| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
The 2nd Department of Internal Medicine (A.S., N.T., K.A.), Tohoku University School of Medicine, Sendai 980 Japan; and Department of Pathology (N.S., R.Y.O.), Tokai University School of Medicine, Isehara-city 259-11, Japan
Address all correspondence and requests for reprints to: Dr. Akira Sugawara, The 2nd Department of Internal Medicine, Tohoku University School of Medicine, 1-1 Seiryo-cho, Aoba-ku, Sendai 980, Japan.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, mainly localized
to the proximal and the distal tubules, but not to the glomeruli. The
serial section staining using anti-VDR antibody showed the
colocalization of RXR and VDR in those tubular cells. In order to
elucidate the functional significance of endogenous receptors in the
tubular cells, we next performed transient transfection studies using
the tubular-cell derived Madin-Darby bovine kidney cells, which express
both endogenous VDR and RXR. We transfected a reporter plasmid
containing direct repeat 3 (DR3) sequence, to which only RXR/VDR
heterodimer can bind, and found that VD and 9-cis retinoic
acid, as well as VD and RXR selective agonist LG100153, had an additive
effect for the DR3 transactivation. Taken together, we speculate that
endogenous RXRs co-localize with VDR, and coregulate VD-dependent genes
in the tubular cells of the kidney as RXR/VDR heterodimer. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,
ß, 
) have been identified so far (2, 5). RXRs not only work as
homodimers (8), but also as heterodimers with other nuclear hormone
receptors including 1, 25-dihydroxyvitamin D3 (VD) receptor
(VDR), retinoic acid receptors (RARs), and thyroid hormone receptors
(1, 2, 3, 4, 5, 9). Among the genes that are regulated by VD, many of them
contain VD response element (VDRE) composed of direct repeated pairs of
motif (consensus: AGGTCA) spaced by three nucleotides called direct
repeat 3 (DR3) (1, 2, 3, 9, 10). It has recently been shown that RXR/VDR
heterodimer, rather than VDR homodimer, preferentially binds to DR3
sequence, and mediates VD-regulated transcriptional activation (9, 10).
Such that, RXRs play important roles in VD signaling. VD acts mainly on
its target tissues such as kidney, intestine, and bone for the
regulation of calcium homeostasis (11). In the kidney, the expression
of VDR in the tubular cells has been well studied (12, 13, 14). Regarding
RXRs, we have previously found the predominant expression of RXR
protein in the kidney nuclear extract using the electrophoretic
mobility shift assay (EMSA) (15). However, their localization and the
functional significance in the kidney are not fully understood.
In the present study, we first demonstrated using isoform-specific
antimouse (m) RXR antibodies we previously reported (16) that all RXR
isoforms, predominantly RXR, mainly localized to the proximal and
the distal tubules, but not to the glomeruli. The serial section
staining using anti-VDR antibody showed the colocalization of RXR
and VDR in those tubular cells. We next transfected a reporter plasmid
containing direct repeat 3 (DR3) sequence, to which only RXR/VDR
heterodimer can bind by EMSA, into the tubular-cell derived Madin-Darby
bovine kidney (MDBK) cells which express both endogenous VDR (17, 18)
and RXR. Interestingly, VD and 9cRA, as well as VD and RXR selective
agonist LG100153 (19), had an additive effect for the DR3
transactivation. We therefore speculate that the endogenous RXRs
colocalize with VDR, and coregulate VD-dependent genes in the tubular
cells of the kidney as RXR/VDR heterodimer.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, ß, were raised against synthetic peptides
containing the following mouse RXR amino acid residues:
RXR-(92-109), RXRß-(78-93), and RXR-(35-54) as previously
described (16). Immunohistochemical preabsorption tests were performed
in order to confirm the specificity by preincubating the diluted
antibodies with the corresponding antigens at 4 C overnight before the
immunohistochemical procedures. Monoclonal (rat) antichicken VDR
antibody (Affinity Bioreagents, Neshanic Station, NJ) was diluted in
1% BSA in PBS. The avidin-biotin complex method was performed as
reported previously (20). In brief, the deparafinized and rehydrated
specimens were incubated in 0.3% hydrogen peroxide in methanol for 30
min and were microwaved at 100 C in citrate buffer (pH 6.0) for the
retrieval of antigenicity. Then the specimens were incubated with
primary antibodies for 18 h (1:4000 dilution for anti-mRXR
antiserum, 1:2000 dilution for anti-mRXRß and anti-mRXR antisera,
and 1:100 dilution for anti-VDR antiserum), and subsequently incubated
with biotinylated antirabbit antibody (1:200 dilution, Vector
Laboratories, Burlingame, CA) for anti-mRXR antibodies and biotinylated
antirat antibody (1:200 dilution, Amersham International, Little
Chalfont, UK) for anti-VDR antibody. Then the sections were rinsed and
incubated with the avidin-biotin complex (Vector Laboratories). Final
colocalization was made by 0.2% 3,3'-diaminobenzidine
tetrahydrochloride (DAB) (Wako Pure Chemicals, Osaka, Japan) containing
0.005% hydrogen peroxide. In order to compare the localization of RXRs
and VDR, serial sections were used for the immunohistochemical
staining.
