Parathyroid Hormone Stimulates Circulating Osteogenic Cells in Hypoparathyroidism
Subjects and Methods
Subjects
Nineteen subjects with documented hypoparathyroidism and 19 controls participated in the study. Each hypoparathyroid subject was matched with an age- (within 3 yr), sex-, and menopause-matched control. The diagnosis of hypoparathyroidism was established by the simultaneous presence of serum calcium and PTH concentrations below the lower limits of normal on at least two prior occasions, separated by an interval of at least 30 d. Hypoparathyroidism had to have been present for at least 3 yr to establish a chronic state of PTH deprivation. All subjects had to be on stable regimens of supplemental calcium and vitamin D intake for 6 months before enrollment. Patients and controls were excluded if they had been on a bisphosphonate within 5 yr before study entry or for more than 6 months’ duration at any time or if they were women within 5 yr of menopause. Patients and controls were also excluded if they used any of the following medications: estrogens, progestins, raloxifene, calcitonin, systemic corticosteroids, fluoride, lithium, statins, loop diuretics, or methotrexate. Potentially confounding disorders were also exclusionary criteria, if present: Paget’s disease of bone, diabetes mellitus, chronic liver or renal disease, acromegaly, Cushing’s syndrome, rheumatoid arthritis, or multiple myeloma.
Patients were recruited from the Metabolic Bone Diseases Unit of Columbia University Medical Center and from the Hypoparathyroidism Association. The control subjects were recruited by local posted announcements. The study was approved by the Institutional Review Board of Columbia University Medical Center. All subjects and controls gave written informed consent.
Protocol
Time course of PTH administration
The hypoparathyroid subjects were given PTH (1-84), provided by NPS Pharmaceuticals (Bedminster, NJ), for 12 months at a sc dose of 100 μg every other day. This dose was selected because we showed previously that this regimen restores suppressed bone turnover markers in hypoparathyroidism to levels that are in the normal range (18). Circulating osteogenic cells were measured at baseline (T = 0) and after 1, 2, 3, 6, and 12 months of PTH. To validate osteogenic cell measurements, biochemical and histomorphometric indices were determined. Biochemical markers of bone formation [amino terminal propeptide of type I procollagen (19), bone specific alkaline phosphatase (20), and osteocalcin (16)] were measured at the same time points. Percutaneous iliac crest bone biopsies were performed in 12 of the 19 hypoparathyroid subjects with a quadruple tetracycline-labeling protocol (21). In contrast to conventional labeling with one set of tetracycline labels, two sets of tetracycline labels were sequentially administered, at baseline and 3 months after PTH (1-84), with the single biopsy obtained after 3 months. This method permits the determination of bone formation indices at baseline and after PTH (1-84) by measuring characteristics of each set of double labels separately and then comparing them with each other (21).
Biochemical markers of bone formation
Intact N-terminal propeptide of type I procollagen (P1NP) was measured by RIA (22). The interassay and intraassay variabilities are 6.0–10.2% (normal range: 19–83 and 16–96 μg/liter for pre- and postmenopausal women, respectively); osteocalcin was measured by ELISA (23) (N-mid osteocalcin; IDS Ltd., Fountain Hills, AZ) (mean ± sd: 17.9 ± 6.5 and 28.4 ± 9.5 ng/ml for pre- and postmenopausal women, respectively). The intraassay and interassay variability is 1.8 and 2.7%, respectively. Bone-specific alkaline phosphatase activity (BAP) was measured by immunoassay (20) (Metra BAP; Quidel Corp, San Diego, CA). This assay has low cross-reactivity with the liver form of alkaline phosphatase (3–8%). Interassay variability is 8.6% at 13.7 U/liter and 6.0% (normal range: 11.6–29.6 U/liter and 14.2–42.7 U/liter in pre- and postmenopausal women, respectively).
Histomorphometric assessment of bone formation with quadruple-labeled protocol
Two tetracycline labels were administered (Sumycin 250 mg four times daily) using a standard format of 3 d on, 12 d off, 3 d on immediately before initiation of PTH (21). After 3 months of PTH administration, the tetracycline labeling protocol was repeated using the same schedule but with a different tetracycline (Declomycin 150 mg four times daily). Percutaneous iliac crest biopsies were performed 1 wk after the second double-label protocol. The method yields two sets of fluorescent labels representing bone formation before (the first set) and after PTH (the second set) administration (21). Each set of double labels was easily distinguishable by color under fluorescent light. Biopsy specimens were processed and analyzed by histomorphometry as previously described in detail from our laboratory (24). Histomorphometry was performed using an OsteoMeasure digitizing image-analysis system (OsteoMetrics, Inc., Atlanta, GA). Bone formation was evaluated on cancellous, endocortical, or intracortical bone surfaces and expressed by the variables of mineralizing perimeter (Md.Pm), mineral apposition rate (MAR), and bone formation rate (BFR). All indices are expressed according to the recommendations of the American Society for Bone and Mineral Research Nomenclature Committee (25).
Flow cytometry and cell sorting
Peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation using Ficoll-Hypaque and were counted with Tryptan blue for viability using a hemacytometer. PBMCs were resuspended in flow-staining buffer [PBS plus 2% fetal bovine serum (FBS)] and the primary antibodies were added. After 30 min incubation at 4 C, the cells were washed twice and fluorochrome-conjugated primary and secondary antibodies were added. The cells were then incubated for an additional 30 min at 4C and washed twice before flow cytometry analysis. The primary unconjugated antibody was a goat polyclonal antihuman osteocalcin (Santa Cruz Biotechnology, Santa Cruz, CA) antibody (a control isotype antibody was used at the same concentrations); secondary conjugated antibodies included fluorescein isothiocyanate-conjugated AffinityPure IgG f(ab′)2 fragment donkey antigoat (Jackson ImmunoResearch, West Grove, PA) antibodies. Primary conjugated antibodies were allophycocyanin-conjugated anti-CD15, phycoerythrin (PE)-conjugated anti-CD146, and anti-PE-Cy7-conjugated CD34 (all from Becton Dickinson, San Diego, CA). Five-color flow cytometry acquisition was performed using a LSR II flow cytometer (Becton Dickinson) and analysis using FLO-JO software (Treestar, Inc., Ashland, OR). Cells were gated for size, shape, and granularity using forward- and side-scatter parameters. The positive populations were identified as cells that expressed specific levels of fluorescence activity above the nonspecific auto fluorescence of the isotype control. The region was set to encompass both the lymphocyte/monocyte-enriched area and the granulocyte-enriched area and to exclude dead cells. All CD15 granulocytes were excluded before gating for specific populations to exclude contamination of isolated mononuclear cells with granulocytes.
For flow sorting, PBMCs were resuspended in flow staining buffer at 1 × 10/ml and labeled with polyclonal antihuman osteocalcin. After 30 min incubation at 4 C, the cells were washed twice and the following fluorochrome-conjugated primary and secondary antibodies were added: fluorescein isothiocyanate-conjugated AffinityPure IgG f(ab′)2 fragment donkey antigoat antibody, allophycocyanin-conjugated anti-CD15, and PE-Cy7-conjugated anti-CD34. After 30 min incubation, cells were washed twice using flow buffer. Flow sorting was performed using FACSAria (BD). Cells were sorted into the following populations; OCN/CD34 and OCN/CD34. Sorted populations were then stored in RNeasy lysis tissue buffer at −80 C for extraction of RNA.
Expression analysis for osteoblast molecular markers
Total RNA was isolated from tissues using TRIZOL reagent (Invitrogen, Carlsbad, CA) followed by a clean-up step using the RNeasy minipurification kit (Qiagen, Valencia, CA). Only the RNA samples that gave A260/A280 about 1.8–1.9 in Nanodrop ND-1000 UV-VIS spectrophotometer (Thermo Scientific, Wilmington, DE) were used. One microgram of total RNA was first treated with deoxyribonuclease at room temperature, and RNA was then reverse transcribed using Superscript III reverse transcriptase at 42 C for 60 min. The resulting cDNAs were used for real-time PCR analysis of various genes using the Stratagene qPCR machine (La Jolla, CA). Primers for the assays were obtained from Superarray Biosciences (Frederick, MD). Reactions were set up in the total volume of 25 μl with the BIORAD 2X qPCR mix in triplicate (Bio-Rad Laboratories, Hercules, CA) for each sample and were measured against standard curves for respective genes. Using real-time PCR, the expression of osteoblast gene markers, including OCN, alkaline phosphatase (ALP), Runt-related transcription factor 2 (Runx2), and TGFβ, was evaluated in sorted OCN/CD34 and OCN/CD34 cells.
