| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Department of Cell Biology (R.H., S.K., N.S., N.B.J.), University of Cincinnati Medical School, Cincinnati, Ohio 45267; Department of Ob-Gyn (R.S.), Indiana University School of Medicine, Indianapolis, Indiana 46202; and Vollum Institute (M.J.L.), Oregon Health Sciences University, Portland, Oregon 97201
Address all correspondence and requests for reprints to: Dr. Nira Ben-Jonathan, Department of Cell Biology, University of Cincinnati Medical School, 231 Bethesda Avenue, Cincinnati, Ohio 45267-0521.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The bioassay for PRF used GH3 cells stably transfected with the PRL promoter ligated to a luciferase reporter gene (GH3/luc). Coculture of mIL5 with GH3/luc cells induced cell-density dependent increases in PRL gene expression and release, whereas mIL39 cells showed negligible PRF activity. Incubation of GH3/luc cells with conditioned media from mIL5, but not mIL39 cells, stimulated PRL gene expression and release up to 10-fold. Coculture of mIL5 cells with primary rat anterior pituitary cells stimulated PRL, but not GH, release. Fractionation of mIL5 cell extracts by reverse phase HPLC resolved PRF activity into one major and one minor peak.
In conclusion, we have developed two novel and distinct cell lines from mouse intermediate lobe tumors. The first reported melanotroph cell line, mIL39, could provide a valuable model for studying dopaminergic regulation of POMC gene expression and release. In contrast, the mIL5 cells do not express POMC, D2 receptors, or SV40 Tag and appear to have been immortalized by a spontaneous mutation(s). These cells produce and secrete a potent PRF and could be used for the purification and biochemical characterization of PRF.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent evidence indicates that the IL participates in the regulation of PRL secretion. PRL release is modulated by both inhibitory and stimulatory factors that originate from the hypothalamus and pituitary (4, 5). Whereas dopamine is well established as the primary inhibitor of PRL secretion (6), the identity of the physiological stimulator of PRL release remains to be determined. Our laboratory has reported that an intact posterior pituitary is necessary for the suckling- and estradiol-induced rises in PRL (7, 8), suggesting the presence of a PRL-regulating factor (PRF) in this tissue. This was supported by in vitro experiments demonstrating that PRF is a potent stimulator of both PRL gene expression (9, 10) and release (11, 12) and is distinct from other PRL secretagogues (12, 13, 14). Cell separation on density gradients indicated that PRF is produced by a subpopulation of IL cells (15).
The small size of the mammalian IL has hampered efforts to isolate and purify PRF. This impediment could be overcome by the availability of PRF-producing cells. Targeted tumorigenesis (16) and gene knockout strategies (17, 18, 19) have generated several lines of transgenic mice that develop IL tumors. In particular, the use of a transgene composed of a truncated POMC promoter ligated to the transforming simian virus 40 large T antigen (POMC-Tag) resulted in tumor formation exclusively in the IL (16). We previously reported that primary tumors from heterozygous POMC-Tag mice, as well as secondary tumors developed in athymic nude mice, exhibited PRF activity that was distinct from known PRL secretagogues or POMC-derived products (14). Given the time and expenses required for generating tumors of relatively small size, an enriched and renewable source of PRF was clearly needed. The objective of the present investigation was to establish a PRF-producing cell line from such tumors and determine its cellular and biochemical characteristics.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rats. Ovariectomized adult female Fischer 344 rats (Harlan) were used as donors of AP cells for the PRF bioassay. All animals were maintained under constant temperature with food and water ad libitum. Animal experimentation was performed under an institutionally approved protocol according to the USPHS Guide for the Care and Use of Laboratory Animals.
Primary and secondary tumors
Primary IL tumors began to develop between 6–12 weeks of age in
both male and female heterozygous mice. Mice bearing large tumors were
easily recognized by lack of grooming and changes in the shape of the
skull as the IL tumor expanded and displaced the brain. Mice were
killed by cervical dislocation and the tumors aseptically removed.
