Identification of AUF1 as a Parathyroid Hormone mRNA 3′-Untranslated Region-binding Protein That Determines Parathyroid Hormone mRNA Stability*

Parathyroid hormone (PTH) mRNA levels are post-transcriptionally increased by hypocalcemia and decreased by hypophosphatemia, and this is mediated by cytosolic proteins binding to the PTH mRNA 3′-untranslated region (UTR). The same proteins are also present in other tissues, such as brain, but only in the parathyroid is their binding regulated by calcium and phosphate. The function of the PTH mRNA 3′-UTR-binding proteins was studied using an in vitro degradation assay. Competition for the parathyroid-binding proteins by excess unlabeled 3′-UTR destabilized the full-length PTH transcript in this assay, indicating that these proteins protect the RNA from RNase activity. The PTH RNA 3′-UTR-binding proteins were purified by RNA affinity chromatography of rat brain S-100 extracts. The eluate from the column was enriched in PTH RNA 3′-UTR binding activity. Addition of eluate to the in vitro degradation assay with parathyroid protein extracts stabilized the PTH transcript. A major band from the eluate at 50 kDa was sequenced and was identical to AU-rich binding protein (AUF1). Recombinant AUF1 bound the full-length PTH mRNA and the 3′-UTR. Added recombinant AUF1 also stabilized the PTH transcript in thein vitro degradation assay. Our results show that AUF1 is a protein that binds to the PTH mRNA 3′-UTR and stabilizes the PTH transcript.

Serum calcium is maintained within a narrow physiological range mainly due to the action of PTH, 1 which acts together with the biologically active metabolite of vitamin D, 1␣,25dihydroxyvitamin D (1). A seven-transmembranous calciumsensing receptor on the parathyroid (PT) cell recognizes small changes in serum ionized calcium to regulate PTH secretion (2). Low serum calcium increases not only PTH secretion, but also PTH mRNA levels (3) and, if prolonged, PT cell prolifera-tion (4). PTH then acts to correct serum calcium by mobilizing calcium from bone and renal reabsorption of calcium. Phosphate also regulates the PT, with low serum phosphate decreasing serum PTH, PTH mRNA levels, and parathyroid cell proliferation (5)(6)(7)(8). Patients with chronic renal failure are unable to excrete the large amounts of phosphorus in the diet. They develop severe complications due to high serum phosphate, including hyperparathyroidism with overactivity of the PT gland, bone pain, and increased mortality (9). There is therefore great interest in understanding the regulation of the PT by calcium and phosphate.
PTH gene expression is markedly increased by hypocalcemia and decreased by hypophosphatemia, and these effects in vivo are post-transcriptional (5,10). The PTH cDNA consists of three exons coding for the 5Ј-UTR (exon I), the prepro region of PTH (exon II), and the structural hormone together with the 3Ј-UTR (exon III) (11,12). The rat 3Ј-UTR is 239 nt long out of the 712 nt of the full-length PTH RNA (12). We have shown that cytosolic proteins from parathyroids bind to the 3Ј-UTR of the rat PTH mRNA and regulate mRNA stability (10). PT proteins from hypocalcemic rats show increased binding to the PTH mRNA 3Ј-UTR by mobility shift and UV cross-linking assays, and this protein-RNA binding is decreased with hypophosphatemic PT proteins. Thus, the level of protein-RNA binding directly correlates with PTH mRNA levels. Since there is no PT cell line, an in vitro PTH RNA stability assay was utilized. This assay showed stabilization of the transcript by hypocalcemic proteins and marked instability with hypophosphatemic proteins (10). These studies indicate that there are instability regions in the PTH mRNA 3Ј-UTR that are protected by RNA-binding proteins. This protein-RNA interaction determines PTH mRNA stability.
In the present study, we have isolated one of the proteins that bind to the PTH mRNA 3Ј-UTR by affinity chromatography and characterized its functional role in stabilizing the PTH mRNA. Addition of recombinant AUF1 stabilized the PTH transcript in an in vitro degradation assay with PT proteins. These results suggest an important function for AUF1 in the transduction of changes in serum calcium and phosphate to the stability of the PTH mRNA.
