Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Instant assembly of collagen for tissue engineering and bioprinting

Nature Materials volume 24, pages 1307–1318 (2025)Cite this article

Subjects

Abstract

Engineering functional cellular tissue components holds great promise in regenerative medicine. Collagen I, a key scaffolding material in bodily tissues, presents challenges in controlling its assembly kinetics in a biocompatible manner in vitro, restricting its use as a primary scaffold or adhesive in cellular biofabrication. Here we report a collagen fabrication method termed as tunable rapid assembly of collagenous elements that leverages macromolecular crowding to achieve the instant assembly of unmodified collagen. By applying an inert crowder to accelerate the liquid–gel transition of collagen, our method enables the high-throughput creation of physiological collagen constructs across length scales—from micro to macro—and facilitates cell self-assembly and morphogenesis through the generation of tunable multiscale architectural cues. With high biocompatibility and rapid gelation kinetics, the tunable rapid assembly of collagenous elements method also offers a versatile bioprinting approach for collagen over a wide concentration range, enabling the direct printing of cellular tissues using pH-neutral, bioactive collagen bioinks and achieving both structural complexity and biofunctionality. This work broadens the scope of controllable multiscale biofabrication for tissues across various organ systems using unmodified collagen.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: MMC-induced instant gelation facilitates rapid, multiscale collagen fabrication.
Fig. 2: Collagen-bundle-scaffolded microvasculature formation.
Fig. 3: In situ differentiation and self-assembly of macroscopic intestinal tubes from TRACE strips.
Fig. 4: Multiscale rapid free-form printing of collagen.
Fig. 5: Enabling robust collagen-based cellular tissue printing.

Similar content being viewed by others

Data availability

All data supporting the results of this study are available within the Article and its Supplementary Information. Source data are provided with this paper. Additional data are available from the corresponding author upon request.

References

  1. Caliari, S. R. & Burdick, J. A. A practical guide to hydrogels for cell culture. Nat. Methods 13, 405–414 (2016).

    Article  CAS  Google Scholar 

  2. Dong, C. & Lv, Y. Application of collagen scaffold in tissue engineering: recent advances and new perspectives. Polymers 8, 42 (2016).

    Article  Google Scholar 

  3. Brown, R. A. In the beginning there were soft collagen-cell gels: towards better 3D connective tissue models? Exp. Cell. Res. 319, 2460–2469 (2013).

    Article  CAS  Google Scholar 

  4. Hong, H. et al. Multi-scale fabrication techniques of collagen hydrogel for developing physiological 3D in vitro barrier model. Int. J. Precis. Eng. Manuf. 23, 227–254 (2022).

    Article  Google Scholar 

  5. Tran-Ba, K.-H., Lee, D. J., Zhu, J., Paeng, K. & Kaufman, L. J. Confocal rheology probes the structure and mechanics of collagen through the sol-gel transition. Biophys. J. 113, 1882–1892 (2017).

    Article  CAS  Google Scholar 

  6. Holder, A. J. et al. Control of collagen gel mechanical properties through manipulation of gelation conditions near the sol–gel transition. Soft Matter 14, 574–580 (2018).

    Article  CAS  Google Scholar 

  7. Yaari, A., Schilt, Y., Tamburu, C., Raviv, U. & Shoseyov, O. Wet spinning and drawing of human recombinant collagen. ACS Biomater. Sci. Eng. 2, 349–360 (2016).

    Article  CAS  Google Scholar 

  8. Mailand, E. et al. Tissue engineering with mechanically induced solid-fluid transitions. Adv. Mater. 34, 2106149 (2022).

    Article  CAS  Google Scholar 

  9. Kajave, N. S., Schmitt, T., Nguyen, T.-U. & Kishore, V. Dual crosslinking strategy to generate mechanically viable cell-laden printable constructs using methacrylated collagen bioinks. Mater. Sci. Eng. C 107, 110290 (2020).

    Article  CAS  Google Scholar 

  10. Lee, A. et al. 3D bioprinting of collagen to rebuild components of the human heart. Science 365, 482–487 (2019).

    Article  CAS  Google Scholar 

  11. Wolf, K. et al. Physical limits of cell migration: control by ECM space and nuclear deformation and tuning by proteolysis and traction force. J. Cell Biol. 201, 1069–1084 (2013).

