A Biofabrication Strategy for a Custom-Shaped, Non-Synthetic Bone Graft Precursor with a Prevascularized Tissue Shell

Abstract

Critical-sized defects of irregular bones requiring bone grafting, such as in craniofacial reconstruction, are particularly challenging to repair. With bone-grafting procedures growing in number annually, there is a reciprocal growing interest in bone graft substitutes to meet the demand. Autogenous osteo(myo)cutaneous grafts harvested from a secondary surgical site are the gold standard for reconstruction but are associated with donor-site morbidity and are in limited supply. We developed a bone graft strategy for irregular bone-involved reconstruction that is customizable to defect geometry and patient anatomy, is free of synthetic materials, is cellularized, and has an outer pre-vascularized tissue layer to enhance engraftment and promote osteogenesis. The graft, comprised of bioprinted human-derived demineralized bone matrix blended with native matrix proteins containing human mesenchymal stromal cells and encased in a simple tissue shell containing isolated, human adipose microvessels, ossifies when implanted in rats. Ossification follows robust vascularization within and around the graft, including the formation of a vascular leash, and develops mechanical strength. These results demonstrate an early feasibility animal study of a biofabrication strategy to manufacture a 3D printed patient-matched, osteoconductive, tissue-banked, bone graft without synthetic materials for use in craniofacial reconstruction. The bone fabrication workflow is designed to be performed within the hospital near the Point of Care.

Keywords: 3D bioprinting; bone; craniofacial; ossification; patient-specific; reconstruction; tissue fabrication; vascularization.

FIGURE 1

Fabrication of graft core and vascularized tissue shell. (A) Schematic of the graft fabrication workflow resulting in a printed graft surrounded by a vascularized, tissue shell. (B) Fabrication of the graft starting with (left to right) the digital TSIM® model and print path used in the 3D printing of the graft core, the printed core as a latticed, 1 cm × 1 cm x 3 cm cuboid, formation of the tissue shell and conditioning the graft, and the complete, vascularized graft ready for implantation. MSC = mesenchymal stromal cell, DBM = demineralized bone matrix, ECM = extracellular matrix.

FIGURE 2

Implants are well integrated and develop vascular leashes. (A) Exterior view of implanted graft at 4 weeks prior to explant (incision scar is located bottom right, graft 684). (B) An exposed implant at 4 weeks (graft 681) showing an intact artery/vein pair (open arrow) arising from the flank musculature that persisted from into week 8 (see panel E). (C) Side view of 8 weeks graft following partial dissection showing thinning (compression) of the graft (graft 688). (D,E) Dermal-side and muscle-side views, respectively, of 8 weeks implants a secondary vein (closed arrow) that developed from the subdermal tissue and the first vascular leash (open arrow, graft 677). (F) Schematic of the vascular leashes associated with the graft over time that arose from the muscle tissue (at week 4) and the subdermis (at week 8).

FIGURE 3

Implants are well vascularized throughout the core. Projected, stitched confocal image stacks of en face preparations of explanted grafts at weeks 4 (graft 681) and 8 (graft 687). Bottom row: Select regions from the first panel at higher magnification highlighting areas involving perfused human and recipient (rat) vessels. Dextran was injected into the recipient circulation as a blood tracer in week 4 groups. Fluorescent UEA-1 lectin was used to stain the human microvasculature while fluorescent GSL-1 lectin was used to stain the rodent vasculature. DBM = demineralized bone matrix.

FIGURE 4

Dynamic remodeling and mineralization of the graft tissue environment. Comparison of representative histology of graft tissue prior to implant (after 3 weeks of in vitro culture), at 4-week explant, and at 8-week explant showing tissue organization (H&E), collagen composition and structure (Masson’s trichrome and picrosirius red), and mineralization (alizarin red, AR). Red blood cells (arrows) are visible in vessel profiles in the 4- and 8-week samples. Scale bar for all images equals 100 µm. 4 and 6 week respective graft numbers: H&E 682, 677; PR/FG 681, 678; Masson’s 681, 677; AR 681, 678.

FIGURE 5

Implanted grafts progressively mineralize over the 8-week period as assessed by microCT imaging. (A) Time course of graft ossification for the 5 grafts implanted for 8 weeks. (B) Projections of rendered microCT scans for graft 687 from week 5 through week 8 showing the spatial progression of ossification within the graft. (C) 3D renderings of graft #687 implant thresholded to show bone-equivalent density and internal structure. Scale bars in all panels equal 1 cm.

FIGURE 6

Human osteoblasts are present in grafts but not in peripheral tissues. (A) Human and rat osteocalcin transcripts, markers of osteoblasts, are present in 4-week implants. (B) No human cells (absence of human β-actin sequence) were detected in grafts implanted for 8 weeks. (C) Rat, but not human (β-actin) sequences were detected in liver (4 weeks) or spleen (8 weeks) genomic DNA. (D) Rat but not human osteocalcin transcripts were detected in peripheral tissues (spleen) of 8-week recipient animals. GAPDH primers were used to confirm intact templates for transcript PCR.

FIGURE 7

Grafts exhibit increased mechanical integrity after implantation. (A) Compression modulus of grafts at each experimental phase, *p < 0.05. (B) Representative stress-strain curves for each experimental group (4 weeks graft 682, 8 weeks graft 678.) (C) Mineralized volume of grafts explanted after 8 weeks plotted against the respective compression moduli.