0013.7227/92/1316-2636$03.00/O Endocrinology Copyright 0 1992 by The Endocrine
Vol. 131, No. 6 Printed in U.S.A.
Society
Individual Parathyroid Cells Exhibit Cyclic Secretion Parathyroid Hormone and Chromogranin-A (as Measured by a Novel Sequential Hemolytic Plaque Assay)* CANDACE LORRAINE
K. RITCHIE, DAVID A. FITZPATRICK
V. COHN,
PETER
B. MAERCKLEIN,
of
AND
Endocrine Research Unit, Mayo Clinic and Mayo Foundation (C.K.R., P.B.M., L.A.F.), Rochester, Minnesota 55905; and the Department of Biological and Biophysical Sciences, University Louisville School of Dentistry, Health Science Center (D. V.C.), Louisville, Kentucky 40292
of
ABSTRACT PTH and chromogranin-A (CgA) are the two major proteins secreted from the parathyroid gland. We investigated the secretory patterns of CgA and PTH using a sequential reverse hemolytic plaque assay (RHPA). The RHPA allows detection of hormone secretion from individual cells after a secretory stimulus. For the sequential RHPA, bovine parathyroid cells were mixed with protein-A-conjugated ovine erythrocytes (oRBC). Antiserum, either anti-PTH or anti-CgA, was added under optimal secretory conditions. The addition of complement caused lysis of oRBC surrounding hormone-secreting cells. In this stage (stage l), individual cells were identified and indexed as secreting or nonsecreting cells. For stage 2, a new lawn of oRBC was established,
and a second RHPA was performed on the same population of cells, allowing for the detection of secretory patterns. In the single stage RHPA, about three fourths of the cells formed CgA plaques compared to only about half that formed PTH plaques, suggesting that CgA and PTH are not always cosecreted. In the sequential RHPA, of the cells that did not secrete CgA or PTH in stage 1, up to half secreted in stage 2. Of those cells that secreted in stage 1, up to one fourth did not secrete in stage 2. These results indicate that the parathyroid cells “cycled” between secretory and nonsecretory phases. Our experimental design precluded our obtaining unequivocal data on whether CgA is an autocrine/paracrine regulator of PTH secretion. (Endocrinology 131: 2638-2642, 1992)
P
In as much as secretory cells that contain CgA have their secretion blocked by CgA-derived peptides (11, 14), it is possible that these derived peptides are autocrine/paracrine modulators of cellular secretion (11, 12). We recently developed and validated two reverse hemolytic plaque assays (RHPA) in order to better detect and quantify PTH and CgA secretion from individual parathyroid cells (15, 16). We noted that under the conditions of the RHPA, a significant fraction of any cell population (from 2045%) did not appear to secrete either PTH or CgA. We wondered whether the nonsecreting cells might be repressed by CgA peptides that these samecells had earlier released. We attempted to test this hypothesis in the present study by means of a two-stage RHPA. In stage 1, we indexed the secretory status of individual cells in terms of whether they were secreting either PTH or CgA. In stage 2, conducted 24 h later, we determined the secretory activity of those same cells that were indexed in stage 1. Our data indicate that parathyroid cells maintained in short term culture go through secretory and nonsecretory phases. We also found that the secretion of CgA and that of PTH are not tightly coordinated. These data, however, did not indicate whether nonsecreting cells are repressedby CgA releasedat an earlier time.
TH AND chromogranin-A (CgA) are the two major proteins secreted by the parathyroid gland in response to hypocalcemic stimulation (1). PTH, a peptide of 84 amino acids, causesrestoration of serum calcium through actions principally on bone and kidney and, hence, plays a key role in calcium homeostasis(2). CgA contains about 450 amino acids and is costored in secretory granules with PTH (3, 4). Whereas PTH is normally confined to the parathyroid cell, CgA is found throughout the endocrine/neuroendocrine system in secretory granules containing the resident hormones (5). Examples include the pituitary, adrenal, and pancreas (6, 7). Although the specific physiological function(s) of CgA remains to be elucidated, there is increasing evidence that it is a precursor of several biologically active peptides that either inhibit or activate exocytosis (8). Among these derived peptides are pancreastatin, which inhibits the secretion of insulin by the pancreas (9, 10) and PTH and CgA by the parathyroid (11, 12); chromostatin, which blocks stimulated secretion of catecholamines by chromaffin cells (13); and parastatin, which depressesPTH/CgA secretion (14). Received July 8, 1992. Address all correspondence and requests for reprints to: Dr. L. A. Fitzpatrick, Endocrine Research Unit, 5-164 West Joseph Building, Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905. . *Presented in uart at the 73rd Annual Meeting of The Endocrine Society, Washington, D.C., 1991. This work was sipported in part by USPHS Grants DK-42802 and DK-42572.
Materials
and Methods
Cell dispersion Bovine parathyroid Meats, Winona, MN)
glands obtained from an abattoir (Ledebuhr were minced in l-mm’ pieces in 5 ml medium
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
SECRETION
[Dulbecco’s Modified Essential Medium (Sigma, St. Louis, MO) containing 20 rnM HEPES, 10 &liter streptomycin, and 100 IU/ml penicillin (DMEM)]. DMEM was added to adjust the volume to 30 ml, and the suspension was centrifuged at 750 x g for 10 min. The pellet was resuspended in 25 ml trypsin (1 mg/ml; wt/vol) in DMEM and digested for 2 h in a siliconized Spinner flask with constant agitation. The tissue fragments were vigorously pipetted with a siliconized pipet at IO-min intervals to aid in cell dispersion. After 2 h, the cell suspension was filtered through 210~pm Teflon mesh (Spectrum, Los Angeles, CA) and centrifuged at 750 x 8 for 10 min. The resulting cell pellet was washed with 40 ml DMEM and centrifuged at 750 X g for 10 min to remove residual trypsin. The total number of cells were determined by counting with a hemacytometer. The cells were resuspended at 2-3 X lo4 cells/ml in DMEM.
Conjugation
of
ovine erythrocytes
(ORBC)
Five hundred microliters (wet weight) oRBC (Colorado Serum Co., Denver, CO) were washed with normal saline and centrifuged at 500 X g for 10 min. This wash/centrifugation was repeated four times. Next, 5 ml 750 pt.4 CrC13. 6HZ0 in saline were mixed with protein-A (5 mg/ml; wt/vol), and the solution was added to the oRBC with gentle mixing. After a l-h incubation at 30 C, the protein-A-conjugated oRBC suspension was washed twice with normal saline, three times with DMEM containing 0.1% BSA, and stored as a ‘2% solution at 4 C.
