JOURNAL OF BONE AND MINERAL RESEARCH Volume 5, Number 8 , 1990 Mary Ann Liebert, Inc., Publishers
Subclones of a Rat Parathyroid Cell Line (PT-r): Regulation of Growth and Production of Parathyroid Hormone-Related Peptide (PTHRP) KAZUSHIGE SAKAGUCHI,' KYOJI IKEDA,' FRANCESCO CURCIO,' GERALD D. AURBACH,' and MARIA LUISA BRANDI'
ABSTRACT Four subclones from a rat parathyroid cell line (PT-r cell) have been isolated, and morphological and functional characteristics have been examined. Subclones 1 and 2 display a polygonal shape, show growth and secretory responses to calcium (half-maximal suppressions at 1.2 and 1.7 mM, respectively), and respond to secretin with CAMP production (14.5-fold and 16.9-fold over basal) and hormone secretion (41 and 58% over basal). Subclone 4 is elongated in form and does not respond to calcium or secretin. Subclone 3 shows mixed morphology, elongated and polygonal shapes, with moderate response to calcium (half-maximal suppression at 1.7 mM) and secretin (CAMP, 3.2-fold increase and hormone secretion, 50% increase over basal). The clones were tested for content of messenger RNA (mRNA) representing parathyroid hormone (PTH) and parathyroid hormone-related peptide (PTHRP). Only PTHRP mRNA was found. The peptide released is virtually all PTHRP. PTH mRNA was not detected even with a sensitive RNA probe. The amount of mRNA for PTHRP closely paralleled the amount of PTH-like bioactivity released into the medium from each clone (144.7 + 12.1, 110.0 + 12.9, 68.0 + 5.6, and 39.9 + 2.4 pgEq of rat PTH-(-34) per lo' cells per 12 h in a medium with 0.7 mM ionized calcium, from subclones 1, 2, 3, and 4, respectively). Culture conditions, low-density passage (less than 150 split ratio) o r high-density passage (greater than 1:lO split ratio), affected morphology and function of the clones 1 and 2. They became elongated and functionally dedifferentiated like subclone 4 after 3 months of high-density culture. Upon dedifferentiation, morphological and functional properties were not reversible with high calcium, 1,25-(OH)*D3,or retinoic acid. These clones may be useful for studies of phenotypic expression, calcium regulation, and gene expression and interrelationships among these functions.
INTRODUCTION a clonal cell line obtained from rat hyperplastic parathyroid tissue.'" This cell line, PT-r, maintained differentiated characteristics through 7 months of serial passages-hormone synthe-
W
E HAVE RECENTLY DEVELOPED
sis, regulation of hormone secretion and growth by calcium, and modulation of secretion by secretin. We recently discovered that the peptide produced and secreted is virtually all parathyroid hormone-related peptide (PTHRP), not parathyroid hormone (PTH) itself.'z1 We have now isolated f o u r subclones (PT-r 1 , 2, 3, and
'Metabolic Diseases Branch, National Institute of Diabetes and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, M D 20892. 'Division of Endocrinology, Department of Medicine, Yale University, New Haven, CT 065 10. Present address: Department of Internal Medicine, University of Tokyo, Tokyo, Japan. 'Laboratory of Cell Biology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892. Present address: Instituto Di Scienze Mediche, Facolta Di Medicina Di Udine, Italy. 'Present address: Florence University School of Medicine, Florence, Italy.
863
864
SAKAGUCHI ET AL.
