The status of the key technologies

The technologies of biomolecular and chemical synthesis are now capable of fabricating a substantial range of complex, atomically precise structures. The most important of these are compact, polymeric structures (foldamers) and larger molecular scaffolds.

• Synthetic foldamers are now approaching the complexity of protein molecules, and can contain monomeric components with a wider range of functional properties.
• Protein engineering has recently reached the milestone of engineering new catalytic structures modeled on natural enzymes.
• Engineering molecular components that self-assemble to form new, complex crystalline materials has become routine.
• Structural DNA nanotechnology now enables the design and assembly of molecular scaffolds on a scale of millions of atoms and hundreds of nanometers.
• A rapidly developing design toolkit for self-assembly of diverse molecules and materials enables the construction of increasingly complex molecular systems.

Current research opportunities

These and related developments now make a range of experimental advances accessible. Some short-term goals and potential applications of the resulting technologies include the following:

• Demonstrate robust artificial foldamers that bind and stabilize complementary proteins.
  – Enables development of enzymatic catalysts for use in relatively harsh industrial process conditions.
• Demonstrate enzyme-like foldamers that bind and determine the activity of synthetic transition metal catalysts.
  – Enables development of highly stable and selective catalysts for the fine chemicals industry.
• Demonstrate self-assembled scaffolding structures that bind diverse components.
  – Enables the organization of nanoscale electronic. optoelectronic, and plasmonic components to form nanoscale sensors and electronic circuits.
• Demonstrate self-assembled scaffolds that promote and direct the growth of inorganic nanocrystals.
  – Enables production of atomically precise nanostructures with diverse materials and shapes for diverse applications in nanomaterials and nanosystems.

Middle-range objectives

Research opportunities today can open the door to the development of a next-generation technology platform that will, in turn, bring a new range of objectives into reach.

• The use of molecular scaffolds to bind and organize diverse components could be developed and elaborated to provide nanoelectronic fabrication methods for the post-Moore’s-law era.
• Devices that link multiple catalytic centers (analogous to polyketide and polypeptide synthases) could be developed and elaborated to provide “molecular assembly lines” that convert small feedstock molecules into high-value macromolecular products in a single, integrated process.
• A capacity for directing the growth of nanocrystals and other non-polymeric structures could be developed and elaborated to provide a capacity for building entirely new classes of complex, high-performance, atomically precise nanoscale components and systems.

Accelerators

Progress toward these objectives can be accelerated by increasing the capacity for innovative design, and for reducing innovations to routine practice. The greatest needs today comprise:

• Design-oriented software that integrates levels of description and physical analysis that range from quantum chemistry through molecular mechanics to the continuum mechanics of materials.
• Design-oriented data repositories that describe available nanoscale and molecular components and fabrication methods.

See also:

• Toward Integrated Nanosystems:
  Fundamental Issues in Design and Modeling [pdf]
• The Molecular Machine Path to Molecular Manufacturing (1):
  Foldamers and Brownian Assembly
• Modular Molecular Composite Nanosystems
• The Physical Basis of High-Throughput Atomically Precise Manufacturing

All ribosomes read genetic data as three-letter words that encode 20 standard amino acids (give or take a few anomalies). This is equally true of the ribosomes in deep-sea bacteria living at 120°C, and the ones in your thumb. This universal code has been a wall that bounds the scope of biosynthetic polypeptide engineering — until now.

Recent developments have cracked the wall by tweaking the code, but Jason Chin’s group in the UK has blasted a wide hole by expanding the address space.

From the abstract of a paper soon to be published in Nature:

[E]very triplet codon in the universal genetic code is used in encoding the synthesis of the proteome….Here we synthetically evolve an orthogonal ribosome (ribo-Q1) that efficiently decodes a series of quadruplet codons…. By creating mutually orthogonal aminoacyl-tRNA synthetase–tRNA pairs and combining them with ribo-Q1 we direct the incorporation of distinct unnatural amino acids…. it will be possible to encode more than 200 unnatural amino acid combinations using this approach.

