Lentivectors encoding immunosuppressive proteins genetically engineer pancreatic beta-cells to correct diabetes in allogeneic mice.

The effectiveness of genetic engineering with lentivectors to protect transplanted cells from allogeneic rejection was examined using, as a model, type 1 diabetes treatment with beta-cell transplantation, whose widespread use has been limited by the requirement for sustained immunosuppressive treatment to prevent graft rejection. We examined whether lentivectors expressing select immunosuppressive proteins encoded by the adenoviral genome early region 3 (AdE3) would protect transplanted beta-cells from an alloimmune attack. The insulin-producing beta-cell line beta TC-tet (C3HeB/FeJ-derived) was transduced with lentiviruses encoding the AdE3 proteins gp19K and RID alpha/beta. The efficiency of lentiviral transduction of beta TC-tet cells exceeded 85%. Lentivector expression of gp19K decreased surface class I major histocompatibility complex expression by over 90%, whereas RID alpha/beta expression inhibited cytokine-induced Fas upregulation by over 75%. beta TC-tet cells transduced with gp19K and RID alpha/beta lentivectors, but not with a control lentivector, provided prolonged correction of hyperglycemia after transplantation into diabetic BALB/c severe combined immunodeficient mice reconstituted with allogeneic immune effector cells or into diabetic allogeneic BALB/c mice. Thus, genetic engineering of beta-cells using gp19K- and RID alpha/beta-expressing lentiviral vectors may provide an alternative that has the potential to eliminate or reduce treatment with the potent immunosuppressive agents necessary at present for prolonged engraftment with transplanted islets.

A new approach for point mutation detection based on a ligase chain reaction.

A new method for the identification of point mutations is proposed. The method is based on ligase chain reaction (LCR) and it includes a procedure for correction of ligation by Cleavase. Reaction products are detected by a colorimetric method after adsorption of the resulting DNA duplexes to the solid phase. One strand of LCR products carries biotin to be bound on a streptavidin-coated microwell. Another strand contains a single-stranded region that is to be coupled with an oligonucleotide carrying a substrate for colorimetric detection. The suggested method has two advantages: (i) use of Cleavase increases the accuracy of ligation and (ii) a template independent ligation does not occur in LCR due to a special design of primers.

Discordant effects of glucosamine on insulin-stimulated glucose metabolism and phosphatidylinositol 3-kinase activity.

The impact of increased GlcN availability on insulin-stimulated p85/p110 phosphatidylinositol 3-kinase (PI3K) activity in skeletal muscle was examined in relation to GlcN-induced defects in peripheral insulin action. Primed continuous GlcN infusion (750 micromol/kg bolus; 30 micromol/kg.min) in conscious rats limited both maximal stimulation of muscle PI3K by acute insulin (I) (1 unit/kg) bolus (I + GlcN = 1.9-fold versus saline = 3.3-fold above fasting levels; p < 0.01) and chronic activation of PI3K following 3-h euglycemic, hyperinsulinemic (18 milliunits/kg.min) clamp studies (I + GlcN = 1.2-fold versus saline = 2.6-fold stimulation; p < 0.01). To determine the time course of GlcN-induced defects in insulin-stimulated PI3K activity and peripheral insulin action, GlcN was administered for 30, 60, 90, or 120 min during 2-h euglycemic, hyperinsulinemic clamp studies. Activation of muscle PI3K by insulin was attenuated following only 30 min of GlcN infusion (GlcN 30 min = 1.5-fold versus saline = 2.5-fold stimulation; p < 0.05). In contrast, the first impairment in insulin-mediated glucose uptake (Rd) developed following 110 min of GlcN infusion (110 min = 39.9 +/- 1.8 versus 30 min = 42.8 +/- 1.4 mg/kg.min, p < 0.05). However, the ability of insulin to stimulate phosphatidylinositol 3,4, 5-trisphosphate production and to activate glycogen synthase in skeletal muscle was preserved following up to 180 min of GlcN infusion. Thus, increased GlcN availability induced (a) profound and early inhibition of proximal insulin signaling at the level of PI3K and (b) delayed effects on insulin-mediated glucose uptake, yet (c) complete sparing of insulin-mediated glycogen synthase activation. The pattern and time sequence of GlcN-induced defects suggest that the etiology of peripheral insulin resistance may be distinct from the rapid and marked impairment in insulin signaling.

Functional analysis of a conditionally transformed pancreatic beta-cell line.

