Kidney Failure Alters Parathyroid Pin1 Phosphorylation and Parathyroid Hormone mRNA-Binding Proteins, Leading to Secondary Hyperparathyroidism

Significance Statement

Secondary hyperparathyroidism (SHP) is a common complication of CKD that when poorly controlled increases morbidity and mortality. In experimental models, the high serum parathyroid hormone (PTH) of SHP is due to increased PTH mRNA stability, mediated by changes in Protein-PTH mRNA interactions that are orchestrated by the isomerase Pin1. It is not known how CKD stimulates the parathyroid to dramatically increase PTH levels. We identify the CKD-induced post-translational modifications that disrupt parathyroid Pin1 isomerase activity and the effects on the Pin1 target and PTH mRNA decay-promoting protein, KSRP. We suggest that CKD-induced changes in Pin1, and hence sustained KSRP phosphorylation and protein-PTH mRNA interactions, are the driving force in overstimulation of the parathyroid glands in SHP.

Visual Abstract

Abstract

Background

Secondary hyperparathyroidism (SHP) is a common complication of CKD that increases morbidity and mortality. In experimental SHP, increased parathyroid hormone (PTH) expression is due to enhanced PTH mRNA stability, mediated by changes in its interaction with stabilizing AUF1 and destabilizing KSRP. The isomerase Pin1 leads to KSRP dephosphorylation, but in SHP parathyroid Pin1 activity is decreased and hence phosphorylated KSRP fails to bind PTH mRNA, resulting in high PTH mRNA stability and levels. The up- and downstream mechanisms by which CKD stimulates the parathyroid glands remain elusive.

Methods

Adenine-rich high-phosphate diets induced CKD in rats and mice. Parathyroid organ cultures and transfected cells were incubated with Pin1 inhibitors for their effect on PTH expression. Mass spectrometry was performed on both parathyroid and PTH mRNA pulled-down proteins.

Results

CKD led to changes in rat parathyroid proteome and phosphoproteome profiles, including KSRP phosphorylation at Pin1 target sites. Furthermore, both acute and chronic kidney failure led to parathyroid-specific Pin1 Ser16 and Ser71 phosphorylation, which disrupts Pin1 activity. Pharmacologic Pin1 inhibition, which mimics the decreased Pin1 activity in SHP, increased PTH expression ex vivo in parathyroid glands in culture and in transfected cells through the PTH mRNA-protein interaction element and KSRP phosphorylation.

Conclusions

Kidney failure leads to loss of parathyroid Pin1 activity by inducing Pin1 phosphorylation. This predisposes parathyroids to increase PTH production through impaired PTH mRNA decay that is dependent on KSRP phosphorylation at Pin1-target motifs. Pin1 and KSRP phosphorylation and the Pin1-KSRP-PTH mRNA axis thus drive SHP.

Secondary hyperparathyroidism (SHP) increases morbidity and mortality in CKD.^1–3 In experimental SHP, parathyroid hormone (PTH) expression is increased post-transcriptionally. Enhanced PTH mRNA stability correlates with changes in the binding of trans-acting proteins, adenine/uridine (AU)-rich binding factor 1 (AUF1) and K-homology splicing regulatory protein (KSRP), to an evolutionary conserved AU-rich element (ARE) in the PTH mRNA 3′-untranslated region (3′-UTR).⁴

AUF1 recognizes AREs in mRNA 3′-UTRs inducing target mRNA stability or decay,^5–7 whereas KSRP promotes rapid mRNA decay. KSRP contains four contiguous K-homology domains that mediate RNA binding, mRNA decay, and interactions with the exosome ribonuclease complex.⁸ The peptidyl prolyl cis/trans isomerase (PPIase) Pin1 binds phosphorylated Ser/Thr residues followed by proline (pSer/Thr-Pro) in target proteins regulating phosphorylation-induced protein conformation.⁹ The isomerization of the peptide bonds changes phosphorylation, stability, localization, and/or activity of target proteins.^9–12 Pin1 is composed of an N-terminal WW protein interaction domain and a C-terminal catalytic PPIase domain.^9,13–15 Protein kinase A (PKA)-mediated Pin1 serine (Ser)16 phosphorylation abolishes Pin1 interaction with its substrates.¹⁶ Death-associated protein kinase 1 (DAPK1)-induced phosphorylation at Ser71 and Pin1 cysteine (Cys)113 oxidation, both in the catalytic domain, inhibit Pin1 isomerase activity.^17–19 Pin1 interacts with RNA-binding proteins, including KSRP and AUF1 to regulate mRNA decay, and other aspects of mRNA lifecycle.^20–24 We have previously identified a novel KSRP phosphorylation at Ser181, serving as Pin1 target site.²⁰ In parathyroid glands and in human embryonic kidney (HEK) 293 cells, Pin1-KSRP interaction leads to KSRP Ser181 dephosphorylation favoring KSRP-PTH mRNA 3′-UTR-binding and thus PTH mRNA decay. In SHP, parathyroid Pin1 isomerase activity is decreased, and thus phosphorylated Ser181 KSRP fails to bind and induce KSRP-mediated PTH mRNA decay, resulting in increased PTH mRNA and serum levels. Pin1^−/− mice have high serum PTH levels and PTH mRNA and protein content in their parathyroids, supporting a central role for Pin1 in determining PTH levels.^20,25

The molecular mechanisms that steer reduction in Pin1 activity, increasing PTH levels in SHP, are not known. We now suggest that CKD-induced Pin1 phosphorylation, reduction in parathyroid Pin1 activity, and hence sustained KSRP phosphorylation are central to the increased PTH expression of SHP.

Methods

Animals, Housing, and Diets

Experimental CKD was induced in male Sprague-Dawley rats (150–170 g) by an adenine-rich (0.75%) high-phosphate (1.5%) diet (ENVIGO, Harlan Laboratories, Madison, WI) given for 14 days.²⁰ Controls received regular chow containing 0.7% phosphate. CKD in mice was induced in male C57BL/6 mice at 11–12 weeks of age by a moderate (0.3%) adenine-rich high-phosphate (1.2%) diet (ENVIGO) given for 14 days.^26,27 AKI was induced in male C57BL/6 mice at 11–12 weeks of age by a single injection of folic acid (240 mg/kg in vehicle, 0.15 mol/L NaHCO₃, pH 7.4; Sigma-Aldrich, St. Louis, MO).^28,29 Control mice were injected with vehicle. Mice were analyzed at 24 and 48 hours. In some experiments, PT-EYFP (parathyroid-specific enhanced yellow fluorescent protein) mice were used to allow identification and microdissection of the mouse parathyroid glands for organ culture experiments. We generated the PT-EYFP mice by Cre-Lox recombination, mating PT-Cre mice, where the Cre recombinase is driven by the human PTH promoter (FVBTg[PT-Cre]; Jackson Laboratory, Bar Harbor, ME),³⁰ with R26-stop-EYFP/R26R-EYFP mice, where a loxP-flanked STOP sequence followed by the EYFP was inserted into the Gt(ROSA)26Sor locus.³¹ Total DNA was extracted from tail samples of offspring and genotyping was performed by PCR. The primers used were the following: for Cre: forward: 5′-TGCCACGACCAAGTGACAGC-3′ and reverse: 5′-CCAGGTTACGGATATAGTTCATG-3′²⁷ and for EYFP: ROSA 26R 5′-AAAGTC GCTCTGAGTTGTTAT-3′; BTG 60: 5′-GAAAGACCGCGAAGAGTT TG-3′; and BTG 62: 5′-TAAGCCTGCCCAGAAGACTC-3′. All animals had free access to food and drinking water. Experiments were approved by the Institutional Animal Care and Use Committee of the Hebrew University-Hadassah Medical School (authorization numbers MD-18-15408 and MD-18-15610).

