Abstract
Introduction
The metalloproteinase, pregnancy-associated plasma protein-A (PAPP-A) functions to enhance local insulin-like growth factor (IGF)-I bioavailability through cleavage of inhibitory IGF binding proteins. Because IGF-I is an important regulator of skeletal growth and remodeling and PAPP-A is highly expressed by osteoblastic cells, we hypothesized that, in the absence of PAPP-A, bone physiology would be compromised because of a blunting of local IGF-I action even in the presence of normal circulating IGF-I levels.
Materials and Methods
pQCT, μCT, histomorphometry, and mechanical strength testing were performed on bones from PAPP-A knockout (KO) mice and wildtype (WT) littermates at 2–12 mo of age. IGF-I levels and bone formation and resorption markers were determined in sera from these animals.
Results
Volumetric BMD in PAPP-A KO mice measured by pQCT at the femoral midshaft, which is primarily cortical bone, was 10% less than WT at 2 mo. This difference was maintained at 4, 6, and 12 mo. Cortical thickness at this site was similarly decreased. On the other hand, trabecular bone at the distal femur (pQCT) and in the tibia (μCT) showed age-progressive decreases in bone volume fraction in PAPP-A KO compared with WT mice. Tibial μCT indicated a 46% relative decrease in trabecular bone volume/total volume (BV/TV) and a 28% relative decrease in trabecular thickness in PAPP-A KO compared with WT mice at 6 mo. These trabecular deficiencies in PAPP-A KO mice corresponded to a weakening of the bone. Serum markers and bone histomorphometry indicated that the primary impact of PAPP-A is on skeletal remodeling resulting in a state of low-turnover osteopenia in adult PAPP-A KO mice. Circulating IGF-I levels were not altered in PAPP-A KO mice.
Conclusions
PAPP-A is a bone growth regulatory factor in vivo and, in its absence, mice show skeletal insufficiency in mass, density, architecture, and strength. The data suggest a primary role for PAPP-A in modulating local IGF bioavailability for trabecular bone remodeling.
Key words: pregnancy-associated plasma protein-A, insulin-like growth factor, knock-out mouse, BMD, skeletal physiology
INTRODUCTION
The insulin-like growth factor (IGF) system is of major importance in skeletal physiology. IGF-I and IGF-II are established bone growth factors, stimulating DNA and matrix protein synthesis through type I IGF receptors present on cells of the osteoblast lineage, as well as influencing cell survival.(1,2) Signaling through the type I IGF receptor is essential for normal skeletal development. Thus, bone formation is retarded in mice lacking functional IGF-I or IGF-I receptor genes.(3–5) In addition, mice homozygous for targeted disruption of the gene for insulin receptor substrate (IRS)-1, a major mediator of IGF signaling, show impaired embryonal and postnatal growth of the skeleton.(6) The importance of local IGF expression for maintenance of bone growth was reinforced by studies in transgenic mice with liver-specific inactivation of IGF-I gene expression.(7,8) However, the complexity of the IGF system with both endocrine and autocrine/paracrine functions has presented a challenge to our full understanding of its role in skeletal development and remodeling.
IGF binding proteins (IGFBPs) are the ultimate regulators of local IGF bioactivity.(9) IGFBPs are produced by osteoblasts in vitro and in vivo.(1,2) IGFBP-4, the predominant IGFBP expressed by human and rodent bone cells, is a negative regulator of IGF action in vitro.(10,11) In addition, studies by Miyakoshi et al.(12) showed an inhibitory effect of locally administered IGFBP-4 on bone growth in vivo. Zhang et al.(13) showed that targeted IGFBP-4 expression in osteoblasts sequesters IGF with consequent impairment of IGF action in skeletal tissue of these transgenic mice.
Importantly, the structure and function of IGFBPs can be modified by specific IGFBP proteases, thereby modulating bone cell response to IGFs.(14,15) An IGFBP-4 protease produced by cultured cells, including osteoblasts, cleaves IGFBP-4 and potentiates the effectiveness of exogenous IGF-stimulated growth.(16,17) Studies using wildtype and protease-resistant IGFBP-4 provided evidence that IGFBP-4 proteolysis may be important for bone formation in vitro and in vivo.(18,19) The IGFBP-4 protease expressed by human osteoblasts was identified as pregnancy-associated plasma protein-A (PAPP-A).(20–22) To address the physiological role of PAPP-A, we developed a mouse model with targeted disruption of the PAPP-A gene achieved through homologous recombination in embryonic stem cells. These mice showed a diminished growth response during early embryogenesis when heightened IGF-II activity is important, resulting in a delay in ossification and reduced overall body size.(23,24) The expression of PAPP-A mRNA and protein in skeletal tissue of wildtype mice predict a role for PAPP-A in postnatal bone growth as well, when IGF-I is dominant.(23) In this study, we investigated the impact of PAPP-A deletion on postnatal skeletal growth and development.
MATERIALS AND METHODS
Wildtype and PAPP-A knockout mice
Generation of mice with targeted disruption of the PAPP-A gene on a mixed C57Bl6/129 genetic background was described previously.(23) Littermates obtained by breeding heterozygous males and females were used for phenotypic analyses. Genotypes were confirmed with liver or tail DNA collected at time of death. All procedures were approved by the Institutional Animal Care and Use Committee of the Mayo Clinic.
pQCT
Volumetric BMD (vBMD) was measured by pQCT at the midshaft and distal metaphysis of the femur using a Stratec XCT Research SA Plus scanner with v 5.40 software (Norland Medical Systems, Fort Atkinson, WI, USA). Each bone at each site was scanned in triplicate using scanning parameters that have been described elsewhere.(25) Resolution of cortical and trabecular bone was considered inadequate to provide reliable information because of the small size of the bone, so only total BMD is reported for each site.
μCT
Tibias from mice were scanned using μCT (MicroCT40; Scanco Medical, Bassersdorf, Switzerland) to evaluate trabecular bone volume fraction and microarchitecture in the metaphyseal region of the proximal tibias.(26,27) The MicroCT40 unit is calibrated weekly with a phantom standard provided by Scanco before beginning bone scans. The tibias were scanned at low resolution, energy level of 55 KeV, and intensity of 145 μA. The proximal trabecular scan started ∼1.2 mm distal to the growth plate and extended distally 1.5 mm. Approximately 150 cross-sectional slices were made at 12-μm intervals, and 100 contiguous slices were selected for analysis. These were contoured inside the endosteal edge of the cortical shell to obtain the total volume (TV) of the space, followed by analysis of the trabecular bone (BV). All scans were analyzed using the Scanco software version 5.0.
