Advertisement
The peptidyl-prolyl isomerase Pin1 determines parathyroid hormone mRNA levels and stability in rat models of secondary hyperparathyroidism
Morris Nechama,1 Takafumi Uchida,2 Irit Mor Yosef-Levi,1 Justin Silver,1 and Tally Naveh-Many1
1Minerva Center for Calcium and Bone Metabolism, Nephrology Services, Hadassah Hebrew University Medical Center, Jerusalem, Israel. 2Molecular Enzymology, Tohoku University, Sendai, Japan.
Address correspondence to: Tally Naveh-Many, Minerva Center for Bone and Mineral Metabolism, Nephrology Services, Hadassah University Hospital, PO Box 12000, Jerusalem 91120, Israel. Phone: 972-2-6776789; Fax: 972-2-6421234; E-mail: tally@huji.ac.il.
Find articles by Nechama, M. in: JCI | PubMed | Google Scholar
1Minerva Center for Calcium and Bone Metabolism, Nephrology Services, Hadassah Hebrew University Medical Center, Jerusalem, Israel. 2Molecular Enzymology, Tohoku University, Sendai, Japan.
Address correspondence to: Tally Naveh-Many, Minerva Center for Bone and Mineral Metabolism, Nephrology Services, Hadassah University Hospital, PO Box 12000, Jerusalem 91120, Israel. Phone: 972-2-6776789; Fax: 972-2-6421234; E-mail: tally@huji.ac.il.
Find articles by Uchida, T. in: JCI | PubMed | Google Scholar
1Minerva Center for Calcium and Bone Metabolism, Nephrology Services, Hadassah Hebrew University Medical Center, Jerusalem, Israel. 2Molecular Enzymology, Tohoku University, Sendai, Japan.
Address correspondence to: Tally Naveh-Many, Minerva Center for Bone and Mineral Metabolism, Nephrology Services, Hadassah University Hospital, PO Box 12000, Jerusalem 91120, Israel. Phone: 972-2-6776789; Fax: 972-2-6421234; E-mail: tally@huji.ac.il.
Find articles by Mor Yosef-Levi, I. in: JCI | PubMed | Google Scholar
1Minerva Center for Calcium and Bone Metabolism, Nephrology Services, Hadassah Hebrew University Medical Center, Jerusalem, Israel. 2Molecular Enzymology, Tohoku University, Sendai, Japan.
Address correspondence to: Tally Naveh-Many, Minerva Center for Bone and Mineral Metabolism, Nephrology Services, Hadassah University Hospital, PO Box 12000, Jerusalem 91120, Israel. Phone: 972-2-6776789; Fax: 972-2-6421234; E-mail: tally@huji.ac.il.
Find articles by Silver, J. in: JCI | PubMed | Google Scholar
1Minerva Center for Calcium and Bone Metabolism, Nephrology Services, Hadassah Hebrew University Medical Center, Jerusalem, Israel. 2Molecular Enzymology, Tohoku University, Sendai, Japan.
Address correspondence to: Tally Naveh-Many, Minerva Center for Bone and Mineral Metabolism, Nephrology Services, Hadassah University Hospital, PO Box 12000, Jerusalem 91120, Israel. Phone: 972-2-6776789; Fax: 972-2-6421234; E-mail: tally@huji.ac.il.
Find articles by Naveh-Many, T. in: JCI | PubMed | Google Scholar
Published September 21, 2009 - More info
J Clin Invest. 2009;119(10):3102–3114. https://doi.org/10.1172/JCI39522.
© 2009 The American Society for Clinical Investigation
Received: April 12, 2009; Accepted: July 15, 2009
Related article:
Abstract
Secondary hyperparathyroidism often occurs in chronic kidney disease (CKD) and vitamin D deficiency, resulting in increased fractures and mortality. Understanding factors that stimulate parathyroid hormone (PTH) synthesis is important for devising methods to treat this condition. Previous work has demonstrated that murine Pth mRNA levels are regulated by proteins that bind AU-rich elements (AREs) within the 3′ UTR region of Pth mRNA and influence Pth mRNA stability. In this issue of the JCI, Nechama et al. demonstrate that in murine secondary hyperparathyroidism associated with CKD or Ca deficiency, the activity of Pin1, a peptidyl-prolyl isomerase, is reduced (see the related article beginning on page 3102). Reduced Pin1 activity resulted in the phosphorylation and degradation of an ARE-binding protein, K-homology splicing regulator protein (KSRP), which normally enhances the degradation of Pth mRNA. The activity of other ARE-binding proteins, such as AU-rich binding factor 1 (AUF1), that increase Pth mRNA stability, was increased, thereby increasing PTH synthesis. This work suggests new ways by which to regulate PTH synthesis in secondary hyperparathyroidism.
