Actions of parathyroid hormone-related protein as an embryonic differentiation factor.
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to R A Nissenson.
Explore the source record for details and available documents.
Predictive and spectroscopic methods were used to develop a model of the structures of the 1-34 peptides of parathyroid hormone (PTH) and the PTH-related protein (PTHrP). Circular dichroism (CD) studies of bovine PTH-(1-34) and human PTHrP-(1-34)amide in the presence of trifluoroethanol suggest the presence of 24-26 alpha-helical residues. For both peptides, interactions between amino- and carboxyl-region alpha-helices are predicted to result in a hydrophobic core with externally facing hydrophilic residues that include probable determinants of receptor binding and activation. Two such residues, Ser3 and Gln6, are conserved in all known members of the PTH/PTHrP family. We have synthesized 13 novel analogues of bovine PTH-(1-34) monosubstituted at positions 3 and 6 and have determined their biological activities in renal and bone cell radioreceptor and adenylyl cyclase assays. Position 3 analogues displayed biological activity that was reduced in direct proportion to the volume of the substituent side-chain. Position 6 analogues also displayed reduced biological activity, but no simple correlation with side-chain volume or hydrophobicity was evident. The analogues fully displaced labeled PTH from binding sites in renal membranes and bone cells, but [Phe3]bPTH-(1-34), [Tyr3]bPTH-(1-34), [Phe6] bPTH-(1-34), and [Ser6]bPTH-(1-34) were only partial agonists in one or both adenylyl cyclase assays. Of these, [Phe3]bPTH-(1-34) and [Phe6]bPTH-(1-34) were tested for antagonist activity and were found to inhibit the activation of adenylyl cyclase in response to bPTH-(1-34) or hPTHrP-(1-34)amide. These results indicate that positions 3 and 6 contribute important determinants of PTH receptor binding and activation. Modification at these positions represents a novel approach to the development of antagonists of PTH action.
A parathyroid hormone-related protein (PTHrP) is the probable cause of humoral hypercalcemia in malignancy, but its normal physiologic role remains unknown. Since current evidence suggests that PTHrP may have a role in embryonic development, we cloned a genomic fragment that encodes chicken PTHrP (cPTHrP) and studied the expression of PTHrP in developing chick embryos. Blot hybridization of chicken genomic DNA with a cPTHrP genomic DNA probe showed a single band, suggesting that a single-copy gene encodes cPTHrP. By screening a chicken genomic library with the human probe an open reading frame was identified that corresponds to the human PTHrP (hPTHrP) exon IV. Compared to the human sequence the 5' splice junction is highly conserved and the two predicted propeptide residues are identical. The sequence predicts a mature peptide of 139 amino acids; all of the first 21 and 94 of the first 112, but only 8 of the final 27 residues of cPTHrP are identical to the human sequence. The structural features required for PTH receptor binding and activation are highly conserved between chicken and hPTHrP. Poly(A)-enriched RNA from 3-15 day chicken embryos was surveyed by hybridization to the chicken probe. A hybridizing band of 1.45 kb was found in tissues derived from all three germ layers, including brain, heart, lung, liver, gizzard, intestine, chorioallantoic membrane, yolk sac, and skeletal muscle. An additional 1.2 kb hybridizing band was found in some tissues. The conservation of the PTHrP sequence between chicken and mammals supports the view that PTHrP has an important physiologic role.(ABSTRACT TRUNCATED AT 250 WORDS)
