclase activity have been examined in membranes purified from normal rat liver and from a series of Morris hepatomas. Prostaglandin E1 gave the greatest stimulation (up to two-fold) in all membranes. However, prostaglandins A1, A2, and F2alpha, although stimulatory in liver and four tumor membranes, were inhibitory of adenylate cyclase activity in membranes from two of the fast-growing tumors. Arrhenius plots yielded broken line curves (at 20 degrees C) for the basal activity of all enzymes. Addition of various prostaglandins caused shifts in the broken line curves and/or produced nonbroken (straight) line curves for the liver and many of the hepatoma adenylate cyclases.

Adenylate cyclase (EC 4.6.1.1) activities were examined in membrane preparations from two rat liver cell lines (62 and 3C4) which were grown in monolayer cultures. The cells were epithelial-like in growth character. Adenylate cyclase from the line 62 was stimulated by epinephrine, Gpp(NH)p, and prostaglandins A1,A2,E1,E2, and F2alpha, but not by glucagon. Arrhenius plots of adenylate cylase activity from line 62 gave straight lines, except when epinephrine was present in the assay; epinephrine-stimulated activity gave a distinct break at 20 degrees C. Adenylate cyclase activity in line 3C4 was stimulated by glucagon ten times greater than by epinephrine. It was responsive to Gpp(NH)p and all the prostaglandins. Arrhenius plots of adenylate cyclase activity of line 3C4 always gave straight line curves. Prostaglandins flattened the straight line curves (allowed temperature independence) of adenylate cyclase activity in membranes from both cell lines.

8-Methoxypsoralen (8-MOP) levels in the blood of vitiligo patients were determined through the use of a reverse-phase high-performance liquid chromatographic method. The overall recovery of the internal standards was 85-94%, with the lower detection limit of 8-MOP at 2 ng. Peak blood levels as low as 130 ng/ml and as high as 3892 ng/ml were obtained in patients at 1-3 h following the oral administration of 0.6 mg/kg body weight of Oxsoralen capsules (Elder Pharmaceuticals Co.). These results are consistent with the clinical observation that maximum response in phototherapy is obtained at about 2 h after oral administration of the drug. Two hours after oral administration of 0.6 mg/kg of Oxsoralen, 8-MOP levels in the epidermis, dermis, and whole skin of the guinea pig (in ng/g) were: epidermis, 330 +/- 20; dermis, 89 +/- 16; whole skin, 379 +/- 19. Also detected were 8-MOP levels of 441 +/- 22 ng/ml in aqueous humor, 166 +/- 18 ng/ml in vitreous gel, 355 +/- 15 ng/g in lens, and 410 +/- 26 ng/g in retina. These results point to the fact that the eyes of the patient must be protected from exposure to sunlight after psoralen UV treatment, and that 8-MOP is absorbed in blood unevenly and varies from patient to patient. The fact that only 50-60% of the patients responded to psoralen photochemotherapy for vitiligo may be related to the variation of absorption of the drug in individual patients.

The binding of adenine nucleotides to liver adenylate kinase was dependent on Mg2+ ions. Citric acid enhanced the binding of all metal-chelated radioactive nucleotides and indicated two observable binding sites for Mg3H-ADP and Mg3-ATP and one-half binding site for Mg3H-AMP. Two binding sites of Mg3H-ADP and one binding site for Mg3H-ATP were also observed in the absence of citric acid. Stoichiometric binding of 14C-citric acid to liver adenylate kinase varied with additions of different nucleotides. AMP prevented whereas ADP and ATP enhanced the binding of 14C-citric acid.

The major enzymatic forms of adenylate kinase have been purified to homogeneity from fetal liver and adult brain of the rat. The two enzymes differ with respect to isoelectric points, Km (ATP), Km (AMP), and Ka (citrate). Antibody to adult liver adenylate kinase does not inhibit either enzyme, while entibody to adult skeletal muscle enzyme inhibits the brain enzyme but not the fetal liver enzyme. It is therefore probable that there are three major forms of adenylate kinases in fetal and adult rat tissues.