Skip to Content

Growing Retinas

Embryonic stem cells growing in a dish ­spontaneously form ­retina-like structures.

Source: “Self-organizing optic-cup morphogenesis in three-dimensional culture”
Mototsugu Eiraku, Yoshiki Sasai, et al.
Nature 472: 51-56

The eyes have it: A soup of chemicals and undifferentiated embryonic stem cells (gray) growing in a dish spontaneously generates two cup-shaped structures (green), which resemble the embryonic retina.

Results: Mouse embryonic stem cells growing in a dish can spontaneously assemble into three-­dimensional structures reminiscent of the early embryonic retina.

Why it matters: Most efforts to grow organlike structures from stem cells involve some kind of scaffolding, often coated with specific signaling molecules, to encourage growth of the proper cell types and tissue architecture. But these findings show that at least some aspects of organ development are pre­programmed into the cells, suggesting that it may be possible to grow some tissue structures much more simply. Such retinal tissue might eventually be able to replace human tissue damaged by degenerative eye diseases.

Methods: Researchers began with about 3,000 mouse embryonic stem cells mixed with a cocktail of chemicals involved in retinal development. Over a two-week period, clusters of stem cells began to grow into balloonlike sacs, which then grew inward. These dual-layer structures resembled the optic cup, an early developmental precursor to the retina.

Next Steps: The group is now working on transplanting these structures into blind mice, in the hope of restoring vision. They are also trying to replicate the research using human cells and hope to have a human version of the system within two years.

Evolving Faster

A new way of directing protein evolution could speed drug development

Source: “A system for the continuous directed evolution of biomolecules”
David R. Liu et al.
Nature 472: 499-503

Results: Scientists used a new approach to create an enzyme designed to bind to a specific target. The process, which involved 200 rounds of protein evolution that would have taken years with conventional methods, was completed in just a week.

Why it matters: Directed evolution—sequentially introducing mutations into a protein to generate a molecule that performs a desired function—can create antibodies and other proteins that fight diseases, including cancer. But current methods are often too slow and labor-intensive to be broadly useful in drug development.

Methods: Researchers engineered M13, a rapidly replicating bacteriophage that infects E. coli, to carry a gene for the protein they wanted to modify. They then grew the viruses inside E. coli cells in an environment designed to boost the number of mistakes made when the viral DNA was copied, generating a library of slightly different proteins. The researchers linked the desired function, such as the ability to bind to the target, to a substance the viruses needed for survival, so only those viruses with the best versions of the protein progressed to subsequent rounds of evolution. Evolution took place at a rate of up to 40 rounds per day, 100 times the rate achieved with other methods.

Next Steps: The team plans to use the system to produce therapeutic proteins and to study seminal questions in evolution, such as whether replicating the same evolutionary conditions will generate different outcomes—and what factors determine these outcomes if so.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.