Our earlier results indicated that the binding moiety of the renal PTH receptor is an 85 kD protein that is susceptible to proteolytic cleavage to a 70 kD form that supports high-affinity binding and Gs coupling, and to a 50-55 kD form that contains the ligand binding domain but does not couple to Gs. In the present study we used [125I]hPTHrP-(1-34)amide and a chemical cross-linking technique to discern the structural features of the intact 85 kD PTH/PTHrP receptor that are retained in the proteolyzed forms to "structurally map" the receptor. The results of lectin chromatography and endoglycosidase treatment show that the partially proteolyzed receptor forms retain the complex, N-linked glycans present on the intact receptor. This conclusion is further supported by the finding that wheat germ agglutinin was equally effective at competitively inhibiting specific [125I]hPTHrP-(1-34)A binding to the 70 kD form and the intact 85 kD receptor. Specific binding of [125I]hPTHrP-(1-34)A to the intact 85 kD receptor or to the 70 kD form was completely abolished by treatment with disulfide reducing agents, and both partially proteolyzed receptor forms (70- and 50 kD) were shown to retain the small (less than or equal to 14 kD) labeled fragment that is released from the intact receptor by disulfide reduction. Lectin chromatography and endoglycosidase treatment revealed that the less than or equal to 14 kD receptor component is not glycosylated. The less than or equal to 14 kD fragment does not contain a transmembrane spanning region, as its release from the membrane can be affected without detergent solubilization. Identical partial proteolytic maps of the receptor were obtained whether the receptor was covalently labeled with [125I]hPTHrP-(1-34)amide or [125I]bPTH-(1-34). These results suggest a model of the renal PTH/PTHrP receptor binding moiety as a single-chain protein in which the sites of glycosylation, ligand binding, and the functionally critical disulfide bonds are in extracellular domains near one end of the protein and the sites of proteolysis reside near the other end of the protein. These studies also provide further confirmation that PTH and PTHrP bind to a structurally indistinguishable renal receptor and validate the use of PTHrP as a ligand for studies designed to characterize and purify the PTH receptor.
The mechanisms by which PTH and thrombin mobilize intracellular Ca2+ (Cai2+) were examined in UMR 106-H5 rat osteosarcoma cells. Bovine PTH-(1-34) (24 pM to 240 nM) produced a dose-dependent increase in Cai2+ (EC50, 3 nM), which returned to baseline within 75 sec. Human alpha-thrombin produced an increase in Cai2+ (ECmax, 10 U/ml) which was similar to that of PTH with respect to both magnitude and time course. Chelation of extracellular calcium with 5.0 mM EGTA did not alter the Cai2+ response to either PTH or thrombin. When added together at maximally effective concentrations, PTH and thrombin produced additive effects on Cai2+ in the presence and absence of EGTA. The additive effects of PTH and thrombin on Cai2+ were confirmed at the single cell level, using laser-based image analysis. Bradykinin (1 microM) produced a significant increase in Cai2+ in UMR 106-H5 cells which was of lesser magnitude than the peak 2- to 3-fold increase elicited by PTH or thrombin. Preexposure of cells to 10 U/ml thrombin for 2 min abolished the Cai2+ response to bradykinin, whereas preexposure to 240 nM PTH had no effect on the Cai2+ response to bradykinin. Thrombin elicited a rapid increase in the accumulation of 3H-labeled inositol phosphates (IP2 and IP3) in UMR 106-H5 cells, with increases in [3H]1,4,5-IP3 detectable as early as 15 sec after the addition of thrombin. Bradykinin increased [3H]IP production to a lesser extent than thrombin, whereas PTH neither increased [3H]IP accumulation nor potentiated the [3H]IP response to thrombin. The results suggest that thrombin and bradykinin mobilize Cai2+ from a shared IP3-responsive calcium pool, whereas PTH may use signals in addition to 1,4,5-IP3 to mobilize calcium from a distinct cellular calcium pool. Alternatively, specific calcium compartmentalization exists, and there is differential coupling of these agonists to the 1,4,5-IP3/Cai2+ pathway.