Development of beta-cell lines for cell therapy of diabetes is hindered by functional deviations of the replicating cells from the normal beta-cell phenotype. In a recently developed cell line, denoted betaTC-tet, derived from transgenic mice expressing the SV40 T antigen (Tag) under control of the tetracycline (Tc) gene regulatory system, growth arrest can be induced by shutting off Tag expression in the presence of Tc. Here, we compared differentiated cell functions in dividing and growth-arrested betaTC-tet cells, both in culture and in vivo. Proliferating cells stably maintained normal glucose responsiveness for >60 passages in culture. Growth-arrested cells survived for months in culture and in vivo and maintained normal insulin production and secretion. After growth arrest, the cells gradually increased their insulin content three- to fourfold. This occurred without significant changes in insulin biosynthetic rates. At high passage numbers, proliferating betaTC-tet cells exhibited an abnormal increase in hexokinase expression. However, the upregulation of hexokinase was reversible upon growth arrest. Growth-arrested cells transplanted intraperitoneally into syngeneic recipients responded to hyperglycemia by a significant increase in insulin secretion. These findings demonstrate that transformed beta-cells maintain function during long periods of growth arrest, suggesting that conditional transformation of beta-cells may be a useful approach for developing cell therapy for diabetes.

cAMP-dependent phosphorylation of the cardiac-type alpha 1 subunit of the voltage-dependent Ca2+ channel in a murine pancreatic beta-cell line.

Two voltage-dependent calcium channels (VDCCs) have been reported in pancreatic islets: the beta-cell/endocrine-brain and cardiac subtypes. The cardiac-type alpha 1 subunit was isolated from cultured beta TC3 cells, a murine pancreatic beta-cell line, by immunoprecipitation with a specific polyclonal antibody. We have examined the effects of 1-isobutyl-3-methylxanthine (IBMX) and forskolin, agonists that elevate cAMP in these cells, on the phosphorylation of this subunit in intact beta TC3 cells using a sensitive back-phosphorylation technique. This technique allows quantitative detection of protein phosphorylation that is specifically stimulated by cAMP. The stimulation of intact beta TC3 cells with forskolin or IBMX resulted in the phosphorylation of the cardiac-type alpha 1 subunit as evidenced by a 40-60% decrease in the ability of the 257-kDa form to serve as a substrate in the in vitro back-phosphorylation reaction with [gamma-32P]ATP and the catalytic subunit of cAMP-dependent protein kinase (PKA). The effects of forskolin were time- and concentration-dependent. The concentration-dependency of forskolin-induced phosphorylation of the cardiac-type alpha 1 subunit and the potentiation of glucose-induced insulin secretion were highly correlated, a finding that is consistent with a role for such phosphorylation in mediating at least some of the effects of cAMP on secretion.

Animal model for maturity-onset diabetes of the young generated by disruption of the mouse glucokinase gene.

Glucokinase catalyzes a rate-limiting step in glucose metabolism in hepatocytes and pancreatic beta cells and is considered the "glucose sensor" for regulation of insulin secretion. Patients with maturity-onset diabetes of the young (MODY) have heterozygous point mutations in the glucokinase gene that result in reduced enzymatic activity and decreased insulin secretion. However, it remains unclear whether abnormal liver glucose metabolism contributes to the MODY disease. Here we show that disruption of the glucokinase gene results in a phenotype similar to MODY in heterozygous mice. Reduced islet glucokinase activity causes mildly elevated fasting blood glucose levels. Hyperglycemic clamp studies reveal decreased glucose tolerance and abnormal liver glucose metabolism. These findings demonstrate a key role for glucokinase in glucose homeostasis and implicate both islets and liver in the MODY disease.

Evidence that Rap1 carboxylmethylation is involved in regulated insulin secretion.

