Spatiotemporal Localization of VEGF-A Isoforms in the Mouse Postnatal Growth Plate
Article first published online: 17 DEC 2007
Copyright © 2007 Wiley-Liss, Inc.
The Anatomical Record
Volume 291, Issue 1, pages 6–13, January 2008
How to Cite
Evans, K. D. and Oberbauer, A. M. (2008), Spatiotemporal Localization of VEGF-A Isoforms in the Mouse Postnatal Growth Plate. Anat Rec, 291: 6–13. doi: 10.1002/ar.20616
- Issue published online: 18 DEC 2007
- Article first published online: 17 DEC 2007
- Manuscript Accepted: 5 OCT 2007
- Manuscript Received: 19 JUL 2007
- growth plate
Vascular endothelial growth factor (VEGF) is implicated as a key angiogenic factor in the development of endochondral long bone. Several studies have evaluated the role of VEGF in prenatal endochondral bone development, but few have evaluated VEGF postnatally. Growth plates from mice at postnatal ages 14 (P14), 35 (P35), 49 (P49), and 77 (P77) days were examined for differential expression of the primary VEGF-A mRNA isoforms: VEGF 120, VEGF 164, and VEGF 188. VEGF 120 isoform expression was stable across all ages, whereas VEGF 164 had significantly less expression at P35 and P49 and VEGF l88 expression increased with increasing age. The proportion of transcript isoforms expressed at a given age also changed with VEGF 120 being expressed more highly at P35 and P49 than the other two isoforms. Changes in VEGF mRNA isoforms across cell types within the growth plate were assessed by Percoll fractionation of growth plate cells at age P28. Cells of the proliferative and early hypertrophic regions had significantly higher total VEGF mRNA expression relative to the resting and late hypertrophic regions. VEGF protein expression assessed by immunohistochemistry showed variable expression patterns with increasing postnatal age. In contrast, FLK-1 (VEGF Receptor-2) expression was restricted to the hypertrophic region. These results indicate that VEGF continues to play a significant role in endochondral bone development throughout the entire growth phase of postnatal bone development. Anat Rec, 291:6–13, 2007. © 2007 Wiley-Liss, Inc.
Vascular endothelial growth factor (VEGF) is considered a major factor initiating angiogenesis in the epiphyseal growth plate driving the cartilage to bone transition (Gerber et al.,1999; Horner et al.,1999; Carlevaro et al.,2000). Postnatal longitudinal bone growth occurs by stimulation of the resting region of chondrocytes to proliferate and differentiate into hypertrophic chondrocytes. Hypertrophic chondrocytes secrete angiogenic signals, including basic fibroblast growth factor, transferrin, and VEGF, to attract metaphyseal vasculature to the chondro-osseous junction (Baron et al.,1994; Carlevaro et al.,1997). The invading vasculature in turn promotes further hypertrophic differentiation to terminal hypertrophic cells and eventual apoptosis thereby creating space for invading osteoclasts and osteoblasts to remodel and build bone on the remaining mineralized hypertrophic extracellular matrix (Hunziker,1994; Karsenty,1999; Olsen et al.,2000; van der Eerden et al.,2003).
While VEGF belongs to a family of genes that include VEGF-A, VEGF-B, VEGF-C, VEGF-D, and placenta growth factor (Ferrara,2004), the predominant form of VEGF within developing bone is VEGF-A expressed as various protein isoforms (Gerber et al.,1999). A single mRNA generates the VEGF-A mRNA isoforms through alternative splicing (Eckhart et al.,1999). In the mouse, three protein isoforms are translated from the individual mRNA isoforms: VEGF 120, VEGF 164, and VEGF 188 (Ferrara and Keyt,1997; Robinson and Stringer,2001). These isoforms exhibit differential binding to bone extracellular matrix. VEGF 120 is diffusible and will not bind the heparin sulfate proteoglycans found in the extracellular matrix of epiphyseal cartilage. In contrast, VEGF 164 and VEGF 188 bind with high affinity to heparin sulfate proteoglycan matrix and the chondrocyte cell surface, although VEGF 164 possesses some diffusibility. All protein isoforms bind to FLT-1 (VEGF receptor-1) and FLK-1 (VEGF receptor -2), whereas VEGF 164 is the only isoform known to bind neuropilin -1 (NRP-1), a third VEGF receptor (Soker et al.,1998; Neufeld et al.,2002).