---

H Neumann et al., “Encoding multiple unnatural amino acids via evolution of a quadruplet-decoding ribosome”, Nature (early online publication).

I’ve selected some examples (image above) to illustrate the scope of these methods. Each of these amino acids (some highly unnatural) has already been used as a building block in ribosomal polypeptide synthesis. Together, they provide a glimpse of the vast new world now opening to molecular engineers. Polypeptides (of the sort usually called “proteins”) are already a family of versatile, high-performance engineering polymers, and an expanded set of building blocks can be exploited to increase thermodynamic stability, extend useful functionality, facilitate self assembly, and enable more systematic design.

Realizing this potential for expanding the scope of protein engineering will require extensive development of new tools, including new aminoacyl-tRNA synthetase–tRNA pairs. Because these are themselves proteins, there will be increasing opportunities for bootstrapping, using the new tools to facilitate development of those that follow. For example, could task-specific side chains (perhaps resembling PNA oligomers) facilitate the development of new aminoacyl-tRNA synthetases? There are complex constraints, but wide room for maneuver.

• Cell-free synthetic biology
• A High-Performance Polymer for Nanosytems Engineering
• Modular Molecular Composite Nanosystems
• Macromolecular Modeling for Molecular Systems Engineering

Here is an attempt to give a useful answer that takes account of these unknowns. My advice centers on fundamentals, outlining areas of knowledge are are universally important, and offering suggestions for how to approach both specialized choices and learning in general. It includes observations about the future of nanotechnology, the context for future careers.

Learn the fundamentals, and not just in science

The most basic requirement for competence in any physical technology is a broad and solid understanding of the underlying physical sciences. Mathematics is the foundation of this foundation, and basic physics is the next layer. Classical mechanics and electromagnetics are universally important, and the concerns of nanotechnology elevate the importance of thermodynamics, statistical mechanics, and molecular quantum mechanics. A flexible competence in nanotechnology also requires a sound understanding of chemistry and chemical synthesis, of biomolecular structure and function, of intermolecular forces, and of solids and surfaces.

These are important areas of science, but science is not technology. As I’ve discussed in “The Antiparallel Structures of Science and Engineering”, science and engineering are in a deep sense opposites, and must not be confused. Nanotechnology today is a science-intensive area of engineering, largely because the problem of designing a nanostructure is often overshadowed by the problem of finding, by experiment, a way to make it.

This has implications for choosing a course of study.

Engineering and progress in nanotechnology

A measure of progress in nanotechnology is growth of the range of physical systems that can be designed and debugged without extensive experimentation. As a basis for implementing nanoscale digital systems, commercial semiconductor fabrication provides a predictable design domain of this sort, and some areas of structural DNA nanotechnology have become almost as predictable as carpentry.

Computational tools are in a class of their own, an area of immaterial technology that applies to every area of material technology. It’s important to understand the capabilities and limitations of these tools, and extending them makes a strategic contribution to progress. Computational tools tools are often the key to transforming reproducible processes and stable structures into reliable operations and building blocks for engineering. Today, better design tools are the key to unlocking the enormous potential of foldamers and self assembly as a basis for implementing complex nanosystems.

Competence in engineering — and understanding how science can support it — requires study of design principles and experience in solving design problems. As with physics, some lessons apply across many domains. Because nanotechnology relies on innovations in macro- and micro-scale equipment, engineering education has immediate and strong relevance. Looking forward, the growth of nanosystems engineering will open increasing opportunities for researchers with backgrounds that provide both the scientific knowledge necessary to understand new nanotechnologies and the engineering problem-solving abilities necessary to exploit them.

Students aiming to pioneer in directions that can open new worlds of nanotechnology should learn enough of both science and engineering to solve crucial problems at the interface between them. The most important of these is the problem of recognizing and developing the means for systematic engineering in new domains, extracting solid toolsets from the flood of novelty-oriented nanoscience.

In considering all of the above, keep in mind that the general direction of nanotechnology leads toward greater precision at the level of nanoscale components, making products of increasing complexity and size, implemented in an increasing range of materials. Molecular-level atomic precision has widespread applications in nanotechnology today, and already provides components with the ultimate precision at the smallest possible length scale. I expect that the road forward will increasingly focus on extending these atomically precise technologies toward greater scale, complexity, and materials quality. I recommend courses of study that prepare for this.

