- Email: [email protected]
Regulation of Parathyroid Hormone-Related Protein Secretion and mRNA Expression in Normal Human Keratinocytes and a Squamous Carcinoma Cell Line
EXPERIMENTAL CELL RESEARCH ARTICLE NO.
232, 79–89 (1997)
EX973481
Regulation of Parathyroid Hormone-Related Protein Secretion and mRNA Expression in Normal Human Keratinocytes and a Squamous Carcinoma Cell Line MICHELLE T. WECKMANN, ANDREA GRO¨NE, CHARLES C. CAPEN,
AND
THOMAS J. ROSOL1
Department of Veterinary Biosciences, The Ohio State University, Columbus, Ohio 43210
(HHM) [1]. HHM is a clinical syndrome resulting from the secretion of PTHrP by certain neoplasms which stimulates osteoclastic bone resorption and renal calcium reabsorption. Increased levels of circulating PTHrP bind to and stimulate parathyroid hormone (PTH) receptors in bone and kidney cells, mimicking the effects of excess PTH. The clinical syndrome of HHM can be reversed experimentally in nude mice with transplanted squamous carcinomas using neutralizing anti-PTHrP antibodies [2]. In addition to the production of PTHrP by human and animal neoplasms [3–5], PTHrP has been demonstrated in many normal tissues, including the epidermis and lactating mammary gland, kidney, bone, placenta, brain, and fetal tissues [6, 7]. In contrast to patients with HHM who have increased circulating concentrations of PTHrP, serum PTHrP in normal adults is very low (õ1 pM) [8]. This suggests that PTHrP functions as an autocrine or paracrine factor in most normal tissues [9]. The mechanisms regulating the secretion, expression, and biologic activity of PTHrP in most normal tissues are poorly understood and may vary between cell types. Normal human keratinocytes (NHFK) and a human squamous carcinoma cell line (A253) were selected for investigating the regulation of PTHrP secretion. Both cell types have been reported to produce PTHrP, and HHM occurs in certain patients with squamous carcinomas [6, 10–12]. In addition to their role in barrier formation and wound repair, keratinocytes function as secretory cells, with secretion of proteins, including cytokines, enzymes, adhesion molecules, and PTHrP [13]. Secretion of proteins occurs by a well-defined series of steps and may be regulated or constitutive. Secretion depends upon the coordinated function of many cellular components, including the cytoskeleton, intracellular environment, and GTP-binding proteins [14]. Cells may produce, modify, and direct both regulated and constitutive transport simultaneously [15, 16]. Monensin, cytochalasin B, colchicine, and GTPgS are agents which can modify secretion by targeting different parts of the secretory pathway. Monensin, an ionophore, pre-
Parathyroid hormone-related protein (PTHrP) has been identified as a causative factor in the pathogenesis of humoral hypercalcemia of malignancy (HHM). The regulation and mechanisms of PTHrP secretion in most normal and malignant cells are unknown. PTHrP secretion, mRNA expression, and transcription were measured in neoplastic human squamous carcinoma cells (A253) and normal human foreskin keratinocytes (NHFK) by radioimmunoassay, RNase protection assay, and transient transfections of the 5*-flanking region of human PTHrP in a luciferase expression vector. Mechanisms of PTHrP secretion were investigated using chemicals (monensin, colchicine, cytochalasin B, guanosine 5*-[g-thio]triphosphate (GTPgS)) that interfere with or facilitate intracellular transport. Monensin inhibited PTHrP secretion in both NHFK and A253 cells. Ultrastructurally, monensin caused dilatation of rough endoplasmic reticulum and the formation of numerous cytoplasmic secretory vacuoles in both cell lines. Colchicine decreased PTHrP production in NHFK cells and stimulated PTHrP production and mRNA levels in A253 cells. Colchicine also stimulated transcription of the PTHrP–luciferase reporter gene. Cytochalasin B stimulated PTHrP secretion and mRNA expression in A253 cells, but had no effect in NHFK cells. GTPgS had no effect on PTHrP secretion in either cell line. It was concluded that PTHrP secretion is dependent on the constitutive movement of secretory vesicles to the cytoplasmic membrane and regulation of PTHrP secretion and mRNA expression are altered in squamous carcinoma cells compared to normal human keratinocytes in vitro. q 1997 Academic Press
INTRODUCTION
Parathyroid hormone-related protein (PTHrP) was originally identified as a causative factor in the pathogenesis of humoral hypercalcemia of malignancy 1 To whom correspondence should be addressed at The Ohio State University, Department of Veterinary Biosciences, 1925 Coffey Road, Columbus, OH 43210. Fax: (614) 292-6473. E-mail: [email protected].
79
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
0014-4827/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
eca
80
WECKMANN ET AL.
vents the acidification of secretory vesicles and their subsequent transport to the cell membrane [17]. Colchicine and cytochalasin B impair the function of microtubules and microfilaments, respectively [18,19]. Guanosine 5*-(g-thio]triphosphate (GTPgS), a nonhydrolyzable form of GTP, activates certain GTP-binding proteins to enhance vesicular secretion [20]. Modification of PTHrP secretion by these chemicals may provide information concerning the mechanisms and regulation of PTHrP secretion. Understanding the control of PTHrP secretion and expression in both normal and neoplastic cells is important in determining how certain malignant cells are able to secrete biologically active PTHrP, resulting in elevated blood levels, and induce hypercalcemia. The purpose of this study was to investigate mechanisms of PTHrP secretion and expression in NHFK and A253 cells using monensin, colchicine, cytochalasin B, and GTPgS, which may interfere with or stimulate intracellular vesicular transport. MATERIALS AND METHODS Materials. Monensin (Calbiochem, San Diego, CA) and cytochalasin B (Sigma Chemical Co., St. Louis. MO) were dissolved in 100% ethanol at 1.0 mM and 5 mg/ml, respectively. Colchicine (Sigma) and lumicolchicine, an inactive analog for colchicine (Sigma), were dissolved in Hanks’ balanced salt solution (HBSS) at 1.0 mM. GTPgS was dissolved in water at 10 mM. All reagents were diluted to final concentration in cell culture medium prior to use. Cell culture. Human squamous carcinoma cells (A253) were obtained from the American Type Culture Collection (Rockford, MD) and grown in McCoy’s 5A media (Sigma) supplemented with 10% fetal bovine serum (FBS)(Life Technologies, Grand Island, NY), 2 mM L-glutamine, and 50 mg/ml gentamicin. Normal human foreskin keratinocytes (NHFK) were prepared from neonatal foreskins following a procedure described by Boyce and Ham [6, 21]. Briefly, neonatal foreskin tissues were cut into 2- to 5-mm2 pieces and incubated for 48 h in 0.2% trypsin (type III, Sigma) at 47C. The epidermis was separated from the dermis and dispersed in MCDB 153 medium (Sigma) containing 10% FBS. The cell suspension was centrifuged and the pellet was resuspended in keratinocyte serum-free basal medium (Life Technologies) containing low calcium (0.08 mM), epidermal growth factor (5 ng/ml), and bovine pituitary extract (35 ml/ ml), placed in 75-cm2 culture flasks, and incubated at 377C in 95% air/5% CO2 . The NHFK were used from passages 1–5 only. Conditioned medium collection. Conditioned medium (CM) was collected from cells grown in 12-well culture plates (Corning). Medium was collected 24 h following treatment or medium change. Conditioned media were stored at 0207C until assayed for PTHrP content by radioimmunoassay. Secretion studies. Treatments were added in 2 ml of medium to cells in triplicate wells of 12-well plates. The PTHrP secretion pattern was evaluated in each cell type over a 2-week culture period without any treatment. The cells were plated at 20% confluence, and medium was collected every 24 h. PTHrP secretion was maximal and stable in A253 cells after confluence was reached (ú 6 days in culture) and was maximal in NHFK near the time of confluence, but decreased in postconfluent cultures of NHFK. Secretion studies were conducted for 24 or 96 h, with CM collected and fresh medium (including treatment) added every 24 h. Experiments were begun near the time of confluence in cultures of NHFK (Days 3–5 after plating)
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
and after confluence in A253 cells (Days 6–7 after plating). In experiments with GTPgS, the cells were permeabilized with Trans-Port kit (Life Technologies) to permit uptake of GTPgS. PTHrP nonequilibrium radioimmunoassay (RIA). The RIA for PTHrP was performed as described by our laboratory using chicken polyclonal anti-PTHrP-(1-36) antibody [22]. [Tyr36]PTHrP-(1-36) (Bachem, Torrance, CA) was radioiodinated with I125 and Iodogen (Pierce Chemical Co, Rockford, IL) and purified using reverse-phase HPLC [22, 23]. Polyclonal chicken anti-PTHrP-(1-36) antibodies (10 mg/ml in PBS, 1:1000 dilution) were added to borosilicate tubes containing standards or medium samples (100 ml) and assay buffer (20 mM sodium phosphate, pH 7.4, 140 mM NaCl, 50 mM EDTA, 0.5% BSA, 0.1 sodium azide) and were incubated for 48 h at 47C. Iodinated PTHrP-(1-36) was added (10,000 cpm/tube) and dextrancoated charcoal (5 mg/ml activated charcoal, 0.5 mg/ml dextran) was utilized to separate the bound from the free PTHrP after 48 h. The supernatant was counted in a gamma-radiation counter and PTHrP content quantified by log-logit transformation using Securia 1.0 (Packard Instruments, Downers Grove, IL). The RIA was sensitive to 0.1 ng PTHrP-(1-36)/ml of medium. All assays included an internal standard of conditioned medium with a PTHrP concentration of 1 ng/ml. The maximum interassay variability was 25% and it was routinely 10– 15%. Isolation of total RNA. For total RNA isolation, cells were plated in 90-mm tissue culture dishes in fully supplemented medium. Total RNA was isolated by the method of Chomczynski and Sacchi [24, 25]. Briefly, the cells were lysed with 4 M guanidium thiocyanate, 25 mM sodium citrate, 0.5% Sarkosyl (pH 7.4), and 2-mercaptoethanol (BME) and either stored at 0207C or extracted three times with water-saturated phenol (pH 4.0) and chloroform:isoamyl alcohol (24:1). The RNA was precipitated with either isopropanol or 2 vol of 100% ethanol with 0.1 vol of 4 M sodium acetate, centrifuged at 30,000g for 15 min, resuspended in diethyl polycarbonate (DEP)treated water, and quantified at A260/280. RNase protection assay (RPA). Total cellular PTHrP mRNA was measured by RPA (Ambion, Austin, TX). Complementary DNA riboprobes labeled with [32P]UTP (Dupont, New England) for human PTHrP mRNA (No. 661) or human glyceraldehyde-3-phosphate dehydrogenase (pTRI-GAPDH, Ambion) were utilized. GAPDH is a constitutively expressed gene in keratinocytes and was used as a loading control to standardize the amounts of RNA present [26, 27]. The antisense GAPDH template contains a 316-bp fragment of the human GAPDH gene derived from exons 5–8 in the pTRIPLEscript transcription vector [28]. The human cDNA riboprobe (661) is a 537bp fragment of human PTHrP kindly obtained from Genentech, Inc. (South San Francisco, CA). The plasmid contains a ClaI–SstI fragment of PTHrP cDNA clone BRF.52. The fragment was cloned between the AccI and SstI sites of the vector pSP65 [29, 30]. The cDNA riboprobes were linearized by digestion with HindIII and transcribed using T7 or SP6 RNA polymerase (Maxiscript In Vitro Transcription kit, Ambion). The RPA was performed as follows. Total RNA from either A253 squamous carcinoma cells (40 mg) or NHFK cells (20 mg) was coprecipitated with [32P]UTP-labeled RNA probes for human PTHrP and GAPDH. After coprecipitation, the RNA was centrifuged, washed in ethanol, dried, resuspended in hybridization buffer, heated to 957C for 3 min, and allowed to hybridize overnight at 427C. The unhybridized RNA was digested with an A/T1 RNase mixture and the doublestranded RNA was precipitated. The double-stranded RNA was resuspended in loading buffer (80% formamide, 1 mM EDTA, pH 8.0, 0.1% bromophenol blue, 0.1% xylene cyanol), heated to 957C, and separated using a 30% urea/8% acrylamide/bis gel at 300 V for 2 h. The gel was exposed to X-ray film with an intensifying screen at 0807C for 8 to 24 h. PTHrP signal intensity was quantified using a laser densitometer (ImageQuant Series 300, Molecular Dynamics, Sunnyvale, CA) and was normalized to GAPDH signal intensity.
