- Email: [email protected]
FMS-like tyrosine kinase 3 interacts with the glucocorticoid receptor complex and affects glucocorticoid dependent signaling
Available online at www.sciencedirect.com
Biochemical and Biophysical Research Communications 368 (2008) 569–574 www.elsevier.com/locate/ybbrc
FMS-like tyrosine kinase 3 interacts with the glucocorticoid receptor complex and affects glucocorticoid dependent signaling Abolfazl Asadi 1, Erik Hedman 1,*, Christina Wide´n, Johanna Zilliacus, ˚ ke Gustafsson, Ann-Charlotte Wikstro¨m Jan-A Department of Biosciences and Nutrition, Division of Medical Nutrition, Karolinska Institutet, NOVUM, S-141 86 Stockholm, Sweden Received 16 January 2008 Available online 7 February 2008
Abstract The glucocorticoid receptor (GR) forms part of a multiprotein complex consisting of chaperones and proteins active in glucocorticoid signaling and other pathways. By immunoaffinity purification of GR, followed by Edman sequencing and Western blotting, we identified the FMS-like tyrosine kinase 3 (Flt3) as a GR-interacting protein in rat liver and hepatoma cells. Flt3 interacts with both non-liganded and liganded GR. The DNA-binding domain of GR is sufficient for Flt3 interaction as shown by GST-pull down experiments. Studies of the effects of Flt3 and its ligand FL in glucocorticoid-driven reporter-gene assays in Cos7 cells, show that co-transfection with Flt3 and FL potentiates glucocorticoid effects. Treatment with FL had no effect on GR location and Dex induced translocation of GR was unaffected by FL. In summary, GR and Flt3 interact, affecting GR signaling. This novel cross-talk between GR and a hematopoietic growth factor might also imply glucocorticoid effects on Flt3-mediated signaling. Ó 2008 Elsevier Inc. All rights reserved. Keywords: Glucocorticoid receptor; Glucocorticoids; Flt3; GST-pull down; Immunoprecipitation
The glucocorticoid receptor (GR) belongs to a large family of nuclear receptors and acts as a ligand-dependent transcriptional regulator [1]. A number of natural and synthetic glucocorticoids can activate GR and thereby affect survival, proliferation and differentiation of cells. These effects are vital, as indicated by the fact that targeted disruption of the GR gene leads to neonatal death [2], mainly as a consequence of impaired lung development. The transcriptional effects of glucocorticoids are mediated via its binding to GR. GR then translocates to the nucleus where it affects gene transcription by binding to certain DNA sequences, called glucocorticoid response elements (GRE). Furthermore, GR also acts via interaction with other proteins in a number of signal transduction pathways. This mode of GR action often leads to transrepression of the other signaling pathway. Notably, this is the *
1
Corresponding author. Fax: +46 8 711 66 59. E-mail address: [email protected] (E. Hedman). The authors contributed equally to this work.
0006-291X/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2008.01.146
mode of action for the NF-jB and the AP-1 signaling pathways [3]. Clinically, glucocorticoids are frequently used as treatment of inflammatory diseases, for immunosuppression after transplantation and to treat certain malignant diseases as acute childhood leukemias. The FMS-like tyrosine kinase 3 (Flt3) belongs to the class III receptor tyrosine kinase (RTK) family [4] in which also the c-KIT, FMS and the PDGF are found. In humans Flt3 is expressed mainly in hematopoietic cells [5] and has been shown to play a role in proliferation and survival of hematopoietic progenitor cells, as well as in differentiation of early B, NK, and dendritic cell progenitors [6]. Binding of the Flt3 ligand FL, which is present both as a membrane bound and as a soluble glycoprotein, to the extra-cellular Flt3-domain, leads to conformational changes that induce and stabilize receptor dimerization [7]. Receptor dimerization brings the kinase domains together and this induces tyrosine auto-phosphorylation and further kinase activity. Mutations in Flt3 leading to constitutive kinase activity have been shown to be present
570
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
in acute myeloid leukemia (AML) as well as in some other hematological malignancies, especially during blast crisis [8], and has been correlated to a poor prognosis in AML patients. In our attempts to understand GR signaling based on protein–protein interactions we have developed an immunoaffinity based purification protocol for GR which allows co-purification of GR-interacting proteins [9]. In this study we report that Flt3 is part of the GR complex and that this has functional consequences for GR activity. Materials and methods SDS–PAGE, Western blotting, and immunostaining. The proteins samples were boiled in SDS sample buffer and resolved on 10% (v/v) SDS– PAGE gels under reducing conditions at 30 mA/gel for 50 min with the mini-Protean III gel system (Bio-Rad, Stockholm, Sweden). Western blotting and immunostaining proceeded as in [10] with the following primary antibodies: anti-GR mAb250 (also called mAb 7) [11] 0.013 lg/ ml, anti-Flt3/Flk2 1:500 (Santa Cruz Biotechnology Inc., Santa Cruz, CA), anti-Lamin A/C (Cellsignaling Technology, MA, USA) 1:1000 and anti-GADPH (Abcam, Cambridge, UK) 1:5000, anti-DBD 3.25 lg/ml, anti-GST 1:1000. Horseradish peroxidase (HRP)-conjugated, -sheep antimouse or -donkey anti-rabbit 1:50,000 (both from GE Healthcare, Uppsala, Sweden) were used as secondary antibodies. The membranes were finally incubated either with SuperSignalÒ (Pierce, Rockford, IL, USA) or with ECLTM Advanced Western Blotting Detection Kit (GE Healthcare, Uppsala, Sweden) according to the instructions from the manufacturer. Co-immunoprecipitation of GR from rat liver cytosol using antibodies directed against Flt3. Liganded/activated- and non-liganded/non-activated GR complexes were prepared as previously described in [9]. Approval by the Ethical Committee was obtained as required. Protein (1 mg) from cytosolic extracts were diluted to a total volume of 300 ll in 10 mM Tris, pH 7.8, 0.1% (v/v) Nonidet P-40 (IGEPAL), 0.15 M NaCl, 1 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride and 0.5 lg/ml of each aprotinin, leupeptin, pepstatin, and 1 lg Flt3/Flk2 antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA) was added. After 1 h, 20 ll of resuspended volume of Protein A/G–Agarose (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) was added and after another 3 h, the pellet was washed three times in PBS followed by SDS–PAGE and Western blotting. Two micrograms of agarose coupled with normal rabbit antibodies were used as a negative control. All steps were carried out at +4 °C. Plasmids. The plasmid pcDNA-hFlt3 containing human Flt3 cDNA was constructed by PCR-recloning of a full-length human Flt3 coding fragment, from the pcDSRa plasmid (a kind gift from Dr. Oliver Rosnet) into the pcDNA3.1 D/V5-HIS-TOPO expression vector (Invitrogen, Carlsbad, CA) using 50 -CACCATGGGCCCAGGAGTTCTGCTGCT-30 as a forward primer and 50 -CGAATCTTCGACCTGAGCCT-30 as a reverse primer. GST-GR fusion proteins were expressed from pGEX-4T-1 plasmids (GE Healthcare, Uppsala, Sweden). The plasmids expressing GST-GR-N-terminal (N+DBD+hinge) domain (aa 1–567), GST-GRDBD (aa 418–503), GST-GR-C-terminal (aa 568–777) domain and GSTGR-full-length (aa 1–777) were constructed by subcloning of BamHI/XhoI fragments corresponding to different domains of GR from the vector pcDNA3.1 D/V5-His-TOPO containing each fragment. The expression plasmid pCMV4-hGR and the luciferase reporter plasmid p19-tk-luc have been described previously [12]. Protein expression and GST-pull down. GST, GST-GR-N-terminal and GST-GR-full-length were expressed in the Escherichia coli strain BL21 (DE3) pLysS whereas GST-GR-DBD and GST-GR-C-terminal were expressed in the E. coli strain BL21 Competent cells. Bacteria were grown to an OD600 of 0.5, and the expression of fusion proteins was accomplished by following the instructions from the manufacturer. GST-pull down was performed essentially as in [13].
