The MATCHIT project context
What are the likely long-term implications of the construction of our simple artificial subcellular matrix made possible by the MATCHIT project? Following the general perspective from Rasmussen et al, (2011), one way of viewing the potential long-term impacts can be seen as follows.
Making living materials from nonliving materials and the implementation of living processes in other media both address and pose fundamental epistemological questions (Rasmussen, 1991). However, the potential usefulness of novel engineered living processes in mixed ChemBio-ICT media stem from the tantalizing properties of life itself. Living systems are characterized by energy efficiency, sustainability, robustness, autonomy, learning, local intelligence, self-repair, adaptation, and most importantly evolution through self-replication (Bedau et al., 2010 & Bedau et al., 2010a). Unfortunately, these are desirable properties current technology lacks, which over the last centuries have created an increasing variety of problems for our societies. It is not our place to make predictions about how future technology could become more alive, but instead we can summarize a vision that part of our scientific community shares. This vision is not yet science and more akin to science fiction.
First a little historical background: During the 19th century, the industrial revolution automated mass production in factories and a vast transportation infrastructure. In the latter part of the 20th century and the start of current century, the information technological revolution automated personal information processing in computers and the Internet. We believe the next major technological revolution will be based on an integration of information processing and material production, which is at the center of the MATCHIT project. Living organisms combine these processes seamlessly and biological organisms are still the only machines that can do this. To find out how they do this is in part why we seek to understand life.
One of our concrete long term visions about living technology is the construction of a personal fabricator (PF) (Gershenfeld, 2005 and Packard, McCaskill & Rasmussen, 2010) as an analog to the personal computer (PC). To get an idea of what it might imply to have a PF at your tabletop in a generation or so, imagine an advanced computer controlled 3D printer, which is able to control micro-fabrication, in part through molecular self-assembly and/or potentially atomic level controls, in order to build macroscopic structures of arbitrary complexity and composition. The PC and the Internet technology have enabled the individual to create and share information. Living technology has the potential to give the individual access to the design, sharing, and production of complex objects in a simple and sustainable manner. Again, the sustainable personal fabricator network is a vision and its implementation still relies on years of basic research and dedicated engineering at the interfaces between nanoscience, biotechnology, production technology, and information & communication technology. Perhaps the greatest science and engineering challenge is to combine the bottom up self-assembling and self-organizing nanoscale processes with top down mesoscale programmable production processes, which we have started to address in the MATCHIT project.
Some of the earlier and ongoing activities within the emerging Chembio-ICT area can be followed e.g. at the European Commission sponsored project web pages for PACE, ECCell and COBRA (Chembio-ICT) and in Amos et al., 2011. Common to these projects is an investigation of how to create and utilize living processes in a variety of hybrid bio-chemical, computational, and robotic systems. As our technology becomes more life-like, it also brings us a variety of new safety, environmental, and ethical challenges. These issues are addressed by the one of the research networks at the Initiative for Science, Society and Policy (ISSP, 2012).
Detailed MATCHIT project impact
Moving down to the technical details of the MATCHIT project, we can discuss the impacts along the lines of the proposed impacts (1) - (6) in the original MATCHIT description:
(1) Programmable information chemistry
This has been achieved in multiple ways ranging from DNA addressing of nano, meso and macro chemtainers as well as the MEMS channels. However, also programmed material production inside or at the surface of the chemtainers is also demonstrated. For more detials see the impact discussion for WPs1 - 6 below.
(2) Nanoscale chemical construction robots
WP1 demonstrates self-assembled nanoscale scaffolding that can encapsulate and release nanoscale materials as well as mitigate nanoscale production.
(3) Self-repair and self-replication
In WPs1-4 we demonstrate self-assembly and self-repair of all used chemtainers. In WPs 2, 3 and 5 (partly) externally enduced chemtainer replication (division) is demonstrated for oil droplets, fatty acid vesicles and water droplets.
(4) Medical applications
In particular WP2 and WP4 have utiliized the MATCHIT technology to develop potential drug delivery vehicles. In WP2 a set of hierachical organized vesicles has been developed, where different internal content that can be mixed for reactions, triggered by external signals (breakage of internal vesicle membranes). In WP4 the concept "Doctor in a Cell" has been further developed.
