SEARCH

SEARCH BY CITATION

Keywords:

  • osteoblasts;
  • apoptosis;
  • craniosynostosis;
  • NELL-1;
  • Nell-1

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

We studied the cellular function of Nell-1, a craniosynostosis-related gene, in craniofacial development. Nell-1 modulates calvarial osteoblast differentiation and apoptosis pathways. Nell-1 overexpression disrupts these pathways resulting in craniofacial anomalies such as premature suture closure.

Introduction: Craniosynostosis (CS), one of the most common congenital craniofacial deformities, is the premature closure of cranial sutures. Previously, we reported NELL-1 as a novel molecule overexpressed during premature cranial suture closure in patients with CS. Nell-1 overexpression induced calvarial overgrowth and resulted in premature suture closure in a rodent model. On a cellular level, Nell-1 is suggested to promote osteoblast differentiation.

Materials and Methods: Different levels of Nell-1 were introduced into osteoblastic cells by viral infection and recombinant protein. Apoptosis and gene expression assays were performed. Mice overexpressing Nell-1 were examined for apoptosis.

Results: In this report, we further showed that overexpression of Nell-1 induced apoptosis along with modulation of apoptosis-related genes. The induction of apoptosis by Nell-1 was observed only in osteoblastic cells and not in NIH3T3 or primary fibroblasts. The CS mouse model overexpressing Nell-1 showed increased levels of apoptosis in the calvaria.

Conclusion: We show that Nell-1 expression modulates calvarial osteoblast differentiation and apoptosis pathways. Nell-1 overexpression disrupts these pathways resulting in craniofacial anomalies such as premature suture closure.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

CRANIOSYNOSTOSIS (CS), the premature closure of the cranial sutures, is a common human congenital craniofacial deformity affecting 1 in 2500–3000 infants.(1) It occurs either familially or sporadically, with neither gender nor ethnic preference. Genetic linkage analyses of CS-related syndromes have provided a wealth of new information regarding the genetic origins of CS(1,2); however, the biological mechanisms of normal suture patency and premature closure remain largely unknown. This is especially true of nonsyndromic, nonfamilial CS.

Mechanism of premature suture closure

Suture closure/patency is a multigenic process involving a complicated orchestration of temporal and spatial specific interactions/expressions of growth factors and receptors.

Premature suture closure can be divided into two possibly distinct processes: calvarial overgrowth and bony fusion. While calvarial overgrowth may be essential in overlapping the opposing osteogenic fronts for bony fusion, their juxtaposition does not necessarily result in bony fusion. Thus, to understand the mechanism of premature suture closure, one should study both abnormal suture overgrowth/overlap and bony fusion.(3) In both processes, the proposed mechanisms for premature suture closure includes an imbalance between osteogenic inhibiting and inducing factors and the dysregulation of proliferation, differentiation, and apoptosis at the abnormally closed osteogenic fronts.(4–6)

The precise role apoptosis plays in CS, however, is not well understood. Current evidence remains controversial. Furtwangler hypothesized that when the bone fronts are adjacent, apoptosis should occur along the bone/osteogenic front to prevent closure.(4) Opperman et al.(7) provided strong evidence that suture patency is achieved by increased apoptosis accompanied by decreased cell proliferation. Opperman's TGF-β study on cranial sutures further suggests that an increase in proliferation and a decrease in apoptosis will induce suture closure.

On the other hand, mutations that induce calvarial osteoblast apoptosis, such as the Apert S252W FRFR2 mutation(5,8) and the TWIST haploinsufficiency in Saethre-Chotzen syndrome,(9) are characterized by the premature onset of suture closure. These findings suggest that an increase in both osteoblast differentiation and apoptosis would induce suture fusion and ossification.

The isolation of Nell-1 from unilateral CS

We previously reported the isolation and identification of a novel gene, NELL-1, in human unilateral coronal CS (identified simultaneously by Watanabe et al.(10) and our group(11)). NELL-1 is a secretory protein that contains an open reading frame encoding 810 amino acids.(12,13)NELL-1 encodes a signal peptide, a NH2-terminal thrombospondin (TSP)-like module,(14) five von Willebrand factor C domains, and six epidermal growth factor (EGF)-like domains. NELL-1 expression was upregulated in the coronal sutures that were fused or prematurely fusing compared with normal coronal sutures. We demonstrated that in sites of premature suture fusion, the human NELL-1 mRNA was localized primarily to the mesenchymal cells and osteoblastic cells at the osteogenic front, along the parasutural bone margins, and within the condensing mesenchymal cells of newly formed bone in sutural sites.(11) The NELL-1 gene is preferentially expressed in cranial intramembranous bone and neural tissue, which are both of neural crest cell origin. We further verified that NELL-1 plays a role in CS by creating a transgenic mouse model exhibiting generalized Nell-1 overexpression.(15) This model displays a similar phenotype to human CS with calvarial overgrowth/overlap and premature suture closure with increased differentiation and decreased proliferation. Infection of osteoblastic cells with Nell-1 adenoviral constructs (AdNell-1) showed that Nell-1 promotes and accelerates differentiation. In addition, Nell-1 downregulation inhibited osteoblast differentiation. Nell-1, therefore, represents a candidate gene involved in cranial suture closure.