Portions of the tumors were either processed for histological
examination or dispersed by trypsinization. The cells were placed in
culture or injected sc into nude mice to generate secondary tumors.
Approximately 1 x 107 primary tumor cells were used
for induction of secondary tumors, which developed as early as 20 days
after inoculation.
Cell culture
mIL cell lines. Of more than 15 primary IL tumors placed in
culture, cells from only two females, designated mIL5 and mIL39,
remained viable. Repeated serial dilutions generated clonal cell lines
that have been carried for over 70 passages. Cultures are maintained in
RPMI 1640 medium supplemented with 10% FBS and penicillin-streptomycin
(Pen-Strep; GIBCO-BRL, Gaithersburg, MD) at 37C under 5%
CO2. Both mIL5 and mIL39 cells were used for the following:
1) generation of secondary tumors in nude mice; 2) determination of
selected gene expression by RT-PCR; 3) assessment of PRF activity by
coculturing with either GH3/luc cells or primary
AP cells; 4) production of conditioned media (CM); and 5) morphological
characterization following growth on glass chamber slides (LabTech,
Naperville, IL).
Primary rat AP cells. Anterior pituitary glands were removed from ovariectomized Fischer 344 female rats and dispersed as previously described (21). Cells were plated at a density of 10,000 cells per well (10K) in 96-well plates (NUNC, Copenhagen, Denmark) and cultured for three days in serum free medium (SFM) composed of DMEM/F10 medium (50/50; GIBCO-BRL) supplemented with 1% ITS+ Premix (Collaborative Research, Bedford, MA), 1% nonessential amino acids and Pen-Strep. The mIL cells were trypsinized, added to the AP cells, and cocultured for 3 additional days. Media were removed and analyzed for PRL and GH by RIA.
GH3/luc cells. GH3 cells, obtained from the American Type Culture Collection (ATCC, Rockville, MD), were stably transfected with 2.5-kb rat PRL promoter ligated upstream of a firefly luciferase reporter gene as previously described (10, 22). The GH3/luc cells were maintained in Ham’s F-10 medium supplemented with 15% gelding serum and 50 µg/ml geneticin (G418; Sigma Chemical Co, St. Louis, MO); basal luciferase activity remained stable for over 18 months. The GH3/luc cells were used to evaluate PRF activity as judged by both PRL gene expression and release as described below.
Cell proliferation
The growth rate of mIL5 and mIL39 cells was determined by the
MTT (3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyltetrazolium bromide)
assay as previously described (23). Briefly, cells were plated at 5K
cells per well in 96-well plates (NUNC) precoated with protamine (2.5
mg/ml; Sigma) and Nu-Serum (Collaborative Research) and cultured in
RPMI 1640 medium containing 10% FBS. At the designated times, MTT (1
mg/ml; Sigma) was added and incubated for 3 h. After removal of
the medium, 100 µl of a developing solution (0.04 M
HCl/isopropyl alcohol) were added and optical density at 540 nm was
determined using a Dynatech (Chantilly, VA) MR700 Microplate
Reader.