eca
81
PTHrP REGULATION IN KERATINOCYTES Transient transfection of the PTHrP gene/luciferase construct. To examine the effects of monensin, cytochalasin B, and colchicine on PTHrP gene expression, normal keratinocytes and A253 squamous carcinoma cells were transfected with plasmids containing portions of the 5*-flanking sequence of the human PTHrP gene ligated to a promoterless bacterial luciferase gene (pGL2-Basic from Promega, Madison, WI). The PTHrP construct consisted of 950 bp of 5*-flanking DNA and included the untranslated exons III and IV and the 5* end of exon V, 20 bp short of the ATG translation start site. The plasmid (pSMR38) with the 5*-flanking sequence contained the P2 and P3 PTHrP promoters and was generously provided by Dr. Larry Suva (Beth Israel Hospital, Boston, MA) [31]. Transfections of the NHFK were conducted in 12-well culture plates at 60% confluency using lipofectin (Life Technologies). Transfections of the A253 cells were conducted using polybrene (Sigma) following the method described by Jiang et al. [32]. Cells were treated 48 h after transfection with supplemented medium containing serum, monensin, cytochalasin B, colchicine, or lumicolchicine. To determine the transfection efficiency of each sample, cells were cotransfected with the CMV– b-galactosidase plasmid (Promega), containing the b-galactosidase gene driven by a CMV (cytomegalovirus) promoter. b-Galactosidase was assayed in cell extracts using the b-Galactosidase Enzyme Assay System (Promega). Cell extracts where assayed for luciferase using the Luciferase Assay System (Promega) and a liquid scintillation counter (Packard). Immunofluorescence of actin and tubulin. To investigate the effects of cytochalasin B and colchicine on the cell cytoskeleton, immunofluorescence for actin and tubulin was performed on NHFK and A253 cells. Normal keratinocytes and A253 cells were grown on Microprobe slides (Fischer Scientific) and treated with either cytochalasin B (5 mg/ml) or colchicine (0.1 mM) for 4 h. Actin was visualized by direct immunofluorescence using fluorescein phalloidin (Molecular Probes Inc., Eugene, OR). The fixation procedure consisted of immersion in acetone at 0207C for 4 min after incubation in 3.7% formaldehyde for 20 min. The cells were then washed twice with DPBS (Dulbecco’s phosphate-buffered saline), incubated for 20 min at room tem-
perature with fluorescein phalloidin at a dilution of 1:40 (1 unit) in DPBS, and washed three times with DPBS. Tubulin was stained by indirect immunofluorescence using rat anti-b-tubulin and anti-rat IgG FITC conjugate (Harlan Sera-Ral Ltd., Sussex, England). The cells were fixed for 20 min at 0207C in methanol and permeabilized for 1 min at 0207C with acetone, washed three times in DPBS, incubated 1 h with a 1:20 dilution of the primary antibody, washed three times with DPBS, incubated 30 min with a 1:20 dilution of secondary antibody, and washed three times with DPBS. The cells were analyzed using an Olympus epifluorescent microscope at 10001 magnification with an oil immersion objective and photomicrographs were obtained with Kodak T-max (ASA 400) black and white film. Electron microscopy. A253 or NHFK cells were grown on 30-mm plates and treated with monensin (0.1–1 mM) for 24, 48, 72, or 96 h for evaluation by transmission electron microscopy. The cells were fixed in 3% glutaraldehyde, postfixed in 1 M cacodylate buffer and osmium, embedded in medcast plastic (Ted Pella Redding, CA), and examined with a Philips 300 electron microscope. Statistical analysis. Numerical data were expressed as means { the standard error of the mean (SEM) for triplicate samples. PTHrP concentration of the conditioned medium was expressed as ng/ml. Statistical differences between means for control and treatment groups were evaluated with Instat 1.0 (Graph PAD software, San Diego, CA) using one-way analysis of variance (ANOVA) and Tukey’s mean separation test, with the level of significance at P õ 0.5, õ 0.01, or õ 0.001.
RESULTS
Secretion of PTHrP by NHFK and A253 Cells To determine the secretory pattern of PTHrP in NHFK and A253 cells in monolayer culture, PTHrP concentration of the medium was measured at 24-h
FIG. 1. Secretion of PTHrP by NHFK and A253 squamous carcinoma cells in vitro. Conditioned medium was collected at 24-h intervals, with cells reaching confluence at Days 4–5. Data for the NHFK and A253 cells represent means { SE (n Å 3).
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
eca
82
WECKMANN ET AL.
FIG. 2. Effects of monensin on PTHrP secretion. NHFK and A253 squamous carcinoma cells were treated with monensin (0.1, 0.3, and 1 mM) for 96 h, and medium was collected every 24 h. Data for NHFK (A) and A253 squamous carcinoma cells (B) represent means { SE (n Å 3). *Significantly different (P õ 0.001) from control.
intervals for 14 days, starting when the cells were 20% confluent. Secretion of PTHrP by normal keratinocytes increased until confluency (Day 4), remained high for 2 days near confluence (Days 5 and 6), and then slowly declined until Day 14 (Fig. 1). Secretion of PTHrP in A253 cells followed a different pattern than did NHFK. PTHrP secretion was low prior to confluence at Day 4, after which secretion increased and remained stable for Days 6–14 (Fig. 1).
eter. In the NHFK there were mild reductions in PTHrP mRNA in monensin-treated cells at 1 h (1.5fold reduction) and 12 h (Fig. 3A). PTHrP mRNA in
Effect of Monensin on PTHrP Secretion, mRNA Levels, and Transcription In order to determine whether changes in vesicle acidity altered PTHrP secretion in normal and neoplastic human cells, NHFK and A253 cells were treated with monensin for up to 96 h. PTHrP secretion by NHFK and A253 cells was significantly (P õ 0.001) inhibited by monensin treatment (0.1 to 1 mM) at 24 to 96 h (Figs. 2A and 2B). The toxicity of monensin resulted in cell death in concentrations greater than 1 mM. Steady-state levels of PTHrP mRNA were measured to determine if monensin decreased expression of PTHrP mRNA in addition to impairing the release of secretory vesicles. PTHrP mRNA was detectable at all time points (1, 6, and 12 h) with the lowest levels at 12 h. All of the lanes were standardized with the GAPDH mRNA loading control using a laser densitom-
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
FIG. 3. The effect of monensin on the steady-state levels of PTHrP and GAPDH mRNA was analyzed by RNase protection assay. NHFK (A) and A253 squamous carcinoma cells (B) were treated with monensin (0.3 mM) or control for 1, 6, and 12 h. Top band, PTHrP probe; lower band, GAPDH probe: T, monensin treatment; C, control.
eca
PTHrP REGULATION IN KERATINOCYTES
83
FIG. 4. Ultrastructure of NHFK and A253 squamous carcinoma cells. Control cultures of NHFK (A) demonstrating abundant free polyribosomes and rough endoplasmic reticulum in the cytoplasm. Control cultures of A253 squamous carcinoma cells (B) demonstrating multiple desmosomes and free polyribosomes in the cytoplasm. Treatment of NHFK (C) with monensin resulted in the presence of dilated secretory vacuoles and endoplasmic reticulum. Treatment of A253 cells (D) with monensin resulted in the accumulation of enlarged secretory vacuoles in the cytoplasm. R, rough endoplasmic reticulum; V, secretory vacuoles; D, desmosomes; N, nucleus. (A) Bar, 0.81 mm; (B) bar, 0.94 mm; (C) bar, 0.95 mm; (D) bar, 1.2 mm.
the A253 cells exhibited a mild decrease in PTHrP mRNA at 12 h and no change at 1 or 6 h (Fig. 3B). Monensin had no significant effect on expression of the PTHrP–luciferase reporter gene in NHFK or A253 cells (data not shown). Effects of Monensin on the Ultrastructure of NHFK and A253 Cells The ultrastructure of NHFK demonstrated a network of intermediate filaments, microtubules, and an abundance of both free polyribosomes and rough endo-
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
plasmic reticulum (Fig. 4A). The A253 cells had multiple desmosomes with anchoring intermediate filaments and an abundance of free polyribosomes (Fig. 4B). Monensin caused dilatation of rough endoplasmic reticulum in the NHFK and dilation of cytoplasmic secretory vacuoles in both the NHFK and A253 cells (Figs. 4C and 4D). Cytoplasmic vacuoles were seen as early as 12 h by electron microscopy in cells treated with 0.3 mM monensin. The number and size of the vacuoles were greater in the A253 cells than in the NHFK. Aggregates of cellular organelles could be seen in the A253
eca
84
WECKMANN ET AL.
