AJP - GI Journal of Applied Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Gastrointest Liver Physiol 295: G542-G551, 2008. First published June 26, 2008; doi:10.1152/ajpgi.00081.2008
0193-1857/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/3/G542    most recent
00081.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lin, W.-Y.
Right arrow Articles by Lee, Y.-H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lin, W.-Y.
Right arrow Articles by Lee, Y.-H.

HORMONES AND SIGNALING

Hepatocyte nuclear factor-1{alpha} regulates glucocorticoid receptor expression to control postnatal body growth

Wan-Yi Lin, Yu-Jie Hu, and Ying-Hue Lee

Laboratory of Molecular Pathology, Institute of Molecular Biology, Academia Sinica, Taipei, Taiwan

Submitted 17 February 2008 ; accepted in final form 19 June 2008


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Hepatocyte nuclear factor 1{alpha} (HNF-1{alpha}) is a homeodomain-containing transcription factor and is important in postnatal growth and development in mice. In the HNF-1{alpha}-deficient liver, the expressions of a large set of growth hormone (GH)-responsive genes were significantly downregulated. By analyzing various HNF-1{alpha} mutant mice, we disclosed a mechanism by which hepatic HNF-1{alpha} regulates the expression of GH-responsive genes that are crucial for growth and development. We found that HNF-1{alpha} is required for the normal expression of glucocorticoid receptor (GR) specifically in livers. In the liver, GR, together with STAT5, is known to mediate the GH action by transactivating the GH-responsive genes that function in body growth and development. We further demonstrated that HNF-1{alpha} modulated GR gene expression by directly transactivating the GR gene promoter via a cryptic regulatory element located 3 bp upstream of the translation start site in exon 2 of the GR gene locus.

transactivation; promoter element

HEPATOCYTE NUCLEAR FACTOR 1{alpha} (HNF-1{alpha}) is a homeodomain-containing transcription factor (2). It regulates the expression of numerous genes crucial in liver and pancreas function and is associated with a form of non-insulin-dependent diabetes mellitus called maturity-onset diabetes of the young (MODY3) (12, 16, 32). In addition, HNF-1{alpha} is required for postnatal growth and development in mice as evidenced by the severe growth defects in the HNF-1{alpha}-null mice (11). Ablation of HNF-1{alpha} expression also causes high serum levels of growth hormone (GH) and low serum IGF-I and IGF binding protein-3, the characteristics of growth hormone insensitivity syndrome, suggesting that the HNF-1{alpha}-related growth retardation involves a defect in the GH signaling (11, 23). Interestingly, liver-specific recovery of HNF-1{alpha} function was able to restore growth to a level similar to that of normal mice, indicating that the hepatic HNF-1{alpha} regulatory network is sufficient for maintaining normal growth and development (12).

Postnatal growth, such as muscle mass and longitudinal growth, is promoted by GH signaling. In the original "somatomedin hypothesis," the growth promoting effects of GH on target tissues were not direct but were instead mediated by a serum factor secreted mostly from liver and known as somatomedin-C (later identified to be IGF-1) (4). However, this hypothesis was contradicted first by the evidence showing the direct effect of GH on bone growth and later by the fact that mice grew normally despite the abolishment of hepatic IGF-1 production (8, 30). Recently, it was also found that the muscle-specific ablation of Stat5 transcription factors for GH signaling caused significant reduction in postnatal growth and skeletal size (9). These findings lessened the importance of liver in the hormone regulation of growth. On the other hand, a lack of IGF-I binding protein acid-labile subunit (ALS) caused 65 and 10% reductions in circulating IGF-1 level and body weight, respectively. Further reduction of circulating IGF-1 levels by hepatic depletion of IGF-1 in the ALS-null mice severely retarded growth, suggesting that liver does play a role in growth control (28, 31). In addition, the liver-specific ablation of glucocorticoid receptor (GR) significantly reduced the expression of numerous Stat5 target genes regulated by GH and caused a growth defect similar to that found in the liver-specific Stat5 knockout mice (6, 27), establishing the crucial role of hepatic GR in the GH-regulated postnatal growth. These observations further support the important role of liver in the GH regulation of postnatal growth, although it is no longer supported that hepatic IGF-1 is a mediator for GH effects on growth.

HNF-1{alpha} is involved in modulating IGF-1 gene expression (11, 15). Liver IGF-1 expression is markedly reduced in Hnf-1{alpha} knockout mice but can be recovered upon reexpression of HNF-1{alpha} in the liver (11, 12). However, hepatic IGF-1 production is dispensable in the postnatal growth (30). This indicates that the regulation of body growth by hepatic HNF-1{alpha} involves an as yet unidentified mechanism. In this study, we examined the mechanism by which hepatic HNF-1{alpha} modulates postnatal growth. By analysis and comparison of the expressions of genes related to growth in the livers of various mutant HNF-1{alpha} strains including HNF-1{alpha}-null mice and liver- and/or pancreatic β cell-specific Hnf-1{alpha}-reexpressed mice, we found that HNF-1{alpha} plays a crucial role in the expression of hepatic GR, which may in turn modulate the expression of a large set of GH-responsive genes found affected by HNF-1{alpha} deficiency. Furthermore, by employing the chromatin immunoprecipitation (ChIP) method and the in vivo promoter activity assays to search for functional HNF-1{alpha} regulatory sites on the GR gene, we revealed that HNF-1{alpha} directly regulates the transcriptional activity of GR gene via a binding sequence located not in the promoter region but 3 bp upstream of the translation start site in Exon 2.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Animals. All animal breeding and experimental protocols were approved by the Institutional Animal Care and Use Committee at Institute of Molecular Biology, Academia Sinica. The HNF-1{alpha} knockout (–) and knock-inactive (kin) heterozygous mice, hnf-1{alpha}+/– and hnf-1{alpha}kin/+, respectively (12), and alb.cre and ins2.cre transgenic mice (20) were maintained in a specific pathogen-free animal facility with 12:12-h light-dark cycles. To reexpress the HNF-1{alpha} specifically in liver cells (LR), pancreatic β cells (βR), or in both types of cells (LβR), hnf-1{alpha}+/– mice were first crossed with alb.cre and ins2.cre transgenic mice to obtain mice carrying an hnf-1{alpha} null gene allele along with either one or both cre transgenes. These mice were than bred with the hnf-1{alpha}kin/+ mice to produce hnf-1{alpha}kin/– mice carrying either one or both cre transgenes. The hnf-1{alpha} kin allele was reactivated in liver or pancreatic β cells of the resultant mice based on the cre transgene they carried.