Disruption of VEGF results in embryonic lethality between embryonic day (E) 11 and E12 of mouse development demonstrating its essentiality in normal development (Ferrara et al.,1996). In addition, several studies have explored the role of VEGF on growth plate function of both mammalian and nonmammalian species (Gerber et al.,1999; Carlevaro et al.,2000; Maes et al.,2002; Zelzer et al.,2004; Haeusler et al.,2005; Bluteau et al.,2007). Mice engineered to selectively remove VEGF-A from embryonic chondrocytes exhibit severe bone abnormalities resulting from reduced chondrocyte survival and extensive cell death (Zelzer et al.,2004). Mice expressing only the VEGF 120 protein isoform have disrupted neonatal vasculature at the metaphysis and decreased hypertrophic chondrocyte differentiation, an increased hypertrophic zone due to reduced hypertrophic cell apoptosis, and decreased osteoclast invasion (Maes et al.,2002). In prenatal mouse models only expressing VEGF 164 or VEGF 188, inactivation of both VEGF 164 and VEGF 188 inhibited the initial embryonic stages of endochondral ossification (Maes et al.,2002). These key studies have characterized the importance of VEGF on prenatal and immediate postnatal long bone development, but little is known about the expression of VEGF and its transcript isoforms throughout postnatal growth.
The majority of VEGF effects appear to culminate at the hypertrophic region of the growth plate. Yet, given the solubility of some of the VEGF protein isoforms (VEGF 120 and VEGF 164), other regions of the growth plate may produce both the mRNA and protein for VEGF. Furthermore, VEGF mRNA and protein are differentially expressed throughout the embryonic growth plate, suggesting differential expression may continue postnatally and that this expression may play a role in cessation of growth plate activity and growth plate closure. The present study evaluated VEGF mRNA transcript isoforms throughout the entire postnatal growth phase of mice along with expression within the different regions of the growth plate to evaluate postnatal spatial mRNA expression. Immunohistochemical staining for VEGF protein and VEGF receptor-2 (FLK-1) was done to correlate VEGF mRNA expression to VEGF protein and receptor expression throughout the postnatal growth phase of mice.
MATERIALS AND METHODS
Ethical Approval and Animals
All animal procedures were done in accordance with a protocol approved by the University of California, Davis, Institutional Animal Care and Use Committee. Stock inbred C57BL/6J breeding animals were obtained from Jackson Laboratories (Bar Harbor, ME) and housed according to current NIH guidelines. Animals were bred and resulting pups weaned at P21 and housed according to sex. All mice were kept at constant temperature (70 ± 2°F) with a 14:10 hr light:dark cycle and fed Purina mouse diet formula 5008 (Purina, St. Louis, MO) and tap water ad lib.
Collection of Growth Plates, RNA Extraction, and cDNA Generation
Three male mice at weekly intervals (postnatal ages P10–P84) were killed by means of CO2 narcosis and costochondral growth plates were removed by fine dissection. Approximately 20 mg of growth plate tissue from each animal was snap frozen in liquid nitrogen and stored at −70°C until RNA extraction. For RNA extraction, whole growth plate tissue from each animal was placed in 500 μl of stabilization solution (nucleic acid purification lysis buffer, Applied Biosystems, Foster City, CA) and stored at −20°C. Proteinase K and two grinding beads (4 mm diameter, stainless steel beads, SpexCertiprep, Metuchen, NJ) were added and the tissues homogenized in a GenoGrinder2000 (SpexCertiprep) for 2 min at 1,000 strokes per min. Protein digest was done at 56°C for 30 min followed by a 30-min period at −20°C to reduce foam and precipitate RNA. Total RNA was extracted from the tissue lysates using a 6700 automated nucleic acid (ANA) workstation (Applied Biosystems) according to the manufacturer's instructions. Complementary DNA (cDNA) was synthesized from DNase digested total RNA using 100 units of SuperScript III (Life Technologies, Invitrogen Corp, Carlsbad, CA), 600 ng of random hexadeoxyribonucleotide (pd(N)6) primers (random hexamer primer), 10 U of RNaseOut (RNase inhibitor), and 1 mM dNTPs (all Invitrogen) in a final volume of 40 μl. The reverse transcription reaction proceeded for 120 min at 50°C. After addition of 60 μl of water, the reaction was terminated by heating for 5 min to 95°C and cooling on ice. The entire age experiment was repeated, and similar transcript patterns were obtained in both experimental runs. The data from both experiments were analyzed together to increase the number of animals in each age group; thus n = 6 for each age assessed.