RT-PCR
Total RNA from cells and tissues was extracted by Tri Reagent
(Molecular Research Center, Cincinnati, OH) and 5 µg were reverse
transcribed using Superscript II (GIBCO-BRL) and random hexamers as
described (22). PCR was performed on 10% of the RT product using the
following primers: 1) POMC, sense primer 5'-TGCCGAGATTCTGCTACAGTCG’3'
and antisense 5'-GGAAGTGACCCATGACGTACTT-3', with an expected product
size of 246 bp; 2) Dopamine receptor (D2 long and short),
sense primer 5'-CGCAGCAGTCGAGCTTTCAGA-3' and
antisense 5'-GCTCATCGTCTTAAGGGAGGT-3' with expected product sizes of
402 bp (long form) and 315 bp (short form); 3) estrogen receptor 
(ER), sense primer 5'-GGTCCAATTCTGACAATCGACG-3' and
antisense 5'-CGTATCCCGCCTTTCATC-3' with an expected product
size of 309 bp; 4) SV40 large T antigen (SV40 Tag), sense primer
5'-GCAATCGAAGCAGTAGCAATC-3' and antisense 5'-CAGCTAATGGACCTTCTAGG-3'
with an expected product size of 395 bp. All PCR reactions also
contained primers for the housekeeping gene ribosomal protein L19
(RPL19), sense primer 5'-AGTAGTCTTAGGCTACAGAAG-3' and
antisense 5'-TTCCTTGGTCTTAGACCTGCG-3' with an expected product size of
500 bp. All primer sets, except for SV40 Tag, were designed to span
introns. Underlined nucleotides represent mismatches between
mouse and rat sequences. For SV40 Tag, RNA was treated with DNase to
remove contaminating DNA. Cycle conditions were: 94 C for 30 sec, 58 C
for 30 sec, and 72 C for 30 sec for 32 cycles. Products were separated
on a 1.5% agarose gel containing ethidium bromide and
photographed.
Bioassay for PRF
PRL gene expression.GH3/luc cells
(20K) were plated on protamine/Nu-Serum-coated 96 well plates and
incubated in SFM for 3 days. The cells were then washed and subjected
to the following treatments: 1) coculture with mIL cells; 2) incubation
with CM from mIL cells; 3) incubation with reconstituted HPLC
fractions. After 18 h, media were removed for PRL and GH analysis
by RIA and the cells were lysed by adding 50 µl of lysis buffer
(Promega, Madison WI). After incubation for 15 min at 37 C, 20-µl
aliquots in duplicate were transferred to black 96-well plates (Packard
Instrument Co, Downers Grove, IL) and 80 µl of luciferin (Promega)
were added. Luciferase activity, as a measure of PRL gene expression,
was quantitated by luminometry using a Packard TopCount.
PRL and GH release. The concentrations of PRL and GH in media from both GH3/luc and primary AP cells were determined by a modified RIA. NIDDK rat PRL and rat GH RIA kits with rPRL RP-3 and GH RP-2 as reference preparations, respectively, were used. Briefly, media aliquots were diluted in PBS containing 0.1% BSA in opaque white 96-well plates (Packard) to a final volume of 100 µl. After adding 50 µl each of primary antibody and iodinated hormone, the plates were incubated for 2 days at 4 C. Protein A (50 µl) was then added and the plates centrifuged at 4000 x g for 10 min. The supernatant was aspirated and the pellet dissolved in 20 µl of 0.1 N NaOH followed by 200 µl of scintillation fluid (Microscint 20, Packard). The plates were sealed with TopSeal (Packard) and after vigorous mixing, radioactivity was counted using a Packard TopCount.
HPLC fractionation
Approximately 18 x 106 mIL5 cells were
pelleted, washed with saline, and extracted by sonication in 1
N acetic acid. After freeze-thaw, extracts were centrifuged
at 10,000 x g for 20 min at 4 C. The supernatant was
loaded on an analytical C-18 reversed phase column (4.6 mm x 25
cm; Rainin, Woburn, MA) and fractionated with 0.1% trifluoroacetic
acid at an increasing gradient of acetonitrile (AcN) from 0–60% over
60 min at 1 ml/min as previously described (14). One-milliliter
fractions were collected and aliquots were pooled from every four
fractions. After lyophilization, fractions were reconstituted in SFM
and incubated with GH3/luc cells as described
above.
Data analysis
All experiments were performed at least three times. PRF values
are expressed as a percentage of control values, i.e.
GH3/luc or primary AP cells incubated alone.