Authors
Rajiv Kumar
-
Results
Pin1 activity is decreased in parathyroid extracts of rats with secondary hyperparathyroidism. We have previously reported that dietary calcium depletion and experimental CKD lead to secondary hyperparathyroidism characterized by a posttranscriptional increase in PTH mRNA levels (22). To determine whether the prolyl isomerase Pin1 is involved in this regulation, we first showed Pin1 expression in parathyroid tissue by immunohistochemistry of rat parathyroid tissue (Figure 1A). We then analyzed Pin1 enzymatic activity in parathyroid extracts from control and secondary hyperparathyroidism rats. Rats were fed either a control diet or diets that were calcium restricted or supplemented with adenine and high phosphorus to induce CKD. Both diets increase serum PTH and PTH mRNA levels compared with a control diet fed to rats, with the effect of calcium depletion being greater than that of the adenine diet (11). At the end of the diet period, Pin1 activity was measured in microdissected parathyroid cytosolic lysates by an isomerase assay (23). PPIase activity was markedly reduced in parathyroid cytoplasmic lysates of both rats fed a calcium-restricted diet and CKD rats compared with parathyroid extracts from control rats (Figure 1, B and D). There was no change in Pin1 protein levels (Figure 1, C and E). These results indicate that Pin1 activity is decreased in response to changes in serum calcium and CKD. The decrease in Pin1 activity inversely correlates with the increased PTH mRNA levels and stability in these models.
Figure 1PPIase activity is decreased in parathyroid extracts from rat with secondary hyperparathyroidism due to a calcium-depleted diet or an adenine and high-phosphorus diet. (A) Immunohistochemistry using an anti-Pin1 antibody or IgG showing specific expression of Pin1 in the parathyroid cells and not in interstitial tissue or blood vessels. Original magnification, ×200. (B and D) Pin1 isomerase assays. PPIase activity of parathyroid cytoplasmic lysates from a pool of 5 rats fed a calcium-depleted diet (B) or an adenine and high-phosphorus diet (D) compared with parathyroid extracts from control rats. The results are representative of 3 independent experiments and are presented as mean ± SEM of 3 measurements from 1 of the experiments (B) and as mean ± SEM of 3 measurements from the 3 experiments combined (D). The decrease in Pin1 activity in B and D was significant (P < 0.05) compared with the respective controls at all the time intervals studied. Control: low Ca or adenine, P < 0.05 at all time intervals. (C and E) Immunoblot analysis using anti-Pin1 or anti-GAPDH antibodies for parathyroid extracts, as in B and D.
Pin1 inhibition by juglone increases PTH mRNA in vivo. The decreased Pin1 activity in parathyroid extracts from secondary hyperparathyroidism rats suggests that Pin1 inhibition would lead to increased PTH mRNA levels. We therefore studied the effect of a specific and irreversible Pin1 inhibitor, juglone (23), on PTH gene expression in vivo. Juglone covalently inactivates Pin1 and accelerates its catabolism by the proteasome (20). Juglone was administered systemically by 2 daily i.p. injections. One hour after the second injection, there was the expected decrease in PPIase activity in the parathyroid cytoplasmic lysates (Figure 2A) and an increase in serum PTH levels (Figure 2B). To study the direct effect of Pin1 inhibition on PTH expression, we applied juglone topically on the parathyroid glands using a method that we have recently developed to study the effect of direct application of a compound to the parathyroid while maintaining the glands in their physiological milieu (24). After 2 hours of juglone or vehicle administration, treated parathyroid glands were removed and analyzed for PPIase activity, Pin1 protein levels, and PTH mRNA levels. Pin1 protein levels and PPIase activity were markedly decreased in the juglone-treated parathyroid cytoplasmic lysates (Figure 2, C and D). Importantly, juglone administration markedly increased PTH mRNA levels by both Northern blots and real-time quantitative RT-PCR (qRT-PCR) (Figure 2, E and F). Therefore, Pin1 inhibition by juglone increases PTH gene expression in vivo.
Figure 2Pin1 inhibition by juglone increases PTH mRNA in vivo posttranscriptionally. (A and B) Juglone i.p. (A) PPIase activity assays of parathyroid cytoplasmic lysates from rats injected i.p. with juglone or vehicle. (B) Serum PTH in rats as in A. (C–F) Juglone applied topically. (C) PPIase activity of parathyroid cytoplasmic lysates from rats treated with juglone or vehicle applied topically; data in A and C are presented as mean ± SEM (n = 3) from 1 of 3 repeat experiments; P < 0.05 at 5–30 seconds. (D) Immunoblots of the extracts in C for Pin1 and GAPDH. (E) Northern blots of PTH mRNA levels after juglone, in rats as in C. The lanes were run on the same gel but were noncontiguous (white line). (F) qRT-PCR analysis for PTH mRNA in rat parathyroids as in C; *P < 0.05. (G) Nuclear run-ons of PTH transcription in isolated nuclei of parathyroids from 5 rats in each group after topical juglone (J) or vehicle (C). (H) Quantification of PTH mRNA transcription in G, showing no difference in PTH transcription. The same result was obtained in a repeat experiment. (I) RNase protection analysis (RPA) of total parathyroid RNA from rats after topical juglone or vehicle. Left panels: RPA using an antisense probe for the PTH mRNA 3′ UTR or for control GAPDH mRNA. Right panel: RPA using an antisense probe for the first intron (newly transcribed pre–PTH mRNA). The intact probes before and after RNase treatment (arrows to the left and right of the gels, respectively) and hybridization with tRNA as a nonspecific control are shown. Juglone increased PTH mRNA but not the pre–PTH mRNA or GAPDH mRNA.
