A microfluidic-based approach to investigate the inflammatory response of macrophages to pristine and drug-loaded nanostructured hydroxyapatite

Abstract

The in vitro biological characterization of biomaterials is largely based on static cell cultures. However, for highly reactive biomaterials such as calcium-deficient hydroxyapatite (CDHA), this static environment has limitations. Drastic alterations in the ionic composition of the cell culture medium can negatively affect cell behavior, which can lead to misleading results or data that is difficult to interpret. This challenge could be addressed by a microfluidics-based approach (i.e. on-chip), which offers the opportunity to provide a continuous flow of cell culture medium and a potentially more physiologically relevant microenvironment. The aim of this work was to explore microfluidic technology for its potential to characterize CDHA, particularly in the context of inflammation. Two different CDHA substrates (chemically identical, but varying in microstructure) were integrated on-chip and subsequently evaluated. We demonstrated that the on-chip environment can avoid drastic ionic alterations and increase protein sorption, which was reflected in cell studies with RAW 264.7 macrophages. The cells grown on-chip showed a high cell viability and enhanced proliferation compared to cells maintained under static conditions. Whereas no clear differences in the secretion of tumor necrosis factor alpha (TNF-α) were found, variations in cell morphology suggested a more anti-inflammatory environment on-chip. In the second part of this study, the CDHA substrates were loaded with the drug Trolox. We showed that it is possible to characterize drug release on-chip and moreover demonstrated that Trolox affects the TNF-α secretion and morphology of RAW 264.7 ​cells. Overall, these results highlight the potential of microfluidics to evaluate (bioactive) biomaterials, both in pristine form and when drug-loaded. This is of particular interest for the latter case, as it allows the biological characterization and assessment of drug release to take place under the same dynamic in vitro environment.

Keywords: Biomaterial; Calcium phosphate cement; Drug release; In vitro; Macrophage; On-chip.

Graphical abstract

Fig. 1

(A) Schematic of the individual layers of the CDHA-on-chip and (B) the top view of the assembled chip.

Fig. 2

(A) Cross-section of disc-F, disc-C, chip-F, and chip-C samples, visualized using fluorescent confocal microscopy. Sorption of albumin-FITC solution is depicted in green. The arrow indicates the direction from which the CDHA substrates were exposed to albumin-FITC. Scale bars correspond to 250 ​μm. (B) Quantification of the albumin-FITC penetration depth. ∗ indicates p ​< ​0.05 between the samples.

Fig. 3

Quantification of (A) calcium and (B) phosphate concentrations in the cell culture medium of disc-F, disc-C, chip-F and chip-C over a period of 4 days. As a control, fresh medium was included. ∗ indicates p ​< ​0.05 between the samples and fresh medium; # indicates p ​< ​0.05 between the chip and respective disc samples (e.g. difference between chip-F and disc-F); ¤ indicates p ​< ​0.05 between F and C substrates of the same type (e.g. difference between chip-F and chip-C).

Fig. 4

Cumulative release profile of Trolox from (A) disc-F and disc-C samples and (B) chip-F and chip-C samples. ∗ indicates p ​< ​0.05 between the samples.

Fig. 5

(A) RAW 264.7 ​cell viability determined by LIVE (green)/DEAD (red) staining after 1 and 3 days of culture on the disc-F, disc-C, chip-F, chip-C and PS samples. Scale bar corresponds to 100 ​μm. (B) Cell proliferation quantified with the LDH assay. The absorbance values were normalized to the surface area (i.e. 0.385 ​cm² for the static samples and 0.975 ​cm² on-chip)∗ indicates p ​< ​0.05 between the different CDHA substrates and PS control; # indicates p ​< ​0.05 between the chip and respective disc samples (e.g. difference between chip-F and disc-F); No statistically significant differences (p ​> ​0.05) were observed between F and C substrates of the same type (e.g. difference between chip-F and chip-C).

Fig. 6

TNF-α secretion of RAW 264.7 ​cells grown on pristine (blue bars) and Trolox-loaded (yellow bars) CDHA substrates after 24 ​h of flow (or culture under static conditions). ∗ indicates p ​< ​0.05 between pristine and Trolox-loaded substrates; - indicates p ​< ​0.05 between CDHA substrates and PS; + indicates p ​< ​0.05 between CDHA substrates and PS+. No statistically significant differences (p ​> ​0.05) were observed between the chip and respective disc samples (e.g. difference between chip-F and disc-F) or between F and C substrates of the same type (e.g. difference between chip-F and chip-C).

Fig. 7

Morphology of RAW 264.7 ​cells on the different pristine and Trolox-loaded CDHA substrates after 1 and 3 days of culture. Scale bar corresponds to 5 ​μm.