Tight binding of the angiotensin AT(1) receptor antagonist.

Angiotensin II induces angiotensin AT(1) receptor internalization via Clathrin coated pits formation. We investigated whether insurmountable inhibition by the non-peptide antagonist 2-ethoxy-1-[(2'-(1H-tetrazol-5-yl) biphenyl-4-yl) methyl]-1H-benzimidazoline-7-carboxylic acid (candesartan) was related to receptor internalization. Mild acid treatment can discriminate between internalized and cell surface bound [(3)H]angiotensin II. In contrast, it provides no information about the subcellular localization of bound [(3)H]candesartan since this binding is acid resistant. The internalization of [(3)H]angiotensin II is rapidly inhibited in the presence of 0.4 M sucrose. Yet, no such rapid effect was noticed for [(3)H]candesartan. [(3)H]candesartan displays insurmountable/long lasting binding to the vast majority of both wild type and L(314) truncated rat angiotensin AT(1A) receptors with impaired receptor internalization. In agreement with previously published AT(1) angiotensin receptor visualization experiments, the present data suggest that non-peptide antagonist-angiotensin AT(1) receptor complexes remain at the cell surface. Insurmountable antagonism of candesartan is therefore independent from receptor internalization via clathrin-coated pits.

Extracellular signal-regulated kinase and the small GTP-binding protein p21Rac1 are involved in the regulation of gene transcription by angiotensin II.

To study the role of extracellular-signal-regulated kinase (ERK) cascade and the small GTP-ase proteins in the activation of the c-fos promoter by angiotensin II (AII), transient transfection experiments were performed in CHO cells stably expressing the rat AT(1A) receptor. In this system AII activated ERK in 1 min and also increased the transcriptional activity of the c-fos promoter-luciferase reporter gene construct. The activation of the promoter proved to be dependent on the Ras-Raf-ERK cascade as cotransfection of expression vectors known to specifically inhibit this cascade blocked the effect of AII. Dominant-negative p21Rac1 mutant partially blocked the activation of the c-fos promoter by AII. However, activation of the c-fos promoter was independent of protein kinase C (PKC) as bisindolylmaleimide I, a specific PKC inhibitor did not block the effect of AII. These results suggest that AII activates the transcription of the c-fos through the Ras-Raf-ERK cascade. Furthermore, p21Rac1 is involved in the modulation of the c-fos promoter by AII.

Functional role of a novel cis-acting element (GAGA box) in human type-1 angiotensin II receptor gene transcription.

GH/growth factors have been shown to increase angiotensin type 1 receptor expression. In the present study we determined the cis-acting regulatory region controlling GH-induced transcription of the human type-1 angiotensin receptor (hAT(1)). In human proximal tubule cells transfected with a chloramphenicol acetyl transferase (CAT) reporter plasmid under the control of the hAT(1) promoter, GH induced CAT activity. Serial deletions of the hAT(1) promoter region indicated that an area between -314 bp and -70 bp upstream of the 5'-end of the cDNA sequence was essential for this activation to occur. Although sequence analysis identified putative multiple nuclear protein binding sites in this region, we determined that a 12 bp sequence (5'-GAGAGGGAGGAG-3', GAGA box) located between -161 bp and -149 bp was important for GH-mediated activation. Using mobility shift assays we demonstrated increased DNA binding activity to the labeled GAGA box in nuclear extracts treated with GH, suggesting this sequence is a GH response element. Southwestern analysis identified an 18 kDa GAGA box-binding protein (GAGA-BP). GH-induced activity of the GAGA-BP occurred within 2.5 min and reached a maximum at 5 min. Activation did not require de novo protein synthesis. Removal of the GAGA box abolished GH-induced transcription as well as basal transcription of the hAT(1) gene. Additional studies demonstrated that epidermal growth factor, platelet-derived growth factor and insulin activate the GAGA-BP, suggesting these growth factors can also regulate the transcription of the hAT(1) gene through the GAGA box. Our data show that the GAGA-BP acts as a trans-acting factor binding to the cis-acting regulatory element in the hAT(1) promoter, which is necessary for the basal and growth factor(s)-mediated transcriptional activation of the hAT(1) gene.

Angiotensin (AT1A) receptor-mediated increases in transcellular sodium transport in proximal tubule cells.