Mechanical testing
Bone strength was measured in Mayo's Orthopedic Biomechanical Laboratory Facility. The left femora were dissected, wrapped in saline-soaked gauze to prevent dehydration, and stored at −20°C in small, sealed freezer bags. On the day of testing, bones were slowly thawed to room temperature and kept in saline. Mechanical testing was performed in a three-point bending configuration to determine the flexural properties using a Dynamic Mechanical Analyzer (DMA 2980, New Castle, DE, USA). Loading was applied at a rate of 0.1 N/s until failure. Using the Euler-Bernoulli beam formulation, the slope of the force-deflection curve was used to calculate the bone's bending rigidity, EI = PL3/48δ, where P = applied load, δ = beam deflection at midspan, L = beam span, E = Young's modulus, and I = area moment of inertia. To determine the Young's modulus of the bone (E), the elliptic cross-section of the femur was imaged by pQCT at the midshaft, and the area moment of inertia (I) calculated using the public domain NIH Image program (developed at the U.S. NIH and available on the Internet at http://rsb.info.nih.gov/nih-image/). The maximum failure load was used to calculate the ultimate stress, σmax = PmaxLr/4I, where r = bone radius.
Bone histomorphometry
A double fluorochrome labeling protocol was used to assess potential age- and genotype-related differences in bone growth. The right femora were harvested and fixed in 10% neutral buffered formalin for 24 h, followed by a minimum of 72 h in 70% ethanol, all at 4°C. The samples were dehydrated in graded ethanols and embedded in glycol methacrylate. Sections of 5 μm thickness were cut in either the frontal (distal femora) or transverse (femoral midshaft) plane using a microtome (Reichert-Jung Supercut 2050) equipped with a tungsten-carbide D-profile blade. Goldner's trichrome stain was used to visualize structures related to static histomorphometric endpoints, including osteoblast and osteoclast number, bone volume, and trabecular architecture. Dynamic histomorphometric endpoints of cortical and trabecular bone were measured under UV light on unstained sections. All tissue measurements were made using Osteomeasure (Osteometrics, Atlanta, GA, USA) and an Olympus XX microscope.
RNA isolation and quantitative PCR
RNA from bone was isolated using the RNase mini kit (Qiagen, Valencia, CA, USA). One microgram of RNA from each sample was reverse-transcribed with Taqman RT reagents, random hexamers, and Multiscribe reverse transcriptase (Applied Biosystems, Foster City, CA, USA). Quantitative PCR reactions were conducted using the primer sequences previously validated and the iCycler iQ Detection System.(28) Amplification plots were analyzed with iCycler iQ Detection System analysis software v3.0.6070 (Bio-Rad, Hercules, CA, USA). Gene expression was normalized to ribosomal protein L19 as an internal control.
Serum chemistries
Serum levels of IGF-I were measured using a rat/mouse IGF-I two-site immunoenzymometric assay kit kindly provided by Immunodiagnostic Systems (IDS, Fountain Hills, AZ, USA). Serum levels of N-terminal propeptide of type I procollagen (PINP, bone formation marker) and TRACP (bone resorption marker) were also determined using IDS immunoassays.
Statistical analysis
Comparisons between wildtype (WT) and PAPP-A knockout (KO) mice were made using Student's t-test, with significance set at p < 0.05.
RESULTS
vBMD
pQCT was performed at the femoral midshaft of PAPP-A KO and WT mice as a measure of vBMD in primarily cortical bone (Fig. 1A). At 2 mo of age, PAPP-A KO vBMD was 10% less than WT. This 10–13% difference in vBMD between PAPP-A KO and WT mice at this site was maintained at 4, 6, and 12 mo of age because vBMD increased equivalently in both groups of mice. Similarly, cortical thickness at the midshaft increased with age in both WT and PAPP-A KO mice and was decreased ∼10% in femora from PAPP-A KO mice compared with WT at all ages tested (Table 1). Also noted in Table 1 are the reduced body weights and femur lengths of PAPP-A KO mice, as previously reported.(23)
FIG. 1.
BMD of (A) midshaft femur and (B) distal femur in WT (black bars) and PAPP-A KO (hatched bars) mice. BMD was measured by pQCT. Results are mean ± SE (n = 7–10). *Significant difference between WT and PAPP-A KO, p < 0.05.
Table 1.
Cortical Thickness, Body Weight, and Femur Length
| Cortical thickness (mm) | Body weight (g) | Femur length (mm) | |
| WT | |||
| 2 mo | 0.434 ± 0.0077 | 22.4 ± 0.64 | 9.4 ± 0.11 |
| 4 mo | 0.488 ± 0.0079 | 26.0 ± 0.57 | 9.6 ± 0.07 |
| 6 mo | 0.529 ± 0.0137 | 28.4 ± 0.83 | 9.9 ± 0.13 |
| PAPP-A KO | |||
| 2 mo | 0.397 ± 0.0048* | 14.1 ± 0.92* | 7.9 ± 0.13* |
| 4 mo | 0.430 ± 0.0072* | 15.9 ± 1.52* | 8.1 ± 0.25* |
| 6 mo | 0.458 ± 0.0008* | 16.3 ± 0.88* | 8.2 ± 0.17* |
Results are mean ± SE, n = 7−16 mice per group.
* Significant difference between WT and PAPP-A KO, p < 0.05.
Figure 1B presents the results for vBMD of the distal femur, where trabecular bone is better represented. At 2 mo, there was no significant difference in vBMD at this site between PAPP-A KO and WT mice. However, at 4, 6, and 12 mo, there were highly significant differences between these two groups. Furthermore, the differences were exacerbated with age with 9%, 11%, and 17% decreases in vBMD at this site in 4-, 6-, and 12-mo-old PAPP-A KO mice, respectively, compared with WT mice.