Fractionation of Chondrocytes
Growth plate chondrocytes from postnatal age P28 mice were fractionated by discontinuous Percoll gradient centrifugation using a protocol modified from that previously described (Oberbauer and Peng,1995). Age P28 was chosen as this age represents the end of the log growth phase for linear growth before the growth plateau in mice (Oberbauer et al.,1992) and to maximize the yield of resting and hypertrophic cells. Pooled costochondral growth plates from six to nine male C57BL/6J mice at age P28 were dissected free of adherent tissue, cut to 2–3 mm pieces and predigested by incubation in Dulbecco's modified Eagle's medium (DMEM)/F12 (Cambrex Bioscience, Walkersville, MD) supplemented with 3 mg/ml collagenase Type II (Worthington Biochemicals, Lakewood, NJ), 10% fetal bovine serum (FBS) (Cambrex Bioscience), 10,000 u/ml penicillin, 10,000 μg/ml streptomycin (Cambrex Bioscience), 0.05 mg/ml ascorbic acid (SIGMA, St. Louis, MO), and 0.2% NaHCO3 (Cambrex Bioscience) at 37°C for one hour with gentle agitation at 100 rpm. At the conclusion of this 1-hr collagenase incubation, excess soft tissue was freed through gentle pipetting. Media with released tissue debris and freed cells were discarded. The growth plates were then digested in supplemented DMEM/F12 containing 1.5 mg/ml collagenase type II and 5% FBS at 37°C for 10 hr. After the 10 hr enzymatic digestion of the extracellular matrix, the freed cells were separated from tissue debris by filtration through a 70-μm nylon cell strainer (Becton Dickinson, Franklin Lakes, NJ) and collected from the filtrate by centrifugation at 150 × g for 5 min. Cells were then washed in 1× Dulbecco's PBS (Ca2+ and Mg2+ free; Cambrex Biosciences) twice and resuspended in 1 ml of DMEM/F12 medium. Growth plate cells were layered on a discontinuous isotonic Percoll (SIGMA) gradient prepared by weight (densities of 1.0154, 1.0398, 1.0491, 1.0657 g/ml) and centrifuged at 300 × g for 20 min in a swinging bucket rotor. Four cell fractions were formed after centrifugation: fraction 4 enriched with 90% hypertrophic cells and 10% proliferative cells; fraction 3 consisted of 75% hypertrophic cells and 25% proliferative cells; fraction 2 consisted of 75% proliferative cells and 25% hypertrophic cells; and fraction 1 consisted of 5% proliferative cells and 95% resting cells. Each of the four cell fractions was collected, washed with 10 ml of ice-cold 1× Dulbecco's PBS (Ca2+ and Mg2+ free) twice and centrifuged at 150 × g for 5 min. Cells were resuspended in 500 μl of stabilization solution (nucleic acid purification lysis buffer, Applied Biosystems, Foster City, CA) and stored at −20°C until procedure for RNA collection. RNA and cDNA preparation of Percoll gradient derived chondrocyte cells was identical to that described for growth plate tissue. The fractionation protocol experiment was repeated four times, yielding four data points for each growth plate region sample.
Analysis of VEGF-A Transcript Isoforms
Real-time PCR was used to quantitate endochondral chondrocyte VEGF transcript expression. Briefly, for each target gene, two primers and an internal, fluorescent-labeled TaqMan probe (5′ end, reporter dye FAM [6-carboxyflourescein], 3′ end, quencher dye TAMRA [6-carboxytetramethylrhodamine]) were designed, using Primer Express software (Applied Biosystems). The TaqMan systems were designed to be splice variant specific by placing the reverse primer over the alternatively spliced exon (Table 1). The splice variant specific TaqMan assays were also specific for cDNA and did not recognize genomic DNA. TaqMan PCR systems were validated using twofold dilutions of cDNA testing positive for the target genes (Leutenegger et al.,1999). The dilutions were analyzed in triplicate and a standard curve plotted against the dilutions. The slope of the standard curve was used to calculate amplification efficiencies using the formula E = 101/-s − 1.
|Target||Primera||Sequence (5′->3′)||Accession no.|
Each PCR reaction contained 20× primer and probes for each respective isoform-specific TaqMan system with a final concentration of 400 nM for each primer and 80 nM for the TaqMan probe and commercially available PCR Mastermix (TaqMan Universal PCR Mastermix, Applied Biosystems) containing 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 5 mM MgCl2, 2.5 mM deoxynucleotide triphosphates, 0.625 U of AmpliTaq Gold DNA polymerase per reaction, 0.25 U of AmpErase UNG per reaction, and 5 μl of the diluted cDNA sample in a final volume of 12 μl. The samples were placed in 96-well plates and amplified in an automated fluorometer (ABI PRISM 7900 HTA FAST, ABI) using standard amplification conditions: 2 min at 50°C, 10 min at 95°C, 40 cycles of 15sec at 95°C. and 60 sec at 60°C. Fluorescent signals were collected during the annealing temperature and CT values extracted with a threshold of 0.04 and baseline values of 3–15. For stronger signals, the baseline was adjusted manually to 3–10.