Angiotensin II (ANG II), acting through angiotensin type 1A receptors (AT1A), is important in regulating proximal tubule salt and water balance. AT1A are present on apical (AP) and basolateral (BL) surfaces of proximal tubule epithelial cells (PTEC). The molecular mechanism of AT1A function in epithelial tissue is not well understood, because specific binding of ANG II to intact PTEC has not been found and because a number of isoforms of AT receptors are present in vivo. To overcome this problem, we developed a cell line from opossum kidney (OK) proximal tubule cells, which stably express AT1A (Kd = 5.27 nM, Bmax = 6.02 pmol/mg protein). Characterization of nontransfected OK cells revealed no evidence of AT1A mRNA (reverse transcriptase-polymerase chain reaction analysis) or protein (125I-labeled ANG II binding studies) expression. In cells stably expressing AT1A, ANG II binding was saturable, reversible, and regulated by G proteins. Transfected receptors were coupled to increases in intracellular calcium and inhibition of cAMP. To determine the polarity of AT1A expression and function in proximal tubules, transfected cells were grown to confluence on membrane inserts under conditions that allowed selective access to AP or BL surfaces. AT1A were expressed on both AP (Kd = 8.7 nM, Bmax = 3.33 pmol/mg protein) and BL (Kd = 10.1 nM, Bmax = 5.50 pmol/mg protein) surfaces. Both AP and BL AT1A receptors underwent agonist-dependent endocytosis (AP receptor: t1/2 = 7.9 min, Ymax = 78.5%; BL receptor: t1/2 = 2.1 min, Ymax = 86.3%). In cells transfected with AT1A, ANG II caused time- and concentration-dependent increases in transepithelial 22Na transport (2-fold over control at 20 min) by increasing Na/H exchange. In conclusion, we have established a stable proximal tubule cell line that expresses AT1A on both AP and BL surfaces, undergoes agonist-dependent receptor endocytosis, and is functional, as evidenced by inhibition of cAMP and increases in cytosolic calcium mobilization and transepithelial sodium movement. This cell line should prove useful for understanding the molecular and biochemical regulation of AT1A expression and function in PTEC.

Functional role for the angiotensin II receptor (AT1A) 3'-untranslated region in determining cellular responses to agonist: evidence for recognition by RNA binding proteins.

We demonstrate a functional role for the 3'-untranslated region (3'-UTR) of the angiotensin II (Ang II) receptor subtype AT1A mRNA in Chinese hamster ovary (CHO-K1) cells by stably transfecting the coding region of the receptor gene with or without the 845 bp 3'-UTR. Two cell lines expressing similar levels of cell-surface receptors (with 3'-UTR, Bmax=571 fmol/mg protein; without 3'-UTR, Bmax=663 fmol/mg protein) were used in the present study. Both cell lines expressed high-affinity receptors (with 3'-UTR, Kd=0.83 nM; without 3'-UTR, Kd=0.82 nM), and binding studies with 125I-labelled Ang II in the presence of GTP[S] demonstrated that both coupled to heterotrimeric G-proteins. Despite these similarities, significant differences were observed for receptor-mediated cell signalling pathways. In cells without the 3'-UTR, Ang II stimulated an increase in cAMP accumulation (11-fold above control) and in cells with the 3'-UTR no stimulation was observed, which was consistent with previous observations in most endogenous Ang II receptor (AT1)-expressing cells. Activation of cAMP by Ang II in cells without the 3'-UTR correlated with an inhibition of DNA synthesis, determined by [3H]thymidine incorporation. Ang II-mediated responses were blocked by EXP3174, a selective non-peptide receptor antagonist. We also observed differences in the transient profiles of intracellular calcium between cells with and without the 3'-UTR in response to Ang II. In cells with the 3'-UTR, a sustained level of intracellular calcium was observed after Ang II stimulation, whereas cells without the 3'-UTR displayed a full return to basal level within 50 s of Ang II treatment. Even though the expressed exogenous gene is under the control of a constitutively expressing promoter (cytomegalovirus promoter), Northern-blot analysis revealed a considerably greater accumulation of AT1A mRNA in cells without the 3'-UTR compared with cells with the 3'-UTR. Analysis of the decay rate of the AT1A mRNA in cells with and without the 3'-UTR revealed that the normally unstable AT1A receptor mRNA became highly stable by removing its 3'-UTR, identifying a role for the 3'-UTR in mRNA destabilization. Interestingly, both cells express similar levels of receptors at the cell surface, suggesting that the 3'-UTR is also involved in the efficient translation and/or translocation of the receptor protein to the plasma membrane. We hypothesized that these 3'-UTR-mediated functions of the receptor are regulated by RNA-binding proteins. To identify possible RNA-binding proteins for the AT1A 3'-UTR, cellular extracts were prepared from parental CHO-K1 cells and 3'-UTR-binding assays, electrophoretic mobility-shift assays and UV crosslinking studies were performed. A major cellular protein of 55 kDa was identified, which specifically interacted with the 3'-UTR. Our data suggest that the 3'-UTR of the AT1A can control specific receptor functions, perhaps via selective recognition of the 3'-UTR by RNA-binding proteins.

