Modular syntheses: Putting function first

Chemist Christian Schafmeister never passes up a chance to reference his favorite episode of The Simpsons. It’s the one in which Homer Simpson says alcohol is “the cause of, and the solution to, all of life’s problems.”

Schafmeister applies this idea not to beer, however, but to steric hindrance.

As a chemist at Philadelphia’s Temple University in 2007, Schafmeister had constructed a small set of building blocks made from proline-derived five-membered rings that he wanted to link together to create larger fused-ring structures. But Schafmeister’s small building block molecules were so tightly packed with carbons and side chains that coupling reactions to link them together simply didn’t happen. There just wasn’t space.

Then Schafmeister recalled a presentation at a peptide symposium where a researcher at Roche had described using an acyl transfer mechanism to create an antiretroviral drug for HIV using native chemical ligation. A graduate student in Schafmeister’s laboratory tried the acyl transfer in a last-ditch effort and held his breath.

It was a wild experiment, but it worked, Schafmeister says.

The transfer didn’t proceed through the thioester intermediate he had initially expected. Instead, the molecule folded back on itself to form a transient cyclic anhydride, which quickly rearranged itself into the desired amide bond. À la Homer Simpson, Schafmeister realized that steric hindrance was not only the cause of his problems but also the solution.

“We learned how to use crowding to temporarily protect the nitrogen and then allow it to couple when we wanted to. We learned how to use steric hindrance to make these really crowded structures, and that’s what cracked open [the process] for us,” Schafmeister says.

Schafmeister’s breakthrough allowed him to construct large, complex molecules dubbed spiroligomers from simpler building blocks. In January 2020, Schafmeister founded Ladder Bio, which was initially known as ThirdLaw Molecular, a biotech company focused on developing and optimizing the ladder-shaped, peptide-like spiroligomers to create new drug candidates and antibody stand-ins that can be synthesized quickly and cheaply for lateral flow assays and diagnostics. In late 2025, the company snagged $5.5 million in seed funding to expand its operations.

And Schafmeister isn’t the only chemist turning to modular methods of synthesis to explore vast regions of chemical space.

It’s a paradigm shift, according to John Moses, a chemist at Cold Spring Harbor Laboratory. Modular synthesis transforms molecular construction from a bespoke craft into a platform capable of generating thousands of complex molecules. “It’s about connectivity,” he says. “Taking modules and connecting [them] together reliably using good chemistry.”

The proponents of modular synthesis say it could accelerate efforts to discover new catalysts, materials, and medicines.

The origins of modularity

Modular, Lego-like chemistry is not such a new idea. Some chemists say it all began with a belt-shaped molecule called kohnkene.

The molecule comprises 12 laterally fused six-membered rings that form a rigid, double-stranded macrocycle. It acts as a precursor to zigzag carbon nanotubes and was first synthesized in 1987 by chemists Franz Kohnke and J. Fraser Stoddart. A subsequent retrosynthetic analysis revealed that the molecule could be built using four repetitive, stereoselective Diels-Alder reactions that provide precise stereoelectronic control.

After graduate student John Mathias joined the Stoddart lab at the University of Sheffield in the late 1980s, he found that slight tweaks in the identity and chirality of the starting building blocks let the team build a range of molecular rings, belts, and waves using a similar stepwise synthesis. Stoddart and Mathias realized that chemists could adapt the process to connect modular building blocks through predictable reactions. They dubbed the process “molecular Lego” in a 1992 Chemical Society Reviews paper.

This and subsequent papers from the Stoddart lab demonstrated that complex molecular systems could be constructed through design principles rather than bespoke target-oriented syntheses. Unlike earlier combinatorial approaches, which generated large libraries through often unpredictable reactions, modular synthetic strategies rely on standardized building blocks and highly reliable chemistries that allow complex, 3D molecules to be assembled in a controlled and reproducible way.

In the intervening years, synthetic chemists have explored strategies for encoding structure directly into molecular building blocks—developing rigid scaffolds, reliable bond-forming reactions, and modular fragments that could be assembled into increasingly complex architectures. But turning that idea into real molecules would require new chemistry.

Building a modular chemistry toolkit

When Schafmeister wanted to develop modular synthetic reactions, his first task was identifying the right building blocks. But that was only the starting point. A lot of time was spent puzzling out how exactly to join them together.

Dilemmas like Schafmeister’s echo across modular synthesis, says chemist Jeffrey Bode of the Swiss Federal Institute of Technology (ETH), Zurich. For modular synthesis to work, assembly reactions must function with a wide range of side chains and other molecular tweaks. A coupling must continue to work even after chemists swap in new fragments with different functional groups and chemical properties. “We don’t always have a really high confidence that a reaction is going to work,” Bode says. “You want to be able to very reliably say, ‘I can put these three building blocks together to make exactly this product.’ ”

Schafmeister’s solution—turn your problem into a key advantage—is one workaround. Bode has focused his attention on avoiding these problems entirely by creating “good enough” synthesis platforms. He has spent years designing chemoselective α-ketoacid–hydroxylamine (KAHA) ligation, which can join complex fragments through amide bond formation.