(5) A novel programming interface with conventional computers
WP6 has together with Microsoft developed a novel computing paradigma based on an extension of the brain calculus, which already containes the components of DNA- and Membrane computing. In WP5 it is demonstrated how this novel compiler can be translated into sequences of elemental electrode operations etc, which in the end conrols the boundary conditions for the self-organizing ChemBio components.
(6) Novel unconventional computing
The MATCHIT project resulsts demonstrate a variety of imbedded computings mainly through an integration of information procesing and material production as well as an integration of processes across multiple scales. The novel developed MATCHIT calculus could easily be expanded to include macroscopic commands e.g. for a 3D printer. This means that our new computing language within a single framework could bridge the controls for nanoscale self-asselbly and self-oreganization at the same time as e.g. sub-milimeter level 3D printing injection or construction of materials. For more details see the WP6 and 5 impact discussion.
Workpackage impact discussion
WP1 - DNA Nanocontainers
DNA nanocontainers, consisting of trisoligonucleotides with selectively addressable overhang sequences, are well defined chemtainers formed by self-assembly. Among the possible applications for such constructs are catalysis, binding, and sensing tasks. Consequently, one main objective in WP1 was the controlled uptake and release of cargo. Depending on the nature of the cargo modules, the linker between the module and the scaffold, as well as the lengths of the junction arms, modules may prefer to orient themselves into the interior or the exterior of the scaffold. While catalysis and small molecule binding tasks may benefit from an endopresentation of modules, binding to objects larger than the scaffold will benefit from their exopresentation. Another appealing target for the application of DNA nanocontainers is the surface of biological cells. Little is known to what extent the arrangement of individual cell-surface molecules and receptors such as integrins, G-protein coupled receptors, lectins, ion channels, and others are spatially organized by the cytoskeleton and to what extent these molecules are presented in a statistical arrangement. If the cytoskeleton “imprints” 2D spatial information to the outside of the cell, this arrangement may vary with the state of cell cycle as well as a function of cell differentiation. Nanoepitope screening through the arraying of the ligands of the cell-surface molecules via an artificial nanoscaffold may shed new light on the classification of cell types and tissues. If cell types including cancer cells can be indeed recognized by cooperative binding to a number of different modules arrayed on the scaffolds, it seems conceivable to load the scaffolds with additional “therapy” modules. Especially, modules with the function to alert the immune system, or, metallic nanoparticles used to dissipate heat upon receival of external radiation seem to be promising candidates for the stepwise implementation of medical nanomachines.
In WP1 we introduced several methods to externally control DNA nanoconstructs. External control may be applied by a change of temperature, light, pH, ionic strength, as well as the concentration of specific factors (e.g., heavy metal ions) for which selective recognition modules can be loaded onto the scaffolds. If one, however, considers a potential use of such nanomachines inside the human body all of the above control parameters are only applicable in limited cases. Most elegant in this context is, however, the possibility to switch conformations by the dissipation of heat from an inductive heating of metallic or magnetic nanoparticles attached to such constructs. It was shown that a gold nanocrystal (1.4 nm) can act as a nanoscale antenna for the receival of the magnetic component of GHz radio frequency radiation (Hamad-Schifferli et al. 2002).
WP2 - Heterophase Microcontainers
The precise control of chemtainer formation, stability, hierarchical architecture and content, as developed in this research program, could have far reaching effect on advanced personalized medicine. Precise temporal spatial control of medicine production, modification or release could both increase the efficacy of a drug and also the side effects. We explored the conditions that would allow chemtainers to release or share contents. We completed a basic study of how a series a salts effect the production and stability of POPC-based vesicles. These are the same vesicles we use for DNA addressing. The determination of salt effects will aid us in developing protocols for chemtainer content release and sharing. The data from this investigation was published in Nature Scientific Reports (Hadorn et al, 2011); and a peer-reviewed proceedings paper showing how this concept in the context of the MATCHIT goals can be used in the next generation of drug therapy was published in Biomedical Engineering Systems and Technologies (Hadorn and Hotz, 2011).