μCT of the tibia provided more information on trabecular bone volume fraction and architecture (Table 2; Fig. 2). Bone volume/total volume (BV/TV) was decreased 33% at 2 mo in PAPP-A KO compared with WT mice (p = 0.0086). Trabecular number was decreased and trabecular spacing was increased in PAPP-A KO mice compared with WT, suggesting an effect on trabecular bone formation during puberty in PAPP-A KO mice. There was no significant effect on trabecular thickness at this time. BV/TV increased in the tibia of WT mice between 2 and 6 mo, but this increase did not occur in PAPP-A KO mice, resulting in a 46% relative decrease in trabecular BV/TV in PAPP-A KO compared with WT mice at 6 mo (p < 0.001). In addition, trabecular thickness increased in the tibia of WT mice, but not PAPP-A KO mice, between 2 and 6 mo; therefore, relative trabecular thickness was diminished by 28% (p < 0.001) in PAPP-A KO mice at 6 mo. There was no significant difference in trabecular number or spacing between the two groups of mice at 6 mo. Thus, there seems to be an impairment in bone turnover in the absence of PAPP-A resulting in an age-related state of reduced trabecular bone mass.
Table 2.
μCT
| WT | PAPP-A KO | p | |
| 2 mo | |||
| BV/TV | 0.093 ± 0.0072 | 0.062 ± 0.0063 | 0.009 |
| Conn.D | 73.4 ± 10.49 | 52.5 ± 10.38 | 0.063 |
| Tb.N | 4.27 ± 0.193 | 3.46 ± 0.240 | 0.018 |
| Tb.Th | 0.042 ± 0.0013 | 0.039 ± 0.0010 | 0.120 |
| Tb.Sp | 0.238 ± 0.0098 | 0.304 ± 0.0198 | 0.006 |
| 6 mo | |||
| BV/TV | 0.112 ± 0.0112 | 0.0601 ± 0.0075 | <0.0001 |
| Conn.D | 43.1 ± 9.28 | 25.2 ± 5.44 | 0.100 |
| Tb.N | 3.63 ± 0.127 | 4.00 ± 0.234 | 0.215 |
| Tb.Th | 0.055 ± 0.0017 | 0.040 ± 0.0010 | <0.0001 |
| Tb.Sp | 0.283 ± 0.0106 | 0.261 ± 0.0186 | 0.363 |
The tibia was assessed by μCT for the following: BV/TV, relative bone volume; Conn.D, connectivity density, normed by TV (1/mm3); Tb.N, trabecular number (1/mm); Tb, Th, trabecular thickness (mm); Tb.Sp, trabecular separation = marrow thickness (mm).
n = 7−9 mice per group, mean ± SE.
FIG. 2.
Bone architecture of tibia from 2- and 6-mo-old WT (right) and PAPP-A KO (left) mice as assessed by μCT.
Mechanical strength testing
Mechanical competence is a fundamental measure of bone phenotype. Young's modulus, a measure of stiffness, and thus, resistance to deformation, was not significantly different in femurs from WT and PAPP-A KO mice at any of the ages tested (Table 3). Ultimate load stress, a measure of the strength of the material and resistance to fracture, was significantly decreased in femurs from PAPP-A KO mice at 4 and 6 mo of age, but not at 2 mo of age, suggesting a weakening of the skeleton after acquisition of peak BMD at puberty.
Table 3.
Mechanical Strength Testing
| WT | PAPP-A KO | p | |
| 2 mo | |||
| E | 2606 ± 131 | 3041 ± 184 | 0.072 |
| Stress | 77.1 ± 2.91 | 72.3 ± 4.82 | 0.397 |
| 4 mo | |||
| E | 4169 ± 237 | 3947 ± 212 | 0.518 |
| Stress | 106.7 ± 3.43 | 88.6 ± 4.38 | 0.005 |
| 6 mo | |||
| E | 5085 ± 257 | 5066 ± 508 | 0.975 |
| Stress | 123.3 ± 4.13 | 110.5 ± 4.26 | 0.047 |
See Materials and Methods section for calculations.
n = 7−10 mice per group, mean ± SE in Mega-Pascals.
E, Young's modulus (measure of stiffness, resistance to deformation); stress: ultimate load stress (measure of strength of the material, resistance to fracture).
Serum bone markers
Serum levels of PINP, a marker of overall bone formation, decreased dramatically between 2 and 4 mo of age in both WT and PAPP-A KO mice in accordance with the normal age-related decline in bone formation rates in mice. However, PINP levels were significantly (20–25%) lower in serum of 4- and 6-mo-old PAPP-A KO compared with WT mice at the same ages (Fig. 3A). Levels of the bone resorption marker, TRACP, did not change with age in WT mice. Interestingly, serum TRACP levels were significantly higher (by 30%) in PAPP-A KO compared with WT mice at 2 mo, and there was a progressive decrease with age so that they were significantly lower (by 40%) in PAPP-A KO mice at 6 mo.
FIG. 3.
Serum markers of (A) bone formation (PINP) and (B) bone resorption (TRACP) in WT (black bars) and PAPP-A KO (hatched bars) mice. Results are mean ± SE (n = 5–7). *Significant difference between WT and PAPP-A KO, p < 0.05.
Histomorphometry
Histomorphometry was performed on cortical bone of 2- and 6-mo-old PAPP-A KO and WT mice (Table 4). There were no significant differences in mineral apposition rate (MAR) or bone formation rate (BFR). Consistent with the observed decreases in BV/TV by μCT, PAPP-A KO mice had significant decreases in trabecular bone volumes at 2 and 6 mo (Table 5). Trabecular number and thickness were modestly but significantly reduced, and spacing was increased in PAPP-A KO mice at 2 mo. At 6 mo, only trabecular thickness was reduced in PAPP-A KO compared with WT mice. These changes were qualitatively similar to those seen with μCT (Table 2). BFR was reduced in PAPP-A KO compared with controls at 2 and 6 mo, but with the considerable intragroup variability differences were not statistically significant. However, there was a significant reduction in osteoblast number and osteoblast number per bone perimeter in trabecular bone of PAPP-A KO mice at 2 and 6 mo. At 6 mo, the reductions were 60% and 55%, respectively. Parameters for bone resorption (osteoclast number and osteoclast number per bone perimeter) were also significantly lower in bones from PAPP-A KO mice at 2 mo (∼60%), but there were no differences between WT and PAPP-A KO mice at 6 mo. Thus, the decrease in bone formation parameters exceeded the decrease in bone resorption parameters at 6 mo, indicating a state of low-turnover osteopenia in adult PAPP-A KO mice.