Parathyroid Glands in Organ Culture

The two glands from microdissected mouse thyroparathyroid tissue, parathyroid glands from PT-EYFP mice or rat parathyroids (n=6), were maintained in 2-ml Eppendorf tubes with needle-punctured caps for aeration, containing 1 ml DMEM (Gibco Life Technologies, Carlsbad, CA) supplemented with 10% FBS, l-glutamine, and penicillin-streptomycin (Gibco Life Technologies). Forskolin at a concentration of 100 μM, H89 (N-[2-p-bromocinnamylamino-ethyl]-5-isoquinolinesulphonamide) at 150 μM, Sulfopin at 50 μM, or vehicle (DMSO) was added to the medium. The tubes were placed in a CO₂ incubator with constant rocking.^27,32 Medium (100 μl) was collected at the time points indicated and analyzed for PTH. In some experiments, thyroparathyroid glands were removed from adenine-rich high-phosphate diet–induced CKD, AKI, and control mice.

Serum Biochemistry and PTH Levels

Serum was analyzed for calcium and BUN using QuantiChrom kits (BioAssay Systems, Hayward, CA). Serum phosphate was analyzed using a Stanbio Phosphorus Liqui-UV kit (Stanbio Laboratories, Boerne, TX). Serum PTH and PTH secreted to growth medium in organ cultures were measured using rat or mouse 1-84 Intact PTH ELISA kit (Quidel, Athens, OH).

RNA-Protein Extraction

Thyroparathyroid glands in organ culture were removed from the medium at the end of the incubation period and homogenized using a bead beater. RNA was extracted from pools of glands from six mice using TRIzol Reagent (Invitrogen, Carlsbad, CA). RNA from HEK293 cells was also extracted with TRIzol Reagent.²⁰

Protein extracts for immunoprecipitation in cultured HEK293 cells were prepared using RIPA buffer containing 150 mM NaCl, 1% NP40, 0.5% sodium deoxycholate, 0.1% SDS, and protease inhibitors cocktail (Roche, Mannheim, Germany). Antibodies used were against Pin1 (Abcam, Cambridge, MA) and FLAG (Sigma-Aldrich).

Mass Spectrometry for Proteomics and Phosphoproteomics

Mass spectrometry (MS) was performed at the Technion Israel Institute of Technology, Haifa, Israel. Pooled parathyroid tissue samples from two rats in every group in triplicate were homogenized in lysis buffer and sonicated. The proteins were digested in 1 M urea with modified trypsin (Promega, Madison, WI) at a 1:50 enzyme-to-substrate ratio, overnight at 37°C. The tryptic peptides were desalted and resuspended in 40% acetonitrile (ACN), 6% trifluoroacetic acid, and enriched for phosphopeptides on titanium dioxide beads. Bound peptides were eluted with 20% ACN with 325 mM ammonium hydroxide followed by 80% ACN with 325 mM ammonium hydroxide. The resulting peptides were desalted and analyzed by liquid chromatography (LC)-MS/MS on a Q Exactive Plus mass spectrometer (Thermo Fisher Scientific) fitted with a capillary HPLC (easy nLC 1000, Thermo Fisher Scientific). MS was performed in a positive ion mode (at mass range of m/z 350–1800 AMU and resolution 70,000) using repetitively full MS scan followed by collision-induced dissociation (at 35 normalized collision energy) of the 10 most dominant ions (greater than one charge) selected from the first full MS scan. For the phosphoproteomic profiling samples, two pools of six rats each from control and kidney failure rats underwent phosphoenrichment using immobilized metal (Fe⁺³) affinity chromatography on a robotic system (Bravo) before LC-MS/MS analysis.

MS Data Analyses

The MS data were analyzed using the MaxQuant software 1.5.2.8 (www.maxquant.org)³³ for peak picking identification and quantitation using the Andromeda search engine, searching against the rat proteome from the Uniprot database with mass tolerance of 20 ppm for the precursor masses and 20 ppm for the fragment ions. Methionine oxidation, phosphorylation (STY), and protein N-terminal acetylation were accepted as variable modifications, and carbamidomethyl on cysteine was accepted as a static modification. Minimal peptide length was set to six amino acids and a maximum of two miscleavage events was allowed. Peptide- and protein-level false discovery rates were filtered to 1% using the target-decoy strategy. The protein table was filtered to eliminate the identifications from the reverse database and common contaminants. The data were quantified by label-free analysis using the MaxQuant software, based on extracted ion currents of peptides enabling quantitation from each LC-MS run for each peptide identified in any of the experiments. More details are provided in the Supplemental material.

The rat parathyroid MS proteomics and phosphoproteomics data have been deposited at the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifiers PXD029368 and PXD029401, respectively.³⁴ For the primary statistical analysis of the MaxQuant proteomic data, we uploaded proteinGroups.txt and evidence.txt files to the ProteoSign webserver.³⁵

Cell Cultures and Transient Transfection

HEK293 cells were transiently transfected in 24-well plates for RNA analysis and in 10-cm plates for protein extractions and immunoprecipitation, respectively, using a TransFectin reagent (Bio-Rad, Hercules, CA). Four hours after transfection, the cells from the 24-well plates were trypsinized and reseeded in 96-well plates. The next day, cells were incubated with vehicle (DMSO), forskolin, or Sulfopin at 5 or 10 μM for the indicated time points. RNA was extracted with TRIzol (Invitrogen) for quantitative RT-PCR (qRT-PCR) analysis.

