Like a good chess player, Beckman Institute researcher Yi Lu isn’t afraid to step around the board and take a look at things from the opposite point of view. Lu has taken that approach to his research and his teaching – to great success in both areas.

Lu, a Professor in the Department of Chemistry at the University of Illinois and member of Beckman’s 3-D Micro- and Nanosystems group, does research that explores science at the nexus where chemistry meets biology. And he does it in a way that is fearless when it comes to changing – sometimes in 180-degree fashion – the direction of a research line.

Lu has a Ph.D. in Chemistry from UCLA but his research often uses biological approaches to advance chemical principles – a method that can provide great insights into biological systems. Recent research results have included practical applications like a new generation of highly sensitive and selective biosensors and bioimaging agents, including one class of sensors for testing toxic levels of metals like lead and another for drug testing in humans.

The sensor aspect of Lu’s work began when he pointed the research he was doing involving metal binding to DNA in a new direction, thanks to a breakthrough discovery about DNA he read about in Nature magazine. The article reported on a completely new method for achieving catalytic reactions that used DNA as an enzyme.

“When my group started, we were not interested in sensing technology at all,” Lu said. “When I started my group in 1994, there was a paper that year in which Breaker and Joyce at the Scripps Institute discovered that DNA can be an enzyme that carries out catalysis. The reason this was exciting is because most people thought that only RNA or proteins, which is the focus of the other part of my group, could perform catalysis.

“But more importantly, they mentioned that a metal can actually be a key in making the DNA function. I thought it was interesting to investigate because it was a brand new field. Let’s see how it works from the fundamental side. We asked ‘can we actually make DNA that can bind to a metal like lead, or a metal like mercury?’”

So Lu leveraged his laboratory’s previous work on how different metal ions can help DNA to function and flipped the direction to see if the DNA, or DNAzyme (deoxribozymes), can help detect metal ions. While attending an out-of-town conference Lu had the idea of reversing course on that line of research. He then called one of his graduate students and said to get started on the new line right away.

“Within a year my student, Jing Li, developed the first sensor,” Lu said.

That work in sensing and his work in catalysis (topics surrounding the rate of a chemical reaction) which involves interests in sustainable energy and biomass conversion, are the main aspects of Lu’s current research. The sensing research line is the big focus of his work at Beckman and has produced applications in the form of dipstick sensors that could soon be commercialized.

“At Beckman, we’re interested in a bottom-up approach, so-called direct assembly of nanomaterials and its applications in sensing and imaging,” Lu said of his research. “Many advances in other groups around the world have produced a variety of nanomaterials such as nanoparticles, nanorods and nanotubes. These nanomaterials are like individual pieces of puzzles. What we’re interested in more importantly is how to put all the pieces together into functional devices. Our idea is inspired by biology, using DNA as a programmable building block to assemble those nanomaterials into unique shapes and forms so that they can have very interesting structures and functions.”

A big breakthrough for Lu’s work was described by him, his student Hee-Kyung Kim and Beckman colleague Taekjip Ha and his postdoctoral associate Ivan Rasnik in a 2007 Nature Chemical Biology paper that reported a lock-and-key mechanism found in nature was also at work in their lead-specific DNAzyme. That discovery, Lu said, was crucial for understanding why their DNAzyme sensors are more selective and sensitive than many other sensing technologies.

“What we’ve been puzzled by is that proteins have been utilizing lock and key mechanisms for catalysis and for sensing for a long time,” Lu said. “DNAzymes, only discovered in 1994, also can catalyze reactions and perform sensing.

“In collaboration with Professor Taekjip Ha’s lab we are able to use single molecule FRET (fluorescence resonance energy transfer) technology to see this lock, which is DNAzyme, and the lead, which comes in like a key. Everyone wondered why this process is so selective and so sensitive and we provided an answer.”

Lu said their technique for detecting lead is based on the rate of the chemical reaction that takes place that does not require any reorganization of the metal-binding site.

“You could put another metal in the DNAzyme besides lead and it will go through a conformational change, but it costs energy and takes time,” he said. “The faster you can see the signal, the more sensitive it is. For the other metal ions, the process takes hours or longer, while for lead, it takes only seconds.”

Lu’s research group has also developed another sensor for uranium that has been used for detecting uranium contamination in soil. Lu said it has a detection limit in parts per trillion, which is better than more sophisticated analytical equipment that costs hundreds of thousands of dollars. He added that the technology can be expanded to include other potentially toxic metals like mercury and cadmium.

Lu said the exciting aspect of his work with sensing technology is that it is broadly applicable to a wide range of targets, including almost all of the small molecules such as metal ions and organic molecules, making it especially valuable in an area where there is a huge technology gap involving current methods. He said methods that employ antibodies are the only other technologies that can be used to detect a broad range of targets, but they have drawbacks that aren’t found with the sensors developed in his lab.

“Antibodies have been powerful in sensing and medical diagnostics,” he said. “One gap in antibodies-based detection is that they cannot detect small molecules very well. Sometimes small molecule markers are more reliable in medical diagnostics.”

