Conference highlighted how molecules are being programmed to carry out specific functions like robots across multiple size scales
On September 21, 2018, the Wyss Institute held its 9th Annual Wyss International Symposium that this year focused on advances in “Molecular Robotics,” a new and rapidly growing discipline that uses the molecular complexity, programmability, and self-assembling ability of DNA, RNA, proteins, and other molecules to design nanoscale and microscale devices that can sense, compute, and actuate, either autonomously or as a collective. Such devices could, for example, function as artificial viruses that can be harnessed for targeted drug delivery; serve as factories for producing chemicals, drug precursors or valuable nanoparticles; or interact with biological systems to control their behavior, record their activities, deliver therapies, or sense and signal the presence of diagnostic disease biomarkers in unprecedented ways.
Organized by the Wyss Institute’s Founding Director Donald Ingber, M.D., Ph.D., and the four leaders of its Molecular Robotics Initiative – William Shih, Ph.D., Peng Yin, Ph.D., Wesley Wong, Ph.D., and Radhika Nagpal, Ph.D. – the symposium provided an overview of how the new field is developing and a vision of how it could impact scientific research and human life in the future. This was showcased in thirteen presentations by speakers from a diverse array of specialties and institutions.
In his welcome remarks to an auditorium packed with attendees from over a dozen countries, Ingber recounted the development of Molecular Robotics as a new research field at the Wyss Institute that emerged through the convergence of ideas and efforts from its very different Programmable Nanomaterials and Bioinspired Robotics platforms. By combining approaches from both of these areas, Molecular Robotics has the potential to unlock entirely new opportunities to design, build, and control systems, as well as read out and store information at the nanoscale, all using biomolecules as building materials. “We’re creating a new technology wave – we’re engineering the future,” concluded Ingber before introducing the first of four presentation sessions.
The first session, entitled “Sense,” was introduced by Wyss Core Faculty member William Shih, and encompassed talks about how molecular nanodevices can be designed to sense elements of their environment – a central requirement for autonomous machines. Shih reported how his group devises new strategies to program DNA to self-assemble into different kinds of structures at increasingly large scales. By building smaller elements with DNA origami technology and finding approaches to overcome the initial nucleation barrier for the first ones to plug into each other, Shih has created entire microscale sheets of DNA that can bind to each other in crisscross patterns. This example represented his group’s largest nanostructures to date, and could find applications in highly sensitive detection tests and zero-noise (“hi-fi”) molecular computation.
Hendrik Dietz, Ph.D., Professor of Biophysics at Technische Universität München, then discussed how learning how to build complex molecular structures and machines from the ground-up is the key to achieving the “dawn of molecular robots.” His lab’s highly accurate molecular validation process encompasses state-of-the-art cryo-electron microscopy and computational advances that can model nanostructures in 3D. Using this extremely high-resolution visualization approach, and an impressive variety of DNA-nanotechnological methods, his team is able to build higher-order structures, such as different virus-like capsid structures that self-assemble from triangular DNA origami building blocks, and actual molecular motors like the “coffee mill,” which could one day interact with biological systems.
Yamuna Krishnan, Ph.D., Professor of Chemistry at the University of Chicago, wrapped up the first session by recounting how her team used DNA nanotechnology to build “CalipHluor,” a synthetic molecular sensor that detects both pH and calcium ion levels in intracellular organelles within cells in living worms with high accuracy. The device enabled her group to determine the function of a protein associated with severe Alzheimer’s disease – the protein was found to shuttle and help store calcium ions in organelles called endosomes and lysosomes, and calcium levels were significantly decreased in tissue samples from patients severe forms of this disease. CalipHluor sets the stage for the development of other such molecular detectors that could help probe the underlying causes of human diseases, and develop highly targeted treatments.