Table 4.
Histomorphometry: Cortical Bone
| WT | PAPP-A KO | p | |
| 2 mo | |||
| MS/BS | 40.3 ± 3.37 | 35.5 ± 4.18 | 0.388 |
| MAR | 1.39 ± 0.164 | 0.96 ± 0.112 | 0.074 |
| BFR/NS | 236 ± 36 | 159 ± 19 | 0.125 |
| 6 mo | |||
| MS/BS | 18.4 ± 1.61 | 21.7 ± 2.79 | 0.324 |
| MAR | 0.670 ± 0.051 | 0.750 ± 0.014 | 0.188 |
| BFR/BS | 86 ± 8 | 91 ± 6 | 0.448 |
Values are mean ± SE, n = 6–9 per group.
MS/BS, mineralizing surface/bone surface (%); MAR, mineral apposition rate (μm/d); BFR/BS, bone formation rate/bone surface (μm3/μm2/yr).
Table 5.
Histomorphometry: Trabecular Bone
| WT | PAPP-A KO | p | |
| 2 mo | |||
| BV/TV | 0.049 ± 0.0049 | 0.025 ± 0.0046 | 0.008 |
| Tb.N | 1.95 ± 0.151 | 1.20 ± 0151 | 0.007 |
| Tb.Th | 0.025 ± 0.0094 | 0.022 ± 0.0082 | 0.033 |
| Tb.Sp | 0.556 ± 0.0059 | 1.003 ± 0.0068 | 0.0005 |
| MS/BS | 12.09 ± 2.038 | 9.50 ± 2.534 | 0.450 |
| MAR | 1.151 ± 0.1080 | 0.985 ± 0.0914 | 0.156 |
| BFR/BS | 176 ± 38 | 95 ± 18 | 0.156 |
| BFR/BV | 1,425 ± 302 | 902 ± 160 | 0.246 |
| N.OB | 178 ± 11 | 98 ± 7 | 0.0001 |
| N.OC | 12 ± 2 | 5 ± 1 | 0.0008 |
| N.OB/B.Pm | 40.3 ± 2.94 | 27.0 ± 1.30 | 0.003 |
| N.OC/B.Pm | 2.9 ± 0.64 | 1.3 ± 0.33 | 0.023 |
| 6 mo | |||
| BV/TV | 0.047 ± 0.0051 | 0.026 ± 0.0039 | 0.008 |
| Tb.N | 1.50 ± 0.148 | 1.19 ± 0.155 | 0.184 |
| Tb.Th | 0.029 ± 0.011 | 0.022 ± 0.0097 | 0.0003 |
| Tb.Sp | 0.658 ± 0.0070 | 1.007 ± 0.0026 | 0.133 |
| MS/BS | 12.41 ± 1.601 | 12.56 ± 2.563 | 0.959 |
| MAR | 0.604 ± 0.0491 | 0.511 ± 0.0728 | 0.287 |
| BFR/BS | 28.8 ± 5.07 | 21.0 ± 2.47 | 0.209 |
| BFR/BV | 215 ± 47.1 | 183 ± 16.5 | 0.546 |
| N.OB | 152 ± 15 | 61 ± 14 | 0.002 |
| N.OC | 6 ± 2 | 4 ± 1 | 0.249 |
| N.OB/B.Pm | 42.0 ± 2.35 | 18.7 ± 4.0 | 0.003 |
| N.OC/B.Pm | 1.8 ± 0.37 | 1.4 ± 0.29 | 0.378 |
Values are mean ± SE, n = 6–9 per group.
The distal femur was assessed by bone histomorphometry (see Materials and Methods section) for the following: BV/TV, relative bone volume; Tb.N, trabecular number (1/mm); Tb.Th, trabecular thickness (mm); Tb.Sp, trabecular spacing (mm); MS/BS, mineralizing surface/bone surface (%); MAR, mineral apposition rate (μm/d); BFR/BS, bone formation rate/bone surface (μm3/μm2/yr); BFR/BV, bone formation rate/bone volume (%/yr); N.OB, number of osteoblasts; N.OC number of osteoclasts; N.OB/B. Pm, number of osteoblasts/bone perimeter; N.OC/B.Pm, number of osteclasts/bone perimeter.
The IGF system
Serum levels of IGF-I were not significantly different in PAPP-A KO and WT mice (Table 6). Neither were skeletal levels of IGF-I mRNA as assessed by real-time PCR in tibias from PAPP-A KO and WT mice (Table 7). Relative gene expression of IGF-I receptor and IGFBP-4 also did not differ between the two mouse types. Therefore, there did not seem to be any compensatory regulation of the IGF system in response to deletion of PAPP-A. However, there was an 18% decrease in the IGF-responsive gene, IGFBP-5, in PAPP-A KO versus WT bone, which may reflect a decrease in IGF-I bioavailability in the absence of PAPP-A.(28,29)
Table 6.
Circulating Levels of IGF-I
|
Serum IGF-I (ng/ml) |
p | ||
| WT | PAPP-A KO | ||
| 2 mo | 849 ± 48 | 800 ± 64 | 0.55 |
| 4 mo | 887 ± 66 | 745 ± 72 | 0.18 |
| 6 mo | 885 ± 78 | 761 ± 54 | 0.22 |
Results are mean ± SE, n = 5–6 per group.
Table 7.
Expression of IGF System Components in WT and PAPP-A KO Bone
|
Copies mRNA/μg total RNA |
|||
| WT | PAPP-A KO | p | |
| IGF-I | 575 ± 118 | 598 ± 69 | 0.86 |
| IGF-IR | 352 ± 108 | 370 ± 85 | 0.90 |
| IGFBP-4 | 1450 ± 546 | 1830 ± 1290 | 0.81 |
| IGFBP-5 | 10,100 ± 1,380 | 8,250 ± 1,360 | 0.36 |
RNA was harvested from tibia of WT (n = 7) and PAPP-A KO (n = 9) mice.