Northern blot and qPCR analyses. To extract tissue RNAs, frozen mouse tissues were homogenized in TRIzol RNA reagent (GIBCO-BRL), and total RNAs were isolated according to the manufacturer's protocol. For Northern blot analysis, total RNA (20 µg) was denatured, electrophoresed, transferred to a nylon membrane, and probed with [32P]dCTP-labeled cDNA probes by a standard protocol. Major urinary protein 1 and ALS cDNAs were isolated from the EST clones, BI329730 [GenBank] and BI456799 [GenBank] , respectively. All other cDNAs used was generated and PCR amplified from the mouse liver mRNA pool, and their sequences were confirmed. For real-time quantitative PCR (qPCR) measurement, total RNAs were reverse transcribed using a high-capacity cDNA reverse transcription kit (Applied Biosystems) with random hexamers as the primers. The real-time PCR measurement of the individual cDNAs was performed in triplicate by using 25 ng cDNA and SYBRgreen dye to measure the double-stranded DNA formation with the ABI 7500 real-time PCR system. 18S rRNA was used as the internal control and the relative expression of the targeted mRNA was calculated by the comparative cycle threshold method. The qPCR primers used were listed in Table 1.


View this table:
[in this window]
[in a new window]

  		Table 1. Primers used for real-time qPCR

 
Western blot analysis. Frozen mouse tissues were homogenized in RIPA buffer. Total protein extracts (75 µg) were separated by SDS-PAGE in a 10% gel and transferred onto a nitrocellulose membrane (NitroPure, OSMONICS) for later use in antibody incubation. The antibodies for GR (sc-1004), MR (sc-11412), HSP90 (sc-7947), and HNF-1{alpha} (sc-4567) were from Santa Cruz Biotechnology, and the antibody for HNF-4{alpha} was from Geneka. The membrane was incubated with a primary antibody then a secondary IgG conjugated to horseradish peroxidase in PBS-1% nonfat dry milk-0.1% Tween 20 solution, and the signal was detected via an ECL detection system (Amersham).

GH treatment. Two-month-old mice were injected intraperitoneally with GH (25 µg/100 g body wt) or PBS after an overnight fast. Twenty minutes after the injection, the mice were euthanized and the livers were immediately removed and snap frozen. Liver proteins were analyzed for STAT5 phosphorylation by immunoprecipitation with a STAT5 antibody (sc-835, Santa Cruz) followed by separation in an 8% SDS-PAGE gel and Western blotting with a phospho-specific antibody for STAT5 (sc-1176).

ChIP. ChIP was performed as described (7). Briefly, liver samples were minced and fixed in 1% formaldehyde at 22°C for 10 min. The fixation was stopped by adding glycine to a final concentration of 0.125 M, and the liver pieces were rinsed in cold PBS before being homogenized in a Dounce homogenizer in ChIP cell lysis buffer (10 mM Tris·HCl, pH 8.0, 10 mM NaCl, 3 mM MgCl2, 0.5% NP-40, 0.5 mM phenylmethylsulfonyl fluoride, and 100 ng/ml of leupeptin). The homogenate was incubated on ice for 5 min, and nuclei were then collected by centrifugation and resuspended in ChIP nuclear lysis buffer [1% sodium dodecyl sulfate, 5 mM EDTA, 50 mM Tris·HCl, 50 mM Tris·HCl (pH 8.1), 0.5 mM phenylmethylsulfonyl fluoride, and 100 ng/ml of leupeptin] and incubated on ice for 10 min. The nuclei samples were then sonicated on ice to obtain DNA fragments of an average length of ~500 to 1,000 bp. The chromatin solution was precleared by use of protein A agarose-salmon sperm DNA (Upstate) before being incubated with 4 µg of affinity-purified anti-HNF-1{alpha} polyclonal antibody (sc-8986, Santa Cruz) or preimmune rabbit serum at 4°C with rotation for 12–16 h. The captured chromatin was precipitated, washed, and then eluted in 100 µl of freshly made ChIP elution buffer (1% SDS, 0.1 M NHCO3). The DNA in the eluted chromatin was then un-cross-linked and treated with proteinase K before being purified with a Qiagen DNA purifying column. For PCR analysis of the purified ChIP DNA in the GR gene region, 20 pairs of oligonucleotide primers were synthesized based on the 4.7-kb sequences upstream of the exon 2 (GeneBank X66367 [GenBank] ). Each pair of primers was designed to amplify a separate 250- to 400-bp area of the 5-kb region with at least a 50-bp area overlapping the sequences amplified by the adjacent upstream or downstream primer pairs. The PCR primers used for the phenylalanine hydroxylase promoter were described previously (17).

Promoter-SEAP reporter constructs. An SEAP reporter system (Clontech) utilizing a secreted form of human placental alkaline phosphatase as a reporter molecule was used here to construct the GR promoter-SEAP reporter plasmids. A 577-bp DNA fragment containing the sequence from –206 to +370 (+1 = start codon; 25) was PCR amplified from the mouse genomic BAC clone (RP23 266E9) and subcloned into pTAL-SEAP plasmid (Clontech) and pSEAP (TATA box-less derived from pTAL-SEAP). A series of pTAL-SEAP plasmids containing various deletion or mutation of this 577-bp fragment were similarly constructed (depicted in Fig. 8A). The sequences of all the plasmid constructs were confirmed before used for the transient expression assay.


