Two-Photon Polymerization of 2.5D and 3D Microstructures Fostering a Ramified Resting Phenotype in Primary Microglia

Abstract

Microglia are the resident macrophages of the central nervous system and contribute to maintaining brain's homeostasis. Current 2D "petri-dish" in vitro cell culturing platforms employed for microglia, are unrepresentative of the softness or topography of native brain tissue. This often contributes to changes in microglial morphology, exhibiting an amoeboid phenotype that considerably differs from the homeostatic ramified phenotype in healthy brain tissue. To overcome this problem, multi-scale engineered polymeric microenvironments are developed and tested for the first time with primary microglia derived from adult rhesus macaques. In particular, biomimetic 2.5D micro- and nano-pillar arrays (diameters = 0.29-1.06 µm), featuring low effective shear moduli (0.25-14.63 MPa), and 3D micro-cages (volume = 24 × 24 × 24 to 49 × 49 × 49 μm³) with and without micro- and nano-pillar decorations (pillar diameters = 0.24-1 µm) were fabricated using two-photon polymerization (2PP). Compared to microglia cultured on flat substrates, cells growing on the pillar arrays exhibit an increased expression of the ramified phenotype and a higher number of primary branches per ramified cell. The interaction between the cells and the micro-pillar-decorated cages enables a more homogenous 3D cell colonization compared to the undecorated ones. The results pave the way for the development of improved primary microglia in vitro models to study these cells in both healthy and diseased conditions.

Keywords: 3D scaffold; effective shear modulus; micro-pillars; nano-pillars; primary microglia; two-photon polymerization.

FIGURE 1

Schematic representation of the fabrication process of the structures. (A) Oxygen plasma cleaning of the substrate. (B) Spin coating of the adhesion promoter Ormoprime^® 08. (C) 2PP printing of the structures. (D) Chemical development of the structure. (E) Oxygen plasma functionalization of the polymer surface.

FIGURE 2

SEM images (at a 45° angle) of (A) the pedestal; (B) micro-pillar array (MP1); (C) nano-pillar array (NP1); (D) nano-pillar array (NP2). The printing of a supporting pedestal for the nano-pillars prevented the collapse of pillars after development. (E) Small cage (SC). (F) Big cage (BC). (G) Small cage with micro-pillar decoration (SC-MP). (H) Big cage with micro-pillar decoration (BC-MP). (I) Big cage with nano-pillar decoration (BC-NP).

FIGURE 3

Sequential printing technique of the BC-NP, using five steps. (1) Base printed. (2) Pillars printed on the base. (3) Vertical beam with pillars printed. (4) Top printed. (5) Pillars printed on the top. CAD model generated by Describe (Nanoscribe GmbH).

FIGURE 4

(A) The phenotype distribution of microglia cultured on laminin-coated fused silica substrates (Lam sub) and uncoated substrates (Unc sub) (n_samples = 11 and n_cells > 1,000 for the uncoated substrates, n_samples = 9 and n_cells > 1,000 for the laminin-coated substrates). (B) Percentage of ramified cells on laminin-coated versus uncoated substrates. (C) Number of primary branches per ramified cell for ramified microglia cultured on laminin-coated versus uncoated substrates (n_samples = 9 and n_{ramified_cells} > 30 for the uncoated substrates, n_samples = 9 and n_{ramified_cells} > 30 for the laminin-coated substrates).

FIGURE 5

Representative confocal microscopy images of primary microglia (maximum projection) cultured on (A) a fused silica substrate; (B) micro-pillar arrays. Blue is Hoechst 33342 staining (nucleus) and IP-Dip micro-pillars. Red is CX3CR1 (cell membrane). (C) Phenotype distribution of primary microglia cultured on flat fused silica substrates (Flat sub), micro-pillar (MP), and nano-pillar (NP) arrays (n_samples = 4 and n_cells > 300 for the substrates, n_samples = 4 and n_cells > 300 for the micro-pillar arrays, n_samples = 3 and n_cells > 300 for the nano-pillar arrays). (D) Percentage of ramified cells. (E) Sholl analysis of ramified cells showing the complexity of the cells by indicating the number of intersections the cells have with the concentric contours. (F) Area under the curve (AUC) calculated from the Sholl plot. Larger AUC corresponds to a more complex ramified cell. (G) Number of primary branches per ramified cell (n_samples = 3 and n_{ramified_cells} = 15 for the flat substrates, n_samples = 3 and n_{ramified_cells} = 15 for the micro-pillar arrays, n_samples = 3 and n_{ramified_cells} = 13 for the nano-pillar arrays).

FIGURE 6

SEM images of primary microglia on: (A) a flat fused silica substrate; (B,C) micro-pillar arrays; (D,E) NP1 nano-pillar arrays (pillars printed directly on the substrate); (F,G) NP2 nano-pillar arrays (pillars printed on a pedestal). Ramified microglia were noticed on both NP patterns, but pillars of NP2 had higher structural integrity compared to those on NP1 due to the support of the pedestal. Interaction of microglial filopodia with (H) the flat substrate; (I) nano-pillar arrays (NP2). (C,E) images are acquired at a 45° angle.

FIGURE 7

SEM images (taken at a 45° angle) of primary microglia cultured on (A) SC; (B) SC-MP; (C) BC; (D) BC-MP; (E) and (F) BC-NP. An amoeboid-like morphology of microglia was observed when cultured on SC and SC-MP. Cells seemed to attempt to enwrap the cages. In presence of the bigger cages, multiple phenotypes were noticed. Detachment of the cages was observed especially for the smaller ones.

FIGURE 8

SEM images showing the interaction of primary microglia and their filopodia with the micro- and nanometric features of the BC (A–C), BC-MP (D–F), and BC-NP (G–I). (C) Filopodia guided by the nano-grooves formed by the hatching lines in the BC. (F) Micro-pillars acting as a competing guidance cue to the filopodia. (I) Nano-pillars providing a mechanical cue and a platform of growth for the cells.

FIGURE 9

(A) 3D reconstruction of a fluorescence z-stack obtained by confocal microscopy of BC-MP cages with microglia. Blue is Hoechst 33342 staining (nucleus) and red is CX3CR1 (cell membrane). (B) Volumetric occupancy of microglia in laminin-coated (blue) and uncoated (red) cages. Laminin coating resulted in doubling the occupancy of cages as compared to uncoated ones for decorated and undecorated cages alike. (C) Microglia distribution along the height of the BC, BC-MP, and BC-NP for laminin-coated and uncoated samples. Microglia were shown to colonize the mid and top sections of decorated structures compared to the undecorated ones (for each structure type n_samples = 2, n_cages > 25, n_cells > 100).