Results (mean ± SE) are expressed as copies of mRNA per microgram total RNA.
DISCUSSION
The results of this study showed a causal relationship between deletion of the PAPP-A gene in mice and skeletal insufficiency in mass, density, microarchitecture, and strength. Thus, PAPP-A, a novel IGFBP metalloproteinase, is a bone growth regulatory factor in vivo. Furthermore, these data support the role of local IGF-I primarily in trabecular bone remodeling.
The absence of PAPP-A had little impact on cortical bone modeling. The modestly reduced vBMD seemed to be appropriate for the smaller bone size and weight of the PAPP-A KO mouse from 2 through 12 mo of age. This finding was somewhat surprising because studies of Mohan et al.,(30) comparing the relative effects of IGF-I, IGF-II, and growth hormone (GH) deficiency, clearly showed a prominent role of IGF-I in acquisition of peak cortical BMD during puberty in mice. In addition, deletion of the IGF-I gene in mice completely prevented the periosteal expansion that occurs during puberty. It is possible that our mice were already postpubertal at 2 mo of age, but additional measurements at 1 mo also showed a diminished vBMD in midshaft femur of PAPP-A KO compared with WT mice. The most likely explanation for the seeming lack of effect of PAPP-A on cortical bone is that circulating IGF-I, produced in liver primarily under GH control, is more critical for modeling of bone, in particular periosteal growth, than is local IGF-I. This is supported by studies in transgenic mice with liver-specific inactivation of IGF-I gene expression(7,8) and in congenic mice with shared genetic determinants for both serum levels of IGF-I and bone acquisition.(31) Thus, the minimal cortical phenotype of the PAPP-A KO mouse is consistent with the relatively normal circulating concentrations of IGF-I in these mice across various ages.
On the other hand, the PAPP-A KO mouse model supports the view that local IGF-I is needed for optimal trabecular bone mass. Both pQCT and μCT measures indicated significant decreases in relative trabecular volume in the absence of PAPP-A, which were progressive with age. With pQCT, there was no difference in volumetric BMD at the distal femur at 2 mo between WT and PAPP-A KO mice, whereas there was a highly significant 17% difference at 12 mo. By μCT, significant and progressive age-related decreases in trabecular bone volume in PAPP-A KO mice were confirmed. The finding that structural differences in distal femur were not detected at 2 mo by pQCT but were detected by μCT is likely because of the cortical component in the pQCT measurements versus a relatively pure analysis of the trabecular compartment by μCT. Furthermore, different bones (femur and tibia) were used for the two analyses. μCT also indicated architectural changes in the bones of PAPP-A KO mice. In particular, the lack of increased trabecular thickness between 2 and 6 mo in PAPP-A KO compared with the 30% increase in WT mice suggests a major impact of PAPP-A deficiency on trabecular bone formation.
Mechanical strength testing, a hallmark of bone quality, indicated a decrease in ultimate load stress in PAPP-A KO mice at 4 and 6 mo that was not apparent at 2 mo, suggesting an age-related decrease in resistance to fracture. For mice, peak structural properties of the bone are attained at 5 mo; peak material properties are attained at ∼4 mo.(32) Thus, PAPP-A KO mice may represent a model of age-related osteoporosis with increased fracture risk.
Serum markers and direct assessment by bone histomorphometry were used to determine the effect of PAPP-A deficiency on bone formation and bone resorption. Serum markers of dynamic bone parameters indicated significantly lower levels of PINP (bone formation) in PAPP-A KO mice at 4 and 6 mo compared with controls. Interestingly, bone resorption markers were significantly increased in PAPP-A KO mice compared with controls at 2 mo but significantly decreased compared with controls at 6 mo. These are whole skeleton evaluations that suggest, at least in the older mice, a low bone turnover state. Histomorphometry of cortical bone showed no significant differences between PAPP-A KO and WT mice in terms of dynamic measures of MAR and BFR, although cortical thickness was reduced by 10%. For trabecular bone, relative bone volume was dramatically decreased in PAPP-A KO mice compared with controls at both 2 and 6 mo. Osteoblast number was decreased by 45% and 60% in trabecular bone from PAPP-A KO mice compared with WT mice at 2 and 6 mo, respectively. Osteoclast number was significantly lower by 50–60% in PAPP-A KO compared with WT mice but only at 2 mo. Thus, serum markers and histomorphometry support the mechanism of PAPP-A deficiency as impaired bone, particularly trabecular bone, remodeling in the adult mice. The reason for the increase in serum bone resorption markers at 2 mo in PAPP-A KO mice, when osteoclast number is significantly reduced, is unclear, but may reflect robust remodeling in other bone types and sites during this time. Alternatively, TRACP levels can increase if there are less active osteoclasts releasing enzyme outside the ruffled border area.