Plasmids

The human PTH gene including exons and introns in pcDNA3 and the GH-PTH mRNA 63-nt plasmid (GH63) containing the 63-nt rat PTH mRNA ARE cloned between the 3′ of the GH mRNA coding sequence and the GH mRNA 3′-UTR were previously described.^4,36,37 The GH expression plasmid was kindly provided by O. Meyuhas (Hebrew University-Hadassah Medical School, Jerusalem, Israel). pCMV-Tag 2B plasmid containing FLAG-full length KSRP or the KSRP phosphorylation null mutants Thr692Ala and Thr100Ala were kindly provided by R. Gherzi (IRCCS Ospedale Policlinico San Martino, Genova, Italy). Ser181Ala, Ser181Ala;Thr100Ala double mutant, and Ser181Ala;Thr100Ala;Thr692Ala triple mutant were prepared using the QuikChange Lightning Site-Directed Mutagenesis Kit (Agilent Technologies, Santa Clara, CA) according to the manufacturer’s instructions. Empty control vector pcDNA3 (Invitrogen) was used as control. GST-Pin1 expression plasmid was a kind gift from K.P. Lu (Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA).

qRT-PCR Analysis

cDNA was synthesized using a qScript cDNA Synthesis Kit (Quantabio, Beverly, MA), and qRT-PCR analysis was performed with PerfeCTa SYBR Green FastMix, Low ROX (Quantabio) in a ViiA 7 Fast Real-Time PCR System (Applied Biosystems, Waltham, MA). Primers are shown in the Supplemental material.

Pin1 Phosphorylation Assay

GST-tagged bait proteins were purified using Pierce GST Protein Interaction Pull-Down kit (Thermo Fisher Scientific). To examine recombinant Pin1 phosphorylation, parathyroids from control and CKD rats were microdissected and homogenized using a bead beater (a pool from 11 to 13 rats in each group) in 80 mM β-glycerophosphate, 20 mM EGTA, 15 mM MgCl₂, 50 mM NaVO₄, and a protease inhibitor cocktail (Roche). The lysates were centrifuged for 5 minutes at 4°C and the kinase assay performed at 37°C for 30 minutes in a 50-μl reaction volume containing 50 mM Tris-HCl (pH 7.5), 10 mM MgCl₂, 50 mM KCl, 1 mM DTT, 1 mM EGTA, 0.16 mCi/ml [γ-³²P]ATP, parathyroid extracts (10–20 μg), and GST or GST-Pin1 bound beads (6 μg). The beads were then washed and the proteins eluted with reduced glutathione (10 mM). Eluted proteins were run on SDS-PAGE. ³²P-labeled GST-Pin1 was detected by autoradiography and GST/GST-Pin1 by Coomassie blue staining of the gels.²⁴ Protein bands were quantified with Quantity One (Bio-Rad).

Immunofluorescence Staining and Quantification

Microdissected rat parathyroid glands and mouse thyroparathyroid tissue were embedded in paraffin and sections prepared. In some experiments, HEK293 cells were grown in 24-well plates. The cells were incubated with 5 μM forskolin for 1 hour and then fixed with methanol for 5 minutes. Immunostaining of glands or cells was performed using the following primary antibodies diluted in Cas block (Zymed Laboratories, San Francisco, CA): p-Pin1 Ser71 (1:1000, a kind gift from K.P. Lu, Harvard Medical School, Boston, MA),¹⁷ p-Pin1 Ser16 (1:300, Bioss, Woburn, MA), total Pin1 (1:500 Cell Signaling, Beverly, MA), and PTH (1:1000, Bio-Rad, Hercules, CA). Fluorochrome-conjugated secondary antibodies Cy3 and Cy5 (Bethyl, Montgomery, TX) and nuclear staining SYTOX (Life Technologies) were used for detection. Images were obtained using a Fluoview 1000 Olympus fluorescence microscope. Immunofluorescence staining was quantified using ImageJ (National Institutes of Health, Bethesda, MD). Immunofluorescence staining of HEK293 cells was visualized using the spinning disk confocal microscopy, which improves the speed of image acquisition; NIS-Elements software (Nikon, Melville, NY) was used for quantification.

RNA-Protein Pulldown and MS

In vitro–transcribed RNA was prepared from linearized plasmid containing the human PTH 3′-UTR³⁸ using a Biotin RNA Labeling Mix (Roche) and T7 RNA polymerase. Biotinylated RNA-protein pulldown was performed according to the protocol of Panda et al.³⁹ Biotinylated RNA was incubated with extracts (1 mg) from HEK293 cells supplemented with forskolin, Sulfopin, or vehicle DMSO, as described above, in triplicate. Nonbiotinylated RNA was transcribed and incubated with cell extracts as background control. Protein-RNA complexes were pulled down using streptavidin-coated beads and eluted with PEB buffer at 100°C for 10 minutes. MS was performed at the Weizmann Institute of Science, Rehovot. The samples were analyzed by LC-MS. The samples were subjected to tryptic digestion using an S-trap. The resulting peptides were analyzed using Waters HSS-T3 column on nanoflow LC (nanoAcquity) coupled to high-resolution, high-mass accuracy MS (Q Exactive Plus). Protein annotations were characterized using the DAVID webserver (https://david.ncifcrf.gov/home.jsp). A heat map of differentially pulled-down proteins was generated with the NMF package in R environment. Additional information is provided in the Supplemental material.

Statistical Analyses

Values are presented as mean±SEM. A two-tailed t test was used to assess differences among groups. P<0.05 was considered significant.

Results

Experimental CKD Induces Changes in Parathyroid Proteome and Phosphoproteome Composition

To characterize molecular changes induced by CKD in SHP parathyroids, we performed the first proteome and phosphoproteome analysis of the minute microdissected rat parathyroid tissue. Kidney failure SHP was induced by an adenine-rich high-phosphate diet.²⁰ After 2 weeks, pools of parathyroids from two rats were analyzed in triplicate by MS. Proteome analysis identified 4950 protein groups, encoded by 3790 genes. These proteins are shown in Supplemental Table 1 alongside respective tissue mRNA expression extracted from the Human Protein Atlas (https://www.proteinatlas.org/ version 21.0)⁴⁰ to expose possible contaminants from neighboring tissues. Parathyroids from CKD rats could be distinguished from controls based on the overall proteome profiles (Figure 1A). A total of 173 and 110 proteins were up- and downregulated, respectively, in SHP gland extracts, compared with controls (Figure 1B and Supplemental Table 1). As expected, hyperparathyroidism was accompanied by changes in ribosomal proteins, translation, mRNA stability, and secretion pathways (Figure 1C). Proteome analysis thus revealed profile alteration and potentially modified pathways for future studies (Supplemental Figure 1).

Figure 1.

Experimental CKD leads to proteomic changes in parathyroid tissue. Rats were fed either a control or an adenine-rich high-phosphate diet for 2 weeks to induce kidney failure and extracts were prepared from microdissected parathyroid glands for MS proteome analysis. (A) Principal component analysis plot of MS intensity data (n=4950 protein groups) showing proteomics-based distinction of adenine (red) from control (blue) glands. (B) Volcano plot showing the landscape of protein dysregulation in adenine glands. Dark and light brown dots represent proteins at adjusted P value cutoffs of 0.05 and 0.005, respectively. Positive log₂ fold change represents proteins that are increased in kidney failure parathyroids. (C) Enrichment of Reactome pathway-associated proteins among the differentially expressed parathyroid proteins (see also Supplemental Figure 1).