Choosing topics and ways to study them

In both science and engineering, a good methodology for selecting an ideal course of study would be to survey a course catalog and note which classes appear in lists of prerequisites for advanced classes in relevant areas of science and engineering. This indicates areas where it is important to study and master the content.

Courses toward the periphery of this network of prerequisites are good candidates for a different mode of study, a mode aimed at understanding the problems an area addresses, the methods used to solve them, and how those problems and methods fit in with the rest of science and technology. I discuss this mode of study in “How to Learn About Everything”. It builds knowledge of a kind that can help a student choose topics that call for deeper, focused learning, and it can later help greatly in practical work — scientists and engineers with broader knowledge will see more opportunities and encounter fewer unanticipated problems. These advantages mean fewer days (months, years) lost and greater strides forward.

Choosing institutions

Beyond topics of study, I’m also asked to recommend universities and programs. It’s difficult to give a specific answer, because a good choice depends on all of the above, and because for each of many areas of science and technology, there are many possible institutions, programs, and research groups. I can only advise that students facing this decision first consider their objectives, and then to look for institutions and people able to help them get there. In particular, universities must either offer a degree program that fits, or provide the flexibility to make one. I found a home in MIT’s Interdisciplinary Science Program (which I can’t recommend, because it no longer exists).

In undergraduate studies, the general breadth, orientation, and quality of a school is more important than any focused undergraduate program that it is likely to have.

Early involvement in research of almost any kind has a special value: It can provide knowledge of kinds that can’t be learned from reading, from classes, or even from lab courses. Pay special attention to research that studies atomically precise structures of significant size and complexity. If that research has an engineering component — designing and making things — so much the better.

• How to Learn About Everything
• How to Understand Everything (and Why)
• The Antiparallel Structures of Science and Engineering

Chemists understood the atomic structure of molecules in the 1800s, yet many say that Einstein established the existence of atoms in a paper on Brownian motion, “Die von der Molekularkinetischen Theorie der Wärme Gefordete Bewegung von in ruhenden Flüssigkeiten Suspendierten Teilchen”, published in 1905.

This is perverse, and has seemed strange to me ever since I began reading the history of organic chemistry. Chemists often don’t get the credit they deserve, and this provides an outstanding example.

For years, I’ve read statements like this:

[Einstein] offered an experimental test for the theory of heat and proof of the existence of atoms….
[“The Hundredth Anniversary of Einstein’s Annus Mirabilis”]

Perhaps this was so for physicists in thrall (or opposition) to the philosophical ideas of another physicist, Ernst Mach; he had odd convictions about the relationship between primate eyes and physical reality, and denied the reality of invisible atoms.

Confusion among physicists, however, gives reason for more (not less!) respect for the chemists who had gotten the facts right long before, and in more detail: that matter consists of atoms of distinct chemical elements, that the atoms of different elements have specific ratios of mass, and that molecules consist not only of groups of atoms, but of atoms linked by bonds (“Verwandtschaftseinheiten”) to form specific structures.

When say “more detail”, I mean a lot more detail than merely inferring that atoms exist. For example, organic chemists had deduced that carbon atoms form four bonds, typically (but not always) directed tetrahedrally, and that the resulting molecules can as a consequence have left- and right-handed forms.

The chemists’ understanding of bonding had many non-trivial consequences. For example, it made the atomic structure of benzene a problem, and made a six-membered ring of atoms with alternating single and double bonds a solution to that problem. Data regarding chemical derivatives of benzene indicated a further problem, leading to the inference that the six bonds are equivalent. Decades later, quantum mechanics provided the explanation.

The evidence for these detailed and interwoven facts about atoms included a range of properties of gases, the compositions of compounds, the symmetric and asymmetric shapes of crystals, the rotation of polarized light, and the specific numbers of chemically distinct forms of molecules with related structures and identical numbers of atoms.