Isolation and Identification of the 50-kDa Protein-S100 extracts were prepared from rat brain tissue. The tissue was removed from the rat under pentobarbital anesthesia and immediately washed in phosphate-buffered saline buffer at 4°C. The tissue was homogenized with a Polytron in one volume of S100 buffer (50 mM Tris, pH 7.5, 25% glycerol, 100 mM KCl, 0.1 mM EDTA, 0.5 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride). The homogenate was centrifuged at 12,000 ϫ g for 15 min at 4°C and the supernatant was centrifuged again at 100,000 ϫ g for 1 h (Beckman type TL-100) at 4°C. The high speed supernatant (S100) was stored at Ϫ80°C until it was used for protein purification and binding assays.
Heparin-Sepharose (6 g) (Amersham Pharmacia Biotech) was used to prepare a 25-ml bed volume column. The heparin-Sepharose column was washed with 250 ml of buffer B (50 mM Tris, pH 7.8, 2 mM EDTA, 5% glycerol, 7 mM ␤-mercaptoethanol) containing 0.1 M NaCl. S100 brain tissue extract from 20 rats (300 mg) was applied to the column (twice). The column was washed with 550 ml of buffer B containing NaCl (0.1 M), and the bound proteins were eluted from the column by a step gradient of buffer B containing increasing NaCl concentrations (0.1-1 M).
The fractions were assayed for binding to the PTH 3Ј-UTR by UV cross-linking. Fractions that showed maximal binding to the PTH 3Ј-UTR eluted at 230 -550 mM NaCl and were pooled. The pooled fractions were then loaded on a CNBr-activated Sepharose column bound to 200 g of PTH mRNA 3Ј-UTR that had been synthesized in vitro. The column was washed with buffer B containing 0.1 M NaCl and the fractions were eluted with increasing NaCl concentrations (0.1-1 M) and assayed by a UV cross-linking assay. Fractions that showed maximal binding were pooled and concentrated using a Centricon 30 filter (Amicon, Beverly, MA). A sample was used to identify the RNA-binding proteins by Northwestern analysis with PTH 3Ј-UTR as a labeled probe. The pooled fractions were run on a preparative polyacrylamide gradient gel (7-12%) and stained with Coomassie Blue. A 50-kDa band was excised from the gel, degraded with the endoprotease LysC, and the peptide products were analyzed by high performance liquid chromatography. Five peptides were microsequenced by Edman degradation.
UV Cross-linking Assay-UV cross-linking assay was performed using 10 g of either brain S100 extracts or microdissected PT extracts as described previously. To measure binding activity of eluates from the RNA column, salt concentrations were normalized (as measured by conductivity). Equivalent aliquots of each fraction were tested. In some experiments, recombinant AUF1 protein was used. The proteins were incubated with 32 P-labeled RNA for the full-length or the 3Ј-UTR of the PTH cDNA. After UV cross-linking, the samples were digested by RNase A, fractionated by SDS-polyacrylamide gel electrophoresis, and autoradiographed as described previously (10).
RNA Electrophoretic Mobility Shift Assays (REMSA)-The PTH 3Ј-UTR RNA probe (5000 cpm) was incubated with microdissected PT or brain extracts (10 g) or rAUF1 as indicated in a final volume of 10 l containing 10 mM Tris, pH 7.5, 0.1 M potassium acetate, 5 mM magnesium acetate, 2 mM DTT, 8 units of RNasin, 2 g of tRNA, 50 g of heparin, and 1 g of BSA for 10 min at 4°C. The samples were run on a native polyacrylamide gel (4% polyacrylamide:bisacrylamide; 70:1) in a cold room. RNA-protein binding was visualized by autoradiography of the dried gel.