4) displaying varied morphological characteristics. In this report we correlate the morphology of these clones with growth, secretory, and calcium regulatory properties.
substances. At appropriate times, media were removed by aspiration, centrifuged to remove cell debris, and stored at -20°C until used for bioassay. Cell number was counted in each dish. Control medium maintained in the incubator at 37°C was also assayed, and the control values were subMATERIALS AND METHODS tracted from those for conditioned media. Ionized calcium Cell culture, subcloning, and chromosomal analysis was determined using a Nova 2 ionized calcium analyzer (Nova Biomedical, Waltham, MA). PTH-like bioactivity was measured by intracellular cAMP The culture medium used for growing PT-r cells and subclones was a calcium- and magnesium-free mixture response of ROS 17/2.8 cells according to the method de(1 :1) of Coon’s modified Ham’s F-12 and Dulbecco’s modi- scribed by Pines et aLL3’NIH Eagle’s 2 medium (0.5 ml, fied Eagle’s minimum essential medium (DMEM) supple- without bicarbonate) containing 20 mM HEPES buffer (pH mented with 5% calf serum (Biofluids), 1% Nutridoma-SP 7.4), 0.5 mM MgCI2, 1 mM CaCI2, 0.3% BSA fraction V, (Boehringer Mannheim), 100 units of penicillin per ml, 100 100 nM 12-0-tetradecanoylphorbol13-acetate (TPA), and 1 (IBMX) was added to each pg streptomycin per ml, 1.0 mM CaCl,, and 0.5 mM mM 3-isobutyl-1-methylxanthine 0.5 ml sample to yield 0.5 mM IBMX. Samples and stanMgS04. Ionized calcium in this medium was 0.7 mM. PT-r cells, passaged at high cell densities (greater than dards [rat PTH-(1-34). Bachem, Torrance, CA] were incu1:lO split ratio) and confluent for 2-3 days, develop an ap- bated for 10 minutes at room temperature with ROS 171 pearance of elongated cells. Passages at high dilution (less 2.8 cells treated with 0.1 pM dexamethasone for 6 days and than 1 5 0 split ratio) before confluency show few elon- 50 ng/ml pertussis toxin for 18 h. Reactions were stopped by gated cells even after continuous culture for 7 months. removing the media and adding 0.1 N HCI containing 0.1 Subcloning was carried out by the limiting dilution method mM CaCI,. cAMP was determined as described later. after 15-20 passages at high cell density. Subclones were initially isolated and then cultured under identical conditions as for the parent strain. From 30 subclones, 4 repre- Intracellular CAMP accumulation with secretin sentative clones were selected based upon difference in Confluent cells in 24 well plates were incubated with morphology. The subclones were frozen in liquid nitrogen at different passages and used for experiments between NIH Eagle’s 2 medium (without bicarbonate) containing 20 mM HEPES buffer (pH 7.4), 0.5 mM MgCl,, 1 mM passages 5 and 12 after subcloning. Chromosomal counting was carried out as described CaCI,, 0.3% BSA fraction V, 0.5 mM IBMX, and graded previously.“’ Actively growing cells were treated with col- concentrations of porcine secretin (Peninsula Laboratochicine (10 pg/ml), centrifuged, resuspended in KC1 solu- ries, San Carlos, CA) for 15 minutes. Reactions were stopped by removing the media and adding 0.1 N HCI contion (5.6 glliter), fixed in methanoVacetic acid (3/1, v/v), spread on slides, and stained with Giemsa reagent; 100 taining 0.1 mM CaCI,. cAMP was determined by automated radioimmunoassay after acetylation according to metaphases were counted. the method previously described.I4’
Cell growth and PHJthymidine incorporation Cells were plated at a density of 4 x lo5 cells per 10 cm dish in complete medium and were counted daily after trypsinization using a hemocytometer. The medium was changed every other day. [’Hlthymidine incorporation was assayed as described previously,“’ with minor modification. Cells were seeded at 5000 cells per well in multiwell plate using complete medium. The cells were incubated with the medium containing varying concentrations of calcium for 30 h, 24 h after plating. [‘Hlthymidine (ICN radiochemicals, 70 Ci/mmol) was added at 4 pCi/ml during the final 6 h.
