Skip to Main Content
Text size: SmallMediumLargeExtra-Large

Welcome to the Academy

May 2012

Gordana Vunjak-Novakovic, Ph.D.

In February, three NIDCR grantees were elected to the National Academy of Engineering. Science Spotlight plans over the coming weeks to interview all three scientists to hear their thoughts on receiving this high honor and learn more about the latest progress in tissue engineering. Our first stop is Gordana Vunjak-Novakovic, Ph.D. She is a professor of biomedical engineering and a professor of medicine at Columbia University in New York, where she serves as director of Columbia’s Laboratory for Stem Cells and Tissue Engineering and is the co-director of the Craniofacial Regeneration Center. Dr. Vunjak-Novakovic also is an associate director of NIH's Resource Center for Tissue Engineering.

Were you surprised by your election to the academy?

Not completely, but I couldn’t be happier. I kept getting requests for biographical information. So I started to connect the dots. Of course, I wasn’t sure. I had made a mental note to look on February 9, the date of the announcement, to see if my hunch was correct.

And it was.

Yes, but first the plot thickened a bit. It turns out that the academy likes to notify its new members the day before the public announcement. Of course, I didn’t know this. On February 8, I remember returning to my office in the afternoon after teaching a class, not thinking at all about the election. A pile of mail was waiting for me with a very large box balanced on top. I thought, “Oh, the box must be something for me to review.” About two hours later, I finally got around to opening it. Surprise. The box was the academy’s notification package for its new members.

I saw that my colleague Tony Mikos [Rice University in Houston] also had been elected. So I immediately called to congratulate him, and he said, “No, I haven’t been elected. I haven’t received anything.” I told him to go to his mailroom. I said, “A large box will be there waiting for you.” And it was. The box contained the formal letter of congratulations and various and sundry materials about the academy and its membership, past and present.

Let’s talk about your NIDCR grant.

Sure, NIDCR has funded me continuously now for about eight years. In the current grant, I’ve proposed to engineer in the laboratory living osteochondral grafts.

Architecture of engineered bone (darker brown color is bone matrix, lighter brown is the scaffold)

In other words, the graft will have cartilage on one end and transition into bone.

That’s right. From an engineering standpoint, we utilize three main building materials. The first is stem cells. What seeds are to a garden, stem cells are to regenerative tissues and organs. In other words, the stem cells are the true tissue engineers. We just coax them into working their magic. For the osteochondral grafts, we utilize different stem cells to produce the committed cell lineages that are required to make new bone and cartilage.

What’s the second building material?

The scaffold. It is a porous, three-dimensional structure that, in this case, mimics the body’s spongy bone matrix that normally guides the biomineralization process. We seed the scaffold with stem cells and various growth-promoting biochemicals. From an engineering perspective, the scaffold offers structural support and defines the signaling for the stem cells and their cell lineages to produce new tissue.


And the third main engineering material?

Number three is the bioreactor. It is essentially a three-dimensional cell culture system. Another way to think of it is as a self-contained biosphere, or pseudo body, for the cells. We control the factors that enter the bioreactor over time. These factors include essential nutrients, oxygen, and physical factors that keep these large pieces of bone alive and functional.

Your NIDCR grant focuses on the mineralized tissues of the face and skull, or the craniofacial region. Is that correct?

Yes. Our thinking is if we can make a piece of the jaw, the cheekbone, or a whole section of the skull, we can make anything.

Why’s that?

The structures of the craniofacial region are much more intricately designed than the long bones. By “long bones,” I’m talking about the weight-supporting skeletal structures. Think of the shin or thigh bones. From an engineering standpoint, long bone is a cylinder with marrow inside. A facial bone is daVinci. Each is anatomically unique in size, shape, and curvature. They also vary from face to face, much like the whorls of our fingerprints differ from person to person.

Craniofacial bones also have a different embryonic origin than the long bones. Correct?

That’s right. The properties of long and craniofacial bones appear to be slightly different. How we don’t know, and that becomes an issue when you ask whether it’s possible to take bone marrow from the ileac crest [the curved ridge of the hip bone] and use it to regenerate craniofacial bone? Or even vice versa. We need to find out.

