From the Ground Up

Instrumental in launching Arabidopsis thaliana as a model system, Elliot Meyerowitz has since driven the use of computational modeling to study developmental biology.

By | February 1, 2017

George W. Beadle Professor of Biology
California Institute of Technology
Investigator, Howard Hughes Medical Institute
In 1980, Elliot Meyerowitz was a newly minted assistant professor of biology at the California Institute of Technology (Caltech) studying Drosophila melanogaster development. He was asked to teach a graduate genetics seminar, and after leading the session on plant development and discussing the subject with his then graduate student Robert Pruitt, he decided, with Pruitt, to dabble in plant genetics using Arabidopsis thaliana. Meyerowitz had become interested in plant genetics in graduate school and thought that there were opportunities to apply molecular cloning—a new technique that he had learned as a postdoc at Stanford University—to plants.

“The literature indicated that the Arabidopsis genome was relatively small, which at the time, was a prerequisite for molecular cloning. And the attempts to do mutagenesis in plants showed that chemical mutagenesis appeared very effective and that the genes segregated in a Mendelian fashion, which is not true in many plants because of polyploidy,” Meyerowitz says. “Arabidopsis self-fertilizes, so you can get homozygous mutants quickly and have more than 10,000 seeds from these small plants. You could grow a million plants in a boiler room instead of needing 50 acres to grow corn!”

“I think if animal developmental biologists were more open-minded about plant research, biology overall would benefit.”

Supplied with Arabidopsis seeds by Pruitt’s uncle, a plant breeder at Washington State University, the Meyerowitz lab began to fill with postdocs and graduate students interested in studying plant biology. Meyerowitz soon became an advocate of making Arabidopsis a model organism for studying plant genetics. The lab also continued to work on Drosophila until the 1990s, when the funding finally shifted completely from fruit flies to the study of Arabidopsis flower and plant development, regeneration, and stem cells. With a focus on developmental biology, Meyerowitz has driven advances in imaging and in computational approaches to understanding how plant stem cells form patterns and employ chemical and mechanical signaling to develop into a mature, flowering plant.

Here, Meyerowitz advocates for more awareness of plant research, talks about why he used to carry DNA in his pockets at plant meetings, and offers advice on how to help postdocs succeed.


Expanding horizons. Meyerowitz was born in Washington, DC, and raised in a Maryland suburb. He was interested in science from as early as he can remember, and in the summer of 1967, his junior year of high school, worked at the National Naval Medical Center in Bethesda, Maryland, analyzing lipids from rats using thin layer chromatography. Meyerowitz’s chemistry teacher encouraged him to apply to other universities besides the local University of Maryland.

A fishy start. Meyerowitz entered Columbia University in 1969. The biology department had recently undergone a number of changes, shifting its focus from traditional zoology and botany to the newly emerging molecular biology, thanks in part to professors such as Cyrus Levinthal, a physicist turned molecular and computational biologist. During his sophomore year, Meyerowitz sought Levinthal’s advice on finding a part-time paid position in a research lab to help defray his college costs. Because Meyerowitz had shown so much enthusiasm for Levinthal’s research during their meeting, Levinthal offered him a spot in his lab. Meyerowitz worked on the Amazon molly (Poecilia formosa), a so-called gynogenetic freshwater fish, in which sperm from males stimulates the eggs to develop but does not contribute genetic material to the offspring. “The question I was addressing was: To what degree was the morphology of neurons the same in animals that were genetically identical?” says Meyerowitz. He learned how to do painstaking serial sections of fish embryo brains and how to analyze neuronal patterns using computer programming. “It turns out that the neuron patterns were identical down to about one micron and that the pattern of midline crossover of the optic nerve was neither random nor all the same, but rather a one-third/two-thirds distribution. I used computational modeling for the analysis, so in some sense my approach has never changed since my undergraduate days.” Meyerowitz also went with Levinthal’s lab to the Woods Hole Marine Biological Laboratory for two summers. There he became fascinated by developmental biology.    