    Article  CAS  Google Scholar 

  12. Murphy, C. M., Haugh, M. G. & O’Brien, F. J. The effect of mean pore size on cell attachment, proliferation and migration in collagen–glycosaminoglycan scaffolds for bone tissue engineering. Biomaterials 31, 461–466 (2010).

    Article  CAS  Google Scholar 

  13. Zhang, Y. et al. 3D printed collagen structures at low concentrations supported by jammed microgels. Bioprinting 21, e00121 (2021).

    Article  Google Scholar 

  14. Hofer, M. & Lutolf, M. P. Engineering organoids. Nat. Rev. Mater. 6, 402–420 (2021).

    Article  CAS  Google Scholar 

  15. Seo, B. R. et al. Collagen microarchitecture mechanically controls myofibroblast differentiation. Proc. Natl Acad. Sci. USA 117, 11387–11398 (2020).

    Article  CAS  Google Scholar 

  16. Li, Y., Asadi, A., Monroe, M. R. & Douglas, E. P. pH effects on collagen fibrillogenesis in vitro: electrostatic interactions and phosphate binding. Mater. Sci. Eng. C 29, 1643–1649 (2009).

    Article  CAS  Google Scholar 

  17. Oh, S., Nguyen, Q. D., Chung, K.-H. & Lee, H. Bundling of collagen fibrils using sodium sulfate for biomimetic cell culturing. ACS Omega 5, 3444–3452 (2020).

  18. Santos, M. I. et al. Endothelial cell colonization and angiogenic potential of combined nano- and micro-fibrous scaffolds for bone tissue engineering. Biomaterials 29, 4306–4313 (2008).

    Article  CAS  Google Scholar 

  19. Dessalles, C. A., Leclech, C., Castagnino, A. & Barakat, A. I. Integration of substrate- and flow-derived stresses in endothelial cell mechanobiology. Commun. Biol. 4, 764 (2021).

    Article  Google Scholar 

  20. Miller, J. S. et al. Rapid casting of patterned vascular networks for perfusable engineered three-dimensional tissues. Nat. Mater. 11, 768–774 (2012).

    Article  CAS  Google Scholar 

  21. Daly, A. C., Davidson, M. D. & Burdick, J. A. 3D bioprinting of high cell-density heterogeneous tissue models through spheroid fusion within self-healing hydrogels. Nat. Commun. 12, 753 (2021).

    Article  CAS  Google Scholar 

  22. Bernal, P. N. et al. Volumetric bioprinting of organoids and optically tuned hydrogels to build liver-like metabolic biofactories. Adv. Mater. 34, 2110054 (2022).

    Article  CAS  Google Scholar 

  23. Nikolaev, M. et al. Homeostatic mini-intestines through scaffold-guided organoid morphogenesis. Nature 585, 574–578 (2020).

    Article  Google Scholar 

  24. Lawlor, K. T. et al. Cellular extrusion bioprinting improves kidney organoid reproducibility and conformation. Nat. Mater. 20, 260–271 (2021).

    Article  CAS  Google Scholar 

  25. Kang, H.-W. et al. A 3D bioprinting system to produce human-scale tissue constructs with structural integrity. Nat. Biotechnol. 34, 312–319 (2016).

    Article  CAS  Google Scholar 

  26. Brassard, J. A., Nikolaev, M., Hübscher, T., Hofer, M. & Lutolf, M. P. Recapitulating macro-scale tissue self-organization through organoid bioprinting. Nat. Mater. 20, 22–29 (2021).

    Article  CAS  Google Scholar 

  27. Lorenzo-Martín, L. F. et al. Spatiotemporally resolved colorectal oncogenesis in mini-colons ex vivo. Nature 629, 450–457 (2024).

    Article  Google Scholar 

  28. Spence, J. R. et al. Directed differentiation of human pluripotent stem cells into intestinal tissue in vitro. Nature 470, 105–109 (2011).

    Article  Google Scholar 

  29. Lancaster, M. A. et al. Cerebral organoids model human brain development and microcephaly. Nature 501, 373–379 (2013).

