Although the idea of 3D printing living tissues may strain our credulity, the basic principles are easy to understand. In the early days, it was a matter of carefully depositing cell suspensions drop by drop, until eventually a 3D structure is obtained. Just like normal 3D printing.
But doing things this way hits some serious barriers. First, it is really slow – by the time the top layer is added, the cells in the bottom layer have been hanging around for some time, and are past their best, if not dead. Secondly, the resolution is pitiful, and certainly nowhere near the spatial precision you need to get realistic living tissues. But there is a third problem, and this is a more fundamental one: how to keep the cells in the middle of the tissue alive. It doesn’t take long before the cells in the middle - starved of oxygen and swimming in their own waste - go necrotic. Not surprising, when you think how much effort living organs take to make sure these deep tissues get perfused. We need a vasculature!
But how can you print a blood supply? One trick involves what has been called “sacrificial” elements. Nothing to do with furtive ceremonies at Halloween – the idea here is to add something to the printed object that you then take away. But finding the right material is tricky: you need something that on the one hand is tough enough to keep its shape during the printing process, and yet can be dissolved to evacuate the vasculature when you are done. Four years ago, Miller and colleagues at the Department of Bioengineering, University of Pennsylvania showed one way you can do this, using a carbohydrate glass. These glasses are made by dissolving a mixture of just the right carbohydrates, then boiling off the solvent. Eventually, after a lot of trial and error, they found the mix that gives the qualities they wanted.
But for all that, there are still problems with existing bioprinting methods. For one, the tissues can’t be perfused directly, limiting the thickness you can achieve with the tissues. And in general, current bioprinting methods just don’t come any where near realistic levels of complexity, geometry or longevity of real tissues. But last month a team headed up by Jennifer Lewis at the School of Engineering and Applied Sciences at Harvard, came up with a novel, and rather clever solution, to this problem. And Lewis’ method seems to allow the printing of tissues more than a centimetre thick, with different cell types all in their right places, and with a vasculature replete with endothelium. What is more, they even kept the printed tissues alive for over 6 weeks.
So what is their trick? Well, as every owner of a cheap deskjet printer will tell you, it is all down to the inks. In 3D bioprinting, the “inks” are actually a gelatin gel with cells embedded in it. Lewis’ trick lies in adding some special ingredients to these inks – namely, fibrinogen, thrombin and transglutaminase. If you know your thrombin, you might guess where we are going with this, but this is how it goes: First, Lewis starts with two inks, one for the cells themselves and one for the vasculature, plus an infilling matrix. To keep up with what follows, bear in mind that the basic idea is to use the thrombin to get the fibrinogen to form a gelatin-fibrin matrix, which is then stabilised by the transglutaminase. OK, so let's look at these inks a bit closer.
The ink for the vasculature is essentially thrombin suspended in pluronic-F127. This is printed just like in an inkjet printer, sprayed out of a nozzle. The cell-containing ink is a mixture of cells and fibrinogen suspended in gelatin. You have to be careful with the gelatin: you have to heat treat it to just the right temperature for just the right amount of time to get it the right stiffness.
Right – we have our inks, so we get printing. Once a lattice of proto-vasculature has been built up, interspersed with another lattice of cells, you pour in the matrix. This is where the clever stuff starts. The matrix also has fibrinogen in it. So matrix with fibrinogen comes into contact with pluronic F-127 with thrombin in it, and - well, you know what is going to happen: the fibrinogen diffuses into the matrix, the thrombin into the pluronic, and before you know it, you have got fibrin. The fibrin holds things together while you remove the pluronic, leaving an empty latticework of channels. Meanwhile the thrombin also makes its way to the cell-ink, where it meets more fibrinogen to turn into fibrin. But fibrin on its own isn’t that strong or stable, hence the transglutaminase (TG), which forms thermostable cross links between the fibrin and the gelatin.
Lewis flushed some human umbilical vein endothelial cells down the holes, and they dutifully took up their place along the linings, forming an effective diffusion barrier. Expression analysis confirmed that something not entirely unlike real vascular endothelium had formed. And by including bone precursor cells in the cell ink and adding the right mix of differentiating factors, they even made dense osteogenic tissue – bone, or at least almost.
The principle is both clever and adaptable – a rare combination. As well as finding such a nifty trick to print deep tissue with built-in vasculature, Lewis’ team has solved the biggest and most tedious challenges of working out the right ink composition and pretreatments, and it would take but very minor modifications to come up with quite diverse types of deep tissue, opening up a host of possibilities for experimental models, drug-testing platforms, perhaps even artificial tissues and organs.
Steven D Buckingham
Photo: K. Herfort