Thyrocalcitonin: stimulation of secretion by pentagastrin.

Administration of a small dose of pentagastrin, a synthetic pentapeptide containing the biologically active portion of the native hormone gastrin, results in a marked, rapid, transitory increase in thyrocalcitonin secretion in the pig. Gastrin or a related gastrointestinal peptide may be important in the physiological secretion of thyrocalcitonin, such as that which occurs when calcium salts are introduced into the gastrointestinal tract.

Nonthrombogenic plastic surfaces.

Heparin has been chemically combined with a number of plastic surfaces rendering them nonthrombogenic as judged by Lee-White coagulation tests in vitro with human blood. Addition of quaternary ammonium groups to the plastic permitted formation of insoluble complexes with heparin. These heparinized surfaces were essentially nonthrombogenic and adsorb blood proteins to a significantly smaller degree from dilute solution than do the unmodified plastic surfaces. The affinity of the formed blood elements for these modified surfaces is much less than for the unmodified surfaces.

Calcitonin inhibition of insulin release from isolated rat pancreatic islets.

Calcitonin is known to inhibit secretion of gastrin and insulin in vivo. The objective of this study was to determine whether calcitonin can act directly on pancreatic islets in vitro to inhibit insulin release. Isolated islets were obtained from collagenase-treated rat pancreas, and three peptides (gastrin-releasing peptide, cholecystokinin-8, bombesin) and glucose were used to stimulate insulin release. All agents caused a significant increase in insulin secretion and calcitonin inhibited these responses, but had no consistent effect on basal release. This study provides evidence that calcitonin is an effective inhibitor of insulin secretion and acts directly on islet tissue.

Down-regulation of calcitonin gene transcription by vitamin D requires two widely separated enhancer sequences.

Transcription of the calcitonin (CT) gene is down-regulated by vitamin D in normal and transformed thyroid C cells. DNA transfer techniques have been previously used to map and characterize a cAMP-induced enhancer at nucleotides -255 to -129 and an enhancer of basal transcription at -1060 to -905 in the CT 5' flanking DNA. The same methods were used to identify a negative response element for vitamin D. Deletion mutants of a genomic fragment of CT extending from nucleotides -1460 to +90 were attached to a promoterless GH gene and transfected individually into the medullary thyroid carcinoma cell line TT. CT nucleotides -1460 to -129 induced significant basal transcription of the GH reporter gene in TT cells. Basal transcription was elevated 3-fold to 4-fold by treatment with cAMP analog. The biologically active metabolite of vitamin D3, 1,25-dihydroxyvitamin D3, had a minor (20%) inhibitory effect on basal transcription but inhibited more than 60% of the cAMP-induced transcription. We further investigated the cAMP-induced response and found that transcriptional activity of the downstream cAMP-induced enhancer was greatly synergized in the presence of the upstream enhancer of basal transcription. The latter enhancer contained three functional CANNTG sequences designated E1 (nucleotides -1060 to -1030), E2 (nucleotides -940 to -920), and E3 (nucleotides -920 to -900). E2 and E3 were essential for maximal cAMP-induced transcription. Detailed mapping of the vitamin D response showed that a minimum requirement for inhibition of the cAMP-induced enhancer by vitamin D was a sequence overlapping E3 (nucleotides -920 to -829). We conclude that a negative response element to vitamin D is located between nucleotides -920 and -829 in the CT 5' flanking DNA. It is possible that vitamin D inhibits transcription by interfering with the synergistic interaction between the cAMP-induced enhancer and the enhancer of basal transcription.

Cosecretion of calcitonin and calcitonin gene-related peptide from cultured rat medullary thyroid C cells.

