Physiological Mineralization during In Vitro Osteogenesis in a Biomimetic Spheroid Culture Model

Abstract

Bone health-targeting drug development strategies still largely rely on inferior 2D in vitro screenings. We aimed at developing a scaffold-free progenitor cell-based 3D biomineralization model for more physiological high-throughput screenings. MC3T3-E1 pre-osteoblasts were cultured in α-MEM with 10% FCS, at 37 °C and 5% CO₂ for up to 28 days, in non-adherent V-shaped plates to form uniformly sized 3D spheroids. Osteogenic differentiation was induced by 10 mM β-glycerophosphate and 50 µg/mL ascorbic acid. Mineralization stages were assessed through studying expression of marker genes, alkaline phosphatase activity, and calcium deposition by histochemistry. Mineralization quality was evaluated by Fourier transformed infrared (FTIR) and scanning electron microscopic (SEM) analyses and quantified by micro-CT analyses. Expression profiles of selected early- and late-stage osteoblast differentiation markers indicated a well-developed 3D biomineralization process with strongly upregulated Col1a1, Bglap and Alpl mRNA levels and type I collagen- and osteocalcin-positive immunohistochemistry (IHC). A dynamic biomineralization process with increasing mineral densities was observed during the second half of the culture period. SEM-Energy-Dispersive X-ray analyses (EDX) and FTIR ultimately confirmed a native bone-like hydroxyapatite mineral deposition ex vivo. We thus established a robust and versatile biomimetic, and high-throughput compatible, cost-efficient spheroid culture model with a native bone-like mineralization for improved pharmacological ex vivo screenings.

Keywords: Fourier-transform infrared spectroscopy (FT-IR); MC3T3-E1; osteoblast progenitors; osteogenic differentiation; physiological biomineralization; spheroid culture; volumetric micro-CT quantification.

Figure 1

General microscopic appearance of spheroids, next to collagen content and cell death. (A) Hematoxylin–Eosin (H.E.) overview of staining, showing nuclei in violet blue and cytoplasm and ECM in shades of pink. (B) Sirius Red staining of collagens. (C) TUNEL/Ki67 co-staining to detect apoptotic cell death and proliferation. Representative stainings are shown (n = 3), scale bars represent 200 µm. (D) 400× g magnifications of the rim region of HE-stained differentiating spheroids at day 28 with polarized light micrograph of Sirius Red staining underneath. Scale bars are 50 µm. (E) Representative SEM images of the outer spheroid layer, cultured for 28 days in differentiation medium. (F) Spheroid volume determination by micro-CT. Significant differences were calculated by 2-way ANOVA (n = 6) followed by Tukey’s post hoc test of BCY-transformed data. (G) Total DNA analyses of the spheroids using BCY-transformed data with two-way ANOVA (n = 6), followed by Tukey’s post hoc test, next to tabular overview. ^# p < 0.05, ^## p < 0.01 as indicated; ^• p < 0.05, ^•• p < 0.01, ^••• p < 0.001 vs. starting day. Data represent geometric mean ± 95% CI.

Figure 2

Expression of osteogenic markers. (A–D) Relative mRNA expression of selected differentiation markers during the first 14 days of culture, with tabular summary of analyses underneath. Shown are Runx2 (A), Bglap (osteocalcin) (B), Spp1 (osteopontin) (C), and Col1a1 (D). Significant differences were calculated by 2-way ANOVA (n = 5), followed by Tukey post hoc test. (E–H) Corresponding immunohistology. IHC staining against Runx2 (E), osteocalcin (F), osteopontin (G) and type I collagen (H). Representative staining from at least three spheroids (n = 3) per condition; scale bars as indicated. (I) Changes in Alpl mRNA abundance until day 14 of culture and (J) respective enzymatic activity over 28 days. Significant differences were calculated by 2-way ANOVA (n = 5) of BCY-transformed data followed by Tukey’s post hoc test (I) and Kruskal–Wallis (n = 6) followed by Dunn’s post hoc test (J). ^# p < 0.05 ^### p < 0.001 as indicated; * p < 0.05, ** p < 0.01, *** p < 0.001, ^••• p < 0.001 vs. starting day. Data represent geometric mean ± 95% CI.