RNA Transcripts and Probes-Labeled and unlabeled RNA was transcribed from linearized plasmids using an RNA production kit (Promega, WI) and the appropriate RNA polymerases. A linearized plasmid construct containing the full-length PTH cDNA in Bluescript KS (In-vitrogen) was used as described previously (10). For the 3Ј-UTR of the PTH cDNA, a PCR product (10) was subcloned into PCRII (Invitrogen, CA) and used for transcribing labeled and unlabeled RNA. The RNA was used for the affinity column, Northwestern analysis, UV crosslinking, REMSA, and in vitro degradation assays. The 63-nt transcript containing the protein binding sequences in the 3Ј-UTR (10) was transcribed from a PCR product of the cDNA subcloned into PCRII (Invitrogen). The sequence of the 63-nt transcript is: 5Ј-GTCTCTTCCAAT-GATTCCATTTCAATATATTCTTCTTTTTAAAGTATTACACATTTCCA -CTTC-3Ј.
In Vitro RNA Degradation Assay-Preparation of S100 parathyroid protein extracts for the RNA degradation assay and the assay itself were performed as before (10) with minor modifications. Transcripts (0.2 ϫ 10 6 cpm) were incubated in 10 g of cytoplasmic extracts and 80 units/ml RNasin (Promega, WI), and at timed intervals samples were removed and extracted by TRI reagent (Molecular Research Center). In some experiments recombinant AUF1, eluate from the RNA column, or Dynein light chain (M r 8000) (LC8) (6.0 g), or BSA (6.0 g) were added together with the cytosolic extracts. In some experiments excess of cold RNA was added. Transcript half-lives were calculated by densitometric imaging of the autoradiographs. The data were plotted and the point where there was a 50% reduction in intact transcript determined.
Preparation of Recombinant p40 AUF1 -Recombinant p40 AUF1 was prepared according to Wilson and Brewer (14) with some modifications. An Escherichia coli DH5␣ clone containing pTrcHisB/p40 AUF1 was induced to express plasmid encoded protein by culturing with 1 mM isopropyl-␤-D-thiogalactopyranoside (MBI, Fermentas). His 6 -AUF1 fusion polypeptide was purified by resuspending the bacterial pellet with HNTA buffer (1 M NaCl, 50 mM NaPO 4 buffer, pH 7.8, 1% Triton X-100, 10 g/ml pepstatin A, 10 g/ml leupeptin, and 0.1 mM phenylmethylsulfonyl fluoride), sonication and centrifugation at 10,000 rpm for 20 min at 4°C. The supernatant was added to 2 ml of ProBond resin (Invitrogen) that had been prewashed twice with double distilled water and once with NTA buffer (300 mM NaCl, 50 mM NaPO 4 buffer, pH 7.8, 1% Triton X-100, 10 g/ml pepstatin A, 10 g/ml leupeptin, 0.1 mM phenylmethylsulfonyl fluoride) and rotated at 4°C for 1 h. The beads were spun down and washed with NTA buffer, and the His 6 -AUF1 protein was eluted with increasing concentration of imidazole (25-300 mM) in NTA buffer. Lysozyme (Sigma) was added at a final concentration of 100 g/ml to the eluates containing the His 6 AUF1 as the main polypeptide, and dialyzed against 100 volumes of 10 mM Tris HCl pH 7.5, 4°C for 5 h. The eluates were then concentrated by Centricon 30 (Amicon) and the concentration of purified rAUF1 was determined by comparison with known amounts of BSA on Coomassie-stained SDSpolyacrylamide gels.
The figures shown are representative gels of at least three repeat experiments.

RESULTS
Excess PTH RNA 3Ј-UTR Leads to Rapid Decay of the Fulllength PTH RNA in Vitro-PTH mRNA levels are determined by RNA-binding proteins that interact with sequences in the 3Ј-UTR of the PTH mRNA. When there is less binding, then the RNA is more accessible to RNase activity (10). Competition for the binding proteins with excess PTH mRNA 3Ј-UTR should lead to more rapid decay in an in vitro degradation assay of PTH RNA transcript and PT cytosolic extract. Excess PTH RNA 3Ј-UTR led to a dramatic decrease in PTH RNA stability (t 1/2 10 min, compared to 60 min with no competitor) (Fig. 1). Excess of a smaller transcript of 63 nt which contains the protein binding sequences in the 3Ј-UTR (10) also led to rapid degradation of the full-length, radiolabeled PTH RNA (t 1/2 Ͻ10 min) (Fig. 1). Addition of unlabeled PTH transcript that did not include the 3Ј-UTR as a control did not lead to a decrease in PTH RNA stability (Fig. 1). In fact, there was a slight increase in PTH RNA stability, which is not readily explainable. These results suggest that the binding region in the 3Ј-UTR competed for stabilizing factors in the PT protein extract, resulting in their depletion from the extract. This resulted in rapid degradation of the PTH transcript. We therefore performed experiments to characterize the binding proteins.
