Nuclear Localization of the Type 1 PTH/PTHrP Receptorin Rat Tissues

P.H. WATSON,1 L.J. FRAHER,1 G.N. HENDY,2 U.-I. CHUNG,3 M. KISIEL,1 B.V. NATALE,1

and A.B. HODSMAN1

ABSTRACT

The localization of PTH/PTH-related peptide (PTHrP) receptor (PTHR) has traditionally been performed byautoradiography. Specific polyclonal antibodies to peptides unique to the PTHR are now available, whichallow a more precise localization of the receptor in cells and tissues. We optimized the IHC procedure for therat PTHR using 5-mm sections of paraffin-embedded rat kidney, liver, small intestine, uterus, and ovary.Adjacent sections were analyzed for the presence of PTHR mRNA (by in situ hybridization) and PTHrPpeptide. A typical pattern of staining for both receptor protein and mRNA was observed in kidney in cellslining the proximal tubules and collecting ducts. In uterus and gut, the receptor and its mRNA are present insmooth muscle layers (PTHrP target) and in glandular cuboidal cells and surface columnar epithelium. Thissuggests that PTH, or more likely PTHrP, plays a role in surface/secretory epithelia that is as yet undefined.In the ovary, PTHR was readily detectable in the thecal layer of large antral follicles and oocytes, and waspresent in the cytoplasm and/or nucleus of granulosa cells, regions that also contained receptor transcripts.PTHR protein and mRNA were found in the liver in large hepatocytes radiating outward from central veins.Immunoreactive cells were also present around the periphery of the liver but not within two or three cell layersof the surface. Clear nuclear localization of the receptor protein was present in liver cells in addition to theexpected cytoplasmic/peripheral staining. PTHR immunoreactivity was present in the nucleus of some cells inevery tissue examined. RT-PCR confirmed the presence of PTHR transcripts in these same tissues. Exami-nation of the hindlimbs of PTHR gene-ablated mice showed no reaction to this antibody, whereas hindlimbsfrom their wild-type littermates stained positively. The results emphasize that the PTHR is highly expressedin diverse tissues and, in addition, show that the receptor protein itself can be localized to the cell nucleus.Nuclear localization of the receptor suggests that there is a role for PTH and/or PTHrP in the regulation ofnuclear events, either on the physical environment (nucleoskeleton) or directly on gene expression. (J BoneMiner Res 2000;15:1033–1044)

Key words: PTH/PTHrP receptor, immunohistochemistry, PTHrP, nuclear localization

INTRODUCTION

THE PTH/PTHRP receptor (PTHR) has a wide tissuedistribution based on analysis of mRNA expression by

Northern blotting.(1,2) Since the receptor was first cloned,

expression of its mRNA has been shown in many “nonclas-sical” PTH target tissues such as liver, lung, intestine,breast, uterus, testis, and brain.(1–9)However, little is knownabout the localization of the receptor protein itself, becausemost studies to date have been limited to radioligand label-

1Departments of Medicine and Biochemistry, University of Western Ontario, and The Lawson Research Institute, London, Ontario,Canada.

2Calcium Research Laboratories, Royal Victoria Hospital, and Department of Medicine, McGill University, Montreal, Quebec, Canada.3Endocrine Unit, Massachusetts General Hospital, and Harvard Medical School, Boston, Massachusetts, U.S.A.

JOURNAL OF BONE AND MINERAL RESEARCHVolume 15, Number 6, 2000© 2000 American Society for Bone and Mineral Research

1033

ing of cell preparations and tissues, a method that does notallow for fine discrimination of the receptor’s positionwithin a tissue or cell. A few studies have employed anti-bodies raised against various extracellular portions of thePTHR to localize the receptor in developing bone(11) andkidney(12) by immunohistochemistry.

Knowledge of where a receptor is in a given tissue cangive important insight into its potential role in that tissue.Little is known of the cellular localization of the PTHR intissues other than bone and kidney. Within the kidney, thereceptor is present in epithelia lining the proximal convo-luted tubules, and to some extent the distal convolutedtubules and cortical collecting ducts.(12–15)This localizationis essential to the calcium-conserving role of PTH. Re-cently, PTHR transcripts were shown in podocytes withinthe glomerulus.(12,13) In bone of fetal and young rats thePTHR is found in cells of the osteoblast and chondro-cyte lineages,(16) where a role for both PTH and PTHrP(in chondrocyte proliferation and maturation) has beenshown.(17)

Several species- and region-specific antibodies to thePTHR have recently become available, which are suitablefor immunohistochemical studies and have been used toidentify PTHRs in chondrocytes.(11) In this report, we showthe tissue and cellular distribution of the PTHR protein inkidney (a classical PTH target) and in liver, small intestine,uterus, and ovary, the latter of which are most likely targetsfor PTHrP.

MATERIALS AND METHODS

Animals

Tissues were removed from 3-month-old, random-cycling female Sprague-Dawley rats (Charles River, St.Constance, Quebec, Canada). Animals were killed by CO2

inhalation and tissues were rapidly removed and fixed inice-cold 4% paraformaldehyde–0.2% glutaraldehyde. Fixa-tion was carried out overnight at 4°C after which tissueswere cleared to 70% ethanol. Cleared tissues were embed-ded in paraffin and 5-mm sections cut and mounted onSuperfrost slides (Fisher, Nepean, Ontario, Canada).