Cardiac effects of AII. AT1A receptor signaling, desensitization, and internalization.

Angiotensin II receptors present in cardiomyocytes, nonmyocytes (predominantly fibroblasts), nerve terminals, and the heart vasculature mediate the multiple actions of angiotensin II (AII) in the heart, including modulation of normal and pathophysiological cardiac growth. Although the cellular processes that couple AII receptors (principally the AT1 subtype) to effector responses are not completely understood, recent studies have identified an array of signal transduction pathways activated by AII in cardiac cells. These include: the stimulation of phospholipase C which results in the activation of protein kinase C and the release of calcium from intracellular stores; an enhancement of phosphaditic acid formation; the coupling to soluble tyrosine kinase phosphorylation events; the initiation of the mitogen activated protein kinase (MAPK) cascade; and the induction of the STAT (Signal Transducers and Activators of Transcription) signaling pathway. It is tempting to speculate that these latter responses, which have been previously associated with growth factor signaling pathways, are involved in AII-induced cardiac growth. Interestingly, some of these novel pathways are apparently not under the same strict control imposed upon the more classical signaling pathways. Thus, while AII-induced calcium transients are rapidly (within minutes) desensitized following exposure to AII, the MAP kinase pathway is not, and activation of the STAT pathway requires hours of agonist exposure for maximal induction. These observations support an emerging picture in which the downstream signal transduction pathways of AII receptors are initiated and terminated with a distinct temporal arrangement. This organization allows appropriate rapid responses (e.g. vascular contraction) to transient AII exposure, some of which are rapidly terminated, perhaps for protective reasons, and others not. In contrast, additional responses (e.g. growth) probably require prolonged exposure to agonist.

Molecular mechanisms of angiotensin II (AT1A) receptor endocytosis.

1. Angiotensin II (AngII) initiates a variety of cellular responses through activation of type 1 (AT1; with subtypes AT1A and AT1B) and type 2 (AT2) cell surface angiotensin receptors. Both AT1 and AT2 receptors couple to heterotrimeric guanyl nucleotide binding proteins (G-proteins) and generate intracellular signals following recognition of extracellular AngII, but only AT1 is targeted for the rapid ligand-stimulated endocytosis (internalization) typical of many plasma membrane receptors. 2. AT1 endocytosis proceeds through clathrin-coated pits and is independent of G-protein coupling which predicts that the AngII-AT1 receptor complex attains a conformation necessary for interaction with the endocytotic machinery, but separate from receptor signalling activation. 3. The function of AT1 endocytosis and the reason for the disparity between AT1 and AT2 endocytosis is not fully appreciated, but the latter probably reflects differences in the primary amino acid sequence of these two receptor types. 4. For many receptors that undergo internalization, it has been established that internalization motifs (2-6 amino acids, often incorporating crucial tyrosine and hydrophobic amino acids) within the cytoplasmic regions of the receptor mediate the selective recruitment of activated receptors into clathrin-coated pits and vesicles. 5. Mutagenesis studies on the AT1A receptor, aimed at identifying such motifs, reveal that sites within the third cytoplasmic loop and the cytoplasmic carboxyl terminal region are important for AngII-stimulated AT1A receptor endocytosis.

Evidence against a role for protein kinase C in the regulation of the angiotensin II (AT1A) receptor.

Three putative protein kinase C phosphorylation sites in the carboxyl-terminal region of the angiotensin II AT1A receptor suggest that protein kinase C is involved in the regulation and desensitisation of this receptor. We investigated this possibility by measuring angiotensin II induced Ca2+ transients in cultures of neonatal rat cardiac fibroblasts which express predominantly the angiotensin AT1A receptor. Stimulating or inhibiting protein kinase C activity had no effect on angiotensin II stimulated Ca2+ transients. In addition, in situ and in vitro kinase assays revealed that a peptide, corresponding to the region of the angiotensin AT1A receptor containing the protein kinase C sites, was a poor substrate for protein kinase C. Thus, a heterologous desensitising role for this kinase on angiotensin AT1A receptors in these fibroblasts appears unlikely.