Purification of the PTH mRNA 3Ј-UTR-binding Proteins-To identify the proteins that bind to the PTH mRNA 3Ј-UTR, we performed affinity chromatography. The proteins that bind the PTH mRNA 3Ј-UTR are present in all tissues examined (10). We therefore used rat brain protein extracts and not the minute parathyroids as the source for the RNA-binding proteins. Rat brain S-100 extracts were chromatographed first on a heparin-Sepharose column to enrich for proteins that bind RNA. The fractions that showed maximum binding to the PTH 3Ј-UTR on UV cross-linking were then chromatographed on a PTH RNA affinity column. The affinity column consisted of cyanogen bromide-activated Sepharose linked to in vitro transcribed PTH RNA 3Ј-UTR. The proteins that bound the 3Ј-UTR column were eluted with increasing salt concentrations and studied by UV cross-linking to the PTH 3Ј-UTR RNA probe ( Fig. 2A). There were three protein-RNA bands, at about 50, 60, and 110 kDa, for brain and parathyroid, consistent with our earlier studies (10). The proteins that eluted between 220 and 500 mM NaCl exhibited maximum binding ( Fig. 2A) and were combined and concentrated. A sample was run on an SDSpolyacrylamide gel and transferred to a nitrocellulose membrane, which was stained for protein by Ponceau. The staining revealed several bands (Fig. 2B). To identify the RNA-binding proteins, the membrane was then incubated with a riboprobe for the PTH 3Ј-UTR for Northwestern analysis (Fig. 2B). The PTH 3Ј-UTR showed the most intense binding to three of the proteins. There was one protein at approximately 60 kDA, two at about 50 kDa, and other less intense bands.
PTH mRNA 3Ј-UTR-binding Proteins Stabilize the PTH RNA Transcript in an in Vitro Degradation Assay with Parathyroid Proteins-To demonstrate the function of the PTH mRNA 3Ј-UTR-binding proteins on PTH mRNA stability, we performed in vitro degradation assays. In the presence of cytosol, there is gradual degradation of the transcript as seen previously in Fig. 1. We measured the effect of added eluate from the RNA column on the ability of PT protein extracts from hypophosphatemic rats to degrade PTH RNA in vitro. Hypophosphatemic PT proteins showed more rapid degradation of PTH RNA in an in vitro degradation assay compared with PT proteins of control rats and also less binding to the PTH mRNA 3Ј-UTR (10). Proteins from hypophosphatemic PTs are therefore depleted in stabilizing factors. Complete depletion of these factors from PT cytosolic proteins by 3Ј-UTR affinity chromatography is not practicable because of the small size of the rat PT gland, which would require the use of Ͼ150 rats for each experiment. The degradation assay was therefore performed with PT proteins from hypophosphatemic rats and increasing amounts of eluate (ϳ200 and 400 ng of protein) from the RNA column. The added eluate had no effect upon transcript stability at lower concentrations; however, higher concentrations of added eluate stabilized the PTH transcript throughout the experiment (t 1/2 80 min, compared to 30 min with no eluant) (Fig. 3). The same stabilizing effect was also found when the eluate was added to PT cytosolic extracts from control rats (data not shown). These results show that proteins eluted from the RNA column stabilize the PTH transcript in vitro and that the eluate can overcome the degrading effect of the PT proteins from low phosphate rats, whose PT proteins show decreased binding to the PTH 3Ј-UTR.