Figure 8
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 8. The HNF-1{alpha}-binding sequence on the GR gene is functional in vivo. A: GR promoter activity assay in HepG2 cells. B: GR promoter activity assay in different cell types. A series of GR promoter-SEAP reporters (depicted at left) were cotransfected with pCMV.HNF-1 and pCH110 into HepG2 and/or CV-1 cells. The SEAP activity for each sample was normalized to galactosidase to control for the transfection efficiency. Histograms are expressed as means ± SE (n = 3) for the emitted signal measured by a luminometer. The X-ray film exposures of the signal captured are displayed beside the histograms. {circ}, SEAP reporter without a minimal TATA box sequence (pSEAP). bullet, SEAP reporter carrying a TATA box-like sequence (pTAL.SEAP, Clontech). The HNF-1{alpha}-binding sequence is boxed, and the translation start codon is underlined. Hatched box, the mutated HNF-1{alpha}-binding sequence based on the Gr-m oligonucleotide sequence listed in Fig. 7B.

 
Transient cotransfection and promoter activity assay. HepG2 and CV-1 cells were transfected essentially as described elsewhere (10). Cells were grown to 50% confluence in 60-mm cell culture dishes. Three micrograms of promoter-SEAP plasmid DNA, 1 µg of internal standard, the β-galactosidase reporter pCH110, and 1 µg of expression vector DNA for HNF-1{alpha}, pCMV.HNF-1{alpha} (14), were mixed with 200 µl of 0.25 M CaCl2, and 200 µl of 50 mM HEPES, 280 mM NaCl, 1.75 mM NaH2PO4, pH 7.1, was added dropwise. The mixture was added to the cells after 15 min of incubation. Cell media were changed 16 h after transfection, and both the cells and media were harvested 72 h thereafter. Twenty-five microliters of the harvested media were then used directly for SEAP activity assay. Cells were lysed and the cell lysates were assayed directly for β-galactosidase activity. The SEAP- and β-galactosidase-initiated light signals were measured in a microplate luminometer (model TR717; PE Biosystems) followed by exposure to X-ray film. The relative SEAP activity was calculated based on the activity of β-galactosidase internal standard.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
We generated a series of HNF-1{alpha}-reexpressing mutant mice by breeding various cre transgenic mice with the previously reported HNF-1{alpha} mutant strain in which the Hnf-1{alpha} gene expression was inactivated by knock-in of a floxed neo gene into its gene locus to disrupt transcription (12, 20). Since the inactivation of Hnf-1{alpha} gene can be reversed in the presence of cre recombinase, each HNF-1{alpha} mutant can reexpress HNF-1{alpha} in the tissue on the basis of the cre transgene promoter that it carries (Table 2). The HNF-1{alpha} reexpression patterns in these mutants were confirmed by immunohistological and Western blotting analyses (Fig. 1A; Fig. 2).


View this table:
[in this window]
[in a new window]

  		Table 2. HNF-1{alpha} mutant mice

 

Figure 1
View larger version (75K):
[in this window]
[in a new window]

  		Fig. 1. Normal growth depends on the hepatic hepatocyte nuclear factor-1{alpha} (HNF-1{alpha}) expression. A: immunohistological analysis of the HNF-1{alpha} reexpression in livers and pancreas islet β cells of 4-wk-old HNF-1{alpha} mutant mice on paraffin sections. The sections were counterstained with hematoxylin and eosin (H&E). Con, control; KO, HNF-1{alpha} complete knockout; LR, HNF-1{alpha} reexpression in liver cells; βR, HNF-1{alpha} reexpression in pancreas β cells; LR, HNF-1{alpha} reexpression in liver. Arrowheads indicate the representative nuclei on both immuno- and H&E-staining micrographs. B: growth curves of various HNF-1{alpha} mutant strains. Values presented are the means ± SE (error bars) for data from 5–7 mice. P values for the level of significance for differences between control/LR/LβR and KO/βR mice at the age of 12 wk old are all <0.001 in either sex.

 

Figure 2
View larger version (39K):
[in this window]
[in a new window]

  		Fig. 2. Liver glucocorticoid receptor (GR) protein levels are reduced by HNF-1{alpha} deficiency. Western blot analysis of representative liver samples of various HNF-1{alpha} mutant mice for the levels of different proteins. Each lane contains 75 µg of liver proteins from an individual animal. Bottom: Coomassie blue staining profile.

 
The HNF-1{alpha}-null mice (KO) are small and diabetic (11). Reexpression of HNF-1{alpha} in liver (LR) is able to reverse the abnormalities in liver glycogen storage and postnatal growth but does not improve their diabetic status (12). As expected, reexpression of HNF-1{alpha} in pancreatic β cells (βR) did not reverse the liver defects, and βR mice developed defects in growth similar to those in KO mice (Fig. 1B). Reexpression of HNF-1{alpha} in both liver and β cells (LβR) was again able to reverse the growth defects (Fig. 1B). On the other hand, despite their blood insulin levels being restored to the levels comparable to those in control mice, both βR and LβR mice still developed varied degrees of hyperglycemia, as did KO and LR mice (Table 2).

The postnatal growth and development in these HNF-1{alpha} reexpression mutants indicated that the HNF-1{alpha}-regulated growth depends only on the expression of HNF-1{alpha} in the liver, whereas the HNF-1{alpha}-regulated glucose homeostasis may require the expression of HNF-1{alpha} in multiple tissues. The hepatic HNF-1{alpha}-regulated growth appears to couple with the GHR signaling pathway, because KO mice exhibit characteristics of GH resistance (11). We examined the expressions of GH responsive genes involved in postnatal growth (NCBI GEO profile search; 22) in the livers of HNF-1{alpha} KO mice and found that most of the GH responsive genes examined were markedly downregulated (Figs. 3 and 4). HNF-1{alpha} is a transcriptional activator, and it might mediate GH action by directly transactivating each GH responsive genes. Indeed, HNF-1{alpha} has been found to interact with promoters of several GH-responsive genes, such as phenylalanine hydroxylase (PAH) and glucose-6-phosphate transporter 1 (G6PT1; Slc37a4) (16). However, it remains unclear how HNF-1{alpha} is involved in the GH action. STAT5 is known to mediate GH signaling in the expression of numerous hepatic genes such as IGF-I; in response to the GH signal, STAT5 is phosphorylated, dimerized, and translocated to the nucleus to transactivate the GH responsive genes (29). However, the expression of STAT5 transcripts appeared to be unaffected in the HNF-1{alpha} KO liver (Fig. 5A). In addition, in the HNF-1{alpha} KO livers, the degree of STAT5 phosphorylation in response to GH stimulation was comparable to that in control liver (Fig. 5B), suggesting that the growth defect in the HNF-1{alpha} KO mice is not due to a reduction in STAT5 expression or activation. Unlike STAT5, there is no molecular link suggesting that HNF-1{alpha} can act as a mediator for the GHR signal pathway, since GH did not affect the HNF-1{alpha} expression in the liver (5) and there is currently no evidence to indicate that GH regulates the transactivation activity of HNF-1{alpha} by modifying HNF-1{alpha} activity. It has been found, however, that HNF-1{alpha} acts cooperatively with STAT5 to transactivate the salmon Igf-1 promoter activity (15). It is therefore possible that HNF-1{alpha} acts in cooperation with STAT5 to transactivate GH-responsive genes.


