The advent of the three-dimensional (3D) printing technology has significantly improved our capacity in fabricating volumetrically well-defined structures at unprecedented ease. The same set of strategies have been further adapted in recent years, to achieve production of 3D constructs made of hydrogels, either alone or containing living cells, with the latter typically referred to as the bioinks, towards applications in tissue biofabrication.

A variety of 3D printing methods are available to process hydrogel-based inks into spatially defined constructs, comprehensively summarized in our recent review (Small 15, 1805510, 2019). Taking the most commonly used extrusion printing as an example, we as well as many colleagues have, demonstrated that hydrogel inks can be directly extruded into different shapes for usage (Advanced Materials 30, 1805460, 2018; Advanced Materials 31, 18006590, 2019), or the extruded patterns can serve as templates that are selectively removed to form hollow channels within another hydrogel volume (Biotechnology Journal 14, 1700703, 2019); alternatively, a co-axial printhead design would allow for single-step extrusion of perfusable fibrous structures (Advanced Materials 30, 1706913, 2018).

Yet, it has also long been apparent to the field that hydrogel-based 3D printing presents limitations, one of which lies in their resolutions attainable (Advanced Materials 25, 5011-5028, 2013; Lab on a Chip 19, 2019-2037, 2019). Again for extrusion printing, the highest resolution that can be faithfully achieved falls in the range of a couple hundred micrometers – although this limit may be pushed to slightly lower through hardware or ink improvements, such methodologies are not always practical, or oftentimes the associated economic/time costs are prohibiting. Therefore, how to effectively resolve the resolution challenge in a simple way has been locking our minds of 3D printing-related research within the lab since quite some time ago, and the interesting concept of shrinking down the size of a hydrogel construct post-printing, as opposed to figuring out how to spend the efforts to design super-printers or super-inks to obtain high resolutions, was then brought up.

While we had already been exploring the different approaches of ‘shrinking’ with varying degrees of success, the key time point came when I had a chance to visit Utrecht University, The Netherlands, initially for a different collaborative effort with Prof. Roos Masereeuw, in the early summer of 2018. Through that trip I was introduced to Prof. Tina Vermonden, who explained to me a then just-accepted paper of theirs (Soft Matter 14, 6327-6341, 2018), where they showed that binding of cationic lysozyme to the negatively charged microgel spheres resulted in deswelling, and thus controlled release of the loaded therapeutic agents. After outlining our perspective in shrinking 3D-printed hydrogel constructs for resolution enhancement, a collaboration team was immediately formed.

Back in Boston, we extensively explored the use of charge complexation to induce stable shrinkage of hydrogel materials that are commonly used in 3D printing, including for example, hyaluronic acid methacrylate (HAMA), gelatin methacryloyl (GelMA), and alginate – all are negatively charged at neutral pH values. Indeed, thrill came when one of the first experiments was successful, where we threw a few microfluidically extruded GelMA/alginate hollow fibers, half or one millimeter in diameter, into an acetic acid solution of positively charged chitosan, and observed their shrinkage that was stable but not reversible (Fig. 1).

Fig. 1. One of the very first feasibility experiments of shrinking printing. © Zhang Laboratory

Through additional joint efforts, we were able to finally optimize the parameters for shrinking of post-printed hydrogel constructs of various types (negatively versus positively charged) and generated by different printing methods (Fig. 2), as well as the effects of shrinking agent properties (e.g., charge density and molecular weight).

Fig. 2. Extending shrinking printing to (a) directly extruded patterns and (b, c) sacrificially generated channels through the use of (b) Pluronic and (c) polycaprolactone as the fugitive templates. The inks are all HAMA and the shrinking agent is chitosan acetic acid solution.

Moreover, utility case towards cell-laden hydrogels was also demonstrated using a neutral solution of quaternary ammonium salt chitosan (Q.chitosan) as the shrinking agent, where it was further found that successive shrinking each at a shorter time led to much higher viability of embedded cells compared to a single shrinking at a longer time (Fig. 3).

Fig. 3. Comparisons of the viability values and proliferation potentials of MCF-7 cells in GelMA/HAMA constructs after a single shrinkage (4 h straight on Day 1) and after two successive shrinkage (2 h each on Day 1 and Day 3) in culture medium supplemented with Q.chitosan.

We hope that our unique strategy of shrinking printing to find broad applications in 3D printing/bioprinting, biofabrication, and tissue engineering.

Mov. 1. The shrinking gels (GelMA and HAMA in chitosan acetic acid solution) replayed at ~3,000X of speed. © Zhang Laboratory

For more details, check out our paper “Complexation-Induced Resolution Enhancement of 3D-Printed Hydrogel Constructs” on Nature Communications.