| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Division of Biomedical Sciences (D.C., C.B.K., L.Y., R.L., A.M.W.), University of California, Riverside, California 92521; and Division of Biochemical and Musculoskeletal Medicine (P.I.), Human Metabolism and Clinical Biochemistry, University of Sheffield Medical School, Sheffield S10 2RX, United Kingdom
Address all correspondence to: Ameae M. Walker, Division of Biomedical Sciences, University of California, Riverside, California 92521.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For PRL, homodimerization of the PRL receptor (PRL-R) is thought to result from PRL binding and to cause activation of a constitutively-associated tyrosine kinase, Janus kinase 2 (JAK 2) (5, 6, 7, 8). This activation of JAK 2 is followed by tyrosine phosphorylation of the PRL-R, a step thought to be involved in efficient recruitment of a transcription factor, signal transducer and activator of transcription 5 (STAT 5), to the transduction complex (9). STAT 5 is then thought to be phosphorylated by JAK 2, to dimerize and then enter the nucleus (10). STAT 5 belongs to a family of transcription factors with seven members to date, STATs 1–4, 5a, 5b, and 6 (10, 11, 12, 13, 14). In addition, STATs may also have splice variants, contributing to the potential diversity of their roles (15). Most STATs have at least semispecific roles in the immune system. STATs 1 and 2, for example, are critical for viral resistance, STAT 6 mediates the effect of interleukin 4, and STAT 4 is critical for interleukin 12 signaling. STAT 5, on the other hand, is activated by a wide variety of cytokines in addition to PRL. These include GH, erythropoietin, thrombopoietin, interleukins 2 and 3, tumor necrosis factor, and granulocyte-macrophage colony-stimulating factor (13, 14, 15). These cytokines can all activate the DNA-binding ability of STAT 5 and/or transactivate a ß-casein luciferase reporter gene in vitro but are unlikely to be involved in modulation of milk protein gene expression in vivo (16). Thus, there must be subtleties to STAT 5 activation, which may include differential activation of 5a and 5b, activation of splice variants of each, and coactivation of other STATs and other transcription factors.
PRL is produced in a number of posttranslationally modified forms (17). Phosphorylated PRL, which is quantitatively the most important posttranslationally modified product of the pituitary (18), has been shown to have antagonistic effects to unmodified hormone (19, 20, 21). Thus, unmodified PRL stimulates cell proliferation, whereas phosphorylated PRL blocks this response (20, 21). To more closely study the effects of phosphorylated PRL, we have produced a recombinant molecular mimic of phosphorylated PRL, which duplicates the antagonistic effects of naturally phosphorylated PRL on PRL-driven cell proliferation (22). The molecular mimic, produced by the substitution of an aspartate residue for the normally phosphorylated serine, has several advantages over the use of phosphorylated material. The most important advantage is removal of the possibility of conversion of phosphorylated PRL to unmodified PRL during the course of an experiment.
As an antagonist, both phosphorylated hormone and the molecular mimic are extremely potent (21, 22). This is shown by the fact that it takes less than one quarter the amount of each to completely block the proliferative activity of unmodified PRL (21, 22). This efficacy led us to consider the possibility that antagonism was not simply attributable to blockade of the unmodified PRL signal transduction pathways but may involve the generation of alternate intracellular signals. In addition, we hypothesized that phosphorylated PRL had a set of biological activities, beyond antagonism of PRL-stimulated cell proliferation, which require appropriate signal transduction pathways.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recombinant protein expression
Recombinant wild-type human PRL (hPRL) and the molecular mimic
of phosphorylated PRL (hPRL aspartate mutant) were prepared, as
previously described, by expression in Escherichia coli
(22). The aspartate mutant substitutes an aspartate for serine 179. To
assure comparability in the protein preparations, wild-type PRL and the
aspartate mutant were expressed at the same time and isolated and
refolded in parallel (22). The proteins were then tested for biological
activity (proliferative or antiproliferative, as appropriate) in an Nb2
bioassay and compared with an NIDDK (B3AFP3855A) preparation of hPRL
(22). The wild-type recombinant hPRL routinely had a higher biological
activity than the NIDDK PRL (22). Proteins were quantified by gel
densitometry, using NIDDK PRL to produce a standard curve. These
proteins have previously been tested for endotoxin contamination, and
none was detectable (22). Antagonist activity is also reversible by the
addition of excess wild-type hormone.