Recently, Qin et al.(33) showed that transgenic mice with overexpression of PAPP-A driven by the rat type I collagen promoter exhibited increased calvarial bone thickness, femoral and tibial bone area, and periosteal circumference but without changes in vBMD. Bone formation, but not bone resorption, was increased in the mid-diaphysis femur. Although these authors reach the same conclusion (i.e., that PAPP-A is a potent anabolic factor in the regulation of bone formation), there seemed to be more of an effect on cortical than trabecular bone in the PAPP-A transgenic mice, opposite to what we found in the PAPP-A KO mice. There may be several reasons for these apparently discrepant results. First, it is unclear from this report whether PAPP-A was expressed in other tissues besides bone and/or whether expression resulted in increased levels of circulating PAPP-A and consequent endocrine effects. Jiang et al.,(34) using the same promoter to drive IGF-I expression in mice, showed that only when local IGF-I expression was high enough to increase circulating levels were femurs affected. Second, the mid-diaphysis of the femur or tibia was assessed in the PAPP-A transgenic mice, which is primarily cortical bone and could mask an effect on trabecular bone. Third, BFR periosteal and endosteal surfaces were measured in 5-wk-old mice when IGF-I is important in the acquisition of peak BMD.(30,35)
Our in vivo data, as well as previous in vitro data,(17,21,22,33) suggest that PAPP-A regulates bone formation by modulating IGF-I bioavailability in the local bone environment. In agreement, bone-specific IGF-I receptor KO mice exhibited decreased trabecular bone volume, trabecular number, trabecular thickness, and connectivity, along with increased mineralization lag time.(36) Deletion of IRS-1, a major mediator of IGF-I signaling, resulted in low turnover osteopenia,(6) similar to what we see in the PAPP-A KO mice. Zhao et al.(37) showed that overexpression of IGF-I in osteoblasts using the human osteocalcin promoter resulted in increased trabecular BFR without a change in osteoblast number. There was little or no effect on cortical bone mass. There was no change in serum IGF-I levels, and the effects of osteoblast-targeted overexpression of IGF-I seemed to be more pronounced in trabecular than in cortical bone, suggesting a role for local IGF-I in remodeling. Why there seem to be mechanistic differences among these models (e.g., osteoblast number) is unclear but may involve the cell type targeted. Overexpression driven by the osteocalcin promoter would affect the mature, well-differentiated osteoblast, whereas osteoblastic (including precursor cells) and osteoclastic cells would potentially be affected in the PAPP-A KO mouse. Furthermore, IGF-I has multiple effects in bone by stimulating proliferation, differentiation, and matrix production, promoting survival of osteoblasts and pre-osteoblasts, as well as promoting the formation and function of osteoclasts—effects that would be cell type and stage specific.(1,2) At seeming odds with ours and these other studies indicating loss of IGF signaling associated with trabecular osteopenia, mice lacking the IGF-I gene showed decreases in cortical bone mass and increased trabecular bone density and connectivity greater than controls.(5) The decrease in cortical bone mass could be caused by the smaller size of the mouse, and thus appropriate, but also to the decrease in circulating IGF-I. It is probable that, in the absence of circulating IGF-I, there are compensatory increases in GH (loss of negative feedback on the pituitary) that stimulated trabecular bone. PAPP-A KO mice have normal circulating IGF-I levels and no increases in GH.(38)
Our working model is that PAPP-A functions to enhance local IGF-I action, without a change in IGF-I or IGF-I receptor gene expression, through cleavage of inhibitory IGFBPs, primarily IGFBP-4. Mouse bones express high levels of IGFBP-4 and -5 mRNA levels as previously reported.(8) IGFBP-4 consistently inhibits IGF action in a number of cell types including osteoblasts, and in vivo overexpression of IGFBP-4 impairs bone growth.(10,11,13,16–19) Miyakoshi et al.(12) reported that local administration of an equimolar dose of IGFBP-4 over the parietal bone inhibited IGF-I-stimulated alkaline phosphatase activity. The role of IGFBP-5 is more controversial, with both inhibitory and enhancing effects on IGF action in bone reported, as well as IGF-independent effects.(39,40) Interestingly, IGFBP-5 is an IGF-response gene in several tissues, including bone,(41) and a decrease in its expression has been suggested to be a marker of diminished local IGF action.(28,29) IGF-I, IGF-I receptor, and IGFBP-4 mRNA levels were not different in PAPP-A KO compared with WT bone. However, IGFBP-5 mRNA levels were decreased. These findings fit with the notion that PAPP-A regulates IGF-I action without changes in IGF-I, IGF-I receptor, or IGFBP-4 gene expression. Although these findings support the notion that PAPP-A increases bone formation through an IGF-dependent mechanism, we cannot at this time rule out the possibility that PAPP-A has IGF-independent effects on bone.
In conclusion, the findings of this study, along with those in our previous study,(23) indicate that PAPP-A is an important regulator of bone development at both embryonic and postnatal stages.
ACKNOWLEDGMENTS
We thank Laurie Bale, Sean Harrington, Kelly Thompson, Stephanie Thomas, and Lindsay Horton for excellent technical assistance. We also thank Dr Edward Johnstone for the early contributions to this study. This work was supported in part by NIH Grant DK-07352 to SJT.
Footnotes
The authors state that they have no conflicts of interest.
REFERENCES
- 1.Rosen CJ. Insulin-like growth factor I and bone mineral density: Experience from animal models and human observational studies. C. lin Endocrinol Metab. 2004;18:423–435. doi: 10.1016/j.beem.2004.02.007. [DOI] [PubMed] [Google Scholar]