FIG. 5. Effect of colchicine on PTHrP secretion. NHFK were treated with colchicine for 96 h, and conditioned medium was collected daily. Data represent means { SE (n Å 3). *Significantly different (P õ 0.01) from control.
cells treated with monensin. Intact membranes were present around the perimeter of some vacuoles. The vacuoles were present at 12, 24, 48, 72, and 96 h in monensin-treated cells. There was an increase in the number of vacuoles from 12 to 24 h, but no increase in vacuoles from 24 to 96 h. Effect of Colchicine on PTHrP Secretion, mRNA Levels, and Transcription The effects of colchicine were evaluated to determine whether disruption of cytoplasmic microtubules would affect PTHrP secretion. Colchicine significantly (P õ 0.01) inhibited PTHrP secretion at concentrations of 0.1, 0.3, and 1 mM at 24 to 96 h in NHFK (Fig. 5). In contrast, colchicine-treated A253 cells had a significant increase in PTHrP production at doses ranging from 0.1 to 1 mM at 24 to 96 h (Fig. 6). Lumicolchicine, an inactive analog of colchicine which does not disrupt microtubules [33], was used as a control. Lumicolchicine treatment (0.1, 0.3, 1 mM) had no effect on PTHrP secretion in either the NHFK or A253 cells (data not shown). Colchicine had a mild stimulatory effect on steadystate levels of PTHrP mRNA in NHFK at 12 h (40% increase) and at 24 h (20% increase) (Fig. 7A). In contrast, colchicine markedly increased PTHrP mRNA in A253 cells at 6 (2.2-fold), 12 (1.6-fold), and 24 h compared to the control cells (Fig. 7B). To determine whether colchicine altered transcription of PTHrP, a segment of the 5*-flanking DNA of the human PTHrP
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
FIG. 6. Effect of colchicine on PTHrP secretion. A253 squamous carcinoma cells were treated with colchicine for 96 h, and conditioned medium was collected daily. Data represent means { SE (n Å 3). *Significantly different (P õ 0.05) from control.
gene containing promoters P2 and P3 was coupled to a luciferase expression vector and used to transiently transfect normal keratinocytes and A253 cells. Colchicine resulted in a 3-fold increase in transcription of the PTHrP–luciferase transcript in A253 cells and a 50%
FIG. 7. Effect of colchicine on steady-state levels of PTHrP and GAPDH mRNA was analyzed by RNase protection assay. NHFK (A) and A253 squamous carcinoma cells (B) were treated with colchicine (1 mM) or control for 1, 6, 12, and 24 h. Top band, PTHrP probe; lower band, GAPDH probe; T, colchicine treatment; C, control.
eca
85
PTHrP REGULATION IN KERATINOCYTES
had a variable response to cytochalasin B treatment (5 mg/ml), with a slight increase in mRNA at 1 h and a slight decrease (1-fold) at 6 h when standardized to the GAPDH controls (Fig. 11A). There was no significant change in PTHrP mRNA levels in the NHFK at 12 h. PTHrP mRNA was markedly increased by cytochalasin B treatment in the A253 cells at 6 (6-fold) and mildly increased at 12 h (Fig. 11B). Cytochalasin B had no significant effect on expression of the PTHrP–luciferase reporter gene in NHFK or A253 cells (data not shown). Effects of Cytochalasin B on Microfilaments in NHFK and A253 Cells
FIG. 8. Transient transfection of A253 squamous carcinoma cells with human 5*-flanking PTHrP DNA in a luciferase expression vector and the effects of 24 h treatment with colchicine, lumicolchicine, or 10% serum were measured. Data represent means { SE (n Å 3). *Significantly different (P õ 0.05) from control.
increase in NHFK. However, there was no effect of the inactive analog, lumicolchicine (Fig. 8). Effects of Colchicine on Microtubules in NHFK and A253 Cells Indirect immunofluorescence was utilized to evaluate the effects of colchicine on microtubules in NHFK and A253 cells. The NHFK and A253 cells arranged their microtubules into a network of cytoplasmic arrays that spanned radially from the nucleus to the cell’s border (Figs. 9A and 9B). The network of microtubules in the cells treated with colchicine collapsed into large aggregates in the cytoplasm, coiled into juxtanuclear caps in the A253 cells, and were entirely disrupted in the NHFK (Figs. 9C and 9D). Effect of Cytochalasin B on PTHrP Secretion, mRNA Levels, and Transcription The effect of cytochalasin B was investigated to determine whether microfilaments were necessary for normal PTHrP secretion. In the NHFK, cytochalasin B mildly inhibited PTHrP production at 24 h while significantly increasing (P õ 0.001) PTHrP secretion after 24 h in the A253 cells (Fig. 10). Cytochalasin B also had an inhibitory affect on cell growth at all concentrations. In addition, cytochalasin B (1 to 10 mg/ ml) significantly increased PTHrP production at 48, 72, and 96 h in the A253 cells (data not shown). Steadystate levels of PTHrP mRNA in normal keratinocytes
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
Direct immunofluorescence was utilized to visualize the effect of cytochalasin B on cytoskeleton microfilaments of normal keratinocytes and A253 cells. The cell shape remained unchanged in both the control and cytochalasin B-treated NHFK and A253 cells. The control groups of both cell types had strong cytoskeletal staining, with a discernable cellular and nuclear outline. Actin filaments were visible in the cytoplasm and along the cytoplasmic membrane (Figs. 12A and 12B). In contrast, the cells treated with cytochalasin B (5 mg/ml) for 4 h had a patchy appearance. There were areas of multiple small and large foci of positive immunofluorescense indicating that cytochalasin B induced depolarization and condensation of the actin filaments (Figs. 12C and 12D). Effects of GTPgS on PTHrP Secretion The A253 and NHFK cells were permeabilized to permit entry of GTPgS into the cytoplasm. The permeabilization process and treatment resulted in up to 100% loss of viability in the A253 cells. The NHFK showed a 20% loss in viability after permeabilization. The uptake and treatment of normal keratinocytes by GTPgS did not affect secretion of PTHrP. DISCUSSION
The mechanisms and regulation of PTHrP secretion are poorly understood. This investigation demonstrated that the regulation of PTHrP secretion and mRNA expression varied between NHFK and A253 squamous carcinoma cells. It has been reported that PTHrP secretion may be regulated or constitutive, depending on cell type [14, 34–36]. In addition, PTHrP is secreted in multiple forms including N-terminal and midregion fragments [37]. Deftos et al. and Ilardi et al. demonstrated PTHrP in secretory granules of atrial myocytes and in the cytoplasm of squamous carcinoma cells, respectively [35, 38]. Rizzoli et al. have reported that PTHrP-(1-34) is secreted via the regulated pathway in a human lung squamous carcinoma cell line
eca
86
WECKMANN ET AL.
FIG. 9. Effects of colchicine on microtubule formation in NHFK and A253 cells. Control cultures of NHFK (A) and A253 squamous carcinoma cells (B) showing the presence of an intact microtubule network. Colchicine resulted in the disruption of the microtubules in NHFK (C) and A253 cells (D).
(BEN) [34]. Using transfected cell lines, it was concluded that PTHrP was secreted in a regulated manner in neuroendocrine cells (e.g., insulinoma cells) and in a constitutive manner in non-neuroendocrine cells (e.g., squamous carcinoma cells and fibroblasts) [36]. Monensin inhibited PTHrP secretion in both NHFK and A253 cells. This was interpreted to be a result of impaired vesicular transport, as evidenced from dilated endoplasmic reticulum and secretory vesicles. In addition, monensin mildly decreased PTHrP mRNA levels. It has been reported that PTHrP secretion was inhibited by monensin in canine and human squamous carcinoma cells [10, 39]. Both the normal and neoplastic keratinocytes responded to monensin in a similar manner, indicating that the mechanisms of PTHrP secretion which involve vesicular acidity function similarly in both the normal keratinocytes and the A253 squamous carcinoma cells. The microtubule disrupter, colchicine, has been reported to inhibit regulated secretion of various proteins in leukocytes, mast cells, adrenal cells, pituitary cells, and parotid cells [40]. We have demonstrated that colchicine inhibited secretion of PTHrP in normal kera-
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
FIG. 10. Effect of cytochalasin B on PTHrP secretion. NHFK and A253 squamous carcinoma cells were treated with cytochalasin B for 24 h. Data represent means { SE (n Å 3). *Significantly different (P õ 0.001) from control.
eca
PTHrP REGULATION IN KERATINOCYTES
FIG. 11. Effect of cytochalasin B on steady-state levels of PTHrP and GAPDH mRNA was analyzed by RNase protection assay. NHFK (A) and A253 squamous carcinoma cells (B) were treated with cytochalasin B (5 m g/ml) or control for 1, 6, 12, and 24 h. Top band, PTHrP probe; lower band, GAPDH probe; T, cytochalasin B treatment; C, control.
87
tinocytes. Unexpectedly, colchicine increased PTHrP secretion and mRNA expression in the A253 squamous carcinoma cells. Microtubules are necessary for regulated secretion in certain cell types [41]. Colchicine inhibited PTHrP secretion in NHFK even though colchicine induced a mild increase in PTHrP mRNA levels and reporter gene expression. This suggests that microtubules were necessary for the secretion of PTHrP in normal keratinocytes. However, in the neoplastic cells (A253), colchicine increased PTHrP secretion, transcription, and steady-state mRNA levels. This demonstrated that PTHrP secretion was not dependent on microtubules in A253 cells. Colchicine has been shown to increase mRNA production of Pro-IL-7b and pituitary adenylate cyclaseactivating peptide gene [42, 43]. Colchicine increased PTHrP mRNA expression and transcription in the A253 cells and NHFK. Colchicine disrupts microtubule structure and stimulates intracellular cyclic AMP production [40, 44, 45]. This is a potential mechanism of increased PTHrP mRNA expression in the A253 squa-
FIG. 12. Effects of cytochalasin B on microfilament formation in NHFK and A253 squamous carcinoma cells. Actin filaments are visible in untreated NHFK (A) and A253 (B) cells. Cytochalasin B treatment induced depolarization and condensation of the actin filaments in the NHFK (C) and A253 cells (D).
