| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Medicine (L.G.S., S.W.S.) and Physiology and Biophysics (S.W.S.), State University of New York and Veterans Affairs Western New York Healthcare System, Buffalo, New York 14215
Address all correspondence and requests for reprints to: Dr. Stephen W. Spaulding, Veterans Affairs Western New York Healthcare System, 3495 Bailey Avenue, Buffalo, New York 14215. E-mail: medspaul{at}acsu.buffalo.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HuR (also called HuA in mice) is a member of the highly conserved ELAV (embryonic lethal abnormal vision) family of proteins and is expressed in many cell types (1, 7, 14, 15, 16). HuR has been shown to shuttle between the nucleus and cytoplasm (17) and has a nuclear shuttling sequence (HNS) similar to the nuclear localization sequence M9 in the well characterized shuttling protein, hnRNP A1 (18, 19). The mechanisms governing the subcellular localization of different RNA-binding proteins remain obscure, although several pathways have been proposed for the modulation of their RNA and protein binding activities to regulate the targeting, translation, and turnover of the AU-rich mRNAs (1, 5, 6, 15, 17, 18, 19).
Overexpression of HuR has been shown to stabilize AU-rich mRNAs in several cell lines (18, 20, 21), and the RNA binding activity of HuR has been shown to protect AU-rich mRNAs from degradation in response to various stresses or stimuli in several in vitro models (4, 14, 21). The binding of HuR to a U-rich sequence in the 5'UTR of p27 mRNA protects this element from endoribonuclease cleavage, indicating a direct mechanism by which HuR can protect mRNA from endonuclease digestion (14). Cytoplasmic HuR in HeLa cells is predominantly associated with polyribosomes (4). In response to heat shock, HuR relocalizes along with mRNA to a protected nuclear/perinuclear compartment (4), suggesting that HuR can protect mRNA from degradation by altering its subcellular localization (14, 15, 18).
We previously demonstrated that androgen alters epidermal growth factor (EGF) mRNA at several posttranscriptional levels in the murine submaxillary salivary gland (SMG) model, and that androgen levels change the activities of several proteins that bind to a unique 23-base AU-rich element in the 3'UTR of EGF mRNA (22). Additionally, we have recently shown that one AU-rich RNA-binding protein that is dramatically altered in the SMG, but not the kidney, is AUF1 (23). The pattern of expression of AUF1 isoforms in the SMG is sexually dimorphic, and the cytosolic level of the p37/p42 pair of isoforms, which have the highest AU-rich binding and destabilizing activities (5, 6), correlates directly with the circulating level of androgen in both male and female BALB/c mice (23). In contrast to the destabilizing activity associated with AUF1, HuR generally stabilizes AU-rich mRNAs (7). We therefore performed experiments to determine whether androgen also regulates HuR in the androgen-responsive HepG2 cell line in vitro (23) and in nonreproductive tissues of mice in vivo.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animal studies
Experiments were conducted in accordance with NIH guidelines for
animal treatment and housing. Litters of young adult (10–12 weeks old)
female BALB/c mice were divided into groups that contained
representatives of each litter. The mice were injected with
testosterone propionate (200 µg, sc, every other day for up to 7 days
and were killed at 3, 5, and 7 days) or DHT (200 µg/day, sc, and
killed at 6, 24, and 48 h) and then given a lethal ip injection of
pentobarbital. The SMGs and kidneys from DHT-treated animals were
removed and either processed immediately (for testosterone and
orchiectomy studies) or snap-frozen in liquid N2
and processed at a later date. The tissue samples were homogenized
(Dounce) in 1 ml freshly prepared iced buffer A [20 mM
HEPES (pH 7.4), 50 mM potassium acetate, 5 mM
magnesium acetate, protease inhibitors (5 µg/ml
phenylmethylsulfonylfluoride, 10 µM soybean trypsin
inhibitor, and 10 µM leupeptin), Prime ribonuclease
inhibitor (1 U/µl; 5 Prime-3 Prime, Inc., Boulder, CO), and 1
mM dithiothreitol]. Litters of young adult male BALB/c
mice were divided into two groups containing equal numbers of
representatives from each litter. Under pentobarbital anesthesia they
underwent either a sham operation or orchiectomy and were killed 2
weeks later with a lethal dose of pentobarbital. The SMG and kidneys
were removed from the male mice and homogenized as described above for
female SMGs and kidneys.
