When physicist John Porter of Johns Hopkins University shone a low-power infrared (IR) laser at a crystal doped with lanthanides, he probably had no idea of the revolution in biology and medicine he had ignited. It was 1961, and these rare-earth metals were being added to the phosphor-coated screens of cathode-ray-tube televisions where they emitted visible light when struck by high-energy electrons. Yet Porter had achieved a similar glow with long-wavelength light, instead of electrons—a surprise discovery that quietly opened the door to the extraordinary world of lanthanides.
The physicist posited that a lanthanide ion in the crystal had absorbed a couple of low-energy photons from the laser, became excited, and then released a single, higher-energy photon—visible light. He’d just witnessed the phenomenon of photon upconversion.
Soon afterwards, Francois Auzel of France Telecom, and Vitaly Ovsyankin and Petr Feofilov of the USSR Academy of Sciences, replicated the effect in lanthanide-doped glass, clearly demonstrating Porter’s result was not an anomaly. Then, in 1979, Jay Chivian and colleagues at Vought Corporation Advanced Technology Center pushed the photon frontier further. By pumping a crystal containing lanthanide ions with an IR laser and unleashing intense visible light, they had not only triggered photon upconversion, but also a highly nonlinear optical process, now known as photon avalanche.
Today, these remarkable optical phenomena are being witnessed in nanoparticles, doped with lanthanide ions, which can shine ever-more brightly for growing numbers of biomedical applications. Upconversion nanoparticles (UCNPs) can track and image cells, deliver drugs to tumors, stimulate brain cells to study neurological diseases—they’ve even been implemented in an analytical process, known as a lateral flow test, to detect covid-19.
“Early studies established that lanthanides really could absorb infrared photons and emit visible light,” highlights Xiaogang Liu of the National University of Singapore (NUS). “These fundamental insights still shape how we design upconversion systems for applications today.”
Keen to activate photon upconversion with as many different lanthanide ions as possible, Liu started work on UCNPs in 2006. By then, more and more researchers were adding the ions to nanoparticles, rather than glass and crystals, as the biomedical potential of these nanomaterials was lost on no one. Miniscule sizes of 10 nm to 100 nm, and the ability to tailor surface chemistry for biocompatibility, made the particles ideal for cellular take-up and intracellular imaging. Also, antibodies, drugs, and other ligands could guide the nanoparticles to cells or tissue for targeted therapy. What’s more, lanthanide-doped nanoparticles, in particular, are less toxic than rival fluorescing nanomaterials like quantum dots and silver nanoparticles.
But perhaps most important for biomedicine, these upconverting lanthonide-doped nanoparticles were activated with near-infrared (NIR) light, rather than the visible and ultraviolet (UV) light used to excite other nanomaterials. This was a big deal for many reasons.
First, longer wavelength NIR light is scattered and absorbed less by biological molecules, allowing it to penetrate tissues more deeply—up to a few centimeters. Second, most biological tissues do not fluoresce in NIR light, so background noise is reduced, and contrast is enhanced during imaging. Third, NIR photons carry less energy than visible and UV light, reducing the risk of cell damage.
So, in 2011 when Liu and colleague Feng Wang introduced a new design for lanthanide-doped nanoparticles that could fluoresce much more intensely than existing nanostructures and was compatible with a wide range of ions, excitement rippled through the UCNP field. With colleagues from the NUS, Chinese Academy of Sciences, and King Abdullah University of Science and Technology, Liu and Wang engineered a core-shell nanoparticle that, living up to its name, comprised a distinct core and shell that held gadolinium and other types of lanthanide ions.
When excited with NIR laser light, photon energy would migrate outwards from the core ions via a network of gadolinium ions to those in the shell, with intense luminescence being emitted. The researchers could also tune the luminescence between violet, blue, and green to red and yellow, by adjusting the specific lanthanide ions they had packed into the nanoparticle. So, with one diminutive particle, Liu and colleagues had thrown wide open the door to deep-tissue imaging as well as stable, multiplexed microscopy, and so much more.
Liu Xiaogang, at the National University of Singapore, has helped to pioneer the field of upconversion nanoparticles (UCNP). Photo credit: Yitong Zhao
Daniel Jaque, of the Universidad Autonoma de Madrid (UAM), at around the same time, was also fascinated by lanthanide-doped UCNPs and their potential to study cells, but he wanted to apply them to sensing as well as imaging. Alongside colleagues, he had developed nanothermometers based on the nanoparticles that would luminesce at different intensities according to the temperature within a cell.
“Early on, we realized that the luminescent properties of lanthanide ions could be strongly dependent on environmental conditions,” Jaque says. These could be changes in stress, pressure, and chemical composition as well as temperature.” So, his team started synthesizing lanthanide-doped nanoparticles. Several years later, and alongside Liu, Jaque also engineered UCNPs to measure changes in the viscosity within cells during chemotherapy to better understand the impact and effectiveness of this type of cancer treatment.