Potential Nell-1 function

This report investigates the function of Nell-1 in osteoblast apoptosis. The results suggest that the overexpression of Nell-1 induces altered bone formation in the calvariae by involvement of the apoptosis process. This increased apoptosis may therefore play a role in premature suture closure. The precise molecular mechanism of this gene is unknown.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Cell culture

The isolation of fetal rat calvarial cells (FRCC) from embryonic day 18 (E18) rat calvaria was performed as previously described.(16) FRCC within passage two and MC3T3 osteoblast cells under passage 25 were used for all experiments under standard culture condition with or without addition of 50 μg/ml ascorbic acid and 10 mM β-glycerophosphate in α-MEM supplemented with 10% fetal bovine serum (FBS; Gibco, Gaithersburg, MD, USA). The differentiation stage of FRCC was determined by either alkaline phosphatase staining (Sigma, St Louis, MO, USA) or ABC (Vector Laboratories, Burlingame, CA, USA) immunocytochemistry of osteocalcin (Hematologic Technologies Inc., Essex Junction, VT, USA).

Nell-1 overexpression

Adenoviruses (AD5 with an E1-A knockout and CMV promoter) were used as the method of gene transfer. Viruses were constructed according to the previous protocol.(15) The appropriate concentration of virus (10, 20, and 50 pfu/cell) was added with the minimum amount of medium.(15) MC3T3, NIH3T3, FRCC, and primary fibroblast cells were infected with AdNell-1, and Adβ-Gal was used as the control. DNase I-treated total RNA was used. The polymerase chain reaction (PCR) was performed by reducing cycle mode to quantify relative gene expression as described previously.(15) The intensities were measured with PhosphorImager (Molecular Dynamics, Sunnyvale, CA, USA), and the densitometry value of Nell-1 expression was normalized against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) value. The specific primer sequences for Nell-1 (GenBank accession no. NM 031096) were as follows: forward, 5′-CTGTGTGGCTCCTAACAAGTGTG-3′; reverse, 5′-GGATTCTGGCAATCACAAGCTGCT-3′; and probe, 5′-CCTACTCACTGTCCGGGGAGTCCTGC-3′.

Western blot detection for Nell-1 protein expression

The infected FRCCs on postinfection days 3 and 6 were used. The protein concentration was measured by the Bradford method using Biorad protein assay reagent (Biorad Laboratories, Hercules, CA, USA). One hundred micrograms of protein from each sample was loaded onto a 7.5% Tris-HCl ready gel under a completely denatured condition and blotted. Anti-β actin antibody was used for normalization purpose. The specificity of Nell-1 antibody was confirmed as described previously.(12,13,15) The relative amount of Nell-1 in cell lysate and culture medium was quantified using the National Institutes of Health Image, with recombinant protein as a standard.

Nell-1 protein addition

The affinity purified Nell-1-flag protein from rat Nell-1 transfected mammalian cells was added to the 70% confluent FRCC cells on 24-well plates at varying amounts ranging from 5 to 1000 ng/ml as indicated. The cells were incubated for 3 days with α-MEM medium containing 10% FBS. Then, the cell culture was grouped into two sets, with and without ascorbic acid and β-glycerophosphate addition in the medium at 50 μg/ml and 10 mM, respectively. The purified Nell-1 protein and fresh medium were replaced every 3 days until the end of the experiment. The morphology of the cells was monitored daily, and photomicrographs were taken.

Cell death analyses

For in situ Annexin-V-PI labeling, the infected FRCC cells grown on coverslips were incubated with both Annexin-V-FITC and propidium iodide (PharMingen, San Diego, CA, USA) followed by fixation with 4% paraformaldehyde PBS. Photomicrographs were taken with 495- and 595-nm fluorescent filters. For trypan blue staining, infected cells were stained with 0.2% trypan blue. The stained cells were counted microscopically under 40× objective. Briefly, 10 fields were randomly selected, and all trypan blue-stained cells were counted for each group. The percentage of positive staining was expressed as total number of stained cells over the total number of cells from all of these 10 fields. The method of estimating cell counts in each field was adopted from Simpson et al.(17) To further verify that MC3T3 cells have similar phenotypic changes, in situ TUNEL was performed on day 12 postinfection. The apoptotic cells were labeled by TACS In Situ Apoptosis Detection Kits (R&D Systems, Minneapolis, MN, USA) on cultured cells according to our previous protocol.(18) The stained cells were viewed under a fluorescence microscope using a 495-nm filter. Triplicates were performed.