Determination of PTH-like bioactivity Confluent cells in 10 cm plastic tissue culture petri dishes were incubated in 3.0 ml of steady-state medium (1:l Coon’s modified Ham’s F-l2/DMEM, without calf serum or Nutridoma-SP, with 0.5 mM MgCI,) containing 500 kallikrein inhibitor units of aprotinin (Sigma) per ml, 0.3% heat-inactivated bovine serum albumin (BSA) fraction V, calcium at indicated concentrations (0.7 mM ionized calcium in the experiments with secretin), and test
Preparation of poly(A)’ RNA a n d Northern blot analysis Total RNA was isolated from cultured cells using a modification of the guanidinium thiocyanate method with ultracentrifugation through a dense cushion of cesium ~ h l o r i d e . ‘ Poly(A)+ ~) RNA was selected by oligo(dT)-cellulose chromatography. RNAs were quantitated by A,,,,. Poly(A)+ RNA was electrophoresed on 1 To agarose-formaldehyde gel and transferred to a nylon membrane (Hybond N, Amersham, Arlington Heights, IL). Blots were prehybridized, hybridized, and washed as previously des ~ r i b e d . ‘ ~The . ~ ’ RNA probes included a pvuII-sac1 frag. ~ apstI-xbaI ) ment of a human P T H R P cDNA ~ l o n e ‘ ~and genomic fragment corresponding to the coding region of rat PTH.‘sl These fragments were subcloned into pGEM3 2 (Promega Biotech, Madison, WI), and antisense RNAs were synthesized using SP6 RNA polymerase and a-[”’P]To show the intactUTP (410 Ci/mmol, Amer~ham).‘~.’) ness of RNA, the same membrane was reprobed with rat cyclophilin cDNA, ~ l B 1 5 . ‘ ~ ) Results were evaluated statistically by Students’ t-test.
865
PTHRP PRODUCTION IN PARATHYROID CELL LINES
RESULTS
Morphology and chroniosotnal analysis The morphological characteristics of the four PT-r subclones at subconlluence are illustrated in Fig. 1. At lowdensity passage, subclones 1 and 2 contained only polygonal cells. The cells of subclone 3 showed a mixed phenotype of polygonal and elongated cells. Subclone 4, in contrast, comprised only spindle-shaped cells. Subclone 4 did not reach true confluence. Instead, the cells tended to detach before reaching high density and appeared to degenerate when left in this state (detached cells were not viable as assessed by exclusion of trypan blue). Regardless of the degree of confluency, these morphological properties remained constant with low cell density passages for at least 6 months. After 3 months of continuous high cell density passages (about 25 passages at 1:5 split ratio), however, the morphology of subclones 1 and 2, which were originally polygonal, gradually became elongated and resembled subclone 3 at the periphery of' islets of polygonal cells. Subclone 3 was nearly completely elongated after 3 months of high-density passages. In low-density cultures the proportion between polygonal and elongated cells did not change after 6 months in continuous culture. Subclone 4 remained stable in morphology with either high- or low-density passage. Neither 1,25-dihydroxyvitarnin D , (I ,25-(OH),D,, lo-' MI, retinoic acid (0.33 FM), or continuous culture in 3
mM calcium medium, nor a combination of these conditions, affected the morphology of subclone 4 during 2 months of observation. Subclones I , 2, 3, and 4 were 93, 90, 92, and 95% diploid, respectively, with a modal distribution of 42 chromosomes.
Growch characterislita In complete medium, the elongated subclone 4 showed the most rapid growth, with a doubling time of 12 h (Fig. 2). Doubling times for subclones 1, 2, and 3 were 20, 17, and 21 h, respectively, during the exponential phase of growth. The saturation densities for subclones 1, 2, 3, and 4 were (2.65 f 0.05) x lo', (2.63 f 0.24) x lo7,(2.14 f 0.34) x lo', and (2.42 f 0.09) x 10' per 10 cm dish, respectively (mean SD of five determinations). Calcium (ionized) at 0.1-2.1 mM caused a dose-dependent decrease in ['Hfthymidine incorporation in subclones I , 2, and 3. Thymidine incorporation was suppressed significantly (p < 0.05) at 2.1 mM calcium. In subclone 4, there was no significant inhibition of thymidine incorporation (Table 1). The growth rate and growth suppression by calcium of elongated subclones 1 and 2 after 5 months of high-density continuous culture were similar to those of subclone 4 (data not shown).