Regarding dissemination, WP2 has produced nine peer-reviewed publications with three more in preparation, ten talks and two posters at international scientific conferences. We succeeded in publishing our work with the controlled and reversible assembly of chemtainers with DNA addresses in the Proceedings of the National Academy of Sciences USA, a high impact factor journal intended for a wide scientific audience. In addition, we received a best presentation in synthetic biology award at the Artificial Life 13 international conference. The underlying basic technology developed by the MATCHIT team also has appeal to a diverse public, not only the specialist. M Hanczyc from SDU was invited to give a prestigious public talk on TED, which currently has over 500,000 views.
WP3 - Container supported reaction / production
The WP3 work has produced new knowledge about several aspects of chemtainers and related chemistry, which have potential for the development of new industrial applications.
First and most, it has shown that chemtainers can be used to alter the outcome of chemical reactions by providing a multiphase system, which is an important proof of principle for future use of chemtainers for the production of (bio-)chemicals and in general Synthetic Biology. The fact that this reaction needs an external inducement (light) could also be implemented toward the development of adaptable, self-repairing materials.
Second, the chemistry of ruthenium complex has been expanded to aqueous media, which per-se is interesting as ruthenium complexes have been gained attention during the last couple of years as potential therapeutic agents. Obviously, the knowledge of its function in chemtainers, such as vesicles, is promising as this type of system is often used a model for the investigation of membrane trafficking. Moreover, the possibility to carry out photochemical reactions in an aqueous environment could also lead to new energy harvesting systems with applications in the development of robust solar cells.
Third, we have successfully developed a new release strategy for caged chemicals during our investigation of the potential alteration of DNA tags. Here, the same exact chemistry could be used to protect chemicals sensitive to degradation, e.g., by hydrolysis, as they would either be provided as stable precursors, caged agents, which would be converted into active agents or be activated to react with other chemicals in a directed manner at any given time.
Finally, the functionalization of chemtainers with single chain hydrocarbon bola-amphiphiles could prove to be applicable for directing therapeutic nano-objects towards target tissues. The type of anchor could also find use in the development of DNA computing or of DNA nanostructures.
In terms of science, the project has been successful, as it has permitted us to directly (one PhD salary) and indirectly supervise the work of two other graduate students and two post-doctoral associates (by paying part of the WP3 leader salary). Indeed, these five young scientists have been able to collaborate within the outstanding groups at the European level during the project development and visits/stays they made to our collaborators. Such an involvement is essential for their future and the future of European science.
In terms of measurable outputs, we have published/will publish 4/3 papers, 2 commissioned articles, 2 book chapters and we (both the collaborators and the WP leader) have presented more than 15 times (mostly talks among them ¾ invited) at international conferences (among other GRC Origins of Life 2010, 2012; ACS 2012 meeting) with the support of the project.
WP4 - DNA computation of container addresses
In this project we ave designed and demonstrated the implementation of a coherent labelling system for Chemtainers, that enables control of specific interactions between Chemtainers and between the Chemtainers and a solid surface. This could enable information chemistry by introducing an addressable Chemtainers production system and interfacing it with electronic computers via MEMS technology with regulatory feedback loops. Thus, Chemtainers based on DNA-labelled heterophase droplets and vesicles, could form microscopic labelled reaction vessels, in which the next processing steps could be determined autonomously, by the system, as chemical reactions occur. The DNA-based addresses could enable parallel chemical programming in a new multilevel architecture through autonomous address modification and resolution at the container-container, container-surface, and container-molecule levels, providing a concrete embedded application for DNA computing.
These features will not only be an enabling technology for ―immersed systems IT applications in the life sciences, chemistry, and nanotechnology, but also promote a deeper understanding of the computational power of coupled production and information processes, as in biology, and provide a platform for building the more organic computers of the future.
The formal calculus developed with WP6 is able to capture the complexity of compartmentalized reaction systems such as colonies of possibly nested vesicular compartments. The overall system organization allows for the setup of programmable chemistry in microuidic or other automated environments. This work demonstrates a simple sequential programming language based on state-of-the-art microfluidic technology and DNA addresses. The formulated semantics could be used to derive the sets of all constructable chemicals and supermolecular structures that emerge from different underlying instruction sets. Since the proofs are constructive, they could be utilized to automatically infer control programs for the construction of target structures from a limited set of resource molecules.