To determine the most stably transcribed housekeeping gene, a housekeeping gene validation experiment was run on a representative number of either whole tissue or fractionated chondrocyte cell samples. Five commonly used housekeeping genes were used: a TaqMan PCR system recognizing 18S rRNA (ssrRNA), GAPDH (glyceraldehyde 3-phosphate dehydrogenase), HPRT1 (hypoxanthine phosphoribosyltransferase 1), B2M (beta 2 microglobulin), and TFR2 (transferrin receptor 2; CD71). The most stably transcribed housekeeping gene, GAPDH, was used to normalize the target gene CT values.
Final quantitation was done using the comparative CT method (User Bulletin #2, Applied Biosystems) and is reported as relative transcription or the n-fold difference relative to a calibrator cDNA (i.e., lowest target gene transcription). In brief, the signal of the endogenous control GAPDH was used to normalize the target gene signals of each sample. The difference in the CT for the target and the CT for the internal control, termed ΔCT, was calibrated against either P10 animals in the case of whole growth plate tissue or VEGF 120 in the case of Percoll fractionated chondrocyte cells. The relative linear amount of target molecules relative to the calibrator, was calculated by 2−ΔΔCt. Therefore, all target gene transcription is expressed as an n-fold difference relative to expression in the calibrator group.
Endochondral tissue and Percoll gradient derived chondrocyte VEGF mRNA transcript expression was analyzed by least-squares analysis of variance procedures for unequal subclass numbers, using PROC GLM of SAS (version 8.0 2001) followed by Bonferroni's post hoc test. VEGF 120 mRNA was the most stable in its expression within each age point analyzed. Given the stability in expression of VEGF 120 mRNA, it was used as the internal comparator for which all other transcripts were compared to evaluate changes in n-fold transcription of the VEGF transcripts within each age point as well as within Percoll gradient growth plate chondrocyte regions. To compare changes in n-fold transcription of total VEGF transcription relative to age, the 10-day time point, the earliest age assessed, was used as the internal comparator for all subsequent ages.
Immunohistochemical Staining of Tissue for VEGF and FLK-1
The left femurs from three animals at P14, P35, P49, and P77 were dissected, fixed in 10% formalin (at least 7 days) and decalcified in 0.5 M ethylenediaminetetraacetic acid for at least 3 weeks. After decalcification, each bone was rinsed for 72 hr under running water and stored in 70% ethanol for a minimum of 24 hr. Tissue dehydration and processing were done with a VIP tissue processor (Tissue-Tek VIP 1000/2000, Miles Scientific, Naperville, IL). The bones were then embedded in paraffin and cut into 6 μm craniocaudal longitudinal sections (LKB Bromma rotary microtome). Sections were dried in a 37°C oven for at least 2 days before tissue staining.
Each slide contained three consecutive slices of bone tissue from the mid-diaphysis of the femur bone. Sections were deparaffinized and rehydrated in PBS then pretreated for antigen retrieval in a 10 mM citrate buffer for 10 min in a water bath heated to 55–60°C. FLK-1 VEGF receptor was detected within the endochondral cartilage using a rabbit anti-mouse FLK-1 primary antibody in a 1:50 dilution per sample (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Serial sections were labeled with the FLK-1 antibody coupled to a novaRED peroxidase color substrate (Vector Laboratories, Burlingame, CA, USA). Detection of VEGF-A was done using a rabbit anti-mouse VEGF primary antibody specific to all protein isoforms of VEGF-A in a 1:100 dilution per sample (Santa Cruz Biotechnology, Inc.). Separate serial sections were stained with the VEGF antibody coupled to a DAB-Ni peroxidase color substrate (Vector Laboratories). The primary antibody was detected using the immunoperoxidase procedure described in the Vectastain ABC kit using goat anti-rabbit secondary antibody (Vector Laboratories). Negative control slides were prepared by excluding the primary antibody but retaining all the other steps. Positive control slides were prepared using sections of liver. After the peroxidase color substrate staining, sections were counterstained with 0.1% methyl green and coverslip mounted with Permount. Slides were viewed under a compound light microscope (Nikon SK, Tokyo Japan) at a magnification of 400×, and photographed using a Nikon Coolpix 3300 camera linked to a computer with Nikon View version 6.2.1 for image conversion (Nikon, Garden City, NY).