Preparation of nuclear extracts/Western immunoblot analyses
MDBK cells were grown in monolayer cultures at 37 C, 5%
CO2 in DMEM with 10% FCS, 100 U/ml penicillin, and 100
mg/ml streptomycin. When cultured MDBK cells were grown to 60%
confluence, media were changed to DMEM with 1% resin and
charcoal-treated calf serum (stripped medium) (21) and incubated for
12 h. Then, fresh-stripped medium was added and the cells were
incubated for additional 48 h in the absence or presence of either
10-7 M VD (Biomol, Plymouth Meeting, PA),
10-6 M 9cRA, or 3 mM
CaCl2 (final CaCl2 concentration including the
media was 5 mM) before preparation of nuclear extracts.
Nuclear extracts from MDBK cells and rat kidney were prepared as
previously described (15, 16). Ten micrograms of cellular nuclear
extracts were subjected to SDS-PAGE (9% acrylamide gel). After
SDS-PAGE, proteins were transferred to polyvinylidene difluoride (PVDF)
(Immobilon P, Millipore) in 20 mM Tris, 150 mM
glycine, 3.5 mM SDS, and 20% (vol/vol) methanol for 3
h at 75 mA constant current. Immunoblot analyses were performed as
previously described (16, 22). Briefly, the blots were blocked in 5%
nonfat dry milk/PBS/0.1% Tween-20 (Bio-Rad, Hercules, CA) at 4 C
overnight, and then incubated either with immune or preimmune
isoform-specific anti-mRXR antisera diluted 1:1000 for 1 h at room
temperature. After incubation with horseradish peroxidase-linked donkey
antirabbit Ig (Amersham International) diluted 1:5000 for 1 h at
room temperature, antibody/protein complexes on the blots were detected
using ECL detection reagents (Amersham International).
Transient transfection studies
DR3 oligonucleotides containing consensus half-sites sequence
AGGTCA arranged as a direct repeat containing nucleotide gaps of 3
(AGCTTACTTATTGAGGTCACTGAGGTCAAGTTACG) were
subcloned into the reporter vector, PT109 (23), which contains a viral
thymidine kinase promoter and the firefly luciferase cDNA as previously
described (24). MDBK cells were grown as described above. When cultured
MDBK cells were grown to 60% confluence, media were changed to the
stripped medium and incubated for 12 h. Then the cells were
transiently transfected using DEAE-dextran method. Briefly, 2 µg
reporter plasmid [either reporter plasmid PT109 itself (23) or above
described DR3-containing PT109 (24)] and 1.2 µg Rous sarcoma
virus-ß-galactosidase control plasmid were mixed with 27 µl of 10
mg/ml DEAE-dextran, 0.85 ml DMEM, and 0.85 µl of 100 mM
chloroquine per 3.5-cm plate. Three hours after transfection, cells
were treated with 10% dimethyl sulfoxide (DMSO), then incubated with
the fresh stripped medium for 24 h. The cells then were incubated with
either 10-6 M 9cRA, 10-7
M VD (Biomol), or both for additional 48 h. Identical
experiment was also performed except that 10-6
M RXR selective agonist LG100153 (named compound 6 g
in Ref. 19, kindly provided by Dr. R. A. Heyman, Ligand
Pharmaceuticals, San Diego, CA) was used instead of 9cRA. After
harvesting, the cell extracts were analyzed for both luciferase (25)
and ß-galactosidase (26) activity in order to correct for
transfection efficiency.
