| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Third Department of Internal Medicine, Yamanashi Medical University, Tamaho, Yamanashi 409-3898, Japan
Address all correspondence and requests for reprints to: Toshimasa Onaya, M.D., Ph.D., Third Department of Internal Medicine, Yamanashi Medical University, Tamaho, Yamanashi 409-3898, Japan. E-mail: onayat{at}res.yamanashi-med.ac.jp
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
On the other hand, physiological roles of aldose reductase are not yet completely understood. Renal medullary cells accumulate large amounts of organic osmolytes, including sorbitol, to compensate for the interstitial hypertonicity (11, 12, 13). Hypertonicity elevates the activity and abundance of aldose reductase by increasing transcription of its gene (13, 14, 15, 16, 17). Induction of aldose reductase by hypertonic media was also demonstrated in kidney mesangial cells (18), renal proximal tubule cells (19), Chinese hamster ovary cells (18), retinal pigment epithelial cells (20, 21), aortic smooth muscle cells (22, 23), and arterial endothelial cells (24).
Recently, the sequence of the putative osmotic response element of the canine betain transporter (GBT1) gene, TonE (tonicity-responsive element), was described (25). Then, the 11-bp osmotic response elements (OREs) of the rabbit, human, and mouse aldose reductase genes were reported (26, 27, 28, 29).
In our previous report (22), we showed that aldose reductase messenger RNA (mRNA) levels as well as its activity were induced by glucose in a concentration-dependent manner up to 205.5 mM in a rat aortic smooth muscle cell line, A7r5 cells. Glucose was more effective than NaCl in the induction of aldose reductase mRNA and activity (22).
In this report, we examined the entire 3.5-kb 5'-flanking sequence of rat aldose reductase by deletion mutagenesis and transfection studies for the presence of a transcription regulatory element to hyperosmolarity or ORE. We also analyzed the nuclear proteins that bind to the element.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-32P]Deoxy-ATP (
6000 Ci/mmol) was purchased
from Amersham (Aylesbury, UK). Promega Corp. (Madison, WI)
provided the Dual Luciferase Assay System and luciferase vectors. Calf
intestinal alkaline phosphatase (CIP), TaKaRa Taq
polymerase, and other restriction enzymes were obtained from Takara
(Kyoto, Japan). Protein tyrosine phosphatase (PTPase) and protein
phosphatase 1, catalytic subunit (PP1), were obtained from Boehringer
Mannheim (Mannheim, Germany).
Genomic cloning
The 1.3-kb rat aldose reductase complementary DNA (cDNA) was
obtained by PCR, using a pair of oligonucleotide primers: AR cDNA
forward and reverse (Table 1
). The PCR
product was cloned into pCR2.1 (Invitrogen, San Diego, CA)
and confirmed that the sequence was identical to that in the previous
report (30). A total of 5.2 x 105 clones of rat
genomic library in EMBL-3 SP6/T7 (CLONTECH Laboratories,
Inc., Palo Alto, CA) were screened with the cDNA, which was labeled
with BcaBest labeling kit (Takara). One clone was isolated from the
library. The clone contained more than 16 kb of the insert. A 4.5-kb
XbaI fragment containing the 3.5 kb 5' of the transcription
initiation site was subcloned into pBluescript to give subclone
p513–1. The nucleotide sequence of the 3'-part of this clone was
identical to the reported 5'-flanking sequence (-1335 to +177) of the
rat aldose reductase gene (31).
|
).
The PCR product was digested with MluI and XhoI,
isolated from agarose gels, and directionally inserted into the pGL3
basic vector (Promega Corp.), which contains the firefly
luciferase sequence downstream from the cloning cassette. The resulting
plasmid, pARLuc-3.5k, was digested with exonuclease III using the
Takara deletion kit (Takara) following the manufacturer’s
instructions, to generate progressively 5'-deleted vectors
(pARLuc-2.9k, pARLuc-2.4k, pARLuc-1.9k, pARLuc-1447, pARLuc-1335,
pARLuc-1216, pARLuc-1047, pARLuc-946, pARLuc-683, pARLuc-541, and
pARLuc-270). Briefly, 20 mg pARLuc-3.5k plasmid DNA were digested with
KpnI and MluI, treated with exonuclease III for
various periods followed by heat inactivation of the enzyme, then mixed
with mung bean nuclease, treated with Klenow fragment, and religated
with T4 DNA ligase. The deletion constructs were sequenced to determine
the extent of deletions.