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
eca
88
WECKMANN ET AL.
mous carcinoma cells, since calcitonin has been reported to increase PTHrP mRNA in a human squamous carcinoma cell line (BEN) by increasing intracellular cAMP and activating a cAMP-responsive element in the PTHrP gene [46]. Cytochalasin B has been reported to increase regulated secretion of various proteins in leucocytes, macrophages, rat pancreatic islets, and chromaffin cells by breaking down the actin web which prevents secretory granules from approaching the cell membrane [40, 47]. In addition, cytochalasin D, a microfilament disrupter, enhanced basal secretion of PTHrP in human squamous carcinoma cells (BEN) by release of stored protein [34]. Cytochalasin B increased PTHrP secretion in A253 cells and mildly inhibited PTHrP secretion in the NHFK, illustrating a second difference in the regulation of PTHrP secretion between normal and neoplastic keratinocytes. Cytochalasin B disrupts microfilaments and actin formation, which can alter secretion and nuclear RNA transport [40, 48–50]. Increased PTHrP secretion in the A253 squamous carcinoma cells was likely due to upregulation of PTHrP mRNA levels. Cytochalasin has been reported to increase intracellular calcium concentrations in T-lymphocytes and mouse soleus muscle [51–53]. It is possible that cytochalasin B increased PTHrP mRNA production in A253 cells by increasing cytosolic calcium concentration and activating the phospholipase C/protein kinase C pathways [54]. Activation of these intracellular regulatory pathways increased transcription of PTHrP mRNA in a human squamous carcinoma cell line (BEN) [16]. It is unknown whether PTHrP is secreted in either a regulated or a constitutive manner or by both mechanisms in normal keratinocytes and squamous carcinoma cells. Plawner et al. has reported that neuroendocrine cells secreted PTHrP in a regulated manner and squamous carcinoma cells in a constitutive manner [36]. Since disruption of the actin web with cytochalasin B did not increase PTHrP secretion in NHFK, constitutive secretion is most likely in these cells. The lack of secretory granules and the high basal levels of PTHrP secreted into the medium in both the untreated normal keratinocytes and A253 squamous carcinoma cells also suggests that PTHrP is secreted constitutively in both of these cell types. In the A253 cells, cytochalasin B and colchicine both increased PTHrP secretion and steady-state mRNA levels. It is likely that increased PTHrP synthesis due to cytochalasin B was the result of a posttranscriptional effect (such as increased mRNA stability) since there was no effect of cytochalasin B on expression of the PTHrP–luciferase reporter gene. Further studies will be needed to determine whether PTHrP secretion follows a regulated pathway in keratinocytes in addition to the constitutive pathway.
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
In conclusion, PTHrP secretion in normal keratinocytes was dependent on microtubules and was constitutively secreted. In contrast, PTHrP secretion in the A253 squamous carcinoma cells was influenced by increased steady-state PTHrP mRNA levels and did not depend on the presence of intact microfilaments or microtubules. However, the A253 cells are poorly differentiated and may not be representative of all squamous carcinomas. The regulation of PTHrP secretion was altered in neoplastic keratinocytes. This may be a result of changes which occurred during transformation and may be important in the autonomous overproduction and secretion of PTHrP that occurs in the pathogenesis of humoral hypercalcemia of malignancy. We thank Dr. James R. DeWille for his collaboration with the ribonuclease protection assays. The investigation was supported by the National Institute of Arthritis, Musculoskeletal Diseases, and Skin Diseases (AR40220, AR01923).
REFERENCES 1. Rosol, T. J., and Capen, C. C. (1992) Lab. Invest. 67, 680–702. 2. Kukreja, S. C., Rosol, T. J., Wimbiscus, S. A., Shevrin, D. H., Grill, V., Barengolts, E. I., and Martin, T. J. (1990) Endocrinology 127, 305–310. 3. Rosol, T. J., and Capen, C. C. (1988) Dom. Anim. Endocrinol. 5, 1–21. 4. Rosol, T. J., Capen, C. C., Danks, J. A., Suva, L. J., Steinmeyer, C. L., Hayman, J. A., Ebeling, P. R., and Martin, T. J. (1990) Vet. Pathol. 27, 89–95. 5. Moseley, J. M. and Gillespie, M. T. (1995) Crit. Rev. Clin. Lab. Sci. 32(3), 299–343. 6. Werkmeister, J. R., Merryman, J. I., McCauley, L. K., Horton, J. E., Capen, C. C., and Rosol, T. J. (1993) Exp. Cell Res. 208, 68–74. 7. Philbrick, W. M., Wysolmerski, J. J., Galbrath, S., Holt, E., Orloff, J. J., Yang, K. H., Vasavada, R. C., Weir, E. C., Broadus, A. E., and Stewart, A. F. (1996) Physiol. Rev. 76, 127–173. 8. Pandian, M. R., Morgan, C. H., Carlton, E., and Segre, G. V. (1992) Clin. Chem. 38(2), 282–288. 9. Stewart, A. F., and Broadus, A. E. (1990) J. Clin. Endocrinol. Metab. 71, 1410–1414. 10. Merryman, J. I., Capen, C. C., McCauley, L. K., Werkmeister, J. R., Suter, M. M., and Rosol, T. J. (1993) Lab. Invest. 69, 347– 354. 11. Orloff, J. J., Ganz, M. B., Ribaudo, A. E., Burtis, W. J., Reiss, M., Milstone, L. M., and Stewart, A. F. (1992) Am. Physiol. Soc. 92, 599–607. 12. Kremer, R., Karaplis, A. C., Henderson, J., Gulliver, W., Banville, D., Hendy, G. N., and Goltzman, D. (1991) J. Clin. Invest. 87, 884–893. 13. Boyce, S. T. (1994) J. Invest. Dermatol. 102, 8–10. 14. Kelly, R. B. (1985) Science 230, 25–31. 15. Burgess, T. L., and Kelly, R. B. (1987) Annu. Rev. Cell Biol. 3, 243–293. 16. Emly, J. F., Hughes, S., Green, E., and Ratcliffe, W. A. (1994) Biochim. Biophys. Acta 1220, 193–198. 17. Nagatsu, A., Takahashi, T., Isomura, M., Nagai, S., Ueda, T.,
eca
PTHrP REGULATION IN KERATINOCYTES Murakami, N., Sakakibara, J., and Hatano, K. (1994) Chem. Pharm. Bull. 42, 2269–2275. 18. Bhisey, A. N. (1986) in Recent Advances in Cytochalasans (Pendse, G. S., Ed.), pp. 157–179, Balodyan Press, India. 19. Vandecandelaere, A., Martin, S. R., Schilstra, M. J., and Bayler, P. M. (1994) Biochemistry 33, 2792–2801. 20. Gomperts, B. D., and Tatham, P. E. R. (1988) Cold Spring Harbor Symp. on Quant. Biol. 53, 983–992. 21. Boyce, S. T., and Ham, R. G. (1985) J. Tissue Cult. Methods 9, 83–93. 22. Rosol, T. J., Steinmeyer, C. L., Merryman, J. I., Werkmeister, J. R., Groene, A., Weckmann, M. T., Swayne, D. E., and Capen, C. C. (1993) Vet. Immunol. Immunopathol. 35, 321–337. 23. McCauley, L. K., Rosol, T. J., Merryman, J. I., and Capen, C. C. (1992) Endocrinology 130, 300–306. 24. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156–160. 25. Merryman, J. I., DeWille, J. W., Werkmeister, J. R., Capen, C. C., and Rosol, T. J. (1994) Endocrinology 134, 2424–2430. 26. Dveksler, G. A., Basile, A., and Dieffenbach, C. (1992) PCR Methods Appl. 1, 283–285. 27. Bosma, P., and Kooistra, T. (1991) J. Biol. Chem. 266, 17845– 17846. 28. Ercolani, L., Florence, B., Denaro, M., and Alexander, M. (1988) J. Biol. Chem. 263, 15335–15341. 29. Suva, L. J., Winslow, G. A., Wettenhall, R. E., Hammonds, R. G., Moseley, J. M., Diefenbach-Jagger, H., Rodda, C. P., Kemp, B. E., Rodriguez, H., Chen, E. Y., Hudson, P. J., Martin, T. J., and Wood, W. I. (1987) Science 237, 893–896. 30. Melton, D. A., Kreig, P. A., Rebagliati, M. R., Maniatis, T., Zinn, K., and Green, M. R. (1984) Nucleic Acids Res. 12, 7035–7057. 31. Kiriyama, T., Gillespie, M. T., Glatz, J. A., Fukumoto, S., Moseley, J. M., and Martin, T. J. (1993) Mol. Cell. Endocrinol. 92, 55–62. 32. Jiang, C. K., Connolly, D., and Blumenburg, M. (1991) J. Invest. Dermatol. 97, 969–973. 33. Chanard, J., Black, R., Purkerson, M., Lewis, J., Klahr, S., and Slatopolsky, E. (1977) Endocrinology 101, 1792–1800. 34. Rizzoli, R., Aubert, M. L., Sappino, A. P., and Bonjour, J. P. (1994) Int. J. Cancer 56, 422–426.
35. Deftos, L. J., Burton, D. W., and Brandt, D. W. (1993) J. Clin. Invest. 92, 727–735. 36. Plawner, L. L., Philbrick, W. M., Burtis, W. J., Broadus, A. E., and Stewart, A. F. (1995) J. Biol. Chem. 270, 14078–14084. 37. Soifer, N. E., Dee, K. E., Insogna, K. L., Burtis, W. J., Matovcik, L. M., Wu, T. L., Milstone, L. M., Broadus, A. E., Philbrick, W. M., and Stewart, A. F. (1992) J. Biol. Chem. 267, 18236– 18243. 38. Ilardi, C. F., and Faro, J. C. (1985) Arch. Pathol. Lab. Med. 109, 752–755. 39. Lo¨wik, C. W., Hoekman, K., Offringa, R., Groot, C. G., Hendy, G. N., Papapoulos, S. E., and Ponec, M. (1992) J. Invest. Dermatol. 98, 198–203. 40. Allison, A. C., and Davies, P. (1974) Adv. Cytopharm. 2, 237– 248. 41. Robin, P., Rossignol, B., and Raymond, M. N. (1995) Eur. J. Cell Biol. 67, 227–237. 42. O’Brien, J. M., Wewers, M. D., Moore, S. A., and Allen, J. A. (1995) J. Immunol. 154, 4113–4122. 43. Hannibal, J., Mikkelsen, J. D., Fahrenkrug, J., and Larsen, P. J. (1995) Endocrinology 136(9), 4116–4124. 44. Galli, A., and DeFelice, L. J. (1994) Biophys. J. 67, 2296–2304. 45. Jasper, J. R., Post, S. R., Desai, K. H., Insel, P. A., and Bernstein, D. (1995) J. Pharmacol. Exp. Ther. 274, 937–942. 46. Chilco, P. J., Gerardi, J. M., Kaczmarczyk, S. J., Chu, S., Leopold, V., and Zajac, J. D. (1993) Mol. Cell. Endocrinol. 94, 1–7. 47. Vitale, M. L., Seward, E. P., and Trifaro, J. M. (1995) Neuron 14, 353–363. 48. Sauman, I., and Berry, S. J. (1994) Eur. J. Cell Biol. 64, 348– 356. 49. Barja, F., and Turian, G. (1994) Cell Biol. Int. 18, 903–906. 50. Evangelisti, R., Becchetti, E., Baroni, T., Rossi, L., Arena, N., Valeno, V., Carinci, P., and Locci, P. (1995) Cell Biochem. Funct. 13, 41–52. 51. Aszalos, A., Pine, P. S., Weaver, J. L., and Rao, P. E. (1994) Cell. Immunol. 157, 81–91. 52. McCall, K., and Duncan, C. J. (1994) Med. Sci. Res. 22, 217– 218. 53. Vidyasagar, P. B. (1986) in Recent Advances in Cytochalasans (Pendse, G. S., Ed.), pp. 149–155, Balodyan Press, India. 54. Brandt, D. W., Pandol, S. J., and Deftos, J. L. (1991) Endocrinology 128, 2999–3004.
