Ontogeny of Phex/PHEX Protein Expression in Mouse Embryo and Subcellular Localization in Osteoblasts


  • The authors have no conflict of interest


PHEX, a phosphate-regulating gene with homologies to endopeptidases on the X chromosome, is mutated in X-linked hypophosphatemia (XLH) in humans and mice (Hyp). Although recent observations indicate that Phex protein is expressed primarily in bone and may play an important role in osteoblast function and bone mineralization, the pattern of the Phex protein expression in the developing skeleton and its subcellular localization in osteoblasts remain unknown. We examined the ontogeny of the Phex protein in the developing mouse embryo and its subcellular localization in osteoblasts using a specific antibody to the protein. Immunohistochemical staining of mouse embryos revealed expression of Phex in osteogenic precursors in developing vertebral bodies and developing long bones on day 16 postcoitum (pc) and thereafter. Calvaria from day 18 pc mice showed Phex epitopes in osteoblasts. No Phex immunoreactivity was detected in lung, heart, hepatocytes, kidney, intestine, skeletal muscle, or adipose tissue of mouse embryos. Interestingly, embryonic mouse skin showed moderate amounts of Phex immunostaining. In postnatal mice, Phex expression was observed in osteoblasts and osteocytes. Moderate expression of Phex was seen in odontoblasts and slight immunoreactivity was observed in ameloblasts. Confocal microscopy revealed the presence of immunoreactive PHEX protein in the Golgi apparatus and endoplasmic reticulum of osteoblasts from normal mice and in osteoblasts from Hyp mice transduced with a human PHEX viral expression vector. PHEX protein was not detected in untransduced Hyp osteoblasts. These data indicate that Phex protein is expressed in osteoblasts and osteocytes during the embryonic and postnatal periods and that within bone, Phex may be a unique marker for cells of the osteoblast/osteocyte lineage.


X-linked hypophosphatemia (XLH) is the most common of several Mendelian disorders of phosphate homeostasis.(1) Patients with XLH have osteomalacia and/or rickets, hypophosphatemia, elevated urinary phosphate excretion, inappropriately normal or decreased serum 1,25-dihydroxyvitamin D concentrations, normal serum calcium concentrations, normal or elevated parathyroid hormone levels, short stature, dental abscesses, and enthesopathy.(2, 3)PHEX (previously known as PEX), a phosphate-regulating gene with homologies to endopeptidases located on the X chromosome, is localized to the Xp22.1-p22.2 region of the X chromosome and is mutated in patients with XLH.(4, 5-12) PHEX is homologous to the membrane-bound zinc metallopeptidase family of proteins that includes neutral endopeptidase (NEP) 24.11, endothelin-converting enzyme 1 and 2, and the Kell blood protein.(6) The mouse with XLH (Hyp) has a biochemical phenotype similar to that of patients with XLH(13, 14) and a large deletion in the 3′ region of the Phex gene.(15, 16) A deletion in the 5′ region of the Phex gene also is present in the gyro (Gy) mouse.(16)

A number of studies have examined the expression of Phex messenger RNA (mRNA) in mouse tissues and cell cultures using polymerase chain reaction, Northern analysis, ribonuclease protection, and in situ hybridization methods.(15, 17-19) In these studies, Phex mRNA expression has been observed in osteoblasts in developing bone on embryonic day 15, just one-half of a day after the onset of ossification.(18) In teeth, Phex mRNA expression is present a day before birth on embryonic day 19.(18) In one study, high levels of Phex mRNA expression were detected in brain, calvaria, kidney, and lung shortly after birth.(19) Recently, Ruchon et al.(20) and Miao et al.(21) have examined the distribution of Phex protein in neonatal and adult mice using monoclonal and polyclonal antisera against Phex, respectively. Immunostaining of osteoblasts, osteocytes, and odontoblasts was noted in these studies.