Protein carboxylmethylation is a reversible posttranslational modification that regulates protein function. We examined the carboxylmethylation of small GTP-binding proteins in a pancreatic beta-cell line (beta TC cells). In vitro assays showed that carboxylmethylation of a membrane 23-kDa protein was induced by guanine nucleotides, best demonstrated by the nonhydrolyzable GTP analog, guanosine 5'-(3-O-thio)triphosphate (GTP gamma S). GTP gamma S also induced translocation of this 23-kilodalton (kDa) protein from cytosol to particulate fractions. Immunoblotting with antiserum sc-65 raised against Rap1 identified the carboxyl-methylated 23-kDa protein as Rap1. 1) The 23-kDa carboxyl-methylated protein separated by two-dimensional electrophoresis overlapped with the 23-kDa protein detected by immunoblotting. 2) GTP gamma S, in the presence of cytosol, increased the amount of detectable membrane-associated Rap1. Studies in intact beta TC cells demonstrated the carboxylmethylation of the 23-kDa protein in response to glucose and depolarizing concentrations of potassium, an effect that was abolished by the calcium channel inhibitor, D600. Similarly, N-acetyl-S-trans,trans-farnesyl-L-cysteine, an inhibitor of in vivo carboxylmethylation at COOH-terminal S-farnesylcysteine by methyltransferase, inhibited carboxylmethylation of the 23-kDa protein in intact cells and reduced insulin secretion in response to glucose and potassium. These data establish a correlation between insulin secretion and carboxylmethylation of a 23-kDa protein that comigrates with Rap1.

Clonal insulinoma cell line that stably maintains correct glucose responsiveness.

A number of pancreatic beta-tumor cell (beta TC) lines have been derived from insulinomas arising in transgenic mice expressing the SV40 T antigen gene under control of the insulin promoter. Some of these lines secrete insulin in response to physiological glucose concentrations. However, this phenotype is unstable. After propagation in culture, these nonclonal lines become responsive to subphysiological glucose levels and/or manifest reduced insulin release. Here we report the use of soft-agar cloning to isolate single-cell clones from a beta TC line, which give rise to sublines that maintain correct glucose responsiveness and high insulin production and secretion for > 55 passages (over a year) in culture. One of these clonal lines, denoted beta TC6-F7, was characterized in detail. beta TC6-F7 cells expressed high glucokinase and low hexokinase activity, similarly to normal islets. In addition, they expressed mRNA for the GLUT2 glucose transporter isotype and no detectable GLUT1 mRNA, as is characteristic of normal beta-cells. These results demonstrate that transformed beta-cells can maintain a highly differentiated phenotype during prolonged propagation in culture, which has implications for the development of continuous beta-cell lines for transplantation therapy of diabetes.

Ribozyme-mediated attenuation of pancreatic beta-cell glucokinase expression in transgenic mice results in impaired glucose-induced insulin secretion.

Phosphorylation of glucose to glucose 6-phosphate by glucokinase (GK; EC 2.7.1.2) serves as a glucose-sensing mechanism for regulating insulin secretion in beta cells. Recent findings of heterozygous GK gene mutations in patients with maturity-onset diabetes of the young (MODY), a form of type II (non-insulin-dependent) diabetes characterized by autosomal dominant inheritance, have raised the possibility that a decrease in beta-cell GK activity may impair the insulin secretory response of these cells to glucose. To generate an animal model for MODY we have expressed in transgenic mice a GK antisense RNA with a ribozyme element under control of the insulin promoter. Mice in two independent lineages had about 30% of the normal islet GK activity. Insulin release in response to glucose from in situ-perfused pancreas was impaired; however, the plasma glucose and insulin levels of the mice remained normal. These mice are likely to be predisposed to type II diabetes and may manifest increased susceptibility to genetic and environmental diabetogenic factors. They provide an animal model for studying the interaction of such factors with the reduced islet GK activity.

Murine insulinoma cell line with normal glucose-regulated insulin secretion.

Pancreatic beta TC lines derived from insulinomas arising in transgenic mice expressing SV40 Tag under control of the insulin promoter manifest a differentiated beta-cell phenotype and secrete insulin in response to glucose. Previously reported beta TC lines respond to subphysiological extracellular glucose levels compared with normal beta-cells. Recently, several beta TC lines were developed with normal glucose-regulated insulin secretion from insulinomas obtained by breeding of the RIP-Tag transgene from the original C57BI/6 mouse strain into the C3HeB/FeJ strain. One of these beta TC lines, beta TC7, was characterized in detail. Beta TC7 cells express GLUT2 and have levels of glucokinase and hexokinase activity similar to those of normal islets. As a result these cells exhibit a normal glucose concentration dependency for glycolysis and insulin secretion, thus representing an accurate model of beta-cell function. On continuous propagation in culture, beta TC7 cells acquired a response to lower extracellular glucose levels. This change was associated with a fourfold increase in hexokinase activity, without significant changes in glucokinase activity and glucose uptake rates. These findings suggest an important role for glucose phosphorylation rates in regulation of the beta-cell insulin secretory response to glucose.

[Val12] HRAS downregulates GLUT2 in beta cells of transgenic mice without affecting glucose homeostasis.