Immunohistochemistry (IHC)

IHC was performed as described.(18) Briefly, tissue sec-tions were postfixed in 4% paraformaldehyde for 15 min-utes, followed by a 10-minute incubation in 10mg/ml pro-teinase K (Gibco, Burlington, Ontario, Canada) at roomtemperature to retrieve fixation-concealed antigens. Sec-tions were stabilized in 4% paraformaldehyde for a further15 minutes before proceeding with the IHC procedure usingthe Vectastain ABC kit (Vector Labs, Burlington, Ontario,Canada) following the manufacturer’s directions. Poly-clonal rabbit anti-rat PTHR antibodies were used: antibodyPTH-II raised against an epitope in the amino terminalextracellular domain (ESKENKDVPTGSRRRGR; aa 91–106) and PTH-IV raised against a portion of the secondextracellular loop (TLDEARLTEEELH; aa 249–262) (Babco,Berkeley, CA, U.S.A.).(19) The PTHrP antibody used was amouse monoclonal raised against amino acids 34–53 of

human PTHrP (Oncogene Research Products, Cambridge,MA, U.S.A.). Primary antibodies were diluted to 100mg/mlin 1% BSA-PBS. As a control, some sections were treatedwith PTHR primary antibody that had been preabsorbedovernight at 4°C with the appropriate peptide. For thePTHrP antiserum, some sections were treated with normalmouse serum instead of primary antibody. As a furthercontrol to ensure the specificity of the PTHR immunostain-ing, sections of hindlimbs from PTHR gene-ablated mice[PTHR (2/2)] and their wild-type littermates [PTHR (1/1)](10,11)were processed with the anti-PTHR antibodies asdescribed earlier. All sections were counterstained withCarazzi’s hematoxylin before dehydration and mounting.

Western and ligand blotting

Confirmation of PTHR immunoreactivity in the nucleuswas obtained after subcellular fractionation of rat liver andcultured mouse osteoblasts (MC3T3-E1 cells), SDS-PAGE,and Western blotting with the PTH II antiserum describedearlier. Briefly, cellss/tissues were homogenized in 3 vol-umes of buffer (20 mM Tris, 330 mM sucrose, 2 mMEDTA, 0.5 mM EGTA, 1% NP-40, 2 mM PMSF, and 0.02TIU/ml aprotinin). Nuclei were collected after a 10-minutecentrifugation at 1000g, washed twice in homogenizationbuffer, resuspended in homogenization buffer, and soni-cated to disrupt nuclear structure. The supernatant wassonicated and then subjected to ultracentrifugation at100,000g for 30 minutes. The supernatant (cytosol fraction)was collected and stored at220°C. The pellet (cell mem-branes) was washed twice in homogenization buffer, resus-pended as described earlier, and sonicated to uniformity.Protein concentration was determined using the Lowrymethod.(20) Each protein lysate (30mg) was separated bySDS-PAGE and transferred to nitrocellulose following stan-dard procedures.(21) Blots were blocked in 5% nonfat drymilk overnight followed by a 1-h incubation in primaryantibody to the PTHR (1/100; PTH II-Babco) and thendeveloped using the Amersham ECL kit (Amersham Phar-macia Biotech, Baie d’Urte, Quebec, Canada) and exposedto X-ray film for 5 minutes. A prominent band at theexpected molecular weight of 66.7 kDa for the PTHR waspresent in all three cell fractions. Biotin-labeled hPTH(1–34) (Peninsula Labs, Belmont, CA, U.S.A.) was preparedusing the Amersham Protein Biotinylation kit following themanufacturer’s directions and was used at a concentrationof 10mg/ml to probe identical blots to the Westerns. Ligandblots were developed with streptavidin-HRP and the Amer-sham ECL kit. As a further test of the specificity of ourresults, 30mg/lane of MC3T3-E1 cell nuclear fraction wasseparated by SDS-PAGE and blotted as described earlier.This blot was cut into strips and each strip probed with 1mg/ml biotin-hPTH(1–34) and either 0, 2.5, 10, or 100mg/ml of unlabeled hPTH(1–34).

In situ hybridization (ISH)

In situ hybridization was carried out using biotin-labeledsense and antisense riboprobes and the Genpoint CSA kit(Dako Diagnostics, Mississauga, Ontario, Canada). Labeledriboprobes were prepared from a plasmid containing a

1034 WATSON ET AL.

2-kilobase cDNA corresponding to rat PTHR mRNA(9,12)

(using T3 and T7 RNA polymerases; Gibco, Burlington,Ontario, Canada) and a nucleotide labeling mix containingbiotin-16-UTP (Boerhinger-Mannheim, Burlington, On-tario, Canada). Tissue sections were deparaffinized and re-hydrated in a standard xylene and alcohol series and ISHperformed according to the manufacturer’s directions.Briefly, slides were heated in Target Retrieval Solution andproteinase K to reveal hidden antigens. Endogenous perox-idase activity was quenched in 0.3% hydrogen peroxide inmethanol. Sections were hybridized for 2 h at50°C with 5ng/ml biotin-labeled riboprobe in the supplied RNA hybrid-ization buffer. After stringent washing, sections weretreated to successive incubations in primary streptavidin–horseradish peroxidase, biotinyl-tyramide solution, and sec-ondary streptavidin–horseradish peroxidase (Dako) for 15minutes each at room temperature. Color was developedwith the supplied diaminobenzidine diluted as directed(Dako) and sections counterstained with Carazzi’s hema-toxylin before dehydration and mounting.