The cellular basis for hormonal control of bone resorption is poorly understood. As the identifiable receptors for bone resorbing agents such as parathyroid hormone (PTH) and 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] are located on osteoblasts rather than osteoclasts, the nature of cellular signaling is obscure. Here it is reported that exposure of fetal rat limb bones to pertussis toxin, a bacterial protein that inhibits certain GTP binding proteins (G-proteins) involved in signal transduction, markedly inhibits bone resorption elicited by PTH, 1,25(OH)2D3 and prostaglandin E2. Pertussis toxin does not block the inhibition of alkaline phosphatase activity by PTH or 1,25(OH)2D3, and it potentiates the cyclic AMP response to PTH. These data support the existence of a pertussis toxin-sensitive G-protein that participates in regulation of bone resorption. The putative G-protein is apparently not involved in the initial transduction of hormonal signals, but it may be part of a final common pathway through which the osteoclast is activated by agents with widely divergent initial actions.
Cyclic AMP is known to enhance retinoic acid-induced differentiation of F9 mouse teratocarcinoma cells to parietal endoderm. Recently, we showed that a parathyroid hormone-related protein (PTHrP), by activating adenylate cyclase, can substitute for exogenous cAMP in promoting retinoic acid-induced differentiation of F9 cells. However, the mechanisms by which endogenous cAMP levels are regulated during F9 differentiation are poorly defined. We therefore assessed whether Gs alpha, a subunit of the stimulatory coupling protein of adenylate cyclase, is induced during this process. Treatment of F9 cells with retinoic acid (1 microM) for 5 days resulted in a 20-fold increase in steady-state levels of a 2.0-kilobase Gs alpha mRNA. This was accompanied by an increase in the expression of 52- and 45-kDa Gs alpha polypeptides. Gs alpha mRNA increases within 24 h of exposure to retinoic acid, whereas the expression of alpha 1 (IV) collagen, a marker for F9 differentiation, did not increase until 48 h of treatment. In the presence of retinoic acid, exogenous human PTHrP-(1-34)-amide (20 nM) produced a further 2-fold increase in Gs alpha mRNA. These effects of retinoic acid and PTHrP were exerted largely if not entirely at the level of Gs alpha gene transcription, as assessed by nuclear run-on assay. Bt2cAMP (1 mM) did not reproduce the stimulatory effects of PTHrP on Gs alpha transcription, mRNA, or protein. These data demonstrate that a marked increase in Gs alpha expression accompanies F9 differentiation induced by retinoic acid and PTHrP, and that the regulation is predominantly transcriptional. The resulting increase in adenylate cyclase responsiveness to PTHrP and perhaps other ligands may be a critical component of the differentiation process. The effect of PTHrP on the expression of Gs alpha appears to be mediated by signals other than cAMP.
Administration of excessive amounts of parathyroid hormone (PTH) in the treatment of osteoporosis can reverse the beneficial effects of a low-dose, intermittent regime. To investigate the direct actions and the possible cellular mechanisms of PTH in inducing desensitization of PTH receptors, we studied the effects of desensitization on rat osteoblastic UMR-106 cells. When the osteoblasts were preincubated with bPTH-(1-34), complete refractoriness to a subsequent challenge with the hormone developed within 1 h and at hormone concentrations as low as 5 nM. When osteoblasts thus desensitized were incubated in hormone-free medium, recovery of the cAMP responses began within 2 h and reached maximum after 16 h. Cycloheximide did not affect the process of desensitization. [Nle8,Nle18,Tyr34]bPTH-(3-34)amide significantly impaired the desensitization process by PTH-(1-34) but did not have stimulatory effect on cAMP responses. No significant heterologous desensitization was obvious after preincubation with isoprenaline (50 microM), prostaglandin E1 (50 microM), or prostaglandin E2 (50 microM) for 2 h. Binding experiments with [125I]PLP-(1-36)amide after desensitization revealed that there was an approximate twofold decrease in receptor affinities as analyzed by Scatchard analysis, showing that the decrease in affinity was prominent in the process of desensitization. When the cells were treated with monensin during desensitization, PTH challenge after desensitization produced significantly lower cyclic AMP responses. Recovery after desensitization occurred over a period of 16 h. Inclusion of monensin, but not cycloheximide, impaired the recovery. The results show that homologous desensitization of rat osteoblasts to PTH is brought about by the occupancy of receptors by PTH-(1-34) but not by cAMP generation itself.(ABSTRACT TRUNCATED AT 250 WORDS)
A parathyroid hormone-like peptide that probably causes hypercalcemia associated with solid tumors was recently characterized. It is a potent hypercalcemic and hypophosphatemic factor whose production is strongly associated with hypercalcemia and whose properties account for most aspects of the clinical syndrome. Diagnostic tests for this peptide have been developed. The parathyroid hormone-like peptide as well as other cytokines, such as tumor necrosis factor-alpha, likely play a role in causing hypercalcemia in multiple myeloma and lymphomas.