VEGF Transcript Isoform Expression With Increasing Postnatal Age
Within the endochondral growth plate of mice, VEGF isoform transcription changed with increasing postnatal age (Fig. 1). For each isoform (VEGF 120, VEGF 164, and VEGF 188), the most significant changes in transcript expression were noted between P49 and P77. Relative transcription of VEGF 120 tended toward increased transcription from P14 to P77 but was not significant (P > 0.3). Conversely, relative transcription of VEGF 164 decreased nearly 50% between ages P14 and P49 before rebounding to P14 levels at P77 (P < 0.02). Increased transcription was seen for VEGF 188 with a twofold increase from P14 to P77 (P < 0.0016). Levels of each isoform at P84 was similar to that for P77 (data not shown).
VEGF Transcript Isoform Expression at Each Postnatal Age
To investigate the relative VEGF isoform expression at a single given age, the data were evaluated using VEGF 120 as the internal comparator (Fig. 2). At P14, transcription of all isoforms was not significantly different from one another, although there was a nonsignificant tendency toward increased transcription of VEGF 188 compared with VEGF 120 and VEGF 164. At P35 and P49 isoform transcription of 164 and 188 were significantly decreased by 75% when compared with the 120 isoform (P < 0.0004). By P77, isoform 164 and 188 transcription had been restored to the same expression pattern seen for P14.
Total and VEGF Isoform Transcription Within Growth Plate Regions
Growth plate chondrocytes from P28 mice were fractionated to characterize VEGF expression within growth plate regions. Total VEGF mRNA transcription was fourfold higher (P < 0.03) in the proliferative (5.88 ± 0.88) and early hypertrophic cells (4.86 ± 0.88) of the growth plate compared with the resting region (1.02 ± 0.88); total VEGF expression in late hypertrophic cells (2.69 ± 0.88) was equivalent to that for the resting region. The relative transcription of each isoform did not differ within the resting zone (P > 0.2, data not shown). Within the proliferative cell region, transcription of both isoform 120 and isoform 164 were greater (P < 0.0001) than isoform 188 (Fig. 3): isoform 164 was 2.4-fold higher than isoform 188, while isoform 120 was 2.9 times higher than isoform 164. This pattern was also seen in the early hypertrophic region (P < 0.008). In the terminal or late hypertrophic region, isoform 164 and isoform 188 were significantly lower in transcription relative to the proliferative and early hypertrophic regions while isoform 120 maintained its level of transcription (P < 0.0008).
VEGF Protein and FLK-1 VEGF Receptor in the Growth Plate With Increasing Postnatal Age
Expression of the VEGF protein was similar to the patterns detected for VEGF mRNA isoform expression over increasing postnatal ages (Fig. 4). At P14 strong expression of VEGF protein was detected intracellularly within the hypertrophic region, but also weakly within the proliferative and resting regions. In addition, low expression was noted in the extracellular matrix, indicating that the soluble form of VEGF (VEGF 120 or VEGF 164) was translated at this age. By age P35, strong expression for VEGF protein continued within the hypertrophic region with lower expression in the proliferative region compared with the preceding P14 age. By P42, expression was restricted to the hypertrophic region and the extracellular matrix no longer exhibited detectable VEGF protein; this expression pattern was maintained until the final age point evaluated, P77. Immunohistochemical staining for FLK-1 at increasing postnatal ages showed that this receptor was expressed strongly within the hypertrophic and late proliferative region in P14 and P35 animals (Fig. 4). However, by age P42, FLK-1 expression was limited to the terminal hypertrophic region where it remained at P77.
The data presented in this study illustrate that VEGF mRNA isoform transcription changes throughout the postnatal growth phase. In addition, the pattern of VEGF protein suggests differential expression of VEGF protein isoforms within the endochondral growth plate throughout the postnatal growth phase. The decreased expression noted for VEGF 164 between ages P35 and P49, in addition to the changes noted near these ages in both the immunohistochemistry and growth plate fractionation (at P28), indicate that this transcript may play a role in growth cessation and growth plate closure. Caution in extending the conclusions must be exercised, however, given the different growth plates analyzed in the present study. Although primate costochondral growth plate possesses similar growth characteristics to that of long bone growth plates (Ellis and Carlson,1986) and mouse fetal rib and long bone growth plates exhibit correlated developmental and gene expression profiles (Schipani et al.,2001), the growth stages of the two growth plates may not be directly comparable. Nevertheless, the changes in VEGF protein expression correspond to reduced proliferative capacity of the femoral growth plate and the alterations in mRNA expression in the costochondral growth plate were consistent with that observed for the protein supportive of a role of VEGF 164 in growth cessation.