Osteogenic cultures
Flow-sorted OCN/CD34 cells were suspended in growth medium (MesenCult basal medium; Stem Cell Technologies, Vancouver, British Columbia, Canada) containing 10% FBS and 1% penicillin-streptomycin mixture and plated in fibronectin-coated plates (Becton Dickinson) at a plating density of 3.5 × 105 per square centimeter. On d 21 the medium was changed to osteogenic differentiation medium containing MesenCult basal medium with 15% osteogenic stimulatory supplements, 3.5 mm β-glycerophosphate, 10m dexamethasone, and 50 μg/ml ascorbic acid (Stem Cell Technologies). Parallel cultures were performed with only MesenCult basal medium containing 10% FBS and no osteogenic differentiation supplements. Throughout the culture of the cells, the total media with the nonadherent cell fraction from each well were aspirated once a week and washed once in the appropriate media, and the nonadherent cells with fresh media were added back to the respective wells. After 3 wk of differentiation, the cultured cells underwent either expression analysis of osteoblast markers or staining for calcium deposition. The expression of osteoblast gene markers in the cultured cells was assessed with real-time PCR and included ALP, Runx2, and Osterix. For calcium staining, the cultured cells were fixed in 10% formaldehyde and then stained for calcium deposits using 2% alizarin red (Millipore Chemicon, Billerica, MA).
Statistical analysis
Data are expressed as mean ± sem. For the case-control study, a matched pair analysis was performed. For the time-course study, estimates of change in indices from baseline were assessed with paired t tests. Linear regression was used to assess the relationship between the change in osteogenic cell populations and the change in biochemical markers of bone formation and to assess the relationship between the change in osteogenic cell populations and the change in dynamic histomorphometric indices of bone formation. A P < 0.05 was considered significant.
Subjects
Nineteen subjects with documented hypoparathyroidism and 19 controls participated in the study. Each hypoparathyroid subject was matched with an age- (within 3 yr), sex-, and menopause-matched control. The diagnosis of hypoparathyroidism was established by the simultaneous presence of serum calcium and PTH concentrations below the lower limits of normal on at least two prior occasions, separated by an interval of at least 30 d. Hypoparathyroidism had to have been present for at least 3 yr to establish a chronic state of PTH deprivation. All subjects had to be on stable regimens of supplemental calcium and vitamin D intake for 6 months before enrollment. Patients and controls were excluded if they had been on a bisphosphonate within 5 yr before study entry or for more than 6 months’ duration at any time or if they were women within 5 yr of menopause. Patients and controls were also excluded if they used any of the following medications: estrogens, progestins, raloxifene, calcitonin, systemic corticosteroids, fluoride, lithium, statins, loop diuretics, or methotrexate. Potentially confounding disorders were also exclusionary criteria, if present: Paget’s disease of bone, diabetes mellitus, chronic liver or renal disease, acromegaly, Cushing’s syndrome, rheumatoid arthritis, or multiple myeloma.
Patients were recruited from the Metabolic Bone Diseases Unit of Columbia University Medical Center and from the Hypoparathyroidism Association. The control subjects were recruited by local posted announcements. The study was approved by the Institutional Review Board of Columbia University Medical Center. All subjects and controls gave written informed consent.
Protocol
Time course of PTH administration
The hypoparathyroid subjects were given PTH (1-84), provided by NPS Pharmaceuticals (Bedminster, NJ), for 12 months at a sc dose of 100 μg every other day. This dose was selected because we showed previously that this regimen restores suppressed bone turnover markers in hypoparathyroidism to levels that are in the normal range (18). Circulating osteogenic cells were measured at baseline (T = 0) and after 1, 2, 3, 6, and 12 months of PTH. To validate osteogenic cell measurements, biochemical and histomorphometric indices were determined. Biochemical markers of bone formation [amino terminal propeptide of type I procollagen (19), bone specific alkaline phosphatase (20), and osteocalcin (16)] were measured at the same time points. Percutaneous iliac crest bone biopsies were performed in 12 of the 19 hypoparathyroid subjects with a quadruple tetracycline-labeling protocol (21). In contrast to conventional labeling with one set of tetracycline labels, two sets of tetracycline labels were sequentially administered, at baseline and 3 months after PTH (1-84), with the single biopsy obtained after 3 months. This method permits the determination of bone formation indices at baseline and after PTH (1-84) by measuring characteristics of each set of double labels separately and then comparing them with each other (21).
Time course of PTH administration
The hypoparathyroid subjects were given PTH (1-84), provided by NPS Pharmaceuticals (Bedminster, NJ), for 12 months at a sc dose of 100 μg every other day. This dose was selected because we showed previously that this regimen restores suppressed bone turnover markers in hypoparathyroidism to levels that are in the normal range (18). Circulating osteogenic cells were measured at baseline (T = 0) and after 1, 2, 3, 6, and 12 months of PTH. To validate osteogenic cell measurements, biochemical and histomorphometric indices were determined. Biochemical markers of bone formation [amino terminal propeptide of type I procollagen (19), bone specific alkaline phosphatase (20), and osteocalcin (16)] were measured at the same time points. Percutaneous iliac crest bone biopsies were performed in 12 of the 19 hypoparathyroid subjects with a quadruple tetracycline-labeling protocol (21). In contrast to conventional labeling with one set of tetracycline labels, two sets of tetracycline labels were sequentially administered, at baseline and 3 months after PTH (1-84), with the single biopsy obtained after 3 months. This method permits the determination of bone formation indices at baseline and after PTH (1-84) by measuring characteristics of each set of double labels separately and then comparing them with each other (21).
Biochemical markers of bone formation
Intact N-terminal propeptide of type I procollagen (P1NP) was measured by RIA (22). The interassay and intraassay variabilities are 6.0–10.2% (normal range: 19–83 and 16–96 μg/liter for pre- and postmenopausal women, respectively); osteocalcin was measured by ELISA (23) (N-mid osteocalcin; IDS Ltd., Fountain Hills, AZ) (mean ± sd: 17.9 ± 6.5 and 28.4 ± 9.5 ng/ml for pre- and postmenopausal women, respectively). The intraassay and interassay variability is 1.8 and 2.7%, respectively. Bone-specific alkaline phosphatase activity (BAP) was measured by immunoassay (20) (Metra BAP; Quidel Corp, San Diego, CA). This assay has low cross-reactivity with the liver form of alkaline phosphatase (3–8%). Interassay variability is 8.6% at 13.7 U/liter and 6.0% (normal range: 11.6–29.6 U/liter and 14.2–42.7 U/liter in pre- and postmenopausal women, respectively).
Histomorphometric assessment of bone formation with quadruple-labeled protocol
Two tetracycline labels were administered (Sumycin 250 mg four times daily) using a standard format of 3 d on, 12 d off, 3 d on immediately before initiation of PTH (21). After 3 months of PTH administration, the tetracycline labeling protocol was repeated using the same schedule but with a different tetracycline (Declomycin 150 mg four times daily). Percutaneous iliac crest biopsies were performed 1 wk after the second double-label protocol. The method yields two sets of fluorescent labels representing bone formation before (the first set) and after PTH (the second set) administration (21). Each set of double labels was easily distinguishable by color under fluorescent light. Biopsy specimens were processed and analyzed by histomorphometry as previously described in detail from our laboratory (24). Histomorphometry was performed using an OsteoMeasure digitizing image-analysis system (OsteoMetrics, Inc., Atlanta, GA). Bone formation was evaluated on cancellous, endocortical, or intracortical bone surfaces and expressed by the variables of mineralizing perimeter (Md.Pm), mineral apposition rate (MAR), and bone formation rate (BFR). All indices are expressed according to the recommendations of the American Society for Bone and Mineral Research Nomenclature Committee (25).
