Using visible light to make pharmaceutical building blocks

The class of compounds, called azetidines, had been previously identified as a good candidate to build therapeutic drugs, but the compounds are difficult to produce in chemical reactions. Now, a team led by University of Michigan chemist Corinna Schindler has developed a method to produce a specific class of azetidines called monocyclic azetidines using visible light and a photocatalyst. Their results are published in the journal Science.

Approximately 60% of pharmaceutical drugs contain building blocks in the form of compounds called nitrogen heterocycles. Nitrogen heterocycles are structures of atoms organized in a ring that contain at least one nitrogen atom, the most common of which have five- and six-membered ring systems. These systems are often used as building blocks in pharmaceuticals.

"These building blocks are very accessible and you can put them together like Legos to build compounds that we can then use for chemical or medicinal testing. But the problem is that a lot of these five or six membered ring systems are not as stable as you'd want them to be," Schindler said.

"The ring systems can break down in the body after a patient has ingested a therapeutic drug. Because the compound can be metabolized by the human body, what you give initially to a patient may not necessarily be what you would find in the body after the patient has taken it, and that is a problem."

Instead, researchers suggest using monocyclic azetidines, a more stable four-membered ring system. But, says Emily Wearing, lead author of the study who recently earned her doctorate from Schindler's lab, the key reactions chemists use to produce azetidines have specific challenges.

The reactions either can't be widely applied or they only produce azetidines with specific substitution patterns. Researchers want to produce azetidines with different substitution patterns because this allows researchers to try a variety of the molecule as building blocks in drug synthesis and drug screening.

Further, the U-M researchers used a method called a [2+2]-cycloaddition to create monocyclic azetidines. This method usually requires photoexcitation, or the excitation of atoms or molecules in a compound through the absorption of energy, according to Schindler. In other words, the reaction needs light.

In the reaction, the researchers used two classes of compounds called acyclic imines and alkenes, which are highly desirable as starting materials because they can be easily varied to produce different products, Wearing says. However, when you use light to excite the imine, the acyclic imine decays from the excited state before it can undergo the cycloaddition, Schindler says.

Previously, there has been a successful example of this reaction, Wearing says, but it used ultraviolet light, which presents safety challenges, and it used different imines and alkenes.

"This also means access to these highly desirable monocyclic azetidine building blocks is much more limited using this approach," Wearing said. "The use of visible light versus UV light is an important benefit, but our key discovery was being able to use a visible light approach to produce monocyclic azetidines."

Their method uses visible light and a photocatalyst to allow access to the required excited state intermediates in what's called an aza Paternò-Büchi reaction. To determine exactly why the reaction worked, Schindler's lab teamed up with the lab of Heather Kulik, associate professor of chemical engineering at the Massachusetts Institute of Technology.

Her lab ran a computational analysis that found using specific classes of the imines and alkenes starting materials would facilitate a better match in energy between those starting materials, which lowered the barrier for reaction. They also analyzed what factors led to high yields of azetidines.

When researchers develop a new reaction like this, they also need to show that it can work for many combinations of substrates, according to Seren Parikh, a graduate student in Schindler's lab. He and postdoctoral research fellow Yu-Cheng Yeh showed that the team's reaction could work on multiple versions of imine and alkene compounds.

"Someone might show that a new reaction works, but if it only works on a single compound, it is not useful to anyone because pharmaceutical companies are likely wanting to use the reaction on their unique compound," Parikh said. "What we can do is show that the reaction works on a diverse range of substrates to essentially prove that the reaction is worth the pharmaceutical company's time to try."

Parikh and Yeh were able to show that they could produce six biologically relevant azetidine compounds, including using the reaction to attach an azetidine to an estrogen derivative, a natural steroid in the human body. Yeh also used this method to synthesize analogues of penaresidin B, which has been shown to be toxic to tumor cells. This is the first total synthesis of this natural product using the [2+2]-cycloaddition

"The synthesis of these azetidine compounds are examples to demonstrate that this synthetic methodology can be applied to make complicated molecules and medicine-like molecules," Yeh said.

Understanding what makes this chemical reaction work will allow the group and the field of medicinal chemistry to design related reactions in the future. New work can build upon this design principle to access other azetidines to be incorporated into new pharmaceuticals, Schindler says.

"Now we can access these types of building blocks that people have wanted for a long time, but couldn't directly access," she said. "The process we have developed can now be used in the future as basically a blueprint for future reaction development."