Pin1 inhibition increases PTH mRNA levels by a posttranscriptional mechanism. To study the mechanism of the effect of Pin1 inhibition by juglone on PTH gene expression, we performed nuclear run-on assays with isolated nuclei from parathyroid glands of rats treated topically with juglone or control rats, as above (Figure 2, C–F). PTH transcription was the same in treated and nontreated parathyroid glands, indicating that the increase in PTH mRNA after Pin1 inhibition by juglone is posttranscriptional (Figure 2, G and H). To further demonstrate the posttranscriptional effect of juglone on PTH gene expression, we performed RNase protection assays using antisense probes for the PTH mRNA 3′ UTR (exon) or for the first intron of PTH mRNA as a marker of newly transcribed pre–PTH mRNA. As shown in Figure 2I, juglone-treated parathyroid glands had higher levels of PTH mRNA compared with control parathyroids based on the RNase protection assay using the PTH mRNA 3′ UTR probe, similar to the results obtained by Northern blots and qRT-PCR (Figure 2, E and F). A probe for GAPDH mRNA 3′ UTR was used as a control that was not affected by juglone (Figure 2I). In contrast, there was no difference in the pre-mRNA levels for the PTH transcript in juglone-treated parathyroid glands compared with controls, as indicated using the probe for the first intron (Figure 2I). Together, these results indicate that the increased PTH mRNA levels after Pin1 inhibition by juglone is posttranscriptional, similar to the regulation of PTH gene expression in secondary hyperparathyroidism in which Pin activity is decreased (Figure 1, B and D). Thus, the physiological function of Pin1 may be to decrease PTH gene expression posttranscriptionally.
Pin1 regulates PTH mRNA stability in cotransfected HEK293 cells, and this is mediated by the PTH mRNA 3′ UTR ARE. Since a parathyroid cell line is not available, the direct effect of Pin1 on PTH mRNA levels was studied in transiently transfected HEK293 cells. Pin1 overexpression decreased PTH mRNA levels with no effect on endogenous L32 or control cotransfected growth hormone (GH) mRNA levels (Figure 3, A–C). The effect of Pin1 overexpression on PTH mRNA stability was measured by in vitro degradation assays (IVDAs) using transfected cell extracts and polyadenylated full-length radio-labeled PTH mRNA transcripts (11). Extracts from Pin1-overexpressing cells led to accelerated PTH mRNA decay in vitro compared with extracts from control transfected cell extracts (Figure 3, D and E). The t1/2 of PTH mRNA decreases from more than 100 minutes with the control transfected extracts to 50 minutes by the Pin1-overexpressing extracts. Pin1 overexpression, however, had no effect on the in vitro decay of a truncated polyadenylated PTH mRNA transcript lacking the PTH mRNA 3′ UTR ARE (Figure 3, D and E), suggesting an ARE-dependent effect of Pin1 to increase PTH mRNA decay.
Figure 3Pin1 regulates PTH mRNA steady-state levels and stability in transfected cells. (A–F) Pin1 overexpression. HEK293 cells were transiently cotransfected with expression plasmids for human PTH or GH and for GFP-Pin1 or control pcDNA3 plasmids. (A) Immunoblots of protein extracts at 48 hours for Pin1 and α-tubulin. (B) Northern blots for PTH mRNA. (C) Northern blots for GH mRNA. (D) IVDA for polyadenylated full-length rat PTH mRNA (top) or polyadenylated PTH mRNA without the ARE (bottom) and extracts from cells transfected with GFP-Pin1 or control plasmids. (E) Quantification of the IVDAs in D and 2 additional IVDAs; mean ± SEM of intact transcript; *P < 0.05. (F and G) Cells were transiently cotransfected with expression plasmids for PTH and GFP-Pin1 or Pin1 mutants at the WW domain (W34A) or at the PPIase domain (K63A,R68/69A). (F) Northern blots for PTH mRNA. (G) Quantification of 3 Northern blots; mean ± SEM; *P < 0.05. (H–L) Pin1 knockdown. Cells were cotransfected with a human PTH plasmid and either Pin1 or control CAT siRNAs. (H) Immunoblots of Pin1 and α-tubulin after transfection. (I) Northern blots for PTH mRNA. (J) Quantification by qRT-PCR analysis for PTH mRNA after Pin1 siRNA. (K) IVDA for full-length rat PTH mRNA (top) or rat PTH mRNA without the ARE (bottom) and extracts from cells with either CAT or Pin1 siRNAs. (L) Quantification of 3 IVDAs; mean ± SEM of intact transcript at time 0; *P < 0.05.
Both the binding (WW) and isomerase (PPIase) activities of Pin1 are normally required for Pin1 to regulate the function of its substrates (16, 25). To determine whether one or both of these activities are required for the decrease in PTH mRNA levels by Pin1, we cotransfected the PTH expression plasmid with expression plasmids for mutated Pin1 at either the binding domain (W34A) or the PPIase domain (R68/69A, K63A) (19). Pin1 overexpression decreased PTH mRNA levels as before (Figure 3B), as did the K63A mutant (Figure 3, F and G). However, mutants for both the binding domain (W34A) and another PPIase domain (R68/69A) attenuated the downregulation of PTH mRNA induced by WT Pin1 (Figure 3, F and G). Therefore, Pin1 overexpression decreases PTH mRNA levels, and both the WW and PPIase domains are necessary for this effect.