Flow cytometry and cell sorting
Peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation using Ficoll-Hypaque and were counted with Tryptan blue for viability using a hemacytometer. PBMCs were resuspended in flow-staining buffer [PBS plus 2% fetal bovine serum (FBS)] and the primary antibodies were added. After 30 min incubation at 4 C, the cells were washed twice and fluorochrome-conjugated primary and secondary antibodies were added. The cells were then incubated for an additional 30 min at 4C and washed twice before flow cytometry analysis. The primary unconjugated antibody was a goat polyclonal antihuman osteocalcin (Santa Cruz Biotechnology, Santa Cruz, CA) antibody (a control isotype antibody was used at the same concentrations); secondary conjugated antibodies included fluorescein isothiocyanate-conjugated AffinityPure IgG f(ab′)2 fragment donkey antigoat (Jackson ImmunoResearch, West Grove, PA) antibodies. Primary conjugated antibodies were allophycocyanin-conjugated anti-CD15, phycoerythrin (PE)-conjugated anti-CD146, and anti-PE-Cy7-conjugated CD34 (all from Becton Dickinson, San Diego, CA). Five-color flow cytometry acquisition was performed using a LSR II flow cytometer (Becton Dickinson) and analysis using FLO-JO software (Treestar, Inc., Ashland, OR). Cells were gated for size, shape, and granularity using forward- and side-scatter parameters. The positive populations were identified as cells that expressed specific levels of fluorescence activity above the nonspecific auto fluorescence of the isotype control. The region was set to encompass both the lymphocyte/monocyte-enriched area and the granulocyte-enriched area and to exclude dead cells. All CD15 granulocytes were excluded before gating for specific populations to exclude contamination of isolated mononuclear cells with granulocytes.
For flow sorting, PBMCs were resuspended in flow staining buffer at 1 × 10/ml and labeled with polyclonal antihuman osteocalcin. After 30 min incubation at 4 C, the cells were washed twice and the following fluorochrome-conjugated primary and secondary antibodies were added: fluorescein isothiocyanate-conjugated AffinityPure IgG f(ab′)2 fragment donkey antigoat antibody, allophycocyanin-conjugated anti-CD15, and PE-Cy7-conjugated anti-CD34. After 30 min incubation, cells were washed twice using flow buffer. Flow sorting was performed using FACSAria (BD). Cells were sorted into the following populations; OCN/CD34 and OCN/CD34. Sorted populations were then stored in RNeasy lysis tissue buffer at −80 C for extraction of RNA.
Expression analysis for osteoblast molecular markers
Total RNA was isolated from tissues using TRIZOL reagent (Invitrogen, Carlsbad, CA) followed by a clean-up step using the RNeasy minipurification kit (Qiagen, Valencia, CA). Only the RNA samples that gave A260/A280 about 1.8–1.9 in Nanodrop ND-1000 UV-VIS spectrophotometer (Thermo Scientific, Wilmington, DE) were used. One microgram of total RNA was first treated with deoxyribonuclease at room temperature, and RNA was then reverse transcribed using Superscript III reverse transcriptase at 42 C for 60 min. The resulting cDNAs were used for real-time PCR analysis of various genes using the Stratagene qPCR machine (La Jolla, CA). Primers for the assays were obtained from Superarray Biosciences (Frederick, MD). Reactions were set up in the total volume of 25 μl with the BIORAD 2X qPCR mix in triplicate (Bio-Rad Laboratories, Hercules, CA) for each sample and were measured against standard curves for respective genes. Using real-time PCR, the expression of osteoblast gene markers, including OCN, alkaline phosphatase (ALP), Runt-related transcription factor 2 (Runx2), and TGFβ, was evaluated in sorted OCN/CD34 and OCN/CD34 cells.
Osteogenic cultures
Flow-sorted OCN/CD34 cells were suspended in growth medium (MesenCult basal medium; Stem Cell Technologies, Vancouver, British Columbia, Canada) containing 10% FBS and 1% penicillin-streptomycin mixture and plated in fibronectin-coated plates (Becton Dickinson) at a plating density of 3.5 × 105 per square centimeter. On d 21 the medium was changed to osteogenic differentiation medium containing MesenCult basal medium with 15% osteogenic stimulatory supplements, 3.5 mm β-glycerophosphate, 10m dexamethasone, and 50 μg/ml ascorbic acid (Stem Cell Technologies). Parallel cultures were performed with only MesenCult basal medium containing 10% FBS and no osteogenic differentiation supplements. Throughout the culture of the cells, the total media with the nonadherent cell fraction from each well were aspirated once a week and washed once in the appropriate media, and the nonadherent cells with fresh media were added back to the respective wells. After 3 wk of differentiation, the cultured cells underwent either expression analysis of osteoblast markers or staining for calcium deposition. The expression of osteoblast gene markers in the cultured cells was assessed with real-time PCR and included ALP, Runx2, and Osterix. For calcium staining, the cultured cells were fixed in 10% formaldehyde and then stained for calcium deposits using 2% alizarin red (Millipore Chemicon, Billerica, MA).
Statistical analysis
Data are expressed as mean ± sem. For the case-control study, a matched pair analysis was performed. For the time-course study, estimates of change in indices from baseline were assessed with paired t tests. Linear regression was used to assess the relationship between the change in osteogenic cell populations and the change in biochemical markers of bone formation and to assess the relationship between the change in osteogenic cell populations and the change in dynamic histomorphometric indices of bone formation. A P < 0.05 was considered significant.
Results
Study population
The hypoparathyroid and control subjects were matched for age (hypoparathyroid: 41 ± 4 yr; controls: 42 ± 3 yr) and gender (each group: four males, nine premenopausal and six postmenopausal females). Hypoparathyroidism was postsurgical (n = 10) or idiopathic (n = 9), with a mean duration of 9.8 ± 3 yr (range 3–40 yr). Baseline serum calcium was 9.0 ± 0.2 mg/dl (2.25 ± 0.1 mmol/liter); PTH was less than 3 pg/ml (<3 ng/liter); baseline calcium supplementation was 2065 ± 263 mg/d (range 600-5000; median 1800 mg/d) and baseline calcitriol supplementation was 0.66 ± 0.1 μg/d (range 0–2; median 0.50 μg/d). Parent vitamin D supplementation was used in nine patients (range 400–100,000 IU/d). Nine of the subjects were on thyroid replacement medication. Of those nine, five had suppressed TSH values (mean 0.11 ± 0.1 μU/ml; normal range 0.34–4.25 μU/ml), and one had a TSH value that was slightly above the normal range (5.23 μU/ml). None of these individuals was clinically hypo- or hyperthyroid. The effect of these abnormal values on the analyses is described below.
Case control comparisons
The percentage of PBMCs that expressed OCN on the surface was significantly lower in the hypoparathyroid subjects, compared with the matched controls (0.7 ± 0.1 vs. 2.0 ± 0.1%; P < 0.0001; Figs. 11 and 2A2A);); a representative pair is shown in Fig. 11.. Although the overall population of OCN cells was lower, the hypoparathyroid subjects tended to show a greater proportion of cells that also expressed the early markers, CD34 and CD146. Although the subpopulation of OCN cells that coexpressed CD34 (OCN/CD34) showed a trend to be higher in the hypoparathyroid subjects (hypoparathyroid: 35.6 ± 1.5 vs. controls 30.9 ± 2.9%; P = 0.08; Fig. 2B2B),), OCN cells that coexpressed CD146 (OCN/CD146) were twice as high in the hypoparathyroid subjects (hypoparathyroid: 31.6 ± 1.8 vs. control: 15.1 ± 1.1%; P < 0.0001; Fig. 2C2C).). Similarly, there were significantly more OCN cells that coexpressed both CD34 and CD146 (OCN/CD34/CD146) in the hypoparathyroid subjects (hypoparathyroid: 11.0 ± 1.0 vs. control: 5.6 ± 0.7%; P = 0.0003; Fig. 2D2D).
Representative example of flow cytometry dot plots looking at OCN population. The isotype control shows the nonspecific binding, or background noise. To measure the specific expression of OCN, the isotype control is subtracted from the total OCN. In the control, the OCN cells are 2.9–0.6% of circulating PBMCs (2.3%), whereas in the matched hypoparathyroid (Hypo) subject, the OCN cells are 1.2–0.3% of circulating PBMCs (0.9%).
Comparison of cell populations in untreated hypoparathyroid subjects and matched controls. Data are mean ± sem. A, Comparison of OCN cells. The percentage of PBMCs that was positive for OCN was lower in the 19 hypoparathyroid subjects compared with the 19 matched controls. B, Comparison of OCN/CD34 cells. The percentage of OCN cells that was also positive for CD34 showed a trend to be higher in the hypoparathyroid subjects. C, Comparison of OCN/CD146 cells. The percentage of OCN cells that was also positive for CD146 was higher in the hypoparathyroid subjects. D, Comparison of OCN/CD34/CD146 cells. The percentage of OCN cells that was also positive for both CD34 and CD146 was higher in the hypoparathyroid subjects.