Cell culture, transient transfections and reporter-gene assay. Cos7 cells were maintained in Dulbecco’s modified Eagle’s medium containing sodium pyruvate and 1 g/l glucose supplemented with 10% (v/v) fetalbovine serum (FBS), 100 U/ml penicillin and 100 lg/ml streptomycin. Rat hepatoma cells H4-II-E-C3 (HTC cells) (ATCC, Manassas, VA), were grown as described in [14]. Cells used for reporter-gene assays were seeded out on 24-well plates at a density of 2 104 cells/well, one day prior to transfection. Transient transfections of cells were performed with FuGENETM-6 transfection reagent (Roche Diagnostics Scandinavia AB, Stockholm, Sweden) at a ratio of 3 ll/lg DNA, according to the manufacturer’s protocol. The total amount of plasmid DNA used was normalized, by the addition of the empty expression plasmid pcDNA3.1 D/ V5-His-TOPO, to a total amount of 112.5 ng plasmid/well. After 18 h incubation, the transfection mixture was replaced with fresh medium, containing 100 nM Dex in 0.005% ethanol or 0.005% ethanol only. 24 h later cells were harvested for enzyme assay. Luciferase assay was performed with the Luc-ScreenÒ System (Applied Biosystems, Stockholm, Sweden) according to the instructions from the manufacturer in a 1450 MicroBeta instrument (Wallac Sverige AB, Stockholm, Sweden). GR and Flt3 distribution in rat hepatoma cells after treatment with dexamethasone and FL. The day of the experiment the HTC cells were washed three times in PBS and then cultivated for an initial 6 h in RPMI media as stated above but in the absence of FBS. Then the cells were exposed to: only media as a control, 106 M Dex or 10 ng/ml FL in various combinations as indicated in Fig. 4. The cells were further cultivated for 2, 15, 30 or 60 min (Fig. 4A and B). Cytosolic and nuclear extracts were subsequently prepared according to [14]. Protein concentrations of the extracts were determined by a Bradford assay and equal protein concentrations for cytosol and nuclear extracts, respectively, were subjected to SDS–PAGE. GADPH and Lamin A were used as loading controls for cytosolic and nuclear extract, respectively. GR immunoprecipitation in rat hepatoma cell cytosol. HTC cells were grown and cytosolic extracts was prepared as described above. Equal protein concentrations, for each individual experiment, were immunoprecipitated according to [14] with the following modifications. The cytosol fractions were preabsorbed with normal mouse IgG coupled to CNBr-activated Sepharose 4B and immunoprecipitated using the mAb250, followed by SDS–PAGE and Western blotting. All steps were carried out at +4 °C.
Results and discussion The immunoaffinity-purified rat liver GR complex contains Flt3 We immunopurified non-liganded/non-activated GR receptor complex from rat liver cytosol. The GR complex was further analyzed by SDS–PAGE and Edman sequencing. One of the few proteins that were not N-terminally blocked contained the amino acid sequence: KWEFPRE NLAEFGKVLGSGAFG that corresponds to aa 602–622 in the human isoform of Flt3, as based upon the protein sequence derived from the NCBI nucleotide database (NCBI Accession No. XM_221874). The GR-Flt3 interaction was also confirmed by a reverse immunoprecipitation, using a Flt3 antibody. Fig. 1A demonstrates that immunoprecipitated cytosolic Flt3, interacted with part of the GR complex. As GR endogenously interacts with hsp90 it is difficult to conclude whether the interaction between GR and Flt3 is direct or not. Especially since hsp90 has been reported to form a protein complex including Flt3 and p23 [16]. However, since our immunoprecipitation results indicated that the
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
Ip
+
S
GR
kDa GR
Flt3
200 116 97
kDa GR 200 116 97
interaction between the two different states of GR ligandation as seen for many other GR-interacting proteins [10].
Flt3
Human Flt3 interacts with the DNA-binding domain of human GR
66
66
Fig. 1. GR co-immunoprecipitates with Flt3 in rat liver cytosol and Flt3 associates with both non-liganded/non-activated and liganded/activated GR from rat liver cytosol. (A) One out of two experiments. Rat liver cytosol containing non-liganded/non-activated GR was prepared from adrenalectomized rats. Immunoprecipitation was performed using 1 lg of Flt3 antibody. The immunoprecipitate (Ip) was resolved on a 10% SDS– polyacrylamide gel and transferred to a PVDF membrane. Immunostaining showed that GR co-immunoprecipitated with Flt3. The positive control (+) contains 50 lg total protein and the total input for the immunoprecipitation (Ip) and the supernatant (S) was 1 mg of protein. The supernatant (S) shows the fraction of GR that has not coimmunoprecipitated with the Flt3 antibody. (B, C) One representative experiment out of four. Cytosol from four livers from adrenalectomized rats was and divided into two equal parts. To one part, sodium molybdate was added, to obtain non-liganded/non-activated GR complexes (B), and to the other part ligand was added and heat-activation was performed to obtain liganded/activated GR complexes (C) before purification by immunoaffinity chromatography was performed. GR and Flt3 antibodies were used, as indicated in the figure, to detect proteins in the immunoprecipitate by Western blotting.
interaction between GR and Flt3 occurs both in complexes where hsp90 is present, i.e. the non-liganded GR complex (Fig. 1B), as well as in complexes were hsp90 is absent, i.e. liganded GR complex (Fig. 1C) a direct interaction seems possible. Previous studies have demonstrated that NF-jB [14], Raf-1 and 14-3-3 [9] preferentially interacts with liganded GR, whereas many chaperone proteins, such as hsp90 and others [15] mainly interact with non-liganded GR. We therefore investigated if the addition of glucocorticoids affected the GR-Flt3 interaction. By using Flt3 antibody to immunodetect Flt3 in liganded/activated GR complexes we found that Flt3 also co-purified with liganded/activated GR (Fig. 1C). This finding does not exclude a variable
kDa G D 150 100 50 37 25
N
To determine which domains of GR were necessary for the interaction between GR and Flt3, we constructed a set of human GST-GR fusion proteins that were incubated with in vitro transcribed/translated Flt3. The autoradiograph of the GST-pull down experiment (Fig. 2C) showed that the C-terminal, ligand-binding domain (LBD) (C) of GR did not interact with Flt3 above the level of the background with non-fused GST (cf. G). However, the DNAbinding domain (DBD) (D), the N-terminal fragment (N), as well as full-length GR (F) could interact with Flt3. Considering that the hsp90 interaction with GR is attributed to the GR-ligand binding domain [17] and that Flt3 interacts with the DNA-binding domain (lacking LBD) those results also speaks in favor of a direct interaction between GR and Flt3. Glucocorticoid dependent reporter-gene activity is affected by Flt3 and FL Cos7 cells have previously been shown to lack endogenous GR [20]. By Western blotting we could demonstrate that Cos7 cells also lacked endogenous Flt3 (data not shown). To study the effects of Flt3 and FL on glucocorticoid signaling, we transfected Cos7 cells with the plasmids pCMV4-hGR, pc-DNA-hFlt3 and the glucocorticoid-driven reporter-gene p19-tk-luc. By addition of FL and Dex in various combinations we studied how glucocorticoid dependent reporter-gene activity was affected. Addition of Flt3 and FL in the absence of Dex did not activate the GR dependent reporter gene (data not shown). However, addition of only Dex induced the reporter gene as expected (Fig. 3B). Neither Flt3 alone nor FL alone had any addiGST-pull down
anti-DBD
anti-GST C F
kDa G D 150 100 50 37
571
N
C F
kDa 150
I
G
D
N
C
F Flt3
100
25
GST Fig. 2. Flt3 interacts with the DNA-binding domain of GR. The figure shows one representative experiment out of two. (A) GST and GST-GR fusion proteins expressed in vitro in Escherichia coli were resolved on 10% SDS–PAGE gels and transferred to PVDF membranes. The membranes were stained with a GST antibody (A); GST-GR fusion proteins are encircled. The same amount of total protein for each protein extract was used in (A–C). In (B) the DBD antibody was used to localize the 37 kDa GST-DBD fusion protein, D. Full-length human Flt3 was in vitro transcribed/translated in rabbit reticulocyte lysates in the presence of 35S-methionine. (C) An autoradiograph of various GST-GR glutathione agarose precipitates in the presence of 35Smethionine labeled Flt3. ‘‘I” represents 10% of input of 35S-methionine labeled in vitro translated Flt3. G contains GST control, D is GST-GR-DBD, N represents GST-GR-N-terminal (including the DBD), C corresponds to GST-GR-LBD and F is GST-full-length GR.