KAHA ligations allow Bode to readily synthesize both ubiquitin and tirzepatide. Because these reactions occur under mild, broadly tolerant conditions, he needs to neither reoptimize the conditions for each new substrate nor protect his functional groups. The goal, he says, is not to perfect individual syntheses but to build systems that continue to work as chemists swap in new building blocks.

At the University of Illinois Urbana-Champaign, chemist Martin Burke developed what he dubbed blocc chemistry as an iterative strategy for forming carbon-carbon bonds. These reactions rely on a system of stable boron-containing building blocks that can be linked together through a repeating sequence of Suzuki cross-coupling reactions. Burke stabilized the fragments using a small ligand called MIDA, which acted as a protecting group for the reactive boronic acids. After he removed MIDA, the fragment was ready for the next cross-coupling step. This method allowed chemists to stitch together complex carbon frameworks one piece at a time.

Burke says that his blocc chemistry forces him to focus on the function instead of the structure of the final molecule. “There are many different natural products that couldn’t look more different, but they can all bind to microtubules, immobilize them, and therefore kill cancer cells. Structurally diverse compounds can have the same function,” he says. “We can think of the blocc not as a structure but as a little piece of molecular function. If we learn how to mix and match them, we can get some emergent properties.”

Other chemists have focused on developing different types of modular reactions that assemble simple building blocks into complex molecular structures. At Cold Spring Harbor Laboratory, Moses has helped expand the family of “click” reactions popularized by Nobel Prize winners Carolyn Bertozzi and K. Barry Sharpless. While the original click chemistry relied on copper, Moses bent his efforts to further develop an analogous set of reactions called sulfur-fluoride exchange, or SuFEx, chemistry.

For Moses, the appeal of these reactions lies in their predictability and simplicity. “All proteins—all life—is assembled from a very few handfuls of building blocks. It’s the connection between them which is important,” Moses says.

Shoring up the scaffolding

Some modular syntheses begin with individual building blocks. But more recently, chemists have also begun creating larger frameworks that establish a molecule’s scaffolding and shape before adding functionality. It’s a strategy that can be used to generate large numbers of related compounds.

The shift is driven in part by a growing focus on molecular shapes. Many traditional drug molecules rely heavily on flat aromatic rings. Not only can these structures be challenging to synthesize, they also often bind imperfectly to the complicated, bumpy surface of a protein target. Rigid 3D scaffolds can position functional groups to more closely bind to irregular protein surfaces, improving both pharmacokinetics and pharmacodynamics.

The scaffold-first approach begins by establishing geometry and only later adds chemical diversity. At the California Institute of Technology, Brian Stoltz incorporates structural information from the outset using spirocycles. These twisted ring structures are connected by a single carbon atom, such that their compact 3D shapes lock functional groups into tightly defined structures.

Stoltz’s scaffolds give chemists a way to introduce shape and structural complexity while still maintaining synthetic control via a process called late-stage diversification.

“It’s an attractive approach because you exhaust your effort and synthesis of one compound. But instead of just getting one compound at the end, you might get 20 just by a single alteration at that last step,” Stoltz says.

At the University of Texas Southwestern Medical Center, Tian Qin and his research group study highly strained frameworks such as bicyclo[1.1.1]pentane, or BCP. BCP has a compact cage-like structure that is being investigated as a replacement for aromatic rings in drug discovery because it is a bioisostere of phenyl groups, meaning that it has similar chemical and biological activity to aromatic rings. The molecule keeps substituents at a regular distance from each other while also acting as a 3D scaffold.

The change from aromatics to BCP can alter how molecules interact with proteins and often improves properties such as solubility and metabolic stability.

“BCP has similar biological activity to many benzene analogs, but it has better pharmacological properties. And because the small ring system has such a nice stability, they cannot easily be metabolized by [cytochrome] P450,” Qin says, referring to a large family of liver enzymes that work to detoxify drugs and other chemicals.

In England, at the University of York, the lab team of chemist Peter O’Brien has begun designing libraries of small molecules that function as tiny, structurally diverse fragments that allow chemists to engineer stereochemistry and structure into a final scaffold from the start.

In fragment-based drug discovery, where the binding of a small molecule to its target is often very weak, O’Brien says the structural complexity of his fragment scaffolds gives better chances of engaging with a target. By making his molecular scaffold flexible, he can begin growing his compounds from a structurally defined starting point.