    Article  CAS  Google Scholar 

  30. Buehler, M. J. Nature designs tough collagen: explaining the nanostructure of collagen fibrils. Proc. Natl Acad. Sci. USA 103, 12285–12290 (2006).

    Article  CAS  Google Scholar 

  31. Connizzo, B. K., Yannascoli, S. M. & Soslowsky, L. J. Structure–function relationships of postnatal tendon development: a parallel to healing. Matrix Biol. 32, 106–116 (2013).

    Article  CAS  Google Scholar 

  32. Salvatore, L. et al. Mimicking the hierarchical organization of natural collagen: toward the development of ideal scaffolding material for tissue regeneration. Front. Bioeng. Biotechnol. 9, 644595 (2021).

    Article  Google Scholar 

  33. Moxon, S. R. et al. Suspended manufacture of biological structures. Adv. Mater. 29, 1605594 (2017).

    Article  Google Scholar 

  34. Hinton, T. J. et al. Three-dimensional printing of complex biological structures by freeform reversible embedding of suspended hydrogels. Sci. Adv. 1, e1500758 (2015).

    Article  Google Scholar 

  35. Osidak, E. O., Kozhukhov, V. I., Osidak, M. S. & Domogatsky, S. P. Collagen as bioink for bioprinting: a comprehensive review. Int. J. Bioprinting 6, 270 (2020).

    Article  CAS  Google Scholar 

  36. Lei, I. M. et al. Soft hydrogel shapeability via supportive bath matching in embedded 3D printing. Adv. Mater. Technol. 8, 2300001 (2023).

    Article  CAS  Google Scholar 

  37. Li, Q. et al. Regulable supporting baths for embedded printing of soft biomaterials with variable stiffness. ACS Appl. Mater. Interfaces 14, 41695–41711 (2022).

    Article  CAS  Google Scholar 

  38. Zhao, Y. et al. A platform for generation of chamber-specific cardiac tissues and disease modeling. Cell 176, 913–927.e18 (2019).

    Article  CAS  Google Scholar 

  39. Skylar-Scott, M. A. et al. Biomanufacturing of organ-specific tissues with high cellular density and embedded vascular channels. Sci. Adv. 5, eaaw2459 (2019).

    Article  CAS  Google Scholar 

  40. Lavrador, P., Moura, B. S., Almeida-Pinto, J., Gaspar, V. M. & Mano, J. F. Engineered nascent living human tissues with unit programmability. Nat. Mater. 24, 143–154 (2024).

  41. Choi, S. et al. Fibre-infused gel scaffolds guide cardiomyocyte alignment in 3D-printed ventricles. Nat. Mater. 22, 1039–1046 (2023).

    Article  CAS  Google Scholar 

  42. Fang, Y. et al. Expanding embedded 3D bioprinting capability for engineering complex organs with freeform vascular networks. Adv. Mater. 35, 2205082 (2023).

    Article  CAS  Google Scholar 

  43. Johnson, T. D. et al. Quantification of decellularized human myocardial matrix: a comparison of six patients. Prot. Clin. Appl. 10, 75–83 (2016).

    Article  CAS  Google Scholar 

  44. Hata, Y., Sawada, T. & Serizawa, T. Macromolecular crowding for materials-directed controlled self-assembly. J. Mater. Chem. B 6, 6344–6359 (2018).

    Article  CAS  Google Scholar 

  45. Saeidi, N. et al. Molecular crowding of collagen: a pathway to produce highly-organized collagenous structures. Biomaterials 33, 7366–7374 (2012).

    Article  CAS  Google Scholar 

  46. Ng, W. L., Goh, M. H., Yeong, W. Y. & Naing, M. W. Applying macromolecular crowding to 3D bioprinting: fabrication of 3D hierarchical porous collagen-based hydrogel constructs. Biomater. Sci. 6, 562–574 (2018).

    Article  CAS  Google Scholar 

  47. Magno, V. et al. Macromolecular crowding for tailoring tissue-derived fibrillated matrices. Acta Biomater. 55, 109–119 (2017).

    Article  CAS  Google Scholar 

  48. Dewavrin, J.-Y., Hamzavi, N., Shim, V. P. W. & Raghunath, M. Tuning the architecture of three-dimensional collagen hydrogels by physiological macromolecular crowding. Acta Biomater. 10, 4351–4359 (2014).