Figure 3
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 3. HNF-1{alpha} is required for the expression of growth hormone (GH)-responsive genes in the liver. Northern blot analyses of expression of GH-responsive and non-GH-responsive genes. Total liver RNAs of the 6-wk-old mice were probed with the indicated cDNA probes. Each lane contains 20 µg RNA from an individual animal. Igfals, insulin-like growth factor binding protein, acid-labile subunit; Slc37a4, solute carrier family 37, member 4 (known as glucose-6-phosphate transporter 1); Crp, C-reactive protein; Apcs, amyloid P component, serum; Pah, phenylalanine hydroxylase; Hsd11β1, 11-β-hydroxysteroid dehydrogenase 1; Mbl2, mannose binding lectin 2; IL1rap, IL-1 receptor accessory protein; Slco1b2, solute carrier organic anion transporter family, member 1b2; Rpl35a, ribosomal protein L35a; MUP1, major urinary protein 1; TMEM49, transmembrane protein 49; FMO5, flavin containing monooxygenase 5; UGT2B1, UDP glucuronosyltransferase 2 family, polypeptide B1; O, regulated by activated GR according to NCBI GEO profiling; V, regulated by GR or STAT5 according to Ref. 6; X, promoter bound activated GR according to Ref. 18; Y, promoter bound HNF-1{alpha} according to Ref. 16.

 

Figure 4
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 4. HNF-1{alpha} is required for the expression of GR-dependent genes in livers. Quantitative PCR (qPCR) analyses of expression of GH-responsive and non-GH-responsive genes. The GH dependency was determined according to the NCBI GEO profile. Total liver RNAs were prepared from various 6-wk-old HNF-1{alpha} mutant mice (n = 4 for each group). The mRNA levels for each gene were quantified using the real-time qPCR method and their relative expression levels were calculated by comparison to that in the control mice. Histograms are expressed as means ± SE. Values above the histograms are P values indicating the level of significance for differences compared with the control mice. IGF1, insulin-like growth factor 1; IGFALS, insulin-like growth factor binding protein, acid labile subunit; EGFR, epidermal growth factor receptor; SOCS2, suppressor of cytokine signaling 2; STAT1, signal transducer and activator of transcription 1; CYP7B1, cytochrome P450, family 7, subfamily b, polypeptide 1; PROM1, prominin 1; SAA2, serum amyloid A 2; GSTM6, glutathione S-transferase, mu 6; ZFP145, zinc finger protein 145; GAS1, growth arrest specific 1; HAO3, hydroxyacid oxidase (glycolate oxidase) 3; O, regulated by activated GR according to NCBI GEO profiling; V, regulated by GR or STAT5 according to Ref. 6; X, promoter bound activated GR according to Ref. 18; Y, promoter bound HNF-1{alpha} according to Ref. 16.

 

Figure 5
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 5. HNF-1{alpha} is required for the normal expression of GR but not STAT5 in livers. A: qPCR analysis of mRNA expression levels. Total liver RNAs were prepared from various 6-wk-old HNF-1{alpha} mutant mice (n = 3–5 for each group). The mRNA levels for each gene were quantified by the real-time qPCR method and their relative expression levels were calculated by comparison to that in the control mice. Histograms are expressed as means ± SE. Values above histograms are P values indicating the level of significance for differences compared with the control mice. B: Western blot analysis of levels of liver STAT5 and the phosphorylated from of STAT5 (p-STAT5). Total liver proteins from 2-mo-old mice receiving GH treatment (25 µg/100 g body wt) were first immunoprecipitated with an antibody against STAT5 and than probed with the antibodies against STAT5 and the phosphorylated form of STAT5.

 
Recently, in mouse livers, STAT5 was found to cooperate with GR to mediate GH regulation on the expression of a large set of GH responsive genes including Igf-1 and its binding proteins necessary for postnatal growth (6, 27). The expressions of this group of genes (dependent on both GR and STAT5) examined were also affected in the HNF-1{alpha} KO liver (Figs. 3 and 4). Interestingly, HNF-1{alpha} deficiency did not alter the expression of those GH responsive genes, such as prolactin receptor and elastin, whose expression depends on STAT5 but not on the cooperation of GR and STAT5 (6; Fig. 4). On the other hand, HNF-1{alpha} deficiency greatly reduced the expression of some non-GH-responsive genes (6), such as MUP1 and ZFP145, whose expression depends on GR or the activated GR (Figs. 3 and 4). The increased expression of STAT1 and SAA2, though not statistically significant, is possibly due to the effect of HNF-4 whose protein levels were increased in the HNF-1{alpha} deficiency livers (Fig. 2). HNF-4 was found to bind the promoters of STAT1 and SAA2 genes (16). These results indicate there is an association between HNF-1{alpha} and GR on the GH signaling and raise the possibility that GR is involved in the expression of HNF-1{alpha}-regulated GH-responsive genes.