Next, we decreased Pin1 protein levels using siRNAs and assessed the resulting effects on cotransfected PTH mRNA levels. A 75% reduction in Pin1 protein levels (Figure 3H) led to a 3-fold increase in PTH mRNA levels by both Northern blot and qRT-PCR (Figure 3, I and J). Similarly, PTH mRNA t1/2 was prolonged in IVDAs with extracts from Pin1-depleted cells compared with extracts from cells transfected with a control bacterial chloramphenicol acetyltransferase (CAT) siRNA (Figure 3, K and L). The t1/2 of PTH mRNA increased from 50 minutes by control extracts to more than 100 minutes by the Pin1 knockdown extracts. As with Pin1 overexpression (Figure 3, D and E), the t1/2 of a PTH transcript lacking the 3′ UTR ARE was not affected by Pin1 depletion (Figure 3, K and L). Together, the Pin1 knockdown and overexpression experiments indicate that Pin1 specifically decreases steady-state PTH mRNA levels and PTH mRNA stability by IVDA and this is dependent upon the PTH mRNA ARE.
To determine whether the regulation of PTH mRNA levels by Pin1 is mediated by the PTH mRNA ARE in the transfected cells, we used a GH reporter gene containing the rat PTH 63-nt ARE within its 3′ UTR (GH63) (Figure 4A) (11). As in our previous reports, the PTH mRNA ARE decreased GH63 mRNA levels compared with WT GH mRNA (Figure 4B) (4). Pin1 overexpression led to a further decrease in chimeric GH63 mRNA levels but had no effect on WT GH mRNA levels (Figure 4, B and C). We then studied the effect of the Pin1 inhibitor juglone on GH63 or WT GH. Juglone decreased endogenous Pin1 protein levels in HEK293 cells, as expected (Figure 4D). Importantly, juglone increased GH63 mRNA levels compared with control cells treated with vehicle but had no effect on WT GH mRNA levels (Figure 4, E and F). Together, our results indicate that Pin1 decreases steady-state PTH mRNA levels, through the PTH mRNA ARE.
Figure 4Pin1 regulates PTH mRNA levels through the PTH mRNA ARE. (A) Schematic representation of mRNAs for PTH, GH, and chimeric reporter GH containing the PTH mRNA ARE (GH63) used for transfections. (B and C) Pin1 overexpression. HEK293 cells were transiently cotransfected with expression plasmids for either GH or a chimeric GH63 (GH63) and GFP-Pin1 or pcDNA3 (control). (B) Northern blot analysis for PTH mRNA. (C) Quantification of 2 Northern blots as in B for GH (left) and GH63 (right); mean ± SEM; *P < 0.05 compared with GH63 without Pin1. (D–F) Effect of juglone on GH63. Cells were transiently transfected with expression plasmids for either GH or GH63 and incubated with juglone or vehicle (–). (D) Immunoblots for Pin1 in cell extracts after juglone or vehicle. (E) Northern blots for GH mRNA. (F) Quantification of Northern blots of 3 independent experiments as in E (*P < 0.05). The lanes in B and E were run on the same gel but were noncontiguous (white lines).
Pin1 interacts with the PTH mRNA decay–promoting protein KSRP, and the effects of KSRP and Pin1 on PTH mRNA are interdependent. PTH mRNA stability is regulated by the coordinated interaction of AUF1 and KSRP with the same PTH mRNA 3′ UTR 63-nt ARE (11). AUF1 interacts with Pin1, resulting in AUF1 isomerization and hyperphosphorylation upon Pin1 activation (20). KSRP-Pin1 interaction has not been reported. To study such interaction, we performed coimmunoprecipitation analysis of FLAG-KSRP–transfected HEK293 cell extracts. Endogenous Pin1 was specifically immunoprecipitated by FLAG-KSRP (Figure 5A). Similarly, Pin1 immunoprecipitated FLAG-KSRP in these extracts (Figure 5A).
Figure 5Pin1 interacts with the PTH mRNA decay–promoting protein KSRP, and the effects of Pin1 and KSRP on the PTH mRNA ARE are interdependent. (A) Pin1-KSRP interaction. HEK293 cells were transiently transfected with FLAG-KSRP plasmid and extracts immunoprecipitated with anti-FLAG or anti-Pin1 antibodies or IgG, run on SDS PAGE, and immunoblotted with the corresponding antibodies. (B and C) Cells were cotransfected with either GH or GH63 and KSRP or control pcDNA3 plasmids. After 48 hours, increasing concentrations of juglone were added for an additional 4 hours. (B) Northern blots for GH mRNA. (C) Quantification of 3 Northern blots as in B; *P < 0.05 compared with GH mRNA without KSRP; **P < 0.05 compared with GH63 mRNA without KSRP. (D–F) KSRP knockdown. Cells were cotransfected with either GH or GH63 and GFP-Pin1 or control pcDNA3 plasmids together with either CAT or KSRP siRNAs. (D) Immunoblots for KSRP and α-tubulin protein levels 48 hours after transfection, showing decreased KSRP in KSRP siRNA–transfected cells. Densitometry analysis of KSRP corrected for α-tubulin is shown below the gels. (E) Northern blots for GH mRNAs. The lanes were run on the same gel but were noncontiguous (white line). (F) Quantification of 3 Northern blots as in E for GH (left) and GH63 (right); *P < 0.05. Data in C and F are represented as mean ± SEM.