*
FIG. 1. Phase-contrast view of four subclones taken from PT-r cells. magnification x 100.
SAKAGUCHI ET AL.
866
Hormone release and cyclic A M P accumulation
Release of PTH-like substance was suppressed in sub.. clones 1, 2, and 3 by increasing concentrations of calcium with half-maximal effects a t 1.2, 1.7, and 1.7 mM, respectively. In subclone 4, however, hormone release was not suppressed even with high calcium concentrations (Fig. 5 ) . Subclones 1 and 2 lost responsiveness to secretin or calcium as they assumed the elongated shape after 5 months of high-density continuous culture. Basal secretion was 41.2 f 4.2 and 43.1 f 6.0 pgEq of rat PTH-(1-34) per lo' cells per 12 h from subclones I and 2, respectively. cAMP production in response to 1 pM secretin was 1.33 + 0.05 and 1.38 f 0.07 compared to basal values, 1.24 i 0.09 and 1.25 f 0.08. Hormone secretion in response to 0.5 pM secretin was 32.8 + 2.15 and 33.4 f 1.6 pgEq of rat PTH(1-34) per lo7 cells per 6 h compared to basal values, 31.3 + 0.9 and 34.6 + 0.8. Hormone secretion at 0.2 and 2.1 mM calcium was 35.6 + 2.7 and 48.4 + 4.7 pgEq per lo' cells per 12 h in subclone 1 and 31.2 + 3.6 and 46.3 f 8.6 in subclone 2. Each value is mean + SEM of at least three determinations.
Release of bioactive PTH-like substance at 0.7 mM ionized calcium in subclones I , 2, 3, and 4 was 144.7 + 12.1, 110.0 f 12.9, 68.0 f 5.6, and 39.9 f 2.4 pgEq of rat PTH-(1-34) per lo7 cells per 12 h, respectively (mean f SEM of six measurements, Fig. 3). Secretin caused a dose-dependent increase in intracellular cAMP accumulation with subclones 1, 2, and 3 ( 1 4 5 , 16.9-, and 3.2-fold increase, respectively, from the basal value), but no effect was detected in subclone 4 during 15 minute incubations (Fig. 4). Secretin (0.5 pM) induced a 41, 58, and 50% increase in the release of PTH-like activity during 6 h incubations from subclones 1 , 2, and 3, respectively; n o response was observed in subclone 4.
-
I0 e 0 I
x
lol
I
,
PT-r 4
I I
'
Y
I
Expression of mRNA f o r PTHRP and PTH
$
H PT-r 2
Northern analysis of poly(A)' RNA showed transcripts for PTHRP in PT-r cells and subclones 1, 2, 3, and 4 (Fig. 6). A renal carcinoma line used as a positive control showed multiple transcripts, including 1.5, 2.1, and 3.2 kb species.''' This single transcript corresponds to that of H500 rat Leydig tumor, a rat model of humoral hypercalcemia of malignancy."' This is probably due to the greater complexity of the human PTHRP gene compared to its rat counterpart, which is capable of transcribing only a single transcript. PT-r subclones revealed the same transcript with decreasing intensities from clones 1 to 4, which is compatible with the bioassay data (Fig. 3). They showed cyclophilin transcripts with similar intensities, dernonstrating that RNAs were intact. Neither the parental line, PT-r, nor the four subclones showed the transcript for PTH (data not shown).
I
I
,
[ I
PT-r 1
I :
PT-r 3
w m
3
z I v)
1-
-!