The demonstration of potential autonomous operation, via products control of program execution, imply that the scope of this project could include autonomous decision making throughout the manufacturing/processing process, based on reactions' execution and products' yields.
WP5 - Electronically programmable chemical matrix
The electronically programmable chemical matrix has cemented a vision of sustainable autonomous chemical processing at the level of addressed chemtainers, under electronic programmable control. This is a major paradigm change from the traditional chemical approach of manual or robotically controlled reaction vessel processing and separation technologies for product cleanup. The ability to tag and jointly process collections of chemicals, and to process the tags themselves, makes the transition to a MIMD (multiple instruction multiple data) at the level of chemtainers possible. This is in fact the biological cell paradigm for parallel production and information processing. In MATCHIT we have only begun to scratch the surface of the potential embedded in this new approach.
The ongoing development and miniaturization of chemical microprocessor technology that was pursued in MATCHIT for general chemtainer processing is steadily bringing about the merger of chemical and computer information science. Early works concentrated on simple parallelization of a limited number of operations. In MATCHIT we have developed a complete set of functional operations to support general iterative chemical processing – of course at this stage still with significant practical limitations. The use of high density electrode arrays and channel networks as employed in MATCHIT will continue to pay dividends for information intensive and programmable lab on a chip applications in biotechnology and chemistry. We have now, through the connection with the MATCHIT calculus a connection with formal computation which should also serve to foster the interaction between computer science work and programmed chemical production.
Perhaps the most concrete first practical output of the electronic microfluidic development in MATCHIT is the application to a specific domain of programmed DNA editing in a subsequent EU project CADMAD. In CADMAD the current state of DNA information processing is compared to the period following Gutenberg’s printing press: what is missing is a general word processing machinery (DNA reuse) allowing users to edit up new DNA libraries from existing ones. The current developments using liquid handling robots are demonstrating the feasibility of this undertaking, and are slowly moving down in volume scale to below microliters (e.g. electrowetting millifluidics) but the true benefit will come from MIMD chemtainer level processing taking advantages of DNA tags for addressing chemtainers with DNA libraries for tagged specific combination and processing of picoliter or femtoliter sized “packets” of DNA. The fundamental primitives of droplet processing with on chip extraction, separation and reinjection developed in MATCHIT for chemtainer level processing, are already being employed in the CADMAD project to build DNA editing processors.
The more long-term implications of this technology for the proposed transition to sustainable personal fabrication, employing Living Technology principles, are also significant. Chemtainer level processing is essential to allow optimized chemical processes to feed into a microscopic fabrication process in which material systems use information to direct local fabrication, through a combination of self-assembly and reactive self-organization. The future connection between 3D printing technology and the processes explored in MATCHIT will open up new possibilities for fine-grained programmed construction that will allow end users to benefit from ongoing construction processes.
MATCHIT has also provided and is providing input to the ChemBioICT roadmap activity organized by the COBRA project. The important overarching goal of this roadmap is clearly also a by product and impact of the MATCHIT project:
“To develop by 2024 a portfolio of emerging ChemBioInfoFab technologies that allow humanly configured, autonomous control of massively parallel information processing and ongoing fabrication in macroscopic artifacts with molecular scale precision that are sufficiently generic and complex to serve as a toolkit for good solutions to a diverse range of technical challenges/problems.”
In particular, the development of addressed chemtainer level chemical processing under distributed electronic control is one of the paths to achieving this objective.
For society, it is important that our ability to mass produce products with high information density (such as computer chips and memory) is complemented with efficient on-demand production of information rich artifacts much close to the domain of application. MATCHIT is opening a path to enabling such information rich production.
The RUB team presented its project results at 26 meetings and conferences, most of them located in Europe, with an international audience. The team presented six posters, two of which were printed in the conference books of abstracts, and delivered seven publications and book sections.
One interview with J.S. McCaskill was published by the popular scientific press on an international level (Chemistry World).
J. S. McCaskill conducted a workshop for a panel of experts to produce a Science & Technology Roadmap for the whole ChemBioIT area, supported by the European Union through the Coordination Action COBRA, and funded by the multidisciplinary Future and Emerging Technologies (FET) programme under FP7, Bochum, 31.01.2013
RUB organised one MATCHIT project workshop in Bochum.