One of the PTH mRNA 3Ј-UTR-binding Proteins Is AUF1-The eluate from the RNA column contained several proteins. One of the proteins at 50 kDa was present in the highest concentration (Fig. 2B). For this reason the 50-kDa protein was gel-purified and microsequenced generating five peptide sequences of 10 -17 residues each. Data base search identified the polypeptide as being identical to AU-rich binding protein (AUF1), which is known to be important to the half-life of other mRNAs (reviewed in Ref. 15). The peptide sequences did not identify which of the AUF1 isoforms had been isolated. However, the binding assays suggests that the PTH RNA 3Ј-UTR bound all isoforms. One of these isoforms, p40 AUF1 , was further studied.
The binding of recombinant AUF1 to the PTH RNA 3Ј-UTR was demonstrated by REMSA. Recombinant p40 AUF1 bound the PTH 3Ј-UTR labeled transcript resulting in a shift of the RNA probe (Fig. 4). The binding was enhanced by increasing concentrations of recombinant p40 AUF1 (Fig. 4). Without protein the labeled transcript ran as two bands. These two bands may represent secondary structures of the RNA molecules because denaturing the RNA by heating it to 80°C and then allowing it to renature at room temperature resulted in a single band on a polyacrylamide gel. This renatured probe bound p40 AUF1 the same as the untreated transcript (data not shown). This indicates that p40 AUF1 alone can bind the PTH 3Ј-UTR in the absence of other cytosolic proteins.
AUF1 Stabilizes the PTH RNA Transcript in an in Vitro Degradation Assay with Parathyroid Proteins-To demonstrate the function of AUF1 in PTH mRNA stability we performed in vitro degradation assays. The degradation assay was performed with PT proteins from hypophosphatemic rats and increasing amounts of p40 AUF1 . When recombinant p40 AUF1 was added to the degradation assay in the presence of hypophosphatemic PT proteins, there was stabilization of the PTH transcript, which was dependent upon the amount of recombinant p40 AUF1 added (Fig. 5A). At 50 ng, added AUF1 had no effect upon transcript stability; however, at higher concentrations addition of AUF1 stabilized the PTH transcript throughout the experiment (t 1/2 of 90 min with AUF1, and 30 min without AUF1) (Fig. 5A). Fig. 5B shows the degradation of the PTH transcript with PT proteins from both normal and FIG. 1. Competitor PTH RNA 3-UTR accelerates PTH RNA degradation in vitro. The full-length, radiolabeled PTH mRNA transcript was incubated with cytosolic parathyroid protein extracts (10 g) from rats fed a normal diet. At timed intervals samples were extracted, run on agarose gels, and autoradiographed to measure the intact transcript remaining. The degradation assay was performed in the presence of PT extracts without competitor, with 25 ng of unlabeled PTH RNA 3Ј-UTR, with unlabeled PTH RNA excluding the 3Ј-UTR (without 3Ј-UTR), or with a 63-nt unlabeled transcript comprising the sequences necessary for binding of proteins to the PTH RNA 3Ј-UTR. Excess 3Ј-UTR or the 63-nt transcripts led to accelerated decay of the full-length PTH transcript by the PT proteins. The competitor transcript that did not contain the 3Ј-UTR had no effect. hypophosphatemic rats, where there is more rapid degradation with hypophosphatemic parathyroid proteins (t 1/2 of 30 min with hypophosphatemic PT proteins, and 60 min with normal PT proteins). Addition of p40 AUF1 to the hypophosphatemic proteins stabilized the transcript (t 1/2 Ͼ120 min) even more than when the degradation assay was performed in the presence of proteins from normal rats. Control proteins had no effect (Fig. 5B). The control proteins used were BSA and LC8. LC8 also binds to the PTH mRNA 3Ј-UTR (16).
To understand the effect of AUF1 and the other proteins in the eluate on the degradation reaction, we added recombinant p40 AUF1 and the eluate to the reaction with hypophosphatemic proteins in concentrations where alone they had no effect on PTH RNA degradation (Figs. 3 and 5A). The PTH RNA was now markedly stabilized (Fig. 5A). Therefore, there is an addi-tive effect of p40 AUF1 and the RNA-binding proteins eluted from the RNA column.