Phosphoproteome analysis of pools of parathyroid extracts from six control and six kidney failure parathyroids recovered 5370 parathyroid phosphopeptides (87.3% pSerine [Ser], 12.0% pThreonine [Thr], and 0.7% pTyrosine [Tyr]), encompassing 3884 phosphosites in 2222 phosphoproteins. Results of biologic theme or function enrichment analysis showed that protein phosphorylation and kinase activity and structure terms and transcription regulation dominate the enrichment lists (Supplemental Figure 2A). Enrichment of ErbB and mTOR signaling-related phosphoproteins is also apparent, which is consistent with our previous findings of mTOR activation in SHP rats.⁴¹ CLOCK protein phosphorylation was also increased in kidney failure (P<0.001) at a site that alters CLOCK function (Supplemental Table 2), supported by findings of a parathyroid circadian clock that is disturbed in SHP.⁴²

The detected phosphoproteome was subjected to phosphosite- and gene-centered enrichment analyses by a contemporary approach, using PhosR package.⁴³ To this end, we processed an additional pool of parathyroids from kidney failure and control rats, and only phosphopeptides identified in both runs were included in the analysis (n=4171, Supplemental Table 2). Phosphosite dysregulation in adenine compared with control rats is shown in Figure 2A. The median log₂ fold change of phosphosite abundance was 0.14 (approximately 10.5% increase), significantly higher than 0 (no change) according to a one-sample Wilcoxon test (P<0.001). Consistent with this shift, there were 460 significantly upregulated phosphosites (11.0%) and 348 significantly downregulated sites (7.4%), indicating a possible global increase in protein phosphorylation. This may be expected in light of increased intracellular phosphate concentration (shown in endothelial cells) in uremic SHP.⁴⁴ There was also enrichment of a potential MAPK phosphorylation motif^45,46 among the recovered phosphosites in the kidney failure pools (median log₂ fold change 0.22, approximately 16.3% increase) (Supplemental Figure 2B). Thus, the phosphoproteome analysis suggests a globally increased phosphorylation state.

Figure 2.

Experimental CKD leads to phosphoproteomic changes in parathyroid tissue. (A) Volcano plot showing phosphosite dysregulation in parathyroid extracts from rats fed adenine-rich (two pools of six rats) versus control diet (two pools). Colored dots represent phosphorylation sites (red, increased; green, decreased) at adjusted P<0.05. Positive log₂ fold change represents phosphoproteins that are increased in kidney failure parathyroids. The top upregulated phosphosites (log₂ fold change ≥5) were Sptbn1:S2099, Bckdk:T32, Hist3h2ba:S5, Tmem69:S13, Hp1bp3:T51, and Virma:S1578, whereas Casr:S1066, Evl:S333, Lmo7:S619, Lpp:T630, Krt86:S47, and Ptbp1:S140 were the most downregulated sites. (B) Two Sample Logo⁴⁸ presenting the only statistically robust sequence-aligned difference between phosphosites increased in adenine rats and phosphosites decreased in adenine rats; the odds ratio for P in the ninth position in adenine versus control was 0.79, Bonferroni-corrected P<0.001. The inset shows additional, less robust differences for which the Bonferroni-corrected P values were insignificant. (C) KSRP phosphorylation sites detected by MS analysis as above. Peptides that are potential target sites for Pin1, Ser182 (VQISP, valine-glutamine-isoleucine-serine-proline), and Thr101 (VNNNTP, valine-asparagine-asparagine-asparagine-threonine-proline) were increased and Ser481 (LFVIRGSP, leucine-phenylalanine-valine-isoleucine-arginine-glycine-serine-proline), were decreased in parathyroids obtained from adenine (CKD) rats.

PhosR prioritizes potential kinases responsible for the phosphorylation change of phosphosites based on kinase recognition motif and phosphoproteome dynamics.⁴³ Kinase activity scores that sum changes in phosphorylation of targets, for which the targeting kinase is prioritized with a probability score, pointed to significantly increased activity of specific kinases in kidney failure parathyroids (Supplemental Figure 2B). Among others, there was increased activity of calcium/calmodulin-dependent protein kinase type II (Camk2a), cGMP-dependent protein kinase 1 (Prkg1), and PKA (Prkaca). Phosphorylated Camk2a blocks PTH secretion ex vivo and negatively correlates with serum calcium levels in patients with primary hyperparathyroidism and thus may act to oppose hypersecretion,⁴⁷ but studies in SHP are lacking. Specifically, cAMP/PKA (Prkaca)-dependent mechanisms inactivate the isomerase Pin1 by phosphorylating Pin1 Ser16 in the WW domain.¹⁶ PKA-activated Pin1 Ser16 phosphorylation could lead to the decreased parathyroid Pin1 activity in SHP reported in experimental CKD or prolonged hypocalcemia.²⁰

Applying the “Two Sample Logo” approach⁴⁸ we found global differences between kidney failure rat parathyroid proteins with enriched and depleted phosphosites (Figure 2B). Phosphorylated Ser, Thr, or Tyr followed by Proline (Pro) was markedly depleted in glands from kidney failure rats (odds ratio 0.79, Bonferroni-corrected P<0.001). Phosphorylated Ser/Thr-Pro motifs are targets for Pin1. The reduced motifs imply fewer available potential Pin1 targets in kidney failure parathyroid extracts that could be due to changes in kinase activity. In light of the latter results and our prior findings,²⁰ we narrowed our search for proteins that were phosphorylated on Ser/Thr-Pro motifs as potential targets for Pin1 isomerase activity. We have previously shown that Pin targets KSRP and induces KSRP dephosphorylation at Ser181. Decreased Pin1 activity, as in CKD, had the opposite effect, to increase Ser181 phosphorylation and PTH mRNA levels.²⁰ Indeed, CKD led to parathyroid KSRP hyperphosphorylation at two Pin1 potential target sites, Ser182 (Ser181 in human) and Thr101 (P=0.016). pSer481, a yet unidentified potential Pin1 target, was decreased in kidney failure parathyroids (P=0.041, Figure 2C and Supplemental Table 2). We therefore studied the changes that affect Pin1 isomerase activity and their role in KSRP-mediated PTH expression.

Chronic and Acute Kidney Failure Induce Parathyroid Pin1-Specific Phosphorylation That Parallels Decreased Pin1 Activity in SHP

Pin1 phosphorylation at Ser16 and Ser71 disrupts its isomerase activity (Figure 3C).¹⁸ We therefore determined Pin1 phosphorylation in CKD parathyroids where Pin1 activity is decreased.²⁰ An in vitro kinase assay using pools of microdissected parathyroid gland extracts from control and kidney failure rats and [γ-³²P]ATP showed that SHP parathyroid extracts were more potent in inducing recombinant GST-Pin1 phosphorylation compared with controls (Figure 3, A and B).

Figure 3.