And chemists not only understood many facts about atoms, they understood how to make new molecular structures, pioneering the subtle methods of organic synthesis that are today an integral part of the leading edge of atomically precise nanotechnology.

All this atom-based knowledge and capability was in place, as I said, before 1900, courtesy of chemical research by scientists including Dalton, van ’t Hoff, Kekulé, and Pasteur.

But was it really knowledge?

By “knowledge”, I don’t mean to imply that universal consensus had been achieved at the time, or that knowledge can ever be philosphically and absolutely certain, but I think the term fits:

A substantial community of scientists had a body of theory that explained a wide range of phenomena, including the many facets of the kinetic theory of gases and a host of chemical transformations, and more. That community of scientists grew, and progressively elaborated this body of atom-based theory and technology to up to the present day, and it was confirmed, explained, and extended by physics along the way.

Should we deny that this constituted knowledge, brush it all aside, and credit 20th century physics with establishing that atoms even exist? As I said: perverse.

But what about quantitative knowledge?

There is a more modest claim for Einstein’s 1905 paper:

…the bridge between the microscopic and macroscopic world was built
by A. Einstein: his fundamental result expresses a macroscopic quantity — the coefficient of diffusion — in terms of microscopic data (elementary jumps of atoms or molecules).
[“One and a Half Centuries of Diffusion: Fick, Einstein, Before and Beyond”]

This claim for the primacy of physics also seem dubious. A German chemist, Johann Josef Loschmidt, had already used macroscopic data to deduce the size of molecules in a gas. He built this quantitative bridge in a paper, “Zur Grösse der Luftmoleküle”, published in 1865.

• A Map of Science
• How to Learn About Everything

Synthetic biology doesn’t require cells, and in several ways, cells are liabilities.

Cells can make engineering difficult. Cell membranes and bacterial walls stand between new genes and the machinery needed to transcribe and translate them. They are barriers to liberating gene products. They contain systems that are complex products of eons of evolutionary history, not systems streamlined to simplify engineering. They are easily poisoned by what would be, to us, useful raw materials and products.

The state of the art in cell-free synthetic biology is already advanced, and moving forward rapidly:

Time and again, decreasing the dependence on cells has increased engineering flexibility with biopolymers and self-copying systems….

Current in vitro methods for synthesizing proteins and evolving protein, nucleic acid, and small-molecule ligands will be improved to accelerate production of new reagents, diagnostics, and drugs. New methods will be developed for synthesizing circular DNAs, modified RNAs, proteins containing unnatural amino acids, and liposomes.

---

Forster and Church, “Synthetic biology projects in vitro”.

A glimpse of some recent developments:

Cell-free systems offer a unique platform for expanding the capabilities of natural biological systems for useful purposes, i.e. synthetic biology. They reduce complexity, remove structural barriers, and do not require the maintenance of cell viability. Cell-free systems, however, have been limited by their inability to co-activate multiple biochemical networks in a single integrated platform. Here, we report the assessment of biochemical reactions in an Escherichia coli cell-free platform designed to activate natural metabolism, the Cytomim system….

---

Jewett et al., “An integrated cell-free metabolic platform
for protein production and synthetic biology”.

Atomically precise self-assembly of complex structures can be engineered by providing for multiple binding interactions that

1. Cooperate to stabilize the correct configuration, in a thermodynamic sense, and
2. Do not stabilize any other configuration, in a kinetic sense

Roughly speaking, in the correct configuration, the parts fit together to allow all the binding interactions to operate simultaneously, and the system doesn’t get stuck in other configurations. It’s easy to see how weak interactions and cooperative binding can implement these conditions, but there are alternatives.

As I’ve discussed elsewhere, recent advances in biomimetic self assembly based on peptide and nucleic acid polymers provide a platform for developing complex, functional self-assembled systems, and in the right environments, some of these structures can be surprisingly robust. However, most of their characteristic binding interactions (hydrogen bonds, hydrophobic interactions, van der Waals interactions in well-packed structures, etc.) are weak in terms of both binding energy and mechanical strength.