To construct aldose reductase gene/reporter plasmids of which the
5'-end of the aldose reductase gene was located between -1216 and
-1047 bp (pARLuc-1148, pARLuc-1128, pARLuc-1111, and pARLuc-1073), PCR
was used with corresponding primers with a MluI restriction
site (-1148Mlu, -1128Mlu, -1111Mlu,
and -1073Mlu; Table 1) and the primer ARXho6. The PCR
products were inserted into the MluI-XhoI site of
the pGL3 basic vector.
To investigate the putative rORE identified in this study and a homologous sequence (rpsuedoORE) found 5' to the putative ORE, we made the following oligonucleotides and their complementary oligonucleotides: rORE, 5'-gatcAACTGGAAAATCACCAG-3'; and rpseudoORE, 5'-gatcAAGTGGAAAATATCTGT-3'. The complementary oligonucleotides were annealed and self-ligated to generate concatemers and inserted into a BamHI site of the pGL3 promoter (Promega Corp.), which contains the heterologous simian virus 40 (SV40) promoter upstream of the luciferase reporter gene. The number of oligonucleotides and the orientation and sequences of the constructs were confirmed by sequencing. pSV40Luc(rORE)1, pSV40Luc(rORE)3, pSV40Luc(rORE)4, and pSV40Luc(rpseudoORE)4 contain one, three, and four copies of the rORE sequence and four copies of the rpseudoORE sequence in tandem, respectively.
To subclone the fragment -1220/-1049 into pGL3 promoter vector, the
region was amplified by PCR, digested with BamHI, and
ligated into a BamHI site of pGL3 promoter vector. The
primers used were -1220Bam and -1049Bam (Table 1). The sequence of the resulting plasmid (pSV40Luc-1220/-1049) was
confirmed.
Cell culture and plasmids transfection
Rat aortic smooth muscle cell line A7r5 cells were maintained in
DMEM containing 5.5 mM glucose supplemented with 10% FCS
(22). A7r5 cells were seeded into 12-well culture plates and
transfected 24 h later using Lipofectamine reagent (Life Technologies) and OptiMEM I (Life Technologies)
following the manufacturer’s instructions with 100 ng luciferase
construct plasmids and 8 ng pRL-CMV (Promega Corp.), a
plasmid containing the Renilla luciferase gene under the
control of cytomegalovirus (CMV) immediate early enhancer/promoter, to
monitor the efficiency of transfection. Transfected cells were
maintained in isotonic medium for 24 h, then switched to
hypertonic medium (150 mM glucose or 75 mM
NaCl) or maintained in isotonic medium for another 20 h. The
osmolarity of those media was 324 ± 13 mosmol/kg H2O
(n = 9) for isotonic medium, 474 ± 17 mosmol/kg
H2O (n = 8) for glucose-supplemented medium, and
471 ± 12 mosmol/kg H2O (n = 9) for
NaCl-supplemented medium, respectively. The transfected cells were
harvested and assayed for luciferase activity using Dual Luciferase
Assay Systems (Promega Corp.) and a luminometer (Lumat
LB9501, Berthold Japan, Tokyo, Japan).
Nuclear extracts preparation
Nuclei were isolated from A7r5 cells cultured for 6 h in
the isotonic medium or the hypertonic medium as described above.
Nuclear extracts were prepared in the presence of protease inhibitors,
pepstatin A (2 mg/ml), leupeptin (2 mg/ml),
phenylmethylsulfonylfluoride (0.5 mM), and DTT (0.5
mM), as descried by Shimura et al. (32). Protein
concentrations were determined by the Bio-Rad protein assay kit
(Bio-Rad Laboratories, Inc., Richmond, CA). When the cells
were treated with cycloheximide (100 µg/ml), it was added 2 h
before switching the medium, and nuclear extracts were prepared in the
same manner.
Electrophoretic mobility shift assay
The synthetic complementary oligonucleotides (rORE, shown in
Fig. 4A) were annealed, labeled with [-32P]deoxy-ATP
by Klenow polymerase, and used as a probe. Nuclear extracts (3 µg)
were incubated with the probe (15,000 cpm) in a total volume of 20
µl binding buffer containing 10 mM Tris-HCl (pH 7.5),
10% glycerol, 50 mM NaCl, 0.5 mM DTT, and
0.05% Nonidet P-40. Binding reactions were carried out in the presence
of 1 µg poly(dI-dC)-poly(dI-dC) (Pharmacia Biotech, Piscataway, NJ) and with or without specific
competitors, as indicated. In experiments with phosphatases or
antiphosphotyrosine monoclonal antibody (PY20, Transduction Laboratories, Inc., Lexington, KY), the reaction mixtures were
incubated with phosphatases or antiphosphotyrosine antibody for 20 min
at 30 C before addition of the probe. Chemical oxidation of the thiols
in the protein was performed with diamide (1 and 3 mM;
Sigma Chemical Co., St. Louis, MO), an inorganic catalyst
of oxidation of thiols [(SH)2] to generate disulfides
(-S-S-) (33). Oxidation was carried out on ice for 5 min in the binding
buffer without DTT. Subsequent chemical reduction of the disulfides to
thiols was carried out with 3 mM DTT at 25 C for 5 min.