Molecular mechanisms of angiotensin II (AT1a) receptor endocytosis.

1. Angiotensin II (AngII) initiates a variety of cellular responses through activation of type 1 (AT(1); with subtypes AT(1a) and AT(1b) ) and type 2 (AT(2) ) cell surface angiotensin receptors. Both AT(1) and AT(2) receptors couple to heterotrimeric guanyl nucleotide binding proteins (G-proteins) and generate intracellular signals following recognition of extracellular AngII, but only AT(1) is targeted for the rapid ligand-stimulated endocytosis (internalization) typical of many plasma membrane receptors. 2. AT(1) endocytosis proceeds through clathrin-coated pits and is independent of G-protein coupling which predicts that the AngII-AT(1) receptor complex attains a conformation necessary for interaction with the endocytotic machinery, but separate from receptor signalling activation. 3. The function of AT(1) endocytosis and the reason for the disparity between AT(1) and AT(2) endocytosis is not fully appreciated, but the latter probably reflects differences in the primary amino acid sequence of these two receptor types. 4. For many receptors that undergo internalization, it has been established that internalization motifs (2-6 amino acids, often incorporating crucial tyrosine and hydrophobic amino acids) within the cytoplasmic regions of the receptor mediate the selective recruitment of activated receptors into clathrin-coated pits and vesicles. 5. Mutagenesis studies on the AT(1a) receptor, aimed at identifying such motifs, reveal that sites within the third cytoplasmic loop and the cytoplasmic carboxyl terminal region are important for AngII-stimulated AT(1a) receptor endocytosis.

A role for cAMP in angiotensin II mediated inhibition of cell growth in AT1A receptor-transfected CHO-K1 cells.

G-protein coupled Angiotensin II receptors (AT1A), mediate cellular responses through multiple signal transduction pathways. In AT1A receptor-transfected CHO-K1 cells (T3CHO/AT1A), angiotensin II (AII) stimulated a dose-dependent EC50 = 3.3 nM) increase in cAMP accumulation, which was inhibited by the selective AT1, nonpeptide receptor antagonist EXP3174. Activation of protein kinase C, or increasing intracellular Ca2+ with ATP, the calcium ionophore A23187 or ionomycin failed to stimulate cAMP accumulation. Thus, AII-induced cAMP accumulation was not secondary to activation of a protein kinase C- or ca2+/calmodulin-dependent pathway. Since cAMP has an established role in cellular growth responses, we investigated the effect of the AII-mediated increase in cAMP on cell number and [3H]thymidine incorporation in T3CHOA/AT1A cells. AII (1 microM) significantly inhibited cell number (51% at 96 h) and [3H]thymidine incorporation of 68% at 24 h) compared to vehicle controls. These effects were blocked by EXP3174, confirming that these responses were mediated through the AT1 receptor. Forskolin (10 microM) and the cAMP analog dibutyryl-cAMP (1 mM) also inhibited [3H]thymidine incorporation by 55 and 25% respectively. We extended our investigation on the effect of AII-stimulated increases in cAMP, to determine the role for established growth related signaling events, i.e., mitogen-activated protein kinase activity an tyrosine phosphorylation of cellular proteins. AII-stimulated mitogen-activated protein kinase activity and phosphorylation of the 42 and 44 kD forms. These events were unaffected by forskolin stimulated increases in cAMP, thus the AII-stimulated mitogen-activated protein kinase activity was independent of cAMP in these cells. AII also stimulated tyrosine phosphorylation of a number of cellular proteins in T3CHO/AT1A cells, in particular at 127 kD protein. The phosphorylation of the 127 kD protein was transient, reaching a maximum at 1 min, and returning to basal levels within 10 min. The dephosphorylation of this protein was blocked by a selective inhibitor of cAMP dependent protein kinase A, H89-dihydrochloride and preexposure to forskolin prevented the AII-induced transient tyrosine phosphorylation of the 127 kD protein. These data suggest that cAMP, and therefore protein kinase A can contribute to AII-mediated growth inhibition by stimulating the dephosphorylation of substrates that are tyrosine phosphorylated in response to AII.