Time course after PTH (1-84) administration
With PTH (1-84) administration, the percentage of OCN cells in hypoparathyroid subjects increased significantly, peaking at 2 months. By 6 and 12 months, the percentage of cells decreased, although they were still higher than at baseline (Fig. 3A3A).). The profile of OCN cells also changed after PTH (1-84) administration. The OCN cells that did not show the early CD34 marker (OCN/CD34) increased, peaking at 3 months and decreasing by 6 and 12 months, although not to baseline levels (Fig. 3B3B).). The OCN cells that were CD146 (OCN/CD146) did not change. OCN cells that did not show either of the early markers (OCN/CD34/CD146) also increased with PTH (1-84) but at 12 months had returned to baseline (Fig. 3C3C).). The hematopoietic cell populations, which included total CD34 cells and OCN/CD34 cells, did not change with PTH administration. Exclusion of the six subjects with abnormal TSH values from the analysis did not alter these findings.
Changes in cell populations in hypoparathyroid subjects with PTH (1-84) treatment over 12 months. Data are mean ± sem. *, P < 0.05 from baseline; **, P < 0.001 from baseline. A, Change in OCN cells. The percentage of PBMCs that was OCN cells increased with PTH (1-84) treatment. B, Change in OCN/CD34 cells. The percentage of OCN cells that was lacking CD34 increased with PTH (1-84) treatment. C, Change in OCN/CD34/CD146 cells. The percentage of OCN cells that was lacking both CD34 and CD146 increased with PTH (1-84) treatment. D, Time course of increase in the OCN cell population compared with time course of increase in biochemical markers of bone formation with PTH (1-84) treatment.
Associations with changes in biochemical markers of bone formation
The increase in percentage of OCN cells from baseline to 3 months of PTH (1-84) administration was associated with increases in biochemical markers of bone formation: P1NP, BAP, and osteocalcin. The increase in the subpopulation of OCN cells lacking the early CD34 marker (OCN/CD34 cells) was also associated with increases in biochemical indices of bone formation (Table 11).). Increases in the numbers of subpopulations lacking early cell markers (OCN/CD146 and OCN/CD34/CD146) were associated with increases in P1NP, the most quickly responsive bone formation marker after PTH administration. The increases in osteoblast lineage populations preceded the increases in biochemical markers of bone formation (Fig. 3D3D).). Moreover, the osteogenic cells declined after 6 months of PTH (1-84), whereas the increase in biochemical markers persisted.
Table 1
Relationship between increases in circulating osteogenic cells from 0 to 3 months of PTH treatment and increases in biochemical indices of bone formation in hypoparathyroid subjects
| Increase in cell population | Increase in biochemical marker | R2 | Slope (β) | P value |
|---|---|---|---|---|
| Total OCN cells (%) | P1NP | 0.36 | 12.63 ± 3.2 | 0.0004 |
| BAP | 0.33 | 1.85 ± 0.5 | 0.0007 | |
| Osteocalcin | 0.35 | 2.12 ± 0.5 | 0.0005 | |
| Subpopulation OCN/CD34 cells (%) | P1NP | 0.20 | 1.20 ± 0.4 | 0.009 |
| BAP | 0.10 | 0.15 ± 0.07 | 0.04 | |
| Osteocalcin | 0.20 | 0.19 ± 0.07 | 0.01 | |
| Subpopulation OCN/CD146 cells (%) | P1NP | 0.10 | 0.92 ± 0.4 | 0.05 |
| BAP | 0.02 | 0.05 ± 0.07 | NS | |
| Osteocalcin | 0.04 | 0.08 ± 0.08 | NS | |
| Subpopulation OCN/CD34/ CD146 cells (%) | P1NP | 0.20 | 0.78 ± 0.3 | 0.007 |
| BAP | 0.10 | 0.08 ± 0.04 | NS | |
| Osteocalcin | 0.10 | 0.09 ± 0.05 | NS |
NS, Not significant.
Associations with changes in dynamic histomorphometric indices of bone formation
To establish further that the increase in the OCN cells after PTH administration reflects changes in dynamic histomorphometric indices, correlations were performed with directly measured mineralized perimeter, mineral apposition rate, and bone formation rate. Histomorphometric analysis showed marked increases in indices of bone formation by both qualitative (Fig. 44)) and quantitative (Table 22)) assessment. The increase in the percentage of OCN cells from baseline to 3 months of PTH (1-84) was associated with increases in all these histomorphometric indices. All three bone envelopes, namely cancellous, endocortical, and intracortical (Table 33),), showed similar changes. Increases in lineage cells that had lost the early CD34 or CD146 cell markers (alone or together) also were associated with increases in dynamic histomorphometric indices (Table 33).
A representative quadruple-label biopsy from a 25-yr-old hypoparathyroid woman before and after PTH treatment. The first set of double labels (bottom arrow) was acquired before PTH treatment, and the second set of double labels (top arrow) was acquired after 3 months of PTH treatment.
Table 2
Absolute values of histomorphometric indices of bone formation in quadruple-labeled bone biopsies in hypoparathyroid subjects at 0 and 3 months of PTH treatment (n = 12)
| Histomorphometric parameter | 0 months | 3 months | P value |
|---|---|---|---|
| Cancellous | |||
| Md.Pm (%) | 0.41 ± 0.2 | 6.55 ± 1.9 | <0.01 |
| MAR (μm/d) | 0.32 ± 0.1 | 0.67 ± 0.1 | <0.001 |
| BFR (μm/μm · d) | 0.003 ± 0.01 | 0.051 ± 0.02 | <0.01 |
| Endocortical | |||
| Md.Pm (%) | 0.82 ± 0.4 | 10.13 ± 2.5 | <0.01 |
| MAR μm/d) | 0.13 ± 0.1 | 0.61 ± 0.1 | <0.001 |
| BFR (μm/μm · d) | 0.004 ± 0.01 | 0.083 ± 0.02 | <0.01 |
| Intracortical | |||
| Md.Pm (%) | 4.04 ± 1.2 | 8.53 ± 1.7 | <0.01 |
| MAR (μm/d) | 0.70 ± 0.2 | 0.81 ± 0.1 | NS |
| BFR (μm/μm · d) | 0.045 ± 0.02 | 0.093 ± 0.02 | <0.01 |
Data are mean ± sem. NS, Not significant.