Sequential
RHPA
This procedure consisted of two standard RHPA analyses performed sequentially on the same parathyroid cells (see flow diagram, Fig. 1). PTH antibody was provided by Dr. Claude Arnaud and has been described previously (15). Antibovine CgA was prepared as described previously (4, 16). Equal volumes of protein-A-conjugated oRBC and dispersed bovine parathyroid cells were mixed (final parathyroid cell concentration, l-l.5 X 104), infused into a Cunningham chamber (17), and incubated at 37 C under 5% CO1 for 45 min to allow cell attachment. In stage 1, antiserum (either anti-PTH at 1:60 or anti-CgA at 1:30) in 0.2 rnM calcium containing DMEM was added to initiate the reaction. As hormone was released from the cell, it bound to the antibody, and the resultant complex then bound to the protein-A-conjugated oRBC.
SEQUENTIAL
PLAQUE
ASSAY
STAGE
P/ PLAQUE (SECRETORY
1
PLAQUE (-) (NONSECRETORY
CELLS)
c STAGE / r’ PLAQUE
Analysis
STAGE #,~,’
PLAQUE
(-)
PLAQUE
PTH and CgA secretion by RHPA
of
PTH secretion
1. Secretion
of PTH
2 \ % of plaques Plaque area (Km* X 103)
‘\ (+)
of
In the standard RHPA, 45-50% of the cellsformed plaques when tested for PTH secretion. In comparison, 70-75% of the cellsreleasedCgA. Plaque area, a function of the amount of hormone released per cell, correlated with the number of plaques formed with either PTH or CgA antisera (Table 1).
TABLE
2 \ ‘\
(+)
Results
PLAQUE
by sequential
RHPA
In stages 1 and 2, overall 62% and 63%, respectively, of the cells formed plaques that were indicative of PTH secretion. This proportion of secreting cells is similar to that reported earlier (19). When the individual cells were examined for plaque formation in stages 1 and 2, a somewhat different pattern became apparent. In stage 1, of the 62% of the cellsthat secretedPTH, only about three quarters secreted PTH in stage 2. Of the cells that did not secrete PTH in stage
‘\
(+) CELLS)
CgA
After 4 h of incubation, the reaction was terminated by the addition of complement (1:15 to 1:30 for 12 min) that itself bound to the antigenantibody complex and lysed the oRBC. This produced an empty area or plaque completely encircling a hormone-secreting cell (17,lS). The slides were flushed with DMEM, and the cells were indexed as plaque positive, i.e. cells secreting hormone, or plaque negative, i.e. cells not secreting hormone. In stage 2, a fresh lawn of oRBC was added to the slide and allowed to attach. Antiserum (anti-PTH at 1:60 or anti-CgA at 1:30) was added in DMEM containing 0.2 rnM calcium, and the slides were incubated for 4 h before the addition of complement (1:15 to 1:30 for 12 min). The slides were fixed with 2% glutaraldehyde, and the coverslips were removed and stained with 0.05% (wt/vol) azure blue and 0.05% (wt/ vol) methylene blue in sodium borate. The doses of PTH or CgA antisera used herein yielded maximal plaque size and maximal plaque area, as determined previously (16, 17). The number of secretory cells (plaque number) was determined on a Nikon Microphot FXA (Nikon Corp., Tokyo, Japan) at ~625 magnification. One hundred cells per slide were counted. For determination of the amount of hormone secreted, the plaque areas were measured with the Bioquant 2 Imaging System (Micro Development, Prior Lake, MN). A minimum of 50 plaques were measured per experiment. The results represent triplicate experiments consisting of at least 2 slides per group/experiment. For the sequential RHPA, experiments were performed 5-7 times. Each experiment consisted of 2-4 slides/treatment group, and a minimum of 20 cells were indexed per slide. The viability of parathyroid cells was monitored throughout the assay with trypan blue exclusion. Statistical analyses were performed by Student’s f test.
Analysis
v /
OF PTH AND
(-)
FIG. 1. Flow diagram for sequential RHPA. In stage 1, dispersed bovine parathyroid cells were incubated with antibody (1:30 for anti-CgA; 1:60 for anti-PTH) in 0.2 mM calcium. After 4 h, complement was added to initiate cell lysis. Cells secreting hormone were indexed as PLAQUE (+), and those not secreting hormone as PLAQUE (-). In stage 2, a fresh lawn of oRBC was established, antibody was added (1:30 for antiCgA; 1:60 for anti-PTH), and slides were incubated in 0.2 mM calcium containing DMEM for 4 h before complement addition. The cells indexed in stage 1 were reindexed as PLAQUE (+) or PLAQUE (-).
and CgA from
parathyroid
cells
PTH
CtzA
47.7 + 2.8 12.5 + 0.6
68.7 + 4.1” 19.6 + 0.8
Dispersed parathyroid cells were plated for RHPA and incubated with antibody (1:60 for anti-PTH; 1:30 for anti-CgA). After 4 h, complement was added to initiate cell lysis. The number of cells forming plaques (percentage of plaques) and the size of plaques (plaque area) were analyzed. For determination of the percentage of plaques formed, 100 cells/slide were analyzed, and for plaque area determination, 50 plaques/slide were measured. The results represent 2 experiments, consisting of quadruplicate slides in each experiment. “P < 0.001 compared to PTH.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
2640
SECRETION
1, about one third secreted PTH in stage 2 (Fig. 2), resulting in a total of 63% cells secreting in stage 2. We include these data for comparison to Figs. 3, 4, and 5. Analysis
of
CgA secretion
by sequential
OF PTH
AND
Endo. Voll31.
