Similarities and differences between porcine mandibular and limb bone marrow mesenchymal stem cells

Abstract

Objective: Research has shown promise of using bone marrow mesenchymal stem cells (BMSCs) for craniofacial bone regeneration; yet little is known about the differences of BMSCs from limb and craniofacial bones. This study compared pig mandibular and tibia BMSCs for their in vitro proliferation, osteogenic differentiation properties and gene expression.

Design: Bone marrow was aspirated from the tibia and mandible of 3-4 month-old pigs (n=4), followed by BMSC isolation, culture-expansion and characterization by flow cytometry. Proliferation rates were assessed using population doubling times. Osteogenic differentiation was evaluated by alkaline phosphatase activity. Affymetrix porcine microarray was used to compare gene expressions of tibial and mandibular BMSCs, followed by real-time RT-PCR evaluation of certain genes.

Results: Our results showed that BMSCs from both locations expressed MSC markers but not hematopoietic markers. The proliferation and osteogenic differentiation potential of mandibular BMSCs were significantly stronger than those of tibial BMSCs. Microarray analysis identified 404 highly abundant genes, out of which 334 genes were matched between the two locations and annotated into the same functional groups including osteogenesis and angiogenesis, while 70 genes were mismatched and annotated into different functional groups. In addition, 48 genes were differentially expressed by at least 1.5-fold difference between the two locations, including higher expression of cranial neural crest-related gene BMP-4 in mandibular BMSCs, which was confirmed by real-time RT-PCR.

Conclusions: Altogether, these data indicate that despite strong similarities in gene expression between mandibular and tibial BMSCs, mandibular BMSCs express some genes differently than tibial BMSCs and have a phenotypic profile that may make them advantageous for craniofacial bone regeneration.

Keywords: BMP-4; Craniofacial bone regeneration; DNA microarray; Marrow stromal cells; Osteogenesis.

Figure 1. Bone marrow aspiration from the pig mandibular symphyseal region

(A) A representative CT image in the sagittal view showing a large marrow space and porous low-density bones inside of the bulged symphyseal process. (B) Histological section stained with H&E showing abundant trabecular bone inside of the bulged symphyseal process. (C) With the pig anesthetized and the symphyseal area disinfected, the symphyseal marrow space was accessed and bone marrow was gradually aspirated into syringes prefilled with heparin. (D) Volumetric surface reconstruction of clinical CT images showing an optimal site for inserting the aspiration needle (arrow), which was 1 cm away from the symphyseal midline and 1 cm above the inferior symphyseal border. The boxes in (A) and (B) and the arrow in (D) indicate the approximate site where bone marrow was aspirated.

Figure 2. Confirmation of MSC identity

(A–B) Brightfield microscopy image of cultured Passage-0 (P0) mandibular BMSCs at day 5 and 10, respectively. (C–D) Brightfield microscopy image of cultured P0 tibial BMSCs at day 5 and 10, respectively. Note P0 mandibular BMSCs tended to be fewer than P0 tibial BMSCs at the beginning but quickly proliferated to comparable levels. (E–F) Flow cytometry analysis showed both mandibular (E) and tibial BMSC (F) were positive for MSC markers CD105, CD90, and CD44, but negative for hematopoietic stem cell markers CD11b and CD45. Scale bar, 200 µm.

Figure 3. Proliferation during in vitro expansion (Passages 1–4)

(A) Mandibular BMSCs demonstrated a higher cell proliferation rate than tibial BMSCs. (B) The mean cell doubling time of mandibular BMSCs was significantly shorter than that of tibial BMSCs (*, p<0.01). Error bars, standard deviations.

Figure 4. Osteogenic differentiation capacity

(A–F) Representative images of ALP staining at 10-day culture of mandibular (A–C) and tibial BMSC (D–F) in regular growth media supplemented with 0 ng (A, D), 5 ng (B, E) and 10 ng (C, F) FGF-2. (G) ALP activity assay at 5- and 10-day cultures of mandibular and tibia BMSCs in the absence or presence of FGF-2. Note the mean ALP activity was higher in mandibular BMSCs than in tibial BMSCs regardless of the FGF-2 concentration. Scale bar, 2 mm. Error bars, standard deviations.

Figure 5. Gene expression reflected by microarray assay

(A) Scatter plot of microarray data overall showed strong correlation (r=0.98) in gene expression between mandibular and tibial BMSCs. Genes that were differentially expressed between mandibular and tibial BMSCs and genes that were highly expressed are indicated by colored dots. (B) Hierarchical clustering showed that mandibular (label 1; yellow) and tibial (label 2; green) BMSCs (both at passage 0) belonged to distinctive clusters when 1.5-fold of difference in gene expression was set as the cut-off threshold. (C) Volcano plot of the microarray data showed the distribution of differentially expressed genes between mandibular and tibial BMSCs (ANOVA tests, p<0.05). Genes showed above 1.5-fold of difference between mandibular and tibial MSCs are indicated by colored dots. Note neural crest-related genes BMP4 and Nestin were upregulated in mandibular BMSCs. (D) Real-time RT-PCR confirmed stronger BMP4 and Nestin expression in mandibular BMSCs, showing a mandibular BMSCs:tibial BMSCs expression ratio greater than 1 (broken line) . Error bars, standard deviations.