The other major complicating factor is craniofacial bones connect to other structures. There can be, for example, a layer of cartilage to consider. There is a muscle that extends from the bone. There is ligament. Then, of course, there is the issue of vascularization. Something you grow in culture that looks beautiful would die in vivo, or in the body, within days without a blood supply. So from a tissue engineering perspective, reconstructing these interfaces is a real challenge.

Human TMJ engineered grown in vitro

So how do you begin to address these issues?

I think some realism is in order first. In computer science, they talk in increments -- 1.0, 2.0., 3.0. But before tissue-engineered bone or any other structure can go live in the clinic, we must work out all of the investigative bugs. Our product must be as near a 10.0 as possible. It can’t crash after transplantation. There is no reboot in reconstructive surgery. Working out these bugs will involve contributions from myriad scientific disciplines. Just as Rome wasn’t built in a day, viable replacement tissues won’t arise overnight.

But you seem very optimistic about the task ahead.

Oh absolutely. Two points come to mind. The first is you don’t need to provide cells with a perfectly recapitulated environment. They don’t need everything. They just need enough. In other words, our task is to give them the most critical components. They will figure out the rest, and we’ve already seen this in the laboratory. In all honesty, I think many of our successes in tissue engineering to date haven’t necessarily been due to anything brilliant that we’ve done. The success is due to the profound ability of our cells to adapt to various conditions. So I think a big question for the field will continue to be: How much is enough? How much do you really need to do in advance? Ideally, a very smart scaffold would allow the graft to recruit its own cells, its own housekeeping and regenerative forces. We’re not there yet. But that’s where we’re headed.

This raises my second point. We live in an unprecedented era of scientific discovery. Researchers can now gather vast amounts of biological information, and our computational power will only grow to cut through the biological complexity and isolate the essential. So the best is very much yet to come for tissue engineering. Our future is bright.

Let’s take a look back at the origins of the field. Whenever I read about replacement tissues, I usually happen upon the phrase, “the new field of tissue engineering.” And yet, this new field already has undergone a pretty profound paradigm shift. That is, it has advanced from empirical, tissue-try-this alchemy to a truly science-based recapitulation of nature’s own recipe to build a tissue, or biomimetics. True?

That’s right. Initially, we measured progress empirically. Here are some cells, here are some biomaterials. Put them together and then hope the cells figure out how to do the job. It was a lot of the same trial and error that a beginning cook might experience in the kitchen. And then over the past 10 to 15 years, more rational approaches emerged. These approaches build on the fundamental premise that I mentioned a moment ago: The stem cells are the ultimate tissue engineers. Our task as scientists is to figure out how to provide them with the right environmental cues to engineer viable new tissue.

Another way to think of it is the cells come equipped with their own machinery, and all of these moving parts have been encoded into their genomes for many, many generations to do the same thing. When exposed to the appropriate environmental cues, make a desired tissue. Our job is to master the environmental cues that occur in developmental time and space, recapitulate them in the bioreactor, and let the stem cells take care of the rest.

And physical cues?

Exactly. Physical forces are present throughout most of development. One example that guides our research is that only three weeks into gestation, the human heart starts beating. The heart is hardly visible, and it’s functional, meaning blood already flows through the body to enable its full development. You see this happening and think the only logical approach is to mimic these developmental conditions and trick the cells into making new tissue.

Tissue engineering seems in some ways like a mixing bowl for science, technology, and innovation. This makes the field fundamentally collaborative. How does collaboration play a role in setting your research course?

Let me give you two examples. We have very strong collaborations in place with stem cell biologists and clinicians. These collaborations allow us conceptually to travel in two directions at once. In one direction, the stem cell biologists help us to delve more deeply into the basic biology. As a result, we understand much better how bone forms. It’s not that there is one generic bone cell. There are many different bone cells, and they have different roles at different times. The matrix, too, is very complex and essential to understand.

What about the other direction?

In the other direction, we can take into account the clinical requirements for our basic discoveries. We can’t produce engineering tools that are profoundly complicated. If a bioreactor is designed like a Rubik’s cube, it’s never going to work in the clinic. The translational path will be too cumbersome, too long, and too expensive for anyone to be interested in it. The clinicians remind us of these facts and keep us grounded. We need them every step of the way.