Whether C cells cosecrete calcitonin (CT) and CGRP was examined by exposing cultured rat medullary thyroid carcinoma 6-23 cells for 2 h to high medium Ca and to agents with a potential for affecting Ca-dependent secretion. In every experiment exposure of cells to high medium Ca (2.0-2.5 mM) provoked an increased release of both peptides that was highly correlated (r = 0.73). With other test substances, also, changes in both hormones occurred in parallel. The Ca-channel activator, BAY-K-8644 (10 microM) increased secretion, and this was inhibited by the Ca channel blocker, nitrendipine (10 microM). The Ca2+ ionophore, ionomycin (5 microM), increased release, and the calmodulin-Ca channel inhibitor, phenytoin (100 microM), inhibited Ca-induced release. The active 4 beta isomer of phorbol-12,13-didecanoate (0.1 microM), but not the inactive 4 alpha isomer, increased secretion. The findings suggest that pathways mediating C cell secretion include plasma membrane Ca channels, intracellular [Ca2+], calmodulin, and protein kinase C. The results show that the secretory process in rat C cells involves the release of CGRP as well as CT.

Binding of calcitonin and calcitonin gene-related peptide to calvarial cells and renal cortical membranes.

Binding of calcitonin (CT) and calcitonin gene-related peptide (CGRP) to rat hemicalvariae and renal membranes was examined in an effort to determine whether CT and CGRP interact with the same bone cell binding site, and to see whether the binding pattern was similar for bone and renal cortex. Specific binding of 125I-salmon CT to rat calvariae was inhibited by unlabeled salmon, porcine, or human CT, but not by rat CGRP. Binding of 125I-rat CGRP to calvariae was inhibited by CGRP and high doses of salmon CT, but not by human or porcine CT. Binding of 125I-salmon CT to renal membranes was inhibited by unlabeled salmon CT or rat CGRP, but no specific binding of 125I-rat CGRP could be detected. The results suggest that separate bone cell receptors for CT and CGRP exist and that CGRP can interact with renal receptors for CT.

Relaxation of rat gastrointestinal smooth muscle by parathyroid hormone.

In this study we investigated whether parathyroid hormone (PTH) can produce relaxation of gastrointestinal (GI) smooth muscle as it has been reported to do for vascular and uterine smooth muscle. Muscle tissue preparations from rat stomach, duodenum, ileum, or colon were mounted in a 37 degrees C tissue bath and perfused with oxygenated medium. Changes in isometric tension were recorded with a force-displacement transducer connected to a polygraph. Decreases in either resting tension or agonist-induced tension (0.5-1.0 microM acetylcholine or carbachol) were observed within 1-2 min of PTH addition and were reversible upon removal of the peptide. All GI regions tested were responsive to PTH. Synthetic rPTH-(1-34) (0.1-100 nM) produced a dose-dependent relaxation of both fundic (ED50 = 5.2 nM) and colonic (ED50 = 2.5 nM) muscle strips. At 100 nM, a 90% decrease in fundic tension and a 70% decrease in colonic tension were seen. At 100 nM, bPTH-(1-34), but not bPTH-(7-34) or rat calcitonin gene-related peptide, also was effective in relaxing fundic or colonic muscle. Similarly, in the fundic muscle, 500 nM bPTH-(3-34) alone was ineffective, but it inhibited the effect of 5 nM rPTH-(1-34) when both peptides were tested in combination. Likewise, 100 nM bPTH-(3-34) also inhibited the relaxation induced by 5, 10, or 100 nM bPTH-(1-34). The results show PTH to be highly effective in nanomolar concentrations in causing relaxation of GI smooth muscle.(ABSTRACT TRUNCATED AT 250 WORDS)

Inhibition by phenytoin of in vitro secretion of calcitonin from rat thyroid glands and cultured rat C cells.

Baby rat thyroid glands and cultured rat medullary carcinoma C cells were incubated acutely with phenytoin (38-100 microM), and the calcitonin (CT) secreted into the serum-free medium was measured by radioimmunoassay (RIA). Phenytoin did not alter CT release from glands or C cells incubated in 1 mM Ca, but, when Ca was raised to 1.75 or 2.5 mM, a marked inhibitory effect of phenytoin was apparent. The inhibitory effect could be negated by including 10 microM BAY-K-8644 in the medium. Inhibitory effects on CT release also were obtained with 100 microM trifluoperazine or 100 microM nitrendipine, and these inhibitory effects also were counteracted by 10 microM BAY-K-8644. The results show that clinically relevant amounts of phenytoin can inhibit CT release, perhaps by interfering with C-cell Ca channels or by inhibiting calmodulin-dependent processes.