Received June 25, 1996 Revised version received December 16, 1996
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
89
eca
232, 79–89 (1997)
EX973481
Regulation of Parathyroid Hormone-Related Protein Secretion and mRNA Expression in Normal Human Keratinocytes and a Squamous Carcinoma Cell Line MICHELLE T. WECKMANN, ANDREA GRO¨NE, CHARLES C. CAPEN,
AND
THOMAS J. ROSOL1
Department of Veterinary Biosciences, The Ohio State University, Columbus, Ohio 43210
(HHM) [1]. HHM is a clinical syndrome resulting from the secretion of PTHrP by certain neoplasms which stimulates osteoclastic bone resorption and renal calcium reabsorption. Increased levels of circulating PTHrP bind to and stimulate parathyroid hormone (PTH) receptors in bone and kidney cells, mimicking the effects of excess PTH. The clinical syndrome of HHM can be reversed experimentally in nude mice with transplanted squamous carcinomas using neutralizing anti-PTHrP antibodies [2]. In addition to the production of PTHrP by human and animal neoplasms [3–5], PTHrP has been demonstrated in many normal tissues, including the epidermis and lactating mammary gland, kidney, bone, placenta, brain, and fetal tissues [6, 7]. In contrast to patients with HHM who have increased circulating concentrations of PTHrP, serum PTHrP in normal adults is very low (õ1 pM) [8]. This suggests that PTHrP functions as an autocrine or paracrine factor in most normal tissues [9]. The mechanisms regulating the secretion, expression, and biologic activity of PTHrP in most normal tissues are poorly understood and may vary between cell types. Normal human keratinocytes (NHFK) and a human squamous carcinoma cell line (A253) were selected for investigating the regulation of PTHrP secretion. Both cell types have been reported to produce PTHrP, and HHM occurs in certain patients with squamous carcinomas [6, 10–12]. In addition to their role in barrier formation and wound repair, keratinocytes function as secretory cells, with secretion of proteins, including cytokines, enzymes, adhesion molecules, and PTHrP [13]. Secretion of proteins occurs by a well-defined series of steps and may be regulated or constitutive. Secretion depends upon the coordinated function of many cellular components, including the cytoskeleton, intracellular environment, and GTP-binding proteins [14]. Cells may produce, modify, and direct both regulated and constitutive transport simultaneously [15, 16]. Monensin, cytochalasin B, colchicine, and GTPgS are agents which can modify secretion by targeting different parts of the secretory pathway. Monensin, an ionophore, pre-
Parathyroid hormone-related protein (PTHrP) has been identified as a causative factor in the pathogenesis of humoral hypercalcemia of malignancy (HHM). The regulation and mechanisms of PTHrP secretion in most normal and malignant cells are unknown. PTHrP secretion, mRNA expression, and transcription were measured in neoplastic human squamous carcinoma cells (A253) and normal human foreskin keratinocytes (NHFK) by radioimmunoassay, RNase protection assay, and transient transfections of the 5*-flanking region of human PTHrP in a luciferase expression vector. Mechanisms of PTHrP secretion were investigated using chemicals (monensin, colchicine, cytochalasin B, guanosine 5*-[g-thio]triphosphate (GTPgS)) that interfere with or facilitate intracellular transport. Monensin inhibited PTHrP secretion in both NHFK and A253 cells. Ultrastructurally, monensin caused dilatation of rough endoplasmic reticulum and the formation of numerous cytoplasmic secretory vacuoles in both cell lines. Colchicine decreased PTHrP production in NHFK cells and stimulated PTHrP production and mRNA levels in A253 cells. Colchicine also stimulated transcription of the PTHrP–luciferase reporter gene. Cytochalasin B stimulated PTHrP secretion and mRNA expression in A253 cells, but had no effect in NHFK cells. GTPgS had no effect on PTHrP secretion in either cell line. It was concluded that PTHrP secretion is dependent on the constitutive movement of secretory vesicles to the cytoplasmic membrane and regulation of PTHrP secretion and mRNA expression are altered in squamous carcinoma cells compared to normal human keratinocytes in vitro. q 1997 Academic Press
INTRODUCTION
Parathyroid hormone-related protein (PTHrP) was originally identified as a causative factor in the pathogenesis of humoral hypercalcemia of malignancy 1 To whom correspondence should be addressed at The Ohio State University, Department of Veterinary Biosciences, 1925 Coffey Road, Columbus, OH 43210. Fax: (614) 292-6473. E-mail: [email protected].
79
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
0014-4827/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
eca
80
WECKMANN ET AL.
vents the acidification of secretory vesicles and their subsequent transport to the cell membrane [17]. Colchicine and cytochalasin B impair the function of microtubules and microfilaments, respectively [18,19]. Guanosine 5*-(g-thio]triphosphate (GTPgS), a nonhydrolyzable form of GTP, activates certain GTP-binding proteins to enhance vesicular secretion [20]. Modification of PTHrP secretion by these chemicals may provide information concerning the mechanisms and regulation of PTHrP secretion. Understanding the control of PTHrP secretion and expression in both normal and neoplastic cells is important in determining how certain malignant cells are able to secrete biologically active PTHrP, resulting in elevated blood levels, and induce hypercalcemia. The purpose of this study was to investigate mechanisms of PTHrP secretion and expression in NHFK and A253 cells using monensin, colchicine, cytochalasin B, and GTPgS, which may interfere with or stimulate intracellular vesicular transport. MATERIALS AND METHODS Materials. Monensin (Calbiochem, San Diego, CA) and cytochalasin B (Sigma Chemical Co., St. Louis. MO) were dissolved in 100% ethanol at 1.0 mM and 5 mg/ml, respectively. Colchicine (Sigma) and lumicolchicine, an inactive analog for colchicine (Sigma), were dissolved in Hanks’ balanced salt solution (HBSS) at 1.0 mM. GTPgS was dissolved in water at 10 mM. All reagents were diluted to final concentration in cell culture medium prior to use. Cell culture. Human squamous carcinoma cells (A253) were obtained from the American Type Culture Collection (Rockford, MD) and grown in McCoy’s 5A media (Sigma) supplemented with 10% fetal bovine serum (FBS)(Life Technologies, Grand Island, NY), 2 mM L-glutamine, and 50 mg/ml gentamicin. Normal human foreskin keratinocytes (NHFK) were prepared from neonatal foreskins following a procedure described by Boyce and Ham [6, 21]. Briefly, neonatal foreskin tissues were cut into 2- to 5-mm2 pieces and incubated for 48 h in 0.2% trypsin (type III, Sigma) at 47C. The epidermis was separated from the dermis and dispersed in MCDB 153 medium (Sigma) containing 10% FBS. The cell suspension was centrifuged and the pellet was resuspended in keratinocyte serum-free basal medium (Life Technologies) containing low calcium (0.08 mM), epidermal growth factor (5 ng/ml), and bovine pituitary extract (35 ml/ ml), placed in 75-cm2 culture flasks, and incubated at 377C in 95% air/5% CO2 . The NHFK were used from passages 1–5 only. Conditioned medium collection. Conditioned medium (CM) was collected from cells grown in 12-well culture plates (Corning). Medium was collected 24 h following treatment or medium change. Conditioned media were stored at 0207C until assayed for PTHrP content by radioimmunoassay. Secretion studies. Treatments were added in 2 ml of medium to cells in triplicate wells of 12-well plates. The PTHrP secretion pattern was evaluated in each cell type over a 2-week culture period without any treatment. The cells were plated at 20% confluence, and medium was collected every 24 h. PTHrP secretion was maximal and stable in A253 cells after confluence was reached (ú 6 days in culture) and was maximal in NHFK near the time of confluence, but decreased in postconfluent cultures of NHFK. Secretion studies were conducted for 24 or 96 h, with CM collected and fresh medium (including treatment) added every 24 h. Experiments were begun near the time of confluence in cultures of NHFK (Days 3–5 after plating)
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
and after confluence in A253 cells (Days 6–7 after plating). In experiments with GTPgS, the cells were permeabilized with Trans-Port kit (Life Technologies) to permit uptake of GTPgS. PTHrP nonequilibrium radioimmunoassay (RIA). The RIA for PTHrP was performed as described by our laboratory using chicken polyclonal anti-PTHrP-(1-36) antibody [22]. [Tyr36]PTHrP-(1-36) (Bachem, Torrance, CA) was radioiodinated with I125 and Iodogen (Pierce Chemical Co, Rockford, IL) and purified using reverse-phase HPLC [22, 23]. Polyclonal chicken anti-PTHrP-(1-36) antibodies (10 mg/ml in PBS, 1:1000 dilution) were added to borosilicate tubes containing standards or medium samples (100 ml) and assay buffer (20 mM sodium phosphate, pH 7.4, 140 mM NaCl, 50 mM EDTA, 0.5% BSA, 0.1 sodium azide) and were incubated for 48 h at 47C. Iodinated PTHrP-(1-36) was added (10,000 cpm/tube) and dextrancoated charcoal (5 mg/ml activated charcoal, 0.5 mg/ml dextran) was utilized to separate the bound from the free PTHrP after 48 h. The supernatant was counted in a gamma-radiation counter and PTHrP content quantified by log-logit transformation using Securia 1.0 (Packard Instruments, Downers Grove, IL). The RIA was sensitive to 0.1 ng PTHrP-(1-36)/ml of medium. All assays included an internal standard of conditioned medium with a PTHrP concentration of 1 ng/ml. The maximum interassay variability was 25% and it was routinely 10– 15%. Isolation of total RNA. For total RNA isolation, cells were plated in 90-mm tissue culture dishes in fully supplemented medium. Total RNA was isolated by the method of Chomczynski and Sacchi [24, 25]. Briefly, the cells were lysed with 4 M guanidium thiocyanate, 25 mM sodium citrate, 0.5% Sarkosyl (pH 7.4), and 2-mercaptoethanol (BME) and either stored at 0207C or extracted three times with water-saturated phenol (pH 4.0) and chloroform:isoamyl alcohol (24:1). The RNA was precipitated with either isopropanol or 2 vol of 100% ethanol with 0.1 vol of 4 M sodium acetate, centrifuged at 30,000g for 15 min, resuspended in diethyl polycarbonate (DEP)treated water, and quantified at A260/280. RNase protection assay (RPA). Total cellular PTHrP mRNA was measured by RPA (Ambion, Austin, TX). Complementary DNA riboprobes labeled with [32P]UTP (Dupont, New England) for human PTHrP mRNA (No. 661) or human glyceraldehyde-3-phosphate dehydrogenase (pTRI-GAPDH, Ambion) were utilized. GAPDH is a constitutively expressed gene in keratinocytes and was used as a loading control to standardize the amounts of RNA present [26, 27]. The antisense GAPDH template contains a 316-bp fragment of the human GAPDH gene derived from exons 5–8 in the pTRIPLEscript transcription vector [28]. The human cDNA riboprobe (661) is a 537bp fragment of human PTHrP kindly obtained from Genentech, Inc. (South San Francisco, CA). The plasmid contains a ClaI–SstI fragment of PTHrP cDNA clone BRF.52. The fragment was cloned between the AccI and SstI sites of the vector pSP65 [29, 30]. The cDNA riboprobes were linearized by digestion with HindIII and transcribed using T7 or SP6 RNA polymerase (Maxiscript In Vitro Transcription kit, Ambion). The RPA was performed as follows. Total RNA from either A253 squamous carcinoma cells (40 mg) or NHFK cells (20 mg) was coprecipitated with [32P]UTP-labeled RNA probes for human PTHrP and GAPDH. After coprecipitation, the RNA was centrifuged, washed in ethanol, dried, resuspended in hybridization buffer, heated to 957C for 3 min, and allowed to hybridize overnight at 427C. The unhybridized RNA was digested with an A/T1 RNase mixture and the doublestranded RNA was precipitated. The double-stranded RNA was resuspended in loading buffer (80% formamide, 1 mM EDTA, pH 8.0, 0.1% bromophenol blue, 0.1% xylene cyanol), heated to 957C, and separated using a 30% urea/8% acrylamide/bis gel at 300 V for 2 h. The gel was exposed to X-ray film with an intensifying screen at 0807C for 8 to 24 h. PTHrP signal intensity was quantified using a laser densitometer (ImageQuant Series 300, Molecular Dynamics, Sunnyvale, CA) and was normalized to GAPDH signal intensity.
