Isolation of highly multidrug-resistant P388 cells from drug-sensitive P388/S cells by flow cytometric cell sorting

To investigate the spontaneous frequency of occurrence of stable multidrug-resistant cells in a population of drug-sensitive cells, we exposed drug sensitive P388/S cells to daunorubicin (dnr) for 1 h, then used fluorescence-activated cell sorting based on intracellular dnr fluorescence to isolate cells within P388/S having different intracellular content of drug. One of the sort windows chosen (low dnr content sort window) isolated only P388/S cells with intracellular drug content equal to or less than that of the known multidrug-resistant subline P388/adr. This sort window constituted approximately 3% of P388/S cells with lowest dnr content. By such a procedure we were able, on one of seven attempts, to isolate and cultivate stable, highly multidrug-resistant cells (comparable to that of P388/adr) from the P388/S cells obtained from the low dnr-content sort window. Net growth of cells in culture was observed 15-20 days after sorting, indicating that of the P388/S cells collected from the low dnr-content sort window, very few were actually highly drug-resistant. On no occasion could resistant cells be cultivated from cells sorted from P388/S with higher dnr content, as would be expected if mutation to a multidrug-resistant phenotype had occurred as a result of exposure to drug. The resistant cells isolated from P388/S by sorting (called P388/LoSort) displayed low intracellular accumulation of dnr that was enhanced by verapamil, were cross-resistant to vincristine and actinomycin-D, and distinct from P388/S, possessed a 150- to 160-kD membrane species identified by Vinca alkaloid photoaffinity labeling.(ABSTRACT TRUNCATED AT 250 WORDS)     

Inductive effect of recombinant human interleukin-1 alpha and beta on differentiation of macrophage-like tumor cell line P388D1

We used the mouse monocyte/macrophage-like tumor cell line P388D1 to test whether or not interleukin-1 (IL-1) stimulates differentiation of monocyte/macrophage progenitors. Incubation of these cells with recombinant human interleukin-1 (rhIL-1) alpha and beta resulted in their increased adherence, stimulation of nonspecific esterase activity, and increased Fc rosette formation. rhIL-1s inhibited cell growth and stimulated Fc rosette formation in a dose-dependent fashion. The cell growth inhibition due to rhIL-1s depended on the concentration of serum in culture medium. Synergism between rhIL-1 and calcium ionophore A23187 was found for the cell growth inhibition and Fc rosette formation. The presence of ethylene glycol bis- (beta-aminoethyl ether) N,N,N,N,-tetraacetic acid(EGTA) in the medium abolished the stimulatory effect of rhIL-1 on Fc rosette formation of the cell line. These results demonstrate that rhIL-1s are a potent inducer of the differentiation of the macrophage-like tumor cell line P388D1.     

PGE2-induced desensitization of adenylate cyclase of a murine macrophage-like cell line (P388D1)

The mode of PGE2-induced desensitization of the adenylate cyclase of a murine macrophage-like cell line, P388D1 was investigated. The exposure of cells to PGE2 for 60 min induced PGE2-specific desensitization of the adenylate cyclase system which still responded normally to other specific ligand such as isoproterenol, 5'-guanylimidodiphosphate (Gpp(NH)p), or forskolin. The exposure of the cells to PGE2 for 6 hr induced heterologous desensitization, as the responses of adenylate cyclase to PGE2 as well as to isoproterenol or Gpp(NH)p were significantly reduced. The lowest concentration of PGE2 to induce both early homologous and late heterologous desensitization was found to be about two-fold over the KD of the low affinity PGE2-binding sites of P388D1 cells. The early homologous desensitization appeared to be due in part to the reduction in number of PGE2 receptors from the cell surface. The late heterologous desensitization may involve functional and/or structural alteration of Gs proteins, in addition to the reduction of PGE2 receptors from the cell surface.     