J
W
u
0
1
2
3
TIME
4 - DAYS
FIG. 2. Growth curves. Cells were plated at a density of 4 x los cells per 10 cm dish in complete medium. Medium
DISCUSSION
was changed every other day after plating. Cell number was counted daily in triplicate. The coefficient of variation of replicate plates was < 10%.
Previously we showed that the parental PT-r cell line releases a bioactive PTH-like molecule and that calcium con-
TABLEI. REGULATION OF THYMIDINE INCORPORATION ~
[Ca"] (mM)
0. I
100.0 f 16.5
81.6
PT-r 2
100.0 100.0 100.0
PT-r 4
BY
~
CALCIUM ~
IN
PT-r SUBCLONESa
~
f
6.0
70.2
f
f
13.5
85.1
f
12.6
ND
f
8.3
83.4
f
6.0
15.5 f
f
9.2
105.8
f
12.6
103.3
f
~
1.7
68.6
f
3.0b
75.0 f 1.2 8.6 19.0
~~
2. I
1.4
0.7
0.4
PT-r 1 PT-r 3
~
ND
88.8 f 15.5
65.6 68.2
f
72.6 87.9
f
+ f
6.0b 1.4b 8.4b 9.1
aEach point represents the mean f SD from three experiments of triplicate measurements. Means of the values at 0.1 mM calcium were taken as 100. ND, not determined. bp < 0.05 in comparison with the data at 0.1 mM calcium.
867
PTHRP PRODUCTION IN PARATHYROID CELL LINES
trols growth and release of the bioactive peptide."' The release of peptide was detected biologically as well as by radioimmunoassay. Further studies"' indicate that the peptide formed and released under calcium control is PTHRP rather than PTH itself. The antiserum that we used in the earlier study"' was unusual in that it had been developed 0 against a human parathyroid extract but was nevertheless sensitive under the conditions used to PTHRP, as we reported previously."' Moreover, the biologically active peptide released into the medium under the control of secretin T as well as calcium, responses that parallel the physiology of Lu the parathyroid gland, make the PT-r cell a useful model for studying parathyroid cell control mechanism. In the present study we evaluated characteristics of subclones of a clonal rat parathyroid cell line and analyzed the Lu possibility that differences in cell morphology, cell growth, U and secretion may be due to subpopulational diversity within the parental cell line. Morphologic and growth properties differed among the selected subclones from the first screening of the cells and remained stable for at least 6 months at low-density passages. After 3 months in culture 2 3 4 using high-density passages, the functional, morphologic, and growth characteristics of two subclones 1 and 2 PT-r SUBCLONES changed; low-density passage did not affect these characFIG. 3. Basal secretion of PTH-like bioactivity at 0.7 teristics. Once the morphology changed, it was not reversimM Ca2+.Confluent cells were incubated with the medium ble with retinoic acid, a differentiation factor for certain described in Materials and Methods for 12 h. Each point cells,'1° l o 1,25-(OH),D,, a differentiation factor for leurepresents mean SEM from six independent experiments kemic cells'lz and a suppressor of PTH secretion"4 1 5 ' of triplicate measurements. and mRNA expression,['61 or calcium, which suppresses parathyroid hormone secretion and cell growth. "' Functional differences accompanied the morphological changes. Four subclones studied in detail here retain phenotypes similar to those of the parental PT-r clone with respect to hormone production, even though there are quantitative differences among the subclones. Diversity among subclonal populations was clear; the fast-growing subclone (subclone 4) released less basal PTh-like bioactivity, and the slower growing subclones expressed higher basal release. Moreover, subclone 4 secretes PTH-like bioactivity independently of calcium or secretin, and in 2 15 these cells there was no effect of secretin on intracellular 5 cAMP accumulation. Subclone 3, with a mixed morphol5j 10 ogy, polygonal and elongated, also showed reduced basal 3 production and reduced sensitivity to calcium and secretin i 5 0 regulation. In this subclone polygonal cells tend to assume i c an elongated phenotype as the cells reach a certain degree z o 0 - 9 - B - 7 - 6 of confluency. Subclone 3 may represent a transitional [SECRETIN], log M 1 2 3 4 state of