Human alpha-thrombin is known to elicit bone resorption in vitro and has been proposed as a mediator of increased bone turnover in inflammatory diseases. We used UMR 106-H5 rat osteoblast-like osteosarcoma cells to explore the signal transduction mechanism utilized by thrombin in bone. Thrombin produced a dose-dependent increase in the accumulation of [3H]inositol phosphates (IPs) in UMR 106-H5 cells prelabeled with [3H]myo-inositol (EC50 15 U/ml). In saponin-permeabilized cells, GTP gamma S increased [3H]IP production, whereas GDP beta S inhibited the response to both GTP gamma S and thrombin, indicating involvement of a G-protein in thrombin action. Thrombin produced a dose-dependent increase in intracellular free calcium (Cai2+) in UMR 106-H5 cells (EC50 1 U/ml; maximal increase 4-fold), as well as a small (20%) increase in [3H]thymidine incorporation. Treatment of UMR 106-H5 membranes with pertussis toxin (PT) and [32P]NAD+ resulted in labeling of a 40-kDa protein. However, pretreatment of cells with a dose of PT sufficient to produce maximal endogenous labeling of this protein failed to influence thrombin action on IP accumulation, Cai2+, or [3H]thymidine incorporation. In contrast, PT treatment of CCL39 hamster lung fibroblasts significantly blunted thrombin-stimulated [3H]IP accumulation and [3H]thymidine incorporation. These results suggest that thrombin raises Cai2+ in UMR 106-H5 cells by activating polyphosphoinositide-specific phospholipase C. Whereas in fibroblasts and platelets, thrombin receptors appear to couple to both PT-sensitive and PT-insensitive G-proteins, only a PT-insensitive G-protein appears to mediate thrombin action in UMR 106-H5 cells. Either these cells lack the relevant PT-sensitive G-protein or they possess thrombin receptors that selectively couple to a pertussis toxin-insensitive G-protein.
Differentiation of mouse embryonal carcinoma cells to the parietal endoderm phenotype is associated with expression of PTH-responsive adenylate cyclase. A PTH-like protein (PLP), which binds to PTH receptors and activates adenylate cyclase in classical PTH target cells was recently isolated and cloned. We assessed whether the parietal endoderm phenotype is associated with the expression of PLP or its receptor. A 1.4-kilobase PLP transcript was detected in the mouse parietal endoderm cell line PYS-2. No hybridizing transcripts were evident in undifferentiated mouse embryonal carcinoma cells PSA-1 or F9. However, differentiation of these cells to parietal endoderm, either spontaneously (PSA-1) or by treatment with retinoic acid and dibutyryl cAMP (F9), resulted in expression of the 1.4-kilobase PLP message. Undifferentiated F9 cells displayed negligible specific binding of [125I]PLP-(1-34)amide. When F9 cells were induced to differentiate to parietal endoderm, specific binding sites for [125I]PLP-(1-34)amide were expressed in parallel with PLP-responsive adenylate cyclase. These receptors, like those in classical PTH target tissues, displayed identical affinity (Kd = 5.2 nM) for bPTH-(1-34) and hPLP-(1-34)amide; with binding capacity (Bmax) of 6.6 x 10(4) sites/cell. In the presence of retinoic acid, exogenous PLP substituted for dibutyryl cAMP in a concentration-dependent fashion in promoting the differentiation of F9 cells to parietal endoderm. Thus, both PLP mRNA and PLP receptors coupled to adenylate cyclase are expressed during the differentiation of mouse embryonal carcinoma cells. Increased cAMP levels produced by autocrine stimulation of PLP receptors by PLP may contribute to differentiation of embryonal carcinoma cells into parietal endoderm.