eca
81
PTHrP REGULATION IN KERATINOCYTES Transient transfection of the PTHrP gene/luciferase construct. To examine the effects of monensin, cytochalasin B, and colchicine on PTHrP gene expression, normal keratinocytes and A253 squamous carcinoma cells were transfected with plasmids containing portions of the 5*-flanking sequence of the human PTHrP gene ligated to a promoterless bacterial luciferase gene (pGL2-Basic from Promega, Madison, WI). The PTHrP construct consisted of 950 bp of 5*-flanking DNA and included the untranslated exons III and IV and the 5* end of exon V, 20 bp short of the ATG translation start site. The plasmid (pSMR38) with the 5*-flanking sequence contained the P2 and P3 PTHrP promoters and was generously provided by Dr. Larry Suva (Beth Israel Hospital, Boston, MA) [31]. Transfections of the NHFK were conducted in 12-well culture plates at 60% confluency using lipofectin (Life Technologies). Transfections of the A253 cells were conducted using polybrene (Sigma) following the method described by Jiang et al. [32]. Cells were treated 48 h after transfection with supplemented medium containing serum, monensin, cytochalasin B, colchicine, or lumicolchicine. To determine the transfection efficiency of each sample, cells were cotransfected with the CMV– b-galactosidase plasmid (Promega), containing the b-galactosidase gene driven by a CMV (cytomegalovirus) promoter. b-Galactosidase was assayed in cell extracts using the b-Galactosidase Enzyme Assay System (Promega). Cell extracts where assayed for luciferase using the Luciferase Assay System (Promega) and a liquid scintillation counter (Packard). Immunofluorescence of actin and tubulin. To investigate the effects of cytochalasin B and colchicine on the cell cytoskeleton, immunofluorescence for actin and tubulin was performed on NHFK and A253 cells. Normal keratinocytes and A253 cells were grown on Microprobe slides (Fischer Scientific) and treated with either cytochalasin B (5 mg/ml) or colchicine (0.1 mM) for 4 h. Actin was visualized by direct immunofluorescence using fluorescein phalloidin (Molecular Probes Inc., Eugene, OR). The fixation procedure consisted of immersion in acetone at 0207C for 4 min after incubation in 3.7% formaldehyde for 20 min. The cells were then washed twice with DPBS (Dulbecco’s phosphate-buffered saline), incubated for 20 min at room tem-
perature with fluorescein phalloidin at a dilution of 1:40 (1 unit) in DPBS, and washed three times with DPBS. Tubulin was stained by indirect immunofluorescence using rat anti-b-tubulin and anti-rat IgG FITC conjugate (Harlan Sera-Ral Ltd., Sussex, England). The cells were fixed for 20 min at 0207C in methanol and permeabilized for 1 min at 0207C with acetone, washed three times in DPBS, incubated 1 h with a 1:20 dilution of the primary antibody, washed three times with DPBS, incubated 30 min with a 1:20 dilution of secondary antibody, and washed three times with DPBS. The cells were analyzed using an Olympus epifluorescent microscope at 10001 magnification with an oil immersion objective and photomicrographs were obtained with Kodak T-max (ASA 400) black and white film. Electron microscopy. A253 or NHFK cells were grown on 30-mm plates and treated with monensin (0.1–1 mM) for 24, 48, 72, or 96 h for evaluation by transmission electron microscopy. The cells were fixed in 3% glutaraldehyde, postfixed in 1 M cacodylate buffer and osmium, embedded in medcast plastic (Ted Pella Redding, CA), and examined with a Philips 300 electron microscope. Statistical analysis. Numerical data were expressed as means { the standard error of the mean (SEM) for triplicate samples. PTHrP concentration of the conditioned medium was expressed as ng/ml. Statistical differences between means for control and treatment groups were evaluated with Instat 1.0 (Graph PAD software, San Diego, CA) using one-way analysis of variance (ANOVA) and Tukey’s mean separation test, with the level of significance at P õ 0.5, õ 0.01, or õ 0.001.
RESULTS
Secretion of PTHrP by NHFK and A253 Cells To determine the secretory pattern of PTHrP in NHFK and A253 cells in monolayer culture, PTHrP concentration of the medium was measured at 24-h
FIG. 1. Secretion of PTHrP by NHFK and A253 squamous carcinoma cells in vitro. Conditioned medium was collected at 24-h intervals, with cells reaching confluence at Days 4–5. Data for the NHFK and A253 cells represent means { SE (n Å 3).
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
eca
82
WECKMANN ET AL.
FIG. 2. Effects of monensin on PTHrP secretion. NHFK and A253 squamous carcinoma cells were treated with monensin (0.1, 0.3, and 1 mM) for 96 h, and medium was collected every 24 h. Data for NHFK (A) and A253 squamous carcinoma cells (B) represent means { SE (n Å 3). *Significantly different (P õ 0.001) from control.
intervals for 14 days, starting when the cells were 20% confluent. Secretion of PTHrP by normal keratinocytes increased until confluency (Day 4), remained high for 2 days near confluence (Days 5 and 6), and then slowly declined until Day 14 (Fig. 1). Secretion of PTHrP in A253 cells followed a different pattern than did NHFK. PTHrP secretion was low prior to confluence at Day 4, after which secretion increased and remained stable for Days 6–14 (Fig. 1).
eter. In the NHFK there were mild reductions in PTHrP mRNA in monensin-treated cells at 1 h (1.5fold reduction) and 12 h (Fig. 3A). PTHrP mRNA in
Effect of Monensin on PTHrP Secretion, mRNA Levels, and Transcription In order to determine whether changes in vesicle acidity altered PTHrP secretion in normal and neoplastic human cells, NHFK and A253 cells were treated with monensin for up to 96 h. PTHrP secretion by NHFK and A253 cells was significantly (P õ 0.001) inhibited by monensin treatment (0.1 to 1 mM) at 24 to 96 h (Figs. 2A and 2B). The toxicity of monensin resulted in cell death in concentrations greater than 1 mM. Steady-state levels of PTHrP mRNA were measured to determine if monensin decreased expression of PTHrP mRNA in addition to impairing the release of secretory vesicles. PTHrP mRNA was detectable at all time points (1, 6, and 12 h) with the lowest levels at 12 h. All of the lanes were standardized with the GAPDH mRNA loading control using a laser densitom-
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
FIG. 3. The effect of monensin on the steady-state levels of PTHrP and GAPDH mRNA was analyzed by RNase protection assay. NHFK (A) and A253 squamous carcinoma cells (B) were treated with monensin (0.3 mM) or control for 1, 6, and 12 h. Top band, PTHrP probe; lower band, GAPDH probe: T, monensin treatment; C, control.
eca
PTHrP REGULATION IN KERATINOCYTES
83
FIG. 4. Ultrastructure of NHFK and A253 squamous carcinoma cells. Control cultures of NHFK (A) demonstrating abundant free polyribosomes and rough endoplasmic reticulum in the cytoplasm. Control cultures of A253 squamous carcinoma cells (B) demonstrating multiple desmosomes and free polyribosomes in the cytoplasm. Treatment of NHFK (C) with monensin resulted in the presence of dilated secretory vacuoles and endoplasmic reticulum. Treatment of A253 cells (D) with monensin resulted in the accumulation of enlarged secretory vacuoles in the cytoplasm. R, rough endoplasmic reticulum; V, secretory vacuoles; D, desmosomes; N, nucleus. (A) Bar, 0.81 mm; (B) bar, 0.94 mm; (C) bar, 0.95 mm; (D) bar, 1.2 mm.
the A253 cells exhibited a mild decrease in PTHrP mRNA at 12 h and no change at 1 or 6 h (Fig. 3B). Monensin had no significant effect on expression of the PTHrP–luciferase reporter gene in NHFK or A253 cells (data not shown). Effects of Monensin on the Ultrastructure of NHFK and A253 Cells The ultrastructure of NHFK demonstrated a network of intermediate filaments, microtubules, and an abundance of both free polyribosomes and rough endo-
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
plasmic reticulum (Fig. 4A). The A253 cells had multiple desmosomes with anchoring intermediate filaments and an abundance of free polyribosomes (Fig. 4B). Monensin caused dilatation of rough endoplasmic reticulum in the NHFK and dilation of cytoplasmic secretory vacuoles in both the NHFK and A253 cells (Figs. 4C and 4D). Cytoplasmic vacuoles were seen as early as 12 h by electron microscopy in cells treated with 0.3 mM monensin. The number and size of the vacuoles were greater in the A253 cells than in the NHFK. Aggregates of cellular organelles could be seen in the A253
eca
84
WECKMANN ET AL.