Table 3
Relationship between increases in circulating osteogenic cells from 0 to 3 months of PTH treatment and increases in histomorphometric indices of bone formation in quadruple-labeled bone biopsies in hypoparathyroid subjects (n = 12)
| Increase in cell population | Increase in histomorphometric parameter | R2 | Slope (β) | P value |
|---|---|---|---|---|
| Total OCN cells (%) | Cancellous | |||
| Md.Pm (%) | 0.5 | 3.041 ± 0.69 | 0.0003 | |
| MAR (μm/d) | 0.4 | 0.107 ± 0.03 | 0.003 | |
| BFR (μm/μm · d) | 0.5 | 0.024 ± 0.01 | 0.0005 | |
| Endocortical | ||||
| Md.Pm (%) | 0.6 | 4.205 ± 0.86 | 0.0001 | |
| MAR (μm/d) | 0.7 | 0.171 ± 0.03 | <0.0001 | |
| BFR (μm/μm · d) | 0.6 | 0.036 ± 0.01 | <0.0001 | |
| Intracortical | ||||
| Md.Pm (%) | 0.6 | 1.748 ± 0.36 | 0.0001 | |
| MAR (μm/d) | 0.2 | 0.067 ± 0.03 | 0.04 | |
| BFR (μm/μm · d) | 0.6 | 0.020 ± 0.01 | 0.0003 | |
| Subpopulation OCN/CD34 cells (%) | Cancellous | |||
| Md.Pm (%) | 0.3 | 0.265 ± 0.10 | 0.01 | |
| MAR (μm/d) | 0.4 | 0.012 ± 0.01 | 0.006 | |
| BFR (μm/μm · d) | 0.3 | 0.002 ± 0.01 | 0.01 | |
| Endocortical | ||||
| Md.Pm (%) | 0.5 | 0.431 ± 0.11 | 0.001 | |
| MAR (μm/d) | 0.6 | 0.017 ± 0.01 | 0.0003 | |
| BFR (μm/μm · d) | 0.5 | 0.004 ± 0.01 | 0.0007 | |
| Intracortical | ||||
| Md.Pm (%) | 0.5 | 0.192 ± 0.05 | 0.0005 | |
| MAR (μm/d) | 0.2 | 0.007 ± 0.01 | NS | |
| BFR (μm/μm · d) | 0.5 | 0.002 ± 0.01 | 0.0005 | |
| Subpopulation OCN/CD146 cells (%) | Cancellous | |||
| Md.Pm (%) | 0.3 | 0.286 ± 0.12 | 0.02 | |
| MAR (μm/d) | 0.1 | 0.007 ± 0.01 | NS | |
| BFR (μm/μm · d) | 0.2 | 0.002 ± 0.01 | 0.03 | |
| Endocortical | ||||
| Md.Pm (%) | 0.3 | 0.386 ± 0.15 | 0.02 | |
| MAR (μm/d) | 0.3 | 0.016 ± 0.01 | 0.008 | |
| BFR (μm/μm · d) | 0.3 | 0.004 ± 0.01 | 0.01 | |
| Intracortical | ||||
| Md.Pm (%) | 0.3 | 0.155 ± 0.06 | 0.03 | |
| MAR (μm/d) | 0.2 | 0.009 ± 0.01 | 0.05 | |
| BFR (μm/μm · d) | 0.3 | 0.002 ± 0.01 | 0.01 | |
| Subpopulation OCN/CD34/CD146 cells (%) | Cancellous | |||
| Md.Pm (%) | 0.3 | 0.166 ± 0.07 | 0.02 | |
| MAR (μm/d) | 0.2 | 0.006 ± 0.01 | 0.04 | |
| BFR (μm/μm · d) | 0.3 | 0.001 ± 0.01 | 0.02 | |
| Endocortical | ||||
| Md.Pm (%) | 0.3 | 0.233 ± 0.09 | 0.02 | |
| MAR (μm/d) | 0.5 | 0.011 ± 0.01 | 0.001 | |
| BFR (μm/μm · d) | 0.4 | 0.002 ± 0.01 | 0.007 | |
| Intracortical | ||||
| Md.Pm (%) | 0.4 | 0.117 ± 0.03 | 0.002 | |
| MAR (μm/d) | 0.3 | 0.006 ± 0.01 | 0.02 | |
| BFR (μm/μm · d) | 0.5 | 0.001 ± 0.01 | 0.001 |
Data are mean ± sem. NS, Not significant.
Expression analysis for osteoblast molecular markers in sorted cells
To provide evidence that the cells are osteogenic, expression profiling of sorted OCN/CD34 and OCN/CD34 cell populations was performed. The expression profiles of OCN, ALP, Runx2, and TGFβ were evaluated using real-time PCR. In addition to expressing osteocalcin (Fig. 55,, A and B), Runx2 was expressed in both populations, whereas the OCN/CD34 population also expressed ALP and TGFβ.
A and B, Expression of osteoblast differentiation markers in sorted osteogenic cells. OCN/CD34 cells and OCN/CD34 cell populations were isolated. The expression of osteoblast differentiation marker genes was assessed by real-time PCR. Bars indicate means of duplicate determinations. C, Expression of osteoblast differentiation markers in cultured OCN/CD34 cells. Sorted OCN/CD34 cells were cultured in growth medium for 3 wk followed by either osteogenic differentiation media (black bars) or vehicle (gray bars) for an additional 3 wk. The expression of osteoblast differentiation marker genes was assessed by real-time PCR. *, P < 0.05. D and E, Alizarin red staining in cultured OCN/CD34 cells. Sorted OCN/CD34 cells were cultured in growth medium for 3 wk followed by either osteogenic differentiation media (D) or vehicle (E) for an additional 3 wk. Panel D shows positive staining as well as the presence of nodules.
Expression analysis for osteoblast molecular markers in cultured cells
To provide further evidence that the cells are osteogenic, OCN/CD34 cells were isolated and cultured in growth medium for 3 wk and then in osteogenic differentiation medium for 3 wk. The expression profiles of Runx2, ALP, and Osterix were evaluated at the end of the 6 wk using real-time PCR. The osteoblast markers were expressed in the cultured cells (Fig. 5C5C).). As a control, OCN/CD34 cells were also cultured without osteogenic differentiation media; these cells had significantly less expression of the osteoblast markers.
Formation of mineralized nodules in cultured cells
In addition to the expression analysis, the cultured OCN/CD34 cells were stained at the end of the 6 wk period with alizarin red to assess calcium deposition. Alizarin red staining and formation of calcium nodules were observed (Fig. 5D5D).). No staining was observed in the control OCN/CD34 cells that were cultured without the osteogenic differentiation media (Fig. 5E5E).
Study population
The hypoparathyroid and control subjects were matched for age (hypoparathyroid: 41 ± 4 yr; controls: 42 ± 3 yr) and gender (each group: four males, nine premenopausal and six postmenopausal females). Hypoparathyroidism was postsurgical (n = 10) or idiopathic (n = 9), with a mean duration of 9.8 ± 3 yr (range 3–40 yr). Baseline serum calcium was 9.0 ± 0.2 mg/dl (2.25 ± 0.1 mmol/liter); PTH was less than 3 pg/ml (<3 ng/liter); baseline calcium supplementation was 2065 ± 263 mg/d (range 600-5000; median 1800 mg/d) and baseline calcitriol supplementation was 0.66 ± 0.1 μg/d (range 0–2; median 0.50 μg/d). Parent vitamin D supplementation was used in nine patients (range 400–100,000 IU/d). Nine of the subjects were on thyroid replacement medication. Of those nine, five had suppressed TSH values (mean 0.11 ± 0.1 μU/ml; normal range 0.34–4.25 μU/ml), and one had a TSH value that was slightly above the normal range (5.23 μU/ml). None of these individuals was clinically hypo- or hyperthyroid. The effect of these abnormal values on the analyses is described below.
Case control comparisons
The percentage of PBMCs that expressed OCN on the surface was significantly lower in the hypoparathyroid subjects, compared with the matched controls (0.7 ± 0.1 vs. 2.0 ± 0.1%; P < 0.0001; Figs. 11 and 2A2A);); a representative pair is shown in Fig. 11.. Although the overall population of OCN cells was lower, the hypoparathyroid subjects tended to show a greater proportion of cells that also expressed the early markers, CD34 and CD146. Although the subpopulation of OCN cells that coexpressed CD34 (OCN/CD34) showed a trend to be higher in the hypoparathyroid subjects (hypoparathyroid: 35.6 ± 1.5 vs. controls 30.9 ± 2.9%; P = 0.08; Fig. 2B2B),), OCN cells that coexpressed CD146 (OCN/CD146) were twice as high in the hypoparathyroid subjects (hypoparathyroid: 31.6 ± 1.8 vs. control: 15.1 ± 1.1%; P < 0.0001; Fig. 2C2C).). Similarly, there were significantly more OCN cells that coexpressed both CD34 and CD146 (OCN/CD34/CD146) in the hypoparathyroid subjects (hypoparathyroid: 11.0 ± 1.0 vs. control: 5.6 ± 0.7%; P = 0.0003; Fig. 2D2D).
Representative example of flow cytometry dot plots looking at OCN population. The isotype control shows the nonspecific binding, or background noise. To measure the specific expression of OCN, the isotype control is subtracted from the total OCN. In the control, the OCN cells are 2.9–0.6% of circulating PBMCs (2.3%), whereas in the matched hypoparathyroid (Hypo) subject, the OCN cells are 1.2–0.3% of circulating PBMCs (0.9%).
Comparison of cell populations in untreated hypoparathyroid subjects and matched controls. Data are mean ± sem. A, Comparison of OCN cells. The percentage of PBMCs that was positive for OCN was lower in the 19 hypoparathyroid subjects compared with the 19 matched controls. B, Comparison of OCN/CD34 cells. The percentage of OCN cells that was also positive for CD34 showed a trend to be higher in the hypoparathyroid subjects. C, Comparison of OCN/CD146 cells. The percentage of OCN cells that was also positive for CD146 was higher in the hypoparathyroid subjects. D, Comparison of OCN/CD34/CD146 cells. The percentage of OCN cells that was also positive for both CD34 and CD146 was higher in the hypoparathyroid subjects.