Phospholipase activities of the P388D1 macrophage-like cell line

The murine macrophage (M phi) cell line, P388D1, was employed as a source of M phi phospholipases in order to characterize the enzymatic properties and subcellular localization of these enzymes because of their importance for prostaglandin biosynthesis. Phospholipase activity was assessed with dipalmitoylphosphatidylcholine (DPPC) as substrate. Phospholipases were characterized with respect to divalent cation dependence, pH optima, and localization in subcellular compartments using linear sucrose gradients. By these criteria a number of different phospholipases were identified. Most importantly, a single Ca2+-dependent activity with a pH optimum of 8.8 was identified in membrane-rich fractions (plasma membrane, mitochondria, and endoplasmic reticulum) and could be clearly separated from the remaining activities, which are Ca2+ independent and exhibit pH optima of 7.5, 5.1, and 4.2. The phospholipases with acidic pH optima may be associated with subcellular components containing lysosomal enzymes and both phospholipase A1 and phospholipase A2 activities are observed. In contrast, the phospholipase activity with a pH optimum of 7.5 sediments with the cytosolic proteins and is inhibited by 5 mM Ca2+. No significant phospholipase C activity was detected in assays performed with or without added Ca2+ at pH's 4.2, 5.1, 7.5, or 8.8 using DPPC as substrate. However, the P388D1 cells do contain a lysophospholipase that is at least 20 times more active than the phospholipase A activities identified. Its presence must be taken into account in evaluating the positional specificities and properties of the macrophage phospholipases.     

Effects of cis-unsaturated fatty acids on doxorubicin sensitivity in P388/DOX resistant and P388 parental cell lines

It has been reported that several cis-unsaturated fatty acids (c-UFAs) could increase doxorubicin (DOX) accumulation in cancer cells and hence elevate its cytotoxicity. However, some researchers showed that c-UFA pretreatment did not affect its cytotoxicity in special cell lines. It is possible that the different results occurred due to different cellular characteristics. We hypothesized that c-UFA treatment might modulate the activities of some antioxidant enzymes to affect the resistance of cells to DOX. In the present study, we examined how c-UFA pretreatment affected DOX cytotoxicity on mouse leukemia cell line, P388, and its resistant subline, P388/DOX, which we found to have significantly higher glutathione peroxidase (GPx) activity as well as P-glycoprotein (p-gp) overexpression. We chose two c-UFAs, gamma-linolenic acid (GLA) (18:3n-6) and docosahexaenoic acid (DHA) (22:6n-3). Cytotoxicity was measured by MTT (3-(4.5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) and trypan blue exclusion assays. DOX accumulation and p-gp expression were measured by flow cytometry. The activities of catalase (CAT), superoxide dismutase (SOD), glutathione S-transferase (GST), and GPx were determined for both cell lines with and without treatment with GLA or DHA. Significant DOX accumulation occurred in both cell lines with GLA or DHA pretreatment, but without any change in p-gp expression in either cell line. Sensitivity to DOX cytotoxicity was improved by GLA or DHA pretreatment in P388/DOX in which only SOD activity was significantly increased, but not in the parental cell line P388 in which both SOD and CAT were significantly increased by the pretreatment. However, combined pretreatment of GLA or DHA with antioxidants, pyrrolidinedithiocarbamate (PDTC) or Vitamin C, could sensitize not only P388/DOX but also P388 cells to DOX. We conclude that the effects of c-UFA pretreatment on the sensitivity of cancer cells to DOX not only depend on the change in drug accumulation but also the change in the levels of antioxidant enzyme activities, and suggest that combined administration of c-UFAs, antioxidants, and DOX may be more effective in treating leukemia.     

Role of self carriers in the immune response and tolerance. X. A lymphoid dendritic-like tumor, P388AD.2, acts as a novel immunogenic carrier for hapten

Hapten-modified spleen cells, peritoneal exudate cells, and certain lymphoid tumors preferentially induce specific tolerance after i.v. administration. In contrast to these tolerogenic carrier cells, we found that a haptenated lymphoid dendritic-like tumor, P388AD.2, acts as a potent immunogen after i.v. injection. The immunogenicity of P388AD.2 was analyzed by measuring the specific augmentation of plaque-forming cell (PFC) responses when spleen cells from mice previously injected with haptenated tumor cells were challenged in vitro with thymus-independent antigens. Optimal immunization was found to be dependent on cell dose and hapten concentrations. Further studies indicated that P388AD.2 elicited a response which was T cell-dependent and which involved both the so-called Lyb-3,5,7- and Lyb-3,5,7+ B cell populations. Injection of haptenated tumor into different mouse strains suggested that H-2 compatibility was required to prime B cells in vivo, although significant augmentation could also be achieved in allogeneic C57B1/6J mice. The enhanced PFC responses elicited in H-2b mice could not be explained by allo-recognition of class I or II MHC determinants. In toto, these results suggest that P388AD.2 acts as a unique accessory cell for the presentation of hapten-modified self.     