Figure 3

Qualitative and quantitative evaluation of calcium deposition within spheroids. (A) Alizarin Red staining over a culture period of up to 28 days. Note that Alizarin sulfonic acid stains calcium deposits deep brick orange-red. (B) Corresponding von Kossa staining, showing calcifications in greyish-black. Note that calcium deposition is not observed in non-differentiating controls. Representative images from sectioning three randomly selected spheroids (n = 3) per condition. Scale bars are 200 µm. (C) Spectrophotometric quantification of the extracted precipitated calcium at 405 nm. Significant differences were evaluated using a two-way ANOVA of the Johnson transformed data (n = 6, mineralizing groups; n = 5, control groups), followed by Tukey’s post hoc test. ^### p < 0.001 as indicated; *** p < 0.001 vs. starting day. Data represent geometric mean ± 95% CI.

Figure 4

Volumetric quantification of biomineralization in spheroids by micro-CT. (A) On top, original micro-CT image; below, pseudo-colored image after segmentation by voxel intensity. Representative 2D (B) and 3D (C) reconstruction with corresponding cross-section (D) showing the mineralized core of a spheroid cultured for 28 d under osteogenic conditions. Note the central trabecular-like mineralization pattern surrounded by a ring-shaped mineralization. (E–H) Micro-CT-based volumetric quantification of biomineralization, with arbitrarily chosen pixel intensity cut-off values, discriminating moderate (E,F) and strong mineralization (G,H) by objective differences in the central calcium deposit volume. Significant differences were evaluated using a one-way ANOVA (n = 6, mineralizing groups; n = 5 control groups), followed by Tukey’s post hoc test. (I) Changes in mean voxel intensities over time. Statistical differences were calculated by Kruskal–Wallis method followed by Dunn’s post hoc test. ^### p < 0.001 as indicated; * p < 0.05, ** p < 0.001 vs. starting day. Data represent geometric mean ± 95% CI.

Figure 5

Native bone-like biomineralization in 3D cultures. (A–C) Representative SEM/EDX analysis of a spheroid cultured for 28 days in differentiation medium, with cross-section (A), magnification (B), and EDX spectrum (C). (D–F) Representative corresponding SEM and EDX analyses of a spheroid cultured under non-mineralizing conditions. (G) Representative FTIR absorption spectra of a murine cortical bone specimen (i.e., femur, red), and spheroids cultured for 28 days under either mineralizing (blue) or non-mineralizing culture conditions (green). a.u., arbitrary units; arrow marks a collagen fiber; “+” marks a cell.

Figure 6

Different responses to cWnt agonist in monolayer and spheroid cultures. (A–E) Effect of LiCl on MC3T3-E1 cells cultured in 2D: determination of alkaline phosphatase (ALP) activity after 14 days of culture without (-) or with (+) 5 mM LiCl (A). Representative Sirius Red stainings after 21 days with macroscopic photographs of wells (top) and respective micrographs (bottom) (B) and corresponding spectrophotometric quantification of total collagen content upon dye extraction (C). Corresponding Alizarin Red staining (D) with quantification of calcium deposition (E). Scale bars show 100 µm. (F–K) Comparison of lithium-mediated effects in 3D-cultured MC3T3-E1 cells. ALP activity after 14 days (F), representative immuno-/histochemically stained sections (i.e., Alizarin Red, Von Kossa, type I collagen, osteocalcin, from top to bottom) after 21 days. Quantified extracellular calcium content in spheroids after 21 days (H) and corresponding micro-CT analyses: mean voxel intensity (I), volumetric quantification of moderate (J) and strong mineralization levels (K). For details see legend to Figure 3. Scale bars are 200 µm. Significant differences were evaluated using Student’s t-test (n = 6). ^# p < 0.05, ^## p < 0.01, ^### p < 0.001 as indicated. Data represent geometric mean ± 95% CI.