RESULTS

Expression of PTHR and PTHrP in liver

In the rat liver, the PTHR was localized to hepatocytes ina pattern radiating outward from the hepatic veins (Figs. 1Aand 1B) and in large hepatocytes within a few cell layers of,but not at, the periphery. No staining of Kupffer cells wasevident. Nuclear staining was present in some, but not all, ofthe positive hepatocytes. Preabsorption of the antibody(PTH-II, see Materials and Methods) with the peptideagainst which it was raised abolished all specific staining forthe receptor in rat liver (Fig. 1C). Messenger RNA for thePTHR was present in the same cells in the liver as theprotein (Figs. 1D and 1E). Hybridization with the PTHRsense probe did not result in any specific staining (Fig. 1F).Staining for the PTHrP peptide was also observed in apattern of hepatocytes radiating outward from hepatic veins,although the staining was much less intense than that for thereceptor (Figs. 1G and 1H). PTHrP was also localized to thenuclei of some hepatocytes and Kupffer cells (Fig. 1H).Substitution of normal mouse serum for the primary anti-body abolished all specific staining (Fig. 1J). The samecontrols for each IHC and the ISH procedure were per-formed (as shown in Figs. 1C, 1J, and 1F, respectively) forall the following tissues with similar results. (To betterpresent the positive data, therefore, the negative controls arenot included in Figs. 2–5.) With respect to immunohisto-chemical analysis with the anti-PTHR antibodies, only datawith antibody PTH-II are shown. An identical pattern ofimmunolocalization was obtained with antibody PTH-IVfor all tissues examined (data not shown).

Expression of PTHR and PTHrP in kidney

Figure 2 shows the results of IHC for the PTHR andPTHrP, and ISH for the PTHR in the rat kidney. Intensebrown staining representing the PTHR protein was presentin epithelial cells of the proximal convoluted tubules andpodocytes within the glomerulus (Figs. 2A and 2B) with

some receptor present in the cortical portions of the distalconvoluted tubules. Nuclear staining for the receptor waspresent in both cell types. In situ hybridization for thereceptor showed the mRNA to be localized to the same celltypes within the proximal tubules and glomerulus of thekidney (Figs. 2C and 2D). PTHrP was localized mainly todistal convoluted tubules (Figs. 2E and 2F), although somestaining was present in proximal convoluted tubules. Nu-clear localization of PTHrP was evident in cells of theproximal convoluted tubules.

Expression of PTHR and PTHrP in small intestine

In the rat small intestine, the PTHR was present in thevillus epithelium, intestinal crypt cells, and in interstitialcells, with each of these cell types exhibiting some nuclearstaining (Figs. 3A and 3B). The PTHR was also detected inthe smooth muscle surrounding the intestine (Fig. 3A). ThePTHR transcript was more sparsely distributed but in thesame areas as the peptide (Figs. 3C and 3D), although littletranscript could be detected in the villus stroma. The PTHrPpeptide was also easily detectable in smooth muscle, inter-stitial cells, and in the villus epithelium (Figs. 3E and 3F).However, PTHrP was not detected in the most apical epi-thelium of the villi nor in the intestinal crypts (Fig. 3F).PTHrP was detected in the nucleus of villus epithelial cells.

Expression of PTHR and PTHrP in uterus

The PTHR was localized to smooth muscle cells, surfaceand glandular epithelium and a subset endometrial stromalcells in the rat uterus (Figs. 4A and 4B). Nuclear localiza-tion of the receptor was especially evident in both surfaceand glandular epithelial cells. PTHR transcripts were local-ized to the same cell types as the protein in the rat uterus(Figs. 4C and 4D). PTHrP was more widely distributed inthe rat uterus (Figs. 4E and 4F) with abundant staining inglandular and luminal epithelium, smooth muscle, andthroughout the endometrial stroma. No nuclear localizationof PTHrP was evident in the uterus preparations.

Expression of PTHR and PTHrP in ovary

In the rat ovary, the PTHR was found in both the thecaland granulosa cell layers of developing and large folliclesand in the corpus luteum (Figs. 5A and 5B). Nuclear stain-ing for the receptor was especially prominent in granulosacells (Fig. 5B). Transcripts for the PTHR were localized totheca, granulosa, and luteal cells (Figs. 5C and 5D). PTHrPwas strongly expressed in thecal cells and was present insome granulosa cells (Figs. 5E and 5F), in which somenuclei were also stained. PTHrP was also present in corporalutea (Fig. 5E).

Confirmation of absence of PTHR expression in PTHR(2/2) mouse tibiae

Figure 6 shows an example of immunohistochemistry ofthe PTHR in hindlimbs from wild-type and PTHR knockout(null) mice taken at day 18.5 of gestation. PTHR was

1035NUCLEAR LOCALIZATION OF PTH/PTHrP RECEPTOR

detected only in chondrocytes from the hindlimb of a wild-type mouse (Fig. 6A). Receptor null mice had no detectablereceptor protein in hindlimb chondrocytes (Fig. 6B), con-firming the specificity of the antibodies used. Data shownare for antibody PTH-II, and similar results were obtainedwith antibody PTH-IV (data not shown).