However, no previous studies have examined the expression of Phex protein in developing embryos, or have evaluated the subcellular distribution of the protein in osteoblasts. This study focused on the expression of Phex protein in the developing embryonic skeleton and on the subcellular localization of the protein in normal osteoblasts, Hyp osteoblasts, and Hyp osteoblasts transduced with a viral vector expressing the normal human PHEX complementary DNA (cDNA). To determine whether Phex expression occurs in specific tissues and cells at defined times during development, we examined Phex protein expression in several embryonic (days 8-19 postcoitum [pc]) and early postnatal (2, 4, and 12 weeks) developing mouse tissues. We used immunofluorescent methods and confocal microscopy to examine the subcellular distribution of human PHEX expression in transformed osteoblast cell lines derived from normal and Hyp mice. We show that Phex protein is first detected in cells of murine developing bone at day 16 pc. Subsequently, it is found in osteoblasts of the growth plate and later in osteocytes of mature bone. Within osteoblasts, the protein is found predominantly in the endoplasmic reticulum and Golgi compartment. Skin also showed modest Phex protein expression.


Polypeptide selection and synthesis

An amino acid sequence spanning amino acids 706-733 (QIGAHSPPQFRVNGAISNSEEFQKAFNC) of the human PHEX protein was synthesized by Fmoc solid-phase peptide synthesis and conjugated to keyhole limpet hemocyanin (KLH).(22–24) There is one amino acid difference between the human and mouse PHEX sequence in this region of the protein (the serine residue at position 724 in the human sequence is replaced by a phenylalanine residue in the mouse protein). The peptide differed from the corresponding region in NEP by 17 amino acid residues.

Production of polyclonal antisera in rabbits

Antibodies were raised at Cocalico Biologicals (Reamstown, PA, USA) using standard methods. Two adult New Zealand white rabbits were inoculated with primary and booster emulsions of the KLH-conjugated polypeptide and Freund's adjuvant. Primary inocula consisting of 0.5 ml of a 1-mg/ml solution of antigen in water mixed to an emulsion with an equal volume of complete Freund's adjuvant were injected intradermally at multiple dorsal sites. Booster injections were administered in a similar manner on days 14, 21, and 49. These injections consisted of 0.25 ml of a 1-mg/ml solution in an equal volume of incomplete Freund's adjuvant. Blood was obtained before the primary inoculation and on days 35, 56, and 86. After confirmation of an antigen-specific response by a direct ELISA,(25) the rabbits were bled by cardiac puncture 7 days after the last booster injection.(25) After each blood draw, serum was separated and stored at −70°C.

Collection of mouse embryos and postnatal mouse tissues

Timed, pregnant mice (SJL/JxC57BL6/J) were killed daily between gestational days 8 and 18.(26) Embryos were harvested and fixed in 4% formaldehyde in phosphate-buffered saline (PBS), pH 7.5, at 4°C. Neonatal (1-day-old) Hyp mice were perfused with a solution of 4% formaldehyde in PBS, pH 7.4. Formaldehyde-fixed tissue specimens were embedded in paraffin and 4-μm-thick sections were prepared for immunohistochemical studies. For studies in postnatal tissues, J-1 mice that were fed regular mouse chow were used. At 2, 4, and 12 weeks of age, mice were killed and tibias, calvarium, and teeth were removed and fixed immediately in 10% formalin. Formalin-fixed tissues were demineralized with 5% formic acid, dehydrated with ethanol and xylene, and embedded in paraffin. Five-micrometer sections were mounted on glass slides for immunohistochemical studies.

Immunohistochemical staining

Immunohistochemical staining was carried out using immune serum or preimmune rabbit serum at a 1:200 dilution.(27–30) Slides were deparaffinized in xylene, rehydrated through a graduated alcohol series, and rinsed in distilled water. Sections were exposed to heat for epitope retrieval in 1 mM of EDTA buffer, pH 8.0, for 30 minutes. After cooling in EDTA buffer for 5 minutes, sections were rinsed in tap water and processed for immunostaining. Endogenous peroxidase was blocked with 0.03% hydrogen peroxide containing sodium azide for 10 minutes. Primary antibody was applied to the tissue sections at room temperature for 30 minutes. The slides were rinsed well with tap water. Peroxidase-labeled polymer conjugated to goat anti-rabbit secondary antibody was added to the sections at room temperature for 20 minutes. After rinsing with water, sections were developed in a buffered substrate containing 3-amino-9-ethylcarbazole and hydrogen peroxide for 5 minutes. Slides were counterstained with Mayer's hematoxylin for 10 minutes and treated with ammonium hydroxide water for 30 s. Coverslips were attached using DAKO Faramount Mounting Medium (DAKO Corp., Carpinteria, CA, USA)