Fc gamma 2b receptor-mediated prostaglandin synthesis by a murine macrophage cell line (P388D1)

The nature of signals transmitted by two types of Fc gamma receptors (one specific for IgG2b and the other for IgG2a) present on the surface of a murine macrophage cell line (P388D1) was investigated. Specific binding of IgG2b (presented as EA2b) to cell surface Fc gamma 2br triggered the release of 3H-arachidonic acid and 3H-prostaglandins (PG) from P388D1 cells that were prelabeled with 3H-arachidonate. The release of 3H-arachidonic acid, which increased in a dose-dependent manner, was enhanced by exogenous Ca++ (1.25 mM) and was completely blocked by ethylenediaminetetraacetate (EDTA) (4 mM) or a phospholipase A2 inhibitor, p-bromophenacylbromide (7 microgram/ml). A cyclooxygenase inhibitor, indomethacin (9 microgram/ml), reduced the 3H-arachidonic acid release and completely blocked the conversion of arachidonate into PG. Cytochalasin D (1 microgram/ml), which inhibited the phagocytosis of immune complexes by 90% of P388D1 cells, did not affect the Fc gamma 2bR-triggered release of arachidonic acid. Specific binding of IgG2a (presented as EA2a) to cell surface Fc gamma 2aR did not trigger the release of either 3H-arachidonic acid or 3H-PG from P388D1 cells. Our data demonstrate a signal for the activation of the arachidonic acid metabolic cascade is transmitted by Fc gamma 2bR, but not by Fc gamma 2aR, on the surface of P388D1 cells, probably through the initial activation of the phospholipase A2 activity associated with Fc gamma 2bR.     

Effect of c-fos overexpression on the murine macrophage cell line P388DI

In order to study the role of Fos on the regulation of proliferation in the monocyte-macrophage lineage we realized a stable transfection of the murine P388D1 cell line by the murine c-fos gene under the control of the human metallothionein IIA promoter. Several clones have been selected by geneticin: they show a variable number of integrated transgene (two to ten copies). Their expression has been analyzed in the presence or absence of cadmium chloride as inducer (5 × 10⁻⁶ M). In one clone especially, the c-fos mRNA and Fos protein levels were respectively 6and 10-fold increased. The study of cell growth by tritiated thymidine incorporation indicates a negative effect of the overexpressed Fos protein in the absence of any other stimulus.     

Properties of prostaglandin-sensitive adenylate cyclase system of a murine macrophage-like cell line (P388D1)

The properties of basal and prostaglandin (PG)-stimulated adenylate cyclase of membrane preparations of P388D1 cells were investigated. Three partially purified membrane fractions were obtained by sucrose density gradient centrifugation at the final step of purification from crude homogenate. About 96% of the basal and 89% of PGE2-stimulated adenylate cyclase activity in the homogenate were recovered in three membrane fractions. Two lighter membrane fractions (I and II), which were enriched 11-fold and 8.4-fold in adenylate cyclase activity over crude homogenate, were pooled and subjected to various studies. Results suggested that the basal activity of the membrane preparations has, as in many other cell types, a relatively broad pH optimum (pH 7.5 to 8.5), requires Mg2+, which must be present in excess ATP, and is inhibited by Ca2+. Highly reactive sulfhydryl group(s), which may be present in the lipid bilayer, is required for the adenylate cyclase activity. Because both fluoride ions and GTP augment the enzymatic activity, P388D1 cell membrane adenylate cyclase must possess stimulatory guanine nucleotide-binding protein. The membrane preparations respond to exogeneously added PG by 1.5-fold to 3-fold increase in adenosine 3'-5' cyclic monophosphate (cAMP) production. The magnitude of PG-responsiveness was dependent on the types of PG and the order of potency in stimulation was PGE1 greater than PGE2 greater than PGI2. PGA1, B1, B2, F1 alpha, and F2 alpha stimulated adenylate cyclase only at the highest concentration tested.     

Enantioselective kappa opioid binding sites on the macrophage cell line, P388d1.