- 2.Conover CA. Insulin-like growth factors and the skeleton. In: Canalis E, editor. Skeletal Growth Factors. Philadelphia, PA, USA: Lippincott Williams & Wilkins; 2000. pp. 101–116. [Google Scholar]
- 3.Clemens TL, Chernausek SD. Genetic strategies for elucidating insulin-like growth factor action in bone. Growth Horm IGF Res. 2004;14:195–199. doi: 10.1016/j.ghir.2003.12.008. [DOI] [PubMed] [Google Scholar]
- 4.Lupu F, Terwilliger JD, Lee K, Segre GV, Efstratiadis A. Roles of growth hormone and insulin-like growth factor 1 in mouse postnatal growth. Dev Biol. 2001;229:141–162. doi: 10.1006/dbio.2000.9975. [DOI] [PubMed] [Google Scholar]
- 5.Bikle D, Majumdar S, Laib A, Powell-Braxton L, Rosen C, Beamer W, Nauman E, Leary C, Halloran B. The skeletal structure of insulin-like growth factor I-deficient mice. J Bone Miner Res. 2001;16:2320–2329. doi: 10.1359/jbmr.2001.16.12.2320. [DOI] [PubMed] [Google Scholar]
- 6.Ogata N, Chikazu D, Kubota N, Terauchi Y, Tobe K, Azuma Y, Ohta T, Kadowaki T, Nakamura K, Kawaguchi H. Insulin receptor substrate-1 in osteoblast is indispensable for maintaining bone turnover. J Clin Invest. 2000;105:935–943. doi: 10.1172/JCI9017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yakar S, Rosen CJ. From mouse to man: Redefining the role of insulin-like growth factor-I in the acquisition of bone mass. Exp Biol Med. 2003;228:245–252. doi: 10.1177/153537020322800302. [DOI] [PubMed] [Google Scholar]
- 8.Sjogren K, Sheng M, Moverare S, Liu J-L, Wallenius K, Tornell J, Isaksson O, Jansson J-O, Mohan S, Ohlsson C. Effects of liver-derived insulin-like growth factor I on bone metabolism in mice. J Bone Miner Res. 2002;17:1977–1987. doi: 10.1359/jbmr.2002.17.11.1977. [DOI] [PubMed] [Google Scholar]
- 9.Holly J, Perks C. The role of insulin-like growth factor binding proteins. Neuroendocrinology. 2006;83:154–160. doi: 10.1159/000095523. [DOI] [PubMed] [Google Scholar]
- 10.Conover CA, Kiefer MC, Zapf J. Posttranslational regulation of insulin-like growth factor binding protein-4 in normal and transformed human fibroblasts: Insulin-like growth factor dependence and biological studies. J Clin Invest. 1993;91:1129–1137. doi: 10.1172/JCI116272. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mohan S, Nakao Y, Honda Y, Landale E, Leser U, Dony C, Lang K, Baylink DJ. Studies on the mechanisms by which insulin-like growth factor (IGF) binding protein-4 (IGFBP-4) and IGFBP-5 modulate IGF actions in bone cells. J Biol Chem. 1995;270:20424–20431. doi: 10.1074/jbc.270.35.20424. [DOI] [PubMed] [Google Scholar]
- 12.Miyakoshi N, Richman C, Qin X, Baylink DJ, Mohan S. Effects of recombinant insulin-like growth factor-binding protein-4 on bone formation parameters in mice. Endocrinology. 1999;140:5719–5728. doi: 10.1210/endo.140.12.7175. [DOI] [PubMed] [Google Scholar]
- 13.Zhang M, Faugere M-C, Malluche H, Rosen CJ, Chernausek SD, Clemens TL. Paracrine overexpression of IGFBP-4 in osteoblasts of transgenic mice decreases bone turnover and causes global growth retardation. J Bone Miner Res. 2003;18:836–843. doi: 10.1359/jbmr.2003.18.5.836. [DOI] [PubMed] [Google Scholar]
- 14.Conover CA. Insulin-like growth factor binding protein proteolysis in bone cell models. Prog Growth Factor Res. 1995;6:301–309. doi: 10.1016/0955-2235(95)00032-1. [DOI] [PubMed] [Google Scholar]
- 15.Bunn RC, Fowlkes JL. Insulin-like growth factor binding protein proteolysis. Trends Endocrinol Metab. 2003;14:176–181. doi: 10.1016/s1043-2760(03)00049-3. [DOI] [PubMed] [Google Scholar]
- 16.Durham SK, Kiefer MC, Riggs BL, Conover CA. Regulation of insulin-like growth factor binding protein 4 by a specific insulin-like growth factor binding protein 4 proteinase in normal human osteoblast-like cells: Implications in bone cell physiology. J Bone Miner Res. 1994;9:111–117. doi: 10.1002/jbmr.5650090115. [DOI] [PubMed] [Google Scholar]
- 17.Oritz CO, Chen B-K, Bale LK, Overgaard MT, Oxvig C, Conover CA. Transforming growth factor-β regulation of the insulin-like growth factor binding protein-4 protease system in cultured human osteoblasts. J Bone Miner Res. 2003;18:1066–1072. doi: 10.1359/jbmr.2003.18.6.1066. [DOI] [PubMed] [Google Scholar]
- 18.Qin X, Byun D, Strong DD, Baylink DJ, Mohan S. Studies on the role of human insulin-like growth factor-II (IGF-II)-dependent IGF binding protein (hIGFBP)-4 protease in human osteoblasts using protease-resistant IGFBP-4 analogs. J Bone Miner Res. 1999;14:2079–2088. doi: 10.1359/jbmr.1999.14.12.2079. [DOI] [PubMed] [Google Scholar]
- 19.Miyakoshi N, Qin X, Kasukawa Y, Richman C, Srivastava AK, Baylink DJ, Mohan S. Systemic administration of insulin-like growth factor (IGF)-binding protein-4 (IGFBP-4) increases bone formation parameters in mice by increasing IGF bioavailability via an IGFBP-4 protease-dependent mechanism. Endocrinology. 2001;142:2641–2648. doi: 10.1210/endo.142.6.8192. [DOI] [PubMed] [Google Scholar]
- 20.Lawrence JB, Oxvig C, Overgaard MT, Sottrup-Jensen L, Gleich GJ, Hays LG, Yates JR, III, Conover CA. The insulin-like growth factor (IGF)-dependent IGF binding protein-4 protease secreted by human fibroblasts is pregnancy-associated plasma protein-A. Proc Natl Acad Sci USA. 1999;96:3149–3153. doi: 10.1073/pnas.96.6.3149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Byun D, Mohan S, Yoo M, Sexton C, Baylink DJ, Qin X. Pregnancy-associated plasma protein-A accounts for the insulin-like growth factor (IGF)-binding protein-4 (IGFBP-4) proteolytic activity in human pregnancy serum and enhances the mitogenic activity of IGF by degrading IGFBP-4 in vitro. J Clin Endocrinol Metab. 2001;86:847–854. doi: 10.1210/jcem.86.2.7223. [DOI] [PubMed] [Google Scholar]