Construction of Moloney murine leukemia virus vector expressing human PHEX cDNA

To reconstitute human PHEX protein in Hyp mouse osteoblasts, we designed a bicistronic retroviral vector using a poliovirus internal ribosome entry site.(31) A DNA fragment containing the PHEX cDNA as well as a short (108 nucleotides [nt]) portion of the untranslated region (UTR) and 0.5 kilobases (kb) of 3′-UTR was first excised from pCDNA 3 with ApaLI and XhoI. A BglI1 linker was blunt end ligated to the ApaLI site and the resulting PHEX-encoding fragment was subsequently substituted for the CXCR4 gene insert of pJZCXCR4.(32) The resulting bicistronic transfer vector pJZ-PHEX was packaged into retroviral particles by transient calcium phosphate-mediated transfection of a stable amphotropic packaging cell line (PhoenixA cells). Vector supernatant was clarified, filtered, and titered on National Institutes of Health (NIH) 3T3 cells by formation of G418 resistant colonies after limiting dilution.

Transduction of Hyp mouse osteoblasts with viral vector pJZ-PHEX

Immortalized Hyp osteoblasts(33) were grown in α-modified essential medium (α-MEM) containing 10% fetal calf serum at 37°C in an atmosphere of 5% CO2/95% air until they were 50% confluent. The 105 viral particles were added to the cells. Twenty-four hours later, the medium was removed and replaced with fresh α-MEM and 10% fetal bovine serum (FBS) containing 800 μg/ml of G418. After 4 weeks, G418-resistant cells were harvested.

Assessment of expression of PHEX Protein and mRNA PHEX protein glycosylation status in normal and Hyp osteoblasts and in Hyp osteoblasts transduced with pJZ-PHEX

Western blot analysis

Western blot analysis of osteoblast (normal, Hyp, or Hyp transformed with full-length human PHEX) cell lysates was performed as follows. Cells were lysed with cell lysis buffer (0.15 M of NaCl, 0.1% NP-40, 1% Triton X-100, and 50 mM of HEPES, pH 7.25). Lysate protein was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE, 7.5%) and separated proteins were transferred electrophoretically to Sequi-Blot polyvinylidene difluoride (PVDF) membrane (Bio-Rad Laboratories, Hercules, CA, USA).(34, 35) After incubating with blocking solution (BM chemiluminescence blotting substrates; Roche Molecular Biochemicals, Indianapolis, IN, USA) for 1-2 h at room temperature, the membrane was treated with PHEX antibody (1:200 dilution) for 1.5 h at room temperature. The blot was washed with Tris-buffered saline-Tween (TBS-Tween) and incubated with goat anti-rabbit immunoglobulins (1:2000 dilution) conjugated to horseradish peroxidase. Blots were developed with chemiluminescence reagents (BM chemiluminescence blotting substrates; Boehringer Mannheim, Roche Molecular Biochemical) and immunoreactive bands were detected by autoradiography.

For experiments in which glycosidase experiments were performed, whole cell lysates were prepared in 6× Laemmli sample buffer and boiled for 3 minutes.(35) Separation of proteins was performed on SDS-PAGE (7.5% acrylamide) and proteins were transferred to nitrocellulose membranes for 1 h.(34) Incubation with a mouse monoclonal anti-Phex antibody at a 1:200 dilution was performed as described previously(20) and was followed by incubation with peroxidase-conjugated anti-mouse immunoglobulin G (IgG; Vector Laboratories, Burlingame, CA, USA) at a 1:3000 dilution. Immune complexes were visualized by chemiluminescence with the enhanced chemiluminescence (ECL) kit (Amersham Pharmacia Biotech, Piscataway, NJ, USA). Both the polyclonal and the monoclonal PHEX antibodies generated similar patterns of PHEX protein expression of Western blots.