PRL treatment of the cells
Cells were placed in lactogen-free medium (complete medium
without FBS) 16–18 h before exposure to the purified PRL preparations.
At the time of the experiment, cells were pelleted and aliquoted in
lactogen-free medium at 108 cells/ml. One ml was placed
into 1.5-ml centrifuge tubes and allowed to equilibrate/recover for at
least 30 min at 37 C before PRL exposure. Cells were then exposed to
prewarmed (37 C) PRL solutions in the same medium by the addition of
100 µl of 11x concentrated material. Mixing was assured by gentle
inversion of the tube. PRL concentrations and times of exposure are
indicated in the figures. At the end of the appropriate exposure, cells
were washed twice with ice-cold stop buffer [20 mM Tris
(pH 7.4); 150 mM NaCl with 10 mM each of sodium
pyrophosphate, sodium fluoride, and sodium vanadate; 10 µg/ml each of
aprotinin, leupeptin, and pepstatin; 1 mM PMSF; 0.02%
sodium azide; and 1 mM EDTA] and then lysed by
resuspension in 1 ml of 1% Triton X-100 in the same buffer. After a
1-h rotation of the tubes at 4 C, lysates were subjected to
centrifugation at 13,000 x g for 10 min, and the
supernatants were saved for immunoprecipitation.
Immunoprecipitation
Total lysates were used for immunoprecipitation. Antibodies were
as follows: polyclonal anti-JAK 2 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA); polyclonal anti-STAT 5b (Santa Cruz Biotechnology, Inc., recognizes both STAT 5a and STAT 5b and
hence is described throughout as antipan STAT 5); polyclonal anti-STAT
5a (Santa Cruz Biotechnology, Inc., specific for STAT 5a);
polyclonal anti-STAT 5b (Zymed Laboratories, Inc., San
Francisco, CA, specific for STAT 5b). Four micrograms of purified
antibody was added to 1 ml of lysate and incubated for 2 h at 4 C.
Anti-PRL-R serum (no. 120) was kindly provided by Patricia Ingleton
(Sheffield University, United Kingdom), and 30 µl of whole serum was
used per milliliter of lysate. This antibody was raised against rat
liver receptor extracellular domain (24). Antibodies were precipitated
by incubation for 1 h further at 4 C after addition of 8 µl
washed protein G Sepharose slurry (Amersham Pharmacia Biotech, Piscataway, NJ). Preliminary experiments determined
that preclearing of the lysates by precipitation with nonspecific
antibodies was unnecessary. After precipitation with protein G, the
pellets were washed three times in stop buffer and then incubated in
reducing SDS sample buffer at 95 C for 10 min before loading on a 7.5%
polyacrylamide gel. Usually, minigels were used with electrophoresis at
40 mA constant current for 1 h. Results from the use of 11-cm
gels, run at 60 mA for 3 h, are shown in Figs. 4
and 5.
|
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
0.02
nM), and a maximum response was achieved at 500 ng/ml
(20 nM, Fig. 1A).
Concentrations up to 3.5 µg/ml were tested (higher concentrations not
shown). Reprobing with a mixture of anti-JAK 2 and antipan STAT 5
showed equivalent loading of the lanes (B). Similar dose response
relationships were seen between PRL and STAT 5 activation (C). Again,
equivalent loading is demonstrated (D). To compare the signaling
response between the wild-type recombinant PRL and NIDDK PRL, a
nonmaximum dose of 5 ng/ml was chosen. Fig. 2A shows phosphorylation of JAK 2 in
response to either NIDDK PRL or the recombinant wild-type PRL. At a
dose of 5 ng/ml, the wild-type PRL results in a greater degree of JAK 2
phosphorylation. This is consistent with a higher biological activity
in the Nb2 bioassay previously described (22) (see later for further
discussion). At this concentration of PRL, no differences were detected
between the two preparations, in terms of their ability to cause the
phosphorylation of STAT 5 (Fig. 2C). Eqivalent loading of lanes is
demonstrated in B and D, which are stripped and reprobed versions of A
and C, respectively. B shows immunoprecipitation with anti-JAK 2
followed by blotting with a mixture of anti-JAK 2 and antipan STAT 5
antibodies. D shows immunoprecipitation with antipan STAT 5 antibody
followed by blotting with the same mixture. E is equivalent to B,
except for the use of only one antibody for Western blotting. F is a
stripped and reprobed version of E, showing even loading. G shows a
composite of overexposed second-antibody controls, which demonstrate
the specificity of staining.