The eyes have it. Meyerowitz entered Yale University’s biology PhD program in 1973, joining Douglas Kankel’s fruit-fly developmental genetics lab as Kankel’s first graduate student. There Meyerowitz examined whether information transmitted from the embryonic fruit-fly eye to the brain is important for brain development and whether neuronal brain connections influence eye development. After first analyzing every available fly mutant with badly developed eyes, he focused on three, which had both a neuronal and an eye phenotype. Meyerowitz used genetic mosaics to show that if the mutation was only in the eye, the axons in the fly’s brain were also highly disorganized, but that the same mutations present only in brain tissue did not result in malformed eyes, providing evidence that the organization of the optic lobe in the brain depends on information from the fruit fly’s eye.

Cloning pioneer. Meyerowitz decided to try a warmer climate for his postdoc and in 1977 joined David Hogness’s lab at Stanford, where molecular cloning was just taking off. “I learned all of the current molecular biology there,” says Meyerowitz. The first Drosophila cDNA library had just been created and Meyerowitz helped to make some of the first fly genomic libraries. To do that, he first developed new vectors into which larger pieces of DNA could be cloned. Using these lambda phage and cosmid vectors, he explored the gene organization of the Drosophila salivary gland polytene chromosomes. Meyerowitz also met his wife, then a graduate student in Arthur Kornberg’s lab, “which is probably more important than anything I learned scientifically,” he says.


A new model system. Meyerowitz chose to begin his career as a principal investigator at Caltech because “there was already a rigorous tradition of developmental biology and genetics. And I would have to be on the ball to keep up with everyone there, which was a challenge I wanted to take.” Not wanting to lose any time while his lab space was being renovated, Meyerowitz conducted fly crosses in the kitchen of his apartment for six weeks. Although he continued to study fly polytene chromosomes, Arabidopsis became a new focus in the lab. By creating a genomic DNA library, his team confirmed that the plant’s genome was relatively small (about 108 base pairs) with few repetitive elements.

By the mid-1980s, Meyerowitz had formed close relationships with other Arabidopsis researchers and, along with Chris Somerville, Maarten Koornneef, David Meinke, and others, used scientific meetings to espouse the adoption of the plant as a research model. “We decided to make sure people shared their materials, and to promote that, I used to bring with me in my pocket the DNA clones that we had mapped along with the restriction polymorphism DNA maps we had created for chromosome walks,” says Meyerowitz. The community grew quickly, and with the help of the National Science Foundation, there were soon Arabidopsis stock centers and an international committee set up to sequence the plant’s genome. Meyerowitz’s graduate student Caren Chang first cloned and sequenced an Arabidopsis gene, the gene for alcohol dehydrogenase, providing proof that molecular genetic tools could be readily applied in plants. “I think because these community resources we developed began along with the community of researchers, it all really took off,” says Meyerowitz.

The ABC model. Graduate student John Bowman and a visiting professor from Monash University in Australia, David Smyth, collected Arabidopsis flower mutants and observed homeotic phenotypes in which one flower organ is transformed into another, such as petals into stamens. “We recognized that the mutants could tell us something fundamental about organ specification in plants,” says Meyerowitz. The group found that their mutants fell into one of three classes they called A, B, and C, which affected the identity of petals, sepals, and stamens, and, in turn, the whorls of the flower—the concentric rings into which the flower organs organize.

“This led us to propose the ABC model in 1991, still the basic model for floral organ specification,” says Meyerowitz. The model predicted the phenotypes of the single, double, and triple mutant flowers the researchers generated. “I don’t think it is technologies that are the primary roadblocks in science,” says Meyerowitz. “I think it’s our own ability to think about how organisms work. An example is the ABC model. There was not a single method or even type of mutation that hadn’t existed 50 years earlier.”

Research offshoots. The Meyerowitz lab continued to work on flower development, focusing on the shoot apical meristem (SAM), which houses plant stem cells capable of regeneration. In 1993, they identified the role of the CLAVATA1 gene, which regulates meristem and flower development. In 1999, the lab found that CLAVATA3, a secreted peptide, and CLAVATA1, a transmembrane receptor kinase, control the balance in the SAM between stem cell renewal and differentiation.

As the list of genes involved in flower development grew, Meyerowitz thought he needed a new approach to developmental genetics, not just for the Arabidopsis model system but for all organisms. Rather than gene diagrams showing the direction of informational flow, Meyerowitz wanted to use computer analyses to create quantitative hypotheses that captured the strength and feedback of gene interactions in order to generate predictive models of how cells interact with each other. His lab collaborated with computer scientists, physicists, and mathematicians. The modeling took Meyerowitz in many different directions, including analyzing the molecular basis for plant phyllotaxis, or leaf and flower patterns. In 2006, the lab showed that, starting in the SAM, polarized transport of the plant hormone auxin results in Arabidopsis flower patterning.