CgA
1992 No 6
STAGE 1 ANTIBODY: CgA STAGE 2 ANTIBODY: PTH
RHPA
Overall, 90% and 83% of the cells secreted CgA in stages 1 and 2, respectively. In stage 1, 90% of the cells indexed secretedCgA, as determined by plaque formation. Of these, STAGE 1 ANTIBODY: PTH STAGE 2 ANTIBODY: PTH
0
Secretory cells Nonsecretory cells Stage 2
FIG. 4. Ability
of cells plaqued with anti-PTH in stage 1 to respond to antiCpA in stage 2. Anti-PTH was used as the stage 1 antibody, and anti-CgA as the stage 2 antibody. In stage 1, 68% of the cells indexed secreted PTH, and of these, 92% (63 cells) were capable of secreting CgA in stage 2. Of the cells not secreting PTH in stage 1, 62% (20 cells) secreted CgA in stage 2. Results are normalized to 100 cells. The SE did not exceed 5%.
u
STAGE 1 ANTIBODY: PTH STAGE 2 ANTIBODY: CgA
Secretory cells Nonsecretory cells
Stage 2
i
FIG. 2. Ability
of cells plaqued with anti-PTH in stage 1 toiespond to anti-PTH in stage 2. PTH was used as the stage 1 antibody and the stage 2 antibody. In stage 1, 62% of the cells indexed secreted PTH, and of these, 79% (49 cells) secreted again in stage 2. Of the cells unable to secrete PTH in stage 1, 37% (14 cells) secreted PTH in stage 2. Results are normalized to 100 cells. The SE of the assay did not exceed 6% [adapted from Sun et al. (19)]. Data are presented as a comparison to those in Figs. 3,4, and 5.
63
STAGE 1 ANTIBODY: CgA STAGE 2 ANTIBODY: CgA Secretory cells Nonsecretory cells FIG. 5. Ability
of cells plaqued with anti-CgA in stage 1 to respond to anti-PTH in stage 2. Anti-CgA was used as the stage 1 antibody, and anti-PTH as the stage 2 antibody. In stage 1,95% of the cells indexed secreted CgA, and of these, 77% (73 cells) were capable of secreting PTH in stage 2. Of the cells not secreting CgA in stage 1,40% (2 cells) secreted PTH in stage 2. Results are normalized to 100 cells. The SE did not exceed 6%.
84% secreted CgA in stage 2. Of those cells that did not secrete CgA in stage 1, about half secreted CgA in stage 2.
Stage 1 Secretory cells Nonsecretory cells FIG. 3. Ability
of cells plaqued antiCpA in stage 2. Anti-CgA the stage 2 antibody. In stage and of these, 84% (78 cells) unable to secrete CgA in stage Results are normalized to 100
Stage 2 with anti-CgA in stage 1 to respond to was used as the stage 1 antibody and 1,90% of the cells indexed secreted CgA, secreted again in stage 2. Of the cells 1, 64% (5 cells) secreted CgA in stage 2. cells. The SE did not exceed 7%.
Analysis of PTH secretion stage 2 or vice versa
in stage 1 and CgA secretion
in
Since the binding of secreted PTH or CgA by antiserum in the stage 1 RPHA might influence the secretion of the active cells in stage 2, we alternated the antiserum in the two stages (Fig. 4). When the stage 1 antiserum was directed toward PTH, 68% of the cells formed plaques. In stage 2,
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
SECRETION
92% of the cells formed plaques when tested with CgA antiserum. This result was a consequence of more than half of the PTH plaque-negative cells of stage 1 forming plaques in stage 2. When CgA antiserum was used in stage 1,95% of the cells formed plaques. Of these, three quarters formed plaques in stage 2 when PTH antiserum was employed. Of the few cells that did not form plaques in stage 1, about half secreted PTH in stage 2 (Fig. 5). We also looked at the secretory patterns of CgA and PTH in parathyroid cells undergoing a 2-h secretory stimulus rather than the 4-h incubation period routinely employed. The secretory patterns were similar to those reported for the 4-h periods (data not shown). Discussion Under all assay conditions, there was always a population of cells indexed that did not secrete hormone in stage 1, but did release hormone during stage 2. Likewise, a small fraction of cells that secreted either PTH or CgA in stage 1 did not secrete in stage 2, although they were viable. These results demonstrate that parathyroid cells under our experimental conditions exhibit both resting and secretory phases and cycle between these phases. The minimum time that elapsed between the end of stage 1 and the end of stage 2 (i.e. the minimum incubation time of 2 h plus set-up time between the end of stage 1 and the beginning of stage 2) was 3 h, as mentioned above. This, then, represents the maximum time frame within which the cells converted from the secretory to nonsecretory phases or vice versa. Parathyroid cell cycling directly confirms a much earlier interpretation based on anatomical examination of the secretory granule content that mouse and human cells exist in both secretory and inactive stages (20, 21). One goal of this study was to determine whether CgA and/or its derived peptides function physiologically in parathyroid cell secretory cycling. In cultured bovine parathyroid cells, CgA and the CgA-derived peptides pancreastatin [equivalent to bovine CgA-(251-294)] and parastatin (14) [equivalent to bovine CgA-(348-393)] are potent inhibitors of PTH secretion (11, 12). Since CgA itself and, presumably, peptides derived by proteolytic cleavage are released from the cells, Fasciotto and co-workers (12) postulated that these peptides served as autocrine or paracrine agents to repress the secretion of actively secreting cells. Our data are not definitive on this point. To comfortably conclude that released CgA or CgA-derived peptides downregulate secretion, we would have expected to find that the majority of those cells that were active in stage 1 were inactive in stage 2. This was not observed. In fact, the percentage of cells that became nonsecretors in stage 2 only ranged from a low of 7% to a high of 23% (average of 16% in the four experiments listed in Figs. 2-5). It is likely, however, that our experimental conditions precluded the detection of down-regulation of secretion. Fasciotto et al. (12) showed that the action of pancreastatin was fully reversible in 90 min or less in cultured parathyroid cells, The minimum time that elapsed from the end of stage
OF PTH
AND
CgA
2641
1 to the beginning of stage 2 was 180 min, a period sufficiently long that cells inhibited by released CgA or a derived peptide could easily have recovered. Indeed, about half of the nonsecreting cells of stage 1 became secretors in stage 2. This could reflect recovery from prior inhibition of released peptides or a cell cycle totally unrelated to secretory events. Our experiments indicate that a greater number of cells formed CgA-related plaques than PTH-related plaques (85% vs. 63%), suggesting that CgA may be released independently of PTH. This noncoordinated secretion supports previous observations of Bajpai and Hamilton (22) that, in addition to being colocalized in secretory granules (23), CgA may be enriched compared to PTH in some secretory granules (22). One caveat to this explanation is that the sensitivity of the PTH and CgA antibodies might be sufficiently different so that the capability to detect PTH secretion is appreciably lower than that for CgA. In the sequential RHPAs in which the same antibody was used in stages 1 and 2, the total number of cells secreting a given hormone remained the same, although different cells were in the secretory phase. For example, when PTH secretion was assessed, 62% of cells in stage 1 and 63% of cells in stage 2 were actively secreting. In the case of CgA, 90% of the cells formed plaques in stage 1, and 83% in stage 2. This suggests that there may be a mechanism in normal tissue that regulates the number of cells that respond to a secretory stimulus at any particular time. The mechanism of this putative regulation remains to be elucidated. References 1.