Proteins structures, however, often include disulfide bonds (R₁–S–S–R₂), and these are covalent and strong. Their role in protein folding illustrates a key point:

Disulfide bonds can shuffle among different pairings through thiol/disulfide exchange,

a process that can be fast in the presence of R–S^– ions. A well-folded structure will strongly favor correct pairings by holding a momentarily displaced R–S^– in a position to reform the bond. In thermodynamic terms, this decreases the entropy cost of the bond-forming reaction, and in kinetic terms, it increases the effective concentration that drives the forward reaction, typically accelerating it by an large factor (> 10³). Exchange can be shut off by decreasing pH or removing free thiols from the folding environment.

The formation and hydrolysis of boronate esters can play a similar role in artificial self-assembling systems. A sample chapter from Boronic Acids (2005, posted by Wiley-VCH Verlag) provides an extensive discussion of the chemistry of boronic acid derivatives; it notes that boronic acids (at high pH, as hydroxyboronate anions) react with diols to form boronate esters with forward rate constants in the 10³ – 10⁴ M ^–1s^–1 range. Hydrolysis is likewise fast. Boronate esters can be stabilized by reducing pH or removing water. They, and boronic acids, are generally biocompatible, and have even been developed as drugs, where they serve to bind carbohydrate moieties.

Here are some recent papers on self assembled systems that discuss boronic acid chemistry, along with other covalent chemistries of similar utility:

• “An Iminoboronate Construction Set for Subcomponent Self-Assembly”
• “Complex Systems from Simple Building Blocks via Subcomponent Self-Assembly”

And a dissertation:

• “Self-Assembly of Boron-Based Supramolecular Structures” [pdf]

Self assembly need not be biomimetic.

• Effective Concentration in Self Assembly,
  Catalysis, and Mechanosynthesis (1)
• Effective Concentration in Self Assembly,
  Catalysis, and Mechanosynthesis (2)
• Self-assembling nanostructures: Building the building blocks

@ Eniac — Thanks, you raise several important questions.

Regarding readiness to build extended, self assembling structures, yes, I think that the existing fabrication abilities (that is, the range of molecular structures that can be synthesized) are now more than adequate. The bottleneck is design software, including the development of rules that adequately (not perfectly) predict whether a given design satisfies a range of constraints. These include synthesis, stability, solubility, and sufficiently strong net binding interactions.

As for specifying face combinations that would result in unique binding, this becomes easier with increasing face size, and more difficult with the number of simultaneously exposed faces. Hierarchical assembly can address both of these, but the most practical schemes require the ability to convert reversible binding interactions into irreversible ones. One approach is to introduce covalent linkages after assembly of the intermediate blocks lower in the hierarchy of sizes. There are several ways to do this.

The problem of enabling motion between self-assembled components can be addressed at the level of interactions between assemblies that are held together by (for example) a combination of large-scale complementary shapes and non-contact colloidal binding interactions.

Flexible hinges in self-assembled structures are also practical, as shown by natural systems. Protein engineers have successfully designed structures that undergo conformational switching.

Downstream, there’s a continuum of assembly approaches that spans the range between free Brownian motion, constrained Brownian motion, and more macro-machine-like devices (discussed in “From Self-Assembly to Mechanosynthesis”, and Motors, Brownian Motors, and Brownian Mechanosynthesis).

You are right that the relative sizes of machines for manipulating matter and for manipulating information become similar (or reversed) at the nanoscale, relative to what we are familiar with in today’s macro-machine, micro-computer world. The resulting design constraints can be met by a various combinations of several techniques, including

• Offloading computation to conventional computers that direct what would typically be large numbers of nanosystems (a good early solution).
• The same single-computer / multiple machine approach with nanosystems for both operations.
• Extensive use of hard automation, in which repetitive operations require no computation at all.

Regarding the last point above, this is how high-throughput manufacturing works today. I’ve discussed this in posts with videos of machines in action: “High-Throughput Nanomanufacturing: Small Parts” and “High-Throughput Nanomanufacturing: Assembly,” with a more quantitative discussion of “molecular mills” on E-drexler.com).