A kappa (kappa) opioid binding site has been characterized on the macrophage cell line, P388d1, using the kappa selective affinity ligand, [3H] (1S,2S)-(-)-trans-2-isothiocyanato-N-methyl-N-[2-(1- pyrrolidinyl) cyclohexyl] benzeneacetamide (-)BD166). The kappa site has a relative molecular mass (Mr) of 38,000 under nonreducing conditions and 42,000 under reducing conditions. Moreover, it exhibits enantioselectivity in that 1S,2S-(-)-trans-3,4-dichloro-N-methyl-N-[2-1-pyrrolidinyl)cyclohexyl] benzeneacetamide ((-)-U-50,488) blocks [3H](5 alpha, 7 alpha, 8 beta)-(-)-N-methyl-N-[7-(1- pyrrolidinyl)-1-oxaspiro-(4,5)-dec-8-yl]benzeneacetamide (U-69,593) binding to P388d1 cells with an IC50 = 7.0 nM whereas 1R,2R-(+)-trans-3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)cyclohexyl] benzeneacetamide ((+)U-50,488) blocks [3H]U-69,593 binding to P388d1 cells with an IC50 = 7000 nM.     

Induction of functional Fc receptors in P388 leukemia cells : Requirement for multiple differentiation signals

The development of functional Fc receptors (FcR) during induced differentiation with the tumor promoter, phorbol myristate acetate (PMA), was studied in the murine tumor cell line, P388. PMA induced the appearance of FcR on the membranes of P388 cells as indicated by the binding of IgG-coated sheep red blood cells (IgG-SRBC). Concentrations of PMA as low as 1 ng/ml were sufficient to induce the expression of FcR as well as to inhibit cellular division and to induce adherence in the P388 tumor cell line; however, optimal FcR induction occurred at PMA concentrations of 10-100 ng/ml. Immunofluorescent analysis with heat-aggregated myeloma proteins indicated that PMA induced FcR which were capable of binding IgG2a and IgG2b immunoglobulins, but not IgG1. Adherence to a substratum was determined to be a second required signal for expression of FcR, since PMA induction of P388 tumor cells in teflon dishes failed to fully develop FcR and adherence of P388 cells to poly-L-lysine-coated culture dishes in the absence of PMA was insufficient for FcR expression. FcR which appeared after PMA induction were non-functional in the sense that membrane-bound IgG-SRBC were not ingested to any significant extent by the tumor cells. However, if FcR induction occurred in the presence conA-induced rat spleen cell culture supernatants, phagocytosis of membrane-bound erythrocytes occurred. These findings suggest that for the expression of FcR which are capable of particle internalization, at least three identifiable membrane-transmitted signals are required during differentiation.     

Lysosome-associated membrane proteins: characterization of LAMP-1 of macrophage P388 and mouse embryo 3T3 cultured cells

Lysosome-associated membrane protein (LAMP)-1, a major glycoprotein of mouse embryo 3T3 cells and specifically associated with the lysosomal membrane, has been identified in P388 macrophage cells and compared with the homologous glycoprotein of NIH 3T3 cells. Immunofluorescence microscopy with anit-LAMP-1 monoclonal antibodies shows that the antigen was distributed throughout P388 cells including the ruffled edges or pseudopodia, identical to the pattern of acridine orange accumulation. LAMP-1 was purified from P388 cells by affinity chromatography with 1D4B monoclonal antibody, yielding a homogeneous glycoprotein comprising 0.1% of the total detergent-extracted cell protein. The apparent mass of P388 LAMP-1 was 130,000 to 150,000 compared to the 3T3 glycoprotein of 105,000 to 115,000. Analysis of tryptic peptides indicated that the two purified glycoproteins were highly homologous. Protein synthesis was analyzed in a variety of cell lines by pulse-chase labeling with [35S]methionine; in every case, LAMP-1 was synthesized as a precursor of apparent Mr 92,000, and then converted to heterogeneous mature forms differing in average Mr from 110,000 to 140,000. The basis for these apparent differences in mass was examined by studies of the biosynthesis and oligosaccharide composition of the glycoprotein. Core polypeptides of 45,000 Da were obtained from both HaNIH and P388 cells by treating immunoprecipitates of [35S]methionine pulse-labeled molecules with endoglycosidase H. Cells treated with monensin contained heterogeneous molecules of 80,000 to 85,000 Da. Isoelectric heterogeneity of mature LAMP-1 was markedly reduced by treatment with neuraminidase whereas there was little effect on the apparent molecular weight of the molecules or the differences between the various cell lines. beta-D-Xyloside inhibition of glycosaminoglycan synthesis had little effect on the apparent mass of LAMP-1.     