Angiotensin II receptor endocytosis involves two distinct regions of the cytoplasmic tail. A role for residues on the hydrophobic face of a putative amphipathic helix.

Following agonist stimulation, many receptors are rapidly internalized from the plasma membrane via a mechanism which presumably involves recognition motifs within the cytoplasmic domains of the receptor. We have previously demonstrated (Thomas, W. G., Thekkumkara, T. J., Motel, T. J., and Baker, K. M. (1995) J. Biol. Chem. 270, 207-213) that truncation of the angiotensin II (AT1A) receptor, to remove 45 amino acids from the cytoplasmic tail, markedly reduced agonist stimulated receptor endocytosis. In the present study, we have stably and transiently expressed wild type and carboxyl terminus mutated AT1A receptors in Chinese hamster ovary cells to identify regions and specific amino acids important for this process. Wild type AT1A receptors rapidly internalized (t1/2 = 2.5 min; Ymax = 76.4%) after AII stimulation. Using AT1A receptor mutants, truncated and deleted at the carboxyl terminus, two distinct regions important for internalization were identified: one membrane proximal site between residues 315-329 and another distal to Lys333, within the terminal 26 amino acids. Point mutations (Y302A, Y312A, L316F, Y319A, and K325A) were performed to identify residues contributing to the membrane proximal site. Mutation of Y302A, Y312A, and K325A had little effect on the rate (t1/2 = 4.3, 2.8, and 2.8 min) and maximal amount (Ymax = 81.7, 67.8, and 73.5%) of AII induced internalization. In contrast, L316F and Y319A mutations displayed an approximately 2.5-fold reduction in rate (t1/2 = 6.1 and 6.2 min) and L316F a decreased maximal level (Ymax = 38.1 and 71.4%, respectively) compared to wild type. Interestingly, Leu316 and Tyr319 are closely aligned within the hydrophobic aspect of a putative amphipathic helix, possibly representing an internalization motif for the AT1A receptor. We conclude that the AT1A receptor does not use NPXXY (NPLFY302) motif, first described for the beta 2-adrenergic receptor, to mediate agonist stimulated endocytosis. Rather, two distinct regions of the carboxyl terminus are utilized: one involving hydrophobic and aromatic residues on a putative alpha-helix and another serine/threonine-rich domain.

Activation of the STAT pathway by angiotensin II in T3CHO/AT1A cells. Cross-talk between angiotensin II and interleukin-6 nuclear signaling.

We recently reported that angiotensin II (AII), acting through the STAT (Signal Transducers and Activators of Transcription) pathway, stimulated a delayed SIF (sis-inducing factor)-like DNA binding activity (maximal at 2-3 h) (Bhat, G.J., Thekkumkara, T.J., Thomas, W.G., Conrad, K.M., and Baker, K.M. (1994) J. Biol. Chem. 269, 31443-31449). Using a cell line transfected with the AT1A receptor (T3CHO/AT1A), we further characterized the AII-induced SIF response and explored the possible reasons for the delay in stimulated SIF activity. In cells transfected with a chloramphenicol acetyltransferase reporter plasmid, under the control of a SIE (sis-inducing element), AII markedly stimulated chloramphenicol acetyltransferase activity. The delayed SIF activation by AII was not due to a requirement for the release of other SIF inducing factors into the medium and contrasts with the rapid (5 min) induction elicited by the cytokine, interleukin-6 (IL-6). Interestingly, both agents stimulated tyrosine phosphorylation of Stat92 and predominantly the formation of SIF complex A. We tested the hypothesis that AII initially activated an inhibitory pathway, which was responsible for delaying the maximal SIF stimulation until 2 h. Pretreatment of cells for 15 min with AII resulted in significant inhibition of the IL-6 induced nuclear SIF response (10 min) and Stat92 tyrosine phosphorylation, which was blocked by EXP3174, an AT1 receptor antagonist. This inhibition was transient with return of the IL-6-induced SIF response at 2 h, suggesting that the delayed maximal activation of SIF by AII occurs following an initial transient inhibitory phase. Pretreatment of cells with phorbol 12-myristate 13-acetate for 15 min, to activate protein kinase C, resulted in inhibition of the IL-6-induced SIF response (10 min). However, down-regulation of protein kinase C activity prevented phorbol 12-myristate 13-acetate, but not AII mediated inhibition of the IL-6-induced SIF response. Although the mechanism is not clear, the results presented in this paper raise the interesting possibility that the activation of SIF/Stat92 by AII is characterized by an initial inhibitory phase, followed by the induction process. The observation that AII and IL-6 utilize similar components of the STAT pathway and that AII can cross-talk with IL-6 signaling through inhibition of IL-6-induced SIF/Stat92, implies a modulatory role for AII in cellular responses to cytokines.