Western and ligand blotting of PTHRThe PTH II antibody used in the immunocytochemical

studies detected a band at 66.3 kDa, the expected size of thePTHR, when 30mg of liver or MC3T3-E1 cell membranes,cytosol and nuclei were subjected to Western blotting (Fig.7A). There is also a prominent band at double this size in the

FIG. 1. Localization of PTHR protein and mRNA, and PTHrP in liver. Immunohistochemistry localized the PTHR in ratliver with an antibody specific to the N-terminal extracellular domain of the receptor (A and B). The receptor was presentin hepatocytes in a pattern radiating outward from hepatic sinuses. Arrows indicate nuclear localization of the receptorprotein. Arrowheads indicate negatively staining Kupffer cells. (C) Shows a section of rat liver treated with antibody thatwas preabsorbed overnight before starting the IHC and no staining was detected. Expression of the mRNA for the PTHRwas detected by in situ hybridization with a biotin-labeled antisense riboprobe. A similar pattern to that for the protein wasobtained, although in a less obviously radiating pattern (D and E). Liver sections hybridized to the correspondingbiotin-labeled sense riboprobe showed that little or no background staining occurs with this procedure (F). Immunohis-tochemical localization of PTHrP in the liver showed a comparable radiating pattern of staining as that for the receptor (Gand H). Arrows indicate nuclear localization of PTHrP in hepatocytes and arrowheads indicate PTHrP in Kupffer cells. Nostaining was evident in liver sections treated with normal serum instead of primary antibody (J). Original magnifications:A, D, G, and J,3120; other panels,3300.

1036 WATSON ET AL.

subcellular fractions of rat liver. When identical blots wereprobed with biotin-hPTH(1–34), a very similar pattern wasobtained (Fig. 7B). Two bands at 66.3 and approximately135 kDa were very prominent in all lanes, with liver againgiving a more heterogeneous result.To be certain that theprotein species detected was indeed the PTHR, ligand blotstrips were probed with both biotin-hPTH(1–34) and in-creasing amounts of unlabeled hPTH(1–34) (Fig. 7C). Botha 10- and 100-fold excess of unlabeled ligand successfullyeradicated the signal obtained with biotin-hPTH(1–34).

DISCUSSION

This study confirms the broad distribution of PTHR ex-pression in rat tissues and is the first to show nuclearlocalization of the receptor protein in liver, kidney, gut,uterus and ovary of any species. Colocalization of PTHrP

and the PTHR within the same tissues and/or cell typesindicates potential paracrine and/or autocrine circuits forPTHrP within these tissues.

PTH binding in vivo to sites in the liver was previouslyshown by autoradiography.(22–24)Although binding of botha 125I-labeled PTH(1–34) peptide analogue and PTH(1–84)to hepatocytes was observed, only intact PTH(1–84), butnot the amino-terminal peptide, bound to Kupffer cells. Thepresent study shows that the PTHR that binds the NH2-terminus of PTH is present on hepatocytes but not onKupffer cells, consistent with the autoradiographic data.The Kupffer cell is thought to be responsible for the hepaticuptake of intact PTH and cleavage between residues 33 and34.(25) Thus binding of PTH to the Kupffer cell is likely toinvolve midregion and carboxyl-terminal portions of themolecule and a receptor/binding protein distinct from thePTHR.

FIG. 2. Localization ofPTHR protein and mRNA,and PTHrP in kidney. (Aand B) PTHR protein im-munostaining in cells liningthe proximal convoluted tu-bules (p) and in podocytes(arrowheads in B) of theglomerulus (g) of the kid-ney. Arrows indicate nu-clear localization of the re-ceptor protein. Transcriptsfor the receptor were alsodetected in proximal convo-luted tubules and podocytes(C and D). In rat kidney,staining for PTHrP was de-tected in cells lining the dis-tal convoluted tubules (d)and podocyte cells (E andF). Arrows indicate nuclearlocalization of PTHrP. Orig-inal magnifications: A, C,and E,3120; B, D, and F,3300.

1037NUCLEAR LOCALIZATION OF PTH/PTHrP RECEPTOR

The present investigations make extensive use of newlyavailable antibodies for the localization of the PTHR. Con-sistent nuclear localization (see Fig. 8 for a detailed view inliver and kidney) was an unexpected result and care wastaken to ensure specific staining of the receptor. Theepitopes to which the antibodies used were raised are eitherin the N-terminal extracellular domain or in the first extra-cellular loop(9,26) and are specific to the PTHR as searchesof protein and DNA databases yielded only matches to thePTHR and no other protein. The inability of the two PTHRantibodies to stain receptors in the hindlimbs of PTHR nullmice despite the fact that tibiae from wild-type littermatesdid stain, further confirms the veracity of these results.Subcellular fractionation of both rat liver and culturedMC3T3-E1 cells showed the presence of immunoreactivityfor the PTHR in membrane, cytosol, and nuclear fractions.The specificity of the immunoreactive bands for PTHR wasconfirmed with ligand blots using biotin-labeled ligand.

Unlabeled hPTH(1–34) successfully competed with biotin-labeled hPTH(1–34) for PTHR binding on ligand blots ofMC3T3-E1 nuclei, thus confirming the specificity of ourresults.

How would the PTHR get to the nucleus? There areseveral possible routes by which the receptor can becomesegregated in the nucleus, none of which is directly an-swered by this study. Robbins and associates were the firstto describe a bipartite type of nuclear localization signal(NLS) in nucleoplasmin.(27) This category of NLS is com-posed of a sequence in which two basic amino acids areseparated by a spacer of at least 10 residues in length froma region where at least 3 of 5 residues are basic. Examina-tion of the rat PTHR sequence with PSORTII (a package ofprograms for the analysis of sorting sequences in primaryprotein structure available on the ExPASy Web site pro-vided by the Swiss Institute for Bioinformatics at http://www.ExPASy.hcuge.ch) revealed a potential bipartite nu-

FIG. 3. Localization ofPTHR protein and mRNA,and PTHrP in small intes-tine. The PTHR was foundthroughout the rat small in-testine (A and B) with nu-clei of epithelial cells liningboth villi (v) and crypts (c)staining positively for thereceptor (arrows). (C and D)Show that mRNA for thePTHR is similarly distrib-uted. PTHrP was found invilli (except at the extremeapical margin) but not crypts(E and F). Both PTHR andPTHrP were abundant insmooth muscle (m) surround-ing the gut (A and E, respec-tively). Original magnifica-tions: A, C, and E,3120; B,D, and F,3300.