Preparation of cytosol, nuclear. and polysome extracts
We centrifuged 1 ml homogenized (Dounce) tissues or cells
prepared in buffer A at 800 x g for 10 min at 4 C. The
800 x g pellets were resuspended in sodium chloride
(0.3 M final concentration) and centrifuged at
12,000 x g for 10 min at 4 C, and the supernatants
(nuclear extracts) were frozen at -70 C. The 800 x g
supernatants were layered on 4-ml cushions of 30% sucrose in iced
buffer A, then centrifuged at 100,000 x g for 2 h
at 4 C, and the supernatants (cytosol extracts) were frozen at -70 C.
The 100,000 x g pellet was also resuspended in buffer
A containing sodium chloride (0.3 M final
concentration), incubated on ice for 1 h, and centrifuged at
10,000 x g for 15 min at 4 C, and then the resultant
supernatant (polysomal extracts) was frozen at -70 C
(22).
Western blotting
Protein concentrations in the samples were determined with
bicinchoninic acid (Pierce Chemical Co., Rockford, IL)
using BSA as a standard. Samples were boiled in SDS-PAGE sample buffer
[final concentrations, 62.5 mM Tris-Cl (pH 6.8), 2% SDS,
10% glycerol, and 5% 2-mercaptoethanol] and briefly centrifuged, and
approximately 50 µg protein/lane were separated on SDS-polyacrylamide
gels (12.5%) as previously described (23). The proteins
were then transferred to polyvinylidene difluoride membranes
(Immobilon) using a Bio-Rad Laboratories, Inc. (Hercules,
CA) Transblot apparatus as previously described (24). We
established that equal amounts of protein had been transferred from
each lane by Ponceau staining of each membrane immediately after
electrophoresis (23, 24). Tris-buffered saline with 0.1%
Tween-20 was used for all incubations and washes, and all incubations
were performed for 1 h at 20 C. The membranes were blocked with
5% membrane blocking agent from Bio-Rad Laboratories, Inc. , then incubated with the appropriate primary antibody (see
figure legends for antibody dilutions). The blots were washed three
times for 15 min each time in Tris-buffered saline with 0.1% Tween-20
after each incubation step. Immunoreactive bands were visualized with
secondary antibody conjugated to alkaline phosphatase and detected
using enhanced chemifluorescence according to the manufacturer’s
instructions (Amersham Pharmacia Biotech, Aylesbury, UK).
Immunoreactivity was quantified on a STORM imaging system
(Molecular Dynamics, Inc., Sunnyvale, CA) and expressed in
arbitrary fluorescence units using ImageQuant software (version 5.0,
Molecular Dynamics, Inc.). Results are expressed as the
mean ± 1 SEM, and statistical significance of
differences was determined by ANOVA (StatView+
graphics, version 1.03, Abacus Concepts, Berkeley, CA). Digital images
were prepared using Microsoft (Redmond, WA) PowerPoint
(MacIntosh 1998 version).
Antibodies
Mouse monoclonal antibody to HuR was obtained from Santa Cruz Biotechnology, Inc. (sc5261, Santa Cruz CA), and alkaline
phosphatase-conjugated rabbit antimouse IgG antibody was purchased from
Bio-Rad Laboratories, Inc. Two new antibodies to AUF1 were
obtained from Tolnay (25): antibody P1b preferentially
detects the AUF1 isoforms that contain exon 2 (p45/p40), and antibody
P3a preferentially detects the AUF1 isoforms missing exon 7
(p37/40).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). Samples
that contained high levels of HuR commonly also contained a slightly
smaller immunoreactive HuR band, which has been described as a
degradation product of HuR (4). Relatively high levels of
HuR were found in HeLa, A431, HepG2, HT29, and Jurkat cells (and mouse
spleen, see below), lower levels were found in LNCaP cells and female
SMG, and the lowest level was in male SMG, which was approximately
1/10th of that found in female SMG. This suggested that HuR might be
androgen responsive.
|
). Extracts from cells treated with
vehicle alone for 48 h had a mean HuR level of 27.1 ± 1.5
relative fluorescence units, whereas the level in DHT-treated HepG2
extracts had fallen to 6.2 ± 5.9 (n = 3; P
< 0.05; Fig. 2B). Thus, total levels of HuR in this human
hepatoblastoma cell line are reduced by androgen.