In vivo growth kinetics of P388 and L1210 leukemias

The P388 lymphocytic leukemia and the L1210 lymphoid leukemia are used as test systems for putative cytotoxic drugs. These leukemias are also used to investigate the perturbation of cell cycle progression of various chemical compounds in more detail. There is little information on the normal growth kinetics in vivo of these leukemias. In the present report we therefore present the results from growth kinetic studies of P388 and L1210 leukemic cells growing in ascites form in mice. We used 3H-TdR autoradiography, DNA flow cytometry and the stathmokinetic method. During exponential growth both leukemias showed a growth fraction of unity. Whereas no significant cell loss was observed during the early growth phase of P388 cells, cell loss was indicated by a discrepancy between potential and actual doubling times during exponential growth of L1210 cells. During the phase of growth retardation, the proportion of G1 and G2 cells increased at the expence of a reduced S phase fraction in the P388 leukemia, whereas only small changes in cell cycle distributions were seen with time after inoculation of L1210 cells. An increasing discrepancy in the reduction of the S phase fraction and the 3H-TdRLI was seen in the P388 cells with time after inoculation. Thus, a majority of P388 cells with S phase DNA content were unlabelled during the late phase of growth restriction, indicating resting cells in S phase. A good correlation was found between the 3H-TdR LI and S phase fraction throughout the life history of L1210 cells, revealing considerable differences in in vivo growth kinetics between the two leukemias. Such differences should be considered when evaluating test results.     

Antigen presentation to T cell hybridomas by a macrophage cell line: an inducible function

We have investigated the ability of an Ia-, nonantigen-presenting macrophage tumor cell line, P388D, (H-2d), to present antigen to T cell hybridomas after incubation in a lymphokine-containing preparation. P388D, cells were incubated in microtiter wells with various concentrations of Con A-stimulated spleen cell supernatants. Antigen-specific stimulation of H-2d-restricted, KLH-specific T cell hybridomas was observed by P388D1 incubated with SUP.P388D1 cells incubated for 3 days in medium or control SUP did not present antigen. In addition, no stimulation of T hybridomas was seen by P388D1 in the inhibited by the appropriate monoclonal anti-Ia reagents. These results demonstrate that a macrophage tumor cell line can be induced to present antigen and provides for large numbers of readily available, homogeneous macrophages for studying the cellular biochemical requirements for antigen processing and presentation.     

Changes in Sensitivity of Leukemia P388 Cells to Oxidative Stress and Platidiam upon Tumor Growth

We studied the effect of oxidative stress induced by hyperoxia, hydrogen peroxide, or menadione on leukemia P388 cells isolated from mice with early (4 days) and late (7 days) stages of tumor growth. Oxidative stress was shown to inhibit cell division and to induce apoptosis. The seven-day leukemia cells feature lower proliferative potential and higher sensitivity to oxidative stress and platidiam.     

Suppression of mitochondrial permeability transition pore and induction of lymphoma P388 cell death by cyclosporin A

Suppression of the mitochondrial permeability transition pore (PTP) and induction of lymphoma P388 cell death were studied in the presence of cyclosporin A (CsA) and its derivatives. In experiments with permeabilized P388 cells, CsA and its nonimmunosuppressive derivative N-methyl-Val-4-CsA, but not cyclosporin H (CsH), enhanced Ca2+ accumulation in mitochondria and suppressed PTP opening. Moreover, CsA was able either itself to induce or to enhance a prooxidant-induced P388 programmed cell death. Blebbing and formation of apoptotic bodies were among the observed CsA effects. N-Methyl-Val-4-CsA showed similar effects, but CsH had no effect on P388 cell death. These results show that initial-stage P388 tumour cell death is not related to PTP opening but can be the result of PTP closing with a corresponding increase in the formation of reactive oxygen species.     

Characterization of lymphocyte-activating factor (LAF) produced by a macrophage cell line, P388D1. II. Biochemical characterization of LAF induced by activated T cells and LPS.