1038 WATSON ET AL.

clear localization signal (NLS)(27) at residues 471–487 inthe cytoplasmic C-terminal tail of the receptor. This se-quence, RKSWSRWTLALDFKRKA (Table 1), could po-tentially bind a-importin and allow translocation of thePTHR to the nucleus.(28) Examination of the human, mouse,opossum, and pig PTHR sequences revealed an NLS in eachof identical or similar structure and position (Table 1). NoNLS was found in the sequences of rat or human PTH2receptors. In studies in which the C-terminal tail of thePTHR was mutated, removal of residues 475–494 resultedin a 50–60% reduction in the internalization of the ligand-bound receptor.(29) Because the sequence removed encom-passes the potential NLS, an involvement of this region ofthe protein in both internalization and nuclear targeting issuggested.

Whether the nuclear translocated receptor is internalizedafter ligand binding [as is true for a number of integralmembrane receptors(30)] or a nascent receptor that lacks

signal sequence (and is therefore a cytoplasmic form) re-mains to be seen. A signal peptide–deficient PTHR tran-script has already been identified in kidney,(31) and it may bethat, at least in the kidney, the PTHR protein lacking thesignal sequence can directly access the nucleus withoutbeing inserted in the plasma membrane first. Alternatively,intracellular PTHrP could act in concert with its cognatereceptor in an intracrine fashion and together go to thenucleus.

PTHrP itself is well known to localize to the nucleus,specifically the nucleolus,(32,33) but less is known of thedetails of its translocation. PTHrP contains a bipartite NLSthat would allow translocation via the importin system.(28,33)

It has been suggested that both secreted PTHrP (autocrine)and cytoplasmic PTHrP (intracrine) find their way to thenucleus.(34) Exogenous, fluorescently labeled PTHrP wasrapidly internalized and translocated to the nucleus ofUMR106.01 cells but not in UMR201 cells, which lack the

FIG. 4. Localization ofPTHR protein and mRNA,and PTHrP in uterus. IHCfor the PTHR in rat uterusshowed the receptor’s pres-ence in uterine glands (ar-rowhead), luminal epithe-lium (e), and some stromalcells (s) (A and B). Arrowsindicate nuclear localizationof the receptor. In situ hy-bridization for the receptormessage in the uterus (Cand D) also gave a strongsignal in uterine glands, lu-minal epithelium, and a sub-set of stromal cells. IHC forPTHrP in rat uterus (E andF) also showed that uterineglands, luminal epithelium,and stromal cells containPTHrP ligand. Original mag-nifications: A, C, and E,3120; B, D, and F,3300.

1039NUCLEAR LOCALIZATION OF PTH/PTHrP RECEPTOR

PTHR.(34) Nascently produced PTHrP could also attain anuclear localization with onlyb-importin as carrier.(34) Inconcert with the results of the present study, these findingssuggest that the PTHR could also be involved in the trans-location of extracellular PTHrP to the nucleus. The possi-bility is also raised that both PTHrP and its cognate receptormay interact in an intracrine manner to affect nuclear func-tion. It has been shown recently that PTHrP can interactwith RNA, although the consequences of this remain un-clear at the present time.(35)

Nuclear targeting of PTHrP in vascular smooth musclecells in culture is associated with increased mitogenesis, aneffect that is diametrically opposite to that elicited byPTHrP binding to cell surface PTHR.(36) Similarly, serum-induced proliferation of aortic smooth muscle cells is ac-companied by abundant expression of PTHrP.(37) Thisraises the possibility that nuclear localization of the PTHR

without membrane association could have a distinct effectfrom that of an internalized, ligand-bound receptor.

Nuclear targeting of protein/peptide receptors and/or theirligands is well documented in the case of tyrosine kinasereceptors (EGF and insulin receptors have a direct effect ongene regulation within the nucleus of target cells) andcytokine receptors (both the interferons and some interleu-kins accumulate in the nucleus).(38–42) Many roles havebeen proposed for nuclear translocated receptors, includingacting as a chaperone for the ligand,(38,30)shuttling of STATmolecules to the nucleus (in concert with their ligand),(43)

and direct regulation of gene function through tyrosinekinase and other activities.(38,30)Our results indicate that thePTHR, a member of the G protein-coupled family of recep-tors (GPCR), is localized to the nucleus of some cells.PTHrP and the PTHR are the only members of the GPCR tohave a shown nuclear localization.(30) The studies reported

FIG. 5. Localization ofthe PTHR protein andmRNA, and PTHrP inovary. PTHR staining wasabundant in the rat ovary (Aand B). All developing fol-licles stained positively forthe receptor in oocyte (o),granulosa (g), and theca (t)compartments. Staining wasalso evident in the corpusluteum (cl). Arrows indicatenuclear localization of thereceptor. PTHR transcriptswere also found in granu-losa, theca, and corporalutea (C and D). No PTHRmRNA was apparent inovarian stroma (C). IHC forPTHrP in the ovary alsoshowed that oocytes (o),granulosa (g), theca (t) cells,and corpora lutea (cl) con-tain the ligand (E and F).Original magnifications: A,C, and E,3120; B, D, andF, 3300.