Glucose-induced insulin release from pancreatic beta cells depends on the beta-cell metabolism of glucose, which generates intracellular signals for secretion. The beta-cell glucose transporter isotype GLUT2 and the glucose phosphorylating enzyme glucokinase have both been implicated in coupling insulin secretion to extracellular glucose levels. Here we present evidence that a pronounced decrease in beta-cell GLUT2 has no immediate effect on glucose homeostasis. Analysis of transgenic mice overexpressing human [Val12]HRAS oncoprotein under control of the insulin promoter reveals a great reduction in plasma-membrane GLUT2 levels. These mice are nonetheless able to maintain normal fed and fasting plasma glucose and insulin levels for a period of several months. Insulin secretion studied in isolated islets and the perfused pancreas is characterized by a normal incremental response to increasing glucose concentrations. Glucose metabolism, as measured by glucose phosphorylation and oxidation in isolated islets, shows a normal dose dependence on extracellular glucose concentrations. These findings suggest that normal GLUT2 expression in beta cells is not essential for glucose sensing. The transgenic mice provide an experimental system for studying the role of glucose phosphorylation in regulation of insulin release in the absence of GLUT2.

Calcium-binding proteins in the parathyroid gland. Detailed studies of parathyroid secretory protein.

In order to identify calcium (Ca2+)-binding proteins in the parathyroid gland, we used electrophoretic blots of proteins separated by a two-dimensional nondenaturing/denaturing gel system and incubated them with 45Ca2+. Parathyroid secretory protein (PSP) and proteins with approximate molecular weights of 98,000, 88,000, 58,000, and 30,000 were noted to bind Ca2+ in cytosolic fractions from bovine parathyroid, adrenal, and pituitary glands. However, differences in the binding affinity and capacity of the various proteins were observed. PSP displayed a low affinity and high binding capacity for Ca2+. In the presence of 5 mM MgCl2 and 60 mM KCl, native PSP (immobilized on nitrocellulose filters) bound 7.5 mol of Ca2+/mol of protein monomer with an apparent Kd of 1.1 mM. Immunoblotting identified the association of PSP with parathyroid cell membranes in a Ca2+-dependent manner. This property, together with its heat stability, distinguished PSP from other cytosolic Ca2+-binding proteins which were identified. There was also evidence for a Ca2+-dependent protein-protein interaction (aggregation) of PSP present in a Nonidet P-40 extract of cell membranes. The high Ca2+ binding capacity of PSP and its Ca2+-dependent membrane association may be features that make PSP a potentially important protein in secretory cells.

Nucleotide sequence of beet western yellows virus RNA.

The nucleotide sequence of the genomic RNA (5641 nt) of beet western yellow virus (BWYV) isolated from lettuce has been determined and its genetic organization deduced. The sequence of the 3'terminal 2208 nt of RNA of a second BWYV isolate, obtained from sugarbeet, was also determined and was found to be very similar but not identical to that of the lettuce isolate. The complete sequence of BWYV RNA contains six long open reading frames (ORFs). A cluster of three of these ORFs, including the coat protein cistron, display extensive amino acid sequence homology with corresponding ORFs of a second luteovirus, the PAV isolate of barley yellow dwarf virus (BYDV) (1,2). The ORF corresponding to the putative viral RNA-dependant RNA polymerase, on the other hand, resembles that of southern bean mosaic virus. There is circumstantial evidence that expression of the BWYV RNA polymerase ORF may involve a translational frameshift mechanism. The ORF immediately following the coat protein cistron may be translated by in-frame readthrough of the coat protein cistron amber termination codon. Similar mechanisms have been proposed for expression of the corresponding ORFs of BYDV(PAV) (1).

Interaction of the regulatory subunit of a type II cAMP-dependent protein kinase with mammalian sperm flagellum.

We have reported previously (Horowitz, J. A., Toeg, H., and Orr, G. A. (1984) J. Biol. Chem. 259, 832-838) that most of the type II cAMP-dependent protein kinases in rat sperm are associated with the flagellum. We have now identified flagellar polypeptides which are capable of forming tight complexes with the regulatory subunit of type II cAMP-dependent protein kinase (RII). Flagellar RII-binding polypeptides were identified using an RII overlay/immunoblot procedure and had apparent subunit Mr of 120,000, 80,000, and 57,000 in rat and 120,000 and 57,000 in bovine flagella. RII is released from the flagellum by disulfide reducing agents, e.g. 1 mM dithiothreitol (DTT). Sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Coomassie Blue staining of the DTT-released material shows that a limited subpopulation of flagellar polypeptides are solubilized by disulfide-reducing agents. Neither tubulin, the dynein ATPase, or any of the RII-binding proteins are released by 1 mM DTT, and thin section electron microscopy revealed that the morphology of the flagellum is unaltered by reducing conditions. Our data established that RII is not linked to the flagellum via a direct disulfide bridge. We propose that RII is released from the flagellum, a highly disulfide cross-linked structure, due to structural changes in the flagellum which disrupts the interaction between RII and its binding proteins.