Experimental CKD leads to increased parathyroid Pin1 Ser16 and Ser71 phosphorylation. (A) In vitro phosphorylation assay of recombinant GST-Pin1 by parathyroid extracts from control and experimental chronic kidney failure (CKD) rats. Male rats were fed an adenine-rich high-phosphate or a control diet for 2 weeks. Extracts from microdissected parathyroid pools of 11–13 rats in each group were incubated with GST or GST-Pin1 and [γ³²P] ATP. ³²P-labeled GST-Pin1 was detected by SDS-PAGE autoradiography (top) and Coomassie blue staining for input GST and GST-Pin1 recombinant proteins (bottom). (B) Quantification of the intensity in autographs as in (A) and additional three repeat experiments. (C) Schematic representation of Pin1 protein showing the WW protein-binding and the catalytic (PPIase) domains. The regulatory Ser16 and Ser71 phosphorylation sites and their respective kinases as well as the Cys113 site and pharmacologic modulators are shown. (D–G) Pin1 immunofluorescence analysis of thyroparathyroid sections from control and CKD rats as above. Sections were stained for SYTOX nuclear staining (green) and antibodies for PTH (blue), Pin1 p-Ser16 (D) or p-Ser71 (F) (red), and merge (purple). (E and G) Quantification of relative fluorescence intensity for p-Pin1 in slides as in (D) and (F) and additional rat sections, measured by Image J software. (H–K) Pin1 immunofluorescence analysis of thyroparathyroid sections from control and folic acid–induced AKI mice at 24 hours (H) or 48 hours (J) post–folic acid or vehicle (control) injection. Sections were stained as above with antibodies for PTH and Pin1 p-Ser16 (H) or p-Ser71 (J). (I and K) Quantification of Pin1 staining in (H) and (J) and additional mice. Results in (B), (E), (G), (I), and (K) are presented as mean±SEM of fold change, compared with controls. *P<0.05.

We then examined Pin1 Ser16 or Ser71 phosphorylation in CKD parathyroids that could not be detected in our MS analysis, because of limitations of the trypsin digestion method.⁴⁹ Immunofluorescence analysis showed increased phosphorylation of both Pin1 Ser16 and Ser71 in kidney failure SHP parathyroid glands compared with controls (Figure 3, D–G), with the expected no changes in total Pin1 protein levels (Supplemental Figure 3).²⁰

To study Pin1 phosphorylation in an additional SHP model, we induced AKI by high-dose folic acid in mice.^28,29 Folic acid led to the expected increase in serum urea and PTH levels (Supplemental Figure 4).^28,50 AKI mice had increased parathyroid Pin1 pSer16 and pSer71 phosphorylation compared with vehicle-injected mice (Figure 3, H–K), with no change in total Pin1 levels (not shown). Therefore, SHP is characterized by increased parathyroid phosphorylation activity toward Pin1. Specifically, Pin1 Ser16 and Ser71 phosphorylation is increased in chronic and acute kidney failure SHP, correlating with the decreased Pin1 activity in SHP.²⁰

PKA Activation Induces Pin1 Ser16 Phosphorylation and Mediates the Increase in PTH Expression in Parathyroid Organ Cultures and Transfected Cells

PKA, which was predicted to be activated by phosphoproteome analysis in CKD parathyroids (Supplemental Figure 2B), phosphorylates Pin1 at Ser16 in the WW domain¹⁶ (Figure 3C). We pharmacologically manipulated Pin1 in parathyroid glands ex vivo, as there is no parathyroid cell line.^27,51 PKA activation by forskolin increased PTH accumulated in the culture media of mouse thyroparathyroids in culture, compared with vehicle-treated glands (Figure 4A). PKA inhibition by H89 decreased PTH secretion in rat microdissected parathyroid glands in culture (Figure 4B). H89 also decreased PTH secretion in mouse thyroparathyroid glands in culture from control and both adenine-rich high-phosphate diet induced kidney failure and AKI (Figure 4, C and D). Therefore, PKA that increases Pin1 Ser16 phosphorylation is central to increased PTH secretion in parathyroids from normal renal function and SHP rodents.

Figure 4.

PKA-induced Ser16 phosphorylation at the Pin1 WW domain increases PTH expression in parathyroid glands in culture and in transfected HEK293 cells. (A) Mouse thyroparathyroid glands were microdissected and the two glands from each mouse maintained in culture with the PKA activator forskolin (100 µM) or vehicle (DMSO). PTH accumulated in the culture medium was measured at 1 and 3 hours. (B) Rat microdissected parathyroid glands (two glands from each rat) were maintained in culture with the PKA inhibitor H89 (150 µM) or vehicle (DMSO) and secreted rat PTH was measured at 24 hours. (C) Mice were fed a normal (Control) or an adenine-rich high-phosphate diet (CKD). At 2 weeks, thyroparathyroid glands were cultured with H89 or DMSO (+D) as in (B). PTH accumulated in the culture medium was measured at 7 and 24 hours. (D) Mice received a single injection of folic acid (240 mg/kg) to induce AKI. At 24 hours, thyroparathyroid glands were removed and cultured with H89 or DMSO (+D) and PTH accumulated in the culture medium was measured. (E) HEK293 cells were transiently transfected with an expression plasmid for the human PTH gene. Transfected cells were incubated with forskolin (10 µM) or DMSO. PTH mRNA levels at 3 and 24 hours, normalized to GAPDH mRNA, were determined by qRT-PCR. Results are presented as mean±SEM of accumulated PTH (A–D) or fold change of PTH mRNA (E), compared with vehicle-treated glands or cells. *, P<0.05 compared with control rodent parathyroids or cells incubated with DMSO; #, P<0.05 compared with control+D.

To further study the effect of PKA activation on PTH expression, we used HEK293 cells transiently transfected with an expression plasmid carrying the human PTH gene. This model recapitulates the regulation of PTH mRNA decay through RNA binding proteins-PTH mRNA interactions.^20,36,37 Forskolin led to the expected increase in Pin1 Ser16 phosphorylation (Supplemental Figure 5).¹⁶ Importantly, forskolin increased PTH mRNA levels (Figure 4E), similar to its effect in parathyroid glands in culture (Figure 4A). Therefore, PKA activation, that increases Pin1 Ser16 phosphorylation, mediates the increased PTH expression ex vivo in parathyroid glands and in transfected HEK293 cells. These findings are consistent with increased parathyroid Pin1 Ser16 phosphorylation in vivo (Figure 3) and decreased Pin1 activity in SHP.²⁰

Pin1 Inhibition by Sulfopin Increases PTH Expression in Parathyroid Organ Cultures

Sulfopin inhibits Pin1 by targeting Cys113 in the catalytic domain (Figure 3C).^15,52 We microdissected parathyroids with minimal adjacent thyroid tissue by fluorescence-guided microsurgery in YFP-expressing mice (PT-YFP) (Figure 5A). Sulfopin increased both PTH secretion and mRNA levels in parathyroids in culture as above (Figure 5, B and C). Altogether, these studies show that Pin1 pharmacologic inhibition at both the protein-binding and catalytic domains, that mimics the decreased Pin1 isomerase activity in SHP, increases PTH levels.

Figure 5.