572
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
Relative luciferace activity (%)
200
150
100
50
0
pCMV4-hGR (2,5ng) p19-tk-luc (100ng) Dex (100nM) pcDNA-hFlt3 (2,5ng) pcDNA-hFlt3 (10ng) FL (10ng/ml) FL (20ng/ml)
A
+ + -
B
+ + + -
C
D
E
F
G
H
+ + + +
+ + +
+ + + +
+ + +
+ + +
-
+ + + +
-
-
+
-
+
+ +
+
-
-
-
+
-
+
-
-
Fig. 3. FL and Flt3 potentiate the Dex effect in a glucocorticoid dependent reporter-gene assay. All data are shown as normalized percentages of B which was set to 100%, corresponding to the relative luciferase activity generated by the reporter gene when 100 nM Dex was added. The median inside each box as well as the whiskers of the same box illustrate the triplicate data points of one representative experiment out of three. The plasmid pCMV4-hGR corresponds to human GR, p19-tk-luc to the GRE-containing luciferase reporter plasmid and pcDNA-Flt3 to human Flt3. A contains 0.005% ethanol as vehicle control. H shows an almost doubled glucocorticoid dependent reporter-gene activity compared with B, when 10 ng/ml Flt3 and 20 ng/ml FL was added.
tional effect on the glucocorticoid-driven reporter-gene activity (Fig. 3C and D). However, the combination of Flt3 and FL potentiated glucocorticoid-driven reportergene activity (Fig. 3G and H) where an almost doubled luciferase activity (Fig. 3H) compared with control (Fig. 3B) was observed. This effect was maintained but not enhanced with increasing amounts of transfected pcDNA-Flt3, upon the addition of FL (data not shown). Previously a functional interrelationship has been shown between other types of class III RTK and glucocorticoids, for example c-KIT and glucocorticoids in erythropoiesis [21,22] and the glucocorticoid dependent up-regulation of c-fms [23] and PDGFR-a [24]. Negative effects by glucocorticoids on cell growth have been shown to be mediated by cell cycle arrest in the G1-phase and even by apoptosis in certain cells [25]. Dex has also been reported to down-regulate the expression of the Flt3 receptor, inhibiting FL induced proliferation in several cell lines [26]. In contrast to this, there are reports about glucocorticoids being positive regulators of survival genes in breast epithelial cell lines [27] and for being one of the factors necessary for the selfrenewal of erythroid progenitors [21]. However, it should be noted that the potentiation of glucocorticoid activity exerted by Flt3 and FL may be a liver-specific phenomenon and further investigations in other cell types are needed to determine the possible general applicability of our results.
FL and Dex effects on the distribution of endogenous Flt3 and GR in rat hepatoma cells Since HTC cells, previously has been successfully used to study GR interactions with other proteins [14] we used this cell line to prepared cytosolic and nuclear extracts. The extracts were subjected to SDS–PAGE and Western blotting by staining with an Flt3 antibody. As shown in Fig. 4A, we found that Flt3 was present in both the cytosolic and in the nuclear extracts. We then went on to study short-term effects of Dex and FL on Flt3 distribution. The cells were incubated for 2, 15, 30 or 60 min, respectively, as indicated in Fig. 4A, with either 106 M Dex or 10 ng/ml soluble FL or a combination of both these ligands. Neither addition of Dex, FL or a combination of these ligands had any effects on the cellular distribution of Flt3 regardless of the time points studied. GR translocated to the nucleus as expected after Dex treatment whereas FL alone did not have any effect on GR localization. The addition of FL and Dex did not differ from Dex alone with regard to GR localization (Fig. 4B). Hence, the glucocorticoid-driven reporter-gene activity potentiated by FL (Fig. 3H) was not caused by enhanced translocation of GR in the presence of FL. Interestingly, we found Flt3 localized both in cytoplasmic and nuclear extracts from HTC cells (Fig. 4A). Since Flt3 is reported to be a cell membrane pro-
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
2 min
Cyt. -
15 min
d f d/f d
f d/f
30 min d f
-
60 min
d/f d f d/ f
Flt3 GADPH 2 min Nuc.
-
d f d/f
-
d
15 min d f d/f
30 min - d
60 min
f d/f d f d/f
Flt3
573
Acknowledgments We gratefully wish to acknowledge Dr. Margaret Warner for the Edman sequencing of Flt3 and Marika Ro¨nnholm for skillful technical assistance. This work was supported by grants from the Swedish Medical Research Council, Grants 12557 and KI 13X-2819 and by grants from the Karolinska Institutet, the Magnus Bergvall Foun˚ ke Wiberg Foundation. A.C.W. was a redation and the A cipient of a clinical research fellowship from the Novo Nordisk Foundation.
LaminA 2 min
Cyt.
f
15 min
d/f d f d/ f
30 min -
d f d/f
60 min d
GR
GADPH 2 min
Nuc. -
d
f d/f d
30 min
15 min f d/ f
-
d f d/f
60 min d
References
f d/f
f d/f
GR
LaminA Fig. 4. GR and Flt3 distribution in rat hepatoma cells after treatment with dexamethasone and FL. HTC cells were treated with either 106 M dexamethasone (d), 10 ng/ml FL (f) or a combination of those concentrations of dexamethasone and FL (d/f) for 2, 15, 30 or 60 min, as indicated. Cytosol and nuclear extracts were resolved on a 10% SDS– PAGE and transferred to a PVDF membrane. Antibodies for Flt3 (A) and GR (B) were used to analyze the cellular distribution of GR and Flt3 on the same PVDF membrane cut into two parts. Extracts from untreated cells () were harvested at 0 min, showing the GR and Flt3 levels of untreated cytosol and nuclear extracts. GADPH and Lamin A were used as loading controls for the cytosolic and nuclear extracts, respectively. The figure shows one out of three representative experiments.
tein [18] our results may reflect the presence of internalized Flt3 as indicated by the demonstrated internalization of FL [5]. Internalization occurs in the normal life cycle of the RTKs and takes place both upon ligand binding and also for the non-liganded RTKs, as addressed in a review mainly concerning the epidermal growth factor receptor (EGFR) [19]. The protein–protein interaction of GR and Flt3 is an example of how cross-talk between glucocorticoid/GR signaling and a hematopoietic growth factor may take place. We show that the physical interaction also allows for a functional interplay, in terms of FL effects on glucocorticoid signaling that however is not connected to GR translocation. The interaction between GR and Flt3 may also imply glucocorticoid effects on FL-mediated signaling. This opens up new venues in GR as well as Flt3 research, as Flt3 is an important regulator of in vivo and in vitro differentiation of specific hematopoietic stem cells and when Flt3 signaling is perturbed in specific hematological malignancies such as AML.