Effect of transforming growth factor beta on parathyroid hormone receptor binding and cAMP formation in rat osteosarcoma cells.

Transforming growth factor beta (TGF-beta) is now recognized as an important growth regulator and modulator in bone, where it apparently acts in an autocrine or paracrine fashion. In an effort to help elucidate how TGF-beta may interact with parathyroid hormone (PTH) to influence bone turnover, we examined the idea that TGF-beta might alter the number or affinity of PTH receptors in osteoblastic bone cells, PTH receptor binding was assessed in cultured ROS 17/2.8 cells using [125I]PTHrP-(1-34) as labeled ligand. Specific binding to intact cells was measured in the presence of up to 1 microM unlabeled rPTH-(1-34), and cAMP in cell extracts was determined by RIA. Incubation of ROS cells with 2 ng/ml of TGF-beta for the maximally effective time of 3 days increased the number of PTH binding sites (Bmax) by 47 +/- 13%, with no change in the KD (3 nM). TGF-beta also increased the intracellular cAMP response to 0.3 nM rPTH-(1-34) (ED50) by 53 +/- 22%. Both effects were dose dependent, with 1-4 ng/ml of TGF-beta producing maximal effects, and both effects were blocked by the protein synthesis inhibitor cycloheximide (2-5 microM). Since TGF-beta induced comparable increases in both PTH binding and cAMP formation, the findings suggest that TGF-beta can increase the number of functional PTH receptors in cultured ROS 17/2.8 cells. This effect may reflect an action of TGF-beta to slow replication and promote differentiated functions in these cells.

Radioiodinated rat parathyroid hormone-(1-34) binds to its receptor on rat osteosarcoma cells in a manner consistent with two classes of binding sites.

Binding of 125I-labeled rat (r) PTH-(1-34) to ROS 17/2.8 osteoblastic bone cells and to membranes from these cells was examined. Competitive binding inhibition experiments were performed using unlabeled rPTH-(1-34) with particular emphasis on concentrations of peptide below 1 nM. In intact cells, binding of labeled rPTH-(1-34) was highly specific, and inhibition of binding by unlabeled ligand suggested the presence of two classes of binding sites, one with high affinity and low capacity (KD = 40 pM, approximately 20% of total binding sites) and the other with lower affinity and high capacity (KD = 2 nM, approximately 80% of the sites). Membranes prepared from ROS cells also exhibited a pattern of binding from competitive inhibition curves consistent with two distinct binding sites (KD = 30 pM and 6 nM). In intact ROS cells, cellular cAMP levels increased over the range of 10(-11)-10(-9) M rPTH-(1-34) with an ED50 intermediate between the two KD values (0.25 nM). These data suggest that osteoblastic bone cells possess two distinct classes of membrane receptors for PTH. Since the KD of the higher affinity site more closely approximates circulating concentrations of PTH, binding to this site may have physiologic relevance.

Identification and characterization of parathyroid hormone receptors in rat renal cortical plasma membranes using radioligand binding.