Then the probe was added. After a further 20 min at room temperature,
the binding mixtures were resolved by electrophoresis on a 4%
polyacrylamide gel (acrylamide-bisacrylamide, 30:0.8) in 40
mM Tris-HCl (pH 8.5), 190 mM glycine, and 1
mM EDTA at 150 V at 4 C. The gels were dried and analyzed
with a Fuji Photo Film Co., Ltd. Bioimage analyzer BAS
2000 (Fuji Photo Film Co., Ltd., Tokyo, Japan).
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, A and
B). The effect of glucose was significantly greater than that of NaCl
(P < 0.001). Other osmolytes (mannose, mannitol,
L-glucose, 3-O-methylglucose, galactose,
-methylglucoside, and LiCl) at the same osmolarity induced
luciferase activities similar to those produced by NaCl; the activities
again were lower than those induced by glucose (Fig. 1B). Figure 1C
shows the Northern blot analysis. Glucose induced aldose reductase mRNA
more than other osmolytes.
|
). However, further deletion up to bp
-1047 totally abolished the response to both glucose and NaCl.
The sequence of the region between bp -1216 and bp -1047 is shown in
Fig. 2A. We found four sequences similar
to ORE in this region. Therefore, we deleted those ORE-like sequences
one by one. A gradual reduction in the induction ratios upon deletion
was observed (Fig. 2B); however, pARLuc-1073 was still capable of
induction of about 2.5-fold. When all four ORE-like sequences
were deleted (pARLuc-1047), induction was abolished.
|
). The responses of
pSV40Luc(rORE)4, pSV40Luc(rORE)3, and
pSV40Luc(rORE)1 to the stimuli decreased as the copy number
of rORE decreased.
|
). When
pSV40Luc(rpseudoORE)4, which contains four copies of the
rpseudoORE sequence in tandem, was transfected, glucose and NaCl did
not increase the luciferase activity (Fig. 3). Deletion of this
sequence from the 5'-flanking region also did not affect the luciferase
activity (pARLuc-1128, Fig. 2B).
When pSV40Luc-1220/-1049, which contains four ORE-like sequences
(Fig. 2A), was transfected, and the cells were treated with glucose or
NaCl, the luciferase activity induction was comparable to that of
pSV40Luc(rORE)1 (Fig. 3).
Identification of DNA-binding activity in nuclear extract
The probes and competitors used in the electrophoretic mobility
shift assay are shown in Fig. 4A. Three
major complexes (I, II, and III) were formed (Fig. 4B). A broad band
(complex III) was formed with the nuclear extracts from cells cultured
in both isotonic and hypertonic media. Slowly migrating bands
(complexes I and II) were formed predominantly with nuclear extracts
from hypertonic cells. A 100-fold molar excess of unlabeled (cold) rORE
or 11-bp ORE (rORE core) eliminated complexes I and II, but not complex
III. The rpseudoORE sequence (-1148/-1132), which was inactive in the
luciferase assay (Fig. 3), did not compete any of the three bands. As
the rpseudoORE sequence differs in only 2 bases from the 11-bp ORE
(rORE core), we made mutated, double stranded oligonucleotides to
delineate the nucleotide sequence responsible for protein binding (Fig. 4B). rORE M1, in which a cytosine at -1062 bp is mutated to an
adenine, did compete for complexes I and II. On the other hand, rORE
M2, in which an adenine at -1061 bp is mutated to a thymine, did not
virtually compete for those complexes. Other nucleotide differences
between rORE and rpseudoORE (5', C-G; 3', CAG-TGT) are not involved in
DNA-protein interaction (Fig. 4B and data not shown).
When rpseudoORE was used as a probe, three bands appeared (a, b, and c,
Fig. 4C). The upper two bands (a and b) were more intense in nuclear
extracts prepared from isotonic and NaCl-treated cells than those from
glucose-treated cells. The bands a and b were competed by unlabeled
probe in a 50-fold molar excess, but were only slightly competed by
rORE in a 50-fold molar excess. Around rORE sequence, there were two
other sequences similar to rORE (Fig. 2A): fragment C (-1213/-1203)
and fragment D (-1125/-1115), which is in the opposite direction.