[1] R. Newton, Molecular mechanisms of glucocorticoid action: what is important? Thorax 55 (2000) 603–613. [2] T.J. Cole, J.A. Blendy, A.P. Monaghan, K. Krieglstein, W. Schmid, A. Aguzzi, G. Fantuzzi, E. Hummler, K. Unsicker, G. Schutz, Targeted disruption of the glucocorticoid receptor gene blocks adrenergic chromaffin cell development and severely retards lung maturation, Genes Dev. 9 (1995) 1608–1621. [3] K. De Bosscher, W. Vanden Berghe, G. Haegeman, The interplay between the glucocorticoid receptor and nuclear factor-kappaB or activator protein-1: molecular mechanisms for gene repression, Endocr. Rev. 24 (2003) 488–522. [4] O. Rosnet, M.G. Mattei, S. Marchetto, D. Birnbaum, Isolation and chromosomal localization of a novel FMS-like tyrosine kinase gene, Genomics 9 (1991) 380–385. [5] A.M. Turner, N.L. Lin, S. Issarachai, S.D. Lyman, V.C. Broudy, FLT3 receptor expression on the surface of normal and malignant human hematopoietic cells, Blood 88 (1996) 3383–3390. [6] S. Fichelson, The FLT3/FLK2 ligand: structure, functions and prospects, Eur. Cytokine Netw. 9 (1998) 7–22. [7] A. Weiss, J. Schlessinger, Switching signals on or off by receptor dimerization, Cell 94 (1998) 277–280. [8] D.G. Gilliland, J.D. Griffin, Role of FLT3 in leukemia, Curr. Opin. Hematol. 9 (2002) 274–281. ˚ . Gustafsson, A.C. Wikstro¨m, Glucocor[9] C. Wide´n, J. Zilliacus, J.-A ticoid receptor interaction with 14-3-3 and Raf-1, a proposed mechanism for cross-talk of two signal transduction pathways, J. Biol. Chem. 275 (2000) 39296–39301. ˚. [10] E. Hedman, C. Wide´n, A. Asadi, I. Dinnetz, W.P. Schro¨der, J.-A Gustafsson, A.C. Wikstro¨m, Proteomic identification of glucocorticoid receptor interacting proteins, Proteomics 6 (2006) 3114– 3126. ˚ . Gustafs[11] S. Okret, A.C. Wikstro¨m, O. Wrange, B. Andersson, J.-A son, Monoclonal antibodies against the rat liver glucocorticoid receptor, Proc. Natl. Acad. Sci. USA 81 (1984) 1609–1613. ˚ . Gustafsson, Role of acidic and [12] T. Almlo¨f, A.P. Wright, J.-A phosphorylated residues in gene activation by the glucocorticoid receptor, J. Biol. Chem. 270 (1995) 17535–17540. [13] T.N. Nguyen, J.A. Goodrich, Protein–protein interaction assays: eliminating false positive interactions, Nat. Methods 3 (2006) 135– 139. ˚ . Gustafsson, A.C. Wikstro¨m, Cytosolic glucocorticoid [14] C. Wide´n, J.-A receptor interaction with nuclear factor-kappa B proteins in rat liver cells, Biochem. J. 373 (2003) 211–220. [15] K.C. Kanelakis, W.B. Pratt, Regulation of glucocorticoid receptor ligand-binding activity by the hsp90/hsp70-based chaperone machinery, Methods Enzymol. 364 (2003) 159–173. [16] Q. Yao, R. Nishiuchi, Q. Li, A.R. Kumar, W.A. Hudson, J.H. Kersey, FLT3 expressing leukemias are selectively sensitive to inhibitors of the molecular chaperone heat shock protein 90 through destabilization of signal transduction-associated kinases, Clin. Cancer Res. 9 (2003) 4483–4493.
574
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
[17] K.J. Howard, S.J. Holley, K.R. Yamamoto, C.W. Distelhorst, Mapping the HSP90 binding region of the glucocorticoid receptor, J. Biol. Chem. 265 (1990) 11928–11935. [18] S.D. Lyman, L. James, J. Zappone, P.R. Sleath, M.P. Beckmann, T. Bird, Characterization of the protein encoded by the flt3 (flk2) receptor-like tyrosine kinase gene, Oncogene 8 (1993) 815–822. [19] D. Teis, L.A. Huber, The odd couple: signal transduction and endocytosis, Cell Mol. Life Sci. 60 (2003) 2020–2033. [20] H. Tazawa, W. Osman, Y. Shoji, E. Treuter, J.A. Gustafsson, J. Zilliacus, Regulation of subnuclear localization is associated with a mechanism for nuclear receptor corepression by RIP140, Mol. Cell. Biol. 23 (2003) 4187–4198. [21] O. Wessely, E.M. Deiner, H. Beug, M. von Lindern, The glucocorticoid receptor is a key regulator of the decision between self-renewal and differentiation in erythroid progenitors, EMBO J. 16 (1997) 267–280. [22] M. von Lindern, W. Zauner, G. Mellitzer, P. Steinlein, G. Fritsch, K. Huber, B. Lowenberg, H. Beug, The glucocorticoid receptor cooperates with the erythropoietin receptor and c-Kit to enhance and sustain proliferation of erythroid progenitors in vitro, Blood 94 (1999) 550–559.
[23] B.M. Kacinski, K.A. Scata, D. Carter, L.D. Yee, E. Sapi, B.L. King, S.K. Chambers, M.A. Jones, M.H. Pirro, E.R. Stanley, et al., FMS (CSF-1 receptor) and CSF-1 transcripts and protein are expressed by human breast carcinomas in vivo and in vitro, Oncogene 6 (1991) 941–952. [24] G.S. Warshamana, S. Martinez, J.A. Lasky, M. Corti, A.R. Brody, Dexamethasone activates expression of the PDGF-alpha receptor and induces lung fibroblast proliferation, Am. J. Physiol. 274 (1998) L499–L507. [25] P. Grandics, A. Miller, T.J. Schmidt, D. Mittman, G. Litwack, Purification of the unactivated glucocorticoid receptor and its subsequent in vitro activation, J. Biol. Chem. 259 (1984) 3173–3180. [26] Z.X. Xu, Y. Xu, J.K. Zhu, C.X. Li, Y. Li, X.G. Zhang, Effects of dexamethasone on Flt3 receptor expression and proliferation induced by recombinant human Flt3 ligand in malignant hematopoietic cells, Acta Pharmacol. Sin. 22 (2001) 534–540. [27] W. Wu, S. Chaudhuri, D.R. Brickley, D. Pang, T. Karrison, S.D. Conzen, Microarray analysis reveals glucocorticoid-regulated survival genes that are associated with inhibition of apoptosis in breast epithelial cells, Cancer Res. 64 (2004) 1757–1764.