- 22.Jia D, Heersche JNM. Pregnancy-associated plasma protein-A proteolytic activity in rat vertebral cell cultures: Stimulation by dexamethasone—a potential mechanism for glucocorticoid regulation of osteoprogenitor proliferation and differentiation. J Cell Physiol. 2005;204:848–858. doi: 10.1002/jcp.20344. [DOI] [PubMed] [Google Scholar]
- 23.Conover CA, Bale LK, Overgaard MT, Johnstone EW, Laursen UH, Fuchtbauer E-M, Oxvig C, van Deursen J. Metalloproteinase pregnancy-associated plasma protein A is a critical growth regulatory factor during fetal development. Development. 2003;131:1187–1194. doi: 10.1242/dev.00997. [DOI] [PubMed] [Google Scholar]
- 24.Bale LK, Conover CA. Disruption of insulin-like growth factor-II imprinting during embryonic development rescues the dwarf phenotype of mice null for pregnancy-associated plasma protein-A. J Endocrinol. 2005;186:325–331. doi: 10.1677/joe.1.06259. [DOI] [PubMed] [Google Scholar]
- 25.Modder UIL, Sanyal A, Kearns AE, Sibonga JD, Nishihara E, Xu J, O'Malley BW, Ritman EL, Riggs BL, Spelsberg TC, Khosla S. Effects of loss of steroid receptor coactivator-1 on the skeletal response to estrogen in mice. Endocrinology. 2004;145:913–921. doi: 10.1210/en.2003-1089. [DOI] [PubMed] [Google Scholar]
- 26.Bouxsein ML, Myers KS, Shultz KL, Donahue LR, Rosen CJ, Beamer WG. Ovariectomy-induced bone loss varies among inbred strains of mice. J Bone Miner Res. 2005;20:1085–1092. doi: 10.1359/JBMR.050307. [DOI] [PubMed] [Google Scholar]
- 27.Ackert-Bicknell CL, Rosen CJ. A chromosomal inversion within a quantitative trait locus has a major effect on adipogenesis and osteoblastogenesis. Ann NY Acad Sci. 2007;1116:291–305. doi: 10.1196/annals.1402.010. [DOI] [PubMed] [Google Scholar]
- 28.Harrington SC, Simari RD, Conover CA. Genetic deletion of pregnancy-associated plasma protein-A is associated with resistance to atherosclerotic lesion development in apolipoprotein E-deficient mice challenged with a high-fat diet. Circ Res. 2007;100:1696–1702. doi: 10.1161/CIRCRESAHA.106.146183. [DOI] [PubMed] [Google Scholar]
- 29.Adamo ML, Ma X, Ackert-Bicknell CL, Donahue LR, Beamer WG, Rosen CJ. Genetic increase in serum insulin-like growth factor-I (IGF-I) in C3H/HeJ compared with C57BL/6J mice is associated with increased transcription from the IGF-I exon 2 promoter. Endocrinology. 2006;147:2944–2955. doi: 10.1210/en.2005-0742. [DOI] [PubMed] [Google Scholar]
- 30.Mohan S, Richman C, Guo R, Amaar Y, Donahue LR, Wergedal J, Baylink DJ. Insulin-like growth factor regulates peak bone mineral density in mice by both growth hormone-dependent and –independent mechanisms. Endocrinology. 2003;144:929–936. doi: 10.1210/en.2002-220948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Delahunty KM, Shultz KL, Gronowicz GA, Koczon-Jaremko B, Adamo ML, Horton LG, Lorenzo J, Donahue LR, Ackert-Bicknell C, Kream BE, Beamer WG, Rosen CJ. Congenic mice provide in vivo evidence for a genetic locus that modulates serum insulin-like growth factor-I and bone acquisition. Endocrinology. 2006;147:3915–3923. doi: 10.1210/en.2006-0277. [DOI] [PubMed] [Google Scholar]
- 32.Brodt MD, Ellis CB, Silva MJ. Growing C57Bl/6 mice increase whole bone mechanical properties by increasing geometric and material properties. J Bone Miner Res. 1999;14:2159–2166. doi: 10.1359/jbmr.1999.14.12.2159. [DOI] [PubMed] [Google Scholar]
- 33.Qin X, Wergedal JE, Rehage M, Tran K, Newton J, Lam P, Baylink DJ, Mohan S. Pregnancy-associated plasma protein-A increases osteoblast proliferation in vitro and bone formation in vivo . Endocrinology. 2006;147:5653–5661. doi: 10.1210/en.2006-1055. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Jiang J, Lichtler AC, Gronowicz GA, Adams DJ, Clark SH, Rosen CJ, Kream BE. Transgenic mice with osteoblast-targeted insulin-like growth factor-I show increased bone remodeling. Bone. 2006;39:494–504. doi: 10.1016/j.bone.2006.02.068. [DOI] [PubMed] [Google Scholar]
- 35.Mohan S, Baylink DJ. Impaired skeletal growth in mice with haploinsufficiency of IGF-I: Genetic evidence that differences in IGF-I expression could contribute to peak bone mineral density differences. J Endocrinol. 2005;185:415–420. doi: 10.1677/joe.1.06141. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Zhang M, Xuan S, Bouxsein ML, von Stechow D, Akeno N, Faugere MC, Malluche H, Zhao G, Rosen CJ, Efstratoados A, Clemens TL. Osteoblast-specific knockout of the insulin-like growth factor (IGF) receptor gene reveals an essential role of IGF signaling in bone matrix mineralization. J Biol Chem. 2002;277:44005–44012. doi: 10.1074/jbc.M208265200. [DOI] [PubMed] [Google Scholar]
- 37.Zhao G, Monier-Faugere M-C, Langub MC, Geng Z, Nakayama T, Pike JW, Chernausek SD, Rosen CJ, Donahue L-R, Malluche HH, Fagin JA, Clemens TL. Targeted overexpression of insulin-like growth factor I to osteoblasts of transgenic mice: Increased trabecular bone volume without increased osteoblast proliferation. Endocrinology. 2000;141:2674–2682. doi: 10.1210/endo.141.7.7585. [DOI] [PubMed] [Google Scholar]
- 38.Conover CA, Bale LK. Loss of PAPP-A extends lifespan in mice. Aging Cell. 2007;6:727–729. doi: 10.1111/j.1474-9726.2007.00328.x. [DOI] [PubMed] [Google Scholar]
- 39.Mohan S, Baylink DJ. Beyond carrier proteins: IGF-binding proteins are multifunctional and act via IGF-dependent and –independent mechanisms. J Endocrinol. 2002;175:19–31. doi: 10.1677/joe.0.1750019. [DOI] [PubMed] [Google Scholar]
- 40.Mukherjee A, Rotwein P. Insulin-like growth factor binding protein-5 in osteogenesis: Facilitator or inhibitor. Growth Horm IGF Res. 2007;17:179–185. doi: 10.1016/j.ghir.2007.01.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Rutter MM, Markoff E, Clayton L, Akeno N, Zhao G, Clemens TL, Chernausek SD. Osteoblast-specific expression of insulin-like growth factor-1 in bone of transgenic mice induces insulin-like growth factor binding protein-5. Bone. 2005;36:224–231. doi: 10.1016/j.bone.2004.10.005. [DOI] [PubMed] [Google Scholar]