Time course after PTH (1-84) administration
With PTH (1-84) administration, the percentage of OCN cells in hypoparathyroid subjects increased significantly, peaking at 2 months. By 6 and 12 months, the percentage of cells decreased, although they were still higher than at baseline (Fig. 3A3A).). The profile of OCN cells also changed after PTH (1-84) administration. The OCN cells that did not show the early CD34 marker (OCN/CD34) increased, peaking at 3 months and decreasing by 6 and 12 months, although not to baseline levels (Fig. 3B3B).). The OCN cells that were CD146 (OCN/CD146) did not change. OCN cells that did not show either of the early markers (OCN/CD34/CD146) also increased with PTH (1-84) but at 12 months had returned to baseline (Fig. 3C3C).). The hematopoietic cell populations, which included total CD34 cells and OCN/CD34 cells, did not change with PTH administration. Exclusion of the six subjects with abnormal TSH values from the analysis did not alter these findings.
Changes in cell populations in hypoparathyroid subjects with PTH (1-84) treatment over 12 months. Data are mean ± sem. *, P < 0.05 from baseline; **, P < 0.001 from baseline. A, Change in OCN cells. The percentage of PBMCs that was OCN cells increased with PTH (1-84) treatment. B, Change in OCN/CD34 cells. The percentage of OCN cells that was lacking CD34 increased with PTH (1-84) treatment. C, Change in OCN/CD34/CD146 cells. The percentage of OCN cells that was lacking both CD34 and CD146 increased with PTH (1-84) treatment. D, Time course of increase in the OCN cell population compared with time course of increase in biochemical markers of bone formation with PTH (1-84) treatment.
Associations with changes in biochemical markers of bone formation
The increase in percentage of OCN cells from baseline to 3 months of PTH (1-84) administration was associated with increases in biochemical markers of bone formation: P1NP, BAP, and osteocalcin. The increase in the subpopulation of OCN cells lacking the early CD34 marker (OCN/CD34 cells) was also associated with increases in biochemical indices of bone formation (Table 11).). Increases in the numbers of subpopulations lacking early cell markers (OCN/CD146 and OCN/CD34/CD146) were associated with increases in P1NP, the most quickly responsive bone formation marker after PTH administration. The increases in osteoblast lineage populations preceded the increases in biochemical markers of bone formation (Fig. 3D3D).). Moreover, the osteogenic cells declined after 6 months of PTH (1-84), whereas the increase in biochemical markers persisted.
Table 1
Relationship between increases in circulating osteogenic cells from 0 to 3 months of PTH treatment and increases in biochemical indices of bone formation in hypoparathyroid subjects
| Increase in cell population | Increase in biochemical marker | R2 | Slope (β) | P value |
|---|---|---|---|---|
| Total OCN cells (%) | P1NP | 0.36 | 12.63 ± 3.2 | 0.0004 |
| BAP | 0.33 | 1.85 ± 0.5 | 0.0007 | |
| Osteocalcin | 0.35 | 2.12 ± 0.5 | 0.0005 | |
| Subpopulation OCN/CD34 cells (%) | P1NP | 0.20 | 1.20 ± 0.4 | 0.009 |
| BAP | 0.10 | 0.15 ± 0.07 | 0.04 | |
| Osteocalcin | 0.20 | 0.19 ± 0.07 | 0.01 | |
| Subpopulation OCN/CD146 cells (%) | P1NP | 0.10 | 0.92 ± 0.4 | 0.05 |
| BAP | 0.02 | 0.05 ± 0.07 | NS | |
| Osteocalcin | 0.04 | 0.08 ± 0.08 | NS | |
| Subpopulation OCN/CD34/ CD146 cells (%) | P1NP | 0.20 | 0.78 ± 0.3 | 0.007 |
| BAP | 0.10 | 0.08 ± 0.04 | NS | |
| Osteocalcin | 0.10 | 0.09 ± 0.05 | NS |
NS, Not significant.
Associations with changes in dynamic histomorphometric indices of bone formation
To establish further that the increase in the OCN cells after PTH administration reflects changes in dynamic histomorphometric indices, correlations were performed with directly measured mineralized perimeter, mineral apposition rate, and bone formation rate. Histomorphometric analysis showed marked increases in indices of bone formation by both qualitative (Fig. 44)) and quantitative (Table 22)) assessment. The increase in the percentage of OCN cells from baseline to 3 months of PTH (1-84) was associated with increases in all these histomorphometric indices. All three bone envelopes, namely cancellous, endocortical, and intracortical (Table 33),), showed similar changes. Increases in lineage cells that had lost the early CD34 or CD146 cell markers (alone or together) also were associated with increases in dynamic histomorphometric indices (Table 33).
A representative quadruple-label biopsy from a 25-yr-old hypoparathyroid woman before and after PTH treatment. The first set of double labels (bottom arrow) was acquired before PTH treatment, and the second set of double labels (top arrow) was acquired after 3 months of PTH treatment.
Table 2
Absolute values of histomorphometric indices of bone formation in quadruple-labeled bone biopsies in hypoparathyroid subjects at 0 and 3 months of PTH treatment (n = 12)
| Histomorphometric parameter | 0 months | 3 months | P value |
|---|---|---|---|
| Cancellous | |||
| Md.Pm (%) | 0.41 ± 0.2 | 6.55 ± 1.9 | <0.01 |
| MAR (μm/d) | 0.32 ± 0.1 | 0.67 ± 0.1 | <0.001 |
| BFR (μm/μm · d) | 0.003 ± 0.01 | 0.051 ± 0.02 | <0.01 |
| Endocortical | |||
| Md.Pm (%) | 0.82 ± 0.4 | 10.13 ± 2.5 | <0.01 |
| MAR μm/d) | 0.13 ± 0.1 | 0.61 ± 0.1 | <0.001 |
| BFR (μm/μm · d) | 0.004 ± 0.01 | 0.083 ± 0.02 | <0.01 |
| Intracortical | |||
| Md.Pm (%) | 4.04 ± 1.2 | 8.53 ± 1.7 | <0.01 |
| MAR (μm/d) | 0.70 ± 0.2 | 0.81 ± 0.1 | NS |
| BFR (μm/μm · d) | 0.045 ± 0.02 | 0.093 ± 0.02 | <0.01 |
Data are mean ± sem. NS, Not significant.
Table 3
Relationship between increases in circulating osteogenic cells from 0 to 3 months of PTH treatment and increases in histomorphometric indices of bone formation in quadruple-labeled bone biopsies in hypoparathyroid subjects (n = 12)
| Increase in cell population | Increase in histomorphometric parameter | R2 | Slope (β) | P value |
|---|---|---|---|---|
| Total OCN cells (%) | Cancellous | |||
| Md.Pm (%) | 0.5 | 3.041 ± 0.69 | 0.0003 | |
| MAR (μm/d) | 0.4 | 0.107 ± 0.03 | 0.003 | |
| BFR (μm/μm · d) | 0.5 | 0.024 ± 0.01 | 0.0005 | |
| Endocortical | ||||
| Md.Pm (%) | 0.6 | 4.205 ± 0.86 | 0.0001 | |
| MAR (μm/d) | 0.7 | 0.171 ± 0.03 | <0.0001 | |
| BFR (μm/μm · d) | 0.6 | 0.036 ± 0.01 | <0.0001 | |
| Intracortical | ||||
| Md.Pm (%) | 0.6 | 1.748 ± 0.36 | 0.0001 | |
| MAR (μm/d) | 0.2 | 0.067 ± 0.03 | 0.04 | |
| BFR (μm/μm · d) | 0.6 | 0.020 ± 0.01 | 0.0003 | |
| Subpopulation OCN/CD34 cells (%) | Cancellous | |||
| Md.Pm (%) | 0.3 | 0.265 ± 0.10 | 0.01 | |
| MAR (μm/d) | 0.4 | 0.012 ± 0.01 | 0.006 | |
| BFR (μm/μm · d) | 0.3 | 0.002 ± 0.01 | 0.01 | |
| Endocortical | ||||
| Md.Pm (%) | 0.5 | 0.431 ± 0.11 | 0.001 | |
| MAR (μm/d) | 0.6 | 0.017 ± 0.01 | 0.0003 | |
| BFR (μm/μm · d) | 0.5 | 0.004 ± 0.01 | 0.0007 | |
| Intracortical | ||||
| Md.Pm (%) | 0.5 | 0.192 ± 0.05 | 0.0005 | |
| MAR (μm/d) | 0.2 | 0.007 ± 0.01 | NS | |
| BFR (μm/μm · d) | 0.5 | 0.002 ± 0.01 | 0.0005 | |
| Subpopulation OCN/CD146 cells (%) | Cancellous | |||
| Md.Pm (%) | 0.3 | 0.286 ± 0.12 | 0.02 | |
| MAR (μm/d) | 0.1 | 0.007 ± 0.01 | NS | |
| BFR (μm/μm · d) | 0.2 | 0.002 ± 0.01 | 0.03 | |
| Endocortical | ||||
| Md.Pm (%) | 0.3 | 0.386 ± 0.15 | 0.02 | |
| MAR (μm/d) | 0.3 | 0.016 ± 0.01 | 0.008 | |
| BFR (μm/μm · d) | 0.3 | 0.004 ± 0.01 | 0.01 | |
| Intracortical | ||||
| Md.Pm (%) | 0.3 | 0.155 ± 0.06 | 0.03 | |
| MAR (μm/d) | 0.2 | 0.009 ± 0.01 | 0.05 | |
| BFR (μm/μm · d) | 0.3 | 0.002 ± 0.01 | 0.01 | |
| Subpopulation OCN/CD34/CD146 cells (%) | Cancellous | |||
| Md.Pm (%) | 0.3 | 0.166 ± 0.07 | 0.02 | |
| MAR (μm/d) | 0.2 | 0.006 ± 0.01 | 0.04 | |
| BFR (μm/μm · d) | 0.3 | 0.001 ± 0.01 | 0.02 | |
| Endocortical | ||||
| Md.Pm (%) | 0.3 | 0.233 ± 0.09 | 0.02 | |
| MAR (μm/d) | 0.5 | 0.011 ± 0.01 | 0.001 | |
| BFR (μm/μm · d) | 0.4 | 0.002 ± 0.01 | 0.007 | |
| Intracortical | ||||
| Md.Pm (%) | 0.4 | 0.117 ± 0.03 | 0.002 | |
| MAR (μm/d) | 0.3 | 0.006 ± 0.01 | 0.02 | |
| BFR (μm/μm · d) | 0.5 | 0.001 ± 0.01 | 0.001 |
Data are mean ± sem. NS, Not significant.
