Two groups of researchers have reported proof-of-principle demonstrations of a new, high-resolution imaging technology that uses infrared lasers to harmlessly penetrate the skin and illuminate special nanoparticle “beacons” inside the body. Scanning systems based on such technology might one day enable the imaging of multiple structures and activities in the nervous system and other organs, at higher resolution than is possible with magnetic resonance imaging and other non-invasive techniques.
The new technique is similar to fluorescence imaging, in which tissues are infused, via subcutaneous injection, with a solution containing special antibodies or other molecules tagged with fluorescent beacons. When a laser of the right wavelength is shone on fluorescent beacons, their light is recorded with a sensitive digital microscope. In essence the fluorescence technique “paints” cells and other targeted structures with tiny specks of light, allowing them to be seen easily.
Fluorescence imaging is increasingly popular, but it has several drawbacks. Among them is the fact that it can’t be used to image deep tissues inside the body, in part because those tissues tend to give off their own fluorescence when illuminated, thus obscuring the signal from the fluorescing target beacons.
The new technique largely overcomes that drawback by relying on a different optical phenomenon known as Raman scattering. Discovered by the Indian physicist C.V. Raman, a contemporary of Einstein, Raman scattering occurs when some of the photons of light that hit a molecule bounce off with a slightly altered frequency.
Normally this type of scattering occurs too rarely to be easily detectable. But researchers have recently learned to design special nanoparticle beacons whose “Raman signatures” are very distinct. By making use of the near-infrared part of the spectrum, where the absorption of light by skin, fat and muscle is relatively weak, researchers can use these Raman beacons to see through a given thickness of biological tissues.
The technique has still another big selling point. “A key advantage for Raman imaging is its multiplexing capability,” says Shuming Nie, whose imaging laboratory, jointly affiliated to Emory University and Georgia Tech, is one of the leaders in this area. “Multiplexing” in this sense means the simultaneous recording of the separate Raman signatures of different beacons attached to different structures in the tissue being observed. Thus, for example, envisioning a future generation of this technology, the dopamine receptors on a particular neuron in the brain could be painted with one set of beacons, while another set paints molecules on a nerve fiber stretching toward a neighboring neuron, and yet another set highlights molecules at the synapses—and illumination by the same laser would reveal all of them distinctly.
Nie’s group published a demonstration of the Raman scanning technique in the January issue of Nature Biotechnology. For their beacons, Nie and his colleagues started with tiny, nontoxic particles of gold, of the kind used for some arthritis treatments. They modified the surfaces of the beacons with “reporter molecules” to boost their Raman signatures and then coated them with polyethylene glycol (PEG) to enhance their ability to circulate in the body. Finally they hooked them to antibodies designed to attach to a receptor normally overexpressed on tumor cells. By injecting a solution containing these Raman beacons into the bodies of mice with tumors and then illuminating them with a near-infrared laser through the skin, they were able to produce an image of the tumors and other sites to which the beacons bound.
A team at Stanford led by radiologist and biomedical engineer Sanjiv Gambhir reported a similar demonstration in a paper published online March 31 in Proceedings of the National Academy of Sciences. To demonstrate the technique’s multiplexing capability, Gambhir and his colleagues used three separate nano-beacons and imaged them simultaneously. The team also performed a whole-body scan of a mouse and showed that the technique could be made to work using carbon “nanotubes” as Raman beacons.
“Raman scanning still has a long way to go before we see clinical applications,” says Xiaoyuan Chen, a Stanford radiologist who has worked with the technology but is not part of Gambhir’s lab. Currently the technology cannot not be used to scan much deeper than a centimeter without running up against safety limits on laser intensity, and at such depths, exposure times of at least several seconds are needed.
But there are reasons to hope that as this technology develops, it will eventually enable much more informative images of cells and other structures in the body, the researchers say.
In addition to being able in principle to penetrate tissue and resolve multiple fine details simultaneously, Raman scanning could be modified for 3-D “tomographic” imaging as X-ray, MRI and PET technologies have been, to precisely pinpoint targeted cells or structures in the body. “The principle is very similar to fluorescence imaging, which is currently being adapted to do tomography,” Chen says. “It’s not optimal, but I think it’s doable.”
To do neurological imaging, nano-beacons would have to be developed that can cross the blood-brain barrier without harming brain cells. The technique also would have to be sensitive enough to image through bone. But Shuming Nie doesn’t consider this impossible. “Since the Raman signals are already in the near-infrared spectral window, I believe that [the technique] does have the potential to penetrate one to two centimeters through the skull.”
“It’s still in a very primitive stage now,” Chen says. “But I’m cautiously optimistic.”