    Article  CAS  Google Scholar 

  49. Ranamukhaarachchi, S. K. et al. Macromolecular crowding tunes 3D collagen architecture and cell morphogenesis. Biomater. Sci. 7, 618–633 (2019).

    Article  CAS  Google Scholar 

  50. Doube, M. et al. BoneJ: free and extensible bone image analysis in ImageJ. Bone 47, 1076–1079 (2010).

    Article  Google Scholar 

  51. Duraivel, S. et al. A silicone-based support material eliminates interfacial instabilities in 3D silicone printing. Science 379, 1248–1252 (2023).

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank M. Wu, C. Kim and A. Gonzalez for access to the CELLINK pneumatic bioprinter and training on the equipment. We also thank Z. Zhu and C. Bergwitz for providing a dissected small intestine from a killed mouse. We thank S. Campbell’s lab, X. Li and I. Gokhan for the hPSC line PGP1 and the technical assistance in cardiac differentiation. We acknowledge the use and research support provided by A. Datye and the Yale Mechanical and Thermal Analysis Instrumentation Core (MTAIC) for the rheometer experiments. We acknowledge the kind assistance with the microCT provided by J. N. VanHouten from the Yale microCT facility. We acknowledge NIH grant R35GM142875 (M.M.), Yale Liver Center Pilot Project under NIH award P30 DK034989 (M.M.), and NIH grant T32EB019941 (R.Y.N.).

Author information

Author notes

  1. These authors contributed equally: Xiangyu Gong, Zhang Wen, Zixie Liang.

Authors and Affiliations

  1. Department of Pharmacological Sciences, Renaissance School of Medicine, Stony Brook University, Stony Brook, NY, USA

    Xiangyu Gong, Hugh Xiao & Michael Mak

  2. Department of Biomedical Engineering, Yale University, New Haven, CT, USA

    Xiangyu Gong, Zhang Wen, Zixie Liang, Hugh Xiao, Sein Lee, Alejandro Rossello‐Martinez, Qinzhe Xing, Thomas Wright, Ryan Y. Nguyen & Michael Mak

  3. Yale Liver Center, Yale University School of Medicine, New Haven, CT, USA

    Michael Mak

Authors
  1. Xiangyu Gong
    View author publications

    Search author on:PubMed Google Scholar

  2. Zhang Wen
    View author publications

    Search author on:PubMed Google Scholar

  3. Zixie Liang
    View author publications

    Search author on:PubMed Google Scholar

  4. Hugh Xiao
    View author publications

    Search author on:PubMed Google Scholar

  5. Sein Lee
    View author publications

    Search author on:PubMed Google Scholar

  6. Alejandro Rossello‐Martinez
    View author publications

    Search author on:PubMed Google Scholar

  7. Qinzhe Xing
    View author publications

    Search author on:PubMed Google Scholar

  8. Thomas Wright
    View author publications

    Search author on:PubMed Google Scholar

  9. Ryan Y. Nguyen
    View author publications

    Search author on:PubMed Google Scholar

  10. Michael Mak
    View author publications

    Search author on:PubMed Google Scholar

Contributions

X.G. and M.M. conceptualized and designed this study. X.G. and Z.L. designed and performed the TRACE bioprinting, including support bath, bioprinting of acellular and cellular tissues, and tissue functional assays. X.G., Z.W. and S.L. performed the multiscale patterning of collagen and EC-patch experiments. Z.W. performed the isolation of mouse small intestinal cells and organoid maintenance. X.G. and Z.W. performed in situ hPSC differentiation and macroscopic mouse intestinal tubes. H.X. developed the collagen–cardiomyocytes bioink and assisted with live imaging. Z.L. and X.G. conducted the cardiac differentiation, bioprinting, maintenance and imaging. X.G., Z.L., Z.W. and S.L. analysed and visualized the data. A.R.-M. instrumented the bioink temperature control system for the Lulzbot bioprinter and helped with the bioprinting of ultrathin filaments and visualization of μ-bundle formation. Q.X. performed the shear-induced bundle alignment experiment. X.G. and T.W. designed and fabricated the PDMS multichannel devices. X.G. and R.Y.N. performed the rheological characterization. Z.L., X.G. and R.Y.N. prepared the SEM samples and performed the SEM measurements. M.M. supervised and directed this study. M.M. secured funding and resources for this study. X.G., Z.L., Z.W. and M.M. drafted the first submission. X.G., Z.L. and M.M. conducted the manuscript revision with input from all authors.