FIG. 5. Effect of colchicine on PTHrP secretion. NHFK were treated with colchicine for 96 h, and conditioned medium was collected daily. Data represent means { SE (n Å 3). *Significantly different (P õ 0.01) from control.
cells treated with monensin. Intact membranes were present around the perimeter of some vacuoles. The vacuoles were present at 12, 24, 48, 72, and 96 h in monensin-treated cells. There was an increase in the number of vacuoles from 12 to 24 h, but no increase in vacuoles from 24 to 96 h. Effect of Colchicine on PTHrP Secretion, mRNA Levels, and Transcription The effects of colchicine were evaluated to determine whether disruption of cytoplasmic microtubules would affect PTHrP secretion. Colchicine significantly (P õ 0.01) inhibited PTHrP secretion at concentrations of 0.1, 0.3, and 1 mM at 24 to 96 h in NHFK (Fig. 5). In contrast, colchicine-treated A253 cells had a significant increase in PTHrP production at doses ranging from 0.1 to 1 mM at 24 to 96 h (Fig. 6). Lumicolchicine, an inactive analog of colchicine which does not disrupt microtubules [33], was used as a control. Lumicolchicine treatment (0.1, 0.3, 1 mM) had no effect on PTHrP secretion in either the NHFK or A253 cells (data not shown). Colchicine had a mild stimulatory effect on steadystate levels of PTHrP mRNA in NHFK at 12 h (40% increase) and at 24 h (20% increase) (Fig. 7A). In contrast, colchicine markedly increased PTHrP mRNA in A253 cells at 6 (2.2-fold), 12 (1.6-fold), and 24 h compared to the control cells (Fig. 7B). To determine whether colchicine altered transcription of PTHrP, a segment of the 5*-flanking DNA of the human PTHrP
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
FIG. 6. Effect of colchicine on PTHrP secretion. A253 squamous carcinoma cells were treated with colchicine for 96 h, and conditioned medium was collected daily. Data represent means { SE (n Å 3). *Significantly different (P õ 0.05) from control.
gene containing promoters P2 and P3 was coupled to a luciferase expression vector and used to transiently transfect normal keratinocytes and A253 cells. Colchicine resulted in a 3-fold increase in transcription of the PTHrP–luciferase transcript in A253 cells and a 50%
FIG. 7. Effect of colchicine on steady-state levels of PTHrP and GAPDH mRNA was analyzed by RNase protection assay. NHFK (A) and A253 squamous carcinoma cells (B) were treated with colchicine (1 mM) or control for 1, 6, 12, and 24 h. Top band, PTHrP probe; lower band, GAPDH probe; T, colchicine treatment; C, control.
eca
85
PTHrP REGULATION IN KERATINOCYTES
had a variable response to cytochalasin B treatment (5 mg/ml), with a slight increase in mRNA at 1 h and a slight decrease (1-fold) at 6 h when standardized to the GAPDH controls (Fig. 11A). There was no significant change in PTHrP mRNA levels in the NHFK at 12 h. PTHrP mRNA was markedly increased by cytochalasin B treatment in the A253 cells at 6 (6-fold) and mildly increased at 12 h (Fig. 11B). Cytochalasin B had no significant effect on expression of the PTHrP–luciferase reporter gene in NHFK or A253 cells (data not shown). Effects of Cytochalasin B on Microfilaments in NHFK and A253 Cells
FIG. 8. Transient transfection of A253 squamous carcinoma cells with human 5*-flanking PTHrP DNA in a luciferase expression vector and the effects of 24 h treatment with colchicine, lumicolchicine, or 10% serum were measured. Data represent means { SE (n Å 3). *Significantly different (P õ 0.05) from control.
increase in NHFK. However, there was no effect of the inactive analog, lumicolchicine (Fig. 8). Effects of Colchicine on Microtubules in NHFK and A253 Cells Indirect immunofluorescence was utilized to evaluate the effects of colchicine on microtubules in NHFK and A253 cells. The NHFK and A253 cells arranged their microtubules into a network of cytoplasmic arrays that spanned radially from the nucleus to the cell’s border (Figs. 9A and 9B). The network of microtubules in the cells treated with colchicine collapsed into large aggregates in the cytoplasm, coiled into juxtanuclear caps in the A253 cells, and were entirely disrupted in the NHFK (Figs. 9C and 9D). Effect of Cytochalasin B on PTHrP Secretion, mRNA Levels, and Transcription The effect of cytochalasin B was investigated to determine whether microfilaments were necessary for normal PTHrP secretion. In the NHFK, cytochalasin B mildly inhibited PTHrP production at 24 h while significantly increasing (P õ 0.001) PTHrP secretion after 24 h in the A253 cells (Fig. 10). Cytochalasin B also had an inhibitory affect on cell growth at all concentrations. In addition, cytochalasin B (1 to 10 mg/ ml) significantly increased PTHrP production at 48, 72, and 96 h in the A253 cells (data not shown). Steadystate levels of PTHrP mRNA in normal keratinocytes
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
Direct immunofluorescence was utilized to visualize the effect of cytochalasin B on cytoskeleton microfilaments of normal keratinocytes and A253 cells. The cell shape remained unchanged in both the control and cytochalasin B-treated NHFK and A253 cells. The control groups of both cell types had strong cytoskeletal staining, with a discernable cellular and nuclear outline. Actin filaments were visible in the cytoplasm and along the cytoplasmic membrane (Figs. 12A and 12B). In contrast, the cells treated with cytochalasin B (5 mg/ml) for 4 h had a patchy appearance. There were areas of multiple small and large foci of positive immunofluorescense indicating that cytochalasin B induced depolarization and condensation of the actin filaments (Figs. 12C and 12D). Effects of GTPgS on PTHrP Secretion The A253 and NHFK cells were permeabilized to permit entry of GTPgS into the cytoplasm. The permeabilization process and treatment resulted in up to 100% loss of viability in the A253 cells. The NHFK showed a 20% loss in viability after permeabilization. The uptake and treatment of normal keratinocytes by GTPgS did not affect secretion of PTHrP. DISCUSSION
The mechanisms and regulation of PTHrP secretion are poorly understood. This investigation demonstrated that the regulation of PTHrP secretion and mRNA expression varied between NHFK and A253 squamous carcinoma cells. It has been reported that PTHrP secretion may be regulated or constitutive, depending on cell type [14, 34–36]. In addition, PTHrP is secreted in multiple forms including N-terminal and midregion fragments [37]. Deftos et al. and Ilardi et al. demonstrated PTHrP in secretory granules of atrial myocytes and in the cytoplasm of squamous carcinoma cells, respectively [35, 38]. Rizzoli et al. have reported that PTHrP-(1-34) is secreted via the regulated pathway in a human lung squamous carcinoma cell line
eca
86
WECKMANN ET AL.
FIG. 9. Effects of colchicine on microtubule formation in NHFK and A253 cells. Control cultures of NHFK (A) and A253 squamous carcinoma cells (B) showing the presence of an intact microtubule network. Colchicine resulted in the disruption of the microtubules in NHFK (C) and A253 cells (D).
(BEN) [34]. Using transfected cell lines, it was concluded that PTHrP was secreted in a regulated manner in neuroendocrine cells (e.g., insulinoma cells) and in a constitutive manner in non-neuroendocrine cells (e.g., squamous carcinoma cells and fibroblasts) [36]. Monensin inhibited PTHrP secretion in both NHFK and A253 cells. This was interpreted to be a result of impaired vesicular transport, as evidenced from dilated endoplasmic reticulum and secretory vesicles. In addition, monensin mildly decreased PTHrP mRNA levels. It has been reported that PTHrP secretion was inhibited by monensin in canine and human squamous carcinoma cells [10, 39]. Both the normal and neoplastic keratinocytes responded to monensin in a similar manner, indicating that the mechanisms of PTHrP secretion which involve vesicular acidity function similarly in both the normal keratinocytes and the A253 squamous carcinoma cells. The microtubule disrupter, colchicine, has been reported to inhibit regulated secretion of various proteins in leukocytes, mast cells, adrenal cells, pituitary cells, and parotid cells [40]. We have demonstrated that colchicine inhibited secretion of PTHrP in normal kera-
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
FIG. 10. Effect of cytochalasin B on PTHrP secretion. NHFK and A253 squamous carcinoma cells were treated with cytochalasin B for 24 h. Data represent means { SE (n Å 3). *Significantly different (P õ 0.001) from control.
eca
PTHrP REGULATION IN KERATINOCYTES
FIG. 11. Effect of cytochalasin B on steady-state levels of PTHrP and GAPDH mRNA was analyzed by RNase protection assay. NHFK (A) and A253 squamous carcinoma cells (B) were treated with cytochalasin B (5 m g/ml) or control for 1, 6, 12, and 24 h. Top band, PTHrP probe; lower band, GAPDH probe; T, cytochalasin B treatment; C, control.
87
tinocytes. Unexpectedly, colchicine increased PTHrP secretion and mRNA expression in the A253 squamous carcinoma cells. Microtubules are necessary for regulated secretion in certain cell types [41]. Colchicine inhibited PTHrP secretion in NHFK even though colchicine induced a mild increase in PTHrP mRNA levels and reporter gene expression. This suggests that microtubules were necessary for the secretion of PTHrP in normal keratinocytes. However, in the neoplastic cells (A253), colchicine increased PTHrP secretion, transcription, and steady-state mRNA levels. This demonstrated that PTHrP secretion was not dependent on microtubules in A253 cells. Colchicine has been shown to increase mRNA production of Pro-IL-7b and pituitary adenylate cyclaseactivating peptide gene [42, 43]. Colchicine increased PTHrP mRNA expression and transcription in the A253 cells and NHFK. Colchicine disrupts microtubule structure and stimulates intracellular cyclic AMP production [40, 44, 45]. This is a potential mechanism of increased PTHrP mRNA expression in the A253 squa-
FIG. 12. Effects of cytochalasin B on microfilament formation in NHFK and A253 squamous carcinoma cells. Actin filaments are visible in untreated NHFK (A) and A253 (B) cells. Cytochalasin B treatment induced depolarization and condensation of the actin filaments in the NHFK (C) and A253 cells (D).
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
eca
88
WECKMANN ET AL.
mous carcinoma cells, since calcitonin has been reported to increase PTHrP mRNA in a human squamous carcinoma cell line (BEN) by increasing intracellular cAMP and activating a cAMP-responsive element in the PTHrP gene [46]. Cytochalasin B has been reported to increase regulated secretion of various proteins in leucocytes, macrophages, rat pancreatic islets, and chromaffin cells by breaking down the actin web which prevents secretory granules from approaching the cell membrane [40, 47]. In addition, cytochalasin D, a microfilament disrupter, enhanced basal secretion of PTHrP in human squamous carcinoma cells (BEN) by release of stored protein [34]. Cytochalasin B increased PTHrP secretion in A253 cells and mildly inhibited PTHrP secretion in the NHFK, illustrating a second difference in the regulation of PTHrP secretion between normal and neoplastic keratinocytes. Cytochalasin B disrupts microfilaments and actin formation, which can alter secretion and nuclear RNA transport [40, 48–50]. Increased PTHrP secretion in the A253 squamous carcinoma cells was likely due to upregulation of PTHrP mRNA levels. Cytochalasin has been reported to increase intracellular calcium concentrations in T-lymphocytes and mouse soleus muscle [51–53]. It is possible that cytochalasin B increased PTHrP mRNA production in A253 cells by increasing cytosolic calcium concentration and activating the phospholipase C/protein kinase C pathways [54]. Activation of these intracellular regulatory pathways increased transcription of PTHrP mRNA in a human squamous carcinoma cell line (BEN) [16]. It is unknown whether PTHrP is secreted in either a regulated or a constitutive manner or by both mechanisms in normal keratinocytes and squamous carcinoma cells. Plawner et al. has reported that neuroendocrine cells secreted PTHrP in a regulated manner and squamous carcinoma cells in a constitutive manner [36]. Since disruption of the actin web with cytochalasin B did not increase PTHrP secretion in NHFK, constitutive secretion is most likely in these cells. The lack of secretory granules and the high basal levels of PTHrP secreted into the medium in both the untreated normal keratinocytes and A253 squamous carcinoma cells also suggests that PTHrP is secreted constitutively in both of these cell types. In the A253 cells, cytochalasin B and colchicine both increased PTHrP secretion and steady-state mRNA levels. It is likely that increased PTHrP synthesis due to cytochalasin B was the result of a posttranscriptional effect (such as increased mRNA stability) since there was no effect of cytochalasin B on expression of the PTHrP–luciferase reporter gene. Further studies will be needed to determine whether PTHrP secretion follows a regulated pathway in keratinocytes in addition to the constitutive pathway.