Unstimulated P388D1 cells, as well as P388D1 cells stimulated with PHA-activated guinea pig T lymphocytes or LPS, produced a lymphocyte activating factor (LAF). In order to have a chemical basis for comparing this LAF with the LAF produced by normal macrophages, we have analyzed several biochemical characteristics of the P388D1-derived LAF. Sephadex G-75 chromatography of concentrated LAF-containing supernatants from cultures of unstimulated and T cell stimulated P388D1 cells demonstrated that the cell line LAF had a m.w. of approximately 16,000. On DEAE cellulose, the T cell-induced LAF fractionated into at least three major peaks and one minor peak. By using hydroxylapatite chromatography, two of the major peaks of LAF activity were separated from residual contaminating Lowry positive material. LPS-stimulated P388D1 also produced LAF with a m.w. of 16,000. However, the LPS-induced LAF appeared to lack one of the DEAE peaks of LAF activity observed with the T cell-derived LAF. In contrast to LPS, T cells may induce the synthesis and/or release of an additional LAF component or enzymatically modify one or more of the LAF species that are produced in response to both stimulants. Based on the results of chemical characterization studies, the P388D1-derived LAF appears to be similar in size and charge to the lymphocyte activating factor produced by normal macrophages.     

Purification and NH2-terminal sequence of a plasmacytoma growth factor derived from the murine macrophage cell line P388D1

Plasmacytoma growth factor (PCT-GF), a putative macrophage-derived lymphokine essential for the in vitro viability and proliferation of early generation plasmacytomas, was purified from conditioned medium of the murine macrophage cell line P388D1. The purification of PCT-GF was accomplished by a batch concentration on trimethylsilyl-controlled pore glass beads, followed by: gel filtration chromatography; hydrophobic interaction HPLC; and reverse-phase HPLC. SDS-PAGE analysis of the purified PCT-GF revealed a single band of Mr 23,000. The amino terminal sequence of PCT-GF was established as NH2-Pro-Thr-Ser-Gln-Val-Arg-Arg-Gly-Asp-Phe-Thr-Glu-Asp-Thr-Thr-Pro-Asn- Arg-Pro-Val-Tyr-Thr. No significant homology was found between this sequence and proteins in the National Biomedical Research Foundation database, suggesting that PCT-GF is a new lymphokine unrelated to previously described growth and differentiation factors.     

Interferon-gamma down-regulates membrane adenylate cyclase activity of a murine macrophage-like cell line (P388D1)

Effects of recombinant murine interferon-gamma (rIFN-gamma) on the membrane adenylate cyclase of a murine macrophage cell line (P388D1) were investigated in order to explore the nature of a signal transmitted by IFN-gamma receptor. Following the incubation of P388D1 cells with 40 U/ml of rIFN-gamma, the intracellular level of cAMP gradually increased about twofold over the control level within 60 min, and then began to gradually decline to about half the control level by 24 h incubation. The initial rise in cAMP level appeared to be due to the modest activation of adenylate cyclase and not due to the inhibition of cAMP-phosphodiesterase. Later decrease of intracellular cAMP may be due to quantitative down-regulation of the adenylate cyclase system. The basal enzymatic activity of the membrane prepared from P388D1 cells exposed to IFN-gamma for 24 h was found to be reduced to about 20% of that of the control membrane. However, the quality of the adenylate cyclase system appeared unchanged, because the relative degree of the response of the down-regulated membrane adenylate cyclase to prostaglandin PGE2, NaF, guanylimidodiphosphate (GppNHp), choleara toxin (CT), or forskolin was found to remain unchanged. This quantitative down-regulation of adenylate cyclase must be due to the action of rIFN-gamma, since the prior treatment of rIFN-gamma with either acid (pH 2) or monoclonal anti-IFN-gamma antibody inhibited the ability of IFN-gamma to induce the down-regulation. The rIFN-gamma-induced down-regulation is a reversible process, since the adenylate cyclase activity of the membrane was found to be restored when the rIFN-gamma-exposed cells were cultured for 72 h in the absence of rIFN-gamma. In addition, the 48 h-incubation of P388D1 cells with rIFN-beta or IFN-alpha was found not to significantly affect the membrane adenylate cyclase system.