Endoglycosidase digestion

Recombinant PHEX proteins in whole cell extracts were boiled for 10 minutes in 10× denaturation buffer (5% SDS and 10% β-mercaptoethanol) and incubated for 1 h at 37°C with either peptide:N-glycosidase F (PNGase F) or endoglycosidase H (endo H) according to the manufacturer's recommendations (New England Biolabs, Missisauga, Ontario, Canada). Digestion products were fractionated on SDS-PAGE and subjected to immunoblot analyses as described previously.

Ribonuclease protection analysis

Total RNA was isolated from normal, Hyp, and Hyp-transduced osteoblasts using Trizol reagent (Life Technologies, Burlington, Ontario), hybridized with33P-labeled 3′-Phex and β-actin riboprobes (2 × 105 cpm) at 50°C for 18 h, and treated with 200 U RNAse T1 for 1 h at 30°C as described previously.(15) The protected fragments were precipitated, heat-denatured, and electrophoresed on 5% denaturing polyacrylamide gels. The gels were dried and exposed to a PhosphorImager screen (Molecular Dynamics, Sunnyvale, CA, USA) for quantification of radioactive signals under conditions in which linearity is achieved. Phex PhosphorImage signals were related to those of β-actin.

Immunofluorescence and confocal microscopy

Immunofluorescence was carried out on three cell lines of transformed mouse osteoblasts:normal osteoblasts (TMOB 23 cells), Hyp osteoblasts (TMOB 20 Hyp cells), and pJZ-PHEX-transduced Hyp osteoblasts (TMOB 20 Hyp PHEX). Each cell line was grown on 22 mm × 22 mm coverslips (Fisher Scientific, Pittsburgh, PA, USA) in Costar 6-well polystyrene sterile cell culture dishes (Corning, Inc., Corning, NY, USA) in α-MEM with 10% FBS and 1% penicillin/streptomycin (Life Technologies). After proliferating to 40-60% confluence, the coverslips were washed in PBS, pH 7.4. Cells were permeabilized and fixed in a 100% methanol (EM Chemicals, Darmstadt, Germany) at −20°C for 10 minutes. Cells on coverslips were air-dried and then rehydrated in PBS, pH 7.4. Cells were incubated in blocking buffer (2.8 mM of KH2PO4, 7.4 mM of K2HPO4, 5% goat serum, 5% glycerol, 1% cold-water fish gelatin, and 0.04% NaN3, pH 7.2) for 30 minutes at 37°C. Primary and secondary antibodies were diluted in blocking buffer, 1:100 and 1:800, respectively, and filtered through a 0.45-μm filter. Slides were incubated with primary antibody for 1 h and washed three times with PBS. A rhodamine-conjugated goat anti-rabbit antibody (Sigma, St. Louis, MO, USA) was used as the secondary antibody. Cells on coverslips were incubated with secondary antibody for 1 h and washed three times with PBS. The Golgi apparatus was labeled with 5 μM of NBDC6-ceramide bovine serum albumin (BSA) complex (Molecular Probes, Eugene, OR, USA) for 1 hour and washed with PBS three times.(36) Endoplasmic reticulum labeling was performed with 50 μM of DiOC6 (Molecular Probes) for 1 h. After labeling, cells were washed with PBS three times.(37) Nuclear staining was performed with vital nuclear stain (Hoechst, Strasbourg, France) for 3 minutes. Coverslips were washed with PBS for 1 minute and 3 minutes. Coverslips were mounted with Prolong (Molecular Probes). Slides were examined using a Ziess Ax10vert 100 M laser scanning confocal microscope and imaging software (LSM510; Carl Ziess Instruments, Thornwood, NY, USA).


