3D cell-laden scaffold printed with brain acellular matrix bioink

Abstract

Background: Intracerebral hemorrhage (ICH) is a severe neurological disorder characterized by bleeding within the brain tissue, typically associated with factors such as hypertension, cerebrovascular disease, and trauma. The transplantation of human umbilical cord-derived mesenchymal stem cells (hUCMSCs) has demonstrated promising effects in restoring neurological function in ICH rats; however, limited retention of these cells significantly impedes their efficacy. To address this limitation, we developed a bioink composed of decellularized extracellular matrix (dECM) and hUCMSCs, which was synthesized into 3D cell-laden scaffold through 3D bioprinting. This approach aims to extend the retention of hUCMSCs and create an early vascular microenvironment, thereby partially compensating for the drawbacks of hUCMSC transplantation and improving neurological function in ICH rats.

Methods: This study aimed to explore the use of a bioink formed by mixing 15% gelatin and 3% sodium alginate with a dECM solution, in conjunction with hUCMSCs, for 3D bioprinting of 3D cell-laden scaffold. The viscosity, morphology, and biocompatibility of the bioink were characterized using rheological analysis, scanning electron microscopy (SEM), and hematoxylin and eosin (HE) staining. Following printing, a live/dead assay kit was employed to assess the viability of hUCMSCs within the 3D cell-laden scaffold. ICH model rats were randomly assigned to four groups: (1) SHAM group; (2) ICH group; (3) ICH + 3D biological scaffold group; and (4) ICH + 3D cell-laden scaffold group.

Results: hUCMSCs exhibited a higher retention rate within the 3D bioprinted 3D cell-laden scaffold. HE staining, immunohistochemistry, and immunofluorescence results indicated that the 3D biological scaffold encapsulating hUCMSCs had a significant impact on the vascularization of the printed 3D cell-laden scaffold. Furthermore, 3D cell-laden scaffold improved nerve function and promoted angiogenesis in rats with cerebral hemorrhage better than 3D biological scaffolds.

Conclusion: Our results suggest that 3D bioprinted 3D cell-laden scaffold hold great potential for restoring impaired neurological function in ICH rats.

Supplementary Information: The online version contains supplementary material available at 10.1186/s12951-025-03644-z.

Keywords: 3D bioprinting; 3D cell-laden scaffold; Angiogenesis; Decellularized extracellular matrix; Human umbilical cord-derived mesenchymal stem cells.

Scheme 1

Preparation of 3D cell-laden scaffold. (A) Sprague-Dawley (SD) Rats. (B) Brain Tissue. (C) Decellularized Brain Matrix. (D) Light-Crosslinkable Hydrogel. (E) Sodium Alginate. (F) Bioink. (G) Human Umbilical Cord Tissue. (H) Human Umbilical Cord Mesenchymal Stem Cells (hUCMSCs). (I) Incorporation of hUCMSCs into the Bioink. (J) 3D Bioprinter. (K) 3D Bioprinting. (L) 3D cell-laden scaffold

Fig. 1

Preparation and characterization of 3D cell-laden scaffold. (A) Images of decellularized brain matrix and hydrogel preparation. (B) Quantitative assessment of DNA content (n = 4). (C) Quantification of collagen retention rate (n = 4). (D) Hematoxylin and eosin (HE) staining and Masson staining demonstrating decellularization, with Sirius Red staining confirming collagen fiber retention. (E) Scanning electron microscopy (SEM) analysis of ECM, dECM, hydrogel, sodium alginate, 3D biological scaffold, and 3D cell-laden scaffold. Scale bars: D 200 μm/100 µm, E 100 μm/3 µm. Data were expressed as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 2

Preparation and characterization of 3D cell-laden scaffold. (A) FTIR analysis of the 3D biological scaffold bioink, showing a peak at 1537 cm⁻¹ corresponding to the NH₂ group, and C = O stretching detected at 1633 cm⁻¹. (B) Variation in viscosity of the bioink as the shear rate increases from 1 s⁻¹ to 100 s⁻¹. (C) Rheological properties measured to evaluate the mechanical performance of decellularized brain matrix (dECM) bioink. (D) Rheological properties measured to assess the mechanical performance of the 3D biological scaffold bioink. (E) Degradation rate of 3D biological scaffold and 3D cell-laden scaffold (n = 4). (F) The 3D biological scaffold exhibits injectability. (G) 3D bioprinter. (H) Preparation of 3D cell-laden scaffold. (I) In situ transplantation of 3D cell-laden scaffold. Data were expressed as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 3

Evaluation of 3D cell-laden scaffold. (A) In vivo biodistribution and degradation of luciferase-labeled 3D cell-laden scaffold. (B) Optical analysis of hUCMSCs. (C) Flow cytometry analysis showed hUCMSCs express CD73, CD90, and CD105 but not CD34 and CD45. (D) 1–14 days to evaluat viability of hUCMSCs within the 3D cell-laden scaffold using an in vitro live/dead assay. hUCMSCs were derived from passage 3 (P3), exhibiting green fluorescence (live cells with calcein AM) and red fluorescence (dead cells with EthD-1) as observed under a laser confocal microscope. (E) Quantitative analysis of the cell viabilities (n = 3). Scale bars: B, D 100 μm. Data were expressed as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 4

Evaluation of 3D cell-laden scaffold. (A) In vitro tube formation assay demonstrating the influence of hUCMSCs on vascular reconstruction within the 3D cell-laden scaffold. (B, C) Evaluate the branche and capollary length of the reconstructed vessels (n = 4). (D, E) Cerebral hemodynamic parameters, including regional cerebral blood flow (rCBF) and vascular permeability, were quantitatively evaluated peri-hematomally using laser speckle contrast imaging (LSCI) and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) at postoperative days 1, 4, 7, and 14. Scale bars: A 200 μm, D 100 μm. Data were expressed as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 5

In vivo evaluation of 3D cell-laden scaffold. (A) Representative images of hemorrhagic lesions in rats from each group, stained with hematoxylin and eosin (HE). (B) Transmission electron microscopy (TEM) analysis of rats from each group, comparing mitochondrial cristae loss, outer membrane rupture, and vascular dilation (indicated by white arrows). (C) Volume of intracerebral hemorrhage in rats from each group (n = 4). (D) Nissl staining of brain tissues from rats in each group (n = 4). (E) Assessment of brain water content in rats from each group to evaluate the impact of treatment on cerebral edema (n = 4). Scale bars: A 400 μm/100 µm, B 4 nm/1 nm. Data were expressed as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 6

In vivo evaluation of 3D cell-laden scaffold. (A) Corner test conducted from days 1 to 14 post-intracerebral hemorrhage (ICH) (n = 12). (B) Forelimb placement test results (n = 12). (C) Bederson score and (D) Longa score were utilized to assess neurological recovery (n = 12). (E, F) Statistical analysis of VEGF and BDNF mRNA (n = 4). Data were expressed as mean ± standard deviation. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 7

Differentiation of CM-Dil-labeled hUCMSCs within 3D cell-laden scaffold. (A, C) Viability of hUCMSCs in 3D cell-laden scaffold at week 4 post-transplantation. (B, D) Expression of CK14 and CD31 within the 3D cell-laden scaffold constructed with encapsulated hUCMSCs, assessed at 4 weeks post-transplantation. Scale bars: A-D 200 μm