cDNA arrayed hybridization

To examine the gene expression pattern, MC3T3 cells from postinfection days 6 and 9 were used. cDNA arrays (ClonTech, Palo Alto, CA, USA) and their application have been described previously in the literature and our previous protocol.(18) The experiments were repeated to verify reproducible data. To prevent a false positive result, only apoptosis-related genes with expression level differences ≥2-fold were reported. Primer-specific reverse transcriptase (RT)-PCR was used to verify the results. To further verify result form the cDNA arrays, we examined the gene expression pattern of Trail (GenBank accession no. U37522) and Fas (GenBank accession no. M83649) with RT-PCR. The specific primer sequences for Trail were as follows: forward, 5′-AGCTAAGTACTCCTCCCTTGCC-3′; reverse, 5′-GGTTCTCACCTTGTCCTTTGAG-3′; and probe, 5′-GCTCTTTAGGAATGGAGAGCTG-3′. The specific primer sequences for Fas were as follows: forward, 5′-TGTGAACATGGAACCCTTGA-3′; reverse, 5′-CCATGAGATTGGTACCAGCA-3′; and probe, 5′-GATCATGCATGACAGCATCC-3′.

Generation of transgenic mice with Nell-1 overexpression

The transgenic mice with Nell-1 overexpression were constructed and reported previously.(15) Newborn mice were killed and fixed in 4% paraformaldehyde. The genotyping was done by PCR. The morphological studies, including von Kossa staining and in situ labeling of apoptotic cells, were performed on paraffin-embedded tissue sections. Tissue sections from prematurely fusing sutures and normal sutures were labeled with the DeadEnd Colorimetric Apoptosis Detection System (Promega Corp., Madison, WI, USA).(18)

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Overexpression of Nell-1 induces cell death in vitro

FRCCs were infected with AdNell-1 at 10, 20, and 50 pfu/cell, and the level of Nell-1 expression was determined by reduced cycle RT-PCR. Compared with endogenous Nell-1 expression, the level of Nell-1 expression was 2-, 5-, and 8-fold higher in the 10, 20, and 50 pfu/cell cultures, respectively (Figs. 1A and 1B). Nell-1 protein expression in the cell extract and medium were quantitated (Fig. 1C) by Western blot. The concentration of Nell-1 protein in the cell extracts were 0.20, 0.45, and 0.60 ng/μg in the 10, 20, and 50 pfu/cell cultures, respectively. The concentration of secreted Nell-1 protein was 14.0, 28.5, and 77.0 ng/ml, respectively. The level of Nell-1 RNA and protein expression was higher than found endogenously but was comparable with concentrations researchers routinely used in TGF-β and bone morphogentic protein (BMP) studies.(19,20) The efficiency of infection was verified with Adβ-Gal infection (Fig. 1D).

thumbnail image

Figure FIG. 1.. Effects of Nell-1 overexpression on cell death in osteoblastic cells. (A) RT-PCR (25 cycle) analysis on Nell-1 expression using 10, 20, and 50 pfu/cell at day 3 and 6 postinfection time points. (B) Histogram of densitometry analysis of Nell-1 expression standardized to Gapdh. (C) Western blot of Nell-1 protein expression in cell extracts and medium from FRCCs infected with 10, 20, and 50 pfu/cell at day 6 postinfection. (D) Infection efficiency with Adβ-Gal at 10, 20, and 50 pfu/cell in FRCCs. (E) FRCCs infected with 20 pfu/cell AdNell-1 day 6 postinfection. Control cell cultures infected with Adβ-Gal. Alkaline phosphatase staining is shown in red. (F) FRCCs infected with 20 pfu/cell AdNell-1 day 9 postinfection. Control cell cultures infected with Adβ-Gal. Sample of aggregates of rounding cells are indicated by dashed lines. Photos are taken under inverted microscope without counter stain. (G) Cells are stained with osteocalcin antibody. AdNell-1-infected cells have significantly stronger staining than the noninfected cells (shown in red). (H) Annexin V-FITC and PI staining of a newly formed nodule at day 9 postinfection. Annexin V staining is green and PI staining is red. Most cells expressing Annexin V are PI negative, with PI presenting early stage of potential cell apoptosis with AdNell-1 infection. Cells infected with Adβ-Gal are negative for Annexin V (diffused green background) with minimal PI staining (diffused red background). (I) Annexin V and PI staining of a matured nodule with AdNell-1 infection. Left: Annexin V staining. Middle: PI staining. Right: Annexin V and PI overlap. Cells stained with both Annexin and PI are shown in yellow. Most cells expressing Annexin V are positive, with PI staining presenting a later stage of apoptosis. (J) FRCCs infected with 20 pfu/cell AdNell-1 day 12 postinfection stained with trypan blue. Dead cells are stained in blue. A significant number of cells in a nodule are in stages of death. Top right: AdNell-1 infection without ascorbic acid (AA) and β-glycerophosphate. Bottom right: control cell cultures infected with Adβ-Gal. (K) FRCCs infected with 50 pfu/cell AdNell-1 day 6 postinfection stained with trypan blue. Significant cell death occurred without nodule formation in the osteoblast differentiation process. D, E, F, H, and I original magnification: 40×. G, J, and K original magnification: 200×.