Targeting Pin1 Cys113 at the catalytic domain by the covalent Pin1 inhibitor Sulfopin increases PTH expression in mouse parathyroid glands in culture. (A) The exposed neck area in a mouse showing YFP expression specifically in the parathyroids (PT-YFP) used for fluorescence-guided microsurgery of the parathyroid glands with minimal contamination of the adjacent thyroid tissue to allow parathyroid-specific mRNA analysis. T, thyroid; tr, trachea. (B) Microdissected parathyroid glands from PT-YFP mice (two glands from each mouse) were maintained in culture with Sulfopin (50 µM) or DMSO and PTH accumulated in the growth medium was measured at 24 hours. (C) PTH mRNA levels, normalized to ACTB mRNA, were determined by qRT-PCR in pooled glands from five mice in each group as in (B) and an additional repeat experiment. Results are presented as mean±SEM compared with vehicle-treated glands. *, P<0.05.

Pin1 Inhibition Increases PTH Expression through the PTH mRNA 3′-UTR 63-nt Protein-Binding Element

To determine whether Pin1 pharmacologic inhibition increases PTH expression through the PTH mRNA ARE, we transfected HEK293 cells with a GH63 reporter plasmid that contains the PTH mRNA 63-nt ARE that is both necessary and sufficient for protein-PTH mRNA interaction and regulation of PTH mRNA stability (Figure 6A).^4,20,36 Both forskolin and Sulfopin increased GH63 mRNA levels (Figure 6, B and C), similar to their effect on PTH mRNA (Figure 4E), but had no effect on wild-type GH mRNA. Therefore, Pin1 inhibition increases PTH expression through the PTH mRNA 63-nt cis element.

Figure 6.

Pin1 inhibition by forskolin or Sulfopin increases PTH mRNA levels through the PTH mRNA 3′-UTR 63-nt protein-binding element. (A) Schematic representation of the constructs used for transfection, showing the PTH, native growth hormone (GH) and the reporter GH containing the PTH mRNA 3′-UTR AU-rich 63-nt cis-acting element (ARE) mRNAs (GH63). CDS, coding sequence; UTR, untranslated region. (B and C) HEK293 cells were transiently transfected with expression plasmids for GH or GH63. Cells were incubated with forskolin (5 µM) (B), Sulfopin (5 µM) (C), or vehicle (DMSO, +D). PTH mRNA levels were measured at 24 hours by qRT-PCR, normalized to TATA box-binding protein (TBP) mRNA. Results are presented as mean±SEM of fold change, compared with GH mRNA in vehicle-treated cells. *, P<0.05 inhibitor versus vehicle.

Pin1 Inhibition Alters Overall Protein-PTH mRNA 3′-UTR Interaction

To characterize the PTH mRNA 3′-UTR–binding proteins and the effect of Pin1 inhibition on binding, we performed LC-MS analysis of human PTH mRNA 3′-UTR–interacting proteins from HEK293 cells incubated with forskolin, Sulfopin, or vehicle. Overall, MS identified 610 proteins, of which 82 (13.4%) were significantly above the background (Supplemental Figure 6A). The majority of these (53%) were RNA-binding proteins and many others were involved in mRNA regulation and processing (Figure 7A, and Supplemental Figure 6B). The top enriched RNA-binding proteins were PRKRA (RNA-dependent protein kinase activator) and IGF2BP1 (mRNA-binding protein, which inhibits both mRNA translation and degradation). Additional highly enriched proteins were the known PTH mRNA-interacting proteins KSRP and AUF1 (HNRNPD),^4,6 confirming the validity of the assay. Of interest, Pin1 was also pulled down by the 3′-UTR (Supplemental Figure 6A). Enrichment of several pulled-down proteins was boosted in extracts from cells incubated with either of the Pin1 inhibitors, and the effect was largely similar (Figure 7, B and C, and Supplemental Figure 6, C and D). Thus, Pin1 inhibition results in altered protein-PTH mRNA interactions. The roles of these newly identified PTH-interacting proteins remain to be determined.

Figure 7.

MS–based discovery of PTH mRNA-binding proteins and the response to Pin1 inhibition mimicking experimental CKD. (A) Characteristics of the PTH 3′-UTR specifically bound (pulled-down) proteins when evaluated for enrichment in Reactome database pathways at an adjusted P value cutoff of 0.01. Supplemental Figure 5B expands this analysis to additional data sources. (B) Results of differential pulldown analysis from Pin1-inhibited cell extracts versus control cells, wherein the y axis shows the effect of Sulfopin and the x axis shows the effect of forskolin. (C) Heat map showing intensity values of the differentially pulled-down proteins (which are color-labeled in panel B). Brighter shades represent higher levels in the pull down extract (log₂ scale).

PKA-Activated Pin1 Inhibition Decreases Pin1-KSRP Interaction

To understand the effect of Pin1 inhibition on KSRP, we first studied Pin1-KSRP interaction. Western blots showed that transfected FLAG-KSRP was pulled down by Pin1, as expected (Supplemental Figure 7).²⁰ Forskolin decreased Pin1-FLAG-KSRP interaction, consistent with its effect on Pin1 protein-binding domain. Sulfopin acting on the catalytic domain, had no effect on Pin1-KSRP interaction (Supplemental Figure 7). Therefore, PKA activation leads to dissociation of Pin1 from KSRP in HEK293 cells.

Pin1-Induced Changes in KSRP Phosphorylation Alter PTH Expression

To characterize the effect of potential KSRP pSer/Thr-Pro Pin1 target sites on PTH mRNA, HEK293 cells were cotransfected with expression plasmids for PTH and wild-type or KSRP phospho-null mutants. We chose KSRP Ser181 and Thr101 that were hyperphosphorylated in parathyroids from CKD rats in the phosphoproteome analysis (Figure 2C) and Thr692 phosphorylation that impairs KSRP’s decay-promoting activity⁵³ (Figure 8A). Cotransfection of PTH and wild-type KSRP expression plasmids led to the expected decrease in PTH mRNA (Figure 8, B and D), consistent with KSRP’s decay-promoting activity.²⁰ Interestingly, Ser181Ala, Thr100Ala, and Thr692Ala single phosphosite null mutants had a similar effect as wild-type KSRP to decrease PTH mRNA, and forskolin only partially reversed this effect (Figure 8, B and C). In contrast to the single phosphorylation null mutants, the ability of wild-type KSRP to decrease PTH mRNA was magnified by overexpression of KSRP Ser181Ala;Thr100Ala double and more so by the Ser181Ala;Thr100Ala;Thr692Ala triple mutants (Figure 8D). Thus, KSRP dephosphorylation at more than one Ser/Thr-Pro motif is required to maximize the KSRP-mediated decrease in PTH expression. Forskolin activated Pin1 Ser16 phosphorylation, did not restore the increase in PTH expression when the double or triple KSRP phosphorylation null mutants were overexpressed (Figure 8D). We next asked whether Pin1 inhibition by Sulfopin and KSRP phosphorylation act via the PTH mRNA ARE. Sulfopin increased GH63 mRNA levels in wild-type KSRP overexpressing cells (Supplemental Figure 8), similar to its effect without KSRP overexpression (Figure 6C). Importantly, Pin1 inhibition failed to increase GH63 mRNA levels in cells over expressing the KSRP triple phosphorylation null mutant (Supplemental Figure 8). Therefore, Pin1 inhibition by forskolin or Sulfopin increases PTH expression in a manner that is dependent on KSRP phosphorylation and the PTH mRNA 3′-UTR cis element. These findings are consistent with loss of Pin1 activity in SHP parathyroids that would increase PTH expression through modified KSRP-PTH mRNA interaction.²⁰

Figure 8.