Cohn DV, Zangerle R, Fischer-Colbrie R, Chu LLH, Eiting JJ, Hamilton JW, Winkler H 1982 Similaritv of secretorv urotein-l from parathyroid Proc Nat1 Acad
gland to chromogranin-A Sci USA 79:6056-6059
adrenal’medulla.
LA 1991 Regulation of paraRev 12:291-301 3. Ravazzola M, Orci L, Habener JF, Potts Jr JT 1978 Parathyroid secretory protein: immunocytochemical localization within cells that contain parathyroid hormone. Lancet 2:371-372 4. Cohn DV, Morrissey JJ, Shofstall RE, Chu LLH 1982 Cosecretion of secretory protein-l and parathormone by dispersed bovine parathyroid cells, Endocrinology 110:625-630 5. Cohn DV, Elting JJ, Frick M, Elde R 1984 Selective localization of the parathyroid secretory protein-l/adrenal medulla chromogranin A protein family in a wide variety of endocrine cells of the rat. Endocrinology 114:1963-1974 6. Cohn DV, Fasciotto BH, Gorr S-U, Parkins F, Shioi J, Levine MA, Greeley JH 1990 The putative role of secretory protein-I/chromogranin A as a precursor for regulatory hormones. In: Peterlik M, Bronner F (eds) Progressive Clinical and Biological Research. WileyLiss, New York, vol 332:51-66 7. Winkler H, Carmichael SW 1982 The chromaffin granule. In: Poisner AM, Trifaro JM (eds) The Secretory Granule. Elsevier, Amsterdam, pp 3-79 8. Cohn DV, Fasciotto BH, Gorr SU, Darkins F, Shioi H, Levine MA, Greely GH 1990 The putative role of secretory protein I/ chromogranin A as a precursor for regulatory proteins. In: Peterlik M, Bonner F (eds) Molecular and Cellular Regulation of Calcium and Phosphate Metabolism. Liss, New York, pp 51-56 9. Tatemoto K, Efendic S, Mutt V Mall G, Friestner GJ, Barchas JD 1986 Pancreastatin, a novel pancreatic peptide that inhibits insulin secretion. Nature (Lond) 274:476-478 10. Greeley GH, Thompson JC, Ishizuha J, Cooper CW, Levine MA, Gorr S-U, Cohn DV 1989 Inhibition of glucose-stimulated insulin 2.
Pocotte S, Ehrenstein
from
thyroid
hormone
J, Fitzpatrick
secretion.
Endocr
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
11.
12.
13.
14.
15.
16.
SECRETION
release in the perfused rat pancreas by parathyroid secretory proteinI (chromogranin A). Endocrinology 124:1235-1238 Fasciotto BH, Gorr S-U, DeFranco DJ, Levine MA, Cohn DV 1989 Pancreastatin, a presumed product of chromogranin-A (secretory protein-I) processing, inhibits secretion from porcine parathyroid cells in culture. Endocrinology 125:1617-1622 Fasciotto BH, Gorr SU, Bourdeau AM, Cohn DV 1990 Autocrine regulation of parathyroid secretion: Inhibition of secretion of chromogranin-A (secretory protein-I) and potentiation of secretion by chromogranin-A and pancreastatin antibodies. Endocrinology 127:1329-1335 Galindo E, Rill A, Bader M-F, Aunis D 1991 Chromostatin, a 20amino acid peptide derived from chromogranin A, inhibits chromaffin cell secretion. Proc Nat1 Acad Sci USA 88:1426-1430 Fasciotto BH, Nishijima A, Cohn DV 1992 Parastatin (porcine chromogranin A 347-391) inhibits parathyroid cell secretion. J Bone Mineral Res 7:S235 (Abstract) Fitzpatrick LA, Leong DA 1990 Individual parathyroid cells are more sensitive to calcium than a parathyroid cell population. Endocrinology 126:1720-1727 Ritchie CK, Cohn DV, Fitzpatrick LA 1992 Chromogranin-A secretion from individual parathyroid cells: effects of 1,25-
OF PTH
AND
CgA
Endo. Voll31.
1992 No 6
(OHhvitamin Da and calcium. Bone Mineral 18:31-40 17. Smith PF, Luque EH, Neil1 JD 1986 Detection and measurement of secretion from individual neuroendocrine cells using a reverse hemolytic plaque assay. Methods Enzymol 124:443-465 18. Neil1 JD, Smith PF, Luque EH, Munoz De Toro M, Nagy G, Mulcahey JJ 1987 Detection and measurement of hormone secretion from individual pituitary cells. Recent Prog Horm Res 43:175-228 19. Sun F, Ritchie CK, Hassager C, Maercklein P, Fitzpatrick LA 1992 Heterogeneous response to calcium by individual parathyroid cells. J Clin Invest, in press 20. Roth SJ, Raisz LG 1966 The course and reversibility of the calcium effect on the ultrastructure of the rat parathyroid gland on organ culture. Lab Invest 15:1187-1211 21. Shannon WA, Roth SI 1974 An ultrastructural study of acid phosphatase activity in normal, adenomatous and hyperplastic (chief cell type) human parathyroid glands. Am J Path01 77:493-501 22. Bajpai S, Hamilton J 1990 The isolation and partial characterization of bovine parathyroid secretory granules. Bone Mineral 9:9-22 23. Arps A, Dietel M, Lauritizen B, Elting JJ, Nierndorf A, Cohn DV 1987 Co-localization of parathyroid hormone and secretory proteinI in bovine parathyroid glands: a double immunocytochemical study at the electron microscopical level. Bone Mineral 2:175-183
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
Vol. 131, No. 6 Printed in U.S.A.
