Recent advances in miniature devices suggest the futuristic vision of medical nanobots – so-called ‘bloodstream submarines’ – patrolling our bodies to detect and repair injury or deliver drugs may be closer than we think, Michael Gross reports
Miniature submarines travelling through the bloodstream have starred in films including Fantastic Voyage (1966) and Innerspace (1987), but real-world nanotechnology is somewhat slow in making this happen. Only in the last few years has the field seen an increasing number of promising candidates for such feats. For instance, Tao Luo and Mingming Wu from Cornell University, US, have recently developed a microscopic swimming robot that is driven and steered by ultrasound and thus could conceivably be used in humans.1
Wu had initially tried to mimic the mechanics of bacterial flagella in an artificial swimmer. When Luo joined her lab as a postdoc, they tried a different approach, using a new laser lithography system, NanoScribe, to produce and optimise a novel microrobot. This system is based on the two-photon polymerisation approach, where a laser triggers polymerisation in specific places in a block of photo-reactive monomer material. This technique enables 3D-printing of structures with micrometre resolution with high precision, albeit quite slowly.
With this method, the researchers created a microscopic vehicle that looks like a twin rocket with a base shaped like the letter ‘m’. Each of the two arches of the ‘m’ carries a small gas bubble at the bottom, which is its motor. Upon irradiation with ultrasound, such bubbles vibrate and create vortices that drive the rocket forward.
Others have used one such bubble to drive microdevices, which means an additional mechanism is necessary to determine the direction. In Luo and Wu’s design, the two bubbles are behind outlet holes of different diameters. This means that the oscillating air-liquid interfaces are tuned to different notes; they can be addressed separately by different frequencies of ultrasound. This way, the device can be both propelled and steered by just one ultrasound transponder, and it could be guided by ultrasound imaging. Thus, the technique could use ultrasound wavelengths and intensities already found safe and approved for medical practice. The researchers are now working towards computer-controlled precision steering, which has to take into account the specific non-Newtonian fluid behaviour at the cellular scale.
The task ahead will be to make the swimmer itself biocompatible and ideally also biodegradable, such that one could apply many of them at a time without worrying too much about their fate in the body.
For drug delivery, you could have a group of micro-robotic swimmers, and if one failed during the journey, that’s not a problem. That’s how nature survives.
Mingming Wu Cornell University, US.
‘For drug delivery, you could have a group of micro-robotic swimmers, and if one failed during the journey, that’s not a problem. That’s how nature survives,’ Wu said. With this, Wu hopes to lead it to real-world applications in targeted drug delivery. This is an important issue especially in cancer treatment, where side effects are typically caused by cytotoxic drugs taken against cancer cells ending up in non-target tissues.
In a recent effort also using the NanoScribe technique, Zemin Liu and colleagues at the Max Planck Institute for Intelligent Systems, at Stuttgart, Germany, created a range of different, magnetically operated microdevices with rotational functionalities, such as stirrers.2 They used the two-photon polymerisation approach both for direct printing of desired shapes and for the production of moulds in which to shape the magnetic parts of the devices. The moulding enables the inclusion of materials that are not suitable for photo-induced polymerisation, including in this case the magnets. The authors anticipate this strategy paves the way for a wider range of materials and functionalities to be incorporated in microscale devices.
Magnetic force is also used in a highly original design of a swimming and rolling microrobot presented by the group of Ruike Renee Zhao at Stanford University. The researchers used the Kresling origami fold, which links two flat surfaces with a collapsible cylinder made of twisted panels.3 The unique movement could serve both propulsion and the delivery of cargo kept inside the cylinder.
A 5x40x70µm semiconductor-based walking robot has been produced with multiple actuator legs to drive it forward. The devices are comparable in size to a Paramecium cell.
Nanoparticles equipped with antibiotics have been attached to natural free-swimming single-cell algae. In animal tests, the swimming motion of the algae allowed the drug to access deep lung tissues that are otherwise hard to reach.
In contrast to the bubble-driven rockets, a separate effort also pursued at Cornell is sticking with mechanical actuators, seeking inspiration in the single cell eukaryote model organism, Paramecium. In 2020, the groups of Itai Cohen and Paul McEuen at Cornell presented a semiconductor-based walking robot with multiple actuator legs to drive it forward.4 With dimensions of 5µm thickness, 40µm in width and up to 70µm in length, the devices are comparable to a Paramecium cell.
The legs are made of thin layers of platinum on an inert titanium carrier. By targeting laser light at specific photovoltaic elements incorporated in the device, the researchers can send positive charges to the legs. These will attract counterions from the solution that will force the legs to bend. A big advantage of this technology is that it can be mass-produced with existing manufacturing devices for electronic devices. A single four-inch silicon wafer can produce up to a million such microrobots.
‘While these robots are primitive in their function, they’re not very fast, they don’t have a lot of computational capability – the innovations that we made to make them compatible with standard microchip fabrication open the door to making these microscopic robots smart, fast and mass producible,’ Cohen said.
A scanning electron microscope image shows a cell-size robotic swimmer that can be powered and steered by ultrasound waves.
Wu Lab, Cornell University/Tao Luo, Xiamen University
If the legs can drag a Paramecium-sized microrobot through space, one can inversely also imagine millions of such legs staying fixed on a surface and waving some sort of cargo past. In the human body, cilia are used in this manner to move body liquids including the cerebrospinal fluid in the brain.
Itai Cohen’s group recently used robotic legs as artificial cilia on a surface.5 The researchers created a device of around 1000 artificial cilia which can be powered individually by charge transfer like the legs of the microrobot. When they produce a coordinated wave, they can transport cargo at a speed of tens of micrometres per second.