|
, and the time course is plotted in
Fig. 3B. HuR levels in both nuclei and cytosol declined, and by 48
h had fallen by 37% and 65%, respectively [from 23 ± 2.28 to
14.5 ± 9.7 (n = 3; P < 0.05) and from
3.4 ± 0.02 to 1.1 ± 0.03 (n = 3; P <
0.05); Fig. 3B]. In contrast, the level of HuR in polyribosomes
steadily increased, reaching 325% above the control value at 48 h
(from 2.28 ± 0.4 to 9.7 ± 0.4; n = 3;
P < 0.05; Fig. 3B). These data indicate that androgen
influences the shuttling of HuR between subcellular compartments and
suggest that this mechanism could be involved in the actions of
androgen on the metabolism of AU-rich mRNAs that display
posttranscriptional regulation by altering their targeted transport,
stability, and/or translational efficiency (22, 26, 27, 28, 29).
|
), although the female SMG contains only
10–20% as much HuR as the female kidney and spleen (Fig. 4B). The
level in the male SMG is 20% of the level in the female SMG (0.23
± 0.1 vs. 1.3 ± 0.05; n = 3; P
< 0.05), and in the male kidney the HuR level is 15% of that in the
female (0.89 ± 0.5 vs. 7.9 ± 0.76; n = 3;
P < 0.05), whereas HuR levels in spleen were not
different between the sexes (13.1 vs. 14.4, means of
duplicate samples; Fig. 4B). HuR levels in female and male thyroid
samples were also not significantly different (not shown).
|
, orchiectomy increased HuR levels in
both SMG and kidney. The relative level of HuR in the male SMG
increased 8-fold (from 0.15 ± 0.003 to 1.32 ± 0.12; n
= 3, P < 0.05) after orchiectomy, approaching the
level of HuR present in control male kidneys (Fig. 5B). In male kidney,
orchiectomy doubled HuR levels (from 1.30 ± 0.15 to 3.63 ±
0.32; n = 6; P < 0.05). Conversely, DHT treatment
of female mice for 48 h decreased the relative level of HuR in the
SMG by 85% (from 1.6 ± 0.14 to 0.23 ± 0.04;
P < 0.05; n = 4) and that in female kidney by
80% (from 9.24 ± 0.4 to 1.86 ± 0.79; P <
0.05; n = 4) without altering the level of HuR in the female
spleen (for pairs of samples, mean of 18.5 vs. 15.3; Fig. 6, A and B). Treatment with testosterone
propionate also decreased HuR levels in female SMGs within 3 days, but
did not change HuR levels in the thyroid (not shown).
|
|
DHT changes HuR levels in polyribosomal extracts of SMG from female
mice
HuR levels in SMG polyribosomal extracts doubled when female mice
were given DHT for 6 h (10.3 ± 1.6 vs. 4.9
± 1.8; n = 6; P < 0.05), whereas HuR levels
decreased substantially in mice given DHT for longer periods (1.06
± 0.18 at 24 h and 1.8 ± 0.35 at 48 h; n = 6;
P < 0.05 for both; Fig 7). HuR levels in both the cytosol and
nuclei of female SMG fell progressively (not shown), similar to the
findings in HepG2 cells reported above. This transient increase in the
level of polyribosomal HuR indicates that subcellular shuttling of
HuR also occurs in vivo in response to androgen
treatment.
|
, A and B). The major
AUF1 isoform detected by reprobing with P3a antibody in control male
SMG cytosol is p37, and orchiectomy decreased its level by 85%
(P < 0.05; n = 3; Fig. 8, A and B). Conversely,
the major AUF1 isoform in SMG cytosol from female mice was p45, and DHT
treatment decreased p45 levels by approximately 50% (from 5.1 ±
0.8 to 2.4 ± 1.1; n = 6; P < 0.05) as
detected with P1b (Fig. 8C). DHT treatment increased p37 levels in
female SMG cytosol by approximately 60% (from 6.6 ± 0.7 to
10.7 ± 0.7; n = 6; P < 0.05), when the same
blot was reprobed with P3a (Fig. 8C). In kidney cytosol, the major AUF1
isoform was p37 (detected with P3a); very little p45 was detected (with
P1b), and the levels of both isoforms did not change after DHT
treatment (not shown). In addition to confirming our previous findings
(23), this further emphasizes the dramatic difference in
androgen responsiveness of AUF1 and HuR between the two tissues. HuR
levels fall in both kidney and SMG when androgen levels rise, and rise
when androgen levels fall. In sharp contrast, AUF1 isoforms respond in
the SMG, but not the kidney (23). The tissue-specific
difference between these two major AU-rich mRNA-binding proteins would
appear to be an important factor in explaining the differences in
androgen-dependent posttranscriptional regulation of mRNAs previously
reported between these two tissues (22, 28, 29).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HuR can bind to a U-rich region, present in p27 mRNA (14), that conforms to the consensus AU-rich element described by Lagnado et al. (30). This element is involved in p27 mRNA turnover, and the binding of HuR prevents degradation of p27 mRNA by a specific endoribonuclease (14). Several androgen-responsive gene transcripts that display posttranscriptional regulation, including EGF, EGF receptor, androgen receptor, and AUF1 itself (22, 23, 26, 27, 31, 32), also contain AU-rich elements in their 3'UTRs, so androgen-dependent changes in HuR levels could affect the stability of these messages by protecting AU-rich elements from endoribonuclease degradation (14).