Preparation of in vitro translated receptors/electrophoretic
mobility shift assay (EMSA)
Previously described cDNA clones of mouse RXR in pBSK, human
RARß in pGEM1 (both are kindly provided by Dr. R. M. Evans, The Salk
Institute, San Diego, CA) (5, 27), and human VDR in pSG (kindly
provided by Dr. W. Hunziker, Hoffman-La Roche, Basil, Switzerland) (24)
were used in these experiments. Unlabeled and
[35S]-methionine labeled receptors were produced from
rabbit reticulocyte lysates according to the manufacturer’s
instructions (Promega, Madison, WI). Unprogrammed reticulocyte lysate
also was incubated under the same conditions.
[35S]-methionine labeled receptor proteins were
quantitated by SDS-PAGE analysis, which showed labeled proteins of
expected molecular weights. EMSA using DR3 as a probe was performed as
described previously (24).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, strong immunostaining was observed in
nuclei of most cells both in the proximal (examples are indicated by
arrows) and the distal (examples are indicated by
arrowheads) tubules by anti-mRXR antibody, which was
completely blocked by peptide preabsorption (Fig. 1C). No
immunostaining was observed in glomeruli (Fig. 1B). Although very
faint, but similar distributions of immunostaining were observed using
anti-mRXRß and anti-mRXR antibodies (data not shown). We next
performed Western immunoblot analyses using rat kidney nuclear extract.
As seen in Fig. 2 (lane 1), anti-mRXR antiserum
specifically detected RXR protein at the expected molecular weight
(44K) in contrast to its preimmune serum (lane 2), further confirming
the specificity of the RXR immunostaining. We could not detect
either RXRß or RXR protein by their respective antiserum by
Western immunoblot analyses (data not shown), probably due to their low
protein expression levels in rat kidney as observed in
immunostaining.
|
|
colocalizes with VDR in the tubular cells, we
next performed immunostaining of the serial section of Fig. 1B
using
anti-VDR antibody. As seen in Fig. 1D, strong immunostaining was
observed in nuclei of most cells both in the proximal and the distal
tubules by anti-VDR antibody. Comparison between Fig. 1, B and D,
revealed the cells immunostained by both anti-mRXR antibody and
anti-VDR antibody in the proximal (indicated by arrows in
Fig. 1, B and D) and the distal (indicated by arrowheads in
Fig. 1, B and D) tubules, indicating the colocalization of RXR and
VDR in those tubular cells.
Expression of endogenous RXR protein in MDBK cells
In order to study the functional significance of these
receptors in the tubular cells, we next performed transient
transfection studies. Among several tubular cell derived cell lines, we
chose MDBK cells for the studies since they were well known to express
endogenous VDR (17, 18). Additionally, the cells also express
endogenous calbindin D-28K protein which is regulated by VD (17, 18),
suggesting that they preserve intact VD signaling pathway. In order to
see the expression of endogenous RXR protein in the cells, we next
performed Western immunoblot analyses using MDBK nuclear extracts. As
observed in lane 1 of Fig. 3, anti-mRXR antiserum
specifically detected RXR protein at the expected molecular weight
(50K), which was not detected by its preimmune serum (Fig. 3, lane 5).