Angiogenesis is a key element in osteogenesis. The invading metaphyseal vasculature brings bone marrow stromal cells to the chondro-osseous junction that serve as progenitors for mature bone forming cells (Bianco and Gehron Robey,2000). It is generally accepted that VEGF protein expression within the hypertrophic region serves as a signaling mechanism for invading vasculature (Ferrara,2004; Fielder et al.,2005). VEGF also has been shown to promote migration and differentiation of osteoblasts as well as formation, survival, and osteoclast resorption in vitro (Deckers et al.,2000; Nakagawa et al.,2000). Previous investigators have speculated that the matrix bound forms of VEGF protein (VEGF 164 and VEGF 188) promote metaphyseal vascularization (Maes et al.,2004). Using knockout mouse models, embryonic mice expressing VEGF 164 protein either alone or with VEGF 188 protein, had normal metaphyseal vascularization and ossification, whereas mice expressing only VEGF 188 protein showed impaired chondrocyte proliferation and survival in response to hypoxia (Maes et al.,2004). Similar studies with mice expressing only VEGF 120 protein revealed disrupted embryonic bone development, further corroborating the role of matrix bound VEGF protein (i.e., that translated from the VEGF 164 or VEGF 188 mRNA isoform) in metaphyseal vascular development (Maes et al.,2004). Furthermore, these embryologic studies show that VEGF 164 protein plays a major role in the vasculature development required for prenatal osteogenesis. The present study showed that VEGF 164 mRNA was differentially expressed throughout postnatal bone development, suggesting that the VEGF 164 protein isoform maintains its role of angiogenic signaling throughout growth. The decrease in VEGF 164 mRNA, along with the increase in VEGF 188 mRNA between P35 and P49, correlates with the normal decrease in growth plate proliferation seen at these ages. The down-regulation of VEGF 164 mRNA and corresponding decrease in VEGF 164 protein could lead to impaired chondrocyte proliferation, resulting in the slowed bone growth that occurs within the last phase of long bone growth before growth plate closure.
The extracellular matrix of the growth plate can function as a reservoir for growth factors controlling the diffusion capacity, which influences the surrounding chondrocyte function (van der Eerden et al.,2003). Both VEGF 120 and VEGF 164 proteins are secreted from cells and are either fully or partially soluble. The VEGF 164 and VEGF 188 protein isoforms have affinity for heparin proteins and, thus, they remain bound to the cell surface and extracellular matrix (Ferrara and Gerber,2001). The higher level of VEGF protein expression within the extracellular matrix at early ages of postnatal endochondral bone growth (P14–P35) versus later ages (P49–P77), indicates that the heparin binding forms of VEGF protein (VEGF 164 and VEGF 188) may play a crucial role in early postnatal endochondral bone development. Although VEGF protein was detected throughout the extracellular matrix at all growth plate regions in early postnatal ages in the present study, it would appear that the effects of the VEGF protein are limited to the late proliferative and early hypertrophic regions of the growth plate as reflected in the location of FLK-1 receptor protein.
The expression pattern in prenatal mice of VEGF protein was similar to that seen in postnatal bone development as detected in the present study. In one study of mouse embryos, VEGF protein was present in fully mature hypertrophic chondrocytes with sporadic expression in prehypertrophic chondrocytes, whereas the VEGF receptor FLK-1 expression was restricted to the hypertrophic region (Carlevaro et al.,2000), similar to the expression pattern seen in the present study. However, VEGF protein expression was not detected in the proliferative and resting chondrocytes in prenatal mice at embryonic age E16 (Carlevaro et al.,2000). This finding is in contrast to the low VEGF protein expression detected in the proliferative region cells of the present study and to a study of VEGF protein expression in the resting and proliferative regions of early human fetal growth plates (Garcia-Ramirez et al.,2000).
The relative amounts of VEGF mRNA for the various VEGF protein isoforms were also evaluated in this study. Previous studies have shown that VEGF 120 makes up approximately 30–50% of the total VEGF mRNA amount in embryonic bone, whereas VEGF 164 constitutes 50–70% of total VEGF mRNA and VEGF 188 mRNA is the least expressed isoform in embryonic bone development at less than 1% (Maes et al.,2002). The present study showed that VEGF 120 mRNA was the predominant form expressed throughout postnatal bone development with low levels of VEGF 188. Although each isoform contributes differently to the total VEGF mRNA expression within the growth plate, the location within the growth plate of isoform transcription and translation is important to understand VEGF's role in endochondral angiogenesis as linear growth slows and the growth plate prepares for closure. In the present study, the proliferative and early hypertrophic regions were the predominant regions exhibiting VEGF mRNA expression. In previous studies, predominant VEGF mRNA expression was noted primarily in the hypertrophic regions with occasional in situ hybridization signal detected in chondrocytes of the proliferative and resting regions (Gerber et al.,1999; Smink et al.,2003; Maes et al.,2004). Although low total VEGF mRNA transcription levels were noted in the resting and late hypertrophic regions in the present study, it is possible that transcription of VEGF mRNA and translation of the various protein isoforms are not correlated. In fact, in the late hypertrophic region of P28 mice, VEGF 164 mRNA showed a trend toward down-regulation relative to VEGF 188, although VEGF protein expression was detected consistently in this region. Further studies are warranted to explore the exact VEGF protein isoform expressed postnatal within this region.