Society
Individual Parathyroid Cells Exhibit Cyclic Secretion Parathyroid Hormone and Chromogranin-A (as Measured by a Novel Sequential Hemolytic Plaque Assay)* CANDACE LORRAINE
K. RITCHIE, DAVID A. FITZPATRICK
V. COHN,
PETER
B. MAERCKLEIN,
of
AND
Endocrine Research Unit, Mayo Clinic and Mayo Foundation (C.K.R., P.B.M., L.A.F.), Rochester, Minnesota 55905; and the Department of Biological and Biophysical Sciences, University Louisville School of Dentistry, Health Science Center (D. V.C.), Louisville, Kentucky 40292
of
ABSTRACT PTH and chromogranin-A (CgA) are the two major proteins secreted from the parathyroid gland. We investigated the secretory patterns of CgA and PTH using a sequential reverse hemolytic plaque assay (RHPA). The RHPA allows detection of hormone secretion from individual cells after a secretory stimulus. For the sequential RHPA, bovine parathyroid cells were mixed with protein-A-conjugated ovine erythrocytes (oRBC). Antiserum, either anti-PTH or anti-CgA, was added under optimal secretory conditions. The addition of complement caused lysis of oRBC surrounding hormone-secreting cells. In this stage (stage l), individual cells were identified and indexed as secreting or nonsecreting cells. For stage 2, a new lawn of oRBC was established,
and a second RHPA was performed on the same population of cells, allowing for the detection of secretory patterns. In the single stage RHPA, about three fourths of the cells formed CgA plaques compared to only about half that formed PTH plaques, suggesting that CgA and PTH are not always cosecreted. In the sequential RHPA, of the cells that did not secrete CgA or PTH in stage 1, up to half secreted in stage 2. Of those cells that secreted in stage 1, up to one fourth did not secrete in stage 2. These results indicate that the parathyroid cells “cycled” between secretory and nonsecretory phases. Our experimental design precluded our obtaining unequivocal data on whether CgA is an autocrine/paracrine regulator of PTH secretion. (Endocrinology 131: 2638-2642, 1992)
P
In as much as secretory cells that contain CgA have their secretion blocked by CgA-derived peptides (11, 14), it is possible that these derived peptides are autocrine/paracrine modulators of cellular secretion (11, 12). We recently developed and validated two reverse hemolytic plaque assays (RHPA) in order to better detect and quantify PTH and CgA secretion from individual parathyroid cells (15, 16). We noted that under the conditions of the RHPA, a significant fraction of any cell population (from 2045%) did not appear to secrete either PTH or CgA. We wondered whether the nonsecreting cells might be repressed by CgA peptides that these samecells had earlier released. We attempted to test this hypothesis in the present study by means of a two-stage RHPA. In stage 1, we indexed the secretory status of individual cells in terms of whether they were secreting either PTH or CgA. In stage 2, conducted 24 h later, we determined the secretory activity of those same cells that were indexed in stage 1. Our data indicate that parathyroid cells maintained in short term culture go through secretory and nonsecretory phases. We also found that the secretion of CgA and that of PTH are not tightly coordinated. These data, however, did not indicate whether nonsecreting cells are repressedby CgA releasedat an earlier time.
TH AND chromogranin-A (CgA) are the two major proteins secreted by the parathyroid gland in response to hypocalcemic stimulation (1). PTH, a peptide of 84 amino acids, causesrestoration of serum calcium through actions principally on bone and kidney and, hence, plays a key role in calcium homeostasis(2). CgA contains about 450 amino acids and is costored in secretory granules with PTH (3, 4). Whereas PTH is normally confined to the parathyroid cell, CgA is found throughout the endocrine/neuroendocrine system in secretory granules containing the resident hormones (5). Examples include the pituitary, adrenal, and pancreas (6, 7). Although the specific physiological function(s) of CgA remains to be elucidated, there is increasing evidence that it is a precursor of several biologically active peptides that either inhibit or activate exocytosis (8). Among these derived peptides are pancreastatin, which inhibits the secretion of insulin by the pancreas (9, 10) and PTH and CgA by the parathyroid (11, 12); chromostatin, which blocks stimulated secretion of catecholamines by chromaffin cells (13); and parastatin, which depressesPTH/CgA secretion (14). Received July 8, 1992. Address all correspondence and requests for reprints to: Dr. L. A. Fitzpatrick, Endocrine Research Unit, 5-164 West Joseph Building, Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905. . *Presented in uart at the 73rd Annual Meeting of The Endocrine Society, Washington, D.C., 1991. This work was sipported in part by USPHS Grants DK-42802 and DK-42572.
Materials
and Methods
Cell dispersion Bovine parathyroid Meats, Winona, MN)
glands obtained from an abattoir (Ledebuhr were minced in l-mm’ pieces in 5 ml medium
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
SECRETION
[Dulbecco’s Modified Essential Medium (Sigma, St. Louis, MO) containing 20 rnM HEPES, 10 &liter streptomycin, and 100 IU/ml penicillin (DMEM)]. DMEM was added to adjust the volume to 30 ml, and the suspension was centrifuged at 750 x g for 10 min. The pellet was resuspended in 25 ml trypsin (1 mg/ml; wt/vol) in DMEM and digested for 2 h in a siliconized Spinner flask with constant agitation. The tissue fragments were vigorously pipetted with a siliconized pipet at IO-min intervals to aid in cell dispersion. After 2 h, the cell suspension was filtered through 210~pm Teflon mesh (Spectrum, Los Angeles, CA) and centrifuged at 750 x 8 for 10 min. The resulting cell pellet was washed with 40 ml DMEM and centrifuged at 750 X g for 10 min to remove residual trypsin. The total number of cells were determined by counting with a hemacytometer. The cells were resuspended at 2-3 X lo4 cells/ml in DMEM.
Conjugation
of
ovine erythrocytes
(ORBC)
Five hundred microliters (wet weight) oRBC (Colorado Serum Co., Denver, CO) were washed with normal saline and centrifuged at 500 X g for 10 min. This wash/centrifugation was repeated four times. Next, 5 ml 750 pt.4 CrC13. 6HZ0 in saline were mixed with protein-A (5 mg/ml; wt/vol), and the solution was added to the oRBC with gentle mixing. After a l-h incubation at 30 C, the protein-A-conjugated oRBC suspension was washed twice with normal saline, three times with DMEM containing 0.1% BSA, and stored as a ‘2% solution at 4 C.