Download figure to PowerPoint

At 20 pfu/cell, we verified that AdNell-1 infected FRCC had increased expression of differentiation markers, including alkaline phosphatase and osteocalcin, compared with Adβ-Gal control (Fig. 1E). AdNell-1 infected FRCC exhibited an increase in cell death, with the previously reported increase in differentiation and mineralization (Figs. 1E–1G). Annexin V staining and propidium iodide (PI) double staining were performed 9 days after infection, when the morphology of cell death first appeared (Fig. 1H). Multiple cells stained with Annexin V were observed in the newly formed nodular areas (cell aggregates). PI staining overlapped the distribution seen with Annexin V, confirming cell death in the later stage (Fig. 1I). Cell death became increasingly pronounced by postinfection day 12. By this point, staining with trypan blue revealed a significant amount of cell death within the well-formed nodules (Fig. 1J). AdNell-1 without the addition of ascorbic acid plus β-glycerophosphate and the Adβ-Gal control showed few dead cells (Fig. 1J). Thus, it seems that Nell-1 overexpression induces cell death in osteoblastic cells when cultured in differentiating medium. In the absence of ascorbic acid plus β-glycerophosphate, undifferentiated osteogenic cells were not driven to cell death by overexpression of Nell-1.

Trypan blue stain was used to calculate the number of nonviable cells at each level of AdNell-1 infection. A dose-dependent response between the infective dose and the cell death rate was observed. On postinfection day 12, 13.6 ± 3.37% of cells were dead at an infective dose of 10 pfu/cell, 15.6 ± 5.63% at 20 pfu/cell, and 32.52 ± 7.83% at 50 pfu/cell. Significant cell death was observed as early as day 6 at the 50 pfu/cell concentration. In contrast, only 3.2 ± 0.56% of cells died in the Adβ-Gal controls, which did not vary significantly at the various levels of infection. There was a statistically significant difference between the percent of cell death seen in all of the AdNell-1 groups relative to the Adβ-Gal controls (p < 0.05). The percentage difference between the AdNell-1 groups at 10, 20, and 50 pfu/cell was also significant (p < 0.05). When ascorbic acid plus β-glycerophosphate, a necessary component for osteoblast differentiation, was eliminated from the media, a death rate of only 0.6% was consistently observed in all groups, regardless of the infecting concentration. This is suggestive that differentiation is required for Nell-1 to induce osteoblast apoptosis.

At an infecting concentration of 50 pfu/cell, a dramatic increase in apoptosis was noted by day 6 (Fig. 1K). At this point, there was no nodular formation, and the pattern of cell death was diffusely distributed. This suggests that at the higher levels of Nell-1 achieved in the 50 pfu/cell group, osteoblastic cell apoptosis was prematurely induced, even before terminal differentiation. In contrast, cell death in the control NIH3T3 and primary fibroblasts showed no significant differences at any of the infective doses or co-ascorbic acid plus β-glycerophosphate culture conditions (data not shown). Collectively, this suggests that Nell-1 overexpression is not, in and of itself, cytotoxic. It would seem that Nell-1 may induce cell death/mineralization through ascorbic acid-dependent promotion and/or dysregulation of osteoblast-specific differentiation pathways. Although flow cytometry is more accurate than field counting in general, flow cytometry was not used because of the rich extracellular matrix characteristics of the osteogenic cell culture and the resulting difficult mechanical obstruction through the cytometer.

Nell-1-induced cell death is linked to apoptosis

Experiments were repeated with MC3T3 cells and similar cell death phenotype was observed. To verify that the observed cell death is linked to apoptosis, in situ TUNEL labeling was performed using the 20 pfu/cell culture at postinfection day 12. Day 12 was chosen because of the high levels of cell death observed on that day. Significant numbers of apoptotic cells were observed by in situ TUNEL (Fig. 2A), showing that apoptosis was the cause of cell death.

thumbnail image

Figure FIG. 2.. Apoptosis analysis on AdNell-1-infected osteoblastic cells. (A) In situ TUNEL assay on postinfection day 12. Left: AdNell-1-infected cells. Right: control Ad β-Gal-infected cells. (B) Expression of apoptosis-related genes in AdNell-1-infected MC3T3 cells by cDNA array analysis. Graph summarizing genes with a 2-fold greater difference in expression after AdNell-1 infection. The Nell-1/β-Gal ratio summarizing the expression measured by PhosphorImager and normalized to Gapdh. (C) RT-PCR verification of cDNA array data. N6 and N9: AdNell-1 infected cells at postinfection days 6 and 9, respectively; β6 and β9: Adβ-Gal infected cells at postinfection days 6 and 9, respectively. (D) FRCCs with 5 ng/ml Nell-1 protein on day 6 post-Nell-1 protein application. (D′) Higher magnification of a newly formed nodule with significant cell death. (E) Increased amount of cell death observed around a nodule (50 ng/ml). (E′) Minimal cell death observed when cultured without ascorbic acid. (F) Number of nodules was reduced and cell death was scattered in the culture (200 ng/ml). (F′) Higher magnification of scattered dead cells. (G) Low cell death observed(400 ng/ml). (H) Control without Nell-1 protein. Minimal dead cells were observed. A, D, E, E′, F, G, and H original magnification: 40×.