Immunostaining of mouse embryo and adult skeletal tissues with PHEX antibody

Figure 1 shows immunostaining of mouse vertebrae at 14, 16, and 18 days pc, respectively. At day 14 pc, Phex immunostaining is absent in the cells of primary centers of ossification in thoracolumbar and lumbar vertebrae of the mouse (Fig. 1A). On day 16 pc, there is staining in osteoblasts within primary ossifying centers (Fig. 1B). In contrast, chondrocytes do not contain epitopes for Phex. At day 18 pc, osteoblasts in primary ossification regions of vertebrae are clearly stained with the Phex antibody (Fig. 1C). Figure 1D shows that no staining is observed with preimmune serum in tissues obtained at day 16 pc. Immunostaining for Phex was noted in ossification centers in the limbs at day 16 pc (data not shown). Figure 1E shows immunohistochemical staining for Phex in calvarial tissue from a day 18 pc mouse embryo. Moderately intense staining of osteoblasts in the calvarium is noted. Interestingly, there is a similar intense staining in the stratum spinosum and granulosum of the overlying epidermis (Fig. 1E). Immunostaining for Phex in 2-week postnatal mouse calvarium reveals moderate staining of osteoblasts and osteocytes (Fig. 1F). At 4 weeks of age, very strong osteocyte staining is evident in compact bone of the calvarium (data not shown). No staining with preimmune serum was detected in pre- or postnatal tissues (data not shown).

Figure FIG. 1..

Immunohistochemical distribution of Phex in embryonic vertebra and in developing prenatal and postnatal mouse calvarium. (A) Day 14 pc mouse embryo lumbar vertebra (magnification ×200). (B) Day 16 pc vertebra. The arrow shows immunostaining of osteoblasts (magnification ×400). (C) Day 18 pc mouse embryo vertebra. The arrow shows staining of osteoblasts (magnification ×300). (D) Day 16 pc mouse embryo, vertebra. Negative control, immunostained with preimmune serum (magnification ×200). (E) Day 18 pc embryonic calvarium (magnification ×200). The large arrow shows Phex immunostaining of osteoblasts. A small arrow shows staining in skin. (F) Two-week postnatal mouse calvarium. Arrow shows Phex immunostaining of osteocyte (magnification ×200).

In Fig. 2, we show patterns of Phex immunohistochemical staining in tibias at 2 weeks and 12 weeks of age. At 2 weeks, there is moderate staining with the Phex antiserum in the osteoblasts and to a lesser degree in osteocytes of both the secondary and the primary centers of ossification (Fig. 2A and 2B). In the zones of the epiphyseal plate, no staining of chondrocytes was noted (Fig. 2A). No immunostaining with preimmune sera was evident (Fig. 2C). At 12 weeks, there was intense staining of osteocytes in compact bone (Fig. 2D).

Figure FIG. 2..

Immunohistochemical distribution of Phex in postnatal mouse tibia and teeth. (A) Two-week postnatal mouse tibia, small arrow shows Phex epitopes in osteoblasts. The large arrow shows chondrocytes which do not immunostain (magnification ×200). (B) Two-week postnatal mouse proximal tibia. The arrow shows Phex immunostaining in osteocytes (magnification ×200). (C) Two-week postnatal mouse tibia. Negative control section, stained with preimmune serum (magnification ×200). (D) Twelve-week postnatal mouse tibia. The arrow in panel D shows Phex epitopes in osteocytes (magnification ×200). (E) Two-week postnatal mouse tooth. The large arrow shows Phex staining in odontoblasts. The small arrow shows immunostaining in ameloblasts (magnification ×200). (F) Twelve-week postnatal mouse tooth. Large arrow shows Phex epitopes in odontoblasts and small arrow shows faint Phex epitopes in ameloblasts (magnification ×100).