Parathyroid hormone (PTH) receptors have been described in renal tissue from several species, but not in the rat. In this study, radioligand binding techniques were used to identify and characterize PTH receptors in rat kidney cortical membranes. The sulfur-free PTH analog [Nle8,18Tyr34]bovine PTH-(1-34)amide was iodinated using the iodogen method. This ligand was suitable for use in identifying PTH receptors in canine renal membranes, but not rat renal membranes. Synthetic, unsubstituted rat PTH-(1-34) was iodinated using the milder, lactoperoxidase technique and was purified by HPLC on a C8 column. [125I]rat PTH-(1-34) bound rapidly to both rat and dog renal membranes. At 22 degrees C reaction reached steady state within 20 minutes, and this level was maintained for at least 3 h. Specific binding was routinely greater than 90% for rat kidney and greater than 95% for dog kidney. Similar results were obtained at 4 degrees C with a longer time required to attain steady state (approximately 45 minutes). Binding was reversible as demonstrated by dissociation of bound ligand after either infinite dilution or displacement with excess nonradioactive PTH. Binding was saturable and of high affinity (rat kidney: Bmax = 2.3 pmol/mg protein, Kd = 3.1 nM, dog kidney: Bmax = 2.1 pmol/mg protein, Kd = 3.7 nM). Rat renal cortical adenylate cyclase activity was stimulated by rat PTH in a dose-dependent manner with an EC50 of 4 nM, a value in good agreement with the binding data. This study demonstrates the feasibility of identifying and characterizing parathyroid hormone receptors in rat renal cortical plasma membranes using radioligand binding techniques.

Parathyroid hormone-related protein relaxes rat gastric smooth muscle and shows cross-desensitization with parathyroid hormone.

Previously, we reported that parathyroid hormone (PTH) is a potent and effective relaxant of rat gastrointestinal smooth muscle. Since the recently discovered PTH-related protein (PTHrP) has amino-terminal homology with PTH and acts like PTH on bone and kidney, we decided to study the effects of synthetic PTHrP analogs on the isometric tension of rat fundic strips. Rat (r) PTH-(1-34), human (h) PTHrP-(1-34), and [Tyr0]hPTHrP-(1-34) relaxed acetylcholine-stimulated fundic strips in a dose-dependent manner, with an IC50 of 6, 10, and 31 nM, respectively. However, maximal doses of [Tyr0]hPTHrP-(1-34) were considerably less effective than the other two peptides. Addition of rPTH-(1-34) or hPTHrP-(1-34) to a maximally effective dose of [Tyr0]hPTHrP-(1-34) produced no further relaxation, indicating that [Tyr0]hPTHrP-(1-34) also has antagonistic properties. Bovine PTH-(3-34) an established in vitro antagonist of PTH, partially inhibited the relaxant effect of PTHrP. Fundic strips that had been desensitized by preincubation for 30-45 minutes with either rPTH-(1-34) or hPTHrP-(1-34) (330-500 nM) were also insensitive to the relaxant action of either peptide, but in the same preparations, the relaxation produced by vasoactive intestinal peptide was unaffected. These studies indicate that PTHrP and PTH can interact with the same receptor. Whether PTHrP influences gastrointestinal motility in normal or tumor-bearing persons remains to be investigated.

Bone disease in burn patients.

Burn patients are at risk for bone disease due to aluminum (Al) exposure from use of antacids and albumin, partial immobilization, and increased production of endogenous glucocorticoids. Moreover, severely burned children are growth impaired up to 3 years after the burn. To determine the extent of bone disease, we studied nine men and three women, ages 18-41 years, with greater than 50% body surface area burn. Seven patients underwent iliac crest bone biopsy following double tetracycline labeling, one additional patient expired after a single label, and three others had postmortem specimens obtained for quantitative Al only. Serial serum and urine samples were obtained weekly until biopsy or death. All biopsied patients had reduced bone formation and osteoid area, surface, and width, with mineral apposition rate, osteoblast surface, and osteoclast number with normal eroded surfaces compared to age- and sex-matched normal ambulatory volunteers. Burn patients also had reduced bone formation, mineral apposition rate, osteoid area, and surface compared to age-matched volunteers at short-term bed rest. Serum levels of osteocalcin were low. Most patients had mild hypercalcemia but only a third had hypercalciuria. All patients had elevated Al in blood or urine; urine Al correlated inversely with serum osteocalcin. In 60% significant bone Al was detectable by stain or quantitation. Our data are compatible with burn patients having markedly reduced bone turnover. Al loading, partial immobilization, endogenous corticosteroids, and cytokine production may be among the etiologic factors.

Identification of C-cells in normal and goitrous rat thyroid tissues using antiserum to rat thyrocalcitonin and the immunoperoxidase bridge technique.