We could not detect either RXR or RXRß protein by their respective
antiserum, probably either due to their low protein expression levels
or species differences (the antisera were raised against mouse RXRs,
and MDBK cells were derived from bovine kidney cell). These data at
least suggest that MDBK cells express endogenous RXR protein in
addition to VDR. In order to see the regulation of RXR protein
expression, we treated MDBK cells with either 10-7
M VD, 10-6 M 9cRA, or 3
mM CaCl2 (final CaCl2 concentration
including the media was 5 mM), and compared the expression
level with the untreated state (Fig. 3, lane 1). As seen in Fig. 3, either VD (lane 2), 9cRA (lane 3), or high CaCl2
concentration (lane 4) did not affect the RXR protein expression
level.
|
, lanes 1–4). In contrast, when we
transfected DR3-containing PT109, incubation with 10-7
M VD alone induced a 5.4-fold increase (Fig. 4A, lane 6)
and 10-6 M 9cRA alone induced a 6.9-fold
increase (Fig. 4A, lane 7) in luciferase activity. Interestingly,
incubation with VD plus 9cRA showed an additive effect, and induced a
13.8-fold increase in luciferase activity (Fig. 4A, lane 8). Because
9cRA is known to bind and activate RARs equally well as RXRs (7), we
next used RXR selective agonist LG100153 (19) instead of 9cRA in order
to make sure that endogenous RXR was activated. As seen in Fig. 4B, 10-6 M LG100153 alone induced a similar
increase (2.6-fold, lane 7) as 10-7 M VD alone
did (3.0-fold, lane 6), and VD plus LG100153 also showed an additive
effect (5.6-fold, lane 8) in luciferase activity. These data suggest
that endogenous RXRs and VDR can co-regulate DR3 sequence in MDBK
cells.
|
, only RXR/VDR heterodimer (lane 6), but not other
combinations among RXR, VDR, and RAR (lanes 1–5, 7) could bind to the
DR3 sequence. These data further support the idea that the ligand
effect observed in the transfection studies were mediated by endogenous
RXR/VDR heterodimer bound to the DR3 sequence.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, localized to the proximal and the distal tubules
of the kidney using isoform-specific anti-mRXR antibodies that we
generated and confirmed their isoform specificities by Western
immunoblot analyses, supershifts using EMSA, immunoprecipitations, and
immunostainings (16). The serial section staining using anti-VDR
antibody demonstrated the colocalization of RXR and VDR in the
tubular cells. Interestingly, in contrast to the previous report
describing the restricted expression of VDR in the distal tubules in
rat kidney (12), we could observe nuclear staining of VDR both in the
proximal and the distal tubules although the same anti-(chicken) VDR
antibody was used. The expression of VDR in the proximal tubules in rat
kidney was previously demonstrated by [3H]-labeled VD
uptake (13). Additionally, Kumar et al. (14) previously
demonstrated the immunostaining of VDR both in the proximal and the
distal tubules in human kidney using anti-human VDR antibody. As we
have used microwave retrieval method in order to increase the
antigenicity for immunostaining (20), we think that the difference of
VDR localization comparing with the previous report (12) was simply due
to the difference of threshold level of immunostaining for detecting
VDR in the proximal tubules. We believe that this is the first
description of the co-localization of VDR and RXR in the VD target
cells. Although it has recently been well recognized that VDR and RXR can heterodimerize and transactivate DR3 type VDRE from in vitro studies (9, 10), it is not fully understood whether endogenous VDR and RXR in VD target cells can also work together for DR3 transactivation. Because the tubular cells in the kidney is one of the important target tissues of VD, and we have demonstrated the colocalization of VDR and RXR there, we considered that the tubular cells might be a good system for the functional analyses of endogenous VDR and RXR. Because it is technically difficult to obtain pure tubular cells for primary culture to perform the transfection studies, we decided to use clonal cell lines instead. There are several renal cell lines, such as MDBK cells, Madin Darby canine kidney (MDCK) cells, LLC-PK1 (pig kidney) cells, and OK (opposum kidney) cells, which are frequently used and well characterized (17). Among these cell lines, MDBK cells are well known to express endogenous VDR (17, 18). Additionally, the cells preserve several characteristics of the distal tubules including the expression of endogenous calbindin D-28K protein which is regulated by VD (17, 18). As we have observed the expression of endogenous RXR protein in MDBK cells, we speculate that the cells preserve intact VD signaling pathway mediated by endogenous VDR and RXR. We therefore chose MDBK cells for the transfection studies. Interestingly, 9cRA alone induced a almost similar increase in luciferase activity as VD alone did, and VD plus 9cRA showed an additive effect. Because 9cRA is known to bind and activate RARs equally well as RXRs (7), we also used RXR selective agonist LG100153 in order to make sure that endogenous RXR was activated. As expected, LG100153 induced a similar increase as 9cRA did both in the absence or presence of VD. Because EMSA demonstrated that only RXR/VDR heterodimer, but not other combinations among RXR, VDR, and RAR, could bind to the DR3 sequence we used, we consider that the DR3 transactivation induced by VD and 9cRA or LG100153 in MDBK cells was mediated by RXR/VDR heterodimer formed by endogenous receptors. The exact mechanisms of VD mediated calcium reabsorption in the distal tubules are still unclear. It has recently been reported that at least calbindin D-28K and 9K are involved in this process (28, 29). Because DR3 like sequences are observed in the promoter region of both of these genes (18, 30), we speculate that VD mediated calcium reabsorption in the distal tubules may be mediated at least in part by transactivation of DR3-containing calcium regulating genes, including calbindins, by endogenous RXR/VDR heterodimer complexes.