PKA activation increases PTH expression through KSRP phosphorylation. (A) Schematic representation of the primary structure of KSRP, including the four hnRNP K-homology (KH) domains and position of the phosphorylation sites relevant to this study. (B–D) HEK293 cells were transiently cotransfected with expression plasmids for human PTH and empty vector, native KSRP, or KSRP phosphorylation null single Thr100Ala (T100A) (B); Thr692Ala (T692A) or Ser181Ala (S181A) (C); or native KSRP, and KSRP Ser181Ala;Thr100Ala (double) or Ser181Ala;Thr100Ala;Thr692Ala (triple) (D) mutants. Transfected cells were incubated with forskolin (FK) or vehicle DMSO (+D). PTH mRNA levels at 24 hours, normalized to GAPDH mRNA, were determined by qRT-PCR. Results are presented as mean±SEM of fold change, compared with cells transfected with human PTH expression plasmid and a control empty vector. *, P<0.05 forskolin versus DMSO; #, P<0.05 forskolin versus KSRP+D. (E) Model for the changes that renal failure induces in the parathyroid cell that lead to increased PTH gene expression in SHP. Under basal conditions, Pin1 is unphosphorylated at Ser16 and Ser71 and thus active, leading to conformational changes and KSRP dephosphorylation. Unphosphorylated KSRP binds to the PTH mRNA 3′-UTR ARE and along with other PTH mRNA-binding proteins determines steady-state PTH mRNA levels. CKD induces parathyroid Pin1 phosphorylation and hence decreased isomerase activity, leaving KSRP phosphorylated at three Pin1 target sites. Phosphorylated KSRP fails to bind PTH mRNA and induce mRNA decay, resulting in increased PTH mRNA stability and high serum PTH levels.

Discussion

SHP is a key player in the devastating clinical consequences of CKD. The study of the molecular signals that induce SHP in CKD has been hampered by the lack of an appropriate cell culture system and the minute size of the parathyroid glands in rodent models. Here we characterized, for the first time, the global changes in protein expression and phosphorylation induced by kidney failure in rat parathyroid glands by MS analysis. We induced kidney failure by an adenine-rich high-phosphate diet given for 2 weeks, that leads to mild renal failure combined with hyperphosphatemia and SHP.⁵⁴ At this time point there was no decrease in parathyroid VDR, CaR, and klotho protein levels in the proteome analysis, as can be found only later, at 6 weeks of the diet.⁵¹ The proteome data therefore highlight new potential pathways of early CKD-induced changes in parathyroid function in SHP. These profile alteration provide data for future studies.

Phosphoproteome analysis identified up- and downregulated phosphosites, with more upregulated than downregulated sites, suggesting an increase in global protein phosphorylation in renal failure parathyroids. We show enrichment of mTOR signaling-related phosphorylated proteins in CKD parathyroids, consistent with parathyroid mTOR activation in SHP induced by either kidney failure or hypocalcemia in rats.⁴¹ There was also increase in parathyroid CLOCK protein phosphorylation, at a conserved site that alters CLOCK function. CLOCK undergoes phosphorylation in a circadian manner, which is coupled to its degradation. Indeed, recent findings showed that a circadian CLOCK operates in parathyroid and is disturbed in SHP, with an increase in CLOCK Ser852 phosphorylation that alters CLOCK function.⁴² These changes in mTOR and CLOCK therefore confirm the validity of the assay. Kinase activity scores that sum changes in phosphorylation of target proteins, pointed to increased activity of several kinases in CKD parathyroids, including PKA. PKA disables Pin1 isomerase activity by Pin1 Ser16 phosphorylation, leading to Pin1 dissociation from its substrates.¹⁶ PKA activation and subsequent Pin1 phosphorylation could explain the decreased parathyroid Pin1 activity that occurs in SHP.²⁰ Accordingly, global differences of enriched and depleted phosphosites showed decreases in potential phosphorylated Ser/Thr-Pro Pin1 target sites in CKD parathyroids. Together with the decreased Pin1 activity,²⁰ this suggests that in CKD there are less available Pin1 target sites, highlighting the role for Pin1 in SHP.

Pin1 targets KSRP phosphorylation at Ser181 that decreased PTH mRNA stability and levels in transfected cells.²⁰ We hypothesized that decreased Pin1 activity in CKD parathyroids²⁰ leaves KSRP in its phosphorylated state. Indeed, phosphoproteome analysis showed parathyroid KSRP Ser182 (Ser181 in human) and Thr101 hyperphosphorylation in CKD parathyroids. These KSRP post-translational modifications would alter KSRP-PTH mRNA interaction contributing to the increased PTH mRNA levels in SHP.²⁰ Of interest, Ser181 KSRP and Pin1 phosphorylation were also identified in PTH regulation of opossum kidney cells. KSRP bound both type IIa sodium phosphate cotransporter (Napi2a) and the sodium/proton exchanger isoform 3 (NHE3) mRNAs in these cells. However, KSRP KO prevented PTH-induced decrease in NHE3 but not Napi2a mRNA expression.⁵⁵ Therefore, although KSRP does not seem to be involved in Npt2a mRNA stability, KSRP-Napi2a mRNA binding suggests that it may yet have another physiologic role in Napi2a gene expression. It is intriguing that KSRP and Pin1 are involved in two organs central to mineral metabolism, both expressing the CaSR. Recently, mRNA profiles by RNA sequencing of porcine parathyroid glands after dietary phosphate intervention showed that PTH abundance is controlled via Pin1, CaSR, MAfB, PLC, and PKA signaling to regulate PTH expression, stability, and secretion.⁵⁶

Based on phosphoproteome analysis, that showed PKA activation and changes in KSRP phosphorylation in kidney failure parathyroids and our previous studies,²⁰ we chose to focus on the parathyroid KSRP-Pin1 axis and its effects on PTH expression in SHP. Both experimental CKD and AKI led to increased parathyroid Pin1 Ser16 and Ser71 phosphorylation. This hyperphosphorylation is consistent with the decreased Pin1 isomerase activity in other systems^17,18 and in parathyroids of SHP rats.²⁰ Pharmacologic inhibition of Pin1 by forskolin-induced PKA activation that would lead to Pin1 Ser16 phosphorylation,¹⁶ increased PTH expression both ex vivo in parathyroid glands in culture and in transfected cells. PKA inhibition, that would leave Pin1 unphosphorylated and active, had the opposite effect to decrease PTH mRNA levels. Forskolin has been shown to increase cellular cAMP and PTH release in dispersed primary bovine parathyroid cells.⁵⁷ Cholera toxin, which indirectly stimulates PKA, increased PTH mRNA levels and secretion in these cells.⁵⁸ The mechanism of increased PTH expression by PKA activation was not studied but may be due to activated PKA-induced Pin1 Ser16 phosphorylation and decreased Pin1 activity, as above.