To determine the functional importance of KSRP-Pin1 interaction, we studied the effect of Pin1 inhibition by juglone on the KSRP-mediated regulation of PTH gene expression in transfected cells. For that, HEK293 cells were transiently cotransfected with expression plasmids for the GH63 reporter gene or WT GH and either a control or KSRP expression plasmid. KSRP decreased GH63-nt mRNA levels while having no effect on WT GH mRNA (Figure 5, B and C), as in our previous studies (11). Pin1 inhibition by juglone specifically prevented the KSRP-mediated decrease in GH63 mRNA levels and restored GH63-nt mRNA levels to those of control cells. There was no effect of either KSRP or juglone on WT GH mRNA levels (Figure 5, B and C).
We then studied the effect of KSRP depletion by siRNA on the regulation of GH63 mRNA levels by Pin1 overexpression. HEK293 cells were cotransfected with GH or GH63-nt and either a control or a Pin1 expression plasmid (Figure 5, D–F). Pin1 overexpression decreased GH63-nt mRNA levels while having no effect on the control GH mRNA as before (Figure 5, E and F, and Figure 4, B and C). siRNA-mediated knockdown of KSRP by approximately 50% (Figure 5D) caused a specific 3-fold increase in GH63-nt mRNA levels compared with control siRNAs (not shown) as before (11). Importantly, the effect of Pin1 in decreasing GH63 mRNA levels was attenuated in the KSRP-depleted HEK293 cells (Figure 5, E and F). Together, these results indicate that Pin1 and KSRP act together to regulate PTH mRNA levels.
KSRP is phosphorylated at S181, and Pin1 overexpression induces dephosphorylation of KSRP at serine residues. KSRP is a phosphoprotein with 2 reported phosphorylation sites, at S193 and T692 (26). To identify a Pin1 interaction consensus site on KSRP, we transfected HEK293 cells with FLAG-KSRP and analyzed immunoprecipitated KSRP by mass spectrometry. This analysis identified a new phosphorylation site at S181 in the KVQI*SPDSGGLPER peptide that has not been previously reported (Figure 6A and Supplemental Figure 1; supplemental material available online with this article; doi: 10.1172/JCI39522DS1). This serine residue is adjacent to proline, making it a putative Pin1-interacting site. To study the role of S181 on KSRP-Pin1 interaction, we cotransfected HEK293 cells with expression plasmids for Pin1 and either FLAG-KSRP or the FLAG-KSRP S181A mutant. Cotransfected Pin1 did not immunoprecipitate FLAG-KSRP S181A but did precipitate FLAG-KSRP (Figure 6B and Figure 5A). In contrast, both KSRP S193A and T692A mutants bound Pin1 as the WT KSRP (Supplemental Figure 2, A and B).
Figure 6Identification of KSRP phosphorylation at S181 and its role in KSRP-Pin1 interaction and PTH mRNA regulation. (A) Mass spectrometry. MS spectra of tryptic mixture of immunoprecipitated FLAG-KSRP from transfected cells. The peak at 677.8555 m/z was assigned as the unphosphorylated form of amino acids 177–197 (left), and the peak at 717.8390 m/z the phosphorylated form of the peptide (right). Asterisk indicates serine phosphorylation. The P probability of the assignment was equal to 7.51E-06. Sequest X score was equal to 3.25. (B) KSRP-Pin1 interaction. HEK293 cells were cotransfected with either FLAG-KSRP or FLAG-KSRP S181A and GFP-Pin1 or control plasmid. Extracts were immunoprecipitated with either anti-Pin1 or IgG. The pellets and input were immunoblotted as indicated. (C) KSRP dephosphorylation. Cells were cotransfected with FLAG-KSRP and GFP-Pin1 or control plasmid. Extracts were immunoprecipitated with anti-FLAG or IgG and immunoblotted with anti–phosphor-serine or anti-FLAG antibody. (D–H) Cells were cotransfected with the PTH plasmid or control and either FLAG-KSRP or FLAG-KSRP S181A. (D) Immunoblots showing equal amounts of KSRP. (E) Northern blots of PTH mRNA. The lanes were run on the same gel but were noncontiguous (white line). (F) qRT-PCR for PTH mRNA as in E. *Compared with control; **compared with KSRP. (G) IVDA using cell extracts and full-length human PTH mRNA (top) or PTH mRNA without the ARE (bottom). (H) Quantification of IVDA results in 3 experiments; mean ± SEM of time 0; KSRP compared with both control and S181A; *P < 0.05. (I) Cells were cotransfected with the PTH and KSRP or S181A plasmids with and without juglone. RNA was analyzed by qRT-PCR for PTH mRNA; *P < 0.05 (n = 3) compared with cells expressing KSRP and not treated with juglone.