We analyzed the expression patterns of GR and STAT5 transcription factors in the livers of various HNF-1{alpha} mutant mice. Quantitative PCR showed that the mRNA level of GR was significantly reduced in livers that lack HNF-1{alpha}, whereas the STAT5 mRNA levels remained unchanged, indicating that the normal expression of the hepatic GR mRNA is dependent on HNF-1{alpha} (Fig. 5A). It has been reported that diabetic state and insulin levels can affect GR expression (19, 26, 33). Different HNF-1{alpha} mutant mouse strains with varying physiological states, such as hyperglycemia and hypoinsulinemia due to the inactivation of the Hnf-1{alpha} gene in different tissues (Table 2), were then used to examine the hepatic GR expression pattern. Both LR and LβR mice showed similar body sizes or growth curves to control mice, but LR mice were severely diabetic and had significantly reduced blood insulin levels, whereas LβR mice initially had normal blood glucose levels but later developed mild hyperglycemia. We found that the hepatic GR protein level was not affected in either the LR or the LβR livers where the HNF-1{alpha} gene was reexpressed (Fig. 2). On the other hand, hepatic GR protein level was markedly reduced in the livers of KO and βR mice in which the HNF-1{alpha} gene remains inactivated in livers. By contrast, the protein levels of mineralocorticoid receptor (MR) and heat shock protein 90 (Hsp90) were increased by varying degrees in the livers of all HNF-1{alpha} mutant mice (Fig. 2). MR is another nuclear receptor that shares considerable structural and functional homology with GR, whereas Hsp90 is a chaperone involved in protecting the ligand-free form of GR from degradation (1, 21). As with GR, the expression of both MR and Hsp90 is known to respond to physiological stress, such as diabetes. Thus their increased expression in the livers of all HNF-1{alpha} mutant mice is likely to be associated with the hyperglycemic state developed in all HNF-1{alpha} mutant mice. To see whether GR expression is specifically affected by HNF-1{alpha} deficiency in the liver, GR protein levels were examined in different tissues of various HNF-1{alpha} mutant mice. Indeed, GR protein was reduced by HNF-1{alpha} deficiency only in the liver, suggesting that the regulatory effect of HNF-1{alpha} on GR expression is liver specific (Fig. 6). Taken together, these results indicate that the normal expression of hepatic GR depends directly and largely on HNF-1{alpha} and is less affected by diabetic state, insulin level, or other abnormalities in HNF-1{alpha} mutant mice.


Figure 6
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 6. GR protein levels are affected by HNF-1{alpha} deficiency specifically in livers. Western blot analyses of GR and Hsp90 protein levels in different tissues of the various HNF-1{alpha} mutant mice. Total proteins (75 µg) from each tissue of 7-mo-old mice were used. Bottom panel in each tissue blot is the Coomassie blue staining profile to indicate equal loading of protein sample. IB, immunoblotting.

 
Our results suggest that hepatic GR expression is directly under HNF-1{alpha} regulation. HNF-1{alpha} is a transcription factor and likely activates the GR gene by interacting with a DNA sequence in the GR promoter/enhancer region. The GR gene structure is highly complex with multiple promoter and exon 1 areas resulting in several GR mRNA species, each with a unique 5'-untranslated region (34). The exact number of promoter that contributes to the GR gene activity is still unclear. In rats, the GR gene may contain up to 11 different promoter/exon 1 regions that span an area of more than 15 kb upstream of exon 2 where the start codon is located (13). Also, it has not been documented whether the reported GR gene promoter regions contain any functional HNF-1{alpha}-binding sequence (34). Therefore, we first used ChIP to see whether HNF-1{alpha} physically interacts with the reported 4.7-kb GR promoter region (X66367 [GenBank] ) (25). The liver DNA fragments captured by the anti-HNF-1{alpha} antibody were then scanned for sequence matches to any of the 4.7 kb area using 20 pairs of primer, each primer pair designed to amplify a separate 250- to 400-bp area. Except for the fragment P1 that contains sequences extending into the 5' part of exon 2, no sequences along the reported promoter/exon 1 regions were amplified (Fig. 7A). The sequences in the P1 fragment contained one potential HNF-1{alpha}-binding sequence, located just 3 bp upstream of the translation start codon ATG. The sequence of this potential HNF-1{alpha}-binding site, 5'-GTTAATATTTGC, is 88% identical to that of the consensus HNF-1{alpha}-binding site, 5'-GTTAATNATTAAC. Although this sequence appears cryptic compared with the consensus sequence, it formed a complex with liver nuclear protein in vitro (Fig. 7B, left), which was readily dissociated in the presence of the consensus HNF-1{alpha}-binding sequence, indicating that this sequence can bind HNF-1{alpha}. To see whether this potential sequence is functional in vivo, a 577-bp DNA fragment containing the sequence from to –206 to +370 (+1 = start codon) that also covers the sequence in the P1 fragment was used to generate a set of SEAP promoter reporter constructs and a series of deletion constructs around the cryptic HNF-1{alpha}-binding site as depicted (Fig. 8A). In the SEAP activity assay, this 577-bp DNA fragment did not affect the basal SEAP activity. However, when the HNF-1{alpha} expression vector was cotransfected, it doubled the activity of the SEAP reporter containing a TATA box sequence but did not increase that of the SEAP reporter without the TATA box sequence (Fig. 8A, top and middle). Moreover, when this cryptic HNF-1{alpha}-binding site was deleted or mutated to abolish the complex formation with nuclear protein (Fig. 7B, right), cotransfection of the HNF-1{alpha} expression vector no longer enhanced the SEAP reporter activity (Fig. 8A, bottom). In addition, cotransfection of the HNF-1{alpha} expression vector in monkey kidney CV-1 cells did not increase the SEAP reporter activity as did in HepG2 cells (Fig. 8B). These results strongly suggest that this cryptic HNF-1{alpha}-binding site located just upstream of the translation start site is functional, and it may mediate the regulatory effect of HNF-1{alpha} on GR gene activity specifically in livers.