|
Acknowledgments |
|---|
Received April 7, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
,25-dihydroxyvitamin D3: an overview. J Nutri 125:1687S–1689S
This article has been cited by other articles:
 |
|
 |
|
  J. Adams, E. Kiss, A. B.V. Arroyo, M. Bonrouhi, Q. Sun, Z. Li, N. Gretz, A. Schnitger, C. C. Zouboulis, M. Wiesel, et al. 13-cis Retinoic Acid Inhibits Development and Progression of Chronic Allograft Nephropathy Am. J. Pathol., July 1, 2005; 167(1): 285 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Liebler, B. Uberschar, H. Kubert, S. Brems, A. Schnitger, M. Tsukada, C. C. Zouboulis, E. Ritz, and J. Wagner The renal retinoid system: time-dependent activation in experimental glomerulonephritis Am J Physiol Renal Physiol, March 1, 2004; 286(3): F458 - F465. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. Sugawara, K. Takeuchi, A. Uruno, Y. Ikeda, S. Arima, M. Kudo, K. Sato, Y. Taniyama, and S. Ito Transcriptional Suppression of Type 1 Angiotensin II Receptor Gene Expression by Peroxisome Proliferator-Activated Receptor-{{gamma}} in Vascular Smooth Muscle Cells Endocrinology, July 1, 2001; 142(7): 3125 - 3134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Haxsen, S. Adam-Stitah, E. Ritz, and J. Wagner Retinoids Inhibit the Actions of Angiotensin II on Vascular Smooth Muscle Cells Circ. Res., March 30, 2001; 88(6): 637 - 644. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Wagner Potential role of retinoids in the therapy of renal disease Nephrol. Dial. Transplant., March 1, 2001; 16(3): 441 - 444. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Li, V. Desai-Yajnik, E. Lo, M. Schapira, R. Abagyan, and H. H. Samuels NRIF3 Is a Novel Coactivator Mediating Functional Specificity of Nuclear Hormone Receptors Mol. Cell. Biol., October 1, 1999; 19(10): 7191 - 7202. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Imaizumi, H. Suzuki, M. Yoshinari, A. Sato, T. Saito, A. Sugawara, S. Tsuchiya, Y. Hatae, T. Fujimoto, A. Kakizuka, et al. Mutations in the E-Domain of RARalpha Portion of the PML/RARalpha Chimeric Gene May Confer Clinical Resistance to All-trans Retinoic Acid in Acute Promyelocytic Leukemia Blood, July 15, 1998; 92(2): 374 - 382. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Prufer, A. Racz, G. C. Lin, and J. Barsony Dimerization with Retinoid X Receptors Promotes Nuclear Localization and Subnuclear Targeting of Vitamin D Receptors J. Biol. Chem., December 22, 2000; 275(52): 41114 - 41123. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Lehrke, M. Schaier, K. Schade, C. Morath, R. Waldherr, E. Ritz, and J. Wagner Retinoid receptor-specific agonists alleviate experimental glomerulonephritis Am J Physiol Renal Physiol, April 1, 2002; 282(4): F741 - F751. [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 |