The mechanisms that regulate the shuttling of HuR to different subcellular compartments are poorly understood. One molecular feature important in regulating HuR distribution is its nuclear shuttling sequence, HNS, located in the hinge region between the second and third RNA recognition motifs present in HuR (18). When cells expressing fusion proteins bearing this HNS sequence are treated with actinomycin D, the level of the proteins in nuclei fall, whereas they rise in cytosol, indicating that active gene transcription is required for the nuclear import of HuR (18). Moreover, the HNS has recently been shown to interact with several nuclear proteins that have acid-rich C-terminal tails to promote nuclear export of HuR (33). The HNS is homologous to the M9 signal in hnRNP A1 (19) as well as to a nuclear retention signal encoded by exon 7 of AUF1 (18, 34). Exon 7 is only present in the p42 and p45 isoforms of AUF1, so enhanced nuclear retention of p45 (25, 34) might partially explain why DHT treatment does not decrease nuclear p45 levels as much as it decreases cytosolic p45 levels in the female SMG (23).
HuR and the p37 isoform of AUF1 display inverse responses to androgen in the cytosol of the mouse SMG. We previously showed that increasing the level of androgen in female mice reduces the level of the p45/p40 pair while raising the level of the p37/p42 pair of AUF1 isoforms in SMG cytosol, whereas orchiectomy caused the opposite response in the cytosol of male SMG (23). Higher levels of the p37/p42 AUF1 isoform pair correlate with enhanced cytoplasmic degradation of AU-rich mRNAs (5, 6). Using two new isoform-specific anti-AUF1 antibodies (25) we have now independently confirmed these androgen-dependent changes in AUF1 isoforms and, more importantly, have correlated these changes with the very different changes in HuR. As HuR and AUF1 do not appear to possess endoribonuclease activity per se (6, 14), it is possible that an androgen-responsive nuclease involved in AU-rich mRNA degradation could be selectively induced in polyribosomes or other subcellular fractions of certain androgen-responsive tissues.
Our observation that HuR levels in both the SMG and kidney respond to androgen sharply contrasts with the absence of an AUF1 response in the kidney (Ref. 23 and this report). The difference between the effects of androgen on AUF1 and HuR in these two tissues could partly explain tissue-specific differences in posttranscriptional regulation of AU-rich mRNAs as a consequence of the subcellular redistribution of HuR that occurs during androgen-dependent differentiation and/or hypertrophy in some nonreproductive tissues (22, 23, 27, 28, 29). A recent survey of tissues from male mice at different stages of development found that the total levels of HuR and AUF1 changed in parallel (35), but sexual dimorphism and the androgen-dependent changes in subcellular distribution of these two hnRNPs would not have been detected.