Sensitive bioassay for the detection and quantification of angiotensin II in tissue culture medium.

We have developed a sensitive, high-throughput bioassay to quantify angiotensin II in culture medium. Using Chinese hamster ovary cells that stably express a transfected angiotensin II receptor as target cells, we demonstrated that an agonist-stimulated myelin basic protein kinase response can be used as a basis of quantitative bioassay for angiotensin II. The assay permits detection of as little as 10 pg of angiotensin II in 1 mL of medium and offers an excellent alternative to HPLC and radioimmunoassay. This approach may also be applicable for quantification of other peptide hormones or growth factors produced by cell cultures.

Stable expression of a functional rat angiotensin II (AT1A) receptor in CHO-K1 cells: rapid desensitization by angiotensin II.

The octapeptide angiotensin II mediates the physiological actions of the renin-angiotensin system through activation of several angiotensin II receptor subtypes; in particular the AT1. In many tissues, the presence of multiple angiotensin II receptor subtypes, together with a low number of receptors, makes it difficult to study biological responses to physiological concentrations (10(-11)-10(-9) M) of angiotensin II. Also, cultured cells show diminished angiotensin II receptor binding with respect to time in culture and passage number. To address these problems, we expressed the recombinant AT1A receptor in CHO-K1 cells. The stably transfected receptor was characterized using radioligand binding studies and functional coupling to cytosolic free calcium. Radioligand binding of [125I] angiotensin II to the angiotensin II receptor was specific, saturable, reversible and modulated by guanine nucleotides. Like the endogenous AT1A receptor, reported in a variety of tissues, the specific, noncompetitive, nonpeptide AII receptor antagonist, EXP3174, blocked binding of [125I] angiotensin II to the transfected receptor. Scatchard analysis demonstrated that the transfected receptor had a dissociation constant of 1.9 nM with a density of 3.4 pmol/mg protein. An important feature of many of the responses to angiotensin II is the rapid desensitization that occurs following agonist occupancy and the development of tachyphylaxis. In AT1A receptor transfected CHO-K1 cells, angiotensin II (10(-9) M) stimulated a rapid increase in cytosolic free calcium that was completely desensitized within 50 sec following receptor occupancy. Agonist induced desensitization was unaffected when receptor internalization was blocked by pretreatment with concanavalin A or incubation at 4 degrees C, and no changes in AT1A receptor affinity or number were observed. Receptor desensitization was also unaffected by inhibition or activation of protein kinase C. Thus, we have established a permanent, high-level transfectant of the AT1A receptor in CHO-K1 cells and have shown that these receptors rapidly desensitize following exposure to physiological concentrations of agonist. The mechanism of rapid desensitization is not related to receptor sequestration, internalization or controlled by PKC phosphorylation. This provides an excellent model for studying AII actions mediated through a specific receptor subtype, at subnanomolar concentrations.

Overexpression and characterization of human tetrameric pyruvate dehydrogenase and its individual subunits.

Pyruvate dehydrogenase (E1), an alpha 2 beta 2 tetramer, is the first component of the pyruvate dehydrogenase complex which catalyzes a two-step oxidative decarboxylation of pyruvic acid. To overexpress human E1 and its subunits individually, cDNAs for the mature forms of human E1 alpha and E1 beta were subcloned either individually or together into a plasmid pQE-9 and expressed in Escherichia coli M15. A polyhistidine extension was added at the NH2-termini of the recombinant E1 alpha and E1 beta for the rapid purification of the proteins by Ni-nitrilotriacetic-agarose chromatography. The polyhistidine extension on either E1 alpha or E1 beta subunit did not affect the activity of the recombinant tetrameric E1. Highly purified recombinant human E1 catalyzed the partial reactions of the oxidative and nonoxidative conversion of pyruvic acid with the same efficiency as E1 purified from bovine kidney. Recombinant human E1 interacted with thiamin pyrophosphate by forming a charge transfer complex band at 330 nm that changed during the catalytic cycle. Recombinant human E1 was phosphorylated by E1-kinase (with concomitant inactivation) by incorporating nearly three phosphoryl groups per mole of E1. When expressed individually, E1 alpha and E1 beta subunits lacked any catalytic activity in the oxidative or nonoxidative reactions. Spectral studies demonstrated that there was no thiamin pyrophosphate binding to either recombinant E1 alpha or E1 beta subunit. The E1 alpha subunit retained the ability to be phosphorylated; however, the incorporation of phosphoryl groups into recombinant E1 alpha alone was only about 12% of that observed with the tetrameric E1. These findings show that both subunits are required for formation of the active center and catalysis.