Download figure to PowerPoint

cDNA array analysis in AdNell-1 infected MC3T3 revealed upregulation of several apoptosis-related genes (Fig. 2B). At postinfection day 6, Nell-1 significantly induced Fas antigen (Fas I receptor/Apo 1),(21)Tnf receptor 1 precursor, Gadd45 (growth arrest protein),(22) and Trail genes by greater than 2-fold (Fig. 2B). These results were confirmed by time course-reduced cycle RT-PCR (Fig. 2C). Notably, the Fas antigen and TNF-α are known to synergistically induce apoptosis in osteoblastic cells,(23) whereas Tnf receptor upregulation has been shown to induce apoptosis in the absence of TNF-α.

Nell-1 protein induces osteogenic cell death

To exclude the possibility that the adenovirus vectors induced cell death nonspecifically, Nell-1 flag protein was purified from a stable expression system and added at 5, 10, 20, 50, 100, 200, 400, 800, 1000 ng/ml to 24-well FRCC plates with and without ascorbic acid plus β-glycerophosphate. Apoptosis was observed only in cultures receiving ascorbic acid plus β-glycerophosphate and appeared 6 days after addition (Figs. 2D–2H). We observed apoptosis in the range of 5–50 ng/ml, with 50 ng/ml being the optimal concentration to induce apoptosis. Concentrations above 100 ng/ml demonstrated reduced activity in inducing apoptosis. This dose-dependent biphasic effect is not uncommon and is seen in anabolic cytokines and growth factors.(24,25)

The bioactivity of the purified Nell-1 flag protein was confirmed by assaying the alkaline phosphatase activity. A total of 50 ng/ml Nell-1 induced 160% alkaline phosphatase compared with control with the presence of differentiating medium (data not shown).

Nell-1 overexpression and apoptosis in transgenic mice

Mice with CMV promoter-driven Nell-1 overexpression were generated.(15) Genotyping was performed (Fig. 3A), and transgenic mice with overlapping sutures were examined (Figs. 3B and 3C). Consistent with our in vitro findings, an increased incidence of apoptosis was seen using in situ TUNEL analysis at the osteogenic fronts of premature suture closure, as well as at the periosteum where active intramembranous bone formation occurred (Figs. 3D, 3D′, 3E, and 3E′). This finding is consistent with several Nell-1 positive littermates. In combination with our previous findings from these Nell-1 transgenic mice,(15) it is clear that their calvarial bone had fewer total number and proportion of cells in the proliferation stage, but increased number in differentiation with increased apoptosis. In the normal littermates, few apoptotic cells were observed in the suture and calvarial bone.

thumbnail image

Figure FIG. 3.. Apoptosis in Nell-1 transgenic mice compared with nontransgenic littermates. (A) Transgene copy number. The transgenic mice that we analyzed has approximately 50 copies of the transgenes. The PCR protocol of establishing transgene copy number is according to http://www.med.umich.edu/tamc/spike.html. (B) Coronal section of a newborn Nell-1 phenotype-positive mouse (left). The sagittal and posterior-frontal sutures (black arrow) are closed. Note the brain protrusion in the parietal area (yellow arrows). A nontransgenic littermate with patent sagittal and PF sutures (black arrow) and normal vasculature underneath the patent sutures (right). (C) von Kossa and H & E staining analysis of the Nell-1 transgenic mouse sagittal suture. Black staining represents mineralized calvarial bone. V represents blood vessel. Original magnification: 100×. (D) In situ TUNEL analysis of the Nell-1 transgenic mouse sagittal suture. Apoptotic cells are stained brown (red arrows). Black dashed lines indicate calvarial bone. Patent sagittal suture from a normal littermate control is also shown as inset (black rectangle). Area depicted within the yellow dashed rectangle is shown further magnified as D′. Original magnification: 200×. (E) In situ TUNEL analysis of the calvarial bone of a Nell-1 transgenic mouse. Apoptotic cells are stained brown (black arrows). Calvarial bone of a nontransgenic littermate is shown as inset (black rectangle). Area depicted within the green dashed rectangle is shown further magnified as E′. Original magnification: 200×. Bar scale: 50 μm.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

Opperman and others proposed that, for suture fusion to occur, an initial phase of proliferation is necessary to achieve a cell density-induced phase of differentiation. This initiates the suture closure process and apoptosis of those cells within the sutural matrix.(7,26) Cytokines and their receptors play important and diverse roles in this process in which cellular response to cytokines depends on several factors.(7,27) The level of stimulation, the inherent differences between in vivo and in vitro osteoblastic cells, the influence of the extracellular milieu, and the various stages of cellular differentiation are all factors to be considered.