We then studied the effect of Pin1 overexpression on KSRP phosphorylation. HEK293 cells were transfected with FLAG-KSRP and a Pin1 expression plasmid. Immunoprecipitated FLAG-KSRP was analyzed by immunoblots with anti–phospho-serine antibody. Pin1 overexpression decreased the amount of phosphorylated KSRP at serine residues (Figure 6C). Therefore, Pin1 decreases KSRP phosphorylation at serine residues, and an intact serine at amino acid 181 but not at 193 is necessary for Pin1-KSRP interaction in transfected cells.
S181A KSRP mutant is more potent than WT KSRP in decreasing PTH mRNA levels in transfected cells. KSRP phosphorylation at S193 by AKT prevents the binding of KSRP to the exosome complex and thereby KSRP-induced mRNA decay (27). We studied the effect of the newly identified S181 phosphorylation site on KSRP-mediated PTH mRNA degradation. HEK293 cells were cotransfected with expression plasmids for PTH and either WT or S181A KSRP. Both WT and S181A KSRP were expressed at the same levels (Figure 6D). WT KSRP overexpression decreased PTH mRNA levels (Figure 6, E and F) as before (26). The overexpressed S181A mutant was more potent than WT KSRP (Figure 6, E and F). These results were confirmed by IVDAs using extracts from HEK293 cells transfected with either KSRP or KSRP S181A mutant and uniformly labeled PTH mRNA as above. As expected, overexpression of KSRP specifically decreased PTH mRNA stability compared with control extracts (Figure 6, G and H). PTH mRNA decay was further accelerated by cell extracts overexpressing KSRP S181A compared with extracts expressing WT KSRP (Figure 6, G and H). There was no difference in mRNA decay of the truncated PTH transcript lacking the PTH mRNA ARE (Figure 6, G and H). These results demonstrate that dephosphorylated KSRP at the Pin1 interaction site, S181, is more potent than WT KSRP in decreasing PTH mRNA levels and stability. They suggest that Pin1 acts through dephosphorylation of KSRP to decrease PTH mRNA levels in an ARE-dependent manner in transfected cells.
We therefore hypothesized that S181A KSRP, which does not bind Pin1, would not be influenced by Pin1 inhibition. HEK293 cells were transiently cotransfected with expression plasmids for PTH and either WT or S181A KSRP and were incubated with juglone, as above (Figure 4, E and F). KSRP overexpression caused a 90% reduction in PTH mRNA levels (Figure 6I). Juglone treatment prevented this decrease dose dependently (Figure 6I), similar to its effect on reporter GH63 (Figure 4, E and F). In contrast, juglone had no effect on PTH mRNA levels at 3 and 5 μM in cells expressing S181A KSRP and only an effect at the highest juglone concentration (Figure 6I). Together, these results indicate that Pin1–KSRP interaction is dependent upon S181 KSRP phosphorylation, suggesting that Pin1 leads to dephosphorylation at this site and thereby activation of KSRP.
Pin1 interacts with KSRP and regulates KSRP–PTH mRNA interaction in parathyroid glands. In vivo in the parathyroid, Pin1 activity was decreased in parathyroid extracts of rats with increased PTH mRNA levels and stability (Figure 1), and Pin1 inhibition by juglone increased PTH mRNA levels posttranscriptionally (Figure 2). We have previously shown that posttranscriptional regulation of PTH gene expression is mediated by differential binding of KSRP to the PTH mRNA ARE (11). To study KSRP–PTH mRNA–Pin1 interactions in vivo, we first showed KSRP-Pin1 interaction in rat parathyroid gland extracts (Figure 7A). KSRP was coimmunoprecipitated by Pin1 (Figure 7A), indicating that these proteins interact also in the parathyroid, similar to their interaction in HEK293 cells (Figure 5A).
Figure 7Pin1 interacts with KSRP and regulates KSRP–PTH mRNA interaction in parathyroid glands. (A) Pin1-KSRP interaction in the parathyroid. Pin1 was immunoprecipitated from microdissected parathyroid extracts using anti-Pin1 antibody or IgG, fractionated on SDS PAGE, and analyzed by immunoblots for KSRP and Pin1. The lanes were run on the same gel but were noncontiguous (white line). (B–D) RIP analysis. Parathyroid glands were subjected to juglone or vehicle applied topically. RIP was performed on rat parathyroid glands using either anti-KSRP or control IgG. RNA from total lysate (input) and KSRP (IP-KSRP) or control IgG–bound fractions was assayed by qRT-PCR for PTH and control HPRT mRNAs. (B) Agarose gel of qRT-PCR products for PTH and HPRT. The size of PCR products was consistent with the predicted lengths of the amplified fragments. (C) PTH mRNA levels in the input. (D) mRNA levels in immunoprecipitated samples, corrected for mRNA levels in input; mean ± SEM of 3 experiments using parathyroids from pools of different rats. (E) Immunoblots of the parathyroid extracts as in C (input). *P < 0.05.