Enhanced activation of cAMP-dependent protein kinase by rapid synthesis and degradation of cAMP.

Activation of cAMP-dependent protein kinase II by static and dynamic steady-state cAMP levels was studied by reconstituting an in vitro model system composed of hormone-sensitive adenylate cyclase, cyclic nucleotide phosphodiesterase, and cAMP-dependent protein kinase II. The rates of cAMP synthesis were regulated by incubating isolated membranes from AtT20 cells with various concentrations of forskolin. In the presence of 3-methylisobutylxanthine, the rate of protein kinase activation was proportional to the rate at which cAMP was synthesized, and there was a direct relationship between the degree of activation and the level of cAMP produced. The activation profiles of protein kinase generated in the presence of exogenous cAMP or cAMP produced by activation of adenylate cyclase in the absence of cAMP degradation were indistinguishable. Dynamic steady-state levels of cAMP were achieved by incubating the membranes with forskolin in the presence of purified cyclic nucleotide phosphodiesterase. Under these conditions, the apparent activation constant of protein kinase II for cAMP was reduced by 65-75%. This increased sensitivity to activation by cAMP was seen when phosphotransferase activity was measured directly in reaction mixtures containing membranes, protein kinase, and histone H2B or when regulatory and catalytic subunits were first separated by immunoprecipitation of holoenzyme and regulatory subunits with specific anti-serum. Our results are consistent with the hypothesis that rapid cAMP turnover may function as a mechanism for amplifying hormonal signals which use the cAMP-dependent protein kinase system.

Somatostatin inhibits corticotropin-releasing factor-stimulated adrenocorticotropin release, adenylate cyclase, and activation of adenosine 3',5'-monophosphate-dependent protein kinase isoenzymes in AtT20 cells.

The mechanisms by which somatostatin (SRIF) inhibits CRF-induced ACTH secretion from AtT20 cells were characterized by comparing the effects of SRIF on cAMP production, adenylate cyclase activity, and activation of cAMP-dependent protein kinase isoenzymes with its effects on ACTH release. In isolated membranes, CRF (100 nM) stimulated adenylate cyclase activity 4- to 5-fold. SRIF inhibited CRF-stimulated adenylate cyclase in a concentration-dependent manner. However, maximal inhibition was 50%. SRIF did not inhibit basal adenylate cyclase or forskolin-stimulated cyclase in the absence of guanine nucleotides and had only small effects on forskolin-stimulated cyclase when assayed in the presence of guanine nucleotides. CRF (100 nM) induced small rises (2-fold) in intracellular cAMP levels which produced maximal ACTH release. SRIF inhibited basal and CRF-stimulated ACTH release in a concentration-dependent manner, and there was a good correlation between inhibition of ACTH release and inhibition of the activation of cAMP-dependent protein kinases in these cells. Thus, the effect of SRIF on CRF-induced ACTH release appeared to result from its effect on inhibition of adenylate cyclase. In the presence of 3-methylisobutylxanthine (MIX), CRF increased cAMP levels 20-fold and activated a greater proportion of cAMP-dependent protein kinase, but did not stimulate ACTH release more than CRF alone. Under these conditions, SRIF (100 nM) inhibited cAMP accumulation by 90%. ACTH release was also inhibited, but higher concentrations of SRIF were required to block ACTH release compared to cells incubated in the absence of MIX. Sufficient cAMP levels were achieved so that activation of cAMP-dependent protein kinases was only partially blocked. There was still sufficient cAMP to activate cAMP-dependent protein kinase to an extent equal to that seen with CRF without MIX. Similar effects of SRIF on cAMP accumulation and protein kinase activation were seen when cells were stimulated with forskolin. Our results demonstrate that SRIF inhibits ACTH release from AtT20 cells by inhibiting hormone-sensitive adenylate cyclase and thereby prevents the activation of cAMP-dependent protein kinases. However, under conditions where cAMP-dependent protein kinases are still sufficiently active to induce ACTH secretion, high concentrations of SRIF can inhibit ACTH release by a mechanism independent of cAMP-dependent protein kinase.