WP6 - IT architecture
WP6 has developed formal methods that integrate autonomous chemical reactions and molecular computing operations (including both DNA and membrane computing primitives) with external control. The frameworks have been tested for their automated chemical manufacturing capabilities, and we have developed a set of operating paradigms that help increasing product yield. Although the frameworks have been taylored to the specific MEMS hardware developed in WP5, the formalisms are much wider in scope and can be readily adapted to other hardware such as conventional microfluidics with automated pumps, high throughput screening facilities based on liquid handling robots and microarray technologies -- all being widespread in contemporary chemical manufacturing.
Recently, small scale personal manufacturing has seen a rapid increase in popularity with emerging technologies such as 3D printing. Personal manufactoring is excepted to have vast societal impact by reshaping our traditional infrastructure of production and distribution of goods. In place of central production and physical distribution of goods, personal manufacturing offers for virtual transfer of design followed by in-place, customizable production. MATCHIT in general, and specifically WP6, aims to open the domain of chemistry for distributed manufactoring. Such technology, when developed into a product, would allow for the small-scale production of personalized medicine directly at the place of demand, for example in local hospitals or pharmacies.
In terms of dissamination, WP6 has published 10 peer-reviewed articles in international high impact journals with four more articles in preparation. Scientific results where presented at 30 international conferences and workshops. In addition, WP6 researchers have engaged in teaching at the 1st COBRA Summer School on Biochemical Information Technologies. Work package leaders have supervised four M.Sc. students who worked directly on work package related issues. These young scientists have been able to collaborate with outstanding European research group. Such an involvement is essential for their scientific future and the future of European science.
M. Amos, Dittrich, P., McCaskill, J., & Rasmussen, S., Biological and chemical information technologies (2011) Proceedings from the 2nd European Future Technologies Conference and Exhibition 2011 (FET 11) - Procedia Computer Science 7, pp 56-60, Elsevier
M. Bedau, McCaskill JS, Packard N, and Rasmussen S, Living technology: Exploiting life’s principles in technology, (2010) Artificial Life 16: 89-97
M. Bedau, Hansen, P. G., Parke, E., and Rasmussen, S. 2010, eds., Living Technology: 5 Questions, Automatic Press/VIP 2010.
Chembio-ICT websites, see e.g. http://fp7-matchit.eu, http://www.cobra-project.eu or http://homepage.ruhr-uni-bochum.de/john.mccaskill/ECCell or http://www.istpace.org/Web_Final_Report/the_pace_report/index.html
M. Hadorn, E. Boenzli, P. Eggenberger Hotz, A Quantitative Analytical Method to Test for Salt Effects on Giant Unilamellar Vesicles Scientific Reports 1 (168) (2011)
M. Hadorn, P. Eggenberger Hotz, Encapsulated Multi-vesicle Assemblies of Programmable Architecture: Towards Personalized Healthcare. in Biomedical Engineering Systems and Technologies, edited by A. Fred, J. Filipe, & H. Gamboa (Springer-Verlag Berlin, Berlin, 2011), Vol. 127, pp. 141-151. (2011)
K. Hamad-Schifferli, J. J. Schwartz, A. T. Santos, S. Zhang, J. M. Jacobson. Nature 415, 152–155 (2002).
ISSP, 2012: Initiative for Science, Society and Policy, see http://science-society-policy.org under living technology.
Packard, McCaskill & Rasmussen, 2010: SPLiT 2010, Sustainable Personal Fabricator Network, see http://www.ecltech.org/LTFlagship/. The SPLiT vision was developed and lead by Packard, N., McCaskill, J., and Rasmussen, S.
S. Rasmussen, S. 1991, Aspects of Information, Life, Reality, and Physics, in Artificial Life II, ed. Langton, C., et al., Addison-Wesley, 1991, 767-773.
S. Rasmussen, A. Albertsen, H. Fellermann, P. Pedersen, C. Svaneborg and H. Ziock, Assembling living materials and engineering life-like technologies, Proceedings of the 13th Annual Conference Companion on Genetic and Evolutionary Computation (GECCO 11). ACM, New York (2011) p15.