Sequential
RHPA
This procedure consisted of two standard RHPA analyses performed sequentially on the same parathyroid cells (see flow diagram, Fig. 1). PTH antibody was provided by Dr. Claude Arnaud and has been described previously (15). Antibovine CgA was prepared as described previously (4, 16). Equal volumes of protein-A-conjugated oRBC and dispersed bovine parathyroid cells were mixed (final parathyroid cell concentration, l-l.5 X 104), infused into a Cunningham chamber (17), and incubated at 37 C under 5% CO1 for 45 min to allow cell attachment. In stage 1, antiserum (either anti-PTH at 1:60 or anti-CgA at 1:30) in 0.2 rnM calcium containing DMEM was added to initiate the reaction. As hormone was released from the cell, it bound to the antibody, and the resultant complex then bound to the protein-A-conjugated oRBC.
SEQUENTIAL
PLAQUE
ASSAY
STAGE
P/ PLAQUE (SECRETORY
1
PLAQUE (-) (NONSECRETORY
CELLS)
c STAGE / r’ PLAQUE
Analysis
STAGE #,~,’
PLAQUE
(-)
PLAQUE
PTH and CgA secretion by RHPA
of
PTH secretion
1. Secretion
of PTH
2 \ % of plaques Plaque area (Km* X 103)
‘\ (+)
of
In the standard RHPA, 45-50% of the cellsformed plaques when tested for PTH secretion. In comparison, 70-75% of the cellsreleasedCgA. Plaque area, a function of the amount of hormone released per cell, correlated with the number of plaques formed with either PTH or CgA antisera (Table 1).
TABLE
2 \ ‘\
(+)
Results
PLAQUE
by sequential
RHPA
In stages 1 and 2, overall 62% and 63%, respectively, of the cells formed plaques that were indicative of PTH secretion. This proportion of secreting cells is similar to that reported earlier (19). When the individual cells were examined for plaque formation in stages 1 and 2, a somewhat different pattern became apparent. In stage 1, of the 62% of the cellsthat secretedPTH, only about three quarters secreted PTH in stage 2. Of the cells that did not secrete PTH in stage
‘\
(+) CELLS)
CgA
After 4 h of incubation, the reaction was terminated by the addition of complement (1:15 to 1:30 for 12 min) that itself bound to the antigenantibody complex and lysed the oRBC. This produced an empty area or plaque completely encircling a hormone-secreting cell (17,lS). The slides were flushed with DMEM, and the cells were indexed as plaque positive, i.e. cells secreting hormone, or plaque negative, i.e. cells not secreting hormone. In stage 2, a fresh lawn of oRBC was added to the slide and allowed to attach. Antiserum (anti-PTH at 1:60 or anti-CgA at 1:30) was added in DMEM containing 0.2 rnM calcium, and the slides were incubated for 4 h before the addition of complement (1:15 to 1:30 for 12 min). The slides were fixed with 2% glutaraldehyde, and the coverslips were removed and stained with 0.05% (wt/vol) azure blue and 0.05% (wt/ vol) methylene blue in sodium borate. The doses of PTH or CgA antisera used herein yielded maximal plaque size and maximal plaque area, as determined previously (16, 17). The number of secretory cells (plaque number) was determined on a Nikon Microphot FXA (Nikon Corp., Tokyo, Japan) at ~625 magnification. One hundred cells per slide were counted. For determination of the amount of hormone secreted, the plaque areas were measured with the Bioquant 2 Imaging System (Micro Development, Prior Lake, MN). A minimum of 50 plaques were measured per experiment. The results represent triplicate experiments consisting of at least 2 slides per group/experiment. For the sequential RHPA, experiments were performed 5-7 times. Each experiment consisted of 2-4 slides/treatment group, and a minimum of 20 cells were indexed per slide. The viability of parathyroid cells was monitored throughout the assay with trypan blue exclusion. Statistical analyses were performed by Student’s f test.
Analysis
v /
OF PTH AND
(-)
FIG. 1. Flow diagram for sequential RHPA. In stage 1, dispersed bovine parathyroid cells were incubated with antibody (1:30 for anti-CgA; 1:60 for anti-PTH) in 0.2 mM calcium. After 4 h, complement was added to initiate cell lysis. Cells secreting hormone were indexed as PLAQUE (+), and those not secreting hormone as PLAQUE (-). In stage 2, a fresh lawn of oRBC was established, antibody was added (1:30 for antiCgA; 1:60 for anti-PTH), and slides were incubated in 0.2 mM calcium containing DMEM for 4 h before complement addition. The cells indexed in stage 1 were reindexed as PLAQUE (+) or PLAQUE (-).
and CgA from
parathyroid
cells
PTH
CtzA
47.7 + 2.8 12.5 + 0.6
68.7 + 4.1” 19.6 + 0.8
Dispersed parathyroid cells were plated for RHPA and incubated with antibody (1:60 for anti-PTH; 1:30 for anti-CgA). After 4 h, complement was added to initiate cell lysis. The number of cells forming plaques (percentage of plaques) and the size of plaques (plaque area) were analyzed. For determination of the percentage of plaques formed, 100 cells/slide were analyzed, and for plaque area determination, 50 plaques/slide were measured. The results represent 2 experiments, consisting of quadruplicate slides in each experiment. “P < 0.001 compared to PTH.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
2640
SECRETION
1, about one third secreted PTH in stage 2 (Fig. 2), resulting in a total of 63% cells secreting in stage 2. We include these data for comparison to Figs. 3, 4, and 5. Analysis
of
CgA secretion
by sequential
OF PTH
AND
Endo. Voll31.