|
|
by comparing A (showing no phosphorylation) and B
(showing immunoprecipitation of STAT 5 with anti-JAK 2) and in the C
lanes of Fig. 2. It is more obvious in Fig. 2B than 2D, probably
because of the relative quantity of JAK 2 vs. STAT 5. When
anti-JAK 2 is used to precipitate, a substantial band of STAT 5 is
evident. When antipan STAT 5 is used to precipitate, a smaller band of
JAK 2 is evident. This preassociation result is unlikely to be
attributable to cross-reactivity of anti-JAK 2 and antipan STAT 5
antibodies, because both were produced against synthetic peptides
corresponding to specific regions of the appropriate molecules.
Additional controls demonstrated that it is not an artifact caused by
coblotting with anti-JAK 2 and antipan STAT 5, because blotting with
only one antibody gave the same result (E and F). G shows that this is
not an artifact caused by second-antibody staining. Similar
preassociation of JAK 2 and STAT 5 has also been observed in our
studies in both the mouse and bovine mammary cell lines, 31EG4 and MAC
T (data not shown).
At the higher dose of 500 ng/ml, JAK 2 phosphorylation reached a
plateau, and it was no longer possible to discern the difference
between wild-type recombinant and NIDDK PRL (Fig. 3). The higher concentration was used,
however, to stringently test for signaling of the aspartate mutant. A
quantity of 500 ng/ml aspartate mutant produced minimal JAK 2
phosphorylation (A) but almost equivalent phosphorylation of STAT 5
(C). Consistent with minimal activation of JAK 2 in response to the
aspartate mutant is the absence of phosphorylated JAK 2 in C after
immunoprecipitation with antipan STAT 5.
|
).
Only minimal JAK 2 phosphorylation occurred in response to the
aspartate mutant at 1 and 5 min, as well as at the 10-min time point
previously used. STAT 5 phosphorylation, on the other hand, was robust
at all time points. Preassociation of JAK 2 and STAT 5 in the absence
of PRL can again be seen in B. In these larger gels, STAT 5 resolves
into several bands. When titrated against one another, the aspartate mutant competes with wild-type hormone for JAK 2 activation (Fig. 5).
Because receptor phosphorylation is considered an integral part of PRL
signal transduction, we also examined receptor phosphorylation in
response to NIDDK PRL B3, wild-type recombinant PRL and the aspartate
mutant. Figure 6 shows phosphorylation of
the receptor in response to NIDDK PRL and wild-type recombinant and
very little, if any, receptor phosphorylation in response to the
aspartate mutant.
|
shows essentially equivalent
phosphorylation of STAT 5a in response to either wild-type or aspartate
mutant, whereas phosphorylation of STAT 5b in response to the aspartate
mutant was much reduced. Because of this, there was also reduced STAT
5a-STAT 5b heterodimerization, as shown on the right side of
B. Consistent immunoprecipitation of STAT 5b is illustrated in C
(right side).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. This is in
accord with previous reports from this lab of greater proliferative
activity of the recombinant wild-type vs. NIDDK PRL in the
Nb2 bioassay (22) and supports the role of JAK 2 in PRL signaling
leading to cell proliferation. Likewise, the essential absence of JAK 2
phosphorylation in response to the aspartate mutant would also support
this concept: a molecule, mimicking a proliferation antagonist
(phosphorylated PRL), would not be expected to phosphorylate JAK 2. By
competition experiments, it can also be shown that the aspartate mutant
blocks the ability of wild-type PRL to cause phosphorylation of JAK 2,
and we know that the aspartate mutant blocks cell proliferation by
wild-type hormone (21). Surprisingly, however, the aspartate mutant results in indistinguishable levels of STAT 5 phosphorylation when compared with NIDDK or recombinant wild-type hormone. To be sure that this result was not simply the consequence of an altered time course of JAK 2 activation, a time course experiment ranging from 1–10 min was conducted. As before, there was essentially no activation of JAK 2. It seems therefore that STAT 5 can be activated in Nb2 cells without, or with very minimal, use of JAK 2. This result demonstrates that the aspartate mutant not only blocks signaling from wild-type hormone but also signals via an alternate pathway.