DISCUSSION
The post-transcriptional regulation of PTH gene expression by calcium and phosphate is mediated by the binding of cytosolic proteins to the PTH mRNA 3Ј-UTR (10). We have now identified one of the proteins that bind to the PTH mRNA 3Ј-UTR by RNA affinity chromatography and microsequencing as AUF1. Recombinant p40 AUF1 bound the PTH mRNA 3Ј-UTR by REMSA. There is no PT cell line; therefore, to demonstrate that AUF1 has a functional role in determining PTH mRNA stability, we performed in vitro degradation assays. Addition of p40 AUF1 , with PT cytosolic extracts, stabilized the PTH transcript. Surprisingly, immunodepletion of rAUF1 from PT cytosolic extracts had no little or no effect on degradation of the transcript (data not shown). This result may reflect functional FIG. 2. Purification of the PTH RNA 3-UTR-binding proteins by affinity chromatography. A, identification by UV cross-linking of eluates from a PTH RNA 3Ј-UTR affinity column. The proteins that bound the 3Ј-UTR were eluted with increasing salt concentrations, and binding to the 3Ј-UTR was examined by UV cross-linking. The fractions that showed maximal binding were eluted at NaCl concentrations of 230 -550 nM.
The arrows indicate the three RNA-protein bands that are also present when parathyroid proteins are studied for binding. Molecular weight markers are indicated on the right. B, Northwestern blot of proteins from concentrated fractions eluted from the affinity columns identified a 50-kDa protein that bound the PTH 3Ј-UTR. A sample of the proteins from the combined positive fractions from Fig. 1 was run on an SDS-polyacrylamide gel and then transferred to a nitrocellulose membrane. The membrane was first stained with Ponceau (left) to identify protein bands, and then the membrane was incubated with radiolabeled PTH 3Ј-UTR for Northwestern analysis (right). There was prominent binding of labeled RNA to several bands, including two proteins of ϳ50 kDa.
FIG. 3. Stabilizing effect of eluate from the RNA column on the degradation in vitro of PTH mRNA by hypophosphatemic rat parathyroid proteins. The full-length radiolabeled PTH mRNA transcript was incubated with cytosolic parathyroid protein extracts (10 g) from hypophosphatemic rats without or with the addition of 200 and 400 ng of protein of the eluate from the RNA column. The proteins used were eluted from an RNA column as in Fig. 2 at a 250 nM salt concentration. At timed intervals samples were extracted, run on agarose gels, and autoradiographed to measure the intact transcript remaining. There was a dose-dependent stabilization with added eluate.

FIG. 4. Recombinant p40 AUF1 binds PTH 3-UTR by REMSA.
Increasing concentrations of recombinant p40 AUF1 (AUF1) resulted in a shift of the PTH 3Ј-UTR transcript, which without protein ran as two bands. We believe that the two bands represent secondary structure within some of the substrate molecules. redundancy in the PT cytosolic extracts, in that only AUF1 was depleted and not the other PTH mRNA-binding proteins. In hypophosphatemia there are decreased binding in three parathyroid protein-PTH RNA species, and a less stable transcript. This suggests that all three protein-RNA species seen by UV cross-linking are involved in determining the RNA stability. To deplete all the PTH RNA 3Ј-UTR-binding proteins from the PT cytosolic extract, we added excess PTH RNA 3Ј-UTR, or a smaller transcript of 63 nt that is sufficient for binding, to the degradation assay of PTH RNA with PT proteins (Fig. 1). This resulted in a rapid degradation of the PTH RNA, suggesting competition for the stabilizing proteins. Polysomal associated in vitro degradation of granulocyte-macrophage colony-stimulating factor (GM-CSF) mRNA was similarly enhanced by AUUUA-containing competitor RNA (17,18). The addition of the competitor RNA depleted the polysome associated AUbinding protein, which stabilized the GM-CSF mRNA by masking AUUUA motifs.