Expression analysis for osteoblast molecular markers in sorted cells
To provide evidence that the cells are osteogenic, expression profiling of sorted OCN/CD34 and OCN/CD34 cell populations was performed. The expression profiles of OCN, ALP, Runx2, and TGFβ were evaluated using real-time PCR. In addition to expressing osteocalcin (Fig. 55,, A and B), Runx2 was expressed in both populations, whereas the OCN/CD34 population also expressed ALP and TGFβ.
A and B, Expression of osteoblast differentiation markers in sorted osteogenic cells. OCN/CD34 cells and OCN/CD34 cell populations were isolated. The expression of osteoblast differentiation marker genes was assessed by real-time PCR. Bars indicate means of duplicate determinations. C, Expression of osteoblast differentiation markers in cultured OCN/CD34 cells. Sorted OCN/CD34 cells were cultured in growth medium for 3 wk followed by either osteogenic differentiation media (black bars) or vehicle (gray bars) for an additional 3 wk. The expression of osteoblast differentiation marker genes was assessed by real-time PCR. *, P < 0.05. D and E, Alizarin red staining in cultured OCN/CD34 cells. Sorted OCN/CD34 cells were cultured in growth medium for 3 wk followed by either osteogenic differentiation media (D) or vehicle (E) for an additional 3 wk. Panel D shows positive staining as well as the presence of nodules.
Expression analysis for osteoblast molecular markers in cultured cells
To provide further evidence that the cells are osteogenic, OCN/CD34 cells were isolated and cultured in growth medium for 3 wk and then in osteogenic differentiation medium for 3 wk. The expression profiles of Runx2, ALP, and Osterix were evaluated at the end of the 6 wk using real-time PCR. The osteoblast markers were expressed in the cultured cells (Fig. 5C5C).). As a control, OCN/CD34 cells were also cultured without osteogenic differentiation media; these cells had significantly less expression of the osteoblast markers.
Formation of mineralized nodules in cultured cells
In addition to the expression analysis, the cultured OCN/CD34 cells were stained at the end of the 6 wk period with alizarin red to assess calcium deposition. Alizarin red staining and formation of calcium nodules were observed (Fig. 5D5D).). No staining was observed in the control OCN/CD34 cells that were cultured without the osteogenic differentiation media (Fig. 5E5E).
Discussion
The clinical model of hypoparathyroidism, in which PTH is chronically absent, has permitted us to gain insights into the effects of PTH on osteoblast number, development, and function. We have shown that circulating osteogenic cells are reduced in hypoparathyroid subjects compared with matched controls. The results also demonstrate that a greater proportion of these cells have an early cell marker and thus may be immature. With PTH (1-84) administration, the number of circulating osteogenic cells increases significantly, peaking at 2–3 months, and returning closer to baseline levels by 12 months. The maturity stage of the osteogenic cells may also increase, with a marked reduction in OCN cells that contain markers associated with earlier stages of osteoblast development. Measurements of the circulating osteogenic cell populations correlated with well-established biochemical, histomorphometric, and molecular parameters of osteoblast function. The positive correlations between the measurements of circulating cells with the osteoblast phenotype and well-established indices of osteoblast function in the circulation as well as histomorphometric correlations help to confirm that PTH stimulates bone formation by actions on osteoblasts. Moreover, our observations that PTH is associated with an apparent increase in the proportion of cells that have lost early markers suggest that it may also help to regulate osteoblast development.
Bone marrow CD34 cells, which have been thought of as hematopoietic stem cells, are also regarded as early osteoblastic cells (15,16). CD34 cells are able to form mineralized nodules (26) . Khosla and colleagues (16) found that CD34 cells constitute 35% of OCN cells in human peripheral blood. Similarly, CD34 cells have been shown to localize to the site of a fracture in rats and to increase the differentiation of local stem cells into osteoblasts, leading to fracture repair (15). We thus used the presence of CD34 as a marker to assess the maturity of the OCN cells. Molecular expression profiling suggested that osteoblastic genes became detectable in cells that lost the CD34 marker, suggesting that they are more mature. We also found the marker CD146 to be useful because bone marrow stromal cells that have this surface marker, also known as MUC18, have been shown by Bianco and colleagues (17) to be osteogenic. In human subjects, CD146 subendothelial cells have been reported to serve as skeletal progenitors capable of generating cells that organize a hematopoietic microenvironment on transplantation (17). Whether the presence of the CD146 marker reflects an earlier or later stage in the osteoblastic lineage than the presence of the CD34 marker is unclear. Both markers appear to be characteristic of progenitor cells. Further expression studies would be necessary to determine the precise lineage sequence.
With PTH, the increases in osteoblast numbers that showed characteristics of more mature cells antedated increases in histomorphometric indices of bone formation, the gold standard by which bone formation is directly ascertained. The quadruple-label protocol allowed histological confirmation on bone formation dynamics both before and after 3 months of PTH administration. The observation that cancellous, endocortical, and intracortical envelopes of bone were similarly affected by PTH suggests that the changes in circulating osteogenic cells are reflected in all three bone envelopes. A conceptual construct is suggested by these data, namely that new circulating osteogenic cells are recruited and differentiated by PTH exposure. The time course indicates an early stimulation of number and differentiation followed by reduction to levels more typical of euparathyroid subjects.
The correlations between osteogenic cells and the serum indices of bone formation were not as strong as the correlations between osteogenic cells and histomorphometric indices of bone formation. This discrepancy could be explained by the timing of the samples. The osteogenic cells peaked by 3 months of PTH treatment, which was also when the bone biopsies were performed. On the other hand, biochemical markers of bone turnover, which were drawn monthly, rose later than the osteogenic cells, peaking at 6 months and remaining elevated thereafter. This time-course discordance between the osteogenic cells and the biochemical markers likely explains the weaker temporal correlations between them.