The authors envisage this function could serve in small and inexpensive diagnostic devices which could transport samples without the need for a conventional pump. What the work shows, according to Cohen, ‘is that once you can individually address these cilia, you can manipulate the flows in any way you want. You can create multiple separate trajectories, you can create circular flow, you can create transport, or flows that split up into two paths and then recombine.’
Meanwhile, Joseph Wang at the University of California, at San Diego keeps his options open studying both biological and synthetic motors. In two recent studies, Wang’s team has used natural microalgae as motors for drug delivery. First his group used click chemistry – quick and easy coupling reactions, as pioneered by the 2022 Nobel laureates among others – to attach nanoparticles to natural free-swimming single-cell algae. The researchers had functionalised the nanoparticles with a membrane carrying antibiotics against bacterial lung infection.6
Upon administering the swimmers into the airways of test animals, the researchers confirmed that the ceaseless swimming motion of the algae ensured the uniform distribution of the drug in deep lung tissues that are otherwise hard to reach.
In the two-photon polymerisation approach, a laser triggers polymerisation in specific places in a block of photo-reactive monomer material. This enables high precision 3D printing of structures with micrometre resolution.
In theranostics, the same micro device can diagnose a problem and dispense the therapy. A theranostic approach to diabetes would measure blood sugar and release insulin as required, much like the healthy metabolism does.
In a related project, Wang’s group embedded a similar alga-driven nanoparticle construct in a pH-sensitive capsule, allowing it to be swallowed and deliver the hybrid swimmer to the small intestines. Tested in mice with a dye payload, and also with a model therapeutic substance, the method showed better distribution of the dye and better retention of the therapeutic in the gastrointestinal tract compared with a passive formulation of the nanoparticles without the motor.7
Another project from Wang’s laboratory uses enzymes encapsulated in a metal-organic framework (MOF) as a motor fuelled with glucose.8 In that case, the study addressed the role of electrode processes in electrochemistry, but similar, chemically driven micromotors are also under consideration for specific medical applications.
As medical applications of microrobots are coming within reach, strategic decisions towards product development have to be made. In a recent comment on the future of microrobotics, Wang pondered the question whether the focus should be on multi-functional or on single purpose robots.9
The trends in big robots are going towards complex, multi-functional machines, and similar developments can be imagined on the microscale, not least for so-called theranostics, meaning that the same micro device can diagnose the problem and dispense the therapy. Thus, a theranostic approach to diabetes would measure blood sugar and release insulin as required, much like the healthy metabolism does.
While complex robots will be developed for theranostics and other complex tasks involving multiple functions, Wang argues that for many more straightforward applications, the limited space in microrobots makes the combination of multiple functions unduly complex. Wang argues that ‘in most cases, many of these extra functions may be of limited use considering the main mission’ and eliminating them will make the design and operation more efficient.
Specialisation could go as far as choosing different propulsion modes in different physiological environments. Wang notes that urea could serve as fuel in the bladder, while in the stomach, acid could be used instead. Where several functions are needed, he suggests ‘it may be possible to combine different single-function microbots in swarms for performing different operational tasks.’ Wang’s team has also demonstrated operation of robot swarms in recent work.10
Despite the recent progress there is still a long way to go to lead the development of microbots into medical applications, Wang concludes: ‘A multidisciplinary collaboration in science, engineering and medicine and researchers trained with cross-disciplinary research skills are vital for addressing critical barriers for translating microbots into the clinical practice toward establishing a new era in the treatment of diseases.’
Top down vs bottom up
Over the past five decades, nanotechnology has followed two main routes to the dream goal of creating controllable machinery on the scale of living cells. The top-down route made things smaller and gave us the incredible technology revolution that broadly kept up with Moore’s law since 1975, doubling the number of transistors on a given space every two years. This enabled the miniaturisation from room-sized computers to smartphones as well as having sophisticated electronics incorporated in many other everyday objects, from contact lenses to cars.
Coming from the opposite direction, scientists working with molecules such as rotaxanes or DNA strands designed for their mechanical connectivity assembled more and more complex molecular machines and learned to manipulate individual molecules. Although much of this research has been on the curiosity-driven and playful side, this molecular approach has also enabled the sequencing of single molecules of DNA being threaded through a nanopore, along with revolutionary sensor technology based on molecules like single-domain antibodies and DNA aptamers (sequences selected for their specific, antibody-like binding abilities).
Neither has delivered on the early, somewhat naïve, promises of nanomachines patrolling our bloodstream to fight disease, eg by cleaning our arteries, delivering drugs, or taking samples for biopsies. Top-down technology mainly driven by the computer industry has remained quite different from biology in its designs and material choices, while many of the impressive achievements of bottom-up molecular nanotechnology are not compatible with industrial mass production methods. Recent advances in miniature devices suggest that the two sides are getting closer, however, and that we may get those promised bloodstream submarines after all.
1 T. Luo, M. Wu, Lab Chip, 2021, 21, 4095.
2 Z. Liu et al, Nat. Comm., 2022, 13, 2016.
3 O. Ze et al, Nat. Comm., 2022, 13, 3118.
4 M. Z. Miskin et al, Nature, 2020, 584, 557.
5 W. Wang et al, Nature, 2022, 605, 681.
6 F. Zhang et al, Nat. Mater.; https://doi.org/10.1038/s41563-022-01360-9
7 F. Zhang et al, Sci. Robot., 2022, 7, eabo416.
8 Z. Guo et al, Angew. Chem. Int. Ed., 2022, 61, e202209747.
9 J. Wang, Nat. Commun., 2021, 12, 7125.
10 K. Yuan et al, Adv. Mater., 2021, 33, 2100070.