1040 WATSON ET AL.

herein shed no light on the potential role of a nuclear PTHR,but we have shown that the PTHR traffics to the nucleusduring the time of DNA duplication in MC3T3-E1 cellswhen synchronized in culture.(44)

Here we have shown that the PTHR can be localized tothe nucleus of cells within the kidney, liver, gut, uterus, and

ovary, although the significance of this observation remainsto be determined. Both PTH and PTHrP have wide-rangingeffects on target cells, mostly mediated by the cAMP andinositol phosphate/diacylglycerol pathways.(45) However,direct nuclear effects cannot be ruled out. PTH modifiesnuclear architecture and gene expression in ROS 17/2.8

FIG. 6. Immunohistochem-istry of the PTH/PTHrP re-ceptor in wild-type mice(WT) and PTHR gene-ablated (2/2) littermates.(A) Positive staining for thePTHR in late proliferative/early hypertrophic chondro-cytes in the tibia of a WTmouse. Arrows indicate nu-clear staining. (B) Chondro-cytes in the tibia of a recep-tor null (2/2) mouse didnot stain for the PTHR, con-firming the specificity of theantibody used in this study.Original magnification3300.

FIG. 7. Western blot of subcellular fractions of rat liver and cultured mouse osteoblast-like MC3T3-E1 cells (A). Blotswere prepared as described in Materials and Methods and were probed with anti-rat PTHR antibody (PTH II) diluted 1/100in 0.1% BSA-PBS. The blot was developed using ECL technology. The position of the PTHR is indicated on the right (N,nuclei; M, membranes; C, cytosol). In (B), an identical blot was probed with 10mg/ml biotinylated-hPTH(1–34). Note thesimilarity of the two blots (A and B) and that rat liver gives a more heterogeneous pattern than does cultured mouseosteoblasts. (C) Shows blot strips of 30mg of MC3T3-E1 cell nuclei separated by SDS-PAGE and probed with 1mg/mlbiotin-hPTH(1–34) in the presence of the indicated amount of unlabeled hPTH(1–34). Note that both a 10- and 100-foldexcess of unlabeled ligand abolishes the PTHR signal at 66.3 kDa.

1041NUCLEAR LOCALIZATION OF PTH/PTHrP RECEPTOR

osteosarcoma cells(46–48) and appears to play a role in theprogression of the cell cycle in ROS cells.(48) Physicalalteration of nuclear structure by the PTHR is one routewhereby gene expression can be modified and this hypoth-esis is under investigation. Potentially, the PTHR could actas a transcription factor/cofactor, or interact with nuclearRNA to modify its function.

ACKNOWLEDGMENTS

We thank Dr. Henry M. Kronenberg for suggesting thatwe use the PTHR gene knockout mice as a control for theseexperiments and for his critical reading of this manuscript.

This work was supported by the Medical Research Coun-cil of Canada through Grant MT-13199 (to A.B.H.) andGrant MT-9315 (to G.N.H.), and a grant from the KidneyFoundation of Canada (to G.N.H.).

REFERENCES

1. Urena P, Kong S-F, Abou-Samra A-B, Jueppner H, Kronen-berg HM, Potts JT Jr, Segre GV 1993 Parathyroid hormone(PTH)/PTH-related peptide receptor messenger ribonucleic ac-ids are widely distributed in rat tissues. Endocrinology133:617–623.

2. Tian J, Smogorzewski M, Kedes L, Massry SG 1993 Parathy-roid hormone-parathyroid hormone-related protein receptor

messenger RNA is present in many tissues besides the kidney.Am J Nephrol13:210–213.

3. Lee K, Deeds JD, Segre GV 1995 Expression of parathyroidhormone-related peptide and its receptor messenger ribonu-cleic acids during fetal development of rats. Endocrinology136:453–463.

4. Tucci J, Beck F 1998 Expression of parathyroid hormone-related protein (PTHrP) and the PTH/PTHrP receptor in the ratuterus during early pregnancy and following artificial decid-uoma induction. J Reprod Fertil112:1–10.

5. Weaver DR, Deeds JD, Lee K, Segre GV 1995 Localizationof parathyroid hormone-related peptide (PTHrP) and PTH/PTHrP receptor mRNAs in rat brain. Mol Brain Res28:296–310.

6. Massry SG, Klin M, Ni Z, Tian J, Kedes L, Smogorzewski M1995 Impaired agonist-induced calcium signalling in hepato-cytes from chronic renal failure rats. Kidney Int48:1324–1331.

7. Downey SE, Hoyland J, Freemont AJ, Knox F, Walls J,Bundred NJ 1997 Expression of the receptor for parathyroidhormone-related protein in normal and malignant breast tissue.J Pathol183:212–217.

8. Li H, Seitz PK, Thomas ML, Selvanayagam P, Rajaraman S,Cooper CW 1995 Widespread expression of the parathyroidhormone-related peptide and PTH/PTHrP receptor genes inintestinal epithelial cells. Lab Invest73:864–870.

9. Pausova Z, Bourdon J, Clayton D, Mattei M-G, Seldin MF,Janicic N, Riviere M, Szpirer J, Levan G, Szpirer C, GoltzmanD, Hendy GN 1994 Cloning of a parathyroid hormone/parathyroid hormone-related peptide receptor (PTHR) cDNAfrom a rat osteosarcoma (UMR106) cell line: Chromosomal

FIG. 8. Higher-power mag-nification of nuclear local-ization of the PTHR in ratliver (A) and kidney (B).Note that some hepatocytes(A) have a strictly nuclear(n) localization, some a cy-toplasmic (c) localization,and some have both. Com-partmentalized localizationof the PTHR in the kidney(B) is less distinct and mostpositive cells have both cy-toplasmic and nuclear stain-ing. Original magnification3480.