When these sequences were used as probes, no specific band was observed
(data not shown).
To investigate whether those proteins bound to the rORE were
phosphorylated, we treated the nuclear extracts with phosphatases (Fig. 5). Nuclear extracts from glucose- or
NaCl-treated A7r5 cells were incubated with CIP (nonspecific
phosphatase), PTPase (phosphotyrosine phosphatase), or PP1
(serine/threonine phosphatase) before addition of the labeled probe.
CIP and PP1 treatment inhibited the formation of complexes I and II to
the unstimulated level, but incubation with PTPase had no effect on the
complexes. When we incubated the nuclear extracts with
antiphosphotyrosine antibody before addition of the probe, the complex
formation was not affected (data not shown).
|
). When 1 mM DTT was added
to the nuclear extracts, the intensities of complexes I and II
decreased slightly. When the nuclear extracts were preincubated with 1
mM diamide to oxidize the thiols of the nuclear proteins,
formation of complexes I and II was not affected, and complex III was
totally abolished. Subsequent addition of 3 mM DTT had no
effect on complexes I and II and restored complex III.
|
, cycloheximide treatment abolished
induction of aldose reductase mRNA by glucose and NaCl. When nuclear
extracts were prepared from A7r5 cells that were pretreated with
cycloheximide and then exposed to glucose or NaCl, the formation of
complexes I and II was virtually abolished (Fig. 7B).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Deletion of the rORE sequence abolished the promoter activity, and the oligomers of rORE enhanced the heterologous promoter when stimulated by glucose or NaCl. In an electrophoretic mobility shift assay, rORE forms two specific complexes (I and II), which were predominantly observed in the nuclear extracts from glucose- or NaCl-treated cells and were competed by unlabeled rORE and the 11-bp rORE core (TGGAAAATCAC).
The rORE core shares a high degree of homology with other
osmolarity-responsive elements; rORE-core is identical to OREs of human
and mouse aldose reductase genes (28, 29) and is different from rabbit
ORE by one nucleotide (27), a cytosine at the first nucleotide in
rabbit and a thymine in rat, human, and mouse OREs (Fig. 4A). When a
cytosine at bp -1062 is substituted by an adenine (rORE M1) and used
as a competitor, formation of complexes I and II is competed, but when
an adenine at bp -1061 is substituted by a thymine (rORE M2),
complexes I and II are not affected. Therefore, an adenine, but not a
thymine, at bp -1061 is required for DNA-protein interaction.
Substitution of the first adenine within the 11-bp rabbit ORE by a
guanine was shown to delete the osmotic response (27).
As shown in Fig. 2B, the magnitude of induction dropped gradually with
decreasing fragment size from -1216 to -1073 bp. There are four
sequences similar to ORE in this region: fragment C, rpseudoORE,
fragment D, and rORE (Fig. 2A). Fragment D is in the opposite
direction. Recently, Ko et al. (28) suggested that the
ORE-like sequences present around human ORE (OreC) cooperatively
interact for the enhancer activity. Therefore, we investigated the
possibility that other elements located close to rORE may potentiate
the response, although rORE is a major cis-element of
induction as mentioned above. However, based on transfection analyzes
and electrophoretic mobility shift assays (Fig. 4, B and C, and data
not shown), we think that rpseudoORE and fragments C and D are not
involved in the transcriptional activation of the aldose reductase
gene. rpseudoORE forms complexes a and b, which are predominant in
nuclear extracts from control and NaCl-treated cells (Fig. 4C). The
meaning of this is unclear at present.
Although osmotic response elements have been identified in aldose
reductase gene and betain transporter gene as described above, the
protein that binds to the element has not been characterized. We showed
here partial characterization of the protein. When nuclear extracts
were treated with CIP or PP1, the formation of complexes I and II was
inhibited (Fig. 5). Incubation with PTPase had no effect on the
complexes. Preincubation of the nuclear extracts with
antiphosphotyrosine antibody also did not affect the complex formation.
Treatment of the nuclear extracts with diamide to oxidize the thiols
did not affect the formation of complexes I and II (Fig. 6), suggesting
that the redox state of thiols is not involved in the interaction of
rORE and rORE-binding proteins. Furthermore as shown in Fig. 7, pretreatment of the cells with cycloheximide abolished the induction of
aldose reductase mRNA and rORE-binding proteins. These results indicate
that phosphorylation at serine/threonine residues of rORE-binding
proteins is involved in the protein-DNA interaction and that new
protein synthesis is required to induce transcription of aldose
reductase gene.