The pathogenesis of hypercalcemia in malignancy has been enigmatic until recent years. Since the realization in 1980 that bioassays for parathyroid hormone detected a cross-reacting substance in malignancy, progress has been remarkably rapid. A parathyroid hormone-related protein was purified and identified by molecular cloning as a 141-amino acid peptide with limited homology to parathyroid hormone itself. Nonetheless, both peptides activate the parathyroid hormone receptor to produce hypercalcemia. It is now clear that the parathyroid hormone-related protein is the cause of hypercalcemia in most solid tumors, particularly squamous and renal carcinomas. New assays for the hormone as well as the related peptide have greatly simplified the differential diagnosis of hypercalcemia. At the same time, new agents for the treatment of hypercalcemia are becoming available, most notably the bisphosphonate drugs.
STUDY OBJECTIVE: To measure the serum levels of a newly described parathyroid hormone-like protein (PLP) which was isolated from malignant tumors associated with hypercalcemia, and determine whether PLP is a humoral factor in malignancy-associated hypercalcemia. DESIGN: A cross-sectional study of serum levels of PLP using a newly developed radioimmunoassay. SETTING: A university-affiliated Veterans Administration hospital in San Francisco, California, a University hospital in Hong Kong, and a private hospital in Danville, Pennsylvania. PATIENTS: Patients with hypercalcemia (calcium greater than 2.65 mmol/L) and a diagnosis of malignancy were studied. Control groups included normocalcemic patients with malignancy, patients with hyperparathyroidism, and normal subjects. MEASUREMENTS AND MAIN RESULTS: Serum immunoreactive PLP (iPLP) levels in normal subjects were less than 2.5 pmol eq/L (10 pg/mL), and 68% of subjects had undetectable levels. The serum concentration of iPLP was normal in 15 of 16 hypercalcemic patients with hyperparathyroidism. Serum iPLP was increased (greater than 2.5 pmol eq/L) in 36 of 65 (55%) patients with malignancy-associated hypercalcemia, with a mean value of 6.1 +/- 0.9 pmol eq/L (24 pg/mL). In a subgroup of patients with solid tumors serum iPLP was increased in 30 (71%) of 42 hypercalcemic patients, with a mean value of 6.5 +/- 0.9 pmol eq/L. Serum iPLP was elevated in only 3 of 23 normocalcemic patients with cancer. In patients with solid malignancies (n = 59), levels of iPLP were positively correlated with the total serum calcium (r = 0.43, P less than 0.01). CONCLUSION: The data indicate a relation between the serum concentration of iPLP and the presence of hypercalcemia in solid malignancies. The results support a role for PLP as a humoral mediator of hypercalcemia in most patients with solid tumors. Measurement of iPLP should be useful in the differential diagnosis of hypercalcemia.