Hypoxia inducible factor-1 (HIF-1), insulin like growth factor-I, and insulin are considered primary factors regulating VEGF production within the growth plate (Semenza and Wang,1992; Goad et al.,1996; Zelzer et al.,1998). Oxygen tension is lowest in hypertrophic chondrocytes (Brighton and Heppenstall,1971). The most hypoxic chondrocytes are those located in the center of the columnar proliferative region and the upper portion of the hypertrophic region corresponding to areas that express relatively high levels of HIF-1 (Schipani et al.,2001). Hypoxic chondrocytes that lack HIF-1 undergo massive cell death (Schipani et al.,2001). A decrease in HIF-1 in the late hypertrophic region of the growth plate could result in a decrease of VEGF 164 mRNA as was seen in the P28 mice of this study. Lowered VEGF 164 may signal late hypertrophic cell apoptosis during postnatal endochondral bone development. In addition, our results demonstrating higher VEGF 164 mRNA expression in the proliferative and early hypertrophic regions relative to VEGF 188 could correlate with the increased HIF-1 expression in these regions as shown by Schipani and colleagues (Schipani et al.,2001). Additional studies are needed to explore the factors controlling the differential expression of VEGF mRNA isoforms and their subsequent protein products.
In addition to proteins inducing VEGF expression, other factors may sequester VEGF protein, thereby preventing its action in various growth plate regions despite its presence. One such protein is connective tissue growth factor (CTGF), which has been shown to be induced by VEGF and binds to the VEGF protein to inhibit its ability to induce angiogenesis. In embryonic mice, CTGF is expressed highly in terminal hypertrophic cells (Ivkovic et al.,2003) and may sequester VEGF in an inactive form in the hypertrophic extracellular matrix. This could lead to the increased presence of protein seen in the hypertrophic region in the postnatal mice of the present study if this protein prevents VEGF protein degradation leading to an increase in protein signal despite the apparent decrease in mRNA expression.
VEGF has a role in endochondral bone development in the embryo. The present study confirms a continued role for VEGF throughout bone elongation and indicates a function in the cessation of bone growth through growth plate closure. The changes in mRNA and protein expression, although in different growth plates at potentially different growth stages, correspond to reduced proliferative capacity of the growth plate. Although the relative expression patterns in VEGF mRNA are similar for prenatal and postnatal development, the age dependent changes in the expression of the VEGF isoforms suggests a role in the postnatal maturation of long bone, particularly at the age of growth cessation preceding growth plate closure.
We thank Christian Leutenegger and the Lucy Whittier Molecular Genomics Laboratory for Taqman analysis and validation.
- 1994. Induction of growth plate cartilage ossification by basic fibroblast growth factor. Endocrinology 135: 2790–2793. , , , , , , .
- 2000. Marrow stromal stem cells. J Clin Invest 105: 1663–1668. , .
- 2007. VEGF and VEGF receptors are differentially expressed in chondrocytes. Bone 40: 568–576. , , , , , , .
- 1971. Oxygen tension in zones of the epiphyseal plate, the metaphysis and diaphysis. An in vitro and in vivo study in rats and rabbits. J Bone Joint Surg Am 53: 719–728. , .
- 1997. Transferrin promotes endothelial cell migration and invasion: implication in cartilage neovascularization. J Cell Biol 136: 1375–1384. , , , , , , , .
- 2000. Vascular endothelial growth factor (VEGF) in cartilage neovascularization and chondrocyte differentiation: auto-paracrine role during endochondral bone formation. J Cell Sci 113(Pt 1): 59–69. , , , .
- 2000. Expression of vascular endothelial growth factors and their receptors during osteoblast differentiation. Endocrinology 141: 1667–1674. , , , , , .
- 1999. Reverse transcription-polymerase chain reaction products of alternatively spliced mRNAs form DNA heteroduplexes and heteroduplex complexes. J Biol Chem 274: 2613–2615. , , , , .