“You can easily attach things, you can vary things, making lots of analogs for testing using a robotic platform,” O’Brien says. “You can make molecules very quickly and test them very quickly.”

Across these approaches—Spiro cycles, BCPs, fragment scaffolds, and spiroligomers—the details may differ, but the underlying strategy is the same. Instead of treating structure as a variable that emerges from a synthesized molecule, chemists are increasingly hard coding structure in from the very beginning by using modular components that carry both geometric and stereochemical information with them.

Realizing the full potential of such modular systems, however, requires more than simple assembly. Chemists must also control the stereochemistry and geometry of these scaffolds with great precision.

Saving spatial relationships

The atoms of molecules interacting with biological targets must have specific shapes and structures. In modular synthesis, this structural information must survive every step of the assembly process.

It’s a requirement that places a heavy emphasis on stereochemistry, the 3D arrangement of atoms within a molecule. When chemists link modular fragments together, they must ensure that the building blocks don’t shift or morph as the molecule grows. If the stereochemistry is lost or scrambled during a reaction, the resulting structure could have the wrong shape and fail to bind its intended target.

Recent advances in stereospecific coupling reactions have made it easier to preserve structural information during modular assembly. At Boston College, James Morken has developed catalytic methods that use boron to transfer stereochemical information directly through reactions that form carbon-carbon bonds. This approach allows the initial geometry to be carried through the reaction and reproduced in the final product. Such reactions allow chemists to combine fragments while maintaining precise control over the 3D structure of the resulting molecule.

“It’s relatively straightforward to replace a boron atom with an OH group,” Morken says. “So you develop one catalytic borylation reaction, but you’re really developing a suite of transformations.”

Other researchers are tackling the challenge by embedding geometry directly into the building blocks themselves. For example, Daniel Blair, a chemical biologist at St. Jude Children’s Research Hospital, has explored strategies for designing fragments whose shapes and stereochemistry are predetermined before the assembly process begins.

“By modularizing the synthesis, the pieces become interchangeable,” Blair says. “It shifts your thinking away from specificity towards generality.”

Testing the approach

Many chemists are increasingly turning to robotic assistants to scale up the number of syntheses that they can run. Automation isn’t just adding fancy whizbang technology to the mix to make experiments look better, Schafmeister says. It’s already making a massive difference in productivity.

“We can make 60 of these spiroligomers in a week,” he says. “I made fewer than 60 in my entire 7 years as a graduate student.”

Schafmeister says he took pains to make the synthetic process as programmable as possible. His team wrote software that would tell them what molecules to make that week based on the building blocks they had in stock. The programs told him how many reagents to weigh out, and then the team programmed the robots to run the synthesis. He can run 12 reactions in parallel on his peptide synthesizer-style robots, repeating the process over and over again.

Burke is also working toward increasing the amount of automation used in his lab. He says it’s part of what attracted him to blocc chemistry in the first place. The modular processes are so simple and so reproducible that they didn’t need a human supervising every single step as would be required in artisanal synthesis. Burke even uses blocc chemistry as a learning tool for high school students. The limited sets of starting materials and repeatable reactions make robot-performed chemistry a natural outgrowth of modular synthesis.

“Historically, synthetic chemistry has been a very human-driven enterprise. You have highly trained chemists making decisions at every step along the way,” Burke says. “What we’re trying to do is develop chemistry where those decisions are no longer required. Where you can define a set of building blocks, define a set of reactions, and then a machine can just carry [the synthesis] out. Only a small number of people in the world can do complex small-molecule synthesis at the highest level. We want anyone to be able to do it.”

The development of modular synthesis represents a subtle but important shift in how chemists think about molecular complexity. Once the building blocks are designed, the assembly process becomes a matter of connecting pieces whose shapes are already defined.

In many ways, the strategy is not unlike software engineering: a small set of simple operations, repeated over and over in the right order, can generate highly complex systems. Similarly, modular synthesis relies on a limited set of reliable reactions to combine building blocks whose structures encode the information needed to produce complex molecular architectures.

“All of a sudden, these new building blocks become things you can plug in to medicinal chemistry approaches, or other syntheses, or whatever,” Stoltz says.

With such systems in place, he says, chemists can begin to generate enormous collections of molecules by mixing and matching modular components.

For much of modern organic chemistry, synthetic success has meant designing an elegant route to a single molecule. Modular strategies suggest a different model, and the result, say its proponents, is faster synthesis and faster discovery.

Ultimately, as modular synthesis becomes more reliable and diverse in its strategies, the range of functions chemists can access will also scale.

“The way we’ve made molecules for the last 100 years, it is really not very well suited for ultimately harnessing all the functional potential that these molecules possess,” Burke says. “We need to think differently about the entire enterprise so that we can put function first.”

Carrie Arnold is a freelance writer based in Virginia.