TABLE 1. NUCLEAR LOCALIZATION SIGNAL (NLS) IN NUCLEOPLASMIN, PTHR,AND PTHRP

Protein NLS Species

Nucleoplasmin 155KRPAATKKAGQAKKKK 170 XenopusPTHR 471RKSWSRWTLALDFKRK A487 Rat/mouse

471KK SWSRWTLALDFKRK A487 Human465KK SWSRWTLALDFKRK A481 Opossum466KK SWSRWTLALDFKRK A482 Pig

PTHrP 88KKKK GKKKPGKRREQEKKKRR T107 Rat/mouse/human

1042 WATSON ET AL.

assignment of the gene in the human, mouse and rat genomes.Genomics20:20–26.

10. Lanske B, Karaplis AC, Lee K, Luz A, Vortkamp A, Pirro A,Karperien M, Defize LHK, Ho C, Mulligan RC, Abou-SamraA-B, Juppner H, Segre GV, Kronenberg HM 1996 PTH/PTHrP receptor in early development and indian hedgehog-regulated bone growth. Science273:663–666.

11. Lee K, Lanske B, Karaplis AC, Deeds JD, Kohno H, Nissen-son RA, Kronenberg HM, Segre GV 1996 Parathyroidhormone-related peptide delays terminal differentiation ofchondrocytes during endochondral bone development. Endo-crinology 137:5109–5118.

12. Amizuka N, Lee HS, Kwan MY, Arazani A, Warshawsky H,Hendy GN, Ozawa H, White JH, Goltzman D 1997 Cell-specific expression of the parathyroid hormone (PTH)/PTH-related peptide receptor gene in kidney from kidney-specificand ubiquitous promoters. Endocrinology138:469–481.

13. Lee K, Brown D, Urena P, Ardaillou N, Ardaillou R, Deeds J,Segre GV 1996 Localization of parathyroid hormone/parathyroid hormone-related peptide receptor mRNA in kid-ney. Am J Physiol270:F186–F191.

14. Riccardi D, Lee W-S, Lee K, Segre GV, Brown EM, HebertSC 1996 Localization of the extracellular Ca21-sensing recep-tor and PTH/PTHrP receptor in rat kidney. Am J Physiol271:F951–F956.

15. Yang T, Hassan S, Huang YG, Smart AM, Briggs JP, Schner-mann JB 1997 Expression of PTHrP, PTH/PTHrP receptor,and Ca(21)-sensing receptor mRNAs along the rat nephron.Am J Physiol272:F751–F758.

16. Lee K, Deeds JD, Bond AT, Jueppner H, Abou-Samra A-B,Segre GV 1993In situ localization of PTH/PTHrP receptormRNA in the bone of fetal and young rats. Bone14:341–345.

17. Lanske B, Divieti P, Kovacs CS, Pirro A, Landis WJ, Drane S,Bringhurst FR, Kronenberg HM 1998 The parathyroid hor-mone (PTH)/PTH-related peptide receptor mediates actions ofboth ligands in murine bone. Endocrinology139:5194–5204.

18. Watson P, Lazowski D, Han V, Fraher L, Steer B, Hodsman A1995 Parathyroid hormone restores bone mass and enhancesosteoblast insulin-like growth factor I gene expression inovariectomized rats. Bone16:357–365.

19. Largo R, Gomez-Garre D, Santos S, Penaranda C, Blanco J,Esbrit P, Egido J 1999 Renal expression of parathyroidhormone-related protein (PTHrP) and PTH/PTHrP receptor ina rat model of tubulointerstitial damage. Kidney Int55:82–90.

20. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ 1951 Proteinmeasurement with the Folin phenol reagent. J Biol Chem193:265–275.

21. Harlow E, Lane A 1988 Antibodies: A Laboratory Manual.Cold Spring Harbor Laboratory, Cold Spring Harbor, NY,U.S.A.

22. D’Amour P, Segre GV, Roth SI, Potts JT Jr 1979 Analysis ofparathyroid hormone and its fragments in rat tissues: Chemicalidentification and microscopical localization. J Clin Invest63:89–97.

23. Bergeron JJM, Tchervenkov S, Rouleau MF, Rosenblatt M,Goltzman D 1981 In vivo demonstration of receptors in ratliver to the amino-terminal region of parathyroid hormone.Endocrinology109:1552–1559.

24. Rouleau MF, Warshawsky H, Goltzman D 1986 Parathyroidhormone binding in vivo to renal, hepatic, and skeletal tissuesof the rat using a radioautographic approach. Endocrinology118:919–931.

25. Segre GV, Perkins AS, Witters LA, Potts Jr. JT 1981 Metab-olism of parathyroid hormone by isolated rat Kupffer cells andhepatocytes. J Clin Invest67:449–456.

26. Abou-Samra AB, Jueppner H, Force T, Freeman MW, KongXF, Schipani E, Urena P, Richards J, Bonventre JV, Potts JrJT, Kronenberg HM, Segre GV 1992 Expression cloning of acommon receptor for parathyroid hormone and parathyroidhormone-related peptide from rat osteoblast-like cells: a single

receptor stimulates intracellular accumulation of both cAMPand inositol triphosphates and increases intracellular free cal-cium. Proc Natl Acad Sci USA89:2732–2736.

27. Robbins J, Dilworth SM, Laskey RA, Dingwall C 1991 Twointerdependent basic domains in nucleoplasmin nuclear target-ing sequence: identification of a class of bipartite nucleartargeting sequences. Cell64:615–623.