Corresponding author

Correspondence to Michael Mak.

Ethics declarations

Competing interests

M.M. and X.G. are inventors on a patent application related to this work filed by Yale University (US18/317,526, filed on 15 May 2023). M.M., X.G. and Z.L. are inventors on a patent application related to this work filed by Yale University (US18/653,162, filed in May 2024). The other authors declare no competing interests.

Peer review

Peer review information

Nature Materials thanks the anonymous reviewers for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 High throughput fabrication of micro-liter disks.

(a) Video recording showing high-throughput micro-liter collagen disk fabrication. Red boxes mark the same location where a 5 μL collagen droplet was deposited at t = 0. Scale bar, 2 mm. (b) Reflectance confocal microscopy reveals pore sizes from the surface to a 49 μm depth of collagen gels assembled via MMC vs. regular conditions (n = 42509, 35998, 18880, 15958, and 14807 pore measurements for five Z-sections of N = 1 crowded gel; n = 2134, 2291, 2158, 2156, 2260 pore measurements for five Z-sections of N = 1 regular gel). Dotted lines in the plots indicate the pore size of 2 μm. Scale bars, 10 μm. Box plot whiskers, boxes, and lines represent minimum, maximum, 25th–75th percentiles, and median. Comparisons were determined by one way ANOVA with a Tukey’s test. (c) Disk area versus collagen precursor droplet volume shows robust, tunable fabrication, with a linear regression passing through the origin. Data represent independent droplets from two independent experiments. (d) Confocal images showing the fluorescent intensity of crowded disks and regularly gelled collagen domes made of GFP-labeled collagen with varied collagen concentration. Scale bar: 500 μm. The fluorescent intensity measurements of the gels normalized against the mean intensity of the 2 mg/mL regular collagen gels. (For the crowding method, n = 22, 18, 21, 19, and 29 disks for 2, 4, 6, 8, and 10 mg/mL concentrations respectively; for the regular method, n = 14, 17, 20, and 17 disks for 2, 4, 6, and 8 mg/mL concentrations respectively. Disks were pooled from two independent replicates.) Data are presented as mean ± s.d. Comparisons were determined by two-way ANOVA with a Tukey’s test. P values indicate the significance of the comparisons between methods for the same collagen concentrations. Comparisons between other conditions are listed in the source data. (e) Collagen microparticle fabrication via pressurized spraying into the MMC bath. The fluorescent images show the 3D morphology of collagen microparticles fabricated with acidic 10 mg/mL collagen solution. Scale bars, 50 μm (left); 10 μm (right).

Source data

Extended Data Fig. 2 Multicellular tissue with mesoscopic collagen architectures.

(a) Schematic of using collagen μ-bundles as a novel ECM scaffold for cell culture. (b) High-resolution imaging of a fibrotic liver spheroid. Scale bar, 200 μm. Bundle and cell morphology in ROI1 and ROI2 are compared with those in local fibrotic regions of liver cancer patients. Scale bars in ROIs and histology slides, 50 μm. (c) High-throughput production of three types of LifeAct-tagged liver cell HepG2 spheroids: (i) without collagen, (ii) mixed with the neutralized collagen solution during formation, and (iii) mixed with thick collagen μ-bundles. Scale bars, 400 μm. (d) Size measurements of the three types of spheroids shown in (c). n = 47 spheroids each condition from N = 3 independent batches (Batches 1, 2, and 3). The lines indicate mean ± s.d. of the sizes of pooled spheroids from all batches. Comparisons were determined by one way ANOVA with a Tukey’s test. The red, blue, and green dots indicate the mean values of each batch.

Source data

Extended Data Fig. 3 Macroscopic liver cell constructs.