differentiation between those of the subclones 1 or PT-r SUBCLONES 2 and 4. Indeed, even originally polygonal subclones (subclones 1 and 2) became mixed and eventually elongated FIG. 4. Effects of secretin on the intracellular accumula- after many passages at high cellular density. The charactertion of cAMP and on release of PTH-like bioactivity. istics of these subclones upon conversion to elongated Confluent cells were incubated with serial dilutions of se- shape were similar to that of subclone 4. These phenotypic cretin for IS minutes to test intracellular cAMP or with 0.5 changes d o not reflect chromosomal instability in that the pM secretin for 6 h to test hormone release. Results for SEM of quad- karyotype was similar among the four subclones studied cAMP and bioactivity represent mean ruplicate and triplicate samples, respectively. Release of here. Heterogeneity within clonal populations is comhormone bioactivity was significantly stimulated by secretin in PT-r subclones 1,2, and 3 (p < 0.01. The bioactivity mon.('8-z01The mechanisms of controlling phenotypic diwas expressed as pg equivalent of rat PTH-(1-34) per lo7 versity in the PT-r subclones are not known. The differcells. ences observed may reflect gene mutation, oncogene acti-
160t
a
*
('
*
868
SAKAGUCHI ET AL.
I? 1 200 b
0
5 150 Q I
w
cn w a 100 1
w
U
50 0 2 U
1 2 [CALCIUM] - mM
0
0
I
FIG. 5. Effect of calcium on release of bioactive PTHlike material. Confluent cells were incubated in medium described in Materials and Methods for 12 h at varying ionized calcium concentrations. Each point represents SEM of three determinations. mean
*
-
Subclones
2 r
2
k
i
2
3
4
3.2 2.1 -
PTHRP
1.5 -
00.m.
Cyclophilin
vation, growth factor production, or other mechanisms. It is well-known that basic fibroblast growth factor (FGF) causes changes in cell morphology to spindle shape and other phenotypes to apparently transformed ones.‘”’ PT-r cells synthesize acid FGF and bear receptors for this factor (K. Sakaguchi et al., unpublished data). Since basic FGF and acidic FGF are similar in function,[21’it is possible that morphologic and functional properties are affected through autocrine regulation by acidic FGF. The differences between highdensity and low-density passage may also be explained by differences in the concentration of this factor or its receptor. In the parental clone, we detected the expression of mRNA for PTHRP, but no PTH mRNA was detected even with a sensitive method.‘>’Originally P T H R P was regarded as a humoral hypercalcemic factor of malignancy, but it is now known that this peptide exists in a variety of tissues and, interestingly, in normal, hyperplastic, or adenomatous parathyroid glands as well.’’ ’’ ’’) In parathyroid adenomas, the PTHRP mRNA is frequently overexpressed.”’ Moreover, PTHRP seems to be the main or only hormone produced by fetal lamb parathyroid glands and stimulates the placental calcium transport to maintain higher ionized calcium in the plasma of the fetus than in that of the PT-r subclones also expressed mRNA for PTHRP, which paralleled in amount of bioactivity found in the conditioned medium from each clone. They did not show the transcript for PTH. PTH and PTHRP act with similar affinity on the same receptor and show similar effects on ROS 1712.8 cells, a rat osteosarcoma cell line.‘”1 Hence our bioassay data presented using ROS 17/23 cells indicate PTHRP activity. Our rat parathyroid cell line appears not to have the capacity to produce PTH, implying this line may be a fetal version of parathyroid cells. It is of particular interest that the PT-r cell nevertheless shows the negative secretory responsivity to calcium that is characteristic of differentiated P T H secreting cells. It will be important to determine whether the PTHRP response to calcium is mediated through a mechanism similar to that normally controlling secretion in the adult parathyroid gland. As such, the PT-r cell and subclones could be useful in studies of phenotypic expression, calcium regulation, and gene expression and possible interrelationships among the three.