Figures 2E and 2F show immunohistochemical staining for Phex in incisor teeth from 2-week-old and 12-week-old mice, respectively. In the 2-week mouse tooth, there was very intense staining of the odontoblasts and lesser staining of the ameloblasts along the basolateral border (Fig. 2E). Similarly, in the 12-week-old mouse, there is continued odontoblast staining but faint to absent staining of the ameloblasts (Fig. 2F). No staining of teeth was evident with preimmune serum (data not shown).

Expression and subcellular distribution of PHEX in transformed osteoblasts

Transformed osteoblasts from normal mice showed Phex epitopes in the perinuclear zone (Golgi) and in the endoplasmic reticulum (Figs. 3A-3C). No nuclear staining was observed when cells were examined by three-dimensional (3D) reconstruction scanning confocal microscopy. Osteoblasts from Hyp mice show background cytoplasmic staining (Figs. 3D-3F). Hyp osteoblasts transduced with the pJZ-PHEX showed endoplasmic reticulum and perinuclear (Golgi) PHEX staining similar to that seen in normal osteoblasts (Figs. 3G-3I).

Figure FIG. 3..

Immunofluorescent distribution of PHEX in normal osteoblasts, Hyp osteoblasts, and Hyp osteoblasts transduced with a PHEX viral expression vector. (A-C) Normal mouse osteoblasts (magnification ×150). (D-F) Hyp mouse osteoblasts (magnification ×150). (G-I) Transduced PHEX Hyp mouse osteoblasts (magnification ×150).

Further confirmation of subcellular localization of Phex protein in normal mouse osteoblasts was obtained with immunofluorescent probes, labeling the nucleus (Fig. 4A), the Golgi apparatus (Fig. 4B), and endoplasmic reticulum (figure not shown). PHEX (Fig. 4C) colocalizes with NBDC6 ceramide, a Golgi apparatus marker. Figure 4D is a composite of various cellular stains, which clearly shows the predominant Golgi localization of PHEX in these cells. Some cell surface staining was noted on confocal imaging. Cytoplasmic staining was noted also. Other work in our laboratories has shown that Phex protein is present at the plasma membrane of nonpermeabilized cells.(38)

Figure FIG. 4..

Immunofluorescent subcellular distribution of PHEX in normal mouse osteoblasts. (A) Nuclear staining (blue; magnification ×150). (B) Golgi apparatus staining (green; magnification ×150). (C) PHEX antibody (red; magnification ×150). (D) Multicolor composite PHEX (red), PHEX in Golgi apparatus staining (yellow), probe in Golgi apparatus (green), and nuclear staining (blue; magnification ×150).

Ribonuclease protection assay, using a riboprobe from the deleted region of the Phex gene, clearly showed abundant PHEX mRNA expression in virally transduced but not untransduced Hyp osteoblasts (Fig. 5A). Phex mRNA was detected also in normal osteoblasts but at levels below that seen in virally transformed Hyp osteoblasts. Western blotting showed a band of ∼97,000 Da in the virally transformed Hyp cells but not in untransformed Hyp cells; a faint band is present in normal osteoblasts (Fig. 5B). To characterize the glycosylation state of the PHEX protein, cell lysates from virally transformed Hyp cells were analyzed by Western blotting before and after treatment with either PNGase F or endo H. The PHEX protein was completely sensitive to digestion with PNGase F and a single protein band with a molecular weight of 86.5 kDa was apparent after the removal of N-linked oligosaccharide side chains (Fig. 5C). In contrast, the 97-kDa band was resistant to endo H digestion. These findings suggest that the expressed PHEX protein had undergone terminal glycosylation and represents a fully processed, mature form of the protein that has exited the endoplasmic reticulum and early Golgi.

Figure FIG. 5..

Assessment of PHEX/Phex mRNA, protein, and protein glycosylation in normal and Hyp osteoblasts and in Hyp osteoblasts transduced with a PHEX viral expression vector. (A) Ribonuclease protection assay. Relative abundance of PHEX/Phex mRNA was determined in virally transduced Hyp osteoblasts, Hyp osteoblasts, and normal osteoblasts (mean ± SEM; n = 3). (B) Western blot of cellular proteins from osteoblasts. Lane 1, virally transduced Hyp osteoblasts; lane 2, Hyp osteoblasts; lane 3, normal osteoblasts. (C) Incubation of PHEX protein from PHEX-transduced Hyp osteoblasts with PNGase F and endo H. Lane 1, no enzyme; lane 2, PNGase F alone; lane 3, endo H alone.

