Flowering

white flower

A seed sprouts. At the growing tip of the green shoot that pokes through the ground is a cluster of undifferentiated cells. These cells could give rise to a leaf. Or to a flower petal. At this point, they have the potential for either. The cells divide and grow and the shoot gets taller, putting out leaves in a spiral, forming a rosette, but always retaining that undifferentiated cluster at the tip.

Then mysteriously the tip itself changes. It becomes a cluster of potential flowers. One by one, by some prearranged pattern, the flowers bloom. Each blossom, whether rose, petunia, or the cabbage-like Arabidopsis, has four types of organs arranged in rings, or whorls. In the outermost whorl are the sepals (protective and leaf-like). Then come the petals, stamens to produce pollen, and, at the center, an ovary (made of carpels). This, when it's fertilized, will become a fruit and bear seeds which fall to the ground and sprout, starting the cycle all over again.

What exactly happens at flowering time? What flips the switch from leaves to sepals, from growing taller to bearing fruit? Precisely which genes are at work? In spite of our fascination with the beauty of flowers and their economic importance as the precursors of fruits and seeds, it's a puzzle biologists have just begun to piece together.

"Traditional biologists have done a lot of descriptive work, using microscopes to see things as small as possible," notes Hong Ma, an associate professor of biology at Penn State, "but until 1980, there was very little information on any floral genes anywhere in the plant kingdom."

Not until then did biologists agree that Arabidopsis thaliana, a spindly little relative of cabbage and cauliflower, made a good experimental system—that it was the fruit fly of the plant kingdom, being small, easy to grow, and producing lots of tiny seeds. In December, an international group published its genome—some 25,500 genes—and the first nearly complete genetic sequence for any type of plant.

"How do we take advantage of this information to push forward our understanding of flower development?" Ma asks.

The first task is simply to identify all the genes that contribute to flower development. "We don't have the complete collection of genes, yet," says Ma. "We have hundreds of them, but the estimates are in the thousands. A large fraction of these genes are probably expressed at very low levels, either in a small number of cells or for a very short duration of time."

While a postdoctoral scholar at Caltech from 1988 to '90, Ma worked on isolating a gene he called AGAMOUS (lacking gametes). Since then Ma has been trying to understand the gene's function and that of other genes it interacts with. He does so by finding mutants in which the gene does not work or works in the wrong part of the plant.

"You don't know what is important to the flowering process until you have a mutant," Ma explains. "To go even further, you don't know if a particular step even exists, in a genetic sense, in that genes actually do something, until you find a mutation. Some processes are spontaneous. Some chemical reactions happen without an enzyme, without a catalyst. So we can't say things are governed by genes until we find a mutant."

AGAMOUS, when it isn't working, produces a very obvious defect. Although Arabidopsis flowers are only two millimeters wide—as small as a pinhead—making it hard to see many other defects without a microscope, the AGAMOUS mutants stand out in a tray of blooming plants. While the others are white dots, these flowers are brighter, fuller, almost frilly. "It's a flower within a flower," Ma explains. The plant produces petals in place of stamens; sepals replace the pistils; and inside the sepal is another flower.

red flower

"One of the important things we've learned is that if you take the gene away, the stamens and carpels cannot form. You have a lot of petals, and no stamens. If you put the gene in a different place in the flower—if it's expressed under a different promoter—it leads to production of reproductive organs in those additional places." So AGAMOUS is required for the plant to be fertile.

Recently, Ma and Yixing Wang, a postdoctoral scholar at Penn State, have devised a way to turn AGAMOUS on and off at different times in the flower's development. "Tagged onto the AGAMOUS protein we have a sequence of a protein that binds to a steroid hormone. Plants don't have this type of protein. So without the hormone, the tagged AGAMOUS is trapped in the cytoplasm of the cell. If we add the hormone to the plant, the tagged protein changes shape and AGAMOUS can go into the nucleus, where it will work. We can observe its effect one to two days later."

The protein made when the AGAMOUS gene is expressed affects several stages in the flower's development, Ma and Wang have learned. Petals, stamens, and carpels all depend on it at different times and for different durations; it may also play a role in regulating pollen development. Yet it's unclear whether AGAMOUS is turning on one gene at different times, to create these effects, or triggering a set of genes. "I believe it's turning on a set of genes," Ma says, "but we don't have any proof yet."

Hong Ma, Ph.D., is associate professor of biology in the Eberly College of Science and a member of the Life Sciences Consortium, 512 Wartik Bldg., University Park, PA 16802; 814-863-6414; hxm16@psu.edu. Collaborating with him are postdoctoral fellows Yixing Wang, Dazhong Zhao, and Changbin Chen; graduate students Qilu Yu and Wei Hu; and undergraduate students Katrina Getz, Eric Harris, Matthew Henry, Niraj Mehta, Megan Petrasek, John Quiles, Jessica Riley, and Andrea Varkonyi. Their work is funded by the Penn State Biology Department and Life Sciences Consortium, the National Science Foundation, the USDA, the National Institutes of Health, and the American Cancer Society.

Last Updated May 01, 2001