Data were analyzed by ANOVA followed by Student’s t
test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, top left panel compares a
brain with attached pituitary from a wild-type mouse with two brains
from transgenic animals with massive IL tumors. These tumors grow at an
unpredictable rate and can become 100-fold larger than a whole
pituitary. In spite of their massive size, the tumors do not appear to
infiltrate the brain proper. At an advanced tumor stage, animals begin
to lose weight and eventually die from complications caused by brain
compression. The two females whose IL tumors are shown in Fig. 1 were 1
yr old at the time they were killed. A representative section of a
primary IL tumor stained with hematoxylin and eosin (H&E) is shown in
Fig. 1, lower left panel. The cells are small and ovoid with
prominent nuclei and variable cytoplasmic staining. The tumor is highly
vascularized, as evident by the presence of blood vessels filled with
red blood cells.
|
, upper right
panel. Within 45 days of innoculation, the tumor reached a 1-cm
diameter and weighed 0.7 g. An H&E stained section of this tumor
(lower right panel) depicts large, spindle-shaped cells with
uniform cytoplasmic staining. In contrast to the slow growth rate of
this tumor, secondary tumors generated from either primary tumor cells
or from mIL39 cells grew rapidly. In fact, mIL39-derived tumors have
reached sizes as large as 4 cm in diameter and weighed as much as 3.5 g
(data not shown).
Cellular characteristics of the mIL cell lines
Although both mIL5 and mIL39 cells originated from primary
IL tumors, they differ dramatically in size and morphology. The mIL5
cells, seen by phase contrast microscopy in Fig. 2A, are gigantic cells with numerous
branched cytoplasmic extensions of variable length that make focal
contacts with neighboring cells. As shown in detail in panel C, these
cells have big polymorphic nuclei containing numerous nucleoli and a
large cytoplasmic volume. Extremely large multinucleated cells are
commonly observed in less confluent cultures. The mIL39 cells, depicted
in panel B and D, are much smaller and bipolar. These cells have a low
cytoplasmic to nuclear ratio and project long and unbranched processes.
Clearly, these photographs underscore the dramatic difference in cell
size and nuclear diameter between mIL5 and mIL 39 cells. The presence
of several mitotic figures in both cell types are indicative of their
fast replication rate.
|
). Under
the standard culture conditions (RPMI containing 10% FBS), the mIL5
cells have a fast doubling time of 15 h, whereas mIL39 cells
double every 21 h. This growth rate remained relatively stable
over a period of 6–8 months. The cells also proliferate, albeit
slower, under reduced serum conditions, e.g. 1% FBS.
|
, upper
panel). However, of the two cell lines, only mIL39 expressed POMC.
Further analysis for expression of SV40 Tag revealed that only mIL39,
but not mIL5 cells, were positive for SV40 Tag (Fig. 4, lower
panel). Although mIL5 cells do not express SV40 Tag messenger RNA,
they carry the POMC-Tag transgene, as judged by PCR of genomic DNA
(data not shown).
|
, upper panel, dopamine
receptors are expressed by all samples examined except for mIL5. The
long form of the D2 receptor predominates, but the short form (315 bp)
is also weakly expressed by the same tissues. Since we previously
reported that estrogen increases PRF activity in the IL (8, 22), we
examined the expression of estrogen receptors in these cells. As
shown in Fig. 5, lower panel, ER is expressed by all
tested samples, except for rPP8, a newly developed rat posterior
pituitary endothelial cell line (unpublished observation) that served
as a negative control.
