A dendritic molecule is one that grows by branching in several directions from its center core. At each branching point, the molecule branches again into a new generation. These molecules can be used for a broad range of biomedical applications, including gene and drug delivery.

In 1983, Nobel Laureate Pierre-Gilles de Gennes predicted that as these molecules continue branching into new generations, the functional groups on the periphery will get so crowded that their reactivity will decrease.

But now, in a paper published in the Proceedings of the National Academy of Sciences, researchers at the University of Pennsylvania have shown that the organic synthesis breaks down completely and the reactions stop altogether. These results could improve the efficiency of gene and drug delivery and could also provide a new plan of attack in the rising antimicrobial-resistance problem.

The research was led by Virgil Percec, the P. Roy Vagelos Professor of Chemistry in the School of Arts & Sciences at Penn; postdocs Davit Jishkariani, Yam Timsina, Silvia Grama, Srujana Yadavalli, Ralph-Olivier Moussodia and Pawaret Leowanawat; graduate students Syeda Gillani and Masoumeh Divar; and undergraduate student Angely Berrios Camacho. Penn alumnus Christopher MacDermaid provided the computational simulations.

​​​​​​​​​​​​​​“This is a fundamental event for the field of dendrimes, as well as the fields of organic chemistry and iterative synthesis,” Percec said. “When you take these molecules and interact them with nucleic acids, the reactivity becomes zero and they stop reacting. Therefore, you can predict what molecule to make at a smaller generation so that all the groups are going to interact.”

Donald Tomalia, the chemist who produced the first of this type of dendrimer, said this paper “provides some of the first quantitative insight into many issues raised by de Gennes seminal hypothesis presented in 1983.”

The paper suggests that the conformation, or structure, of the core, not the multiplicity (how many times it grows from a single point), may play a role in this self-limiting growth phenomena.

After synthesizing and studying these molecules, the researchers did a year of supercomputer simulations trying to elucidate the mechanism of this interaction.

“Typically when you do simulations,” MacDermaid said, “you're limited by how long you can actually look at the molecule. We're looking at snapshots that are 9 billion times within a second, and we simulated this essentially for 100 nanoseconds. In that time, we were actually able to provide valuable insight that helped them determine what to do next or to at least help explain the phenomenon that they're seeing.”

Percec said there are “an infinite number of opportunities coming from this very fundamental science that we discovered.”

He hopes to exploit this research for various applications, as well as to figure out mechanisms that will allow them to overcome this limitation.

In addition to guiding scientists in synthesizing molecules with just the right generation number to effectively transport genes or drugs, this work could lead to new agents in the fight against antimicrobial resistance.

​​​​​​​“There are a lot of bugs in clinical settings and healthcare settings,” Yadavalli said, “that are constantly evolving resistance mechanisms and are very intractable to the current antimicrobial agents. It’s great to find these new dendritic molecules with both stable properties and antimicrobial properties that can be useful for developing new classes of antimicrobial agents that can attack this problem.”

The research was supported National Science Foundation grants DMR-1066116 and DMR-1120901, the P. Roy Vagelos Chair at Penn, the Humboldt Foundation, the Joseph and Josephine Rabinowitz Award for Excellence in Research from the Penn School of Dental Medicine and National Institutes of Health Grant R01-GM080279. Computer time was provided through National Science Foundation Grant ACI- 1614804, the Petascale Computing Resource Allocations and support from the National Center for Supercomputer Application.