Scientists at Johns Hopkins University have uncovered how humans develop sharp central vision before birth, identifying a carefully timed interaction between a vitamin A-derived molecule and thyroid hormones in the retina. The discovery, published in the Proceedings of the National Academy of Sciences, challenges a three-decade-old explanation for how the eye's key light-sensing cells form.
The research focused on the foveola — the tiny region at the center of the retina responsible for the sharpest vision. Though it makes up only a small portion of the retina, the foveola is responsible for about half of all human visual perception. Unlike the rest of the retina, which contains all three cone types (blue, green, red), the foveola contains only red and green cones — a specialization that has puzzled scientists for decades.
The prevailing theory suggested that blue cones formed in the center of the retina and later migrated outward. Instead, the new evidence reveals those cells remain in place but change their identity.
Using lab-grown retinal organoids — small clusters of tissue grown from fetal cells that closely mimic parts of the retina — the team observed a remarkable transformation. During weeks 10 through 12 of fetal development, a small number of blue cones appear in the developing foveola. By week 14, however, those cells have changed into red and green cones.
The researchers found this happens through two coordinated mechanisms. First, retinoic acid — a molecule derived from vitamin A — is broken down, reducing the formation of new blue cones. Then, thyroid hormones drive the remaining blue cones to convert into red and green cones.
"First, retinoic acid helps set the pattern. Then, thyroid hormone plays a role in converting the leftover cells," said Robert J. Johnston Jr., an associate professor of biology at Johns Hopkins who led the research. "That's very important because if you have those blue cones in there, you don't see as well."
The findings could eventually support new approaches to treating vision loss. The team is continuing to improve its retinal organoids so they more closely resemble the function of the human retina. The goal is to produce healthier photoreceptor cells for future cell replacement therapies targeting diseases such as macular degeneration, which currently has no cure.
"The main model in the field from about 30 years ago was that somehow the few blue cones you get in that region just move out of the way," Johnston said. "Our data supports a different model. These cells actually convert over time, which is really surprising."