Application of the immunoperoxidase bridge technique to the light microscopic localization of C-cells in rat thyroid tissue is described. Guinea pig antisera to rat thyrocalcitonin (TCT) were produced by the injection of highly purified rat TCT (100-300 MRC U/mg) emulsified in complete Freund's adjuvant. A 1:1000 dilution of the antiserum used in this study gave a strong positive reaction with rat C-cells, and 1 ml of undiluted antiserum provided sufficient material for staining approximately 5000 slides. The substitution of nonimmune guinea pig serum for the anti-rat TCT serum or the prior absorption of anti-rat TCT serum with increasing amounts of highly purified rat TCT both eliminated the staining of thyroid C-cells. Likewise, no staining was observed in tissue sections from rat parathyroid, ovary, pituitary gland, and skeletal muscle. Antiserum to synthetic human TCT also could be used to identify rat thyroid C-cells. The method revealed abundant C-cells in goiters from rats fed a low-iodine diet for more than 1 year. This finding was supported by electron microscopic evaluation of goitrous tissue and by the detection, by radioimmunoassay, of TCT in thyroid tissue and in peripheral blood from goitrous rats.

Acute and chronic fluctuations of immunoreactive and biologically active plasma calcitonin in the rat.

Using a sensitive calcitonin (CT) immunoassay and a newly developed bioassay capable of detecting 0.025 MRC mU CT, we have studied acute and chronic plasma CT fluctuations in male and female rats. Immunoassay of serial plasma samples revealed progressive increases in plasma CT concentrations during aging; female rats have higher CT concentrations than age-matched males. Acute periodic CT fluctuations were discovered by immunoassay of plasmas obtained at 3-h intervals; the greatest values occurred just before and during feeding. Fed rats have higher CT than starved rats. We have used immunoadsorbent chromatography to concentrate specifically CT moieties from large volumes of plasma for concurrent immunoassay and bioassay measurements of circulating CT. These concurrent measurements of immunoextracted plasma CT demonstrate that for normal rats, our immunoassay measurements correspond to bioassay measurements. In 1-yr-old rats on a regulated feeding schedule, the biological (hypocalcemic) activity of CT recovered from 15 ml peripheral plasma ranged from less than 0.15 mU in starved males to 0.78 mU in feeding females. We conclude that biologically active CT circulates in normal rats and that the blood concentration of biologically active CT progressively increases during somatosexual maturation, being highest in old females, and increases acutely just before and during feeding.

Effects of the calcium channel activator BAY-K-8644 on in vitro secretion of calcitonin and parathyroid hormone.

The recently discovered calcium (Ca) channel activator BAY-K-8644 [methyl-1,4-dihydro-2,6-dimethyl-3-nitro-4-(2-trifluoromethylphenyl) pyridine-5-carboxylate], an analog of the calcium channel blockers nifedipine and nitrendipine, was tested to determine its potential for altering hormone secretion in an in vitro system designed to study concurrent secretion of calcitonin (CT) and PTH. Addition of BAY-K-8644 (10(-4)-10(-5) M) to medium (1 mM Ca) bathing baby rat thyroparathyroids enhanced secretion of CT at least 2- to 4-fold and suppressed PTH release by as much as 75-85%. Addition of BAY-K-8644 alone to medium containing high (2.5 mM) Ca did not further enhance the already high rate of CT release, nor did it cause any further suppression of PTH secretion. BAY-K-8644 did not stimulate CT release or suppress PTH release in the absence of medium Ca. Addition of the Ca channel blocker nitrendipine (10(-5) M) inhibited CT release at either 1 or 2.5 mM Ca, and at 1 mM Ca, nitrendipine negated the simulatory effect of 10(-5) M BAY-K-8644 on CT release. However, at 2.5 mM Ca, 10(-5) M BAY-K-8644 reversed the marked inhibitory effect of 10(-5) M nitrendipine on CT release. At 1 mM Ca, PTH secretion was inhibited equally well by BAY-K-8644 and nitrendipine, and both agents together caused a further suppression of PTH release. The results indicate that Ca entry into the thyroid C-cell and parathyroid chief cell may occur via classical voltage-sensitive Ca channels and that the newly described Ca channel activator BAY-K-8644 should prove useful as a probe for studying hormone secretion in Ca-dependent secretory systems.