Observations from the in vitro system reveals that cells differentiating to the mature osteoblast phenotype go onto programmed cell death.(28) As Lian's group has shown, apoptosis begins to increase in the late differentiation stage and later significantly increases during the mineralization stage.(28) The addition of stimuli to enhance differentiation and mineralization of osteoblast, such as dexamethasone, will increase apoptosis. The absence of ascorbic acid, thus preventing differentiation, and the absence of β-glycerophosphate, thus preventing mineralization, significantly reduces the rate of apoptosis.

The role of Nell-1 in the induction of apoptosis therefore may be caused by either (1) increased osteoblast differentiation leading to mineralization or (2) a different mechanism independent of the differentiation and mineralization pathways. At 20 pfu/cell, the Nell-1-induced apoptosis is similar to the increased apoptosis demonstrated by Lian's group with the addition of dexamethasone. Apoptosis is associated with differentiated nodules. Therefore, the increased apoptosis observed is likely because of the accelerated differentiation caused by Nell-1. This was verified by analysis of specific bone markers, such as type 1 collagen, alkaline phosphatase, osteopontin, and osteocalcin. It seems as though there is a generalized shift of these bone differentiation markers toward an earlier time point.(15)

During high levels of Nell-1 overexpression, however, a significant percent of the osteogenic cells underwent apoptosis as early as 4–6 days postinfection. The apoptosis was generalized and not specifically associated with bone nodule formation. This suggests the possibility that Nell-1 might also have a separate distinct apoptotic process bypassing the terminal differentiation and mineralization stage. The surviving cells showed much delayed mineralization, suggesting that Nell-1 may selectively induce apoptosis in differentiating osteoblastic cells with mineralization potential. This hypothesis was further supported by the failure of Nell-1 to induce apoptosis in undifferentiated osteogenic cells deprived of ascorbic acid and β-glycerophosphate. As we have previously reported, Nell-1 can induce progression to the early stages of differentiation in the absence of ascorbic acid and β-glycerophosphate. This suggests that Nell-1 exerts its differentiating function on immature osteoblasts. On the other hand, Nell-1 induces apoptosis only when cells are exposed to the proper conditions for differentiation. Thus, Nell-1 likely plays different roles in osteoblastic cells depending on their stage of differentiation.

Nell-1-induced cell death may be caused by a supraphysiological level of cytotoxic Nell-1. However, in the absence of ascorbic acid, Nell-1 failed to induce cell death even at high expression levels. Cell death also did not occur in other cell types exposed to Nell-1, such as fetal fibroblasts and NIH3T3 cells. These results suggest that a generalized cytotoxic induced cell death is unlikely and that Nell-1 induced cell death may require specific receptor(s)/protein interaction to be present for apoptosis to occur. Apoptosis was demonstrated to be the main process of cell death.

Another concern regarding Nell-1-induced cell death is that overexpression may not reflect the true physiological function of Nell-1, but rather an interference or cross activity with other TSP-like molecules with high homology. Overexpression of Nell-1 may induce osteoblast apoptosis by binding or sequestering ligands or by triggering nonspecific receptor-mediated signaling.(29) For example, the Nell-1 protein may function extracellularly, similar to the TSP-1 molecule or chordin which interact with TGF-β superfamilies.(30,31) As we have shown in previous Nell-1 knock-down experiments, Nell-1's interaction with osteoblastic cells is not likely caused by the mimicking of or interference with other TSP-like molecules. However, the exact molecular mechanism of Nell-1-induced apoptosis is not yet understood. Apoptosis-related genes, including the Fas antigen (Fas I receptor/Apo 1), the Tnf receptor 1 precursor, Trail, and Gadd45, were transiently upregulated by greater than 2-fold. These genes implicate the potential pathway for Tnf/Fas induced apoptosis. The Fas pathway has been shown to play a major role in normal osteoblast apoptosis.(21,23) The Tnf and Fas pathways were also shown to be involved in osteoblast apoptosis in the S252W FGFR2 Apert CS and TWIST haploinsufficiency in Saethre-Chotzen syndrome.(8,9)Gadd45 participates in cell differentiation, organogenesis, anti-proliferation, and apoptosis.(22) This further suggests that the observed cell death induced by Nell-1 is mediated through apoptosis. Our recent unpublished data suggested that Nell-1 protein induces phosphorylation of Jun N-terminal kinase (JNK), but not extracellular signal-regualted protein (ER) or p38, and therefore, Nell-1 may function through the JNK pathway.

Apoptosis is implicated in the regulation of bone formation both in vitro and in vivo. However, the exact role of apoptosis in the development of the osteoblast is not known. It was suggested that apoptosis in the late mineralization stage might actually facilitate mineralization. Reduced osteoblastic cell apoptosis in vitro and in vivo was found to increase mineralization and bone mass(32,33); therefore, the timing of apoptosis during osteoblast maturation may be crucial. If apoptosis occurs prematurely before differentiation, the population density will decrease, preventing bone formation, and the suture remains patent. On the other hand, if apoptosis occurs after differentiation and is part of an accelerated differentiation process, bone formation may increase, and calvarial overgrowth may occur. Nell-1's role, therefore, may depend on the stage of osteoblast maturation (Fig. 4). We hypothesize that in the early differentiation stage, the expression level of Nell-1 and its potential receptors will determine the final fate of the cell to continue through either differentiation or premature apoptosis. This also implies that Nell-1 may be involved in normal suture patency by eliminating selective osteoblastic cells that express high levels of Nell-1 or specific receptors.