ACKNOWLEDGMENTS
We thank Dr. A.E. Broadus for critical reading of this FIG. 6. Northern blot analysis for transcripts of P T H R P in subclones. Poly(A)+RNA extracted from PT-r cells and manuscript, subclones 1 , 2, 3, and 4 (4 pg each), as well as from HHMassociated renal carcinoma line (2 pg), was analyzed with probes for PTHRP and cyclophilin as described in MateriREFERENCES als and Methods. Autoradiography exposure time was 15 and 6 h for PTHRP and cyclophilin, respectively, and the 1. Sakaguchi K, Santora A, Zimering M, Curcio F, Aurbach sizes of the transcripts in renal carcinoma are shown on the GD, Brandi ML 1987 Functional epithelial cell line cloned left in kb. from rat parathyroid glands. Proc Natl Acad Sci USA 84: 3269. 2. Ikeda K, Weir EC, Sakaguchi K, Burtis W J , Zimering M, Mangin M , Dreyer BE, Brandi ML, Aurbach GD, Broadus
PTHRP PRODUCTION IN PARATHYROID CELL LINES
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
869
17. Raisz LG 1963 Regulation by calcium of parathyroid growth AE 1989 A clonal rat parathyroid cell line expresses a paraand secretion in vitro. Nature (Lond) 197:1115. thyroid-related peptide but not parathyroid hormone itself. 18. Smith JR, Hayflick L 1974 Variation in the life-span of Biochem Biophys Res Commun 162:108. clones derived from human diploid cell strains. J Cell Biol Pines M, Santora A, Gierschik P , Menczel J , Spiegel A 1986 62:48. The inhibitory guanine nucleotide regulatory protein modu19. Korn J H , Torres D, Downie E 1984 Clonal heterogeneity in lates agonist-stimulated CAMP production in rat osteosarthe fibroblast response to mononuclear cell derived mediacoma cells. Bone Min 1:15. tors. Arthritis Rheum 27:174. Eastman S, Aurbach G D 1982 Determination of plasma 20. Longenecker J P , Kilty LA, Ridge JA, Miller DC, Johnson cyclic AMP with automated radioimmunoassay system LK 1983 Proliferative variability of endothelial clones de(Gamma-Flo). J Cyclic Nucleotide Res 8297. rived from adult bovine aorta: Influence of fibroblast growth Chirgwin JM, Przybyla AE, Raymond J, MacDonald RJ, factor and smooth muscle cell extracellular matrix. J Cell Rutter WJ 1979 Isolation of biologically active ribonucleic Physiol 114:7. acid from sources enriched in ribonuclease. Biochemistry 18: 21. Gospodarowicz D, Ferrara N, Schweigerer L, Neufeld G 5294. 1987 Structural characterization and biological functions of lkeda K, Weir EC, Mangin M, Dannies PS, Kinder B, Deftos fibroblast growth factor. Endocr Rev 8:95. LJ, Brown EM, Broadus AE 1988 Expression of messenger ribonucleic acids encoding a parathyroid hormone-like pep- 22. Mangin M, Webb AC, Dreyer BE, Posillico JT, lkeda K, Weir EC, STewart AF, Bander NH, Milstone L, Barton DE, tide in normal human and animal tissues with abnormal exFrancke U, Broadus AE 1988 Identification of a cDNA enpression in human parathyroid adenomas. Mol Endocrinol2: coding a parathyroid hormone-like peptide from a human tu1230. mor associated with humoral hypercalcemia of malignancy. Ikeda K, Mangin M, Dreyer BE, Webb AC, Posillico JT, Proc Natl ACad Sci USA 85597. Stewart AF, Bander NH, Weir EC, Insogna KL, Broadus AE 1988 Identification of transcripts encoding a parathyroid 23. Thiede MA, Rodan G A 1988 Expression of a