This post is prompted by a set of interrelated advances in chemistry that hold great promise for advancing the art of atomically precise fabrication. In this post, I’ll describe an emerging class of modular synthesis methods for making a diverse set of small, complex molecular building blocks.

The road to complex self-assembled nanosystems starts with stable molecular building blocks, and the more choices, the better. Self-assembly and the folding of foldamers are similar processes: They work when parts fit together well, and in just one way. Having building blocks to choose from at the design stage will typically make possible a better fit, resulting in a denser, more stable structure.

Building blocks for building blocks for building blocks

I often think in terms of four levels of molecular assembly:

• Specialized covalent chemistry to synthesize monomers
  (~1 nm)
• Modular covalent chemistry to link monomers to make oligomers
  (~10 nm length)
• Intramolecular self-assembly (folding) to make 3D objects
  (< 10 nm diameter)
• Intermolecular self-assembly to make functional systems
  (~10–1000 nm)

Recent developments are blurring the first level into the second, however, because new modular chemistries can make complex structures that can serve a monomers at the next level of assembly. Perhaps the most outstanding example comes from Marty Burke’s lab, which has pioneered a new, combinatorial methodology for piecing together small molecules of enormous diversity. From the lab website:

To most effectively harness the potential impact of complex small molecules on both science and medicine, it is critical to maximize the simplicity, efficiency, and flexibility with which these types of compounds can be synthesized in the laboratory.

…the process of peptide synthesis is routinely automated. As a result, this highly enabling methodology is accessible to a broad range of scientists. In sharp contrast, the laboratory synthesis of small molecules remains a relatively complex and non-systematized process. We are currently developing a simple and highly modular strategy for making small molecules which is analogous to peptide synthesis…

Our long term goal is to create a general and automated process for the simple and flexible construction of a broad range of complex small molecules, thereby making this powerful discovery engine widely accessible, even to the non-chemist.

In outline, the Burke group’s method exploits iterative Suzuki-Miyaura coupling, a mild and increasingly general technique in which (in Burke’s approach) carbon-carbon bond formation plays the role of amide bond formation in making peptides. In peptide synthesis, suitably-protected amino acids are iteratively coupled, deprotecting the terminal amine at each step. In Burke’s method, suitably-protected boronic acids play the analogous role.

The key advance is the N-methyliminodiacetic acid (MIDA) protecting group, a trivalent ligand that rehybridizes the boron center from sp² to sp³, thereby filling and blocking access to the open p orbital that makes trivalent boron compounds so wonderfully, gently reactive. The resulting complex is stable to a wide range of aggressive conditions, including powerful oxidants and strong acids. It can be removed, however, by an aqueous base (e.g., sodium bicarbonate in water).

For more information, good places to start are the Burke lab’s research overview page, and the MIDA boronate technology spotlight page at Sigma-Aldrich, which also provides off-the-shelf MIDA-protected building blocks. Sigma-Aldrich offers a larger universe of boronic acids and boronic esters, as does CombiPhos Catalysts. It’s worth looking through one of these documents to get a gut sense of what’s now available. Impressive diversity, compared to the 20 standard amino acid side chains.

(For a general perspective on this direction of development, see “Controlled Iterative Cross-Coupling: On the Way to the Automation of Organic Synthesis”, Angew. Chem. Int. Ed. 2009.)

More than a protecting group

The MIDA boronate ester is an example of a broader class of structures that are important in their own right. The demands of organic synthesis have brought forth a vast range of commercially available boronate esters (see links above), and this investment gives a free ride to scientists aiming to exploit them as building blocks. As linkers for self-assembled structures, boronate esters are both extraordinary and underexploited.

Relying a little less on hydrogen bonds, and a little more on bonds that can hold a self-assembled solid together at 600°C — dull red heat — could increase the robustness of self-assembled products. A fast, reversible, aqueous, biocompatible boron chemistry opens a door.

More later.

See also:

• Modular Molecular Composite Nanosystems
• A High-Performance Polymer for Nanosytems Engineering
• CAD for Nanoengineering: DNA, proteins, and search-intensive design