Lu cited as an example the common method of measuring levels of PSA as a marker for prostate cancer – a method that is now the center of a debate in medicine.

“It has been shown that the protein marker PSA is not a very good marker because high levels of PSAs do not necessarily indicate the person will develop prostate cancer,” Lu said. (“In a recent issue of Nature) it was demonstrated that a small molecule marker called sarcosine gives a much better correlation between its presence in the body, such as in urine, and prostate cancer. Now we need a new technology to detect small molecules that antibodies are not very good at.

“I think we can fill in that gap using our technology developed at Beckman. We can use this DNA that can bind to small molecules; we have already demonstrated that we can convert the binding into changes of color, florescence, and MRI images. So now we can apply this new technology for detecting this and other small molecules to complement the current antibodies-based technology.”

Lu and his group are looking to apply their technology to improving magnetic resonance imaging (MRI) technology. He said the current clinical application of MRI sensors is good for diagnostics in that it can detect differences between damaged tissue (such as a tumor) and undamaged tissue. But, he added, the future of medicine will focus on prevention and that is going to require techniques that are able to detect disease at much earlier stages.

“What we want to do is use MRI as a powerful tool for not only detecting tumors when they develop but before they develop,” Lu said. “There are small molecular markers or protein markers that actually give you signs of tumor development before the tumor develops. Right now the current MRI contrast agents cannot see those small molecules because the contrast is not good enough to do that.”

Lu’s technique couples DNA with a non-toxic iron-oxide nanoparticle in a contrast agent that patients will ingest. The resulting MRI images would reveal differences at the level of small molecules, or molecular markers, for cancer and other diseases. Lu said they have demonstrated that the technique works in human serum, and they are now in the process of seeking funding for the research line in order to advance the work toward applications.

Lu’s group also produced a dipstick sensor in 2006 that was, using DNA-gold nanoparticle technology, able to detect cocaine and other drugs in saliva, urine and blood serum.

Lu’s work with sensing technology may soon be on display on store shelves in the form of inexpensive, reliable, and accurate colorimetric dipstick sensors. The research has spawned a start-up company, Dzyme Tech Inc., located in the University of Illinois Research Park. Lu and the company are working with the Environmental Protection Agency (EPA) on getting a lead sensor certified for use in EPA testing procedures.

“They are very excited about our results,” Lu said. “It seems to be innovative and better than the current technology. By the end of the year we should have an EPA-certified lead detection kit that can be marketed.”

Lu said there are testing kits such as those for lead on the market, but many of those sensors give high rates of false positives or false negatives, and are not EPA-certified. And there are none for detecting small molecules that will give results at the exact levels needed for determining whether a substance such as water is safe to drink or not.

“We are looking for the next generation of technologies and we believe we have that,” Lu said. “Another unique aspect of our technology is that we can tune the color change region to exactly where people care about.”

Lu gave as an example the EPA’s desire to have their detection kit for lead in paint to change color at one milligram per square centimeters (the threshold for leaded paint), while wanting another component of the same kit for lead in water to change color at 15 parts per billion (the maximum contamination level for safe drinking water).

“We have been able to develop a technology to meet the requirements so that we have one sensor for detecting lead in paint, one in dust, one in water,” he said. “So an EPA inspector or a homeowner can buy one kit for detecting toxic metal ions in a number of places.”

Like his approach to research, Lu has been unafraid to take a completely new path when it comes to teaching chemistry. A few years ago he received a grant from the Howard Hughes Medical Institute to develop a four-year course that teaches chemistry from a perspective that takes the students’ particular interests into account.

“We have outstanding students at Illinois and all of them came to Illinois full of curiosity,” Lu said. “But a lot of them are discouraged sometimes by the so-called very hard science courses like chemistry. Many of them, such as those who are going to medical school, consider chemistry as a “killer” course for their careers. What I wanted to figure out was a way to motivate them, because they are smart students.”

Lu said that instead of approaching things from the traditional point of view of a professor with his or her own particular interests in a field, he geared the course toward the interests of the individual students in different areas of chemistry.

“What I wanted to do was turn it upside down; it is the same thing as with the sensor idea,” he said. “Can we have a course that starts with the student interest instead of interests of a particular area, such as inorganic, organic, or physical chemistry? I use that as a starting point for instruction. Then I can utilize other courses to link their interests with that other course material.”

Students stay in the course for four years, with juniors and seniors helping to mentor freshmen and sophomores in similar areas of interest before moving on to lab work that lets them further explore their interests.

“This way, even if the course is very hard, they will be more likely to stick to it and excel in it,” Lu said. “The first element of the course is that its contents are based on interests of the students. The second element is peer mentoring, that is mentoring by students in the class who do not have any authority over their grades. This kind of advice is perceived as being much more acceptable. The same kind of advice coming from a peer mentor can be valued much, much more than advice coming from a higher level.

“We’re very happy with the course. HHMI wants the teaching to be like a research project, and it allows us to have a hypothesis, explore, and fail. The kids really are engaged. Obviously there are frustrations and a lot of work. But I’ve enjoyed it because I can see the results.”

Just like with his research work.