Because STAT 5 was equally activated by the aspartate mutant and the aspartate mutant is antiproliferative, this result also suggests that STAT 5 activation can result from both proliferative and nonproliferative signaling from the PRL-R. What receptor-associated kinase is responsible for STAT 5 activation in response to the aspartate mutant is currently the subject of investigation. Other investigators have reported an association between the PRL-R and other tyrosine kinases (27, 28, 29) and activation of some in response to NIDDK PRL (27, 28, 29). Because NIDDK PRL contains both unmodified PRL (equivalent to wild-type) and phosphorylated PRL (equivalent to aspartate mutant), one would expect use of both proliferative and nonproliferative signaling pathways. STAT 5 phosphorylation by one of these previously described kinases, in response to the aspartate mutant, therefore is a reasonable possibility.
Receptor phosphorylation is considered important in efficient recruitment of STAT 5 to the transduction complex (9, 30). When examining receptor phosphorylation, we found no detectable phosphorylation of the receptor in response to the aspartate mutant. Thus, it seems that, with the aspartate mutant, we have efficient STAT 5 phosphorylation in the absence of receptor phosphorylation. Certainly, STAT 5 can be phosphorylated in the absence of receptor phosphorylation (31, 32, 33), but the current results suggest the existence of an alternate mechanism for efficient recruitment of STAT 5 to the transduction complex. This is consistent with previous reports, in PRL signaling and homologous systems, showing the presence of two distinct mechanisms of STAT 5 activation: one dependent on receptor tyrosine phosphorylation and the other not (31, 32, 33, 34).
Another unexpected finding in the current study was evidence of JAK 2 and STAT 5 association before any addition of PRL or phosphorylation of either JAK 2 or STAT 5. This implies an interaction between JAK 2 and STAT 5 without the involvement of SH2 domain-phosphotyrosine interactions. A search of the literature found no other description of an experiment in which both JAK 2 and STAT 5 had been localized by Western blot after immunoprecipitation for either JAK 2 or STAT 5. This is presumably the reason why such preassociation has not been previously reported. JAK 2 and recombinant STAT 5 have been shown to form a stable complex in vitro by Flores-Morales et al. (35). These authors, however, maintain the necessity for JAK 2 phosphorylation for this interaction to occur. Our results, on the other hand, show no necessity for this and no evidence that phosphorylation of JAK 2 increases the amount of STAT 5 that coimmunoprecipitates. The results are instead consistent with the idea that JAK 2 can be either a kinase or an adapter molecule, as suggested previously by Fujitani et al. (36).
Because both STAT 5a and STAT 5b have been demonstrated to be involved in PRL signaling (37, 38, 39, 40), we investigated the possibility that the aspartate mutant (through use of a second-receptor-associated kinase) would differentially activate one or the other STAT 5. The results show no distinction between the wild-type PRL and aspartate mutant in the activation of STAT 5a, but they do show a much reduced activation of STAT 5b in response to the aspartate mutant. This reduced activation of STAT 5b also results in reduced STAT 5a-STAT 5b heterodimers. Nonproliferative signaling, therefore, primarily activates STAT 5a. It will be of interest to investigate the actions of the aspartate mutant in tissues with well-defined nonproliferative responses to PRL.
PRL is normally secreted from the anterior pituitary as a combination of posttranslationally modified forms (17). In the rat, where the picture is not complicated by the presence of N-glycosylated forms (41), we have demonstrated the ratio of unmodified to phosphorylated PRL to vary reproducibly according to the stage of the estrus cycle and with pregnancy and pseudopregnancy (42, 43). Estrogen, for example, increases not only the total amount of PRL produced, but it also increases the proportion of it that is unmodified (42, 43, 44, 45). Estrogen therefore increases the Nb2 proliferative activity of the PRL (21, 22). We hypothesize therefore that when a nonproliferative activity of PRL is desired, the amount of phosphorylated PRL increases. By working through an alternate kinase to JAK 2, phosphorylated PRL would not activate other JAK 2-connected signaling cascades in the cell, which normally collectively result in proliferation.