Biochemical and Biophysical Research Communications 368 (2008) 569–574 www.elsevier.com/locate/ybbrc
FMS-like tyrosine kinase 3 interacts with the glucocorticoid receptor complex and affects glucocorticoid dependent signaling Abolfazl Asadi 1, Erik Hedman 1,*, Christina Wide´n, Johanna Zilliacus, ˚ ke Gustafsson, Ann-Charlotte Wikstro¨m Jan-A Department of Biosciences and Nutrition, Division of Medical Nutrition, Karolinska Institutet, NOVUM, S-141 86 Stockholm, Sweden Received 16 January 2008 Available online 7 February 2008
Abstract The glucocorticoid receptor (GR) forms part of a multiprotein complex consisting of chaperones and proteins active in glucocorticoid signaling and other pathways. By immunoaffinity purification of GR, followed by Edman sequencing and Western blotting, we identified the FMS-like tyrosine kinase 3 (Flt3) as a GR-interacting protein in rat liver and hepatoma cells. Flt3 interacts with both non-liganded and liganded GR. The DNA-binding domain of GR is sufficient for Flt3 interaction as shown by GST-pull down experiments. Studies of the effects of Flt3 and its ligand FL in glucocorticoid-driven reporter-gene assays in Cos7 cells, show that co-transfection with Flt3 and FL potentiates glucocorticoid effects. Treatment with FL had no effect on GR location and Dex induced translocation of GR was unaffected by FL. In summary, GR and Flt3 interact, affecting GR signaling. This novel cross-talk between GR and a hematopoietic growth factor might also imply glucocorticoid effects on Flt3-mediated signaling. Ó 2008 Elsevier Inc. All rights reserved. Keywords: Glucocorticoid receptor; Glucocorticoids; Flt3; GST-pull down; Immunoprecipitation
The glucocorticoid receptor (GR) belongs to a large family of nuclear receptors and acts as a ligand-dependent transcriptional regulator [1]. A number of natural and synthetic glucocorticoids can activate GR and thereby affect survival, proliferation and differentiation of cells. These effects are vital, as indicated by the fact that targeted disruption of the GR gene leads to neonatal death [2], mainly as a consequence of impaired lung development. The transcriptional effects of glucocorticoids are mediated via its binding to GR. GR then translocates to the nucleus where it affects gene transcription by binding to certain DNA sequences, called glucocorticoid response elements (GRE). Furthermore, GR also acts via interaction with other proteins in a number of signal transduction pathways. This mode of GR action often leads to transrepression of the other signaling pathway. Notably, this is the *
1
Corresponding author. Fax: +46 8 711 66 59. E-mail address: [email protected] (E. Hedman). The authors contributed equally to this work.
0006-291X/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2008.01.146
mode of action for the NF-jB and the AP-1 signaling pathways [3]. Clinically, glucocorticoids are frequently used as treatment of inflammatory diseases, for immunosuppression after transplantation and to treat certain malignant diseases as acute childhood leukemias. The FMS-like tyrosine kinase 3 (Flt3) belongs to the class III receptor tyrosine kinase (RTK) family [4] in which also the c-KIT, FMS and the PDGF are found. In humans Flt3 is expressed mainly in hematopoietic cells [5] and has been shown to play a role in proliferation and survival of hematopoietic progenitor cells, as well as in differentiation of early B, NK, and dendritic cell progenitors [6]. Binding of the Flt3 ligand FL, which is present both as a membrane bound and as a soluble glycoprotein, to the extra-cellular Flt3-domain, leads to conformational changes that induce and stabilize receptor dimerization [7]. Receptor dimerization brings the kinase domains together and this induces tyrosine auto-phosphorylation and further kinase activity. Mutations in Flt3 leading to constitutive kinase activity have been shown to be present
570
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
in acute myeloid leukemia (AML) as well as in some other hematological malignancies, especially during blast crisis [8], and has been correlated to a poor prognosis in AML patients. In our attempts to understand GR signaling based on protein–protein interactions we have developed an immunoaffinity based purification protocol for GR which allows co-purification of GR-interacting proteins [9]. In this study we report that Flt3 is part of the GR complex and that this has functional consequences for GR activity. Materials and methods SDS–PAGE, Western blotting, and immunostaining. The proteins samples were boiled in SDS sample buffer and resolved on 10% (v/v) SDS– PAGE gels under reducing conditions at 30 mA/gel for 50 min with the mini-Protean III gel system (Bio-Rad, Stockholm, Sweden). Western blotting and immunostaining proceeded as in [10] with the following primary antibodies: anti-GR mAb250 (also called mAb 7) [11] 0.013 lg/ ml, anti-Flt3/Flk2 1:500 (Santa Cruz Biotechnology Inc., Santa Cruz, CA), anti-Lamin A/C (Cellsignaling Technology, MA, USA) 1:1000 and anti-GADPH (Abcam, Cambridge, UK) 1:5000, anti-DBD 3.25 lg/ml, anti-GST 1:1000. Horseradish peroxidase (HRP)-conjugated, -sheep antimouse or -donkey anti-rabbit 1:50,000 (both from GE Healthcare, Uppsala, Sweden) were used as secondary antibodies. The membranes were finally incubated either with SuperSignalÒ (Pierce, Rockford, IL, USA) or with ECLTM Advanced Western Blotting Detection Kit (GE Healthcare, Uppsala, Sweden) according to the instructions from the manufacturer. Co-immunoprecipitation of GR from rat liver cytosol using antibodies directed against Flt3. Liganded/activated- and non-liganded/non-activated GR complexes were prepared as previously described in [9]. Approval by the Ethical Committee was obtained as required. Protein (1 mg) from cytosolic extracts were diluted to a total volume of 300 ll in 10 mM Tris, pH 7.8, 0.1% (v/v) Nonidet P-40 (IGEPAL), 0.15 M NaCl, 1 mM EDTA, 0.5 mM phenylmethylsulfonyl fluoride and 0.5 lg/ml of each aprotinin, leupeptin, pepstatin, and 1 lg Flt3/Flk2 antibody (Santa Cruz Biotechnology Inc., Santa Cruz, CA) was added. After 1 h, 20 ll of resuspended volume of Protein A/G–Agarose (Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA) was added and after another 3 h, the pellet was washed three times in PBS followed by SDS–PAGE and Western blotting. Two micrograms of agarose coupled with normal rabbit antibodies were used as a negative control. All steps were carried out at +4 °C. Plasmids. The plasmid pcDNA-hFlt3 containing human Flt3 cDNA was constructed by PCR-recloning of a full-length human Flt3 coding fragment, from the pcDSRa plasmid (a kind gift from Dr. Oliver Rosnet) into the pcDNA3.1 D/V5-HIS-TOPO expression vector (Invitrogen, Carlsbad, CA) using 50 -CACCATGGGCCCAGGAGTTCTGCTGCT-30 as a forward primer and 50 -CGAATCTTCGACCTGAGCCT-30 as a reverse primer. GST-GR fusion proteins were expressed from pGEX-4T-1 plasmids (GE Healthcare, Uppsala, Sweden). The plasmids expressing GST-GR-N-terminal (N+DBD+hinge) domain (aa 1–567), GST-GRDBD (aa 418–503), GST-GR-C-terminal (aa 568–777) domain and GSTGR-full-length (aa 1–777) were constructed by subcloning of BamHI/XhoI fragments corresponding to different domains of GR from the vector pcDNA3.1 D/V5-His-TOPO containing each fragment. The expression plasmid pCMV4-hGR and the luciferase reporter plasmid p19-tk-luc have been described previously [12]. Protein expression and GST-pull down. GST, GST-GR-N-terminal and GST-GR-full-length were expressed in the Escherichia coli strain BL21 (DE3) pLysS whereas GST-GR-DBD and GST-GR-C-terminal were expressed in the E. coli strain BL21 Competent cells. Bacteria were grown to an OD600 of 0.5, and the expression of fusion proteins was accomplished by following the instructions from the manufacturer. GST-pull down was performed essentially as in [13].