Figure 7
View larger version (24K):
[in this window]
[in a new window]

  		Fig. 7. An HNF-1{alpha}-binding sequence is located 3 bp upstream of the translation start site on the GR gene locus. A: chromatin immunoprecipitation (ChIP) and PCR analyses of the 4.7-kb GR gene promoter region (GeneBank Accession no. X66367). The purified ChIP DNA by HNF-1{alpha} antibody was analyzed by PCR using the primer pairs for the phenylalanine hydroxylase and GR gene promoters. Twenty pairs of primers for the GR gene were synthesized on the basis of the sequence (GeneBank X66367 [GenBank] ), and each was used to amplify a region in X66367 [GenBank] as depicted. The locations of exon 1 and exon 2 (E1 and E2) are based on Ref. 25. The sequences displayed are the 3' of intron and 5' of exon 2, and the potential HNF-1{alpha}-binding sequence and translation start codon are boxed and underlined, respectively. Only the PCR results from the first 3 pairs of primers are shown. B: gel mobility shift analysis of the potential HNF-1{alpha}-binding sequence on the GR gene. The synthetic oligonucleotide "Gr" based on the sequence boxed in A was 32P-labeled and used as the probe for assaying formation of DNA-protein complex with liver nuclear protein. The unlabeled synthetic "H1" oligonucleotide based on the consensus HNF-1{alpha}-binding sequence was used in the assay to interfere with the complex formation by the labeled probe. NE, nuclear extract. Arrowheads indicate DNA-protein complex.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
By analyzing various HNF-1{alpha} mutant mice, we uncovered a mechanism possibly involved in the HNF-1{alpha}-regulated, GH-dependent postnatal growth and development. HNF-1{alpha} directly transactivates the GR gene in the liver through a cryptic regulatory element located just upstream of the translation start site. By directly regulating GR expression, HNF-1{alpha} may efficiently control the expression of a large set of hepatic GH-responsive genes that depend on the cooperation between GR and STAT5, without having to interact individually with their promoter/enhancer regions.

Our results, however, do not exclude the possibilities that HNF-1{alpha} by itself directly modulate the promoter activity of each of those GH-responsive genes affected in the HNF-1{alpha} deficient liver, since HNF-1{alpha} has also been found to associate with the promoters of several GH-responsive genes like G6PT1 and CRP (16). Likewise, HNF-1{alpha} may also regulate the expression of GH-responsive genes in cooperation with STAT5 as reported for salmon IGF-1 gene regulation (15). Regardless, the finding that HNF-1{alpha} is required for the normal expression of GR in liver not only further strengthens the crucial role of HNF-1{alpha} in the postnatal growth but also suggests a regulatory role of HNF-1{alpha} in the hepatic expression of glucocorticoid-responsive and GR-dependent genes involved in may aspects of physiological functions (18). For example, similar to that in the STAT5-deficient liver (3), the expression of peroxisome proliferators-activated receptor-{gamma} (PPAR{gamma}) was also increased, although not as significantly, in the GR-deficient liver (6). It was considered that the elevated expression of the hepatic PPAR{gamma} in combination with low IGF levels had led to fatty liver development in the STAT5-deficient liver (3). HNF-1{alpha} null mice also develop fatty liver and express more PPAR{gamma} than control mice (11, 24), which could be reversed upon reexpression of HNF-1{alpha} specifically in the liver (12). It is thus possible that the fatty liver development in HNF-1{alpha}-deficient liver is due in part to the decrease of GR expression, which in turn reduces GR-STAT5 cooperation and leads to an enhancement in PPAR{gamma} expression.

Considering its regulatory role on the expressions of genes involved in growth, immune system, and metabolism (18), GR is likely to be controlled tightly at the levels of both expression and activity. The GR gene possesses multiple exon 1 and promoters, and it produces several mRNA species (34). The regulation of GR expression at the transcription level is expected to be highly complex and finely tuned to reflect its complicated gene structure. HNF-1{alpha} has not been reported to be required for GR activity or to be involved in the transcriptional regulation of GR gene expression. Our results indicate clearly that, in the liver, the normal expression of GR mRNA is dependent on HNF-1{alpha} and that a functional, though cryptic, HNF-1{alpha}-binding sequence is located in the GR gene locus. Interestingly, the HNF-1{alpha}-binding sequence identified here is not located in any of the prospective promoter regions upstream of the respective exon 1s. Instead, this cryptic binding site is located just 3 bp upstream of the translation start site in exon 2. Nevertheless, in the in vivo functional assay, this cryptic HNF-1{alpha}-binding site was capable of increasing the activity of the promoter reporter twofold, despite the fact that it is unclear at present whether this HNF-1{alpha}-binding site located in the exon 2 region can enhance transcriptional activity driven by any of the promoters or selectively by some promoters on the GR gene locus. The mouse GR gene spans an area of more than 80 kb in chromosome 18, and it is possible that the gene locus contains additional functional regulatory sequences for HNF-1{alpha} to efficiently control its activity. Nevertheless, our result has established a regulatory role of HNF-1{alpha} in controlling GR gene expression via a functional interaction with its binding sequence on the gene.

The GH-regulated gene expression and postnatal growth appear less straightforward than previously proposed. Recent progress has revealed that the original somatomedin hypothesis is inadequate to explain the GH action on growth and that STAT5 is not the sole signal factor required to transactivate those genes in response to GH stimulation (6, 9, 27, 30). Our finding that HNF-1{alpha} controls the normal expression of GR further indicates the complexity of gene regulation that has evolved to ensure proper responses to GH signals for growth. Thus it is tempting to speculate that HNF-1{alpha}, in addition to controlling GR levels for the expression of GH-responsive genes, might be involved in other as yet undiscovered aspects of GH-dependent growth. Interestingly, the regulation of GR expression by HNF-1{alpha} appears to be liver specific. Although the tissue-specific mechanism enabling or preventing HNF-1{alpha} to control GR expression in liver or in extrahepatic tissues remains to be elucidated, this result has reinforced the importance of the liver in modulating postnatal body growth.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
This work was supported in part by research grants from the National Health Research Institute (NHRI-EX93-9327SI) and National Science Council (NSC-94-2311-B-001-031) to Y. H. Lee.