We also need young talent to keep the field vibrant and innovative. I can’t stress this enough. There are two pillars of progress in tissue engineering: Young talent and collaboration.

Given the tall scientific task ahead, though, are you aiming for small, incremental advances? Or, do you have your sights set on hitting a reconstructive home run?

I think we are going for the home run. Here’s why. One of my colleagues is a surgeon who does more than 800 reconstructive surgeries per year. That’s roughly two surgeries per day. He said, “Give me something.” He often reminds me that there is such a tremendous need for new and better surgical alternatives. Today, surgeons typically extract a piece of rib or bone from your leg. They then very precisely carve the bone in the surgery room, reinsert it into the problem area, and wrap muscle around the graft to cushion it and enhance blood flow. It’s just not the optimal approach. That’s why I don’t want to settle for incremental improvement. That said, I’m not talking here about engineering a huge, integral graft. We are looking at the pieces that are most needed. We also are looking very much into a modular approach, so that you can sort of construct what you need from individual pieces and make the surgeon’s life easier. You always try to make a difference. I want to change the way people are reconstructing bone.

Pig TMJ reconstruction

Shifting gears a bit, we’ve talked about tissue engineers leveraging the latest progress in science and technology. But doesn’t tissue engineering also give back to science and technology? The process must be reciprocal.

That’s actually an extremely important point. As we apply our lessons learned, our engineering systems have improved tremendously. We have produced much better model systems that can be exported across the disciplines.

What do you mean?

Well, take the Petrie dish. Julius Petrie, who cultured bacteria for the famous Robert Koch, invented this standard agar plate in the 1880s to study cell behavior. When cultured in a Petrie dish, the bottom of the cell anchors to the substrate. The top of the cell faces the path of the growth medium. Well, this situation is totally non-physiological. A lot of evidence shows that a cell’s responses under these two-dimensional conditions are much different than in the body’s three-dimensional space. Now take the bioreactor that I mentioned earlier. Let’s say it houses a little three-dimensional piece of tissue. The tissue may be very small, but it is still three-dimensional and surrounded by other cells in matrix. Everything is pulling, pressing, and electro-stimulating, and then you flow in culture medium to mimic the flow of blood. This is a very in vivo-like situation, and an experimental system that will benefit any discipline.

The other thing – and NIDCR is playing a big role in it – we are getting much better at designing bioreactors that generate gradients. In other words, we are no longer confined to studying one factor or compressive force at a time. We can generate gradients in which, say, one region is exposed to one set of conditions while the other is exposed to another set of conditions. This is how the body naturally generates its boundaries and interfaces.

What about screening platforms?

That’s another great example. We reverse the engineering paradigm and instead create large numbers of very small tissues. We call them microtissues. Now these microtissues can be used to construct models of disease for drug screening assays. Instead of doing the screening in animals, the work can be done in a more precise way in a human system comprised of these microtissues. We are now talking to pharmaceutical companies that are interested in having this human-in-a-dish system.

What about bioprinting, the 3D printing out of human cells?

We do bioprinting. In fact, in 2011, we had an extremely interesting result. We printed cells in extremely small patterns, something on the scale of 100 to 200 micrometers. You can’t even see the pattern with the naked eye. We wanted to study the behavior of cells in communities of roughly 200 or 300 cells.

What did you discover?

We discovered that these cells expressed directional motion, which is one of the fundamental prerequisites for development. In other words, groups of cells follow a programmed directional motion in the embryo that allows them to polarize, a fundamental step in forming a tissue or organ. In our bioprinting study, we found that these teeny-tiny patterns of cells express the same type of behavior. We also discovered that different cell types act in different ways. Let’s say you’re looking at a muscle cell, it turns clockwise. If you looked at a blood vessel cell, it turns counterclockwise.

But the big surprise is if you have a specific cell type and a corresponding cancer cell, their polarity has a different orientation. These changes are actually seen much sooner than other changes indicative of disease. So we are now looking into whether this bioprinting system might serve as a diagnostic tool.

Dr. Vunjak-Novakovic, congratulations again on your election to the National Academy of Engineering, and thanks for the conversation.

Share This Page

GooglePlusExternal link – please review our disclaimer

LinkedInExternal link – please review our disclaimer


This page last updated: February 26, 2014