Scientists had also started exploring the use of UCNPs in optogenetics. They realized that by injecting these nanoparticles into the brain and shining NIR light from outside the skull, the particles could produce visible light to activate deep brain cells. The big appeal was that this method didn’t require optical fibers or surgery. In 2015, prolific nanophotonics developer Paras Prasad at the State University of New York at Buffalo, engineered UCNPs with novel core-shell architectures to shine brightly for this brain stimulation method. A few years later, Liu and colleagues from RIKEN Brain Science Institute injected the nanoparticles into the brain of a mouse, directed NIR light towards the skull to trigger light emission and activate neurons associated with a fear-response. The mice froze with fear—the approach had worked.
Eight years on, and very many developments have followed, underlining the weird yet wonderful nature of
lanthanide-doped nanoparticles. In a taste of what was to come, Liu and colleagues from Fuzhou University and Hong Kong Polytechnic recently engineered the nanoparticles to emit visible light—for up to 30 days—when irradiated with X-rays, rather than NIR. They embedded these persistent luminescence nanoparticles in rubber to create a flexible X-ray detector for X-ray imaging of curved 3D objects. “This could enable point-of-care X-ray detectors and mammography devices,” says Liu.
At around the same time as core-shell development, UCNP activity was picking up pace at Lawrence Berkeley National Laboratory’s Molecular Foundry. There, a team led by Bruce Cohen and P. James Schuck was intrigued by the UCNPs being synthesized by Liu and others. Little did they know how their developments would also shape the field.
“What caught our eye is that these nanoparticles were eight orders of magnitude brighter than your best two-photon dye molecules,” recalls Schuck. “They were seen by many as novel material…but we wanted to show they had properties people really cared about.”
Cohen, Schuck, and colleagues went on to synthesize ytterbium and erbium-doped nanoparticles that emitted bright, visible light, and that were perfectly photostable and stayed continuously lit. Knowing these qualities predisposed the UCNPs to single-molecule imaging, they coated the nanoparticles with a biocompatible polymer, internalized them in mouse fibroblasts, shone NIR laser light at the cells, and watched as the cells glowed bright green.
Development continued, and by 2014, the researchers had developed even brighter UCNPs by raising the concentration of lanthanide ions in nanoparticles way beyond what was thought to be useful. But their research was about to take a different turn.
Thulium-doped avalanching nanoparticles separated by 300 nm. Photo credit: Berkeley Lab/Columbia Engineering.
According to Schuck, his colleague Emory Chan had been writing computational models of nanoparticle synthesis for better-performing luminescent probes. Calculations suggested that a particular lanthanide ion—thulium—showed promise, so when initial imaging experiments supported this the researchers started to raise the concentration of thulium ions in single core-shell nanoparticles in earnest.
Within a couple of years, they could trigger powerful chain reactions of photon absorption and energy emission amongst the thulium ions with NIR light. And by 2021, as part of an international effort between Schuck, now at Columbia University, Berkeley Lab colleagues, and collaborators at the Korea Research Institute of Chemical Technology, Sungkyunkwan University, and the Polish Academy of Sciences, they had realized photon avalanching in a nanoparticle for the first time. The powerful light from these avalanching nanoparticles was also in the NIR range, and all in all, the results were dazzling.
To achieve photon avalanching in any material, a doubling of the irradiating laser intensity should increase the intensity of emitted light by 30,000-fold. But the avalanching nanoparticles met each doubling of exciting laser intensity with an emission increase of 80-million-fold, far exceeding that of existing lanthanide-doped nanoparticles and other nonlinear optical materials.
The phenomenon of photon avalanching had only ever been observed in millimeter to centimeter-scale crystals. But to witness this extremely nonlinear optical response at the nanoscale in UCNPs opened the door to higher resolution bioimaging, precision sensing, and more.
For example, during imaging with the UCNPs, bright light emissions take place at tightly focused spots, revealing fine, nanoscale structures such as viruses and organelles that conventional imaging can miss. Meanwhile, in a diagnostics context, the UCNPs can be precisely positioned to specifically target cancer cells at locations deep within tissue.
“We just love thulium,” says Cohen. “It’s got this particular organization of luminescent states that allows us to excite the excited states, so the energy of the photons keeps adding and adding, leading to avalanching. We got the cover of Nature with this—it was a big deal for us.”
Yet there was more. Within months of realizing photon avalanching, the researchers discovered they could switch the luminescence of the UCNPs on and off with certain intensities of NIR light indefinitely. They were shocked. “For years we’d thought lanthanide-doped nanocrystals didn’t photo-blink,” says Schuck.