CgA
1992 No 6
STAGE 1 ANTIBODY: CgA STAGE 2 ANTIBODY: PTH
RHPA
Overall, 90% and 83% of the cells secreted CgA in stages 1 and 2, respectively. In stage 1, 90% of the cells indexed secretedCgA, as determined by plaque formation. Of these, STAGE 1 ANTIBODY: PTH STAGE 2 ANTIBODY: PTH
0
Secretory cells Nonsecretory cells Stage 2
FIG. 4. Ability
of cells plaqued with anti-PTH in stage 1 to respond to antiCpA in stage 2. Anti-PTH was used as the stage 1 antibody, and anti-CgA as the stage 2 antibody. In stage 1, 68% of the cells indexed secreted PTH, and of these, 92% (63 cells) were capable of secreting CgA in stage 2. Of the cells not secreting PTH in stage 1, 62% (20 cells) secreted CgA in stage 2. Results are normalized to 100 cells. The SE did not exceed 5%.
u
STAGE 1 ANTIBODY: PTH STAGE 2 ANTIBODY: CgA
Secretory cells Nonsecretory cells
Stage 2
i
FIG. 2. Ability
of cells plaqued with anti-PTH in stage 1 toiespond to anti-PTH in stage 2. PTH was used as the stage 1 antibody and the stage 2 antibody. In stage 1, 62% of the cells indexed secreted PTH, and of these, 79% (49 cells) secreted again in stage 2. Of the cells unable to secrete PTH in stage 1, 37% (14 cells) secreted PTH in stage 2. Results are normalized to 100 cells. The SE of the assay did not exceed 6% [adapted from Sun et al. (19)]. Data are presented as a comparison to those in Figs. 3,4, and 5.
63
STAGE 1 ANTIBODY: CgA STAGE 2 ANTIBODY: CgA Secretory cells Nonsecretory cells FIG. 5. Ability
of cells plaqued with anti-CgA in stage 1 to respond to anti-PTH in stage 2. Anti-CgA was used as the stage 1 antibody, and anti-PTH as the stage 2 antibody. In stage 1,95% of the cells indexed secreted CgA, and of these, 77% (73 cells) were capable of secreting PTH in stage 2. Of the cells not secreting CgA in stage 1,40% (2 cells) secreted PTH in stage 2. Results are normalized to 100 cells. The SE did not exceed 6%.
84% secreted CgA in stage 2. Of those cells that did not secrete CgA in stage 1, about half secreted CgA in stage 2.
Stage 1 Secretory cells Nonsecretory cells FIG. 3. Ability
of cells plaqued antiCpA in stage 2. Anti-CgA the stage 2 antibody. In stage and of these, 84% (78 cells) unable to secrete CgA in stage Results are normalized to 100
Stage 2 with anti-CgA in stage 1 to respond to was used as the stage 1 antibody and 1,90% of the cells indexed secreted CgA, secreted again in stage 2. Of the cells 1, 64% (5 cells) secreted CgA in stage 2. cells. The SE did not exceed 7%.
Analysis of PTH secretion stage 2 or vice versa
in stage 1 and CgA secretion
in
Since the binding of secreted PTH or CgA by antiserum in the stage 1 RPHA might influence the secretion of the active cells in stage 2, we alternated the antiserum in the two stages (Fig. 4). When the stage 1 antiserum was directed toward PTH, 68% of the cells formed plaques. In stage 2,
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
SECRETION
92% of the cells formed plaques when tested with CgA antiserum. This result was a consequence of more than half of the PTH plaque-negative cells of stage 1 forming plaques in stage 2. When CgA antiserum was used in stage 1,95% of the cells formed plaques. Of these, three quarters formed plaques in stage 2 when PTH antiserum was employed. Of the few cells that did not form plaques in stage 1, about half secreted PTH in stage 2 (Fig. 5). We also looked at the secretory patterns of CgA and PTH in parathyroid cells undergoing a 2-h secretory stimulus rather than the 4-h incubation period routinely employed. The secretory patterns were similar to those reported for the 4-h periods (data not shown). Discussion Under all assay conditions, there was always a population of cells indexed that did not secrete hormone in stage 1, but did release hormone during stage 2. Likewise, a small fraction of cells that secreted either PTH or CgA in stage 1 did not secrete in stage 2, although they were viable. These results demonstrate that parathyroid cells under our experimental conditions exhibit both resting and secretory phases and cycle between these phases. The minimum time that elapsed between the end of stage 1 and the end of stage 2 (i.e. the minimum incubation time of 2 h plus set-up time between the end of stage 1 and the beginning of stage 2) was 3 h, as mentioned above. This, then, represents the maximum time frame within which the cells converted from the secretory to nonsecretory phases or vice versa. Parathyroid cell cycling directly confirms a much earlier interpretation based on anatomical examination of the secretory granule content that mouse and human cells exist in both secretory and inactive stages (20, 21). One goal of this study was to determine whether CgA and/or its derived peptides function physiologically in parathyroid cell secretory cycling. In cultured bovine parathyroid cells, CgA and the CgA-derived peptides pancreastatin [equivalent to bovine CgA-(251-294)] and parastatin (14) [equivalent to bovine CgA-(348-393)] are potent inhibitors of PTH secretion (11, 12). Since CgA itself and, presumably, peptides derived by proteolytic cleavage are released from the cells, Fasciotto and co-workers (12) postulated that these peptides served as autocrine or paracrine agents to repress the secretion of actively secreting cells. Our data are not definitive on this point. To comfortably conclude that released CgA or CgA-derived peptides downregulate secretion, we would have expected to find that the majority of those cells that were active in stage 1 were inactive in stage 2. This was not observed. In fact, the percentage of cells that became nonsecretors in stage 2 only ranged from a low of 7% to a high of 23% (average of 16% in the four experiments listed in Figs. 2-5). It is likely, however, that our experimental conditions precluded the detection of down-regulation of secretion. Fasciotto et al. (12) showed that the action of pancreastatin was fully reversible in 90 min or less in cultured parathyroid cells, The minimum time that elapsed from the end of stage
OF PTH
AND
CgA
2641
1 to the beginning of stage 2 was 180 min, a period sufficiently long that cells inhibited by released CgA or a derived peptide could easily have recovered. Indeed, about half of the nonsecreting cells of stage 1 became secretors in stage 2. This could reflect recovery from prior inhibition of released peptides or a cell cycle totally unrelated to secretory events. Our experiments indicate that a greater number of cells formed CgA-related plaques than PTH-related plaques (85% vs. 63%), suggesting that CgA may be released independently of PTH. This noncoordinated secretion supports previous observations of Bajpai and Hamilton (22) that, in addition to being colocalized in secretory granules (23), CgA may be enriched compared to PTH in some secretory granules (22). One caveat to this explanation is that the sensitivity of the PTH and CgA antibodies might be sufficiently different so that the capability to detect PTH secretion is appreciably lower than that for CgA. In the sequential RHPAs in which the same antibody was used in stages 1 and 2, the total number of cells secreting a given hormone remained the same, although different cells were in the secretory phase. For example, when PTH secretion was assessed, 62% of cells in stage 1 and 63% of cells in stage 2 were actively secreting. In the case of CgA, 90% of the cells formed plaques in stage 1, and 83% in stage 2. This suggests that there may be a mechanism in normal tissue that regulates the number of cells that respond to a secretory stimulus at any particular time. The mechanism of this putative regulation remains to be elucidated. References 1.