   FOOTNOTES
 

Address for reprint requests and other correspondence: Y. H. Lee, Academia Sinica, Institute of Molecular Biology, Taipei 115, Taiwan (e-mail: yinghue{at}gate.sinica.edu.tw)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin BL, Housman DE, Evans RM. Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237: 268–275, 1987.[Abstract/Free Full Text]
  2. Baumhueter S, Mendel DB, Conley PB, Kuo CJ, Turk C, Graves MK, Edwards CA, Courtois G, Crabtree GR. HNF-1 shares three sequence motifs with the POU domain proteins and is identical to LF-B1 and APF. Genes Dev 4: 372–379, 1990.[Abstract/Free Full Text]
  3. Cui Y, Hosui A, Sun R, Chen K, Gavrilova O, Chen W, Cam MC, Gao B, Robinson GW, Hennighausen L. Loss of signal transducer and activator of transcription 5 leads to hepatosteatosis and impaired liver regeneration. Hepatology 46: 504–513, 2007.[CrossRef][Web of Science][Medline]
  4. Daughaday WH, Hall K, Raben MS, Salmon WD Jr, van den Brande JL, Van Wyk JJ. Somatomedin: a proposed designation for the "sulfation factor." Nature 235: 107, 1972.[CrossRef][Medline]
  5. Eleswarapu S, Jiang H. Growth hormone regulates the expression of hepatocyte nuclear factor-3 gamma and other liver-enriched transcription factors in the bovine liver. J Endocrinol 184: 95–105, 2005.[Abstract/Free Full Text]
  6. Engblom D, Kornfeld JW, Schwake L, Tronche F, Reimann A, Beug H, Hennighausen L, Moriggl R, Schütz G. Direct glucocorticoid receptor-Stat5 interaction in hepatocytes controls body size and maturation-related gene expression. Genes Dev 21: 1157–1162, 2007.[Abstract/Free Full Text]
  7. Fong CJ, Burgoon LD, Zacharewski RT. Comparative microarray analysis of basal gene expression in mouse Hepa-1c1c7 wild-type and mutant cell lines. Toxicol Sci 86: 342–353, 2005.[Abstract/Free Full Text]
  8. Friedman JR, Larris P, Le TH, Peiris A, Arenlis J, Schug JW, Tobias KH, Greenbaum LE. Orthogonal analysis of C/EBPbeta targets in vivo during liver proliferation. Proc Natl Acad Sci USA 101: 12986–12991, 2004.[Abstract/Free Full Text]
  9. Iida K, Del Rincon JP, Kim D-S, Itoh E, Nass R, Coschigano KT, Kopchick JJ, Thorner MO. Tissue-specific regulation of growth hormone (GH) receptor and insulin-like growth factor-I gene expression in the pituitary and liver of GH-deficient (lit/lit) mice and transgenic mice that overexpress bovine GH (bGH) or a bGH antagonist. Endocrinology 145: 1564–1570, 2004.[CrossRef][Web of Science][Medline]
  10. Isaksson OG, Jansson JO, Gause IA. Growth hormone stimulates longitudinal bone growth directly. Science 216: 1237–1239, 1982.[Abstract/Free Full Text]
  11. Klover P, Hennighausen L. Postnatal body growth is dependent on the transcription factors signal transducers and activators of transcription 5a/b in muscle: a role for autocrine/paracrine insulin-like growth factor I. Endocrinology 148: 1489–1497, 2007.[Abstract/Free Full Text]
  12. Lee YH, Yano M, Liu SY, Matsunaga E, Johnson PF, Gonzalez FJ. A novel cis-acting element controlling the rat CYP2D5 gene and requiring cooperativity between C/EBP beta and an Sp1 factor. Mol Cell Biol 14: 1383–1394, 1994.[Abstract/Free Full Text]
  13. Lee YH, Sauer B, Gonzalez FJ. Laron dwarfism and non-insulin-dependent diabetes mellitus in the Hnf-1 knockout mouse. Mol Cell Biol 18: 3059–3068, 1998.[Abstract/Free Full Text]
  14. Lee YH, Magnuson MA, Muppala V, Chen SS. Liver-specific reactivation of the inactivated Hnf-1 gene: elimination of liver dysfunction to establish a mouse MODY3 model. Mol Cell Biol 23: 923–932, 2003.[Abstract/Free Full Text]
  15. McCormick JA, Lyons V, Jacobson MD, Noble J, Diorio J, Nyirenda M, Weaver S, Ester W, Yau JLW, Meaney MJ, Seckl JR, Chapman KE. 5'-Heterogeneity of glucocorticoid receptor messenger RNA is tissue specific: differential regulation of variant transcripts by early-life events. Mol Endocrinol 14: 506–517, 2000.[Abstract/Free Full Text]
  16. Mendel DB, Hansen LP, Graves MK, Conley PB, Crabtree GR. HNF-1 alpha and HNF-1 beta (vHNF-1) share dimerization and homeodomains, but not activation domains, and form heterodimers in vitro. Genes Dev 5: 1042–1056, 1991.[Abstract/Free Full Text]
  17. Meton I, Boot EP, Sussenbach JS, Steenbergh PH. Growth hormone induces insulin-like growth factor-I gene transcription by a synergistic action of STAT5 and HNF-1. FEBS Lett 444: 155–159, 1999.[CrossRef][Web of Science][Medline]
  18. Noguez DA, Yagi K, Venable S, Darlington G. Gene expression profile of long-lived Ames dwarf mice and Little mice. Aging Cell 3: 423–441, 2004.[CrossRef][Web of Science][Medline]
  19. Odom DT, Zizlsperger N, Gordon DB, Bell GW, Rinaldi NJ, Murray HL, Volkert TL, Schreiber J, Rolfe PA, Gifford DK, Fraenkel E, Bell G, Young RY. Control of pancreas and liver gene expression by HNF transcription factors. Science 303: 1378–1381, 2004.[Abstract/Free Full Text]
  20. Parrizas M, Maestro MA, Boj SF, Paniagua A, Casamitjana R, Gomis R, Rivera F, Ferrer J. Hepatic nuclear factor 1 directs nucleosomal hyperacetylation to its tissue-specific transcriptional targets. Mol Cell Biol 21: 3234–3243, 2001.[Abstract/Free Full Text]