(a) Macrograph and (b) schematic showing the fabricating procedure of the parallel channel device for rapid strip production. (c) Schematic of fabricating cell-lade macroscopic strips. (d) Viability assay on HepG2 strips after fabrication (Day 0) and after 24-hour culture. Scale bars, 200 μm. (e) Histological staining of PAS and PAS with diastase (PAS/D) on a fixed 8-day HepG2 strip. Scale bars, 200 μm. (f) A HepG2 strip (1 × 107 cell/mL) cultured freely in the medium for 8 days. Scale bar, 1 mm. In the ROIs, actin and nucleus were stained. Scale bars, 200 μm. (g) Assembling the Day-7 HepG2 strips into a macroscopic complex pattern “YALE” in a culture medium-rich granular support. Scale bar, 2 mm.

Extended Data Fig. 4 Characterizations of μ-bundles as vascular scaffolds.

(a) Tracking EC-patch compaction dynamics normalized against Day 1 patch area over Day 1 to Day 5. Statistical significance between two patch types or patch areas between each day is determined by two-way ANOVA with Šídák’s multiple comparisons (n = 14 and 15 patches for the “Thin” and “Thick” bundle conditions respectively, from N = 3 biological replicates per condition). (b) Live confocal images showing the daily tracking of thick bundle patches and demonstrating cell infiltration and vascular network formation. Images correspond to the max intensity projection of a z-stack around 200 μm. Top, images showing an entire thick μ-bundle patch formation. Scale bars, 500 μm. Bottom, magnified images showing cell infiltration and network formation. Scale bars, 200 μm. (c) Fluorescent confocal image of a representative region on a thick bundle patch on Day 5, VE-Cad (red), nuclei (blue), actin filaments (green), μ-Bundles (gray). The ECs form intact junctions and wrap around the bundles to form vessel-like structures. (d) Live confocal images showing long-term (28 days) angiogenic sprouting of thick μ-bundle patches. Close up image shows the interconnected, luminal features of the angiogenetic sprouts. Scale bars, 300 μm. (e) EC strip compaction. Scale bar, 1000 μm. Lengths of strips (mixed with or without bundles) normalized by Hour 0 were tracked within (f) hours and (g) days with low or high-dosing VEGF treatments. (h) Strip length comparison on Day 5. Two-way ANOVA with Tukey’s multiple comparisons. For (f), n = 6, 5, 6, and 6 strips for “Control, low VEGF”, “Bundle, low VEGF”, “Control, high VEGF”, and “Bundle, high VEGF” conditions respectively. For (g) and (h), n = 15 strips without bundles for both low and high VEGF conditions, n = 9 strips with bundles for both low and high VEGF conditions, pooled from N = 2 biological replicates for “Control”, and 3 for “Bundle”. Data in all plots are presented as mean ± s.d.

Source data

Extended Data Fig. 5 hPSC macroscopic strips.

(a) Bright-field images tracking the expansion of an hPSC strip within the Geltrex encapsulation. Over time, the strip exhibited pronounced outgrowth of organized, luminal colonies. Scale bar, 500 μm. (b) The hPSC strips were allowed to expand for up to 7 days. Immunofluorescent staining of F-actin, Oct4, and Nanog demonstrate hPSC well preserved their proper apical/basal polarity and pluripotency after the long-term culture. Scale bars: 500 μm. (c) Bright-field images showing EB induction of an hPSC strip freely cultured in EB induction medium on Day 3 and Day 5. The strip exhibited a solid structure and brightened surface cells. Scale bars, 500 μm. (d) Day 5 EB strips can be further differentiated towards the neural lineage. Visible tissue folding of the neural epithelium is observed on Day 4 of the neural induction. Scale bars, left: 500 μm; right (close-up view): 250 μm.

Extended Data Fig. 6 Development and characterization of bioprinting in the TRACE bath.