In the symposium’s “Compute” session, four speakers presented different approaches for enabling molecular robots to compute complex behaviors. Lulu Qian, PhD., Assistant Professor of Bioengineering at the California Institute of Technology, started out by showing the audience DNA “walkers” that are programmed to pick up, transport, and drop off molecular cargo in a testing area, and that work together to perform even more complex tasks. Her lab is also developing new computing approaches inspired by decision-making processes in neural networks to allow molecular robots to respond to their environments, compute specific inputs, and perform varying tasks, such as recognizing patterns that contain 100 bits of information. In addition, her team has pioneered a strategy of “tile displacement” that allows modular 2D structures to reconfigure themselves in response to specific signals by displacing large DNA-origami tile components from higher-order structures. “Our hope is that molecular programming languages will let anyone write their own molecular apps,” Qian concluded.
Similar complexity was also achieved in work reported by Justin Werfel, Ph.D., Senior Research Scientist at the Wyss Institute, but at a different size scale. He illustrated how principles developed by Radhika Nagpal’s research group in creating the Kilobot system, which works as a swarm of quarter-sized, electromechanical robots that demonstrates collective higher-order behaviors, can be translated into a molecular robotic system. The Wyss researchers used the “autocycling proximity recording” (APR) DNA nanotechnology developed in Peng Yin’s lab to program a swarm of DNA “robots” to move across a board covered with DNA-origami structures presenting one of two different molecular labels, record which labels they encountered, and recognize neighboring robots. These studies are the first steps to creating swarm molecular robots.
Rebecca Schulman, Ph.D., Assistant Professor of Chemical and Biomolecular Engineering at Johns Hopkins University, highlighted a different facet of molecular programming in her presentation. She focused on work on DNA-programmable hydrogels and their versatile abilities. By crosslinking individual polymers of a hydrogel network with synthetic DNA elements and adding actuating DNA-hairpins, the hydrogel can be induced to swell and combined into more complex multi-hydrogel arrangements that can change shape. The group devised a “hydrogel operating system” that incorporates ideas from cell biology and computer science to enable the hydrogels to sense and respond to changes in their environment with responsive changes in 3D form, thus creating “biomolecular controllers for active soft materials.”
As the last speaker of the set, George Church, Ph.D., Founding Core Faculty at the Wyss Institute and Professor at Harvard Medical School and MIT, gave an overview of different projects in his lab that are all based on multiplexed approaches to direct molecular motions and behaviors. His lab is developing DNA polymerases that, when attached to a nanopore, can be used as sequencing devices in highly multiplexed systems. He then introduced different uses of CRISPR systems to work as molecular machines that write information in the form of DNA into the genome of bacteria that can be used, for example, to create a short movie of a galloping horse that had been converted from a digital format into a sequence of DNA base pairs. Other projects included CRISPR-based approaches to engineer pigs’ organs to be more immune-compatible with humans, engineering synthetic AAV capsids for more effective gene therapies, a full human transcription factor library that can be deployed to engineer stem cells into virtually any mature cell type, and the Human Genome Project Write that aims to create virus-resistant cell lines.
As most existing robots’ purpose is to perform physical tasks, the third session focused on projects that aim to direct the motion and movement of molecules to achieve similar goals on the nanoscale. Itai Cohen, Ph.D., Associate Professor of Physics at Cornell University, explained his lab’s inspiration as “the idea that at the size scale of 50 microns, there’s an entire universe of organisms. The question we’re asking ourselves is, could we build one?” He then showcased a graphene origami machine composed of “paper” made from graphene sheets one atom thick whose folding can be directed by precisely designing the “creases” using materials with different strengths and properties. Such devices could be integrated with semiconductor manufacturing to create fully actuatable nanodevices for a variety of functions.