- 1986. Histologic comparison of the costochondral, sternoclavicular, and temporomandibular joints during growth in Macaca mulatta. J Oral Maxillofac Surg 44: 312–21. , .
- 2004. Vascular endothelial growth factor: basic science and clinical progress. Endocr Rev 25: 581–611. .
- 2001. The role of vascular endothelial growth factor in angiogenesis. Acta Haematol 106: 148–156. , .
- 1997. Vascular endothelial growth factor: basic biology and clinical implications. EXS 79: 209–232. , .
- 1996. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380: 439–442. , , , , , , , , .
- 2005. VEGF-A and PlGF-1 simulate chemotactic migration of human mesenchymal progenitor cells. Biochem Biophys Res Commun 334: 561–568. , , , , .
- 2000. Vascular endothelial growth factor is expressed in human fetal growth cartilage. J Bone Miner Res 15: 534–540. , , , , .
- 1999. VEGF couples hypertrophic cartilage remodeling, ossification and angiogenesis during endochondral bone formation. Nat Med 5: 623–628. , , , , , .
- 1996. Enhanced expression of vascular endothelial growth factor in human SaOS-2 osteoblast-like cells and murine osteoblasts induced by insulin-like growth factor I. Endocrinology 137: 2262–2268. , , , , .
- 2005. Localization of matrix metalloproteinases, (MMPs) their tissue inhibitors, and vascular endothelial growth factor (VEGF) in growth plates of children and adolescents indicates a role for MMPs in human postnatal growth and skeletal maturation. Calcif Tissue Int 76: 326–335. , , , .
- 1999. Immunolocalisation of vascular endothelial growth factor (VEGF) in human neonatal growth plate cartilage. J Anat 194(Pt 4): 519–524. , , , , , , .
- 1994. Mechanism of longitudinal bone growth and its regulation by growth plate chondrocytes. Microsc Res Tech 28: 505–519. .
- 2003. Connective tissue growth factor coordinates chondrogenesis and angiogenesis during skeletal development. Development 130: 2779–2791. , , , , , , , .
- 1999. The genetic transformation of bone biology. Genes Dev 13: 3037–3051. .
- 1999. Quantitative real-time PCR for the measurement of feline cytokine mRNA. Vet Immunol Immunopathol 71: 291–305. , , , , , .
- 2002. Impaired angiogenesis and endochondral bone formation in mice lacking the vascular endothelial growth factor isoforms VEGF164 and VEGF188. Mech Dev 111: 61–73. , , , , , , , .
- 2004. Soluble VEGF isoforms are essential for establishing epiphyseal vascularization and regulating chondrocyte development and survival. J Clin Invest 113: 188–199. , , , , , , , .
- 2000. Vascular endothelial growth factor (VEGF) directly enhances osteoclastic bone resorption and survival of mature osteoclasts. FEBS Lett 473: 161–164. , , , , , , , , .
- 2002. The neuropilins: multifunctional semaphorin and VEGF receptors that modulate axon guidance and angiogenesis. Trends Cardiovasc Med 12: 13–19. , , , , , .
- 1995. Fractionation of growth plate chondrocytes: differential expression of IGF-I and growth hormone and IGF-I receptor mRNA in purified populations. Connect Tissue Res 31: 179–187. , .
- 1992. Linear bone growth of oMT1a-oGH transgenic male mice. Am J Physiol 262: E936–E942. , , , , .
- 2000. Bone development. Annu Rev Cell Dev Biol 16: 191–220. , , .
- 2001. The splice variants of vascular endothelial growth factor (VEGF) and their receptors. J Cell Sci 114: 853–865. , .
- 2001. Hypoxia in cartilage: HIF-1alpha is essential for chondrocyte growth arrest and survival. Genes Dev 15: 2865–2876. , , , , , .
- 1992. A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol 12: 5447–5454. , .
- 2003. Short-term glucocorticoid treatment of piglets causes changes in growth plate morphology and angiogenesis. Osteoarthritis Cartilage 11: 864–871. , , , , , , , , .
- 1998. Neuropilin-1 is expressed by endothelial and tumor cells as an isoform-specific receptor for vascular endothelial growth factor. Cell 92: 735–745. , , , , .
- 2003. Systemic and local regulation of the growth plate. Endocr Rev 24: 782–801. , , .
- 1998. Insulin induces transcription of target genes through the hypoxia-inducible factor HIF-1alpha/ARNT. EMBO J 17: 5085–5094. , , , , , .
- 2004. VEGFA is necessary for chondrocyte survival during bone development. Development 131: 2161–2171. , , , , , .