Effects of vasoactive intestinal peptide on adenosine 3',5'-monophosphate, ornithine decarboxylase, and cell growth in a human colon cell line.

Vasoactive intestinal peptide (VIP) is a widely distributed neuropeptide that has been considered a potential regulator of cell growth and differentiation in various tissues, including the gut. To examine this idea, we used a human colon carcinoma cell line (LoVo) as a model system and measured ornithine decarboxylase (ODC), because this is the rate-limiting enzyme for the formation of polyamines, which are thought to be key factors in regulating cell growth. LoVo cells, grown to about 80% confluence in F-12 medium containing 10% fetal bovine serum, were preincubated for 5 h in low serum medium (1% fetal bovine serum in F-12), and ODC activity was determined by measuring 14CO2 liberated from 14C-labeled ornithine. VIP caused a dose-related biphasic change in ODC, with activity increased at 10 pM, maximal (5-fold increase) at 10 nM, and decreased toward basal at 100 nM to 1 microM. Incubation of cells for 6 days with VIP in low serum medium showed similar changes in cell numbers, with growth being increased by doses in the 1 pM to 100 nM range and decreased at higher doses (greater than or equal to 100 nM). Exposure of cells to 5 mM alpha-difluoromethylornithine blocked both the VIP-induced increase in cell number and the VIP-induced increase in ODC activity. Increased ODC mRNA was detected after 2 h of exposure to VIP, a time at which ODC activity peaked after treatment, and the increase in ODC mRNA caused by VIP was dose-dependent. In related experiments LoVo cells were found to have high affinity VIP receptors (Kd = 0.4 nM), as assessed by examination of [125I]VIP binding in the presence of varying concentrations of unlabeled VIP. Studies of intracellular cAMP revealed a dose-related increase in cAMP in response to VIP (ED50 = 11 pM), and the adenylate cyclase activator forskolin increased both ODC activity and ODC mRNA. The findings support the idea that LoVo cells have VIP receptors linked to cAMP which can stimulate cell growth at least in part by increasing ODC synthesis and activity, thereby altering the production of polyamines. The decreased growth and ODC activity observed with high doses of VIP may involve a second messenger other than cAMP.

Overexpression of parathyroid hormone-related protein enhances apoptosis in the rat intestinal cell line, IEC-6.

We used the rat intestinal cell line, IEC-6, to study potential effects of overexpression of PTH-related protein (PTHrP) on apoptosis. A clonal line of PTHrP-overexpressing cells was established by stably transfecting parental cells with PTHrP complementary DNA in a sense orientation (sense). A similarly transfected line stably, transfected with empty vector, served as control (vector). Immunoreactive PTHrP, measured in culture medium, showed that sense cells secreted approximately 30 times as much PTHrP as did vector control cells. Apoptosis induced by serum withdrawal was evaluated by several methods. DNA laddering was demonstrable in sense-transfected cells as early as 12 h after serum withdrawal but not until later time points in vector-transfected control cells. Flow cytometric analysis of propidium iodide-stained cells showed a greater increase in the sub-G1 (apoptotic) population in sense cells, compared with vector. Fluorescent microscopy with Hoechst 33258 dye showed increased nuclear fragmentation and condensation in sense cells. Studies of apoptotic gene expression by ribonuclease protection assay, and protein by Western blot analysis, showed an enhanced ratio of Bax to Bcl-x(L) in sense cells. Mutation of the PTHrP nuclear localization amino acid sequence negated the ability of PTHrP to enhance apoptosis.

Effects of parathyroid hormone-related peptide on adenosine 3',5'-monophosphate and ornithine decarboxylase in a human colonic cell line.