Again, this was a big deal. Luminescent materials, including fluorescent dyes and proteins, which could be switched on and off with light had been widely used in super-resolution imaging as well as optogenetics and targeted drug delivery. However, many such light-sensitive, photo-switchable probes would break down over time, and typically required UV or visible light to work. These photostable, photo-switchable UCNPs offered a more stable alternative, and could be switched on and off with NIR light. “Not only were we using these gentler, longer wavelengths of NIR light to control photo switching, we were controlling it much better than anyone had done before,” says Cohen.
Since the landmark breakthroughs, myriad researchers have continued to engineer a mind-boggling array of UCNPs for an ever-growing number of applications. For example, Dayong Jin of the University of Technology Sydney, has been engineering UCNPs for more than a decade to target structures inside cells and to explore their inner processes in ever greater detail. With a purpose-built super-resolution microscope, his team, over many hours, imaged and measured the real-time temperature changes of single mitochondria in live cells. Abnormal temperature fluctuations signal mitochondrial stress, which is linked to diseases like cancer, diabetes, and neurodegenerative diseases. The researchers have also imaged small extracellular vesicles associated with cancer development.
“We need to know exactly where to send our nanoparticles within the cell, and then, where they actually end up,” says Jin. “We may then want to look at how sub-cellular organelles network together, how they respond to drugs and nutrition, or how they maintain their metabolisms in different conditions.”
Dayong Jin and colleagues are exploring how UCNPs can measure temperatures in cells. Their purpose-built system can super-resolve individual mitochondria at work and detect the upconversion emissions to calculate the temperature. Photo credit: Ou Xiangyu.
Jaque’s focus on sensing also continues. Working with Wenzhou Medical University colleagues, he has unveiled UCNPs that detect hypoxic tumor cells, to provide insight into cancer severity. As part of a team from UAM, the Lithuanian National Cancer Institute, and the University of Quebec, he’s also shown how UCNPs can enhance cancer photodynamic therapy by carrying photosensitizer molecules to tumors. The researchers created a UCNP-photosensitizer complex guided by stem cells that successfully destroyed tumors with just two NIR treatments.
And just this year, two teams—one led by Columbia researchers Schuck and Natalie Fardian-Melamed, and the other by Stanford University’s Jennifer Dionne—independently developed UCNP-based force sensors. The Stanford researchers discovered that applying force to nanoparticles doped with both ytterbium and erbium alters the intensity ratio of red and green light emitted when excited by NIR light. They actually fed polystyrene-encased UCNPs to Caenorhabditis elegans, measuring the micronewton (µN)-scale forces exerted by the nematode as it digested the sensor.
Similarly, Fardian-Melamed discovered that she could alter photon-avalanching UCNP luminescence intensity and color by applying a gentle force with the tip of an atomic force microscope. The researcher says the nanoparticles respond to forces from pico- to µN and could be used to detect events as subtle as synapses forming between neurons. “We have this huge dynamic range that means you could also use the same sensor to measure tissue-level forces,” she adds.
Despite the lightning-fast progress, biomedical application mostly remains bound to the lab. For example, in biosensing applications, researchers work with mice and rats because today’s microscopes can capture clear images at depths of up to a few centimeters in these small animals. But as Jaque puts it: “To do this with humans, we need a penetration depth of 10 centimeters—our current technology needs improving.”
However, change is afoot. Jin—named one of “100 rockstars of Australia’s new economy” by The Australian newspaper—has been collaborating with a microscope manufacturer to integrate his UCNPs into commercial super-resolution imaging. “Our goal is to demonstrate the first working system by the end of the year,” he says. “We cannot limit the technology to ourselves. We have to integrate upconversion [technologies] into commercial systems so that people who want to use the materials have a system to work on.”
In another encouraging development, UCNPs have reached diagnostic markets. In-vitro diagnostics firm Uniogen has packaged UCNPs into a labeling system for biochemical assays, lateral flow tests, and imaging. Meanwhile, Jin has also launched a lateral flow test for covid-19 with start-up Alcolizer. Instead of the usual antibodies, the test uses core-shell UCNPs to bind to coronavirus proteins in saliva and is said to be more than two orders of magnitude more sensitive than antigen-based tests. He’s excited by the results but notes that reproducibility—long a challenge in nanotechnology—hasn’t been easy to master.
“We have had no problem publishing beautiful papers, but when you talk to industry, everyone says [the nanomaterial] has to be the same every time it’s reproduced,” Jin says. “This has pushed us to practice [making the UCNPs] over and over again in the lab. We ask ourselves, where are the risks? How many ways can we make this wrong?”
Jin’s team had already developed a UCNP-based lateral flow test to detect prostate-specific antigen proteins, but re-engineering it to detect coronavirus took more than two years. Undeterred, they succeeded and are now translating the test technology to detect biomarkers for preeclampsia during pregnancy, as well as proteins and bacteria that can contaminate milk.
“You know, what really excites me these days is the commercialization opportunities,” says Jin. “I think it’s time to push our field massively towards applications, particularly in biomedicine.”
Rebecca Pool is a science and technology journalist based in the UK.