thumbnail image

Figure FIG. 4.. Hypothetical model of Nell-1 in craniofacial development. Upregulation of Nell-1 induces osteoblast differentiation. This results in calvarial bone overgrowth and consequently premature suture closure. When Nell-1 is overexpressed at a higher level, Nell-1 induces premature/early osteoblast/osteogenic cell apoptosis and therefore reduces the number of osteoblast to form bone. Defect in calvarial bone formation results in craniofacial anomalies such as exencephaly. Dash lines represent potential modulation.

Download figure to PowerPoint

In conclusion, our results suggest that Nell-1 overexpression promotes osteoblastic cell apoptosis. The actual effect varies depending on the level of expression and the stage of cell differentiation. These effects may result in premature suture closure and other craniofacial developmental deformities.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES

We thank Dr Samson Cheng for editing the manuscript. This research is supported by the Wunderman Family Foundation, March of Dimes Birth Defect Foundation 6-FY02–163, National Institutes of Health/NIDCR RO3 DE 014649–01, and National Institutes of Health/NIDCR K23DE00422. We thank the Jane and Jerry Weintraub Center for Reconstructive Biotechnology, UCLA, for technical support.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. Acknowledgements
  8. REFERENCES
  • 1
    Cohen M 2000 Craniosynostosis. In: CohenM (ed.) Craniosynostosis Diagnosis Evaluation and Management, 2nd ed. Oxford, New York, NY, USA, P 112117.
  • 2
    Jabs EW, Muller U, Li X, Ma L, Luo W, Haworth IS, Klisak I, Sparkes R, Warman ML, Mulliken JB, Snead ML, Maxson R 1993 A mutation in the homeodomain of the human MSX2 gene in a family affected with autosomal dominant craniosynostosis. Cell 75:443450.
  • 3
    Liu YH, Tang Z, Kundu RK, Wu L, Luo W, Zhu D, Sangiorgi F, Snead ML, Maxson RE 1999 Msx2 gene dosage influences the number of proliferative osteogenic cells in growth centers of the developing murine skull: A possible mechanism for MSX2-mediated craniosynostosis in humans. Dev Biol 205:260274.
  • 4
    Furtwangler JA, Hall SH, Koskinen-Moffett LK 1985 Sutural morphogenesis in the mouse calvaria: The role of apoptosis. Acta Anat (Basel) 124:7480.
  • 5
    Mansukhani A, Bellosta P, Sahni M, Basilico C 2000 Signaling by fibroblast growth factors (FGF) and fibroblast growth factor receptor 2 (FGFR2)-activating mutations blocks mineralization and induces apoptosis in osteoblasts. J Cell Biol 149:12971308.
  • 6
    Rice DP, Kim HJ, Thesleff I 1999 Apoptosis in murine calvarial bone and suture development. Eur J Oral Sci 107:265275.
  • 7
    Opperman LA, Adab K, Gakunga PT 2000 Transforming growth factor-beta 2 and TGF-beta 3 regulate fetal rat cranial suture morphogenesis by regulating rates of cell proliferation and apoptosis. Dev Dyn 219:237247.
  • 8
    Lemonnier J, Haÿ E, Delannoy P, Fromigué O, Lomri A, Modrowski D, Marie PJ 2001 Increased osteoblast apoptosis in apert craniosynostosis: Role of protein kinase C and interleukin-1. Am J Pathol 158:18331842.
  • 9
    Yousfi M, Lasmoles F, Lomri A, Delannoy P, Marie PJ 2001 Increased bone formation and decreased osteocalcin expression induced by reduced Twist dosage in Saethre-Chotzen syndrome. J Clin Invest 107:11531161.
  • 10
    Watanabe TK, Katagiri T, Suzuki M, Shimizu F, Fujiwara T, Kanemoto N, Nakamura Y, Hirai Y, Maekawa H, Takahashi E 1996 Cloning and characterization of two novel human cDNAs (NELL1 and NELL2) encoding proteins with six EGF-like repeats. Genomics 38:273276.
  • 11
    Ting K, Vastardis H, Mulliken JB, Soo C, Tieu A, Do H, Kwong E, Bertolami CN, Kawamoto H, Kuroda S, Longaker MT 1999 Human NELL-1 expressed in unilateral coronal synostosis. J Bone Miner Res 14:8089.
  • 12
    Kuroda S, Oyasu M, Kawakami M, Kanayama N, Tanizawa K, Saito N, Abe T, Matsuhashi S, Ting K 1999 Biochemical characterization and expression analysis of neural thrombospondin-1-like proteins NELL1 and NELL2. Biochem Biophys Res Commun 265:7986.
  • 13