Cell culture, transient transfections and reporter-gene assay. Cos7 cells were maintained in Dulbecco’s modified Eagle’s medium containing sodium pyruvate and 1 g/l glucose supplemented with 10% (v/v) fetalbovine serum (FBS), 100 U/ml penicillin and 100 lg/ml streptomycin. Rat hepatoma cells H4-II-E-C3 (HTC cells) (ATCC, Manassas, VA), were grown as described in [14]. Cells used for reporter-gene assays were seeded out on 24-well plates at a density of 2 104 cells/well, one day prior to transfection. Transient transfections of cells were performed with FuGENETM-6 transfection reagent (Roche Diagnostics Scandinavia AB, Stockholm, Sweden) at a ratio of 3 ll/lg DNA, according to the manufacturer’s protocol. The total amount of plasmid DNA used was normalized, by the addition of the empty expression plasmid pcDNA3.1 D/ V5-His-TOPO, to a total amount of 112.5 ng plasmid/well. After 18 h incubation, the transfection mixture was replaced with fresh medium, containing 100 nM Dex in 0.005% ethanol or 0.005% ethanol only. 24 h later cells were harvested for enzyme assay. Luciferase assay was performed with the Luc-ScreenÒ System (Applied Biosystems, Stockholm, Sweden) according to the instructions from the manufacturer in a 1450 MicroBeta instrument (Wallac Sverige AB, Stockholm, Sweden). GR and Flt3 distribution in rat hepatoma cells after treatment with dexamethasone and FL. The day of the experiment the HTC cells were washed three times in PBS and then cultivated for an initial 6 h in RPMI media as stated above but in the absence of FBS. Then the cells were exposed to: only media as a control, 106 M Dex or 10 ng/ml FL in various combinations as indicated in Fig. 4. The cells were further cultivated for 2, 15, 30 or 60 min (Fig. 4A and B). Cytosolic and nuclear extracts were subsequently prepared according to [14]. Protein concentrations of the extracts were determined by a Bradford assay and equal protein concentrations for cytosol and nuclear extracts, respectively, were subjected to SDS–PAGE. GADPH and Lamin A were used as loading controls for cytosolic and nuclear extract, respectively. GR immunoprecipitation in rat hepatoma cell cytosol. HTC cells were grown and cytosolic extracts was prepared as described above. Equal protein concentrations, for each individual experiment, were immunoprecipitated according to [14] with the following modifications. The cytosol fractions were preabsorbed with normal mouse IgG coupled to CNBr-activated Sepharose 4B and immunoprecipitated using the mAb250, followed by SDS–PAGE and Western blotting. All steps were carried out at +4 °C.
Results and discussion The immunoaffinity-purified rat liver GR complex contains Flt3 We immunopurified non-liganded/non-activated GR receptor complex from rat liver cytosol. The GR complex was further analyzed by SDS–PAGE and Edman sequencing. One of the few proteins that were not N-terminally blocked contained the amino acid sequence: KWEFPRE NLAEFGKVLGSGAFG that corresponds to aa 602–622 in the human isoform of Flt3, as based upon the protein sequence derived from the NCBI nucleotide database (NCBI Accession No. XM_221874). The GR-Flt3 interaction was also confirmed by a reverse immunoprecipitation, using a Flt3 antibody. Fig. 1A demonstrates that immunoprecipitated cytosolic Flt3, interacted with part of the GR complex. As GR endogenously interacts with hsp90 it is difficult to conclude whether the interaction between GR and Flt3 is direct or not. Especially since hsp90 has been reported to form a protein complex including Flt3 and p23 [16]. However, since our immunoprecipitation results indicated that the
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
Ip
+
S
GR
kDa GR
Flt3
200 116 97
kDa GR 200 116 97
interaction between the two different states of GR ligandation as seen for many other GR-interacting proteins [10].
Flt3
Human Flt3 interacts with the DNA-binding domain of human GR
66
66
Fig. 1. GR co-immunoprecipitates with Flt3 in rat liver cytosol and Flt3 associates with both non-liganded/non-activated and liganded/activated GR from rat liver cytosol. (A) One out of two experiments. Rat liver cytosol containing non-liganded/non-activated GR was prepared from adrenalectomized rats. Immunoprecipitation was performed using 1 lg of Flt3 antibody. The immunoprecipitate (Ip) was resolved on a 10% SDS– polyacrylamide gel and transferred to a PVDF membrane. Immunostaining showed that GR co-immunoprecipitated with Flt3. The positive control (+) contains 50 lg total protein and the total input for the immunoprecipitation (Ip) and the supernatant (S) was 1 mg of protein. The supernatant (S) shows the fraction of GR that has not coimmunoprecipitated with the Flt3 antibody. (B, C) One representative experiment out of four. Cytosol from four livers from adrenalectomized rats was and divided into two equal parts. To one part, sodium molybdate was added, to obtain non-liganded/non-activated GR complexes (B), and to the other part ligand was added and heat-activation was performed to obtain liganded/activated GR complexes (C) before purification by immunoaffinity chromatography was performed. GR and Flt3 antibodies were used, as indicated in the figure, to detect proteins in the immunoprecipitate by Western blotting.
interaction between GR and Flt3 occurs both in complexes where hsp90 is present, i.e. the non-liganded GR complex (Fig. 1B), as well as in complexes were hsp90 is absent, i.e. liganded GR complex (Fig. 1C) a direct interaction seems possible. Previous studies have demonstrated that NF-jB [14], Raf-1 and 14-3-3 [9] preferentially interacts with liganded GR, whereas many chaperone proteins, such as hsp90 and others [15] mainly interact with non-liganded GR. We therefore investigated if the addition of glucocorticoids affected the GR-Flt3 interaction. By using Flt3 antibody to immunodetect Flt3 in liganded/activated GR complexes we found that Flt3 also co-purified with liganded/activated GR (Fig. 1C). This finding does not exclude a variable
kDa G D 150 100 50 37 25
N
To determine which domains of GR were necessary for the interaction between GR and Flt3, we constructed a set of human GST-GR fusion proteins that were incubated with in vitro transcribed/translated Flt3. The autoradiograph of the GST-pull down experiment (Fig. 2C) showed that the C-terminal, ligand-binding domain (LBD) (C) of GR did not interact with Flt3 above the level of the background with non-fused GST (cf. G). However, the DNAbinding domain (DBD) (D), the N-terminal fragment (N), as well as full-length GR (F) could interact with Flt3. Considering that the hsp90 interaction with GR is attributed to the GR-ligand binding domain [17] and that Flt3 interacts with the DNA-binding domain (lacking LBD) those results also speaks in favor of a direct interaction between GR and Flt3. Glucocorticoid dependent reporter-gene activity is affected by Flt3 and FL Cos7 cells have previously been shown to lack endogenous GR [20]. By Western blotting we could demonstrate that Cos7 cells also lacked endogenous Flt3 (data not shown). To study the effects of Flt3 and FL on glucocorticoid signaling, we transfected Cos7 cells with the plasmids pCMV4-hGR, pc-DNA-hFlt3 and the glucocorticoid-driven reporter-gene p19-tk-luc. By addition of FL and Dex in various combinations we studied how glucocorticoid dependent reporter-gene activity was affected. Addition of Flt3 and FL in the absence of Dex did not activate the GR dependent reporter gene (data not shown). However, addition of only Dex induced the reporter gene as expected (Fig. 3B). Neither Flt3 alone nor FL alone had any addiGST-pull down
anti-DBD
anti-GST C F
kDa G D 150 100 50 37
571
N
C F
kDa 150
I
G
D
N
C
F Flt3
100
25
GST Fig. 2. Flt3 interacts with the DNA-binding domain of GR. The figure shows one representative experiment out of two. (A) GST and GST-GR fusion proteins expressed in vitro in Escherichia coli were resolved on 10% SDS–PAGE gels and transferred to PVDF membranes. The membranes were stained with a GST antibody (A); GST-GR fusion proteins are encircled. The same amount of total protein for each protein extract was used in (A–C). In (B) the DBD antibody was used to localize the 37 kDa GST-DBD fusion protein, D. Full-length human Flt3 was in vitro transcribed/translated in rabbit reticulocyte lysates in the presence of 35S-methionine. (C) An autoradiograph of various GST-GR glutathione agarose precipitates in the presence of 35Smethionine labeled Flt3. ‘‘I” represents 10% of input of 35S-methionine labeled in vitro translated Flt3. G contains GST control, D is GST-GR-DBD, N represents GST-GR-N-terminal (including the DBD), C corresponds to GST-GR-LBD and F is GST-full-length GR.