The effect of PTH on hematopoietic stem cell niche expansion requires further investigation. We found that circulating hematopoietic stem cells, as represented by total CD34 and OCN/CD34 cells did not change with PTH administration. Our expectation was that these populations would have increased; that in addition to mobilizing osteoblast lineage cells, which support hematopoiesis, PTH would similarly mobilize hematopoietic stem cells. Such a finding would have been consistent with recent observations in mice illustrating that mobilization of hematopoietic stem cells occurs within 14 d of PTH administration (27). It is in fact possible that such an increase did occur but that it occurred before the 1-month time point that was measured. Another possibility is that PTH is altering hematopoietic stem cell engraftment (29,30). Bianco and colleagues (17) have shown that CD45/CD34/CD146 marrow cells contain a subpopulation of cells that can reconstitute a complete bone that is hospitable to bone marrow. These observations have recently been extended to include an even more specific subset of cell markers essential for bone regeneration (31). Further investigations delineating the effects of PTH on these cell types will help to address this point.
A limitation to this study is that the characterization of the circulating osteogenic cells, including their relationship to mature osteoblasts on the bone surface, is still preliminary. It is possible that the OCN cells might be providing information about other cells of the mesenchymal lineage, such as adipocytes and chodrocytes, instead of osteoblasts. However, the evidence for the osteogenic identity of these cells includes our observed correlations with established serum and histomorphometric markers of osteoblast function as well as the pertubation in the cells coincident with PTH administration. Further supportive evidence comes from the molecular expression data in the sorted cells. Although the expression of osteocalcin mRNA is to be expected because the cells were sorted using an antibody to osteocalcin, the expression of other key osteoblastic genes in cell cultures supports the osteogenic nature of these cells. In addition, cultures of the sorted OCN cells showed that they also expressed osteoblastic genes and had the potential to form mineralized nodules in vitro. Our data thus suggest that selection for these peripheral cells enriches for an osteogenic population capable of mineralization. Measurement of osteoblast numbers and how they correlate with circulating osteogenic precursor cells could be an additional way to validate the measurement of these cells. Furthermore, recent data suggest that the mesenchymal marker CD73 might identify early osteogenic cells (28). Human embryonic stem cells that are initially double positive for both CD34 and CD73 (CD34/CD73) subsequently lose CD34 expression. The cells that remain, namely those that express only CD73, are capable of multilineage mesenchymal differentiation, including into osteoblasts (28). Use of this marker, as well as investigations to confirm that circulating osteogenic cells ultimately return to the marrow as osteoblast precursors, are important future areas of investigation.
In conclusion, these results provide evidence that the number and possible maturity of circulating osteogenic cells are reduced in hypoparathyroidism, a clinical model in which PTH is absent. The actions of PTH administration to increase the number and possibly the maturity of these cells, in the context of this clinical model, help to support the hypothesis that PTH stimulates bone formation by actions on the number and maturation of osteogenic cells.
Abstract
Context: The osteoanabolic properties of PTH may be due to increases in the number and maturity of circulating osteogenic cells. Hypoparathyroidism is a useful clinical model because this hypothesis can be tested by administering PTH.
Objective: The objective of the study was to characterize circulating osteogenic cells in hypoparathyroid subjects during 12 months of PTH (1-84) administration.
Design: Osteogenic cells were characterized using flow cytometry and antibodies against osteocalcin, an osteoblast-specific protein product, and stem cell markers CD34 and CD146. Changes in bone formation from biochemical markers and quadruple-labeled transiliac crest bone biopsies (0 and 3 month time points) were correlated with measurements of circulating osteogenic cells.
Setting: The study was conducted at a clinical research center.
Patients: Nineteen control and 19 hypoparathyroid patients were included in the study.
Intervention: Intervention included the administration of PTH (1-84).
Results: Osteocalcin-positive cells were lower in hypoparathyroid subjects than controls (0.7 ± 0.1 vs. 2.0 ± 0.1%; P < 0.0001), with greater coexpression of the early cell markers CD34 and CD146 among the osteocalcin-positive cells in the hypoparathyroid subjects (11.0 ± 1.0 vs. 5.6 ± 0.7%; P < 0.001). With PTH (1-84) administration, the number of osteogenic cells increased 3-fold (P < 0.0001), whereas the coexpression of the early cell markers CD34 and CD146 decreased. Increases in osteogenic cells correlated with circulating and histomorphometric indices of osteoblast function: N-terminal propeptide of type I procollagen (R = 0.4, P ≤ 0.001), bone-specific alkaline phosphatase (R = 0.3, P < 0.001), osteocalcin (R = 0.4, P < 0.001), mineralized perimeter (R = 0.5, P < 0.001), mineral apposition rate (R = 0.4, P = 0.003), and bone formation rate (R = 0.5, P < 0.001).
Conclusions: It is likely that PTH stimulates bone formation by stimulating osteoblast development and maturation. Correlations between circulating osteogenic cells and histomorphometric indices of bone formation establish that osteoblast activity is being identified by this methodology.
The mechanisms by which PTH increases bone formation are not well understood. When PTH acts on bone, communication between several different specialized cell types occurs, including osteoblasts, bone marrow and stromal cells, hematopoietic precursors of osteoclasts, and mature osteoclasts and osteocytes (1). PTH increases the number of osteoblasts by decreasing apoptosis of preosteoblast cells and osteoblasts (2,3) and increasing osteoblast proliferation (4). Increasing the supply of osteoblast precursor cells in the bone marrow is another way that PTH might stimulate bone formation (5).
Osteoblast precursor cells were previously identified by specific qualities, namely their ability to adhere to plastic, to stimulate the expression of osteoblast-related genes and to form mineralized nodules (6,7). In the circulation, such plastic-adherent osteoblastic cells have been demonstrated (8,9), but they are present in exceedingly low concentrations (9). Thus, until recently, osteoblast lineage cells were difficult to study in the peripheral blood. Long et al. (10,11) demonstrated that in addition to the adherent bone marrow-derived osteoblastic cells, there is another population of nonadherent cells with osteogenic potential in the circulation. Using flow cytometry with antibodies to the osteoblast-specific protein products osteocalcin and bone-specific alkaline phosphatase to identify osteogenic cells in the circulation, the investigators demonstrated that these osteogenic cells do, in fact, circulate in physiologically significant numbers (12). This concept has also been supported by studies in mice, which have similarly shown that osteogenic cells can be readily found in nonadherent circulating stromal cells (13,14).
Potential markers of osteoblast progenitor cells include other cell surface markers, besides osteocalcin. CD34, for example, a marker of hematopoietic stem cells, is also a marker for cells that give rise to functional osteoblasts that are capable of forming mineralized nodules and can heal fractures (15). Cells that are positive for both osteocalcin and CD34 (OCN/CD34) are probably more immature than cells that are positive only for osteocalcin (OCN/CD34). The cells that are positive for OCN and CD34 diminish when osteoblasts develop (16). Bianco and colleagues (17) suggested that CD 146 stromal cells also function as self-renewing, clonogenic skeletal progenitors and that CD146 may be a marker of early osteoblasts.
Hypoparathyroidism is a useful model to study the actions of PTH on circulating osteogenic cells. Because endogenous PTH is absent, this disorder can be used to evaluate the effects of PTH deficiency and its administration on these cells. To this end, we studied 19 untreated hypoparathyroid subjects before and after 1, 2, 3, 6, and 12 months of PTH (1-84) administration. At each time point, we measured the number of circulating osteogenic cells as well as early cell markers, CD34 and CD146. We hypothesized that the number of circulating osteogenic cells would be lower in hypoparathyroid subjects, compared with controls, with an altered profile of early cell markers that would change with PTH exposure. To validate our methodology, we correlated changes in osteoblast lineage populations in the hypoparathyroid subjects after PTH (1-84) administration with changes in circulating bone formation markers and data from percutaneous iliac crest bone biopsies. For histomorphometric assessment, we used the quadruple-labeled iliac crest bone biopsy, a method that permits time-dependent assessment of changes after PTH (1-84) administration with only a single biopsy. Our data support the hypothesis that PTH regulates osteoblast development and may facilitate the maturation of osteogenic cells.
NS, Not significant.
Data are mean ± sem. NS, Not significant.
Data are mean ± sem. NS, Not significant.
Footnotes
This work was supported by Grant DK077696 and Florence Irving Research Award FD-R-02525.
Disclosure Summary: The authors have nothing to declare.
First Published Online September 29, 2010
Abbreviations: ALP, Alkaline phosphatase; BAP, bone-specific alkaline phosphatase activity; BFR, bone formation rate; FBS, fetal bovine serum; MAR, mineral apposition rate; Md.Pm, mineralizing perimeter; OCN, osteocalcin; PBMC, peripheral blood mononuclear cell; PE, phycoerythrin; P1NP, N-terminal propeptide of type I procollagen; Runx2, Runt-related transcription factor 2.