calcium-mobilizing parathyroid hormone-like peptide in lactating mamhormone-like peptide in messenger RNAs from a variety of mary tissue. Science 242:278. human and animal tumors associated with humoral hypercal24. Rodda CP, Kubota M, Heath JA, Ebeling PR, Moseley J M , cemia of malignancy. J Clin Invest 81:2010. Care AD, Caple IW, Martin TJ 1988 Evidence for a novel Heinrich G, Kronenberg HM, Potts JR. JT, Habener J F parathyroid hormone-related protein in fetal lamb parathy1984 Gene encoding parathyroid hormone- nucleotide seroid glands and sheep placenta: Comparisons with a similar quence of the rat gene and deduced amino acid sequence of protein implicated in humoral hypercalcaemia of maligrat preproparathyroid hormone. J Biol Chem 259:3320. nancy. J Endocrinol 117:261. Danielson PE, Forss-Petter S, Brow MA, Calavetta L, Douglass J , Milner RJ, Sutcliffe JG 1988 plB15: A cDNA 25. Care AD, Abbas SK, Caple IW, Heath JA, Loveridge N, Pickard DW, Rodda C , Martin TJ 1988 Evidence for a novel clone of the rat mRNA encoding cyclophilin. DNA 7:261. hormone in the parathyroid glands of fetal sheep. In: Jones Strickland S, Mahdavi V 1978 The induction of differentiaC T (ed) Fetal and Neonatal Development. Perinatology tion in teratocarcinoma stem cells by retinoic acid. Cell 15: Press, Ithaca, NY, p. 103. 393. 26. Abbas SK, Pickard DW, Rodda C P , Health JA, Hammonds Siddel N, Altnian A, Haussler MR, Seeger RC 1983 Effects RG, Wood WI, Cale l W , Martin T J , Care AD 1989 Stimulaof retinoic acid (RA) on the growth and phenotypic exprestion of ovine placental calcium transport by purified natural sion of several human neuroblastoma cell lines. Exp Cell Res and recombinant parathyroid hormone-related protein 148:21. (PTHrP) preparations. Q J Exp Physiol 74549. Abe E, Miyaura C , Sakagami H , Takada M, Konno K, 27. Shigeno C, Yamamoto I , Kitamura N, Noda T , Lee K, Sone Yamazaki T , Yoshiki S, Suda T 1981 Differentiation of T, Shiomi K, Ohtaka A, Fujii N, Yajima H , Konishi J 1988 mouse myeloid leukemia cells induced by lu,25-dihydroxyInteraction of human parathyroid hormone-related peptide vitamin D,. Proc Natl Acad Sci USA 78:4990. with parathyroid hormone receptors in clonal rat osteosarMiyaura C, Abe E, Kuribayashi T, Tanaka H, Konno K, coma cells. J Biol Chem 263:18369. Nishii Y, Suda T 1981 lu,25-Dihydroxyvitamin D, induces differentiation of human myeloid leukemia cells. Biochem Biophys Res Commun 102:937. Au WYW 1984 Inhibition by 1,25-dihydroxycholecalciferol Address reprint requests to: of hormonal secretion of rat parathyroid gland in organ culKazushige Sakaguchi ture. Calcif Tissue Int 36:384. National Institute of Diabetes and Cantley LK, Russell J, Lettieri D, Sherwood LM 1985 1,25Digestive and Kidney Diseases Dihydroxyvitamin D, suppresses parathyroid hormone secreNational Institutes of Health tion from bovine parathyroid cells in tissue culture. EndocriBethesda, M D 20892 nology 117:2114. Silver J , Russell J, Sherwood LM 1985 Regulation by vitamin D metabolites of messenger ribonucleic acid for preproparathyroid hormone in isolated bovine parathyroid cells. Received for publication October 24, 1989; in revised form JanuProc Natl Acad Sci USA 82:4270. ary 2, 1990; accepted January 19, 1990.