In several human cell lines previously reported to have high proliferation rates (Jurkat, HT29, HepG2, A431, and HeLa), we found relatively high basal levels of HuR, a finding similar to earlier reports on HuR levels in both human and murine cells (7, 17, 36). In contrast, we found the level of HuR to be substantially lower in LNCaP cells, a prostate line derived from reproductive tissue that displays a low proliferation rate and high androgen responsiveness (37). Thus, it will be interesting to determine how the metabolism and distribution of AU-rich mRNA correlate with the relative levels of HuR expressed in androgen-unresponsive human prostate cancer cell lines that display different rates of cell proliferation and metastasis (38, 39), such as PC-3 cells, which have HuR levels 10 times greater than those in LNCaP cells (Sheflin, L. G., and S. W. Spaulding, unpublished). In summary, despite the broad tissue distribution of HuR (4, 7, 17, 36), androgen-mediated cellular differentiation plays a pivotal role in regulating the expression and subcellular targeting of HuR in several tissues. Disruption of hormone regulation of HuR could result in alterations in the turnover of early response gene transcripts involved in cellular proliferation, transformation, and/or carcinogenesis (1, 2, 17, 40, 41, 42).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received October 30, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
6ß4 integrin-mediated
invasive phenotype. Endocrinology 141:3172–3182This article has been cited by other articles:
 |
|
 |
|
  Z. Hu, B. G. Zimmermann, H. Zhou, J. Wang, B. S. Henson, W. Yu, D. Elashoff, G. Krupp, and D. T. Wong Exon-Level Expression Profiling: A Comprehensive Transcriptome Analysis of Oral Fluids Clin. Chem., May 1, 2008; 54(5): 824 - 832. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. H. Ing, D. A. Massuto, and L. A. Jaeger Estradiol Up-regulates AUF1p45 Binding to Stabilizing Regions within the 3'-Untranslated Region of Estrogen Receptor {alpha} mRNA J. Biol. Chem., January 18, 2008; 283(3): 1764 - 1772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Lu, L. Zheng, L. Viera, E. Suswam, Y. Li, X. Li, A. G. Estevez, and P. H. King Mutant Cu/Zn-Superoxide Dismutase Associated with Amyotrophic Lateral Sclerosis Destabilizes Vascular Endothelial Growth Factor mRNA and Downregulates Its Expression J. Neurosci., July 25, 2007; 27(30): 7929 - 7938. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N.S. Treister, S.M. Richards, T. Suzuki, R.V. Jensen, and D.A. Sullivan Influence of Androgens on Gene Expression in the BALB/c Mouse Submandibular Gland J. Dent. Res., December 1, 2005; 84(12): 1187 - 1192. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. R. Manjithaya and R. R. Dighe The 3' Untranslated Region of Bovine Follicle-Stimulating Hormone {beta} Messenger RNA Downregulates Reporter Expression: Involvement of AU-Rich Elements and Transfactors Biol Reprod, October 1, 2004; 71(4): 1158 - 1166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Yasuda, S. Wada, Y. Arao, M. Kogawa, F. Kayama, and S. Katayama Interaction between 3' Untranslated Region of Calcitonin Receptor Messenger Ribonucleic Acid (RNA) and Adenylate/Uridylate (AU)-Rich Element Binding Proteins (AU-Rich RNA-Binding Factor 1 and Hu Antigen R) Endocrinology, April 1, 2004; 145(4): 1730 - 1738. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Auboeuf, D. H. Dowhan, Y. K. Kang, K. Larkin, J. W. Lee, S. M. Berget, and B. W. O'Malley Differential recruitment of nuclear receptor coactivators may determine alternative RNA splice site choice in target genes PNAS, February 24, 2004; 101(8): 2270 - 2274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Denkert, W. Weichert, S. Pest, I. Koch, D. Licht, M. Kobel, A. Reles, J. Sehouli, M. Dietel, and S. Hauptmann Overexpression of the Embryonic-Lethal Abnormal Vision-like Protein HuR in Ovarian Carcinoma Is a Prognostic Factor and Is Associated with Increased Cyclooxygenase 2 Expression Cancer Res., January 1, 2004; 64(1): 189 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. B. Nabors, E. Suswam, Y. Huang, X. Yang, M. J. Johnson, and P. H. King Tumor Necrosis Factor {alpha} Induces Angiogenic Factor Up-Regulation in Malignant Glioma Cells: A Role for RNA Stabilization and HuR Cancer Res., July 15, 2003; 63(14): 4181 - 4187. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. B. Yeap, D. C. Voon, J. P. Vivian, R. K. McCulloch, A. M. Thomson, K. M. Giles, M. F. Czyzyk-Krzeska, H. Furneaux, M. C. J. Wilce, J. A. Wilce, et al. Novel Binding of HuR and Poly(C)-binding Protein to a Conserved UC-rich Motif within the 3'-Untranslated Region of the Androgen Receptor Messenger RNA J. Biol. Chem., July 19, 2002; 277(30): 27183 - 27192. [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 |
, female tissues; 
, male tissues).