Differential binding of the regulatory subunits (RII) of cAMP-dependent protein kinase II from bovine brain and muscle to RII-binding proteins.

The association of regulatory subunits (RII) of Type II cAMP-dependent protein kinase from bovine cerebral cortex (RII-B) and bovine cardiac and skeletal muscle (RII-H) with specific binding proteins in bovine brain cytosol and purified brain microtubules was demonstrated using a solid phase binding assay. RII-binding proteins present in bovine cerebral cortex were immobilized on nitrocellulose filters after separation by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Incubation of the filters with 32P-labeled regulatory subunits showed that both RII-B and RII-H interact with the 75,000-dalton calmodulin-binding protein (P75) and microtubule-associated protein 2 (MAP-2). However, significant differences in binding affinities and capacities were observed. RII-B displayed a higher affinity for P75 compared to RII-H while RII-H preferentially bound to MAP-2. Quantitation of radioactive RII bound to MAP-2 showed that MAP-2 bound 4-6 times more RII-H than RII-B. The differential binding affinities and capacities of RII-H and RII-B for MAP-2 were not affected by autophosphorylation since both phospho and dephospho forms of RII displayed the same binding characteristics. Competitive binding studies suggest that RII-H and RII-B bind to the same sites on MAP-2. The biochemical basis for the differential binding of RII-B and RII-H to the same sites of MAP-2 is unknown. However, other high affinity RII-binding proteins present in cerebral cortex (i.e. P75) might affect the affinity of RII-B for MAP-2. 32P-RI did not bind to P75 nor MAP-2 under the conditions used.

Postillumination adenosine triphosphate synthesis in Rhodospirillum rubrum chromatophores. I. Conditions for maximal yields.

The very low level of postillumination ATP synthesis in chromatophores was markedly stimulated when permeant anions (thiocyanate or perchlorate) or permeant cations (potassium in the presence of valinomycin) were added to the light stage. Although these compounds stimulated also light-induced proton uptake in chromatophores the pH dependence of both photoreactions was different. Proton uptake peaked at pH 6.5 while the amount of postillumination ATP was maximal when the light stage was carried out around pH 7.7. The increased yield of ATP at the more alkaline pH could not be explained by a slower decay of the high energy state at this pH, since the decay rate was faster at pH 7.7 than at pH 6.5. The proton concentration gradient which is maintained across the chromatophore membrane in the light was also found to increase when the external pH was raised from 6.0 to 8.0. Only a minimal amount of postillumination ATP was formed when this gradient was below 2.1 pH units, but above this value the ATP yield rose steeply as a function of the increasing pH gradient. In light of these results it is suggested that in order to obtain a high yield of postillumination ATP synthesis in chromatophores two conditions are required: the particles have to be loaded with a sufficient number of protons and a light-induced pH gradient above a certain threshold value has to be maintained across their membrane. The low yield of postillumination ATP in chromatophores and the increase obtained by adding permeating ions, is thus explained by similar variations in the extent of the pH gradient, which exceeded the threshold value only in the presence of the permeating ions.

Postillumination adenosine triphosphate synthesis in Rhodospirillum rubrum chromatophores. II. Stimulation by a K+ diffusion potential.

Addition of valinomycin, nonactin, or monactin plus KCl in the dark to preilluminated chromatophores induced the synthesis of a large amount of ATP. This stimulation of postillumination ATP synthesis by a dark-imposed K+ diffusion potential was different from the stimulation caused by addition of permeant anions or cations in the light, since it increases when the pH of the light stage decreased from 8.0 to 6.0. It was thus most pronounced when the chromatophores were preloaded with protons but the light-induced proton concentration gradient (deltapH) was low. Imposition of a Kplus diffusion potential resulted however in stimulation of ATP synthesis even when the light-induced deltapH was already above the threshold value required to initiate postillumination ATP synthesis. This situation was realized when valinomycin plus KCl were added in the dark to chromatophores preilluminated above pH 6.7 with thiocyanate as the permeant anion, and the amount of ATP formed was the sum of the yields obtained with each of these affectors by itself. On the other hand addition of thiocyanate together with valinomycin plus KCl in the dark led to inhibition of ATP synthesis. In this case the permeant anion could not affect the light-induced deltapH but it did eliminate the diffusion potential by decreasing the difference between the permeabilities of Kplus and the anion present in the reaction mixture.