Sulfopin is a highly selective Pin1 inhibitor that targets Cys113 in the catalytic domain and phenocopies Pin1^−/− mice phenotypes.⁵² Sulfopin had a similar effect as PKA activation to increase PTH expression. Sulfopin also increased PTH mRNA levels in organ culture, as measured in fluorescence-guided microdissected parathyroids from PT-YFP mice. Therefore, pharmacologic inhibition of Pin1 that mimics the decreased Pin1 activity in vivo in CKD SHP parathyroids leads to increased PTH expression in vitro in parathyroid glands and in transfected cells. These effects of Pin1 inhibition were dependent upon the conserved PTH mRNA 3′-UTR cis-acting protein-binding element,³⁶ confirming their effect on PTH mRNA stability. The effect of Pin1 inhibition to increase PTH expression is consistent with the increased serum PTH and parathyroid gland PTH mRNA and protein levels in Pin1^−/− mice.²⁰ Altogether these findings support a role for Pin1 inhibition to increased PTH expression. Of interest, in a human study of single nucleotide polymorphisms in the Pin1 gene promoter, Pin1 C667T genetic variants were associated with CKD SHP.⁵⁹ Analysis of global human PTH mRNA 3′-UTR–interacting proteins by MS confirmed the binding of KSRP, AUF1,^4,6,20 and other proteins that are involved in mRNA fate to PTH mRNA. Pin1 also bound the PTH mRNA 3'-UTR, which may be indirect through other RNA-binding proteins. Pin1 inhibition at either the WW or PPIase domain had a similar effect on PTH mRNA-binding protein profiles.

KSRP dephosphorylation is crucial for its mRNA decay–promoting capabilities.^20,22 KSRP overexpression that induces PTH mRNA decay in transfected cells²⁰ was more potent when KSRP phosphorylation null mutations at Pin1 potential target sites were introduced. The maximal effect on PTH mRNA decay was achieved by mutations at three Pin1 target motifs of KSRP: Ser181, Thr100, and Thr692. Multisite phosphorylation, where several residues on the same protein must all be phosphorylated for the change in protein activity to occur, is a common pathway to regulate protein function, activity, and interaction pattern.^60,61 Upon Toll-like receptor activation, Pin1 binds and isomerizes multiple IRAK1 protein pSer-Pro motifs that are required for Pin1 binding and activity.⁶² KSRP phosphorylation at multiple Pin1 target sites may similarly magnify the effect of Pin1 on KSRP and hence increase PTH mRNA levels in CKD. Pin1 pharmacologic inhibition failed to increase PTH expression when KSRP phosphorylation null mutants were present. Thus, CKD-induced Pin1 inhibition results in sustained KSRP phosphorylation at Pin1 target sites, as was also shown by phosphoproteome analysis, resulting in high PTH mRNA stability and levels.

In summary, experimental CKD induces global changes in expression and phosphorylation of parathyroid proteins. We propose that impaired renal function leads to parathyroid Pin1 phosphorylation, at least partly by PKA activation. The resulting decreased Pin1 activity increases PTH expression in a manner that is dependent on KSRP phosphorylation at Pin1 target residues and the PTH mRNA-protein-binding element (Figure 8E). This triggers the parathyroids to overproduce PTH through modified Pin1 and KSRP-PTH mRNA interaction and KSRP phosphorylation, contributing to SHP. This study demonstrates, for the first time, what elicits the increased PTH mRNA stability at the molecular level and may identify new potential targets for the prevention and management of CKD-induced SHP.

Disclosures

N. London reports consultancy with FoRx Therapeutics; reports ownership interest with Larkspur Biosciences and Totus Medicines; reports research funding from Ayala Pharmaceuticals, Monte Rosa Therapeutics, and Pfizer; reports patents or royalties from the Weizmann Institute of Science; and reports advisory or leadership role with Larkspur Biosciences, MetaboMed, Monte Rosa Therapeutics, and Totus Medicines. All remaining authors have nothing to disclose.

Funding

This work was supported by grants from the Israel Science Foundation (ISF 642/16 to T.N.-M.) and the United States-Israel Binational Science Foundation (BSF 2019300 to T.N.-M. and I.Z.B.-D.).

Author Contributions

I. Ben-Dov, A. Hassan, T. Naveh-Many, and M. Nechama conceptualized the study and were responsible for visualization; I. Ben-Dov, A. Hassan, and T. Naveh-Many were responsible for data curation; I. Ben-Dov, A. Hassan, R. Kilav-Levin, and Y. Pollak were responsible for investigation; A. Hassan and T. Naveh-Many were responsible for formal analysis and project administration; A. Hassan and N. London were responsible for methodology; A. Hassan, N. London, and T. Naveh-Many were responsible for validation; T. Naveh-Many was responsible for funding acquisition, resources, and supervision and wrote the original draft; and I. Ben-Dov, A. Hassan, T. Naveh-Many, and M. Nechama reviewed and edited the manuscript.

Data Sharing Statement

Original data reported in this paper of type Experimental Data have been deposited to PRoteomics IDEntifications database (PRIDE), under accession numbers PXD029368 (rat proteomics), PXD029401 (rat phosphoproteomics), and PXD029456 (human HEK293 proteins that bind the PTH mRNA).

Supplemental Material

This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2022020197/-/DCSupplemental.

Supplemental Methods.

Supplemental Table 1. Protein group differential expression in adenine versus control parathyroid glands (Proteosign 2.0 output) (Excel file).

Supplemental Table 2. Joined phosphoproteomic results from two experiments (Excel file).

Supplemental Figure 1. Bar plot showing enriched characteristics of proteins detected in rat parathyroid glands by MS, arranged by data source and P value.

Supplemental Figure 2. Phosphoproteomic results from control and experimental CKD parathyroids.

Supplemental Figure 3. No change in Pin1 protein levels by immunofluorescence analysis of thyroparathyroid sections from experimental CKD rats.

Supplemental Figure 4. Serum biochemistry and PTH levels in folic acid–induced acute kidney failure.

Supplemental Figure 5. Forskolin increases Pin1 Ser16 phosphorylation in HEK293 cells.

Supplemental Figure 6. PTH mRNA-binding protein pulldown MS results.

Supplemental Figure 7. PKA-activated Pin1 inhibition decreases Pin1-KSRP interaction.

Supplemental Figure 8. PKA activation increases PTH gene expression through KSRP phosphorylation and the PTH mRNA 3′-UTR ARE.