PTH-related peptide (PTHrP) is widely distributed in normal tissues, including the gut, and is considered a potential autocrine or paracrine regulator of cellular growth and differentiation. With this in mind, a human colonic cell line (LoVo) was used to study the effect of PTHrP on ornithine decarboxylase (ODC), because ODC is known to have profound effects on the growth and differentiation of many cell types via stimulation of synthesis of polyamines. cAMP also was measured, because this second messenger has been implicated in the regulation of ODC activity. Nearly confluent LoVo cells, grown in F-12 medium and 10% fetal bovine serum (FBS), were preincubated in 1% FBS for at least 5 h, and then PTHrP-(1-34) was added, and the incubation was continued for up to 6 h. Cell extracts were analyzed for ODC activity by measuring 14CO2 liberated from 14C-labeled ornithine, for cAMP by RIA, and for ODC mRNA by Northern analysis. PTHrP produced dose-related increases in both cAMP (2- to 3-fold) and ODC (3- to 5-fold), with a maximal effect at 0.1-1 microM and an ED50 of 1-10 nM. Comparison of the cAMP and ODC responses to PTHrP showed a strong correlation (r = 0.96; P less than 0.001). The effects of 1 microM PTHrP-(1-34) to increase cAMP and ODC were completely inhibited by 10-20 microM of the specific antagonist [Asn10,Leu11]PTHrP-(7-34). PTHrP-(1-34) did not stimulate ODC activity when cells were incubated without FBS. The stimulation of ODC activity by PTHrP-(1-34) was maximal at 2 h, a time at which an increase in ODC mRNA also was evident. PTH-(1-34) and forskolin also stimulated ODC activity, but PTHrP-(67-86) amide was ineffective. The results indicate that the N-terminal portion of the PTHrP molecule can stimulate ODC activity in a human colon cell line and that the effect is probably mediated by cAMP. The results are consistent with the idea that PTHrP may influence cell growth and differentiation in the gut via an effect on polyamine biosynthesis. Since LoVo cells also express PTHrP mRNA, this gastrointestinal cell line may serve as a useful model for studying autocrine regulation of gut cell growth and differentiation by PTHrP.

Effect of endogenously produced parathyroid hormone-related peptide on growth of a human hepatoma cell line (Hep G2).

A well differentiated human hepatoma cell line (Hep G2) was used to explore potential roles for PTH-related peptide (PTHrP) as an autocrine/paracrine growth factor. Using Northern analysis or reverse transcription-PCR, Hep G2 cells were found to express messenger RNAs for both PTHrP and the cloned PTH/PTHrP receptor, and the cells exhibited specific binding for [125I]PTHrP(1-36). Hep G2 growth medium was found to contain relatively large amounts of immunoreactive PTHrP (30 vs. 1-2 pM in medium not exposed to cells), and the PTHrP in growth medium (conditioned medium) was shown to contain N-terminal PTHrP biological activity, as assessed by the ability of the medium to stimulate cAMP production in rat osteosarcoma cells (ROS 17/2.8). Conditioned medium produced a dose-dependent severalfold increase in ROS cell cAMP that could be blocked by the PTHrP receptor antagonist [Asn10,Leu11,DTrp12]PTHrP-(7-34). PTHrP in Hep G2 cells also was detected by immunocytochemistry using antiserum to either synthetic PTHrP-(1-34) or recombinant PTHrP-(-5 to 139). Furthermore, these antisera were found to inhibit the ability of PTHrP-(1-34) to stimulate ROS cell cAMP production. When either antiserum (1:800-1:100 dilution) was added to subconfluent Hep G2 cells in medium containing 5% FBS for 3 days, a dose-related 40-50% increase in cell number occurred that could be inhibited by concurrent addition of 10 microM synthetic PTHrP-(1-36). The results show that Hep G2 cells synthesize and secrete both immunoreactive and biologically active PTHrP. As neutralization of PTHrP secreted by these cells by the addition of antiserum to PTHrP resulted in increased cell growth, the findings suggest that PTHrP can function as an autocrine or paracrine growth factor to suppress the growth of these human hepatoma cells.