|
, left panel). As few as
2.5 x 103 mIL5 cells stimulated PRL gene expression
2.5-fold. In contrast, as many as 40 x 103 mIL39
cells were needed to increase PRL gene expression 2-fold, with little
evidence for density dependence. To determine whether PRF is a secreted
product, 4 x 105 mIL5 or mIL39 cells were grown in
SFM and the CM collected after 3 days. As shown in Fig. 6, right
panel, 12.5, 25, and 50% CM from mIL5 cells induced 4-, 5-, and
10-fold increases in PRL gene expression, respectively. In contrast, at
all concentrations tested, CM from mIL39 cells resulted in only 2- to
3-fold increases. The pattern of PRL release from the same
GH3/luc cells paralleled the luciferase
activity, supporting the presence of substantial PRF activity in mIL5,
but not mIL39 cells (Fig. 7). As
determined by both rat and mouse PRL RIAs, PRL itself was undetectable
in CM collected from either mIL5 or mIL39 cells. The release of GH was
unchanged by coculture with mIL cells or by incubation with their CM
(data not shown). Coculture of mIL cells with nontransfected
GH3 cells induced a similar rise in PRL release to that
obtained using GH3/luc cells (data not
shown).
|
|
, upper panel). PRF activity
of mIL39 cells, compared with mIL5, was significantly lower
(P < 0.01 at 20K). As evident, GH release from these
cocultures was not significantly changed (Fig. 8, lower
panel), further supporting the hormonal specificity of PRF
activity.
|
, PRF activity resolved into a major
peak at 45–48% AcN and a minor peak at 38–40%. PRL release, both in
elution pattern and magnitude, closely paralleled luciferase gene
activation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The mouse IL is a minute tissue that contains only 2 x 105 cells (28). Unlike most of its neighboring AP cells, IL cells maintain a robust proliferative capability, as evident by increased cell division in response to pituitary stalk section (3) or treatment with haloperidol (29) or interleukin-1ß (30). The IL expresses a very high density of D2 receptors (26, 27) and dopamine exerts an inhibitory control over both POMC gene expression (29, 31) and melanotroph proliferation (32, 33). Indeed, the onset of dopaminergic innervation of the IL in early postnatal life coincides with cessation of cellular proliferation (32). Whether maintenance of a relatively constant number of IL cells during adulthood reflects a balance between proliferation and apoptosis, remains to be determined.
A common strategy for targeted tumorigenesis is via transgenes encoding a transforming viral protein (e.g. SV40 Tag) under the transcriptional control of tissue-specific gene promoters. The rat 5'-flanking sequences (-706 to +64) of the POMC gene used to generate the POMC-Tag transgenic mice appear insufficient for transcriptional activity in extrapituitary sites (34). However, the reason for tumor formation in IL melanotrophs, but not AP corticotrophs, is not clear. Presumably, SV40 Tag induces neoplastic transformation by binding to the protein products of the tumor suppressors p53 and retinoblastoma (Rb) genes (35), which are involved in cell cycle regulation, differentiation and survival. It is of interest that large IL tumors also develop in 95% of Rb+/- heterozygous mice (19) subsequent to a loss of the remaining wild-type RB allele (28). Further, disruption of the p27kip1 gene, whose protein product inactivates cyclin/cyclin-dependent kinase complexes, also caused selective neoplastic growth in the IL (17, 18). These, together with the high incidence of spontaneous IL tumors in several mammalian species (36), suggest that IL cells, for yet unknown reasons, are especially sensitive to loss of negative regulators of cell cycle progression.
In spite of the general notion that the IL is composed of homogeneous cells, two distinct subpopulations of melanotrophs have been identified that differ in secretory activity, staining properties and receptor expression (37, 38, 39). The IL also contains several non-POMC expressing cells, including marginal, folliculo stellate and interstitial cells (40, 41, 42), whose function is poorly understood. Microscopic and biochemical observations revealed that IL tumors in either POMC-Tag (16) or Rb+/- (19) heterozygous mice begin as multifocal nodules that progress into large tumors by clonal expansion. Given the heterogeneity of the IL and the polyclonal origin of IL tumors, the emergence of two cell lines with different properties was not unexpected. Yet, although both mIL cell lines were derived from transgenic mice with IL tumors, SV40 Tag was expressed only by mIL39 cells. Presumably, only the POMC-expressing mIL39 cells should drive the production of Tag, resulting in cell transformation. Therefore, it appears that mIL5 cells have originated from a cell other than melanotroph that became immortalized by a spontaneous mutation(s). Alternatively, mIL5 cells represent a stem cell or a dedifferentiated melanotroph that no longer expresses typical cellular markers. Future studies will explore cellular markers that might reveal the origin and identity of mIL5 cells
Pituitary hormone research has benefitted from the availability of
immortalized cell lines such as the rat somatomammotroph
GH3 cells (43) and the ACTH-producing mouse AtT20 cells
(44). Cell lines provide a renewable homogeneous population of
cells that can be manipulated under controlled conditions. Although
both melanotrophs and corticotrophs produce POMC, they differ in their
expression of POMC processing enzymes (24), main secretory products
(45), and expression of receptors for glucocorticoids (46), CRH (47)
and dopamine D2 receptors (26, 27). Work is underway to
determine whether mIL39 cells process POMC in a melanotroph-specific
pattern and whether they secrete POMC peptides such as MSH and
ßendorphin in a regulatable manner.