    Kuroda S, Tanizawa K 1999 Involvement of epidermal growth factor-like domain of NELL proteins in the novel protein-protein interaction with protein kinase C. Biochem Biophys Res Commun 265:752757.
  • 14
    Bornstein P 1995 Diversity of function is inherent in matricellular proteins: An appraisal of thrombospondin 1. J Cell Biol 130:503506.
  • 15
    Zhang X, Kuroda S, Carpenter D, Nishimura I, Soo C, Moats R, Iida K, Wisner E, Hu FY, Miao S, Beanes S, Dang C, Vastardis H, Longaker M, Tanizawa K, Kanayama N, Saito N, Ting K 2002 Craniosynostosis in transgenic mice overexpressing Nell-1. J Clin Invest 110:861870.
  • 16
    Ting K, Petropulos LA, Iwatsuki M, Nishimura I 1993 Altered cartilage phenotype expressed during intramembranous bone formation. J Bone Miner Res 8:13771387.
  • 17
    Simpson JF, Dutt PL, Page DL 1992 Expression of mitoses per thousand cells and cell density in breast carcinomas: A proposal. Hum Pathol 23:608611.
  • 18
    Sayah DN, Soo C, Shaw WW, Watson J, Messadi D, Longaker MT, Zhang X, Ting K 1999 Downregulation of apoptosis-related genes in keloid tissues. J Surg Res 87:209216.
  • 19
    Lomri A, Marie PJ 1990 Bone cell responsiveness to transforming growth factor beta, parathyroid hormone, and prostaglandin E2 in normal and postmenopausal osteoporotic women. J Bone Miner Res 5:11491155.
  • 20
    Fromigue O, Marie PJ, Lomri A 1998 Bone morphogenetic protein-2 and transforming growth factor-beta2 interact to modulate human bone marrow stromal cell proliferation and differentiation. J Cell Biochem 68:411426.
  • 21
    Kawakami A, Eguchi K, Matsuoka N, Tsuboi M, Koji T, Urayama S, Fujiyama K, Kiriyama T, Nakashima T, Nakane PK, Nagataki S 1997 Fas and Fas ligand interaction is necessary for human osteoblast apoptosis. J Bone Miner Res 12:16371646.
  • 22
    Price BD, Calderwood SK 1992 Gadd45 and Gadd153 messenger RNA levels are increased during hypoxia and after exposure of cells to agents which elevate the levels of the glucose-regulated proteins. Cancer Res 52:38143817.
  • 23
    Tsuboi M, Kawakami A, Nakashima T, Matsuoka N, Urayama S, Kawabe Y, Fujiyama K, Kiriyama T, Aoyagi T, Maeda K, Eguchi K 1999 Tumor necrosis factor-alpha and interleukin-1beta increase the Fas-mediated apoptosis of human osteoblasts. J Lab Clin Med 134:222231.
  • 24
    Hay E, Lemonnier J, Fromigue O, Marie PJ 2001 Bone morphogenetic protein-2 promotes osteoblast apoptosis through a Smad-independent, protein kinase C-dependent signaling pathway. J Biol Chem 276:2902829036.
  • 25
    Nasatzky E, Grinfeld D, Boyan BD, Dean DD, Ornoy A, Schwartz Z 1999 Transforming growth factor-beta1 modulates chondrocyte responsiveness to 17beta-estradiol. Endocrine 11:241249.
  • 26
    Opperman LA 2000 Cranial sutures as intramembranous bone growth sites. Dev Dyn 219:472485.
  • 27
    Marie P 2000 Significance of growth factors in bone repair. Growth factors and bone tissue. Rev Chir Orthop Reparatrice Appar Mot 86(Suppl 1):150151.
  • 28
    Lynch MP, Capparelli C, Stein JL, Stein GS, Lian JB 1998 Apoptosis during bone-like tissue development in vitro. J Cell Biochem 68:3149.
  • 29
    Garcia Abreu J, Coffinier C, Larrain J, Oelgeschlager M, De Robertis EM 2002 Chordin-like CR domains and the regulation of evolutionarily conserved extracellular signaling systems. Gene 287:3947.
  • 30
    Crawford SE, Stellmach V, Murphy-Ullrich JE, Ribeiro SM, Lawler J, Hynes RO, Boivin GP, Bouck N 1998 Thrombospondin-1 is a major activator of TGF-beta1 in vivo. Cell 93:11591170.
  • 31
    Piccolo S, Agius E, Lu B, Goodman S, Dale L, De Robertis EM 1997 Cleavage of Chordin by Xolloid metalloprotease suggests a role for proteolytic processing in the regulation of Spemann organizer activity. Cell 91:407416.
  • 32
    Jilka RL, Weinstein RS, Bellido T, Roberson P, Parfitt AM, Manolagas SC 1999 Increased bone formation by prevention of osteoblast apoptosis with parathyroid hormone. J Clin Invest 104:439446.
  • 33
    Plotkin LI, Weinstein RS, Parfitt AM, Roberson PK, Manolagas SC, Bellido T 1999 Prevention of osteocyte and osteoblast apoptosis by bisphosphonates and calcitonin. J Clin Invest 104:13631374.