Meyerowitz also works on the role of mechanical signaling between cells and tissues. Work from his lab suggests that mechanical signaling is as important for plant patterning and growth as peptide and hormone signaling. “The work on peptide signaling, hormone signaling, phyllotactic patterns, and mechanical signaling is all converging on a computational model that might explain everything about how the SAM works,” says Meyerowitz. His lab also developed live-imaging methods using confocal microscopy that are now widely used to observe plant development. The impetus, says Meyerowitz, was to test some of the lab’s computational models of gene function and cell-cell interactions. In 2005, the lab observed, in real time, cycles of auxin buildup and depletion in the SAM that facilitate the spiral pattern of leaves around the stem of Arabidopsis.

Looking to the future. “The Arabidopsis revolution has pretty much answered the fundamental plant biology questions that existed 30 years ago. We know how most of the plant hormones work; we know a lot of the modalities by which plants interact with the environment; we know ever so much more about plant differentiation, stem cells, and development,” says Meyerowitz. “Today, we have new ways of thinking about these topics. It’s now a question of going from the gene to the whole-organism level and integrating information at every level. We can start to put the whole picture together, using the sorts of computational methods those in the field have worked so hard to develop. Maybe before the end of the century, we’ll have computational models that represent everything that’s happening so as to understand the full complexity of plants.”


Plant promotion. “About 15 years ago, I gave a talk on plant development in a session focused on animal development at the American Society for Cell Biology meeting. Before I even got up on stage, about half of the 10,000 people in the audience started to walk out. That’s an example of animal scientists not having any interest in plants. And I don’t care that they learn from me necessarily, but I think if they were more open-minded about plant research, biology overall would benefit. I think a lot of people don’t realize that many things found in animals were first discovered in plants—viruses, the cytoskeleton, transposable elements, microRNAs. Of the three major theories in biology—on cells, genes, and evolution—the cell and gene theories originated from the study of plants.”

Seeking balance. “Funding for plant research is another issue. Only about 2 percent of funding in the U.S. goes to fund plant research, which is not well balanced if you consider the importance of these organisms for our lives. The way our funding situation is set up, most of the money goes to health issues. But being able to eat is also an important health issue, as is living in air without too much carbon dioxide in it.”

Scientific pursuit. “I try not to compete with my former postdocs. They take what they want with them to their new academic positions, and it’s my job to find new ideas for my lab. The goal of having a postdoc is to train that person to go out and be an independent researcher or industry researcher or whatever they wish to be. You have to do whatever you can to help that. If you don’t, you are not contributing as much to the academic system as you ought to be. It’s the same when you review a paper. The goal is to help the authors, to tell them what they could do to make the paper better, not give the editor an excuse to turn down the paper. If you think about what goals are necessary to advance science and the scientific community, it becomes clear how we need to deal with these things.” 

Greatest Hits

  • Instrumental in establishing Arabidopsis thaliana as a widespread model plant system  
  • First cloning of a plant hormone receptor, for ethylene, with grad student Caren Chang and postdoc Tony Bleecker
  • First identification of a peptide signaling pathway in plants, with postdocs including Steven Clark and Jennifer Fletcher
  • With John Bowman and David Smyth, proposed the ABC model, the basic model of plant organ specification
  • Collaborating with computational biologists Eric Mjolsness and Henrik Jönsson, initiated a novel computational modeling–based approach to understanding plant development

Add a Comment

Avatar of: You



Sign In with your LabX Media Group Passport to leave a comment

Not a member? Register Now!

LabX Media Group Passport Logo

Popular Now

  1. Consilience, Episode 3: Cancer, Obscured
  2. RNAi’s Future in Drug-Target Screening
    News Analysis RNAi’s Future in Drug-Target Screening

    A recent CRISPR study contradicted years of RNA interference research on a well-studied cancer drug target. But is it the last nail in the coffin for RNAi as a screening tool? 

  3. A History of Screening for Natural Products to Fight Cancer
  4. Human Cord Plasma Protein Boosts Cognitive Function in Older Mice