Recent evidence suggests that guanyl nucleotide binding (G) proteins are involved in receptor-mediated bone resorption and in osteoblastic function, but the nature of the G protein coupled to effectors that are involved in these skeletal effects is unknown. The purposes of this study were to determine (1) whether a G protein mediates activation of phosphoinositide-specific phospholipase C in UMR-106 rat osteosarcoma cells, and (2) whether parathyroid hormone (PTH) and a PTH-like protein (PLP) associated with humoral hypercalcemia of malignancy promote GTP-dependent PIP2 hydrolysis. Addition of GTP (10(-4) M) or guanosine 5'-0-(3-thiotriphosphate, GTP gamma S, 10(-5) M) to membranes prepared from UMR-106 cells labeled with [3H]myo-inositol increased both [3H]inositol trisphosphate (IP3) and [3H]inositol bisphosphate (IP2) formation. The increases in [3H]IP2 and [3H]IP3 produced by GTP were 8.6- and 4.3-fold, respectively. GTP gamma S produced a 17.6- and 11.9-fold increase in [3H]IP2 and [3H]IP3, respectively. The stimulatory effects of GTP and GTP gamma S were dose dependent (GTP ED50 = 3.9 x 10(-6) M; GTP gamma S ED50 = 2.5 x 10(-7) M) and progressive over 10 minutes and required the presence of Mg2+.GTP (10(-4) M) and GTP gamma S (10(-5) M) decreased membrane [3H]phosphoinositides concomitantly with increased [3H]IP2 and [3H]IP3. The GDP analog guanosine 5'-O-(2-thiodiphosphate, GDP beta S) alone did not alter [3H]IP2 or [3H]IP3 production but at 10(-4) M blocks the stimulatory effects of GTP and GTP gamma S. NaF (3 x 10(-2)M) produced a 2.8- and 2.0-fold stimulation of [3H]IP2 and [3H]IP3, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)
Expression of a parathyroid hormone-like protein (PLP), which is associated with hypercalcemia in malignancy, has recently been localized to normal lactating mammary tissue. We examined the possibility of an extramammary role of PLP by measuring its levels in serum and milk of lactating women. The levels of PLP by radioimmunoassay in serum of lactating and nonlactating women were indistinguishable [4.2 +/- 1.8 and 3.6 +/- 1.0 pg equivalents (eq) of PLP-(1-34) amide per ml, respectively]. As PLP was undetectable in some serum samples, this result does not exclude the possibility that lactation results in a small increase in serum levels of PLP. In contrast, high concentrations of immunoreactive PLP [40,000-75,000 pg eq of PLP-(1-34) amide per ml] and correspondingly high concentrations of bioactive PLP were found in human, rat, and bovine milk. A variety of processed bovine milk products had a PLP content similar to that of fresh bovine milk, whereas infant formulas had lower concentrations, ranging down to undetectable. Although the physiological role of PLP in lactation is unknown, the data establish the presence of PLP in milk and suggest the possibility that PLP may be important in neonatal calcium homeostasis.
A tumor-derived protein with a spectrum of biologic activities remarkably similar to that of parathyroid hormone (PTH) has recently been purified and its sequence deduced from cloned cDNA. This PTH-like protein (PLP) has substantial sequence homology with PTH only in the amino-terminal 1-13 region and shows little similarity to other regions of PTH thought to be important for binding to receptors. In the present study, we compared the actions of two synthetic PLP peptides, PLP-(1-34)amide and [Tyr36]PLP-(1-36)amide, with those of bovine parathyroid hormone (bPTH)-(1-34) on receptors and adenylate cyclase in bone cells and in renal membranes. Synthetic PLP peptides were potent activators of adenylate cyclase in canine renal membranes (EC50 = 3.0 nM) and in UMR-106 osteosarcoma cells (EC50 = 0.05 nM). Bovine PTH-(1-34) was 6-fold more potent than the PLP peptides in renal membranes, but was 2-fold less potent in UMR-106 cells. A competitive PTH receptor antagonist, [Tyr34]bPTH-(7-34)amide, rapidly and fully inhibited adenylate cyclase stimulation by the PLP peptides as well as bPTH-(1-34). Competitive binding experiments with 125I-labeled PLP peptides revealed the presence of high affinity PLP receptors in UMR-106 cells IC50 = 3-4 nM) and in renal membranes (IC50 = 0.3 nM). There was no evidence of heterogeneity of PLP receptors. Bovine PTH-(1-34) was equipotent with the PLP peptides in binding to PLP receptors. Likewise, PLP peptides and bPTH-(1-34) were equipotent in competing with 125I-bPTH-(1-34) for binding to PTH receptors in renal membranes. Photoaffinity cross-linking experiments revealed that PTH and PLP peptides both interact with a major 85-kDa and minor 55- and 130-kDa components of canine renal membranes. We conclude that PTH and PLP activate adenylate cyclase by binding to common receptors in bone and kidney. The results further imply that subtle differences exist between PTH and PLP peptides in their ability to induce receptor-adenylate cyclase coupling.