572
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
Relative luciferace activity (%)
200
150
100
50
0
pCMV4-hGR (2,5ng) p19-tk-luc (100ng) Dex (100nM) pcDNA-hFlt3 (2,5ng) pcDNA-hFlt3 (10ng) FL (10ng/ml) FL (20ng/ml)
A
+ + -
B
+ + + -
C
D
E
F
G
H
+ + + +
+ + +
+ + + +
+ + +
+ + +
-
+ + + +
-
-
+
-
+
+ +
+
-
-
-
+
-
+
-
-
Fig. 3. FL and Flt3 potentiate the Dex effect in a glucocorticoid dependent reporter-gene assay. All data are shown as normalized percentages of B which was set to 100%, corresponding to the relative luciferase activity generated by the reporter gene when 100 nM Dex was added. The median inside each box as well as the whiskers of the same box illustrate the triplicate data points of one representative experiment out of three. The plasmid pCMV4-hGR corresponds to human GR, p19-tk-luc to the GRE-containing luciferase reporter plasmid and pcDNA-Flt3 to human Flt3. A contains 0.005% ethanol as vehicle control. H shows an almost doubled glucocorticoid dependent reporter-gene activity compared with B, when 10 ng/ml Flt3 and 20 ng/ml FL was added.
tional effect on the glucocorticoid-driven reporter-gene activity (Fig. 3C and D). However, the combination of Flt3 and FL potentiated glucocorticoid-driven reportergene activity (Fig. 3G and H) where an almost doubled luciferase activity (Fig. 3H) compared with control (Fig. 3B) was observed. This effect was maintained but not enhanced with increasing amounts of transfected pcDNA-Flt3, upon the addition of FL (data not shown). Previously a functional interrelationship has been shown between other types of class III RTK and glucocorticoids, for example c-KIT and glucocorticoids in erythropoiesis [21,22] and the glucocorticoid dependent up-regulation of c-fms [23] and PDGFR-a [24]. Negative effects by glucocorticoids on cell growth have been shown to be mediated by cell cycle arrest in the G1-phase and even by apoptosis in certain cells [25]. Dex has also been reported to down-regulate the expression of the Flt3 receptor, inhibiting FL induced proliferation in several cell lines [26]. In contrast to this, there are reports about glucocorticoids being positive regulators of survival genes in breast epithelial cell lines [27] and for being one of the factors necessary for the selfrenewal of erythroid progenitors [21]. However, it should be noted that the potentiation of glucocorticoid activity exerted by Flt3 and FL may be a liver-specific phenomenon and further investigations in other cell types are needed to determine the possible general applicability of our results.
FL and Dex effects on the distribution of endogenous Flt3 and GR in rat hepatoma cells Since HTC cells, previously has been successfully used to study GR interactions with other proteins [14] we used this cell line to prepared cytosolic and nuclear extracts. The extracts were subjected to SDS–PAGE and Western blotting by staining with an Flt3 antibody. As shown in Fig. 4A, we found that Flt3 was present in both the cytosolic and in the nuclear extracts. We then went on to study short-term effects of Dex and FL on Flt3 distribution. The cells were incubated for 2, 15, 30 or 60 min, respectively, as indicated in Fig. 4A, with either 106 M Dex or 10 ng/ml soluble FL or a combination of both these ligands. Neither addition of Dex, FL or a combination of these ligands had any effects on the cellular distribution of Flt3 regardless of the time points studied. GR translocated to the nucleus as expected after Dex treatment whereas FL alone did not have any effect on GR localization. The addition of FL and Dex did not differ from Dex alone with regard to GR localization (Fig. 4B). Hence, the glucocorticoid-driven reporter-gene activity potentiated by FL (Fig. 3H) was not caused by enhanced translocation of GR in the presence of FL. Interestingly, we found Flt3 localized both in cytoplasmic and nuclear extracts from HTC cells (Fig. 4A). Since Flt3 is reported to be a cell membrane pro-
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
2 min
Cyt. -
15 min
d f d/f d
f d/f
30 min d f
-
60 min
d/f d f d/ f
Flt3 GADPH 2 min Nuc.
-
d f d/f
-
d
15 min d f d/f
30 min - d
60 min
f d/f d f d/f
Flt3
573
Acknowledgments We gratefully wish to acknowledge Dr. Margaret Warner for the Edman sequencing of Flt3 and Marika Ro¨nnholm for skillful technical assistance. This work was supported by grants from the Swedish Medical Research Council, Grants 12557 and KI 13X-2819 and by grants from the Karolinska Institutet, the Magnus Bergvall Foun˚ ke Wiberg Foundation. A.C.W. was a redation and the A cipient of a clinical research fellowship from the Novo Nordisk Foundation.
LaminA 2 min
Cyt.
f
15 min
d/f d f d/ f
30 min -
d f d/f
60 min d
GR
GADPH 2 min
Nuc. -
d
f d/f d
30 min
15 min f d/ f
-
d f d/f
60 min d
References
f d/f
f d/f
GR
LaminA Fig. 4. GR and Flt3 distribution in rat hepatoma cells after treatment with dexamethasone and FL. HTC cells were treated with either 106 M dexamethasone (d), 10 ng/ml FL (f) or a combination of those concentrations of dexamethasone and FL (d/f) for 2, 15, 30 or 60 min, as indicated. Cytosol and nuclear extracts were resolved on a 10% SDS– PAGE and transferred to a PVDF membrane. Antibodies for Flt3 (A) and GR (B) were used to analyze the cellular distribution of GR and Flt3 on the same PVDF membrane cut into two parts. Extracts from untreated cells () were harvested at 0 min, showing the GR and Flt3 levels of untreated cytosol and nuclear extracts. GADPH and Lamin A were used as loading controls for the cytosolic and nuclear extracts, respectively. The figure shows one out of three representative experiments.
tein [18] our results may reflect the presence of internalized Flt3 as indicated by the demonstrated internalization of FL [5]. Internalization occurs in the normal life cycle of the RTKs and takes place both upon ligand binding and also for the non-liganded RTKs, as addressed in a review mainly concerning the epidermal growth factor receptor (EGFR) [19]. The protein–protein interaction of GR and Flt3 is an example of how cross-talk between glucocorticoid/GR signaling and a hematopoietic growth factor may take place. We show that the physical interaction also allows for a functional interplay, in terms of FL effects on glucocorticoid signaling that however is not connected to GR translocation. The interaction between GR and Flt3 may also imply glucocorticoid effects on FL-mediated signaling. This opens up new venues in GR as well as Flt3 research, as Flt3 is an important regulator of in vivo and in vitro differentiation of specific hematopoietic stem cells and when Flt3 signaling is perturbed in specific hematological malignancies such as AML.