In our previous studies (22), we showed that in A7r5 cells, glucose
induced aldose reductase mRNA and its activities more efficiently than
NaCl. Here we confirmed the results by transient transfection assays.
When pARLuc-3.5k was transfected, the luciferase activity induced by
glucose was 1.5-fold of that induced by NaCl. Glucose-specific
induction became obscure upon shortening of the test fragment in the
reporter vectors (Figs. 1A and 2B) and is not observed in transfection
with pARLuc(rORE)4 (Fig. 3). Furthermore, as shown in Fig. 1B, other osmolytes induced the luciferase activities no more than
NaCl, and the results in Northern blot analysis were similar to those
in transient transfection assays (Fig. 1C). Induction by
L-glucose, which is unable to enter the cells, and that by
3-O-methylglucose, which is not phosphorylated and further
metabolized, were also less than that by glucose. All of these results
suggest that the glucose-specific induction mechanism involves the
metabolism of glucose and is not medicated by rORE. Experiments to
elucidate its precise mechanism are under way.
Studies in yeast have revealed a two distinct transmembrane osmosensors, SLN1 and SHOP1, that regulate the mitogen-activated protein kinase cascade (PBS2-HOG1 pathway) (35, 36, 37). However, there is little information available regarding how mammalian cells recognize and transduce hyperosmotic stimuli to transcriptional machinery. Hypertonicity stimulates three cascades of MAP kinase homologs: ERKs (38, 39, 40), Jun N-terminal kinase-1/stress-activated protein kinase, and p38 (40, 41, 42, 43, 44, 45). In MDCK cells, however, the induction of the sodium/myo-inositol transporter mRNA by hypertonicity does not involve the ERK pathway (46). Recently, Kültz et al. (45) reported that although p38 and Jun N-terminal kinase-1/stress-activated protein kinase cascades are activated by hyperosmolarity in PAP-HT25 cells, activation of those pathways is not necessary for transcriptional regulation of the aldose reductase gene through the ORE. We also observed that inhibition of ERK does not affect the aldose reductase induction by glucose and NaCl (Aida, K., M. Tawata, and T. Onaya, manuscript in preparation). These results suggest that hyperosmotic stress signals through divergent pathways.
Isolation of the rORE-binding proteins and characterization of their interaction with the rORE should clarify the mechanisms of regulation of aldose reductase gene expression.
Received May 4, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-hydroxylase gene in male mouse. J Mol Endocrinol 11:213–222[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  C. E. Irarrazabal, C. K. Williams, M. A. Ely, M. J. Birrer, A. Garcia-Perez, M. B. Burg, and J. D. Ferraris Activator Protein-1 Contributes to High NaCl-induced Increase in Tonicity-responsive Enhancer/Osmotic Response Element-binding Protein Transactivating Activity J. Biol. Chem., February 1, 2008; 283(5): 2554 - 2563. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Burg, J. D. Ferraris, and N. I. Dmitrieva Cellular Response to Hyperosmotic Stresses Physiol Rev, October 1, 2007; 87(4): 1441 - 1474. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Yang, A. D. Hodgkinson, P. J. Oates, H. M. Kwon, B. A. Millward, and A. G. Demaine Elevated activity of transcription factor nuclear factor of activated T-cells 5 (NFAT5) and diabetic nephropathy. Diabetes, May 1, 2006; 55(5): 1450 - 1455. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. S. Galvez, J. A. Ulloa, M. Chiong, A. Criollo, V. Eisner, L. F. Barros, and S. Lavandero Aldose Reductase Induced by Hyperosmotic Stress Mediates Cardiomyocyte Apoptosis: DIFFERENTIAL EFFECTS OF SORBITOL AND MANNITOL J. Biol. Chem., October 3, 2003; 278(40): 38484 - 38494. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Kultz Evolution of Osmosensory MAP Kinase Signaling Pathways Integr. Comp. Biol., August 1, 2001; 41(4): 743 - 757. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Tian and D. M. Cohen Urea inhibits hypertonicity-inducible TonEBP expression and action Am J Physiol Renal Physiol, May 1, 2001; 280(5): F904 - F912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. N. Henry, R. N. Frank, S. R. Hootman, S. E. Rood, C. W. Heilig, and J. V. Busik Glucose-Specific Regulation of Aldose Reductase in Human Retinal Pigment Epithelial Cells In Vitro Invest. Ophthalmol. Vis. Sci., May 1, 2000; 41(6): 1554 - 1560. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