In summary, by using a nonproliferative form of PRL, which has nevertheless been demonstrated to signal in Nb2 cells, the results presented show a dissociation between JAK 2 and STAT 5 activation. Moreover, they support an important role for JAK 2 in proliferative responses and for STAT 5a in both proliferative and nonproliferative responses to PRL.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received March 18, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. L. Williams, A. DeGuzman, H. Dang, M. Kawaminami, T. W. C. Ho, D. G. Carter, and A. M. Walker Common and specific effects of the two major forms of prolactin in the rat testis Am J Physiol Endocrinol Metab, December 1, 2007; 293(6): E1795 - E1803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Ueda, U. Ozerdem, Y.-H. Chen, M. Yao, K. T. Huang, H. Sun, M. Martins-Green, P. Bartolini, and A. M Walker A molecular mimic demonstrates that phosphorylated human prolactin is a potent anti-angiogenic hormone. Endocr. Relat. Cancer, March 1, 2006; 13(1): 95 - 111. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Goffin, S. Bernichtein, P. Touraine, and P. A. Kelly Development and Potential Clinical Uses of Human Prolactin Receptor Antagonists Endocr. Rev., May 1, 2005; 26(3): 400 - 422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Schroeder, J. L. Brockman, A. M. Walker, and L. A. Schuler Inhibition of Prolactin (PRL)-Induced Proliferative Signals in Breast Cancer Cells by a Molecular Mimic of Phosphorylated PRL, S179D-PRL Endocrinology, December 1, 2003; 144(12): 5300 - 5307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Badger, M. J. J. Ronis, S. J. Frank, Y. Chen, and L. He Effects of Chronic Ethanol on Hepatic and Renal CYP2C11 in the Male Rat: Interactions with the Janus-Kinase 2-Signal Transducer and Activators of Transcription Proteins 5b Pathway Endocrinology, September 1, 2003; 144(9): 3969 - 3976. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Y. H. Mak, J. J. Brosens, M. Christian, F. A. Hills, L. Chamley, L. Regan, and J. O. White Regulated Expression of Signal Transducer and Activator of Transcription, Stat5, and its Enhancement of PRL Expression in Human Endometrial Stromal Cells in Vitro J. Clin. Endocrinol. Metab., June 1, 2002; 87(6): 2581 - 2588. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A M Walker Unmodified and phosphorylated prolactin and gamma delta T cell development and function Lupus, October 1, 2001; 10(10): 735 - 741. [Abstract] [PDF]
|
|
|
|
|
|
S. Bernichtein, S. Kinet, S. Jeay, M. Llovera, D. Madern, J. A. Martial, P. A. Kelly, and V. Goffin S179D-Human PRL, a Pseudophosphorylated Human PRL Analog, Is an Agonist and Not an Antagonist Endocrinology, September 1, 2001; 142(9): 3950 - 3963. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Glasow, L.-C. Horn, S. E. Taymans, C. A. Stratakis, P. A. Kelly, U. Kohler, J. Gillespie, B. K. Vonderhaar, and S. R. Bornstein Mutational Analysis of the PRL Receptor Gene in Human Breast Tumors with Differential PRL Receptor Protein Expression J. Clin. Endocrinol. Metab., August 1, 2001; 86(8): 3826 - 3832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Xu, E. Kreye, C. B. Kuo, and A. M. Walker A Molecular Mimic of Phosphorylated Prolactin Markedly Reduced Tumor Incidence and Size When DU145 Human Prostate Cancer Cells Were Grown in Nude Mice Cancer Res., August 1, 2001; 61(16): 6098 - 6104. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. S. Bridges, B. A. Rigero, E. M. Byrnes, L. Yang, and A. M. Walker Central Infusions of the Recombinant Human Prolactin Receptor Antagonist, S179D-PRL, Delay the Onset of Maternal Behavior in Steroid-Primed, Nulliparous Female Rats Endocrinology, February 1, 2001; 142(2): 730 - 739. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Coss, L. Yang, C. B. Kuo, X. Xu, R. A. Luben, and A. M. Walker Effects of prolactin on osteoblast alkaline phosphatase and bone formation in the developing rat Am J Physiol Endocrinol Metab, December 1, 2000; 279(6): E1216 - E1225. [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]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
JAK
2, followed by Western blotting with antiphosphotyrosine (