[1] R. Newton, Molecular mechanisms of glucocorticoid action: what is important? Thorax 55 (2000) 603–613. [2] T.J. Cole, J.A. Blendy, A.P. Monaghan, K. Krieglstein, W. Schmid, A. Aguzzi, G. Fantuzzi, E. Hummler, K. Unsicker, G. Schutz, Targeted disruption of the glucocorticoid receptor gene blocks adrenergic chromaffin cell development and severely retards lung maturation, Genes Dev. 9 (1995) 1608–1621. [3] K. De Bosscher, W. Vanden Berghe, G. Haegeman, The interplay between the glucocorticoid receptor and nuclear factor-kappaB or activator protein-1: molecular mechanisms for gene repression, Endocr. Rev. 24 (2003) 488–522. [4] O. Rosnet, M.G. Mattei, S. Marchetto, D. Birnbaum, Isolation and chromosomal localization of a novel FMS-like tyrosine kinase gene, Genomics 9 (1991) 380–385. [5] A.M. Turner, N.L. Lin, S. Issarachai, S.D. Lyman, V.C. Broudy, FLT3 receptor expression on the surface of normal and malignant human hematopoietic cells, Blood 88 (1996) 3383–3390. [6] S. Fichelson, The FLT3/FLK2 ligand: structure, functions and prospects, Eur. Cytokine Netw. 9 (1998) 7–22. [7] A. Weiss, J. Schlessinger, Switching signals on or off by receptor dimerization, Cell 94 (1998) 277–280. [8] D.G. Gilliland, J.D. Griffin, Role of FLT3 in leukemia, Curr. Opin. Hematol. 9 (2002) 274–281. ˚ . Gustafsson, A.C. Wikstro¨m, Glucocor[9] C. Wide´n, J. Zilliacus, J.-A ticoid receptor interaction with 14-3-3 and Raf-1, a proposed mechanism for cross-talk of two signal transduction pathways, J. Biol. Chem. 275 (2000) 39296–39301. ˚. [10] E. Hedman, C. Wide´n, A. Asadi, I. Dinnetz, W.P. Schro¨der, J.-A Gustafsson, A.C. Wikstro¨m, Proteomic identification of glucocorticoid receptor interacting proteins, Proteomics 6 (2006) 3114– 3126. ˚ . Gustafs[11] S. Okret, A.C. Wikstro¨m, O. Wrange, B. Andersson, J.-A son, Monoclonal antibodies against the rat liver glucocorticoid receptor, Proc. Natl. Acad. Sci. USA 81 (1984) 1609–1613. ˚ . Gustafsson, Role of acidic and [12] T. Almlo¨f, A.P. Wright, J.-A phosphorylated residues in gene activation by the glucocorticoid receptor, J. Biol. Chem. 270 (1995) 17535–17540. [13] T.N. Nguyen, J.A. Goodrich, Protein–protein interaction assays: eliminating false positive interactions, Nat. Methods 3 (2006) 135– 139. ˚ . Gustafsson, A.C. Wikstro¨m, Cytosolic glucocorticoid [14] C. Wide´n, J.-A receptor interaction with nuclear factor-kappa B proteins in rat liver cells, Biochem. J. 373 (2003) 211–220. [15] K.C. Kanelakis, W.B. Pratt, Regulation of glucocorticoid receptor ligand-binding activity by the hsp90/hsp70-based chaperone machinery, Methods Enzymol. 364 (2003) 159–173. [16] Q. Yao, R. Nishiuchi, Q. Li, A.R. Kumar, W.A. Hudson, J.H. Kersey, FLT3 expressing leukemias are selectively sensitive to inhibitors of the molecular chaperone heat shock protein 90 through destabilization of signal transduction-associated kinases, Clin. Cancer Res. 9 (2003) 4483–4493.
574
A. Asadi et al. / Biochemical and Biophysical Research Communications 368 (2008) 569–574
[17] K.J. Howard, S.J. Holley, K.R. Yamamoto, C.W. Distelhorst, Mapping the HSP90 binding region of the glucocorticoid receptor, J. Biol. Chem. 265 (1990) 11928–11935. [18] S.D. Lyman, L. James, J. Zappone, P.R. Sleath, M.P. Beckmann, T. Bird, Characterization of the protein encoded by the flt3 (flk2) receptor-like tyrosine kinase gene, Oncogene 8 (1993) 815–822. [19] D. Teis, L.A. Huber, The odd couple: signal transduction and endocytosis, Cell Mol. Life Sci. 60 (2003) 2020–2033. [20] H. Tazawa, W. Osman, Y. Shoji, E. Treuter, J.A. Gustafsson, J. Zilliacus, Regulation of subnuclear localization is associated with a mechanism for nuclear receptor corepression by RIP140, Mol. Cell. Biol. 23 (2003) 4187–4198. [21] O. Wessely, E.M. Deiner, H. Beug, M. von Lindern, The glucocorticoid receptor is a key regulator of the decision between self-renewal and differentiation in erythroid progenitors, EMBO J. 16 (1997) 267–280. [22] M. von Lindern, W. Zauner, G. Mellitzer, P. Steinlein, G. Fritsch, K. Huber, B. Lowenberg, H. Beug, The glucocorticoid receptor cooperates with the erythropoietin receptor and c-Kit to enhance and sustain proliferation of erythroid progenitors in vitro, Blood 94 (1999) 550–559.
[23] B.M. Kacinski, K.A. Scata, D. Carter, L.D. Yee, E. Sapi, B.L. King, S.K. Chambers, M.A. Jones, M.H. Pirro, E.R. Stanley, et al., FMS (CSF-1 receptor) and CSF-1 transcripts and protein are expressed by human breast carcinomas in vivo and in vitro, Oncogene 6 (1991) 941–952. [24] G.S. Warshamana, S. Martinez, J.A. Lasky, M. Corti, A.R. Brody, Dexamethasone activates expression of the PDGF-alpha receptor and induces lung fibroblast proliferation, Am. J. Physiol. 274 (1998) L499–L507. [25] P. Grandics, A. Miller, T.J. Schmidt, D. Mittman, G. Litwack, Purification of the unactivated glucocorticoid receptor and its subsequent in vitro activation, J. Biol. Chem. 259 (1984) 3173–3180. [26] Z.X. Xu, Y. Xu, J.K. Zhu, C.X. Li, Y. Li, X.G. Zhang, Effects of dexamethasone on Flt3 receptor expression and proliferation induced by recombinant human Flt3 ligand in malignant hematopoietic cells, Acta Pharmacol. Sin. 22 (2001) 534–540. [27] W. Wu, S. Chaudhuri, D.R. Brickley, D. Pang, T. Karrison, S.D. Conzen, Microarray analysis reveals glucocorticoid-regulated survival genes that are associated with inhibition of apoptosis in breast epithelial cells, Cancer Res. 64 (2004) 1757–1764.