The short half-life of many proto-oncogenes and cytokines is mediated in part by the rapid turnover of their mRNAs (15). AϩU-rich elements in the 3Ј-untranslated regions of these mRNAs serve as one recognition signal targeting the mRNAs for rapid degradation. AUF1 is a cytosolic protein that both binds to the proto-oncogene c-myc AϩU-rich element (ARE) and specifically destabilized c-myc mRNA in a cell-free mRNA decay system, which reconstituted mRNA decay processes found in cells (19). Moreover, the ARE binding affinities of AUF1 correlated with the potency of an ARE to direct degradation of a heterologous mRNA (20). These studies established a role for AUF1 in ARE-directed mRNA decay that is based upon its affinity for different AREs.
The role of AUF1 in mRNA decay is not restricted to protooncogenes. The developmental immaturity of neonatal phagocytic function is associated with a shorter half-life of GM-CSF mRNA. In vitro the decay of the GM-CSF in mononuclear cells is also unstable, and this instability was accelerated by protein fractions enriched for AUF1 (21). Moreover, this accelerated ARE-dependent decay of the GM-CSF 3Ј-UTR was attenuated by immunodepletion of AUF1, thereby demonstrating that the in vitro RNA decay is ARE-and AUF1-dependent (21). Anti-AUF1 immunoblotting showed significantly higher levels of two AUF1 protein isoforms and lower levels of one in cord than in adult mononuclear cell extracts. The results suggested that increased levels of specific AUF1 isoforms in cord mononuclear cells destabilized the GM-CSF mRNA by targeting it for increased degradation (22). The ARE destabilizing function in K562 cells was dramatically impeded during hemin-induced erythroid differentiation (23). Ectopic expression of hnRNP D/AUF1 in hemin-treated K562 cells restored the rapid decay directed by the ARE. The extent of destabilizing effect varied among the four isoforms of AUF1, with p37 and p42 displaying the most profound effect. These results demonstrated a specific cytoplasmic function for AUF1 as an RNA-destabilizing protein in ARE-mediated decay pathway.
In contrast to the role of AUF1 in the rapid degradation of mRNAs, it may have a role in the stabilization of other mRNAs, such as ␣-globin mRNA. Kiledjian et al. (24) identified AUF1 as one of the proteins that, together with two other proteins, ␣ CP1 and ␣ CP2, bind to the 3Ј-UTR of the ␣-globin mRNA and regulate the erythrocyte-specific accumulation of ␣-globin mRNA. Alone, none of these proteins can bind the ␣-globin 3Ј-UTR, and they only bind when they are complexed with the other proteins of the ␣-complex. However, Chkheidze et al. (25) showed that in ␣-globin 3Ј-UTR the poly(C)-binding proteins, ␣ CPs, were quantitatively incorporated into the ␣-complex in the absence of AUF1, suggesting that AUF1 may not be essential for the protein-RNA complex formation.
We have now shown that AUF1 binds PTH mRNA 3Ј-UTR FIG. 5. Stabilizing effect of p40 AUF1 on the degradation in vitro of PTH mRNA by hypophosphatemic rat parathyroid proteins. The full-length radiolabeled PTH mRNA was incubated with cytosolic parathyroid protein extracts (10 g) from hypophosphatemic rats and at timed intervals samples were extracted, run on agarose gels and autoradiographed to measure the intact transcript remaining. A, degradation in the presence of increasing doses of recombinant p40 AUF1 . p40 AUF1 stabilized the PTH transcript dose-dependently. Addition of eluate (200 ng), prepared as in Fig. 3, together with 10 ng of p40 AUF1 stabilized the PTH transcript at doses that alone had no effect. B, degradation with PT proteins from normal and hypophosphatemic (ϪP) rats, without and with added recombinant p40 AUF1 (200 ng), or BSA (6 g), or LC8 (6 g). Recombinant p40 AUF1 , but not BSA or dynein light chain (LC8) stabilized the PTH transcript. and determines PTH mRNA stability. The PTH mRNA 3Ј-UTR has a region that is rich in A and U but does not have the classical ARE configuration. The PTH mRNA ARE is an example of a regulatory element, which is stabilized by AUF1 and other parathyroid cytosolic RNA-binding proteins. RNA-protein binding regulates PTH mRNA levels in response to changes in serum calcium and phosphate. The role of AUF1 in the regulation of PTH mRNA stability in response to changes in serum calcium and phosphate remains to be determined.