(a) The TRACE bath is prepared by combining agarose slurry and PEG solution in syringes connected via a 90° lure-lock connector. The agarose slurry is termed the “regular bath.” (b) Rheological characterization of shear moduli (G’: storage modulus; G”: loss modulus) of TRACE bath and regular bath as a function of shear strain at the oscillation frequency of 1 Hz. (c) Shear moduli of both baths as the function of shear frequency at the strain of 0.1%. (d) Viscosity characterization of collagen bioinks at various concentrations (2, 6, and 24 mg/mL) across various shear rate. All three inks exhibit shear-thinning behavior (N = 3 independently prepared inks per concentration; mean ± s.d.). (e) Assessing (top view) the structural integrity and shape retention of 2 mg/mL neutralized collagen blocks (L: 1 cm, W: 1 cm, H: 0.5 cm) with grid infill printed in TRACE, regular agarose slurry, and gelatin+HEPES support baths, after releasing in PBS. Scale bars, 5 mm. (f) Slicing and macroscopic photography of a collagen block (1 cm × 1 cm × 1 cm) with grid infill were printed using 6 mg/mL acetic collagen in the TRACE support bath. The construct is released in PBS. (g) A weight bearing test to assess structural integrity of bioprinted constructs using 6 mg/mL acetic collagen. Blocks (1 cm × 1 cm × 1 cm) with grid infill were printed in “Regular” or TRACE support baths, with an additional variant crosslinked with 10 mM genipin for 24 h post-released from TRACE bath. The sequence of images records structural integrity and deformation recovery for each construct before application of a 2.5 g weight, during the load, and after its removal. Scale bar, 5 mm. Collagen printed in the TRACE bath shows much higher resistance to compressive force. The unmodified collagen can be further reinforced by biocompatible crosslinkers, such as genipin.

Source data

Extended Data Fig. 7 Resolution of TRACE printed filaments using 2 mg/mL neutralized collagen bioink.

(a) Representative images of single filament printed in PEG only bath (brightfield microscopy), and filament with bioink mixed with fluorescent beads printed in PEG+agarose granular support bath (confocal microscopy). Filament diameter was tuned by varying the combination of flow rate Q and printing speed v. Scale bars, 100 μm. (b) Filament diameter d was measured at 16 distinct combinations of flow rate Q and velocity v, from lowest to highest, with three filaments examined per combination. The results are shown as mean ± s.d., with sample sizes of n = 22, 15, 11, 11, 12, 12, 12, 12, 10, 10, 10, 10, 10, 10, 10, and 10 for each combination, respectively. The exact Q and v values used are listed in the source data. A strong agreement is observed between d and \(2\sqrt{Q/\pi v}\), highlighting the effectiveness of the conservation model in predicting TRACE printed collagen filament dimensions. The line in the plot represents \(d=2\sqrt{Q/\pi v}\).

Source data

Extended Data Fig. 8 Shear flow induced alignment of the collagen μ-bundles.

(a) Fabrication procedure of the bundle-infused collagen ink. Final bulk concentration of the bundles is 4 mg/mL. (b) Mini-disks formed via MMC using the bundle-infused ink droplets. (c) Extruded collagen filament via fast drawing (speed v = 10,000 mm/min) in a MMC bath. MMC bath is 800 mg/mL PEG8000 in PBS. (d) Comparison in μ-bundle alignment in the filament and the disks. Angular distribution of μ-bundle orientation in the shear-induced vs. bundles. The angle of 0° indicates the direction of the shear in shear-induced extrusion. n = 22 and 21 fields of view for 9 disks and 12 extruded filaments respectively, pooled from N = 3 independent experiments. Scale bars, 100 μm.

Source data

Extended Data Fig. 9 TRACE-printed centimeter-scale constructs with various collagen bioinks and the comparison with their original CAD designs.

(a) Heart design printed with neutralized, 2 mg/mL acellular collagen ink. (b) Heart design printed with acidic, 6 mg/mL acellular collagen ink. (c) Stomach design printed with acidic, 6 mg/mL acellular collagen ink. (d) 3D artery tree printed with HUVECs in 2 mg/mL neutral collagen. (e) Whole kidney design printed with HepG2 cells in 2 mg/mL neutral collagen. (f) A half kidney (cut plane indicated by the red dotted line in (e)) printed using the same cellular bioink showcasing the internal structures. Daily fluorescent imaging demonstrates that the construct is bioactive and undergoes tissue compaction without compromising its original design. The HepG2 cell line was chosen for its ease of expansion and cost-effectiveness, making it suitable for proof-of-concept studies. Scale bars, 1 cm.

Extended Data Fig. 10 Imaging-based evaluations of TRACE-printed cardiac tissues.