Previous studies have established the presence of parathyroid hormone (PTH)-sensitive adenylate cyclase activity in cultured human skin fibroblasts. The present study was undertaken to identify and quantitate PTH receptors directly in such cells. Human dermal fibroblast cell line CRL 1564 was found to possess specific binding sites for [125I]PTH(1-34). These sites bound PTH selectively; bovine and human PTH(1-34) and PTH(1-84) competed for [125I]PTH(1-34) binding sites, whereas the unrelated peptides calcitonin, insulin, AVP, angiotensin II, and ACTH(1-24) were inactive even at micromolar concentrations. Competitive binding experiments demonstrated the presence of binding site heterogeneity. These data fit a "two-site" model (p less than 0.001) in which one binding component has high affinity (Kd = 2.5 ng/ml = 0.6 nM) and low capacity (10(4) sites/cell) while the other has low affinity (Kd = 5.9 micrograms/ml = 1.5 microM) and high capacity (greater than 10(7) sites/cell). Similar high- and low-affinity [125I]bPTH(1-34) binding sites were seen also in CRL 1564 membranes containing a PTH-responsive adenylate cyclase. The Kd of the high-affinity sites was identical to the concentration of unlabeled bPTH(1-34) (4.2 ng/ml = 1.0 nM) required to half-maximally elevated cyclic AMP in CRL 1564 cells. Affinity labeling of specific PTH binding sites revealed the presence of multiple components with Mrs of 85, 70, 40, 33, and 23 kD on SDS-PAGE. Competition experiments did not disclose structurally discrete high- and low-affinity sites. Thus, structurally homologous PTH receptors in human skin fibroblasts apparently can assume two affinity states: (i) a high-affinity state coupled to adenylate cyclase and (ii) a low-affinity state that may represent uncoupled receptors.
A peptide secreted by tumors associated with the clinical syndrome of humoral hypercalcemia of malignancy was recently purified from human renal carcinoma cell line 786-0. The N-terminal amino acid sequence of this peptide has considerable similarity with those of parathyroid hormone (PTH) and of peptides isolated from human breast and lung carcinoma (cell line BEN). In this study we obtained the nucleotide sequence of a 1595-base cDNA complementary to mRNA encoding the PTH-like peptide produced by 786-0 cells. The cDNA contains an open reading frame encoding a leader sequence of 36 amino acids and a 139-residue peptide, in which 8 of the first 13 residues are identical to the N terminus of PTH. Through the first 828 bases the sequence of this cDNA is identical with one recently isolated from a BEN cell cDNA library; however, beginning with base 829 the sequences diverge, shortening the open reading frame by 2 amino acids. Differential RNA blot analysis revealed that 786-0 cells express two major PTH-like peptide mRNAs with different 3' untranslated sequences, one of which hybridizes with the presently described sequence and the other one with that reported for the BEN cell PTH-like peptide cDNA. Primer-extension analysis of 786-0 poly(A)+ RNA together with Southern blot analysis of human DNA confirmed the presence of a single-copy gene coding for multiple mRNAs through alternate splicing. In addition, the 3' untranslated sequence of the cDNA described here has significant similarity to the c-myc protooncogene.