The next talk was given by David Baker, Ph.D., Professor of Biochemistry at the University of Washington, whose efforts in de novo protein design have led to the creation of synthetic proteins that go beyond what natural proteins can do in the same way that engineering synthetic DNA has gone beyond what natural DNA can do. His lab uses computational approaches to enumerate all possible 3D structures that can be generated from shorter protein sequences. Building on these insights, he showed newly created proteins with the ability to integrate into membranes and form pores that allow the passage of ions and small molecules. His group also found solutions for assembling multimeric protein molecules into various structures including hexagonal lattices, protein logic gates, nanocages, and artificial viral capsids that could bind to tumors in vivo, package genomes, serve as detection agents, or work as nanoscale delivery vehicles.
Finally, Khalid Salaita, Ph.D., Associate Professor of Chemistry at Emory University and Georgia Institute of Technology presented a DNA-based motor that is extremely fast, strong and sensitive. The motor is an example of a synthetic system that converts chemical energy into mechanical energy via an RNAse that cleaves RNA molecules attached to the surface on which it moves. This autochemophoresis can be scaled up or down to different sizes, and could be used for single-molecule detection in future sensing/diagnostic technologies.
The final session was dedicated to Molecular Robotics projects whose goal is to be translated into products that can solve problems outside the laboratory. Wyss Associate Faculty member Wesley Wong reported on his lab’s work to develop advanced “optical tweezers” that perform single-molecule analysis but also can be multiplexed. By combining centrifugal force microscopy with DNA-nanoswitches that are attached to molecules of interest, his team is investigating conformational changes and stabilities in molecules like DNA, performing up to 400 measurements in one centrifuge spin cycle. This approach is being used to study interactions between ligands and their receptors, monoclonal antibodies and their target proteins, and other molecular interactions, and could be harnessed in the development of biological drugs. Among other projects, Wong’s DNA puppeteered calipers can measure distances between molecular structures with atomic precision, and an NLISA (Nanoswitch-Linked Immunoabsorbent Assay) is capable of detecting biomarkers in whole human blood with femtomolar sensitivity.
David Zhang, Ph.D., Assistant Professor of Biomedical Engineering at Rice University and Hong Kong University of Science and Technology, outlined a molecular diagnostics approach for fast and multiplexed pathogen detection that could be performed in a doctor’s office or at a hospital patient’s bedside, rather than requiring long-lasting blood cultures. The platform integrates two components: a donut-shaped PCR fluidic chip that uses convection to cycle a reaction between 95 and 60 degrees centigrade and can be built much smaller and cheaper than a regular PCR machine; and a label-free microarray in which the presence or absence of 40 different pathogen targets can be detected with engineered DNA strand-displacement reactions within the chip. A battery-powered prototype that can work almost everywhere is currently being developed into a smaller instrument for use in settings outside the lab.
The session was closed by Wyss Core Faculty member Peng Yin, who began his talk with DNA toeholds, programmable nanodevices that can activate the transcription of reporter genes and have become the basis of multiple translational efforts. These include molecular diagnostics that can detect rare point mutations in patient DNA with high sensitivity, and “Ribocomputing” devices that allow cells to be programmed with computer-like logic. He then touched on a number of other developments from his lab with clear translational potential. The group’s DNA-PAINT super-resolution technology enables a simpler, more affordable and multiplexed way to visualize molecules at super-high resolution, and is currently being commercialized. He showed examples in which the approach can be used to see specific molecules in up to eight subcellular organelles, and to fingerprint up to ten molecules in complete tissues, like the retina. Finally, his lab’s autocycling proximity recording (APR) and primer exchange reaction (PER) methods use dynamic molecular robots to determine the geometry of single molecules and autonomously synthesize DNA in response to molecular signals, offering great promise for the detection of disease and the characterization of single molecules with high precision.
Ingber closed the symposium by thanking the speakers and commenting on how far Molecular Robotics has come despite its short life as a distinct research area, and its future possibilities. “We have seen today how a biophysical form of synthetic biology is opening up a totally new design space. It took genetic research 50 years to get to the first real-world applications. Molecular robotics is a really, really early discipline, but after only 10 years we are already seeing incredible applications in this field.”