Lab-grown retinas reveal how the eye builds its sharpest vision

Overview

Updated 1 hour ago

The foveola is a sliver of tissue at the very center of the retina, and the researchers say it accounts for roughly half of human visual perception. It also has an odd composition: it holds red and green cones but no blue ones. Johns Hopkins scientists grew that region in a dish and watched it assemble itself, then described a mechanism that contradicts an explanation the field has used for about 30 years.

Between weeks 10 and 12 of fetal development, the foveola contains a small number of blue-sensitive cones. By week 14 those cells appear to be red and green cones instead. The old model said the blue cones simply moved out of the way. The new data suggest they changed identity, with a vitamin A derivative setting the pattern and thyroid hormones driving the conversion.

Why it matters

If the conversion holds up, researchers gain molecular instructions for building the exact photoreceptor mix that macular degeneration destroys.

Questions about this story

Free account needed to ask — your question is kept and asked for you right after sign-up. Answers are public.

No questions yet — be the first to ask.

Key Indicators

50%

Share of visual perception from the foveola

The foveola is a tiny fraction of the retina but handles detailed central vision.

Human cone types

Blue, green and red cones, each most sensitive to a different wavelength band.

Weeks 10-12

When blue cones appear in the foveola

A small number are present at this stage of fetal development.

Week 14

When they have become red and green cones

The change in identity is visible in lab-grown retinal tissue by this point.

~30 years

Age of the model being challenged

The field's main explanation for the missing blue cones dated to about three decades ago.

Voices

Curated perspectives — historical figures and your fellow readers.

People Involved

Robert J. Johnston Jr.

Associate professor of biology, Johns Hopkins University; lead author

Leading the retinal organoid program at Johns Hopkins

Katarzyna A. Hussey

First author on the study; former doctoral student in Johnston's lab

Molecular and cell biologist at CiRC Biosciences in Chicago

Organizations Involved

Johns Hopkins University

Research university

Home institution of the study

Baltimore university whose biology department runs the retinal organoid program behind the findings.

Proceedings of the National Academy of Sciences (PNAS)

Peer-reviewed journal

Published the study

Journal of the US National Academy of Sciences, where the cone-patterning paper appeared.

CiRC Biosciences

Cell therapy company

Employer of the study's first author

Chicago cell therapy company where first author Katarzyna Hussey now works.

Timeline

February 2026 October 2026

3 events Latest: Today

October 2026
Johns Hopkins researchers describe conversion of blue cones
Today Research

ScienceDaily's report on the study says the findings challenge a model that has stood for about 30 years.
Today
July 2026
ScienceDaily reports the retinoic acid and thyroid hormone sequence
Coverage

Coverage describes the two-step mechanism: retinoic acid limits new blue cones, thyroid hormones convert the rest.
July 8th, 2026
February 2026
Johns Hopkins publishes the foveolar cone study
Research

The paper appears in PNAS, reporting that blue cones in the foveola become red and green cones rather than migrating away.
February 18th, 2026

Scenarios

Independent labs reproduce the cone conversion

Possible Resolves by Q2 2028

Discussed by: Johnston and colleagues, who note their data cannot yet fully exclude cell migration

A second group growing human retinal tissue finds the same transition from blue to red and green cones in the foveolar region. Replication would move the conversion model from one lab's organoids toward a general account of how the human fovea forms.

Follow-up work finds migration, not conversion, explains most of the pattern

Possible Resolves by End of 2029

Discussed by: The authors themselves, who say they cannot yet rule out the older model

Tracing individual cells over time shows that most blue cones leave the central region, with identity switching accounting for only a small share. That would restore a modified version of the three-decade-old explanation and narrow the clinical implications.

Organoid-derived photoreceptors enter a first human safety trial

Unlikely Resolves by End of 2031

Discussed by: Hussey, who calls cell replacement a viable but long-term path

A sponsor files to test transplanted retinal cells in people with advanced vision loss. The step requires showing the cells survive, wire into existing circuitry and do not form tumors, which the researchers say is years of work.

Historical Context

3 moments from history that rhyme with this story — and how they unfolded.

1990s

The cone-translocation model (c. mid-1990s)

Studies of primate and human retinas concluded that the few blue cones present in the developing fovea simply move outward as the region matures. The same work held that a cone's subtype is fixed once assigned.

Then

The migration account became the field's default explanation for why the foveola has no blue cones.

Now

It shaped how researchers interpreted foveal anatomy for roughly three decades, largely untested against living human tissue.

Why this matters now

The new organoid data supports conversion instead of displacement, which is the specific claim under dispute.

2018

Thyroid hormone and cone subtypes in human organoids (2018)

Researchers in Johnston's lab used lab-grown human retinal tissue to show that thyroid hormone signaling helps determine whether a cone becomes blue-sensitive or red and green-sensitive.

Then

It established that human cone identity is set by chemical signals rather than fixed in advance.

Now

It made retinal organoids a working tool for questions that animal models could not answer.

Why this matters now

The new study extends that finding: thyroid hormones do not just assign cone identity, they appear to reassign it.

2011-2012

Self-organizing retinal organoids (2011-2012)

Stem cell researchers reported that cultured cells could assemble on their own into layered optic-cup and retinal structures, without a scaffold dictating the architecture.

Then

The results showed that key steps of eye development could be observed outside an embryo.

Now

Labs including Johns Hopkins built on the technique to model human-specific retinal features.

Why this matters now

The foveola findings depend on this lineage: without self-organizing human tissue, the blue-cone transition could not be watched as it happened.