Immunostaining of mouse embryo extraskeletal tissue with PHEX antibody

A survey of tissues from internal organs of the day 18 pc mouse embryo reveals the absence of Phex immunostaining in kidney tubules and glomeruli (Fig. 6A), intestinal villi (Fig. 6B), skeletal muscle (Fig. 6C), and myocardium (Fig. 6E). Immunostaining for Phex also was absent in the developing lung and brain (data not shown). Interestingly, there is moderate staining of macrophages within the liver (Fig. 6D) and minimal expression in the brown adipose tissue (Fig. 6F). Of note, staining for Phex was observed in the suprabasal layers of the skin (Fig. 1E).

Figure FIG. 6..

Immunohistochemical distribution of various internal organs in a day 18 pc mouse. (A) Kidney (magnification ×200); (B) small intestine with villi (magnification ×100); (C) skeletal muscle (magnification ×100); (D) liver (magnification ×200), note macrophage staining; (E) heart (magnification ×100); (F) brown adipose tissue (magnification ×100).

Immunostaining of tissues from neonatal Hyp mice with PHEX antibody

Phex epitopes were not detected in skin (Fig. 7A, large arrow), calvaria (Fig. 7A, small arrow), vertebral region mesenchyme (Fig. 7B, large arrow), and vertebral osteoblast-like cells (Figs. 7B and 7C, small arrow). Some minimal staining was detected in brown adipose tissue (Fig. 7D), suggesting that the staining seen in normal and Hyp brown fat (Fig. 6F) is nonspecific.

Figure FIG. 7..

Immunostaining of tissues of neonatal Hyp mice with a PHEX/Phex antibody. (A) Skin and underlying calvarium (magnification ×l00). Note absent immunostaining in skin (large arrow) and calvarium (small arrow). (B) Vertebral body (magnification ×200). Note absent immunostaining in mesenchymal cells (large arrow) and osteoblasts (small arrow). (C) Vertebral body (magnification ×200). Note absent staining in osteoblasts (magnification ×200). (D) Brown adipose tissue (magnification ×200).


The aim of this study was to examine Phex protein expression patterns in the developing mouse embryo to obtain insights into the role of this protein in skeletal function. Additionally, we examined the pattern of Phex protein expression in osteoblasts and odontoblasts. In the developing mouse embryo, Phex expression is first seen in osteoblasts of the vertebral bodies and limbs at day 16 pc. In our study, Phex protein expression became more intense in vertebrae and limb long bones as development progressed. By day 18 pc, Phex expression was clearly evident in osteoblasts of vertebrae and calvarial tissues. Our findings of Phex protein expression patterns in skeletal tissues are similar to the patterns of Phex mRNA expression in mouse embryos using in situ hybridization.(18) In the latter study, Phex mRNA was first observed in vertebral regions and the calvaria around embryonic day 16. It is of interest that Phex protein is expressed in osteoblasts as soon as these cells appear in vertebrae and in ossifying centers in the limb buds. This suggests that the protein may be involved in early events of osteoblast maturation or mineralization in vivo. The lack of immunostaining in chondrocytes also is notable.

Although Phex protein expression is present in osteoblasts in various regions of the developing embryo, visceral abdominal organs such as the kidney, liver hepatocytes, and intestine do not express Phex. Cardiac muscle and skeletal muscle also are devoid of Phex epitopes. However, liver macrophages express Phex epitopes. No Phex epitopes are noted in the lung or brain. The absence of Phex staining in kidney, brain, and lung is in contrast to a previous study in which Phex mRNA was found in the tissues.(19) Phex protein may not be expressed in these tissues despite Phex mRNA expression. Alternatively, levels of protein expression are low and undetectable by our antibody. In normal mouse embryos, some immunostaining for Phex is observed in the stratum spinosum and granulosum of skin overlying the embryonic calvarium. The staining is attenuated in Hyp neonatal skin. The significance of Phex immunostaining in normal skin is unknown at present.