We then asked whether the effect of juglone to increase PTH mRNA levels is mediated by an effect of Pin1 on KSRP–PTH mRNA interaction in vivo. Juglone was applied topically to rat parathyroid glands as above (Figure 2, C–F). The parathyroid glands were then cross-linked and analyzed by RNA immunoprecipitation (RIP) using anti-KSRP antibody and qRT-PCR analysis. As shown in Figure 7B, RIP analysis showed a specific interaction of KSRP with PTH mRNA and not with a control HPRT mRNA, as we have previously published (11). Juglone increased PTH mRNA levels in the input fraction (Figure 7C) as above (Figure 2, E and F). Pin1 inhibition led to a marked decrease in the amount of KSRP-bound PTH mRNA compared with vehicle-treated parathyroid glands (Figure 7D). This decrease in KSRP–PTH mRNA interaction was not a result of changes in KSRP protein levels that were unchanged by juglone treatment (Figure 7E). Juglone led to the expected decrease in Pin1 protein levels (Figure 7E) (23). AUF1 protein levels were also dramatically decreased in the juglone-treated parathyroid glands, as reported by Shen et al. (20, 21). Due to the marked decrease in AUF1 protein levels here, it was not possible to study AUF1–PTH mRNA interactions in the juglone-treated parathyroids. Our results indicate that Pin1 inhibition by juglone in vivo decreases KSRP–PTH mRNA interaction in the parathyroid. This may contribute to the increased PTH mRNA levels induced by juglone (Figure 7C and Figure 2, E and F).
Pin–/– mice have increased PTH expression. Our results indicate a correlation between Pin1 activity and PTH mRNA levels in vivo and regulation of PTH gene expression by Pin1 in vitro. To study the role of Pin1 in PTH gene expression in vivo more directly, we measured serum PTH, calcium, and phosphorus levels in mice with genetic deletion of Pin1 (Pin–/– mice) (28). Serum PTH levels in the Pin1–/– mice were 3-fold higher than in WT littermates (Figure 8A). There was no change in serum calcium and phosphorus levels (Figure 8, B and C) nor in CYP27B1 (1α-hydroxylase) mRNA levels (Supplemental Figure 3). Therefore, the increased serum PTH was not secondary to changes in serum calcium or phosphorus levels. Immunohistochemistry using an anti-PTH antibody demonstrated a 2-fold increase in PTH protein content in the parathyroid gland of the Pin–/– mice compared with control WT littermates (Figure 8, D and E), in agreement with the high–serum PTH levels in the Pin1–/– mice. Similarly, in situ hybridization demonstrated an approximately 2-fold increase in PTH mRNA levels in the parathyroid gland of the Pin–/– mice (Figure 8, F and G). Together, these results indicate that Pin1 determines basal PTH levels in vivo.
Figure 8Pin–/– mice have high serum PTH and PTH mRNA levels. (A–C) Serum PTH, calcium, and phosphate levels in WT (n = 18) and Pin1–/– (n = 14) mice. (D) Immunostaining of thyroparathyroid section for PTH. (E) Quantification of data presented in 6 sections from each group. (F) In situ hybridization for PTH mRNA in thyroparathyroid sections. (G) Quantification of the in situ hybridization staining in 6 sections from each group. Original magnification, ×100 (D and F). *P < 0.05. PT, parathyroid; T, thyroid tissue.
-
Discussion
The increased PTH mRNA levels of secondary hyperparathyroidism due to calcium depletion or CKD is mediated by the interaction of trans-acting RNA-binding proteins with a defined cis-acting ARE in the PTH mRNA 3′ UTR. AUF1 and Unr (upstream of N-ras) stabilize and KSRP destabilizes the PTH mRNA (11, 22). Many RNA-binding proteins, including KSRP and AUF1, are phosphoproteins, and their activity is tightly regulated through phosphorylation and dephosphorylation signaling cascades. In particular, AUF1 is posttranslationally modified in the parathyroid in correlation with PTH gene expression in secondary hyperparathyroidism (29). Here, we demonstrate for what we believe is the first time that Pin1 is a critical mediator of PTH expression in vivo and in vitro in transfected cells. Pin1–/– mice have increased levels of serum PTH, parathyroid PTH protein, and PTH mRNA levels, indicating that Pin1 determines basal PTH expression in vivo. Moreover, Pin1 isomerase activity is decreased in parathyroid extracts of rats fed a calcium-depleted diet or CKD rats in which PTH mRNA levels and stability are increased. Juglone inhibits Pin1 activity and leads to its degradation by the proteasome (20). Juglone has also recently been shown to inhibit postmitotic dephosphorylation and the exit of mitosis independently of Pin1 by covalent modification of sulfhydryl groups in proteins essential for metaphase/anaphase transition (30). We show that Pin inhibition by juglone increases PTH mRNA levels in vivo posttranscriptionally. The effect of juglone occurs when it is applied topically on the parathyroid or given systemically, indicating a direct effect of the Pin1 inhibitor on the parathyroid. We used HEK293 cells, which reproduce certain aspects of the regulation of a transfected PTH gene, to study the effect of Pin1 on PTH gene expression (11, 31, 32). Pin1 regulates PTH gene expression in transiently transfected HEK293 cells, and this regulation is dependent upon the PTH mRNA 3′ UTR ARE and KSRP. We identify KSRP as a new target protein for Pin1. Pin1 interacts with KSRP and leads to KSRP dephosphorylation, which enhances the KSRP-mediated decrease in PTH mRNA levels. In vivo, Pin1 inhibition decreases KSRP–PTH mRNA interaction, which would lead to the increased PTH mRNA and serum PTH levels after juglone administration.