Regardless of its cellular origin or stage of differentiation, the mIL5 cell line has remained stable for many generations. In validation of PRF production by these cells, several criteria were fulfilled, including stimulation of PRL gene expression and release and lack of effect on GH. PRL release in response to coculture with mIL5 cells increased in both the somatomammotroph GH3 cell line and primary rat AP cells. The robust stimulation of PRL gene expression and release by serum-free CM from mIL5 cells confirms that PRF is a secreted product, as would be expected from our in vivo demonstration that the IL participates in the control of PRL release (7, 8). Furthermore, in support of our previous report (14), the HPLC elution profile of PRF extracted from mIL5 cells resembles that extracted from primary IL tumors and differs from POMC products and known PRL secretagogues. Finally, the small size of the IL and the time-consuming task of producing sufficient IL tumors have made previous purification attempts an arduous task. Given the development of the PRF-producing cell line, the challenge of PRF isolation and structural determination should be finally met.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received August 8, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-melantropin. Proc Nat Acad Sci USA 90:4922–4926This article has been cited by other articles:
 |
|
 |
|
  M. Miller, S. Chen, J. Woodliff, and S. Kansra Curcumin (Diferuloylmethane) Inhibits Cell Proliferation, Induces Apoptosis, and Decreases Hormone Levels and Secretion in Pituitary Tumor Cells Endocrinology, August 1, 2008; 149(8): 4158 - 4167. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Lin Chang, J. Roh, J.-I. Park, C. Klein, N. Cushman, R. V. Haberberger, and S. Y. T. Hsu Intermedin Functions as a Pituitary Paracrine Factor Regulating Prolactin Release Mol. Endocrinol., November 1, 2005; 19(11): 2824 - 2838. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bertrand, G. Y. K. Ng, M. G. Purisai, S. E. Wolfe, M. W. Severidt, D. Nouel, R. Robitaille, M. J. Low, G. P. O'Neill, K. Metters, et al. The Anticonvulsant, Antihyperalgesic Agent Gabapentin Is an Agonist at Brain gamma -Aminobutyric Acid Type B Receptors Negatively Coupled to Voltage-Dependent Calcium Channels J. Pharmacol. Exp. Ther., July 1, 2001; 298(1): 15 - 24. [Abstract] [Full Text]
|
|
|
|
|
|
S. Khurana, S. Ranmal, and N. Ben-Jonathan Exposure of Newborn Male and Female Rats to Environmental Estrogens: Delayed and Sustained Hyperprolactinemia and Alterations in Estrogen Receptor Expression Endocrinology, December 1, 2000; 141(12): 4512 - 4517. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Freeman, B. Kanyicska, A. Lerant, and G. Nagy Prolactin: Structure, Function, and Regulation of Secretion Physiol Rev, October 1, 2000; 80(4): 1523 - 1631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Dannies Editorial: A New Releasing Factor? with Biotechnology and a Little Bit of Luck Endocrinology, December 1, 1997; 138(12): 5085 - 5086. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