In adult tissues, Phex expression is present in osteoblasts and osteocytes. These expression patterns are similar to those noted previously using both immunohistochemical methods as well as antisense riboprobes.(17–21) Both Ruchon et al.(20) and Miao et al.(21) showed Phex immunostaining in osteoblasts and osteocytes of postnatal mice. Of particular interest in our study is the presence of Phex expression in osteoblasts of both primary and secondary ossification centers. Phex appears to be absent from chondrocytes in the ossification zone. Our findings also are consistent with previous studies in which Phex expression was observed in osteocytes.(20, 21)

We found intense Phex protein staining in odontoblasts as previously reported.(20) However, moderate staining was observed also in the ameloblasts along the basolateral border, particularly at 2 weeks of age. Phex epitopes were absent in ameloblasts of 12-week postnatal mice. This staining of ameloblasts at an early postnatal stage has not been reported before. Our findings show strong persistent staining in postnatal incisors, which in mice are continuously growing.

Analysis of the subcellular distribution of PHEX/Phex protein expression was performed on cultured osteoblasts. As seen in Figs. 3 and 4, Phex protein is present in the endoplasmic reticulum and Golgi apparatus of normal osteoblasts. Hyp osteoblasts do not show detectable Phex protein expression, whereas virally transformed osteoblasts show PHEX expression patterns similar to those found in normal cells. Because our antibody recognizes an epitope in the carboxy-terminal portion of the Phex protein, and because the epitope is within the three deleted regions of the Phex gene in Hyp mice, Phex protein expression would not be detectable in Hyp tissues. Thus, the minimal staining seen in the cytoplasm of normal and Hyp osteoblasts is nonspecific. Our findings that PHEX protein expressed in Hyp-transduced osteoblasts is sensitive to PNGase F digestion but resistant to endo H digestion indicate that the expressed PHEX protein has undergone terminal glycosylation and probably represents a fully processed, mature form of the protein. Some expression of PHEX was noted at the surface of the cell as well, consistent with the plasma membrane localization that characterizes this class of proteins.(6, 38) Moreover, Sabbagh et al. recently indicated that wild-type protein is localized to the plasma membrane of nonpermeabilized cells whereas proteins harboring disease-causing missense mutations in the PHEX gene are not detected at the cell surface under these conditions.(38) However, in permeabilized cells the PHEX proteins are detected in the endoplasmic reticulum.

In conclusion, our results show that Phex protein expression, as detected by immunohistochemical staining, is present in osteoblast-like cells of developing embryonic bone and in osteoblasts and osteocytes of maturer mouse bone. The protein is strongly expressed in osteoblasts in both primary and secondary ossification centers of growing bone. Additionally, there is strong expression of the Phex protein in odontoblasts of continuously growing incisors. With immunofluorescence, the subcellular distribution of Phex protein appears to be localized predominantly to the Golgi apparatus and to the endoplasmic reticulum although some staining is observed at the plasma membrane. In bone, Phex protein is a marker for cells of the osteoblast/osteocyte lineage.


We thank Susan Barrett for assistance with immunofluorescence and confocal microscopy, Linda Murphy for immunohistochemical staining studies, Claude Gauthier for PHEX mRNA analysis, and Hien Chau for harvesting newborn Hyp mice. This work was supported by NIH grants DK 25409, DK 58546 (to R.K.), AR 27032 (to M.K.D. and R.K.), and AI 47536 (to E.M.P.); the Canadian Institutes for Health Research grant 14107 (to H.S.T.); and an NIH training grant DK-07013 (to D.L.T.). Yves Sabbagh is the recipient of Studentship Awards from the Canadian Institutes of Health Research and FRSQ-FCAR-Santé.