Biochemists uncover new rules of mitochondrial protein import

Mitochondria are cellular organelles that play an important role in making ATP (adenosine triphosphate), the molecular fuel that powers most cellular functions. These organelles originated over a billion years ago when a primitive archaeal cell entered into a symbiotic relationship with an ancestral bacterium. Over time, mitochondria became essential for metabolism and energy production, while transferring most of their genes to the host. As a result, they now rely on the host cell to supply most of their proteins, which are synthesized by ribosomes outside the organelle and must be properly delivered to mitochondria.

Now, Caltech scientists have uncovered new details about how mitochondrial proteins get delivered from ribosomes in the cytosol, the fluid around the nucleus, to mitochondria. In a surprising twist, the process is largely shaped by the technicalities of protein folding.

“It turns out that localizing proteins to mitochondria involves a multilayered, complex pathway that is wired around the biophysical principles of protein folding,” says Shu-ou Shan, the Altair Professor of Chemistry at Caltech.

For decades, the dominant model in biochemistry has held that mitochondrial proteins are imported only after protein synthesis, or translation, has completely finished. (That ribosome-driven process involves adding amino acids one by one to a growing chain, following the sequence laid out by the cell’s genetic code.) In a new paper that appears in the journal Cell, Shan and her colleagues offer a revision to this model, showing that up to 20% of mitochondrial proteins can be cotranslationally imported, meaning they enter into mitochondria during translation when the proteins are still in the process of being synthesized by the ribosome.

“Once we identified these mitochondrial proteins that are cotranslationally imported, we asked, “What is special about this subset of proteins?'” says Zikun Zhu (Ph.D. ’24), Shan’s former grad student and lead author of the paper.

It turns out that the most prominent feature of these proteins is that they are large molecules that fold in complex ways. Such topologically complex proteins are rich in residues—amino acids in the chain that makes up the protein—that, while distant from one another in linear sequence, need to bind together for the protein to fold into the proper three-dimensional structure. “That becomes a much more difficult process than just folding through interactions between neighboring residues,” Shan says.

As a result, the system for cotranslational import into mitochondria prioritizes these really difficult-to-fold proteins. This makes sense if you consider that the large structures have to eventually go through narrow channels on the mitochondrial membrane during import. “There is going to be a problem if you let these large, very complex proteins finish translation in the cytosol,” says Shan. “They will get stuck in irreversible structures, and then you will not only block import, you will clog all the channels.”

But how does the cell know which proteins need to be imported during translation?

The team found that nearly all such proteins carry a mitochondrial targeting sequence, which is a signal that directs proteins to mitochondria. Yet, surprisingly, this alone is not enough to tell this subset of proteins to be delivered during translation. Zhu conducted experiments that showed that the system waits for a second molecular signal to move a protein to the mitochondria early. That signal comes in the form of the first large protein domain, or foldable structural unit within the sequence, that emerges from the ribosome.

“It’s like having your boarding pass locked in a suitcase,” Zhu says. “The targeting sequence is the boarding pass, but to access it, you need the code to open the suitcase. In this case, the large domain is that code.”

The scientists were even able to transplant examples of such large protein domains to other mitochondrial proteins that are normally imported after translation and showed that the domains indeed served as transferable signals capable of rerouting proteins to be imported during translation.

“Cotranslational targeting to mitochondria turns out to be completely different from targeting to other organelles,” Zhu says. “Going forward, it will be exciting to uncover more mechanistic details and, ultimately, to manipulate the timing of mitochondrial protein import. This will not only help us understand why cells evolved such a sophisticated targeting pathway for mitochondrial proteins but also open the door to potential therapeutic applications.”

The paper is titled, “Principles of cotranslational mitochondrial protein import.” Additional Caltech authors are Taylor A. Stevens (Ph.D. ’24), a postdoctoral scholar research associate in biology and biological engineering, and Riming Huang, a graduate student in biochemistry and molecular biophysics. Saurav Mallik of the Weizmann Institute of Science in Israel and Emmanuel D. Levy of the University of Geneva in Switzerland are also authors.