Stable expression of a truncated AT1A receptor in CHO-K1 cells. The carboxyl-terminal region directs agonist-induced internalization but not receptor signaling or desensitization.

Phosphorylation of serine and threonine residues in the carboxyl-terminal region of many G-protein-coupled receptors directs the rapid uncoupling from signal transduction pathways. In Chinese hamster ovary cells, we have stably expressed a truncated mutant of the angiotensin II (AT1A) receptor devoid of the carboxyl-terminal 45 amino acids, encompassing 13 serine/threonine residues. One clone, designated TL314 to indicate truncation after leucine 314, expressed a single class of angiotensin II receptors with a dissociation constant of 1.08 nM and a receptor density of 560 fmol/mg of protein (approximately 75,000 receptors/cell). A nonhydrolyzable analog of GTP accelerated the angiotensin II-induced dissociation of [125I]angiotensin II from TL314 plasma membranes 3.6-fold, indicating G-protein coupling. In TL314 cells, angiotensin II stimulated the release of intracellular calcium and the induction of mitogen-activated protein kinase activity, the level of which were comparable with the full-length AT1A receptor. The AII-stimulated calcium response was rapidly desensitized in both full-length and truncated AT1A receptors. Interestingly, angiotensin II-induced endocytosis of the truncated receptor was almost completely inhibited, suggesting that a recognition motif within the carboxyl-terminal 45 amino acids of the AT1A receptor promotes sequestration. Thus, truncation of the AT1A receptor after leucine 314 inhibits agonist-induced internalization without affecting the capacity of the expressed protein to adopt the correct conformation necessary for high affinity binding of angiotensin II, coupling to G-proteins, and activation of signal transduction pathways. The rapid desensitization and refractoriness of the angiotensin II-induced calcium transient in the TL314 cell line, in which putative carboxyl-terminal phosphorylation sites are absent, suggests that the mechanism of AT1A receptor desensitization differs from that of other prototypical G-protein-coupled receptors.

Angiotensin II stimulates sis-inducing factor-like DNA binding activity. Evidence that the AT1A receptor activates transcription factor-Stat91 and/or a related protein.

Recent studies on cytokine and growth factor stimulated signal transduction have defined a direct pathway (Stat91) linking cell surface receptors to target genes in the nucleus. The Stat91 pathway regulated c-fos gene transcription involves activation by tyrosine phosphorylation of the DNA binding factor SIF (sis-inducing factor) in the cytoplasm, its nuclear translocation, and interaction with the regulatory element SIE (sis-inducing element). SIF is a complex of proteins containing members of the STAT family of transcription factors. We determined whether angiotensin II (AII), which acts as a growth factor in many cell types, could activate the Stat91 pathway. We used neonatal rat cardiac fibroblasts expressing G-protein linked AII receptors and CHO-K1 cells expressing stably transfected angiotensin type 1A (AT1A) receptors to address this question. Angiotensin II induced SIF-like activity in both cell types, with initial induction at 15-30 min, maximal around 2-3 h, and undetectable at 6 h. Cytoplasmic and nuclear fractions from cells exposed to AII contained DNA binding activity to SIE. The SIF activity was insensitive to protein synthesis inhibitors and sensitive to the tyrosine kinase inhibitor genistein. Stat91 or a related protein was identified as a component of the AII-induced SIF complex and increased levels of this tyrosine-phosphorylated protein were found in nuclear extracts of cells treated with AII. This is the first evidence that a seven transmembrane, G-protein-coupled receptor, namely AT1A, activates the Stat91-nuclear signaling pathway.

Identification of novel isoforms of the delta subunit of Ca2+/calmodulin-dependent protein kinase II. Differential expression in rat brain and aorta.