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
In conclusion, PTHrP secretion in normal keratinocytes was dependent on microtubules and was constitutively secreted. In contrast, PTHrP secretion in the A253 squamous carcinoma cells was influenced by increased steady-state PTHrP mRNA levels and did not depend on the presence of intact microfilaments or microtubules. However, the A253 cells are poorly differentiated and may not be representative of all squamous carcinomas. The regulation of PTHrP secretion was altered in neoplastic keratinocytes. This may be a result of changes which occurred during transformation and may be important in the autonomous overproduction and secretion of PTHrP that occurs in the pathogenesis of humoral hypercalcemia of malignancy. We thank Dr. James R. DeWille for his collaboration with the ribonuclease protection assays. The investigation was supported by the National Institute of Arthritis, Musculoskeletal Diseases, and Skin Diseases (AR40220, AR01923).
REFERENCES 1. Rosol, T. J., and Capen, C. C. (1992) Lab. Invest. 67, 680–702. 2. Kukreja, S. C., Rosol, T. J., Wimbiscus, S. A., Shevrin, D. H., Grill, V., Barengolts, E. I., and Martin, T. J. (1990) Endocrinology 127, 305–310. 3. Rosol, T. J., and Capen, C. C. (1988) Dom. Anim. Endocrinol. 5, 1–21. 4. Rosol, T. J., Capen, C. C., Danks, J. A., Suva, L. J., Steinmeyer, C. L., Hayman, J. A., Ebeling, P. R., and Martin, T. J. (1990) Vet. Pathol. 27, 89–95. 5. Moseley, J. M. and Gillespie, M. T. (1995) Crit. Rev. Clin. Lab. Sci. 32(3), 299–343. 6. Werkmeister, J. R., Merryman, J. I., McCauley, L. K., Horton, J. E., Capen, C. C., and Rosol, T. J. (1993) Exp. Cell Res. 208, 68–74. 7. Philbrick, W. M., Wysolmerski, J. J., Galbrath, S., Holt, E., Orloff, J. J., Yang, K. H., Vasavada, R. C., Weir, E. C., Broadus, A. E., and Stewart, A. F. (1996) Physiol. Rev. 76, 127–173. 8. Pandian, M. R., Morgan, C. H., Carlton, E., and Segre, G. V. (1992) Clin. Chem. 38(2), 282–288. 9. Stewart, A. F., and Broadus, A. E. (1990) J. Clin. Endocrinol. Metab. 71, 1410–1414. 10. Merryman, J. I., Capen, C. C., McCauley, L. K., Werkmeister, J. R., Suter, M. M., and Rosol, T. J. (1993) Lab. Invest. 69, 347– 354. 11. Orloff, J. J., Ganz, M. B., Ribaudo, A. E., Burtis, W. J., Reiss, M., Milstone, L. M., and Stewart, A. F. (1992) Am. Physiol. Soc. 92, 599–607. 12. Kremer, R., Karaplis, A. C., Henderson, J., Gulliver, W., Banville, D., Hendy, G. N., and Goltzman, D. (1991) J. Clin. Invest. 87, 884–893. 13. Boyce, S. T. (1994) J. Invest. Dermatol. 102, 8–10. 14. Kelly, R. B. (1985) Science 230, 25–31. 15. Burgess, T. L., and Kelly, R. B. (1987) Annu. Rev. Cell Biol. 3, 243–293. 16. Emly, J. F., Hughes, S., Green, E., and Ratcliffe, W. A. (1994) Biochim. Biophys. Acta 1220, 193–198. 17. Nagatsu, A., Takahashi, T., Isomura, M., Nagai, S., Ueda, T.,
eca
PTHrP REGULATION IN KERATINOCYTES Murakami, N., Sakakibara, J., and Hatano, K. (1994) Chem. Pharm. Bull. 42, 2269–2275. 18. Bhisey, A. N. (1986) in Recent Advances in Cytochalasans (Pendse, G. S., Ed.), pp. 157–179, Balodyan Press, India. 19. Vandecandelaere, A., Martin, S. R., Schilstra, M. J., and Bayler, P. M. (1994) Biochemistry 33, 2792–2801. 20. Gomperts, B. D., and Tatham, P. E. R. (1988) Cold Spring Harbor Symp. on Quant. Biol. 53, 983–992. 21. Boyce, S. T., and Ham, R. G. (1985) J. Tissue Cult. Methods 9, 83–93. 22. Rosol, T. J., Steinmeyer, C. L., Merryman, J. I., Werkmeister, J. R., Groene, A., Weckmann, M. T., Swayne, D. E., and Capen, C. C. (1993) Vet. Immunol. Immunopathol. 35, 321–337. 23. McCauley, L. K., Rosol, T. J., Merryman, J. I., and Capen, C. C. (1992) Endocrinology 130, 300–306. 24. Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156–160. 25. Merryman, J. I., DeWille, J. W., Werkmeister, J. R., Capen, C. C., and Rosol, T. J. (1994) Endocrinology 134, 2424–2430. 26. Dveksler, G. A., Basile, A., and Dieffenbach, C. (1992) PCR Methods Appl. 1, 283–285. 27. Bosma, P., and Kooistra, T. (1991) J. Biol. Chem. 266, 17845– 17846. 28. Ercolani, L., Florence, B., Denaro, M., and Alexander, M. (1988) J. Biol. Chem. 263, 15335–15341. 29. Suva, L. J., Winslow, G. A., Wettenhall, R. E., Hammonds, R. G., Moseley, J. M., Diefenbach-Jagger, H., Rodda, C. P., Kemp, B. E., Rodriguez, H., Chen, E. Y., Hudson, P. J., Martin, T. J., and Wood, W. I. (1987) Science 237, 893–896. 30. Melton, D. A., Kreig, P. A., Rebagliati, M. R., Maniatis, T., Zinn, K., and Green, M. R. (1984) Nucleic Acids Res. 12, 7035–7057. 31. Kiriyama, T., Gillespie, M. T., Glatz, J. A., Fukumoto, S., Moseley, J. M., and Martin, T. J. (1993) Mol. Cell. Endocrinol. 92, 55–62. 32. Jiang, C. K., Connolly, D., and Blumenburg, M. (1991) J. Invest. Dermatol. 97, 969–973. 33. Chanard, J., Black, R., Purkerson, M., Lewis, J., Klahr, S., and Slatopolsky, E. (1977) Endocrinology 101, 1792–1800. 34. Rizzoli, R., Aubert, M. L., Sappino, A. P., and Bonjour, J. P. (1994) Int. J. Cancer 56, 422–426.
35. Deftos, L. J., Burton, D. W., and Brandt, D. W. (1993) J. Clin. Invest. 92, 727–735. 36. Plawner, L. L., Philbrick, W. M., Burtis, W. J., Broadus, A. E., and Stewart, A. F. (1995) J. Biol. Chem. 270, 14078–14084. 37. Soifer, N. E., Dee, K. E., Insogna, K. L., Burtis, W. J., Matovcik, L. M., Wu, T. L., Milstone, L. M., Broadus, A. E., Philbrick, W. M., and Stewart, A. F. (1992) J. Biol. Chem. 267, 18236– 18243. 38. Ilardi, C. F., and Faro, J. C. (1985) Arch. Pathol. Lab. Med. 109, 752–755. 39. Lo¨wik, C. W., Hoekman, K., Offringa, R., Groot, C. G., Hendy, G. N., Papapoulos, S. E., and Ponec, M. (1992) J. Invest. Dermatol. 98, 198–203. 40. Allison, A. C., and Davies, P. (1974) Adv. Cytopharm. 2, 237– 248. 41. Robin, P., Rossignol, B., and Raymond, M. N. (1995) Eur. J. Cell Biol. 67, 227–237. 42. O’Brien, J. M., Wewers, M. D., Moore, S. A., and Allen, J. A. (1995) J. Immunol. 154, 4113–4122. 43. Hannibal, J., Mikkelsen, J. D., Fahrenkrug, J., and Larsen, P. J. (1995) Endocrinology 136(9), 4116–4124. 44. Galli, A., and DeFelice, L. J. (1994) Biophys. J. 67, 2296–2304. 45. Jasper, J. R., Post, S. R., Desai, K. H., Insel, P. A., and Bernstein, D. (1995) J. Pharmacol. Exp. Ther. 274, 937–942. 46. Chilco, P. J., Gerardi, J. M., Kaczmarczyk, S. J., Chu, S., Leopold, V., and Zajac, J. D. (1993) Mol. Cell. Endocrinol. 94, 1–7. 47. Vitale, M. L., Seward, E. P., and Trifaro, J. M. (1995) Neuron 14, 353–363. 48. Sauman, I., and Berry, S. J. (1994) Eur. J. Cell Biol. 64, 348– 356. 49. Barja, F., and Turian, G. (1994) Cell Biol. Int. 18, 903–906. 50. Evangelisti, R., Becchetti, E., Baroni, T., Rossi, L., Arena, N., Valeno, V., Carinci, P., and Locci, P. (1995) Cell Biochem. Funct. 13, 41–52. 51. Aszalos, A., Pine, P. S., Weaver, J. L., and Rao, P. E. (1994) Cell. Immunol. 157, 81–91. 52. McCall, K., and Duncan, C. J. (1994) Med. Sci. Res. 22, 217– 218. 53. Vidyasagar, P. B. (1986) in Recent Advances in Cytochalasans (Pendse, G. S., Ed.), pp. 149–155, Balodyan Press, India. 54. Brandt, D. W., Pandol, S. J., and Deftos, J. L. (1991) Endocrinology 128, 2999–3004.
Received June 25, 1996 Revised version received December 16, 1996
AID
ECR 3481
/
6i1d$$$141
03-13-97 08:17:22
89
eca