The Pin1–/– mice have high serum PTH and PTH mRNA levels without the expected hypercalcemia or hypophosphatemia or increase in CYP27B1 (1α-hydroxylase) mRNA levels. This may represent downregulation of the PTH receptor in these knockout mice. In fact, the bones of Pin1–/– mice are markedly abnormal, which was attributed to a premature aging phenotype and not hyperparathyroidism (33).
It was recently shown in HeLa cells that Pin1 shuttles between nuclear and cytoplasmic domains. A nuclear localization signal located at the PPIase domain is based on 3 residues: Lys63, Arg68, and Arg69 (34). Interestingly, our results indicate that even though mutation at Lys63 did not affect Pin1 activity, double-point mutation at Arg68 and Arg69 residues dramatically disrupted Pin1 activity on PTH mRNA levels. The effect of the R68/69 mutation on Pin1 activity here may be due to damage to the Pin1 PPIase domain itself by the mutation. The WW-binding domain mutation (W34A) functions as a dominant negative mutation by preventing endogenous Pin1 from binding to Ser/Thr-Pro targets (35). Mutation at the WW domain attenuated the decrease in PTH mRNA levels by Pin1. Together, our results suggest that both the Pin1 PPIase and WW domains are necessary for its effect on PTH mRNA levels.
AUF1 was shown to be a Pin1 target protein (20, 21). AUF1 consists of 4 isoforms: p37, p40, p42, and p45 (36). Pin1 inhibition by juglone in eosinophil cells leads to proteosome-mediated degradation of 3 of the 4 AUF1 isoforms, p40, p42, and p45, while isoform p37 remains stable and destabilizes TGF-β and GM-CSF mRNA (20, 21). The stabilizing action of AUF1 on the PTH mRNA is in contrast to its effect on many other ARE-containing mRNAs in which it is a destabilizing factor (10, 37). In the parathyroid, juglone led to degradation of all 4 AUF1 isoforms to undetectable levels (Figure 7E). We could therefore not determine a contribution of any of the AUF1 isoforms to the effect of Pin1 on PTH gene expression. It is paradoxical that juglone increased PTH mRNA levels both in transfected cells and in the parathyroid while decreasing protein levels of AUF1, which is a PTH mRNA–stabilizing protein. It is possible that small amounts of AUF1 that remain bind with a higher affinity to the PTH mRNA 3′ UTR and stabilize the PTH mRNA.
We show that the effect of KSRP in decreasing PTH gene expression is dependent upon Pin1. Moreover, Pin1 inhibition by juglone prevents KSRP-mediated PTH mRNA decay through the PTH mRNA ARE and decreases KSRP–PTH mRNA interaction in vivo with no change in KSRP protein levels. KSRP is a phosphoprotein with 2 characterized phosphorylation sites, S193 and T692 (26, 27). T692 is adjacent to a proline, making it a putative consensus Pin1-interacting site. Mutations at these 2 sites did not interfere with KSRP-Pin1 interaction, indicating that these 2 phosphorylated sites are not the Pin1-interacting sites. We identify a new phosphorylation site at S181 in the N terminus of the protein. S181 is adjacent to proline, forming a potential Pin1 interaction site. Indeed, the S181A mutation prevented Pin1-KSRP interaction and enhanced PTH mRNA destabilization by KSRP compared with WT KSRP. Moreover, Pin1 inhibition by juglone did not affect the decrease in PTH mRNA levels by KSRP S181A.
Our results suggest that phosphorylated KSRP at S181 is inactive. Upon interaction with Pin1, cis-trans isomerization of the proline bond in KSRP leads to conformational change, exposing the phosphorylated S181 residue and possibly additional phosphorylation sites. This leads to dephosphorylation, thus activating KSRP, which then interacts with PTH mRNA and enhances its decay (Figure 9). A low calcium diet and CKD led to decreased Pin1 isomerase activity in the parathyroids of these rats. This decreased Pin1 activity would prevent KSRP dephosphorylation, resulting in decreased KSRP–PTH mRNA interaction, inhibition of PTH mRNA degradation, and increased PTH mRNA levels. Consistent with this finding, Pin1 inhibition by juglone increases PTH mRNA levels and decreases KSRP–PTH mRNA interaction. Pin1 is therefore a regulator of PTH gene expression. Pin1 determines basal PTH levels and the response of the parathyroid to chronic hypocalcemia and CKD.
Figure 9Model for the regulation of PTH mRNA stability by PTH mRNA 3′ UTR ARE binding proteins. Under basal conditions, there is a balanced interaction of the PTH mRNA with its stabilizing proteins AUF1 (isoforms p37, p40, p42 and p45) and Unr and the destabilizing protein KSRP. In hypocalcemia or CKD, Pin1 is inactive, resulting in KSRP phosphorylation and hence its inactivation. This would allow AUF1 and Unr to bind the PTH mRNA 3′ UTR ARE with a greater affinity, leading to increased PTH mRNA stability.
Copyright © 2024 American Society for Clinical Investigation
ISSN: 0021-9738 (print), 1558-8238 (online)