Two novel isoforms of the Ca2+/calmodulin-dependent protein kinase II delta subunit were detected in rat aorta. Identification of the subunits was based on two independent lines of evidence, i.e. detection by immunoblotting of differently sized delta subunits and DNA sequence analysis of partial cDNA clones of the kinase. Cytosolic extracts from rat brain, aorta, and cultured aortic cells were analyzed by Western blotting using a delta subunit-specific antipeptide antibody. Aortic extracts demonstrated a single 53-kDa cross-reactive band approximately 7 kDa smaller than the cross-reactive band seen in brain. To ascertain the structural basis for this difference, reverse-transcribed RNAs from rat aorta and brain were analyzed by polymerase chain reaction (PCR), and the PCR fragments were cloned and sequenced. When aortic cDNA was analyzed with a primer pair that spanned the known variable region of the brain kinase subunit, the amplified PCR products were smaller than the major product obtained from brain cDNA. The aortic PCR product was cloned and sequenced and found to represent two novel subunit sequences, designated delta 2 and delta 3 to distinguish them from the previously described delta sequence (now called delta 1) from brain. delta 2 was identical to the predicted delta 1 sequence except for a deletion of 102 base pairs (bp). This deletion corresponded to nearly the entire variable domain. In the sequence of delta 3, this 102-bp region was replaced by a sequence of 33 bp that had 79% nucleotide sequence identity to a portion of the gamma subunit variable domain. A fourth form of the delta subunit (delta 4) was identified in rat skeletal muscle. The delta 4 isoform was characterized by the deletion of a 42-bp sequence identical to the 42 bp at the 3' end of the 102-bp deletion of delta 2. Reverse-transcription PCR analysis of additional rat tissues indicated that alternatively spliced variants of the delta subunit of Ca2+/calmodulin-dependent protein kinase II are expressed in a tissue-specific pattern.

Use of alternative polyadenylation sites for tissue-specific transcription of two angiotensin-converting enzyme mRNAs.

The pulmonary isozyme of rabbit angiotensin-converting enzyme (ACE) is encoded by an mRNA of about 5 kb. cDNA clones corresponding to different parts of this mRNA have been isolated and the complete nucleotide sequences of both the coding and non-coding regions of the mRNA have been determined. The encoded protein has 1309 residues with a 33 amino acids-long signal peptide at the amino terminus and a potential membrane-anchoring domain near the carboxyl terminus. There is a strong sequence homology between two regions of the rabbit cDNA and between the rabbit, human, and mouse cDNAs. Comparison of the nucleotide sequences of the 3' untranslated regions of rabbit pulmonary and testicular ACE cDNAs revealed that the testicular cDNA is nested within the pulmonary cDNA at the 3' end. A rabbit genomic clone encompassing this region was isolated and partially sequenced. It was shown that the gene contains two potential polyadenylation sites 628 bp apart within one exon. Northern analyses with an appropriate oligonucleotide probe confirmed that the proximal polyadenylation site is used exclusively for terminating the testicular mRNA whereas the distal one is used exclusively for the pulmonary mRNA. These results demonstrated that the transcription of the two mRNAs encoding the two ACE isozymes not only initiates at two alternative tissue-specific sites which are 5.7 kb apart but the mRNAs also get polyadenylated at two alternative sites which are 628 bp apart.

The mRNAs encoding the two angiotensin-converting isozymes are transcribed from the same gene by a tissue-specific choice of alternative transcription initiation sites.

The two tissue-specific isozymes of angiotensin-converting enzyme are encoded by two mRNAs. The 5-kilobase pulmonary mRNA (mRNAP) and the 2.5-kilobase testicular mRNA (mRNAT) have identical sequences near their 3'-ends, whereas each has a unique sequence toward its 5'-end. Here we report that the two mRNAs originate from the same gene by initiation of transcription at two alternative sites. We have isolated a rabbit genomic clone which encodes these mRNAs. The sequence organization of the genome is such that the unique sequence of mRNAP precedes that of mRNAT, which in turn precedes the sequence common to both species. For generation of mRNAP, the mRNAT-specific sequence is eliminated as an intron from the primary transcript by splicing. Transcription of mRNAT, on the other hand, initiates within this intron at the middle of the gene. The exact initiation sites of mRNAT and mRNAP have been determined by DNA sequencing; DNA, oligonucleotide, and antisense RNA protection assays; and primer extension assays. Analyses of the sequence upstream of the transcription initiation sites revealed the presence of putative binding sites of several known transcription factors including SP-1, AP-2, and IID.