  21. Phuc Le P, Friedman JR, Schug J, Brestelli JE, Parker JB, Bochkis IM, Kaestner KH. Glucocorticoid receptor-dependent gene regulatory network. PloS Genet 1: 1–12, 2005.[Medline]
  22. Pierreux CE, Ursø B, De Meyts P, Rousseau GG, Lemaigre FP. Inhibition by insulin of glucocorticoid-induced gene transcription: involvement of the ligand-binding domain of the glucocorticoid receptor and independence from the phosphatidylinositol 3-kinase and mitogen-activated protein kinase pathways. Mol Endocrinol 12: 1343–1354, 1998.[Abstract/Free Full Text]
  23. Postic C, Shiota M, Niswender KD, Jetton TL, Chen Y, Moates JM, Shelton KD, Lindner J, Cherrington AD, Magnuson MA. Dual roles for glucokinase in glucose homeostasis as determined by liver and pancreatic cell-specific gene knockouts using cre recombinase. J Biol Chem 274: 305–315, 1999.[Abstract/Free Full Text]
  24. Pratt WB. The role of heat shock proteins in regulating the function, folding, and trafficking of the glucocorticoid receptor. J Biol Chem 268: 21455–21458, 1993.[Free Full Text]
  25. Rowland JE, Lichanska AM, Kerr LM, White M, d'Aniello EM, Maher SL, Brown R, Teasdale RD, Noakes PG, Waters MJ. In vivo analysis of growth hormone receptor signaling domains and their associated transcripts. Mol Cell Biol 25: 66–77, 2005.[Abstract/Free Full Text]
  26. Savage MO, Burren CP, Blair JC, Woods KA, Metherell L, Clark AJL, Camacho-Hübner C. Growth hormone insensitivity: pathophysiology, diagnosis, clinical variation and future perspectives. Horm Res 55: 32–35, 2001.[Medline]
  27. Shih DQ, Bussen M, Sehayek E, Ananthanarayanan M, Shneider BL, Suchy FJ, Shefer S, Bollileni JS, Gonzalez FJ, Breslow JL, Stoffel M. Hepatocyte nuclear factor-1{alpha} is an essential regulator of bile acid and plasma cholesterol metabolism. Nat Genet 27: 375–382.
  28. Son D-S, Roby KF, Terranova PF. Tumor necrosis factor-{alpha} induces serum amyloid A3 in mouse granulosa cells. Endocrinology 145: 2245–2252, 2004.[Abstract/Free Full Text]
  29. Strahle U, Schmidt A, Kelsey G, Stewart AF, Cole TJ, Schmid W, Schutz G. At least three promoters direct expression of the mouse glucocorticoid receptor gene. Proc Natl Acad Sci USA 89: 6731–6735, 1992.[Abstract/Free Full Text]
  30. Thompson A, Arany EJ, Hill DJ, Yang K. Glucocorticoid receptor expression is altered in pancreatic beta cells of the non-obese diabetic mouse during postnatal development. Metabolism 51: 765–768, 2002.[CrossRef][Web of Science][Medline]
  31. Tronche F, Opherk C, Morigg R, Kellendonk C, Reimann A, Schwake L, Reichardt HM, Stang K, Gau D, Hoeflich A, Beug H, Schmid W, Schütz G. Glucocorticoid receptor function in hepatocytes is essential to promote postnatal body growth. Genes Dev 18: 492–497, 2004.[Abstract/Free Full Text]
  32. Ueki I, Ooi GT, Tremblay ML, Hurst KR, Bach LA, Boisclair YR. Inactivation of the acid labile subunit gene in mice results in mild retardation of postnatal growth despite profound disruptions in the circulating insulin-like growth factor system. Proc Natl Acad Sci USA 97: 6868–6873, 2000.[Abstract/Free Full Text]
  33. Waters MJ, Hoang HN, Fairlie DP, Pelekanos RA, Brown RJ. New insights into growth hormone action. J Mol Endocrinol 36: 1–7, 2006.[Abstract/Free Full Text]
  34. Yakar S, Liu JL, Stannard B, Butler A, Accili D, Sauer B, LeRoith D. Normal growth and development in the absence of hepatic insulin-like growth factor I. Proc Natl Acad Sci USA 96: 7324–7329, 1999.[Abstract/Free Full Text]
  35. Yakar S, Rosen CJ, Beamer WG, Ackert-Bicknell CL, Wu Y, Liu JL, Ooi GT, Setser J, Frystyk J, Boisclair YR, LeRoith D. Circulating levels of IGF-1 directly regulate bone growth and density. J Clin Invest 110: 771–781, 2002.[CrossRef][Web of Science][Medline]
  36. Yamagata K, Oda N, Kaisaki PJ, Menzel S, Furuta H, Vaxillaire M, Southam L, Cox RD, Lathrop GM, Borirai VV, Chen X, Cox NJ, Oda Y, Yano H, Le Beau MM, Yamad S, Nishigori H, Takeda J, Fajans SS, Hattersley AT, Iwasaki N, Hansen T, Pedersen O, Polonsky KS, Turner RC, Velho G, Chevre JC, Froguel P, Bell GI. Mutations in the hepatocyte nuclear factor-1 gene in maturity-onset diabetes of the young (MODY3). Nature 384: 455–458, 1996.[CrossRef][Web of Science][Medline]
  37. Yourick JJ, Beuving LJ. The effects of insulin on hepatic glucocorticoid receptor content in the diabetic rat. J Recept Res 5: 381–395, 1985.[Web of Science][Medline]
  38. Yudt MR, Cidlowski JA. The glucocorticoid receptor: coding a diversity of proteins and responses through a single gene. Mol Endocrinol 16: 1719–1726, 2002.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/3/G542    most recent
00081.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lin, W.-Y.
Right arrow Articles by Lee, Y.-H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lin, W.-Y.
Right arrow Articles by Lee, Y.-H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.