28. Moroianu J 1997 Molecular mechanisms of nuclear proteintransport. Crit Rev Eukaryot Gene Expr7:61–72.

29. Huang Z, Chen Y, Nissenson RA 1995 The cytoplasmic tail ofthe G-protein-coupled receptor for parathyroid hormone-related protein contains positive and negative signals for en-docytosis. J Biol Chem270:151–156.

30. Jans DA, Hassan G 1998 Nuclear targeting by growth factors,cytokines, and their receptors: A role in signaling? Bioessays20:400–411.

31. Joune H, Lanske B, Karperien M, Qian F, Defize L, Abou-Samra A 1997 Tissue-specific transcription start sites andalternative splicing of the parathyroid hormone (PTH)/PTH-related peptide (PTHrP) receptor gene: A new PTH/PTHrPreceptor splice variant that lacks the signal peptide. Endocri-nology 138:1742–1749.

32. Henderson JE, Amizuka N, Warshawsky H, Biasotto D, Lan-ske BMK, Goltzman D, Karaplis AC 1995 Nucleolar localiza-tion of parathyroid hormone-related peptide enhances survivalof chondrocytes under conditions that promote apoptotic celldeath. Mol Cell Biol15:4064–4075.

33. Nguyen MTA, Karaplis AC 1998 The nucleus: A target site forparathyroid hormone-related peptide (PTHrP) action. J CellBiochem70:193–199.

34. Lam MHC, Briggs LJ, Hu W, Martin TJ, Gillespie MT, JansDA 1999 Importin (beta) recognizes parathyroid hormone-related protein with high affinity and mediates its nuclearimport in the absence of importin (alpha). J Biol Chem274:7391–7397.

35. Aarts MM, Levy D, He B, Stregger S, Chen T, Richard S,Henderson JE 1999 Parathyroid hormone-related protein in-teracts with RNA. J Biol Chem274:4832–4838.

36. Massfelder T, Dann P, Wu TL, Vasavada R, Helwig JJ, Stew-art AF 1997 Opposing mitogenic and anti-mitogenic actions ofparathyroid hormone-related protein in vascular smooth mus-cle cells: A critical role for nuclear targeting. Proc Natl AcadSci USA 94:13630–13635.

37. Hong T, Kupfer J, Enomoto H, Sharifi B, Gianella-Neto D,Forrester JS, Singer FR, Goltzman D, Hendy GN, Pirola C,Fagin JA, Clemens TL 1991 Abundant expression of parathy-roid hormone-related protein in primary rat aortic smoothmuscle cells accompanies serum-induced proliferation. J ClinInvest6:1841–1847.

38. Radulescu RT 1995 Insulin receptora-subunit: A putativegene regulatory molecule. Med Hypotheses45:107–111.

39. Csaba G, Hegyesi H 1994 Immunocytochemical verification ofthe insulin receptor’s specificity in the nuclear envelope ofTetrahymena. Comparison with receptors of the plasma mem-brane. Biosci Rep14:25–31.

40. Burwen SJ, Jones AL 1987 The association of polypeptidehormones and growth factors with the nuclei of target cells.Trends Biochem Sci12:159–162.

41. Burysek L, Houstek J 1996 Multifactorial induction of geneexpression and nuclear localization of mouse interleukin 1a.Cytokine6:70–78.

42. Curtis BM, Widner MB, de Roos P, Qwarnstorm EE 1990 IL-1and its receptor are translocated to the nucleus. J Immunol144:1295–1303.

43. Johnson HM, Torres BA, Green MM, Szente BE, Siler KI,Larkin III J, Subramaniam PS 1998 Cytokine-receptor com-plexes as chaperones for nuclear translocation of signal trands-ducers. Biochem Biophys Res Commun244:607–614.

44. Watson PH, Fraher LJ, Natale BV, Kisiel M, Hendy GN,Hodsman AB 2000 Nuclear localization of the type I PTH/

1043NUCLEAR LOCALIZATION OF PTH/PTHrP RECEPTOR

PTHrP receptor in MC3T3–E1 cells: Association with serum-induced cell proliferation. Bone (in press).

45. Chakravarthy BR, Durkin JP, Rixon RH, Whitfield JF 1990Parathyroid hormone fragment [3–34] stimulates protein ki-nase C (PKC) activity in rat osteosarcoma and murineT-lymphoma cells. Biochem Biophys Res Commun171:1105–1110.

46. Bidwell J, Van Wijnen A, Banerjee C, Fey E, Merriman H,Penman S, Stein J, Lian J, Stein G 1994 Parathyroid-responsive modifications in the nuclear matrix of ROS 17/2.8rat osteosarcoma cells. Endocrinology134:1738–1744.

47. Bidwell J, Feister H, Swartz D, Onyia J, Holden J, Hock J1996 Parathyroid hormone regulates the expression of ratosteoblast and osteosarcoma nuclear matrix proteins. J CellBiochem63:374–383.

48. Torrungruang K, Feister H, Swartz D, Hancock EB, Hock J,Bidwell JP 1998 Parathyroid hormone regulates the expressionof the nuclear mitotic apparatus protein in the osteoblast-likecells, ROS 17/2.8. Bone22:317–324.

Address reprint requests to:Patricia H. Watson, Ph.D.

Lawson Research Institute, Room H425268 Grosvenor Street

London, Ontario N6A 4V2, Canada

Received in original form August 13, 1999; in revised form De-cember 13 1999; accepted February 14, 2000.

1044 WATSON ET AL.