Cohn DV, Zangerle R, Fischer-Colbrie R, Chu LLH, Eiting JJ, Hamilton JW, Winkler H 1982 Similaritv of secretorv urotein-l from parathyroid Proc Nat1 Acad
gland to chromogranin-A Sci USA 79:6056-6059
adrenal’medulla.
LA 1991 Regulation of paraRev 12:291-301 3. Ravazzola M, Orci L, Habener JF, Potts Jr JT 1978 Parathyroid secretory protein: immunocytochemical localization within cells that contain parathyroid hormone. Lancet 2:371-372 4. Cohn DV, Morrissey JJ, Shofstall RE, Chu LLH 1982 Cosecretion of secretory protein-l and parathormone by dispersed bovine parathyroid cells, Endocrinology 110:625-630 5. Cohn DV, Elting JJ, Frick M, Elde R 1984 Selective localization of the parathyroid secretory protein-l/adrenal medulla chromogranin A protein family in a wide variety of endocrine cells of the rat. Endocrinology 114:1963-1974 6. Cohn DV, Fasciotto BH, Gorr S-U, Parkins F, Shioi J, Levine MA, Greeley JH 1990 The putative role of secretory protein-I/chromogranin A as a precursor for regulatory hormones. In: Peterlik M, Bronner F (eds) Progressive Clinical and Biological Research. WileyLiss, New York, vol 332:51-66 7. Winkler H, Carmichael SW 1982 The chromaffin granule. In: Poisner AM, Trifaro JM (eds) The Secretory Granule. Elsevier, Amsterdam, pp 3-79 8. Cohn DV, Fasciotto BH, Gorr SU, Darkins F, Shioi H, Levine MA, Greely GH 1990 The putative role of secretory protein I/ chromogranin A as a precursor for regulatory proteins. In: Peterlik M, Bonner F (eds) Molecular and Cellular Regulation of Calcium and Phosphate Metabolism. Liss, New York, pp 51-56 9. Tatemoto K, Efendic S, Mutt V Mall G, Friestner GJ, Barchas JD 1986 Pancreastatin, a novel pancreatic peptide that inhibits insulin secretion. Nature (Lond) 274:476-478 10. Greeley GH, Thompson JC, Ishizuha J, Cooper CW, Levine MA, Gorr S-U, Cohn DV 1989 Inhibition of glucose-stimulated insulin 2.
Pocotte S, Ehrenstein
from
thyroid
hormone
J, Fitzpatrick
secretion.
Endocr
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
PARATHYROID
11.
12.
13.
14.
15.
16.
SECRETION
release in the perfused rat pancreas by parathyroid secretory proteinI (chromogranin A). Endocrinology 124:1235-1238 Fasciotto BH, Gorr S-U, DeFranco DJ, Levine MA, Cohn DV 1989 Pancreastatin, a presumed product of chromogranin-A (secretory protein-I) processing, inhibits secretion from porcine parathyroid cells in culture. Endocrinology 125:1617-1622 Fasciotto BH, Gorr SU, Bourdeau AM, Cohn DV 1990 Autocrine regulation of parathyroid secretion: Inhibition of secretion of chromogranin-A (secretory protein-I) and potentiation of secretion by chromogranin-A and pancreastatin antibodies. Endocrinology 127:1329-1335 Galindo E, Rill A, Bader M-F, Aunis D 1991 Chromostatin, a 20amino acid peptide derived from chromogranin A, inhibits chromaffin cell secretion. Proc Nat1 Acad Sci USA 88:1426-1430 Fasciotto BH, Nishijima A, Cohn DV 1992 Parastatin (porcine chromogranin A 347-391) inhibits parathyroid cell secretion. J Bone Mineral Res 7:S235 (Abstract) Fitzpatrick LA, Leong DA 1990 Individual parathyroid cells are more sensitive to calcium than a parathyroid cell population. Endocrinology 126:1720-1727 Ritchie CK, Cohn DV, Fitzpatrick LA 1992 Chromogranin-A secretion from individual parathyroid cells: effects of 1,25-
OF PTH
AND
CgA
Endo. Voll31.
1992 No 6
(OHhvitamin Da and calcium. Bone Mineral 18:31-40 17. Smith PF, Luque EH, Neil1 JD 1986 Detection and measurement of secretion from individual neuroendocrine cells using a reverse hemolytic plaque assay. Methods Enzymol 124:443-465 18. Neil1 JD, Smith PF, Luque EH, Munoz De Toro M, Nagy G, Mulcahey JJ 1987 Detection and measurement of hormone secretion from individual pituitary cells. Recent Prog Horm Res 43:175-228 19. Sun F, Ritchie CK, Hassager C, Maercklein P, Fitzpatrick LA 1992 Heterogeneous response to calcium by individual parathyroid cells. J Clin Invest, in press 20. Roth SJ, Raisz LG 1966 The course and reversibility of the calcium effect on the ultrastructure of the rat parathyroid gland on organ culture. Lab Invest 15:1187-1211 21. Shannon WA, Roth SI 1974 An ultrastructural study of acid phosphatase activity in normal, adenomatous and hyperplastic (chief cell type) human parathyroid glands. Am J Path01 77:493-501 22. Bajpai S, Hamilton J 1990 The isolation and partial characterization of bovine parathyroid secretory granules. Bone Mineral 9:9-22 23. Arps A, Dietel M, Lauritizen B, Elting JJ, Nierndorf A, Cohn DV 1987 Co-localization of parathyroid hormone and secretory proteinI in bovine parathyroid glands: a double immunocytochemical study at the electron microscopical level. Bone Mineral 2:175-183
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 28 October 2015. at 00:37 For personal use only. No other uses without permission. . All rights reserved.