(a) High resolution confocal imaging showing the cellular organization the printed cardiac tissue at Day 7. The cell density measured from this image is ~1.8 × 108 cells/cm3, similar to that of a native heart. (b) 3D-reconstructe TRACE-printed cardiac dual chamber from microCT imaging. Scale bar, 1 mm. (c) Orthogonal views of the microCT imaging showing two compartments in the dual chamber. The cutting planes and the viewing directions of these views are indicated by the dotted lines and the arrows respectively. Scale bar, 1 mm.

Supplementary information

Supplementary Information

Supplementary Figs. 1–6, Tables 1 and 2, captions for Supplementary Videos 1–13, Methods, notes, discussion, G-code for ultrathin collagen filament printing, sources of organ models for 3D printing and references.

Reporting Summary

Supplementary Video 1

High-throughput, rapid formation of microlitre collagen discs. Neutralized, 2 mg ml–1 collagen droplets (5 μl) are dispensed into the PEG8000 bath (200 mg ml–1 in PBS) and form defined disc shapes in seconds with increasing opacity.

Supplementary Video 2

Microlitre collagen droplets dispensed into PBS. Neutralized, 2 mg ml–1 collagen droplets (5 μl) are dispensed onto the PBS and instantly disappear. No visible gelation is observed within 5 min. Around minute 6, some droplets start to form gels with undefined shapes.

Supplementary Video 3

Rapid fabrication of collagen strip in the channel device. Neutralized collagen solution is injected from the left side into a PDMS channel prefilled with PEG8000 solution. A distinct interface can be observed between the collagen solution and PEG bath.

Supplementary Video 4

Rapid single-line printing of collagen ink (neutralized, 2 mg ml–1) in the TRACE printing bath versus the regular agarose granular bath. The TRACE bath enables the rapid gelation of the deposited collagen ink and fast extraction (1 min), whereas the printed collagen does not solidify in the regular granular support bath and cannot be extracted from the bath 1 min or 5 min after printing.

Supplementary Video 5

High-speed (4,000 mm min–1) printing of a macroscopic windmill pattern (2 cm in diameter) consisting of microscopic ultrathin collagen filaments and material charging points in PEG8000 bath.

Supplementary Video 6

Structural folding of a macroscopic windmill pattern (diameter, 2 cm) consisting of an ultrathin collagen filament. The visible charging points are connected to the central point by the ultrathin filaments and can be bundled and manoeuvred by pulling only the central point.

Supplementary Video 7

Rapid releasing of the 3D collagen construct printed in the TRACE bath. A millimetre-scale 3D hollow gourd shape printed with acidified collagen ink (6 mg ml–1) in the TRACE bath shows expedited gelation at room temperature and can be released from the bath in 3 min. Room temperature for 3 min in the regular support bath (without PEG) is not sufficient to lead to full gelation and structural integrity on releasing.

Supplementary Video 8

Rapid releasing of a 3D collagen tube (diameter, 3 mm; height, 5 mm) printed in the TRACE bath. The millimetre-scale collagen tube printed with acidified collagen ink (6 mg ml–1) in the TRACE bath gels at room temperature and can be released from the bath in 3 min.

Supplementary Video 9

Strength comparison between a single filament printed in the TRACE support bath and the counterpart printed in a regular agarose slurry bath.

Supplementary Video 10

Spontaneous calcium transient and beating of cardiac tubes printed in the TRACE support bath versus the regular bath without MMC.

Supplementary Video 11

Whole-tissue spontaneous calcium and beating of a TRACE-printed cardiac ventricle construct. The calcium signal and tissue contraction are synchronized.

Supplementary Video 12

Cardiac cycles of a TRACE-printed cardiac ventricle construct (left, recording of the fluid output visualized by fluorescent beads; right, particle imaging velocimetry of the fluid field).

Supplementary Video 13

Cyclically beating cardiac dual chamber printed with physiological cell density.

Supplementary Data

Source data

Source Data Figs. 1–5 and Extended Data Figs. 1, 2, 4, 6–8

Source data for Figs. 1–5 and Extended Data Figs. 1, 2, 4, 6–8, organized in separate tabs and compiled in a single Excel file.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gong, X., Wen, Z., Liang, Z. et al. Instant assembly of collagen for tissue engineering and bioprinting. Nat. Mater. 24, 1307–1318 (2025). https://doi.org/10.1038/s41563-025-02241-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41563-025-02241-7

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing